8-Azapurines as new inhibitors of cyclin-dependent kinases

Article abstract of DOI:10.1016/j.bmc.2005.06.007

Purine inhibitors of cyclin-dependent kinases (CDK) seem to be a potential anticancer drug candidate as one of the first representatives, roscovitine, is passing Phase II clinical trials for cancer and glomerulonephritis. In this article, we describe a no

Full text of DOI:10.1016/j.bmc.2005.06.007

Bioorganic & Medicinal Chemistry 13 (2005) 5399–5407
8-Azapurines as new inhibitors of cyclin-dependent kinases
Libor Havlicek,a Kveta Fuksova,b Vladimir Krystof,c Martin Orsag,c
Borivoj Vojtesekd and Miroslav Strnadc,*
aIsotope Laboratory, Institute of Experimental Botany, Academy of Sciences of the Czech Republic, Videnska 1083,
142 20 Prague 4, Czech Republic
bFirst Faculty of Medicine, Charles University, 121 08 Prague, Czech Republic
cLaboratory of Growth Regulators, Palacky University and Institute of Experimental Botany, Slechtitelu 11,
783 71 Olomouc, Czech Republic
dDepartment of Experimental Oncology, Masaryk Memorial Cancer Institute, Zluty kopec 7, 656 53 Brno, Czech Republic
Received 25 January 2005; revised 31 May 2005; accepted 3 June 2005
Available online 29 June 2005
Abstract—Purine inhibitors of cyclin-dependent kinases (CDK) seem to be a potential anticancer drug candidate as one of the first
representatives, roscovitine, is passing Phase II clinical trials for cancer and glomerulonephritis. In this article, we describe a novel
modification of the purine scaffold influencing CDK2 inhibitory activities as well as anticancer properties in cell lines of different
histopathological origin. The introduced N at position 8 of the purine ring generally lowered CDK2 inhibitory activity of new
8-azapurines (1,2,3-triazolo[4,5-d]pyrimidines) in comparison to the model trisubstituted purines, whereas the antiproliferative
potential of some derivatives remained very high, reflecting their ability to activate p53 tumor suppressor.
ꢀ 2005 Elsevier Ltd. All rights reserved.
1. Introduction
not only confined to cell cycle control, but also function
in the regulation of transcription. CDK2, 7, 8, and 9
have been shown to be implicated in transcriptional reg-
ulation through activating and deactivating phosphoryl-
ations of the C-terminal domain of RNA polymerase
II.7 Despite the facts mentioned above, new CDK-inhib-
itory pharmacophores and compounds have to be devel-
oped and reported. The lack of selectivity as well as
incomplete understanding of the selectivity of the
CDK inhibitors makes interpretation of their pharma-
cological effects difficult and the question of which
CDK(s) should be inhibited for optimal anticancer
activity remains largely unanswered.8
All cancers are characterized by an abnormal control of
cell proliferation. This is caused by mutation or mis-reg-
ulation of cell-cycle regulatory genes and proteins.1 It
has been shown that it deregulates G1/S cyclin-depen-
dent kinase (CDK) activity in tumor cells, while en-
abling these cells to enter S-phase unchecked, also
renders them selectively susceptible to CDK2 inhibi-
tion,2 because CDK2 activity is not only instrumental
in initiating S-phase (CDK2/cyclin E), but it is also crit-
ically required to terminate E2F-mediated transcription-
al activity in S-phase (CDK2/cyclin A). Failure of a cell
to terminate this activity in timely fashion constitutes a
powerful apoptotic signal.3 Recent findings also suggest
that CDK2 activity—at least in colorectal cancers—may
be redundant4 and that other kinases, possibly CDK3 or
AURORA kinases5, can step in upon CDK2 activation.
Even CDK1/cyclin B1 activity, which had traditionally
been associated with regulation of M-phase entry, was
recently shown to possess S-phase promoting potential.6
Furthermore, it is now clear that the roles of CDKs are
The crucial role of CDK within cell division cycle and
their frequent deregulations in human tumor and cancer
cells have encouraged an active search for chemical
CDK inhibitors (CDKI).9 Systematic screening of the
adenine-derived plant hormone cytokinins led to the dis-
covery of 2,6,9-trisubstituted purines, a group of highly
active CDK inhibitors.10 The first specific inhibitor
olomoucine,10 was quickly followed by derivatives with
enhanced efficiency, like roscovitine, purvalanol A,
olomoucine II and others.11–15 Several of them have al-
ready undergone in vivo efficacy studies, preclinical and
clinical evaluation, confirming CDKs as real targets for
development of new anticancer drugs.15–17
Keywords: CDK2; Inhibitor; Anticancer drug; p53.
*
Corresponding author. Tel.: +420 585 634 850; fax: +420 585 634
870; e-mail: strnad@aix.upol.cz
0968-0896/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2005.06.007

