Stereoselective synthesis of 1′-C-branched arabinofuranosyl nucleosides via anomeric radicals generated by 1,2-acyloxy migration

Article abstract of DOI:10.1021/jo020620d

Stereoselective C-C bond formation at the anomeric position of uracil and adenine nucleoside has been accomplished through reaction of the anomeric radical, generated by 1,2-acyloxy migration, with a radical acceptor. The present method consists of the fo

Full text of DOI:10.1021/jo020620d

Substrates for 1,2-acyloxy migration were synthesized
through bromo-pivaloyloxylation of 3′,5′-O-(1,1,3,3-tetra-
isopropyldisiloxane-1,3-diyl) (TIPDS)-protected 3 and
3′,5′-O-(di-tert-butylsilylene) (DTBS)-protected 1′,2′-un-
saturated uracil nucleosides 4, which were prepared by
selenoxide syn-elimination of the corresponding 2′-
phenylseleno derivatives 1 and 2.9 The bromo-pivaloyl-
oxylation of 3 gave the R-anti-adduct 5 (52%) as the major
product along with the isomeric 6 (26%). It was found
that the R-anti-adduct 7 (67%) was the sole product in
the reaction of 4. The stereochemical assignments of 5
and 6 can be deduced from their J 2′,3′ values (5, J ) 4.8
Hz; 6, J ) 0.7 Hz), since the smaller J value corresponds
to the dihedral angle close to 90° that is possible only
for 6. The stereochemistry of 7 was determined on the
basis of X-ray crystallography.6
Ster eoselective Syn th esis of 1′-C-Br a n ch ed
Ar a bin ofu r a n osyl Nu cleosid es via
An om er ic Ra d ica ls Gen er a ted by
1,2-Acyloxy Migr a tion
Kazuhiro Haraguchi,* Yoshiharu Itoh,
Kouichiro Matsumoto, Kyoko Hashimoto,
Kazuo T. Nakamura, and Hiromichi Tanaka
School of Pharmaceutical Sciences, Showa University,
1-5-8 Hatanodai, Shinagawa-ku, Tokyo 142-8555, J apan
harakazu@pharm.showa-u.ac.jp
Received September 24, 2002
Abstr a ct: Stereoselective C-C bond formation at the
anomeric position of uracil and adenine nucleoside has been
accomplished through reaction of the anomeric radical,
generated by 1,2-acyloxy migration, with a radical acceptor.
The present method consists of the following steps: (1)
electrophilic addition (bromo-pivaloyloxylation) to 3′,5′-O-
(1,1,3,3-tetraisopropyldisiloxane-1,3-diyl)-protected 1′,2′-
unsaturated nucleoside, (2) tin radical-mediated reaction of
the resulting adduct with a radical acceptor. The use of allyl-
(tributyl)tin gave the 1′-C-allylated uracil nucleoside 14 in
66% yield together with the unrearranged 2′-C-allylated
product 15 (6%). Radical acceptors such as styryl(tributyl)-
tin and 3-bromo-2-methylacrylonitrile can also be used in
the reaction of 5, giving 16 (70%) and 17 (76%) without the
formation of unrearranged product. The radical-mediated
C-C bond formation of the adenine counterpart 12 was also
investigated.
Free radical reaction processes are recognized as being
increasingly important as a synthetic method in organic
chemistry,1 due to the mild and neutral reaction condi-
tions that permit stereoselective entry to highly func-
tionalized molecules such as natural products.2 Glycosyl
anomeric radicals have been used extensively for the
synthesis of C-glycosides through reaction with acceptors
such as alkenes and alkynes.3 In the field of nucleoside
chemistry, however, much effort regarding anomeric
radicals has been devoted to the studies of DNA damage.4
In this paper, we describe the stereoselective synthesis
of arabinofuranosyl nucleosides branched at the anomeric
position by way of reaction of the nucleoside anomeric
radical, generated by 1,2-acyloxy migration,5-7 with a
radical acceptor.8
(1) (a) Giese, B. Radicals in Organic Synthesis: Formation of
Carbon-Carbon Bonds; Pergamon: Oxford, 1986. (b) Motherwell, W.
