E. J. Bourne, Ph.D., D.Sc., F.R.I.C., and P. Finch, BSc,, Ph.D. 45 46 49 52 .. 57 . . . . .. . . . . . . References . . . . . . . . .. . . . . . . . . 57 THE GLYCOSYL TRANSFER REACTION POLYSACCHARIDES-ENZYMIC SYNTHESIS AND DEGRADATION Dept of Chemistry, Royal Holloway College, Englefield Green, Surrey The glycosyl transfer reaction . . .. . . Polysaccharide biosynthesis . . . . . . . . . . . . . . . . * . * . Homopolysaccharides, 46 Branched and heteropolysaccharides, 48 Enzymic degradation of polysaccharides Mechanisms of action of glycosyl transfer enzymes . . Conclusions . . . . . . .. . . * . The wide occurrence and diversity of function of polysaccharides in nature is well documented,l as is the importance of this group of natural products in industry.2 These factors, coupled with the vast number of different structural examples so far encountered, are responsible for the considerable research effort which has been directed towards carbohydrate polymers.Perhaps the greatest problems to be overcome in any study of an enzyme- polysaccharide system involve purification of the two components. In addition to a variety of organic and inorganic contaminants of unrelated structure, each of the polymeric reactants is usually associated with other polymeric material of very similar structure. Ideally the reactants should be purified until constant physical and chemical properties (e.g. molecular weight, electrical charge, type of linkage, degree of branching) are achieved. During recent years greatly improved purifications have resulted from the wider application of procedures involving such techniques as ion-exchange chroma- tography, molecular-sieve chromatography, selective complex formation and electrophoresis.The successful elucidation of polysaccharide structures has led to con- sideration of the relations between chemical structure and physical and biological pr~pertiesla,~ and, in addition, has provided the basis for investiga- tions of the biological pathways of polysaccharide synthesis and degradation. This brief survey illustrates some general features of these pathways and indicates some topics of 'current interest by reference to examples selected from the many available. It is convenient to consider polysaccharide synthesis and degradation together for a number of reasons.The most important of these possibly is the fact that Bourne and Finch 45 both processes are examples of a single general reaction, termed the glycosyl transfer reaction : (1) G-OR + G-OR’ DONOR + H-OR’ ACCEPTOR The general applicability of this reaction to carbohydrate metabolism was first realized during the 1940s4 and, indeed, all subsequent observations may be discussed in terms of it. Thus, if the acceptor (H-OR’) is a polysaccharide chain the net result of the forward reaction is synthesis; if the acceptor is water (R’=H) the net result is hydrolysis; in each case the group transferred is the glycosyl group (G). POLYSACCHARIDE BIOSYNTHESIS One approach to a discussion of polysaccharide synthesis in terms of the transfer reaction is via an appraisal of the various possibilities for the nature of the glycosyl donor (substrate) G-OR, e.g.is G a single glycose residue or a number of residues previously linked by another process, and what possi- bilities exist for the structure of R? Clearly during synthesis the acceptor (H-OR’) in each successive step (except the first) is the growing poly- saccharide chain. Four main types of donor aglycone (OR) have been en- countered-phosphate, glycoses, nucleoside diphosphates and polyisoprenoid phosphates. 0 It (2) + HOR + GIC-O--(GIC), Homopolysaccharides The first examples of polysaccharide synthesis in vitro to be established con- clusively were the formation of a-l,4-linked amylose-type polymers by incuba- tion of a-D-glucose- 1 -phosphate with phosphorylases from muscle, yeast, liver, peas and potatoes (see ref.4 and refs therein): 0 ll Glc-0-P-0- + H-O-(Glc), I + HO-P-0- 0- A- The reactions are readily reversed by varying the relative proportions of inorganic phosphate and a-D-glucose- 1 -phosphate, and from these observa- tions the concept of transglycosylation was developed. It was later realized that the ratio of phosphate to a-D-glucose- 1 -phosphate in vivo is unfavourable to synthesis, and it is now postulated6 that in mammals the enzyme phos- phorylase is probably concerned with polysaccharide (glycogen) degradation only. However the stimulus provided by these observations resulted in the isolation of other cell-free extracts which would catalyse the incorporation of sugar units into polysaccharides.An early example was that of Hehre and Sugg,7 who demonstrated that an extract of the bacterium Leuconostoc mesenteroides catalysed the synthesis from sucrose (R = fructofuranosyl) of dextran, an a- 1,6-linked polyglucose. Further examples of polysaccharide synthesis by glycosyl transfer from donors containing carbohydrate ‘agly- cones’ are shown in Table 1. The last two examples in Table 1 refer to the action of enzymes which are R.I.C. Reviews 46 Table I. Glycosyl transfer from purely carbohydrate donors ~~ 10 Acceptor and linkage Dextran a- I ,6 Levan /3-2,6 Amylopectin a- I ,6 Glycogen 01- I ,6 Donor and linkage Sucrose a- I ,8-2 Sucrose a- I ,&2 Amylose a- I ,4 Glycogen a-1,4 ribofuranosyl or deoxyribofuranosyl Ref.7 8, 9 I 1 responsible for the synthesis of branch points in amylopectin and glycogen. Glycogen is generally held to possess a tree-like structure consisting of chains of a-174-linked glucose units connected by a-1,6-branch points.12 Glycogen branching enzyme catalyses the transfer of a chain of about six or seven glucose residues from an a-1,4-position to an a-l,(i-p~sition;~~ thus in this reaction glycogen is acting as both donor and acceptor. Amylopectin, the major energy reserve material in plants, is thought to possess a structure