Synthesis and?biological activity of?some?new 5′-O-acyl tiazofurin derivatives

Article abstract of DOI:10.1016/j.ejmech.2006.01.005

Three new 5′-O-acyl tiazofurin derivatives 2-4 were synthesized and evaluated for their antiproliferative activity against different tumour cell lines as well as for their ability to induce apoptosis in C6 cells in vitro. Apart of the antitumour assays, t

Full text of DOI:10.1016/j.ejmech.2006.01.005

European Journal of Medicinal Chemistry 41 (2006) 503–512
Original article
http://france.elsevier.com/direct/ejmech
Synthesis and biological activity
of some new 5′-O-acyl tiazofurin derivatives
Vjera Pejanović a,*, Vesna Piperski b, Dragana Uglješić-Kilibarda b, Jelena Tasić b,
Mirjana Dačević b, Ljubica Medić-Mijačević a, Esmir Gunić c, Mirjana Popsavin a,
Velimir Popsavin a
a Department of Chemistry, Faculty of Sciences, University of Novi Sad,Trg D. Obradovića 3, 21000 Novi Sad, Serbia & Montenegro
b Galenika AD Institute, Batajnički drum bb, 11000 Belgrade, Serbia & Montenegro
c Research and Development, Valeant Pharmaceuticals International, 3300 Hyland Avenue, Costa Mesa, California, USA
Received in revised form 16 January 2006; accepted 17 January 2006
Available online 07 March 2006
Abstract
Three new 5′-O-acyl tiazofurin derivatives 2–4 were synthesized and evaluated for their antiproliferative activity against different tumour cell
lines as well as for their ability to induce apoptosis in C6 cells in vitro. Apart of the antitumour assays, the cell membrane permeation of 2–4 and
their intracellular metabolism in C6 cells in vitro was also studied in order to evaluate their potential as possible tiazofurin bioisosteres or
prodrugs.
© 2006 Elsevier SAS. All rights reserved.
Keywords: Tiazofurin derivatives; Prodrugs; Anticancer agents; Apoptosis; Cytotoxicity
1. Introduction
its clinical use [1,8,9]. In order to provide an access to deriva-
tives of reduced toxicity, a number of tiazofurin analogues
have been synthesized bearing either a modified heterocyclic
aglycon [10–15] or a modified sugar moiety [16–21]. Our cur-
rent interest in this field is focused onto the 5′-O-acyl deriva-
tives 2, 3 and 4 (Fig. 1), designed as possible tiazofurin bioi-
sosteres, or as the tiazofurin prodrugs cleavable by cellular
esterases. A number of nucleoside prodrugs of improved selec-
tivity and/or better uptake by target cells have already been
described [22–26]. The less hydrophilic derivatives 2, and 3
were expected to show improved transport abilities across cell
membrane, while the tiazofurin/D-glucose conjugate 4, was in-
itially designed as a possible selective agent that would be able
to pass across tumour cell membrane through D-glucose trans-
porters. However, our preliminary studies suggested that 4
competed with free tiazofurin for the nucleoside transport sites
in C6 cells, whereby its uptake was significantly lower than
that observed for tiazofurin [27]. Continuing work in this field,
we report herein on the synthesis of 2–4 for evaluation of their
biological activity against some human and murine tumour cell
lines (Scheme 1).
Tiazofurin (1, Fig. 1) is a synthetic C-nucleoside, which
shows significant antitumour activity in a variety of tumour
systems [1]. In the phase II clinical trials, it induced dramatic
haematological responses in patients with acute myelogenous
leukaemia, or chronic myeloid leukaemia in blast crisis [2,3].
The biological activity of tiazofurin derives from a combina-
tion of cytotoxicity and maturation-inducing activities [4–6].
Both effects are attributed to inhibition of inosine 5′-monopho-
sphate dehydrogenase (IMPDH) by the tiazofurin adenine di-
nucleotide (TAD), which induces the shutdown of guanine nu-
cleotide synthesis [7]. This results in inhibition of proliferation,
induction of differentiation and apoptosis in different types of
malignant cells. Despite the remarkable efficacy of tiazofurin,
lack of specificity and significant toxicity remains a problem in
* Corresponding author. At present address: Hemofarm Group, Hemofarm
Institute, Beogradski put bb, 26300 Vršac, Serbia & Montenegro;
Tel.: +381-13-821-068; fax: +381-11-23-51-994.
E-mail address: vpejanovic@hemofarm.co.yu (V. Pejanović).
0223-5234/$ - see front matter © 2006 Elsevier SAS. All rights reserved.
doi:10.1016/j.ejmech.2006.01.005

504
V. Pejanović et al. / European Journal of Medicinal Chemistry 41 (2006) 503–512
Fig. 1. Tiazofurin (1) and the corresponding 5-O-acyl derivatives 2-4.
2. Results and discussion
allowed to react with 10 in DMSO, in the presence of sodium
hydride as a proton acceptor, whereupon the 3-O-(6′-tosyloxy)
hexyl derivative 11 was obtained in 66% yield. The intermedi-
ate 11 was treated with an excess of potassium carbonate in
aqueous DMF to give a moderate yield of the corresponding
primary alcohol 12. In this way, compound 12 was prepared in
an overall yield of only 24% from two synthetic steps. How-
ever, when the last two-step sequence was carried out without
purification of the intermediate 11, the desired product 12 was
obtained in a more acceptable overall yield (41% from 10).
Oxidation of the primary hydroxyl group in 12 gave the corre-
sponding carboxylic acid 13 (67%), a key intermediate in the
synthesis of target 4. With the requisite intermediate 13 in
hands, we next focused to its coupling reaction with protected
tiazofurin derivative 5. Attempted esterification of 13 with 5,
under the conditions already used for preparation of ribavirin/
D-glucose conjugate (DCC, DMAP, Py, 60 °C) [30] did not
give the expected product 14. The corresponding dicyclohexy-
lurea derivative 13a was a major product under these reaction
conditions. However, when the coupling reaction of 5 and 13
was carried out with ethyl chloroformate (Et3N, CH2Cl2), the
corresponding ester 14 was obtained in 36% yield. Final depro-
tection of 14 with aqueous trifluoroacetic acid furnished the
target molecule 4 (65%), as a 1:1 mixture of the corresponding
α- and β-anomers. NMR and high-resolution MS data con-
firmed the structure and purity of product 14.
2.1. Chemistry
Our first approach to 5′-O-acyl tiazofurin derivatives 2 and
3 involved a direct treatment of 1 with an appropriate acyl
chloride, under the conditions similar to those reported for es-
terification of ribavirin [28]. Although the preparation of cer-
tain 5′-O-acyl ribavirin derivatives was successfully achieved
by an action of 1.1 molar equivalent of the appropriate acyl
chloride in a mixture of pyridine and N,N-dimethylformamide,
this procedure failed to afford the expected 5′-O-acyl tiazofurin
derivatives 2 and 3. Under these reaction conditions a mixture
of the corresponding mono-, di-, and tri-O-acyl derivatives was
obtained. To overcome this problem, a simple three-step se-
quence for the preparation of targets 2 and 3 was elaborated.
In the first step, the secondary hydroxyl groups of 1 were first
protected by acetonation with 2,2′-dimethoxypropane in boil-
ing acetone, in the presence of p-toluenesulfonic acid as a cat-
alyst. The known [29] 2,3-O-isopropylidene derivative 5 was
thus obtained in 51% yield. Reaction of 5 with acetyl chloride
in pyridine afforded the expected 5′-O-acetyl derivative 6 in
86% yield, while treatment of 5 with benzoyl chloride in pyr-
idine gave the corresponding 5′-O-benzoyl derivative 7 in 96%
yield. Both reactions must be carried out at 0 °C, in order to
avoid transformations of the amide function observed at higher
temperatures. Hydrolytic removal of the isopropylidene protec-
tive group in 6 and 7 gave high yields of target compounds 2
(88%) and 3 (80%) ready for biological testing. The structure
and purity of 2 and 3 thus obtained was confirmed by the cor-
responding spectral and analytical data.
