Preparative and Stereoselective Synthesis of the Versatile Intermediate for Carbocyclic Nucleosides: Effects of the Bulky Protecting Groups to Enforce Facial Selectivity

Article abstract of DOI:10.1021/jo0356762

The preparative and stereoselective synthesis (45-50% overall yields) of the target compound 17 has been accomplished from D-ribose. The bulky protecting groups such as TBDPS and Trityl enforced the facial selectivity during Grignard reaction to give the

Full text of DOI:10.1021/jo0356762

P r ep a r a tive a n d Ster eoselective Syn th esis
of th e Ver sa tile In ter m ed ia te for
Ca r bocyclic Nu cleosid es: Effects of th e
Bu lk y P r otectin g Gr ou p s to En for ce F a cia l
Selectivity
Won J un Choi, Hyung Ryong Moon, Hea Ok Kim,
Byul Nae Yoo, J eong A Lee, Dae Hong Shin, and
Lak Shin J eong*
Laboratory of Medicinal Chemistry, College of Pharmacy,
Ewha Womans University, Seoul 120-750, Korea
lakjeong@ewha.ac.kr
F IGURE 1. Potent carbocyclic nucleosides with cyclopentenyl
sugar moiety.
Received November 14, 2003
suffered from inconsistent and low overall yields, lengthy
synthetic routes, racemization, lack of large-scale prepa-
rations, and sensitivity to reaction conditions such as
temperature and moisture. Thus, a short and efficient
procedure to the D-carbocycle is highly desirable. Re-
cently, J acobson and co-workers have published the
elegant synthesis of the carbocycle moiety of neplanocin
A from D-ribono-γ-lactone, using olefin ring closing me-
tathesis (RCM) as a key step.6 More recently, short and
efficient syntheses of D- and L-3-unsubstituted cyclopen-
tenones, which employ RCM as the key step, have been
reported from our laboratory7 and by Chu et al.,8 and
these substrates serve as versatile precursors for the
synthesis of D- and L-carbocyclic nucleosides. Thus, we
are very interested in developing a new and efficient
synthetic route to the D-cyclopentenone derivative with
a 3-hydroxymethyl side chain for the extensive modifica-
tion of the cyclopentenyl sugar moiety. The highlights of
our synthesis are the stereoselective formation of the
tertiary â-allylic alcohol 16 during the Grignard reaction,
which is enforced by the bulky protecting group, and the
oxidative rearrangement of the tertiary â-allylic alcohol
to the key molecule 17.
Abst r a ct : The preparative and stereoselective synthesis
(45-50% overall yields) of the target compound 17 has been
accomplished from D-ribose. The bulky protecting groups
such as TBDPS and Trityl enforced the facial selectivity
during Grignard reaction to give the tertiary â-allylic alcohol
16 as the sole product, which was oxidatively rearranged to
the key molecule 17 in excellent yield.
A naturally occurring nucleoside, neplanocin A (1)1 is
a representative of the carbocyclic nucleosides and shows
potent antiviral and antitumor activities by inhibiting
S-adenosylhomocysteine hydrolase (Figure 1). Its fluo-
rocyclopentene analogue, fluoroneplanocin A (2),2 was
also reported to be a potent irreversible inhibitor of
S-adenosylhomocysteine hydrolase and exhibit potent
antitumor and antiviral activities. On the other hand,
the cytidine analogue 3 showed potent antitumor activity
by reducing cytidine-5′-triphosphate (CTP) pools and in
addition displayed potent antiviral activity by inhibiting
CTP synthetase.3
However, although a variety of biological activities of
these carbocyclic nucleosides have stimulated medicinal
chemists to carry out structure-activity relationship
(SAR) studies in carbocyclic nucleosides, synthetic dif-
ficulties in preparing the D-carbocycle have hampered
them from pursuing the extensive SAR study. Thus,
modifications have mainly been done on the base moiety,
not on the cyclopentenyl sugar part. Chu and co-workers
reported the structure-activity relationship study, based
on the modification of the adenine base, in which 5-fluo-
rocytosine analogue 4 exhibited potent antiviral activity
against West-Nile virus.4
First, we tried to synthesize 10 with a benzyl protecting
group, which is widely used in nucleoside chemistry
(Scheme 1). D-Ribose was converted to the acetonide 5.9
Wittig reaction of 5 with methyl ylide followed by
selective benzyl protection, using organotin chemistry,
yielded benzyl ether 6. Direct benzylation of 5 or diol
derivative obtained after Wittig reaction with NaH/BnBr
(5) (a) Arita, M.; Adachi, K.; Ito, Y.; Sawai, H.; Ohno, M. J . Am.
