New comprehensive biochemistry vol 06 the chemistry of enzyme action

11 where AG is the Gibbs free energy change for the reaction under the given conditions and AGO is the change in Gibbs energy for the reaction with reactants and products at the standa

New Comprehensive Biochemistry

The Chemistry of Enzyme Action

All rights reserved N o part of this publication may be reproduced, stored

in a retrieval system, or transmitted, in any form or by any means, elec- tronic, mechanical, photocopying, recording or otherwise without the prior permission of the copyright owner

ISBN for the series: 0444 80303 3

ISBN for the volume: 0444 80504 4

Published by:

Elsevier Science Publishers B.V

P.O Box 1527

lo00 BM Amsterdam, The Netherlands

Sole distributors for the U.S.A and Canada

Elsevier Science Publishing Company Inc

52 Vanderbilt Avenue

New York, N Y 10017, U.S.A

Printed in The Netherlands

Recognition is of fundamental importance to living systems How do proteins and

other macromolecules distinguish between molecules of similar shape or ions of similar size? Recognition is controlled by the intermolecular forces between the

‘host’ and ‘guest’ The binding energy resulting from the mutual satisfying of these forces are ultimately responsible for the catalysis and specificity of enzyme-catalysed reactions Understanding how enzymes efficiently transform their substrates is not only a question of reaction mechanisms, describing the routes of bond making and breaking processes, but also one of recognising that the interactions between the

‘non-reacting’ parts of the substrate and enzyme play a crucial role in the activation step The forces responsible for the chemical mechanism adopted by the enzyme are closely related to those which account for recognition of the ‘non-reacting’ parts The interplay of these forces is fundamental to an appreciation of enzymic catalysis This volume describes the physical and organic basis of enzyme action The background knowledge required to understand the chemistry of enzyme action is presented by major scientists in their own field The borderline area between

disciplines are stimulating and rewarding and this is reflected by the high calibre of

the contributors to this volume The level of understanding enzyme-catalysed reactions is dependent upon the techniques employed Determining reaction mecha- nisms requires a detailed knowledge of kinetic techniques and discussions of these topics are followed by examples of their applications The level of understanding enzyme catalysis that has been reached by using physical organic methods is illustrated by some biologically important examples Finally the important contribu- tion that biomimetic studies have made to understanding the recognition and catalysis exhibited by enzymes is emphasised by leading exponents in the field

Michael I Page

Huddersfield, October 1983

Preface

Contents f

Chapter I The energetics and specificity of enzyme - substrate interactions Michael I Page (Huddersfield)

1 Introduction

2 Enzyme structure

3 Michaelis - Menten kinetics

5 Regulation and thermodynamics

6 Specificity and k , , , / K ,

7 Rate enhancement and specificity

8 Specificity, induced fit and non-productive binding 9 Approximation, entropy and intramolecular reactio 4 Intra- and extra-cellular enzymes

10 Decreasing the activation energy

11 Utilisation of binding energy

12 Intramolecular force fields

Bond stretching, 35 - Bond angle bending, 35 - Torsion, 35 - Disulphide links, 36 - Non-bonded interaction, 37 - 13 Intermolecular force fields

Hydrogen bonding, 39 - Electrostatic interactions, 40 - Hydrophobicity, 44 - Dispersion forces, 45 - 14 Stress and strain

15 Estimation of binding energies

References

Chapter 2 Non-covalent forces of importance in biochemistry Peter Kollman (San Francisco)

1 Introduction

1.1 The basis

2 The thermodynamics of non-covalent interactions

2.2 Solution phase association

3.1 Electrostatic forces

2.1 Gas phase interactions

3 Examples of biological 3.2 Dispersion forces

3.3 Hydrophobic interactions

4 Summary

Acknowledgement

References

V

1

2

6

7

9

11

12

14

16

22

27

34

38

47

50

53

5 5

55

55

57

57

59

62

62 65

66

69

70

70

Chapter 3

Enzyme kinetics

Paul C Engel (Sheffield)

1 Introduction: aims and approaches in enzyme kinetics 2 Steady-statekinetics

2.1 Michaelis-Menten equation

2.2 Relationship between K , and dissociation constant

2.3 Experimental determination of kinetic constants

(a) Experimental design, 78 - (b) Lineweaver-Burk plot, 79 - (c) Eadie-Hofstee plot, 79 - (d) Hanes plot, 79 - (e) Eisenthal and Cornish-Bowden plot, 82 - (f) Does the choice of plotting method matter? 82 - 2.4 Non-linearity

2.5 Inhibition

(a) Definition, 86 - (b) Competitive inhibit Non-competitive inhibition, 89 - (e) Mixed inhibition, 90 - (a) Types of mechanism, 91 - (b) Overall strategy, 93 - (c) Deriving a rate equation, 93 - (d) Experimental determination of the rate equation for an individual enzyme, 97 - (e) Drawing conclusions from the experimentally determined rate equation, 99 - (f) Inhibition experiments, 104 - (g) Isotope exchange at equilibrium, 106 - (h) Whole time-course studies, 106 - 2.6 Multi-substrate kinetics

3 Rapid reaction kinetics

References

Chapter 4 Aspects of kinetic techniques in enzymology Kenneth T Douglas and Michael T Wilson (Colchester)

1 Introduction

2 Use of steady-state techniques

2.1 Note on measurement of initial velocities

3 Experimental treatment of transients

3.1 Determination of k,, _ _

4 Stopped-flow methods

4.1 Binding reactions

5.1 Ligand binding

4.2 Burst kinetics

_ _ 5.2 Coupled reactions, linked redox reactions and structural rearrangements

6 Conclusion

References

Chapter 5 Free-energy correlations and reaction mechanisms Andrew Williams (Canterbury)

1 Introduction

2 Brensted relationships

73

73

76

76

78

78

82

86

91

107

109

111

111

113

114

115

116

119

119

121

123

123

124

125

125

127

127

128

2.1 Simple proton transfer 128

2.2 Molecular basis of the Bransted relationship and interpretation of the exponents 129

2.3 Statistical treatments 130

2.4 The extended Brcanste 2.5 Meaning of the Brans (a) Effective charges in transition states, 134 - (b) Additivity of ‘effective’ charge, 135 - (c) Transition state index (a), 136 - (a) Eigen curvature, 137 - (b) Marcus curvature, 139 - 2.6 Curvature in Bransted correlations 137

141

3 Indices of el 142

143

(a) Additivity of sigma values, 147 - (b) Resonance and inductive effects, 148 - (c) Sigma-minus parameters, 148 - (d) Sigma-plus parameters, 149 - (e) More than one transmission path for u , 151 - (f) The Yukawa-Tsuno equation, 155 - (a) Taft’s polar ( 0 *) and steric ( E , ) parameters, 158 - (b) Taft’s steric parameter ( E s ) , 161 - (c) Relationship between u * and u , , 162 - (d) Meaning and use of E, and 6, 163 - (e) Other steric parameters, 164 - ( f ) Values of u * for alkyl groups, 166 - 4 Hydrophobic interactions 166