B.; Crich, D. Free Radical Chain Reactions in Organic Synthesis;
Academic Press: London, 1992. (c) Renaud, P.; Sibi, M. P. Radicals in
Organic Synthesis; Wiley-VCH: Weinheim, Germany, 2001.
(2) J asperse, C. P.; Curran, D. P.; Fevig, T. L. Chem. Rev. 1991, 91,
1237-1286.
(3) (a) Collins, P. M.; Ferrier, R. J . Monosaccharides; Their
Chemistry and Their Roles in Natural Products; J ohn Wiley & Sons:
Chichester, U.K., 1995. (b) Postema, M. H. D. C-GlycosideSynthesis;
CRC Press: Boca Raton, FL, 1995.
(4) 1′-C-Acylated derivatives of 2′-deoxyuridine were synthesized as
anomeric radical precursors. (a) Greenberg, M. M.; Yoo, D. J .;
Goodman, B. K. Nucleosides Nucleotides 1997, 16, 33-40. (b) Chat-
gilialoglu, C.; Gimisis, T. Chem. Commun. 1998, 1249-1250. (c)
Chatgilialoglu, C.; Ferreri, C.; Gimisis, T. Tetrahedron Lett. 1999, 40,
2837-2840.
(5) For a review, see: Beckwith, A. L.; Crich, D.; Duggan, P. J .; Yao,
Q. Chem. Rev. 1997, 97, 3273-3312.
(6) Haraguchi, K.; Tanaka, H.; Matsumoto, K.; Nakamura, K. T.;
Miyasaka, T. Tetrahedron Lett. 1995, 36, 3867-3870.
(7) Kinetic studies of the 1,2-acyloxy migaration have been re-
ported: (a) Gimisis, T.; Ialongo, G.; Zamboni, M.; Chatgilialoglu, C.
Tetrahedron Lett. 1995, 36, 6781-6784. (b) Gimisis, T.; Ialongo;
Chatgilialoglu, C. Tetrahedron 1998, 54, 573-592. (c) Chatogilialoglu,
C. Nucleosides Nucleotides 1999, 18, 547-553.
(8) Recently, intra- and intermolecular C-C bond formation of
nucleoside-1-yl has been reported: (a) (Intramolecular reaction) Kodama,
T.; Shuto, S.; Nomura, M.; Matsuda, A. Chem. Eur. J . 2001, 7, 2332-
2340. (b) (Intermolecular reaction) Kumamoto, H.; Murasaki, M.;
Haraguchi, K.; Anamura, A.; Tanaka, H. J . Org. Chem. 2002, 67,
6124-6130.
10.1021/jo020620d CCC: $25.00 © 2003 American Chemical Society
Published on Web 01/30/2003
2006
J . Org. Chem. 2003, 68, 2006-2009

SCHEME 1
Although the 3′,5′-O-TIPDS-protected 1′,2′-unsaturated
adenosine 8, prepared according to a reported proce-
dure,10,11 showed a stereochemical trend similar to that
of 3 in its bromo-pivaloyloxylation, the yield of the
adducts was rather low (10, 39%; 11, 3%) and their
chromatographic separation from succinimide was dif-
ficult. The best results were obtained by employing the
N6-pivaloyl derivative 9 as a substrate and 1,3-dibromo-
5,5-dimethylhydantoin (dibromatin) as a brominating
agent. Thus, when 9 was reacted with dibromatin (4
equiv) in the presence of t-BuCO2H (10 equiv) and Et3N
(10 equiv) in CH2Cl2, the R-anti-adduct 12 was formed
in 57% yield along with the â-anti-adduct 13 (12%).12
dition-elimination mechanism, such as styryl(tributyl)-
tin and 3-bromo-2-methylacrylonitrile, worked more ef-
ficiently in the reaction of 5 under the photochemical
conditions to give the respective products uniformly in
higher yields: 16, 76%; 17, 70%.15
We first investigated the synthesis of 1′-C-branched
uracil nucleosides. When 5 was treated with allyl-
(tributyl)tin (3.4 equiv)/AIBN (0.7 equiv) in refluxing
benzene for 2 h, initially formed C-2′ radical A was
transformed into nucleoside anomeric radical B13 via 1,2-
