New carbocyclic nucleosides: Synthesis of carbocyclic pseudoisocytidine and its analogs

Article abstract of DOI:10.1016/j.tetlet.2014.05.030

Cyclopentane-containing nucleoside analogs with a CC connection between the (heterocyclic) base and the carbocyclic scaffold are quite rare. Herein, we report the synthesis of previously unknown racemic carbocyclic pseudoisocytidine and its analogs, which

Full text of DOI:10.1016/j.tetlet.2014.05.030

Tetrahedron Letters 55 (2014) 3713–3716
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
New carbocyclic nucleosides: synthesis of carbocyclic
pseudoisocytidine and its analogs
a
a
a
b
Lukáš Maier , Ondrej Hylse , Marek Necas , Martin Trbušek , Mari Ytre-Arne c,d, Bjørn Dalhus c,d
,
ˇ
ˇ
Magnar Bjorås c,d, Kamil Paruch a,
⇑
a Department of Chemistry, Faculty of Science Masaryk University, Kamenice 5/A8, 62500 Brno, Czech Republic
b Central-European Institute of Technology, Molecular Medicine, Kamenice 753/5, 62500 Brno, Czech Republic
c Department for Microbiology, Oslo University Hospital, N-0424 Oslo, Norway
d Department for Medical Biochemistry, Oslo University Hospital, N-0424 Oslo, Norway
a r t i c l e i n f o
a b s t r a c t
Article history:
Cyclopentane-containing nucleoside analogs with a CAC connection between the (heterocyclic) base and
the carbocyclic scaffold are quite rare. Herein, we report the synthesis of previously unknown racemic
carbocyclic pseudoisocytidine and its analogs, which were prepared in 13 steps from commercially avail-
able materials. Pseudoisocytidine and its sulfur analog were moderately active against the mantle cell
lymphoma cell line, JVM-3. We also prepared a versatile cyclopentanone intermediate, which can be con-
verted into novel carbocyclic nucleosides via highly stereoselective addition of organometallic nucleo-
philes; the adduct with phenyllithium, the stereochemistry of which was unambiguously confirmed by
X-ray crystallography, inhibits glycosylase NEIL1 in a dose-dependent manner.
Received 6 February 2014
Revised 18 April 2014
Accepted 7 May 2014
Available online 20 May 2014
Keywords:
Nucleoside analogs
Pseudoisocytidine
Diastereoselective synthesis
Glycosylases
Ó 2014 Elsevier Ltd. All rights reserved.
NEIL1
Nucleoside analogs represent a diverse group of organic com-
pounds. Appropriate modifications of the nucleoside scaffold can
result in significantly altered biological activity.1 As natural
nucleosides contain the relatively labile aminal motif (structure
A in Fig. 1), significant effort has been invested in order to identify
more stable analogs.
C-nucleosides with a CAC connection between the (heterocyclic)
base and the carbocyclic scaffold. It is conceivable that, at least
in some cases, these compounds might be more robust versions
of nucleoside analogs B and C. Furthermore, the installation of cer-
tain substituents (e.g., R = OH) is meaningful only in this series, as
this would lead to chemically unstable ketals and aminals in the
other series. Interestingly, compounds with general structure D
(where R = H) are quite rare6 and we have been unable to find
any analogs of type D (containing R = OH) with nucleoside-like
substitution patterns.
One strategy consists of replacing the tetrahydrofuran ring with
an appropriate carbocyclic isostere. The resulting carbanucleosides
have been synthesized by diverse synthetic methods,2 and in many
cases, it has been observed that the tetrahydrofuran ring can be
replaced with cyclopentane (structure B in Fig. 1) without a signif-
icant loss of biological activity.3 Naturally occurring representa-
tives of this series include aristeromycin and its unsaturated
analog (À)-neplanocin A.4 Another strategy is based on attachment
of the heterocyclic base to the tetrahydrofuran via a CAC bond
linkage (structure C in Fig. 1). While the resulting C-nucleosides
are in general more stable than natural nucleosides, the synthesis
of even relatively simple systems (e.g., tiazofurin and its analogs)
in this series is often not trivial.5 In addition, some C-nucleosides
can still undergo ring-opening of the furan (see below). Structure
D in Figure 1 combines elements of structures B and C: carbocyclic
R5
R5
O
N
N
C
R3 R2
R3 R2
A
B
R5
R5
O
C
R
R3 R2
R3 R2
C
D
⇑
Corresponding author. Tel.: +420 549495477; fax: +420 549492443.
