New cytosine derivatives as inhibitors of DNA methylation

Article abstract of DOI:10.1016/j.ejmech.2012.07.024

DNA cytosine methylation catalyzed by DNA methyltransferase 1 (DNMT1) is an epigenetic method of gene expression regulation and development. Changes in methylation pattern lead to carcinogenesis. Inhibition of DNMT1 activity could be a good strategy of safe and efficient epigenetic therapy. In this work, we present a novel group of cytosine analogs as inhibitors of DNA methylation. We show new methods of synthesis and their effect on in vitro reaction of DNA methylation. Almost all of analyzed compounds inhibit DNA methyltransferase activity in the competitive manner. K_i values for the most potent compound 4-N-furfuryl-5,6-dihydroazacytosines is 0.7 μM. These compounds cause also a decrease of 5-methylcytosine (m⁵C) level in DNA of mammalian HeLa and HEK293 cells.

Full text of DOI:10.1016/j.ejmech.2012.07.024

European Journal of Medicinal Chemistry 55 (2012) 243e254
Contents lists available at SciVerse ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
New cytosine derivatives as inhibitors of DNA methylation
Beata Plitta 1, Ewelina Adamska 1, Ma1gorzata Giel-Pietraszuk,
Agnieszka Fedoruk-Wyszomirska, Miros1awa Naskre˛t-Barciszewska,
*
*
Wojciech T. Markiewicz , Jan Barciszewski
ꢀ
Institute of Bioorganic Chemistry, Polish Academy of Sciences, Z. Noskowskiego 12/14, 61-704 Poznan, Poland
a r t i c l e i n f o
a b s t r a c t
Article history:
DNA cytosine methylation catalyzed by DNA methyltransferase 1 (DNMT1) is an epigenetic method of
gene expression regulation and development. Changes in methylation pattern lead to carcinogenesis.
Inhibition of DNMT1 activity could be a good strategy of safe and efficient epigenetic therapy. In this
work, we present a novel group of cytosine analogs as inhibitors of DNA methylation. We show new
methods of synthesis and their effect on in vitro reaction of DNA methylation. Almost all of analyzed
compounds inhibit DNA methyltransferase activity in the competitive manner. Ki values for the most
Received 24 January 2012
Received in revised form
11 July 2012
Accepted 16 July 2012
Available online 24 July 2012
potent compound 4-N-furfuryl-5,6-dihydroazacytosines is 0.7
a decrease of 5-methylcytosine (m5C) level in DNA of mammalian HeLa and HEK293 cells.
Ó 2012 Elsevier Masson SAS. All rights reserved.
mM. These compounds cause also
Keywords:
Inhibitors of DNA methyltransferase 1
Cytosine derivatives
Cancer therapy
Furfural
4-N-Furfurylcytosine
1. Introduction
Until now several DNMT1 inhibitors have been identified. They
can be divided into four groups. The first one includes cytidine
analogs like 5-azacytidine, 5-aza-2-deoxycytidine and zebularine
[6,9]. After phosphorylation they are incorporated into newly
synthesized DNA at cytosine sites [10]. The lack of the proton at
position 5 of these cytosine derivatives stabilizes the covalent bond
between DNA and enzyme (Fig. 1). Thus, the trapped DNMT1
molecules are excluded from the pool of active enzymes and
formation of these covalent proteineDNA complexes is considered
as a reason of a high toxicity of those inhibitors [6]. This leads to
a decrease in the level of methylation of replicated DNA [9]. It has
been suggested that 5-azacytidine affects also other cellular
processes via altering expression of genes which promoters were
not regulated by DNA methylation prior to the treatment [11]. The
another group of DNMT1 inhibitors includes small non-nucleoside
natural compounds, like EGCG (epigallocatechin-3-gallate) and
curcumin however, it has been suggested that the observed inhi-
bition of DNMT1 might be effected by free radicals formed during
degradation of EGCG rather than its direct interaction with enzyme
[12e14]. Another compound of that group, phthalimide RG108
functions as a competitive inhibitor and binds directly to DNMT1
active site [15]. A broad application of RG108 is however limited
due to its high hydrophobicity [16]. The third group of DNMT1
inhibitors consists of SAM (S-adenosyl-L-methionine) analogs
lacking the methyl group [17]. One of them, sinefungin, increases
the rate of m5C deamination to thymine [18]. Finally, the fourth
The expression of the genetic information is regulated through
epigenetic mechanisms such as methylation of cytosine in DNA,
modifications of histones and noncoding RNAs. Methylation of
cytosine residues at C-5 position is catalyzed by DNA methyl-
transferases [1]. In eukaryotic DNA about 3e8% of all cytosine
residues are methylated. This modification occurs predominately at
50-CpG-30 dinucleotides that are methylated up to ca. 70%. Aber-
rations in specific DNA methylation pattern are among the main
reasons of cancer development [2e5]. The crucial enzyme in this
process is DNA methyltransferase 1 (DNMT1) that maintains an
altered methylation pattern by copying it from a parent to
a daughter DNA strand after replication [3]. Correlation of hyper-
methylation of promoter regions with transcriptional silencing of
tumor suppressor genes (TSG) in carcinogenesis is very well
established [6]. Inactivation of DNMT1 may lead to a passive
demethylation of promoter and in a consequence might restore an
expression of TSG and inhibit tumorogenesis [2,7,8]. Thus, down
regulation of DNMT1 seems to be a promising strategy in the anti-
cancer therapy.
* Corresponding authors. Tel.: þ48 61 852 85 03; fax: þ48 61 852 05 32.
E-mail addresses: markwt@ibch.poznan.pl (W.T. Markiewicz), Jan.Barciszewski@
ibch.poznan.pl (J. Barciszewski).
1
Both authors contributed equally to this manuscript.
0223-5234/$ e see front matter Ó 2012 Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.ejmech.2012.07.024

244
B. Plitta et al. / European Journal of Medicinal Chemistry 55 (2012) 243e254
Fig. 1. Mechanism of reaction catalyzed by cytosine C5 DNA methyltransferase. The target cytosine in DNA interacts with active site residues (I) of DNMT1 to facilitate cysteine
nucleophilic attack at the C6 position. Nucleophilic attack disrupts the pyrimidine’s aromaticity, generating the reactive covalent adduct (II) which readily undergo electrophilic
addition of methyl group through methylation (III). Finally deprotonation of position C5 leads to restoration of CeC double bond due to
b-elimination (IV). Lack of proton at position
5 of 5-aza-dC (a) as well as the presence of a poor leaving group in 5-F-dC (b) or lack of amino group at C4 in zebularine (c) prevent the restoration of double bond.
group of inhibitors consists of procaine and its derivative procai-
namide. Through binding to CpG-rich sequences these compounds
induce a spatial hindrance preventing DNA methylation [17,19e21].
The persistent shortage of effective anticancer drugs makes
searching for new inhibitors of DNMT1 a widely recognized chal-
lenge. The aim of our study was to find a new, effective and non-toxic
inhibitors of the enzyme, which could be used in anti-cancer therapy.
We focused our interest on the new analog of cytosine, 4-N-
furfurylcytosine (FC), that has been found by us in DNA [manuscript
in preparation]. It is a product of cellular oxidation of DNA in which
furfural moiety resulted from hydroxylation at the carbon 50 of
deoxyribose units of DNA reacts with N-4 of cytosine residues
nearby [22e24]. We considered FC as a leading compound for
searching inhibitors of DNA methyltransferase. Here we show that
4-N-furfurylcytosine and its derivatives inhibit reaction of DNA
methylation. In addition to that, for all compounds either new
methods of synthesis or modifications of earlier published proce-
dures are presented.
advantage of plethora of synthetic procedures that lead to nucleo-
side modifications. After an acid-catalyzed hydrolysis of N-nucleo-
sidic bond the following nucleobases: 2, 4, 29, 30 (Schemes 1 and 5)
were obtained [25]. Synthesis of cytosine derivatives starting from 4-
N-p-toluenesulfonyl-20-deoxycytidine (1) was carried out according
to procedure described earlier [26]. The reaction of 1 with a series of
primary alkyl amines carried out at higher temperature (70 ꢀC) in
pyridine gave expected cytidine derivatives in very high yield
(Scheme 1). 4-N-substituted cytosines: 2, 4 (Scheme 1) were ob-
tained by the cleavage of N-glycosidic bond in 1 or 3 (Scheme 1) with
hydrochloric acid in refluxing methanol [25].
The second approach was based on reaction of cytosine with
aldehydes leading to formation of Schiff base derivatives followed
by their reduction with sodium borohydride in methanol (Scheme
2). This method had been already described [27,28] but we have
used other conditions of this reaction. With this approach we
received many compounds in short time without protecting of
cytosine at N-1 position step. The reaction of cytosine with
‘aromatic’ aldehydes was found to proceed smoothly and quanti-
tatively in anhydrous methanol in the presence of magnesium
methanolate as water ‘consumer’ (4e9, 24, 25) (Schemes 2 and 4).
In the case of alkyl aldehyde as a substrate, the reaction was found
to be efficiently catalyzed by acetic acid (10e12) (Scheme 2). The
use of basic conditions (methanolate) for alkyl aldehydes led to
condensation products. The Schiff bases were reduced either with
2. Chemistry
A series of cytosine derivatives was synthesized using two
approaches. In the first one a nucleoside was used as a substrate in
which the sugar residue is treated as a “protecting group” of
nucleobase (1, 3, 27, 28) (Schemes 1 and 5). Thus, one can take an

