Synthesis and cytotoxic evaluation of novel pyrimidine deoxyapiothionucleosides

Article abstract of DOI:10.1016/j.bmcl.2013.03.091

The synthesis of novel pyrimidine deoxyapiothionucleosides of d- and l-series was realized following application of a versatile and high-yielding scheme, which utilized inexpensive l- and d-arabinose as starting materials, respectively, and which makes us

Full text of DOI:10.1016/j.bmcl.2013.03.091

Bioorganic & Medicinal Chemistry Letters 23 (2013) 3364–3367
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and cytotoxic evaluation of novel pyrimidine
deoxyapiothionucleosides
a
b
b
a,
⇑
´
´
Stefanos S. Kotoulas , Vesna V. Kojic , Gordana M. Bogdanovic , Alexandros E. Koumbis
a Laboratory of Organic Chemistry, Department of Chemistry, Aristotle University of Thessaloniki, Thessaloniki 541 24, Greece
b Experimental Oncology Department, Oncology Institute of Vojvodina, University of Novi Sad, Novi Sad 21000, Serbia
a r t i c l e i n f o
a b s t r a c t
Article history:
The synthesis of novel pyrimidine deoxyapiothionucleosides of
D- and
L-series was realized following
Received 2 February 2013
Revised 19 March 2013
Accepted 22 March 2013
Available online 1 April 2013
application of a versatile and high-yielding scheme, which utilized inexpensive
starting materials, respectively, and which makes use of a regio- and stereo-selective Pummerer rear-
rangement reaction for the coupling of the nucleobase with the thiosugar moiety. Some of the targeted
L
- and -arabinose as
D
compounds have shown selective cytotoxic effects (with IC50 <10
lM) against specific cancer cell lines.
All of the tested compounds had no cytotoxic effect on the normal cell line tested.
Keywords:
Thionucleosides
Apiose
Ó 2013 Elsevier Ltd. All rights reserved.
Pummerer rearrangement
Pyrimidine
Cytotoxic activity
Modified nucleosides represent a large and diverse class of com-
pounds which, due to their bioisosterism with the natural ana-
logues, constitute the main target pool for antiviral and
anticancer therapeutic agents.1–3 More specifically, thionucleo-
sides are considered to be one of the most biologically active sub-
groups of the modified nucleosides. After the first synthesis of
thionucleosides by Reist et al.,4 numerous synthetic attempts affor-
ded a variety of structurally diverse thionucleosides, often accom-
panied by data for their biological activity.2,5 The term
thionucleoside is used for both thionucleobase- and thiosugar-con-
taining nucleosides. Among the latter, some 40-thionucleosides
(e.g., 40-thioribonucleosides 1, Fig. 1), having the furanose oxygen
atom replaced by a sulfur atom, exhibit good to excellent biological
properties such as antibiotic, antitumor and antiviral activities.2,5,6
These properties are believed to result from the inhibition of viral,
or cellular, DNA, or RNA, polymerase after their conversion to the
corresponding 50-triphosphates. As a result, several elegant synthe-
ses of 40-thionucleosides have been reported,2,5 using mainly chiral
pool approaches.
of the Pummerer rearrangement in the synthesis of simple thionu-
cleosides was published by O’Neil and Hamilton10 in 1992. Since
then, this reaction has been frequently used to reach a plethora
of more complicated thionucleosides.2,5,9
D
-Apiose (2, Fig. 1), a branched-chain sugar found as residues in
galactopyrans-type pectins,11 resembles
D-ribose having the 4-
hydroxymethyl group migrated to C-3. Adenine apionucleosides
were originally prepared by Tronchet et al.12 Other apionucleosides
bearing natural or unnatural nucleobases were synthesized la-
ter,13,14 including 20-deoxy-derivatives.14,15 None of those com-
pounds was, however, found to exhibit significant antiviral or
anticancer activity. In contrast to apiose and apionucleosides, there
is only one known synthetic thioanalogue of
D
-apiose,16 and, to the
best of our knowledge, neither parent 30-deoxyapiose (3) nor thio-
nucleosides of
thus far.
