The new and efficient synthesis of a heptose moiety of spicamycin

Article abstract of DOI:10.1246/cl.2003.190

The new and efficient synthesis of 7-O-acetyl-4-azido-2,3,6-tri-O-benzyl-4-deoxy-L-glycero-α-L-manno- heptopyranosyl acetate (5), a key intermediate for the synthesis of novel anti-cancer antibiotic, spicamycin (1), starting from D-ribose is described.

Full text of DOI:10.1246/cl.2003.190

190
Chemistry Letters Vol.32, No.2 (2003)
The New and Efficient Synthesis of a Heptose Moiety of Spicamycin
Tamotsu Suzuki and Noritaka Chidaꢀ
Department of Applied Chemistry, Faculty of Science and Technology, Keio University,
Hiyoshi, Kohoku-ku, Yokohama 223-8522
(Received November 7, 2002; CL-020947)
The new and efficient synthesis of 7-O-acetyl-4-azido-2,3,6-
the natural product, it is now required for us to develop more
practical synthetic way to 5 for the large-scale preparation
and structure-activity relationship study of novel spicamycin
derivatives. It is also an important task to establish an
effective synthetic route to heptopyranoses from readily
available material, since several seven-carbon sugars struc-
turally related to 5 are known to be constituents of biologi-
tri-O-benzyl-4-deoxy-L-glycero-a-L-manno-heptopyranosyl
acetate (5), a key intermediate for the synthesis of novel anti-
cancer antibiotic, spicamycin (1), starting from D-ribose is
described.
cally
significant
antibiotic
(septacidin)7a
and
Spicamycin (1), isolated by Hayakawa and co-workers from
culture broth of Streptomyces as a differentiation inducer of HL-
60 human promyelocytic leukemia cells, have been reported to
show high anti-cancer activities.1 Spicamycin consists of a novel
aminoheptose (4-amino-4-deoxy-L-glycero-L-manno-hepto-
pyranose), adenine, glycine, and fatty acids. Numerous
spicamycin derivatives possessing various fatty acid moieties
were prepared2;3 from spicamycin amino nucleoside 3 (SAN,
obtained by acid hydrolysis of natural spicamycin) to
generate clinically promising compounds, SPM VIII2 2 and
KRN 5500.4
lipopolysaccharides.7b In this paper, we report a new and
efficient synthesis of 5 starting from D-ribose.
The known methyl 2,3-di-O-benzyl-b-D-ribofuranoside8
(7ꢀ), chosen as the starting material, was prepared from D-
ribose in 4step reactions (Scheme 2). Thus, O-benzylation of
an anomeric mixture of methyl 6-O-trityl-a- and b-riborufa-
nosides (obtained from D-ribose in quantitative yield, a:b =
1:3.7), followed by chromatographic separation afforded 6ꢀ
and 6ꢁ in 74% and 20% isolated yields, respectively.
Deprotection of a trityl ether in 6ꢀ with p-TsOH in MeOH
gave 7ꢀ in 91% yield along with the isomerized product 7ꢁ in
4% yield. Similar treatment of 6ꢁ also gave 7ꢀ (90%) and 7ꢁ
(9%), respectively, and further amount of 7ꢀ was obtained in
88% yield by acid-catalyzed isomerization of 7ꢁ. An
aldehyde 89 derived from alcohol 7ꢀ by Swern oxidation
(94%) was subjected to the two-carbon homologation by
reactions with vinylmagnesium bromide or vinyllithium.
However, low yields of the desired product 9 and observed
poor stereoselectivities (up to 65% combined yields, 9:10 =
1:3–1.3:1) led us to explore other approaches.
Recently, we reported the first total synthesis of one of
spicamycin congeners, SPM VIII 2,5 in which Pd-catalyzed
coupling reaction6 of heptopyranosylamine 4 with a 6-chloro-
purine derivative was employed as the key transformation, and
successfully confirmed its proposed absolute structure.1 The
novel heptopyranosylamine 4 with L-glycero-L-manno config-
uration was synthesized by stereoselective introduction of an
amino function into a-pyranosyl acetate 5, which was
prepared from naturally abundant cyclitol, myo-inositol5
(Scheme 1). Although the first (and to date only) synthesis of
5 from myo-inositol, which involved the optical resolution
during the synthetic process, was useful for the determination
of the synthetically unconfirmed absolute stereochemistry of
Scheme 2. Tr = –CPh3. Reagents and conditions: i HCl–MeOH, 0 ꢃC;
ii TrCl, pyridine, DMAP, rt; iii BnCl, NaH, DMF, rt; iv p-TsOHꢁH2O,
MeOH, rt; v DMSO, (COCl)2, CH2Cl2, ꢂ78 ꢃC to ꢂ50 ꢃC then Et3N,
ꢂ50 ꢃC; vi CH2=CHMgBr or CH2=CHLi.
Jones oxidation of 7ꢀ, followed by condensation with
MeONHMeꢁHCl provided Weinreb’s amide 11 in 80% yield
from 7ꢀ (Scheme 3). The amide 11 was then treated with
vinylmagnesium bromide to afford 12 in 98% yield. Stereo-
selective reduction of 12 was achieved with NaBH4–CeCl3ꢁ7H2O
at ꢂ78 ꢃC to provide allylic alcohols 9 and 10 in 85% and 11%
Scheme 1. Bn = –CH2Ph.
Copyright Ó 2003 The Chemical Society of Japan

