Synthesis and biological evaluation of guanidino analogues of roscovitine

Article abstract of DOI:10.1016/j.ejmech.2013.01.021

A series of 2,9-substituted 6-guanidinopurines, structurally related to the cyclin-dependent kinase (CDK) inhibitors olomoucine and roscovitine, has been synthesized and characterized. A new copper-catalyzed method for the synthesis of 2-substituted 6-guanidino-9-isopropylpurines under mild reaction conditions has been developed. All prepared compounds were screened for their CDK1 and CDK2 inhibitory activities, cytotoxicity and antiproliferative effects in the breast cancer-derived cell line MCF7. The most active derivative 16g possessed an identical side chain in the C2 position to roscovitine; this compound displayed approximately five fold higher inhibitory activity towards CDK2/cyclin E and more than ten fold increase in cytotoxicity in MCF7 cells. Interestingly and in contrast to previously described findings, (S)-6-guanidinopurine derivatives were generally more active than their (R)-counterparts. Kinase selectivity profiling of (R)- and (S)-enantiomers 16e and 16g, respectively, revealed that introduction of a guanidino group at the C6 position of the purine moiety decreased selectivity towards protein kinases compared to roscovitine. Nevertheless, increased inhibitory activity and decreased selectivity offer a good starting point for further development of new protein kinase inhibitors.

Full text of DOI:10.1016/j.ejmech.2013.01.021

European Journal of Medicinal Chemistry 62 (2013) 443e452
Contents lists available at SciVerse ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Synthesis and biological evaluation of guanidino analogues
of roscovitine
a
, ,
1
b
,
1
b
b
c
b
ꢀ
*
ꢀ
ꢀ
ꢀ
ꢀ
Iva Dolecková
, Michal Cesnek , Martin Dracinský , Jirí Brynda , Jirí Voller , Zlatko Janeba ,
a
ꢀ
Vladimír Krystof
a
ꢀ
ꢁ
Laboratory of Growth Regulators, Faculty of Science, Palacký University & Institute of Experimental Botany ASCR, Academy of Science, Slechtitelu 11, 783 71 Olomouc, Czech Republic
b Institute of Organic Chemistry and Biochemistry ASCR, v.v.i, Flemingovo nam. 2, 166 10 Praha 6, Czech Republic
c
ꢀ
ꢁ
Centre of the Region Haná for Biotechnological and Agricultural Research, Faculty of Science, Department of Growth Regulators, Palacký University, Slechtitelu 11, 783 71 Olomouc,
Czech Republic
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of 2,9-substituted 6-guanidinopurines, structurally related to the cyclin-dependent kinase (CDK)
inhibitors olomoucine and roscovitine, has been synthesized and characterized. A new copper-catalyzed
method for the synthesis of 2-substituted 6-guanidino-9-isopropylpurines under mild reaction condi-
tions has been developed. All prepared compounds were screened for their CDK1 and CDK2 inhibitory
activities, cytotoxicity and antiproliferative effects in the breast cancer-derived cell line MCF7. The most
active derivative 16g possessed an identical side chain in the C2 position to roscovitine; this compound
displayed approximately five fold higher inhibitory activity towards CDK2/cyclin E and more than ten
fold increase in cytotoxicity in MCF7 cells. Interestingly and in contrast to previously described findings,
(S)-6-guanidinopurine derivatives were generally more active than their (R)-counterparts. Kinase
selectivity profiling of (R)- and (S)-enantiomers 16e and 16g, respectively, revealed that introduction of
a guanidino group at the C6 position of the purine moiety decreased selectivity towards protein kinases
compared to roscovitine. Nevertheless, increased inhibitory activity and decreased selectivity offer
a good starting point for further development of new protein kinase inhibitors.
Received 10 September 2012
Received in revised form
11 January 2013
Accepted 13 January 2013
Available online 22 January 2013
Keywords:
6-Guanidinopurine
Olomoucine
Roscovitine
CDK inhibitors
Mitosis
Ó 2013 Elsevier Masson SAS. All rights reserved.
1. Introduction
rates of proliferation such as cancer. Although mutations of CDKs
have been described in some tumours (CDK4 [2,3], CDK6 [4], CDK9
Cyclin-dependent kinases (CDKs) are specific serine/threonine
protein kinases which play a key role in regulation of the cell cycle,
transcription, apoptosis and differentiation [1]. Their activity is
regulated on multiple levels such as synthesis or degradation of
activating subunit cyclins, phosphorylations, inhibition by specific
proteins or intracellular localization. Overactivation of CDKs is
a frequent feature of human pathologies associated with abnormal
[5]), more frequent is overexpression of cyclins or inactivation of
endogenous CDK inhibitors [4,6]. For these reasons, CDKs have
become a promising target in anticancer research and intensive
search for small molecule inhibitors of CDKs has been encouraged.
Some CDK inhibitors have already been evaluated in clinical trials
[6,7].
In spite of large chemical diversity, the vast majority of CDK
inhibitors are low molecular weight, flat, hydrophobic heterocyclic
molecules that compete with ATP for binding at the kinase
ATP-binding site by both hydrophobic interactions and hydrogen
bonding [8]. The purine skeleton was the first systematically
modified scaffold that led to early discoveries of olomoucine and
roscovitine [9e11] (Fig. 1), followed by synthesis of purvalanols
[12,13], olomoucine II [14] or CR8 [15], DRF053 [16] and other biaryl
derivatives [17,18]. Modification of purine core was another
approach to obtain new potent CDK inhibitors such as pyrazolo[1,5-
a]pyrimidines, pyrazolo[4,3-d]pyrimidines or pyrazolo[1,5-a]pyri-
dines [19]. Dinaciclib is a pyrazolo[1,5-a]pyrimidine which has
Abbreviations: ATP, adenosine 50-triphosphate; CDK, cyclin-dependent kinase;
DABCO, 1,4-diazabicyclo[2.2.2]octane; DMEM, Dulbecco’s modified Eagle’s medium;
DMF, dimethylformamide; DMSO, dimethyl sulfoxide; DTT, dithiothreitol; EDTA, eth-
ylenediaminetetraacetic acid; HEPES, 4-(2-hydroxyethyl)-1-piperazineethanesulfonic
acid; Km, Michaelis constant; NMP, N-methyl-2-pyrrolidone; PEG, polyethylene gly-
col; pH3Ser10, histone H3 phosphorylated on serine 10; RESP, restrained fit to the
electrostatic potential;r.m.s.d.,rootmeansquaredeviation; RT, roomtemperature;SAR,
structureeactivity relationship; TEAB, triethylammoniumbicarbonate; THF, tetrahy-
drofuran; TMS, tetramethylsilane.
* Corresponding author. Tel.: þ420 585 634 854; fax: þ420 585 634 870.
ꢀ
E-mail address: iva.doleckova@centrum.cz (I. Dolecková).
Both authors contributed equally.
1
0223-5234/$ e see front matter Ó 2013 Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.ejmech.2013.01.021

ꢀ
444
I. Dolecková et al. / European Journal of Medicinal Chemistry 62 (2013) 443e452
Fig. 1. Some purine and related CDK inhibitors.
already entered phase III clinical trials for the treatment of chronic
lymphocytic leukaemia (Fig. 1).
of starting 2-amino-6-chloro-9-isopropylpurine in the presence of
CuI according to published procedure [27] and subsequent guani-
dinolysis of obtained 6-chloro-2-iodo-9-isopropylpurine (14) in the
presence of 1,4-diazabicyclo[2.2.2]octane (DABCO) and NaH
(Scheme 1). A catalyst-free reaction of compound 15 with amines at
room temperature (RT) gave no desired products, while at elevated
temperature only decomposition of starting material was observed.
In this way we modified a copper mediated nucleophilic sub-
stitution described by Nair and Sells [28] for the introduction of
substituted amines into the C2 position of the purine moiety. The
All potent purine and related CDK inhibitors contain an aromatic
amine moiety at position 6, such as benzylamino or phenylamino
groups, that (i) binds to the hydrophobic pocket near the active site
of CDKs (and thus deliver their selectivity) and (ii) forms an
essential H-bond to carbonyl of leucine 83 in CDK2 (or to corre-
sponding residues in other CDKs).
Compounds containing the guanidino group such as amino
acids or other natural compounds are widespread throughout
biological systems with a wide spectrum of biological activities
[20]. The ability of the guanidino group to form strong non-covalent
interactions at the active site of enzymes is a key factor in their
biological activities. For this reason, we decided to explore the in-
fluence of the guanidino group at position C6 of the purine and we
prepared a library of compounds consisting of 2,9-disubstituted-6-
guanidinopurines. These compounds were screened for CDK
inhibitory activities and antiproliferative effects in the human
breast cancer-derived cell line MCF7.
Table 1
Structures of prepared derivatives.
2. Chemistry
Firstly, various substituted 2-amino-6-guanidino-9-(hetero)
alkylpurines (Table 1) were synthesized to evaluate the effect of
substitution in the N9 position on the activity of CDKs and to confirm
whether the introduction of a methyl or isopropyl group to 6-
guanidinopurine derivatives is essential for the activity of the
compounds, as in the case of olomoucine and roscovitine, respec-
tively [9,10]. 2-Amino-6-guanidinopurine derivatives 1e5 bearing
the nonpolar aliphatic chains and derivatives 6e9 containing the
polar functional groups were readily prepared by alkylation of
commercially available 2-amino-6-chloropurine with the appro-
priate alkyl halogenides in the presence of NaH and subsequent
guanidinolysis of the obtained intermediates [21]. Compound 10
was prepared by the guanidinolysis of 6-chloro-9-methylpurine. 9-
Alkyl-2,6-diaminopurines 11 and 12 were prepared byammonolysis
of appropriate 2-amino-6-chloropurines and compound 13 was
Compound
R1
R2
R3
1
2
3
4
5
6
7
8
9
10
11
12
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NH2
H
NH2
NH2
Guanidino
Guanidino
Guanidino
Guanidino
Guanidino
Guanidino
Guanidino
Guanidino
Guanidino
Guanidino
NH2
Me
Pr
iPr
iBu
Et
CH2CF3
CH2CH2OH
CH2CH2NH2
CH2CH2N3
Me
iPr
NH2
(Me)2Ne
Et
13
NH2
CH2CH2N3
16a
16b
16c
16d
16e
16f
16g
16h
16i
16j
16k
16l
16m
16n
16o
16p
NH(CH2)2OH
NH(CH2)3OH
NH(CH2)6OH
(R)-OCH2CH(Et)NH2
(R)-NHCH(Et)CH2OH
(S)-OCH2CH(Et)NH2
(S)-NHCH(Et)CH2OH
(R)-OCH2CH(iPr)NH2
(R)-NHCH(iPr)CH2OH
(S)-OCH2CH(iPr)NH2
(S)-NHCH(iPr)CH2OH
Cyclopropylamino
(S)-O-Prolinol
(S)-N-Prolinol
(R)-O-Prolinol
(R)-N-Prolinol
Guanidino
Guanidino
Guanidino
Guanidino
Guanidino
Guanidino
Guanidino
Guanidino
Guanidino
Guanidino
Guanidino
Guanidino
Guanidino
Guanidino
Guanidino
Guanidino
iPr
iPr
iPr
iPr
ꢀ
isolated as a side product as described by Cesnek and co-workers
[21].
iPr
iPr
The SAR study was continued by the derivatization of the active
6-guanidino derivative 3 at the purine position C2. The synthesis of
several 2-O and 2-N derivatives was achieved. The usual synthetic
access to substituted 2-aminopurines was based on the treatment
of appropriate 2-halogenopurine with corresponding alkyl or
arylamine [10,16,22e24]. The coupling reaction of 2-chloro-6-
guanidino-9-isopropylpurine with appropriate amines failed. 6-
Guanidino-2-iodo-9-isopropylpurine (15), as an excellent starting
compound for coupling reactions [25,26], was prepared. The most
reactive intermediate 15 was prepared by the diazotization reaction
iPr
iPr
iPr
iPr
iPr
iPr
iPr
iPr
iPr
iPr
ꢀ
Synthesis of compounds 1e13 was published by Cesnek and co-workers [21].

