12 Enzyme Chemistry By D. GANl Department of Chemistry The University Southampton SO9 5NH 1 Introduction Annual Reports last reviewed enzyme chemistry in 1985.' This review follows the same format and thus the main objective is to highlight some of the more important findings published in 1986 rather than to provide exhaustive coverage ofthe literature. Again emphasis is placed upon papers describing structural and mechanistic aspects of enzymic catalysis rather than those describing the isolation and/or purification of the proteins. Over the past year many reviews have bcen published in the area of enzyme chemistry. Battersby has reviewed studies largely from the Cambridge Group on the biosynthesis of vitamin B12.2 Breslow has surveyed the use of artificial enzymes and enzyme models3 and Kuhn Schewe and Rapoport4 have reviewed the stereochemistry of the reactions of the lipoxygenase enzymes and their metabolites.The use of enzymes in synthesis has been reported by Jones;' other reviews will be cited in context. 2 Racemase Enzymes Proline racemase operates by a two-base mechanism and is not pyridoxal 5'-phosphate dependent. The enzyme has been studied in detail by Knowles and his co-workers.6-'2 Following on from earlier work by Abeles13 who showed that there are two forms of the free enzyme Knowles was able to demonstrate that the proline racemase system shows three kinetic regimes -unsaturated saturated and oversatur- ated -depending upon the concentration of substrate. At low concentrations of substrate the initial rates for racemization under equilibrium conditions increased ' D.Gani Annu. Rep. hog. Chem. Sect. B ch. 11 Enzyme Chemistry. A. R. Battersby Acc. Chem. Res. 1986 19 147. R. Breslow Ado. Enzymol Relat. Areas Mol. Biol. 1986 58 1. H. Kuhn T. Schewe and S. M. Rapoport Adv. Enzymol. Relat. Areas Mol. Biol. 1986 58 273. J. B. Jones Tetrahedron 1986 42 3351. L. M. Fisher W. J. Albery and J. R. Knowles. Biochemistry 1986 25 2529. ' L. M. Fisher W. J. Albery and J. R. Knowles Biochemistry 1986 25 2538. * L. M. Fisher J. G. Belasco T. W. Bruice W. J. Albery and J. R. Knowles Biochemistry 1986,25,2543. J. G. Belasco W. J. Albery and J. R. Knowles Biochemistry 1986 25 2552. lo J. G. Belasco T. W. Bruice W. J. Albery and J.R. Knowles Biochemistry 1986 25 2558. " J. G. Belasco T. W. Bruice L. M. Fisher W. J. Albery and J. R. Knowles Biochemistry 1986 25 2564. W. J. Albery and J. R. Knowles Biochemistry 1986 25 2572. '3 G. Rudnick and R. H. Abeles Biochemistry 1975 14 4515. 303 304 D. Gani with increasing substrate concentration. When the concentration of substrate was increased still further the enzyme showed saturation kinetics and the initial rate was unaffected by substrate concentration. At high substrate concentration the enzyme entered the oversaturated regime where an increase in substrate concentra- tion led to a decrease in the reaction velocity after the initial phase where the rates were the same. Given that all trivial explanations for the observation of an oversatur- ated regime can be discounted Knowles suggested6 that at high substrate concentra- tion the amounts of free E and E2 were small and thus the interconversion of the free enzyme from the product releasing form (E2) to the substrate binding form (E,) became rate-limiting Scheme 1.k2 EIP r EZP Scheme 1 Using the tracer perturbation method which involves measurement of the time- dependent distribution of radiolabelled substrate and product initially at equili- brium when the system is perturbed by the addition of a large amount of one of the unlabelled substrates Knowles was able to estimate the rate constant for the enzyme intercon~ersion.~ Furthermore it was shown that the enzyme is bound-state saturated and thus the rate determining step for interconversion does not involve k4 and k-4 for the saturated regime but rather steps 1-3.In a further two papers’.’ using [2-2H]- and [2-3H]-substrates Knowles showed that the transition state that involves the protonation and deprotonation of proline the TS of step 2 (Scheme 2) must be at least partially rate-limiting. To test for the concertedness of step 2 a double fractionation experiment was undertaken.’ The results indicated that the on/off rates for the substrate and product were faster than the racemization step and that either the racemase reaction proceeded in a concerted manner or that the Substrate Product J H Step 1 --b c- Scheme 2 Enzyme Chemistry 305 reaction was stepwise and involved catalytic groups such as thiols with ground-state fractionation factors of about 0.5.The fractionation factors of the protons bound to the essential catalytic groups of proline racemase were determined using two competitive deuterium wash-out experiments." The results allowed Knowles and co-workers to conclude that two thiol groupsI4 of cysteine residues mediate enzymic racemization. These findings confirmed earlier proposals by Abeles.13 In order to probe the nature of the interconversion of the two unliganded forms of proline racemase,' a number of experiments including competitive deuterium wash-out were conducted under oversaturating conditions where it has been shown that this interconversion was rate lirnitir~g.~?~ The results supported a stepwise mechanism for the interconversion of the free enzyme forms in which a proton is abstracted from a bound water molecule to give a reaction intermediate possessing a hydroxide ion bound to the diprotonated form of the enzyme Scheme 3.H H H I 0 I 0 H' Scheme 3 E2 Finally Albery and Knowles collated the results from their studies of proline racemase to allow the construction of the complete free energy profile of the reaction for the unsaturated saturated and oversaturated regimes.12 The pyridoxal 5'-phosphate-dependent enzyme alanine racemase has been the subject of much attention from Walsh and his co-workers. The biosynthetic enzyme encoded by the alr gene of Salmonella typhimurium has now been purified and characterized" and compared to the catabolic enzyme encoded by the dad B The two enzymes share 43% amino acid sequence homology" and active- site decapeptides of each enzyme which contain lysine are identical.The biosynthetic enzyme was inhibited through treatment with both D-and L-P-haloalanines to give a ternary inactivated enzyme ~omplex'~ similar to that of the catabolic enzyme,19 Scheme 4. Walsh has also studied alanine racemase from Gram-positive bacteria.*' Recently these studies have been extended to the enzyme from Bacillus stearothermophilus where the gene has now been cloned and expressed in E. coli21 Investigation of the time-dependent inhibition of the enzyme with both antipodes of ( 1-aminoethy1)phos-phonate the phosphonate analogues of alanine reyealed that inhibition was due l4 R.J. Szawelski C. W. Wharton and S. White Biochem. SOC.Trans. 1982 10 232. l5 N. Esaki and C. T. Walsh Biochemistry 1986 25 3261. 16 S. A. Wasserman E. Daub P. Grisafi D. Bolstein and C. T. Walsh Biochemistry 1984 23 5182. l7 B. Badet D. Roise and C. T. Walsh Biochemistry 1984 23 5188. l8 N. G. Galakatos E. Daub D. Botstein and C. T. Walsh Biochemistry 1986 25 3255. 19 D. Roise K. Soda T. Yagi and C. Walsh Biochemistry 1984 23 5195. 20 B. Badet and C. Walsh Biochemistry 1985 24 1333. 21 K. Inagaki K. Tanizawa B. Badet C. T. Walsh H. Tanaka and K. Soda Biochemistry 1986,25,3268. 306 D. Gani H H H 1 1 T' -N+ Me H I H Scheme 4 to the formation of non-covalent slowly dissociating enzyme-inhibitor complexes.22 Interestingly the time-dependent loss of activity by this inhibitor appears to be general for alanine racemases from Gram-positive bacteria" but not from Gram- negative bacteria to which the racemases are not s~sceptible.'