Skip to main content

Full text of "Enzyme and metabolic inhibitors"

^ 



Enzyme 
and Metabolic 
Inhibitors 

Volume II 

Malonate, A nalogs, Dehydroacetate, 

Sulfhydryl Reagents, o-Iodosobenzoate, Mercurials 



Volume I General Principles of Enzyme Inhibition 

Volume III lodoacetate 
Maleate 

A^-Ethylmaleimide 
Alloxan 
Ouinones 
Arsenicals 



Volume IV 



in preparation 



Uncouplers of Oxidative Phosphorylation 

Dinitrophenol 

Arsenate 

Cyanide 

Carbon Monoxide 

Azide 

Sulfide 

Antimycin 

Fluoroacetate 

Parapyruvate 

Urethanes 

Chelating Agents 



Volume V 



in preparation 



Protein Group Reagents 

Heavy Metals (Copper, Zinc, Cadmium, Silver, etc.) 

Dyes 

Aldehydes 

Dimercaprol 

Antienzymes 

Phlorizin 

Selenite and Tellurite 

Naturally Occurring Inhibitors 

Cholinesterase Inhibitors 

Monoamine Oxidase Inhibitors 

Drugs as Inhibitors 

Carbonic Anhydrase Inhibitors 

Borate 



Enzyme 
and Metabolic 
Inhibitors 



Volume II 

Malonate, Analogs, Dehydroacetate, 

Sulfhydryl Reagents, o-lodosobenzoate. Mercurials 



J. LEYDEN WEBB 

School of Medicine 

University of Southern California 

Los Angeles, California 





1966 



ACADEMIC PRESS New York and London 



Copyright © 1966, by Academic Press Inc. 
all rights reserved. 

no part of this book may be reproduced in any form, 
by photostat, microfilm, or any other means, without 
writt'^n permission from the publishers. 



ACADEMIC PRESS INC. 

Ill Fifth Avenue, New York, .s'ew York 10003 



United Kingdom Edition published by 
ACADEMIC PRESS INC. (LONDON) LTD. 
Berkeley Square House, London W.l 



Library of Congress Catalog Card Number: 62-13126 



PRINTED IN THE UNITED STATES OF AMERICA 



This volume is dedicated with sincere gratitude to the medical librarians 
— Vilma, Ruth, Lilian, Clara, Esther, Michelle, Rose, Shari, Nancy, Nahida, 
and others — who have not only helped me to the limit, hut have made each 
visit to the library a pleasure and often the most delightful experience of 
the day. 



PREFACE 



For those rare readers who may feel inclined to pursue their way through 
Volumes II and III from beginning to end, I have tried to arrange the 
chapters and sections in a logical and interdependent order. Malonate has 
been approached first because its actions so well illustrate some of the 
general principles covered in Volume I, and, indeed, malonate is discussed 
in greater detail than any other inhibitor in order to suggest how one 
would like to deal with all inhibitors if one had either the time or space. 
Inasmuch as malonate is the classic substrate analog, the next chapter 
takes up various types of analogs and here we are able to obtain some 
rough idea of the energies involved in the interactions of inhibitors with 
enzyme surfaces, as well as study some of the factors which determine 
specificity. Some readers may feel that too much attention has been given 
to these analogs, but I believe they represent a very important group of 
inhibitors and illustrate many principles — competitive behavior, group 
specific interactions, protection and reversal, and even nuituil depk Lion 
kinetics since some analogs are extremely potent inhibitors — and, in 
addition, contribute to our understanding of feedback inhibition and me- 
tabolic regulation. Most of the remainder of the volumes is devoted to sub- 
stances considered to react with SH groups, certainly one of the most 
commonly used and important classes of inhibitors, about which it is sur- 
prisingly difficult to find adequate and comprehensive treatment. Certain 
aspects of inhibition have been treated in detail, not necessarily because 
of any intrinsic importance, but because of the information which is 
provided to help us comprehend the general phenomena of inhibition. 
There are many ways of writing about inhibitors, and I have tried to alter 
the approach according to what I believe to be the most interesting aspects 
of each inhibitor. These aspects may not happen to be those which would 
have been chosen by the reader, but it is impossible to cover any inhibitor 
completely and present it from all viewpoints. On the other hand, there 
are certain sections which I have been unable to make very interesting, 
to organize into a coherent picture, sometimes because the data are insuf- 
ficient or too heterogeneous, but nevertheless some worthwhile material 



vu 



Vlll PREFACE 

can often be included in these areas. One finds much of the subject to be 
somewhat disconnected and it is seldom possible to present an orderly- 
clear version of any inhibition because of the gaps in our knowledge, but 
one must consider that these isolated strands may some day be woven 
into a durable fabric. Each chapter has in general been organized so that 
the treatment proceeds from the simplest system to the higher levels of 
organization, since this generates progressive understanding it may be 
hoped, although one occasionally wishes that the effects on the simpler 
systems could be appreciated against a background of the actions on tissues 
and animals. Perhaps to some extent the historical introductions at the 
beginning of most chapters may serve as provisional backgrounds. As in 
many fields of science one is confronted with the problem of vertical or 
horizontal presentations. Efforts, however inadequate, have been made to 
correlate the results at the different levels, and it is hoped that unlike 
bacteria under certain conditions this volume does not too much exhibit 
the phenomenon of accumulation without synthesis, or suffer from an even 
worse danger, that in the psychosynthesis of concepts and over-all pictures 
some abnormal or spurious units have been lethally incorporated. 

This periplus of the field of enzyme inhibition presents a rather large 
and often heterogeneous group of information, but everything has been 
selected for some reason; the reasons may be debatable, since different 
readers come to a book for different purposes, but occasionally one detects 
something in a report, perhaps intuitively, which others would not, and 
hence includes it for reasons difficult to express. The half-life for the general 
use of a book is, indeed, determined in part by the ability or good fortune 
of the author to select that which will have the most value or pertinence 
in the pseudopodal fronts of science. One has no time for justifications, 
since some decisions have to be made, and only the naive think they can 
please or help everyone, but there is perhaps one justification I feel impelled 
to make. Certainly there will be those who ask why the effects of an inhib- 
itor on the blood pressure or the central nervous system have been pre- 
sented when there is little or no obvious correlation with any metabolic 
inhibition, or why I have made up tables of tolerated or lethal doses, and, 
in general, some may criticize the discussion of inhibitor actions which are 
likely to be unrelated to enzyme inhibition, or at least for which there is 
no direct evidence. In defense of this, I can only say that I believe we 
should not so rigorously categorize the actions of inhibitors. The refusal 
to consider the nonmetabolic actions has led many investigators to very 
biased interpretations of their data. If we are interested in the mercurials, 
we are, I assume, interested in all their possible actions, whether they are 



PREFACE IX 

based on metabolic disturbances or not. To be narrow here would be like 
discussing only the beneficial effects of drugs and omitting the toxic actions. 
Of course, space limitations make it impossible to treat all these actions 
equally, and I have tried to emphasize those actions in which a disturbance 
of metabolism is the most likely mechanism. But we must never ignore the 
possibilities of other mechanisms with any inhibitor, particularly those 
reacting with groups on proteins and other cell components. The mercurials 
offer an especially clear example of enzyme inhibitors producing character- 
istic effects on many tissues (e.g. kidney, heart, central nervous system, 
liver, muscle, etc.) and where not a single action can be definitely corre- 
lated with a mechanism involving enzyme inhibition. Nevertheless, with 
further improvements in techniques and more knowledge, it is quite pos- 
sible that in the future at least some of these actions will be related to 
effects on enzymes. To be perfectly honest, at the present time we cannot 
say in the majority of cases just how substances called enzyme inhibitors 
act to produce their interesting and often clinically or industrially important 
effects on microorganisms or tissues, and it is necessary to realize our igno- 
rance so that progress in understanding may take place. Inhibitors do 
produce some very intriguing effects on tissue function or in whole animals, 
and many of these effects are unknown to those who look upon inhibitors 
merely as biochemical tools; so by reading of these effects some may be 
activated to study the ultimate causes in greater detail. Incidentally, 
since inhibitors will be used more and more frequently in animals, infor- 
mation on dosage ranges to produce various effects may serve a very prac- 
tical purpose. 

I would like to express my gratitude to those who have written to me 
saying they have found the first volume of interest or of some value to 
them, who have sent me unpublished manuscripts or difficultly obtainable 
material, and who have given me encouragement during those periods 
when I sincerely wished I were in a monastery in Kyoto. 

J. Leyden Webb 
November, 1965 




CONTENTS 



Dedication v 

Preface vii 

Introduction xv 

Conventions xix 

Symbols xix 

CHAPTER 1 

Malonate 1 

Early Historical Development 1 

Chemical Properties 3 

Inhibition of Succinate Dehydrogenase 15 

Inhibition of Succinate Oxidation in Cellular Preparations .... 50 

Inhibitions of Enzymes Other Than Succinate Dehydrogenase ... 58 

Effects of Malonate on the Operation of the Tricarboxylic Acid Cycle . 69 

Accumulation of Succinate during Malonate Inhibition 90 

Accumulation of Cycle Substrates Other than Succinate 104 

Antagonism of Malonate Inhibition with Fumarate 112 

Specificity of Malonate Inhibition in the Cycle 117 

Effects of Malonate on Oxidative Phosphorylation 118 

Effects of Malonate on Glucose Metabolism 122 

Effects of Malonate on Lipid Metabolism 135 

Effects of Malonate on Amino Acid and Protein Metabolism .... 151 

Effects of Malonate on Porphyrin Synthesis 158 

Effects of Malonate on Miscellaneous Metabolic Pathways 163 

Effects of Malonate on the Endogenous Respiration 166 

Permeability of Cells to Malonate 186 

Growth, Development, and Differentiation 192 

Cellular and Tissue Function 202 

Effects of Malonate in the Whole Animal 217 

Effects of Malonate on Bacterial Infections 221 

Metabolism of Malonate 224 

Inhibitors Structurally Related to Malonate 235 



XI 



xii CONTENTS 

CHAPTER 2 

Analogs of Enzyme Reaction Components 245 

Terminology 246 

Possible Sites and Mechanisms of Inhibition 246 

Kinetics of Analog Inhibition 248 

Means of Expressing Results 252 

Important Types of Molecular Alteration Producing Inhibiting Analogs . 255 

Development of the Concept of Inhibition by Analogs 259 

Analog Inhibition of Membrane Transport 261 

Analogs Which Are Isomers of Substrates 268 

Fumarase 274 

Inhibition of Xanthine Oxidase by Purine Analogs and Pteridines . . 279 

Choline Oxidase 290 

Inhibition of Nitrogen Fixation by Other Gases 291 

Phenol Oxidases 296 

Tyrosine Metabolism 302 

Tryptophan Metabolism 321 

Glutamate Metabolism 327 

Arginase 335 

L-Amino Acid Oxidases 338 

D-Amino Acid Oxidase 340 

Analog Inhibition of the Metabolism of Various Amino Acids .... 350 

Diamine Oxidase (Histaminase) 360 

Carboxypeptidase, Aminopeptidases, and Dipeptidases 365 

Chymotrypsin and Other Proteolytic Enzymes 368 

Hexokinases 376 

Effects of 2-Deoxy-D-Glucose on Carbohydrate Metabolism .... 386 

Effects of 6-Deoxy-6-Fluoro-D-Glucose on Metabolism 403 

Various Analog Inhibitors of Carbohydrate Metabolism 405 

Glycosidases 415 

Pyruvate Metabolism 429 

Lactate Metabolism 432 

Phosphatases 439 

Sulfatases 443 

Adenosinetriphosphatases and Transphosphorylases 444 

Hydroxysteroid Dehydrogenases 447 

Nitrite and Sulfite Metabolism 450 

Simple Ion Antagonisms 452 

Inhibition by Macroions 453 

Inhibitions by Nucleotides and Related Substances 465 

Inhibitions by Coenzyme Analogs 482 

Analogs of Nicotinamide and the Pyridine Nucleotides 484 

Analogs of Thiamine 514 

Analogs of Riboflavin and FAD 534 

Analogs of Pyridoxal 561 

Analogs of Pteroylglutamate (Folate) 579 



CONTENTS Xni 

Analogs of Other Vitamins, Coenzymes, and Their Components . . . 586 

Miscellaneous Analog Inhibitions 590 

CHAPTER 3 

Dehydroacetate 617 

Chemical Properties 618 

Inhibition of Enzymes 620 

Effects on Respiration and Glycolysis 623 

Effects on Tissue Functions 624 

Effects on the Whole Animal 627 

Distribution and Metabolism 629 

Antimicrobial Activity 631 

CHAPTER 4 

Sulfhydryl Reagents 635 

Role of SH Groups in Metabolism and Function 636 

Chemical Properties of SH Groups 637 

Types of SH Reaction Important in Inhibition 642 

Factors Determining the Reactivities of SH Groups 643 

Interpretation of Inhibitions by SH Reagents 647 

Protection and Inhibition Reversal by Thiols 650 

General Considerations of the Uses of SH Reagents 651 

CHAPTER 5 

Oxidants 655 

Disulfides 661 

Porphyrexide and Porphyrindin 664 

Ferricyanide 670 

Iodine 678 

Peroxides 690 

Tetrathionate 696 

CHAPTER 6 

o-lodosobenzoate 701 

Chemistry 701 

Reaction with Protein SH Groups 703 

Inhibition of Enzymes . 704 

Inhibition of Metabolism 721 

Effects on Animal Tissue Functions 723 

Effects in Whole Animals 724 

Effects on Sea Urchin Egg Development 726 

Effects on Bacteria and Viruses 727 



XIV CONTENTS 

CHAPTER 7 

Mercurials 729 

Chemical Properties 730 

Reactions with Proteins 751 

Inhibition of Enzymes 768 

Electron Transport and Oxidative Phosphorylation 870 

Fermentation and Glycolysis 874 

Tricarboxylate Cycle 877 

Respiration 879 

Various Metabolic Pathways 886 

The Cell Membrane as a Site for Mercurial Action 892 

Effects on Permeability and Active Transport 907 

Effects on the Kidney 917 

Effects on Tissue Functions 937 

Effects Observed in the Whole Animal 950 

Effects on Mitosis, Growth, and Differentiation 963 

Effects on the Growth of Microorganisms 970 

Development of Resistance to Mercurials 983 

References 987 

Author Index 1071 

Subject Index 1125 



INTRODUCTION 



Certain principles or prejudices in the approach should be clearly stated 
since these occasionally heretical opinions have been the basis for much of 
the organization of this volume. 

What is hoped to be pertinent information on the physical and chemical 
properties of the inhibitors has been given in the belief that the proper use 
of any inhibitor requires as much knowledge of its properties as possible. 
Such data are often difficult to find and I am afraid that much important 
material has been omitted. 

The quantitative formulation of inhibitions has been stressed because 
here, as in every science, progress often depends on accurate recording and 
reporting of observations. It is, for example, not only not informative but 
actually misleading to state that aldehyde oxidase is or is not inhibited 
by p-mercuribenzoate. What is meant by " inhibited " — 10%, 50%, or 
100%? What is the concentration of the inhibitor? If it is 0.01 mM it may 
mean something, but if it is 10 mM probably nothing. What is the source 
of the enzyme? There are many aldehyde oxidases and they differ quite 
markedly according to their sources. What substrate and acceptor were 
used? Is the substrate acetaldehyde, glyceraldehyde, formaldehyde, reti- 
nene, or even hypoxanthine, and are electron acceptor dyes used or the Og 
uptake measured? These and other factors must be made explicit. Of 
course, it is impossible in a book like this to give a complete picture of 
each inhibition mentioned, but I have tried to state the source of the 
preparation, the substrate, the inhibitor concentration, and the per cent 
inhibition in every case, as well as the pH and the incubation times when 
necessary. 

One can understand a phenomenon better when it can be visualized in 
some manner and much of the recent work on the inhibition of enzymes 
has been done with the purpose of clarifying the topography of the en- 
zyme surface and the nature of the interactions occurring there. Thus I 
have tried to emphasize the interpretation of data in terms of an accurate 
delineation of group orientation and intermolecular forces, although in the 
present state of our knowledge this can seldom be done satisfactorily. 



XV 



XVI INTRODUCTION 

Metabolism within cells is almost always a matter of multienzyme sys- 
tems and so the effects of inhibitors on such systems have been discussed 
fully wherever possible, although this is even more difficult to describe 
quantitatively than the behavior of single enzymes. 

The importance of the specificity of inhibition was sufficiently emphasiz- 
ed in the previous volume and it should be clear that this is a critical prob- 
lem which has been neglected, ignored, or abused extensively. It is not 
an easy matter to evaluate the specificity of an inhibitor under various 
conditions, particularly when the necessary data are lacking, but it is 
hoped that at least a provisional picture has been presented in some in- 
stances. 

Certain aspects of metabolism (e.g. glucose utilization, respiration, pho- 
tosynthesis, protein synthesis, or oxidative phosphorylation) and cellular 
activity (e.g. active transport, membrane potentials, movement, mitosis, 
or proliferation) are obviously of general significance, and the effects of 
inhibitors on these have been emphasized. This is not to say that other 
pathways or functions are unimportant, and indeed where necessary they 
have been treated as adequately as possible, but one cannot discuss all 
the actions of each inhibitor, so that some compromises must be made. 

A major use of inhibitors is in the attempt to correlate cellular functions 
with particular enzymes or metabolic pathways, and for this reason, as 
well as the fact that this represents one of the most fascinating aspects of 
inhibitor study, these correlations have been discussed fully if the infor- 
mation has been available, and the effects on certain organisms or processes 
have often been given in the hope that some correlation will emerge or 
further work will be stimulated. It is believed that conceiving inhibitor 
actions in terms of deviations in the energy flow is of some value although 
an accurate formulation of this must await the development of a new 
terminology. 

It is simpler to restrict the treatment of an inhibitor's action to a par- 
ticular organism or tissue, but it is felt that a great deal may be learned 
from comparative inhibitor enzymology. Therefore, in the tables, the ef- 
fort has been made to present the results from as many sources as possible 
for a particular enzyme or metabolic pathway since by doing this one is 
better able to see the great extent of the variability in responses; only a 
distorted view is obtained if a limited range of action is considered. 

Paradoxical actions have been both the despair and delight of scientists 
in many fields, and it is recognized that some of our finest theories have 
originated in the observation and study of anomalies. There is an inherent 
desire in most of us to eliminate anomalies and perhaps devote a good deal 



INTRODUCTION XVU 

of effort to this, since we feel that an anomaly really is something we would 
expect if we knew the system or mechanism better, or as Henry Miller 
has said in the " Tropic of Capricorn," " confusion is a word we have in- 
vented for an order which is not understood." I have thus brought up 
certain so-called anomalies, not only for their interest but again because 
they often stimulate deeper investigation, although at present they may 
to some only confuse the picture. 

Many of the results have been put into tabular form, first because this 
is the most efficient way of presenting certain types of data, second because 
such simple observations are often the sole information on the inhibitors 
provided in the reports, third because this allows a more convenient com- 
parison of results (e.g., for those interested in possible phylogenetic rela- 
tionships, for which reason the source organisms have usually been given 
in the classic taxonomic sequence, or for studying the variability in re- 
sponses on a comparative basis), fourth because this is the clearest way 
to provide information from which specificity may be evaluated, and fifth 
because these tables may serve as reference sources for those interested in 
the actions of a particular inhibitor on a certain enzyme or organism. 
There is much more in these tables than anyone can assimilate or under- 
stand or interpret today, but it is these data which could possibly con- 
tribute to some idea or concept if placed against the proper experience or 
background. Nothing makes some data look more miserable or incomplete 
than putting them in tables, but perhaps this is an asset, since it shows 
what is missing, what should have been done, and what more there is to 
do. A great deal of information could not be included in the tables, for, 
although some of them look formidably long, they represent only a frac- 
tion of what is available in reports. One tries to include only that which 
is important, but the definition of this word becomes more difficult as one 
applies it. There are so many very specialized and unique enzymes being 
isolated and studied these days that it becomes more of a problem each 
year to determine which of the enzymes are generally significant. An en- 
zyme which at first sight might seem esoteric, if for no other reason than 
its gargantuan name, implying a specificity of catalysis incommensurate 
with anything but a very limited role in metabolism, may well be of great 
importance in a particular pathway, a pathway perhaps as yet undiscover- 
ed. Every enzyme is of some importance to some organism or tissue, or it 
would not be there. And we often take a limited viewpoint; one of the 
numerous enzymes in the pathway of steroid biosynthesis is recognized as 
important in cholesterol or adrenal corticoid formation, but it may be equal- 
ly important to some microorganism in producing steroids which function 



XVIU INTRODUCTION 

in their membranes, the inhibition of the formation of which could lead 
to a suppression of growth. In view of the past history of science, anyone 
is presumptuous to claim they can distinguish what is important from what 
is not — we have to do this much of the time, of course, but we should 
realize we are presumptuous. There are probably some errors in the tables, 
since it is often difficult to determine exactly the conditions used; one is 
sometimes referred to a previous report, but cannot be certain that all the 
conditions have been maintained throughout the work. One must often 
guess a parameter from other work the investigators have done, and some- 
times calculate results from heterogeneous data. There has been a good 
deal of calculation, and recalculation, and averaging, and I take full re- 
sponsibility for anything right or wrong I may have done. A number of 
curves have been replotted or data represented in a way that differs from 
that of the original investigator, and I fully realize that this usuallj' results 
in nothing but animosity. 



CONVENTIONS 



The naming of enzymes is not an easy task. On the one hand, there are the more 
trivial names with their occasional confusions — on the other, there are the official 
names in the " Report of the Commission on Enzymes " (1961) which are reasonably 
precise but often unwieldy. I have usually chosen the former because I feel most 
readers will recognize these more readily, but frequently I have taken an interme- 
diate course which probably will not please anyone. It is much more accurate to 
write NADH:menadione oxidoreductase than to use the designation NADH oxidase 
or NADH dehydrogenase, since the former name indicates the substrate and acceptor 
used. In addition it is cumbersome to use D-xylulose-5-phosphate D-glyceraldehyde- 
3-phosphate-lyase (phosphate-acetylating) instead of phosphoketolase, yet there is no 
doubt that this longer terra accurately describes the enzyme. There are also prefer- 
ences in nomenclature, for various reasons. I never cared much for the term invertase; 
I prefer to call it [ii-fructofuranosidase, although it is clumsier, but not as much so 
as P-D-fructofuranoside fructohydrolase. In other instances the older and shorter 
names are more pleasing to me and I imagine to others. I have tried to use enzyme 
names which, at least, can be found in the index of the " Report of the Commission 
on Enzymes," and some cross referencing of names has been included in the index. 
There are certain instances of inconsistency which I do not particularly regret. 

As in the first volume, concentrations have been given as millimolar (m.M) except 
when designated otherwise, and in other matters the conventions given there have 
been retained. 



SYMBOLS 



A 


absorbance 


DQ 


ADP 


adenosinediphosphate 


DQH, 


AMP 


adenosinemonophosphate 


Eo 


ATP 


adenosinetriphosphate 




9,10-AQ 


9,10-anthraquinone 


Eo' 


BAL 


dimercaprol 




ChE 


cholinesterase 


ED, 


CoA 


coenzyme A 




6-DFG 


6-deoxy-6-fluoro-D-glucose 


EDTA 


2-DG 


2-deoxy-D-glucose 


EI 


DNA 


deoxyribonucleic acid 


EM 


DNP 


2,4-dinitrophenol 





duroquinone 
durohydroquinone 
standard oxidation-reduc- 
tion potential (pH = 0) 
oxidation-reduction poten- 
tial at specified pH (usually 7) 
effective dose or concentra- 
tion for X per cent 
ethylenediaminetetraacetate 
enzyme-inhibitor complex 
Embden-Meyerhof (path- 
way) 



XX 


SYMBOLS 




EP 


enzyme-product complex 


NEM 


iV-ethylmaleimide 


Epi 


epinephrine 


1,2-NQ 


1 ,2-naphthoquinone 


ES 


enzyme-substrate complex 


1,4-NQ 


1 ,4-naphthoquinone 


FAD 


flavin-adenine dinucleotide 


pl 


- log (I) 


FDP 


fructose- 1 ,6-diphosphate 


p-MB 


p-mercuribenzoate ion 


FMN 


flavin mononucleotide 


p-MPS 


j9-mercuriphenylsulfonate 


GSH 


reduced glutathione 




ion 


GSSG 


oxidized glutathione 


P-Q 


2)-benzoquinone 


i 


fractional inhibition 


P-QH, 


p-benzohydroquinone 


if 


final fractional inhibition 




(hydroquinone) 


lA 


iodoacetate 


pS 


- log (S) 


lAM 


iodoacetamide 


P-XQ 


p-xyloquinone 


IC 


intracutaneous 


9,10-PAQ 


9, 10-phenanthraquinone 


IM 


intramuscular 


3-PGDH 


3 - phosphogly ceraldehyde 


IMP 


inosinemonophosphate 




dehydrogenase 


IP 


intraperitoneal 


PM 


phenylmercuric ion 


ISBZ 


o-iodosobenzoate 


Pyr 


pyruvate 


IV 


intravenous 


SC 


subcutaneous 


Ka 


ionization constant 


SH 


sulfhydryl 


Ki 


inhibitor constant 


S/M 


slice/medium ratio 


Km 


Michaelis constant 


s— s 


disulfide 


Ks 


substrate constant 


TD 


tolerated dose or concentra- 


LD. 


lethal dose or concentration 




tion 




for X per cent 


T/M 


tissue/medium ratio 


MD 


menadione 


TQ 


toluquinone 


MHD2 


menadiol 


TQH3 


toluhydroquinone 


MLD 


minimal lethal dose 


V 


rate 


MM 


methylmercuric ion 


Vm 


maximal rate 


NAD 


nicotinamide-adenine 


P-XQ 


p-xyloquinone 




dinucleotide 


£ 


molar extinction coefficient 


NADP 


nicotinamide-adenine 


(p-AsO 


phenylarsenoxide 




dinucleotide phosphate 


99-ASO2 


phenylarsonic acid 



Enzyme 
and Metabolic 
Inhibitors 

Volume II 

Malonate, Analogs, Dehydroacetate, 

Suljhydryl Reagents, o-Iodosobenzoate, Mercurials 



CHAPTER 1 

MALONATE 



Malonate is one of the most interesting, specific, useful, and well-known 
enzyme inhibitors, and for these reasons will be discussed in detail in order 
to illustrate some of the general principles delineated in the first volume. 
It will be valuable perhaps to take up one inhibitor to the degree necessary 
to consider many of the various problems in the application of these prin- 
ciples, and to examine the pitfalls that may appear even in the use of an 
inhibitor that in several ways approaches what one would want ideally. 
Much of what will be said concerning malonate may be applied to the other 
inhibitors. This discussion will emphasize the often overlooked fact that 
the effect of relatively simple inhibitors in cells may constitute a very 
complex problem and that their use in elucidating metabolic relationships in 
tissues or whole organisms should not be undertaken lightly. It is a simple 
matter to apply an inhibitor such as malonate but it is often very difficult 
to use it properly and to interpret the results accurately. The treatment 
of malonate will, furthermore, provide a foundation for the more general 
discussion of competitive inhibitions produced by analogs in the following 
chapter. 

EARLY HISTORICAL DEVELOPMENT 

The first report of the use of malonate in a biological system was made 
by Heymans (1889) to the Physiological Society in Berlin. The toxicity 
of oxalate to animals had been known for many years and Heymans be- 
lieved that an investigation of the higher homologs of the dicarboxylate 
series might be interesting. Although sodium oxalate was quite poisonous 
when injected into the frog dorsal lymph sac, the sodium salts of malonate, 
succinate, and glutarate were essentially without effect, sodium malonate 
being nonlethal at a dose as high as about 8 g/kg. However, Pohl (1896), 
working in Prague, found that the urinary excretion of oxalacetate was 
increased by administering malonate to dogs and, furthermore, that only 
a small portion of the malonate given could be recovered in the urine, 
indicating that the dog can metabolize malonate. The first experiments 
showing the metabolic inhibitory action of malonate were done by Thun- 



2 1. MALONATE 

berg (1909) in Lund. He had observed the inhibition of minced frog muscle 
respiration by oxalate and decided to study the higher homologs. Thus the 
inhibitory action of malonate on muscle respiration was demonstrated, 
whereas succinate instead stimulated the oxygen uptake. Apparently this 
observation went unnoticed and the inhibitory activity had to be rediscov- 
ered later. Rose (1924) at Illinois showed that malonate exhibits no nephro- 
toxic action, as does glutarate, when given orally to rabbits, although he 
later (Corley and Rose, 1926) found that the methyl and ethyl derivatives 
depress renal function. At about the same time, Momose (1925) in Japan, 
continuing the work of Pohl, observed that malonate when perfused through 
dog liver, gives rise to acetoacetate, acetone, and aldol. He postulated that 
these substances arise from malonate after decarboxylation to acetate, but 
it is more likely from our present knowledge that malonate gives rise to 
these substances by a disturbance of the metabolism. During the next 
few years evidence was accumulated that malonate can arise from normal 
tissue metabolism — occurring in alfalfa (Turner and Hartman, 1925) 
and wheat (Nelson and Hasselbring, 1931), and appearing during citrate 
fermentation in the mold Aspergillus (Challenger et al., 1927) — and be 
metabolized by certain microorganisms, such as Escherichia coli (Grey, 
1924). 

Our present concepts of the inhibitory action of malonate, however, 
arose from the work on bacterial dehydrogenations by Quastel and Whetham 
(1925) at Cambridge. They tested the abilities of various dicarboxylic acids 
to reduce methylene blue in suspensions of E. coli and found that only suc- 
cinate is active. Malonate inhibited this reduction by succinate. As stated 
in their own words, "Oxalic, glutaric, and adipic acids (when mixed with 
succinic acid) do not retard the reduction due to the succinic acid, but 
malonic acid has a definite retarding effect. It is difficult to explain the 
anomalous behaviour of malonic acid, but there is no doubt as to the reality 
of the effect." They found the methylene blue reduction time with succinate 
bo be tripled in the presence of 77 mM malonate. Quastel and Wooldridge 
(1928) extended this work to show that the action on succinate oxidation is 
rather specific in that malonate does not appreciably inhibit the oxidation 
of several other substrates by E. coli. But in addition they demonstrated 
that increasing succinate concentrations would counteract the malonate 
inhibition, leading them to suggest that both substances are adsorbed to 
the enzyme reversibly, probably competing for the same active site. Final- 
ly, Quastel and Wheatley (1931), now at the Cardiff City Mental Hospital, 
reported that the malonate inhibition of succinate oxidation occurs in 
many bacteria, and in mammalian brain and muscle as well, the enzymes 
from the mammalian tissue being even more sensitive. 

The concept of the competitive inhibition of an enzyme by a substance 
structurally related to the normal substrate was first clearly demonstrated 



CHEMICAL PROPERTIES 6 

and expressed for this inhibition of succinate oxidation by malonate in the 
work of Quastel and Wooldridge, although competition specifically was not 
mentioned. Cook (1930), also at Cambridge, however, stated that a "com- 
petitive" mechanism had been established, presumably referring to the 
work of Quastel inasmuch as Cook performed no experiments indicating 
a competitive relationship. The competitive nature of the malonate inhibi- 
tion has been substantiated many times and placed on a quantitative basis, 
so that malonate has come to be recognized as the classical example of 
inhibition by a pvirely competitive mechanism. The development and ap- 
plications of this concept will be discussed in more detail in Chapter 2. 
For 20 years malonate was the only available specific inhibitor of succinate 
dehydrogenase, and later of the tricarboxylic acid cycle, and actually played 
an important role in the elucidation of the cycle sequence. Other cycle inhib- 
itors have been described recently, but no other inhibitors of the succinate 
oxidation step as specific and useful as malonate have been found. 

CHEMICAL PROPERTIES 

Malonic acid and its salts when obtained commercially are often not 
sufficiently pure for accurate work and it has been the practice in our lab- 
oratory to recrystallize all material. Malonic acid may be recrystallized 
from ethyl acetate and benzene (Adell, 1940), or ether and benzene con- 
taining 5% light petroleum (Vogel, 1929), or simply from a hot concentrated 
benzene solution by cooling to 50-10°. The sodium and potassium salts may 
be dissolved in small amounts of warm water and precipitated by the ad- 
dition of ethanol, as is commonly done with other dicarboxylate salts 
(Potter and Schneider, 1942), or, with somewhat less yield, may be crystal- 
lized by cooling hot concentrated aqueous solutions. In all cases we have 
decolorized with activated charcoal in the solutions before recrystallization 
and have washed the products with ether preparatory to drying. It should 
be emphasized that the choice of the sodium or the potassium salt will 
depend on whether the preparation to be tested is cellular or subcellular. 

Stability 

Malonic acid and its salts are quite stable and chemically unreactive. 
Decarboxylation to acetate proceeds very slowly under ordinary conditions. 
Aqueous solutions of sodium malonate heated to 125° for 48 hr show no 
perceptible decomposition (Fairclough, 1938), and the half-life of sodium 
hydrogen malonate in solutions 5-50 mM is at 80° around 40 days, cal- 
culated from the rate constant for decarboxylation (Hall, 1949). The free 
energy change for the reaction 

Malonate -> acetate + CO, 



4 1. MALONATE 

is approximately —7 kcal/mole but the activation energy is 27.9 kcal/mole 
(Gelles, 1956). Thus at physiological temperatures one may consider mal- 
onate as completely stable. Nevertheless, there are enzyme systems which 
catalyze the decarboxylation (see page 227) and one should be reasonably 
certain in the use of malonate that it is stable in the system investigated, 
since the formation of acetate might well confuse the results. 

Molecular Structure 

In crystals of malonic acid the molecules are arranged in zigzag chains 
with the carboxyl groups linked through two hydrogen bonds (Goedkoop 
and MacGillavry, 1957). The following bond parameters were observed 
(the two values refer to the two carboxyl groups, since the molecule in the 
crystal is apparently not symmetrical): C — C — C angle = 110^; C — C 
distance = 1.54, 1.52; C— distance = 1.29, 1.31; C=0 distance = 1.24, 
1.22; 0— C— angle = 128^, 128°; and H-bond distance = 2.68-2.71. 
The malonate ion in solution would probably deviate somewhat from this 
configuration but not a great deal inasmuch as malonate is fixed in a rather 
rigid structure, because the C — C — C angle is determined by the electronic 
tetrahedral orbitals and can be distorted only with difficulty. The carbox- 
ylate groups can rotate around the C — C axis but they presumably ster- 
ically interfere with each other when both lie in the plane of the molecule, 
since the centers of the oxygen atoms would be 2.2 A apart and the van 
der Waals' radius of the oxygen atom is 1.4 A (Goedkoop and MacGillavry, 
1957). In malonic acid crystals one carboxyl seems to be in the molecular 
plane and the other is at right angles; in the malonate ion it may well be 
that neither is in the C — C — C plane. However, another factor must be 
considered; it is possible that in solution there is intramolecular hydrogen 
bonding (Gelles, 1956), at least for the hydrogen malonate ion. When the 
carboxyl group ionizes, the equivalence of the structures: 

R-cf R_c--° 

allows a greater resonance than in the unionized state (equivalent to an 
extra 8 kcal/mole energy), and this high resonance would indicate an inter- 
mediate structure in which the center of negative charge lies midway be- 
tween the two oxygen atoms. Although keto-enol tautomerism occurs 
(Hofling et al., 1952) in the esters of malonic acid: 

O OH 

-CH^-C-OEt ^ -CH=C-OEt 

it is probably not significant in the malonate ion because it would reduce the 
electronic resonance. 



CHEMICAL PROPERTIES 5 

The distance between the two centers of negative charge in malonate 
is of importance in the binding to succinate dehydrogenase. Calculation of 
this distance (using the following values: C — C — C angle = 111.7°; — C — 
angle = 125.8°; C— C distance = 1.544, and C— distance = 1.273) 
leads to a value of 3.28 A. Intercharge distances for various dicarboxylates 
and other compounds known to inhibit succinate dehydrogenase are given 
in Table 1-1, and these values will be of interest in comparisons of inhibi- 
tory activity. Dicarboxylic acids with more than one methylene group are 
flexible and the intercharge distances may vary between the limits of great- 
est bending and extension of the molecules. As the chain length increases 
there will be greater tendency for the intercharge distance in the ions to 
be less than that of the maximal extension, since the electrostatic repulsion 
will decrease. The mean statistical intercharge distance in the succinate ion 
will probably be closer to 4.75 A than the contracted distance of 3.81 A. 
Indeed, it may be calculated that it would require at least 2.3 kcal/mole 
to bring succinate from the extended to the contracted configuration, 
using the dielectric constant obtained from D = 6d — 7 (Eq. 1-6-72)*; 
since the dielectric constant is probably less due to the hydrocarbon groups 
between the charges, this would be a minimal energy value. The mean 
intercharge distance in the succinate ion may be estimated as not less than 
4.20 A (Gane and Ingold, 1931; Eyring, 1932; Westheimer and Shookhoff, 
1939) and possibly closer to 4.75 A. The glutarate intercharge distance is 
probably around 5.2 A. These considerations are of importance in comparing 
the interactions of these substances with the active center of succinate 
dehydrogenase. It must be remembered that the bound ions are undoubt- 
edly held in a configuration different from the statistical mean in solution. 
The flexibility of the higher homologs allows them to adjust to a specified 
configuration, but at the expense of the energy and entropy changes neces- 
sary to bring them from their free configurations. 

Acidic Ionization 

The ionizations of malonic and succinic acids are important in the inter- 
actions with succinate dehydrogenase and with regard to the penetration of 
these substances into cells. The pKJs for the simple dicarboxylic acids, in- 
cluding various derivatives of succinate and malonate, and related succinic 
dehydrogenase inhibitors, are given in Table 1-2. The dissociation constants 
change slightly with ionic strength; for malonic acid, dpK^ Ids = — 0.32, 
and dpK^ jds = — 0.98, for ionic strengths around 0.15, where s is the 

* In cross references of this type, Eq. 1-6-72, the roman number indicates the volume 
of this treatise in which the equation, table, or figure (as the case may be) may be 
found, the first arabic number indicates the chapter number, and the second arabic 
jiumber the equation, table, or figure number. 



1. MALONATE 



Table 1-1 
Intekchakge Distances in Dianions op Interest " 



Compound 



Oxalate 
Malonate 



Succinate 



Glutarate 



(c) 



Adipate 



Hydroxy malonate 
(tartronate) 



Fumarate 

Maleate 

Acetylene-dicar boxy late 
o-Phthalate 

Isophthalate 
Terephthalate 



Structure 



'ooc-coo 

'ooc coo" 

\ / 

CHj 
"ooc -CHj 



"OOC 



CH,- COO" 
COO" 



\ / 

CH2 CH2 



"OOC— CHj 



CH2 CH2 

COO" 



"OOC-CHj 

"ooc coo" 

/ \ 

CH9 CHa 

^ / 

CHj 
"OOC-CHj 

CHj— CH. 

\ 



COO 



CHj- COO" 



"OOC COO 

\ / 

CH 

I 
OH 

"OOC H 

\ / 

c=c 

/ \ - 

H COO 

"OOC COO" 

\ / 
c = c 

/ \ 

H H 

"OOC-CEC-COO" 

COO" 



^^^^-^ COO- 




COO" 



Distance (A) 



2.42 
3.28 

4.75 

3.01 

5.83 

4.84 

1.70 

6.87 
3.28 

4.87 

3.66 
5.16 
3.55 

5.84 
7.33 



COO 



CHEMICAL PROPERTIES 



Table 1-1 (continued) 



Compound 



Structure 



Distance (A) 



Cyclobutane-dicar boxy late 



Cyclopentane-dicar boxy late 




COO 



coo 

COO' 



Cyclohexane-dicarboxylate 



COO' 
COO' 



COO 



/3-Sulfopropionate < 


(a) 
(b) 


OOC-CH, 
\ 
CHj-SO. 

"OOC SO,' 

CHj— CHj 


Methanedisulfonate 
(methionate) 




'O3S SO3' 
CH, 


1,2- Ethanedisulf onate < 


(a) 
(b) 


'OjS-CHj 

CHj— SO3' 

'0,8 SO,' 
' \ / ' 
CHj— CHj 

/::?X^COO" 


-Sulfobenzoate 




^SO,' 


Arsonoacetate 




'OOC ASO3H' 
CH, 



^-Phosphonopropionate < 



(a) 



(b) 



'OOC-CHj 



"OOC 



CH,— PO; 

po: 



\ / 

CH2 CH2 



3.39 



cis- 


3.52 


trans- 


4.91 


cis - 


3.05 


trans - 


4.51 




5.05 




3.13 




3.40 




5.38 




3.25 



3.72 

4.05 
5.09 
3.14 



The distances were calculated on the basis of bond lengths and angles given In Tables 1-6-12 and 1-6-13 
except where direct measurements were available. It was assumed that the center of negative charge lies 
midway between the resonating oxygen atoms. For o-phthalate and cyclohexane-dicarboxylate a 3° distortion 
of the bond angles due to electrostatic repulsion was assumed. For cyclopentane-dicarboxylate a further 
5° widening was estimated from the bending of the ring angles. The value for cyclobutane-dicarboxylate 
is only approximate and is based on a 7° distortion of the bond angle in comparison to malonate. It should 
be pointed out that the values in this table are smaller than generally used; one reason is that Intercarboxy- 
late distances have usually been based on the distances between associating or dissociating protons , ra- 
ther than between centers of negative charge. 

ionic strength. Since AH for the dissociation of weak acids is quite small, 
the constants do not change much with temperature; d^K^JdT = 0.0031, 
and d])Kg jdT = 0.0038, approximately. Three species will be present in 
any solution of malonate — HOOC— CHg— COOH, HOOC— CHg— C00-, 
and -OOC — CHg — COO" — the relative concentrations being determined 



1. MALONATE 

Table 1-2 
Ionization Constants of Dicarboxylic Acids " 



Acid 



pKa, 



P^., 



Oxalic 

Malonic 

Succinic 

Glutaric 

Adipic 

Pimelic 

Methylmalonic 

Ethylmalonic 

Dimethylmalonic 

Diethylmalonic 

n.-Propylmalonic 

i«o-Propylmalonic 

Methylethylmalonic 

Di-w-propylraalonic 

Phenylmalonic 

Hydroxymalonic (tartronic) 

Maleic 

Methylmaleic (citraconic) 

Fumaric 

Methylfumaric (mesaconic) 

Malic 

Tartaric 

2,2 '-Dimethylsuccinic 

2,2 '-Diethylsuccinic 

Tetramethylsuccinic 

Methylenesuccinic (itaconic) 

Cyclopropane- 1 , 1 -dicarboxylic 
Cyclobutane- 1 , 1 -dicarboxylic 
Cyclopentane- 1 , 1 -dicarboxylic 
Cyclohexane- 1 , 1 -dicarboxylic 
<raw5-Cyclopropanedicarboxylic 
cis-Cyclopropanedicarboxylic 
frarw-Cyclopentanedicarboxylic 
<raws-Cyclohexanedicarboxylic 

Phthalic 
Isophthalic 



1.09 


3.79 


2.58 


5.17 


3.95 


5.16 


4.07 


4.93 


4.17 


4.95 


4.23 


4.98 


2.86 


5.24 


2.79 


5.34 


2.97 


5.59 


2.04 


6.90 


2.83 


5.38 


2.78 


5.46 


2.71 


6.07 


1.92 


7.13 


2.43 


4.68 


2.93 


— 


1.67 


5.75 


2.23 


5.89 


2.85 


4.00 


2.93 


4.82 


3.21 


4.62 


2.83 


3.88 


3.77 


5.82 


3.34 


6.22 


3.33 


6.90 


3.67 


5.19 


1.77 


7.37 


3.08 


5.82 


3.18 


6.02 


3.40 


6.05 


3.49 


4.75 


3.16 


6.09 


— 


4.22 


— 


4.45 


2.95 


5.23 


2.15 


4.49 



" These values have been obtained from a variety of sources and have been corrected 
to a temperature of 37° and an ionic strength of 0.15 so as to be applicable to physio- 
logical conditions. These corrections were obtained from studies on the variations of 
dicarboxylic ionization constants with temperature and ionic strength (e.g. AdeU, 
1940). They are not absolutely correct but probably allow a closer approximation 
than the values in the literature, which are usually for 25° and extrapolated to zero 
ionic strength. 



CHEMICAL PROPERTIES 



by the pH (see Eqs. 1-14-6 to 1-14-8). The variations of these species with 
pH are shown in Fig. 1-1, and actual concentrations are given in Table 1-3 
for malonate and succinate at a total concentration of 10 mM. In the 
usual range of physiological pH, over 95% of these acids are in the form 
of the completely dissociated doubly-charged anion; this is the active form 
for the inhibition of succinate dehydrogenase. However, it is the concen- 
trations of the other forms which are important in the rates and degrees 
of penetration into cells, and these change appreciably with pH (e. g., 
from pH 7.4 to 6.8 there is a 4-fold increase in the singly dissociated form 
and a 16-fold increase in the undissociated form). 



"x 






/^ 


,cooh\ 

^COOH \ 


^< 


_ cooA 

' COOH \ 


/ COO' 

/ ^coo- 


; ^ 


I 




' Malonate 


J 


\ 




V 



pH 

Fig. 1-1. Concentrations of the various ionic species of malonic 

acid at different values of the pH expressed as per cent as of 

the total concentration. 



Das and Ives (1961) on the basis of thermodynamic evidence suggested 
that an internal symmetrical hydrogen bond occurs in H-malonate~, and 
that this would affect to some extent the piiC^ values and reduce hydration. 
However, Lloyd and Prince (1961) examined the infrared spectra of malonic 
acid and its ions in DgO, compared the data with those obtained with 
fumaric acid (in which no hydrogen bonds could occur), and concluded that 
if hydrogen bonding exists in H-malonate~, the bond is very weak and not 
symmetrical. Eberson and Wadsd (1963) determined the ionization enthal- 
pies in water and ethanol, and also concluded that hydrogen bonding is 
not important in the dicarboxylates when zlpiiC^ is less than 4. It would 
thus appear that intramolecular hydrogen bonding is not a significant 
factor in stabilizing the H-malonate~ ion. 



10 



1. MALONATE 



> 

ft 

E-t 



< ^ 



d d 



o o 



-H 05 -< 



^ -^ oo 



ft 



CHEMICAL PROPERTIES 11 

Metabolic studies of the substituted malonates and malonic esters wiU be 
taken up after the actions of malonate have been discussed (see page 235). 
It is interesting to note (Table 1-2) that the pK^ 's of the substituted 
malonic acids are generally higher than for malonic acid itself. This is 
mainly the result of the reduction of the dielectric constant of the region 
between the interacting carboxyl groups, and is particularly evident for 
the disubstituted ethyl and w-propyl derivatives. This increase in pK^ 
should facilitate penetration of these compounds into cells; in addition they 
are more lipid-soluble, which will also favor penetration. The esters are not 
active inhibitors of succinate dehydrogenase, at least by the same mechanism 
as malonate, but have been used because of their ability to enter cells and 
tissues readily, some hydrolysis to active malonate within the cells being 
assumed. The presence of two keto groups on either side of the methylene 
group makes this latter group more reactive and, indeed, imparts some 
acidic character to it, malonic diethyl ester having a p^^ of approximately 
5 X 10~^^ (Pearson and Mills, 1950). The rate of ionization is, however, 
quite slow (A; = 1.8 X 10-^ min-^). 

Chelation with Metal Cations 

Malonate is able to form fairly stable complexes with various cations 
normally present in media used in metabolic studies. The importance of 
this in malonate inhibition will be discussed later (see page 66), and in the 
present section we shall investigate the magnitudes of the effects expected. 
These complexes are chelates with a six-membered ring structure, accounting 

o or o. /O 

for the relatively high stability compared to complexes with the monocar- 
boxylates. The chelate dissociation constant is given by: 

(M++) (A=) 

K=- — - (1-1) 

(MA) 

where A= represents the anion of any dicarboxylic acid. The values for the 
piii's of some of the more important chelates are given in Table 1-4. These 
constants are dependent on temperature and ionic strength. The pK for 
Mg-malonate is related to the temperature at zero ionic strength in the 
following way: pZ = 2.92 - 0.008 (35 - «oC) (Evans and Monk, 1952). 
Thus the pii's at 37° are approximately 0.1 unit higher than at 25°, the 
temperature at which the constants are most commonly determined. It 
was calculated from the data on the complexes of malonate with Mg++, 
Ca++, Ba++, and Zn++ that the pK at an ionic strength of 0.15 is about 



12 



1. MALONATE 



« >S 






O 



-M — I -- 



O cc -^ 
CI -^ — 1 



(N -H rt 



-H (M P- 





c 


2 


c5 


o 


o 


+s 


P< 


cS 


3 


3 


T3 


^ 


02 


O 


<Jl 



P^ 



^ CS 



a 


c 

CO 


>> 

S 




^ 


ki 


■^ 


'-^S 


cS 


43 


fcT 


3 


O 


i 


^ 
V 


.S 


V 


s 


is 


E3 


3 


CO 


o 


O 


'53 

CD 


W 


^ 


o 


-1-i 




§ 


ft 


C 


CO 


M 




g 


S 


a 




c3 


•r" 


1 


O 

o 

CO 

a; 


O 
S 




o 


^ 


_C 


to 


o3 






3 






to 


o 


to 


to 


c 




c3 


C 

o 


_o 


'-P 


73 




'-3 


^ 


01 


+3 


eS 


4^ 




^ 


s 


S 
« 


OJ 


s 


V. 


o 


iH 


o 


e 
o 

o 


s 


o 


o 


o 


1 


a 


c 


ft 

>. 


_bi 


J2 


5b 


'S 


CO 


[o 


"c 


c 


eS 


'm 


o 


o 


C 
c3 




CO 

73 


CO 
1 


"s 


P 


<D 


^ 
S 


s 




4^ 


,::— ' 


2 


C 


t4H 


."'^ 


JS 


o 


o 


II 


-»^ 


o 




^ 


o 






+ 
+ 






§ 


3 

03 


to 
3 


aj 


■ ' 


O 


3 


le 


II 


^ 


^ 


3 




C 
> 


+3 
S 

a 


>. 


0; 


'So 


aj 


^ 


J2 


CO 


j3 


'T3 


o 


-t> 


« 


-ti 


_s 


Cm 


C 
O 






o 


CO 


> 


to 

c 


-C 


O 


S) 


_o 


03 


CO 


-C 


'-5 




»o 


-tJ 


S 


C 

CO 

C 
O 


d 
o 


a 

73 


a) 

C 

o 
o 




■t^ 


C 


(h 


C 


M 


a 


,o 


_o 


C 




c« 


"-3 


2 


m 


w 


2 




(0 


a> 


'o 


c3 


CO 


2 


'S 


J5 


a; 


73 


.2 


o 


> 


s 




j; 


'-3 


c 


-p 


cS 




eS 


bl) 


-i-i 


H 


C 




e 


TS 


?> 


13 




C 




3 




cS 


CO 


a* 



CHEMICAL PKOPERTIES 13 

0.74 unit lower than at an ionic strength of zero. The free energy of forma- 
tion of Mg-malonate is — 3.90 kcal/mole, while .1//° = 3.2 kcal/mole, 
and J>S'o = 23.9 cal/degree (Evans and Monk, 1952; Chaberek and Martell, 
1959, p. 139). The entropy term is large and probably results mainly from 
displacement of water from the charged groups. 

The importance of this chelation in inhibition work lies in the reductions 
in the concentrations of the free ions it may bring about, both the metal 
ions and malonate. The decrease in metal ion concentration can easily alter 
enzyme activity or cellular function and such changes are apt to be attrib- 
uted to a direct action of malonate. Conversely, the malonate concentra- 
tion may be reduced appreciably. Examples of mutual concentration re- 
duction are given in Table 1-5. The concentrations of Mg++ chosen approxi- 
mate those in Tyrode solution (0.11 niM), Krebs-Ringer phosphate medium 
(1.18 mM), the usual media for mitochondria (5 mM), and sea water (53.6 
TCiM), while the concentrations of Ca"'""'" correspond to Krebs-Ringer phos- 
phate medium (2.54 mM) and sea water (10.24 mM). It may be noted that 
very significant reductions in Mg^^ and Ca++ can occur with concentrations 
of malonate commonly used; e.g., malonate at 10 mM will reduce the Mg++ 
49% and the Ca++ 23% in Krebs-Ringer medium, and higher concentra- 
tions may almost deplete these ions from the solution. When the concen- 
trations of these cations are high, as in sea water, the effective malonate 
concentration may be reduced markedly; e. g. malonate added to sea water 
at a total concentration of 10 mM will result in a 1.5 mM solution of free 
malonate ion. Such phenomena have usually been ignored or forgotten 
despite their possibly large magnitudes. One way of determining the impor- 
tance of cation reduction in malonate studies is to calculate the reduction 
to be expected in the medium used and at the malonate concentration, and 
then to test the effects of lowering the cation concentration to this extent 
(Rice and Berman, 1961). Other metal cations may be reduced to a greater 
extent than Mg++ and Ca++. Media initially 1 mM in Co++, Mn++, or Cu++ 
will in the presence of 5 mM malonate contain these ions at concentrations 
of 0.65 mM, 0.34 mM, and 0.0052 mM, respectively. If such metal ions are 
normally bound to enzymes, the degree of removal from the enzyme will, 
of course, depend on the relative affinities of the metal ion for the enzyme 
and malonate. There is also the possibility that malonate may chelate with 
metal ions combined with the enzyme, inactivating them for their catalytic 
role. It should also be remembered that similar phenomena may occur with 
succinate, and the inhibition kinetics may be distorted when malonate is 
used due to the differential reduction in the concentrations of these anions. 

Detection and Determination of Malonate 

Methods have been developed for the separation and identification of 
organic acids from animal and plant tissues. Earlier determinations involved 



14 



1. MALONATE 



r^ q 



K 









,— V 






















c 


lO 


o 


CO 


CO 














t^ 


CD 


00 


LCl 


(M 


c<l 


o 




+* 




o 


o 


r- 


00 


o: 


in 




LO 




+ 

O 


g 




d 


d 


CO 


t^ 


d 




C5 




Vm 






















O 


o 


'^ 


lO 


o 


CO 


CO 


CO 


Oi 


o 








+ 


t- 


CD 


GO 


LO 


l-H 




(^ 






+ 


ai 


t- 


00 


OJ 


i-O 


^_ 


lO 




"-3 




O 


05 


05 


00 


r^ 


CD 


■* 


(M 




cS 






















;-i 






















-fj 






















c 












































0) 






















o 






t^ 
















C3 




g 


<M 


CD 


c- 


t- 










O 

o 


^ 


Is 


05 


t^ 


CD 


o 


CO 


LO 


CD 




o 


GO 


CD 


Tt< 


o 


Tt< 


O 




„_ 


g 


o 


d 


Tji 


d 


d 


GO 


oo 




o3 
g 


-* 


^ 












Tt< 


05 




'Si 


lO 










































°3 


C<l 


+ 


CO 


CD 


r^ 


t^ 


i?q 


CI 


o 




o 




+ 


fO 


r-H 


o 


Tt< 


t^ 


00 


o 








cS 


lO 


Tt< 


(M 


05 


LO 


C5 


CD 








O 


c^ 


■M 


Csl 


-' 


-< 


d 


d 




























II 


lO 














g 






c 


CD 


lO 


t- 


o 


'N 










_o 




CD 


t^ 


C5 


'M 


CD 


CD 




^ 


Is 


O 

d 


d 


00 

d 


00 


^ 
^ 


o 

CO 


(M 
t^ 






j;^ 


^^- 












-M 


»c 














































_o 




o 


^_^ 
















'-3 




IC 


+ 

+ 


'M 


CD 


00 


Ci 


(M 


CD 


CD 






h 


Ci 




GO 


00 


"* 


O 


(M 


"S 






^ 


Ol 


C5 


O' 


_h' 


tH 


CO 


t^ 










Tl* 


^ 


-<* 


'* 


CO 


'>^ 
























o 






















o 






"ip 
c 
o 


t^ 


t^ 


'^^ 


CO 








^ 






CC 


t^ 


•* 


GO 


Tt< 


00 


CD 


,o 


+' 




o 


CD 


CD 


t~ 


00 


00 


Tj< 


Qi) 


+ 

SB 


^ 


1 


d 


O 


CO 


c- 


d 


LO 


lO 


£ 


§ 


^ 










'^ 


-* 


c: 


CJ 






















13 


O 


>o 


4^ 


t^ 


l^ 


^M 


CO 


CD 


t^ 


LO 


c2 


o 




+ 


CD 


t^ 


rti 


00 


CO 


t^ 


CD 


E 






Ci 


CD 


CD 


t— 


00 


00 


■* 


'-3 

(-1 




"* 


■^ 


CO 


•M 


-i 


d 


d 
























-iJ 






















!=l 




^„^ 


















0) 

o 

C 

o 




II 


_l 


-* 


t- 


t^ 












C 


CI 




C^ 


O 


•* 


CD 


CS 






^ 


o 


C5 


CD 


in 


CO 




o 




o 


^ 


'^ 


d 


d 


^ 


Ci 


C5 


o 
•^ 


C5 
C5 




"^ 






















_g 


S 










































'Si 


^H 


+ 


,_, 


rH 


r- 


t^ 


CD 


LO 


o 




'S 




+ 


C5 




t- 


o 


CO 


CD 


C5 




o 




M 


o 


C2 


CD 


>o 


CO 




o 








S 


d 


d 


d 


d 


d 


d 


o 








■~" 






















11 


t- 




















c 


00 


o 


o 


■* 


















C5 


t^ 


ri* 


CO 


<— < 


1— t 






^ 


o 


C5 


C5 


C5 


05 


C5 


CI 






o 


C' 


■* 


d 


d 


d 


d 






g 












•* 


c:> 






^_ 


_^ 


t^ 


















d 


+ 


GO 


(^ 


O 


^ 


o 


LO 


t- 






+ 


OS 


o 


CD' 


Tf 


CO 


i-H 


o 








t* 
^ 


o 


o 


O 


o 


o 


o 










d 


d 


d 


d 


d 


d 


d 




Oi 




















1 1 i 
? ^ 1 




o 




lO 


o 


o 


o 


o 














OJ 


lO 


o 




<: s ^ 






















" 



















o 



hI 



INHIBITION OF SUCCINATE DEHYDROGENASE 15 

oxidative titrations with hot acid permanganate or eerie sulfate (Willard 
and Young, 1930; Christensen and Ross, 1941) and such methods have 
been used for the analysis of malonate in the presence of proteins (Ross and 
Green, 1941). Colorimetric tests, for example with tetrahydroquinoline-iV^- 
propenal to form blue- violet compounds (sensitive to 0.01 mg malonic acid) 
(Dieterle and Wenzel, 1944), have been used, and several microcolorimetric 
and spot tests are available (Feigl, 1960). However, the most valuable 
methods are chromatographic. A large variety of solvents and spraying 
agents have been utihzed, and different techniques have been applied: 
ion-exchange resin columns and partition chromatography (Stark et al., 1951; 
Phares et al, 1952; Owens et al, 1953; Shkol'nik, 1954; Reinbothe, 1957), 
strip paper chromatography, both one- and two-dimensional (Buch et al., 
1952; Cheftel et al, 1952; Denison and Phares, 1952; Duperon, 1956; van 
Duuren, 1953; Jermstad and Jensen, 1950; Kalyankar et al, 1952; Ladd 
and Nossal, 1954; Overell. 1952; Smith and Spriestersbach, 1954), and cir- 
cular paper chromatography (Airan and Barnabas, 1953; Airan et al, 1953; 
Barnabas, 1955). The use of chromatographic methods for the determina- 
tion of succinate and malonate in animal tissues is well illustrated in the 
work of Busch and Potter (1952 a, b; Busch et al., 1952). A moderately sen- 
sitive method for estimating 10-100 //g of malonate by applying iodine 
vapor to paper chromatograms was developed by Dittrich (1963). 



INHIBITION OF SUCCINATE DEHYDROGENASE 

The oxidation of succinate in cells is catalyzed by a mitochondrial 
multienzyme system which is structurally organized into units that can be 
isolated as a particulate suspension capable of transferring electrons from 
succinate to oxygen: this complex is known as succinate oxidase. Evidence 
will be presented that the inhibition of this system by malonate is related 
to the binding of the malonate to the most proximal site in the sequence, 
namely, the active center for the attachment and dehydrogenation of succi- 
nate: this component is known as succinate dehydrogenase. It is impossible 
at the present time to define the succinate dehydrogenase accurately, be- 
cause it is assayed with various electron-accepting dyes and the basic mini- 
mal unit has not been completely characterized. The inhibition of succinate 
dehydrogenase by malonate has generally been determined with various 
preparations of succinate oxidase and not with the isolated dehydrogenase, 
and thus it is necessary to discuss briefly the nature of the entire system. 

Properties of Succinate Oxidase 

Knowledge of the components of succinate oxidase and the pathways of 
electron flow has been advanced markedly in the past few years (Mahler, 



16 1. MALONATE 

1956; Singer, et al., 1957; Green and Crane, 1958; Singer and Lara, 1957; 
Green and Flieseher 1960; Green, 1960; Redfearn, 1960). The particulate elec- 
tron transport particle (ETP) of Green, obtained from heart mitochondria, is 
a succinate oxidase preparation and has been shown to contain the following 
components (per molecular weight of about 5 x 10^): flavin dinucleotide 2, 
nonheme iron 64, heme (equal amounts of cytochrome a, cytochrome 
b, and cytochromes c + Cj) 6, copper 8, ubiquinone (coenzyme Q) 10, and 
lipid 34,5%. In addition, there are the several proteins with which these 
substances are bound. Such preparations also oxidize NADH, ascorbate, 
p-phenylenediamine, and hydroquinones, these substrates supplying elec- 
trons at various sites in the electron transport chain. The use of various 
electron donors and acceptors, the application of specific inhibitors, and the 
fragmentation of the succinate oxidase complex have led to several postu- 
lates of the pathways of electron flow. One difficulty is the probable dif- 
ference in pathways between mitochondrial phosphorylating systems and 
submitochondrial nonphosphorylating preparations. A second difficulty is 
the possibility of alternate pathways of electron flow rather than a single lin- 
ear sequence. Scheme 1-2, (Green, 1960; Redfearn, 1960) might be assumed 
provisionally: 

UQ UQ 

cs . i'«/ ^ ^^ ^ ^^f,| 

Succinate »- ASF ^ -• NADH 

iFe) \^ /^ I >\ ^ lFe\ 



(1-2) 



Fe represents nonheme iron, f, is the flavoprotein associated with succinate 
dehydrogenase, f^ is the flavoprotein associated with NADH dehydrogenase, 
UQ is ubiquinone (coenzyme Q), ASF is the antimycin-sensitive factor, and 
the other symbols indicate the usual cytochromes. It is possible that cyto- 
chrome b and ubiquinone are common to the two pathways from succinate 
and NADH, but Green believes the evidence points to fusion of the chains 



INHIBITION OF SUCCINATE DEHYDROGENASE 17 

only at the antimycin-sensitive factor. It is also possible that ubiquinone 
and cytochrome b are on a linear pathway rather than on alternate path- 
ways. In nonphosphorylating systems, the role of cytochrome b is debat- 
able. The sites of oxidation of p-phenylenediamine and ascorbate are distal 
to ASF. It should also be pointed out that succinate oxidase from other 
sources may have different ratios of components and somewhat different 
pathways. Thus the oxidation of succinate and NADH by the electron 
transport system of Azotobacter is not sensitive to antimycin (Bruemmer et 
al., 1957). Two recent observations have extended our knowledge of suc- 
cinate oxidase and related pathways. Azzone and Ernster (1961) have 
demonstrated an ATP requirement for mitochondrial phosphorylating 
succinate oxidation and propose reactions such as the following: 

Succinate + A -l ATP -> fumarate + AH ~ P + ADP 
AH ~ P + B + ADP -> A + BH2 + ATP 

Succinate + B -> BHj + fumarate 

A is possibly a flavoprotein, a quinone, or some other component of succinate 
dehydrogenase, and B may be cytochrome b. In nonphosphorylating systems 
the electrons may reach B more directly. An alternate pathway for AH -^ P 
is the reduction of NAD: 

AH ~ P + NAD -> A + NADH + P^ 

Succinate has been shown to reduce NAD in mitochondria by Chance and 
Hollunger (1961 a, b, c) and this is dependent on ATP. Since cndogen^asly 
formed succinate is more effective than added succinate, it is possible that 
this reaction may involve succinyl-CoA. The reduction of NAD passes 
through antimycin-sensitive and Amytal-sensitive links and seems to involve 
a third flavoprotein component. These findings are not only important with 
respect to the behavior of succinate oxidase, but also have possible bearing 
on the responses to malonate in mitochondrial and cellular systems. 

The succinate oxidase particles have been fragmented in various ways to 
yield smaller particles or soluble preparations with different compositions 
and properties. One such preparation is the succinate dehydrogenase complex 
(SDC) from heart mitochondria or ETP, which perhaps represents that frac- 
tion of the complex up to cytochrome c, since it oxidizes both succinate and 
NADH and possesses an antimycin-sensitive step. Of more interest for 
malonate inhibition are the several forms of soluble succinate dehydrogenase 
that have been prepared. The purest contain no heme or lipid; four atoms 
of tightly bound nonheme iron and one flavin occur in a molecule (assuming 
a molecular weight of around 200,000). The flavin is apparently not ribofla- 
vin but occurs in a dinucleotide form covalently attached to peptide chains 
of the apoenzyme (Kearney, 1960). There is spectral evidence that both 



18 1. MALONATE 

flavin and nonheme iron are reduced by succinate, and the functional role 
of iron in the catalysis is further indicated by the inhibition produced by 
complexing the iron with 1,10-phenanthroline or /J^-globulin (Singer et al., 
1957). Electron spin resonance studies of succinate oxidase have demonstrat- 
ed signals when succinate is added; these free radicals seem to be associat- 
ed with the dehydrogenase and possibly reflect changes in the states of 
iron or flavin (Commoner and Hollocher, 1960; Hollocher and Commoner, 
1960). The high sensitivity of succinate dehydrogenase to most sulfhydryl 
reagents indicates the presence of an SH group at or near the succinate- 
binding site. One may, therefore, characterize succinate dehydrogenase 
from our present knowledge as containing two cationic groups for the 
binding of succinate, an SH group nearby, some nonheme iron, and a unique 
flavin dinucleotide in a tight peptide complex. 

Relationship of Malonate Inhibition to the Electron Acceptor Used 

Measurement of malonate inhibition involves either the oxygen uptake of 
the complete succinate oxidase system, or the determination spectroscopi- 
cally of the reduction of one of the normal components (such as cytochrome 
c), or the reduction of some artificial electron-acceptor dye. The complete 
system can reduce a variety of substances in the presence of succinate, and 
malonate has been shown to inhibit such reductions whatever the acceptor 
used: methylene blue (Quastel and Whetham, 1925; Hopkins et al., 1938; 
Forssman, 1941; Franke 1944 a; Kaltenbach and Kalnitsky, 1951 a; Wad- 
kins and Mills, 1955), ferricyanide (Stoppani, 1948; Thorn, 1953), tetrazolium 
dyes (Barker, 1953; Becker and Rauschke, 1951; Zollner and Rothemund 
1954; Waterhouse, 1955), manganese dioxide (Hochster and Quastel, 1952), 
2,6-dichlorophenolindophenol (Repaske, 1954; Wadkins and Mills, 1955, 
Millerd, 1951), janus red (Agosin and von Brand, 1955), brilliant cresyl 
blue (Agosin and von Brand, 1955), and lY-methylphenazine sulfate (Singer 
et al., 1956 b). Inhibition of cytochrome c reduction has also been observed 
Seaman, 1954). There is thus substantial evidence that malonate blocks 
electron flow very early in the sequence, as would be expected if it prevents 
the binding of succinate to the dehydrogenase. The most proximal location 
of the site of malonate inhibition comes from the work of Ziegler (1961) on 
electron transport particles from heart mitochondria. Some of the nonheme 
iron is reduced by succinate and this is blocked by 20 raM malonate. It 
may also be noted that succinate reduces NAD and NADP in submito- 
chondrial particles from heart through an ATP-dependent system and this 
is readily inhibited by malonate (Snoswell, 1962; Hommes, 1963; Lee et al., 
1964), which possibly indicates that succinate dehydrogenase is involved. 
The addition of malonate to succinate dehydrogenase brings about changes 
in the absorption spectrum in the flavin region, as do succinate, fumarate, 
and other competitive inhibitors (Dervartanian and Veeger, 1962). There is 



INHIBITION OF SUCCINATE DEHYDROGENASE 19 

a decrease in absorption between 400 and 470 m// and an increase between 
480 and 540 m//, with a maximum at 510 m// in the difference spectrum. 
It is not known what this implies relative to the site of malonate binding. 
The dyes may accept electrons from various sites in the succinate oxi- 
dase system. The fragments obtained from the oxidase particles differ in 
their abilities to reduce these dyes, and as one approaches the purest succi- 
nate dehydrogenase the number of possible electron acceptors is reduced. 
Indeed, soluble succinate dehydrogenase reduces only the A^-alkylphenazines 
and ferricyanide at appreciable rates. Some dyes accept electrons chiefly 
from the nonheme iron, some accept more efficiently from the flavin com- 
ponent, and some, such as the indophenol dyes, involve ubiquinone. It may 
well be that none of the commonly used dyes is completely selective. Fer- 
ricyanide, for example, may react at other sites down the chain in addition 
to the nonheme iron, since its reduction has a partially antimycin-sensitive 
component, and even the A^-alkylphenazines, considered to be the most 
reliable acceptors at the dehydrogenase level, may be able to react at other 
sites. 

The question of importance with respect to malonate inhibition is whether 
all of the methods of measuring succinate dehydrogenase activity are equiva- 
lent for the purpose of obtaining accurate kinetic results. Unfortunately, 
there have been very few reliable investigations wherein different methods 
or acceptors have been compared. The inhibitions of succinate oxidase 
(manometric) and succinate dehydrogenase (dye acceptors) at the same 
concentrations of succinate and malonate have been reported sporadically. 
It has generally been found that the oxidase is inhibited somewhat more 
strongly: Arum spadix (Simon, 1957), Limidus gill cartilage (Person and 
Fine, 1959), and potato tubers (Millerd, 1951). But in the enzymes from 
Xanthomonas, this is reversed (Madsen, 1960). The results of Millerd are 
particularly difficult to understand, inasmuch as she found a 40% inhibi- 
tion of the oxidase at 0.1 raM malonate whereas the dehydrogenase (using 
2,6-dichlorophenolindophenol as acceptor) was not inhibited at all. The 
author is not aware of any quantitative studies comparing malonate inhi- 
bition with different dye acceptors. It would appear that most workers 
have assumed there would be no difference. 

Actually, from kinetic consideration, it is not at all necessary that the 
inhibition produced by a chosen concentration of malonate be the same 
when different acceptors are used. In fact, the inhibition may depend on the 
acceptor concentration. Thorn (1953) showed that the inhibition of pig heart 
succinate dehj^drogenase by 0.536 mM malonate increases with the con- 
centration of methylene blue: at 0.15 n\M methylene blue the inhibition 
is 15.9% and at 3 vaM methylene blue it is 28.8%. Succinate oxidase, is 
inhibited 52.6% at the same malonate concentration. The transfer of hydro- 
gen atoms from succinate to a dye acceptor always involves the interaction 



20 1. MALONATE 

of these two substances with the enzyme surface, undoubtedly at different 
sites. At low concentrations of the acceptor, or with weak acceptors, the 
over-all rate may not be determined by the rate at which the hydrogen atoms 
are removed from the succinate, but may depend also on the rate of transfer 
to the acceptor. The malonate inhibition will thus vary with the degree of 
saturation of the enzyme with acceptor. On this basis it would seem reason- 
able to use those acceptors which are the most active and react with 
sites closest to the succinate site. A further consideration is the inhibition 
produced by the acceptors themselves. Both ferricyanide and iV-methylphe- 
nazine begin to inhibit succinate dehydrogenase as the concentration is 
raised above certain levels. Such systems would then constitute examples 
of multiple inhibition and the kinetics of the inhibition due to malonate 
alone may be distorted. The choice of the acceptor and its concentration 
is thus of some significance. 

Site of Inhibition by Malonate in the Succinate Oxidase Sequence 

The results discussed in the preceding section point clearly to the site 
of inhibition as succinate dehydrogenase. Indeed, inhibition of soluble 
succinate dehydrogenase by malonate has been demonstrated. The competi- 
tive nature of the inhibition, to be treated in the following section, indicates 
the inhibition to be at the active site at which succinate is bound. There is 
thus no question but that the major site of inhibition is at the very begin- 
ning of the electron transport sequence in succinate oxidase. The question 
that now must be considered is whether malonate can inhibit at any other 
step of the electron transport chain. 

There are two obvious ways to examine this. One is to test the action 
of malonate on the succinate oxidase system, using substrates that donate 
electrons at more distal sites than succinate. The other way is to determine 
the response of other oxidases that utilize most of the electron carriers 
in the succinate oxidase. Quastel and Wheatley (1931) observed that ma- 
lonate at 67 mM does not inhibit the oxidation of p-phenylenediamine 
and hence concluded that the cytochrome region of the sequence is immune 
in their preparations. Actually, many oxidases, comprising varying segments 
of the electron transport chain, have been found to be insensitive to malo- 
nate at concentrations from 25 mlf to 50 mM. All this evidence points to 
a rather specific action on the dehydrogenase. However, there are data in 
the literature which indicate that, at least in some species and at high enough 
malonate concentrations, inhibition at other sites may occur. Although 30 
mM malonate does not inhibit NADH oxidation in beet mitochondria 
(Wiskich et al., 1960), some inhibition has been observed in mosquito parti- 
cles (10-15% at 1-10 mM) (Gonda et al, 1957) and in Tetrahymena NADH 
oxidase (35% at 6.2 mM and 70% at 18 mM) (Eichel, 1959). Such inhi- 
bition could, of course, be on the NADH dehydrogenase rather than on 



INHIBITION OF SUCCINATE DEHYDROGENASE 



21 



enzymes common to the succinate pathway. Malonate does not usually 
interfere with ascorbate oxidation, but in the silkworm it was found both 
spectroscopically and manometrically that malonate inhibits the oxida- 
tions of succinate and ascorbate equally (Sanborn and Williams, 1950). 
Observations such as these, coupled with those showing greater inhibition 
of succinate oxidase compared to the dehydrogenase, make it necessary 
to exert some caution in assuming a completely specific action in all cases. 

Competitive Nature of the Inhibition 

It has been often stated that competitive inhibition was first demonstrat- 
ed by Quastel and Wooldridge (1928) for the inhibition of E. coli succinate 
dehydrogenase by malonate. However, inhibitions were not calculated 
and the data presented do not lend themselves to quantitative interpreta- 
tion. Indeed, when their data are plotted on a l/w, — 1/(S) graph (Fig. 1-2) a 
straight line is not obtained. Since no comparable control studies were done 
in the absence of malonate, the fact alone that increasing the succinate con- 
centration increases the rate in the presence of malonate does not prove 
competition. These points are brought out not to criticize pioneering work 
but to illustrate that conclusions about the type of inhibition cannot be 
made so readily as many imagine. 




Fig. 1-2. A double reciprocal plot for the 
inhibition of E. coli succinate dehydrogenase 
by malonate at 1.43 mM. Succinate concen- 
trations are in vaM. (Data from Quastel and 
Wooldridge, 1928). 



22 



1. MALONATE 



Competition between succinate and malonate has been claimed to occur in 
the following: rat liver homogenates (Potter and DuBois, 1943), oyster 
muscle homogenates (Humphrey, 1947), oyster egg homogenates (Cleland, 
1949), carrot root (Hanly et al, 1952), yeast (Krebs et al, 1952), pig heart 
particulates (Thorn, 1953), cockroach muscle homogenates (Harvey and 
Beck, 1953), the trypanosome Crithidia (Hunter, 1960), and the soluble 
succinate dehydrogenase from heart (Keilin and King, 1960). In all cases, 
the inhibition produced by a certain malonate concentration is reduced by 
increasing the succinate concentration, but a quantitative analysis of the 
data has been seldom carried out. Most of these studies were made with 
the usual assay methods, but competition has recently been shown by meas- 
uring the reductions in the electron spin resonance signals produced by 
malonate at different succinate concentrations (Commoner and Hollocher, 
1960). 




Fig. 1-3. A single-curve plot for the inhibition of 

cockroach muscle succinate oxidase by malonate at 

0.33 mM. Ki = 0.105 mM and KJK^ = 275. (Data 

from Harvey and Beck, 1953). 



Single-curve plots (type F, see Chapter 1-5) were made for two of the 
inhibitions mentioned above (Figs. 1-3 and 1-4). In the case of Crithidia, 
Hunter showed competition with a 1/'U-1/(S) plot and the single-curve plot 
confirms this, K^ having the value of 0.22 mM and K^^JK^ calculated from 
the slope a value of 53. The results with cockroach muscle likewise fall 
roughly on a straight line, giving K, as 0.11 mM and KJK^ as 275, values 
differing somewhat from those calculated by Harvey and Beck using a 



INHIBITION OF SUCCINATE DEHYDROGENASE 



23 



different plotting procedure {K^ = 0.13 mM, and KJK^ = 200). In most 
studies insufficient data are available for plotting. It must be emphasized 
that a change in inhibition observed at two or three succinate concentra- 
tions is not adequate to prove a purely competitive inhibition. It seems 
to be rarely considered that an inhibition may not be either completely 
competitive or completely noncompetitive. The conditions for partially 
competitive inhibition were given (Eqs. 1-3-14 and 1-3-15) and the types 
of plot to be expected discussed (Chapter 1-5). An interesting example 
of this is provided by the work of Honda and Muenster (1961) on the inhi- 
bition of succinate oxidation in lupine mitochondria. Here the osmolarity 
of the preparation and assay media was varied with sucrose, and it was 
found that the interaction constant, a, defined in Eqs. 1-3-5 and 1-3-6, 




Fig. 1-4. A single-curve plot for the inhibition of 

Crithidia succinate dehydrogenase by malonate at 

1 mM. Ki = 0.22 mM, and KJE^ = .53. (Data from 

Hunter, 1960). 



varies quite markedly from values indicating nearly completely competitive 
inhibition to those showing noncompetitive inhibition. This work will be 
discussed in greater detail in a later section (see page 46), but it suffices 
to show that partial competitive inhibition by malonate is possible and that 
the type of inhibition may vary with the experimental conditions. 

The most elegant treatment of malonate inhibition is by Thorn (1953) 
at the St. Thomas's Hospital Medical School in London, using succinate 
oxidase preparations from pig heart muscle. The activity was measured by 
the reduction of ferricyanide in the presence of cyanide to block the cyto- 



24 



1. MALONATE 



chrome pathway. The usual Ijv - 1/(S) plots (Fig. 1-5) show apparently 
completely competitive inhibition, and values of the substrate and inhibi- 
tor constants, to be discussed in the next section, were calculated. Using 
average values of K„^ and K^ obtained by Thorn, the inhibition curves in 



(Malonate) = 0089mM 




No Malonate 



Fig. 1-5. Double reciprocal plots for the inhibition 
of pig heart succinate dehydrogenase by malonate, 
showing pure competitive inhibition. Succinate con- 
centrations are in mM and v in spectrophotometric 
units. (Data from Thorn, 1953). 



Fig. 1-6 were plotted. These curves show the expected reduction in inhi- 
bition as the succinate concentration is increased at constant values of 
malonate concentration. It may be noted that overcoming the inhibition 
is much more difficult at high inhibitor concentrations and this must be 
taken into account in experiments designed to show a competitive type of 
action. K„i was used rather than K,. because under the usual experimental 
conditions it is this constant that determines the behavior. 

A word must now be said about malonate reversibility. One of the cri- 
teria of competitive inhibition is that the inhibitor should leave the active 
site readily when its concentration is reduced, or that it should be displaced 
rapidly when more substrate is added. These points have seldom been 



INHIBITION OF SUCCINATE DEHYDROGENASE 



25 



tested with malonate. However, it was demonstrated quite early that the 
inhibition of rabbit muscle succinate dehydrogenase is reversible by washing; 
the preparation was incubated for 30 min. with 100 milf malonate anaerobi- 
cally and then washed 3 times on a filter — the reduction times were 15 
min for the control, 61 min with the malonate, and 17 min for the washed 
preparation (Hopkins et al., 1938). A trypanosomal succinate dehydroge- 
nase, however, showed no reversal of the inhibition, even when succinate 
was added to 50 times the malonate concentration, which is particularly 
surprising since in the living cells a good reversal was observed (Agosin and 
von Brand, 1955). The concentrations of malonate used here were not 
unduly high and so these data are unexplainable. 




Fig. 1-6. Curves showing the calculated reductions 
of the malonate inhibition of succinate dehydrogen- 
ase by various concentrations of succinate, using 
K„ = 0.366 mM, and K^ = 0.0076 mM, as given 
by Thorn (1953). 



Constants of the Inhibition 

It is not surprising that the values of the inhibitor constant, K^, for 
malonate inhibition are quite variable in the literature, because not only 
do the experimental conditions affect this constant markedly but the suc- 
cinate dehydrogenase varies in its properties from species to species. From 
the values in the accompanying tabulation and from reasonable estimates 
based on assumed or determined substrate constants for the data in Table 
1-6, it is seen that the K^'s for most preparations vary between 0.005 mM 



26 



1. MALONATE 



< 


1? 

8 s 


g 


s "~" 


5 


C/3 


o 




u 




M 






C 


[i- 


_o 


O 


■-S 


Z 


(1 


o 


c3 




& 


? 


0; 






S 


li 


5 




;z 





P5 



s ^ 







2 
'o 






3 


05 






:3;. 


-2 


c 

c5 




o 




S 


13 




:^ 


"e 


2 

c3 






-^ 




(» 


C 






C5 


"^ 


(h 


J3 
o 




c 

cS 

03 






2 








3 


c 

N 


'^ 




S 




7^ 


_« 


03 




0; 


(-1 


W 




O" 




<l 


o 


<; 




Qi 


m 


m 


O 


o 


(M LO 


lO 


t- 


QO 


lO 


o 


t~ 


o 


(N O <N 


c5 


IC 


t> Ol 


A 


05 


GO 


<r> 


o 


c- 


o 


^ C^J t> 



13 


^ 




C 


e 


•*-* 


^ 


<u 


e 


73 
C 




'«3 


>. 


s 


o 
c 

3 


73 

o 


s 



O O O -H o -^ 



o o o 

Tt< (M Tt< 



a> 73 



w TO ^ W M 



c? s; .ii 



H 



H c^ 



O c3 
C» Ph 



w 



w 



S 



fej 



^ W 



d d 



O-^t^ coo 'MCC'M CO — O 



s s 



^ ^ 





-S 


ei 






"3 


TO 




f, 


_o 


o 


cS 


Ch 


'-^ 




b4 




Ch 


"c 


-»^ 


X 


c3 


o 


X 


W 


|1h 


M 


W 





g 






•ra 


s 




s 




o 


^«e 


S) 


a 


?■ 



INHIBITION OF SUCCINATE DEHYDROGENASE 



27 



^ pq 



>> 

J 



Tf C5 O C-1 

o ^ 00 05 



§ 




o 


■* r- ci o 
Tt< o u-j t- ^ 


o 


o 


o o 


o 


o o -< o o 


o 



•— I .^ c^ »o 



LO L-^ CO CO CC M 
C-l (M CO CO CO CO 



-M (N Tff 00 O 



P^ 



a 3 



-^ ''I 



Ti 


o 


C 


o 


CS 


^ 


0) 


cS 


-ti> 


T-! 


c 


C 


p 


O 


W 


a 


00 


lO 


CO 


00 



o 
d -H 



P '-3 
5 Is 



05 



a 



s ts 



PM 



« 




Ol 


•^ 


« 




S 


^ 


s 


>, 


■e 




a; 


s 








s 



s a 



or, e 
s s 



^ -^ a, ^ 



fin P4 



M 2 



^ g 



05 O C5 CO Oi O 
00 C<| t^ CO o 



-H O O — I o ^ 



,S ^ (U 



s (^ 



« 



* a 



H ^ 



M 



28 



1. MALONATE 



05 



1 S 



PL, 



OO 00 00 

M 00 CC 



a 

ft 



,0 .2 

c o 






00 










^ 










05 










i-H 










'•-*' 










>> 




, , 






V 




CO 






^4 




lO 






^ 




Oi 






ft 










s 




■-^ 






s 




2 






W 




o 
o 




t^ 


T3 




s 




2 


C 
c3 


OO 


t3 


^^ 




>. 


05 


e3 


IN 

O 


>. 


<D 






05 


p 


(h 




c 




tx 


^ 


b 


o 


^_^ 


j3 


ft 


"^ 




ft 




-*^ 




T3 




C3 


c 


^ 


rt 


S 


^ 


^ 


1-5 


0) 


3 


« 


w 


05 


w 



i-O-^C- M 00 f- OOOCO OOC<I 
MI>05 Oi C5 >Cl t-t>00 CDOO 



-^ O OO-H -HC^i o — 



m o -^ ic s*5 



lO W lO 



« 2 



cS 


cS 


C 


C 


0) 


a> 


bC 


M 


O 


O 


a 


S 


o 


o 


w 


X 



■i 


•S 

Si 




>> 

O 


=0 




1 




1 


Si 

s 






5 




g 


e 


o 


=0 


>, 


s 


OQ 


+i 


g 




"S) 




S 




CI 




§ 


SQ 


^ 


^"^ 


a: 




O 


w 



INHIBITION OF SUCCINATE DEHYDROGENASE 29 






m 



s 

i> •- .s; 

»c s ^ 

c^ ^ i: 












O W CO -H 



o 


eS 


s 


« 


rj 


a 


cS 


o 


Ut 






o 


s 


o 


O 


cc 


p^ 


QO CO 




Oi <u a 


fO X 


CO 


lO 00 



CO 73 



00 


CO 




IM 






o 


CO CO 


t^ 


CO 


ut 


»o 


9 


O CO 


t^ 


o 


(M 


■<* o 



s 


cS 




o3 


eS 


eS 


s 


^ 


c 


a 




C 


C 


C 


cS 


cS 


v 


o 




0) 


a> 


a> 






b£ 


M 




bD 


bc 


bc 


3 


"3 


O 


O 




O 


o 


O 


« 


o 


s 


S 




S 


S 


S 


'•S 


'•S 


o 


o 




o 


o 


o 


c3 


ei 


W 


W 


be 


W 


a 


K 


PLi 


Ph 




^ 


'-3 


j2 


CO 

3 


00 




3 


O 


o 


^ 


S 


s 


^ 


§ 



OOOCOO— 'O'CO— <o O OOO OO-^iOO ^ 









2 § £ g > 



^ -^ -8 





s 




s 




■4) 


1) 








e 












-iS 




s 




^ 


"a 




QB 












1 




bc 

3 


Co 

-2. 


J3 




•£ 
£ 


^^ 


3 




i 

t 










o 

1 
2 
o 


to 

-2 


a; 


o 


o 
O 

o 

o 


8 

i;5 


o 

o 


g 


3 

cr 

03 

o 


1 


a 

S-i 

O 


a 


c 
IS 


c 
o 

b£ 




W 




1 






s 


cc 


« 


PM 


§ 



30 



1. MALONATE 



tf 



fe 



o 

M 

03 



OS ,-. 

Oi lO ^^ 

^^ 05 CO 



-ii 



^ -^ 

i; O O 

< ffi § 



Ah 



;?? 



I 






»CC0(M05O'^00'^20O 



Tti>p-Hco-*t-?cOQOirjTt*i>os 



o o o o 



o o o o — < 



oooooooo-^ooo 
o^ooooooooooo 






OOOO-HOOOO^ 



^ ■^ lO CO CO CO CO 

-^OOO'-HO^C0O'-<C0 



^ 



G Si S 
O («1 o 

ffi H W 



c 
? 
bo 
o 

s 

o 

w 



w 



INHIBITION OF SUCCINATE DEHYDROGENASE 



31 



C5 fi^ 2 






^ ^ ^ 

CO CO o 

IC »0 lO 

05 05 



^ ^ -Q 

^ X! ^ 

(B <U 4) 

T3 T3 TS 

C C C 



b TJ t^. t^. >•, 



<U ^ C. G S Oi <b a 

S Si -t^ o £ S S 

O-SfQMCoOO 
MO!. O * MMM 

GOSOoOOO 



<N (M 






«c »o 






05 05 






i-H t-H 


^_^ 


^.„^ 


^ — ^ — 


uo 


>-o 


1— 1 f— 


lO 


o 


<0 <D 


05 


05 


-^ -^ 






03 tC 






c5 cS 






3 S 


o 


o 


GfC? 


m 


<0 






CO 


eS c3 


73 


-o 


;h Ld 


C 


c 



o o 



COOO»CCO(MfOC<lfOM-^Ot'^ 00050COIOO 
00O51O00O5GOO500 -hC000O5CO -HF^cOlOOt^M 



>> >> 







j3 


-S3 










^^ 


^ 








T3 


73 


00 


o 






C 


C 


05 






eS 


c3 


^ 








^^ 


^^ 




^~^ 






<U 


OJ 


;h 








^ 


-IJ 


IB 


in 








^ 










P 


p 


cS 


O 






a o- PM 


Ph 


t- -H 


o 




»o 


o 


o 


Tf 'O 


t- 


C5 


C5 


o 


o 



^H O O-HfCCOO-^fOO 

CO O 0-H(N lOcOCOOOOOOOOO 



lO -^ CO lO o 



o 
in lo lo o 



o-Hioo>o -H-Moo-^eoo 



CO. CO CO 

O ^ CO 



^iW 1.,/ -V ^ig ^g I.U 



T3 




C C 


rt -3 -r! -c 
c a c c 


O 


+2 


o o 


S<. o o o 


j: 


o 


js j= 




o 


rt 


o o 


2 o o o 


o 




o o 


s o o o 






■i^ -t^ 


— -u HJ -u 










S 


w 


s s 


W § S S 





g =« 


eS 








r; 


bC P 


3 


1 


O o 


■-3 



>< O cS 

w W pm 



^1 



3 § w 



■ti '-H r-' -*^ 



•3 " .3 



w 



Oh S pq W 



t- CI .3 

b. 1^ 3 

.£ fc; (B 

i-:i M cc 



2 2 



-5 rH -tJ 






pq M 



(30 

■ft 

eS 
(D 

C 

'3 



32 



1. MALONATE 



tf 





c8 
C 


^ 


s 


_o 




•r^ 


"3 


s 


o 


§ 




^ 












CO 


+3 




J^ 


eS 


Q 




, J5 


"^i 


u 


o 


c 


tS) 


o 


s 


n 


3 


"" — 


< 


W 




H 







Oh 






fq 






Pi ;^ 



aj 3 



-C ^ k-H 



CO 05 






« o 









CO 


__^ 






S 


05 


;o 






tH 




»o 






m 


■^-^ 


05 






C 
O 


>> 
-2 


^ 






o 


a;) 






pq 




_N 






rt 


« 


"3 


CO 
CO 






s 




52 !- 






c 


73 

c 






o 


3 


ci 


^ 


c 




03 


^ 


oj 3 


c3 




cS 


b 


d ^ 


Lh 


o 


3 


03 


Q H 


fe 


W 


o* m 


O CO CO 


o o 


!M 


t- 


-H -H CO 


lO 05 05 


05 oo 


00 


00 


O 00 00 



rt O -H O 



rt (M -H 



■* 


(M 


»0 >^ ■* 'M 


o 


o Tf< c^ CO 


o 



O <M ^ O '^1 O -H 



CO -^ -^ "^ CO 

CO t^ t^ t-~ o -^ 



-i3 a> 


2 ^ 




3 






c ce 


55 '^ 




eS 






O -3 


0) — . 
M 3 




-2 -3 


o 


+3 


3 t 


O .O 
3 -k^ 




11 




1 


V3 cS 


O c3 


3 


O ci 


"S 


>1 


S Ah 


W P^ 


;^ 


M Ah 


w 


w 






>. 


3 


a) *3 


p 


3 >- 




■C S5 


K»> 


.« o 




M W 


H 



M 



Ul 


<*- 




o 


"eS 


X 


Q 


o 


o 



lO -H (N 

-H CO CO 



-^ -H ^ CO 



3 -►^ 



"3 s 





13 


o 

_3 


3 




'o 




^3 


;h 


"i 


'& 




PQ 


W 



INHIBITION OF SUCCINATE DEHYDROGENASE 



33 



Preparation 



Ki {mM) 



Reference 



Pig heart 


0.0076 


Rat heart 


0.01 


Yeast 


0.0105 


Ehizobium japonicum 


0.017 


Claviceps purpurea 


0.03 


Corynebacterium diphtheriae 


0.037 


Beef heart 


0.041 


Beef heart 


0.045 


Mytilus edulis 


0.06 


Beef heart 


0.13 


Micrococcus lactilyticus 


0.23 


Phaseolus vulgaris 


0.24 


Lupinus albus 


0.91 



Thorn (1953) 

ITyhn and Matsumoto (1964) 

Ryan and King (1962a) 

Cheniae and Evans (1959) 

McDonald et al. (1963) 

Strauss and Jann (1956) 

Kearney (1957) 

Keihn and King (1960) 

Ryan and King (1962 a) 

Lee et al. (1964) 

Warringa and Giuditta (1958) 

Hiatt (1961) 

Honda and Muenster (1961) 



and 0.05 vaM, and mammalian tissues generally yield quite low constants. 
The interest in the K, lies in its relationship to the binding energy of mal- 
onate to the active center, so that variations in the K^, unless due to exper- 
imental conditions, may be attributed to differences in the topography 
of the enzyme surface. The values of £",„ for different preparations are not 
particularly significant, except for characterizing a certain preparation 
under specified conditions, because .K",,, does not usually equal K^.. Inasmuch 
as the ratio of K^ to K j is of some significance in interpreting interactions 
with the active center, it will be worthwhile to discuss the only instance in 
which this has been determined. 

The ratio, K„,!K„ has been given by various investigators as ranging 
from 10 to 60. Thorn (1953) pointed out that K,„ for succinate dehydroge- 
nase is often quite different than K, (Slater and Bonner, 1952). K,„ thus 
equals (A-j + A'2)/A:j, and A-j is dependent on the experimental conditions. 
Thorn obtained values of KJK^ from 3 to 60 by varying the electron- 
acceptor dye and the reaction rate. The faster the rate, the greater the de- 
viation of ^,„ from K^. Thus extrapolation to zero rate from a series of 
experiments at different methylene blue concentrations enabled Thorn to 
determine the true ratio of succinate and malonate affinities; KJK, turned 
out to be 3. Since K, is approximately 0.0076 milf . K, = 0.023 mM. a value 
which checks well with directly determined values of the rate constants 
(A'i = 3.35x 10^, and A-_i = 0.99). Similar low ratios would probably be found 
in most succinate dehydrogenase preparations. The difference in binding 
between succinate and malonate is thus not so great as previously believed. 

It is of great practical importance to realize that the degree of inhi- 
bition of succinate oxidase by malonate varies with the rate of succinate 
oxidation and the electron acceptors present. This may explain some of the 



34 1. MALONATE 

differences in Table 1-6. If one is to compare the inhibitory potencies of 
malonate on succinate oxidases from different tissues or species, equivalent 
rates of succinate oxidation should be used and similar methods of deter- 
mining the rate should be employed. 

Inhibition of Succinate Dehydrogenase by Other Dicarboxylates 

Before discussing the more intimate nature of the inhibition and the 
possible ways by which malonate is bound to the active center, it will be 
useful to consider the inhibitory potencies of other dicarboxylate ions. 
The configuration of an active center may often be approached by comparing 
the relative affinities of analogous compounds for the enzyme. At the end of 
this chapter a more complete discussion of inhibitors related to malonate 
will be given; for the present we shall be interested only in the inhibition 
of succinate dehydrogenase. Inhibitions and inhibitor constants are summa- 
rized in Tables 1-7 and 1-8. 

(a) U nsubstituted dicarboxylate ions. It was stated by Quastel and Whet- 
ham (1925) that oxalate, glutarate, and adipate do not interfere with 
succinate oxidation, in contrast to malonate, and, although in later work 
they found some inhibition, this has generally been confirmed. Accurate 
comparisons must be made using inhibitor constants, and these are not 
available, but there is no doubt that in the series ~00C — (CHa)^ — C00~ 
the inhibition reaches a sharp maximum at w = 1. One must assume that 
malonate best fits the intercationic distance on the active center of suc- 
cinate dehydrogenase, and this includes succinate itself since K^ is generally 
larger than K, for malonate. 

(6) Unsaturated dicarboxylate ions. It is interesting to compare the two 
isomers, fumarate and maleate, since they differ in the intercarboxylate 
distance (Table 1-1). For mammalian succinate dehydrogenase it would 
appear that fumarate is bound much less tightly than either succinate or 
malonate (Table 1-8), but this does not necessarily hold for the bacterial 
enzymes, providing evidence that the active center configurations may be 
quite different in different dehydrogenases. One might expect a rather low 
affinity for fumarate because of the fairly long intercarboxylate distance, 
but it is surprising that maleate is such a poor inhibitor inasmuch as its inter- 
carboxylate distance in only 0.38 A greater than in malonate. A complicating 
factor is the ability of maleate to react with SH groups, and actually the 
inhibitions observed by Hopkins et al. (1938) and Morgan and Friedmann 
(1938 b) were obtained only after prolonged incubation. The possible 
significance of these observations will be discussed in the next section. 
On the other hand, acetylene-dicarboxylate, with a distance of 5.16 A 
between carboxylate groups and a restriction to linearity, does inhibit and 
this is completely competitive (Thomson, 1959). 



INHIBITION OF SUCCINATE DEHYDROGENASE 



35 



P5 



IS ^ 






«> lii 



S^ 2 



P T3 



s 


15 


5 


T3 




T) 


S 


3 


C 


cS 




c3 


t4 


^2 

CO 


S 


.^ 


(S 


OI 


o 


^ 


O 


(1^ 


c? 


tf 



w 



Q Ec5 



<M 2 r- 05 

Oi _ eo -^ 



- QU -— - 

S "« - 

S 'E S 

<u "o X 

^ ^ ^ 

^ ^ H 

XS ^5 _ 

eS cS C 

— — ** 

11^ 

5 3 O 



t> o 



■<* 00 o o o 
ec <M ^ lo o 



Eq 



« ^ +3 

C o t- 

W § K 



■" O o3 
O ffi Ph 



e<5 <3^ M 



.1 


s 


s 


cT 


c 
o 
o 
pq 


c 
So i 

CO S 


'a 

> 


5s 


> 


2 


o 


2 1 


T3 


"^ 


Tj 


"e 


T3 


<4> r< 


C 


s 


fi 


<u 


cS 


cS 


d 




cS 






CO 




'3 

3 


o 

(1h 


2. 




o o 



t^ 



fcii ffi ffi S K 



-M 



o &3 o P5 Ph Ph Pm 



o 





a 


'^ 


3 


<3 


fM 



36 



1. MALONATE 



^ 





CD 


CO 




lO 


U5 


__^ 


05 


OS 


<s 


^^ 


1— 1 


■* 


■ — ■ 


' — ' 


05 


Xi 


Xi 


^ 


X5 


-S 




a 


-^ aj 


_c 


^ 




fli 




CJ 


J« 


73 


-H 73 


^ 


C 


^ c 


^ 












o 


ll 


9^ 


M 


(h 


-ki 


t-i -p 


0^ 


o 


.£ o 


fc 


§ 


Q § 



5 2 - 



fcH 2 



W 



-0 

c 


73 

C 


-c 


03 










a; 


01 




« 


-k^i 


TS 


CO 


;h 


c3 


-t-l 


cS 


S 


o 


fin 


o- 


fin 



.S cc 



i>aooc<iooiMOiio 



:5 ^ 



CO M IC O 
rt CO (M 



T»<C^) -hM:0000<NO '-h 



TO r 



-* o 


CO 


CO o 


o 


o 


CO K> 


CO 


-H lO 


CO 


IC 



PQ 



g a 



.2 « 



m 



t^ 



05 



Ph Oh 



O Oh 



a, 0:5 



&q 



3 -^ 



o3 
C 

o ^^ 

S ce 

>< S 

2 -^ 



INHIBITION OF SUCCINATE DEHYDROGENASE 37 



^ lO 






CC ^ CO 

- - -* - - CO CO ^"^ ''^ ''^ ^^ '•^ "-^ 






O OOOOOOOOOO O "^^iSkS 



-x ^ c 



-u .-. 



c5 




o 


(S 


C 
O 




e3 


c 
o 


eS 


c 


O 


cS 


5= 


o 


o3 














a 




s 

42 








M 


< 


M 


W 



-T « S « c ;>> 



5-3 



o tc CO 



-^ ^ o 



"'2 - -_, ._. <~- 

_ ..^_ „ - « C O >)'rt '^T-l B-^ J3 



(D 


CS 


C 


CC 


& 




>. 


OJ 


>> 


o 


CS 


y, 


P- 


y. 






o 
Id 




o 

XI 
o 




O 


t3 




-0 


1 



38 



1. MALONATE 



Table 1-8 
Inhibitor Constants for Dicarboxylate Ions on Succinate Dehydrogenase 



Source of enzyme 


Inliibitor 


Ki (mM) 


Reference 


Corynebacterium 


Malonate 


0.037 


Strauss and Jann 


diphtheriae 


Fumarate 


1.8 


(1956) 


Micrococcus lactilyticus 


Malonate 


0.23 


Warringa and Giix- 




Fumarate 


0.22 


ditta (1958) 


Claviceps purpurea 


Malonate 


0.03 


McDonald et al. 


(ergot) 


Fumarate 


0.93 


(1963) 


Yeast 


Malonate 


0.0105 


Ryan and King 




Fumarate 


1.03 


(1962 a) 


Mytilus edulis (mussel) 


Malonate 


0.06 


Ryan and King 




Fumarate 


0.15 


(1962 a) 


Rat kidney 


Oxalacetate 


0.0015 


Pardee and Potter 

(1948) 




Acetylene-dicar- 


0.81 


Thomson (1959) 




boxylate 






Pig heart 


Malonate 


0.05 


Hellerman et al. 




Fumarate 


3.5 


(1960) 




Oxalacetate 


0.0016 






Ethyloxalacetate 


0.04 






Diethyloxalacetate 


No inh. 






Acetylene-dicar- 


1.4 






boxylate 








ci5-Cyclohexane- 


132 






1,2-dicarboxylate 






Beef heart 


Malonate 


0.041 


Kearney (1957) 




Fumarate 


1.9 






Itaconate 


1.8 


Dervartanian and 

Veeger (1962) 



(c) Oxalacetate. The marked inhibition exerted by this substance was first 
noted by Das (1937 b) and this has been confirmed on the succinate dehydro- 
genases from a variety of organisms. It is at present recognized as the most 
potent succinate dehydrogenase inhibitor among the dicarboxylate ions. 
The binding of the oxalacetate to the enzyme would appear to be at the 
active center because the inhibition is competitive with succinate (Pardee 
and Potter, 1948; Kearney and Singer, 1956; Hellerman et al., 1960), and 
the presence of oxalacetate on the enzyme protects the active center from 
various sulfhydryl reagents (Stoppani and Brignone, 1957). The reason for 



INHIBITION OF SUCCINATE DEHYDROGENASE 39 

this strong inhibition is still not clear but Hellerman and his group have 
advanced some interesting speculations. Oxalacetate in aqueous solution 
may exist around neutrality in three forms in equilibrium. It was suggested 
that the enolate form may be the potent inhibitor, especially the trans- 

'CX)C coo" ^ "OOC^ H 

^c=c'^ -* »^ 'ooc-c-CH,— coo" -* ^ ,c=c^ 

-Q^ ^n 'o^ ^coo" 

cJs -Enolate Keto tautomer trans -Enolate 

tautomer tautomer 

enolate tautomer where the enolate group and the carboxylate group are 
on the same side. Evidence for this was provided by showing that the 
oxalacetate monoethyl ester is inhibitory (about as potent as malonate) 

EtOCO^ H 

^C = C^ 

~o^ ^coo' 

^raws- Enolate 
/ tautomer of 

monoethyl ester 

(Table 1-8). The diethyl ester in inactive. Actually there is a difference in 
binding energy of about 2.1 kcal/mole between oxalacetate and either 
malonate or the monoethyl ester. This extra energy might be due to the 
doubly negative charge on one end of the molecule. It could well be that 
both the cis- and ^ra ws-enolate tautomers are bound, the charge distri- 
butions being almost equivalent if one takes the center of negative charge 
as lying between the carboxylate and enolate groups. It is unfortunate that 
the per cent of the oxalacetate in the enolate forms at pH 7.6 and 30° in 
aqueous solution is not known. Another possibility is that the keto tautomer 
combines with the enzyme, the extra binding energy arising from an inter- 
action between the keto group and the enzyme; a hydrogen bond could ac- 
count for the 2.8 kcal/mole difference between the binding energies of ox- 
alacetate and succinate. One difficulty in assuming the enolate form as the 
active inhibitor is the fact that maleate inhibits very poorly and fumarate 
not a great deal better. The difference in binding energy between oxalace- 
tate and fumarate for the pig heart succinate dehydrogenase is close to 
4.8 kcal/mole, and it would be difficult to account for this large difference 
simply on the basis of additional ionic interactions. A final possibility must 
be entertained, although no evidence for it exists, namely, the complexing 
of the enolate group with one of the nonheme iron atoms near the cationic 
groups of the enzyme surface, since this could provide the extra energy 
for the binding of oxalacetate. It should be noted that although oxalacetate 
strongly inhibits succinate oxidation in rapidly respiring liver mitochondria, 
both coupled and uncoupled, it stimulates when the mitochondria are in a 



40 1. MALONATE 

state of respiratory control (Kunz, 1963). Malonate does not stimulate 
under these conditions and thus the effects of oxalacetate on intact mito- 
chondrial succinate oxidation differ in some manner from those of malonate. 
This, however, is probably explained by other reactions of oxalacetate, e.g., 
the oxidation of NADH to NAD, which would release the mitochondria 
from respiratory control. 

{d) Substituted malonates and succinates. Franke (1944 a) found that alkyl- 
malonates are inactive on heart succinate dehydrogenase, confirming the 
weak inhibitions reported by Thunberg (1933). Alkylsuccinates likewise are 
inactive until the length of the alkyl chain is greater than eight carbon 
atoms. These higher alkylsuccinates may exhibit some degree of competitive 
inhibition but there is also inhibition of the oxidation of p-phenylenediamine 
and, hence, of the cytochrome system. Therefore, it is clear that additions 
of even small alkyl groups to malonate and succinate depress or abolish 
the normal inhibitory activity. Hydroxymalonate (tartronate) also is es- 
sentially inactive. These results indicate the importance of that part of the 
molecule between the two carboxylate groups and would seem to argue 
against a simple interaction of the molecules with cationic groups on a 
flat surface, the substituted groups protruding outward. Of course, it is 
necessary that the — CHgCHo — grouping of succinate interact with the 
enzyme in order for dehydrogenation to take place. The failure of short- 
chain alkylmalonates to inhibit appreciably must be attributed to some 
manner of steric interference by the alkyl groups. 

(e) Cyclic dicarhoxylate ions. Cyclobutane and cyclopentane dicarboxylates 
are weak inhibitors of succinate dehydrogenase. Since the intercarboxylate 
distance in cyclobutane- 1,1-dicarboxy late and CT's-cyclopentane-l,2-dicar- 
boxylate are not too far from that in malonate (Table 1-1), the poor inhibi- 
tion may be due to the bulkiness of the rings interfering sterically, as do the 
alkyl groiips discussed in the previous section. It is strange, however, that 
there is no inhibitory difference between the cis and trans isomers of cyclo- 
pentane- 1,2-dicarboxylate, since the intercarboxylate distances differ by 
1.39 A. Thus the inhibition may not be by the same mechanism as for mal- 
onate; indeed, cw-cyclohexane-l,2-dicarboxylate inhibits succinate dehy- 
drogenase noncompetitively (Hellerman et al., 1960). 

Nature of the Active Center and the Binding of Malonate 

The evidence indicates the presence at the active center of two cationic 
groups and a nearby SH group. The cationic groups, perhaps 3-4 A apart, 
are suggested by the very weak inhibitions exerted by monocarboxylates 
(Quastel and Wooldridge, 1928; Dietrich et al., 1952) and the complete 
lack of a competitive inhibition by compounds in which the negative 
charges on the carboxylate groups are eliminated. Malondialdehyde (Holt- 



INHIBITION OF SUCCINATE DEHYDKOGENASE 41 

kamp and Hill, 1951), malondiamide (Fawaz and Fawaz, 1954), and var- 
ious derivatives of succinate, in which one or both of the carboxylate 
groups have been replaced with nonionic groups (Dietrich et al., 1952), 







/COO 


/OH 


^CHO 


CONH, 


H,C 


H-C 


H^C^ 


H,C^ 


1 


1 


^CHO 


CONHj 


"^^-Br 


"^*^^CN 



Malondialdehyde Malondiamide /3-Bromopropionate /3- Hydroxy propiononitrile 

are all lacking in inhibitory activity. An SH group close to the cationic 
attachment points is proved by the high sensitivity of the dehydrogenase to 
substances reacting with SH groups and the protection that malonate af- 
fords against such substances. The latter is actually better evidence for the 
proximity of the SH group because sulfhydryl reagents can alter protein 
structure and exert effects for some distance over the enzyme, whereas the 
blockade of these substances by a small molecule such as malonate would 
be almost certain proof. Hopkins et al. (1938) showed that malonate, can 
protect succinate dehydrogenase from oxidized glutathione (GSSG), which 
is a potent inhibitor. Incubation of the enzyme with GSSG increased the 
methylene blue reduction time from 10 min to 3 hr; malonate at concentra- 
tions from 0.2 to 100 mM gave almost complete protection. That the 
SH groups are protected by malonate was also shown by titration of these 
groups with iodine. Potter and DuBois (1943) reported protection against 
quinone inhibition and Barron and Singer (1945) against arsenicals. It is 
interesting that malonate also protects against oxygen poisoning of suc- 
cinate oxidase, which was interpreted to mean that the SH groups of the 
enzyme are involved in this inactivation (Dickens, 1946 b). Oxalacetate 
has also been shown to be protective against mercurials and arsenicals 
(Stoppani and Brignone, 1957), providing evidence that it binds to the 
same site as succinate and malonate. The degree of protection depends on 
several factors, including the concentrations of the inhibitors and the rates 
at which they react with the enzyme; the less protection seen against the 
mercurials is attributed to their comparatively rapid action whereas the 





Inhibitor 


Oxalacetate 






Inhibitor 


concentration 
{mM) 


concentration 
(mM) 


0/ 

/o 


Protection 


jj-MB 


0.5 


1.3 




21.8 


p-Chloromercuriphenol 


0.76 


3.4 




24.0 


HgCl, 


0.38 


3.4 




53.8 


Methylarsenoxide 


2.0 


3.4 




67.2 


Oxophenarsine 


0.36 


1.3 




81.2 



42 1. MALONATE 

arsenicals require at least 30 min to reach their maximal inhibition. Such 
protection on a competitive basis was called interference inhibition by 
Ackermann and Potter (1949). Once the sulfhydryl reagents have reacted 
with the enzyme, malonate will not reverse the inhibition, but only slows 
down the rate at which the substance acts on the enzyme. (I find it difficult 
to understand how, in some cases, such low concentrations of malonate 
afford protection against irreversible inhibitors. For example. Potter and 
DuBois reported that 0.33 mM malonate protects quite well against 
p-quinone, and yet this concentration of malonate inhibits only around 
20%, showing most of the enzyme uncombined with malonate). 

Succinate dehydrogenase also contains nonheme iron and flavin dinucleo- 
tide but the locations of these components relative to the succinate-binding 
site are not known. Since the iron and the flavin both participate in the 
electron transfer, it is reasonable to assume that at least one of them is 
close, or even part of the active center. Most formulations have pictured 
the initial step as a transfer of hydrogen atoms from succinate to the fla- 
vin; if this is so, the topography of the active center must be rather 
complex. 

We shall now turn to the energetics of the binding of malonate in order 
to determine if the ionic interactions generally assumed are reasonable. 
An immediate difficulty is the variability in the values of K^ reported, even 
for the same tissue; for example, 0.0076 mM (Thorn, 1953) and 0.05 milf 
(Hellerman et al., 1960) for the enzyme from pig heart, and 0.041 mM for 
beef heart (Kearney, 1957). A K^ of 0.0076 mM would indicate an over all 
binding energy of 7.24 kcal/mole, or 3.62 kcal/mole for each carboxylate 
interaction assuming only ion-ion contribution. This is a reasonably high 
value for the interaction of C00~ and NH3+ groups and corresponds to 
an intercharge separation of around 4.30 A (Fig. 1-6-16), which is not far 
from contact of the groups. It is unlikely that other types of interaction 
are important for malonate. The corresponding K, for succinate is 0.0028 
mM, giving an interaction energy of 3.28 kcal/mole per carboxylate group 
and a separation of 4.45 A. A difference of fit of 0.15 A would thus account 
for the relative bindings of malonate and succinate. However, in the case 
of succinate it is more likely that other energy terms are involved. There 
is undoubtedly some interaction between the — CH2CH2 — region and the 
enzyme, and it is probable that distortion of the succinate molecule occurs 
upon binding. In any event, these rough estimates point to a fairly close 
fit of malonate to the enzyme cationic groups for the pig heart enzyme. 
The ability of malonate to bind to the active center of succinate dehydro- 
genases from bacteria is apparently much less (Table 1-8). 

The next question is: what is the most probable distance between the 
two enzyme cationic binding sites? Information on this must be obtained 
from the relative bindings of substances having negatively charged groups 



INHIBITION OF SUCCINATE DEHYDROGENASE 43 

different distances apart. Since malonate is usually bound more tightly 
than succinate, and much more than oxalate, it is reasonable to assume an 
intercationic distance approximating the intercarboxylate distance in mal- 
onate. It is by no means necessary that the substrate in its free configu- 
ration exactly fits the enzyme site. Pauling (1946, 1948) has suggested, 
" an active region of the surface of the enzyme... is closely complementary 
in structure not to the substrate molecule itself, in its normal configuration, 
but rather to the substrate molecule in a strained configuration, correspond- 
ing to the activated complex for the reaction catalyzed by the enzyme." 
Now, the fact that malonate fits the active site well does not mean that the 
enzyme cationic groups are the same distance apart as the carboxylate 
groups (3.28 A). The calculations above indicate a distance of 4.3 A 
between carboxylate and cationic groups and thus, depending on the 
geometry of the binding, the cationic groups could be much farther apart 
than 3.28 A. Extreme situations are shown in Fig. 1-7, where the intercat- 





FiG. 1-7. Representations for the ex- 
treme situa/tions in the interaction of 
malonate with the two cationic groups 
on the surface of succinate dehydro- 
genase. In both cases (A and B) the 
interaction distances and the energies 
between the ionic groups are the same. 

ionic distance may vary from 3.28 to 13.1 A, approximately the same energy 
of binding being expected in either case. Situation A is not very likely 
because it is improbable that protein cationic groups would occur so close, 
and, furthermore, in this case oxalate might be expected to bind quite well. 
Also, situation B would provide more opportunity for succinate to be dehy- 
drogenated at the enzyme surface. Of course, the enzyme surface at the 



44 1. MALONATE 

active center may not be smoothly curved as shown, and we shall soon 
examine evidence that it is not. 

The problem of the interactions of di-ionic substances with receptor 
groups has been recently treated by Schueler (1960, p. 448) and on the basis 
of statistical calculations he has concluded, "The most dramatic alteration 
in activity should occur upon approaching that agent in the series which 
possesses a length distribution just capable of overlapping the negative- 
charge spacing in the receptor, and this should be followed by a relatively 
slow rate of loss in activity with respect to increasing length" (he is assum- 
ing a positively charged drug). In view of what has just been said above, 
there is some doubt if predictions like this can be made with confidence. 
The large distances between the interacting charged groups make it very 
difficult to assign receptor configuration and much will depend on the over 
all configuration of the protein surface. Other factors, such as distortion of 
long-chain molecules, interactions of regions between the end charged 
groups the surface, and possible steric repulsions, must be considered. 

Inasmuch as little accurate information on the intercationic distance can 
be obtained from Kj and K^. values alone, let us now turn to more profitable 
considerations of the topography of the active center. There is evidence 
from several lines that the active center is not a flat or slightly convex 
surface. In the first place, alkylsuccinates are bound to the enzyme very 
poorly; methylsuccinate is oxidized at 23% the rate for succinate, and 
ethylsuccinate at 18% the rate for succinate, while higher members are 
neither oxidized nor are they inhibitory (Franke, 1944 a). In the second 
place, as we have already seen, alkylmalonates are very poor inhibitors. 
Indeed, even the introduction of a hydroxyl group (tartronate) reduces the 
inhibition markedly. These observations indicate a rather close fit for 
malonate and succinate at the active center, additional groups giving rise 
to steric repulsion, as frequently reported for antigen-antibody reactions. 
In the third place, fumarate is bound fairly well while maleate is not, indi- 
cating again some steric repulsion since the intercarboxylate distances 
alone would certainly allow predictions that maleate would be bound more 
tightly. All molecules that bind appreciably to succinate dehydrogenase 
seem to be simple linear substances, or substances capable of assuming a 
linear configuration. All of this evidence points to a slit or tubular structure 
for the active center, such that compounds with added groups or rigid non- 
linear molecules cannot enter. Such a situation is pictured in Fig. 1-8 in 
two dimensions. This is not to be construed as an attempt to represent 
the actual configuration but merely to show the steric barriers impeding 
attachment of larger or nonlinear molecules. Glutarate and adipate could 
not fit well, not because of unsatisfactory intercarboxylate distances, but 
because of the bulkiness of the longer hydrocarbon chains. Fumarate is 
able to bind because its configuration is much like that of succinate in the 



INHIBITION OF SUCCINATE DEHYDROGENASE 



45 



extended form shown in the figure, whereas maleate might not fit because 
of its nonlinear structure. Acetylene-dicarboxylate would be expected to in- 
hibit to some extent because of its linearity. Such a model would also ex- 
plain why small alkyl groups added to succinate do not completely abolish 
the binding. It may also be mentioned that this type of configuration would 
allow the flavin and iron components of the dehydrogenase to be in posi- 
tions close to the — CH2CH2 — group and thus able to participate in the 
removal of the hydrogen atoms. 




Succinate 




Fig. 1-8. Representations of the binding 
of malonate and succinate at the active 
site of succinate dehydrogenase, indicat- 
ing the steric barriers possibly surround- 
ing the region of the two cationic sites. 
The actual situation must be visualized 
in three dimensions. 



Activation of Succinate Dehydrogenase by Malonate 

Preparations of beef heart succinate oxidase obtained using borate buf- 
fer are not fully active but may be activated by the addition of phosphate. 
This interesting discovery by Kearney (1957, 1958) may have important 
bearings on the understanding of the active center of this enzyme, especially 
as she later found that succinate, fumarate, and malonate also activate, 
and indeed are much more potent than phosphate (see accompanying 
tabulation). Once the enzyme has been activated, the activator can be 
removed without loss of the activity; in fact, malonate must be dialyzed 
away if the full activity of the enzyme is to be measured. The activation 
constants are quite different from the Michaelis or inhibitor constants for 
these substances. It would appear that malonate binds more tightly to the 
less active form of the enzyme. Kearney favors the view that these activa- 



46 1. MALONATE 

tors convert a less active form of the enzyme to a more active form, and 
that this transformation may involve a localized change in the protein 



Activator 



(mJf) (mM) 



Malonate 


0.0072 


0.025 


Succinate 


0.12 


0.52 


Fumarate 


5.6 


0.80 


Phosphate 


100 


— 



structure because of the high energy of activation. Whatever the explanation 
it is important to remember that malonate can exert an activating effect, 
as weU as inhibiting, in media low in phosphate. 

Effects of Various Factors on the Inhibition of Succinate Dehydrogenase 

Very few illuminating studies on the modification of malonate inhibition 
are available. The effects of temperature were mentioned in Volume I, where 
the following thermodynamic parameters for the inhibition of beef heart 
succinate dehydrogenase at 38° were calculated: AF = — 6.26 kcal/mole, 
AH = — 5.48 kcal/mole, and AS = 2.6 cal/mole/degree. The K^^ for malonate 
was found to be 0.025 mM at 20o-23o and 0.041 mM at 38o (Kearney, 1957). 
The effects of osmolarity on the inhibition are surprisingly large (Honda and 
Muenster, 1961). Lupine mitochondria were prepared in media of different 
sucrose concentrations and assayed in media of two osmolarities (Table 
1-9). These results were obtamed on mitochondria and it is possible that 
the effects are not directly on succinate oxidase but on the permeability 
or structural properties of the mitochondria. In this connection it has been 
pointed out by Singer and Lusty (1960) that iV-methylphenazine measures 
the full activity of succinate dehydrogenase in mitochondrial fragments, 
but in intact mitochondria it measures only a fraction of the activity. 
Various ways of damaging the mitochondria lead to increased succinate 
dehydrogenase activity (such as increase in Ca++ concentration). This was 
interpreted in terms of permeability barriers to A^-methylphenazine (and 
also FMNHj), limiting the rate in intact mitochondria. It could also be 
explained on the basis of structural changes in the enzyme complexes. In 
any case, these observations demonstrate the importance of the mitochon- 
drial state in the functioning of succinate oxidase, and it would not be 
surprising if malonate inhibition were similarly sensitive. The inhibition 
of succinate oxidation by malonate in rat heart mitochondria in KCl me- 
dium is not altered by either halving or doubling the KCl concentration in 
the assay medium (Montgomery and Webb, 1956 b). However, the effects 



INHIBITION or SUCCINATE DEHYDROGENASE 



47 



Table 1-9 

Effect of Osmolarity in the Properties of Succinate Oxidase in Lupine 

Mitochondria " 



Preparation 


Assay 


K 


K- 






osmolarity 


osmolarity 
(M) 


(mM) 


(mM) 


KJKi 


a 


0.15 


0.22 


5.10 


0.91 


5.6 


2 




0.60 


12.34 


0.64 


19.3 


1 


0.40 


0.22 


2.87 


0.19 


15.1 


8 




0.60 


5.47 


0.16 


34.2 


191 


0.60 


0.22 


1.20 


0.05 


24.4 


36 




0.60 


5.47 


0.11 


49.7 


268 



" Kj„ is the Michaelis constant for succinate, iT, the inhibitor constant for malonate, 
and a is the interaction constant defined in Eqs. 1-3-5 and 1-3-6, indicating the type 
of inhibition. The osmolarity is given in terms of sucrose concentration. (From Honda 
and Muenster, 1961.) 

of Ca'^^ concentration on both succinate oxidation and malonate inhibition 
are complex (Fig. 1-9) and difficult to explain. The decrease in the rate 
of oxidation beyond Ca+"^ concentrations around 5 m.M might be attributed 
to a complexing of the succinate, but for the same reason, namely, the 
complexing of malonate, which has a higher affinity for Ca+''" than does 
succinate, the inhibition would be expected to decrease. Whether the 




0.001 01 



Fig. 1-9. Effects of Ca++ on the rate of succinate oxidation and 
the inhibition by malonate in rat heart mitochondria. Succinate is 
5 mM and malonate is 1 mM. (From Montgomery and Webb, 1956 b). 



48 1. MALONATE 

modest stimulation of succinate oxidation by low concentrations of Ca^^ 
is a permeability or structure-opening effect on the mitochondria, or due 
to a more direct effect on the enzyme, is not known; certainly the oxidations 
of other cycle substrates and pyruvate are strongly depressed by Ca++. 
The effects of Mg++ concentratioti from 6.2 to 12.4 mM on trypanosomal 
succinate dehydrogenase inhibition by malonate have been reported as 
negligible (Agosin and von Brand, 1955). In connection with the relation- 
ship between mitochondrial structure and malonate inhibition, it is interest- 
ing to examine the effect of ATP, inasmuch as ATP protects or stabilizes 
mitochondria after isolation from cells. ATP at 1 mM has very little effect 
on the rate of succinate oxidation in rat heart mitochondria (in five exper- 
iments a mean depression of 3%), but in the presence of succinate (5 
mM) and malonate (5 mM) it stimulated the rate 67%, thus antagonizing 
the inhibition by malonate (Montgomery and Webb, 1956 b). Similar re- 
sults were seen with acetylene-dicarboxylate inhibition. If the effect of 
ATP is to reduce the permeability to these inhibitors, it is surprising that 
interference with succinate penetration does not also occur. Thus some 
other explanation may have to be sought. 

Inhibition of Fumarate Reduction 

If succinate, fumarate, and malonate bind at the same site on succinate 
dehydrogenase, the reverse reaction — the hydrogenation of fumarate to 
succinate — should be inhibited by malonate; that is, malonate should 
compete with fumarate as well as with succinate. The relative potencies 
of the inhibitions on the two reactions would depend on the Michaelis 
constants for succinate and fumarate, so that the inhibitions would not 
necessarily be identical. Malonate was, indeed, found to inhibit the reduction 
of fumarate by horse muscle succinate dehydrogenase, but less potently 
than the oxidation of succinate (18 mM malonate required to inhibit 50% 
in the former case and 3.6 mM in the latter) (Das, 1937 b). A more detailed 
analysis was made by Forssman (1941) in Lund, using pig heart succinate 
dehydrogenase and leucomethylene blue as a hydrogen donor, in this case 
the inhibition by malonate being essentially equivalent for both forward 
and backward reactions. More recent studies of beef heart succinate dehy- 
drogenase, however, have given different values for K,: 0.025 mM for suc- 
cinate oxidation and 0.12 mM for fumarate reduction (Singer et al., 1956 a). 
This difference is unexpected and the suggestion was made that the binding 
of malonate may be effected by the state of oxidation of the electron trans- 
port components adjacent to the binding site. The affinities of fumarate 
for the enzyme were also found to be different for each reaction. It may also 
be of significance that iV-methylphenazine was the acceptor in the oxida- 
tion of succinate, and FMNH, or leucodiethylsafranin the donor in the re- 
duction of fumarate. 



