The First Direct Evaluation of the Two-Active Site Mechanism for Chitin Synthase

Article abstract of DOI:10.1021/jo035100c

Chitin synthase polymerizes UDP-GlcNAc to form chitin (poly-β (1,4)-GlcNAc) and is essential for fungal cell wall biosynthesis. The alternating orientation of the GlcNAc residues within the chitin chain has led to the proposal that chitin synthase possesses two active sites. We report the results of the first direct test of this possibility. Two simple uridine-derived dimeric inhibitors are shown to exhibit 10-fold greater inhibition than a monomeric control, consistent with the presence of two active sites. This observation has important implications for the development of antifungal agents, as well as the understanding of polymerizing glycosyltransferases.

Full text of DOI:10.1021/jo035100c

Th e F ir st Dir ect Eva lu a tion of th e Tw o-Active Site Mech a n ism for
Ch itin Syn th a se
Adam R. Yeager and Nathaniel S. Finney*
Department of Chemistry and Biochemistry, University of California, San Diego,
9500 Gilman Dr., La J olla, California 92093-0358
nfinney@chem.ucsd.edu
Received J uly 28, 2003
Chitin synthase polymerizes UDP-GlcNAc to form chitin (poly-â(1,4)-GlcNAc) and is essential for
fungal cell wall biosynthesis. The alternating orientation of the GlcNAc residues within the chitin
chain has led to the proposal that chitin synthase possesses two active sites. We report the results
of the first direct test of this possibility. Two simple uridine-derived dimeric inhibitors are shown
to exhibit 10-fold greater inhibition than a monomeric control, consistent with the presence of two
active sites. This observation has important implications for the development of antifungal agents,
as well as the understanding of polymerizing glycosyltransferases.
In tr od u ction
We wish to report the first direct evaluation of the two-
active site mechanism for chitin synthase (CS), a glyco-
syltransferase essential for fungal cell wall biosynthesis.1
Comparison of monomeric and dimeric inhibitors pro-
vides evidence of bivalent binding of a dimeric inhibitor,
consistent with the presence of two active sites in close
proximity. These results suggest that polymerizing trans-
ferases such as chitin synthase are mechanistically
distinct from the comparatively well-understood single-
sugar transferases.2,3 In the case of chitin synthase, this
mechanistic divergence has important implications for
the design of new antifungal therapeutic agents.
Ba ck gr ou n d . Chitin synthase converts cytosolic uri-
dine diphosphoryl-N-acetylglucosamine (UDP-GlcNAc) to
chitin, poly-â(1,4)-GlcNAc (Figure 1), with simultaneous
extrusion of the polymer through the membrane. Chitin
is an integral structural component of the fungal cell wall
but is not found in mammals, and thus CS is an
attractive therapeutic target. However, like other poly-
merizing transferases, CS is an integral membrane
protein (for which no crystal structure has been deter-
mined) and exhibits relatively low affinity for its sub-
strate (KM ≈ 1 mM).1,4 These factors preclude the effective
F IGURE 1. Chitin synthase polymerizes UDP-GlcNAc to form
chitin.
development of inhibitors by structure-based design or
the synthesis of substrate analogues.5-7
Mech a n ism -Ba sed In h ibit ion ? In addressing this
problem, we were drawn to the long-standing debate as
to whether polymerizing transferases such as CS possess
two active sites and are thus mechanistically distinct
(5) For previous studies on rationally designed UDP-GlcNAc ana-
logues, see: (a) Behr, J .-B.; Gourlain, T.; Helimi, A.; Guillerm, G.
Bioorg. Med. Chem. Lett. 2003, 13, 1713-1716. (b) Xie, J .; Thellend,
A.; Becker, H.; Vidal-Cros, A. Carbohydr. Res. 2001, 334, 177-182.
(c) Behr, J .-B.; Gautier-Lefebvre, I.; Mvondo-Evina, C.; Guillerm, G.;
Ryder, N. S. J . Enzyme Inhib. 2001, 16, 107-112. (d) Grugier, J .; Xie,
J .; Duarte, I.; Valery, J .-M. J . Org. Chem. 2000, 65, 979-984. (e)
Schafer, A.; Thiem, J . J . Org. Chem. 2000, 65, 24-29. (f) Obi, K.; Uda,
J .-I.; Iwase, K.; Sugimoto, O.; Ebisu, H.; Matsuda, A. Bioorg. Med.
Chem. Lett. 2000, 10, 1451-1454.
(6) Polyoxins and nikkomycins are the predominant known natural
product inhibitors of chitin synthase. For a review, see: Zhang, D.;
Miller, M. Curr. Pharm. Des. 1999, 5, 73.
(7) Crystal structures of several single-sugar glycosyltransferases
have recently been determined. For discussion, see: (a) Davies, G. J .
Nature Struct. Biol. 2001, 8, 98-100. (b) Charnok, S. J .; Henrissat,
B.; Davies, G. J . Plant Physiol. 2001, 125, 527-531. (c) Thomas, S.;
Yen, T.-Y.; Macher, B. A. Glycobiology 2002, 12, 4G-7G.
(1) Representative overviews of fungal chitin synthase (EC
2.4.1.16): (a) Munro, C. A.; Gow, N. A. R. Med. Mycol. 2001, 39, 41-
53. (b) Valdivieso, M. H.; Duran, A.; Roncero, C. EXS 1999, 87, 55-
69. (c) Merz, R. A.; Horsch, M.; Nyhlen, L. E.; Rast, D. M. EXS 1999,
87, 9-37. (d) Bulawa, C. E. Annu. Rev. Microbiol. 1993, 47, 505-534.
(e) Cabib, E. Advances Enzymol. 1987, 59, 59-101.
(2) For an overview of common glycosyltransferases, see: Essentials
of Glycobiology; Varki, A., Cummings, R., Esko, J ., Freeze, H., Hart,
G., Marth, J ., Eds.; Cold Spring Harbor Laboratory Press: Spring
Harbor, NY, 1999.
(3) The term “polymerizing glycosyltransferase” is used here to refer
to processive transferases that synthesize polymeric sugars without
releasing the growing chain and the term “single-sugar transferases”
to refer to transferases that add a single sugar to an acceptor substrate
and release the glycosylated product.