E-mail address: paruch@chemi.muni.cz (K. Paruch).
Figure 1. The common scaffold of natural nucleosides (A) and the structures of
their analogs B, C, and D.
http://dx.doi.org/10.1016/j.tetlet.2014.05.030
0040-4039/Ó 2014 Elsevier Ltd. All rights reserved.

3714
L. Maier et al. / Tetrahedron Letters 55 (2014) 3713–3716
H
N
H
a
N
R
HO
NH2
HO
COOCH3
MeOOC
R
O
N
R
N
7a: R = Br
6a: R = Br
6b: R = SO2Ph
O
O
OH
HO
HO
OH
7b: R = SO2Ph
2a: R = NH2
: R = OH
: R = SH
1
2b
2c
b
O
c
HO
HO
Scheme 1. Opening of pseudoisocytidine
carbocyclic analog 2a and related compounds 2b and 2c.
1 and the structures of its direct
O
COOCH3
COOCH3
R
R
One attractive biologically active candidate for the tetrahydro-
furan-cyclopentane replacement is pseudoisocytidine (1), which
has been shown to be active against cytarabine-resistant leuke-
mias,7 but hepatotoxic in vivo,8 which may be the result of opening
of the tetrahydrofuran core (Scheme 1).9
The direct carbocyclic analog 2a cannot undergo such a ring-
opening process and its toxicological profile might therefore be
superior to that of pseudoisocytidine, while its biological activity
could be retained (Scheme 1). Herein, we report the first synthesis
of the previously unknown carbocyclic pseudoisocytidine analog
2a and its derivatives 2b and 2c.
The overall retrosynthetic strategy is depicted in Scheme 2. It
includes the previously described substituted norbornene interme-
diate 5,10 which could yield ketoester 4 after oxidative cleavage.
The desired stereochemistry of 4 is dictated by the geometry of
the bicyclo[2.2.1]heptene scaffold produced in a Diels–Alder reac-
tion between cyclopentadiene and appropriately substituted die-
9a: R = Br
8a
: R = Br
9b: R = SO2Ph
8b: R = SO2Ph
d
O
O
O
e
COOCH3
OCH3
O
H
O
10
O
f
O
11
12
TBDPSO
OH
O
OH
O
HO
g
OCH3
OCH3
O
O
O
O
13
nophile 6, possessing
a leaving group that is utilized in a
subsequent elimination-diastereoselective cis-dihydroxylation
sequence. Compound 4 was envisioned as a precursor to the novel
aldehyde-ester 3, which upon reaction with urea, thiourea or gua-
nidine, and global deprotection would provide the target com-
pounds 2a–c.
Scheme 3. Reagents and conditions: (a) EtAlCl2, cyclopentadiene, CH2Cl2, 0 °C; 80%
for 7a, 70–95% for 7b. (b) OsO4, NMO, acetone/H2O (4:1), 40 °C; 80% for 8a, 90% for
8b. (c) Me2CH(OMe)2, cat. TsOH, acetone, rt; 99% for 9a and 9b. (d) DBU, Et2O, 0 °C
to rt; 95%, for 9a, DBU, CH3CN, 90 °C; 86% for 9b. (e) O3, CH2Cl2, À78 °C then Me2S
À78 °C to rt. (f) Li(Al-O-tBu)3H, THF, 0 °C to rt; 50–80% from 10. (g) TBDPSCl,
imidazole, CH2Cl2, rt; 70–92%.