D
-apiose and 30-deoxyapiose have been reported
Accordingly, thioapionucleosides seem to be synthetically and
biologically unexplored targets. Actually, two out of the three main
modifications applied to nucleosides, for example the removal of
50-hydroxyl group (which plays a key role in the elongation of
the nucleotide chain), and the alteration of the sugar moiety
(which changes the affinity with the building enzymes and conse-
quently causes their inactivation), are already present in thioapio-
nucleosides. In this context, the syntheses, as well as, the cytotoxic
The Pummerer rearrangement reaction, first reported by Pum-
merer7 in 1909, has been extensively studied over the past century
both mechanistically and as a synthetically useful process.8 It con-
tinues to serve as an indispensable tool for synthetic organic chem-
ists with recent applications in natural product synthesis,
nucleoside synthesis, and fluorous methods.9 The first application
evaluation of pyrimidine thio-deoxyapio-derivatives of both
-series, D-4/L-4 (Fig. 1), are both described herein.
For development of a synthesis for the targeted compounds, we
undertook the retrosynthesis analysis shown in Scheme 1. Since it
D- and
L
⇑
Corresponding author. Tel.: +30 2310 997839; fax: +30 2310 997679.
E-mail address: akoumbis@chem.auth.gr (A.E. Koumbis).
0960-894X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2013.03.091

S. S. Kotoulas et al. / Bioorg. Med. Chem. Lett. 23 (2013) 3364–3367
3365
S
O
S
O
B
OH
B'
HO
RO
RO
OR
X
HO
HO
O
HO
i
iii
HO
OH
HO
OH
X = OH
HO
OH
O
O
O
O
OH
2
1
D-4/L-4
OH
3 X = H
L-10/D-10
L-11/D-11 R = H
L-13/D-13 R = H
L-14 D-14
R = OMs
ii
iv
L-12 D-12
/
R = OTs
/
Figure 1. Structures of 40-thioribonucleosides (1), -apiose (2), 30-deoxyapiose (3)
D
and targeted novel thionucleosides D-4/L-4 (B = nucleobase, B0 = pyrimidine
v
nucleobase).
O
O
S
S
S
5
2
vi
O
+
S
S
S
O
O
O
O
O
O
B'
(PB')
deprotection
Pummerer
rearrang.
D-16 L-16
D-15 L-15
/
HO
OH
PO
HO
OP
D-5/L-5
PO
OP
/
D-17/L-17
D-4/L-4
D-6/L-6
Scheme 2. Synthesis of key-sulfoxides 16 and 17. Reagents and conditions: (a) for
D: Ref. 20 and for L: Refs. 20,21; (b) TsCl, pyridine, 25 °C, D 90%/L 87%; (c) LiAlH4,
THF, 80 °C, D 75%/L 80%; (d) MsCl, Et3N, CH2Cl2, 25 °C; (e) Na2SꢀxH2O, DMF, 60 °C, D
80%/L 78% (overall from 13); (f) m-CPBA, CH2Cl2, 0–25 °C, for D: 81% of 16 and 8% of
17/for L: 80% of 16 and 9% of 17.
oxidation
O
S
HO
GLG
hydride
thioether
formation
OH
reduction
substitution
PO
L-9 D-9
OP
PO
OP
PO
OP
isomers 16 and 17 in a ratio of about 10:1. 1H chemical shifts of
the three methyl groups for each isomer were assigned using
NOESY experiments. This helped us to identify that the major iso-
mers (16) were the ones bearing the sulfoxide O in trans position to
the fused dioxolane ring since different methyl groups are
deshielded by the sulfoxide bond22 in each case (4-CH3 for 16
L-8 D-8
D-7 L-7
/
/
/
Scheme 1. Retrosynthetic analysis (P = protecting group, (PB0) = protected or
unprotected pyrimidine nucleobase, GLG = good leaving group).
and
a-acetonide-CH3 for 17). Nevertheless, both isomers (16 and
17) gave independently the same results in the initial pilot Pum-
merer reaction experiments. For simplicity, we have conducted
all the coupling reactions with only the major ones (16).
was highly desirable to be able to easily prepare both enantiomers
of each thionucleoside, the use of a common plan was one of the
key-points of this analysis. According to the general plan, the final
nucleosides 4 might be obtained after the removal of sugar- and,
wherever needed, nucleobase-protecting groups. Fully masked
nucleosides 5 may be derived from the coupling of a thiosugar unit
with the appropriate nucleobase. This could be realized either
through a Vorbrüggen or a Pummerer rearrangement reaction.
