3 Reaction Mechanisms Part (iii) Enzyme Mechanisms By M. AKHTAR and D. C. WILTON Dept. of Physiology and Biochemistry The University of Southampton SO9 5NH From a mechanistic viewpoint an enzymic reaction may be separated into two broad stages. The first stage is concerned with the formation of the Michaelis Complex in which the substrate is attached to the enzyme surface in an arrange- ment that is appropriate for the second stage the catalysis to occur. Understand- ing the formation of the Michaelis Complex requires knowledge of the physico- chemical forces which direct the approach of the substrate to the active site the molecular interactions between substrate(s) and the binding groups which hold the complex together and the features which regulate the effectiveness of the enzyme binding-site groups.The investigations pertinent to the catalytic stage include the identification of the catalytic groups determination of the chemical nature of the intermediates involved in the overall process the elucidation of the ideal special arrangement of the substrates with respect to the catalytic groups and determination of any factors which may influence the reactivity of the catalytic groups. This report summarizes some of the major approaches currently being used which shed light on the nature of the active sites of enzymes and proteins. Par- ticular emphasis is laid on the identification function and regulation of groups involved in enzyme catalysis. 1 The Binding of Substratesto the Active Sites of Proteolytic Enzymes and the Influence of Secondary Interactions on Catalysis It has been recognized for a long time that low molecular weight synthetic substrates may be hydrolysed by proteolytic enzymes such as pepsin through a mechanism and at a site normally used for the hydrolysis of protein substrates ; however the rates of hydrolysis for the latter are usually several orders of magni- tude faster than those for the best synthetic substrates.These observations led to the view that substrate groups remote from the bond to be cleaved may play an important role in modulating the activity of the catalytic groups of enzymes. A more quantitative examination of this view has now been made possible by rate specificity and X-ray crystallographic studies carried out using systematically modified peptide substrates.167 M. Akhtar and D.C. Wilton Papain can accommodate at least seven amino-acid residues at the active site’ (Scheme 1). The regions involved in the binding are termed ‘subsites’ and are designated as S,-S and S;-S; the hydrolysis occurring between S and S; . The nature of the products formed from the hydrolysis of the substrates (lH4) by papain has permitted the deduction that the substrates align at the active site as shown in Scheme 1. An aromatic amino-acid residue when present always occupies the subsite S2 and the hydrolytic cleavage occurs not at the peptide bond linking this aromatic residue but at the peptide bond adjacent to it.’ This method of mapping the active site has been extended2 to carboxypeptidase A.1 H-Ala-Ala-Phe-Lys-Ala.NH (4) H.Ala-Phe-Ala-Ala.OH (3) HePhe-Ala-Ala-Ala-OH (2) H-Phe-Ala-LysOH (1) t 4 I s I s I s 1 s; I ///////1//// 11111 Enzyme Scheme 1 The studies carried out using compounds in which the peptide bond under- going cleavage is replaced by ester or amide linkages (5)have shown that papain- catalysed hydrolysis occurs via the mechanism of Scheme 2 (for details see ref. 3). The first step is the formation of the Michaelis Complex (5)(6),which then rearranges to give the acyl enzyme intermediate (7) and finally hydrolysis yields the product (8). 0 //R-C-X + Enz-S-H (5)(6) 0 //R-C-S-Enz + HX (5) (6) (7) 1lH20 0 R-C-OH// + (6) (8) X = leaving group; R = peptide chain Scheme 2 We will now consider the experimental approach for assessing the contribution to the overall catalysis which each part of the substrate molecule may make by binding to the enzyme.The ratio k,,,/K has been suggested4 to be a suitable parameter for determining the catalytic specificity of proteolytic enzymes ; as A. Berger and I. Schechter Phil. Trans. Roy. Sue. 1970 B257,249. ’ I. Schechter European J. Biochem. 1970 14 5 16. M. R. Hollaway Ann. Reports (B) 1968 65 601. M. L. Bender and F. J. Kezdy Ann. Rev. Biochem. 1965 34 49. Reaction Mechanisms -Part (iii) Enzyme Mechanisms an approximation a higher k,,,/K value indicates a better substrate. In the case of papain hydrolysis an impressive influence on productive binding is noted when structural modifications are introduced at groups remote from the bond to be cleaved.’ The k,,,/K data in Scheme 3 show the adverse effect of the progressive replacement of the amide group adjacent to the bond to be cleaved by methylene groups [(10)--(12)].Also to be noted is the improvement in the k,,,/K value when the N-terminal residue in (lo) PhCO- is replaced by PhCH2CH(NHAc)-CO-as in (9). (9) AcNH 0 0I II II 1Ph-CH,-CH-C-NH-CH2-C-O-C6H4N02 kcaJKm 1.7 x lo7 (10) 0 0 II II Ph-C-NH-CH2-C-O-C6H4N02 2.2 x lo5 (1 1) 0 0 II IIPh-C-CH2-CH2-C-0-C6H4N02 4.5 x lo3 0 Scheme 3 The available kinetic data,5 together with the knowledge of the X-ray structure of papain,6 have been married into a pictorial view of the enzyme mechanism (Scheme 4).5,7 The model building used L-phenylalanylalanine amide (13) as the ~ubstrate.