(4) We limit our discussion to fungal CS, as much less is known
about CS in insects and other organisms.
10.1021/jo035100c CCC: $27.50 © 2004 American Chemical Society
Published on Web 01/03/2004
J . Org. Chem. 2004, 69, 613-618
613

Yeager and Finney
F IGURE 4. Potential bivalent inhibitors and a monomeric
control.
F IGURE 2. Two-active site mechanism for chitin synthase.
F IGURE 3. Proposed two-site mechanism can be probed with
bivalent inhibitors.
from single-sugar transferases.8 At the origin of this
debate is the extended structure of polysaccharides such
as chitin, in which adjacent sugar residues have opposed
orientations (Figure 1). It has been proposed that CS has
two active sites, one for each sugar orientation, in order
to facilitate the simultaneous polymerization and extru-
sion of the linear polymer (Figure 2). This proposal has
been pointedly debated, with arguments for and against
based on the same indirect evidence: primarily protein
sequence analysis but also mutagenesis and the crystal
structures of several single-sugar glycosyl-transferases.9
Significantly, no direct test of this hypothesis has been
reported.
A defining characteristic of the two-active site hypoth-
esis is that it is chemically testable: if there are two
UDP-GlcNAc binding sites in close proximity, then ap-
propriate dimeric nucleoside inhibitors should show
bivalent inhibition and should be more potent than the
corresponding monomers (Figure 3).10
Biva len t In h ibitor s of Ch itin Syn th a se. Potential
bivalent inhibitors were constructed on the basis of 5′-
deoxy-5′-aminouridine fragments connected by ethylene
glycol-based carbamate linkers (Figure 4).11 Compounds
1-4 vary in maximum uridine-uridine 5′-5′ distance
F IGURE 5. Synthesis of 1-5.
from 14 to 22 Å, which spans the calculated maximum
possible separation (∼18 Å) for two active sites.12 Com-
pound 5 is similar in structure to 1-4 but is a methyl-
terminated monomer and serves as a control. All five
compounds were prepared in five steps from uridine
(Figure 5, Chart 1).
Inhibition of CS activity by 1-5 was determined using
an established assay based on incorporation of 3H-UDP-
GlcNAc in precipitated chitin.13 The shortest dimers, 1
and 2, exhibited significant inhibition of chitin synthase
at 1 mM (32 and 45%, respectively, Figure 6). The longer
(8) For one of the first formal statements of the two-active site
hypothesis, see: (a) Saxena, I. M.; Brown, R. M., J r.; Fevre, M.;
Geremia, R. A.; Henrissat, B. J . Bacteriol. 1995, 177, 1419-1424. For
representative further discussion, see: (b) Saxena, I. M.; Brown, R.
M., J r.; Dandekar, T. Phytochemistry 2001, 57, 1135-1148. (c) Saxena,
I. M.; Brown, R. M., J r. Curr. Opp. Plant Biol. 2000, 3, 523-531.
(9) For a summary of arguments against the two-site mechanism,
see refs 7a,b and: Kamst, E.; Spaink, P. H. Trends Glycosci. Glycotech.
1999, 11, 187-199.
(12) Maximum 5′-5′ distances were calculated using ChemBats3D
3.0 (Cambridgesoft, Cambridge, MA). The maximum distance was
determined by constructing
a model of the transition state for
condensation of two molecules of UDP-GlcNAc and manually adjusting
dihedral angles to attain the most extended conformation possible. The
ethylene glycol linkers are sufficiently flexible that they do not
constrain the relative orientation of the uridine residues.
(13) Orlean, P. J . Biol. Chem. 1987, 262, 5732-5739. 100% activity/
0% inhibition is typically ≈ 20 000 cps/h. A time-independent back-
ground of ≈ 500 cps/h is observed, equivalent to ∼2.5% apparent
inhibition. Reported values are the average of g3 measurements.
Experimental variation is consistently (5%.
(10) For a recent review of polyvalency in protein-carbohydrate
association, see: Kiessling, L. L.; Young, T.; Mortell, K. H. Glycoscience
2001, 2 1817-1861. For reviews of equally relevant bisubstrate
analogue inhibition strategies, see: (a) Broom, A. D. J . Med. Chem.
1989, 32, 2-7. (b) Radzicka, A.; Wolfenden, R. In Methods in Enzymol-
ogy; Academic Press: New York, 1995; pp 284-312.
(11) See Experimental Section. All relevant 1H NMR spectra can
be found in Supporting Information.
614 J . Org. Chem., Vol. 69, No. 3, 2004

Two-Active Site Mechanism for Chitin Synthase
Con clu sion
In conclusion, we have presented the first chemical
evaluation of a two-active site mechanism for chitin
synthase and have provided evidence that supports this
mechanism. While it remains impossible to prove a
mechanism, the alternative single-site mechanism pro-
vides no explanation for our observations. These observa-
tions have several important implications. First, it
suggests that dimerization of more potent monomeric
inhibitors should lead to high-affinity inhibitors of CS.
Second, it provides for the first time a potential strategy
by which CS could be selectively targeted in the presence
of the numerous mammalian single-sugar GlcNac trans-
ferases. Finally, it suggests that other important poly-
merizing transferases such as the cellulose, heparin, and
hyaluronan synthases may also function by a two-site
mechanism and could be selectively inhibited by this
same strategy. Efforts are underway to identify superior
monomers for dimerization as well as to develop other
mechanistic probes for chitin synthase.
F IGURE 6. Inhibitory activity of dimers 1-4 and monomer
5.
CHART 1. Lin k er P r ecu r sor s for th e Syn th esis of
1-5
Exp er im en ta l Section .
For general experimental information, see Supporting In-
formation.
