(1967) and Vernon (1967) have proposed that the. ES' intermediate .... Roy. Soc. B, 167, 416. Vernon, C. A. (1967). Proc. Roy. Soc. B, 167, 389. Wenzel, M.
Biochem. J. (1967) 104, 893
893
Lysozyme-Catalysed Hydrolysis of some p-Aryl Di -N-acetylchitobiosides By G. LOWE, G. SHEPPARD, M. L. SINNOTT AND A. WILLIAMS The Dy8on Perrins Laboratory, Univer8ity of Oxford
(Received 1 February 1967) 1. Four fl-aryl di-N-acetylchitobiosides and ,-S-phenyl di-N-acetylthiochitobioside have been prepared and shown to be substrates for hen's-egg-white lysozyme. 2. The lysozyme-catalysed hydrolysis of these substrates obeys MichaelisMenten kinetics. 3. The Michaelis constants, K., for the ,-aryl di-N-acetylchitobiosides are almost independent of the aglycone, whereas the catalytic constants, k,,t., show a marked dependence, giving a Hammett reaction constant, p, equal to 1-2; this suggests the rate-determining step involves concerted acid-base or acid-nucleophilic catalysis. 4. This conclusion is supported by the MichaelisMenten constants found for ,B-S-phenyl di-N-acetylthiochitobioside. 5. A threestep reaction pathway is proposed, and mechanisms are suggested that would account for the evidence currently available.
Hen's-egg-white lysozyme (N-acetylmuramide CH2-OH CH2-OH CH2-OH1 glycanohydrolase, EC 3.2.1.17) is structurally the ,-0\ HO L X most thoroughly characterized enzyme at the HO I\ present time. Its primary amino acid sequence has OH Ac NAH been determined (Joll6s, Jauregui-Adell & Jolles, 1963, 1964; Canfield, 1963; Canfield & Liu, 1965; NH-Ac NH -Acj n NH -Ac]'n L L Brown, 1964) and its secondary and tertiary strucAc= *CO-CH3 R = - CH(Me) - C02H tures have since been established (Blake et at. 1965). Ac= -CO-CH3 Further, the tertiary structure of a lysozyme(I) substrate complex has been established (Blake et (II) al. 1967). By contrast with this unique situation CH2 OH CH2 * OH comparatively little is known about lysozyme's enzymic activity and no substrate has yet been /1o -Ox shown to obey Michaelis-Menten kinetics. ConseHO"H' quently the kinetic data that are available have NH-Ac NH-Ac not been separated into binding and catalytic components. Ac= *CO*CH3 The reason for this unusual situation can be (III) attributed to the fact that the natural substrate for this enzyme is the cell wall of certain sensitive bacteria, e.g. Micrococcus ly8odeikticu8, or, more specifically, the polysaccharide 'backbone' of the mucopeptide component of the bacterial cell wall, hydrolysis of chitin that tri-N-acetylchitotriose which does not lend itself to quantitative kinetic (II, n=3) was found to be the smallest substrate analysis. This polysaccharide is a regular polymer for lysozyme (Wenzel, Lenk & Schiitte, 1961; of the disaccharide, N-acetyl-p-D-glucosaminyl- Rupley, 1964). To undertake a detailed kinetic investigation of (1 -.4)-N-acetylmuramnic acid (I) linked together ,-(1-4) (Jeanloz, Sharon & Flowers, 1963; Leyh- the mechanism of action of lysozyme it was necessary to find water-soluble substrates in which one Bouille, Ghuysen, Tipper & Strominger, 1966). Chitin, the polysaccharide of N-acetyl-p-D- identifiable bond was cleaved. From the knowledge glucosamine (II) linked together P-(1 -*4), although that tri-N-acetylchitotriose was cleaved by lysonot a natural substrate for lysozyme, is nevertheless zyme into di-N-acetylchitobiose and N-aeetylhydrolysed by it. It was indeed through the use of glucosamine, it seemed likely that ,-aryl di-Noligosaccharides obtained by the acid-catalysed acetylchitobiosides (III, X = OAr) might prove to
894
G. LOWE, G. SHEPPARD, M. L. SINNOTT AND A. WILLIAMS
be suitable substrates for our purpose. During the course of our synthetic programme, we learned from Dr T. Osawa (to whom we are grateful for information before publication) that he had prepared f-p-nitrophenyl di-N-acetylchitobioside (III; X=p-NO2 C6H4.0) and that lysozyme catalysed its hydrolysis into di-N-acetylchitobiose and pnitrophenol; he could find no evidence for the production of N-acetylglucosamine (Osawa, 1966). Several /-aryl di-N-acetylchitobiosides and ,B-Sphenyl di-N-acetylthiochitobioside have now been prepared and investigated as potential substrates for lysozyme. It was expected that, if several variously substituted aryl glycosides were found to be substrates for lysozyme, useful information pertinent to the mechanism of action of this enzyme could be obtained.