B. Plitta et al. / European Journal of Medicinal Chemistry 55 (2012) 243e254
245
R
HN
N
N
R
N
HN
O
a
O
S
O
b
HO
H3C
Ts =
O
N
H
O
O
CH2
OH
R= Ts
3 R= Fur
Fur =
R= Ts
2
4 R= Fur
1
Scheme 1. Synthesis of cytosine derivatives via acid-catalyzed hydrolysis N-glycosidic bond of nucleosides. Reagents and reaction conditions: (a) furfurylamine (5 equiv), pyridine,
70 ꢀC, 12 h; (b) HCl (2.4 equiv) in MeOH aq., reflux, 4 h.
sodium borohydride or borane dimethyl sulfide complex to corre-
sponding products: 4e9, 13, 14, 24, 25 (Schemes 2 and 4).
Modification of 4-N-furfurylcytosine (4) and 4-N-benzylcyto-
sine (7) at C-5 position with paraformaldehyde done in the pres-
ence of triethylamine led to synthesis of 5-hydroxymethyl-4-N-
furfurylcytosine (15) in high yield and 5-hydroxymethyl-4-N-ben-
zylcytosine (16) in lower yield (Scheme 3) [29]. These compounds
were used furthermore as substrates in synthesis of a few other
new compounds (17e23).
Acetylation of 15 and 16 with acetic anhydride gave
5-acetyloxymethyl derivatives 17 and 18 respectively (Scheme 5). The
reaction of 15 and 16 with sodium metabisulfite in water led to
formation of 5-methyl sulphonic acid derivatives 19 and 20 respec-
tively in good yields (Scheme 3).
followed by reduction with sodium borohydride gave 5-(N-buty-
loamine)methyl-4-N-furfurylcytosine (23).
On the other hand, 5-hydroxymethyl-4-N-furfurylcytosine (15)
was transformed into
a methanesulfonyl ester with meth-
anesulfonyl chloride in pyridine and subsequently reacted with
amine to obtain 23 in high yield (Scheme 3).
Another approach was applied for synthesis of 4-N-furfuryl-5-
methylcytosine (29) and 4-N-benzyl-5-methylcytosine (30).
5-Methylcytosine and its nucleosides are rather expensive although
commercially available. Thus, we have taken advantage of using 30,50-
di-O-acetylthymidine and transforming it into 5-methyl-4-(1,2,4-
triazol-1-yl)-1-(b-D-3,5-di-O-acetyl-2-deoxyribofuranosyl)pyr-
imidin-2(1H)-one (26) [30e32]. Reaction of triazole nucleoside 26
and furfurylamine or benzylamine resulted in 5-methyl-4-N-fur-
furyl-20-deoxycytidine (27) and appropriate 5-methyl-4-N-benzyl-
20-deoxycytidine (28) and these could be easily transformed into
desired 5-methyl-4-N-furfurylcytosine (29) and 5-methyl-4-N-
benzylcytosine (30) after cleavage of N-nucleoside bond (Scheme 5)
[25].
The 5-hydroxymethyl group of 15 and 16 could be oxidized with
manganese dioxide in water giving appropriate 5-formyl deriva-
tives 21 and 22 (Scheme 3) in moderate yields. The derived alde-
hydes could be then used to obtain Schiff bases in reaction with
primary amines. Thus, reaction of 21 with n-butylamine (Scheme 3)
CH2R
CHR
3. Results
NH2
N
HN
N
c or d
3.1. Inhibitory activity of cytosine analogs in in vitro DNA
methylation assay
a or b
N
O
N
O
N
H
O
N
H
N
H
Inhibitory activity of each compound was analyzed in in vitro
DNA methylation reaction catalyzed by DNA methyltransferase
from Spiroplazma. Sequence homology between human DNMT1
(Gen Bank Acc no X63692) and M.SssI (X17195.1) is low. However
Reaction
conditions
R
O
O
R1
N
R1
N
HO
R1
Fur
Bnz
Compound
15, 17, 19, 21
HN
HN
R2
4
5
6
NO2
a
a, c
16, 18, 20, 22, 23
N
H
O
N
H
O
H3C
R2 = HOCH2
N
CH2
15
b
8
9
7
16 R2 = HOCH2
Bnz =
c
R2 = CH3CO2CH2
R2 = CH3CO2CH2
17
18
19
20
b
CH3- CH3CH2- CH3CH2CH2-
10 11 12
e
c
b
e
R2 = HSO2CH2
R2 = HSO2CH2
d
21 R2 = CHO
22
23
R2 = CHO
R2 = CH3(CH2)3NHCH2
CH3-
13
CH3CH2CH2-
14
b, d
f
Scheme 3. Synthesis of 4-N-furfurylcytosine and 4-N-benzylcytosine derivatives with
modification at 5-C position. Reagents and reaction conditions: (a) CH2O (2.3 equiv),
Et3N (10 equiv), H2O, 65 ꢀC, 12 h; (b) Ac2O (1.2 equiv), Py, rt, 3 h; (c) sodium meta-
bisulfite (2 equiv), water, 55 ꢀC, 24 h; (d) 1. Py, MsCl (1.5 equiv), rt, 0.5 h; 2. n-butyl-
amine (3 equiv), 55 ꢀC, 1 h; (e) water, MnO2 (3 equiv), 55 ꢀC, 24 h; (f) 1), n-Butylamine
(5 equiv), HCl (2.5 equiv), MeOH, reflux, 8 h; 2) NaBH4 (2 equiv), reflux, 1 h.
Scheme 2. Synthesis of cytosine derivatives via acidic/basic route. Reagents and
reaction conditions: (a) Mg(MeO)2 (5 equiv), aromatic aldehyde (6 equiv), MeOH,
55 ꢀC, 3 h; (b) AcOH (5 equiv), CH3COH or CH3CH2COH or CH3CH2CH2COH (4 equiv),
MeOH, reflux, 3 h; (c) NaBH4 (9 equiv), 0.5 h, MeOH, rt; (d) 1) 2 equiv BH3-DMS, THF,
CH2Cl2, 12 h. 2) HCl, MeOH aq.

246
B. Plitta et al. / European Journal of Medicinal Chemistry 55 (2012) 243e254
superposition of crystal structure of mouse DNMT1 (sequence
homology of mouse AF162282 and human X63692 DNMT1 is 82%)
and M.HhaI in their DNA-bound complexes shows that five out of
six conserved sequence motifs of their catalytic domain adopt very
similar conformation [33]. Moreover comparison of a modeled
tertiary structure of M.SssI with M.HhaI shows that amino acid
residues which take part in cofactor binding, target recognition and
catalysis occupy the same position in three dimensional structure
despite a sequence heterogeneity of the genes [33e36]. Based on
those findings we have used DNA methyltransferase M.SssI as
a model system for analysis of DNA methylation inhibitors.
experiment EGCG inhibit 42% of human DNMT1 at the same
concentration as 24 (10 M). In DNMT1 assay both compounds are
less potent than in M.SssI inhibition assay (Fig. 2).
m
3.3. Effect of test compounds on DNA methylation of different cell
lines
HeLa and HEK293 cells were treated with compounds 7, 24, 29,
to study their influence on the global level of DNA methylation.
Genomic DNA was isolated, digested into nucleotides and labeled
with [g-
32P]ATP. The mixture of nucleotides was analyzed by two-
Our cytosine derivatives were divided into three groups
according to modification at exocyclic amino group. The first group
consists of furfuryl derivatives of cytosine (Table 1). Ki for 4-N-
dimensional thin layer chromatography on cellulose coated plates
(Fig. 3a). Amount of m5C was calculated according to equation:
R ¼ m5C/(m5C þ dC þ dT) ꢁ 100 [40]. The compounds 7, 24 and 29
reduce m5C level in DNA of HeLa and HEK293 cells (Fig. 3b). IC50
values (concentration of inhibitor which causes 50% decrease in
methylation level) for these compounds determined for HeLa cells
furfurylcytosine (4) was 70
mM. This compound was further
modified in different ways. Substitution of furfuryl moiety of 4 with
hydroxymethyl group (5) caused tenfold increase of Ki. Introduction
of methyl group at C-5 position (29) resulted in 4.6 times increase
of inhibitory potency while addition of hydroxymethyl group (15)
resulted in decreasing of inhibitory potency (30 times) relative to
activity of 29. The presence of 20-deoxyribose (3, 27) increased Ki
values.
Substitution at C-5 with acetoxymethyl group (17) slightly (1.5
times) decreases Ki value comparing to leading compound.
Exchange of 5-(N-butyloamine)methyl group in 23 into formyl
group (21) or modification with sulfonyl group (19) resulted in
decreasing of inhibitory activity. The highest value of Ki of all
were 31, 26 and 88
compounds 7, 24 and 29 were above 100
furfuryl-5,6-dihydro-5-azacytosine this value is lower (75
m
M, respectively. For HEK293 IC50 for
M, whereas for 4-N-
M).
m
m
Detailed analysis of the products of hydrolysis of DNA isolated
from cell growing in the presence of inhibitors did not show
detectable amount of analyzed compounds (Fig. 3a).
4. Discussion
Until now the best DNMT1 inhibitor which is being used in
clinical practice is 5-azacytidine, which is a simple derivative of the
substrate. However, that compound is too toxic to meet the
demands of modern medicine [41]. Therefore we have been looking
for a new analogs of the substrate of DNA methylation. We chose N-
4-furfurylcytosine as a starting compound.
Analysis of the crystal structure of DNA methyltransferase HhaI
showed that the cytosine binding site is located in largely hydro-
phobic pocket composed of polar amino acids residues [42]. Thus,
to fulfill the prerequisite that DNMT1 inhibitors should be
a hydrophobic derivative of a substrate, we used 4-N-furfur-
ylcytosine as a starting compound. Recently we have found that it
occurs naturally in DNA from calf thymus [manuscript in prepara-
tion]. Plausible mechanism of its formation by the intramolecular
rearrangement of oxidatively modified deoxyribose at C-50, has
been already proposed for kinetin, another furfural derivative
[22e24].
The inhibitors which directly interact with DNMT1 enzyme are
expected to be less toxic comparing to those which are incorpo-
rated into the DNA like 5-azacytidine, 5-aza-20-deoxycytidine and
zebularine and form DNAeprotein adduct through the formation of
covalent bond with the enzyme [1]. Dixon plot analysis showed
that almost all compounds act as competitive inhibitors, with
exception of compounds 6 and 10e14 which are noncompetitive
and uncompetitive ones, respectively. In our in vitro assay, analyzed
compounds show significant inhibition of the enzyme activity
without being incorporated into DNA chain, in contrast to 5-aza-
cytidine which is active only when the inhibitor is within the DNA
chain. In our test 5-aza-cytosine inhibit SssI activity with Ki
analyzed compounds which is 0.7 mM for 4-N-furfuryl-5,6-dihydro-
5-azacytosine (24): is 100 times lower than Ki of a starting
compound 4. The increase of the hydrophobicity of compound 4
caused by benzyl moiety (7) (Table 2) enhances its inhibitory
potential sevenfold. However 7 is less active than 24. Similarly by
changes of furfuryl (24) with benzyl group in 25 did not improve
the activity. Substitution of benzyl moiety in 7 with methyl group in
para position (8) resulted in 2.6 fold increase of Ki. Changes of
benzyl into 3-picolyl group (9) resulted in decrease of the inhibitory
activity. Among benzyl derivatives the most active compound is 30.
Comparison of furfuryl and benzyl derivatives (16, 18, 20, 22) shows
that these substituted with benzyl group are more potent, but still
their Ki values are higher than observed for 24. Within this group
the derivative 6 is noncompetitive inhibitor.
The third group consists of derivatives substituted with aliphatic
chains at 4-N (Table 3). All of them act as uncompetitive inhibitors
as determined from Dixon plot [37]. In these cases Ki0 values were
determined by Cornish-Bowden method where S/V (concentration
of substrate vs velocity of reaction) were plotted against
I
(concentration of inhibitor) concentrations [38]. Because Ki0 values
were very high these compounds were not investigated further. For
comparison 5-aza-cytosine (31), RG108 (32) and EGCG (33) inhibit
the activity of SssI in our test with Ki e 5500, 0.23 and 0.028 mM,
respectively (Table 1).
3.2. Inhibition of human DNMT1 activity in colorimetric assay
The most potent inhibitor of M.SssI from furfuryl derivative of
M). In order to
cytosine is compound 24 (Ki ¼ 0.7
m
M; IC50 ¼ 2.73
m
5500 mM, while Ki for RG108 is 0.23 mM. It means that cytosine
derivative might effectively inhibit methyltransferase activity
through direct interaction with the enzyme.
check its inhibitory potency in reaction catalyzed with human
DNMT1 we used non radioactive in vitro enzymatic assay. Analysis
was carried out in presence of 5
and 500 M concentration compound 24. Under assay conditions
24 shows dose-dependent activity and inhibit of human DNMT1
M. In concentration of 10 M 24 inhibits 55% of
human DNMT1 activity. Because one of the known best inhibitors,
RG108, was shown to inhibit only 11% of DNMT1 at concentration of
m
g of nuclear extract and 10, 100
Inhibitory activity of N-4-furfuryl-5,6-dihydro-5-azacytosine
(24) is 3 times weaker than e RG108. However 24 an analog of
5,6-dihydro-5-azacytosine with saturated pyrimidine double bond
between N-5 and C-6 forms reversible complex with DNMT1 and
reduces its toxicity [43]. To check its activity against human
DNMT1, additional enzymatic assay was performed. In this assay
we used nuclear extract from HeLa cells and hemimethylated DNA
m
with IC50 ¼ 22
m
m
100
mM [39], to compare our results we used EGCG. In our