~ It is suggested5 that the aromatic residue of (13)is accommodated in the hydrophobic part of the cleft comprising the side-chains of the amino-acid residues tyrosine-67 tryptophan-69 phenylalanine-207 alanine-160 valine-133 and valine-157.The -NH group (marked as NA)of the substrate (13) is then within hydrogen-bonding distance of the side-chain carboxy-group of aspartic acid-158. The interaction of the aromatic ring with the hydrophobic groups allows the -NH-and -CO-groups of the amide linkage of the substrate to hydrogen-bond more effectively with C=O of aspartic acid-158 and -NH-of glycine-66 respectively. The fixing of the amide bond in this fashion results in severe non-bonded interaction between the bond about to be cleaved and the a-CH group of histidine-159.From the model of the enzyme-substrate complex it would appear that non-bonded interaction could be relieved by forcing the carbon atom of the amide or ester bond about to be cleaved towards sp3 hybri-dized configuration through attack by the thiol group of cysteine-25. Thus a crucial step in the overall reaction has been performed. G. LoweandY. Yuthavong Biochem. J. 1971 124 107. J. Drenth J. N. Jansonius R. Koekoek H. M. Swen and B. G. Wolthers Narure 1968 218 929; J. Drenth J. N. Jansonius R. Koekoek L. A. A. Sluyterman and B. G. Wolthers Phil. Trans. Roy. Soc. 1970 B257 231. B. G. Wolthers J. Drenth J. N. Jansonius R. Koekoek and H. M.Swen Proceedings of the International Symposium on Structure-Function Relationships of Proteolytic Enzymes Munksgaard Copenhagen 1970 p. 272; J. Drenth J. N. Jansonius R. Koekoek and B. G. Wolthers Adv. Protein Chem. 1971 25 79. M.Akhtar and D. C. Wilton Scheme 4 The model could rationalize5 why papain shows stereospecificity for L-amino- acids in both positions of the substrate (13). With L-configuration the a-CH bond (*H)points into the body of the enzyme; amino-acids with D-configuration will not be tolerated since this would require interchanging the position of the a-hydrogen atom (*H)with the amino-acid side-chain thus giving rise to sterically unfavourable interactions. Similar reasons can be envisaged for the requirement of L-configuration for the aromatic amino-acid.Attention is drawn to a related X-ray crystallographic study' with chymotryp- sin A. The chloromethyl ketones of the structure (14a) irreversibly inactivate chymotrypsin through alkylation at histidine-57. The crystallographic structure of the enzyme-inhibitor complex (14b) has highlighted the importance of the secondary interactions existing between the active site and the amide bonds a p and y of the substrate (14) away from the bond to be cleaved. 0 Me 0 Me 0 Ph 0 I1 1 I1 I I1 I I1 Me.C-NHCHC-NHCH.C-NH.CH.C-CHC-CH2-X Y B o! (14a) X = C1 (14b) X = protein D. M. Segal J. C. Powers G. H. Cohen D. R. Davies and P. E. Wilcox Biochemistry 1971 10 3728. Reaction Mechanisms -Part (iii)Enzyme Mechanisms The specificity and the mechanism of action of another proteolytic enzyme pepsin has also been extensively inve~tigated,~ though studies on its structure are at present at a rather less advanced stage.Pepsin consists of a single poly- peptide chain and has a molecular weight of about 35 000. Unlike the cysteine and serine proteases which catalyse the hydrolysis of an amide linkage through the formation of an acyl enzyme intermediate pepsin catalysis appears to involve the formation of an amino-enzyme intermediate. An abbreviated pathway for the overall reaction using benzyloxycarbonyl-L-tyrosyl-L-tyrosine(15) as substrate is shown in Scheme 5. The mechanism has received support from trans- peptidation experiments" and from the fact that incubation of unlabelled (15) with labelled (17) or (18) results in the incorporation of radioactivity in the recovered starting material only from (17) and not from (18)." Detailed discus- sions on the mechanism of action of pepsin are available.' ',12 RO R I II I ZNHCHC-NHCH-COOH + Enz -XH (15) R R I I CEnz -NHCH-COOH + ZNH-CHCOOH (16) (17) R 1 I Enz -XH + NH,.CHCOOH (18) R = tyrosine residue ;Z = benzyloxycarbonyl Scheme 5 With small synthetic substrates of structure A-X-Y-B where the X-Y bond is cleaved pepsin exhibits a preference for Phe in the X-position and for Trp Phe and Tyr in the Y-position.Scheme 6 shows some of the available on the peptic cleavage of the Phe-Phe bond in substrates (19H22). The replacement of the N-terminal group from benzyloxycarbonyl (Z) to Z-Gly- to Z-Gly-Gly- has progressive favourable effect on the k,,,/K values.It should be noted that there is a decrease in the kinetic parameter when the benzyloxycarbonyl group in (21) is replaced by a hydrogen atom as in (22). Thus it would appear that the presence of a hydrophobic benzyl group two amino-acid residues away from the amino terminus of Phe-Phe has a marked J. S. Fruton Adv. Enzymol. 1970 33 401. lo H. Neumann Y. Levin A. Berger and E. Katchalski Biochem. J. 1959 73 33. " J. S. Fruton S. Fujii and M. H. Knappenberger Proc. Nat. Acad. Sci. U.S.A. 1961 47 759. l2 J. R. Knowles Phil. Trans. Roy. SOC. 1970 B257,135. G. P. Sachdev and J. S. Fruton Biochemistry 1969 8 423 1. M.Akhtar and D. C. Wilton kcatlLl Z-Phe-Phe-OX (19) 3.7 Z-Gly-Phe-Phe-OX (20) 7.8 Z-Gly-Gly-Phe-Phe-OX (21) 180 H-Gly-Gly-Phe-Phe-OX (22) 3.0 Z = benzyloxycarbonyl ;OX = leaving group Scheme 6 effect on the catalytic efficiency. These and related results on pepsin14 coupled with those from similar studies on carboxypeptidase A,2 thermolysine,l5 throm-bin,16 rennin,14 and bacterial proteina~es'~ have highlighted the role of group(s) remote from the bond to be cleaved in modulating the activity of the catalytic groups of enzymes. The most dramatic illustration