Stereoselective synthesis of fluoro-homoneplanocin A as a potential antiviral agent

Article abstract of DOI:10.1021/ol300667q

Fluoro-homoneplanocin A (4) was synthesized from d-ribose, via the enyne ring-closing metathesis of 9, the stereoselective opening of epoxide 23a with fluoride, and a simultaneous oxidation-elimination reaction. The key intermediate 8 is expected to serve

Full text of DOI:10.1021/ol300667q

ORGANIC
LETTERS
2012
Vol. 14, No. 8
2134–2137
Stereoselective Synthesis of
Fluoro-homoneplanocin A as a Potential
Antiviral Agent
Girish Chandra, Mahesh S. Majik, Ji Yee Lee, and Lak Shin Jeong*
Department of Bioinspired Science and Laboratory of Medicinal Chemistry,
College of Pharmacy, Ewha Womans University, Seoul 120-750, Korea
lakjeong@ewha.ac.kr
Received March 16, 2012
ABSTRACT
Fluoro-homoneplanocin A (4) was synthesized from D-ribose, via the enyne ring-closing metathesis of 9, the stereoselective opening of epoxide
23a with fluoride, and a simultaneous oxidationꢀelimination reaction. The key intermediate 8 is expected to serve as a versatile intermediate for
the synthesis of carbanucleosides.
Neplanocin A (1, Figure 1),1 isolated from Aspergillus
niger, is one of the representative carbocyclic nucleosides
that inhibits S-adenosylhomocysteine (SAH) hydrolase.
SAH hydrolase catalyzes the hydrolysis of S-adenosylho-
mocysteine to adenosine and L-homocysteine. The inhibi-
tion of SAH hydrolase results in the accumulation of SAH
in the cell, which in turn inhibits S-adenosyl-L-methionine
(SAM)-dependent transmethylase.2 Because SAM-depen-
dent transmethylase is essential for the formation of the
capped methylated structure at the 50-terminus of viral
mRNA, SAH hydrolase is an attractive target for the
development of broad-spectrum antiviral agents.3 Nepla-
nocin A showed potent antiviral activities against several
RNA and DNA viruses but could not be further developed
as a clinical agent because of high cytotoxicity due to the
phosphorylation by adenosine kinase. Neplanocin A is
also metabolized by adenosine deaminase, yielding an
inactive hypoxanthine derivative.4
Based on the structure of neplanocin A (1), fluoro-
neplanocin A (2) was designed and synthesized as a
mechanism-based inhibitor of SAH hydrolase.5
For the synthesis of 2, electrophilic vinyl fluorination
was utilized asa key step. Compound 2 was two times more
potent than neplanocin A (1) against SAH hydrolase.
Unlike neplanocin A, which reversibly inhibits SAH hy-
drolase, fluoro-neplanocin A was found to inhibit the
enzyme through both mechanism-based irreversible inhi-
bition and mechanism-based reversible cofactor (NADþ)-
depletion.5 In addition, homoneplanocin A (3), a one-carbon
homologated analog of neplanocin A (1), also exhibited
potent inhibitory activity against SAH hydrolase.6
(4) Keller, B. T.; Borchardt, R. T. Metabolism and mechanism of
action of neplanocin A ꢀ A potent inhibitor of S-adenosylhomocysteine
hydrolase. In Biological methylation and drug design; Borchardt, R. T.,
Creveling, C. R., Ueland, P. M., Eds.; Humana Press: Clifton, NJ, 1986; pp
385ꢀ396.
(1) (a) Yaginuma, S.; Muto, N.; Tsujino, M.; Sudate, Y.; Hayashi,
M.; Otani, M. J. Antibiot. (Tokyo) 1981, 34, 359–366. (b) Hayashi, M.;
Yaginuma, S.; Yoshioka, H.; Nakatsu, K. J. Antibiot. (Tokyo) 1981, 34,
675–680.
(2) (a) Turner, M. A.; Yang, X.; Yin, D.; Kuczera, K.; Borchardt, R. T.;
Howell, P. L. Cell Biochem. Biophys. 2000, 33, 101–125. (b) Cantoni, G. L.
The centrality of S-adenosylhomocysteinase in the regulation of the
biological utilization of S-adenosylmethionine. In Biological methylation
anddrugdesign; Borchardt, R. T., Creveling, C. R., Ueland, P. M., Eds.; Humana
Press: Clifton, NJ, 1986; pp 227ꢀ238.
(5) (a) Jeong, 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, 201–203. (b) Lee, K. M.; Choi, W. J.; Lee,
Y.; Lee, H. J.; Zhao, L. X.; Lee, H. W.; Park, J. G.; Kim, H. O.; Hwang,
K. Y.; Heo, Y.-S.; Choi, S.; Jeong, L. S. J. Med. Chem. 2011, 54, 930–
938.
(3) Wolfe, M. S.; Borchardt, R. T. J. Med. Chem. 1991, 34, 1521–1530
and references cited therein.
(6) Shuto, S.; Obara, T.; Saito, Y.; Andrei, G.; Snoeck, R.; De Clercq,
E.; Matsuda, A. J. Med. Chem. 1996, 39, 2392–2399.
r
10.1021/ol300667q
Published on Web 04/11/2012
2012 American Chemical Society

