Total synthesis and structural revision of TMG-chitotriomycin, a specific inhibitor of insect and fungal β-N-acetylglucosaminidases

Article abstract of DOI:10.1021/ja9055245

(Chemical Equation Presented) TMG-chitotriomycin, a potent and selective inhibitor of the β-N-acetylglucosaminidases that possesses an unique N,N,N-trimethyl-D-glucosamine (TMG) residue, is revised to be the TMG-β-(1→4)-chitotriose instead of the original

Full text of DOI:10.1021/ja9055245

Published on Web 08/07/2009
Total Synthesis and Structural Revision of TMG-chitotriomycin, a Specific
Inhibitor of Insect and Fungal ꢀ-N-Acetylglucosaminidases
You Yang,†,‡ Yao Li,† and Biao Yu*,†
State Key Laboratory of Bio-organic and Natural Products Chemistry, Shanghai Institute of Organic Chemistry,
Chinese Academy of Sciences, Shanghai 200032, China, and Department of Chemistry, UniVersity of Science and
Technology of China, Hefei, Anhui 230026, China
Received July 6, 2009; E-mail: byu@mail.sioc.ac.cn
Scheme 1
TMG-chitotriomycin (1) was recently disclosed by Kanzaki and
co-workers from the culture filtrate of Streptomyces anulatus
NBRC13369, which exhibited potent and selective inhibition against
the ꢀ-N-acetylglucosaminidase (GlcNAcase) of insects and fungi
but no activity against the enzyme of humans and plants.1 This
unique inhibitory spectrum could lead to new insights for under-
standing the molecular mechanisms of the important chitin-
degrading systems occurring in a wide variety of organisms and
the development of new antifungal and insecticidal agents.2 The
proposed structure of 1 (Figure 1), determined by spectral and
constitutive sugar analyses of the authentic sample and its alditol
derivatives, was also intriguing in that this tetrasaccharide possesses
a unique N,N,N-trimethyl-D-glucosamine (TMG) residue R-(1f4)-
linked at the nonreducing end of a chitotriose;3 moreover, the
presence of the trimethylammonium could astonishingly result in
epimerization at the remote C2 of the reducing-end GlcNAc unit.1
Figure 1
In this work, we developed a convergent [2 + 2] approach to
synthesize the unique tetrasaccharide 1 (Scheme 1),4 where the
sterically demanding (1f4)-glycosidic linkages were assembled
successfully by our newly developed glycosylation protocol with
glycosyl ortho-hexynylbenzoates as donors and Au(I) as the
catalyst.5 The 2-azido and 2-N-phthalimido (Phth) groups in the
monosaccharide units were employed to secure the formation of
the R- and ꢀ-glycosidic linkages, respectively, and to distinguish
the final elaboration of the trimethylammonium and acetamide
functions. The p-methoxyphenyl (MP) group was used to protect
the anomeric hydroxyl groups.
Thus, the required three monosaccharide building blocks 3-5
were readily prepared from glucosamine.6 The coupling of the
2-azidoglucopyranosyl o-hexynylbenzoate 3 with glucosamine-
4-OH derivative 4 in Et2O (at -30 °C to rt) demanded an excess
amount of the donor 3 (1.8 equiv) and a larger loading of the catalyst
Ph3PAuOTf (0.5 equiv) for complete consumption of the acceptor
4, providing the desired R-linked disaccharide 6 in 96% yield with
no ꢀ-anomer being detected. The coupling of the 2-N-Phth-
glucopyranosyl o-hexynylbenzoate 5 (1.3 equiv) with acceptor 4
in CH2Cl2 (-30 °C to rt) proceeded smoothly in the presence of
0.2 equiv of Ph3PAuOTf, affording the ꢀ-linked disaccharide 7
exclusively (98%). Disaccharide 6 was subjected to selective
removal of the anomeric O-MP protection with ceric ammonium
nitrate (CAN) and subsequent ester formation (with o-hexynylben-
zoic acid 8) to afford the disaccharide o-hexynylbenzoate 9 (76%
for two steps). Disaccharide 7 was converted into a new acceptor
10 via selective opening of the 4,6-O-benzylidene acetal (Et3SiH,
BF3OEt2, CH2Cl2, rt, 78%).7 Coupling of disaccharides 9 and 10
under the catalysis of Ph3PAuOTf (0.2 equiv) in CH2Cl2 at -30
°C to rt provided the desired ꢀ-linked tetrasaccharide 11 in a
satisfactory 79% yield.
To convert the fully masked tetrasaccharide 11 into the target
TMG-chitotriomycin (1), model studies were carried out on
disaccharide 6 to find the optimal sequence and conditions for those
† Shanghai Institute of Organic Chemistry.
‡ University of Science and Technology of China.
9
12076 J. AM. CHEM. SOC. 2009, 131, 12076–12077
10.1021/ja9055245 CCC: $40.75  2009 American Chemical Society