of the effect of substrate binding on the con- formation of active-site groups remains the example of carboxypeptidase A in which it was noted that the binding of the substrate glycyl L-tyrosine to the enzyme promotes the displacement of the hydroxy-group of Tyr-248 through a distance of 12A so that it may participate in catalysis.l7 Comprehensive treatment of the structure and function of proteolytic enzymes is available in reviews," reports of symposia," and books.20 2 Bisubstrate Reactions The Effect of the Binding of One of the Substrates on the Activity of the Catalytic Group(s) We have seen above how the substrate groups away from the sensitive bond in- fluence the catalytic activity of the hydrolytic enzymes. A related phenomenon is noted with enzymes which participate in reactions involving two substrates. Isocitrate dehydrogenase catalyses the oxidative decarboxylation of isocitrate (23) into a-oxoglutarate (28) ;the overall conversion occurs through the sequence of Scheme 7.This involves the NADP-dependent oxidation of (23) to give the hypothetical intermediate (26) which decarboxylates to the enol (27) ; the latter after rearrangement then furnishes the product (28). In the formal reversal of the reaction one of the P-hydrogen atoms of (28) would be expected to exchange l4 I. M. Voynick and J. S. Fruton Proc. Nat. Acad. Sci. U.S.A. 1971 68 257. l5 K. Morihara and H. Tsuzuki European J. Biochem. 1970 15 374. l6 R. K. H. Liem R. H. Andreatta and H. A. Scheraga Arch. Biochem. Biophys. 1971 147 201. l7 W.N. Lipscomb G. N. Reeke J. A. Hartsuck F. A. Quiocho and P. H. Bethge Phil. Trans. Roy. Soc. 1970 B257 177. I' D.M. Blow and T. A. Steitz Ann. Rev. Biochem. 1970 39 63; also see G.P. Hess and J. A. Rupley Ann. Rev. Biochem. 1971 40 1031. l9 A Discussion on the Structure and Function of Proteolytic Enzymes organized by D. C. Phillips D. M. Blow B. S. Hartley and G. Lowe Phil. Trans. Roy. SOC.,1970 B257 65-265 ;see also ref. 7. 'O 'The Enzymes,' ed. P. D. Boyer Academic Press New York and London 1971 vol. 4; C. J. Gray 'Enzyme-Catalysed Reactions,' Van Nostrand Reinhold London 1971. Reaction Mechanisms -Part (iii) Enzyme Mechanisms 173 with the protons of the medium. Such a reaction however is observed2’ only when a-oxoglutarate (28) and the enzyme are incubated in the presence of NADPH. The experiment highlights the fact that the catalytic group(s) par- ticipating in equilibrium c (Scheme 7) become available only after the coenzyme site is occupied by NADPH.HO-CH-COOH +?=C-COOH 1 I //O H-C-C + NADP NADPH + \ I 0-H (24) (25) CH2-COOH CH,-COOH] (23) 0 COOH HO CH2-COOH CH2-COOH (28) (27) Scheme 7 The pyridoxal phosphate dependent serine transhydroxymethylase (SHM) catalyses the transfer of a C ,-unit from methylenetetrahydrofolate (30) to glycine to give tetrahydrofolate (29) and serine. It has been suggested22 that the conver- sion occurs through the sequence a-d (Scheme 8). The first event in the con- version is the reaction of glycine with the enzyme-pyridoxal complex (32) to give the Schiff-base intermediate (33) which undergoes deprotonation to give a resonance-stabilized carbanion or its equivalent species (34).The carbanion then reacts with ‘formaldehyde’ released at the active site from (30) to give the pyridoxal derivative of serine (35). There is a four-fold increase23 in the rate of reaction of glycine with the pyridoxal-enzyme complex (equilibrium a) when the second- substrate site is occupied by tetrahydrofolate. An even more dramatic influence of the binding of tetrahydrofolate to (32) is reflected on an exchange rea~tion~~.~~ occurring through the combination of equilibria a and b. No detectable cleavage of the C-H bond of glycine is observed when glycine is incubated with SHM ; however the addition of tetrahydrofolate results in a rapid exchange of the S-hydrogen atom of glycine with the protons of the medium.22 The experiments show that the activity of the catalytic groups participating in equilibria a and b is greatly enhanced by the binding of tetrahydrofolate to the enzyme.Related ” G. E. Lienhard and I. A. Rose Biochemistry 1964 3 185; Z. B. Rose J. Biol. Chem. 1960,235 928. 22 P. M. Jordan and M. Akhtar Biochem. J. 1970 116 277. 23 L. Schirch and M. Ropp Biochemistry 1967 6 253. 24 L. Schirch and W. T. Jenkins J. Biol. Chem. 1964 239 3801. 174 M. Akhtar and D. C. Wilton H Rl&R2 I HsTH&)OH + CH,-NHR NH2 N R3 I H ENZ (31) (29) N-X = N-H + (30) N-X = N=CH H HYHCOOH . -yCOOH oH-~Hy~oOH Serine 2 7-t 7 7 ‘HCHO’ generated (32) from (30) f” (29) and (30) R = Ph-C-(gIutamyl),; (32) R’ = CH,.043 R2 = OH R3 = Me I1 0 Scheme 8 phenomena showing the effect of cosubstrates on hydrogen atom exchange reactions have been noted with several enzymes including deoxycytidylate hydr~xymethylase,~~ malate synthetase,26 and citrate ~ynthetase.~~ 3 The Effect of Changes in Enzyme Structure on Catalytic Activity We have so far dealt with the effect of substrate structures on the activity of the catalytic groups of enzymes.The availability of rapid methods for the synthesis of peptides and proteins has permitted the study of another aspect. The staphylo- coccal nuclease catalyses the hydrolysis of 3‘,5’-phosphodiester linkages in both DNA and RNA to give the corresponding 3’-phosphomononucleotides. A limited tryptic hydrolysis of the nuclease yields two large fragments nuclease- T-(6-48) and nuclease- T-(49- 149).These two fragments bind non-covalently to form nuclease-T’ a complex structurally similar to the nuclease but only 8-10% as active. The reported three-dimensional structure of the nuclease suggests that the amino-acid residues Asp-21 Arg-35 Asp-40 and Glu-43 are functionally important.28 Peptide analogues corresponding to the nuclease region 6-47 have been prepared with single site substitutions at the four positions (residues 21 35,40 and 43). These peptides were tested for their ability to form complexes with nuclease-T-(49-149) and for catalytic activity when the complex formation o~curred.’