2.2. Cytotoxic activity
Compounds 2–4 were evaluated for their cytotoxic activity
against murine C6 and 9L glioma, human MCF7 breast adeno-
carcinoma, human HTB177 lung carcinoma, human K562 and
NB4 leukaemia cells, and against normal human dermal fibro-
blasts (NHDF). In vitro cytotoxicity was evaluated after 72-h
cell treatment by using the NRU assay. The results, including
the data for the reference compound 1, are presented in Table 1.
Remarkably, the 5′-O-acetyl derivative 2 has shown a potent
cytotoxic activity against C6, MCF7, K562 and NB4 cell lines,
comparable to that observed for tiazofurin. This compound also
Synthesis of tiazofurin/D-glucose conjugate 4 was achieved
by using an adopted procedure, recently applied for the pre-
paration of ribavirin and thymidine analogues of 4 [30]. Com-
mercially available 1,6-hexane diol (8) and 1,2:5,6-di-O-iso-
propylidene-α-D-glucofuranose (10) have been used as
convenient starting compounds in this part of the work
(Scheme 2). Treatment of 8 with tosyl chloride in pyridine
gave the expected ditosylate 9 (84%), which was further
Scheme 1. Reagents and conditions: (a) Me2C(OMe)2, TsOH, ME2CO, reflux, 1 h, 51%; (b) AcCl, Py, 0 °C, 20 h; 86%; (c) BzCl, Py, 0 °C, 20 h; 96%; (d) 9:1 TFA-
H2O, rt, 1 h, 88% of 2, 80% of 3.

V. Pejanović et al. / European Journal of Medicinal Chemistry 41 (2006) 503–512
505
Scheme 2. Reagents and conditions: (a) TsCl, Py, CH2Cl2, −15 °C →0 °C, 24 h 84%; (b) 9, NaH, DMSO, rt, 2 h, 66%; (c) K2CO3, H2O, DMF, 125 °C, 4 h, 40%; (d)
KMnO4, PDC, Me2CO, reflux, 6 h, 30%; (e) 5, EtOCOCl, Et3N, CH2Cl2, −7 °C, for 0.5 h, then rt for 19 h, 36%; (f) 9:1 TFA-H2O, rt, 1 h, 65%.
2.3. Detection of apoptosis
Table 1
In vitro cytotoxicity of compounds 1–4
As the induction of apoptosis in cancer cells is an important
outcome of anticancer therapy, we performed additional studies
to explore the ability of tiazofurin ester derivatives 2–4 to pro-
mote apoptotic cell death in murine C6 glioma. The effect of
tiazofurin 5′-O-ester derivatives 2–4 on the induction of apop-
tosis in C6 rat glioma cells was evaluated by microscopic ana-
lysis of May-Grunwald-Giemsa stained cells. Morphological
changes characteristic for apoptosis – loss of cell contacts, cell
rounding, nuclear condensation and fragmentation, as well as
phagocytosis of apoptotic by neighbouring cells [31], were
found after treatment with all tested compounds 1–4. Thus, in
concentration of 60 μM, 5′-O-acetyl derivative 2 was shown to
induce 23.8% apoptotic cells (Fig. 2), almost 5 times more than
5′-O-benzoyl derivative 3 (4.9% apoptotic cells). The conju-
gate 4 induced only 2.9% of apoptotic cells, while tiazofurin
itself caused 18.9% apoptotic cells after 72 hours of incubation.
Morphological evidence of apoptosis was confirmed with the
TUNEL assay that detected DNA internucleosomal fragments
in individual cells. The number of TUNEL positive cells was
proportionally higher than the number of morphologically
apoptotic cells (data not shown). Such results are expected,
considering the sensitivity of TUNEL method and detection
of earlier stages of apoptosis [32].
Compds
IC50, μMa
MCF7 HT-
C6
9L
K562
NB4
NHDF
B177
28.7
38.1
73.1
83.8
1
2
3
4
a
3.5
6.4
27.8
58.4
18.4
19.1
74.2
83.9
5.1
6.8
37.8
23.3
6.3
NDb
8.0
13.9
11.9
NDb
> 100
> 100
> 100
14.7
> 100
86.0
IC50 is a compound concentration required to inhibit the cell growth by
50% compared to untreated control.
b
ND – not determined.
exhibited essentially the same activity as tiazofurin against
murine 9L glioma cells. Compounds 3 and 4 exhibited a sig-
nificant antiproliferative activity only against leukaemia NB4.
The tiazofurin/D-glucose conjugate 4 was also active against
MCF7 cells, but it was more than 4-fold less active with re-
spect to the reference compound 1.
Generally, in all murine and human tumour cells, the 5′-O-
acetyl derivative 2 was more active than the 5′-O-benzoyl de-
rivative 3, which in turn was more active than the tiazofurin/D-
glucose conjugate 4, against both murine glioma and HTB177
cells. However, the conjugate 4 was slightly more active than
5′-O-benzoyl derivative 3 towards the MCF7, K562 and NB4
cell lines. Overall, these results indicate that only 5′-O-acetyl
derivative 2 brings together structural characteristics that
enable it to exert a potent antiproliferative activity towards all
investigated malignant cells, but in the same time, it is y inac-
tive against the normal human dermal fibroblast cells.
2.4. Evaluation of growth inhibition
As the C6 cells were the most sensitive to tiazofurin, this
cell line was used as a model for further biological testing of

506
V. Pejanović et al. / European Journal of Medicinal Chemistry 41 (2006) 503–512
Fig. 2. Morphological evidence of apoptosis induced by 5′-O-acetyl-derivative 2 in C6 cells. May-Grunwald-Giemsa staining. (a) Control untreated cells. (b) Cells
treated with 2 for 24 hours. Note change in cell shape – rounding and loss of cell-cell contacts. (c) and (d) Fragmentation and condensation of nuclear material seen
after 48 and 72 hours of incubation. (e) Phagocytosed apoptotic cell inside one normal looking cell.
1–4. The viability of C6 cells was first evaluated by the trypan-
blue exclusion assay. It has been found that the viability of C6
cells after their treatment with highest concentrations of 2–4
was higher than 90%, thus implying to their cytostatic, rather
than cytotoxic effects. The similar mode of action was recently
observed for tiazofurin itself [33]. The cytostatic effects were
further evaluated by the NRU test at 24, 48 and 72 h after
using the indicated concentrations of 1–4 (1–120 μM). After
24-hours treatment (Fig. 3a), both tiazofurin and the 5′-O-ben-
zoyl derivative 3 inhibited the cells growth in dose-dependent
manner in all tested concentrations, while the 5′-O-acetyl deri-
vative 2 showed a similar dose-dependent effect in the concen-
trations higher than 7.5 μM. Tiazofurin/D-glucose conjugate 4
showed a weak growth inhibition only in the concentration
range of 30–120 μM. Compound 4 was completely inactive
in the concentration of 15 μM, but it showed a growth stimu-
latory activity in the concentration range of 1–7.5 μM. The
most potent antiproliferative effect was induced by tiazofurin,
which in concentration of 3.75 μM decreased the cells growth
for 21%. The 5′-O-acetyl derivative 2 showed a similar activity
as tiazofurin in concentrations higher than 30 μM, while com-
pounds 3 and 4 exhibited similar antiproliferative effects at the
highest concentration tested. An increased dose-dependent
growth inhibition (in all concentrations) was observed when
C6 cells were treated with 1 and 2 for 48 h (Fig. 3b). Conver-
sely, compounds 3 and 4 showed a similar dose-dependent ef-
fects only in concentrations higher than 3.75 μM. Tiazofurin
again showed the most potent antiproliferative activity, where-
by at a concentration of 3.75 μM it induced an inhibition of
27%. In the same concentration, compound 2 was almost 3-
fold less active than tiazofurin, but in the concentration of
60 µM it showed almost the same antiproliferative effect as
tiazofurin itself. In the same concentration, both 3 and 4
showed significant antiproliferative effects, although almost
Fig. 3. Effects of tiazofurin (1) and ester derivatives 2-4 on proliferation of rat glioma C6 cells: (a) influence of concentration after 24 h, (b) influence of concentration
after 48 h, (c) influence of concentration after 72 h, (d) time-dependent effects in the concentration of 60 μM.

