Structure based design, synthesis and in vitro antitumour activity of tiazofurin stereoisomers with nitrogen functions at the C-2′ or C-3′ positions

Article abstract of DOI:10.1016/j.ejmech.2019.111712

Three novel tiazofurin analogues having D-arabino stereochemistry and nitrogen functionalities at the C-2′ position (5–7) have been designed and synthesized in multistep sequences, starting from D-glucose. The known D-xylo stereoisomer of 1 (compound 2) a

Full text of DOI:10.1016/j.ejmech.2019.111712

European Journal of Medicinal Chemistry 183 (2019) 111712
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Research paper
Structure based design, synthesis and in vitro antitumour activity of
tiazofurin stereoisomers with nitrogen functions at the C-20 or C-30
positions
b
,
*
a
ꢀ a
ꢁ
ꢁ
ꢀ b
ꢁ
ꢀ b
Vesna Kojic , Mirjana Popsavin , Sasa Spaic , Dimitar Jakimov , Ivana Kovacevic ,
b
b, c
ꢁ
ꢀ a
ꢀ
ꢀ b
Milos Svircev , Lidija Aleksic , Bojana Sreco Zelenovic , Velimir Popsavin
a Oncology Institute of Vojvodina, Faculty of Medicine, University of Novi Sad, Put Dr Goldmana 4, 21204, Sremska Kamenica, Serbia
b
ꢀ
Department of Chemistry, Biochemistry and Environmental Protection, Faculty of Sciences, University of Novi Sad, Trg Dositeja Obradovica 3, Novi Sad,
Serbia
c Serbian Academy of Sciences and Arts, Knez Mihajlova 35, 11000, Belgrade, Serbia
a r t i c l e i n f o
a b s t r a c t
Article history:
Three novel tiazofurin analogues having
D
-arabino stereochemistry and nitrogen functionalities at the C-
-glucose.
20 position (5e7) have been designed and synthesized in multistep sequences, starting from
D
Received 13 August 2019
Received in revised form
5 September 2019
Accepted 16 September 2019
Available online 18 September 2019
The known D-xylo stereoisomer of 1 (compound 2) along with two new analogues bearing nitrogen
functions at the C-30 (3 and 4) has also been synthesized from the same sugar precursor. The synthetic
sequence consisted of the following three stages: (i) the multistep synthesis of suitably protected pen-
tofuranosyl cyanides, (ii) the construction of ethyl thiazole-4-carboxylate part by cyclocondensation of
thus obtained glycofuranosyl cyanides with L-cysteine ethyl ester followed by dehydrogenation, and (iii)
the final transformation of the ethyl thiazole-4-carboxylates into the target tiazofurin analogues using
the esters ammonolysis. The tiazofurin analogues were evaluated for their antitumour activities in cell-
culture-based assays. Compounds 3, 4 (D-xylo) and 7 (D-arabino), showed remarkable antitumour ac-
Keywords:
Tiazofurin analogues
De novo synthesis
Cytotoxicity
tivities, with IC50 values in the range of 4e7 nM. Preliminary structure-activity relationship allowed
identification of two analogues with antiproliferative activities exceeding that of the parent compound 1
for several orders of magnitude (e.g. 4: 1354-fold against Raji, 7: 309-fold against K562). Flow cytometry
Flow cytometry
Comet assay
Double fluorescent staining
data and Western blot analysis suggested that cytotoxic effects of D-xylo stereoisomers in the culture of
K562 cells caused changes in the cell cycle distribution, as well as the induction of apoptosis in caspase-
dependent way. The increase of apoptotic cells percentage in treated samples is also confirmed with
fluorescent double-staining method. Genotoxicity testing showed that the analogues with the xylo-
configuration (2e4) are far less genotoxic than tiazofurin.
© 2019 Elsevier Masson SAS. All rights reserved.
1. Introduction
Supplementary data), namely, thiazole-4-carboxamide adenine
dinucleotide (TAD) and selenazole-4-carboxamide adenine dinu-
C-Nucleosides represent a class of nucleoside analogues of sig-
nificant biological interest. Their chemistry and biomedicinal po-
tential has been reviewed [1e5]. C-Nucleoside tiazofurin (1, Fig. 1)
and its selenium analogue selenazofurin (1a) are potent anti-
tumour agents [6,7]. They are prodrugs, which are metabolically
activated by a two-step sequence comprised from an initial phos-
phorylation to the corresponding 50-monophosphates followed by
their subsequent conversion to analogues of NAD (Scheme S1 in the
cleotide (SAD). Both TAD and SAD potently inhibit inosine 50-
monophosphate dehydrogenase (IMPDH) that results in cancer cell
death [6e9]. Due to its severe toxicity, selenazofurin was abolished
from clinical testing, but tiazofurin proceeded to phase III clinical
trials in patients with chronic myelogenous leukemia [6]. This led to
the approval of 1 as an orphan drug for the treatment of patients
with chronic myelogenous leukemia in blast crisis [6]. However, the
lack of specificity and occasional toxicity found for tiazofurin dur-
ing clinical trials [10e12] prompted the search for structural ana-
logues, especially those with variations in the furanose ring. In this
* Corresponding author.
E-mail address: mirjana.popsavin@dh.uns.ac.rs (M. Popsavin).
context, analogues with D-xylo and D-arabino configurations have
https://doi.org/10.1016/j.ejmech.2019.111712
0223-5234/© 2019 Elsevier Masson SAS. All rights reserved.

ꢀ
2
V. Kojic et al. / European Journal of Medicinal Chemistry 183 (2019) 111712
Fig. 1. Structures of tiazofurin (1, X ¼ S), selenazofurin (1a, X ¼ Se), xylo-tiazofurin (2), and analogues 2ae7.
been prepared [13]. Their structural features [14] and antitumour
properties [15] have been studied, but no significant antitumour
activity was recorded in the cell assays. In the preliminary
communication [16] we have described a novel synthesis of D-xylo-
tiazofurin (2), as well as the re-investigation of its antiproliferative
activity against a panel of human malignant cells. A remarkable
cytotoxicity was recorded in the culture of several tumour cell lines,
including K562 cells. Additionally, a D-xylo-C-nucleoside containing
pyrrolo[2,1-f] [1,2,4]triazin-4-amine has been recently synthesized
and shown to exhibit a strong antiproliferative activity against a
panel of tumour cell lines [17].
Based upon these observations, we wanted to prepare new
xylo (2e4) and -arabino tiazofurin analogues (5e7) with nitrogen
functionalities at the C-20 or C-30 position (N3, NH2 and NHAc).
D-
D
Furthermore, some D-ribo-tiazofurin analogues with similar nitro-
gen functions at the C-20 position have shown interesting activities
in the cultures of certain tumour cells [18]. Finally, we were
Fig. 2. SAD in the active site of IMPDH (binding site of NADþ) with regions suitable for
the formation of H-bonds (purple: proton donor regions, green: proton acceptor re-
gions). (For interpretation of the references to colour in this figure legend, the reader is
referred to the Web version of this article.)
intrigued by the possibility of applying
a molecular design
approach using the available X-ray crystallographic data of human
enzyme IMPDH (type II) in complex with SAD (PDB code 1B3O) [19]
to produce novel IMPDH inhibitors.
Herein, we wish to report structure based design and full details
compound 8a to the NADþ binding site of human type II IMPDH
(1B3O). Conformationally rigid ligand 8a (see Fig. S2 in the Sup-
plementary data) and two neighbouring subunits of IMPDH with
five flexible amino acid residues in the active site (Asp274, Thr252,
Ser275, Ser276 and Met325) were used as input files for molecular
docking. Indeed, the docking analysis revealed the existence of an
additional hydrogen bond between the amino group at the C-30
position of the ligand and NH group of Ser276 from the peptide
backbone, which additionally stabilizes the potential inhibitor 8a in
the active site (for details see, the Supplementary data). These re-
sults encouraged us to synthesize analogues 2e7 and to evaluate
their antitumour properties.
of the synthesis of -xylo tiazofurin (2) [16], its isosteres 3 [20] and 4
D
[21], as well as the attempted synthesis of azido derivative 2a. We
also disclosed the synthesis of three novel tiazofurin analogues
having D-arabino stereochemistry and nitrogen functionalities at
the C-20 position (5e7). Their effects onto the proliferation of some
malignant cell lines will also be discussed in details.
2. Results and discussion
2.1. Chemistry
Our approach to design of novel inhibitors of IMPDH included an
initial inspection of the active site within the X-ray crystal structure
of the enzyme containing the bound inhibitor (SAD), which sug-
Our strategy for the synthesis of target C-nucleosides 2e7 in-
volves the preparation of suitably protected pentofuranosyl cya-
nides 9b and 15e19 (Table 1), followed by their cyclocondensation
0
0
gested that it may be possible to change of the C2 eC3 sugar portion
of the ligand whilst still retaining binding due to the presence of the
ligand/protein contacts within the remaining part of the molecule
(Fig. 2).
with L-cysteine ethyl ester, and by dehydrogenation [23] whereby
the protected C-nucleosides should be obtained. The final step of
the sequences will include ester ammonolysis of intermediary ethyl
thiazole-4-carboxylates into the corresponding tiazofurin ana-
logues 2e7. The preparation of initial intermediates 9e13 is shown
in Scheme 1.
Since this synthesis is described in detail in our preliminary
communication [20], it will not be further commented. However,
the corresponding experimental details are given in the Supple-
mentary data.
Preparation the desired pentofuranosyl cyanides 15e19 is
shown in Table 1. It is a three-step sequence that commenced with
the hydrolytic removal of acetal ring followed by subsequent
aldehyde oximination and the final oxime dehydration. In addition
to the required pentofuranosyl cyanides, this sequence provides a
variable quantity of furan 20, mainly formed as a side product
during the oxime dehydration reactions. Unfortunately, in the case
The active site was further analyzed using the Discovery Studio
Visualizer (DSV) [22] molecular design programme. This revealed
the presence of a proton donating region within the binding cavity
(Fig. 2, the central pink region) which was not being utilized for
binding of the ligand. Further analysis using DSV indicated that the
binding of the SAD analogues could be increased by placing the
0
0
proton acceptor groups above the C2 eC3 sugar region. This means
that -xylo and -arabino stereoisomers of tiazofurin, with proton
D
D
acceptor functions at the C-20 or C-30 positions can bind to the
enzyme by means of additional attractive interactions, which
would result in stronger antitumour effects. It was further assumed
that analogues 2e7 can be metabolically converted to the corre-
sponding SAD analogues 8a or 9a (Scheme S2 in the Supplementary
data), which can be bound to the active centre of the enzyme. To
check this assumption we performed molecular docking of
of D-gulo isomer 9 (entry 1) elimination product 20 was obtained as

ꢀ
V. Kojic et al. / European Journal of Medicinal Chemistry 183 (2019) 111712
3
Table 1
Preparation of 2,5-anhydro-D-aldononitriles 15e19.
