Novel carbohydrate scaffolds. Assembly of a uridine-mannose scaffold based on tunicamycin

Article abstract of DOI:10.1016/S0040-4039(99)02209-1

A disaccharide scaffold based on tunicamycin was synthesized from D- uridine and L-mannose. The key step in disaccharide assembly was a β- mannosylation performed using Crich's modification of the sulfoxide glycosylation method. The scaffold described contains two orthogonal derivatization sites and will be used in the search for novel biologically active compounds. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(99)02209-1

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 855–858
Novel carbohydrate scaffolds. Assembly of a uridine–mannose
scaffold based on tunicamycin
Domingos J. Silva ∗,† and Michael J. Sofia
Incara Research Laboratories, 8 Cedar Brook Drive, Cranbury, NJ 08536, USA
Received 9 September 1999; revised 19 November 1999; accepted 22 November 1999
Abstract
A disaccharide scaffold based on tunicamycin was synthesized from D-uridine and L-mannose. The key step in
disaccharide assembly was a β-mannosylation performed using Crich’s modification of the sulfoxide glycosylation
method. The scaffold described contains two orthogonal derivatization sites and will be used in the search for novel
biologically active compounds. © 2000 Elsevier Science Ltd. All rights reserved.
Keywords: carbohydrates; antibiotics; combinatorial chemistry.
As part of our drug discovery program, we are interested in the construction of scaffolds based on
therapeutically interesting carbohydrates and glycoconjugates.1 Such scaffolds can be utilized as building
blocks in the assembly of combinatorial libraries, with the objective of exploring the SAR of the original
drug and identifying novel biologically active agents.2
This paper reports work in the construction of one such scaffold based on the tunicamycins (1), a
family of nucleoside antibiotics isolated from the fermentation broths of Streptomyces lysosuperificus.3
These drugs have a general structure composed of a fatty acid chain, uracil, N-acetyl-D-glucosamine and
an undecose sugar named tunicamine.4 The tunicamycins have been shown to inhibit a wide variety
of lipid carrier-dependent protein glycosylations, having great potential as antibiotic and antitumor
agents.5 For example, in bacteria the tunicamycins inhibit the conversion of UDP-MurNAc to lipid
I catalyzed by the MraY enzyme, inhibiting cell wall biosynthesis and leading to bacterial death.6
Unfortunately, tunicamycins have not enjoyed use as human therapeutic agents because they are also
toxic to mammalian cells, interfering with dolichol–diphosphoryl–GlcNAc synthesis and inhibiting
oligosaccharide biosynthesis.7 Nevertheless, since tunicamycins inhibit enzymes with distinct substrate
requirements, it is plausible that distinct features of the drug may be recognized by each affected enzyme.
Tunicamycin analogs may thus have differential inhibitory effects towards eukaryotic and prokaryotic
cells, allowing for the possible targeting of pathogenic cells over mammalian cells. Because of the
∗
†
Corresponding author. E-mail: domingos_j_silva@sbphrd.com (D. J. Silva)
Current address: SmithKline Beecham Pharmaceuticals, 709 Swedeland Road, PO Box 1539, UW2421, King of Prussia,
PA 19406, USA.
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(99)02209-1
tetl 16137

856
great structural complexity of the drugs, limited work besides chemical degradation has been devoted
to establishing a structure–activity relationship for these compounds.8
The disaccharide scaffold described here (2, Fig. 1) has the general structure of tunicamycin, with
modifications in the two-carbon bridge and the glucosamine ring. Besides simplifying the overall
synthesis of 2, these modifications may shed light on the importance of the specific groups in biological
activity. Furthermore, 2 incorporates two sites (azide and hydroxyl groups) that can be derivatized
orthogonally (Fig. 1). Retrosynthetic analysis of 2 showed that the key synthetic step was formation of
a β-glycosidic linkage between a L-mannosyl donor and a D-uridyl acceptor. Synthesis of β-mannosides
is still one of the most difficult challenges in modern carbohydrate chemistry.9 Since Crich and Sun have
recently shown that β-mannosides can be synthesized in high yield using a modification of the sulfoxide
glycosylation method,10 this glycosylation method was chosen for construction of the disaccharide core.
Fig. 1.
Synthesis of the D-uridyl acceptor is shown in Fig. 2. Uridine 3 was reacted with trityl chloride in
pyridine in the presence of DMAP to afford the 50-trityl derivative 4, which was peralkylated with
benzyloxymethyl chloride to generate 5. Treatment of 5 with formic acid in acetonitrile led to clean
removal of the trityl group, without loss of the BOM groups, yielding acceptor 6.11
Fig. 2.
Synthesis of the L-mannosyl donor is shown in Fig. 3. L-Mannose 7 was peracetylated and reacted with
PhSH and BF3·OEt2 to generate sulfides 9a–b. The α-sulfide 9a was deacetylated to the tetraol 10 and
reacted with benzaldehyde dimethyl acetal to afford the 4,6-di-O-benzylidene compound 11. Treatment
with excess BnCl and NaH yielded the fully protected sulfide 12, which was cleanly oxidized using
mCPBA to sulfoxide 13. Compound 13 was recrystallized as a single diastereoisomer.12
Fig. 3.