B. Plitta et al. / European Journal of Medicinal Chemistry 55 (2012) 243e254
247
Table 1
Inhibition of SssI enzyme in DNA methylation in vitro assay by furfuryl and benzyl derivatives of cytosine.
.
c
R1
R2
R3
X
C
Ki (
m
M)a
M.W. (Da) H-bond acceptor/donor atomd c Log Pb tPSA (A)
ꢁ
Compound
HO
O
O
O
CH2
3
H
170
307
8/3
ꢂ0.22
103.6
OH
CH2
O
4
H
H
H
H
H
C
C
C
C
70
700
440
50
191
221
221
263
5/2
6/3
6/3
7/2
0.3
62.7
82.9
82.9
89
CH2
5
H
ꢂ0.22
ꢂ0.41
ꢂ0.18
HO
O
CH2
CH2
15
17
HOCH2e
O
O
CH2
19
H
C
112
285
8/3
ꢂ0.96
117
O
O
O
O
CH2
CH2
CH2
CH2
21
23
24
27
H
H
HCOe
C
C
N
C
83
219
276
194
321
6/2
6/3
6/3
8/3
ꢂ0.55
0.97
0.53
0.13
79.7
74.7
CH2
960
N
H
H
H
0.7
74.7
HO
O
CH3e
250
103.6
OH
O
CH2
29
31
32
H
H
CH3e
C
15
205
112
334
5/2
5/3
0.65
ꢂ1.24
2.7
62.7
84.6
H
e
N
5500
RG108
EGCG
0.23
6/2
92.1
33
0.028 458
11/8
2.2
197.3
a
All derivatives are competitive inhibitors with Ki measured according to Dixon plot.
c Log P is calculated with ChemDraw Ultra 11.0 programme.
tPSA (topological polar surface area) is calculated with ChemDraw Ultra 11.0 programme.
b
c
d
The numbers of Lipinski’s H-bond acceptors/donors were calculated with Discovery Studio 2.5.
as a substrate. N-4-furfuryl-5,6-dihydro-5-azacytosine (24) inhibits
measured) nor DNA substrate, which makes impossible to deter-
minate the Ki value.
Analysis of the effect of these three potent compounds on DNA
methylation in cell lines (7, 24, 29) (Schemes 2, 4 and 5) showed
that they reduce the methylation level in HEK293 and much more
human DNMT1 with IC50 ¼ 22
m
M, this result is higher than the one
observed in the case of SssI (IC50 ¼ 2.72
mM) but in the experiment
with human DNMT1 we could not precisely determine neither the
concentration of DNMT1 (only total amount of protein was

248
B. Plitta et al. / European Journal of Medicinal Chemistry 55 (2012) 243e254
Table 2
Inhibition of SssI enzyme in DNA methylation in vitro assay by benzyl derivatives of cytosine.
c
R1
R2
R3
X
Ki (
m
M)a
M.W. (Da)
265
H-bond acceptor/donor atomsd
6/2
c Log Pb
tPSA (A)
ꢁ
Compound
O
S
2
H
H
H
C
550
1.45
87.6
H3C
O
O
NO2
CH2
6
H
C
6400
246
7/2
0.74
105.3
7
8
9
H
H
H
H
H
H
C
C
C
10
26
201
215
202
4/2
4/2
5/2
1.68
2.17
0.35
53.4
53.4
65.8
CH2
H3C
N
CH2
CH2
400
16
18
H
H
HOCH2e
H
C
135
23
231
273
5/3
6/2
0.97
1.2
73.7
79.7
CH2
CH2
O
CH2
H3C
O
O O
S
20
H
C
33
295
7/3
0.43
107.8
CH2
HO
CH2
22
25
H
H
HCOe
C
35
20
229
204
5/2
5/3
0.84
1.91
70.5
65.5
CH2
CH2
H
N
30
H
CH3e
C
3.6
215
4/2
2
53.4
CH2
a
Almost all derivatives are competitive inhibitors with Ki measured according to Dixon plot. Compound 6 is noncompetitive inhibitor.
c Log P is measured as indicated above.
tPSA is measured as indicated above.
b
c
d
Numbers of Lipinski’s H-bond acceptors/donors are measured as indicated above.
in HeLa cells. Having in mind that copying of the methylation
acceptors, Mw lower than 500 Da and log P (octanolewater parti-
tion coefficient) less than 5 [44]. Therefore a perfect inhibitor of
DNMT1 should be a small molecule, able to bind specifically to
catalytic domain of enzyme and inhibit its activity without any side
effects [20]. All analyzed compounds clearly fulfill the Lipinski’s
rules and have less than 5 and 10 hydrogen bond donors and
acceptors, respectively. Their molecular weight and c log P (calcu-
lated log P) are less than 500 Da and 5, respectively (Table 1).
pattern follows replication process, one can expect that increased
rate of cancer cells divisions will result in greater reduction of DNA
methylation in comparison to normal cells after treatment with
these inhibitors.
All presented compounds were checked to fulfill Lipinski’s rules
[44]. According to them the active drug should have not more than
5 hydrogen bond donors and not more than 10 hydrogen bond
Table 3
Inhibition of SssI enzyme in DNA methylation in vitro assay by aliphatic derivatives of cytosine.
c
R1
R2
R3
X
Ki (m
M)a
M.W. (Da)
H-bond acceptor/donor atomsd
c Log Pb
tPSA (A)
ꢁ
Compound
10
11
12
13
14
CH3CHe
H
H
H
H
H
H
H
H
H
H
C
C
C
C
C
5600
3400
800
6000
1000
137
151
165
139
167
4/1
4/1
4/1
4/2
4/2
ꢂ0.17
0.48
0.9
0.15
1.05
53.8
53.8
63.8
53.4
53.4
CH3CH2CHe
CH3CH2CH2CHe
CH3CH2e
CH3CH2CH2e
a
All derivatives are uncompetitive inhibitors with Ki measured according to Cornish-Bowden’s plot.
c Log P is measured as indicated above.
tPSA is measured as indicated above.
b
c
d
Numbers of Lipinski’s H-bond acceptors/donors are measured as indicated above.