~"~~~~~~~ 3 Other mridoxal Phosphate-dependent Enzymes The identification of structure (l) Scheme 4 as the heat hydrolysis-denatured product of the inactivated complexes formed from ~acemase,'~~'~ tran~aminase,~~ and decarb~xylase~~ enzymes and P-nucleofuge-substituted alanines (2; X = F C1 OSO or OAc) tends to suggest that the depicted inactivation mechanism is general.However it should be pointed out that these suicide substrates all lack a distal enzyme-binding group after the elimination of HX has occurred and thus there may be less translationary restriction than for suicide inhibitors containing additional active-site binding groups e.g.(S E ) -4-amino-5-fluoropent-2-enoic acid.26 These suicide substrates may be unable to dissociate in order to become involved in enamine condensation with the coenzyme aldamine and thus could potentially act 22 B. Badet K. Inagaki K. Soda and C. T. Walsh Biochemistry 1986 25 3275. 23 E. Wang and C. Walsh Biochemistry 1978 17 1313. 24 H. Veno J. J. Likos and D. E. Metzler Biochemistry 1982 21 4387. 25 J. J. Likos H. Veno R. W. Feldhaus and D. E. Metzler Biochemistry 1982 21 4377. 26 R. B. Silverman B. J. Invergo and J. Mathew J. Med. Chem. 1986 29 1840. Enzyme Chemistry as Michael acceptors for enzyme-bound nu~leophiles.~~~~~ Clearly the structures of several more inactivated ternary complexes need to be determined before a single inactivation mechanism can be assumed general.Recently Grigg has presented a new perspective of pyridoxal chemistry relevant to enzymic inhibiti~n.~’ Churchich has studied homogeneous porcine brain glutamate decarboxylase.28 The enzyme catalyses a slow decarboxylative transamination reaction with L-glutamic acid at ca. lop4times the rate of normal a-decarboxylation to give succinic acid semialdehyde and presumably inactive apoenzyme-PMP. Treatment of the inactive enzyme with phosphopyridoxyl-ethanolaminephosphate restores activity through formation of pyridoxal5’-phosphate. This reaction (Scheme 5)is mechanisti- cally interesting and shows similarities to observations reported by Riva and co- worker~.~~.~’ Gani has shown that L-methionine decarboxylase from the fern species Dryopteris felixmas has a wide substrate specificity and catalyses the decarboxylation of H H H 1 1 TEnz Me -NH, Active Ct H H Scheme 5 27 P.Armstrong D. T. Elmore R. Grigg and C. H. Williams Biochem. SOC. Trans. 1986 14 404. 28 S. Y. Chai and J. E. Churchich Eur. J. Biochem. 1986 160 515. 29 F. Eva D. Carolti A. Giartosio and C. Turano J. Biol. Chem. 1980 285,9230. 30 D. Carolti F. Riva R. Santucci F. Ascoli and P. Fasella Eur. J. Biochem. 1982 124 589. 308 D. Gani L-methionine with retention of configuration at C-2.31Christen and co-workers have shown that chicken mitochondrial aspartate aminotransferase catalyses the stereo-specific exchange of the c-4'-pro-s hydrogen of [4'-3H]-pyridoxamine 5'-phosphate with solvent in the absence of ketoacid substrate.32This result will be of particular use in investigating the acid-base properties of C-4' and the modulating effects of the active-site environment independently of aldimine and/or ketimine formation.Gehring has investigated the mechanism of the mitochondrial chicken heart enzyme using cryoenzymological methods.33In the cryosolvent (80% methanol) the kinetic parameters for the overall reaction with the substrates cysteine sulphinate and oxaloacetic acid were essentially unaltered. At -44 "C mixing PLP-holoenzyme and substrate resulted in the formation of an intermediate absorbing at 430nm (probably the external aldimine) which decayed in a biphasic process.Quinoid formation was not observed indicating that protonation at C-4' is not rate-limiting (Step 3 Scheme 6) and /or that the equilibrium strongly favours the aldimine. Further analysis of the kinetic data indicated that hydrolysis of the ketamine is probably rate-limiting (step 4 Scheme 6). "r"" NH2 L RljjCo; Step 1 Step 2 PLP-Holoenzyme-Substrate -+ H H 430 nm 490 nm 11~:~ RDS "r" NH2 RYco2-i. Step 4 H\ RDS He "H PMP-Holoenzyme-Product H 340 nm Scheme 6 " D. E. Stevenson M. Akhtar and D. Gani Tetrahedron Lett. 1986 27 5661. 32 H. P. Tobler P. Christen and H. Gehring J. Biol.Chem. 1986 261 7105. 33 H. Gehring Eur. J. Biochem. 1986 159 291. Enzyme Chemistry 309 The differential inhibition of the porcine heart enzyme with D-and L-hydrazinosuccinate has been studied recently.34 The inhibition of L-enantiomer was found to involve a two-step inactivation mechanism in contrast to that for the D-antipode and resulted in a lower inhibition constant. Morino and co-workers have identified the 4’-H internal aldimine ‘H-n.m.r. chemical shift value for both cytosolic and mitochondrial aspartate amin~transferase.~’ Remarkably the resonance for 4’-H of the cytosolic holoenzyme showed no variation with pH whereas the chemical shift of the mitochondrial holoenzyme varied considerably with pH reflecting significant differences in the respective coenzyme microenvironments.Comparison of the tryptophan synthase and tryptophanase reactions has been investigated by Miles.36 These workers were interested in establishing why the synthase although capable of catalysing several similar p-elimination and P-replace- ment reactions to tryptophanase was not apparently able to catalyse the elimination of indole from L-tryptophan the physiological reaction for tryptophanase Scheme 7. Careful examination of the tryptophanase-free synthase both in the and p2 complexes revealed that L-tryptophan was indeed converted into indole pyruvate and ammonia. H H H + Scheme 7 In a further mechanistic study to assess substrate binding using ‘’F-n.m.r. spectro- scopy Miles showed that D-and ~-5-fluorotryptophan tryptophan and (3s)-2,3-dihydro-5-fluorotryptophanare all slowly isomerized by racemization-epimerization reaction^.^' These reactions occur lo3-10’ times more slowly than the &replacement and p-elimination reactions.Possible mechanisms for proton translocation during the racemization-epimerization process a single base swinging door or a two-base system vide supra are discussed in the light of recent findings with other PLP- dependent enzymes.38 34 R. Yamada Y. Wakabayashi A. Iwashima and T. Hasegawa Biochim. Biophys. Actu 1986,871 279. 35 Y. Morino F. Nagashima S. Tamase M. Yamasaki and T. Higaki Biochemistry 1986 25 1917. 36 S. A. Ahmed B. Martin and E. W. Miles Biochemistry 1986 25 4233. 37 E. W. Miles R. S.Phillips H. J. C. Yeh and L. A. Cohen Biochemistry 1986 25 4240. S. A. Ahmed N. Esaki H. Tanaka and K. Soda Biochemistry 1986 25 385. 310 D. Gani Recently Matthews has studied synergism in the folding of b double mutant of the a subunit of tryptophan synthase from E. COZ~.~~ Using ultraviolet difference spectroscopy and urea-induced unfolding techniques two single inactive mutant proteins Tyr-175 + Cys and Gly 211 + Glu and the active double mutant Cys- 175/Glu-211 were examined. The sum of the changes in stability for the single mutants was found not to be equal to the change observed for the double mutant. As an equality would be expected only if the residues at positions 175 and 211 did not inter-react Matthews concluded that a structural interaction occurred.Kinetic studies revealed that the synergism occurs only after the final rate-limiting step of domain association. Nihira and co-workers have studied the catalytic role4' of an essential tyrosine residue4' in tryptophanase using chemical modification techniques. It appears that the tyrosine is involved in coenzyme binding probably through an H-bonding interaction with the pyridine heteroatom. Although the modified holoenzyme is able to catalyse transaldimination in the presence of substrate C" -H labilization does not occur. 