5′-Deoxy-5′a m in ou r id in e Dim er 1. Silylated dimer 14
(0.46 g, 0.43 mmol) was dissolved in 3:1:1 AcOH/THF/H2O (5
mL) and allowed to stir at room temperature for 9 h. The
reaction was concentrated to give dimer 1 (0.25 g, 98%) as a
white solid. Dimer 1 was purified by reverse-phase silica gel
chromatography (10% CH3CN/H2O) prior to use in chitin
dimers, 3 and 4, and the monomeric control, 5, showed
minimal activity at the same concentration. Notably,
doubling the concentration of 5 ([5] ) 2 mM) did not lead
to a significant increase in inhibition, indicating that 1
and 2 are not better inhibitors simply by virtue of double
the effective concentration of uridine derivative. Inhibi-
tion by dimers 1 and 2 and monomer 5 was measured
over 0.05-20 mM, allowing determination of IC50 ) 2.2
mM for 1, IC50 ) 1.1 mM for 2 and IC50 ) 11.8 mM for
5.14 At [2] ) 1 mM, increasing [UDP-GlcNAc] to 15 mM
led to full recovery of enzymatic activity (Vmax), indicating
that inhibition by 1 is competitive.15,16
1
synthase. H NMR (400 MHz, CD3OD): δ 3.36 (m, 4H), 3.98
(s, 4H), 4.18 (t, 2H, J ) 5.2 Hz), 4.23 (s, 4H), 5.71 (d, 2H, J )
8.0 Hz), 5.77 (d, 2H, J ) 4.8 Hz), 7.64 (d, 2H, J ) 8.0 Hz). 13C
NMR (100 MHz, CD3OD): δ 42.2, 64.2, 70.8, 73.8, 82.8, 90.7,
102.7, 142.3, 151.8, 158.8, 166.4. IR (KBr): ν 3336 (br), 1728
(s), 1553 (s), 1475 (s), 1396 (s), 1274 (s), 1073 (s), 776 (s). HRMS
calcd for C22H28N6O14Na (MNa+) 623.1556, found 623.1575. Rf
) 0.32 (15% water/acetonitrile).
5′-Deoxy-5′-a m in ou r id in e Dim er 2. Silylated dimer 15
(0.24 g, 0.22 mmol) was dissolved in 3:1:1 AcOH/THF/H2O (25
mL) and allowed to stir at room temperature for 9 h. The
reaction was concentrated to give dimer 2 (0.14 g, 99%) as a
white solid. Dimer 2 was purified by reverse-phase silica gel
chromatography (10% CH3CN/H2O) prior to use in chitin
That dimers 1 and 2 are an order of magnitude more
potent than 5 and 3 or 4 is consistent with the two-active
site model for CS. Comparing 1 or 2 and 5, the enhanced
inhibition may be ascribed to simultaneous binding of the
two uridine groups by two active sites. In compounds
1-4, the only difference is the distance between the two
covalently linked uridine groups. Thus, the inhibitory
nature of 1 and 2 is consistent with the idea that these
shorter linkers best match the inter-active site distance
in CS while the linkers in 3 and 4 are too long. In
contrast, the single-site mechanism fails to offer a
rationale for either the increased efficacy of dimeric
inhibitors relative to a monomer or for the tether length
dependence observed in the dimeric inhibitors.17
1
synthase. H NMR (400 MHz, D2O): δ 3.34 (m, 4H), 3.61 (s,
4H), 3.93 (m, 2H), 3.97 (m, 2H), 4.05 (m, 4H), 4.18 (m, 2H),
5.65 (m, 2H), 5.72 (d, 2H, J ) 8.0 Hz), 7.50 (d, 2H, J ) 8.0
Hz). 13C NMR (100 MHz, CD3OD): δ 43.5, 65.2, 70.5, 72.1,
74.8, 84.2, 91.4, 102.8, 142.7, 152.1, 158.9, 165.9. IR (KBr):
ν 3388, 1702, 1545, 1457, 1265 cm-1. HRMS: calcd for
C
24H32N6O15Na (M - Na+) 667.1823, found 667.1805. Rf ) 0.50
(20% H2O/CH3CN).
5′-Deoxy-5′-a m in ou r id in e Dim er 3. Silylated dimer 16
(1.73 g, 1.51 mmol) was dissolved in 3:1:1 AcOH/THF/H2O (50
mL) and allowed to stir at room temperature for 5 h. The
mixture was concentrated to give dimer 3 (1.01 g, 98%) as a
white solid. Dimer 3 was purified by reverse-phase silica gel
chromatography (10% CH3CN/H2O) prior to use in chitin
1
synthase assays. H NMR (400 MHz, D2O): δ 3.30 (dd, 2H, J
(14) IC50 values were determined by nonlinear least-squares fitting
of % inhibition vs log [inhibitor] plots using Prism3CX (Graphpad, Inc.,
San Diego, CA). Inhibitor concentrations were spaced evenly on the
log [inhibitor] axis to minimize residual errors in fitting. Complete
details are provided in Supporting Information.
) 6.0, 14.8 Hz), 3.40 (dd, 2H, J ) 3.2 Hz), 3.55 (s, 4H), 3.61
(m, 4H), 3.94 (m, 2H), 3.99 (t, 2H, J ) 5.4 Hz), 4.06 (m, 4H),
4.2 (t, 2H, J ) 4.6 Hz), 5.67 (d, 2H, J ) 4.0 Hz), 5.74 (d, 2H,
(15) Segel, I. H. Enzyme Kinetics, J ohn Wiley & Sons: New York,
1975. See Supporting Information for details.
(16) Intra- and interprotein variants of the two-site mechanism
cannot be distinguished by the approach described here. However, the
linker length dependence argues against bivalent intermolecular
binding.
(17) It is possible that the enhanced binding could reflect adventi-
tious association of the second uridine with a portion of the enzyme
that normally binds GlcNAc. While this possibility cannot be excluded,
three factors weigh against it: (1) GlcNAc itself exhibits no inhibitory
activity; (2) uridine and GlcNAc are not particularly similar in
structure; and (3) GlcNAc is present at 40 mM in the assay.
J . Org. Chem, Vol. 69, No. 3, 2004 615

Yeager and Finney
8.0 Hz), 7.53 (d, 2H, J ) 8.4 Hz). 13C NMR (100 MHz,
CD3OD): δ 43.6, 65.3, 70.5, 71.5, 71.5, 72.2, 74.7, 84.2, 91.5,
102.8, 142.7, 152.1, 158.9, 165.8. IR (KBr): ν 3380, 3091, 2951,
1693, 1562, 1466, 1274 cm-1. HRMS calcd for C26H36N6O15Na
(M - Na+) 711.2085, found 711.2087. Rf ) 0.65 (20% H2O/
CH3CN).
