14 Enzyme Chemistry By C. A. ROSS Department of Biochemistry University College Cork Lee Makings Prospect Row Cork Ireland 1 Introduction New texts devoted exclusively or almost so to enzymes and enzyme chemistry have continued to appear during the past year particularly from the publishing house of Springer-Verlag. Their first’ is devoted to the techniques of enzyme preparation while the second,* sub-titled ‘A Chemical Approach to Enzyme Action’ deals with enzyme mechanisms and has been described by one reviewer as ‘an exciting book’. The third3 is an account of the 32nd Colloquium held in Mosbach and is devoted to the structural and functional aspects of enzyme catalysis. The fourth4 is the subject of a separate section covalent catalysis of this Report.From the pen of the distinguished enzymologist G. G. Hammes has come a book,’ based upon his lecture course and review articles which follows a now familiar format. Following a discussion of protein structure enzyme kinetics is dealt with before a review of selected enzymes and the book concludes with chapters on regulatory enzymes multi-enzyme complexes and membrane-bound enzymes. Two other books with identical titles6 follow very much this same pattern. In the case of the book by Price and Stevens,60 the authors’ stated aim is to place enzyme studies in the context of the cell whereas Royer6’ wishes to stress the application of enzymes. In the event we are faced with yet more texts which differ only in detail from those which have appeared over the past few years.An unusual aspect of enzymology namely the effect of pressure has recently received attention. Thirty years after a review of the phenomenon by Laidler,’ a comprehensive theoretical treatment has now been published’ which is concerned with high-pressure effects such as might be encountered in deep sea exploration. It is conventional when discussing in general terms the control of enzyme-catalysed reactions to mention the concentration of enzyme available. Certainly in eukaryotes little work has been done on this important parameter which results from the ‘ R. K. Scopes ‘Protein Purification’ Springer-Verlag Berlin 1982. H. Dugas and C. Penney ‘Bioorganic Chemistry’ Springer-Verlag Berlin 1981. ‘Structural and Functional Aspects of Enzyme Catalysis’ ed.H. Eggerer and R. Huber Springer-Verlag Berlin 1982. L. B. Spector ‘Covalent Catalysis by Enzymes’ Springer-Verlag Berlin 1982. ’ G. G. Hammes ‘Enzyme Catalysis and Regulation’ Academic Press New York 1982. ‘(a)N. C. Price and L. Stevens ‘Fundamentals of Enzymology’ O.U.P. Oxford 1982; (b)G. P. Royer ‘Fundamentals of Enzymology’ John Wiley and Sons New York 1982. ’K. J. Laidler Arch. Biochern. 1951,30 226. E. Morild in ‘Advances in Protein Chemistry Vol. 34’,ed. C. B. Anfinsen J. T. Edsall and F. M. Richards Academic Press New York 1981. 322 C. A. Ross opposing rates of enzyme synthesis and of degradation. Now a fairly extensive study has been reported' on the turnover profile of one enzyme lactate dehy- drogenase in a variety of animal tissues and under several physiological perturba- tions.As the chosen enzyme exists in five isoenzymic forms whose proportions are tissue specific the investigation has produced a considerable amount of data. Appearing too late for inclusion in last year's Report came a monograph" from Cold Spring Harbor which promises to be for the time being at least the definitive work on the nucleases. Also in last year's Report was included reference to alcohol dehydrogenase. News has now been received'' of the death at the age of 79 of Hugo Theorell whose name will always be associated with that enzyme. 2 Covalent Catalysis Enzyme catalysis has long been assumed to occur via either a single-displacement or a double-displacement mechanism. Spector4 has now put forward a theory that all enzymes obey a double-displacement mechanism by forming a covalent inter- mediate with the substrate or with some fragment of the substrate.This is in contrast with the conclusion reached by Bell and Koshland'* some ten years ago when they stated that covalent intermediates were not essential for enzyme action. Spector opens his case with a consideration of the nature of enzymic catalysis compared with homogeneous and heterogeneous catalysis. The latter has tradi- tionally not been considered to be highly appropriate to biological systems but Spector points out that physisorbed and chemisorbed states