Synthesis of 5-thiodidehydropyranylcytosine derivatives as potential anti-HIV agents

Article abstract of DOI:10.1016/j.bmcl.2011.04.006

As a part of our ongoing efforts to identify new anti-HIV agents, a 5′-thiopyrano-nucleoside derivative 4, designed based on 4′-thioD4C 1 and cyclohexenylnucleoside 3, was synthesized. The dihydrothiopyran skeleton of 4 was constructed by the ring closing metathesis of 21 which was synthesized from but-2-yne-1,4-diol. After converting the protecting group from MOM to TBS followed by oxidation, a Pummerer-type thioglycosylation reaction of 24 with persilylated uracil gave the desired 5-thiodihydrothiopyranyluracils 25 and 26 as a mixture of anomers. The conversion of 25 to a cytosine derivative and subsequent deprotection gave a 5-thiodidehydropyranosylcytosine derivative 4 in good yield. The anti-HIV activity of 4 was also evaluated.

Full text of DOI:10.1016/j.bmcl.2011.04.006

Bioorganic & Medicinal Chemistry Letters 21 (2011) 3313–3316
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis of 5-thiodidehydropyranylcytosine derivatives as potential anti-HIV
agents
Yuichi Yoshimura a, , Yoshiko Yamazaki a,b, Yukako Saito a, Yoshihiro Natori a, Tomozumi Imamichi c,
⇑
Hiroki Takahata a,
⇑
a Faculty of Pharmaceutical Sciences, Tohoku Pharmaceutical University, 4-4-1 Komatsushima, Aoba-ku, Sendai 981-8558, Japan
b Research and Development Department, SEED CO. Ltd, 1030-7 Fukuro, Kounosu, Saitama 369-0131, Japan
c Laboratory of Human Retrovirology, Applied and Development Research Program, National Institute of Allergy and Infectious Diseases at Frederick, Science Applications
International Corporation-Frederick, Inc., Frederick, MD 21702, USA
a r t i c l e i n f o
a b s t r a c t
As a part of our ongoing efforts to identify new anti-HIV agents, a 50-thiopyrano-nucleoside derivative 4,
designed based on 40-thioD4C 1 and cyclohexenylnucleoside 3, was synthesized. The dihydrothiopyran
skeleton of 4 was constructed by the ring closing metathesis of 21 which was synthesized from
but-2-yne-1,4-diol. After converting the protecting group from MOM to TBS followed by oxidation, a
Pummerer-type thioglycosylation reaction of 24 with persilylated uracil gave the desired 5-thiodihydro-
thiopyranyluracils 25 and 26 as a mixture of anomers. The conversion of 25 to a cytosine derivative and
subsequent deprotection gave a 5-thiodidehydropyranosylcytosine derivative 4 in good yield. The
anti-HIV activity of 4 was also evaluated.
Article history:
Received 19 February 2011
Revised 2 April 2011
Accepted 5 April 2011
Available online 9 April 2011
Keywords:
Nucleoside
40-Thionucleoside
Anti-HIV
Ó 2011 Elsevier Ltd. All rights reserved.
Antiviral
Pummerer reaction
HAART (Highly Active Anti-Retroviral Therapy) is the most suc-
cessful AIDS treatment to date for controlling HIV replication and
has greatly improved the lifespan of AIDS patients.1 However, the
emergence of drug resistant viruses during the course of chemo-
therapy for AIDS continues to be a serious problem even after
the establishment of HAART.2 The HAART protocol involves the
use of a cocktail of anti-HIV drugs including nucleoside reverse
transcriptase inhibitors (NRTIs), non-nucleoside reverse transcrip-
tase inhibitors (NNRTIs), and protease inhibitors (PIs).1 Since NRTIs
play an important role in HAART,2 the development of an effective
NRTI for HIV, which is also effective for a drug-resistant AIDS virus,
would be highly desirable. As a result, a number of nucleoside ana-
logues that are effective for treating HIV have been reported to
date.3 Our efforts in this area have involved attempts to prepare
new antiviral nucleoside derivatives that have the potential for
use in HIV therapy. Our strategy involved the construction of a
novel pseudosugar skeleton that could serve as a structurally novel
NRTI. Such a compound could then be used in place of currently
used NRTIs that are ineffective against viruses that are resistant
to the NRTIs now in use.4 We previously reported on the synthesis
of dihydrothiopyrano-nucleoside 24c,e which was designed based
thiopyrano-nucleoside 2 did not show anti-HIV activity. Cyclohex-
enylcytosine 3, however, was found to have weak anti-HIV
activity.4d Although these results were insufficient to obtain
antiviral nucleosides, we hypothesized that the transformation of
the carbocyclic derivative 3 into its 50-thio counterpart might
potentiate anti-HIV activity. To confirm this, we envisioned syn-
thesizing a dihydrothiopyranocytosine derivative 4 as a potential
HIV-nucleoside. In this Letter, we report on the synthesis of a race-
mic mixture of 5-thio-2,3-didehydropyranosylcytosine 4 starting
NH2
NH2
S
N
N
S
N
N
HO
O
O
HO
O
O
1
2
NH2
NH2
N
N
S
N
N
HO
HO
HO
HO
on the known anti-HIV L
-40-thioD4C 1.5 Unfortunately, dihydro-
3
4
⇑
Corresponding authors.