Bis(silyl eth er ) 7. Azide 6 (2.04 g, 7.58 mmol) was
azeotropically dried with THF and dissolved in dry pyridine
(12 mL) under N2. DMAP (0.01 g) was added; the solution was
cooled to 0 °C, and TES-Cl (7.30 mL, 45.48 mmol) was added.
After stirring for 6 h, the reaction mixture was concentrated
and purified by silica gel chromatography (40% EtOAc/
hexanes) to afford bis(silyl ether) 7 (3.50 g, 93%) as a white
foam. 1H NMR (400 MHz, CDCl3): δ 0.61-0.65 (m, 12H), 0.94
(m, 18H), 3.60 (dd, 1H, J ) 3.4, 13.4 Hz), 3.80 (dd, 1H, J )
3.4, 13.4 Hz), 3.96 (t, 1H, J ) 5.2 Hz), 4.12 (m, 1H,), 4.20 (t,
1H, J ) 3.8 Hz), 5.64 (d, 1H, J ) 3.2 Hz), 5.77 (d, 1H, J ) 8.0
Hz), 7.66 (d, 1H, J ) 8.4 Hz), 9.51 (s, 1H). 13C NMR (100 MHz,
CDCl3): δ 4.9, 4.9, 6.8, 6.9, 51.2, 71.2, 75.1, 81.3, 90.8, 102.2,
140.1, 149.9, 163.2. IR (Film): ν 2960, 2890, 2113, 1457, 1387,
1169 cm-1. HRMS calcd for C21H39N5O5Si2Na (M - Na+)
520.2382, found 520.2372. Rf ) 0.76 (40% EtOAc/hexanes).
5′-Deoxy-5′-a m in ou r id in e (8). To a solution of bis(silyl
ether) 7 (4.84 g, 9.72 mmol) in CH3OH (20 mL) was added
10% w/w Pd/C (0.09 g). The flask was evacuated and back-
filled with H2 three times and allowed to stir under H2 (1 atm)
for 7 h. The reaction was filtered through Celite (rinsing with
CH3OH) and concentrated to provide 5′-deoxy-5′-aminouridine
(8) (4.09 g, 89%) as a white foam. When necessary, further
purification was effected by silica gel chromatography (8-15%
CH3OH/CH2Cl2). 1H NMR (400 MHz, CD3OD): δ 0.59-0.71
(m, 12H), 0.95 (t, 9H, J ) 7.8 Hz), 1.00 (t, 9H, J ) 7.8 Hz),
2.87 (m, 2H), 3.98 (m, 1H), 4.03 (t, 1H, J ) 3.8 Hz), 4.41 (t,
1H, J ) 5.4 Hz), 5.74 (d, 1H, J ) 8.0 Hz), 5.81 (d, 1H, J ) 6.0
Hz), 7.74 (d, 1H, J ) 8.0 Hz). 13C NMR (100 MHz, CD3OD): δ
5.8, 6.0, 7.2, 7.3, 44.7, 74.6, 75.4, 87.2, 91.2, 103.0, 143.3, 152.2,
165.9. IR (KBr): ν 3388, 2969, 2873, 1693, 1475, 1379, 1265,
1151 cm-1. HRMS calcd for C21H41N3O5Si2Na (M - Na+)
494.2477, found 494.2484. Rf ) 0.37 (15% CH3OH/CH2Cl2).
5′-Deoxy-5′-a m in ou r id in e Dim er 4. Silylated dimer 17
(1.26 g, 1.06 mmol) was dissolved in 3:1:1 AcOH/THF/H2O (50
mL) and allowed to stir at room temperature for 5 h. The
mixture was concentrated to give dimer 4 (0.78 g, 99%) as a
white solid. Dimer 4 was purified by reverse-phase silica gel
chromatography (10% CH3CN/H2O) prior to use in chitin
1
synthase. H NMR (400 MHz, D2O): δ 3.30 (m, 2H), 3.40 (m,
2H), 3.54 (m, 12H), 3.60 (t, 4H, J ) 3.8 Hz), 3.96 (m, 4H), 4.07
(m, 4H), 4.20 (t, 2H, J ) 5.0 Hz), 5.67 (d, 2H, J ) 4.0 Hz),
5.74 (d, 2H, J ) 8.0 Hz), 7.553 (d, 2H, J ) 8.4 Hz). 13C NMR
(100.594 MHz, CD3OD): δ 43.6, 65.3, 70.5, 71.5, 71.5, 72.2,
74.7, 84.2, 91.4, 102.8, 142.7, 152.1, 158.9, 165.8. IR (KBr): ν
3401, 3107, 2943, 1739, 1543, 1469, 1395, 1264 cm-1. HRMS
calcd for C28H40N6O17Na (M - Na+) 755.2347, found 755.2334.
Rf ) 0.50 (20% H2O/CH3CN).
5′-Deoxy-5′-a m in ou r id in e Mon om er 5. Silylated mono-
mer 18 (0.91 g, 1.47 mmol) was dissolved in 3:1:1 AcOH/THF/
H2O (5 mL) and allowed to stir at room temperature for 20 h.
The mixture was concentrated and purified by silica gel,
chromatography (15% CH3OH/CH2Cl2) to give monomer 5 (0.47
g, 82%) as a white solid. Monomer 5 was purified by reverse-
phase silica gel chromatography (10% CH3CN/H2O) prior to
use in chitin synthase. 1H NMR (400 MHz, D2O): δ 3.22 (s,
3H), 3.32 (dd, 1H, J ) 5.8, 15.2 Hz), 3.41 (dd, 1H, J ) 4.0,
21.6 Hz), 3.46 (m, 3H), 3.55 (m, 2H), 3.60 (t, 2H, J ) 4.0 Hz),
3.97 (m, 2H), 4.09 (m, 2H), 4.21 (t, 1H, J ) 4.4 Hz), 5.69 (d,
1H, J ) 4.4 Hz), 5.75 (s, 1H, 8.0 Hz), 7.54 (s, 1H, J ) 8.4 Hz).
