NAG-thiazoline, an N-acetyl-β-hexosaminidase inhibitor that implicates acetamido participation

Full text of DOI:10.1021/ja960826u

+
+
6804
J. Am. Chem. Soc. 1996, 118, 6804-6805
Scheme 1. Proposed N-Acetyl-â-hexosaminidase
Mechanisms
NAG-thiazoline, An N-Acetyl-â-hexosaminidase
Inhibitor That Implicates Acetamido Participation
Spencer Knapp,*,† David Vocadlo,‡ Zhinong Gao,†
Brian Kirk,† Jianping Lou,† and Stephen G. Withers*,‡
Departments of Chemistry
RutgerssThe State UniVersity of New Jersey
New Brunswick, New Jersey 08903
UniVersity of British Columbia, VancouVer
British Columbia, Canada V6T 1Z1
ReceiVed March 14, 1996
N-Acetylhexosaminidases (NAGases) are enzymes that pro-
mote the cleavage of N-acetylhexosaminides, as during glyco-
protein processing and glycolipid catabolism.1 They are highly
specific inasmuch as gluco- and galactopyranosides lacking the
2-acetamido functionality are poor substrates.2 A number of
inhibitors of NAGases have been discovered; these typically
mimic some aspect of an enzyme-protonated 2-acetamidopy-
ranoside substrate or a derived transition state (flattened C-1
conformation, positive or partially positive charge), and they
uniformly possess an acetamido group at the C-2 site.3-6 By
analogy to a widely cited mechanism for retaining â-glucosi-
dases,7 retaining N-acetyl-â-hexosaminidases may be said to
operate by stabilizing a transition state leading to a covalent
enzyme-substrate complex (Scheme 1, upper path).8 Alterna-
tively, the configuration-retaining aspect of certain NAGases
may be attributed to participation of the neighboring C-2
acetamido group, leading initially to a cyclized oxazoline
intermediate (Scheme 1, lower path).9 None of the existing
NAGase inhibitors points to a choice between these two
mechanistic possibilities, as the latter both feature similar
transition state characteristics in the vicinity of C-1.6 However,
allosamidin (2), an inhibitor of chitinase, incorporates a cyclic
Scheme 2. Synthesis of NAG-Thiazoline and a Precursor
Substrate
† Rutgers University.
‡ University of British Columbia.
(1) (a) Kennedy, J. F.; White, C. A. BioactiVe Carbohydrates in
Chemistry, Biochemistry, and Biology; John Wiley & Sons, New York,
1983; pp 230-242. (b) Conzelmann, E.; Sandhoff, K. AdV. Enzymol. Relat.
Areas Mol. Biol. 1987, 60, 89-216. (c) Kresse, H.; Glossl, J. AdV. Enzymol.
Relat. Areas Mol. Biol. 1987, 60, 217-311.
(2) The contribution of the acetamido group to NAGase binding has been
estimated at 4.2 kcal/mol. Lai, E. C. K.; Withers, S. G. Biochemistry 1994,
33, 14743-14749.
(3) 2-Acetamido-2-deoxy-D-gluconolactone: (a) Conchie, J.; Hay, A. J.;
Strachan, I.; Levvy, G. A. Biochem. J. 1967, 102, 929-941. (b) Li, S. C.;
Li, Y. T. J. Biol. Chem. 1970, 245, 5153-5160. (c) Sandhoff, K.; Wassle,
W. Hoppe-Seyler’s Z. Physiol. Chem. 1971, 352, 1119-1133. (d) Villar,
E.; Cabezas, J. E.; Calvo, P. Biochimie 1984, 66, 291-304. (e) Horsch,
M.; Hoesch, L.; Fleet, G. W.; Rast, D. M. J. Enzyme Inhib. 1993, 7, 47-
55.
(4) 2-Acetamido-(2-deoxy and 1,2-dideoxy)nojirimycin: (a) Kappes, E.;
Legler, G. J. Carbohydr. Chem. 1989, 8, 371-388. (b) Kajimoto, T.; Liu,
K. K.-C.; Pederson, R. L.; Zhong, Z.; Ichikawa, Y.; Porco, J. A.; Wong,
C.-H. J. Am. Chem. Soc. 1991, 113, 6187-6196. (c) Legler, G.; Lullau,
E.; Kappes, E.; Kastenholz, F. Biochim. Biophys. Acta 1991, 1080, 89-95.
(d) Fleet, G. W. J.; Smith, P. W.; Nash, R. J.; Fellows, L. E.; Parekh, R.
