Synthesis of a bisubstrate-type inhibitor of N- acetylglucosaminyltransferases

Article abstract of DOI:10.1002/anie.200460388

Transfer interference: Synthesis of the bisubstrate-type N-acetylglucosaminyltransferase (GnT) inhibitor 1 was achieved by a polymer-resin hybrid capture-release strategy for the construction of the acceptor component. One-pot ligation in aqueous media pr

Full text of DOI:10.1002/anie.200460388

Communications
Glycosyltransferase Inhibitor
Scheme 1. a) Substrate specificities of GnT-V and GnT-IX; b) bisub-
strate-type inhibitors consist of acceptor trisaccharide and UDP–
GlcNAc moieties. UDP=uridine diphosphate.
Synthesis of a Bisubstrate-Type Inhibitor of N-
Acetylglucosaminyltransferases**
structures, as well as to O-linked GlcNAcb1-2Man.[4] The
GlcNAcb1-6-branched glycans are often extended by the
polylactosamine ((Galb1-4GlcNAc)n, Gal = galactose)struc-
ture and present ligands for cell-adhesion molecules. It is well
known that levels of b1-6-branched glycans are increased in
tumor cells,[5] and GnT-V is involved in cancer metastasis.[6a–e]
Also, relationships with T-cell activation and angiogenesis
have been revealed.[6f,g] The exclusive expression of GnT-IX
in the brain is in sharp contrast with the ubiquitous expression
of GnT-V, and this difference suggests roles for GnT-IX in
neuronal development and functions.[3]
It is expected that suppression of GnT-V may be useful for
the treatment of cancer.[7] In addition to a gene-knockout
strategy,[8] chemical inhibition would be an promising
approach. In fact, several attempts to develop inhibitors of
GnT-V that mimic acceptor substrates have been reported.[9]
In this paper, we report the synthesis of compound 1a, which
is a prototype for bisubstrate-type inhibitors of type 1
(Scheme 1b).[10] These inhibitors were designed to contain
both donor (UDP–GlcNAc)and acceptor components, with
the proposed mechanism of inverting GnTs[11] taken into
consideration. As the acceptor component, the trisaccharide
(GlcNAcb1-2Mana1-6Manb)was incorporated, because it
was previously reported by Tahir and Hindsgaul to serve as an
efficient acceptor substrate of GnT-V.[12] In order to approach
our target 1a, a convergent route was adopted; namely, the
trisaccharide and donor components were constructed sepa-
rately and combined together by chemoselective ligation.
Synthesis of the acceptor trisaccharide was conducted by
using solution-phase polymer-support technology,[13] which
utilized low-molecular-weight (MW ꢀ 750)poly(ethylene
glycol)monomethyl ether (MPEG,) as shown in Scheme 2.
Shinya Hanashima, Shino Manabe, Kei-ichiro Inamori,
Naoyuki Taniguchi, and Yukishige Ito*
Glycosyltransferases are a group of enzymes responsible for
the biosynthesis of glycoconjugate oligosaccharides.[1] Among
them, N-acetylglucosaminyltransferases (GnTs)are key
enzymes in the production of highly branched complex N-
glycan structures. GnT-V transfers an N-acetylglucosamine
(GlcNAc)residue to the core a1-6-mannose (Man)arm to
form a b1-6 linkage (Scheme 1a).[2] The recently identified
GnT-IX is a homologue of GnT-V that is exclusively
expressed in the brain.[3] GnT-IX has a broader specificity
and transfers GlcNAc to both a1-6- and a1-3-mannose
[*] Dr. S. Hanashima, Dr. Y. Ito
RIKEN (The Institute of Physical and Chemical Research)
2–1 Hirosawa, Wako-shi, Saitama 351-0198 (Japan)
Fax: (+81)48-462-4680
E-mail: yukito@riken.jp
Dr. S. Manabe
RIKEN and PRESTO
Japan Science and Technology Agency (JST) (Japan)
Dr. K.-i. Inamori, Prof. Dr. N. Taniguchi
Department of Biochemistry
Osaka University Medical School
2–2 Yamadaoka, Suita, Osaka 565-0871 (Japan)
[**] S.H. is a recipient of the Special Postdoctoral Researcher Program
Fellowship in RIKEN. Additional financial support from CREST (JST)
is acknowledged. We thank Dr. T. Chihara and his staff for elemental
analyses and Dr. H. Koshino and Ms. T. Chijimatsu for NMR
measurements. We thank Ms. Akemi Takahashi for her technical
assistance.