elimination of the tertiary alcohol (Route B). Epoxide 11
would be synthesized from cyclopentenol 8 by manipulation
of the vinyl group followed by epoxidation.
Scheme 1. Retrosynthetic Analysis of the Desired Nucleoside 4
Figure 1. Rationale for the design of the target nucleoside 4.
Compound 3 exhibited potent antiviral activity against
hepatitis B virus (HBV) without apparent cytotoxicity.7
Based on these results, it was of great interest to synthe-
size fluoro-homoneplanocin A (4), which combines the
properties of fluoro-neplanocin A (2) and homoneplano-
cin A (3). In the synthesis of the target nucleoside, enyne
ring-closing metathesis (RCM), nucleophilic fluorination
to an epoxide, and simultaneous oxidationꢀelimination
were the key steps. Herein, we report the highly efficient
synthesis of fluoro-homoneplanocin A (4) from D-ribose,
along with the synthesis of homoneplanocin A (3).
We first synthesized the common intermediate 8, which
is utilized in both Routes A and B, as shown in Scheme 2.
D-Ribose was converted into known lactol 109 in three
steps. The conversion of D-ribose into 2,3-acetonide 12
under acidic conditions followed by treatment with vinyl-
magnesium bromide gave triol 13. The oxidative cleavage
of 13 with NaIO4 yielded lactol 10.9 The nucleophilic
10a
Scheme 1 illustrates the retrosynthetic analysis for de-
sired nucleoside 4. Fluoro-homoneplanocin A (4) could be
synthesized by condensing allylic alcohol 5 with a purine
base under the Mitsunobu conditions. It was envisaged
that fluoroallylic alcohol 5 might be synthesized by two
routes. In Route A, compound 5 is derived from ketone 6
using electrophilic fluorination8 and phenylseleneylation
followed by syn-elimination. The oxidation of allylic alco-
hol 7 followed by reduction could lead to ketone 6.
Compound 7 might be easily derived from the chemo-
selective hydroborationꢀoxidation of 8 followed by pro-
tection of the resulting hydroxyl group. The cyclopentenol
8 could be derived from 9 via ring-closing metathesis
(RCM). Compound 9 could be synthesized by reacting
lactol 10 with trimethylsilyl diazomethane (TMSCHN2) un-
der basic conditions. The lactol 10 could be easily synthesized
from D-ribose using a known procedure.9 The other route for
the synthesis of fluoroallylic alcohol 5 involves the regiose-
lective opening of epoxide 11 with fluoride followed by the
addition of 10 with TMSCHN2 in the presence of n-
BuLi and N,N-diisopropylamine (DIPA) afforded enyne 9
(65%) and its TMS-protected derivative 14 (16%). The
TMS group in compound 14 was removed with tetra-n-
butylammonium fluoride (TBAF) to give 9. It is worth
noting here that the use of TMSCHN2/LDA provided
the desired derivative 9 without any epimerization at C-5,
which is usually observed when using the Bestmannꢀ
Ohira reagent/condition.10b The structure of 9 was
confirmed by the coupling constant (JH4,H5 = 5.6 Hz)
between H-4 and H-5 and the NOE enhancement
(2.11%) that indicated the cis-orientation of these
two protons. The ring-closing metathesis (RCM)11
of enyne 9 with a first generation Grubbs catalyst
afforded cyclopentenol 8 (60%), which can serve as a
(10) (a) Miwa, K.; Aoyama, T.; Shioiri, T. Synlett 1994, 107–108.
(b) Pandey, G.; Bharadwaj, K. C.; Khan, M. I.; Shashidhara, K. S.;
Puranik, V. G. Org. Biomol. Chem. 2006, 6, 2587–2595.
(11) (a) Grubbs, R. H.; Miller, S. J.; Fu, G. C. Acc. Chem. Res. 1995,
28, 446–452. (b) Li, J.; Lee, D. Eur. J. Org. Chem. 2011, 4269–4287.
(c) Diver, S. T.; Giessert, A. J. Chem. Rev. 2004, 104 (3), 1317–1382.
(d) Mori, M.; Sakakibara, N.; Kinoshita, A. J. Org. Chem. 1998, 63,
6082–6083. (e) Ye, W.; He, M.; Schneller, S. W. Tetrahedron Lett. 2009,
(7) Yang, M.; Schneller, S. W.; Korba, B. J. Med. Chem. 2005, 48,
5043–5046.
(8) Kumamoto, H.; Kobayashi, M.; Kato, N.; Balzarini, J.; Tanaka,
H. Eur. J. Org. Chem. 2011, 2685–2691.
(9) Moon, H. R.; Choi, W. J.; Kim, H. O.; Jeong, L. S. Tetrahedron:
ꢀ
50, 7156–7158. (f) Lopez, J. C.; Plumet, J. Eur. J. Org. Chem. 2011, 1803–
Asymmetry 2002, 13, 1189–1193.
1825.
Org. Lett., Vol. 14, No. 8, 2012
2135