ꢀ
I. Dolecková et al. / European Journal of Medicinal Chemistry 62 (2013) 443e452
445
Scheme 1. Synthesis of 6-guanidino-9-isopropylpurine derivatives 16aep.
best results were achieved using
2
equivalents of CuI and
3. Pharmacology
10 equivalents of amine in the presence of N,N-diisopropylethyl-
amine at RT.
CDK inhibitoryactivity of the prepared compounds was evaluated
using purified recombinant enzymes CDK1/cyclin B and CDK2/cyclin
E. Cytotoxicity assays were then performed on MCF7 breast cancer
cells. Flow cytometry analysis of the cell cycle showed that several
compounds arrested cells in G2 þ M phases. In an effort to dis-
tinguish between G2 and M phase arrest we examined the intra-
cellular level of histone H3 phosphorylated on serine 10 (pH3Ser10).
This phosphorylation, catalyzed by aurora kinases A and B, is
essential for chromatin condensation during mitosis and is often
detected as a mitotic marker [31,32]. All the data are summarized in
The desired 2-N-substituted purine derivatives were obtained in
moderate yield because the reaction with hydroxylamines was
accompanied by the formation of the corresponding 2-O-substituted
purine isomers as side-products. The yields were also influenced by
the multistep purification process.
Since there seemed to be problem with the solubility of the
starting compounds in tetrahydrofuran (THF), the influence of the
solvent on the reaction course was studied. Further, preferential
formation of undesired O-isomer was observed in THF. On the other
hand, 2-(N,N-dimethylamino) derivative was isolated as a major
by-product of the reactions run in dimethylformamide (DMF).
Finally, N-methyl-2-pyrrolidone (NMP) proved to be the most
convenient solvent for this type of reaction. Compounds 16aep
were prepared in NMP according to the above mentioned proce-
dure (2 equivalents of CuI and 10 equivalents of amine in the
presence of N,N-diisopropylethylamine at RT).
Table 2. CDK inhibitor roscovitine (IC50 values: CDK2 ¼ 0.17
mM,
CDK1 ¼ 2.4
m
M, MCF7 proliferation ¼ 20.2
m
M) inducing specifically
G2 arrest was used as a positive control throughout the study.
4. Results and discussion
4.1. Structureeactivity relationship
We also attempted to use 2-chloro-6-guanidino-9-isopropylpu-
rine (17) as a starting material for the above mentioned amination
reaction both with and without the copper catalyst. This derivative
was prepared by the guanidinolysis [21] of known 2,6-dichloro-9-
isopropylpurine [29]. The reaction of the compound 17 with
amines at RT failed and elevated reaction temperature again caused
only the decomposition of the guanidino group, similar to the
previously described observations with the amido [25] and gua-
nidinopyrimidine derivative [30].
Substitution at position 6 is a critical element determining po-
tency and selectivity of purine derivatives towards CDKs [19,33].
Replacement of benzylamine in the roscovitine molecule e.g. with
biaryl derivatives improved the biological activity and led to dis-
covery of even more potent compounds such as CR8 [15], DRF053
[16], and others [17,18]. The abilityof the guanidino group at position
C6 to stabilize binding interaction with the active site of CDKs can be
demonstrated by the increased activities of 6-guanidinopurines 3
Transformation of copper halogenide into the soluble complex
caused problems with purification of these reaction mixtures. The
guanidino group probably makes a complex with a copper ion.
Classical silica gel purification was not efficient due to the high
polarity of products. The polarity and equal distribution between
organic and water phases also did not allow the application of
extraction using EDTA or dithisone (diphenylthiocarbazone).
Removal of copper from its complex with amines by H2S gas bub-
bled into the reaction mixture was left due to formation of by-
products. High affinity of the guanidino group on the purine moi-
ety to the Dowex 50 ꢀ 8 was used to overcome this problem. After
adsorption of neutralized residue on Dowex 50 ꢀ 8, the copper
residues were removed very efficiently by washing of the column
with aqueous HCl:H2O (1:10). N- and O-Isomers were subsequently
separated on the HPLC column to yield products 16aep.
and 5 (IC50 CDK2 < 2 mM, CDK1 <15 mM) compared to corresponding
6-amino derivatives (11, 12). Similarly, comparison of inactive
compound 13 with a weak CDK2 inhibitor 9 showed that the
replacement of 6-dimethylamino substituent (13) with the guani-
dino group (9) improved CDK inhibitory activity.
As a result, we prepared and evaluated N9-substituted 2-amino-
6-guanidinopurines for their biological activity as was previously
done with other sets of purine derivatives [10,13,34]. Our results
show that the most active compounds of this set in terms of CDK
inhibition possess small nonpolar alkyl substituents in sequence
from the most active ethyl (5) and isopropyl (3) as in the roscovitine
molecule, then methyl (1), propyl (2) and the least active
was compound 4 with isobutyl. Generally, substitution with polar
heteroalkyls led to loss of both CDK inhibitory and cytotoxic
activity.

ꢀ
I. Dolecková et al. / European Journal of Medicinal Chemistry 62 (2013) 443e452
446
Table 2
Biological activities of prepared compounds.
Compound
IC50
[
m
M]
Effects on the cell cycle of MCF7 cells dose 100 mM, 24 h
CDK1/cyclin B
CDK2/cyclin E
MCF7
G2 þ Ma (compound/control)
pH3 (Ser10)b (compound/control)
1
2
3
4
5
6
7
8
9
10
11
12
38.2 ꢁ 5.5
55.9 ꢁ 1.5
14.8 ꢁ 3.3
>100
6.8 ꢁ 2.8
10.7 ꢁ 4.9
1.3 ꢁ 0.2
66.5 ꢁ 30.0
1.7 ꢁ 1.0
>100
>100
1.5 ꢁ 0.7
4.0 ꢁ 1.0
2.0 ꢁ 0.8
3.5 ꢁ 0.6
1.6 ꢁ 0.2
0.8 ꢁ 0.5
0.9 ꢁ 0.3
1.0 ꢁ 0.6
1.4 ꢁ 0.8
1.0 ꢁ 0.4
0.9 ꢁ 0.4
1.2 ꢁ 0.6
1.0 ꢁ 0.4
2.9 ꢁ 1.1
2.5 ꢁ 1.2
2.0 ꢁ 0.8
0.8 ꢁ 0.4
1.8 ꢁ 1.1
1.5 ꢁ 0.4
1.6 ꢁ 0.8
2.1 ꢁ 0.7
2.3 ꢁ 1.0
1.7 ꢁ 0.4
1.8 ꢁ 0.7
2.4 ꢁ 0.6
1.7 ꢁ 0.2
1.7 ꢁ 0.8
1.3 ꢁ 0.5
2.0 ꢁ 0.7
1.5 ꢁ 0.1
0.81 ꢁ 0.09
6.71 ꢁ 2.49
0.18 ꢁ 0.01
7.00 ꢁ 1.00
0.10 ꢁ 0.03
ND
ND
ND
ND
ND
36.5 ꢁ 9.8
97.2 ꢁ 24.2
96.6 ꢁ 38.2
54.3 ꢁ 25.7
>100
>100
>100
>100
>100
12.6 ꢁ 1.3
>100
>100
>50
>100
>100
64.9 ꢁ 17.4
37.5 ꢁ 10.7
>100
15.1 ꢁ 9.2
4.1 ꢁ 0.3
4.2 ꢁ 1.7
47.6 ꢁ 11.0
2.1 ꢁ 0.4
8.1 ꢁ 5.1
0.9 ꢁ 0.5
15.6 ꢁ 6.8
2.0 ꢁ 1.0
1.1 ꢁ 0.3
1.1 ꢁ 0.6
6.4 ꢁ 1.5
>100
>100
>100
74.9 ꢁ 14.1
>100
22.2 ꢁ 17.2
7.9 ꢁ 3.7
>100
>100
>100
>100
ND
ND
ND
13
16a
16b
16c
16d
16e
16f
16g
16h
16i
16j
16k
16l
16m
16n
16o
16p
Roscovitine
0.82 ꢁ 0.34
0.18 ꢁ 0.07
1.6 ꢁ 0.1
13.3 ꢁ 7.0
0.14 ꢁ 0.06
6.2 ꢁ 3.4
0.037 ꢁ 0.031
8.0 ꢁ 3.3
0.14 ꢁ 0.03
1.8 ꢁ 0.8
0.071 ꢁ 0.046
0.50 ꢁ 0.18
25.2 ꢁ 4.8
0.16 ꢁ 0.04
23.6 ꢁ 4.9
0.67 ꢁ 0.44
0.17 ꢁ 0.05
12.4 ꢁ 6.4
8.8 ꢁ 3.3
5.2 ꢁ 3.1
>100
4.5 ꢁ 1.3
16.3 ꢁ 2.9
1.6 ꢁ 0.3
79.4 ꢁ 22.1
10.5 ꢁ 3.8
3.4 ꢁ 0.7
2.1 ꢁ 0.4
5.4 ꢁ 0.2
>100
0.15 ꢁ 0.02
0.18 ꢁ 0.02
0.33 ꢁ 0.01
1.08 ꢁ 0.21
0.08 ꢁ 0.06
0.07 ꢁ 0.01
0.09 ꢁ 0.03
0.39 ꢁ 0.06
0.13 ꢁ 0.09
0.09 ꢁ 0.07
0.09 ꢁ 0.08
0.17 ꢁ 0.03
0.80 ꢁ 0.10
0.11 ꢁ 0.08
ND
1.5 ꢁ 0.1
81.5 ꢁ 27.0
6.2 ꢁ 2.2
2.4 ꢁ 0.94
6.2 ꢁ 3.3
>100
20.4 ꢁ 16.8
20.2 ꢁ 12.3
0.10 ꢁ 0.07
0.11 ꢁ 0.01
a
Ratio of G2 þ M populations in treated cells and G2 þ M populations in control cells.
b
Ratio of pH3-positive cells treated and pH3-positive control cells.