Entry
1
Starting compound
Conditions
Products (yield, %)
9 R ¼ N3
(a) þ4 ꢀC, 24 h, rt, 24 h; (b) rt, 5 h;
(c) ꢁ18 ꢀC, 0.5 h, then rt, 2 h.
9b R ¼ N3 (0)
20 (30)
2
3
4
5
6
10 R ¼ OBz
11 R ¼ NHCOCF3
12 R ¼ NHAc
13 R ¼ N3
(a) þ4 ꢀC, 168 h; (b) rt, 2 h;
15 R ¼ OBz (59),
(c) ꢁ15 ꢀC, 45 min, then rt, 2 h.
20 (5)
(a) þ4 ꢀC, 70 h; (b) rt, 2 h;
16 R ¼ NHCOCF3 (59)
20 (4)
(c) ꢁ18 ꢀC, 144 h, then rt, 1 h.
(a) þ4 ꢀC, 168 h; (b) rt, 2 h, þ4 ꢀC, 70 h; (c) ꢁ15 ꢀC, 0.5 h, then rt, 2 h.
17 R ¼ NHAc (79),
20 (traces)
(a) þ4 ꢀC, 161 h; (b) rt, 6 h;
(c) ꢁ15 ꢀC, 45 min, then rt, 2 h.
(a) þ4 ꢀC, 140 h; (b) rt, 2 h;
(c) ꢁ15 ꢀC, 0.5 h, then rt, 2 h.
18 R ¼ N3 (41)
20 (6)
14 R ¼ NHAc
19 R ¼ NHAc (45)
20 (5)
Scheme 1. Reagents and conditions: (a) (i) BF3$OEt2, CH2Cl2, rt, 46 h; (ii) BzCl, Py, rt, 4 days, 36% of 10, 29% of 10a; (b) NaN3, DMSO, 108e112 ꢀC, 26 h; (c) BzCl, Py, rt, 24 h, 36% of 13
(from 8), 44.5% of 9 (from 8); (d) H2ePtO2, Ac2O, AcOH, rt, 21 h for 9, 57% of 12, 48 h for 13, 56% of 14; (e) (i) H2ePd/C, EtOH, rt, 24 h; (ii) TFAA, Py, CH2Cl2, ꢁ10 ꢀC for 0.5 h, then þ4 ꢀ
C
for 72 h, 62%.
the only reaction product. No desired nitrile 9b could be detected in
the reaction mixture. Due to the unfavourable outcome of this re-
action, we were forced to temporarily abandon the synthesis and
biological testing of isostere 2a. The remaining reactions (entries
2e6) took place successfully giving generally good yields of the
desired nitriles 15 (59%), 16 (59%), 17 (79%), 18 (41%) and 19 (45%).
conversion of nitriles 16 and 17 into the corresponding xylofur-
anosyl thioamides, followed by their subsequent cyclo-
condensation with ethyl bromopyruvate using a modified Hantzsch
thiazole synthesis protocol [24] (for details see the Supplementary
data). The desired intermediates 22 and 23 were obtained in similar
overall yields, independently of procedures used for their
preparation.
Having obtained desired 2,5-anhydro- -aldononitriles 15e19,
D
we next focused on their conversion to protected C-nucleosides
21e25 (Table 2) using the two-step sequence that includes cyclo-
The final transformation of the ethyl thiazole-4-carboxylates
21e25 into the corresponding thiazole-4-carboxamides 2e5 and
condensation of glycofuranosyl cyanides with
L-cysteine ethyl
7 are shown in Table 3. The ammonolysis of
D-xylo isomers 21 and
ester, followed by dehydrogenation of the resulting thiazolines
with BrCCl3 in the presence of DBU. This procedure provided
acceptable yields of protected C-nucleosides 21 (31%), 22 (39%), 23
(50%), 24 (56%) and 25 (58%). Arabinofuranosyl nitriles 18 and 19
gave somewhat higher yields of the corresponding C-nucleosides
24 and 25 (entries 4 and 5), when compared with xylofuranosyl
series (entries 1 and 2). Although generally lower, the yields of
xylofuranosyl derivatives 21 and 22 are acceptable.
23 (entries 1 and 3) and -arabino isomers 24 and 25 (entries 4 and
D
5) with saturated ammonia in MeOH selectively cleaved only ester
groups to afford good yields of target C-nucleosides 2 (89%), 4
(85%), 5 (85%) and 7 (58%). The acetamido functionalities in both
compounds 23 and 25 remained unchanged under these reaction
conditions (entries 3 and 5). However, the amide CeN bond in
trifluoroacetamido derivative 22 was easily cleaved under the
applied ammonolysis conditions, to afford the corresponding 3-
amino derivative 3 in 69% yield (entry 2).
The protected C-nucleosides 22 and 23 were alternatively pre-
pared using
a
two-step sequence comprised from previous
The isomeric 20-amino C-nucleoside 6 was prepared starting

ꢀ
V. Kojic et al. / European Journal of Medicinal Chemistry 183 (2019) 111712
4
Table 2
Conversion of 2,5-anhydro-D-aldononitriles 15e19 to protected C-nucleosides 21e25.
Entry
1
Starting compound
Conditions
Product (yield, %)
15 R ¼ OBz
(a) rt, 2 h;
21 R ¼ OBz (31)
22 R ¼ NHCOCF3 (39)
23 R ¼ NHAc (50)
24 R ¼ N3 (56)
(b) 0 ꢀC, 5 h, þ4 ꢀC, 66 h.
(a) rt, 3.5 h;
2
3
4
5
16 R ¼ NHCOCF3
17 R ¼ NHAc
18 R ¼ N3
(b) 0 ꢀC, 5 h, þ4 ꢀC, 24 h.
(a) rt, 4 h;
(b) 0 ꢀC 5 h, þ4 ꢀC, 24 h.
(a) rt, 4 h;
(b) 0 ꢀC, 5 h, þ4 ꢀC, 43 h.
(a) rt, 2 h;
19 R ¼ NHAc
25 R ¼ NHAc (58)
(b) þ4 ꢀC, 16 h.
Table 3
Conversion of protected C-nucleosides 21e25 to tiazofurin analogues 2e5 and 7.
Entry
Starting compound
conditions
Products (yield, %)
1
2
3
21 R ¼ OBz
rt, 6 days
rt, 8 days
rt, 8 days
2 R ¼ OH (89)
3 R ¼ NH2 (69)
4 R ¼ NHAc (85)
22 R ¼ NHCOCF3
23 R ¼ NHAc
4
5
24 R ¼ N3
rt, 4 days
rt, 5 days
5 R ¼ N3 (85)
25 R ¼ NHAc
7 R ¼ NHAc (58)
from 20-azido-20-deoxy C-nucleoside 5. Thus, catalytic hydrogena-
tion of 5 over 10% Pd/C, gave the desired amino derivative 6 in 82%
yield (Scheme 2).
and tiazofurin (1) were used as reference compounds. The results
are presented in Table 4.
D
-xylo-Tiazofurin (2) was more potent than DOX against two cell
lines (K562, 2-fold; HL-60, 7-fold) and more active than tiazofurin
2.2. Antiproliferative activities and SAR
Table 4
Cytotoxicities of synthesized analogues were recorded in the
cultures of six human tumour cell lines (K562, HL-60, Jurkat, Raji,
HT-29 and HeLa) and a single normal cell line (MRC-5) employing
MTT assay [25]. The cells were treated with test compounds for 72 h
[26]. The usage of MRC-5 cells serves to demonstrate the selectivity
of synthesized compounds toward tumour cells. Doxorubicin (DOX)
In vitro cytotoxicity of tiazofurin (1) and analogues 2e7 after 72 h.
Compounds
IC50
K562
1 (tiazofurin) 1.89
(
m
M)a
HL-60 Jurkat
Raji
HT-29 HeLa
MRC-5
0.36
0.19
0.04
5.28
0.26
>100
0.005
3.82
2
3
4
5
6
7
DOX
0.12b
0.13b
9.39
3.77
7.55
1.02
0.53b
3.11b
0.08b >100b
0.012 >100
>100 42.35
29.87 >100
0.035
0.34
6.28
1.97
0.0061 12.86
0.25 0.92
0.0036 3.85
0.0072 0.0039 0.26
>100
0.04
40.52
1.03
2.56
2.98
2.48
23.45
>100
0.15
1.03
>100 3.21
0.065 0.10
>100
>100
0.03
a
IC50 is the concentration of compound required to inhibit the cell growth by 50%
compared to an untreated control. Values are means of three independent experi-
ments. Coefficients of variation were less than 10%.
b
Scheme 2. Reagents and conditions: (a) H2ePd/C, EtOH, rt, 48 h, 82%.
Taken from Ref. [16].

ꢀ
V. Kojic et al. / European Journal of Medicinal Chemistry 183 (2019) 111712
5
(1) against four cell lines (K562, HL-60, Raji, HeLa; between 2 and
48 fold). Amino derivative 3 exhibited highly potent activities to-
ward K562, Jurkat, HT-29 and HeLa cells in low nanomolar range
being more potent than both DOX (5e30 fold) and 1 (11e318 fold).
It is worth pointing out that cytotoxicity of compound 3 towards
Jurkat cells was the highest recorded in this assay (IC50 3.6 nM).
Acetamido-derivative 4 is more potent inhibitor of growth of two
and three tumour cell lines compared to DOX and to tiazofurin,
respectively. It showed particularly potent activity in the culture of
T-cells leukemia (Jurkat, IC50 7.2 nM) exhibiting the 5- and 4-fold
higher potency than 1 and DOX, respectively. This compound was
remarkably potent toward Burkitt's lymphoma (Raji, IC50 3.9 nM)
being 1353- and 764-fold more active than lead 1 and DOX,
cells in the G2/M phase; after 72 h, this percentage is similar to
controls and 2e2.4 times greater than the values obtained for tia-
zofurin (1). Compounds 2 and 3 induce significant apoptosis of
K562 cells (subG1 fraction), as well as tiazofurin (1), after 72 h
(Fig. 3). The formation of the subG1 peak originates from a popu-
lation of cells with fragmented DNA molecules. This hypodiploid
DNA content is the basic characteristic of the cells that entered the
apoptosis, and the population of these cells is clearly distinguished
in the form of subG1 peak on an one-parameter histogram of the
cell cycle.