E-mail address: yoshimura@tohoku-pharm.ac.jp (Y. Yoshimura).
Chart 1.
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.04.006

3314
Y. Yoshimura et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3313–3316
from but-2-yne-1,4-diol and a preliminary evaluation of its anti-
HIV activity (Chart 1).
of diastereomers. The resulting sulfoxide 14, in hand, was treated
with bis(TMS)uracil in the presence of TMSOTf and excess iPr2NEt
to give the desired 50-thionucleoside 15 in 55% yield as an insepa-
As discussed above, we previously reported on the synthesis of
dihydrothiopyrano-nucleoside 2,4c,e which involves two key reac-
tions: (1) the construction of a dihydrothiopyran skeleton by
ring-closing metathesis (RCM) and (2) a Pummerer-type thiogly-
cosylation reaction. It should be possible to apply this basic
synthetic concept to the synthesis of 5-thio-2,3-didehydropyrano-
sylcytosine 4. Thus, we started the synthesis from but-2-yne-1,4-
diol to construct a dihydrothiopyran skeleton as a pseudosugar.
The synthesis of the homoallyl alcohol derivative 8 was accom-
plished in four steps: (1) the conversion of but-2-yne-1,4-diol 5 to
trans-but-2-ene-1,4-diol by hydroaluminamtion, (2) benzylation
of the hydroxyl groups, (3) epoxidation by mCPBA treatment, and
(4) cleavage of the resulting oxirane ring by an organocopper re-
agent.6 To obtain the intermediate for the RCM reaction, it is neces-
sary to introduce an allyl sulfide unit at the 2-position of 8. This was
achieved as follows. After mesylation of the secondary hydroxyl
group of 8, a nucleophilic substitution reaction by treatment with
potassium thioacetate in the presence of crown ether gave the thio-
acetate 10 in good yield. The acetyl group of 10 was removed by
treatment with NaOMe and the resulting thiol derivative was subse-
quently treated with allyl bromide to give the diene 11 in 85% yield.
The diene 11 was then subjected to the RCM reaction. From our pre-
vious results, it appeared possible that the use of the Grubbs 1st cat-
alyst might not be efficient in RCM reactions for compounds
containing a sulfur atom in their structure.4c,e Therefore, the Grubbs
2nd catalyst was used for the RCM reaction of 11. As expected, when
11 was reacted in the presence of 10 mol % of 2nd Grubbs catalyst in
refluxing benzene, the dihydrothiopyran 12 was produced in 78%
yield (Scheme 1).
rable mixture of a- and b-anomers (Scheme 2).
Although attempts were made to remove benzyl groups of 15,
the free nucleosides could not be obtained.8 Alternatively, depro-
tection of the dihydrothiopyran derivative 12 was also attempted,
which gave the deprotected products, but in poor yields. Failure to
obtain free nucleosides or pseudosugar derivatives can be attrib-
uted to the instability of 15 and 12 under the reductive or Lewis
acid conditions used (data not shown). These results prompted
us to synthesize the target compound using other protecting
groups.
After serious consideration, we re-examed the entire synthesis
starting from the MOM protected derivative 16. As can be seen
in Scheme 1, all reactions starting from 5 to 12 were reproducible
using a series of the MOM protected compounds and gave the de-
sired products 16–22 in good yields. It was necessary to convert
the MOM group of 22 into another protecting group that tolerates
the conditions for the Pummerer-type thioglycosylation reac-
tion.4c,e,7 To accomplish this, we attempted to convert 22 into a
TBS-protected derivative. Since the pseudosugar skeleton, as de-
scribed above, was assumed to be unstable under acidic conditions,
we used Fujioka’s method9 to deprotect the MOM group, since con-
ditions close to neutral are used for this method. Compound 22
was reacted with TMSOTf in the presence of 2,20-bipyridyl and
was subsequently treated with H2O to give the desired diol 13 in
55% yield.9 Protection of 13 with a TBS group using standard meth-
odology (TBSCl, imidazole) resulted in the predominant formation
of the monosilylated product (data not shown). Instead of TBSCl
and imidazole, the use of TBSOTf and 2,6-lutidine resulted in an
improved yield, giving the bis(TBS) derivative 23 in 83% yield
(Scheme 1).