3.2 Separation of inductive, steric and resonance effects 156

4.1 Other hydrophobic parameters 4.2 Non-linear hydrophobic relation 4.3 Molar refractivity 170

4.4 Additivity 171

4.5 Ambiguities arising from interrelationships between parameters 172

4.6 Application to n 172 5 Solvent effects 173

5.1 Reporter groups 175

6 General equations of 177

6.1 Swain-Scott and Edwards relationship 177

7 Cross-correlation and selectivity-reactivity 179

7.1 Cross-correlation 179

7.2 Reactivity-selecti 183 8 Estimation of ionisation constants 183

9 Elucidation of mechanism 186

9.1 Mechanistic identity 186

9.2 Changes in mechanism 189

9.3 Change in rate-limiting step 191

9.4 Dependece on concentration as a free energy correlation 9.5 Distinction between kinetic ambiguities

9.6 Proton transfer

(a) Correlations with model reactions, 187 - References 197

General references 200

Chapter 6 Isotopes in the diagnosis of mechanism Andrew Williams (Canterbury) 203

1 Theoretical background 203

2 Measurement of isotope effects 205

3 Equilibrium isotope effects

4 Primary isotope effects

4.1 Variation in primary isotope effects

4.3 Heavy atom isotope effects

4.2 Solvent isotope effects

(a) Nucleophilic versus general base catalysis, 213 - (b) Fractionation factors, 214 - 5 Secondary isotope effects

6 Labelling techniques ,

6.2 Proton transfer

6.3 Detection of intermediates by isotope exchange

6.4 Isoracemisation

6.6 Isotopic enrichment ,

References

General references

6.1 Position of bond cleavage

6.5 Double-labelling experiments

Chapter 7 The mechanisms of chemical catalysis used by enzymes Michael I Page (Huddersfield)

1 Introduction

2 General acid base 2.1 Catalysis by stepwise proton tra 2.3 Catalysis by preassociation

2.4 Concerted catalysis

4 Metal-ion catalysis ,

5 Catalysis by coenzymes

5.1 Pyridoxal phosphate coenzyme

5.2 Thiamine pyrophosphate coenzy 5.3 Adenosine triphosphate (ATP) 6.2 Nicotinamidecoenzymes

6.3 Flavin coenzymes

(a) Oxidation of amino and hy thiols, 261 - (c) Reductive activation of oxygen by dihydroflavins, 261 - (d) Flavomono- oxygenase, 262 - 6.4 Electron transfer with metals

(a) Thermodynamic stability of metal complexes, 263 - (b) edox metalloproteins and oxygen, 264 - (d) Iron-containing proteins and enzymes, 265 - (e) Copper-containing oxidases and monooxygenases, 261 - References

207

209

21 1

213

218

219

220

220

222

222

223

225

226

226

227

229

229

230

23 1

235

236

236

231

240

243

246

246

249

252

253

256

256

251

259

262

269

Chapter 8

Enzyme reactions involving imine formation

Donald J Hupe (Rahway)

1 Introduction 2 Iminium ion formation

4 Aldolases

5 Transaldolase 6 Acetoacetate decarboxylase

I Pyruvate-containing enzymes

9 Conclusions

References

8 Dehydratases

Acknowledgement

Chapter 9 Pyridoxal phosphate-dependent enzymic reactions: mechanism and stereochem- istry Muhammad Akhtar, Vincent C Emery and John A Robinson (South- ampton)

1 Historic background: Braunstein-Snell hypothesis

2 Pyridoxal phosphate-dependent reactions involving C,-CO, H bond cleavage

3 Pyridoxal phosphate-dependent enzymic reactions involving C,-H bond cleavage

3.1 Aminotransferases

(a) Metabolic background, 314 - (b) Transaminations at the C, of amino acid, 315 - (c) Mechanistic studies on miscellaneous transaminations, 3 15 -

3.4 5-Aminolevulinate synthetase (ALA synthetase

4.1 General introduction

4.2 P-Replacement reactions

(a) Tryptophan synthetase, 331 -

4.3 8-Elimination-deamination reactions

(a) Tryptophan synthetase-/3, protein, 33 enzymes, 339 - ) Tryptophanase, 338 - (c) Miscellaneous 5 Pyridoxal phosphate-dependent reactions occurring at C,

5.1 Enzymic aspects

5.2 Stereochemical aspects

6 Structure and molecular dyn rnary complexes

6 1 Electronic spectrum of the coenzyme chromophore

6.2 Chemical studies on binary (coenzyme-enzyme) complexes

6.3 Stereochemical aspects of the reduction of Schiff base at C-4' with NaBH,

6.4 Structure and stereochemistry of the substrate-coenzyme bond in ternary complexes

6.5 Stereochemical and mechanistic events at C, of the substrates and at C-4' of the coenzyme 6.6 Orientation of the pyridinium ring of the coenzy rnarycomplexes

during catalysis

27 1

27 1

272

216

279

286

288

29 1

294

298

298

299

303

306

314

314

319

320

327

33 1

33 1

331

336

343

343

344

349

349

353

354

355

359

366

367

Chapter 10

Transformations involving folate and biopterin cofactors

S.J Benkovic and R.A Lazarus (University Park)

I Introduction

2 Structure

3 Reduction

5 Formyltransfer

6 Methyl transfer

Chapter 1 1 Glycosyl transfer - The Physicochemical Background Michael L Sinnott (Bristol)

1 Introduction

2 Effects of the gen in space

2.1 The anomeric effect

2.3 Stereoelectronic control of reactions of acetals?

3 The chemistry of processes occurring with electrophiles or acids

3.1 Lifetimes of oxocarbonium ions

3.2 Preassociation mechanisms

3.3 Chemical synthesis of glycosides

3.4 Effect of oxocarbonium ion structure

3.5 Intramolecular nucleophilic assistance

3.6 Electrostatic stabilisation?

4 Processes occurring via the application of acidic oxygen-leavinggroups

2.2 Geometrical changes in oxocarbonium ion formation

4.1 Specific acid catalysis of acetals, 4.2 Intramolecular nucleophilic assistance in specific acid-catalysed processes

4.3 Intermolecular general acid catalysis of the hydrolysis of acetals and ketals

4.4 Intramolecular general acid catalysis of the hydrolysis of acetals, ketals and glycosides 4.5 Intramolecular general acid catalysis concerted with intramolecular nucleophilic (or elec- 4.6 Electrophilic catal 5 Acid- and electrophi sulphur and fluorine