13C NMR (100 MHz, CD3OD): δ 43.5, 65.2, 70.5, 71.3, 72.1,
74.7, 84.2, 91.4, 102.8, 142.7, 152.1, 158.9, 165.8. IR (KBr): ν
3397, 3318, 3205, 3065, 2978, 2917, 1684, 1553, 1483, 1335,
1256 cm-1. HRMS calcd for C15H23N3O9Na (M - Na+) 412.1332,
found 412.1331. Rf ) 0.58 (20% H2O/CH3CN).
5′-Deoxy-5′-a zid ou r id in e (6). A solution of hydrazoic acid
(HN3) was prepared by the slow addition of 14.4 mL of sulfuric
acid to a suspension of NaN3 (33.8 g, 520 mmol) in 16.9 mL
H2O and 206 mL of toluene at 0 °C. (Caution: HN3 and NaN3
are toxic and potentially explosive. The reaction temperature
must be maintained at or below 0 °C during addition.) The
mixture was stirred for 30 min, at which time Na2SO4
precipitated. (Note: the precipitate may hinder stirring.) The
toluene layer was decanted, dried over Na2SO4, and stored over
Na2SO4 until used. Uridine (5.00 g, 20.48 mmol) was sus-
pended in 350 mL of dry THF under an atmosphere of
nitrogen. Triphenylphosphine (16.53 g, 61.40 mmol) and 125
mL of fresh HN3 (∼2.5 M in toluene) were added, and the
mixture was cooled to 0 °C. Diethylazodicarboxylate (DEAD,
10.92 mL, 61.40 mmol) was slowly added dropwise, and the
reaction was allowed to warm to room temperature. After 15
h, the reaction was concentrated under reduced pressure.
(Note: the temperature was maintained at or below 45 °C
during concentration.) The resulting viscous solid was dis-
solved in minimal CH2Cl2 (∼500 mL) and extracted into water
(2 × 200 mL). After washing with with CH2Cl2 (6 × 300 mL),
the product was obtained by concentrating the aqueous phase.
(Note: the temperature was maintained at or below 45 °C
during evaporation.) Column chromatography (10-15% CH3OH/
CH2Cl2) afforded 6 (4.95 g, 90% yield) as a white foam. 1H NMR
(400 MHz, CD3OD): δ 3.59 (dd, 1H, J ) 4.6, 13.4 Hz), 3.68
(dd, 1H, J ) 3.0, 13.4 Hz), 4.04 (m, 1H), 4.09 (t, 1H, J ) 5.6
Hz), 4.21 (t, 1H, J ) 5.0 Hz), 5.73 (d, 1H, J ) 8.0 Hz), 5.83 (d,
1H, J ) 4.4 Hz), 7.72 (d, 1H, J ) 8.0 Hz). 13C NMR (100 MHz,
CD3OD): δ 53.5, 72.1, 75.2, 84.1, 92.1, 103.4, 142.9, 152.5,
166.3. IR (KBr): ν 3231, 2104, 1728,1667, 1483 cm-1. HRMS
calcd for C9H12N5O5 (M + H+) 270.0833, found 270.0833. Rf )
0.51 (15% CH3OH/CH2Cl2).
Bis(p-n itr op h en yl ca r bon a te) 9. Ethylene glycol (0.13
mL, 2.07 mmol) was azeotropically dried with THF and
dissolved in dry pyridine (2 mL) under N2. The reaction was
cooled to 0 °C and p-nitrophenylchloroformate (1.30 g, 6.21
mmol) was added. After 40 h, the reaction was concentrated
and purified by silica gel chromatograpy (2% EtOAc/hexanes)
1
to yield 9 (0.38 g, 47%) as a white solid. H NMR (400 MHz,
CDCl3): δ 4.60 (s, 4H), 7.37 (d, 4H, J ) 5.2 Hz), 8.26 (d, 4H,
J ) 6.8 Hz). 13C NMR (100 MHz, CDCl3): δ 66.1, 121.6, 125.2,
145.3, 152.1, 155.0. IR (Film): ν 3677 (br), 3135 (s), 3109 (s),
1781 (s), 1623 (s), 1588 (s), 1527 (s), 1483 (s), 1352 (s), 1213
(s), 863 (s) cm-1. HRMS calcd for C16H12N2O10 (M+) 392.0492,
found 392.0509. Rf ) 0.64 (5% EtOAc/CH2Cl2).
Bis(p-n itr op h en yl ca r bon a te) 10. Diethylene glycol (0.40
mL, 4.21 mmol) was azeotropically dried with THF and
dissolved in dry pyridine (3 mL) under N2. The reaction was
cooled to 0 °C and p-nitrophenyl chloroformate (2.04 g, 10.10
mmol) was added. After 40 h, the reaction was concentrated
and purified by silica gel chromatography silica (2% EtOAc/
1
hexanes) to yield 10 (0.84 g, 46%) as a white solid. H NMR
(400 MHz, CDCl3): δ 3.85 (m, 4H), 4.47 (m, 4H), 7.38 (d, 4H,
J ) 8.0 Hz), 8.25 (d, 4H, J ) 8.0 Hz). 13C NMR (100.594 MHz,
CDCl3): δ 67.9, 68.6, 121.6, 125.1, 145.2, 152.2, 155.1. IR
(Film): ν 3677, 3135, 3083, 2969, 1781, 1623, 1588, 1527, 1483,
1352, 1221 cm-1. HRMS calcd for C18H16N2O11Na (M - Na+)
459.0646, found 459.0642. Rf ) 0.52 (5% EtOAc/CH2Cl2).
Bis(p-n itr op h en yl ca r bon a te) 11. Triethylene glycol (1.00
mL, 7.49 mmol) was azeotropically dried with THF, dissolved
in dry CH2Cl2 (5 mL), and cooled to 0 °C under N2. Pyridine
(2.40 mL, 30.00 mmol) and p-nitrophenyl chloroformate (3.80
g, 18.70 mmol) were added. After stirring for 24 h, the reaction
mixture was concentrated and purified by silica gel chroma-
tography (3-7% EtOAc/hexanes) to provide 11 (2.93 g, 81%)
1
as a white solid. H NMR (400 MHz, CDCl3): δ 3.73 (s, 4H),
3.83 (m, 4H), 4.45 (m, 4H), 7.37 (d, 4H, J ) 8.4 Hz), 8.26 (d,
4H, J ) 8.4 Hz). 13C NMR (100 MHz, CDCl3): δ 68.2, 68.6,
68.7, 70.7, 121.6, 125.1, 145.2, 152.2, 155.2. IR (Film): ν 3677,
3117, 3083, 2873, 1772, 1623, 1588, 1536, 1492, 1352, 1231
616 J . Org. Chem., Vol. 69, No. 3, 2004

Two-Active Site Mechanism for Chitin Synthase
cm-1. HRMS calcd for C20H20N2O12Na (M - Na+) 503.0908,
found 503.0914. Rf ) 0.44 (5% EtOAc/CH2Cl2).