acyloxy migration, and subsequent reaction of B with
allyltributyltin gave the â-anomer of 1′-C-allylarabino-
furanosyl derivative 1414 in 52% yield stereoselectively,
along with eliminated product 3 (11%) (Scheme 1). Under
photochemically initiated conditions using (Bu3Sn)2 at
room temperature in benzene, 14 was formed with an
improved yield of 66% as well as the 2′-C-allylated
byproduct 15 (6%, an isomeric mixture at C-2′) derived
from A. The fact that 15 was isolated instead of 3 in this
latter reaction indicates that the higher temperature
employed in the former reaction caused the elimination
pathway. Radical acceptors that react through an ad-
The 3′,5′-O-DTBS derivative 7 was not a suitable
substrate for 1,2-acyloxy migration due to its rigid
conformation.6 However, its ready accessibility from 4 as
well as the right stereochemistry led us to replace the
3′,5′-O-protecting group of 7 with acetyl groups to make
the molecule much more flexible. Compound 7 was
treated with Bu4NF in THF containing Ac2O at room
temperature for 2 h to give the 3′,5′-di-O-acetyl derivative
18 in 78% yield. When a benzene solution of 18 was
irradiated in the presence of allyl(tributyl)tin/(Bu3Sn)2,
the 1′-C-allylated product 19 was obtained in 62% yield
along with 20 (10%) (Scheme 2). From the observed yield
of 19, which is comparable to that of 14, it would be
reasonable to say that the 3′,5′-O-TIPDS protection
provides enough flexibility for the present 1,2-acyloxy
migration. As was seen for 5, styryl(tributyl)tin reacted
with 18 more efficiently than allyl(tributyl)tin, giving 21
(a diastereomeric mixture, E/Z ) 10/1) in 71% yield
without forming the 2′-C-styrylated byproduct. This
result may be a reflection of the difference in reactivity
of these two radical acceptors. Namely, allyl(tributyl)tin
(9) Haraguchi, K.; Tanaka, H.; Maeda, H.; Itoh, Y.; Miyasaka, T. J .
Org. Chem. 1991, 56, 5401-5408.
(10) Fukukawa, K.; Ueda, T.; Hirao, T. Chem. Pharm. Bull. 1983,
31, 1582-1592.
(11) Robins, M. J .; J ones, R. A. J . Org. Chem. 1974, 39, 113-115.
(12) The stereochemistry of these adducts was determined on the
basis of NOE experiments: 12, H-2′/H-3′ and H-4′/H-8; 13, H-2′/H-4′.
(13) Although B is depicted as an sp2-hybridized radical, an alterna-
tive possibility is that B exists in equilibrium of two sp3-hybridized
species with opposite configurations. For examples of 2′-deoxyuridin-
1-yl radical, see: Chatgilialoglu, C.; Ferreri, C.; Bazzanini, R.; Buerra,
M.; Choi, S.-Y.; Emanuel, C. J .; Horner, J . H.; Newcomb, M. J . Am.
Chem. Soc. 2000, 122, 9525-9533.
(15) Although styryl(tributyl)stannane and 3-bromo-2-methylacrylo-
nitrile used were a mixture of geometric isomers, the respective
products 16 and 17 appeared to consist of only the depicted isomer.
The configurations about the 1′- and 2′-positions of these products were
confirmed on the basis of the observation of NOE correlations between
H-3′ and H-6 and H-2′ and H-4′, respectively.
(14) The stereochemistry of 14 was confirmed on the basis of the
observation of NOEs between H-3′/H-6 and H-2′/H-4′.