In spite of CDK inhibitory activities of several N9-substituted 2-
amino-6-guanidinopurines, the majority had no significant cyto-
toxic effect on MCF7 cells (IC50 > 100 M) or only moderate
(IC50 > 50 M). Nonetheless, some were still able to accumulate
a
methylene group (16b) increased CDK inhibitory activity
approximately 4-times, but no significant increase in cytotoxicity
was observed (Table 2). Further, elongation by methylene groups
(16c) did not potentiate the biological activity.
m
m
cells in G2 þ M phases (compounds 2e5). In comparison with
roscovitine, derivatives 3 and 5 were weaker CDK inhibitors with
IC50 at least 5-times higher and their cytotoxicity was also lower
(4ꢀ and 2ꢀ, respectively). However, these compounds induced
increase in the G2 þ M population (2ꢀ and 1.6ꢀ, respectively) and
more than 5-times decrease in mitotic marker pH3Ser10 in a similar
manner to G2-block induced by roscovitine. On the other hand,
propyl (2) and isobutyl (4) derivatives caused an increase in the
mitotic cell subpopulation as evidenced by a strong (7-times)
accumulation of pH3Ser10. It is interesting that such a small change
in the structure of the N9 substituent apparently leads to alter-
native mechanisms of action resulting in different responses in
biological systems.
Having confirmed isopropyl as one of the most active sub-
stituents in the group of various N9 substituents, we synthesized
substituted 6-guanidino-9-isopropylpurines and investigated the
effects of the introduction of various substituents related to the
olomoucine or roscovitine C2 side chain to position C2. The
importance of the C2 substituent was demonstrated by comparison
of the biological activities of 2-amino-6-guanidino-9-methylpurine
(1), which inhibited CDKs in micromolar range, and inactive, C2
unsubstituted 6-guanidino-9-methylpurine (10). Compound 16a
containing 2-hydroxyethylamino group, the olomoucine C2 side
chain, did not significantly improve the CDK inhibitory activity of
the corresponding 2-amino derivative (3), but it potentiated its
antiproliferative properties towards MCF7 almost 8-times. This
observation suggests that the increased potency is not due to in-
hibition of CDKs. Elongation of olomoucine C2 side chain by
A set of 6-guanidinopurines with various N- or O-linked C2
substituents was then prepared. Generally, derivatives with the
high CDK inhibitory activity tended to be more active against MCF7.
This trend was observed for both 2-N- and 2-O-subgroups. This
observation suggests that at least part of the activity is due to CDK
inhibition. 2-O-Substituted derivatives were several times less
potent than their 2-N-regioisomers. The only exception was 16j,
one of the most active compounds of our set. Although the 25ꢀ less
potent CDK2 inhibitor than its N-regioisomer 16k, its effects on
CDK1, MCF7 cells and on the cell cycle were surprisingly nearly
identical.
Compound 16g was the most active 6-guanidinopurine bearing
the same substituent at position C2 as roscovitine (in addition to
the roscovitine N9 substituent). This compound was more active in
CDKs (2.7ꢀ for CDK1 and 4.6ꢀ for CDK2) as well as on MCF7 cells
(IC50 more than 10ꢀ lower) than roscovitine itself. In contrast to
roscovitine, which is a more potent CDK inhibitor as an (R)-enan-
tiomer [35], (S)-enantiomer (16g) seems to be more active than its
(R)-counterpart 16e. Similarly, the other guanidinopurines in (S)-
configuration also proved to be more active in CDK inhibition as
well as in the antiproliferative effects in MCF7 cells. We found the
same with 2-N-enantiomers 16e and 16g, 16i and 16k as well as 2-
O-enantiomers 16d and 16f, 16h and 16j.
The presence of a cyclopropylamino group at position C2 (16l)
improved both CDK inhibitory (about 2.5-times) and cytotoxic ac-
tivity (almost 20-times) in comparison with the similar 2-amino
derivative 3. (S) and (R)-Prolinol were other cyclic C2 substituents
tested. In a group of various prolinol isomers and enantiomers at

ꢀ
I. Dolecková et al. / European Journal of Medicinal Chemistry 62 (2013) 443e452
447
Table 3
position C2 (16mep) we observed that the most active compound
16n possessed N-linked (S)-prolinol. The biological activity of this
derivative was very similar to that of roscovitine.
Selectivity of compounds 16e and 16g against a panel of recombinant protein
kinases. The compounds were screened at 1 M concentration in duplicate.
m
Kinase
Remaining
activity [%]
Kinase
Remaining
activity [%]
4.2. Kinase selectivity
16e
16g
6 ꢁ 3
16e
16g
75 ꢁ 4
The most active compounds 16e and 16g were further evaluated
for selectivity on a panel of protein kinases. As selectivity and po-
tency are usually inversely correlated for kinase inhibitors [36] it
was not surprising that both these derivatives were found to inhibit
a broad range of various kinase targets (Table 3). In comparison,
roscovitine has been shown to be more selective, inhibiting only
ABL
AMPK
ASK1
AURA
AURB
BRK
BRSK1
BRSK2
BTK
CAMK1
CAMKKb
CDK2/cyclin A
CK1
5 ꢁ 0
29 ꢁ 1
100 ꢁ 6
39 ꢁ 1
47 ꢁ 12
50 ꢁ 15
44 ꢁ 2
77 ꢁ 12
53 ꢁ 1
35 ꢁ 6
17 ꢁ 0
5 ꢁ 1
MINK1
MKK1
MKK2
MKK6
MLK1
64 ꢁ 0
41 ꢁ 0
53 ꢁ 4
72 ꢁ 12
25 ꢁ 5
37 ꢁ 7
173 ꢁ 15
119 ꢁ 8
36 ꢁ 2
46 ꢁ 1
35 ꢁ 5
42 ꢁ 19
73 ꢁ 11
95 ꢁ 4
8 ꢁ 1
22 ꢁ 1
81 ꢁ 7
30 ꢁ 4
70 ꢁ 3
44 ꢁ 9
45 ꢁ 18
67 ꢁ 12
48 ꢁ 11
24 ꢁ 5
13 ꢁ 1
7 ꢁ 1
28 ꢁ 9
39 ꢁ 8
88 ꢁ 2
31 ꢁ 10
37 ꢁ 14
178 ꢁ 10
122 ꢁ 7
28 ꢁ 9
38 ꢁ 3
43 ꢁ 8
40 ꢁ 18
57 ꢁ 6
85 ꢁ 4
7 ꢁ 3
MLK3
MNK1
MNK2
MPSK1
MSK1
MST2
MST4
NEK2a
NEK6
a few kinases besides CDKs as CaM Kinase 2, CK1a, CK1d, DYRK1A,
EPHB2, ERK1, ERK2, FAK and IRAK4 as published previously [35,37].
The reason for this decreased selectivity is probably the removal of
aromatic substitution at C6 that was designed as a key determinant
of CDK selectivity [33]. In accordance with our other data, (S)-
enantiomer 16g inhibited selected kinases more efficiently than its
(R)-enantiomer 16e.
15 ꢁ 6
43 ꢁ 1
4 ꢁ 2
21 ꢁ 2
33 ꢁ 4
4 ꢁ 2
CK2
CLK2
CSK
NUAK1
OSR1
58 ꢁ 6
54 ꢁ 8
1 ꢁ 0
59 ꢁ 0
36 ꢁ 5
1 ꢁ 0
16 ꢁ 2
87 ꢁ 14
64 ꢁ 3
37 ꢁ 9
77 ꢁ 2
45 ꢁ 1
15 ꢁ 2
18 ꢁ 1
26 ꢁ 3
68 ꢁ 10
14 ꢁ 2
85 ꢁ 10
88 ꢁ 7
64 ꢁ 10
90 ꢁ 3
99 ꢁ 8
111 ꢁ 19
71 ꢁ 1
68 ꢁ 2
73 ꢁ 1
17 ꢁ 1
108 ꢁ 4
76 ꢁ 16
85 ꢁ 22
72 ꢁ 5
61 ꢁ 23
18 ꢁ 1
35 ꢁ 7
49 ꢁ 2
60 ꢁ 1
49 ꢁ 3
22 ꢁ 22
22 ꢁ 1
11 ꢁ 0
58 ꢁ 6
7 ꢁ 2
16 ꢁ 1
82 ꢁ 17
66 ꢁ 9
23 ꢁ 2
71 ꢁ 7
38 ꢁ 5
12 ꢁ 0
15 ꢁ 3
17 ꢁ 5
62 ꢁ 18
12 ꢁ 0
71 ꢁ 20
91 ꢁ 5
44 ꢁ 9
83 ꢁ 6
107 ꢁ 24
103 ꢁ 8
68 ꢁ 14
61 ꢁ 10
61 ꢁ 7
18 ꢁ 1
108 ꢁ 19
74 ꢁ 23
84 ꢁ 34
76 ꢁ 15
65 ꢁ 18
12 ꢁ 4
17 ꢁ 1
36 ꢁ 3
51 ꢁ 15
32 ꢁ 2
6 ꢁ 3
DAPK1
DYRK1A
DYRK2
DYRK3
EF2K
EIF2AK3
EPH-A2
EPH-A4
EPH-B1
EPH-B2
EPH-B4
ERK1
ERK2
ERK8
FGF-R1
GCK
GSK3b
HER4
p38a MAPK
p38b MAPK
p38d MAPK
p38g MAPK
PAK2
PAK4
PAK5
PAK6
PDK1
PHK
PIM1
PIM2
PIM3
PKA
PKBa
PKBb
PKCa
4.3. Docking study
27 ꢁ 3
18 ꢁ 2
94 ꢁ 1
17 ꢁ 1
22 ꢁ 5
22 ꢁ 6
8 ꢁ 0
16 ꢁ 1
11 ꢁ 3
94 ꢁ 6
11 ꢁ 0
17 ꢁ 8
41 ꢁ 13
7 ꢁ 2
In order to acquire insight into the location and molecular
interaction of the most active compounds in the binding site,
a docking study was performed. When we compared the X-ray
structure of CDK2/roscovitine complex (PDB code 2A4L) with the
results of docking of two enantiomer pairs 16e/16g and 16i/16k, we
found that the position of the purine moiety is conserved as well as
the orientation of isopropyl group at position N9 (see Fig. 2A, B).
