18 Enzyme Mechanisms By M.AKHTAR and D. C. WILTON Department of Physiology and Biochemistry University of Southampton SO9 5NH THElast Report on Enzyme Mechanisms in 1968 dealt with reactions involving the hydrolysis of peptide and glycosidic linkages. The present review deals with the mechanism* of action of enzymes requiring the participation of coenzymes such as pyridine nucleotides and 5’-deoxyadenosyl-B 2 and of a newly discovered group of ‘modified amino-acid’ prosthetic groups. Since Annual Reports (B)are directed primarily towards readers interested in organic chemistry we have confined ourselves to the selection of material which is relevant to the ‘bio- organic’ aspect of the enzyme mechanisms. We apologize in advance to those authors whose work has been omitted either due to the limitation of space or over- sight on our part.1 Basis of Enzyme Catalysis Intramolecular reactions involving compounds in which the ground-state structure resembles the geometry of the transition state show spectacular reacti- vity when compared to their intermolecular counterparts. Thus a major contribu- tion to the efficiency of biological reactions may be made by bringing together the substrate molecules and the catalytic groups at the active site of the enzyme in the perfect arrangement for the given reaction to occur. This concept has been the basis of instruction in enzyme mechanisms at many institutions for several years. The universal acceptance of this view was hindered by the assertion of Koshland and co-workers who calculated that the contribution of proximity and proper orientation of the substrates to the rate of enhancement could not satis- factorily explain the efficiency of enzymes.’ The latter position has now been modified2 and the revised view accepting the importance of ‘proper orientation’ of substrates in enzymic reactions has received wide publicity3 on the basis of experiments carried out on model systems2 and given the title ‘Orbital Steering’.It has been shown2 that the acid-catalysed esterification of (2)is about lo6 times faster than the bimolecular reaction between methyl alcohol and acetic acid and about 1.3 x lo4 times faster than the lactonization of (1). Bruice and co-workers ’ D. E. Koshland and K. E. Neet Ann. Rev. Biochem. 1968 37 359.* D. R. Storm and D. E. Koshland Proc. Nat. Acad. Sci. U.S.A. 1970,66,445. Chem. in Brit. 1970,6 372 and 551 ; Chem. Eng. News July 1970 54. * The term mechanism in the present context refers to the bond-forming events which ensue after the formation of the enzyme-substrate complex (ES). M. Akhtar and D. C. Wilton had previously shown4 that the intramolecular reaction of the monophenyl ester of dicarboxylic acid anion (4) was about 5 x lo4times faster than the analogous reaction with (3). This study has recently been e~tended.~ In another report6 the lactonization of (6)has been found to be 2.5 x lo7times faster than that of (5). The obvious explanation for the dramatic rate enhancement in compounds (2) (4) and (6)is the juxtaposition of the reacting atoms in the ground-state structure so that the transition state for the reaction is reached with a minimal loss of trans- lational entropy.These studies highlight the contribution made to rate enhance- ment by the presence of reactive centres in an intramolecular environment and further emphasise the stringent geometrical requirement that must be fulfilled to achieve the maximal reactivity. It may be noted that these features are already well recognized in the literature and form the basis of several methods used in structure determination and selective syntheses. 2 Coenzyme-B, The biosynthesis of vitamin-B ,has been recently reviewed.' Coenzyme-B (7) (also called cobamide coenzyme or 5'-deoxyadenosyl-B Jis formed enzymically * from vitamin-B 2 a reducing agent and ATP (Scheme 1).A non-enzymic syn- thesis of the coenzyme has also been reported.' The participation of coenzyme- B, has been demonstrated for several enzymes and these are discussed below. Reviews" and a monograph' covering this and related aspects are available. Dio1dehydrase.-This enzyme catalyses the conversion of both the (R)-and (9-enantiomorphs of propanediol into propionaldehyde and also of ethylene glycol 'T. C. Bruice and U. K. Pandit J. Amer. Chem. SOC. 1960,82 5858. ' T. C. Bruice and A. Turner J. Amer. Chem. SOC. 1970,92 3422. ' S. Milstien and L. A. Cohen Proc. Nut. Acad. Sci. U.S.A. 1970,67 1143. H. C. Friedmann and L. M. Cagen Ann. Rev. Microbiol. 1970 24 159. A. Peterkofsky and H. Weissbach Ann.New York Acud. Sci. 1964 112 622; G. A. Walker S. Murphy and F. M. Huennekens Arch. Biochem. Biophys. 1969 134 95. A. W. Johnson L. Mervyn N. Shaw and E. L. Smith J. Chem. SOC. 1963,4146. lo (a)H. P. C. Hogenkamp Ann. Rev. Biochem. 1968,37,225;(b)H. A. Barker Biochem. J. 1967,105 I ;(c)T. C.Stadtman Ann. Rev. Microbiof. 1967,21 121 ;(d)F. Wagner, Ann. Rev. Biochem. 