Prior to carrying out the Pummerer-type thioglycosylation reac-
tion,4c,e,7 the dihydrothiopyran derivative 12 was converted into
the corresponding sulfoxide by treatment with NaIO4. The reaction
gave sulfoxide 14 in excellent yield as an inseparable 1:1 mixture
By the method described above, compound 23 was oxidized to
the corresponding sulfoxide 24, which was subjected to the
RO
RO
HO
OH
d
e
a, b or c
RO
RO
O
OR
OR
OH
6
: R = Bn (95%, 2 steps)
7
: R = Bn (99%)
8
: R = Bn (92%)
5
16
: R = MOM (90% 2 steps)
17
: R = MOM (95%)
18
: R = MOM (93%)
OMs
g
S
SAc
h
f
RO
RO
RO
RO
RO
RO
9
: R = Bn (97%)
10
20
: R = Bn (72%)
11
: R = Bn (85%)
19
: R = MOM (81%)
: R = MOM (79%)
21
: R = MOM (77%)
S
S
j
k
S
i
HO
RO
TBSO
HO
RO
TBSO
23
(83%)
13
(55%)
12
22
: R = Bn (78%)
: R = MOM (84%)
Scheme 1. Reagents and conditions: (a) LiAlH4, THF, reflux; (b) BnCl, NaH, THF, DMF, rt; or (c) MOMCl, iPr2NEt, CH2Cl2, 40 °C; (d) mCPBA, CH2Cl2, rt; (e) H2C@CHMgCl, CuI,
Et2O, ꢀ40 °C; (f) MsCl, Et3N, CH2Cl2, rt; (g) KSAc, 18-crown-6, CH3CN, reflux; (h) NaOMe, allyl bromide, CH3CN, rt; (i) 2nd Grubbs cat., benzene, reflux; (j) TMSOTf, 2,20-
bipyridyl, CH2Cl2, 0 °C, then H2O, Et2O, rt; (k) TBSOTf, 2,6-lutidine, CH2Cl2, rt.

Y. Yoshimura et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3313–3316
3315
O
O
S
a
a
S
12
23
BnO
BnO
TBSO
TBSO
24 (59%)
14 (97 %)
b
b
63 % (α: β = 1:1)
O
O
O
S
N
NH
BnO
BnO
S
N
NH
S
N
NH
TBSO
TBSO
O
TBSO
O
+
TBSO
O
α β
= 1:1)
15 (55 %,
:
25
26
Scheme 2. Reagents and conditions: (a) NaIO4, CH3CN, rt; (b) bis(TMS)uracil,
TMSOTf, iPr2NEt, PhCH3/CH2Cl2.
c
c
Pummerer-type thioglycosylation reaction.4c,e,7 The reaction of 24
with bis(TMS)uracil, TMSOTf and iPr2NEt afforded an anomeric
mixture of the 50-thionucleosides 25 and 26 in 63% yields. After
careful separation by silica gel column chromatography, the ste-
reochemistry at the 10-position of the isolated 25 and 26 was deter-
mined in NOE experiments as depicted in Figure 1.10
NH2
NH2
S
N
N
S
N
N
RO
RO
RO
RO
O
O
The uracil ring of the b-50-thionucleoside 25 was converted into
cytosine by treatment with triisopropylbenzenesulfonyl chloride
(TPSCl) and triethylamine in the presence of DMAP followed by
treatment with ammonium hydroxide11 to give 27 which con-
tained inseparable impurities. Finally, the TBS groups of 27 were
deprotected by treatment with TBAF to furnish the desired 5-thio-
27: R = TBS
4: R = H (72%, 2 steps)
28: R = TBS
29: R = H (74%, 2 steps)
d
d
Scheme 3. Reagents and conditions: (a) NaIO4, CH3CN, rt (recovery of 23: 20%); (b)
bis(TMS)uracil, TMSOTf, iPr2NEt, PhCH3/CH2Cl2; (c) TPSCl, DMAP Et3N, CH3CN, rt,
then NH4OH; (d) TBAF, THF.
didehydropyranosylcytosine 412 in 72% yield from 25. The
a-ano-
mer of 2913 was prepared from 26 following the same sequence
of reactions (Scheme 3).