The synthesis started from commercially available methyl pro-
piolate, which was converted into methyl (Z)-3-bromo-2-propeno-
ate (6a) (Scheme 3).11
Since we have found that the bromo compound 6a irritates the
skin (especially upon repeated exposure), we have utilized an alter-
native starting material for the Diels–Alder reaction, sulfone 6b.12
Thermal Diels–Alder reaction between cyclopentadiene and 6a
was very sluggish—the conversion after 24 h in refluxing benzene
or toluene was negligible and significant dimerization of cyclo-
pentadiene occurred. On the other hand, the reaction proceeded
smoothly at low temperature (0 °C), catalyzed by EtAlCl2, with very
good diastereoselectivity (9:1 endo/exo). Sulfone 6b underwent an
uncatalyzed Diels–Alder reaction quite efficiently (60% yield, rt,
14 h), although a higher conversion and yield were obtained in
the presence of the catalyst (EtAlCl2). The structure of the major
endo diastereomer 7b was confirmed by X-ray crystallography
(see Supporting information). Diastereoselective cis-dihydroxyla-
tion of both adducts 7a and 7b provided the corresponding diols
8a and 8b, which were subsequently protected as acetonides. Elim-
ination of the bromide or phenylsulfonyl group under basic condi-
tions afforded the key intermediate 10, which underwent
ozonolytic cleavage to produce the rather unstable aldehyde 11 that
was immediately used in the next step without purification. It
should be noted that, in principle, compound 10 could be prepared
more directly by dihydroxylation and protection of the Diels–Alder
adduct of methyl propiolate and cyclopentadiene, which we pre-
pared in 60–80% yield (cat. AlCl3, PhH, 0 °C). Attempted dihydroxy-
lations of the adduct, however, yielded complex hydroxylation
mixtures that contained the desired product together with the
unwanted diastereomer, the regioisomer with a dihydroxylated
double bond in the vicinity of the ester group as well as a tetrahydr-
oxylated product. In accordance with the published results,10 our
attempts to selectively reduce the aldehyde in the presence of an
H
H
N
O
HO
R
PG´O
N
OR
O
O
HO
2a
OH
: R = NH2
O
O
2b: R = OH
: R = SH
3
2c
O
PG´O
PGO
PGO
COOR
OR
O
5
PGO
OPG
a-ketoester were not successful. On the other hand, reduction with
4
excess Li(Al-O-t-Bu)3H yielded an inseparable mixture of epimeric
diols in which the primary hydroxyl group could be selectively sily-
lated to provide a-hydroxyester 13 (Scheme 3).
Oxidation of the hydroxyl group in 13 proved challenging. Many
standard methods (e.g., MnO2, PDC, KMnO4, Swern oxidation)
including RuO2 plus NaIO4 conditions that were previously used
ROOC
R
6
Scheme 2. Retrosynthetic analysis of pseudoisocytidine analogs (PG = protecting
group).

L. Maier et al. / Tetrahedron Letters 55 (2014) 3713–3716
3715
for a structurally similar intermediate,10 failed to give the desired
product 14 in acceptable yield. Fortunately, oxidation with Dess–
intermediates 17b and 17c as well as 16a by HPLC on a chiral sta-
tionary phase (see Supporting information).
Martin periodinane yielded pure
and purity (Scheme 4).
a
-ketoester 14 in very good yield
Clearly, the strategy described above can only be applied to con-
struct the (hetero)cyclic bases by elaboration of the ketoester 14. In
order to target compounds that are unaccessible by the methodol-
ogy described above, we envisioned a different and perhaps more
general route that utilizes a versatile cyclopentanone intermediate
(19, Scheme 5).
One-carbon homologation was accomplished via the Wittig
reaction with methoxymethylenetriphenylphosphorane. The reac-
tion produced enol ether 15 in 37–65% yield as a separable mixture
of Z and E isomers (Z:E ꢀ7:5). It should be noted that the Wittig
olefination was rather sensitive to the type and quality of base.
Generation of the phosphonium ylide with LDA gave the most con-
sistent and reproducible results, while reactions with LiHMDS,
KHMDS, or t-BuOK afforded olefination products in substantially
lower yields. Attempts to hydrolyze selectively enol ether 15 into
the desired aldehyde 3 in the presence of the acetonide and TBPDS
groups met with only limited success. With PPTS or acetic acid,
partial cleavage of the acetonide and/or TBDPS groups was
observed, while the enol ether moiety remained intact. We thus
attempted direct transformation of 15 into pyrimidine 16b by reac-
tion with urea. With sodium ethoxide in ethanol or NaH in THF, we
observed mainly cleavage of the TBDPS group and the desired
product 16b was formed in very low yield. However, using t-BuOK
in t-BuOH as the base, the desired product was formed in good
yield. Reactions of 15 with thiourea and guanidine under similar
conditions yielded compounds 16c and 16a, respectively. Selective
deprotection of the TBDPS group in pyrimidines 16a–c with TBAF
revealed the primary hydroxyl group, which could be utilized for
further selective derivatization, for example, the preparation of
phosphates and its isosteres. Final hydrolysis of the acetonide
under acidic conditions provided the target compounds 2a, 2b,
and 2c in good overall yields. The relative configurations of 2a–c
were confirmed by 2D NMR experiments (shown in Supporting
information). We were able to separate the enantiomers of
In order to access quickly and evaluate the potential of com-
pound 19, we converted one of the synthetic intermediates, ketoes-
ter 14, into the corresponding silyl enol ether 18. Subsequent
ozonolysis of this, rather unusual,13 substrate afforded 19 in an
acceptable yield (Scheme 5). The two-step sequence provided suffi-
cient amounts of material for preliminary studies. We initially stud-
ied the introduction of a phenyl group using PhMgBr or PhLi under a
variety of conditions (e.g., variable temperature and solvent, pres-
ence or absence of CeCl3) and found that the best results were
obtained when PhLi was added to a solution of 19 in THF at 0 °C.