V. Pejanović et al. / European Journal of Medicinal Chemistry 41 (2006) 503–512
507
2-fold lower with respect to tiazofurin. The growth inhibition
was further increased after 72 h (Fig. 3c). At the concentration
of 3.75 μM tiazofurin and 5′-O-acetyl derivative 2 induced
growth inhibition of 60 and 42%, respectively. The 5′-O-ben-
zoyl derivative 3, at the same concentration, induced a growth
inhibition of 5%, while the tiazofurin/D-glucose conjugate 4
again showed a weak growth stimulatory activity. At the con-
centration of 60 µM tiazofurin and 5′-O-acetyl derivative 2 in-
duced the same growth inhibition of 62%, while both 3 and 4
induced growth inhibitions of 44 and 53%, respectively.
All tested compounds 1–4 inhibited the growth of C6 cells
in time dependent manner. The growth inhibition pattern of 1–
4 for the concentration of 60 µM is shown in Fig. 3d. The
activity of all tested compounds was gradually increased in
time, whereupon tiazofurin and the corresponding 5′-O-acetyl
derivative 2 showed the most potent and the same degree of
growth inhibition (cca. 30% after 24 h, 60% after 48 h, and
over 80% after 72 h). After 24 h however, the 5′-O-benzoyl
derivative 3 was almost 2-fold more active than the conjugate
4, although the induced inhibition was very low after this time
(9 and 5%, respectively). After treatment for 48 hours, com-
pound 3 induced for 10% lower growth inhibition with respect
to that induced by 4 (32 and 42%, respectively). Both 3 and 4
induced the most pronounced inhibition after 72-h treatment
(44 and 53%, respectively).
2.5. In vitro uptake and intracellular conversion studies
Our previous work revealed that the actual uptake of 4 into
rat C6 glioma cells was significantly lower than the uptake of
tiazofurin [27]. Therefore, only the ability of 2 and 3 to enter to
the murine C6 glioma cells was evaluated in the present work.
Tiazofurin was used as a reference compound. Hence, the cells
were incubated in the cell culture medium that contained
60 mM of tested compounds, and after 4 and 48 h of treatment
the cells were collected by tripsinisation and centrifugation,
homogenized and centrifuged again. The supernatant was ana-
lysed by the HPLC method recently developed in our labora-
tory [34]. The results are presented in Fig. 4.
The obtained results confirmed that the tiazofurin ester de-
rivatives 2 and 3 were able to penetrate through the membrane
of C6 cells, then underwent to subsequent intracellular hydro-
lysis to 1, which was finally converted to TAD. The same se-
quence of intracellular transformations had already been estab-
lished for the tiazofurin/D-glucose conjugate 4 [27]. When the
C6 cells were treated with 5′-O-acetyl derivative 2 for 4 h, the
cells contained essentially the same amounts of 1 and TAD
(28.3 and 28.8 nM/mg of cells, respectively) and a somewhat
lower amount (22.7 nM/mg) of the unchanged 5′-O-acetyl de-
rivative 2. The sum of intracellular concentrations of 2 and its
metabolites (1 and TAD) was for 39% higher than that ob-
served inside the cells treated with tiazofurin for 4 h. This re-
sult implies to a faster cellular uptake of 2 with respect to tia-
zofurin. The cell membrane permeation of 5′-O-benzoyl
derivative 3 is essentially the same as that observed for the
parent compound 1 after 4 h. The concentration of the intact
Fig. 4. Intracellular content of C6 cells after treatment with: (a) free tiazofurin 1,
(b) 2-O-acetyl derivative 2, (c) 5-O-benzoyl derivative 3.
5′-O-benzoyl derivative 3 in the cells was very low with re-
spect to the concentrations of tiazofurin and TAD, thus imply-
ing to a rapid intracellular conversion of 3 into the correspond-
ing metabolites (1 and TAD). Conversely, the intracellular
concentration of 5′-O-acetyl derivative 2 was very similar to
the concentrations of tiazofurin and TAD indicating that the
hydrolysis of its ester bond in the cells after 4 h might be rather
slow. After 48-h of cells treatment however, the C6 cells trea-
ted with tiazofurin contained essentially the same amounts of 1
(14.5 nM/mg) and TAD (14 nM/mg). The total intracellular
content was decreased for 50% compared to that established
after 4-h of cells treatment, presumably due to a tight binding
of TAD to the IMPDH. When the C6 cells were treated with
5′-O-acetyl derivative 2, the intracellular content of TAD
(18.1 nM/mg) was almost 2-fold higher than that of tiazofurin
(9.8 nM/mg). In the same time, this value is for 30% higher
than the concentration of TAD recorded inside the cells treated

508
V. Pejanović et al. / European Journal of Medicinal Chemistry 41 (2006) 503–512
Table 2
indicated to a faster cellular uptake of 2 with respect to tiazo-
furin, while the cell membrane permeation of 5′-O-benzoyl de-
rivative 3 is essentially the same as that observed for the parent
compound 1. The results of intracellular conversion studies of
2–4, pointed to the fact that these esters could certainly serve as
tiazofurin depot forms (at least in the C6 cells), but their poten-
tiality as tiazofurin prodrugs should be further evaluated
through additional in vitro and in vivo experiments. Namely,
the efficacy of the reported compounds as prodrugs in vivo is
strongly dependent on their stability towards plasma esterases at
least. Finally, on account of the lack of the free 5′-hydroxy
group in their ribose moiety, derivatives 2–4 should devoid of
the side effects of tiazofurin on the central nervous system.
Namely, it is believed that neurotoxicity of 1 originates from
its ability to interact with the central nervous system adenosine
receptors, whereby the presence of free 5′-hydroxyl group is
essential for the ligand-receptor binding [35].
Correlations between the intracellular content of active metabolites (1+TAD)
and the observed C6 cells growth inhibition
a
Compds
1+TAD
Estimated amount of
IMPDH-bound TAD
28.8
29.2
10.5
% of inhibition
a
4 h
48 h
1
2
3
a
57.3
57.1
53.9
28.5
27.9
43.4
59
56
32
In nM/mg of cells.
with tiazofurin for 24 hours. When the C6 cells were treated
with 5′-O-benzoyl derivative 3 the intracellular content of TAD
was essentially the same as that observed after 4 hours of cells
treatment (cca. 25 nM/mg). Moreover, the total amount of ac-
tive metabolites (1 and TAD) in the cells treated for 48 h with
ester derivative 3 was significantly higher with respect to that
observed inside the C6 cells treated with tiazofurin or 5′-O-
acetyl derivative 2. These data did not apparently correlate with
the observed cytotoxic patterns (Fig. 3), but enabled us to es-
timate the amount of IMPDH-bound TAD. Namely, it was as-
sumed that this amount is related to the difference of intracel-
lular concentrations of the active metabolites (1+TAD)
detected after 4 and 48 hours of cells treatment.
4. Experimental section
4.1. Chemistry
As shown in Table 2, the estimated amounts of IMPDH-
bound TAD after treatment with 1 and 2 are essentially the
same and significantly higher with respect to that observed
for 3. These data correlate well with the growth-inhibitory ac-
tivity of 1, 2, and 3 observed after the C6 cells treatment for
48 hours (Figs. 3b and d).