After exposure of the K562 cells to compounds of arabino-
configuration (5, 6 and 7) for 72 h, all the investigated compounds
and tiazofurin (1) increase the percentage of apoptotic cells (subG1
peak) with a reduced percentage of cells in G0/G1 phase relative to
control, while the percentage of cells in the S phase is almost the
same as after treatment with tiazofurin (1). The percentage of cells
in the G2/M phase for all investigated compounds, other than tia-
zofurin, is similar to control (Fig. 3).
respectively. -arabino-Derivative 5 showed the modest activity
D
toward K562, HL-60 and HT-29 cells. Compound 6 was the most
cytotoxic against Jurkat cells where it exhibited the similar activity
as DOX and lead 1. Acetamido derivative 7 exhibited a low nano-
molar activity towards myelogenous leukemia cells (K562, IC50
6.1 nM), with 309-fold higher potency than 1 and 40-fold higher
potency than DOX. Moreover, compound 7 was more active than
DOX and 1 in the culture of Burkitt's lymphoma cells.
2.4. Detection of apoptosis
Conclusively, it is important to highlight that in contrast to tia-
zofurin (1) and DOX, compounds 2, 3, 5 and 6 are not toxic against
normal foetal fibroblasts (MRC-5), while compound 4 showed
Percentage of apoptotic, necrotic and live cells was determined
with flow cytometry after staining of treated cells with Annexin V-
FITC/PI and the results are shown in Fig. S3 (see also Table S2 in the
Supplementary data).
modest activity against MRC-5 cells (IC50 42.35 mM). This could
imply their potential selectivity toward tumour cells, but it remains
to be further investigated toward wider panel of normal cells. Only
The apoptotic response, presented as a percentage of specific
apoptosis (Fig. 3B), shows that analogues 3, 6 and 7 after 72 h
multiply the percentage of Annexin-V positive K562 cells compared
to tiazofurin. Analogue 3 was the most active (34.37%), while 6 and
7 caused a 2-fold higher percentage of specific apoptosis than tia-
zofurin (1).
compound 7 is toxic against these cells (IC50 3.21
9-fold higher than 1 (IC50 0.36 M) and 32-fold higher than DOX
(IC50 0.1 M).
mM), being almost
m
m
The next step in our work was establishing correlations between
the structures of synthesized compounds and their antitumour
activity in an attempt to identify structural features beneficial for
potency. In order to investigate the influence of configuration on
activity, first we compared activity of tiazofurin (1) with xylo ana-
logues 2e4 (see the Supplementary data for details, Fig. 1Sa). The
results showed that compounds with xylo-configuration 2e4 were
more active than 1 toward the majority of tumour cell lines.
However, comparison of IC50 values of tiazofurin (1) and arabino
analogues 5e7 (Fig. S1b) showed that change to arabino-configu-
ration lead to decrease of potency. These results indicate that xylo-
configuration is more beneficial for antiproliferative activity.
Further, we considered substituent influence at C-30 of xylo-se-
ries (Fig. S1c) and at C-20 of arabino-derivatives (Fig. S1d) on anti-
tumour activity. The data revealed that xylo compounds bearing
amino (3) and acetamido (4) group at C-30 were more active than
corresponding compound with OH group (2) against at least 50% of
cell lines, and amino-derivative 3 was more potent than NHAc
analogue 4 toward majority of investigated tumour cell lines.
Similarly, in the arabino series, the presence of amino group at C-20
influenced increase in potency compared to 20-azido (5) and 20-
NHAc (7) derivatives. Based upon these observations, it appears
that the amino group has the most beneficial effects on antitumour
properties of this type of C-nucleosides.
The percentage of specific necrosis after 72 h for analogues 2, 3
and 4 was between 14% and 21%, while for other analogues it was
slightly lower (6e10%) with respect to tiazofurin (Fig. 3B).
2.5. Western blot analysis
Western blot analysis of samples treated with compounds 2e7
(Table 5 and Fig. S6 in the Supplementary data) shows that
analogue 5 after 72 h most reduced the expression of the anti-
apoptotic Bcl-2 protein relative to control and to 1, while the ana-
logues 2, 3, 4, 6 and 7 increased the total amount of Bcl-2 protein
relative to control. When compared to tiazofurin, the expression of
Bcl-2 is higher only for analogues 2 and 6.
The expression of BAX protein was increased after 72 h treat-
ment by all the tested substances. These results are consistent with
measured increase in specific apoptosis after 72 h.
The expression level of Caspase 3 precursor and its active sub-
unit in cells exposed to the action of tiazofurin and its analogues
2e7 were determined to verify whether these compounds induced
apoptosis by activating the enzyme Caspase 3. Western blot anal-
ysis showed the expression of Caspase 3 precursor in K562 cells
treated with all of the examined analogues (2e7). However, the
expression level of Caspase
3 active subunit was different,
depending on the nature of analogues. A lower effect on the Cas-
pase 3 content in cells after 72-h treatment was observed in tested
substances 4 and 7.
2.3. Cell cycle analysis
The cell cycle profile of exponentially growing K562 cells treated
with synthesized compounds for 24 and 72 h was analyzed by flow
cytometry in cells stained with propidium iodide. Untreated cells
served as a control. The results are presented in Fig. 3.
Western blot analysis showed that the proteolytic cleavage of
PARP protein in K562 cells after treatment with all analogues de-
pends on the analogue structure. Among others, the analogue 3 (
D-
xylo-configuration) induces the lowest PARP cleavage.
In order to determine the antiproliferative effect of synthesized
analogues of tiazofurin, their effect on the cell cycle of K562 cells
was studied using flow cytometry.
2.6. Double fluorescent staining
All compounds of xylo-configuration reduce the percentage of
Morphological features present in apoptotic cells can be

ꢀ
V. Kojic et al. / European Journal of Medicinal Chemistry 183 (2019) 111712
6
Fig. 3. (A) Results of cell cycle phases distribution for control sample (Ctrl) and samples treated with analogues 1e7 for 72 h, obtained by flow cytometry. (B) Specific apoptosis and
necrosis calculated from the results obtained by Annexin-V flow cytometry after cells treatment for 72 h.
Table 5
Results of Western blot analysis.
% of control (72 h)a
Bcl-2
BAX
Caspase3 precursor (32 kDa)
Caspase3 active subunit (18 kDa)
PARP (112 kDa)
PARP (85 kDa)
control
100.00
344.33
436.51
164.90
274.16
35.23
100.00
338.37
180.18
240.62
126.68
186.25
193.00
141.43
100.00
100.00
89.69
230.09
172.46
87.05
156.25
153.73
59.32
100.00
299.68
665.40
84.76
336.99
355.91
387.73
477.15
100.00
214.99
552.68
104.95
364.31
169.74
570.91
534.18
1
2
3
4
5
6
7
1302.79
4433.80
1465.64
432.96
3222.35
1144.41
777.09
414.66
149.59
a
For data recorded after 24-h of cells treatment see the Supplementary data.
visualized by staining with appropriate selective dyes. The loss of
membrane integrity of cells is detectable with fluorescent double
staining, when one of the fluorescent dyes enters the cell freely
(acridine orange) and the other (ethidium bromide) is excluded. In
late apoptosis, both dyes entered the cell and the nuclei are stained
orange or red.
Presence of apoptotic cells was evident in all samples (Fig. S5 in
the Supplementary data). Induced apoptotic changes by tested
compounds were quantified and compared by the ratio of the red
and green colour intensity. The results are presented in Fig. 4A.
The obtained results confirm that tested compounds induce
apoptotic changes in K562 cells. Tiazofurin (1) has the highest
DNA fragmentation and degradation is one of the stages in this
complex process.
3. Conclusions
Six novel tiazofurin mimics (2e7) bearing nitrogen functionalities
at the C-20 or C-30 position have been designed and synthesized in
multistep sequences starting from D-glucose. The synthetic strategy
to targets 2e7 assumed the initial multistep synthesis of suitably
protectedpentofuranosylcyanides, theconstructionofethyl thiazole-
4-carboxylate ring by cyclocondensation of thus obtained glycofur-
anosyl cyanides with L-cysteine ethyl ester followed by dehydroge-
nation. Final exposure of the resulting protected C-nucleosides to
methanolic ammonia provided targets 2e7 in good yields.
impact. Among the tested analogues of D-xylo configuration, com-
pound 2 was the most active.
Synthesized compounds were evaluated for their anti-
proliferative activity against a panel of human malignant cell lines.
The most active compounds that displayed low nanomolar activ-
2.7. Comet assay
Finally, we used the alkaline method of comet assay [29] to
evaluate DNA damage of K562 cells after treatment with synthe-
sized compounds. Comet assay is based on electrophoresis of
nucleic acids from a single cells immobilized in agarose layer,
where the DNA fragments cross a distance in an electric field
depending on their mass, leaving a trail (comet) after staining with
fluorescent dyes. Comet Tail Moment is taken as a parameter for
evaluating DNA fragmentation.
ities from
Jurkat and 0.005
0.0039 M, Raji and 0.0072
that exhibited a nanomolar activity against K562 cells is 20-acet-
amido derivative 7 (IC50 0.0061 M). Unfortunately, this compound
showed a potent cytotoxicity against the normal cell line, MRC-5. A
brief SAR analysis has shown that -xylo configuration increases
antiproliferative activity, while -arabino stereochemistry reduces
D
-xylo series are: 30-amino derivative 3 (IC50 0.0036
m
M,
M, HT-29) and 30-acetamido derivative 4 (IC50
M, Jurkat). The only -arabino isomer
m
m
m
D
m
D
D
Results of the comet assay for samples treated with the tested
derivatives are shown in Fig. 4B. Tiazofurin (1) was the most gen-
otoxic compound to the K562 sells. Analogues of D-xylo configura-
tion (2e4) caused less than half damage to the DNA molecule than
tiazofurin.
the activity of the analogues against the majority of cell lines under
evaluation. All tested compounds increased the percentage of
apoptotic cells, which is confirmed with the consistent results from
flow cytometric cell cycle analysis, Western blot analysis of
apoptotic protein expression and fluorescent double-staining
In addition to assessing genotoxicity, the results of comet assay
can be considered also as an indicator of cell apoptosis, since the
method. Genotoxicity testing showed that the analogues of D-xylo
configuration (2e4) are far less genotoxic than tiazofurin.