Under these conditions, a single diastereomer of addition product
20, resulting from attack of the reagent from the less sterically hin-
dered side of 19, was obtained in 75% yield (Scheme 6). We were
unable to detect the other diastereomer by NMR spectroscopy.
The relative stereochemistry of the addition product 20 (sup-
ported by 2D NMR; see Supporting information) was unambigu-
ously assigned by X-ray crystallography of its p-bromobenzoyl
derivative 24 (Scheme 6 and Fig. 2).
We tested the effect of compounds 2a, 2b, and 2c on the viabil-
ity of leukemia cell lines that were available to us: SU-DHL-4 (dif-
fuse large B-cell lymphoma, del/mut TP53), JEKO-1 (mantle cell
lymphoma, del/mut TP53), and JVM-3 (mantle cell lymphoma,
wt-TP53). The viability of SU-DHL-4 and JEKO-1 was not affected
by 10
JVM-3 was more sensitive: 90% viability was observed upon treat-
ment with 100 M 2c; with 10 M and 100 M 2a we observed
90% and 83% viability, respectively.
lM or 100 lM concentrations of the compounds. However,
l
l
l
OH
O
O
TBDPSO
TBDPSO
Since the arrangement around the tertiary alcohol carbon of
compound 22 mimics that of the acetal product of glycosylase-med-
iated cleavage,14 this compound was tested against glycosylases
NEIL1, NEIL2, NTH1, and hOGG1, and was found to inhibit selec-
tively NEIL1 in a dose-dependent manner: at 1 mM we observed
46% inhibition, and at 0.5 mM and 0.125 mM concentrations 22%
and 2% inhibition, respectively.
OCH3
OCH3
a
O
O
O
O
O
13
14
b
In summary, we have completed the first syntheses of three new
racemic carbocyclic nucleoside analogs (2a–c) of pseudoisocyti-
dine, each in 13 steps. The synthetic approach builds on a user-
friendly preparation of sulfone 7b, which can be diastereoselective-
ly dihydroxylated and ultimately elaborated into tetrasubstituted
chiral cyclopentanes 11 with good diastereoselectivity. While
H
H
N
OCH3
OCH3
R
N
TBDPSO
TBDPSO
c
O
O
O
O
O
O
15
16a
: R = NH2
16b: R = OH
16c
: R = SH
O
OTBS
OCH3
TBDPSO
TBDPSO
OCH3
a
d
O
O
O
O
O
O
H
N
H
N
R
N
R
18
14
N
HO
e
HO
b
O
O
O
O
HO
OH
TBDPSO
2a: R = NH2
2b
17a: R = NH2
17b
O
: R = OH
2c: R = SH
: R = OH
17c: R = SH
O
O
Scheme 4. Reagents and conditions: (a) Dess–Martin periodinane, CH2Cl2, 0 °C to
rt; 80–90%. (b) Ph3P+CH2OMeClÀ, LDA, THF, 0 °C to rt; 37–65% (Z:E 7:5). (c)
guanidiniumÁHCl for compound 16a (20–35%), urea for 16b (30–45%) and thiourea
for 16c (40–50%), t-BuOK, t-BuOH, reflux. (d) TBAF, wet THF, rt; 90% for 17a, 96% for
17b, 90% for 17c. (e) HCl/H2O/MeOH 1:1:1, rt; 69% for 17a, 76% for 17b, 71% for 17c.
19
Scheme 5. Reagents and conditions: (a) LiHMDS, TBSOTf, THF, À78 °C. (b) O3,
CH2Cl2, À78 °C, then Me2S À78 °C to rt; 52% from 14.