B. Plitta et al. / European Journal of Medicinal Chemistry 55 (2012) 243e254
249
solvents were evaporated. The dry residue was dissolved with
water. Product was extracted with methylene chloride and cleaned
with active carbon. Evaporation of organic solvents gave 4-N-p-
toluenesulfonylcytosine (2) as a white powder.
Yield 78%, 269 mg. MS (ESI, pos.) m/z: 288 [M þ Na]þ, 553
[2M þ Na]þ. 1H NMR:
2.4 (s, 3H, CH3); 6.4 (d, J ¼ 7.5 Hz, 1H, H-5);
d
7.4 (d, J ¼ 8.1 Hz, 1H, H-6); 7.6 (d, J ¼ 7.5 Hz, 2H, H-8, H-10); 7.7 (d,
J ¼ 8.1 Hz, 2H, H-9, H-11); 7.6 (s, 1H, NH-7). 13C NMR (75 MHz,
CD3CN-d3):
d 22.13; 98.28; 127.85; 131.16; 141.10; 145.08; 145.48;
153.70; 161.90.
5.1.2. 4-N-Furfuryl-20-deoxycytidine (3)
4-N-p-Toluenesulfonyl-20-deoxycytidine (1) [26] (1.1 g,
2.88 mmol, 1 equiv) and furfurylamine (1.4 g, 14.39 mmol, 5 equiv)
were added to anhydrous pyridine (25 mL). The reaction flask was
closed and kept in an oven at 80 ꢀC for 12 h. Next, dichloromethane
(25 mL) was added and the mixture was extracted with water
(3 ꢁ 25 mL). The aqueous layer was evaporated under reduced
pressure to give a product 3 as a white foam.
Fig. 2. Plot of DNMT1 activity in the presence of different concentration of inhibitors:
N-4-furfuryl-5,6-dihydro-5-azacytosine (24) and EGCG (33). Percentage of change in
enzymatic activity was determined in relation to DNMT1 without inhibitor. Light gray
bars e EGCG, dark gray bars e compound 24.
Yield85%,752mg. MS(ESI,pos.)m/z:308[Mþ H]þ, 330[Mþ Na]þ,
615 [2M þ H]þ, 637 [2M þ Na]þ. 1H NMR:
d
1.9 (m, 1H, H-20); 2.0 (m,
1H, H-200); 3.5 (d, J ¼ 2.4 Hz, 2H, H-50); 3.7 (q, J ¼ 3.9 Hz, J ¼ 6.8 Hz,1H,
H-40);4.2 (q, J ¼ 3.2 Hz, J ¼ 5.5 Hz,1H, H-30);4.5 (d, J ¼ 5.5 Hz, 2H, H-8);
4.9 (m,1H, 50-OH); 5.2 (m,1H, 30-OH);5.8 (d, J ¼ 7.4 Hz,1H, H-5);6.1 (t,
J ¼ 6.1 Hz, 1H, H-10); 6.3 (q, J ¼ 0.7 Hz, J ¼ 3.2 Hz, 1H, H-9); 6.4 (q,
J ¼ 1.8 Hz, J ¼ 3.2 Hz, 1H, H-10); 7.3 (d, J ¼ 7.9 Hz, 1H, H-6); 7.6 (q,
J ¼ 0.7 Hz, J ¼ 1.8 Hz, 1H, H-11); 8.0 (t, J ¼ 5.3 Hz, 1H, NH-7). 13C NMR
Similarly, calculated tPSA (topological polar surface area) for all
2
ꢁ
compounds is less than 140 A what is necessary for efficient
transport through the membranes. Moreover, tPSA values lower
2
ꢁ
than 60 A as calculated i.e. for 7, 8, 30 (Schemes 2 and 5) and
indicates the possibility of crossing the bloodebrain barrier. The
analyzed derivatives fulfill also a criterion of aromaticity which is
extremely important for bioavailability and drugetarget interac-
(75 MHz, DMSO-d6): d 36.42; 38.67; 61.34; 70.37; 84.89; 87.19; 94.41;
107.24; 110.48; 140.24; 142.23; 151.75; 154.92; 163.11.
5.1.3. 4-N-Furfurylcytosine (4)
tion via van der Waals forces or p-stacking. A number of aromatic
4-N-Furfuryl-20-deoxycytidine (3) (731 mg, 2.38 mmol, 1 equiv)
was dissolved in water (18 mL) and concentrated hydrochloric acid
(1.8 mL, 8.4 equiv) was added. The mixture was refluxed for ca. 1 h
while the reaction mixture turned yellowish. After cooling to room
temperature the pH of the mixture was adjusted to ca. 7 with 1 M
aqueous solution of potassium hydroxide. Then the solvents were
removed by evaporation under the reduced pressure. The 4-N-
furfurylcytosine (4) was purified by silanized silica gel column
chromatography and was eluted with up to 10% of the methanol in
warm water (40e45 ꢀC). Product was obtained as a white powder in
yield 90%, 409 mg.
rings larger than two, reduce hydrophilicity and lead to a poor drug
distribution within the body [45]. Moreover, our cytosine deriva-
tives take advantage of being small molecules which can be used in
a rapid, dose-dependent and reversible manner [45].
In conclusion, we showed a new family of cytosine derivatives
having a big potential to inhibit DNA methyltransferase activity and
methods of their synthesis.
5. Experimental section
5.1. Chemistry
5.1.4. Derivatives of cytosine with aryl substituent at 4-N position
Cytosine (300 mg, 2.70 mmol,1 equiv) and appropriate aromatic
aldehyde (16.20 mmol, 6 equiv) were added to methanol (ca. 25 mL)
containing magnesium methylate (1.2 g, 13.50 mmol, 5 equiv). The
reaction was carried for 3 h at 55 ꢀC. Then sodium borohydride
(923 mg, 24.30 mmol, 9 equiv) was slowly added to the mixture and
the reduction reaction was carried for 30 min at room temperature.
After that, the pH of the mixturewas adjusted to7 with conc. HCl and
then the solvents were removed by evaporation under the reduced
pressure. Water (25 mL) was added to the residue and the mixture
was extracted with ethyl acetate (25 mL). The organic layer was
extracted with aqueous solution of hydrochloric acid (0.05 M,
25 mL). The combined aqueous extracts were neutralized with aq.
potassium hydroxide (1 M, 1.3 mL) and extracted with ethyl acetate
(25 mL). The organic layer was dried with magnesium sulfate and
then evaporated under the reduced pressure to give a product as
a white solid. The yields range: from 55 to 86%.
All reagents were commercially available or their synthesis has
been described [26]. Solvents except of methanol were used
without further purification. Reaction progresses were usually
monitored with TLC using Merck silica gel 60 F254 plates and UV
light at 254 nm. All chromatographic purifications were carried out
on the silica gel 60 0.0063e0.200 mm (Merck) and on the silica gel
60 0.040e0.063 mm or on the silanized silica gel 60
0.063e0.200 mm (Merck). 1H NMR (300 MHz) spectra were
recorded on a Varian 300 MHz spectrometer using DMSO-d6 as
a solvent. 13C NMR spectra were recorded on a Bruker 400 MHz
spectrometer or a Varian 300 MHz spectrometer using DMSO-d6,
D2O-d2 or MeOH-d4 as solvents. Tetramethylsilane (TMS) was
used as an internal standard. Mass spectra (EMS) were measured
with ESI Micro Q-TOF (Bruker).
5.1.1. 4-N-p-Toluenesulfonylcytosine (2)
To the flask containing 4-N-toluenesulfonyl-20-deoxycytidine
(1) (495 mg, 1.3 mmol) acetonitrile (10 mL), methanol (2 mL) and
hydrochloric acid (0.26 mL, 3.12 mmol, 2.4 equiv) were added. The
reaction was refluxed for 3 h. After that the pH was adjusted with
KOH (175 mg, 3.12 mmol, 2.4 equiv) in methanol (10 mL), and the
5.1.4.1. 4-N-Furfurylcytosine (4). Yield 56%, 289 mg. MS (ESI, pos.)
m/z: 192 [M þ H]þ, 214 [M þ Na]þ, 230 [M þ K]þ. 1H NMR:
d 4.5 (d,
J ¼ 5.4 Hz, 2H, H-8); 5.6 (d, J ¼ 7.1 Hz, 1H, H-5); 6.3 (q, J ¼ 0.7 Hz,
J ¼ 3.2 Hz, 1H, H-9); 6.4 (q, J ¼ 1.9 Hz, J ¼ 3.2 Hz, 1H, H-10); 7.2 (d,

250
B. Plitta et al. / European Journal of Medicinal Chemistry 55 (2012) 243e254
Fig. 3. Panel A, two-dimensional thin layer chromatography (TLC) analysis of [50-32P] labeled deoxynucleotides obtained by enzymatic hydrolysis from HeLa cell line treated with
DNA methyltransferases activity inhibitor, dA e 20-deoxyadenosine phosphate, dmC e 5-methyl-20-deoxycytosine phosphate, dC e 20-deoxycytosine phosphate, dT e 20-deoxy-
tymidine phosphate, dG e 20-deoxyguanosine phosphate. A comparison of m5C amount in DNA from HeLa (gray bars) and HEK293 cells (black bars) treated with various
concentration (1, 10, 100 mM) of 4-furfuryl-5-methylcytosine (28) panel B, 4-N-benzylcytosine (7) panel C and 4-furfuryl-5,6-dihydroazacytosine (24) panel D.
J ¼ 7.1, 1H, H-6); 7.6 (q, J ¼ 0.7 Hz, J ¼ 1.9 Hz, 1H, H-11); 7.9 (t,
J ¼ 5.6 Hz, 1H, NH-7); 10.3 (s, 1H, NH-1). 13C NMR (75 MHz, DMSO-
5.1.4.2. 4-N-(5-Hydroxymethylfurfuryl)cytosine
(5). Yield
62%,
370 mg. MS (ESI, pos.) m/z: 222 [M þ H]þ, 244 [M þ Na]þ, 443
d6):
d
40.33; 90.10; 107.16; 110.47; 141.82; 142.19; 151.96; 156.55;
[2M þ H]þ, 465 [2M þ Na]þ. 1H NMR:
4.3 (d, J ¼ 7.2 Hz, 2H, H-11);
d
164.29.
4.4 (d, J ¼ 5.7 Hz, 2H, H-8); 5.2 (t, J ¼ 7.6 Hz, 1H, OH); 5.6 (d,
Scheme 4. Synthesis of 5-azacytosine derivatives via basic route. Reagents and conditions: (a) Mg(MeO)2 (5 equiv), aromatic aldehyde (6 equiv), MeOH, 55 ꢀC, 3 h; (b) NaBH4
(9 equiv), 0.5 h, MeOH, rt.