INHIBITION OF SUCCINATE DEHYDKOGENASE 49 

The succinate dehydrogenase of Micrococcus lactilyticus behaves quite 
differently and is poorly inhibited by malonate, a ratio of (malonate)/(fu- 
marate) = 20 being required for 29% inhibition (Peck et al, 1957). It might 
be thought that this enzyme is not succinate dehydrogenase, but another 
enzyme that could be called " fumarate reductase," especially as fumarate 
is reduced at a faster rate than succinate is oxidized, in contrast to the 
mammalian enzymes. However, it has been conclusively demonstrated 
it is not a separate enzyme and that the failure of malonate to inhibit 
is to be attributed to a very high affinity for fumarate coupled with a rela- 
tively low affinity for malonate (Warringa et al., 1958). The configuration 
of the active center of the bacterial enzyme must differ from that of the 
mammalian enzymes. This may also explain the "fumarate reductases" 
obtained from yeast (Fischer and Eysenbach, 1937; Kovac, 1960) which are 
rather insensitive to malonate. 

Variations of Malonate Inhibition of Succinate Dehydrogenases 
from Different Tissues and Species 

The comparative biochemistry of enzyme inhibition is in its infancy and 
accurate comparison of results is usually impossible due to the different 
conditions under w^hich the inhibitions w^ere studied. Examination of Table 
1-6 with a view to establishing phylogenetic relationships is made difficult 
by the different types of preparation and assay procedure used. A correla- 
tion graph, made by plotting inhibitions against (I)/(S) ratios, shows a very 
marked scatter of the points. For example, at an (I)/(S) ratio of 0.1. the 
inhibitions range from 10% to 100%. This variation cannot all be due to 
the differences in technique. All of those cases in which the malonate inhibi- 
tion is significantly below the mean turn out to be in the bacteria, inverte- 
brates, or plants. However, this is not a strict correlation because some of the 
potent inhibitions have been found in such organisms {e.g., Azotobacter, E. 
colt, and Trypanosoma). The possibility of a relationship between succinate 
dehydrogenase type in the bacteria and the oxygen requirements for growth 
has been proposed. The enzyme from the obligate anaerobe Micrococcus 
lactilyticus has low affinities for succinate and malonate, as discussed in the 
previous section, whereas the enzyme from the facultative anaerobe, 
Propionibacterium pentosaceum is intermediate in properties between Mi- 
crococcus and the aerobic mammalian tissues (Singer and Lara, 1958). Cer- 
tainly many invertebrate and plant tissues can withstand anaerobiosis 
better than mammalian tissues, but at the present state of our knowledge 
such a correlation is dangerous to make. 

Comparisons of the malonate inhibitions of succinate dehydrogenases 
from different tissues have been reported in a few cases. The epithelium 
and muscle of guinea pig seminal vesicle were separated and the inhibitions 
by malonate at four different concentrations were similar (Levey and Szego, 



50 1. MALONATE 

1955). However, the inhibitions of epithelial dehydrogenase appear to be 
about 5% higher than for the muscle enzyme, although this may not be 
statistically significant. Malonate was found to inhibit Hepatoma 134 
tumor succinate dehydrogenase more than the enzyme from normal mouse 
liver (Fishgold, 1957) over a range of five malonate concentrations; for 
example, 0.21 mM malonate inhibited the liver enzyme 42% and the he- 
patoma enzyme 76% at a succinate concentration of 1 vaM. Killer and 
sensitive stocks of paramecia may have succinate dehydrogenases with 
different sensitivities to malonate, but it is difficult to draw conclusions 
from the data published (Simonsen and van Wagtendonk, 1956). The Og 
uptake of homogenates of the two strains was increased to different de- 
grees by 50 mM succinate and malonate inhibited both quite well (see 
accompanying tabulation). The authors concluded that the enzyme from 
the killer strain, is inhibited more, based on absolute reduction, but actually 

Increase in Oj uptake from succinate 



Sensitive strain Killer strain 



Control 1.4 14.0 

With malonate 80 mM 0.2 3.8 

% Inhibition 86 73 

the enzyme from the sensitive strain seems to be inhibited as w^ell. Our 
work with succinate dehydrogenase from various rat tissues (Table 1-6) 
indicates no significant difference in susceptibility to malonate. 

INHIBITION OF SUCCINATE OXIDATION 
IN CELLULAR PREPARATIONS 

Attention will now be turned to the inhibition of the succinate oxidase 
system when it is located in the normal cellular structure and succinate is 
added exogenously to the preparations. When succinate is added to most 
cell suspensions, minces, or slices, there is an increase in the O2 uptake, 
and this response is inhibited to varying degrees by malonate (Table 1-10). 
It is particularly important in cellular preparations to take account of the 
endogenous respiration and the effect of malonate on it (see Chapter 1-9). 
In many studies this has not been done and this is one factor that makes 
it difficult to compare accurately the malonate inhibitions m vitro and in 
vivo. Since the endogenous respiration is generally inhibited less than 
succinate oxidation by malonate, failure to correct for endogenous respi- 
ration usually leads to low values for the inhibition. This is illustrated in 



INHIBITION OF SUCCINATE OXIDATION 51 

the four examples given in Table 1-11. The importance of the endogenous 
correction is seen to vary with the effect of malonate on the endogenous 
respiration. When malonate inhibits the endogenous O2 uptake poorly, 
as in rat liver, or actually stimulates the O2 uptake (due to its metabolism), 
as in Euglena, the true inhibition of succinate oxidation is much higher than 
would be calculated simply from the data on succinate and succinate + 
malonate. 

Comparison of the inhibitions in Tables 1-6 and 1-10 for the same species 
or tissues, although this is qualitative only, shows that in several instances 
the inhibitory potency of malonate seems to be significantly less in cellular 
preparations. This is true for E. coli, Rhodospirilliim, Crithidia, Zygor- 
rhynchus, pigeon muscle, and rat liver. Moses (1955) points out that in 
Zygorrhynchus the inhibition of succinate oxidation in cell suspensions is 
very weak, even at the low pH of 3.4, but when the cells are treated with 
liquid nitrogen to destroy their structure, malonate inhibits normally. 
The oxidation of succinate by cell susj^ensions of Bacterium succinicum is 
not inhibited at all by 5 mM malonate whereas such oxidation in cell-free 
extracts is inhibited completely (Takahashi and Nomura, 1952). There are 
also several reports in which malonate was found to be ineffective but the 
extracted succinate oxidase system was not directly tested; however, in 
these cases one would certainly expect the enzyme to be sensitive to mal- 
onate. For example, malonate (2 mM) does not inhibit the oxidation of 
succinate by barley roots (Honda, 1957), or at 5-20 mM in dried cells of 
CJdorella (Millbank, 1957), while in beech roots malonate (28.6 mM) actually 
stimulates the rate of succinate oxidation (Harley and Ap Rees, 1959). 
However, in some cases comparable inhibitions have been observed in vitro 
and in vivo. Danforth (1953) showed in Euglena that malonate is quite ef- 
fective in intact cells if the pH is low enough (around 4.5) and, although it 
is difficult to compare the results with those obtained from homogenates 
because of different concentrations of succinate and malonate, it would 
appear that malonate is equally inhibitory in the two preparations. Similar 
effects of malonate were also observed in our work (Montgomery and Webb, 
1956 b) on rat heart slices and mitochondrial suspensions. 

The failure of malonate to inhibit the oxidation of added succinate in 
cellular preparations well, or at all, has usually been attributed to permea- 
bility factors. However, it is difficult to understand how permeability could 
explain these results, inasmuch as the penetration of both succinate and mal- 
onate would be controlled by the same factors, presumably. That is, if there 
is some barrier to malonate reaching the succinate dehydrogenase, how can 
succinate pass this barrier? It is true that the "pK,, for succinate is higher 
than for malonate but almost identical values of ^K,, have been reported; 
thus the distribution of the ionic forms around neutrality would be approxi- 
mately the same (Table 1-3). If only the uncharged forms of these acids — 



52 



1. MALONATE 



« 






<i) 



13 — 






m 



3 



3 



X 


-C 




^ 


^ 




T3 


T3 


,^ 


C 


c 


CD 


cS 


cS 


05 


S 


"a; 


^-' 


5S 


tc 


T3 


e8 


eS 




S 


S 


^ 


o- 


a 


t-i 



.S '^ 






O —< -* CD 
CO »c t^ 



i? 



r^ o --< o ic o 
CD '—I -H irj ic 



— H 






o 


O 


_o 


_o 










M 


CO 


'S 


'« 


c 


C2 




C 


« 


£ 


5 


OJ 


ft 


ft 


ft 


ft 


ai 


t» 


m 


CO 



M W MM 



S S 



5 e 



B3 



q; 



CQ 



INHIBITION OF SUCCINATE OXIDATION 



53 



o o o 
'^ 2 S 

I— I CO 



o o o 



P5 



- — • 




:3 




^"^ 


ft 


o 


'e 




^ 




"e 


-< 


« 


'«3 




T3 

c 




>« 






c 
o 




eS 




02 


c3 


cS 




to 




_c 


>■. 


cS 










tn 






cS 




p 




eS 


cS 


'3 


O 




m 




W 


ffi 


§ 


m oi 


o 


o 


GO 


O 


oo 


o >o IC 


iM CO 


00 


IM 


t> 


o 


M 


Tt< O 00 






a^ 



eo CD o 



CQ 



2Q 



I— I '^ 









c3 



0) o 



0) .^ 



4) <» 



ft ft 
O O 

'o _ 






I 



ft 
ft 



ft 
02 





s 






s 
■a 

1 


1 


6q 


Q. 



54 



1. MALONATE 



tf 









CM 



P^ P5 



W 



Oi CO -H Oi o 

t^ i> !> CO ;d 



-o 


73 


^ 


C5 


73 


^ 


c 


C 


^ 


^_^ 


C 


<u 


cS 


c3 


Q) 




c3 






s 


s 










o 


'? 




4^ 
0^ 


CO 

1 


to 




2 


O 
1-5 


IS 


-* 


(M 


CO 


t^ 


00 


t^ 


t- 


00 


(N 


CO 


CO 


IC 



o 



73 
C 
cS 

73 

O 



3 
(1h 



K g 



^ 



^- 



O 



1^ 




o 


eS 


J3 










■iJ 


-u 


t^ 


a 


ce 


P^ 


O 



S 




?5, 




."^ 






g 




3- 




'« 




£ 


o 




« 

tH 


S 


^ 


•§ 




!-) 


§ 


ci 


-§ 


+i 


^ 


a) 


CO 

o 




C/2 




to 
3 
o 
O 


C2s 
C3) 


-is! 
O 



INHIBITION OF SUCCINATE OXIDATION 



55 



pq 






lO CO 00 IC 
CO O ^ "O l> 

O^HfOCSClCDOOO 

(M 00 -^ -< — I 



-H Tj< O 



Ph 



o ^ 



o 





cS 


^-v 


f^ 


o 




fC 


T3 


Ci 


S 


^ 


eS 




SI 


T3 


c« 


W 


fe 



■75 'IS C5 






0) « 



(Xi a^ ^ lo 'Tit 
C5 :o t-^ uo C5 



CO tJ^ -Tt^ -^ 

c^ c^ c^ t^ 



t- c- t> 



g 


t^ 


4J 


-p 


c 


O 


'3 


<0 

> 


c3 




1 


CO 


m 


i^ 


W 


K 


M 


1 



^ 3 



O^ 2 



^ H 



^ T3 



J 1 



_ -t^ — , 



O" O" Ph O" i: W 



o 


o 


o 


i-O o 


t^ 


t- 


>o 


t- 


o 


o 




(M 


(N 


<N 


-O 


o 


'M 


cr; 


o 


^H 



^ S CC S M S 



<» c? 



W Ph Ph 



•- .SP 



-J^ 


>. 


r^ 


,0 






o 


_^ 




>w' 


cS 


S 



<t) OS 



56 1. MALONATE 



Table Ml 



Effects of Endogenous Respikation on Calculations of Malonate Inhibition 
OF Succinate Oxidation " 





Pigeon 
muscle 


Rat kidney 


Rat liver 


Euglena 


O2 Uptake 










Endogenous 


11.8 


14.3 


14.7 


11.4 


Malonate 


3.0 


4.5 


12.4 


13.9 


Succinate 


19.0 


58.5 


20.2 


28.0 


Succinate + malonate 


6.0 


15.8 


14.1 


19.7 


% Inhibition 










Endogenous respiration 


74.6 


68.6 


15.6 


Stim21.9 


Total respiration in the 










presence of succinate 


68.4 


73.0 


30.2 


29.6 


Succinate oxidation 


58.3 


74.4 


69.1 


65.1 



° The true inhibition of succinate oxidation is given in the bottom row, and is to 
be compared to the figures in the row immediately above where the correction for the 
endogenous ejRFect has not been made. The concentrations were: pigeon muscle — suc- 
cinate 1 vaM and malonate 10 iwM (Stare and Baumann, 1939); rat kidney — suc- 
cinate 20 n\M and malonate 20 mM (Fawaz and Fawaz, 1954); rat liver — succinate 
10 n\M and malonate 20 mM (Edson 1936); Euglena — succinate 10 mM and malonate 
10 ml/ (Danforth, 1953). 

HOOC — R — COOH — penetrated, succinate would enter cells somewhat 
better than malonate, but, at least at pH's above 7, it seems unlikely 
that this is the situation. Also, the entrance of enough of the undissociated 
caid to be effective would presumably decrease the intracellular pH signi- 
ficantly (see Chaper 1-14). Besides, some of these experiments in which 
malonate was inactive were done at low pH's (3.4 to 5.5). A question that 
must be considered is whether succinate oxidation is always entirely intra- 
cellular. It is possible that succinate oxidase occurs both in the mitochondria 
and in the plasma membrane. In many cases, malonate has little or no effect 
on tissue metabolism or function at concentrations capable of inhibiting 
the oxidation of added succinate completely. This is well seen in rat ven- 
tricle slices (Webb et al, 1949) where the succinate may be oxidized at 
the cell surface, and it is interesting that in this tissue the potency of mal- 
onate in vivo and in vitro is the same. However, in most instances, the 
succinate oxidase seems to be protected in some manner in intact cells. 
There are several possible factors that could modify the malonate inhi- 



INHIBITION OF SUCCINATE OXIDATION 57 

bition of succinate dehydrogenase when the enzyme is isolated from the 
cells. It will be well to mention some of these in order to emphasize that 
there are usually many ways, other than by permeability, by which unex- 
pected phenomena may be explained. 

(a) The enzyme environment within the cell is different from the artificial 
media used with isolated enzymes (see Chapter 1-9). Many substances may 
be able to alter malonate inhibition; we have noted the effects of Ca+"'" 
and ATP. The concentrations of such substances may be different in cell 
and medium. The intracellular pH is also not that of most media used in 
enzyme study and may be easily changed by the addition of external 
substrate or inhibitor. 

(b) The addition of succinate to cells may influence the endogenous respiration; 
that is, the change in Og uptake upon adding succinate may not represent 
accurately the rate of succinate oxidation. In other words, the correction 
for endogenous respiration may be in error. Also, the addition of succinate 
or malonate can secondarily alter the complex balance of the metabolic 
systems. For example, oxalacetate is a very potent inhibitor of succinate 
dehydrogenase and its concentration in the cell may be a controlling factor 
in the operation of the cycle. Since oxalacetate can be formed from succin- 
ate, and its formation inhibited by malonate, secondary changes in succinate 
dehydrogenase activity may occur that are easily interpreted as due to 
the direct effects of malonate. 

(c) The concentration of succinate in the cell may already be appreciable 
and the addition of more may reduce the inhibition by malonate because 
of the competitive nature of the inhibition. 

(d) The rate of succinate oxidation may be limited by the rate at which it 
can enter into the cells or tissus; that is, the succinate oxidase is so active 
that the succinate is oxidized as rapidly as it enters. In such a case, mal- 
onate would be quite ineffective until the enzyme is inhibited sufficiently 
to make it limiting. 

(e) The ratio (malonate) I (succinate) may be different in the cell than in the 
medium. This could be due to differences in the permeabilities of the cell 
membrane towards these substances for, despite the fact that succinate and 
malonate have similar properties, many cases of differential permeabilities 
to more closely related ions are known. It could also be due to the somewhat 
different p/C^ values, since the internal concentrations of the active ions are 
determined by the ionization constants (see Eq. 1-14-146 for buffered 
cells). 

Whether these factors, or others, are responsible for the anomalous results 
mentioned above is not known. It would be very useful to have data on the 



58 1. MALONATE 

rates of penetration of succinate and malonate into cells, obtained prefer- 
ably with radioactive material and under the same conditions. Quantitative 
studies on the malonate inhibition of intracellular dye reduction resulting 
from succinate oxidation might also be informative in certain instances. 
It has been shown that malonate inhibits the succinate-induced reduction of 
neotetrazolium in adipose tissue cells (Fried and Antopol, 1957), but noth- 
ing otherwise is known about the succinate dehydrogenase from this tissue. 

Competitive Nature of the Inhibition in Cellular Systems 

It has been shown in several tissues that the addition of succinate will 
reverse the inhibition of respiration produced by malonate. Thus in Avena 
coleoptile the inhibition by 50 mM malonate is reduced from 57.4% to 
25.8% upon adding succinate (Bonner, 1948), and in spinach leaves from 
75.4% to 20.5% (Bonner and Wildman, 1946). In chick embryonic carti- 
lage, the depression of respiration by 10 mM malonate is reversed by 100 
mM succinate but not by 10 mM (Boyd and Neuman, 1954). Such results 
have occasionally been stated to prove the competitive nature of the inhibi- 
tion but this reasoning is not completely valid. The mere increase in O2 up- 
take seen on addition of succinate to malonate-inhibited tissues is alone 
not evidence for competition. The effects of succinate on uninhibited tissue 
must also be tested and it must be shown that the actual inhibition is de- 
creased. A decrease in the inhibition brought about by increasing succinate 
concentrations has indeed been reported in two tissues, pigeon breast muscle 
(Krebs and Johnson, 1948) and the trypanosome Crithidia (Hunter, 1960) 
and in the latter a true competitive inhibition was demonstrated by 1/v — 1 /(S) 
plots. The data are given in Table 1-10. It is probable that the inhibition 
of succinate oxidation in cellular systems by malonate would frequently 
not obey strictly competitive kinetics, due to the various complexities that 
arise, as discussed in the previous section, even though the primary inhi- 
bition on the succinate dehydrogenase were competitive. Some of the prob- 
lems involved in the determination of the type of inhibition in cells have 
been discussed in Chapter 1-9. 



INHIBITIONS OF ENZYMES 
OTHER THAN SUCCINATE DEHYDROGENASE 

It is very important to establish the degree of specificity that may be 
achieved in the use of malonate under various conditions. To this end 
we shall first discuss the direct evidence for the inhibition of enzymes 
other than succinate dehydrogenase, and then proceed to the effects on the 
operation of the tricarboxylic acid cycle, the accumulation of succinate and 
other intermediates, and finally the antagonism of the malonate inhibition 



ENZYMES OTHER THAN SUCCINATE DEHYDROGENASE 59 

by fiimarate. We shall then be in a position to evaluate the specificity of 
malonate. There are other reasons, of course, for taking up these subjects; 
for example, malonate is frequently used to block the cycle in living tissue 
and in this connection it is essential to understand how malonate can alter 
cycle activity under various conditions. 

Some effects of malonate on miscellaneous enzymes are presented in 
Table 1-12. There are three major difficulties in the establishment of the 
over all spectrum of action of malonate. First, there are many quite impor- 
tant enzymes whose response to malonate has never been investigated 
directly. Second, inspection of the table will show that in few cases has 
more than one concentration been used, and often the single concentration 
reported is either too high or too low to be of much value. Third, the same 
enzyme from different species often shows widely varying susceptibility 
to malonate (e.g. NADH oxidase, /5-glucuronidase, lactate dehydrogenase, 
oxalacetate decarboxylase, and others in the table), making it clear that 
there are many different spectra of malonate inhibition and that statements 
on specificity must be qualifield by naming the source of the enzymes in 
question. 

Ideally, the results on the inhibition of an enzyme by malonate should 
be given for several concentrations, preferably convering the range from that 
concentrations just sufficient to produce some inhibitions, through that 
causing approximately 50% inhibition to higher inhibitions (unless these 
latter concentrations are unreasonably high). Another way of looking at the 
problem is to consider that range of malonate concentrations most likely 
to give useful information when tested on enzymes. This will depend 
on the source of the enzymes. For example, in mammalian tissues it usu- 
ally requires malonate concentrations between 2 and 5 mM to inhibit suc- 
cinate dehydrogenase around 90% in the presence of 5-10 mM succinate. 
In the interests of establishing the degree of specificity, it would thus be 
most important to test the effects of malonate at concentrations around 
5 mM on enzymes from such sources. When the organism studied possesses a 
succinate dehydrogenase less sensitive to malonate, correspondingly higher 
concentrations must be applied to the other enzymes. 

Instances of Competitive Inhibition 

Until it is time to discuss the matter of specificity, there is little to say 
about these inhibitions since the results in the table speak for themselves. 
It is evident that several enzymes other than succinate dehydrogenase 
are readily inhibited. One would not be surprised if enzymes attacking the 
dicarboxylate anions, where the carboxylate groups are separated by two 
carbon atoms, were inhibitable by malonate to some extent, since it is 
likely that these enzymes also possess cationic groups appropriately spaced. 
Actually, fumarase, malate dehydrogenase, the malic enzyme, oxalacetate 



60 



1. MALONATE 



P5 






W 



^H ^ 



^ .5 C 






m 



o 



^ iln c3 



^ lO ^ 



^^="•13 



C ^ > 



c 


_ 


03 


S 

S 


O! 




S 


2 


>. 


"o 


§ 


O 



4j :_, 



S <D 5 ■:S ^ 



~ .22 



fe 5 CO 



3 o 



-5 * 

t/3 W N 



o 
o 


3 




g 

•S 


>> 

C 

15 


2 




> 


^ 
"o 

2 

-li 


-§ 


O 


k< 




^ 


IS 


Ld 
Q 


^ 


'JS 


so 




(0 
> 

3 


^ 
^ 




P5 


P^ 


05 


P5 






Q 



y *- 



G t« .5 



2 'O 



'x 


05 




o 


cS 


<D 


T3 


r2 








"S 


o 


ft 




cS 





'x 




ft 


o 



-.SO) 



OJ 03 03 OJ C r- 



X S G c ^ 



<j <ii 



!- O uj 

O O Q 






o 



o o o o o 



G 







^ 


(B 




a 


03 




03 


cS 




O 


3 


0^ 


-0 


'x 


03 


ft 


12 




o 


^ 


» 


'x 


<© 



P fM 



^ 1 



fe5 ^ CM 



ENZYMES OTHER THAN SUCCINATE DEHYDROGENASE 61 







OS 














2 








1—4 
1— I 


„_^ 




>> 








^— ' 


00 




> 








b 


Tt^ 




> 








« 


c: 




o 








J3 


i-H 




J 




^.^ 




O 


'^-^ 








c^ 


^— V 


:S 


2 


05 


s 




o 








T3 


s 


1 - 

is 


3 

a 


<H-I 

ft 

3 


C 
M 



T! 
>> 
jS 



5^ S 
CO >o 

OS OS 



2 X 






:r; c o 







■* 






lO 






OS 














^-v 


r- 




t^ 


tJ< 


fl 


iC 


OS 


o 






-»^ 








"-' 


— ^ 






e 




fe 




t3 


«^ 


"S 


s:; 


ilJ 


U-i 


c3 


cS 


& 


^ 


TS 


r^ 


V 






4^ 


O 


5 


^ 


o 


M 


S 



-•3 2- ;3n-S 4) =;o'- 






ic ^ ^^ ec 

X X 






~ ^ s cs .5 



cS a> o « o 3 7-^^ O o > =1 .r? .Sf =c c3 a> ^2 v- ^ 



® 


c3 




c 


cj 


« 


« 


"i 


00 


^ 


C 


T3 


ci 


>. 



o ^ o ^» o o 



3 




c3 


05 






>> 




c3 


p 




3 






_c3 

o 

5 


O 
>> 


X 


2 

CD 


-a 
>, 

X 


p 
>■, 

"S 

-^3 


cS 

u 
■*^ 

O 

CO 

1— 1 



<^ ^ 



g +1 g Ph o 

3 3 rt O cS 

O 3 -li tn -2 

3 cS 3 «U O 

-3 +j— .CJO ^'Q.,^,_^_ocj .-J 



62 



1. MALONATE 



tf 



-^ -^ o 



J2 W) 
O Z 



■r! 


-Q 


>jj 


C3 




t^ 




C5 




CO 


3 




,__ 


C5 






0^ 




2 


CO 







cS 


O 




03 


s 


3. 


^ 


C^ 


(S 


'S" 


a 


Q 


>^ 


o 



^ So 
O cS iS 



M Q C 



A A 



o o o o 

-M Tt 00 Oi 



CD O O CO O O — H 
rt -H C<l «N CO lO ^ 



1^ 



K 



O K-l 



i^ ^ S 



^ 



* -, - 
CO S i: 

?5 ^ e 



•»^ O V. 



^ ^ 



• ^ " (H 



c 


C 


« 


o 


o 


-=5 


(» 


Oj 




M 


tuD 


M 








£ 


e^ 


S 



S cS 



ft 



ENZYMES OTHER THAN SUCCINATE DEHYDROGENASE 



63 



o 2 2 



73 



S 2 2 ^ 

-H ^ "^ Ph 






ffi - -7- 



m 






P- ^ 



m 



c ^ 5 



-a rt 



^ ^ t; 






«: ^ o 



^ 



?^ r- r- rH > > 



Si o 



o 



2 

73 


1 


1^ 
>. 


4> 

CO 

a 




>-. 


">. 




:2 


ffl 


o 


X 

o 




o 


_cS 

13 


^ 




£ 


Q 


Is 


-c 


c« 


<i 


X 


§ 


^ 


§ 


"^ 


o 



Ph CM eu 2^ si 



C5 |l4 



i-H 03 



■^ .2 



« c § 2 



o 2 



9 c- c 



;j S O 



lO O -H ^H ,—( ^H 






p. & S C tj) 



- -«^^a4^ 



S « — =s 



!;^ p^ 



X -S -ii 



o c3 3 3 CZ2 c:c CO 

s Pi w a ^ ^ ^ 



« 3 



3 5 






O fc 



-d 




>> 




^ 




0) 




S 








'S 




93 


S 


S 


OQ 


_>;. 


e3 




S 


■fci 


« 


o 


M 




O 


03 


73 


O 


>. 


-C 


-C 


Ph 


^ 



64 



1. MALONATE 



cj \2 



t— I « HH 



tf 






fe 



E^ 



5" S 



S P5 



O PQ 



e «5 



-^ .S 



(1-1 j!^ '-1 






=« .2 



11^ eiH 



-c ^ ju 



m 





^►, 




;j 




<D 


£2 


^ 


4^ 


73 


c3 





IB 


)3 


J3 


1 




,4^ 


-ii! 


s 












p M) 



H 



-C3 






y, g 



cs rt 



o o 



22 C 5 ^ 



-^ 


a 


a 






oi 




_c 


aj 




-t^ 


'c 


<33 


9J 


-u 


£ 


ft 




< 


M 



;^ p 



ENZYMES OTHER THAN SUCCINATE DEHYDROGENASE 



65 



decarboxylase, and the condensing enzyme are inhibited by malonate, al- 
though usually not as potently as is succinate dehydrogenase. Enzymes 
catalyzing reactions of the dicarboxylates in which the charges are farther 
apart (e.g. a-ketoglutarate) would be expected to be less susceptible. It is 
likely that these inhibitions are mosth^ competitive but sufficient data to 
establish this are generally lacking. 

The inhibition oifumarase by malonate has been shown to be competitive 
in the thorough study of Massey (1953 b) but the affinity of the enzyme for 
malonate is not very high {K^ = 40 n\M). This implies either a very dif- 
ferent intercationic distance in this enzyme from that in succinate dehydro- 
genase or a different configuration of the enzyme surface surrounding these 
cationic groups, probably the latter. Both directions of the reaction cata- 
lyzed by the malic enzyme are inhibited by malonate, which competes 
with either malate (Stickland, 1959 b) or p\Tuvate (Stickland, 1959 a) (see 
accompanying tabulation). The inhibition of this enzyme might well alter 



Malate 


% Inhibition by malonate at: 


Pyruvate 


% Inhibition by malonate at: 


(mif) 


2 mM 


5 mM 


20 mM 


{mM) 


1 mM 


10 mil/ 


0.1 


71 


88 


100 


1 


43 


62 


0.3 


48 


74 


91 


10 


37 


70 


1 


19 


48 


84 


50 


10 


46 



the operation of the tricarboxylic acid cycle under certain conditions. The 
situation is different for lactate dehydrogenase, since malonate is competitive 
with respect to lactate but noncompetitive with respect to pyruvate (Ot- 
tolenghi and Denstedt, 1958), leading to the suggestion that these two sub- 
strates react with different sites on the enzyme. The X/s are 6.4 m.M for 
the oxidation of lactate and 27 mM for the reduction of pyruvate. Oxalate 
and tartronate are much better inhibitors of this enzyme. The D-a-hydroxy 
acid dehydrogenase of yeast, which oxidizes lactate, malate, a-hydroxybuty- 
rate, and glycerate, is competitively inhibited by malonate with a K^ of 
0.9 mil/ (Boeri et al., 1960). In this case, oxalate is a very potent inhibitor 
{Ki = 0.0025 mM) while tartronate is of similar potency to malonate. 
Finally, in reducing systems in which succinate can serve as an electron 
donor, malonate maj^ inhibit competitively. This is the case for the particu- 
late nitrate reductase of soybean root nodules, the K^ for malonate being 
0.017 milf (Cheniae and Evans, 1959). The nitrate reductase is not inhibited 
by malonate directly but the results on the over aU system might make it 
appear to be the case. In all cases where malonate inhibits competitively, 
the susceptibility of the reaction in complex systems will depend on the 
concentration of the substrate, and thus may be quite high in living systems 
where the concentrations of intermediates are frequently low. 



66 1. MALONATE 

Inhibition Due to Chelation with Metal Ion Cofactors 

Many enzymes are dependent on dissociable metal ions for their activity 
and the operation of most of the important metabolic systems thus requires 
the presence of these cofactors. The list of enzymes requiring Mg++ is a 
long one and includes the oxidases and decarboxylases for the keto acids, 
most of the enzymes involved in phosphate metabolism (e.g., the kinases, 
the transphosphorylases, the phosphatases, and the acyl — CoA synthetases), 
some dehydrogenases e.g., phosphoglucose dehydrogenase, phosphogluco- 
nate dehydrogenase, and isocitrate dehydrogenase), some peptidases, 
phosphoglucomutase, and enolase. Several enzymes require Zn++, such as 
lactate dehydrogenase, glutamate dehydrogenase, alcohol dehydrogenases, 
carboxy peptidase, and carbonic anhydrase. Since malonate is able to chelate 
effectively with these metal ions, inhibition may result from the reduction 
of metal ion concentration in the medium or the removal of the metal 
ions from the enzyme. The ability of malonate to inhibit by this mechanism 
will depend on the affinity of the enzyme for the metal ion. The binding 
of Zn++ to enzymes is usually rather strong and it is difficult for malonate 
to deplete the enzyme of this metal, but Mg++ is more loosely bound in most 
cases and the activity of enzymes dependent on it is generally related to 
the Mg+'*" concentration in the medium. The effect of malonate on Mg++- 
dependent enzymes will thus depend on the concentrations of Mg++ and 
malonate, and on the relationship between enzyme activity and Mg+*^ 
concentration. In the use of malonate, especially at higher concentrations, 
it is imperative to consider the possibility of such effects. It is likely that 
some of the inhibitions in Table 1-12 are due to metal ion depletion. 

Inhibition by reaction with an activator was discussed briefly in Chapter 
1-3 and it was seen that in the general case no simple expression for the 
inhibition is possible. Nevertheless, it should be reasonably easy to deter- 
mine if the inhibition is purely the result of activator depletion, since the 
concentration of free activator can be calculated from the dissociation 
constant of the activator-inhibitor complex by an equation similar to 
Eq. 1-3-72: 

(Mg) = ^ V [(I,) - (Mg,) + K-\- + 4(Mg)if - -^ [(I,) - (Mg,) + K] (1-3) 

If the effect on the enzyme is only to reduce the Mg++ concentration, the 
addition of malonate should bring the activity to that value corresponding 
to the reduced free Mg++. Another possibility is that the Mg-malonate 
complex is the active inhibitor, in which case the kinetics should be investi- 
gated with the calculated concentrations of this complex at different con- 
centrations of Mg++ and malonate. 

The most thorougly studied instance of the possible relationship of Mg++ 
to inhibition by malonate is the work on the utilization of oxalacetate 



ENZYMES OTHER THAN SUCCINATE DEHYDROGENASE 



67 



in rat tissue homogenates by Pardee and Potter (1949). The formation 
of citrate here probably involves decarboxylation of some of the oxal- 
acetate to pyruvate, with subsequent condensation to enter the cycle. Al- 
though a single enzyme was not studied, it is possible that the utilization 
of oxalacetate was limited by the decarboxylase. This reaction is activated 
by low concentrations of Mg++ and inhibited by higher concentrations 
(upper curve in Fig. 1-10). It is clear that the inhibition by malonate de- 




( Mg* • ). 



30 
mM 



Fig. 1-10. Utilization of oxalacetate by a rat 

kidney homogenate at diflferent concentrations 

of Mg++, with and without malonate. (From 

Pardee and Potter, 1949). 



creases as the Mg"*"*" concentration is raised (lower curve). This was interpret- 
ed to mean that the inhibition is mainly due to depletion of Mg++. Some 
arguments may be brought against this interpretation. Table 1-13 gives 
my calculations of the concentrations of free Mg^^, free malonate, and the 
complex at the different levels of total Mg++ used in the experiment shown 
in Fig. 1-10. It is seen that the malonate at 10 mM does indeed reduce the 
free Mg++, but in the higher range of Mg++ concentrations this should in- 
crease the activity rather than decrease it, because in this range Mg"'"+ is 
somewhat inhibitory. Pardee and Potter did not consider the reduction in 
free malonate concentration, which is very marked as shown in Table 1-13. 
This could well acount for the decrease in the inhibition from 47% at zero 
Mg++ concentration to 12% at 30 roM Mg++. One might speculate that the 
reduction in rate at high Mg++ concentrations could be due to the complexing 
of oxalacetate so that it is unable to react with the enzyme. The question 
of the mechanism of the malonate inhibition is thus not settled. It was 
claimed that the inhibition is not typically competitive because increase 
in the oxalacetate concentration actually increases the inhibition somewhat, 



68 1. MALONATE 



Table 1-13 



Effects of Total Mg++ Concentration on the Concentrations of Free Mg++, 
Free Malonate, and Ms-Malonate Complex " 



(Mg++,) 


(Mg++) 


(Mg 


-malonate) 


(Malonate) 


(milf) 


(mif) 




(mif) 


(mM) 













10 


3 


1.59 




1.41 


8.59 


10 


6.13 




3.87 


6.13 


20 


14.09 




5.91 


4.09 


30 


22.98 




7.02 


2.98 



" Conditions as in Fig. 1-10 from the work of Pardee and Potter (1949). The total 
malonate concentration is 10 mM in all cases. The values were calculated from Eq. 
1-3 using 9.77 X 10^^ M for the dissociation constant of the Mg-malonate complex. 

although no figures were given so the magnitude of the effect is unknown. 
An increase of oxalacetate might reduce the amount of Mg++ bound to 
malonate and thereby increase the free malonate concentration. It would 
appear unlikely that the Mg-malonate complex is inhibitory since its con- 
centration increases with Mg++ concentration (Table 1-13), whereas the 
inhibition decreases. However, it is evident that malonate inhibits this 
enzyme system in some manner directly, not only from the above consid- 
erations but also because of the marked inhibition observed in the absence 
of Mg++. 

Dialkylfluorophosphatase is activated by Mn++ and its inhibition by 
malonate was attributed by Mounter and Chanutin (1953) to the chelation 
of the activator. Since no concentrations of either Mn++ or malonate were 
given, it is impossible to evaluate the results. However, their data show 
that malonate inhibits just as well, if not better, in the absence of Mn++ 
(21% with Mn++ added and 28% without Mn++ at 15 min). These results 
might better be interpreted as due to removal of free malonate by the 
Mn++. If Mn-malonate is incubated with the Mn-free enzyme, activity 
slowly appears, indicating that the Mn++ is transferred from the malonate 
to the enzyme, which is not surprising since the affinity of the enzyme for 
the Mn++ is much greater, as shown by the dissociation constants {pK for 
enzyme-Mn++ complex is 7.7). It would be very surprising under these 
circumstances if malonate were able to reduce the Mn++ sufficiently to 
inhibit the enzyme. 

There are several other claims for this mechanism of malonate inhibi- 
tion but in all cases there is either inadequate evidence or no evidence at 



EFFECTS ON TRICAKBOXYLIC ACID CYCLE 69 

all. The examples discussed above indicate the impossibility of establishing 
such a mechanism without considering the changes in malonate concentra- 
tion or treating the data quantitatively. Despite this lack of positive evi- 
dence, there is certainly no doubt but that this type of inhibition can occur 
and may be sometimes very important. In the oxidation of pyruvate by 
rat heart mitochondria, the Mg++ concentration must fall below 1 mM 
before there is any significant decrease in the rate (Montgomery and Webb, 
1956 b). We usually used 5 mM Mg++ in the medium so that it would have 
required at least 40 mM malonate to produce a detectable inhibition by this 
mechanism. Malonate at 50 mM did indeed inhibit around 50% but this 
must certainly be due to other actions to a large extent. These experiments 
were done with the a-ketoglutarate oxidase blocked by parapyruvate so 
that any inhibition of succinate oxidation would not be involved. 

EFFECTS OF MALONATE ON THE OPERATION OF THE 
TRICARBOXYLIC ACID CYCLE 

Malonate is usually assumed to produce its major effects on cellular meta- 
bolism and function by disturbing the operation of the cycle* and reducing 
the rate of formation of ATP. Malonate has often been used to establish if 
the cycle is operative in a tissue or if a particular functional activity is 
dependent on the cycle. It is thus important to examine critically the 
nature of the cycle block and the effects it may have on the over-all oxida- 
tive metabolism. There are two aspects that are especially relevant to this 
question. There is the problem of the specificity of action of malonate on 
succinate dehydrogenase and this will be considered later. In the present 
section we shall assume that the only inhibition is on the oxidation of suc- 
cinate and discuss the problems relative to the interpretation of such 
a block. Before treating the actual results obtained with malonate, the 
nature of the cycle and its responses to inhibition will be outlined. 

Some General Principles of Cycle Block 

The primary function of the cycle is to incorporate and oxidize acetyl-CoA, 
whether this arises from pyruvate, acetate, fatty acids, or elsewhere, and 
thus it is particularly important to discuss the effects of malonate block on 
this. The situation is relatively clear in suspensions of isolated mitochondria, 
in which the concentrations of the cycle intermediates are low and the endo- 
genous respiration is generally negligible. Pyruvate, or other substances 
giving rise to acetyl-CoA, may be oxidized through the cycle only if some 

* In this chapter the term "cycle" will always refer to the tricarboxylic acid cycle 
for convenience and other cycles will be designated by their special names. 



70 1. MALONATE 

cycle intermediate (sparker) is provided to furnish oxalacetate to condense 
with the acetyl-CoA. A very small amount of such a sparker may suffice to 
initiate the entry of the acetyl-CoA into the cycle, and the cycle will then 
perpetuate itself through the continuous formation of oxalacetate, in which 
case pyruvate will be completely oxidized to COg and water. Such a system 
should be quite sensitive to malonate, because an inhibition of the oxidation 
of succinate will reduce the amount of oxalacetate formed and consequently 
the amount of acetyl-CoA entering the cycle. On the other hand, if an 
approximately molar equivalent of fumarate, malate, or oxalacetate is 
initially present with the pyruvate, there wiU be an adequate concentration 
of oxalacetate to incorporate acetyl-CoA at a rapid rate, and the process will 
not depend on a regeneration of oxalacetate. This system will not be very 
sensitive to malonate, because a block of succinate oxidase will not apprec- 
iably reduce the amount of oxalacetate present. The first important princi- 
ple is, therefore, that the degree of cycle inhibition by malonate will depend 
on the source of oxalacetate. 

When the cycle is operating in a steady state, the concentrations of 
intermediates are low, and oxalacetate is formed just as rapidly as pyruvate 
is incorporated, the cycle rate being limited by the entry of acetyl-CoA, 
This may be the normal state of the cycle (Krebs and Lowenstain, 1960) but 
probably in cells, and certainly in isolated preparations, there are times when 
the cycle is not in a steady state. There is an initial rise in citrate concentra- 
tion during the oxidation of pyruvate by heart mitochondria in the presence 
of malate (Montgomery and Webb, 1956 a), indicating that the tricarboxy- 
lates cannot be handled as rapidly as pyruvate can enter the cycle when the 
supply of oxalacetate is not limiting. The rate of oxygen uptake is initially 
very high, falls to a new level during the first 40 min, maintains this level 
for 2-3 hr, and then suddenly fails when the pyruvate is completely utilized. 
The first phase occurs when oxalacetate is readily available, and corresponds 
to the accumulation of citrate; the rate of oxygen uptake during this period 
is not an accurate measure of the rate of operation of the entire cycle — 
block of succinate oxidation may have very little effect on the oxygen up- 
take because relatively little of the respiration arises from this region of the 
cycle. The second steady-state phase should be more sensitive to malonate 
because the oxidations of succinate and malate now contribute 2 of the 
total of 5 oxygen atoms taken up per molecule of pyruvate. The second 
principle is thus that malonate inhibition will sometimes depend on the time 
interval during which the oxygen uptake is measured, particularly whether 
it is the initial rate or the total oxygen consumed. 

Succinate usually accumulates during malonate inhibition (see page 90) 
and this will progressively reduce the degree of inhibition due to the com- 
petitive nature of the inhibition. Eventually a new steady state may be 
reached during which the succinate concentration remains constant. This 



EFFECTS ON TRICARBOXYLIC ACID CYCLE 71 

will always tend to lessen the effect of malonate and under certain circum- 
stances it might effectively overcome the inhibition. The third principle 
is that the degree of malonate inhibition will depend on the level of suc- 
cinate accumulation in the system studied. 

Oxalacetate can often be formed by reactions outside the cycle. Pyruvate 
and phosphoenolpyruvate can be carboxylated to oxalacetate in the presence 
of oxalacetate decarboxylases or oxalacetokinase (Bandurski, 1955), and 
transamination between a-ketoglutarate and aspartate may also give rise 
to oxalacetate. In such cases the inhibition of pyruvate oxidation in the 
cycle by malonate will be reduced because the incorporation of pjTuvate 
will not be dependent only on the regeneration of oxalacetate (Holland and 
Humphrey, 1953). Such reactions may occur in isolated mitochondria, as 
well as in cells, since in heart mitochondria, where pyruvate alone is not 
oxidized at all, the presence of bicarbonate or COg allows a substantial rate 
of pyruvate oxidation (Montgomery and Webb, 1956 a), presumably 
through the carboxylation of some of the pyruvate to oxalacetate. The 
fourth principle is that the degree of malonate inhibition will depend on 
noncycle sources of oxalacetate. 

Alternate metabolic pathways involving cycle substrates or intermediates 
may occur in some tissues. There are many opportunities for the metabolism 
of pyruvate, in addition to its oxidation through the cycle, and in the pres- 
ence of malonate these pathways may become important. This is particu- 
larly true in microorganisms but the ability to decarboxylate pyruvate to 
acetate is common to most species and tissues. Thus, in the presence of 
high concentrations (50 mJf ) of malonate, pyruvate is quantitatively trans- 
formed into acetate by rabbit heart mitochondria (Fuld and Paul, 1952). 
An alternate pathway for succinate that would circumvent a malonate 
block is the cleavage of succinate (in the presence of NADH, CoA, and 
ATP) to 2 acetyl-CoA molecules. This succinate-cleaving enzyme was dis- 
covered in Tetrahymena (Seaman and Naschke, 1955) but it is also active 
in several rat tissues and in certain bacteria. This reaction will not, of course, 
restore cycle activity but it can lead to the formation of acetate or other 
products from acetyl-CoA, as well as reduce the concentration of succinate. 
Finally, the recently delineated glyoxylate cycle (Kornberg and Krebs, 
1957) could bypass that region of the cycle containing succinate oxidase, 
malate being formed from isocitrate through the condensation of glyoxylate 
and acetyl-CoA, the over-all process being the formation of succinate from 

2 PjTuvate + 3/2 O2 -> succinate + 2 CO2 + HgO 

pyruvate. This shunt would allow a greater utilization of pyruvate and a 
greater oxygen uptake in the presence of malonate than would be the case 
with the tricarboxylic acid cycle alone. The glyoxylate cycle has been 
found in many microorganisms and there is some evidence for its occurrence 



72 1. MALONATE 

in certain plants, but its role in animal tissues is as yet unknown (Krebs and 
Lowenstein, 1960). As a result of these considerations, the fifth principle 
of cycle block is that the degree of inhibition by malonate will depend on 
the activity of various alternate pathways and shunts; in addition, it will 
depend on what is measured, e.g., oxygen uptake, CO2 production, or 
pyruvate disappearance. 

There is no doubt, therefore, tha the operation of the cycle and any 
ancillary pathways will vary with the experimental or physiological condi- 
tions, and that one must expect marked differences in the behavior of the 
cycle in different species or tissues. In addition to the factors discussed a- 
bove, there are several other reasons for variability in response to malonate; 
the different susceptibilities of succinate dehydrogenase to inhibition (see 
page 49), the failure of malonate to penetrate readily into cells, and the pos- 
sibility that malonate can inhibit other enzymes. The reliability of malonate 
as an indicator of cycle activity in a tissue must be evaluated in the light 
of these considerations. Certainly the lack of an expected response to mal- 
onate cannot be immediately interpreted as indicating the absence of the 
cycle, and the production of a significant effect by malonate should be 
substantiated by other more direct evidence before the operation of the 
cycle is established. 

Inhibition of Cycle Substrate Oxidation by Malonate 

A summary of some of the effects of malonate on cycle oxidations is 
given in Table 1-14. In many cases the concentration of malonate is too 
high to act specifically on succinate dehydrogenase, and the results are to 
some extent meaningless. It is very difficult to make any generalizations 
but malonate concentrations above 10 mM in subcellular preparations must 
be looked upon as probably not completely specific, whereas in cellular 
systems it is impossible to evaluate the specificity because the intracellular 
concentration is not known. In attempting to interpret the inhibitions 
observed, it is often necessary to know the pathway of metabolism of the 
substrate and how much oxygen is normally taken up per molecule utilized. 
For example, when a-ketoglutarate is added to a mitochondrial suspension, 
it may be oxidized to fumarate (or malate) taking up 2 atoms of oxygen, 
or to oxalacetate taking up 3 atoms of oxygen, or completely taking up 8 
atoms of oxygen. If succinate oxidation is completely blocked by malonate, 
only 1 atom of oxygen will be taken up. Thus the maximal inhibitions in 
these three cases are 50%, 67%, and 87.5%, respectively. Similar reasoning 
applies to each substrate and in many cases the exact fate of the substrate 
is not known so that it is difficult to estimate what effect might be expected 
from malonate. It is also important in this connection, as pointed out in 
the previous section, to distinguish between inhibition over an initial short 
period of oxidation and inhibition of the total oxygen uptake. 



EFFECTS ON TRICARBOXYLIC ACID CYCLE 73 

(a) Inhibition of pyruvate oxidation. The oxidation or disappearance of 
pyruvate in cellular preparations is usually not depressed very much by mal- 
onate at concentrations less than 10 miH , whereas in mitochondrial prepara- 
tions the expected degree of inhibition is usually observed. This may be 
partly explained by poor penetration into the cells and partly by the alter- 
nate pathways that may reduce the importance of the cycle. One type of 
correction that can be applied for a more accurate determination of cycle 
inhibition by malonate is that used by Speck et al., (1946). In malarial 
parasitized erythrocytes, pyruvate is oxidized without the appearance 
of acetate, but in the presence of malonate, some acetate in formed. Cor- 
rection was made for that pyruvate that went to acetate, since this fraction 
of the pyruvate utilization is not dependent on the cycle. The inhibition of 
over all pyruvate utilization was 12% but corrected for the acetate it was 
31%. In the free parasites, the over all inhibition was 33% and the cor- 
rected inhibition 76%. Of course, pyruvate here or in other cells may be 
metabolized in other ways, so that the correction for acetate alone may 
not give the true cj^cle inhibition, but at least is provides a better value. 

Malonate should inhibit the oxidation of pyruvate more strongly when 
there is a low concentration initially of oxalacetate or a substance forming 
oxalacetate (see page 70). This was shown in homogenates of rat tissues 
by Pardee and Potter (1949). In each case the inhibition by 4 mM malonate 



Tissue 


Substrates 


% Inhibition 


Heart 


Pyruvate 


91 




Pyruvate + oxalacetate 


56 


Kidney 


Pyruvate 


93 




Pyruvate -|- oxalacetate 


55 


Brain 


PjTuvate 


74 




Pyruvate + oxalacetate 


47 


Liver 


Pyruvate 


28 




Pyruvate + oxalacetate 


15 



is less when oxalacetate, is present. Since the oxygen uptake was deter- 
mined from 10 to 30 min after the start of the experiments and inasmuch as 
the concentrations of pyruvate and oxalacetate were 3.5 mM, it is unexpect- 
ed that so much inhibition would be exerted when the mixture is present. 
This might be due to the decarboxylation of sufficient oxalacetate so that 
it was less effective than anticipated, or even at this low concentration 
malonate may have been inhibiting some reaction other than the oxidation 
of succinate. In rat heart mitochondria we found that 5 mM malonate 
inhibits the oxidation of 5 mM pyruvate only about 10% in the presence of 



74 



1. MALONATE 



P5 



»0 r-H 

03 ^- 



M OS 

^ -« -H 






M 



ft ^ 

ft 2 

O ^-' 

5 'O 



o3 _ 



c 




eS 


T--J 




e 


2 


.^ 


'« 


aj 


CC 


02 


-C 


pQ 



-c i-i 3 M 



=D & 



Oi ^ -e 



ft 

ft -« 

O C 

CZ2 1-1 



^ ft 

c S 



< Q 



a 



3 TP 

C5 - — • 



-5 ^ S -3 

M W CO CO 






CO 
H 

Q 

o 



M O -H (M lO M O 



O O 
O ^ 



oooooooo 



o o o o 



OOOOmOOOOOOOoj 



o 

Q 
O 

K 








Q 


•§ 


5^ 


-§ 

.u 


o 






-< 


s 


«:> 


<u 


=0 


«o 


f^ 


Q^ 



o 5j t3 ;::; 



O — ' o 

2 -S -S cs 



■^ .!2 "^ <» "^ 
i=^ ^ :f^ ^'^ .-"^ c 

Oh ^ ft^ >H kH H 



c 
o 

o 
o 

C S ft 



o 


b4 

T3 


6C 


-§ 


'E 


1 


ji 


C 




S 


S 


o 
o 


O 


43 




s 




o 

o 

S 


to 


S 


^ 


s 
•^ 


o 




o 


.g 


'« 


73 




'ft 


« 


o 


o 
O 


2 




s 


J, 


Q 


> 


a 


3 


J2 


"o 




<I1 


o 


02 


a, 


o 


s^ 



03 

> 



EFFECTS ON TRICARBOXYLIC ACID CYCLE 75 



CD 




O 














CO 

05 






CD 

o 






^^ 




2 












05 


^ 


C<) 


Oi 


^ 






ft 
ft 

C 


i 


c 

on 
c3 


ST 


O 
CO 

1—1 


§ 

o 


CO 


CO 


(H 

o 

p-l 

T3 


-D 

C 


CO 

s 

T3 
C 


o 

73 


-D 
ID 

e 


CO 

C5 


2 


c 


"e 






73 


^^ 


x 


c 


a; 




C 


ID 




-c 


c8 
S 

s 


-2 


M 
C 
c3 


s 




c 
o 

CO 

is 


Is 




s 

c 
o 


O 

c 

M 
15 




B 
o 
«> 
c 
o 


Si 

s 
2 


c 
.2 


o 


M 


w 


m 


w 


§ 


fi 


^ 


PM 


s 


Q 


PM 


§ 


o 


H 



^ D 



CO 




o 




05 




^H 




-— ^ 






1 


^ 


^ 


73 


c 
ce 




a 




Ch 


>> 


pq 


<» 




a 

o 


-0 


bj) 


cS 


■1^ 




O 


.1 



§ p^ 



lO-HF-HCOCOCOOO-^OO-^iCOlCOiOCO^H-^COCDC^l^OCOCOt^^H 
05(MI>'<1< (M Ttl(NTt<Q0»OO5O5 Tti00t^Tj<Tt<t>C0COCO f-^ 



lCrti0-*-<*<O'*C0OO-*OlCOTt<Ok0OO©OOlC(Nl0O 



00202 OOccO O O OOO OOomOmOcc OO 



oc o.S^ ■" a c 



^ C S 73 



-c 

^ c 
o 



(B C o; 



a.sSo^fi_os M 

3 3 o o c -^ o 



.a .a » "^ 

cS eS -C M 
-D 42 M ^4 



<» ;5 ;^ 



bl 


C 


C 


C 


0^ 


73 


73 


-^ 










^ 


^ 


13 


-*J 


-t^ 


-1-3 


-i-3 


eg 


c6 


c3 


c3 


P5 


P5 


P5 


P5 



o o ^ S 

« a> o a 

OPh SSqooc4p5p5p5p5 Ci P^O^ 



76 



1. MALONATE 



M 









«:> 


CO 








lO 


»o 






00 


05 


05 






C5 








05 


^-' 


•-' 




^ 

o 


^ 


-o 
^ 






tH 




4> 




05 


<v 


s 








o 




-0 

c3 


n3 

a 


"S 


TJ 


W 


>> 






T3 
C 


S 


0) 

s 




4) 


cS 


o 


o 


a; 




CO 


o 




^ 


Ph 


P5 S 


§ 



3 S^l 



G? 








-o 


s 








CO 


05 








05 


^ 


CO 




Oi 


^ 


a 


OS 




■^ 


^ 


'o 


•■"I 




CJ 


^ 




" — 






<D 


"n 










0^ 






CD 


73 


c 
c 

05 


o 

m 

a 




-t> 


c 


> 


2 




O 


cS 


CO 


o 


?? 


Ph 


>, 


Ti 


X 


CO 


c 


^ 
« 


c 


T3 
c3 


+3 


4) 


O 


3 


C 


o 




a 


> 


O 


a 


cS 


o 


o 


O 


3 


Ph 


S 


H^l 


O 


Q 



H 



O -H fo 
t^ O O 



fcao'-'^-^'MCiao30fOt^Tt<-^i>0'M03DC505 

fO 1— i'>)COiOCOTt(OOCOt--.t^'MO'— iCCfOCOCO 



0-*0 0(MiOOOC'OOOOOOtJ<0»0 

^1 c-i -H ^H '^^ CO ■* LO F^ Tt Tt< Tf< im 



t(< O O O O 

■ri o o o 

r-l (M -^ 



O O «} o 



QOOojO OOO 



O 



O ^ o 



o3 c3 



Pi 



>::^ o 



P5 P5 P5 



EFFECTS ON TRICARBOXYLIC ACID CYCLE 77 



n -Q 



u5 ii 






jj cj 3 ^ ci ^ JS 

^ ffi PQ c« H o a 



O O O C-1 







:3^ 




-O 








C5 


^^ 


1 




00 








f-H 


'M 


rt 




lo 








3 

0) 


O 

CO 

ii 


o 

-^ 
5 

o 


CO 


05 


05 >■/} 


c2 


05 

C 
4J 




> 


"o 


S 


0) (U 


c 


s 


o 


o 


-*-• 


>-. 


o 


b u 


Cw 


o 


K 


« 


J 


h:j 


w 


M fc^ Q 


o 






OOOOOa; O OOOOOcc OCOO Oo 



2 s ^ 

2 S I 1 i - ^ 



rt 2 cj o 00 "-^ «*• 






^1 S-2 



2 -5 t* 

S "S '^ 

S ""^ 
ace 

ai c3 cS 
— £ S 



S^ =53S"uo =S ;:§S See rtce!:>:S®£ 



78 



1. MALONATE 



p:5 



If 


IC lO o ic o o 

rf* — < 'M -H (M 


le S 








o "^ ^_ 



M M <5 






2 2 2 





05 *^ 


o 

CO 


<4J 


"a 


05 


5 
> 

73 






CO 
CO 


eS * 


s._^ 


o 






.^^ 


C 


ce 


13 ^ 




to eS 

o ^ 




o 
o 
o 




o 
o 




c6 




CO tc 

14 « 


»r5 



t- Xh 



^ 3 



m 


s 




T3 


T3 


C 


C 


eS 


cS 






k( 


fi 


<D 


o 


c 



< M 



O O 





"2 




3 


4i 


■1^ 


d 


cS 


tf 


tf 



lO o 



O -^ O C^ 



O O O o o 
M O O ^H i^j 



O O O M 



o 


O 


.« 


^ 


-u 


s 


CO e3 




^ 


-kJ 


,^ 




11 

1-^ C3 


1 


^ 


e 




■^ ft 






-ko c 


.Vi 


a, 

40 


3 s 

03 3 






Pi W 


6q 


ft? 






»0 3 00 00 o 
.S 00 ■<# <N 



o o ^ 
o o 



o o o o o 



OO MOOOO OOOOO 



(^ c s .^ ^ 



cS o 



lU « 0) ^ 



»> ai s 



S P -S 






S 



S S o qH 



"C -tj -u -kj 

5 c^ c^ c3 

^ Im t-< Sh 

-" -ki +i 4J .-ti 

^ o o o - 

is 
o 



J> Q ^ ^ ^ 



5U .M ;— 



5 -H ic o 



Q, CQ o 



-0 

a 


'a 


o 


=c 


o 


S 
^ 


-u 


s 


■§ 


e 




Of) 


O 


5j 


^ 




o 




o 


Q 


> 


-< 


<1 


Oh 



o 



EFFECTS ON TRICARBOXYLIC ACID CYCLE 79 



05 _ 



CD 

i ^ S" s^ ?;r o^ f? 



ececg^ji; "^ ^^C 55 



jn ^ 05 CO CO S 

ST r- ^ C ^-^ i-H ^H f** 



oscocog,^;,^ rr 3 — S— ' 



tj&ciSK© CD'S o SS^mS, -g rt j=iJ .^«3 



CO IC W CD L'5 O C<1 CO 



s s 



CQ 



O OOOOOO COCOCDOOOO 



OOOojOO mO O OOOOOO OOOOO002OOOO 



.2 


Lh 




c 


c 




73 


-a 




a 


a 




c 






s 


.°ci 





CJ 





S 



S 


1 






1 

3 
S3 


'3 




.g 


=0 


1 


ttj 








"C 


5 









« 





A 








c3 




1 


.H 




0) 


04 





e 


« 


« 


bC 


<; 


H 


cS ^ a: 


E 





P-i 


Q 













< 










-< 


S5 








.2 


^ 


_^ 







"0 



•2 






.2 
"E 

c 



T3 
C 


-C 
c^ 

s 





S 


V 

e 


"0 


'g 


^ 


_&> 


^ 





£ 


s 


"s 


.s 








s 


s 






tH 









(u 


s 


« 


-ts 


00 


=0 


V 




_> 


rt 


S 


S 


S 







Ci 


<u 


<» 


>» 







"0 


"o 


"S 


Eh 


4^ 


> 


V 


i; 


e 


c4 


c3 


< 


a. 


tt. 


Q, 





P5 



Eq ■-* ^ ^ ;^ OQ 



^ 







^ 








ce 








c 




s 

S 

s 






s 




.2 




8 


§ 


13 


J3 



S 


^ 


3 
JO 





-§ 


_e 


bi 


s 


"Ss 


5j 



^ 




-«i 


». 





?^ 


_o 





'^ 





"o 


en 


eS 






^ 


"5 

cS 





> 


CQ 


05 





< 



s 



M 



80 



1. MALONATE 



tf 



05 t— 



^ "" o o 



cm lo 
o 



>o O 



^ -2 r" 



e ^' fi 



*) 



c 
c 

m o W 



o 



'A m m 



C S 



ffi 






•;r -^ yZt 
Cm cc W 



O O -H 



C5 A 



a 



Oi 



>> 


>. 


0) 


<» 


X 


tc 


X 


03 


cfi 


03 



S S W O 



— H 









OiCiOI~-l~-lOfOOO-^— * 



s 1 



CO 



O O O I- O O O O M -< 



c o o o ■ri 



o o 

LO O 



OOOOOOOOOO OooOo: 



O M 



cS 


^ 






-73 


_^ 




•s 






-c 


e 


o3 


cS 




S 


ft 

ft 

-0 
C 


o 

s 


2 
o 


c 
o 

"o 
o 

'3 


C 
O 

3 

o 


0) 

c 
■ft 


s 

1 

e 

•i 
S 


s 

to 

a 
1 


O 
C 

'3 
1 


ft 
o 

-(J 

ft 

D 

•2 
e 

1 


o 
"3 

s 

2 
't 


s 

H 

o 


fi 
O 

"o 
o 

> 


■fcl 

-c 
c 
o 

"o 
o 

!-. 


Hi 


C 
1 

'ft 


bO 


«i 


_c 


« 


o 


tH 


^ 


e 


£ 




=0 


•S* 






o 


» 


o 


00 


' r\ 


"ft 


o 






5 
^ 






"e 


o 

3 


rv 

c3 


'3 




^ 


J 


■^ 


H 


CO 


a, ^ 


^ 


&3 


~p 


O 




o 


S 


O 



o 



tf 



EFFECTS ON TRICARBOXYLIC ACID CYCLE 81 






-"'■^ fet^ io ^-^ Z-, ^ a 2 



_^ J f-^ C5 5 "^ 



^ T3 ^ M • • ''^ 



-.^ c 






O5-*(NO5C5f0LOTti o^-'M'^^'00■<^^orJ^oo5(^^■*^0'— o 



CC CC O' 

(M T^ I -H "M lO -^ CO O O 

-H CO 



0.0)000 O Occ coQOOOO O Ox 






eP i3:C55s S 






OJ s c c 



P5 05P5WCI, Cl,q^ >^'^<iici,ll;fi^ p^fe^;^ 



s 


^ 


s 


OJ 




•2 


S 


03 
_0 


s 


^ 




'm 


S 


i 


3 
S 


03 






C 








o 

be 


4^ 








Su 


o ^**. 


^si 


5s •« 


A 



CC 



82 



1. MALONATE 



K 






2 2 S 



w 



< W 



^ c 

05 O 



o OJ 
Cm 02 



t-i a t- 

03 C 
cS 

o 

"3 o 



<x> 


^ 


05 


^ 




a> 


-(J 


^ 


tH 




M 


T3 




eS 


T3 




C 


> 



O g 05 



Q § PQ 



-H CO 



oo^oooooo 









13 


















C 


















o 
















.2 


J= 


















o 














_c8 


"C 


Q 














'E 


T3 


^^ 








_cS 




a 


T3 


C! 


'r- 




se 


_c3 


'C 




C 
O 


C 

o 


O 

o 
o 


a 
5P 


2 

73 


5 

1 


o 


T3 
C 
O 


OJ 


o 

o 


o 

S 


S 


^ 
« 

o 


O 

■a 




3 


o 
o 


-4J 


-1-^ 


Cm 


1/3 


o 






"g 


k< 


'rj 




03 




o 


g 


a 




03 


?H 


'3 


-2 


05 

c 




.S 


b4 


+3 




O 
73 


chl 


o 


"S- 


14) 


> 


o3 


c 


c3 


<B 


o 


tH 


o 


g 


'^ 


^ 


03 


o 
O 
> 


"ft 

-< 


o 
H 


o3 


Para 
Rat 
Rat 
Hum 



05 05 
^- CI 



CI |f^ 05 ro 






hJ Ph 



|1h 



Ph 



Ph 



OQOfOiMOOiMOOO 
O^iCC0l>f0-<*lCiXi 



O ■* O -* O Tt o 
C^l <M "M (M 



OOOOOOOOO «:'0 



% 



P fi jS 



-C ^ 



!2 ^ ^ 



^ « 



p:; p5 



P2 



73 

o 

ft 

O . 



-2 crt 



I ^ ^ 

^ -^ 'm 

C ° I 

<1^ 43 M 

>> C m 

° ^ ^ 

ft Is 

O 03 13 

CO 



c3 



CO c 



o a> 

o _ 



« T) 



<u ;:? 



;^ 



5. 



=" 73 ® 

:S ^ o 

.2 " ts 

03 TJ aJ 



.2 ^ 



•— ., O I— I 
C 1^ C t 

C > o ^ 



s -^ 



o 


03 


-|J 


■3 


C 


eS 


a 


o 


Lh 


u 




+3 




cS 


c 


o 


s 


8 


o 


43 


fl 






8 


43 

s 




cS 


s 


D 


T3 




O 


C 


-a 


o3 



^ s 



EFFECTS ON TRICARBOXYLIC ACID CYCLE 83 

5 mM malate. However, when the malate concentration is between 0.1 and 
1 mM, the inhibition is close to 50% (Montgomery and Webb, 1956 b). 

(6) Inhibition of a-ketoglutarate oxidation. The possible inhibition of a-ke- 
toglutarate oxidase by malonate is important not only because of the bearing 
it has on the effects of malonate on the operation of the cycle, but also 
because malonate has been frequently used to block succinate oxidation 
in order to study in particulate systems the oxidation of a-ketoglutarate 
uncomplicated by further oxidations. This technique was first proposed by 
Ochoa (1944), who showed that high concentrations (25-50 mM) of malon- 
ate would allow a-ketoglutarate to be oxidized to succinate in enzyme pre- 
parations from cat heart. In four experiments with 50 mM malonate, 0.86 
mole of succinate was formed for every mole of a-ketoglutarate utilized, 
indicating that even here an appreciable fraction of the succinate formed 
is further oxidized. No data were given as to whether these concentrations 
of malonate inhibit the utilization of a-ketoglutarate. Slater and Holton 
(1954) used 10 mM malonate to study the oxidation of a-ketoglutarate 
in heart mitochondria, and it was shown that this concentration does not re- 
duce the utilization of a-ketoglutarate, although 20-40 mM does inhibit it. 
Malonate was also used to investigate the formation of a-ketoglutarate from 
citrate in Micrococcus sodonensis (Perry and Evans, 1960), but the rationale 
for this is obscure since malonate by inhibition of succinate oxidation would 
not depress the disappearance of a-ketoglutarate. However, at the concen- 
tration of malonate used (75 mM), it is quite possible that the a-ketoglu- 
tarate oxidase was inhibited. 

The oxygen uptake with a-ketoglutarate as the substrate in particulate 
preparations from several sources has been shown to be reduced by malonate 
as expected if the succinate formed is partially protected from oxidation 
(see Table 1-14). Malonate concentrations around 10 mM inhibit 40-60% 
in most cases. No definite information on the possible inhibition of the 
a-ketoglutarate oxidase can be obtained from such studies. 

If the oxidation of a-ketoglutarate stops at fumarate, the system is a 
two-step linear chain (neglecting the other reactions involved in the forma- 
tion of succinate). An atom of oxygen is taken up in each step: 

1/2 Oa 1/2 O. 

a-Ketoglutarate -> succinate ->■ fumarate 
(1) (2) 

SO that a complete and specific inhibition of reaction (2) would result in a 
maximal inhibition of 50% with respect to the oxygen uptake. However, in 
case the first reaction is much faster than the second, malonate would not 
inhibit the initial rate of the reaction, even though it inhibited the final total 
oxygen uptake. Therefore, the inhibition may theoretically vary from to 



84 1. MALONATE 

50%, depending on the relative rates of the reactions and the period during 
which the oxygen uptake is measured. If the fumarate is further oxidized, 
more oxygen will be consumed and the inhibition by malonate may be 
greater than 50%. Furthermore, it is essentially impossible to inhibit the 
succinate dehydrogenase completely, especially as succinate will accumulate 
and progressively overcome the inhibition. 

The use of malonate to study a-ketoglutarate oxidation in mitochondria 
involves the assumption that malonate does not significantly affect the a- 
ketoglutarate oxidase directly. Unfortunately, no investigations of the in- 
hibition of a-ketoglutarate dehydrogenase by malonate have been report- 
ed, and thus it is difficult to compare the sensitivities of the two dehydro- 
genases. Several studies have determined the effects of malonate on the 
disappearance of a-ketoglutarate (a-KG) during periods when the oxygen 
uptake is depressed, and it is rather strange that some effect, either positive 
or negative, has always been reported. The pertinent data have been sum- 
marized in the accompanying tabulation, and it is seen that an inhibition 



Preparation 


Malonate 

(mM) 








Inhibition of 


O2 uptake 


a-KG utilization 


Streptomyces coelicolor 


10 


41 




Stim 9 


Pea seedling mitochondria 


10 


66 




21 


Blowfly sarcosomes 


10 


10 




Stim 25 


Heart mitochondria 


10 


33 




Stim 8 


Rat brain homogenate 


3.3 


59 




29 



of a-ketoglutarate disappearance is observed in some cases and a stimulation 
in others. The oxidation of a-ketoglutarate depends on Mg++ and it is 
possible that the differences are related to the degree of Mg++ requirement 
and the concentrations of Mg++ used in the assay media. Price (1953) 
found that even 1 mM malonate would inhibit a-ketoglutarate utilization 
in pea seedling mitochondria and that 30 raM had a very marked effect 
(60% inhibition). This is not a competitive type of inhibition because 
increasing the a-ketoglutarate concentration does not lower the inhibition, 
and even increases it somewhat. The possibility of Mg++ reduction was 
explored and it was found that the calculated drop in the Mg++ concen- 
tration could not have been responsible for the inhibition. Furthermore, it 
was not possible to reverse the inhibition by increasing the Mg^^ concen- 
tration from 1 mM to 4 mM. An inhibition by a Mg-malonate complex 
was considered to be the most likely explanation, but yet no inhibition 
was seen when the concentration of Mg++ in the medium was reduced to 
zero, although the enzyme was still partially complexed with Mg+"'". What- 



EFFECTS ON TRICARBOXYLIC ACID CYCLE 85 

ever the explanation, it is obvious that in such mitochondria malonate 
could not be used to isolate a-ketoglutarate oxidation. 

Another quantitative study on the effects of malonate on a-ketoglutarate 
oxidation was made by Grafflin et al. (1952), who were attempting to find 
a good assay system for a-ketoglutarate oxidase in rabbit kidney homogen- 
ates. They concluded that the use of malonate is unsatisfactory and aban- 
doned this procedure. The difficulty lies particularly in the inability, 
except at high malonate concentrations (around 30 mM), to inhibit com- 
pletely the oxidation of succinate, as determined from the total oxygen 
uptake compared with the theoretical value for a one-step oxidation of 
a-ketoglutarate. Although no evidence on the effect of malonate on the 
a-ketoglutarate oxidase was presented, it would be surprising if concentra- 
tions of malonate above 20 mM had no effect. Lewis and -Slater (1954) 
also remarked that even in the presence of 10 milf malonate, the oxygen 
uptake from the oxidation of a-ketoglutarate greatly exceeds the disap- 
pearance of a-ketoglutarate, .lO/.Ja-KG ratios generally being above 2, 
in blowfly sarcosomes. Also, the oxygen uptake over 35-45 min is depressed 
only 10% at this concentration of malonate. 

It is difficult to understand why the oxidation of succinate is so resistant 
to malonate under these circumstances. Taking the values of Kj that 
have been found in mammalian tissue studies, malonate at 10 mM should 
inhibit well over 95% even at succinate concentrations that might occur 
experimentally. For example, in beef heart preparations, where K^ is about 
0.04 WlM, 10 vnM malonate would inhibit over 99% at a succinate concen- 
tration of 2 mM. It may be that in the intact system the oxidation of endo- 
genously formed succinate from a-ketoglutarate via succinyl-CoA is ki- 
neticaUy different than the oxidation of exogenous succinate, or that local 
concentrations of succinate can reach much higher levels than predicted 
on the basis of over all analyses. One must conclude at least at the present 
time that the specific inhibition of succinate oxidation, even in these rela- 
tively simple systems, is generally impossible. 

This problem has been approached recently by Jones and Gutfreund 
(1964), who experimentally determined the steady-state concentrations of 
succinate in guinea pig liver mitochondria during the oxidation of a-keto- 
glutarate. The O2 uptake of uninhibited mitochondria is 40-45% due to 
a-ketoglutarate oxidation, 40-45% due to succinate oxidation, and 10-20% 
due to other oxidations. The effects of malonate on the oxidation of exoge- 
nous succinate were compared with the effects on the oxidation of succin- 
ate-C^* formed from a-ketoglutarate-C^^. The rate of utilization of a-keto- 
glutarate is not altered up to 8 mM malonate. The steady-state succinate 
concentration by total analysis is 0.04 mM, and this is not affected by 
malonate until its concentration is higher than 0.04 mM; half the succinate 
formed from a-ketoglutarate accumulates with 0.7 mM malonate, and 



86 1. MALONATE 

80% accumulates at 8 mM malonate. The Og uptake due to the oxidation 
of succinate formed from a-ketoglutarate is reduced 50% by 0.2 mM mal- 
onate. The data indicate that endogenously formed succinate is at a much 
higher concentration than that determined by total analysis. The inhibition 
studies indicate the steady-state succinate concentration in the mitochon- 
dria to be 4.6-13 mM and the isotopic studies suggest concentrations ex- 
ceeding 4 mM. Such high succinate concentrations would protect the suc- 
cinate dehydrogenase and result in less malonate inhibition than expected. 
Since there appear to be no permeability barriers in the mitochondria to 
succinate, the authors suggested that there is a spatial relation between 
the succinate dehydrogenase and the enzyme forming the succinate. If 
this is true, it raises the interesting possibility that certain enzymes are 
specific not only for their substrates but also for other enzymes with 
which they interact to form functional metabolic complexes. 

(c) Effects on tricarboxylate utilization. Malonate has variable affects on 
the oxidation of citrate and isocitrate, often inhibiting rather well, but 
sometimes having little effect or actually stimulating (Table 1-14). The 
relationship between malonate inhibition and isocitrate oxidation is prob- 
ably complex in most instances. There seems to be minimal direct inhibi- 
tion of isocitrate dehydrogenase, although data are lacking. Hiilsmann 
(1961) showed that malonate at 11 mM reduces quite potently the utiliza- 
tion of isocitrate by rabbit heart mitochondria when acetate or /?-hydroxy- 
butyrate is the additional substrate (these accelerating the utilization of iso- 
citrate). The stimulation of isocitrate utilization by these substrates was 
claimed to be due to the formation of acetoacetyl-CoA, an intermediate in 
fatty acid synthesis, and this alters the oxidation-reduction states of NAD- 
NADH and NADP-NADPH through the transhydrogenase reaction, 
isocitrate oxidation rate being dependent on the concentration of NADP. 
Kasamaki et ol. (1963) showed that malonate inhibits citrate and isocitrate 
oxidations in Proteus vulgaris much less when NADP is added. Chappell 
(1964 a) suggested that the effects of malonate on tricarboxylate utilization 
in rat liver mitochondria are due to the block in the formation of malate 
from succinate, malate being necessary for the oxidation of isocitrate, since 
it provides a means for reoxidation of NADPH mediated through the 
transhydrogenase and malate dehydrogenase. We see here a possibly im- 
portant control in the operation of the cycle and how malonate can exert 
effects indirectly on steps distant from the succinate oxidation reaction, 

{d) Effect of cycle substrate concentration on the inhibition. Some claims 
have been made that increases in the citrate concentration will tend to over- 
come inhibition of the oxygen uptake by malonate, and sometimes this 
has been attributed to a competitive effect, it being assumed that the higher 
concentration of substrate will give rise to a higher succinate level. Thus 



EFFECTS ON TRICARBOXYLIC ACID CYCLE 87 

Laties (1953) found that increasing the citrate from 1 raM to 10 mM 
would reduce the inhibition by 10 mM malonate from around 52% to zero in 
cauliflower homogenates, and Pierpoint (1959) reported that a 10-fold rise 
in citrate concentration brings about a decrease from 62 to 45% in the 
inhibition by 21 mM malonate in tobacco leaf mitochondria. Laties pointed 
out that this could not be due to a competitive effect because of an increased 
production of succinate, since at the malonate concentration used the 
oxidation of 10 mM succinate is completely inhibited. However, if what 
was suggested in the previous section regarding the difficulty in the inhi- 
bition of the oxidation of endogenously produced succinate is valid, data 
on the inhibition of exogenous succinate may not be applicable. Laties felt 
that the explanation might lie in the interference between electron-trans- 
port systems, so that when citrate concentration is high, the contribution 
made by succinate oxidation to the total respiration is less. Clear-cut effects 
of concentration on inhibition have not been observed with a-ketoglutarate 
(Grafflin et al., 1952; Pierpoint, 1959), pyruvate (Smyth, 1940), acetate 
(Jowett and Quastel, 1935 c), or malate (Pierpoint, 1959). 

In experiments of this type, it is well to remember that the pattern of 
oxygen uptake may change with the concentration of the substrate. The rela- 

xO yO 

S -> succinate ->• P 

tive amounts of oxygen taken up before and after the succinate step may be 
altered by substrate concentration due to factors previously discussed in 
this chapter. Thus the ratio x/y in the above equation will vary and hence 
the effect of an inhibitor of succinate oxidation. If the inhibition on succinate 
oxidation is i and the ratio xjy = r, the over all inhibition on the total oxy- 
gen uptake will be ijil + r), so that anything that changes r will change the 
inhibition. If high concentrations of the substrate produce an accumulation 
of some intermediate (as citrate rises when pyruvate enters the cycle rapidly), 
r will increase, at least over the initial period, and the inhibition will be 
less at lower substrate concentrations. 

(e) Stimulation of cycle substrate utilization by malonate. In a number of 
cases there is clearly a stimulation of the utilization of pyruvate, acetate, 
or citrate by malonate. Sometimes this is recognized in an increased oxygen 
uptake but occasionally the disappearance of the substrate is accelerated 
while the oxygen uptake is inhibited. Often this effect is very marked. In 
the mycelia of Ashbya gossypii, oxygen uptake due to addition of acetate is 
stimulated 48% by 4 m.M malonate, whereas 40 m.M malonate inhibits 
53% (Mickelson and Schuler, 1953). The increase in the respiration from 
pyruvate in bull sperm by 10 mM malonate is almost as great (Lardy and 
Phillips, 1945). Table 1-14 cites a number of other instances. 



88 1. MALONATE 

The mechanisms for such stimulations must vary with the particular sub- 
strate used, but it may be useful to suggest some possible ways in which 
malonate could produce this apparently anomalous effect. (1) If the prepara- 
tion has an active oxalacetate decarboxylase, this may reduce the concentra- 
tion of oxalacetate for condensation with acetyl-CoA. Malonate is able to 
inhibit this enzyme in some cases (Table 1-12) (Pardee and Potter, 1949), 
in which case oxalacetate may be protected so that cycle entry of acetyl- 
CoA is facilitated. (2) If ADP concentration is low and ATP concentration 
high normally in the preparation, malonate by inhibiting certain phases of 
the cycle might increase ADP concentration and thus stimulate electron 
transport in other oxidative processes by providing more phosphate ac- 
ceptor. (3) If there is competition between the different oxidative reactions 
in the cycle for some common coenzyme or cofactor (e.g. NAD, NADP, or 
Co A), inhibition of some oxidations by malonate might allow these factors 
to be used more readily by other systems. Such competition between the 
pyruvate and a-ketoglutarate systems has been suggested in heart mito- 
chondria (Montgomery and Webb, 1956 a). (4) If malonate is metabolized by 
the preparation, it is possible that a product would accelerate the utiliza- 
tion of some cycle intermediate. In some organisms malonate can form acetyl- 
CoA and acetate, as well as other products. (5) In intact cells, malonate 
might increase the permeability of the cell membrane so that certain sub- 
strates, such as pyruvate, citrate, or a-ketoglutarate, could enter more 
readily. (6) By chelation with inhibitory metal ions that may occur in the 
preparation, malonate might accelerate the rates of certain reactions. All of 
these mechanisms are purely hypothetical, since in no case has the actual 
mechanism been established. 

Intracellular Concentrations of Cycle Intermediates 

Interpretation of intracellular inhibition by malonate and other inhi- 
bitors acting on the cycle should ideally involve in many cases a knowledge 
of the concentrations of certains intermediates. Data collected for different 
types of cells are shown in Table 1-15. These figures were calculated on 
the basis of intracellular water contents. However, it is likely that these 
substances are not distributed homogeneously throughout the cell water. 
There is evidence that some intermediates may occur within the mito- 
chondria at different concentrations than in the surrounding medium. Thus 
the mitochondria/medium ratios for sheep kidney mitochondria under 
certain conditions were found to be: pyruvate 0.84, fumarate 7.42, a-keto- 
glutarate 1.0, citrate 0.83, and oxalacetate 0.13 (Bartley and Davies, 
1954). Furthermore, the values given in the table are all abnormal since 
truly normal tissues were not used. The rats were fasted for 24 hr while 
the suspensions of E. coli and yeast were metabolizing acetate rather than 
a more normal substrate. In normal rat tissues the concentrations may well 



EFFECTS ON TRICARBOXYLIC ACID CYCLE 



89 



hJ 


•'" 


& 


>-^ 


o 


^ 


^ 


s 


■■ 




w 


r! 


H 


.9 


< 


cS 


O 




H 


fl 


§ 


o 


Z 


o 

Q 



o 



fe) 



w 



m 



C5 CO — 
■<* -^ IC 



o o o o 



COCSCOCOJOO^M'M 
OOO-H-<G0 — -H 

oooooooo 



V 



o o o o 



O'^^OOOOC-'OO 



O iM O O O O O 






ci IB 



-S 8 I 



P^ h^ O o h5 



bn 




rl 


o 
-to 


C 


cS 


4; 






W 


o 


S 


3 


M 


[^ 



§ o 



o 


^ 


H-J 


cfl 


>. 


-o 


-Q 


SJ 


^ 


_o 




rt 


> 


-a 



.2 .2 



«) 


.a 


« 




c 


J2 


o 


"2 


R 


^ 



_ r- ^ 

^ JZ^ TO 



:^ *i ^ 



2 ^ 



.o -^ ., -^ 



~ o 



"c -o g a c 



3 "* D 






p •« oj c 
fS 'S H ^ 



'J^ 



tH ^ ^ 



.2 — M 



^ S^ 



— cc ,0 ~ 



P §^ I 5 



o ■— 



73 



3 S rf -« 



<U Q o "S 

-§ 2 t^ ?^ ^ 

3 ft 3 -43 N 

o TO o oj ;z; 

13 » 2 <" ° 

" i3 ^ -2 

, ci ai a> rt 

1 .5 :^ 1 1 

cS cS o ? S 

^ ^ = ^ « 



t- o 



^ J= -^ .S <D C 



oj OJ t- — 



c 



o „ o C t- ^ 
S ~ tc O 4^ 

o > 2 O c ^ 



•r cs 



S ^ t2 



o Sti 



o 51 



S g 



Q " 
o t 



S S. » s 

CO a) 2 '^ 

"'- -c -5 A 

-^ ^ ^ -J3 



c5 rt 



Q 15 Q 



90 1. MALONATE 

be higher, and in yeast metabolizing sugar the values may be lower. One 
value for oxalacetate in normal rat liver is available (0.036 mikf) and in 
this case fasting for 24 hr did not alter this appreciably (Kalnitsky and 
Tapley, 1958). 

These values do not represent a thermodynamic equilibrium based on dif- 
ferences in free energy, but rather a dynamic or kinetic equilibrium, depend- 
ing mainly on the relative rates of the cycle reactions and competing process- 
es. The higher concentration of fumarate compared to malate in rat tissues 
illustrates this because in a thermodynamic equilibrium there would be 
about one-fourth the concentration of malate. It is interesting that the values 
differ so widely from tissue to tissue and certainly this must be one factor in 
determining the different responses to malonate or other competitive inhi- 
bitors. The levels of succinate are generally low, except in yeast, but as 
pointed out above the concentrations within the mitochondria or at the re- 
gion of the active center of succinate dehydrogenase may well be higher. 
The effects of malonate on the concentrations of these intermediates in cells 
will be taken up in the following section. 

Analyses of plant tissues have not been presented here because there is 
some doubt as to the significance of the figures. Most plants contain large 
amounts of organic acids, including the cycle intermediates. Beevers (1952) 
postulated that the cj^cle in plants is less readily blocked than in animal 
tissues because of these high concentrations of succinate and other cycle 
intermediates. This could well be an important factor, but actually the 
concentrations of these acids in the plant cytoplasm are not known in 
most cases, total analyses including the vacuolar fluid, which is often of 
greater volume than the cytoplasm and contains most of the organic acids. 
There is another way by which these plant acids could protect against 
malonate. The presence of large amounts of fumarate or malate, or of any 
substance capable of forming oxalacetate, would allow pyruvate to be incor- 
porated into the cycle even in the state of complete block of succinate oxi- 
dation. Examples of the overcoming of malonate inhibition by fumarate 
and malate will be presented shortly. 

ACCUMULATION OF SUCCINATE DURING MALONATE 

INHIBITION 

An effective inhibition of succinate oxidation should lead to a rise in 
the concentration of succinate under conditions in which succinate can still 
be formed. Such accumulation of succinate has been frequently observed 
and some of the more quantitative results are summarized in Table 1-16. 
In addition to the examples in the table, accumulation of succinate has 
been reported in the following species and tissues: Shigella (Yee et al., 
1958), Nocardia (Cartwright and Cain, 1959), Aspergillus (Shimi and Nour 



ACCUMULATION OF SUCCINATE 91 

El Dein, 1962), tobacco leaves (Vickery, 1959; Vickery and Palmer, 1957), 
potato slices (Romberger and Norton, 1961), avocado mitochondria (Avron 
and Biale, 1957), pea leaf particulates (Smillie, 1956), barley roots (Laties, 
1949 b), Colpidium (Seaman, 1949), Trypanosoma (Bowman et al., 1963), 
carp liver mitochondria (Gumbmann and Tappel, 1962 b), rat heart ho- 
mogenates (Lehninger, 1946 b), rat liver slices (Elliott and Greig, 1937), 
human heart slices (Burdette, 1952), ascites carcinoma cells (Dajani et al., 
1961), and in many tissues of rats and rabbits (Busch and Potter, 1952 a; 
Forssman, 1941). In the experiments leading to the results in Table 1-16, the 
preparations were incubated for one to several hours with malonate and the 
succinate analyzed at the and of the incubation, so that the rates of succinate 
formation at any time are difficult to evaluate, and may well have been 
greater initially. The over all succinate concentrations may be estimated 
from the volumes in which the expriments were run and in most cases the 
final succinate concentrations range between 0.5 and 2.5 mM. 

Several points are brought out by the results in Table 1-16. It is seen 
that succinate can be formed from essentially all the cycle substrates and 
intermediates in the presence of malonate. Rapid rates are found when 
oxalacetate or some substance forming oxalacetate is added with pyruvate, 
as would be expected, because in the absence of a source of oxalacetate, the 
malonate would reduce the incorporation of pyruvate into the cycle and 
hence the rate of formation of succinate. It may be noted in some cases 
that, in the absence of added substrates or malonate, some succinate ac- 
cumulates (yeast, Avena coleoptile, spinach leaves, and dog heart), which 
implies that under the experimental conditions succinate is formed more 
rapidly than it can be oxidized. This is somewhat surprising because it is 
usually assumed that the activity of succinate oxidase is quite high in 
most tissues. The possibility of the accumulation of sufficient oxalacetate 
to inhibit succinate dehydrogenase when little acetyl-CoA is available 
cannot be ignored. This phenomenon is also evident in the analyses for 
succinate given in Table 1-15. The interesting effects of malonate concen- 
tration are seen in two investigations. In spinach leaves, the maximal suc- 
cinate accumulation occurs at malonate concentrations around or below 
50 milf; at higher concentrations, the malonate is apparently acting on 
other enzymes in the cycle and reducing the rate of formation of succinate. 
Likewise, in brain minces, the high concentration of 200 mM malonate is 
seen to depress succinate accumulation. 

Quantitative conversion of cycle substrates to succinate in the presence 
of malonate is generally not observed. In fact, in most cases in which the 
disappearance of substrate was determined simultaneously with the for- 
mation of succinate, only a small fraction appeared as succinate. For exam- 
ple, Speck et al. (1946) found in malarial parasitized erythrocytes that 
only 22% of the pyruvate utilized in the presence of 20 mM malonate was 



92 



1. MALONATE 



:^ ^ 



05 



>. 



m 



o 

03 



0000'i^C<ICO'005-H(MfOTti-HCC'^ 



11 

cS 3 






+ 



< < 



CO 


JO 






C<1 


1^1 






o 


o 


2- 


2i 



^ cj o 



c c ai OJ 
S 3 o o 

&H fe <: <: 



CLi (i, 



Oj © -H -H 



r-i rS cS C3 
r- 3 C3 C6 

fa p& s s 



ACCUMULATION OF SUCCINATE 



93 



02 



1^ 7^1 C5 l^ O 

— (M -* O -^ ~ 



C*^ ^H ■»!^ 3 CO 



M M O Lt O O 

r: --^ 't c rc C5 
o — ':o — -H tJ< 



+ + 



<! <j3 



a; <B ^ 






Ch Ph § 



+ + 

S" o 3 2 S^ S" 



i' t^ Z ~ C C 



Ch Ph 



PL, 



o 
-1-2 



^ ^ Ph fl. 



fa 



94 



1. MALONATE 






(X 



bO 

o 

C 
c8 



o 


M 


K 




T1 






TJ 



K 



tiJ 



a 



rt (M (M — < 



O — ' O — ' 



(^rtO-^-*CO-Ht>C5 



lO lO LO >c 

■ri "M 5^1 C^l o 



+ + + + + + 



+ 



o 


o 


(3, 


.— , 








. 




. 





^^ 


2. 


■" 


-^ 


^ 


"^ 


"-^ 


:;;^ 




~-^ 




:;^ 


M5 


OJ 


« 


« 


0) 


Oi 





OJ 


» 


© 


<u 


(Kl 


^ 
















f« 








2 


C3 


* 


03 


03 


CS 


ce 


o3 


(S 


ee 






^ 


> 


^ 


^ 


^ 


> 


r^ 


> 


> 


-tS 


oa 


S 


^ 




^ 


3 


D 


^ 




3 


^ 


te 


a 


k. 


u 


;-i 




b< 




^ 




b 






>-. 




>, 


>- 




>•- 


>. 


2 


>> 


>> 




s 


CM 


Cm 


fl^ 


ei^ 


<^ 


;a^ 


^ 


t>H 


^ 


eu 





Pm 






03 



ACCUMULATION OF SUCCINATE 



95 






00 o 
lO 00 ■<* 



O 00 CO 

■* -H C^ 



<MOI>'*00'-< OCa<M 



iC >c o o o o 

(M <M ifi »0 O O 



IC lO o o o o 

Cq <M lO iO O O 



O 

Q 



W 



H 5 



o 'C 

V0 0) 



0) « c si 



c ^ 





cS 


c« 




a 


rt 




S 


c 






Si 




3 


d 










<«4 


<4H 




3 


s 




+ 


+ 




Ci 


c«^ 


^^ 


^^ 


, . 




+ 


+ 


lO 


IC 


lO 




















<M 


•M 


(M 


O 








::;i 


:3i 


:;::;, 


■3;!, 


:;i 


^ 


ffii 


(D 


i» 





« 


(D 


-f^ 


4J 


*i 


-u 


-ti 


•ii 


a 


eS 


cS 


ci3 


cS 


ce 


> 


> 


> 


> 


> 


> 


? 


S 


3 


3 


3 


3 



Ph PL| Pm Ph Ph Pm 



-5 








r5 








rj 


<M 


<M 




— ■ 


^-^ 


<D 


CJ 


<» 






-t^ 


crt 


cS 


crt 


> 


> 




3 




3 


ki 




t- 


>i 


>. 


>. 


Ih 


P4 


(^ 



w 



w 



l:i1 



oj _tn 



>■, * 



96 1. MALONATE 

recoverable as succinate. The highest conversion efficiency was observed by 
Krebs and Eggleston (1940) in pigeon muscle brei, where 75-85% of the 
pyruvate utilized in the presence of fumarate and 12.5 mM malonate went 
to succinate. Complete conversion to succinate would not, of course, be ex- 
pected unless one could inhibit succinate oxidation completely and specifi- 
cally, which in most instances cannot be done. 

Factors Determining Succinate Accumulation 

The effects of inhibition on the concentrations of intermediates in multi- 
enzyme systems have been treated in Chapter 1-7. Some of the most impor- 
tant factors involved in malonate inhibition will be summarized. 

(I) Degree of inhibition of succinate oxidase: hence, the concentration 
of malonate, the affinity of the enzyme for malonate, and the ability 
of malonate to penetrate if the preparations are cellular. 

(II) Rate of formation of snccinate: this will depend primarily on the availa- 
bility of cycle substrates and their concentrations. 

(III) Action of malonate on enzymes other than succinate dehydrogenase: 
such actions may slow down the formation of succinate, as discussed 
above. 

(IV) Other pathways of succinate metabolism: several reactions of succinyl- 
CoA or succinate are known and these would tend to prevent accumula- 
tion. 

(V) Diffusion of succinate from cells: when only cells or tissues are analyzed 
for succinate, intracellular accumulation will be reduced by the loss of 
succinate into the medium or the blood. 

(VI) Time after addition of malonate: although this has never been studied, 
it is probable that the succinate concentration will follow characteristic 
time courses in each case, in some cases perhaps decreasing after a peak 
level has been reached. 

Another factor, about which nothing is known, is the possible effect of 
rising succinate concentration on the reactions forming succinate. The 
oxidation of a-ketoglutarate forms succinyl-CoA, and succinate can arise in 
at least three different ways from succinyl-CoA. 

P-enzymes 



Q -Ketoglutarate — >- sue cinyl - Co A *- sue einate 

transaeylases 



ACCUMULATION OF SUCCINATE 97 

What effects succinate concentration might have on these reactions are 
unknown, but some inhibition during succinate accumulation is possible, 
although a simple backing-up of the a-ketoglutarate -^ succinate reaction 
would be thermodynamically unlikely. 

The relative potential rates of a-ketogluti^rate oxidase and succinate 
oxidase under normal intramitochondrial conditions are not known, but the 
oxidation of succinate is certainly one of the most rapid reactions seen in 
mitochondrial suspensions. It may be that there is no accumulation of suc- 
cinate in the cycle under physiological conditions, and that the small 
amounts of succinate found in tissues do not truly indicate the situation in 
the regions of succinate oxidase. Succinate oxidase is, perhaps, the one 
enzyme that has never been considered as normally limiting the cycle rate. 
If the maximal rate of succinate oxidation is much higher than the rate at 
which succinate can be formed, it would require a fairly high inhibition of 
the oxidase before succinate accumulates markedly. For example, if the rate 
of succinate formation is one-tenth the rate at which it can be oxidized, 
90% inhibition of the succinate oxidase would make the rates equivalent, 
and the succinate concentration would not rise very much (probably not 
more than 0.02 0.05 n\M). Under any likely conditions, calculations from 
Eqs. 1-7-8 and 1-7-9 make it clear that succinate oxidation must be inhibited 
fairly strongly to produce a significant succinate accumulation. The com- 
m.on assumption that succinate must accumulate rather quantitatively 
when sufficient malonate has been added to inhibit succinate oxidase 
75-90% is thus unjustified. If alternate pathways for succinyl-CoA or suc- 
cinate exist, the accumulation of succinate would be even less evident. 

Several instances of failure of succinate to accumulate during malonate 
inhibition have been reported. Hanly et al. (1952) found that in only two 
of six experiments with carrot root slices did succinate accumulate, and 
in these the rise was insignificant. One might suspect a lack of penetration, 
but 15-50 mM malonate was used at pH 4; under these conditions, malonate 
depressed respiration strongly. Weil-Malherbe (1937) found no succinate 
accumulation in guinea pig brain slices with malonate 4-40 mM, respiration 
being markedly reduced at the higher concentrations. A depression of suc- 
cinate level in Streptomyces due to 10 vaM malonate was noted by Coch- 
rane (1952), with either malate or citrate as substrate. Since the incubation 
was 16 hr and the pH 5.1-5.4, it is possible that a nonspecific acid damage 
from malonic acid penetration was responsible for the cycle depression, or 
it might be that in this organism malonate is not specific at 10 mM, or, 
as Cochrane suggested, succinate may not be formed via the cycle. The fail- 
ure of succinate to accumulate, even when the respiration is suppressed by 
malonate, is difficult to explain except on the basis of actions other than 
on succinate dehydrogenase. 



98 1. MALONATE 

Succinate Accumulation in the Whole Animal 

Some of the most interesting and suggestive experiments on succinate 
accumulation resulting from malonate inhibition have been performed with 
whole animals and, although such work is often difficult to interpret in a 
quantitative fashion, the results have demonstrated that malonate can 
partially block the succinate oxidase in various tissues of the living animal. 
Such inhibition has obvious implications for developments in the study of 
drug actions and chemotherapy. The first work of this type was done by 
Krebs et al. (1938), who determined the urinary excretion of succinate, cit- 
rate, and a-ketoglutarate following injections of cycle intermediates and mal- 
onate into rats and rabbits. Some of their results are shown in Table 1-17. 
The effects on citrate and a-ketoglutarate will be discussed later (see page 
104). Although malonate increases the succinate excretion some 5-fold, 
only 2.9% of the injected fumarate is recovered as succinate compared to 
0.6% in the controls. The injection of 10 millimoles of a substance into a 
rabbit will lead to a maximal extracellular concentration of approximately 
30 TdM, so that reasonably high concentrations of malonate were probably 
achieved. The effect of the malonate had mainly disappeared after 24 hr 
due to the excretion and destruction of the malonate. An almost 10-fold 
increase in the urinary succinate was seen in the more recent experiments 
of Thomas and Stalder (1958), in which 3.7 millimoles/kg of sodium mal- 
onate were fed to rats, the succinate over a 40-hr period rising from a 
control value of 2.35 mg to 28.0 mg. 

The blood concentration of succinate is increased in rabbits following 
the injection of malonate (Forssman, 1941). Intravenous injection of 2.8 
millimoles/kg of malonate leads to a slow rise in the blood succinate to 
around 0.20 mM at 3 hr, while injections of 3.5-5.1 millimoles/kg give 
levels as high as 0.77 mM. A lethal dose of 8.25 millimoles/kg produces 
death in 35 min and at the time of death the succinate concentration is 
1.1 mM. The lower doses produce no obvious effects on the animals. The 
normal values for blood succinate are about 0.025 milf. 

The succinate found in the urine and blood in these studies originated 
mainly in the tissues of the animals. Are malonate inhibition and succinate 
accumulation especially related to a particular tissue, or do all the tissues 
contribute to the metabolic disturbance? Can differences in the metabolic 
patterns of the various tissues be demonstrated by their responses to the 
administration of malonate? How do tumors compare with normal tissues in 
their susceptibility to malonate? It was to answer such questions as these 
that Busch and Potter (1952 a, b) at the McArdle Memorial Laboratory 
at Wisconsin undertook their excellent series of studies on the accumulation 
of succinate in various tissues of rats following injections of malonate. 
Analyses for malonate and succinate were made by anion exchange chrom- 
atography (Busch et al., 1952) at various times after the subcutaneous 



ACCUMULATION OF SUCCINATE 



99 



a 



W 



O 






"l^ 


S 


-^ 




cS 


'o 


b 


2 


en 




J2 


^ 


3 

CO 


's 



« 



to 


a 


S 


ft 


o 


!«! 


c 


V 


> 


-C 


cS 





i:i5 



« 


T3 


^ 


03 


03 


3 


-l-i 




3 


£ 


-D 


4) 




^ 


» 


J2 


-c! 


Is 


H 














c 


00 


cS 


M 


Oi 


Id 




3 




o 




Ch 






e 


73 


■•«& 


C 


■» 


eS 








3 
O 




0) 


P5 


C 


o 


^ 


fe 


3 
o 


e 


,Q 




g 



100 



1. MALONATE 



injection of 12 millimoles/kg of malonate (Fig. 1-11). Maximal concentra- 
tions of malonate are reached in most tissues 1-2 hr after the injection and 
the time course of the succinate levels is well correlated with that of malo- 
nate. The tissue succinate concentration is linearly related to the malonate 
concentration, except in the Flexner-Jobling tumor (Busch and Potter, 




Fig. 1-11. Tissue levels of malonate and succinate 
in rats injected with sodium malonate (12 milli- 
moles/kg). (From Potter et al., 1952). 



1952 a) (Fig. 1-12). The slopes of the lines give some measure of the degree 
of malonate effect in the particular tissue, but several factors are involved 
so that they are not quantitative indications of succinate oxidase inhibition. 
The urinary excretion of malonate, succinate, and citrate is shown in Fig. 
1-13. The malonate and succinate levels in several normal tissues and tu- 
mors following 24 millimoles/kg malonate subcutaneously are given in 
Table 1-18. If the figures in the table are multiplied by approximately 
0.7, one will obtain the millimolar concentrations in the cell water, except 
for kidney and blood, the former tissue having extracellular fluid high in 
succinate and malonate. 

The ratios of (succinate)/(malonate) in the tissues give essentially the 
slopes of the lines in plots such as Fig. 1-12, except for tumors where a 



ACCUMULATION OF SUCCINATE 



101 



Liver 


/ 

Tumor 


y^ Kidney 


/ 


^^ 


Q\oo^^^^^^^ 



10 20 

Malonote (millimoles/ kg) - 

Fig. 1-12. Relationships between malonate and suc- 
cinate concentrations in the tissues of rats injected 
with malonate. (From Busch and Potter, 1952 a). 




Time(hr)- 



FiG. 1-13. Urinary excretions of malonate, succinate, and citrate 

following the injection of 12 millimoles/kg sodium malonate into 

rats. (From Busch and Potter, 1952 a). 



102 



1. MALONATE 



Table 1-18 

Tissue Succinate and Malonate Concentrations Following Injections of 

Malonate in Rats " 





Tissue 


Succinate 


Malonate 


Ratio 




Control 


After 
malonate 


Control 


After 
malonate 


succinate 




malonate 


Spleen 







12.8 





12.8 


1.00 


Liver 




2.2 


12.0 


2.4 


14.2 


0.85 


Brain 







2.0 





2.8 


0.71 


Thymus 







9.0 





21.0 


0.43 


Kidney- 




0.6 


16.4 


0.3 


44.0 


0.37 


Lung 







5.6 





16.0 


0.35 


Muscle 




1.0 


3.0 


2.0 


8.8 


0.34 


Heart 







5.0 





18.4 


0.27 


Blood 







2.6 





24.6 


0.11 


Tumors 














Flexner 


-Jobling 












carcinoma 





9.2 





17.6 


0.52 


Walker 


256 carcinoma 





8.0 





5.2 


1.54 


Jensen 


sarcoma 





8.8 





8.0 


1.10 


Hepatoma 





6.0 





12.0 


0.50 


Papilloma 





4.4 





4.4 


1.00 


Average 







7.3 





9.4 


0.78 



" The figures are in /<equivalents/g wet weight of tissue. Malonate was injected 
subcutaneously at a dosage of 12 millimoles/kg, and after 1 hr a similar amount was 
again injected; 1 hr following the last injection the animals were sacrificed. (From 
Busch and Potter, 1952 b.) 



linear relationship may not be followed. These ratios indicate the amount 
of succinate formed per unit concentration of malonate and do not relate 
directly to the degree of inhibition of succinate oxidase but more to the 
ability of the tissue to form succinate in the presence of the inhibition. 
As discussed previously, for the same degree of block, succinate will be form- 
ed at greatly different rates in different tissues, depending on the succin- 
ate-forming substrates available and the activity of the pathways leading 
to succinate. The concentrations of malonate in the tissues vary greatly 
and this must be a reflection of the differing permeabilities of the tissues to 
malonate. It may be noted that the concentration in the brain is quite low, 



ACCUMULATION OF SUCCINATE 103 

a phenomenon seen with most ionic substances, and this must account for 
the poor accumulation of succinate in this organ and the relative lack of 
effect of malonate on central nervous system function. 

The low concentration of succinate in the blood is interesting since it 
implies that succinate does not leave the tissues readily. The conclusion 
that must be reached is that the rate at which succinate diffuses out of the 
tissues into the blood is slower than the rate of renal excretion of the 
succinate. Substances that are not resorbed by the renal tubules are excret- 
ed rapidly and their concentrations in the blood can be maintained at a 
low level despite a continuous influx. Nevertheless, it shows that succinate 
leaves the tissue cells rather slowly under physiological conditions. The slow 
penetration of malonate into the tissues is suggested by the fact that the 
peak levels in the blood occur around 30 min after administration whereas 
the peak levels in liver and tumor occur 30 min later (Fig. 1-11), the kid- 
ney concentration paralleling the blood levels because of the excretory 
function of this organ. 

The degree of succinate accumulation does not necessarily reflect the 
cycle activity in a tissue. For one reason, succinate can often be formed 
from other pathways. In certain tissues the amino acid content rises mark- 
edly (e.g. + 215% in thymus and + 160% in spleen) during malonate 
inhibition, while in others, especially the Flexner-Jobling carcinoma, the 
amino acids decrease as the succinate increases. It is likely in the latter 
tissues that some of the succinate is derived from amino acids, probably 
mainly glutamate. Thus the cycle activity in this tumor may be quite low 
and the normal accumulation of succinate due to other sources for the suc- 
cinate. The other tumors do not show such marked decreases in amino acid 
content and this was attributed to their greater necrosis. It may be recall- 
ed that the incorporation of acetate by Flexner-Jobling tumor is slower 
than in most tissues and very little labeled succinate is formed from la- 
beled acetate (Busch and Potter, 1953), indicating a low degree of cycle 
activity. We have seen that this tumor also differs from normal tissues in 
the nonlinearity of the plot of tissue succinate against tissue malonate 
(Fig. 1-12). At low levels of malonate the ratio (succinate)/(malonate) is 
near 3 but at high concentrations diminishes to 0.5. This means that as 
the malonate concentration rises, the ability of the tumor to accumulate 
succinate decreases. This is the type of curve expected if malonate at higher 
concentrations is inhibiting the reactions forming succinate. If the supply 
of cycle substrates in this tumor is low, a relatively small block of the cycle 
might reduce the formation of succinate through the cycle to zero. The 
slope approaches that of the blood, and it is possible that above the inflection 
point the slow rise in succinate may be due only to the rise in the blood. 

This type of investigation could well be applied to other inhibitors 
and certain chemotherapeutic agents. First, one is able to correlate tissue 



104 1. MALONATE 

concentrations of inhibitor with the metabolic disturbance produced. 
Second, the inhibition occurs under physiological conditions, rather than 
in slices or minces or other preparations in which the cell metabolism may 
be very abnormal. Last, one is able to compare the different tissues with 
respect to their metabolic patterns and perhaps determine some of the 
reasons for the selective actions of inhibitors or drugs. These methods of 
investigation, called "m vivo metabolic blocking techniques" by Busch 
and Potter, if applied properly, would help to provide a more rational 
basis for development in chemotherapy and the selective depression of 
tumor growth. 

ACCUMULATION OF CYCLE SUBSTRATES 
OTHER THAN SUCCINATE 

Specific inhibition of succinate oxidase would be expected to lead to the 
accumulation of succinate but of no other cycle intermediates, because 
the free energy differences between them are of such magnitude that no 
backing-up from succinate would be anticipated. When other members of 
the cycle are found to accumulate in the presence of malonate, it is generally 
considered to be evidence that either the action of malonate is not specific 
or that secondary reactions are proceeding. Malonate has been shown many 
times to cause an accumulation of certain cycle intermediates, especially 
citrate and a-ketoglutarate, in cell suspensions, slices, and whole animals. 
The nature of these effects will first be summarized and then some possible 
mechanisms will be considered. 

Accumulation of Citrate 

The administration of malonate to dogs (Orten and Smith, 1937), rabbits 
(Krebs et al., 1938),- and rats (Busch and Potter, 1952 a) leads to an increased 
urinary excretion of citrate (Table 1-17 and Fig. 1-13). There is also a rise 
in plasma citrate following injections of malonate in rabbits (Forssman, 
1941) and dogs (Stoppani, 1946). Tissue citrate also rises in mice injected 
with 10 millimoles/kg malonate: kidney (16 to 20), heart (40 to 70), liver 
(5 to 10), and diaphragm (70 to 225) (values in milligrams per kilogram 
wet weight) (Chari-Bitron, 1961). Brain, however, shows no increase in cit- 
rate, perhaps due to the poor penetration of malonate. Some accumula- 
tion of citrate has also been observed in suspensions of Ashbya gossypii 
mycelia metabolizing acetate and oxalacetate (Mickelson and Schuler, 
1953), Schizophyllum commune mycelia metabolizing pyruvate and malate 
(J. G. H. Wessels, 1959), and Ehrlich ascites tumor cells metabolizing fu- 
marate (Kvamme, 1958 c) in the presence of malonate. Thus this phenom- 
enon is widespread, occurring in different types of organism and under a 
variety of conditions. 



SUBSTRATES OTHER THAN SUCCINATE 105 

On the other hand, 4-10.5 millimoles/kg of malonate injected intra- 
venously into rabbits does not increase plasma citrate appreciably (Forss- 
man and Lindsten, 1946), and a number of reports have indicated a depres- 
sion of citrate formation by malonate. Eat brain and liver homogenates oxi- 
dizing oxalacetate, or pyruvate and oxalacetate, form less citrate in the 
presence of 10 mM and 30 milf malonate, respectively, this being attributed 
to an inhibition of oxalacetate decarboxylase (Pardee and Potter, 1949). 
The formation of citrate from acetate in yeast is inhibited 73% by 17 mill 
malonate, while simultaneously succinate accumulates markedly (Barron 
and Ghiretti. 1953). The incorporation of C^^ from glucose into citrate in 
potato tuber slices is also depressed 71% by 50 mM malonate (Table 
1-19) (Romberger and Norton, 1961). Although there is an increase in 
citrate in excised tobacco leaves during culture with malonate, this increase 
is generally less than in the controls, so that this probably represents an 
inhibition of citrate formation (Table 1-20) (Vickery, 1959; Vickery and 
Palmer, 1957). Finally, malonate inhibits the formation of citrate from a 
variety of substrates in kidney and testis breis, the effects being surprisingly 
large for the reasonable concentrations of malonate used (Table 1-21) 
(Hallman, 1940). It is, therefore, evident that citrate levels may be affected 
by malonate in a variety of ways, depending on the malonate concentration, 
the type of preparation, and the conditions of the experiment. It is not dif- 
ficult to explain the falls in citrate level brought about by malonate, since 
this could arise either by a depression of succinate oxidation (reducing the 
rate of entry of acetyl-CoA into the cycle) or inhibitions of other reactions 
(such as the condensation of oxalacetate and acetyl-CoA), especially at the 
high malonate concentrations often used. It is, on the other hand, difficult 
to interpret the accumulation of citrate and to this end we must direct 
our efforts, although only suggestions can be offered because of the lack 
of sufficient data. 

Citrate is being formed and metabolized continuously and thus, generally 
speaking, a rise in the citrate level implies an inhibition of citrate utili- 
zation or an acceleration of its formation, or both. Although there is little 
evidence for a direct affect of malonate on isocitrate utilization, we have 
noted that an inhibition of succinate oxidation can interfere with isocitrate 
oxidation by depletion of NADP mediated by a fall in malate concentration 
(Jones and Gutfreund, 1964). This could certainly contribute to the accu- 
mulation of the tricarboxylates in some instances. 

The inhibitions of citrate oxidation in Table 1-14 can be mostly explained 
on the basis of a block at the succinate oxidase step. On the other hand, 
there is certainly no evidence that the formation of citrate via the cycle 
can be stimulated by malonate, most data pointing instead to a depression 
if there is any effect. It would appear that effects of malonate on the cycle 
alone are not sufficient to explain an accumulation of citrate. Other path- 



106 



1. MALONATE 



Table 1-19 

Distribution of Radioactivity Following 3-Hr Incubation of Potato Tuber 
Slices with Glucose-u-C^* "■ 



Component 




Radioactivity (cpm) 


Control 


Malonate 


% Change 


Sucrose 


344,000 


306,000 


- 11 


Glucose 


31,900 


26,800 


- 16 


Fructose 


5,950 


5,830 


- 2 


Oligosaccharides 


8,050 


6,000 


- 25 


Weak acids 








Citrate 


4,350 


1,280 


- 71 


Isocitrate 


235 


757 


+222 


Succinate 


925 


6,300 


+582 


Fumarate 


190 


125 


- 34 


Malate 


8,900 


1,740 


- 80 


Glycolate 


2,750 


1,830 


- 33 


Total 


67,300 


41,500 


- 38 


Acidic amino acids 


56,100 


16,700 


- 71 


Neutral amino acids 


32,100 


56,200 


+ 75 


Basic amino acids 


950 


750 


- 21 


Phosphorylated compounds 


28,000 


10,000 


- 64 


Lipids 


111 


367 


+230 


Respiratory COj 


22,600 


17,300 


- 23 



" Fresh slices of potato tubers incubated at 28° and pH 5 for 3 hrs with uniformly 
labeled glucose, in the absence of and presence of 50 xnM malonate. (From Romberger 
and Norton, 1961.) 



ways for the formation of citrate are known, e.g., the citrase reaction from 
acetate and oxalacetate, or through isocitrate by the isocitrase reaction 
from succinate and glyoxylate. However, the free energy changes for these 
reactions are such that citrate would not accumulate in significant amounts 
even though the substrates for its formation accumulated. Also the results in 
Table 1-22 show that the effect is not specific for malonate, but is seen with 
succinate, malate, fumarate, and other organic anions. The marked effect 
of glutarate, which is a poor inhibitor of succinate oxidase, suggests that 
the citrate accumulation may not be related to the block of this enzyme. 
It may be noted that part of the augmented citrate excretion may be attri- 



SUBSTRATES OTHER THAN SUCCINATE 



107 







S" 




GO 








o 




CO 








o 


w 




CO -^ 


o 




o o 


TlH (M 


OO CO 1 


S^l 


CO 


--^ 




5 ^ 










Ci o 


<N 


i> 00 1 






^H 




a 






C<1 








^H 






^H 


ID 








1 + 






+ 7 


+ 1 


+ + 


1 




























t-i 
3 




lO 




■* 








• o 




(M 


t^ 




-*^ 


W^ 


W 




o 00 


CO 




-^ o 


I> 


LO K) 1 


d 


00 





'3 




I— 1 


o 




lO CI 




Tf* OO 1 






(M 


o 




a 






(N 




I-H 




I-H 






^H 


a 








1 + 






+ 1 


+ 


+ + 


_\_ 










CO 








o 


^~, 


lO 






Ti 




IC ■*' 


1— ( 




Oi o 


CO o 


<N ^ 1 




1 


1 








'^^ 




00 o 


-* 


CO 1 


(M 


I 


1 


bJD 
C 


Ksi O ft 




+ + 


-M 




+ + 


+ 


+ + 


_J_ 






(-1 




(M 








C5 


»o 


»o 






o 


"^ 














(C 










-*-:> 




00 d 


1 




lO ^ 


CO CO 


o O 1 


(M 


1 


1 




k5* 






I 




CO 


IM 


<s 




1 


1 




^ 














^ 












+ + 






1 1 


+ + 


^ + 


_^ 






"o ff> • 




CO 








^ 


Ci 


CO 






S t- -Q 














a; 








-t^ ,o s 




o c- 


1 




-H CO 


lO fM 


o O 1 


0<l 


1 


1 


" ^ " 






1 




O CO 




S 




1 


I 


5-S-S 










(M -H 






























,3 


"^^ 










t^ 




Tt< 








^-1 


CO 


CO 










w 




O ci 


CO 




Tt< cc 


^ O 


Oi CO 


^ 


'X 


-^, 










CO 




CO C5 


Tin rt 


Tt< 






5 






ft 






i-H 




I-H 




I-H 










^ 






+ 






+ 1 


+ 1 


+ + 1 


1 




















CO 


(^, 








s 


S" 




lO 








'M 


LO 


CO 








<3) 


w 




C<J C2 


•* 




O -H 


O -H 


LO C5 


,_) 


r- 







o 






CO 




T)< O 


CO I-H 


lO CO 






00 


_c 


1— 1 


3 






I-H 




1-H I-H 




»-H 














1 + 






+ 1 


+ 1 


+ + 1 


1 






0) 


-2 
















00 









3 


1 


o 




CO 








OO 


CO 


CO 






+s 


W 




(M 00 


,—, 




-H ^ 


CO -H 


'>^ ^ <Z> 


^H 








"3 






(N 


(M 




t- — < 




LO OO 






Tti 


o 


Is 


ft 






C<1 




I-H I-H 














s 






1 + 






+ 1 


+ 1 


+ + 1 


J_ 




















_H 


t^ 






s 




^ 




O 








CI 


CO 


(M 






Tj 




w 




O Tt< 


C5 




r- c<) 


oi C5 


t- d 


I-H 
















-* 




a> -H 




CO c^ 






00 


« 




ft 






I-H 




i-H 




f— ( 
















1 + 






+ 1 


+ 1 


+ + 1 


1 






8 ^ e 




?M 








CO 





CO 

CO 








^ o W 
bi! o P. 




Tj< IC 


o 




CO Ci 


CO CO 


10 1 


F-H 


1 


1 












-M >0 


c^ 


(M 1 




1 


1 






+ + 


^ 




+ + + + + 1 


1 




















t- 


05 








t^ 




i-H 








<M 


CO 










Oj 














<x, 










-^ 




Ci o 


1 




lo CO 


O' tJH 


1 


I-H 


1 


I 










1 




^ t^ 


>o 


2 1 




1 


1 






+ 1 






1 1 


+ + 


^ 1 


1 




















10 


Oi 






"o a> -^ 












'* 


»o 


CO 






u u Xi 














<B 








-►^ ,0 3 




o o 


1 




CO GO 


JO -* 


1 CO 


I-H 


1 


1 


G ^-' o 




(N 


1 




r^ lO 


CO ->* 


cS i 




1 


1 


a-s -s 










(M rt 




















CO 






^ 




N 


, 




J3 






'^ 




^— V 


Ef tJD 






C 




CO 






eg 
Ci 












_8i 

5 s 


c 




^ ^ 






~a^ 


C (D 


^ 2 M 


'm 


QJ -H 


G 


o 

a 

a 


,G 

« 
C 


i5 ^ =« 

:G M_g 


"a- 


s 

O 

13 


'5c ^ 


^ -1 


-2 J? 

g tl G 




G 3 
^ in 
_g 

ce ~- 

G o~ 


-9 G 




o3 

a 


> M ft 


s 


o 


a Is 




s ■^ 2 


s 


"tS 7 


"3 S 








" <1 P 




H 


-S 


b o 


03 S Ph 


M 


UH <!' 


k?H C^ 



1-4 O 



P^ 



£ CI 






108 



1. MALONATE 





H 




:z 




H 




P5 




H 




P4 




p:^ 


Oi 


Q 


-^ 


s 


1^ 


o 


IJI 


P3 


m 


h 







w 






O 





^ 


O 


a 






"S 


o 


S 


(N 


>. 




Xi 




a 




_o 








'-ij 




^ 


^ 


2 


s 


c 




1— 1 


o 



CO -< 



O 



.a 

'-3 






-H i> 00 ^ 

CO ^ -H ^ 





1 

> 


o 






CI 

o 


3 


» 


'o 


-i: 




w 


o 
3 


5 


^ 


flH 


c 


M 


E^ 



t3 



SUBSTRATES OTHER THAN SUCCINATE 



109 



Table 1-22 

Urinary Excretion of Citrate Following Administration of the Sodium Salts 

OF Various Weak Acids 



Substance 



Urinary citrate " after dose '' of : 



4.35 



26 



26 



39 



None 

NaCl 

NaHCOg 

Malonate 

Succinate 

Pyruvate 

a-Ketoglutarate 

Fumarate 

Malate 

Oxalacetate 

Glutarate 

Adipate 

Maleate 

Acetate 

Citrate 

Aconitate 



0.47 


16 


0.50 


— 


0.90 


44 


14.76 


583 


14.80 


418 





336 


17.10 


275 


18.50 


287 


— 


216 


— 


1150 


— 


200 


61.30 


210 


1.53 


— 


15.2 


— 



413 



2.4 
9-25 
54.2 
29.5 
10.2 
12.3 
25.8 
13.1 

6.7 
44.3 

3.7 
11.4 



26.3 



10-17 

72.4 

39.4 

42.3 

36.6 

20.1 

45.7 

16.2 

65.8 

17.6 

45.0 

34.5 

52.5 

42.0 



" The second column shows the 24-hr urinary citrate (mg/kg) in dogs. From Orten 
and Smith, (1937.) 

The third column shows the urinary concentration of citrate (mg%) in rats. 
(From Simola and Kosunen, 1938.) 

The fourth and fifth columns show the 24-hr urinary citrate (mg) in rats. (From 
Krusius, 1940.) 

* Dose in miUimoles per kilogram. 



buted to alkalosis. Crawford (1963) confirmed that the injection of 10 mil- 
limoles/kg of malonate in rats causes a marked rise in urinary citrate 
(1.7 -> 47 /^moles/kg/hr), but found that succinate, malate, and sodium 
bicarbonate also have this effect. Serum citrate simultaneously rises very 
moderately. It was concluded that all the effects are due to an alkalosis 
induced by the administration of sodium. However, this cannot account for 
all of the actions of malonate, nor could it be responsible for the citrate ac- 
cumulation in cell suspensions and isolated preparations. 

Another possibility is that the citrate arises from the substances that 
are administered. This was favored by Orten and Smith (1937), but it was 



110 1. MALONATE 

difficult then with incomplete knowledge of the cycle and related pathways 
to understand how such conversions could take place; it still is. That is, 
if there is no significant impairment of the utilization of citrate by mal- 
onate, it is difficult to conceive of a pathway for the formation of citrate from 
malonate that would be so rapid as to lead to a large rise in the citrate level. 
A possibility that seems not to have been considered is that the substance 
determined as citrate may not have been citrate but a related tricarboxylic 
acid or some other compound giving a positive test. Although Hallman 
(1940) examined the specificity of the determination, there are many 
substances that have not been tested. For example, it is easy to formulate 
reactions in which malonyl-CoA could react with various carbonyl sub- 
stances, such as glyoxylate or pyruvate, to form tricarboxylate anions which 
might be oxidized to pentabromacetone in the citrate test and be mistaken 
for citrate. Certain dicarboxylates. such as itaconate, also are determined in 
this test. Such substances may not be readily metabolized and hence would 
accumulate much more readily than citrate. Although this posibility may 
seem far-fetched, it would be well to make certain that it actually is citrate 
that is accumulating during the action of malonate. It would be necessary 
to convert only a small fraction of the administered malonate to such a 
compound, since a dose of 26 millimoles/kg (see column 3 of Table 1-22) 
would theoretically give rise to almost 1 g of a substance with a molecular 
weight near that of citrate, whereas actually only around one-twentieth 
of this was determined as citrate. Of course, such estimations depend on the 
degree of sensitivity of the test to the compound. If there is any validity in 
this suggestion, it may be that depressions of citrate levels may occur in 
those preparations or tissues where such reactions of malonate or its meta- 
bolic products do not occur, i.e., where the response to malonate is the one 
expected on the basis of its inhibition of the functioning of the cycle. 

Accumulation of a-Ketoglutarate 

Very large increases in urinary a-ketoglutarate following the administra- 
tion of malonate to rabbits and rats were reported by Krebs et al. (1938) 
(Table 1-17) and this has been confirmed by El Hawary (1955), who found 
a 3.7-fold increase in serum a-ketogluterate 30 min after the intraperitoneal 
injection of 20 millimoles/kg malonate. As in the case of citrate, Krusius 
(1940) found that a-ketoglutarate excretion is increased not only by mal- 
onate but by many organic anions: malonate (46.4), maleate (42.0), malate 
(17.7), succinate (17.7), /5-hydroxybutyrate (15.7), acetate (14.3), pyruvate 
(12.1), fumarate (9.3), and sodium bicarbonate (0.6-7.8) (the figures give 
urinary excretion in milligrams/day). He concluded that essentially all the 
substances that increase citrate also raise the a-ketoglutarate excretion. 
However, glutarate is a notable exception, for it potently augments citrate 
formation but has no effect on a-ketoglutarate excretion. This would seem 



SUBSTRATES OTHER THAN SUCCINATE 111 

to disprove the theory that glutarate is active because it undergoes /?-oxi- 
dative decarboxylation to malonate. Little has been done with isolated 
preparations, but in three cases accumulation of a-ketogkitarate has been 
observed in the presence of malonate: in locust sarcosomes from malate 
(Rees, 1954), in suspensions of ascites cells from fumarate (Kvamme, 
1958 c), and in ascites cells and rat heart mitochondria from glutamate and 
malate (Borst, 1962), in all instances the malonate concentration being 
rather high (20-50 mM). 