B.; Rademacher, T. W. Chem. Lett. 1986, 1051-1054.
(5) Some recent NAGase inhibitors: (a) Heightman, T. D.; Ermert, P.;
Klein, D.; Vasella, A. HelV. Chim. Acta 1995, 78, 514-532. (b) Liessem,
B.; Giannis, A.; Sandhoff, K.; Nieger, M. Carbohydr. Res. 1993, 250, 19-
30. (c) Wolk, D. R.; Vasella, A.; Schweikart, T.; Peter, M. G. HelV. Chim.
Acta 1992, 75, 323-334. (d) Aoyagi, T.; Suda, H.; Uotani, K.; Kojima, F.;
Aoyama, T.; Horiguchi, K.; Hamada, M.; Takeuchi, T. J. Antibiot. 1992,
45, 1404-1408. (e) Aoyama, T.; Naganawa, H.; Suda, H.; Uotani, K.;
Aoyagi, T.; Takeuchi, T. J. Antibiot. 1992, 45, 1557-1558. (f) Horsch,
M.; Hoesch, L.; Vasella, A.; Rast, D. M. Eur. J. Biochem. 1991, 197, 815-
818.
isourea functionality that, when protonated, resembles the
cyclized oxazoline shown in Scheme 1. A recent crystal-
lographic study of a chitinase/2 complex shows a position of
the inhibitor and a hydrogen-bonding pattern consistent with
acetamido participation in the natural substrate.10 Although the
“NAG-oxazoline” itself is too hydrolytically unstable for use
as an inhibitor of NAGases,6 we have prepared a stable version,
“NAG-thiazoline” 1, and report its powerful inhibitory effect
(6) An acetamido conduritol epoxide showed kinetics with jack bean
NAGase that suggested possible formation of an oxazoline intermediate.
Legler, G.; Bollhagen, R. Carbohydr. Res. 1992, 233, 113-123.
(7) Sinnott, M. L. Chem. ReV. 1990, 90, 1171-1202.
on jack bean N-acetyl-â-hexosaminidase and its enzyme-
(8) Koshland, D. E. Biol. ReV. 1953, 28, 416-436.
mediated formation from a precursor substrate.
(9) Lowe, G.; Sheppard, G.; Sinnott, M. L.; Williams, A. Biochem. J.
1967, 104, 893-899. Jones, C. S.; Kosman, D. J. J. Biol. Chem. 1980,
255, 11861-11869.
(10) van Scheltinga, A. C. T.; Armand, S.; Kalk, K. H.; Isogai, A.;
Henrissat, B.; Dijkstra, B. W. Biochemistry 1995, 34, 15619-15623.
S0002-7863(96)00826-8 CCC: $12 00 © 1996 American Chemical Society

+
+
Communications to the Editor
J. Am. Chem. Soc., Vol. 118, No. 28, 1996 6805
Figure 2. Time course of release of 4-methylumbelliferone from 0.65
mM MuTag (7) upon reaction with jack bean NAGase (O); Spontaneous
decomposition of 7 (2).
Figure 1. Inhibition of jack bean NAGase catalyzed hydrolysis of
PNPGlcNAc by NAG-thiazoline 1. The concentrations of 1 used were
0.000 (]), 0.087 (1), 0.203 (0), 0.434 (9), 0.868 (4), and 2.026 (2)
µM. Inset: graphical analysis of Ki by plotting Km apparent against
[NAG-thiazolin].
(Scheme 2). It was isolated after crystallization as a white solid,
mp 144.5-146 °C, and then was further purified by HPLC to
remove traces of thiazoline 1. Incubation of jack bean N-
acetylglucosaminidase with 7, while monitoring the release of
4-methylumbelliferone fluorometrically,16 resulted in slow, time-
dependent loss of activity, as shown in Figure 2. Reaction over
a 300 min period resulted in a 21-fold reduction in rate with
release of 2.2 µmol of 4-methylumbelliferone, consistent with
stoichiometric conversion of 7 to 1. No loss of enzyme activity
occurred in the absence of 7, and the putative precursor 7 is
itself a poor inhibitor of jack bean N-acetylglucosaminidase
when measured over time periods too short for formation of
significant quantities of 1.
These results on jack bean NAGase provide strong evidence
for a mechanism involving acetamido participation and an
oxazoline intermediate (Scheme 1, lower path) and complement
the chitinase crystallographic study.9 Furthermore, 2-deoxy-2-
thioacetamido glucosides such as 7 may prove valuable as
enzyme-activated inhibitor precursors with adjustable properties
according to their aglycon portions.