4 Carboxylation Reactions Transcarboxylase is a biotin-dependent enzyme which catalyses the reversible forma- tion of (S)-methylmalonyl CoA from propionyl CoA (using oxaloacetic acid as a source of carbon dioxide) in two steps involving the intermediacy of carboxybiotin Scheme 8.The reaction is unique among biotin-dependent enzymes in that it is not driven by ATP and bicarbonate. The kinetic mechanism is two-site ping-pong and the reaction has an overall equilibrium constant of about 1. COSCoA COSCoA 0 Transcarbox ylase M,i-H + ,&/co2- + Me+-CO; + via H -02c H Scheme 8 In order to gain insight into the mechanism of the carboxylation of pyruvate to oxaloacetate catalysed by the enzyme Knowles measured the deuterium ["( V/ K)] and l3C [13( V/K)] isotope eff e~ts.~~ By comparison of the slopes of double-reciprocal plots of the initial rates obtained using [2H,]pyruvate and unlabelled material as substrates D( V/K)was found to be 1.39.The I3C kinetic isoWpe effect for carboxy- lation was found to be 1.0227 when measured using the competitive method. These results together suggested concerted or balanced stepwise mechanisms for the carboxylate reaction. In order to determine whether the removal of the proton from 39 M. R. Harte N. B. Tweedy and C. R. Matthews Biochemistry 1986 25 6356. 40 T. Kakizono T. Nihira and H. Taguchi Biochem. Biophys. Res. Commun. 1986 137 964. 41 T. Nihira T. Toraya and S. Fukui Eur. J. Biochem. 1981 119 273. 42 S.J. O'Keefe and J. R. Knowles Biochemistry 1986 25 6077. Enzyme Chemistry 311 pyruvate and the addition of the carboxyl group occur in the same or different steps Knowles made use of the double-isotope fractionation test.This method predicts that the ‘3C-isotope effect for carboxylation will decrease in magnitude when protium is replaced by deuterium in the methyl group of pyruvate if the overall reaction occurs via a stepwise mechanism. Such a decrease would occur because the C-C bond formation step would be less rate-limiting in a system where the free energy at the transition state for C-H bond cleavage was increased. A concerted mechan- ism however would predict no change in the observed I3C isotope effect. Knowles observed that the value of 13( V/ K) dropped from 1.0227 (for undeuteriated sub- strate) to 1.0141 (for deuteriated pyruvate) and thus the results clearly indicated that the reaction was stepwise. The stereochemistry of the carboxylation of phosphoenolpyruvate catalysed by phosphoenolpyruvate carboxykinase from both chicken liver and Ascaris (helminth) muscle has been investigated by Nowak using (2)-3-fluorophosphoenolpyruvate.43 Analysis of the carboxylated products which had not incorporated deuterium through enolization using 19F-n.m.r.spectroscopic techniques showed that carboxylation had occurred at the 3-si face of enzyme-bound substrate. These results parallel those obtained by Roseu using specifically labelled [3-’H,3H]-phosphoenolpyruvate and the enzyme from pigeon liver. Frey and co-workers have studied the stereochemical course of the phosphoenol- pyruvate carboxykinase reaction using inosine 5‘4 3-thiotriphosphate) (ITP,S) as the ~ubstrate.~’ Incubation of the rat liver cytosolic enzyme with (Rp)-[ y-1802] ITP,S and oxaloacetate gave (Sp)-thio[180]phosphoenolpyruvate.Thus the reaction proceeded with overall inversion at the phosphorus atom as for the mitochondria1 enzyme46 indicating that an odd number of phosphoryl transfers (probably one) were involved in the reaction.The catalytically essential cysteine residue in phos- phoenolpyruvate carboxylase has been tentatively identified through protection studies with the substrate analogue 2-pho~pholactate.~~ The oxygen dependence of vitamin K-dependent carboxylase has been studied by S~ttie.~~ The carboxylase reaction of ribulose 1,5-bisphosphate carboxylase/oxy- genase has been investigated by Schloss using deuteriated and tritiated substrates for the enzymes from spinach and Rhodospirillum r~brum.~~ With the spinach enzyme isotope effects at high pH on V,, and V/K did not vary with C02 concentration from K to 100 times K,.This result was interpreted in favour of a mechanism whereby Cot adds to the ene-diol of ribulose bisphosphate in a Theorell-Chance- type bimolecular manner after abstraction of the C-3 proton of the sugar bisphos- phate. The isotope effect on V,, was found to be pH-dependent 2 at high pH up to about 9 at low pH. Inhibition by the substrate analogue xylulose 1,5-bisphosphate was also pH-dependent. In contrast to the spinach enzyme both V,, and V/K 43 S. H. Hwang and T. Nowak Biochemistry 1986,25 5590. 44 I. A. Rose E. L. O’Connell P. Noce M. F. Utter H. G. Wood J. M. Willard T. G. Cooper and M. Benziman J.Biol. Chem. 1969 224 6130. 45 J. M. Knopka H. A. Lardy and P. A. Frey Biochemistry 1986 25 5571. 46 K. F. Sheu H. T. Ho L. D. Nolan P. Mankovitz,J. P. Richard M. F. Utter and P. A. Frey Biochemistry 1984 23 1779. 47 S. Ishijima K. hi and H. Katsuki J. Biochem 1986 99,1299. 48 J. J. McTigue and J. W. Suttie FEBS Lett. 1986 200 71. 49 D. E. Van Dyk and J. V. Schloss Biochemistry 1986 25 5145. 312 D.Gani were found to be pH-dependent for the enzyme from R. rubrum. From this data Schloss suggested that both enzymes contain an essential base of about pK 7.5 probably imidazole of histidine or e-NH2 of lysine which abstracts the C-3 proton in the first step of the reaction. Pierce and coworkers have shown that the enzymic reaction proceeds via the ordered addition and enolization of ribulose bisphosphate followed by reaction with gaseous C02.50 The inability to detect an enzyme-C02 complex directly by 13C-n.m.r.spectroscopy in these studies supports the view that a Theorell-Chance type mechanism (no Michaelis complex for C02) occurs.49 5 Coenzyme BIz-dependent Rearrangements The stereochemical course of the reaction catalysed by coenzyme B12-dependent 2-methyleneglutarate mutase (Scheme 9) has been investigated by Buckel.'l H Isomerase Mutase -___, MeIco; Me CO; D2O D2O c CO2H (2s) Scheme 9 In order to determine the absolute configuration of biologically active 3-methyl- itaconic acid the racemic material was incubated with a cell-free extract of Clos-trzdiurn barkeri until half was consumed.Re-isolated methylitaconic acid gave a specific optical rotation of [a12 + 3.58" indicating by comparison with published values that the absolute configuration at C-3 was (S). The (R)-antipode was synthesized using the light-inactivated cell free extract; AdoCbl-dependent enzymes are light sensitive. This preparation contained an isomerase activity which was able to convert 2,3-dimethylmaleic acid into (3R)-3-methylitaconic acid [a]',"-3.53" via si-face protonation. Thus the biologically active isomer possessed (3R) -absolute stereochemistry. To determine the stereochemical course of the rearrange- ment of (3 R)-3 -methylitaconate to 2-methyleneglutaride the cell-free extract in J. Pierce G. H. Lorimer and G. S.Reddy Biochemistry 1986 25 1636. 5' G. Hartrampf and W. Buckel Eur. J. Biochem. 1986 156 301. Enzyme Chemistry 313 deuterium oxide was incubated with 2,3-dimethylmaleic acid. The resulting deuteri- ated 2-methyleneglutaric acid was isolated purified and oxidized with nitric acid to give deuteriated succinic acid. Comparison of the CD spectrum of the succinic acid with published data revealed that the sample possessed (2s)-absolute stereochemistry and thus the migration of the acryloyl group had occurred with inversion of configuration at C-3 of the substrate. This stereochemical result is similar to that for glutamate mutase but differs from that of methylmalonyl CoA mutase. Retey has recently investigated some of the unusual side reactions observed with methylmalonyl CoA m~tase.