~Only the substitution that did not alter the net charge allowed the complex formation i.e. the replacement of Asp-21 with glutamic acid allowed the 25 Yun-Chi Yeh and G.R. Greenberg J. Biol. Chem. 1967 242 1307. 26 H. Eggerer and A. Klette European J. Biochem. 1967 1 447. 27 H. Eggerer Biochem. Z. 1965,343 11 1. 28 A. Arnone C. J. Bier F. A. Cotton V. W. Day E. E. Hazen D. C. Richardson J. S. Richardson and A. Yonath J. Biol. Chem. 1971 246 2302. 175 Reaction Mechanisms -Part (iii) Enzyme Mechanisms complex formation but not when it was replaced with asparagine. The study29 also showed that the formation of the catalytically active complex required essentially the entire amino-acid sequence of nuclease- T(6-48). In another study3’ by the same group it has been shown that a catalytically active complex is formed when nuclease- 7349-149) binds to nuclease- T-(1-126). The redun- dant portion (residues 49-126) of nuclease-T-(1-126) hangs away from the molecule as shown in the Figure and does not interfere with the ordered structure of the complex.I26 and nuclease-T-( 1-126) (I). Figure Interaction of nuclease-T-(49-149) (0) The ordered structure is surrounded by the circle. (Simpl$ed adaptation of aJigure from ref 30) Studies31 on leucyl-tRNA synthetase also indicate that the enzyme may be broken into two fragments by proteolysis and then rejoined non-covalently to give full activity. The larger of the two fragments is able to catalyse the ATP/PP exchange but has no synthetase activity. 4 The Identification of Catalytic Groups Active-site-directed Inhibitors.-Since the early days of enzymology a major approach to the study of enzyme mechanisms has been the use of inhibitors that are able to chemically modify reactive groups on the protein.This line of research which has been extensively reviewed in the past has evolved the concept of the active-site-directed irreversible inhibit~r.~~,~~ These reagents are designed such that they have a very close similarity to the normal substrate and therefore will form a Michaelis Complex with the enzyme. However the presence of a suitably reactive group on the inhibitor should allow it to react chemically with residues at the active site that are involved in the catalytic process. Such inhibitors are 29 I. M. Chaiken and C. B. Anfinsen J. Biol. Chem. 1971,246 2285. 30 H. Taniuchi and C. B. Anfinsen J. Biol. Chem. 1971 246 2291 ; I.Parikh L. Corley and C. B. Anfinsen J. Biol. Chem. 1971 246 7392. 31 P. Rouget and F. Chapeville European J. Biochem. 1971 23 459. 32 L. Wofsy H. Metzger and S. J. Singer Biochemistry 1962 1 1031. B. R. Baker ‘Design of Active-Site-Directed Irreversible Enzyme Inhibitors,’ Wiley New York 1967. 176 M. Akhtar and D. C. Wilton eagerly sought by both mechanistic enzymologists and chemical pharmacologists since by their very nature they have tremendous potential both as tools for investigating enzymes and for therapeutic purposes. The object of this part of the review is to highlight recent developments in the field of active-site-directed irreversible inhibitors. The literature up to 1969 is available in reviews.34 For convenience we have discussed in separate sections the reactions of inhibitors on each class of enzyme.Oxidoreductases. There has been much interest in the development of alkylating inhibitors structurally related to NADf. The pyridine analogue (36) has been successfully used to alkylate both a histidine and a cysteine at the active site of lactic dehydr~genase.~ More recently the NAD analogue 5-bromoacetyl-4- methylimidazole dinucleotide (37) in which the alkylating bromoketone is in the adenine part of the cofactor has been incorporated into liver alcohol de- hydrogenase. Diazo-compounds are finding increasing use in the labelling of amino-acids at the active site of enzymes. Their special advantage lies in the fact that once the inhibitor is bound at the active site then photolysis of the diazo-group will generate a highly reactive carbene RR'C=N 5 RRT This carbene has the potential to react with any amino-acid residue and in particular it is able to insert into the C-N bond of aliphatic amino-acids and phenylalanine.These amino-acids are immune to modification by any normal reagent. 0 0 I1 II uC-CH2Br ("d' N C-CH,Br 0-0-i (36) I I Ribose-0-P-0-P-0-Ribose I1 11 00 (37) + Me,N-CH,-CH-CH,-COO-0 I II 0 \ C-CH,Br II 0 (38) (39) 34 E. Shaw Physiological Rev. 1970 50 244; B. E. Vallee and J. F. Riordan Ann. Rev. Biochem. 1969 38 733. 35 C. Woenckhaus J. Berhauser and G. Pfleiderer Z. physiol. Chem. 1969 350 473. 