similar to that of glycogen, but one that is less highly branched and in fact glycogen branching enzyme can convert amylopectin into a glycogen-like molecule.14 The search for an alternative pathway to the glucose- 1 -phosphate/phos- phorylase system for the synthesis of the a-1,4-linked chains in glycogen was not rewarded until 1957, when Leloir and Cardin95 reported the isolation of an enzyme from liver tissue which catalysed the incorporation of glucose from uridine diphosphate glucose (UDPG) into glycogen.Uridine diphosphate glucose was the first example to be characterized16 of the general class of compounds known as the nucleoside diphosphate sugars. They possess the general formula : -5’-pyrophosphate- 1-sugar and act as glycosyl donors by cleavage of the linkage between the sugar anomeric carbon and the pyrophosphate oxygen, e.g.: UDPG + glycogen giycogen synthetase, G- I ,4-glycogen +- UDP Enzyme Dextransucrase Levansucrase Q-enzy me Branching enzyme purine or Bourne and Finch (3) In addition to their role as glycosyl donors nucleoside diphosphate sugars are important intermediates in metabolic interconversions of sugars, e.g. the formation of UDP-galactose from UDP-glucose and the formation of deoxy sugars.17 Some examples of polysaccharides believed to be synthesized by trans- glycosylation from nucleoside diphosphate sugars are given in Table 2. In the past few years evidence has been accumulating for the involvement, so far only in the synthesis of bacterial polysaccharides, of another class of sugar derivative in which the sugar aglycone (OR) takes the form of a poly- isoprenoid chain linked to the sugar through one26 or t ~ 0 ~ 7 + ~ * phosphate ester groups.The possible involvement of lipid intermediates in polysaccharide biosynthesis was perhaps first suggested by Colvin in 1961 in connection with the biosynthesis of cell~lose,~9 but it was not until 1965 that their importance was established by workers who were studying the cell-wall constituents of a number of Staphylococcus,30 Micrococcus3O and Salrn0nella~~9~1 species 47 4 -~ Ref. Rous chicken sarcoma Potato, pea, maize Acetobacter xylinum 19 20 Hyaluronic acid Starch Cellulose Mung beans 21 Table 2. Glycosyl transfer from nucleotide diphosphate sugars Source C, cytidine Donor UDP-N-acetyl-glucosamine ADP-gl u cose U DP-gl ucose G D P-g I ucose UDP-xylose TDP-rhamnose CDP-abequose A, adenosine Polysaccharide Xylan Lipopolysaccharide U, uridine Salmonella Hen oviduct typhimurium 1;: 22 G, guanosine T, thymidine Table 3.Proposed pathways of biosynthesis of some branched and heteropol ysaccharides Enzyme source Liver Potato Pathway iv (branching) iv (branching) iv (desu I phation) i , o r ii i , or ii, or iii and iv i, or ii Polysaccharide Glycogen Amylopectin Galactan Hyaluronic acid Chond roiti n Pneumococcus Type 111 capsule GI ycoprotei n Glycoprotein Lipopol ysaccharide Porphyra umbilicalis Staphylococcus hemolyticus Hen oviduct Pneumococcus Type 111 Colostrum Salmonella typhimurium Staphylococcus aureus ii iii iii 18 [24,25 Ref.35 10 36 37 38 39 40 25 32 of bacteria. Matsuhashi et al. have suggested32 that the formation of lipid intermediates may provide a means whereby intracellularly manufactured components can be transported through the hydrophobic cell membrane prior to synthesis of the outer wall. Branched and heteropolysaccharides Consideration of the biosynthesis of branched and heteropolysaccharides poses additional problems. Such polymers appear to be divisible into two types according to structure, namely those with regular, repeating structures and those with more random arrangements. A recent re-examination33 of Aerobacter aerogenes A3 (Sl) polysaccharide led to the formulation of a strict repeating tetrasaccharide unit, in conflict with earlier work, and it was suggested that other polysaccharides for which random structures have been proposed may merit re-examination.On the other hand, many polysaccharides definitely possess non-repeating structures, e.g. some dextrans are polymers of a-1 ,&linked glucose units having randomly arranged single unit branches.34 Four possible pathways may be envisaged for the formation of branched and heteropolysaccharides : ( i ) transfer of single units by a single enzyme having a number of specific binding sites; (ii) transfer of single units by a number of enzymes ; (iii) transfer of preformed oligosaccharides by one or more enzymes; and (iv) operation of transglycosylases, or of other enzymes, on a presynthesized chain of sugar units.Regular repeating structures would probably result from the operation of processes (i) and (iii) (where only one R.I.C. Reviews 48 enzyme is involved), while the regularity of polymers resulting from the other processes would depend on the specificities and possibly on the relative concentrations of the enzymes involved. All four types of system have been postulated; some examples are shown in Table 3. It is clear that this aspect of heteropolysaccharide biosynthesis is subject to considerable uncertainty, and that substantial developments can be expected in the near future by the use of the mild purification procedures now avail- able.A recent example is the isolation of Q-enzyme from potato juice by chromatography on DEAE-cellulose.41 A related aspect of polysaccharide biosynthesis is the question of direction of chain propagation, i.e. are units added to the reducing or to the non- reducing end of the growing polysaccharide chain? It has sometimes been assumed, by analogy with many other transglycosylation reactions, that the non-reducing chain terminus acts as the acceptor in polysaccharide biosyn- thesis, and this has been shown to be correct in the majority of cases which have been examined.42 However, application of the technique known as pulse labelling, in which incorporation of radioactively labelled substrate (the ‘pulse’) is followed by exposure of the synthetic system to a relatively large amount of unlabelled substrate (the ‘chase’), has revealed that in at least two i n ~ t a n c e s ~ ~ f ~ ~ polysaccharide elongation takes place other than at non-reducing chain ends.These observations are not exceptions to the gly- cosy1 transfer reaction; for example, the growing polysaccharide may be considered as the donor and the extra unit added as the acceptor in the elongation of SaZmoneZZa 0-antigen chains :42 M an-R ha-Gal-P-P-AC (Man-Rha-Gal),+r-P-P-ACL L -* (or P + P-ACL) (4) L + (Man-R ha-Gal),-P-P-AC + P-P-ACL In this example the bond cleaved is that between galactose (Gal) and phosphate (P) in the growing chain-antigen carrier lipid (ACL) complex. Gahan and Conrad43 succeeded in isolating a glycogen synthetase fraction from Aerobacter aerogenes which, although devoid of glycogen, retained the capacity to catalyse de novo glycogen synthesis from ADPG in the presence of an activator protein.The labelled glucose residues incorporated in the early stages of the reaction remained in external chains after an additional incorporation of five to six times as much glucose into the product. It appears that, at least in Aerobacter aerogenes, de novo glycogen synthesis may proceed via a different mechanism from the accepted glucosyl transferase-branching enzyme process. On storage the enzyme becomes more like a typical glycogen- dependent glycogen synthetase. ENZYMIC DEGRADATION OF POLYSACCHARIDES Enzymes which catalyse the hydrolysis of glycosidic linkages have attracted attention for more than 150 years.However, until recently their detailed investigation in mechanistic terms was somewhat neglected while extensive kinetic studies on the protein-hydrolysing enzymes, especially on chymotryp- sin by Bender and co-w~rkers,~~ were providing the basis for a molecular interpretation of enzyme action. Without doubt one of the major reasons Bourne and Finch 49 for this was the difficulty of purifying the glycosidases, particularly their reluctance to crystallize. In 1965 the picture changed when lysozyme, one of the few glycosidases to be crystallized, became the first enzyme to have its secondary and tertiary structure elucidated by x-ray cry~taIlography.~5 Furthermore, the detailed topography of the active site and the location of a substrate molecule therein was determined by x-ray diffraction analysis of the lysozyme-tri-N-acetyl- chitotriose complex.46 This brilliant work has stimulated some elegant experi- ments designed to elucidate the detailed mechanism of action of this enzyme, and has led to a resurgence of interest in glycosidases generally.One of the most obvious and intriguing properties of enzymes is their specificity, which may be defined as the range of compounds which a given enzyme will utilize successfully. Modern understanding of specificity stems from Fischer’s famous ‘lock and key’ hypothesis of 1894,47 but the original idea has been extended in three ways, two of which derive from the concept that enzyme molecules are flexible.The induced-fit theory48 postulates that an enzyme can adapt its conformation to that of a potential substrate, but that not all such induced conformations may be active. Secondly, there is the recognition of allosteric sites,49 which are distinct from the active site and catalytically inactive, but which may influence activity via conformational changes induced by the binding of small mo1ecules.49~50 The third additional factor is the recognition51 that most enzyme reactions are multi-step pro- cesses, and that selectivity or specificity may be exercised at any step including the initial (binding) one. Furthermore, selectivity may be due to electronic effects on the reactions involved as well as to steric effects.Thus the most quantitative approach to a discussion of specificity is in terms of the rate constants and energy parameters for the individual reaction steps, but since the nature of the steps is not known, except in a few cases, a more qualitative approach has often been used. A direct approach to the study of specificity is provided by x-ray crystallo- graphy, a difficult technique experimentally and one which may be applicable only to some enzymes (those which yield heavy-metal derivatives and form stable enzyme-substrate complexes). An alternative technique is to study the rates of interaction of an enzyme with a number of possible substrates of different structure. Carbohydrate substrates offer unusual opportunities for studying the factors involved since their structures may be varied between wide limits.The results obtained52 from studies of this type have been valuable in the classification of enzymes and in their application to the structural analysis of polysaccharides. Under natural conditions, water acts as the acceptor for degradative enzymes, except for some debranching enzymes and for phosphorylases. However if compounds which contain an alcohol group are added to the system they will often compete successfully with water for the role of acceptor; such compounds may for example be other carbohydrates or simple alcohols. One application of the use of simple alcohols is that the determination of anomeric configuration of the product glycoside reveals the stereochemistry of the enzyme reaction.53 If another sugar molecule is used as acceptor there will be a number of possible positions of linkage in the resultant disaccharide.R.I.C. Reviews 50 Wallenfels et aZ.s4 made the interesting observation that for transfer to glucose by ,B-galactosidase the disaccharide produced in greatest amount is that which is hydrolysed most readily, and it was concluded from this that the aglycone and the acceptor occupy the same position on the enzyme molecule. It has been rec0gnized~55~ for a long time that transglycosylation provides a means of preparing