Disaccharide pyrimidine nucleosides and their derivatives: A novel group of cell-penetrating inhibitors of poly(ADP-Ribose) Polymerase 1

Article abstract of DOI:10.1080/15257770.2013.827793

Nearly 30 synthetic nucleosides were tested with human recombinant poly(ADP-ribose) polymerase 1 as potential inhibitors of this enzyme. The most active compounds were some disaccharide analogues of thymidine: 3′-O-β-D-ribofuranosyl-5-iodo-dUrd (2d; IC

Full text of DOI:10.1080/15257770.2013.827793

This article was downloaded by: [University of Calgary]
On: 05 October 2013, At: 01:47
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Nucleosides, Nucleotides and Nucleic
Acids
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/lncn20
Disaccharide Pyrimidine Nucleosides
and Their Derivatives: A Novel Group of
Cell-Penetrating Inhibitors of Poly(ADP-
Ribose) Polymerase 1
Anna S. Efremova a , Alexandra L. Zakharenko b , Stanislav I. Shram
a , Irina V. Kulikova c , Mikhail S. Drenichev c , Maria V. Sukhanova b
,
Svetlana N. Khodyreva b , Nikolay F. Myasoedov a , Olga I. Lavrik b
Sergey N. Mikhailov c
&
a Institute of Molecular Genetics , Russian Academy of Sciences ,
Moscow , Russia
b Institute of Chemical Biology and Fundamental Medicine , Siberian
Branch of the Russian Academy of Sciences , Novosibirsk , Russia
c Engelhardt Institute of Molecular Biology , Russian Academy of
Sciences , Moscow , Russia
Published online: 30 Sep 2013.
To cite this article: Anna S. Efremova , Alexandra L. Zakharenko , Stanislav I. Shram , Irina V.
Kulikova , Mikhail S. Drenichev , Maria V. Sukhanova , Svetlana N. Khodyreva , Nikolay F. Myasoedov ,
Olga I. Lavrik & Sergey N. Mikhailov (2013) Disaccharide Pyrimidine Nucleosides and Their Derivatives:
A Novel Group of Cell-Penetrating Inhibitors of Poly(ADP-Ribose) Polymerase 1, Nucleosides,
Nucleotides and Nucleic Acids, 32:9, 510-528, DOI: 10.1080/15257770.2013.827793
To link to this article: http://dx.doi.org/10.1080/15257770.2013.827793
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or

howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions

Nucleosides, Nucleotides and Nucleic Acids, 32:510–528, 2013
Copyright ꢀC Taylor and Francis Group, LLC
ISSN: 1525-7770 print / 1532-2335 online
DOI: 10.1080/15257770.2013.827793
DISACCHARIDE PYRIMIDINE NUCLEOSIDES AND THEIR
DERIVATIVES: A NOVEL GROUP OF CELL-PENETRATING
INHIBITORS OF POLY(ADP-RIBOSE) POLYMERASE 1
Anna S. Efremova,1 Alexandra L. Zakharenko,2 Stanislav I. Shram,1
Irina V. Kulikova,3 Mikhail S. Drenichev,3 Maria V. Sukhanova,2
Svetlana N. Khodyreva,2 Nikolay F. Myasoedov,1 Olga I. Lavrik,2
and Sergey N. Mikhailov3
1Institute of Molecular Genetics, Russian Academy of Sciences, Moscow, Russia
2Institute of Chemical Biology and Fundamental Medicine, Siberian Branch of the Russian
Academy of Sciences, Novosibirsk, Russia
3Engelhardt Institute of Molecular Biology, Russian Academy of Sciences, Moscow, Russia
Nearly 30 synthetic nucleosides were tested with human recombinant poly(ADP-ribose) poly-
2
merase 1 as potential inhibitors of this enzyme. The most active compounds were some dis-
accharide analogues of thymidine: 3ꢁ-O-β-D-ribofuranosyl-5-iodo-dUrd (2d; IC50 = 45 μM),
3ꢁ-O-β-D-ribofuranosyl-2ꢁ-deoxythymidine (2e; IC50 = 38 μM), and 3ꢁ-O-β-D-ribofuranosyl-2ꢁ-
deoxythymidine oxidized (4; IC50 = 25 μM). These compounds also reduced H2O2-induced synthe-
sis of poly(ADP-ribose) in cultured human ovarian carcinoma (SKOV-3) cells in a dose-dependent
manner. Furthermore, compounds 2d or 2e until a concentration of 1 mM did not affect growth
of SKOV-3 cells, whereas dialdehyde compound 4, as well as thymidine, exhibited a significant
cytotoxicity.
Keywords Disaccharide pyrimidine nucleosides; poly(ADP-ribose) polymerase 1
inhibitors; DNA repair; cytotoxicity; SKOV-3 human ovarian carcinoma cell line
INTRODUCTION
Nucleoside analogues may be considered as one of the most promis-
ing group of organic compounds in medicinal chemistry. Numerous drugs
have been developed, which have or resemble nucleoside structure. Most
of nucleosides should be phosphorylated to target polymerases and to exert
Received 29 May 2013; accepted 18 July 2013.
This work was supported by the Program “Molecular and Cell Biology” of the Russian Academy of
Sciences and by the Russian Foundation for Basic Research (grant nos. 10-04-01770, 11-04-12099, 11-04-
01434, and 12-04-92601). The authors gratefully acknowledge the assistance of Artem A. Tikhomirov,
PhD, in preparing the article.
Address correspondence to Sergey N. Mikhailov, Engelhardt Institute of Molecular Biology, Vavilova
Str. 32, Moscow 119991, Russia. E-mail: smikh@eimb.ru
510

Disaccharide Nucleosides as Inhibitors of PARP-1
511
FIGURE 1 Structure of poly(ADP-ribose).
their biological effect.[1,2] In spite of great achievements in this area, there
is still a room for finding new targets for nucleoside analogues. Poly(ADP-
ribose) polymerase 1 (PARP-1) has been an actively pursued drug discovery
target for almost three decades. Often referred to as a “guardian angel of
DNA,” this nuclear enzyme has been the focus of several medicinal chem-
istry programs in a wide range of therapeutic areas. Targeting DNA repair
with PARP inhibitors has great promise as anticancer therapeutic strategy
either as a single agent or as a resistance modifier in conjunction with other
DNA-damaging agents including radiation therapy.[3,4]
PARP-1 converts nicotinamide adenine dinucleotide (NAD+) into a
polymer—poly(ADP-ribose) (PAR; Figure 1). It is known that PAR is in-
volved in modulation of the chromatin structure, replication, and transcrip-
tion, as well as in DNA repair and cell differentiation.[5,6] PAR molecules
isolated from natural sources or synthesized in vitro with PARP contain up
to 200–300 monomeric units. Linear regions of 20 to 50 units alternate with
branched fragments.[7,8] The development of methods for the preparation
of PAR is still a challenging problem. Only recently two syntheses of 2ꢁ-O-α-
D-ribofuranosyladenosine, the monomeric unit of PAR, were published.[9,10]
Disaccharide nucleosides comprise an important group of natural com-
pounds. About one hundred disaccharide nucleosides and their analogues
have been isolated to date from various sources.[11–13] These compounds
have a broad range of biological activities and display antibacterial, an-
timycotic, herbicidal, insecticidal, antitumor, and antiviral properties. Com-
pounds of this type have an additional saccharide residue attached to one
of the nucleoside hydroxyl groups via an O-glycoside bond. The presence
of a disaccharide residue and a heterocyclic base makes their properties