~~ Using homogeneous enzyme preparations and prepar- ations contaminated with methylmalonyl CoA epimerase Retey was able to show that the previously observed high wash-out of labels3 from methyl- and ethyl-malonyl CoA substrates54 and low wash-in of solvent label resulted from three combined factors.Infidelity of the mutase with respect to radical abstraction of the 3-pro-R and 3-pro-S H-atoms high intramolecular isotopic discrimination and contamination with the epimerase Scheme 10. Thus where a deuterium atom only occupies the preferred 3-pro-R position the isotope effect and steric preference work against each other and a significant amount of substrate is turned over uia abstraction of the 3-pro-S protium atom.Some of the label in this product then occupies the acidic position of methylmalonyl CoA which is rapidly exchanged with solvent protium. If deuterium occupies the 3-pro-S position where both steric preferences and the isotope effect favour the removal of the 3-pro-R hydrogen the deuterium will end up almost entirely in an exchangeable position. R I 5 *H *H COSCoA *H ,COSCoA *H-hCO; eco; *H OH *H OH R \ I *H major Przzi!tion 2 pathway 2 5 OH COSCoA pro-S-H migration C0; *-__-_ *-----*H-* minor pathway *H OHs QH Solvent proton R.=5’-deoxyadenosyl-5’-radical exchange Scheme 10 52 K. Wolfe M. Michenfelder A. Konig W. E. Hull and J. Retey Eur. J. Biochem. 1986 156 545. 53 A. Gaudemer J.Zylber N. Zylber M. Baran-Harszac W. E. Hull M. Fountaoulakis A. Konig K. Wolfe and J. Retey Eur. J. Biochem. 1981 119 279. 54 J. Retey E. H. Smith and B. Zagalak Eur. 1. Biochem. 1978 83 437. 314 D. Gani Leadlay has recently studied the subunit structure of the mutase from Propionibac-terium sherma n ii. 6 Redox Reactions The chemistry and structure of NAD(P)-dependent alcohol dehydrogenase have been reviewed by Biellmann.56 Recent work by Biellmann5’ has examined the effect of the modified coenzyme 3-benzoylpyridine-adeninedinucleotide upon the catalytic properties of horse liver alcohol dehydrogenase. In the modified system only primary alcohols are accepted as substrates. Kroneck has studied the flavin-dependent alcohol oxidase of yeast.58 This enzyme converts lower alcohols and oxygen into the corresponding aldehydes and hydrogen peroxide.Kinetic data obtained using methanol and deuteriomethanol as substrates under both single and multiple turnover conditions allowed a four-step mechanism to be proposed (1) formation of a Michaelis complex between the enzyme and the alcohol; (2) partially rate-limiting scission of the substrate C-H bond; (3) reaction of the reduced enzyme-aldehyde complex with dioxygen; and finally (4) dissociation of the product from the reoxidized enzyme complex at a rate slow enough to affect the overall reaction rate. The C-H bond cleavage appears to occur homolyti- ally.^^,^^ Ghisla has investigated the kinetic properties of acyl CoA dehycrogenase from pig kidney using a range of deuteriated butyryl CoA substrates labelled at C-2 and C-3 and in both positions.60 In turnover catalysis isotope effects of 2 3.6 and 9 were observed for each substrate respectively while in the reductive half-reaction the corresponding values were 2.5 14 and 28.No intermediates were apparent during the reduction of oxidized enzyme to the presumed reduced enzyme crotonyl CoA complex indicating a high degree of concertedness during C-2-H and C-3-H bond rupture. The results are compatible with a mechanism in which simultaneously the C-2 hydrogen is removed as a proton and the C-3 hydrogen is transferred to the flavin as hydride. Dihydroorotate dehydrogenase61 and dihydroorotate oxi- dase62,63appear to catalyse reactions via a stepwise hydride-transfer mechanism.Glutathione reductase is a disulphide-containing flavoprotein for which the X-ray crystal structure has been determined.64965 Lively and McFarland have studied the structure of the EH2 intermediate that is formed on reaction of the flavoprotein with NADPH using resonance Raman spectroscopy.66 The EH2 species is important as it is kinetically competent for the reaction with glutathione. The EH intermediate is an oxidized flavin and furthermore s5 F. Francalanci N. K. Davis J. Q. Fuller D. Murfitt and P. L. Leadlay Biochem. J. 1986 236 489. 56 J. Biellmann Acc. Chem. Res. 1986 19 321. 57 J. Samama D. Hirsch P. Gaulas and J. Biellmann Eur. J. Biochem. 1986 159 375. 58 J. Geissler S. Ghisla and P. M. H. Kroneck Eur.1.Biochem. 1986 160 93. 59 B. Sherry and R. H. Abeles Biochemistry 1985 24 2594. 60 B. Pohl T. Raichte and S. Ghisla Eur. J. Biochem. 1986 160 109. 61 P. Blattrnann and J. Retey Eur. J. Biochem. 1972 30 130. 62 R. A. Pascal N. Trang A. Cerami and C. T. Walsh Biochemistry 1983,22 171. R. A. Pascal and C. T. Walsh Biochemistry 1984 23 2745. 64 G. E. Schulz R.H. Schirmer and E. F. Pai J. Mol. Bid. 1982 160 287. 65 E. F. Pai and G. E. Schulz J. Biol. Chem. 1983 258 1752. 66 C. R. Lively and J. T. McFarland Biochem. Biophys. Res. Commun. 1986 136 22. Enzyme Chemistry the most likely structure involves charge transfer donation of electrons from the thiolate anion of Cys-63 to the N-5 flavin heteroatom (cf. ref. 1). Dihydrolipoamide dehydrogenase a related protein has now been purified from Halobacterium h~lobiurn.~~ This enzyme is not (unusually) associated with 2-oxoacid dehydrogenase multienzyme complex.Recently Walsh has described a new member of the disulphide-containing flavoprotein group that includes lipoamide dehydrogenase and glutathione and mercuric reductase which reduces the macrocyclic disulphide trypanothione to the dithiol. The new enzyme trypanothione reductase shows considerable active-site peptide amino-acid homology with glutathione reductase.68 Walsh and co-workers have also investigated the interaction of the EH form of mercuric reductase with Hg2+.69 Although the substrate (Hg2+) is bound very tightly by the active-site thiol groups of the reduced enzyme no reduction of the substrate occurs unless additional reducing equivalents are provided.Lindskog has recently published the results of rapid-scan stopped-flow studies of the pH-dependence of the reaction between the enzyme and NADPH.70 Rate constants and isotope effects are reported. Mercuric reductase is the second enzyme in the microbial mercury detoxification pathway. The first organomercurial lyase catalyses the protonolysis of carbon mercury bonds Scheme 11. 0rganornercurial lyase R-Hg-X R-H+HgZ+ Mercuric reductase HgZ+ -* Hg O NADPH NADP’ Scheme 11 Walsh has overproduced the lyase to the level of 3% of the soluble cell protein in E. coli by a construction using the l7 promoter. The homogeneous enzyme is a monomer (M 22 400) and requires no detectable cofactors or metal ions.71 The enzyme is able to catalyse protonolysis of the C-Hg bond in a wide range of organomercurial salts including primary secondary and tertiary alkyl vinyl allyl and aryl mercurials to give the hydrocarbon and mercuric ion at turnover rates in the range 1-240 min-’.Walsh has studied the mechanism of the enzymic reaction using a wide range of structurally diverse organomercury substrate^.^ The protonoly- sis products are summarized in Scheme 12. The results indicate that the enzyme operates via an SE2 mechanism and thus is novel in this respect. Ferry Walsh and co-workers have also investigated the mechanism of the 5-deazaflavin coenzyme F,,,-dependent formate dehydrogenase rea~tion,’~ Scheme 67 M.T. Danson A. McQuatties and K. J. Stevenson Biochemistry 1986 25 3880. 68 S. L. Shomes A. H. Fairlamb A. Cerami and C. T. Walsh Biochemistry 1986 25 3519. 