MATERIALS AND METHODS LysozyMe. Hen's-egg-white lysozyme (3 x crystallized) was obtained from Seravac Laboratories Ltd., Maidenhead, Berks. Chitin. Prawn or shrimp shells (7kg.) were kept in 5% (w/v) NaOH solution at 800 for 2 days. The shells were then washed with water and kept at 200 with 2N-HCI for a further 2 days. The residual shells were then washed with water and dried to give 300g. of chitin. Chitobiose octa-acetate. This was prepared from chitin (30g.) by the method of Barker, Foster, Stacey & Webber (1958). The product (6g.) had m.p. 3050, [a]D+ 530 (in acetic acid). a-Acetochlorochitobiose. Chitobiose octa-acetate (4 0g.) was converted into a-acetochlorochitobiose (4.0g.), m.p. approx. 1950 (decomp.), La]D+ 40° (in CHCl3), by the method of Osawa (1966). P-Phenyl di-N-acetylchitobioside (III; X= O-C6H5). A solution of x-acetochlorochitobiose (0 3g.), phenol (0-12g.), acetone (lOml.) and 0-5N-NaOH (2ml.) was kept at 50 for 16hr. The P-phenyl hepta-acetylchitobioside crystallized out as the acetone was removed in a Rotovac and had m.p. 1700, []D- 18-9' (in acetic acid). This material in methanol (2-5 ml.) containing a trace of sodium (to give approx. 0-01 M-sodium methoxide) was warmed to clarify the suspension, and kept at 200 for 16hr. The /-phenyl di-N-acetylchitobioside (0 08g.), which crystallized out, had m.p. 3040, [a]2 -15-10 (in water) (Found: C, 49-4; H, 6-8; N, 4-6. C22H31N2011,2H20 requires C, 49-4; H, 6-6; N, 5.2%). ,B-p-Nitrophenyl di-N-acetylchitobioside (III; X= O C6H4 NO2-p). a-Acetochlorochitobiose was converted into P-p-nitrophenyl di-N-acetylchitobioside by the method of Osawa (1966). f-o-Nitrophenyl di-N-acetylchitobioside (III; X= O.C6H4.NO2-o). A solution of o-nitrophenol (0-13g.), a-acetochlorochitobiose (0 3g.), acetone (lOml.) and 0 5N-NaOH (2ml.) was kept at 50 for l6hr. Removal of the acetone in a Rotovac gave 3-o-nitrophenyl hepta-acetylchitobioside (0 2g.) as a white powder, m.p. 229-230°. To a suspension of this material (0-14g.) in methanol (5mi.) was added a small amount of sodium (to give approx. 0-01 M-sodium methoxide). The suspension was clarified by
1'367
heating to approx. 600, and the solution was kept at 200 for 2 days, during which time the product (0.05g.) crystallized out, m.p. 2400, [oc]D -17 (in water) (Found: C, 46-6; H, 5-8; N, 6-8. C22H31N3013,H20 requires C, 47-0; H, 5-9; N, 6.8%). fl-2,4-Dinitrophenyl di-N-acetylchitobioside [III; X= O * C6H3(NO2)2]. A solution of cz-acetochlorochitobiose (1-2g.), 2,4-dinitrophenol (10g.), acetone (20ml.) and 0-5N-NaOH (20ml.) was kept at 00 for 24hr. The acetone was removed in a Rotovac and water (lOOml.) added. The precipitate was washed with saturated NaHCO3 solution and recrystallized from methanol to give /3-2,4-dinitrophenyl hepta-acetylchitobioside (0 06g.) as colourless needles, m.p. 191-192° (decomp.), [a]" 7.5' (c 0 -I in methanol) (Found: C, 48-0; H, 4-6; N, 7-0. C32H40N4020 requires C, 48-0; H, 5-0; N, 7.0%). The hepta-acetylglycoside (20mg.) was shaken with methanolic 0.5% magnesium methoxide (2 ml.) at 20° until a homogeneous solution was obtained, and the solution kept at 50 for 20hr. (cf. Yamamoto,Miyashita & Tsukamoto, 1965). The product (6mg.) separated as white crystals, but decomposed on heating and was light-sensitive, Vmax. (Nujol) 1665 cm.