ꢀ
V. Kojic et al. / European Journal of Medicinal Chemistry 183 (2019) 111712
7
Fig. 4. (A) Results of the double-fluorescence test for control sample (Ctrl) and samples treated with derivatives 1e4 for 72 h, obtained by analysis of red and green channel density
ratio in the ImageJ computer program. (B) The quantified values of the comet tail moment for control (Ctrl) and treated samples (1e4, 72 h) obtained by the CometScore computer
program.
Due to their potent nanomolar activities, compounds 3,4 and 7
have serious potential as possible antitumour drugs. This is espe-
cially true for analogue 3, which is completely inactive against the
normal cell line (MRC-5). However, in order to prove such a claim,
additional in vivo tests must be done, followed by detailed clinical
trials.
(d, 1 H, J2,3 ¼ 3.7 Hz, H-2), 6.01 (dd, 1 H, J2,3 ¼ 3.7, J3,4 ¼ 3.9 Hz, H-3),
7.37e8.19 (m, 15 H, 3 ꢂ Ph).
13C NMR (62.9 MHz, CDCl3):
d 61.75 (C-5), 70.69 (C-1), 74.97 (C-
2), 79.96 (C-3), 80.39 (C-4), 115.82 (C≡N), 127.89, 128.02, 128.34,
128.46, 128.66, 129.23, 129.68, 129.92, 130.11, 133.23, 133.99 and
134.16 (3 ꢂ Ph), 164.59, 164.73 and 165.95 (3 ꢂ PhC ¼ O).
HRMS (ESI): m/z 472.1380 (MþþH), calcd for C27H22NO7:
472.1391.
4. Experimental section
A minor amount of unreacted starting compound 10 (0.0475 g,
11%) was recovered.
For general experimental procedures and for the preparation of
initial intermediates 9e14, see the Supplementary data.
4.1.2. (2,5-Di-O-benzoyl-3-deoxy-3-trifluoroacetamido-
b-D-
xylofuranosyl)cyanide (16)
4.1. General procedure for the synthesis of glycofuranosyl cyanides
Acetal 11 (0.450 g, 0.93 mmol) was converted into crude 16 ac-
cording to the above general procedure. Pure 16 (0.114 g, 59%), was
15e19
isolated by flash column chromatography (4:1 cyclohexane/
A solution of acetal 9e14 (1 equiv) in a 4:1 mixture of TFA/6 M
HCl (0.2e0.3 M) was kept at þ4 ꢀC for 2e7 days. The mixture was
concentrated to a third of the initial volume and poured into
saturated aq NaHCO3. The aqueous solution was rendered alkaline
with solid NaHCO3 to pH 8e9 and extracted with CH2Cl2. The
combined extracts were washed successively with saturated aq
NaHCO3 and water, dried and evaporated. The remaining crude
aldehydes were immediately dissolved in a mixture of EtOH/CH2Cl2
(0.2e0.3 M) and treated with NaOAc (~3 equiv) and NH2OH ꢂ HCl
(~2 equiv) while stirring at room temperature for 2e6 h. The
mixture was evaporated and the residue distributed between water
and CH2Cl2. The organic layer was separated and the aqueous phase
extracted with CH2Cl2. The combined organic solutions were
washed with water, dried and evaporated to afford crude oximes, as
mixtures of the corresponding E- and Z-isomers. To a cooled (ꢁ15
to ꢁ18 ꢀC) and stirred solution of crude oximes in anhydrous pyr-
idine (~0.2 M) was added dropwise during 0.5 h a cold solution of
MsCl (~4 equiv) in dry pyridine (1 M). The mixture was first stirred
at ꢁ15 ꢀC for 0.5 h, then at room temperature for 2 h and then
poured into a 1:1 mixture of ice and concentrated HCl (pH ~ 2). The
emulsion was extracted with CH2Cl2, the combined extracts were
washed with water, saturated aq NaHCO3 and again with water. The
extract was dried and evaporated, and the residue was purified by
flash column chromatography or by preparative TLC.
23
Me2CO) as a colourless oil, [
(20:1 CH2Cl2/EtOAc).
a
]
D
¼ þ20.5 (c 1.3, CHCl3), Rf ¼ 0.55
IR (film): nmax 3330 (NH),1726 (C¼O, ester),1662 (amide I band),
1554 (amide II band).
1H NMR (400 MHz, CDCl3):
d
4.63 (dd, 1 H, J5a,5b ¼ 12.6,
J4,5a ¼ 4.0 Hz, H-5a), 4.72 (dd, 1 H, J5a,5b ¼ 12.6, J4,5b ¼ 4.2 Hz, H-5b),
4.79e4.85 (m, 2 H, J1,2 ¼ 3.8, J3,4 ¼ 5.5 Hz, H-1 and H-4), 5.00 (ddd,
1 H, J3,NH ¼ 7.6, J2,3 ¼ 3.8, J3,4 ¼ 5.5 Hz, H-3), 5.75 (t, 1 H,
J1,2 ¼ J2,3 ¼ 3.8 Hz, H-2), 7.44e8.16 (m, 10 H, 2 ꢂ Ph), 7.76 (d, 1 H,
J3,NH ¼ 7.6 Hz, NH).
13C NMR (100 MHz, CDCl3):
d 56.2 (C-3), 61.8 (C-5), 70.0 (C-1),
1
78.5 (C-4), 80.0 (C-3), 115.4 (q, CF3, JC,F ¼ 287.6 Hz), 116.0 (C≡N),
127.6, 128.7, 128.72, 128.8, 129.0, 129.8, 130.1, 133.8, 134.4 (2 ꢂ Ph),
157.8 (q, 2JC,F ¼ 38.6 Hz, COCF3), 165.5 and 166.2 (2 ꢂ PhC ¼ O).
LRMS (CI): m/z 463 (MþþH).
4.1.3. (3-Acetamido-2,5-di-O-benzoyl-3-deoxy-b-D-xylofuranosyl)
cyanide (17)
Compound 12 (0.620 g, 1.39 mmol) was converted into crude 17
according to the above general procedure. Pure 17 (0.409 g, 72%
from 3 steps, 79% calculated to reacted 12), was isolated by flash
chromatography (7:3 toluene/EtOAc) as
a
colourless oil,
23
[a]
¼ þ20.7 (c 2.0, CHCl3), Rf ¼ 0.54 (7:3 toluene/EtOAc).
D
IR (film): nmax 3283 (NH), 1723 (C¼O, ester),1661 (amide I band),
4.1.1. (2,3,5-Tri-O-benzoyl-
b-
D-xylofuranosyl)cyanide (15)
1541 (amide II band).
Acetal 10 (0.450 g, 0.93 mmol) was converted into crude 15 ac-
cording to the above general procedure. Pure 15 (0.114 g, 59%), was
1H NMR (250 MHz, CDCl3):
d 2.04 (s, 3 H, CH3CO), 4.61 (m, 2 H,
2 ꢂ H-5), 4.72 (m, 2 H, J1,2 ¼ 3.4, J3,4 ¼ 5.2 Hz, H-2 and H-4), 5.02 (m,
1 H, J1,2 ¼ 3.4 Hz, J3,NH ¼ 7.9 Hz, H-4), 5.66 (t, 1 H, J1,2 ¼ J2,3 ¼ 3.4 Hz,
H-3), 6.98 (d, 1 H, J3,NH ¼ 7.9 Hz, NH), 7.39e8.14 (m, 10 H, 2 ꢂ Ph).
isolated by flash column chromatography (4:1 cyclohexane/
23
Me2CO) as a colourless oil, [
(9:1 toluene/EtOAc).
a]
¼ þ34.6 (c 0.2; CHCl3), Rf ¼ 0.55
D
13C NMR (62.9 MHz, CDCl3):
d 22.8 (CH3CO), 55.3 (C-3), 62.4 (C-
IR (film): nmax 1725 (C¼O), 1263 (CeO).
1H NMR (250 MHz, CDCl3):
4.69 (d, 2 H, J4,5 ¼ 5.9 Hz, 2 ꢂ H-5),
4.88 (m, 1 H, J4,5 ¼ 5.9, J3,4 ¼ 3.9 Hz, H-4), 4.97 (br s, 1 H, H-1), 5.79
5), 70.1 (C-1), 79.1 (C-4), 80.7 (C-2), 116.4 (C≡N), 127.9, 128.1, 128.4,
129.1, 129.6, 129.8, 133.3 and 134.0 (2 ꢂ Ph), 165.3 and 166.0
(2 ꢂ PhC ¼ O), 170.8 (NHCOCH3).
d

ꢀ
V. Kojic et al. / European Journal of Medicinal Chemistry 183 (2019) 111712
8
HRMS (ESI): m/z 409.1389 (MþþH), calcd for C22H20N2O6:
at þ4 ꢀC until the starting materials were consumed (TLC). The
409.1394.
mixture was evaporated, and the residue was purified by flash
A minor amount of unreacted starting compound 12 (0.054 g,
column chromatography or by preparative TLC.
9%) was recovered.
4.2.1. Ethyl 2-(2,3,5-tri-O-benzoyl-b-D-xylofuranosyl)thiazole-4-
4.1.4. (2-Azido-3,5-di-O-benzoyl-2-deoxy-
b
-D
-arabinofuranosyl)
carboxylate (21)
cyanide (18)
Compound 15 (0.2850 g, 0.60 mmol) was converted to crude 21
(0.3532 g) according to the above general procedure. Purification by
Acetal derivative 13 (1.362 g, 3.10 mmol) was converted into
crude 18 according to the above general procedure. Pure product 18
(0.464 g, 41% calculated to reacted 13) was isolated by preparative
preparative TLC (17:3 toluene/EtOAc) gave pure 21 (0.112 g, 31%
23
from 15) as a colourless oil, [
(23:2 toluene/EtOAc).
a
]
¼ þ19.20 (c 1.0, CHCl3), Rf ¼ 0.25
D
TLC (49:1 toluene/EtOAc, developed twice) as colourless crystals,
23
mp 110 ꢀC (CH2Cl2/hexane), [
(9:1 toluene/EtOAc).
a
]
¼ ꢁ20.7 (c 0.5, CHCl3), Rf ¼ 0.59
IR (film): nmax 1724 (C¼O), 1262 (CeO).
D
1H NMR (250 MHz, CDCl3):
d
1.36 (t, 3 H, J ¼ 7.2 Hz, CH2CH3),
0
0
IR (KBr): nmax 2262 (C≡N), 2148 (N3), 1721 (C¼O).
