Chemistry Letters Vol.37, No.4 (2008)
449
[ꢂext 253 (ꢀ" ꢁ20:3) and 236 (ꢀ" +1.9) nm, MeOH] of
dibenzoate 9 was similar to that of 10 [ꢂext 253 (ꢀ" +20.4)
and 237 (ꢀ" ꢁ8:1) nm, MeOH]15 except for the sign. Therefore,
the absolute configuration of the 5-deoxy-ꢀ-xylofuranose unit
of 1 was determined to be D.
Compounds 1 and 6 were isolated from two unrelated
marine organisms, the ascidian Diplosoma sp. and the alga
Hypnea valendiae, which supports the possibility of microbial
origin of these compounds. Compound 1 was found to cause
complete inhibition of cell division in fertilized sea urchin eggs
at a concentration of 1 mg/mL and showed weak activity against
HCT116 cells (human colorectal cancer cells) with an IC50
of >20 ppm.
NH2
I
N
H
N
N
H
CH3
O
OH
H
H
H
NOE
H
OH
Figure 1. Selected NOEs for compound 1.
50-deoxypentose. Major differences in the 1H–1H coupling con-
stants of 1 and 6 were observed for H10 [5.95 (d, J ¼ 1:8 Hz) in
1, 6.00 (d, J ¼ 5:2 Hz) in 6], H20 [4.12 (dd, J ¼ 1:8, 3.7 Hz) in 1,
4.39 (q, J ¼ 5:2 Hz) in 6], H30 [3.81 (dd, J ¼ 3:7, 4.3 Hz) in 1,
3.84 (q, J ¼ 5:2 Hz) in 6] and H40 [4.20 (dq, J ¼ 3:7, 6.7 Hz)
in 1, 3.89 (dq, J ¼ 5:2, 6.3 Hz) in 6]. The sugar in 1 differed
significantly from that of 6 in terms of 13C NMR chemical shifts
(Table 1).
The relative stereochemistry of the sugar moiety in 1
was determined by NOE experiment (Figure 1), chemical trans-
formation (acetalization), and comparison of the appropriate
1H NMR resonances with those of the known nucleoside 7
and synthetic ꢀ-xylofuranosyl nucleosides.5b,8 The coupling
constants for H10/H20, H20/H30, H30/H40, and H40/H50 in
References and Notes
1
Nat. Prod. Rep. 2002, 19, 1, and previous reports in this series. c)
J. W. Blunt, B. R. Copp, M. H. G. Munro, P. T. Northcote, M. R. J.
2
3
4
Taxonomical assignment was performed by Prof. Euichi Hirose,
University of the Ryukyus.
To the best of our knowledge, there have been no reports on naturally
occurring xylofuranosyl or 50-deoxyxylofuranosyl nucleosides.
1991, 52, 269. c) Y. Kato, N. Fusetani, S. Matsunaga, K. Hashimoto,
`
928. g) K. Kondo, H. Shigemori, M. Ishibashi, J. Kobayashi, Tetra-
a) R. Kazlauskas, P. T. Murphy, R. J. Wells, J. A. Baird-Lambert,
D. D. Jamieson, Aust. J. Chem. 1983, 36, 165. b) G. Cimino, A. Crispino,
N. Lindquist, W. Fenical, D. F. Sesin, C. M. Ireland, G. D. V. Duyne,
1: [ꢃ]2D6 ꢁ69ꢂ (c 0.1, MeOH); IR (film) 3461, 3317, 3132, 1633, 1584,
1474, 1084, 755 cmꢁ1; UV (MeOH) ꢂmax 283 nm (" 3900); LR-EIMS
m=z (rel. %) 376 (Mþ, 7), 303 (3), 289 (13), 261 (33), 260 (100),
233 (18).
0
0
0
0
0
0
0
0
1 (J1 ;2 ¼ 1:8 Hz, J2 ;3 ¼ 0 Hz, J3 ;4 ¼ 3:7 Hz, and J4 ;5
¼
6:7 Hz) were comparable to those of a ꢀ-xylofuranoside deriva-
0
0
0
0
0
0
tive 7 (J1 ;2 ¼ 1:4 Hz, J2 ;3 ¼ 0 Hz, J3 ;4 ¼ 6:7, 3.2 Hz, and
J4 ;5 ¼ 6:8 Hz). A cis relationship between H10 and H40 was
inferred from the NOE data, while the data obtained for H10/
OH20, H20/OH30, H40/OH20, CH350/OH30, and CH350/H2
implied that the sugar in 1 was ꢀ-50-deoxyxylose (Figure 1).