-1 (NH*CO*CH3). ,B-S-Phenyl di-N-acetylthiochitobioside (III; X= S*C6H5). A solution of oc-acetochlorochitobiose (0-7g.), thiophenol (0-24ml.), acetone (20ml.) and N-NaOH (2-6ml.) was kept at 200 for 24hr. and then poured into water (50ml.). The solid was filtered off and washed with water to give an amorphous powder (0-3g.), m.p. 288-300o (decomp.). The hepta-acetylglycoside in methanolic 0- M-sodium methoxide (20ml.) was kept at 200 for 24hr., during which time the product crystallized out. Recrystallization from water gave /-S-phenyl di-N-acetylthiochitobioside as colourless needles (0.20g.), m.p. 289-293' (decomp.), [oc]" -2.5° (c 0-2 in water), Am,. 2450A (e 693) (Found: C, 47-6; H, 6-2; N, 5-1; S, 5-6. C22H32N2010S,2H20 requires C, 47-7; H, 6-5; N, 5-1; S, 5.8%). Kinetic&. (a) A solution of lysozyme in citrate buffer was equilibrated in a thermostatically controlled water bath. The substrate in water (to give a final buffer concentration of 0-1 M), also thermally equilibrated at the same temperature, was added and the solution gently shaken. Samples (0-5ml.) were withdrawn from the reaction vessel every 2hr. and added to O- lN-NaOH (2-5ml.). The extinction was measured at an appropriate wavelength: p-nitrophenol, 4000A (E 18000); o-nitrophenol, 4000A (e 4800). By determining the initial velocities (less than 2% of the substrate was hydrolysed) at different substrate concentrations (range indicated in Table 1) the Michaelis-Menten parameters, kcat. and K., were obtained from the equation: v
=
kcat.[Eo][So]/(Km+[So])
by using a Lineweaver-Burk plot. A modified Autocode computer programme employing an English Electric KDF9 digital computer was used to calculate the MichaelisMenten parameters and the standard deviations by the method of least squares. (b) A solution of lysozyme (3 ml.) in 0-1 M-citrate buffer, pH5-0, was thermally equilibrated in a modified Unicam SP. 500 spectrophotometer fitted with a photomultiplier, the output being fed to a potentiometric pen recorder. A solution of P-2,4-dinitrophenyl di-N-acetylchitobioside in 0- M-citrate buffer, pH5-0 (0. ml.), was added and the transmission at 3600k was recorded as a function of time.
LYSOZYME-CATALYSED HYDROLYSES
Vol. 104
Infinity readings were determined by heating stock solutions at 1000 for 15min., diluting samples (0- Iml.) in 0- IMcitrate buffer solution, pH5-0, and measuring the extinction at 36001. Because of the relatively high blank rate for the ,B-2,4-dinitrophenyl di-N-acetylchitobioside it was necessary to use the conditions [EO] > [So]. Under these conditions, and allowing for the blank rate kb determined separately, it can be shown (cf. Kezdy & Bender, 1962) that for the three-step pathway [eqn. (1) below]: d[P] - [(kt -kb)[E ] + kb] ([S]dt K+[0
[PD])
895
15
0
5
and therefore the observed first-order rate constant:
K,i
- kb)[Eo] ko - (kt Km+ [Eo] +kb
iI 0
The kinetics were never followed beyond 10% hydrolysis. A plot of 1/(ko-kb) against 1/[EO] therefore allowed the determination of the Michaelis-Menten parameters. The computer programme referred to above was used for this purpose.