B. Plitta et al. / European Journal of Medicinal Chemistry 55 (2012) 243e254
251
N
were added. Then all solvents were evaporated and then conc. HCl
(1 mL) in methanol (10 mL) and water (10 mL) were added to
reaction mixture. The mixture was left for 12 h. Then the solvents
were evaporated and product was separated by silanized silica gel
column chromatography with mixture of water and acetone.
N
N
N
R
N
HN
N
N
O
HO
2'dRib =
a
AcO
O
O
5.1.5.1. 4-N-Ethylidenecytosine acetate (10). Yield 92%, 652 mg. MS
(ESI, pos.) m/z: 138 [M þ H]þ; 170 [M þ MeOH þ H]þ; 192
O
OH
R1
OAc
[M þ MeOH þ Na]þ. 1H NMR:
1.2 (d, J ¼ 5.6 Hz, 3H, H-9); 3.1 (s, 3H,
d
26
27 R = Fur, R1 = 2'dRib
R = Bnz, R1 = 2'dRib
R = Fur, R1 = H
CH3 of anion); 5.2 (q, JA ¼ 14.3 Hz, JB ¼ 6.2 Hz, 1H, H-8); 5.5 (d,
J ¼ 7.2 Hz, 1H, H-5); 7.2 (d, J ¼ 7.2 Hz, 1H, H-6); 7.8 (d, J ¼ 8.8 Hz, 1H,
28
29
30
b
b
NH-7); 10.5 (s, 1H, NH-1). 13C NMR (100 MHz, DMSO-d6):
d 18.96;
R = Bnz, R1 = H
34.82; 61.47; 87.73; 120.23; 148.98; 154.96; 180.14.
Scheme 5. Synthesis of 5-methylcytosine derivatives via route of acid-catalyzed
hydrolysis of N-glycosidic bond. Reagents and reaction conditions: (a) 1) furfuryl-
amine or benzylamine (1.5 equiv), CH3CN, 50 ꢀC; 2) MeOH, 35% aq. ammonia, 1 h; (b)
conc. HCl (3 equiv), MeOH aq., reflux, 4 h.
5.1.5.2. 4-N-Propylidenecytosine acetate (11). Yield 76e80%,
578e608 mg. MS (ESI, pos.) m/z: 152 [M
þ
H]þ; 184
0.8 (t,
[M þ MeOH þ H]þ; 206 [M þ MeOH þ Na]þ. 1H NMR:
d
J ¼ 7.5 Hz, 1H, H-5); 6.1 (m, 2H, H-9, H-10); 8.1 (t, J ¼ 5.5 Hz, 1H, NH-
J ¼ 7.6 Hz, 3H, H-10); 1.5 (m, 2H, H-9); 3.1 (s, 3H, CH3 of anion); 5.2
(q, JA ¼ 14.8 Hz, JB ¼ 6.0 Hz, 1H, H-8); 5.5 (d, J ¼ 6.8 Hz, 1H, H-5); 7.2
(d, J ¼ 7.2 Hz, 1H, H-6); 7.7 (d, J ¼ 9.2 Hz, 1H, NH-7); 10.5 (s, 1H, NH-
7); 10.3 (s,1H, NH-9). 13C NMR (100 MHz, DMSO-d6):
d 40.33; 55.62;
93.24; 107.60; 107.75; 141.82; 151.17; 154.75; 156.67; 164.29.
1). 13C NMR (100 MHz, DMSO-d6):
d 13.86; 22.02; 34.47; 54.60;
5.1.4.3. 4-N-o-Nitrobenzylocytosine (6). Yield 54%, 359 mg. MS (ESI,
pos.) m/z: 247 [M þ H]þ, 269 [M þ Na]þ, 285 [M þ K]þ.
80.51; 92.91; 142.55; 156.58; 165.24; 174.66.
1H NMR:
d
4.7 (d, J ¼ 5.8 Hz, 2H, H-8); 5.8 (d, J ¼ 7.3 Hz, 1H, H-5);
5.1.5.3. 4-N-Butylidenecytosine acetate (12). Yield
48e65%,
389e526 mg. MS (ESI, pos.) m/z: 166 [M
þ
H]þ; 198
7.3 (d, J ¼ 5.8 Hz, 1H, H-6); 7.6 (m, 2H, H-11, H-12); 7.8 (m, 2H, H-9,
H-10); 8.1 (m, 1H, NH-7); 10.3 (s, 1H, H-1). 13C NMR (100 MHz,
[M þ MeOH þ H]þ; 220 [M þ MeOH þ Na]þ. 1H NMR:
d 0.8 (t,
DMSO-d6):
d 42.58; 96.11; 125.94; 129.43; 131.33; 134.72; 135.12;
J ¼ 7.3 Hz, 3H, H-11); 1.2 (m, 2H, H-10); 1.4 (m, 2H, H-9); 3.1 (s, 3H,
CH3 of anion); 5.4 (q, JA ¼ 15.1 Hz, JB ¼ 6.3 Hz, 1H, H-8); 5.5 (d,
J ¼ 6.8 Hz, 1H, H-5); 7.2 (d, J ¼ 6.8 Hz, 1H, H-6); 7.8 (d, J ¼ 8.7 Hz, 1H,
142.79; 149.74; 160.49; 166.97.
NH-7); 10.5 (s, 1H, NH-1). 13C NMR (100 MHz, DMSO-d6):
d 13.65;
5.1.4.4. 4-N-Benzylcytosine (7). Yield 61%, 331 mg. MS (ESI, pos.) m/
z: 202 [M þ H]þ, 224 [M þ Na]þ, 240 [M þ K]þ, 403 [2M þ H]þ, 425
17.79; 18.81; 34.66; 54.58; 80.23; 92.91; 142.52; 156.54; 165.54;
174.67.
[2M þ Na]þ, 441 [2M þ K]þ. 1H NMR:
4.5 (d, J ¼ 5.7 Hz, 2H, H-8);
d
5.7 (d, J ¼ 7.1 Hz, 1H, H-5); 7.2e7.3 (m, 6H, H-6 and H of Ar); 8.0 (t,
J ¼ 5.9 Hz, 1H, NH-7); 10.3 (s, 1H, H-1). 13C NMR (100 MHz, MeOD-
5.1.5.4. 4-N-Ethylcytosine (13). Yield 57%, 158 mg. MS (ESI, pos.) m/
z: 140 [M þ H]þ; 162 [M þ Na]þ; 301 [2M þ Na]þ. 1H NMR:
d
1.1 (t,
d4): d 45.57; 96.75; 128.78; 129.36; 130.03; 130.44; 130.71; 140.24;
142.83; 161.19; 167.13.
J ¼ 7.3 Hz, 3H, CH3); 3.2 (m, 2H, H-8); 5.6 (d, J ¼ 6.8 Hz, 1H, H-5); 7.4
(d, J ¼ 6.8 Hz,1H, H-6); 7.9 (s, 1H, NH-7); 10.5 (s,1H, NH-1). 13C NMR
5.1.4.5. 4-N-p-Methylbenzylcytosine (8). Yield 86%, 499 mg. MS
(ESI, pos.) m/z: 216 [M þ H]þ, 238 [M þ Na]þ, 431 [2M þ H]þ. 1H
(100 MHz, DMSO-d6): d 14.57; 34.31; 93.25; 141.24; 156.85; 156.91;
164.30; 166.65.
NMR:
d
2.2 (s, 3H, CH3); 4.5 (d, J ¼ 5.6 Hz, 2H, H-8); 5.6 (d, J ¼ 7.1 Hz,
1H, H-5); 7.2 (d, J ¼ 6.8 Hz, 1H, H-6); 7.5 (m, 2H, H-9, H-10); 7.7 (m,
5.1.5.5. 4-N-Butylcytosine (14). Yield 88%, 294 mg. MS (ESI, pos.) m/
z: 168 [M þ H]þ; 190 [M þ Na]þ; 335 [2M þ H]þ; 357 [2M þ Na]þ. 1H
2H, H-11, H-12); 8.1 (t, J ¼ 5.9 Hz, 1H, NH-7); 10.2 (s, 1-H, NH-1). 13C
NMR:
d
0.9 (t, J ¼ 7.2 Hz, 3H, CH3); 1.3 (m, 2H, H-10, H-100); 1.4 (m,
NMR (100 MHz, DMSO-d6):
d 20.64; 42.62; 93.20; 127.39; 128.82;
2H, H-9, H-90); 3.2 (q, JA ¼ 12.4 Hz, JB ¼ 6.4 Hz, 2H, H-8); 5.6 (d,
135.87; 136.12; 141.62; 156.72; 164.45.
J ¼ 7.2 Hz, 1H, H-5); 7.2 (d, J ¼ 6.8 Hz, 1H, H-6); 7.5 (t, J ¼ 4.8 Hz, 1H,
NH-7); 10.2 (s, 1H, NH-1). 13C NMR (100 MHz, DMSO-d6):
d 14.12;
5.1.4.6. 4-N-Picolylcytosine (9). Yield 86%, 469 mg. MS (ESI, pos.) m/
z: 202 [M]; 225 [M þ Na]þ, 240 [M þ K]þ.1H NMR:
d 4.8 (d,
21.11; 32.15; 41.25; 96.31; 141.87; 166.71; 180.37.
J ¼ 5.9 Hz, 2H, H-8); 5.8 (d, J ¼ 7.3 Hz, 1H, H-5); 7.3 (d, J ¼ 7.8 Hz, 1H,
H-6); 7.4 (m, 1H, H-10); 7.7 (d, J ¼ 1.5 Hz, 1H, H-9); 8.1 (t, J ¼ 5.9 Hz,
1H, NH-7); 8.4 (m, 1H, H-11); 8.5 (d, J ¼ 1.3 Hz, 1H, H-12); 10.3 (s,
5.1.6. Derivatives of 4-N-arylmethylcytosine with hydroxymethyl
substituent at C-5
1H, NH-1). 13C NMR (100 MHz, DMSO-d6):
d 40.57; 93.15; 123.44;
4-N-Arylmethylcytosine (4, 7) (2.62 mmol, 1 equiv) and para-
formaldehyde (180 mg, 6.02 mmol, 2.3 equiv) were added to water
(ca. 25 mL) containing triethylamine (2.6 g, 26.00 mmol, 10 equiv).
The mixture was refluxed for 8 h and then the flask was closed and
kept in oven at 60 ꢀC for 12 h. After that the solvents were removed
by evaporation under the reduced pressure and ethanol (25 mL)
was added to a residue causing to precipitate a white powder. The
powder was filtered and dissolved in methanol (0.2e0.5 mL) and
again precipitated with ethanol (25 mL). This procedure was
repeated two more times.
134.72; 135.21; 141.92; 148.10; 148.87; 156.56; 164.50.
5.1.5. Derivatives of cytosine with alkyl substituent at external
amino group
Cytosine (400 mg, 3.60 mmol, 1 equiv) and alkyl aldehyde
(14.40 mmol, 4 equiv) were added to methanol (ca. 35 mL) con-
taining acetic acid (1 mL, 18.00 mmol, 5 equiv). The mixture was
refluxed for 3 h. Then, the solvents were removed by evaporation
under the reduced pressure. Then, products were isolated using
reverse chromatography with mixture of water and acetone. The
Schiff bases (2 mmol) were dissolved in methylene chloride (5 mL)
and then were reduced with 2 M borane dimethyl sulfide complex
in THF (2 mL, 4 mmol, 2 equiv) for 12 h at ambient temperature.
After that to the reaction flask water (3 mL) and acetone (10 mL)
5.1.6.1. 4-N-Furfuryl-5-hydroxymethylcytosine (15). Yield 85e95%,
492e550 mg. MS (ESI, pos.) m/z: 222 [M þ H]þ; 244 [M þ Na]þ;
465 [2M þ Na]þ. 1H NMR:
d
4.1 (d, J ¼ 4.9 Hz, 2H, H-13); 4.5 (d,
J ¼ 5.3 Hz, 2H, H-8); 5.0 (t, J ¼ 5.2 Hz, .1H, OH); 6.2 (m,1H, H-12); 6.3