of heterogenous cataly- sis have their counterparts in the Michaelis complex and the covalent enzyme- substrate intermediate of enzymic catalysis.Furthermore the migration of groups on a solid surface in order to react with other groups has its counterpart in those enzymes which require the migration ('surface walk') of covalently fixed fragments of substrate. Thus a parallel may be drawn between the hydrogenation of ethylene on a metal surface and the transfer of electrons from well separated donor to acceptor sites on the surface of a redox enzyme. Larger fragments are also capable of migration and Spector cites the example of so-called oxidative decarboxyl- ation (Scheme 1) where multienzyme complexes exhibit a 'triple-displacement' mechanism. The existence of a covalent enzyme-substrate intermediate was first demonstrated by Doudoroff et af.13in 1947 in the enzymic phosphorolysis of sucrose.Isotope- a-glucosyl-fructose + Pi $ a-glucosyl-1-P + fructose (1) exchange studies proved that a glycosylated enzyme was formed in the course of the reaction (Scheme 2). That the reaction proceeded by a double-displacement mechanism might have been anticipated since there is a net retention of configur- ation about C-1 of the glucose moiety following two Walden inversions. This might be thought to preclude covalent catalysis where an inversion of configuration occurs C. J. Masters Int. J. Biochem. 1982,14 685. lo 'Nucleases' ed. S. M. Linn Cold Spring Harbor Laboratory New York 1982. l1 B. Chance and B. L. Vallee Trends Biochem. Sci. 1983,8,45. l2 R. M. Bell and D. E. Koshland Science 1971 192 1253. l3 M. Doudoroff H.A. Barker and W. Z. Hassid J. Biol. Chem.. 1947 168,725. Emyme Chemistry + H+ \ (Reproduced by permission from ‘Biochemistry’ by D. E. Metzler J. Wiley & Sons New York,1977) Scheme 1 a-glucosyl-enzyme E a-glucosyi-I-P fructose fructose /3-gluWsyl-E Scheme 2 (such as would result from a single displacement reaction). Spector however cites the example of the adenine phosphoribosyltransferasereaction (Scheme 3) in which there is an inversion of configuration about C-1 of the 5’-phosphoribosyl group 324 C. A. Ross and yet a phosphoribosyl-enzyme intermediate has been isolated. It therefore follows that a 'triple-displacement' reaction has occurred to account for both the inversion and the covalent enzyme-substrate complex formation.Biochemical reasoning has long been influenced by the tenets of organic reaction mechanisms occurring in solution. Spector argues that the concept of the enzyme merely providing a special surface on which the single-displacement reaction may proceed at an enhanced rate is not tenable for over 400 enzymes for which covalent catalysis has been proved. There is an entropic advantage to be had if the enzyme is concerned with only one substrate at a time and also an enthalpic advantage when it comes to forming a covalent bond between two atoms if one of the atoms is part of the enzyme itself. Whereas enzymes have for convenience been classified by reaction types into six main groups by the Enzyme Commission of the I.U.B. on reflection one may argue that all enzymes belong to group 2 -the transferases.The bulk of Spector's book is devoted to an impressive documentation group by group of some 465 enzymes for which covalent catalysis is said to have been proved. On the grounds that a quarter of all listed enzymes have thus been shown to act by covalent catalysis and that in no case has single-displacement reaction been unambiguously proved Spector claims that all enzymes will be found to act in this manner. 3 Kinetic Analysis The enzyme kineticist has been well plied with texts on his subject in recent years. The latest title comes from Professor Lam in a very poorly printed form14 which belies the value of the contents. The book is intended as a practical guide to the devising of kinetic experiments and interpreting the resulting data in investigations of enzyme mechanisms.It has already been pointed out that many enzyme mechan- isms do not conform to the Michaelis mode1I5 and so this book deals largely with non-linear mechanisms a topic which is made relatively easy by the use of com-puters. Following a conventional opening chapter on the generation of steady-state rate equations a discussion of detailed balance and constraint equations follows. The law of micro-reversibility which states that the product of rate constants in one direction