The mechanism of such accumulation is poorly understood. We have seen 
that in some tissues the a-ketoglutarate oxidase can be inhibited by malo- 
nate, especially at concentrations above 10 mM, so that the results can be 
partially explained in this way (at least for the ascites cells and locust 
particulate preparations). The moderate increases in a-ketoglutarate excre- 
tion brought about by the cycle intermediates and substrates might well be 
due to a greater rate of formation with unchanged utilization rate. However, 
the possibility of a formation of c-ketoglutarate from glutamate cannot be 
ignored. It will be recalled that in potato slices the succinate formed in the 
presence of malonate comes partly from such a source (Romberger and 
Norton, 1961). Permeability effects causing a leakage of a-ketoglutarate 
and other anions from the tissues must also be considered. El Hawary (1955) 
found that several inhibitors (arsenite, maleate, iodoacetate, alloxan, and 
fluoride) increase the serum a-ketoglutarate levels in rats, and simulta- 
neously raise pyruvate levels. It may well be that any severe metabolic 
disturbance causes a release of cycle substrates from the tissues and an 
increased excretion, as well as secondary changes such as hyperglycemia. 

Effects on the Levels of Other Cycle Substrates 

The tissue concentrations of all the cycle substrates are probably altered 
by malonate and it is sufficient here to mention the results with potato tuber 
slices (Table 1-19) and tobacco leaves (Table 1-20). The reduction in the 
incorporation of C^** from glucose into malate in the former and the marked 
falls in malate level in the latter would be anticipated from a block of suc- 
cinate oxidation. Fumarate and oxalacetate levels probably are changed 
similarly. In this connection, one wishes that more information were avail- 
able on the factors that control the tissue pools of cycle intermediates, since 
it is evident that these substances do not occur only in the mitochondria 
in kinetic equilibria depending on the relative rates of the cycle reactions, 
but must also be present in cellular compartments. The transfer of the 
substances between these compartments and the active cycle regions must 
depend on processes that could be affected by inhibitors. Such compart- 
ments are well known in plant cells but it is probable that similar situations 
are applicable to animal cells. 



112 1. MALONATE 

Sequential Inhibition with Malonate and Fluoroacetate on Citrate Levels 

Before leaving the subject of accumulation of cycle intermediates, a 
few words must be said on the effects of malonate on the increases in citrate 
brought about by fluoroacetate. This was discussed in Volume I (page 502) 
and the results obtained by Potter (1951) presented (Fig. 1-10-5). The prin- 
ciple of the experiments is simply that fluoroacetate blocks the utilization 
of citrate so that the tissue citrate levels rise at different rates and to dif- 
ferent degrees. If malonate is administered to the animals prior to the 
fluoroacetate, the accumulation of citrate may be modified. These studies 
thus provide information on the effects of malonate on the rates of forma- 
tion of citrate and supplement the results discussed above. 

The tissues differ greatly in their response to malonate in the presence of 
fluoroacetate. In thymus, for example, malonate blocks the formation of 
citrate completely and no citrate accumulation at all occurs. The kidney 
behaves similarly but some citrate begins to accumulate an hour after 
the fluoroacetate is injected. In spleen and brain the inhibition of citrate 
formation is around 50%. Heart responds quite differently from the other 
tissues. Here malonate actually increases the citrate formation somewhat. 
Potter believed these results to indicate that in heart there are pathways 
other than the cycle for the synthesis of oxalacetate, and suggested the con- 
version of ketone bodies (malonate induces ketonemia) to citrate by a 
pathway involving oxalacetate, the ketone bodies arising from fatty acids. 
Liver was not studied but if similar reactions occur in this tissue, they 
could,. at least in part, account for the increases in serum citrate and urinary 
excretion of citrate. In any event, these experiments illustrate very well 
the inherent differences between tissues with respect to their metabolic 
pathways. The stimulation of citrate formation in the heart in the presence 
of fluoroacetate was confirmed by Fawaz and Fawaz (1954), whereas in the 
kidney only a depression was observed. The use of fluoroacetate to block 
the utilization of citrate is a useful technique by which to study the effects 
of malonate uncomplicated by possible effects on the rate of disappearance 
of citrate. 



ANTAGONISM OF MALONATE INHIBITION 
WITH FUMARATE 

The overcoming of an inhibition by the addition of an intermediate nor- 
mally arising distal to the site of the block is often excellent evidence 
for the locus of action of the inhibitor and for the specificity of the inhi- 
bition. For this reason, fumarate has frequently been used in malonate- 
inhibited preparations and a reversal of the inhibition taken as proof for 
the specific action of malonate on succinate oxidase. The first acceptable 



ANTAGONISM WITH FUMARATE 113 

data for fumarate reversal were reported by Quastel and Wheatley (1935), 
who showed that the inhibition of acetoacetate utilization by rat liver in 
the presence of malonate is mainly abolished when fumarate is added, 
and they used this as evidence that malonate does not act directly on the en- 
zyme involved in the breakdown of acetoacetate. The use of fumarate im- 
mediately became popular and many studies since 1935 have included its 
addition for the purpose of demonstrating that the observed effect of mal- 
onate is indeed due to its block of succinate oxidation. 

Most of the tests for fumarate reversal, it must be admitted, have not 
been done properly and the results have been evaluated uncritically, so that 
little of value has been demonstrated. There are several points that must 
be considered in the planning and interpretation of such experiments. 

(a) An increase in the oxygen uptake upon addition of fumarate to a 
malonate-inhibited preparation is not, by itself, very meaningful. Let us take 
a typical experiment similar to many reported in the literature. The nor- 
mal Qq of a tissue preparation is 10 and this is decreased to 4.5 by mal- 
onate. When malonate and fumarate are added together, the Qq is 8.7. 
It has been generally stated in such cases that fumarate is capable of 
antagonizing the malonate inhibition. Actually, an increase in the respira- 
tion could have been brought about by the addition of any substrate that 
is oxidized by the preparation, including substrates completely unrelated 
to the cycle. One can say that fumarate has overcome the malonate inhibi- 
tion but the results are of no particular significance with respect to malonate 
specificity or the site of the inhibition. What one has shown is that fumarate 
can be oxidized in the presence of malonate and this need not imply that 
the operation of the cycle has been even partially restored. If fumarate 
had been added to the uninhibited preparation and a Qq of 14.2 found, the 
following might have been concluded: malonate inhibits the endogenous 
respiration 55% and the respiration in the presence of fumarate either 13% 
(with respect to the endogenous control) or 39% (with respect to the rate 
with fumarate alone present), in both cases the inhibition being less than that 
for the endogenous respiration. Actually, it has been shown that the oxi- 
dation of fumarate is unaffected by malonate, since the same absolute 
rise in the Qq^ is obtained from fumarate in the uninhibited and inhibited 
preparations. If the oxygen uptake resulting from the addition of fumarate 
results mainly from the oxidation of malate, the results would have little 
bearing on the mechanism or selectivity of malonate inhibition. 

(6) The addition of fumarate would not in any case restore the complete 
cycle, since oxidation of succinate would still be depressed and less energy 
would be available from this step. It is possible in some instances that the 
energy from succinate oxidation, rather than the over-all energy production 
from all oxidations, is important, and this would not be restored by fu- 
marate. 



114 1. MALONATE 

(c) The accumulation of succinate due to malonate inhibition would, of 
course, not be reversed by fumarate; instead, it is usually increased. If some 
response to malonate is dependent on the rise in succinate concentration 
(e.g., a direct effect of succinate on other enzymes or some cell function, or 
the increased formation of some substances derived from the succinate), 
this would not be antagonized by fumarate. 

(d) The response to fumarate will often depend on what is being measured. 
One example will be used here to illustrate this and others will be men- 
tioned later. Malonate inhibits the oxidation of trilaurin and octanoate 
in liver and kidney slices, and also reduces the amount of C^^Og formed from 
labeled substrates (Geyer et al., 1950 a). It was found that fumarate is very 
ineffective in counteracting the inhibition of C^^O, production, and this 
might be attributed to a direct effect of malonate on fatty acid oxidations. 
However, the results can be explained on the basis of an inhibition of suc- 
cinate oxidase. Much of the C^* taken into the cycle from acetyl-CoA 
would accumulate in succinate and fumarate would have no effect on this. 
For the full release of CO2, the operation of the entire cycle is required, and 
thus fumarate would increase the C^^Og formation only moderately in the 
presence of malonate. If the oxygen uptake had been determined, fumarate 
could well have shown a complete reversal of the malonate inhibition. 

(e) Absence of a reversal by fumarate, or the failure to achieve a complete 
reversal, can be due to a variety of causes. In an experiment, such as the one 
shown in the following tabulation, done on Aplysia muscle slices: 



Additions Qq^ 



Endogenous 


0.33 


Succinate 


1.80 


Malonate 


0.33 


Succinate -f malonate 


0.96 


Succinate + malonate + fumarate 


0.90 



although it was stated that fumarate was unable to relieve the malonate 
inhibition (Ghiretti et al., 1959), it may simply be that fumarate would not 
have been oxidized if added to the uninhibited tissue. Certainly the direct 
inhibition of succinate oxidase would not be expected to be overcome by 
fumarate, but in any case a control with fumarate alone must be run to 
give any significance to the results. It is even possible for fumarate to in- 
crease the inhibition produced by malonate. This was observed for the respi- 
ration of Helix hepatopancreas (see accompanying tabulation) (Rees, 1953). 
Here the endogenous respiration apparently is little dependent on the cycle, 
whereas upon the addition of fumarate, through the formation of oxalace- 



ANTAGONISM WITH FUMARATE 115 

tate and its partial decarboxylation to pyruvate, the cycle presumably 
becomes activated. The high inhibition of fumarate oxidation by malonate 
cannot, of course, be attributed solely to an inhibition of succinate oxidase, 

Additions Og uptake % Inhibition 



Endogenous 


30 


Malonate 


25 


Fumarate 


167 


Malonate + fumarate 


47 



16.7 
71.9 



and an effect on oxalacetate decarboxylase or some other enzyme at the 
relatively high malonate concentration (33 mM) must be assumed. It is 
desired to point out that the effects with fumarate are unrelated to the 
inhibition of the endogenous respiration observed with malonate and 
throw no light on the mechanism of the inhibition. 

Let us now turn to some experiments in which the addition of fumarate 
provides results indicative of the mechanism of malonate inhibition. The 
most significant studies usually have involved the determination of the 
utilization or formation of a particular substance. In the original work 
of Quastel and Wheatley (1935) mentioned above, the disappearance of 
acetoacetate in rat liver slices was determined, and fumarate was found to 
increase its utilization in the presence of malonate, whereas in the absence 
of malonate it had essentially no effect. Krebs and Eggleston (1940) likewise 
showed that fumarate would increase the utilization of pyruvate in pigeon 
muscle in the presence of malonate, at high malonate concentrations the 
amount of pyruvate utilized being equivalent to the fumarate added. The 
results obtained by Stare et at. (1941) with pigeon muscle show definitely 
that the block of pyruvate utilization produced by malonate is effectively 
overcome by fumarate (see accompanying tabulation). Results such as 



Additions PjTuvate utilized 

Endogenous 16 . 3 

Malonate 7 . 2 

Fumarate 21.4 

Malonate + fumarate 23.3 

these demonstrate that fumarate can counteract the effect of malonate on 
a specific process and, from the nature of the experiments, it is likely that 
the cycle block is being overcome in a sense, that is, that oxalacetate is 
being made available for condensation with acetyl-CoA. They also provide 



116 1. MALONATE 

some information on the site and specificity of the malonate inhibition. 
Quite different results on acetate utilization by yeast were obtained by 
Stoppani et at. (1958 b), who found that fumarate is completely unable to 
reverse the inhibition by malonate (see accompanying tabulation). These 

Additions Acetate utilized 



Acetate 30 
Acetate + malonate 4.4 

Acetate + fumarate 27 
Acetate + malonate + fumarate 3.8 



data would nuply that the inhibition of acetate utilization by malonate is 
not mediated through a block of succinate oxidation but by another jnecha- 
nism. It is difficult to explain this by a failure of fumarate to penetrate 
into the cells since malonate seems to enter readily. It was suggested that 
malonate might interfere with acetate activation by depleting the system 
of coenzyme A, due to the formation of relatively stable malonyl-CoA. 
Whatever the mechanism, these results point to an action of malonate 
other than on succinate oxidase. 

An interesting illustration of the useful information that may be ob- 
tained from the use of fumarate is given in the inhibition of urea formation 
by malonate. The formation of both arginine and urea from citrulline and 
glutamate in liver homogenates is potently inhibited by malonate (Cohen 
and Hayano, 1946; Krebs and Eggleston, 1948). Fumarate is able to counter- 
act this block completely. These results were difficult to understand ini- 
tially, but it is now known that transamination must occur between gluta- 
mate and oxalacetate to form aspartate, which reacts with the citrulline 
to form arginosuccinate, from which arginine and urea are derived. The 
effect of malonate is to reduce the supply of oxalacetate for transamination 
and it is clear why fumarate will abolish this inhibition (Ratner, 1955). 
Krebs and Eggleston (1948) observed the formation of 3-5 molecules of 
urea for each molecule of fumarate added. This may be explained by the 
fact that the formation of arginine from arginosuccinate involves the release 
of fumarate, which can again go to oxalacetate. 

The demonstration that fumarate will counteract the inhibitory action 
of malonate on some tissue function is indicative of a primary block of the 
succinate oxidase, but even complete reversal does not prove a specific ac- 
tion of malonate. One example would be the inhibition of Br~ uptake in 
barley roots by malonate (Machlis, 1944). Malonate (10 mM) inhibits the 
uptake 57% but if fumarate is present the uptake is 35% above the control. 
On the surface this would imply an effective reversal of the malonate inhi- 



SPECIFICITY OF MALONATE INHIBITION IN THE CYCLE 117 

bition, but actually fumarate alone increases the uptake to 71% above the 
control. Thus malonate inhibits a significant amount even in the presence 
of fumarate, pointing to an action other than on succinate oxidation. 
Another example would be the malonate inhibition of cell division oiArhacia 
eggs (Barnett, 1953). Cleavage is inhibited almost completely by 60 milf 
malonate and an equimolar concentration of fumarate abolishes this. It is 
likely that fumarate overcomes the cycle block but one cannot conclude 
that the action of malonate is specific on succinate oxidase. Malonate, can 
also inhibit to some extent steps in the utilization of fumarate, but due to 
the high concentration of fumarate enough oxalacetate is formed to allow 
cleavage. Indeed, succinate atGOmJf also abolishes the inhibition, indicating 
that as a result of the competitive nature of the inhibition enough succinate 
has broken through the block to restore cleavage. It must be remembered 
that a complete reversal of a metabolic block is not always necessary for a 
cell function to proceed normally. 

SPECIFICITY OF MALONATE INHIBITION IN THE CYCLE 

At this point we may summarize some of the conclusions with respect to 
the specificity of action of malonate on the succinate dehydrogenase. Possi- 
ble effects of malonate outside the cycle will be discussed in a later section. 
We have seen that malonate can inhibit enzymes other than succinate dehy- 
drogenase rather potently (Table 1-12), that the oxidations of certain cycle 
substrates are suppressed more than predicted on the basis of a selective 
action on succinate oxidation (Table 1-14). that the accumulation patterns 
of cycle intermediates are sometimes distorted by malonate in ways implying 
inhibition at more than one site, and that fumarate is seldom able to reverse 
the actions of malonate completely. Of particular significance are the clear 
demonstrations of the inhibition of reactions related to the entry of acetyl- 
CoA into the cycle, particularly those of Pardee and Potter (1949) pointing 
to an inhibition of the condensation reaction, those of Stoppani et al. 
(1958 b) reporting a marked inhibition of acetate utilization apparently 
unrelated to an inhibition of succinate oxidase, and our own results on rat 
heart mitochondria where malonate inhibits p^Tuvate oxidation in the 
presence of malate and with the a-ketoglutarate oxidase completely blocked. 
Another susceptible site is a-ketoglutarate oxidation, especially in view 
of the clear proof by Price (1953) that specific inhibition can not even be 
obtained in the simple system, 

a-Ketoglutarate -^ succinate ->■ fumarate 

There is thus a large amount of evidence that malonate can at certain 
concentrations in various conditions inhibit other reactions than succinate 
oxidation. 



118 1. MALONATE 

It would be convenient if one could specify a malonate concentration, 
or range of concentrations, which would most likely be specific, but there 
are too many factors involved to do this with any confidence. Some of the 
factors may be listed: (a) the species, tissue, or preparation used, (b) the 
enzymes or metabolic pathways involved in what is measured, (c) the 
conditions of the experiment, e.g., the pH or the Mg++ concentration, (d) 
the degree to which succinate can accumulate and antagonize the inhibi- 
tion, (e) the effective concentration of malonate within cells, and (f) the 
possibility of nonenzyme effects on cell membranes or other structures. 
Specificity for an inhibitor is not a constant to which can be given a value 
for all cases, but a characteristic that must be evaluated for each experi- 
ment. Aside from the direct actions of malonate on enzymes, there are the 
problems of metal cation depletion, the possible effects of Na+ or K+ added 
with the malonate, and the inactivation of the coenzyme A in some pre- 
parations through the formation of malonyl-CoA. 

Bearing in mind these difficulties, a few general remarks may be made. 
It is probable that malonate usually does not inhibit any cycle enzyme more 
strongly than succinate dehydrogenase, so that the major effect will be re- 
lated to the inhibition at this site, but it will be recalled that in certain 
species the succinate dehydrogenase is not very susceptible to malonate. It 
is impossible to achieve a nearly complete block of succinate oxidation 
without affecting other cycle reactions; if one wishes a specific effect, one 
must be satisfied with a moderate inhibition of succinate oxidation. If a 
single malonate concentration for general usefulness had to be chosen, 5 mM 
might be taken provisionally. Although in some cases rather incomplete inhi- 
bition will be obtained, this concentration will probably not inhibit other 
cycle reactions significantly. This applies to noncellular preparations where 
penetration is not a factor but otherwise higher external concentrations 
may have to be used. In any case, good evidence for a specific action should 
be obtained under the conditions of the investigation, and reliance should 
not be based on generalities. 

EFFECTS OF MALONATE 
ON OXIDATIVE PHOSPHORYLATION 

An uncoupling action on oxidative phosphorylation has been claimed for 
malonate several times and it is important to determine if this is actually 
so. This is quite difficult because pertinent or reliable data are generally 
lacking. There are four important ways in which malonate could alter the 
P : ratio: (1) a direct effect on the coupling between oxidation and phos- 
phorylation, (2) an alteration of the pattern of substrate oxidation, since 
different substrates may have different P : ratios, (3) a differential inhi- 
bition of electron transport pathways for a single substrate but with differ- 



EFFECTS ON OXIDATIVE PHOSPHORYLATION 119 

ent P : ratios, and (4) an effect on the hydrolysis of ATP or other high 
energy substances. Only the first mechanism should be considered as true 
uncoupling, although it is often difficult to determine the exact mechanism. 
Included in the first mechanism would be the chelation of malonate with 
Mg++ or Mn++, since these cations are beheved to be cofactors in phospho- 
rylation. 

Claims for an uncoupling action will be discussed first. Lehninger (1951) 
stated that malonate uncouples oxidative phosphorylation associated with 
the oxidation of /5-hydroxybutyrate by rat liver mitochondria but no data 
were given. In a previous report (Lehninger, 1949) 7.5 mM malonate was 
shown to inhibit oxygen uptake 48%, the formation of acetoacetate from 
/?-hydroxybutyrate 64.7%, and phosphorylation 37.3% (as determined by 
the incorporation of P''- into the ester fraction). Since phosphorylation is 
inhibited less than oxidation, no uncoupling is evident, and indeed the P:0 
ratio should increase. Berger and Harman (1954) claimed that malonate 
inhibits phosphorylation associated with the one-step oxidation of a- 
ketoglutarate and completely suppresses phosphorylation during the oxida- 
tion of L-glutamate by muscle mitochondria. However, the malonate con- 
centration is not given and the absence of data prevents evaluation of the 
results. An inhibition of phosphorylation does not necessarily mean an 
uncoupling action. Malonate at 30 milf drops the P:0 ratio from 1.6 to 0.5 
in the oxidation of choline by rat liver mitochondria (Rothschild et al., 
1954). Although no control with malonate alone was reported, it would 
appear that this is the most valid instance of uncoupling by malonate. 
The malonate concentration was high and a reduction of Mg++ (total con- 
centration was 5.7 mM) must be considered. Finally, the phosphorylation 
associated with succinate oxidation in lupine mitochondria was shown to 
be strongly depressed by 10 mM malonate (Conn and Young, 1957), but 
it may be observed that the oxygen uptake was inhibited even more (Ta- 
ble 1-23), so that no uncoupling occurred. 

All of the reports in which P:0 ratios were calculated are summarized 
in Table 1-23 and in all cases, except for E. coli, Carcmvs maenas, and the 
oxidation of choline, it is seen that the P:0 ratio is actually increased by 
malonate. However, most of these only illustrate mechanism (2) above, 
because in the oxidation of a-ketoglutarate a rise in the P:0 ratio would 
be expected upon blocking succinate oxidase due to the fact that P:0 for 
succinate oxidation is 2, whereas for the one-step oxidation of a-ketoglu- 
tarate to succinate usually it is experimentally between 3 and 4. All one can 
say from such data is that there is no evidence for an uncoupling action by 
malonate. Copenhaver and Lardy (1952) used 3-20 mM malonate in all 
their media in the study of the phosphorylation associated with a-keto- 
glutarate oxidation and obtained high P:0 ratios, again providing evidence 
against any uncoupling activity. It was shown by Slater and Holton in 



120 



1. MALONATE 



05 



pq 



\4 



m 



M 



X 



CO -^ t^ o 00 o 

do d -H d ^ 



lO CO C3 CO o >c 
—I CO — I ■* iM (M 



•M IC C^ C2 CO 

^ rt ^ d -H 



^ 



+ 



Cm 



1^ 



W 



o 



^ 



^ 



H 



fe] 



s 


c 


c 


03 


t>n 


Xi 


S 




O 


C 


t<n 


P 


J 


s 


tin 


~ 



§^^ 

^ 



p* 


o 


<5 


p. 


M 


+3 










1 


03 


►-H 





EFFECTS ON OXIDATIVE PHOSPHORYLATION 121 



'-~. lO 



Oi 


ft 


I— ( 


ft 


— 


o3 


^^ 


H 


-t^ 


t3 


cS 




w. 


c3 










T3 


C 


fi 


c3 


cS 





Q 



t4 



o -^ ao o;^ft-c_go 







GO 
05 


CO 




'M 
iM 




CO 

C5 


^ 00 

-H CO 


00 


o 


CO 


cc 


00 

CO 


o ^ 


o 


§ 




CO 


CO 




00 

CO 


CJ 


•M 


-H 


o 


o 


S^ 


CO 


CO 


fc ec 


CO 


CO 


CO 


CO 


CO 


CO C^l 


-H 


o 


rt C^l 


o 


o 


o 


o 



I I I 



+ I I I I I I I I I I + 



a a as csccOs 



?^ 


?i< 


^^ 






;>■. 




o 


rV. 






> 














^ 




cs 


5^ 


x> 


>i) 





o^o^^d PM pm § 0^ Ph p5 m 



122 1. MALONATE 

rat heart (1954) that malonate increases the P:0 ratio with a-ketoglutarate 
as the substrate as the malonate concentration is increased up to 20 mM; 
at 40 mM the P:0 ratio decreases somewhat, so that at this high concen- 
tration a small degree of uncoupling may occur. Azzone and Carafoli (1960) 
found the same in pigeon muscle. Evidence against uncoupling by malonate 
intracellularly is provided by the study on ascites carcinoma cells by 
Greaser and Scholefield (1960). Comparison of the changes brought about 
by malonate (20 mM) in the respiration and the sum of the concentrations 
of ADP and ATP led to the estimation of a 15% increase in the P:0 ratio, 
whereas the classic uncoupler, 2,4-dinitrophenol, depressed the P:0 ratio 
70%. Malonate may actually stimulate the esterification of phosphate. In 
addition to the two examples in Table 1-23, Jackson et al. (1962) reported 
an 8% elevation in phosphate uptake by 0.1 mM malonate in barley root 
mitochondria oxidizing succinate, and Rosa and Zalik (1963) found such a 
stimulation in pea seedling mitochondria, which is maximal around 0.01 
mM malonate, the oxygen uptake not being altered. Above this concentra- 
tion, both phosphorylation and oxidation are depressed in a parallel fashion; 
respiration is hence always depressed somewhat more than is phosphate 
esterification, and no uncoupling is seen at any malonate concentration. 
The transfer of phosphate from phosphorylated coenzyme Q to ADP to 
form ATP in mitochondrial preparations is not altered by even 10 milf 
malonate (Gruber et al., 1963), whereas dinitrophenol inhibits this readily. 
It would therefore appear to be legitimate to conclude that malonate 
does not exhibit uncoupling activity except possibly at high concentrations 
(above 30 mM). The evidence for uncoupling at concentrations commonly 
used is considered to be inadequate and outweighed by the mass of indirect 
evidence that P:0 ratios are not reduced in the oxidations of citrate, a- 
ketoglutarate, and succinate. 



EFFECTS OF MALONATE ON GLUCOSE METABOLISM 

The actions of malonate on carbohydrate, lipid, amino acid, porphyrin, 
and other types of metabolism will now be considered, after which it will be 
possible to evaluate the specificity of malonate more broadly. The effects 
on glucose metabolism are important but difficult to analyze. In the first 
place, the interrelationships between the cycle and the glycolytic pathways 
are complex and secondary effects on glucose utilization must be expected. 
In the second place, much of the work on the alteration of glucose utiliza- 
tion by malonate has not been adequate for the determination of mecha- 
nisms nor do the results even provide useful information in many cases, and 
for this reason only certain reports will be discussed. There are three basic 
ways by which malonate, might alter glucose oxidation. (1) Malonate will 
usually cause a depression of the oxygen uptake related to glucose metabo- 



EFFECTS OF MALONATE ON GLUCOSE METABOLISM 123 

lism because in the total oxidation of glucose to CO, and water, five-sixths 
of the oxygen is taken up in cycle reactions. This effect would be due to a 
block of the cycle and would be roughly equivalent to the inhibition of 
of pyruvate oxidation via the cycle. (2) Malonate may cause an increased 
uptake and utilization of glucose as a result of inhibition of oxidative reac- 
tions in the cycle, the mechanism being essentially that of the Pasteur reac- 
tion. (3) Malonate may have direct actions on the cellular uptake of glucose 
or on the glycolytic pathways. Evidence for each of these mechanisms will 
be presented, and then a more detailed analysis of the possible shifts in 
metabolic patterns brought about by malonate will be given. 

Effects on Oxygen Uptake Associated with Glucose Oxidation 

There are many reports on the effects of malonate on respiration with 
glucose as the substrate, but in few have the data provided a correction for 
an effect on the endogenous respiration. Indeed, an endogenous correction 
to determine the oxygen uptake associated only with glucose oxidation 
is particularly unreliable, even when the data are available. This is because 
of the well-known effect of glucose in suppressing endogenous respiration and 
mitochondrial oxidations (Crabtree effect). Thus the nonglucose respiration 
may be significantly changed when glucose is added and, perhaps, completely 
suppressed in some cases. 

Let us assume that the oxygen uptake measured derives only from glucose 
and that malonate acts only on the cycle. What inhibitions could one theo- 
retically expect? The inhibition will depend, other than on the degree of 
cycle block, on the final products of glucose oxidation before and after the 
addition of malonate. The following equations give the molar oxygen up- 
takes for the oxidation of glucose to varying degrees of completeness: 

To CO2 and water 

CeHi^Oe + 6 O2 ^ 6 CO, + 6 H,0 

To pyruvate 

CgHi^Oe + O2 -> 2 CH3COCOOH + 2 H2O 

To acetate 

CeHiaOe + 2 0, -> 2 CH3COOH -f- 2 CO, + 2 H^O 

To succinate 

CfiHi^Oe + 5/2 O2 -> HOOCCH2CH2COOH + 2 CO2 + 3 H2O 

CsHiaOg + 2 HOOCCH2COCOOH + 4 0, -> 2 HOOCCH2CH2COOH + 6 CO2 + 4 H^O 

To lactate 

CsHi^Oe + 2 H -). 2 CH3CHOHCOOH 

The second equation for the formation of succinate might express the situa- 
tion in which there is a source of oxalacetate other than from pyruvate 



124 1. MALONATE 

carboxylation, whereas the first equation assumes no external oxalacetate 
source. If the inhibition on succinate oxidase is complete and glucose is 
oxidized completely in the control, we may calculate the inhibitions for the 
situations where in the presence of malonate the glucose is transformed into 
the various oxidation products: to succinate with external oxalacetate 
source (33%), to succinate in a closed system (58%), to acetate (67%), 
to pyruvate (83%), and to lactate (100%). Of course, experimentally it is 
probably impossible to achieve a complete cycle block with malonate, so 
that the inhibitions will usually be less than the theoretical. These calcula- 
tions also assume that the rate of glucose disappearance is not altered by 
the inhibition. The degree of inhibition, therefore, even under these simple 
conditions, will vary a good deal, depending on the terminal product of 
glucose metabolism and thus on the enzymes and pathways occurring in 
the cells under consideration. Actually, it is likely that various substances 
will accumulate during glucose metabolism, and not only those given above 
but others into which these substances can be metabolized. 

A strong inhibition of the oxygen uptake related to glucose metabolism 
probably indicates the participation of the cycle in the oxidation, at least 
when the malonate concentration is not unreasonably high. Thus, in the 
instances shown in the tabulation and selected at random of inhibition of 



Preparation 




Malonate 
(mif) 


% 


Inhibitior 


1 Reference 


Sarcina lutea 




10 




50 


Dawes and Holms (1958) 


Yeast 




12.5 




73 


Krebs et al. (1952) 


Bull sperm 




10 




57 


Lardy and Phillips (1943 a) 


Malarial parasitized 


RBC 


20 




69 


Speck et al. (1946) 


Brain mince 




17 




100 


Huszak (1940) 


/iL+-Stimulated brain slices 


10 




62 


Kimura and Niwa (1953) 



glucose respiration, one would conclude that the role of the cycle is impor- 
tant, but no information on the exact mechanism or on the fate of glucose 
in the presence of malonate may be obtained. However, in the cases of brain 
mince and yeast, fumarate is unable to overcome the inhibition, so that 
some doubt is introduced even into this interpretation. The inhibition of 
the glucose oxygen uptake in parasitized erythrocytes is very similar to 
that for the oxygen uptake associated with pyruvate oxidation, as expected 
if the cycle is the terminal pathway for glucose metabolism. On the other 
hand, there have been many reports that malonate, even at high concentra- 
tions, does not inhibit the glucose respiration at all, and in these instances 
little can be concluded because of the possibility that malonate does not 
penetrate and that an adaptation of the glucose utilization occurs. The 



EFFECTS OF MALONATE ON GLUCOSE METABOLISM 125 

endogenous respiration of Ehrlich ascites cells is inhibited around 35% by 
30 vaM malonate, but when glucose is present there is very little effect of 
malonate (Seelich and Letnansky, 1960). Put in another way, in the pres- 
ence of malonate the addition of glucose increases the 0, uptake somewhat 
instead of depressing it as it does in uninhibited cells. Another possible 
reason for a failure of malonate to depress glucose respiration is the oc- 
currence of oxidative pathways unassociated with the cycle. Caldariomyces 
fumago has no hexokinase and the usual Embden-Meyerhof glycolytic 
pathw^ay is absent; the oxidation of glucose occurs more directly through 
glucose and gluconate oxidases (Ramachandran and Gottlieb, 1963). The 
respiration in this organism is not affected by malonate at 10 mM. 

Effects on Glucose Utilization 

A depression of cycle oxidations by malonate would be expected to cause 
an increased utilization of glucose, as do hypoxic conditions, in cells capable 
of exhibiting a Pasteur reaction. This is a manifestation of one interrela- 
tionship between the cycle and the glycolytic pathway, and is partially 
mediated through changes in the concentrations of inorganic phosphate 
and phosphate acceptors, such as ADP. For such a response to occur, the 
utilization of glucose must have been initially limited by the coupled 
phosphorylation at the 3-phosphoglyceraldehyde oxidation step. Another 
mechanism for the influence of cycle activity on glucose utilization involves 
the membrane transport of glucose and its phosphorylation. The inward 
transport of glucose under certain conditions may limit glucose utilization, 
and it has been shown that anoxia accelerates this transport in rat heart 
(Morgan et al., 1961 a, b). It is quite possible that malonate by its depres- 
sion of cycle activity, can alter these transport processes. A third mechanism 
might involve the rate of oxidation of NADH. The addition of pyruvate to 
ascites cells stimulates the formation of C^^02 from labeled glucose (Wenner 
and Paigen, 1961) Initially the rate of pyruvate oxidation is limited by the 
rate of NADH oxidation and the exogenous pyruvate acts as an electron 
acceptor, 1 mole of lactate appearing for each mole of pyruvate oxidized 
through the cycle. The accumulation of pyruvate as a result of the cycle 
block by malonate could initiate this dismutation so that more glucose 
would be utilized than otherwise. Indeed, lactate formation is often increased 
during malonate inhibition. In all these ways, and perhaps others, a cycle 
block might affect glucose uptake and utilization. 

A stimulation of glycolysis by malonate was first observed by Kutscher 
in Heidelberg (Kutscher and Sarreither, 1940; Kutscher and Hasenfuss, 
1940). Malonate was injected into guinea pigs, the muscle removed later, 
and the formation of lactate determined in a brei. In some cases, glucose, 
succinate, or fumarate was also injected. Malonate accelerates lactate for- 
mation and this is overcome by both succinate (see accompanying tabula- 



126 1. MALONATE 

tion) and fumarate. It was also shown that the rate of glycogen disappear- 
ance in muscle brei is stimulated some 35% by malonate. These responses 
were equated with the Pasteur reaction. Similar effects have been found in 



Injection 


Lactate formation 


Control 


2.95 


Malonate 


9.2 


Glucose 


5.1 


Glucose + malonate 


9.4 


Succinate + malonate 


-1.0 


Glucose + succinate + malonate 


-0.1 



other tissues more recently. The rates of glucose utilization and of lactate 
formation in brain slices are stimulated 13% and 44%, respectively, by 
10 mM malonate (Takagaki et at., 1958), an effect much like that produced 
by azide. Simultaneously less glucose is oxidized, the increased utilization 
being diverted to lactate. Ehrlich ascites tumor cells exhibit a more rapid 
rate of glycolysis when the oxygen tension is reduced and a similar response 
is seen with malonate (Kvamme, 1957, 1958 d). When fumarate is added 
to the malonate-blocked cells, glucose utilization and lactate formation are 
suppressed. Strictly speaking, it is fumarate that gives rise to a Pasteur 
reaction in the presence of malonate, just as the addition of oxygen does 
in preparations previously anaerobic. Kvamme obtained data which led him 
to conclude that this effect is mediated through changes in the concentra- 
tions of inorganic phosphate and phosphate esters. The relationships 
between glycolysis and mitochondrial oxidations are particularly well seen 
in the reconstructed systems of Aisenberg et al. (1957). A supernatant frac- 
tion from rat liver forms lactate from glucose and the addition of a mito- 
chondrial suspension suppresses this markedly. Malonate partially prevents 
this suppression; that is, added to the complete system, lactate formation is 
increased. This stimulation of glycolysis occurs despite the fact that mal- 
onate inhibits the glycolytic rate 15.7% in the supernatant fraction. This is 
a good illustration of how malonate can produce different effects on glucose 
metabolism, depending on the conditions and the factors controlling gly- 
colysis. Stimulation of aerobic glycolysis by malonate, and a variety of 
other inhibitors, is particularly well seen in thymocytes; malonate ac- 
celerates lactate formation sigmoidally from 3 mM to 100 mM (Araki 
and Myers, 1963). 

Malonate may have no effect on glucose uptake, or may inhibit it, in 
other tissues. Chick chorioallantoic membrane infected with influenza 
virus exhibits a 47.5% reduction in the endogenous respiration in the pre- 
sence of 6 mM malonate, and the final virus titer drops to zero, but the 



EFFECTS OF MALONATE ON GLUCOSE METABOLISM 127 

uptake of glucose is unaffected (Ackermann, 1951). An example of a marked 
inhibition was reported for guinea pig brain slices, glucose utilization being 
reduced by 10 mM malonate in both normal and K+-stimulated slices (see 
accompanying tabulation) (Tsukada and Takagaki, 1955). It may be ob- 



KCl 

(70 mM) 


Malonate 

(10 mM) 


Glucose 
utilization 
(//mole/g) 


Lactate 
formation 
(//mole/g) 


O2 uptake 
(//mole/g) 






35.9 


34.9 


54.1 


— 


+ 


4.71 


28.0 


61.5 


+ 


— 


112 


99.0 


104 


+ 


+ 


63.7 


117 


44.9 



served that the amount of lactate formed per glucose consumed is increased 
by malonate, but some of the lactate must be derived from other sources 
than glucose. It is difficult to understand the discrepancy between these 
results and those obtained later on the same tissue (Takagaki et al., 1958), 
where a slight stimulation of glucose utilization by 10 mM malonate was 
reported. There are so many different results obtained relative to the utili- 
zation of glucose that it often appears each organism or tissue exhibits a 
characteristic pattern of response. Malonate at 40 mM has relatively little 
effect on the uptake of glucose by brain and kidney slices, depressing it 
slightly in the former and perhaps accelerating it in the latter, but reduces 
the C^^Og formed from uniformly labeled glucose 79% and 52%, respective- 
ly, the O2 uptake being suppressed comparably (Cremer, 1962). Since much 
less of the glucose goes to amino acids and proteins in the presence of mal- 
onate (Cremer, 1964), it is likely that here there is an accumulation of 
certain cycle intermediates, such as succinate, and of lactate. Succinate is 
normally formed and released into the medium by Trypanosoma cruzi. 
It is formed aerobically through both the glycolytic pathway and the cycle, 
and by CO2 fixation; some must be metabolized through the cycle since 
malonate elevates the succinate level even further (Bowman et al., 1963). 
The uptake of glucose is increased almost 20% by malonate but there is no 
change in C^^Oo formation, most of the excess glucose probably appearing 
as succinate, acetate, and related anions. 

Direct Effects on the Glycolytic Pathways 

There is no evidence that any enzyme of the Embden-Meyerhof glycolytic 
pathway is significantly inhibited by malonate at concentrations below 20 
mM (Table 1-12), although some of the enzymes have never been investigat- 
ed. Several of the enzymes in this pathway require Mg++, or a related cation. 



128 1. MALONATE 

and under certain conditions malonate could inhibit through the chelation 
of these ions. A possibly sensitive enzyme is lactate dehydrogenase, which 
functions in anaerobic glycolysis, although insufficient quantitative work 
has been done, and it is likely that the enzyme from different sources would 
be inhibited to different degrees by malonate. It is unfortunate that so 
little work on the effects of malonate on anaerobic glycolysis has been done, 
since studies under aerobic conditions are always complicated by the sec- 
ondary reactions discussed above. In a number of cases, an inhibition of 
aerobic glycolysis has been observed, i.e., a decreased formation of lactate 
in air or 95% oxygen, and these inhibitions are usually small. In bull sperm 
(Lardy and Phillips, 1943 b), 10 mM malonate inhibits 6.7%; in beef 
thyroid homogenates (Weiss, 1951), 33 mM malonate inhibits 8%; and in 
rat liver supernate (Aisenbergef oi., 1957), 25 mM malonate inhibits 15.7%. 
Although rises in lactate can be explained on an indirect basis, a decrease 
in the rate of lactate formation must usually be attributed to some inhibi- 
tion along the glycolytic pathway, since it is not very likely that malonate 
would shift the fate of pyruvate from lactate to the cycle. Greater inhibi- 
tions have been observed: in mouse brain, 40 mM malonate inhibits lactate 
formation 57% and in mouse liver mitochondria 100%, this being taken as 
evidence of some direct effect on glycolysis (du Buy and Hesselbach, 1956). 
This concentration is, of course, rather high and could have depleted the 
Mg++ from either the 3-phosphoglyceraldehyde dehydrogenase or enolase 
systems, or could have inhibited lactate dehydrogenase. Since the substrate 
in these cases was 3-phosphoglyceraldehyde, an action earlier in the pathway 
is impossible. There are also miscellaneous reports which might be inter- 
preted as indicating an inhibition of glycolytic pathways, for example the 
results of Greville (1936) on rat brain, in which 20 mM malonate inhibits 
glucose oxidation around 50% and pyruvate oxidation only around 15%. 
Only one instance of a direct test on anaerobic glycolysis has come to my 
attention, that of Covin (1961), who found that 5 mM malonate inhibits 
lactate formation in rat ventricle slices, but the rate of lactate formation in 
this tissue is so slow, that Covin expressed some doubts as to the reliability 
of the measurements. One must conclude from the incomplete data, that 
there is some evidence for a minor inhibition of glycolysis by malonate, 
especially at the higher concentrations. 

The problem of the direct inhibition of glycolysis by malonate has been 
studied particularly well by Eva and George Fawaz at the American Univer- 
sity of Beirut. They had observed that 30 mM malonate almost completely 
blocks the cycle, leads to the accumulation of succinate, and yet does not 
depress the dog heart significantly (Fawaz et ol., 1958). However, 60 mM 
malonate causes rapid reduction in cardiac frequency and contractile 
failure, although no more succinate accumulates than with 30 mM. This de- 
pression of cardiac function must be related to an action other than on the 



EFFECTS OF MALONATE ON GLUCOSE METABOLISM 129 

cycle. The formation of lactate from glucose in extracts of rat muscle is 
inhibited over 90% by 60 mM malonate; simultaneously there is a decrease 
in inorganic phosphate, creatine-P, and readily hydrolyzable phosphate, 
accompanied by an increase in nonhydrolyzable phosphate (Fawaz and 
Fawaz, 1962). It was concluded that the acid-resistant phosphate fraction 
occurring in the presence of malonate must be made up of glycolytic inter- 
mediates, and it was then shown by analyses of the incubated extracts at 
various times that there is accumulation of 3-P-glycerate and glycerol- 1-P 
particularly, with smaller contributions from phosphoenolpyruvate and 
2-P-glycerate (see accompanying tabulation). Furthermore, the addition 



Incubation 


Accumulation of intermediates (mg ^ 


P/100 g 


tissue) 




time (min) 


P-enolpyruvate 


2-P-glycerate 


3-P-glycerate 


Glycerol- 1-P 


6 


6.12 


1.84 


19 


.00 


22 


.30 


30 


8.60 


2.45 


29 


.10 


30. 


,80 


120 


10.90 


3.18 


45 


.60 


33 


,40 



of these intermediates to a malonate-treated extract resulted in the ap- 
pearance of a major fraction as 3-P-glycerate, whereas in control incuba- 
tions they break down to inorganic phosphate and pyruvate or lactate. 
This might imply a block at the pyruvate kinase and some reversal of the 
glycolytic pathway which cannot proceed beyond 3-P-glycerate due to the 
lack of ATP. The addition of pyruvate to an inhibited extract leads mainly 
to the formation of lactate. Pyruvate kinase from rabbit muscle is inhibited 
86% by 60 mM malonate and this correlates quite well with the results 
seen in the extracts. Malonate at 30 mM only partly inhibits glycolysis, 
causes less accumulation of the phosphorylated intermediates, and inhibits 
pyruvate kinase 67%. The relative lack of affect of 30 mM malonate on 
cardiac function is probably due to a combination of two factors: the intra- 
cellular malonate concentration is undoubtedly less than 30 mM, and it 
is likely that glycolysis in the heart can be depressed to a certain degree 
before failure occurs, the myocardium having other sources of energy 
available.* Glycolysis in dog muscle extracts proceeds somewhat differently 
than in rat muscle extracts (e.g. accumulation of hexose phosphates occurs) 
and the response to malonate is consequently different (Fawaz et at., 
1963). High concentrations of malonate cause the accumulation of the same 

* Since the heart is generally considered to obtain much of its energy from fatty 
acid oxidation under certain circumstances, it would be important to know the effects 
of these high concentrations of malonate on this pathway. However, with the cycle 
inhibited, the generation of energy from the oxidation of fatty acids should be reduced. 



130 1. MALONATE 

intermediates in dog muscle as described above for rat muscle, but in ad- 
dition there is an inhibition of the accumulation of fructose-l,6-diP, so 
that some inhibition of a proximal step in glycolysis seems likely. Of the 
four enzymes involved previous to fructose-l,6-diP, only phosphogluco- 
mutase is sensitive to malonate, 40% inhibition being produced by 60 mM 
malonate and 76% inhibition by 120 mM malonate. Thus a secondary site 
for the inhibition of glycolysis is likely. Other mechanisms may be involved 
in intact muscle cells. 

Effects on the Distribution of C" from Labeled Glucose 

If glucose is metabolized exclusively by the Embden-Meyerhof glycolytic 
pathway and no initial decarboxylation of glucose or the hexose phosphates 
occurs (as it would if the pentose-P pathway were operative), malonate 
should depress the formation of C^'^Og from glucose-1-C^* and glucose-6-C^* 
equally. However, if the pentose-P pathway is important, malonate should 
decrease the formation of C^^Og from glucose-6-C^* relatively more than from 
glucose-1-C^*, and hence increase the C-l/C-6 ratio. The difference between 
the C^'^Oo formed from these precursors is often taken as a measure of the 
activity of the pentose-P pathway; this may not be strictly true because 
every hexosephosphate which is decarboxylated may not pass through the 
pentose-P pathway completely, but it is certainly the best evidence for 
the relative importance of these two pathways. 

Experiments of this type were performed with slices of various tissues 
from the rat (van Vals et al., 1956). The specific activities of the C^^Og 
formed from C-1 and C-6-labeled glucose are essentially the same in the 
controls, indicating the pentose-P pathway to be unimportant (see ac- 
companying tabulation). Malonate increases the C-l/C-6 ratios, which was 



C-l/C-6 ratio 
Tissue 



Control Malonate 30 mM 



Heart 


0.95 


2.38 


Brain 


0.93 


2.18 


Kidnev 


0.99 


1.92 


Diaphragm 


1.02 


1.32 



taken as evidence for a malonate-induced appearance of the pentose-P 
pathway, although it would account for only a small fraction of the glucose 
oxidized. In the lung and various mouse tumors, in which the pentose-P 
pathway is operative normally (C-l/C-6 rations between 1.4 and 5.6), 
malonate further augments the importance of the pentose-P pathway as 



EFFECTS OF MALOXATE ON GLUCOSE METABOLISM 131 

determined by C-l/C-6 ratios, which are increased to vakies between 4 
and 125. A similar situation has been encountered in sheep thyroid shoes 
(the results of three experiments are averaged in the accompanying tabu- 
lation) (Dumont, 1961). The formation of C^^Oofrom glucose-6-C^* is inhibit- 

Ci*0, Control Malonate 100 mM 



From glucose-1-C^^ (cpm) 


10.7 


9.72 


From glucose-6-Ci* (cpm) 


4.1 


0.47 


C-l/C-6 


2.58 


20.5 


(C-1) — (C-6) 


6.6 


9.25 



ed strongly, whereas that from ghicose-1-C^^ is scarcely affected. This, of 
course, indicates an almost complete block of the cycle, which is not sur- 
prising at this high malonate concentration, but it also suggests a greater 
participation of the pentose-P pathway in the presence of malonate. These 
results are perhaps more understandable in the light of the glycolytic 
inhibitions by high malonate concentrations discussed in the previous 
section. Quite different results were obtained in electrically stimulated rat 
ventricle strips (see accompanying tabulation) in which the pentose-P 

Ci^O, Control Malonate 5.6 mM 



From glucose- 1-C^* (cpm) 


0.318 


0.271 


From glucose-6-C''' (cpm) 


0.285 


0.264 


C-l/C-6 


1.12 


1.03 


(C-1) - (C-6) 


0.033 


0.007 



pathway is presumably not important, malonate at this concentration 
having little effect on glucose utilization (Rice and Berman, 1961). It is 
possible that higher concentrations of malonate would produce changes 
such as observed with other tissues. However, this concentration of mal- 
onate is quite effective in modifying ventricular function. 

An indirect mechanism for the acceleration of the pentose-P pathway by 
malonate may involve the levels of NADP and ATP in the tissues. The oxida- 
tive decarboxylation of glucose-6-P to initiate this pathway requires NADP, 
the concentration of which may be changed due to the action of malonate on 
the cycle. Also the phosphorylation of fructose-6-P in the Embden-Meyerhof 
pathway requires ATP, the level of which may be reduced by high concen- 
trations of malonate. However, little is known about the control of the 
pentose-P pathway and further experiments are needed to elucidate its 



132 1. MALONATE 

role during malonate inhibition. Nothing is known of the possible direct 
effects of malonate on the pentose-P pathway. The production of C^^Og 
from ribose-l-C^* is inhibited strongly by malonate in heart homogenates 
(Jolley et al., 1958) and it is believed that ribose is metabolized in this 
pathway, but the inhibition may reflect an action on the cycle. It is inter- 
esting that fluoroacetate does not inhibit C^^Og formation very potently, 
except at very high concentrations, so that some direct effect on ribose meta- 
bolism is possible. The results of the experiments discussed above would 
argue against this. D-Xylose and D-ribose-5-P are oxidized through se- 
doheptulose-7-P in extracts of Pseudomonas hydrophila and this is not af- 
fected by malonate at 20 mikf (Stone and Hochster, 1956). 

The only investigation of the effects of malonate on the general distri- 
bution of C^^from labeled glucose is that of Romberger and Norton (1961) in 
potato tuber slices incubated with uniformly labeled glucose for 3 hr (Table 
1-19). The situation in this tissue is complex inasmuch as fresh slices do not 
metabolize glucose appreciably, whereas 36-hr-old slices oxidize it quite 
rapidly. In the aged tissue, CO2 production is inhibited 92% by 50 mM mal- 
onate at pH 5, while the formation of CO2 in fresh tissue is stimulated 28%. 
In the fresh tissue, they suggest that CO2 is formed mainly in the pentose-P 
pathway and little through the Embden-Meyerhof sequence, but glycolysis 
contributes more and more with time, so that the marked inhibition by 
malonate in aged tissue is not surprising. The synthesis of sucrose accounts 
for over half of the labeling from glucose-u-C^* and this is inhibited only 
11% by malonate. After 3-hr incubation, however, one can deduce little 
about the initial attack on glucose. It may be mentioned in this connection 
that in Acetobacter xylmum, where the sole product of glucose assimilation 
is cellulose, malonate at 10 mM does not inhibit the formation of cellulose. 
Laties (1964) has investigated this problem in detail and found that dif- 
ferent methods of aging result in metabolically different potato slices, 
in that some exhibit a malonate-sensitive and some a malonate-resistant 
respiration. In the malonate-sensitive slices the formation of C^^Og from 
labeled glucose is almost completely abolished by malonate, whereas in the 
malonate-resistant slices there is little effect by malonate on the production 
of C^^Og. There is a similar correlation with respect to the effects of malonate 
on glucose uptake. Since dinitrophenol does not interfere with glucose 
uptake, one can eliminate depression of ATP formation by malonate as 
responsible for the inhibition of the uptake in sensitive slices. Laties con- 
sidered the possibility that increased citrate levels might inhibit phos- 
phofructokinase, but the mechanism is not yet well understood. 

Effects of Tissue Age on the Response of Glucose Metabolism to Malonate 

We have seen above that aging of potato slices increases their sensitivity 
to malonate with respect to glucose oxidation. It might be expected that 



EFFECTS OF MALONATE ON GLUCOSE METABOLISM 133 

the metabolic characteristics of tissues, would change with age and that 
this would be reflected in difl"erent susceptiblies to inhibitors. The altered 
response to inhibitors could provide some information on the nature of 
metabolic aging. The results obtained on rat brain are, however, discordant. 
Tyler (1942), using a minced preparation, found that malonate inhibition 
of the respiration in the presence of glucose increases up to a rat age of 10 
days, after which it remains at the adult level (see accompanying tabula- 
tion). Although the control respiration rises, the malonate-resistant fraction 



Rat age (days) 


Control 


Malonate 10 


mi¥ 


"o Inhibition 


1 


656 


547 




16.9 


2 


582 


507 




12.9 


6 


814 


590 




27.6 


8 


712 


559 




34.3 


10 


1080 


530 




50.9 


16 


1500 


810 




46.0 


31 


1900 


960 




49.5 


Adult 


1723 


864 




50.0 



of the respiration remains constant up to 10 days, i.e., the increased respi- 
ration is all due to the development of activation of a system sensitive to 
malonate, presumably the cycle. On the other hand, Muir et al. (1959) 
reported that the glucose respiration of adult rat brain slices is less sensitive 
to 10 inM malonate (30% inhibition) than the respiration of tissue from 
young animals of 1-3 days (70%). In this case, the young brain respires 
almost 2.5 times as rapidly as adult brain. The differences in these obser- 
vations may be related to the preparations used (mince or slice). There are 
several reasons why malonate sensitivity would change with age, for exam- 
ple, an alteration of cycle activity, the development or loss of pathways 
other than the cycle for the metabolism of acetyl-CoA, or a change in the 
ability to demonstrate a Pasteur effect. The effect of age should also be 
studied on the electrically stimulated or K+-stimulated respiration of the 
brain, since it is more sensitive to malonate and the results might have 
more physiological pertinence. 

The respiratory rates and patterns of fungus spores are altered during 
the initiation of germination and the subsequent development of the germ 
tube (Gottlieb, 1964). During the incubation of the spores of Penicillium 
oxalicum and Ustilago maydis and the progress of germination, the respira- 
tion in the presence of glucose rises markedly and this is accompanied by an 
increasing sensitivity to malonate (see accompanying tabulation) (Caltrider 
and Gottlieb, 1963). It is somewhat difficult to determine if this implies 



134 



1. MALONATE 



an increase in cycle activity since the concentration of malonate was 100 
mM and more than the cycle might be inhibited. 



Incubation (hr) 


Respiration 


0/ 

/o 


Inhibition 





1.3 







6 


6.0 




16 


9 


11.1 




22 


12 


23.5 




32 



Effects on Resting and Stimulated Glucose Metabolism in Brain 

The effects of malonate on a tissue may depend on the activity of the 
tissue as well as the age. Stimulation of a tissue such as brain alters the 
metabolic pattern and this is exhibited in altered responses to inhibitors. 
The results obtained by Heald (1953) on guinea pig cerebral cortex slices 
are shown in Fig. 1-14. The resting respiration and aerobic glycolysis are 




( Malonote ! 



10 100 
mM 



Fig. 1-14. Effects of malonate on guinea pig brain 
slices with glucose as the substrate. Electrical stimula- 
tion through grids. The respiration and aerobic glycol- 
ysis in //moles/g wet weight/hr. (From Heald, 1953.) 



EFFECTS OF MALONATE ON LIPID METABOLISM 135 

little affected by malonate up to 10 mM, whereas the electrically stimulated 
respiration is readily depressed (down to the resting level at 10 mM mal- 
onate) and the stimulated lactate formation markedly increased. This means 
that the glucose metabolism appearing upon stimulation is quite sensitive 
to malonate and perhaps involves a greater participation of the cycle. 
These results were confirmed by Kimura and Niwa (1953) in guinea pig 
brain stimulated by K+, and a stimulation of lactate formation by 10 mM 
malonate was observed by Tsukada and Takagaki (1955). An abolition of 
the inhibition of respiration upon addition of fumarate occurs (Takagaki et 
at., 1958). Rat brain slices behave similarly, the resting respiration in the 
presence of glucose being unaffected by malonate up to 0.8 mM, while the 
stimulated respiration is readily suppressed (Wallgren, 1960). A malonate 
concentration as low as 0.2 mM inhibits the stimulated respiration 15%. 
Pyruvate utilization and the associated oxygen uptake are also inhibited 
more strongly in stimulated slices than in resting slices (Takagaki et al., 
1958). The C^^Og from labeled pyruvate is formed about twice as rapidly in 
high K+ medium compared to the controls (Kini and Quastel, 1959), and 
this is inhibited more strongly by malonate in the K+-stimulated slices, 
while the stimulated respiration is depressed to the endogenous level. 

These results taken together clearly indicate a dependency of malonate 
inhibition on the metabolic activity of brain tissue, whether altered by 
electrical stimulation or K+. A Pasteur effect is observed and it is possible 
that the inhibition by malonate would have been greater if it had not 
induced a greater utilization of glucose. The data do not necessarily imply 
a specific activation of the cycle; a greater uptake or utilization of glucose 
would impose a greater load on the cycle, and this might be inhibited more 
readily. Whatever the explanation for these effects, such results have 
important bearing on the actions of malonate on intact and functioning 
nervous tissue. 



EFFECTS OF MALONATE ON LIPID METABOLISM 

The major pathway for fatty acid oxidation is a helical degradation into 
acetyl-CoA, which normally enters the cycle by condensation with oxal- 
acetate. Each turn of the helix, releasing one acetyl-CoA, takes up 2 atoms 
of oxygen, and the complete oxidation of acetyl-CoA through the cycle 
takes up 4 more atoms of oxygen. Thus, approximately two-thirds of the 
oxygen uptake due to fatty acid oxidation occurs in the cycle,* and one 
would expect malonate to depress this fraction in proportion to the cycle 

* The term "cycle." as before, will indicate the tricarboxylate cycle only, and the 
pathway of degradation of fatty acids to acetyl-CoA and other terminal products 
will be designated the "helix" for convenience. 



136 1. MALONATE 

block it induces. The situation is vey similar to that of ghicose oxidation in 
this respect. Generally speaking, malonate could act on either the cycle, 
or the helix, or both. Despite the extensive work that has been done on the 
effects of malonate on fatty acid oxidation, direct information on the actions 
on the helix is lacking, since the five reactions involved in each turn of the 
helix and the enzymes associated with these have not all been tested for 
susceptibility to malonate, nor has the operation of the helix dissociated 
from the cycle been studied. Our evidence on this point must be indirect. 

Before considering this evidence, let us outline some of the possibilities 
for mechanisms of helix inhibition. Just as in the oxidation of glucose, ATP 
is required for the initiation of the helix reactions and Mg++ is a necessary 
cofactor (e.g., for the fatty acid thiokinase), so that malonate might depress 
the operation of the helix by depleting the system of either of these sub- 
stances. The extent of such an inhibition will depend on the availablity of 
ATP or the presence of systems generating it, and on the concentration of 
Mg++. Possibly a more important factor is the requirement for coenzyme A. 
Malonate could deplete conzyme A by at least two mechanisms. The for- 
mation of a relatively stable malonyl-CoA would remove some of the 
coenzyme A from participating in the helix. A block of the cycle would 
impede the entrance of acetyl-CoA into the cycle and the regeneration of 
coenzyme A will depend on the enzymes present for the metabolism of 
acetyl-CoA. The usual pathways for acetyl-CoA are (1) a simple splitting 
to form acetate, (2) a transfer of the coenzyme A to another acid, and (3) 
a condensation of two acetyl-CoA's to form acetoacetyl-CoA and eventually 
acetoacetate. As in the oxidation of pyruvate through the cycle, the fate 
of acetyl-CoA will depend also on the presence of reactions forming oxal- 
acetate by pathways other than the cycle. The rate of fatty acid oxidation 
can thus be limited by the rate of regeneration of coenzyme A. These 
considerations lead one to ])redict that the effects of malonate on fatty 
acid oxidation would be variable and dependant on the metabolic charac- 
teristics of the tissue studied and the conditions of the experiment. This 
prediction is borne out. 

There is a good deal of evidence that malonate in concentrations up to 
20 raM does not directly inhibit the reactions of the helix. Although an 
inhibition of the oxygen uptake or the CO2 production during fatty acid oxi- 
dation is not indicative of the site of inhibition when the helix and the 
cycle are operating together, the absence of inhibition implies a lack of 
action on the helix. Malonate at 16.8 mM has no effect on the C^^Oa arising 
from palmitate-1-C^* in soluble extracts of peanut cotyledons (Castelfranco 
et al., 1955), nor does 1 roM malonate have an effect on the oxygen uptake 
associated with palmitate oxidation in peanut microsomes (Humphreys 
et al., 1954). The anaerobic dehydrogenation of C4-C18 fatty acids in liver 
homogenates with methylene blue as an acceptor is not inhibited by 



EFFECTS OF MALONATE ON LIPID METABOLISM 137 

10 raM malonate (Blakley, 1952). The rate of oxidation of decanoate by 
Serratia marcescens is also not affected by 10 mM malonate, although 
20 m.M inhibits somewhat (Waltman and Rittenberg, 1954). Geyer and 
Cunningham (1950) stated that their data indicated no inhibition directly of 
octanoate oxidation in liver by 5 mM malonate (this work will be discussed 
in greater detail later). 

On the other hand, Lehninger and Kennedy (1948) reported that 10 mil/ 
malonate stronglj' inhibits octanoate oxidation in particulate suspensions, 
from rat liver. Not only is the respiration from the oxidation inhibited 
but the utilization of octanoate is almost completely suppressed. The ad- 
dition of malate or oxalacetate reduces the inhibition only partially, the 
utilization of octanoate still being inhibited around 70%. It may be noted 
that the total Mg+"'" concentration in these experiments was 0.25 mM, 
which is quite low, so that malonate could have inhibited by depletion of 
this cofactor. An interesting point is that the strain of rats used is very 
important, since 10 mM malonate inhibits octanoate oxidation 25% in 
preparations from livers of Sprague-Dawley rats, but in preparations made 
from a heterogeneous stock colony 2 mM malonate inhibits 50-75%. 
Such differences in strain behavior may explain some of the discrepancies in 
the reports on malonate inhibition. Weinhouse et al. (1949) found that 
20 mM malonate almost completely blocks the oxidation of octanoate in rat 
liver slices and that fumarate only partially overcomes this, suggesting to 
them that malonate must have some action other than on succinate oxidase. 
In several instances malonate has been found to inhibit the oxygen uptake 
from fatty acid oxidation very potently. Butyrate respiration in peanut 
mitochondria is inhibited 75% by 6 mM malonate and the formation of 
C^^Og from butyrate-l-C^* is depressed even more strongly (Stumpf and 
Barber, 1956). Malonate at 10 mM inhibits the oxidation of octanoate by 
carp liver mitochondria 80% (Brown and Tappel, 1959). In suspensions of 
particulates from desert locust thorax, butyrate oxidation is inhibited 70% 
by malonate at concentrations as low as 1 mM and maximally 85% (Meyer 
et al., 1960). The oxidation of octanoate- 1-C^* and myristate-1-C^* by sub- 
cellular particles from the lateral line of the rainbow trout, as measured 
by the C^^Oj released, is reduced 95% by 10 mM malonate (Bilinski and 
Jonas, 1964). One of the most sensitive systems is found in the oxidation of 
linolenate in liver mitochondria of vitamin E-deficient chicks, 0.25 mM 
malonate inhibiting 40% (Kimura and Kummerow, 1963). These examples 
must be interpreted as indicating either a direct or an indirect inhibition 
of the helix by malonate. Finally, it was stated by Mudge (1951), on the basis 
of unpublished experiments, that malonate inhibits fatty acid oxidation 
more strongly than succinate dehydrogenase in kidney particulate prepa- 
rations. The possibility of effects on the helix must therefore be entertained 
on the basis of our present knowledge. 



138 



1. MALONATE 



Effects on the Formation of Acetoacetate and Other Ketones 

It was known before 1912 that acetate, oxalate, and maleate can either 
be metabolized to acetoacetate or so alter metabolism that acetoacetate ac- 
cumulates, and for this reason Momose (1914) in Berlin studied the effects 
of malonate perfused through starved dog livers at a concentration of 
approximately 13 mM. He found that acetone appears and detected a sub- 
stance which he only later, after returning to Japan (Momose, 1925), proved 
was acetoacetate. However, he postulated that malonate -^ acetate — >■ 
acetoacetate -^ acetone, which was not unreasonable considering the inhi- 
bitory action of malonate was unknown. The appearance of acetone in the 
urine of rabbits fed malonate or rats injected subcutaneously with malonate 
was observed by Huszak (1935), and simultaneously Annau (1935) dem- 
onstrated that malonate causes the formation of acetone in slices and 
breis of rabbit kidney. Acetoacetate has been shown to accumulate in 
tissues as a response to malonate (see accompanying tabulation). Since 
acetoacetate and acetone are the most important ketonic substances ap- 
pearing in the tissues, these results clearlj^ show that malonate is ketogenic. 



Preparation 


Substrate 


Malonate 


Reference 


Whole rabbits (blood) 


Endogenous 


1.6g/kg 


Handler (1945) 


Whole rats (blood) 


Endogenous 


O.S 


!g/kg 


Mookerjea and Sadhu (1955) 


Rat liver slices 


Acetate 


40 


mM 


Jowett and Quastel (1935 c) 




Butyrate 


40 


mM 


Jowett and Quastel (1935 c) 


Guinea pig liver slices 


Propionate 


40 


mM 


Jowett and Quastel (1935 c) 




Butyrate 


40 


mM 


Jowett and Quastel (1935 c) 


Rat liver slices 


Endogenous 


10 


mM 


Edson (1936) 


Rat liver slices 


Fatty acids 


5 


mM 


Geyer and Cunningham (1950) 


Rat liver suspension 


Fatty acids 


10 


mM 


Lehninger (1946 a) 


Rat liver homogenates 


Pyruvate 


4 


mM 


Recknagel and Potter (1951) 


Rabbit liver mitochondria 


Fatty acids 


15 


mM 


Cheldelin and Beinert (1952) 


Jensen rat sarcoma mince 


Glucose 


6.7 mM 


Boyland and Boyland (1936) 


Peanut mitochondria 


Butyrate 


6 


mM 


Stumpf and Barber (1956) 



Acetoacetate is an important substance in intermediary metabolism and 
the pathways for its formation and utilization are often complex. The con- 
centration of acetoacetate will depend on the relative rates of its formation 
and utilization. The accumulation of acetoacetate in the presence of mal- 
onate could result from either an acceleration of its formation or an inhibi- 
tion of its utilization, or both. The earliest concept that malonate itself 
gives rise to the acetoacetate was soon abandoned, and several investigators 
assumed that malonate interferes with the disposal of acetoacetate, while 



EFFECTS OF MALONATE ON LIPID METABOLISM 



139 



recently more emphasis has been placed on the diversion of carbohydrate 
and fatty acid metabolism to acetoacetate by malonate. We may summarize 
some of the major pathways of acetoacetate before discussing the mecha- 
nisms for the action of malonate (Ac-CoA = acetyl-CoA, and AcAc-CoA = 
acetoacetyl-CoA). Reaction (1) for the formation of Ac Ac-Co A from aceto- 



Fatty acids ( 
Pyruvate ij 



Butyrate 



-»- Ac -Co A 



AcAc — CoA 



-»- Ac Ac — CoA 



Even-numbered 
fatty acids 

•-Butyryl —CoA *- Ac Ac — CoA 

Succinate + Ac Ac— CoA 

/3-Hydroxybutyrate 

Amino acids (phenylalanine, 
tyrosine, leucine, and 
isoleucine) 



Acetone + COj 
.AcAc-CoA (1) 

I acetoacetate I ' *-AcAc— CoA (2) 

Sterols 
/3-Hydroxybutyrate 



acetate is catalyzed by an activating enzyme in the presence of CoA and 
ATP, while reaction (2) is catalyzed by a CoA transferase in the presence 
of succinyl-CoA. All of these reactions do not occur in a single tissue and 
the response to malonate depends in part on which reactions are possible 
in any case. 

A block of the cycle restricts the entrance of acetyl-CoA, derived from 
pyruvate and fatty acids, into the cycle, unless there is an adequate synthe- 
sis of oxalacetate from a noncycle source, which is seldom the case. If the 
acetyl-CoA accumulates, coenzyme A soon becomes tied up and the oxida- 
tion of pyruvate and fatty acids would cease. However, in most tissues 2 
molecules of acetyl-CoA condense to form acetoacetate and coenzyme A is 
regenerated; in other situations, hydrolysis to acetate may occur. Malonate 
may thus divert acetyl-CoA from the cycle to acetoacetate. Quantitative 
conversion to acetoacetate has been observed (Recknagel and Potter, 1951). 
Another reaction possibly favoring acetoacetate formation during malonate 
inhibition results from the accumulation of succinate, which can now react 
more readily with acetoacetyl-CoA in a transfer of coenzyme A. The effec- 
tiveness of such a mechanism depends on the continued formation of suc- 
cinate and, hence, usually on a noncycle source of oxalacetate. Acetoacetyl- 
CoA is also formed as the terminal product of the helical oxidation of even- 
numbered fatty acids. These relationships are illustrated in Fig. 1-15 
where a block of succinate oxidation induces accumulation of acetoacetate 
by accelerating its formation through two mechanisms. The other pathways 
for the formation of acetoacetate are probably less important in most tis- 
sues and would not be accelerated by malonate. Accumulation of aceto- 
acetate implies that its utilization must not be too rapid. Liver is notable in 
this respect because it lacks enzymes to metabolize acetoacetate rapidly, 



140 1. MALONATE 

especially the activating system for the formation of acetoacetyl-CoA, and 
possesses an active deacylase to split acetoacetyl-CoA. Therefore, acetoace- 
tate accumulation is most readily observed in liver and most investi- 
gations have been on this tissue. The urinary acetoacetate found after the 
administration of malonate is probably derived mainly from liver. In heart, 
on the other hand, the enzyme balance is such as to favor the rapid me- 
tabolism of acetoacetate and it does not accumulate. Acetate rather than 
acetoacetate accumulates in some cells, for example in heart mitochondrial 
suspensions metabolizing pyruvate in the presence of 8.8 mM malonate 
(Fuld and Paul, 1952). 



(Malonate) 
a -Ketoglutorate ^ Succmyl- Co A Succinate -X"^ Fumorcte — ^ —^Citrate 




Fatty acids Pyruvate 



Fig. l-l.'j. Diagram of some pathways involved in the effects 
of malonate on the metabolism of acetoacetate. 

One would predict that fumarate should counteract the ketogenic activi- 
ty of malonate because, by supplying oxalacetate, acetyl-CoA will again be 
able to enter the cycle. However, it may be noted that fumarate may lead 
to an even greater accumulation of succinate and if the formation of aceto- 
acetate by the transfer of coenzyme A from acetoacetyl-CoA to succinate is 
important, fumarate will only augment the malonate effect. Administration 
of fumarate with malonate to rabbits abolishes the appearance of acetone 
that arises with malonate alone (Huszak, 1935). Addition of fumarate to 
malonate-inhibited minces of rat sarcoma likewise prevents the accumula- 
tion of acetone bodies (Boyland and Boyland, 1936). However, fumarate 
has very little effect on the appearance of acetoacetate in rat liver slices 
inhibited by malonate (Edson, 1936), and this might indicate a mechanism 
for malonate action other than the inhibition of succinate oxidation, or the 
importance of the coenzyme A transfer reaction. 

It is now easy to see how malonate can reduce the oxygen uptake and 
the CO2 production from fatty acid oxidation without necessarily decreasing 
the utilization of the fatty acids. A fraction that would normally be com- 
pletely oxidized is diverted into acetoacetate (or acetate, acetone, and other 
products). One of the best indications that malonate does not inhibit the 
helix directly is the fact that the C^Oo appearing in the end products from 
labeled fatty acid is not reduced by malonate. To illustrate this it will be con- 
venient to turn to the excellent studies of Geyer and his group at Harvard. 



EFFECTS OF MALONATE ON LIPID METABOLISM 141 

The basic procedure in these investigations was to incubate carboxyl- 
labeled fatty acids with rat liver and kidney slices, and determine the dis- 
tribution of C^* in acetoacetate and COg. Malonate at 5 mM depresses the 
formation of C^^Og from octanoate-C^* 00^ around 60% and fumarate is 
able to overcome this inhibition only partially (Geyer et al., 1950 a). Fu- 
marate and other cycle intermediates increase the total COg formed but 
have little effect on the C^^Oa- This was explained by the accumulation of 
some of the C^* as succinate, this not being relieved by fumarate, and we 
have previously cited this as an example of the importance of considering 
what is measured in demonstrating a reversal by fumarate. 

Where does the C^* go that does not appear as C^^Oo in inhibited slices? 
They found that in the presence of malonate much of the C^* appears in 
acetoacetate (Table 1-24) (Geyer and Cunningham, 1950). The ratio AcAc- 
QujQuQ^ is near 1.21 in the controls and is increased to around 4.51 by 
malonate, averaging the results from the five fatty acids used. It may 
also be noted that malonate generally increases the total C" recovered, even 
though succinate was not determined, showing that malonate does not in- 
hibit the fatty acid oxidation directly. Later they determined both the car- 
boxyl and carbonyl C^* in acetoacetate and the more complete results are 
summarized in Table 1-25, where I have taken the liberty of averaging 
the data for all the fatty acids used, inasmuch as the effects are always 
in the same direction although differences between the different fatty acids 
are evident. These results show clearly the diversion of fatty acid metabo- 
lism into acetoacetate by malonate. Weinhouse et al. (1949) reported that in 
rat liver slices 10 mM malonate inhibits COo formation and no acetoacetate 
appears, which was so contradictory to the results obtained by Geyer that 
the latter studied malonate in concentrations up to 20 mM, but found only 
that even more acetoacetate accumulates. Also they tested three different 
strains of rat and the results were the same. The reason for this discrepancy 
could not be explained. 

The differential labeling in the carboxyl and carbonyl groups of aceto- 
acetate is difficult to explain. If acetoacetate arises by a condensation of 
acetyl-CoA units, the labeling in these positions should be uniform. However, 
the ratio is seldom unity as may be seen in the results summarized by Chai- 
koff and Brown (1954). In the work of Geyer with rat liver slices, the ratio 
CHgC^^ — / — CHoC^* 00" is less than 1 in the controls and increases with 
the length of the fatty acid chain. Malonate increases this ratio, that is, it 
increases relatively the labeling in the carbonyl group. Chaikoff and Brown 
have given a detailed analysis of the possible factors determining this ratio, 
and the explanation is based on the existence of two types of 2-carbon 
fragment formed from fatty acids, one designated the CH3CO — fragment 
and the other the — CH2CO — fragment. These fragments are assumed 
to arise from different portions of the fatty acid chain and only the — CHg 



142 



1. MALONATE 



Q 


T 




o 






>H 


<; 




H- 


< 




fo 


O 


Q 




W 


6 


iJ 




w 




K 




< 




J 




ij 




>H 




X 




o 




m 




P3 


•rf 


< 


6 
1 


O 


o 

<1 


53 


o 



o 



o 



o 



Q 


(B 


-P 


-kJ 


tS 


cS 


Cj 


O 


_« 


^ 


> 




-2 


a) 


o 

c 


+3 

O 

c 



o 



fa 






EFFECTS OF MALONATE ON LIPID METABOLISM 



143 



Table 1-25 

Effects of 5 mM Malonate on the Oxidation of Fatty Acids in Rat Liver 

Slices " 





Control 


Malonate 


Change 


C^Og produced 


10,694 


4,536 


- 6,158 


AcAc-carbonyl-C* formed 


3,572 


8,488 


+ 4,916 


AcAc-carboxyl-Ci^ formed 


7,268 


13,438 


+ 6,170 


CHaCi^O— /— CH^Ci^OO- 


0.47 


0.63 




Total AcAc-Ci* formed 


10,840 


21,926 


+ 11,086 


C'^Oa + AcAc-C" formed 


21,534 


26,462 


+ 4,928 


AcAc-CiVC'^Oii 


0.83 


4.55 





" Slices incubated 1 hr with carboxyl-labeled fatty acids shown in Table 1-24 at 
38° and pH 7.1. The data from the five fatty acids were averaged to indicate the 
general effects of malonate. (From Geyer et al., 1950 b.) 



CO— fragments are believed to enter the cycle. The CHgC^* o_/— CHaC^^ 
00" ratio in acetoacetate will depend on the rates of production and utili- 
zation of these two fragments. As pointed out, these two types of 2-carbon 
fragment may be only convenient designations for two reactive forms of 
acetyl-CoA. Malonate is assumed to increase the formation of acetoacetate 
by condensation of randomized fragments of the — CHaC^^O — type, so 
that the ratio rises. The extra acetoacetate formed over the control when 
malonate is present does indeed exhibit a ratio of unity for hexanoate and 
octanoate oxidation (Geyer et al., 1950 b). If malonate does this by blocking 
the cycle, a preferential accumulation of — CH2CO — fragments would occur, 
and a greater proportion of the acetoacetate would be formed from them. 
The effect of malonate on acetoacetate accumulation will depend on the 
pathway of fatty acid oxidation in the uninhibited tissue and, hence, on the 
experimental conditions. For example, Witter et al. (1950) found that 3 mM 
malonate inhibits acetoacetate formation from hexanoate 4% and 10 mM 
malonate inhibits 9% in suspensions of washed particles from rat liver. 
However, hexanoate is quantitatively converted to acetoacetate in the con- 
trols, presumably because no cycle intermediates, are present to form oxal- 
acetate for condensation of the acetyl-CoA units. Under such circumstances 
malonate would not be expected to increase acetoacetate and the small 
inhibitions must be attributed to actions directly on the helix. The rela- 
tionship between acetoacetate formation in malonate-inhibited systems 
and the presence of cycle intermediates was illustrated and discussed by 
Cheldelin and Beinert (1952). 



144 1. MALONATE 

We have seen that the accumulation of acetoacetate in the presence of 
malonate can be attributed to an increased rate of formation of the aceto- 
acetate. Is the accumulation due entirely to this or can malonate also inhibit 
the utilization of acetoacetate in some tissues? The rise in the acetoacetate 
in liver slices in the presence of malonate was believed by Jowett and 
Quastel (1935 c) to be due to the inhibition of the decomposition of aceto- 
acetate, since at that time the pathways for the formation of acetoacetate 
were not understood. However, Quastel and Wheatley (1935) soon provided 
evidence that malonate can interfere with the disappearance of acetoacetate 
in rat liver and kidney slices. In kidney slices, malonate at 8 mM inhibits 
around 42% and at 16 mM 64%, and in liver slices an inhibition of 74% 
was observed with 40 mM malonate. Fumarate is able to counteract this 
inhibition partially and it was concluded that acetoacetate oxidation must 
be coupled with other oxidations inhibited by malonate. Very similar results 
were reported by Edson and Leloir (1936); indeed, 20 mM malonate inhibits 
disappearance of acetoacetate in rat kidney slices 93% and it was stated, 
"Malonate is a powerful and relatively specific inhibitor of respiration and 
of aerobic disappearance of acetoacetic acid in kidney." Both Handler 
(1945) and Mookerjea and Sadhu (1955) in their work with whole animals, 
favored the concept that malonate interfered with acetoacetate metabolism 
accounting for the rises in blood acetoacetate. Inasmuch as several different 
pathways are open to acetoacetate and these vary with the tissue used, it is 
difficult to interpret accurately these results. In some tissues, acetoacetate 
can be split into acetyl-CoA fragments that enter the cycle and here malo- 
nate might inhibit by blocking the cycle and the formation of oxalacetate, 
which is, of course, esentially the same mechanism adduced to explain the 
increased formation of acetoacetate. There is evidence that malonate does 
not inhibit the reduction of acetoacetate to /?-hydroxybutyrate (Edson and 
Leloir, 1936), nor does it seem to interfere with the formation of sterols from 
acetoacetate (Mookerjea and Sadhu, 1955). It is probably best in the 
present state of our knowledge to attribute the accumulation of acetoacetate 
in the presence of malonate primarily to a diversion of 2-carbon units away 
from oxidation through the cycle, without eliminating the possibility that 
malonate may interfere in other pathways for the utilization of acetoacetate. 

Effects on Propionate Metabolism 

Propionate arises terminally from the /^-oxidation of odd-numbered fatty 
acids and in certain tissues, such as the liver, can be oxidized completely 
through the cycle. However, the oxidation of propionate differs from that 
of other fatty acids. The following sequence of reactions has been suggested: 

CO, 

+ 

. ATP ATP B,j 

Propionate — > propionyl-CoA — > methylmalonyl-CoA — > succinyl-CoA 



EFFECTS OF MALONATE ON LIPID METABOLISM 145 

The over-all reaction is the carboxylation of propionate to succinate. Other 
pathways occur in bacteria, e.g. 

Propionate -> propionyl-CoA -> acrylyl-CoA -> lactyl-CoA ->■ pyruvate -> COg + H2O 

Such a sequence may also operate in animal tissues, since lactate was iden- 
tified chromatographically after incubation of mouse liver slices with pro- 
pionate (Daus et al., 1952). If the principal pathway of propionate is via 
succinate, malonate would be expected to inhibit its oxidation readily but, 
if acetyl-CoA is formed, the inhibition will vary with the conditions as 
discussed for the effects of malonate on pyruvate oxidation. 

Malonate has been shown to inhibit strongly the C^^Oo formation from 
labeled propionate in mouse liver slices (Daus et al., 1952), rat liver slices 
(Katz and Chaikoflf, 1955), suspensions of rabbit liver particles (Wolfe, 
1955), and peanut mitochondria (Giovanelli and Stumpf, 1958), as anticipat- 
ed. In the rabbit liver particulate preparation, 10 mM malonate suppresses 
the formation of C^^O, from both propionate-l-C^* and propionate-2-C^^ 
almost completely, and at the same time leads to the accumulation of suc- 
cinate, and in rat liver slices malonate also causes succinate accumulation. 
From these data alone it is impossible to say whether the succinate arises 
directly from propionate or is formed via the cycle, but the marked inhibi- 
tion of C^^O, formation would indicate the former. This is substantiated 
by the demonstration of labeled methylmalonate in the rat liver slices. 
Another possible site for malonate inhibition is suggested by the work of 
Flavin et al. (1955) on rat tissues. The intercon version of methylmalonate 
and succinate was shown to be inhibited completely by 5 mM malonate and 
thus malonate leads to the accumulation of methylmalonate during propion- 
ate metabolism. However, it is not known if malonate can inhibit the 
methylmalonyl-CoA isomerase, which catah^zes the interconversion in the 
normal pathway, or if malonate only inhibits the formation of methyl- 
malonyl-CoA from methylmalonate. The latter is reasonable because mal- 
onate could compete with methylmalonate for the active site on the enzyme. 

In peanut mitochondria the situation may well be different. Malonate at 
6 mM inhibits the formation of C^^Oa from propionate- 1-C^* 41% (Giovanelli 
and Stumpf, 1958). It was felt that this inhibition is not as much as would 
be expected if the pathway from propionate leads to succinate. Further- 
more, fluoride, which inhibits the carboxylation of propionyl-CoA, does 
not depress the C^^Oa significantly. The pathway through methylmalonyl- 
CoA to succinate may not be operative here, and the following pathway was 
proposed: 

Propionate ->• propionyl-CoA -> acrylyl-CoA -> ^-hydroxypropionyl-CoA ->• 
/?-hydroxypropionate -> malonic semialdehyde -> malonyl-CoA -> acetyl-CoA 

The COo formed in the last step derives from the carboxyl group of propion- 



146 1. MALONATE 

ate so that an inhibition of C^^02 formation from proprionate-l-C^* would 
imply an action of malonate somewhere along this pathway. It is possible 
that malonate, or malonyl-CoA formed from it, could compete with the 
malonyl-CoA from propionate and in this way reduces the formation of 
0^*02- It is known that the addition of methylmalonate simultaneously 
with propionate depresses the propionate utilization strongly (Feller and 
Feist, 1957). Lactating rat mammary gland preparations convert propion- 
ate to fatty acids in part, the principal pathway being direct condensation 
with malonyl-CoA to form the odd-chain fatty acids. Malonate at 10 vnM 
inhibits this incorporation 50-65%, whatever the position of C^^ in pro- 
pionate, and simultaneously C^^Og is depressed around 30% from proprion- 
ate-l-C^* and propionate-2-C^^, and nearly 50% from priopionate-3-C^^ 
(Cady et al., 1963). This was interpreted not as a direct action on the pro- 
pionate pathway but as a reduction of ATP or NAD(P)H, these being nec- 
essary for fatty acid synthesis, as a consequence of inhibition of the cycle. 
In Rhodospirillum rubrum, both the oxidation (Clayton et al., 1957) 
and the photosynthetic dissimilation (Clayton, 1957) of propionate are 
inhibited by malonate to approximately the same extent as are the similar 
reactions of succinate, and this was given as evidence to support the meta- 
bolism of propionate to succinate in these organisms. 

Effects on Synthesis of Fatty Acids 

There are at least three systems for the synthesis of fatty acids; one is the 
reversal of the /^-oxidation in the helix and the other two involve the for- 
mation of malonyl-CoA from acetyl-CoA, one mitochondrial and the other 
nonmitochondrial (Green and Wakil, 1960). There are obvious relationships 
between fatty acid synthesis and oxidative metabolism of various sub- 
strates. The controls that establish the rates of fatty acid synthesis, or the 
balance between oxidation and synthesis, have not been elucidated and it is 
difficult to determine in a particular case what the effect of a cycle block 
would probably be. The level of acetyl-CoA and the availability of the var- 
ious pathways for its metabolism must be an important factor, but the 
concentrations of coenzyme A, ATP, NADH and NADPH could also play 
a significant role. 

Malonate has been found to produce a variety of effects. Most of the 
studies have involved the incubation of tissue preparations with acetate- 
1-C^* and the subsequent determination of labeled fatty acids formed from 
the acetate. In some cases a marked stimulation of fatty acid synthesis 
in the presence of malonate has been observed. Malonate at 50 milf inhib- 
its the O2 uptake of nonparticulate extracts of rat mammary gland and 
yet increases the formation of fatty acids, sometimes as much as 10-fold 
(Popjak and Tietz, 1955), The addition of oxalacetate or a-ketoglutarate 
with the malonate increases the synthesis aven more (see accompanying 



EFFECTS OF MALONATE ON LIPID METABOLISM 147 

tabulation). The stimulating action of a-ketoglutarate was attributed to 
the generation of NADH by which hydrogen atoms are provided for fatty 
acid synthesis. Dils and Popjak (1962) claimed that malonyl-CoA is not 



Incorporation of acetate into 
fatty acids (m/< moles/ 100 mg) 



Control 18.2 

Malonate 118 

Oxalacetate 72 . 7 

Malonate + oxalacetate 241 

a-Ketoglutarate 51.7 

Malonate + a-ketoglutarate 517 

formed from malonate in these extracts and that the stimulation of fatty 
acid synthesis must be an indirect effect, possibly the suppression of the 
deacylation of malonyl-CoA formed from acetyl-CoA, or the inliibition of 
the decarboxylation of malonyl-CoA. Kallen and Lowenstein (1962) pointed 
out that if this were the mechanism by which malonate acts, it should also 
stimulate the synthesis of fatty acids from malonyl-CoA, which it does not; 
indeed, malonate at 10 niM inhibits the conversion of malonyl-CoA into 
fatty acids 33%. There is actually a stimulation of the formation of fatty 
acids from acetyl-CoA. Furthermore, Spencer and Lowenstein (1962) found 
that malonate is incorporated into fatty acids in an extramitochondrial 
extract from rat mammary gland; acetate stimulates malonate incorpora- 
tion just as malonate stimulates acetate incorporation. All of the stimula- 
tion by malonate, however, cannot be explained by its conversion to mal- 
onyl-CoA since several times more acetate than malonate is incorporated. 
The varying effects of malonate on different preparations from a single 
tissue are well illustrated in the studies of Abraham et al. (1961) with rat 
mammary gland, where malonate stimulates fatty acid synthesis markedly 
in cell-free systems (maximal stimulation around 130% at 17 mM malon- 
ate), inhibits the synthesis 63% in slices, and has very little affect when 
glucose is present. Glucose was assumed to provide NAD(P)H by forming 
cycle substrates and also to augment the ATP level, which in the absence 
of glucose might have been reduced by malonate. In the homogenates ATP 
was added so that this aspect of the action of malonate could not be exhibit- 
ed. The response to malonate is markedly dependent on the experimental 
conditions, as shown by Hosoya and Kawada (1961) with human placental 
slices, additions of estradiol, ATP, NAD, or bicarbonate altering the fatty 
acid synthesis and its modification by malonate. It may be noted that fatty 
acid synthesis in particulate preparations from the locust fat body occurs 
rapidly only in the presence of malonate (Tietz, 1961). 



148 1. MALONATE 

So far we have considered total fatty acid synthesis. Separation of the 
different fatty acids from animal tissues in malonate experiments has not 
been done, but in mycobacteria malonate shifts the incorporation of acetate 
into the higher fatty acids (Kusunose et al., 1960). The synthesis of total 
fatty acid is moderately increased and this was attributed to the formation 
of malonyl-CoA (see accompanying tabulation). Differential effects of mal- 

Acetate-l-C* incorporation 

Fatty acid % Change 

Control Malonate 3.3 mM 

Palmitate 7627 

Stearate 2887 

Arachidate 1144 

Behenate 1089 

Lignocerate 1862 

Total acids 14609 17234 + 18 



onate on the synthesis of short-chain and long-chain fatty acids are also 
seen in rat liver slices metabolizing octanoate-1-C^* (Lyon and Geyer, 1954). 
Although the over all effect of malonate on lipid synthesis in a particulate 
preparation from avocado mesocarp is not marked, the incorporation of 
acetate is shifted from stearate to oleate (see accompanying tabulation) 



3488 


- 54 


1497 


- 48 


1332 


+ 16 


2110 


+ 94 


8807 


-f373 



Malonate 


Acetate incorporation 
into lipid 


Distribution of label 




(mM) 


Palmitate 


Stearate 


Oleate 





8.20 


26 


33 


41 


5 


8.50 


23 


12 


65 


10 


8.15 


24 


14 


62 


30 


7.10 


21 


9 


70 



(Mudd and Stumpf, 1961). Although malonate may inhibit the cycle, this 
may be counteracted by the formation of malonyl-CoA, which dilutes the 
labeled malonyl-CoA formed from labeled acetate. It is interesting that 
malonate is formed from acetate in avocado and this may be one regulatory 
factor in fatty acid synthesis. Malonate has been found in three instances 
to exert only inhibitory effects on fatty acid synthesis: in cell-free prepara- 
tions from pigeon liver, 20 mM malonate inhibits the incorporation of 
acetate into fatty acids 32% (Brady and Gurin, 1952); in various tumor 
tissues (mammary and testicular carcinomata, and a sarcomatoid ovarian 



EFFECTS OF MALONATE ON LIPID METABOLISM 149 

tumor), 30 mM malonate inhibits 8-73% (van Vals and Emmelot, 1957); 
and in rat liver extracts containing mitochondria, 10 mM malonate in- 
hibits 51% (Iliffe and Myant, 1964). 

These divergent observations are difficult to explain satisfactorily and 
one must conclude that the final effects of malonate must depend on many 
factors. In cellular preparations malonate may alter the levels of ATP, 
NAD(P)H, and coenzyme A, as well as divert the metabolism of acetyl- 
CoA by its inhibition of the cycle. In nonmitochondrial soluble enzyme 
systems these actions would be minimized or absent, and the most important 
factors might be the facilitation of fatty acid synthesis through the forma- 
tion of malonyl-CoA or direct effects on the enzymes involved, although 
there is no evidence for such direct affects at present. When malonate is 
itself incorporated into fatty acids, as in several examples above and in 
spinach chloroplasts, where malonate incorporation occurs at about half 
the rate for acetate (Mudd and McManus, 1964), additional complications 
must be considered. Since the incorporation of acetate- 1-C^* into lipid in 
chloroplasts is reduced 71% by 0.67 mM malonate (Mudd and McManus, 
1962), it would appear that malonate also inhibits some step or steps in the 
pathway. The compartmentalization of the pools of acetyl-CoA, malonyl- 
CoA, acetoacetate, and the various enzymes and cofactors within the cell 
must be borne in mind in trying to explain certain differential affects of 
malonate. 

Effects on the Metabolism of Fats, Phospholipids, and Sterols 

Several observations on total lipid response to malonate are interesting 
even though it is impossible to assign a mechanism. In rat liver homogenates 
incubated with palmitate-1-C^*, the lipids other than phosphohpids increase 
22% in the presence of 10 mM malonate compared to controls (Jedeikin 
and Weinhouse, 1954). Whether this is direct utilization of palmitate or 
lipid synthesis with the C^^Og formed from palmitate is difficult to say. 
Malonate at 50 mM also increases the total lipid content of potato tuber 
slices some 230% (Table 1-19) (Romberger and Norton, 1961) and this 
could be due mainly to an increased synthesis of fatty acids. On the other 
hand, lipid synthesis from glucose-C^* in human placenta is depressed 75% 
by 20 mM malonate (Hosoya et al, 1960). These results again show that 
the action of malonate on lipid metabolism is quite variable. It will be more 
profitable to turn to the synthesis of particular lipid fractions. 

Injections of malonate lead to elevation of the free and esterified cho- 
lesterol in the liver, kidney, and blood of the rat (see tabulation) (Mook- 
erjea and Sadhu, 1955). Injections of 800 mg/kg of sodium malonate were 
made daily for 3-4 weeks, some toxic effects being noted, and the animals 
then sacrificed. Simultaneously, the blood glucose rose from 92 mg% to 
196 mg% and the blood acetoacetate from 0.8 mg% to 3.6 mg%. Kidney 



150 



1. MALONATE 







Free cholesterol 
(mg/100 g wet wt.) 


Esterified cholesterol 
(mg/100 g wet wt.) 




Contro: 


Malonate 


% Change 


Control 


Malonate % Change 


Liver 

Kidney 

Blood 


205 

360 

43 


426 
724 

83 


+ 107 
+ 101 
+ 93 


73 

72 
68 


104 + 42 

228 +216 

80 + 17 



and liver slices showed impaired respiration with succinate, acetate, and 
acetoacetate as substrates. This augmentation of tissue cholesterol is clear 
and is reasonable on the basis of diversion of acetyl-CoA metabolism by a 
block of succinate oxidase. However, in vitro work has shown only inhibi- 
tion of cholesterol synthesis. The formation of labeled cholesterol from 
octanoate-1-C^* in rat liver slices is consistently depressed by 5.84 mM 
malonate, and fumarate was very ineffective in counteracting this inhibition 
(Lyon and Geyer, 1954). The total lipids rise and this is partly attributable 
to the increased synthesis of short-chain fatty acids. The formation of 
labeled cholesterol from acetate- l-C^* in the same tissue is inhibited 73% 
by 50 mM malonate (Kline and DeLuca, 1956) and 78% by 30 mM mal- 
onate (van Vals and Emmelot, 1957). Cholesterol synthesis in rat tumors is 
even more strongly depressed. The discrepancy between the in vivo and in 
vitro results might be due to several factors. In the intact animal many 
secondary effects may occur, e.g. as a result of the marked rise in blood 
glucose. Also the malonate concentration in the tissues of the rats is undoubt- 
edly less than in the work with slices. It is unfortunate that most of the 
studies have been made with unreasonably high malonate concentrations 
so that a specific inhibition of succinate oxidation is doubtful. The catabolism 
of cholesterol, as determined by the formation of C^'^Og from the labeled 
terminal methyl groups of cholesterol, in suspensions of rat liver mitochon- 
dria is inhibited 78% by 10 mM malonate (Whitehouse et al., 1959), so 
that this factor must also be considered in explaining changes in tissue levels 
over longer periods of time. The synthesis of other sterols has been studied 
very little. Pieces of rat adrenal form corticosteroids in the presence of 
glucose and this is markedly stimulated by the addition of ACTH. Malonate 
at 10 mM stimulates the formation of sterols in the absence of ACTH 
from 17 to 22 //g/100 mg/2 hr ( + 29%) but depresses the synthesis in the 
ACTH-activated preparations from 81 to 72 //g/100 mg/2 hr (-11%) 
(Schonbaum et al., 1956). Fluoroacetate also inhibits very little and it was 
concluded that the cycle does not play a major role in sterol synthesis, 
glucose metabolism and particularly the pentose phosphate pathway being 
of more importance. The bearing of such studies on the metabolic basis of 
cholesterol and hormonal sterol levels in animals, especially the relationship 



EFFECTS ON AMINO ACID AND PROTEIN METABOLISM 151 

to the activity of the cycle and the other pathways for the utilization of 
acetyl-CoA, warrants further investigations of the actions of malonate and 
other cycle inhibitors both in vitro and in vivo. One approach to the met- 
abolic defect in hypercholesteremia could be made in this way. 

The incorporation of inorganic P^^ into phospholipids is almost invari- 
ably inhibited strongly by malonate. This has been shown in peanut mito- 
chondria (Mazelis and Stumpf, 1955), mycobacteria (Tanaka, 1960), guinea 
pig brain dispersions (R. M. C. Dawson, 1953), rat liver mitochondria 
(Marinetti et al., 1957), and other tissues. In cat brain slices, the effects of 
malonate are very slight and it is possible that malonate does not penetrate 
well (Strickland, 1954). Yet 3 mM malonate inhibits such incorporation 
87% in K+-stimulated rat brain slices, although this may be due to a more 
active cycle participation in the active tissue, inasmuch as respiration is 
93% inhibited (Yoshida and Quastel, 1962). The phosphorylation of phos- 
pholipid precursors probably involves the formation of high-energy phos- 
phate compounds and malonate could depress this as the result of a block of 
the cycle. A direct effect on the phosphorylation is unlikely. On the other 
hand, the incorporation of activity into phospholipids from palmitate-1-C^^ 
in rat liver homogenates (Jedeikin and "Weinhouse, 1954) or from acetate- 
1-C^* in rat liver slices (Kline and DeLuca, 1956) is affected scarcely at aU by 
malonate. The phospholipids comprise a very heterogenous group and the 
response to malonate probably depends on which type of phospholipid is 
under investigation. 

EFFECTS OF MALONATE ON AMINO ACID 
AND PROTEIN METABOLISM 

The pathways of amino acid metabolism often lead to or from the cycle 
so that malonate would be expected to influence amino acid utilization and 
formation by its inhibition of succinate oxidase. The intracellular accumu- 
lation of amino acids and their incorporation into proteins are processes 
requiring energy and consequently malonate could depress these important 
reactions involved in cellular growth by a depletion of high-energy phosphate 
derived from the cycle. Finally, malonate might act directly on the enzjones 
catalyzing amino acid transformations. Information on these matters is 
fragmentary but enough work has been done to demonstrate some interest- 
ing effects of malonate on this phase of metabolism. 

Effects on Amino Acid Metabolism 

None of the enzymes involved directly in amino acid metabolism seems 
to be very sensitive to malonate (Table 1-12) but a number of important 
reactions have never been studied. Enz>Tnes catalyzing the reactions of 



152 1. MALONATE 

the dicarboxylic amino acids particularly might be expected to bind mal- 
onate to some extent, but there is only indirect evidence for this. The 
dehydrogenation of glutamate with methylene blue as an acceptor in 
toluene-treated E. coli is inhibited about 20% by 71 vaM malonate (Quastel 
and Wooldridge, 1928), but it may be that all of the hydrogen atoms do 
not arise from glutamate here and that some other reaction is inhibited. 
In Walker carcinosarcoma, kidney, and liver, glutamate is metabolized 
readily to succinate in the presence of 6.3 milf malonate (Nyhan and 
Busch, 1957), but no controls are avilable so that some inhibition is pos- 
sible. Aspartate and glutamate are metabolized by Hemophilus parain- 
fluenzae; malonate does not interfere with the oxidation of the former but 
does inhibit glutamate oxidation (Klein, 1940). Contrary to these results, 
malonate completely inhibits aspartate oxidation in rat liver homogenate 
(Nakada and Weinhouse, 1950). It was believed rightly that this could not 
be entirely attributed to an inhibition of succinate oxidase. 

Glutamate may be converted to a-ketoglutarate by either glutamate 
dehydrogenase or transamination, or it may be decarboxylated to y-amino- 
butyrate; the decarboxylase is limited mainly to certain bacteria and the 
nervous system of animals, so the major product is usually a-ketoglutarate, 
which can be oxidized through the cycle or participate in transaminations 
whereby it is reconverted to glutamate (this occurs also with y-amino- 
butyrate so that the net reaction forms succinic semialdehyde, ammonia, 
and CO2 from glutamate). The pattern of glutamate metabolism will depend 
on the relative activities of these various enzymes, the availability of other 
amino acids for transamination, and the supply of NAD for the glutamate 
and a-ketoglutarate dehydrogenases; likewise, the response to malonate 
inhibition will depend on these factors. If malonate selectively inhibits 
succinate oxidation, the O2 uptake due to glutamate should be reduced 
moderately (perhaps around 25-50%) unless much of the a-ketoglutarate 
formed is transaminated and does not enter the cycle. Malonate, however, 
occasionally inhibits the formation of ammonia from glutamate, indicating 
some effect on the oxidative deamination. Malonate also inhibits the oxi- 
dation of glutamate by guinea pig mammary gland mitochondria completely 
(Jones and Gutfreund, 1961), which would not be the case if only succinate 
oxidation were blocked. The glutamate respiration of rat brain mitochon- 
dria is depressed 88% by 17.3 mM malonate (see note in Table 1-14) 
(Levtrup and Svennerholm, 1963), which would indicate that glutamate 
is being converted mainly to a-ketoglutarate by transamination (glutamate 
decarboxylase is not present in brain mitochondria). Similar high inhibi- 
tions by 20 mM malonate are observed in the mitochondria from pigeon 
muscle, rat heart, rat liver, and ascites cells (64-99%) (Borst, 1962). Con- 
clusions as to the pathway of glutamate catabolism based on the results 
with malonate depend on the assumption that the inhibition is specifically 



EFFECTS ON AMINO ACID AND PROTEIN METABOLISM 153 

on succinate oxidation, and at these high concentrations this may not be 
true. On the other hand, Das and Roy (1961, 1962) claim that transamina- 
tion contributes little to the metabolism of glutamate in mitochondria from 
Vigna sinensis, and since the decarboxylase is absent, oxidation by gluta- 
mate dehydrogenase would seem to be the major route. Glutamate is con- 
verted primarily to aspartate in rat brain homogenate via the pathway 
glutamate -^ a-ketoglutarate -^ succinate -^ oxalacetate -^ aspartate 
(Haslam and Krebs, 1963). The addition of fumarate removes this inhi- 
bition, as expected. 

Certain amino acids appear to be involved in the functioning of nerve 
tissue and the effects of inhibitors on the metabolism of these substances 
are of particular interest in this connection. Glutamate is accumulated in 
brain and plays a role in the active transport of ions, while y-aminobutyrate 
and A^-acetylaspartate have recently attracted attention because of their 
ability to modify central nervous system activity. Glutamate and K"*" are 
taken up by retina and brain slices in approximately equivalent amounts. 
Malonate at 20 raM depresses the formation of glutamate + glutamine 
only 12% while it reduces K+ uptake 40% (Terner et al., 1950), indicating 
that the major effect of malonate is not mediated through interference with 
glutamate. When guinea pig brain slices are incubated with glucose-u-C^*, 
a good deal of the C^'* appears in amino acids, the most important of which 
is glutamate (Tsukada et al., 1958). Malonate at 10 mTlf inhibits glutamate 
formation around 25%, ^-aminobutyrate formation around 75%, and the 
formation of aspartate appreciably. The total C^* incorporation into amino 
acids from glucose-u-C^* in rat brain slices is inhibited 64% by 10 mM 
malonate at normal K+ concentration but 83% in the presence of 105 
mM K+, which produces an activation of brain metabolism (Kini and 
Quastel, 1959). Such results can be readily explained on the basis of a mal- 
onate-reduced pool of amino acid precursors due to the reduction in cycle 
activity. Glutamate is a central substance in amino acid formation through 
transaminations and anything which decreases the formation of a-keto- 
glutarate would be expected to impair these pathways. Cremer (1964) has 
recently found that 40 mM malonate not only reduces drastically the 
incorporation of glucose-u-C^* into glutamate, aspartate, y-aminobutyrate, 
and protein in brain slices, but also causes a loss of amino acids from the 
cells. This concentration, of course, is probably not specifically inhibiting 
succinate oxidation. A disputation type of reaction occurs in certain 
tissues: 

2 a-Ketoglutarate + NH3 + ADP + P, ^ glutamate + succinate + CO, + ATP 

Tager (1963) used malonate to block succinate dehydrogenase and surpris- 
ingly found that it augments the formation of glutamate in suspensions 
of rat liver mitochondria (see accompanying tabulation). It was suggested 



154 



1. MALONATE 



Control 



Malonate 20 mM 



A O2 (/^ atoms) 


-1.2 


- 1.2 


A a-Ketoglutarate (^ moles) 


-3.9 


-10.4 


A Glutamate (/< moles) 


+2.0 


+ 4.9 


A Esterified phosphate (/< moles) 


+ 1.9 


+ 4.8 



that malonate is converted to oxalosuccinate via malonyl-CoA. The oxalo- 
succinate might function in the NAD- and NADP-dependent isocitrate 
dehydrogenase systems to form a transhydrogenase so that NADPH is 
the eventual donor for the formation of glutamate, oxalosuccinate acting 
catalytically. Such studies show how complex the effects of malonate on 
amino acid metabolism can be. The oxidation of certain amino acids pro- 
ceeds via an initial transamination followed by degradation of the deaminat- 
ed acids. The oxidation of y-aminobutyrate is completely inhibited by 
1 n\M malonate in rat brain mitochondria (Sacktor et al., 1959) but in 
Bacillus pvmilus is not affected even by 40 mM malonate (Tsunoda and 
Shiio, 1959). Whether the former inhibition is the result of an indirect 
suppression of transamination by cycle block or a direct effect on the oxi- 
dative pathway of this amino acid is not known. 

iV-Acetylaspartate occurs at a relatively high concentration in mamma- 
lian and avian brain, increasing rapidly after birth. Its formation involves 
direct acetylation of aspartate and the brain has little ability to metabolize 
this substance. When acetate-1-C^* is injected intracerebrally, some of the 
C^* is later found in A^-acetylaspartate (Jacobson, 1959). The injection of 
malonate with the acetate reduces the incorporation of the C^* about 50%. 
The injection of acetate depresses the level of total A^-acetylaspartate and 
malonate counteracts this. These effects are quite complex and difficult to 
interpret. The concentration of malonate injected was high (1.34 M) and 
could have caused a severe fall in ATP so that acetate activation prior to 
acetylation would be depressed. The rise in ^V-acetylaspartate seen with 
malonate might have been due to a cycle block counteracting the effect 
of the injected acetate, whereby cycle intermediates involved in trans- 
aminations would be decreased, the level of aspartate being maintained with 
more aspartate available for acetylation. There are so many pathways 
associated with aspartate metabolism and acetylation reactions that the 
final effects of a cycle block are difficult to predict. A good example of the 
complex effects of malonate on amino acid metabolism is seen in Table 
1-19, where certain types of amino acid in potato slices increase and other 
types decrease during incubation with malonate (Romberger and Norton, 
19C1). 



EFFECTS ON AMINO ACID AND PROTEIN METABOLISM 155 

Effects on Protein Synthesis 

The intracellular synthesis of protein requires the simultaneous opera- 
tion of many metabolic pathways and thus is susceptible to inhibition on a 
variety of reactions. Some of the processes involved in protein synthesis 
are: (1) the active uptake or accumulation of exogenous amino acids, (2) 
the production of high-energy substances such as ATP from the oxidative 
reactions of the cycle (except in anaerobes), (3) the formation of amino acid 
precursors, again mainly by the operation of the cycle, and (4) all the com- 
plex reactions for the activation and assemblage of the amino acids into 
proteins. There are thus a multitude of possible sites for malonate action 
but, at reasonable concentrations, the most important mechanism must 
be a cycle block leading to both depletion of energy supplies and decrease 
in amino acid precursors. There is no evidence that malonate can interfere 
significantly either with the proteases or peptidases involved in the break- 
down of proteins to amino acids or with the terminal assembling reactions 
for the formation of protein. 

The effects of malonate on the uptake and accumulation of amino acids 
by cells have been studied in three types of tissue. Excised diaphragm main- 
tains the same tissue/medium ratio for glycine as in the whole animal, and 
the marked effects of 2,4-dinitrophenol indicate that glycine is concentrated 
actively (Christensen and Streicher, 1949). Malonate, however, at concen- 
trations of 3-55 TaM does not uniformly alter the tissue/medium ratio. It 
is possible that malonate does not penetrate adequately, because muscle 
is often rather impermeable to anions. The situation is different in Ehrlich 
mouse ascites carcinoma cells. Glycine is accumulated so that tissue/medium 
ratios are often 10-15. In two experiments, malonate at 37 mM decreased 
this ratio from 13.0 to 5.1 and at 40 mM from 13.9 to 4.0 (Christensen and 
Riggs, 1952). This occurred despite the fact that malonate increased the 
synthesis of glycine. In cell suspensions of Gardner lymphosarcoma the 
uptake of labeled glycine is inhibited 73% and the uptake of alanine 56% 
by 10 mM malonate (Kit and Greenberg, 1951). These studies demonstrate 
that malonate can interfere with protein s>Tithesis, at least in some cells, 
by inhibiting the initial process of amino acid uptake. 

The synthesis of protein is usually strongly inhibited by malonate, but 
no analyses of the block have been made and the mechanisms are un- 
known (see accompanying tabulation). The formation of adaptive enzymes 
has often been taken as indicative of the synthesis of general cell proteins, 
but this is not necessarily so, as pointed out by Mandelstam (1961). In E. 
coli any substance acting as a substrate and source of energy represses en- 
zyme synthesis, whereas inhibitors, such as malonate and 2,4-dinitrophenol, 
counteract such effects and stimulate the synthesis. Furthermore, under 
conditions in which /5-galactosidase synthesis is inhibited, the incorporation 
of leucine-C^* into cell protein is not affected. The lack of inhibition in 



156 



1. MALONATE 



Process 


Malonate 
(mi/) 


% Change 


Reference 


Formation of induced /5-galacto- 








sidase in E. coli 


33.5 


+ 31 


Mandelstam (1961) 




100 


- 8 






134 


- 42 






167 


- 79 




Incorporation of leucine-C* 








into tobacco leaf proteins 


10 





Stephenson et al. (1956) 


Protein formation in chick embryo 








tissue culture 


10 


-100 


Gerarde et al. (1952) 


Incorporation of glycine-2-C^* 








into rat liver homogenates 


45 


- 86 


Peterson and Greenberg 


Incorporation of glycine-C'' 






(1952) 


into antibody in rabbit lymph 








nodes 


1.2 


- 67 


Ogata et al. (1956) 


Incorporation of acetate- l-C^^ 








into rat liver slices 


30 


- 50 


van Vals and Emmelot 


Incorporation of acetate- l-C* 






(1957) 


into various tumor slices 


30 


— 74 to —90 


van Vals and Emmelot 


Incorporation of glutamate-u-C^* 






(1957) 


into Walker carcinosarcoma 


6.25 


-23 to -56 


Nyhan and Busch (1957) 


Incorporation of glycine- 1-C" 








into chick embryo proteins 


20 


- 57 


Quastel and Bickis (1959) 


Incorporation of glycine- 1-C^* 








into ascites protein 


20 


- 92 


Quastel and Bickis (1959) 



tobacco leaves was attributed to the presence of preformed precursors or 
energy donors, so that interference with metabolism during the 2-hr in- 
cubation does not modify the assembling of the proteins (Stephenson et al., 
1956). In two instances, glucose is able to partially reverse the effects of 
malonate. Glucose addition to the Walker carcinosarcoma slices reduces the 
inhibition by malonate, sometimes restoring the normal rate of protein 
synthesis (Nyhan and Busch, 1957), and in ascites cell suspensions glucose 
decreases the malonate inhibition from 92% to 14% (Quastel and Bickis, 
1959), although the inhibition, is even increased slightly in chick embryo. 
The marked glycolytic activities of tumor tissue may be responsible for this 
phenomenon, sufficient energy for protein synthesis being obtained from 
noncycle pathways. 

The inhibition of amino acid uptake and the synthesis of proteins and 
enzymes by malonate must be considered in long-term experiments or in 
whole animal experiments, since this could secondarily affect many other 



EFFECTS ON AMINO ACID AND PROTEIN METABOLISM 157 

metabolic systems. Most enzymes are probably in a state of simultaneous 
formation and degradation, so that an inhibitor of synthesis would induce 
a steady fall of the enzyme level in the cells. This could apply, of course, 
to all inhibitors of protein synthesis. 

Effects on Urea Formation 

The terminal product of much protein and amino acid metabolism is urea 
and it has been found that under certain circumstances malonate inhibits 
the formation of urea quite potently. The inhibition has been mentioned in 
connection with its antagonism by fumarate (page 116). The most important 
reactions of the urea cycle comj^rise the following, assuming that glutamate 
is the immediate amino-group donor: 

(1) Glutamate + oxalacetate ->• aspartate + a-ketoglutarate (transaminase) 

(2) Aspartate + citrulline -j- ATP ->■ argininosuccinate + ADP + P, 

(argininosuccinate synthetase) 

(3) Argininosuccinate -> arginine + fumarate {argininosuccinase) 

(4) Fumarate -> oxalacetate (fumarase and lyialate dehijdrogenase) 

(5) Arginine -> ornithine + urea (arginase) 

(6) Ornithine + NH3 + CO2 ~> citrulline {citrulline synthetase) 

Glutamate + NH3 + CO2 + ATP -► a-ketoglutarate + urea + ADP + P, 

This urea cycle thus makes contact with the tricarboxylate cycle at several 
points. The a-ketoglutarate formed in the over all reaction can be oxidized 
through succinate to oxalacetate or can be transaminated to regenerate glu- 
tamate. The operation of the urea cycle thus requires sources for oxalace- 
tate and ATP, both of which may be blocked by malonate. 

Cohen and Hayano (1946) found that 5.7 raM malonate inhibits the con- 
version of citrulline to arginine 90% in liver homogenates when glutamate 
is the amino donor. The mechanism of the inhibition was not apparent 
at that time. These results were confirmed by Fahrlander et al., (1947) 
and, in addition, they showed that fumarate or malate, can counteact the 
inhibition, indicating a block of succinate oxidation. They interpreted 
the mechanism as a depletion of ATP and a consequent inhibition of reac- 
tion (2). Subsequently, they showed that low malonate concentrations (1-2 
ToM) inhibit urea formation as much as 75% and felt this was evidence for 
a specific action on succinate dehydrogenase (Fahrlander et al., 1948). 
The ATP level in the homogenates drops from 130 to 49.6 in the presence 
of 2.5 yrM malonate and fumarate restores the ATP level to normal. It 
was believed that glutamate not only furnishes the amino group but 
also cycle substrates from which the energy is derived; a block by malonate 
at the succinate level would reduce the amount of ATP formed. Krebs and 



158 



1. MALONATE 



Eggleston (1948) then demonstrated a differential effect of malonate on 
the formation of urea depending on whether glutamate or aspartate is 
used as the amino donor, the inhibition being less in the latter case. An 
elucidation of the true mechanism of the inhibition was presented by 
Ratner and Pappas (1949), who showed a very definite differential effect 
of malonate when glutamate and aspartate are used (see tabulation). The 





% Inhibition 


by malonate 20 laM 


Substrate 


Aspartate 






Glutamate 




Arginine synthesis Oj 


uptake 


Arginine synthesis Oj uptake 


None 

Pyruvate 

Oxalacetate 

Fumarate 

a-Ketoglutarate 


6 
11 

8 

1 
Stim 6 


27 
32 

20 

7 
16 




73 38 

57 37 

Stim 22 5 

Stim 2 9 

66 40 



transamination forming aspartate from glutamate is not inhibited by mal- 
onate so the mechanism must be sought elsewhere. It was proposed that 
malonate prevents the formation of oxalacetate and thus indirectly blocks 
the formation of aspartate; fumarate would, of course, overcome this block. 
They opposed the idea that ATP depletion is important and felt that the 
ATP derived from a-ketoglutarate oxidation would be sufficient. However, 
they did not by any means disprove the ATP depletion hypothesis and it is 
quite possible that it also plays a role in assigning an over all mechanism 
for the inhibition. Miiller and Leuthardt (1950) extended these observations 
by showing chromatographically that malonate inhibits the formation of as- 
partate by reducing the formation of oxalacetate from a-ketoglutarate, and 
also demonstrated conclusively that the transamination reaction itself is 
not sensitive to malonate. It should be noted that in the reactions written 
above, oxalacetate appears to be regenerated in the arginosuccinase reac- 
tion followed by the hydration and oxidation of fumarate. but this is appar- 
ently not sufficient to maintain the cycle, probably because much of the 
oxalacetate disappears in other reactions. This is why an external source 
of oxalacetate is necessary. 

EFFECTS OF MALONATE ON PORPHYRIN SYNTHESIS 



The pathway for the synthesis of porphyrins in both animals and plants 
originates in the cycle in the condensation of succinyl-CoA with glycine 
(Fig. 1-16). The succinyl-CoA can be formed either from a-ketoglutarate 



EFFECTS OF MALONATE ON PORPHYRIN SYNTHESIS 159 

or from succinate; the latter reaction requires ATP and is catalyzed by 
succinyl-CoA synthetase (P-enzyme). A total of 8 molecules of succinate and 
8 molecules of glycine is required for the synthesis of a molecule of proto- 
porphyrin. The close connection between this pathway and the succinate 
steps of the cycle, and the great iniportance of porphyrin synthesis in all 
tissues, make the study of the action of malonate on this system interesting. 
We may speculate on the various ways in which malonate could modify 
porphyrin synthesis. (1) If the succinyl-CoA is formed in the cycle through 
a-ketoglutarate, malonate could restrict its formation by blocking the cycle 



Acetyl -Co A 

Fumarote 

(Malonate) 



I 



a-Ketoglutarote Succmote 




-Succinyl -Co A 

^^ Glycine 

a -Ammo- /9 - ketoadipote 
S - Aminolevulinote 

I 

Porphobilinogen 



Protoporphyrin 

Fig. 1-16. The pathways involved in 
porphyrin biosynthesis. 

and reducing the rate of acetyl-CoA entry, especially if no noncycle source 
of oxalacetate is available. (2) If succinyl-CoA can be formed through the 
cycle readily in spite of a malonate block, malonate might divert more suc- 
cinate into the synthesis of porphyrin by inhibiting succinate oxidation. 
(3) If the succinyl-CoA arises from succinate, this requires ATP and mal- 
onate could deplete the system of ATP. (4) Malonate might deplete the 
system of coenzyme A by the formation of malonyl-CoA. (5) It is possible 
in some way that malonate might inhibit the formation of glycine, although 
this is rather unlikely because there are usually several pathways avail- 
able for glycine synthesis. The effects of malonate will thus depend on the 
type of preparation used and the conditions of the experiment. 

Duck erythrocytes (intact or hemolyzed) incubated with succinate and 
glycine form porphyrin. Succinyl-CoA could be formed from succinate either 
directly or through the cycle and the relative importance of these pathways 
may be demonstrated by the use of succinate-C^'* with subsequent deter- 



160 1. MALONATE 

mination of the porphyrin labeling (Shemin and Kumin, 1952). Succinate- 
C^^OO" when oxidized through the cycle gives rise to a-carboxyl-labeled 
a-ketoglutarate and hence to unlabeled succinyl-CoA; therefore, no porphy- 
rin labeling should result from this pathway. However, succinate-C"00~ 
could also directly form succinyl-CoA, which in this case would be labeled 
and C^* would be found in porphyrin. On the other hand, succinate-C^^Hg 
would form labeled succinyl-CoA by both pathways. If it is assumed that 
malonate inhibits the oxidation of succinate and the cycle pathway only, 
malonate should not inhibit any porphyrin labeling after incubation with 
succinate-C^*00~, but should inhibit appreciably the porphyrin labeling 
from succinate-C^^Hg. This was found by Shemin and Kumin, as the aver- 
aged results in the accompanying tabulation show (figures are counts/ 

Hemin Control Malonate 20 mif % Change 

From succinate-C'^OO- 215 210 - 2 

From succinate-Ci^Hj 1025 421 -59 

minute for intact erythrocytes). It would appear that both pathways are 
operative in these cells. The failure of malonate to increase the porphyrin 
labeling from succinate-C^*00~ is rather surprising because one might 
expect malonate to divert some of the succinate from the cycle into the 
formation of succinyl-CoA. It is possible that this effect is somewhat coun- 
teracted by an inhibition on succinyl-CoA synthetase. 

Further information on porphyrin synthesis and the effects of malonate 
were obtained by Wriston et al. (1955) by the use of labeled acetate. Different 
malonate effects were obtained when methyl-labeled and carboxyl-labeled 
acetate were incubated with glycine, the inhibition of porphyrin labeling 
being much greater with the former (see tabulation). This is the expected 

Hemin Control Malonate 20 mM % Change 

From acetate-Ci^Ha 376 174 -54 

From acetate-C'^OO- 72.5 67 - 8 



result, because the formation of labeled succinyl-CoA from acetate-C^*00~ 
does not involve the complete cycle and the C^* pathway does not go through 
the succinate oxidation step, whereas porphyrin labeling from acetate- 
C^^Hg depends on the operation of the entire cycle (except for the contri- 
bution from the y-C atom of a-ketoglutarate). Furthermore, the labeling 



EFFECTS OF MALONATE ON PORPHYRIN SYNTHESIS 161 

in the porphyrin from acetate-C^^Hg should be altered by malonate. 
Labeling in the carbon atoms of the A and B pyrrole rings of protoporphyrin 
occurs after incubation of duck erythrocytes with acetate-C^^Hg and glycine. 

CH, 9 

II 
6 H3C CH 8 

' I I 

' fi^N \ 

I 

H 





Control 


Malonate 10 


mM 


% Inhibition 


Total porphyrin 


186,000 


54,000 




71 


Pyrroles A and B 


88,000 


27,500 




69 


Carbon 4 


13,000 


3,700 




72 


Carbon 5 


12,000 


900 




93 


Carbon 6 


20,000 


9,400 




53 



Acetate-Ci^Hg will lead directly to -OOC-C^^HaCHa-CO-CoA and if the 
cycle is blocked completely by malonate, carbons 6 and 9 only will be 
labeled, except for some labeling of carbons 4 and 8 due to the reversible 
reaction succinyl-CoA :^ succinate (as long as ATP is available). Carbons 2, 
3, and 5 should not be labeled. This is essentially seen in the tabulation. 
The cycle, of course, is not blocked completely so that some labeling in 
carbon 5 occurs. The over all inhibition is due to a depression of the entry 
of acetate into the cycle. These experiments not only show the variable 
effects of malonate on a pathway associated with the cycle but well illus- 
trate the use of an inhibitor to elucidate a metabolic pathway. 

The analysis of the action of malonate on porphyrin synthesis was ex- 
tended by Granick (1958) in his work with chicken erythrocytes. The for- 
mation of protoporphyrin is innibited 90% by 10 mM malonate when only 
glycine is present, 85% when succinate is added, and 80% when a-keto- 
glutarate is added. The effects of different concentrations of malonate are 
shown in Fig. 1-17. Malonate could decrease the incorporation of succinate 
into porphyrin by blocking the cycle and reducing the ATP level, and thus 
inhibit both pathways of succinate-CoA formation from succinate. However, 
the quite strong inhibition of protoporphyrin formation from glycine 
-f a-ketoglutarate is surprising. Inhibition of the step a-ketoglutarate — * 
succinyl-CoA is not likely as the only explanation, because 1 raM malonate 
inhibits protoporphyrin synthesis 32% and there is no reason for thinking 
that this low concentration would inhibit a-ketoglutarate oxidase. The for- 



162 



1. MALONATE 



mation of protoporphyrin. from S-aminolevulinate is not inhibited by mal- 
onate so that an action on this part of the pathway is excluded. Granick 
suggested that malonate reacts with coenzyme A and thus depletes the 
system so that succinyl-CoA cannot be so readily formed. These effects 
were confirmed in lysed chicken erythrocytes by Brown (1958). Porphyrins 
are not formed in these preparations but (5-aminolevulinate is formed from 
glycine and succinyl-CoA derived from a variety of cycle substrates. Mal- 
onate at 10 mM inhibits this reaction 30% when the incubation is with 
glycine and citrate, substantiating the action on this region of the pathway. 






1 50 







2 I 

A Arsenite 

B DNP 

C Fluoroacetote 

D Malonate 




pi ► 

Fig. 1-17. Effects of four inhibitors on the sjti thesis 

of protoporphyrin in chicken erythrocytes with the 

substrates as indicated. (From Granick, 1958.) 



It is interesting that the addition of succinate to glycine and citrate in the 
incubation medium leads to an inhibition of (5-aminolevulinate synthesis. 
This was shown to be due to the formation of oxalacetate, which inhibits 
a-ketoglutarate oxidase. Malonate is able to overcome this inhibition by 
succinate through the prevention of oxalacetate formation. This indicates 
another minor mechanism for the effect of malonate on porphyrin synthesis, 
namely, the reduction in oxalacetate concentration and a consequent release 
from any inhibition on the oxidation of a-ketoglutarate. Finally, we may 
note that coproporphyrin synthesis from glycine and a-ketoglutarate in 



EFFECTS ON METABOLIC PATHWAYS 163 

Rhodopseitdomonas spheroides is inhibited 50% by 20 mif malonate and 
75% by 40 mM malonate (Lascelles, 1956), indicating again some inhibition 
of the formation of succinyl-Co A. 

The incorporation of iron into heme, as demonstrated with Fe^^, is not 
inhibited by 10 mM malonate in canine reticulocytes (Yoshiba et al., 1958) 
but is inhibited 26% in chicken erythrocytes (Kagawa et al., 1959). It is 
possible in the latter case that the inhibition is due to the chelation of part 
of the Fe^^, making it unavailable for incorporation. 