252
B. Plitta et al. / European Journal of Medicinal Chemistry 55 (2012) 243e254
(m, 1H, H-11); 7.2 (t, J ¼ 5.2 Hz, 1H, NH-7); 7.3 (s, 1H, H-6); 7.5 (m,
1H, H-10); 10.3 (s, 1H, NH-1). 13C NMR (100 MHz, DMSO-d6):
DMSO-d6): d 45.48; 68.79; 99.89; 126.51; 127.12; 127.64; 128.12;
129.41; 142.40; 156.26; 163.94.
d
40.33; 57.13; 104.87; 106.88; 110.44; 140.02; 141.94; 152.27;
156.39; 163.08.
5.1.9. Oxidation of 4-N-arylmethyl-5-hydroxymetylcytosine
The
4-N-arylmethyl-5-hydroxymethylcytosine
(15,
16)
5.1.6.2. 4-N-Benzyl-5-hydroxymethylcytosine (16). Yield 80e90%,
484e545 mg. MS (ESI, pos.) m/z: 232 [M þ H]þ;254 [M þ Na]þ;
(0.91 mmol) and manganese oxide (IV) (MnO2) (237 mg,
2.73 mmol, 3 equiv) were added to flask followed by water (15 mL)
and ethylene dichloride (15 mL). The reaction mixture was inten-
sively stirred at ambient temperature for 24 h. After that, the
organic layer was separated and dried with anhydrous magnesium
sulfate, and then evaporated to give a product as a white powder.
270 [M þ K]þ; 485 [2M þ Na]þ. 1H NMR:
4.2 (d, J ¼ 4.8 Hz, 2H, H-
d
15); 4.5 (d, J ¼ 5.7 Hz, 2H, H-8); 7.2e7.3 (m, 5H, Ar); 7.6 (s, 1H, H-6);
10.5 (s, 1H, NH-1). 13C NMR (100 MHz, DMSO-d6):
d 45.48; 57.15;
104.96; 126.82; 127.08; 127.39; 128.18; 128.31; 139.53; 141.70;
156.56; 163.22.
5.1.9.1. 4-N-Furfuryl-5-formylcytosine
(21). Yield
65e75%,
5.1.7. Acetylation of 4-N-arylmethyl-5-hydroxymethylcytosine
derivatives
130e150 mg. MS (ESI, pos.) m/z: 220 [M þ H]þ; 242 [M þ Na]þ;
461 [2M þ Na]þ. 1H NMR:
4.6 (d, J ¼ 3.4 Hz, 2H, H-8); 6.3 (s, 1H, H-
d
The 4-N-aryl-5-hydroxymethylcytosine (15, 16) (0.45 mmol)
and acetic anhydride (0.05 mL, 0.54 mmol, 1.2 equiv) were added to
pyridine (0.2 mL, 2.71 mmol, 5 equiv). The reaction flask was kept at
ambient temperature for 3 h. After that the water (0.5 mL) was
added, and all solvents were evaporated. The dry residue was dis-
solved with water (15 mL). Product was extracted by ethyl acetate
(3 ꢁ 15 mL). The organic part was dried by sodium sulfate, and then
solvent was evaporated receiving clean, white powder product.
10);6.4 (s,1H, H-9); 7.6 (s,1H, H-11);8.4 (s,1H, H-6); 8.7 (t, J ¼ 5.6,1H,
NH-7), 9.4 (s,1H, COH). 13C NMR (100 MHz, MeOH):
108.93; 111.61; 143.83; 152.20; 157.26; 157.91; 163.16; 190.14.
d 38.07; 106.76;
5.1.9.2. 4-N-Benzyl-5-formylcytosine
(22). Yield
60e70%,
125e146 mg. MS (ESI, pos.) m/z: 230 [M þ H]þ; 252 [M þ Na]þ;
481 [2M þ Na]þ. 1H NMR:
4.6 (d, J ¼ 5.6 Hz, 2H, H-8); 7.3e7.2 (m,
d
5H, Ar); 8.4 (s,1H, H-6); 8.8 (t, J ¼ 5.2 Hz,1H, NH-7); 9.5 (s,1H, COH).
13C NMR (100 MHz, DMSO-d6):
d 45.35; 104.13; 126.85; 127.01;
5.1.7.1. 4-N-Furfuryl-5-acetyloxymethylcytosine (17). Yield 90e95%,
107e113 mg. MS (ESI, pos.) m/z: 264 [M þ H]þ; 286 [M þ Na]þ; 549
127.37; 128.16; 128.46; 138.52; 154.09; 157.02; 161.10; 189.04.
[2M þ Na]þ. 1H NMR:
d
1.9 (s, 3H, CH3); 4.5 (d, J ¼ 5.6 Hz, 2H, H-8);
5.1.10. 4-N-Furfuryl-5-(n-butylamine)methylcytosine
4.7 (s, 2H, H-13); 6.2 (d, J ¼ 3.2 Hz, 1H, H-12); 6.3 (m, 1H, H-11); 7.5
(s, 1H, H-6); 7.5 (m, 1H, H-10); 7.6 (t, J ¼ 5.21 Hz, 1H, NH-7); 10.5 (s,
5.1.10.1. Method A. To flask with solution of 4-N-furfuryl-5-
hydroxymethylocytosine (15) (300 mg, 1.36 mmol) in pyridine
(20 mL) the methanesulfonyl chloride (0.16 mL, 232 mg, 2.04 mmol,
1.5 equiv) was added. The reaction was kept at ambient temperature
for 30 min, after that the appropriate amine (4.07 mmol, 3 equiv)
was added and mixture was kept in oven at temperature 55 ꢀC for
1 h. After that to reaction flask water was added. Then all solvents
were evaporated and product was isolated by silanized silica gel
column chromatography and was eluted with water and acetone.
1H, N-H-1). 13C NMR (100 MHz, DMSO-d6):
d 20.86; 36.57; 57.12;
104.87; 106.87; 110.44; 140.02; 141.94; 152.26; 156.39; 163.08.
5.1.7.2. 4-N-Benzyl-5-acetyloxymethylcytosine
(18). Yield 95%,
117 mg. MS (ESI, pos.) m/z: 274 [M þ H]þ; 296 [M þ Na]þ; 569
[2M þ Na]þ. 1H NMR:
2.0 (s, 3H, CH3); 4.5 (d, J ¼ 5.2 Hz, 2H, H-8);
d
4.7 (s, 2H, H-14); 7.2e7.3 (m, 5H, Bz); 7.5 (s, 1H, H-6); 7.8 (t,
J ¼ 6.0 Hz, 1H, NH-7); 10.5 (s, 1H, H-1). 13C NMR (100 MHz, DMSO-
d6):
d
20.89; 45.21; 60.31; 99.36; 126.54; 126.89; 127.08; 127.37;
5.1.10.2. Method B. 4-N-Furfuryl-5-formylcytosine (21) (300 mg,
1.36 mmol) and appropriate amine (6.38 mmol, with methanol
(50 mL)) were refluxed for 8 h. After cooling of the mixture sodium
borohydride (103 mg, 2.72 mmol, 2 equiv) was added and then
refluxed for 1 h. After that the reaction mixture was adjusted to pH
7 with hydrochloric acid, solvents were evaporated under reduced
pressure and a product was isolated by silica gel column silanized
chromatography and was eluted with water and acetone.
128.14; 139.53; 143.90; 156.16; 162.80; 170.52.
5.1.8. Sulfonylation of 4-N-arylmethyl-5-hydroxymethylcytosine
The
4-N-arylmethyl-5-hydroxymethylcytosine
(15,
16)
(4.83 mmol) and sodium metabisulfite (1.8 g, 9.66 mmol, 2 equiv)
were dissolved in water (15 mL) or mixture of water and acetonitrile
(2:1, by vol., 15 mL). The reaction was kept in oven at 55 ꢀC for 24 h.
After that the solvents were evaporated, the product was isolated by
column chromatography using silanized silica gel chromatography
with mixture of water and acetone. Then the product was dissolved
with methanol (15 mL) and cleaned with active carbon.
5.1.10.3. 4-N-Furfuryl-5-(n-butylamine-)methylcytosine (23). Yield
60e75%, 225e282 (method A) mg. MS (ESI, pos.) m/z: 277
[M þ H]þ; 299 [M þ Na]þ; 575 [2M þ Na]þ. 1H NMR: 0.89 (t,
J ¼ 7.2 Hz, 3H, CH3-18); 1.26 (m, 2H, CH2-17); 1.45 (m, 2H, CH2-16);
3.1 (t, J ¼ 7.5 Hz, 1H, NH-14); 3.5 (m, 2H, H-15); 3.8 (m, 2H, H-13);
4.5 (s, 2H, H-8); 6.2 (m, 1H, H-11); 6.4 (m, 1H, H-10); 7.5 (s, 1H, H-6);
7.6 (m, 1H, H-12), 8.1 (m, 1H, NH-7). 13C NMR (100 MHz, DMSO-d6):
5.1.8.1. 4-N-Furfurylcytosine 5-methyl sulfonic acid sodium salt
(19). Yield 85%, 1171 mg. MS (ESI, pos.) m/z: 330 [M þ 2NaeH]þ. 1H
NMR:
d
3.5 (s, 2H, H-12); 4.4 (d, J ¼ 5.1 Hz, 2H, H-8); 6.3 (d,
J ¼ 2.4 Hz, 1H, H-11); 6.4 (d, J ¼ 2.9 Hz, 1H, H-9); 7.3 (s, 1H, H-6); 7.5
d 13.45; 19.136; 28.93; 36.66; 42.72; 45.95; 96.16; 107.02; 110.43;
128.08; 141.77; 152.18; 155.89; 162.49.
(m, 1H, H-10); 8.1 (t, J ¼ 5.3 Hz, 1H, NH-7). 13C NMR (100 MHz,
DMSO-d6):
d 37.12; 50.43; 99.92; 106.77; 110.42; 141.94; 142.62;
152.21; 156.01; 163.93.
5.1.11. Derivatives of 5-azacytosine with aryl substituent at 4-N
5-Azacytosine (500 mg, 4.46 mmol, 1 equiv) and appropriate
aromatic aldehyde (26.77 mmol, 6 equiv) were added to methanol
(ca. 40 mL) containing magnesium methylate (521.80 mg,
22.30 mmol, 5 equiv). The reaction was carried for 3 h at 55 ꢀC. Then
sodium borohydride (1.525 mg, 40.14 mmol, 9 equiv) was slowly
added to the mixture and the reduction reaction was carried for
30 min at room temperature. After that, the pH of the mixture was
adjusted to 7 with conc. HCl and then the solvents were removed by
5.1.8.2. 4-N-Benzylcytosine 5-methylsulfonic acid sodium salt
(20). Yield 75%, 1069 mg. MS (ESI, pos.) m/z: 296 [M þ H]þ; 318
[M þ Na]þ; 340 [2M þ NaeH]þ. 1H NMR:
d 4.2 (s, 2H, H-15); 4.5 (d,
J ¼ 5.6 Hz, 2H, H-8); 7.2e7.4 (m, 5H, Ar þ H-6); 8.1 (t, J ¼ 5.6 Hz, 1H,
NH-7). 1H NMR (MeOD)
d 3.88 (s, 2H, H-15); 4.65 (s, 2H, H-8); 7.20
(t, J ¼ 7.6, 1H, H-14); 7.28 (t, J ¼ 7.6 Hz, 2H, H-10, H-12); 7.38 (d,
J ¼ 7.6 Hz, 2H, H-11, H-13); 7.42 (s, 1H, H-6). 13C NMR (100 MHz,