Anti-HIV-1 activities of 5-thiodidehydropyranosylcytosine 4
and its anomer 29 were evaluated by the inhibition of HIV replica-
tion in peripheral blood mononuclear cells (PBMC), which was
determined by a p24 HIV antigen capture assay.14 The results are
summarized in Figure 2. As expected, the b-5-thiodidehydropyr-
anosylcytosine derivative 4 showed significant anti-HIV-1 activity
at concentrations of 10 and 100 nM. It is noteworthy that the
inhibitory activity of 4 at a concentration of 10 nM suppressed
HIV replication to a level of 65% of the control while Lamivudine
suppressed it to 37% at the same concentration. Unexpectedly,
the suppression of HIV replication for 4 at a concentration of
100 nM was reduced to the level of 75% of the control. Although
the reason for this is not clear at this time, the assay was performed
using a racemic mixture of the 50-thiopyrano-nucleoside and the
inactive enantiomer of 4 might have had an influence on the
Figure 2. Summary of anti-HIV-1 activities of 4 and 29. Anti-HIV-1 activity was
evaluated by the inhibition of HIV replication in peripheral blood mononuclear cells
(PBMC) from healthy volunteers, which was determined by a p24 HIV antigen
capture assay (Perkin–Elmer)14 using Lamivudine as a positive control (⁄p <0.01,
⁄⁄p <0.05). Statistical analyses were performed using the Student’s t-test.
anti-HIV activity of the mixture.15 The
a-anomer 29, on the other
hand, showed only weak activity compared with 4. This result is
consistent with the fact that the nucleosides having antiviral and
antitumor activities generally possess b-configurations, analogous
to naturally occurring nucleosides.
O
HN
H
NOE
H
H
In conclusion, we report on the design and synthesis of the no-
vel 50-thiodidehydropyrano-nucleosides 4 and 29. The findings
indicate that 4 has significant anti-HIV activity. To obtain more po-
tent analogues for inhibiting HIV replication, investigations on
structure–activity relationships including the enantioselective syn-
thesis of both enantiomers of 4 will be needed. These studies are
currently underway and the results will be reported in future
reports.
H
O
N
H
NOE
N
H
S
O
OTBS
N
H
OTBS
O
S
OTBS
26
H
OTBS
25
Figure 1. NOE experiments of 25 and 26.

3316
Y. Yoshimura et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3313–3316
7. (a) Yoshimura, Y.; Kuze, T.; Ueno, M.; Komiya, F.; Haraguchi, K.; Tanaka, H.;
Acknowledgments
Kano, F.; Yamada, K.; Asami, K.; Kaneko, N.; Takahata, H. Tetrahedron Lett. 2006,
47, 591; (b) Yoshimura, Y.; Kitano, K.; Yamada, K.; Satoh, H.; Watanabe, M.;
Miura, S.; Sakata, S.; Sasaki, T.; Matsuda, A. J. Org. Chem. 1997, 62, 3140; (c)
Yoshimura, Y.; Kitano, K.; Satoh, H.; Watanabe, M.; Miura, S.; Sakata, S.; Sasaki,
T.; Matsuda, A. J. Org. Chem. 1996, 61, 822.
This work was supported in part by a Grant-in-Aid for Scientific
Research (No. 21590119), JSPS (Y.Y.) and by High Technology Re-
search Program from Ministry of Education, Culture, Sports, Sci-
ence and Technology of Japan.
8. Attempts to remove the benzyl group involved the use of a Lewis acidic (BCl3 in
CH2Cl2) as well as reductive conditions (Li, NH3 and Na, naphthalene). The
reactions resulted in the formation of a complex mixture.
9. (a) Fujioka, H.; Kubo, O.; Senami, K.; Minamitsuji, Y.; Maegawa, T. Chem.
Commun. 2009, 4429; (b) Fujioka, H.; Okitsu, T.; Ohnaka, T.; Li, R.; Kubo, O.;
Okamoto, K.; Sawama, Y.; Kita, Y. J. Org. Chem. 2007, 72, 7898; (c) Fujioka, H.;
Sawama, Y.; Murata, N.; Okitsu, T.; Kubo, O.; Matsuda, S.; Kita, Y. J. Am. Chem.
Soc. 2004, 126, 11800.
References and notes
1. Esté, J. A.; Cihlar, T. Antiviral Res. 2010, 85, 25.
2. (a) Meadows, D. C.; Gervey-Hague, J. ChemMedChem 2006, 1, 16; (b) Imamichi,
T. Curr. Pharm. Des. 2004, 10, 4039.
10. Compound 25 showed
a
4.0% NOE at H-10 when irradiated at H-50 and
compound 26 a 7.7% NOE at H-50 when H-6 of the pyrimidine ring was
irradiated.