EFFECTS OF MALONATE 
ON MISCELLANEOUS METABOLIC PATHWAYS 

There have been many reports on the actions or lack of action of malonate 
on enzyme reactions or metabolic pathways of varying degrees of impor- 
tance. Some of these are worth mentioning, either because they indicate 
areas where further study might be profitable or because they provide some 
evidence for noncycle actions of malonate. 

One might expect very little effect of malonate on photosynthesis but, 
although very little work has been done, in every case some effect has been 
observed. Even the Hill reaction is susceptible to inhibition (Ehrmantraut 
and Rabinowitch, 1952). This reaction is the photochemical oxidation 
of water with the production of oxygen and the reduction of a substance, 
usually quinone, other than COg. In Chlorella this reaction is inhibited 
30% by 6 mM malonate and 50% by 60 mM malonate. The inhibition is, 
surprisingly, prevented by fumarate, indicating that the site of action is 
succinate dehydrogenase and that this enzyme takes part in the transport 
of hydrogen in the Hill reaction, which would not be the case if quinone 
were serving as the immediate hydrogen acceptor. It may also be that 
malonate does not inhibit the Hill reaction directly, but depletes the cells 
of cycle intermediates or other cycle products necessary for the Hill reaction 
to proceed. Malonate not only inhibits the total incorporation of C^^Og in 
Scenedesmus by about 20%, but almost completely blocks the formation of 
labeled malate (Bassham et al., 1950). This was taken as evidence that mal- 
ate is not on the direct line of phosphoglycerate synthesis, but it also dem- 
onstrates that by some mechanism malonate can inhibit COg incorporation. 
An inhibition of glucose formation in Chlorella by malonate has also been 
reported (Kandler, 1955), although there is less inhibition in the light than 
in the dark. The synthesis of glucose was believed to be closely related to 
the formation of high energy phosphate intermediates, and it is thus inter- 
esting that malonate inhibits the photosynthetic phosphorylation of ADP 
in Rhodospirillum, although the inhibition is only 17% at the very high 
concentration of 100 mM (Smith and Baltscheffsky, 1959). With these lim- 
ited observations on the effects of malonate, it must be admitted that it 



164 1. MALONATE 

is difficult to fit the data into the modern concepts of the carbon pathway in 
photosynthesis, which does not directly involve the cycle, and particularly 
to understand the mechanism whereby the Hill reaction is inhibited. 

Malonate is also able to interfere in the metabolism of glycerol. Glycerol 
is fermented to succinate, accompanied by the uptake of CO2, in Propioni- 
hacterium pentosaceum, and 30 mM malonate inhibits both the glycerol 
fermentation and the CO2 uptake around 10% (Wood and Werkman, 1940). 
It is impossible to attribute this to an action on succinate dehydrogenase. 
The oxidation of glycerol may involve an initial phosphorylation with 
subsequent formation of pyruvate and entry into the cycle: 

Glycerol + ATP -> glycerophosphate -> pyruvate -> cycle 

The phosphorylation is inhibited in rat liver homogenates (Ruffo and D'A- 
bramo, 1952), but the total oxidation is not markedly affected in the 
mycobacteria (G. J. E. Hunter, 1953). Malonate at 10 mM inhibits 5% in 
M. stercoris and stimulates 4-6% in M. smegmatis and M. butyricum. It is 
surprising that greater inhibition is not observed if the oxidation does 
involve the cycle. In castor bean cotyledons, malonate at high concentra- 
tions has very marked effects on the utilization of glycerol (Beevers, 1956). 
At 70 mM the C^^Oa formation from labeled glycerol is inhibited 97% and 
the sucrose formation is inhibited 82%, while at 130 mM the oxygen 
uptake is inhibited 60%. Such high concentrations may interfere with the 
formation of ATP and hence depress phosphorylation and also block the 
cycle CO2 release. 

The stimulation of muscle respiration by insulin is inhibited potently by 
malonate (Stare and Baumann, 1940). Insulin almost doubles the respiration 
of minced breast muscle from depancreatized pigeons and 1 mM malonate 
inhibits this increase 92%. Fumarate is able to overcome both this inhibi- 
tion and the inhibition of nonstimulated respiration completely. This 
interesting action in vitro led to a study of the antagonism in the whole 
animal. Solutions of sodium malonate were injected subcutaneously in 
rabbits either before or with insulin and the drop in blood glucose was much 
less than with insulin alone. Insulin (4 units) decreases the blood glucose 65% 
in 4 hr, whereas with malonate present the reduction is only 13% and none 
of the rabbits goes into convulsions. Malonate alone increases the blood 
glucose 26%. The respiratory inhibition in the muscle mince could be 
explained on the basis of a typical cycle block (although the degree of inhi- 
bition is surprising for a concentration of 1 mM), but the inhibition of glu- 
cose utilization in the animal is more complicated. The initial phosphoryla- 
tion of glucose and its uptake could have been inhibited indirectly by a re- 
duction of the available ATP, or it could have resulted, at least in part, 
from a hyperglycemic action of malonate unrelated directly to the insulin 
stimulation. 



EFFECTS ON METABOLIC PATHWAYS 165 

Succinate and propionate are formed anaerobically in Ascaris muscle 
from glucose and lactate, presumably by the following pathway: 

Glucose ^ 

^>fc^ +CO, +4H „ -COj ^ 

Pyruvate »=- Oxalacetate *- Succinate *~ Propionate 



Lactate -""^ 

Malonate at 20 mM does not appreciably inhibit the decarboxylation of suc- 
cinate to propionate (about a 13% reduction in total radioactivity) but the 
small inhibition indicates a possible competition with succinate for the 
enzyme. However, the incorporation of lactate-2-C'-^ into succinate is inhibited 
almost 90%. If succinate is formed by reduction of fumarate derived from 
oxalacetate, malonate would be expected to inhibit well, not only because 
of the effect on the succinate dehydrogenase but also by an inhibition of 
oxalacetate formation. Malonate inhibits the formation of labeled propionate 
from lactate-2-C^* 65%. The smaller inhibition compared to that for suc- 
cinate formation implies another less important pathway for the formation 
of propionate, perhaps by direct reduction, as shown in several bacteria. 

The metabolism of glyoxylate by Kver mitochondria is rather complex; 
it is decarboxylated to formate by a devious route, it may be oxidized to 
oxalate, or it may be aminated to glycine (Crawhall and Watts, 1962). 
Malonate inhibits the decarboxylation competitively but does not interfere 
with the formation of oxalate or glycine; indeed, the latter may be stimul- 
ated slightly due to diversion in a branched chain. The decarboxylation 
reaction, which requires glutamate, is quite sensitive to malonate, around 
50% inhibition occurring at 0.15 mM, both substrates being at 3 mM. 
This would certainly appear to be one system in which a marked effect can 
be exerted by malonate at low concentrations and which is unrelated to 
succinate oxidation. 

The synthesis of acetylocholine is an endergonic process and is related 
to the cycle both for the supply of energy and with respect to the utilization 
of acetyl-CoA. The effects of inhibitors on acetylcholine synthesis and hy- 
drolysis are particularly important when considering the mechanisms by 
which malonate can alter nerve and muscle function. Unfortunately, only 
one study of the action of malonate has been made (Torda and Wolff, 
1944 a). The formation of free acetylcholine in minced frog brain, in the 
presence of physostigmine to prevent hydrolysis, is inhibited 32% by 0.08 
mM, 46% by 0.8 mM, and 49% by 8 mM; the inhibition of total acetyl- 
choline is about the same. Succinate, fumarate, and citrate increase acetyl- 
choline formation. It would be interesting to investigate the effects of mal- 
onate on the purer enzyme systems now available for acetylcholine synthe- 
sis to determine if the inhibition is a direct effect or secondary through 
ATP depletion. The effects of malonate in the intact cell may be quite 
complex, because malonate might suppress the incorporation of acetyl-CoA 



166 1. MALONATE 

into the cycle and thereby lead to a greater availability of acetyl-CoA for 
choline acetylation. However, it is not known if the acetyl-CoA pool is 
common to both the cycle and the synthesis of acetylcholine. In many 
cases the effects of malonate are due only to a depression of the cycle oper- 
ation and a decreased formation of ATP. For example, in the synthesis of 
chondroitin sulfate in tibial condyles of chick embryos, the fixation of sulfate 
is inhibited by malonate in a parallel fashion to the inhibition of respiration 
(Boyd and Neuman, 1954). The fixation of sulfate requires ATP, as shown 
by the marked inhibition with 2,4-dinitrophenol, so that here the mechanism 
of malonate action is simply an inhibition of energy formation. Other proc- 
esses, such as calcium deposition in tibial cartilage (Hiatt et al., 1953), 
do not require energy and are not inhibited by malonate. 

EFFECTS OF MALONATE 
ON THE ENDOGENOUS RESPIRATION 

The alterations of the most important metabolic pathways by malonate 
have been discussed, and we shall now conclude this aspect of the subject 
with a survey of the effects on the total oxygen uptake of cells respiring in 
the absence of any external substrate. Although the interpretation of the 
results of such studies is very difficult, the changes in the endogenous respi- 
ration have been examined more frequently than any other response to 
malonate. There is, thus, a vast and variable mass of data, some of which 
is summarized in Table 1-26. The aim of most of these investigations has 
been to demonstrate the absence or presence of the cycle in the types of 
cells tested, and we must attempt to assess the validity of conclusions based 
on the response to malonate. The most unsatisfactory work has been done, 
and the most unjustified conclusions have been drawn, in studies of this 
type, inasmuch as the inherent complexities of the situations have seldom 
been appreciated. Although the cycle has a wide distribution in the cells of 
microorganisms, plants, and animals, its operation during the metabolism 
of endogenous substrates is quite variable and dependent on the state and 
past history of the cells. 

Factors That May Determine the Degree of Malonate Inhibition 

Certain basic factors should be considered in every investigation of the 
susceptibility of the endogenous respiration to malonate. Although some 
of these have been mentioned previously and some will be taken up in 
greater detail later, it may be convenient to enumerate here the most im- 
portant. 

(a) Intrinsic susceptibility of succinate dehydrogenase to malonate. This en- 
zyme from different species varies a good deal in its ability to bind mal- 



EFFECTS ON THE ENDOGENOUS RESPIRATION 167 

onate, as is evident from the range of K^ values observed (page 33). It is, 
perhaps, too often assumed that in every organism the succinate dehydro- 
genase wiU be readily blocked by malonate and that the inhibition of the 
endogenous respiration depends only on the importance of the enzyme in 
the total oxygen uptake. It is mandatory to demonstrate the sensitivity 
of succinate dehydrogenase to malonate in the preparation being studied. 

(b) Degree to which intracellular succinate dehydrogenase is inhibited. In 
addition to the intrinsic susceptibility, there are other factors which can 
alter the inhibition occurring within the cell. The concentration of succinate, 
both initially and following the accumulation resulting from the inhibition, 
may be high enough to oppose the malonate effect appreciably. The relative 
stability of plant respiration to malonate has been attributed to the high 
concentrations of succinate and other organic anions in plant cells. It is 
probably very seldom that an inhibition even approaching completeness 
can be achieved in cells at concentrations likely to be specific. 

(c) Specificity of malonate inhibition. Inhibition of the endogenous res- 
piration can, of course, arise from actions other than on succinate dehydro- 
genase. If the object of the study is to evaluate the contribution of the 
cycle to the oxygen uptake, inhibitions on noncycle pathways must be 
eliminated. At the high malonate concentrations often used (see Table 1-26), 
there is certainly no assurance that the inhibition is specific. 

(d) Intracellular concentration of malonate. Malonate does not penetrate 
readily into most cells, especially at physiological external pH values, so 
that the internal concentrations of malonate may be far below those in 
the surrounding medium (see page 190). The degree of respiratory inhibition 
observed is probably often more of a measure of malonate penetrability 
than of the nature or susceptibility of the metabolic systems. There are 
several instances in Table 1-26 in which the inhibition rises with destruction 
of the normal tissue structure or the removal of permeability barriers. In 
general, the inhibition is greater in homogenates than in minces, and greater 
in minces than in slices or intact cells. The results of Bonner (1948) on 
Avena coleoptiles are interesting in this regard. Soaking in water for 24 hr 
increases the susceptibility to malonate and removal of the endosperm 
further increases the inhibition. The many observations that a lowering 
of the pH augments the inhibition also provide evidence of the importance 
of permeability. 

(e) Metabolism of malonate. Many tissues and organisms can metabolize 
malonate to acetyl-CoA, the oxidation of which contributes to the oxygen 
uptake (see page 228). Some of the respiratory stimulations noted with 
malonate must be due to this, and it is likely that the experimentally de- 
termined inhibition in other cases is reduced from that which would be ob- 
served if malonate were not metabolized. The best way to test for and 



168 



1. MALONATE 



^ 



1^ 




1— 1 




05 'ki 


(N 


t^ 




1— 1 CO 


-* 


-^ 


■ — - -k^ 


05 




o 


!h 




'a 


o 




" — ' 


S 


ft 




"e 




'^ 




>« 


-c 


S 


!3 a; 


u 


c 


c 


cS O 


S 


ce 


o 


-iU t^ 


"m 






u |3 


S3 






o 

> 













m 



it >« 



-d 


rt 


c 




rt 


TJ 


C 




o 




m 


t^ 








o 


i 


PQ 



s a s s 

"-J3 '-iS +3 -t3 

M 02 02 03 



« 






ft 



CO ■* QO "M O "M O 'M 



GO 00 






'MOO-HOOOTt<OiCOO'OiC 



O -Tit 



11^ 



« (B 


O) 03 


ft ft 




t» tc 


s s 


^ r*^ 


02 02 


02 W 






>. 

% 



ft 

02 



oq ^ 



g^ 





•i 




as 

=0 


^ 

^ 




_£ 


=0 


s*. 


O 


'S. 




O 


C 


"^ 


e 




i 


1 


=5 



0^ -^ 



EFFECTS ON THE ENDOGENOUS RESPIRATION 



169 



c3 ^' 



2 OJ ^ 



Oi 






o 



Z o 



W 



t^ O <M 00 Tj< <M 






-a 



^ 3 c 









c 


c 


C 


C 


c 


£ 


j5 


c 


c 


g 


c 


3 


C 







_o 


_o 


_o 








_o 


_o 


_o 


_o 


.2 





_o 














a 


















tr. 


■5 


'S 


'S 


'S 


'S 


'm 


'm 


"m 


'ro 




"m 




s 


s 


c 


c 


c 


05 

-2 


C 


C 


fi 


C 


C 


f^ 


C 




0) 


(L> 


V 


D 


v 


<a 


a:i 


v 


«> 


0) 


ID 





CO 


a 


a 


a 


a 


a 


15 


a 


a 


a 


a 


a 


a 


a 


« 


CO 


m 


m 


03 


m 


P 


CO 


CO 


m 


(O 


(C 


03 


ai 


_o 


S 


S 


3 


^ 


3 


^ 
P 


3 


3 


3 


3 


3 


3 


3 




CO 


CB 


J» 


M 


M 


M 


m 


M 


m 


m 


CC 


CO 


s 



03 ro 

3 3 

- 3 



I i 

O S 
CO «i) 



.e -3 



M ® 'S -g 



&q 



o o cS ' « „ „ 






3 S 

«= J t; S 



-3 -3 



« e6 



S 3 2 






<=« ^ a .'^ 

m ft w w 



^ ^ 


^^^ 


W 


^_^ 


00 


00 




05 


-* 


Tl< 


T3 


■* 


05 


Oi 


C 


Oi 


)-H 


y—i 


cS 


1— 1 


^-^ 


^— ^ 




^— ^ 


Lh 


tH 


3 


tl 


« 


<U 


3 
-0 


« 


3 


3 


3 


3 


c 


3 











m 


pq 


< 


m 



COI>l>^(Nf-ifCt^C0005 



00 

0-^0>CO»0000000»OM 



-H O 10 

00 >c Tt< 



10 >o o »c 

■^ T)H O ■<* 






o 



,— V 









-0 


-o 






<o 


c 






M 


Ci 






ee 


















3 


c 






_« 


^ 


-2 


s. 


















-1-^ 


-»^ 


■k^ 


-u 


a 


a 


a 


a 














-2 


® 


V 


0) 










"o 


"o 

















Q 



S c3 












O 



170 



1. MALONATE 



P5 



a 


Tt< 


05 


■? 


■* 


05 


Oi 


i-H 


r— 1 


^"^ 


















OJ 


^ 


■-3 




J 


§ 



Oi -^ 



w o 



c3 X! ^ 

.£ r- t- 

M IC >0 

'O ri rl 

H-) ^ ^ 

-§ ^ ^ 



-5 ,« ,::i; 
S cS cS 

1:15 PH fe 



a 2 



P^ Q 



o 00 00 o 10 

10 00 I-H ffO 10 



iCTj<iO00O5f0l>iO 



000^000 






iC o 



O >C lO >0 Tj< 

t^ TjH -^ ■<* CO 



10 ira o 

TjH O l> 



flH 



o o 
o o 

Pi 05 



o o 
o o 






g oj o 



til a P, ^ -72 ^ 
03 pR O W CB 






pq 



^ 






EFFECTS ON THE ENDOGENOUS RESPIRATION 171 



^ 




3 




ert 




-tJ 




C/J 


<N 


'O 








cS 


^^ 


C 


^ 




c 










01 




^3 

c 


c 









O) 






<M O •>* CO O O C^ 



00«0->*0 »C»CiOiO'0>OlOOOOOOOOOOOOCO-*CDTf<0 
•^ Oi -^ ^ »ClO»OOiClCiMlOOOO CO M 

—1 -H CO GO 



»CO»CiOOOiCOOOOOOO »0 CO CO o 

eO'*'*>CCOl:-»0-<*CO->*OCOI>>C -* »C CD ;o 



OC Coo O S rf 



o •£ 'I 

i g I ^ 



172 



1. MALONATE 






m 



2? <» 



M 



P. 


a 


T3 


-d 


< 


< 


i 


G 
c6 


T3 


T3 


bl 


b 


C 


C 


<u 


0) 


c3 


cS 


M 


M 






Lh 


(H 


>^ 


>> 


<D 


4) 


-2 


(D 


-Q 


,0 


"t^ 


" 


s 


g 


ce 


C8 


o 


O 


K 


ffi 


tf 


tf 



s -^ 









(M C<1 O O O 
Ttt Tt »0 lO lO 



-£3 -a 



.2 S .S 



PL, 



§ >> 3 S g 

P? S H H 1-^ 



J< 



a 

CO 



to 


3 






S 


M 






i< 


eS 






S 


fe 




o 


e 




S 


« 


r^ 


^ 


o 


^i:) 


o 


c3 


c3 


^ 




+3 

o 


X5 
O 




pq 


(X, 


H 



EFFECTS ON THE ENDOGENOUS RESPIRATION 



173 



W5 

OS r\ 



m ^ —H 









o 

s 






>■ o 



05 eg 



^ t^ 1- ^ 

1^ O O Ci 
0^0 05 — ' 



05 1— I 



'O 43 

(^ Pi 






■>* 


b 


o> 


« 






>_^ 


c 




3 


S 


X 



m is S =* 



?►. O O « MO X^^ 

a5fiQW<:!i!a50 



o^ 



tf 



05 



T3 
05 



o 



CCi Tfi 00 <M O Oi »C 
^H ^ -^ ^H CO CC "^ 



O CO 00 O -M O O 



.s .s .s 

'-3 '-3 ^3 
CO M OQ 



O O -H U5 

lO lO i> ■* 



JC '^ '^ '^ 
t^ CD CC t^ 



o o 



m xn " 



O O 



o o c, "5 

05 P5 fe (1^ 





















s 




s. 


c 


c 


C 


C 


C 


C 


C 


j2 


J^ 


_3 




cS 


_o 


o 


_o 


o 


o 


o 


.2 


_o 


.2 




C 


'ro 


■^ 


'S 


'm 


'm 


■^ 


"S 


■^ 


"« 


O 




« 


C 


c 


c 


fl 




c 


c 


c 


c 


© 


S 


<» 


(D 


« 


(B 


s 





« 


<» 


© 






& 


& 


c 


ft 


& 


ft 


ft 


ft 


ft 


^ 


"o 


m 


n 


OJ 


OJ 


CO 


tc 


05 


02 


tc 






o 




5 


3 


3 


3 


3 


3 


3 


3 


s 


> 


m 


a; 


CO 


02 


CB 


02 


02 


02 


02 


> 



Oh 



W 



a? 



a S 

ft o 

<i1 03 



00 


00 




'S. 




•^ 
-< 




=0 




Q 


B 


S£ 


S 


p 



OQ &i 









_s 2? 






*>. *s. \,«i -o (O -_,- 

E-1 ^ ft; Oh o S 






^ « © 






rfi cS 00 



K3 



- t ° 

cS ?5 e8 



01 B Oi 



174 



1. MALONATE 



lO 2 



tf 



o 

05 


O 

P5 


o 

P3 


t3 
C 


C 


T3 

c 



Ph 



SU 


r] 


lO 


" — 




O 


05 


_ 


■5 


a 




^ 


-*:i 

2 




03 


'$ 

2 


5 
3 















c 






-O 


^ 


05 


Oi 


s 


"o 

O 


•«* 


-* 






Oi 


05 


t3 


t3 




I-H 


fi 


C 


■ — ■ 


^-' 


a 


cS 


-o 


73 


>i 


c 


c 


C 


« 


o 


-2 


-S 


L4 


h 


"S 


"3 


-^ 




o 


o 


'TS 


w 



X 



«0 T(< Tt* 

t> I> t^ 






>0 LO ^ <N 

i> i> 00 00 



Ph 



i i i I 



o ^ ^ 



N .5 CD 2 lU 3 

6 § W ^ W S 



o 


C 




'm 


4J 




c 


M 




ID 


O 






S 




oo 


o 


^ 

WJ 


M 


M 


c 


bC 


M 


cS 


W 


w 


§ 



f»H 







•c* 










S 








!^ 




to 














<U 










e 








«0 








V 








e 








»o 




^ 




"to 




J 




-S 


5s 


.e 


"e 


^_^ 


e 


^^.^ 


<a 


^— ^ 


'ft. 




'•^ 


§> 


® 


s 


(D 


?^ 


CD 


«0 


Q 


00 


t3 
S 

s 
s 

c 




T3 
S 

S 


1 


o 

1 


'ft, 


1 


1 


^ 


^"^ 


fe; 




(^ 


"^"^ 


'^ 


^ 


&5 



CQ 



O 



— . CI 



O e •= 



EFFECTS ON THE ENDOGENOUS RESPIRATION 175 



O ^ C5 

•^ n n 

Oi Oi Oi 






2 _ H^ir^i^^J^s^' -25 



5 >^ 



N 3 



: i 






^ >> 3 C C 






Me'iOitNTti-^ccaor-CLoao-^'i^ioOfcOt^^ioo 



OOOOO-^OOOOOO-^-^OOO 
-Hrt-*OOfC'MTtiO — O-^'N'MOO-H 



5 



^ -^ 



c 





S 




cS 




bi 




X! 


2 


g 


c3 


43 


<u 




bo 


O 


o 

g 


1 


o 






M 










>> 


ei 


« 


_o 



t3 



^ 




M 


'S 




tc 




o 


'£ 


O 

o 






o 

C/3 


S 


o 


^ 




■^ 








>> 






c« 


_» 


^ 


^ 


Lri 












^ 


'3 


'S 


O 


to 

3 


73 

2 


o 

CO 

3 



WW ^p:S tiio WWfqpq^SSS 



I I g § 

;^ I a 5 i 

C =c C .S -2 -^ 



S i; 



S -^ 



6 t: ]B .3' 5 



176 



1. MALONATE 



P5 



w 



n 




to 




05 


^— V 


i-H 


l~- 


^—^ 


lO 




<35 


s? 




^ 


^^ 






s 


"e 


-d 


.^ 


c 


<u 


cS 


.^ 




^ 


c3 


cS 


3 

s 





w 



to Ji 

OJ o 

-H > 

^ o 

-*^ 0-1 






^ a 

lO W CQ 



0) CO 



£-2 2 

l-J cc c» 






-^ 73 M 

t- ti Sh 

cS <S 3 

J h-l pq 









« o o o o 



a 

02 



1 o 
o -* 



K 

a 






<o 


f—i 


o 


_o 


3 


_o 


"m 




"oo 


3 


bo 


C 




3 
3 




cS 


'S 




t-i 


PQ 




m 



3 ^ 

8 8 :>. § § I g .£ .£ 

CO GO w fefl ^7 TT! ""^ '-" '~> 



o > 



«« y .S 3 



pq pa g "- J J 



a; 4) 



3 ^ 0? -3 S 



3 3 S ^ ^ £" p 



c P c 



cc cc cS 



^^ 






w 



a 

CO 



EFFECTS ON THE ENDOGENOUS RESPIRATION 177 






TjH 2 



^ cd c^ ^ 









cS 



cS 



^^ >-* -^ r^ 1-^ /•■\ I--1 (•-! r:) r^ fi r-i ,-^ ~.' ~.* Q 



3 3 



— . 


o 


3 




05 


>. 


C 


_ 


c 


o 






o 


>. s 


C 


T3 


T3 


CO 


o 


C 

0) 

73 




Oi 

3 
O 


C 1* 












a 


cS 




T1 


cS O 


« 


Ph 


|-^ 


W 


Q W 



WKKpqpqpqpqpqpqpqpqpqm -^-SWp^oS Go 

COCD^OOO>OOCC'000 iCOiMOOiC 00-<*-^OCO'*fO'MOO 
(M-HCOiMQOOSOOt^CO'-fOCQDiO »OfOMTt<»OfO lO 00Or-<M-HC0'*>0 



t^ t^ t^ o o 

-^ -^ -H O O 



CO eo C5 I ■* 

t- C^ l> o 



c .z ■-. ? •- 













^ 






.2 


c8 


c 


^TO 












XH 


'S 


"S 


'E 


'C 


o 


IS 














L4 


u 




73 


o 


2 




§ 




'S 


-2 

3 

s 




Xi 

M 
3 
_3 


S 
o 

•*3 


3 
O 

o 
O 


o 
00 


3 

S 

o 


JD 




^ 

^ 


(h 

^ 


o 


o 


O 


s 


3 


0) 


00 




> 


o 
3 


3 
3 


x> 
S 


a 


J2 

s 


'S 




o 

2 


-2 



•-- r 5; •■- S; ^ H <D y 2^ X! -Q J2 -3 ni C3 .ii — H .3 .3 .3 .3 






178 



1. MALONATE 



P5 



1^ 

cs -3- 






PLh 



CO Ci 

C5 -H 



<o 


.5 


05 


■*^ 


'(C 


-* 


o 


73 


05 


73 






C 


S 


c 


eS 




^^ 




(S 


« 




73 


-Q 









•^ o 
P5 T '2 



M T3 ii^ ^ S ^ - 



CLh 






c5 g:i 









Oi 




rj 




O) 


TjH 


















Ol 




O 






00 "-' 


fO 


o 


o 


"m 


■ — - -u 




05 

73 


CO 




o 




C 






>> PL, 


>. 


eS 




73 


o 

1^ "^ 


o 
i:i5 


£? 






73 c3 


73 


0) 




73 


C «^ 


c 


C 




i:« 


2 0) 


en 


c 




c3 


03 ^ 


t« 


i^s 




o3 


Q P^ 


Q fc^ 




W 



Tt< c-1 CO 

■^ »0 00 



— C<l'>^<MO00OC0C0»Ct^O000000C30O 

c^oo-^^ooc^t^ooccioecoo-^ooooT^os 



« IB 





i§ 


-S 








cS 




cS 


ce 








c 




C 


c 












m 


0^ 






o 




M 


M 


03 


OO 


o 


O 


8 


o 


O 


o 


<B 


_C 


S 


c 


o 


S 


"So 


'x 


's 


o 
















^ 


^*^ 


^^ 


>) 


>. 


>. 


>■. 


4^ 


.JJ 


+:j 


<D 


a> 


QJ 


« 


;4 


;-! 








c 


c 


cS 
« 


03 




rs 


2 


73 


-0 



o 



HH )-H hH k> ky* ls> Vv< 

HH HH H-t HH hH t-M HH 



EFFECTS ON THE ENDOGENOUS RESPIRATION 



179 



CO CO 

CO CO 
05 05 



2 o 



-H o 



pq ^ 



O -H O C CO CD O 



ce 2 



CO ? S 



QQ 


s 




O 


ti 




;i3 

GQ 


s 


cS 


2 


2 


S 


£2 


C3 


o 


o 


o 


CJ 
c8 


o 


o 


tl 


L. 


cS 


rt 


cc 


cc 


tc 


cS 


^ 


C 


TS 


» 


<D 


'^ 


OJ 


OQ 




0) 


C 
4) 





CO uo 73 
2 ^ eS 



IC lO lO -H ic 

CO CO CO lo CO 

C^ Oi O Oi Ot! 



03 


CD 


cS 


-O 


T3 




S 


S 


M 


o 


^ 


eg 


eg 


eg 


eg 


eg 


.2f 


-4-^ 


+2 


">> 


>> 


C 


C 


C 


'E 


"3 


^ 


O 


O 


eg 


C 


C 


fQ 


^ 


h-1 


pq 


m 


H 


<1 


<: 



~ ,2 









tf 



eg 



eg 
H 



<» 03 



eg eg 

>■. >. 

O O 

CO 03 



^O^^'C— " C C O 



K pq <5 ffi ffi 



H P^ d ^ ^ ^ 



lO O lO o o o o 
C<I ^ O O <M Oq 



•<* Tjt CO Tt< (M -^ 05 

i> t^ t^ t> i> r- CO 



C eo U 

O <0 .-, -^ -^ 



Q a> .-. .2 



2 ^ 



_^ O K 03 ^ 

"^ '^ >■■>>>■. 

^ ^ QJ 2 05 

^ fl^ rS -S -S 

> > T3 ^ T3 

;j ;3 S 5 i5 



O 03 "3 .X ;:; O 



o -d .ii -X 



^ S ^ =« 



^ -35 Q =? ?? 






s :g 



05 fe M <3 fiH 



P eg 



PM 



"B 


K 


.4J 


& 


-2 


<B 


X 


-j3 




-^ 


-bS 







S-i 


C 


a 



£ S 



y. H 



li T3 OJ 



« s 



T3 



r" ^, ■" 
S <5^ 3 



^- O J2 



bC 


c 





eg 


T3 


•+i 




03 


5 




_>■. 





3 


c 


t-i 




-t^J 


_c 






03 






4^ 


c 



3 
,0 






1 


4i 
03 


ft 


03 



« C 05 



c 


eg 


-Q 


'g 


^ 


r^ 


05 


eg 


05 

> 


-t^ 




05 


S 


'Si 


5 


<» 
c;> 


<» 
45 




_3 


05 


3 



"3) 


> 
eg 

J3 


5£ 


oT 


^ 


03 


>> 



05 



eg 


^ 


c 

eg 
4-:> 


_C 


T3 


03 




45 


c 


8 


T5 


05 


05 


73 


r^ 


w 


c 



r-. JC> 05 03 
" 'S 05 S 
'^ •" -O 01 



180 1. MALONATE 

correct for this phenomenon is to determine the C^^02 formed from labeled 
malonate. 

(f) Nature of the cycle operation and the presence of alternate pathways. 
The inhibition of the oxygen uptake associated with the cycle will depend 
on a number of factors in addition to the inhibition of succinate dehydro- 
genase. The availability of a large pool of organic acids to form oxalacetate, 
or the presence of pathways from which oxalacetate may arise (e.g., by 
carboxylation of pyruvate, or from aspartate by transamination), will re- 
duce the inhibition of oxygen uptake from that which would be observed 
if all the oxalacetate had to be derived from the cycle. Other pathways 
for the metabolism of succinate may circumvent the block to some extent. 
These matters have been discussed in some detail (see pages 72-88). 

(g) Adaptive changes in the presence of malonate. Inhibition of the cycle 
may accelerate other pathways. The increased uptake and metabolism of 
glucose brought about by malonate have been noted in several types of 
cells, and such a phenomenon will tend to counteract the malonate inhi- 
bition on the oxygen uptake. Adaptive changes in enzyme concentrations 
probably are seldom important in short-term experiments but cannot be 
completely ignored in work with certain microorganisms. Inhibitions by 
malonate have occasionally been noted to decrease with time, and adaptive 
changes are the most obvious explanation. 

These and other more subtle factors determine the effect of malonate on 
the total oxygen uptake of a preparation, and it should be apparent that 
deductions based exclusively on the inhibitions of endogenous respiration 
are frequently untenable. A definite inhibition with a reasonable malonate 
concentration is more significant than a negative result, because there are 
many factors which can reduce or abolish the action of malonate even 
though the cycle is present and active. 

The Time Course of Malonate Inhibition 

The inhibition of respiration by malonate may occur fairly rapidly and 
remain constant, or it may increase slowly to a level at which it is main- 
tained, or it may gradually disappear, or it may vary in quite complex 
fashion with time. A slowly developing inhibition would not be unexpected 
with a substance which does not penetrate readily. The inhibition of suc- 
cinate dehydrogenase is essentially instantaneous so that an approximately 
linear increase in the inhibition to a constant level would imply that the 
rate of inhibition is determined by the penetration. On the other hand, 
secondary effects, such as would result from the depletion of ATP, may 
also contribute to a progressive inhibition. Greville (1936) noted that mal- 
onate does not immediately inhibit the respiration of rat diaphragm but 



EFFECTS ON THE ENDOGENOUS RESPIRATION 181 

that the inhibition developes over 1-2 hr. A very similar time course was 
observed in barley roots by Laties (1949 a). After the addition of 10 mM 
malonate, the respiration drops linearly for 60-90 min and then becomes 
relatively constant at about 40%. However, over the next 5 hr, the inhi- 
bition lessens somewhat. This was postulated to be due to increasing suc- 
cinate concentration, but actually such changes in succinate take place 
much more rapidly in most cases. It must be admitted that in spinach 
leaves the maximal succinate accumulation occurs at 4 hr (Laties, 1949 b), 
so that this explanation can by no means be eliminated. The inhibition 
of the spinach leaf respiration by malonate is greater at 6 hr than at 3 hr 
but not enough data are available to correlate the changes in inhibition 
with succinate levels. Malonate requires about 1 hr to produce its maximal 
inhibition of sea-urchin egg homogenate respiration (Yeas, 1950). This is 
the only report of such a slowly developing inhibition in subcellular pre- 
parations and no explanation is evident. Another situation seems to exist 
in bovine kidney culture cells, where the inhibition by 100 mM malonate 
slowly increases from 23% at 1 hr to 35% at 5 hr (Polatnick and Bachrach, 
1960). It is more likely here that secondary changes are responsible. 

Definite decrease in the respiratory inhibition of pigeon brain dispersions 
(probably similar to homogenates) with time was reported by Banga et al. 
(1939). The inhibition is 41.7% at 10 min, 29.5% at 30 min, and 20.5% 
at 50 min. Since the malonate concentration was 24 mM, it seems unlikely 
that metabolism of malonate could have reduced its concentration signifi- 
cantly. Adaptive changes in homogenates are improbable and sufficient 
accumulation of succinate from the relatively limited substrate supply is 
scarcely possible. Besides, the inhibition of pyruvate oxidation increased 
over this interval. Sometimes changes in tissue metabolism occur during 
the course of an experiment independent! j^ of malonate action. When mal- 
onate is added to human brain slices immediately, the endogenous respira- 
tion is inhibited around 25% at 5 mM, but if malonate is added after 
90 min incubation of the slices, there is no inhibition (Elliott and Suther- 
land, 1952). The role of succinate oxidase in the respiration must change 
as a result of the slicing or the abnormal medium. 

The inhibition of the respiration of rat ventricle slices by malonate 
5-20 mM follows a more complex course (Webb et al., 1949). Following 
an initial inhibition, the respiration rises for approximately 1 hr and then 
begins to fall again. There is thus a maximum or hump in the respiration 
curve. After 1 hr, the respiratory level with malonate is higher than in 
the controls. The reversal of the inhibition is inhibited by fluoride, which 
would indicate that the hump is due to augmented glucose oxidation or to 
the metabolism of malonate. Calcium is also necessary for the typical re- 
sponse to malonate. Very complex effects of malonate were also found by 
Turner and Hanly (1947) and Hanly et al. (1952) in carrot slices. The var- 



182 1. MALONATE 

iation of the inhibition depends on the pH. At pH 4, the inhibition de- 
velops over 1 hr and remains constant, but at higher pH's the inhibition 
may disappear or only stimulation may be seen. The value of these interest- 
ing experiments is greatly reduced by the unaccountable use of potassium 
malonate rather than the sodium salt. Potassium at 50 mM (which is the 
concentration of malonate generally used by them) stimulates the respira- 
tion and alters its character, so that aU the results must be the summation 
of two usually opposing actions. This illustrates how the incorrect choice 
of an inhibitor salt can vitiate the results of an otherwise excellent in- 
vestigation. 

Most of the work on the malonate inhibition of endogenous respiration 
has been done without regard for possible alterations in the inhibition 
with time. The inhibitions have simply been determined over an arbitrary 
interval. Inasmuch as changes in the inhibition by malonate occur quite 
frequently, it is likely that over all inhibitions, such as are presented in 
Table 1-26, are often mean values and do not reflect either the initial inhi- 
bition or the maximal inhibition. As was pointed out in Chapter 1-12, the 
value of many studies on inhibitors would be increased by determinations 
of the variation of the inhibitions with time. 

Effects of Different Conditions on the Inhibition of Endogenous Respiration 

One of the most important variables affecting the response of the endog- 
enous respiration to malonate is the age of the tissue, particularly as it 
relates to the stage of development or the interval between the preparation 
of the tissue and the experimental testing. The respiration of plant tissues 
usually becomes more sensitive to malonate with time. This indicates a 
progressive change in the metabolic pattern in the direction of a greater 
participation of the cycle. The changes in the inhibition during malonate 
inhibition, discussed in the previous section, can be due to the effects of 
the malonate or to inherent metabolic alterations. It is thus important in 
such studies to determine both the changing inhibition in the presence of 
malonate and the changing susceptibility as malonate is added at various 
intervals. The malonate inhibition rises with time in the Avena coleoptile 
(Bonner, 1948), rose petals (Siegelman et al., 1958), chicory root slices (La- 
ties, 1959 a), Arum spadix slices (Simon, 1959), and potato tuber slices 
(Romberger and Norton, 1961). These changes are usually associated with 
an increase in the total uninhibited respiration. For example, in potato 
slices there is a 4-fold rise in the respiration during incubation for 30 hr; 
the malonate resistant fraction doubles and the malonate-sensitive fraction 
increases 10- to 15-fold. Carrot slice inhibition by malonate, on the other 
hand, decreases steadily up to 376 hr after cutting the sections (Hanly 
et al., 1952), although the uninhibited respiration first rises and then falls. 
The results obtained with animal tissues are less striking and more variable. 



EFFECTS ON THE ENDOGENOUS RESPIRATION 183 

The inhibition of rabbit ova respiration at various times postcoitum does 
not change significantly (Fridhandler et al., 1957), while the inhibition of 
trematode respiration decreases with time from excystment (Vernberg and 
Hunter, 1960). When 20 raM malonate is added to rat ventricle slices 1 hr 
after slicing, the inhibition of the respiration is only 25%, whereas initially 
the inhibition is near 50% (Webb et al., 1949), and this relationship holds 
for all malonate concentrations up to 100 vaM. We have seen that in the 
presence of malonate the inhibition has been replaced by stimulation at 
1 hr. Therefore the metabolism changes differently during the action of 
malonate and the maxima in the time curves cannot be explained by 
inherent alterations of the metabolic pattern. All of these results point to 
the importance of considering the time factor in studies of malonate inhi- 
bition. 

Another apparently important factor is the ion and buffer composition of 
the medium, although no thorough studies have been done and the mecha- 
nisms are not understood. Many years ago Annau (1935) observed that the 
inhibition of the respiration of both rabbit liver and kidney slices is less 
in Ringer than in phosphate medium. The results were quite variable but 
on the average the inhibition is 30% in phosphate and 15-20% in Ringer 
medium. Unfortunately, Annau did not state what form of malonate was 
used nor did he mention pH control, so the results are perhaps unreliable. 
The presence of bicarbonate abolishes the inhibition by malonate in ox 
retina homogenates (Burgess et al., 1960), and it is probable that inhibition 
in intact cells would also vary with the bicarbonate concentration. Bicarbon- 
ate can, of course, facilitate the formation of oxalacetate through carboxyla- 
tion reactions. When Ca++, Mg++, or K"*" is removed from the medium, the 
inhibition of rat brain slice respiration by 10 mM malonate is not altered, 
but in the presence of fumarate the inhibition becomes progressively greater 
as these ions are successively removed (Greville, 1936). The addition of 
Ca+"'" to nematode minces increases both the respiratory inhibition and the 
inhibition of succinate oxidation (Massey and Rogers, 1950). The sensitivity 
of chicory root slice respiration to malonate is markedly affected by K+ and 
Li+ (Laties, 1959 b). Slices incubated with 50 milf K+ are inhibited more 
and with 50 mM Li+ inhibited less than the fresh slices. It is also interesting 
that increase in CO2 tension results in progressive disappearance of the mal- 
onate inhibition, whereas increase in Og tension augments the malonate- 
sensitive fraction of the respiration. It is thus clear that the medium can 
play an important role in the response to malonate. Much work has been 
done in quite nonphysiological media and the results are thus difficult to 
apply to the actions of malonate in situ. Much more effort should be di- 
rected at creating approximately physiological conditions. 

The functional activity of the tissue determines the level and type of 
respiration, and therefore is often a major factor in the sensitivity to mal- 



184 1. MALONATE 

onate. This has been shown particularly clearly in brain slices, stimulated 
both electrically (Heald, 1953) and by K+ (Kimura and Niwa, 1953; Yoshida 
and Quastel, 1962). The stimulated respiration is readily inhibited by mal- 
onate (Fig. 1-14) whereas the resting respiration is insensitive. This be- 
havior is probably exhibited by many tissues. It is often very difficult to 
determine exactly the functional state of isolated tissues, such as slices, 
but where possible this should be attempted. We shall find later that 
active tissues are more easily functionally depressed by malonate and 
this may have a metabolic basis. Indoleacetate stimulates the growth of 
Avena coleoptiles and increases the respiration simultaneously. This ad- 
ditional respiration brought about by indoleacetate is readily inhibited by 
malonate (Bonner, 1949), and it is likely that the respiration of rapidly 
growing tissue is generally inhibited more strongly by malonate than that 
of resting or slowly proliferating tissue. 

Consideration must also be given to the history of the tissue. Bonner (1948) 
has shown that the nutritional state of the Avena coleoptile determines the 
inhibition by malonate, and it is probable that the same applies to animal 
tissues. The inhibition of wheat seedling respiration by malonate depends 
on a number of factors, including the type and duration of irradiation, 
the nutrition, and the region from which the plants come (Farkas et al., 
1957 a, b). This is a field that has been very little explored. The changes 
in the respiratory inhibition of animal tissues with nutrition might not only 
provide information on the metabolic patterns under various conditions, 
but be important in the use of the inhibitor to selectively depress the me- 
tabolism and growth of neoplastic tissues. 

Effects on the Respiratory Quotient 

The effects of an inhibitor on the respiratory quotient (R.Q. = COg form- 
ed/Og uptake) are often indicative of shifts in metabolic pathways. Let us 
first consider the theoretical values of the R.Q. for the metabolism of 
various substrates (see tabulation) in the presence and absence of malonate, 
assuming that malonate is able to block succinate oxidation completely. 
Cases in which oxalacetate is formed in the cycle and from noncycle sources 
must be separated. Summarizing these results, one would expect malonate 
to increase or decrease the R.Q., depending on the substrate and the nature 
of the cycle operation. Since a complete block of succinate oxidation would 
prevent the formation of oxalacetate through the cycle, malonate may 
shift the pathway from cycle oxalacetate to externally formed oxalacetate, 
if the latter reaction is possible. If this is so, the R.Q. should rise in every 
case. 

This prediction is quite consistently borne out experimentally. The R.Q. 
of rat liver slices rises from 0.72 to 0.77 in the presence of 20 mM malonate, 
at which concentration the respiration is inhibited 14% (Elliott and Greig, 



EFFECTS ON THE ENDOGENOUS RESPIRATION 185 



Substrate metabolism R.Q. 



Glucose -> CO2 + H2O 1.0 

Glucose -> succinate -f CO2 + H^O 0.8 

Glucose + 2 oxalacetate ->• CO2 + HgO 1.27 

Glucose + 2 oxalacetate -> 2 succinate + CO2 + H2O 1.5 

Pyruvate -> COj + HjO 1.2 

2 PjTuvate -> succinate + CO2 + H2O 1 . 33 

Pyruvate + oxalacetate -> COj + H2O 1.4 

Pyruvate + oxalacetate ->■ succinate + COj + O2 2.0 

ButjTate -^ CO2 + H2O 0.8 

ButjTate -r oxalacetate -> COj + HjO 1 . 23 

Butyrate + oxalacetate -> succinate + CO2 + HjO 1.0 

Butyrate + 2 oxalacetate -> CO2 + HjO 1.2 

But>Tate + 2 oxalacetate -> 2 succinate + CO2 + H2O 1.33 



1937). However, malonate decreases the R.Q. of kidney slices, both en- 
dogenous and with pyruvate as the substrate. Malonate elevates the R.Q. 
of rat adipose tissue from 1.0 to 1.13 endogenously and from 1.14 to 1.41 
in the presence of glucose (Haugaard and Marsh, 1952). In frog muscle, 
the R.Q. first rises from 0.9 to 0.97 at 10 mM malonate, but then progres- 
sively decreases as the malonate concentration is raised so that at 200 mM 
malonate the R.Q. is 0.39 (Thunberg, 1909). In plant tissues, the effects 
are less variable. Malonate has been shown to increase the R.Q. of barley 
roots from 0.97 to 1.14 (Machlis, 1944), of maize roots (Beevers, 1952), 
of carrot roots up to values as high as 3 (Hanly et al., 1952), of chicory 
roots from 1.03 to 1.14 (Laties, 1959 a), and of rhubarb leaves at pH 5.3 
(Morrison, 1950). 

Of course, there are many factors which must be taken into account, 
since malonate can secondarily alter several metabolic pathways. A stimula- 
tion of glucose uptake could change the R.Q. in either direction, depending 
on the nature of the substrates used in the uninhibited tissue; in the pre- 
sence of a significant cycle block, this would usually depress the R.Q. and 
counteract the more direct effects described above. On the other hand, 
metabolism of malonate would tend to elevate the R.Q. since the complete 
oxidation would give R.Q.'s of 1.50-1.55 and the oxidation to succinate 
3.0-4.0. A final factor of importance is the relative dependence of glucose 
and fatty acid metabolism on the operation of the cycle and the levels of 
ATP, since malonate could alter the oxidative contribution from these sub- 
strates secondarily. 



186 1. MALONATE 

Significance of Respiratory Inhibition 

Does the degree of malonate inhibition indicate the contribution of the 
cycle to the total oxygen uptake? This must certainly be answered in the 
negative. Lack of inhibition can be due to a failure to penetrate, the me- 
tabolism of malonate, a source of oxalacetate external to the cycle, met- 
abolic adaptations of the cells, and many other factors. Positive evidence 
of inhibition is more valuable than absence of inhibition, but even when 
definite inhibition is observed the possibility of actions other than in the 
cycle must be considered, especially when the malonate concentration must 
be high to achieve an effect. It is doubtful if anyone examining Table 1-26 
would attempt to correlate the inhibitions with the importance of the 
cycle in the organisms and tissues. For example, in general there is greater 
inhibition of mammalian endogenous respiration than of the respiration of 
microorganisms or plants. This might indicate a greater role of the cycle 
in mammalian tissues, but it could also be attributed to a poorer penetration 
in the plants and microorganisms, or to a greater metabolic flexibility and 
adaptability in these more resistant forms. It must also be clear that the 
degree of respiratory inhibition bears no necessary relationship to the de- 
gree of inhibition of succinate dehydrogenase. A significant inhibition by a 
reasonable concentration of malonate is evidence for the operation of the 
cycle, but the quantitative aspects of the contribution cannot be derived 
from these data alone. The effects of malonate on the endogenous respira- 
tion are sometimes of greater physiological significance than effects on the 
oxygen uptake in the presence of high concentrations of often abnormal 
substrates, since the endogenous metabolism may be representative of a 
more normal balance of substrates. In this connection, studies of inhibitors 
would often be improved if the attempt were made to provide the cells 
with a mixture of physiologically pertinent substrates at the concentrations 
normally occurring in the cellular environment. 

PERMEABILITY OF CELLS TO MALONATE 

One of the major problems in the use of malonate has always been the 
degree of penetration of the inhibitor into the cells or tissues, and it has 
been frequently stated that this is the primary factor responsible for the 
low inhibitions observed in many cases. It is true that the plasma mem- 
brane is relatively impermeable to most ions, particularly anions and those 
carrying two or more charges, but if this is so how can one explain the 
marked respiratory stimulations usually seen with succinate or other di- 
carboxylate ions? Furthermore, malonate is often metabolized readily by 
tissues and this presupposes entrance into the cells. Since there are other 
possible reasons for a resistance to malonate, the permeability hypothesis 
must be examined critically. 



PERMEABILITY OF CELLS TO MALONATE 187 

Experimental Evidence Relating to the Penetration of Malonate 

Malonic acid is more than 10 times as lethal on injection into frogs as 
is sodium malonate (Heymans, 1889). No explanation was offered for this 
observation but it could have been due to the greater permeability of the 
cells to the acid or, on the other hand, to a nonspecific acidification of the 
animals. Malonate administered to rabbits circulates initially in a volume 
equivalent to the extracellular compartment and the intracellular transfer 
occurs slowly (Wick et al., 1956). The failure of malonate to alter the me- 
tabolism of labeled acetate was attributed to both the slow penetration into 
the tissues and the simultaneous metabolism of the malonate, both factors 
keeping the intracellular concentration at low levels. The inability of mal- 
onate to alter gastric acid secretion in frogs, even at lethal doses, was sim- 
ilarly attributed to these factors (Davenport and Chavre, 1956). Inasmuch 
as succinate oxidase is present in the secretory cells and the cycle is im- 
portant in secretion (as shown by the inhibition with fluoroacetate), the 
lack of action must be due to an insufficient concentration within the cells. 

Turning to isolated tissues and cell suspensions, the augmentation of 
malonate effects by procedures designed to reduce or abolish the perme- 
ability barriers has been demonstrated many times (Table 1-26). Rat dia- 
phragm respiration is inhibited slowly by malonate, but if the diaphragm 
is cut into small pieces, and hence presumably damaged, the inhibition is 
immediate (Greville, 1936). The oxygen uptake of pigeon brain brei is 
inhibited rather poorly by 24 mM malonate, but when the brain is dispersed 
more completely in the form of a homogenate the inhibition is more marked 
(Banga et al., 1939). The succinate dehydrogenase of yeast incubated for 
5 hr in liquid nitrogen is much more susceptible to malonate than in normal 
cells (Lynen, 1943), and the same holds for E. coli treated with toluene 
(Ajl and Werkman, 1948). Sensitivity to malonate can be induced by liquid 
nitrogen treatment in the fungus Zygorrhynchus (Moses, 1955) and by dry- 
ing Pseudomonas (Gray, 1952). Malonate does not inhibit glutamate oxida- 
tion in intact cells of Pasteurella, but inhibits well in sonic lysates (Kann 
and Mills, 1955). All of these phenomena have been interpreted in terms of 
permeability. This is certainly the most obvious explanation and it is prob- 
ably generally correct, but it must be admitted that such drastic treat- 
ments could affect many other things; for example, alter the organized 
enzyme structure so that the attacked enzyme is more exposed, or reduce 
the ability of the cells to metabolize malonate. 

Only one investigation of the relative permeabilities of the dicarboxy- 
late anions has been made. Giebel and Passow (1960) determined the half- 
times for penetration of these ions into bovine erythrocytes and the results 
are given in Table 1-27. Giebel and Passow attempted to correlate the per- 
meabilities with the ionic sizes and the acidic ionization constants. The 
ionic volumes and lengths presented in the table, which are somewhat 



188 



1. MALONATE 



Table 1-27 

Erythrocyte Permeabilities and Molecular Properties of Dicarboxylate 

Anions " 



Anion 



Relative 
permeability 



Ionic 
volume 



Ionic 
length 

(A) 



(H,B) 



(HB-) 



Oxalate 


100 


54 


5.1 


0.00015 


276 


Malonate 


17 


66 


7.5 


0.112 


6590 


Maleate 


14 


75 


7.9 


0.051 


252 


Fumarate 


1.4 


75 


7.6 


0.014 


445 


Succinate 


0.45 


78 


7.5 


2.56 


6410 


Malate 


0.31 


91 


7.5 


0.135 


1860 


Glutarate 


0.085 


90 


8.5 


1.99 


3780 


Tartrate 


0.045 


104 


7.5 


0.0103 


338 



" The permeabilities were determined in bovine erythrocytes by Giebel and Passow 
(1960). The values given here for the relative permeabiHties are the reciprocals of the 
entrance rate half-times (ti/^) multipHed by 100. Tlie calculations of the ionic volumes 
and lengths are approximate and usually depend on the configuration of the ions. 
The concentrations of HjB and HB are given for a total concentration of 1 M. 



different than those given by Giebel and Passow, must be considered as 
only relative values, neglecting hydration and special configurations of the 
ions. Their experiments were run at a pH of 7.35 so the concentrations of 
the undissociated and singly dissociated forms of the acids are given for 
this pH. The ionization constants are sometimes quite different from those 
assumed by Giebel and Passow and are those in Table 1-2. There is certainly 
little or no general correlation between permeabilities and the concentra- 
tions of either H2B or HB". Succinate, for example, penetrates one-thirty 
eighth as rapidly as malonate and yet (H2B) for succinate is 23 times higher 
than for malonate. This does not necessarily invalidate the assumption 
that for a single substance the unionized forms penetrate more rapidly than 
the ionized, but it shows that there are other factors which are quite im- 
portant. There is also no correlation with the ionic length and it is unlikely 
that one would be expected. However, there is some correlation with ionic 
volume, leading Giebel and Passow to suggest that the dicarboxylate anions 
penetrate through pores in the membrane, whereas the monocarboxylates 
pass through the lipid phase of the membrane. They calculate the pore 
radius to be between 3.8 and 4.5 A. If these ions pass through the pore 
channels in the extended form, which is likely, there are two major factors 
which may contribute to the permeability: the cross-sectional area perpen- 



PERMEABILITY OF CELLS TO MALONATE 189 

dicular to the direction of passage and the degree of interaction of the 
molecules with the walls of the pores. This interaction may be of various 
types and includes steric repulsion and attractive forces (such as van der 
Waals' forces or hydrogen bonds). The configuration of the ion must be 
important in this connection. Maleate penetrates 10 times faster than fu- 
marate and this must be mainly due to the structure of maleate wherein 
the carboxylate groups are much closer than in fumarate. That both of 
these ions penetrate more rapidly than succinate may be due to the greater 
rigidity of the former, energy perhaps being required for succinate to change 
from its statistically most probable configuration to that necessary for 
penetration. It is evident, however, that none of these explanations satis- 
factorily fits the experimental data and that we need to know much more 
about the membrane before accurate interpretations can be made. It should 
be mentioned that these results on erythrocytes do not apply to other 
types of cells or tissues, inasmuch as erythrocyte permeability is in some 
senses unique. 

Malonate inhibits the succinate dehydrogenase of calf thymus nuclei and 
yet at 10 mM has no effect on the respiration or ATP level of intact nuclei 
(McEwen et al., 1963 b). This indicated a failure to penetrate and it was 
shown with labeled malonate that this is indeed the case, which is some- 
what surprising in view of the usual concepts of the nuclear membrane. 

Malonic acid does not enter organic solvents from water readily, due 
probably to the dipolar nature of the unionized carboxyl groups. The par- 
tition ratios for malonic acid are given as (concentration in solvent/concen- 
tration in water): oleyl alcohol 0.049 (Collander, 1951), ether 0.083, iso- 
butanol 0.62, methylisobutylketone 0.15, and methylisobutylcarbinol 0.37 
(Pearson and Levine, 1952). The partition ratios for succinic acid are always 
somewhat higher, as expected. These data would indicate that even the 
unionized malonic acid would not penetrate through the lipid phase of the 
membrane too readily. The fact that the un-ionized forms penetrate better 
than the ionized does not imply that passage through a lipid phase occurs. 
The negative charge on the ions could impede movement through pores, 
especially when it is considered that in most cells these ions must move 
up an electrical potential gradient to cross the membrane. 

Variation of Malonate Inhibition with pH 

One of the strongest arguments for the preferential uptake of the less 
ionized forms of malonic acid is the rise in the inhibition observed with a 
lowering of the pH. This has been examined particularly in plant tissues 
and the results are quite uniform. Such effects have been observed in to- 
mato stem slices (Link et al., 1952), maize roots (Beevers, 1952), rhubarb 
leaves (Morrison, 1950), barley roots (Laties, 1949 a), spinach leaves (Bon- 
ner and Wildman, 1946), carrot root slices (Hanly et al, 1952), and Avena 



190 1. MALONATE 

coleoptile (Cooil, 1952). These results are plotted in Fig. 1-14-19. On the 
other hand, rather insignificant effects of pH have been noted in fungi, 
such as Microsporum, Trichophyton, Epidermophyton (Chattaway et al., 
1956), and Pullularia (Clark and Wallace, 1958). In Pullularia malonate 
is readily oxidized; the oxygen uptake from malonate increases with a 
lowering of the pH along with the inhibition of the cycle and the effects 
tend to cancel one another. The effects of malonate in tobacco leaves are 
not changed greatly by lowering the pH from 7 to 4 (Vickery and Palmer, 
1957), although down to pH 5 the uptake of malonate becomes progressively 
greater. The incubation here was very long (48 hr) and hence there was 
more opportunity for malonate to penetrate than in shorter experiments. 
It is unfortunate that no quantitative work on pH has been done with 
animal tissues. 

It is also regrettable that in those instances in which malonate inhibits 
more strongly at low pH values the reversibility of the inhibition has not 
been adequately tested. Lowering the pH of the medium in the presence 
of a weak acid increases the amount of unionized acid entering the cells 
and can decrease the intracellular pH to a degree causing cell damage. It 
was noted in carrot root tissue that injury to the cells occurred, including 
loss of turgor and release of some of the cell contents, with malonate at a 
pH around 4 (Hanly et al., 1952). It is very difficult in experiments of this 
type to distinguish between a specific malonate effect and a nonspecific 
acid damage. Examination of the reversibility of the inhibitions might 
provide some evidence on this point. 

In Volume I it was shown that the intracellular concentration of a di- 
carboxylate anion is related to the total external concentration in the follow- 
ing manner (Eq. I-i4-178): 

(i=), = I J; (!,)„ (1-4) 

oJi \n. )'i 

where the subscripts o and i refer to outside and inside the cells and ^f-' 
is the appropriate pH function for the external inhibitor (see Eq. 1-14-12). 
Two assumptions are involved in this formulation: (1) only the Hgl form 
of the inhibitor penetrates, and (2) the cells are internally completely buf- 
fered. The variation of the intracellular active 1= form with external pH 
is very marked, as shown in the accompanying tabulation, assuming an 
intracellular pH, of 6.8. Of course, cells are not completely buffered and, 
as the internal pH, drops, the entrance of the inhibitor is slowed, so that 
with decrease in the buffering capacity the ratios given will be lessened. 
Here one is presented with the dilemma that at low pH values one must 
either assume a high internal inhibitor concentration or a significant fall 
in pH,. Since the inhibitions observed are not as high as would be predicted 
on the basis of the above equation and tabulation, one is forced to conclude 



PERMEABILITY OF CELLS TO MALONATE 191 

that the intracellular pH^ must fall. This may not only damage the met- 
abolic systems but wiU tend to reduce the inhibition on succinate dehydro- 
genase by decreasing the concentration of dicarboxylate anion. 



pHo 


a-)i/(it)o 


8.2 


0.0016 


7.4 


0.063 


7.0 


0.39 


6.0 


34.8 


5.0 


1610 



If one assumes that the HI~ form can also penetrate, the internal con- 
centration of the active 1= form will not be so strongly dependent on the 
external pH^, and the inhibition will increase significantly as the pH^ is 
lowered from 6 to 5, but otherwise the same behavior will be expected. It 
is usually dif&cult to distinguish between penetration by the Hgl form only 
and penetration by the HI" form also. Some arguments have arisen on this 
point, Bonner and Wildman (1946) believing that the HI~ form penetrates 
and Beevers (1952) holding that only the Hgl penetrates. A Simon-Beevers 
plot of the data from maize roots (see Chapter 1-14) shows that (Hgl) does 
not remain constant for 50% inhibition of respiration over the pH range 
3-6.5, which would indicate a possible contribution from the HI~ form. 
Actually, it might be better to express the total entry rate of malonate as: 

Entry rate = Phj(H2I)o + Phi-(HI-)o + Pi=(I=)<, (1-5) 

where the P's represent the permeabilities to the various forms of the inhi- 
bitor. Although Pjj I > Pri- > Pi-^ above pH 4 (HI~) is much greater 
than (Hoi) so that the contribution of the second term to the total rate may 
be significant. One would like to know the relative values of the P's for a 
particular tissue and these could be determined if the entry rate of mal- 
onate were determined (for example, with labeled malonate) at different 
pH values. One must also bear in mind that a change of pH could alter 
the permeability properties of the membrane. 

The only data on the effects of pH on malonate action in animal tissues 
were obtained on rat ventricle strips by Masuoka et at. (1952). Malonate 
stimulates the amplitude of hypodynamic strips much more at pH 6.2 
than at 7.4. This positive inotropic action may be unrelated to the inhi- 
bition of succinate dehydrogenase, since succinate at pH 6.2 gives essenti- 
ally the same response, and could depend on the oxidation of malonate 
(see page 216). Whatever the mechanism, these results indicate that mal- 
onate penetrates more readily at the lower pH. 



192 1. MALONATE 

In comparing the actions and penetrations of malonate with succinate, 
or with other dicarboxylate anions, it is necessary to consider that the 
relative concentrations of the ionic species can be quite different. Thus 
the HgB form of succinate is at a much higher concentration than the same 
form of malonate at the same total concentrations (Table 1-3). It is also 
possible that some cells possess active transport or carrier systems for the 
substrate dicarboxylate anions, allowing succinate to penetrate more readi- 
ly than malonate. Permeability in some cases seems to be as specific as 
are enzyme reactions, and can be quite dependent on the configurations 
and charge distributions of the transported substances. This could be due 
to either the structure of the membrane pores or the nature of a carrier. 
For these reasons one must anticipate striking differences between different 
tissues with regard to the relative permeabilities to the various ions, and 
it is particularly important not to apply the results on plants unreservedly 
to animal tissues or microorganisms. 



GROWTH, DEVELOPMENT, AND DIFFERENTIATION 

The responses of growth, cleavage, and histogenesis to inhibitors are 
interesting because they often demonstrate the nature of the metabolic 
basis for these important biological processes. The results may also have 
bearing on the possible use of the inhibitors for the selective depression 
of the growth of organisms or abnormal cells which are detrimental to the 
host. These processes all require energy from the metabolism so that any 
reduction of either the exergonic reactions or their coupled phosphoryla- 
tions would be expected to interfere in some manner. In addition, more 
specific effects may occasionally be observed. The selective inhibition of 
the growth of certain cells can result from different rates of growth of the 
cells involved, or from differences in the metabolic requirements for growth. 
It is generally true that rapidly proliferating cells are more readily affected 
by inhibitors than are the same or other cells growing or multiplying at 
slower rates. It has also been demonstrated that various types of cells may 
utilize different enzymes or metabolic pathways to support proliferation. 
With respect to malonate, one might anticipate that cells whose growth 
is in one way or another significantly dependent on the cycle would be 
inhibited more than cells not requiring the operation of the cycle. However, 
other factors, such as the degree of penetration of the malonate or the 
susceptibility of the succinate dehydrogenases to malonate, may be im- 
portant. 

Virus Multiplication 

Malonate is able in some instances to suppress the intracellular formation 
of virus without permanently damaging the host cells. The results obtained 



GROWTH, DEVELOPMENT, AND DIFFERENTIATION 



193 



with influenza type A virus isolated from man and cultured in chick chorio- 
allantoic membrane illustrate this well. Ackermann (1951) found that mal- 
onate inhibits the formation of virus and simultaneously reduces the oxygen 
uptake of the host cells (see accompanying tabulation). Malonate is not 



Malonate 


% Inhibition 


Final 


Infectivity 


(mi/) 


of respiration 


virus titer 


titer 


None 




213 


107.8 


20 


3 


192 


107.8 


40 


18 


69 


106.5 


60 


47 





103.3 



virucidal since the original virus can be recovered from the infected cul- 
tures. The major effect of malonate is not to prevent infection of the cells, 
since essentially the same results were obtained when malonate was added 
4 hr after the inoculation of the virus. The inhibition is thus exerted on 
the synthesis of new virus material. Furthermore, the host cells are not 
damaged; if infected chorioallantoic tissue is exposed to 60 mM malonate 
for 24 hr, virus proliferation is completely inhibited but, if the tissue is 
washed free of malonate, becomes susceptible to infection and supports 
virus multiplication. These results were confirmed and extended by Eaton 
(1952), who reported that 42 niM malonate completely inhibits virus mul- 
tiplication, inhibits respiration around 50%, and does not alter aerobic 
glycolysis. The formation of virus in minced chorioallantoic membrane is 
inhibited 85% by 27 mM malonate when glucose is the substrate and 95% 
when glutamate is the substrate (Eaton and Scala, 1957). The injection 
of 1 mg sodium malonate into chick embryos 1 hr after inoculation with 
virus reduces the amount of virus after 48 hr by 55% (Hannoun, 1952). 
The energy for the synthesis of new virus material must come mainly from 
the host cell tricarboxylate cycle. It is impossible to reduce the energy 
supply for virus synthesis without simultaneously restricting the energy 
for host cell activities. However, it appears that the virus propagation is 
selectively depressed because it is the most endergonic process occurring, 
the chorioallantoic cells otherwise not being very functionally active. As 
long as the low maintenance energy requirement is satisfied, the cells are 
not damaged (see Fig. 1-9-6). If one were dealing with virus in an active 
tissue, such as nerve or muscle, it would be much more difficult, or im- 
possible, to selectively block virus multiplication. 

Other viruses growing in animal cells have been studied with various 
results. Malonate at 6.7 mM has a definite depressant effect on the proli- 
feration of vaccinia virus in chick embryo cultures (R. L. Thompson, 1947) 



194 1. MALONATE 

and similar results were reported for psittacosis virus (Morgan, 1954). In 
neither case does malonate depress the growth of the tissue cultures, noi 
does it inhibit the cellular contractions of chick heart cultures. Very slight 
effects of malonate at concentrations from 10 to 100 mM were observed 
with foot-and-mouth disease virus in bovine kidney culture cells (Polatnick 
and Bachrach, 1960). Since there is little inhibition of the respiration, it is 
possible that malonate does not penetrate adequately. At 100 mM, respira- 
tion is inhibited 23%, virus yield perhaps 20%, and there is a 10 min delay 
in the appearance of virus. Finally, malonate was found to actually stimu- 
late feline pneumonitis virus proliferation in the isolated chick embryo yolk 
sac, 10 mM increasing the virus titer about 33% (Moulder et al., 1953). 
Thus a wide variety of effects have been observed with different viruses 
and host cells, no general conclusions being possible at this time. I am not 
aware of any studies on the effects of malonate on the course of virus in- 
fections in whole animals. 

Plant viruses have been inadequately investigated and results on the 
tobacco mosaic virus only are available. Although malonate at 0.5 mM de- 
creases the number of lesions/cm^ in detached tobacco leaves from 28.9 
to 21.9, Chiba et al. (1953) felt that this result is statistically insignificant. 
Ryzhkov and Marchenko (1954, 1955) reported that malonate inhibits 
multiplication of this virus and that this is reversed by fumarate, but 
Schlegel (1957) found only variable effects of 3 mM malonate on the yield 
of virus in leaf discs. The spraying of 10 mM malonic acid solutions (pH 
2.7-3.6) onto the leaves of bean plants decreases the number of virus lesions 
68% without leaf damage (Matthews and Proctor, 1956), but this may be 
unrelated to the action of malonate on the cycle inasmuch as succinic acid 
is even more inhibitory. This is probably a nonspecific acid effect because 
the cycle intermediates usually increase the virus yield when they are added 
to leaf cultures. 

It is somewhat surprising that malonate has no demonstrable effect on 
the multiplication of E. coli phage. Spizizen (1943) found no effect at 10 mM 
under any conditions of virus growth, and Czekalowski (1952) reported no 
actions on either T2 phage or host cells at concentrations from 0.1 to 100 mM. 
It may well be that phage proliferation is not directly dependent on the 
energy derived from the cycle, but inhibition by 2,4-dinitrophenol, cyanide, 
and fluoride is observed. In fact, Czekalowski stated that all the inhibitors 
that depress phage selectively seem to act in some manner on the cyto- 
chrome system and are able to inhibit succinate oxidase; yet the most 
specific inhibitor for this enzyme is inactive. Lack of penetration is an 
unlikely explanation and this failure of malonate to inhibit deserves further 
study. 



GROWTH, DEVELOPMENT, AND DIFFERENTIATION 195 

Multiplication of Bacteria, Fungi, and Other Microorganisms 

Bacterial growth is apparently fairly resistant to malonate, despite the 
many observations of enzyme and metabolic inhibitions in these organisms. 
The absence of any effect on E. coli at malonate concentrations up to 100 mM 
reported by Czekalowski was mentioned above, but Loveless et al. (1954) 
found 50% inhibition of growth at 19.2 ml/ malonate, with no effect on cell 
size. Malonate at 300 roM produces somewhat elongated E. coli cells and 
at 800 mM they are markedly lengthened and often U-shaped: division is 
abolished but the cells continue to elongate slowly (Schweisfurth and 
Schwartz, 1959). The effects of malonate must depend on many factors, 
including the nutrient medium, the duration of the growth phase studied, 
and the pH. Although Rosenberg (1948) found the growth of Clostridium 
saccharobutyricum to be inhibited by malonate, a concentration of 100 mM 
was used, so that the mechanism of the effect is not clear. His observation 
that the inhibition is overcome by meso-inositol and borate must be inter- 
preted as indicating a unique approach to the study of malonate inhibition. 
Malonate at 10 mM has a slightly depressant action on rate of germination 
of Bacillus subtilis spores but does not affect growth in culture (Hachisuka 
et al., 1955). The bacteriostatic concentrations of malonate were given as 
3.2 mM for Pseudomonas fluorescens and 7.7 mM for B. aerogenes (Cooper 
and Goddard, 1957) but the acid was used, the pH being 2.5 and 1.5, re- 
spectively, so that a nonspecific acid effect is the most likely explanation, 
especially as succinic acid is as inhibitory. It is clear that the investigations 
of the effects of malonate on bacterial growth leave everything to be desired. 

The sporulation and growth of yeast are inhibited by malonate quite 
potently and in parallel fashion, 50% depression of each occurring at 5 mM 
(Miller and Halpern, 1956). On the other hand, Hensenula ellipsoidospora, 
a vellum-forming yeast, grows more rapidly in the presence of malonate 
(Luteraan, 1953). It would certainly not be too surprising to find certain 
microorganisms stimulated by malonate inasmuch as many can oxidize 
malonate readily (page 228). Malonate arrests sporulation of Aspergillus 
niger without suppressing growth (Behal, 1959) but germination of Neuro- 
spora ascospores is not blocked by 10 mM malonate even at pH 2.3 and 
after 24 hr (Sussman et al., 1958), nor is the germination of Puccinia uredo- 
spores affected significantly by 20 mM malonate at pH 4.8, although the 
respiration is inhibited around 37% (Farkas and Ledingham, 1959). The 
formation of conidia is often essential for the spread of the scab disease 
of apple caused by Venturia inaequalis and so Kirkham and Flood (1963) 
investigated the effects of various respiratory inhibitors on ascosporulation. 
Malonate was found to inhibit at high concentrations, the inhibition sur- 
prisingly increasing with increase in the pH (see accompanying tabulation). 
This might imply an action on or within the membrane; this is supported 
by the relative lack of effect on the respiration and the rather potent inhi- 



196 1. MALONATE 

bition produced by ^ra ws-aconitate. The injection of 50 milf malonate into 
the leaf petioles, however, increases infectivity so that a more significant 



Malonate 
(mil/) 


Initial pH 


0/ 

/o 


Inhibition of 
sporulation 


50 


4.2 




4 


100 






39 


40 


5.0 




Stim 7 


100 






71 


50 


6.2 




68 


100 






96 



effect on the host tissue is evident. The development of Puccinia rust on 
wheat seedling leaves is inhibited by 10 mM malonate, but the leaf tissue 
is damaged (Samborski and Forsyth, 1960). In this particular case the 
phytotoxicity is greater than the rust suppression so that malonate could 
not be used commercially. Malonate esters have been tested for inhibition 
of mold growth in syrups, but 13 mM does not have much effect over 144 hr 
(Lord and Husa, 1954); however, these esters are used as fungistatic agents 
in soy sauce in Japan (Tsukamoto, 1951). Another instance of growth stim- 
ulation by malonate was reported for Endamoeba histolytica (Nakamura 
and Baker, 1956); the average cell count per field at the end of 3-4 days 
was 1 in the control and 16 in the presence of 12.8 mM malonate, indicating 
possible metabolism of the malonate by these organisms. 

Plant Growth 

Avena coleoptile growth is sometimes stimulated and sometimes depressed 
by malonate, the response depending on the strain of oats used, the pH, 
and whether the sodium or potassium salt of malonate is used. The marked 
inhibition (61%) reported by Commoner and Thimann (1941) for 10 mM po- 
tassium malonate over 24 hr has not been observed by others. Albaum and 
Eichel (1943) found only stimulation (around 30%) from 1-5 niM potassium 
malonate over a period of 160 hr, and it was felt that malonate was serving 
as a substrate, which was substantiated by the higher respiration in the 
presence of malonate. Thimann and Bonner (1948) provided further evi- 
dence for this by the finding that malonate at 1 mM, having little effect by 
itself, antagonizes the marked inhibition produced by iodoacetate. How 
much of this is due to malonate and how much to potassium is difficult 
to say. Cooil (1952) confirmed the counteraction of iodoacetate inhibition 
by potassium malonate, but found that the sodium salt is not nearly as 
potent, implicating the potassium ion. The failure of malonate to inhibit 
the growth is probably the result of poor penetration, as shown by the 



GROWTH, DEVELOPMENT, AND DIFFERENTIATION 197 

effects of pH (see accompanying tabulation); the pH alone has little effect 
on growth. The malonate inhibition is satisfactorily reversed by fumarate. 



pH Mean growth (mm) in malonate 3 mM 

6.5 1.56 

6.0 1.70 

5.5 1.55 

5.0 0.22 

4.5 0.16 

The mitotic activity of the excised roots of the garden pea {Pisum sativum) 
is stimulated by glucose. This is very strongly blocked by malonate at pH 
5.5. A concentration of 0.01 mM delays the action of glucose 2 hr but does 
not inhibit mitosis; 0.1 mM inhibits mitotic activity around 50%; 0.5 milf 
almost completely inhibits mi^-^ses; and 1 mi!f not only inhibits completely 
but produces some toxic effects (Wilson et al., 1959). It was postulated 
that the initiation of mitosis is dependent on the cycle, since once mitosis 
began it proceeded to telophase normally. The progression through mitosis 
may be coupled with a more anaerobic type of metabolism. The growth 
and cell proliferation in tissue cultures of the crown galls of various plants 
(marigold, Paris daisy, periwinkle, and sunflower) are inhibited to different 
degrees by malonate. Cultures from normal tobacco stem are inhibited 
similarly. At 10 mM, the following inhibitions may be estimated from the 
curves given by Hildebrandt et al. (1954): sunflower 0%, marigold 29%, 
tobacco 40%, Paris daisy 60%, and periwinkle 67%. At 80 mM malonate, 
all are inhibited completely. A question arises again as to whether these 
effects are related to cycle inhibition, since succinate, pyruvate, acetate, 
and other organic anions inhibit also. The pH was 6.0 so that acid effects 
should not be important. 

Egg Cleavage and Embryogenesis 

The best and most interesting work on the growth responses to malonate 
has been done with marine invertebrate eggs and embryos. Since this work 
was done in sea water, we must bear in mind that the concentration of 
free malonate is much less than the total concentration due to the high 
amounts of Ca++ and Mg+"'". When malonate is added at a total concen- 
tration of 25 mM, it is likely that the free malonate is around 4 mM (see 
Table 1-5). In addition, the pH of sea water is near 8.2 and this is unfavor- 
able to malonate penetration into the cells. Considering these factors, it is 
surprising that such definite and characteristic effects of malonate have 
been observed. 



198 1. MALONATE 

Egg cleavage is generally rather sensitive to malonate. Although Arbacia 
eggs divide normally in 1 mM malonate (Krahl and Clowes, 1940), Dend- 
raster eggs are inhibited quite well at this concentration (Pease, 1941). The 
development of bilaterality in Dendraster, seen with many inhibitors, does 
not occur with malonate even at 10 mM, demonstrating a true differential 
effect on cleavage. Division of Arbacia and Chaetopterus eggs is inhibited 
completely by malonate at 70 mM, although 40 mM is essentially without 
action, and this is completely reversed by fumarate, indicating a block of 
the cycle (Brust and Barnett, 1952; Barnett, 1953). This is not a sodium 
effect since NaCl does not inhibit. Such high concentrations of malonate 
are not unreasonable in work in sea water and the inhibitions here may be 
quite specific. Egg cleavage seems to depend on the ATP generated in the 
cycle, because concentrations of malonate that completely inhibit the cleav- 
age of Chaetopterus and Lytechinus eggs cause an immediate drop in the 
high-energy phosphate to the unfertilized levels, and inorganic phosphate 
is actually lost from the cells (Barnett and Downey, 1955). 

The effects of malonate on the development of Arbacia eggs were 
thoroughly investigated by Rulon (1948), who demonstrated abnormal dif- 
ferentiation with low concentrations of malonate. If fertilized eggs in the 
1-2-cell stage are placed in 1.2 mM malonate, a slight retardation of cleav- 
age is observed, and at 13 hr (when the controls are swimming bilateral 
gastrulae) there is no evidence of gastrulation, development having pro- 
gressed only to spherical blastulae with no ventral flattening. After 24 hr 
(when the controls are plutei), abnormal gastrulae with thickened apical 
ends and long active cilia are seen. At 48 hr some had developed into ab- 
normal plutei with ciliated apical knobs rather than normal arms. When 
malonate is removed after 13 hr, quite normal plutei are formed, showing 
that the effect is readily reversible. It is interesting that eggs exposed to 
malonate for varying times, then washed free of malonate and fertilized, 
show abnormal development, demonstrating that malonate can so disturb 
egg metabolism that the effects are made evident later after the malonate 
is no longer present. Exposure of unfertilized eggs to 1.44 mM malonate 
for 12 hr, for example, leads to only 30 40% cleavage with irregular cleav- 
age furrows and cells of unequal sizes, development not progressing 
beyond irregular blastulae. Rulon postulated a gradient of succinate de- 
hydrogenase throughout the cells and embryos, paralleling the physiological 
activity gradient, but it is not necessary to assume this to explain effects on 
differentiation. In connection with our work on parapyruvate (Montgomery 
and Bamberger, 1955), we examined the development of Strongylocentrotus 
purpuratus in 25 mM malonate. Up to 24 hr no discernible differences from 
the controls are seen, but at 44 hr a slight inhibition of development can be 
detected, with less formation of the primary mesoderm. Incoordination of 
ciliary activity is also evident, most of the blastulae simply rotating rather 



GROWTH, DEVELOPMENT, AND DIFFERENTIATION 199 

than swimming. At 64 hr, when the controls are beginning to gastrulate, 
the treated embryos are still spherical blastulae, and at 72 hr have not 
progressed beyond this point. In this species, malonate would appear to 
be a rather specific inhibitor of gastrulation without altering cleavage pri- 
marily. It may be noted that other cycle inhibitors, such as parapyruvate 
and fluoroacetate, also specifically block gastrulation in Strongylocentrotus. 
It is difficult to understand these marked differences between the behaviors 
of various sea urchin eggs, and especially the striking effects obtained by 
Rulon with such low malonate concentrations. The free malonate concen- 
trations in his work would have been around 0.17 milf (he used Ca++-free 
sea water) and the intracellular concentration presumably much less. 

There is evidence that insect spermatogenesis is an aerobic process with 
the terminal electron transport through the cytochrome system, and that 
the cycle is the primary pathway for substrate oxidations. Yet no inhi- 
bition of meiosis or differentiation into spermatids and spermatozoa by 
50 ToM malonate is observed in hanging-drop cultures from the Cecropia 
silkworm (Schneiderman et at., 1953). These experiments were carried out 
at pH 6.8-7.2 and it is possible that malonate failed to penetrate. 

Amphibian gastrulation is inhibited by high concentrations of malonate. 
Frog embryos at the early dorsal lip stage were dissected to give explants 
which were exposed to 40 raM malonate at pH 8.0 for 18 hr. Development 
does not proceed beyond the next stage (Ornstein and Gregg, 1952; Gregg 
and Ornstein, 1953). There is no differential effect on the respiration of 
dorsal and ventral explants, both being inhibited about 60%. Unfortunately, 
there is some doubt that the block of development is due to any specific 
effect of the malonate since 45 vciM NaCl apparently inhibits to the same 
degree. Thus the mechanism of the block could have been osmotic or a Na"*" 
effect. The chick embryo seems to be much more sensitive to malonate. 
Explants of chick embryo in the presence of glucose undergo morphogenesis 
and differentiation to the formation of the central nervous system and an 
actively beating heart. Malonate at 1-2 mJf exerts striking differential ef- 
fects (Spratt, 1950). Although no differences were noted during the first 
20 hr, afterwards the central nervous system degenerates completely while 
the heart forms normally and beats. Malonate is the most specific inhibitor 
for the development of the nervous system. Some antagonism of this effect 
was seen with succinate but none with fumarate. No inhibition of mitoses 
in cultures of chick bone is observed with malonate from 65 to 138 nxM 
(A. F. W. Hughes, 1950). 

Mitoses in Mammalian Tissues 

Epidermal mitotic activity in mouse ear fragments was determined over 
4 hr periods at pH 7.4 and 38^ by BuUough and Johnson (1951). From the 
effects of anaerobiosis and various substrates it was concluded that the 



200 1. MALONATE 

energy for mitosis is derived from cycle oxidations. Malonate at 20 mM 
prevents the cell from entering mitosis for 3 hr but evidence of recovery is 
seen after this time. However, at 4 hr, although some cells are progressing 
through mitosis, the number of mitoses is definitely less than in the controls. 
There is no evidence that malonate can stop mitosis once it has begun. 
Epidermal cells adjacent to the wound show a higher number of mitoses 
than normal epidermis and malonate depresses both strongly (see accom- 
panying tabulation). (Bullough and Laurence, 1957). Malonate can thus 



Malonate 



Average number of mitoses 



(milf) Epidermis adjacent to wound Normal epidermis 

8.20 0.96 

10 1.23 0.22 

20 0.05 0.18 

30 0.05 0.01 



inhibit healing without significant damage to the tissue, since no necrosis 
is seen following incubation with malonate. Similar results were obtained 
by Gelfant (1960) and concentrations as low as 0.5 mM are definitely inhi- 
bitory. After 4 hr 50 m M malonate produces some necrosis. The mitotic 
rate in the germinal epithelium of rat ovaries is also suppressed by malo- 
nate, 2 mM having variable effects but inhibiting 21% on the average and 
10 mi/ inhibiting 59% (Weaver, 1959). The relatively high sensitivity of 
mammalian tissue mitosis to malonate would implicate the cycle as an 
important source of energy for this process. 

Growth of Neoplastic Tissues 

The respiration of various types of tumor cells is inhibited by malonate 
(Table 1-26) but comparisons with the appropriate normal tissues have 
seldom been made. Amino acid uptake is also inhibited (Kit and Greenberg, 
1951) and high-energy phosphate compounds reduced (Greaser and Schole- 
field, 1960), but whether tumor tissue is more or less sensitive than normal 
tissue to malonate is not known. Fishgold (1957) obtained evidence that 
hepatoma succinate oxidase is inhibited more readily than the enzyme from 
normal mouse liver, but it is not certain if this is a true difference in the 
affinities of the dehydrogenase active center for malonate or the result of 
other factors. Other than this, there is no demonstrated metabolic difference 
between tumor and normal tissues with respect to inhibition by malonate. 
The time course of the accumulation of succinate in the Flexner-Jobling 
tumor is essentially the same as in other tissues (Fig. 1-11), but the relation- 



GROWTH, DEVELOPMENT, AND DIFFERENTIATION 



201 



ship of the accumulation to malonate concentration is not linear for the 
tumor (Fig. 1-12). As pointed out by Busch and Potter (1952 b), the ac- 
cumulation of succinate in the tumor may be superficially indistinguishable 
from that in normal tissues, but there is reason to believe that the succinate 
in the tumor arises by somewhat different pathways, mainly from gluta- 
mate and related compounds. There is little reason to believe from the 
known metabolic characteristics of tumor tissues that malonate would se- 
lectively depress their growth; indeed, one might expect them to be less 
sensitive to malonate, except for the fact that tumor cells are often more 
rapidly proliferating and more active metabolically than normal tissues. 

Neoplastic growth in general seems to be relatively resistant to malonate. 
Malonate at 30 mM is not toxic to cultures of various tumors but 50 mM 
is toxic to all types of cells in a few hours (Chambers et al., 1943). It is 
interesting that no differences in susceptibility of lymphocytes and other 
wandering cells, whether from normal or neoplastic tissues, are seen. Eagle's 
KB strain of human carcinoma cells is more sensitive, the 50% inhibitory 
concentration of malonate being around 4 ml/ (Smith et al., 1959). Malo- 
nate, like many metabolic inhibitors, is capable of producing acentric blebs 
on Sarcoma 37 ascites cells (Belkin and Hardy, 1961). Although this indi- 
cates some alteration of the membrane properties and the permeability to 
water, the relationship to growth inhibition is unknown. 

Studies of the action of malonate on tumors growing in whole animals 
are more pertinent to the problem of the possible value of this inhibitor 
in therapy. The earliest work was done by Boyland (1940) following his 
investigations of the effects of malonate on tumor respiration. Definite sup- 
pression of growth was observed with malonate at a dose well below the 
lethal, the carcinoma being more sensitive than the sarcoma (see accom- 
panying tabulation). Actual regression of the carcinomata was observed. 



Inhibitor 



Dose 



% Inhibition of growth of 



LD,„ 



(mg/day) Qrafted sarcomata Spontaneous carcinomata 



Malonate 


20 


32 


Malonamide 


25 


36 


Ethylmalonate 


40 


27 


Glutarate 


25 


8 


Adipate 


25 






31 


100 


44 


150 


25 


160 


48 


150 


35 


200 



as indicated by the 131% inhibition. Whether ethylmalonate and malon- 
amide are metabolized to malonate and are active for this reason is not 
known, but the lesser potencies compared to malonate do not suggest that 
these uncharged substances penetrate better. Malonate, fluoride, iodoacetate 



202 1. MALONATE 

and azide were administered to patients with advanced neoplasia and tem- 
porary suppressive effects were noted (Black and Kleiner, 1947; Black et at., 
1947). Sodium malonate was given orally at doses of 1-1.5 g/day. It is 
difficult to state clearly the effects of malonate, since the inhibitors were 
usually given sequentially or together, but it was stated that hematological 
remissions occurred in acute myeloblastic leukemia and that shrinkage of 
solid tumors, with relief of pain, was evident. The tumor cells usually be- 
come refractory to these inhibitors. After resistance to fluoride and iodo- 
acetate has developed, a beneficial effect is seen with malonate. It would 
seem that these results are encouraging enough to warrant further study, 
particularly with combinations of the inhibitors to prevent or reduce the 
development of resistance. Several derivatives of malonate were tested 
against mammary Carcinoma 755 in mice and suppressive action was dem- 
onstrated (Freedlander et al., 1956). Malonic acid at 1.2% in the diet did 
not affect either the tumor size or the growth of the mice. The most active 
ethyl ester was diethylethoxymethylenemalonate (the group =CH — — 
— CH2CH3 on C-2), which at 1.2% in the diet reduces the surface area of 
the tumors 83% while causing minimal loss of body weight. This substance 
is as effective as 8-azaguanine and is less depressant on the total body 
growth. Some diamides are also active, iV-dimethylmalondiamide being the 
most active, reducing tumor area 68% with no effects on body growth. 
It was thought that these substances may be inhibitory to succinate de- 
hydrogenase, probably after hydrolysis, but there is no evidence at present 
for this and it is quite possible that the mechanism is entirely different. 
Despite the lack of evidence for a high susceptibility of the metabolism or 
growth of isolated neoplastic cells to malonate, the in vivo work has brought 
out interesting effects that deserve more thorough investigation. 



CELLULAR AND TISSUE FUNCTION 

Many studies of the effects of malonate on physiological function with 
the object of relating the cellular activity to succinate oxidase or the cycle 
have been reported, but in only a few instances have the necessary data 
been obtained and a relationship adequately established. The general rela- 
tions between enzyme inhibition and changes in cellular function were 
discussed in Chapter 1-9, and several methods for demonstrating correla- 
tions were presented. The complexities of such studies were emphasized 
and the difficulties commonly encountered are well illustrated in the results 
of malonate inhibition. In addition to the various possible metabolic effects 
of malonate, one must bear in mind that malonate, or the cations added 
with it, can directly alter functional processes, actions which can be dis- 
tinguished by the proper controls. 



CELLULAR AND TISSUE FUNCTION 203 

Single Cell Motility 

A thorough study of the respiration and motility of the ciliate Paramecium 
caudatum was made by Holland and Humphrey (1953). Malonate at 20 mM 
inhibits the endogenous respiration 31% but has no effect on the ciHary 
activity over 1 hr. The oxidation of cycle substrates, such as citrate, a- 
ketoglutarate, fumarate, and malate, is well inhibited, and it is likely that 
the cycle is present. Furthermore, the succinate oxidase in homogenates 
is quite sensitive to malonate. One must conclude either that malonate 
does not penetrate sufficiently or that the motility is not entirely dependent 
on the cycle for an energy supply. The answer to this problem may lie in 
the observation that malonate does not inhibit the oxidation of pyruvate 
or acetate. Holland and Humphrey point out that there are many alternate 
pathways for metabolism in paramecia. Thus it is possible that normally 
energy is derived from the cycle but during a cycle block energy is provid- 
ed by other alternate reactions. In the human parasitic ciliate Balantidium 
coli, malonate depresses both respiration, which is probably endogenous, 
and motility (Agosin and von Brand, 1953). The flagellar motility of Ba- 
cillus brevis is not affected by 10 mM malonate (De Robertis and Peluffo, 
1951). The motility of bull sperm is reduced appreciably by 10 mM mal- 
onate and the endogenous respiration is simultaneously inhibited 55% 
(Lardy and Phillips, 1945). However, the results are quite different when 
various substrates are present and motility is depressed very little or not 
at all. Since motility is not depressed with glucose or pyruvate as sub- 
strate, whereas it is with acetate, it is likely that energy sources other 
than the cycle can be used. Ciliary and flagellar activities in different or- 
ganisms are thus affected in various ways by malonate and no uniform 
picture emerges from the data at present available. 

Chemotaxis and phagocytosis in guinea pig leucocytes are partially inhi- 
bited by malonate but the concentration used (140 mM) was so high that 
little significance can be attached to these results (Lebrun and Delaunay, 
1951). Bacterial phagocytosis by human neutrophiles is depressed moder- 
ately by malonate from 33.5 to 100 mM but no effect is seen with 6.7 mM 
(Berry and Derbyshire, 1956). Malonate at 1 mM has no effect on the mi- 
gration of amphibian chromatophores in cultures of the neural crest (Flick- 
inger, 1949). At 10 mM, malonate is toxic to these preparations and not 
specifically depressant to the motility. These limited results point to a 
relative insusceptibility of ameboid-type movements to malonate, which is 
not surprising considering the known metabolic characteristics of such cells. 

Renal Tubular Transport 

Most of the studies of the effects of malonate on the kidney have involved 
the active transport of p-aminohippurate. The accumulation of this sub- 



204 1. MALONATE 

stance by kidney cortex slices, which is quite marked (slice/medium ratios 
around 10), is reduced by malonate: in slices from dogs, 5 raM inhibits 25%, 
10 mM inhibits 35%, and 20 mM inhibits 55% (Shideman and Rene, 1951 b) 
and in slices from rabbits 20 mM inhibits 72% while reducing the respiration 
with acetate by 53% (Cross and Taggart, 1950). This action can be shown 
to occur in the whole animal also. Dominguez and Shideman (1953, 1955) 
removed one kidney from rats, administered solutions of sodium malonate, 
removed the other kidney, and determined the accumulation of j^-amino- 
hippurate. When approximately 10 millimoles/kg of malonate are injected 
subcutaneously, the uptake of p-aminohippurate is depressed 52%, the 
average slice/medium ratio falling from 9.28 to 4.47. The decrease in the 
slice/medium ratio is linear with intravenous doses of 4-7 millimoles/kg. 
Some effect occurs at 15 min after the injections, the maximal inhibition 
is around 60 min, and the transport mechanisms have returned to normal 
by 150 min, indicating the ready reversibility of the action. The transport 
inhibition can also be demonstrated by the renal clearance technique in 
dogs (Shideman and Rene, 1951 b). Malonate at a dose of 0.96 millimole/kg 
depresses the p-aminohippurate Tm* 73%. It was stated that 50% inhi- 
bition of renal succinate dehydrogenase is produced by 1.32 mM malonate 
and thus the dose given would be expected to inhibit in vivo, but the con- 
centration of malonate in the tubular fluid is probably much higher than 
in the blood and permeability factors must also be important. Farah and 
Rennick (1954, 1956) studied the effects of many inhibitors on the p-amino- 
hippurate uptake in guinea pig kidney slices and the results are summarized 
in Fig. 1-18. Malonate is one of the weakest inhibitors but yet exhibits a 
marked effect at 10 mM. Koishi (1959 a) confirmed the inhibition by mal- 
onate on p-aminohippurate accumulation in rat kidney slices, obtaining 
slight inhibition at 1 mM and 65% inhibition at 5 mM. The effects of mal- 
onate are expressed in the following equation: 

log (S/M) = 1.452 + 0.457 log (/) (1-6) 

where S/M is the slice/medium ratio and (/) is the molar concentration of 
malonate. The active transport of p-aminohippurate is thus definitely relat- 
ed in some manner to succinate dehydrogenase and the cycle, assuming a 
specific action of malonate, which is likely at the generally low concentra- 
tions used. 

This conclusion is somewhat substantiated by the findings that the renal 
transport of other substances is frequently not inhibited potently by mal- 
onate. The accumulation of tetraethylammonium ion is scarcely affected 
by malonate up to 40 mM (Farah and Rennick, 1956; Farah, 1957), the 
metabolic requirements apparently being different than for jj-aminohippu- 

* Tm is the tubular transport maximal rate for a substance. 



CELLULAR AND TISSUE FUNCTION 



205 



rate. Phenol red accumulation is also not depressed (Shideman and Rene, 
1951 b), although in isolated flounder tubules it is suppressed by 10-20 roM 
malonate (Forster and Goldstein, 1961). It was claimed that there is a cor- 
relation between succinate oxidase activity and transport in different spe- 
cies. Clearance studies with glucose and phosphate show that there is little 
effect of malonate on their transport: e.g., when p-aminohippurate trans- 
port is inhibited 73%, glucose Tm is decreased only 10%. On the other 
hand, Malvin (1956) reported that malonate quite definitely depresses the 
phosphate Tm, although no data were given. Since malonate interferes 
with the uptake of inorganic phosphate in kidney homogenates, it was 
concluded that malonate in some manner suppresses the esterification of 
phosphate during its transport. 




Fig. 1-18. Effects of inhibitors on the slice/medium (S/M) ratio 
for p-aminohippurate in kidney shces. DNP = 2,4-dinitrophenol, 
FA = fluoroacetate, lA = iodoacetate, DHA = dehydroacetate, 
CN = cyanide, and F = fluoride. (Modified from Farah and 
Rennick, 1956.) 



Renal electrolyte transport is disturbed by malonate (Mudge, 1951). 
Rabbit renal cortex slices were leached for 2.5 hr in 0.15 M NaCl, this 
lowering the tissue K+ concentration and reducing the endogenous respir- 
ation. The slices were then incubated for 30 min in medium containing 
10 mM K+, 10 mM acetate, NaCl to provide a constant osmotic pressure, 
and phosphate buffer, during which time K+ enters the cells. Malonate at 
50 mM almost completely blocks this return of K+ into the cells and re- 
verses the movement of water (see accompanying tabulation). The Na+ 
changes are not so significant because the substitution of divalent anions 
for chloride increases the external, and presumably the internal, Na+ con- 
centration. If fresh slices are treated with malonate, a loss of cell K+ and 



206 1. MALONATE 



Initial Control Malonate 50 xnM % Change 



espiration (Qo^) 


— 


0.69 


0.37 


'^ater content (%) 


79.0 


78.2 


81.1 


+ (meq/kg wet wt.) 


24.7 


58.0 


28.1 


a+ (meq/kg wet wt.) 


112 


93.8 


139 


+ + Na+ 


136.7 


151.8 


167.1 



- 46 

- 90 
+ 100 



an equivalent gain of Na+ would be expected due to the inhibition of the 
transport mechanism responsible for K+ accumulation. The effect on water 
transport is really quite marked and greater than with some fifty other 
inhibitors used. Mudge suggested that malonate acts by depression of 
aerobic metabolism in general and not necessarily by a specific inhibition 
of succinate oxidase, which at the high concentration used is quite possible. 
The results on malonate in vivo occasionally do not correspond to the 
in vitro experiments, nor do they always correspond to each other. The in- 
jection of 10 millimoles/kg malonate into rats leads to a considerable diu- 
resis which lasts for several days (Angielski et al., 1960 a). On the other 
hand, infusion of malonate into the renal artery of a dog at a concentration 
of 8.7 mM causes no significant change in creatine, p-aminohippurate, or Na+ 
clearances, and produces a 15% suppression of urinary volume (Strickler 
and Kessler, 1963). The effects in intact animals are related to acid-base 
imbalance in addition to direct renal action, as shown by unpublished ex- 
periments by Goldberg (1963) in which rats were water loaded and received 
17 millimoles/kg sodium malonate with 17 ml 10% mannitol/kg, this pro- 
ducing certain toxic symptoms (e.g., respiratory difiiculty, sluggishness, 
and mild cyanosis). The results are summarized in the accompanying tab- 
ulation and it is clear that a systemic acidosis was produced. A rather 

Determination Control Malonate 



Urinary flow (ml/min) 0.054 0.058 

Urinary pH 

Titratable acidity (meq/liter) 

Creatinine clearance (ml/min) 

Na+ excretion (meq/liter) 

K+ excretion (meq/liter) 

NH4+ excretion (meq/liter) 

Plasma pH 

Plasma Na+ (meq/liter) 

Plasma K+ (meq/liter) 



6.72 


5.45 


17.0 


46.7 


1.1 


0.89 


9.5 


110 


10.3 


57.5 


25.7 


16.3 


7.39 


7.23 


50 


198 


5.4 


9.5 



CELLULAR AND TISSUE FUNCTION 207 

clear increase in phosphate excretion with a fall in plasma phosphate was 
also observed. 

The effects of malonate on the renal excretion of malate are interesting, 
but it is not known if this phenomenon is related to an action on transport 
mechanisms or to more general metabolic effects. Vishwakarma (1957, 
1962) showed that malonate has no effect on the excretion df malate in- 
duced by malate infusion. However, when succinate is infused, the in- 
creased malate excretion is due to both an increased filtration and a marked 
tubular secretion. Malonate inhibits the tubular secretion of malate and 
converts the excretion to a purely filtration process or causes a net resorp- 
tion. Vishwakarma and Lotspeich (1960) continued this study in chickens 
and found that when malonate is infused with succinate, instead of block- 
ing the formation of malate, it further increases the malate excretion. 
Malonate was infused at a rate of about 6.8 //moles/kg/min. This could 
mean that malonate (1) enhances the formation of malate from succinate, 
(2) facilitates the tubular secretion of malate, or (3) gives rise to malate by 
metabolic conversion. Since malonate infused alone did not significantly 
increase malate excretion, the last explanation is unlikely. In the dog, mal- 
onate inhibits the tubular secretion rather than stimulating it and does not 
inhibit the resorption of malate. The mechanism for this paradoxical effect 
is unknown. Reference may be made to studies of Lotspeich and Woron- 
kow (1964), who unilaterally perfused chicken kidneys and found complex 
effects on the excretions of various organic acids, and concluded that the 
cycle must be involved in some manner. 

Transintestinal Transport 

Quastel has studied the effects of various inhibitors on the transfer of 
glucose across the guinea pig intestinal wall. This is an active transport 
and depends strongly on the aerobic metabolism (as shown by the marked 
inhibition by cyanide and azide) and the associated phosphorylations (as 
shown by the 2,4-dinitrophenol inhibition). When malonate at 20 mM is 
present in the lumen, there is 18.5% inhibition of the glucose transported, 
but if malonate is present both inside and outside, the inhibition is 44.3% 
(Darlington and Quastel, 1953). An increase of K"'" from 6 to 15.6 n\M ac- 
celerates glucose transport about 50%. Malonate inhibits the K^-stimulated 
transport completely at concentrations as low as 2 mM (Riklis and Quastel, 
1958). This result may be related to the greater sensitivity of K+-stimulated 
brain slices to malonate. It was claimed that 20 mM malonate depresses 
both the accumulation of L-monoiodotyrosine-I^^^ in the intestinal cells and 
its transport across the intestine, but no data were given (Nathans et al., 
1960). There is a marked difference between transintestinal transport and 
tissue accumulation of triiodothyroacetate, the former being inhibited much 



208 1. MALONATE 

more readily by a number of substances; malonate at 10 mM inhibits up- 
take 16% and transport 70% (Herz et al., 1961). 

Everted segments of the rat duodenum transport iron from the mucosal 
to the serosal surface against concentration gradients and this process is 
dependent on oxidative phosphorylations (Dowdle et al., 1960). Malonate 
at 50 mM reduces the inside/outside ratio of Fe++ from 4.0 to 0.6. Ca++ is 
also transported actively and malonate at 20 mM reduces the inside/outside 
ratio of Ca*^ from 5.0 to 2.5 (Schachter and Rosen, 1959). Ca++ is also ac- 
cumulated by the intestinal cells, the tissue/medium ratio being 5.8, which 
is decreased by 20 mM malonate to 3.4 (Schachter et al., 1960). The question 
of the chelation of Ca++ and Fe+++ by the malonate arises, since the high 
malonate concentrations would certainly reduce the free ions appreciably. 
This must play some role but in the case of Ca++ cannot explain the reduc- 
tion in the transport, inasmuch as the inside/outside ratio is increased as 
the Ca++ concentration is lowered. 

Gastric Acid Secretion 

The secretion of hydrochloric acid by the parietal cells is dependent on 
oxidations and the formation of ATP, since it is strongly inhibited by cya- 
nide, antimycin, and 2.4-dinitrophenol. However, the secretion in isolated 
rat stomachs is not affected by 10 mM malonate (Patterson and Stetten, 
1949). Injection of malonate in mice subcutaneously inhibits the accumu- 
lation of p-aminohippurate in the kidney but does not inhibit acid secretion: 
4.8 millimoles of malonate reduces the kidney/medium p-aminohippurate 
ratio from 6.1 to 2.9 but inhibits the secretion of hydrochloric acid only 
6% (Davenport and Chavre, 1956). This is near the fatal dose of malonate 
and many of the mice did not live long enough to perform the test. Inas- 
much as succinate oxidase activity is high in the stomach and is readily 
inhibited by malonate, and since fluoroacetate inhibits acid secretion, the 
most likely explanation for the lack of a malonate effect is a failure to 
penetrate into the parietal cells sufficiently. Some evidence for the partici- 
pation of succinate oxidase in acid secretion was obtained by Vitale et al. 
(1956), who showed that stimulation of guinea pig or human gastric mucosa 
with histamine leads to significant increases in the succinate oxidase activity, 
although histamine has no such effect in liver or duodenum. Furthermore, 
succinate oxidase is concentrated in those regions of the stomach where 
the parietal cells are abundant. 

Active Transport of Ions in Various Cells and Tissues 

The effects of malonate on nerve and muscle, to be discussed in the fol- 
lowing sections, depend in part on the modification of the active transport 
of ions in these tissues. Malonate depresses many types of active transport. 



CELLULAR AND TISSUE FUNCTION 209 

as we have seen for kidney and gastric mucosa, and the mechanism may be 
either a simple reduction in the energy supply or a more direct interference 
with electron transport associated with ionic movements across the mem- 
brane. One must also attempt to distinguish between effects on the active 
transport and increases in permeability. If the permeability to an ion is 
significantly increased, its intracellular accumulation may drop because the 
ion pump is no longer able to maintain the normal concentration; such an 
action would appear superficially to be an inhibition of active transport. 

Malonate up to 10 mM has no effect on the transport of Na+ and K+ 
across the human erythrocyte membrane (Maizels, 1951), which is not sur- 
prising since the principal energy source in such cells is glycolytic. In ascites 
tumor cells, substrates (for example, glucose and succinate) increase the 
efflux of Na+. Cyanide at a concentration inhibiting respiration 70% has 
no effect on either the influx or efiiux of Na+, presumably because the rate 
of aerobic glycolysis is simultaneously doubled, this compensating for the 
oxidative depression (Maizels et al., 1958). Malonate at 12.5 mM, on the 
other hand, inhibits respiration 35% but produces only a 5% increase in 
the glycolysis, which may explain why the rate coefficient for Na+ efflux 
drops from 6.5 to 5.1 hr-^ The accumulation of intramitochondrial K+ in 
preparations from rabbit heart is dependent on oxidative phosphorylation 
but is unaffected by 0.2 mM malonate (Ulrich, 1960). Inasmuch as a-keto- 
glutarate was the substrate used, even a complete block of succinate oxida- 
tion might not be expected to have much effect on ion movements because 
sufficient ATP may be generated in the single-step oxidation of a-keto- 
glutarate. Ca++ uptake and binding by kidney mitochondria depend on an 
oxidizable substrate and ATP; it is depressed 75% by 10 mM malonate, 
which suggests interference with the operation of the cycle, but could 
relate to the chelation of the Ca++ by the malonate (Vasington and Murphy, 
1962). The uptake of iodide is inhibited by rather high concentrations of 
malonate in the brown alga Ascophyllum, nodosum (79% inhibition at 25 mM) 
(Kelly, 1953), the rabbit ciliary body (50% inhibition at 50 mM) (Becker, 
1961), and the rabbit choroid plexus (50% inhibition at 20 mM) (Welch, 
1962), but 1 mM malonate has no effect on the uptake or incorporation of 
iodide in sheep thyroid particulate fractions (Tong et al., 1957). 

The accumulation of P,^^ in the roots of the loblolly pine Pinus taeda 
during a 3 hr incubation is inhibited 5% at pH 4.75 but stimulated 54% 
at pH 5.75 (Kramer, 1951). Similar results are obtained in mycorrhizal 
root tips but the inhibition is somewhat greater. It is possible that at the 
concentration (25 mM) of malonate, the stimulation is an ionic effect which 
is partially counteracted by a malonate inhibition at the lower pH. The 
uptake of K+ and Br~ by barley roots is quite strongly inhibited by mal- 
onate at pH 4.5 (see accompanying tabulation) (Ordin and Jacobson, 1955). 
The inhibition is overcome to some extent by malate and fumarate; sue- 



210 1. MALONATE 

cinate, however, actually increases the inhibition. It is likely in this tissue 
that ion accumulation is obligatorily coupled to the operation of the cycle. 



Malonate 




% Inhibition of: 




(mM) 


K+ uptake 


Br- uptake 


Respiration 


5 

10 


39 

92 


42 

70 


30 
55 



Effect of Malonate on Mitochondrial Swelling 

Eat liver mitochondria swell quite readily, as measured by changes in 
light scattering or optical density, when treated with various substances, 
and the effects of malonate on this phenomenon are interesting. Raaflaub 
(1953) established that succinate and phosphate promote swelling. This 
swelling is counteracted by ATP in both cases, but malonate prevents the 
swelling from succinate only. This was confirmed by Tapley (1956), who 
extended the list of substances causing swelling to fumarate, malate, gluta- 
mate, acetate, and a-ketoglutarate. Swelling is prevented by citrate, pyru- 
vate, and oxalacetate, as well as malonate. Since malonate can prevent the 
swelling from substrates other than succinate, there is some question as to 
the specificity of the effect. It was claimed that the same results are obtain- 
ed in the absence of oxygen and thus that the swelling is not related to the 
utilization of these substrates. Quite different conclusions were reached by 
Chappell and GreviUe (1958) inasmuch as they found a good correlation 
between swelling and utilizable substrates. Malonate prevents the swelling 
from succinate but not from a-hydroxybutyrate, and, in general, inhibitors 
blocking oxidations reduced swelling. Matters were further complicated by 
the results of Keller and Lotspeich (1959 b). They found that phlorizin 
caused swelling of kidney mitochondria and that this could be counteracted 
by Mg++, ATP, 2,4-dinitrophenol, and malonate. Hunter et al., (1959 a, b) 
considered the possibility that swelling is related to the fraction of NAD in 
the oxidized form, since amobarbital prevents oxidation of NADH and 
prevents swelling. However, succinate in the presence of amobarbital causes 
swelling and this is blocked by malonate. Glutamate-induced swelling is not 
prevented by malonate. It was concluded that swelling depends on electron 
flow between the substrate and oxygen, and whether or not an inhibitor will 
prevent swelling is determined by where in the electron transport chain 
the substrate and the inhibitor act. This does not very well explain the 
prevention of swelling by 2,4-dinitrophenol and it was suggested that there 
are at least two different types of mitochondrial swelling. Further confusion 
was introduced by Sabato and Fonnesu (1959), who found that swelling is 



CELLULAR AND TISSUE FUNCTION 211 

prevented by oxidizable substrates such as succinate, glutamate, and a- 
ketoglutarate, and that malonate counteracts this preventive action. These 
results were confirmed by Kaufman and Kaplan (1960), who again ob- 
served, in contrast to the earlier workers, that succinate inhibits swelling 
and malonate reverses this protection. They believe that swelling is correl- 
ated with the mitochondrial release of pyridine nucleotides (see tabulation). 



Pyridine nucleotide released 
(//g/20 min) 



Optical density 



No substrate 64 —0.710 

Succinate (20 mM) 14 —0.095 

Malonate (20 mM) 68 -0.740 

Succinate + malonate 38 — 0.520 



Succinate reduces the loss of the pyridine nucleotides and malonate anta- 
gonizes this effect. One must conclude that there must be different mecha- 
nisms of swelling and that the mitochondrial behavior perhaps depends on 
the metabolic state and the nature of the suspension medium. The effects 
of malonate and other anions on the concentrations of free Ca++ and Mg++ 
should also not be ignored, inasmuch as EDTA has usually been shown to 
modify the swelling. 

Conduction and Membrane Potentials of Nerve 

Penetration of malonate into nerve axons in the physiological pH range 
must be very poor. This may account for the failures of Shanes and Brown 
(1942) to observe an eft'ect of 20 mM malonate on the resting potential of 
frog nerve, and of Greengard and Straub (1962) to find an effect of 10 mM 
malonate on nonmyelinated nerve posttetanic hyperpolarization, despite 
the fact that this phenomenon is quite sensitive to other inhibitors. How- 
ever, Jenerick (1957) reported some effect of 10 mM malonate on frog sciatic 
nerve although the action was presumably slow in developing. When the 
action potential spike amplitude is reduced by 80-90%, the threshold for 
stimulation begins to rise rapidly. Conduction block occurs when the rest- 
ing potential has fallen by 30-40%. It is doubtful if the decrease in external 
Ca++ concentration, which was 1.3 mM initially, resulting from chelation 
by malonate could be held responsible for these effects, and it was felt 
that metabolic interference must occur. The preganglionic and postgan- 
glionic action potentials in preparations of cat sympathetic ganglia are 
depressed equally (75-80%) by 14 mM malonate and transmission through 
the ganglia is reduced (Larrabee and Bronk, 1952). The excitability of the 



212 1. MALONATE 

isolated cat carotid body is lowered by perfusion with malonate (Anichkov, 
1953). These meager data are all we have to understand the actions of mal- 
onate on nerve function. Unfortunately, little has been done on junctional 
transmission, inasmuch as it might be predicted that the synapses would 
be more sensitive to malonate than are the axons, because of both a higher 
permeability of such regions to anions and a greater energy requirement 
for the synthesis of acetylcholine. 

Skeletal and Smooth Muscle Function 

Essentially nothing is known of the effects of malonate on skeletal muscle. 
Beckmann (1934) claimed that 6.7 mM malonate causes a swelling of muscle, 
indicating an alteration of permeability. This was termed a membrane- 
loosening effect. In the initial work of Ling and Gerard (1949) with intra- 
cellular microelectrodes, it was observed that 10 toM malonate drops the 
resting potential of frog sartorius muscle from 78 mv to 65.3 mv over a 
period of 3 hr. This may be correlated with the suppression of Na+ ex- 
trusion observed by Kernan (1963) in the same muscle, 30% inhibition 
being produced by 1 mM malonate over 2 hr, an effect similar to that oc- 
curring in brain slices (Bilodeau and Elliott, 1963). No direct work on the 
contractile response to malonate has been done. 

The contractions of isolated rabbit intestine are not inhibited by 10 mM 
malonate, whether in the absence of substrate or in the presence of either 
acetate or glucose (Weeks and Chenoweth, 1950; Weeks et al., 1950). In- 
deed, there is a tendency for malonate to increase the contractile activity 
slightly, especially with glucose as the substrate. There is also no interfer- 
ence with the recovery of substrate-depleted strips produced by the ad- 
dition of acetate or pyruvate. It was suggested that a lack of penetration 
of malonate into the smooth muscle cells might be responsible. Fluoroacetate 
is quite inhibitory under the same conditions so that some relationship of 
the contractility to the cycle is likely. The contractile properties of the 
vascular smooth muscle in the cat hind limb are not affected by 1 mM 
malonate (Hitchcock, 1946), and the behavior of electrically stimulated pig 
carotid artery is not altered by 10 mM malonate (Jacobs, 1950). 

It would be important to know more about the possible effects of mal- 
onate on the formation and release of the neurohormones, such es acetyl- 
choline and the catecholamines, but the data are not available. It is in- 
teresting to note, however, that malonate is reasonably effective in inhi- 
biting the release of histamine from guinea pig lung slices during an ana- 
phylactic reaction (Moussatche and Prouvost-Danon, 1958). The inhibition 
is 10% at 20 mM, 40% at 40 mM, and 50% at 60 mM. The inhibition was 
attributed to the effect on succinate dehydrogenase. Nevertheless, malonate 
at 40-60 ToM has virtually no effect on the release of histamine brought 
about by the application of the histamine-releaser Compound 48/80 (Mous- 



CELLULAR AND TISSUE FUNCTION 213 

satche and Prouvost-Danon, 1957), although respiration of the lung slices 
is depressed fairly strongly. It would appear that malonate interferes with 
the anaphylactic release of histamine by a mechanism other than a direct 
effect on the formation or release of histamine. It may be noted that suc- 
cinate accelerates the oxygen uptake but has no effect on the release of 
histamine. 

Cardiac Membrane Potentials and Function 

The physiological disturbances produced by malonate have been most 
thoroughly studied in the heart. Although the effects are often very slight, 
despite the evident importance of the cycle in the myocardium, under cer- 
tain conditions the responses to malonate are very interesting. The earliest 
investigation was made by Forssman and Lindsten (1946) at Lund, who 
noted a marked discrepancy between the effects of malonate in the whole 
animal and on isolated hearts. Intravenous injections of malonate at doses 
around 3.7-7.5 millimoles/kg to cats and rabbits lead to an increase in the 
venous blood pressure and a fall in the arterial blood pressure, indicating 
cardiac depression. In rabbits the cardiac failure begins about 20 min after 
the injection whereas in cats the changes are immediate. In rabbits the 
heart may stop after 40 min but in cats recovery is the rule. At autopsy 
the heart is found to be dilated. The effects of malonate on the isolated 
perfused rabbit heart, however, are rather small and inconsistent (see ac- 
companying tabulation). Moreover, succinate at the same concentrations 

Malonate °''» ^^^^^^ ^^ 



(mM) 


Amplitude 


Coronary flow 


Rate 


10 


-14 


-23 


- 5 


20 


-19 


-16 


- 8 


40 


-34 


-29 






acts similarly. These are the immediate effects of malonate and it is possible 
that the heart would recover from this depression after several minutes, as 
do rabbit atria (Webb, 1950). It is doubtful if these effects are related to 
inhibition of succinate oxidase; they are more likely ionic actions on the 
membrane. The reduction in the coronary flow may result from a vascular 
constriction, but is more probably the response to the decreased functional 
activity. 

Isolated rabbit atria are depressed only by high concentrations of mal- 
onate, 30-40 mM producing a 30% decrease in contractile amplitude and 
rate at 2 min; the depression is temporary, complete recovery being ob- 
served after 8-10 min (Webb, 1950b). Atria can continue to beat normally 



214 1. MALONATE 

for hours in 50 mikf malonate. The temporary depression brought about 
by malonate is not counteracted by fumarate added either before, with, or 
after the malonate. In fact, fumarate, along with pyruvate, acetate, suc- 
cinate, and malate, has an action very similar to that of malonate on the 
atria. It is not known if this inhibition and recovery are related to the 
somewhat slower but similar time course of ventricular respiration under 
the influence of malonate (page 181) (Webb et al, 1949). The depression is 
not due to chelation of Ca++ or Mg++ since reduction in the concentrations 
of these ions produces a different response. Gardner and Farah (1954) con- 
firmed the resistance of rabbit atria to malonate, finding that 10-20 milf 
has no significant effects on contractility, spontaneous rate, excitability 
threshold, refractory period, and conduction rate. 

The effects of malonate were investigated more thoroughly on rat atria 
(Webb and Hollander, 1959). The contractility is depressed 21% imme- 
diately but slow recovery occurs: the inhibition is 13% during 5-25 min, 
9% during 25-45 min, and 5% during 45-60 min. The malonate concen- 
tration used was 15 mM. The addition of 15 mM NaCl produces a rapid 
contractile depression about half as great as from malonate, so that at least 
part of the initial malonate effect is attributable to the Na+ ion. A slight 
initial rise in the magnitude of the action potential is observed with both 
malonate and NaCl, but in the case of malonate this is soon replaced by a 
small depression. There is also some shortening of the action potential and 
a decrease in its area after the first 5 min, which could be responsible for 
the fall in tension. In summary, the addition of 15 ml/ malonate produces 
a rapid initial effect attributable mainly to the Na+ and this is progressively 
replaced by changes due to the malonate, these latter changes being mod- 
erate depressions of the action potential and contractility. The importance 
of the cycle in the atrial function is indicated by the marked changes brought 
about by fluoroacetate, and thus the resistance to malonate is probably due 
to a low intracellular concentration of malonate. Greater effects on the con- 
tractility of rat atrium were observed by Venturi and Schoepke (1960), 
5 mM depressing 22%, 10 mM 44%, and 20 mM 90%. It was stated that 
NaCl at these concentrations does not alter the contractility. Furthermore, 
succinate is as inhibitory as malonate. The greater depression observed here 
compared to my work is difficult to explain. Venturi and Schoepke used 
Locke solution at pH 7 whereas I used Krebs-Ringer-bicarbonate medium 
at pH 7.4. Part of the larger inhibition seen by Venturi and Schoepke thus 
might be due to the lower pH. In any event, these effects seem to be un- 
related to the inhibition of succinate oxidase and again must be attributed 
to some action directly on the membrane. Venturi and Schoepke found that 
increasing Ca++ concentration can completely overcome the depressant ac- 
tions of malonate, succinate, and the other organic anions used, leading 
them to suggest that the negative inotropic action is due to the chelation 



CELLULAR AND TISSUE FUNCTION 



215 



of Ca+"'". However, there are some reservations in accepting this explanation 
completely. In the first place, Ca++ stimulates atrial contractility and would 
be expected to counteract most depressants in a nonspecific manner. In 
the second place, reducing the Ca++ from 1.22 mM to 0.91 mM does not 
alter the contractility, although further reduction to 0.61 mM depresses 
43%. Malonate at 15 mM would reduce the Ca++ from 1.22 mM to 0.82 mM 
and a small contractile depression may result from this. The total Ca++ in 
Locke solution is 2.16 mM and 20 mM malonate would reduce the free 
Ca++ to 1.31 mM, which alone could not produce the 90% depression seen 
by Venturi and Schoepke. In the third place, monocarboxylate ions, such 
as acetate, lactate, and pyruvate, at 20 mM depress the contractile am- 
plitude 40-45% and these do not deplete the Ca++. 

The modifications in the electrocardiogram following intravenous injec- 
tions of malonate into turtles were studied by Lenzi and Caniggia (1953). 
At a dose of 4.4 millimoles/kg malonate the following occurred: brady- 
cardia, slowing of the a-v conduction, a tendency for the shortening of 
repolarization and electric systoles, with eventually a total a-v block 
and a prolongation of the depolarization time (see accompanying tabula- 
tion). Pacemaker and conduction depression are thus evident, and it is 





Control 


Malonate 
at 30-35 min 


Control 


at 


Malonate 
57-65 min 


Rate 


50 


23 


78 




32 


pq interval 


0.30 


0.465 


0.24 




block 


qrs interval 


0.15 


0.18 


0.10 




0.18 


st-t interval 


0.57 


0.895 


0.44 




1.06 


qt interval 


0.72 


1 . 075 


0.54 




1.24 



quite possible that similar changes would be observed in mammals, con- 
sidering the general behavior of the heart in cats and rabbits treated with 
malonate (Forssman and Lindsten, 1946). The electrocardiogram from the 
embryonic chick heart is not altered by 4 mM malonate (Boucek and Paff, 
1961). 

In contrast to the depressant effects of malonate on the whole heart and 
isolated atria, the rat ventricle strip is usually strongly stimulated, as first 
noticed by Masuoka et al. (1952). Substrate-depleted and hypodynamic strips 
recover to almost the initial contractile amplitude upon addition of 10 mM 
malonate at pH 6.2 (which was used to facilitate penetration of the mal- 
onate), and simultaneously the stimulatory action of succinate is blocked. 
The ability of glucose to induce recovery is augmented by malonate, and 
that of pyruvate is slightly reduced (Berman and Saunders, 1955). This 
interesting positive inotropic action was analyzed in detail because it was 



216 1. MALONATE 

felt that such an action might have bearing on the mechanisms whereby 
the cardioactive glycosides stimulate the failing heart. Only the major 
results will be summarized here. The positive inotropic action occurs most 
strongly when glucose is present, less in substrate-free medium, and not at 
all with pyruvate or a-hydroxybutyrate as substrate (Covin and Berman, 
1956). These results suggested that malonate might stimulate the Embden- 
Meyerhof glycolytic pathway, resulting in an accelerated conversion of glu- 
cose and glycogen to pyruvate. If this were so, pyruvate should produce a 
comparable positive inotropic effect and it does in both substrate-depleted 
and glucose-supplemented strips. Furthermore, iodoacetate at 0.2 mM 
blocks the stimulation by malonate, whereas it does not affect the response 
to pyruvate. The chelation of Ca++ was shown to contribute to the depression 
produced by malonate at high concentrations (20-50 mM), and it probably 
reduces the amount of stimulation seen at the lower concentrations since 
lowering the Ca++ to the degree calculated to occur in 10 mM malonate 
depresses the contractile activity 23%. The effects of malonate on the oxida- 
tion of C^*-labeled substrates by ventricle strips were then studied in cham- 
bers in which the respiration and contractile activity could be determined 
simultaneously (Rice and Berman, 1961). Malonate at 5.6 mM under con- 
ditions in which a positive inotropic effect is observed has very little effect 
on the utilization of glucose- 1-C'^, glucose-6-C^*, and pyruvate-2-C^*, slight 
inhibition of glucose oxidation being noted although this is possibly not 
significant. These results indicate that the stimulatory action is not related 
to (1) acceleration of glucose metabolism, (2) inhibition of the cycle, or (3) 
stimulation of the pentose-phosphate pathway. It was found that C^^Og is 
produced from malonate-2-C^* in ventricle strips, and possibly part of the 
positive inotropic action in substrate-depleted strips is related to the oxida- 
tion of malonate via formation of acetyl-CoA and its incorporation into 
the cycle. However, the explanation for the greater effect of malonate in 
the presence of glucose and the inhibition of its action by iodoacetate is 
not immediately evident. It may be noted that other metabolic inhibitors, 
such as fluoride, arsenate, fluoroacetate, and dehydroacetate, can exert po- 
sitive inotropic actions under the appropriate conditions, so this paradoxical 
effect of malonate is not unique. 

Wenzel and Siegel (1962) determined the dose-response curves for the 
positive inotropic effects of malonate and ouabain on the rat ventricle strip, 
and then constructed an isobologram, plotting the malonate concentration 
against the ouabain concentration for a chosen contractile stimulation. 
Since the isobol sags, i.e., is concave upwards, they claimed it is clear that 
potentiation occurs and that this indicates the sites of action of malonate 
and ouabain are different. There is some doubt that a moderately sagging 
isobol can be interpreted as potentiation, inasmuch as pure summation 
often elicits such a curve (see Figs. 1-10-7, 8). 