B. Plitta et al. / European Journal of Medicinal Chemistry 55 (2012) 243e254
253
evaporation under the reduced pressure. Methanol (30e40 mL)
was added to the residue and the mixture was filtered. Then
solvents were evaporated and product was isolated by silica gel
column chromatography and was eluted with ethyl acetate and
methanol. The yields range: 60%.
5.1.13.1. 4-N-Furfuryl-5-methylcytosine (29). Yield 74%, 189 mg. MS
(ESI, pos.) m/z: 206 [M þ H]þ, 228 [M þ Na]þ, 244 [M þ K]þ. 1H
NMR: d 1.3 (s, 3H, CH3); 4.6 (s, 2H, H-8); 6.2 (m, 1H, H-9); 6.3 (m, 1H,
H-10); 7.3 (m, 1H, H-11); 7.4 (s, 1H, H-6). 13C NMR (75 MHz, DMSO-
d6): 13.40; 37.38; 104.66; 108.82; 111.97; 140.57; 143.53; 153.76;
d
161.01; 166.39.
5.1.11.1. 4-N-Furfuryl-5,6-dihydro-5-azacytosine (24). Yield 60%,
520 mg. MS (ESI, pos.) m/z: 195 [M þ H]þ. 1H NMR: 4.2 (s, 2H, H-6);
4.5 (s, 2H, H-8); 6.2 (m, 1H, H-11); 6.4 (m, 1H, H-10); 7.1 (s, 1H, NH-
5.1.13.2. 4-N-Benzyl-5-methylcytosine (30). Yield 74%, 199 mg. MS
(ESI, pos.) m/z: 216 [M þ H]þ; 238 [M þ Na]þ; 453 [2M þ Na]þ. 1H
3); 7.5 (m, 1H, H-9). 13C NMR (100 MHz, DMSO-d6):
d
37.14; 56.14;
NMR:
d
1.9 (s, 3H, CH3); 4.5 (d, J ¼ 6.0 Hz, 2H, H-8); 7.1 (s, 1H, H-6);
106.72; 110.44; 142.05; 152.46; 154.63; 167.43.
7.2e7.3 (m, 5H, Ar); 7.5 (t, J ¼ 6.0 Hz, 1H, NH-7). 13C NMR (75 MHz,
DMSO-d6):
d 12.75; 42.97; 100.21; 126.60; 127.10; 128.14; 128.20;
5.1.11.2. 4-N-Benzyl-5,6-dihydro-5-azacytosine (25). Yield 63%,
573 mg. MS (ESI, pos.) m/z: 205 [M þ H]þ. 1H NMR: 4.2 (s, 2H, H-6);
4.5 (s, 2H, H-8); 7.1 (s,1H, NH-3); 7.3 (m, 5H, Ar). 13C NMR (100 MHz,
128.45; 139.78; 156.73; 163.10.
5.2. Biology
DMSO-d6):
d 43.91; 55.69; 127.27; 127.61; 127.92; 128.72; 129.44;
139.78; 155.17; 165.37.
5.2.1. Plasmid DNA in vitro methylation assay
SssI, CpG methyltransferase, isolated from a strain of Escherichia
coli transfected with the SssI gene from Spiroplasma sp. strain MQ1
was purchased from New England Biolabs. [3H]CH3-adenosyl-L-
methionine (SAM) (specific activity 10 Ci/mmol) was purchased
from PerkinElmer Life and Analytical Sciences. The scintillation
cocktail ingredients: toluene, 2,5-diphenyloxazole (PPO), 2,20-p-
Phenylene-bis(5-phenyloxazole) (POPOP) were purchased from
SigmaeAldrich. pUC18 plasmid was isolated using Qiagen Plasmid
Plus Maxi Kit. The purity of DNA was estimated by UV spectroscopic
measurement and the A260/A280 ratio was 1.9e2.0.
5.1.12. 4-N-Arylmethyl-5-methyl-20-deoxycytidine
5-Methyl-4-(1,2,4-triazol-1-yl)-1-(b-D-3,5-di-O-acetyl-2-
deoxyribofuranosyl)pyrimidin-2(1H)-one (26, 388 mg, 1.33 mmol,
1 equiv) and appropriate amine (1.99 mmol, 1.5 equiv) were dis-
solved in anhydrous acetonitrile (15 mL) [26]. The reaction flask
was closed and kept at 50 ꢀC for 2 h. The part of the product in the
form of a beige solid was filtered off. To remove the acetyl groups
from rest of the product, the filtrate was evaporated down and the
residue was dissolved in methanol (15 mL) and 32% aqueous
ammonia (15 mL). The mixture was heated till boiling for an hour.
Then the solvents were evaporated under reduced pressure. The
residue was extracted with dichloromethane (20 mL) and water
(3 ꢁ 20 mL). The aqueous layer (with product) was evaporated
down and the product as a white foam was obtained.
To optimize the assay, different concentrations of SAM
(0.1e25 mM), pUC18 plasmid (0.1e1 mg), SssI enzyme (0.5e4 U) and
reaction time (5e180 min) were evaluated. To determine the rate of
methylation reaction (V0 [nM/min]), the reaction was carried out in
total volume of 20 mL containing 0.1e1 mg of pUC18 plasmid, 2 mM
[3H]CH3-SAM, 1 U of SssI in 10 mM TriseHCl pH 7.9 reaction buffer
5.1.12.1. 4-N-Furfuryl-5-methyl-20-deoxycytidine (27). Yield 94%,
containing 50 mM NaCl, 10 mM DTT for 1, 3, 5 and 10 min at 37 ꢀC.
399 mg. MS (ESI, pos.) m/z: 322 [M þ H]þ, 360 [M þ K]þ.1H NMR:
d
1.8
The reaction was stopped by loading 18 mL of reaction mixture on
(s, 3H, CH3); 1.9 (m,1H, H-20); 2 (m,1H, H-200); 3.5 (d, J ¼ 2.4 Hz, 2H, H-
50); 3.7 (q, J ¼ 3.8 Hz, J ¼ 6.7 Hz,1H, H-40); 4.2 (t, J ¼ 2.9 Hz, 1H, H-30);
4.5 (d, J ¼ 5.8 Hz, 2H, H-8); 5.0 (m, 1H, 50-OH); 5.1 (m, 1H, 30-OH); 6.1
(t, J ¼ 5.8 Hz, 1H, H-10); 6.2 (d, J ¼ 2.9 Hz, 1H, H-9); 6.3 (m, 1H, H-10);
7.3 (m, 1H, H-11); 7.8 (s, 1H, H-6); 8.0 (m, 1H, NH-7). 13C NMR
Whatmann GF/C filters, washed with 3 mL of 10% TCA, twice with
3 mL of 5% TCA, and 3 mL of ethanol. The amount of methylated
DNA precipitated on dried filters was measured in 4 mL of scintil-
lation cocktail in Beckman 5000TA counter. All reactions were
repeated twice.
(100 MHz, DMSO-d6):
89.39; 108.49; 110.89; 113.56; 140.20;145.50;153.97;160.28;166.23.
d
15.22; 40.24; 42.10; 64.23; 73.51; 88.63;
Optimized SssI methyltransferase inhibition activity assay was
performed with three different amounts of substrate (pUC18
plasmid) e 0.1, 0.5 and 1
concentration of tested compounds in reaction buffer described
above. Concentration of CpG in pUC18 is 4.4 M in 1 g (0.56 pmol)
mg, 2 mM SAM, 1 U of SssI and increasing
5.1.12.2. 4-N-Benzyl-5-methyl-20-deoxycytidyne (28). Yield 96%,
423 mg. MS (ESI, pos.) m/z: 332 [M þ H]þ; 357 [M þ Na]þ; 685
[2M þ Na]þ.
m
m
of plasmid. The reactions were carried out at 37 ꢀC for 10 min. Ki
values were determined from Dixon plot (1/V vs I e inhibitor
concentration) [37]. Ki0 value was calculated according to Cornish-
Bowden method [38]. All reactions were repeated twice.
1H NMR: (MeOH) 1.9 (s, 3H, CH3); 2.1 (m, 1H, H-20); 2.2 (m, 1H,
d
H-200); 3.9 (q, JA ¼ 6.8 Hz, JB ¼ 3.2 Hz, 1H, H-30); 4.1 (s, 2H, H-50); 4.3
(m,1H, H-40); 4.6 (s, 2H, H-8); 6.2 (t, J ¼ 6.8 Hz,1H, H-10); 7.4 (m, 6H,
Ar); 7.8 (s, 1H, H-6). 13C NMR (100 MHz, DMSO-d6):
d 14.34; 35.56;
53.31; 62.54; 77.76; 79.23; 95.53; 118.54; 119.02; 119.91; 120.40;
120.66; 129.40; 130.75; 149.20; 155.39.
5.2.2. Inhibition of human DNMT1
Nuclear extracts containing human DNMT1 were prepared
using EpiQuik Nuclear Extraction Kit (Epigentek) according to the
manufacturer’s protocol. HeLa cells were grown up to 80% con-
fluency in 75 mm2 culture flask. The protein concentration was
determined by Bradford method [46]. The inhibitory activities of
compounds were measured in in vitro enzymatic assay using
EpiQuik DNMT1 Activity/Inhibitor Screening Assay Core Kit
(Epigentek). Analysis is based on ELISA using anti-5-methycytosine
antibodies to detect methylated substrate. According to manu-
facturer’s protocol nuclear extract with substrate was incubated
for 1 h in the presence or absence of inhibitor, then antibody was
added to samples and analyzed colorimetrically in plate reader
(BioTek Synergy 2) at 450 nm. Each experiment was repeated
twice.
5.1.13. 4-N-Arylmetyl-5-methylcytosine
4-N-Arylmethyl-5-methyl-20-deoxycytidine
(27,
(1.25 mmol,1 equiv) was dissolved in the mixture of water:methanol
(15 mL, 2:1, v/v) and (115 L) concentrated hydrochloric acid (3 equiv)
28)
(10)
m
was added. The resulted mixture was refluxed for ca. 4 h and after
cooling toroom temperaturethepHofthe mixturewasadjustedto ca.
7 with 1 M methanol solution of potassium hydroxide. Then the
solvents were removed by evaporation under the reduced pressure.
The residue was partitioned between water (15 mL) and n-butanol
(3ꢁ15 mL), and thenthe active carbon (300 mg, Norit A, 4e7
mm) was
added to the organic layer. The filtered organic layer was evaporated
to give a product as a pale yellow solid.