4.37 (q, 2 H, J ¼ 7.2 Hz, CH2CH3), 4.69 (dd, 1 H, J5 a,5 b ¼ 11.7,
1H NMR (250 MHz, CHCl3):
d
4.48 (m, 1 H, H-4), 4.55 (dd, 1 H,
J4 ,5 a ¼ 5.1 Hz, H-50a), 4.78 (dd, 1 H, J5 a,5 b ¼ 11.7, J4 ,5 b ¼ 6.4 Hz, H-
0
0
0
0
0
0
0
50b), 4.93 (m, 1 H, 0H-40), 5.69 (d, 1 H, J1 ,2 ¼ 1.4 Hz, H-1 ), 5.90 (d, 1 H,
0
0
J1,2 ¼ 5.1, J2,3 ¼ 2.4 Hz, H-2), 4.61 (dd, 1 H, J5a,5b ¼ 12.0, J4,5a ¼ 5.3 Hz,
H-5a), 4.67 (dd, 1 H, J5a,5b ¼ 12.0, J4,5b ¼ 5.2 Hz, H-5b), 4.96 (d, 1 H,
J1,2 ¼ 5.1 Hz, H-1), 5.54 (dd, 1 H, J3,4 ¼ 3.4 Hz, J2,3 ¼ 2.4 Hz, H-4),
7.38e8.16 (m, 10 H, 2 ꢂ Ph).
0
0
0
0
0
J3 ,4 ¼ 3.3 Hz, H-3 ), 5.98 (d, 1 H, J1 ,2 ¼ 1.4 Hz, H-2 ), 7.29e8.14 (m,
15 H, 3 ꢂ Ph), 8.21 (s, 1 H, H-5). NOE contact: H-10 and H-40.
13C NMR (62.9 MHz, CDCl3):
d 14.2 (CH2CH3), 61.4 (CH2CH3),
13C NMR (62.9 MHz, CDCl3):
d
63.0 (C-5), 66.1 (C-2), 69.2 (C-1),
62.12 (C-50), 76.0 (C-30), 79.9 (C-40), 81.8 (C-20), 82.9 (C-10), 127.7 (C-
5), 128.3, 128.32, 128.5, 128.7, 129.3, 129.6, 129.7, 129.8, 133.1, 133.6
and 133.6 (3 ꢂ Ph), 147.7 (C-2), 161.1 (C-4), 164.4, 164.5 and 166.0
(3 ꢂ PhC ¼ O), 170.92 (CO2Et).
77.8 (C-3), 82.1 (C-4), 114.2 (C≡N), 120.0, 128.2, 128.3, 128.5, 129.6,
129.65, 133.1 and 133.9 (2 ꢂ Ph), 165.1 and 165.9 (2 ꢂ PhC ¼ O).
HRMS (ESI): m/z 393.1212 (MþþH), calcd for C20H17N4O5:
393.1194.
Anal. Found: C, 60.54; H, 4.14; N, 14.13. Calcd for C20H16N4O5: C,
61.22; H, 4.11; N, 14.28.
HRMS (ESI): m/z 602.1471 (MþþH), calcd for C32H28NO9S:
602.1479.
A minor amount of unreacted starting compound 13 (0.089 g,
6.5%) was recovered.
4.2.2. Ethyl 2-(2,5-di-O-benzoyl-3-deoxy-3-trifluoroacetamido-
-xylofuranosyl)thiazole-4-carboxylate (22)
Compound 16 (0.020 g, 0.04 mmol) was converted in to crude 22
b-
D
4.1.5. (2-Acetamido-2-deoxy-3,5-di-O-benzoyl-
b-
D-
according to the above general procedure. Pure product 22 (0.010 g,
arabinofuranosyl)cyanide (19)
39%) was isolated by preparative TLC (100:3 CH2Cl2/EtOAc, devel-
23
Compound 14 (0.413 g, 0.91 mmol) was converted into crude 19
according to the above general procedure. Pure 19 (0.166 g, 45%
from 14) was obtained after purification by flash column chroma-
tography (8:3 light petroleum/Me2CO) as a colourless syrup. Crys-
oped twice, elution with EtOAc), as a colourless oil, [
a
]
¼ þ25.0 (c
D
0.1, CHCl3), Rf ¼ 0.37 (100:3 CH2Cl2/EtOAc).
IR (film): nmax 3240 (NH), 1721 (C¼O), 1602 (amide I band), 1554
(amide II band).
tallization from CH2Cl2/hexane gave an analytical sample 19, as
1H NMR (400 MHz, CDCl3):
d
1.42 (t, 3 H, J ¼ 7.1 Hz, CH2CH3),
23
colourless crystals, mp 113e114 ꢀC, [
Rf ¼ 0.54 (1:1 toluene/EtOAc).
a
]
¼ ꢁ18.5 (c 1.3, CHCl3),
4.46 (q, 2 H, CH2CH3), 4.57 (dd, 1 H, J4 ,5 a ¼ 7.4, J5 a,5 b ¼ 11.9 Hz, H-
0
0
0
0
D
50a), 4.68 (dd, 1 H, J4 ,5 b ¼ 4.8 Hz, H-50b), 4.90 (m, 1 H, J3 ,4 ¼ 4.1 Hz,
0
0
0
0
H-40), 5.19 (dd, 1 H, J3 ,NH ¼ 9.3 Hz, H-30), 5.37 (d, 1 H, J1 ,2 ¼ 0.9 Hz,
0
0
0
IR (film): nmax 3282 (NH), 1731 (C¼O, ester),1667 (amide I band),
1539 (amide II band).
H-10), 5.66 (d, 1 H, H-20), 7.38e8.12 (m, 10 H, 2 ꢂ Ph), 8.24 (s, 1 H, H-
1H NMR (400 MHz, CDCl3):
d
2.00 (s, 3 H, CH3CO), 4.49 (m, 1 H,
5), 9.23 (d, 1 H, J3 ,NH ¼ 9.3 Hz, NHCOCF3).
0
J4,5 ¼ 6.8 Hz, H-5), 4.63 (dd, 1 H, J5,6a ¼ 4.6, J6a,6b ¼ 12.3, Hz H-6a),
4.67 (dd, 1 H, J6a,6b ¼ 12.3, J5,6b ¼ 4.2 Hz, H-6b), 4.89 (m, 1 H,
J3,NH ¼ 6.3, J3,4 ¼ 6.8, J2,3 ¼ 6.9 Hz, H-3), 5.27 (d, 1 H, J2,3 ¼ 6.8 Hz, H-
2), 5.64 (t, 1 H, J3,4 ¼ 6.8, J2,3 ¼ 6.8 Hz, H-4), 7.15 (bs, 1 H, NH),
7.34e8.10 (m, 10 H, 2 ꢂ Ph).
NOE contact: H-20 and NH, H-50 and NH.
13C NMR (100 MHz, CDCl3):
d
14.2 (CH3CH2), 55.8 (C-30), 61.7
(CH2CH3), 62.1 (C-50), 80.6 (C-40), 81.4 (C-10), 82.0 (C-20), 115.7 (q,
1JC,F ¼ 286.8 Hz, CF3), 128.4 (C-50) 128.3, 128.7, 129.4, 129.8, 130.0,
2
133.2 and 134.1 (2 ꢂ Ph), 148.4 (C-2), 157.4 (q, JC,F ¼ 40.0 Hz,
13C NMR (100 MHz, CDCl3):
d
22.8 (CH3CO), 56.7 (C-3), 63.2 (C-
COCF3), 160.6 (C-4), 165.5 and 166.1 (2 ꢂ PhC ¼ O), 167.4 (CO2Et).
HRMS (ESI): m/z 593.1195 (MþþH), calcd for C27H23F3N2O8S:
593.1206.
6), 69.6 (C-2), 75.6 (C-4), 80.1 (C-5), 115.7 (C≡N), 128.2, 128.5, 128.7,
129.3, 129.8, 129.9, 133.5 and 134.1 (2 ꢂ Ph), 166.2 and 166.3
(2 ꢂ PhC ¼ O), 171.2 (NHCOCH3).
Anal. Found: C, 62.99; H, 4.95; N, 6.63. Calcd for
4.2.3. Ethyl 2-(3-acetamido-2,5-di-O-benzoyl-3-deoxy-b-D-
C
22H20N2O6 ꢂ 0.5H2O: C, 63.25; H, 5.03; N, 6.71.
xylofuranosyl)thiazole-4-carboxylate (23)
Compound 17 (0.020 g, 0.04 mmol) was converted in to crude 23
according to the above general procedure. Pure product 23 (0.013 g,
4.2. General procedure for the conversion of glycofuranosyl
cyanides 15e19 to the protected C-nucleosides 21e25
50%) was isolated by flash column chromatography (3:2 toluene/
23
EtOAc as a colourless oil, [
toluene/EtOAc).
a]
¼ ꢁ19.8 (c 2.3, CHCl3), Rf ¼ 0.45 (3:2
D
To a solution of glycofuranosyl cyanides 15e19 (1 equiv) in a
mixture of absolute MeOH and CH2Cl2 (0.04e0.05 M) was added
L
-
IR (film): nmax 3320 (NH), 1717 (C¼O), 1669 (amide I band), 1543
cysteine ethyl ester hydrochloride (1.5e1.8 eq) and dry Et3N
(1.5e1.8 eq). The mixture was stirred at room temperature (2e4 h)
and then evaporated. The residue was distributed between CH2Cl2
and water, the organic layer was separated washed with saturated
aq NaHCO3 and brine, then dried and evaporated. To a cooled (0 ꢀC)
and stirred solution of crude thiazolines in dry CH2Cl2 (0.1 M) were
added DBU (~2 eq) and BrCCl3 (~1.2 eq). The solution was left
(amide II band).
1H NMR (250 MHz, CDCl3):
d
1.41 (t, 3 H, J ¼ 7.3 Hz, CH2CH3),
2.20 (s, 3 H, CH3CO), 4.43 (q, 2 H, CH2CH3), 4.55 (dd, 1 H,
J4 ,5 a ¼ 7.0 Hz, J5 a,5 b ¼ 12.0 Hz, H-50a), 4.69 (dd, 1 H,0 J4 ,5 b ¼ 3.8,
0
0
0
0
0
0
J5 a,5 b ¼ 12.0 Hz, H-50b), 4.84 (m, 1 H, J3 ,4 ¼ 3.1 Hz, H-4 ), 5.17 (dd0d,
0
0
0
0
1 H, J2 ,3 ¼ 1.2, J3 , NH ¼ 9.5 Hz, H-30), 5.31 (d, 1 H, J1 ,2 ¼ 0.9 Hz, H-1 ),
0
0
0
0
0
5.52 (bs, 1 H, H-20), 6.58 (d, 1 H, NH), 7.25e8.01 (m, 10 H, 2 ꢂ Ph),

ꢀ
V. Kojic et al. / European Journal of Medicinal Chemistry 183 (2019) 111712
9
8.19 (s, 1 H, H-5), 8.63 (2 ꢂ s, 1 H each, NH2).