1966 35 405. E. L. Smith 'Vitamin-B I l'r Methuen's Monographs on Biochemical Subjects London 1965. Enzyme Mechanisms n B" g.j-f/oy 2e co -+ cb' -+ co HO OH HO OH Coenzyme-B Vitamin-B Vitamin-B 2s ATP "CH2-Rib-I KO) Scheme I X = Triphosphate (7) into acetaldehyde.12 The reaction proceeds with the transfer of one of the C-1 hydrogen atoms of propanediol to C-2 without exchange with protons of the medium.' Using (R)-propanediol it is the HRthat migrates to C-2 (8a)-+ (9a); however the hydrogen atom with the opposite stereochemistry H, is involved when (S)-propanediol is the substrate (8b) -+(9b).14 The displacement of the 2-position hydroxy-group in both the isomers occurs with inversion of configura-ti~n.'~.'~ It has been suggested that conformations of the two isomers at the enzyme surface as shown in structures (8a) and (8b) permit the observed steric course.' With respect to the status of oxygen atoms in the dioldehydrase reac- tion it has been shown that in the conversion of (R)-propanediol into propion- aldehyde the oxygen atom originally present at C-1 is lost and is replaced by the oxygen from C-2 (8a)-+(9a).On the other hand in the conversion of (S)-propanediol the C-1 oxygen atom is retained in the product (8b)-+(9b).16 These experiments have been rationalized by assuming the intermediacy of the gem-diol(22) followed by stereospecific dehydration.16 Me H H. A. Lee and R. H. Abeles J. Biol. Chem. 1963 238 2367. l3 A. M. Brownstein and R. H. Abeles J. Biol.Chem. 1961,236 1199. l4 B. Zagalak P. A. Frey G. L. Karabatsos and R. H. Abeles J. Biof. Chem. 1966,241 3028. l5 J. Retey A. Umani-Ronchi and D. Arigoni Experienriu 1966 22 72. l6 J. Retey A. Umani-Ronchi J. Seibel and D. Arigoni Experientiu 1966,22 502. 560 M. Akhtar and D.C.Wilton A contribution to the understanding of the mechanism of action of coenzyme- B, linked enzymic reactions was made by the demonstration that when a mixture of [l-3H,]propanediol and unlabelled ethylene glycol was incubated with dioldehydrase both the products propionaldehyde and acetaldehyde contained tritium in the a-position.” Analogous results were obtained in the conversion of a mixture of [l-2H,]propanediol and propanediol when mono- deuterio-propionaldehyde was the main product.’ These results highlighted for the first time that hydrogen transfer from C-1 to C-2 in these reactions was not strictly intramolecular. Furthermore the incubation of [l-3H,]propanediol and unlabelled coenzyme-B, with the enzyme resulted in the incorporation of tritium at C-5’ of the adenosyl moiety of the recovered coenzyme.” These results suggested that in the dioldehydrase reaction the hydrogen atom from C-1 is transferred to C-2 cia the coenzyme.This was confirmed by showing that chemically synthesised coenzyme-B containing tritium at C-5’ transferred tritium to the product when added to the enzyme and unlabelled s~bstrate.’~ Glycerol Dehydrase.-This enzyme catalyses the conversion2’ of glycerol to P-hydroxypropionaldehyde ;however mechanistic studies of the type described above for the dioldehydrase have not yet been reported. Ethanol Deaminase.-This enzyme has been isolated as a homogenous protein and catalyses the conversion of ethanolamine (10)into acetaldehyde.21.22 In this reaction one of the hydrogen atoms from the carbinol carbon atoms is trans- ferred to the amino carbon atom without exchange with water,23 the alcohol C-atom becomes the carbonyl C-atom of acetaldehydeZZb and the oxygen atom of the substrate is retained in the product.23 When [5’-3H,]coenzyme-Bl is used in the enzymic reaction the isotopic hydrogen is transferred to the a-position of a~etaldehyde,~ thus suggesting mechanistic similarity between the dioldehydrase and ethanolamine deaminase reaction.Ribonucleotide Reductase. 5-In the three enzymic reactions discussed above the reaction centres of the substrate undergoing oxidation and reduction were part of the same molecule. An interesting variation on this theme is provided by R. H. Abeles and B. Zagalak J. Biol. Chem. 1966 241 1245. Is J. Retey and D. Arigoni Experientia 1966 22 783. l9 P. A. Frey M. K. Essenberg and R.H. Abeles J. Biol. Chem. 1967 242 5369. 2o Z. Schneider E. G. Larsen G. Jacobson B. C. Johnson and J. Pawelkiewicz J. Bid. Chem. 1970 245 3388. ’‘ C. Bradbeer J. Biol. Chem. 1965 240 4675. 22 (a)B. H. Kaplan and E. R. Stadtman J. Biol. Chem. 1968,243,1787;(b)B. H. Kaplan and E. R. Stadtman J. Biol. Chem. 1968,243 1794. ” B. M. Babior J. Biol. Chem. 1969 244 449. 24 Unpublished work quoted in Ref. 23. ’’ The literature on this enzyme is reviewed in Ref. 10a and by R. L. Blakley and E. Vitols Ann. Rev. Biochem. 1968 37 201. Enzyme Mechanisms 561 the enzyme ribonucleotide reductase which catalyses the coenzyme-B depend-ent reduction of ribonucleotide triphosphates (1 1) into the corresponding de- oxyribonucleotide triphosphates (12).This reaction compulsorily requires the oxidation of one mole of a dithiol per mole of ribonucleotide reduced.26 In the conversion (1 1) -+ (12) the new hydrogen atom at C-2’ of ribose is derived from the medium2’ and the overall reaction occurs with the retention of con- figuration.28 Reduction of [Z-’ 80]adenosine triphosphate to deoxyATP results in complete loss of the isotope from the nucleotide whereas [3’-180]ATP is reduced to [3’-180]deoxyATP with the same isotopic content.28c When the ribonucleotide reductase reaction is carried out in the presence of tritiated water the recovered coenzyme-B contains tritium at the S-position.