36 C. Woenckhaus and R. Jeck Z.physiol. Chem. 1971 352 1417. Reaction Mechanisms -Part (iii) Enzyme Mechanisms 177 The [''C]diazoacetate ester of 3-hydroxymethylpyridine (38) has been enzymi- cally exchanged with NAD to give the 3-diazoacetoxymethyl analogue of NAD. This conversion was achieved by the enzyme diphosphopyridine nucleosidase which normally hydrolyses NAD to give adenosine diphosphoribose (ADPPR) and nicotinamide but will also catalyse an exchange whereby the diazo-derivative of nicotinamide (38) may be incorporated into the ADPPR moiety Nicotinamide Adenine (38) Adenine \ I + (38) --* I I Ribose-P-P-Ribose Ribose-P-P-Ribose +Nicotinamide The diazo-analogue of NAD was allowed to react with yeast alcohol dehydrogen- ase and then photolysed ;subsequent hydrolysis of the enzyme gave a number of radioactive products.37 This lack of specificity in the reaction of the carbene with groups on the enzyme is disappointing.It is hoped that more conclusive results will be obtained by synthesizing suitable diazo-analogues that are more rigidly held to the coenzyme binding site of the enzyme. Transferases. Identification of a histidine residue at the active site of the enzyme carnitine acetyl transferase has been achieved by the inhibitor bromo- acetylcarnitine (39) ; to date however no catalytic function has been ascribed to this residue.38 Hydrolysases. The hydrolysases and in particular the proteolytic enzymes and cholinesterase have long been a major focus of attack for the classical alkylating inhibitors.More recently particular use has been made of the potentially long- lived acyl intermediate formed during hydrolysis. Reactive groups are attached to the acyl moiety in order that they might react with other residues at the active site of the enzyme. Westheimer's group has formed acyl intermediates of both chymotryp~in~~ and trypsin4' where a diazo-group is attached to the acyl moiety. In the case of trypsin photolysis resulted in the radioactive inhibitor (40) being inserted into an alanine residue since subsequent hydrolysis of the enzyme gave radioactive glutamic acid. The overall reaction is shown in Scheme 9. This is the first time a non-polar amino-acid has been identified at the active site of an enzyme by a purely chemical approach. A major achievement in the chemical modification of proteins is to introduce bifunctional alkylating inhibitors.Notable successes have been achieved with this approach in the case of papain ficin and bromelain using dibrom~acetone.~ Although available evidence suggests that there are two catalytically important carboxy-groups in pepsin only one of these has been identified using 14C-labelled dia~oacetylphenylalanine.~~ Fruton and his co-workers have now reported the 37 D. T. Browne S. S. Hixson and F. H. Westheimer J. Biol. Chem. 1971 246 4477. '* J. F. A. Chase and P. K. Tubbs Biochem. J. 1970 116 713. 39 J. Shafer P. Baronowsky R. Laursen F. Finn and F. H. Westheimer J. Biol. Chrm. 1966 241 42 1. 40 R. J. Vaughan and F. H. Westheimer J. Amer. Chem. SOC.,1969,91 217.41 S. S. Husain and G. Lowe Biochem. J. 1968 108 855; 1970 117 333 341. J2 R. S. Bayliss J. R. Knowles and E. Wybrandt Biochern. J. 1969 113 377. 178 M. Akhtar and D.C. Wilton Trypsin + + 0 C0,Et \c* CH,I CH N CH \ /H0-c-c 1 YH2 (I I C=O/\ NH HN/'C=O 0 C0,Et\ I0-c-c:I CH CH CH2 * -1 Iy 2 I\ C=O NH Hi \C=O COOH I I *CH2 YHZ + I NH,-C-H I COOH (38) Scheme 9 use of a bifunctional inhibitor l,l-bis-(diazoacetyl)-2-phenylethane(41) to inactivate pepsin.43 The precise site of cross-linking has yet to be identified. A lysine has been implicated at the active site of adenosine deaminase by using the inhibitor 9-(p-bromoacetamidobenzyl)adenine (42).44 C0,Et 0 0 I II II C=N NZCH-C C-CHN, I \/ c=o CH I I 0 CH NO (40) Me I p N%N> c=o 'N I 0 0 NH-C-CH,Br II (42) 43 S.S. Husain J. B. Ferguson and J. S. Fruton Proc. Nat. Acad. Sci. U.S.A. 1971 68 2765. 44 G. Ronca M. F. Saettone and A. Lucacchini Biochim. Biophys. Acra 1970 206 414. Reaction Mechanisms -Part (iii) Enzyme Mechanisms Lyases. The irreversible inhibition of acetoacetate decarboxylase by 2,4-dini- trophenyl propionate (43) has been successfully used to investigate the pK of the active-centre lysine re~idue.~’ The pH dependence of acylation of this group by (43) indicated that it had a pK of 5.9 about 4 pH units lower than that of a normal lysine. The anomalous pK allows this lysine to be in the non-protonated form at pH 6 the enzyme pH which is normally required for Schiff-base formation between enzyme and substrate.Isornerases. Studies on triose phosphate isomerase have included the use of two types of active-site-directed inhibitors 3-halogenoacetol phosphate (44)46947 and glycidol phosphate (45).48 In both cases the site of esterification is a glutamic acid residue and sequencing of the amino-acids around this residue has estab- lished that it is in fact the same glutamic acid residue that is modified with each inhibitor. This glutamate residue has been ascribed a basic function in the reaction mechanism. I PX Ll H 0 (‘6 It /\ Enz -C-O-CH2-CHZ-CH2O@ +chemical reduction 0 II Enz -C-O-CHZcX I c=o I CH208 (44) AIlosteric-site-directed Inhibitors. There is now considerable interest in the design of allosteric-site-directedinhibitors of key regulatory enzymes.Work on glutamic dehydrogenase (GDH) has already established the importance of a tyrosine residue at the binding site of the allosteric effector GTP.49 Now the covalent attachment of the oestrogen analogue diethylstilboestrol to the allosteric site of this enzyme has been reported,” Tritium-labelled bromoacetyldiethyl-45 D. E. Schmidt and F. H. Westheimer Biochemistry 1971 10 1249. 46 F. C. Hartmann Biochemistry 1971 10 146. 47 A. F. W. Coulson J. R. Knowles J. D. Priddle and R. E. Offord Nature 1970 227 180; A. F. W. Coulson J. R. Knowles and R. E. Offord Chem. Comm. 1970 7. 