13C NMR (62.9 MHz, CDCl3):
14.4 (CO2CH2CH3), 23.1 (CH3CO),
4.3.1. 2-(b-D-Xylofuranosyl)thiazole-4-carboxamide (2)
Compound 21 (0.0445 g, 0.082 mmol) was converted to crude 2
according to the above general procedure. Pure 2 (0.0192 g, 89%)
d
55.2 (C-30), 61.6 (CH2CH3), 63.3 (C-50), 81.1 (C-40), 81.3 (C-10), 82.7
(C-20), 128.0 (C-5), 128.4, 128.6, 129.7, 129.72, 129.8, 133.4 and 133.8
(2 ꢂ Ph), 147.7 (C-2), 160.5 (C-4), 165.6 and 166.2 (2 ꢂ PhC ¼ O),
167.8 and 170.6 (NHCOCH3 and CO2CH2CH3).
was isolated by preparative TLC (5:2 CHCl3/MeOH, elution with 1:1
23
EtOAc-iPrOH) as a colourless oil, [
Rf ¼ 0.18 (5:1 CHCl3/MeOH).
a
]
D
¼ ꢁ5.3 (c 0.5; MeOH),
LRMS (FAB): m/z 561.8 (MþþNa).
IR (Nujol): nmax 1666 (amide I band), 1592 (amide II band).
HRMS (ESI): m/z 561.1305 (MþþNa), calcd for C27H26N2NaO8S:
561.1308.
1H NMR (250 MHz, methanol-d4):
0
d
0
3.90 (dd, 1 H, J4 ,5 a ¼ 6.1,
0
0
J5 a,5 b ¼ 11.7 Hz, H-50a), 3.95 (dd,1 H, J4 ,5 b ¼ 5.0, J5 a,5 b ¼ 11.8 Hz, H-
0
0
0
0
50b), 4.15 (dd, 1 H, J2 ,3 ¼ 1.7, J3 ,4 ¼ 3.6 Hz, H-3 ) 4.30 (ddd, 1 H,
0
0
0
0
0
J3 ,4 ¼ 3.7, J4 ,5 a ¼0 6.1, J4 ,5 b ¼ 5.0 Hz, H-40), 4.37 (t, 1 H, J1 ,2 ¼ 1.9,
0
0
0
0
0
0
0
0
4.2.4. Ethyl 2-(2-azido-3,5-di-O-benzoyl-2-deoxy-b-D-
arabinofuranosyl)thiazole-4-carboxylate (24)
Compound 18 (0.384 g, 0.98 mmol) was converted into crude 24
according to the above general procedure. Purification by prepar-
0
0
0
0
0
J2 ,3 ¼ 1.9 Hz, H-2 ), 5.04 (d, 1 H, J1 ,2 ¼ 2.1 Hz, H-1 ), 8.13 (s, 1 H, H-
5).
13C NMR (62.9 MHz, methanol-d4): 61.8 (C-50), 78.6 (C-30), 84.5
d
(C-20 and C-40), 86.4 (C-10), 125.6 (C-5), 150.5 (C-2), 165.9 (C-4),
ative TLC (17:3 toluene/EtOAc) gave pure 24 (0.286 g, 56% from 18)
23
174.9 (CONH2).
as a colourless oil, [
toluene/EtOAc).
a
]
¼ ꢁ6.7 (c 2.9, CHCl3), Rf ¼ 0.43 (17:3
D
HRMS (ESI): m/z 261.0536 (MþþH), calcd for C9H13N2O5S:
261.0540.
IR (film): nmax 2112 (N3), 1724 (C¼O), 1270 (CeO).
1H NMR (250 MHz, CHCl3):
1.35 (t, 3 H, J ¼ 7.1 Hz, CH2CH3),
4.38 (q, 2 H, J ¼ 7.1 Hz, CH2CH3), 4.55 (ddd, 1 H, J4 ,5 a ¼ 5.1,
d
4.3.2. 2-(3-Amino-3-deoxy-b-D-xylofuranosyl)thiazole-4-
carboxamide (3)
Compound 22 (0.018 g, 0.003 mmol) was converted to crude 3
according to the above general procedure. Pure 3 (0.006 g, 69%) was
0
0
0
0
0
0
0
0
0
0
J4 ,5 b ¼ 5.3, J3 ,4 ¼ 2.0 Hz, H-4 ), 4.62 (d,1 H, J1 ,2 ¼ 4.0 Hz, H-2 ), 4.67
(dd,1 H, J4 ,5 a ¼ 5.1, J5 a,5 b ¼ 11.6 Hz, H-50a), 4.74 (dd,10H, J4 ,5 b ¼ 5.3,
0
0
0
0
0
0
J5 a,5 b ¼ 11.6 Hz, H-50b), 5.51 (d, 1 H, J3 ,4 ¼1.8 Hz, H-3 ), 5.68 (d, 1 H,
0
0
0
0
isolated by preparative TLC (5:1 CHCl3/MeOH, elution with 1:1
0
0
0
J1 ,2 ¼ 3.9 Hz, H-1 ), 7.30e8.15 (m, 10 H, 2 ꢂ Ph), 8.21 (s, 1 H, H-5).
23
EtOAc-iPrOH) as a colourless oil, [
Rf ¼ 0.17 (2:1 CHCl3/MeOH).
a]
¼ ꢁ0.82 (c 0.24, MeOH),
13C NMR (62.9 MHz, CDCl3):
d 14.2 (CH2CH3), 61.3 (CH2CH3), 63.4
D
(C-50), 67.8 (C-20), 79.2 (C-30), 80.5 (C-10), 82.6 (C-40), 128.2 (C-5),
128.4, 128.6, 128.9, 129.5, 129.6, 133.0 and 133.7 (2 ꢂ Ph), 146.8 (C-
2), 160.9 (C-4), 165.1 and 166.0 (2 ꢂ PhC ¼ O), 166.6 (CO2CH2CH3).
HRMS (ESI): m/z 523.1277 (MþþH), calcd for C25H23N4O7S:
523.1282.
IR (film): nmax 3344 (OH, NH2), 1669 (amide I band), 1592 (amide
II band).
1H NMR (400 0MHz, methanol-d4):
d
3.44 (dd, 1 H, J2 ,3 ¼ 5.2,
0
0
0
0
0
0
0
0
J3 ,4 ¼ 6.4 Hz, H-3 ), 3.80 (dd, 1 H, J4 ,5 a ¼ 4.7, J5a ,5 b ¼ 12.0 Hz, H-
50a), 3.85 (dd, 1 H, J4 ,5 b ¼ 4.3 Hz, J5a ,5 b ¼ 11.9 Hz, H-50b), 4.20 (pt,
0
0
0
0
1 H, J ¼ 5.2 Hz, H-20), 4.35 (m, 1 H, J3 ,4 ¼ 6.4 Hz, J4 ,5 ¼ 4.5 Hz, H-4 ),
0
0
0
0
0
4.2.5. Ethyl 2-(2-acetamido-3,5-di-O-benzoyl-2-deoxy-b-D-
arabinofuranosyl)thiazole-4-carboxylate (25)
Compound 19 (0.157 g, 0.38 mmol) was converted into crude 25
according to the above general procedure. Pure 25 (0.120 g, 58%
from 19) was obtained after purification on a column of flash silica
0
0
0
4.95 (d, 1 H, J1 ,2 ¼ 5.5 Hz, H-1 ), 8.18 (s, 1 H, H-5).
13C NMR (100 MHz, methanol-d4):
d
61.5 (C-30), 62.1 (C-50), 82.6
(C-40), 84.2 (C-10), 84.5 (C-20), 125.7 (C-5), 150.9 (C-2), 165.7 (C-4),
173.4 (CONH2).
HRMS (ESI): m/z 260.0703 (MþþH), calcd for C9H13N3O4S:
(2:1 light petroleum/Me2CO) in the form of colourless oil,
260.0705.
23
[
a
]
¼ ꢁ40.8 (c 1.2, CHCl3), Rf ¼ 0.19 (2:1 light petroleum/Me2CO).
D
IR (film): nmax 1722 (C¼O), 1602 (amide I band), 1537 (amide II
band).
1H NMR (400 MHz, CDCl3):
4.3.3. 2-(3-Acetamido-3-deoxy-b-D-xylofuranosyl)thiazole-4-
carboxamide (4)
d
1.41 (t, 3 H, J ¼ 7.1 Hz, CH2CH3), 1.70
Compound 23 (0.065 g, 0.24 mmol) was converted to crude 4
according to the above general procedure. Pure 4 (0.031 g, 85%) was
(s, 3 H, CH3CO), 4.43 (q, 2 H, J ¼ 7.1 Hz, CH2CH3), 4.49 (m, 1 H,
J4 ,5a ¼ 4.5 Hz, J4 ,5b ¼ 3.9, H-40), 4.79 (dd, 1 H, J5a ,5b ¼ 12.2,
0
0
0
0
0
obtained after flash column chromatography (7:3 EtOAc/PriOH) as a
J4 ,5a ¼ 4.5 Hz, H-50a), 4.84 (dd, 1 H, J5a ,5b ¼ 12.2, J4 ,5b ¼ 3.90Hz, H-
23
0
0
0
0
0
colourless oil, [
MeOH).
a]
¼ ꢁ3.9 (c 1.3, MeOH), Rf ¼ 0.24 (5:1 CHCl3/
D
50b), 5.21 (ddd, 1 H, J1 ,2 ¼ 5.2, J2 , NH ¼ 8.7, J2 ,3 ¼ 3.9 Hz, H-2 ), 5.51
0
0
0
0
0
0
0
0
0
0
0
0
0
(t, 1 H, J2 ,3 ¼ J3 ,4 ¼ 3.9 Hz H-3 ), 5.65 (d, 1 H, J1 ,2 ¼ 5.2 Hz, H-1 ),
IR (film): nmax 3334 (OH, NH), 1668 (amide I band), 1590 (amide
II band).