Treatment of 2-acetamido-2-deoxy-1,3,4,6-tetra-O-acetyl-â-
D-glucopyranose11 (3, Scheme 2) with Lawesson’s reagent12 led
to selective formation of the thioamide 4 (observable by TLC),
which then cyclized by displacement of acetate to provide the
thiazoline triacetate 5. The structure of 5 was confirmed by
the close match of its 1H NMR spectrum with that of the
corresponding NAG-derived oxazoline triacetate.13 The R-ano-
mer of 3 did not cyclize under these conditions, presumably
because direct participation of the thiocarbonyl with inversion
at C-1 would require a strained trans-fused transition state.
Methanolysis of the acetyls and then column chromatography
gave the NAG-thiazoline 1 as a hygroscopic off-white solid.
Not only is 1 stable at basic pH, but it also survives protonation
with trifluoromethanesulfonic acid in chloroform solution fol-
lowed by neutralization.
NAG-thiazoline 1 was found to be a potent competitive
inhibitor of jack bean N-acetylhexosaminidase as measured
against p-nitrophenyl 2-acetamido-2-deoxy-â-D-glucopyranoside
(Km ) 0.62 mM).14 Reciprocal replot of apparent Km values
vs inhibitor concentration, as shown in Figure 1, reveals a Ki
for 1 of 280 nM. It therefore binds almost 20 000 times more
tightly than the parent sugar N-acetyl-â,D-glucosamine (Ki ) 5
mM) and may be compared to the best inhibitors of this enzyme,
such as 2-acetamido-1,2-dideoxynojirimycin (Ki ) 140-230
nM), and 2-acetamido-2-deoxynojirimycin (1.2 nM).4a
Acknowledgment. We are grateful to American Cyanamid, Berlex
Corporation, Hoffmann-La Roche, the National Sciences and Engineer-
ing Council of Canada, and the Protein Engineering Network of Centres
of Excellence for support of this research. We thank Rutgers
undergraduates Peter Blomgren and Suzanne Moeller for assistance with
the preparation of starting materials and American Cyanamid for a
Faculty Award to S.K.
Supporting Information Available: Experimental procedures for
the synthesis of 1 and 7 (2 pages). See any current masthead for
ordering and Internet access instructions.
If the N-acetyl-â-hexosaminidase uses the acetamido group
for anchimeric assistance in the cleavage of the â-glycosidic
linkage, then a glycoside of 2-deoxy-2-thioacetamido-â-D-
glucose might be converted by the enzyme to the stable,
inhibitory, thiazoline 1, resulting in a time-dependent loss of
enzyme activity. The 4-methylumbelliferyl 2-thioacetamido
glucoside 7 was prepared from 4-methylumbelliferyl 2-acet-
amido-2-deoxy-3,4,6-tri-O-acetyl-â-D-glucopyranoside6,15 6 by
treatment with Lawesson’s reagent and then sodium methoxide
JA960826U
(15) Roy, R.; Tropper, F. D. Can. J. Chem. 1991, 69, 817-821.
(16) The time course of release of 4-methylumbelliferone from 7 by jack
bean NAGase was followed by establishing a series of reaction mixtures,
each containing NAGase (2.4 µg/mL) and 7 (0.65 mM) in 420 µL of 50
mM citrate buffer containing 100 mM NaCl and 0.1% BSA at pH 5.0.
These mixtures were quenched at various incubation times by adding 1.26
mL of 0.2 M glycine buffer, pH 10.65. The fluorescence due to 4-methyl-
umbelliferone was then measured at 450 nM. The resulting time-dependent
decrease in activity was shown to be due to the buildup of a reversible
inhibitor rather than covalent inactivation by repeating the experiment at a
higher enzyme concentration and then diluting the sample prior to assay.
Under these conditions essentially no time-dependent loss of enzyme activity
was observed.
(11) Available from Aldrich Chemical Company.
(12) Cava, M. P.; Levinson, M. I. Tetrahedron 1985, 41, 5061-5087.
(13) Srivastava, V. K. Carbohydr. Res. 1982, 103, 286-292.
(14) Jack bean NAGase was obtained from Sigma Chemical Company
and was assayed by using p-nitrophenyl N-acetyl-â-D-glucosaminide in 50
mM citrate buffer containing 100 mM NaCl and 0.1% BSA, pH 5.0.
Competitive inhibition studies were performed as described in ref 2.