512
A. S. Efremova et al.
similar to those of carbohydrates and nucleosides. We investigated 2ꢁ-O-α-D-
ribofuranosyladenosine and some other disaccharide nucleosides as poten-
tial inhibitors of PARP-1, having in mind that thymidine (Thd) and some of
its analogues may inhibit this enzyme.[14]
Most of the known competitive inhibitors of PARP-1 are heterocyclic
compounds containing primary or secondary amide group.[3,15] Ability of
these compounds to inhibit PARP-1 is associated with their structural ho-
mology with nicotinamide, which is known to be one of the endogenous
inhibitors of PARP-1. Thd is thought to be another endogenous inhibitor
of this enzyme. Among all natural nucleosides, Thd is unique to possess
notable PARP-1 inhibitory activity.[16] Furthermore, Thd has a stronger in-
hibitory effect on PARP-1 than thymine, and some other nucleic acid compo-
nents similar to Thd—such as uridine (Urd), 2ꢁ-deoxyuridine (dUrd), thymi-
dine monophosphate, and 2ꢁ-deoxythymidine triphosphate—completely
lack such activity.[16] Later structural and functional studies have shown
that a number of 5-substituted derivatives of Urd and dUrd possess signifi-
cantly higher activity than Thd.[14] It was found that changes in the structure
of the sugar residue can also significantly affect the inhibitory potential of
nucleosides.[15,17]
This study was aimed to investigate properties of several disaccharide nu-
cleoside analogues as potential inhibitors of human PARP-1. To do this, ex-
periments on the isolated recombinant enzyme and cultured human tumor
cells were performed and cytotoxicity of these compounds was evaluated.
RESULTS
Inhibition of Human Recombinant PARP-1
by Nucleoside Analogues
For primary screening, effects of about 30 analogues of nucleosides
(mainly disaccharides) on human recombinant PARP-1 have been evaluated.
The most important results are summarized in Table 1 (for an overview on
all compound structures also, see Table 1). It was shown that a number of
disaccharide analogues of Thd (compounds 1d, 2e, 2f, 2h, and 4), like the
Thd itself, exerts a considerable inhibitory effect on PARP-1. Estimated IC50
values for these compounds vary from 25 to 216 μM. On the contrary, Urd
disaccharide derivative 1c, as well as Urd itself, did not show any PARP-1 in-
hibitory activity. However, disaccharide derivative of 5-iodo-dUrd (2d), but
not 5-fluoro-dUrd (2c), was as effective in inhibiting PARP-1 (IC50 = 45 μM)
as the thymidine derivatives. Interestingly, several disaccharide nucleosides
(compounds 2d, 2e, and 4) demonstrated a similar or even higher inhibitory
activity than 3-aminobenzamide (3-AB), a widely used inhibitor of PARP-1.
Disaccharide nucleoside derivatives in which the nitrogenous bases are Ade

Disaccharide Nucleosides as Inhibitors of PARP-1
513
TABLE 1 Inhibition of human recombinant PARP-1 by some synthetic nucleoside analogues with
different carbohydrate moiety
Carbohydrate moiety
Compound
Base
Modifications
IC50, μM
nH
D-Ribose or
Cyd
Cyt
Cyt
> 2000
> 2000
> 2000
54 2a
> 2000
> 2000
> 2000
2ꢁ-Deoxy-D-ribose
dCyd
Urd
Thd
1a
1b
1c
Ura
Thy
Ade
Cyt
Ura
Thy
0.92 0.04a
1.03 0.04
1d
178
6
2a
2b
2c
2d
2e
2f
Ade
Cyt
> 2000
320
> 2000
1.20 0.20
Ura
Ura
Thy
Thy
Thy
Thy
5-fluoro
5-iodo
45
38
3
4
0.80 0.05
1.16 0.14
0.97 0.13
5ꢁ-β-D-ribofuranosyl
5ꢁ-phosphate
110 16
> 2000
216 56
2g
2h
5ꢁꢁ-phosphate
1.06 0.25
3ꢁ-Rib-α-Thd
3
4
Thy
Thy
> 2000
25
3
1.23 0.17
5
Ura
Ura
139 22
1.47 0.28
6
> 2000
3-Aminobenzamide
57
8
0.91 0.09
a The values of IC50 and nH in the table are presented as mean SE.

514
A. S. Efremova et al.
or Cyt were not active, with an exception of compound 2b that demonstrated
a weak inhibitory activity (IC50 = 320 μM). It is worth noting that 2ꢁ-O-α-
D-ribofuranosyladenosine,[9] a monomeric unit of PAR, was not also active
towards enzyme (IC50 > 2000 μM).
Oxidized Thd disaccharide derivative 4, which can be readily prepared by
treatment of 2e with a small excess of NaIO4,[18–21] exerts most pronounced
inhibitory activity (IC50 = 25 μM) among all nucleoside analogues tested.
Its precursor, compound 2e, demonstrated only slightly weaker inhibitory
activity (IC50 = 38 μМ). It should be noted that α-anomer of 2e, compound
3, did not show any inhibitory effect at concentrations up to 1 mM. Com-
pounds 1d, which is the 2ꢁ-O-β-D-ribofuranosyl derivative of ribothymidine,
demonstrated substantially lower inhibitory activity (IC50 = 178 μМ) than its
3ꢁ-analogue compound 2e. Interestingly, the 3ꢁ-O-β-D-ribofuranosyl deriva-
tive of 2ꢁ-O-deoxycytidine (dCyd), compound 2b, also showed an appreciable
inhibitory activity, while dCyd, cytidine (Cyd), or its 2ꢁ-O-disaccharide deriva-
tive 1b were completely inactive. In addition, it was shown that analogues of
compound 2e phosphorylated in 5ꢁ or 5ꢁꢁ position (2g and 2h) were signif-
icantly less active (2h) or not active at all (2g). Furthermore, it was shown
that the trisaccharide derivative of Thd, which contains two ribose residues
at the 2ꢁ and 5ꢁ positions (2f), markedly inhibited PARP-1 (IC50 = 110 μM),
although its inhibitory potential was lower than that of compound 2e.
Finally, PARP-1 inhibitory potential of two nucleoside analogues of Urd
with oxidized ribose (5) or erythrose (6) sugar residue was investigated.
These compounds have structural homology; however, only compound 5
showed significant inhibitory activity (IC50 = 139 μM). This could be ex-
plained by the structural features of carbohydrate moiety, because neither
Urd nor its disaccharide analogue 1c inhibited PARP-1.
Since we were dealing with the novel compounds, it was appropriate
to determine the values of Hill coefficient (nH), which not only identifies
the various cooperative effects but also helps to detect nonideal inhibition.
Estimated values of nH for almost all active nucleosides tested were close to
1.00 (Table 1). The only exception was compound 5, for which a noticeable
positive cooperativity was observed (nH = 1.47).
Effects of Nucleoside Analogues on H2O2-Induced
Poly(ADP-ribose) Synthesis in Cultured Tumor Cells
The ability of some nucleoside inhibitors of PARP-1 to suppress H2O2-
induced elevation of PAR synthesis in SKOV-3 human ovarian carcinoma cell
line was assessed. It was demonstrated in control experiments that preincu-
bation of SKOV-3 cells with the compounds tested did not stimulate basal
PAR synthesis (Figure 2). After five-minute incubation with 1 mM H2O2, we
observed a 20–25-fold increase of PAR level in cell nuclei as compared with
the control (Figures 2 and 3).

Disaccharide Nucleosides as Inhibitors of PARP-1
515
FIGURE 2 Inhibition of H2O2-induced synthesis of poly(ADP-ribose) (PAR) in cultured SKOV-3 human
ovarian carcinoma cells by compound 4. SKOV-3 cells were preincubated with 0.01 mM, 0.1 mM, 1 mM,
or without compound 4 for one hour prior to five-minute treatment of cell cultures with 1 mM H2O2. The
fixed cells were stained with anti-PAR antibodies (green) and propidium iodide (red) for counterstaining
nuclei and then subjected to the fluorescent microscopy. The scale bar in the bottom right corner of the
figure corresponds to 100 μm and refers to all images. (Color figure available online).
Preincubation of cells with compounds 2d, 2e (Figures 3A, 3B), or 4
(Figures 2 and 3C) for one or 24 hours resulted in dose-dependent down-
regulation of PAR formation. However, the inhibitory effect of these com-
pounds was weaker than that of Thd (Figure 3E). One-hour preincubation
with 0.1 mM of Thd (Figure 3E), compound 4 (Figure 3C), or compound 2e
(Figure 3B) led to a decrease of stimulated synthesis of PAR by 72%, 50%,
and 25%, respectively. It is obvious that the lower inhibitory potential of the
Thd derivatives is determined by an additional ribose residue in 3ꢁ-position.
However, a compound containing an oxidized second ribose residue (4)
inhibits PAR production to a greater extent than its counterpart, compound
2e (Figures 3B and 3C). It is of interest that one-hour preincubation of cells
with all tested substances, with an exception of compound 2e, exerted a
more pronounced effect on PAR synthesis than a 24-hour exposure. The
one-hour preincubation of SKOV-3 cells with 0.1 mM of oxidized Urd (5)
slightly inhibited stimulated PAR synthesis, but at higher concentrations this
nucleoside intensified the H2O2-induced PAR synthesis (Figure 3D). More-
over, a marked cytotoxic effect of compound 5 after 24-hour preincubation
was observed as well (data not shown).
In addition, the nucleoside analogues tested also exerted their inhibitory
effect on PARPs in cell culture. It is noteworthy that their inhibitory efficacy
in cell culture proved to be somewhat lower than what was observed with
the recombinant PARP-1. In contrast, 3-AB demonstrated the same efficacy
in culture and with the recombinant PARP-1 (Table 1, Figure 3F), whereas
efficacy of Thd was even higher in culture than with the recombinant protein.