(c) A solution of lysozyme in 01 M-phosphate buffer,
pH5*2, was thermally equilibrated at 35°. The substrate was added and the solution gently shaken. Samples (lml.) were withdrawn at suitable time-intervals up to 5% hydrolysis and shaken with 'spectrosol' cyclohexane (4ml.) for 30sec., and the mixture was centrifuged for 90sec. The extinction of the organic layer was measured at 2360A for ,B-S-phenyl di-N-acetylthiochitobioside and 2710A for ,B-phenyl di-N-acetylehitobioside. Thiophenol in cyclohexane has A,,.. 23601 (c 10000) (Koch, 1949), and phenol in cyclohexane has A.... 2710A (e 2100) (Mizushima, Tsuboi, Shimanouchi & Tsuda, 1953). The partition coefficients between the phosphate buffer and cyclohexane are 3: 2 and 3: 1 respectively. The kinetics were performed under the conditions [Eo]> [So]; infinity data were calculated and first-order constants obtained by using a leastsquares programme. Values of Il/ko against [EO] were plotted for a fixed initial substrate concentration and the Michaelis-Menten parameters determined by using the
equation:
d[PI] dt
kjat[Eo]([SO]-[P1]) K.+[Eo]
Although the conditions (i.e. [Eo] > [So]) only approximate to those that lead to this equation, the Michaelis-Menten parameters were insignificantly different from those derived by using the more complex rate equation for the conditions [Eo] [So], which contains both first-order and zero-order terms.
RESULTS AND DISCUSSION The change of initial velocity, v, for the lysozymecatalysed hydrolysis of f-p-nitrophenyl di-Nacetylchitobioside (III; X=OC6H4.NO2) at different substrate concentrations [So] (see Fig. 1) demonstrates for the first time a substrate of lysozyme obeying Michaelis-Menten kinetics. The theoretical curve was obtained from the statistically derived Michaelis-Menten parameters. Four ,B-aryl di-N acetylchitobiosides have been found to be
10
5
15
[So] (mM) Fig. 1. Change in initial velocity, v, with substrate concentration, [So], for the lysozyme-catalysed hydrolysis of /-p-nitrophenyl di-N-acetylchitobioside (III; X= O.C6H4.NO2) in 0 1 M-citrate buffer, pH5-2, at 35°.
substrates for lysozyme, and the Michaelis-Menten parameters, kcat. and K., have been determined under the conditions indicated in Table 1. Because of the high blank rate with P-dinitrophenyl di-Nacetylchitobioside [III; X = O *C6H3(NO2)2] the Michaelis-Menten parameters were determined under the somewhat unusual conditions in which the enzyme concentration was larger than the substrate concentration (cf. Kezdy & Bender, 1962). With fl-phenyl di-N-acetylchitobioside (III; X= OPh) and with P-S-phenyl di-N-acetylthiochitobioside (III; X = SPh) it was also necessary to use a very high enzyme concentration to achieve a rate of phenol and thiophenol production that would allow kinetic data to be obtained. The results shown in Table 1 indicate that the four aryl di-N-acetylchitobiosides (III; X = OAr) have very similar Km values whereas the kt,t. values differ markedly, those for the phenyl glycoside and the dinitrophenyl glycoside differing by over 1000fold. These results strongly suggest that the catalytic constants (kC,,t.) incorporate the step in which the glycosidic group is cleaved and that this step is excluded from K,. These kinetic results could formally be interpreted in terms of the simple twostep Michaelis-Menten equation in which the free sugar and the aglycone were generated simultaneously. However, in view of the fact that lysozyme is known (Rupley, 1967; G. Lowe & G. Sheppard, unpublished work) to cleave the C(1)-O glycosidic bond with overall retention of configuration, a three-step reaction pathway (1) involving two catalytic steps is much more probable
896
1967
G. LOWE, G. SHEPPARD, M. L. SINNOTT AND A. WILLIAMS
(Koshland, 1953), P1 and P2 representing the aglycone and sugar respectively: C)
-
-1
lo:
k.
k-1
_ o
00 _
0 x x x x
*'S3
o 0_
0
o
C)
000 0 0x x
+1+1+1+1+ '-40
0)0
_
*C;l
o Co Co
+l +1 +1 +1 +1
40 0D
.
.