EFFECTS OF MALOXATE IN THE WHOLE ANIMAL 217 

The response of the heart to neurohormones is not altered by malonate. 
The positive chronotropic effect of epinephrine on the frog heart is not 
changed by 0.1 roM malonate (Nickerson and Nomaguchi, 1950), which is 
not surprising considering the concentration. A brief report by Ellis and 
Anderson (1951 a) stated that malonate does not affect the stimulation by 
epinephrine except after prolonged treatment when the frog heart is de- 
pressed. The malonate concentration was not given. It would seem likely 
that any severe metabolic disturbance producing marked cardiac depression 
would prevent, or at least reduce, any type of stimulation, since the ad- 
ditional functional activity would demand more energy, so that an an- 
tagonism of epinephrine by malonate is not of much significance unless it 
occurs when the heart is not too much depressed. Malonate has no demon- 
strable effect on the response of the heart to acetylcholine, with respect to 
reduction of either rate or contractility (Webb, 1950 b), whereas fluoroace- 
tate alters the response markedly, indicating again that the cycle is of im- 
portance in these myocardial functions but that malonate does not reach 
sufficiently high intracellular concentrations. 



EFFECTS OF MALONATE IN THE WHOLE ANIMAL 

A summary of the miscellaneous results relating to toxic and lethal ef- 
fects of malonate is given in Table 1-28. One may conclude that in mammals 
injected doses of 1.5 2.5 g/kg (10 17 millimoles/kg) of sodium malonate are 
generally lethal. Such doses, especially when given intravenously, probably 
produce plasma levels in excess of 10 inM malonate at peak concentra- 
tions. A dose of 12 millimoles/kg subcutaneously in rats leads to a plasma 
concentration of 4.5 mill at 30 min, and another similar dose raises the 
plasma level to around 8 mM (Busch and Potter, 1952 a). Intravenous 
injection would give higher peak levels. When compared in the same ex- 
periments, the acid is more toxic than the sodium salt. It is difficult to 
know if this is due to a nonspecific acid effect or to better penetration into 
the tissues. It would appear that malonate is more toxic to mice at 38° than 
at 30° environmental temperature (Gruber et al., 1949). 

The sequence of symptoms resulting from the injection of malonate into 
rats and mice was described by Gruber et al. (1949) as: champing, air hunger, 
maintenance of the head in a dorsally flexed position, and clonic convulsions. 
Busch and Potter (1952 a) found dyspnea and convulsions in rats follow- 
ing injection of toxic doses. The cause of death has been attributed to var- 
ious actions. Forssman (1941) and Forsmann and Lindsten (1946) believed 
that death is due to cardiac failure (the cardiovascular effects observed were 
discussed in the previous section). Handler (1945) also favored a cardiac 
mechanism for death and found the succinate oxidase to be inhibited 50- 
75% in homogenates prepared from poisoned animals. He also noted that 



218 



1. MALONATE 



"So 



ft 

CO 



lO '« 







c3 




:;^ 






JU 










C J 




r«^ 












05 


S3i 




^—^ 




(T) 


GO 


o 


H.^ 


CD 


TO 


Tt* 








05 








-1^ 


^ 




a 




^ 


T3 




S 




c 


eS 


T3 


t*. 


l>» 


> 


>> 


<» 






3 O 



^ t- 



(» 4) o3 

W W P 



o 

p< 

(_; tn ^ ^ 

m p^ Q o 



Ci 


Oi 


C5 


xs 


"^ 


'IH 


•* 


-* 


ctf 


(-1 


02 


C5 


c: 


f/; 


ii 












" — ' 


" — ' 


^^ 


73 


o 


t**" 


^ 


^ 




^ 


w 


« 


« 










cS 


t3 
C 






«> 


01 










CS 


u 


b 


u 






01 


0) 


0) 




^ 


-O 


-O 


^ 




o 


D 


3 


3 




m 


;h 




t4 




3 


O 


O 


o 


S 


M 



^ ^ rW "" 

a a ^ . 

2 ^ ^ § "e 

cd c^ c^ 



CO ^ 05 

Ift /-T% ^H 



f "O 03 03 CD 
C^ ^ CO M «1 






f=^ O ^ 



cs -r? c^ 



'T3 -C 



oj o d 



hJ Z H h^ ^ S 






_ X 



o o o cS 



Q Q Q Q 

hj ^-5 H-5 h^ 



o <u o 0) c 
H h^l 12; hJ M 






(N |ci-*->*OOiO(M»OOOOOOCO«DTj< 



OG0OC0-Ht-rt(M(N-H 



O-hO-hO^OOO 



-t3 +3 ^o -u -u 



09 03 CO 



i c^ cd ^ ^ cd cd c6 

^ vj vj vj _^ vj _^ CO CO _^ CO CO CO CO CO 

'oo3cScS*ocS'oc3cS'Sc8(:3«Sc8cS 



'i^ O^ 



P 



Ph 



o o 






> CM 



PL| O O Q > 

h-i CO CO W M 



Ci2 



IS ,o 



43 T) 

03 ^ 



bO 



to C 
O — ' 
P ^o 



03 

>> "a 
a 






W) bO 





O 


O 


O 


03 

cS 


.2 

cS 
Xi 






o 


C 

o 


T3 


<D 



03 cS 



P P 



EFFECTS OF MALONATE IN THE WHOLE ANIMAL 219 

the heart is dilated at death and that there is an accumulation of ascitic 
and pleural fluid. Forssman and Lindsten, from their failure to obtain appre- 
ciable direct action on the isolated heart, suggested an indirect mechanism, 
possibly mediated through the effect of malonate on liver metabolism. The 
dyspnea and various central nervous system effects may result from a direct 
action of malonate but could also arise from the ionic and acid-base im- 
balances produced. Handler (1945) showed that malonate causes a marked 
fall in CO2 capacity in the blood from 53 to 6 vol%. Although no study of 
the alterations in the plasma electrolytes undoubtedly produced by mal- 
onate has been made, Wick et at. (1956) noted that the ability of the blood 
to coagulate is reduced, and attributed many of the actions of malonate to 
the chelation of divalent cations. Other changes in the blood composition 
may also be responsible for some of the toxic effects. Handler (1945) re- 
ported that 1.6 g/kg malonate given subcutaneously to rabbits increases 
the blood glucose 368%, blood lactate 545%, blood pyruvate 163%, serum 
inorganic phosphate 262%, and serum organic phosphates 155%. (The keto- 
nemia produced by malonate was discussed previously). There is thus much 
opportunity for secondary mechanisms to play a role in the toxicity of 
malonate. At the present state of our knowledge it is even difficult to evaluate 
the importance of succinate oxidase inhibition in these effects. 

The kidneys have the highest concentration of malonate after administra- 
tion and therefore the renal effects and nephrotoxicity have been investigat- 
ed. Early arguments about the renal toxicity of glutarate were published 
between 1907 and 1912. Rose (1924) "reinvestigated this and found that 
glutarate is a nephrotoxic substance in rabbits, as indicated by the increases 
in nonprotein nitrogen, urea, and creatinine, and the almost complete disap- 
pearance of renal function as measured by the phenolsulfonphthalein test. 
A single experiment with malonate was reported. No renal damage was 
observed after 2 g given on successive days and 3 g 2 days later, nor was 
there a change in the rate of dye excretion. Corley and Rose (1929) reported 
that methylmalonate and ethylmalonate are slightly toxic to the kidneys 
at doses of about 1 g/kg in rabbits, there being a definite increase in the 
nonprotein nitrogen and some reduction in dye secretion, although both 
effects are transitory. Extensive renal damage was observed by Becker and 
Rieken (1954) following the intraarterial injection of 20 mg potassium mal- 
onate (Fig. 1-19 a). The vessel walls become edematous, potocytosis is evi- 
dent, and many perinuclear vacuoles appear in the loops of Henle. However, 
it requires much higher doses to depress the respiration of kidney slices 
prepared from injected animals, 80 mg potassium malonate giving no effect 
and 150 mg inhibiting 30.5%. Similar histological changes occur after in- 
cubating kidney tissue in vitro with 110 mil/ potassium malonate for 20 min 
(Fig. 1-19 b), vacuolization being intense. Tinacci (1953) found not only 
kidney damage but widespread degenerative changes 2-8 days after sub- 



220 



1. MALONATE 





Fig. 1-19. Renal damage resulting from (a) intraarterial injection of 20 mg potas- 
sium malonate into rabbits, and (b) incubation of slices with 110 mif malonate for 
20 min. (From Becker and Rieken, 1954.) 



EFFECTS OF MALONATE ON BACTERIAL INFECTIONS 221 

cutaneous injections of 0.4-0.8 g/kg of sodium malonate, almost all organs 
being involved. Malonate is diuretic when subcutaneously injected into 
hydrated rats at a dose of 11.5 millimoles/kg, this dose being sufficient to 
inhibit kidney succinate oxidase (Fawaz and Fawaz, 1954). However, the 
effect is probably osmotic rather than due to enzyme inhibition by the mal- 
onate, since KNO3 at the same dosage induces the diuresis, and also be- 
cause diethylmalonate, which inhibits kidney succinate oxidase, has no 
diuretic activity. Summarizing these results, one may conclude that renal 
damage may occur from high doses of malonate, but that minimal changes 
in the kidney occur after the usual toxic doses. It seems unlikely that the 
renal action is of major importance in poisoning or death from malonate. 

EFFECTS OF MALONATE ON BACTERIAL INFECTIONS 

The influence of enzyme inhibitors on the course of bacterial infections 
is of great interest because the results have bearing on the fundamental 
question of the metabolic basis of the resistance to infection. The effects 
of malonate on Salmonella typhimurium infection in mice have been studied 
by Berry and co-workers at Bryn Mawr in a series of excellent investiga- 
tions. Mice injected intraperitoneally with a Salmonella suspension show 
evidence of the infection on the third day and most succumb by the sixth 
day. If mice are given intraperitoneal injections of 20 mg sodium malonate 
in saline every hour for 8 hr, they die much more rapidly from the infection 
than animals given only saline injections (Berry and Mitchell, 1953 a). 
The striking effects is illustrated in Fig. 1-20. Noninfected mice show no 
effects of the malonate. Thus sublethal doses of malonate are able either 
to accelerate bacterial proliferation or to decrease the resistance of the 
host markedly. These are the basic observations and the later work attempts 
to elucidate the mechanisms by which these effects are brought about. 

The reduction of the survival time in mice infected with Salmonella is 
not unique. Similar effects of malonate on infections with Proteus morganii, 
Staphylococcus aureus, Streptococcus pyogenes (Berry and Mitchell, 1954), 
Diplococcus pneumoniae, and Klebsiella pneumoniae (Berry et al., 1954 a) 
have been observed. Furthermore, reduced survival times have been found 
in Salmonella infected rats, guinea pigs, and chickens (Berry and Beuzeville, 
1960). Finally, the phenomenon has been seen with other inhibitors, such 
as fluoroacetate and arsenite (Berry et al., 1954 a, b). This, then, is a gen- 
eral effect of certain types of inhibitor, especially those affecting the cycle 
in some manner, and the problem is thus more important because it must 
relate to some fundamental metabolic relationship between host and bacteria. 

We shall now examine in greater detail some of the characteristics of 
the malonate effect. Malonate not only can reduce survival time, but in 
some instances can change a nonlethal infection into a lethal one, this being 



222 



1. MALONATE 



observed with Diplococcus, Staphylococcus, and Proteus. That is, an infec- 
tion which does not kill plus a dose of malonate which is non-toxic may 
cause the death of all the animals within several hours (Berry et al., 1954 a). 
This is a true case of synergism. It has been found that Salmonella bactere- 
mia in mice is much greater in malonate-treated animals than in the con- 
trols (Berry and Mitchell, 1953 b, 1954). The bacteremia in the controls 
reaches a low peak value soon after inoculation and then falls off, whereas 
in the presence of malonate it continues to progress. At 9 hr, the blood of 



No. 10 
Dead 



/ Malonate 


/ 


j 


/Saline 




/ controls 


. 


y 



24 30 36 42 48 

Time after injection ( Hours) 



Fig. 1-20 Effect of malonate given in 8 hourly injections of 20 mg 
sodium malonate on the mortality of mice intraperitoneally in- 
jected with suspensions of Salmonella typhimurium . (From Berry 
and Mitchell, 1953 a.) 



the controls contains 10,000 to 20,000 bacteria/ml whereas the blood of 
malonate-treated animals has a count of around 3,000,000 bactcria/ml. 
Thus the bacteremia is over 100-fold as severe in the treated animals as in 
the controls. This bacteremia is a reflection of the situation throughout 
the body. The total number of bacteria in the body 6.5 hr after the injec- 
tions is 1.65 X 10^ in the controls and 32.1 X 10^ in the malonate-treated 
mice (Berry, 1955). The ratio of counts in the treated and control series is 
20 for the whole body and 19 for the blood at this time. Now, the interesting 
thing is that, at the time of death, the number of bacteria in the body is 
the same in both the controls and malonate-treated mice. This clearly shows 
that malonate does not alter the susceptibility of the mice to the bacteria, 
but reduces the time required for the bacteria to multiply to that number 
necessary to kill. 



EFFECTS OF MALONATE ON BACTERIAL INFECTIONS 223 

The rapid proliferation of bacteria could be due to a weakening of the 
body defenses for disposing of the bacteria. This does not seem to be the 
case, inasmuch as malonate has no effect on the uptake of thorotrast by 
the reticuloendothelial system (Berry, 1955) nor does it depress phagocytosis 
except at very high concentrations (Berry and Derbyshire, 1956). Instead it 
would appear that malonate disturbs metabolism in such a way that it 
creates a more favorable environment in the host for bacterial growth. 
Malonate itself may be metabolized slightly by Salmonella, but not to the 
extent required to explain the explosive proliferation (Berry and Beuzeville, 
1960). Growth medium was prepared from the eviscerated carcasses of con- 
trol and malonate-treated animals and it was found that the bacteria grow 
more rapidly in the latter (Berry, 1955). It has also been shown that Sal- 
monella grows more rapidly in the peritoneal fluid of malonate-treated mice 
than in the controls (Berry and Beuzeville, 1960). Citrate is known to ac- 
cumulate following the administration of malonate. This was confirmed in 
mice given the doses of malonate capable of reducing survival times of 
infected animals (Berry et al., 1954 b). Both malonate and endotoxin from 
Salmonella increase citrate levels in most tissues, and together the increases 
are often greater than with either alone (see accompanying tabulation). It 



Treatment 



Citrate (//g/g) in 



None 


42 


94 


76 


HI 


Malonate 


43 


245 


225 


120 


Endotoxin 


50 


273 


190 


187 


Malonate 










+ endotoxin 


70 


173 


173 


407 



Blood Spleen Kidney Heart Duodenum Liver 

132 109 

115 170 

170 443 

543 240 



is thus possible that a summation of effects on the cycle could be partially 
responsible for the increased mortality. However, Salmonella infection does 
not increase citrate levels (Berry and Beuzeville, 1960). Could the increased 
citrate be favorable to the growth of the bacteria? It is unlikely that this 
is a major factor because malonate is the most potent inhibitor for reducing 
survival times and yet both arsenite and fluoroacetate cause greater accu- 
mulations of citrate. The primary cause of the augmented bacterial proli- 
feration has not been found but the range of possible mechanisms has been 
narrowed. Since there are many other possible substrates for Salmonella 
that accumulate during malonate inhibition, it will be necessary to examine 
these in mice and their effects on the growth of Salmonella. 

Some work on this problem in other laboratories may be briefly mention- 
ed. Malonate reduces the antibacterial activity of guinea pig blood toward 



224 1. MALONATE 

Salmonella enteritidis but this is not due to a decrease in the number of 
leucocytes (Yamauchi, 1956). The survival times of chicks infected with 
Salmonella pullorum are reduced by 500-800 mg/kg of malonate injected 
simultaneously or shortly after the bacterial inoculation (Gilfillan et al., 
1956). On the other hand, the diethyl ester of malonate increases the sur- 
vival time of mice infected with Mycobacterium, tvbercvlosis, when admin- 
istered daily for 2 weeks at oral doses of 250-500 mg/kg (Davies et al., 
1956). Compounds of this type are thus considered worthy of study as 
chemotherapeutic agents in tuberculous infections. 



METABOLISM OF MALONATE 

Malonate occurs normally in many types of organism and occasionally 
at concentrations possibly inhibitory to succinate dehydrogenase. Many 
organisms are capable of metabolizing malonate by various pathways and 
some are able to utilize it for growth or cell functions. In some cases, in- 
deed, it is difficult to demonstrate the inhibitory action of malonate in the 
presence of its own oxidation. The metabolism of malonate must always be 
considered in studies of the effects of malonate on any type of cellular ac- 
tivity. It is often impossible to detect and correct for the metabolism of 
malonate without using labeled malonate. It is possible that many studies 
of the inhibition of respiration or cycle activity by malonate have been 
complicated by the oxidation of the malonate. 

Occurrence of Malonate 

Malonate has been isolated or demonstrated chromatographically from a 
number of microorganisms, plants, and animals, and it is likely, consider- 
ing the recent demonstration of its role and the role of malonate derivatives 
in fatty acid metabolism that its occurrence is widespread. The accom- 
panying incomplete tabulation will serve to illustrate this. Malonate has 
also been found in winter wheat, barley, oats, alfalfa, kidney bean leaves, 
clover, pea leaves, vetch leaves (Soldatenkov and Mazurova, 1957), sake 
(Kawano and Kawabata, 1953), and several products prepared from plants. 
Although no thorough studies of animal tissues have befen made, it is 
evident that malonate must occur in rat, dog, and human tissues to some 
extent if it is found in the urine. Although the name of malonate comes 
ultimately from the Latin mains, it has never been identified in apples 
or other fruit. 

Methylmalonate has been found in Propionibacterium (Stjernholm and 
Wood, 1961; Wood and Stjernholm, 1961), pigeon liver (Bressler and Wakil, 
1961), pig heart (Flavin et al., 1955), mouse adipose tissue (Feller and Feist, 
1957), and rat and human urine (Boyland and Levi, 1936; Barness et al., 



METABOLISM OF MALONATE 



225 



Source of malonate 



Reference 



Achromobader guttatum 
Nocardia corallina 
Penicillium funiculosum 
Aspergillus niger 



Phaseolus vulgaris (bush bean) 



Phaseolus coccineus (runner bean) 
Wheat plants 

Bunias orientalis (Cruciferae) 
Tobacco leaves 



Lucerne (green alfalfa) 

Hevea brasiliensis 

Helianthus annus 

Umbelliferae {Anthriscus and Apium) 

Leguminosae (18 species) 

Rat urine 

Dog urine 
Human urine 



Sguros and Hartsell (1952a) 

i.ara (1952) 

Igarasi (1939) 

Challenger et al. (1927) Walker 

et al. (1927), Subramanian 

et al. (1929) 
Young and Shannon (1959), 

Rhoads and Wallace (1957), 

Huffaker and Wallace (1961) 
Bentley (1952) 

Nelson and Hasselbring (1931) 
Jermstad and Jensen (1951) 
Wada and Kobashi (1953), Bel- 

lin and Smeby (1958), Vickery 

and Palmer (1956 b, 1957), 

Vickery (1959) 
Turner and Hartman (1925) 
Fournier et al. (1961) 
Bentley (1952) 
Bentley (1952) 
Bentley (1952) 
Stalder (1958), Thomas and 

Stalder (1958) 
Thomas and Kalbe (1953) 
Stalder (1958) 



1957; Stalder, 1958; Thomas and Stalder, 1958). Ethylmalonate has been 
found in rat and human urine (Stalder, 1959). Hydroxymalonate (tartronate) 
occurs in Acetohacter (Stafford, 1956) and malonic semialdehyde in Pseudo- 
monas (Nakamura and Bernheim, 1961). 

The concentrations of malonate in plant tissues are often surprisingly 
high. The legumes and umbellifers analyzed by Bentley (1952) contain 
0.5-2 mg/g fresh tissue. These values correspond to 7-30 raM malonate if 
distributed uniformly throughout the tissue water. The stems of the runner 
bean {Phaseolus coccineus) contain 2.1 mg/g and the expressed juice of the 
stem is 30 milf in malonate. Since 20 raM malonate at pH 4.5 inhibits the 
respiration of these stems 50% and causes accumulation of succinate, one 
would expect the metabolism in these plants to be constantly suppressed 
by the malonate, unless the malonate is in some manner segregated from 
the metabolic systems. Soldatenkov and Mazurova (1957) reported similar 



226 1. MALONATE 

values in legumes (2-3% of the plant dry weight) and that in kidney-bean 
and clover leaves malonate represents 45% of the total di- and tricarbox- 
ylates present. Bush-bean {Phaseolus vulgaris) leaves often contain as 
much as 10 mg/g dried tissue and malonate is more concentrated than fu- 
marate or succinate, although less than malate and citrate (Young and 
Shannon, 1959). In man malonate is excreted in the urine at an average 
rate of 0.0047 mg/kg/day and in the rat at 10 times this rate (Stalder, 1958). 
This amounts to only 0.32 mg/day in man (only two individuals were tested 
so these averages are not accurate). Since malonate is apparently metabol- 
ized in mammals, the tissue concentration or rate of excretion will reflect a 
balance between formation and destruction. In other words, these excretion 
values do not necessarily represent the rates of malonate formation. 

Relatively little is known about the pathways for the formation of mal- 
onate, but the miscellaneous observations make it likely that different reac- 
tions are involved in various organisms. Malonate can arise from many 
different substrates but in most cases the pathways are complex and the 
immediate precursors are not known. Malonate can be formed from pyri- 
midines and barbiturates in the mycobacteria (Hayaishi and Kornberg, 
1952), from pyrimidines in Nocardia (Lara, 1952), from acetate in Hevea 
hrasiliensis (Fournier et al., 1961) and avocado (Mudd and Stumpf, 1961), 
from citrate in Aspergillus niger (Challenger et al., 1927), from succinate 
in Aspergillus niger (Subramanian et al., 1929), from asparagine in rats 
(Thomas and Stalder, 1959), from oxalacetate in pig heart extracts catalyzed 
by metmyoglobin and Mn++ (Vennesland and Evans, 1944; Vennesland 
et al., 1946), and from malonyl-CoA in Penicillium cyclopium, (Bentley and 
Keil, 1961). The high concentrations of malonate in bush-bean plants led 
Huffaker and Wallace (1961) to study the mechanism of the accumulation. 
They found that the malonate synthesis is related to the dark COg fixation 
in the roots, phosphoenolpyruvate being carboxylated to oxalacetate and 
this going to malonate with the help of one or more enzymes. The addition 
of phosphoenolpyruvate and Mg++ to root homogenate leads to the for- 
mation of labeled malonate from 0^*02- It was also found that any other 
reactions utilizing oxalacetate decrease the yield of malonate. It is very 
interesting that frogs accumulate malonate-C^* from C^Oa, along with other 
dicarboxylates (Cohen, 1963). In normal strains the malonate accounts for 
only 0.3-0.5% of the total incorporation but in hybrids {R. pipiens X R. 
sylvatica) the value is 6-23%. This increased accumulation in the hybrids 
was attributed to some defect in the metabolism of malonate. 

Methylmalonate can be formed from propionate (Flavin et al., 1955) in a 
variety of tissues, and in rat liver the pathway has been shown to go through 
succinate (Katz and Chaikoff, 1955). The feeding of isobutyrate and valine 
to rats leads to the formation of methylmalonate (Thomas and Stalder, 
1958) and the feeding of isoleucine leads to ethylmalonate (Stalder, 1959). 



METABOLISM OF MALONATE 227 

The first reaction appears to be an oxidative deamination. Since methyl- 
malonyl-CoA is a common intermediate in many tissues, methylmalonate 
probably arises from any substance forming the coenzyme A derivative. 
(This will be discussed in greater detail in the following sections). 

General Occurrence and Nature of Malonate Metabolism 

The metabolism of malonate by many organisms and tissues has been 
conclusively demonstrated by a variety of techniques. The ideal method is 
the determination of C^^Og or other labeled products formed from labeled 
malonate, but in some instances good evidence is provided by studies of 
oxygen uptake, especially when the endogenous respiration is very small, 
or by growth in media containing only malonate as a utilizable substrate. 
In other cases, the evidence is more indirect. For example, a stimulation of 
growth rate or a rise in respiration in the presence of other substrates may 
suggest the utilization of malonate but does not prove it. In the tabulation 
on page 228 of organisms in which malonate metabolism has been claimed 
to occur, those that are probable and based on indirect evidence are des- 
ignated by (P). It may be noted in addition that Shannon et al. (1959) found 
malonate to be metabolized by the excised leaves of 15 different common 
plants (such as fig, peach, eucalyptus, azalea, and lantana) and 30 other 
plants representing 27 families, from which it must be concluded that plants 
are generally capable of utilizing malonate. 

The bacterial oxidation of malonate occasionally shows a lag period, first 
observed by Lineweaver (1933). The rate of oxidation by Azotobacter is 
very low for 2-3 hr, rises to a maximum around 6-8 hr, and falls off by 
10 hr. The malonate is eventually 99% metabolized with an R.Q. of 1.6. 
The theoretical R.Q. for complete oxidation: 

CHaiCOOH), + 2 O2 -» 3 CO2 + 2 H,0 

is 1.5. Lineweaver postulated that two separate reactions are involved: the 
decarboxylation of malonate to acetate, and the oxidation of the acetate. 
He attributed the lag phase to the slow decarboxylation, which was sup- 
ported by the progressive decrease in the R.Q. with time. This interpreta- 
tion was criticized by Karlsson (1950) because Lineweaver had not used 
cells adapted to malonate. Malonate-grown Azotobacter does not decarbox- 
ylate malonate appreciably under anaerobic conditions, so Karlsson con- 
cluded that oxygen is required for the initial attack on malonate, either 
because malonate must be oxidized before decarboxylation or because 
oxygen may be required for some activation of malonate (for example, by 
a phosphorylative mechanism). A lag phase was also demonstrated for 
Aerobacter by Barron and Ghiretti (1953), the maximal oxidative rate oc- 
curring around 2-3 hr after malonate addition. Only 40% of the malonate 



228 



1. MALONATE 



Organisms and tissues 
metabolizing malonate 



Reference 



Pseudomonas aeruginosa 
Pseudomonas fluorescens 



Escherichia coli (P) 

Aerobacter aerogenes 
Azotobacter agilis 
Salmonella typhimurium (P) 
Mycobacterium tuberculosis 



Mycobacterium phlei 
Aspergillus niger 

Aspergillus (6 species) (P) 

Penicillium cyclopium (P) 

Penicillium (3 species) (P) 

Streptomyces olivaceus 

Pullularia pullulans (P) 

Avena coleoptile (P) 

Chlorella pyrenoidosa (P) 

Pollen of Camellia, Thea, and Lilium 

Tobacco leaves 

Bush bean leaves 

Peanut mitochondria 

Euglena gracilis 

Hymenolepis diminuta (cestode) (P) 

Locusta migratoria fat body 

Chickens (P) 

Pigeon liver 

Mice 

Rats 



Rat liver 

Rat ventricle 

Rabbits 

Dogs 

Dog heart and muscle 

Human placenta 

Human prostate 



Gray (1952) 

Hayaishi (1953, 1954, 1955 a), Wolfe 

and Rittenberg (1954), Wolfe et al. 

(1954 a, b, 1955) 
Grey (1924), Quastel and Whetham 

(1925), Cook (1930) 
Barron and Ghiretti (1953) 
Lineweaver (1933), Karlsson (1950) 
Berry and Beuzeville (1960) 
Hayaishi and Kornberg (1952), 

Bernheim et al. (1953), Horio and 

Okunuki (1954), Kusunose et al. 

(1960) 
Miiller et al. (1960) 
Walker et al. (1927), Challenger et al. 

(1927) 
Berk et al. (1957) 
Bentley and Keil (1961) 
Berk et al, (1957) 
Maitra and Roy (1961) 
Clark and Wallace (1958) 
Albaum and Eichel (1943) 
Eny (1951) 
Okunuki (1939) 
Vickery (1959), Vickery and Palmer 

(1957) 
Young and Shannon (1959) 
Giovanelli and Stumpf (1957) 
Danforth (1953) 
Read (1956) 
Tietz (1961) 

Clementi (1929), Pupilh (1930) 
Menon et al. (1960) 
Lifson and Stolen (1950) 
Lee and Lifson (1951), Busch and 

Potter (1952a), Nakada etal. (1957), 

Thomas and Stalder (1959) 
Menon el at. (1960) 
Rice and Berman (1961) 
Wick et al (1956) 
Pohl (1896) 
Menon et al. (1960) 
Hosoya and Kawada (1958), Hosoya 

et al. (1960) 
Andrews and Taylor (1955) 



METABOLISM OF MALONATE 



229 



is oxidized but the R.Q. is 1.47, close to that for complete oxidation to 
CO2 and water. Again, no decarboxylation is observed in nitrogen. Horio 
and Okunuki (1954) reported a 30-60 min lag period for Mycobacterium and 
also found that decarboxylation presumably precedes oxidation, because 
CO2 formation always is ahead of Oo uptake, and acetate can be demon- 
strated in the culture after 2 hr. The explanation of this behavior is found 
in the work of Gray (1952) on Pseudomonas. Unadapted cells show a lag 
period of 2 hr whereas cells cultured for 70 hr in 22 mM malonate are able 
to oxidize malonate immediately (Fig. 1-21). It was postulated that mal- 




Time ( hours) 



Fig. 1-21. Oxidation of malonate by normal and 
adapted Pseudomonas aeruginosa. (From Gray, 1952.) 



onate decarboxylase is an adaptive enzyme, and it was pointed out that 
the resistance of certain microorganisms to malonate may be due to such 
an enzyme as well as to permeability barriers. Horio and Okunuki showed 
that streptomycin does not directly inhibit the decarboxylation of mal- 
onate or the oxidation of acetate, but inhibits malonate oxidation in un- 
adapted cells, probably by preventing the synthesis of the decarboxylase. 
The oxygen requirement may also relate to the adaptive enzyme syn- 
thesis. 



230 



1. MALONATE 



Pathways of Malonate Metabolism in Microorganisms 

An analysis of the pathways of malonate oxidation was made simultane- 
ously in the laboratories of Hayaishi and Rittenberg between 1953 and 1955. 
The work was done on Pseudomonas fluorescens, a soil isolate capable of 
utilizing malonate as the sole carbon source, and adapted to malonate by 
culture in 27-33 nvM malonate media. Hayaishi (1953) observed that the 
decarboxylation of malonate requires ATP and CoA and postulated that 
malonate must first be activated, probably to malonyl-CoA. Using partially 
purified extracts from Pseudomonas, it was shown that malonate is quanti- 
tatively converted to COg and acetate, no other products being detectable 
chromatographically. The proposed scheme (1-7) may be represented as 
follows (Hayaishi, 1954, 1955 a): 



Malonate 



Acetate 



Malonate 




(1-7) 



Acetyl — CoA 



The cyclic process thus involves the transfer of CoA back and forth be- 
tween the acetyl and malonyl groups. Wolfe et at. (1954) also ruled out the 
direct decarboxylation to acetate by showing a dependence on ATP and 
CoA, and in later work (Wolfe et al., 1954 b, 1955; Wolfe and Rittenberg, 
1954) proposed the following scheme (1-8) based mainly on chromatographic 
analyses of intermediates and products: 



Acetate + Malonyl-diCoA- 




Malonate 

ATP + CoA 



t 
Malonyl - CoA 



Acetyl - CoA 



>■ 



(1-8) 



The principal difference between the two schemes is the participation of 
malonyl-diCoA. Hayaishi's results do not exlude it but provide no evidence 
for it, while Wolfe et al. claim to have detected it chromatographically. 
Although the decarboxylation of malonate is characterized by AF = — 7 
kcal/mole, the activation energy is presumably so high that the more com- 
plex mechanisms above are necessary. The malonate decarboxylase and 



METABOLISM OF MALONATE 231 

CoA-transferase have been partially purified. The acetyl-CoA formed from 
malonate may enter the cycle directly or participate in other reactions, such 
as the formation of acetoacetate if the cycle is blocked by malonate, or 
transfer its CoA to malonate, or simply be hydrolyzed to acetate. It is very 
interesting that a lag period was noticed in the oxidation of malonate by 
cell-free extracts (Wolfe et al., 1955). Little oxygen uptake occurs with 
malonate until it is all decarboxylated; during the period of the most rapid 
CO2 evolution, the respiration is low. Yet acetate is activated and oxidized 
rapidly. One possible explanation is a block of the cycle by malonate so 
that acetyl-CoA cannot enter until most of the malonate has been me- 
tabolized. Another explanation involves a distribution of CoA in favor of 
the malonyl derivatives with little acetyl-CoA present during the active 
decarboxylation reaction. 

Cryptococcus terricolus can grow on malonate as the sole source of carbon 
and malonate stimulates the endogenous respiration and CO2 formation after 
a lag phase (Pedersen, 1963). It would appear that malonate is completely 
oxidized to CO2 and water, since the R.Q. in the presence of malonate is 
1.54. In other microorganisms where the oxidation is not complete, mal- 
onate may be incorporated into a variety of substances, especially lipids 
(Bu'Lock et al., 1962), although the rate of incorporation is seldom very 
rapid. 

Pathways of Malonate Metabolism In Plants and Animals 

Despite the widespread occurrence of malonate metabolism in plant tis- 
sues, little is known of the reactions involved, although perhaps they are 
not significantly different from those described for microorganisms. Peanut 
mitochondria supplemented with ATP, CoA, and other factors are able to 
oxidize malonate (Giovanelli and Stumpf, 1957). Incubation with malo- 
nate-2-C^* for 2 hr leads to the appearance of labeled citrate, malate, and 
succinate, indicating the sequence 

Malonate ->■ malonyl-CoA -> acetyl-CoA -> CO2 + HjO 

the last step occurring in the cycle. The participation of malonyl-CoA in 
the oxidation of propionate by peanut mitochondria is suggested by the 
tracing of the label from propionate- 1-C^* (Giovanelli and Stumpf, 1958). 
The following sequence involving malonic semialdehyde may be formulated: 

Propionate -> propionyl-CoA ->■ acrylyl-CoA -> |3-hydroxypropionyl-CoA -> 

^-hydroxypropionate -> malonic semialdehyde -> malonyl-CoA -> 

acetyl-CoA -> CO^ + H2O 

It is also possible that CoA derivatives are retained throughout, since a 
malonic semialdehyde-CoA dehydrogenase which catalyzes the formation 



232 1. MALONATE 

of malonyl-CoA has been found in Clostridium kluyveri (Vagelos, 1960). 
In bush bean {Phaseolus vulgaris) leaves, incubation with malonate-2-C^* 
leads to labeled citrate, isocitrate, and malate, indicating a pathway through 
acetyl-CoA and the cycle (Young and Shannon, 1959). Malonate is incor- 
porated in isolated spinach chloroplasts at about half the rate for acetate, 
much of the label appearing in lipids (Mudd and McManus, 1964). 

Metabolism of malonate was first described in the dog by Pohl (1896), 
who found that only a fraction of the malonate administered to the animals 
can be recovered in the urine. This subject was not taken up again until 
1950 and since that time much has been learned of how the body deals 
with malonate, and of the role of malonate and its derivatives in normal 
metabolism. It would appear from the limited data that about 30% of 
the administered malonate is metabolized (Lifson and Stolen, 1950; Busch 
and Potter, 1952 a). However, the rate of oxidation is relatively slow and 
in rabbits represents less than 2% of the total respiration (Wick et al., 
1956). The rate of oxidation may be in part limited by the transfer from 
the extracellular space into the tissues, since this is slow. 

Various mammalian tissues can decarboxylate malonate and utilize the 
acetate formed. This has been investigated most completely in rat tissue 
slices by Nakada et al. (1957), who determined the 0^*0, arising from mal- 
onate-1-C^^ during 1 hr incubation at pH 7.4, the total concentration of 
malonate being 5 xnM (see accompanying tabulation). Kidney, liver, and 



Tissue Added C^* as Qi^Oa (%) 

Kidney 27.0 

Liver 18.0 

Heart 15.2 

Diaphragm 6 . 8 

Spleen 1 . 7 

Brain 1 . 5 

Lung 1 . 

Testis 0.6 



heart are particularly active, and the tissues show a wide range of decar- 
boxylative ability, part of which may be due to different rates of penetra- 
tion. The variation of malonate oxidation with concentration is shown in 
Fig. 1-22 for rat kidney slices, and an inhibition of its own metabolism is 
seen at concentrations above 5 vaM. The inhibition of acetate- 1-C^* oxida- 
tion is shown for comparison. The relative rates of activation of malonate, 
succinate, and glutarate by several tissues were determined by measuring 
the rates of formation of hydroxamic acid from hydroxylamine during the 



METABOLISM OF MALONATE 



233 



incubations (see accompanying tabulation) (Menon et al., 1960). The results 
show not only differences between the tissues, but also that the activating 
system for malonate is different from that for succinate and glutarate. 



Tissue 


Hydroxamic 


acid formed (m/<moles/mg/30 min) 










Malonate 


Succinate 


Glutarate 


Dog heart 


10.3 


7.4 


28.4 


Dog muscle 


2.6 


8.2 


16.0 


Pigeon liver 


4.7 


20.5 


31.7 


Rat liver 


11.5 


62.9 


86.7 



The metabolic pathways for malonate in mammalian tissues appear to 
be very similar to those in bacteria. The oxidation of malonate requires 
ATP, coenzyme A, and Mg++ in extracts or homogenates of rat kidney 
(Nakada et al., 1957), human placenta (Hosoya and Kawada, 1958), and, 
incidentally, locust fat body (Tietz, 1961). However, the kinases, decarbox- 




FiG. 1-22. Effects of malonate concentration on the oxidation 

of malonate and acetate by rat kidney slices at pH 7.4 and 

37° during 2-hr incubation. (From Nakada et al., 1957.) 



ylases, and CoA-transferases for these reactions have not been studied. 
An important transfer of coenzyme A between acetoacetyl-CoA and mal- 
onate has been shown to occur in pig heart extracts (Beinert and Stansly, 
1953). Acetoacetyl-CoA is formed by the condensation of two acetyl-CoA 



234 1. MALONATE 

molecules, and various carboxylate anions can accept the coenzyme A so 
that acetoacetate is formed: 

2 Acetyl-CoA :^ acetoacetyl-CoA + CoA 
Acetoacetyl-CoA -|- malonate :;^ malonyl-CoA + acetoacetate 

succinate and butyrate being normally the most active. In this way mal- 
onate can give rise to acetoacetate as well as by its block of the cycle. 
Labeled malonate forms labeled acetoacetate in rat liver (Nakada et al., 
1957). This type of reaction has also been shown to occur in yeast, and dog 
heart and skeletal muscle (Menon and Stern, 1960). The enzyme, succinyl- 
/3-ketoacyl-CoA transferase, was purified from pig heart and catalyzes the 
transfer to both malonate and glutarate. The following reaction: 

Succinyl-CoA + malonate :^ malonyl-CoA + succinate 

also occurs. It is interesting to speculate that some of the succinyl-CoA 
formed in the oxidation of a-ketoglutarate transfers its coenzyme A to 
malonate when it is present; if the malonyl-CoA is not readily metabolized, 
this could deplete the a-ketoglutarate oxidase of coenzyme A and slow down 
the reaction. Condensation of malonyl-CoA with coenzyme A derivatives 
may be important in fatty acid synthesis. In pigeon liver and carrot roots 
there is an enzyme catalyzing the condensation of malonyl-CoA with either 
acetyl-CoA or butyryl-CoA; although the product is unknown, it was isolat- 
ed chromatographically (Steberl et al., 1960). This product can form pal- 
mitate with other enzyme fractions. When malonyl-2-C^^-CoA is incubated 
with extracts from various rat tissues, various Cig-Cig acids are formed, 
depending on the acyl-CoA acceptor used (Horning et al., 1960). One acyl- 
CoA unit is incorporated into long-chain fatty acids and the rest of the 
C-chain in supplied from malonyl-CoA. Labeled fatty acids are also formed 
from malonate- 1-C^* in particle suspensions of the locust fat body (Tietz, 
1961). A highly purified preparation from pigeon liver, which converts 
malonyl-CoA and acetyl-CoA to palmitate in the presence of NADPH, has 
been reported (Bressler and Wakil, 1961). In the absence of NADPH, mal- 
onyl-CoA and acetyl-CoA condense to form an unknown product (which 
is not acetoacetate, butyrate, or /^-hydroxybutyrate). The conversion of 
malonyl-CoA to fatty acids is perhaps mediated through such condensa- 
tions to Cg acids, forming butyryl-CoA, which would again condense with 
malonyl-CoA, and so lengthen the chain. 

A few words should be said about the pathways of methylmalonate me- 
tabolism since it has been recently found to be an important intermediate 
in fatty acid |metabolism, and malonate can interfere markedly with at 
least one of these reactions. Methylmalonate was shown to be an interme- 
diate in the metabolism of propionate in various tissues (Flavin et al., 1955; 
Katz and Chaikoff, 1955; Feller and Feist, 1957). In the course of this work 



INHIBITORS STRUCTURALLY RELATED TO MALONATE 235 

a novel reaction was discovered, namely, the interconversion of methyl- 
malonate and succinate. The enzyme, which has been called a methyl- 
malonyl-CoA isomerase, catalyzes the reaction: 

Methylmalonyl-CoA :^ succinyl-CoA 

and has been purified from ox liver (Stern and Freidman, 1960) and Pro- 
fionihacterium shermanii (Wood and Stjernhohn, 1961; Stjernholm and 
Wood, 1961). At equilibrium the ratio (succinyl-CoA)/(methylmalonyl-CoA) 
is 10.5 (pH 7 and 25°). This reaction functions in both the oxidation and 
formation of propionate. Another reaction of importance in propionate 
metabolism is 

Methylmalonyl-CoA + pyruvate :^ propionyl-CoA + oxalacetate 

It is catalyzed by methylmalonyl-oxalacetate transcarboxylase, whereby a 
carboxylation may be effected without the intervention of COg or the ex- 
penditure of energy to activate COg. It may finally be noted that the feed- 
ing of malonate to dogs leads to a marked excretion of methylmalonate in 
the urine (Thomas and Stalder, 1958). This does not necessarily mean that 
the methylmalonate is formed from malonate, since malonate inhibits the 
interconversion of methylmalonyl-CoA and succinyl-CoA very strongly 
(Flavin et al., 1955). 



INHIBITORS STRUCTURALLY RELATED TO MALONATE 

The inhibition of succinate dehydrogenase by various dicarboxylate ions 
was treated earlier (page 34). Maleate will be dealt with in Chapter III-2 
and oxalate will be discussed in Volume IV. Inhibitors related to malonate 
will also be found in the following chapter on analogs. It then remains to 
take up the esters of malonate, hydroxy malonate, and those compounds 
in which the carboxylate groups have been replaced by other anionic groups. 
As mentioned previously, there is often confusion in the nomenclature; we 
shall adopt the following (using the ethyl derivatives as examples): 

COO" Et^ /COO" 

Et— Hc:" . /C^ _ 

coo Et COO 

Ethylmalonate Diethylmalonate 



/COO /COO— Et 

HgC Hz^s, 

^COO— Et ^COO — Et 



Malonic Malonic 

monoethyl ester diethyl ester 



236 1. MALONATE 

Malonic Esters 

Malonic monoethyl and diethyl esters have been found to have some in- 
teresting actions. They can increase the survival period of mice infected 
with mycobacteria (Davies et al., 1956), are occasionally carcinostatic (Freed- 
lander et al., 1956), can inhibit the breakdown of hexobarbital in liver homo- 
genates and prolong the narcotic action (Kramer and Arrigoni-Martelli, 
1960), and inhibit sporulation of Bacillus cereus (Nakata and Halvorson, 
1960). However, the relation of these effects to succinate dehydrogenase 
inhibition is obscure. In most cases, malonic esters have been used to cir- 
cumvent the permeability barriers to malonate, inasmuch as the esters 
should penetrate into cells readily. It has often been assumed that hydrolysis 
to malonate occurs within the cells. This hydrolysis must be enzymatic 
because the esters are quite stable. A lipase from pig liver hydrolyzes one 
ethyl group from malonic diethyl ester but does not remove the other 
ethyl group, the product being malonic monoethyl ester (Christman and 
Lewis, 1921). I have been able to find no direct evidence for the hydrolysis 
to malonate. Malonic esters are neutral at physiological pH's, since they 
are very weak acids with pK„ values around 15.75 (Rumpf et al., 1955) 
and ionize very slowly with a rate constant of 1.8 X 10~^ min~^ (Pearson 
and Mills, 1950). 

The effects of the malonic esters on metabolism will now be discussed in 
order to determine if there is any indirect evidence for the intracellular 
hydrolysis to malonate and the inhibition of succinate dehydrogenase. 
When injected into fluoroacetate-poisoned rats, malonate and malonic di- 
ethyl ester have approximately the same effects on the accumulation of 
citrate in the heart and kidneys (Fawaz and Fawaz, 1954). Furthermore, in 
kidney slices, the diethyl ester inhibits succinate oxidation 92% at 20 mM, 
and malonate at the same concentrazion inhibits 75%. Less inhibition by 
the ester compared to malonate is observed in heart slices. Evidence for 
hydrolysis by both tissues was adduced from the decreases in the pH ob- 
served. The respiration of the fungus Zygorrhynchus moelleri is not inhibited 
readily by malonate although the succinate dehydrogenase from this orga- 
nisms is quite sensitive, indicating a failure to penetrate (Moses, 1955). 
Malonic diethyl ester was tested and found to inhibit the oxidation of both 
glucose and acetate, but at high concentrations (see tabulation). It was felt 



Malonic diethyl 


ester 


% Inhibition of: 


{mM) 


Glucose oxidation 


Acetate oxidation 


10 

30 

100 




Stim 49 
30 
96 


10 

76 
98 



INHIBITORS STRUCTURALLY RELATED TO MALONATE 237 

that the greater inhibition of acetate oxidation is evidence for the hydro- 
lysis of the ester and that the inhibition is exerted by malonate. Malonic 
diethyl ester was also used to facilitate penetration into Penicillium chryso- 
genum, since even 100 roM malonate does not effect acetate metabolism 
(Goldschmidt et al., 1956). The ester at 20 milf inibits the production of 
C^*02 from labeled acetate 75-85% and simultaneously decreases the in- 
corporation of C^* into cellular materials, the labeling of glutamate being 
particularly depressed. The utilization of acetate by Bacillus cereus is also 
interfered with by malonic diethyl ester, so that acetate and p>Tuvate ac- 
cumulate (Nakata and Halvorson, 1960). Malonate and the diethyl ester 
have been compared with respect to their effects on the respiration of 
Mycobacterium pJdei (Miiller et al., 1960). Malonate stimulates the endoge- 
nous respiration, presumably through its oxidation, and the ester stimulates 
even more potently at 1-10 raM, although at 100 mM the ester inhibits 
and malonate still stimulates. The respiration with glycerol as the substrate 
behaves similarly. Finally, malonic diethyl ester markedly stimulates the 
endogenous respiration of Chlorella vulgaris at 4-10 mM, but inhibits the 
oxidation of glucose and acetate, the latter more strongly (Merrett and 
Syrett, 1960). All of these results show that the diethyl ester is inhibitory 
but certainly do not constitute conclusive evidence for a hydrolysis to mal- 
onate. This is a subject that should be pursued further and more extensive 
tests should be made for enzymes hydrolyzing the esters. Until the intra- 
cellular hydrolysis can be established, the results obtained with the malonic 
esters cannot be interpreted. 

Hydroxy malonate (Tartronate) 

The substitution of a methylene hydrogen of malonate by any group 
seems to reduce rather strongly the ability to inhibit succinate dehydrogen- 
ase. Even the small hydroxyl group almost abolishes the inhibitory activity 
and this is evidence that the binding of malonate to the active center of 
succinate dehydrogenase must involve severe steric restrictions. Although 
tartronate does not inhibit succinate dehydrogenase, it has other actions 
of some interest. Quastel and Wooldridge (1928) found that 71.4 roM tar- 
tronate does not inhibit succinate oxidation at all whereas it inhibits the 
oxidation of lactate 90%, as measured by methylene blue reduction in tol- 
uene-treated E. coli. In fact, the marked differences in the effects of mal- 
onate and tartronate led Quastel and Wooldridge to postulate the specific 
structure of the active centers of enzymes. Tartronate inhibits the respira- 
tion of rat liver slices and is strongly ketogenic, acetoacetate being formed 
to a greater extent that with equivalent concentrations of malonate (Ed- 
son, 1936). Pig heart malate dehydrogenase is inhibited 24% by 60 mM 
tartronate (Green, 1936) but in pigeon liver extracts it is about 1000 times 
as effective, inhibiting malate oxidation competitively with a ^^ of 0.09 mM 



238 1. MALONATE 

(Scholefield, 1955). Tartronate is also a competitive inhibitor of the decar- 
boxylating malate dehydrogenase (malic enzyme) of pigeon liver with a K^ 
of 0.1 mM, the inhibition being stronger than with malonate (Stickland, 
1959 b). The carboxylation of pyruvate, catalyzed by the same enzyme, 
is also strongly inhibited, but there is a large noncompetitive element 
(Stickland, 1959 a). The inhibition of lactate and malate oxidations and 
decarboxylations is not surprising since tartronate is structurally similar. 
Although tartronate occurs in plant and animal tissues to the extent of 
8-15 //g/g wet weight of tissue, which would correspond to about 0.1 xnM 
in the tissue water (Veitch and Brierley, 1962), it is doubtful if it could 
exert a regulatory action on the metabolism. Tartronate is not metabolized 
in the rat and consequently is near 8 raM in the urine. 

The respiration of guinea pig brain slices is depressed by tartronate at 
concentrations of 67-75 rcvM (Jowett and Quastel, 1937). The degree of inhi- 
bition depends on the substrate provided and is maximal with lactate and 
minimal with pyruvate. Since lactate is probably metabolized through 
pyruvate, the inhibition here may be mainly on lactate dehydrogenase. 
However, the anaerobic breakdown of pyruvate and anaerobic glycolysis 
are also well inhibited. The respiration of Mycohacterium. phlei with lactate 
as substrate is inhibited 65% with 66 mM tartronate, although the endo- 
genous respiration is stimulated and the oxidation of glucose unaffected 
(Edson and Hunter, 1947). This relatively specific effect on lactate dehydro- 
genase was used by Fiume (1960) to inhibit aerobic glycolysis in tumor cells, 
inasmuch as lactate dehydrogenase is involved. It was postulated that 
aerobic glycolysis might be inhibited more strongly in the tumor than in 
normal tissues, depleting the ATP supply more severely. It was found that 
tartronate inhibits aerobic glycolysis of the Yoshida ascites hepatoma — 
26% at 10 mM, 34% at 20 mM, and 58% at 50 mM - but comparisons 
with normal tissue were not made. 

The inhibition of phosphatases should perhaps also be considered in 
work with tartronate, since prostatic acid phosphatase is inhibited com- 
petitively with a K^ of around 50 mM. The inhibition by tartronate is 
much greater than by malonate and about twice as potent as by ketomal- 
onate (Kilsheimer and Axelrod, 1957). 

Aminomalonate 

This substance was considered to be a possible substrate for the synthesis 
of S-aminolevulinate but was found upon examination to be a potent inhib- 
itor of S-aminolevulinate synthetase (Matthew and Neuberger, 1963). The 
inhibition of the enzyme from Rhodopseiidomonas spheroides and chicken 
erythrocytes is competitive; the K, for the bacterial enzyme is 0.0225 mM. 
Pyridoxal-P is a coenzyme in these systems and the inhibition depends on 
its concentration. Aminomalonate may be considered to be an analog of 



INHIBITORS STRUCTURALLY RELATED TO MALONATE 239 

glycine (it is carboxyglycine) and presumably inhibits 8-aminolevulinate 
synthetase by binding to the glycine site and complexing with pyridoxal-P 
as well. Aminomalonate condenses with aldehydes nonenzymatically in 
the presence of pyridoxal-P. Furthermore, other pyridoxal-P-dependent 
enzymes are inhibited, e.g. serine hydroxy methyltransferase, whereas en- 
zymes involved in glycine metabolism but not requiring pyridoxal-P are not 
inhibited. Aminomalonate can be formed in the tissues by transamination 
between ketomalonate and glutamate, and can be decarboxylated by an 
enzyme found in silkworm larvae and rat heart and liver to glycine. This 
derivative of malonate is a good illustration of how a simple change in the 
structure can create an inhibitor with quite different properties and inhib- 
itory spectrum. 

Substituted Malonates 

Although the alkylmalonates are not particularly interesting as inhibi- 
tors, there are two malonate derivatives that may warrant further investi- 
gation. Fluoromalonate was studied by Chari-Bitron (1961) on the principle 
that if malonate is metabolized through acetyl-CoA, fluoromalonate might 
follow the same pathway and enter the cycle as fluoroacetyl-CoA, produc- 
ing the same effects as fluoroacetate, namely, a block of the cycle at the 
aconitase step. The toxicity of fluoromalonate is a good deal less than 
fluoroacetate but the ester is as toxic in mice (see accompanying tabulation). 



LD50 (mg/kg) 

Animal 

Fluoromalonate Fluoromalonic diethyl ester Fluoracetate 



Mouse 


80 


Rat 


60 


Guinea pig 


2 



15 15 

70 5 

— 0.25 



Death is associated with a marked accumulation of citrate in the tissues 
and it differs strongly from malonate in this respect. Accumulation of cit- 
rate also occurs in kidney mitochondria with fluoromalonate at concentra- 
tions around 1 mM. It was further established that decarboxylation of 
fluoromalonate occurs in kidney preparations. Finally, fluoromalonate is 
only about one tenth as effective as malonate in the inhibition of succinate 
dehydrogenase. The results thus conform quite well to the predicted mech- 
anism. Difluoromalonate and its amide inhibit quite readily the oxida- 
tions of succinate and fumarate by Pseudomoyias (around 70% inhibition 
at 0.7 mM), but there is no inhibition of succinate dehydrogenase in soni- 
cates; the mechanism is unknown (Bernheim, 1963). 



240 1. MALONATE 

Inasmuch as the active center of succinate dehydrogenase possesses a 
sulfhydryl group close to the cationic binding sites, this being the basis 
for the inhibition by mercurials and other sulfhydryl agents, the possibility 
of combining a sulfhydryl inhibition with malonate presents itself. Mercuri- 
malonamide and mercurimalonic diethyl ester have been prepared (Naik 
and Patel, 1932) and these compounds, or particularly the hydrolyzed 

^COO-Et 
^C(X>-Et 

forms if they are stable, might be interesting to examine as inhibitors of 
succinate dehydrogenase. 

Acetylene-dicarboxylate and Propane-tricarboxylate 

Succinate dehydrogenase is inhibited competitively by acetylene-dicar- 
boxylate (Dietrich et at., 1952). The order of addition of the succinate and 
acetylene-dicarboxylate is important; for example, if acetylene-dicarboxy- 
late is added 20 min after the succinate, the rate of oxygen uptake de- 
creases slowly and does not become equal to that observed when the suc- 
cinate is added after the inhibitor until 5 hr (Thomson, 1959). Acetylene- 
dicarboxylate is about as effective as malonate on long incubation with the 
enzyme. The constants obtained on rat kidney succinate dehydrogenase 
are ir,„ = 4.12 mM and K^ = 0.81 roM when substrate and inhibitor are 
added together; after 18 hr incubation with the inhibitor, iC, =^ 0.171 vaM. 
It is difficult to understand why the rate of inhibition is so slow. Succinate 
dehydrogenase from pig heart is less readily inhibited, K^ being 1.4 mM 
with A'^-methylphenazine as electron acceptor and 16.5 when ferricyanide is 
the acceptor (Hellerman et al., 1960). Possibly insufficient time for equilib- 
rium was allowed. The inhibitions of succinate and pyruvate oxidations 
by acetylene-dicarboxylate in suspensions of rate heart mitochondria are 
shown in Fig. 1-23 (Montgomery and Webb, 1956 b). Acetylene-dicarboxy- 
late is less potent than malonate against succinate oxidation and more po- 
tent against pyruvate oxidation, indicating that an inhibition is exerted 
at some other point in the cycle. 

Propane-tricarboxylate is a rather weak inhibitor of succinate dehydro- 
genase from rat heart, 5 mM inhibiting around 30% when the succinate is 
also 5 mM. The oxidation of a-ketoglutarate is inhibited 60% under the 
same conditions, suggesting some effect on the a-ketoglutarate oxidase. 
The oxidation of pyruvate is inhibited even more strongly (Fig. 1-24). It 
might be thought that propane-tricarboxylate would inhibit aconitase or 
isocitrate dehydrogenase, but very little inhibition is noted, either of cit- 
rate oxidation or the ability of citrate to function as a source of oxalac- 
etate for the oxidation of pyruvate. 



INHIBITORS STRUCTURALLY RELATED TO MALONATE 



241 



Traws-l,2-Cyclopentane-dicarboxylate inhibits the succinate dehydroge- 
nase of Tetrahymena geleii homogenates, around 50% inhibition being ob- 
served when the inhibitor/substrate ratio is 0.5 (Seaman and Houlihan, 
1950). On the other hand, in intact cells this substance increases the utili- 
zation of pyruvate, acetate, and succinate. The uptake of acetate is in- 



80 


- 


Pyruvote + Malate 


^^ 


%60 
INH 


- 




/ 


40 


■ 




/ 


20 


- 


1 1 i 


Succinate 



I 5 10 50 

(Acetylene - Dicarboxylate) 

Fig. 1-23. Effects of acetylene-dicarboxylate (in milf) 

on the oxidations of p>Tuvate -f malate and succinate 

by rat heart mitochondria. (From Montgomery and 

Webb, 1956 b.) 

creased as much as 50% by ^ra??5-l,2-cyclopentane-dicarboxylate and in 
its presence succinate is taken up and oxidized whereas succinate does not 
enter the cells normally. It was postulated that trans-1, 2 cyclopex\':ine- 
dicarboxylate increases the permeability of the membrane to these sub- 
strates, possibly by an action on the metabolic systems involved in the 




)5 I 2 5 

( Propone - Tncarboxylote) 

Fig. 1-24. Effect of propane -tricarboxylate (in 

inM) on the oxidation of pyruvate + malate by 

rat heart mitochondria. (From Montgomery and 

Webb, 1956 b.) 



242 1. MALONATE 

inward transport. It is not known if such an effect is observed with other 
dicarboxylate anions. 

Sulfonate, Phosphonate, and Arsenate Analogs of Succinate 

Succinate and malonate are bound to the active center of succinate de- 
hydrogenase in part through electrostatic forces involving the negatively 
charged carboxylate groups. It might be expected that substances in which 
the carboxylate groups are replaced by other anionic groups would also be 
inhibitory. The structures for some of the compounds and the intercharge 
distances are given in Table 1-1. Klotz and Tietze (1947) first demonstrated 
the inhibition of succinate dehydrogenase (rat liver) by 1,2-ethanedisulfo- 
na'te and /5-sulfopropionate, 50% inhibition being observed in both cases 
when the inhibitor is 13 mM and succinate is 20 mM, these inhibitions 
being approximately equivalent to those of malonate. Intact E. coli cells 
are not affected by 1,2-ethanedisuIfonate but the oxidation of succinate 
by cell-free preparations is well inhibited, corresponding to the results with 
malonate (Ajl and Werkman, 1948). Some inhibitor constants for these 
substances are shown in Table 1-29. 

There appears to be definite differences in the susceptibility of the suc- 
cinate dehydrogenases from different sources. Seaman (1952) noted that 
/?-phosphonopropionate inhibits the Tetrahymena enzyme more strongly 
than does malonate, and yet no inhibition is observed on the enzymes 
from rat heart, liver, brain, or muscle. The inhibitions are competitive 
wherever they have been tested and there is no evidence that the mecha- 
nism is in any way different from that of malonate. Since these groups are 
larger than the carboxylate group, the interchange distances are greater 
than in malonate or succinate, and this must play some role in determining 
their binding to the enzyme. However, the distances for the malonate ana- 
logs, methanedisuffonate and arsonoacetate, are between those for malon- 
ate and succinate, so that binding should be as tight as for these latter 
substances if this were the only factor. Another factor of importance is the 
total net charge on these inhibitors, inasmuch as each of the sulfonate, 
phosphonate, and arsonate groups can ionize more than once. In other 
words, a disulfonate can exist in five different forms with charges 0,-1, 
— 2, —3, and —4. Furthermore, the third and fourth ionization constants 
are in the physiological range of pH (see accompanying tabulation). The 



vK„ vK„ 



Arsonoacetate 


7.7 


— 


1 ,2-Ethanediphosphonate 


6.84 


8.17 


1 ,4-Butanediphosphonate 


7.28 


9.05 


Pyrophosphate 


5.69 


7.76 



INHIBITORS STRUCTURALLY RELATED TO MALONATE 



243 



Table 1-29 

Inhibitor Constants for Sulfonate, Phosphonate, and Arsonate Analogs of 

Substrates " 



Inhibitor 


Preparation 


K, 


Reference 


Malonate 


Tetrahymena succinate 


6.67 


Seaman (1952) 


/S-Phosphonopropionate 


dehydrogenase 


1.51 




Arsonoacetate 




9.25 




1,2-Ethanedisulfonate 


Rat liver succinate 


2.73 


Klotz and Tietze 


/?- Sulfopropionate 


dehydrogenase 


3.69 


(1947) 


Malonate 


Mouse liver succinate 


0.19 


Tietze and Klotz 


1 ,2-Ethanedisulfonate 


dehydrogenase 


26.5 


(19.52) 


o-Sulfobenzoate 




11.7 




/5-Phosphonopropionate 




No inh 




Arsonoacetate 




No inh 




Methionate 




20.6 




Malonate 


Beef heart succinate 


0.014 


Rosen and Klotz 


Methanediphosphonate 


dehydrogenase 


0.5 


(1957) 


Phosphonoacetate 




0.15 




Pyrophosphate 




0.0011 




Hs^pophosphate 




0.14 




1,2-Ethanediphosphonate 




9.6 




1,4-Butanediphosphonate 




8.9 




Arsonoacetate 




15.3 




Malonate 


Fumarase 


40 


Massey (1953b) 


a-Hydroxy-/3-sulfo- 








propionate 




16.5 





" The A', values for rat and mouse liver succinate dehydrogenases were cal- 
culated assuming A'^ = 6 X 10~^ M, and for beef heart succinate dehydrogenase 
K^ = 4 X 10"* M. Although these values are undoubtedly inaccurate, the Aj values 
so calculated are useful for comparisons since within each experiment the relative 
values are reliable. 



rather poor inhibition produced by the arsono and phosphono derivatives 
was postulated by Tietze and Klotz (1952) as possibly due to the extra 
negative charge carried bj^ these substances. In the more recent work of 
Rosen and Klotz (1957), the ionization was taken into account and the 
inhibitor constants calculated on the basis of the concentration of doubly 
charged anion present. Evidence that too high a charge reduces the inhib- 
itory potency is provided by the falling off of the inhibition as the pH is 
raised. On the other hand, it is difficult to understand why there should 



244 1- MALONATE 

be less affinity between cationic groups on the enzyme and doubly charged 
anions, compared to singly charged anions, and one cannot help but wonder 
if the alterations in pH in the experiments of Kosen and Klotz were not 
affecting the ionizable groups on the enzyme. These workers suggested that 
the binding of these substances involves an iron ion on the enzyme and 
correlated affinities with the ability to chelate iron. This is certainly a type 
of binding that should be kept in mind, but it is difficult to reconcile with 
the fact that iron-chelating agents such as 1,10-phenanthroline and 2,2'-bi- 
pyridine do not interfere with the binding of succinate to the dehydrogenase 
even though their attachment to the enzyme can be demonstrated spectro- 
scopically. 

An interesting succinate dehydrogenase inhibitor, not fundamentally re- 
lated to the compounds previously discussed, is 3-nitropropionate (hiptage- 
nate), found to be the toxic principle of Indigofera endecaphylla (Morris 
et al., 1954). It inhibits succinate dehydrogenase competitively with K^ == 
0.19 n\M (Hylin and Matsumoto, 1964,) although no inhibition of the 
enzyme is found after administration of the substance to animals, so that 
it is not possible to correlate the toxicity with an effect on this enzyme. It 
is rather surprising that 3-nitropropionate binds so well to succinate de- 
hydrogenase, lacking two anionic groups, and it would be interesting to 
know more of the nature of the interaction of the nitro group with the 
enzyme. 

With respect to inhibitors related to malonate in one way or another, 
it must be concluded that none possesses particular advantages over mal- 
onate for the specific inhibition of succinate dehydrogenase, although cer- 
tain derivatives have interesting properties and metabolic actions. Progress 
could be made by the finding of forms of malonate, or related inhibitors, 
that are uncharged, reasonably stable outside the cell, and easily split 
to the active inhibitor intracellularly. It would also be valuable to have an 
inhibitor which would initially bind to the succinate dehydrogenase at the 
substrate site, because of its complementary configuration, and then react 
chemically with an adjacent group, so that the inhibition of this enzyme 
would be not only specific but slowly reversible. 



CHAPTER 2 

ANALOGS OF ENZYME REACTION 
COMPONENTS 



An enzyme-catalyzed reaction involves the combination of the com- 
ponents with specific sites on the apoenzyme protein surface, these areas 
possessing the particular molecular configuration and the electrical field 
distribution required for the attachment and the electronic displacements 
characterizing the activated complex. If one of these components is modified 
in any way, its behavior in the system will usually be altered, due primarily 
to the new pattern of interaction between the modified component and 
the enzyme. The development and study of such analogs of substrates and 
coenzymes have been very active fields during the past few years for several 
reasons. First, the determination of the relative aflfinities of analogs that 
are substrates or inhibitors for enzymes is one of the most effective means 
for analyzing the topography of active centers and establishing the types 
of interaction involved in the catalysis. Second, it is hoped that inhibitors 
more specific for blocking certain enzymes than the inhibitors previously 
available will be found and this has been justified to a certain extent. 
Third, it has been realized that analog inhibition has direct bearing on the 
important phenomenon of feedback control of metabolic sequences and on 
the general regulation of cellular metabolism. Last, it is anticipated that 
the use of proper analogs may be useful in the specific correction of certain 
abnormal metabolic patterns and growth processes, such as occur in here- 
ditary enzyme defects or neoplastic changes. 

The previous chapter is concerned with malonate, a classic analog inhib- 
itor, and in the present chapter it is proposed to extend this principle to 
a variety of enzymes in order to establish some basic concepts. There are 
a number of inhibitors which act, at least occasionally, because they are 
structurally related to some enzyme reaction component, but which for 
various reasons will be discussed in separate chapters. Such are carbon 
monoxide, fluoroacetate and fluorocitrate, parapyruvate, arsenate, pyro- 
phosphate, monoamine oxidase inhibitors, certain inhibitors of cholinester- 
ase, and various drugs. Furthermore, it is necessary to point out that no 
attempt will be made to review the vast literature on the depression of 

245 



246 2. ANALOGS OF ENZYME REACTION COMPONENTS 

growth and proliferation exerted by many analogs, inasmuch as my pur- 
pose here is to restrict the discussion to enzymic and metabolic levels. 

TERMINOLOGY 

The term analog is defined very broadly as any substance that is in some 
way structurally related to a substrate, coenzyme, or cofactor.* It may 
either participate in the enzyme reaction to a greater or lesser extent than 
the normal components, or inhibit the reaction by interfering with the 
functioning of these normal components. The commonly used term anti- 
metabolite generally implies that the substance is a biologically abnormal 
compound synthesized in the laboratory and capable of interfering with the 
reactions of some cellular metabolite. We shall in this chapter frequently 
be concerned with inhibitions produced by substances such as carbohy- 
drates, amino acids, purines, and nucleotides, which naturally occur in most 
cells, and thus the more general term analog is preferred. The use of the 
term isostere has generally been restricted to a substance produced by the 
substitution of an atom or group in the normal compound by another atom 
or group with similar electronic or steric properties. A homolofj is a member 
of a series in which some part or property of a basic chemical type is progres- 
sively varied, as is the case when an aliphatic chain is lengthened by adding 
successive methylene groups. Most of the analogs to be discussed therefore 
fall in to one or more of these latter categories, but it is felt that there is 
little benefit to be derived from using these more specific terms. 

POSSIBLE SITES AND MECHANISMS OF INHIBITION 

The most common mechanism of inhibition is a competition between 
the analog and the normal reactant for a specific site on the enzyme surface. 
However, the frequently made assumption that this is the only mechanism 
involved is often unjustified. Other mechanisms which should be borne in 
mind will be mentioned here; they will be illustrated and discussed in greater 
detail later in the chapter. Let us first consider the mechanisms which may 
apply particularly to the inhibition of pure enzymes. 

(A) Binding of the analog to the enzyme sites for substrate, coenzyme, 
or activator by interactions which are at least in part those involved in 
the binding of the normal reactant and which allow reversibility. 

(B) An irreversible, or practically irreversible, reaction with the enzyme 

* The definition of analog must be imprecise because it is impossible to limit accu- 
rately how much structural deviation can occur before the derivative can no longer 
be thought of as related to the parent compound. 



POSSIBLE SITES AND MECHANISMS OF INHIBITION 247 

site subsequent to binding. Inasmuch as substrates occasionally form a 
temporary chemical bond with the enzyme, an analog may do likewise but 
fail to complete the reaction, remaining chemically attached to the site. 

(C) Binding of the analog to an enzyme site other than that with which 
the normal reactant interacts. Such binding may be simply fortuitous or 
the site may be specifically for the purpose of allowing feedback inhibition 
by a product formed in the sequence in which the enzyme participates. 
Regions outside the catalytic areas with which inhibitors can react are of- 
ten called allosteric sites. 

(D) The analog may be a substrate of the enzyme and will inhibit the 
reaction of the normal substrate to a degree dependent on the relative 
binding affinities and reaction rates. 

{E) Binding of the analog to a complex of the enzyme with the normal 
substrate, coenzyme, or activator. 

(F) The formation of a molecular complex of the analog with the normal 
reactant as a result of their structural complementarity. Although such 
complexes are probably uncommon and have seldom been considered in 
work with analogs, we shall see that examples of this mechanism are known. 

(G) Inhibition by a mechanism only indirectly related, or completely 
unrelated, to the structural similarity of the analog to the normal substra,te. 
An analog may, for example, possess chelating properties not exhibited 
by the substrate, or it may react with SH or carbonyl groups. 

When one is investigating more complex systems, particularly cellular 
preparations, a number of other mechanisms for analog inhibition may be 
proposed, and these should be added to the above list. 

(H) The analog may interfere with the transport mechanism by which the 
normal substance is taken through the cell membrane, since the two sub- 
stances may both combine with some membrane carrier or enzyme system 
required for efficient transport or accumulation. 

(/) The analog may not be the actual inhibitor, but may be transformed 
through a metabolic sequence into a substance which blocks a later reaction, 
a process frequently termed lethal synthesis. In certain instances the analog 
may complete a long and complex metabolic journey to terminate as a 
component in some important cellular product. The incorporation of pyrim- 
idine and purine analogs (e.g., the 5-halouracils, 2-thiouracil, 8-azagua- 
nine, and 8-azathymine into RNA and DNA) and amino acid analogs 
(e.g., tryptazan, 7-azatryptophan, ethionine, p-fluorophenylalanine, and 
/5-2-thienylalanine into proteins) has been frequently demonstrated. The 
products containing the analogs may be so abnormal as to fail to function 
properly in the cells, thereby producing far-reaching and complex dis- 
turbances. 



248 2. ANALOGS OF ENZYME REACTION COMPONENTS 

(J) The analog may act not on the enzyme attacking the normal sub- 
strate but on an enzyme involved in the formation of this substrate, since 
the precursors of the substrate will usually be structurally similar to it. 
In the linear sequence: 

E, Ej 
X -^S-> P 

an analog of S may have been designed to inhibit Eg but actually in cellular 
metabolism acts primarily on Ej to reduce the rate of formation of P. 



KINETICS OF ANALOG INHIBITION 

The kinetics of competitive inhibition have been presented in Chapter 
1-3, and the graphical analyses for the proof of competition and the deter- 
mination of the constants discussed in Chapter 1-5. Type A plots of l/v 
against 1/(S) have almost invariably been used to demonstrate the types 
of inhibition produced by analogs, but other types of plotting may be more 
satisfactory in certain situations, especially when the inhibition is not 
clearly and completely competitive. It is my opinion that many kinetic 
analyses of inhibition would be improved if several types of plotting proce- 
dure were used, allowing comparison of the results and more accurate cal- 
culations of the constants. 

It is quite often the case that the inhibition by an analog is not, by the 
usual methods of analysis, competitive with the substrate or coenzyme 
to which the analog is structurally related, and such results have puzzled 
many workers. The plotting may indicate noncompetitive, coupling, mixed, 
or indeterminable inhibition mechanisms. It has been pointed out (Chap- 
ter 1-3) that true noncompetitive inhibition must be rather rare and this 
is particularly true for inhibition by analogs. Coupling or uncompetitive 
inhibition is perhaps more common among analogs than with other in- 
hibitors, especially with regard to coenzymes or cofactors, inasmuch as 
the substrate combines with enzyme-coenzyme or enzyme-cofactor com- 
plexes in many reactions. It is important to determine in any case if the 
inhibition is really competitive and the kinetics modified to obscure this, 
or whether the mechanism is actually other than competitive, in those 
instances in which the graphical analysis does not demonstrate the typical 
and expected competitive picture. 

There are several reasons why an analog inhibition may not turn out to 
be competitive by the usual plotting procedures, and it may be useful to 
list them at this point. 

(A) The analog may be acting by some mechanism other than specific 
attachment to an active site on the enzyme, that is, by one of the mecha- 
nisms listed in the previous section. 



KINETICS OF ANALOG INHIBITION 249 

(B) The analog may be bound very tightly to the enzyme, in which case 
the order of addition of the substrate and the analog may be of great im- 
portance. If the analog is added to the enzyme and the mixture is incubated 
before the substrate is added, the inhibition may be very marked and the 
substrate may be unable to displace the inhibitor from the enzyme in a 
reasonable time. On the other hand, if both substrate and analog are added 
together, the inhibition may be very low initially but progress slowly, due 
to the relatively small fraction of the active centers available for reaction 
with the inhibitor. In either case secondary changes in the enzyme may 
occur and complicate the kinetics (Chapter 1-12). The basic problem in 
the interpretation of such inhibitions is the inability experimentally to 
achieve satisfactory equilibrium conditions, and it must be stressed that 
the usual inhibition equations and plotting procedures apply only to equi- 
librium conditions. There has been confusion between noncompetitive and 
irreversible inhibitions. The arsenicals, the mercurials, and iodoacetate, 
for example, are often thought to inhibit succinate dehydrogenase non- 
competitively, and yet succinate and these inhibitors react at the same site 
on the enzyme, as shown by the protection afforded by the presence of 
succinate when it is added with the inhibitors. These inhibitions are indeed 
competitive under certain conditions and during specific time intervals, 
but once the inhibition has been established it is difficult to demonstrate 
a competitive effect. Several examples of this situation using analog inhi- 
bitors will be encountered in this chapter. 

(C) The concentration of free inhibitor may be depleted due to its com- 
bination with the enzyme or other materials and one is then dealing with 
a mutual depletion system (Chapter 1-3). The quantitative aspects of com- 
petition may be modified quite markedly in such cases. This behavior must 
be looked for particularly when one is working with very potent analog 
inhibitors. 

(D) An irreversible or semi-irreversible change in the configuration or 
properties of the active center may occur following reaction with the inhibi- 
tor, so that even after dissociation of the inhibitor from the enzyme the 
affinity of the enzyme for the substrate is altered. The active centers of 
certain enzymes appear to be flexible and adapt in some way to the in- 
teracting molecules, and it is possible that such a change would not be readily 
reversible. The active center, or at least the immediately adjacent region, 
of the penicillinase of Bacilltis cerevs is altered by combination with the 
competitive analog of benzylpenicillin, in that there is a marked increase 
in the sensitivity to iodination, and this is prevented by the presence of 
substrate (Citri and Garber, 1961). Although this alteration is presumably 
reversible, since the ability to hydrolyze benzylpenicillin after removal of 
the inhibitor is unimpaired, it is easy to imagine changes only slowly re- 
versible. 



250 2. ANALOGS OF ENZYME KE ACTION COMPONENTS 

(E) 111 cellular systems, or possibly in subcellular preparations of some 
structural complexity, the failure of the analog to penetrate to the site of 
inhibition as readily as the substrate would obviously distort the kinetics, 
although the inhibition itself is truly competitive. 

{F) If the substrate possesses several groups through which it is bound 
to the enzyme, an analog which has only one of these groups (perhaps an 
analog comprising only a part of the normal substrate molecule) may block 
off one enzyme attachment point but allow the substrate to bind through 
the remaining groups. In many cases this will prevent catalysis, but in 
others it could only reduce the rate at which the substrate is reacted. 
Simultaneously there will be a reduction in the affinity of the enzyme for 
the substrate. The plots will give evidence of a mixed inhibition, which is 
the actual situation, but nevertheless competition of a type is occurring. 

Analogs of coenzymes often present a special problem in this connection 
since the coenzyme may be bound quite tightly to the apoenzyme. The 
addition of analog to the complete enzyme may not result in significant 
displacement of the coenzyme and little inhibition will be observed. How- 
ever, if the enzyme is resolved into its components by dissociating the coen- 
zyme in some manner, competition between the coenzyme and analog may 
be demonstrated in recombination experiments in which the analog reduces 
the ability of added coenzyme to reactivate the enzyme. 

The type of inhibition may depend on the concentration of the analog. 
It has been noted several times that an inhibition may be competitive at 
low analog concentrations and partially or completely noncompetitive at 
higher concentrations. The noncompetitive elements of the inhibition 
probably reflect the increasing unselectivity of action that is a common 
property of all inhibitors when the concentration is raised beyond a cer- 
tain level. 

An analog of a substrate will occasionally undergo reaction in the presence 
of the enzyme and, since both substrate and analog bind at the same site 
on the enzyme, competition will occur. The behavior in such a situation 
was discussed in Chapter 1-3 (page 96). The inhibition observed will depend 
on what is determined. If we designate the substrate by S and its analog 
by S': 

E + S ^ES— >E + P (2-1) 

E + S' ;?t ES' ^ E + P' (2-2) 

and the individual rate expressions may be written as: 

F„.(S) 
{S)+K, 1 + Vv 

A, 



KINETICS OF ANALOG INHIBITION 251 

F.'(S') 



(S') + K, 



1+^ 



(2-4) 



If the disappearance of S or the formation of only P is measured, the inhi- 
bition by S' will be typically competitive, but the inhibitor constant de- 
termined by the plotting procedures, K^,, wiU not necessarily be a true dis- 
sociation constant. The result wiU depend on the ratio kg.jkg, which in 
most cases will be considerably less than unity. If P and P' are identical 
and determined together, the inhibition wiU be less than in the previous 
case. It is important in work with analogs to establish whether they are 
catalytically reacted and, if so, to plan the inhibition experiments accord- 
ingly. It is also necessary in many instances to determine initial rates, since 
the concentration of the analog may be reduced significantly because of 
its conversion to the product. 

The product, or one of the products, of the enzyme-catalyzed reaction 
may generally be considered as an analog of the substrate, or of part of 
the substrate molecule. Inhibition by products was taken up in Chapter 
1-4 (page 140) and it was pointed out there that the inhibition is not 
necessarily competitive, since the product can react with the enzyme or 
other components of the reaction in a variety of ways. Several instances 
of product inhibition will be encountered in this chapter and in none of 
these is the inhibition due to a simple reversal of the forward reaction, 
but to actual combination with the enzyme at or near the active center. 

A somewhat more complex situation, in which the analog is transformed 
into a product that inhibits a subsequent reaction of the substrate, is fairly 
common and warrants some discussion although a complete kinetic analysis 
is difficult. The simplest system may be represented by: 

A --^ B A C (2-5) 

E. t 

A' — ^ B' (2-6) 

The substrate, A, and its analog, A', both are reacted by E^ and the analog 
product, B', inhibits E,. There are several ways in which the rate can vary 
with time and the behavior of the system will depend on many factors. 
In the first place, the presence of A' may slow down reaction 1 whereby 
A is transformed to B, this being a case of competing substrates. The for- 
mation of C may thus be initially slowed for this reason. The concentra- 
tion of B' will progressively rise and the inhibition on Ej increase. However, 
the concentration of A' will decrease and the competition w^th A be reduced, 
leading to a relative acceleration of reaction 1. The change in the rate of 
formation of C will depend on the balance between these two types of 
inhibition. For example, if A' is reacted fairly rapidly but B' is not a very 



252 2. ANALOGS OF ENZYME REACTION COMPONENTS 

potent inhibitor of Eg, the rate may first rise as the inhibition on E^ is 
released, and then fall later when the concentration of B' rises sufficiently. 
On the other hand, if A' is not readily depleted and B' inhibits well, the in- 
hibition will steadily increase. It is clear that the kinetics may not indicate 
competitive inhibition when the rate of formation of C is measured. The 
concentration of the intermediate B may rise and fall in a complex manner, 
as discussed in Chapter 1-9 (page 438). This is actually the simplest case 
of lethal synthesis complicated by competition in the initial reaction. 



MEANS OF EXPRESSING RESULTS 

The results in the study of analogs have usually been given in terms of 
inhibitions at different concentrations of substrate and analog.* Such 
results are often valuable but rather difficult to interpret, especially when 
different analogs at different concentrations must be compared. It would 
be more valuable if inhibitor constants were calculated and presented, 
along with Michaelis or substrate constants, and fortunately this practice 
is becoming more common. With values of K,„ and K^ it is possible to de- 
termine the inhibition expected at any combination of substrate and analog 
concentrations. Furthermore, it is possible to calculate relative interaction 
energies from a series of K-q obtained for a group of analogs, and thereby 
put the inhibition on a more molecularly interpretable basis. 

There are fundamentally two types of /i",. The actual dissociation constant 
for the EI complex may be called the true inJiihitor constant and refers to 
a particular free and active form of the inhibitor. The experimentally 
determined K^, on the other hand, often differs from the true Ki and may 
be termed the apparent inhibitor constant. This apparent K^ may depend 
on a number of factors, the most important of which is usually the pH 
since many inhibitors are weak acids or bases. This problem has been 
discussed in some detail in Chapter 1-14. If the inhibitor is a weak acid 
only one form may be the inhibitor, say the ionized I~ form, in which case 
the apparent K^ will vary with pH in the range around the p^^ of the 
inhibitor. The true K^, which refers to the equilibrium 

E + I- ^ EI- 

will not vary with the pH because of changes in the ionization of the inhi- 
bitor, although it may for other reasons. The apparent K^, in fact, will 
depend on any type of equilibrium between active and inactive forms of 
the inhibitor, for example an enol-keto isomerism, or on the binding of 
the inhibitor to nonenzyme components of the preparation. The Kj's given 

* The substrate concentration has been omitted in some reports and this vitiates 
the results and makes them quantitatively meaningless. 



MEANS OF EXPRESSING RESULTS 253 

in this chapter will in almost all cases be apparent inhibitor constants be- 
cause the true K/s have generally not been calculated in the reports, 
although in many instances they must be very close to the true K/s because 
of the nature of the analog or the conditions of the experiments. I have 
taken the liberty in certain cases where the data are adequate of calculating 
the K/s from the inhibitions reported. In every instance where plotting 
procedures have been used to determine the type of inhibition, it would 
have been possible to determine the appropriate constants, but these con- 
stants have been seldom reported. 

An especially unsatisfactory means of expressing the results is to give the 
inhibition for a certain ratio of inhibitor to substrate concentrations, 
(I)/(S), which has often been called the inhibition index. If the actual con- 
centrations are not given, it is not possible to visualize the inhibition quan- 
titatively, because (I)/(S) is not constant for a certain degree of inhibition 
even when the inhibition is competitive. This has been clearly pointed out 
in Chapter 1-3 (page 106) but it is important to re-emphasize it here. The 
ratio, (I)/(S), is constant and meaningful only at high substrate concen- 
trations which saturate the enzyme. This may be seen from the following 
expression, obtained by rearranging the equation for competitive inhibition: 






1 1 

~k7 ^'Wi 



1 



(2-7) 



It has sometimes been assumed tliat for 50% inhibition, (I)/(S) = KJKg, 
but this is not necessarily true, which can be easily seen by rewriting Eq. 
2-7 for i = 0.5 and (S) = nK,: 

(I) 



l:[v + ^l 



, -. (2-8) 

(S) Ks ' ' 

(I)/(S) = KJK^ only when the substrate concentration is high relative to 
Kf. (i.e., when n is much greater than unity). This can also be seen in another 
way: If K, = 1, K^ = 0.3, and (I)/(S) is kept constant at 0.3, the inhibition 
will vary with the absolute magnitudes of the concentrations as shown in 
the following tabulation: 



(S) 


(I) 


i 


0.1 


0.03 


0.083 


0.3 


0.09 


0.19 


0.5 


0.15 


0.25 


1 


0.3 


0.33 


2 


0.6 


0.40 


5 


1.5 


0.45 


10 


3 


0.48 


20 


6 


0.49 



254 2. ANALOGS OF ENZYME REACTION COMPONENTS 

It is obvious that a statement that the inhibition was a certain value at 
(I)/(S) = 0.3 would be of little significance. 

When a series of analogs is tested, quantitative expression of the relative 
affinities of the enzyme for the various analogs is desirable when possible. 
The term affinity implies no units and has a vague meaning, having been 
used in a variety of ways. A common method of expressing the affinity 
is to equate it tolIK^ in the case of an inhibitor, and to HKg for a substrate. 
The ratio of the affinities of an inhibitor and a substrate is thus often ex- 
pressed as KJKi. If K, = 1 mM, and K, = 0.001 mM, it would be stated 
that the affinity of the enzyme for the inhibitor is 1000 times that for the 
substrate. This may sound dramatic but is misleading in a way. It would 
seem that affinity might be better expressed in terms of binding energies. 
The ratio of the free energies of binding of inhibitor and substrate is 
given by: 



AF, vKs 



(2-9) 



In the hypothetical case above, AFJAF^ = 2, and the inhibitor is bound 
twice as tightly as the substrate, which is a more reasonable way of desig- 
nating the relative affinities. 

One might consider three types of AF for the binding of an inhibitor to 
an enzyme. There is the true AF corresponding to the true K^^ and an ap- 
parent AF corresponding to the apparent experimental K^. In addition 
there is the theoretical AF corresponding to the interaction energy of the 
enzyme and inhibitor in a vacuum, uncomplicated by solvent and ions. In 
order to compare different analogs with respect to their affinities for the 
enzyme, it is actually this last AF one would wish in most cases, but it is 
impossible to obtain. Lacking this, one must use the true AF values, but, 
as pointed out above, true K/s are not often available. It may be quite 
misleading to compare calculated AF values for a series of analogs if these 
analogs have different p-ff^'s and the experimental pH is in the region of 
these p-fi'./s. The differences in the AF's may reflect mainly the different 
degrees of ionization rather than the differences in binding energies of the 
inhibitory forms. In several instances in this chapter, I have calculated the 
AF values for such series of analogs and it must be remembered that the 
validity of comparing these values is sometimes questionable. It is some- 
times impossible to calculate even an apparent AF and one must be satisfied 
with values that are relative to some chosen compound; these will be called 
relative AF values. Although the AF itself may not be meaningful, occa- 
sionally the difference of the AF values for two inhibitors will be significant 
in attributing interaction energies to certain groups, as discussed in Chapter 
1-6 (page 268), since all of the other factors involving the solvent and ionic 
atmosphere may remain relatively constant for the inhibitors. The relative 
binding energies may be calculated in some cases even though the K^'a 



IMPORTANT TYPES OF MOLECULAR ALTERATION 255 

are not known, since for two inhibitors: 

AF, — AF, = 1.422 log [' il ~ \'l (2-10) 

ii (1 —I2) 

if(Ij) = (l2)- 

It is easy to derive an expression for the relationship between the true 

AF difference and the apparent JF difference for two inhibitors when the 

sole factor involved is the ionization of the inhibitors. If we designate the 

true free energy of binding as AF and the apparent free energy of binding 

as AF', the difference in the true binding energies for the two inhibitors 

will be AFi — AF^, and the difference in the apparent binding energies 

will be AF^ — AF^'. The relationship between these for analogs that are 

singly ionizing weak acids is: 

(AF, - AF,) = {AF,' - AF,') - 1.422 log \ + [|h^)/^?] ^^'^^^ 

It is therefore possible to correct for the pH effect if the p/C^'s of the inhibi- 
tors are known and the true interaction energy difference may be obtained. 

IMPORTANT TYPES OF MOLECULAR ALTERATION 
PRODUCING INHIBITING ANALOGS 

A substrate or coenzyme might generally be considered to have three 
different types of molecular region: (1) groups involved primarily in the 
binding to the enzyme, (2) groups involved in the catalytic reaction, and 
(3) groups or regions not directly involved in either binding or reaction. 
There is overlap between these in some cases, of course, because the groups 
undergoing chemical change usually participate to a certain extent in the 
binding. Furthermore, some substrates, such as succinate, do not possess 
the third type of group, all of the molecule being directly involved in bind- 
ing and reaction. The properties of a substance produced by altering a single 
group or region of a substrate wiU depend on the type of group or region 
that is modified, that is, its function in the reaction of the substrate with 
the enzyme. A change in a binding group will usually alter the interaction 
energy in the combination of the substance with the enzyme and may or 
may not affect the susceptibility to chemical reaction, whereas a change in 
a group directly involved in the catalysis will generally reduce the reactivity 
without necessarily modifying the binding to the enzyme. It is also evident 
that a change in the third type of neutral group will not be so likely to 
alter the behavior of the substrate, unless such a change in some way 
secondarily modifies the interactions of the other groups. The aim in the 
design of analogs for enzyme inhibition is to produce a compound which 
will bind reasonably tightly to the enzyme (preferably more tightly than 



256 2. ANALOGS OF ENZYME REACTION COMPONENTS 

the substrate), but which is resistant to chemical reaction in its complex 
with the enzyme. It would appear that the most effective inhibitors would 
result from modifications of the reactive groups, or of groups adjacent 
to the reactive region, rather than changes in binding groups. Malonate 
illustrates this in a simple way because here the — CH2CH2 — group of 
succinate has been altered to the nonoxidizable — CHg — group, while 
the binding — COO" groups remain. Amidation of the — C00~ groups of 
either succinate or malonate, thereby eliminating the negative charges, 
produces a substance that is neither a substrate nor an inhibitor because 
the affinity for the enzyme has been lost. 

The total interaction energy between a substrate or an inhibitor and the 
enzyme active center is the result of all the forces of attraction and repulsion 
summed over all the participating groups. Every atom or group of a sub- 
strate or its analog contributes to some degree to the interaction energy 
but, practically, the binding may be attributed usually to two or three 
groups that serve to orient the molecules on the enzyme surface. The sub- 
traction, addition, or alteration of substrate groups may change the binding 
energy in various ways. Modifying a region vicinal to a binding group may 
sterically interfere with the normal approach of this group to the enzyme 
group with which it interacts, or it may by inductive or resonance effects 
alter the properties of the binding group, as discussed in Chapter 1-6 
(page 304). It is worthwhile emphasizing again that a change in a particular 
region of a molecule may produce variations in the electronic configurations 
throughout the entire molecule, and that a change in the interaction energy 
cannot generally be attributed solely to this altered region. Furthermore, 
the volume and configuration of a substrate and its analog may involve 
water of hydration, so that a group change can secondarily affect the bind- 
ing by modifying the disposition of the bound water molecules. The in- 
troduction of so-called neutral groups, such as hydrocarbon chains, can 
bring about an increase in the binding energy through nonspecific van der 
Waals' interactions, providing these groups do not interfere sterically with 
the approach of the important binding groups to the enzyme surface. 

Most analogs of substrates have less affinity for the enzymes than do the 
natural substrates, which is reasonable in view of the enzyme active center 
conformation to the substrate configuration. However, occasionally an 
analog will exhibit a much tighter binding than the substrate, the extra 
binding energy being more than could be attributable simply to a new group 
introduced into the molecule. In such cases it is likely that a qualitative 
change in the binding is involved. A substrate frequently forms a covalent 
bond with the enzyme during the catalytic reaction, and normally this 
constitutes only an intermediate state in the sequence of changes. Certain 
analogs may be able to form this type of bond but are unable to complete 
the sequence, so that the analogs remain firmly attached to the enzyme. 



IMPORTANT TYPES OF MOLECULAR ALTERATION 257 

This is known to occur in the case of diisopropylfluorophosphate and re- 
lated cholinesterase inhibitors, as well as with monoamine oxidase inhibi- 
tors such as iproniazid. 

An analog may be related to the corresponding substrate in one of two 
general ways: it may be either an isomer of tne substrate, or a substance 
obtained by the replacement of one or more groups on the substrate. 
An isomeric analog may be a geometric isomer (e.g., one of a cis and trans 
pair), optical isomer, or any stereoisomer of the substrate. It might be 
thought that such analogs would often be specific and useful inhibitors but 
actually, except for certain optical isomers (see page 268), this is seldom 
the case, the reason being that the configuration of the analog is more 
important than a simple equivalence of all the atoms and groups. A sub- 
stitution analog can result from a variety of molecular changes in the sub- 
strate. What is often called addition of groups is usually only a substitution 
of the new group for a H atom (e.g., the replacement of a H atom with a 
F atom or a CH3 group), and what is called deletion of groups is usually 
a substitution of a H atom for the group that is removed. On the other hand, 
an important type of analog is derived by the substitution of one functional 
group with another group (e.g., the replacement of an OH group with a 
SH group, or of a CH3 group with a CI atom). Some commonly interchangea- 
ble groups might be put in the following families: 

(a) — NH2 —OH — SH — CH3 —CI — F — H 

(b) —COO- — SO3- — ASO3H- — PO3- 

(c) — CONH2 — SO2NH2 

{d) — S— —0— — NH— — CH=CH— — CH2— 

(e) — Phenyl — benzyl — pyridyl — pyrimidyl — cyclohexyl 

A good deal has been written about isosteric and isomorphic groups in the 
production of analogs (for an excellent review see Schatz, 1960), especially 
with respect to the development of new drugs, but this has limited bearing 
on the elaboration of enzyme inhibitors. The replacement of substrate 
groups with isosteric and approximately isomorphic groups usually leads 
to substances that are also substrates. It is generally necessary to alter 
the proper region of the substrate molecule significantly in order to produce 
an effective inhibitor. There are actually at the present time no general 
rules for the most efficient procedures to be used for the modification of 
substrates to produce inhibitors. Various enzymes exhibit quite different 
reaction mechanisms and an effective transformation of the substrate in 
one case will not work for other enzymes. For this reason the most important 
thing to establish initially is the nature of the particular enzyme mechanism, 
if an attempt is to be made to design analogs rationally. The binding groups, 
reactive groups, and relatively neutral groups in the substrate must be 



258 2. ANALOGS OF ENZYME REACTION COMPONENTS 

determined so that modifications in the structure may be made in the proper 
regions. The examjjles discussed in this chapter will clearly demonstrate 
that many analog inhibitors are not isosteric or isomorphic with the sub- 
strate and that, indeed, many of the most useful inhibitors appear to dif- 
fer quite markedly from the substrate. In this connection it is necessary 
to call attention to the danger of visualizing molecules on the basis of their 
classic two-dimensional formulas. One must also realize that usually not 
all of a substrate molecule is involved in the binding and reaction with the 
enzyme; the side of the molecule more distant from the enzyme may be 
relatively less important than the rest of the molecule and modifications 
on this side may give the appearance of producing radically different sub- 
stances, whereas from the standpoint of the enzyme surface these sub- 
stances may be very similar to the substrate. Conversely, just because two 
substances look alike when written in the usual structural formulas is not 
enough to ensure that they will exhibit comparable interactions with the 
active center of an enzyme. Ideally, one should learn to conceive enzyme 
reactions on an electronic and molecular level and to visualize analogs 
with three-dimensional molecular imagination, in other words, to approach 
the problem of analog design from the point of view of a rational active 
center. 

Some analogs are not directly inhibitory but are metabolically trans- 
formed into inhibitory substances that block a sequence at a later step. 
Such analogs are often very interesting and useful, being in many cases 
specific and potent. The design of an analog to be an inhibitor precursor 
presents a slightly different problem than in the general case. If the analog 
is to enter into the metabolic sequence it must be a substrate of the initial 
enzymes and, hence, structurally similar to the natural substrate in the 
region of the reactive groups. Isosteric and isomorphic substitution is often 
useful in this situation. This probably accounts for the popularity of fluo- 
rine as a replacement for a H atom in the design of this type of analog. 
Many F-substituted analogs enter into metabolic sequences and interfere 
at a more distal region, for example, fluoroacetate, 5-fluorouracil, 2>-fluoro- 
phenylalanine, and 6-deoxy-6-fluoro-D-glucose. The F atom is the smallest 
of the common atoms that may be substituted for a H atom, and approaches 
the H atom in size (van der Waals' radii for H and F atoms being 1.2 and 
1.35 A, respectively) (see Table 1-6-8). The F atom is also relatively un- 
reactive and forms a stable bond to carbon. However, it is strongly electro- 
negative and alters the electronic configuration in comparison to the parent 
compound. The dipole moment of the C — F bond will be quite different 
from the C — H bond, this altering neighboring bonds as well as inducing 
an ability to form hydrogen bonds with the enzyme. Thus the F analog 
not only may be much like the substrate in over all size and configuration, 
allowing it to be metabolically reactive, but eventually may be transformed 



THE CONCEPT OF INHIBITION BY ANALOGS 259 

into a substance in which the F atom, because of its electronegative char- 
acter, interferes in some way with the catalytic reaction. Another consid- 
eration arises when phosphorylation of the substrate is an initial step in 
the metabolic sequence. Here one must be careful not to alter the groups 
involved in the phosphorylation, but to modify groups that are reactive in 
a later enzymic step. 

DEVELOPMENT OF THE CONCEPT OF INHIBITION 
BY ANALOGS 

This important concept, which today plays a major role in the fields 
of enzymology and biochemistry and is becoming more and more important 
in pharmacology, chemotherapeutics, and pathology, has an interesting 
history illustrating a typical growth pattern of a scientific idea. The de- 
velopment of the general concept has been traced by Martin (1951), Wool- 
ley (1952), and Albert (1960), so that here it is necessary only to present a 
cursory exposition related particularly to enzyme inhibition. However, 
it must be realized that many fields of study — including immunology, 
drug antagonism, and microbial growth inhibition among others — con- 
tributed in one way or another to this concept. 

Despite the fact that numerous examples of competitive inhibition by 
analogs had been demonstrated since 1910, and that the analog concept 
had been quite clearly stated around 1930, general recognition of the basic 
principles did not occur until after 1940. Wohl and Glimm (1910) reported 
the inhibition of amylase by glucose and galactose, as well as by the reac- 
tion product, maltose, and shortly Michaelis described the inhibition of 
/5-fructofuranosidase by fructose and a-methylglucoside (Michaelis and 
Pechstein, 1914), and the inhibition of a-glucosidase by glucose (Michaelis 
and Rona, 1914). Isolated, unpremeditated, and unrecognized discoveries, 
such as the inhibition of arginase by ornithine (Gross, 1921) and the inhi- 
bition of /?-fructosidase by /^-glucose but not by cf-glucose (Kuhn, 1923) un- 
fortunately gave no impetus to the formulation of a general concept. Even 
the classic work of Quastel on malonate inhibition, described in the previous 
chapter, wherein competitive inhibition by structural analogs was considered 
in a modern fashion, apparently did not activate anyone outside Cambridge, 
where Bernheim (1928) studied the aconitate inhibition of citrate oxidation, 
Murray (1929) and Murray and King (1930) applied the principles of 
analogs (although in a somewhat naive way) to the inhibition of lipase by 
various ketones and alcohols, Richter (1934) proved the inhibition of ca- 
techol oxidase by resorcinol to be competitive, Keilin and Hartree (1936) 
reported potent competitive inhibition of uricase by the methylurates, and 
Green (1936) extended the malonate inhibition of succinate dehydrogenase 
to the inhibition of malate dehydrogenase by several dicarboxylates. 



260 2. ANALOGS OF ENZYME REACTION COMPONENTS 

Several accidental observations were made of the toxicity of metabolite 
analogs while they were being tested for activity in animals, for example, 
the discovery that ethionine is toxic to rats (Dyer, 1938) and that pyridine- 
3-sulfonate is lethal to dogs suffering from nicotinic acid deficiency (WooUey 
et al., 1938), but the importance was not immediately recognized. Appar- 
ently the stimulus for the formulation of the general theory had to come 
from a discovery of clinical importance, and this was provided by the find- 
ing of the antagonism by p-aminobenzoate of the action of the sulfona- 
mides by Woods (1940). Between 1940 and 1943 many examples of analog 
antagonism were reported, and from 1943 to 1946 it was shown that vi- 
tamin deficiency symptoms could be produced in animals by the administra- 
tion of the appropriate analogs of all the known vitamins. It was then 
possible to return to the enzyme level and apply the principles formulated 
long before to the new results. Since 1946 there has been a steadily increas- 
ing interest in analog inhibitors, which is reflected in the large numbers of 
publications on this subject each year (Fig. 2-1). At the present time a paper 
on analog enzyme inhibition as covered in this chapter appears every other 
day; this does not include all of the investigations on analog inhibitors 
that are to be treated separately (such as fluoroacetate, arsenate, carbon 
monoxide, parapyruvate, and others), or analogs commonly used clinically 




Fig. 2-1. Curve showing the annual number of publications on analog inhibition 

of enzymes. 



ANALOG INHIBITION OF MEMBRANE TRANSPORT 261 

(sucli as the cholinesterase and monoamine oxidase inhibitors), or antime- 
tabolites used in microbial growth suppression or tumoristasis. 

A few general references treating aspects of the subject not included 
in the present work are suggested for additional information: Welch (1945), 
Work and Work (1948), Woolley (1950 b, 1952), Martin (1951), Rhoads 
(1955), Matthews (1958), Albert (1960), Schueler (1960), Schatz (1960), 
Kaiser (1960), and the Symposium on Antimetabolites sponsored by the 
National Vitamin Foundation (1955). A great deal of information on the 
biological actions of many types of analog will be found in Volume I of 
"Metabolic Inhibitors" edited by Hochster and Quastel (1963), and this 
aspect of the subject will be mainly omitted in the present work so that 
the enzymic effects may be discussed in sufficient detail. 



ANALOG INHIBITION OF MEMBRANE TRANSPORT 

Before discussing specific enzyme inhibitions by analogs, we must turn 
our attention to the possibility that the depression of the utilization of 
some substrate or metabolite by an analog is not due to an action on any 
enzyme involved in the metabolism, but is the result of a specific inter- 
ference with the transport of the substrate or metabolite into the cell. 
It is quite probable that some of the actions of analogs on metabolism, at- 
tributed to competition at the enzyme level, are actually exerted at the 
cell membrane; indeed, certain instances will be discussed. It is becoming 
more and more evident that many substrates and metabolic precursors 
are taken up by cells by processes other than simple diffusion, and this 
applies particularly to certain carbohydrates, amino acids, coenzymes, 
or coenzyme precursors. It is not necessary that the transport be active 
to be influenced by analogs; any movement of a substance through a mem- 
brane which involves a carrier, a special size or configuration of pores, or a 
specific type of mechanism, active or passive, can be slowed in the presence 
of an analog of this substance. It is possible, o^ course, that the membrane 
trasport is mediated through an enzyme reaction, such as a phosphoryla- 
tion, and the competition by the analog is then truly an enzymatic one. 

It is sometimes difficult to determine if there is inhibition of a transport 
process. The demonstration of a reduced uptake of a substrate from the 
medium is generally not sufficient evidence, inasmuch as a decreased in- 
tracellular utilization might also be responsible. The best procedure is to 
determine directly the concentration of the substrate within the cells in 
the absence and presence of the analog, but sometimes the concentration 
is too low to measure accurately, especially when the substrate is rapidly 
metabolized. The demonstration of typical competitive inhibition with 
respect to the action of an analog on some metabolic process is also not 
sufficient evidence for an enzymic site of action, because the inhibition of 



262 2. ANALOGS OF ENZYME REACTION COMPONENTS 

membrane transport may exhibit competitive kinetics. The kinetics will 
often depend on whether the transport is limiting the metabolic utilization 
of the substrate or not. Membrane transport involving a carrier (C) molecule 
can often be represented by: 

S + C ^ SCo -> SC, -► C + S (2-12) 

where the subscripts refer to the outside and inside of the membrane. The 
later reactions may be written in one direction only because of the utili- 
zation of the substrate as it enters the cell. If an analog also combines 
with the carrier: 

S' + C ^ S'C (2-13) 

whether it is transported into the cell or not, typical competitive behavior 
will be observed since the forms of Eqs. 2-12 and 2-13 are the same as 
those for competitive enzyme inhibition. 

Plots of transport rates against the external concentrations of the trans- 
ported substance usually yield hyperbolic curves, and double reciprocal 
plots are often linear, allowing the calculation of a constant which corres- 
ponds to the Michaelis-Menten constant in enzyme kinetics. It is frequently 
assumed that this is the dissociation constant for the complex of the sub- 
stance with the carrier, but it is not necessarily true for the same reasons 
that K^,^ is not always K,. The kinetics of transport inhibition are likewise 
commonly similar to those observed with enzymes and values of K^ may 
be determined by appropriate plotting, this constant representing the dis- 
sociation constant of the carrier-analog complex. The kinetics of carrier 
transport and its inhibition have been elaborated by Wilbrandt and Rosen- 
berg (1961) and Rosenberg and Wilbrandt (1962) for facilitated diffusion 
and certain restricted types of active transport. It is interesting in connec- 
tion with certain types of inhibition work to note that the accumulation 
ratio, (X),/(X)^ = VJkK„j, where F^„ is the maximal transport rate, K,„ 
is the Michaelis-Menten constant for transport, and k is the passive diffu- 
sion constant for the membrane. Thus the cell/medium ratio may be altered 
by the inhibitor as a result of changes in any of these three parameters. 

Carbohydrate Transport 

Competition between sugars for entrance into cells has been observed 
in many tissues but has seldom been studied quantitatively, so that in most 
cases it is impossible to know if true competition kinetics are followed. 
Occasionally a reduction in the inhibition with increase in the concentration 
of the transported substrate has been noted; for example, the active ac- 
cumulation of D-galactose by rabbit kidney cortex slices is inhibited 61% 
by 5.6 rciM glucose when D-galactose is 0.1 mM and only 28% when d- 
galactose is 0.2 mM (Krane and Crane, 1959), but these results do not fit 



ANALOG INHIBITION OF MEMBRANE TRANSPORT 263 

a simple competitive formulation. Indeed, the exact site of inhibition has 
not been established in any case. The entrance of L-arabinose into rat 
heart cells is inhibited 92% by glucose at equimolar concentration and, 
since L-arabinose is not metabolized, the inhibition is presumably on some 
phase of membrane transport (Morgan and Park, 1958). In the same tissue 
the competition between glucose and 3-methylglucose at the outer but 
not the inner surface of the cell membrane leads to a net outward transport 
of 3-methylglucose against a concentration gradient, and it was concluded 
that there are stereospecific combining sites at both surfaces. A compari- 
son between cellular and subcellular preparations can occasionally indicate 
a membrane site for an inhibition. Galactose markedly inhibits the utili- 
zation of fructose by intact ascites tumor cells but not in homogenates, 
pointing to a competition before the hexokinase step and probably in 
transport (Nirenberg and Hogg, 1957). The effects of 2-deoxy-D-glucose 
on the fermentation of glucose, fructose, and mannose by yeast (to be 
discussed in more detail on page 391) led Scharff (1961) to assume a trans- 
port system which is stable, since it still functions in acetone-dried cells, 
and perhaps bound to a complex or "bundle" of fermentation enzymes 
somewhere in the outer regions of the yeast cells. 

Although the nature of the transport mechanisms and the site of inhi- 
bition are generally not known, it is clear that the interference is quite 
specific and dependent on the molecular configurations of the sugars. The 
transport of D-galactose (5 mM) by hamster jejunum is inhibited to varying 
degrees by other sugars and derivatives at 25 nxM (see tabulation) (Wilson 



Sugar % Inhibition 

D-Mannose Stim 4 

D- Xylose 7 

3-0-Methyl-D-glucose 59 

a-Methyl-D-glucoside 95 

et al, 1960), and the uptake of glucose by lymph node cells is likewise 
inhibited differently by several sugars at 9 mM (see tabulation) (Helm- 
reich and Eisen, 1959), both observations pointing to stereospecific trans- 



Sugar % Inhibition 

D-Arabinose 7 

D-Galactose 9 

D-Fructose 40 

D-Mannose 61 



264 2. ANALOGS OF ENZYME REACTION COMPONENTS 

port systems presumably involving carriers in the membranes. The pat- 
terns of competition between various sugars have often led to the assumption 
of two or more different transport mechanisms for carbohydrates and that 
these systems can operate simultaneously and independently. Competition 
for transport can be demonstrated in rat diaphragm muscle for members 
of the group including D-glucose, D-mannose, D-xylose, D-arabinose, l- 
arabinose, D-lyxose, and 3-0-methyl-D-glucose, but not between these 
and D-galactose, D-fructose, maltose, a-methyl-D-glucoside, and /?-methyl- 
D-glucoside (Battaglia and Randle, 1960), so that different sites for entry 
were postulated. A number of sugars are transported into erythrocytes: 
all the aldoses penetrate by a transport system characterized for glucose, 
while the ketoses penetrate according to a pattern of passive diffusion 
(LeFevre and Davies, 1951). The aldoses compete with each other, e.g., 
the uptake of glucose is strongly inhibited by mannose, but any aldose 
delays the entrance of a ketose, e.g., the uptake of fructose is prevented 
by glucose and galactose, while the ketoses do not perceptibly alter the 
transport of the aldoses. One carrier system in the erythrocyte has been 
characterized as reacting only with those monosaccharides in which the 
pyranose ring tends to assume the "chair" configuration, which illustrates 
a unique type of stereospecificity (LeFevre and Marshall, 1958). On the 
other hand, in hamster intestine only a single transport mechanism or car- 
rier seems to be present, mutual inhibition occurring between D-glucose 
and D-galactose, D-glucose and 1,5-anhydro-D-glucitol, D-galactose and 1,5- 
anhydro-D-glucitol, D-glucose and 6-deoxy-D-glucose, and 6-deoxy-D-glucose 
and 1,5-anhydro-D-glucitol, whereas sugars that are not transported do not 
interfere with those that are (R. K. Crane, 1960). 

Amino Acid Transport 

The situation here is very much the same as in sugar transport and there 
is good evidence for stereospecific systems and multiple pathways. There 
is competition between L-leucine and DL-isoleucine for renal tubular resorp- 
tion in the dog, these being well resorbed amino acids, and there is also 
competition between the poorly resorbed L-arginine and L-lysine (Beyer 
et at., 1947). However, no interference is observed between L-leucine and 
L-arginine. It is possible to classify the amino acids into groups with respect 
to their mutual interference in resorption. In the rat kidney, dibasic amino 
acids (arginine, lysine, and cystine) are actively accumulated and there 
is mutual inhibition of the transport (Rosenberg et al., 1962). Monobasic 
amino acids (alanine, phenylalanine, and histidine) do not interfere with 
the uptake of the dibasic amino acids, nor does arginine depress the uptake 
of the monobasic amino acids. It seems likely that separate transport sys- 
tems are present. The transport system for basic amino acids in the hamster 
intestine is distinct from that for other amino acids and similar to the renal 



ANALOG INHIBITION OF MEMBRANE TRANSPORT 



265 



system (see accompanying tabulation) (Hagihira et al., 1961). Arginine 
and cystine interfere more with lysine transport than with glycine trans- 
port, whereas methionine behaves in the reverse manner, and there is 

/-I J. A.- % Inhibition of transport of : 

, , ., . Concentration 



inniDitor 


{raM) 


Glycine 


L-Lysine 


L-Arginine 


2 


16 


89 


L-Cystine 


0.8 





45 


L-Methionine 


1 


73 


32 


L-Lysine 


1 


14 


— 


Glycine 


1 


— 






little or no interference between glycine and lysine. Proline, histidine, and 
glycine are actively transported across the intestinal wall and methionine 
at eqiiimolar concentration completely inhibits this (Wiseman, 1954). This 
intestinal transport carrier is limited to monoamine-monocarboxylates and 
they compete with each other. In addition to a common carrier, there may 
be an additional carrier for glycine and proline (Newey and Smyth, 1964), 
and it was pointed out that although each carrier system would conform 
to Michaelis-Menten kinetics, the total transport with two or more carriers 
involved would not necessarily. 

Intestinal transport systems may react with only the l- or the D-form 
of an optically isomeric pair. D-Methionine is accumulated and transported 
against a concentration gradient by the rat intestine and this is blocked 
completely by equimolar concentrations of L-methionine ( Jervis and Smyth, 
1960). Similarly, the transport of L-I^^^-monoiodotyrosine, which involves 
active accumulation of the amino acid in the gut wall, is inhibited by 
many L-amino acids (the most effective being L-tryptophan, L-methionine, 
L-leucine, and L-isoleucine) but scarcely at all by any of the four D-amino 
acids tested (Nathans et al., 1960). L-Tryptophan, for example, at 10 mM 
reduces the tissue/medium ratio from 8.85 to 1.55 and the inhibition is 
apparently competitive. 

The active cumulative uptake of amino acids by ascites carcinoma cells 
comprises several transport systems, each with a specific range of substrates. 
The uptake of glycine-1-C^* {K,,, = 6.4 mM) is most strongly inhibited by 
1-aminocyclopentanecarboxylate {K, = 1.47 mM), and this is competitive, 
while the uptake of DL-methionine-S^^ {K„, = 1.7 mM) is inhibited best 
by allylglycine {K^ = 0.86 mM). Glycine transport is moderately inhib- 
ited by aUylglycine and less readily by furylglycine and thienylglycine 
(Scholefield, 1961). On the other hand, the uptake of DL-leucine-1-C^^ 
is stimulated by most of these inhibitors. Transport of L-tryptophan in 



266 2. ANALOGS OF ENZYME EEACTION COMPONENTS 

ascites cells occurs by both diffusion and active transport (Jacquez, 1961). 
Certain amino acids accelerate this at lower concentrations (1 mM) and 
competitively inhibit at higher (5 mM), while other amino acids (such 
as L-alanine, L-lysine, and L-arginine) only inhibit. Oxender and Christen- 
sen (1963) thoroughly studied the effects of many amino acids on the up- 
take of neutral amino acids by ascites cells and found they fall into two 
overlapping clusters, the transport systems apparently not being very 
specific. 

The penetration of amino acids into the brain is probably important 
for the metabolism and function of that tissue, and there appear to be 
several transport systems available. Tyrosine enters the brain readily in 
vivo and a specific transport is probably involved, since L-tyrosine pene- 
trates more rapidly than D-tyrosine and the entry is potently inhibited by 
certain other amino acids, particularly L-tryptophan, L-leucine, L-valine, 
/5-fluorophenylalanine, and L-histidine (Chirigos et al., 1960). In phenyl- 
ketonuria the blood levels of phenylalanine are high due to the inability 
of the tissues to metabolize it to tyrosine. It is possible that these high 
concentrations can interfere with the entry of other amino acids into the 
brain and partially account for the central nervous system disturbances. 
The uptake of five amino acids by rat brain slices is indeed inhibited by 
L-phenylalanine (see accompanying tabulation) and it was felt that such 

Amino acid (2 raM) % Decrease of concentration gradient 



L-Proline 9 

L-Histidine 42 

L-Arginine 46 

L-Ornithine 47 

L-Tvrosine 70 



could occur in vivo (Neame, 1961). The transport of L-histidine is inhib- 
ited by neutral aliphatic amino acids and short-chain diamino acids to a 
degree dependent on the length of the carbon chain (Neame, 1964). Inhi- 
bition by the dicarboxylic amino acids is not dependent on the chain length. 
In general, the L-isomers inhibit more potently than the D-isomers. It was 
suggested that histidine is transported by a system which transports most 
other amino acids, but with different affinities, since the inhibitions are all 
competitive. The synthetic amino acid, 1-aminocyclopentanecarboxylate, 
is not metabolized but is actively transported in brain slices and ascites 
cells, and the system involved must be the same as for methionine since 
it is affected similarly by the same competitive amino acids (Ahmed and 
Scholefield, 1962). Such transported but nonmetabolized amino acids may 
well be of use in studying transport inhibition, since effects can be clearly 



ANALOG INHIBITION OF MEMBRANE TRANSPORT 267 

distinguished from possible inhibitions of amino acid incorporation in the 
cell. Reference should be made to the very complete investigation of amino 
acid transport in brain slices by Abadom and Scholefield (1962), in which the 
many competitive inhibitions established point to several separate amino 
acid transport systems. 

The entrance of valine, proline, and hydroxyproline into the human 
erythrocyte is not mutually competitive, but is inhibited markedly by 
certain sugars, such as glucose, galactose, and xylose, although not by 
fructose (Rieser, 1961). These results would indicate that some amino acids 
and sugars follow the same transport pathway. If this is a general phenome- 
non, one must consider in the use of analogs of these substances the possi- 
bility that an amino acid analog might depress glucose uptake, and thereby 
secondarily interfere with the transport by suppressing energy generation. 

The accumulation of L-histidine by the parasitic fungus Botnjtis fabae 
is inhibited by most other amino acids, and one transport system seems to 
be available to all the amino acids (Jones, 1963). Substitution at the NH2 
or COOH groups lessens or abolishes the inhibitory activity, indicating 
that the binding to the carrier is at least partly electrostatic. 

Miscellaneous Transport Systems 

The oxidation of protocatechuate by a Flavohacterium is competitively 
inhibited by ^-aminosalicylate and one might conclude that mutual in- 
teraction with some enzyme is responsible. However, 59-aminosalicylate 
does not affect the rate or extent of the oxidation in extracts, measured in 
different ways (Hubbard and Durham, 1961). These results thus point to 
competition for a transport system in the membrane, rather than the more 
usual explanation. The active transport of biotin across the hamster in- 
testine is inhibited by various analogs (e.g., biocytin, desthiobiotin, di- 
aminobiotin, and biotin methyl ester), but the nature of the inhibition was 
not investigated so it may not be competitive (it is not with lipoate) (Spen- 
cer and Brody, 1964). The active influx of urate into erythrocytes is com- 
petitively inhibited by hypoxanthine with K^ = 0.1 mM, whereas the 
efilux consists of two components, one sensitive to hypoxanthine (Lassen 
and Overgaard-Hansen, 1962). 

There are several instances in which the transport of inorganic ions is 
inhibited by other related ions. The transfer and exchange of phosphate 
across the membrane of S. aureus are inhibited by chlorate, borate, and ar- 
senate (Mitchell, 1954), although only arsenate is able to substitute com- 
pletely for phosphate in the exchange. The uptake of sulfate by yeast is 
competitively inhibited by thiosulfate (Kleinzeller et al., 1959). Nitrate 
inhibits quite well the accumulation of iodide in the rabbit ciliary body, 
50% reduction of the tissue/medium ratio occurring at 3 vaM, but it is 
not known if this is truly competitive (Becker, 1961). 



268 2. ANALOGS OF ENZYME REACTION COMPONENTS 

The examples chosen to ilhistrate transport inhibition do not always 
involve analogs, except in the most general sense, but clearly demonstrate 
the importance of considering such a type of interference whenever analogs 
are used in cellular preparations. 

ANALOGS WHICH ARE ISOMERS OF SUBSTRATES 

The behavior of the isomers of normal metabolites, particularly optical 
isomers, constitutes, in a way, a special field and therefore some of the more 
interesting results will be discussed in this section rather than under the 
specific enzymes that are involved. The concept that a proper fit of a sub- 
strate to the enzyme surface is necessary for reaction implies that enzymes 
will usually be stereospecific, and this has been demonstrated many times. 
Since an enzyme commonly attacks only one form of an isomeric pair, 
the unreactive form may be either an inhibitor or completely inert. In most 
cases the unreactive form does not bind to the enzyme at all, as might be 
expected from the different spatial configurations of most isomeric pairs, 
and is not inhibitory. 

Enantiomeric Analogs 

These analogs are related to the corresponding substrates on the basis 
of molecular asymmetry and the most common examples are optically 
active due to an asymmetric carbon atom. Unnatural enantiomers often 
exhibit no affinity for the enzymes. Indeed, it has been generally found 
that D-amino acids do not interfere with the microbial growth-promoting 
activity of L-amino acids, although there are exceptions. Some examples 
of a lack of inhibition by optical isomers may be mentioned. The oxidation 
of L-phenylalanine by the L-amino acid oxidase of Neurospora is not in- 
hibited by D-phenylalanine, even when the latter is present at 500 times 
the concentration of the substrate (Burton, 1951 b), and D-tryptophan 
does not inhibit E. coli L-tryptophanase even though it interferes with 
growth (Gooder and Happold, 1954). D-Malate is not oxidized by the malate 
dehydrogenase of Mycobacterium tuberculosis nor does it inhibit the oxida- 
tion of L-malate (Goldman, 1956 b). A rather unusual case is presented by 
potato tyrosinase in that both l- and D-tyrosine are attacked at the same 
rate, but there is no evidence of mutual inhibition, perhaps because of the 
limited range of concentrations used (Spencer et al., 1956). Occasionally 
a slight inhibition is noted but one which would not be of any practical 
significance, as in the just detectable inhibition by D-leucine of the oxida- 
tion of L-leucine by the L-amino acid oxidase of the hepatopancreas of 
Cardium tuberculatum, an inhibition actually much less than exerted by 
other amino acids and hence probably not specific (Roche et al., 1959). 



ANALOGS WHICH ARE ISOMERS OF SUBSTRATES 269 

However, the primary purpose of this section is to discuss instances in 
which significant inhibition by optical isomers is observed. 

The asparaginase of Mycobacterium phlei attacks only L-asparagine and 
this deamidation is quite well inhibited by D-asparagine (Grossowicz and 
Halpern, 1956 a). It is a particularly clear and straightforward instance 
of stereomeric inhibition and a 1/v — 1/(S) plot shows it to be completely 
competitive. Ai'ound 75% inhibition is produced by 40 mM D-asparagine 
when L-asparagine is 10 mM. On the other hand, the asparaginases of 
Bacillus coagulans and B. stearothermophilus are inhibited by D-asparagine 
but not competitively (Manning and Campbell, 1957). The type of inhi- 
bition is difficult to designate since the double reciprocal plots intersect 
to the right of the ordinate, that is, the inhibition does not tend toward 
noncompetitive kinetics. A yet more complex situation is presented in the 
depression of the formation of a-amylase by D-aspartate in Pseudomonas 
saccharophila, 0.2 mM blocking the protein synthesis completely (Eisen- 
stadt et a]., 1959). The inhibition is readily reversed by L-aspartate and is 
characterized by a fairly long lag period before inhibition is observed. 
The inhibition was assvimed to be on the reaction: 

L- Aspartate + IMP + GTP -> fumarate + AMP + GDP + P 

thus producing an impairment in AMP synthesis, this secondarily disturbing 
the formation of ATP and the activation of amino acids for protein synthesis. 
Examination of this reaction in cell-free extracts showed that D-aspartate 
does indeed inhibit competitively. Another type of inhibition is exhibited 
by pea glutamine synthetase with L-glutamate, ammonia, and ATP as 
substrates (Varner, 1960). D-Glutamate inhibits the formation of L-gluta- 
mine. However, D-glutamate is also a substrate and actually has a lower 
K„^ although the reaction rate is slower than with L-glutamate (^,„ for 
L-glutamate is 10 mM and for D-glutamate is 2 mM). This is then an exam- 
ple of a competitive inhibition between isomeric substrates. It may also 
be mentioned that D-glutamate inhibits, although quite weakly, the de- 
carboxylation of L-glutamate by bacterial glutamate decarboxylase (Ro- 
berts, 1953). 

The splitting of L-histidine by rat liver histidase is inhibited by D-histi- 
dine, 11% inhibition being given by 2 mM, 36% by 12 mM, 55% by 24 
mM, and 85% by 48 mM when the L-histidine is 12 mM (Edlbacher et al., 
1940). The formation of L-histidine from L-histidinol occurs in two steps: 

L-Histidinol -^ L-histidinal -> L-liistidine 

The enzymes catalyzing both these reactions are inhibited by D-histidinol 
and D-histidinal competitively (Adams, 1955). The Ki for D-histidinol is 
0.05 mM for both oxidations by a yeast preparation and it is possible that 
only a single enzyme is involved. 



270 2. ANALOGS OF ENZYME REACTION COMPONENTS 

Germination of Bacillus cereus spores is induced by certain amino acids, 
such as L-alanine, and it would appear in this case that L-alanine dehydro- 
genase is essential for the activation process (O'Connor and Halvorson, 
1961 b). D- Alanine and some other analogs, such as D-a-amino-w-butyrate, 
inhibit the germination when it is induced by L-alanine, and also inhibit 
the oxidative deamination of L-alanine and to a lesser extent other amino 
acids, there being a good correlation between these two inhibitory actions 
in the series of analogs used (see accompanying tabulation). The inhibition 



% Inhibition of deamination 
Substrate ^^ D-alanine (100 mi/) 



L-Alanine 61 

L-a-Amino-?i-butyrate 59 

L-Norvaline 48 

L-Serine 28 

L-Valine 24 

L-Cysteine 15 

L-Isoleucine 12 

L-Leucine 

L-Phenylalanine 



by D-alanine is, however, by no means specific for L-alanine. It is possible 
that several enzymes, with different susceptibilities to D-alanine, are in- 
volved in the deamination of the various amino acids. 