Hydrogen bonding of adenine moiety is also conserved and in
comparison to roscovitine is extended by one hydrogen bond
donated by a guanidine secondary amino group to the carbonyl
oxygen of leucine 83. Interestingly, the docking study of compounds
16e, 16g, 16i and 16k showed that modification on C2 by [1-(1S)-
(hydroxymethyl)propyl]amino (compound 16g) gave a superior
binding due tothe two hydrogen bonds of the hydroxyl, one donated
to the carboxy group of aspartate 145 and one accepted from side
chain amino group of lysine 33 (Fig. 3).
2 ꢁ 2
2 ꢁ 1
6 ꢁ 2
6 ꢁ 4
50 ꢁ 3
32 ꢁ 4
16 ꢁ 8
4 ꢁ 0
28 ꢁ 7
15 ꢁ 3
18 ꢁ 6
5 ꢁ 0
15 ꢁ 2
37 ꢁ 1
27 ꢁ 2
15 ꢁ 1
8 ꢁ 0
10 ꢁ 1
28 ꢁ 12
29 ꢁ 12
11 ꢁ 2
6 ꢁ 1
PKCz
PKCg
PKD1
PLK1
HIPK1
HIPK2
HIPK3
CHK1
38 ꢁ 6
58 ꢁ 8
59 ꢁ 8
10 ꢁ 2
74 ꢁ 6
66 ꢁ 8
5 ꢁ 0
24 ꢁ 0
46 ꢁ 16
42 ꢁ 15
5 ꢁ 0
PRAK
PRK2
RIPK2
ROCK 2
RSK1
RSK2
S6K1
SGK1
SMMLCK
SRC
SRPK1
STK33
SYK
TAK1
TAO1
TBK1
TIE2
CHK2
IGF-1R
IKKb
IKKe
76 ꢁ 12
52 ꢁ 14
3 ꢁ 0
5. Conclusion
IR
In this study we evaluated a library of 2,9-disubstituted-6-
guanidinopurines, structurally related to the CDK inhibitors olo-
moucine and roscovitine. From this series, 2-substituted 6-
guanidino-9-isopropylpurines were newly synthesized. SAR anal-
ysis proved that small changes in the structure of the N9 substit-
uent yielded compounds with a different mode of action. Although
the majority of the compounds arrested cells in the G2 phase of the
cell cycle in a similar manner to roscovitine, a few of them induced
significant accumulation of mitotic cells. Moreover, in contrast to
roscovitine, our compounds were more active in (S)-configuration
as supported by our docking study. The introduction of the guani-
dino group at position C6 increased the inhibitory potential to-
wards protein kinase targets several times more than with
roscovitine but with a concomitant decrease in selectivity. Never-
theless increased potency and decreased selectivity offer a good
starting point for further development of new, more selective
protein kinase inhibitors.
IRAK1
IRAK4
IRR
JAK2
JNK1
JNK2
JNK3
LCK
LKB1
MAPKAP-K2
MAPKAP-K3
MARK1
MARK2
MARK3
MARK4
MEKK1
MELK
26 ꢁ 1
21 ꢁ 1
19 ꢁ 2
16 ꢁ 3
88 ꢁ 8
85 ꢁ 4
96 ꢁ 20
13 ꢁ 2
13 ꢁ 4
41 ꢁ 9
80 ꢁ 18
31 ꢁ 7
45 ꢁ 16
24 ꢁ 3
38 ꢁ 3
91 ꢁ 12
27 ꢁ 2
17 ꢁ 6
11 ꢁ 0
16 ꢁ 2
16 ꢁ 8
78 ꢁ 7
73 ꢁ 19
80 ꢁ 19
12 ꢁ 3
16 ꢁ 9
48 ꢁ 27
78 ꢁ 4
39 ꢁ 0
62 ꢁ 24
17 ꢁ 7
29 ꢁ 8
96 ꢁ 5
19 ꢁ 2
14 ꢁ 1
9 ꢁ 3
50 ꢁ 4
13 ꢁ 10
9 ꢁ 3
9 ꢁ 2
92 ꢁ 1
39 ꢁ 12
9 ꢁ 1
90 ꢁ 45
40 ꢁ 14
9 ꢁ 5
TLK1
TRKA
TTK
VEGFR
YES1
ZAP70
24 ꢁ 5
40 ꢁ 1
6 ꢁ 1
28 ꢁ 14
32 ꢁ 4
4 ꢁ 0
13 ꢁ 2
86 ꢁ 15
11 ꢁ 4
109 ꢁ 54
6. Experimental protocols
recorded on Bruker Avance 500 (1H at 500 MHz, 13C at 125.8 MHz)
and Bruker Avance 400 (1H at 400 MHz, 13C at 100.6 MHz) spec-
trometerswith TMS as internal standard or referenced totheresidual
solvent signal. Mass spectra were measured on a ZAB-EQ (VG Ana-
lytical) spectrometer. The chemicals were obtained from commercial
sources (SigmaeAldrich) or prepared according to published
6.1. General chemical procedures
Unless otherwise stated, solvents were evaporated at 40 ꢂC/2 kPa,
and the compounds were dried over P2O5 at 2 kPa. NMR spectra were

ꢀ
I. Dolecková et al. / European Journal of Medicinal Chemistry 62 (2013) 443e452
448
Fig. 2. Overlay of ligands poses and X-ray structure of CDK2eroscovitine complex, based on CDK2 superposition: (A) view perpendicular and (B) parallel to purine plane. Colour
coding: carbon atoms green for roscovitine, magenta for 16g, light-blue for 16e, orange for 16k and light-yellow for 16i; nitrogen atoms blue, oxygen atoms red.
procedures. DMFandacetonitrilewere distilled from P2O5 andstored
over molecular sieves (4 A). Preparative HPLC purifications were
performed on columns packed with 7 mm C18 reversed phase resin
residue was dissolved in MeOH (150 ml) and filtered through the
Celite pad once again. The filtrate was adsorbed on SiO2 and
purified with column chromatography (HUB III); yield 0.7 g, 65%.
FABMS: 345.9 [MHþ] (50). 1H NMR (DMSO-d6): 1.48 (d, 6H, J(CH3,
CH) ¼ 6.8, CH3); 4.65 (sept, 1H, J(CH, CH3) ¼ 6.8, NeCH); 7.40 (br,
4H, NH); 8.13 (s, 1H, H-8). 13C NMR (DMSO-d6): 22.36 (2C, CH3);
46.45 (NeCH); 118.32 (C-5); 125.28 (C-2); 138.91 (C-8); 150.24
(C-4); 159.16 and 159.74 (NeC and C-6). For C9H12IN7 (345.14)
calcd: C, 31.32; H, 3.50; I, 36.77; N, 28.41. Found: C, 31.55; H, 3.59; I,
36.67; N, 28.46.
ꢁ
(Waters Delta 600 chromatograph column), 17 ꢀ 250 mm; in ca.
200 mg batches of mixtures using gradient MeOH/H2O as eluent or
using a 0.1 mol triethylammoniumbicarbonate (TEAB) as a buffer.
Deionization was performed on Dowex 50 ꢀ 8 (Hþ-form) columns
by the following procedure: after application of crude product the
column was washed with water until the UV absorption dropped.
Thereafter, the column was eluted with 2.5% aqueous NH3. Chro-
matography on Dowex 1 ꢀ 2 (acetate form) was as follows: after
application of the aqueous solution of the crude product onto the
column, it was washed with water until the UV absorption dropped.
The column was then eluted with a gradient of dilute acetic or formic
acid (0e1 M). All tested compounds were characterized by 1H NMR,
13C NMR and mass spectrometry. Representative examples of 1H and
13C spectra are provided as supplementary material (Figs. S1eS12).
6.1.2. Generalprocedureforpreparationof2-substituted6-guanidino-
9-isopropyl-9H-purines
N,N-Diisopropylethylamine (0.48 ml; 2.8 mmol) and appro-
priate amine (14 mmol) were added to a mixture of compound 15
(0.5 g; 1.4 mmol) and CuI (0.52 g; 2.8 mmol) in NMP (7 ml) under
Ar. The reaction mixture was stirred at RT by the time until the
starting compound disappeared. The reaction mixture was neu-
tralized with a solution of HCl in DMF and evaporated, co-distilled
by DMF and subsequently with toluene. The residue was dissolved
in 20% aqueous MeOH (25 ml), applied on column of Dowex
(50 ꢀ 8) and washed subsequently with 20% aqueous MeOH and
water. Impurities (mainly copper salts) were eluted with 10%
aqueous HCl and the column was then washed with water to
remove the acid. The elution was made with 10% aqueous Et3N (20%
MeOH in H2O). The collected UV absorbing fractions were evapo-
rated and purified by HPLC column chromatography to give desired
products in moderate yields.
6.1.1. 6-Guanidino-2-iodo-9-isopropyl-9H-purine (15)
A filtered solution of guanidine [38] in NMP (15.5 ml,15.5 mmol)
was added to a mixture of 6-chloro-2-iodo-isopropyl-9H-purine
[27] (1 g, 3.1 mmol) and DABCO (0.35 g, 3.1 mmol). The reaction
mixture was stirred at RT for 4 h. The resulting mixture was
neutralized with aqueous HCl, the solids were removed by
filtration. The filtrate was evaporated in vacuo, co-distilled with
DMF (3 ꢀ 50 ml) and subsequently with toluene (3 ꢀ 50 ml). The
6.1.3. 6-Guanidino-2-[(2-hydroxyethyl)amino]-9-isopropyl-9H-purine
(16a)
General procedure; ethanolamine (1.5 ml, 24 mmol); 2 h, RT;
yield 0.41 g, 69%; FABMS: 279.1 [MHþ] (100). 1H NMR (DMSO-d6):
1.43 (d, 6H, J(CH3, CH) ¼ 6.7, CH3); 3.32 (brq, 2H, OeCH2); 3.55 (t,
2H, J(CH2, CH2) ¼ 5.7, NeCH2); 4.56 (sept, 1H, J(CH, CH3) ¼ 6.7, Ne
CH); 4.70 (brt,1H, OH); 6.60 (brs,1H, NH) and 7.80 (br, 4H, NH); 7.81
(s, 1H, H-8). 13C NMR (DMSO-d6): 22.21 (2C, CH3); 44.20 (NeCH2);
45.75 (NeCH); 60.39 (OeCH2); 118.90 (C-5); 136.02 (C-8); 151.88
(C-4); 158.09, 2C, (C-6; C-2); 159.25 (NeC). For C11H18N8O-
$EtOH$HCl (278.31) calcd: C, 43.27; H, 6.98; N, 31.05; Cl, 9.81.
Found: C, 43.29; H, 6.75; N, 30.85; Cl, 10.12.