J . Org. Chem, Vol. 68, No. 5, 2003 2007

SCHEME 2
(arabinofuranosyl)uracil and -adenine nucleosides based
on a combination of bromo-pivaloyloxylation to 1′,2′-
unsaturated nucleoside and C-C bond formation of the
nucleoside anomeric radical resulting from 1,2-acyloxy
migration. The present radical migration (from C-2′ to
the anomeric position) is unique in that it is the reverse
pathway to that observed in carbohydrate systems (from
the anomeric position to C-2). Reaction of the nucleoside
anomeric radical with radical acceptors proceeded
stereoselectively, due to the steric repulsion exerted by
the 2′-“up”-pivaloyloxy group (the anti-rule).16 It is note-
worthy that a radical acceptor, which reacts through an
addition-elimination mechanism such as styryl(tributyl)-
tin and 3-bromo-2-methylacrylonitrile, is a suitable choice
of reagent for this tandem 1,2-acyloxy migration/inter-
molecular C-C bond formation methodology. Since the
compounds synthesized in this study are medicinally
interesting arabinofuranosyl nucleosides, we are cur-
rently evaluating their biological activities.
could be more reactive than styryl(tributyl)tin, and
therefore, allylation takes place before migration of the
1′-C-pivaloyloxy group.
We next extended the radical-based C-C bond forma-
tion to the synthesis of 1′-C-branched arabinofuranosyl
adenines by using the 3′,5′-O-TIPDS derivative 12. When
12 was reacted with 5.0 equiv of allyl(tributyl)tin in
benzene for 4 h under photochemical conditions as
for the allylation reaction of TIPDS-protected uracil
nucleoside 5, unrearranged 2′-C-allylated byproduct 23
was formed in 24% yield and the yield of the desired
product 22 was only 42%. This is a likely consequence of
Exp er im en ta l Section
Melting points are uncorrected. 1H NMR was measured at
400 or 500 MHz. Chemical shifts are reported relative to that
of Me4Si for 1H NMR. Mass spectroscopy (MS) was done in FAB
mode (m-nitrobenzyl alcohol as a matrix). Column chromato-
graphy was carried out on silica gel (silica gel 60). Thin-layer
chromatography (TLC) was performed on silica gel (precoated
silica gel plate F254). HPLC was carried out on a Shimadzu LC-
6AD with a shim-pack PREP-SIL(H)‚KIT column (2 × 25 cm).
1-[2-O-P iva loyl-1-C-(E )-st yr yl-3,5-O-(1,1,3,3-t e t r a iso-
pr opyldisiloxan e-1,3-diyl)-â-D-ar abin ofu r an osyl]u r acil (16).
A benzene solution (15 mL) of 5 (117.7 mg, 0.18 mmol), Bu3-
SnCHdCHPh (0.36 mL, 0.90 mmol), and (Bu3Sn)2 (98 µL, 0.18
mmol) was irradiated with a high-pressure mercury lamp under
an Ar atmosphere at rt for 4 h. Column chromatography of the
reaction mixture (hexane/AcOEt ) 5/1) gave 16 (85.3 mg, 70%)
as a syrup: UV (MeOH) λmax 255 (ꢀ 23700), λmin 225 (ꢀ 6800);
1H NMR (CDCl3) δ 1.05-1.06 (28H, m), 1.12 (9H, s), 3.90 (1H,
dd, J ) 8.4 and 11.4 Hz), 4.19 (1H, dd, J ) 3.7 and 11.4 Hz) δ
4.04-4.08 (1H, m), 4.38-4.41 (1H, m), 5.68 (1H, d, J ) 8.4 Hz),
5.74 (1H, d, J ) 2.6 Hz), 6.79 and 6.88 (2H, each as d, J ) 16.1
Hz), 7.27-7.34 and 7.42-7.53 (6H, each as m), 7.84 (1H, d, J )
the highly reactive nature of the allylating agent, as
mentioned above. A simple solution would be, therefore,
to reduce the amount of the reagent. In fact, by using
2.5 equiv of allyl(tributyl)tin, the yield of 22 was im-
proved to 52% at the expense of 23 (11%). The best re-
sult was obtained by reducing the amount to 1.5 equiv:
22, 60%; 23, 5%.