An interesting illustration of optical specificity is provided by the 0- 
phosphoserine phosphatase from chicken liver (Neuhaus and Byrne, 1960), 
Both L-phosphoserine and D-phosphoserine are substrates and both l- 
serine and D-serine inhibit. The L-serine is a much more potent inhibitor 
(see accompanying tabulation), l- Alanine also inhibits, being between 





K, 


(railf) 


Substrate 


L-Serine 


D-Serine 


L-Phosphoserine 
D-Phosphoserine 


0.68 
0.70 


27 
29 



l- and D-serine in potency, but D-alanine does not inhibit. The inhibition 
by L-serine seems to be uncompetitive from a double reciprocal plot, but 



ANALOGS WHICH ARE ISOMERS OF SUBSTRATES 271 

this is apparent only and the type of inhibition does not fit into the usual 
classical categories. The following scheme was proposed: 

E + PS ^ EPS :^ EP + S 

E + P 

where PS is phosphoserine. It was shown to fit the data kinetically if 
k_Jki is small. This example shows well the danger of uncritically accepting 
the usual interpretation of a plotting procedure since, as emphasized 
previously, there are types of inhibition different from those included 
in the classic formulations. 

a-Chymotrypsin hydrolyzes the L-isomers of various tryptophanamides 
and tyrosinamides, and these reactions are usually inhibited by the D-iso- 
mers (Huang and Niemann, 1952; Manning and Niemann, 1958). When 
the substrate is nicotinyl-L-tryptophanamide (Kg = 2.7 mM), the reaction 
is inhibited by nicotinyl-D-tryptophanamide (K^ =1.4 mM) and a number 
of other derivatives of D-tryptophanamide. The hydrolysis of several de- 
rivatives of L-tyrosinamide is similarly inhibited by the D-isomers (see ac- 
companying tabulation). It is to be noted that in every case the D-isomer 
is bound more tightly than the L-isomer, assuming that K^ does indeed 
represent a dissociation constant. The possible forces binding these sub- 
stances to the enzyme will be discussed in a later section (pages 370-375). 

Tyrosinamide Kg for L-isomer K^ for D-isomer 



Nicotinyl- 


12 


9 


Chloroacetyl- 


27 


6.5 


Trifluoroacetyl- 


26 


20 


Acetyl- 


32 


12 



Anomeric Analogs 

Michaelis and Pechstein (1914), in their early work on /5-fructofuranosi- 
dase, and Michaelis and Rona (1914), studying yeast maltase inhibition, 
concluded that the configuration around carbon 1 of carbohydrates (i.e., 
a- and /5-anomers) is of importance in determining the affinity of these 
substances for the enzymes, since a-methylglucoside inhibits both enzymes 
quite potently whereas /?-methylglucoside inhibits very little or not at 
aU. The splitting of phenol-/?-glucosides by taka-/?-glucosidase is also in- 
hibited by phenol-a-glucoside (Ezaki, 1940). The synthesis of polysaccha- 
ride from a-D-glucose-1 -phosphate by muscle phosphorylase is not inhib- 
ited by /?-D-glucose-l-phosphate; however, it is interesting that a-methyl- 



272 2. ANALOGS OF ENZYME REACTION COMPONENTS 

glucoside inhibits while /5-methylglucoside does not, indicating the impor- 
tance of the carbon 1 configuration (Campbell et al., 1952). In general, the 
/5-anomers cannot act as either substrates or inhibitors of phosphorylase. 
The /^-glucuronidase of mouse liver is also stereospecific, since menthyl- 
/?-glucoronide is a substrate but menthyl-a-glucuronide only a very weak 
inhibitor (Levvy and Marsh, 1952). 

Positional Analogs 

Isomers in which a ring group is moved from one position on the ring 
to another are generally not inhibitory due to the fairly marked structural 
changes involved. There are exceptions, however, and one of the most 
striking is the inhibition of the oxidation of p-hydroxyphenylpyruvate to 
homogentisate by m-hydroxyphenylpyruvate in preparations from dog li- 
ver (La Du and Zannoni, 1955). A depression of 50% is seen with 0.2 mM 
and over 90% with 0.5 mM m-hydroxyphenylpyruvate when the substrate 
concentration is presumably 1.2 raM, indicating a tighter binding to the 
enzyme of the m-isomer. Other examples of positional isomers will be 
encountered in later sections. 

Geometric Isomeric Analogs 

One would not expect that potent inhibitors would be found in cis 
and trans pairs because of the different molecular configurations. We have 
already seen that fumarate and maleate differ markedly in their reactions 
with succinate dehydrogenase (page 34). The outstanding exception to 
this rule is the well-known inhibition of aconitase by fraws-aconitate. This 
enzyme catalyzes the interconversion of the tricarboxylates: 

Citrate ^ cts-aconitate ;fi isocitrate 

although perhaps cis-aconitate is not an obligatory intermediate between 
citrate and isocitrate. Bernheim (1928), impressed by the results obtained 
by Quastel with malonate, tested the effect of frans-aconitate on liver 
"citric dehydrogenase" (the reduction of the methylene blue used in this 
system was actually due to the oxidation of isocitrate formed from 
citrate via aconitase) and found definite inhibition. He believed the inhi- 
bition to be related to the structural similarity between citrate and trans- 
aconitate, stating, "The curve obtained seems to indicate that the aconitic 
acid is adsorbed on the enzyme so that part of the surface is unavailable 
for citric acid." Twenty years later a thorough study of this inhibition was 
made by Saffran and Prado (1949), using aconitase from pigeon breast 
muscle. Both the conversion of m-aconitate to citrate and the disappearance 
of citrate are inhibited by irans-aconitate. However, trans-acomta.te is 



ANALOGS WHICH ARE ISOMERS OF SUBSTRATES 273 

not bound as tightly to the enzyme as are the substrates. When citrate is 
3.3 mM, 50% inhibition is found with 16 xnM trans-a.comta.te; since K^^ 
for citrate is roughly 1 mM, K, is approximately 4 mM. The inhibition is 
competitive although the Ijv — 1/(S) plots are not ideal, perhaps because 
of some enzyme inactivation or failure to achieve equilibrium. The inhibi- 
tion of rat mammary gland aconitase seems to be somewhat more potent, 
since equimolar concentrations of citrate and ^raws-aconitate lead to around 
50% inhibition (Abraham et al., 1960). Studies on the stereospecificity and 
deuterium transfer during reactions catalyzed by aconitase (Speyer and 
Dickman, 1956; Englard and Colowick, 1957) point to a three-point at- 
tachment of the tricarboxylates (and perhaps an intermediate carbonium 
ion) to the apoenzyme and Fe++. The carboxyl groups in trans-aconitate 
would not appear to be in such a favorable position as in m-aconitate for 
the formation of this complex and this might explain the relatively weaker 
binding. The effects of trans-acomtate on other enzymes have been little 
investigated but it has been found to be a fairly potent inhibitor of fu- 
marase, Z^ being 0.63 mM at pH 6.35 (Massey, 1953 b). The K, increases 
with rise in the pH (Fig. 1-14-11) and, as with malonate, the formation of 
the EI complex is exothermic at low temperatures and endothermic at 
high temperatures. 

It is somewhat surprising that tro ns-aconitate is a reasonably effective 
inhibitor of the respiration of intact cells, inasmuch as penetration into the 
cells should be difficult. The following inhibitions have been observed: 
22-36% of endogenous respiration of various tumor shces and 28% of 
endogenous respiration of liver slices at unspecified concentration (Wein- 
house et al, 1951), 25% of Paramecium respiration at 10 mM (Holland and 
Humphrey, 1953), and 40% of the ion-linked respiration of barley roots 
at 20 mM (Ordin and Jacobson, 1955). However, no inhibition of the 
respiration of Australorbis mince at 10 mM was reported (Weinbach, 
1953). The oxygen uptake resulting from the addition of citrate or cis- 
aconitate to rat liver slices is strongly inhibited by 30 mM ^raws-aconitate 
(Sherman and Corley, 1952). The most complete study of respiratory in- 
hibition is by Saffran and Prado (1949) with rat liver and kidney slices. 
In the latter the inhibition is 27% at 2 mM and 73% at 20 mM, which is 
quite comparable to malonate. The inhibition by 2 mM ^rans-aconitate 
is not altered by adding malate or fumarate, but is reversed with 5 mM 
citrate or cis-aconitate. In other experiments the sensitivity to trans- 
aconitate is unexplainably less. The inhibition of aconitase in liver and 
mammary gland homogenates leads to a fairly marked depression of the 
conversion of citrate to CO2 and of acetate to fatty acids by fmws-aconitate 
(Abraham et al., 1960). The synthesis of mammary fatty acids is inhibited 
45% by 7.1 mM and 75% by 21.4 mM. Accumulation of citrate accompa- 
nies the inhibition in kidney, liver, and tumor slices (Weinhouse et al., 



274 2. ANALOGS OF ENZYME REACTION COMPONENTS 

1951; Saffran and Prado, 1949), and this is augmented by the addition of 
cycle intermediates such as pyruvate or malate. These results indicate 
clearly that aconitase is being inhibited intracellular ly. 

Essentially nothing is known of the possible effects of ^rans-aconitate on 
cellular functions. No depression of Paramecium motility is seen at 10 mM 
(Holland and Humphrey, 1953). However, the active transport of ions by 
barley roots is markedly reduced (Ordin and Jacobson, 1955). K+ and Br" 
uptakes are inhibited 32% and 33%, respectively, by 10 milf ^raws-aco- 
nitate, and 63% and 47%, respectively, by 20 mM. These inhibitions are 
probably not specific but the result of the depression of respiration. 

^rans- Aconitate is known to occur naturally in many plant tissues and 
is abundant in sugar cane juice. It is formed from acetate-C^^ in corn tis- 
sues and 95% of the aconitate which accumulates is in the trans form, it 
being out of equilibrium with the cycle acids; further evidence for its com- 
partmentalization during endogenous formation is provided by the fact 
that it is metabolized quite readily when it is added to corn roots (Mac- 
Lennan and Beevers, 1964). It was suggested by Rao and Altekar (1961) 
that it may arise from ci'.s-aconitate through the mediation of an aconitate 
isomerase, which they isolated from soil organisms. Some pseudomonads 
are capable of metabolizing fraw^^-aconitate without previous exposure to 
it and other strains can adapt to utilizing it (Altekar and Rao, 1963). 



FUMARASE 

Fumarase has been studied more intensively than most enzymes with 
regard to interactions with competitive inhibitors, the effects of pH on 
these interactions, and the nature of the active center. A generalized 
representation of the bindings of fumarate and L-malate to the enzyme is 
shown in Fig. 1-6-2, the pH effects are discussed in Chapter 1-14 (page 
691), the apparent p^,'s for various competitive inhibitors are given in 
Table 1-14-2, and the P-K'^'s of the two catalytically active sites for 
fumarase and its substrate complexes are given in Table 1-14-3. 

Emphasis in this section will be directed to a more accurate delineation 
of the active center configuration and to a more quantitative expression 
of the ways in which competitive inhibitors interact with the active center. 
Fumarase possesses four important groups: two cationic groups for binding 
the COO" groups of the substrate in the trans position, and two ionizable 
groups interacting with the groups on the a- and /5-carbon atoms and in- 
volved in the addition or removal of water. The latter enzyme groups wiU 
be designated as Rl and R^ in conformity with Wigler and Alberty (1960); 
each may exist in the protonated form, RlH or Rj^H. The p^^'s of these 
groups, which are 6.3 and 6.9 in the free enzyme, point to their phenolic 
or imidazole nature; indeed, it is possible that these two groups are identical, 



rUMARASE 



275 



but evidence against this comes from the study of inhibition by the tar- 
trates. The catalytically active form of fumarase may be represented by 
EH, in which one of these groups is protonated, although both E and EHg 
are capable of binding both substrates and inhibitors. 

The values of K^ for several competitive inhibitors at pH 6.35 and 23° 
(Table 2-1) may be used as a rough and provisional means of evaluating the 
relative energies of ])inding, bearing in mind that these are apparent K-&, 
that the enzyme exists in three different ionized states for each of which 
the binding is different (so that the K^s are in a sense composite), and 
that the various inhibitors alter the pA^^'s of the enzyme groups in different 
ways in the EI complexes. However, some reasonable conclusions may 
be drawn from these AF values. 

Table 2-1 

Inhibitor Constants" and Relative Binding Energies 
FOR Competitive Inhibitors of Fumarase 



Inhibitor 



Apparent K^ 
(mM) 



Relative —AF 
(kcal/mole) 



Adipate 

Succinate 

Glutarate 

Malonate 

D -Tartrate 

Mesaconate 

Maleate 

L-a-Hydroxy-/5-sulfopropionate * 

D-Malate 

Citrate 

<rans-Aconitate 



100 


1.35 


52 


1.73 


46 


1.80 


40 


1.90 


25 


2.16 


25 


2.16 


11 


2.64 


10 


2.70 


6.3 


2.97 


3.5 


3.32 


0.63 


4.32 



° Values of Ki determined at pH 6.35 and 23°. 

* The Ki for L-a-hydroxy-^-sulfopropionate was changed from 16.5 mM as given 
in the table (Massey, 1953 b) to correspond to the value in the curve presented (16.5 
is probably a misprint for 10.5.) 



[A) Since all monocarboxylates and the methyl ester of fumarate are 
without inhibitory activity, it must be assumed that at least two negatively 
charged groups are necessary for binding. However, it is evident that for 
the more tightly bound substances other attraction forces are involved. If 
we assume that these additional forces arise from hydrogen bonding be- 



276 2. ANALOGS OF ENZYME REACTION COMPONENTS 

tween hydroxyl groups and the Rl and R^ enzyme groups, polarization 
of double bonds, and interactions of a third C00~ group (in the tricar- 
boxylates), the tentative values shown in the following tabulation may be 
assigned for the contributions made by the various interactions to the 
total binding: 

Two C00~ groups 1.75 kcal/mole 

— OH group hydrogen bonding 0.5-1.5 kcal/mole 

— C=C — polarization 1.9 kcal/mole 

Additional CHgCOO- group 0.5 kcal/mole 



The hydrogen bonding and polarization values are minimal since, in part, 
they were derived from the K„'s of the substrates (for fumarate K„^ = 
= 1.78 mM and — JF = 3.71 kcal/mole, and for L-malate Z,„ == 4.0 mM 
and — AF = 3.24 kcal/mole). The ^„/s may not represent dissociation 
constants but in any case the true K^'s would be equivalent to or smaller 
than the K„'s, so that the binding energies for the substrates may be 
somewhat higher. The cis configuration of maleate reduces the attraction, 
but the value for maleate in the table should be corrected since the pK^ 
— 5.9 (at 23° and around 0.1 ionic strength) and only 74% of the total 
maleic acid would be in the form of maleate=: this increases — JF to 2.82 
kcal/mole. The difference in binding between fumarate and maleate is 
thus at least 0.9 kcal/mole. It is also interesting that the introduction of 
a methyl group into fumarate to form mesaconate brings about a 1.55 
kcal/mole or greater reduction in the binding energy, resulting possibly 
from a steric displacement and lowered polarization interaction. The 1.9 
kcal/mole estimated for electrical polarization of the double bond is not 
unreasonable and actually corresponds fairly closely to that calculated, 
using appropriate molar refractions and an interaction distance of 4 A. 
A factor of unknown importance is the possible deformation of the dicar- 
boxylates to fit the active site and the energies that would be involved 
with the different inhibitors. 

(B) If these conclusions are valid, the interaction energy for fumarate is 
approximately half due to coulombic ion-ion forces and half due to the 
inductive polarization by a strong dipole. It is possible that one of the R 
groups on the enzyme is positively charged and the other negatively 
charged on the active enzyme, as suggested by Massey (1953 b). The ionic 
interactions serve to orient the fumarate at the active center, and the 
polarization not only stabilizes the complex but initiates the addition of 
water. 

(C) The third COO" group of citrate and trans-aconitate seems to be 
able to interact with an adjacent positive group on the enzyme, but rela- 



FUMARASE 277 

tively weakly (0.3-0.6 kcal/mole). In fact, the low interaction energies of the 
terminal COO" groups (0.87 kcal/mole for each group, which may be com- 
pared with the 3.3-3.6 kcal/mole binding per COO" group of malonate or 
succinate on succinate dehydrogenase) might indicate that the distance 
between them and the enzyme cationic groups is relatively great (perhaps 
12-15 A), or could even point to a type of interaction other than ion-ion 
attraction. 

(D) L-a-Hydroxy-/?-sulfopropionate is bound at least 0.54 kcal/mole less 
tightly than L-malate, much of the affinity of this analog resulting from 
the OH group interaction. This would indicate that the sulfonate group 
is not a very good substitute for a COO" group in this case. It would be 
interesting, in this connection, to have inhibition data on L-/5-sulfopropio- 
nate. 

(E) When one turns to the effects of pH on the binding of these inhi- 
bitors, it is evident that the situation is more complex than assumed from 
the data at a single pH (see Fig. 1-14-11). Several inhibitors exhibit a pro- 
gressive decline in binding with increase in the pH, but in the case of fu- 
marate the pZ,„ rises between pH 7 and 8, and the pZ,„-pH curve for l- 
malate shows several changes of slope. The inhibitor D-malate also shows 
an increase in binding between pH 7 and 8. Massey (1953 b) suggested that 
binding to different sites might be involved. However, since deviant be- 
havior is noted with substances containing a double bond or OH group, 
it is possible that the affects of pH on the polarization and hydrogen bonding 
interactions may be involved. Succinate exhibits a linear decrease in jiK^ 
with pH whereas fumarate behaves quite differently. Mesaconate and ma- 
leate have succinate-type pH dependences and it is possible that the pola- 
rization interaction is sterically prevented in these substances, as postulated 
above. Apparently some change in the active center ionization occurs 
between pH 7 and 8 which alters these interactions, and we shall return to 
this problem later when more recent inhibition data have been presented. 
(If the crude approach to these problems in the preceding paragraphs 
serves either to irritate or activate others to further theoretical or experi- 
mental study, a purpose will have been accomplished.) 

One would expect the most potent competitive inhibitors of fumarase 
to have either a polarizable group (such as — C=C — ) or a group capable 
of forming hydrogen bonds (such as OH). Substitution of groups at the 
double bond of fumarate seems to reduce the binding, and acetylene-di- 
carboxylate has not been studied. Thus one is left with the tartrates as 
possibly interesting inhibitors, and they were investigated by Wigler and 
Alberty (1960) in an excellent study designed to establish the more intimate 
nature of the catalysis. The variation of the inhibitions with pH allowed the 
determination of the p^^'s of the enzyme groups, the changes in these pro- 
duced by complex ing with the inhibitors, and the dissociation constants 



278 2. ANALOGS OF ENZYME REACTION COMPONENTS 



COO 
H^i .H 
^C^ 

1 


COO' 

1 


COO 
H. i OH 

1 


COO 
H^ i .OH 

^C^ 

1 


1 

H-^^^H 

coo- 


1 
H-^ ; OH 

coo- 


1 
HO^ i ^H 

coo- 


1 

Q 

U^ 1 ^OH 

coo- 


Succinate 


L-Tartrate 


D -Tartrate 


wieso- Tartrate 



of the inhibitors with the variously protonated forms of the enzyme. It 
was found that meso-tartrate is the most potent inhibitor of this group 
and it was concluded that this configuration of OH groups allows the 
formation of two hydrogen bonds with the Rl ^^^ ^d enzyme groups 
(Fig. 2-2). The singly protonated form of the enzyme, EH, binds the 

C-—C FUMARATE 

/ \ 

Rp-H COO 



© 



ooc 



Rjj-H 




L-MALATE 



MESO- TARTRATE 



Fig. 2-2. Simplified scheme of the fumarase active site described by Wigler and 

Alberty (1960). The cationic, R^, and R^ groups occur on different levels and are so 

located they can interact with certain isomers of substrates and inhibitors. 

meso-tartrate most tightly. The binding energies contributed by the hy- 
drogen bonds can be estimated from the relative interactions of succinate 
and the tartrates. Weak hydrogen bonds are indicated for the doubly pro- 
tonated form of the enzyme, EHgl, with D- and L-tartrates, while stronger 
bonds (— 2.8 to — 3.3 kcal/mole) are formed with meso-tartrate and the 
less protonated enzyme (see tabulation). 



INHIBITION OF XANTHINE OXIDASE 279 



AF for displacement of succinate 
(kcal/mole) 

EH2I EHI EI 

D-Tartrate —0.5 +0.4 

L-Tartrate —0.5 +0.8 +0.7 

meso-Tartrate — 1.6 — 2.8 — 3.3 

The fumarase from, liver is inhibited differently from the heart enzyme, 
in that mesaconate (methylfumarate) is inactive whereas citraconate 

H3C ^COO' H COO" V*^ 

C C C=CH2 

II II I ' 

/C^ . ^C^ CH3 

H COO H3C H i,^- 

Citraconate Crotonate Itaconate 

(methylmaleate) inhibits well (Jacobsohn, 1953). Itaconate inhibits less 
potently and crotonate even less potently (about the same as succinate). 
The fact that crotonate inhibits at all is interesting, in that it suggests 
that one C00~ is sufficient if a double bond is present. czs-Aconitate and 
^raws-aconitate inhibit equally and weakly. DL-/5-Fluoromalate inhibits com- 
petitively the conversion of malate to fumarate by fumarase (Krasna, 
1961). Assuming that both optical isomers inhibit equally (which may 
not be true), K^ = 28 mM, and K„^ for malate is 3.5 niM. The inhibition 
of malate dehydrogenase is much stronger, if ^ being 0.16 mM, which may 
be compared to a K,,^ of 11 mM for malate. 

INHIBITION OF XANTHINE OXIDASE 
BY PURINE ANALOGS AND PTERIDINES 

Some potent and specific inhibitors of xanthine oxidase have been dis- 
covered and have proved to be interesting not only on the enzyme level 
but because of the disturbances in purine metabolism produced in whole 
animals. Xanthine oxidase catalyzes the oxidation of hypoxanthine and 



H3C ^coo 

^c 

II 

OOC H 


Mesaconate 






1 1 


H 

1 






H 




Xanthine 



N N' 



Hypoxanthine Xanthine Uric acid 




280 



2. ANALOGS OF ENZYME REACTION COMPONENTS 



xanthine to uric acid, each step essentially involving the addition of water 
and the removal of two hydrogen atoms which are transferred to oxygen 
along a typical electron transport sequence. Hydroxypurines exhibit keto- 
enol tautomerism, and it is only recently that spectroscopic evidence point- 
ing to the predominance of the keto form has been obtained (Mason, 
1957). The structures of the purines and pteridines to be discussed will 
be written as far as possible in conformity to these results. However, in 
some cases it is difficult to assign the most important structure, especially 
in multiply substituted compounds. Ionization may also be a complicating 
factor. Most purines appear to be predominantly neutral at physiological 
pH (piiC^ between 1 and 4; pK^, between 8 and 12), but some, such as uric 
acid, lose a proton in the acid pH range and physiologically exist as anions 
(e.g., pK^ for uric acid is 5.4). The site of loss of the proton is not known 
and it is quite possible that the anions should be considered as being 
equilibrium mixtures of several structures. The form in which they are 
written, hence, does not imply that this structure is the only one present, 
or even that it is necessarily the most important structure. Such considera- 
tions become important in treating the forces between these compounds 
and the enzyme. A final factor must be borne in mind: the most stable 
form in solution is possibly not the form in which the purine is bound to 
the enzyme surface, inasmuch as the interaction may modify the structure 
appreciably. 

Purine Analogs 

Dixon and Tliurlow (1924) reported that xanthine oxidase is inhibited 
by various purines, such as adenine and uric acid, but no quantitative data 
were given. There was no further investigation until the introduction of 
the azapurines. 2-Azaadenine and 2-azahypoxanthine, like many purine 






Purine 



Guanine 



8-Azaguanine 



NH. 




NH, 



NH. 





Pyrazoloisoguanine 



INHIBITION OF XANTHINE OXIDASE 281 

analogs, are oxidized by the enzyme in the 8-position (Shaw and Woolley, 
1952). Eqiiimolar concentrations of 2-azaadenine prolong the formation of 
urate from xanthine 2-fold and from hypoxanthine 4-fold, the inhibition 
being competitive. The kinetics in such situations may be complicated 
by two factors: (1) the disappearance of the inhibitor (Shaw and Woolley 
found, for example, that the azaadenine essentially all disappeared before 
much urate was formed), and (2) the inhibition produced by the product 
of the inhibitor oxidation. 8-Azapurine and all of its monohydroxyl and 
monoamino derivatives are oxidized by xanthine oxidase and the products 
are frequently inhibitory not only to xanthine oxidase but to other enzymes. 
2-Amino-8-azapurine is converted to 8-azaguanine and hence can be used 
as a precursor of this inhibitor (Bergmann et al., 1959). 

Certain inhibitions of xanthine oxidase by purine compounds are sum- 
marized in Table 2.2 The inhibitions are not always competitive despite 
the close similarity of substrate and inhibitor structures. Some of the simple 
analogs are bound more tightly to the enzyme than are the normal sub- 
strates. 6-Chloropurine and pyrazoloisoguanine are bound particularly well 
and this brings up questions regarding the forces involved. Very little is 
known about these forces. Ionic forces must be unimportant and it is pos- 
sible that hydrogen bonds, coupled with an appropriate fit of the bonding 
groups, play a major role. PjTazoloisoguanine is the 4-amino-6-hydroxy 
derivative of pyrazolopyrimidine, and it is interesting to note that the 
4-amino derivative is a very weak inhibitor relatively, as are the 4-methyl- 
amino and l-methyl-4-amino derivatives (Feigelson et al., 1957). It has 
been stated that there is some correlation between the potency of the xan- 
thine oxidase inhibition and the carcinostatic activity of these and related 
compounds. 

When inhibitory purine analogs are administered to animals it is often 
difficult to determine the toxic mechanisms because of the multiple possible 
sites for interference. The biological effects of 6-mercaptopurine seem to 
be related to its conversion to the ribonucleotide, which inhibits inosinic 
acid metabolism, rather than to any direct enzyme inhibition (Silberman 
and Wyngaarden, 1961). On the other hand, it has been postulated that 
8-azaguanine induces a guanine deficiency by inhibiting xanthine oxidase, 
which operates in one guanine biosjoithetic pathway (i.e., hypoxanthine 
-^ xanthine -^ guanine) (Feigelson and Davidson, 1956 a). It has been 
shown in one instance that purine metabolism can be inhibited in vivo. 
6-Chloropurine given to rats at 80 mg kg inhibits the formation of C^^Og 
from xanthine-6-C^* about 40% when administered 20 min before the xan- 
thine (Duggan et al., 1961). This is probably not due to a direct action 
on xanthine oxidase but to the formation of 6-chlorourate and the resulting 
inhibition of uricase. 6-Chloropurine also depresses the conversion of ace- 
tate to lipid, of glycine to protein, and nucleic acid synthesis. 



282 



2. ANALOGS OF ENZYME REACTION COMPONENTS 



P5 









O ^ 

6 6 



^ ^ 



o § 



+ 






CO '^ 






tH c :r cr 



112 1^ 



■^ ^ o 



c3 c3 ^ 'i) 

e c g c c s 

O TO Co CO o f 

C U) M M c ^ 

§ 5 5 5 § -2 



^ C C 
™ Oi (D 



o3 c« 

cS cS 

bC bD 

C C 



^pq hJQQPh-1S^^ 









S fc — I SO s S ;::! 



-ti -ti +i 



!^ ^ ^ 



,ifO -< o 






I I 



+ + i 






.:- .- (D 



.S ^ . " — ^ 



773 O S) ■:3 773 



^ c3 ce 
§ 05 P^ 



O S 



m 



2 5 



(U 



:a ^ 



111 



cS 



CK pJH PQ 



"-'do 



t< t< 






+ 



f=^ 



Q 



s i^ 



o J4 M M 

13 ^ ^ " ^ 
?? fer* ^T-f w P- 



O 



§ 




.g 


§ 


^ 




^ 




s 




s 


s£ 


1^ -^ 


-« 


"^ 


;2 ^ 


f^ T^ 


772 


o 


Si rs 


" g 


S 


O 


" § 



u ^ tl, 



o 



o § 



'S 'S >> ^ 

c c X o 

cs ce o ,i2 

5 5 ^ >. •£ 



<D 



- -3 -^ 

0) cS c 

■d 3 cS 

c3 bC X 

N N SI 






>> K 3 oi <; <c 

'-* & . I I 

C-] GO 00 



:s a 



^ .2 






"5 .-:< 



INHIBITION OF XANTHINE OXIDASE 283 

A new xanthine oxidase competitive inhibitor, 4-hydroxypyTazolo(3,4-d) 
pyrimidine (allopurinol, Zyloprim), is now being clinically tested in hyper- 
uricemia and for the potentiation of the antitumor activity of the 6-substi- 
tuted purines (e.g. 6-mercaptopurine) (Elion et al., 1963; Information for 
Investigators report from the Burroughs Wellcome Company). It is a 




N 

I 
H 

4-Hydroxypyrazolo(3, 4-d)pyrimidine 

very potent inhibitor, with K^ = 0.000032 mM, being bound around 100 
times more tightly to the enzyme than is xanthine, but it is also a substrate 
for xanthine oxidase and its oxidation product is likewise a potent inhib- 
itor, with Ki = 0.000054 mM. Mice and dogs given 100 mg/kg intraperi- 
toneally show an increased urinary excretion of xanthine and hypoxanthine, 
with a decrease in aUantoin, and in man a similar action has been demon- 
strated, serum and urinary urate being depressed. It is well tolerated by 
man at 200-1000 mg/day orally up to several weeks. It apparently has no 
antitumor activity itself but is able to potentiate the action of 6-mercapto- 
purine by interfering with its metabolism. Mice given 20 mg/kg intraperi- 
toneally along with 6-mercaptopurine exhibit a reduced urinary excretion 
of thiourate. The value of this analog in neoplastic disease and gout is 
not yet known. 

Inhibition of Uricase by Purine Analogs 

It is appropriate at this time to refer to certain studies on uricase (urate 
oxidase) before proceeding with the inhibition of xanthine oxidase by the 
pteridines. Uricase catalyzes the oxidative opening of the pyrimidine ring 
to form aUantoin. Many methyl and ethyl derivatives of urate were tested 
by Keilin and Hartree (1936) on an enzyme from pig liver; they found that 
none is a substrate but that several inhibit quite potently. They considered 
the mechanism to be competitive and stated, "The fact that the methyl 
compounds of uric acid, although not oxidizable by the enzyme, inhibit 
the oxidation of uric acid shows that these methyl compounds react with 
the same active grouping of the enzyme molecule as uric acid itself." It 
is very interesting that the 1,3,7-derivative is so much more potent than 
the 1,3,9-derivative, particularly in view of the greater potency of the 
monomethyl compounds compared to the latter derivative (the urate was 
10 mM in all cases) (see tabulation). 2,6,8-Trisubstituted purines were 



284 2. ANALOGS OF ENZYME REACTION COMPONENTS 



Inhibitor 


Concentration 
(mif) 


% 


Inhibition 


1,3,9-Trimethylurate 


7.7 




16 


7-Methylurate 


9 




41 


3-methylurate 


9 




49 


1-methylurate 


9 




58 


1,3,7-Trimethylurate 


7.7 




68 



studied by Mahler et at. (1956) (see tabulation) and the results give some 
information on the nature of the binding to the active site. Three binding 
sites were recognized: (1) a cationic group binding the 2-substituent, 



Substituent in position: 


A', {mM) 








2 


6 


8 




CI 


CI 


CI 


0.0008 


CI 


CI 


OH 


0.0013 


OH 


NH, 


NH2 


0.0018 


OH 


OH 


H 


0.012 


OH 


OH 


OH 


0.025 (A', for urate) 


CI 


NHa 


OH 


0.04 


OH 


NHa 


OH 


0.15 


NH2 


NHa 


OH 


0.5 


NH2 


OH 


NH2 


No inhibition 


NH2 


NH, 


XH, 


\o inhibition 



(2) a neutral group binding the 8-substituent, and (3) the copper which 
chelates with the 6- substituent and the 7-N atom. It is difficult to under- 
stand on this basis the high affinity of the trichlorourate for the enzyme, 
since one would predict that substitution of chlorine in the 2-position would 
reduce binding to the cationic group and substitution in the 6-position would 
interfere with the chelation. The increased binding produced by chlorine 
substitution might be in part the result of a stabilization of conjugative 
resonance and an increased hydrogen bonding (and perhaps an increased 
chelation of two N atoms with the copper). Also resonance with structures 
in which one or more of the chlorine atoms are in the form 

C1+ 

— N-— C— 

would produce strong dipoles. The most potent inhibitor apparently is 
6-chloro-2,8-dihydroxy purine (usually misnamed 6-chlorourate), which is 



INHIBITION OF XANTHINE OXIDASE 285 

formed from 6-chloropurine by the action of xanthine oxidase, K^ being 
0.00006 mif (Duggan and Titus, 1959). It inhibits urate degradation in 
the rat 75% at a dose of 20 mg/kg (Duggan et al., 1961). The over all AF 
for the binding to uricase is approximately 10 kcal/mole and this would 
indicate the formation of some type of stable bond. It was pointed out that 
this analog is very stable and might be useful in studying the metabolism 
and disposition of urate. It should induce urate accumulation in the tissues 
and the effects of this might have some bearing on the manifestations of 
gout. The inhibition of uricase by xanthine (2,6-dihydroxypurine) in the 
tabulation above is interesting because it illustrates a novel effect in a multi- 
enzyme system. In the sequence: 

xanthine 

oxidase uricase 
Xanthine > urate > allantoin 

the initial substrate inhibits the second enzyme in the series, causing ac- 
cumulation of urate in increasing amounts as the xanthine concentration 
rises (e.g., in rat liver homogenates). At high concentrations the urate con- 
centration falls due to the substrate inhibition of xanthine oxidase (Van 
Pilsun, 1953). 

Further studies on uricase have been reported by Bergmann et al. (1963 a) 
and some of the results, including calculations of the approximate apparent 
relative binding energies, are presented in the following tabulation. The 
xV-methylpurines are relatively weak inhibitors and are not included in 
the tabulation. It is clear that the best inhibitors contain a 2-OH group 
(designated as OH for convenience but the keto form is probably dominant), 
and this position is the most important in the binding; substitution of the 
2-0II with a 2-SH group lowers the inhibitory activity markedly. The 
weakening of the binding by A^-substitution points to the importance of 
the imino group for attachment. The inhibitions are generally" sensitive to 
the pH and for most analogs increase with a rise in pH, although for 6- 
SH-8-OH-purine there is a decrease, and with 8-OH-purine there is no 
effect. The pH probably influences the tautomerism, which is quite impor- 
tant since the binding depends on the states of the N and OH groups. The 
K^ for 2,8-diOH-6-SH-purine is 0.0026 niM and for 2,6-diOH-8-SH-purine 
is 0.00039 mM (Bergmann et al, 1963 b). The 2-OH-6,8-diSH-purine (which 
is 6,8-dithiourate) is a much more potent inhibitor and, although its K^ 
was not given, it must be at least around 0.00004 mil/. 

Pteridines 

The inhibition of xanthine oxidase by synthetic folate, reported by 
Kalckar and Klenow in 1948, was soon found to be due to some impurity, 
and the simultaneous observation, by Lowry and Bessey, of the very po- 



2. ANALOGS OF ENZYME REACTION COMPONENTS 



Substituent in position: 


(1)50 (mif) Relative —AF (kcal/mole) 








2 


6 


8 






OH 


CH3S 


OH 


0.00013 


6.80 


OH 


SH 


SH 


0.0004 


6.12 


OH 


H 


Aza 


0.0016 


5.25 


OH 


SH 


H 


0.0027 


4.94 


OH 


OH 


SH 


0.0050 


4.55 


OH 


H 


OH 


0.0052 


4.52 


OH 


OH 


Aza 


0.0059 


4.45 


OH 


CH3S 


H 


0.0060 


4.44 


OH 


H 


H 


0.012 


4.01 


OH 


H 


OH 


0.012 


4.01 


OH 


SH 


OH 


0.014 


3.91 


OH 


CH3 


H 


0.017 


3.78 


OH 


OH 


H 


0.018 


3.76 


H 


CH3S 


OH 


0.032 


3.40 


OH 


OH 


CH3S 


0.038 


3.30 


H 


OH 


OH 


0.066 


2.95 


H 


SH 


OH 


0.070 


2.91 


SH 


OH 


SH 


0.080 


2.84 


H 


H 


OH 


0.11 


2.65 


SH 


SH 


OH 


0.15 


2.44 


SH 


OH 


H 


0.19 


2.30 


H 


OH 


H 


0.22 


2.21 


SH 


OH 


OH 


0.25 


2.13 


SH 


H 


OH 


0.50 


1.70 


H 


SH 


OH 


0.50 


1.70 


H 


SH 


H 


0.70 


1.19 



tent inhibition produced by 2-amino-4-hydroxy-6-pteridyl aldehyde, a 
photolytic product of folate, led to the conclusion, subsequently verified, 
that this was the contaminant (Kalckar et al., 1948; Lowry et al., 1949 a, 
b). For convenience we shall follow Hofstee (1949) in designating 2-amino- 
4-hydroxypteridine as pterin and the inhibitor thus as pterin-6 -aldehyde* 
Pterin and many of its substituted derivatives are oxidized to varying 
degrees by xanthine oxidase while other derivatives are only inhibitory. 

* This substance has been variably called 2-amino-4-hydroxy-6-pteridyl aldehyde, 
2-amino-4-hydroxy-6-formylpteridine, 6-formylpteridine, pteridylaldehyde, 2-amino- 
4-hydroxy-6-pteridine carboxaldehyde, and 2-amino-4-hydroxy-6-formylpterine, in 
most cases without either justification or accuracy. 




INHIBITION OF XANTHINE OXIDASE 
O 



H,N 



287 





H,N 



H,N 




«^N^ 


1 "^ 


HsN-^N-^ 


^N^ 


Pterin-6- 


-aldehyde 


O 


H 

1 


H-nA 


^Ny-o 


H2N^^^N-^ 


1 




1 

H 



Xanthopterin Isoxanthopterin Leucopterin 

Actually all the photolytic oxidation products of folate are inhibitory 
but pterin-6-aldehyde, the primary product: 

Folate -> pterin-6-aldehyde -> pterin-6-carboxylate -> pterin -> isoxanthopterin 

is by far the most potent; indeed, it is one of the most potent inhibitors 
known. 

Inhibition of xanthine oxidase by pterin-6-aldehyde is observed at a 
concentration of 2 X 10^* //g/ml or roughly 10"^ M (Lowry et at., 1949 a). 
Competitive inhibition with respect to both xanthine and pterin has been 
demonstrated, and a /iC, of 6 x 10"' mM calculated for the milk enzyme 
(Lowry ef al., 1949b). It was shown that 35% inhibition occurs when enzyme- 
FAD = 9.3 X 10-6 jnM, the substrate pterin = 78 x 10-« mM, and pte- 
rin-6-aldehyde = 2.26 X 10-« mM, this indicating that 2.26 X 10-« mM 
inhibitor completely blocks 3.3 X 10-^ mM enzyme (on the basis of FAD 
content). It may have been that all the FAD was not catalytically active 
or that more than one FAD molecule was associated with one enzyme 
molecule. However, there is no doubt that this is a mutual depletion system 
and that pterin-6-aldehyde titrates the enzyme. A few other reports will 
be mentioned to illustrate the potency. Both the milk and rat liver enzymes 
are inhibited, 40-50% inhibition occurring at 5-8 X 10"^ mM when xan- 
thine is 0.07 mM (Kalckar et al, 1950). The xanthine oxidase from Clostri- 
dium cylindrosporum is potently inhibited by 0.0002 mM (Bradshaw and 
Barker, 1960). Milk xanthine oxidase is completely inhibited by pterin-6- 
aldehyde at 0.0033 mM when hypoxanthine is 3.33 mM (Petering and 
Schmitt, 1950), and the rat intestine enzyme is inhibited completely by 
0.067 mM when hypoxanthine is 6.6 mM (Westerfeld and Richert, 1952). 
In many experiments the substrate concentrations have been unnecessarily 
high since maximal rates are usually obtained at concentrations well below 



288 2. ANALOGS OF ENZYME REACTION COMPONENTS 

0.1 niM, and hence the true potency of the inhibition has been somewhat 
obscured. Pterin-6-aldehyde is actually oxidized very slowly by xanthine 
oxidase and reversal of the inhibition develops gradually. The oxidations 
of aldehydes (Kalckar et al., 1950) and sulfite (Fridovich and Handler, 
1957) by xanthine oxidase are also inhibited by pterin-6-aldehyde, but the 
oxidation of NADH is not affected (Lowry et al., 1949 b). 

Although it has generally been stated that the inhibition by pterin-6- 
aldehyde is competitive, the Ijv — 1/(S) plots are not linear (Bradshaw 
and Barker, 1960), and others (Hofstee, 1949) have presented only quali- 
tative evidence for competition. Deviations from the classic kinetics might 
be expected because of the mutual depletion of free enzyme and inhibitor 
concentrations, the difficulty in achieving true equilibrium, and the oxida- 
tive removal of the inhibitor. There is no evidence against a competitive 
mechanism, however, and competition has been more clearly demonstrated 
for some other pteridines. 

The inhibition by pterin-6-aldehyde is quite specific although it must 
be admitted that not many enzymes have been tested. Uricase, glucose 
oxidase, and 3-phosphoglyceraldehyde dehydrogenase are not affected, but 
the quinine oxidase of liver is inhibited (Kalckar et al. 1950; Villela, 1963). 
Mouse liver guanase, using 8-azaguanine as a substrate, is inhibited but 
not potently when the substrate is 11 mM and 30-min preincubation is 
allowed (see accompanying tabulation) (Shapiro et al., 1952). In fact, 



Pterin-b-aldehyde n, n ■ i i x- 

/n Guanase inhibition 
(mi/) 



1 32 

6 62 

11 80 

16 88 



xanthopterin is a more potent and more rapidly acting inhibitor (Dietrich 
and Shapiro, 1953 b). Pterin-6-aldehyde is not carcinostatic itself, but 
potentiates the action of 8-azaguanine and a suppression of 8-azaguanine 
destruction was claimed as the mechanism, although the relatively low 
inhibitory potency coupled with the low doses (20 mg/kg) necessary makes 
it difficult to accept this explanation. Byers (1952) investigated the effects 
of pterin-6-aldehyde injections in rats (200 mg/kg intraperitoneally) on 
the tissue urate levels and found no significant changes, which might in- 
dicate a rapid destruction of the inhibitor in the animal. Daily injections 
of 30 //g pterin-6-aldehyde in chicks also does not alter liver xanthine oxi- 
dase activity despite the potent inhibition in vitro (Dietrich et al., 1952). 
Other pteridines, although less potent than pterin-6-aldehyde, are nev- 



INHIBITION OF XANTHINE OXIDASE 289 

ertheless effective inhibitors of xanthine oxidase. The following tabulation 
shows the inhibitions after 20 min incubation at pH 8.5 when the substrate 



Inhibitor 


Concentration 
(mM) 


% Inhibition 


Pterin-6-aldehyde 


0.031 


100 


Pteroate 


0.033 


100 


Xanthopterin 


0.033 


72 


Isoxanthopterin 


0.066 


95 


Leucopterin 


0.032 


34 


7-methylxanthopterin 


0.077 


91 


6-MethyUsoxanthopteiin 


0.077 


89 


Xanthopterin-7-carboxylate 


0.050 


62 


Isoxanthopterin-6-carboxylate 


0.075 


23 


Pterin-6-carboxylate 


0.055 


15 


2,4-Diamino-6,7-dihydroxypteridine 


0.040 


76 



concentration is 0.063 mM (Hofstee, 1949). The potency relative to pterin- 
6-aldehyde is not seen here since 0.001 mM inhibits 82% under these con- 
ditions. From the K„, for xanthine of 0.02 mM (Hofstee, 1955), a K^ for 
pterin-6-aldehyde of 5.1 X 10"^ mM may be calculated, which is higher 
than was obtained by Lowry et al. (1949 b). The inhibition by xanthopterin 
appears to be completely competitive (i^,„ for xanthine 0.0053 milf , and K, 
= 0.0016 mM), and xanthopterin-7-carboxylate is of comparable potency 
(Krebs and Norris, 1949). Bovine serum xanthine oxidase is inhibited 94% 
by 0.052 mM pterin-6-aldehyde but only 3% by xanthopterin at a com- 
parable concentration (Villela et al., 1956). There is no doubt that the al- 
dehyde group at the 6-position confers a strong affinity for the enzyme, 
since when it is reduced to a CHgOH group, oxidized to a COO" group, 
or altered to a CHg group, the inhibitory activity is markedly reduced (Pe- 
tering and Schmitt, 1950). Pterin-6-aldehyde is bound to xanthine oxidase 
approximately 4.9 kcal/mole more tightly than xanthopterin, and it would 
appear that pterin itself is bound somewhat less tightly than xanthopterin. 
It is tempting to relate the augmenting action of the aldehyde group to 
the fact that xanthine oxidase oxidizes simple aldehydes. There must be 
a site on the enzyme capable of reacting with aldehyde groups, and it is 
possible that the pterin-6-aldehyde is bound in a configuration such that 
the aldehyde group interacts in this manner. Support for this interpretation 
comes from the observation by Lowry et al. (1949 b) that pterin-6-aldehyde 
is slowly oxidized to pterin-6-carboxylate by xanthine oxidase. 



290 



2. ANALOGS OF ENZYME REACTION COMPONENTS 



CHOLINE OXIDASE 

The inhibitions of liver choline oxidase reported by Wells (1954) (Table 
2-3) recall the interferences exerted by these and similar analogs on cho- 
linesterase and tissue acetylcholine receptor groups. It would appear that 
a certain critical distance between the N+ group and the terminal CH2OH 



Table 2-3 
Inhibition of Rat Liver Choline Oxidase by Choline Analogs" 





Inhibitor 


Relative 

rate of 

oxidation 


Equimolar 

% 
inhibition 


Relative 
AF of 


Series 


Ri 


R2 


R3 


binding 
(kcal/mole) 


Ethanolamine 


H 


H 


H 





23 







H 


Me 


Me 





53 


- 0.81 


R 


H 


Et 


Et 





25 


- 0.06 


R-N^^CHoCHgOH 
R^ 


Me 
Me 


Me 

Et 


Et 
Et 


79 
24 


- 


- 




Et 


Et 


Et 





11 


+ 0.55 


3- Aminopropanol- 1 


H 


H 


H 





25 


- 0.06 


R 


Me 


Me 


Me 


5 


15 


+ 0.33 


R— N^!— CH2CH2CH2OH 
R 


H 

Me 


Et 
Et 


Et 
Et 



8 


18 

7 


+ 0.19 
+ 0.88 




Et 


Et 


Et 





14 


+ 0.38 


1- Aminopropanol -2 


Me 


Me 


Me 


66 


- 


- 


R CH3 
R— N^— CH.— CH- OH 
R 


H 

Me 
Et 


Et 
Et 
Et 


Et 
Et 
Et 




14 

4 


13 

6 

16 


+ 0.43 
+ 0.96 
+ 0.29 


2-Amino-2-methylpropanol-l 














R CH, 

R— N^— C-CH2OH 

/ 1 
R CH3 


H 

Me 
Et 


H 
Me 
Et 


H 
Me 

Et 


4 

105 

8 


64 
65 


- 1.09 

- 1.12 


2 - Amino - 2 - methy Ipropanediol 














R CH, 

R— N^-C-CHjOH 

R CH.OH 


H 
Me 

Et 


H 

Me 
Et 


H 

Me 

Et 


3 

41 

5 


41 
50 


- 0.51 

- 0.74 



"Experiments done with rat liver homogenates. Choline and all analogs at 37.5 mM. The 
rates of oxidation are given relative to that for choline (100). (From Wells, 1954.) 



INHIBITION OF NITROGEN FIXATION BY OTHER GASES 291 

group is necessary for substrate activity. The most reactive substrates 
are dimethyl or trimethyl compounds, indicating that an exact fit of the 
cationic head is necessary to place the hydroxyl group in position for 
oxidation. Substitution of groups on the C-1 atom reduces the binding 
whereas substitution on the C-2 atom increases the affinity even though 
the groups are fairly bulky. The simplest interpretation is that the cationic 
head anchors the molecules in position so that the CH2OH group can react 
with an enzyme group on the opposite side of a hole or slit in the protein. 
When the analogs are too long they do not readily fit into this region, 
whereas groups protruding from C-2 interact by van der Waals' forces 
with the walls of the cavity. A three-point attachment of the cationic head 
is suggested by the reduction in affinity brought about by altering only 
one of the R groups, this perhaps tilting the molecules so that the hydro- 
carbon chain is not in the normal direction. The differences in binding 
energies between these analogs are rather small and this might indicate 
that dispersion forces are mainly involved, but it may also be that changes 
in the electrostatic interactions (resulting from the different volumes of 
the R groups, for example) are offset by opposite changes in the dispersion 
energy. Since choline must be oxidized to betaine before it can serve as 
a methyl donor, it is interesting that Wells demonstrated the inhibition 
of methionine synthesis in liver homogenates by 2-amino-2-methyl-l- 
propanol and its triethyl derivative. Niemer and Kohler (1957) studied 
analogs of choline in which one of the methyl groups is substituted by var- 
ious radicals (e.g., — CHoCH.^OH, — CHaCH^Br, — CH2CH=CH2, — CH2= 
=CH2, and —CH2COO-) and found 10-20% inhibition of liver choline 
oxidase at concentrations approximately equimolar with choline (11.5 mM). 
None of these analogs is a potent inhibitor, confirming the importance of 
fit at the cationic head. 



INHIBITION OF NITROGEN FIXATION BY OTHER GASES 

Some simple instances of competitive interference between gases in 
nitrogen fixation and hydrogen evolution have been observed, and are 
reminiscent of the suppression of hemoglobin oxygenation by carbon mon- 
oxide, nitric oxide, and other gases. The primary product of nitrogen 
fixation in microorganisms is probably ammonia, and the enzyme system 
responsible for this is generally termed nitrogenase. A few examples of 
inhibition are given in Table 2-4. Most of these have been shown to be 
strictly competitive. CO and NO are the most potent inhibitors while H2 
and NgO are relatively weak. O2 is a special case in that as pOg is increased 
from zero the nitrogen fixation accelerates, but above a certain value, 
depending on the organism, the rate falls off (Burris, 1956). Ethane, neon, 
argon, and helium have no significant effects (Molnar et al., 1948). 



292 



2. ANALOGS OF ENZYME REACTION COMPONENTS 



03 



O 



Zl- 




05 


05 










3 




-ti 


cS 




Tin 


oo 






3 


C 




C5 


05 


40 




^ 


o 

03 


o 


^ 




05 


cS 


lO 


73 




lO 


"e 


'e 






05 


c 


T3 


Oi 






^ 


o 


^ 


c« 




:^ 


«u 


<» 


e 


"C 


t4 




.52 


O 

03 


Sh 

C3 


Si) 


a 


^ 


^ 


U 


'C 


_a 


..^ 


>> 


;h 


tH 


(D 


Sh 






ci3 








O 
O 


3 

m 


s 


o 


K 



2 "^ 



m 



Zji 


^ 


O 


Oi 


•^ 


73 






6; 


3 


73 


— 


Si) 


cS 


i 


c 




ai 




0) 




m 


o 




O 









N — a ja 



03 



pq PQ 





s 


q 
d 




^"^ 


c3 












-S 




4^ 












d 


Y 
















eS 




o3 




X 
















SO 






cc 


















g 
d 


S 
o 

m 


o 
d 

II 


a 

c6 

CO 

O 

d 


'3 

a 


o 
d 

II 




+3 

O 

d 


o 

d 

II 




C 
cS 
CD 

d 

II 


O 
? 

d 


o 
d 

II 

O 


O 

d 
II 

s 




O 

3 
O 


a 

CO 

d 


-2 
cS 

<M 

O 
O 

d 
II 
O 


II 

d 


51, 


]3 

.2 


d 


d 
II 

d" 


+2 
c 


a 




+3 
_0 




03 


O 


3 

_2 


a 


3 
o 


O 


4^ 




.^ 


-»^ 


> 


'■+3 


^^ 




-l-i 


g 




4^ 






4^ 




-p 


-ts 




> 


|S 


03 


IS 




03 


IS 


c3 


_2 






IS 
IS 




IS 


cS 


3 




.2 


^ 
» 


IS 


o 




d 
II 


2 


03 
> 
'-2 


II 


IS 


IS 


3 

,2 


IB 


o 
d 

II 




.2 

'+3 
IS 


.2 
IS 


IS 


|S 


> 


IS 


_2 


;^ 


IS 

IS 


.3 


"S 


IS 


> 
'-+3 


"-I3 


IS 

3 


"S 




[c 


'42 


■"" 


'S 


'■+3 


^~ 


_c 


'■S 


^^ 


-is! 
03 




[c 


'-3 


^" 


+3 






vO 




& 


-.o 


IS 


& 






& 




^ o 




& 




a 


■^ 


^ 


o~- 


^o 




o"- 




p 


TJ 


^o 




T3 


O"" 


-^o 




TJ 


3 


o^ 


o^ 


o 


o"-- 


s 


O 


^ 




j2 


o - 


^ 


C 


O 


o^ 


3 


3 


O 


O 


C' 


«C 


o 


O 


c 


5 


ce 


lO 


o 


03 


O 


O 


O 


cS 


o 


o 


o 




CD 


O 




1 — 1 


o 




M 


O 






^ 


o 




o 







o 


-« 


'^ 


s 




e 


-2 




"S 




o 




3 


*, 





s 






s 






e 


•.!^ 






'« 




s 


s 


r2 


CO 


-2 


"S 

73 


e 


.s 


?5, 


§ 


O 






3 


S 






s 


3 


•«^ 


ts 


^ 


e 


£ 


o 


>. 


1 


o 



^ Ul "% 



o 



o 



o 


o 


3 


3 


3 


3 


03 


o3 


<o 


<D 


X> 


XI 


>> 


>. 


O 


O 


02 


03 



INHIBITION OF NITROGEN FIXATION BY OTHER GASES 293 

Inhibition of Azotohacter growth by N2O occurs if N2 is the sole source 
of nitrogen, but does not if ammonia (Repaske and Wilson, 1952) or nitrate 
(Mozen et al., 1955) is present, indicating that the site of the block is prev- 
ious to these substances. Likewise, in Clostridium H^ inhibits uptake of 
N2 but not of ammonia (Hiai et al., 1957). These inhibiting gases are oc- 
casionally utilized. For example, NgO is assimilated by Azotohacter, although 
slowly, and this is inhibited by N, and Hg (Burris, 1956). It is quite likely 
that the competition in all these cases, with the possible exception of Og, 
is at the nitrogenase active site binding Ng. Nitrogenase is a metalloflavo- 
protein containing molybdenum and it is reasonable that these gases are 
bound to the metal, the catalysis of Ng reduction being of similar type to 
those mediated by various inorganic metal preparations (which are also 
inhibited by other gases). 

It will be necessary before discussing mechanisms of inhibition in greater 
detail to consider the enzyme hydrogenase, which has recently been closely 
linked to nitrogen fixation, and its inhibitions. This enzyme catalyzes the 
reduction of some unknown primary acceptor by molecular Ho and the 
reduced acceptor then transfers the hydrogen atoms to other acceptors, 
such as dyes, NAD, or eventually oxygen. It has been postulated that it 
may in some instances participate in the reduction of Ng. The inhibition 
by O2 is primarily due to oxygenation of the enzyme: 

E + nO, ^ E(02)„ 

and this inhibition is reversible upon removal of the O2 by dialysis (Krasna 
and Rittenberg, 1954; Fisher et al., 1954). Prolonged exposures to Og 
lead to progressive inactivation of bacterial hydrogenase (Shug et al., 
1956). Although n has generally been assumed to be 1, Atkinson (1956) 
obtained rather complex data possibly indicating a value of 2 for the 
Hydrogenomonas facilis enzyme. The hydrogenase-catalyzed evolution of 
H2 in Rhodospirillum rubrum (Lindstrom et al, 1949) and soybean root 
nodules (Hoch et al, 1960) is inhibited by Ng. Although this might be at- 
tributed in part to a diversion of the flow of hydrogen atoms to the reduction 
of N2, Bregoff and Kamen (1952) observed that 1 mole of N2 prevents 
the release of several moles of Hg. One of the difficulties in assuming a direct 
competition between Hg and Ng for the hydrogenase active site is the 
fact that, despite the inhibition of Hg evolution by N2, the exchange reac- 
tion whereby HD is formed from D2 and a hydrogen donor is actually accel- 
erated by N2 (Hoch et al., 1960). The Hj inhibition of nitrogen fixation 
was previously claimed to be competitive, but Parker and Dilworth (1963) 
found that Hg causes a lag in the N2 uptake at low pN2, whereas in cells 
of Azotohacter vinelandii adapted to Hg the lag is abolished. Taking the 
lag phase into account, the inhibition is not competitive; from the reciprocal 



294 2. ANALOGS OF ENZYME REACTION COMPONENTS 

plot presented, although the point scatter is marked, the inhibition might 
be uncompetitive. The inhibition of nitrogen fixation in soybean root nod- 
ules by O2 is complex, due to both plant and bacterial components of the 
respiration, but is competitive when the O2 is above 80% (Bergersen, 
1962). NgO is a somewhat more potent inhibitor of H2 evolution than is 
Ng, but the most potent is NO, complete and irreversible inhibition being 
produced by concentrations of 1% or greater (Shug et al., 1956). The hy- 
drogenase in cell-free extracts of Proteus vulgaris is inhibited 87% by 
0.002% or 0.00004 m.M NO, and the inhibition at these low concentrations 
is partially reversible (Krasna and Rittenberg, 1954). The NO is neither 
oxidized nor reduced by the enzyme. 

Although the configurations, electronic structures, and physical prop- 
erties of these simple gases must be important in determining the inter- 
action with nitrogenase and hydrogenase, it is difficult to establish correla- 
tions. Some structural and physical properties that might relate to the 
interactions of these molecules are given in Table 2-5. Comparing the effects 
of H2, N2O, ethane, and the rare gases on Azotobacter nitrogen fixation, 
Molnar et al. (1948) concluded there is no correlation with the van der 
Waals constants and doubted if any mechanism could be based on physical 
properties alone. Wilson and Roberts (1954) postulated that N2O is inhi- 
bitory because the N — N distance is close to that in Ng', since NgO is linear, 
the oxygen might neither interfere nor be involved in the binding. If the 
binding is to metal groups on the enzymes, the degree of interaction would 
depend more on the types of bond possible and hence on the electronic 
structures of both the gases and the metal, as it is in the interactions of 
O2, CO, and NO with hemoglobin and cytochrome oxidase. 

The inhibition of nitrogen fixation by O2 has been explained as a compe- 
tition between N2 and 0, as terminal acceptors for electrons originating 
in the oxidation of substrates by various dehydrogenases, nitrogen fixation 



SH, »- XH, 



being considered as a form of respiration (Parker, 1954; Parker and Scutt, 
1958, 1960). It is likely that the interrelationships between N, and Hg 
metabolism, and the inhibitions on these systems, must be considered in 
the light of a hydrogen or electron pool with all the possible pathways for 
formation and utilization of hydrogen atoms. The scheme below, modified 
from Gest et al. (1956), may serve as a means of visualizing some of these 
pathways. Some of the inhibitions observed are due to competition for 




INHIBITION OF NITROGEN FIXATION BY OTHER GASES 



295 



^ 






;0 :0: 



+ ^ +;?; 



;^ ^ 



^ o 



X 
O 



o 



p-( 



°<; 



S I T^ o o 8 2 
§ I ^ X X "^ X 



m 



s ^ 






§ 5 



fi S- -^ O) 



o ;=; 



1 ^ 



296 



2. ANALOGS OF ENZYME REACTION COMPONENTS 




Photolysis of water 



SH, 



(nitrogenase) 

N2 (forming NH3) 

available hydrogen atoms, while others are based on direct competition 
at the enzyme active sites. 



PHENOL OXIDASES 

These enzymes usually hydroxylate monophenols in the oriho position 
and further oxidize these to o-quinones. They have often been named ac- 
cording to the particular substrate chosen: e.g., catechol oxidase (cate- 
cholase), cresolase, tyrosinase, phenolase, o-diphenol oxidase, or polyphenol 
oxidase. Different specificities are observed with enzymes from various 
sources. The role such enzymes play in tyrosine metabolism will be dis- 
cussed in the following section. 

The competitive inhibition of potato catechol oxidase by resorcinol was 
first observed by Richter (1934). He noticed that the enzymes from various 
sources exhibited quite different susceptibilities to resorcinol, those from 
elder (Sambucus) and lilac {Syringa) being more sensitive than the potato 
enzyme, and those from mushroom (Polyporus) and mealworm {Tenebrio) 
less sensitive. The respiration of apple skin is inhibited about 35% by 
50 mM resorcinol and thus Hackney (1948) studied an extracted catechol 
OH OH OH 






"OH H3C ^-^ OH 

Resorcinol Orcinol 




COOH 



C^ 



CH-CH— COOH 



Nicotinic acid 



Cinnamic acid 



PHENOL OXIDASES 297 

oxidase. Although the inhibition of this enzyme is reduced by increasing 
catechol concentration, her data do not correspond to pure competitive 
inhibition and the derived K^ seems to be in error, as pointed out by War- 
ner (1951). A potato enzyme oxidizing p-cresol is inhibited by resorcinol, 
phloroglucinol, and orcinol (see tabulation), the inhibition being completely 

Inhibitor (10 mM) % Inhibition (p-cresol = 10 mif) 



Resorcinol 78 

Phloroglucinol 43 

Orcinol 20 



reversible by dialysis and apparently competitive (Schneider and Schmidt, 
1959). 4-Chlororesorcinol is a much more potent inhibitor of the potato 
enzyme, oxidizing either catechol or p-cresol, and it has been claimed that 
the initial inhibition is competitive, although progressive inactivation oc- 
curs (Heymann et al., 1954). A K^ of 0.024 tqM was calculated. However, 
the double reciprocal plots seem to me to be of the perfectly noncompetitive 
type. Bonner and Wildman (1946) postulated that the bulk of spinach leaf 
respiration passes through a polyphenol oxidase. p-Nitrophenol is a rather 
potent inhibitor of this respiration, 1 voM inhibiting 94%, and of the po- 
lyphenol oxidase, whether it is oxidizing catechol or p-cresol. On the other 
hand, o-nitrophenol is only a weak inhibitor of both. It was felt that p-nitro- 
phenol might well be an analog of naturally occurring substrates, and o- 
nitrophenol an analog of o-phenols against which the enzyme is inactive. 
It is interesting, finally, to note that dihydroxymaleate is an inhibitor of 
catechol oxidase (Florkin and Duchateau-Bosson, 1939), and it was sug- 
gested that the 

OH OH 

— C = C— 

grouping complexes with the enzyme in a manner similar to the analogous 
catechol grouping. Most of this work on inhibiting phenolic compounds 
is unsatisfactory from the quantitative standpoint and clear-cut proofs 
of competitive inhibition are lacking. Nevertheless, effective analog in- 
hibition for this class of enzymes has heen demonstrated. 

We shall now turn to a more potent, more interesting, and more 
thoroughly studied type of inhibitor, namely, the benzoates. The inhibition 
of mushroom catechol oxidase by benzoate itself was attributed to a com- 
petition with the substrate for the active center, although no direct evidence 
for this was adduced (Ludwig and Nelson, 1939; Gregg and Nelson, 1940), 
but good competitive kinetics for the inhibition by m-hydroxybenzoate 



298 



2. ANALOGS OF ENZYME REACTION COMPONENTS 



were observed later (Warner, 1951), with K^ 2.5 mM for the potato en- 
zyme and 0.6 mM for the mushroom enzyme. An excellent investigation 
by Kuttner and Wagreich (1953) of the inhibition of a catechol oxidase 
from Psalliota campestris provides data from which some ideas of the me- 
chanism may be obtained. Some of the inhibition data are presented in 
Table 2-6, and have been used to calculate the apparent relative binding 
energies. However, these inhibitors are mostly weak acids and the degree 

Table 2-6 

Inhibition of a Phenol Oxidase from Psalliota campestris with Catechol ( 1 .82 mM) 
AS Substrate (at pH 5.2 and 25°) 



Inhibitor 


Concentration 

(mM) 


% Inhibition 


Relative 
— zli^ of binding" 






(kcal/mole) 


Benzoate 


0.012 


50 


6.96 


p-Chlorobenzoate 


0.023 


50 


6.56 


p-Methylbenzoate 


0.04 


50 


6.23 


o-Chlorobenzoate 


0.21 


50 


5.21 


p-Methoxybenzoate 


0.26 


51 


5.10 


Nicotinate 


0.34 


48 


4.88 


o-Hydroxybenzoate 


0.38 


50 


4.84 


p-Nitrophenol 


0.48 


50 


4.70 


p-Hydroxybenzoate 


0.49 


50 


4.69 


irans-Cinnamate 


0.81 


62 


4.43 


Phenylacetate 


0.77 


50 


4.40 


o-Methylbenzoate 


1.0 


50 


4.25 


o-Nitrophenol 


1.1 


50 


4.19 


Benzoate methyl ester 


0.65 


30 


3.94 


Hydroquinone 


2.2 


50 


3.76 


p-Nitrobenzoate 


0.72 


19 


3.55 


Orcinol 


3.2 


50 


3.53 


Resorcinol 


4.5 


50 


3.32 


o-Nitrobenzoate 


0.72 


8 


2.95 


o-Methoxybenzoate 


8.0 


49 


2.94 



" The relative energies of binding to the enzyme were calculated assuming competi- 
tive inhibition and without taking into account the state of ionization. (Data from 
Kuttner and Wagreich, 1953.) 



PHENOL OXIDASES 



299 



of ionization at the experimental pH of 5.2 varies. It was found that the 
inhibition by benzoate decreases with rise in the pH (see tabulation) and 



pH 



% Inhibition by benzoate (0.0123 mil/) 



% Un-ionized 



5.2 
5.8 
6.4 
7.0 



56 

28 



9.0 
2.5 
0.6 

0.2 



similar results were obtained with some substituted benzoates. This was 
interpreted to mean that the un-ionized form of the inhibitor is the active 
one, for example that benzoic acid is the inhibitor and not the benzoate 
ion. The relative binding energies have been recalculated (Table 2-7) on 
this basis and a somewhat different order of potency is obtained. This il- 

Table 2-7 

Inhibition of Mushroom Phenol Oxidase Corrected for the State of Ionization 

OF THE Inhibitors " 









Corrected relative 


Inhibitor 


P^a 


(HA)/(A,) 


— AF of binding 
(kcal/mole) 


Benzoate 


4.203 


0.0914 


8.44 


o-Chlorobenzoate 


2.943 


0.0055 


8.39 


p-Chlorobenzoate 


3.978 


0.0565 


8.32 


o-Hydroxybenzoate 


3.001 


0.0063 


7.95 


p-Methylbenzoate 


4.371 


0.129 


7.47 


o-Nitrobenzoate 


2.173 


0.00094 


7.24 


p-Methoxybenzoate 


4.470 


0.157 


6.24 


o-Methylbenzoate 


3.909 


0.0487 


6.10 


p-Nitrobenzoate 


3.425 


0.0165 


6.07 


p-Hydroxybenzoate 


4.559 


0.186 


5.71 


Nicotinate 


4.854 


0.311 


5.59 


o-Methoxybenzoate 


4.092 


0.0724 


4.56 



° Relative binding energy corrected on the basis that the un-ionized forms of the 
inhibitors are active. The ionization constants were obtained from Dippy (1939); 
inhibition data are given in Table 2-6 (Kuttner and Wagreich, 1953.) 



300 



2. ANALOGS OF ENZYME REACTION COMPONENTS 



lustrates the importance of taking ionization into account in the comparison 
of inhibitors, as discussed earlier in the chapter. It must be emphasized 
that the absolute values of the binding energy are meaningless; it is only 
the differences between the — JF values that are significant. 

Before discussing the implications of these results we will examine the 
data reported by Krueger (1955) on the inhibition of a mushroom enzyme 
oxidizing p-cresol (Table 2-8). Confirming the work of Kuttner and Wag- 



Table 2-8 
Inhibition of p-Cresol Oxidation by Mushroom Tyrosinase ' 



Inhibitor 


Concentration 
(mif) 


% Inhibition 


Relative 
— AF oi binding 








pH 5.3 


pH7.0 


(kcal/mole) 


Benzoate 


4* 


90 


27 


5.20 


Oxalate 


4* 


83 


— 


4.78 


Cyclohexanecarboxylate 


16 


84 


— 


3.57 


Phenylacetate 


4 


54 





3.51 


Fluoride 


40* 


72 


14 


2.99 


Butyrate 


40 


73 


10 


2.60 


Bromide 


40* 


56 


5 


2.56 


Iodide 


40* 


54 


8 


2.51 


Benzamide 


4* 


8 


— 


2.33 


Lactate 


20 


40 





2.16 


Terephthalate 


4 


11 


— 


2.12 


Phthalate 


4 


10 


— 


2.05 


Chloride 


40 


45 


5 


1.85 


Formate 


4 


6 


— 


1.71 




20 


24 


— 






40* 


10 







Acetate 


20 


21 


— 


1.68 




40 


42 


6 




Trimethylacetate 


40 


30 


4 


1.46 


Chloroacetate 


20 


16 





1.39 


Benzenesulfonate 


40 


20 





1.13 


Trichloroacetate 


20 


8 


— 


0.91 



" Concentration of p-cresol was 4.63 mM except where indicated by asterisks, in 
which cases it was 9.26 n\M. The relative binding energies were calculated for pH 5.3 
and were not corrected for ionization. (Data from Krueger, 1955.) 



PHENOL OXIDASES 301 

reich, a marked decrease in the inhibition with an elevation of the pH 
from 5.3 to 7 is observed, but Krueger interpreted this as indicating an 
ionizing group on the enzyme with a pK^ around 6 and possibly an imida- 
zole group. One reason for assuming the ionizing group to be on the enzyme 
is the pH effect on inhibitions by the inorganic anions; however, the in- 
hibition by these ions may be through a different mechanism than the 
benzoates, and indeed Krueger found noncompetitive kinetics for chloride. 
The substrates for the enzyme are, of course, un-ionized and this might 
favor the concept of the acid form of the inhibitors being active and the 
important ionizing group on the inhibitors. It is impossible at the present 
time to decide which is the correct interpretation and hence the signifi- 
cance of the pH effects. Ionizing groups on both enzyme and inhibitors 
might also be considered. It should be added that the following were found 
to be without effect: sulfate (40 mM), nitrate (40 mM), pyrophosphate (40 
mM), pyruvate (20 mM), succinate (20 mM), maleate (20 mM), fumarate 
(20 mM), and ethyl benzoate (4 mM). 

The lack of information on the exact catalytic mechanism involved in 
these enzymes, and particularly our ignorance of the state and role of 
copper, make it difficult to understand the binding of the inhibitors. There 
are two copper ions at the active center but we do not know if they com- 
plex with oxygen, or the substrates, or both, or whether one copper com- 
plexes with oxygen and the other with the substrates. Some of the ionic 
inhibitions might be due to the formation of complexes with the copper; 
such complexes might be more difficult to form at higher pH's because of 
competition with hydroxyl ions. The strong inhibition by oxalate might 
be due to chelation of the copper, but it is odd that pyrophosphate does 
not inhibit since it also chelates copper weU. It is also surprising that sul- 
fate and nitrate do not inhibit at all, unless the ionic size is as critical as 
Krueger believes. 

It would appear that inhibitory activity is related to the presence of 
a carboxyl group (excepting the inorganic ions). Benzoate ester and benza- 
mide are bound much less tightly than benzoate (around 3 kcal/mole 
difference). The weak action of benzenesuLfonate might be explained in 
three ways; (1) if the un-ionized form of the acid is necessary for activity, 
there would be less in the case of benzenesulfonic acid than with benzoic 
acid, (2) the sulfonate group might be too bulky, according to Krueger 
(although it is certainly not much larger than the carboxyl group), and 
(3) the sulfonate group does not have the ability to form bonds with the 
copper or hydrogen bonds with an enzyme group. Separation of the car- 
boxyl from the benzene ring reduces the binding, as in phenylacetate or 
cinnamate. The second requirement for potent inhibitory activity is a 
benzene ring. This may be seen by comparing acetate and phenylacetate, 
benzoate and nicotinate, and benzoate and cyclohexanecarboxylate; in 



302 2. ANALOGS OF ENZYME REACTION COMPONENTS 

the last example the increased binding of benzoate may be due to its great- 
er polarizability. In general the substitution of groups on the benzene 
ring reduces the affinity for the enzyme. This may be due to steric inter- 
ference with the approach of the ring or to inductive effects on the carboxyl 
group's interaction with the enzyme. It is rather odd that an oriho chlorine 
does not disturb binding much while an ortho methoxy group reduces the 
binding some 4 kcal/mole. 

The forces binding the substrates and inhibitors to these enzymes are 
thus vague at the present time. It is possible that hydrogen bonds between 
the OH or COOH groups and the enzyme are important, and it is equally 
possible that bonds to the copper ions are involved. One might conceive 
of the inhibitor's COOH group reacting with either the two copper ions, 
or with a copper ion and a vicinal — NH — group. Copper ions are able to 
catalyze the oxidation and hydroxylation of phenols nonenzymically, and 
it might be interesting to study the inhibition of such reactions by some 
of the compounds active in the enzymic reaction. 



TYROSINE METABOLISM 

Many interesting and practically important examples of analog inhibi- 
tion are to be found in the general field of amino acid metabolism, and we 
shall begin the discussion of this subject by considering the inhibitions of 
the various pathways of tyrosine metabolism. Tyrosine may be hydroxyl- 
ated to form dihydroxyphenylalanine (dopa), oxidatively deaminated or 
transaminated to form p-hydroxyphenylpyruvate, decarboxylated to form 
tyramine, or activated prior to incorporation into proteins; inhibition 
of all of these reactions by analogs has been reported. The scheme on page 
303 indicates the major pathways of tyrosine metabolism. Interference 
with these reactions might be expected to bring about physiological changes 
due to the acceleration or suppression of active amine synthesis, and also 
to affect melanin formation. 

Tyrosinase (Phenol Oxidase) 

These enzymes hydroxylate tyrosine in the oriho position to form dopa 
and further oxidize dopa to dopa-quinone. The enzymes discussed in 
the previous section generally possess this activity. However, mammalian 
tyrosinases are much more specific than the plant enzymes and oxidize 
tyrosine and dopa more rapidly than other phenols. Inhibitions of the en- 
zymes with tyrosine as the substrate will be considered in this section, 
but the results with p-cresol or catechol as substrate are probably generally 
applicable to tyrosinase activity. The high inhibitory potency of 4-chloro- 
resorcinol on potato tyrosinase, as determined by the rate of formation of 



TYROSINE METABOLISM 



303 




n ^ -2 — ■ 
2 rt tn D M G 
•5 ^ O -rl O 5 



oS 



2 >> e 



a 



304 2. ANALOGS OF ENZYME REACTION COMPONENTS 

melanin from tyrosine, was noted by KuU et al. (1954), for example (see 
following tabulation). Other substitutions in the 4-position usually reduce 





Lowest 


Inhibitor 


inhibitory concentration 




(mil/) 


4-Chlororesorcinol 


0.000069 


Resorcinol monobenzoate 


0.0047 


m-Aminophenol 


0.0092 


4-«.-Hexylresorcinol 


0.103 


Naphthoresorcinol 


0.125 


Orcinol 


0.141 


j)-Aminophenol 


0.183 


Resorcinol 


0.183 


Phloroglucino! 


6.17 


2-Nitroresorcinol 


65 



or abolish the inhibitory activity. The formation of melanin involves several 
steps and the inhibitions observed are not necessarily entirely on tyrosinase. 
Hydroquinone was found to be a weak inhibitor but the monobenzyl 
ether of hydroquinone is as potent as resorcinol. This latter substance, 




"°A\ /r°-'''^-A\ // 



Monobenzone (Benoquin) 



known also as monobenzone (Benoquin), is an inhibitor of melanin for- 
mation in the skin when applied topically, can produce leucoderma in 
Negroes, and is used in various conditions of melanosis. Some have thought 
that it releases hydroquinone after penetration into the skin but this is 
questionable in view of its own inhibitory activity. 

An active tyrosinase occurs in the Hardin-Passey mouse melanoma and 
is probably responsible for pigment formation. It is competitively inhib- 
ited by various tyrosine analogs (Lerner et al., 1951). The values of K^ 
shown in the tabulation were calculated on the basis of a K„, of 0.60 mM 



Inhibitor J^, (mM) 

iV-Acetyl-L-tyrosine 0.140 

iV-Formyl-L-tyrosine 0.177 

3-Amino-L-tyrosine 0.314 

3-Fluoro-L-tyrosine 1.25 



TYROSINE METABOLISM 305 

obtained from the reciprocal plots. 3-Nitro-L-tyrosine and O-methyl-L- 
tyrosine are not inhibitory. The effects of 3-substitution may be mediated 
through inductive effects on the 4-OH group and its interaction with the 
enzyme, whereas iV-substitution must lead to an altered position of binding 
to prevent oxidation. L-Phenylalanine and phenylpjTuvate inhibit com- 
petitively the tyrosinase from melanoma, inhibit the incorporation of ty- 
rosine-C^* into melanin, and depress the respiration of tumor tissue with 
tyrosine as the substrate (Boylen and Quastel, 1962). The high concentra- 
tions of these inhibitors in phenylketonuria might be responsible for the 
reduced pigment formation in these individuals. 

Tyrosine : a-Ketoglutarate Transaminase 

The effects of numerous analogs on the formation of p-hydroxyphenyl- 
pyruvate from tyrosine and a-ketoglutarate by a dog liver transaminase 
were reported by Canellakis and Cohen (1956 b); some of the results are 
given in Table 2-9. Certain of these analogs are transaminated (e.g., the 
3-substituted tyrosines) and the inhibitory activity varies inversely with 
their abilities to act as substrates. The rapid transamination of 3-fluoro- 
tyrosine may partly explain its toxic effects and inhibition of growth, 
since fluorofumarate or fluoroacetoacetate may be formed. Comparing the 
hydroxyl-substituted phenylalanines, it is seen that a m- or p-hydroxyl 
is necessary for tight binding, the contribution to the binding energy 
being over 2 kcal/mole. The if,,; for L-tyrosine is 0.71 mM, so that its rel- 
ative binding energy is at least — 4.47 kcal/mole, and that of m-hydroxy- 
DL-phenylalanine is at least — 4.75 kcal/mole (— 5.18 kcal/mole if only 
the L-isomer is active), which may be compared with the — 2.55 kcal/ 
mole for L-phenylalanine. Hydroxyl groups in the o-positions, on the other 
hand, do not augment binding very much. The tighter the binding between 
a basic hydroxyl and an acidic enzyme group, the greater the inhibitory 
activity; ring substituents modify the electronic character or basicity of 
the phenolic group. A carboxylate group is necessary for strong binding, 
as may be seen by comparing tjTosine with tyramine, and p-hydroxyben- 
zoate with p-cresol. The a-amino group may also be involved in the binding 
(— AF for p-hydroxyphenylacetate is 2.87 kcal/mole), but lacking data on 
p-hydroxyphenylproprionate it is not possible to evaluate this accurately. 
The strong inhibition by epinephrine is rather surprising in the light of 
the absence of a carboxylate group, but the /?-hydroxyl or iV-methyl group 
may contribute to make up for this deficiency. A similar study on the rat 
liver enzyme has been reported by Jacoby and La Du (1964). 

p-Hydroxyphenylpyruvate Oxidase 

The further oxidation of the product of tyrosine transamination is ca- 
talyzed by an enzyme from dog liver and is inhibited markedly by phenyl- 



306 



2. ANALOGS OF ENZYME REACTION COMPONENTS 



Table 2-9 
Inhibition of Tyrosine : a-KETOGLUTARATE Transaminase from Dog Liver " 









Relative 


Inhibitor 


(I)/(S) 


% Inhibition 


— zJi^ of binding 
(kcal/mole) 


3-Araino-DL-tyTosine 


1 


85 


6.00 


3,4-Dihydroxy-L-phenylalanine 


2 


70 


5.03 


Epinephrine 


2 


70 


5.03 


m-Hydroxy-DL-phenylalanine 


2 


60 


4.75 


2,5-Dihydroxy-L-phenylalanine 


2 


35 


4.13 


3-Fluoro-DL-tyrosine 


2 


20 


3.65 


3,5-Dibromo-L-tyrosine 


2 


10 


3.15 


o-Hydroxy-DL-phenylalanine 


2 


7 


2.91 


p-Hydroxybenzoate 


10 


26 


2.87 


p-Hydroxyphenylacetate 


10 


26 


2.87 


p-Aminobenzoate 


10 


21 


2.70 


L-Tryptophan 


10 


21 


2.70 


p-Nitro-DL-phenylalanine 


2 


5 


2.69 


L-Phenylalanine 


2 


4 


2.55 


o-Hydroxybenzoate (saHcylate) 


10 


17 


2.53 


p-Nitrobenzoate 


10 


15 


2.44 


3,5-Diiodo-L-tyrosine 


2 


3 


2.36 


D-Tyrosine 


2 


3 


2.36 


j)-Cresol 


10 


13 


2.34 


Tjrramine 


10 


9 


2.09 



" Relative binding energies were calculated on the basis of competitive inhibition 
with K^ = 0.71 mM for L-tyrosine. No correction for ionization was made. It should 
be noted that the different isomers of the DL-compounds may have different activities, 
so that the binding energy of the active form should be increased. (Data from Canel- 
lakis and Cohen, 1956 b.) 

pyruvate, essentially complete inhibition occurring with 0.4 mM in the 
presence of 2 mM substrate (Zannoni and La Du, 1959). The inhibition is 
negligible for the first 5 min but then increases to become complete at 
around 20 min. This might be due to protection by the substrate and pre- 
incubation experiments would have been informative. An equally good 
inhibitor is m-hydroxyphenylpyruvate but phenylacetate, 2,5-dihydrox- 
yphenylpyruvate, p-hydroxyphenyllactate, p-hydroxybenzoate, and ho- 
mogentisate are not inhibitory. 

Other Pathways of Tyrosine Metabolism 

Little is known about the effects of analogs on tyrosine decarboxyla- 
tion, although this might well be an important site to block if one wished 



TYROSINE METABOLISM 



307 



to reduce the tissue tyramine concentration. Mardashev and Semina 
(1961) found that the tyrosine decarboxylase from Streptococcus fecalis is 
inhibited 20% by 8.3 mM cysteine and 25% by 8.3 mM homocysteine 
when substrate concentration is 2.8 mM. This is a general phenomenon seen 
with several amino acid decarboxylases and is presumably due to the for- 
mation of complexes of the inhibiting amino acids with pyridoxal phosphate. 
The tyrosine-activating enzyme of pig pancreas catalyzes the first 

Amino acid + ATP + E -> E-amino acyl-AMP + PP 

step in the incorporation of amino acids into proteins. This also would 
be a very interesting step to investigate from the standpoint of analog 
inhibition, but our present information is meager. Schweet and Allen 
(1958) found that 3-fluoro-lL-tyrosine activated 50% the rate of L-ty- 
rosine and does not inhibit. L-tyrosinamide inhibits weakly but tyramine 
inhibits the phosphate exchange reaction 80% at 0.2 mM, this inhibition 
being reduced by increase in substrate concentration. The importance of 
the p-hydroxyl group is indicated by the fact that no inhibition is seen 
with phenylethylamine. 



Dihydroxyphenylalanine (Dopa) Decarboxylase 

This enzyme is on the pathway leading from tyrosine to the important 
catecholamines, epinephrine and norepinephrine, and lately has been the 
subject of much investigation because of the possible clinical applications 
of producing a selective block at this step. Inhibitors have indeed been 
found which are effective in vivo, reduce amine formation, and lower the 
blood pressure. The sequence of reactions for the formation of amines 
from dopa is the following: 



HO 



HO 




NH3 
-CH— COO' 



dopa 



decarboxylase 



HO 




CH^ CH,— NH, 



Dopa 



Dopamine 



dopamine /3-oxidase 




OH 

1 H 

CH— CH,— NH, 



Epinephrine 



308 



2. ANALOGS OF ENZYME REACTION COMPONENTS 



A block of dopa decarboxylase would thus decrease the rate of formation 
of these three physiologically active amines in the tissues. 

Of historical interest are the following observations on the analog inhi- 
bition of this enzyme: dopamine (Blaschko, 1942), epinephrine (Schapira, 
1946), various aromatic amines (tryptamine, tyramine, phenylethylamine, 
etc.) (Polonovski et al., 1946), and the various hydroxy, methoxy, and 
dimethoxy derivatives of phenylethylamine, the dimethoxy analogs being 
the most potent (Gonnard, 1950). 

These studies were extended in important ways by Sourkes (1954), 
who discovered the potent inhibiting activity of certain a-methylphenyl- 
alanines (Table 2-10). Especially inhibitory is a-methyldopa and this sub- 



Table 2-10 
Inhibition of Dopa Decarboxylase from Pig Kidney Cortex " 



Inhibitor 


Concentration 

(mM) 


% Inhibition 


Relative 

— /iF oi binding 

(kcal/mole) 


a-Methyldopa 


0.01 


22 


5.94 




0.1 


71 






0.5 


98 




a-Methyl-3-hydroxy-PA 


0.05 


45 


5.57 




0.5 


74 






5 


95 




a-Methyl-3-hydroxy-4- 


4.3 


92 


4.86 


methoxy-PA 








a-Methyl-3,4-dimethoxy-PA 


0.01 


16 


4.65 




0.1 


20 




iV-Acetyl-3,4-dimethoxy-PA 


3.6 


44 


3.32 


A'^-Methyldopa 


0.8 


11 


3.12 




1.7 


30 




Diiodotyrosine 


2.3 


15 


2.68 


3,4-Dimethoxy-PA 


5 


25 


2.59 


2,4-Dimethoxy-PA 


5 


22 


2.49 


3-Methoxy-4-hydroxy-PA 


5 


18 


2.33 


a-Methyl-3-methoxy-PA 


5 


18 


2.33 


a-Methyltyrosine 


5 


16 


2.25 


a-Methyl-PA 


2 


6 


2.13 


A'^-Methyl-3-methoxy-4- 


1.6 





— 


hydroxy-PA 









" Concentration of DL-dopa 4 mM, pH 6.8, preincubation with inhibitor 15 min. 
PA — phenylalanine, and dopa = 3,4-dihydroxyphenylalanine. Relative binding ener- 
gies calculated on the basis of competitive inhibition, which may not be strictly true; 
in any event, these values give a better indication of the relative inhibitory potency 
than the per cent inhibition at different concentrations. (Data from Sourkes, 1954.) 



TYKOSINE METABOLISM 309 

stance has been thoroughly studied biochemically and pharmacologically 
during recent years. It is interesting that these analogs have very little 
inhibitory activity toward tyrosine decarboxylase and that a-methyl- 
tyrosine does not inhibit dopa decarboxylase strongly, both facts pointing 
to the importance of the 3-hydroxyl group in the binding to the enzyme. 
This is also seen by comparing or-methylphenylalanine and its hydroxylated 
derivatives: The addition of a 4-hydroxyl has little effect, whereas a 
3-hydroxyl increases the binding energy over 3 kcal/mole. A 3-methoxy 
group seems to be ineffective. 

The inhibition by a-methyldopa was shown to be pseudoirreversible 
by varying the enzyme concentration and using the graphic procedure of 
Ackermann and Potter (1949). At concentrations of 0.01-0.03 mM, the 
inhibition being 15-25%, the behavior is fairly reversible, but at concen- 
trations of 0.1 mM or above there is marked nonlinearity of the curves. 
As pointed out by Sourkes, these data indicate merely that K^ is low and 
the affinity for the enzyme is high. The binding might be to the apoenzyme, 
to a great extent through the phenolic groups, or the inhibition could be 
the result of reaction with pyridoxal phosphate. The former mechanism 
was favored by Sourkes on the basis of the following evidence against a 
reaction with the coenzyme. (1) The inhibition is reversible by dialysis. 
(2) The rate of nonenzymic reaction of a-methyldopa with pyridoxal 
phosphate is too slow at inhibiting concentrations to be significant. (3) In- 
crease in pyridoxal phosphate concentration does not alter the inhibition 
significantly. (4) Analysis for pyridoxal phosphate at the end of inhibition 
experiments showed no loss. (5) Tyrosine decarboxylase is also a pyridoxal 
phosphate enzyme and is not inhibited. None of this evidence is completely 
conclusive and it is possible that a-methyldopa can form a reversible 
complex with pyridoxal phosphate on the enzyme surface, so that increase 
in coenzyme concentration would not be effective and analysis for total 
coenzyme would not detect the small amount combined. 5-Hydroxytryp- 
tophan decarboxylase is also potently inhibited by a-methyldopa (it is 
possible that the decarboxylases for dopa, 5-hydroxytryptophan, trypto- 
phan, tyrosine, and phenylalanine in mammalian tissues represent a single 
enzyme) and S. E. Smith (1960 a) has investigated the mechanism, using 
the mouse brain enzyme. Plots of 1/(S) against l/v showed pure competitive 
inhibition with respect to substrate at higher coenzyme concentrations 
(above 0.01 mM), but at low coenzjTne concentrations the inhibition be- 
comes noncompetitive with substrate. In contrast to dopa decarboxylase, 
increase of coenzyme concentration leads to a reduction in the inhibition 
(Fig. 2-3). Smith inclines to a coenzyme inactivation mechanism but admits 
that the inhibition is incompletely explained. If a-methyldopa forms a 
complex with pyridoxal phosphate on the enzyme surface, which it can do 
because it is decarboxylated slowly, it might be considered to be an in- 



310 



2. ANALOGS OF ENZYME REACTION COMPONENTS 



hibitor which enters into" the catalytic sequence of reactions but is not 
able to complete the process readily. This mechanism is also suggested by 
the results of Lovenberg et al. (1963) on kidney aromatic amino acid de- 
carboxylase, using tryptophan as the substrate. When a-methyldopa is 
preincubated with the enzyme in the absence of pyridoxal-P the inhibition 
is noncompetitive and potent, but when pyridoxal-P is added during the 
preincubation period the inhibition is competitive with respect to substrate 




0.001 

■ METHYL - OOPA 

Fig. 2-3. Inhibition of mouse brain 5-hydroxytryptophan 
decarboxylase by a-methyldopa. The sohd curves show 
the rate of formation of serotonin in //g/g/hr, and the 
dashed curves show the fractional inhibition. The curves 
X-X-X-X show the results without addition of pyridoxal-P, 
and the curves 0-0-0-0 show the results after addition 
of 0.008 mM pyridoxal-P. (From S. E. Smith, 1960 a.) 



and is less potent. If no preincubation is done and a-methyl dopa is added 
with the substrate, competitive inhibition is observed. The data suggest 
that a-methyldopa interacts specifically with the enzyme-pyridoxal-P com- 
plex (the enzyme as isolated contains tightly bound pyridoxal-P), and the 
protection or reversal of the inhibition by exogenous pyridoxal-P may be 
due to the reactivation of the enzyme-bound coenzyme. The reaction of 
the a-methyldopa with pyridoxal-P may involve the cyclization of a 
Schiff base (Mackay and Shepherd, 1962). 

Another comprehensive study of dopa decarboxylase was made by Hart- 
man et al. (1955), who determined the inhibitory activities of some 200 
compounds. The results, some of which are presented in Table 2-11, enable 



TYROSINE METABOLISM 



311 



Table 2-11 
Inhibition of Dopa Decarboxylase from Pig Kidney Cortex" 



Inhibitor 



Concen- 
tration 
(mM) 



% 
Inhibition 



Relative 

- AF of 

binding 

(kcal/mole) 



Cinnamates 



O 



CH=CH— COO' 



3-Mercapto- 


0.1 


78 


6.90 


3,4-Dihydroxy- (ethyl ester) 


0.3 


90 


6.78 


2, 6-Dihydroxy- 


0.2 


84 


6.70 


3, 4-Dihydroxy- 


0.4 


74 


5.90 


2-Hydroxy-3, 5-dibromo- 


0.4 


65 


5.63 


3-Hydroxy- (methyl ester) 


0.4 


60 


5.50 


2-Hydroxy- (methyl ketone) 


0.2 


31 


5.19 


3-Hydroxy- (methyl ketone) 


0.2 


30 


5.15 


3-Hydroxy- 


0.4 


37 


4.92 


a-Methyl-3-hydroxy- 


0.4 


31 


4.75 


3-Hydroxy-6-sulfonate- 


0.4 


31 


4.75 


a-Ethyl-3-hydroxy- 


0.4 


27 


4.64 


3 -Hydroxy- (amide) 


0.4 


19 


4.35 


2-Hydroxy- 


2 


52 


4.30 


2, 4-Dichloro- 


2 


41 


4.03 


2-Chloro- 


2 


36 


3.91 


4-Hydroxy- 


0.4 


7 


3.65 


3-Nitro- 


2 


15 


3.19 


3-Amino- 


2 


10 


2.91 


2, 4-Dihydroxy- 


2 


9 


2.83 


2, 3-Dimethoxy- 


2 


8 


2.75 


4-Chloro- 


2 


8 


2.75 


Unsubstituted 


2 





~ 


Hydrocinnamates <( \— CH2CH2— COO" 

V / 








3-Hydroxy-2, 4, 6-triiodo- 


0.2 


90 


7.04 


a-Methyl-3-hydroxy-2, 4, 6-triiodo- 


0.4 


91 


6.68 


3, 4-Dihydroxy- 


0.44 


63 


5.52 


a-Methyldopa 


0.2 


37 


5.35 


4-Hydroxy- 


2 


9 


2.83 


3-Hydroxy- 


2 





- 



312 2. ANALOGS OF ENZYME REACTION COMPONENTS 

Table 2-11 (continued) 



Inhibitor 



Concen- 
tration 
(mA'i) 



% 
Inhibition 



Relative 

- AF of 

binding 

(kcal/mole) 



d-Hydroxy- 
Unsubstituted 

Phenylacetates 

3, 4-Dihydroxy- 
2, 5-Dihydroxy- 
3, 4-Dimethoxy- 
3 -Hydroxy - 
Unsubstituted 



CH,— COO" 



2 


88 


5.49 


2 


28 


3.68 


2 


18 


3.32 


2 


11 


2.97 


2 





_ 



Phenylglycines 

3, 4-Dihydroxy- 
3-Hydroxy- 
4-Hydroxy- 
Unsubstituted 



=\ NH, 

CH 
\ 
COO 



2 


16 


2 





2 





2 






3.24 



Phenylpyruvates 

3, 4-Dihydroxy- 

3-Hydroxy- 

4-Hydroxy- 

2, 5-Dihydroxy- 

Unsubstituted 



CH5— CO— COO' 



0.2 

1 

2 

2 

2 



60 
72 
60 
60 
60 



5.92 
5.26 
4.50 
4.50 
4.50 



Benzoates (\ n — - COO" 

2-Hydroxy-3, 5-diiodo- 
2-Hydroxy-3, 5-dibromo- 
3, 5-Dibromo- 
3,4, 5-Trihydroxy- 
3-Hydroxy-4, 6-dibromo- 
2 -Hydroxy- 5-amino- 
2-Hydroxy-4-nitro- 



0.2 


47 


5.60 


0.2 


33 


5.24 


0.2 


10 


4.33 


2 


42 


4.07 


2 


36 


3.91 


2 


28 


3.68 


2 


26 


3.62 



TYROSINE METABOLISM 



313 



Table 2-11 (continued) 



Inhibitor 






Concen- 
tration 
(niiW) 


% 
Inhibition 


Relative 

- AF of 

binding 

(kcal mole) 


2, 5-Dihydroxy- 






2 


21 


3.44 


2, 4, 6-Trihydroxy- 






2 


16 


3.24 


3, 4-Dihydroxy- 






2 





- 


2. 4-Dihydroxy- 






2 





- 


Miscellaneous 












5- (3. 4-Dihydroxycinnamoyl)salicylate 


0.002 


87 


9.69 



HO 



COO" 




CH=CH-CO 



OH 



° Concentration of dopa 2 mM and pH 6.8. (Data from Hartman et al., 1955. ) 

one to speculate further about the nature of the binding to the active 
center. Several inhibitors more potent than a-methyldopa were found. 
The basic structure for inhibition was written as: 



HO 



(HO) 




O 

I I II 

c=c-c— X 



where X is OH, 0-alkyl, alkyl, or aryl. It is rather surprising that the 
negatively charged carboxylate group is not necessary, esters and amides 
being as potent as the acids, and it may be that the CO group is critical. 
The positively charged amino group is also not necessary, since 3,4-dihy- 
droxyhydrocinnamate is a good inhibitor, and this would make it likely 
that the binding of the inhibitors is not too much dependent on pyridoxal 
phosphate. The importance of the 3- and 4-hydroxyls is again evident and 
all the potent inhibitors have phenolic groups; apparently only the sulf- 
hydryl group can replace the hydroxyl. Halogens have the ability to aug- 
ment binding when they are the only substituents but particularly when 
a hydroxyl group is also present; 3,5-dibromobenzoate and 2,4-dichloro- 
cinnamate are bound reasonably well (at least 2 kcal/mole more than the 
unsubstituted compounds). The only unsubstituted inhibitor is phenyl- 
pyruvate, which must be significant, although of what is not clear. The 
interaction of the side chain must be complex and involve different types 
of forces. If one compares all the 3,4-dihydroxy derivatives, it is seen that 



314 2. ANALOGS OF ENZYME REACTION COMPONENTS 

inhibitory activity increases with the length or bulk of the side chain. 
Also it may be noted that the Hnear cinnamate derivatives are generally 
more potent than the hydrocinnamates. One must conclude that the most 
important binding groups are the hydroxyls, the ring, the side-chain car- 
bonyl, and any more terminal groups, which just about includes all of the 
molecule. The specificity of these inhibitors may well be quite high, since 
3-hydroxycinnamate and 3,4-dihydroxycinnamate (caffeate) were tested 
on tyrosine decarboxylase, glutamate decarboxylase, histidine decarbo- 
xylase, and succinate dehydrogenase, and found to be without effect. 
Tyrosinase is inhibited somewhat, especially by caffeate. 

There is evidence in patients with phenylpyruvic oligophrenia of a dis- 
turbance in tyrosine metabolism (hypopigmentation), and it is possible 
that the phenyl acids which are abnormally high might be inhibiting some 
step or steps in these pathways. Fellman (1956) therefore studied the ef- 
fects of such substances on the dopa decarboxylase from beef adrenal 
medulla. The order of inhibitory potency is phenylpyruvate > phenyllac- 
tate > phenylacetate > phenylalanine. Phenylpyruvate inhibits 77% when 
equimolar (3.3 milf ) with L-dopa. The low plasma epinephrine levels found 
in these patients thus might be due to such an inhibition, but another point 
of attack would have to be adduced for a suppression of melanin formation. 
It is interesting that this enzyme is inhibited quite strongly by norepi- 
nephrine and dopamine, whereas epinephrine exerts no effect (Fellman, 
1959). The susceptibilities of dopa decarboxylases from various tissues 
are obviously different, since the results of Fellman are often different 
from those of previous workers. 

cf-Methyldopa and related analogs can effectively inhibit decarboxylases 
in vivo, thereby interfering with amine formation and modifying tissue 
function. Direct evidence for an in vivo inhibition was obtained by intra- 
muscular injection of «-methyldopa into guinea pigs and demonstration 
of a marked depression of the decarboxylation of both dopa and 5-hydroxy- 
tryptophan in isolated kidney 15-30 min afterward (Westermann et at., 
1958). Indeed, the inhibition of 5-hydroxytryptophan decarboxylation is 
complete and after 90 min is 83%. More indirect evidence has been obtain- 
ed by showing that these analogs prevent the pharmacological actions of 
dopa and 5-hydroxytryptophan, these actions being dependent on decar- 
boxylation of these substances to dopamine and serotonin, respectively. 
Injection of dopa leads to a rise in the blood pressure which is probably 
primarily due to dopamine (although some norepinephrine and epinephrine 
may also be formed). This pressor response can be blocked by several de- 
carboxylase inhibitors, including 5-(3-hydroxycinnamoyl)salicylate (Po- 
grund and Clark, 1956) and a-methyldopa (Dengler and Keichel, 1958). 
There is no effect on the response to dopamine or norepinephrine. The 
increase in cardiac contractility induced by dopa is also completely blocked 



TYROSINE METABOLISM 315 

by a-methyldopa. 5-Hydroxytryptoplian causes bronchoconstriction in 
guinea pigs and central excitation in mice (if brain monoamine oxidase is 
blocked), these effects being due to serotonin, and pretreatment of the 
animals with a-methyldopa prevents these actions (Westermann et al., 
1958). 

Decarboxylase inhibition should lead to a decrease in the tissue concen- 
trations of certain amines and this has been demonstrated. The degree 
of reduction will depend on the relative rates of formation and metabolism 
of the amines, as well as on the magnitude of the decarboxylase inhibition 
(which will depend in part on the penetration of the analogs into the tis- 
sues), since we are dealing with steady-state multienzyme systems. Injec- 
tion of a-methyldopa (200 mg/kg) into dogs leads to a lowering of serotonin 
in the caudate nucleus (1.03 to 0.43 //g/g at 3 hr), a more prolonged lowering 
of norepinephrine (2.41 to 1.77 /ngjg at 24 hr) (Goldberg et al., 1960), and 
a fall of total catecholamines in the brain stem (0.17 ^- 0.08 //g/g), heart 
(0.59 -^ 0.26 //g/g), and spleen (0.94 -^ 0.43 //g/g) (Stone et al, 1962). In 
the mouse, brain serotonin is reduced but norepinephrine is unaffected 
(S. E. Smith, 1960 a). The urinary amines in four hypertensive patients 
were decreased by a-methyldopa (1-4 g per day): the reductions were 
81% for t>Tamine, 63% for serotonin, and 55% for tryptamine (Gates 
et al, 1960). 

Altering these amine levels should produce physiological disturbances. 
It has been found that a-methyldopa lowers the blood pressure and causes 
sedation in the dog (Goldberg et al, 1960), decreases coordinated activity 
and produces miosis in mice (S. E. Smith, 1960 a), and in a variety of animals 
induces a syndrome similar to that produced by reserpine (including hy- 
pothermia), a drug releasing amines from the tissues (S. E. Smith, 1960 b). 
a-Methyldopa is being studied clinically for the reduction of hypertension 
and a preliminary report (Gates et al., 1960) indicated its effectiveness, 
doses of 1-4 g/day for 1 week leading to a fall in supine blood pressure 
from 187.0/115.4 to 173.4/108.1 and in standing blood pressure from 177.7/ 
119.3 to 138.6/98.0, the controls being given a placebo in a double-blind 
study. 

It is thus clear that a-methyl-dopa can inhibit certain amino acid de- 
carboxylases in vivo, can alter amine levels in tissues, and can produce 
physiological disturbances that could reasonably be attributed to the in- 
hibition. Recently, however, more detailed studies of tissue amines have 
indicated that other mechanisms are possibly operative. Injections of 
a-methyl-3-hydroxyphenylalanine into guinea pigs lead to a reduction 
in brain amines (Fig. 2-4), the degree of lowering and the duration of the 
effect depending on the amine. a-Methyldopa acts similarly but is slightly 
more potent. Simultaneously there is an inhibition of amino acid decar- 
boxylase (Fig. 2-5). Cardiac norepinephrine is even more potently reduced 



316 



2. ANALOGS OF ENZYME REACTION COMPONENTS 



100 



75 



50 



25- 



% OF 
NORMAL 



SEROTONIN / 


/ 




/ / 


/dopamine 


^^ 


/ 




y"^ norepinephrine 


\y 




1 1 



20 

time (HOURS) — 



40 



60 



80 



100 



Fig. 2-4. Effects of a-methyl-w-tyrosine on brain amine concentrations in the 
guinea pig following an intraperitoneal dose of 400 mg/kg. (From Hess e< a?., 1961.) 



and remains at a lower level for a longer time than in brain. The results 
on serotonin and dopamine levels in the brain correspond as expected to 
the time course of decarboxylase inhibition, but the prolonged depletion 
of norepinephrine is difficult to explain on this basis. Since dopamine levels 
return to normal long before norepinephrine, there must be either an in- 
hibition of the /?-hydroxylation of dopamine to norepinephrine or an in- 
terference with the tissue binding of norepinephrine. Hess et al. (1961) 
showed that these analogs inhibit /5-hydroxyIation only at relatively high 
concentrations, which might have been produced soon after injection but 
certainly would not occur several hours later, and thus inclined to the sec- 
ond explanation. The lack of inhibition of dopamine /5-oxidase has been 
confirmed by Creveling et al. (1962) 

The time course for catecholamine depletion in mouse brain and heart 
following administration of these analogs is similar to that following re- 
serpine, except the return toward normal in the brain is somewhat faster. 
Porter et al. (1961) examined different analogs for ability to reduce nor- 
epinephrine in the brain and heart, and compared these results with their 
effectiveness in inhibiting decarboxylation of 5-hydroxytryptophan in 
kidney (Table 2-12). Some lack of correlation between the two activities is 



TYROSINE METABOLISM 



317 



Table 2-12 

Effective Doses of Analogs in Inhibiting Kidney Decarboxylase 
AND Lowering Brain and Heart Norepinephrine in Mice " 



Analog 



ED^o (mg/kg) 



Inhibition 

of renal 

decarboxylase 



Depletion 
of norepinephrine 



Brain 



Heart 



L-a-Methyl-3,4-dopa 2.63 32 21 

L-a-Methyl-2,3-dopa 1.35 >100 >100 

L-a-Methyl-3-hydroxy-PA 11.7 12 1 

L-a-Methyldopamine No inhibition >100 5 

L-a-Methyl-3-hydroxyphenylethylamine No inhibition 33 0.7 



" Analogs injected intraperitoneally. Kidney decarboxylase activity determined 
45 min after injection. Doses required to half deplete tissues of norepineplirine in 
last two columns. Since it has been shown that only the L-isomers are active, results 
are given on this basis for more convenient comparison. (Data from Porter et al., 1961.) 



100 




5 

TIME (HOURS) 

Fig. 2-5. Inhibition of amino acid decarboxylase in guinea pig tissues by a-methyl- 
TO-tyrosine injected intraperitoneally at 400 mg/kg. (From Hess et al., 1961.) 



318 2. ANALOGS OF ENZYME REACTION COMPONENTS 

evident and led to the conclusion that an action other than decarboxylase 
inhibition is involved, this probably being an interference with the binding 
of amines in the tissues. Since the a-methylamino acid analogs can be 
slowly decarboxylated to the corresponding a-methylamines in the body, 
and since these amines have the ability to deplete norepinephrine, it was 
suggested that at least part of the tissue amine lowering is due to displace- 
ment by the a-methyl analogs of the amines. This mechanism has been 
subscribed to by several recent workers. Maitra and Staehelin (1963) ad- 
ministered a-methyldopa to rats and guinea pigs and found the cardiac 
catecholamine levels to be insignificantly altered. They detected an increase 
in the a-methylnorepinephrine level, however, and a corresponding de- 
crease in norepinephrine, indicating the displacement of the normal catechol- 
amine with its analog. MuschoU and Maitra (1963) further demonstrated 
that a-methylnorepinephrine stored in the sympathetic nerve endings is 
released by nerve stimulation and is active on various adrenergic receptors. 
Pletscher et al. (1964) after injecting a-methyldopa into rats, found marked 
reduction of brain serotonin several hours later and felt that inhibition of 
the decarboxylase could not explain the results. They inclined to the view 
that a-methyldopa must also release or displace stored amines, and might 
also interfere with the uptake of amino acids by the brain. However, S. E. 
Smith (1963) had shown that a-methyldopa is only a very weak inhibitor 
of 5-hydroxy tryptophan uptake in brain slices (50% inhibition at 6.7 vaM), 
although it inhibits the decarboxylase potently (50% inhibition at 0.00056 
mM). It may be noted that other analogs may inhibit uptake more than 
decarboxylation. Day and Rand (1964) showed that a-methyldopa can 
restore the activity in animals whose catecholamine levels have been de- 
pleted by treatment with reserpine, presumably by the formation of a- 
methylnorepinephrine, which is generally only 1/9-1/2 as pharmacologically 
potent as norepinephrine; this does not provide direct evidence for the 
mechanism of inhibition by a-methyldopa, but clearly shows that it forms 
an active amine analog. 

The principles involved in the interpretation of these results are impor- 
tant in the general field of analog inhibition and the disturbances produced 
in tissue function, and, furthermore, the foregoing experiments might be 
carelessly construed as invalidating the decarboxylase inhibition mechanism; 
thus some critical comments may not be out of place. 

(1) It is unfortunate that Porter etal. (1961) did not determine decarbox- 
ylase inhibition in brain and heart for comparison with amine depletion 
in these tissues, since the inhibition in kidney may be quite different. In 
the first place, the penetration of the analogs into the three tissues may 
vary. Indeed, Hess et al. (1961) found that a-methyldopa concentrations 
in brain, heart, and kidney are in the ratio 1:1.66:3.28 at 1 hr and 1:1.8:9.8 
at 5 hr after injection. The concentrations of the other analogs in the tis- 



TYROSINE METABOLISM 319 

sues are not known, but it was demonstrated that a-methyldopa does not 
penetrate into brain. In the second place, the decarboxylases from the var- 
ious tissues may have different susceptibilities to the analogs, as we have 
noted above. Perhaps cardiac decarboxylase is resistant to a-methyldopa 
while the renal enzyme is more sensitive. 

(2) Monoamine oxidase inhibitors were administered during norepineph- 
rine depletion, and norepinephrine levels immediately rose (Hess et al., 
1961). It was concluded that "biosynthesis of norepinephrine can still 
occur in animals treated with or-methylamino acids." All these results 
mean is that inhibition of the decarboxylase was not complete. In any 
sequence 

AillB^C 

an inhibition of (1) will lower B and an inhibition of (2) will raise B (see 
Chapter 1-7). 

(3) It is stated that the decarboxylase step is the most rapid in the 
over-all sequence and therefore cannot be rate-limiting (Hess et al., 1961). 
The conclusion was that the decarboxylase must be inhibited very strongly 
for any effect to be observed. First, it is very difficult to establish that 

very very 

Tyrosine — > dopa — > dopamine — > norepinephrine 

sloiv fast slow 

these are the relative rates of the reactions in vivo, where the concentra- 
tions and states of the enzymes are quite different than when isolated from 
the cells. Second, the rate of formation of norepinephrine in a steady state 
is controlled by the first reaction (or a previous reaction) since these 
reactions are virtually irreversible. An inhibition of dopa decarboxylase 
will not alter the rate of norepinephrine formation as long as a steady 
state is maintained; the concentration of dopamine will also be unchanged 
in the steady state. However, it has been demonstrated that dopamine 
concentration falls, indicating that a departure from a steady state has 
occurred. One must also consider the other possible metabolic pathways 
for dopamine (e.g., oxidation and 0-methylation), since this is a divergent 
sequence. A certain depression of the decarboxylation need not be reflected 
in the same depression of norepinephrine formation, even under nonsteady- 
state conditions; the latter can be either greater or less than the inhibition 
of dopamine formation. In the case of serotonin formation, the decar- 
boxylation is the last step, and whether it will be inhibited or not will 
depend on the degree to which 5-hydroxytryptophan concentration can 
rise to overcome the block. In any event, it has been shown that the 
in vivo inhibition of decarboxylase by these analogs can be very high 
and sometimes complete. 



320 2. ANALOGS OF ENZYME REACTION COMPONENTS 

(4) The fact that certain a-methyl analogs of the catecholamines can 
deplete tissues of the amines, although they do not inhibit the decarboxy- 
lase, is not evidence against a decarboxylase inhibition mechanism for the 
a-methylamino acids, but indicates another mechanism, which may play 
a role in the prolonged lowering of norepinephrine levels without neces- 
sarily being involved in the initial rapid fall in tissue amines. The relatively 
rapid return of serotonin and dopamine levels to normal (Fig. 2-4) suggests 
that there is no generalized disturbance in tissue amine binding, but that 
the effect is specifically on norepinephrine. The most satisfactory position 
at the present time might be the following: the initial marked fall in tissue 
amines brought about by the a-methyl analogs is primarily due to an in- 
hibition of decarboxylation (perhaps supplemented at peak concentrations 
by inhibition of other steps, such as /?-hydroxylation), and further distur- 
bances in amine binding are progressively produced by the a-methylamines 
formed from the inhibitors, so that even when the decarboxylase is normally 
active again the tissues cannot concentrate certain of the normal catechol- 
amines. 

Dopamine p-Hydroxylase 

This enzyme catalyzes the synthesis of norepinephrine from dopamine 
and, as we have seen, its inhibition by analogs could be both theoretically 
and practically important. Hess et al. (1961) found that a-methyl-3-hy- 
droxyphenylalanine does not inhibit at 2 tclM. but inhibits 50% at 4 toM. 
The concentration of a-methyldopa 1 hr after injection is given as 376 
//g/g in the heart and this could mean a concentration around 750 //g/ml 
of intracellular fluid (assuming extracellular fluid has a low concentration 
at this time). This is approximately equivalent to 4 mM so that appreciable 
inhibition might occur. Inhibition data indicate that 3-methyl-3-hydroxy- 
phenylalanine concentrations in the tissues are roughly the same as a- 
methyldopa concentrations. Until more is known about the nature of the 
inhibition of this enzyme, it might be safe to conclude that it plays some 
role in the effects of the or-methyl analogs of phenylalanine. 

Various amines can inhibit this enzyme (Goldstein and Contrera, 1961). 
When dopamine concentration is 0.26 niM, the following inhibitions are 
observed: tyramine at 2.9 raM (75%), /?-phenylethylamine at 3.3 vaM 
(45%), amphetamine at 5.9 milf (35%), and 3-methoxydopamine at 4.8 
mM (15%). None of these inhibitors appears to be potent enough to be 
practically important in reducing norepinephrine synthesis and, further- 
more, these amines are so pharmacologically active that their use is limited. 
Benzyloxy amine, and particularly the p-hydroxyl derivative, inhibit this 
enzyme rather potently, 0.01 mM of the latter blocking almost completely 
after 90 min (van der Schoot et al., 1963), this being attributed to the iso- 
steric relation between phenethylamines and benzyloxyamines. 



i 



TRYPTOPHAN METABOLISM 321 

TRYPTOPHAN METABOLISM 

Tryptophan is involved in several important metabolic pathways, form- 
ing active substances as well as being incorporated into proteins, so that 
many attempts to block these pathways specifically with analogs have 
been made. Growth inhibition and physiological disturbances are readily 
produced by many of these analogs. (See scheme on page 322). 

Synthesis of Tryptophan 

L-Tryptophan is a potent feedback inhibitor of the conversion of 5- 
phosphoshikimate to anthranilate, an early reaction in tryptophan biosyn- 
thesis, and 5-methyltryptophan also inhibits, although not so strongly, 
a phenomenon (i.e., inhibition of a biosynthetic step by an analog) termed 
false feedback inhibition by Moyed (1960). It is likely that this mechanism 
explains the bacteriostatic activity of this analog. The condensation of 
anthranilate and 5-phosphoribosyl-l-pyrophosphate is not inhibited by 
5-methyltryptophan, but 6-fiuorotryptophan is inhibitory. A later reac- 



COO 



NH, 



Tryptophan Anthranilate 

tion in this sequence, the conversion of anthranilic deoxyribonucleotide to 
indoleglycerol-3-phosphate, is inhibited by a variety of anthranilate de- 
rivatives, especially the 3- and 4-methyl analogs (Gibson and Yanofsky, 
1960). The final reaction, the condensation of indole and serine to form 
tryptophan, catalyzed by tryptophan synthetase, is a major site of the 
attack by 4-methyltryptophan, which is a bacterial growth inhibitor 
(Trudinger and Cohen, 1956). The 5- and 6-methyl indoles are fairly potent 
competitive inhibitors, with K^ values near 0.1 roM (Hall et al., 1962). 
They are also antibacterial. The growth depression of E. coli is counteracted 
by tryptophan (Fig. 2-6). At least two sites for the inhibition have been 
demonstrated. Tryptophan synthetase is inhibited competitively, but there 
is also a block of the much earlier formation of anthranilate. There are no 
effects on the immediate metabolism of anthranilate or on tryptophanase, 
which indeed readily splits the analog to 4-methylindole. The bacteriostatic 
action is probably due mainly to suppression of tryptophan synthesis rather 
than to a disturbance of tryptophan utilization. Thus several steps in the 
biosynthesis are susceptible to analogs and it is quite possible that other 





322 



2. ANALOGS OF ENZYME EEACTION COMPONENTS 




™ rt 



O) T3 



+ a 



x: 
a 
o 



S S S 



^ eg CO 



o 

x: 
a, 



TRYPTOPHAN METABOLISM 



323 



unstudied reactions are likewise inhibited. A more indirect mechanism is 
the inhibition of the synthesis of tryptophan synthetase in growing cells, 
whereby tryptophan formation is further reduced. 



100 




Fio. 2-6. The inhibition of E. coli growth by 4-methyltryptophan 
at 0.01 mM and its antagonism by increasing tryptophan con- 
centrations. (From Trudinger and Cohen, 1956.) 

Tryptophanase 

This bacterial and fungal enzyme may be involved in the fermentation 
of tryptophan and is responsible for the putrefactive reaction in the in- 
testines. It is potently and competitively inhibited by the product indole 
and less potently by some analogs (as shown in the accompanying tabula- 



Inhibitor 



Relative — AF of binding 
(kcal/mole) 



Indole 

/?-3-Indolylpropionate 

3-Indolylacetate 

/?-3-Indolylethylamine 

DL-/3-l-Methyl-3-indolylalanine 



5.40 
4.05 
3.76 
2.55 
1.44 



324 2. ANALOGS OF ENZYME REACTION COMPONENTS 

tion) (Gooder and Happold, 1954). The importance of the indole N in 
binding is indicated by the weak inhibition with the last analog (tryptophan 
with the indole N methylated) and the failure of indene (the hydrocarbon 




/ 
.CHoCOO" /\ /CH,— CH 



NH^ 




IJ COO' 

N' "^ N 

I I 

H CH3 

3-Indolylacetate 0-1 -Methyl- 3 -indolylalanine 



.CHoCHoNH, 




N 
I 
H 

0-3-Indolylethylamine 

analog of indole) to inhibit, while the importance of the carboxylate group 
is reflected in the weak inhibition by the indolylethylamine. The potency 
of indole may be attributed to the fact that it may not have to be oriented 
in a manner necessary for reaction of the side chain. 

Tryptophan Pyrrolase (Tryptophan Peroxidase) 

This enzyme initiates one of the most important catabolic pathways of 
tryptophan and is readily inhibited by certain analogs. Hayaishi (1955 b) 
found the Pseudomonas enzyme to be sensitive to the hydroxytryptophans, 
and calculated the values of K; shown in the following tabulation. Since 



Substance 



Ki or Kg Relative — Zli^ of binding 

(milf) (kcal/mole) 



5- Hydroxy tryptophan . 002 8 . 06 

7-Hydroxytryptophan 0.12 5.55 

L-Tryptophan 0.4 4.81 



5-hydroxytryptophan is normally formed from tryptophan on the pathway 
to serotonin, its potent inhibition of the pyrrolase suggests that it may 
play a role in regulating tryptophan metabolism. The enzyme from rat 
liver is also inhibited by 5-hydroxytryptophan and even more potently 



TRYPTOPHAN METABOLISM 325 

by serotonin (Frieden et al., 1961). The K/s in the following tabulation 
indicate 3-indolylacrylate to be the most effective inhibitor. Other inhibi- 
tions observed when (S) = (I) = 3 mM are: indole 69%, tryptazan 50%, 



Inhibitor 


K, 


Relative — Ji" of binding 




(mM) 


(kcal/mole) 


3-Indolylacrylate 


0.012 


6.99 


Serotonin 


0.067 


5.93 


5-HydroxytrjT)tophan 


0.094 


5.71 


3-Indolylbutyrate 


0.16 


5.39 


Tryptamine 


0.20 


5.25 


3-Indolylpropionate 


0.29 


5.02 


3-Indolylacetate 


0.91 


4.32 


^-Methyltryptophan 


1.1 


4.20 


D-Tryptophan 


1.6 


3.97 


6-Fluorotr>i)tophan 


2.0 


3.83 


5-Fluorotryptophan 


2.2 


3.77 



5-methyltryptophan 38%, and 6-methyltryptophan 33%. The analogs with 
altered side chains are competitive while the others are mainly noncompeti- 
tive. The roughly equivalent binding of tryptamine and 3-indolylpropionate, 
and of serotonin and 5-hydroxytryptophan, might indicate that the binding 
is primarily with the indole ring, the side chains contributing little, and this 
is substantiated by the fact that indole is bound approximately as well 
as 3-indolylpropionate. The stronger binding of 3-indolylacrylate compared 
to 3-indolylpropionate (about 2 kcal/mole) is thus difficult to account for 
unless there is a modification of the interaction of the indole N. Tryptophan 
pyrrolase is an inducible enzyme in the rat but none of these analogs is 
active, although Sourkes and Townsend (1955) found a-methyltryptophan 
to induce after subcutaneous injection. 

Tryptophan Hydroxylase (Phenylalanine Hydroxylase) 

Excessive feeding of phenylalanine leads to low blood serotonin, a low 
excretion of 5-hydroxyindoleacetate, and a decrease in brain serotonin, 
and this has usually been attributed to an inhibition of 5-hydroxytrypto- 
phan decarboxylase. However, no accumulation of 5-hydroxytryptophan 
has been demonstrated and it is possible that the site of the inhibition might 
be earlier, perhaps on the hydroxy lation of tryptophan (Freedland et al., 
1961). Hydroxylating preparations from rat liver are indeed quite potently 
inhibited by L-phenylalanine, and also by phenylpyruvate and phenyUac- 
tate (K,„ for L-tryptophan is 29 mM, and K^ for L-phenylalanine is 0.22 mM). 



326 2. ANALOGS OF ENZYME REACTION COMPONENTS 

It is likely that the same enzyme is responsible for the hydroxylation of 
both phenylalanine and tryptophan, since the K^ for phenylalanine is close 
to the iii,,, when it is the substrate; the affinity for tryptophan is, however, 
much less. These results may help to explain some of the changes observed 
in phenylpyruvic oligophrenia (see pages 329 and 429). 

Tryptophan-Activating Enzyme 

An activating enzyme from pancreas is specific for tryptophan with 
respect to other normal amino acids but can activate certain analogs of 
tryptophan (Sharon and Lipmann, 1957). The analogs tested fall into three 
categories: 

Group I are activated (tryptazan, azatryptophan, 5-fluorotryptophan, 
and 6-fluorotryptophan). 

Group 11 are inhibitory (tryptamine, D-tryptophan, /5-methyltryptophan, 
5-hydroxytryptophan, 5-methyltryptophan, and 6-methyltryptophan). 

Group III are inactive (indole, indoleacetate, 6-methyltryptazan, and 
iV-acety Itry ptophan ) . 




CH,— CH ^-^ ^ CH^— CH 

^coo' f H H coo' 

N N N 

I I 

H H 



r 



Tryptazan Azatryptophan 

Reciprocal plots were said to indicate competitive inhibition but actually 
do not show pure competition, and it might better be designated as partially 
competitive inhibition. For analogs to be activated they must match the 
size of tryptophan, and the introduction of bulkier groups prevents reaction 
with ATP but allows binding and inhibition. Azatryptophan and tryptazan 
are both incorporated into proteins. In E. coli, azatryptophan permits syn- 
thesis of proteins and nucleic acids but many of the enzymes are not in 
the normal forms (Pardee and Prestidge, 1958). The synthesis of adaptive 
enzymes, such as /?-galactosidase, is inhibited very rapidly, and phage for- 
mation is blocked more readily than bacterial growth. The induced synthe- 
sis of maltase in yeast is also very potently suppressed by tryptazan, in- 
corporation of all amino acids being simultaneously blocked (Halvorson 
etal., 1955). 5-Methyltryptophan and 6-methyltryptazan inhibit moderate- 
ly, while 6-methyltryptophan is inactive. 