48 J. C. Miller and S. G. Waley Biochem. J. 1971 123 163; S. G. Waley J.C. Miller 1. A. Rose and E. L. O’Connell Nature 1970 227 181. j9 N. C. Price and G. K. Radda Biochem. J. 1969 114 419. J. Kallos and K. P. Shaw Proc. Nut. Acad. Sci. U.S.A.. 1971.68. 916. M. Akhtar and D.C. Wilton stilboestrol (46) was prepared and successfully incorporated into the enzyme to the level of one mole of steroid per mole of enzyme sub-unit. The radioactivity was removed by treatment with alkali or hydroxylamine indicating that the steroid was covalently bound through an ester linkage. The alkylation had ‘fixed’ the enzyme in its Y conformation where it has little glutamate dehydro- genase activity but enhanced dehydrogenase activity towards alanine GTP oestrogen GDHW)4 ADP GDH(Y) glutamate activity alanine activity The synthesis of diazomalonyl derivatives of cyclic-AMP has led to the successful labelling of the cyclic-AMP binding site of the enzyme phosphofructokinase.Tritiated 02’-(ethyl-2-diazomalonyl)cyclic-AMP (47) was photolysed in the presence of the enzyme to generate a reactive carbene and resulted in covalent binding. Prior to photolysis (47) was able to mimic the effect of cyclic-AMP and a direct competition between (47) and cyclic-AMP for the allosteric site was also observed. The authors did not measure the enzyme activity of the covalently modified phosphofructokinase. The Active Sites of Antibodies. The major problem in synthesizing active-site- directed irreversible inhibitors capable of chemical modification of the enzyme is the chemical limitation imposed by the nature of the natural substrate.A unique alternative offers itself in the field of immunology where the substrate may be first decided upon and then the antibody is produced to that substrate. In practice the substrate (hapten) is linked to a protein in order to elicit an immune response. The dinitrophenol hapten has been extensively used to produce the necessary antibody and then suitably activated DNP derivatives have been prepared to investigate the active site of the antibody. Compounds of the type (48) have allowed the identification of a tyrosine residue present in the light chain of the immunoglobulin.52 Bromoacetyl derivatives of the DNP hapten have now been used53 in which the bromoacetyl moiety is placed at varying distances from the 51 D.J. Brunswick and B. S. Cooperman Proc. Nut. Acud. Sci. U.S.A. 1971,68 1801. 52 E. J. Goetzl and H. Metzger Biochemistry 1970 9 3862. 53 D. Givol P. H. Strausbach E. Hurwitz M. Wilchek J. Haimovich and H. N. Eisen Biachemistr.y 1971 10 346 1. Reaction Mechanisms -Part (iii) Enzyme Mechanisms DNP grouping (49a4) (see Table). With this series of active-site-directed reagents both a tyrosine in the light chain and a lysine in the heavy chain were alkylated. However the proportion of each amino-acid that is alkylated depends on the distance between the DNP grouping and the bromoacetyl grouping as shown in the Table. Diazonium fluoroborate NO (48) Table of the radioactivity in TYr LYS Compound (light (heavy chain) chain) DNP.NH-CH2CH,-NH.C0.CH,Br (49a) 96 4 DNP-NHCH,-CH,CH.NH.COCW,Br (49b) 87 13 COOH DNP.NH.CH ,-CH ,CH ,-CH-NHCO-CH ,Br (49c) 66 34 COOH DNP-NHCH ,.CH 2.CH2.CH ,.CHON HCO-CH2Br (49d) 5 95 COOH 0 I1 -CH-C-(NH),I 0 II -C-CH,Br Lysine I ENH A further extension of this work was to make a bifunctional alkylating in- hibitor (50) in which the two bromoacetyl groups were at the critical distances as determined from the above results (see Table).With this reagent the authors were able to cross-link light and heavy chains through their respective tyrosine and lysine residues. Thus it would appear that these two residues located in two different chains at the active site of the immunoglobulin are about 5 A apart. Another approach to the structural determination of the active site of anti-bodies is the use of nitrophenyl derivatives (51) capable of generating nitrenes on photolysis.Nitrophenyl azides (51) have been successfully incorporated into antibodie~.~~ s4 G. W. J. Fleet R. R. Porter and J. R. Knowles Nature 1969 224 51 I 182 M. Akhtar and D. C. Wilton The Trapping of Covalent Intermediates.-There are at least three bio-organic approaches through which an intermediary stage in an enzymic reaction may be frozen to identify the group(s) involved in catalysis. In a hydrolytic reaction of the type Enz-XH + A-B Enz-X-A + BH HO Enz-XH + A-OH the group -XH may be identified if the reaction is performed under conditions (temperature or pH) which favour the formation of the complex Enz-X-A but are unsuitable for its further decomposition by the reaction of equilibrium b.This approach has been used to identify the serine residue on the active site of alkaline phosphatase.” Alternatively in some cases the use of a substrate analogue A-B has permitted the isolation of the complex Enz-X-A’ because its formation by the reaction of equilibrium a is faster than its hydrolytic decom- position (Eq. b). This approach has been successfully employed in the case of some protolytic enzymes and the appropriate examples are discussed in ref. 3. A second approach may be envisaged for identifying catalytic groups of enzymes which participate in group transfer reactions occurring through the sequence A-B + Enz-X-H Enz-X-A + BH Enz-X-H + A-Y In these cases the intermediate Enz-X- A may accumulate when the incuba- tion is conducted in the absence of the acceptor Y-H.An example of this principle is found in the case of phosphoglycerate kinase which catalyses the conversion of 3-phosphoglycerate into 1,3-diphosphoglycerate through a sequence of two reactions a Enz + ATP 2 Enz-8 + ADP EnzB + 3-phosphoglycerate 6 Enz + 1,3-diphosphoglycerate When in the above reaction the enzyme was incubated in the absence of the substrate but with ATP and the ADP formed by equilibrium a was removed by a coupled enzyme system the accumulation of the phosphorylated enzyme intermediate (Enz-P) o~curred.’