5400
L. Havlicek et al. / Bioorg. Med. Chem. 13 (2005) 5399–5407
The pharmacological improvement of the olomoucine
structure was till date now mainly based on modifica-
tions at positions 2, 6 and 9, respectively.11–18 In the ef-
fort to enhance the affinity of CDKIs, we tried to extend
the inhibitor-CDK mutual interactions by introduction
of another nitrogen to the purine ring. There are already
reports describing rearrangement of the purine het-
eroatoms.19,20 One of them characterized successful
repositioning of heteroatoms of purine to pyrazolo[4,3-
d]pyrimidine that lead to improved anti-CDK activity.19
The shift of heteronitrogen from position 9 to 8 prompt-
ed us to further explore and modify the purine ring.
Therefore, we prepared several 2,6,9-trisubstituted 8-
azapurines as analogs of potent CDK inhibitors. The
impact of modification of the purine skeleton on CDK
inhibitory activity was consequently verified by p53-
transcriptional and anticancer activity testing.
derivatives II was a crucial step in the synthesis which
was achieved by the action of POCl3/2,6-lutidine at
120 ꢁC. Decomposition of the evaporated reaction resi-
dues by coincident extraction in benzene or chloroform
at 0 ꢁC afforded crude dichloroderivatives III. While the
benzyl derivative IIIa was able to be crystallized, the iso-
propyl analogue IIIb seems to be very unstable and was
therefore used immediately in subsequent nucleophilic
substitution of C1–C6. Furthermore, the yield of 2,6-
dichloro-9-isopropyl-8-azapurine was not entirely repro-
ducible as a satisfactory yield (20–25%) was obtained
only under the small-mass conditions (up to 2 g). Subse-
quent regioselective substitution of C1–C6 (III!IV) was
then achieved at 0–20 ꢁC using only one equivalent of
R6-NH2 in the presence of tertiary amine. At room tem-
perature, the highly reactive amines like benzylamines
were however, able to substitute C1–C2 to a certain ex-
tent. The final substitution reaction of C1–C2 was then
performed at elevated temperature (>100 ꢁC, IV!V).
The 2-methyl-8-azaderivative 2 was synthesized by a
modified procedure adopted from the synthesis of simi-
lar 8-azapurine compounds.23 Here the cyclization of
isopropylderivative Ib with ethyl acetate gave rise to
2. Results and discussion
2.1. Chemistry
We used several approaches to synthesize the target
compounds. Owing to the low regioselectivity we
abandoned the synthesis starting either from 2,6-dichlo-
ro-8-azapurine or 5-nitro-2,4,6-trichloropyrimidine,
respectively. For example, the reaction of isopropyl
amine with 5-nitro-2,4,6-trichloropyrimidine proceeds
with all three chlorine atoms simultaneously in contrast
to the selective reaction of ammonia.21 A series of
new 2,6,9-trisubstituted-8-azapurines (1,2,3-triazol-
o[4,5-d]pyrimidines) was therefore synthesized using a
general synthetic route outlined in Figure 1. In this
approach an alkylazide (R9N3) was cyclized with cya-
nacetamide in the presence of sodium ethoxide to 1-al-
kyl-4-amino-5-carboxamido-1,2,3-triazole (I). Reaction
of derivative I with diethyl carbonate afforded 9-alkyl-
2,6-dihydroxy-8-azapurine (8-azaxanthine derivative
II). The synthesis of 9-benzyl-2,6-dihydroxy-8-azapurine
(IIa) followed the cited procedures.22 This approach also
proved to be useful for 9-isopropylderivative II despite
the lower yields. Chlorination of 8-azaxanthine
6-hydroxy-9-isopropyl-2-methyl-8-azapurine
which
after chlorination by Bosshard reagent (SOCl2/DMF)
and subsequent nucleophilic substitution at C1–C6
afforded the target compound 2.
2.2. Biological activity
To verify the effect of an additional nitrogen in position
8 of the active purine inhibitors, new derivatives (sum-
marized in Table 1) were screened for their ability to
inhibit recombinant CDK2/cyclin E. The results showed
that the introduction of the fifth nitrogen instead of
C8H functionality in the purine ring reduces CDK2
inhibitory activity in all cases (30–60-fold). Straightfor-
ward structural point of view would consider 8-azapu-
rines to be weaker CDK inhibitors than respective
purines; as co-crystals of CDK2 with purine inhibitors
show (e.g., purvalanol B or H717),12,13 hydrogen at C8
of the purine ring interacts with the backbone oxygen
of Glu81. In contrast, guanine NU2058 binds to
Figure 1. Scheme of synthesis of 2,6,9-trisubstituted-8-azapurines.