Scheme 2. Synthesis of the Key Intermediate 8 via Enyne Ring-
Closing Metathesis
Scheme 3. Synthetic Approach to the Glycosyl Donor 5 via
Route A
Recently, we have used the electrophilic fluorination on
vinyllithium (from iodovinyl) for the synthesis of fluoro-
neplanocin A.5 Using the same strategy, we tried iodina-
tion on 16, but it was found that conventional iodination
(I2/pyridine in CCl4) was slow. To solve this problem, we
tried various iodinating agents and solvents, but the de-
sired iodo compound was obtained in a very low yield
(10ꢀ15%) (see pp S8ꢀS9 of the Supporting Information).
After having trouble with well-known methods, we devel-
oped a new method for the synthesis of the fluorovinyl
compound via Route B, as depicted in Scheme 4.
The key intermediate 8 was protected with TBDPSCl to
give 19. The chemoselective hydroborationꢀoxidation
produced homoallylic alcohol 20 (78%). The treatment
of 20 with pivaloyl chloride followed by the removal of the
TBDPS group afforded cyclopentenol 22. The treatment of
allylic alcohol 22 with mCPBA at ꢀ40 °C to room tempera-
ture yielded R-epoxide 23a (71%) and β-epoxide 23b (8.6%)
because of the directing effect of the R-hydroxyl group.
With 23 in hand, we explored the epoxide ring opening
using tetra-n-butylammonium hydrogenfluoride (n-Bu4NH2F3)
in the presence of KHF2.14 The heating of R-epoxide 23a
with n-Bu4NH2F3 in DMF at 130 °C afforded the desired
fluorodiol 24a in a highly regio- and stereoselective man-
ner. The stereochemistry of the fluorine atom was con-
firmed by NOE experiments. Under modified Pfitznerꢀ
Moffatt conditions,15 fluorodiol 24a underwent elimination
simultaneously with oxidation to give the R,β-unsaturated
ketone 25 in good yield. The minor isomer 23b was also
converted into compound 25 under the same reaction
conditions. The Luche reduction of 25 yielded glycosyl
donor 5 which is ready for condensation with a purine base.
versatile intermediate for the synthesis of carbocyclic
nucleosides.
Several novel methods have been developed for the
synthesis of fluorovinyl compounds, including the Hornerꢀ
WadsworthꢀEmmons condensation of fluorophospho-
nates with carbonyls, Julia olefination, electrophilic fluor-
ination of vinyllithium or stannane, Peterson olefination,
fluoro-olefination via a Reformatsky-type reaction, and
the fluorodesilylation of vinylsilanes.12 Despite there being
various methods for the synthesis of fluorovinyl com-
pounds, a universal method to produce these compounds
remains elusive. Thus, for the synthesisofglycosyl donor 5,
Route A was used starting from the key intermediate 8, as
shown in Scheme 3. The chemoselective hydrobora-
tionꢀoxidation of 8 was achieved with 9-BBN and sodium
perborate to give desired diol 15. Selective TBDPS protec-
tion of 15 gave 7 (60%) with concomitant formation of
disilyl derivative 7a (20%).
The oxidation of the resulting allylic alcohol 7 with
TPAP/NMO13 afforded cyclopentenone16. The reduction
of 16 in the presence of Pd/C yielded cyclopentanone 6 as
the single stereoisomer. The treatment of 6 with LiHMDS
followed by trapping with TESCl gave a silylenol ether,
which was treated with Selectfluor to give fluoroketone
17.8 However, the phenylseleneylation of 17 under various
reaction conditions did not give the desired fluoroselenide
18, which could be smoothly converted into glycosyl donor
5 by syn-elimination and reduction.
(14) (a) Barbier, P.; Mohr, P.; Muller, M.; Masciadri, R. J. Org.
Chem. 1998, 63, 6984–6989. (b) Landini, D.; Penso, M. Tetrahedron Lett.
1990, 31, 7209–7212.
(15) Johnson, C. P.; Penning, T. D. J. Am. Chem. Soc. 1988, 110,
4726–4735.
(12) (a) Yanai, H.; Taguchi, T. Eur. J. Org. Chem. 2011, 5939–5954.
(b) Qiu, X.-L.; Xu, X.-H.; Qing, F. L. Tetrahedron 2010, 66, 789–843. (c)
Liu, P.; Sharon, A.; Chu, C. K. J. Fluor. Chem. 2008, 129, 743–766.
(13) Griffith, W. C.; Ley, S. V. Aldrichimica Acta 1990, 23, 13–19.
2136
Org. Lett., Vol. 14, No. 8, 2012