5.1 Hydrolysis of gl

5.3 Hydrolysis of hemithioacetals, hemithioketals, and thio- and thia-glycosides

5.4 Hydrolysis of glycosyl fluorides

6 Envoi References

373

373

373

375

376

379

38 1

381

385

389

389

390

39 1

395

396

398

398

399

402

403

406

407

408

408

413

413

417

419

420

42 1

422

423

425

426

421

427

Vitamin B12

Kenneth L Brown (Arlington) ,

1 Introduction and scope of this chapter

2 Structure

3 Oxidation states

5 Chemical reactivity of organocobalt complexes

5 I Reactions in which carbon-cobalt bonds are formed (a) Synthesis of organocobalt complexes via cobalt(1) reagents, 439 - (b) Synthesis of organocobalt complexes via cobalt(I1) reagents, 441 - (c) Synthesis of organocobalt complexes via cobalt(II1) reagents, 443 - 5.2 Reactions in which carbon-cobalt bonds are cleaved

(a) Mode I cleavage of carbon-cobalt bonds, 445 - (b) M bonds, 447 - (c) Mode 111 cleavages of carbon-cobalt bonds, 450 -

cobalt 5.3 Axial ligand substitution reactions

5.4 Reactions of cobalt-bound organic ligands

6 Concluding remarks

References

Chapter 13 Reactions in micelles and similar self-organized aggregates Clifford A Bunton (Santa Barbara)

1 Introduction

2 Formation of normal micelles

3 Micellar structure in water

4 Kinetic and thermodynamic effects

4.1 Micellar effects upon reaction rates and e

4.2 Quantitative treatments of micellar effects in aqueous solution

4.3 Quantitative treatment of bimolecular reactions

4.4 Second-order rate constants in the micellar pseudophase

5 Reactive counterion micelles

6 Reactions in functional micelles

7 Stereochemical recognition

8 Submicellar and non-micellar aggregates

9 Micelles in non-aqueous systems

9.1 10 Related systems

10.1 Reactions in microemulsions ,

10.2 Reactions in vesicles

11 Photochemical reactions

12 Isotopic enrichment

13 Preparative and practical aspects

Noteaddedinproof

Acknowledgements

References

Normal micelles in non-aqueous media 9.2 Reverse micelles in aprotic solvents

433

433

433

435

437

438

439

444

452

455

457

457

46 1

46 1

462

464

468

468

47 1

472

475

479

482

487

487

490

490

49 1

493

493

495

496

497

498

499

500

500

Chapter 14

Cyclodextrins as enzyme models

Makoto Komiyama and Myron L Bender (Tokyo and Evanston)

1 Introduction

2 Formation of an inclusion complex

3 Catalysis by cyclodextrins 3.1 Catalysis by the hydro 3.2 Effect of reaction field 4.1 Substrate specificity

4.2 Product specificity

4 Specificity in cyclodextrin 5.1 Models of hydrolytic enzymes

5.2 Model of carbonic anhydrase

5.3 Model of metalloenzymes ,

5.4 Introduction of a coenzyme moiety

6 Conclusion

Acknowledgements

Chapter I5 Crown ethers as enzyme models J Fraser Stoddart (Sheffield)

1 Introduction

2 Ground-state binding and reco 2.1 Binding forces

2.2 Complexation

2.3 Enantiomeric differentiation ,

2.4 Substrate recognition

2.5 Allosteric effects

3.1 Enzyme mimics: hydrogen-transfer reactions

3.2 Enzyme mimics: acyl transfer reactions

3.3 Enzyme analogues: Michael addition reactio 3 Binding and recognition at 4 Conclusion

References

505 505 506 511 511 513 514 514 517 519 520 520 524 525 525 526 526 526 529 529 530 530 538 540 542 544 546 546 550 556 558 559 Subject Index 563

The energetics and specificity of enzyme-substrate

interactions

MICHAEL I PAGE

Department of Chemical and Physical Sciences, The Polytechnic, Queensgate, Hudders-

field HD1 3HD, Great Britain

1 Introduction

Life is a dynamic process which depends upon the recognition (interaction) between inanimate molecules Living organisms have the capacity to utilize external energy and matter to maintain and propagate themselves The interactions between sub- strate and enzyme that account for the catalysis and specificity of the reactions of intermediary metabolism also account for the energy-coupling processes which enable the exchange of chemical energy and allow the system to do work

Because of their relative masses the electron density distribution controls the movement of nuclei Electronic interactions and distortions of electron density distributions are responsible not only for the formation of structures and complexes but also for chemical reactions themselves It is thus logical to examine the forces that hold enzymes in their unique 3-dimensional structure, then the energetics of enzyme-substrate interactions and finally the forces or mechanism by which the bond-making and -breaking processes occur during the reaction catalysed by the enzyme

Enzymes are usually globular proteins and are distinguished from fibrous proteins

by the ability of the pieces of secondary structure to associate and give a stable 3-dimensional structure The forces responsible for protein folding are similar to those used in the formation of antibody-antigen or hormone-receptor complexes and to those giving rise to enzymic catalysis The problem with protein folding is how does the system gain enough energy from hydrophobicity, hydrogen bonds and all the various electrostatic interactions to overcome the loss of conformational entropy and steric strain that occurs in the folded state? The problem with complex formation is how does the system gain enough energy to overcome the loss of translational, rotational and vibrational entropy that occurs upon complexation? The problem with enzymic catalysis is how are these interactions expressed in the transition state but not in the ground state of the enzyme-substrate complex?

It is necessary to understand the structures of folded proteins so that we are aware of the environment or force field generated by the enzyme which in turn controls the interaction between substrate and enzyme This chapter will briefly

Michael I Page (Ed.), The Chemistry of Enzyme Action

1984 Elsevier Science Publishers B V

review the problems associated with understanding the energetics of enzymic cataly- sis, the strength and geometrical requirements of the various forces available to

molecular systems (which is dealt with in detail in Chapter 2) and how these forces

stabilise transition states

2 Enzyme structure

Proteins, with a specific function and isolated from a single source, usually have a homogeneous population of molecules all with the same unique amino acid se-

quence Yet with 20 different amino acids possible at each position in a polypeptide

chain of n residues, 20" different primary structures are theoretically possible Furthermore, the great majority of all molecules of a natural protein may exist in a unique conformation despite the degrees of freedom formally permitted by rotation about the peptide backbone (motility) and side chains (mobility) For example, with only 3 conformations defined per residue, a polypeptide chain of 210 residues would

have a theoretical possibility of existing in 1 O 1 O 0 different conformations

Reversible unfolding of proteins has been known for some time The folding of several proteins occurs spontaneously showing that the required information for folding is present in the protein's primary structure [ 11

The forces stabilising the folded state are presumably similar to those that bind the substrate to enzymes The conventional stress with protein structures is on (i) the primary structure of the peptide sequence, i.e the linear covalent linkage, (ii) the secondary structure of the hydrogen bonds between the peptide links, (iii) the tertiary structure formed by hydrophobic interaction which is largely responsible for many protein folds, together with charge-charge interactions and disulphide lin- kages

The production of a disordered polypeptide chain by removing just a few of the interactions which normally contribute to the stability of the folded state is well illustrated by breaking the 4 disulphide bonds in ribonuclease A - even in the absence of a denaturant, the reduced protein is fully unfolded despite all other favourable interactions, such as hydrogen bonds and hydrophobic forces, still being possible [2]