Chem. Soc. 1983, 105, 4049. (b) Medich, J . R.; Kunnen, K. B.; J ohnson,
C. R. Tetrahedron Lett. 1987, 28, 4131. (c) Ali, S. M.; Ramesh, K.;
Borchardt, R. T. Tetrahedron Lett. 1990, 31, 1509. (d) Marquez, V. E.;
Lim, M.-I.; Tseng, C. K. H.; Markovac, A.; Priest, M. A.; Khan, M. S.;
Kaskar, B. J . Org. Chem. 1988, 53, 5709. (e) Wolfe, M. S.; Anderson,
B. L.; Borcherding, D. R.; Borchardt, R. T. J . Org. Chem. 1990, 55,
4712. (f) Bestmann, H. J .; Roth, D. Angew. Chem., Int. Ed. Engl. 1990,
29, 99.
Although many synthetic methods to the carbocyclic
moiety have been reported so far,5 they sometimes
(1) (a) Yaginuma, S.; Muto, N.; Tsujino, M.; Sudate, Y.; Hayashi,
M.; Otani, M. J . Antibiot. (Tokyo) 1980, 34, 359. (b) Hayashi, M.;
Yaginuma, S.; Yoshioka, H.; Nakatsu, K. J . Antibiot. (Tokyo) 1981,
34, 675.
(2) J eong, L. S.; Yoo, S. J .; Lee, K. M.; Koo, M. J .; Choi, W. J .; Kim,
H. O.; Moon, H. R.; Lee, M. Y.; Park, J . G.; Lee, S. K.; Chun, M. W. J .
Med. Chem. 2003, 46, 201.
(3) Marquez, V. E.; Lim, M.-I.; Treanor, S. P.; Plowman, J .; Priest,
M. A.; Markovac, A.; Khan, M. S.; Kaskar, B.; Driscol, J . S. J . Med.
Chem. 1988, 31, 1687.
(6) Lee, K.; Cass, C.; J acobson, K. A. Org. Lett. 2001, 3, 597.
(7) (a) Choi, W. J .; Park, J . G.; Yoo, S. J .; Kim, H. O.; Moon, H. R.;
Chun, M. W.; J ung, Y. H.; J eong, L. S. J . Org. Chem. 2001, 66, 6490.
(b) Moon, H. R.; Choi, W. J .; Kim, H. O.; J eong, L. S. Tetrahedron:
Asymmetry 2002, 13, 1189.
(8) J in, Y. H.; Chu, C. K. Tetrahedron Lett. 2002, 43, 4141.
(9) Moon, H. R.; Choi, W. J .; Kim, H. O.; J eong, L. S. Tetrahedron:
Asymmetry 2002, 13, 1189 and references therein.
(4) Song, G. Y.; Paul, V.; Choo, H.; Morrey, J .; Sidwell, R. W.;
Schinazi, R. F.; Chu, C. K. J . Med. Chem. 2001, 44, 3985.