around a loop must equal the product of those in the opposite direction has to be taken into account in all but the simplest linear models. Chapters 3 and 4 are devoted to the estimation of rate and kinetic constants and a systematic approach to formulating kinetic mechanisms.The book concludes with an account of time course studies based on differential equations derived from mass-balance principles as in fast reaction methods and those based on integrated steady-state equations from progress curve methods. Another plot has been presentedI6 for the determination of initial rates from progress curves in which AP/t the chord from l4 C. F. Lam 'Techniques for the Analysis and Modelling of Enzyme Kinetic Mechanisms' Research Studies Press John Wiley and Sons Ltd. Chichester 1981. W. G. Bandsley P. Leff J. Kavanagh and R. D. Waight Biochem. J. 1980,189,739. '' E. A. Boeker Biochem. J. 1982 203 117. Enzyme Chemistry Po to to P t on a product-versus-time plot is plotted against product formed.The integrated rate equation for uncatalysed first-order reactions is In[1 -AP/(P -Po)] = -kl(l + l/Ke)r (2) = -(dP/dt,)r/(P -Po) (3) when rewritten to show the dependence on the initial rate dP/dt (the subscripts 0 and e indicate initial and equilibrium concentrations). If the left-hand side of equation (2) is approximated by AP/[-(Pe -Po) + AP/2] which has been shown to hold for up to 50% completion of reaction equation (3) can be rewritten AP/t = (dP/dto)[l -AP/2(P -Po)] (4) This is the equation of a straight line with an intercept dP/dto equal to the initial velocity. Similarly the integrated Michaelis-Menten equation may be closely approximated” by AP/t = Km(1 + Km/So) -APVmKm/2(Km + (5) Plots of AP/t against product formed result in curves approximating very closely to straight lines with an intercept at Po of dP/dt, the initial rate.Discrepancies between true initial rates and dP/dto are shown to be less than 1%for the most common form of integrated rate equation. The usual form of the integrated Michaelis-Menten equation expressed in terms of substrate concentrations is Vmax.r= (So -S) + Km.In(3 The time taken for an enzyme-catalysed reaction to achieve half-completion is obtained by substituting the relation S = S0/2 when t = t1/2into equation (6) Vmax*tl/2 = KmIn 2 + S0/2 (7) t1/2 = (Km/ Vmax) In 2 + S/2 Vmax (8) While the substrate concentration at the commencement of a reaction is the initial substrate concentration local substrate concentration refers to the substrate con- centration at any time t.Thus a plot of tlIZ against local So yields a straight line of slope $Vmaxand intercept (&,/ Vmax)In 2. The half-time plot is constructed by taking local So values at various stages of the reaction and calculating each corre- sponding value.’’ The authors have also investigated the performance of the direct linear form of the half-time plot. Equation 6 may be writtenlg in the form into which may be substituted the half-time relationship Vmax = S0/2t1/2 + (In 2Km)/t1/2 (10) Thus if local S0/2 values are plotted on the abscissa and S0/2f1/2 (which represents the slope of the chord joining a point at So to a point at S0/2 on the progress curve) ” (a)H. Goldenberg Arch.Biochem. Biophys.,1954,52 288; (b)M. R. J. Morgan Enzymologia 1972 42 219. ’* C.W. Wharton and R. J. Szawelski Biochem. J. 1982,203,351. ” R. Eisenthal and A. Cornish-Bowden Biochem. J. 1974 139,715. 326 C.A. Ross on the ordinate the intercept in the first quadrant of the plot is given by V,, In 2K (Figure 1).Data from progress curves of several enzyme-catalysed reactions have been analysed by both these plots and excellent results obtained provided certain conditions were adhered to. local IS,] -BJ/2 In 2 K Figure 1 (a) Half-time plot; (b) Direct linear half-time plot A method for determining kinetic parameters at high enzyme concentrations has been described.’’ Using the relationship [ES] = i{Et + St + K -J(Et + St + K,)’ -4E& } (11) which arises if it is not assumed that free substrate concentration is equal to total substrate concentration and by writing v/V,, for [ES]/Et it can be shown that ’(’ C.J. Halfman and F. Marcus Biochem. J.. 1982,203,339. Enzyme Chemistry Equation (12) describes the relationship between velocity and total substrate enzyme ratio as a function of enzyme concentration relative to K,. When data is plotted in these terms (Figure 2) free and bound substrate concentrations may be determined from and V Figure 2 Initial velocity curves predicted when Eo > K and enzyme concentration (b) > (c). The broken curve results when E >> K and represents the stoicheiometric relationship between u and St/E [equation (12)] The many corresponding values of o [S] and [ES]/Et which are obtained from velocity measurements at several ligand enzyme ratios and at two or more enzyme concentrations may then be analysed by conventional means.Halfman and Marcus state that the method is applicable to non-linear kinetics and they also apply it to the inhibition of fructose 1,6-bisphosphatase by the tightly-binding AMP. It has long been maintained that it is not possible to evaluate the rate constants k+l and k- by steady-state kinetic measurements for the simple Michaelis-Menten mechanism for one-substrate enzyme reactions E + S & ES % E + products k-1 328 C.A. Ross Now Wong and his collaborators2’ propose a means whereby these constants may be measured by determining the effect of temperature on Vmaxand K utilizing the constraint on the rate constants of the Arrhenius equation The procedure does require the measurement of isotopic rate effects in order to distinguish between k+l and k-l.Furthermore it is essential that a linear Arrhenius plot of log V,, versus 1/T obtains and that K varies markedly with tem-perature. Given that practical limitations may prevent the universal application of the method the authors nevertheless claim the principle disproves the contention that steady-state kinetic measurements are incapable of determining rate constants in this the simplest of enzyme mechanisms. Following his treatment of progress curves to obtain initial rate measurements as recorded in these Reports last year Waley2* has now produced a quick method for the determination of inhibition constants by the comparison of progress curves recorded in the presence and absence of an inhibitor.From the integrated form of the Michaelis-Menten equation for the case of a competitive inhibitor it can be shown that where (t -t,) is the time difference taken in the inhibited and uninhibited reactions for the initial substrate concentration So to fall to a chosen concentration S. When (t -t,) is plotted uersus In (So/S) a straight line results of slope K,/V,, * [I]/Ki from which Kimay be determined (Figure 3). The procedures for product inhibition and mixed inhibition are also dealt with. t-t (b) [PI Time Figure 3 (a) Progress curves in the absence and presence of inhibitor (I); (b) Time differences from (a) plotted against In (So/S) 21 S.Lin K. Chou and J. T. Wong Biochem. J. 1982 207 179. 22 S. G. Waley Biochem. J. 1982,205 631. Enzyme Chemistry 329 4 Regulatory Molecules Biochemists continue to identify new and surprising compounds which often have far-reaching properties in regulating enzyme action where it might have been supposed there was nothing new to discover. Possibly the best known example is that of adenosine 3',5'-monophosphate (cyclic AMP CAMP) which was first repor- ted by Sutherland in 1958 and which is now recognized to have a widespread role in mediating cellular response to external hormonal stimuli. Sites on the plasma membrane receptive to hormone molecules bring about the activation of the enzyme adenyl cyclase which causes the cyclization of ATP to cAMP which then acts as a cellular hormone or so-called 'second messenger'.The resulting effect is often the phosphorylation of a catalytic protein. The chemistry of cAMP and analogous cyclic nucleotides has been extensively reviewed;23 recently an interesting study has been published by van 001 and These authors have performed quantum chemical calculations on the formation of intermediates with trigonal-bipyramidal (TBP) configurations in the hydrolysis of cAMP with phosphodiesterases and in the activation of protein kinase by CAMP. In the case of hydrolysis of the analogue adenosine 3',5'-[thio]monophosphate (cAMP[S]) it is shown that the involvement of diequatorial ring-positioned intermediates would always result in the apical location of sulphur (la-d).The energy difference between the resulting diastereomers is calculated to be over 500 kJ mol-' and they could not therefore be hydrolysed at similar rates. The energy difference between the intermediates with an equatorial-apical cyclic phosphate ring (le) (If) is only 53 kJ mol-' and TBP configurations diequatorial ring position equatorial-apical ring position OMe (la) X = S,Y = 0 (lb) X = 0,Y = S HOKAd from endo-attack XH -0,OMe U-b0\ I '\ P-0 EWAd HS'b OH 0 (lc) x = 0,Y = s (Id) X = S,Y = 0 &Ad from exo-attack HO (If) Sp isomer of cAMP S gives rise to (la) (Id) (le) Rp isomer gives rise to (lb) (lc) (If) 23 'Cyclic 3',5'-Nucleotides Mechanisms of Action' ed.H. Cramer and J. Schultz John Wiley and Sons London 1977. 