5674
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/anie.200460388
Angew. Chem. Int. Ed. 2004, 43, 5674 –5677

Angewandte
Chemie
Scheme 2. Polymer-resin hybrid synthesis of trisaccharide 7. Tr=triphenylmethyl=trityl, Bn=benzyl, TFA=trifluoroacetic acid, Phth=phthaloyl,
NIS=N-iodosuccinimide, TfOH=trifluoromethanesulfonic acid, HDTC=hydrazinedithiocarbonate, Fmoc=9-fluorenylmethoxycarbonyl.
During the whole process, MPEG functioned as a polar tag
and products were retrieved by chromatography on short
silica gel columns.[14] In the first place, MPEG-supported 2
was prepared by using Hodosi and Kovµcꢀs b-mannosylation,
as described before.[13d,15] The trityl group was removed with
TFA (40 equiv)and Et 3SiH (20 equiv), and the liberated
alcohol was glycosylated with phenylthiomannoside 3 with
NIS and TfOH (1.1 equiv)to afford disaccharide 4.[16,17]
Cleavage of the chloroacetyl group[18] and glycosylation with
5 provided trisaccharide 6. At this stage, the product was
subjected to capture–release purification. Namely, compound
6 was captured with cysteine-conjugated Wang resin 8
through the reaction between the chloroacetyl and thiol
groups. Liberation of the amino group from the Fmoc
protection with 4-aminomethylpiperidine initiated cyclization
and trisaccharide 7 was obtained in 46% overall yield and
with high purity.
Further deprotection[13d] and introduction of the thiol
group were conducted as depicted in Scheme 3. Acidic
cleavage of the benzylidene acetal, ethylenediamine treat-
ment, and acetylation afforded compound 9 in 77% yield.
Subsequent deacetylation and esterification gave 10 in 85%
yield. The contaminating aGlcNAc isomer (< 10%)was
separated at this stage. Tosylation of the primary hydroxy
group was best performed by using DMAP as a base to afford
11, and cleavage of the benzyl ethers gave 12. The tosyl group
of 12 was substituted with SAc by using AcSK (5 equiv)in
DMF at 708C. Deacetylation with NaOMe was accompanied
by disulfide formation to provide 13.
Scheme 3. Preparation of acceptor 13: a) 60% aqueous AcOH, 608C,
87%; b) 1. 1m KOH, EtOH/THF, reflux; 2. ethylenediamine, 1-BuOH,
1008C; 3. Ac2O, pyridine, 77%; c) 1. 0.05m NaOMe/MeOH;
2. TMSCHN2, PhH, MeOH, 85%; d) TsCl, DMAP, CH2Cl2, 60%; e) H2,
20% Pd(OH)2/C, AcOH, MeOH, 88%; f) 1. AcSK, DMF, 708C;
2. 0.05m NaOMe/MeOH, 90%. Ts=tosyl=toluene-4-sulfonyl,
TMS=trimethylsilyl, DMAP=4-dimethylaminopyridine, DMF=N,N-
dimethylformamide.
Glucosamine phosphate 21 was synthesized as shown in
Scheme 4. Compound 14 was subjected to anomeric de-
Angew. Chem. Int. Ed. 2004, 43, 5674 –5677
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5675

Communications
Scheme 5. Chemoselective one-pot ligation and coupling with UMP:
a) TCEP-HCl, MeOH, H2O; b) 21, iPr2NEt; c) Et3N, 51% (three steps);
d) UMP–morpholidate, 1H-tetrazole, pyridine, 78%. UMP=uridine
monophosphate, TCEP-HCl=tris(carboxyethyl)phosphine hydrochlo-
ride.
Scheme 4. Preparation of phosphate 21: a) NH2NH2/AcOH, THF,
94%; b) TBSCl, imidazole, DMF, 95%; c) 1. H2, 10% Pd/C, EtOAc;
2. (BrCH2CO)2O, pyridine, CH2Cl2, 98%; d) 47% aqueous HF, CH3CN,
93%; e) 1. 19, 1H-tetrazole, CH2Cl2, ꢁ108C; 2. TBHP, ꢁ40!08C,
57%; f) [Pd(PPh3)4], Et3SiH, AcOH, toluene, 85%. Cbz=benzyloxycar-
bonyl, TBS=tert-butyldimethylsilyl, All=allyl, TBHP=tert-butylhydro-
peroxide.
only modest in comparison with the acceptor substrate (Km =
150 mm). However, its activity toward GnT-IX was much
greater (Ki = 7.2 mm). In order to investigate the kinetic
mechanism and physiological role of GnT-IX, compound 1a
would therefore be a useful molecular probe.