1265, 1169, 1073 cm-1. HRMS calcd for C48H88N6O15Si4Na (M
- Na+) 1123.5277, found 1123.5277. Rf ) 0.4 (10% CH3OH/
CH2Cl2).
Bis(p-n itr op h en yl ca r bon a te) 12. Tetraethylene glycol
(0.51 mL, 2.63 mmol) was azeotropically dried with THF,
dissolved in dry CH2Cl2 (8 mL), and cooled to 0 °C under N2.
Pyridine (0.85 mL, 10.50 mmol) and p-nitrophenyl chloro-
formate (0.99 g, 6.58 mmol) were added. After stirring at room
temperature for 21 h the reaction was concentrated and
purified by silica gel chromatography (7-10% EtOAc/hexanes)
to afford 12 (0.85 g, 61%) as a white solid. 1H NMR (400 MHz,
CDCl3): δ 3.67 (s, 8H), 3.79 (m, 4H), 4.41 (m, 4H), 7.38 (d,
4H, J ) 9.2 Hz), 8.28 (d, 4H, J ) 9.2 Hz). 13C NMR (100 MHz,
CDCl3): δ 68.2, 68.5, 70.5, 70.6, 121.6, 125.0, 145.0, 152.1,
155.2. IR (Film): ν 3135, 3083, 2917, 1772, 1615, 1588, 1527,
1492, 1352, 1265, 1221 cm-1. HRMS calcd for C22H24N2O13Na
(M - Na+) 547.1171, found 547.1158. Rf ) 0.32 (10% EtOAc/
CH2Cl2).
Silyla ted Dim er 16. 5′-Deoxy-5′-aminouridine (8) (0.94 g,
1.95 mmol) and bis(p-nitrophenyl carbonate) 11 (2.20 g, 4.70
mmol) were azeotropically dried with THF and dissolved in
dry CH2Cl2 (8 mL) under N2. Triethylamine (1.10 mL, 4.10
mmol) and DMAP (0.01 g, 0.10 mmol) were added, and the
reaction mixture was allowed to stir for 20 h. It was then
diluted with CH2Cl2 (200 mL), washed with water (6 × 100
mL), dried over Na2SO4, and concentrated, and the resulting
solid was purified by silica gel chromatography (3% CH3OH/
CH2Cl2) to provide silylated dimer 16 (1.73 g, 78%) as a white
solid. 1H NMR (400 MHz, CDCl3): δ 0.48-0.63 (m, 24H), 0.86-
0.96 (m, 36H), 3.41 (s, 4H), 3.63 (m, 8H), 3.95 (s, 2H), 4.03 (s,
2H), 4.20 (m,4H), 4.54 (s, 2H), 5.43 (d, 2H, J ) 5.6 Hz), 5.73
(d, 2H, J ) 8.0 Hz), 5.84 (s, 2H), 7.31 (d, 2H, J ) 8.0 Hz), 9.79
(s, 2H). 13C NMR (100 MHz, CDCl3): δ 4.7, 4.8, 4.9, 5.0, 6.7,
6.8, 6.9, 6.9, 42.8, 64.3, 69.4, 70.4, 72.9, 73.0, 84.2, 93.5, 102.3,
142.8, 150.0, 156.5, 163.3. IR (Film): ν 3371, 3109, 2934, 2873,
1693, 1553, 1466, 1387, 1256 cm-1. HRMS calcd for C50H92N6O16-
Si4Na (M - Na+) 1167.5539, found 1167.5520. Rf ) 0.65 (15%
CH3OH/CH2Cl2).
Silyla ted Dim er 17. 5′-Deoxy-5′-aminouridine (8) (1.59 g,
3.36 mmol) and bis(p-nitrophenyl carbonate) 12 (0. 85 g, 1.60
mmol) were azeotropically dried with THF and dissolved in
dry CH2Cl2 (11 mL) under N2. Triethylamine (0.98 mL, 3.52
mmol) and DMAP (0.01 g, 0.10 mmol) were added, and the
reaction was allowed to stir for 20 h. It was then diluted with
CH2Cl2 (400 mL), washed with water (6 × 300 mL), dried over
Na2SO4, and concentrated, and the resulting solid was purified
by silica gel chromatography (3% CH3OH/CH2Cl2) to afford
silylated dimer 17 (2.21 g, 61%) as a white solid. 1H NMR (400
MHz, CDCl3): δ 0.49-0.67 (m, 24H), 0.87-0.99 (m, 36H), 3.42
(s, 4H), 3.65 (m, 12H), 3.96 (s, 2H), 4.05 (s, 2H), 4.21 (m,4H),
4.54 (s, 2H), 5.42 (d, 2H, J ) 5.6 Hz), 5.75 (d, 2H, J ) 8.0 Hz),
5.90 (s, 2H), 7.31 (d, 2H, J ) 8.0 Hz), 9.83 (s, 2H). 13C NMR
(100 MHz, CDCl3): δ 4.7, 4.9, 4.9, 6.7, 6.9, 6.9, 42.7, 64.3, 69.4,
70.5, 70.6, 72.9, 73.0, 84.3, 93.6, 102.3, 142.8, 150.1, 156.6,
163.3. IR (Film): ν 3321, 3214, 3065, 2951, 2890, 1719, 1553,
1457, 1379, 1256, 1169 cm-1. MS calcd for C52H96N6O17Si4Na
(M - Na+) 1211.5801, found 1211.5816. Rf ) 0.42 (10%
CH3OH/CH2Cl2).