254
B. Plitta et al. / European Journal of Medicinal Chemistry 55 (2012) 243e254
[16] A. Mai, L. Altucci, Epi-drugs to fight cancer: from chemistry to cancer treat-
ment, the road ahead, Int. J. Biochem. Cell Biol. 41 (2009) 199e213.
[17] N.O. Reich, N. Mashhoon, Inhibition of EcoRI DNA methylase with cofactor
analogs, J. Biol. Chem. 265 (1990) 8966e8970.
[18] J.M. Zingg, J.C. Shen, A.S. Yang, H. Rapoport, P.A. Jones, Methylation inhibitors
can increase the rate of cytosine deamination by (cytosine-5)-DNA methyl-
transferases, Nucleic Acids Res. 16 (1996) 3267e3275.
5.2.3. Inhibition of methylation process in mammalian cell based
assay
HeLa, derived from a cervical carcinoma (Sigma, Munich, Germany)
andHEK293cells,humanembryonickidneycellsderivedfromhealthy
fetus transformed with adenovirus (Sigma) were seeded at a density of
1.25 ꢁ 105 per well in 6-well tissue culture plates. The HeLa cells were
grown in RPMI-1640 medium, HEK293 in DMEM medium (Sigma,
Munich, Germany) supplemented with 10% fetal bovine serum (Gibco,
Paisley, UK),1% antibiotics (Sigma) and 1% RPMI vitamin mix (Sigma) at
37 ꢀC under 5% CO2 atmosphere. Cells were incubated for 72 h with
[19] A. Villar-Garea, M.F. Fraga, J. Espada, M. Esteller, Procaine is
a DNA-
demethylating agent with growth-inhibitory effects in human cancer cells,
Cancer Res. 63 (2003) 4984e4989.
[20] B.H. Lee, S. Yegnasubramanian, X. Lin, W.G. Nelson, Procainamide is a specific
inhibitor of DNA methyltransferase 1, J. Biol. Chem. 280 (2005) 40749e40756.
[21] M. Nieto, E. Samper, M.F. Fraga, G. González de Buitrago, M. Esteller,
M. Serrano, The absence of p53 is critical for the induction of apoptosis by 5-
aza-20-deoxycytidine, Oncogene 23 (2004) 735e743.
inhibitors at 1,10,100 mM concentrations. DNA from HeLa and HEK293
cells was isolated with a commercial kit (A&A Biotechnology, Gdynia,
[22] J. Barciszewski, G.E. Siboska, B.O. Pedersen, B.F. Clark, S.I. Rattan, Evidence for the
presence of kinetin in DNA and cell extracts, FEBS Lett. 393 (1996) 197e200.
[23] J. Barciszewski, G.E. Siboska, B.O. Pedersen, B.F. Clark, S.I. Rattan, A mechanism
for the in vivo formation of N6-furfuryladenine, kinetin, as a secondary
oxidative damage product of DNA, FEBS Lett. 414 (1997) 457e460.
Poland). The analysis of m5C level in DNAwas carried out as previously
described [40]. Dried DNA (1
and 0.02 U of micrococcal nuclease in 20 mM succinate buffer con-
taining 10 mM CaCl2 for 6 h at 37 ꢀC. After that the hydrolysate (0.3
g)
was labeled with 1.6 Ci [
32P]ATP (6000 Ci/mmol Hartmann
mg) was digested with 0.001 U of spleen II
m
ꢀ
[24] E. Wyszko, M.Z. Barciszewska, M. Markiewicz, M. Szymanski,
W.T. Markiewicz, B.F. Clark, J. Barciszewski, “Action-at-a distance” of a new
DNA oxidative damage product 6-furfuryl-adenine (kinetin) on template
properties of modified DNA, Biochim. Biophys. Acta 1625 (2003) 239e245.
[25] E.R. Garrett, J.K. Seydel, A.J. Sharpen, The acid-catalyzed solvolysis of pyrim-
idine nucleosides, J. Org. Chem. 31 (1966) 2219e2227.
[26] W.T. Markiewicz, G. Gröger, R. Rösch, A. Zebrowska, M. Markiewicz, M. Klotz,
P. Godzina, H. Seliger, New method of synthesis of fluorescently labelled
oligonucleotides and their application in DNA sequencing, Nucleic Acid Res.
25 (1997) 3672e3680.
m
g-
Analytic, Braunschweig, Germany) and 1.5 U T4 polynucleotide kinase
in 10 mM bicine-NaOH pH 9.7 buffer containing 10 mM MgCl2,10 mM
DTTand 1 mM spermidine. After incubation for 30 min at 37 ꢀC, 0.03 U
of apyrase in 10 mM bicine-NaOH buffer was added and incubated for
30 min. Next, 0.2
pH 4.5 was used for 30 phosphate cleavage. Analysis of [
done by 2D TLC on cellulose plates (Merck) in isobutyric acid:-
NH4OH:H2O (66:1:17) and 0.1 sodium phosphate pH
m
g of RNase P1 in 500 mM ammonium acetate buffer
g-
32P] m5C was
[27] S. Dincer, Synthesis of some cytosine Schiff bases, Indian J. Chem. B Org. 35B
(1996) 1335e1336.
M
[28] Y. Kawai, K. Uchida, T. Osawa, 20-Deoxycytidine in free nucleosides and
double-stranded DNA as the major target of lipid peroxidation products, Free
Radic. Biol. Med. 36 (2004) 529e541.
6.8eammonium sulfateen-propanol (100 mL/60 g/1.5 mL). Radioac-
tivity was measured with Fluoro Image Analyzer FLA-5100 with Multi
Gauge 3.0 Software. Every analysis was repeated three times.
[29] R.H.E. Hudson, Y. Liu, F. Wojciechowski, Hydrophilic modifications in peptide
nucleic acid synthesis and properties of PNA possessing 5-hydroxymethyluracil
and 5-hydroxymethylcytosine, Can. J. Chem. 85 (2007) 302e312.
[30] W.L. Sung, Chemical conversion of thymidine into 5-methyl-20-deoxycytidine,
J. Chem. Soc. Chem. Commun. (1981) 1089.
Acknowledgments
[31] T.R. Webb, M.D. Matteucci, Hybridization triggered cross-linking of deoxy-
oligonucleotides, Nucleic Acids Res. 14 (1986) 7661e7674.
[32] C. Cristescu, F. Czobor, As-triazine derivatives with potential therapeutic
action. XXVI. Syntheses of 5-substituted-6-azauracil acyclonucleosides,
Nucleosides Nucleotides 17 (1998) 1319e1324.
[33] J. Song, O. Rechkoblit, T.H. Bestor, D.J. Patel, Structure of DNMT1eDNA
complex reveals a role for autoinhibition in maintenance DNA methylation,
Science 331 (2011) 1036e1040.
[34] J.B. Margot, A.M. Aguirre-Arteta, B.V. DiGiacco, S. Pradhan, R.J. Roberts, M.C. Cardoso,
H. Leonhardt, Structure and function of the mouse DNA methyltransferase gene:
Dnmt1 shows a Tripartite structure, J. Mol. Biol. 297 (2000) 293e300.
[35] P. Siedlecki, R.G. Boy, S. Comagic, R. Schirrmacher, M. Wiessler,
P. Zielenkiewicz, S. Suhai, F. Lyko, Establishment and functional validation of
a structural homology model for human DNA methyltransferase, Biochem.
Biophys. Res. Commun. 306 (2003) 558e563.
This work was supported with grant from Polish Ministry of
Science and Education no. PBZ-MNiSW-07/I/2007/11.
References
[1] F. Lyko, R. Brown, DNA methyltransferase inhibitors and the development of
epigenetic cancer therapies, J. Natl. Cancer Inst. 97 (2005) 1498e1506.
[2] H. Gowher, A. Jeltsch, Mechanism of inhibition of DNA methyltransferases by
cytidine analogs in cancer therapy, Cancer Biol. Ther. 3 (2004) 1062e1068.
[3] B. Brueckner, F. Lyko, DNA methyltransferase inhibitors: old and new drugs
for an epigenetic cancer therapy, Trends Pharmacol. Sci. 25 (2004) 551e554.
[4] P.J. Bastian, S. Yegnasubramanian, G.S. Palapattu, C.G. Rogers, X. Lin, A.M. De
Marzo, W.G. Nelson, Molecular biomarker in prostate cancer: the role of CpG
island hypermethylation, Eur. Urol. 6 (2004) 698e708.
[36] E.V. Koudan, J.M. Bujnicki, E.S. Gromova, Homology modeling of the CG-
specific DNA methyltransferase SssI and its complexes with DNA and
AdoHcy, J. Biomol. Struct. Dyn. 22 (2004) 339e345.
[5] P.W. Laird, R. Jaenisch, The role of DNA methylation in cancer genetic and
epigenetics, Annu. Rev. Genet. 30 (1996) 441e464.
[37] M. Dixon, The determination of enzyme inhibitor constants, Biochem. J. 55
(1953) 170e175.
[38] A. Cornish-Bowden, Statistical considerations in the estimation of enzyme
kinetic parameters by the direct linear plot and other methods, Biochem. J.
139 (1974) 721e731.
[6] C. Mund, B. Brueckner, F. Lyko, Reactivation of epigenetically silenced genes
by DNA methyltransferase inhibitors: basic concepts and clinical applications,
Epigenetics 1 (2006) 7e13.
[7] P.A. Jones, S.B. Baylin, The fundamental role of epigenetic events in cancer,
Nat. Rev. Genet. 6 (2002) 415e428.
[39] J. Yoo, J.H. Kim, K.D. Robertsona, J.L. Medina-Franco, Molecular modeling of
human DNA methyltransferase with a crystal structure: discovery of novel
DNMT1 inhibitor, Adv. Protein Chem. Struct. Biol. 87 (2012) 219e247.
[40] M.Z. Barciszewska, A.M. Barciszewska, S.I.S. Rattan, TLC-based detection of
[8] P.A. Jones, Overview of cancer epigenetics, Semin. Hematol. 42 (Suppl. 2)
(2005) S3eS8.
[9] M. Szyf, The DNA methylation machinery as a therapeutic target, Curr. Drug
Targ. 1 (2000) 101e118.
methylated cytosine: application to aging epigenetics, Biogerontology
(2007) 673e678.
8
[10] C.B. Yoo, P.A. Jones, Epigenetic therapy of cancer: past, present and future, Nat.
Rev. Drug Discov. 5 (2006) 37e50.
[41] Y.G. Zheng, J. Wu, Z. Chen, M. Goodman, Chemical regulation of epigenetic
modifications: opportunities for new cancer therapy, Med. Res. Rev. 28 (2008)
645e687.
[11] V.M. Komashko, P.J. Farnham, 5-Azacytidine treatment recognizes genomic
histone modification patterns, Epigenetics 5 (2010) 1e12.
[12] M.Z. Fang, Y. Ai, N. Wang, Z. Hou, Y. Sun, H. Lu, W. Welsh, C.S. Yang, Tea
polyphenol (ꢂ)-epigallocatechin-3-gallate inhibits DNA methyltransferase
and reactivates methylation-silenced genes in cancer cell lines, Cancer Res. 63
(2003) 7563e7570.
[13] H. Nakagawa, K. Hasumi, J.T. Woo, K. Nagai, M. Wachi, Generation of hydrogen
peroxide primarily contributes to the induction of Fe(II)-dependent apoptosis in
Jurkat cells by (ꢂ)-epigallocatechin gallate, Carcinogenesis25 (2004) 1567e1574.
[14] Z. Liu, Z. Xie, W. Jones, R.E. Pavlovicz, S. Liu, J. Yu, P.K. Li, J. Lin, J.R. Fuchs,
G. Marcucci, C. Li, K.K. Chan, Curcumin is a potent DNA hypomethylation
agent, Bioorg. Med. Chem. Lett. 19 (2009) 706e709.
[15] B. Brueckner, R.G. Boy, P. Siedlecki, T. Musch, H.C. Kliem, P. Zielenkiewicz,
S. Suhai, M. Wissler, F. Lyko, Epigenetic reactivation of tumor suppressor
genes by a novel small-molecule inhibitor of human DNA methyltransferases,
Cancer Res. 65 (2005) 6305e6311.
[42] X. Cheng, Structure and function of DNA methyltransferases, Annu. Rev.
Biophys. Biomol. Struct. 24 (1995) 293e318.
[43] J.K. Christman, 5-Azacytidine and 5-aza-20-deoxycytidine as inhibitors of DNA
methylation: mechanistic studies and their implications for cancer therapy,
Oncogene 21 (2002) 5483e5495.
[44] C.A. Lipinski, Drug-like properties and the causes of poor solubility and poor
permeability, J. Pharmacol. Toxicol. Methods 1 (2000) 235e249.
[45] P.B.R. Kumar, M. Soni, B.U. Bhikhalal, I.R. Kakkot, M. Jagadeesh, P. Bommu,
M.I. Nanjan, Analysis of physicochemical properties for drugs from nature,
Med. Chem. Res. 19 (2010) 984e992.
[46] M.M. Bradford, A rapid and sensitive method for the quantitation of micro-
gram quantities of protein utilizing the principle of proteinedye binding,
Anal. Biochem. 72 (1976) 248e254.