10.1021/jo0356762 CCC: $27.50 © 2004 American Chemical Society
Published on Web 03/06/2004
2634
J . Org. Chem. 2004, 69, 2634-2636

SCHEME 1a
SCHEME 2
SCHEME 3
a
Reagents and conditions: (a) c-H2SO4, acetone, rt, 2.5 h; (b)
Ph3PCH3Br, KOt-Bu, THF, rt, 12 h; (c) (i) Bu2Sn(O), toluene, 15
h, (ii) TBAI, BnBr, 75 °C, 16 h; (d) (COCl)2, DMSO, CH2Cl2, -78
°C, 1 h, then Et3N, rt, 1 h; (e) CH2dCHMgBr, THF, -78 °C, 1 h;
(f) Grubbs catalyst, CH2Cl2, rt, 2 d; (g) (1) PCC, MS, benzene or
DMF, (2) PDC, DMF, ClCH2CH2Cl, CH2Cl2 or DMSO, or (3) CrO3,
pyridine, Ac2O, CH2Cl2; (h) PDC, MS, DMF, rt, 18 h.
resulted in no differentiation between primary alcohol
and secondary alcohol, giving a low yield of primary
benzyl ether. Swern oxidation of 6 gave ketone 7 (92%),
which was subjected to the Grignard reaction with
vinylmagnesium bromide to give diene 8 (95%) as an
inseparable mixture of diastereomeric tertiary alcohols.
Ring-closing metathesis (RCM) reaction of diene 8 with
use of the 2nd generation Grubbs catalyst10 afforded 9a
(79%) and 9b (17%) after separation on silica gel. Relative
stereochemistry of 9a and 9b was easily assigned on the
Thus, the next goal was to increase the formation of
the tertiary allylic â-alcohol-like 9b. This was achieved
by changing the benzyl protecting group to bulky groups
such as tert-butyldimethylsilyl (TBS), tert-butyldiphen-
ylsilyl (TBDPS), or trityl (Tr), which were envisaged to
control the stereoselectivity in the Grignard reaction. As
shown in Scheme 2, acetonide 5 was protected as bulky
ethers 11a -c, which were converted to the dienes 14a -
c, respectively, by a similar procedure as before. Grignard
reactions of 13a -c with vinylmagnesium bromide af-
forded the dienes 14a -c as inseparable diastereomers
but with the opposite predominant diastereomer from
that observed from the benzyl-protected precursors. The
ratios of diastereomers were determined after RCM
reaction, as illustrated in Scheme 3.
1
basis of H NOE experiment. Although 9a or 9b can be
converted to the desired intermediate 10 in 4 steps
according to J ohnson’s procedure,5b we pursued one-step
conversion of 9a or 9b to 10 using oxidative rearrange-
ment to increase overall yields. However, oxidative
rearrangement of the tertiary allylic alcohol 9a to cyclo-
pentenone 10 failed under the various oxidation condi-
tions,11 while the minor isomer 9b was smoothly con-
verted to 105d on stirring with PDC in DMF in 75% yield.
This result clearly indicates that the tertiary chromate
ester of 9a could not form the six-membered transition
state to be rearranged to 10 because of the steric
hindrance by the 2,3-isopropylidene group.
RCM reaction of 14a with the TBS protecting group
gave the undesired 15a 12 (8%) and desired 16a 12 (75%),
while use of the bulkier protecting groups TBDPS and
(12) While preparing this manuscript, the same approach to com-
pounds 15a and 16a was reported by Chu et al., who obtained 15a
and 16a in a 1/10 ratio: (a) Chu, C. K.; J in, Y. H.; Baker, R. O.;
Huggins, J . Bioorg. Med. Chem. Lett. 2003, 13, 9. (b) J in, Y. H.; Liu,
P.; Wang, J .; Baker, R.; Huggins, J .; Chu, C. K. J . Org. Chem. 2003,
68, 9012.
(10) (a) Grubbs, R. H.; Miller, S. J . Acc. Chem. Res. 1995, 28, 446.
(b) Grubbs, R. H.; Chang, S. Tetrahedron 1988, 54, 4413.
(11) Harrowven, D. C.; Lucas, M. C.; Howes, P. D. Tetrahedron Lett.
2000, 41, 8985.