*’ These results have been rationalized as shown in Scheme 2 involving the reduction of the coenzyme by dithiol followed by the displacement of the C-2’ hydroxy-group of the substrate by a hydride from the reduced coenzyme29 (see also Scheme 3).The exchange of the 5’-hydrogens of the coenzyme with the protons of the medium has also been observed in a partial reaction without the substrate ribonucleotide triphosphate but in the presence of allosteric activator^.^^' * S-H + Coenzyme F El + Coenzyme.*H + *Hf Is-; + Coenzyme-H* -+ x - ~ o ~ sCoenzyme ~ HOx-cQBase OH HO H* (11) (12) Scheme2 X = Triphosphate Methylmalonyl CoA Mutaw.-This enzyme catalyses the interconversion of (R)-methylmalonyl CoA (L-methylmalonyl CoA) and succinyl CoA (13a) (14a). The reaction involves the counter transfer of a hydrogen atom and the carbonyl thioester group and occurs without exchange with the protons of the medium.30 The retention of configuration during the conversion (13a) -+( 14a) was established by using the deuterio-analogue (13).The resulting succinyl CoA 26 E. Vitols and R. L. Blakley Biochem. Biophys. Res. Comm. 1965 21 466. ’7 M. M. Gottesman and W. S. Beck Biochem. Biophys. Res. Comm. 1966,24,353; R. L. Blakley R. K. Ghambeer T. J. Batterham and C. Brownson Biochem. Biophys. Res. Comm. 1966,24,418. ” (a) A. Larsson Biochemistry 1965,4 1984; (b)T. J. Batterham R. K. Ghambeer R. L. Blakley and C. Brownson Biochemistry 1967 6 1203; (c) H. Follmann and H. P. C. Hogenkamp Biochemistry 1969,8,4372. ” (a) W. S. Beck R. H. Abeles and W. G. Robinson Biochem. Biophys. Res. Comm. 1966 25 421; (b) R. H. Abeles and W.S. Beck J. Biol. Chem. 1967 242 3589; (c) H. P. C. Hogenkamp R. K. Ghambeer C. Brownson R. L. Blakley and E. Vitols J. Biol. Chem. 1968 243 799. 30 Literature on this aspect is covered in Ref. 10a and by J. W. Cornforth and G. Ryback Ann. Reports 1965 62 428. M. Akhtar and D.C. Wilton (14) on hydrolysis yielded (S)-monodeuteriosuccinic acid (15)? The use of [5’-3H,Jcoenzyme-Bl in the mutase reaction resulted in the incorporation of tritium into (13a) and (14a).’*~~~ Me D 27 HO2C CO-SCOA HO2C H H02C H (13)(13a) D =H CH( NH 2)C02H / CH H ->; + H02C CH(NH2)C02H HO,C D HO2C D (16) (17) (18) (16a) D =H (17a) D =H Glutamate Mutaw.-This enzyme catalyses the reversible conversion of L-glutamate (17a) to threo-/?-methyl-L-aspartate(16a).The enzymic reaction in- volves the counter transfer of a glycine moiety and a hydrogen atom.’Ob When the deuteriated analogue (16) was used in the enzymic reaction and the resulting glutamic acid (17) degraded (R)-monodeuteriosuccinic acid (18) was ~btained.~ This suggested that the conversion (16) -+(17) occurs with the inversion of c~nfiguration.~~ The involvement of 5’-hydrogen atoms of coenzyme-B in the glutamate mutase reaction has been claimed. lob L-fl-Lysine Mutase-This enzyme catalyses the reversible conversion of L-p-lysine (19) to 3,5-diaminohexanoate (20).34 Two other related enzymes3 participate in the interconversions of D-a-lysine (19a) 2,5-diaminohexanoate (20a) and L-ornithine (19b)*2,4-diaminopentanoic acid (20b).These transfor- mations involve the counter transfer of a hydrogen atom and an amino-group which in the case of the L-P-lysine mutase reaction occurs without exchange with the amino-group of free ammonia34a or the hydrogen of water.36 When the L-p-lysine mutase reaction is carried out in the presence of [5’-3H,]coenzyme-B,2 the tritium is transferred to C-6 of 3,Sdiaminohexanoate (20) and to C-5 of L-/3-lysine (19)?’ Another analogous enzymic interconversion of L-lysine and L-/?-lysine surprisingly does not involve3’ the participation of coenzyme-B12. 3’ M. Sprecher M. J. Clark and D. B. Sprinson J. Bid. Chem. 1966 241 872. 32 G. J. Cardinale and R. H. Abeles Biochim. Biophys. Actu 1967 132. 517. ” M. Sprecher R. L. Switzer and D. B. Sprinson J.Biol. Chem. 1966,241 864. 34 (a)R. C. Bray and T. C. Stadtman J. Biol. Chem. 1968,243,381; (b)E. E. Dekker and H. A. Barker J. Biol. Chem. 1968,243 3232. 35 (a)T. C. Stadtman and L. Tsai Biochem. Biophys. Res. Comm. 1967,28,920; (6)J. K. Dyer and R. N. Costilow J. Bacteriul. 1970,101,77 (c)Y.Tsuda and H. C. Friedmann J. Biol. Chem. 1970 245 5914. 36 J. Retey F. Kunz T. C. Stadtman and D. Arigoni Experientia 1969 25 801. 37 T. C. Chirpich V. Zappia R. N. Costilow and H. R. Barker J. Biol. Chem. 1970 245 1778. Enzyme Mechanisms R-CH,-CH,-NH R-CH(NH,)-Me (19 19a 19b) (20 20a 20b) (19) and (20) R = H0,C-CH2-CH(NH,)- (19a) and (20a) R = HOzC-CH(NH,)-(CH2)2-(19b) and (20b) R = HO,C-CH(NHJ-CH,-Mechanism.-The cumulative evidence presented suggests that in the above transformations coenzyme-B, acts as a hydrogen carrier'3*'4*' 'and that during the course of the enzymic reactions one of the hydrogen atoms of the substrate molecules and both the C-5' hydrogen atoms of the coenzyme become enzymic- ally indistingui~hable.~~~*~ 8,39 This could be achieved through the transfer of a hydride ion or its equivalent from the substrate to the coenzyme thus resulting in the formation of a C-5' methyl group and a reduced cobalt species (21).29'*38 This mode of cleavage of the carbon-cobalt bond was originally deduced38 by considering the formal reversal of the process involved in the biosynthesis of the coenzyme (Scheme I).The re-formation of the coenzyme (7) from the reduced species (21) may then occur through a number of closely related processes.The suggested me~hanisrn~~.