Treatment of 1 with 2,2-dimethoxypropane and CSA (24 h
at rt and then 4 h at 45 ꢂC) afforded acetal 89 rather than the ace-
0
0
5
6
7
1
tonide. In the H NMR spectrum of this compound, the proton
signal [5.77 (d, J ¼ 3:7 Hz)] corresponding to OH20 in 1 disap-
peared, and two new methyl proton signals [1.17 (3H, s), 1.32
(3H, s)] and a methoxy proton signal [2.95 (3H, s)] appeared,
indicating the presence of an acetal group in 8. In addition, a
proton signal [4.12 (dd, J ¼ 1:8, 3.7 Hz)] assigned as H20 in 1
was shifted to ꢁH 4.22 (s) in 8. These changes revealed that
the sterically less hindered OH group at C20 underwent addition
to 2,2-dimethoxypropane. This result, with no formation of
acetonide, provided evidence for a trans relationship between
OH20 and OH30. Thus, the sugar moiety in 1 was concluded to
be ꢀ-50-deoxyxylose. All attempts to hydrolyze 1 were unsuc-
cessful, resulting in decomposition of the reaction products.
Naturally occurring xylose is known to be a D-series sugar. How-
ever, in view of the fact that a small but significant amount of
both (+)- and (ꢁ)-isomers are present in marine natural prod-
ucts,6,10 we attempted to determine the absolute stereochemistry
of marine metabolite 1 by CD measurement. A pronounced
negative Cotton effect which was seen in the CD spectrum of
1 [ꢂext 242 (ꢀ" ꢁ1:9) and 210 (ꢀ" ꢁ2:6) nm, EtOH] suggested
purin-9-yl ꢀ-D-xylofuranosides.11 In addition, to confirm the
absolute stereostructure of the 5-deoxy-ꢀ-xylofuranose unit
of 1, we synthesized dibenzoates 9 and 1012 by treatment of
methyl 5-deoxy-ꢀ-L-xylofuranoside13,14 and 1 with 4-bromoben-
zoyl chloride, DMAP and pyridine (24 h at rt). The CD spectrum
8
9
G. Gosselin, M.-C. Bergogne, J. de Rudder, E. De Clercq, J.-L. Imbach,
8: HR-FABMS m=z 449.0677 (M + H)þ (calcd for C15H22IN4O4,
449.0686); 1H NMR [(CD3)2SO, 500 MHz] ꢁ 1.17 (3H, s, Me), 1.23
(3H, d, J ¼ 4:6 Hz, Me50), 1.32 (3H, s, Me), 2.95 (3H, s, OMe), 3.88
(1H, dd, J ¼ 3:4, 4.6 Hz, H30), 4.12 (1H, qd, J ¼ 3:4, 6.4 Hz, H40),
4.22 (1H, brs, H20), 5.83 (1H, d, J ¼ 4:6 Hz, OH30), 6.04 (1H, d,
J ¼ 2:0 Hz, H10), 7.63 (1H, s, H8), 8.11 (1H, s, H2).
10 O. Richou, V. Vaillancourt, D. J. Faulkner, K. F. Albizati, J. Org. Chem.
12 9: HR-ESIMS m=z 534.9340 (M + Na)þ (calcd for C20H18Br2NaO6:
534.9368); 1H NMR (CDCl3, 400 MHz) ꢁ 7.92 (d, J ¼ 8:8 Hz, 2H),
7.88 (d, J ¼ 8:8 Hz, 2H), 7.59 (d, J ¼ 8:8 Hz, 2H), 7.58 (d, J ¼
8:8 Hz, 2H), 5.51 (d, J ¼ 5:4 Hz, 1H), 5.44 (s, 1H), 5.01 (s, 1H), 4.69
(m, 1H), 3.46 (s, 3H), 1.32 (d, J ¼ 6:8 Hz, 3H); 10: HR-ESIMS m=z
740.8857 (M + H)þ (calcd for C25H20Br2IN4O5: 740.8845); 1H NMR
(CDCl3, 400 MHz) ꢁ 8.21 (s, 1H), 7.90 (d, J ¼ 8:3 Hz, 2H), 7.89 (d,
J ¼ 8:3 Hz, 2H), 7.66 (d, J ¼ 8:3 Hz, 2H), 7.60 (d, J ¼ 8:3 Hz, 2H),
7.43 (s, 1H), 6.51 (s, 1H), 5.74 (s, 1H), 5.64 (d, J ¼ 3:4 Hz, 1H), 4.69
(m, 1H), 1.42 (d, J ¼ 6:4 Hz, 3H).
´
15 Positive chirality between the p-bromobenzoyl chromophores of 10
also indicated that the sugar moiety in 1 was of the D-xylose series
(20R and 30S configuration).16