516
A. S. Efremova et al.
FIGURE 3 Effect of exposure of the SKOV-3 human ovarian carcinoma cells to increasing concentrations
of tested nucleosides (A-E) or 3-aminobenzamide (3-AB; F) on H2O2-induced PAR synthesis. SKOV-3
cells were preincubated for one or 24 hours with compounds 2d (A), 2e (B), 4 (C), 5 (D), Thd (E),
3-AB (F), or without these compounds prior to five-minute treatment of cell cultures with 1 mM H2O2.
The level of PAR synthesis in nuclei of cells treated with 1mM H2O2 alone was taken as 100%. Intact
cells, which were not subjected to any treatments, were considered as a negative control. The values of
H2O2-induced PAR synthesis are presented as mean SD. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 (Student’s
t-test). (Color figure available online).
Cytotoxic Effects of Nucleoside Analogues on Cultured
Tumor Cells
Some of the nucleoside analogues were selected for cytotoxicity study.
Among them, compounds 2d, 2e, and 4 were taken due to their high in-
hibitory activity with respect to PARP-1, whereas other compounds have
lower activity or none (2a, 5, and 6). Thd, dAdo, and 3-AB were used as
reference compounds. Almost two-fold decrease of the cell density was ob-
served after 72-hour incubation of SKOV-3 cells with 1 mM dAdo or Thd as
compared with the control (Figure 4). In contrast, disaccharide derivatives

Disaccharide Nucleosides as Inhibitors of PARP-1
517
FIGURE 4 Effect of some tested nucleosides and 3-aminobenzamide (3-AB) on growth of the cultures of
the SKOV-3 human ovarian carcinoma cells. SKOV-3 cells were cultured for 72 hours in growth medium
containing 0.1 or 1 mM of 3-AB, dAdo, Thd, or compounds 2a, 2d, 2e, 4, 5, 6, or contained 0.1% DMSO
(the amount of DMSO equivalent to that added to growth medium of cell cultures treated with nucleoside
analogues in concentration of 1 mM), or without any additives (control). Cell density in each premarked
fields prior to adding the test substances was taken as 100%. The values of the number of MTT-positive
cells are presented as mean SD. ∗∗∗p < 0.001 (Student’s t-test).
of dAdo and Thd containing a second ribose residue at the 3ꢁ-position
(2a and 2e) were virtually nontoxic in concentrations up to 1 mM (Fig-
ure 4). Moreover, 5-iodo-dUrd derivative 2d with similar structure was not
toxic. Thus, disaccharide derivatives of Thd, dAdo, and 5-iodo-dUrd with an
additional ribose moiety at 3ꢁ-position did not display any toxicity, unlike
their prototypes.
In contrast to Thd disaccharide analogue 2e, its oxidized derivative 4 ap-
peared to be toxic for SKOV-3 cells. It was shown that 72-hour incubation of
SKOV-3 cells with 1 mM of compound 4 decreased cell density by 75% versus
control, whereas this nucleoside had no effect on growth of cell culture in a
concentration of 0.1 mM. It is worth noting that the oxidized Urd derivatives
5 and 6 exhibited even greater cytotoxicity. When SKOV-3 cells were incu-
bated for 72 hours in the presence of 1 mM of these compounds, almost all
cultured cells died. At a concentration of 0.1 mM, these compounds caused
approximately two-fold decrease of cell density (Figure 4).
These results indicate that the cytotoxic effect of the nucleosides tested
is not determined by their PARP-1 inhibitory activity. This suggestion is
confirmed by an observation that a conventional PARP-1 inhibitor, 3-AB, at
a concentration of 1 mM completely inhibited PARP-1 but did not affect
proliferative activity of SKOV-3 cells (Figure 4).

518
A. S. Efremova et al.
DISCUSSION
It is known that Thd, unlike other natural nucleosides and nitroge-
nous bases, is able to effectively suppress PARP activity in cells.[16] The de-
pendence of PARP-1 inhibition efficiency on the structure of nitrogenous
bases for different synthetic nucleosides has been investigated in numerous
works.[14,16,22] However, the influence of carbohydrate moiety of nucleosides
and its modifications on the inhibition of PARP-1 has not been thoroughly
established.
Earlier, we developed approaches for the synthesis of the disaccharide
nucleoside analogues (see[11–13] for details). In this study, we examined these
analogues as potential inhibitors of human PARP-1. In addition, cytotoxic
effects of nucleosides on cultured SKOV-3 human ovarian carcinoma cells
were assessed.
Structure of the substituent at the C5-position of nitrogenous base seems
to be the main factor governing inhibitory efficacy of pyrimidine nucleoside
analogues.[14] As it was expected, disaccharide derivative of Thd (compounds
2e and 4) and 5-iodo-dUrd (2d) inhibit the human recombinant PARP-
1 most efficiently, whereas disaccharide analogues with other heterocyclic
bases were almost not active. Comparison of IC50 values of Thd and com-
pound 2e indicates that attachment of an additional ribose residue at the
3ꢁ-position of Thd molecule increases the inhibitory efficacy. The oxidized
disaccharide nucleoside 4 inhibits PARP-1 more effectively than the parent
2e or Thd. Based on these results, we suppose that dialdehyde group in the
carbohydrate moiety of the nucleoside may participate in the formation of
more stable protein-nucleoside complex, possibly through Schiff bases for-
mation. However, other results obtained in experiments with oxidized Urd
(5) suggest direct interaction of dialdehyde group with the substrate-binding
site of PARP-1. It was found that oxidized nucleoside 5, in contrast to the
parent Urd or dUrd ([14,16], our data), displayed marked PARP-1 inhibitory
efficacy, while dialdehyde derivative of 1-β-D-erythrofuranosyluracil (6) was
inactive. It should be noted that from all of the tested nucleosides, consider-
able positive cooperativity in PARP-1 inhibition was found only for 5. This fact
may indicate that there are additional interactions of 5 with PARP-1, possibly
due to the presence of the dialdehyde group, apart from the direct inter-
1
action with the active center of the enzyme through nitrogenous base. H
NMR spectra of the dialdehyde nucleoside derivatives in water are substan-
tially more complex than those of the starting nucleosides. This is explained
by the fact that dialdehyde derivatives exist in water as hydrates forming
internal hemiacetals and diastereomeric dioxane derivatives,[19] which may
interact with functional groups of enzyme. Natural β-configuration of nucle-
oside seems to be a critical factor for manifestation of the inhibitory proper-
ties of nucleosides. Indeed, α-anomers of the Thd disaccharide derivatives