E-4-E 0
0D
0
ks
ks
E+S "-ES = ES'
(1)
E+P2
+PI
The approximate similarity of the Km values for the four O-aryl glycosides indicates that k2 < k3, but the fact that there is a slight drift may mean that k3 is not very much larger than k2 for the dinitrophenyl glycoside. For our present discussion, however, we shall assume that the slight variation in Km arises from small differences in binding and therefore kcat. represents the k2 step. From an X-ray-crystallographic investigation of the lysozyme-tri-N-acetylchitotriose complex, Blake et al. (1967) have suggested that the catalytic site of lysozyme is located close to and probably between the carboxylic acid residues of glutamic acid-35 and aspartic acid-52; no other functional amino acids occur in this region. In view of the susceptibility of the glycosidic bond to acid- rather than base-catalysed hydrolysis, it seems most probable that, if this suggestion is correct, either glutamic acid-35 or aspartic acid-52 is acting as an intra-complex acid catalyst. The observation made by Capon (1963), that o-carboxyphenyl fl-D-glucoside hydrolyses about 104 times faster than p-carboxyphenyl ,-D-glucoside at 350, is particularly relevant. Evidence has
o o Co Co Co
---
£1 v
C)
-3
000_
.
-4
0; 0
00
0-
-56 0
-6
o
v-
OtI
-7
e~Zmzz
,00^
O~ * .QO OI OI
11 -1
V
1 1
1
* ^ * -^
0
1
2
3
a
Fig. 2. Hammett plot for lysozyme-catalysed hydrolysis of some ,B-aryl di-N-acetylchitobiosides at 350; p=1i23 ±0-17.
LYSOZYME-CATALYSED HYDROLYSES
Vol. 104
been presented (cf. Capon & Smith, 1965) that shows that this rate enhancement is probably due to intramolecular general acid catalysis by the undissociated o-carboxylic acid group. However, it has already been pointed out that, in spite of this considerable rate enhancement, this form of catalysis alone is inadequate to account for lysozyme's catalytic efficacy (Lowe, 1967). The results shown in Table 1 provide clear evidence in support of this contention. The fact that the catalytic constant (k.t.) for the lysozyme-catalysed hydrolysis of ,-2,4-dinitrophenyl di-N-acetylchitobioside is approx. 1000 times that of the phenyl glycoside suggests that nucleophilic or basic catalysis or both must be assisting the acid catalysis. This follows from the fact that the acid- and base-catalysed hydrolyses of aryl ,B-D-glucosides have p values
897 (Hammett, 1940) of - 0-66 and + 2-48 respectively (Nath & Rydon, 1954). The p value found from the kc.at values in Table 1 by using a constants (p-NO2, 1-27; Hammett, 1940; o-NO2, 1-24; Barlin & Perrin, 1966) is + 1-23 + 0-17 (see Fig. 2), suggesting that concerted acid-base or acid-nucleophilic catalysis is involved in the rate-determining step. This conclusion is supported by the Michaelis-Menten parxaneters found for fi-S-phenyl di-N-acetylthiochitobioside (III; X = SPh). The Michaelis constant, Kin, is greater (i.e. binding is weaker) than for the ,-phenyl di-N-acetylchitobioside (III; X = OPh), as would be expected if the glycosidic oxygen or sulphur is hydrogen-bonded to a carboxylic acid residue, sulphur forming a much weaker hydrogen bond (Heafield, Hopkins & Hunter, 1942). The greater catalytic constant,
ES
ES'
Glu-35
09 (la) E+S
-ROH
>
' =H
E+P2
Glu-35
-ROH
(lb) E+S >
NO
HO - CH2
0
H\ fH OH
R'O Ac-NH
O, e -o
' E+P2
O, . eO
Asp-52 Asp-52
H
C9 R R
(2) E+S
HOCH2 o
N0
R'O \
~0
p
eCH3 29
H
-ROlH
HO-CH2
R'0 \
O
E+P2
OH3
H>CH3
Scheme 1. Possible mechanisms for lysozyme-catalysed hydrolys. Ac, CH-CO. Bioch. 1967, 104
898
G. LOWE, G. SHEPPARD, M. L. SINNOTT AND A. WILLIAMS
kc&t for the thiochitobioside (III; X = SPh) compared with ,B-phenyl. di-N-acetylchitobioside (III; X= OPh) supports the view that the rate-determining step involves, in addition to acid catalysis, either general base or nucleophilic catalysis. The relative rate constants for the acid- and basecatalysed hydrolysis of phenyl 2-acetamido-2(ko) and S-phenyl 2deoxy-p-D-glucopyranoside acetamido-2-deoxy-,-D-1-thioglucopyranoside (kg) are iceks = 580 (2N-hydrochloric acid at 1000) and ksl/ko= 16 (N-sodium hydroxide at 61.50) (G. Lowe, G. Sheppard & M. L. Sinnott, unpublished work). Since lysozyme-catalysed cleavage occurs at the C(j)-O glycosidic bond with retention of configuration, the possibility that general base catalysis can occur at C(l) of the substrate can be excluded. The cleavage of the substrate's glycosidic bond (k2 step) by lysozyme must therefore arise from concerted nucleophilic-general acid catalysis. The nucleophilic assistance to the general acid catalysis could arise in three possible ways. First, the second carboxylic acid group in the catalytic site could in its ionized form participate and so stabilize the transition state