GLUTAMATE METABOLISM 



327 



GLUTAMATE METABOLISM 

Glutamate occupies a central position in many important metabolic 
pathways and serves to link amino acid metabolism with the tricarboxylic 
acid cycle. Its relationship to the biochemically active glutamine and the 
physiologically active /-aminobutyrate (GABA) makes possible specific 
inhibitions of glutamate reactions of great interest. The reactions catalyzed 
by enzymes studied with respect to analog inhibition are shown in the 
following scheme. 



glutamylhydroxamate 



D -glutamate 



(4) 
o-ketoglutarate 



(7) 
(3) (8) 



glutamine 



(6) 



(1,2) 



L-glutamate — 

(5) 

r 
>' -aminobuty rate 



(10) 



proline 



A'-pyrroline-5-carboxylate 

a 

(10) 
glutamic -7 -semialdehyde 



(1) L-glutamate dehydrogenase 

(2) glutamate transaminases 

(3) glutamate racemase 

(4) D-glutamate oxidase 

(5) glutamate decarboxylase 



(6) glutamine synthetase 

(7) glutaminase 

(8) formylglycinamidine phosphoriboside synthetase 

(9) 7 -glutamyl transferase 

(10) A'-pyrroline-5-carboxylate dehydrogenase 



Glutamate Decarboxylase 

Several analogs of glutamate inhibit its utilization by Lactobacillus ara- 
binosus, and thus a study of decarboxylase from E. coli was undertaken 
by Roberts (1953). The two most potent inhibitors are a-oximinoglutarate 
and a-methylglutamate, but the former is probably active by virtue of 
its hydrolysis to hydroxylamine which inactivates pyridoxal phosphate. 
The latter analog inhibits competitively when added with the substrate, 
but if it is preincubated with the enzyme the inhibition becomes progres- 
sively more noncompetitive and difficulty reversible. The rates of combina- 
tion with the enzyme and dissociation from the enzyme are very slow. 
The methyl group interferes with the normal binding of the molecule so 
that decarboxylation does not occur, but by some unknown mechanism 
brings about a type of binding that is very strong, a behavior seen with 
some other a-methylamino acids. The decarboxylation of glutamate in 
rat brain homogenates is also inhibited competitively by aspartate {K„^ 



328 2. ANALOGS OF ENZYME REACTION COMPONENTS 

= 21 mM, and K^ = 23 milf ) (Wingo and Awapara, 1950), which is rather 
surprising because of the shorter intercarboxylate distance. 

Ghitamate decarboxylase from the squash Curcurbita moscliata is inhib- 
ited competitively by a variety of organic acids (see accompanying tabu- 
lation), and the results are of some interest with regard to structure and 

Inhibitor (13.6 mM) % Inhibition 



Monocarboxylates 

Formate 17 

Acetate 42 

Propionate 10 

7i-Butyrate 18 

Isobutyrate 8 

n-Valerate 25 

Iso valerate 15 

n-Caproate 24 

Isocaproate 20 

Dicarboxylates (saturated) 

Oxalate 

Malonate 

Succinate 

Glutarate 24 

Adipate 49 

Pimelate 58 

Suberate 37 

Dicarboxylates (unsaturated) 

Fumarate 

Maleate 14 

Citraconate 

Mesaconate 7 

Itaconate 11 

Tricarboxylates 

cis-Aconitate 29 

irans-Aconitate 18 



intercarboxylate distance (Ohno and Okunuki, 1962). Glutamate concen- 
tration was 27.2 mM in all cases. Many amino acids examined are weakly 
inhibitory or without effect. The maximal inhibition by pimelate in its 
series probably indicates that a cambering of the molecule is necessary for 
binding of the two carboxylate groups, or possibly that the cationic groups 
of the enzyme are farther apart than in glutamate. The relatively high 





Relative 




— Zli^of bindin 


g (kcal/mole)" 


Hanson (1958) 


Tashian (1961) 


3.15 






4.16 


— 






4.26 


2.62 






2.62 


2.53 






3.73 


1.75 






3.83 


0.60 






— 


0.53 






— 



GLUTAMATE METABOLISM 329 

inhibition by acetate is surprising; if the monocarboxylates interact with 
the cationic groups, one might expect propionate or butyrate to be more 
inhibitory. 

The problem of the abnormal brain development in phenylpyruvic oligo- 
phrenia prompted an investigation of the effects of the phenyl acids on 
brain glutamate decarboxylase by Hanson (1958); the results are presented 
in the tabulation below in which they are compared with those of Tashian 



Inhibitor 



p-Hydroxyphenylacetate 

o-Hydroxyphenylacetate 

Phenylpyruvate 

Phenylacetate 

p-Hydroxyphenylpyruvate 

Phenylalanine 

Phenyllactate 

° Relative — AF's of binding adjusted so that value for phenylpyruvate is the same 
in each series. 

(1961), who also used rat brain. There are some rather striking differences, 
part of which may be due to the procedures used since these analogs, 
although stated to be competitive, present deviations from classic kinetic 
formulations (and for this reason the binding energies calculated from ap- 
parent K/s are probably not very reliable). If these analogs, which are 
present in high concentrations in the blood enter the brain readily, it is 
possible that they depress the formation of y-aminobutyrate (GABA) 
which may be essential for normal neurological development. In branched- 
chain ketonuria (maple sugar urine disease) various keto and hydroxy 
fatty acids accumulate in the body, which Tashian (1961) showed also 
inhibit glutamate decarboxylase (relative — JF's of binding for a-hydroxy- 
isovalerate, a-ketoisovalerate, and the corresponding isocaproates from 
3.58 to 4.50 kcal/mole on the scale above). The enzyme from E. coli, on 
the other hand, is relatively insensitive to any of these analogs. 

L-Glutamate Dehydrogenase 

The oxidative deamination of L-glutamate in beef liver homogenates is 
catalyzed by a NAD-linked dehydrogenase, the inhibition of which at 
pH 8.4 was thoroughly studied by Caughey et al. (1957). The accompanying 
tabulation shows the K^ values for competitive analogs, from which the 



330 2. ANALOGS OF ENZYME REACTION COMPONENTS 



Inhibitor 


(mM) 


Relative — Zli*' of binding 
(kcal/mole) 


5-Bromofuroate 


0.059 


6.00 


5-Chlorofuroate 


0.063 


5.96 


5-Nitrofuroate 


0.17 


5.35 


m-Iodobenzoate 


0.46 


4.74 


»n-Bromobenzoate 


0.54 


4.65 


Isophthalate 


0.66 


4.62 


Glutarate 


0.58 


4.60 


a-Ketoglutarate 


0.73 


4.45 


w -Chlorobenzoate 


1.02 


4.25 


D-Glutamate 


2.0 


3.83 


m-Nitrobenzoate 


3.4 


3.51 


Trimesate 


4.0 


3.40 


Fumarate 


6.8 


3.08 


Succinate 


11 


2.78 


Adipate 


16 


2.55 



relative binding energies have been calculated. Inhibitors were classed as 
competitive if the interaction constant a is greater than 10; trimesate and 
D-glutamate are only partially competitive, with a = 1.7. It was postulated 

CCXD' 
^O. ^COO" 




COO 

Furcate Isophthalate 



COO 





COO' 



OOC \^ COO 



Trimesate Indole-3-carboxylate 

that the enzyme contains two cationic groups at a separation optimal for 
interaction with the carboxylate groups of glutamate, glutarate, and iso- 
phthalate. The effects of some other inhibitors for which the ^/s were 
not calculated are shown in the following tabulation; most of these are 



GLUT AM ATE METABOLISM 331 

relatively weak (fumarate is included for comparison with values in the 
preceding table). L-Glutamate was 2 mM in each case. The following are 
not inhibitory at 10 mJ-l: L-glutamine, L-diethylglutamate, /?-methylglu- 
tarate, citrate, o- and p-hydroxybenzoate. 



Inhibitor 


Concentration 
(mM) 


% Inhibition 


Benzoate 


10 


12 


Furoate 


10 


26 


Phthalate 


10 


17 


Terephthalate 


6.7 


16 


Isophthalate 


2 


50 


<ran.5-Aconitate 


8 


20 


Indole-2-carboxylate 


7.5 


21 


Indole-3-carboxylate 


8.3 


38 


m-Hydroxybenzoate 


10 


27 


Pyridine-2,6-dicarboxylate 


10 


27 


Fumarate 


10 


29 



In the various substituted benzenes the 7neta compounds are invariably 
the most potent inhibitors, presumably because the intergroup distances 
are close to the enzyme intercationic separation. The dipoles of the halogen 
and nitro compounds may interact with the cationic group since they may 
be represented as: 

However, there is no correlation of inhibitory activity with dipole moment. 
One might think that the dipole-cation interaction would be weaker than 
the carboxylate-cation interaction, but the hydration of the inhibitors must 
also be considered. Less water needs to be displaced when the dipoles ap- 
proach the enzyme cationic group. It was pointed out that all good in- 
hibitors are reasonably planar and the presence of bulky groups protruding 
lowers the affinity. The particularly good binding of the furoates may be 
related to some interaction of the ring with the enzyme. The reverse 
reaction from or-ketoglutarate to glutamate is inhibited less than the for- 
ward reaction by glutarate, isophthalate, and 5-bromofuroate, and the 
inhibitions are noncompetitive. 

The L-glutamate dehydrogenase from cockroach muscle mitochondria 



332 2. ANALOGS OF p:;nzyme reaction components 

is similarly inhibited (see' accompanying tabulation), and here glutarate 
also appears to present a relatively good fit to the active site (Mills and 
Cochran, 1963). Glutamate was 15 mM in all cases. The reverse reaction 
catalyzed by the glutamate dehydrogenases (both NAD- and NADP- 



Inhibitor (3 mM) 


% Inhibition 


Succinate 


15 


Fumarate 


15 


Malate 


15 


Glutarate 


65 


Adipate 


30 


D-Glutamate 


55 


L- Aspartate 


20 


D-Aspartate 


10 



linked) from Fusarium oxysporum is also inhibited by glutarate, the K,^ 
for a-ketoglutarate being 2.1 mM, and the K^ for glutarate 1.52 mM 
(Sanwal, 1961). 

Glutaminase 

The deamidation of glutamine is inhibited by the product glutamate 
and this is not a reversal of the equilibrium but a competition for the 
active center, as first pointed out by Krebs (1935). L-Glutamate and d- 
glutamate inhibit guinea pig kidney glutaminase equally (98% at 25 mM 
when glutamine is 8.7 mM). Inhibition by glutamate has been confirmed 
for the enzyme from guinea pig kidney (van Baerle et al., 1957), pig kidney 
(Klingman and Handler, 1958), dog kidney (Sayre and Roberts, 1958), 
and rat brain (Blumson, 1957). The inhibition has generally been found 
to be noncompetitive with respect to glutamine, but oddly is competitive 
with phosphate on the phosphate-activated glutaminase from dog kidney. 
The ammonium ion is, however, strictly competitive with glutamine on 
both the pig and dog kidney enzymes. Another type of glutaminase (called 
glutaminase II), which is transaminating in the presence of pyruvate and 
is obtained from guinea pig kidney, is not inhibited by even 100 mM 
glutamate (Goldstein et al., 1957). Sayre and Roberts (1958) pictured the 
active center as containing two cationic groups, one binding the phosphate 
and one the glutamine carboxylate group; the negatively charged phosphate 
also interacts with the positive or-amino group of glutamine. Since the active 
enzyme is the phosphate complex, it is easy to see why phosphate would 
antagonize inhibitions produced by certain substances (e.g., dyes such as 
bromosulfalein or bromcresol green which complex with both enzyme 
cationic sites), but it is difficult to understand why glutamate inhibits 



GLUTAMATE METABOLISM 333 

competitively with respect to phosphate. It was stated that for a substance 
to compete with glutamine it should have affinity for the enzyme-phosphate 
complex, and it would seem that glutamate may fall into this category. 

Few other analogs have been tested on this enzyme. Krebs (1935) ob- 
served a mild inhibition by DL-/?-hydroxyglutamate (22% at 80 mM when 
glutamine 40 mM), and Girerd et al. (1958) reported inhibition by ethyl 
D-glutamate and DL-/5-methylglutamate. Because of the postulated role 
of glutaminase in renal function, these two latter analogs, along with 
L-glutamate and bromosulfalein, were tested in vivo. These inhibitors re- 
duce diuresis in rats around 50% at 50 mg/kg subcutaneously, whereas a 
group of five less potent glutaminase inhibitors actually increase diuresis. 

Formylglycinamidine Phosphoriboside Synthetase 

Glutamine participates in purine biosynthesis by contributing its amide 
N. Azaserine and 6-diazo-5-oxo-L-norleucine (DON) are potent inhibitors 
of inosinate biosynthesis and lead to the accumulation of formylglycina- 
mide phosphoribotide (FGAR). These substances may be considered as 
analogs of glutamine and have been shown to inhibit formylglycinamide 
ribonucleotide amidotransferase competitively with respect to glutamine 



N:=N=CH- 


-CO— O-CH2- 
Azaserine 


-CH 

COO' 


N=N=CH- 


NH3 
/ 
-CO— CH2CH2— CH 

^COO 
DON 



(Levenberg et al., 1957). The K„, for glutamine is 0.615 mM and the K^s 
for azaserine and DON are 0.034 mM and 0.0011 mM, respectively. Once 
azaserine binds to the enzyme, however, an irreversible reaction occurs, 
due perhaps to an alkylation of the enzyme. French et al. (1963 a) pointed 
out that 50% inhibition can be obtained with a (S)/(I) ratio of 2100 with 
DON. Phosphoribosyl-PP amidotransferase, another enzyme catalyzing 
the transfer of the amide nitrogen of glutamine, is also inhibited compe- 
titively by DON, with a K^ of 0.019 mM, and much more weakly by aza- 
serine (Hartman, 1963 b). A slow covalent binding of DON to the enzyme 
occurs following the initial reversible attachment, and this is accelerated 
by the presence of phosphoribosyl-PP and Mg++ on the enzyme, indicating 
that the active site for the reaction of glutamine, or the binding of DON, 
is partly dependent on the other substrate and the cofactor. Blocking of 
an SH group prevents the attachment of DON to the enzyme, suggesting 
that this SH group is catalytically functional in the nitrogen transfer and 
the irreversible binding of DON, as French et al. (1963 b) concluded from 
their work with azaserine on the formylglycinamide ribonucleotide amido- 
transferase. 



334 2. ANALOGS OF ENZYME REACTION COMPONENTS 

Glutamate Transaminases 

This enzyme apparently possesses two cationic groups properly separated 
to interact with glutamate and the other dicarboxylates, because it is inhib- 
ited best by glutarate of all the saturated dicarboxylates, as shown in the 
accompanying tabulation (aspartate =1.7 mM and a-ketoglutarate = 6.7 
mM) (Jenkins et al., 1959). This is one instance in which the «-methyl 
analog has no affinity for the enzyme. Very similar results were obtained 



Inhibitor (40 mM) 

Malonate 

Succinate 

Glutarate 

Adipate 

Piraelate 

Suberate 

Maleate 

a-Methylaspartate 

a-Methylglutaniate 



by Velick and Vavra (1962), glutarate inhibiting the most potently of the 
saturated dicarboxylates; from their values for K, one can calculate that 
glutarate is bound 0.85 kcal/mole more tightly than succinate, and 0.32 
kcal/mole more tightly than adipate. Of the three phthalates, o-phthalate 
is the most inhibitory {K^ = 5 mM), m-phthalate intermediary (K^ = 8 
mM), and j^-phthalate the least active {K, = 10 mM). The results of the 
studies above were obtained on pig heart transaminase, and from the lim- 
ited data reported by Goldstone and Adams (1962) it would appear that 
the enzyme from rat liver is different in that glutarate is about 10 times 
more potent than maleate as an inhibitor. The alanine:a-ketoglutarate 
transaminase from rat liver is inhibited moderately and competitively by 
certain amino acids, such as leucine and valine, and also by maleate, but 
no data were given for comparison with the aspartate: a-ketoglutarate trans- 
aminase (Segal et al., 1962). Fluorooxalacetate inhibits the latter enzyme 
from heart competitively with respect to oxalacetate (and to a-ketoglutarate 
and aspartate when the reverse reaction is run), K^,, being 3.5mMforof- 
ketoglutarate and 0.5 mM for aspartate, and /f, being 0.1 mM (Kun et 
al., 1960). It is also slowly transaminated to fluoroaspartate which likewise 
inhibits the enzyme. 



Inhibition 


Relative 


- AF (kcal/mole) 







<0.17 


31 




1.49 


72 




2.56 


62 




2.29 







<0.17 







<0.17 


78 




2.76 


35 




1.60 







<0.17 



ARGINASE 335 

Other Enzymes in Glutamate Metabolism 

Analog inhibition has been reported for several other enzymes involved 
in the metabolism of glutamate but quantitative studies from which struc- 
ture-action relationships may be derived have not been made. Some of 
these results are summarized in Table 2-13. An interesting inhibitor is the 
convulsant isolated from agenized flour, methionine sulfoximine, which is 
inhibitory to the incorporation of methionine into proteins in bacteria 
and brain. Since these actions are to a great extent antagonized by methio- 
nine, this substance has generally been considered as a methionine antago- 
nist, but glutamine also is antagonistic. Sellinger and Weiler (1963) have 
shown that methionine sulfoximine inhibits brain glutamine synthetase 
competitively with respect to glutamate, some inhibition being seen at 
0.01 mM and around 50% inhibition at 1 mM with low glutamate concen- 
trations {K^ = 0.05-0.064 mM). The relation between this inhibition and 
the convulsant action is not clear but it was postulated that methionine 
sulfoximine interferes in some vague manner with glutamine synthesis 
in an intracellular compartment in the brain. 

ARGINASE 

The hydrolysis of arginine: 

\ / (+H2O) + / \ 

C— NH— CHoCH^CH,— CH ^ ^-^- H3N— CHoCHoCHz— CH + CO 

+ <^ \ - \ - / 

HgN COO COO H2N 

Arginine Ornithine Urea 

is catalyzed by the Mn++-activated enzyme arginase and is a step in the 
urea cycle. Arginase is inhibited by the product ornithine, as first shown 
by Gross (1921) and confirmed by Bach and Williamson (1942), who found 
that the inhibition is much more marked in liver extracts than in slices 
(50% inhibition given by 5.3 mM ornithine in extracts and by 15.9 mM 
in slices when argmine is 3.56 mM). Edlbacher and Zeller (1936) noted that 
several amino acids inhibit, but ornithine is the most potent with lysine 
running a close second. 

These inhibitions were compared and subjected to quantitative treat- 
ment by Hunter and Downs (1945) in the publication wherein the first 
use of the single-curve plot (type F) was made (see Chapter 1-5). Orni- 
thine and lysine inhibit competitively {K , = 4.1 mM and 4.8 mM, respec- 
tively) but other amino acids are usually only partially competitive (since 
the DL-forms of the inhibitors used and only the L-isomers are active, it is 
likely that these constants should be halved). Calculations of relative 



336 



2. ANALOGS OF ENZYME REACTION COMPONENTS 



tf 



Oh 



c^ <N 



O 






o ~ 



a> Oi ir- 



-tj I— ( 



Xi 


J2 


^ 




o 


O 


e 




05 


05 




, . 


■o 


-73 


"^S 


O 


c 


C 


N 


C5 


c3 


cS 


-t^ 




tH 


tM 


'> 


^^ 


eS 


cS 


o 


^ 


M 


M 


_c 


v 


C 


fi 




c 


<D 


01 


.A 


tH 


>, 


>, 


c3 


cS 


< 


< 


05 


> 



-ki 



rO 



U; o c 



c II 






"H '^ -fc^ -Li 






^ rz rz .— ."S 

" ■" :2 :2 :2 :2 

e "^ IS IS IS 2 

S — e c c c 



p. 5 c. 



o ;5 



o 



& c 






4J O 



o 2 



^ c ^ ^ '^ ^ 

S S S 2 

rit CO CO CO cc CO 

^ O CO CO CO CO 



- o° 2 

■S o .S s? 

•S t^ ^ Sj 

(N ,2 '=^. 

|-H iM ,_, lO 



S *« 

O (M 



.S -a 

ft -^ 

S ic 
o ^ 



s 




s 

c8 


g 

13 


S 


o 


J 


2 

o 

c 


o 


1 

2 

T3 


>> 

3 
O 




C 




-2 

g 
3 


5 

o3 

S 

3 

5 


3 

5 


j 


"ao 


>, 

w 






3 

5 


-< 


> 


>> 


S 


1 

Ph 


< 




_3 


u 





«iX 


=Ci 


Q 


'O 


'^ 


j 


?^ 


u 


<ai. 





Q 



13 O 



o 





eS 


-1-3 




"2 


o 






JS 




'x 


-^ 


(a 


o 


3 


S 


-2 


>> 

ou 


N 


c8 


aj 


3 


s 


_s 




■k^ 


c 




3 


c« 






-»j 




O 


3 




c 


5 



eS 



s ^1 



O 



ARGINASE 337 

interaction energies from the constants given are shown in the following 
tabulation. Two types of inhibition may be possible: (1) competitive inhi- 
bition by diamino compounds that are probably oriented as the substrate, 
and (2) partially competitive or noncompetitive inhibition by monoamines 



^ . ., . Relative — Ji^ of binding 

Inhibitor ,, ,, , , 

(kcal/mole) 



L-Ornithine 3.82 

L-Lysine 3 . 72 

L-Norvaline 2 . 53 

L-Isoleucine 2 . 48 

L- Valine 1.98 

L-Cysteine 1 . 94 

L-Leucine 1.81 

L-a-AminobutjTate 1.40 

L-Phenylalanine 1 . 09 

L-Xorleucine . 87 

L-Proline 0.78 

L- Aspartate . 67 

L- Alanine 0.61 

L-Citrulline 0.47 

L-Serine 0.25 

L-Tryptophan 0.21 

L-Histidine -0.82 

Glycine —0.99 



that may be oriented otherwise. Inhibitory activity increases in the straight- 
chain series from glycine to norvaline; each additional methylene group 
contributes 1.17 kcal/mole to the binding energy, a value similar to that 
found in other series, and undoubtedly due to dispersion forces. It is odd 
that there is a sudden and marked drop in the affinity on adding another 
methylene group to form norleucine. As pointed out by Hunter and Downs, 
it is difficult to establish structural correlation in the series of substituted 
alanines (shown in the accompanying table); why, for example, is cysteine 
bound so much more tightly than serine? Substitution in the a-amino group 
(carbamyl or formyl) always reduces the activity; hence this probably con- 
stitutes one binding group. The experiments were done at pH 8.4 and 
therefore all carboxyl groups were essentially completely ionized, but there 
would be some variation between the inhibitors with respect to the fraction 
of the amino groups protonated {pK^'s for these inhibitors run from 8.2 
to 10.6). 



338 2. ANALOGS OF ENZYME REACTION COMPONENTS 



_,,,., , Relative — AF oi bmdim 

bubstituent group 

(kcal/mole) 



—Imidazole —0.82 

—Indole 0.21 

-OH 0.25 

-H 0.61 

-COO- 0.67 

-Phenyl 1.09 

-CH3 1.40 

-SH 1.94 



The relative sensitivities to these amino acids probably vary with the 
source of the arginase. For example, the mouse liver enzyme is inhibited 
somewhat more strongly by L-lysine than L-ornithine, and even D-lysine 
is a weak inhibitor (inhibitions at 1 mM are 47% for L-lysine, 42% for orni- 
thine, and 8% for D-lysine) (Nadai, 1958). Johnstone (1958) stated that 
there is evidence in intact ascites cells that ornithine can interfere with the 
transport of arginine into the cells, as well as inhibit arginase intracellularly, 
so that this additional site of inhibition must be borne in mind. 

An attempt was made by Sen (1959) to determine the effects of L-lysine in 
vivo in order to evaluate its possible use in uremia. Bilaterally nephrec- 
tomized dogs show an increase of blood urea at a rate of 15-16 mg/day and 
die in 80-84 hr. When L-lysine is injected daily and intravenously at a 
dose of 1 g, the blood urea rise only 3-4 mg/day and the animals survive 
for 274-278 hr. Thus it would seem possible to reduce urea formation in 
vivo with this competitive inhibitor, but the clinical benefit of this remains 
to be tested. 

L-AMINO ACID OXIDASES 

These enzymes from snake venoms and mammalian tissues oxidize the 
L-isomers of the monoamino-monocarboxylate amino acids, the substrate 
used often being L-leucine. The snake venom enzyme was shown by Zeller 
and Maritz (1944) to be inhibited by various aromatic sulfonates. Some of 
the results are shown in Table 2-14, from which it may be seen that 
p-nitrodiphenyl sulfonate is the most potent inhibitor studied. The inhi- 
bitions seem to be competitive and a complex between the sulfonate group 
and an enzyme amino group was postulated. Benzoate is a rather weak 
inhibitor of rat kidney L-amino acid oxidase, 10 mM inhibiting 28% 
(Blanchard et al., 1944). However, the ammonium ion is a surprisingly good 
inhibitor, 12 mM producing 69% depression of activity. 



L-AMINO ACID OXIDASES 



339 



Table 2-14 
Inhibition of Snake Venom l-Amino Acid Oxidase by Sulfonates" 



Inhibitor 



Sulfonate 
(mM) 



L-Leucine 

(mA/) 



Inhibition 



HO 



N=N- 



SO, 



2.9 



0.01 






//_\Vn=n^/ Vso; 



2.9 



0.01 



76 




2.9 



2.9 



2.9 



0.01 



0.01 



0.01 



66 



55 



50 



0,N 




0.33 



1.67 



31 



0,N 



// \V_N„_// V-so; 



0.07 



0.01 



21 



a From Zeller and Maritz (1944). 



340 2. ANALOGS OF ENZYME REACTION COMPONENTS 

The L-amino acid oxidase from the hepatopancreas of Cardium tubercula- 
tum is more specific than the venom or kidney oxidases, since many L-amino 
acids are not oxidized but are inhibitory (Roche et al., 1959). The inhibitions 
by 16.7 mM L-leucine are shown in the following tabulation. At pH 7.6 
the inhibitions are competitive, but not at pH 9.2. Apparently this enzyme 
can complex with both l- and D-isomers although in no case is a direct 
comparison possible. 





% Inhibition at: 


Amino acid 








pH7.6 


pH9.2 


DL- Alanine 


89 


71 


D-Alanine 


39 


— 


L-Serine 


54 


73 


Glycine 


.51 


56 


L-Glutamate 


26 


— 


L- Proline 


24 


74 


L- Valine 


24 


30 


L-Threonine 


20 


14 


L- Aspartate 


13 


12 


D-Histidine 


12 


— 


D- Leucine 


5 


— 



D-AMINO ACID OXIDASE 

D-Amino acid oxidase is also inhibited by certain amino acids, but no 
thorough studies have been reported. The enzyme from pig kidney oxidizing 
D-leucine is inhibited by DL-leucinamide, DL-leucylglycine, glycyl-DL- 
leucine, and DL-leucylglycylglycine, but not glycylglycine (Heimann- 
Hollaender and Lichtenstein, 1954). It is interesting that the oxidation 
of D-phenylalanine is inhibited by DL-iV-ethylphenylalanine and the oxida- 
tion of D-leucine by DL-A^-ethylleucine, since A'-substituted amino acids are 
usually not inhibitory for any enzymes acting on amino acids. D-Lysine is 
a good inhibitor of this enzyme oxidizing D-alanine (Ky„ = 3.3 mM, and 
K^ = 5 mM), but L-lysine is completely inactive (Murachi and Tashiro, 
1958). The D-amino acid oxidase from pig kidney with glycine as a substrate 
is competitively inhibited by L-leucine {K, = 1 mM) (Neims and Hellerman, 
1962). Pyruvate not only competes with D-alanine {K^ = 43 mM) but accel- 
erates the photodecomposition of FAD (Yagi and Natsume, 1964). However, 
the most interesting and best studied inhibition of D-amino acid oxidase is 
that of benzoate and its derivatives, and the opportunity will be taken in 
this section of discussing not only the actions of the benzoates on this en- 
zyme but also on other enzymes and metabolism in general. 



D-AMINO ACID OXIDASE 



341 



Benzoates and Related Compounds on D-Amino Acid Oxidase 

The oxidation of D-alanine in slices and homogenates of rat liver and 
kidney was shown to be markedly inhibited by 1 mM benzoate by Klein and 
Kamin (1941). A preparation of D-amino acid oxidase from pig kidney was 
thus obtained and found to be inhibited 79% by 0.1 mM, this being rever- 
sible upon dialysis. Several substituted benzoates are also inhibitory but 
all are less potent than benzoate; benzamide is inactive. The inhibitions of 
a lamb kidney D-amino acid oxidase by benzoate and j9-amino-benzoate 
were shown by Hellerman et al. (1946) to be competitive with substrate. 
The rate of spontaneous inactivation of the apoenzyme is reduced by either 
substrate or FAD, and benzoate was shown by Burton (1951 a) to have a 
comparable action, indicating combination with the active center. 

Before discussing the more detailed mechanism of the inhibition we 
shall turn to three studies providing information on the structural require- 
ments for inhibition. Bartlett (1948) compared many substituted benzoates 
(Table 2-15) and found only four to be more potent inhibitors than benzoate, 

Table 2-15 
Inhibition of Pig Kidney d-Amino Acid Oxidase by Substituted Benzoates " 



Substituent 


Relative 


— Ji^ of binding 


(kcal/mole) 










ortho 


meta 


para 


F 


4.26 




6.25 


CI 


3.27 


6.81 


5.39 


Br 


2.11 


6.40 


5.05 


I 


2.84 


5.39 


3.90 


OH 


4.55 


5.20 


3.27 


NHa 


4.26 


4.26 


2.84 


NO, 


2.52 


5.12 


4.90 


CH3 


2.11 


6.25 


5.12 


OCH3 


2.11 


3.90 


3.78 


COOH 


1.85 


2.11 


2.84 


None 


— 


5 . 57 


— 



" The substrate is DL-alanine at 30 mM and the pH 7.6. Binding energies calculated 
from concentrations for 50% inhibition. (Data from Bartlett, 1948.) 



pure competitive inhibition with respect to substrate being observed. 
J. R. Klein (1953, 1957) demonstrated inhibition, often potent, by various 
aromatic carboxylates (Table 2-16), and Frisell et al. (1956) provided further 



342 2. ANALOCxS OF ENZYME REACTION COMPONENTS 

Table 2-16 
Inhibition of Pig Kidney d-Amino Acid Oxidase by Aromatic Acids " 



Inhibitor 



Ki Relative — AF oi binding 

(mM) (kcal/mole) 



2-Chloromethyl-5-hydroxy-l,4-pyrone 

Kojate 

TO-Toluate 

Pjrrrole-2-carboxylate 

Benzoate 

Furan-2 -acrylate 

p-Toluate 

Nicotinate 

Cinnamate 

Furan-2-carboxylate 

l,2-Pyrone-5-carboxylate (coumalate) 

Indole-3-acetate 

Hydrocinnamate 

Phenylacetate 

l,4-Pyrone-2,6-dicarboxylate (chelidonate) 

o-Toluate 



0.004 


7.68 


0.021 


6.65 


0.022 


6.62 


0.026 


6.52 


0.048 


6.14 


0.052 


6.08 


0.08 


5.82 


0.11 


5.63 


0.56 


4.62 


0.60 


4.58 


0.60 


4.58 


2.3 


3.74 


5.5 


3.21 


6.4 


3.12 


15 


2.59 


29 


2. IS 



"The substrate is DL-alanine and the pH 8.0-8.3. (Data from J. R. Klein, 1953, 1957.) 

data on aliphatic and heterocyclic carboxylates (Table 2-17). Some of the 
conclusions regarding relations between structure and inhibition derived 
from these investigations will be summarized. 

(1) A negatively charged anionic group is necessary for activity. This is 
seen from the lack of inhibition by benzamide, nicotinamide, and cinnama- 
mide. It is likely that the COO" group of the inhibitors reacts with the en- 
zyme cationic site normally reacting with the amino acid C00~ group. A 
SO3"" group can replace the COO" but is less effective. 

(2) Klein has emphasized the importance of a positive charge at a distance 
from the COO" group approximating the separation in amino acids. Most 
of the potent inhibitors can be written in structures possessing such a 
positive charge by virtue of resonance effects. Benzoate, for example, 
resonates between the following structures: 

- -J- 

/ \ / / \ 



v>\ ^% 






/ \ 





/ \ 





/ 


// ^ 


V / 


/ \ 


/ 


_ 


)=c - 


+( ) 


= c 


\ 


\ / 


'^ \ 


\ / 


\ 





\ / 





\ / 






D-AMINO ACID OXIDASE 



343 



Table 2-17 
Inhibition of Lamb Kidney d-Amino Acid Oxidase" 



Inhibitor (3 mM) 



Structure 



% Inhibition 



Crotonate 

Butyrate 

Dimethylacrylate 



Isovalerate 



Fumarate and maleate 



CHj— CH=CH— COO' 

CH3— CH2— CHj— COO' 

H,C 
^ \ 

C=CH— COO 

H3C 

H3C 

CH— CH,— COO" 
H3C 

'OOC - CH= CH— COO' 



99 




70 



Cinnamate 



r y— CH=CH— COO' 



100 



Hydrocinnamate 



CH,— CHo^COO 



55* 



Cinnamamide 



(^ y — CH = CH-CONH2 



Phenylacetate 



CHo— COO 



15 



/>-Toluenesulfonate 



H,C- 



SO, 



34 



Sulfanilate 



-h,nV/ V 



so. 



344 



2. ANALOGS OF ENZYME REACTION COMPONENTS 
Table 2-17 (continued) 



Inhibitor (3 mM) 



Nicotinate 



Structure 



N- 
// W 



COO 



Inhibition 



65 



Nicotinamide 



N — V 



CONH, 



2-Furoate 



O^ ^COO 



W 



100 



Indol e - 2 - ca rbox yl ate 




H 

N^ COO' 



100 



Benzoate 

/»-Methoxybenzoate 
o - Methoxybenzoate 
/J-Carboxybenzoate 
o - Carboxybenzoate 
Hz-Carboxyben^oate 

Cyclohexanecarboxylate 



€y 



COO 



( V-coo' 



100 

100 

67 

74 

3 

20 

53* 



Hydroquinone 

Tribromophenol 

/>-Aminophenol 

Triiodophenol 

Catechol 

3, 5-Dihydroxyphenol 

2, 3-Dihydroxyphenol 

Resorcinol 



73 

40 

30 

19 

19 

5 







" The substrate is o-alanine at 6.25 mM and the pH 8.3. The inhibitions marked 
with an asterisk are at least partially due to contamination with the unsaturated 
compounds. (Data from Frisell (7 (v/., 1956.) 



D-AMINO ACID OXIDASE 



345 



and some of the other inhibitors may be written as: 



,/0 




Nicotinate 



^O 



2-Furancarboxylate 



HOH,C 




W A-""-""=V 



HCH^=CH-CH=C:f - 
O 




O 



Cinnamate 



Crotonate 



Indole - 2 - carboxylate 



Such structures would be less possible or impossible for hydrocinnamate, 
l,4-pyrone-2,6-dicarboxylate, cyclohexanecarboxylate, phenylacetate, and 
some of the other weaker inhibitors. The orientation relative to the sub- 
strate, according to Klein, would be represented as: 



+ 
/C-CC 






Substrate 



Inhibitor 



where X is carbon, oxygen, or nitrogen. 

(3) Frisell and his co-workers, on the other hand, emphasized the impor- 
tance of a double bond near the carboxylate group. Saturation of crotonate, 
cinnamate, dimethylacrylate, and benzoate definitely reduces the inhibition. 
They postulated that this double bond might correspond in position on the 
enzyme to the double bond of the iminoketonic form of the dehydrogenated 
product, and thus according to their theory the structural correspondence 
would be: 



Product 



— C> 



^c-c: 



,o 



Benzoate 



-'^--c-cr" 



Aliphatic carboxylate 



(4) These two theories are not incompatible, since we have seen that 
the presence of a positive charge generally depends on resonance, which in 
turn requires double bonds and either conjugation or hyperconjugation. 



346 2. ANALOGS OF ENZYME REACTION COMPONENTS 

(5) The different potencies of the substituted benzoates would be explained 
on the basis of the effects, such groups would have on resonance and the 
magnitude of the positive charge, but in addition other factors must play 
a role, for example the dipole moments of the ring X bonds and the possible 
interactions of the substituent groups themselves. One might expect ortho 
groups to decrease the binding and it is true that all are less inhibitory than 
benzoate. Bartlett pointed out that the inhibition increases markedly with 
the electronegativity of the halogens. It is very interesting to attempt to 
interpret the results in Table 2-15 but without more knowledge about the 
nature of the binding any hypotheses must be vague for the time being. 

(6) It is unlikely that the degree of ionization of the carboxyl group is 
important here because all of this work was done between pH 7.6 and 8.3 
where a negligible fraction is undissociated. The p/iT^'s of all the substituted 
benzoates tested run from 2.85 to 4.65. However, the series of phenols 
studied by Frisell et al. (1956) may have to be considered in terms of the 
ionization of the OH groups when relating structure to activity. 

(7) The K-^ for some selected inhibitors were determined by Frisell 
et al. (1956) and from these the relative binding energies may be estimated 
(see accompanying tabulation). The 1.45 kcal/mole difference between cin- 



Source of enzyme Inhibitor 



A", Relative — AF oi binding 
{mM) (kcal/mole) 



Lamb kidney 


Cinnamate 




0.076 


5.85 




Crotonate 




0.038 


6.28 


Pig kidney 


Cinnamate 




0.22 


5.20 




Benzoate 




0.021 


6.65 




Indole-2-carb 


axylate 


0.0034 


7.77 



namate and benzoate (it is 1.52 kcal/mole from Klein's data) may be 
attributed to a more satisfactory location of the positive charge in benzo- 
ate. The lesser affinity of the enzyme for nicotinate compared with benzoate 
(0.51 kcal/mole) could be due to the reduction in resonance brought about 
by the asymmetry of the former. If the 1.94 kcal/mole difference between 
pyrrole-2-carboxylate and furan-2-carboxylate is related to the different 
amounts of positive charge on the nitrogen and oxygen atoms, this would 
not be surprising since pyrrole should resonate more effectively. Finally, 
if the positive charge theory is valid, the 3.02 kcal/mole difference between 
benzoate and phenylacetate might indicate the amount of binding contri- 
bution from the positive charge. The positive charge may, of course, be 
stabilized somewhat by an enzyme anionic site with which it interacts. 



D-AMINO ACID OXIDASE 347 

Although the concept that these inhibitors bind to the enzyme and com- 
pete with the substrate is very reasonable, some recent evidence may 
indicate that the situation is a little more complex. The values of K^ for 
benzoate should be the same for every substrate, according to the classic 
treatment, but Klein (1956, 1960) has found that they are not, although 
all the 1/v — 1/(S) plots are definitely competitive. The A;/s may vary as 
much as almost 3-fold (see tabulation below). One possibility is that in 



Substrate 


Relative F,„ 


Km 

[mM) 


{mM) 


Alanine 


1.00 


6.3 


0.059 


Proline 


1.66 


5.8 


0.070 


Phenylalanine 


1.39 


14.0 


0.092 


Valine 


0.60 


4.6 


0.096 


Isoleucine 


0.79 


4.1 


0.104 


Methionine 


1.32 


5.3 


0.163 



the aqueous extracts of pig kidney there are different oxidases for each 
substrate but Klein prefers to assume that the inhibitors may react with 
the ES complex to release the substrate: 

ES + I ^ EI 4- S 

The equilibrium constant K = (ES) (I)/(EI) (S) depends on the substrate 
used and in the usual rate equation for competitive inhibition, {l)KJK^ 
would be replaced by {l)IK. Since K = KJK„ the determined values of K 
should be inversely proportional to K,. They are not inversely proportional 
to K„„ but is K„, = K, in this case? It may be noted that this mechanism is 
not quite the same as uncoupling inhibition, since there an ESI complex is 
formed and the substrate is not forced off. I must admit that I cannot 
easily visualize how a competitive inhibitor can actively displace an enzyme- 
bound substrate molecule. 

Although Hellerman et al. (1946) reported the inhibition of D-amino acid 
oxidase by benzoate to be independent of FAD concentration, there is 
more recent evidence that certain aromatic carboxylates and phenols not 
only can compete with substrate, but can also either compete with FAD 
or complex directly with FAD (Yagi et al., 1957, 1959, 1960). The constants 
for each of these reactions are given for a few of these inhibitors in the fol- 
lowing tabulation. Only the carboxylates compete with substrate, while 
the phenols act by the other two mechanisms; the substances with both 
COO- and OH groups react in all three ways, although competition with 
the substrate is the most important. Yagi and his group have recently 



348 2. ANALOGS OF ENZYME REACTION COMPONENTS 



K^ (mi/) 



Inhibitor 



Competition 


Competition 


Complex 


with substrate 


with FAD 


with FAD 


0.0145 






0.35 


1.05 


0.65 


6.35 


15 


75 


0.305 


7.25 


— 


— 


0.85 


— 


— 


9.15 


— 


— 


56.5 


32 


— 


28 


110 


— 


5.0 


0.5 


— 


3.9 


0.02 


— 


2.3 


0.04 



Benzoate 

Salicylate 

^-Aminosalicylate 

p-Arainobenzoate 

p-Nitrobenzoate 

Aniline 

Phenol 

m-Aminophenol 

p-Nitrophenol 

2,4-Dinitrophenol 

2,6-Dinitrophenol 

2,4,6-Trinitrophenol — 0.64 0.012 



crystallized the complex of apoenzyme, FAD, and benzoate, and found 
equimolar amounts of each component present. 

It would be interesting to have more data on the effects of these inhibitors 
on L-amino acid oxidases. Benzoate inhibits the L-amino acid oxidases 
from rat kidney (Blanchard et al., 1944) and snake venom (Zeller and Maritz, 
1945), but does not inhibit the enzyme from Neurospora even at 10 niM 
(Burton, 1951 b). 

Benzoate on Other Enzymes and Metabolism 

The endogenous ammonia formation and respiration of rat kidney slices 
are inhibited 52% and 66%, respectively, by 25 mM benzoate (Herner, 
1944). One might assume that the former might be attributed to inhibition 
of amino acid oxidases, but this is unlikely because benzamide, phenyl- 
acetate, and /?-phenylpropionate are even more potent inhibitors than 
benzoate. The deamination of various l- and D-amino acids in kidney is, 
however, strongly inhibited by benzoate, at least in part competitively. 

The inhibition of respiration by benzoate was first observed by Grif- 
fith (1937) in slices and minces of various tissues in the presence of glucose, 
but no analysis of the site of action has been published, although in con- 
nection with recent studies on the salicylates some effects on mitochondria 
have been investigated. Benzoate inhibits the oxidations of citrate, a- 
ketoglutarate, and succinate in rat kidney homogenates but is invariably 
less active than salicylate (E. H. Kaplan et al., 1954); and neither the Og 
uptake nor the P : ratio is affected markedly in rat liver mitochondria, 
although salicylate uncouples strongly (Brody, 1956). Endogenous phosphor- 



D-AMINO ACID OXIDASE 349 

ylation in liver mitocliondria is inhibited 55% by 10 mM benzoate (Wein- 
bacli, 1961). One can conclude from this limited material that benzoate is 
certainly a weak inhibitor of cycle oxidations and phosphorylations. Bosund 
(1959, 1960 a, b) has investigated the effects of benzoate on the metabolism 
of glucose and pyruvate in Proteus vulgaris in attempting to elucidate 
the mechanisms for the bacteriostatic activity. There is no interference 
with glucose metabolism to the acetate level, and acetate was found to 
accumulate. The respiratory quotient is increased from 1.24 to 1.82 by 
benzoate during the oxidation of pyruvate and the Oa/pyruvate ratio is 
decreased. The oxidation of pyruvate in yeast is quite strongly inhibited 
by benzoate (50% at 0.4 mM), especially at low pH's where penetration is 
better, but acetate oxidation is less sensitive. It is quite possible that these 
inhibitions play a role in the suppression of growth, which for yeast requires 
5 mM benzoate at pH 5.1 and 60 mM at pH 6. It is clear that much more 
work must be done before the mechanisms of respiratory inhibition are 
understood. 

Benzoate can also interfere in lipid metabolism, as demonstrated many 
years ago by Jowett and Quastel (1935 a, b), but the mechanism is still 
unknown. In liver slices it was claimed that benzoate at around 0.5-2 
mM inhibits specifically the oxidation of fatty acids, and the oxidation of 
crotonate 63% at 1 mM. There is progressively less effect on the higher 
fatty acids, little inhibition of decanoate being observed. It is possible 
that the benzoate ring simulates the aliphatic chains of butyrate or croton- 
ate enabling it to compete with these substrates for some enzyme; it would 
be interesting to know if benzoate can participate in any of these reactions 
(e. g., if benzoyl-CoA is formed) and deplete the systems of some cofactor. 
Benzoate is a weak inhibitor of tyrosinase (Ludwig and Nelson, 1939), 
chymotrypsin (Foster and Niemann, 1955 b), p-aminobenzoate acetylation 
(Koivusalo and Luukkainen, 1959), and NADPH dehydrogenase (Kasa- 
maki et al., 1963); it does not effect shikimate dehydrogenase (Balinsky and 
Davies, 1961 b) or D-glutamate oxidase (Mizushima and Izaki, 1958) at 
1 mM, or a-ketoisocaproate decarboxylase at 4 mM (Sasaki, 1962). 

Kojic Acid 

The potent inhibition of D-amino acid oxidase by kojic acid is interesting 
in light of the central nervous system effects observed in dogs, rabbits, and 
rats, namely, ataxia, excitement, and convulsions (Friedemann, 1934). 
Kojic acid was first isolated by Saito in 1907 from Aspergillus oryzae and 
has since been found in many species of Aspergillus. It is a weak antibiotic, 
inhibiting growth of most bacteria at 2-15 mM, but is particularly active 
against Leptospira, complete growth inhibition being observed at 0.007 m.N 
(Morton et al., 1945). Toxic effects are produced in dogs by 150 mg/kg and 
in mice by 250 mg/kg when injected parenterally; the LD50 for mice is 



350 2. ANALOGS OF ENZYME EEACTION COMPONENTS 

1.5-2.0 g/kg. Leucocytic activity and phagocytosis are not affected by 18 
raM kojic acid. A biochemical study was undertaken by Klein and Olsen 
(1947), who found that the respiration of muscle and heart mince is resistant 
to kojic acid, whereas 10 mM suppresses the respiration of liver 40%, 
kidney 20%, and brain 15%. The convulsant dose corresponds to a tissue 
concentration around 4-50 mM. The oxidation of both l- and D-amino 
acids in liver homogenates is quite strongly inhibited in a competitive 
fashion: for example, 50% inhibition is given by 0.04 niM for L-methionine 
and by 0.12 mM for l- and D-phenylalanine. Xanthine oxidation is also 
inhibited (50% at 0.7 mM). It was suggested that kojic acid may be an 
inhibitor of flavin enzymes, and it is possible that some direct complexing 
with FAD may occur. Nevertheless, FAD does not influence the inhibition 
of D-amino acid oxidase. There is no inhibition of the oxidation of succinate, 
tyramine, L-proline, choline, or urate at 5 mM kojic acid. Although the 
metabolic effects are interesting, it is impossible to correlate any of these 
inhibitions with either the central effects in animals or the bacteriostatic 
activity. 

ANALOG INHIBITION OF THE METABOLISM 
OF VARIOUS AMINO ACIDS 

Many reports indicate the influence of analogs on various enzymes con- 
cerned with amino acid metabolism, but in most cases insufficient work 
has been done to draw clear conclusions about the mechanism of the bind- 
ing to the enzymes. It will suffice to present some of the results in Table 
2-18. Probably many of these inhibitions are competitive but they have 
been so indicated only when graphical analysis has shown this to be true. 
A few of these inhibitions may be significant in feed-back control or in the 
general regulation of amino acid metabolism. 

Generally speaking, there are certain types of amino acid analog that 
have proved to be effective inhibitors. Mcllwain (1941) pointed out that 
aminosulfonate analogs of amino acids are frequently bacteriostatic and 
that this inhibition is reduced by adding the normal amino acids. If staphy- 
lococci are trained to be independent of exogenous amino acids, the a- 
aminosulfonates no longer inhibit. However, the exact sites of action 
of these analogs have not been determined. Umbreit (1955 b) has discussed 
the general inhibitory properties of the a-methylamino acids and pointed 
out that the inhibitions are often competitive only for a short interval 
if both substrate and inhibitor are present, whereas they are noncompetitive 
if the inhibitor is added first. This is a matter of terminology; the inhibitions 
are probably competitive but appear to be noncompetitive because of the 
very high affinities of some analogs for the enzymes (an inhibition can be 
competitive even though irreversible but the substrate must be given an 



ANALOG INHIBITION OF THE METABOLISM OF VARIOUS AMINO ACIDS 351 

opportunity to compete). A third group of interesting analogs is the halogen- 
substituted amino acids, which are often potent inhibitors of protein syn- 
thesis and growth. The fluoroamino acids are particularly active. The 
m-, 0-, and p-fluorophenylalanines all inhibit the formation of the adaptive 
maltase in yeast, the last being the most potent (Halvorson and Spiegelman, 
1952). On the other hand the p-chloro- and js-bromophenylalanines do not 
inhibit. The incorporation of phenylalanine and other amino acids into 
ascites cell proteins is competitively inhibited by o-fluorophenylalanine; 
this is in part due to a depression of transport into the cells and in part due 
to block of some unknown steps in the incorporation (since labeled phenyl- 
alanine accumulates in cells) (Rabinovitz et al., 1954). There is also an inhi- 
bition of protein synthesis in rat liver in vitro by the fluorophenylalanines, 
this leading to a net breakdown of tissue proteins since the constant bal- 
ance of synthesis and degradation is disturbed (Steinberg and Vaughan, 
1956). a-Amino-/?-chlorobutjTate is an analog of valine and inhibits valine 
incorporation into rabbit reticulocyte proteins, including hemoglobin; it 
was suggested that the analog enters a precursor protein which is unable 
to assume the proper configuration of hemoglobin and thus there is accu- 
mulation of protein intermediates (Rabinovitz and McGrath, 1959). As 
pointed out previously, some of these analogs are incorporated into cell 
proteins. p-Fluorophenylalanine-C^* is incorporated into the proteins of 
muscle, blood, and liver when fed to rabbits, this being a replacement of 
phenylalanine (Westhead and Boyer, 1961). The replacement of phenyl- 
alanine in aldolase is 25% and in 3-phosphoglyceraldehyde dehydrogenase 
16%, and in each case the enzyme activities are normal. Despite this ap- 
preciable incorporation, the rabbits suffer no obvious biochemical or physi- 
ological disturbances, so that mammals may well differ from microorgan- 
isms in the response to this analog. 

Feedback inhibition in the biosynthetic pathways of amino acids is an 
important aspect of regulation but we can touch only briefly on this prob- 
lem. An interesting example of this has been studied in connection with 
the synthesis of histidine, since the enzyme inhibited is the first in the 
pathway and catalyzes a reaction not involving substrates structurally 
similar to histidine (Martin, 1963). This enzyme is phosphoribosyl-ATP 
pyrophosphorylase and the reaction catalyzed is: 

5'-P-ribosyI-PP + ATP :^ N-l-(5'-P-ribosyl)-ATP + PP 

Histidine is a surprisingly potent and specific inhibitor with K, = 0.1 mM. 
Related compounds inhibit weakly or not at all; 2-methylhistidine, for 
example, exhibits weak inhibition with K^ = 2.4 mM. The inhibition varies 
with the pH but maximal inhibition is exerted at physiological pH. HgClg 
at 0.03 mM does not inhibit the enzyme but blocks almost completely the 
inhibition by histidine. This coupled with the fact that the inhibition by 



352 



2. ANALOGS OF ENZYME REACTION COMPONENTS 



P5 






CC 73 



A Cm 
>> O 

H 



S 

N 

c 



GO 05 

-H (T^ l> 



o o o 



3 






s 


O 


s 




o 


CO 

'o 


o 




tJ 


c 


c 










<D 


>, 


S 


■g 


C 


<0 


<? 


<5 




"1 


a 


a 


hJ 


j 


ij 


h!] 


^ 



-^ .s 



3 03 t- 

o o ce 

to CO Kj 

I— I t— I K* 



M 





^ 








>. 




X 




O 


(U 


x> 


C 


kl 




<D 


4) 


T! 


l-J 





« Xh 

^ " 






O O O O <M 



r^f»f^t^i--.i--.t>t-i>i> OO 



P O 



c 

"a 

c 
■3 

o 



c 

^ cS 

o jh 

S ft 






S -Si <s 






'2 

eS 

s 

O 

c 
S 



T3 Is 



ANALOG INHIBITION OF THE METABOLISM OF VARIOUS AMINO ACIDS 353 






e8 05 



CO 


CO 


'0 ^ 




eS 


o3 


_ TS 


- "O 


C b. 


c b 


o <o 


o « 


CO j: 


aj ^ 


e ft 


fi ft 


:s 2 


^ 1 



73 



>> 

^ 



C^ - Pc; --^ f^ - j^ ^ (^ ^ 



©O-HCOOMMCOOO 0I>0 O (NTt''*'*-*OpO0 O-hgOCO 

o CO CO CO CO CO 

ll||l|||00 OOO O (M(M(M(N(MC^'M rt^_^ 



c 

ft S 3 

a ^ftx3ft><:2 ® 

-^ c •-ft'oSft'^e -g.s-S 

>> ft oi=ip>>^cP>.g'c ^^-sSib c c 

»o H H Pm ^- ^ ^ -^ >^- « >: -^ '« '« i :S c -S •= 5 

a Q QQoa Wuoc K ^ ^=t O ^ Pi ^ O O 



•1= » "o. 



ft -ks 



■S 2 .S -ta -»i §" S 

S3 S e| c3 ^ ^ 

« O (23 05 ^ ^ 



» 2 



ffi 



354 



2. ANALOGS OF ENZYME REACTION COMPONENTS 



tf 



fc< 



Si 



ft 



T3 




C 


^— ' 


cS 


c 


u 
O 


o 


c 


o 


c 


> 






p 


w 



-H lO o o 

IC (M lO 00 

rt Csl (M 



o o o o o o o 



C -i:^ S « <» .S r* 

S E? .5 -C -S -3 ^ 



-— >^ o • 0) 



9 5« o 



^ O ^.CC ^ > ^ 
Q Q O Q Q O O 



Cq 








■<* 


^ 


CO 


73 ^ 


C5 


e8 »C 






'e 


C ^ 


V 


•- Si 


!>. 


II 


« 


^ 


|6 


1 



>o 


<>J 


<:d 


C<l 


O 


o 


-* 


Tt* 


-* 


-* 


■* 


% 


C^ 


d 


(M 


f>\ 


C<1 


d 



ec CO >c CO lo 

'o^fod—iMCOfM 



O 



O 



O 



o 









_g 




"2 


4i 




^ 




"? 


'c 




3 


"2 


Q 


^ 


<o 


T. 


eS 


^ 




1 


3 




">> 


<I> 


A 


">> 


.2 




ft 
O 


45 


cS j2 


IS 


c 


o 


12 


< -S 


H 


c 


3 


H 


1 o 


(N 


^ 


S 


*? 


o 


<^ 


<^ 


9, 


cii. 



.a ro 


)M 




C !« 


"2 M 




JS '>. 


^ 2 






a 


ce y. 


's -^ 


^^ o 


--S ee 


>. 2 


b > 


f>J 


? '« 


9. '43 


N 








PM ^ 


CL| 





ANALOG INHIBITION OF THE METABOLISM OF VARIOUS AMINO ACIDS 355 







« 


-Q 






(^ 


(^ 


" in 




05 


05 


s ^ 




^ 


^ 


CS -H 


T3 


S 


-O « 


tH , 


C 




C fl 




c3 
a: 


1 

GQ 


03 o 

CO OS 




s 


T3 


S 2 




cS 


"o 


c6 o 


< 


o 


^o 



O OS 

O •-H Oi 

O O Oi Tf* rt t^ CD 



O rt (M O 



lOt^CO <lDt^(M>OTtf(N 0>OC<I 100505 10 I I I 0»000 

■>*M-*i f0»O<M-*CO<M »OCOM lO >0 III OOI><M 



(M«0 C<ICO lOlOlO -^ 



CO CO CO o 

oooooo I I I oddd 



O O Q O O 



2 9 



s =1 i i ■&! 

'-^ >> rt O O 



3 


ft 
o 




01 


-0 




'>^ 


d 










t^, 


s 


o 


>. 


c^ 


<i 


H 


H 



ra cc J- o< >, a 







ft 




ft 








's 


^ 




« 


s 


!«! 


-2 


X 


a> 


jj 


0) 


00 

C 

HP 


c3 


e3 




1 


O 






Ph 


'o 

SI 

cS 

1— 1 


1 

3 


o 

"3 


cS 
ft 

CO 

< 


J 


Q 


J 


1— 1 


►J 


ij 


fe 


w 


OJ 


p 



73 


»c 


a 


Oi 


cS 




c 


''■— ' 






4J 


fl 




3 






Cj 


Tl 


Ih 


c 



o 








o 


;^ 


o o o o 


OOO 


OOO 


o ^ 





eS 




C 


(1> 


O 


rrt 


ft 


-t:^ 


o 


^ 




crt 


ft 


ft 


o 


cc 


c 


ai 




>■. 


w 


X 


ri*. 


n 




t: 


Q 


>. 


fO 



^ pq pq pq a, *^ CM 

§s >>^ ^^-^ P^^-Sft ft^l 



356 



2. ANALOGS OF ENZYME REACTION COMPONENTS 



05 



^ 























00 

so 


o 


o 


S<l 


o 


o 

GO 


j2 
C 
1— 1 












i 


_ 



^^ 



ft ^ .ti 



w 



o 


03 


m 


Oi 


o 


O 




-C 


^ 




O 


O 



rt F^ ^ o CO 






^ 



03 



O 



o 







05 

o 












en b 


o 


:?; 


CO 


:^ 


1^ 




1-3 H 


w 


o 




o 


O 




J Jj 


J] 


Q 


<! 


Q 


Q 



D5 



EsJ 



o o o o o o 

o o o o o o 



o o o o o o 



o o 



■fi 




c2 


0) 






s 




to 


r> 


>■. 





o 



K O) ,iS 



^ - ^ G 



4^ .X .-= 



O .S 



m H cc Q a H 



» «? -2 



c i s 'p 

'3 to -C — 

S O & 

5 -*^ to 

-2 ti- -o 






0) 


s 




>, 








-l^ 


>^ 




o 


J- 


o 




c 





li 



o 






M 



2 o 



ANALOG INHIBITION OF THE METABOLISM OF VARIOUS AMINO ACIDS 357 



c3 



C C5 

s ^ 

o X- 









rt -^ 



fe 



I I 



-§ =3 "S o •= 

c3 r- C3 ^ Gj 

u C -If c ^ 

^ ci ^ ::: »3 

^ ^ & W c.' 

c c ■< j J 



05 









S "^ 



Ph 



Ti 


C3 


C 


7^, 


IS 






hn 






tH 


<u 


S 


^ 


fl 


c 


.^ 


0) 










Z 


C 



o 


O 


O 


2 


o 


s 


O 


2 






LI 


C-1 




'M 


m 
-M 





a o o 



Q o o o 



o 



•- .S 4^ 



Z^ •■-' -*^ r- '^ ^7^ 



a; .= .= :^ 






J C J K 



-s ^ -< 






&q 



c 








c 


d 


rt 


c 


o 


t^ 




» S 


S 


S c 


» 





T3 
>> 



H « 



fe 



1-1 »c 



~ o 
P- i 



® 


fi") 


c 


71 




rrt 






o 


0; 


0) 








^ 


c 


H 


>> 



.2 ^ 



a c 



j3 



O H 



358 2. ANALOGS OF ENZYME REACTION COMPONENTS 

histidine is noncompetitive indicates that the histidine site is different 
from the catalytically active site. ATP and 5'-P-ribosyl-PP protect the 
enzyme against inactivation by trypsin, and histidine eliminates this pro- 
tection, so it is possible that histidine modifies the enzyme configuration 
by combining with the feedback site. 

Aminooxyacetate 

An interesting analog studied recently in connection with the physiolog- 
ically important y-aminobutyrate (GABA) is aminooxyacetate (+H3N — 
— CH2COO"). This substances inhibits the GABA: a-ketoglutarate trans- 
aminase of E. coli very potently (40% at 0.0033 mM and 100% at 0.33 
mM when both substrates are 27 mM), so it was tested on a similar enzyme 
from brain and found to inhibit as strongly (Wallach, 1960). Aminooxyace- 
tate in doses of 6-50 mg/kg elevates the brain GABA in several species as 
much as 4- to 5-fold, peak levels being reached about 6 hr after administra- 
tion and high levels remaining up to 24 hr (Wallach, 1961 a; Schumann et al., 
1962). Reinvestigation of the transaminase inhibition led to a K^ of 0.0075 
mM (Z,„ is 9.66 mM for a-ketoglutarate and 27.6 mM for GABA); the 
inhibition is competitive with respect to both substrates. The alanine: 
«-ketoglutarate transaminase from rat liver and heart is also potently 
inhibited, around 60-80% by 0.0001 mM aminooxyacetate at pH 6.8 
(Hopper and Segal, 1964). Indeed, this transaminase seems to be more 
sensitive than either the GABA:a-ketoglutarate or aspartate:or-ketoglutar- 
ate transaminase. 

Inasmuch as GABA has been implicated in certain convulsive disorders 
(e. g., GABA formation in epileptic brains is apparently depressed), amino- 
oxyacetate was administered to animals made convulsive with thiosemi- 
carbazide and methionine sulfoximine (DaVanzo et al., 1961). Anticonvul- 
sant activity was observed but there is some doubt if this is correlated with 
the brain GABA levels since the time relations are not correct. The effects 
of aminooxyacetate on central nervous system function are complex. 
There is first a progressive depression and muscular relaxation with loss of 
certain reflexes, but at high doses tonic-clonic convulsions occur (DaVanzo 
et al., 1964 a). Pyridoxal-P antagonizes these convulsions and it was postu- 
lated that oxime formation occurs between aminooxyacetate and pyri- 
doxal-P. Pyridoxal-P does not reverse the transaminase inhibition in the 
brain (Wallach, 1961 b). W^allach also suggested that a depletion of suc- 
cinic semialdehyde, which arises from GABA by transamination, might 
also play a role in the convulsant action. Since a pyridoxal deficiency is pro- 
duced in rats by the administration of aminooxyacetate, DaVanzo et al. 
(1964 b) postulated another possible mechanism of action, namely, the 
inhibition of pyridoxal kinase, inasmuch as McCormick and Snell (1961) 
had shown this enzyme to be rather potently inhibited by the condensation 



ANALOG INHIBITION OF THE METABOLISM OF VARIOUS AMINO ACIDS 359 

product of aminooxyacetate and pyridoxal. It may be that aminooxyacetate 
should be classed as a carbonyl reagent which reacts with pyridoxal-P, 
rather than strictly as an amino acid analog, but it is sometimes difficult 
to distinguish these actions. Nevertheless, the effects of aminooxyacetate 
on GABA levels in the tissues will undoubtedly make it a useful compound 
for the study of the physiological role of GABA. 

Cycloserine 

D-Cycloserine (orientomycin, Oxamycin) is a tuberculostatic antibiotic 
isolated from Streptomyces which can be considered as a cyclic form of 
aminooxyalanine or 0-aminoserine. It is written in the zwitterion state be- 

I I 
H nh; 

Cycloserine 

cause there is some evidence that this is the inhibitory form (Neuhaus and 
Lynch, 1964). Its actions are similar to aminooxyacetate in many respects 
but differ occasionally in interesting ways. D-Cycloserine inhibits the growth 
of many bacteria and this is often well antagonized by D-alanine, and 
competitively inhibits the incorporation of D-alanine into a uridine nucleo- 
tide necessary for the synthesis of cell wall material. The growth of myco- 
bacteria is 50% suppressed, for example, by D-cycloserine at 0.03-0.045 mM 
and this is reversed by D-alanine but not by L-alanine (Zygmunt, 1963). 
L-Cycloserine is also inhibitory to certain bacteria and this is antagonized 
by L-alanine. It is interesting that *S. aureus can develop a 50-fold resistance 
to D-cycloserine, no cross-resistance with other antibiotics being observed 
(Howe et al., 1964). In animals it produces sedation, lethargy, muscular 
relaxation, ataxia, an accentuated startle response, and, above all, epi- 
leptic convulsions (Dann and Carter, 1964; Holtz and Palm, 1964). In these 
respects it at least superficially acts like aminooxyacetate and, furthermore, 
these effects are antagonized by pyridoxal. 

D-Cycloserine inhibits certain enzymes dependent on pyridoxal-P, such 
as the transaminases and glutamate decarboxylase, and it has been postu- 
lated that it simply reacts with pyridoxal-P to form a substituted oxime. 
However, this is not the usual reaction in which a Schiff base is produced, 
but involves an opening of the cycloserine ring. The inhibition of GABA: 
or-ketoglutarate transaminase is initially competitive with respect to GABA 
{K^ is 0.25 TCiM for the enzyme from cat brain and around 0.6 mM for the 
enzyme from E. coli), but a secondary progressive irreversible inhibition 
also occurs (Dann and Carter, 1964). This may be related to the hypothesis 
of Khomutov et al. (1961) that the decyclicized cycloserine forms an oxime 



360 2. ANALOGS OF ENZYME KE ACTION COMPONENTS 

bond with enzyme-bound pyridoxal-P, and also an acyl bond with a cat- 
ionic group at the active site, these two groups thus being connected by a 
bridge would prevent access to the site. The L-asparagine:a-ketoglutarate 

I 
HC = NH— 0— CHa— CH— CO 

I I 

Pyr X 



Apoenzyme 

transaminase is inhibited by L-cycloserine with respect to L-asparagine 
{K^ = 0.0001 mM) but D-cycloserine is a much weaker inhibitor {K^ = 
1 mM) (Braunstein et al., 1962). L-Cycloserine likewise competitively 
inhibits the L-alanine: a-ketoglutarate transaminase {K^ = 0.008 mM). 
This taken with previous work indicates that the cycloserines inhibit 
specifically those enzyme attacking substrates of the same optical isomerism. 
Another possible site of action of D-cycloserine in bacteria would be the 
D-alanyl-D-alanine synthetase, a block of which would prevent the incor- 
poration of D-alanine into cell wall material. Alanylalanine is often able to 
reverse the cycloserine inhibition of bacterial growth, sometimes more ef- 
fectively than alanine. One example of this is the inhibition of the prolif- 
eration of agents of the psittacosis group in chick embryo yolk sac (Moulder 
et al., 1963). Chick embryos infected with the mouse pneumonitis organism, 
for example, are well protected by D-cycloserine at 0.004-0.008 mM, and, of 
all the possible reversors tested, only alanylalanine is effective. The d- 
alanyl-D-alanine synthetase of Streptococcus fecalis is inhibited competi- 
tively with respect to D-alanine {K^ = 0.022 mM), and Neuhaus and 
Lynck (1964) felt that this enzyme may well be the major site of inhibition 
in certain bacteria. It is unfortunate that cycloserine and aminooxyacetate 
have not been accurately compared in any study. 



DIAMINE OXIDASE (HISTAMINASE) 

Enzymes in this group exhibit different degrees of substrate specificity 
depending on the source, but most oxidatively deaminate diamines of the 
type +H3N — (CH2),i — NH3+ (with maximal rates when n is around 5) and 
histamine, in all cases primary amines being attacked. The usual substrate 
in most studies has been cadaverine {n = 5). 

The diamines have been written as cations because the p^^'s are usually 
between 8.5 and 10.5. The amidines and guanidines exist almost entirely 
as cations at physiological pH since the p^^'s for these groups are around 
13 to 14. The structures have been written in a rather unconventional way 



DIAMINE OXIDASE (hISTAMINASE) 



361 



H3N— (CH2)— NHa"" 
Cadaverine 



Putrescine 



H3N— (CH2 )3— ^fH— (C H2 )4— NH — (CH2 )3— NH^ 



*2M 

Spermine 



CH3— (CH3)^C^ + 
Monoamidines 



HjN^ , , /NH2 

HzN'^ NH2 

Diamidines 






/=\ /NH3 

Dibenzamidines 



H,N— C 

^NH— CH3 

Methylguanidine 



H,N-C 



'NH^ 



-NH— NH2 
Aminoguanidine 



H^N^ /NH2 

+ C— NH— (CH,);,— NH— C + 

Diguanidines 



+ /NH2 

H3N— (CHs).— NH-C + 

Agmatine 



H^N^ /NH2 

+ C— S — (CHs)^— S-C- + 
HgN^ NH2 

Diisothioureas 



H2N /NH2 

+ C-NH — (CH2)4— NH-C + 
HgN^ NH2 

Arcaine 



to indicate the equivalence of the C — N bonds and the states of the amino 
groups, since there is resonance between forms such as the following for 
the guanidinium ion: 



+ /NH2 



H,N— C 



+ 
NH, 



^NH, 



H,N-C, 



,NH, 



NH, 

+ ^ 



Alkyl substitution does not reduce the resonance appreciably. 

Aliphatic monoamines, such as amylamine, are not substrates nor are 



362 



2. ANALOGS OF ENZYME REACTION COMPONENTS 



they readily bound to the enzyme since they inhibit weakly (Zeller, 1940). 
The short aliphatic diamines are very poor substrates but inhibit quite 
well; thus both ethylenediamine and trimethylenediamine inhibit the 
oxidation of cadaverine by pig kidney diamine oxidase (Zeller, 1938). 
If the amino groups of these inhibitors react with the same anionic enzyme 
sites as does cadaverine, these anionic groups must be fairly close so that 
cadaverine would have to assume a very bowed configuration. On the other 

NH3+ 
\ ^ 
CH, 
\ 

CH, 

/ ' 

NH+ 







H. 








C 










HgC 




CH2 


/ 




\ 


HgC 




CH2 


+ \ 




/+ 


H3N 




NH3'^ 




- 




- 





H,C- 
^H,N 



-CH, 

NH, 



(a) 



(b) 



(c) 



hand, the anionic groups may be separated by a distance corresponding 
to the amino groups in cadaverine and the inhibitors react with only one 
of the anionic sites ,the other amino group interacting with some anionic 
site outside the active center (as in (c)). 

Certain guanidine derivatives are more potent inhibitors. Guanidine 
itself is rather weak, inhibiting cadaverine (2 mM) oxidation 42% at 10 
mM, but methylguanidine inhibits 63% at 1 mM (Zeller, 1938). Although 
Zeller noted that in the pig kidney preparation methylguanidine did not 
inhibit histamine oxidation very well, Waton (1956) found marked inhibi- 
tion of histaminase activity in cat kidney, 0.01 mM inhibiting 42% and 
0.1 mM 75%. The oxidations of putrescine and agmatine are both well 
inhibited by methylguanidine (Zeller, 1940). An even more potent inhibitor, 
however, is aminoguanidine. Apparently the diamine oxidases differ in 
sensitivity to aminoguanidine; 50% inhibition is given by 0.00005 mM for 
the enzyme from pig kidney (Schuler, 1952), by 0.001 mM for the enzyme 
from cat kidney (Waton, 1956), by 0.01 mM for the enzyme from rabbit 
liver (Kobayashi, 1957), and the enzyme from mouse liver is not inhibited 
even by 0.1 mM. The nature of the inhibition is not clear, inasmuch as 
aminoguanidine is also a derivative of hydrazine and might act by attacking 
carbonyl groups: hydrazine and semicarbazide are, indeed, potent inhibitors 
of diamine oxidase (Schuler, 1952; Waton, 1956). A further complication is 
that aminoguanidine hydrolyzes to form semicarbazide and eventually 
hydrazine. It behaves chemically more like a hydrazine than a guanidine, 
and reacts with carbonyl groups without being hydrolyzed (Lieber and 
Smith, 1939). It is also possible that the NHNHg group simulates the 
CH2NH2 substrate group, as in the monoamine oxidase inhibitors, and forms 
a tight bond to the enzyme. In vivo inhibition of diamine oxidase by amino- 



DIAMINE OXIDASE (HISTAMINASE) 



363 



guanidine was demonstrated by Schayer et al., (1954) by injecting labeled 
cadaverine into mice and determining the C^^Og respired (Fig. 2-7). It is 
much more effective than isoniazid or agmatine. The metabolism of hista- 
mine should also be blocked by aminoguanidine. This was shown in three 
psychiatric patients by administering labeled histamine and finding that 



100 




-1.5 -0.5 

LOG INH DOSE (^g/G) 



+ 0.5 



+ 1.5 



+ 2.5 



+ 3.5 



Fig. 2-7. Effects of diamine oxidase inhibitors on the oxidation of cadaverine in 
mice, as determined by the formation of C'^Oj from labeled cadaverine. (Data from 

Schayer et al., 1954.) 



urinary imidazoleacetate-C^^ is reduced by aminoguanidine at 0.1-1 mg/kg 
(Lindell et al., 1960). A larger fraction of the histamine is excreted as 
methylhistamine, demonstrating a diversion of metabolic pathways by 
this inhibitor. 

Diamine oxidase is not strongly inhibited by monoamidines, but diami- 
dines, diguanidines, and diisothioureas of the proper chain lengths are 
quite potent inhibitors (Blaschko et al., 1951). The data for these series 
are summarized in Fig. 2-8. The correlation between chain length and 
inhibition is not nearly so clear as for monoamine oxidase (Blaschko and 
Duthie, 1945; Blaschko and Himms, 1955), and in some cases, as the dia- 
midines, there is surprisingly little variation of inhibition with chain length. 
For diamidines of n = 10-16, one wonders if the binding is actually to 



364 



2. ANALOGS OF ENZYME KEACTION COMPONENTS 




Fig. 2-8. Inhibition of pig kidney diamine oxidase with cadav- 
erine (5 niM) as the substrate and all the inhibitors at 1 mM. 
(From Blaschko et ah, 1951.) 
NH, 



(A) Monamidines 



H3C — (CHa)^ C 



NH 



(B) Diamidines 



H,N 



NH, 






(C) Dibenzamidines 



(D) Diguanidines 



H2N 
HN 
H,N 




0-(CH2)„-0— ( 



NH, 



NH, 



W 
NH 



C-NH-(CH2)— HN-C^ 
HN ^NH 



(E) Diisothioureas 



H,N 



NH, 



C-S- CH,L— S— C 

// \\ 

HN NH 



CARBOXYPEPTIDASE, AMINOPEPTIDASES, DIPEPTIDASES 365 

the substrate site entirely, or possibly to two substrate sites, or even to 
anionic groups outside the substrate site. The enzyme may have binding 
sites for the imidazole ring of histamine since imidazole inhibits 11%, 
imidazolelactate 20%, and histidine 4% at 6.7 mM when cadaverine is 
3.3 mM (Zeller, 1941). Urate also competitively inhibits the oxidation of 
histamine, but rather weakly. An excellent review of the structure-action 
relationships among the amidine derivatives is by Fastier (1962). 

CARBOXYPEPTIDASE, AMINOPEPTIDASES, 
AND DIPEPTIDASES 

Certain aspects of the inhibition of carboxypeptidase by substrate 
analogs were discussed in Volume I (page 292) to illustrate how certain 
interaction contributions could be estimated. We shall now attempt to 
visualize more clearly the orientation of these analogs on the enzyme sur- 
face. The data indicate that a three-point attachment of the substrate 
is necessary for catalysis. The enzyme sites may be indicated as follows 
(see Fig. 2-9 for hypothetical orientation of substrate): (A) the peptidatic 
site contains the mechanism of the electron displacement necessary for hy- 
drolysis and is probably positively charged, (B) the cationic site is a positively 
charged group that interacts electrostatically with the C00~ group, and 
(C) the electrokinetic site is perhaps a lipophilic region capable of reacting 
with alkyl or phenyl groups by dispersion forces. It is easy to see why 
D-substrates are not reacted since the peptide bonds would not be able to 
approach the peptidatic site. There is also an enzyme region near the projec- 
tion direction of the fourth asymmetric carbon bond that sterically prevents 
attachment of groups larger than an amino group, and thus the D-isomers 
usually do not bind and are not inhibitors. Only two-point attachment is 
necessary for inhibitors, and most that have been studied bind at the cationic 
and electrokinetic sites. 

The relative binding energies for inhibitors in Table 1-6-26 were calculated 
from the data of Smith etal. (1951). Earlier studies by Elkins-Kaufman and 
Neurath (1949) provide additional information on the competitive inhibitors 
in the accompanying tabulation. It is interesting that D-phenylalanine is 



Inhibitor 


(mM) 


Relative — AF oi binding 
(kcal/mole) 


^-Phenylpropionate 


0.062 


5.96 


Phenylacetate 


0.39 


4.83 


y - Phenylbutyrate 


1.13 


4.17 


D - Phenylalanine 


2 


3.82 


2?-Nitrophenylacetate 


2.5 


3.68 



366 



2. ANALOGS OF ENZYME REACTION COMPONENTS 

A. 




Acyl-L-phenyl- 
alanine (substrate) 





(3"'-^l 



L- Phenylalanine 



/^ X NH3 Vc 




r hV 



Benzoate 



\^ y CHj 



D- Phenylalanine 





Ql 




Phenylacetate 



Propionate 



Q- 




3 - Pheny Ipr opionat e 
(hydrocinnamate) 





Valerate 




y- Pheny Ibuty rate 3-Indoleacetate 

Fig. 2-9. Possible orientations of substrate and inhibitors at the active center of 
carboxypeptidase. (A) is peptidatic site, (B) is cationic site, and (C) is electrokinetic 
site. The molecular configurations are only approximate. 



CARBOXYPEPTIDASE, AMINOPEPTIDASES, DIPEPTIDASES 367 

a better inhibitor than L-phenylalanine under the usual conditions, since 
the NH3+ group of the latter is repelled by the positively charged pep- 
tidatic site while in the former it does not encounter serious steric in- 
terference (Fig. 2-9). However, /5-phenylpropionate (hydrocinnamate) is 
bound more tightly than D-phenylalanine by 2.14 kcal/mole, suggesting 
that some steric repulsion of the latter analog occurs. iV- Substitution in- 
creases the repulsion markedly and inhibitory activity is lost. A comparison 
of the possible orientations of some inhibitors in Fig. 2-9 with the relative 
binding energies may give some idea of the structural requirements for 
potent inhibition. It may be noted that benzylmalonate is an effective 
inhibitor but is somewhat less well bound than /5-phenylpropionate; this is 
surprising because it might be thought that additional energy would be 
contributed by interaction of one of the C00~ groups with the peptidatic 
site. Neither cis- nor <rans-cinnamate inhibits and it was suggested that the 
double bond restricts the orientation of the ring so that adequate binding 
cannot occur. The linearity of these molecules may also be a factor, since 
the active site is probably not flat as is implied by the two-dimensional 
representations in the figure. 

Competitive inhibition by the following analogs has been demonstrated 
more recently: 3-indolepropionate, e-aminocaproate, (5-amino-w-valerate 
(Folk, 1956; Greenbaum and Sherman, 1962), y-aminobutyrate, S-guani- 
dinovalerate, argininate (Folk and Gladner, 1958), iV-acetyl-L-tyrosine, 
D-leucinyl-L-tyrosine, glycyl-L-tyrosine, and other dipeptides (Yanari and 
Mitz, 1957). The inhibition apparently sometimes depends on the substrate 
used; for example, 3-indolepropionate inhibits the hydrolysis of carbo- 
benzoxyglycyl-L-phenylalanine but not the hydrolysis of a-iV-benzoyl- 
glycyl-L-lysine, where s-aminocaproate exhibits just the opposite behavior. 

Relatively little work has been done on analog inhibition of dipepti- 
dases and aminopeptidases, and it will suffice to mention a few isolated 
observations. Yeast dipeptidase is inhibited by various amino acids; for 





Concentration for 


Amino acid 


50% inhibition 




(mM) 


L-Leucine 


1.5 


L-Isoleucine 


1.8 


L-Tryptophan 


6.0 


L-Histidine 


8.0 


L-Leucinamide 


9.0 


L-Arginine 


18 


DL- Valine 


22 


DL-Phenylalanine 


22 


DL-Serine 


>100 



368 2. ANALOGS OF ENZYME REACTION COMPONENTS 

example, L-leucine, D-alanine, and glycine inhibit the hydrolysis of alanyl- 
glycine and glycylglycine (Grassmann et al., 1935). This has been investi- 
gated more quantitatively by Nishi (1960) and his results are summarized 
in the accompanying tabulation (substrate is glycylglycine at 50 mM). 
These inhibitions are competitive with respect to substrate and uncompe- 
titive with respect to Co++. Some interesting inhibitions of pig kidney 
leucine aminopeptidase have been reported by Hill and Smith (1957). 
The hydrolysis of substrates of the type R— CH(NH3+)— CONH— R' 
depends on a three-point attachment of the R group, the NH3+ group, 
and the amide N. The R groups interact by van der Waals' forces; further 
energy is contributed from the hydrogen bonding of water molecules dis- 
placed from the hydrophobic surfaces. The inhibitions given in the follow- 
ing tabulation are for L-leucinamide as substrate at 50 mM and at pH 8.50- 



Inhibitor 


Concentration 
{mM) 


% Inhibition 


L-Leucine 


50 


45 




100 


61 


L-Leucinol 


50 


38 




100 


48 


Isocaproamide 


25 


41 




50 


56 


Isocaproate 


100 


47 




200 


65 


w-Hexylamine 


100 





a-Ketoisocaproamide 


25 


21 


L-a-Hydroxyisocaproamide 


50 


68 



8.65. Every good inhibitor contains an R group that should give nearly 
optimal interaction with the electrokinetic site of the enzyme; in addition, 
at least one group that will react with the bound Mn++ is present. 



CHYMOTRYPSIN AND OTHER PROTEOLYTIC ENZYMES 

Chymotrypsin hydrolyzes various amides and esters of the general type: 

NH— R2 
Rj — CH2 CH 

bo— R3 

where R^ represents the side chains of amino acids (phenylalanine, tyrosine, 
and tryptophan most commonly used), Rg is an acyl group (acetyl, benzoyl, 



CHYMOTRYPSIN AND OTHER PROTEOLYTIC ENZYMES 369 

or nicotinyl), and Eg is a group forming either an amide or ester bond. 
A typical substrate is benzoyl-L-tyrosinamide: 



NH— CO 
KO—(. />— CH,— CH 

\\ // Vo-NH 



o 



In addition, the NH — R2 chain may be replaced by H, CI, or OH groups. 
Only the derivatives of L-amino acids are hydrolyzed. The R^ and Rg 
groups are important in binding to the enzyme and thus, with the esteratic 
(paptidatic) site, one may again visualize a three-point attachment. Analogs 
either devoid of susceptible amide or ester bonds, or having in their 
place bonds resistent to hydrolysis, are often inhibitory. The R^ group is the 
most important for binding, as is shown by the strong inhibitory activity 
of /5-phenylpropionate (hydrocinnamate) (Kaufman and Neurath, 1949). 
The necessity for at least one aromatic ring in one of the side chains was 
pointed out by Neurath and Gladner (1951). Their data on the /^-substituted 
propionates indicate the ring groups to have the following order of inhibitory 
activity: 

Indole > napthyl > phenyl > 2,4-dinitrophenyl > cyclohexyl 

The distance between the COO" group and the Rj group is also of impor- 
tance. The inhibitions are summarized in Table 2-19. The weaker binding 
of cyclohexyl derivatives compared to phenyl compounds (0.5-1 kcal/mole) 
could be explained by either the smaller polarizability of the cyclohexyl 
ring or the inability of the cyclohexyl ring to approach the enzyme surface 
as close as the phenyl ring. The strong binding of the indole compounds 
was explained on the basis of the enhancement of hydrogen bonding by the 
ring N. In fact, Neurath and Gladner interpreted the inhibitions by most 
of the analogs in terms of hydrogen bonding. Even the C00~ may not 
interact electrostatically with an enzyme cationic group since the binding 
energies are quite low; indeed, it may serve as a hydrogen acceptor. The 
equivalent bindings of 



. /y — CH2CH2— COO and (^ /)— O-CH^CH,— OH 

l3-Phenylpropionate 2-Phenoxyethanol 

would be difficult to explain otherwise; however, the latter compound can 
act as a hydrogen donor in forming a hydrogen bond. These two com- 
pounds have essentially the same molecular dimensions but the electronic 
configurations of the terminal groups differ. 



370 2. ANALOGS OF ENZYME REACTION COMPONENTS 

Table 2-19 

Competitive Inhibition of a-CnYMOXRYPSiN by Analogs " 



Inhibitor 



Apparent iC, Relative — AF oi binding 

(mM) (kcal/mole) 



3-Indolepropionate 2.5 3.57 

3-Indolebutyrate 3.6 3.35 

a-Naphthylpropionate 4.0 3.28 

2,4-Dinitro-^-phenylpropionate 5.3 3.08 

^-Phenylpropionate 5.5 3.07 

2-Phenoxyethanol 5.8 3.05 

y-Phenylbutyrate 14 2.53 

Cyclohexylpropionate 30 2.08 

Cyclohexylbutyrate 35 1.99 

Benzoate 42 1.88 

Phenylacetate 42 1 . 88 

a-Naphthylmethylmalonate 55 1 . 72 

a-Benzylmalondiamide 78.8 1.51 

Cyclohexylacetate 86 1.46 

" The substrate is acetyl-L-tyrosinamide (K^ = 27 mM). Experiments at pH 7.8 
and 25°. (Data from Neurath and Gladner, 1951.) 

For the past several years Niemann and his associates have conducted 
excellent quantitative investigations of the inhibition of chymotrypsin 
by a variety of analogs and their results, some of which are presented in 
Table 2-20, provide a basis for the interpretation of the binding mechanisms. 
Although final conclusions must await completion of their work, some tenta- 
tive ideas may be expressed. 

(A) Despite the fact that the derivatives of L-amino acids are hydro- 
lyzed by chymotrypsin, the D-isomers of several inhibitors are bound on an 
average of 0.45 kcal/mole more tightly than the corresponding L-isomers 
(see also page 271). Although three-point attachment may be important for 
substrate binding, it is evident from this difference and other data that 
it is not for inhibitor binding. 

{B) Comparing the derivatives of the three amino acids, it is seen that 
the phenylalanine and tyrosine analogs are equally bound, whereas the 
tryptophan analogs are bound some 1.19 kcal/mole more tightly, and this is 
probably to be attributed to the greater affinity of the enzyme for the 



CHYMOTRYPSIN AND OTHER PROTEOLYTIC ENZYMES 



371 



Table 2-20 
Competitive Inhibition of q-Chymoteypsin by Analogs " 



Inhibitor 



Apparent K^ Relative — AF oi binding 
(mil/) (kcal/mole) 



Tryptophan series 
L-Tryptophanamide 
D-Tryptophanamide 
Acetyl-D-tryptophanamide 
Trifluoroacetyl-D-tryptophanamide 
Acetyl-L-tryptophanmethylamide 
Acetyl-D-tryptophanmethylamide 
Benzoyl-D-tryptophanamide 
Nicotinyl-D-tryptophanamide 
p-Methoxybenzoyl-D-tryptophanamide 
Acetyl-L - tryptophanate 
Acetyl-D -tryptophanate 
Acetyl-D-tryptophan isopropylester 
Tryptamine 
Acetyltryptamine 
Trifluoroacetyltryptamine 
Indole 

Indoleacetate 
Indolepropionate 
Indolebutyrate 
Indolepropionaraide 

Tyrosine series 

Acetyl-D-tyrosinamide 

Trifluoroacetyl-D-tyrosinamide 

Chloroacetyl-D-tyrosinamide 

Nicotinyl-D-tyrosinamide 

Acetyl-L-tyrosinemethylamide 

Formyl-L-tyrosinemethylamide 

Nicotinyl-L-tyrosinemethylamide 

Benzoyl-L-tyrosinemethylamide 

Acetyl-L-t>Tosinate 

Fluoroacetyl-L-tyrosinate 

Chloroacetyl-L-tyrosinate 

Acetyl-D-tyrosine ethyl ester 

Nicotinyl-D-t>Tosine ethylester 

D-Tyrosinehydroxamide 

Acetyl-D-tyrosinehydroxamide 



8.5 


2.93 


4.0 


3.40 


2.4 


3.72 


4.0 


3.40 


6.5 


3.10 


1.8 


3.89 


0.7 


4.46 


1.6 


3.96 


0.6 


4.55 


9.5 


2.87 


7.5 


3.01 


0.8 


4.39 


2.3 


3.74 


1.8 


3.89 


1.2 


4.13 


0.8 


4.34 


18 


2.47 


15 


2.58 


23 


2.32 


2.3 


3.74 


12 


2.72 


20 


2.40 


6.5 


3.10 


9 


2.90 


61 


1.72 


31 


2.14 


8.8 


2.92 


6.4 


3.12 


110 


1.36 


120 


1.30 


150 


1.16 


4.7 


3.30 


0.8 


4.39 


40 


1.98 


7.5 


3.01 



372 2. ANALOGS OF ENZYME REACTION COMPONENTS 

Table 2-20 {continued) 



Inhibitor 



Apparent K^ Relative — Zli^ of binding 
(mM) (kcal/mole) 



Phenylalanine series 

Acetyl-D-phenylalaninamide 12 2 . 72 

Nicotinyl-D-phenylalaninamide 9 2.90 

Acetyl-D-phenylalanine methylester 2.3 3.74 

Benzoate 150 1.16 

Benzamide 10 2 . 83 

Phenylacetate 200 0.99 

Phenylacetamide 15 2.58 

Phenylpropionate 25 2 . 26 

Phenylpropionamide 7.0 3 . 05 

Phenylbutyrate 60 1.73 

Phenylbutyramide 12 2.72 

" The values of K^ are taken from work at pH 7.9. A", varies with the pH and thus 
the degree of ionization may be of importance in some instances. The relative binding 
energies are therefore subject to some error but may provide an initial basis for discus- 
sion. Various substrates have been used in the different studies and this may introduce 
further uncertainties. (Data from Foster et al., 1955; Foster and Niemann, 1955 a, 
b; Lands and Niemann, 1959.) 

indole ring system. This is confirmed by comparing the phenyl and indole 
acids and amides, the indole derivatives being bound 0.77 kcal/mole more 
strongly on the average. 

(C) Comparing the iV-substituted Rg groups one may calculate that the 
order of binding is: 

A = 0.35 A = 0.68 

(kcal/mole) (kcal/mole) 

Benzoyl > nicotinyl > acetyl 

and it appears that the acetyl derivatives are better inhibitors than the 
analogs with a free NH3+ group. It is likely that these groups interact 
with the enzyme surface in a nonspecific fashion. 

(D) Analogs with a CONH2 group are bound around 1.10 kcal/mole 
more tightly than those with a free C00~ group. This might indicate that 
the peptidatic site is in an electric field arising from surrounding negative 
charges, but it could also mean that hydrogen bonds between the pep- 
tide linkage and the enzyme are important. The rather tight binding of 



CHYMOTRYPSIN AND OTHER PROTEOLYTIC ENZYMES 373 

tryptaraine indicates also a more favorable field for positive than negative 
ionic groups, but the positive charge is not important since acetyltrypt- 
amine is bound even more readily. 

Foster and Niemann (1955 a) determined the values of K^ for several 
inhibitors at pH 6.9 and 7.9. The acetyltryptophanates and indolepropio- 
nate are bound 1.0-1.4 kcal/mole more tightly at pH 6.9 than at pH 7.9, 
whereas the binding of various amides is not significantly affected. This is 
interpreted as pointing to the development of a negative charge in the vicin- 
ity of the active site as the pH rises from 6.9 to 7.9; this charge would repel 
the negatively charged tryptophanates and other carboxylates. It is pos- 
sible that the lack of effect of pH on amide binding is due to the simulta- 
neous deprotonation of the CONH3+ with rise in pH (p^^ = 7.5). The con- 
cept of a negative charge on or near the active center was first postulated 
by Neurath and Schwert (1960) on the basis of the suppressing effect of a 
carboxylate group on the hydrolysis of an adjacent peptide bond. Whether 
this enzyme negative group participates in the hydrolysis along the lines 
suggested by Steam (1935) and Vaslow and Doherty (1953) is not certain. 

The nitrogen analogs of substrates are usually not hydrolyzed and can 
act as inhibitors (Kurtz and Niemann, 1961). In these compounds an a- 
methine group is replaced by a N atom, or an or-methylene group by an 
NH group. Thus ethyl l-acetyl-2-benzylcarbazate is an analog of ethyl 

^COO— Et 

CH2— N 

NH- COCH3 

acetyl-L-phenylalaninate and is an inhibitor, with K^ = 20 raM. Also 
9)— CH2CH2— COO— CH3 and (7^— NH— CH2— COO— CH3 are substrates of 
chymotrypsin whereas cp — CHg — NH — COO — C2H5 is an inhibitor [K^ = 
6 m.M). It is suggested that an a-N= or a-NH — group leads to a restriction 
of the rotation around the bond joining it to the CO group compared to 
that of a C — C bond, this constraint leading to loss of substrate activity or 
weakening of inhibitor binding. 

Wallace et al. (1963) published the results of their studies of the interac- 
tions of 136 compounds with chymotrypsin, and it is seen that several 
substances not directly related to amino acids are fairly potent inhibitors; 
such are quinoline and the methylquinolines, hydroxyquinolines, and ami- 
noquinolines, various acridines, a-naphthol, and the naphthylamines, all 
with K^ values less than 1 mM . They summarized their conclusions with re- 
spect to structure and inhibitory activity in ten postulates, from which the 
following comments are extracted. Aromatic compounds are more effective 
than the corresponding saturated derivatives, and monosubstituted benzene 
derivatives with polarizable uncharged groups are more inhibitory than the 



374 2. ANALOGS OF ENZYME REACTION COMPONENTS 

parent compounds; both facts indicate the importance of polarization of 
the inhibitor molecule in the field of ionic groups on the enzyme. The ac- 
tive site has a locus for interaction with the aromatic nucleus, and vicinal 
to this there is at least one anionic group; the orientation of the inhibitor 
is determined by the interaction of the polarizable group with a sublocus. 
Molecules presenting a larger planar area are more inhibitory, and it seems 
that the site at which the aromatic compounds act is mainly flat, of greater 
length than breadth, and not straight but curved along the enzyme surface. 
The different loci involved in the interactions perhaps have different prop- 
erties; e. g., it was postulated that the active site may be hermaphroditic* 
in that one locus may be electron-deficient and another electron-rich. 

The chymotrypsin reaction proceeds in two steps, the acetylation of the 
enzyme: 

EH + AcB z^ (EH— AcB) ^ (EAc-HB) ^ EAc + HB 

and the solvolysis of the acetylchymotrypsin: 

EAc + HOR ^ (EAc-HOR) ^ (EH— AcOR) 5=t EH + AcOR 

The competitive inhibitors tabulated here were shown by Foster (1961) 
to block the acetylation reaction and not the solvolysis. The inhibition by 
indole is not strictly competitive (Applewhite et al., 1958). 





A'i 


Relative — AF of binding 


Inhibitor 


(mi/) 


(kcal/mole) 


Skatole 


0.5 


4.69 


Indole 


0.7 


4.43 


p-Nitrobenzoyl - d - try ptophan 


1.5 


4.01 


Tryptamine 


2.0 


3.84 


p-Nitrobenzoyl-L-tryptophan 


2.3 


3.74 


Acetyl-D-tryptophan 


4.0 


3.41 


Acetyl-L-tryptophan 


6.0 


3.16 


D-Tryptophan 


10 


2.84 


L-Tryptophan 


20 


2.41 



The relation of chymotrypsin or a chymotrypsin-like enzyme to hista- 
mine release and the anaphylactic reaction was examined by Austen and 
Brocklehurst (1960) in guinea pig lung sensitized to ovalbumin. Inhibitors 
such as /5-phenylpropionate, indoleacetate, and indolepropionate depress 
anaphylactic histamine release more than 50% at 2.5 mM, and indole and 
skatole are effective at 1 mM or below. Antigen apparently activates some 
proteolytic enzyme necessary for the release of histamine. 

* This terminology connotes an entirely new way of looking at enzyme catalysis. 



CHYMOTRYPSIN AND OTHER PROTEOLYTIC ENZYMES 375 

The hydrolysis of a-benzoyl-L-argininamide by papain is inhibited com- 
petitively by carbobenzoxy-L-glutamate (Kimmel and Smith, 1954; Stockell 
and Smith, 1957). The inhibition is very sensitive to pH between 3.9 and 
5, the inhibitory activity almost disappearing at the upper end of this range 
(see Fig. 1-14-6). The y-carboxyl group has a pTiT,, of 4.4 so it is possible 
that the active form is un-ionized: 

coo" 

/ 
HOOC— CH2CH2— CH /^^ 

Vh^— COO-CH2 — (\ /) 

It is rather surprising that the ionization of this group should be so im- 
portant in the binding, especially as it is at some distance from the other 
probable binding groups, and the enzyme as a whole is quite positively 
charged at these pH's (isoelectric point near 8.75). Perhaps a hydrogen- 
bonding function must be attributed to the COOJI group instead. Benzoyl- 
L-arginine {K^ = 60 mM) and benzoyl-L-ar'gininamide (iiC, = 54 mM) 
also inhibit the hydrolysis of benzoyl-L-arginine ethyl ester (K,,, =23 mM) 
by ficin at pH 5.5 (Bernhard and Gutfreund, 1956). 

Beef spleen cathepsin C is competitively inhibited by various amides, 
esters, and dipeptides when glycyl-L-tyrosinamide is the substrate (Fruton 
and Mycek, 1956). L-Phenylalanine amides and esters are good inhibitors 
(as shown in the accompanying tabulation) but L-phenylalanine, acetyl-L- 



Inhibitor 


(mM) 


L-Phenylalaninamide 


8.3 


L-Phenylalanine ethyl ester 


14 


L-Phenylalanyl-L-phenylalanine 


17 


L-Tyrosinamide 


22 


DL-Phenylalanylglycine 


25 


D-Phenylalaninamide 


68 



phenylalanine, and glycyl-DL-phenylalanine are inactive. ^-Acetylation or 
a free C00~ group seems to prevent binding. The interaction between the 
inhibitors and the enzyme may involve several groups with possible hydro- 
gen bonding of the carbonyl group in a manner similar to that proposed for 
chymotrypsin. Decarboxylation and A^-acylation of amino acids can lead 
to inhibitors, as in the competitive inhibition by tosylagmatine of thrombin 
(Ki = 13.2 mM) and trypsin {K, = 3.45 mM) (Lorand and Rule, 1961). 
This inhibitor markedly slows clotting in a thrombin-fibrinogen system and 
in whole plasma. 



376 2. ANALOGS OF ENZYME KEACTION COMPONENTS 

HEXOKINASES 

We shall now turn to various aspects of carbohydrate metabolism and 
begin with those enzymes responsible for the initial phosphorylation of 
sugars. Hexokinases catalyze the reaction between two types of substrate 
— hexoses and ATP — and analogs of either may inhibit. Discussion of cer- 
tain particularly important glucose analogs (2-deoxy-D-glucose, 6-deoxy-6- 
fluoro-D-glucose, and related compounds), whose actions may involve not 
only hexokinases but other early steps in carbohydrate metabolism, will 
be postponed to the following section. The remaining analog inhibitors may 
be classed as (A) hexoses, (B) hexose phosphates, (C) glucosamine and deri- 
vatives, and (D) nucleotides and polyphosphates. It should be remembered 
that the values of K„, and K^ may often be composite in that the sugar or 
its derivative may occur in solution in a variety of forms (for example, a- 
or /^-isomers, or pyranose or furanose rings). These forms may have quite 
different activities and different individual constants. 

Inhibition by Hexoses 

Most hexokinases are not specific for the phosphorylation of one sugar 
but act at varying rates with different hexoses. Thus yeast hexokinase 
phosphorylates glucose {K„i = 0.16 mM), fructose {K„j = 1.7 mM), and 
mannose (K,,, = 0.1 mM); brain hexokinase is very similar (Slein et al., 
1950). If one active site on the enzyme is responsible, competition between 
the substrates should be observed. At equimolar concentrations, glucose 
and mannose almost completely inhibit the phosphorylation of fructose 
(88-98%), whereas fructose inhibits the phosphorylation of the former 
hexoses very little (15-18%), which would be expected on the basis of their 
relative iiC,„'s. The calculated K/s are essentially the same as the J^,„'s. 
The inhibition of Schistosoma fructokinase* by glucose and mannose is also 
quite marked, whereas galactose is much less potent (Bueding and Mac- 
Kinnon, 1955). The fructokinase of rat intestinal mucosa shows similar 
specificity, the following inhibitions being observed with 6 niM inhibitor 
when fructose is 10 mM: glucose (90%), mannose (80%), mannoheptulose 
(65%), xylose (50%), allose (0%), and galactose (0%) (Sols, 1956). Ascites 
cell glucokinase is inhibited rather well by talose (jK", = 3.2 mM) and altrose 
{K^ = 6 mM), but not by allose, gulose, or idose (Lange and Kohn, 1961 b). 
Yeast glucokinase is competitively inhibited by mannose {K^ = 0.11 mM) 
(Fromm and Zewe, 1962). Mannoheptulose induces a temporary diabetic 
state by blocking the uptake of glucose into the tissues through inhibition 
of glucokinase (Coore and Randle, 1964). 

* The terms "glucokinase," "fructokinase," etc., will be used to designate kexo- 
kinases when the corresponding substrates are used without implying that the en- 
zymes are specific for these substrates. 



HEXOKINASES 377 

Such inhibitions may be of importance in the metabolism of mixtures 
of sugars. Fructose is usually phosphorylated more rapidly than glucose, 
but in mixtures of the two the phosphorylation of fructose is markedly 
suppressed and essentially only glucose is metabolized. Although little 
is known of the nature of the binding of these hexoses to the enzyme, it 
would appear that the configurations at C-3 and C-4 are particularly im- 
portant; for example, allose differs from glucose only at C-3 and is not 
readily bound, and galactose differs from glucose only at C-4 and is much 
less bound. 

Inhibition by Hexose Phosphates 

Three hexose phosphates have been found to be fairly specific and inter- 
esting inhibitors of hexokinases: these are a-D-glucose-6-P, a-L-sorbose-1-P, 
and l,5-anhydro-D-glucitol-6-P. Inhibiting hexose phosphates should be 
visualized in their pyranose or furanose forms, since it is likely that in- 
teraction with the enzyme occurs with one side of these ring structures. 
(See formulas on page 378). 

Glyceraldehyde has been known for many years to be an inhibitor of 
glycolysis and Lardy et al. (1950) in a study to determine the site of action 
found that L-glyceraldehyde prevents the phosphorylation of glucose or 
fructose in brain extracts, but yet has no direct effect on the hexokinase. 
This paradox was resolved by showing that aldolase catalyzes the conden- 
sation of L-glyceraldehyde with glyceraldehyde-3-P to give a mixture of 
D-fructose-1-P and L-sorbose-1-P. The latter compound was found to be a 
potent inhibitor of hexokinase (for example, 0.08 mM inhibits 67% the 
phosphorylation of glucose at 35 mM), whereas L-sorbose and L-sorbose-6-P 
are inactive. The conventionally written structure for L-sorbose- 1-P 
(as above) bears little obvious resemblance to D-glucose-6-P, but if the 
former structure is inverted it is seen that the molecules are identical from 
one side and differ only by the transposition of a hydroxyl group, as illus- 
trated by Lardy et al. with molecular models. D-Glucose-6-P is the product 

CH3— O— PO'a' 
O 



HO^ 



Q- L-Sorbopyranose- 1-P 

of the hexokinase phosphorylation of glucose and is an inhibitor of the 
reaction (see below). It is surprising that the inhibition by L-sorbose-1-P 
is not competitive with glucose; indeed, as the glucose concentration is 
increased, the inhibition becomes somewhat greater. It is possible that 



378 



2. ANALOGS OF ENZYME REACTION COMPONENTS 




















7^ 




/ 


"^o^ 

K 


pK 


\ 


^^^ 


— O 


u- 

i 


\^ 







HEXOKINASES 



379 



competition might be seen if rates of inhibition in the presence of various 
concentrations of glucose were determined, but once L-sorbose-1-P has 
combined with the enzyme the inhibition is essentially irreversible. It is 
possible that glucose forms an intermediary tightly bound complex with 
the enzyme- ATP during the reaction and that L-sorbose-1-P forms a similar 
complex that is stable. Actually the affinities of the brain enzyme for 
D-glucose-6-P (Kj = 0.4 mM) and L-sorbose-1-P {K^ = 0.7 mM) are close 
(Crane and Sols, 1954). Other hexokinases may not be so susceptible to 
L-sorbose-1-P, since Taylor (1960) found only slight inhibition by 0.5 mM of 
glucose uptake by Scenedesmus, the primary transfer site being hexokinase 
on the outside of the membrane. 

A closely related inhibitor is l,5-anhydro-D-glucitol-6-P (1,5-D-sorbitan- 
6-P), this lacking the 2-OH group in L-sorbose-1-P and binding somewhat 
less tightly {K^ = 1 mM) to the brain hexokinase (Crane and Sols, 1954). 
The nonphosphorylated compound is a very weak inhibitor (Sols, 1956). 
Since there are very few potent and specific hexokinase inhibitors, Ferrari 
et al. (1959) have recently investigated l,5-anhydro-D-glucitol-6-P in some 
detail to see if it might be useful as a blocking agent of this enzyme in ho- 
mogenates. It is stable to the enzymes attacking D-glucose-6-P, except for 
hydrolysis by liver glucose-6-phosphatase. No inhibition of glucose-6-P de- 
hydrogenase is evident, but it inhibits phosphoglucomutase variably (de- 
pending on the concentrations of glucose-1-P, glucose-l,6-diP, and Mg++) 
and phosphoglucose isomerase noncompetitively at higher concentrations. 
At 6.25 mM it blocks glucose respiration in heart homogenates but has 
no influence on the oxidation of glucose-6-P, indicating under these con- 
ditions a rather specific inhibition of hexokinase. 

Brain hexokinase is inhibited by glucose-6-P whereas yeast hexokinase 
is not (L-sorbose-1-P also does not inhibit the yeast enzyme), and the inhi- 
bition has been found to be noncompetitive with respect to both glucose 
and ATP (Weil-Malherbe and Bone, 1951). Inhibitions by various hexose 
phosphates have been studied thorougly by Crane and Sols (1953, 1954); 
the accompanying tabulation summarizes their data. The following are 
noninhibitory: /?-D-glucose-l,6-diP, D-mannose-6-P, D-fructose-6-P, D-fruc- 
tose-l,6diP, D-arabinose-5-P, D-ribose-5-P, D-galactose-6-P, a-glucose-l-P, 



Inhibitor 


imM) 


Relative 


! — JjP of binding 
(kcal/mole) 


a-D-Glucose-6-P 


0.4 




4.80 


a-D-Glucose-l,6-diP 


0.7 




4.46 


a-L-Sorbose-1-P 


0.7 




4.46 


l,5-Anhydro-D-glucitol-6-P 


1.0 




4.25 


a-D-Allose-6-P 


7.0 




3.05 



380 2. ANALOGS OF ENZYME REACTION COMPONENTS 

D-altrose-6-P, glucuronate, and glucuronate-6-P. It may be noted that glu- 
cose is the only hexokinase substrate that forms an inhibitory phosphate, 
indicating the importance of the configuration at C-2 for inhibition. Thus 
glucose phosphorylation in a closed system slows down progressively while 
that of mannose is linear, a phenomenon which may be significant in 
regulating the rate of sugar utilization. In a purer preparation of brain 
hexokinase. Crane and Sols confirmed that the inhibition is not formally 
competitive but that a reversible EI complex is formed. Phosphorylation 
at C-6 (or at C-1 in the sorbose structure) seems necessary for inhibition; 
e. g., glucose-1-P and 1,5-anhydro-D-glucitol lack inhibitory activity. It is 
interesting that the inversion of the phosphate-carrying group at C-1 to 
form /?-glucose-l,6-diP abolishes the inhibition, possibly due to a static 
interference of the now closely apposed phosphate groups. Inversion of the 
groups on C-2 (mannose-6-P), C-3 (allose-6-P), or C-4 (galactose-6-P) re- 
duces or abolishes the inhibition; it was felt by Crane and Sols that the 
configuration at C-3 influences the effect of an adjacent group and is not 
directly concerned in the binding. It is difficult in most cases to decide if 
the change in affinity on inversion of the groups is related to the hydroxyl 
group as a binding site or as producing steric hindrance; thus the lack 
of inhibition by galactose-6-P could be due either to the loss of hydrogen 
bonding through the OH group (occurring in glucose-6-P) or to a protru- 
sion of the OH group preventing approach of the pyranose ring. Compar- 
ison with the corresponding deoxyglucose-6-P's might be informative. We 
cannot do this for C-4, but at C-2 removal of the OH group (2-deoxy- 
glucose-6-P) abolishes inhibition, pointing to the OH group as a binding 
site. The weak inhibitory activity of 3-deoxyglucose-6-P substantiates the 
idea that the 3-OH group is not involved in binding. The retention of 
inhibition in l,5-anhydro-D-glucitol-6-P likewise indicates that the 1-OH is 
not a binding site, but the loss of inhibition on C-1 methylation (a-meth- 
ylglucoside-6-P) shows that steric repulsion occurs when the C-1 group 
becomes too large. One may conclude that binding sites are at the 2-OH 
and 6-phosphate groups, and possible at the 4-OH group. The inhibitors 
thus attach to a different set of enzyme sites than the substrates, only 
C-4 being common to both, and Crane and Sols visualized these differences 
in the following structures, where the solid circles indicate necessary binding 
positions: 

► H,OH •H2— O— PO",' 

H )r — \ 

OH 



Substrate Inhibitor 





HEXOKINASES 



381 



Rat intestinal mucosa hexokinase is inhibited by glucose-6-P but only 
about one-tenth as readily as the brain enzyme (Sols, 1956). The hexokinase 
of Schistosoma is strongly inhibited by glucose-6-P when glucose or mannose 
is the substrate, but fructose phosphorylation is unaffected (Bueding and 
MacKinnon, 1955). Ascites tumor hexokinase behaves like the brain enzyme 
and the K, for glucose-6-P is 0.4 mM (McComb and Yushok, 1959). Thus 
inhibition of various hexokinases by glucose-6-P has been observed, but the 
original observation that the yeast enzyme is resistant cannot as yet be 
explained. 

Inhibition by D-Glucosamine and Derivatives 

Glucosamine (2-amino-D-glucose) is phosphorylated by brain hexokinase 
(Harpur and Quastel 1949) and it was postulated by Quastel and Cantero 
(1953) that it might be carcinostatic through ATP depletion. However, it 
also competitively inhibits glucose phosphorylation and any carcinostatic 
activity it possesses would be more likely related to this. Maley and Lardy 
(1955) thus attempted to find a more potent inhibitor among the A^-acylated 
derivatives and were quite successful, as shown in the accompanying tabu- 
lation. Furthermore, these derivatives are not phosphorylated. 

Ki (mM) 





Glucokinase 


Fructokinase 


iV^-(3,5-Dinitrobenzoyl)- 


0.011 


0.004 


A''-(m-Nitrobenzoyl)- 


0.033 


0.0084 


iV-(p-Nitrobenzoyl)- 


0.04 


0.05 


iV-Benzoyl- 


— 


0.036 


iV-( w? - Aminobenzoyl)- 


0.15 


0.081 


i\/^-(p-Aminobenzoyl)- 


0.2 


0.11 


iV-Acetyl 


— 


0.46 


iV-Phenylacetyl- 


— 


0.86 



CH2OH 




nA 

/H 
\OH 
HON 


\ 

H 

1/ 



OH 



NH, 



CHoOH 




NH— CO 



O 



Q-D-Glucosamine 



iV- Benzoyl- o-D -glucosamine 



Before considering the nature of this inhibition, let us examine other 
hexokinases to determine how widespread is the susceptibility. The fructo- 



382 2. ANALOGS OF ENZYME REACTION COMPONENTS 

kinases of Schistosoma (Bueding and MacKinnon, 1955) and rat intestinal 
mucosa (Sols, 1956) and the glucokinase of Spirochaeta recurrentis (P. J. C. 
Smith, 1960 b) are moderately sensitive to glucosamine (50-65% inhibition 
by 6-10 mM) and more sensitive to A^-acetylglucosamine (75% inhibition 
by 1-2 mM). The iC, for iV-acetylglucosamine and the glucokinase of ascites 
tumor cells is 0.074 mM (McComb and Yushok, 1959), indicating a binding 
about 1 kcal/mole tighter than for glucose-6-P. Furthermore, both glucos- 
amine and iV-acetylglucosamine inhibit the metabolism of glucose-C^* and 
fructose-C^* to glycogen and COg in rat liver slices (Spiro, 1958), and the 
A'^-(2?-nitrobenzoyl) and A^-(3,5-dinitrobenzoyl) derivatives inhabit glucose 
uptake by Scenedesmus (Taylor, 1960). The phosphorylation of glucosamine 
in liver extracts is competitively inhibited by glucose {K^ =0.11 mM), 
fructose, A'^-acetylglucosamine, and hexose and glucosamine phosphates, 
illustrating mutual interference by these substrates and products (McGar- 
rahan and Maley, 1962). 

In all these instances the inhibitions are strictly competitive with glucose 
or fructose. The question arises as to why the iV-acylated derivatives are 
not phosphorylated. Maley and Lardy (1955) showed by molecular models 
that the iV-acyl groups do not overlap the 6-position so that some other 
explanation must be sought. It was suggested that the A^-acyl groups might 
interfere with the binding of ATP to the enzyme, but it is also possible that 
they shift the position of the pyranose or furanose rings sufficiently so 
that the 6-position is not favorably oriented for phosphorylation. It may be 
mentioned that no carcinostatic activity was noted with any of these sub- 
stances when tested in sarcoma-bearing mice, perhaps due to the hydrolysis 
of the iV-acyl compounds by tissue cathepsins. 

Kono and Quastel (1962) confirmed the glucosamine inhibition of glyco- 
gen formation in rat liver slices (50% inhibition by around 0.8 mM) and 
showed there to be no depression of the entry of glucose into the cells. Hexo- 
kinase, phosphoglucomutase, and UDP-glucose pyrophosphorylase are in- 
hibited quite weakly by glucosamine, significant effects being exerted only 
at concentrations over 20 mM. UDP-glucose-glycogen glucosy transferase 
in inhibited by glucosamine but not by iV-acetylglucosamine, which inhi- 
bits glycogen synthesis as does glucosamine. The isolated phosphorylase 
is also resistant to glucosamine. Thus the enzymes involved in glycogen 
formation are not directly inhibited by glucosamine and iV-acetylglucosa- 
mine. However, each of these substances stimulates phosphorylase activity 
in slices when added with glucose. Thus the explanation for the reduced 
glycogen formation may be an acceleration of glycogen breakdown and not 
a true inhibition. Silverman (1963) found a significant reduction of glucose 
oxidation by glucosamine in the rat epididymal fat pad in the presence of 
insulin and felt that an inhibition of glucokinase could not entirely account 
for the results, so that one must assume some action from nonmetabolized 



HEXOKINASES 383 

products formed from glucosamine, possibly glucosamine-P. Glucosamine 
depresses the respiration and oxidation of pjTuvate in ascites cells, as 
does glucose (Crabtree effect), and this is relieved by 2,4-dinitrophenol 
(Ram et al., 1963). Since this is presumably related to the phosphorylation 
of glucosamine and the effects on ADP-ATP levels, it constitutes another 
mechanism whereby glucosamine can alter carbohydrate metabolism. 

Inhibition by Adenine Nucleotides and Polyphosphates 

The reaction rate of hexokinases falls with time due to the accumulation 
not only of glucose-6-P but also of ADP (Sols and Crane, 1954). Phosphate, 
pyrophosphate, and AMP do not inhibit. The nature of the ADP inhibition 
appears to vary with the source of the hexokinase. With yeast hexokinase 
the inhibition is noncompetitive with respect to ATP since around 50% 
inhibition is produced by 0.5 mM ADP at all levels of ATP used (Gamble 
and Najjar, 1955), and with Schistosoma glucokinase the inhibition actually 
increases slightly with ATP concentration (Bueding and MacKinnon, 1955). 
The inhibition of liver fructokinase by ADP is reduced slightly by increasing 
the ATP concentration from 5 to 10 mM, but not enough to indicate pure 
competitive inhibition; Parks et al. (1957) stated it is noncompetitive but it 
might better be designated as mixed. Echinococcus fructokinase, on the 
other hand, is inhibited competitively (Agosin and Aravena, 1959). 

Tripolyphosphate (PgOjo^") inhibits the fermentation of glucose by in- 
tact yeast cells, glycolysis in cell-free extracts, and pure hexokinase (Vish- 
niac, 1950). The inhibition of hexokinase is quite potent when (ATP) = 3.75 
mM: 13% at 0.47 mM, 31% at 1.4 mM, 74% at 4.7 mM, and 93% at 14 
mM tripolyphosphate. The inhibition is reversed by both ATP and Mg++. 
Wheat germ hexokinase appears to be more resistant to tripolyphosphate, 
only 8% inhibition being given by 5 mM (Saltman, 1953). 

Inhibition by Giucosone 

D-Glucosone may be formed from D-glucose by mild oxidation (e. g. 
with Cu++) at C-2 but, although it has been known for over 75 years, its 
structure is not completely understood (Becker and May, 1949). The 
following forms are possible and it is difficult to choose between them: 

CH,OH 

O 




a) 




384 2. ANALOGS OF ENZYME KEACTION COMPONENTS 





OH H OH 

an) (IV) 

There is some evidence that the enolic tautomer II is not important and the 
behavior with enzymes might favor structure I. Hynd (1927) at St. Andrews 
tested D-glucosone to determine if it could counteract insulin hypoglycemic 
convulsions, as glucose does, but found that, if anything, the effect of 
insulin is increased. Glucosone was then administered to normal mice giving 
toxic symptoms within 5 min and a well-developed insulin-like reaction in 
20 min. The lethal dose range is very narrow, 2.4 mg/kg being nonlethal and 
2.6 mg/kg generally lethal. Glucose injected before or with the glucosone 
reduces the effects somewhat. Moribund mice following lethal doses show 
an elevation in blood glucose from 0.161 to around 0.240 mg%; thus the 
symptoms are not due to a hypoglycemia. Although these results would 
point to glucosone interference with the utilization of glucose, Hynd un- 
fortunately assumed that glucosone is formed from glucose by the action 
of insulin and that, indeed, it is responsible for the effects of insulin, the 
raised blood glucose levels being unexplained. Similar reactions to glucosone 
are seen in several species (Herring and Hynd, 1928). The theory that insulin 
induces glucosone formation was made untenable by Dixon and Harrison 
(1932), who found no glucosone in the blood during insulin convulsions. 

The problem rested at this stage for 20 years and then was taken up at St. 
Andrews (Bayne, 1952; Mitchell and Bayne, 1952; Johnstone and Mitchell, 
1953), but the results were published in a series of short and incomplete 
communications without adequate data. D-Glucosone effects in mice were 
not seen with up to 10 mg/kg of any other osone, including D-galactosone, 
D-arabinosone, D-xylosone, L-glucosone, and 3-methyl-D-glucosone. Turning 
to yeast glucose fermentation, they found no inhibition by 50 mM D-gluco- 
sone but marked inhibition at 200 mM, whereas L-glucosone has no effect 
at 200 mM. Becker (1954) reported almost complete inhibition of the aero- 
bic and anaerobic utilization of glucose by yeast when (glucosone)/(glu- 
cose) = 5. 

It was realized finally by Eeg-Larsen and Laland (1954) in Oslo that the 
structural similarity of glucosone to glucose might allow the former to 
interfere with the utilization of the latter by blocking its phosphorylation. 
This was demonstrated with ox brain hexokinase; glucosone is not phospho- 
rylated but inhibits glucose phosphorylation 50% at 0.35 mM and 100% 
at 2.4 mM when glucose is 2.4 mM. They concluded that the inhibition is 
noncompetitive, but the small range of glucose concentrations used makes 



HEXOKINASES 385 

it impossible to determine the type of inhibition; actually some decrease 
in the inhibition with increasing glucose was observed. As would be expected, 
glucosone does not produce a Crabtree effect but blocks it (Yushok and 
Batt, 1957). The inhibition of glucose fermentation in yeast depends on the 
pH and the buffer system present (Hudson rnd Woodward, 1958). At pH 
6.5 in phosphate buffer an inhibition of 73% of the anaerobic fermentation 
of glucose was found at (glucosone) /(glucose) = 2, whereas no effect was 
found at pH 3.5-4.5. The fermentation and phosphorylation of fructose are 
inhibited more readily than with glucose, due to the higher ^,„ for fructose, 
and the inhibitions are completely competitive with K^ for glucosone around 
0.061 mM. Marked inhibition of anaerobic glycolysis in rat tissues by glu- 
cosone was noted, brain being much more sensitive than tumor tissue; in 
brain complete inhibition occurs with (glucosone)/(glucose) = 0.0067. The 
susceptibility of brain glycolysis to glucosone is certainly much greater than 
of any hexokinase studied and possibly there is an additional site of action. 
In any event, these results provide sufficient explanation for the central 
toxic actions of glucosone. Despite the fact that glucosone can be formed 
in certain organisms (e. g., molluscan crystalline styles and red algae) and 
can be metabolized in streptococci and mammals, it would appear that it 
is not an important substance in intermediary metabolism and is not gen- 
erally on the pathway for the synthesis of glucosamine (Becker and Day, 
1953; Topper and Lipton, 1953; Dorfman et at., 1955; Bean and Hassid, 
1956). There is thus no evidence that glucosone can participate in the reg- 
ulation of carbohydrate metabolism, but the high susceptibility of brain 
glycolysis suggests that one should withhold final judgment until the ab- 
sence of glucosone in the body under various conditions has been demon- 
strated. 

Inhibition of Phosphafructokinase by Cycle Intermediates 

The phosphofructokinase from sheep brain is quite potently inhibited by 
certain cycle intermediates (see accompanying tabulation) (Passonneau and 
Lowry, 1963). Although this is not strictly analog inhibition, it is worth 



Inhibitor 


Ki (mJ/) 


Citrate 


0.03 


cis-Aconitate 


0.1 


Isocitrate 


0.2 


Malate 


0.6 


Succinate 


1.5 


a-Ketoglutarate 


2.5 


Fumarate 


>10 



386 2. ANALOGS OF ENZYME REACTION COMPONENTS 

mentioning because of the implications such actions have for a feedback 
control of carbohydrate oxidation. A rise in the levels of the inhibitory- 
intermediates would reduce the rate of formation of pyruvate. The steady- 
state concentrations of the cycle intermediates are certainly high enough 
to inhibit significantly, but the problem of compartraentalization arises 
since it is generally assumed that the tricarboxylates particularly are 
mostly confined to the mitochondria. Whatever the significance of this 
type of inhibition, it emphasizes the importance of compartmentalization 
in regulatory control of metabolism, a factor which has not always been 
taken into account. 

EFFECTS OF 2-DEOXY-D-GLUCOSE 
ON CARBOHYDRATE METABOLISM 

An inhibitor capable of specifically blocking the glycolytic pathway 
would not only be a valuable tool in biochemical investigation but might 
play a role in the chemotherapy of certain neoplasms. Glucose analogs, 
especially those entering the pathway and forming inhibitory intermediates, 
would be the most likely candidates, and 2-deoxy-D-glucose (2-DG) is the 
most interesting and best understood substance of this type. The volume 
of literature during the past 10 years on this analog precludes a complete 
discussion and emphasis will be placed on the sites and mechanisms of the 
inhibition in the glycolytic pathway. 2-DG was first examined by Cramer 
and Woodward (1952) at the Franklin Institute in the course of searching 
for carcinostatic glucose analogs, and they found that it does indeed produce 
some regression of Walker carcinoma and terminates embryonic develop- 
ment in rats. 2-DG differs from glucose in the substitution of the 2-OH 
group by a hydrogen atom and may be represented in the pyranose form as: 




2 -Deoxy-D -glucose 

Absorption, Distribution, and Metabolism of 2-DG 

It will be well to discuss the uptake and fate of 2-DG in cells before turn- 
ing to the metabolic disturbance produced. 2-DG enters most cells readily; 
this may involve a phosphorylation at the membrane in some cases, but 
in others it is phosphorylated only after entry. Whatever the transport 
mechanism there is usually competition between 2-DG and glucose, the 



EFFECTS OF 2-DEOXY-D-GLUCOSE 387 

uptake of 2-DG being depressed progressively as the glucose concentration 
is increased in rat diaphragm (Nakada and Wick, 1956; Kipnis, 1958) 
and lymph node cells (Helmreich and Eisen, 1959). Nakada and Wick 
(1956) showed that insulin can double the rate of 2-DG uptake by dia- 
phragm, and Kipnis and Cori (1959, 1960) studied this in greater detail. In 
normal diaphragm the 2-DG taken up appears as 2-deoxy-D-glucose-6-phos- 
phate (2-DG-6-P), the rate of phosphorylation being apparently greater 
than the rate of penetration. Diabetic diaphragm takes up and phosphory- 
lates 2-DG at a reduced rate but 2-DG does not accumulate in the cells, 
indicating the penetration is still rate-limiting. Addition of insulin acceler- 
ates the uptake and some free 2-DG appears in the cells so that the phos- 
phorylation is not increased proportionately. Epinephrine does not in- 
fluence the penetration but slows phosphorylation of 2-DG. Transport of 
2-DG across the entire diaphragm is slow, being about one-fifth the rate 
for glucose and one-twenty-fifth the rate for galactose (Ungar and Psy- 
choyos, 1963). It is possible that it is trapped in the muscle as 2-DG-6-P 
since insulin depresses the transfer. The uptake of 2-DG by yeast in glucose- 
phosphate medium is 5-10 times faster aerobically than anaerobically; 
when glucose is omitted the aerobic uptake of 2-DG is not altered, but an- 
aerobically there is a loss of 2-DG from the cells, so that the uptake is de- 
pendent on aerobic processes and probably on the level of ATP since 2,4- 
dinitrophenol acts like anaerobiosis (Kiesow, 1959). Certain fungi, such as 
Neurospora crassa and Aspergillus oryzae, can grow with 2-DG as the sole 
source of carbon (Sols et al., 1960 b) but not as rapidly as with glucose; 
indeed, growth with glucose or other sugars is inhibited by 2-DG. E. coli 
will not grow with 2-DG as the only carbon source and there is some evi- 
dence that it does not penetrate into the cells (Gershanovich, 1963). The 
evidence for the lack of entrance is that glycolysis is not inhibited in intact 
cells where it is in extracts. 2-DG diffuses across the intestinal wall but is 
not actively transported as are glucose, 1-DG, and 3-DG, indicating the 
importance of the 2-position in transport (Wilson and Landau, 1960), 
nor does 2-DG have an effect on the short-circuit current through the in- 
testinal wall, such being associated with transport (Schultz and Zalusky, 
1964; Barry et al, 1964). 