24 P. J. J. M. Van 001and H. M.Buck Eur. J. Biochem. 1982 121,329. 330 C.A. Ross these would therefore be hydrolysed at similar rates. The activation of protein kinases is assumed to proceed via diequatorial-ring-positioned TBP intermediates. The enzyme-nucleotide covalent intermediate is said to bring about the conforma- tional changes and dissociation of the catalytic and regulatory subunits which are necessary for activation of the enzyme. In glycolysis the metabolic pathway leading to the formation of pyruvate from glucose the most important step from the point of view of the control of the process is the conversion of fructose 6-phosphate to fructose-1,6-bisphosphate (Fru-1,6-P,) by phosphofructokinase (PFK or PFK 1).The bisphosphate was first discovered by Harden and Young in 1909 and the ester named after them until its chemical structure was identified by Levene and Raymond some twenty years later.Now over half a century later another fructose bisphosphate Fru-2,6-P2 (2) has been di~covered,~~~~~ which while not being an intermediate of glycolysis is a powerful regulator of PFK. This enzyme which was discussed in these Reports in 0 II O--P-O-CH20-I H Y o H 0 II 0-P-0-F 6- CHZOH OH H (2) 1980 continues to be the subject of active inve~tigation.~~" The discovery of a new allosteric effector will require a major revision of the roles played by the many controlling factors to which this enzyme responds.The identification of Fru-2,6-P2 has been and was aided by its lability to mild acid conditions. This property has been exploited in a specific method for the measurement of the ester28 in which after the removal of endogenous hexose-6-phosphate the extract is subjected to mild acid hydrolysis (pH2 for ten minutes at room temperature) when more than 95 O/O hydrolysis to fructose-6-phosphate will have occurred as compared with less than 1% with Fru-1,6-P,. The resulting Fru-6-P is then assayed in a coupled enzymic assay with bacterial NADH-dependent l~ciferase.,~ With the discovery of Fru-2,6-P2 has come identification of the enzymes respon- sible for its formation (6-phosphofructo-2-kinase PFK2) and its hydrolysis (fruc- tose-2,6-bisphosphatase FBPase 2).Just as PFK 1 is stimulated by AMP and inorganic phosphate and inhibited by phosphoenol-pyruvate and citrate so too is PFK 2. The K for Fru-6-P is 50 pM which is within the physiological range found in liver and so substrate concentration contributes to the control of biosynthesis of Fru-2,6-P2. On the other hand unlike the effects upon FBPase 1 FBPase 2 is " E. Van Schaftingen L. Hue and H.-G. Hers Biochem. J. 1980 192,897. 26 H.-G. Hers and E. Van Schaftingen Biochem. J. 1982,206 1. 27 (a) A. Sols J. G. Castano J. J. Aragon C. Domenech P. A. Lazo and A. Nieto in 'Metabolic Interconversion of Enzymes' ed. H. Holzer Springer-Verlag Berlin 1980; (b) H.-G. Hers E. Van Schaftingen and L. Hue ibid. '' L. Hue P. F. Blackmore H.Shikama A. Robinson-Steiner and J. H. Exton I. Bid. Chem. 1982 251,4308. 29 S. Golden and J. Katz Biochem. J. 1980 188 799. Enzyme Chemistry not dependent upon Mg2+ nor inhibited by AMP but is strongly inhibited by the product of the reaction Fru-6-P. The effect of glucagon a hormone which is known to stimulate glycogenolysis 'by the activation of adenyl cyclase and hence the activation of protein kinases by CAMP also causes the phosphorylation of PFK2 and of FBPase 2 with the consequent disappearance in Fru-2,6-P2 (Scheme 4). The Glucpp I I I t cyclic AMP I I I + ~ --Protein kinase ---. // \ 0 \ 0 \ / \ / \ / / \ (PEP = phosphoenolpyruvate; P-glycerol = glycerol phosphate) Scheme 4 interesting aspect of the control of Fru-2,6-P2 concentrations is that the activation- deactivation of PFK2 by dephosphorylation-phosphorylation operates in the opposite manner to that for phosphorylase and in similar manner to that for glycogen synthase.Thus CAMP prevents the formation and favours the destruction of Fru-2,6-P2. Evidence has been produced3' for a distinct mechanism of control mediated by Ca2' and calmodulin (see Scheme 1 in these Reports for 