On the basis of the results in the synthesis of 16 and
17, one would anticipate that styryl(tributyl)tin and
3-bromo-2-methylacrylonitrile might react with the
adenine nucleoside anomeric radical more efficiently. In
fact, by using 5 equiv of these radical acceptors, the
desired 1′-C-styryl(arabinofuranosyl)adenine 24 and
1′-C-acrylyl(arabinofuranosyl)adenine 25 compounds
were obtained as a mixture of geometric isomers in 65%
and 73% yield, respectively, without the formation of
unrearranged product.
8.4 Hz), 8.51 (1H, br); FAB-MS m/z 695 (M+ + Na), 673 (M+
+
H). Anal. Calcd for C34H52N2O8Si2: C, 60.68; H, 7.79; N, 4.16.
Found: C, 60.58; H, 7.94; N, 4.13.
1-[1-C-Me t h a cr ylyl-2-O-p iva loyl-3,5-O-(1,1,3,3-t e t r a -
isop r op yld isiloxa n e-1,3-d iyl)-â-D-a r a bin ofu r a n osyl]u r a cil
(17). A benzene solution (10 mL) of 5 (69.8 mg, 0.11 mmol),
BrCHdC(CH3)CN (53.0 µL, 0.55 mmol), and (Bu3Sn)2 (0.30 mL,
0.55 mmol) was irradiated with a high-pressure mercury lamp
under an Ar atmosphere at rt for 4 h. Purification of the reaction
mixture by preparative TCL (hexane/AcOEt ) 4/1) gave 17 (52.9
mg, 76%) as a syrup: UV (MeOH) λmax 261 (ꢀ 10400), λmin 236 (ꢀ
(16) Renaud, P. Stereoselectivity of Radical Reactions: Cyclic
Systems. In Radicals in Organic Synthesis; Renaud, P., Sibi, M. P.,
Eds.; Wiley-VCH: Weinheim, Germany, 2001; Vol. 1, pp 400-415.
In conclusion, we have developed a novel method for
the stereoselective synthesis of 1′-C-carbon-substituted
2008 J . Org. Chem., Vol. 68, No. 5, 2003

4500); 1H NMR (CDCl3) δ 0.97-1.10 (28 H, m), 1.11 (9H, s), 2.03
(3H, d, J ) 1.8 Hz), 3.88 (1H, dd, J ) 7.0 and 11.7 Hz), 4.17
(1H, dd, J ) 3.7 and 11.7 Hz), 4.05-4.08 (1H, m), 4.31 (1H, dd,
J ) 3.7 and 4.9 Hz), 5.63 (1H, d, J ) 3.7 Hz), 5.74 (1H, d, J )
8.4 Hz), 6.94 (1H, d, J ) 1.8 Hz), 7.99 (1H, d, J ) 8.4 Hz), 8.37
(1H, br); FAB-MS m/z 658 (M+ + Na), 636 (M+ + H). Anal. Calcd
for C30H49N3O8Si2: C, 56.66; H, 7.77; N, 6.61. Found: C, 56.82;
H, 7.90; N, 6.43.
(7 mL) of 12 (50.9 mg, 0.07 mmol), Bu3SnCHdCHPh (131.7 mg,
0.34 mmol), and (Bu3Sn)2 (84.6 µL, 0.17 mmol) was irradiated
with a high-pressure mercury lamp under an argon atmosphere
at rt for 4 h. Purification of the reaction mixture by preparative
TCL (hexane/AcOEt ) 2/1) gave trans-24 (23.8 mg, 46%, syrup)
and cis-24 (10.4 mg, 19%, syrup).