There is some information on the disposal of 2-DG in intact animals. 
Injected into rabbits, it is rapidly distributed in the extracellular space 
and some enters the tissues in eviscerated and nephrectomized animals, the 
uptake being markedly stimulated by insulin (Wick et al., 1955). Since it 
produces a block of glucose uptake for at least 8 hr, it is evident that little 
2-DG is metabolized beyond the 2-DG-6-P stage in extrahepatic tissues. 
This is confirmed by the finding that little or no C^^Oa is expired following 
injection of 2-DG-Ci* (Wick et al, 1957). Blood levels of 2-DG are more 
consistent after subcutaneous injection than when it is given intraperito- 



388 



2. ANALOGS OF ENZYME REACTION COMPONENTS 



neally, due to erratic absorption from the peritoneum (Ball and Saunders, 
1958). After subcutaneous administration there is a blood peak at 15 min, 
after which there is a gradual fall over 6 hr. In human subjects infused 
intravenously with 50-200 mg/kg of 2-DG over 30-min periods, approxi- 
mately 30% is excreted in the urine (Landau et al., 1958). 

The pathways of 2-DG metabolism have not been completely worked out. 
The accompanying scheme shows some of the reactions encountered. The 
phosphorylation by hexokinase to 2-DG-6-P would seem to be the most 
important reaction, especially as 2-DG-6-P is not metabolized in most cells 
and tends to accumulate. HeLa cells can oxidize 2-DG-6-P but much more 
slowly than glucose-6-P (Barban and Schulze, 1961). Hexokinases for the 

deoxy-disaccharides 



2-deoxy-D-glucose 



6 -deoxy-D-glucono lactone 



^^ 2-deoxy-D-gluconate 

\ 

2-deoxy-D-glucose-6-P »- 2-deoxy-D-gluconate- 6- P 

i \ 

(oxidized) D-mannonate-6-P 

formation of 2-DG-6-P have been found in brain (Sols and Crane, 1953), 
kidney, intestine, liver (Lange and Kohn, 1961 a), skin (Brooks etal., 1959), 
diaphragm( Kipnis and Cori, 1959), HeLa cells (Barban and Schulze, 1961), 
ascites carcinoma cells (McComb and Yushok, 1959; Lange and Kohn, 1961 
b), and Neurospora crassa (Sols et al., 1960 b). The Michaelis constants for 
2-DG are usually higher than for glucose (see tabulation) but the rates of 





Hexokinase K^ {mM) 


Source 












Glucose 


2-DG 


Brain 


0.01 


0.024 


Ascites carcinoma 


0.04 


0.069 


Intestine 


0.065 


0.09 


Kidney 


0.048 


0.04 


Liver 


0.04 


0.09 



phosphorylation are often comparable. It is interesting that a strain of 
HeLa cells resistant to 2-DG has been obtained, and they are defective in 
hexokinase or contain a hexokinase inhibitor; the phosphorylation rates for 
2-DG, glucose, fructose, and mannose are all lower than normal (Barban, 
1961). Resistance is also associated with a 5- to 10-fold increase in alkaline 
phosphatase activity and this may partly account for the slower rate of 



EFFECTS OF 2-DEOXY-D-GLUCOSE 389 

accumulation of 2-DG-6-P (Barban, 1962 a, b). How much of the 2-DG is 
oxidized directly is generally unknown, but the glucose oxidase from 
As'pergillus niger oxidizes it fairly well: the relative rates of oxidation are 
glucose (100), 2-DG (20), 3-DG (1), 4-DG (2), 5-DG (0.05), and 6-DG (10) 
(Pazur and Kleppe, 1964). The direct oxidation of 2-DG is apparently 
catalyzed by a variety of enzymes, some of the notatin type (Sols and de 
la Fuente, 1957) and some of the glucose dehydrogenase type (Williams 
and Eagon, 1959). The further metabolism of 2-deoxy-D-gluconate probably 
varies with the tissue and has been shown in the scheme above for skin 
(Brooks et at., 1960). Other pathways for 2-DG metabolism may occur in 
plants, since Kocourek et al. (1963) have provided evidence for (1) /?- 
glucosidation probably on C-6, (2) oxidation on C-1 to form 2-deoxyhexo- 
nate lactone, and (3) epimerization to 2-deoxygalactose in tobacco plants 
taking up 2-DG through the roots. The last reaction involves three enzymes 
and the epimerization occurs in a complex with UDP. The abnormal di- 
saccharide, /5-D-fructofuranosyl-2-deoxy-D-glucose, has been insolated from 
the excised leaves of several plants following incubation with 2-DG (Barber, 
1959). It is possible that a number of abnormal polysaccharides containing 
2-DG will eventually be found. 

Effects of 2-DG and 2-DG-6-P on Glycolytic Enzymes 

2-DG inhibits the utilization of glucose and other sugars in many organ- 
isms and tissues, and we shall now attempt to localize the site of this 
inhibition in the early phases of glycolysis. We must consider not only 
2-DG but also its primary metabolic product, 2-DG-6-P, as inhibitors. 
The most likely sites for inhibition would be (1) 2-DG on hexokinases, or 
(2) 2-DG-6-P on phosphoglucose isomerase, 6-phosphofructokinase, or al- 
dolase with respect to glucose metabolism. Since 2-DG is phosphorylated 
about as well as glucose by hexokinases it is clear that some competitive 
inhibition would be observed under certain circumstances, a suggestion first 
made by Cramer and Woodward (1952). However, this would appear to 
be generally an unimportant factor in the over-all glycolytic inhibition, 
since 2-DG equimolar with glucose does not inhibit the glucokinase of ascites 
cells (Nirenberg and Hogg, 1958) and at 10 times the glucose concentration 
does not inhibit HeLa cell glucokinase (Barban and Schulze, 1961). The 
situation may be somewhat more complex in certain tissues, however, in- 
asmuch as rat liver contains two glucose-phosphorylating enzymes, called 
glucokinase and hexokinase (Walker and Rao, 1963). The sensitivities of 
these enzymes are quite different (see accompanying tabulation) and a 
kinetic analysis was made, the hexokinase being studied without interfer- 
ence by the glucokinase since fetal liver contains only the former. The 
effect of the various inhibitors, which are all competitive, varies in a com- 
plex fashion as the glucose concentration is changed because of the vary- 



390 2. ANALOGS OF ENZYME REACTION COMPONENTS 







Ki (mM) 




Inhibitor 


Rat liver 


Rat liver 


Guinea pig 




glucokinase 


hexokinase 


hexokinase 


2-DG 


14 


0.3 


0.6 


D-Glucosamine 


0.8 


0.3 


0.2 


iV-Acetyl-D-glucosamine 


0.5 


0.2 


0.3 


Glucose (K,„) 


10 


0.04 


0.03 



ing importance of each enzyme, and it was pointed out that it is very dif- 
ficult to assess the effects of such analogs in adult liver. 

Most of the emphasis recently has been placed on the inhibition of 
phosphoglucose isomerase by 2-DG-6-P. This inhibition is competitive on 
the enzyme from rat kidney (Wick et al., 1957), rat muscle (Ferrari et al., 
1959), and ascites cells (Nirenberg and Hogg, 1958). The inhibition is reas- 
onably potent (see tabulation for kidney enzyme) and it would be quite 



Glucose-6-P 2-DG-6-P 

{mM) (mM) 



% Inhibition 



I 0.5 9 

1 1 24 

0.5 1 79 



easy for the 2-DG-6-P concentration to become greater than the glucose-6-P 
concentration in cells, especially as the former is usually not metabolized 
and accumulates. The inhibition of the skeletal muscle enzyme is very simi- 
lar. Unfortunately the inhibitions of 6-phosphofructokinase and aldolase 
by 2-DG-6-P have not yet been adequately examined, so one is left with 
phosphoglucose isomerase as the site of the primary block, the conclusion 
of Wick et al. (1957). However, another possible site for inhibition is the 
membrane transport system for glucose, as suggested by the work of Kipnis 
and Cori (1959), and this might be by either 2-DG or 2-DG-6-P. Furthermore, 
Nirenberg and Hogg (1958) reported that the metabolism of fructose-1,6- 
diP is blocked by 2-DG-6-P and stated that some inhibition must occur 
after the phosphofructokinase step (which could be on aldolase). Other 
inhibitions on enzymes metabolizing glucose but not on the main glycolytic 
pathway may be mentioned. Rat liver microsomal glucose-6-phosphatase 
is inhibited weakly by 2-DG (Hass and Byrne, 1960), HeLa cell glucose-6-P 



EFFECTS OF 2-DEOXY-D-GLUCOSE 391 

dehydrogenase is inhibited noncompetitively by 2-DG-6-P (Barban and 
Schulze, 1961), 2-DG-6-P competitively inhibits the activation of rat liver 
glycogen synthetase by glucose-6-P (Steiner et al., 1961), and UDP6: 
a- l,4-glucan-o;-4-glucosy transferase from dog muscle is inhibited by 2-DG- 
6-P with Kj =1.3 mM (Rosell-Perez and Larner, 1964). The importance 
of these inhibitions in the interference produced by 2-DG on glucose me- 
tabolism is not understood. 

The block of fructose utilization by 2-DG may well present a different 
problem. Fructose phosphorylation is inhibited much more readily than 
glucose phosphorylation, presumably due to the lower affinity of the hexo- 
kinases for fructose (Sols, 1956; Nirenberg and Hogg, 1958; Barban and 
Schulze, 1961). The inhibition of phosphoglucose isomerase could not explain 
the suppression of fructose utilization inasmuch as the fructose pathway 
bypasses this step. In rat adipose tissue 2-DG has very little effect on the 
metabolism of fructose although glucose metabolism is quite strongly 
depressed (Fain, 1964). With glucose at 2.8 mM and 2-DG at 1.4 mM, the 
formation of COg is reduced 70% and of fatty acids 89%. Nevertheless, the 
stimulatory effect of insulin on fructose utilization is blocked by 2-DG 
whereas the effects of insulin on glucose are unaltered. Certainly different 
tissued and organisms must have various transport and enzyme systems for 
the metabolism of fructose, so one should not expect a uniform action of 
2-DG. The enzymes involved in the metabolism of fructose- 1-P or fruc- 
tose-6-P have not been studied with respect to 2-DG inhibition. 

Effects of 2-DG on Carbohydrate Metabolism and Respiration 

Investigations on isolated enzymes have indicated an important block 
of phosphoglucose isomerase by 2-DG-6-P and contributory inhibition of 
hexokinases under certain conditions. Let us now turn to studies on carbo- 
hydrate metabolism in intact cells and tissues in order to determine if the 
effects of 2-DG can be explained adequately on this basis, or to accumulate 
evidence of blocks elsewhere. Anaerobic glycolysis, aerobic glycolysis, 
and glucose respiration are inhibited by 2-DG but to very different degrees 
(Fig. 2-10). Indeed, respiration is inhibited only with high concentrations, 
usually 30-100 times that required to inhibit anaerobic glycolysis compara- 
bly (Woodward and Hudson, 1954; Tower, 1958), so that some workers 
have reported that respiration is not inhibited (Fridhandler, 1959; Taylor, 
1960). In the case of sea urchin eggs, the inhibition can be almost completely 
counteracted by increasing glucose concentration (Bernstein and Black, 
1959). However, glucose must be present when the 2-DG is added and is 
ineffective when the inhibition has developed. The respiration of guinea pig 
skin in the presence of various substrates is inhibited by 2-DG moderately 
and progressively (see accompanying tabulation) (Carney et al., 1962). 
All substrates and 2-DG were 20 mM. There is no effect on the endogenous 



392 



2. ANALOGS OF ENZYME REACTION COMPONENTS 



100 



50 



% 

INH 



ANAEROBIC 
GLYCOLYSIS 



0.1 




(2-06) 



Fig. 2-10. Effects of 2-DG on the glucose metabolism in cat brain slices. 
(From Tower, 1958.) 

respiration. The small effect on galactose respiration was felt to be due 
perhaps to a different hexokinase being used for the phosphorylation of 
galactose, since inhibition is exerted by 2- and 4-deoxy galactose. 



Substrate 


% Respiratory inhibition at: 










0-2 hr 


22 


-24 hr 


Glucose 


12 




53 


Mannose 


10 




43 


Fructose 


8 




50 


Galactose 


3 




21 


PjTuvate 







1 



Glycolysis as measured by the formation of C^^Og from glucose-u-C^^ 
is depressed by 2-DG in diaphragm (Nakada and Wick, 1956), kidney (Serif 
and Wick, 1958), lymph node cells (Helmreich and Eisen, 1959), and adipose 
tissue (Brooks et at., 1961). The variation of inhibition with 2-DG concen- 
tration is shown in Fig. 2-11 for rat kidney slices. Glycolysis as measured 
by unlabeled COg or lactate formation is also inhibited in yeast (Cramer 
and Woodward, 1952), ascites carcinoma and leukemic cells (Laszlo et al., 
1958), and cerebral cortex slices (Tower, 1958). The depression of aerobic, 
and anaerobic glycolysis in tumor tissue is counteracted by increasing glu- 
cose concentrations (Woodward and Hudson, 1954). These results are quite 



EFFECTS OF 2-DEOXY-D-GLUCOSE 



393 



consistent in showing an inhibition of the glycolytic pathway by 2-DG. 
However, the formation of CO2 from glucose is not always depressed. Frid- 
handler (1959) found that although 2-DG inhibits anaerobic glycolysis 
in rabbit blastocysts, the rates of respiration and COg formation are not 
significantly affected. The formation of C^^Og from glucose-1-C^* in human 
fetal liver is actually increased by 2-DG, but this may be attributed to a 
partial inhibition of glycolysis (Villee and Loring, 1961). In ascites carci- 
noma cells 2-DG simultaneously inhibits the glycolysis of glucose and 



X 

INH 



100 






^^^--''^ - DE ox Y - D - GLUCOSE 


. 


^^^"^"^^^ 


50 


■ /^^^ 


< 


/ ^^ 6- DEOXY- 6- FLUORO-D-GLUCOSE 



10 



20 



30 



40 



50 



60 



(I) 



70 

mM 



Fig. 2-11. Inhibition of the formation of C^^Oa from glucose-u-C* 

by glucose analogs in rat kidney slices. Glucose was 10 vaM. (From 

Serif and Wick, 1958.) 



increases its oxidation (Christensen et al., 1961), while the disappearance of 
glucose from the medium is reduced (Fig. 2-12). The increased C^^Oj formed 
from glucose coupled with the depressed glucose uptake must be taken to 
mean that the pathway of glucose utilization has been markedly altered, 
i. e., less glucose is going to lactate and more is being oxidized. One impor- 
tant factor in such tissues must be the activity of the pentose-P pathway, 
which is apparently not directly inhibited but is indirectly stimulated by 
2-DG. The formation of C^^Og from glucose-6-C^* in calf thymus nuclei is 
inhibited essentially completely by 2-DG (McEwen et al., 1963 b). However, 



394 



2. ANALOGS OF ENZYME REACTION COMPONENTS 



the C-l/C-6 ratio in brain slices remains the same when ghicose uptake is 
reduced to one third by 2-DG (Tower, 1958). 

Since hexose uptake into cells is often coupled with phosphorylation, 
one would expect 2-DG to inhibit this uptake by suppressing kinase activity 
directly or indirectly. Lymph node cells treated with 2-DG until 2-DG-6-P 
is formed and then washed free of 2-DG do not accumulate glucose, fructose, 
or mannose, and lactate formation is markedly reduced (Helmreich and 
Eisen, 1959). The uptake of glucose into chick embryo hearts is 35-45% 



ISO 



0, FORMED 




30 



60 



90 



(2-06) 



120 

mM 



Fig. 2-12. Effects of 2-DG on the glucose metabolism in Ehrlich ascites carcinoma 
cells. Glucose-u-C* was 10 mM. Control for residual glucose taken from nonincubated 
flasks with no 2-DG. Experiments 2 and 4 were averaged. (Data from Christensen 

et ah, 1961.) 



reduced by 40 mM 2-DG (Modignani and Foa, 1963) and into carrot slices 
is reduced to about the same degree by 10 mM 2-DG, this inhibition not 
being overcome by increase in glucose concentration (Grant and Beevers, 
1964). 2-DG interferes with galactose uptake into mouse strain L cells but 
not potently {K^ = 7.2 niM) (Maio and Rickenberg, 1962). It is quite likely 
that these inhibitions are exerted predominantly by 2-DG-6-P. On the 
other hand, the accumulation of a-methylglucoside, which is transported 
into E. coli by the glucose carrier, is well inhibited by 1-DG, poorly by 
6-DG, and not at all by 3-DG (Hagihira et al, 1963). Apparently 2-DG 
is not as potent an inhibitor here as 1-DG. 

Another mechanism by which 2-DG could alter carbohydrate uptake and 
metabolism is by changing the levels of Pj, ADP, and ATP, since the rates 



EFFECTS OF 2-DEOXY-D-GLUCOSE 395 

of hexose phosphorylation and glycolytic breakdown are controlled by these. 
In brain slices, 10 mM 2-DG causes a 50% fall in creatine-P and almost 
complete disappearance of the adenosine polyphosphates (Tower, 1958). 
There is also a diversion of phosphate to the stable 2-DG-6-P so that a cer- 
tain amount of phosphate is removed from glycolytic participation, as also 
pointed out by Kiesow (1960 c). Furthermore, 2-DG has been shown to 
inhibit the incorporation of P/^ into ADP and ATP in ascites carcinoma 
cells 70% at 10 mM (Greaser and Scholefield, 1960). El'tsina and Beresot- 
skaya (1962) determined P, and nucleotide levels in tumor cells exposed to 
11 raM 2-DG and found marked decreases in ATP and ADP (see accompa- 
nying tabulation). On the other hand, rat liver and kidney slices show no 



Tumor ( 


components 


Control 


2-DG 


Zahdel 


hepatoma 






ATP 




69.2 


7.7 


ADP 




28.9 


13.2 


AMP 




— 


11.7 


P. 




157 


39 


Sarcoma 37 






ATP 




74.7 


1.7 


ADP 




40.5 


5.4 


AMP 




14.2 


3.5 


P, 




140 


67 



significant changes, a difference attributed to variations in the activity 
and cellular location of hexokinases. McComb and Yushok (1964 a) also 
reported marked falls in ATP in ascites cells within 12 min after exposure 
to 2-DG, and a 65% net loss of the cellular adenine nucleotides. The disap- 
pearance of nucleotides is at least partly accounted for by the phosphory- 
lation of 2-DG by hexokinase, the formation of AMP mediated by adeny- 
late kinase, the deamination of AMP to IMP, and the splitting of IMP 
to inosine by 5 '-nucleotidase (McComb and Yushok, 1964 b). They observed 
a steady rise in inosine, correlated with a rise and subsequent fall in IMP. 
Changes in nucleotide levels should affect various phases of metabolism 
which involve these substances, and this is well seen in the effects of 2-DG 
on the oxidation of ethanol in yeast (Maitra and Estabrook, 1962). When 
2-DG is added to previously starved yeast in the presence of ethanol, 
there is acceleration of respiration and the oxidation of NADPH, accompa- 
nied by a fall in ATP with an elevation of ADP. This stimulatory phase 
lasts less than a minute and is followed by a depressed phase characterized 



396 2. ANALOGS or enzyme reaction components 

by very low levels of P^. During the oxidation of ethanol, the ATP/ADP 
and P,/ADP ratios are high; when 2-DG is added it temporarily augments 
respiration by lowering these ratios, but within a minute enough phosphate 
has been trapped in 2-DG-6-P to cause a marked fall in the Pj/ADP ratio 
(perhaps from 17 to 1.5). The P, may now be so low that it limits the respi- 
ration. 

The respiration of certain tissues is diminished by the addition of glucose, 
a phenomenon often called the Crabtree, or reversed Pasteur, effect; it 
is particularly evident in Ehrlich ascites carcinoma cells and most of 
the studies of the mechanisms involved have been on these cells. It has 
been stated that an acceleration of glycolysis inhibits the oxidation of 
pyruvate, but there was no real evidence to link the entire EM pathway 
with respiratory control. The effects of 2-DG are thus of particular impor- 
tance, since it is phosphorylated but not further metabolized to any extent. 
It was shown that 2-DG inhibits respiration to about the same degree as 
glucose (Ibsen et al., 1958). Respiration and pyruvate decarboxylation are 
reduced 50% by 10 milf 2-DG (Ram et al, 1963). There has been dis- 
agreement as to whether glucose and 2-DG act by the same mechanism or 
differently. Let us briefly compare the responses to these sugars. (1) 2-DG 
depresses the respiration more slowly than does glucose. Yushok (1964) 
has shown in a group of sugars that the rate of respiratory inhibition is 
correlated with the rate of phosphorylation. One would thus expect 2-DG 
to act more slowly than glucose, so this does not constitute a real difference 
in action. (2) The inhibition by glucose is released when it is all glycolyzed 
but the inhibition by 2-DG remains (Ibsen et al., 1962; Hofmann et al., 1962). 
This does not seem to me to be valid evidence for different mechanism of 
action. (3) The addition of glucose leads to the formation of lactate whereas 
2-DG does not (Ram et al., 1963). This is what would be expected, of course, 
but emphasizes that glycolysis, as defined classically, is not necessary for the 
effect. (4) Glucose at 10 mM inhibits the respiration 40%, 2-DG at 20 mM 
inhibits it 48%, and both together inhibit it only 23% (Wenner and Cereijo- 
Santalo, 1962). This was interpreted to mean that the inhibitory mechanisms 
are quite different. (5) It has been stated that amobarbital prevents the 
inhibition of respiration by 2-DG but not by glucose (Wenner and Cereijo- 
Santalo, 1962). This is true, however, only in the presence of succinate, 
since the endogenous respiration is not affected by either glucose or 2-DG 
in the presence of amobarbital (there is very little to be affected). (6) The 
respiratory inhibition by glucose is released by 2,4-dinitrophenol, but 
there is some disagreement as to the effect of the uncoupler with 2-DG, 
Ibsen et al. (1962) stating that it releases the inhibition and Ram et al. 
(1963) stating that it does not. The latter workers, however, did not feel 
that this is evidence for different mechanisms and were inclined to attrib- 
ute the differences to the availability of glycolytic intermediates. (7) Both 



EFFECTS OF 2-DEOXY-D-GLUCOSE 397 

glucose and 2-DG reduce the ATP level immediately and the ADP level 
very soon (Ibsen et al., 1962). The most commonly accepted explanation 
of the Crabtree effect is a depletion of ADP, since the rate of mitochondrial 
oxidation in ascites cells depends on the level of ADP. (8) 2-DG causes the 
loss of enzymes from the ascites cells (measured with lactate dehydrogen- 
ase) whereas glucose does not (Hofmann et al., 1962). Indeed, glucose 
seems to antagonize this action of 2-DG. Whether this observation has any 
bearing on the Crabtree effect is not known. (9) Both glucose and 2-DG 
still exert respiratory inhibition in the presence of sufficient iodoacetate 
to block almost completely the glycolytic pathway (Ibsen et al., 1958; 
Wenner and Cereijo-Santalo, 1962). It seems that although the Crabtree 
effect is not abolished by iodoacetate, it may be diminished. If the initial 
phosphorylation of hexoses is responsible for the Crabtree effect, iodoacetate 
would not be expected to inhibit it, except as it might reduce ATP for the 
kinase reactions. (10) Glucosone inhibits the Crabtree effect produced by 
both glucose and 2-DG (Yushok, 1964), but this is probably the result of 
the inhibition of hexokinase by glucosone. Summarizing these results, it 
would appear that glucose and 2-DG inhibit respiration by basically the 
same mechanism and that this is related to their phosphorylation. It is 
difficult to understand the diminished Crabtree effect in the presence of 
glucose and 2-DG together, observed by Wenner and Cereijo-Santalo (1962), 
but this should be investigated further, inasmuch as Ram et al. (1963) 
stated that the respiratory inhibition by glucose is not enhanced by 2-DG, 
apparently no antagonism being noted. 

Glucose and 2-DG not only depress the oxidation of pyruvate in ascites 
cells but even more strongly the oxidation and C^'^Oa formation from label- 
ed palmitate (Sauermann, 1964). Inhibition of palmitate oxidation occurs 
to the extent of around 80% at the relatively low concentration of 1.8 
mM 2-DG. The inhibitions are approximately 29% for acetate, 58% for 
pyruvate, and 92% for palmitate at 18 mM 2-DG. There would thus appear 
to be some effect on fatty acid oxidation which is exerted prior to the uti- 
lization of acetyl-CoA. This action is not related to the inhibition of glyco- 
lysis for several reasons, including the demonstration that it occurs in the 
presence of iodoacetate. Some effect on fatty acid oxidation might be 
expected from a lowering of the ATP level, but it was felt that this is not 
the entire explanation because of the relatively small effect on acetate 
oxidation. It is not necessary, however, that a fall in ATP should affect 
acetate and palmitate oxidations equally. One awaits with interest the 
elucidation of this interesting effect. 

The Crabtree effect has been discussed in some detail because it illus- 
trates one way in which 2-DG can alter carbohydrate metabolism through 
the alteration of levels of P, and the adenine nucleotides. It is quite possi- 
ble that part of the effect of 2-DG on tissues is the result of a lowering 



398 



2. ANALOGS OF ENZYME REACTION COMPONENTS 



of available ATP, this secondarily reducing hexokinase activity, and the 
uptake of sugars. The greater resistance of aerobic glycolysis and respira- 
tion to 2-DG, compared to anaerobic glycolysis, may be due in part to the 
higher ATP levels aerobically. The differential effects on the uptakes of 
different hexoses (Fig. 2-13) might be explained on the basis of whether a 
hexose is actively transported or not, and the steady-state rate of its 
phosphorylation (e. g., galactose metabolism is not depressed as much as 
that of glucose), and the different affinities of the hexoses for the hexoki- 
nases. 



20 



15 



10 



5 ■ 



c 0^ 
(% OF 
ADDED ) 



^ 


--^ 


■"-..^GLUCOSE 


\ 


FRUCTOSE 




GALACTOSE 


^ 


-^-^^__ 





10 



15 



20 



[ 2 - DG ) 



Fig. 2-13. Inhibition of hexose oxidation by 2-DG in isolated rat 
diaphragm. Hexoses were 10 mM. (From Nakada and Wick, 1956.) 



A summary of the sites and mechanisms in the inhibition of carbohydrate 
utilization by 2-DG would then include: (1) primary competitive inhibition 
of certain hexokinases by 2-DG, (2) possible direct interference in the ac- 
tive transport of hexoses into the cell, (3) inhibition of phosphoglucose 
isomerase by 2-DG-6-P, (4) secondary reduction in transport and hexose 
phosphorylation through depletion of ATP, and (5) possible inhibitions by 
2-DG-6-P of glycolytic enzymes not yet examined. 



EFFECTS OF 2-DEOXY-D-GLUCOSE 399 

Effects of 2-DG on Various Metabolic Pathways 

Infusion of acetate-C^^OO" into rabbits and determination of the respired 
C^^Og were done prior to and after 2-DG injections; no alteration of acetate 
oxidation was observed (Wick d al., 1957). The total COg produced decreas- 
ed, partly due to the hypothermia brought about by 2-DG. This was taken 
as evidence that any block by 2-DG is previous to the tricarboxylate cycle. 
An effect of 2-DG on citrate levels in ascites carcinoma cells in the presence 
of pyruvate was reported by Letnansky and Seelich (1960). The citrate 
begins to rise 2 min after addition of 18.3 mM 2-DG and eventually reaches 
levels definitely higher than in untreated suspensions. The utilization of 
pyruvate is also depressed around 50%. These observations might be in- 
terpreted as originating from some action on the cycle, but more recently 
(Seelich and Letnansky, 1961) it was shown that methylene blue reduces 
the high citrate levels and promotes pyruvate utilization in the presence of 
2-DG. It was postulated that this is caused by oxidation of NADPH to 
NADP, which is necessary for isocitrate oxidation. The rise in citrate may 
be due to a deficiency of NADP, possibly because lactate is not formed in 
the presence of 2-DG and NADPH is not oxidized, and also due to a drop in 
ATP. The results can thus be satisfactorily explained on the basis of the 
mechanisms previously discussed without assuming any direct action on 
the cycle. 

Lipogenesis in human fetal liver from glucose is depressed by 2-DG, 
incorporation of l-C^* being lowered 33% and of 6-Ci* 18% at 6.1 mM 
(Villee and Loring, 1961). This can, of course, be attributed to an inhi- 
bition of glycolysis. Plasma fatty acids in man rise as much as 330% 
following intravenous infusion of 2-DG at 60 mg/kg, and this might be 
interpreted as a depression of fatty acid synthesis so that the normal equi- 
librium is disturbed and fatty acids are mobilized from the tissues (Laszlo 
et al., 1961). On the other hand, there is evidence that 2-DG augments 
epinephrine release, so that this could be partly responsible. 

The appearance of labeled amino acids from labeled glucose in brain 
slices is also depressed by 2-DG (Tower, 1958). 2-DG not only prevents the 
accumulation of glutamate but causes a profound fall in the intracellular 
level, glutamine decreasing less markedly. It is believed that glutamate 
accumulation is important in K+ uptake by brain and it was found, as 
expected, that this is inhibited around 60% by 10 mM 2-DG. The role of 
such changes in the central actions of 2-DG in the whole animal is not as 
yet understood. The incorporation of L-valine-C^* into protein in ascites 
cells is almost abolished by 10 mM 2-DG; since this may be reversed by 
glucose, it is likely that the inhibition is related to glycolytic suppression 
(Riggs and Walker, 1963). The DNA content of a rat carcinoma is reduced 
by feeding 2-5% 2-DG in the diet and possibly this is related to the observed 
inhibition in growth (Sokoloff et al., 1956). 



400 2. ANALOGS OF ENZYME REACTION COMPONENTS 

Effects of 2-DG on Cell Division and Growth 

The growth rate of E. coli is depressed at least 50% by 10 mM 2-DG 
but the differential rate of inducible /5-galactosidase synthesis is not altered 
(Cohn and Horibata, 1959). Although Neurospora and Aspergillus can 
grow slowly on 2-DG alone, the growth is inhibited when the usual sugars 
are present (Sols et al., 1960 b). The mutiplication of influenza virus in 
chick embryos is markedly inhibited by 2-DG but in preliminary studies 
there was no evidence that the development of lesions in mice is altered 
(Kilbourne, 1959). Glucose and, to some extent, pyruvate are able to 
counteract this depression. Sea-urchin egg cleavage is delayed by 2-DG 
and development is stopped at various stages, depending on the concen- 
tration: 1-10 mM prevents gastrulation and the eggs reach swimming 
blastulae, 100 mM delays first cleavage but the early blastula stage is 
eventually reached, 200 mM causes greater cleavage delay and develop- 
ment stops before the blastula stage (Bernstein and Black, 1959). Glucose 
can counteract the cleavage delay. The most sensitive tissue investigated 
appears to be chick embryo heart fibroblasts, since the mitotic index is 
reduced from 2.65 to 0.15 by 1.52 mM 2-DG (Ely et al, 1952). Glucose at 
approximately equimolar concentration is not able to reverse this inhibition 
significantly. The migration of cells in the 2-DG-treated cultures is also 
limited and the cells become vacuolated. These observations all point to a 
general growth-inhibiting action of 2-DG, which is not unexpected, but 
there has been little study of differential growth effects. 

The ability of tumors to derive a good part of their energy for growth 
from glycolysis prompted the original study of 2-DG as a possible carcino- 
static agent, but surprisingly little work has been done on this aspect. 
It has been established that glycolytic inhibition of tumors in vivo is pos- 
sible, in that patients with chronic myelogenous leukemia infused with 
60 mg/kg 2-DG show 35-40% inhibition of glycolysis in the leukemic cells 
(Laszlo et al., 1958). The growth of cultured HeLa cells from human carci- 
noma is inhibited readily by 2-DG, 5 mM producing essentially complete 
depression which is reversible up to 3 days but not afterward (Barban 
and Schulze, 1961). Glucose or mannose will counteract the growth inhibi- 
tion. Cells in a fructose medium are inhibited more readily than when 
grown on glucose or mannose, which corresponds to the greater inhibition 
of fructose utilization discussed previously. When 2-DG is added at 2-5% 
to the diet of rats, the growth of carcinoma G-175 in reduced (Sokoloff et al., 
1956). Adult rats tolerate this dosage well but the growth of young rats is 
retarded. The 2-DG can also be injected subcutaneously at 2-4 mg/kg/day. 
Mouse sarcoma is similarly affected. The inhibition in either case is not 
very marked. Solid, transplanted, and ascitic tumors in mice grow more 
slowly when 2-DG is administered (Laszlo et al., 1960). A modest prolonga- 
tion of the survival time of the animals was noted. Definite carcinostatic 



EFFECTS OF 2-DEOXY-D-GLUCOSE 401 

activity in vivo has thus been demonstrated but there should be much more 
work to establish if sufficient differential depression can be achieved, and 
other types of neoplasm should be studied. 

Effects of 2-DG on Whole Animals 

The intravenous infusion of 2-DG at doses of 50-200 mg/kg in cancer 
patients produces a feeling of warmth, flushing, diaphoresis, headache, 
drowsiness, tachycardia, a rise in blood glucose, and a fall in white cell 
count (Landau et al., 1958). Hyperglycemia has been noted in all studies 
and might be attributed to a reduced utilization of glucose brought about 
by glycolytic inhibition. However, other factors must be considered. Brown 
and Bachrach (1959) showed that the rise in blood glucose from 2-DG can 
be partially prevented by demedullation of the adrenals, indicating that 
2-DG may stimulate the release of epinephrine. Increases in urinary cate- 
cholamines during 2-DG infusion have also been noted (Laszlo et al., 1961). 
Hokfelt and Bydgeman (1961) felt that this epinephrine release is the 
primary cause of the hyperglycemia and showed that 2-DG can deplete 
the adrenals of half their epinephrine. Spinal transection reduces the 2-DG 
effect, indicating epinephrine release to be mediated through the central 
nervous system. Pretreatment with dihydroergotamine, which blocks the 
effects of epinephrine, abolishes the 2-DG hyperglycemia (Altszuler et al., 
1963; Sakata et al., 1963). It is interesting in connection with the possible 
effects of 2-DG on the central nervous system that 2-DG is transported from 
the blood into the cerebrospinal fluid faster than glucose; there is also com- 
petition between glucose and 2-DG for the carrier (Fishman, 1964). Landau 
et al. (1958), on the other hand, were inclined to discount the role of epine- 
phrine since no rise in blood pressure is observed during maximal hyper- 
glycemia. The hyperglycemia probably is responsible for the decreased 
stainability and degranulation of pancreatic islet cells produced by 2-DG, 
indicating increased activity of these cells, rather than a direct action 
(Hokfelt and Hultquist, 1961). The symptoms listed above must be due 
in part to the restricted glucose utilization caused by 2-DG. The rise in 
blood glucose must tend to counteract this inhibition, since glucose in- 
fusions reduce mortality from 2-DG, but not sufficiently. It is interesting 
that the anaphylactoid reaction to dextran and ovomucoid in rats is 
inhibited by 2-DG at 200 mg/kg intravenously, although the response to the 
histamine releasers is not affected, indicating an important role of glucose 
uptake and metabolism in certain inflammatory reactions (Goth, 1959). 
The LD50 in mice is around 2.5 g/kg for single intravenous injections and 
5 g/kg subcutaneously or intraperitoneally (Laszlo et al., 1960). The animals 
survive several hours to a day in most cases but may die within 10 min af- 
ter intravenous injection. The major toxic effects are related to the central 
nervous system and are weakness, convulsions, and coma. 



402 2. ANALOGS OF ENZYME REACTION COMPONENTS 

Effects of 2-DG on the Heart * 

The contractile tension of rat atria is progressively depressed by 10 mM 
2-DG when glucose is present at its usual concentration of 5.5 mM; the inhi- 
bition is 25% at 20 min, 50% at 40 min, and 70% at 90 min (A. Gimeno and 
M. Gimeno). If pyruvate is present, the rate of depression is slower and the 
inhibition is 30% at 90 min; if pyruvate is added at 30 min when the depres- 
sion by 2-DG is around 40-45%, there is partial recovery to the — 30% 
level; if glucose and pyruvate are present and 2-DG is added at 30 min, 
there is a depression to the — 30% level at 90 min. The atria in the absence 
of glucose progressively fail so that at 30 min the contractile tension is 
50% depressed (actually the course is quite similar to that when glucose 
and 2-DG are present), but with 2-DG the rate of fall is much faster, in- 
dicating that 2-DG can effect the endogenous metabolism. Addition of 
glucose at 30 min when the depression is 80% or more results in partial 
recovery, pyruvate is less effective, and both allow return to near the 
— 30% level. The rapid cessation of the 2-DG depression brought about 
by either glucose or pyruvate is noteworthy. The contractile levels reached 
in 90 min may be summarized in the following tabulation. The failure of 



(I) Glucose-free and glucose added at 30 min — 0% 

(II) Glucose-free and pyruvate added at 30 min —15% 

(III) Glucose + pyruvate + 2-DG (in any order) —30% 
(IV^ Glucose + 2-DG (in any order) -70% 

(V) 2-DG alone -90% to -100% 



pyruvate to maintain normal contractions might indicate that some 25-30% 
of the tension is dependent on glycolysis, but it is also possible that 2-DG is 
interfering in some way with the utilization of pyruvate. The ability of 
glucose to stimulate 30 min after depression by 2-DG at 10 mM shows 
that the block of glycolysis is only partial or that glucose is acting by some 
other mechanism. The atrial depression by anoxia (— 85% at 10 min) is 
accelerated by 2-DG (— 93% at 5 min and — 100% at 10 min), and re- 
covery upon readmission of Og is much less when 2-DG is present (J. La- 
cuara). Rat ventricle strip contractility is not affected over 160 min by 
0.5-2 milf 2-DG, but 4 mM causes a slow depression and partially prevents 
the positive inotropic effect of ouabain (E. Majeski). 

* Inasmuch as so little is known of the effects of 2-DG on tissue functions and 
nothing has been reported relative to the heart, this section summarizes briefly some 
of the recent work done in our department and not yet published at the time of sub- 
mission of the manuscript. 



EFFECTS OF 6-DEOXY-6-FLUORO-D-GLUCOSE 403 

It is very interesting that the atrial depression is completely unassoci- 
ated with demonstrable changes in the membrane electrical characteristics 
(E. Ruiz-Petrich). 2-DG at 10 mM depresses the contractile tension around 
50% at 20-30 min under the conditions of these experiments, glucose being 
present, but there are no changes at all of the resting potential, the action 
potential magnitude, or the rates of depolarization and repolarization. The 
addition of 5 mM pyruvate rapidly stopped the progression of the contractile 
depression and allowed slight recovery, again without detectable altera- 
tions of the membrane characteristics. 2-DG is the only inhibitor with which 
we have worked that is able to affect the contractile processes so selectively, 
all other inhibitors decreasing the action potential duration to varying 
degrees and producing other correlated changes in the potentials. If 2-DG 
acts here by reducing the utilization of glucose or glycogen, these results 
would point to a close relation between the contractile process and some as- 
pect of glycolysis other than the generation of ATP. It has also been shown 
that atrial K+ influx and efflux are only very slightly altered by 11 mM 
2-DG, and that intracellular K+ is unchanged over a period during which 
the contractile activity is depressed 50% (Chin, 1963). 

EFFECTS OF 6-DEOXY-6-FLUORO-D-GLUCOSE 
ON METABOLISM 

Modification of hexoses at the 6-position should interfere with their 
phosphorylation by hexokinase, and hence any inhibition of glucose metab- 
olism observed would probably not be exerted beyond the hexokinase step. 
Thus inhibition produced by 6-substituted sugars should be simpler than 
inhibition by 2-substituted sugars, which can be phosphorylated and may 
block at several sites. Brooks et al. (1961) compared 6-deoxy-D-glucose 
with 2-DG on the oxidation of glucose-u-C^* by various tissues and found 
it to be somewhat less potent. When glucose is 10 mM and 6-DG is 30 mM, 
the inhibitions of C^'^Og formation are 43% for adipose tissue, 23% for 
kidney, and 19% for diaphragm. The inhibition in adipose tissue is reversed 
by increasing glucose concentration and appears to be competitive. 6-DG 
is not metabolized by adipose tissue and no C^^Oa arises from 6-DG-u-C^^. 
The site of inhibition was considered to be either the membrane transport 
system or hexokinase. It has been found that galactose transport across 
the intestine is inhibited by 6-DG (47% at 5 mM) and that the transport of 
6-DG is depressed by glucose (Wilson et al., 1960). 

The replacement of the 6-0II group with fluorine to give 6-deoxy-6- 
fluoro-D-glucose (6-DFG) leads, as expected, to an interesting inhibitor 
of glucose utilization. The original idea was apparently to produce a gly- 
colytic inhibitor with fluorine analogous to the cycle inhibitor fluoroacetate. 
The initial work was done by Blakley and Boyer (1955) at Minnesota; they 



404 2. ANALOGS OF ENZYME REACTION COMPONENTS 

showed that fermentation of glucose and fructose by yeast is competitively 
inhibited by 6-DFG. The apparent constants are shown in the following 
tabulation. 6-DFG is not fermented by intact cells or yeast extracts. In 





Glucose 


Fructose 


Yeast 


Km 

(milf) 


Ki 

{mM) 


Km 

{n\M) 


K, 

(mM) 


Baker's 
Brewer's 


1.8 
6.9 


7.3 

2.7 


5.0 

27 


3.3 

2.3 



extracts 6-DFG has essentially no effect even at 36 milf and hexokinase 
is very weakly inhibited. Glucose utilization by rat diaphragm is not in- 
hibited as well as yeast fermentation, and 6-DFG is not taken up as readily 
as glucose by the muscle cells. Glucose oxidase (notatin) oxidizes 6-DFG at 
about 3% the rate of glucose oxidation so that direct oxidation in the 
tissues is probably negligible. It was concluded that the rate-limiting re- 
action for glucose utilization is different in intact cells and extracts, and 
that 6-DFG probably inhibits the membrane transport of normal hexoses. 

The uptake and metabolism of 6-DFG may vary from tissue to tissue. 
For example, although 6-DFG is not actively transported into rat diaphragm 
and insulin has no effect on this transfer, it is well transported across the 
intestinal wall whereas 2-DG is not (Wick et at., 1959; Wilson and Landau, 
1960). And, although 6-DFG is not metabolized in most tissues, a particulate 
preparation from Aerobacter aerogenes is able to oxidize it to the corres- 
ponding gluconate (Blakley and Ciferri, 1961). It is fairly certain that the 
compound is quite stable and that release of fluoride does not occur suffi- 
ciently to inhibit glycolysis (Serif et al, 1958). 

6-DFG inhibits the formation of C^^Oa from glucose-u-C^^ in liver, kidney, 
and adipose tissue, without modifying the metabolism of acetate or lactate 
(Serif and Stewart, 1958; Serif and Wick, 1958; Serii et al, 1958). In general 
6-DFG is somewhat less potent than 2-DG (Fig. 2-11), but the relative 
sensitivities depend on the tissue studied. Nevertheless, in the eviscerated 
rat 6-DFG is able to inhibit the intracellular transport of glucose quite 
appreciably, e. g., 42% from 200 mg/kg (Wick et al, 1959). There is no direct 
evidence that 6-DFG acts elsewhere than on membrane transport and the 
comparable inhibitions produced on the metabolism of glucose- 1-C^^, glu- 
cose-6-C^*, and glucose-u-C^* would point to a block prior to the glyco- 
lytic-pentose phosphate shunt division (Serif and Wick, 1958). However, 
hexokinases from other than yeast have not been adequately tested. Al- 
though the toxicities of 2-DG and 6-DFG have not been directly compared, 
Blakley and Boyer (1955) found that 250 mg/kg 6-DFG intraperitoneally 



INHIBITORS OF CARBOHYDRATE METABOLISM 405 

in rats produces toxic symptoms with recovery. Since the LD50 for 2-DG 
is probably around 10 times this, it would indicate that 6-DFG must af- 
fect the central nervous system more than the other tissues which have 
been studied. 

VARIOUS ANALOG INHIBITORS 
OF CARBOHYDRATE METABOLISM 

Numerous inhibitions of enzymes in the glycolytic and pentose phosphate 
pathways by analogs have been reported. Some of these may be important 
in attempts to block carbohydrate metabolism specifically and some are 
undoubtedly significant in mechanisms regulating the rates in these path- 
ways. We shall discuss a few of the more important enzymes and inhibi- 
tions, no effort being made to include all of the observations. 

Phosphorylases 

The enzymes involved in the synthesis and phosphorolysis of polysac- 
charides, such as starch or glycogen, are very stereospecific with respect 
to substrates and inhibitors. The fairly potent inhibition of glycogen phos- 
phorolysis by the product glucose- 1-P in preparations from rabbit muscle 
was reported by Cori et al. (1939), 7 mM inhibiting 93%, whereas glucose- 
6-P at the same concentration inhibits only 17%. The reverse reaction of 
glycogen synthesis from glucose-1-P is inhibited competitively by glucose, 
and to a lesser extent by mannose, galactose, and maltose, but the ajffinity 
of the enzyme for these sugars is low since 30% inhibition occurs when 
glucose and substrate are approximately equimolar at 17 m.M (Cori and 
Cori, 1940). However, the lobster muscle phosphorylase is inhibited only 
25% by 250 mM glucose when glucose-1-P is 100 mM (Cowgill, 1959). 

The only known substrate for rabbit muscle phosphorylase is a-D-glu- 
cose-l-P with respect to glycogen synthesis, and Cori and Cori (1940) had 
found that only a-D-glucose inhibits, /?-D-glucose having little if any ef- 
fect, so the question of a, /5-specificity was studied in detail on the crystal- 
line enzyme by Campbell et al. (1952). It was found that /?-D-glucose-l-P 
is neither a substrate nor an inhibitor, and that whereas a-methyl glucoside 
inhibits, /?-metliylglucoside is without activity. The /?-anomers appear to 
have no affinity for the enzyme. Furthermore, a large number of sugars 
and derivatives at 50 mM were found to be inactive. The configurations of 
the hydroxyl groups on all the positions of glucose seem to be necessary 
for combination with the enzyme. A pyranose structure seems also to be 
a requirement for inhibition, since sorbitol and inositol are without effect, 
and a primary alcohol group on C-5 is necessary since D-xylose is inactive. 
The failure of fructose to inhibit might be explained in several ways: (1) 
the more planar furanose form is dominant, (2) the configurations on the 



406 2. ANALOGS OF ENZYME REACTION COMPONENTS 

C-1 of D-glucopyranose and C-2 of D-fructopyranose are different, (3) 
there is a primary alcohol group on the C-5 of glucose but not on the C-6 
of fructose, and (4) the /?-anomer of fructose may predominate. 

An interesting type of inhibition on the phosphorolysis of starch by 
monofluorophosphate was observed by Rapp and Sliwinski (1956). The 

0- o- 

-O— P+— OH -0— P+— F 

o- o- 

Orthophosphate Monofluorophosphate 

sizes and electronic configurations of orthophosphate and monofluoro- 
phosphate are quite similar so that interference with many reactions in- 
volving phosphate might be anticipated. The inhibition of potato phosphory- 
lase is completely competitive {Kj„ = 3.2 mM and K, = 2.8 mM calculated 
from their ijv-ijiS) plot) and the affinity of the enzyme for the two substances 
corresponds to this tectonic resemblance. The inhibition is not due to the 
release of fluoride since 20 mM fluoride inhibits only 6.4% and 2.1 mM 
monofluorophosphate inhibits 50%. 

Phosphoglucose Isomerase 

This enzyme is important in regulating carbohydrate metabolism since 
its activity, along with other factors, determines how much glucose-6-P 
enters the glycolytic pathway; in other words, this enzyme represents a 
branching point of metabolism in the terminology of Krebs. We have seen 
that inhibition by 2-DG-6-P is probably an important component of the 
mechanism of action of 2-DG. The potent inhibition by 6-phosphogluconate 
is particularly interesting because this substance is formed from glucose-6-P 
in the pentose phosphate pathway and could determine to some extent the 
diversion at the branching point. Parr (1956, 1957) reported inhibition of the 
enzymes from blood, liver, muscle, and potato and found it to be competi- 
tive; in the reaction glucose-6-P -> fructose-6-P, 1 mM inhibits 75% and 

fructose-6-P »^ glycolytic pathway 

Glucose »- glucose-6-P 



-phosphogluconate ■*- pentose- P pathway 

2 mM 95% (glucose-6-P 2 mM). Similar competitive inhibitions of the en- 
zymes from yeast (Noltmann and Bruns, 1959) and Trichinella spiralis 
(Mancilla and Agosin, 1960) have been noted. Rabbit muscle phosphoglu- 
cose isomerase may be even more sensitive to 6-phosphogluconate, since the 



INHIBITORS OF CARBOHYDRATE METABOLISM 407 

Ki is around 0.005 mM (Kahana et al., 1960). Another potent inhibitor that 
is an intermediate in the pentose-P pathway is erythrose-4-P, for which 
the K^ is 0.0007- 0.001 mM {K„, for fructose-6-P is 0.08 mM) (Grazi et al, 
1960). Under conditions that would lead to an accumulation of erythrose- 
4-P, more glucose-6-P might be diverted into the pentose-P pathway or, 
if fructose-6-P is being metabolized, a negative feedback effect would be 
exerted. A third potent inhibitor is glucosamine-6-P (Wolfe and Nakada, 
1956), which on the T. spiralis enzyme is around 40 times more inhibitory 
than 6-phosphogluconate, 0.06 mM inhibiting 64% when glucose-6-P 
is 5 mM (Mancilla and Agosin, 1960). The deamination of glucosamine-6-P 
in E. coli: 

Glucosamine-6-P -> fructose-6-P :f± glucose-6-P 

thus terminates at fructose-6-P initially because of the inhibition of the 
second reaction, but eventually disappearance of glucosamine-6-P relieves 
the inhibition and glucose-6-P is formed, another example of metabolic 
regulation through inhibition. 

Aldolase 

This enz^Tne splitting fructose- 1,6-diP to glyceraldehyde-3-P and dihy- 
droxyacetone-P has unfortunately been studied very little from the stand- 
point of inhibition by hexose or triose phosphates. Yeast aldolase binds sev: 
eral hexose phosphates quite tightly but splits them very slowly (Richards 
and Rutter, 1961), as may be seen from the K/s and the relative reac- 
tion rates (rate with fructose-l,6-diP as 1) in the accompanying tabulation. 



Inhibitor , ', Pielative rate 

(mM) 



L-Sorbose-l,6-diP 


0.13 


0.0014 


L-Sorbose-1-P 


0.2 


0.0002 


D-Fructose-l-P 


1.0 


0.0004 


D-Fructose-6-P 


3.8 


0.0001 



Muscle aldolase splits these analogs, niueh more readily. The aldolase from 
rabbit Ihuscle was found by Herbert et al. (1940) to be competitively 
inhibited by fructose-6-P (31%), fructose (16.3%), and glucose (5.4%), the 
percentages being for 20 mM inhibitor and fructose-l,6-diP. The difficulty 
in interpreting these results lies in our ignorance of the preferred form of 
the substrate (i. e., a or j3, p>Tanose, furanose, or linear) for the enzyme, 
and the distribution of the substrates and inhibitors among these forms 
under the experimental conditions. The values of K^ may be quite misleading 



408 2. ANALOGS OF ENZYME REACTION COMPONENTS 

because the inhibiting form could represent only a small fraction of the total 
concentration of the inhibitor. 

Glyceraldehyde-3-Phosphate Dehydrogenase 

A new, potent, and apparently specific inhibitor of this enzyme has 
been found and is likely to be useful as a blocking agent of this step in 
glycolysis. Glycolaldehyde-2-P freshly prepared does not inhibit glyceral- 
dehyde-3-P dehydrogenase, but either aging the preparation or allowing it 
to react in 1 N NaOH for a short time leads to the formation of a potent 
inhibitor, termed tetrose-diP by Racker et al., (1959) and isolated as d- 
threose-2,4-diP by Fluharty and Ballou (1959). Both the d- and L-isomers 

CHO 



CHO 


"OgP— 0— C— H 


CHa— 0— P03= — 


H— C— OH 




CH^— 0— PO 


ycolaldehyde-2-P 


D-Threose-2,4-diP 



of the inhibitor were synthesized by the latter workers and only the D-iso- 
mer was found to inhibit strongly. The inhibition is reversible but non- 
competitive with respect to glyceraldehyde-3-P; the inhibition may actually 
increase with NAD concentration. The value of K, for rabbit muscle en- 
zyme is 0.0001-0.0002 milf. D-Threose-2,4-diP is oxidized by the enzyme 
but only as much as the NAD present. Fluharty and Ballou postulated 
that it might react with a site other than the normal catalytic site for 
oxidation of glyceraldehyde-3-P but Racker etal., showed spectroscopically 
that a stable acyl-enzyme complex is formed, probably with the SH group 
known to be involved in the catalysis and the binding of NAD. The sub- 
strate reactions might be written as: 

Enzyme-NAD -f Glyceraldehyde-3-P :f± phosphoglyceryl-enzyme-NADH 
Phosphoglyceryl-enzyme-NADH + P 5± glycerate-l,3-diP + enzyme-NADH 

whereas the reaction with the inhibitor is: 

Enzyme-NAD + threose-2,4-diP :^ diphosphothreonyl-enzyme-NADH 

Normally an inorganic phosphate is transferred from a second site to the 
phosphoglyceryl-enzyme to form glycerate-l,3-diP, but the location of the 
2-phosphate group on the inhibitor is such as to occupy this second site 
so that no phosphate can enter the reaction. This is thus an example of 
an inhibition in which a substance enters the reaction sequence in the same 



INHIBITORS OF CARBOHYDRATE METABOLISM 409 

manner as the substrate, forming a stable complex with the enzyme, but is 
unable to complete the sequence. 

The use of D-threose-2,4-diP to block glycolysis in intact cells or tissues 
will probably not meet with general success due to the poor penetration. 
Extracts of ascites tumor cells are inhibited readily, but glycolysis in intact 
ascites or HeLa cells is unaffected (Racker et al., 1959). Preparations of 
glycolaldehyde-2-P, containing the tetrose-diP, inhibit the growth of some 
bacteria and not others, probably depending on the degree of penetration. 
The possibility was considered that the inhibitor, can be formed intracellu- 
larly but so far appropriate precursors have not been found. 

If photosynthesis involves the reversal of the glycolytic sequence, 
D-threose-2,4-diP should inhibit, and it has been found that the total C^^Og 
fixation in sonically ruptured spinach chloroplasts is reduced 57% by this 
analog at 0.1 mM (Park et al., 1960). The photoreduction of 3-phospho- 
glycerate is inhibited and this substance accumulates in the presence of the 
inhibitor. This inhibition is due primarily to an action on glyceraldehyde-3-P 
dehydrogenase leading to a deficiency of ribulose-l,5-diP, the CO2 acceptor. 
Carboxy dismutase is inhibited only slightly by 0.1 mM D-threose-2,4-diP 
but higher concentrations inhibit appreciably. This inhibitor may thus play 
a role in photosynthetic studies. 

Keleti and Telegdi (1959) examined various glyceraldehyde-3-P dehy- 
drogenases and found inorganic phosphate to stimulate the activity at low 
concentrations but to inhibit progressively above 20-30 mM. The inhi- 
bition of the yeast enzyme is competitive with both glyceraldehyde-3-P 
and NAD. Taylor et al., (1963) also found phosphate inhibition of the 
hydrolysis of p-nitrophenylacetate by glyceraldehyde-3-P dehydrogenase, 
the muscle enzyme being much more sensitive than the yeast enzyme. 
Indeed, the muscle enzyme is 56% inhibited already at 10 mM phosphate. 
Phosphate inhibitions of the glycolytic enzymes have seldom been consid- 
ered as playing a role in the regulation of carbohydrate metabolism. 

Enolase 

Enolase catalyzes the reaction 

D-2-phosphoglycerate :^ phosphoenolpyruvate 

and various analogs of these substances inhibit competitively (see tabula- 
tion) (Wold and Ballou, 1957). The following substances do not inhibit: 
D-lactate, D-glyceraldehyde-3-P, dihydroxyacetone-P, and glycerol-2-P. An 
inhibitor must have a carboxyl and a phosphate group, and the distance 
between them is of some importance, although not critical. The 3-methyla- 
tion of 2-phosphoglycerate does not lead to much reduction in binding, but 
3-methylation of the 3-phosphoglycerate reduces the afiinity by approxi- 



410 



2. ANALOGS OF ENZYME REACTION COMPONENTS 



Inhibitor 




{mM) 


D-Phospholactate 




0.35 


^-Hydroxypropionate-P 




0.45 


D - 3 - Phosphogly cerate 




0.45 


D-er;/<Aro-2,3-Dihydroxybutyrate 


-2-P 


0.60 


D-er2/<7»ro-2,3-Dihydroxybutyrate-3-P 


3.3 



COO- 

HCOP03H- 
H2COH 

D-2-Phosphoglycerate 



coo- 

I 

CH, 

H2COPO3H- 

/9-Hydroxypropionate-P 



coo- 

HCOH 

H2COPO3H- 
D -3 - Phosphogly cerate 

coo- 

HCOPO3H- 
HCOH 

CH3 

D-er7/<A/-o-2,3-Dihydroxy- 
butvrate-2-P 



COO- 

HCOP03H- 
CH3 

D-Phospholactate 

coo- 

HCOH 
HCOPO3H- 

CH3 

T>-erythro-2,Z-T)'\\ry- 
droxybutyrate-3-P 



mately 1.2 kcal/mole. This would indicate that 2- and 3-phosphates fit 
comparably as long as there is no bulky group on the 3-position, but when 
there is it sterically interferes with the bending of the 3-phosphate to fit the 
active site. 

Glucose and Glucose-6-P Dehydrogenases 

Beef liver glucose dehydrogenase is inhibited strongly and competitively 
by glucose-6-P (Strecker and Korkes, 1952) and fructose- 1,6-diP (Brink, 
1953 a), the K-b being 0.0025 mM and 0.062 mM, respectively. The K^ 
for glucose is around 31 mM at pH 7, so that if this represents a dissociation 
constant the phosphorylated compounds are bound much more tightly 
(around 5.8 kcal/mole). The rat liver enzyme is similarly inhibited: glucose- 
6-P at 0.015 mM inhibits 78% when glucose is 200 mM (Metzger et al, 
1964). Glucose-1-P also inhibits but is about one-tenth as effective as glu- 
cose-6-P. Ribose-5-P and fructose-6-P are also less inhibitory (Brink, 1953 a). 
The potency of the glucose-6-P inhibition is surprising and it is quite likely 
that it may be important in metabolic regulation or conserving glucose for 
phosphorylation. Not enough is known about this enzyme or the reaction 
mechanism to speculate on the nature of the interaction of these inhibitors. 



INHIBITORS OF CAEBOHYDRATE METABOLISM 411 

Yeast glucose-6-P dehydrogenase is inhibited competitively by glucos- 
amine-6-P but in this case the affinity for the inhibitor is not as great 
as for the substrate {K,,, = 0.058 mM, and K^ = 0.72 mM) (Glaser and 
Brown, 1955). Neither mannose-6-P nor N-acetyl-D-gkicosamine-6-P inhi- 
bits. Phosphate inhibits the yeast enzyme rather weakly but it is competitive 
with respect to NADP. The enzyme from Prototheca zopfii, however, is inhi- 
bited 13% by 0.07 mM phosphate, 43% by 0.14 mM, and 70% by 0.7 
mM (glucose-6-P = 3 mM and NADP = 0.67 mM) (Ciferri, 1962). There 
needs to be much more study of the inhibition of such enzymes if we are 
to understand the controlling factors of carbohydrate metabolism. 

Phosphopentose Isomerases 

The phosphoribose isomerase of alfalfa is j^otently inhibited by 5-phospho- 
ribonate and much less so by a variety of related substances (as shown in 
the accompanying tabulation) (Axelrod and Jang, 1954). Since ribose-5-P 



Inhibitor 


Concentration 
{mM) 


% Inhibition 


5-Phosphoribonate 


0.13 


50 


Glucose-6-P 


11.3 


32 


Phosphate 


25 


49 


Ribose 


11.5 





Ribose-3-P 


12.5 






was 2.5 mM, the 5-phosphoribonate is bound more tightly. However, even 
19 mM 5-phosphoribonate does not inhibit the growth of Leuconostoc mes- 
enteroides or Lactobacillus arabinosus in glucose medium; this could mean 
that the pentose phosphate pathway is not necessary for growth of these 
organisms or that the inhibitor does not penetrate readily. The Ophiodon 
elongatus muscle enzyme is also inhibited by 5-phosphoribonate but less 
potently (28% inhibition at 0.5 mM and 45% at 5 mM) (Tarr, 1959). 

The phosphoarabinose isomerase of Propionibacterium pentosaceum is not 
inhibited by D-ribose, D-xylose, D-arabinose, L-arabinose, ribose-5-P, and 
glucose-6-P at 10 times the concentration of arabinose-5-P (Volk, 1960), 
and the phosphopentose isomerase of EcJiinococcus granvlosiis hydatid cysts 
is not inhibited by glucose, fructose, glucosamine, glucose-6-P, fructose- 
6-P, and mannose-6-P at 4 mM (Agosin and Aravena, 1960). However, the 
latter enzyme is inhibited 51% by 1.2 mM dihydroxyacetone-P and 54% by 
4 mM dihydroxyacetone, so that conditions favoring accumulation of these 
substances might suppress the alternate pentose-P pathway, which could 
be important in the results obtained with iodoacetate or D-threose-2,4-diP. 



412 2. ANALOGS OF ENZYME REACTION COMPONENTS 

Glucose-6-Phosphatase 

The hydrolysis of glucose-6-P in rat liver preparations (probably microso- 
mal) is inhibited by glucose with a K^ around 29 mM (Langdon and Weak- 
ley, 1957). This inhibition is noncompetitive with respect to glucose-6-P, 
and it was postulated by Segal (1959) that the glucose competes with the 
second substrate, water, for its site; the transfer of phosphate could occur 
either to water or another molecule of glucose. If this is true, incorporation 

(1) Enzyme + glucose-6-P -^ enzyme-gIucose-6-P 

(2) Enzyme-glucose-6-P :^ enzyme-P + glucose 

(3) Enzyme-P + water ^ enzyme + Pt 

(4) Enzyme-P + glucose 5=^ enzyme-glucose-6-P 

of C^* from glucose-C^* into glucose-C^'*-6-P should occur and this was dem- 
onstrated. It was shown that with the appropriate rate constants the re- 
actions (l)-(4) (reaction (4) is of course the reverse of (2) and is included to 
visualize competition between water and glucose) lead to noncompetitive 
kinetics. A study of the inhibition of this incorporation by Hass and Byrne 
(1960) showed that glucose is the most potent inhibitor of various sugars 
tested. 

Miscellaneous Inhibitions 

Numerous other analog inhibitions of enzymes involved in carbohydrate 
utilization have been reported, some of which have been summarized in 
Table 2-21. These inhibitions along with those previously discussed point 
to many mechanisms for the control of glucose metabolism. The complex 
interplay between all the sugars and their phosphorylated derivatives with 
respect to the inhibition of various enzymes in the different available path- 
ways must always be borne in mind in work on intact cells. Many of these 
enzymes are also inhibited to varying degrees by inorganic phosphate; hence 
the level of phosphate can also be a regulating factor. Enzymes inhibited 
by phosphate include phosphodeoxyribomutase, phosphoribose isomerase, 
phosphoglucose isomerase, triosephosphate isomerase, glucose-6-P dehydro- 
genase, enolase, glyceraldehyde-3-P dehydrogenase, phosphorylase, trans- 
aldolase, transketolase, glucose-6-phosphatase, and ribulose-P carboxylase. 
In many instances appreciable inhibition is exerted by 5-20 niM phosphate. 
An interesting study of phosphate inhibition of transaldolase was made by 
Bonsignore et al. (1960) in which the following reactions were examined: 

Fructose-6-P + glyceraldehyde -> glyceraldehyde-3-P + fructose 
Sedoheptulose-7-P + glyceraldehyde-3-P -> fructose-6-P + erythrose-4-P 

Phosphate inhibits the first reaction competitively with respect to fruc- 



INHIBITORS OF CARBOHYDRATE METABOLISM 



413 



P5 



- -r. m 



O 3 



P-I (=^ 



O 



K 



M 










O 


Pm 


^ 






cc 


ci 


ci 


*^ 


Ph 


O 




i 


ci 






i 


CB 




o 


& 


CO 

o 


o 




0) 


^ 


+i 


o 


o 


5£ 


-^ 


o 


rt 


o 


O 


' 


3 






X5 


ro 




oj 






i-f 


t^ 


o 


O 


P5 



-tj 


<B 


e 


]5 


C 


•§ 


Xi 


'> 


g 


di 


o 


e 


P5 


m 


^ 



c 
■ft 



O O 



'^ 


t3 




c 


s 


cS 


0) 


0) 


o 


o 


>. 


t>-. 


o 


O 


^ 


^ 



w 





^ 






^ 






















Ph 
















'? 






<a 


<ii 


« 












I> 


Ph 




















rH 


t^ 


fij 


3 




fe 










o 




C/3 


'S 


H 














^ 


_o 


o 


e 


'bi 


"m 










o 


^ 


5 


»-H 


D 


o 


o 










"a 

cc 
O 


ST 


p 


o 


o 


O 


03 

O 


0^ 
O 
o 


0) 
CO 


'o 


1 




"o 


o 


o 


01 


^ 


r^ 


j2 


O 




^ 


^ 


13 




3 


Q 


Ph 


Ph 


'S 


■>> 


T. 


Lh 

o 


iC 


m 


m 


fe 


cc 


CO 


'^^ 


O 


X 


XI 


M 



fej 



Ph 



Q 



P5 



O 



-7=^. O -3 



W 



O 



414 2. ANALOGS OF ENZYME REACTION COMPONENTS 

tose-6-P but inhibits the second reaction noncompetitively with respect to 
sedoheptulose-7-P or glyceraldehyde-3-P. Since the ^/s are roughly the 
same for all inhibitions (around 60 vnM), a single binding site for phosphate 
was postulated. Phosphate prevents the formation of the transaldolase- 
dihydroxyacetone complex but does not interfere with the transfer of the 
dihydroxyacetone to its acceptor. 

Multivalent anions in general are inhibitors of glycolysis. The anaerobic 
formation of lactate from glucose in pigeon hemoly sates is depressed 84% 
by 40 mM sulfate, 64% by 20 mM phosphate, 100% by 4.2 mM oxalate, 
and 48% by 0.1% ribonucleate (Dische and Ashwell, 1955). The reactions 
from glucose -^ lactate are inhibited more strongly than from fructose-1,6- 
diP — > lactate; therefore there is inhibition previous to fructose-l,6-diP. The 
sequence from glucose -^ glyceraldehyde-3-P is inhibited strongly by sulfate 
and ribonucleate. Aldolase is inhibited about 20% by 40 mM sulfate but is 
not affected by oxalate. It was concluded that there are three sites of action, 
the strongest on hexokinase, next on glyceraldehyde-3-P dehydrogenase, 
and the last possibly on pyruvate kinase, the inhibitions being competitive 
with ATP or NAD. Actually little positive evidence was provided for these 
sites and other possibilities are just as likely; furthermore, competition with 
hexose phosphate and glycerol phosphate is also possible. In addition the 
complexing of Mg++ by these anions must be considered since several gly- 
colytic enzymes are activated by this ion. 

It will suffice to mention four additional interesting examples of analog 
inhibition on this phase of metabolism. Cataractogenic sugars and polyols 
inhibit lens mutarotase while noncataractogenic sugars do not; the K,'s are 
4 mM for galactose, 15 mM for xylose, and 100 mM for sorbitol (Keston, 
1963). Despite the weak inhibitory activity of sorbitol, there is a large 
amount in the lens in certain conditions, such as diabetes (perhaps around 
30 mM). Mannose is quite toxic to honeybees; of bees offered 1 M mannose 
solution, 50% were dead in 90 min and over 90% in 3 hr (Sols et al., 1960 a). 
It was found that bees have a hexokinase very active toward mannose 
coupled with a negligible amount of phosphomannose isomerase, so that 
mannose not only may interfere with phosphorylation of glucose and fruc- 
tose, but many accumulate as mannose-6-P, which could disturb glycolysis 
in a number of ways. Xylose appears to be in some manner a specific inhibi- 
tor of photosynthesis, since Chlorella propagation is not inhibited by 0.5- 
1.5% xylose when glucose, fructose, or mannose is present (thus it is not 
inherently toxic), but under photosynthetic conditions the cells rapidly lose 
their color and ability to divide, an effect that can be reversed by glucose 
(Hassall, 1958). It was postulated that xylose may compete with xylulose- 
5-P for an enzyme in the transketolase pathway and block photosynthesis; 
on the other hand, a phosphorylated product may be the active inhibitor. 
Analogs without the usual hexose structure may also inhibit glycolysis and 



GLYCOSIDASES 415 

the pentose-P pathway. Sahasrabudhe et al. (1960) in looking for carcino- 
static analogs found that thiophene-2,5-dicarboxylate, which might be con- 
sidered as an analog of substances such as ribose-5-P, inhibits the formation 
of C^^Og from glucose- 1-C^* and glucose-6-C^'* around 43% (concentration 
unspecified) in tumor tissue, and suppresses the growth in vivo of rat sar- 
coma. No evidence was given as to the site of action and so the assumption 
is tenuous, but the concept of using heterocyclic substances analogous to 
the furanose and pyranose structures may be important. 

GLYCOSIDASES 

This large group of enzymes hydrolyzing the glycosidic bonds of simple 
glycosides, oligosaccharides, and polysaccharides has been studied for many 
years and it is not surprising that numerous instances of analog inhibition 
have been observed. Most of the reports, although important in themselves, 
do not lend themselves to interpretations on the molecular level; some of 
the data have been briefly summarized in Table 2-22. Most of the inhibi- 
tions are relatively weak but a few could definitely be significant in meta- 
bolic regulation. The a, (3 configuration of the inhibitor is seen to be im- 
portant in some instances. However, the enzymes are seldom completely 
specific for the a- or /?-forms and the affinities often diff"er very little, as 
with maltose transglucosylase (amylomaltase) from E. coli where (see ac- 
companying tabulation) the binding difference between the u- and /5-glu- 



Inhibitor 


(mM) 


Relative 


(k 


- AF of binding 
cal/mole) 


/3-Methylmaltoside 


2.5 






3.70 


a-Methylglucoside 


8 






2.98 


^-Methylglucoside 


10 






2.84 


a-Phenylglucoside 


10 






2.84 


/S-Phenylglucoside 


30 






2.16 



cosides is 0.14-0.68 kcal/mole (Wiesmeyer and Cohn, 1960). It appears that 
the hydroxyl groups on C-2, C-4, and C-6 of ring A are involved in the 

CH,OH CH,OH 

-Ott XT J O 



ring A 




416 



2. ANALOGS OF ENZYME REACTION COMPONENTS 



O 



H K 



P5 



^-1 



s^ 



ft ^ rS 

c 



g 

SI 

C 



S ^ 



c 

-S3 
o 
O 

C 
>> 



^ ^ C fl 



o o 



w 



O Q 



rs "c 



o o 



o o o o o 



(B 03 -u o M o 

O O O C «', -'-' 

^ ^ le c« '^ 5 , 

O O O S ^ tR «x 



"? w 


o 








^ 


<c 








'm 




2 o 


fl 








"m 


^ 


OJ 






O 




i5 o 


a 


2 

o 


s 

O 

eg 


CO 

O 

C 

3 




'35 
o 

& 


O 

o 


a) 
O 
'n 


CO 


'be 


i 


=i. a 


'ai. 


H^ 


H 


^ 




1— 1 


S 


H 


a 


P5 



Cl^ 



GLYCOSIDASES 



417 



W 



w 



TS 


lO 


a 


Oi 


c3 


^i;;;^ 


2 


T3 




!» 


K} 


c< 


g 


c 




<» 


o 


s 


^ 


o 


J 


D5 



in ic »o >c 



— ^ -^ c 



o o o 






O >< CK S O 




































<» 


03 


















C 


^ 














">> 







8 






CI 








^ 











£ 












r5 


_s 











^ 


5 


















r5 


© 





c« 


a 

















w; 




C 


C 




'2 


"0 






c 


O 


3 








Ch 





^ 


Q 





■4i 


_o 














_o 








■3 




S 


3 


02 




<» 




_3 




'S 






OJ 




-C 


"S 






^ 








C 





Pn 












O ra o o 



^ -=; ^ O 



O ^3 S M 










^ 








_5 




'S 








^ 




^ 






03 












O! 


">> 




'>. 









-»^ 


03 


■t^ 


03 


03 





03 



a 


8 


c 


8 


^ 


<^ 


ff 


<1 


s 


Is 




^ 


S 


fe; 


eS 


3 
ft* 



'00 

o 

o 

O 
B 



Q::^. 


^^ 


^ 


10 










03 

n3 


s 


S 


43 


« 


^ 




4 







c 


M 


o3 


t-i 


(^ 


->J C3 


-S 


t? -2 





>> 


^^ 


c 


03 


03 


r2 


r-< 




ft 





P 


-u 



^ s, 



03 cS 
CO 



C? 



►«; S 



418 



2. ANALOGS OF ENZYME REACTION COMPONENTS 



P5 



1^ '2 



. o ^ 



-^ 



H 2 



2 Xh 



O O O O O (M 



■S 1^ 



03 £ 






o 'Ti 



y o 






>■- 




j3 


<x> 


0) 


73. 




'w 


"ft 


o 


o 


-u 


s 


o 


4^ 


riS 






g 


'rt 



t*5 



t*3 



OJ rfr 



^^(5^3^^^:^ 53 






o 



GLYCOSIDASES 419 



.s .s ^ ^ 



^ Pm ^ 



<N O O ^ ■* 

O t^ O fO o 



Q O O O O O O 



I 03 

-5 'cS 









c5 K-^ c3 



jg.a: fe;=';^ra fe;''<lj<t3 





(M 


o 






6 


c 
o 






T3 


o 






a 

4^ 







o 


^ 


c 


c^ 


<u 


X 


o 


s 

03 




o 

T3 


o 



0) y< 



^ • o 






^:2 



fl -= O G cS Qi 

2 -^-l 2 ->. = 






OJ 03 






•<S J 



e8 


r-2 


eS 


S 


cs as 


r2 




M 03 




«i '3 




■ 'S 


^S 


«i. 'O 


2--S 


'?>;§ 


>> eS 


>» c6 


-»^ m 


4J r- 


*^ CO 


© O 


43 S 


® O 


o o 


■^ eS 


O o 


< ,B 


*? O 


< ^ 


fe, '^ 


fe; ^ 


)^ ^ 



420 



2. ANALOGS OF ENZYME REACTION COMPONENTS 



P5 



!^ 



- a 



>> o S, 

c 



C- -H 



o 



o 



o 



o 



o 



-u 

s 


IS 


o 

o 








C<1 


O 

c 




'^ 


5 




o 


'2 


3 








6 


o 




o 


o 




c3 








Is 




tj 


o 




'2 


eS 




^ 


a 










'a 


'bjb 


0) 


's 


Is 


« 


c 


CO 

O 

o 


< 






a 


o 


X 

o 


c 
o 

o 

-2 


c3 


X 
O 

t3 


o 
o 



c ce (» 

£ ■>. a 



a 



be 


CO 


oi. 


c3 
"2 






T. 


'c 




'a 


o 


a 


< 


S 



o o o o o o 



03 M^ ^ 



_3 




^ 


cc 














O 

o 


^ 




$ 


a) 


o 


0) 




2 

o 




95 


^ 


3 






cS 




3 


5 


<^ 


a 


s 


><1 



a 



GLYCOSIDASES 



421 



S 

s 

o 

c 

^ O 



^ 



."" --I 



O >M 05 C5 
CO ■<* IC -^ 



1^ 05 ic 00 

>o CC Oi t^ 

^ 1— I Oq 



•c:. c 



o 



aj o o' © 



O c« O t< -*i 

-T 1 2 § 3 

o O fi< oj s 



> 

c 

CO 55 



o 



o 



t3 <D aj 

2 ^ :« 



te 


73 


Xh 


eS 


C 


■ ■ 


43 


c3 


c 


Ph 


ro 


73 




'J 


c 


"s 


CO 


cS 


eS 


45 




4m 


o 


-C 


o 


v 


3 
1-5 


S 


(1h 



Cm 



•^ .-^ O IC t- IC 



o o o o 



73 Ti 



h 


3 


_s 


0) 


K 


OJ 


13 


o 


^ 










O 


-l-i 


^Ji 


"bX) 


"m 


Q 


o 


o 


Eh 


o 


15 


i 




o 
S 




5 


3 


o 





■^i. 


P^ 


a 


■ai. 


c? 


=Ci. 






_o 


43 


'-3 


^ 






'2 


.,— ^ 


;s 


^ 


_c 


a 






^ 




o~" 


c 


bl 


_o 


OJ 


'-3 


73 


f« 


G 






v 




o 


aJ 


fl 


_> 


o 


■-13 


o 








-2 


ft 


cS 




bi 


c 


-kJ 


o 


oc 


o 


42 


i=; 


3 


o 


01 


c 


(X> 


11 


43 


H 


o 


C 


^ 


_o 




■-!3 


b 




o 


'2 


<» 


:a 


> 


.g 






•^ 




'-3 

4J 


o 


& 


cc 


g 


^ 




-fi 


O 


c3 


o 


o 




• ^ 


11 


73 



o 



73 

a 

c3 



'bi T) 



.2 73 



a -i5 



42 4) 

-I 



ft 42 



4) 



.o <u 



422 2. ANALOGS OF ENZYME REACTION COMPONENTS 

binding since alteration of the positions, as in a-methylmannoside or a- 
methylgalactoside, abolishes the inhibition, and omission, as in a-methyl- 
xyloside, also prevents binding. The substitution of a methyl group on C-1 
of ring B (/?-methylmaltoside) does not interfere with the binding, since 
this analog has approximately the same affinity for the enzyme as the sub- 
strate maltose. Alteration of the glycosidic link from the o;-l,4 in maltose 
to the /?-l,4 in cellobiose does not result in an inhibitor, and other changes 
(as in trehalose, sucrose, dextran, or lactose) likewise reduce the affinity. 
The requirements for binding may be summarized as (1) a glucose-like con- 
figuration of hydroxyl groups in ring A, (2) a glycosidic link of the a- 1,4 
type, and (3) a widely variable glycosidic group. 

An a-mannosidase of Streptomyces griseus with a-phenylmannoside as the 
substrate is inhibited competitively (probably) by several sugars and gly- 
cosides (Hockenhull et al., 1954 c) and the relative binding energies have 
been estimated on this basis (see tabulation), although the reliability of 



Inhibitor 


% Inhibition at 
50 mM 


Relative — Zli^ of binding 
(kcal/mole) 


a-Methylmannoside 


96 


5.23 


Mannose 


83 


4.25 


Cellobiose 


82 


4.20 


Maltose 


73 


3.88 


a-Methylglucoside 


14 


2.15 


Xylose 


11 


1.98 


Arabinose 


10 


1.91 


Sucrose 


7.5 


1.72 


Glucose 


5 


1.45 


Fructose 


5 


1.45 


Ribose 


4 


1.31 


Mannitol 


2 


0.87 


Rhamnose 





— 



these figures is quite low due to variations between experiments. The man- 
nose configuration seems to confer strong binding, as expected, but the high 
inhibitory activities of maltose and cellobiose {a and /5 glucosides, respec- 
tively), especially in view of the weak inhibitions produced by a-methyl- 
glucoside and glucose, are unexpected and perhaps indicate that these sub- 
stances are oriented on the enzyme surface in a different way than the 
substrate, although there is no necessity to postulate an irreversible com- 
plex as did Hockenhull and his co-workers. The active centers for such 
enzymes probably possess several binding groups for the hydroxyls arrang- 
ed in a certain pattern; glycosides other than the substrate could conceivably 



GLYCOSIDASES 423 

interact satisfactorily with this pattern by appropriate translation or ro- 
tation of the molecules (indeed, the other sides of these roughly planar 
molecules could possibly fit the pattern in some cases). 

One of the most interesting studies of analog inhibition was reported 
by Halvorson and Ellias (1958) for the a-glucosidase of Saccharomyces 
italicus, the data from which are given in the accompanying tabulation. 



Inhibitor 


{mM) 


Relative — AF oi binding 
(kcal/mole) 


a-Phenylgliicopyranoside* 
Glucose 


0.3 
1.2 


5.00 
4.15 


a-Butylglucopyranoside 

a-Ethylthioglucopyranoside 

Turanose* 


1.9 

4 

9.5 


3.86 
3.40 

2.87 


Isomaltose 


11 


2.78 


Xylose 
Sucrose* 


12 
23 


2.72 
2.32 


/S-Methylmaltoside* 

a-Ethylglucopyranoside 

Maltose* 


37 

45 
81 


2.04 
1.91 
1.55 


a-Ethylthioglucofuranoside 


86 


1.51 


Arabinose 


110 


1.36 


a-Methylglucopyranoside 


400 


0.56 



Several of the substances (marked by *) are also substrates; maltose, in 
fact, is probably the natural substrate. Inversion of hydroxyl groups on 
C-2 and C-4, and substitution or oxidation at C-6, abolish the affinity. 
Increasing the size of the aglycone groups increases the affinity. The rather 
potent inhibition by glucose is surprising and indicates that the alkyl agly- 
cone groups actually reduce the affinity. This might mean that the C-1 
hydroxyl group can be involved in the binding; a small substituent prevents 
this but the binding increases as the alkyl group is lengthened. 

Inhibition of (3-Glucuronidases by D-Glucaro-1,4-lactone 
and Related Compounds 

These widely distributed enzymes catalyze the hydrolysis of /5-glucur- 
onides occurring naturally in plants and animals, and in addition may play 
a role in mucopolysaccharide metabolism. They are not involved in the 
glucuronide syntheses of detoxification, these reactions being catalyzed by 
glucuronyl transferases. There is evidence that /^-glucuronidases are related 
in some manner to growth and the activities of certain tissues, and it was 



424 



2. ANALOGS OF ENZYME REACTION COMPONENTS 



to establish their role in metabolism that Karunairatnam and Levvy (1949) 
searched for specific inhibitors. The most potent of the analogs tested is 
glucarate (K^ = 0.06 mM, and K,^, = 3.5 for phenyl-/?-glucuronide) but 
subsequent studies in different laboratories showed great variability in ac- 
tivity. The problem was solved by Levvy (1952), who found that the ac- 
tive inhibitor is glucaro-l,4-lactone,* which is formed from glucarate and 
occurs to varying extents in different preparations. The tabulation shows 



Inhibitor 


(mM) 


Relative — AF oi binding 
(kcal/mole) 


Glucaro- 1 ,4-lactone 


0.00054 


8.87 


3-Methylglucaro- 1 ,4-lactone 


0.13 


5.50 


Glucarate 


0.17 


5.33 


Glucaro-3,6-lactone 


0.48 


4.70 


Glucuronate 


1.6 


3.96 


Galacturonate 


6.0 


3.14 


Galactarate (mucate) 


6.0 


3.14 



the inhibitor constants obtained on mouse liver /^-glucuronidase hydrolyz- 
ing phenolphthalein-/?-glucuronide, and the particular potency of the glu- 
caro- 1,4-lactone is evident."^ No inhibition at 1 mM is observed with man- 
narate, mannaro-l,4-3,6-dilactone, 4-methylglucuronate, and ascorbate, or 
at 10 mM with 3-methylglucarate, glucurone, 3-methylglucuronate, man- 
nuronate, mannurone, and 2-keto-L-gulonate. Heat and acid treatments of 
glucarate and galactarate increase the inhibitory activity greatly and it is 
possible that the dicarboxylates are completely inactive. The inhibition by 
glucaro- 1,4-lactone is competitive and the high potency probably due to 
the structural similarity between the inhibitor and the /5-glucofuranuronide 
form of the substrate. The hydroxyl configuration in the furan ring is very 
important since the glucaro-3,6-lactone is bound much less tightly to the 
enzyme, and the 3-OH must also be involved since methylation reduces 
the binding by over 3 kcal/mole. These rather large energy differences argue 



* This substance has also been called saccharo- 1,4-lactone, 1 ,4-saccharolactone, 
glucosaccharo- 1,4-lactone, and other names. The generic name for the dicarboxylic 
acids derived from sugars is saccharic acid and that specifically from glucose is glu- 
cosaccharic acid. However, it seems that glucaric acid is used most commonly today 
and this terminology is more consistent with the naming of the monocarboxylic acids, 
so that the lactones will here be named accordingly. 

t The figures in this and other reports for inhibitors such as glucarate, galactarate, 
glucuronate, and related acids may reflect to varying degrees the presence of lactones, 
since, due to the high potency of the lactones, only small amounts need be present. 



GLYCOSIDASES 



425 



5 






1 

o 








o- 

/ 


'^ 


SB 

-o 


murom 

tone 

one) 


/ 


^ 


r-O 


O— A. 
O- 






fur; 
lac 
cur 


O 

\ 






-^o 


-K 


-Gluco 
3,6- 
(glu 


\ 

o- 




-K 


1 


B3 




oa 


k" 


s 




o 








2 






s 








a 

















o 

as 
u- 


s 
o 

1 

-u- 


EC 
1 

-o- 


a 

o 

1 

-u- 


O 

1 

-o- 


8 

-o 






1 


1 

o 


1 


1 




3 4) 

1 



'o O X O O O 

5 I I I I x" 
o-u-u-u-u— u 

I I I I 

X O X X 
X 



-22 


8 

o- 


X 

o 

1 

-u- 


X 
X o 

1 1 

-u-u- 


?8 

-u-u 


3 <U 

5^ 




1 

X 


1 i 

O X 

X 


1 

X 



* 


2 


'? 


o 


a 


rt 


u 


0) 


3 


v 



OiS. 



426 2. ANALOGS OF ENZYME REACTION COMPONENTS 

against van der Waals' forces being a major factor in the hydroxyl inter- 
actions, and suggest hydrogen bonding. The terminal carboxylate group 
also participates, since glucaro-3,6-lactone is much more inhibitory than 
glucurone. (See formulas on page 425). 

Some inhibitor constants for various /^-glucuronidases are shown in Table 
2-23. Not enough /^-glucuronidases have been tested with glucaro-l,4-lac- 
tone to determine the variations in susceptibility, but from the limited 
data it appears that the animal enzymes are quite sensitive whereas the 
plant enzyme baicalinase is much less readily inhibited. It is clear that 
glucaro-l,4-lactone is one of the most potent inhibitors known and that 
none of the other analogs so far examined on /^-glucuronidase is comparable, 
although galactaro-l,4-lactone is undoubtedly a very effective inhibitor. It 
is interesting that the /?-galacturonidases of limpet and preputial gland are 
approximately as susceptible to these lactones as are the /^-glucuronidases 
(Marsh and Levvy, 1958). 

Limpet or-glucuronidase shows quite a different pattern of inhibition (see 
accompanying tabulation for inhibitions at 5 mM with phenyl-a-glucur- 



Inhibitor 


% Inhibition 


Relative — AF of binding 
(kcal/mole) 


Glucuronate 


68 


3.73 


Mannuronate 


43 


3.09 


Galacturonate 


28 


2.68 


Glucurone 


9 


1.84 


Xylono- 1 ,4-lactone 


22 


2.48 


Glucono- 1 ,5-lactone 


7 


1.67 


Arabono- 1 ,4-lactone 


4 


1.31 


Glucono- 1 ,4-lactone 


3 


1.13 


a-Glucuronate- 1 -phosphate 


55 


3.39 


/?-Glucuronate- 1 -phosphate 


22 


2.48 


Menthyl-a-glucuronide 


55 


3.39 


Menthyl-/S-glucuronide 


6 


1.57 


Borneol-a-glucuronide 


41 


3.04 


Veratroyl-/3-glucuronide 


9 


1.84 



onide at 1 raM with relative binding energies calculated on the basis of 
competitive inhibition). a-Glucuronides are more inhibitory than /5-glucur- 
onides, as expected, but no particularly potent inhibitors were found {Kj^ 
for glucuronate is 0.54 mM and for galacturonate 4.7 mM). The following 
do not inhibit: mannurone, glucaro-l,4-lactone, mannono-l,4-lactone, and 
galactarate. The difference between the a and /? enzymes with respect to 
glucaro-l,4-lactone is especially striking. 



GLYCOSIDASES 



427 



A 






00 -^ (M — O -* 

• c 

O -H CD — O O 



Id 

_^ CO <M C5 05 

rZT o c-i t^ ^ CO 

c 



c c c c 



ooooco-*-^-^ — 



'H :2 



O GO CO ., 

O Oi o (^i 

o o o «^ 

O O O O >0 CO 

o o o o ^ -<t 



P -:, ci:i 

>» = -^ ^ 

■*^ Si A lA 






03 03 eS ^^ -*^ 






o o o o o o 



> a 



a O 



o ^ p 

> :§ o 



3 S O ^ 



T3 7:= 
« o 



-r TTn =5 3 -T _. -Q. 






^ Is « C3 5 



^ 6 rt "ci 5 g P 



2 ^ "3 ^ le -2 Is 
S O a O O O O 



0) 






-r) 


r2 


'6 
















'S 


*2 


"S 






o 


o 


o 






^ 


tH 


«H 






3 


3 


3 






O 


CJ 




o 






_3 


'S 




"3i 


bC 


tiC 












'S 


^^ 


oi. 


^ 


**?- 
%. 


o 

3 
o 


s 


'S 




-a 
o 
u 


>> 






^ 


N 


"^ 


"eg 


'm 


c 


j3 




<ai. 


■>* 


^ 

Bf 




T. 


ec 


'o 


"o 


3 
a 


3 
<» 




3 


3 

4) 


a 


Ph 




Pl, 


PlH 



M 


M 


oi. 


<ai. 


^ 


r- 










r2 


« 






'3 


13 


jil 


43 


-►i 


4i 


^ 


J3 


a. 


ft 






"o 


o 


3 


3 


05 


(» 




43 


^ 


Ph 



M 



GQ 





is 




^ 


+3 






OJ 




^ 




3 


'^^ 




<u 


-ti 


^ 


e 


cS 




^ 


05 


^ 



3 C 



03 


^ 


(C 


3 


"^ .^ 




^ 


O 






ai 




3 


0) 


-tj 


-S 


-0 


CJ 


c 




<A 


^ 






03 


.s 


OJ 




3 


' — V 


Q 


"3 



-3 CO 

2 "o 



^H 



cS o 



f* cS 



eS 73 



c 


<1) 


o 


ft 






cS 


3 


frt 


o 










4J 


Tl 






t^ 


tS 




ft 



428 2. ANALOGS OF ENZYME REACTION COMPONENTS 

It was previously stated that /^-glucuronidases do not seem to partici- 
pate in glucuronide synthesis in tissues and the evidence is mainly the 
lack of inhibition by glucaro-l,4-lactone usually observed (Levvy and 
Marsh, 1960). For example, Lathe and Walker (1958) found no effects of 
2 mM glucaro-l,4-lactone on the formation of o-aminophenylglucuronide 
or bilirubin-glucuronide in liver suspensions. However, Fishman and Green 
(1957) reported that glucarate potently inhibits glucuronyl transfer (50% 
inhibition by 0.05 mM), and Sie and Fishman (1954) found glucarate and 
glucurone to inhibit glucuronide synthesis in liver slices. It is not definitely 
known if these inhibitions relate to a /^-glucuronidase or a transferase, but 
the question of the role of the /^-glucuronidases in glucuronide synthesis 
should probably be left open. 

The urinary bladder cancer that occurs in people employed in certain 
industries may be related to the formation of glucuronides of aromatic 
amines in the body and their subsequent hydrolysis by urinary /5-glucur- 
onidase. The oral administration of inhibitory analogs to patients is a 
possible approach to the prevention of such cancers. Boy land et al. (1957) 
found that the urinary enzyme can be inhibited by administration of glu- 
conate and glucarate at a dose of 10 g/day, but the most potent inhibition 
is produced by glucaro-l,4-lactone, 73% inhibition occurring from 1.5-2 
g/day and 90% inhibition from 4 g/day. Nevertheless, these inhibitions are 
less than expected and the urinary pH was reported later to be an important 
factor, the inhibition decreasing as the pH rises above 6 (Boyland et al., 
1959). Inhibition of liver /^-glucuronidase in mice by the administration 
of glucaro-l,4-lactone at doses from 50 to 800 mg/kg (17% to 76%) was 
found by Akamatsu et al. (1961). The maximal inhibition occurs at 30-60 
min and after 2-4 hr most of the activity has returned. Similar results 
were found in rat liver and kidney. Since /^-glucuronidase and /?-iV-acetyl- 
glucosaminidase hydrolyze products from chondroitin and hyaluronate and 
may participate in the metabolism of connective tissue mucoproteins, and 
since certain tumors have relatively large amounts of these enzymes, Carr 
(1963) administered the two inhibitors — glucaro-l,4-lactone and 2-acet- 
amido-2-deoxygluconolactone — to mice bearing Tumor 2146. At 150 mg/kg 
these substances are nontoxic and cause regression of the tumors. It was 
suggested that the inhibitors prevent the penetration of the tumor cells 
through the intercellular cement substance, but this is admittedly only a 
tenuous hypothesis. Whatever the explanation, it represents an interesting 
approach to tumor chemotherapy. 

Inhibition of Various Glycosidases by Glucono- and Glucaro Lactones 

The inhibition of the /^-glucuronidases by the saccharo-l,4-lactones sug- 
gested that the /^-glycosidases might be inhibited by the aldonolactones, 
and this was found to be so by Conchie (1953, 1954). A sheep rumen /?- 



PYRUVATE METABOLISM 429 

glucosidase, possibly involved in the digestion of cellulose, is strongly inhi- 
bited by glucono-l,4-lactone (^, = 0.094 mM) and glucono-l,5-lactone 
(^. = 0.091 milf ) in a competitive manner, the affinity of the enzyme for 
these analogs being about 10 times that for the substrate o-nitrophenyl- 
/?-glucoside, whereas gluconate itself has no action. Ox liver /?-galactosidase 
is inhibited much more potently by the galactono- and fucono-l,5-lactones 
than the corresponding 1,4-lactones, and this can be readily explained on 
the basis of the relationship to substrate configuration (Lewy et al., 1962). 
A remarkable degree of specificity is exhibited by the a and /? glycosidases, 
inhibition usually resulting only from the corresponding aldonolactone in 
a number of enzymes from different sources (Conchie and Lewy, 1957). 
For example, limpet a-mannosidase is inhibited markedly by mannono- 
1,4-lactone but not by the glucono-, galactono-, arabono-, or xylonolactones. 
On the other hand, galactono- 1,4-lactone is specific for /?-galactosidase 
(Conchie and Hay, 1959). Cellulytic rumen enzymes are inhibited to vary- 
ing degrees by glucono- 1,4-lactone, depending on the substrate chain length; 
hydrolysis of cellobiose is inhibited 99%, of cellotriose 90%, and of cellote- 
traose 75% by 0.5 mM (Festenstein, 1959). Glucono- 1,4-lactone also inhibits 
a yeast debranching isoamylase (Gunja et al., 1961) and a mammalian 
thioglycosidase (Goodman et al., 1959), so that this type of inhibition ap- 
pears to be widespread. 

Very potent and specific inhibitions are exerted on N-acetyl-/?-glucos- 
aminidase and N-acetyl-/?-galactosaminidase by the corresponding lactones 
(Marsh and Lewy, 1957; Findlay et al, 1958). The former enzyme from 
epididymis is inhibited by N-acetylglucosaminolactone competitively, with 
K- = 0.000072 mil/, but the limpet enzyme is less sensitive, the K^ being 
0.027 mM. The N-acetyl-a-glucosaminidase, on the other hand, is inhibited 
much less. The acetyl group is essential for the inhibition and cannot be 
replaced by other acyl groups. Certainly the use of the corresponding lac- 
tones for specific and potent inhibition of the glycosidases has been one of 
the most successful endeavors in the application of analogs. 



PYRUVATE METABOLISM 

The usefulness of a direct and specific inhibitor of pyruvate utilization 
is obvious and the natural occurrence of certain pyruvate analogs makes 
this field of inhibitor study an important one, but relatively little work 
has been done. Fluoropyruvate will be taken up with other fluorinated 
compounds, such as fluoroacetate, in a separate chapter. Phenylpyruvate 
is the best studied of the other pyruvate analogs. It is readily formed from 
phenylalanine and is usually decarboxylated to phenylacetate, although 
some may be reduced to phenyllactate. In phenylketonuria (phenylpyruvic 
oligophrenia), the accumulation of phenylpyruvate could be responsible for 



430 2. ANALOGS OF ENZYME REACTION COMPONENTS 

some of the disturbances by interfering with pyruvate metabolism, but no 
work has been done on the metabolic changes in tissues resulting from 
such levels of the analog. Hydroxyphenylpyruvate is similarly formed from 
tyrosine. 




(. ))— CH2— CH- COO" 
Phenylalanine 



The formation of acetoin from pyruvate in Streptococcus fecalis is inhibited 
13% by 1 mM and 75% by 10 mM phenylpyruvate, whereas the formation 
of acetoin from acetoacetate is unaffected by 10 mM (Dolin and Gunsalus, 
1951). Pyruvate oxidase is inhibited to about the same degree. Several 
anaerobic pyruvate pathways are inhibited by phenylpyruvate in several 
bacteria and yeast, including the formation of acetoin, the phosphoroclastic 
reaction, and decarboxylation, whereas the oxidative metabolism of pyru- 
vate is not so readily affected (Watt and Werkman, 1954). The concentra- 
tions of pyruvate and phenylpyruvate used (120 mM) are unfortunately 
too high to be physiologically significant, but further study on extracts 
of Aerobacter aerogenes showed competitive inhibition of acetoin formation 
with K,„ = 123-197 mM, and K^ = 0.59-1.1 mM, so that in this case phen- 
ylpyruvate is bound to the enzyme much more tightly than pyruvate. 
The reduction of hydroxypyruvate to glycerate by glycerate dehydrogenase 
from spinach is inhibited by phenylpyruvate (5%), pyruvate (34%), and 
bromopyruvate (52%) at 10 mM (Holzer and Holldorf, 1957). These inhi- 
bitions are competitive but rather weak. 

The most pertinent study with respect to blocking an important pyruvate 
pathway is that of Gale (1961) on yeast pyruvate decarboxylase, which 
reports the inhibitions given in Table 2-24. The following compounds are 
inactive: pyruvic ethyl ester, oxalacetate, propionate, phenyllactate, phen- 
ylalanine, acetamide, oxamate, and oxalate. One might infer that (1) the 
C=0 group is necessary for inhibition (reduction or substitution abolishes 
activity), and (2) the COO" group is necessary for strong inhibition (amides 
and esters inactive). The nature of the R group in R — CO — COO" can vary 
quite widely and it is difficult to correlate structure with activity; for 
example, it is surprising that ketomalonate is bound so well and chloro- 
pyruvate relatively poorly, and that oxanilate is bound so very weakly. 



PYRUVATE METABOLISM 



431 



Table 2-24 
Inhibition of Yeast Pyruvate Decarboxylase by Analogs" 



Inhibitor 



Structure 



Concen- 
tration 
(mM) 



Inhibition 



Relative 
activity 



Glyoxylate 



o-Nitrophenyl- 
pyruvate 



Ketomalonate 



/)-Hydroxyphe- 
nylpyruvate 



Phenylpyruvate 



Chloropyruvate 



2, 3-Butanedione 



O 

II 
H-C-COO 



i> 



CH, 



O 

II 

c-coo 



o 

II 

ooc-c-coo 



HO 



CHj— C-COO' 



O 

II 

CH,— C-COO 



O 

II 
CI— CH,— C-COO 



O O 

II II 
HoC C C CHo 



0.023 


33 


0.45 


95 


0.09 


61 


0.14 


70 


0.23 


71 


0.34 


82 


0.45 


62 


1.1 


71 


1.1 


68 


2.3 


80 


0.45 


27 


2.3 


73 



32 



17 



12 



2.9 



56 



1.0 



0.14 



Oxanilate 



a-Ketoglutarate 



O 

II 
NH- C-COO' 



OOC-CH,— CH,— C— COO 



27 



31 



24 



0.05 



0.012 



o Pyruvate concentration 4.5 mW and preincubation with inhibitor 15 mln. Relative inhibitory 
activity calculated from the formula t/(l - i)(I) and roughly would be inversely proportional 
to Kf for noncompetitive Inhibition. (From Gale, 1961.) 



The inhibition by ketomalonate is more surprising when one considers that 
oxalate and oxalacetate are inactive. The kinetics of the inhibitions pro- 
duced by the five most potent substances were studied in greater detail 
and a typical noncompetitive mechanism was established. However, pre- 
sence of the substrate prevents development of the inhibitions, suggesting 
that the inhibitors combine at the substrate site. These results indicate an 
irreversible or pseudoirreversible type of inhibition, and it was indeed de- 
monstrated that the inhibitions are all progressive with time. Phenylpy- 
ruvate is the only inhibitor whose effects are even partially reversible by 



432 2. ANALOGS OF ENZYME REACTION COMPONENTS 

dialysis. It was considered that the true inhibitors are the aldehydes cor- 
responding to the keto acids, and it was shown that COg is evolved from 
phenylpyruvate and p-hydroxyphenylpyruvate but not from the other 
three inhibitors. Although this does not completely invalidate the aldehyde 
proposal, it makes it unlikely. On the other hand, it is possible that the 
tight complex is formed with some intermediate prior to decarboxylation. 
Hydroxypyruvate is decarboxylated by yeast decarboxylase at a rate 
1/76 that of pyruvate and strongly inhibits the pyruvate reaction (Holzer 
et al., 1955 d). It was thought that hydroxypyruvate might exert some 
regulatory function on pyruvate metabolism in yeast. Formaldehyde and 
acetaldehyde inhibit pyruvate decarboxylase but the mechanism may not 
be competitive with substrate (Bauchop and Dawes, 1959). The oxidation 
of pyruvate in rat kidney slices is inhibited around 70% by 20 mM meso- 
tartrate, but not at all by d- and DL-tartrates (after correction for the inhi- 
bition of endogenous respiration) (Quastel and Scholefield, 1955). This 
inhibition is completely reversed by fumarate and malate. However, when 
mitochondria were examined it was found that there is little direct effect on 
pyruvate oxidation but a rather strong inhibition of or-ketoglutarate oxida- 
tion, which is progressive with time. In the presence of bicarbonate, pyru- 
vate is oxidized and this is inhibited by meso-tartrate. It was assumed 
that the inhibition is upon the incorporation of CO.2, and it was stated that 
meso-tartrate is a specific inhibitor of pyruvate oxidation in slices of rat 
kidney cortex, a conclusion that would seem to be unjustified since so few 
systems were tested. 

LACTATE METABOLISM 

A specific inhibitor of lactate dehydrogenase would be useful in studying 
the effects of a block of glycolytic lactate formation and for determining 
the role of this enzyme in the functioning of tissues. One of the most in- 
teresting lactate dehydrogenase inhibitors is oxamate, first reported by 
Hakala et al. (1953), who stated that it is the most potent inhibitor of many 
lactate and pyruvate analogs examined. The inhibition is competitive with 

O 

+H3N— C— coo- 

Oxamate 

respect to pyruvate,* noncompetitive with respect to lactate and NAD, 
and uncompetitive with respect to NADH (Papaconstantinou and Colo- 

* Examination of the double reciprocal plot reveals that the situation is not purely 
competitive but mainly so. 



LACTATE METABOLISM 433 

wick, 1957; Novoa et al., 1959). The substrate and inhibitor constants have 
been summarized by Papaconstantinou and Colo wick (1961 a), as shown 
in the accompanying tabulation; if the KJs, are dissociation constants, 



Source 



A',„ (pyruvate) K^ (oxamate) 
{mM) (mif) 



Beef heart 0.137 0.0374 

Ascites carcinoma 0.212 0.0563 

Rabbit muscle 0.302 0.10 



oxamate is bound around 0.77 kcal/mole more tightly than pyruvate. The 
difficulty in interpreting K^ is that oxamate complexes with the apode- 
hydrogenase, E, with E-NAD, and with E-NADH, the dissociation con- 
stants being different for each (Novoa et al., 1959). The values for three 
different pH's are shown in the accompanying tabulation. Such ternary 







K 


i (mM) at: 




Complex 


pH 6.40 




pH 8.45 


pH 9.70 


E-oxamate 


0.10 




0.78 


20 


E-NAD-oxamate 


0.069 




0.57 


11 


E-NADH-oxamate 


0.026 




0.17 


1.6 



complexes have recently been demonstrated by ultracentrifugal separation, 
1 mole of oxamate being bound for each mole of NAD or NADH (Novoa 
and Schwert, 1961). The dissociation constant for the E-NADH-oxamate 
complex determined ultracentrifu gaily is 0.011 mM at pH 7.4. No evidence 
could be found for a complex with the apoenzyme alone; whether such a 
complex occurs or not, the inhibition is mainly due to the ternary complexes. 
The lactate dehydrogenase from human liver and heart is also inhibited by 
oxamate (Plummer and Wilkinson, 1963). The reduction of 2-ketobutyrate 
is inhibited more strongly than pyruvate reduction, presumably because 
2-ketobutyrate is bound to the enzyme less tightly, but the reduction of 
succinic semialdehyde by a rat brain lactate dehydrogenase is inhibited 
to the same degree as the reduction of pyruvate, namely, 88% by 0.1 mM 
(Fishbein and Bessman, 1964). Not all lactate dehydrogenases are sensitive 
to oxamate; that from L. mesenteroides is inhibited only 50% by 7 mM 
(Papaconstantinou and Colowick, 1961 a). Oxamate specifically inhibits the 
L( + )-lactate dehydrogenase and does not affect d( — )-lactate dehydroge- 



^- 



434 2. ANALOGS OF ENZYME REACTION COMPONENTS 

nase up to 8 raM (Dennis and Kaplan, 1960). The active site of the l( + )- 
lactate dehydrogenase was represented as containing a cationic group for 
electrostatic interaction with the C00~ group, and a hydrogen bonding 
group which interacts with the OH group of lactate and the NHg group of 
oxamate (it might also bond to the CO or enolized COH group of oxamate). 
The role of the amino group of oxamate on the binding is not known, nor 
can the importance of the enolic tautomer of oxamate be evaluated. The 
marked decrease in inhibition between pH 8.45 and 9.70 (see tabulation 
above) could indicate the deprotonation of an amino group, but it could 
be on the enzyme as well as the inhibitor. 

The aerobic lactate production in human leucocytes is inhibited less than 
25% by 10 mM oxamate, even in broken cell suspensions, suggesting that 
lactate dehydrogenase is not solely responsible for lactate formation, al- 
though the sensitivity of the leucocytic enz^^me to oxamate has not been 
examined (McKinney et al., 1955). The effects of oxamate on tumor cell 
metabolism and growth have been studied thoroughly by Papaconstantinou 
and Colowick (1957, 1961 a, b). Anaerobic glycolysis in ascites carcinoma 
cells is inhibited 50% by 8 mM oxamate and aerobic glycolysis is similarly 
depressed. This may indicate that oxamate does not penetrate into cells 
readily, since the K^ of 0.0563 mM for ascites cell lactate dehydrogenase 
would lead one to expect a greater effect at this concentration. The inhi- 
bition of anaerobic glycolysis decreases with time due to the accumulation 
of pyruvate, whereas no accumulation of pyruvate occurs aerobically, in- 
dicating that oxamate has little effect on pyruvate oxidase (an 18% inhi- 
bition of pyruvate oxidation by 10 mM oxamate was observed). A decrease 
in the inhibition anaerobically with time was also noted in Tetrahymena 
pyriformis (Warnock and van Eys, 1963). The growth of HeLa cells is 
completely inhibited by 40-80 mM oxamate and this is paralleled by de- 
creases in glucose uptake and lactate formation, so that lactate dehydro- 
genase appears in some manner to be essential for the growth of these cells 
(assuming that oxamate acts specifically on lactate dehydrogenase). It was 
proposed that oxamate might be a useful inhibitor for selectively blocking 
glycolysis in mammalian cells. Mice can tolerate quite large doses (1 g/kg), 
however. A block of glycolysis, of course, refers here only to an inhibition 
of lactate formation, and the formation or utilization of pyruvate should 
not be significantly affected, so it would seem that aerobic glucose meta- 
bolism, at least with respect to the generation of energy, would be resistant 
to oxamate. 

A number of disturbing observations have appeared which cast some 
doubt on the simple concept that oxamate specifically inhibits lactate de- 
hydrogenase. Leached HeLa cells restored to normal conditions actively 
extrude Na+ and accumulate K+; these processes are inhibited 50% and 
77%, respectively, by 38 mM oxamate (Wickson-Ginzburg and Solomon, 



LACTATE METABOLISM 435 

1963). Inasmuch as the conditions were aerobic here, it is difficult to un- 
derstand how an inhibition of lactate dehydrogenase would account for the 
marked effects observed, unless there is an elevation of the NADH/NAD 
ratio due to the prevention of pyruvate reduction, this slowing the oxidation 
of glyceraldehyde-3-P and reducing the generation of ATP. This does not 
seem to occur in Ehrlich ascites carcinoma cells inasmuch as oxamate stim- 
ulates the formation of C^^Og from glucose-6-C^* without affecting that 
from glucose- 1-C^'* (Christensen and Wick, 1963). The stimulation is thus 
associated only with oxidation through the cycle and there is no effect on 
the fraction going through the pentose-P pathway. More pyruvate enters 
the cycle since less goes to lactate. The effects of oxamate on glucose utiliza- 
tion will depend for one thing on how rapidly pyruvate can be oxidized. 
The Crabtree effect is abolished almost completely by 40 uiM oxamate, i.e., 
in the presence of glucose, oxamate stimulates the respiration in ascites 
cells (Papaconstantinou and Colowick, 1961 a). Simultaneously, glucose up- 
take is depressed 40% and lactate formation 70%. In view of the conclusion 
about the nature of the Crabtree effect in the section on 2-DG, it would 
seem that inhibition of lactate dehydrogenase could not be responsible for 
this effect of oxamate. It is possible that oxamate diverts more pyruvate 
into the cycle and hence stimulates respiration under these conditions, but 
this would not be a true abolition of the Crabtree effect. One would not 
expect glucose respiration to be depressed by oxamate in any case if the 
only action is on lactate dehydrogenase, but 31% respiratory depression 
is produced by 10 mM oxamate in guinea pig alveolar macrophages (Oren 
et at., 1963). Despite the statements relative to the specificity of oxamate, 
it must be admitted that very few enzymes or metabolic pathways have 
been studied. It is quite likely that certain phases of amino acid metabolism 
might also be inhibited, since oxamate could be considered as an amino 
acid analog. It will also be noted that all the effects discussed in this para- 
graph were produced by oxamate at the high concentration of 10 raM 
or above. 

Oxalate often inhibits lactate-metabolizing enzymes as potently as does 
oxamate and similarly forms ternary complexes with lactate dehydrogenase 
and NAD. However, it differs from oxamate in being competitive with 
lactate instead of pyruvate; this has been shown on beef heart lactate 
dehydrogenase (^, = 0.015 mM at pH 6.7) (Novoa et al, 1959), yeast 
D-lactate dehydrogenase (^, = 0.007) (Labeyrie and Stachiewicz, 1961), 
yeast D-lactate cytochrome c reductase {K^ = 0.0016 mM) (Nygaard, 1961 
b), and yeast D-hydroxy acid dehydrogenase (/if, = 0.0025 mM) (Boeri et 
al., 1960), although the inhibition seems to be uncompetitive on the lactate 
dehydrogenase oi Propionihacterium pentosaceum (Molinari and Lara, 1960). 
The complex kinetics have been treated in detail by Novoa et al. (1959), 
but it is still rather puzzling that oxalate inhibits all these enzymes so 



436 2. ANALOGS OF ENZYME REACTION COMPONENTS 

potently and competes with lactate rather than pyruvate. Tartronate and 
malonate also inhibit competitively with lactate and noncompetitively with 
respect to pyruvate, although by no means as strongly as oxalate, and 
Ottolenghi and Denstedt (1958) concluded that pyruvate and lactate react 
with different sites on the enzyme surface, but this is not necessary, as 
Novoa et al. (1959) have shown, since the configurations of the active sites 
on E-NAD and E-NADH are different. The five lactate dehydrogenase 
(LD) isoenzymes from human tissues are inhibited to different degrees by 
oxalate at 0.02 vaM (see accompanying tabulation) (Emerson et al., 1964). 



Isoenzyme 


0/ 

/o 


Inhibition 


LDi 




70 


LD^ 




64 


LD3 




56 


LD4 




46 


LD, 




32 



There are A and B monomers, the B monomer being more sensitive to 
oxalate. LD^ is a pure B tetramer, LD5 is a pure A tetramer, and the others 
are intermediate in the proportion of A to B. This is an illuminating example 
of how enzymes may respond differently to inhibitors by reason of varied 
composition. 

There are several interesting studies on lactate dehydrogenases from 
which deductions on the nature of the active site and certain interaction 
energies may be derived. Dikstein (1959) found that yeast lactate dehydro- 
genase is not inhibited potently by monocarboxylates: the concentrations 
for 50% inhibition are 6000 vaM for formate, 2300 mM for acetate, 800 milf 
for propionate, 300 mM for butyrate, 120 mM for valerate, and 10 mM for 
caprylate. By plotting log (1)50 against the number of carbon atoms in the 
aliphatic chains he obtained a straight line, from the slope of which it was 
possible to calculate that the transfer of a methylene group from the solvent 
to the enzyme surface involves an over-all energy change of 0.5 kcal/mole. 
It was assumed that the interaction energy between the COO" group and 
an enzyme cationic group could be calculated from the intercept of this 
line, but it is doubtful that at the high concentrations used the residual 
energy can be attributed with certainty to ion-ion interactions, since non- 
specific effects cannot be excluded. 

The D-of-hydroxy acid dehydrogenase of yeast is competitively inhibited 
by several dicarboxylates (see accompanying tabulation) (Cremona, 1964). 
Monocarboxylates inhibit much more weakly and the inhibitions are usually 
not purely competitive. Oxalate is bound approximately 3.7 kcal/mole more 



LACTATE METABOLISM 437 

tightly than malonate, suggesting that there are two cationic groups quite 
close on the enzyme. 



Inhibitor Ki (mil/) 



Oxalate 


0.0025 


Tartronate 


0.84 


Malonate 


0.95 


L-Malate 


1.05 


a-Ketoglutarate 


1.4 



The D-lactate dehydrogenase of yeast studied by Boeri et al. (1960) is 
inhibited potently by oxalate {K^ = 0.0025 mM), moderately by malonate 
(^ . = 0.9 mM), and not at all by 16 mM succinate or fumarate, indicating 
the importance of the position of the second C00~ group. The enzyme is 
inactivated gradually by EDTA and reactivated with Zn++. Although this 
does not prove that Zn++ is the normal metal ion involved, it points to the 
possibility of chelation between a-hydroxy carboxylates, or dicarboxylates, 
with an enzyme-bound metal ion. The configuration around the a-carbon 
atom is important since L-lactate binds very weakly to the enzyme {K^ = 
= 62 mM). 

The lactate cytochrome c reductases of yeast are flavoproteins and are 
competitively inhibited by a variety of lactate analogs (Nygaard, 1961 a, 
b, c). Fatty acid inhibitions lead to the calculation of 0.37 kcal/mole for 
the interaction of methylene groups with the enzyme and 2.80 kcal/mole 
for the C00~ group in the case of the D-lactate cytochrome c reductase, 
and of 0.37 kcal/mole and 0.88 kcal/mole, respectively, for the L-lactate 
cytochrome c reductase. Neither eftzyme is inhibited by dicarboxylates, 
with the exception of oxalate, and from the differences in inhibition patterns 
it is probably safe to assume that the active sites of the two enzymes are 
quite different. 

Some analog inhibitors of lactate dehydrogenases need only be listed for 
reference: pyruvate (Green and Brosteaux, 1936; Das, 1937 b; Neilands, 
1952; Labeyrie and Stachiewicz, 1961), bromopyruvate, chloropyruvate, 
hydroxypyruvate (Busch and Nair, 1957), phenylpyruvate (Dikstein, 1959), 
a-ketoglutarate (Boeri et al., 1960), malate (Boeri et al., 1960; Busch and 
Nair, 1957), tartronate (Green and Brosteaux, 1936; Lehmann, 1938), tar- 
trate (Labeyrie and Stachiewicz, 1961), mandelate (Lehmann, 1938), a, 
y-diketovalerate (Meister, 1950), benzenesuLfonate (Baptist and Vestling, 
1957), mercaptoacetate, mercaptosuccinate, or-mercaptopropionate, of-mer- 
captobutyrate, and a-mercaptovalerate (Chaffee and Bartlett, 1960). The 
benzenesulfonates, where the COO" group is replaced by a SO3" group. 



438 2. ANALOGS OF ENZYME REACTION COMPONENTS 

and the mercapto fatty acids, where the a-OH is replaced by an a-SH 
group, are particularly interesting and deserve further study to determine 
their specificity on lactate metabolism. 

The oxidation of glycolate is catalyzed by glycolate oxidase and the 

CHO-CHO- + NADH + H+ ;fi CH^OH-COO- + NAD+ 
Glyoxylate Glycolate 

reverse reaction by glyoxylate reductase, both enzymes being found in 
plants. This reaction is similar to the pyruvate ±5 lactate interconversion 
and, indeed, L-lactate is slowly oxidized by the oxidase. Since these enzymes 
bear some relationship to those involved in lactate metabolism due to this 
similarity, it is not inappropriate to discuss them at this point, particularly 
as Zelitch at the Connecticut Agricultural Experiment Station has reported 
some very interesting inhibitions by analogs. Glyoxylate reductase is ap- 
parently not especially susceptible to analogs: the following inhibitions were 
observed with 16.5 mM glyoxylate and 10 mM analogs — phenylglyoxy- 
late 21%, oxalacetate 22%, oxamate 31%, and pyruvate 56% (Zelitch, 
1955). Glycolate oxidase, on the other hand, is well inhibited by a-hydrox- 
ysulfonates. Competitive inhibition was found with hydroxymethanesul- 
fonate {K, = 0.0018 mM), a-hydroxyethanesulfonate (^,; = 0.0023 mM), 
and sulfoglycolate {K^ = 0.0021 mM); since K,,^ = 0.38 mM, these analogs 
are reasonably potent (Zelitch, 1957). Rabbit muscle lactate dehydrogenase 
is inhibited comparably, D-glycerate dehydrogenase less strongly, and ma- 
late dehydrogenase not at all, so that some specificity toward enzymes 
oxidizing a-hydroxy acids is evident. The inhibition of lactate oxidation 
by the glycolate oxidase is inhibited very strongly because lactate is bound 
less tightly to the enzyme. Another competitive inhibitor of comparable 
potency is of-hydroxy-2-pyridinemethanesulfonate (Zelitch, 1959) which 
has been used in most of the in vivo work apparently because it is more 
effective in cells, although sulfoglycolate w^ould seem to act very similarly 
(Zelitch, 1958). 

OH 

I 

CH3— CH— SO3 H0-CH2— so; 

Q-Hydroxyethanesulfonate HydroxymethanesuKonate 



9" ^N OH 

I /^ \ I 

'OOC — CH-SO; \\ A>— CH-SO3 

Sulfoglycolate ■^^'^^^!'u''^'K . 

pyridinemethanesulfonate 



PHOSPHATASES 439 

Normal tobacco leaves contain around 0.5-1 //mole/g wet weight glyco- 
late and this level can be increased as much as 10-fold by placing them in 
solutions containing the a-hydroxysuLfonates (Zelitch, 1959). Since glyco- 
late is formed photosyntheticaUy, marked accumulation occurs only in the 
light. The concentration of a-hydroxy-2-pyridinemethanesulfonate for max- 
imal inhibition is near 10 m.M, indicating that penetration into the cells 
is quite poor. At higher concentrations of inhibitor the glycolate level falls, 
due presumably to inhibition of glycolate formation; in fact, even at 10 mM 
there must be some inhibition since photosynthetic incorporation of C^Og 
is inhibited about 33%. The pattern of C^* distribution is, however, more 
markedly altered; in controls, glycolate-C^^ accounts for 5.5% of the labeling 
but in inhibited leaves it is almost 50%. This is a good example of a spe- 
cific analog inhibitor useful in studying the importance of an intermediate 
in a complex metabolic pathway, and valuable information may result 
from more detailed studies on photosynthesis. 



PHOSPHATASES 

These enzjines are commonly inhibited by the products of the hydroly- 
sis. We have already noted the inhibitions of phospho-L-histidinol phos- 
phatase, O-phosphoserine phosphatase (page 270), and glucose-6-phos- 
phatase (page 442) by dephosphorylated products, and there are other 
examples related to analog inhibition. Orthophosphate also frequently in- 
hibits: The following may be cited as instances of well-marked inhibition 
of different types of enzyme — calf intestinal phosphatase (Schmidt and 
Thannhauser, 1943), mouse liver acid phosphatase (Macdonald, 1961), 
mouse liver pyrophosphatase (Rafter, 1958), calf brain carbamyl and acyl 
phosphatases (Grisolia et al., 1958), sweet potato phosphatase (Ito et al., 
1955), and E. coli alkaline phosphatase (Garen and Levinthal, 1960). An- 
other class of inhibitor comprises the phosphates that are substrates. Thus 
sweet potato phosphatase with phenyl phosphate as substrate is competi- 
tively inhibited by /^-glycerophosphate {Kf = 2 mM), pyrophosphate (K, = 
= 0.33 mM), metaphosphate (^, = 3.2 mM), and ATP {K^ = 0.67 mM), 
all of which are also substrates (Ito et al., 1955). Likewise the E. coli 
phosphatase with p-nitrophenyl phosphate as substrate is competitively 
inhibited by uridine phosphate {K^ = 0.044 raM), guanosine phosphate 
{K^ = 0.046 mM), /^-glycerophosphate {K^ = 0.05 vnM), glucose- 1 -phos- 
phate {K^ = 0.063 mM), and adenosine-5'-phosphate (Z,; = 0.093 mM) 
(Garen and Levinthal, 1960). 

More interesting are the inhibitions by various anions that may be con- 
sidered as analogs of either phosphate or the substrate phosphates. Thus 
arsenate (Garen and Levinthal, 1960; Ito et al., 1955; Macdonald, 1961), 
borate (Ito et al., 1955), and silicate (Umemura et al., 1961) inhibit various 



440 2. ANALOGS OF ENZYME REACTION COMPONENTS 

phosphatases about as potently as phosphate and probably combine with 
the enzymes in a similar manner. Certain carboxylates (oxalate, malonate, 
malate, citrate, glucarate, gluconate, lactate, and others) have been found 
to be inhibitory, but most of these are not remarkably effective, probably 
binding to enzyme cationic groups to various degrees or chelating with 
metal ions. However, ( + )-tartrate* is a very potent inhibitor of certain 
phosphatases, as first shown by Abul-Fadl and King (1949) and confirmed 
by Anagnostopoulos (1953), and is so much more active than other anions 
that the mechanism has been investigated in several excellent studies. 

The phylogenetic relationships of ( + )-tartrate inhibition and the vari- 
able susceptibilities of phosphatases from different tissues are quite inter- 
esting. Abul-Fadl and King (1949) found that although prostatic acid phos- 
phatase is very sensitive, no effects are exerted on the acid phosphatases of 
erythrocytes or plasma, and Anagnostopoulos (1953) noted no inhibition 
with the phosphatases of mustard, wheat germ, or Aspergillus. Kilsheimer 
and Axelrod (1958) investigated the effects of ( + )-tartrate on the phospha- 
tases from many sources and found that at 20 mM the inhibitions vary 
from to 93%. They concluded that animal phosphatases are more suscep- 
tible than plant phosphatases, and that bacterial phosphatases are generally 
resistant. It was suggested that ( + )-tartrate may be of taxonomic use in 
those more primitive organisms where it is difficult to decide whether they 
are plants or animals. 

The earliest work demonstrated that the inhibition is stereospecific, nei- 
ther ( — )-tartrate nor meso-tartrate exerting appreciable effects on prostatic 
phosphatase, and this has been confirmed in all the recent studies. In the 
series of phosphatases tested by Kilsheimer and Axelrod (1958) it was ob- 
served that very few of the enzymes are affected by (— )-tartrate and 
these are inhibited only slightly. The iiT/s were determined by London et 
at. (1958) as 0.13 mM for ( + )-tartrate and 93 mlf for (-)-tartrate on 
prostatic acid phosphatase. It is strange that they found meso-tartrate to 
be an effective inhibitor {K^ = 0.4 mM) in contrast to all other work. 
The acid phosphatase from Neurospora crassa is inhibited completely by 
12 mM ( + )-tartrate but is unaffected by 50 mM ( — )-tartrate or meso- 
tartrate (Kuo and Blumenthal, 1961). In all cases the inhibition by ( + )- 
tartrate is competitive with respect to substrate. The inhibition is thus more 
marked with /^-glycerophosphate as the substrate than with phenyl phos- 
phate or p-nitrophenyl phosphate, since the former substrate is bound less 
tightly (Nigam at al, 1959; Kuo and Blumentha