1980).The role of Fru-2,6-P2 would appear to be one of maintaining glucose levels rather than mimicking the action of phosphorylase. In this way glycolysis and glycogenoly- sis may be co-ordinated and not merely be similar routes with alternative starting points leading to the production of pyruvate (Scheme 5).Twenty five years have elapsed since Isaacs and Lindenmann first reported the isolation of a protein fraction from virus-infected animal cells which conferred resistance to viral attack in unaffected cells. The nature of the resistance appeared to be an interference with the normal development of the virus within the host cell and so they named the protein fraction interferon. Today the study of interferons '(' C.S.Richards and K. Uyeda J. Biol. Chem. 1982,257,8854. 332 C.A. Ross (bold arrows represent metabolic reactions broken arrows indicate regulatory effects) Scheme 5 has become one of the most active areas in biochemistry mainly due to the wide ranging implications for the control of viral infections and disease. In 1981 a new journal the Journal of Interferon Research commenced.There is a continuing series entitled Interferon3la which has already run to three volumes and two recent volumes in Methods in Enzymology31b have been devoted to this subject. During the past year 500 publications on interferon research have appeared together with a number of review Interferons (IFNs) are a family of proteins found in vertebrates and are divided into three antigenically distinct classes a (at least eight types) p (at least two types) and an unknown number of yIFNS. The first two classes may be induced in a variety of cells by certain viruses bacteria or double-stranded RNA (ds RNA) whereas y-IFNs are induced in lymphoid cells by mitogens and antigens to which the cells have been sensitized.It has been estimated that the genes specifying a-and @-IFNs are as old as vertebrates themselves. Originally interferons were produced in cell lines leucocytes for (Y -1FNs and fibroblasts for p-IFNs. Recently 31 (a)'Interferon' ed. I. Gresser Academic Press London and New York 1979 1980 1981 Vols. 1-3; (b) 'Methods in Enzymology Interferons Part A' ed. S. Pestka Academic Press London and New York 1981 Vol. 78; 'Part B' ibid. 1982 Vol. 79. 32 P. Lengyel Ann. Reu. Biochern. 1982,51,251. Enzyme Chemistry however using recombinant DNA technology human IFNgenes have been isolated and inserted into E. coli and shown to specify biologically active human IFNs. The protein sequences of interferons are being elucidated. Owing to the initial scarcity of material micro-sequencing proceeded slowly but now amino-acid sequences are being predicted from nucleotide sequences of cloned IFW cDNAs.~~ Interferon affects viral infection of cells by two distinct mechanisms both affecting the protein- synthesizing machinery.In one the enzyme catalysing the unusual reaction (n + 1)ATP + (2’-5’)-pppA(pA) + n pyrophosphate (18) (where n can be between 1 and 15) is stimulated in the presence of ds RNA (of at least 30 base pairs and maximally of 65 base pairs or more). The enzyme has been isolated and purified from various sources. The only well-established function of (2‘-5p)(A) is the activation of a latent endoribonuclease RNase L which cleaves only single-stranded regions of ribonucleic acid such as mRNA.Whereas in in uitro experiments RNase L is not specific in the cell it does appear to act preferentially on viral RNA. The oligonucleotide is resistant to many nucleases by virtue of its unusual (2’-5’)-phosphodiester linkage but it is degraded by a specific (2’- 5’) -phosphodiesterase. The other mechanism of interference by interferon in protein synthesis is by the phosphorylation (and hence inactivation) of an initiation factor eIF-2. The action of interferon is to bring about the activation of a protein kinase in the presence of double stranded RNA (Scheme 6). Because of the central role played by cal- mod~lin~~ in regulating cellular processes already referred to in this Section it is Interferon 1 Induction (2’-5’)synthetase Protein kinase ATP (2‘-5’)A elF-2 elF-2 phosphorylated unphosphorylated 10 Endonuclease (inactive) (active) 1 1 RNA degradation No initiation of protein synthesis I/ INo protein synthesis I Scheme 6 33 T.Taniguchi N. Mantei M. Schwarzstein S. Nagata M. Muramatsu and C. Weissmann Nature (London),1980,285 547. 