0
6.52 (d, 1 H, J2 , NH ¼ 8.7 Hz, NH), 7.42e8.08 (m, 10 H, 2 ꢂ Ph), 8.16 (s,
1 H, H-5). NOE contact: H-10 and H-40, H-30 and H-50.
1H NMR (250 MHz, methanol-d4):
d 1.97 (s, 3 H, CH3CO), 3.68
NMR (100 MHz, CDCl3):
d 14.3 (CH2CH3), 22.8 (CH3CO), 57.0 (C-
(dd, 1 H, J4 ,5 a ¼ 4.6, J5 a,5 b ¼ 12.3 Hz, H-50a), 3.79 (dd, 1 H,
0
0
0
0
20), 61.6 (CH2CH3), 63.6 (C-50), 78.4 (C-10), 78.8 (C-30), 81.9 (C-40),
127.9 (C-5), 128.4,128.6, 128.7, 128.8,129.6,129.7, 129.8,129.9,133.4
and 133.8 (2 ꢂ Ph), 147.6 (C-2), 161.0 (C-4), 165.9 and 166.2
(2 ꢂ PhC ¼ O), 166.9 and 169.8 (NHCOCH3 and CO2CH2CH3).
LRMS (CI): m/z 539 (MþþH).
J5 a,5 b ¼ 12.3, J4 ,5 b ¼ 3.5 Hz, H-50b), 4.36 (t, 1 H, J1 ,2 ¼ J2 ,3 ¼ 5.7 Hz,
0
0
0
0
0
0
0
0
H-20), 4.40 (m, 10H, J3 ,4 ¼ 6.6 Hz, H-4 ), 4.47 (dd, 1 H, J3 ,4 ¼ 6.6,
0
0
0
0
0
0
0
0
0
0
J2 ,3 ¼ 5.7 Hz, H-3 ), 4.98 (d, 1 H, J1 ,2 ¼ 5.7 Hz, H-1 ), 8.23 (s, 1 H, H-
5).
13C NMR (62.9 MHz, methanol-d4):
d
22.5 (CH3CO), 60.0 (C-30),
HRMS (ESI): m/z 539.1490 (MþþH), calcd for C27H27N2O8S:
539.1488.
61.9 (C-50), 81.2 (C-40), 81.8 (C-20), 83.8 (C-10), 125.9 (C-5), 150.8 (C-
2), 165.7 (C-4), 172.6 (CONH2), 173.8 (NHCOCH3).
Anal. Found: C, 44.02; H, 5.18; N, 13.67; S, 10.43. Calcd for
4.3. General procedure for the preparation of C-nucleosides 2e5
and 7
C11H15N3O5S: C, 43.85; H, 5.02; N, 13.95; S, 10.64.
4.3.4. 2-(2-Azido-2-deoxy-b-D-arabinofuranosyl)thiazole-4-
A solution of protected thiazole 21e25 (1 equiv) in saturated
methanolic ammonia was kept at room temperature for 4e8 days,
then evaporated and the residue was purified on a column of silica
gel or by preparative TLC.
carboxamide (5)
Compound 24 (0.286 g, 0.55 mmol) was converted to crude 5
according to the above general procedure. Pure 5 (0.133 g, 85%) was
isolated by preparative TLC (5:1 CHCl3/MeOH, elution with 1:1

ꢀ
V. Kojic et al. / European Journal of Medicinal Chemistry 183 (2019) 111712
10
23
EtOAc/iPrOH) as a colourless oil, [
Rf ¼ 0.42 (5:1 CHCl3/MeOH).
a
]
¼ þ84.6 (c 0.6, MeOH),
4.5. MTT assay
D
IR (film): nmax 2110 (N3), 1664 (amide I band), 1588 (amide II
The colorimetric MTT assay was carried out following the re-
ported procedure [25]. The cytotoxicity of tested compounds was
examined using six human tumour cell lines: K562 (ATCC CCL 243,
chronic myeloid leukemia), HL-60 (ATCC CCL 240, promyelocytic
leukemia), Jurkat (ATCC CCL 1435, T cell leukemia), Raji (ATCC CCL
86, Burkitt's lymphoma), HT-29 (ATCC HTB38, colorectal adeno-
carcinoma) and HeLa (ATCC CCL2, human cervix adenocarcinoma),
as well as one normal human cell line MRC-5 (ATCC CCL 171, lung
fibroblast).
band).
1H NMR (250 MHz, methanol-d4):
d
3.74 (d, 2 H, J4 ,5 ¼ 5.8 Hz,
0
0
2 ꢂ H-50), 3.97 (m, 1 H, J3 ,4 ¼ 3.8, J4 ,5 ¼ 5.9 Hz, H-4 ), 4.21 (dd, 1 H,
0
0
0
0
0
0
0
0
0
0
0
0
0
0
J2 ,3 ¼ 2.6, J3 ,4 ¼ 3.5 Hz, H-3 ), 4.34 (dd,1 H, J1 ,2 ¼ 4.8, J2 ,3 ¼ 2.4 Hz,
H-20), 5.54 (d, 1 H, J1 ,2 ¼ 4.8 Hz, H-1 ), 8.25 (s, 1 H, H-5).
0
0
0
13C NMR (62.9 MHz, methanol-d4): 63.2 (C-50), 71.9 (C-20), 77.7
d
(C-30), 81.0 (C-10), 88.1 (C-40), 126.2 (C-5), 150.7 (C-2), 165.6 (C-4),
170.1 (CONH2).
HRMS (ESI): m/z 286.0599 (MþþH), calcd for C9H12N5O4S:
286.0604.
4.6. Cell cycle analysis
Anal. Found: C, 38.10; H, 3.94; N, 24.68; S, 11.17. Calcd for
C9H11N5O4S: C, 37.89; H, 3.89; N, 24.55; S, 11.24.
4.3.5.
carboxamide (7).
Compound 25 (0.0293 g, 0.054 mmol) was converted to crude 7
according to the above general procedure. Pure 7 (0.0095 g, 58%)
was isolated by preparative TLC (5:1 CHCl3/MeOH, elution with 5:1
Cell cycle analysis of K562 cell line was performed on FACS
Calibur E440 (Becton Dickinson) flow cytometer following the re-
ported procedure [27].
2-(2-Acetamido-b-D-arabinofuranosyl)thiazole-4-
4.7. Detection of apoptosis
23
CHCl3/MeOH) as a colourless oil, [
a
]
¼ ꢁ132.7 (c 0.7, MeOH),
D
Apoptosis of K562 cells was evaluated with an FACS Calibur
E440 (Becton Dickinson) flow cytometer according to the reported
procedure [27]. Results were presented as percent of Annexin V
positive gated cells. Specific apoptosis and necrosis were calculated
according to Bender et al. [28].
Rf ¼ 0.23 (5:1 CHCl3/MeOH).
IR (film): nmax 3445 (OH, NH), 1652 (amide I band), 1558 (amide
II band).
1H NMR (400 MHz, methanol-d4):
d 1.76 (s, 3 H, CH3CO), 3.82
(dd, 1 H, J5 a,5 b ¼ 12.0, J4 ,5 ¼ 3.8 Hz, H-50a), 3.94 (dd, 1 H,
0
0
0
0
0
0
0
0
0
0
0
0
J5 a,5 b ¼ 12.0, J4 ,5 ¼ 2.4 Hz, H-5 b), 3.98 (m, 1 H, J3 ,4 ¼ 4.5 Hz, H-4 ),
4.8. Western blot analysis
0
0
0
0
0
0
0
4.18 (t, 1 H, J2 ,3 ¼0 4.3, J3 ,4 ¼ 4.5 Hz, H-3 ), 4.63 (dd, 1 H, J1 ,2 ¼ 5.8,
0
0
0
0
0
J2 ,3 ¼ 4.3 Hz, H-2 ), 5.51 (d, 1 H, J1 ,2 ¼ 5.8 Hz, H-1 ), 5.82 and 6.12
(2 ꢂ bs, the minor OH signals remained after partial exchange with
methanol-d4), 7.53, 7.93 and 8.18 (3 ꢂ bs, the minor NH2 and NH
signals remained after partial exchange with methanol-d4), 8.20 (s,
1 H, H-5). NOE contact: CH3CO and H-50.
Western blot analysis was performed after treatment of
K562 cells with analogues 2e7 according to the procedure
described in the literature [27].
13C NMR (100 MHz, methanol-d4):
d
22.4 (CH3CO), 61.1 (C-20),
4.9. Double fluorescent staining
62.6 (C-50), 76.9 (C-30), 80.4 (C-10), 87.2 (C-40), 125.8 (C-5), 150.7 (C-
2), 165.7 (C-4), 170.4 (CONH2), 172.7 (NHCOCH3).
LRMS (CI): m/z 302 (MþþH).
We used acridine orange and ethidium bromide staining
method for discrimination of cells on the basis of cell membrane
integrity. The K562 cells were seeded for double fluorescence
staining test and the experimental samples were incubated with
tested compounds at their IC50 concentrations, for 72 h, along with
untreated cells (control). After staining, cells were collected, placed
on clean microscope slides with cover slips and immediately
examined on a fluorescence microscope. Microphotographs with
the fluorescent signal were analyzed in the ImageJ computer pro-
gram (NIH Image, http://imagej.nih.gov), measuring density of
separated red and green colour channels, which ratio is chosen as a
parameter for comparison of samples. Detailed experimental pro-
cedure is given in the Supplementary data.
HRMS (ESI): m/z 324.0628 (MþþNa), calcd for C11H15N3NaO5S:
324.0630.
4.4. 2-(2-Amino-2-deoxy-b-D-arabinofuranosyl)thiazole-4-
carboxamide (6)
A solution of azido derivative 5 (0.133 g, 0.47 mmol) in absolute
EtOH (14 mL) was hydrogenated over 10% Pd/C (0.043 g) for 2 days
at room temperature and normal pressure of H2, then filtered and
the catalyst washed with MeOH. The organic solutions were com-
bined and evaporated and the residue was purified by preparative
TLC (5 plates, 1:2 CHCl3/MeOH, eluted with iPrOH). Pure 6 (0.099 g,
4.10. Comet assay
23
82%) was obtained as a colourless syrup, [
a]
¼ þ61.2 (c 0.3,
D
MeOH), Rf ¼ 0.30 (1:1 CHCl3/MeOH).
For the comet assay, K562 cells were seeded and treated with
IC50 concentrations of tested compounds during 72 h in an incu-
bator. Untreated cells were used as a control sample. A modified
version of the alkaline Comet assay [29] was performed to evaluate
DNA damage (single and double breaks). Cells were immobilized on
slides in agarose layers, prepared according to the protocol and
exposed to electric field in horizontal electrophoresis unit. Washed
and dried slides were then stained with ethidium bromide fluo-
rescent dye and analyzed using a fluorescence microscope. A series
of microphotographs were taken by a digital camera and 50 comets
per sample were analyzed by CometScore computer program. The
mean value of the parameter “Tail Moment" was selected to
compare the samples. Complete experimental procedure used for
the comet assay is described in the Supplementary data.