Scheme 4. Synthesis of Glycosyl Donor 5 via Route B
Scheme 5. Synthesis of Fluoro-homoneplanocin A (4)
Scheme 6. Synthesis of Homoneplanocin A (3)
The glycosyl donor 5 was condensed with 6-chloropur-
ine under the standard Mitsunobu conditions to give the
desired N-9 derivative 26 without the concomitant forma-
tion of the N-7 isomer (Scheme 5).
After removing the isopropylidene group under acidic
conditions, the N-9 isomer was confirmed by comparison of
the 1H and13C NMR spectra reported in the literature16 and
by the HMBC correlation between 10-H and C-4 of 6-chlor-
opurine. Treatment of 27 with ammonia in tert-butanol,
followed by the removal of the pivaloyl group with NaOMe,
yielded the final fluoro-homoneplanocin A (4).
In addition to the synthesis of fluoro-homoneplanocin A
(4), the synthesis of homoneplanocin A (3) was also easily
accomplished from compound 7 (Scheme 6). The glycosyl
donor 7 was condensed with 6-chloropurine under the
standard Mitsunobu conditions to give the desired N-9
derivative 28 as a major isomer along with minor amounts
of the N-7 derivative (9%). Careful removal of the protect-
ing groups of 28 with TFA produced the 6-chloropurine
derivative 29 which was treated with ammonia in tert-
butanol gave the final product, homoneplanoicn A (3).
In summary, we have accomplished the stereoselective
synthesis of fluoro-homoneplanocin A (4) along with the
synthesis of homoneplanocin A (3) from D-ribose as a
potential inhibitor of S-adenosylhomocysteine hydrolase.
For the synthesis of compounds 3 and 4, enyne ring-closing
metathesis (RCM), the regio- and stereoselective opening of
the epoxide with fluoride, and a simultaneous oxida-
tionꢀelimination reaction were employed as key steps.
The key intermediate 8 is expected to serve as an excellent
template for the synthesis of carbocyclic nucleosides such as
neplanocin A, neplanocin C, and aristeromycin. Analysis of
the generality, scope, and limitations of the methodology
using vinyl fluoride is underway in our laboratory, and the
results of these studies will be reported in due course.
Acknowledgment. This work was supported by a grant
from the WCU project (R31-2008-000-10010-0) of Na-
tional Research Foundation (NRF), Korea, and the Ewha
Global Top5 Grant 2011 of Ewha Womans University.
Supporting Information Available. Complete experi-
1
mental procedure and H NMR copies of all new com-
pounds and 19F and 13C NMR copies of 4 and 24a des-
cribed herein. This material is available free of charge via
the Internet at http://pubs.acs.org.
The authors declare no competing financial interest.
(16) Paquette, L. A.; Dong, S. J. Org. Chem. 2005, 70, 5655–5664.
Org. Lett., Vol. 14, No. 8, 2012
2137