Chemistry Letters Vol.32, No.2 (2003)
191
other biological important natural heptopyranoses starting
from readily available carbohydrates.
References and Notes
1
a) Y. Hayakawa, M. Nakagawa, H. Kawai, K. Tanabe, H.
Nakayama, A. Shimazu, H. Seto, and N. Otake, J. Antibiot., 36,
934(1983). b) Y. Hayakawa, M. Nakagawa, H. Kawai, K. Tanabe,
H. Nakayama, A. Shimazu, H. Seto, and N. Otake, Agric. Biol.
Chem., 49, 2685 (1985). c) T. Sakai, K. Shindo, A. Odagawa, A.
Suzuki, H. Kawai, K. Kobayashi, Y. Hayakawa, H. Seto, and N.
Otake, J. Antibiot., 48, 899 (1995).
2
3
M. Kamishohara, H. Kawai, A. Odagawa, T. Isoe, J. Mochizuki, T.
Uchida, Y. Hayakawa, H. Seto, T. Tsuruo, and N. Otake, J.
Antibiot., 46, 1439 (1993).
a) M. Kamishohara, H. Kawai, A. Odagawa, T. Isoe, J. Mochizuki,
T. Uchida, and N. Otake, J. Antibiot., 47, 1305 (1994). b) T. Sakai,
H. Kawai, M. Kamishohara, A. Odagawa, A. Suzuki, T. Uchida, T.
Kawasaki, T. Tsuruo, and N. Otake, J. Antibiot., 48, 504(1995). c)
T. Sakai, H. Kawai, M. Kamishohara, A. Odagawa, T. Tsuruo, and
N. Otake, J. Antibiot., 48, 1467 (1995).
i
Scheme 3. MOM = –CH2OMe, Pr = –CHMe2, Tf = –SO2CF3.
Reagents and conditions: i Jones reagent, acetone, 0 ꢃC; ii MeONH-
i
MeꢁHCl, WSC, HOBt, Pr2NEt, DMF, rt; iii CH2=CHMgBr, THF,
ꢂ78 ꢃC; iv NaBH4, CeCl3ꢁ7H2O, MeOH–CH2Cl2, (1/1), ꢂ78 ꢃC; v
i
CSA, PrOH, 120 ꢃC, in a sealed tube, 48 h; vi MOMCl, NaH, THF,
4a) M. Kamishohara, H. Kawai, T. Sakai, T. Isoe, K. Hasegawa, J.
Mochizuki, T. Uchida, S. Kataoka, H. Yamaki, T. Tsuruo, and N.
Otake, Oncol. Res., 6, 383 (1994). b) M. Kamishohara, H. Kawai, T.
Sakai, T. Uchida, T. Tsuruo, and N. Otake, Cancer Chemother.
Pharmacol., 38, 495 (1996).
0 ꢃC; vii K2OsO4ꢁ2H2O, NMO, tBuOH–H2O (3/1), rt; viii BnBr, NaH,
Bu4NI, DMF, rt; ix 4mol/l aqueous HCl–THF (1/2), rt; x Tf 2O, DMAP,
pyridine, CH2Cl2, 0 ꢃC; xi Me3SiN3, KF, 18-crown-6-ether, MeCN, rt;
xii Ac2O, AcOH, FeCl3ꢁ6H2O, H2SO4, rt.
5
a) T. Suzuki, S. Tanaka, I. Yamada, Y. Koashi, K. Yamada, and N.
Chida, Org. Lett., 2, 1137 (2000). b) T. Suzuki, S. T. Suzuki, I.
Yamada, Y. Koashi, K. Yamada, and N. Chida, J. Org. Chem., 67,
2874(2002).
N. Chida, T. Suzuki, S. Tanaka, and I. Yamada, Tetrahedron Lett.,
40, 2573 (1999).
a) H. Agahigian, G. D. Vickers, M. H. von Saltza, J. Reid, A. I.
Cohen, and H. Gauthier, J. Org. Chem., 30, 1085 (1965). b) J. F.
Kenedy and C. A. White, ‘‘Bioactive Carbohydrates: In Chemistry,
Biochemistry and Biology,’’ Ellis Horwood Limited, Chichester
(1983), Chap. 8, p 172.
a) R. R. Schmidt and A. Gohl, Chem. Ber., 112, 1689 (1979). b) E.
Kawashima, K. Umabe, and T. Sekine, J. Org. Chem., 67, 5142
(2002).
All new compounds were characterized by 1H and 13C NMR, IR,
and mass spectrometric and/or elemental analyses.
isolated yields, respectively. Conversion of methyl furanoside 9
into a pyranoside form was successfully carried out by transgly-
cosylation of 9 with 2-propanol in the presence of CSA10 (120 ꢃC
in a sealed tube) to give a-pyranoside 13 as the major product in
60% yield along with its b-furanoside isomer (35% yield).11
Protection of thehydroxy function in 13 asa methoxymethylether