4.1.1. General methods
Melting points were determined using a Büchi SMP-20 appa-
ratus and were not corrected. Optical rotations were measured
on a Perkin Elmer 141 MC polarimeter. IR spectra were re-
corded on Perkin-Elmer 457 and Perkin-Elmer FT-IR 1725 X
spectrometers and the band positions are expressed in cm−1.
NMR spectra were recorded on a Bruker AC 250 E instrument
and chemical shifts are expressed in ppm values (δ-scale),
downfield from tetramethylsilane. Low-resolution mass spectra
were recorded on Finnigan-MAT 8230 (CI and EI) and VG
AutoSpec (FAB) mass spectrometers. High-resolution mass
spectra were taken on a Micromass LCT KA111 spectrometer.
HPLC analysis was performed on a Hewlett Packard 1100 liquid
chromatograph, equipped with quaternary pump and diode-array
detector. The compounds of interest were separated on a Supel-
cosil LC-18 RP column (250×4.6 mm, 5 mm) by gradient elu-
tion. The mobile phase was composed of 0.1 M sodium-hydro-
gen phosphate buffer pH 5.1, and methanol. Peak identification
was achieved by comparison of retention times and UV spectra
of the eluting components with known standards. TLC was per-
formed on DC Alufolien Kieselgel 60 F254 (E. Merck). Flash
column chromatography was performed using ICN silica 32–
63. Organic extracts were dried with anhydrous Na2SO4. Organ-
ic solutions were concentrated in a rotary evaporator under di-
minished pressure at a bath temperature below 30 °C. Tiazofurin
[36] and TAD [37] were prepared as previously reported. All
other chemicals were obtained commercially (guarantied reagent
grade) and used without further purification. The trypan-blue
exclusion assay was carried out as earlier reported [33].
3. Conclusion
In conclusion, we have synthesized three novel tiazofurin 5′-
O-acyl ester derivatives 2, 3 and 4, which have shown a potent
cytotoxic activity against some human leukaemia and solid tu-
mour cell lines, but did not exhibit any significant cytotoxicity
towards the normal human dermal fibroblast cells. All newly
synthesized compounds induced apoptosis in C6 rat glioma
cells in vitro, whereupon the 5′-O-acetyl derivative 2 was
shown to be the most effective (23.8% of apoptotic cells). The
5′-O-benzoyl derivative 3 and the conjugate 4 were signifi-
cantly less effective (4.9 and 2.9% of apoptotic cells, respec-
tively) than tiazofurin (18.9% of apoptotic cells). Compounds
2–4 inhibited the C6 cells growth in dose- and time-dependent
manner. After 72 h and at the concentration of 60 μM, both
tiazofurin and 5′-O-acetyl derivative 2 showed the same cyto-
toxicity pattern (62% growth inhibition), while compounds 3
and 4 induced growth inhibitions of 44 and 53%, respectively.
Accordingly, all three esters 2-4 might be considered as tiazo-
furin bioisosteres, where by the 5′-O-acetyl derivative 2 has
shown a similar cytotoxicity as the parent compound 1. It was
further confirmed that all tiazofurin esters 2–4 were intracellu-
larly metabolised to tiazofurin and its active metabolite TAD.
The intracellular concentration of these metabolites (1 and
TAD) was used to estimate the amount of IMPDH-bound
TAD. The observed cytotoxicity pattern of tested compounds
correlated well with these data. The sum of intracellular concen-
trations of 2 and its metabolites (1 and TAD) inside the C6 cells
4.1.2. 2-(2,3-O-isopropylidene-β-D-ribofuranosyl)thiazole-4-
carboxamide (5)
To a solution of tiazofurin (1, 20 g, 74.3 mmol) in dry acet-
one (400 mL), was added 2,2′-dimethoxypropane (30 mL) and

V. Pejanović et al. / European Journal of Medicinal Chemistry 41 (2006) 503–512
509
TsOH×H2O (0.8 g, 4.27 mmol). The mixture was refluxed for
1 h and then neutralized by stirring with Amberlyst IR-45 resin
at room temperature for 0.5 h. The mixture was filtered and the
resin washed with acetone. The combined organic solutions
were evaporated to appearance of first crystals. The suspension
was left at +4 °C over night to accomplish the crystallization.
Pure 5 (11.7 g, 51%) was obtained as colourless crystals, mp
119–121 °C, [α]25D = −28.2 (c, 0.4 in H2O); lit. [29] mp 119–
Hz, H-2′), 5.30 (d, 1H, H-1′), 6.29 and 7.20 (2×bs, 2H,
CONH2), 7.3–7.85 (m, 5H, Ph), 8.01 (s, 1H, H-5); 13C NMR
(CDCl3): δ 25.40 and 27.20 (Me2C), 64.60 (C-5′), 82.30 (C-3′),
83.58 (C-4′), 84.90 (C-1′), 85.75 (C-2′), 114.43 (Me2C),
124.77 (C-5), 149.67 (C-4), 128.35, 129.08, 129.35 and
133.30 (Ph), 162.84 (C-2), 165.93 (PhCO), 170.80 (CONH2);
CI MS: m/z 405 (M++H), 404 (M+).
1
120 °C, [α]25D = −32.0 (c, 0.5 in H2O), H NMR (CDCl3): δ
4.1.5. 2-(5-O-Acetyl-β-D-ribofuranosyl)thiazole-4-carboxamide
(2)
1.39 and 1.63 (2×s, 3H each, Me2C), 2.99 (dd, 1H, exchange-
able with D2O, J5′a,OH = 7.4, J5′b,OH = 4.9 Hz, OH), 3.73 (ddd,
J4′,5′a = 4.0, J5′a,5′b = 12.2 Hz, H-5′a), 3.89 (ddd, J4′,5′b = 3.3
Hz, H-5′b), 4.39 (m, 1H, J3′,4′ = 2.9 Hz, H-4′), 4.85 (dd, 1H,
J2′,3′ = 6.4 Hz, H-3′), 4.96 (dd, 1H, J1′,2′ = 3.8 Hz, H-2′), 5.25
(d, 1H, H-1′), 5.99 and 7.16 (2×bs, 2H, exchangeable with
D2O, CONH2), 8.14 (s, 1H, H-5); 13C NMR (CDCl3): δ
25.38 and 27.31 (Me2C), 63.02 (C-5′), 82.09 (C-3′), 84.27
(C-1′), 85.96 (C-2′), 86.38 (C-4′), 114.50 (Me2C), 124.83 (C-
5), 149.70 (C-4), 162.64 (C-2), 170.52 (CONH2); CI MS: m/z
301 (M++H), 300 (M+), 285 (M+–Me).
A solution of 6 (0.684 g, 2 mmol) in 9:1 TFA/H2O (10 mL)
was stirred for 1 h at room temperature. The solvent was eva-
porated and the residue was co-evaporated with 1:1
MeOH–CH2Cl2 (2×10 mL). The remaining crude 2 was crys-
tallised from 9:1 CH2Cl2–MeOH, to yield the pure 2 (0.53 g,
88%) as colourless crystals, m.p. 156–158 °C; [α]20 +9.8 (c,
D
2.96 in MeOH); IR (KBr): vmax 3484, 3462, 3366, 1726, 1651,
1
1246; H NMR (DMSO-d6): δ 2.04 (s, 3H, MeCO), 3.91 (t,
1H, J2′,3′ = 5.3, J3′,4′ = 5.2 Hz, H-3′), 4.03–4.17 (m, 3H, J1′,2′
= 3.9, J5′a,5′b = 13.9 Hz, H-2′, H-4′ and H-5′a), 4.35 (dd, 1H,
H-5′b), 5.00 (d, 1H, H-1′), 5.30 (bs, 2H, exchangeable with
D2O, 2×OH), 7.59 and 7.72, (2×bs, 2H, NH2), 8.22 (s, 1H,
H-5); 13C NMR (DMSO-d6): δ 20.77 (MeCO), 63.80 (C-5′),
71.19 (C-3′), 76.56 (C-2′), 80.91 (C-4′), 82.64 (C-1′), 124.51
(C-5), 150.40 (C-4), 162.35 (C-2), 170.27 (MeCO), 171.72
(CONH2); CI MS: m/z 303 (M++H); Anal. Calcd. for
C11H14N2O6S: C, 43.70; H, 4.63; N, 9.27; Found: C, 43.82;
H, 4.87; N, 9.12.