Disaccharide Nucleosides as Inhibitors of PARP-1
519
(compound 3 and its oxidized form), in contrast to their β-anomers (com-
pounds 2e and 4), were inactive.
Introduction of a phosphate group to the 5ꢁ or 5ꢁꢁ position of disac-
charide nucleoside markedly reduced inhibitory potential of disaccharide
nucleosides that follows from comparison of IC50 values for compounds
2e, 2g, and 2h. Possibly, the presence of negatively charged group in the
carbohydrate moiety of molecule impedes its interaction with active cen-
ter of PARP-1. Previously published data confirm the lack of any signifi-
cant inhibitory activity of thymidine monophosphate and 2ꢁ-deoxyuridine
monophosphate.[14,16]
To assess the efficacy of PARP inhibition in living cells by the compounds
studied, a test system based on the determination of PAR accumulation in
cultured cells after short-term exposure to H2O2 has been used.[23] Com-
pounds 2d, 2e, and 4 were found to be the most effective inhibitors of
recombinant PARP-1 among all tested synthetic nucleosides. Besides, they
suppress H2O2-induced synthesis of PAR in SKOV-3 cells. Differences in the
nucleoside inhibitory efficacy determined with recombinant PARP-1 were
also revealed in cell study. However, PARP-1-inhibitory capacity estimated
on cell culture was lower than that obtained for recombinant protein for
all compounds with exception of Thd. These findings agree with the data
obtained by Rankin et al. in experiments on 5-BrdU.[24] Based on these re-
sults, we claim that the presence of an additional residue in the 3ꢁ position
of Thd derivatives slightly deteriorates PARP-inhibiting effect shown in the
experiments on cultured cells.
The inhibitory efficacy of tested compounds depended on the duration
of cell preincubation. In all cases, except for compound 2e, one-hour prein-
cubation resulted in a significantly stronger suppression of PAR synthesis
than 24-hour preincubation. These differences could be caused by different
permeability, degradation rate, and/or metabolism of these compounds. In
contrast to the results obtained with the recombinant protein, compound
5 did not show substantial inhibitory effect on the cells, probably due to its
high toxicity.
Thus, it is likely that the disaccharide nucleosides examined, despite
their large size and hydrophilicity, are capable of easy penetration through
both plasma and nuclear membranes. It is likely that transfer of these com-
pounds through plasma membrane into the cell is carried out by nucleoside
transporters and is similar to the transport of various antiviral (acyclovir,
zidovudine, etc.) and antineoplastic (cytarabine, cladribine, gemcitabine,
etc.) nucleoside analogues.[25–27]
Incubation of SKOV-3 cells with dAdo or Thd in concentration of 1 mM
significantly suppressed culture growth and provoked cell death. At the same
time, attachment of second ribose residue at the 3ꢁ-position of these nucleo-
sides (compounds 2a and 2e) notably lessened their cytotoxicity. The nature

520
A. S. Efremova et al.
of heterocyclic base had no effect on the toxic properties of the tested
compounds, since different disaccharide nucleosides were not toxic at con-
centrations up to 1 mM.
Alternatively, nucleosides bearing dialdehyde group in the carbohydrate
part of molecule (compounds 4, 5, and 6) strongly suppressed cell growth in
millimolar concentrations. However, their cytotoxicity was not based on the
ability to inhibit PARP-1, as their inhibitory efficacy increases in an order 6
< 5 < 4, whereas cytotoxic activity elevates in the reverse order: 4 < 5 < 6.
In addition, the most potent PARP-1 inhibitor, 3-AB, taken in concentration
of 1 mM, in which it completely inactivates PARP-1, had no effect on culture
growth in comparison with the control.
Pharmacological properties of nucleoside dialdehyde derivatives ob-
tained by periodate oxidation of parent nucleosides were extensively stud-
ied. For example, oxidized derivatives of inosine,[28,29] 5ꢁ-deoxyinosine,[30]
methylpurine,[31] and several other nucleosides[32] have been reported to
display pronounced antitumor effects. Therefore, dialdehyde derivatives of
disaccharide nucleosides are believed to be promising candidates for ap-
plication in antitumor chemotherapy, as they combine properties of both
inhibitors of PARP-1 and cytostatics.
In conclusion, we have described a lot of disaccharide analogues of
pyrimidine nucleosides as prototypes for novel group of PARP-1 inhibitors.
Advantageous properties of these agents are their low cytotoxicity and the
ability to penetrate into cells.
EXPERIMENTAL
Chemistry
The solvents and materials of reagent grade were used without additional
purification. Thin layer chromatography (TLC) was performed on Alu-
gram SIL G/UV254 (Macherey-Nagel) with UV visualization. Column chro-
matography was performed on Kieselgel 0.040–0.063 mm. Reversed-phase
chromatography was performed on chromatograph UVICORD-II (LKB),
equipped with column (diameter 25 mm), containing modified silica gel
˚
(Polygosil C18, pore diameter 60 A, sorbent volume 30 mL). Chromato-
1
graphic separation was controlled by UV-detector. H NMR spectra were
recorded on Bruker AMX 400 NMR instrument at 32◦C. Chemical shifts
in ppm were measured relative to the residual solvent signal as internal
standard; coupling constants J are given in Hertz (Hz). The position of
NH resonances was additionally confirmed by their exchange with the ad-
dition of D2O. The UV spectra were recorded on a Cary 300UV/VIS spec-
trophotometer (Varian). LC-MS analysis was performed on Surveyor MSQ
instrument (Thermo Finnigan, USA), operating in atmospheric pressure
chemical ionization (APCI) mode with detection of positive and negative

Disaccharide Nucleosides as Inhibitors of PARP-1
521
ions and equipped with Onyx Monolithic C18 25 × 4.6 mm Part No CHO-
7645 column. The eluent was a gradient of 0.1% HCOOH aqueous solution
in MeCN. Chromatographic peaks were detected simultaneously with evap-
orative light scattering detector (ELSD), photodiode array detector (PAD),
and total ion current (TIC) detector. ESI-spectroscopy was performed on Q-
Tof micromass instrument (Great Britain) equipped with standard electron-
spray ionization (ESI) interface.
The following compounds were prepared according to the procedures
described elsewhere: 2ꢁ-(O-β-D-ribofuranosyl)-nucleosides (1a–d),[33,34] 3ꢁ-
(O-β-D-ribofuranosyl)-2ꢁ-deoxyadenosine (2a),[35] 3ꢁ-(O-β-D-ribofuranosyl)-
2ꢁ-deoxycytidine
(2b),[36]
3ꢁ-(O-β-D-ribofuranosyl)-2ꢁ-deoxythymidine
(2e),[36] O-β-D-Ribofuranosyl-(1ꢁꢁ-5ꢁ)-O-β-D-Ribofuranosyl-(1ꢁꢁ–3ꢁ)-thymid-
ine (2f),[36] O-β-D-Ribofuranosyl-(1ꢁꢁ–3ꢁ)-thymidine-5ꢁ-O-phosphate (2g),[37]
O-β-D-Ribofuranosyl-(1ꢁꢁ–3ꢁ)-thymidine-5ꢁꢁ-O-phosphate (2h).[37]
General Procedure for Preparation of 3ꢀ-O-modified
Disaccharides 2c and 2d
Compounds 2c and 2d were prepared from corresponding 5ꢁ-O-TBDPS-
deoxyuridines by their glycosylation followed by deprotection of silyl and
benzoyl functionalities according to Scheme 1.
O
X
O
N
NH
O
N
O
N
X
X
X
NH
NH
N
O
NH
t-BuPh2SiO
O
O
O
O
HO
t-BuPh2SiO
HO
O
O
O
iii
ii
i
HO
+
O
O
O
O
OAc
HO
BzO
BzO
BzO
O
O
O
OH
OBz
OBz
HO
BzO
OBz
BzO
BzO
2c: X=F
2d: X=I
SCHEME 1 Synthesis of disaccharide nucleosides. Reagents and conditions: (i) SnCl4, ClCH2CH2Cl,
0◦C, 10 minutes; (ii) Bu4NF/THF, r.t., 45 minutes; (iii) NH3/MeOH, r.t., four days.
1-[5-O-tert-butyldiphenylsilyl-3-O-(2,3,5-tri-O-benzoyl-β-D-ribofuranosyl)-2-
deoxy-β-D-ribofuranosyl]-5-fluorouracil (Step i)
To a cold solution (0◦C) of 1-O-2,3,5-tri-O-benzoyl-β-D-ribofuranose
(484 mg, 0.96 mmol) in 1,2-dichloroethane (12 mL) under nitrogen, tin
tetrachloride (0.14 mL, 1.18 mmol) was added and the solution was kept at
0◦C for 10 minutes. After addition of 1-[5-O-tert-butyldiphenylsilyl-2-deoxy-β-
D-ribofuranosyl]-5-fluorouracil (387 mg, 0.8 mmol), the resulting solution