~' for the dioldehydrase reaction is outlined in an H 1 5' / -C -H Scheme 3 H CO-SCOA CO.SCoA H (7) + (13a) (21) + +S.H H*q (7) + (14a) [H C02H H C02 H I (231 (24) abbreviated form in Scheme 3 and may be adopted for glycerol dehydrase ethanolamine deaminase and ribonucleotide reductase reactions. In the case of the last enzyme the reduced coenzyme (21) is formed by the reaction:2gc (7) + dithiol e(21) + disulphide. A similar mechanism for the methylmalonyl -'8 M. Akhtar Comp. Biochem. Physiof. 1969 28 1. 39 A mechanism similar to that in Scheme 3 appears to have been considered by R. H. Abeles and R. Williams Abstracts of 8th International Congress of Biochemistry Switzerland 1970 p.135. 564 M. Akhtar and D.C. Wilton CoA mutase reaction has been proposed38 and briefly it involves the formation of a carbonium ion intermediate (23) which after rearrangement to (24) followed by reduction with (21) yields the product. That the methyl intermediate as in structure (21) may participate in the coenzyme-B ,dependent reactions was originally suggested by the important observation that in the dioldehydrase reaction the use of a pseudo-substrate capable of supporting only the initial reaction (corresponding to the first reaction of Scheme 3) resulted in the formation of 5'-deo~yadenosine.~' Similar results were later obtained with ethanolamine deamina~e.~~ Another type of mechanism has also been considered involving a homolytic process for the cleavage of the carbon-cobalt bond42 of (7) followed -Rib-&2 -Rib-kH2 / / H ./ (C'iI) ;-+H-C-+ -Rib-C-H + C-\ \ \ (25) (26) by a hydrogen-atom transfer to give (25) and (26).42" This principle,42a when generally extended to coenzyme-B ,dependent enzymic reactions will require in several cases radical species to undergo unorthodox rearrangements displace- ments or eliminations to furnish products.On the other hand a heterolytic cleavage of the carbon-cobalt bond can rationalize not only the mechanisms of the coenzyme-B, dependent enzymes but also that of methyl-B, linked meth- ionine bio~ynthesis.~~.~~ However a clear choice between the two types of mechanism cannot be made at present.It is conceivable that the final solution may incorporate features from both types of mechanism. Vitamin-B also participates in the biosynthesis of methionine methane and acetic acid. In these reactions methyl-B ,[-(Co)-Me] is the crucial intermediate This aspect of the situation has been reviewed,"" up to 1968. 3 Modified Amino-acids as Prosthetic Groups It has recently come to light that the traditional r61e of pyridoxal phosphate in some enzymic reactions is performed by modified amino-acid residues. Histidine decarboxylases catalyse the decarboxylation of hystidine to yield histamine (29) and CO . One such enzyme from Lactobacillus 30a contains covalently bound pyruvate as the prosthetic group (27).44 It has been suggested that the carbonyl group of the pyruvyl residue forms a Schiff base (28) with the substrate which 'O 0.W.Wagner H. A. Lee P. A. Frey and R. H. Abeles J. Biol. Chem. 1966,241 1751. '' B. M. Babior J. Biol. Chem. 1970 245 1755. 42 (a)B. M. Babior J. Biol. Chem. 1970 245 6125; (b) M. A. Foster H. A. 0.Hill and R.J. P. Williams in 'Biochemical Society Symposia No. 31 ;Chemical Reactivity and Biological Role of Functional Groups in Enzymes' ed. R. M. S. Smellie Academic Press London 1970. 43 R. T. Taylor and H. Weissbach Arch. Biochem. Biophys. 1969 129 728 745. 44 W. D. Riley and E. E. Snell Biochemistry 1968 7 3520; P. A. Recsei and E. E. Snell Biochemistry 1970 9,1492. Enzyme Mechanisms facilitates decarboxylation. Biosynthetic experiments suggest that the pyruvyl residue in histidine decarboxylase originates by the modification of a serine m~iety.~’ A pyruvyl residue has also been shown to be involved in the analogous conversion catalysed by S-adenosyl-methionine decarb~xylase.~~ Amino-acid H Me + I I CH,-C-C-NH-protein -+ R-C-N-C--+ $ 03rotein R-CHz-NHz + (27) 00 (29) I1 II €-I-i \o A R=N+ ,NH (27) (28) C D-Proline reductase which catalyses the conversion of proline into 5-aminovaleric acid in the presence of a dithiol also contains a pyruvyl prosthetic gro~p.~’ A hypothetical mechanism for the conversion has been ~uggested.~’ a-Ketobutyrate has been identified as the prosthetic group of urocanase (30)which catalyses the hydration of urocanate to give (32).48 A mechanism of reaction involving the intermediacy of (31) which permits an attack of OH-at position 4 has been ~uggested.~’ Urocanic acid + Et -C -NH I1 I100 -Protein -+ N 7CH=CH-COzH -C-C-NH- f’rotein (30) r( J 0 Histidine and phenylalanine ammonia lyases catalyse the reaction (33)+(34).In the case of the histidine ammonia lyase reaction the elimination involves the removal of H from the p-carbon atom.49 The inacti~ation~~-’~ of these enzymes with NaB3H4 followed by acid hydrolysis of the protein led to the isolation of radioactive alanine containing most but not all of the tritium in the p-po~ition.~~-”These results have been interpreted in terms of the presence of a dehydroalanyl residue in the enzymes and the partial structure (35) for the active site has been propo~ed,4~3~~ though no evidence for the existence of >C=N-45 W.D. Riley and E. E. Snell Biochemistry 1970,9 1485. 