69 S. M. Miller D. P. Ballou V. Massey C. H. Williams and C. T. Walsh J. Biol. Chem. 1986 261 8081. 70 L. Sahlman A. Lambeir and S. Lindskog Eur. J. Biochem. 1986 156 479. 71 T. P. Begley A. E. Walts and C. T. Walsh Biochemistry 1986 25 7186. 72 T. P. Begley A. E. Walts and C. T. Walsh Biochemistry 1986 25 7192. 73 N. L. Schauer J. G. Ferry J. F. Homek W. Orme-Johnson and C. Walsh Biochemistry 1986,25,7163. 316 D. Guni HgCl H Scheme 12 13. The enzyme is specific for the si-face hydride transfer to C-5 of F420 and thus joins three other F,,,-recognizing methogen enzymes in this stereospecificity.While catalysis probably occurs via hydride-transfer from formate to the enzyme to generate an EH2 species and then by hydride-transfer back out to coenzyme F420 the formate-derived hydrogen exchanges with solvent hydrogen before transfer to F42p The kinetics of hydride-transfer from formate revealed that the step was not rate- limiting suggesting that an internal electron transfer may be. Villafranca has determined kinetically the order of substrate binding for the dopamine P-hydrolylase reaction using a range of substrates of differing structure including unlabelled and [2,2-2H2]tyramine.74 The results suggested that the reaction occurred via a ping-pong mechanism in which tyramine binds to the enzyme after the release of oxidized ascorbate.Subsequently oxygen binds to form a ternary complex. Benkovic and co-workers have studied iron-binding in phenylalanine hydroxylase (the enzyme which catalyses the conversion of phenylalanine into tyrosine) using e.p.r. spectro~copy.’~ It appears that iron can be bound in two environments which are both populated in the crude enzyme. The two environments are not interconvert- ible. Benkovic has also reported on studies of phenylalanine hydroxylase from Chromobacterium viol~ceum.~~ This pterin-dependent enzyme is a copper-containing 74 P. F. Fitzpatrick M. R. Harpel and J. J. Villafranca Arch. Biochem. Biophys. 1986 249 70. 75 L. M. Bloom S. J. Benkovic and B. J. Gamey Biochemistry 1986 25 4204. 76 S.0. Pember J. J. Villafranca and S. J. Benkovic Biochemistry 1986 25 6611. Enzyme Chemistry F420 Fo R=H monooxygenase which contains 1 mole of Cu2+ per mole of enzyme. The enzyme shows some features in common with the mammalian iron-containing enzyme such as the intermediacy of the 4a- hydrate of 6-methyltetrahydropterin suggesting that the mechanism of oxygen activation is similar for both enzymes. The copper-containing monooxygenase tyrosinase has been reported to catalyse 7-an unusual oxidative decarboxylation of 3,4-dihydroxymandelate. ' ' A mechanism for the reaction has been proposed. Muller has investigated the role of active-site residues in the p-hydroxybenzoate hydroxylase reaction by chemical modification and sequence determination.Residues 343-5 Ser-Trp-Trp were previously assigned erroneo~sly.'~ Ghisla and co-workers have described the preparation and properties of 6-substituted flavins as active-site probes for flavin enzymes" and Ghisla and Massey have reviewed the use of such probes.*' Severin et al. have reviewed the chemistry of the dehydrogenases of a-ketoacids.81 Finally Walsh has studied reactions cata- lysed by the flavoenzyme cyclohexanone oxygena~e.~~~ In the oxidation of a chiral boronic acid the chiral secondary alcohol is produced with complete retention of configuration Scheme 14.82 It appears that enzyme-bound FAD-4a-OOHS3 is the actual oxygenation agent in this versatile and probably prototypic biological Bayer- Villiger catalyst. The enzyme is also reported to catalyse the oxidation of organoselenides to ~elenoxides.~~ Enzymically-formed allylic selenoxide rearranged in a 2,3-sigmatropic 77 M.Sugumaran Biochemistry 1986 25 4489. 78 R. A. Wijnands W. J. Weijer F. Muller P. A. Jekel W. J. H. \-an Berkel and J. J. Beinlema Biochemistry 1986 25 421 1. 79 S. Ghisla V. Massey and K. Yagi Biochemistry 1986 25 3282. S. Ghisla and V. Massey Biochem. J. 1986 239 1. S. E. Severin L. S. Khailova and V. S. Gomazkova Adu. Enz. Re& 1986 25 347. 82 J. A. Latham and C. Walsh J. Chem. Soc. Chem. Commun. 1986 527. 83 C. C. Ryerson D. P. Ballou and C. Walsh Biochemistry 1982 21 2644. 84 J. A. Latham B. P. Branchaud Y. J. Chen and C. Walsh J. Chem. SOC.,Chem. Commun. 1986 528. 318 D.Gani Scheme 14 manner to give after hydrolysis racemic alcohols. It is not clear that the initial oxidation at selenium is non-enantioselective since selenoxides readily racemize through hydrate formation and this process could outcompete the 2,3-sigmatropic rearrangement rate. The oxygenating component of 2,5-diketocamphane 1,2-monooxygenase from Pseudornonas putida has been purified to homogeneity by Tr~dgill.~~ This enzyme is also a simple flavoprotein and operates in a similar manner to the cyclohexanone oxygenase enzyme. 7 Ammonia-lyase and Dehydrase Enzymes Over the past few years advances in genetic techniques have allowed the convenient determination of the primary structure of enzymes using relatively simple gene nucleic acid sequencing techniques.Very recently the application of these techniques and the comparison of deduced amino-acid sequences have produced exciting results. Aspartase (L-aspartate ammonia-lyase) catalyses the reversible deamination of L-aspartic acid to give ammonia and fumaric acid. The enzyme is thought to follow a carbanion (E 1cb)-type mechanism in which an active-site thiolate86 (from Cys-140 or -43087) abstracts the pro-R-hydrogen as a proton from C-3 before rate-limiting C- N bond cleavage.88 Studies of the binding affinity of transition-state intermediates has suggested that fumarase which catalyses an analogous process with the same stereospecificity but with water instead of ammonia also follows the same mechanistic pathway.88 However no catalytically essential active-site thiolates have been implicated.Compari~on~~.~~ of the deduced amino-acid sequence from two genes for each enzyme has shown that there is a striking degree of homology. Alignment of the sequences shows that the active-site cysteine residues of L-aspartase are replaced by serine and alanine in fumarase Figure lag9 Clearly aspartase and fumarase have evolved from a common gene; it will be interesting to compare the other chemically related members of the group for example 3-methylaspartase. Gani" has shown that this enzyme can utilize substrates containing halogens in place of the methyl group of mesaconic acid the best substrate for the retrophysiological amination reaction and that chiral highly functionalized aspartic acids can be prepared in good yield.The observed suicide inhibition shown by bromofumaric acid has been rationalized (Scheme 15) in terms of the alkylation 85 D. G. Taylor and P. W. Trudgill J. Bacferiol. 1986 165 489. 86 N. Ida and M. Tokushige J. Biochem. 1985 98 793. 87 J. S. Takagi N. Ida M. Tokushige H. Sakamotom and Y. Shimura Nucl. Acids Res. 1985 13 2063. 88 D. J. T. Porter and H. J. Bright 1.Biol. Chem. 1980 225 4772; I. I. Nuiry J. D. Hermes P. M. Weiss C. Chen and P. F. Cook Biochemistry 1984 23. 5168. 89 S.A. Woods J. S. Miles R. E. Roberts and J. R.Guest Biochem. J. 1986 237 547. 90 T. Takagi and M. Kisumi J. Bacreriol 1985 161 1. 91 M. Akhtar M. A. Cohen and D. Gani J. Chem. SOC. Chem. Commun. 1986 1290. 