Detailed models of the folded states of proteins depend almost entirely on X-ray

diffraction analysis of the protein crystal Although side chains and flexible loops on

the surface may be mobile in solution, protein conformation in solution is essentially that determined in the solid crystal The atoms of folded proteins are generally well fixed in space

The overall structure- of a folded protein is remarkably compact The extended

chain of carboxypeptidase, with 307 residues, would be 1114 A long but the

maximum dimension of the folded molecule is only about 50 A However, the

polypeptide topology in roughly spherical, globular proteins has never been observed

to form knots [3]

The non-polar side chains tend to be on the inside, shielded from the solvent, and

generate globular structures The ionised hydrophilic side chains are nearly always

on the outside and help maintain solubility Many polar groups are buried in the hydrophobic interior but these tend to be “neutralised” by hydrogen bonding [3]

However, internal hydrogen bonds probably provide no substantial net stabilisation energy to the folded state because similar hydrogen bonds would be formed with water in the unfolded state Hydrogen bonds are probably important in limiting conformational fluctuations in the folded protein The major difference between folded and unfolded molecules is probably the decreased exposure of the non-polar groups to water

Upon folding, there is a large decrease, amounting to several thousand A2, in the surface area of the protein exposed to water Even choosing the transfer of amino acids from water to a non-polar solvent as an estimation of these changes (which, as

we shall see later, underestimates the change) yields hundreds of kJ/mole The free energy of folding amounts to only 15-60 kJ/mole which shows the large unfavoura- ble entropic contribution which accompanies folding as a result of the loss of internal rotations

The polar external surface of a protein which is in contact with water is usually much more mobile than the interior Generally, the higher the proportion of charged amino acids in a protein the greater is its flexibility Most enzymes have a high proportion of hydrophobic amino acids relative to charged ones Together with disulphide bridges this makes proteins have low motility and mobility An exception

is the kinases whch are not globular, have a high content of charged amino acids, are not cross-linked, and are relatively flexible [6]

Charge-charge interactions between two opposite charges in a buried hydro- phobic region of low dielectric constant could be important in strengthening the fold

Proteins found in membranes can experience very different solvation environ- ments In the membrane the environment may be similar to a hydrocarbon solvent while at its surface the medium slowly changes to an aqueous region Proteins or parts of proteins located in the non-polar region of the membrane are expected to have very few exposed charged groups

The atoms within a protein molecule are very closely packed with few holes - about 75% of the volume of the interior is filled with atoms [4] However, the packing density does appear to vary somewhat which may be important for the flexibility of the molecule Any holes present within the molecule tend to be occupied by solvent molecules as in chymotrypsin and carboxypeptidase In micelles,

by contrast, the packing density is low with the volume occupied by the surfactant atoms being large Enzymes are not like micelles and the “oil drop” model of protein structure is incorrect The low packing density of micelles accounts for the much lower efficiency of micellar catalysis (Chapter 13) compared with enzymic catalysis; the surface area of contact between micelle and substrate is much less than that between enzyme and substrate The close packing of protein interiors prevents water molecules being trapped in non-polar cavities and maximises the packing energy

Most of the larger protein structures are composed of 2 or 3 globular units, each

of 40- 150 residues However, disulphide bonds always link cysteine residues within the same domain Each domain is closely packed with flexibility between the domains

Proteins from different species are often functionally indistinguishable; even though their primary sequences may be different their folded conformations are often very similar Even more surprising is the observation that functionally unre- lated proteins sometimes have similar folded conformations, e.g superoxide dis- mutase and immunoglobulin domain [5] Structural homology between proteins of different amino acid sequence may be a packing phenomenon

Although the high degree of time-averaged order of the individual atoms of folded proteins in the crystalline state permits their location in space many protein molecules exhibit varying degrees of flexibility The motions have very different

activation energies ranging from 5 to over 100 kJ/mole, but their importance, if any,

is often unknown For example, in most proteins aromatic rings of tyrosine and phenylalanine residues flip by 180’ rotations at a rate greater than 104/s In general,

experimental techniques such as fluorescence quenching and relaxation, phosphores-

cence and NMR indicate a rather fluid, dynamic structure for globular proteins [ 6 ]

It is of interest to note that the smaller the substrate, apparently, the greater is the requirement for a rigid enzyme Cytochrome c transfers electrons and is very immobile, the solution structure is almost identical to that found in the solid state

[ 6 ] The flavodoxins, transferring hydride ion; catalase, of molecular weight 284 000 with hydrogen peroxide as substrate and carbonic anhydrase, of molecular weight

31000 with carbon dioxide as substrate show a slight increase in mobility Lyso- zyme, of molecular weight 14000 with a polysaccharide as substrate has the same outline fold in solution and in the solid state but many aromatic rings and aliphatic groups are mobile Kinases, molecular weight 45 000, transfer phosphate from ATP and show even greater mobility but like all enzymes still have well defined struc- tures

NMR relaxation parameters, by virtue of their sensitivity to both the frequencies and the amplitudes of motions, are potentially the best source of experimental information on macromolecular dynamics, largely for motions in the frequency range lo6- 10’’/sec However, it is difficult to find a unique model and often several

models, differing largely in the formulation of amplitude factors, account equally well for the experimental data Physically non-existent motions may erroneously be assumed to exist or motions which actually do contribute to relaxation may be suppressed in the analysis [7] Succinctly, some techniques indicate a rigid inflexible system whereas others suggest that enzymes are fluid and flexible The apparent dichotomy may be attributable to the application of concepts derived from macro- scopic observations (density, elasticity, heat capacity, etc.) of an assembly of a large number of molecules or atoms to one molecule of an enzyme There is no clear-cut or sharp boundary between statistical mechanics and thermodynamics A macroscopic system undergoes incessant and rapid transitions among its microstates Extensive parameters, such as energy and volume, have average values equal to the sum of the values in each of the subsystems and undergo macroscopic fluctuations Statistical

distribution functions give the probability that any specified value of a fluctuating extensive parameter will be realised at a given moment The distribution function for macroscopic systems is so sharply peaked that average values and most probable values are nearly identical The most probable values are those that maximise the distribution function The average value of the deviation from the average parameter

is clearly zero, but the average square of the deviation is non-zero and is called the mean square deviation, a convenient measure of the magnitude of the fluctuations, although it is only a partial specification of the distribution The distribution

becomes increasingly sharp as the size of the system increases [8]

molecules per mole, and their thermodynamic properties are very sharp In contrast the individual molecules of proteins are very small systems consisting of relatively few particles, a few thousand atoms, and statistical fluctuations will cause the thermodynamic parameters to be blurred (Fig 1)

A typical globular protein may have a molecular weight of 25000 and each

square volume fluctuation is about 0.2% per molecule A system with similar gross

thermal properties but one hundredth of the size of the protein molecule, i.e 3.2 X cm3 would show a volume fluctuation of about 28, whereas a system of volume 3.2 cm3 would show a volume fluctuation of only 2 X lo-"%