522
A. S. Efremova et al.
was kept at 0◦C for 40 minutes. Then 10% aqueous solution of sodium bicar-
bonate (5 mL) was added and the suspension was stirred at 0◦C for 20 min-
utes. The suspension was diluted with methylene chloride (20 mL) and fil-
tered through Hyflo Super Cel; the organic layer was separated, washed with
water (10 mL), dried, and evaporated to dryness. The residue was purified by
column chromatography on silica gel (50 g). The column was washed with
methylene chloride (500 mL) and then eluted with 1% ethanol in methylene
chloride to give the resulting product as foam. The yield was 446 mg (60%).
Rf 0.50 (methylene chloride:ethanol, 97:3). 1H NMR (CDCl3): 8.48 br s (1H,
NH Ura5F), 8.10-7.25 m (26H, Ph, Bz, H-6 Ura5F), 6.21 dd (1H, J 1 ,2 a = 6.8,
ꢁ
ꢁ
J 1 ,2 b = 5.8, H-1ꢁ dUrd5F), 5.82 dd (1H, J 3 ,2 = 4.8, J 3 ,4 = 6.4, H-3ꢁ Rib),
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
5.62 d (1H, H-2ꢁ Rib), 5.24 s (1H, H-1ꢁ Rib), 4.76-4.60 m (3H, H-4ꢁ, H-5ꢁa,
H-5ꢁb Rib), 4.55 ddd (1H, J 3 ,2 a = 7.0, J 3 ,2 b = 3.3, J 3 ,4 = 2.7, H-3ꢁ dUrd5F),
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
4.09 ddd (1H, J 4 ,5 a = 3.0, J 4 ,5 b = 2.5, H-4ꢁ dUrd5F), 3.92 dd (1H, J 5 a,5 b
=
=
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
−11.5, H-5ꢁa dUrd5F), 3.78 dd (1H, H-5ꢁb dUrd5F), 2.60 ddd (1H, J 2 a,2 b
ꢁ
ꢁ
−13.5, H-2ꢁb dUrd5F), 2.15 ddd (1H, H-2ꢁa dUrd5F), 1.05 s (9H, t-Bu).
1-[5-O-tert-butyldiphenylsilyl-3-O-(2,3,5-tri-O-benzoyl-β-D-ribofuranosyl)-2-
deoxy-β-D-ribofuranosyl]-5-iodouracil (Step i)
Analogous glycosylation of 1-[5-O-tert-butyldiphenylsilyl-2-deoxy-β-D-
ribofuranosyl]-5-iodouracil (474 mg, 0.8 mmol). The yield was 639 mg (77%)
as foam. Rf 0.48 (methylene chloride:ethanol, 97:3). 1H NMR (CDCl3): 8.30
br s (1Н, NH Ura5I), 8.15-7.25 m (26H, Ph, Bz, H-6 Ura5I), 6.15 dd (1H,
J 1 ,2 a = 8.2, J 1 ,2 b = 5.6, H-1ꢁ dUrd5I), 5.81 dd (1H, J 3 ,2 = 5.0, J 3 ,4 = 6.5,
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
H-3ꢁ Rib), 5.63 d (1Н, H-2ꢁ Rib), 5.26 s (1H, H-1ꢁ Rib), 4.78-4.68 m (2H,
H-4ꢁ, H-5ꢁa Rib), 4.61 dd (1H, J 5 b,5 a = −11.7, J 5 b,4 = 5.1, H-5ꢁb Rib), 4.52
ꢁ
ꢁ
ꢁ
ꢁ
dt (1H, J 3 ,2 a = 6.2, J 3 ,2 b = J 3 ,4 = 2.2, H-3ꢁ dUrd5I), 4.13 ddd (1H, J 4 ,5 a
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
= 3.4, J 4 ,5 b = 3.2, H-4ꢁ dUrd5I), 3.90 dd (1H, J 5 a,5 b = −11.5, J 5 b,4 = 3.2,
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
H-5ꢁa dUrd5I), 3.79 dd (H-5ꢁb dUrd5I), 2.63 ddd (1H, J 2 a,2 b = −13.7, H-2ꢁb
ꢁ
ꢁ
dUrd5I), 2.11 ddd (1H, H-2ꢁa dUrd5I), 1.07 s (9H, t-Bu).
1-[3-O-(2,3,5-tri-O-benzoyl-β-D-ribofuranosyl)-2-deoxy-β-D-ribofuranosyl]-5-
fluorouracil (Step ii)
1-[5-O-tert-butyldiphenylsilyl-3-O-(2,3,5-tri-O-benzoyl-β-D-ribofuranosyl)-
2-deoxy-β-D-ribofuranosyl]-5-fluorouracil (446 mg, 0.48 mmol) was
dissolved in 0.5 M tetrabutylammonium fluoride trihydrate in tetrahydro-
furan (1.3 mL). The solution was kept for 45 minutes at 20◦C, evaporated
to dryness, co-evaporated with chloroform (10 mL), and applied onto a
silica gel column (20 g). The column was washed with methylene chloride
(300 mL) and with 1% ethanol in methylene chloride (200 mL) and then
eluted with 2.5% ethanol in methylene chloride to give the resulting product
as foam. The yield was 308 mg (93%). Rf 0.37 (methylene chloride:ethanol,
97:3). 1H NMR (CDCl3): 8.38 br s (1H, NH Ura5F), 8.20-7.75 m (7H, Bz,

Disaccharide Nucleosides as Inhibitors of PARP-1
523
H-6 Ura5F), 7.70-7.30 m (9H, Bz), 6.06 dd (1H, J 1 ,2 a = 6.6, J 1 ,2 b = 6.2,
ꢁ
ꢁ
ꢁ
ꢁ
H-1ꢁ dUrd5F), 5.81 dd (1H, J 3 ,2 = 4.9, J 3 ,4 = 6.9, H-3ꢁ Rib), 5.65 d (1H,
ꢁ
ꢁ
ꢁ
ꢁ
H-2ꢁ Rib), 5.33 s (1H, H-1ꢁ Rib), 4.80-4.65 m (2H, H-3ꢁ dUrd5F, H-5ꢁa Rib),
4.65-4.50 m (2H, H-4ꢁ dUrd5F, H-5ꢁb Rib), 4.13 ddd (1H, J 4 ,5 a = 3.2, J 4 ,5 b
ꢁ
ꢁ
ꢁ
ꢁ
= 2.4, J 4 ,3 = 2.8, H-4ꢁ dUrd5F), 3.97 dd (1H, J 5 a,5 b = −11.8, H-5ꢁa dUrd5F),
ꢁ
ꢁ
ꢁ
ꢁ
3.85 dd (1H, H-5ꢁb dUrd5F), 2.50 ddd (1H, J 2 b,3 = 3.6, J 2 a,2 b = −13.9,
ꢁ
ꢁ
ꢁ
ꢁ
H-2ꢁb dUrd5F), 2.33 ddd (1H, J 2 a,3 = 6.8, H-2ꢁa dUrd5F).
ꢁ
ꢁ
1-[3-O-(2,3,5-tri-O-benzoyl-β-D-ribofuranosyl)-2-deoxy-β-D-ribofuranosyl]-5-
iodouracil (Step ii)
Analogous desilylation of 1-[5-O-tert-butyldiphenylsilyl-3-O-(2,3,5-
tri-O-benzoyl-β-D-ribofuranosyl)-2-deoxy-β-D-ribofuranosyl]-5-iodouracil
(639 mg, 0.62 mmol). The yield was 409 mg (83%) as foam. Rf 0.34
(methylene chloride:ethanol, 97:3). 1H NMR (CDCl3): 8.37 br s (1H, NH
Ura5I), 8.09 s (1H, H-6 Ura5I), 8.07 dd (2H, J ortho-meta = 8.0, J ortho-para = 1.0,
o-Bz), 8.00 dd (2H, o-Bz), 7.91 dd (2H, o-Bz), 7.61-7.50 m (3H, p-Bz), 7.42
ꢁ
ꢁ
dd (4H, J meta-para = 7.6, m-Bz), 7.35 dd (2H, m-Bz), 6.00 dd (1H, J 1 ,2 a
=
J 1 ,2 b = 6.5, H-1ꢁ dUrd5I), 5.82 dd (1H, J 2 ,3 = 4.8, J 3 ,4 = 6.7, H-3ꢁ Rib), 5.66
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
d (1H, H-2ꢁ Rib), 5.37 s (1H, H-1ꢁ Rib), 4.79–4.71 m (1H, H-4ꢁ Rib), 4.73 dd
(1H, J 5 a,5 b = −13.2, J 5 a,4 = 3.9, H-5ꢁa Rib), 4.63–4.54 m (2H, H-5ꢁb Rib,
ꢁ
ꢁ
ꢁ
ꢁ
H-3ꢁ dUrd5I), 4.13 q (1H, J 4 ,5 a = J 4 ,5 b = J 4 ,3 = 2.6, H-4ꢁ dUrd5I), 3.96 dd
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
(1H, J 5 a,5 b = −12.0, H-5ꢁa dUrd5I), 3.83 dd (H-5ꢁb dUrd5I), 2.53 ddd (1H,
ꢁ
ꢁ
J 2 a,2 b = −14.0, J 2 b,3 = 3.7, H-2ꢁb dUrd5I), 2.39 dt (1H, J 2 a,3 = J 2 a,1 = 6.5,
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
H-2ꢁа dUrd5I).
1-[3-O-(β-D-ribofuranosyl)-2-deoxy-β-D-ribofuranosyl]-5-fluorouracil
(2c, Step iii)
A solution of 1-[3-O-(2,3,5-tri-O-benzoyl-β-D-ribofuranosyl)-2-deoxy-β-D-
ribofuranosyl]-5-fluorouracil (308 mg, 0.44 mmol) in 5 M ammonia in
methanol (9 mL) was kept for four days at 20◦C and then concentrated
in vacuo to dryness. The residue was partitioned between methylene chlo-
ride (10 mL) and water (20 mL), and the aqueous layer was washed with
methylene chloride (2 × 10 mL) and evaporated to dryness; the residue
was purified by reversed-phase chromatography. The column was washed
with water (100 mL) and the product was eluted with 5% ethanol in water.
The fractions, containing the product, were collected and evaporated. The
residue was dissolved in methanol (10 mL), filtered from mechanical impu-
rities, evaporated in vacuum, and then co-evaporated with methanol (2 ×
10 mL) to give white hygroscopic foam. The yield was 118 mg (71%). Rf 0.40
(methylene chloride:ethanol, 80:20). UV (pH 1–7): λmax 268 nm (ꢀ 8900);
(pH 13): λmax 270 nm (6900). ESI: (C14H19FN2O9+Na+). Calc. 401.1000.
1
Found 401.0972. H NMR (CD3OD): 8.21 d (1H, J H-6,F = 6.8, H-6 Ura5F),
6.19 dd (1H, J 1 ,2 a = 7.4 Hz, J 1 ,2 b = 6.0, H-1ꢁ dUrd5F), 4.99 s (1H, H-1ꢁ
ꢁ
ꢁ
ꢁ
ꢁ