oligosaccharides which is often superior to available chemical methods of synthesis. Thus the action of Aspergillus niger a-glucosi- dase on glucose led to the first isolation of kojibiose (a-1,2-link).56 Further recent applications using different donor and acceptor molecules with extracts of yeast57 and of Tetrahymena pyriformi~~~ have afforded a number of unusual mixed disaccharides.A study of the transglycosylation reactions of fysozyme by Sharon and others59 has provided information on the mode and energies of binding of substrates to the active region of the enzyme. Examination of the mixture of higher oligosaccharides formed by the action of lysozyme on a tritium- labelled tetrasaccharide revealed that transglycosylation to give higher oligosaccharides plays a vital part in the overall process of hydrolysis of tetrasaccharide to disaccharide. The tree-like structure of glycogen, in which non-reducing end groups comprise about 10 per cent of all units,60 was elucidated largely by the use of degradative enzymes.The major mode of degradation in vivo is the stepwise cleavage of glucosyl units from the non-reducing ends by phosphorolytic transglycosylation to give a-D-g~ucosyl phosphate. However this process is halted in the vicinity of the branch points, and a phosphorylase-resistant ‘limit dextrin’ results, which has residual chains of four glucose units attached at the branch points.61 Further degradation is made possible by the action of a debranching enzyme, first isolated from rabbit muscle,62 which (i) transfers an a-maltotriosyl fragment to another part of the molecule61 and (ii) catalyses the hydrolysis of the 1,6-linkage to the remaining residue.Attempts to separate the transferase and amylo- 1,6-glucosidase activities have so far been U ~ S U C C ~ S S ~ U ~ , ~ ~ ~ and it has been suggested that the two functions may be carried out by the same enzyme. However the discovery636 of a glycogen storage disease type IIID, in which transferase activity is missing but glucosi- dase activity is present, is in conflict with this conclusion. For the degradation of a polysaccharide three possible action patterns may be envisaged: (i) single chain, in which all the linkages are broken in one chain before the enzyme forms an active complex with another polymer molecule; (ii) multiple chain, in which only one link is broken per effective enzyme-substrate encounter ; and (iii) multiple atta~k,~4 a general case in which the enzyme catalyses the hydrolysis of several but not all bonds in a chain before dissociating from it.One may define the degree of multiple attack as the number of bonds broken per single enzyme-substrate encounter. The situation is complicated by the fact that some enzymes catalyse the cleavage of glycosidic linkages stepwise from the non-reducing end of a polysaccharide chain (exo-enzymes), while others do so at the interior of a chain (endo-enzymes). One of the first enzymes to be examined was the exo- enzyme p-amylase, and a number of workers proposed a single chain action pattern. This view was criticized by Bourne and Whelan65 who held that the Bourne and Finch 51 experimental evidence was more in accord with a multichain process.The question was resolved seven years later in favour of a multiple attack mecha- nism of degree 4.3 for a short chain amylose containing 44 units.66 For an amylose of higher molecular weight but of narrow molecular weight distribution Husemann and Pfannemuller67 observed a decrease in molecular weight during IS-amylolysis, thus confirming a multichain or multiple attack mode of action. French64 has pointed out that single and multichain patterns may be thought of as opposite extreme cases of multiple attack, and suggested that the degree of multiple attack depends on the relative values of the rate of cleavage of glycosidic linkages and the rate of dissociation of the enzyme- polysaccharide complex.The studies have been extended recently to endo- enzymes by Robyt and French68 who have examined the action of three a-amylases and of 1M sulphuric acid on recrystallized amylose of degree of polymerization 1000 & 50. The degrees of multiple attack for the three a-amylases were 7.0 (porcine pancreatic at pH 6.9), 3.0 (human salivary at pH 6.9), and 2.9 (Aspergillus oryzae at pH 5.5). An unexpected result was the value of 1.9 for 1M sulphuric acid, which was expected to give a value corresponding to purely random attack, i.e. 1 .O. The related problem of the ‘mode of action’ of enzymes which catalyse polysaccharide synthesis has been explored using starch phosphorylase,65~69 levansucrase70~71 and dextrans~crase~l-73 but the results obtained so far are conflicting, even for the same enzyme.This may be attributed, at least in part, to the experimental difficulties involved in the characterization of the synthesis products as reaction proceeds, and also to the possible controlling influence of external factors such as polymer solubility or crystallizability. MECHANISMS OF ACTION OF GLYCOSYL TRANSFER ENZYMES The ultimate goals of mechanistic interpretations of enzyme action must be to explain in chemical terms the enormous catalytic power and subtle selec- tivity of enzymes towards substrates and products. A particular requirement of any mechanism proposed for transglycosylation is that the stereochemistry of the enzyme-catalysed reaction must be accounted for.All transglycosylases which have been examined appear to operate with rigid stereochemistry (inversion or retention of configuration) with respect to the anomeric carbon of the donated sugar unit; some examples are given in Table 4. In common with other classes of enzymes, transglycosylases bring about a reduction of the free energy of activation for the reaction catalysed.79 This is generally thought to occur by the involvement of one or more functional groups at the enzyme active site which lowers the potential energy of activa- tion AH$. The concomitant increase in kinetic energy of activation --TASZ, due to participation of the extra groups, may be offset by the precise orienta- tion of bonds undergoing reaction with respect to the functional group(s).