Physical data for trans-24: UV (MeOH) λmax 273 (ꢀ 19000),
1
λmin 233 (ꢀ 9600); H NMR (CDCl3) δ 0.70 and 1.39 (18H, each
1-[3,5-Di-O-a cetyl-1-C-(E a n d Z)-styr yl-2-O-p iva loyl-â-D-
a r a bin ofu r a n osyl]u r a cil (21). A benzene (15 mL)/THF (0.2
mL) solution of 18 (155.8 mg, 0.32 mmol), Bu3SnCHdCHPh
(623.2 mg, 1.59 mmol), and (Bu3Sn)2 (31 µL, 0.06 mmol) was
irradiated with a high-pressure mercury lamp under an argon
atmosphere at rt for 4 h. Column chromatography of the reaction
mixture (hexane/AcOEt ) 2/1) gave 21 (115.8 mg, E/Z ) 10/1,
71%) as a syrup: UV (MeOH) λmax 255 (ꢀ 20100) and 225 (ꢀ
11200), λmin 226 (ꢀ 5600); 1H NMR (CDCl3) δ (E-isomer) 1.13 (9
H, s), 1.99 and 2.14 (6H, each as s), 4.32-4.36 (2H, m), 4.47
(1H, dd, J ) 7.4 and 14.2 Hz), 4.90 (1H, d, J ) 2.1 Hz), 5.73
(1H, dd, J ) 1.9 and 8.2 Hz), 5.84 (1H, s), 6.71 and 6.87 (2H,
each as d, J ) 16.2 Hz), 7.31-7.36, 7.41-7.44 (5H, each as m),
7.84 (1H, d, J ) 8.2 Hz); (Z-isomer/selected data) 1.06 (1H, s),
2.04 and 2.19 (6H, each as s), 5.71 (1H, s), 6.21 (1H, d, J ) 12.2
Hz); FAB-MS m/z 515 (M+ + H). Anal. Calcd for C26H30N2O9‚
1/4H2O: C, 60.17; H, 5.92; N, 5.44. Found: C, 60.07; H, 5.98; N,
5.21.
1-[1-C-Allyl-2-O-p iva loyl-3,5-O-(1,1,3,3-tetr a isop r op yld i-
s ilo x a n e -1,3-d iy l)-â-D-a r a b in o fu r a n o s y l]-N 6-p iv a lo y l-
a d en in e (22) a n d 1-[2-C-Allyl-2-d eoxy-1-p iva loyloxy-3,5-O-
(1,1,3,3-tetr aisopr opyldisiloxan e-1,3-diyl)-r-D-r ibofu r an osyl]-
N6-p iva loyla d en in e (23). A benzene solution (7 mL) of 12 (88.0
mg, 0.12 mmol), Bu3SnCH2CHdCH2 (57 µL, 0.18 mmol), and
(Bu3Sn)2 (60 µL, 0.12 mmol) was irradiated with a high-pressure
mercury lamp under an Ar atmosphere at rt for 4 h. Purification
of the reaction mixture by preparative TCL (hexane/AcOEt )
2/1) gave 22 (51.1 mg, 60%, syrup) and 23 (4.3 mg, 5%, syrup, a
mixture of diastereomers).
as s), 0.91-1.26 (28H, m), 3.99-4.03 (1H, m), 4.18 (1H, dd, J )
1.5 and 12.5 Hz), 4.32 (1H, dd, J ) 5.2 and 12.5 Hz), 5.29 (1H,
dd, J ) 6.4 and 7.6 Hz), 5.57 (1H, d, J ) 6.4 Hz), 7.09 (1H, d, J
) 15.9 Hz), 7.31 (1H, d, J ) 15.9 Hz), 7.36-7.39 and 7.52-7.54
(5H, each as m); FAB-MS m/z 561 (M+ + H). Anal. Calcd for
C40H61N5O7Si2‚3/4CH3CO2C2H5: C, 61.03; H, 7.98; N, 8.28.
Found: C, 60.66; H, 8.15; N, 8.14.