524
A. S. Efremova et al.
Rib), 4.42 ddd (1H, J 3 ,2 a = 6.3, J 3 ,2 b = 3.2, J 3 ,4 = 6.0, H-3ꢁ dUrd5F), 4.07
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
dd (1H, J 3 ,2 = 4.7, J 3 ,4 = 6.9, H-3ꢁ Rib), 4.03 dd (1H, J 4 ,5 a = 3.2, H-4ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
dUrd5F), 3.95 dd (1H, J 4 ,5 b = 3.4, H-4ꢁ Rib), 3.90 d (1H, H-2ꢁ Rib), 3.80 dd
ꢁ
ꢁ
(1Н, J 5 a,5 b = −12.0 Hz, H-5ꢁb dUrd5F), 3.72 dd (1Н, J 5 a,5 b = −11.9, H-5ꢁb
ꢁ
ꢁ
ꢁ
ꢁ
Rib), 3.67–3.53 m (2H, H-5ꢁa dUrd5F, H-5ꢁa Rib), 2.53 ddd (1H, J 2 b,2 a
=
ꢁ
ꢁ
−13.7, H-2ꢁb dUrd5F), 2.19 ddd (1H, H-2ꢁa dUrd5F). 13C NMR (DMSO-d6):
149.26 (C-4), 141.62 (C-2), 138.56 (C-6), 124.54 d (J C-5,F = 136, C-5), 106.40
(C-1ꢁ Rib), 84.85 (C-4ꢁ dUrd5F), 84.54 (C-1ꢁ dUrd5F), 83.67 (C-4ꢁ Rib), 76.55
(C-3ꢁ dUrd5F), 74.46 (C-2ꢁ Rib), 70.65 (C-3ꢁ Rib), 62.80 (C-5ꢁ Rib), 61.03 (C-5ꢁ
dUrd5F), 37.91 (C-2ꢁ dUrd5F).
1-[3-O-(β-D-ribofuranosyl)-2-deoxy-β-D-ribofuranosyl]-5-iodouracil
(2d, Step iii)
Analogous debenzoylation of 1-[3-O-(2,3,5-tri-O-benzoyl-β-D-ribo-
furanosyl)-2-deoxy-β-D-ribofuranosyl]-5-iodouracil (409 mg, 0,51 mmol).
The yield was 178 mg (72%) as white hygroscopic foam. Rf 0.38 (methylene
chloride:ethanol, 80:20). UV (pH 1–7): λmax 287 nm (ꢀ 8300); (pH 13):
λmax 280 nm (ꢀ 6400). LC-MS (APCI): retention time (RT), 1.25 minutes;
(C14H19IN2O9-H++HCOOH). Calc. 531.01. Found 530.54. 1H NMR (D2O):
8.32 s (1H, H-6 Ura5I), 6.28 t (1H, J 1 ,2 a = J 1 ,2 b = 6.7, H-1ꢁ dUrd5I), 5.19 s
ꢁ
ꢁ
ꢁ
ꢁ
(1H, H-1ꢁ Rib), 4.53 dt (1H, J 3 ,2 b = J 3 ,4 = 4.0, J 3 ,2 a = 6.5, H-3ꢁ dUrd5I),
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
4.30 dd (1H, J 3 ,2 = 4.8, J 3 ,4 = 6.8 Hz, H-3ꢁ Rib), 4.23 dt (1H, J 4 ,3 = J 4 ,5 b
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
= 4.0, J 4 ,5 a = 4.7, H-4ꢁ dUrd5I), 4.18 d (1H, H-2ꢁ Rib), 4.12 td (1H, J 4 ,3
ꢁ
ꢁ
ꢁ
ꢁ
= J 4 ,5 b = 6.8, J 4 ,5 a = 3.4, H-4ꢁ Rib), 3.96-3.91 m (2H, H-5ꢁa dUrd5I, H-5ꢁa
ꢁ
ꢁ
ꢁ
ꢁ
Rib), 3.86 dd (1H, J 5 b,5 a = −12.4, H-5ꢁb dUrd5I), 3.75 dd (1H, J 5 b,5 a
=
ꢁ
ꢁ
ꢁ
ꢁ
−12.3, H-5ꢁb Rib), 2.67 ddd (1H, J 2 b,2 a = −14.7, H-2ꢁb dUrd5I), 2.45 dt
(1H, H-2ꢁa dUrd5I). 13C NMR (D2O): 172.28 (C-4), 159.14 (C-2), 148.22
(C-6), 130.00 (C-5), 109.32 (C-1ꢁ Rib), 88.56 (C-4ꢁ dUrd5I), 87.36 (C-1ꢁ
dUrd5I), 85.64 (C-4ꢁ Rib), 80.03 (C-3ꢁ dUrd5I), 77.26 (C-2ꢁ Rib), 73.47 (C-3ꢁ
Rib), 65.54 (C-5ꢁ Rib), 63.83 (C-5ꢁ dUrd5I), 40.54 (C-2ꢁ dUrd5I).
ꢁ
ꢁ
Dialdehyde Derivative 4
To 0.3 mL of aqueous solution of compound 2e (112 mg, 0.3 mmol),
0.7 mL of 0.5 M sodium periodate was added dropwise at 20◦C. The reaction
mixture was kept for 20 minutes at 20◦C and then diluted with ethanol
(10 mL). The precipitate of NaIO3 was filtered-off and washed with ethanol
(10 mL). The filtrates were evaporated in vacuum to dryness. The residue was
dissolved in methanol (2 mL), filtered from traces of NaIO3, and evaporated
in vacuum to dryness. The residue was co-evaporated with methanol (2 ×
5 mL) and dried over P2O5 in vacuum dessicator during several days. The
yield was 110 mg (98%) as white hygroscopic foam. Rf 0.65 (extending spot)
in methylene chloride:ethanol, 80:20.