C O M M U N I C A T I O N S
transformations (data not shown). It was noted that the trimethy-
lammonium residue could not survive under the conditions required
for removal of the N-Phth groups. Thus, the three N-Phth groups
in 11 were first removed (NH2CH2CH2NH2, EtOH, reflux)8 and
the resulting -NH2 acetylated to give 12 (65%). The azido group
in 12 was reduced into an amino group with 1,3-propanedithiol
(NEt3, pyridine, H2O, rt)9 in an excellent yield of 96%; other
conditions, such as reduction with PPh3 or Zn/HOAc, led to much
lower yields (∼50%) because of migration of the O-Ac group onto
the nascent amino group. The resulting amino group was then
converted readily into the trimethylammonium residue with excess
amounts of MeI in the presence of iPr2NEt in THF at rt, leading to
13 (82%). The O-Ac groups in 13 were then removed with K2CO3
in a MeOH/CH2Cl2 mixed solvent at rt (81%), and the remaining
O-Bn groups were cleaved via hydrogenolysis over Pd(OH2)2/C in
the same solvent in the presence of a tiny amount of HCl(aq) at rt
(81%), affording the p-methoxyphenyl ꢀ-glycoside of TMG-
chitotriomycin (14). The final removal of the anomeric MP group
proved to be difficult, as treatment of 14 with CAN under a variety
of conditions led to complex mixtures. Fortunately, a mild oxidizing
agent, bis(hydrogen dipicolinate)silver(II),10 effected the cleavage
of the anomeric MP group, furnishing the target tetrasaccharide 1
(78%).
reducing-end C2; however, further research is required to clarify
this unusual property of TMG-chitotriomycin.
Scheme 2
In summary, the tetrasaccharides 1 and 2 with the unique TMG
unit R- and ꢀ-(1f4)-linked to a chitotriose, respectively, have been
efficiently synthesized in a convergent [2 + 2] manner in which
the construction of the sterically demanding (1f4)-glycosidic
linkages was achieved by the newly developed Au(I)-catalyzed
glycosylation protocol with stable glycosyl o-hexynylbenzoates as
donors. The present total synthesis has led to the unambiguous
revision of the structure of TMG-chitotriomycin from 1 to 2 and
shall provide ready access to this intriguing natural metabolite and
its derivatives for further exploration of the selective inhibition of
GlcNAcases, which are important and widely occurring in the
chitin-degrading systems in nature.
The synthetic compound 1 was apparently not the natural TMG-
1
chitotriomycin, as determined by a comparison of their H NMR
spectra. The biggest discrepancy arose from the chemical shift of
the anomeric proton of the TMG residue: that of the natural product
appeared at 5.38 ppm (J ≈ 3.0 Hz) and that of synthetic 1 at 6.24
ppm (J ) 2.7 Hz). After carefully examining the present synthesis
and the structural evidence provided by Kanzaki and co-workers,1
we suspected that the TMG might be ꢀ-linked to the chitotriose in
the natural product instead of R-linked as in the previously
assignment. The small H1,2 coupling constant might be attributed
Acknowledgment. This work was financially supported by the
NSFC, MOST, and CAS.
4
to the conformational deviation of the TMG residue from the C1
commonly adopted by glucosamine residues. This is strongly
supported by the assignment of the ꢀ-TMG residue in the alditol
derivatives of TMG-chitotriomycin to be in a twist-boat conforma-
tion.6
Supporting Information Available: Experimental details, charac-