6.1.4. 6-Guanidino-2-[(3-hydroxypropyl)amino]-9-isopropyl-9H-
purine (16b)
General procedure; 3-aminopropan-1-ol (2.48 ml, 32.6 mmol);
5 h; RT; yield 0.46 g, 57%; FABMS: 293.2 [MHþ] (100). 1H
NMR (DMSO-d6): 1.42 (d, 6H, J(CH3, CH) ¼ 6.8, CH3); 1.50 (m, 2H,
CH2-20); 3.19 (bq, 2H, J(10e20) ¼ J(10eNH) ¼ 6.7 CH2-10); 3.32 (m, 2H,
Fig. 3. Complex of CDK2 with compound 16g, detail of active site. CDK2
shown in cartoon representation, ligand in stick representation. Colour coding:
nitrogen atoms blue, oxygen atoms red, hydrogen atoms white, carbon atoms
grey for CDK2 and magenta for 16g. Hydrogen bonds are depicted by green
dash-line.

ꢀ
I. Dolecková et al. / European Journal of Medicinal Chemistry 62 (2013) 443e452
449
OeCH2); 4.50 (br, 1H, J(CH, CH3) ¼ 6.8, NeCH); 4.60 (brs, 1H, OH);
6.40 (br, 1H, NH); 7.31 (vbs, 4H, NH); 7.75 (s, 1H, H-8). 13C NMR
(DMSO-d6): 22.25 (2C, CH3); 29.85 (C-20); 42.45 (C-10); 45.60 (Ne
CH); 60.50 (C-30); 119.45 (C-5); 135.90 (C-8); 151.80 (C-4); 158.10
(C-2); 159.50 (NeC); 159.70 (C-6). For C12H20N8O$1/3H2O$1/3 HCl
(292.34) calcd: C, 46.42; H, 6.82; N, 36.09; Cl 3.81. Found: C, 46.37;
H, 6.71; N, 35.92; Cl 4.18.
Compound 16g: yield 0.2 g, 29%; FABMS: 307 [MHþ] (100).
D ꢃ33.3ꢂ (c 0.21, methanol); 1H NMR (DMSO-d6): 0.89 (t, 3H, J(40,
[a]
30) ¼ 7.4, CH3-40); 1.45 (d, 3H, J(CH3, CH) ¼ 6.8, CH3); 1.46 (d, 3H,
J(CH3, CH) ¼ 6.8, CH3); 1.46 (m, 1H, 30b); 1.65 (m, 1H, 30a); 3.38 (dd,
1H, J(10be20) ¼ 5.9, Jgem ¼ 10.6, 10b); 3.49 (dd, 1H, J(10ae20) ¼ 5.0,
Jgem ¼ 10.5, 10a); 4.52 (br, 1H, J(CH, CH3) ¼ 6.8, NeCH); 4.63 (brs,
1H, OH); 6.12 (brs, 1H, NH); 7.25 (vbs, 4H, NH); 7.71 (s, 1H, H-8). 13
C
NMR (DMSO-d6): 10.87 (C-40); 22.21 and 22.27 (2C, CH3); 24.18 (C-
30); 45.63 (NeCH); 54.37 (C-20); 63.30 (C-10); 119.84 (C-5); 135.32
(C-8); 151.67 (C-4); 158.06 (C-2); 159.82 (NeC); 160.17 (C-6). For
6.1.5. 6-Guanidino-2-[(6-hydroxyhexyl)amino]-9-isopropyl-9H-
purine (16c)
General procedure; 6-amino-1-hexanol (1.99 g, 17 mmol); 24 h;
not fully converted, RT; yield 0.13 g, 25%; FABMS: 335 [MHþ] (100).
1H NMR (DMSO-d6): 1.32 (m, 4H, CH2-30,40); 1.42 (m, 2H, CH2-50);
1.45 (d, 6H, J(CH3, CH) ¼ 6.8, CH3); 1.53 (m, 2H, CH2-20); 3.20 (bq,
2H, J(10e20) ¼ J(10eNH) ¼ 6.6 CH2-10); 3.37 (m, 2H, CH2-60); 4.36
(brs, 1H, OH); 4.54 (br, 1H, J(CH, CH3) ¼ 6.8, NeCH); 6.46 (brt, 1H,
J(NH-10) ¼ 5.3, NH); 7.3 (vbs, 4H, NH); 7.73 (s, 1H, H-8). 13C NMR
(DMSO-d6): 22.30 (2C, CH3); 25.65 (C-40); 26.84 (C-30); 29.54 (C-20);
32.80 (C-50); 41.36 (C-10); 45.62 (NeCH); 60.94 (C-60); 119.62 (C-5);
135.47 (C-8); 151.88 (C-4); 158.17 (C-2); 159.74 (NeC); 159.84 (C-6).
For C15H26N8O$MeOH (334.42) calcd: C, 52.44; H, 8.25; N, 30.58.
Found: C, 52.84; H, 7.88; N, 30.63.
C12H20N8O$½ HCl (306.37) calcd: C, 48.10; H, 6.99; N, 34.52. Found:
C, 48.40; H, 7.01; N, 34.29.
6.1.8. 2-{[(2R)-2-Amino-3-methylbutyl]oxy}-6-guanidino-9-isopro-
pyl-9H-purine (16h) and 6-guanidino-2-{[(1R)-1-(hydroxymethyl)-2-
methylpropyl]amino}-9-isopropyl-9H-purine (16i)
General procedure; (2R)-(ꢃ)-2-amino-3-methyl-1-butanol (1 g,
10 mmol); 10 h; RT; compound 16h: yield 0.09 g, 28%; FABMS: 321.0
[MHþ] (40). [
a
]
D ꢃ9.4ꢂ (c 0.19, methanol); 1H NMR (DMSO-d6): 0.89
(d, 3H, J(40,30) ¼ 6.8, CH3); 0.93 (d, 3H, J(40,30) ¼ 6.8, CH3); 1.48 (d,
6H, J(CH3, CH) ¼ 6.8, CH3); 1.72 (m, 1H, CH-30); 2.83 (m, 1H, CH-20);
3.99 (dd, 1H, J(10be20) ¼ 7.1, Jgem ¼ 10.4, 10b); 4.19 (dd, 1H, J(10ae
20) ¼ 5.2, Jgem ¼ 10.4, 10a); 4.61 (br, 1H, J(CH, CH3) ¼ 6.8, NeCH);
7.30 (vbs, 4H, NH); 7.96 (s, 1H, H-8). 13C NMR (DMSO-d6): 17.52
and 19.74 (C-40); 22.29 (2C, CH3); 30.34 (C-30); 46.15 (NeCH); 55.07
(C-20); 70.10 (C-10); 122.08 (C-5); 137.60 (C-8); 151.08 (C-4); 159.92
6.1.6. 2-[2-(2R)-(Aminomethyl)propoxy]-6-guanidino-9-isopropyl-
9H-purine (16d) and 6-guanidino-2-{[1-(1R)-(hydroxymethyl)propyl]
amino}-9-isopropyl-9H-purine (16e)
General procedure; R-(ꢃ)-2-amino-1-butanol (1.32 ml,14 mmol);
(NeC); 160.21 (C-2); 160.75 (C-6). For C16h24N8O$2/3 MeOH
6 h; RT; compound 16d: yield 0.2 g, 45%; FABMS: 307.0 [MHþ] (20).
(320.39) calcd: C, 51.54; H, 7.86; N, 32.79. Found: C, 51.35; H, 7.59;
N, 33.09.
[a]
D ꢃ5.4ꢂ (c 0.34, methanol); 1H NMR (DMSO-d6): 0.93 (t, 3H, J(40,
30) ¼ 7.5, CH3-40); 1.28 (m,1H, CH2-30b); 1.48 (d, 6H, J(CH3, CH) ¼ 6.8,
CH3); 1.52 (m, 1H, 30a); 2.90 (m, 1H, CH-20); 3.98 (dd, 1H, J(10be
20) ¼ 6.6, Jgem ¼ 10.2, 10b); 4.07 (dd, 1H, J(10ae20) ¼ 5.4, Jgem ¼ 10.2,
10a); 4.61 (br, 1H, J(CH, CH3) ¼ 6.7, NeCH); 7.3 (vbs, 4H, NH); 7.96 (s,
1H, H-8).13C NMR (DMSO-d6): 10.60 (C-40); 22.31 (2C, CH3); 26.97 (C-
30); 46.11 (NeCH); 51.63 (C-20); 71.71 (C-10); 122.07 (C-5); 137.58 (C-
8); 151.08 (C-4); 159.93 (NeC); 160.20 (C-2); 160.75 (C-6). For
Compound 16i: yield 0.1 g, 31%; FABMS: 321.0 [MHþ] (100).
[
a
]
þ37.5ꢂ (c 0.35, methanol); 1H NMR (DMSO-d6): 0.90 (d, 3H,
J(40D,30) ¼ 6.9, CH3); 0.91 (d, 3H, J(40,30) ¼ 6.9, CH3); 1.45 (d, 3H, J(CH3,
CH) ¼ 6.9, CH3); 1.46 (d, 3H, J(CH3, CH) ¼ 6.8, CH3); 1.97 (m, 1H, CH-
30); 3.49 (m, 2H, CH2-10); 3.75 (brs, 1H, CH2-20); 4.52 (br, J(CHe
CH3) ¼ 6.8, CHeN); 4.58 (brs, 1H, OH); 6.13 (br, 1H, NH); 7.4
(vbs, 4H, NH); 7.71 (s,1H, H-8). 13C NMR (DMSO-d6): 18.55 and 19.83
(C-40); 22.18 and 22.33 (2C, CH3); 28.82 (C-30); 45.71 (NeCH); 57.71
(C-20); 61.67 (C-10); 119.74 (C-5); 135.38 (C-8); 151.67 (C-4); 158.42
(C-2); 159.76 (NeC); 159.97 (C-6). For C16h24N8O$1/3 EtOH$1/3 HCl
(320.39) calcd: C, 50.63; H, 7.63; N, 32.21. Found: C, 50.61; H, 7.50; N,
32.55.
C
13H22N8O$MeOH (306.37) calcd: C, 49.69; H, 7.74; N, 33.11. Found: C,
49.82; H, 7.39; N, 33.53.
Compound 16e: yield 0.15 g, 33%; FABMS: 307.0 [MHþ] (100).