Disaccharide Nucleosides as Inhibitors of PARP-1
Dialdehyde Derivative 5
525
To 1.0 mL of aqueous solution of Urd (244 mg, 1.0 mmol), 2.4 mL of
0.5 M sodium periodate was added dropwise at 20◦C. The reaction mixture
was kept for 20 minutes at 20◦C and then diluted with ethanol (30 mL).
The precipitate of NaIO3 was filtered-off and washed with ethanol (10 mL)
and the filtrate was evaporated in vacuum to dryness. The residue was dis-
solved in methanol (7 mL), filtered from inorganic salts, and the filtrate
was evaporated in vacuum to dryness. The residue was co-evaporated with
methanol (2 × 10 mL) and dried over P2O5 during several days. The yield
was 235 mg (97%) as white hygroscopic foam. Rf 0.71 (extending spot) in
1
methylene chloride:ethanol, 80:20. H NMR Spectrum in D2O is very com-
plex due to the presence of hydrated species of dialdehyde that may form
diastereomeric internal hemiacetals and dioxane derivatives and it is very
similar to the published one.[18]
Dialdehyde Derivative 6
Analogous oxidation of 1-(β-D-erythrofuranosyl)uracil[38] (65 mg,
0.3 mmol) with sodium periodate yielded 62 mg (96%) of compound 6.
1
Rf 0.76 (extending spot) in methylene chloride:ethanol, 80:20. H NMR
(D2O): 7.70 d (1H, J 6,5 = 7.8, H-6 Ura), 5.86 d (1H, H-5 Ura), 5.56 d (1H,
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
J 2 ,1 = 4.0, H-1 oxidized erythrose), 5.15-5.00 m (2H, H-2 , H-4 oxidized ery-
throse), 3.57 dd (1H, J 5 a,5 b = −10.6, J 5 a,4 = 4.4, H-5ꢁa oxidized erythrose),
ꢁ
ꢁ
ꢁ
ꢁ
3.50 dd (1H, J 5 a,5 b = −10.6, J 5 b,4 = 4.7, H-5ꢁb oxidized erythrose).
ꢁ
ꢁ
ꢁ
ꢁ
PARP-1 Activity Assay
Purified human recombinant PARP-1 and [3H] NAD+ were prepared as
described elsewhere.[39–41] The PARP-1 inhibiting activities of nucleosides
were assayed as described previously.[42] Briefly, reaction mixtures (15 μL)
containing 2 optic units (A260)/mL DNAse I-activated calf thymus DNA,
300 μM NAD+, 0.4 μCi [3H] NAD+, 10% DMSO, and tested compounds in
the various concentrations if required in the buffer (50 mM Tris, pH 8.0,
20 mM MgCl2, 150 mM NaCl, 7 mM β-mercaptoethanol) were preincubated
for five minutes at 37◦C, and then recombinant human PARP-1 was added
to the final concentration of 200 nM. After one minute, the reaction was
stopped by dropping the 12 μL aliquote on the Whatman 1 paper filters
moistened in 5% trichloroacetic acid. Filters were washed four times with
150 mL of 5% trichloroacetic acid, once with 90% ethanol, and then air-
dried. [3H] ADP-ribose incorporation into the acid insoluble material was
quantified using a scintillation counter Tri-Carb 2800 (Perkin Elmer) in the
toluene scintillator.

526
A. S. Efremova et al.
Cell Culture
SKOV-3 human ovarian carcinoma cell line was kindly provided by Dr.
Galina A. Posypanova (Moscow Research Institute of Medical Ecology, Rus-
sia). The cell line was cultured in RPMI 1640 (BioloT), containing 10%
fetal bovine serum (PAA Laboratories), 2 mM L-glutamine, and 50 μg/mL
gentamicin at 37◦C in a humidified atmosphere with 95% air and 5% CO2.
Cytotoxicity Analysis
Cytotoxicity of all tested compounds was determined by the change in
the proliferative status of cultured cells using MTT test. Exponentially grow-
ing SKOV-3 cells were seeded in 96-well plates at a density of 2.5 × 103
cells/well (200 μL per well); 24 hours later, the number of attached cells in
premarked fields were counted and nucleosides or 3-AB were added to the
growth medium. The cells were cultured for 72 hours and then the medium
in each well was replaced with the medium containing 0.5 mg/mL 3-[4,5-
dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT). After 3-hour
incubation at 37◦C, the number of the living (i.e., accumulating formazan)
cells was counted in each premarked field. Finally, the density of living cells
in the selected fields after 72-hour incubation with the tested compound
was calculated and expressed as a percentage of live cells relative to their
original number 24 hours after seeding.
Determination of Poly(ADP-Ribose) in Cultured Cells
Exponentially growing SKOV-3 cells were seeded in 4-well plates on cov-
erslips at a density of 30 × 103 cells/well in 1 mL growing medium. After
48 hours, nucleosides or 3-AB were added and the cultures were further
incubated either for one or 24 hours. Then the cells were treated with hy-
drogen peroxide (1 mM, 5 minutes) in the presence of tested compounds
and fixed by methanol. The fixed cells were incubated with mouse mono-
clonal antibody to PAR (10H; Santa Cruz Biotechnology) at 1:300 dilution,
followed by incubation with goat antimouse antibody conjugated with Alexa-
488 (Molecular Probes) at 1:1000 dilution. Nuclei were counterstained with
propidium iodide (PI) and photomicrographs were obtained using confocal
microscope LSM-510 (Carl Zeiss). The photomicrographs were analyzed by
AxioVision 4.8 (Carl Zeiss) and ImageJ 1.42q (National Institutes of Health,
USA) programs.
Data Analysis
All data were analyzed using the Sigma Plot 11.0 (Systat Software Inc.) or
GraphPad Prism 5.00 (GraphPad Software Inc.) software. The IC50 and nH

Disaccharide Nucleosides as Inhibitors of PARP-1
527
values were calculated using the four parameter logistic fit equation. Statis-
tical analysis of experimental data was performed by the Student’s unpaired
t-test. The difference between the values was considered significant when
p < 0.05.
REFERENCES
1. Parker, W.B. Enzymology of purine and pyrimidine antimetabolites used in the treatment of cancer.
Chem. Rev. 2009, 109, 2880–2893.
2. De Clercq, E. Highlights in antiviral drug research: antivirals at the horizon. Med. Res. Rev. 2012,
DOI:10.1002/med (Epub), 1–34.
3. Ferraris, D.V. Evolution of poly(ADP-ribose) polymerase-1 (PARP-1) inhibitors: from concept to
clinic. J. Med. Chem. 2010, 53, 4561–4584.
4. Sandhu, S.K.; Yap, T.A.; de Bono, J.S. Poly(ADP-ribose) polymerase inhibitors in cancer treatment:
a clinical perspective. Eur. J. Cancer 2010, 46, 9–20.
5. Sukhanova, M.V.; Lavrik, O.I.; Khodyreva, S.N. Poly(ADP-ribose) polymerase-1 as a regulator of
protein-nucleic acid interactions in the processes responding to genotoxic action. Mol. Biol. (Mosk).
2004, 38, 834–847.
6. Luo, X.; Kraus, W.L. On PAR with PARP: cellular stress signaling through poly(ADP-ribose) and
PARP-1. Genes Dev. 2012, 26, 417–432.
7. D’Amours, D.; Desnoyers, S.; D’Silva, I.; Poirier, G.G. Poly(ADP-ribosyl)ation reactions in the regu-
lation of nuclear functions. Biochem. J. 1999, 342, 249–268.
8. Hassa, P.O.; Haenni, S.S.; Elser, M.; Hottiger, M.O. Nuclear ADP-ribosylation reactions in mammalian
cells: where are we today and where are we going? Microbiol. Mol. Biol. Rev. 2006, 70, 789–829.
9. Mikhailov, S.N.; Kulikova, I.V.; Nauwelaerts, K.; Herdewijn, P. Synthesis of 2ꢁ-O-α-D-ribo-
furanosyladenosine, monomeric unit of poly(ADP-ribose). Tetrahedron 2008, 64, 2871–2876.
10. Van der Heden van Noort, G.J.; Overkleeft, H.S.; van der Marel, G.A.; Filippov, D.V. Ribosylation of
adenosine: an orthogonally protected building block for the synthesis of ADP-ribosyl oligomers. Org.
Lett. 2011, 13, 2920–2923.
11. Efimtseva, E.V.; Mikhailov, S.N. Disaccharide nucleosides. Russ. Chem. Rev. 2004, 73, 401–414.
12. Efimtseva, E.V.; Kulikova, I.V.; Mikhailov, S.N. Disaccharide nucleosides and their incorporation into
oligonucleotides. Curr. Org. Chem. 2007, 11, 337–354.
13. Efimtseva, E.V.; Kulikova, I.V.; Mikhailov, S.N. Disaccharide nucleosides as an important group of
natural compounds. Mol. Biol. (Rus). 2009, 43, 301–312.
14. Pivazyan, A.D.; Birks, E.M.; Wood, T.G.; Lin, T.-S.; Prusoff, W.H. Inhibition of poly(ADP-
ribose)polymerase activity by nucleoside analogs of thymidine. Biochem. Pharmacol. 1992, 44, 947–953.
15. Banasik, M.; Ueda, K. Inhibitors and activators of ADP-ribosylation reactions. Mol. Cell Biochem. 1994,
138, 185–197.
16. Preiss, J.; Schlaeger, R.; Hilz, H. Specific inhibition of poly ADPribose polymerase by thymidine and
nicotinamide in HeLa cells. FEBS Lett. 1971, 19, 244–246.
17. Steinhagen, H.; Gerisch, M.; Mittendorf, J.; Schlemmer, K.-H.; Albrecht, B. Substituted uracil deriva-
tives as potent inhibitors of poly(ADP-ribose)polymerase-1 (PARP-1). Bioorg. Med. Chem. Lett. 2002,
12, 3187–3190.
18. Howarth, O.; Jones, A.S.; Walker, R.T.; Wyatt, P.G. Solution structures of some uridine dialdehyde
derivatives. J. Chem. Soc. Perkin Trans. 1984, 261–265.
19. Mikhailov, S.N.; Florentiev, V.L.; Pfleiderer, W. Convenient synthesis of partially blocked oxidized-
reduced nucleosides. Synthesis 1985, 399–400.
20. Mikhailov, S.N.; Yakovlev, G.I. The use of periodate oxidation reaction in combination with PMR
spectroscopy for the determination of nucleoside, monosaccharide and their analogs structure. Khim.
Prirodnykh Soedinen. (USSR). 1987, 40–43.
21. Ermolinsky, B.S.; Mikhailov, S.N. Periodate oxidation in the chemistry of nucleic acids: dialdehyde
derivatives of nucleosides, nucleotides, and oligonucleotides. Russ. J. Bioorg. Chem. 2000, 26, 483–504.
22. Banasik, M.; Komura, H.; Shimoyama, M.; Ueda, K. Specific inhibitors of poly(ADP-ribose) synthetase
and mono(ADP-ribosyl)transferase. J. Biol. Chem. 1992, 267, 1569–1575.