~ In the intermediate the phosphoryl moiety was shown to be linked to a carboxy-group of the enzyme.Other examples of this type are the cases of acetate kina~e,~’ and succinyl coenzyme A ~ynthetase,’~ ATP-citrate lya~e.’~ SchifS Base Intermediates. When an intermediate enzyme-substrate complex contains a reactive covalent linkage which may be stabilized by a chemical modification this then permits the identification after a suitable degradation ’’ J. H. Schwartz Proc. Nut. Acad. Sci. U.S.A. 1963,49 871 ;L. Engstrom Arkiu Kemi. 1962 19 129. 56 C. T. Walsh and L. B. Spector J. Biol. Chem. 1971 246 1255. 57 R. S. Anthony and L. B. Spector J. Biol. Chem. 1970 245 6739. 58 R. F. Ramaley W. A. Bridger R. W. Moyer and P. D. Boyer J. Biol. Chem. 1967 242 4287. 59 C. T. Walsh and L. B. Spector J. Biol. Chem. 1969,244,4366. Reaction Mechanisms -Part (iii) Enzyme Mechanisms of the catalytic group.The approach has proved useful in identifying the active- site groups of enzymes where catalysis occurs through the formation of Schiff base intermediates. The earliest example of this is the enzyme fructose-1,6- diphosphate aldolase which catalyses the reversible reaction between dihydroxy- acetone-1-phosphate (52)and glyceraldehyde-3-phosphate The reaction (53).60761 may be considered to involve the cleavage of the a-C-H bond of the ketone (52) to give a carbanion intermediate which condenses with the aldehyde (53)to furnish 0 H-C / I (53) R /H-C I R 0 H-C I R 11 - Fructose + Enzyme S C=NH- I HCOH CH,-OP0,2 -R= I HO-LH HX-EI I (56) R H-C-OH Scheme 10 the product.Mechanistic studies have however highlighted the elegant process evolved by biological systems to improve the electron-withdrawing property of a carbonyl group through its conversion into a protonated Schiff base (54) which facilitates the cleavage of a C-H bond to yield the species (55). The overall reaction thus occurs through the sequence of Scheme 10. The intermediacy of the species (54) was demonstrated when the addition of NaBH to a solution of the enzyme and the ketone (52) resulted in inactivation of the enzyme activity and the stable linking of the substrate to enzyme. Suitable degradative experiments established that the N-atom of (54)belonged to an E-amino-group of lysine.60 Schiff base intermediates have also been established for transaldolase aceto- acetic decarboxylase 2-keto-6-deoxy-6-phosphogluconate aldolase 2-deoxy-~- ribose-5-phosphate aldolase and 2-keto-4-hydroxyglutarate aldolase.The literature on these enzymes has been reviewed up to 1968.61Recently the in- volvement of a Schiff base has also been established for N-acetylneuraminic 6o J. C. Speck P. T. Rowley and B. L. Horecker J. Amer. Chem. SOC.,1963 85 1012. 61 D. E. Morse and B. L. Horecker Ado. Enzymol. 1968 31 125. 184 M. Akhtar and D. C. Wilton acid aldola~e.~~?~~ The nature of the group X in intermediates of the type (54) has been investigated and preliminary experiments suggest that the group is a histidine residue in the case of fructose-l,6-diphosphatealdola~e,~~ a cysteine in N-acetylneuraminic acid aldolase,62 and a carboxylic acid group in 2-keto-6-deoxy-6-phosphogluconate ald~lase.~ Attention is drawn to another type of aidolases (class I1 ;Schiff-base aldolases are class 1) in which the electron-withdrawing property of the ketone group is enhanced not through a Schiff-base formation but through complexing with a metal ion.This sub.ject has been reviewed.61 A Schiff base intermediate is also involved in b-amino-levulinic acid dehydra- tase which catalyses the condensation of two molecules of b-amino-levulinic acid (57) to give porphobilinogen (59). This was shown66 by the incubation of the enzyme with the substrate (57) followed by treatment with sodium borohydride which led to the inactivation of the enzyme. The sequence of Scheme 11 for the conversion has been proposed to account for this and related observations.66 COOH COOH I I y2 YHz A v 7H2 Enz-NH2 + FH2 Enz -N=C P=* I H2N-CH2 y2 NH (57) (58) p57 COOH COOH I I HOOC CH HOOC CH, II II CH CH CH CH + Enz Enz -NH AH2 H,N-CH G O N H -H,N-CH 10 H (59) Scheme 11 We have discussed above the chemistry of the active site of those proteins whose sole function is to catalyse chemical transformations.Another type of protein molecule of biological interest is that in which a chemical conversion is intimately 62 J. E. G. Barnett D. L. Corina and G. Rasool Biochem. J. 1971 125 275. 63 R. Schauer 2.physiol. Chem. 1971,352 1517. 64 P. Hoffee C. Y. Lai E. L. Pugh and B. L. Horecker Proc.Nut. Acud. Sci. U.S.A. 1967 57 107. 65 H. P. Meloche Biochemistry 1970 9 5050. 