** However, opinions differ as to whether the rates of enzyme-catalysed reactions can be accounted for quantitatively in these t e r m ~ .~ ~ ~ ~ ~ ~ 8 ~ A frequently used qualitative approach is to identify functional groups at the enzyme active site, and then to postulate how these groups might be involved in the enzyme reaction. The postulations are usually made on the basis of the known chemistry of the types of substrate and reaction under consideration. Such a R.I.C. Reviews 52 j?-Amy lase a- Amy I ase Phosphor y i ase Inversion Retention Retention , ' f+rr Table 4. Stereochemistry of some transglycosylation reactions Enzyme I ,3-Giucanase Celiulase L y soz y m e Dextransucrase Levansucrase Q-enzyme UDPG/starch synthetase U DPG/cellulose synthetase 10 15 procedure can be quite successful, particularly if a stable enzyme-substrate intermediate is formed which can be characterized, as in the case of some protein hydrcllases.81 Unfortunately, this is not normally the case with glycolytic enzymes and mechanistic interpretations have so far been based on more indirect information, and in most cases must be regarded as specula- tive. Non-enzymic hydrolysis of glycosides is known to proceed by a variety of mechanisms depending on the nature of the substrate and on the reaction conditions. The hydrolysis and alcoholysis of glycosides derived from sugars and simple alcohols proceeds via an acid-catalysed mechanism, which is usually formulated as :83 20 Stereochemistry Inversion Retention Retention Retention Retention Retention Retention Inversion -H+ NH Ref. 74,75 75 76 77 78 79 7 8 ~ H, ' 0 'H Glycosides of phenols (which are often used as enzyme substrates, since the product phenol may be estimated spectrophotometrically) may also be hydrolysed via base-catalysed processes.Three types of mechanism have been proposed which involve ( i ) nucleophilic (anion) attack at the anomeric carbon OH Bourne aid Finch (5) 53 atom with the formation of anhydro-compounds ;84 (ii) nucleophilic substitu- tion at the aromatic carbon atom involved in the glycosidic linkage;85~~~ and (iii) ionic di~sociation.8~ The relative rates of acid- and alkali-catalysed hydrolyses of a series of substituted phenyl-a- and ,8-D-glucosides were measured by Rydon and co-workers,85Jj6 and under alkaline conditions both a- and ,8-glucosides exhibited positive Hammett p-values, while in acid zero (a-glucosides) and small negative (p-glucosides) values were obtained.When these studies were carried out using a ,8-glucosidase85 and an ~glucosidase~~ positive p-values were obtained in both cases when the substituent a-values were correlated with an enzyme-substrate affinity constant. Although the interpretation of substituent effects is hampered by the difficulty of assessing possible steric contributions, the results certainly suggest that basic or nucleophilic catalysis is involved in some way. The effect of the structure of the glycose residue on non-enzymic hydrolysis has been studied by a number of workers.A particularly pertinent case is that of the substituted phenyl-2-acetamido-2-deoxy-~-glucopyranosides which have been studied by Bruice and co-workers.88 Following the sugges- tions of Capon89 and of Inch and Fletcher,go the results were interpreted in terms of intramolecular general acid catalysis by a carboxyl substituent in the phenyl ring and intramolecular nucleophilic catalysis by a neutral 2-acetamido group. Lowe et aE.91 measured the rates of lysozyme-catalysed hydrolysis of several p-aryf-N-acetyl-chitobiosides and analysed the tesults in terms of the Michaelis-Menteng2 equation. It was found that the Michaelis constant Km was independent of phenyl substituents but that the catalytic constant kcat showed a marked dependence, with p equal to + 1.2.It would seem that the action of lysozyme involves nucleophilic or basic participation in some way. With few exceptions,85@ all the mechanisms proposed for non-enzymic hydrolysis of glycosides involve glycosyl-oxygen fission. This has been observed experimentally for reactions catalysed by acid,93 and also by the enzymes sucrose phosphorylase, 94 yeast a-glucosidase,g5 almond ,8-gluco- sidase,95 invertase, 96 takamaltase, 97 ,&amylase, 98-100 a-amylase,100-101 fl-glucuronidase102 and lys0zyrne.7~ The acid-catalysed hydrolysis of methyl a-D-glucopyranoside is about twice as fast in deuterium oxide as in water, and this can be explained in terms of the mechanism shown in (5).lo3 However several enzymic hydrolyses proceed more slowly in deuterium oxide,104 and it has been argued105 that a different mechanism must be in operation.An alternative explanation106 is that heavy water may adversely affect the conformation of enzymes in solution. Consideration of the various mechanisms proposed for glycoside hydrolysis reveals that the rate determining step is that which involves the loss of the aglycone moiety. Thus one may write a general reaction sequence for an enzyme-catalysed hydrolysis which embodies all the features of the mecha- nisms proposed : E + S + ES 4 E.Glycose + Aglycone a__). H*O fast slow E + Glycose + Aglycone (6) R.I.C. Reviews 54 Some or all of the individual steps may be catalysed by acidic, basic, or nucleophilic groups at the enzyme active site and the enzyme complex E.Glycose could be ionic or covalent in nature.One would not expect to be able to isolate the enzyme substrate complex in the presence of acceptor, and as far as the authors are aware there have been only two reports of the isolation of glycolytic enzyme-substrate intermediates. Silverstein et aZ.107 have submitted