857
Compounds 6 and 13 were coupled using Crich’s protocol for the sulfoxide glycosylation reaction
(Fig. 4). A solution of 13 and base in CH2Cl2 was treated with Tf2O at −78°C, and after 5 min 6 was
added. The reaction afforded 53% of the desired β-mannoside 14,13 along with 29% of the corresponding
α-mannoside and 10% of mannosyl lactols. The benzylidene group in 14 was selectively removed with
aqueous HOAc at 50°C, and the 60-hydroxyl group was tosylated to afford 16. Finally, the tosyl group
was displaced using sodium azide, affording target scaffold 17.14
Fig. 4.
Diversity could, in principle, be introduced in 17 by selective derivatization of the 60-azide and the 40-
hydroxyl groups on the L-mannose ring, followed by removal of protective groups (Fig. 5). In order
to validate such chemical diversity strategy, 17 was treated with PMe3 in THF–EtOH–H2O and the
resulting amine 18 was selectively acetylated using HOAc, HATU and DIPEA to afford 19. Reaction
with n-octadecyl isocyanate afforded the urethane 20, which was hydrogenated to generate the expected
bis-derivatized product 21.
Fig. 5.
In summary, we have reported the efficient synthesis of a disaccharide scaffold 17 based on the
tunicamycins. Scaffold 17 was synthesized from two commercially available building blocks (L-mannose
and D-uridine) and the key step in the synthesis involved the use of Crich’s modification of the sulfoxide
glycosylation method to obtain a β-mannoside in good yield. Compound 17 presents a functional group
dyad (the azide and the free hydroxyl groups) that can be orthogonally derivatized to generate di-
substituted disaccharide scaffolds. Based on this combinatorial flexibility, we are currently pursuing the
synthesis of tunicamycin analogs based on 17 and other related scaffolds. Biological evaluation of such