34 C. B. Klee and T. C. Vanaman in ‘Advances in Protein Chemistry’ ed. C. B. Anfinsen. J. T. Edsall and F. M. Richards Academic Press New York 1982 Vol. 35. 334 C.A.Ross not surprising to find that its involvement in interferon induction has been investi- gated.35 Numerous questions still remain to be answered such as why there is such a diversity among the interferons and how it is that widely differing substances are capable of inducing interferon synthesis.It is also not known if interferon penetrates into the receptor cell or if not what is the nature of the ‘secondary messenger’ which relays the signal from the membrane receptor site. Secondary to this problem is the nature of the activation of two very dissimilar enzymes the (2’-5’)(A), synthetase and the protein kinase. 5 Conformational Changes It has now come to be widely recognized that many enzymes are composed of subunits and that practically all regulatory enzymes are multimeric and the question therefore must arise as to the functional significance of quaternary structure. Of the two original models proposed to account for co-operativity between sub-units that by Monod Wyman and Change~x~~ had as its central feature that the subunits were identical and that symmetry was conserved.As a result the model was unable to account for the phenomenon of negative co-operativity where clearly the subunits were behaving in a non-identical fashion. More recently Viratelle and Seydo~x~~ have modified the two-state model by introducing into one of the states two classes of binding sites with differing affinities for the same ligand. Negative co-operativity can now be accommodated but only at the expense of the basic criterion of the original model. On the other hand the Koshland Nemethy and Filmer3* model based on induced fit theory can readily account for negative co-operativity and can also be adapted to the situation where the subunits are initially non-identical i.e.pre-existing asymmetry. Since apparent co-operative effects may occur for other reasons than those described by the above models it has proved in practice extremely difficult to allocate many individual enzymes to one or the other model. There have been attempts to associate subunit interaction with the catalytic process. Thus the flip-flop model originally proposed by Lazdunski to account for extreme cases of negative co-operativity exhibiting Michaelian kinetics,39 envisages that binding sites act in pairs so that when one is occupied with substrate the other has a greatly diminished affinity. When a chemical event occurs at the first site the second site is reactivated to bind substrate with the consequent catalysis and release of product at the first site.Thus the two sites alternate as the catalytic site in a flip-flop cycle. The ‘alternate site’ model of Boyer4’ is basically the same but is more general not being confined to extreme negative co-operativity and can be adapted to the simultaneous binding of substrate at both sites. 35 D. Gurari-Rotman FEBS Lett. 1982,148 17. 36 J. Monod J. Wyman and J.-P. Changeux J. Mol. Biol. 1965 12 88. 37 0.M. Viratelle and F. J. Seydoux J. Mol. Biol. 1975,92 193. 38 D. E. Koshland G. Nkmethy and D. Filmer Biochemistry 1966 5 365. 39 M. Lazdunski Curr. Top. Cell. Regul. 1972,6,267. ‘’ P. D. Boyer M.Gresser C. Vinkler D. Hackney and G. Choate in ‘Structure and Function of Energy-Transducing Membranes’ ed. K. van Dam and B. F. van Gelder Elsevier Amsterdam 1977.Enzyme Chemistry 335 The whole area of subunit co-operation and enzymic catalysis has recently been reviewed.41 One enzyme in particular glyceraldehyde-3-phosphatedehydrogenase (GPDH) which catalyses reversibly the oxidative phosphorylation of D-glyceral- dehyde 3-phosphate to 1,3-bisphosphoglycerate (Scheme 7) is the subject of much GAP*&+ 1,3-BPG NAD' NADH GAP = glyceraldehyde 3-phosphate 1,3-BPG = 1,3-bisphosphoglycerate Scheme 7 investigation and debate concerning the role of subunit interaction in the catalytic mechanism. GPDH is a tetramer being composed of four chemically identical subunits each with a uniquely reactive sulphydryl group and each binding a molecule of coenzyme. However the GPDH's from different sources exhibit quite different properties and the yeast enzyme for example binds the coenzyme in a positively cooperative manner and conforms to the concerted model of Monod et al.42In contrast the binding of coenzyme by the enzymes isolated from various vertebrate muscles and from Bacillus stearothermophilus exhibits negative co-operativity although there is some disagreement among the reported dissociation constants.The differences between subunits in the affinity for NAD' and for NADH have been ascribed to two independent pairs of binding sites pre-existing in the tet~amer.