66 D. L. Nandi and D. Shemin J. Biol. Chem. 1968,243 1231 1236. Reaction Mechanisms -Part (iii) Enzyme Mechanisms coupled to the transmission of a physiological impulse. These proteins are termed receptor molecules. The work on the visual receptor of bovine retina rhodopsin has highlighted the existence of two broad biochemical processes (Scheme 12).67 The first (equation a) involves the combination in a dark reaction of 11-cis-retinal with the protein opsin to give rhodopsin E,,, 498 nm (the absorption maximum is species-dependent and is in the range 440-560 nm). (60) hv H,O + 7-+ all-trans-retinal + NH,-opsin Scheme 12 The bathochromic shift accompanying this reaction permits the complex (60) to absorb the wavelength abundantly available in the sunlight when rhodopsin (60) participates in the transmission of the impulse responsible for visual sensation and is converted into opsin and all-trans-retinal (equation b).Two features of equation a (Scheme 12) of particular interest in the present context are the nature of the retinal-opsin linkage in (60) and the factor contributing to the batho- chromic shift of 6&180 nm accompanying the formation of rhodopsin. The first of these was investigated when it was shown that rhodopsin upon exposure to light in the presence of sodium borohydride gave a reduced derivative for- mulated as dihydrometarhodopsin 11. In this derivative the retinyl moiety was shown to be linked to an E-amino-group of ly~ine.~~.~~ This and related experiments7' showing the existence of a Schiff-base linkage [as in (60)] at the active site of rhodopsin could partly rationalize the bathochromic shift accompanying equation a since it is known that Schiff bases formed from retinal and primary amines upon protonation have A,, 430 nm.The elucidation of the factor(s) which may make further contribution to the red shift in visual proteins has stimulated a great deal of work on model systems. It has recently been shown that retinylideniminium ions show solvent-dependent bathochromic shift^^',^' which may be related to the refractive index of the solvent.71 Further- more protonated Schiff bases of retinal (&,, 430nm) undergo dramatic red shifts (up to 530 nm) in non-polar environments in a frozen ~tate.~'?~~ The model '-For a review see G.Wald Nature 1968 219 800. 68 (a) M. Akhtar P. T. Blosse and P. B. Dewhurst Chenz. Corrrm. 1967 631 ; (b)Biochem. J. 1968 110 693. 69 D. Bownds Nature 1967 216 1178. 7o M. D. Hirtenstein and M. Akhtar Biochem. J. 1970 119 359. " C. S. Irving G. W. Byers and P. A. Leermakers Biochemistry 1970 9 858. l2 M. D. Hirtenstein and M. Akhtar Nature New Bid. 1971 233 94. 73 W. Waddell and R. S. Becker J. Amer. Chem. SOC.,1971,93 3788. 186 M. Akhtar and D. C. WiIton experiments although not providing a precise physical explanation for the absorption spectra of visual pigments do however suggest that the presence of a protonated retinylidene chromophore in a specialized non-polar environment at the active site may make a significant contribution to the red shift.Thus in visual proteins the profound modification of the absorption properties of the chromophoric prosthetic group has allowed speculations on the nature of the environment at the active site. An analogous experimental approach of broad application involves the introduction of an ion or a small molecule called ‘probe’ into a special site of an enzyme or a protein. The measurement of a suitable spectroscopic property of the probe may then give useful information about its environment and about relatively minor changes around the probe- binding site. The literature on the use of emis~ion,~~’.~ e.~.r.,’~‘ ab~orption,~~‘,~ and n.m.r.74u,d probes has been reviewed and is not discussed here.5 Surface Structure of Proteins and Enzymes Work on rhodopsin has drawn attention to another facet of the chemistry of proteins. Most extensively studied from a crystallographic viewpoint have been the water-soluble proteins which catalyse the reaction of or participate in the transport of polar substrates. The examination of the tertiary structures of these proteins has allowed the general conclusion to be drawn that they contain non- polar (hydrophobic) residues in the core of the molecule and the polar side- chains (charged groups) are located on the surface. l8 This structural arrangement accounts for the water solubility of these proteins. Rhodopsin on the other hand is a membrane protein which resides in the rod outer segments where it is held by lipid molecules and participates in the reaction of a highly non-polar substrate.In view of these features it is noteworthy that rhodopsin is completely insoluble in water and dissolves only in detergent solutions. Similar features are also noted for other membrane-bound proteins which are involved in the reactions of non-polar substrates. Particular mention may be made of a bacterial mem- brane enzyme C ,-isoprenoid alcohol phosphokinase. The enzyme which catalyses the ATP-dependent phosphorylation of C ,-isoprenoid alcohols is soluble in butanol and has a high content (58%) of non-polar amino-acid~.’~ These examples suggest that these membrane-bound proteins acting on non-polar substrates may in fact contain more non-polar amino-acid residues on the surface.In closing this chapter we draw attention to how current knowledge of enzyme mechanisms has led to the chemical synthesis of a polymer with enzymic activity. A derivative of polyethyleneimine containing dodecyl groups to bind small molecules and methyleneimidazole side-chains to provide nucleophilic catalytic groups has been shown to hydrolyse p-nitrophenyl acetate at a rate appr~aching~~ that of the enzyme chymotrypsin. ’‘ (a) G. K. Radda Biochem. J. 1971 122 385; (6) R. B. Freedman Quart. Rev. 1971 25 431 (c) H. M. McConnell and B. G. McFarland Quart. Reu. Biophys. 1970 3 91 ; (6) A. S. Mildvan and M. Cohn Adc. Enzymol. 1970 33 1. ’’ H. Sandermann and J. L. Strominger Proc.Nat. Acad. Sci. U.S.A. 1971 68 2441. 76 I. M. Klotz G. P. Royer and I. S. Scarpa Proc. Nut. Acad. Sci. U.S.A. 1971 68 263.