After some consideration the second option was selected, with
the hope of achieving in conjunction with a suitable diol moiety
protecting group (P), better diastereoselectivities for this key reac-
tion. The required sulfoxides 6 could be reached upon oxidation of
the corresponding cyclic thioethers 7, which in turn are the prod-
ucts of consecutive double mesylation and thio-substitution of
diols 8.17 Finally, it was expected that the latter might be derived
from lactols 9 upon hydride reduction/substitution. An acetonide
of lactol D-9 (with GLG = TsO) was known to our group from pre-
vious syntheses.18,19 This knowledge gave us an indication that
the acetonide might be a suitably rigid and relatively bulky diol
With sulfoxides 16 in our hands, we then sought to investigate
the regio- and stereo-selectivity of the next transformation under
classical Pummerer reaction conditions with acetic anhydride, be-
fore utilizing the persilylated pyrimidines. This reaction could pro-
vide us with information about which of the two methylene groups
adjacent to the sulfur atom is preferred for the initial proton
abstraction and which face of the thiosugar ring is more accessible
with the respect to the subsequent nucleophilic attack. Taking into
account the steric effects exerted on these sulfoxides, it is obvious
that C-5 is less favorably approached by the base since both sides
are blocked (methyl and acetonide groups). Based on previous re-
sults,23 we may also assume that the two possible intermediate
a-
thiocarbocations do not isomerize. Moreover, as was mentioned
above, the acetonide group, due to its rigid and bulky nature, is ex-
pected to control the nucleophilic attack favoring the approach
from the b-face of the thiosugar ring.
protecting group, capable of blocking the
a-side of the reacting sul-
fonium intermediate in the Pummerer reaction. Moreover, it was
OAc
H
H
obvious that two inexpensive pentoses (
be used as starting materials.
L- and
D
-arabinose) could
S
S
AcO
AcO
Indeed, it is known that lactols 11 (Scheme 2) may be easily pre-
pared on a multigram scale from - and -arabinose (10) in practi-
O
O
O
O
O
O
O
O
O
S
L
D
cally two steps.20,21 Notably, the quaternary center is formed
during a stereospecific aldol reaction between the corresponding
erythrose acetonide and formaldehyde. Diols 11 were left to react
with TsCl18,19 to give mono-tosylates 12. Substitution of the tosyl
group occurred simultaneously with the lactol reduction upon
exposure18 to LiAlH4 to provide key-diols 13. Then, dimesylates
14 were smoothly formed and, although they were found to be
quite stable enough to be isolated, they were directly used in the
next step. Heating of 14 in the presence of Na2S gave cyclic
thiothers 15 in very good yields. Oxidation to the corresponding
sulfoxides took place when employing m-CPBA, yielding the two
5
2
3
18
19
4
S
S
O
O
OAc
D-16
20
21
Scheme 3. Pilot Pummerer reaction of sulfoxide D-16. Reagents and conditions: (a)
Ac2O, 110 °C, 66% of 18, 23% of 19, 3% of 20, 8% of 21.