528
A. S. Efremova et al.
23. Ryabokon, N.I.; Cie´slar-Pobuda, A.; Rzeszowska-Wolny, J. Inhibition of poly(ADP-ribose) polymerase
activity affects its subcellular localization and DNA strand break rejoining. Acta Biochim. Pol. 2009,
56, 243–248.
24. Rankin, P.W.; Jacobson, E.L.; Benjamin, R.C.; Moss, J.; Jacobson, M.K. Quantitative studies of in-
hibitors of ADP-ribosylation in vitro and in vivo. J. Biol. Chem. 1989, 264, 4312–4317.
25. King, A.E.; Ackley, M.A.; Cass, C.E.; Young, J.D.; Baldwin, S.A. Nucleoside transporters: from scav-
engers to novel therapeutic targets. Trends Pharmacol. Sci. 2006, 27, 416–425.
26. Molina-Arcas, M.; Casado, F.J.; Pastor-Anglada, M. Nucleoside transporter proteins. Curr. Vasc. Phar-
macol. 2009, 7, 426–434.
27. Koczor, C.A.; Torres, R.A.; Lewis, W. The role of transporters in the toxicity of nucleoside and
nucleotide analogs. Expert Opin. Drug Metab. Toxicol. 2012, 8, 665–676.
28. Kaufman, J.; Mittelman, A. Clinical phase I trial of inosine dialdehyde (NSC-118994). Cancer
Chemother. Rep. 1975, 59, 1007–1014.
29. Plagemann, P.G.; Graff, J.C.; Behrens, M. Mechanism of action of inosine dialdehyde (NSC 118994)
in the inhibition of proliferation of tumor cells in culture. Cancer Res. 1977, 37, 2188–2195.
30. Cory, J.G.; Parker, S.H. Dialdehyde derivative of 5ꢁ-deoxyinosine as a more potent analog of the
dialdehyde derivative of inosine (NSC 118994). Biochem. Pharmacol. 1979, 28, 867–871.
31. Bell, J.P.; Faures, M.L.; LePage, G.A.; Kimball, A.P. Immunosuppressive and antitumor activity of
periodate oxidation product of beta-D-ribosyl-6-methylthiopurine. Cancer Res. 1968, 28, 782–787.
32. Mirkin, B.L.; O’Dea, R.F.; Hogenkamp, H.P. Cytotoxic action of adenosine nucleoside and dialde-
hyde analogues on murine neuroblastoma in tissue culture: structure-activity relationships. Cancer
Res. 1987, 47, 3650–3655.
33. Mikhailov, S.N.; DeBruyn, A.; Herdewijn, P. Synthesis and properties of some 2ꢁ-O-β−D-
ribofuranosylnucleosides. Nucleosides Nucleotides 1995, 14, 481–484.
34. Mikhailov, S.N.; Efimtseva, E.V.; Gurskaya, G.V.; Zavodnik, V.E.; De Bruyn, A.; Rozenski, J.; Herdewijn,
P. An efficient synthesis and physico-chemical properties of 2ꢁ-O-β-D-ribofuranosyl-nucleosides, mi-
nor tRNA components. J. Carbohydr. Chem. 1997, 16, 75–92.
35. Gulyaeva, I.V.; Neuvonen, K.; Lonnberg, H.; Rodionov, A.A.; Shcheveleva, E.V.; Bobkov, G.V.; Efimt-
seva, E.V.; Mikhailov, S.N. Effective anomerisation of 2ꢁ-deoxyadenosine derivatives during disaccha-
ride nucleoside synthesis. Nucleosides Nucleotides Nucleic Acids 2004, 23, 1849–1864.
36. Mikhailov, S.N.; De Clercq, E.; Herdewijn, P. Ribosylation of pyrimidine 2ꢁ-deoxynucleosides. Nucle-
osides Nucleotides 1996, 15, 1323–1334.
37. Efimtseva, E.V.; Shelkunova, A.A.; Mikhailov, S.N.; Nauwelaerts, K.; Rozenski, J.; Lescrinier, E.;
Herdewijn, P. Synthesis and properties of phosphorylated 3ꢁ-O-β-D-ribofuranosyl-2ꢁ-deoxythymidine.
Nucleosides Nucleotides Nucleic Acids 2003, 22, 359–371.
38. Zavgorodniy, S.G.; Efimtseva, E.V.; Mikhailov, S.N.; Tsilevitch, T.L.; Yavorsky, A.E.; Florentiev, V.L.
Acyclic analogues of nucleosides. Synthesis of 1,5-dihydroxy-3-oxapent-2-yl derivatives of nucleic
bases. Khim. Geterotsikl. Soedinen. (USSR). 1988, 223–228.
39. Sukhanova, M.V.; Khodyreva, S.N.; Lavrik, O.I. Poly(ADP-ribose) polymerase-1 inhibits strand-
displacement synthesis of DNA catalyzed by DNA polymerase beta. Biochemistry (Mosc.) 2004, 69,
558–568.
40. Shram, S.I.; Rybakova, I.G.; Lazurkina, T.Iu.; Sidorov, G.V.; Shilovsk˘ı, G.A.; Khokhlov, A.N.; Miasoe-
dov, N.F. Enzymatic synthesis of beta-NAD+ selectively marked with tritium at adenine and its use for
determining the activity of poly(ADP-ribose) polymerase. Prikl. Biokhim. Mikrobiol. 1999, 35, 638–646.
41. Sidorov, G.V.; Zverkov, Yu.B.; Shram, S.I.; Lazurkina, T.Yu.; Myasoedov, N.F. Chemical and enzymatic
synthesis of tritium labelled coenzymes. J. Labelled Comp. Radiopharm. 2003, 46, 465–473.
42. Zeifman, A.A.; Stroylov, V.S.; Novikov, F.N.; Stroganov, O.V.; Zakharenko, A.L.; Khodyreva, S.N.;
Lavrik, O.I.; Chilov, G.G. Hit clustering can improve virtual fragment screening: CDK2 and PARP1
case studies. J. Mol. Model. 2012, 18, 2553–2566.