[
a]
D þ32.3ꢂ (c 0.38, methanol); 1H NMR (DMSO-d6): 0.89 (t, 3H, J(40,
30) ¼ 7.4, CH3-40); 1.45 (d, 3H, J(CH3, CH) ¼ 6.8, CH3); 1.46 (d, 3H,
J(CH3, CH) ¼ 6.8, CH3); 1.46 (m, 1H, 30b); 1.65 (m, 1H, 30a); 3.38 (dd,
1H, J(10be20) ¼ 5.9, Jgem ¼ 10.6, 10b); 3.49 (dd, 1H, J(10ae20) ¼ 5.0,
Jgem ¼ 10.5, 10a); 4.52 (br, 1H, J(CH, CH3) ¼ 6.8, NeCH); 4.63 (brs,
6.1.9. 2-{[(2S)-2-Amino-3-methylbutyl]oxy}-6-guanidino-9-isopro-
pyl-9H-purine (16j) and 6-guanidino-2-{[(1S)-1-(hydroxymethyl)-2-
methylpropyl]amino}-9-isopropyl-9H-purine (16k)
1H, OH); 6.12 (brs, 1H, NH); 7.25 (vbs, 4H, NH); 7.71 (s, 1H, H-8). 13
C
NMR (DMSO-d6): 10.87 (C-40); 22.21 and 22.27 (2C, CH3); 24.18 (C-
30); 45.63 (NeCH); 54.37 (C-20); 63.30 (C-10); 119.84 (C-5); 135.32
(C-8); 151.67 (C-4); 158.06 (C-2); 159.82 (NeC); 160.17 (C-6). For
General procedure; (2S)-(ꢃ)-2-amino-3-methyl-1-butanol (1 g,
10 mmol); 10 h; RT; compound 16j: yield 0.12 g, 38%; FABMS: 321.0
[MHþ] (100). [
a
]
þ12.6ꢂ (c 0.17, methanol); 1H NMR (DMSO-d6):
D
C
12H20N8O$2/3 MeOH (306.37) calcd: C, 50.09; H, 7.59; N, 34.19.
Found: C, 49.90; H, 7.29; N, 34.19.
0.89 (d, 3H, J(40,30) ¼ 6.8, CH3); 0.93 (d, 3H, J(40,30) ¼ 6.8, CH3); 1.48
(d, 6H, J(CH3, CH) ¼ 6.8, CH3); 1.72 (m, 1H, CH-30); 2.83 (m, 1H, CH-
20); 3.99 (dd,1H, J(10be20) ¼ 7.1, Jgem ¼ 10.4,10b); 4.19 (dd,1H, J(10ae
20) ¼ 5.2, Jgem ¼ 10.4,10a); 4.61 (br,1H, J(CH, CH3) ¼ 6.8, NeCH); 7.30
(vbs, 4H, NH); 7.96 (s, 1H, H-8). 13C NMR (DMSO-d6): 17.52 and
19.74 (C-40); 22.29 (2C, CH3); 30.34 (C-30); 46.15 (NeCH); 55.07 (C-
20); 70.10 (C-10); 122.08 (C-5); 137.60 (C-8); 151.08 (C-4); 159.92
(NeC); 160.21 (C-2); 160.75 (C-6).
6.1.7. 2-[2-(2S)-(Aminomethyl)propoxy]-6-guanidino-9-isopropyl-
9H-purine (16f) and 6-guanidino-2e{[1-(1S)-(hydroxymethyl)propyl]
amino}-9-isopropyl-9H-purine (16g)
General procedure; S-(þ)-2-amino-1-butanol (1.32 ml,
14 mmol); 6 h; RT; compound 16f: yield 0.3 g, 44%; FABMS: 307.0
[MHþ] (100). [
a
]
þ15.6ꢂ (c 0.39, methanol); 1H NMR (DMSO-d6):
Compound 16k: yield 0.12 g, 28%; FABMS: 321.0 [MHþ] (100).
D
0.93 (t, 3H, J(40, 30) ¼ 7.5, CH2-40); 1.28 (m, 1H, CH2-30b); 1.48 (d, 6H,
J(CH3, CH) ¼ 6.8, CH3); 1.52 (m, 1H, 30a); 2.90 (m, 1H, CH2-20); 3.98
(dd, 1H, J(10be20) ¼ 6.6, Jgem ¼ 10.2, 10b); 4.07 (dd, 1H, J(10ae
20) ¼ 5.4, Jgem ¼ 10.2, 10a); 4.61 (br, 1H, J(CH, CH3) ¼ 6.7, NeCH);
7.3 (vbs, 4H, NH); 7.96 (s, 1H, H-8). 13C NMR (DMSO-d6): 10.60 (C-
40); 22.31 (2C, CH3); 26.97 (C-30); 46.11 (NeCH); 51.63 (C-20); 71.71
(C-10); 122.07 (C-5); 137.58 (C-8); 151.08 (C-4); 159.93 (NeC);
160.20 (C-2); 160.75 (C-6).
[
a
]
ꢃ35.6ꢂ (c 0.24, methanol); 1H NMR (DMSO-d6): 0.90 (d, 3H,
J(40D,30) ¼ 6.9, CH3); 0.91 (d, 3H, J(40,30) ¼ 6.9, CH3); 1.45 (d, 3H, J(CH3,
CH) ¼ 6.9, CH3); 1.46 (d, 3H, J(CH3, CH) ¼ 6.8, CH3); 1.97 (m, 1H,
CH-30); 3.49 (m, 2H, CH2-10); 3.75 (brs, 1H, CH2-20); 4.52 (br, J(CHe
CH3) ¼ 6.8, CHeN); 4.58 (brs, 1H, OH); 6.13 (br, 1H, NH); 7.4 (vbs,
4H, NH); 7.71 (s, 1H, H-8). 13C NMR (DMSO-d6): 18.55 and 19.83
(C-40); 22.18 and 22.33 (2C, CH3); 28.82 (C-30); 45.71 (NeCH); 57.71
(C-20); 61.67 (C-10); 119.74 (C-5); 135.38 (C-8); 151.67 (C-4); 158.42

ꢀ
I. Dolecková et al. / European Journal of Medicinal Chemistry 62 (2013) 443e452
450
(C-2); 159.76 (NeC); 159.97 (C-6). For C16h24N8O$4/5 MeOH
(320.39) calcd: C, 51.37; H, 7.92; N, 32.38. Found: C, 51.23; H, 7.41; N,
32.31.
159.95 (NeC); 160.09 (C-6). For C16h22N8O$½ EtOH (318.38) calcd:
C, 52.77; H, 7.38; N, 32.82. Found: C, 52.75; H, 7.06; N, 32.98.
6.1.13. 2-Chloro-6-guanidino-9-isopropyl-9H-purine (17)
6.1.10. 2-Cyclopropylamino-6-guanidino-9-isopropyl-9H-purine (16l)
General procedure; cyclopropylamine (2.3 ml, 32.6 mmol); 4 h,
RT; yield 0.25 g, 25%; FABMS: 275.1 [MHþ] (100). 1H NMR (DMSO-
d6): 0.43 and 0.62 (m, 2H, 2ꢀ CH2); 1.48 (d, 6H, J(CH3, CH) ¼ 6.7,
CH3); 2.60 (m, 1H, NeCH); 4.55 (sept, 1H, J(CH, CH3) ¼ 6.7, NeCH);
6.72 (brs, 1H, NH); 7.50 (brs, 4H, NH); 7.77 (s, 1H, H-8). 13C NMR
(DMSO-d6): 6.68, 2C, (CH2); 22.27 (2C, CH3); 24.20 (NeCH); 45.50
(NeCH); 119.79 (C-5); 135.70 (C-8); 151.86 (C-4); 158.75 (C-6);
159.68 and 160.02 (NeC and C-2). For C12H18N8$0.5H2O$HCl
(274.32) calcd: C, 45.37; H, 6.26; N, 32.66; Cl 11.09. Found: C, 45.07;
H, 6.30; N, 35.04; Cl 10.92.
2,6-Dichloro-9-isopropyl-9H-purine [29] was treated with
a solution of guanidine [21] at RT. The precipitate was filtered and
filtrate was purified by column chromatography (5e15% ethano-
l:EtOAc followed by EtOAc:acetone:EtOH:H2O, 6:1:1:0.5). Yield
3.21 g, 75%. FABMS: 254.0 [MHþ] (100). 1H NMR (DMSO-d6): 1.49
(d, 6H, J(CH3, CH) ¼ 6.8, CH3); 4.67 (sept, 1H, J(CH, CH3) ¼ 6.8, Ne
CH); 7.40 (br, 4H, NH); 8.19 (s, 1H, H-8). 13C NMR (DMSO-d6): 22.30
(2C, CH3); 46.56 (NeCH); 124.50 (C-5); 139.36 (C-8); 150.57 (C-2);
150.87 (C-4); 160.16 (C-6); 160.34 (NeC). For C9H12ClN7$1/4H2O
(253.69) calcd: C, 41.87; H, 4.88; N, 37.97; Cl, 13.73. Found: C, 42.16;
H, 4.68; N, 37.82 Cl, 14.05.
6.1.11. 6-Guanidino-9-isopropyl-2-(2S)-(pyrrolidin-2-ylmethoxy)-
9H-purine (16m) and 6-guanidino-2-[2-(2S)-(hydroxymethyl)pyrr-
olidin-1-yl]-9-isopropyl-9H-purine (16n)
6.2. Enzyme inhibition assay
CDK1/cyclin B and CDK2/cyclin E kinases were produced using
the baculoviral expression system and purified on a Ni2þ-NTA
column (Qiagen). The kinase reaction was performed with 1 mg/ml
General
procedure;
(S)-(þ)-2-(hydroxymethyl)pyrrolidine
(1.43 ml, 14 mmol); 10 h; RT; compound 16m: yield 0.07 g, 16%;
FABMS: 318.0 [MHþ] (10). [
a
]
D
þ5.0ꢂ (c 0.28, methanol); 1H NMR
histone H1 in the presence of 15
with a test compound in a final volume of 10
buffer (60 mM HEPESeNaOH, pH 7.5, 3 mM MgCl2, 3 mM MnCl2,
M sodium orthovanadate, 1.2 mM DTT, 2.5 g/50 l PEG20.000).
The reaction was stopped by adding 5 l of 3% aqueous H3PO4.
m
M ATP, 0.05
m
Ci [
g-
33P]ATP, and
(DMSO-d6): 1.43 (m, 1H, 30b); 1.48 (d, 6H, J(CH3, CH) ¼ 6.8, CH3);
1.65 (m, 2H, CH-40); 1.82 (m, 1H, 30a); 2.79 (m, 2H, CH2-50); 3.68 (m,
1H, CH-20); 4.02 (dd, 1H, J(CH2e20) ¼ 6.1, Jgem ¼ 10.3, OeCH2b); 4.06
(dd, 1H, J(CH2e20) ¼ 6.7, Jgem ¼ 10.3, OeCH2a); 4.61 (br, 1H, J(CH,
CH3) ¼ 6.8, NeCH); 7.40 (vbs, 4H, NH); 7.95 (s, 1H, H-8). 13C NMR
(DMSO-d6): 22.27 (2C, CH3); 25.24 (C-40); 28.65 (C-30); 46.09 (Ne
CH); 46.25 (C-50); 56.70 (C-20); 70.32 (OeCH2); 122.05 (C-5);
137.49 (C-8); 151.06 (C-4); 159.89 (NeC); 160.23 (C-2); 160.72 (C-6).