Although globular proteins appear to be only marginally stable in the folded state this does not imply conformational flexibility Indeed environmental changes do not easily induce changes in the folded state Crystallisation under different conditions does not produce large changes in conformation and even the binding of ligands tends only to alter the relative orientations of sub-units or domains In no instance has a substantial change of conformation of a globular protein been observed even

Familiar macroscopic systems consist of many discrete particles, e.g 6 x

g and a volume of 3.2 X

Thermodynamic value

Fig 1 The distribution function for thermodynamic values showing the difference between the most probable and average value of the thermodynamic function

when the binding energies of the ligands are comparable to the net stability of the folded conformation [9] Substantial energy barriers must therefore separate alterna- tive conformations of globular proteins Of course, proteins have considerable thermal energy and consequently their structures, like all molecules, fluctuate but the atoms within the molecule do not appear to deviate far from their positions determined crystallographically Of course this is not true of all proteins, surfaces

and structural units, for example, may show high mobility and motility [6]

3 Michaelis-Menten kinetics

Experimentally, the initial rate, v, of enzyme catalysed reactions is found to show saturation kinetics with respect to the concentration of the substrate, S At low concentrations of substrate the initial rate increases with increasing concentration of

S but becomes independent of [S] at high or saturating concentrations of S (Fig 2 )

This observation was interpreted by Michaelis and Menten in terms of the rapid and reversible formation of a non-covalent complex (ES), from the substrate (S) and enzyme (E), which then decomposes into products (P) (Eqn 1)

This scheme led to the familiar Michaelis-Menten equation

where K , is the Michaelis constant which is the concentration of substrate at which

the initial rate is hay the maximal rate at saturation, V,,, The first-order rate

constant for the decomposition of ES is k,,, (the turnover number) At low

concentration of substrate, where [S] << K,, v is given by Eqn 3, with k , , , / K ,

being an apparent second-order rate constant

At high concentrations of S, where [S] >> K , , v is given by Eqn 4 and becomes

of ES to E and S is comparable or slower than the forward rate of reaction of ES

(Briggs-Haldane kinetics) The measured value of k,,, may also be a function of various and several microscopic rate constants (see Chapter 3)

When the activation energies for the catalytic steps have been sufficiently lowered, the binding of the substrate or the desorption of the product may become at least partially rate-limiting

4 Intra- and extra-cellular enzymes

Extracellular enzymes show little motility and are not very sensitive to the ionic strength of the solutions in which they are dissolved They are synthesised in one environment but are then placed in another outside the cell In particular, salt concentrations vary between the two environments; for example, calcium concentra-

tion outside the cell is lo4 times greater than in the cell Furthermore many

extracellular enzymes are produced as zymogens and are activated by removing a section of the protein Insensitivity to the environment is an essential requirement of

these enzymes if activity is to be maintained in extracellular fluids of uncertain composition Extracellular enzymes are usually cross-linked several times by dis- ulphide bridges The more oxidising environments generally encountered outside cells make the disulphide bonds correspondingly more stable Stability depends upon the thiol redox potential of the environment but, of course, the disulphide bond could be formed inside the cell

Intracellular enzymes exist in a more controlled medium and as the requirement for rigidity is less very few intracellular enzymes have disulphide cross-links Some enzymes, particularly those involved in electron transfer, have a very fixed fold Intracellular enzymes, in distinction to most extracellular ones, have a quaternary structure

Individual enzymes in vivo have different constraints and requirements Gener- ally, intracellular enzymes are required to maintain a constant concentration of the various metabolites and this may be achieved by having a wide variation in the reaction flux of the material through the various metabolic pathways The reaction rate will vary with [S] if the enzyme is working below saturation, K , >> [S], and the

rate is given by Eqn 3 However, extracellular enzymes are often faced with

dramatic changes in the concentration of their substrates and yet are required to maintain a steady flow of material for absorption and use by the cell The reaction rate will be independent of [S] if the enzyme is working under saturation conditions, [S] >> K,, and the rate is given by Eqn 4

Given the obedience to Michaelis-Menten kinetics one may obtain information

Fig 3 A reaction rate showing saturation kinetics has different rates, u , and u2, at different substrate concentrations, S, and respectively

about the optimal physiological concentration for the substrate [ 101 Obviously, the maximal change in rate for a change in substrate concentration is when S and v are

minimal (dv/dS is maximal at S = 0, v = 0) If there is an optimal physiological concentration for the substrate, enzymes could have evolved to bind the substrate more or less tightly to maximise the changes in rate with respect to changes in

substrate concentration The problem is illustrated in Fig 3, with two substrate concentrations S , and S, with respective rates v I and v, The minimal fractional change in substrate concentration (S,/S,) to obtain a given change in velocity x,

occurs at the K , for the reaction

From Eqn 2 the substrate concentration is given by

If v, = v, + x, the concentration of S, is given by

The derivative of S2/S, is zero when v l + v, = V (Eqns 7-9), i.e at the K , of the reaction

V = x + 28, = v, + v2 (9)

The optimal physiological concentration of the substrate is at its K , [10,1 I]

5 Regulation and thermodynamics

Biochemical reactions exist to bring about the net formation of a compound which

may be required either of itself or as the starting material of a further process A

metabolic pathway could involve the conversion of a large, or constantly maintained,

concentration of substrate S into an equally well maintained pool of product P (Eqn

10)

The sequence may involve the formation of several intermediates, B, C, D and

require the conversion of coenzymes w and y into x and z , respectively The chemical flux out of S and P may be controlled by: (i) changing the concentrations of S or P,

(ii) changing the concentrations of the coenzymes, (iii) changing the activities of the enzymes involved in the pathway

Although the thermodynamics of the step S -+ B may be unfavourable, net

synthesis can occur from left to right if the overaN equilibrium constant is favoura- ble Thermodynamically, Eqn 10 may be described by Eqn 11 where AG is the

Gibbs free energy change for the reaction under the given conditions and AGO is the

change in Gibbs energy for the reaction with reactants and products at the standard state concentrations

At thermodynamic equilibrium there can be no net flow Enzymes do not alter the position of equilibrium between unbound substrates and products If B, C or D are

removed rapidly their concentration will not correspond to their equilibrium value but rather to their steady state concentrations The total chemical flux out of C is given by the sum of the fluxes out of C into B and D The total chemical flux into C

is given by the sum of the flux of B into C and that of D into C The rate of

appearance of C is then given by the difference of these two sums (Eqn 12)

The flux through a sequence of reactions cannot generally be controlled by suppress- ing the activity of the enzyme catalysing any one of the reactions The positions of the reaction with respect to its equilibrium value and with respect to the degree of enzyme saturation are important If any of the steps in Eqn 10 are near equilibrium the rate of the reverse reaction will be similar to that of the forward reaction and this

step cannot, therefore, limit the rate of production of P It is often suggested that

control of metabolism through alteration of enzyme activity should be exerted at

reactions which are far from thermodynamic equilibrium If the substrate concentra- tion is not in excess and if the dissociation constants for substrate and product, K ,

and Kp, respectively, are very different it is conceivable that the overall equilibrium

constant K (Eqn 14) is not a good guide to the value of the equilibrium constant, K ,