4.1.3. 2-(5-O-Acetyl-2,3-O-isopropylidene-β-D-ribofuranosyl)
thiazole-4-carboxamide (6)
To a stirred solution of 5 (3 g, 10 mmol) in dry pyridine (70
mL) cooled at 0 °C in an ice bath, freshly distilled acetyl chlor-
ide (11 mmol) was added dropwise. The reaction mixture was
stirred at 0 °C for 20 hours. After that, methanol was added
and the mixture evaporated to dryness. The residue was co-
evaporated twice with chloroform. Chromatographic separation
on a column of silica gel (20:1 CH2Cl2–MeOH) gave pure 6
(2.93 g, 86%) as colourless crystals, mp 104–107 °C (from
4.1.6. 2-(5-O-Benzoyl-β-D-ribofuranosyl)thiazole-4-
carboxamide (3)
CH2Cl2–MeOH), [α]20 −22.1 (c, 1.0 in CHCl3); IR (KBr):
By using the same procedure as described above (prepara-
tion of 2), pure 3 (0.58 g, 80%) was obtained, m.p. 136–138 °C
D
vmax 3462, 3344, 3120, 2988, 1744, 1682, 1590, 1381, 1236,
1108, 1078; 1H NMR (CDCl3): δ 1.39 and 1.61 (2×s, 3H each,
Me2C), 1.93 (s, 3H, MeCO), 4.13 (dd, 1H, J5′a,5′b = 12.0, J4′,5′a
= 4.6 Hz, H-5′a), 4.30 (dd, 1H, J4′,5′b = 3.9 Hz, H-5′b), 4.47
(m, 1H, J3′,4′ = 3 Hz, H-4′), 4.72 (dd, 1H, J2′,3′ = 6.4 Hz, H-
3′), 5.03 (dd, 1H, J1′,2′ = 3.5 Hz, H-2′), 5.25 (d, 1H, H-1′),
6.22 and 7.24 (2×bs, 2H, CONH2), 8.14 (s, 1H, H-5); 13C
NMR (CDCl3): δ 20.55 (MeCO), 25.38 and 27.19 (Me2C),
64.13 (C-5′), 82.24 (C-3′), 83.33 (C-4′), 84.63 (C-1′), 85.62
(C-2′), 114.54 (Me2C), 124.83 (C-5), 149.69 (C-4), 162.36
(C-2), 170.38 and 170.78 (2×C = O); CI MS: m/z 685 (2M+
+H), 343 (M++H).
(from CH2Cl2–MeOH); [α]20 +14.55 (c, 2.9 in MeOH); IR
D
1
(KBr): vmax 3478, 3356, 1715, 1682, 1277; H NMR (DMSO-
d6): δ 4.08 (m, 1H, J2′,3′ = 5.2, J3′,4′ = 5.1 Hz, H-3′), 4.25 (m,
2H, J1′,2′ = 4.0, J4′,5′a = 4.6, J4′,5′b = 2.4 Hz, H-2′ and H-4′),
4.41 (dd, 1H, J5′a,5′b = 12.1 Hz, H-5′a), 4.60 (dd, 1H, H-5′b),
5.05 (d, 1H, H-1′), 5.38 (d, 1H, exchangeable with D2O, J3′,OH
= 5.3 Hz, OH-3′), 5.59 (d, 1H, exchangeable with D2O, J2′,OH
= 5.6 Hz, OH-2′), 7.45–7.95 (m, 5H, Ph), 7.62 and 7.74 (2×bs,
2H, NH2), 8.18 (s, 1H, H-5); 13C NMR (DMSO-d6): δ 64.47
(C-5′); 71.29 (C-3′), 76.55 (C-2′), 81.19 (C-4′), 82.46 (C-1′),
124.45 (C-5), 128.82, 129.24, 129.49 and 133.52 (Ph),
150.41 (C-4), 162.38 (C-2), 165.63 (PhCO), 171.48
(CONH2); CI MS: m/z 364 (M+); Anal. Calcd. for
C16H16N2O6S×H2O: C, 50.26; H, 4.71; N, 7.32; Found: C,
50.03; H, 4.93; N, 7.31.
4.1.4. 2-(5-O-Benzoyl-2,3-O-isopropylidene-β-D-ribofuranosyl)
thiazole-4-carboxamide (7)
Following the same procedure as described above (prepara-
tion of 6), by using bezoyl chloride as an acylation agent, pure
7 was obtained (3.90 g, 96%) as colourless crystals, m.p. 114–
116 °C (from CH2Cl2–MeOH), [α]20D −28.1 (c, 1.0 in CHCl3);
IR (KBr): vmax 3464, 3333, 2990, 1723, 1682, 1586, 1382,
1271, 1093; 1H NMR (CDCl3): δ 1.41 and 1.63 (2×s, 3H each,
Me2C), 4.43 (dd, 1H, J5′a,5′b = 12, J4′,5′a = 4.2 Hz, H-5′a), 4.56
(dd, 1H, J4′,5′b = 3.5 Hz, H-5′b), 4.64 (m, 1H, J3′,4′ = 2.6 Hz, H-
4′), 4.86 (dd, 1H, J2′,3′ = 6.3 Hz, H-3′), 5.15 (dd, 1H, J1′,2′ = 3.4
4.1.7. 1,6-Ditosyloxy-hexane (9)
To a cooled (–15 °C) and stirred solution of dry 1,6-hexane
diol (8; 50 g, 0.42 mol) in a mixture of dry pyridine (107 mL)
and CH2Cl2 (180 mL) was added tosyl chloride (177.4 g, 0.93
mol) in small portions providing that temperature does not ex-
ceed 10 °C. The mixture was left for 24 h at 0 °C, then diluted
with CH2Cl2 (550 mL) and the solution was washed succes-

510
V. Pejanović et al. / European Journal of Medicinal Chemistry 41 (2006) 503–512
sively with 10% aq HCl (5×100 mL) and with 10% aq NaCl
(2×100 mL). The organic layer was dried and evaporated to a
colourless solid. Recrystallization from EtOH gave an analyti-
cal sample 9 (151.3 g, 84%), mp 69–70 °C; 1H NMR (CDCl3):
δ 1.20–1.74 (m, 8 H, 2×H-2, 2×H-3, 2×H-4 and 2×H-5), 2.46
(s, 6 H, 2×MeC6H4SO2), 3.98 (m, 4H, 2×H-1 and 2×H-6),
7.32–7.80 (m, 8 H, 2×MeC6H4SO2); 13C NMR (CDCl3): δ
21.44 (2×MeC6H4SO2), 24.56 (C-3 and C-4), 23.38 and
28.44 (C-2 and C-5), 70.17 and 70.27 (C-1 and C-6), 127.64,
129.72, 132.82 and 144.67 (2×MeC6H4SO2); CI MS: m/z 427
(M++H). HR MS (ES+): m/z 427.1249 (M++H). Calcd for
C20H27O6S2: 427.1249.
NaCl (3×60 mL), dried and evaporated. Chromatographic pur-
ification on a column of flash silica (3:2 hexane–EtOAc) gave
pure 12 (5.2 g, 40%) as a colourless oil.