3366
S. S. Kotoulas et al. / Bioorg. Med. Chem. Lett. 23 (2013) 3364–3367
The pilot Pummerer reaction of D-16 (Scheme 3) proceeded
O
O
O
S
S
quantitatively and resulted in a mixture of all possible isomers
(18–21). The major product (18) was obtained pure after chro-
matographic separation whereas the other three (19–21) were
practically inseparable and mixtures of them in different ratios
were collected. The relative stereochemistry of the less polar 18
(2b-isomer) was clearly deduced from its 1H NMR spectrum (iso-
lated AB system for H-5 and no coupling constant between H-2
and H-3) and was also confirmed using NOESY experiments (selec-
tive diagnostic correlations are given in Scheme 3). The less fa-
vored isomer, having a coupling constant between H-2 and H-3,
N
N
NH
HN
HN
i-iii
+
O
O
O
O
O
O
O
N
H
26
L-28 D-28
D-27/L-27
/
iv
O
S
N
NH
was assigned as 20 (2a-isomer). In contrast, the 5-regio-isomers
19 and 21 (both having H-2 AB system coupled to H-3) gave very
similar spectra and their structural assignment was not an easy
task. In NOESY spectra of both these isomers H-5 gives a correla-
tion peak with the adjacent C-4 methyl group but not with the
neighboring acetonide methyl group, as it was expected for the
O
HO
OH
D-29/L-29
Scheme 5. Pummerer reaction with uracil (26). Reagents and conditions: (a)
TMSOTf, Et3N, PhH, 25 °C; (b) D-16/L-16, CH2Cl2, 25 °C; (c) Et3N, BnH, 25 °C, from D-
16: 56% of D-27 and 14% of L-28/from L-16: 55% of L-27 and 15% of D-28; (d)
Amberlyst-15, CH2Cl2/MeOH (1:2), 40 °C, D-29 73%/L-29 73%.
5
a
-isomer. However, the ambiguous correlation peak between H-
5 and C-4 methyl is substantially stronger for isomer 21, indicating
that this is most probably the 5 -isomer.
a
Examining the ratio of the obtained isomers, it is notable that
the combined yield of the b-isomers (18 and 19) is clearly higher
than that of the a-isomers (20 and 21) and their ratio (89:11) is lar-
more difficult to separate, but once again only the b-isomers were
observed, with a slightly decreased ratio of the 2-/5-regioisomers
and slightly lower combined yields (70%) than was the case with
the thymine derivatives.
ger than the ratio of the combined yields of the 2-/5-regioisomers
(69:31). This observation leads to the conclusion that methyl group
on C-4 is less capable of directing the nucleophilic attack than the
acetonide group. These results were really encouraging for the pro-
grammed synthesis of the desired thionucleosides, since the acet-
oxy group as a nucleophile is significantly smaller compared to
the persilylated pyrimidines.
The Pummerer reaction with thymine (22) was performed first
and, to our delight, only two isomers D-23/L-23 and L-24/D-24
were obtained in each case (Scheme 4). Their separation was con-
siderably easier in comparison to that encountered with the prod-
ucts of the pilot reaction. As had been expected, the products were
both identified as the b-isomers using their 1D and 2D NMR spec-
tra. The combined yield (72%) was decreased due to the formation
of a mixture of the corresponding trimethylsilylthioglycosides.
However, the ratio of the regio-isomers was definitely larger in fa-
vor of the desired thionucleoside (2b-isomer), confirming the
hypothesis that the pyrimidine as a larger nucleophile would at-
tack C-1 faster.
Cytosine thionucleosides can be prepared using the Pummerer
reaction employing either the free base, or a protected derivative.