(Eqn 13)

In this case, it is possible for an enzyme to apparently “alter” the position of equilibrium between substrate and product

6 Specificity and k , , , / K ,

Obedience to Michaelis-Menten kinetics yields interesting conclusions about the specificity of enzyme-catalysed reactions In vitro “non-specific” substrates are sometimes described as “poor” because they show a low value of kcat or a high value

of K, However, in vivo specificity results from a competition of substrates for the

active site of the enzyme If two substrates, S and S‘, compete for the same enzyme different conclusions could be reached about their relative specificity if rates of reaction or the K,s of the individual substrates are compared instead of their relative values of k C a t / K , If the enzyme catalyses the reaction of both S and S‘

(Eqn 15) the relevant equations may be obtained by the usual procedures (Eqns 16-18)

The relative rate of the two reactions is given by Eqn 19 whether the enzyme is working below or above saturation for both substrates Furthermore Eqn 18 reduces

to Eqn 19 even if the individual K,s or substrate concentrations are such that the enzyme is working below saturation for one substrate but above saturation for the other

Specificity between competing substrates is therefore given simply by relative values

of k , , , / K , and not by the individual values of k,,, or K , [ l l ] Specificity can

apparently be reflected in poor binding (high K , ) and/or slow catalytic steps (low

k , , , ) but specificity between competing substrates is controlled only by their relative

7 Rate enhancement and specificity - the two striking phenomena associated with enzyme-catalysed reactions are the rate enhancement and specificity

Enormous rate enhancements can be aclueved by enzymes compared with non-en- zyme-catalysed reactions, but the quantitative evaluation of the rate difference is not always straightforward For example, hydrolytic enzymes often exhibit rate enhance- ments of 108-10’2 compared with the spontaneous water-catalysed - or the acid- or base-catalysed - reaction at around neutral pH However, the mechanism by which the enzyme-catalysed reaction occurs is different from these relatively simple non- enzyme-catalysed reactions Enzymes generally utilize their functional groups to act

as nucleophilic, electrophilic, general acid or base catalysts (Chapter 5) It has been tempting, therefore, to speculate that it is the “chemical catalysis” or “mechanism”

which is responsible for the large rate enhancement brought about by enzymes There are thus two aspects of enzymic catalysis which should be distinguished: (i) the rate enhancement brought about by “chemical catalysis” relative to the “ un- catalysed” or “solvent-catalysed” reaction; (ii) the rate enhancement broughht about

by the reaction occurring within the enzyme-substrate complex compared with the

same chemical reaction in the absence of enzyme There appears to be nothing

unusual in the pathways by which covalent bonds are formed and broken in enzyme-catalysed reactions; the mechanisms used to account for ordinary chemical reactions are also applicable to nature’s reactions Chemical catalysis alone cannot explain the rate enhancement brought about by enzymes The forces of interaction

between the non-reacting parts of the substrate and enzyme are used to lower the

activation energy of the reaction The enzyme succinyl-CoA-acetoacetate transferase catalyses reaction 22, (R,CO; = acetoacetate and R,CO; = succinate) and pro- ceeds by the initial formation of an enzyme-CoA intermediate in which the coenzyme A is bound to the enzyme as a thiol ester of the y-carboxyl group of glutamate (Eqn 23)

(22) (23)

In turn this intermediate is generated by nucleophilic attack of the glutamate carboxylate on succinyl-CoA to give an anhydride intermediate (I) The second-order

rate constant for this reaction I is 3 X 10I3-fold greater than the analogous reaction

of acetate with succinyl-CoA (11) [ 121 It seems unlikely that the chemical reactivity

of acetate and the enzyme’s glutamate will be vastly different Similar chemical

reactions are therefore being compared and yet the non-reacting part of the enzyme

lowers the activation energy by 78 kJ/mole ( R T In 3 X lOI3)

pionate, should be able to fit into the active site Therefore the non-reacting part of

succinyl-CoA of molecular weight ca 770 lowers the activation energy by 72

kJ/mole ( R T In 3 x lo’,)

It is relatively easy to rationalise how enzymes discriminate against substrates

which are larger than the specific substrate - the substrate is simply too large to fit into the active site It is more difficult to explain why some small non-specific substrates that can bind to the active site react very slowly In the example cited above, it is not too difficult to imagine that the non-covalent interactions between the non-reacting part of succinyl-CoA and the enzyme can give rise to binding energies on the order of 70-80 kJ/mole which in turn can be utilised to increase the reaction rate However, it is not so easy to understand how, for example, an enzyme whch has iso-leucine (IV) as a specific substrate can discriminate against valine (V)

CH3CH ,z ,NH3 H3C ,NH3

‘Cti-CH H3C ,””-“< c 0; H3C / ‘CO;

It is of interest to note that the efficient enzymes, catalase, carbonic anhydrase and the nitrogenases have very small substrates (H,O,, CO, and N,, respectively)

for which it is not easy to distinguish between “reacting” and “non-reacting” parts

It is a great challenge to quantitatively understand the forces of interaction between these substrates and their enzymes

8 Specificity, induced fit and non -productive binding

According to the Induced Fit model of specificity the active site of the free enzyme (E) is in the “wrong” conformation and is catalytically inactive The binding of a

“g00d” substrate induces a conformational change in the enzyme making it catalyti- cally active (E’) (Eqn 24)

A “poor” substrate does not have enough favourable binding energy to compensate for the unfavourable conformational change in the enzyme Whereas some of the binding energy between the “good” substrate and the enzyme is used to “pay for” the conversion to the unfavourable, but active conformation of the enzyme (E’) The

free energy of binding good substrates is greater than the free energy of distortion of the enzyme Induced fit can be defined to explain specificity between very good and very poor substrates but not between substrates that all have sufficient binding energy to compensate for the unfavourable conformational change It is an energeti- cally expensive, but sometimes necessary, mechanism of specificity It mediates

against catalysis in the sense that the enzyme is not as catalytically effective as it would be if the enzyme existed in the catalytically active conformation in the free state [ 11,121

From Eqn 24, where K = (E)/(E), it can be seen that the observed k C a , / K , is reduced by a factor K , equivalent to the free energy required to distort the enzyme,

compared with the situation where the enzyme is initially present in the active conformation

Provided that the substrates being compared have sufficient binding energy to compensate for the unfavourable conformational change of the enzyme, then in- duced fit cannot explain specificity between competing substrates Compared with the situation where the enzyme is initially in the active conformation, the induced fit mechanism reduces the value of k , , , / K , for all substrates by the same fraction and therefore does not affect their relative rates

If the conformational change occurs upon initial binding to form the Michaelis

complex then the observed binding constant, K , , will be increased by a factor 1 / K but k,,, will be the same as it would be if all the enzyme were in the active