Silyla ted Mon om er 18. 5′-Deoxy-5′-aminouridine (8) (0.22
g, 0.46 mmol) was dissolved in CH2Cl2 (3 mL) under N2, and
triethylamine (0.14 mL, 0.50 mmol) and DMAP (0.01 g, 0.03
mmol) were added. p-Nitrophenyl carbonate 13 (0.14 g, 0.5
mmol) in dry pyridine (5 mL) was added, and the reaction
mixture was allowed to stir for 18 h, after which it was
concentrated and purified by silica gel chromatography (5%
CH3OH/CH2Cl2) to yield silylated monomer 18 (0.24 g, 85%)
as a white solid. 1H NMR (400 MHz, CDCl3): δ 0.51-0.65 (m,
12H), 0.63-0.96 (m, 18H), 3.36 (s, 3H), 3.42 (m, 2H), 3.53 (s,
2H), 3.62 (m, 4H), 3.93 (t, 1H, J ) 4.0 Hz), 4.03 (s, 1H), 4.23
(m, 2H), 4.49 (t, 1H, J ) 5.2 Hz), 5.44 (d, 2H, J ) 5.6 Hz),
5.60 (s, 1H), 5.74 (d, 2H, J ) 8.0 Hz), 7.30 (d, 2H, J ) 8.4 Hz).
13C NMR (100 MHz, CDCl3): δ 4.7, 4.9, 6.7, 6.9, 42.9, 59.0,
64.2, 69.4, 70.3, 71.8, 73.0, 73.1, 84.0, 93.3, 102.3, 142.3, 149.9,
156.5, 163.2. IR (Film): ν 3318, 1763, 1615, 1536, 1352, 1256,
1213 cm-1. HRMS calcd for C27H51N3O9Si2Na (M - Na+)
640.3056, found 640.3085. Rf ) 0.4 (10% CH3OH/CH2Cl2).
Ch itin Syn th a se Assa y. The assay protocol used is based
on the procedure of Orlean,13 modified after helpful discussions
with Prof. Peter Orlean (University of Illinois) and Dr. Enrico
Cabib (NIH).18 Saccharomyces cerevisiae strains (PP-1D, wild
type) were kindly provided by Professor Orlean and stored at
-70 °C on freezer stabs. Active yeast cultures were temporarily
maintained on agar plates, stored at 4 °C. Cells were cultured
in 200 mL YEPG (1% yeast extract, 2% bactopeptone, 2%
glucose) medium at 30 °C and allowed to grow to saturation.
p-Nitr op h en yl Ca r bon a te 13. 2-(2-Methoxyethoxy)etha-
nol (0.77 mL, 6.42 mmol) was azeotropically dried with THF
and dissolved in dry CH2Cl2 (6 mL) under N2. Triethylamine
(1.30 mL, 9.63 mmol) and p-nitrophenyl chloroformate (1.45
g, 9.63 mmol) were added. After stirring at room temperature
for 3 h, the reaction was concentrated and purified by silica
gel chromatography (4-5% EtOAc/hexanes) to afford 13 (1.62
1
g, 88%) as a white solid. H NMR (400 MHz, CDCl3): δ 3.37
(s, 3H), 3.56 (t, 2H, J ) 4.8 Hz), 3.66 (t, 2H, J ) 4.8 Hz), 3.79
(t, 2H, J ) 4.8 Hz), 4.43 (t, 2H, J ) 4.4 Hz), 7.36 (d, 2H, J )
8.8 Hz), 8.25 (d, 2H, J ) 8.8 Hz). 13C NMR (100 MHz, CDCl3):
δ 29.7, 59.0, 68.2, 68.6, 70.5, 71.8, 121.6, 125.1, 145.1, 152.2,
155.2. IR (Film): ν 3659, 3528, 3135, 3083, 2899, 1772, 1623,
1588, 1536, 1457, 1352, 1221 cm-1. HRMS calcd for C12H15
-
NO7Na (M - Na+) 308.0741, found 308.0745. Rf ) 0.35 (5%
EtOAc/CH2Cl2).
Silyla ted Dim er 14. 5′-Deoxy-5′-aminouridine (8) (0.88 g,
1.60 mmol) and bis(p-nitrophenyl carbonate) 9 (0.29 g, 0.73
mmol) compounds were azeotropically dried with THF and
dissolved in dry CH2Cl2 (3 mL) under N2. Triethylamine (0.22
mL, 1.60 mmol), and (dimethylamino)pyridine (0.01 g, 0.07
mmol) were added, and the reaction mixture was allowed to
stir for 20 h, after which it was concentrated and dissolved in
CH2Cl2 (500 mL). After washing with aqueous NaHCO3
(saturated, 2 × 300 mL) and H2O (1 × 300 mL), the organic
phase was dried over Na2SO4 and concentrated and purified
by silica gel chromatography (2-4% CH3OH/CH2Cl2) to yield
silylated dimer 14 (0.52 g, 31%) as a white solid. H1 NMR (400
MHz, CDCl3): δ 0.54-0.68 (m, 24H), 0.93 (t, 18H, J ) 7.6 Hz),
0.98 (t, 18H, J ) 7.6 Hz), 1.28 (t, 4H, J ) 7.2 Hz), 3.49 (s, 2H),
4.04 (s, 2H), 4.22 (m, 4H), 4.71 (s, 2H), 5.25 (d, 2H, J ) 5.3
Hz), 5.78 (d, 2H, J ) 8.0 Hz), 7.24 (d, 2H, J ) 7.6 Hz). 13C
NMR (100 MHz, CDCl3): δ 4.8, 5.0, 6.8, 7.0, 42.0, 62.3, 63.5,
72.5, 72.6, 84.4, 96.2, 102.6, 144.1, 150.5, 156.5, 163.4. IR
(Film): ν 3345 (br), 3196 (br), 2969 (s), 2873 (s), 1719 (s), 1597
(s), 1518 (s), 1466 (s), 1335 (s), 1248(s), 1169 (s), 1082(s). HRMS
calcd for C46H84N6O14Si4Na (MNa)+, 1079.5020, found 1079.5000.
Rf ) 0.35 (8% MeOH/CH2Cl2).