In our case, an attempt to perform the coupling reaction using
cytosine led to a very polar mixture of products, which was diffi-
cult to manipulate and to separate. Thus, commercially available
acetyl-cytosine (30) was used to uneventfully obtain the desired
products (Scheme 6). Again, only two isomers were found in the
reaction mixture (b-isomers D-31/L-31 and L-32/D-32). Compared
to the thymine and uracil reactions the combined yields (53–54%)
and the 2-/5-regioisomeric ratio were both lower.24,25
Having achieved the preparation and isolation of the protected
desired thionucleosides, the end of the synthesis was quite
straightforward. In the cases of thymine (23, Scheme 4) and uracil
(27, Scheme 5) the acetonide group was cleanly cleaved using an
acidic resin (Amberlyst-15) to obtain the targeted thionucleosides
25 and 29 (respectively) in good yields. Surprisingly, an attempt to
cleave the acetonide group from the cytosine derivatives 31 under
the same reactions conditions did not give any product. Thus,
stronger conditions were applied (TMSBr) to obtain thionucleo-
sides 33 (Scheme 6) in excellent yields. It is worth mentioning that
Moving from thymine to uracil (26), the mixtures of isomeric
products formed, D-27/L-27 and L-28/D-28 (Scheme 5), became
NHAc
AcHN
NHAc
S
S
O
O
O
N
N
S
S
N
N
N
N
N
i-iii
NH
HN
+
HN
i-iii
O
O
+
O
O
O
O
O
N
H
30
O
O
O
O
O
O
O
N
H
22
D-31 L-31
L-32/D-32
/
L-24 D-24
D-23/L-23
/
v (only for D-31)
iv
iv
NH2
NH2.HBr
O
S
S
N
N
N
N
O
S
N
NH
O
HO
OH
D-33/L-33
HO
OH
O
HO
OH
D-25/L-25
34
Scheme 6. Pummerer reaction with acetylcytosine (30). Reagents and conditions:
(a) TMSOTf, Et3N, PhH, 25 °C; (b) 16D/L, CH2Cl2, 25 °C; (c) Et3N, BnH, 25 °C, from
16D: 42% of 31D and 11% of 32L/from 16L: 44% of 31L and 11% of 32D; (d) TMSBr,
CH2Cl2, 0–25 °C then evaporation then MeOH, D 95%/L 94%; (e) TMSBr, CH2Cl2, 0–
25 °C then MeOH then evaporation, 89%.
Scheme 4. Pummerer reaction with thymine (22). Reagents and conditions: (a)
TMSOTf, Et3N, PhH, 25 °C; (b) D-16/L-16, CH2Cl2, 25 °C; (c) Et3N, BnH, 25 °C, from D-
16: 60% of D-23 and 12% of L-24/from L-16: 60% of L-23 and 11% of D-24; (d)
Amberlyst-15, CH2Cl2/MeOH (1:2), 40 °C, D-25 75%/L-25 76%.

S. S. Kotoulas et al. / Bioorg. Med. Chem. Lett. 23 (2013) 3364–3367
3367
with the TMSBr promoted deprotection, and, depending in the
work-up protocol used, it is possible to directly obtain the corre-
sponding hydrobromide salt, as is shown for D-31, which gave 34.
Cytotoxic activities of the synthesized thionucleosides 25, 29,
33 and 34 were evaluated after 48 h by MTT assay against five hu-
man cancer cell lines (MCF7, MDA-MB-231, A549, HeLa, and HT-
29) and one human normal cell line (MRC-5) (Table 1). Only
MDA-MB-231 cells were found to be sensitive to all seven tested
compounds, whereas another breast carcinoma cell line, MCF7,
toxic agents to fight cancer, the exploration of apiothionucleosides
may have a significant potential in the discovery and development
of novel drug candidates.
Acknowledgments
S.S.K. would like to thank the Greek State Scholarships Founda-
tion for a PhD scholarship. A.E.K. would like to thank Professor J.K.
Gallos for his inspirational comments.
was sensitive only to one of the tested compounds, cytosine L-thi-
onuleoside L-33. The same compound induced similar growth inhi-
bition in MDA-MB-231, HeLa, and MCF7 cells (IC50 values were
Supplementary data
11
D-25 was the most active of the tested compounds inducing a cyto-
toxic response of 57% and 62% (IC50 values of 3 M and 5 M in
MDA-MB-231 and HT29, respectively). The same compound was
17-fold and 20-fold times more active compared to its -enantio-
mer (L-25) in the MDA-MB-231 and HT29 cells, respectively. A
comparison between the - and -enantiomers reveals either sim-
ilar (e.g. compounds D-33 and L-33 on HeLa and MDA-MB-231)
or quite different (e.g. compounds D-33 and L-33 on A549 and
MCF7) activity against the same cell line. However, D-nucleosides
lM, 12 lM, and 13 lM, respectively. Thymine D-thionucleoside
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2013.
03.091.
l
l
L
References and notes
1. Yokoyama, M. In Modified Nucleosides in Biochemistry, Biotechnology and
Medicine; Herdewijn, P., Ed.; Wiley-VCH: Weinheim, Germany, 2008; pp
173–221. Chapter 8.