Non-productive binding is sometimes suggested to account for the low reactivity

of “poor” substrates by suggesting that they bind at sites on the enzyme where the catalytic reaction cannot take place

i.e the free energy of activation, represented by k , , , / K , , is not changed by

non-productive binding

9 Approximation, entropy and intramolecular reactions

It has long been thought that enzymes are effective by bringing reactants together at the active site of the enzyme whch is referred to as approximation or proximity However, it was not until 10 years ago that this effect was fully understood [14] The importance of this effect may be illustrated by considering a reaction between 2 molecules A and B to give a transition state AB” and comparing this with the same reaction occurring at the surface of an enzyme but without involving chemical catalysis by the enzyme (Eqn 27) i.e the enzyme is used to bring the reactants together but none of the functional groups on the enzyme are used to facilitate the reaction (see also p 32)

Historically, there was a delay in appreciating the magnitude of the entropic contribution to intramolecular and enzyme-catalysed reactions because of the wide variation observed in the rate enhancements of intramolecular reactions It is well known that intramolecular reactions (Eqn 28) in which the reactants are covalently bonded to one another, often proceed at very much faster rates than those of the analogous intermolecular reactions (Eqn 29) These intramolecular reactions are frequently taken as models for enzyme-catalysed reactions where the reactants are held close together in the enzyme-substrate complex (Eqn 30).It is therefore often

(29)

Typical rate enhancements and favourable equilibria of intramolecular reactions

are illustrated by some reactions of succinic acid and its derivatives [15] The

equilibrium constant for succinic anhydride formation from succinic acid (Eqn 3 1)

is 3 x lo5 moles/l more favourable than that for acetic anhydride formation from acetic acid (Eqn 32), an analogous intermolecular reaction A similar situation exists

analogous intermolecular formation of acetic anhydride (VII) A special explanation

for the rate differences which is peculiar to the activated complex is not required since the favourable reactions of intramolecular systems over their intermolecular counterparts are manifested in both rates and equilibria [ 161

The rate enhancements and favourable equilibria have units of concentration because a unimolecular reaction is being compared with a bimolecular one For this reason the rate enhancement is sometimes called the “effective concentration” or

“effective molarity”, which is the hypothetical concentration of one of the reactants

in the intermolecular reaction required to make the intermolecular reaction proceed

at the same rate as that of the intramolecular one [ 151

In general, however, the effective concentrations or rate enhancements of in- tramolecular reactions do not show a constant value but cover a very wide range from about 1 to 10’’ moles/] [15] For example, the intramolecular general base

catalysed enolization reaction of 4-oxopentanoate (VIII) in which the neighbouring carboxylate group removes a proton in a 5-membered ring transition state, shows an effective concentration of 0.5 mole/l compared with the analogous intermolecular reaction of (IX) However, the equilibrium constant for the dehydration of dimethyl-

amaleic acid (X) to give the 5-membered ring acid anhydride (XI) is about lo'*

moles/l times more favourable than that for acetic anhydride formation from acetic acid (Eqn 32)

There are different degrees of freedom lost in intramolecular and analogous intermolecular reactions which gives rise to large differences in the entropy change between the two systems For a non-linear molecule containing n atoms there are 3

degrees of translational freedom, i.e its freedom to move along the 3 axes in space;

there are also 3 degrees of rotational freedom representing the freedom of the whole

molecule to rotate about its centre of gravity This leaves 3n - 6 degrees of freedom associated with internal motions in the molecule, i.e vibrations and internal rota-

tions On forming the product of a bimolecular association reaction (Eqn 33) there

is a loss of 3 degrees of translational freedom and a loss of 3 degrees of rotational

freedom There is a gain of 6 new vibrational modes in the product However, in the

uni- and intramolecular reaction (Eqn 34) there is no net change in the number of

degrees of freedom of translation, rotation, and vibration upon forming the product The entropy associated with these motions may easily be calculated from the

In the gas phase, for a standard state of 1 mole/l and at 298 K the total loss of

TABLE 1

Typical entropy contributions from translational, rotational, and vibrational motions at 298 K [ 141

(J/K/mole) Three degrees of translational freedom; molecular weights 20-200;

Three degrees of rotational freedom

translational and rotational entropy for a bimolecular association reaction is about

- 220 J/ K/ mole and this value has only a small dependence upon the masses, sizes

and structures of the molecules involved However, this loss is often compensated to varying extents by low frequency vibrations in the product or transition state For several reactions having “ tight” transition states or covalently bonded products the change in internal entropy is about +50 J/K/mole so the total entropy change is predicted to be about - 170 J/K/mole, which is the value observed experimentally for many bimolecular gas phase reactions [14,16] In solution, the entropy change for

a bimolecular reaction is estimated to be only about 20 J/K/mole less than that in

the gas phase, at the same standard state of 1 mole/l

Since this large loss of entropy for a bimolecular reaction is avoided if the reactants are bound to an enzyme active site (Eqn 30) or converted to an intramo-

lecular reaction (Eqn 28) the maximum entropic advantage from approximation

may now be estimated (Eqn 35)

“Loose” “ Tight” Maximum

transition transition state or state or product product

Effective concentration (mole/l) lo2 lo8 10”

The theoretical most negative entropy change for a bimolecular reaction is about

-200 J/K/mole which is equivalent to 60 kJ/mole at 298 K and makes such a

reaction unfavourable by antilog (60/2.303 R T ) , a factor of 10” moles/l However,

this is very rare and a more general situation is that this large loss of translational and rotational entropy is compensated for by an entropy change of about +50

J/K/mole, resulting from changes in internal motions on going from the reactants

to the product or transition state and from differences in entropies of vaporisation of reactants and products giving bimolecular reactions a total entropy change of about

- 150 J/K/mole Since this loss of entropy does not take place for a reaction in which the reactants are bound to the active site of an enzyme (Eqn 30) or for an

intramolecular reaction (Eqn 28), the approximate maximum effective concentration

or rate enhancement for these reactions from entropic factors alone is about 10’ moles/l In the comparatively rare situation of a bimolecular reaction having a very

“loose” transition state or product then association will be even less entropically unfavourable since the entropy of the low frequency vibrations in the “loose” complex will counterbalance the large loss of translational and rotational entropy

[ 14- 161 The rate enhancement for the analogous intramolecular reaction will,

therefore, be smaller, perhaps in the range of 100 moles/l or less

In summary, rate enhancements of about 10’ moles/l may occur for intramolecu-

lar and enzymic reactions simply on the basis of the difference in entropy changes

between bimolecular and unimolecular reactions Reactions which show rate en- hancements greater than this are probably the result of additional contributions from potential energy differences Smaller rate enhancements may result either from

a “loose” transition state or product or from unfavourable entropy changes, such as the loss of internal rotations, and/or potential energy changes in the intramolecular reaction [15,16]

When allowance is made for the changes in strain energy and loss of entropy of internal rotations that occur upon cyclisation in intramolecular reactions the varia- tions in “effective concentrations” previously mentioned may be nicely accounted for [15,16]