IR (film): nmax 3334 (OH, NH2), 1666 (amide I band), 1559 (amide
II band).
1H NMR (250 0MHz, methanol-d4):
d
3.59 (dd, 1 H, J1 ,2 ¼ 4.8,
0
0
0
0
0
0
0
0
J2 ,3 ¼ 2.4 Hz, H-2 ), 3.76 (dd, 1 H, J4 ,5 a ¼ 3.8, J5 a,5 b ¼ 12.0 Hz, H-
50a), 3.87 (dd, 1 H, J4 ,5 b ¼ 3.0, J5 a,5 b ¼ 12.0 Hz, H-50b), 3.99 (m, 1 H,
0
0
0
0
J3 ,4 ¼ 3.4, J4 ,5 a ¼ 3.8, J5 a,5 b ¼ 12.0, Hz, H-40), 4.08 (dd, 1 H,
0
0
0
0
0
0
0
0
0
0
0
0
0
0
J3 ,4 ¼ 3.4, J2 ,3 ¼ 2.4 Hz, H-3 ), 5.39 (d, 1 H, J1 ,2 ¼ 4.9 Hz, H-1 ), 8.20
(s, 1 H, H-5).
13C NMR (62.9 MHz, methanol-d4): 62.0 (C-50), 62.9 (C-20), 79.9
d
(C-30), 81.9 (C-10), 88.6 (C-40), 125.8 (C-5), 151.0 (C-2), 165.8 (C-4),
170.9 (CONH2).
HRMS (ESI): m/z 260.0696 (MþþH), calcd for C9H14N3O4S:
260.0699.

ꢀ
V. Kojic et al. / European Journal of Medicinal Chemistry 183 (2019) 111712
11
Erythrocytotoxicity of tiazofurin in vivo and in vitro detected by scanning
probe microscopy, Toxicol. Lett. 146 (2004) 275e284.
Author contributions
ꢀ
ꢀ
ꢀ
[12] V. Vranic, K. Savovski, N. Dedovic, B. Dimitrijevic, Hematological toxicity
associated with tiazofurin influence on erythropoiesis, Toxicol. Lett. 14 (2000)
181e190.
The manuscript was written through contributions of all au-
thors. All authors have given approval to the final version of the
manuscript.
[13] D.T. Mao, V.E. Marquez, Synthesis of 2-b-d-ara- and 2-b-d-xylofuranosylth-
iazole-4-carboxamide, Tetrahedron Lett. 25 (1985) 2111e2114.
[14] B.M. Goldstein, D.T. Mao, V.E. Marquez, Ara-tiazofurin: conservation of
structural features in an unusual thiazole nucleoside, J. Med. Chem. 31 (1988)
1026e1031.
Acknowledgement
[15] W. Zhen, H.N. Jayaram, V.E. Marquez, B.M. Goldstein, D.A. Cooney, G. Weber,
Cytotoxicity of tiazofurin and its arabinose and xylose analogues in K562 cells,
Biochem. Biophys. Res. Commun. 180 (1991) 933e938.
Financial support from the Ministry of Education, Science and
Technological Development of the Republic of Serbia (Grant No
172006) is gratefully acknowledged. This work has also received
funding from the Serbian Academy of Sciences and Arts under
strategic projects programme (grant agreement No 01-2019-F65).
ꢀ
ꢁ
ꢀ
ꢀ
[16] M. Popsavin, S. Spaic, M. Svircev, V. Kojic, G. Bogdanovic, V. Popsavin, Syn-
thesis and in vitro antitumor screening of 2-( -d-xylofuranosyl)thiazole-4-
b
carboxamide and two novel tiazofurin analogs with substituted tetrahy-
drofurodioxol moiety as a sugar mimic, Bioorg. Med. Chem. Lett 22 (2012)
6700e6704.
[17] P. Nie, E. Groaz, D. Daelemans, P. Herdewijn, Xylo-C-nucleosides with a pyr-
rolo[2,1-f][1,2,4]triazin-4-amine heterocyclic base: synthesis and anti-
proliferative properties, Bioorg, Med. Chem. Lett. 29 (2019) 1450e1453.
Appendix A. Supplementary data
ꢀ
ꢀ
ꢁ
ꢀ
ꢀ
[18] M. Popsavin, V. Kojic, Lj Torovic, M. Svircev, S. Spaic, D. Jakimov, L. Aleksic,
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.ejmech.2019.111712.
ꢀ
G. Bogdanovic, V. Popsavin, Synthesis and in vitro antitumour activity of
tiazofurin analogues with nitrogen functionalities at the C-20 position, Eur. J.
Med. Chem. 111 (2016) 114e125.
[19] T.D. Colby, K. Vanderveen, M.D. Strickler, G.D. Markham, B.M. Goldstein, Proc.
Natl. Acad. Sci. U.S.A. 96 (1999) 3531e3536.
References
ꢀ
ꢁ
ꢀ
ꢀ
[20] M. Popsavin, S. Spaic, M. Svircev, V. Kojic, G. Bogdanovic, V. Popsavin, 2-(3-
Amino-3-deoxy- -d-xylofuranosyl)thiazole-4-carboxamide: a new tiazofurin
[1] K. Temburnikar, K.L. Seley-Radtke, Recent advances in synthetic approaches
for medicinal chemistry of C-nucleosides, Beilstein J. Org. Chem. 14 (2018)
772e785.
[2] G.R. Ott, D.A. Favor, Pyrrolo[2,1-f][1,2,4]triazines: from C-nucleosides to ki-
nases and back again, the remarkable journey of a versatile nitrogen hetero-
cycle, Bioorg, Med. Chem. Lett. 27 (2017) 4238e4246.
[3] E. De Clercq, C-Nucleosides to be revisited, J. Med. Chem. 59 (2016)
2301e2311.
[4] P. Merino, M. Ghirardello, T. Tejero, I. Delso, R. Matute, Recent advances on the
enantioselective synthesis of C-nucleosides inhibitors of inosine mono-
phosphate dehydrogenase (IMPDH), Curr. Top. Med. Chem. 14 (2014)
1212e1224.
[5] M.F.A. Adamo, R. Pergoli, Synthesis and medicinal properties of 2-deoxyribose
and ribose C-nucleosides, Curr. Org. Chem. 12 (2008) 1544e1569.
[6] M. Grifantini, Tiazofurin ICN pharmaceuticals, Curr. Opin. Investig. Drugs 1
(2018) 80e97.
[7] V. Gandin, P. Khalkar, J. Braude, A.P. Fernandes, Organic selenium compounds
as potential chemotherapeutic agents for improved cancer treatment, Free
Radic. Biol. Med. 127 (2004) 1657e1669.
[8] L. Hedstrom, IMP dehydrogenase: structure, mechanism, and inhibition,
Chem. Rev. 109 (2009) 2903e2928.
[9] K.W. Pankiewicz, S.E. Patterson, P.L. Black, H.N. Jayaram, D. Risal,
B.M. Goldstein, L.J. Stuyver, R.F. Schinazi, Cofactor mimics as selective in-
hibitors of NAD-dependent inosine monophospate dehydrogenase (IMPDH) -
the major therapeutic target, Curr. Med. Chem. 11 (2004) 887e900.
b
analogue with potent antitumour activity, Bioorg. Med. Chem. Lett 16 (2006)
5317e5320.
ꢀ
ꢀ
ꢀ
ꢀ
[21] M. Popsavin, Lj Torovic, V. Kojic, G. Bogdanovic, S. Spaic, V. Popsavin, De novo
synthesis of two new cytotoxic tiazofurin analogues with modified sugar
moieties, Bioorg. Med. Chem. Lett 13 (2003) 3167e3170.
[22] DiscoveryStudioVisualizer, Bovia Corp, 2016 v. 17.2.0.16349.
[23] R.S. Brown, J. Dowden, C. Moreau, B.V.L. Potter, A concise route to tiazofurin,
Tetrahedron Lett. 43 (2002) 6561e6562.
[24] E. Aguilar, A.I. Meyers, Reinvestigation of a modified Hantzsch thiazole syn-
thesis, Tetrahedron Lett. 35 (1994) 2473e2476.
[25] D.A. Scudiero, R.H. Shoemaker, K.D. Paull, A. Monks, S. Tierney, T.H. Nofziger,
M.J. Currens, D. Seniff, M.R. Boyd, Evaluation of a soluble tetrazolium/for-
mazan assay for cell growth and drug sensitivity in culture using human and
other tumor cell lines, Cancer Res. 48 (1988) 4827e4833.
[26] Antiproliferative Activities of Analogues 3 and 4 against Four Cell Lines Were
Reported in the Preliminary Communications, but after 24-hours of cells
treatment (see, references 20 and 21).
ꢁ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
[27] I. Kovacevic, M. Popsavin, G. Benedekovic, J. Kesic, V. Kojic, D. Jakimov,
ꢀ
ꢀ
ꢀ
T. Srdic-Rajic, G. Bogdanovic, V. Divjakovic, V. Popsavin, Synthesis and in vitro
antitumour activity of crassalactone D, its stereoisomers and novel cinnamic
ester derivatives, Eur. J. Med. Chem. 134 (2017) 293e303.
[28] A. Bender, D. Opel, I. Naumann, R. Kappler, L. Friedman, D. Von Schwinitz, K.-
M. Debatin, S. Fulda, PI3K inhibitors prime neuroblastoma cells for chemo-
therapy by shifting the balance towards pro-apoptotic Bcl-2 proteins and
enhanced mitochondrial apoptosis, Oncogene 30 (2011) 494e503.
[29] O. Merk, G. Speit, Detection of crosslinks with the Comet assay in relationship
to genotoxicity and cytotoxicity, Environ. Mol. Mutagen. 33 (1999) 167e172.
€
ꢀ
ꢀ
[10] G. Weber, F. Shen, T.I. Orban, S. Kokeny, E. Olah, Targeting signal transduction,
Adv. Enzym. Regul. 43 (2003) 47e56.
[11] V. Vranic-Mandusic, V. Subota, K. Savovski, L. Medic, T. Dramicanin,
ꢀ
ꢁ ꢀ
ꢀ
ꢀ
ꢀ
ꢁ
ꢀ
ꢀ
O. Jozanov-Stankov, D. Popov-Celeketic, M. Jokanovic, B. Dimitrijevic,