46 R. W. Wickner C. W. Tabor and H. Tabor J. Biol. Chem. 1970,245,2132. 4’ D. Hodgins and R. H. Abeles J. Biol. Chem. 1967,242,5158; Arch. Biochem. Biophys. 1969 130,274. O8 D. J. George and A. T. Phillips J. Biol. Chem. 1970 245 528. 4y I. L. Givot T. A. Smith and R. H. Abeles J. Biol. Chem. 1969 244 6341 ;J. Retey H. Feirz and W. P. Zeylemaker F.E.B.S.Letters 1970,6 203. 50 R. B. Wickner J. Biol. Chem. 1969 244,6550. 51 K. R. Hanson and E. A. Havir Arch. Biochem. Biophys. 1970,141 1. M. Akhtar and D.C. Wilton linkage [encircled in (35)Jis at present available. A mechanism of action for the enzymic reactions has been proposed.This involves the addition of the amino- group of the substrate (33) to the p-carbon atom of the dehydroalanyl residue followed by an elimination reaction and finally the regeneration of the enzyme in several steps.51 An alternative formulation of the active site compatible with the available evidence is shown in structure (36)which allows the enzymic reaction to occur via (37) in two steps. CO,HI R-CH2-CH(NH,)+ R-CH=CH-CO,H + PjH (33) (34) + (33) -+ (35) Protein Protein (36) R I 1 -(34) + (36) (37) 4 Pyridine Nucleotides A large group of enzymes catalyse the general reaction involving the transfer of a hydrogen between the coenzyme and substrate. The enzymes taking part in the \ I C=X + NAD(P)H* + H+ H*-C-X-H + NAD(P)’ / I usually reversible reactions where X is oxygen and nitrogen are called dehydro- genases and have been the subject of recent excellent reviewsand discussion^.^^-^ However the pyridine nucleotide linked saturation of carbon double bonds has yet to be shown to be reversible and the enzymes are termed reductases.Conformation of Pyridine Nucleotides-The three-dimensional structure of pyri -dine nucleotides in solution has recently been subjected to critical evaluation using 220 MHz n.m.r.56 Although the pyridine ring of reduced pyridine nucleotides is planar,57 n.m.r. data reveal that the two hydrogens at C-4 are non-equivalent. 52 A. Frankfater and I. Fridovich Biochem. Biophys. Acta 1970,206,457. 53 S. P. Colowick J.van Eys and J. H. Park in ‘Comprehensive Biochemistry’ ed. M. Florkin and E. H. Stotz Elsevier Amsterdam 1966. 54 H. Sund in ’Biological Oxidations’ ed. T. P. Singer Interscience New York 1968. ‘Pyridine Nucleotide Dependent Dehydrogenases’ ed. H. Sund Springer-Verlag Berlin 1970. 56 R. S.Sarma and N. 0. Kaplan Biochemistry 1970,9 539 549 557. 57 I. L. Karle Acta Crvst. 1961 14 497. Enzyme Mechanisms 567 This non-equivalence is due to interaction of the adenine and nicotine rings which results from two possible folded conformations P and M of the pyridine nucleotide with the intermediacy of an open-chain form. If folded conformations u M-HELIX P-HELIX are relevant to the enzyme mechanism then the coenzyme should be bound to the A and B specific dehydrogenases in the M and P helices respectively.Fluorescence studies suggest that alcohol lactic glutamic and glycerol-3-phosphate dehydro- genases bind the coenzyme in the open config~ration.~~ Only glyceraldehyde-3- phosphate dehydrogenase appears to bind in the closed config~ration.~~ The recent 220 MHz n.m.r. data on the binding of NADH to lactic dehydrogenase suggest a closed configuration for the coenzyme ;59 however these observations are inconsistent with the 2.8 X-ray crystallographic structure.60 Structure of Dehydrogenases.-In the past few years rapid advances have been made in the elucidation of the structure of dehydrogenases. These advances have included the amino-acid sequence of glyceraldehyde-3-phosphatedehydrogen-ase,6 liver alcohol dehydrogenase,62 and glutamic dehydr~genase.~~ The 2.8 8 three-dimensional X-ray structure of lactic dehydrogenase has now been deter- mined.60 At the time of writing neither the amino-acid sequence nor the high- resolution X-ray structure of the same dehydrogenase were yet available.Active Site Groups.-Sulphur amino-acids. Elucidation of mechanisms for a group of enzymes is facilitated if unifying features within the group are discovered (cj serine proteases). Similarities do exist between certain dehydrogenases in particular with respect to the ‘essential thiol peptide’. Treatment of nine species of ” S. F. Velick ‘Light and Life’ ed. W. D. McElroy and B. Glass Johns Hopkins Balti- more 1961. ‘9 R. S. Sarma and N. 0.Kaplan Proc.Nut. Acad. Sci. U.S.A. 1970,67 1375. 6o M. J. Adams G. C. Ford R. Koekoek P. J. Lentz A. McPherson M. G. Rossmann I. E. Smiley R. W. Schevitz and A. J. Wonacott Nature 1970 227 1098. 61 B. E. Davidson M. Sajgo H. F. Noller and J. I. Harris Nature 1967 216 1181 ;J. I. Harris and R. N. Perham ibid. 1968 219 1025. 62 H. Jornvall European J. Biochem. 1970 16 25. 63 E. L. Smith M. Landon D. Piszkiewicz W. J. Brattin T. L. Langley and M. D. Melamed Proc. Nut. Arad. Sci. U.S.A. 1970 67 724. M. Akhtar and D. C. Wilton glyceraldehyde-3-phosphatedehydr~genase~~ and also yeast and liver alcohol dehydrogena~e~’ with iodoacetate resulted in the and lactic dehydrogena~e~~.