The formation of the lactone (XIII) from the hydroxy-amide (XII) is free of strain and solvation effects and is, incidentally, a good model for the mechanism of hydrolysis of amides catalysed by chymotrypsin The rate enhancement for this intramolecular reaction is 10’ M and therefore agrees well with the theoretical prediction [40]

If the reactants A and B are bound to the enzyme tightly and in close proximity (Eqn 27) there will be little loss of entropy upon forming the transition state Reaction within the enzyme-substrate complex therefore has an entropic advantage over the uncatalysed reaction and k,,, may be 10’ times greater than k,,,,, even

though there is no chemical catalysis by the enzyme However, initial binding of the reactants A and B to the enzyme is, of course, entropically unfavourable and any increase in the rate of reaction brought about by the enzyme is given by the energy

of binding the enzyme and substrates, A and B, less the entropy of association of the

transition state and the enzyme [13,17] (see p 32) The binding energy may not appear as enthalpy as it is likely to be hydrophobic or electrostatic or have compensating enthalpy/entropy effects (see Section 13) Any excess of the intrinsic binding energy between the enzyme and substrates over this entropy loss will appear

as catalysis of the reaction

The idea of the binding energy being directly responsible for the rate enhance- ment brought about the enzyme has been substantiated by the observation that P-galactosidase catalyses the S,1 hydrolysis of P-galactopyranosyl pyridinium salts

The enzyme increases the rate of reaction by a factor of 10” and yet there is no chemical catalysis by the enzyme [ 181 Similarly, enzymes which catalyse [3,3]-sigma- tropic rearrangements, such as the Claisen type found in the conversion of choris- mate (XIV) to prephenate (XV), are unlikely to involve chemical catalysis by the enzyme The binding energy between the substrate and enzyme in the transition state must be responsible for catalysis

10 Decreasing the activation energy

In principle, the activation energy for any chemical reaction may be decreased by raising the energy of the ground state or lowering that of the transition state The

important contribution of enzyme-substrate binding energy in decreasing the activa-

tion energy of an enzyme-catalysed compared with a non-enzyme-catalysed reaction has been recognised for many years However, there has been much discussion as to whether the binding energy between the enzyme and substrate, between the enzyme and product or between the enzyme and some intermediate is the most important It

is now generally considered that maximum binding energy, i.e stabilisation, occurs between the substrate and enzyme in the transition state of the reaction It has also become common practice to use energy diagrams to rationalise aspects of enzymic catalysis [ 101

A simple enzyme-catalysed reaction (Eqn 36)

(36)

k‘,,

K %

may be depicted as in Fig 4 The “pit” into which the enzyme-substrate complex

falls reflects the favourable exergonic process of binding, the free-energy change associated with this is dependent upon the choice of standard states For a standard state of 1 mole/l, the free-energy change reflects the difference in free-energy between I mole of enzyme-substrate complex, at a concentration of I M, and 1 mole each of enzyme and substrate, also each at a concentration of 1M Fig 4 does not

exemplify the actual free energy of the species in the reaction mixture as carried out under the conditions of an experiment and as computed from their chemical potentials If the system is at equilibrium then E, S and ESc are at the same energy level as there is zero free-energy difference between the states at equilibrium For

example, if the equilibrium constant for binding, I / K s , is 103/M then the free

energy change at 298 K when the standard state is chosen as I M is - 17.3 kJ/mole and Fig 3 could be used to illustrate this negative free-energy change However, if the standard state is chosen as l o p 5 M then the corresponding free-energy change would be + 11.5 kJ/mole, which would give the free-energy difference between I

mole of enzyme-substrate complex, at a concentration of IO-’ M, and I mole each of

enzyme and substrate, also each at a concentration of I O - ’ M This could be illustrated by Fig 5 Altering the standard state changes the entropy difference between reactants and products in a bimolecular reaction

Fig 5 Standard state free-energy changes for an enzyme-catalysed reaction showing saturation kinetics,

(Eqn 36), with the standard state chosen so that the reaction occurs below saturation

The free-energy change accompanying the conversion of the enzyme-substrate complex to the enzyme-bound transition state is independent of the choice of standard states The standard free energy of activation for this process represents the

difference in free energy between I mole of ES' and I mole of ES and, apart from

Henry's law effects, it does not matter whether the concentration of these species is say, lo-' M or l o p 3 M If the reaction is plotted on a free-energy diagram such as

those illustrated in Figs 4 and 5 the ordinate does not indicate the actual free energy

of the species in the reaction mixture According to the Transition State Theory,

reactants, either E and S or ES, are in equilibrium with the transition state, ES' , and therefore there is zero free-energy difference between them

If energy diagrams are used to answer questions about enzymic catalysis some- times it is necessary to specify the standard states chosen or the working concentra- tions and sometimes it is not There are two cases commonly used, one is the comparison of enzyme-catalysed with non-enzyme-catalysed reactions and the other

is a series of enzyme-catalysed reactions in which the enzyme or substrate is assumed

to be modified in a particular manner [ 10,13,19,20]

Consider the hypothetical case where an enzyme binds equally well to the ground state and the transition state

Let us first compare a reaction where the transition state is the same in the

absence and presence of enzyme (Eqn 37)

It is necessary to state whether the comparison is made above or below saturation Above saturation, two first-order processes are being compared, k s and k,,, and it is

not necessary to specify the standard state The rate of reaction will be the same in

the enzyme substrate complex as it is in the absence of enzyme There can be no

catalysis or rate enhancement brought about by the enzyme (Fig 6a) Below

saturation the difference in the free energies of activation to give S* and ES' will

depend upon the choice of standard state because the uncatalysed reaction is a first-order process and the enzyme-catalysed reaction is a second-order one (Fig 6b)

However, there can be no catalysis by the enzyme as the free energy of activation to

give ES' will always be greater than that to give S'

Let us now compare a series of enzyme-catalysed reactions in which the enzyme

ES*

1 -

or substrate is modified so that the ground state and transition state are bound more

strongly by equal amounts If the rates of enzyme-catalysed reactions are to be

compared it is necessary to specify the working concentrations of enzyme and substrate to determine whether the reaction occurs above or below saturation This is independent of the arbitrary standard state chosen, the enzyme and substrate do not know the standard state which we have prescribed!

Above saturation, there can be no difference in the rates of reaction upon binding the ground state and transition state more tightly by the same amount (Fig 7a)

However, below saturation there would be an increased rate, (Fig 7b), but there is a

limit to this type of catalysis because if the enzyme binds the transition state and ground state very tightly, although the free energy of ES' will be reduced, the concentrations of enzyme and substrate required to maintain non-saturation condi- tions will be decreased These concentrations may be so low that catalysis would not

be observed

Specificity, therefore, could be reflected by the enzyme stabilising both the

transition state and the ground state for the better substrate, but this is true for only small changes in stabilisation energy If a non-reacting substituent of a specific substrate contributes a large amount of binding energy it is essential that this is not expressed in the ground state or intermediate states in order to avoid saturation