Compound 16n: yield 0.22 g, 48%; FABMS: 319 [MHþ] (100).
ml, all in the reaction
3
m
m
m
m
Aliquots were spotted onto P-81 phosphocellulose paper (What-
man), washed 3ꢀ with 0.5% aqueous H3PO4, and finally air-dried.
Kinase inhibition was quantified using digital image analyzer
FLA-7000 (Fujifilm). The IC50 values were determined from dosee
response curves.
Kinase selectivity profiling was carried out under the conditions
used as described previously [37]. Compounds 16e and 16g were
screened at a concentration of 1
[
a]
ꢃ83.4ꢂ (c 0.33, methanol); 1H NMR (DMSO-d6): 1.47 (d, 6H,
D
J(CH3, CH) ¼ 6.8, CH3); 1.83e1.99 (m, 4H, CH2-30, 40); 3.37 (m, 1H,
50b, CH2eO); 3.49 (m, 1H, 50a); 3.66 (bdt, J(CH2eOH) ¼ J(CH2-
20) ¼ 4.1, Jgem ¼ 10.2, CH2eO); 4.04 (brs, CH2-20); 4.54 (br, 1H,
J(CH, CH3) ¼ 6.8, NeCH); 4.90 (brs, 1H, OH); 7.3 (vbs, 4H, NH); 7.74
mM in duplicate. The assays were
initiated with ATP (800 cpm/pmol [
g-
33P]ATP at 5, 20 or 50
mM in
order to be at or below the Km for ATP for each enzyme), stopped by
the addition of phosphoric acid and spotted onto P81 filter plates.
Inhibition was expressed as residual kinase activity.
13
(s, 1H, H-8). C NMR (DMSO-d6): 22.01 (2C, CH3); 23.21 (C-40);
28.00 (C-30); 45.85 (NeCH); 47.83 (C-50); 59.46 (C-20); 62.48 (Oe
CH2); 119.73 (C-5); 135.81 (C-8); 151.64 (C-4); 156.48 (C-2);
159.95 (NeC); 160.09 (C-6). For C12H20N8O$1/3 MeOH (292.34)
calcd: C, 52.32; H, 7.15; N, 34.05. Found: C, 52.45; H, 6.97; N, 34.15.
6.3. Cytotoxicity assay
The cytotoxicity of the studied compounds was determined on
MCF7 breast cancer cell line. The cells were cultivated in Dulbecco’s
Modified Eagle’s Medium (DMEM) supplemented with 10% (v/v)
fetal bovine serum, 2 mM glutamine, 100 IU/ml penicillin and
100 mg/ml streptomycin and maintained at 37 ꢂC, in a humidified
environment with 5% CO2. Approximately 10,000 cells were seeded
into a well of a 96-well microtiter plate. After 4 h stabilization,
a tested compound in various concentrations was added in tripli-
cates. The final concentration of DMSO never exceeded 0.5%. After
72 h incubation, calcein AM solution (Molecular Probes) was added
6.1.12. 6-Guanidino-9-isopropyl-2-(2R)-(pyrrolidin-2-ylmethoxy)-
9H-purine (16o) and 6-guanidino-2-[2-(2R)-(hydroxymethyl)pyrr-
olidin-1-yl]-9-isopropyl-9H-purine (16p)
General
procedure;
(R)-(ꢃ)-2-(hydroxymethyl)pyrrolidine
(1.43 ml, 14 mmol), 10 h; RT; compound 16o: yield 0.07 g, 15%;
FABMS: 319.0 [MHþ] (40). [
a]
ꢃ8.6ꢂ (c 0.21, methanol); 1H NMR
D
(DMSO-d6): 1.43 (m, 1H, 30b); 1.48 (d, 6H, J(CH3, CH) ¼ 6.8, CH3);
1.65 (m, 2H, CH-40); 1.82 (m, 1H, 30a); 2.79 (m, 2H, CH2-50); 3.68 (m,
1H, CH-20); 4.02 (dd, 1H, J(CH2-20) ¼ 6.1, Jgem ¼ 10.3, OeCH2b); 4.06
(dd, 1H, J(CH2e20) ¼ 6.7, Jgem ¼ 10.3, OeCH2a); 4.61 (br, 1H, J(CH,
CH3) ¼ 6.8, NeCH); 7.40 (vbs, 4H, NH); 7.95 (s, 1H, H-8). 13C NMR
(DMSO-d6): 22.27 (2C, CH3); 25.24 (C-40); 28.65 (C-30); 46.09 (Ne
CH); 46.25 (C-50); 56.70 (C-20); 70.32 (OeCH2); 122.05 (C-5);
137.49 (C-8); 151.06 (C-4); 159.89 (NeC); 160.23 (C-2); 160.72 (C-6).
Compound 16p: yield 0.23 g, 50%; FABMS: 319.0 [MHþ] (50).
to the final concentration of 1 mg/ml for 1 h and the fluorescence of
free calcein was measured at 485/538 nm (excitation/emission)
using Fluoroskan Ascent reader (Labsystem). The viability was
expressed as a percentage of the fluorescence of the control cells
and the IC50 values (the concentration of an inhibitor that is
required for 50% decrease of the intracellular esterase activity) were
determined by graphical analysis from doseeresponse curves. The
IC50 values represent averages of at least three independent
experiments.
[
a]
þ94.3ꢂ (c 0.53, methanol); 1H NMR (DMSO-d6): 1.47 (d, 6H,
D
J(CH3, CH) ¼ 6.8, CH3); 1.83e1.99 (m, 4H, CH2-30, 40); 3.37 (m, 1H,
50b, CH2eO); 3.49 (m, 1H, 50a); 3.66 (bdt, J(CH2eOH) ¼ J(CH2-
20) ¼ 4.1, Jgem ¼ 10.2, CH2eO); 4.04 (brs, CH2-20); 4.54 (br, 1H,
J(CH, CH3) ¼ 6.8, NeCH); 4.90 (brs, 1H, OH); 7.3 (vbs, 4H, NH); 7.74
6.4. Cell cycle analysis
13
(s, 1H, H-8). C NMR (DMSO-d6): 22.01 (2C, CH3); 23.21 (C-40);
28.00 (C-30); 45.85 (NeCH); 47.83 (C-50); 59.46 (C-20); 62.48 (Oe
Approximately million cells were fixed with 90% methanol. Af-
CH2); 119.73 (C-5); 135.81 (C-8); 151.64 (C-4); 156.48 (C-2);
ter rehydration the cells were incubated with primary antibody

ꢀ
I. Dolecková et al. / European Journal of Medicinal Chemistry 62 (2013) 443e452
451
against pH3Ser10 and then with secondary antibody conjugated
with Alexa Fluor 488. DNA was stained with propidium iodide and
the fluorescence signal was measured at 488/575 nm (ex/em) for
propidium iodide and 488/525 nm (ex/em) for Alexa Fluor 488
using Cell Lab QuantaTM SC flow cytometer (Beckman Coulter). The
data were analyzed using the program Multicycle AV for Windows.
[8] M. Knockaert, P. Greengard, L. Meijer, Pharmacological inhibitors of cyclin-
dependent kinases, Trends Pharmacol. Sci. 23 (2002) 417e425.
[9] J. Vesely, L. Havlicek, M. Strnad, J.J. Blow, A. Donella-Deana, L. Pinna,
D.S. Letham, J. Kato, L. Detivaud, S. LeClerc, Inhibition of cyclin-dependent
kinases by purine analogues, Eur. J. Biochem. 224 (1994) 771e786.
[10] L. Havlicek, J. Hanus, J. Vesely, S. LeClerc, L. Meijer, G. Shaw, M. Strnad,
Cytokinin-derived cyclin-dependent kinase inhibitors: synthesis and cdc2
inhibitory activity of olomoucine and related compounds, J. Med. Chem. 40
(1997) 408e412.
[11] L. Meijer, A. Borgne, O. Mulner, J.P. Chong, J.J. Blow, N. Inagaki, M. Inagaki,
J.G. Delcros, J.P. Moulinoux, Biochemical and cellular effects of roscovitine,
a potent and selective inhibitor of the cyclin-dependent kinases cdc2, cdk2
and cdk5, Eur. J. Biochem. 243 (1997) 527e536.
[12] N.S. Gray, L. Wodicka, A.M. Thunnissen, T.C. Norman, S. Kwon, F.H. Espinoza,
D.O. Morgan, G. Barnes, S. LeClerc, L. Meijer, S.H. Kim, D.J. Lockhart,
P.G. Schultz, Exploiting chemical libraries, structure, and genomics in the
search for kinase inhibitors, Science 281 (1998) 533e538.
[13] Y.T. Chang, N.S. Gray, G.R. Rosania, D.P. Sutherlin, S. Kwon, T.C. Norman,
R. Sarohia, M. Leost, L. Meijer, P.G. Schultz, Synthesis and application of
functionally diverse 2,6,9-trisubstituted purine libraries as CDK inhibitors,
Chem. Biol. 6 (1999) 361e375.
[14] V. Krystof, I.W. McNae, M.D. Walkinshaw, P.M. Fischer, P. Muller, B. Vojtesek,
M. Orsag, L. Havlicek, M. Strnad, Antiproliferative activity of olomoucine II,
a novel 2,6,9-trisubstituted purine cyclin-dependent kinase inhibitor, Cell.
Mol. Life Sci. 62 (2005) 1763e1771.
6.5. Docking study
The human CDK2, roscovitine and inhibitors 16e/16g and 16i/
16k were prepared for docking in the YASARA modelling package
[39] on the basis of deposited structure 2A4L [33]. H atoms were
added to the protein to mimic neutral pH and their positions were
optimized. The glycerol and water molecules were removed from
the model. The parameter set used for the protein was AMBER ff03
[40]. The ligand was optimized in a vacuum and partial charges on
its atoms were obtained by a restrained fit to the electrostatic po-
tential (RESP) at the AM1BCC level [41]. The ligand was then docked
to the protein using the AutoDock program [42]. 1000 poses were
obtained using a local search protocol, these were subsequently
[15] K. Bettayeb, N. Oumata, A. Echalier, Y. Ferandin, J.A. Endicott, H. Galons,
L. Meijer, CR8, a potent and selective, roscovitine-derived inhibitor of cyclin-
dependent kinases, Oncogene 27 (2008) 5797e5807.
ꢁ
clustered based on similarity (r.m.s.d. < 5 A) and then scored.
[16] N. Oumata, K. Bettayeb, Y. Ferandin, L. Demange, A. Lopez-Giral, M.L. Goddard,
V. Myrianthopoulos, E. Mikros, M. Flajolet, P. Greengard, L. Meijer, H. Galons,
Roscovitine-derived, dual-specificity inhibitors of cyclin-dependent kinases
and casein kinases 1, J. Med. Chem. 51 (2008) 5229e5242.
Contributions
ID, MC, MD, VK and ZJ conceived and designed the experiments.