evidence for the formation of a complex of sucrose phos- phorylase with glucose on reaction of sucrose with this enzyme, and the kinetic data were consistent with a double displacement mechanism originally proposed by Doudoroff et a1.108 Leglerl O9 has reported the isolation of a radioactive p-glucosidase complex after reaction with 14C-labelled conduritol B epoxide (2,3-anhydro-myo- inositol). The radioactivity could be released as (+)-inositol by treatment with 0.05M hydroxylamine, but not by treatment with sodium carbonate/hydrogen carbonate buffer of the same pH.It was suggested that complex formation occurred by acid-catalysed trans opening of the epoxide ring via attack of an enzyme carboxylate anion to give an ester of (+)-inositol, which released inositol when treated with hydroxylamine. However the operation of the same process during the hydrolysis of p-glucosides would presumably lead to a-glucose, which is in conflict with experimental evidence that the action of /I-glucosidase proceeds with retention of configuration.l1° It is possible that the stability of the inositol-enzyme intermediate may be due to the presence of a cyclohexane rather than a pyranosyl ring in the substrate, and it will be interesting to see if this type of experiment can be extended to other glycosi- dases. It has been postulated that transglycosylation proceeding with inversion occurs by a single-step nucleophilic .displacement. Products with retained configuration are supposed to arise by two successive displacements, the first of donor aglycone by an enzyme functional group, the second of the enzyme functional group by the acceptor.ll1 This latter explanation was proposed as the mechanism of sucrose phosphorolysis, and strong evidence was pro- vided by the observation of an exchange reaction between a-D-glUCOSyl phosphate and H32POz- catalysed by sucrose phosphorylase in the absence of fructose :I08 sucrose phosphorylase -..A a-D-glucose-1 -phosphate , postulated glucosyl enzyme + phosphate (7) This double inversion hypothesis has been extended to cover retention during polysaccharide synthesis.lf2 However, phosphate exchange does not occur in the case of polysaccharide phosphorylase unless the acceptor (glycogen or starch) is present.113 Lack of exchange may be explained in terms of Kosh- land’s induced fit theory by postulating that the enzyme active site adopts the correct configuration only in the presence of acceptor.48 Retention of configuration has been rationalized in two other ways: ( i ) formation of a carbonium ion followed by an addition whose stereo- specificity is controlled by the enzyme;ll4 or (ii) the operation of an SNi process.115 However it is difficult to explain lack of exchange in terms of Bourne and Finch 55 E + P E + S HOCHZ H \ ke Fig.I . Proposed mechanism of action of lysozyme. these last two theories. Present research in the field of polysaccharide bio- synthesis is largely concentrated on the isolation of donors and enzymes from a wide variety of sources, but it appears that substantial progress would result from the detailed examination of one or two systems. The x-ray crystallographic studies of lysozyme and of its complex with N-acetyl-chitotriose45s46 have provided a basis for further investigations of the mechanism of action of this enzyme.Under this stimulus, detailed studies of the enzymic and non-enzymic hydrolysis of glycosides of various mono- and oligosaccharides structurally related to the natural substrates have been carried out. As a result the possible mechanisms have been narrowed down to a small number of alternatives, although within these limitations some uncertainties and anomalies still remain. A large amount of experimental data45~46@~919116 supports the assignment of an enzyme-catalysed pathway involving general acid catalysis by the carboxyl group of glutamic acid 35 and intramolecular nucleophilic catalysis by the substrate 2-deoxy-2-acetamido group as represented in Fig. 1. The mechanism proposed postulates a double inversion at the substrate anomeric carbon atom, leading to retention of con- figuration, as has been 0bserved.7~ However the mechanism is not yet established with the certainty which this brief description might suggest, since the importance of intramolecular acetamido group participation is challenged by other experimental evidence.Raftery and Rand-Meir reported117 that, in the presence of the tetra- saccharide chitotetraose, lysozyme will catalyse the release of p-nitrophenol from p-nitrophenyl-/3-glycosides of N-acetyl-D-glucosamine, D-glUCOSe, and 2-deoxy-~-glucose. The relative rates of release of p-nitrophenol were 2 : 1 : 16 respectively, and it was shown that release ofp-nitrophenol occurred via synthesis of p-nitrophenyl oligosaccharides by transglycosylation.Thus it was proposed that acetamido group participation is not necessary to explain catalysis, and that the alternative pathways involve (i) a carbonium ion possibly stabilized by an enzyme basic group, e.g. aspartic acid 52,469118 or (ii) a covalent intermediate resulting from participation by an enzyme basic group. The steric distortion of the substrate proposed by Phillips and co- workerslls would presumably facilitate the production of a carbonium ion but not necessarily that of a covalent intermediate. It may be noted that the rate of acid-catalysed hydrolysis of methyl-2-deoxy-~-~-glucoside is ca 103 R. I. C. Reviews 56 times that of methyl-~-~-glucoside,~~~ and this has been ascribed to the greater ease of production of a carbonium ion (due to both electronic and steric effects).CONCLUSIONS In this brief survey we have attempted to demonstrate the extreme variety which has been encountered in the biochemical pathways of polysaccharide synthesis and degradation, while emphasizing how the majority of these pathways are unified by the concept of glycosyl