leading to glycolysis. Secondly, the 2-acetamido group in the sugar itself could act as a neighbouring nucleophilic group, and so stabilize the transition state leading to glycolysis. Thirdly, it is conceivable that the second carboxylic acid group in the catalytic site could in its ionized form further assist the neighbouring 2-acetamido group either by weakening the N-H bond or by directly attacking the carbon atom of the acetamido group in the transition state. This third mechanism, however, would require a considerably different conformation of the catalytic site from that found for the lysozyme-tri-N-acetylchitotriose complex. In view of the fact that oligosaccharides of N-acetylglucosamine with more than three sugar residues do not bind better than triN-acetylchitotriose (Rupley, 1967), it would seem that the conformation of the protein revealed by the crystallographic studies of the lysozyme-tri-Nacetylchitotriose complex (Blake et al. 1967) closely represents the conformation in which catalysis takes place. This then renders the third mechanism improbable, and we are left therefore with essentially two mechanisms to account for lysozyme catalysis: these are shown in Scheme 1. An inspection of the model of the lysozyme-triN-acetylchitotriose complex reveals that in the first of these mechanisms (la in Scheme 1), in which glutamic acid-35 acts as a general acid and aspartic acid-52 the nucleophile, there would need to be considerable movement (approx. 2k) of aspartic acid-52 for participation in the transition state (Fig. 3). There can be little doubt, however, that the breakdown of the glycosyl-enzyme (ES') would be rapid compared with the catalytic constants
1967
OH2- OH
Asp-52 Ac.NHV)g /
0
0 og1
3.5 A
o0
Glu-35 Fig. 3. Projection of the proposed catalytic site in the enzyme-substrate complex as required by mechanisms la and lb (Scheme 1), showing that, when glutamic acid-35 is hydrogen-bonded to the glycosidic oxygen, C(1) is 3-51 from aspartic acid-52, assuming the conformation found in the lysozyme-tri-N-acetylchitotriose complex. Ac,
CH3-CO.
Fig. 4. Projection of the proposed catalytic site in the enzyme-substrate complex, as required by mechanism 2 (Scheme 1), showing the sugar residue being hydrolysed in the 'boat' conformation and so allowing aspartic acid-52 to hydrogen-bond with the glycosidic oxygen and the neighbouring N-acetyl group to participate in the cleavage of the glycosidic bond. The conformation found in the lysozyme-tri-N-acetylchitotriose complex is assumed.
observed (k,, in Table 1) (cf. Fife, 1965). To avoid the necessity for this considerable conformational change in reaching the transition state, Blake et al. (1967) and Vernon (1967) have proposed that the ES' intermediate remains as an ion pair (lb in Scheme 1), which never collapses to a covalent bond. In the second mechanism (2 in Scheme 1), in which general acid catalysis by aspartic acid-52 is intramolecularly assisted by the neighbouring 2-acetamido group, it is necessary that the glycosidic residue undergoing cleavage should take up the 'boat' conformation (Fig. 4). Inspection of the model of the lysozyme-tri-N-acetylchitotriose complex reveals that such a conformational change of the sugar in the catalytic site may well be compensated for by the additional hydrogen-bonding acquired between the substrate and enzyme. It is perhaps pertinent to note that, in the Ag+-catalysed
Vol. 104
LYSOZYME-CATALYSED HYDROLYSES
displacement reactions of tetra-acetyl-fl-D-glucosyl halides (Koenigs-Knorr reaction), the 2-acetoxy group is known to participate, leading to retention of configuration (Frush & Isbell, 1941). Amides are also known to be effective neighbouring groups (Capon, 1964). Of particular relevance to the present discussion is the investigation by Inch & Fletcher (1966), who showed that, in the solvolysis of 1-0-acyl-N-acetyl-D-glucosamine and l-0-acylN-acetyl-D-galactosamine, the I,-anomers (i.e. the 1-acyl group tranm to the 2-acetamido group) are solvolysed with anchimeric assistance and with retention of configuration. The la-anomers (i.e. the 1-acyl group ci8 to the 2-acetamido group), however, showed no evidence of anchimneric assistance and no stereochemical control. Both mechanism (1) and (2) (in Scheme 1) account for the evidence currently available on the mechaism of lysozyme catalysis. It should, however, be possible to devise experiments that would enable a clear distinction to be made between them. The authors thank the Science Research Council for a research studentship (to G.S.), the Ministry of Education for a State Scholarship (to M. L. S.) and the Science Research Council for a N.A.T.O. Fellowship (to A.W.).