In summary, we have synthesized bisubstrate-type ana-
logue 1a as a GnT-V inhibitor by using a polymer-resin hybrid
strategy. Ligation of bromoacetamide 21 and disulfide 13,
after coupling with UMP, gave the desired compound 1a.
Simple extension of this chemistry would provide access to
various homologues differing in the structure of the acceptor
component (for example, Man, GlcNAcb1-2Man, Mana1-
6Man)or the donor–acceptor distance. The latter task may
well be achieved immediately simply by changing the
bromoacyl group of 17 (see Scheme 4). Further studies are
in progress along these lines to discover more potent and
selective inhibitors of GnT-V and GnT-IX.
acetylation, and the hemiacetal was masked with a TBS group
to give 16. Removal of the Cbz group was conducted by
hydrogenolysis in ethyl acetate,[19] and subsequent N-bro-
moacetylation gave 17 in 98% yield.
The TBS group was removed with 47% aqueous HF in
CH3CN to afford hemiacetal 18 in 93% yield. Direct
phosphorylation turned out to be problematic because of
base sensitivity of the bromoacetamide moiety. However,
treatment with 19 and 1H-tetrazole at low temperature
(ꢁ108C)readily produced a phosphite compound that was
oxidized with TBHP to afford phosphate 20 in 57% yield.[20,21]
Cleavage of the allyl groups by using [Pd(PPh3)4] and Et3SiH
afforded 21 in 85% yield.
Sequential ligation and deprotection were conducted in a
one-pot procedure (Scheme 5). Disulfide 13 was reduced to
the thiol derivative with TCEP in MeOH/H2O. Subsequent
addition of bromoacetamide 21 and iPr2NEt afforded crude
22, which was deacetylated with excess Et3N to give 23 in 51%
yield from 13 over 3 steps (see the Experimental Section).
The modified morpholidate method reported by Wittmann
and Wong[22] was used to introduce UMP; namely, mono-
phosphate 23 was treated with UMP–morpholidate in the
presence of 1H-tetrazole in pyridine for 4 days to afford 1a in
78% yield. The structure of 1a was rigorously confirmed by
2D NMR spectroscopy (DQF-COSY, HMQC)and mass
spectrometry (MALDI-TOF).[23]
Experimental Section
Preparation of 22 by a one-pot procedure: TCEP (3.5 mg, 17 mmol)
was added to a solution of disulfide 13 (4.0 mg, 2.7 mmol)in MeOH/
H2O (7:3, 1 mL). After stirring for 1 h, GlcNAc derivative 21 (9.3 mg,
18 mmol)and iPr2NEt (20 mL)were added, and the mixture was
stirred for 28 h. Et3N (0.1 mL)was then added, and the mixture was
stirred for an additional day. This was followed by concentration and
purification by chromatography with
a Sep-Pak C18 cartridge
(MeOH/H2O 0:100!40:60)afforded 22 (2.9 mg, 0.00283 mmol,
51%).
Enzyme assay: GnT-V activity was assayed by using pyridyl
aminated acceptor substrate and cell lysate from COS-1 cells
transfected with the GnT-V expression vector as an enzyme
source.[3,25] GnT-IX activity was assayed by using pyridyl amino
ethyl succinamyl acceptor substrate and partially purified recombi-
nant soluble GnT-IX.[4] For kinetic analyses, these enzyme sources
were incubated at 378C for 2 h with
Inhibitory activities of 1a toward GnT-V and GnT-IX
were evaluated, and the results are summarized in
Table 1.[24,25] The affinity of 1a to GnT-V (Ki = 103 mm)was
20 mm acceptor substrate (GnGn-bi-PA
or GnM-S-PAES)and various concen-
trations of UDP–GlcNAc and the
inhibitor in 100 mm b-morpholino-
ethanesulfonic acid (pH 6.25)or 3-( N-
Table 1: Inhibitory activities of 1a toward GnT-V and GnT-IX.