transfer. It may be concluded that further progress towards the elucidation of the details of such pathways will be facilitated by the multitude of mild separation and purification procedures now available, coupled with the use of radioactively labelled enzyme substrates. As far as the mechanisms of the enzymically-catafysed reactions are concerned the situation is rather more uncertain. It would appear that further significant progress will require the detection and study of individual reaction steps and intermediates, perhaps by application of the rapid reaction techniques which have already been used with considerable success with a number of enzymes.12* Another area in which further informa- tion is desirable is in the non-enzymic catalysis of glycosyl cleavage reactions.A number of functional groups have been proposed as participants in enzy- mically catalysed transglycosylations but, apart from hydrogen ion, experi- mental data concerning the reactivity of such groups towards glycosides is extremely fragmentary. REFERENCES 1 See, for example, (a) D. A. Rees, The shapes of molecules. London: Oliver and Boyd, 2 See, for example, Industrialgums, ed.R. L. Whistler. New York and London: Academic 3 (a) D. A. Rees and J. Skerrett, Carbohydrate Res., 1968,7,334; (b) G. N. Ramachandran 1967; (b) M. Stacey, Chem. Brit., 1970, 6, 113 Press, 1959. 87. San Francisco and London: W. H. Freeman, 1968. and London: Academic Press, 1967. in Structural chemistry and molecular biology (ed. A. Rich and N. Davidson), pp. 83- 4 S. A. Barker and E. J. Bourne, Q. Rev. chem. SOC., 1953,7, 56. 5 W. Z. Hassid in Metabolic pathways, Vol. 1 (ed. D. M. Greenberg), 316. New York 6 E. E. Smith, P. M. Taylor and W. J. Whelan in Carbohydrate metabolism and its disorders, Vol. 1 (ed. F. Dickens, P. J. Randle and W. J. Whelan), 99-100. London: Academic Press, 1968. 7 (a) E. J. Hehre, Science, N. Y., 1941, 93, 237; (b) E. J.Hehre and J. Y. Sugg, J. expl Med., 1942, 75, 339. 8 S. Hestrin, S. Avineri-Shapiro and M. Aschner, Biochem. J., 1943, 37, 450. 9 S. Hestrin and S. Avineri-Shapiro, Biochem. J., 1944, 38, 2. 10 W. N. Haworth, S. Peat and E. J. Bourne, Nature, Lond., 1944, 154, 236. 11 G. T. Cori and C. F. Cori, J. biol. Chem., 1943,151, 57. 12 K. H. Meyer, Natural and synthetic high polymers, 2nd ed., pp. 468-469. New York: Interscience, 1950. 13 W. Verhue and H. G. Hers, Biochem. J., 1966, 99, 222. 14 J. Larner in Methods'in enzymology, vol. 1 (ed. S. P. Colowick and N. 0. Kaplan), 222-225. New York and London: Academic Press, 1955; C. Krisman, Biochim. biophys. Acta, 1962, 65, 307. 15 L. F. Leloir and C. E. Cardini, J. Am. chem.SOC., 1957, 79, 6340. 16 R. Caputto, L. F. Leloir, C. E. Cardini and A. C. Paladini, J. biol. Chem., 1950, 184, 333. 17 L. F. Leloir, Biochem. J., 1964, 91, 1. 18 L. Glaser and D. H. Brown, Proc. natn. Acad. Sci. U.S.A., 1955, 41, 253. Bourne and Finch 57 21 A. D. Elbein, G. A. Barber and W. Z. Hassid, J. Am. chem. SOC., 1964, 86, 309; G. A. Barber, A. D. Elbein and W. Z . Hassid, J. biol. Chem., 1964, 239,4056. 22 A. Bdolah and D. S. Feingold, Biochem. biophys. Res. Commun., 1965, 21, 543. 19 R. B. Frydman, Archs Biochem. Biophys., 1963,102, 242. 20 L. Glaser, J. biol. Chem., 1958, 232, 627. 23 1. M. Weiner, T. Higuchi, L. Rothfield, M. Saltmarsh-Andrew, M. J. Osborn, and 57, 1878. B. L. Horecker, Proc. natn. Acad. Sci. U.S.A., 1965, 54, 228.24 H. Nikaido and K. Nikaido, Biochem. biophys. Res. Commun., 1965, 19, 322. 1313. 25 M. J. Osborn and J. M. Weiner, J. biol. Chem., 1968, 243, 2631. 26 M. Scher, W. J. Lennarz and C. C. Sweeley, Proc. natn. Acad. Sci. U.S.A., 1968, 59, 27 Y. Higashi, J. L. Strominger and C. C. Sweeley, Proc. natn. Acad. Sci. U.S.A., 1967, 28 A. Wright, M. Dankert, P. Fennessey and P. W. Robbins, Proc. natn. Acad. Sci. Acad. Sci. U.S.A., 1965,53, 881. 235. U.S.A., 1967, 57, 1798. 29 J. R. Colvin, Can. J. Biochem. Physiol., 1961, 39, 1921. 30 J. S. Anderson, M. Matsuhashi, M. A. Haskin and J, L. Strominger, Proc. natn. 31 A. Wright, M. Dankert and P. W. Robbins, Proc. natn. Acad. Sci. U.S.A., 1965, 54, 32 M. Matsuhashi, C. P. Dietrich and J. L. Strominger, Proc.natn. Acad. Sci. U.S.A., 33 H. L. Conrad, J. R. Bambury, J. D. Epley and T. J. Kindt, Biochemistry, 1966,5,2808. 1965,54,587. 34 D. Abbott, E. J. Bourne and H. Weigel, J. chem. SOC. (C), 1966, 827. 35 L. F. Leloir, 6th Int. Congr. Biochem., New York, 1964, IUB, 35, 15. 36 D. A. Rees, Biochem. J., 1961, 81, 347. 37 A. Dorfman, Fedn Proc. Fedn Am. SOCS exp. Biol., 1962, 21, 1089; A. Markovitz, J. A. Cifonelli and A. Dorfman, J. 6ioI. Chem., 1959, 234, 2343. 38 S. Suzuki and J. L. Strominger, J. biol. Chem., 1960, 235, 257, 267, 274. 39 E. E. B. Smith, G. T. Mills, H. P. Bernheimer and R. Austrian, J. biol. Chem., 1960, 235, 1876. 40 I. R. Johnston, E. J. McGuire, G. W. Jourdian and S. Roseman, J. biol. Chem., 1966, Sarma, Nature, Lond., 1965, 206, 757.241,5735. 41 H. L. Griffin and Y. V. Wu, Biochemistry, 1968,7, 3063. 42 P. W. Robbins, D. Bray, M. Dankert and A. Wright, Science, N. Y., 1967, 158, 1636. 43 L. C. Gahan and H. E. Conrad, Biochemistry, 1968,7, 3979. 44 See, for example, M. L. Bender and F. J. Kezdy, J. Am. chem. SOC., 1964, 86, 3704. 45 C. C. F. Blake, D. F. Koenig, G. A. Mair, A. C. T. North, D. C. Phillips and V. R. 46 C. C. F. Blake, N. L. Johnson, G. A. Mair, A. C. T. North, D. C. Phillips and V. R. Sarma, Proc. R. SOC., 1967, B167, 378. 47 E. Fischer, Chem. Ber., 1894,27,2985; Hoppe-Seyler’s 2. physiol. Chem., 1898,26, 60. 48 D. E. Koshland Jr in The enzymes, vol. 1, 2nd edn (ed. P. D. Boyer, H. Lardy and K. Myrback), 332-336. New York: Academic Press, 1959.49 J. Monod, J. P. Changeux and F. Jacob, J. mol. Biol., 1963,6, 306. 50 J. Gerhart and A. B. Pardee, J. biol. Chem., 1962, 237, 891. 51 M. L. Bender, F. J. Kkzdy and C. R. Gunter, J. Am. chem. SOC., 1964,86, 3714. 