J . Org. Chem, Vol. 69, No. 7, 2004 2635

Hz), 5.25 (m, 2 H), 5.13 (dt, 1 H, J ) 1.2, 10.0 Hz), 4.68 (t, 1 H,
J ) 7.6 Hz), 4.47 (d, 1 H, J ) 6.8 Hz), 3.81 (d, 1 H, J ) 9.6 Hz),
3.49 (d, 1 H, J ) 9.6 Hz), 1.50 (s, 3 H), 1.40 (s, 3 H), 1.08 (s, 9
H); 13C NMR (100 MHz, CDCl3) δ 138.4, 135.8, 135.8, 135.4,
133.1, 133.0, 130.0, 130.0, 127.9, 127.9, 117.9, 116.1, 108.4, 79.5,
78.9, 75.7, 68.7, 27.5, 27.0, 25.1, 19.5. Anal. Calcd for C27H36O4-
Si: C, 71.64; H, 8.02. Found: C, 71.71; H, 8.42.
Tr resulted in the exclusive formation of the desired
tertiary â-alcohols, 16b and 16c, respectively.13 Thus, it
is evident that the size of the protecting group played a
major role in controlling the stereochemistry of the
carbon carrying the tertiary hydroxyl group during the
Grignard reaction. More systematic studies are needed
to elucidate the origin of the distinct different stereo-
chemical outcome of vinylmagnesium bromide addition
to the carbonyl group of R′,R,â-trioxygenated ketones (7,
13a -c), since several chelated as well as nonchelated
transition states are feasible.14
With the desired isomers in hand, various oxidation
reactions using the chromium reagents11 (PCC, PDC, and
CrO3) in various solvents (CH2Cl2, DMSO, ClCH2CH2Cl,
and DMF) were tried to convert the tertiary allylic
alcohols, 16a , 16b, and 16c, to their corresponding
cyclopentenones, 17a , 17b, and 17c,15 among which PDC
oxidation in DMF gave the best yields (84-92%).
In summary, we have developed a short, efficient, and
preparative synthesis of our target compound 17 with
various protecting groups, starting from D-ribose in 7
steps and in 45-50% overall yields. To the best of our
knowledge, our synthetic method can be regarded as an
excellent procedure from the viewpoint of number of
steps, overall yields, large-scale preparation, and mild
reaction conditions and has a great potential to be
utilized extensively in the SAR study of the carbocyclic
nucleosides.
A Typ ica l P r oced u r e for th e RCM Rea ction . To a stirred
solution of 14b (14.42 g, 31.86 mmol) in methylene chloride (100
mL) was added tricyclohexylphosphine[1,3-bis(2,4,6-trimeth-
ylphenyl)-4,5-dihydroimidazol-2-ylidene][benzylidine]ruthenium-
(VI) dichloride (270 mg, 0.32 mmol), and the reaction mixture
was stirred at room temperature for 2 d. The volatiles were
removed under reduced pressure and the residue was purified
by silica gel column chromatography (hexane:ethyl acetate )
5:1) to give (1S,4S,5S)-(+)-4,5-O-isopropylidene-1-(tert-butyl-
diphenylsilyloxymethyl)-2-cyclopenten-1-ol (16b) (12.11 g, 95%)
as a colorless oil: [R]25 +5.58 (c 1.26, CHCl3); 1H NMR (400
D
MHz, CDCl3) δ 7.74-7.35 (m, 10 H), 5.97 (dd, 1 H, J ) 1.6, 5.6
Hz), 5.74 (d, 1 H, J ) 5.6 Hz), 5.33 (d, 1 H, J ) 5.6 Hz), 4.54 (d,
1 H, J ) 5.2 Hz), 4.02 (d, 1 H, J ) 10.4 Hz), 3.67 (d, 1 H, J )
10.4 Hz), 1.32 (s, 3 H), 1.26 (s, 3 H), 1.09 (s, 9 H); 13C NMR (100
MHz, CDCl3) δ 136.0, 135.9, 135.7, 134.7, 133.2, 133.0, 130.0,
130.0, 128.0, 127.8, 112.2, 85.3, 85.1, 84.8, 66.0, 27.6, 27.0, 26.3,
19.5. Anal. Calcd for C25H32O4Si: C, 70.72; H, 7.60. Found: C,
70.98; H, 7.99.
A Typ ica l P r oced u r e for th e Oxid a tive Rea r r a n gem en t.