56 Clt G 55 59 i15 114 119 175 173 178 235 233 238 295 293 298 355 353 WLASGPRCGlGEIVlPENEPGSSlMPGKVNPTQSEALTMl~AQ N DlTuVTIMA A &IS GP R d N L LQ AIG S S IMPmK V N P~P~~VflFl;t?lll~G 358 415 413 418 466 461 411 (ReproducedbypermissionfromBiochem.J. 1986,237,547) Figure 1 Alignments of amino acid sequences for the Fume CitG and AspA proteins. The sequences have been aligned for maximum homology based on the DIAGON comparisons. Identical residues shared in two or more of the sequences are boxed. The asterisks mark the methionine histidine and cysteine residues that are conserved in all three sequences the arrows denote the reactive cysteine residues in AspA. The polypeptides are numbered from the residue immediately following the N-terminal methionine D.Gani 320 Enz -Enzl Scheme 15 of an enzyme-bound nucleophile at the active-site by the enzymic amination product 3-bromoaspartic acid. Viola92 has recently shown that L-aspartase contains an activation binding-site for L-aspartate distinct from the active-site. Other substrate analogues which are not substrates can bind to the activation site. The mechanism of the argininosuccinate lyase reaction has been investigated by Rau~hel,~~ Scheme 16. No primary deuterium isotope effect was observed for either V,, or V/K for (2S,3R) -[3-2H,]argininosuccinate while a primary "N isotope effect on V/K of 0.9964 was observed. The "N isotope effect on the equilibrium constant was 1.018. A deuterium solvent isotope of 2.0 was observed on V,,,.The data is consistent with a carbanion mechanism where the C-3 proton abstraction step is not rate-limiting. The inverse "N primary isotope effect and the solvent deuterium isotope effect suggest that protonation of the guanidino group and C-N bond cleavage are kinetically significant. Raushel has also used a new method for the determination of dissociation rates of enzyme-substrate complexes to study the lyase reaction.94 The method which is referred to as 'Dynamic Isotope Exchange Enhancement' involves measuring the 92 W. E. Karsten R. B. Gates and R. E. Viola Biochemistry 1986 25 1299. 93 S. C. Kim and F. M. Raushel Biochemistry 1986 25,4744. 94 S. C. Kim and F. M. Raushel J. Biol. Chem. 1986 261 8163.Enzyme Chemistry Scheme 16 rate of exchange of a labelled product back into substrate during catalysis of the forward reaction when the forward reaction is kept far from equilibrium by the enzymic removal of the non-exchanging product. The values of the ratio of the exchange rate and the net rate for product formation at various concentrations allow estimation of the relative rates of product dissociation from binary and ternary enzyme complexes. Application of this technique led to the conclusion that argininosuccinate-lyase has a random kinetic mechanism. The calculated lower limit for arginine release from enzyme-fumarate-arginine was 0.35 of V,, for arginosuc- cinate formation while the rate of release from the binary complex enzyme-arginine was 210 times V,,,.Schwab has investigated the stereochemical course of the hydration-dehydration reaction catalysed by p-hydroxydecanoyl thioester dehydrase. Synfacial elimina- tion-addition is ~bserved.~’ Schwab has also determined the fate of the N-acetylcys- teamine thioester of 3-decynoic acid a suicide inhibitor for the enzyme,96 Scheme 17. H H I Scheme 17 95 J. M.Schwab A. Habib and J. B. Klassen J. Am. Chem. SOC.,1986 108 5304. 96 J. M. Schwab C. Ho W. Li C. A. Townsend and G. M. Salituro J. Am. Chem. SOC.,1986 108 5309. 322 D. Gani Thorpe and co-workers have disclosed that the flavoprotein medium-chain acyl CoA dehydrogenase from porcine kidney exhibits an intrinsic hydrolase activity towards crotonyl-CoA which leads to the formation of ~-3-hydroxybutyryl-CoA.~~ The hydrolase activity which is FAD or FAD analogue-dependent is about 10-fold lower than the dehydrogenase activity.The activity is not due to contamination. It is proposed that both the activities utilize the same active-site and share some common mechanistic features Scheme 18. R RH I II Dehydrogenase H tH COSCoA *. COSCoA I ‘kH Me H Me -H + \:B-H-B- R R I 0 0 H\. COSCoA H COSCoA M e n H HO +H-B- :B- Scheme 18 8 Proteases and Related Enzymes The seine proteases are a relatively well-studied group of enzymes. Over recent years a detailed picture of the active-site and catalytic mechanism of these enzymes has emerged. Bryan et aL98 working with subtilisin have tested the proposed role of Am-155 as the potential hydrogen-bond donor for the tetrahedral active-site serine-substrate intermediate by replacing asparagine with the isosteric residue leucine using site-directed mutagenesis.Kinetic experiments with the mutant sub- tilisin containing the modified oxyanion hole revealed that K for substrate was unaltered but K,, was 200-300 times lower. These results are consistent with the proposed role of Asn-155 Scheme 19. Meyer et al. have studied the X-ray structure of the product complex of acetyl-Ala- Pro-Ala with porcine pancreatic elastase at 1.65 8 res01ution.~~ Unfortunately the 97 S. Lau P. Powell H. Buettner S. Ghisla and C. Thorpe Biochemistry,1986 25 4184. 98 P. Bryan M.W. Pantoliano S.G. Quill H. Hsiao and T. Poulos Roc. Natl. Acud. Sci. USA 1986 83 3743. 99 E. F. Meyer R. Radpakrishnan G. M. Cole and L. G. Presta J. Mol. Biol. 1986 189,533. Enzyme Chemistry 323 Asn-155 \ N Ser-221 R NH Substrate A Scheme 19 peptide was not present in productive binding mode in the crystal. Abeles"' has synthesized peptidyl fluoromethyl ketones that are specific inhibitors of the serine proteases a-chymotrypsin and porcine pancreatic elastase. By analogy with the corresponding aldehydes it is believed that the fluoromethylketones react with the hydroxyl group of the active-site serine residue to form stable hemiacetals. "F-N.m.r. spectroscopic studies of chymotrypsin-bound trifluoromethyl ketone inhibitors clearly indicate that the carbonyl carbon is tetrahedral at the active-site of the enzyme.Abeles rationalized the observed potency of the transition-state inhibitors (the trifluoro- and difluoro-methyl ketones are better than monofluoro- etc.) in terms of the degree of hydration of the carbonyl group and the pK of the hemiacetal hydroxyl. This latter consideration is important for binding at the anionic hole Scheme 19. Stein has studied the mechanism of inactivation of human leukocyte elastase using the chloromethylketone MeO-Suc-Ala-Ala-Pro-Val-CH2Cl.101 The kinetic data for inactivation suggested that a Michaelis complex was formed initially which reacted reversibly to give a covalent hemiketal. Further reaction via nucleophilic attack by the active-site histidine at the chloromethyl group then gave the fully inactivated enzyme.Abelesio2 has also examined the inactivation of chymotrypsin by 3-benzyl-6- chloro-2-pyrone using X-ray diffraction analysis at 1.9 A resolution. Analysis of the inactivator-enzyme complex showed that the oxygen of active-site Ser- 195 is covalently attached to C-1 of (2)-2-benzylpentenedioicacid that the benzyl group of the inactivator is held in the hydrophobic specificity pocket of the enzyme and that the free carboxylate forms a salt bridge with active-site His-57 (Scheme 20). Bachovchin et ~1"~ have used "N-n.m.r. spectroscopy to examine the H-bonding interactions in the active-site of a-lytic protease. The enzyme was isolated from a histidine-requiring mutant of Lysobacter enzymogenes grown on [i5N-irnidazoZe]his-tidine and thus His-57 the central residue in the catalytic triad was "N-enriched.Spectroscopic analysis of the resting enzyme revealed (in contrast to X-ray diffraction data) that a strong H-bond links the active-site histidine and serine residues. In addition the "N-chemical shifts demonstrated that protonation of the histidine imidazole ring at low pH in the transition-state or similar complexes triggers the 100 B. Imperiali and R. H. Abeles Biochemistry 1986 25 3760. lo' R. Stein and D. A. Trainor Biochemistry 1986 25 5414. 102 D. Ringe J. M. Mottonen M. H. Gelb and R. H. Abeles Biochemistry 1986 25 5633. 