Enzyme
Acceptor
Km [mm][a]
UDP–GlcNAc
Ki [mm][b]
1a
Acceptor
GnT-V
GnT-IX
GnGn-bi-PA[c]
0.150
–
4.0–6.0
–
103
7.2
morpholine)propanesulfonic
(pH 7.5)buffer containing 200 m
GlcNAc, 0.5% Triton X-100, and
10 mm ethylenediaminetetraacetate.
acid
GnMSer-PAES[d]
m
[a] Michaelis constant. [b] Inhibition constant. [c] GnGn-bi-PA
ref. [25]. [d] GnMSer-PAES = pyridylaminoethylsuccinimyl GnM; see Scheme 1a and ref. [4].
= bi-pyridylaminated Gn-Gn; see
5676
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 5674 –5677

Angewandte
Chemie
The reaction was terminated by boiling for 3 min and then centrifu-
gated at 15000 rpm for 5 min. The resulting supernatant was analyzed
by HPLC.
[18] C. A. A. van Boeckel, T. Beetz, Tetrahedron Lett. 1983, 24,
3775 – 3778.
[19] Hydrogenolysis in EtOH caused 3-O-deacetylation as a side
reaction.
[20] Preparation of bis(allyloxy)diisopropylaminophosphine 19: W.
Bannwarth, E. Kung, Tetrahedron Lett. 1989, 30, 4219 – 4222.
[21] M. M. Sim, H. Kondo, C.-H. Wong, J. Am. Chem. Soc. 1993, 115,
2260 – 2267.
Received: April 21, 2004
Revised: June 9, 2004
Keywords: bisubstrates · glycoproteins · inhibitors · solid-phase
.
[22] V. Wittmann, C.-H. Wong, J. Org. Chem. 1997, 62, 2144 – 2147.
synthesis · transferases
[23] Spectroscopic data for 1a: 1H NMR (500 MHz, D2O with a small
amount of dioxane as an internal standard, 408C): d = 7.78 (d,
3
3JH,H = 7.8 Hz, 1H), 5.81 (m, 2H), 5.37 (dd, 3JH,H = 3.0, JH,P
=
[1] Handbook of Glycosyltransferases and Related Genes (Eds.: N.
Taniguchi, K. Honke, M. Fukuda), Springer, Tokyo, 2002.
[2] a)I. Brockhausen, J. P. Carver, H. Schachter, Biochem. Cell Biol.
1988, 66, 1134 – 1151; b)R. D. Cummings, I. S. Trowbridge, S.
Kornfeld, J. Biol. Chem. 1982, 257, 13421 – 13427.
[3] K. Inamori, T. Endo, Y. Ide, S. Fujii, J. Gu, K. Honke, N.
Taniguchi, J. Biol. Chem. 2003, 278, 43102 – 43109.
[4] K. Inamori, T. Endo, J. Gu, I. Matsuo, Y. Ito, S. Fujii, H. Iwasaki,
H. Narimatsu, E. Miyoshi, K. Honke, N. Taniguchi, J. Biol.
Chem. 2004, 279, 2337 – 2340.
7.1 Hz, 1H), 4.76 (brs, 1H; anomeric proton of aMan), 4.42 (brs,
1H; anomeric proton of bGlcNAc; overlapped in HOD signal,
confirmed by HMQC spectra)4.20 (brs, 1H; anomeric proton of
bMan) 4.11–3.22 (m), 3.59 (s, 3H; methyl ester), 2.96 (dd, 3JH,H
=
14.6, 1.8 Hz, 1H), 2.58 (dd, 3JH,H = 14.6, 8.3 Hz, 1H), 2.22 (t,
3JH,H = 7.3 Hz, 2H), 1.90 (s, 3H), 1.42 (brs, 4H), 1.15 (brs,
8H)ppm; MALDI-TOF MS (negative mode, CHCA): m/z:
calcd for C47H77N4O34P2S [MꢁH]ꢁ: 1335.36; found: 1335.32.
[24] K. Sasai, Y. Ikeda, T. Fujii, T. Tsuda, N. Taniguchi, Glycobiology
2002, 12, 119 – 127.
[5] J. W. Dennis, M. Granovsky, C. E. Warren, Biochim. Biophys.
Acta 1999, 1473, 21 – 34.