Physical data for cis-24: UV (MeOH) λmax 263 (ꢀ 17200), λmin
232 (ꢀ 7800); 1H NMR (CDCl3) δ 0.71 and 1.40 (18H, each as s),
0.91-1.15 (28H, m), 3.72-3.75 (1H, m), 3.88-3.89 (2H, m), 4.86
(1H, t, J ) 6.5 Hz), 5.66 (1H, d, J ) 6.5 Hz), 6.66 (1H, d, J )
12.5 Hz), 6.90 (1H, d, J ) 12.5 Hz), 7.02-7.04, 7.09-7.10 (5H,
each as m), 8.22 (1H, s), 8.71 (1H, s), 8.44 (1H, br); FAB-MS
m/z 561 (M+ + H). Anal. Calcd for C40H61N5O7Si2‚2/3H2O: C,
59.52; H, 7.99; N, 8.68. Found: C, 59.49; H, 8.04; N, 8.42.
1-[1-C-(E)-Meth a cr ylyl-2-O-p iva loyl-3,5-O-(1,1,3,3-tetr a -
isop r op yld isiloxa n e-1,3-d iyl)-â-D-a r a b in ofu r a n osyl]-N6-
p iva loyla d en in e (tr a n s-25) a n d 1-[1-C-(Z)-Meth a cr ylyl-2-
O-p iva loyl-3,5-O-(1,1,3,3-tetr a isop r op yld isiloxa n e-1,3-d iyl)-
â-D-a r a bin ofu r a n osyl]-N6-p iva loyla d en in e (cis-25). A ben-
zene solution (7 mL) of 12 (53.2 mg, 0.07 mmol), BrCHd
CH(CH3)CN (34.0 µL, 0.35 mmol), and (Bu3Sn)2 (0.19 mL, 0.35
mmol) was irradiated with a high-pressure mercury lamp under
an argon atmosphere at rt for 4 h. Purification of the reaction
mixture by preparative TCL (hexane/AcOEt ) 2/1) gave a
mixture of trans-25 (14.1 mg, 28%, syrup) and cis-25 (23.1 mg,
45%, syrup).
Physical data for trans-25: UV (MeOH) λmax 271 (ꢀ 16800),
1
λmin 236 (ꢀ 5500); H NMR (CDCl3) δ 0.73 and 1.40 (18H, each
Physical data for 22: UV (MeOH) λmax 264 (ꢀ 14000), λmin 230
(ꢀ 2900); 1H NMR (CDCl3) δ 0.73 and 1.41 (18 H, each as s),
0.98-1.34 (28H, m), 3.37 (1H, dd, J ) 7.0 and 14.8 Hz), 3.52
(1H, dd, J ) 7.3 and 14.8 Hz), 3.98-4.02 (1H, m), 4.14-4.18
(2H, m), 4.57 (1H, t, J ) 7.3 Hz)), 5.09 (1H, dd, J ) 1.4 and 10.3
Hz), 5.16 (1H, J ) 1.4 and 17.2 Hz), 5.69 (1H, d, J ) 7.3 Hz),
5.64-5.75 (1H, m), 8.41 and 8.71 (2H, each as s), 8.64 (1H, br);
FAB-MS m/z 718 (M+ + H). Anal. Calcd for C35H59N5O7Si2: C,
58.54; H, 8.28; N, 9.75. Found: C, 58.81; H, 8.25; N, 9.92.