2. (a) Merino, P. Curr. Med. Chem. 2006, 13, 539; (b) Romeo, G.; Chiacchio, U.;
Corsaro, A.; Merino, P. Chem. Rev. 2010, 110, 3337.
D
L
3. (a) Seo, N.; Yamashiro, H.; Tadaki, T. Curr. Med. Chem. 2008, 15, 991; (b)
Moreno, S.; Aldeguer, J. L.; Arribas, J. R.; Domingo, P.; Iribarren, J. A.; Ribera, E.;
Rivero, A.; Pulido, F. J. Antimicrob. Chemother. 2010, 65, 827; (c) Huryn, D. M.;
Okabe, M. Chem. Rev. 1992, 92, 1745; (d) Burgess, K.; Cook, D. Chem. Rev. 2000,
100, 2047; (e) Wagner, C. R.; Iyer, V. V.; McIntee, E. J. Med. Res. Rev. 2000, 20,
417; (f) Tan, X.; Chu, C. K.; Boudinot, F. D. Adv. Drug Deliv. Rev. 1999, 39, 117; (g)
De Clercq, E. Rev. Med. Virol. 2000, 10, 255; (h) Jonckheere, H.; Anne, J.; De
Clercq, E. Med. Res. Rev. 2000, 20, 129; (i) Fischer, B. Drug Dev. Res. 2000, 50, 338.
4. Reist, E. J.; Gueffroy, D. E.; Goodman, L. J. Am. Chem. Soc. 1964, 86, 5658.
5. (a) Yokoyama, M. Synthesis 2000, 12, 1637; (b) Chambert, S.; Décout, J.-L. Org.
Prep. Proced. Int. 2002, 34, 27; (c) Gunaga, P.; Moon, H. R.; Choi, W. J.; Shin, D.
H.; Park, J. G.; Jeong, L. S. Curr. Med. Chem. 2004, 11, 2585.
6. Tsai, C.-H.; Lee, P.-Y.; Stollar, V.; Li, M.-L. Curr. Pharm. Des. 2006, 12, 1339.
7. (a) Pummerer, R. Ber. Dtsch. Chem. Ges. 1909, 42, 2282; (b) Pummerer, R. Ber.
Dtsch. Chem. Ges. 1910, 43, 1401.
8. (a) Bur, S. K.; Padwa, A. Chem. Rev. 2004, 104, 2401; (b) Feldman, K. S.
Tetrahedron 2006, 62, 5003.
were generally more active. Comparing the different nucleobases
it seems that cytosine thionucleosides have a broader cytotoxic
activity in contrast to the thymine analogues which are more selec-
tive and potent. These results indicate that, depending on the cell
line, either the nucleobase or the absolute stereochemistry of sugar
moiety are the key structural features for the observed activity. A
very important observation is that hydrobromide 34 is far less ac-
tive than its free cytosine thionucleoside D-33. This difference
could be explained by a reduced uptake of the salt by the cells
and thus introduces another key factor regarding the observed
cytotoxicity. Finally, it is interesting that none of the tested com-
pounds inhibited cell growth of normal fetal fibroblasts (MRC-5)
by 50% after 48 h.
9. (a) Akai, S.; Kita, Y. Top. Curr. Chem. 2007, 274, 35; (b) Smith, L. H.; Coote, S. C.;
Sneddon, H. F.; Procter, D. J. Angew. Chem., Int. Ed. 2010, 49, 5832.
10. O’ Neil, I. A.; Hamilton, K. M. Synlett 1992, 791.
In conclusion, novel pyrimidine deoxyapiothionucleosides of
both D- and L-series were synthesized from L- and D-arabinose,
11. Hudson, C. In Advances in Carbohydrate Chemistry; Pigm, W., Wolfro, M., Eds.;
Academic Press, 1949; Vol. 4, pp 57–74.
12. (a) Tronchet, J. M.; Tronchet, J. Helv. Chim. Acta 1970, 53, 853; (b) Tronchet, J.
M.; Tronchet, J. Helv. Chim. Acta 1971, 54, 1466; (c) Tronchet, J. M.; Tronchet, J.