Procedure B: A solution of 10 (50 g, 192.1 mmol) in
DMSO (150 mL), was allowed to react with a solution of 9
(300 g, 703.3 mmol) in DMSO (400 mL) in the presence of
NaH (80% in mineral oil, 11.5 g, 287.5 mmol) according to
procedure 4.1.7. The remaining crude 11 (118 g) was dried
under high vacuum overnight and dissolved in a mixture of
DMF (215 mL) and water (30 mL). To the solution was added
K2CO3 (13 g, 95 mmol) and the obtained suspension was
heated at 90 °C for 6 hours. A workup as described above,
followed by flash column chromatography (procedure A) gave
1
pure 12 (28.52 g, 41%) as a colourless oil. H NMR (CDCl3):
4.1.8. 1,2:5,6-di-O-isopropylidene-3-O-(6-tosyloxy-hexyl)-α-D-
glucofuranose (11)
δ 1.28, 1.31, 1.39 and 1.46 (4×s, 3H each, 2×Me2C), 1.35–1.53
(m, 8 H, 4×CH2), 1.96 (bs, 1H, exchangeable with D2O, OH),
3.41–3.63 (m, 2H, CH2O-3), 3.58 (m, 2H, CH2OH), 3.81 (d,
1H, J3,4 = 3.1 Hz, H-3), 3.94 (dd, 1H, J6a,6b = 8.5, J5,6a = 5.9
Hz, H-6a), 4.04 (dd, 1H, J5,6b = 6.1 Hz, H-6b), 4.08 (dd, 1H,
J4,5 = 7.0 Hz, H-4), 4.27 (m, 1H, H-5), 4.50 (d, 1H, J1,2 = 3.7
Hz, H-2), 5.84 (d, 1H, H-1); 13C NMR (CDCl3): δ 25.09,
25.94, 26.45 and 26.53 (2×Me2C), 25.26, 25.57, 29.37 and
32.29 (4×CH2), 62.10 (CH2OH), 66.83 (C-6), 70.20 (CH2O-
3), 72.23 (C-5), 80.83 (C-4), 81.78 (C-3), 82.17 (C-2),
104.94 (C-1), 108.58 and 111.40 (2×Me2C); FAB MS: m/z 383
(M++Na), 359 (M++H), 345 (M+–Me).
To a suspension of NaH (80% in mineral oil, 1.15 g, 28.75
mmol) in dry DMSO (10 mL) was added 10 (5 g, 19.2 mmol)
in small portions providing that temperature does not exceed
40 °C. After stirring the mixture at room temperature for
30 min a warm solution (35 °C) of 9 (28.4 g, 67 mmol) in
DMSO (43 mL) was added. The mixture was stirred at room
temperature for 2 h, and then partitioned between 10% aq NaCl
(60 mL) and EtOAc (60 mL). The organic layer was separated
and the aqueous phase was extracted with EtOAc (2×50 mL).
The extracts were combined, washed with 10% aq NaCl (2×50
mL) dried and evaporated. Flash column chromatography
(2:1 hexane–EtOAc) of the residue gave pure 11 (6.58 g,
1
4.1.10. 1,2:5,6-di-O-isopropylidene-3-O-(6-carboxy-hexyl)-α-
D-glucofuranose (13)
66%) as a colourless syrup. H NMR (CDCl3): δ 1.32, 1.33,
1.42 and 1.49 (4×s, 3H each, 2×Me2C), 1.24–1.73 (m, 8H,
4×CH2), 2.45 (s, 3H, MeC6H4SO2), 3.42–3.64 (m, 2H, CH2O
-3), 3.83 (d, 1H, J3,4 = 3.0 Hz, H-3), 3.93–4.07 (m, 4H,
CH2OTs and 2×H-6), 4.11 (dd, 1H, J4,5 = 7.6 Hz, H-4), 4.27
(ddd, 1H, J5,6a = 5.2, J5,6b = 4.3 Hz, H-5), 4.51 (d, 1H, J1,2
= 3.6 Hz, H-2), 5.86 (d, 1H, H-1), 7.35–7.78 (m, 4H,
MeC6H4SO2); 13C NMR (CDCl3): δ 21.58 (MeC6H4SO2),
25.38, 26.21, 26.74 and 26.80 (2×Me2C), 25.14, 25.43, 28.77
and 29.43 (4×CH2), 67.22 (C-6), 70.31 (CH2OTs), 70.44
(CH2O-3), 72.44 (C-5), 81.13 (C-4), 82.09 (C-3), 82.49 (C-
2), 105.22 (C-1), 108.86 and 111.70 (2×Me2C), 127.83,
129.77, 133.20 and 144.63 (MeC6H4SO2); CI MS: m/z 515
(M++H). HR MS (ES+): m/z 515.2318 (M++H). Calcd for
C25H39O9S: 515.2315.
A mixture of 12 (1 g, 2.8 mmol), pyridinium dichromate
(2.12 g, 5.6 mmol) and KMnO4 (0.89 g, 5.6 mmol) in dry
Me2CO (50 mL) was stirred at 50 °C for 8 h. The solution
was decanted and evaporated to a yellow oil. Chromatographic
purification on a column of flash silica (1:1 hexane–EtOAc)
1
gave pure 13 (0.7 g, 67%) as a pale yellow oil. H NMR
(DMSO-d6): δ 1.24, 1.25, 1.31 and 1.38 (4×s, 3H each,
2×Me2C), 1.30–1.56 (m,
6 H, 3×CH2), 2.20 (t, 2H,
CH2CO2H), 3.72 (m, 2H, CH2O-3), 3.81 (m, 2H, J3,4 = 3,
J6a,6b = 8.5, J5,6a = 6.2 Hz, H-3 and H-6a), 3.91 (dd, 1H, J5,6b
= 6.5 Hz, H-6b), 4.05 (dd, 1H, J4,5 = 6.6, J3,4 = 3 Hz, H-4),
4.21 (m, 1H, H-5), 4.58 (d, 1H, J1,2 = 3.7 Hz, H-2), 5.81 (d,
1H, H-1); 13C NMR (DMSO-d6): δ 24.25, 25.15, 28.90 and
33.67 (4×CH2), 25.22, 26.04, 26.57 and 26.66 (2×Me2C),
66.05 (C-6), 69.31 (CH2O-3), 72.24 (C-5), 80.40 (C-4), 81.57
(C-2 and C-3), 104.69 (C-1), 108.00 and 110.81 (2×Me2C),
174.51 (CO2H); FAB MS: m/z 375 (M++H), 359 (M+–Me).
4.1.9. 1,2:5,6-di-O-isopropylidene-3-O-(6-hydroxy-hexyl)-α-D-
glucofuranose (12)
Procedure A: A mixture of 11 (17.9 g, 34 mmol), K2CO3
(9.6 g, 70 mmol), water (5.5 mL) and DMF (54 mL) was stir-
red at 125 °C for 4 h. The mixture was cooled to room tem-
perature and divided in three portions. The first portion was
poured into EtOAc (300 mL) and washed once with 10% aq
NaCl (100 mL). Aqueous layer was collected and the second
portion was poured into the first organic layer, which was
again washed with 10% aq NaCl (100 mL). The procedure
was repeated with the third reaction portion. The combined
aqueous layers were extracted once with EtOAc (90 mL). Fi-
nally, the organic layers were combined, washed with 10% aq
4.1.11. 2-{5-O-[6-(1,2:5,6-di-O-isopropylidene-α-D-
glucofuranos-3-yl)-hexanoyl]-2,3-O-isopropylidene-β-D-
ribofuranosyl}thiazole-4-carboxamide (14)
To a stirred and cooled (–7 °C) solution of 13, (1.3 g,
3.5 mmol) in CH2Cl2 (25 mL) was added ethyl chloroformate
(1.34 mL, 14 mmol) and Et3N (2 mL, 14 mmol). The mixture
was stirred at –7 °C for 30 min and then compound 5 (0.5 g,
1.67 mmol) was added to the solution. The resulting mixture
was first stirred at –7 °C for 2 h, then at room temperature for

V. Pejanović et al. / European Journal of Medicinal Chemistry 41 (2006) 503–512
511
additional 19 h. The volatiles were removed by evaporation,
the remaining mixture was treated with CH2Cl2 (3×25 mL)
and evaporated. The oily residue was treated with EtOAc (8
mL), filtered and the solvent evaporated. Flash column chro-
matography (160:40:1 hexane–EtOAc–Et3N) of the residue
ferent concentrations (1–100 μM) of tested compounds 1-4.