Physical data for 23: UV (MeOH) λmax 259 (ꢀ 13900), λmin 227
(ꢀ 2800); 1H NMR (CDCl3) δ (major isomer) 1.02-1.16 (28H, m),
1.08 (9H, s), 1.41 (9H, s), 1.81-1.94 (1H, m), 3.12-3.15 (1H, m),
3.42-3.46 (1H, m), 4.14 (1H, dd, J ) 3.3 and 11.9 Hz), 4.18 (1H,
dd, J ) 3.7 and 11.9 Hz), 4.29-4.33 (1H, m), 4.44 (1H, dd, J )
1.5 and 15.9 Hz), 4.51-4.57 (2H, m), 5.43-5.54 (1H, m), 8.30
(1H, s), 8.60 (1H, br), 8.75 (1H, s); (minor isomer/selected data)
1.21 (9H, s), 1.40 (9H, s), 2.17-2.24 (1H, m), 2.48-2.53 (1H, m),
2.87-2.92 (1H, m), 3.88 (1H, dd, J ) 8.1 and 12.0 Hz), 3.98 (1H,
dd, J ) 7.3 and 12.0 Hz), 4.42-4.46 (1H, m), 4.91-4.99 (2H,
m), 5.09 (1H, dd, J ) 1.5 and 17.2 Hz), 5.63-5.70 (1H, m), 8.17
(1H, s), 8.54 (1H, br), 8.71 (1H, s); FAB-MS m/z 718 (M+ + H),
as s), 0.91-1.11 (28H, m), 1.89 (3H, d, J ) 1.8 Hz), 3.96-3.99
(1H, J ) 3.4 and 12.7 Hz), 4.14 (1H, dd, J ) 3.4 and 12.7 Hz),
4.22 (1H, dd, J ) 4.3 and 12.7 Hz), 4.92 (1H, t), 5.58 (1H, d,
J ) 7.0 Hz), 7.36 (1H, q, J ) 1.8 Hz), 8.23 (1H, s), 8.50 (1H, br),
8.70 (1H, s); FAB-MS m/z 781 (M+ + K). Anal. Calcd for
C
36H58N6O7Si2‚1/2AcOEt: C, 57.98; H, 7.94; N, 10.67. Found: C,
58.02; H, 8.00; N, 10.88.
Physical data for cis-25: UV (MeOH) λmax 271 (ꢀ 15900) λmin
1
261 (ꢀ 10000); H NMR (CDCl3) δ 0.77 and 1.41 (18H, each as
s), 0.91-1.16 (28H, m), 2.13 (3H, d, J ) 1.5 Hz), 4.18 (1H, dd,
J ) 3.1 and 13.1 Hz), 4.21-4.23 (1H, m), 4.31 (1H, dd, J ) 2.8
and 13.1 Hz), 4.74 (1H, t, J ) 7.8 Hz), 5.57 (1H, d, J ) 7.8 Hz),
7.65 (1H, d, J ) 1.5 Hz), 8.45 (1H, s), 8.58 (1H, br), 8.70 (1H, s);
FAB-MS m/z 781 (M+ + K). Anal. Calcd for C36H58N6O7Si2‚
1/2AcOEt: C, 57.98; H, 7.94; N, 10.67. Found: C, 57.68; H, 8.08;
N, 10.81.
Ack n ow led gm en t. Financial support from the Min-
istry of Education, Culture, Sports, Science and Tech-
nology (Grant-in-Aid No. 11771391 to K.H.) is gratefully
acknowledged.
-
616 (M+ t-BuCOO). Anal. Calcd for C35H59N5O7Si2: C,58.54;
H, 8.28; N, 9.75. Found: C, 58.69; H, 8.11; N, 10.09.
Su p p or tin g In for m a tion Ava ila ble: Detailed description
of the experimental procedures and compound characterization
of 1-15 and 18-20. This material is available free of charge
via the Internet at http://pubs.acs.org.
1-[2-O-P iva loyl-1-C-(E )-st yr yl-3,5-O-(1,1,3,3-t e t r a iso-
pr opyldisiloxan e-1,3-diyl)-â-D-ar abin ofu r an osyl]-N6-pivalo-
yla d en in e (tr a n s-24) a n d 1-[2-O-P iva loyl-1-C-(Z)-styr yl-3,5-
O-(1,1,3,3-t et r a isop r op yld isiloxa n e-1,3-d iyl)-â-D-a r a b in o-
fu r a n osyl]-N6-p iva loyla d en in e (cis-24). A benzene solution
J O020620D
J . Org. Chem, Vol. 68, No. 5, 2003 2009