3716
L. Maier et al. / Tetrahedron Letters 55 (2014) 3713–3716
TBDPSO
found to be moderately active against wt-TP53—mantle cell lym-
TBDPSO
Ph
O
phoma cell line JVM-3, which is generally the most sensitive to
chemotherapeutic treatment. The ability of compound 22 to inhibit
glycosylase NEIL1 has served as the starting point for a more thor-
ough exploration of the biological activity of this series of novel
carbocyclic nucleoside analogs. Further studies are currently in
progress and the results will be published elsewhere.
a
OH
O
O
O
O
b
19
20
HO
Ph
c
OH
Acknowledgments
TsO
HO OH
Ph
O
22
This project was funded from the SoMoPro programme
(SRGA771). Research leading to these results has received a finan-
cial contribution from the European Community within the Seventh
Framework Programme (FP/2007-2013) under Grant Agreement
No. 229603. The research was also co-financed by the South Mora-
vian Region. The authors would like to thank Dr. Jakub Švenda for
helpful comments and a review of the manuscript.
O
O
O
Br
24
Scheme 6. Reagents and conditions: (a) PhLi, THF, 0 °C; 75%; (b) TBAF, wet THF, rt;
then PPTS, MeOH rt; 42% from 20. (c) TBAF, wet THF, rt, then TsCl, Et3N, DMAP,
CH2Cl2, 0 °C; then NaH, 4-BrC6H4COCl, THF, 0 °C to rt, 42% from 20.
Supplementary data
Supplementary data (experimental procedures, spectral charac-
terization, and copies of 1H and 13C NMR data) associated with this
article can be found, in the online version, at http://dx.doi.org/
10.1016/j.tetlet.2014.05.030.
References and notes
1. Modified Nucleosides in Biochemistry, Biotechnology and Medicine; Herdewijn, P.,
Ed.; Wiley-VCH, 2008.
2. (a) Agrofoglio, L.; Suhas, E.; Farese, A.; Condom, R.; Challand, R. S.; Earl, R. A.;
Guedj, R. Tetrahedron 1994, 50, 10611; (b) Hildbrand, S.; Troxler, T.; Scheffold,
R. Helv. Chim. Acta 1994, 77, 1236; (c) Altmann, K. H.; Kesselring, R. Synlett
1994, 853; (d) Crimmins, M. T. Tetrahedron 1998, 54, 9229; (e) Zhu, X.-F.
Nucleosides, Nucleotides, Nucleic Acids 2000, 19, 651; (f) Agrofoglio, L. A. Curr.
Org. Chem. 2006, 10, 333.
3. (a) Borchardt, R. T.; Wu, Y. S.; Huber, J. A.; Wycpalek, A. F. J. Med. Chem. 1976,
19, 1104; (b) Secrist, J. A.; Clayton, S. J.; Montgomery, J. A. J. Med. Chem. 1984,
27, 534; (c) Neres, J.; Labello, N. P.; Somu, R. V.; Boshoff, H. I.; Wilson, D. J.;
Vannada, J.; Chen, L.; Barry, C. E.; Bennett, E. M.; Aldrich, C. A. J. Med. Chem.
2008, 51, 5349; (d) Choi, W. J.; Chung, H.; Chandra, G.; Alexander, V.; Zhao, L.
X.; Lee, H. W.; Nayak, A.; Majik, M. S.; Kim, H. O.; Kim, J.-H.; Lee, Y. B.; Ahn, C.
H.; Lee, S. K.; Jeong, L. S. J. Med. Chem. 2012, 55, 4521.
4. (a) Arita, M.; Adachi, K.; Ito, Y.; Sawai, H.; Ohno, M. J. Am. Chem. Soc. 1983, 105,
4049; (b) Arai, Y.; Hayashi, Y.; Yamamoto, M.; Takayema, H.; Koizumi, T. J.
Chem. Soc., Perkin Trans. 1 1988, 3133.
Figure 2. X-ray crystal structure of compound 24 (CCDC Ref. No. 937690).
ˇ
5. (a) Popsavin, M.; Spaic, S.; Svircev, M.; Kojic, V.; Bogdanovic, G.; Popsavin, V.
ˇ
´
Bioorg. Med. Chem. Lett. 2006, 16, 5317; (b) Štambasky´, J.; Hocek, M.; Kocovsky,
ˇ
ˇ
P. Chem. Rev. 2009, 109, 6729; (c) Bárta, J.; Slavetínská, L.; Klepetárová, B.;
Hocek, M. Eur. J. Org. Chem. 2010, 5432.