858
compounds should shed light on the mechanism of action of the tunicamycins and help refine our initial
choice of scaffold structure.
References
1. For previous work in this area, see: Sofia, M. J.; Hunter, R.; Chan, T. Y.; Vaughan, A.; Dulina, R.; Wang, J.; Gange, D. J.
Org. Chem. 1998, 63, 2802–2803.
2. Sofia, M. J. Med. Chem. Res. 1998, 8, 362–378.
3. Takatsuki, A.; Arima, K.; Tamura, G. J. Antibiotics 1971, 24, 215–223.
4. Takatsuki, A.; Kawamura, K.; Okina, M.; Kodama, Y.; Ito, T.; Tamura, G. Agric. Biol. Chem. 1977, 41, 2307–2310.
5. For reviews, see: (a) Elbein, A. Trends Biochem. Sci. 1981, 219–221. (b) Tunicamycin; Tamura, G., Ed.; Japan Scientific
Press: Tokyo, Japan, 1982. (c) Echardt, K. J. Nat. Prod. 1983, 46, 544–550.
6. Brandish, P. E.; Kimura, K. I.; Inukai, M.; Southgate, R.; Lonsdale, J. T.; Bugg, T. D. Antimicrob. Agents Chemother. 1996,
40, 1640–1644.
7. Heifetz, A.; Keenan, R. W.; Elbein, A. D. Biochem. 1979, 18, 2186–2192.
8. For a total synthesis, see: (a) Suami, T. In Mycotoxins and Phycotoxins; Steyn, P. S.; Vieggaar, R., Eds.; Elsevier Science:
Amsterdam, The Netherlands, 1985. (b) Danishefsky, S.; Barbachyn, M. J. Am. Chem. Soc. 1985, 107, 7761–7762. (c)
Myers, A. G.; Gin, D. Y.; Rogers, D. H. J. Am. Chem. Soc. 1994, 116, 4697–4718. (d) Myers, A. G.; Gin, D. Y.; Rogers,
D. H. J. Am. Chem. Soc. 1993, 115, 2036–2038. For biological evaluation of homologues and degradation products, see:
(e) Duksin, D.; Mahoney, W. C. J. Biol.Chem. 1982, 257, 3105–3109. (f) Hashim, O. H.; Cushley, W. Biochim. Biophys.
Acta 1987, 923, 362–370. (g) Sarabia-García, F.; Lópes-Herrera, F. J. Tetrahedron 1996, 52, 4757–4768. For synthesis and
evaluation of analogues, see: (h) Komimato, K.; Ogawa, S.; Suami, T. Carbohydr. Res. 1988, 174, 360–368.
9. (a) Paulsen, H. Angew. Chem., Int. Ed. Engl. 1990, 29, 823–938. (b) Toshima, K.; Tatsuta, K. Chem. Rev. 1993, 93,
1503–1531.
10. (a) Crich, D.; Sun, S. J. Org. Chem. 1996, 61, 4506–4507. (b) Crich, D.; Sun, S. J. Org. Chem. 1997, 62, 1198–1199.
11. Analytical data for 6: 1H NMR (300 MHz, CDCl3): δ 7.55–7.20 (m, 20H), 5.64 (1H, s), 4.82 (1H, d, J=12 Hz), 4.66 (1H, d,
J=12 Hz), 4.58 (2H, d, J=1.8 Hz), 4.50 (1H, br), 4.39 (1H, dd (br), J=1.2, 3 Hz), 4.34–4.20 (3H, m), 4.10 (1H, td, J=4.5, 9, 9
Hz), 3.76 (1H, t, J=9.9 Hz). 13C NMR (75.4 MHz, CDCl3): δ 141.5, 138.2, 137.3, 137.2, 131.6, 129.4, 129.0, 128.3, 128.2,
127.9, 127.8, 127.6, 126.0, 124.3, 101.6, 97.6, 78.0, 76.2, 73.5, 73.2, 72.8, 70.0, 68.2. MS (ES) m/z obs. 574 [M+NH4+].
1
12. Analytical data for 13: H NMR (300 MHz, CDCl3): δ 7.79 (1H, d, J=8.1 Hz), 7.40–7.20 (15H, m), 5.89 (1H, d, J=3.3
Hz), 5.66 (1H, d, J=8.1 Hz), 5.38 (2H, dd, Japp=9.9, 13.2 Hz), 4.98 (1H, d, J=6.9 Hz), 4.87 (1H, d, J=6.9 Hz), 4.83 (2H, m),
4.68–4.55 (5H, m), 4.45 (1H, t, J=3.3 Hz), 4.36 (1H, t, J=5.1 Hz), 4.22 (1H, d, J=6.0 Hz), 3.97 (1H, d (br), J=12.0 Hz),
3.74 (1H, dd (br), J=3.0, 12.0 Hz), 3.11 (1H, br). 13C NMR (75.4 MHz, CDCl3): δ 162.7, 150.8, 139.7, 137.6, 137.1, 128.4,
128.3, 128.2, 127.8, 127.6, 127.5, 127.5, 127.4, 101.5, 94.4, 94.2, 90.5, 82.8, 78.0, 73.0, 72.1, 69.9, 69.7, 60.7. MS (ES) m/z
obs. 622 [M+NH4+].
13. The determination of the stereochemistry of the new glycosidic linkage in 14 was hampered by severe overlap in NMR
1
spectra. Removal of all protective groups in 14 (H2, Pd/C, 1 atm) afforded 22, for which the H-coupled 13C NMR was
recorded in CD3OD (75.4 MHz). The 13C resonance of the mannosyl anomeric carbon (101.7 ppm) showed JH–C=159 Hz,
characteristic of β-mannosides: Bock, K.; Pedersen, C. J. Chem. Soc., Perkin Trans. 2 1974, 293–297. The β:α selectivity
of this glycosylation (1.8:1) is lower than the selectivities generally obtained with Crich’s procedure (Ref. 10) and may be
associated with particular stereoelectronic characteristics of the uridyl acceptor, since the mannosyl donor has been shown
to afford high β/α selectivities with primary glycosyl acceptors (Ref. 10b).
14. Analytical data for 17: 1H NMR (300 MHz, CDCl3): δ 7.48 (1H, d, J=8.4 Hz), 7.4–7.2 (25H, m), 5.96 (1H, d, J=2.4 Hz),
5.50 (1H, d, J=8.1 Hz), 5.41 (1H, d, J=9.9 Hz), 5.36 (1H, d, J=9.9 Hz), 4.99 (1H, d, J=6.9 Hz), 4.9–4.8 (4H, m), 4.7–4.5
(9H, m), 4.4–4.3 (5H, m), 4.22 (1H, dd, J=2.7, 11.4 Hz), 3.9–3.8 (3H, m), 3.5–3.4 (3H, m), 3.32 (1H, dd, J=2.7, 9.3 Hz).
13C NMR (75.4 MHz, CDCl3): δ 162.3, 150.7, 138.4, 138.1, 137.8, 137.2, 128.7, 128.4, 128.3, 128.2, 127.9, 127.8, 127.8,
127.7, 127.7, 127.6, 127.5, 101.6, 101.4, 94.3, 89.9, 81.6, 80.8, 78.1, 76.3, 74.5, 74.2, 73.0, 72.2, 71.6, 70.2, 69.9, 69.9, 67.7,
67.5, 51.5. MS (ES) m/z obs. 989 [M+NH4+].