A solution of 16b (12.06 g, 28.40 mmol), 4 Å molecular sieves
(14.2 g), and pyridinium dichromate (32.05 g, 85.20 mmol) in
DMF (100 mL) was stirred at room temperature for 2 d. After
the mixture was diluted with diethyl ether and ethyl acetate,
the mixture was filtered through a short pad of a mixture of
silica gel and Celite. The filtrate was evaporated and the
resulting residue was purified by silica gel column chromatog-
raphy (hexane:ethyl acetate ) 4:1) to give (4R,5R)-(+)-3-tert-
butyldiphenylsilyloxymethyl-4,5-O-isopropylidene-2-cyclopen-
tenone (17b) (10.08 g, 84%) as a colorless oil: [R]25D +6.1 (c 2.29,
CH2Cl2); 1H NMR (400 MHz, CDCl3) δ 7.68-7.37 (m, 10 H), 6.34
(ps t, 1 H, J ) 2.0 Hz), 4.99 (d, 1 H, J ) 5.6 Hz), 4.70 (dd, 1 H,
J ) 2.0, 19.2 Hz), 4.51 (dd, 1 H, J ) 1.6, 19.2 Hz), 4.50 (d, 1 H,
J ) 5.6 Hz), 1.36 (s, 3 H), 1.34 (s, 3 H), 1.10 (s, 9 H); 13C NMR
(100 MHz, CDCl3) δ 201.9, 177.0, 135.7, 135.6, 132.8, 132.8,
130.3, 128.1, 128.1, 115.6, 78.3, 77.9, 62.6, 27.6, 26.9, 26.5, 19.5.
Anal. Calcd for C25H30O4Si: C, 71.05; H, 7.16. Found: C, 70.98;
H, 6.95.
Exp er im en ta l Section
A Typ ica l P r oced u r e for th e Gr ign a r d Rea ction . To a
stirred solution of 13b (14.53 g, 34.22 mmol) in THF (150 mL)
was added dropwise vinylmagnesium bromide (68.44 mL, 68.44
mmol, 1.0 M solution in THF) at -78 °C, and the reaction
mixture was stirred for 1 h at the same temperature. The
reaction mixture was quenched by saturated ammonium chloride
solution and brine and extracted with ethyl acetate. The organic
layer was dried over anhydrous magnesium sulfate, filtered, and
evaporated. The resulting oil was purified by column chroma-
tography (hexane:ethyl acetate ) 9:1) to give (1S,4R,5S)-(+)-1-
(2,2-dimethyl-5-vinyl-1,3-dioxolan-4-yl)-1-(tert-butyldiphenylsi-
lyloxymethyl)-2-propen-1-ol (14b) (13.01 g, 84%) as a colorless
Ack n ow led gm en t. This work was supported by a
grant from the Korea Health R&D Project, Ministry of
Health & Welfare, Korea (HMP-01-PJ 1-PG1-01CH13-
0002).
oil: [R]25 +17.99 (c 1.45, CHCl3); 1H NMR (400 MHz, CDCl3) δ
D
7.70-7.36 (m, 10 H), 6.11 (m, 2 H), 5.44 (dd, 1 H, J ) 1.6, 17.6
(13) Relative stereochemistry of RCM products was also determined,
based on 1H NOE experiments as in the case of 9a and 9b.
(14) (a) Mead, K.; Macdonald, T. L. J . Org. Chem. 1985, 50, 422. (b)
Marco, J . A.; Carda, M.; Gonzalez, F.; Rodriguez, S.; Castillo, E.; Murga,
J . J . Org. Chem. 1998, 63, 698. (c) Hui, C. W.; Lee, H. K.; Wong, H. N.
C. Tetrahedron Lett. 2002, 43, 123.
Su p p or tin g In for m a tion Ava ila ble: Full experimental
details and characterization data for all other compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(15) Shiozaki, M.; Kobayashi, Y.; Arai, M.; Haruyama, H. Tetrahe-
dron Lett. 1994, 35, 887.
J O0356762
2636 J . Org. Chem., Vol. 69, No. 7, 2004