afforded 14 in 84% yield.
Dihydroxylation of 14 with catalytic amount of potassium
osmate in the presence of N-methylmorpholin-N-oxide (NMO)
proceeded stereoselectively to afford 15 and its diastereomer in
82% and 18% isolated yields, respectively. Diol 15 was
transformed into tetra-O-benzyl ether 16 in 93% yield. Deprotec-
tion of a MOM ether in 16 gave 17, and the resulting hydroxy
group in 17 was transformed into an inverted azide via
trifluoromethanesulfonate12 18 to give 19 in 52% yield from 16.
Treatment of 19 with AcOH–Ac2O inthe presence of sulfuric acid
and FeCl313 induced the acetolysis of the anomeric center as well
as the primary benzyl ether to furnish the known a-pyranosyl
acetate 5 in 64% yield. Spectral and physical data of 514 were fully
identical with those of the authentic sample prepared from myo-
inositol.5 Compound 5 had been converted into SPM VIII 2 via 4
in 7 step reactions.5b
6
7
8
9
10 When this reaction was carried out in MeOH (CSA, 100 ꢃC, in a
sealed tube for 60 h), the desired methyl a-pyranoside was obtained
as the minor product (33% yield), and the starting methyl furanoside
9 was recovered in 57% yield.
11 Acidic treatment (CSA in PrOH, 120 ꢃC) of b-furanoside gave
i
additional 13 in 50% yield (28% recovery of the starting b-
furanoside).
ꢀ
12 A. Martꢁn, T. D. Butters, and G. W. J. Fleet, Tetrahedron:
Asymmetry, 10, 2343 (1999).
In summary, the new synthetic way to a-pyranosyl acetate 5,
a key intermediate for the synthesis of spicamycin, starting from
D-ribose has been established. The present synthesis, pro-
ceeded in 16 steps and in 7.3% overall yield from D-ribose, is
more efficient than our previous approach5 (24steps, 0.4%
overall yield from myo-inositol), and provides a practical
method for the synthesis of novel spicamycin derivatives
possessing various heterocyclic bases with an exocyclic
amino group (guanine, cytosine, and so on), which are
expected to show clinically interesting activities. Further
studies on preparation and biological assessment of such
derivatives are currently underway and will be reported in due
course. Moreover, it is noteworthy that the synthetic method
developed in this work should be applicable to preparation of
13 D. R. McPhail, J. R. Lee, and B. Fraser-Reid, J. Am. Chem. Soc.,
114, 1905 (1992).
14Data for 5: [a]23 ꢂ50 (c 0.43, CHCl3) {lit.5b [a]22 ꢂ47 (c 0.72,
D
D
1
CHCl3)}; H NMR (300 MHz, CDCl3) d 2.01 and 2.04(2s, each
3H), 3.66 (dd, 1H, J ¼ 1:3 and 10.0 Hz), 3.67 (dd, 1H, J ¼ 2:0 and
2.9 Hz), 3.72 (dd, 1H, J ¼ 2:9 and 10.0 Hz), 3.89 (ddd, 1H, J ¼ 1:3,
4.4 and 7.1 Hz), 4.16 (dd, 1H, J ¼ 10:0 and 10.0 Hz), 4.24 (dd, 1H,
J ¼ 7:1 and 12.0 Hz), 4.34 (dd, 1H, J ¼ 4:4 and 12.0 Hz), 4.57 and
4.70 (2s, each 2H), 4.62 and 4.71 (2d, each 1H, J ¼ 12:0 Hz), 6.17
(d, 1H, J ¼ 2:0 Hz), 7.26–7.39 (m, 15H); 13C NMR (75 MHz,
CDCl3) ꢀ 20.9, 21.0, 57.8, 64.0, 71.8, 71.9, 72.5, 72.9, 73.9, 77.4,
77.5, 91.5, 127.6, 127.8, 127.9, 128.0, 128.0, 128.3, 128.5, 137.2,
137.4, 137.9, 168.7, 170.7. HRMS (EI) Calcd for
C32H35O8N3(Mþ): 589.2424, Found: m=z 589.2436.