52 See, for example, A. Gottschalk, Adv. Carbohyd. Chem., 1950, 5, 49; M. A, Jermyn, 53 J. H. Hash and K. W. King, J. biol. Chem., 1958, 232, 395. 54 K. Wallenfels and 0. P. Malhotra in The enzymes, vol. 4, 2nd edn (ed. P. D. Boyer, Rev. pure appl. Chem., 1961, 11, 92. H. Lardy and K. Myrback), 414. New York: Academic Press, 1960. 55 S. Peat, W. J. Whelan and K. A. Hinson, Nature, Lond., 1952, 170, 1056. 56 S. Peat, W. J. Whelan and K. A. Hinson, Chemy Ind., 1955, 385. 57 M. J. Clancy and W. J. Whelan, Archs Biochem.Biophys., 1967,118,724,730. 58 D. J. Manners and J. R. Stark, Carbohyd. Res., 1966, 3, 102. 59 D. M. Chipman, J. J. Pollock and N. Sharon, J. biol. Chem., 1968, 243, 487 and refs therein. 60 E. E. Smith, P. M. Taylor and W. J. Whelan in Carbohydrate metabolism and its disorders, vol. I (ed. F. Dickens, P. J. Randle and W. J. Whelan), 94. London: Academic Press, 1965. 61 G. J. Walker and W. J. Whelan, Biochem. J., 1960, 76, 264. 62 G. T. Cori and J. Larner, J. biol. Chem., 1951, 188, 17. 63 (a) D. H. Brown and B. Illingworth in Control of glycogen metabolism (ed. W. J. Whelan and M. P. Cameron), 139-150. London: Churchill, 1964; (b) H. G. Hers, W. Verhue and M. Mathieu, ibid., 151-175. R.Z. C. Reviews 58 64 D. French in The enzymes, vol.4,2nd edn (ed. P. D. Boyer, H. Lardy and K. Myrback), 359-361. New York: Academic Press, 1960. 65 E. J. Bourne and W. J. Whelan, Nature, Lond., 1950, 166, 258. 66 J. M. Bailey and D. French, J. biol. Chem., 1957, 226, 1. 67 E. Husemann and B. Pfannemiiller, Makromol. Chem., 1961, 49, 214. 68 J. F. Robyt and D. French, Archs Biochem. Biophys., 1967, 122, 8. 69 W. J. Whelan and J. M. Bailey, Biochem. J., 1954, 58, 560; E. E. Smith and W. J. Whelan, ibid., 1963, 88, 50P. 70 G. Rapoport, R. Dionne, E. Toulouse and R. Dedonder, Bull. SOC. Chim. Biol., 1966, 48, 1323. 71 K. H. Ebert and G. Schenk, Adv. Enzymol., 1968, 30, 179. 72 F. A. Bovey, J. Polym. Sci., 1959, 35, 167. 73 H. M. Tsuchiya, N. N. Hellman, H. J. Koepsell, J. Corman, C. S. Stringer, S.P. Rogovin, M. 0. Bogard, G. Bryant, V. H. Feger, C. A. Hoffman, F. R. Senti and R. W. Jackson, J. Am. chem. SOC., 1955, 77, 2412. 74 H. T. Brown and G. H. Morris, J . chem. SOC., 1895,67, 309. 75 R. Kuhn, Ber., 1924, 57B, 1965, Justus Liebigs Annln Chem., 1925,443, 1. 76 C. F. Cori, S. P. Colowick and G. T. Cori, J . biol. Chem., 1937, 121, 465. 77 F. W. Panish and E. T. Reese, Carbohyd. Res., 1967,3, 424. 78 D. R. Whitaker, Archs Biochem. Biophys., 1954, 53,436. 79 See, for example, J. A. Rupley and V. Gates, Proc. Natn Acad. Sci. U.S.A., 1967, 57, 496. 80 See, for example, R. Lumry in The enzymes, vol. 1,2nd edn (ed. P. D. Boyer, H. Lardy and K. Myrback), 197-199. New York: Academic Press, 1959, pp. 197-199. 81 D. E. Koshland Jr and K. E.Neet, A. Rev. Biochem., 1968, 37, 359. 82 J. A. Thoma, J. theor. Biol., 1965, 19, 297. 83 J. T. Edward, C’emy Ind., 1955, 1102. 84 C . E. Ballou, Adv. Carbohyd. Chem., 1954, 9, 59. 85 R. L. Nath and H. N. Rydon, Biochem. J., 1954, 57, 1. 86 A. N. Hall, S. Hollingshead and H. N. Rydon, J . chem. Soc., 1961,4290. 87 A. N. Hall, S. Hollingshead and H. N. Rydon, Biochem. J., 1962,84, 390. 88 (a) D. Piszkiewicz and T. C. Bruice, J . Am. chem. SOC., 1967,89,6237; (b) ibid., 1968, 90, 2156; (c) ibid., 1968, 90, 5844. 89 B. Capon, Tetrahedron Lett., 1963, 911. 90 T. D. Inch and H. G. Fletcher, J. org. Chem., 1966,31, 1810. 91 G. Lowe, G. Sheppard, M. L. Sinnott and A. Williams, Biochem. J., 1967, 104, 893. 92 L. Michaelis and M. L. Menten, Biochem. Z., 1919,49, 333. 93 C. A. Bunton, T. A. Lewis, D. R. Llewellyn and C. A. Vernon, J . chem. SOC., 1955, 4419. 94 M. Cohn, J. biol. Chem., 1949, 180, 771. 95 C. A. Bunton, T. A. Lewis, D. R. Llewellyn, H. Tristram and C. A. Vernon, Nature, Lond., 1954, 174, 560. 96 D. E. Koshland Jr and S. S. Stein, J. biol. Chem., 1954, 208, 138. 97 M. Halpern and J. Leibowitz, Bull. Res. Coun. Israel, 1957, A6, 131. 98 F. C. Meyer and J. Larner, Biochim. biophys. Acta, 1958, 29,465. 99 M. Halpern and J. Leibowitz, Biochim. biophys. Acta, 1959, 36, 29. 100 F. C. Mayer and J. Larner, J. Am. chem. SOC., 1959,81, 188. 101 M. Halpern and J. Leibowitz, Bull. Res. Coun. Israel, 1959, AS, 41. 103 (a) M. M. Kreevoy and R. W. Taft Jr, J. Am. chem. SOC., 1955,77,3146; (b) W. J. C . Orr and J. A. V. Butler, J . chem. SOC., 1937, 330; (c) M. Kilpatrick, J. Am. chem. Soc., 102 F. Eisenberg, Fedn Proc. Fedn Am. Socs exp. Biol., 1959, 18, 221. 1963,85, 1036. 1 04 D. E. Koshland Jr, Y. A. Yankeelov Jr., and J. A. Thoma, Fedn Proc. Fedn Am. Socs exp. Biol., 1962, 21, 1031. 105 M. Flasher and A. Lukton, Biochim. biophys. Acta., 1967, 146, 596. 106 M. L. Bender, G. E. Clement, J. F. KCzdy and H. D’A. Heck, J. Am. chem. SOC., 1964, 86, 3680. 107 R. Silverstein, J. Voet, D. Reed and R. H. Abeles, J. bid. Chem., 1967, 242, 1338. 108 M. Doudoroff, H. A. Barker and W. 2;. Hassid, J , biol. Chem., 1947, 168, 725. 1968, 151,728. 109 G. Legler, Hoppe-Seyler’s 2. Physiol. Chem., 1968, 349, 767; Biochim. biophys. Acta, 111 D. E. Koshland Jr., Biol. Rev., 1953, 28,416. 112 W. 2. Hassid in The structure and biosynthesis of macromolecules, Biochem. SOC. Symp., 110 G. Legler, Hoppe-Seyler’s 2. Physiol. Chem., 1967, 348, 1359. 1962, 21, 63. Bourne and Finch 59 115 D. E. Koshland Jr in The mechanism of enzyme action (ed. W. D. McElroy and B. 113 M. Cohn and G. T. Cori, J. biol. Chem., 1948, 175,89. 114 F. C. Mayer and J. Lamer, J. Am. chem. SOC., 1959, 81, 188. Glass), 608. Baltimore: Johns Hopkins Press, 1954; M. Cohn, J. cell. comp. Physiol., 1959, 54, Suppl. 1, 17. 116 G. Lowe and G. Sheppard, Chem. Communs, 1968,9, 529. 117 M. A. Raftery and T. Rand-Meir, Biochemistry, 1968, 7, 3281. 118 D. C. Phillips, Proc. Natn Acad. Sci, U.S.A., 1967, 57, 484. 119 W. G. Overend, C. W. Rees and J. S. Sequira, J . chem, Soc., 1962, 3429. 120 G. G. Hammes, Adv. Protein Chem., 1968,23, 1. 60 R.Z. C. Reviews