~~ inactivation of the enzymic activity and selective hydrolysis of the alkylated proteins furnished peptide fragments which contained carboxymethylated cyste- ine.Possible similarities of amino-acids between the ‘essential thiol peptides’ of the enzymes have been but as yet it is not possible to draw any general conclusion regarding the common genetic origin of or mechanistic function for amino-acids in the various fragments. In the case of lactic dehydrogenase the cysteine may be required for the binding of the substrate but not of the coenzyme.68 Iodoacetate inactivated cytoplasmic malate dehydrogenase through the alkyla- tion of a methionine4’ (cf. mitochondria1 enzyme below). Lysine. Inactivation of glutamic dehydrogenase with 4-iodoacetamido-salicylic acid involves the alkylation of a ~ysteine~’.~’ Incubation of as well as a ly~ine.~’ the enzyme with the inhibitor in the presence of a-oxoglutarate protects the enzyme from inactivation and results in the alkylation of cysteine but not of lysine.This result indicates the important r61e of a lysine in the binding of substrate. The lysine also reacts with a number of other reagents and has been identified as lysine 93 and the amino-acid sequence around it determined.63 A similar sequence surrounds lysine 212 of glyceraldehyde-3-phosphatedehydro-genase.61 A mechanistic r61e for lysine has also been suggested for lactic dehydr~genase~’ Attention is drawn to recent and alcohol dehydr~genase.~~ work in which it has been shown that the modification of the amino-group(s) (presumably c-amino-groups of lysine) of alcohol dehydrogenase with (38) resulted in a dramatic increase in the enzymic 0 II 0C-CH2Br 64 R.N. Perham and J. I. Harris J. Mol. Biol. 1963,7 3 16; W. S. Allison and J. I. Harris Abstracts 2nd F.E.B.S. Meeting 1965 p. 140. 65 J. 1. Harris Nature 1964 203 30. 66 T. P. Fondy J. Everse G. A. Driscoll F. Castillo F. E. Stolzenbach and N. 0.Kaplan J. Biol. Chem. 1965 240 421 9. 67 J. J. Holbrook G. Pfleiderer K. Mella M. Volz W. Leskovac and R. Jeckel European J. Biochem. 1967 1 476. 68 J. J. Holbrook and R. A. Stinson Biochem. J. 1970,120 289. 69 V. Leskovac and G. Pfleiderer 2.physiol. Chem. 1969,350,484. 70 A. D. B. Malcolm and G. K. Radda European J. Biochem. 1970 15 555. ’’ J. J. Holbrook P. A. Roberts B. Robson and R. A. Stinson in ‘Abstracts of 8th International Congress of Biochemistry’ 1970 p.83. 72 A. D. Winer and G. W. Schwert J. Bid. Chem. 1958,231 1065; 1959,234 11 55. 73 E. M. Kosower ‘Molecular Biochemistry’ McGraw-Hill New York 1962. 74 B. V. Plapp J. Biol. Chem. 1970 245 1727. Enzyme Mechunisms 569 Histidine. The inactivation of lactic dehydrogenase with (39) involves the alkyla- tion of a cysteine in the essential thiol peptide and also a hi~tidine.'~ A histidine has also been directly demonstrated to be at the active site of mitochondria1 malate dehydr~genase.~~ Less direct evidence for the involvement of a histidine has been obtained in the case of rabbit muscle cytoplasmic a-glycerol phosphate dehydrogenase.77 Histidines have long been implicated in mechanisms for dehydrogenases.Winer and Schwert proposed histidine as the proton source during reduction in the case of a lactic dehydr~genase~~ while Ringold has presented a general mech- anism for dehydrogenases involving a histidine residue. This mechanism allows for the activation of both substrate and coenzyme by a hi~tidine.~~ Tryptophan. The role of tryptophan at the active site of dehydrogenases is at present under debate. Schellenberg has shown that incubation of both yeast alcohol dehydr~genase~~ and lactic dehydrogenase" with tritiated substrate I (3H-C-OH) in the presence of NAD at high pH resulted in the incorporation I of tritium into the protein. The label was shown to be located in the methylene group ofa tryptophan to which has been assigned a mechanistic r61e.Attempts to confirm the work using liver alcohol dehydrogenase,62 glyceraldehyde-3-phos- phate dehydrogenase,8 malate dehydrogenase,82 and lactic dehydrogenase7 1*7 have been unsuccessful. Using yeast alcohol dehydrogenase Palm83 has shown a time-dependent incorporation of radioactivity in the protein not only from [l-3H]ethanol but also from [4B-3H]NADH (this enzyme utilises the A side). Of possible pertinence in this connection is the observation that non-enzymic tritium exchange may be observed between NAD and NADH as a result of complex formation.84 Similar interactions between NAD+ and tryptophan were facili- tated at high pH.85 Although no direct tritium exchange has been observed in the latter case it would seem reasonable that the specialised environment of the enzyme surface coupled with the non-ionised nature of tryptophan at high pH might well produce such an exchange.Thus Schellenberg's observation could reflect the interaction of the coenzyme and tryptophan at the active site. Trypto- phan has already been implicated as being in close proximity to the coenzyme at the active site of dehydrogenases as the result of fluorescence and inactivation studies. *v8 75 V. C. Woenckhaus J. Berghauser and G. Pfleiderer 2.physiol. Chem. 1969,350,473. 76 B. H. Anderton European J. Biochem. 1970 15 562. 77 R. Apitz-Castro and Z. Suarez Biochirn. Biophys. Acra 1970 198 176. " H. J. Ringold Nature 1966 210 535. 79 K. A. Schellenberg J. Biof. Chem. 1965 240 1165; 1966 241,2446.'O K. A. Schellenberg J. Biol. Chem. 1967 242 1815. " W. S. Allison M. J. Connors and D. J. Parker Biochem. Biophys. Res. Comm. 1969 34 503. 