REFERENCES Barker, S. A., Foster, A. B., Stacey, M. & Webber, J. M. (1958). J. chem. Soc. p. 2218. Barlin, G. B. & Perrin, D. D. (1966). Quart. Rev. chem. Soc., Lond., 20, 75. Blake, C. C. F., Johnson, L. N., Mair, G. A., North, A. C. T., Phillips, D. C. & Sarma, V. R. (1967). Proc. Roy. Soc. B, 167, 378.
899
Blake, C. C. F., Koenig, D. F., Mair, G. A., North, A. C. T., Phillips, D. C. & Sarma, V. R. (1965). Nature, Lond., 206, 757. Brown, J. R. (1964). Biochem. J. 92, 13P. Canfield, R. E. (1963). J. biol. Chem. 238, 2698. Canfield, R. E. & Liu, A. K. (1965). J. biol. Chem. 240,1997. Capon, B. (1963). Tetrahedron Lett. no. 14, 911. Capon, B. (1964). Quart. Rev. chem. Soc., Lond., 18, 45. Capon, B. & Smith, M. C. (1965). Chem. Commun. p. 523. Fife, T. H. (1965). J. Amer. chem. Soc. 87, 271. Frush, H. L. & Isbell, H. S. (1941). Bur. Stand. J. Re., Wash., 27, 413. Hammett, L. P. (1940). Physical Organic Chemistry, p. 184. New York: McGraw-Hill Book Co. Heafield, T. G., Hopkins, G. & Hunter, L. (1942). Nature, Lond., 149, 218. Inch, T. D. & Fletcher, H. G. (1966). J. org. Chem. 31, 1810. Jeanloz, R. W., Sharon, N. & Flowers, H. M. (1963). Biochem. biophys. Re8. Commun. 13, 20. Jolles, J., Jauregui-Adell, J. & Jolles, P. (1963). Biochim. biophys. Acta, 78, 668. Joll6s, P., Jauregui-Adell, J. & Jolles, J. (1964). C.R. Acad. Sci., Paris, 258, 3926. Kezdy, F. & Bender, M. L. (1962). Biochemistry, 1, 1097. Koch, H. P. (1949). J. chem. Soc. p. 391. Koshland, D. E. (1953). Biol. Rev. 28, 416. Leyh-Bouille, M., Ghuysen, J.-M., Tipper, D. J. & Strominger, J. L. (1966). Biochemistry, 5, 3079. Lowe, G. (1967). Proc. Roy. Soc. B, 167,431. Mizushima, S., Tsuboi, M., Shimanouchi, T. & Tsuda, Y. (1953). Spedcrochim. Acta, 7, 103. Nath, R. L. & Rydon, H. N. (1954). Biochem. J. 57, 1. Osawa, T. (1966). Carbohyd. ReB. 1, 435. Rupley, J. A. (1964). Biochim. biophys. Acta, 83, 245. Rupley, J. A. (1967). Proc. Roy. Soc. B, 167, 416. Vernon, C. A. (1967). Proc. Roy. Soc. B, 167, 389. Wenzel, M., Lenk, H. P. & Schutte, E. (1961). Hoppe-Seyl. Z. 827, 13. Yamamoto, A., Miyashita, C. & Tsukamoto, H. (1965). Chem. Pharm. Bull., Tokyo, 13, 1036.