Diels–Alder reactions between cyclopentadiene with a beta-sulfo-
nyl enoate are known,15 the strategic elaboration of readily
accessed sulfones 7b into cyclopentanes seems largely undevel-
oped, and can be used for the construction of nucleoside analogs
6. (a) Playtis, A. J.; Fissekis, J. D. J. Org. Chem. 1975, 40, 2488; (b) Chu, C. K. I.;
Wempen, I.; Watanabe, K. A.; Fox, J. J. J. Org. Chem. 1976, 41, 2793; (c) Just, G.;
Kim, S. Tetrahedron Lett. 1976, 14, 1063; (d) Just, G.; Kim, S. Can. J. Chem. 1977,
55, 427; (e) Saksena, A. K.; Ganguly, A. K. Tetrahedron Lett. 1981, 22, 5227; (f)
Takahashi, T.; Kotsubo, H.; Koizumi, T. Tetrahedron: Asymmetry 1991, 2, 1035;
(g) Dishington, A. P.; Humber, D. C.; Stoodley, R. J. J. Chem. Soc., Perkin Trans. 1
1993, 57; (h) Díaz, M.; Ortuño, R. M. Tetrahedron: Asymmetry 1997, 8, 3421.
7. Burchenal, J. H.; Ciovacco, K.; Kalaher, K.; O’Toole, T.; Kiefner, R.; Dowling, M.
D.; Chu, C. K.; Watanabe, K. A.; Wempen, I.; Fox, J. J. Cancer Res. 1976, 36, 1520.
8. Woodcock, T. M.; Chou, T. C.; Tan, C. T. C.; Sternberg, S. S.; Philips, F. S.; Young,
C. W.; Burchenal, J. H. Cancer Res. 1980, 40, 4243.
9. Grierson, J. R.; Shields, A. F.; Zheng, M.; Kozawa, S. M.; Courter, J. H. Nucl. Med.
Biol. 1995, 22, 671.
10. (a) Just, G.; Reader, G. Tetrahedron Lett. 1973, 17, 1521; (b) Just, G.; Reader, G.;
Faure, B. C. Can. J. Chem. 1975, 54, 849.
11. (a) Ma, S.; Lu, X. Org. Synth. 1999, 9, 415; (b) Rossi, R.; Bellina, F.; Catanese, A.;
Mannina, L.; Valensin, D. Tetrahedron 2000, 56, 479.
12. Hirst, G. C.; Parsons, P. J. Org. Synth. 1990, 69, 169.
13. We have been unable to find any examples of the ozonolysis of such
tetrasubstituted alkenes.
14. (a) Slupphaug, G.; Mol, C. D.; Kavli, B.; Arvai, A. S.; Krokan, H. E.; Tainer, J. A.
Nature 1996, 384, 87; (b) Parikh, S. S.; Mol, C. D.; Slupphaug, G.; Bharati, S.;
Krokan, H. E.; Tainer, J. A. EMBO J. 1998, 17, 5214.
15. Downey, C. W.; Craciun, S.; Vivelo, C. A.; Neferu, A. M.; Mueller, C. J.; Corsi, S.
Tetrahedron Lett. 2012, 53, 5766.
16. Tanaka, M.; Yoshioka, M.; Sakai, K. Tetrahedron: Asymmetry 1993, 4, 981.
and for target directed syntheses. Analogs of a-ketoester intermedi-
ate 14 have been previously used for the construction of the hetero-
cyclic ring directly,6d or after a two carbon homologation;6f our one-
carbon homologation enables the synthesis of additional (het-
ero)cycles. Furthermore, we have performed preliminary studies
toward a potentially more versatile strategy for the preparation of
carbocyclic nucleoside analogs that utilizes highly diastereoselec-
tive additions of organometallic reagents onto cyclopentanone 19,
which itself is available in two-steps from a-ketoester intermediate
14. While analogs of 19 are known and have been used in synthesis
of carbocyclic nucleosides,16 tetraol 22, to our knowledge, is the
only carbocyclic C-nucleoside represented by generic structure D,
where R is an oxygenated substituent.
Unlike the analogs in the series A, B, and C (Fig. 1), compounds
such as 22 are stable and their biological evaluation should enable
mapping of part of the chemical space that is currently unaccessi-
ble. Along this line, we have tested the prepared carbocyclic ana-
logs in three leukemia cell lines, and compounds 2a and 2c were