82 L. M. Allen and R. G. Wolfe Biochem. Biophys. Res. Comm. 1970,41 1518. '' D. Palm Biochem. Biophys. Res. Comm. 1966,22 151. 84 J. Ludowieg and A. Levy Biochemistry 1964,3 373. 85 G. G. A. Alivisatos F. Ungar A. Jibril and G. A. Mourkides Biochirn. Biophys. Acta 1961,51 361. '6 P. L. Luisi and R. Favilla European J. Biochem. 1970 17 91. M. Akhtar and D.C. Wilton Metal Ions.-Zinc is associated with some dehydrogenases. In the case of the liver alcohol dehydrogenase this metal can be replaced by Co2+ or CdZ+ without any appreciable loss of enzymic a~tivity.~’ Mechanisms involving Zn2 + have been propo~ed~~,’~,~~ for dehydrogenases but lack experimental verification.Geometry of the Active Site.4ther highlights in the study of the active site of dehydrogenases include the use of a spin-labelled analogue (40) of NAD.89*90 This analogue in which the N-O bond has been suggested to correspond to the C-N bond between the nicotinamide and the ribose of NAD binds to liver alcohol dehydrogenase and is competitive with NADH.89 The n.m.r. proton relaxation times of either ethanol acetaldehyde or isobutyramide when bound to the enzyme in the presence of (40) have been used to establish the spatial geometry of the substrate and coenzyme at the active site. With each compound the oxygen atom occupies the same point in space relative to the c~enzyme.’~ In addition e.s.r.studies provide evidence that the substrate binds to form a ternary complex on the solvent side of the coenzyme.89 The nature of the binding site of the 3a-steroid dehydrogenase from Pseudo-rnonus testosteroni has been subjected to a rigorous investigation.” This work has shown how the enzyme becomes more specific and efficient as the substrate used changes from cyclohexanone up through substituted two- and three-ringed ketones to the normal steroid nucleus. Finally Eisele and Wallenfels have observed stereoselective inactivation of certain dehydrogenases using the two antipodes of a-iodopropionic acid and OL-iod~propionamide.~’ A-specific dehydrogenases are more rapidly inhibited by the D-an tipode whereas the L-antipode is more effective against B-specific enzymes.This correlation must reflect the nature of the coenzyme binding site on the enzyme as in the tentative Scheme shown ~pposite.’~ The Order of Bond Formation During the Reduction of C=C.-The mechanism of pyridine-nucleotide-linked reduction of C=C has been extensively studied. These reductions may be divided into two types depending upon whether or not the olefin linkage is conjugated to an electron-withdrawing group. In the enzymic ” D. E. Drum and B. L. Vallee Biochem. Biophys. Res. Comm. 1970,41,33. N. Evans and B. R. Rabin European J. Biochem. 1968,4 548. 89 A. S. Mildvan and H. Weiner Biochemistry 1969 8 552. 90 A. S. Mildvan and H. Weiner J. Biol. Chem. 1969 244 2465. 91 H. J. Ringold J.M. H. Graves A. Clark and T. Bellas Recent Prugr. Hormone Res. 1967 23 349. 92 K. Wallenfels and B. Eisele European J. Biochem. 1968 3 267; B. Eisele and K. Wallenfels in ‘Pyridine Nucleotide Dependent Dehydrogenases’ ed. H. Sund Springer- Verlag Berlin 1970. Enzyme Mechanisms A enzyme + B enzyme + A enzyme + B enzyme + coenzyme coenzyme D(+)a-iodo-L(-)qr-i?do-propionic acid proplonlc acid or amide or amide (R = 0- NH2) (R = 0; NHZ) reduction of conjugated ketones a hydrogen from the medium always adds at the position CY to the carbonyl while the hydrogen from the pyridine nucleotide is transferred to the p po~ition.~~,'~' From cofactor 1 1 H \/ Ill c=c -+-C-C-C-/a fl\ I1 I I -C OH II T 0 From medium Examples of double bonds not conjugated to electron-withdrawing groups are found during the biosynthesis of cholesterol from lanosterol and occur at the A7.8 A14.15 7 3 and ~24.25positions.The complete stereochemistry and orienta- tion of addition of hydrogens to these double bonds has now been determined and in each case the orientation of addition is consistent with a Markownikoff mechanism. Protonation of the double bond occurs first to give the most stable carbonium ion which is subsequently neutralised by the addition of a hydride ion from the coenzyme.94 Thus activation of the substrate in the reduction of C=C is achieved by an enzyme-mediated protonation step and the coenzyme simply completes the process initiated by the enzyme. Because of the formal similarity between all pyridine nucleotide linked reduc- tions it is possible to draw a general mechanism (Scheme 4) in which activation Enz.(H+) Enz.I,,+ NAD(P>\ ,TD(P),+ \ H-C-X-H /C-x-H I Scheme 4 93 J. S. McGuire and G. M. Tomkins Fed. Pror.. 1960 19 A-29; 0. Berseus and I. Bjorkhem Europeun J. Biochenr. 1967 2 503; D. C. Wilton and H. J. Ringold Proc. 3rd Int. Congr. Endocrinology 1968 no. 263; I. Bjorkhem European J. Biuchern 1969 8 345;J. D. Robinson R. 0.Brady and C. R. Maxwell J. Lipid Res. 1962 3 243. 94 D. C. Wilton K. A. Munday S. J. M. Skinner and M. Akhtar. Biochem. J. 1968 106. 803; M. Akhtar K. A. Munday A. D. Rahimtula I. A. Watkinson and D. C. Wilton Chem. Comni. 1969 1287; 1. A. Watkinson D. C. Wilton K. A.Munday and M. Akhtar Eiochern. J. 1971 121 131. 