~~ Similarly half-of-the-sites reactivity of the sulphydryl groups towards acylation has been interpreted in terms of a 'dimer of dimers' structure." Recently Cardon and B~yer~~ have re-examined the evidence for equivalent catalytic sites and found that tightly bound NAD' is preferentially reduced in the presence of glyceraldehyde 3-phosphate.They have proposed a scheme (Scheme 8) in which the enzyme has two interconvertible sites in an attempt to reconcile evidence for the equivalent participation of four catalytic In Scheme 8 the change in the binding of NAD' at one site from loose to tight not only promotes oxidation- reduction but may also promote phosphorolysis accompanying NAD' binding and NADH release at an alternate site. The authors readily admit that firm evidence for the 'alternate site' model is still required but they express the hope that subunit interactions contribute to catalysis and that many oligomeric enzymes will proceed by similar mechanisms.41 C. Y. Huang S. G. Rhee and P. B. Chock Annu. Rev. Biochem. 1982,51,935. 42 K. Kirschner E. Gallego I. Schuster and D. Goodall J. Mol. Biol. 1971 58 29. 43 (a)N. Kekernen N. Kellershohn and F. J. Seydoux Eur. J. Biochem. 1975,57,69;(6)N. Kellershohn and F. J. Seydoux Biochemistry 1979 18 2465. 44 0.P. Malhotra and S. A. Bernhard J. Biol. Chem. 1968,243 1243; Biochemistry 1981,20,5229. " J. W. Cardon and P. D. Boyer J. Biol. Chem. 1982 257,7615. 46 (a)D. R. Trentharn Biochem. J. 1971,122.59 71; (6) B. D. Peczon and H. 0.Spivey Biochemistry 1972 11,2209. 336 C. A. Ross RcHol 0 RC S /NA~H. E-&+ \NADH rapid redox rate limiting // release L* NAD+ f = high affinity site Scheme 8 A contribution towards techniques in detecting conformational changes in en- zymes has been presented by Christen and Gehri~~g.~’ They define two types of conformational change; that in which the change is induced by ligand binding and that in which change occurs as enzyme-substrate becomes enzyme-product.The latter has been termed syncatalytic change. The techniques discussed are the ‘differential chemical modification’ detection of the change in reactivity of side-chains to various labelling reagents as the protein passes through transitional configurations. By ‘differential chemical modification’ is meant the treatment of the protein in different states such as unliganded or complexed with substrates substrate analogues and inhibitors of various sorts.Such an approach to GPDH for example has been The case of aspartate aminotransferase is cited49” as an illustration of the value of the technique. Bromopyruvate a substrate analogue was found to inactivate the enzyme in the presence of the second substrate an amino-acid. However the presence of a second keto acid substrate did not protect the enzyme from inactivation by the haloacid which was subsequently found to be binding to a cysteine tesidue not in the active site. Hence it is concluded that bromopyruvate forms a covalent enzyme-ligand complex with consequent confor- mational change such that a sulphydryl side-chain is activated to react with 47 P. Christen and H. Gehring Methods Biochem. Anal. 1982 28 151. 48 L. D. Byers and D.E. Koshland Biochemistry 1975,14 3661. 49 (a) W. Birchmeier and P. Christen J. Biol. Chem. 1974 249 6311; Methods. Enzymol. 1977 46 41; (b)P. Christen M. Cogoli-Greuter M. J. Healy and D. G. E. Lubini Eur. J. Biochem. 1976 63 223; (c)D. G. E. Lubini and P. Christen,Proc. Natl. Acad. Sci. U.S.A. 1979,76 2527. Enzyme Chemistry bromopyruvate. The phenomenon of paracatalytic enzyme modification is also described in which a substrate becomes activated on complexing with enzyme and may then react with an extrinsic reagent. Such a situation has for example been described for the trapping by suitable oxidants of the intermediate in the fructose- 1,6-bisphosphate aldolase reaction (Scheme 9).49b*c The inactivation of the enzyme results from the crosslinking of two vital lysine residues by a substrate derivative.CH2W I c=o I H2N-E Fru-1,6-P2 DHAP ferrocyanide +2H' GAP +H20 CH2O I C=NH-E I -CHOH ferricyanide GAP = glyceraldehyde-3-phosphate1DHAP = dihydroxyacetone phosphate Scheme 9