terization data, and 1H NMR spectra for new compounds. This material
is available free of charge via the Internet at http://pubs.acs.org.
References
With a modification of the synthetic approach leading to
tetrasaccharide 1, the synthesis of the revised structure 2 was
straightforward (Scheme 2). Thus, the ꢀ-linked disaccharide 17 was
assembled stereoselectively via coupling of 2-azidoglucopyranosyl
R-trichloroacetimidate 15 with glucosamine-4-OH derivative 16
in the presence of BF3OEt2 in CH2Cl2 at -30 °C (72%).4b,6 The
p-methoxybenzyl disaccharide 17 was then converted into the
corresponding o-hexynylbenzoate 18 (69% for two steps) in a
manner similar to that for the 6 f 9 conversion. The Au(I)-
catalyzed glycosidic coupling of the disaccharides 18 and 10 worked
equally as well as that of 9 and 10 to give 11, providing the whole
ꢀ-linked tetrasaccharide 19 in 76% yield. Steps similar to those
for 11 f 1 were then employed for the conversion of 19 into the
newly proposed TMG-chitotriomycin (2) (seven steps, 28%).
(1) Usuki, H.; Nitoda, T.; Ichikawa, M.; Yamaji, N.; Iwashita, T.; Komura,
H.; Kanzaki, H. J. Am. Chem. Soc. 2008, 130, 4146.
(2) (a) Merzendorfer, H.; Zimoch, L. J. Exp. Biol. 2003, 206, 4393. (b) Horsch,
M.; Mayer, C.; Sennhauser, U.; Rast, D. M. Pharmacol. Ther. 1997, 76,
187.
(3) The synthetic TMG has been reported. See: (a) Reckendorf, W. M.; Sandner,
S. Tetrahedron Lett. 1988, 29, 2047. (b) Falkowski, L.; Beszczynski, M.;
Jarzebski, A.; Stefanska, B. Pol. J. Chem. 1983, 57, 1353.
(4) For some relevant examples of the synthesis of chitooligosaccharide
derivatives, see: (a) Nicolaou, K. C.; Bockovich, N. J.; Carcanague, D. R.;
Hummel, C. W.; Even, L. F. J. Am. Chem. Soc. 1992, 114, 8701. (b) Wang,
L. X.; Li, C.; Wang, Q. W.; Hui, Y. Z. Tetrahedron Lett. 1993, 34, 7763.
(c) Ikeshita, S.; Sakamoto, A.; Nakahara, Y.; Nakahara, Y.; Ogawa, T.
Tetrahedron Lett. 1994, 35, 3123. (d) Tailler, D.; Jacquinet, J. C.; Beau,
J. M. J. Chem. Soc. Chem. Commun. 1994, 1827.
(5) Li, Y.; Yang, Y.; Yu, B. Tetrahedron Lett. 2008, 49, 3604.
(6) See the Supporting Information for details.
(7) (a) Deninno, M. P.; Etienne, J. B.; Duplantier, K. C. Tetrahedron Lett.
1995, 36, 669. (b) Debenham, S. D.; Toone, E. J. Tetrahedron: Asymmetry
2000, 11, 385.
(8) (a) Kanie, O.; Crawley, S. C.; Palcic, M. M.; Hindsgaul, O. Carbohydr.
Res. 1993, 243, 139. (b) Hansson, J.; Garegg, P. J.; Oscarson, S. J. Org.
Chem. 2001, 66, 6234.
(9) (a) Stauch, T.; Greilich, U.; Schmidt, R. R. Liebigs Ann. 1995, 2101. (b)
Galan, M. C.; Venot, A. P.; Glushka, J.; Imberty, A.; Boons, G. J. J. Am.
Chem. Soc. 2002, 124, 5964.
1
The H NMR spectrum of the synthetic tetrasaccharide 2 was
virtually identical to that of the authentic TMG-chitotriomycin.1,6
Also, the ESI-MS spectrum of 2, as described for the authentic
sample, developed an intensified [M+ + 1] peak (at m/z 832) after
CD3OD treatment, indicating the deuterium exchange at H2 of the
reducing-end GlcNAc; however, the same was not observed for
synthetic 1. These results indicated that the ꢀ-orientation of the
TMG unit is also critical in causing the epimerization at the
(10) (a) Kloc, K.; Mlochowski, J.; Syper, L. Chem. Lett. 1980, 725. (b) Noshita,
T.; Sugiyama, T.; Kitazumi, Y.; Oritani, T. Tetrahedron Lett. 1994, 35,
8259.
JA9055245
9
J. AM. CHEM. SOC. VOL. 131, NO. 34, 2009 12077