MC performed organic syntheses. ID performed biological assays
and ID, VK and JV analyzed biological data set. ID, MC, VK, JV and ZJ
wrote the paper. JB performed the docking study. All authors have
approved the article.
[17] M.P. Trova, K.D. Barnes, C. Barford, T. Benanti, M. Bielaska, L. Burry,
J.M. Lehman, C. Murphy, H. O’Grady, D. Peace, S. Salamone, J. Smith, P. Snider,
J. Toporowski, S. Tregay, A. Wilson, M. Wyle, X. Zheng, T.D. Friedrich, Biaryl
purine derivatives as potent antiproliferative agents: inhibitors of cyclin
dependent kinases. Part I, Bioorg. Med. Chem. Lett. 19 (2009) 6608e6612.
[18] M.P. Trova, K.D. Barnes, L. Alicea, T. Benanti, M. Bielaska, J. Bilotta, B. Bliss,
T.N. Duong, S. Haydar, R.J. Herr, Y. Hui, M. Johnson, J.M. Lehman, D. Peace,
M. Rainka, P. Snider, S. Salamone, S. Tregay, X. Zheng, T.D. Friedrich, Hetero-
biaryl purine derivatives as potent antiproliferative agents: inhibitors of cyclin
dependent kinases. Part II, Bioorg. Med. Chem. Lett. 19 (2009) 6613e6617.
[19] R. Jorda, K. Paruch, V. Krystof, Cyclin-dependent kinase inhibitors inspired by
roscovitine: purine bioisosteres, Curr. Pharm. Des. 18 (2012) 2974e2980.
[20] R.G. Berlinck, M.H. Kossuga, Natural guanidine derivatives, Nat. Prod. Rep. 22
(2005) 516e550.
[21] M. Cesnek, A. Holy, M. Masojidkova, Z. Zidek, Synthesis of 9-alkyl and 9-
heteroalkyl substituted 2-amino-6-guanidinopurines and their influence on
the NO-production in macrophages, Bioorg. Med. Chem. 13 (2005) 2917e
2926.
[22] A.J. Hutchison, M. Williams, J.R. de, R. Yokoyama, H.H. Oei, G.R. Ghai,
R.L. Webb, H.C. Zoganas, G.A. Stone, M.F. Jarvis, 2-(Arylalkylamino)adenosin-
50-uronamides: a new class of highly selective adenosine A2 receptor ligands,
J. Med. Chem. 33 (1990) 1919e1924.
Acknowledgements
This work is a part of the research project #AVOZ40550506. It
was supported by the “Centre for New Antivirals and Antineo-
plastics” (1M0508) and by Gilead Sciences, Inc. (Foster City, CA,
USA). The authors from Palacky University were supported by grant
P305/12/0783 from the Czech Science Foundation and Ministry of
Education, Youth and Sports, Czech Republic (ED0007/01/01 Centre
of the Region Haná for Biotechnological and Agricultural Research).
The authors thank Dita Parobková from Palacky University for the
technical assistance and the staff of the mass spectrometry and
analytical departments of the Institute of Organic Chemistry and
Biochemistry ASCR.
[23] C.H. Hocart, D.S. Letham, C.W. Parker, Inhibitors of cytokinin metabolism 4.
Substituted xanthines and cytokinin analogs as inhibitors of cytokinin N-
glucosylation, Phytochemistry 30 (1991) 2477e2486.
[24] V. Brun, M. Legraverend, D.S. Grierson, Traceless solid-phase synthesis of
2,6,9-trisubstituted purines from resin bound 6-thiopurines, Tetrahedron 58
(2002) 7911e7923.
Appendix A. Supplementary material
[25] S. Piguel, M. Legraverend, Selective amidation of 2,6-dihalogenopurines:
application to the synthesis of new 2,6,9-trisubstituted purines, J. Org. Chem.
72 (2007) 7026e7029.
Supplementary data related to this article can be found at
http://dx.doi.org/10.1016/j.ejmech.2013.01.021.
[26] N. Boudet, S.R. Dubbaka, P. Knochel, Oxidative amination of cuprated py-
rimidine and purine derivatives, Org. Lett. 10 (2008) 1715e1718.
[27] M. Legraverend, O. Ludwig, S. LeClerc, L. Meijer, Synthesis of a new series of
purine derivatives and their anti-cyclin-dependent kinase activities,
J. Heterocycl. Chem. 38 (2001) 299e303.
[28] V. Nair, T.B. Sells, Copper mediated reactions in nucleoside synthesis, Tetra-
hedron Lett. 31 (1990) 807e810.
[29] M. Rypka, J. Vesely, Z. Chmela, D. Riegrova, K. Cervenkova, L. Havlicek, K. Lemr,
J. Hanus, B. Cerny, J. Lukes, K. Michalikova, In vitro biotransformation of 2,6,9-
trisubstituted purine-derived cyclin-dependent kinase inhibitor bohemine by
mouse liver microsomes, Xenobiotica 32 (2002) 1017e1031.
References
[1] M. Malumbres, M. Barbacid, Mammalian cyclin-dependent kinases, Trends
Biochem. Sci. 30 (2005) 630e641.
[2] L. Zuo, J. Weger, Q. Yang, A.M. Goldstein, M.A. Tucker, G.J. Walker, N. Hayward,
N.C. Dracopoli, Germline mutations in the p16INK4a binding domain of CDK4
in familial melanoma, Nat. Genet. 12 (1996) 97e99.
[3] K.S. Smalley, R. Contractor, T.K. Nguyen, M. Xiao, R. Edwards, V. Muthusamy,
A.J. King, K.T. Flaherty, M. Bosenberg, M. Herlyn, K.L. Nathanson, Identification
of
a
novel subgroup of melanomas with KIT/cyclin-dependent kinase-4
[30] D.L. Ladd, Synthesis of some substituted guanidinopyrimidines and their
structural assignment by C-13 and H-1-NMR, J. Heterocycl. Chem. 19 (1982)
917e921.
[31] C. Crosio, G.M. Fimia, R. Loury, M. Kimura, Y. Okano, H.Y. Zhou, S. Sen,
C.D. Allis, P. Sassone-Corsi, Mitotic phosphorylation of histone H3: spatio-
temporal regulation by mammalian aurora kinases, Mol. Cell. Biol. 22 (2002)
874e885.
overexpression, Cancer Res. 68 (2008) 5743e5752.
[4] S. Ortega, M. Malumbres, M. Barbacid, Cyclin D-dependent kinases, INK4 in-
hibitors and cancer, Biochim. Biophys. Acta 1602 (2002) 73e87.
[5] G. Romano, A. Giordano, Role of the cyclin-dependent kinase 9-related
pathway in mammalian gene expression and human diseases, Cell Cycle 7
(2008) 3664e3668.
[6] S. Lapenna, A. Giordano, Cell cycle kinases as therapeutic targets for cancer,
Nat. Rev. Drug Discov. 8 (2009) 547e566.
[7] V. Krystof, S. Uldrijan, Cyclin-dependent kinase inhibitors as anticancer drugs,
Curr. Drug Targets 11 (2010) 291e302.
[32] B. Perez-Cadahia, B. Drobic, J.R. Davie, H3 phosphorylation: dual role in
mitosis and interphase, Biochem. Cell. Biol. 87 (2009) 695e709.
[33] W.F. DeAzevedo, S. LeClerc, L. Meijer, L. Havlicek, M. Strnad, S.H. Kim, In-
hibition of cyclin-dependent kinases by purine analogues e crystal structure

ꢀ
I. Dolecková et al. / European Journal of Medicinal Chemistry 62 (2013) 443e452
452
of human cdk2 complexed with roscovitine, Eur. J. Biochem. 243 (1997)
518e526.
[38] M. Cesnek, M. Masojidkova, A. Holy, V. Solinova, D. Koval, V. Kasicka, Synthesis
and properties of 2-guanidinopurines, Collect. Czech. Chem. Commun. 71
(2006) 1303e1319.
[39] E. Krieger, K. Joo, J. Lee, J. Lee, S. Raman, J. Thompson, M. Tyka, D. Baker,
K. Karplus, Improving physical realism, stereochemistry, and side-chain ac-
curacy in homology modeling: four approaches that performed well in CASP8,
Proteins 77 (2009) 114e122.
[40] Y. Duan, C. Wu, S. Chowdhury, M.C. Lee, G.M. Xiong, W. Zhang, R. Yang,
P. Cieplak, R. Luo, T. Lee, J. Caldwell, J.M. Wang, P. Kollman, A point-charge
force field for molecular mechanics simulations of proteins based on
condensed-phase quantum mechanical calculations, J. Comput. Chem. 24
(2003) 1999e2012.
[34] D.S. Williamson, M.J. Parratt, J.F. Bower, J.D. Moore, C.M. Richardson, P. Dokurno,
A.D. Cansfield, G.L. Francis, R.J. Hebdon, R. Howes, P.S. Jackson, A.M. Lockie,
J.B. Murray, C.L. Nunns, J. Powles, A. Robertson, A.E. Surgenor, C.J. Torrance,
Structure-guided design of pyrazolo[1,5-a]pyrimidines as inhibitors of human
cyclin-dependent kinase 2, Bioorg. Med. Chem. Lett. 15 (2005) 863e867.
[35] S. Bach, M. Knockaert, J. Reinhardt, O. Lozach, S. Schmitt, B. Baratte, M. Koken,
S.P. Coburn, L. Tang, T. Jiang, D.C. Liang, H. Galons, J.F. Dierick, L.A. Pinna,
F. Meggio, F. Totzke, C. Schachtele, A.S. Lerman, A. Carnero, Y.Q. Wan, N. Gray,
L. Meijer, Roscovitine targets, protein kinases and pyridoxal kinase, J. Biol.
Chem. 280 (2005) 31208e31219.
[36] S.L. Posy, M.A. Hermsmeier, W. Vaccaro, K.H. Ott, G. Todderud, J.S. Lippy,
G.L. Trainor, D.A. Loughney, S.R. Johnson, Trends in kinase selectivity insights
for target class-focused library screening, J. Med. Chem. 54 (2011) 54e66.
[37] J. Bain, L. Plater, M. Elliott, N. Shpiro, C.J. Hastie, H. Mclauchlan, I. Klevernic,
J.S.C. Arthur, D.R. Alessi, P. Cohen, The selectivity of protein kinase inhibitors:
a further update, Biochem. J. 408 (2007) 297e315.
[41] A. Jakalian, D.B. Jack, C.I. Bayly, Fast, efficient generation of high-quality
atomic charges. AM1-BCC model: II. Parameterization and validation,
J. Comput. Chem. 23 (2002) 1623e1641.
[42] G.M. Morris, R. Huey, W. Lindstrom, M.F. Sanner, R.K. Belew, D.S. Goodsell,
A.J. Olson, AutoDock4 and AutoDockTools4: automated docking with selective
receptor flexibility, J. Comput. Chem. 30 (2009) 2785e2791.