3. (a) Mehellou, Y.; De Clercq, E. J. Med. Chem. 2010, 53, 521. and references cited
therein; (b) Cihlar, T.; Ray, A. S. Antiviral Res. 2010, 85, 39. and references cited
therein; (c) Wang, Q.; Li, Y.; Song, C.; Qian, K.; Chen, C.-H.; Lee, K.-H.; Chang, J.
Bioorg. Med. Chem. Lett. 2010, 20, 4053; (d) Boyer, P. L.; Vu, B. C.; Ambrose, Z.;
Julias, J. G.; Warnecke, S.; Liao, C.; Meier, C.; Marquez, V. E.; Hughes, S. H. J. Med.
Chem. 2009, 52, 5356.
4. (a) Yamada, K.; Hayakawa, H.; Sakata, S.; Ashida, N.; Yoshimura, Y. Bioorg. Med.
Chem. Lett. 2010, 20, 6013; (b) Yoshimura, Y.; Asami, K.; Imamichi, T.; Kuroda,
T.; Shiraki, K.; Tanaka, H.; Takahata, H. J. Org. Chem. 2010, 75, 4161; (c)
Yoshimura, Y.; Yamazaki, Y.; Saito, Y.; Takahata, H. Tetrahedron 2009, 65, 9091;
(d) Yoshimura, Y.; Ohta, M.; Imahori, T.; Imamichi, T.; Takahata, H. Org. Lett.
2008, 10, 3449; (e) Yoshimura, Y.; Yamazaki, Y.; Kawahata, M.; Yamaguchi, K.;
Takahata, H. Tetrahedron Lett. 2007, 48, 4519; (f) Yoshimura, Y.; Asami, K.;
Matsui, H.; Tanaka, H.; Takahata, H. Org. Lett. 2006, 8, 6015.
11. Sekine, M.; Matsuzaki, J.; Satoh, M.; Hata, T. J. Org. Chem. 1982, 47, 571.
12. Data for 4: 1H NMR (400 MHz, CD3OD) d 2.69–2.73 (m, 1H), 2.98–3.04 (m, 1H),
3.45–3.59 (m, 4H), 5.78 (ddd, J = 1.4, 3.4, 11.1 Hz, 1H), 5.81 (d, J = 7.7 Hz, 1H),
5.89–5.90 (m, 1H), 6.14 (ddd, J = 2.4, 5.8, 11.1 Hz, 1H), 7.52 (d, J = 7.7 Hz, 1H).
13C NMR (100 MHz, CD3OD) d 38.8, 41.3, 52.1, 63.6, 66.2, 96.6, 124.8, 134.9,
143.5, 158.3, 167.5. EI-MS m/z 269 (M+). HRMS Calcd for C11H15N3O3S:
269.0834, Found 269.0829.
13. Data for 29: 1H NMR (400 MHz, CD3OD) d 2.25–2.42 (m, 1H), 3.00 (dd, J = 5.8,
13.0 Hz, 1H), 3.65–3.75 (m, 4H), 5.75 (ddd, J = 1.9, 4.3, 10.6 Hz, 1H), 5.79 (d,
J = 7.7 Hz, 1H), 6.09–6.14 (m, 2H), 7.61 (d, J = 7.7 Hz, 1H). 13C NMR (100 MHz,
CD3OD) d 40.5, 41.1, 52.9, 63.8, 63.9, 96.5, 125.3, 137.6, 143.8, 158.2, 167.5. EI-
MS: m/z 269 (M+). HRMS Calcd for C11H15N3O3S: 269.0834, Found 269.0830.
14. Brann, T. W.; Dewar, R. L.; Jiang, M.-K.; Shah, A.; Nagashima, K.; Metcalf, J. A.;
Falloon, J.; Lane, H. C.; Imamichi, T. J. Virol. 2006, 80, 6136.
5. Young, R. J.; Saw-Ponter, S.; Thomson, J. B.; Miller, J. A.; Cumming, J. G.; Pugh, A.
W.; Rider, P. Bioorg. Med. Chem. Lett. 1995, 5, 2599.
6. Hanazawa, T.; Koiwa, M.; Hareau, G. P.-J.; Sato, F. Tetrahedron Lett. 2000, 41,
2659.
15. To clarify an active enantiomer of 4 and influence to anti-HIV activity by the
inactive enantiomer, the study on the enantioselective synthesis of
currently under way.
4 is