[25] N. Taniguchi, A. Nishikawa, S. Fujii, J. G. Gu, Methods Enzymol.
1989, 179, 397 – 408.
[6] a)J. W. Dennis, S. Laferte, C. Waghorne, M. L. Breitman, R. S.
Kerbel, Science 1987, 236, 582 – 585; b)S. Ihara, E. Miyoshi, J.
Ko, K. Murata, S. Nakahara, K. Honke, R. B. Dickson, C.-Y. Lin,
N. Taniguchi, J. Biol. Chem. 2002, 277, 16960 – 16967; c)M.
Demetriou, I. R. Nabi, M. Coppolino, S. Dedhar, J. W. Dennis, J.
Cell Biol. 1995, 130, 383 – 392; d)H.-B. Guo, I. Lee, M. Kamar,
S. K. Akiyama, M. Pierce, Cancer Res. 2002, 62, 6837 – 6845;
e)H.-B. Guo, I. Lee, M. Kamar, M. Pierce, J. Biol. Chem. 2003,
278, 52412 – 52424; f)M. Demetriou, M. Granovsky, S. Quaggin,
J. W. Dennis, Nature 2001, 409, 733 – 739; g)T. Saito, E. Miyoshi,
K. Sasai, N. Nakano, H. Eguchi, K. Honke, N. Taniguchi, J. Biol.
Chem. 2002, 277, 17002 – 17008.
[7] M. Granovsky, J. Fata, J. Pawling, W. J. Muller, R. Khokha, J. W.
Dennis, Nat. Med. 2000, 6, 306 – 312.
[8] J. W. Dennis, J. Pawling, P. Cheung, E. Partridge, M. Demetriou,
Biochim. Biophys. Acta 2002, 1573, 414 – 422.
[9] a)P.-P. Lu, O. Hindsgaul, H. Li, M. M. Palcic, Carbohydr. Res.
1997, 303, 283 – 291; b)I. Brockhausen, F. Reck, W. Kuhns, S.
Khan, K. L. Matta, E. Meinjohanns, H. Paulsen, R. N. Shah,
M. A. Baker, H. Schachter, Glycoconjugate J. 1995, 12, 371 – 379.
[10] Previous reports on bisubstrate-type glycosyltransferase inhib-
itors: a)H. Hashimoto, T. Endo, Y. Kajihara, J. Org. Chem. 1997,
62, 1914 – 1915; b)B. Waldscheck, M. Streiff, W. Notz, W. Kinzy,
R. R. Schmidt, Angew. Chem. 2001, 113, 4120 – 4124; Angew.
Chem. Int. Ed. 2001, 40, 4007 – 4011; c)H. Hinou, X.-L. Sun, Y.
Ito, J. Org. Chem. 2003, 68, 5602 – 5613.
[11] I. Tvaroska, I. Andre, J. P. Carver, J. Am. Chem. Soc. 2000, 122,
8762 – 8776.
[12] S. H. Tahir, O. Hindsgaul, Can. J. Chem. 1986, 64, 1771 – 1780.
[13] a)H. Ando, S. Manabe, Y. Nakahara, Y. Ito, J. Am. Chem. Soc.
2001, 123, 3848 – 3849; b)H. Ando, S. Manabe, Y. Nakahara, Y.
Ito, Angew. Chem. 2001, 113, 4861 – 4864; Angew. Chem. Int. Ed.
2001, 40, 4725 – 4728; c)Y. Ito, S. Manabe, Chem. Eur. J. 2002, 8,
3076 – 3084; d)S. Hanashima, S. Manabe, Y. Ito, Synlett 2003,
979 – 982.
[14] L. Jian, C. Hartley, T.-H. Chan, Chem. Commun. 1996, 2193 –
2194.
[15] G. Hodosi, P. Kovµc, Carbohydr. Res. 1998, 308, 63 – 75.
[16] a)P. Konradsson, D. R. Mootoo, R. E. McDevitt, B. Fraser-Reid,
J. Chem. Soc. Chem. Commun. 1990, 270 – 272; b)G. H. Veene-
man, S. H. van Leeuwen, J. H. van Boom, Tetrahedron Lett.
1990, 31, 1331 – 1334.
[17] In this particular case, use of a catalytic amount of TfOH
afforded the corresponding orthoester exclusively.
Angew. Chem. Int. Ed. 2004, 43, 5674 –5677
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5677