Carbohydr. Res. 1974, 34, 2630; (d) Tronchet, J. M. J.; Tronchet, J.; Graf, R. J. Med.
Chem. 1974, 17, 1055.
13. (a) Parikh, D. K.; Watson, R. R. J. Med. Chem. 1978, 21, 706; (b) Taylor, M. D.;
Moos, W. H.; Hamilton, H. W.; Szotek, D. S.; Patt, W. C.; Badger, E. W.; Bristol, J.
A.; Bruns, R. F.; Heffner, T. G.; Mertz, T. E. J. Med. Chem. 1986, 29, 346; (c) Kim,
M. J.; Jeong, L. S.; Kim, J. H.; Shin, J. H.; Chung, S. Y.; Lee, S. K.; Chun, M. W.
Nucleosides Nucleotides Nucl. 2004, 23, 715.
respectively. The efficient 9-step synthetic scheme delivers the tar-
geted compounds in 13–14% overall yields. Some of the synthe-
sized compounds have shown selective cytotoxic effects (with
IC50 <10 lM) on specific cancer cell lines, whereas all of them
had no inhibitory effect on the normal cell line tested. The syn-
thetic strategy presented herein could be appropriately modified
to give the corresponding apiothionucleosides, which are expected
to show equal or even better cytotoxic activities. In view of the
continuous need to investigate and introduce new specific cyto-
14. Hammerschmidt, F.; Öhler, E.; Polsterer, J.-P.; Zbira, E.; Balzarini, J.; DeClercq, E.
Liebigs Ann. 1995, 551.
15. Kim, J.; Hong, J. H. Carbohydr. Res. 2003, 338, 705.
16. Halford, M. H.; Ball, D. H.; Long, L., Jr. J. Org. Chem. 1971, 36, 3714.
Table 1
17. Going from
7
to
8
the designation of
D
-/L
-series is reversed since
7 are
Cytotoxic activity evaluation of synthesized thionucleosides against human cancer
and normal cell lines
considered deoxyapiose analogues whereas
derivatives.
8
are considered erythritol
a
Compounds (base)
IC50
(l
M)
18. Koumbis, A. E.; Kaitaidis, A. D.; Kotoulas, S. S. Tetrahedron Lett. 2006, 47, 8479.
19. Koumbis, A. E.; Kotoulas, S. S.; Gallos, J. K. Tetrahedron 2007, 63, 2235.
20. (a) Thompson, D. K.; Hubert, C. N.; Wightman, R. H. Tetrahedron 1993, 49, 3827;
(b) Koumbis, A. E.; Dieti, K. M.; Vikentiou, M. G.; Gallos, J. K. Tetrahedron Lett.
2003, 44, 2513.
21. Gallos, J. K.; Stathakis, C. I.; Kotoulas, S. S.; Koumbis, A. E. J. Org. Chem. 2005, 70,
6884.
22. Abraham, R. J.; Byrne, J. J.; Griffiths, L. Magn. Reson. Chem. 2008, 46, 667.
23. Naka, T.; Minakawa, N.; Abe, H.; Kaga, D.; Matsuda, A. J. Am. Chem. Soc. 2000,
122, 7233.
24. The observed drop of regioselectivity could be rationalized considering that the
three nucleobases react with the initially formed cations with different rates
while competing attack by other nucleophiles occurs.
25. Although it was rather easy to obtain pure the major regioisomers (23, 27, 31),
the minor ones (24, 28, 32) were always contaminated (ca. 5%) by the former.
A549 HeLa MCF7 MDA-MB-
231
HT29 MRC-
5
D-25 (T)
L-25 (T)
D-29 (U)
L-29 (U)
D-33 (C)
L-33 (C)
34 (CꢀHBr)
41
30
>100 >100
>100 >100
3
52
19
14
10
11
58
5
>100
>100 >100
>100
>100 >100 >100
30
9
>100 12
>100 >100 >100
8
>100 >100
>100 >100
>100 >100
>100 >100
>100 >100
17
>100
13
a
Two independent experiments were set out with quadruplicate wells for each
concentration of the compound. SDs are given in the Supplementary data.