103 W. W. Bachovchin Biochemistry 1986 25 7751. 324 D. Gani ___ ~ph H+ 00 I I Ser-195 specificity pocket -on,\ 00Po I H Ser-195 His-57 I IN$ N H Scheme 20 disruption of the aspartate- histidine H-bond.The results suggest a catalytic mechan- ism involving directed movement of the imidazole ring of His-57. Zinc-dependent carbopeptidase A has also been subjected to directed mutagenesis experiments. Replacement of Tyr-248 by phenylalanine has revealed that the phenolic hydroxyl is not required for catalysis although it is probably important for ligand binding. 104,105 Lipscomb has examined the X-ray structure of the enzyme-2- benzyl-4-oxo-5,5,5-trifluoropentanoic acid (trifluoromethyl ketone inhibitor) com- plex.'06 The inhibitor is bound as its hydrate vide supra. The structure has also yielded the first direct interaction of Arg-127 with a zinc-bound oxygen of an inhibitor.Lipscomb has also shown that the Co2+reconstituted enzyme (which has a co-ordination structure essentially identical to the native enzyme as determined by X-ray diffraction) has the same co-ordination environment in crystals and in sol~tion.'~' The stereospecificity of the formation of thiohemiacetal inhibitor complexes has been studied with the cysteine protease papain.'" Both D-and L-N-acetyl phenyl- alanyl [l-'3C]glycinal react stereospecifically with papain and each gives one diastereomer of the thiohemiacetal only. The 13C-n.m.r. chemical shifts are 75.1 and 74.7 p.p.m. respectively. Storer has shown also using 13C-n.m.r. spectroscopy that '3CN-labelled benzoylamidoacetonitrile forms a covalent adduct with the thiol group of cysteine-25 in the active-site of papain.'" It is proposed that the adduct is a thioimidate.104 S. J. Gardell C. S. Craik D. Hilvert M. S. Ordea and W. J. Rutter Noture (London) 1985 317 551. 105 D. Hilvert S. J. Gardell W. J. Rutter and E. T. Kaiser J. Am. Chem. Soc. 1986 108 5298. I06 D. W. Christianson and W. N. Lipscomb J. Am. Chem. Soc. 1986 108 5003. L. C. Kuo W. N. Lipscomb and M. W. Makinen J. Am. Chem. Soc. 1986 108 5003. 108 N. E. MacKenzie S. K. Grant A. I. Scott and J. P. G. Malthouse Biochemistry 1986 25 2293. 109 J. Brisson P. R. Carey and A. C. Storer J. Biol. Chem. 1986 261. 9087. Enzyme Chemistry 325 9 Phosphoryl Transfer Reactions Sowadski has reported on the refined structure of alkaline phosphatase (phos- phomonoester hydrolase EC 3.1.3.1) from E.coli at 2.8 8 resolution."' Alkaline phosphatase is a metalloenzyme that forms an isologous dimer with two reactive centres 32 8 apart. Despite some similarities with the a/P class of proteins the enzyme does not have a characteristic binding cleft at the carboxyl end of the parallel sheet but rather an active pocket that contains a cluster of three functional metal sites. Alkaline phosphatase is a non-specific phosphomonoesterase that hydrolyses small phosphomonoesters as well as the phosphate termini of DNA. The active pocket barely accommodates inorganic phosphate; thus the organic portion of substrates must occupy exposed positions on the surface of the enzyme.Two metal sites M-1 and M-2 3.9 8 apart are occupied by zinc. The third M-3 5 8 and 7 8 away from the other sites is occupied by magnesium or zinc. The imidazole side-chain of histidine residues are ligands to the zinc sites M-1 (three) and M-2 (one). Ligand assignment indicates that sites M-1 M-2 and M-3 correspond to the spectroscopically deduced sites A B and C respectively. The inhibitor arsenate a product analogue binds between Ser-102 and M-1 and M-2. Arg-166 is within H-bonding of the arsenate site. This structural arrangement suggests that despite the lack of protein acid and base functions metals can activate both nucleophiles Ser- 102 and water necessary for double in-line nucleophilic displace- ment on phosphorus Scheme 21.has also studied the metal sites using n.m.r. spectroscopic techniques. The entire primary structure has been rep~rted."~ Enz' Enz+ Enz+ Enz+ R = alkyl or aryl,or H Scheme 21 Hall and Williams have recently studied leaving-group dependence in the phos- phorylation of E. coli alkaline phosphatase using a range of monophosphate ester~."~ The results of these kinetic studies indicate that the leaving groups (alkoxides or phenoxides) bind in a lipophilic site and that the leaving group in the enzyme- substrate complex points away from the surface of the enzyme. Arguments are also advanced to exclude a dissociative mechanism (involving metaphosphate) for the enzyme-catalysed substitution at phosphorus. 110 J. M. Sowadski M.D. Handschumacher H. M. K. Murthy B. A. Foster and H. W. Wyckoff J. Mol. Biol. 1985 186 417. 111 J. E. Coleman K. Nakamura and J. F. Chlebowski J. Eiol. Chem. 1983 258 386. 112 P. Gettins and J. E. Coleman J. Biol. Chem. 1983 258 396. 113 R. A. Bradshaw F. Cancedda L. H. Ericsson P. A. Neuman S. P. Piccoli M. J. Schlesinger K. Schriefer and K. A. Walsh Roc. Natl. Acad. Sci. USA 1981,78 3413. 114 A. D. Hall and A. Williams Biochemistry 1986 25 4784. 326 D. Gani Of related interest Dunaway-Mariano has investigated the regiospecificity and stereospecificity of proton-transfer in the yeast inorganic pyrophosphatase-catalysed rea~tion.''~ The enzyme was shown only to catalyse the hydrolysis of Rp enantiomers of substrate complexes [ e.g.Co(NH3)4PPS]. The reported results are accommodated by a reaction mechanism involving enzyme-mediated proton transfer to the pro-R 0-atom of the incipient phosphoryl leaving-group of the P',p-bidentate Mg(H,O),PP-complex Scheme 22. -+ Em-? H I / H-&Enz Pro-R Scheme 22 Frey'16 has shown that the transfer of the terminal thiophosphate group of chirally labelled [ y-'702180]ATPS in the mevalonate-5-diphosphate decarboxylase reaction proceeds with overall inversion of configuration of phosphorus Scheme 23. Culp et a!."' have shown that the active-site residue of bovine intestinal 5'-nucleotide phosphodiesterase is threonine.' l7 This is the first reported example of the involvement of threonine in covalent phosphoryl enzyme intermediates.Potter' l8 has shown the enzyme-catalysed reaction proceeds with retention of configuration at phosphorus as is expected for double-transfer reactions. The crystal structure of RNase A complexed with d(pA) has been examined by McPherson et a/.at 2.5 8,re~olution."~ The analysis reveals many important interac- tions and explains why the protein can cover or protect 11-12 base segments within long strands of nucleic acid. Over the recent past many groups have turned their attention to the mode of action of E. coli DNA polymerase I. The enzyme has five activities and catalyses polymerization pyrophosphorolysis pyrophosphate exchange 3' +5' exonu-cleolytic degradation and 5' .-+ 3' exonucleolytic degradation. Polymerization pyrophosphorolysis and PPi exchange are associated with the polymerase activity and indeed pyrophosphorolysis and PP exchange are reverse reactions.Limited proteolysis of DNA Polymerase I (Pol I) results in cleavage of the protein into a 115 I. Lin W. B. Knight A. Hsueh and D. Dunaway-Manano Biochemistry 1986 25 4688. 116 P. Iyengar E. Cardemil and P. A. Frey Biochemistry 1986 25 4693. 117 J. S. Culp H. J. Blytt M. Itermodson and L. G. Butler J. Biol. Chem. 1985 260 8320. J. E. Cummins and B. V. L. Potter Biochem. SOC.Trans. 1986 14 1289. 