The capacity of cells to concentrate Neutral Red dye (3-ami-
no-7-dimethylamino-2-methylphenazine hydrochloride) in their
lysosomes was measured at wavelength 540 nm using Micro-
plate cell reader (Biorad). All growth inhibition experiments
were done in triplicates. Concentrations inhibiting cell growth
by 50% (IC50) were calculated using FORECAST programme.
gave pure 14 (0.305 g, 36%) as a colourless oil, [α]20
=
D
−15.0 (c, 1.0 in CHCl3); 1H NMR (CDCl3): δ 1.30, 1.32,
1.38, 1.40, 1.47 and 1.60 (6×s, 3H each, 3×Me2C), 1.3–1.4
(m, 6 H, 3×CH2), 2.13 (m, CH2CO2), 3.57 (m, 2H, CH2O-3′)
3.82 (d, 1H, J3′’,4′’ = 3.0 Hz, H-3′’), 3.96 (dd, 1H, J5′’,6′’a = 5.8,
J6′’a,6′’b = 8.6 Hz, H-6′’a), 4.04–4.20 (m, 3H, J5′’,6′’b = 6.1,
J4′’,5′’ = 6.6, J5a’,5b’ = 12.0, J4′,5′a = 4.6 Hz, H-6′’b, H-4′’ and
H-5′a), 4.23–4.31 (m, 2H, J4′,5′b = 4.1 Hz, H-5′b and H-5′’),
4.47 (m 1H, J3′,4′ = 2.9 Hz, H-4′); 4.52 (d, 1H, J1′’,2′’ = 3.7
Hz, H-2′’); 4.70 (dd, 1H, J2′,3′ = 6.3 Hz, H-3′), 5.04 (dd, 1H,
J1′,2′ = 3.4 Hz, H-2′), 5.25 (d, 1H, H-1′), 5.87 (d, 1H, H-1′’),
6.19 and 7.19 (2×bs, 2H, CONH2), 8.12 (s, 1H, H-5); 13C
NMR (CDCl3): δ 24.35, 25.34 and 29.22 (3×CH2), 33.65 (
CH2CO2), 25.38, 25.48, 26.15, 26.71, 26.73 and 27.17 (3×
Me2C), 64.02 (C-5′), 67.13 (C-6′’), 70.09 (CH2O-3′’), 72.38
(C-5′’), 81.02 (C-4′’), 82.00 (C-3′’), 82.27 (C-3′), 82.36 (C-
2’), 83.36 (C-4′), 84.66 (C-1′), 85.58 (C-2′), 105.15 (C-1′’),
108.82, 111.63 and 114.45 (3×Me2C), 124.71 (C-5), 149.78
(C-4), 162.68 (C-2), 170.83 (CO2), 172.86 (CONH2); EI MS:
m/z 641 (M+–Me).
4.3. Morphologic signs of apoptosis
To visualise morphological changes in C6 cells indicative of
apoptosis induced by tiazofurin and the esters 2-4, cells grown
in chamber slides and treated with testing compounds in con-
centration of 60 μM for 72 hours, were fixed in 4% parafor-
maldehyde, post fixed in methanol, stained with May-Grun-
wald-Giemsa, according to the standard haematological
procedure, and examined under the light microscope (Olym-
pus). The criteria for apoptosis were the following: (1) decrease
in cell size and rounding-up; (2) chromatin condensation as
evidenced by the presence of dark, heavily stained nuclei; (3)
chromatin fragmentation as observed by the appearance of at
least three discrete masses of heavily stained nuclear material.
Cells were considered apoptotic when at least two of these cri-
teria were met. The number of morphologically apoptotic cells
was counted under the microscope and the percentage of apop-
totic cells was determined. At least 1000 cells were counted for
each compound tested.
4.1.12. 2-{5-O-[6-(α,β-D-glucopyranos-3-yl)-hexanoyl]-β-D-
ribofuranosyl}thiazole-4-carboxamide (4)
A solution of 14 (1.63 g, 2.5 mmol) in 90% aq TFA (8.1
mL) was stirred for 1 h at room temperature. The mixture was
evaporated by co-distillation with toluene. Flash column chro-
matography (4:1 CHCl3–MeOH) of the residue gave pure 4
(0.87 g, 65%) as a colourless syrup, [α]20D = +23.6 (c, 1.0 in
4.4. TUNEL in situ labelling of nuclear DNA fragments
To confirm morphological evidence of apoptosis, the term-
inal deoxynucleotidyl transpherase (TdT) mediated dUTP nick-
end labelling (TUNEL) assay was performed, which detected
DNA internucleosomal fragments in situ [32]. Cells under-
going apoptosis were detected with an in situ cell death detec-
tion kit (Boehringer-Mannheim). Procedure was followed as
recommended by the manufacturer. A negative control (cells
incubated with labelling solution without TdT) was included
in each experiment. Cells were considered positive if they
had distinct brown staining of nuclei, with or without apoptotic
morphology. The number of TUNEL-positive cells was
counted under the immersion objective and percentage of po-
sitive cells determined. At least 1000 cells were counted for
each compound tested.
1
MeOH); H NMR (DMSO-d6): δ 5.32, 5.56, 6.35 and 6.66
(4×bs, 4H, exchangeable with D2O, 4×OH), 7.58 and 7.72
(2×bs, exchangeable with D2O, NH2), 4.87 (d, 1H, J = 4.1
Hz, H-1′α), 4.91 (d, J = 5.2 Hz, H-1′’β), 4.98 (d, 1H, J = 3.4
Hz, H-1′), 8.17 and 8.20 (2×s, 1H, H-5); 13C NMR
(DMSO-d6): δ 25.13, 24.42 and 22.46 (3×CH2), 33.59
(CH2CO2), 81.96 (C-1′), 92.38 (C-1’α), 96.95 (C-1′’β),
124.32 (C-5), 150.36 (C-4), 162.40 (C-2), 171.70 (CO2),
172.90 (CONH2); FAB MS: m/z 559 (M++Na), 537 (M++H).
HR MS (ES+): m/z 559.1555 (M++Na). Calcd for
C21H32N2O12SNa: 559.1574.
4.2. Evaluation of growth inhibition
4.5. Determination of intracellular concentrations
The target cells were maintained into the appropriate culture
media as follows: murine C6 and 9L glioma cells in DMEM,
human MCF7 and HTB177 cells in EMEM, leukaemia K562
cells in RPMI-1640, leukaemia NB4 cell line in SSM and
NHDF cells in FGM. All media were supplemented with
10% FBS, except FGM that contained 2% of FBS. The cells
were grown in a humidified atmosphere with 5% CO2 at 37 °C.
Viable cell number was determined using NRU assay (Clo-
netics) in 96-well plates after 72 hours of treatment with dif-
To the C6 cells (cca. 150 mg) grown in an appropriate cul-
ture medium in triplicate was added 60 μM of tested com-
pound. After treatment for 4 or 48 hours, the cells were col-
lected by trypsinisation and centrifugation at 1600 rpm for
10 min. Cells were homogenised in 0.15 M KCl buffer,
pH 7.4 at +4 °C and centrifuged at 40000 rpm for 0.5 h in
the cold. The supernatant was heated at 100 °C for 1 min and
centrifuged at 15000 rpm for 10 min. The supernatant was fil-

512
V. Pejanović et al. / European Journal of Medicinal Chemistry 41 (2006) 503–512
De Montis, A.G. Loi, P. La Colla, M. Grifantini, J. Med. Chem. 40
(1997) 1731–1737.