572 M. Akhtar and D. C. Wilton of the substrate is the crucial first step. Support for this mechanism is provided by evidence resulting from the study of model chemical systems. This work has been discussed by a number of a~thors~~,~~ and therefore it is our intention only to summarise the general conclusions that may be drawn. Dihydropyrjdine compounds have been used with only limited success for the reduction of non-activated double-bond species. However compounds contain- ing activated double bonds such as Malachite green (41)9”and riboflavin (42),”’ CH,OH I Ph are rapidly reduced at room temperature. Furthermore it is interesting to note that riboflavin is much more rapidly reduced under acid conditions due to initial protonation of the riboflavin facilitating hydride transfer.97 These obser- vations re-emphasise the requirement for protonation of the substrate prior to hydride transfer in pyridine nucleotide reductions as proposed above for C=C reductions.Pyridine Nucleotides in Carbohydrate Transformations.-UDP-D-glucose 4‘-epimerase catalyses the NAD linked reaction UDP-D-glucose (43) UDP-D-+ galactose (49 involving the inversion of the hydroxy-group at C-4. The reaction proceeds without incorporation of hydrogen or oxygen from the solvent98 while a hydrogen isotope effect is observed when UDP-[4-3H]-~-glucose is used.99 The mechanism shown in (43) -+(44) +(45) involving the intermediacy of enzyme- bound NADH and UDP-4-oxo-~-glucose has been proposed for this reaction.98 Recent observations using an equal mixture of perdeuteriated and non-deuteriated (43)have established that the hydrogen transfer at C-4 is intramolecular and uses CHzOH CHzOH CH20H Em.-NAD’ +HO (or) ]eNA:4y3.u1 0-UDP-[.*y+r) 0-UDP OH OH OH (43) (44) (45) 95 T.C. Bruice and J. Benkovic ‘Bio-organic Mechanisms’ vol. 11 Benjamin New York 1966; K. A. Schellenberg in ‘Pyridine Nucleotide Dependent Dehydrogenases’ ed. H. Sund Springer-Verlag. Berlin I970 ;K. Wallenfels ibid. 96 D. H. Mauzerall and F. H. Westheimer J. Amer. Chem. SOC., 1955,77,2261. 97 C. H. Suelter and D. E. Metzler Biochim. Biophys. Acfa. 1960,44,23. 98 I. A. Rose Ann. Rev. Biochem. 1966,35 23. 99 S. Kirkwood and G. L.Nelsestuen Biochim. Biophys. Acta 1970 220 633. Enzyme Mechanisms (Y!*UDP +[ H(-T-UDP CH20H HO (46) NHAc NADH 0 NHAc + Enz.-NAD J + CHzOH CH OH (LqH +UDj -3 OH HN HO NADH .o NAD+ (47) only one face of NADH.'" A related mechanism is involved in the conversion (46)-+(47).lo I Glucose-6-phosphate-myo-inositol-1-phosphate cyclase catalyses the NAD' linked cyclisation (48) --* (50). This reaction occurs without re- arrangement of the carbon skeleton and with retention of configuration at each carbon atom."' By double-labelling techniques C-1 C-3 and C-5 hydrogen atoms were shown to be retained during the cycli~ation,'~~ whereas one of the two hydrogen atoms at C-6 is lost.'03*'04 The r61e of a 5-0x0 intermediate (49) has been suggested on the basis of an isotope effect observed with the C-5 tritiated substrate.lo3 UDP-D-glucuronic acid decarboxylase catalyses the conversion of CH20@ .O (q--+ IDH$fF -NADH p v H HO OH HO HO OH OH OH ~ (48) (49) + 1 Enz.-NAD+ OH (50) UDP-D-glucuronic acid (51) into UDP-D-xylose (54).In the reaction (51)+ (54) the hydrogen atoms at C-3 C-4 and C-5 of the substrate are retained. The conversion of the 4-tritiated analogue (51) is attended by a significant isotope loo L. Glaser and L. Ward Biochim. Biophys. Actu. 1970 198 613. 'O' W. L. Salo and H. G. Fletcher Biochemistry 1970,9 882. lo' F. Eisenberg A. H. Bolden and F. A. Loewus Biochern. Biophys. Res. Comm. 1964 14,419; I. W. Chen and F. C. Charalampous J.Biol. Chem. 1965,240,3507; H. Kindi J. Biedl-Neubacher and 0.Hoffman-Ostenhoff Z. physiol. Chem. 1965,341 157. J. E. G. Barnett and D. L. Corina Biochem. J. 1968,108 125. Io4 I. W. Chen and F. C. Charalampous Biochim. Biophys. Acfu 1967 136 568. M. Akhtar and D.C. Wilton C02H CO2H OH (51) + Enz.-NAD + NADHo$-> 1- OH "*D+$!q) O*UDP HO 0* UDP OH OH (53) (54) effect which has been attributed to reversible oxidation at C-4 to facilitate de- carbo~ylation.'~~ Compound (52) has also been implicated in the biosynthesis of apiose. O6 Pyridine nucleotides are involved in the conversion of TDP-D-glucose (55) into TDP-4-oxo-6-deoxyglucose(56).'O7 This reaction involves the transfer of a hydrogen from C-4 to C-6 while a hydrogen from the medium is incorporated at C-5.Incubation of [4'-3H]-TDP-6-deoxyglucose with the enzyme resulted in transfer of tritium to bound NAD+ giving [4B-3H]NADH. It is this 4B tritium that is transferred intramolecularly to the C-6 position of the sugar.'07d Similar mechanisms seem to be involved in the removal of the 6-OH group during the bio- synthesis of cytidine diphosphate 3,6-dioxy-~-glucose. O8 CHZOH CH,OH HO O.TDP (q O*TDP -lADHo+q) OH OH (55) J + Enz.-NA D + OH OH (56) lo' J. S. Schutzbach and D. S. Feingold J. Biol. Chem. 1970 245 2476. H. Grisebach and V. Dobereiner Biochem. Biophys. Res. Comm. 1964 17 737. lo' (a)A. Melo W. H. Elliott and L. Glaser J. Biol. Chem. 1968,243 1467; (b)0.Gabriel and L. Lingquist J. Biol.Chem. 1968 243 1479; (c) H. Zarkowsky and L. Glaser J. Biol. Chem. 1969,244,4750; (d)S. F. Wang and 0.Gabriel J. Biol. Chem. 1970,245,8. lo* H. Pape and J. E. Strominger J. Biol. Chem. 1969 244 3598.