119 A. McPherson G. D. Brayer and R. D. Morrison J. Mol. BioL 1986 189 305. Enzyme Chemistry m N 328 D. Gani large (Klenow) fragment and a small fragment. The Polymerase and 3’ -+ 5’ exonu-clease activities are associated with the Klenow fragment (KF) while 5’ --* 3’ exonu- clease activity is displayed by the small fragment.Only the mechanistic properties of the Klenow fragment will concern us here. Polymerization proceeds in the 5’ -+ 3’ direction and requires a template strand a 3’-hydroxy primer terminus and 2’-deoxynucleoside 5’-triphosphates. Synthesis of the chains antiparallel to the template is a direct consequence of the initial antiparallel orientation of the primer strand to the template Scheme 24. P/ Template DNA Scheme 24 The 3’ -+ 5’ exonuclease activity is thought to edit-out mismatched base-pairs at the primer terminus before further polymerization (see Kornberg’” for further background). The amino-acid sequence of the protein has been reported12‘ and the X-ray crystal structure of the complex with thymidine monophosphate has been obtained at 3.3 8 resolution.’22 Brody and Fre~’~~ have shown that the polymerase reaction occurs with overall inversion of configuration at the a-P-atom of the incoming dNTP while Benk~vic’~~ has shown that the 3’ -+ 5’ exonuclease activity (hydrolysis of the phosphate diester) also proceeds with overall inversion.Thus both processes probably involve direct one-step phosphoryl transfers. Benkovic has also investigated the mechanism of activities associated with poly- merization. Using the Klenow fragment it was shown that dTTP and corresponding a-thiotriphosphate (dTTPa S) were incorporated at equal initial rates during tem- 120 A.Kornberg ‘DNA Replication’ Freeman San Francisco 1980. 121 C. M. Joyce W. S. Kelley and N. D. F. Grindley J. Biol. Chem. 1982 257 1958. 122 D. L. Ollis P. Brick R. Hamlin N. G. Xuong and T. A. Steitz Nature (London) 1985 313 762. 123 R. S. Brody and P. A. Frey Biochemistry 1981 20 1245. 124 A. P. Gupta and S. J. Benkovic Biochemistry 1984 23 5874. Enzyme Chemistry 329 plate-directed synthesis thus indicating that a chemical step was not rate-1imiti11g.l~~ Further positional isotope exchange experiments showed that no label from [u-~*O~] dATP (labelled in the bridging position) entered a non-bridging position at the &phosphorus atom at dATP during template-directed reaction catalysed by Pol I indicating that PPi release rapidly follows the chemical step.The stereochemical course of PP exchange was found to occur with overall retention of configuration. Thus PPi attack on the elongated primer must occur with inversion of configuration since polymerization is known to proceed with inversion vide supra. Several groups have examined the mechanism of the idling-turnover reaction described below.'26-128 In previous studies conversion of a fraction of the available dNTP pool into a corresponding dNMP pool indicated that both the polymerase and 3' +5' exonuclease activities were expressed during the course of DNA syn- thesis. It was suggested that the extent of action of the 3'-5' exonuclease in producing nucleoside monophosphate could reflect the degree of proof-reading accompanying replication.In the absence of the correct following dNRP the rate of nucleoside monophosphate appearance was enhanced and the enzyme was forced to 'idle' at the primer terminus until the dNTP pool was depleted. Experiments conducted in this idling-mode have allowed the evaluation of the base misinsertion frequency12' and ongoing mechanistic studies are providing detailed insight. Ben- kovic has shown that when the idling reaction is conducted in a pool of [3H]dATP correctly template-matched 32P-deoxyadenosine is excised from the primer terminus and is replaced with tritiated material.'26 This result suggests that a mode of excision/incorporation rather than misincorporation/excision operates Scheme 25. 5'-G*A* A 3'-C?TAA Pathway A dAM P" dA'TP Pathway B dA'MP ' = 3H P * ~ 32 -G*A*A -CITAA -C'ITAA Scheme 25 The 3-[32P]end-label from the excision was traced to two products the expected exonuclease hydrolysis product [32P]dAMP and also [32P]dATP.The formation of the [32P]dATP was shown to occur via the attack of PPi on the terminal phos- phodiester linkage of the primer strand by PPi derived from the nucleoside triphos- phate pool. The pyrophosphorolysis reaction which occurs at extremely low PPi concentrations (0.1 pM),and the 3' --* 5' exonucleolytic degradation of DNA operate at similar rates. Benkovic has also shown that the DNA substrate does not dissociate from the enzyme during the switch in activity from exonuclease -P polymerase demanded by 125 V. Mizrahi N. Henrie J.F. Marlier K. A. Johnson and S. J. Benkovic Biochemistry 1985 24 4010. '26 V. Mizrahi P. A. Benkovic and S. J. Benkovic Proc. Nail. Acad. &I. USA 1986 83 231. 330 D. Gani excision/incorporation reaction in the idling-turnover mode.'27 Also the rate of pyrophosphorolysis was found to depend significantly upon the DNA sequence within the duplex region upstream of the primer-template junction. Evidence for the misincorporation/excision during idling-turnover was also presented. Mildvan has studied the conformation and interactions of substrate and ribonu- cleotide templates bound to the Klenow fragment using n.0.e. technique^.'^^ Finally Papanicolaou et al. have compared Pol I to the Klenow fragment and report that the differences are more profound than thought previously.Apparently the 3'-5' exo/pol activity ratios and hence error rates are subject to multiple influences which could cause mutational hot-spots in vivo whenever processive replication is inter- rupted.13' 10 Other Enzymes The application of site-directed mutagenesis to the study of the mechanism of tyrosyl tRNA synthetase was discussed in some detail last year.' Fersht and co-workers have continued to apply these techniques as is described in several recent publica- tion~.'~~-' Finally Baldwin and co-w~rkers'~~'~~ and others14' have studied the properties of isopenicillin N synthetase in some detail. Specifically the stereospecificity of carbon-sulphur bond'36 and p-lactam ring formation'37 has been determined as well as the structure reactivity profiles for ~nsaturated'~~ substrates.and alleni~'~~ Demain and co-w~rkers'~~ have now detected enzymic activity in cell-free extracts of Cephalosporium acremonium which catalyses the formation of 8-(L-W aminoadipy1)-L-cysteine the first intermediate in penicillin and cephalosporin bio- synthesis. The isolated yields are similar to those reported by Ba1d~in.l~~ The biosynthesis of p-lactam antibiotics has been reviewed re~ent1y.l~~ 127 V. Mizrahi P. A. Benkovic and S. J. Benkovic hoc. Natf. Acad. Sci. USA 1986 83 5769. lZ8 A. R. Fersht J. P. Shi and W. C. Tsui J. Mol. Biof. 1983 165 655. 129 L. J. Ferrin and A. S. Mildvan Biochemistry 1986 25 5131. I30 C. Papanicolaou P. Lecomte and J. Ninio J. Mol.Biof. 1986 189 435. 131 T. N. C. Wells and A. R. Fersht Biochemistry 1986 25 1881. 132 M. D. Jones D. M. Lowe T. Borgford and A. R. Fersht. Biochemistry 1986 25 1887. 133 C. K. Ho and A. R. Fersht Biochemistry 1986 25 1891. 134 A. R. Fersht R. J. Leatherbarrow and T. N. C. Wells Nature (London) 1986 322 284. 135 T. N. C. Wells C. K. Ho and A. R. Fersht Biochemistry 1986 25 6603. 136 J. E. Baldwin R. M. Adlington B. P. Domayne-Hayman H. Ting and N. J. Turner J. Chem. SOC. Chem. Commun. 1986 110. 137 J. E. Baldwin R. M. Adlington N. G. Robinson and H. Ting J. Chem. SOC. Chem. Cornmun. 1986,409. 138 J. E. Baldwin R. M. Adlington A. Basak S. L. Flitsch A. K. Forrest and H. Ting J. Chem. Soc. Chem. Cornmun.,1986 273. 139 J.E. Baldwin R.M. Adlington A. Basak and H. Ting J. Chem. SOC. Chem. Commun. 1986 1280. 140 J. M. Castro P. Liras J. Cortes and J. F. Martin FEMS Microbiol. Lett. 1986 34 349. 141 G. Banko S. Wolfe and A. L. Demain Biochem. Biophys. Res. Commun. 1986 137 528. 142 R. M. Adlington J. E. Baldwin M. Lopez-Nieto J. A. Murphy and N. Patel Biochem. J. 1983,213,573. 143 J. A. Robinson and D. Gani Nut. Prod. Rep. 1985 2 293.