Silyla ted Dim er 15. 5′-Deoxy-5′-aminouridine (8) (0.49 g,
1.05 mmol) and bis(p-nitrophenyl carbonate) 10 (0.22 g, 0.50
mmol) compounds were azeotropically dried with THF and
dissolved in dry CH2Cl2 (3 mL) under N2. Triethylamine (0.31
mL, 1.10 mmol) and DMAP (0.01 g, 0.03 mmol) were added,
and the reaction mixture was allowed to stir for 20 h, after
which it was concentrated and purified by silica gel chroma-
tography (1-8% CH3OH/CH2Cl2) to provide silylated dimer 15
1
(0.47 g, 85%) as a white solid. H NMR (400 MHz, CDCl3): δ
0.52-0.65 (m, 24H), 0.90 (t, 18H, J ) 8.0 Hz), 0.96 (t, 18H, J
) 7.8 Hz), 3.45 (t, 4H, J ) 5.4 Hz), 3.69 (s, 4H), 3.99 (t, 4H, J
) 3.6 Hz), 4.06 (t, 4H, J ) 4.4 Hz), 4.21 (d, 4H, J ) 4.4 Hz),
4.60 (t, 4H, J ) 5.4 Hz), 5.41 (d, 2H, J ) 5.6 Hz), 5.76 (d, 2H,
J ) 8.0 Hz), 5.88 (s, 2H), 7.51 (d, 2H, J ) 8.0 Hz). 13C NMR
(100 MHz, CDCl3): δ 4.6, 4.6, 4.8, 4.9, 6.7, 6.8, 6.9, 42.7, 64.2,
69.4, 72.9, 84.1, 93.2, 102.3, 142.5, 150.1, 156.5, 163.4. IR
(Film): ν 3318, 3196, 2951, 2873, 1702, 1536, 1457, 1414, 1379,
(18) Orlean, P.; Cabib, E. Personal correspondence.
J . Org. Chem, Vol. 69, No. 3, 2004 617

Yeager and Finney
An aliquot (10-12 mL) of the saturated medium was trans-
ferred to 400 mL of YEPG medium to give an optical density
at 600 nm of 0.15-0.20 and allowed to grow to an optical
density of 0.65-0.70. The cells were washed with cold water
and TM buffer (50 mM tris‚HCl, 2.5 mM MgCl2, pH 7.5) by
suspension and centrifugation (15 min, 2000 × g). The wet
weight of the cells at this point was typically around 1 g; this
weight was used to determine the volume of buffer in which
the final pellet was suspended (vida infra).
The cells were suspended in 2 mL of TM buffer in a 50 mL
plastic centrifuge tube, and glass beads (0.45 mm) were added
until the volume of beads reached about 3 mm below the
liquid’s surface. The tube was then vortexed 20 × 30 s, with
30 s of cooling on ice between each vortex; vortexing was
performed in a 4 °C cold room. The broken cells were removed
from the bottom of the tube with a glass Pasteur pipet, and
the glass beads were rinsed 5-7 × 1.5 mL with TM buffer.
The pooled rinsings were centrifuged at 2000 × g for 4 min;
the supernatant was removed, and the remaining cell wall
precipitate was washed once more with TM buffer. The cell-
wall free supernatants were combined and centrifuged at
60 000 × g for 1 h.
substrate (typically 0.125 µCi, transferred to the eppendorf as
a solution that was then evaporated to dryness under vacuum)
in 40 µL of assay solution (containing the necessary inhibitor)
was transferred to each tube. Trypsin-treated membrane (20
µL) was then added, and the mixture was incubated for 1 h at
30 °C. The reaction was stopped by the addition of 1 mL of
cold (0 °C) aqueous trichloroacetic acid (10% v/v), filtered onto
glass fiber filter disks (Whatman GF/C, 25 mm), and rinsed
with 7:3 EtOH/1 M acetic acid (4 × 1 mL), and the remaining
radioactivity on the filter paper was measured by scintillation
counting.
Enzyme activity (in the absence of inhibitor) was typically
25000-30000 cpm/h, with a time-independent background of
500-600 cpm. Under these conditions, enzyme activity was
linear (on the basis of plots of incorporated radioactivity vs
time) to at least 3 h. Control reactions run in in the presence
of 0.1 mM polyoxin D or nikkomycin Z, both of which were
known competitive inhibitors (Ki ≈ 10 µM) of chitin synthase,
always showed g99% inhibition.
Ack n ow led gm en t. We thank the NIH (GM60875),
the Hellman Foundation, and the UC Systemwide
Biotechnology Program for direct support and the NIH
and NSF for infrastructure support (GM62116, GM61894,
CHE-9709183). We thank Prof. Peter Orlean (University
of Illinois) and Dr. Enrico Cabib (NIH) for assay advice,
Prof. Orlean for donation of yeast strains, Prof. Scott
Singleton for helpful discussion, and Prof. J ack Kyte and
Mr. Steven Adams (University of California, San Diego)
for technical assistance.
The enzyme pellet was suspended in 1.6 mL × {gram wet
weight of cells} of TM buffer and homogenized throughly with
a glass Dounce homogenizer. The membranes were pretreated
with trypsin (quantified by weight, 10 min, 30 °C) and then
treated with 3.0 × {mass of trypsin} trypsin inhibitor. Typi-
cally, four different concentrations of trypsin/trypsin inhibitor
were tested for each new membrane preparation, and the
combination with the highest activity at 30 min was used in
the assay. Concentrations of trypsin typically tested were 0.5,
1.0, 2.0, and 4.0 mg/mL, and trypsin was added to the
membrane preparation at a concentration of 1 µL of trypsin
solution for every 5 µL of membrane preparation. Individual
assays were performed in 1.5 mL eppendorf centrifuge tubes.
The assay solution used contained UDP-GlcNAc (1.0 mM),
GlcNAc (40mM), and digitonin (0.2% w/v) dissolved in pH 7.5
tris buffer (50 mM) containing MgCl2 (5.0 mM). Radioactive
Su p p or tin g In for m a tion Ava ila ble: Copies of 1H and 13C
NMR spectra for all compounds; IC50 plots for 1, 2, and 5; and
a Vmax recovery plot for 2. This material is available free of
charge via the Internet at http://pubs.acs.org.
J O035100C
618 J . Org. Chem., Vol. 69, No. 3, 2004