[15] P. Franchetti, L. Cappellacci, M. Grifantini, A. Barzi, G. Nocentini, H.
Yang, A. O’Conor, H.N. Jayaram, C. Corell, B.M. Goldstein, J. Med.
Chem. 38 (1995) 3829–3837.
tered through a Sartorius filter (0.2 m) and the solution was
subjected to HPLC analysis as described in general methods.
Acknowledgments
[16] V. Nair, T. Wenzel, ARKIVOC 14 (2004) 128–132.
[17] M. Popsavin, Lj. Torović, V. Kojić, G. Bogdanović, V, Popsavin, Tetra-
hedron Lett. 45 (2004) 7125–7128.
[18] C.W. Liang, M.J. Kim, L.S. Jeong, M.W. Chun, Nucleosides Nucleotides
Nucleic Acids 22 (2003) 2039–2048.
[19] M. Popsavin, Lj. Torović, V. Kojić, G. Bogdanović, S. Spaić, V. Popsa-
vin, Bioorg. Med. Chem. Lett. 13 (2003) 3167–3170.
[20] L. Cappellacci, G. Barboni, P. Franchetti, C. Martini, H.N. Jayaram, M.
Grifantini, Nucleosides Nucleotides Nucleic Acids 22 (2003) 869–872.
[21] H.Y. Zhang, H.W. Yu, L.T. Ma, J.M. Min, L.H. Zhang, Tetrahedron
Asymmetry 9 (1998) 141–149.
[22] F. Rivault, I. Tranoy-Opalinski, J.-P. Gesson, Bioorg. Med. Chem. 12
(2004) 675–682.
[23] T.B. Cai, X. Tang, J. Nagorski, P.G. Brauschweiger, P.G. Wang, Bioorg.
Med. Chem. 11 (2003) 4971–4975.
[24] M. Nomura, S. Shuto, A. Matsuda, Bioorg. Med. Chem. 11 (2003)
2453–2461.
[25] K. Breistol, J. Balzarini, M.L. Sandvold, F. Myhren, M. Martinsen, E. De
Clercq, O. Fodstad, Cancer Res. 59 (1999) 2944–2949.
[26] C. Meier, Synlett (1998) 233–242.
[27] M.P. Dačević, J.S. Tasić, V.M. Pejanović, M.B. Segal, D.D. Uglješić-Ki-
libarda, A.I. Isaković, D.J. Begley, L.M. Rakić, Z.B. Redžić, J. Drug
Targ. 10 (2002) 633–636.
[28] N.B. Hanna, K.G. Upadhya, C.R. Petrie, R.K. Robins, G.R. Revankar,
Nucleosides Nucleotides 5 (1986) 343–362.
[29] M. Fuertes, M.T. Garcia-Lopez, G. Garcia-Munoz, M. Stud, J. Org.
Chem. 41 (1976) 4074–4077.
[30] D. Miljković, Z. Pietrzkowsky, E. Gunić, W. Seifert, US patent 5 (723)
(1998) 589.
[31] A.H. Willie, J.F.R. Kerr, A.R. Currie, Int. Rev. Cytol. 68 (1980) 251–
306.
[32] Y. Gavrieli, Y. Sherman, S.A. Ben-Sasson, J. Cell Biol. 119 (1992) 493–
501.
This work was supported by a research grant from the Min-
istry of Science and Environment Protection of the Republic of
Serbia (Grant No. 142005). The authors are grateful to Mr. D.
Djoković (Faculty of Chemistry, University of Belgrade,
S&M) and to Mrs. T. Marinko-Covell (Department of Chem-
istry, University of Leeds, UK) for recording the mass spectra.
References
[1] M. Grifantini, Curr. Opin. Invest. Drugs 1 (2000) 257–262.
[2] K. Malek, M. Boosalis, K. Waraska, B.S. Mitchell, D.G. Wright, Leuke-
mia Res. 28 (2004) 1125–1136.
[3] D.G. Wright, M. Boosalis, K. Malek, K. Waraska, Leukemia Res. 28
(2004) 1137–1143.
[4] D.G. Wright, M.S. Boosalis, K. Waraska, L.J. Oshry, L.R. Weintraub, E.
Vosburgh, Anticancer Res. 16 (1996) 3349–3351.
[5] G. Tricot, G. Weber, Anticancer Res. 16 (1996) 3341–3347.
[6] G. Weber, M. Nagai, Y. Natsumeda, J.N. Eble, H.N. Jayaram, E. Paulik,
W.N. Zhen, R. Hoffman, G. Tricot, Cancer Commun. 3 (1991) 61–66.
[7] K.W. Pankiewicz, S.E. Patterson, P.L. Black, H.N. Jayaram, D. Risal,
B.M. Goldstein, L.J. Stuyver, R.F. Schinazi, Curr. Med. Chem. 11
(2004) 887–900 (For a recent review see).
[8] V. Vranić-Mandušić, V. Subota, K. Savovski, L. Medić, T. Dramićanin,
O. Jozanov-Stankov, D. Popov-Čeleketić, M. Jokanović, B. Dimitrijević,
Toxicol. Lett. 146 (2004) 275–284.
[9] G. Tricot, H.N. Jayaram, G. Weber, R. Hoffman, Int. J. Cell Cloning 8
(1990) 161–170.
[10] J.-M. Navarre, D. Guianvarc’h, A. Farese-Di Giorgio, R. Condom, R,
Benhida, Tetrahedron Lett. 44 (2003) 2199–2202.
[11] P. Franchetti, S. Marchetti, L. Cappellacci, J.A. Yalowitz, H.N. Jayaram,
B.M. Goldstein, M. Grifantini, Bioorg. Med. Chem. Lett. 11 (2001) 67–
69.
[12] P. Franchetti, S. Marchetti, L. Cappellacci, H.N. Jayaram, J.A. Yalowitz,
B.M. Goldstein, J.-L. Barascut, D. Dukhan, J.-L. Imbach, M. Grifantini,
J. Med. Chem. 43 (2000) 1264–1270.
[13] P. Franchetti, S. Marchetti, L. Cappellacci, M. Grifantini, B.M. Gold-
stein, D. Dukhan, J.-L. Barascut, J.-L. Imbach, Nucleosides Nucleotides
18 (1999) 679–680.
[33] V. Piperski, M. Vračar, M. Jokanović, P. Stukalov, L. Rakić, Apoptosis
3 (1998) 345–352.
[34] J. Tasić, S. Ristić, V. Piperski, M. Dačević, V. Pejanović, M. Jokanović,
L. Mandić, J. Pharm. Biomed. Anal. 30 (2002) 993–999.
[35] P. Franchetti, L. Cappellacci, S. Marchetti, C. Martini, B. Costa, K. Var-
ani, P.A. Borea, M. Grifantini, Bioorg. Med. Chem 8 (2000) 2367–2373
(and references cited therein).
[36] K. Ramasamy, A. Lewis, R.K. Robins, N. Raju, N. Bharadway, V. Peja-
nović, US patent 5 (907) (1999) 36.
[14] P. Franchetti, L. Cappellacci, G. Abu Seikha, H.N. Jayaram, V.V. Guru-
dutt, T. Sint, B.P. Schneider, W.D. Jones, B.M. Goldstein, G. Perra, A.
[37] G. Gebeyehu, V.E. Marquez, J.A. Kelley, D.A. Cooney, H.N. Jayaram,
D.G. Johns, J. Med. Chem. 26 (1983) 922–925.