L. Havlicek et al. / Bioorg. Med. Chem. 13 (2005) 5399–5407
5401
Table 1. Structure–activity relationships of 8-azapurines
Compound Substitution
R6
IC50 (lM)a
p53a
G361 Arn8
R2
Roscovitineb [1-(Hydroxymethyl)propyl]
CDK2 MCF7 K562 HOS
Benzyl
Benzyl
Benzyl
Benzyl
0.45
11.3
100
4.1
11. 1
11.4
112
40
32
34
8.3
++(20)
++(20)
—
1
3
(4-Aminocyclohexyl)
(3-Hydroxypropyl)
[1-(Hydroxymethyl)propyl]
19.2
142
15.6
159
123
129
32.6
14.3
126
84.7
28.2
42.6
25.4
4
55.3
32.8
31.8
13.1
22.1
13.6
23.4
20.2
>167
22.3
26.9
29.1
18.7
17.8
>167
5.5
68.0
50.0
68. 1
30.3
33.1
18.9
19.4
21.3
>167
39.8
63.2
54.3
24.3
25.3
>167
5.7
+(80)
++(50)
++(40)
++(20)
+(80)
++(30)
—
5
[1-(Hydroxymethyl)-2-methylpropyl] Benzyl
7.3
(3-Hydroxy-4-methoxybenzyl) 2.3
6
[1-(Hydroxymethyl)propyl]
(2-Aminocyclohexyl)
[1-(Hydroxymethyl)propyl]
7
(3-Hydroxy-4-methoxybenzyl) 13.7
(2-Hydroxybenzyl)
8
1.2
101.7 49.2
9
[1-(Hydroxymethyl)-2-methylpropyl] (2-Hydroxybenzyl)
0.84
4.1
35.7
69.7
33.1
>167
35.9
30.3
72.2
26.7
31.8
>167
8.8
19.1
20.5
11.8
>167
41.1
77.4
34.1
24.1
26.9
>167
5.5
10
11
12
13
14
15
16
17
18
19
20
21
(2-Aminocyclohexyl)
(4-Aminocyclohexyl)
(2-Hydroxybenzyl)
(2-Hydroxybenzyl)
(2-Hydroxybenzyl)
(2-Hydroxybenzyl)
16.4
>100
8.0
—
—
[1-(Hydroxymethyl)-2-methylpropyl] (3-Hydroxybenzyl)
++(30)
++(30)
++(50)
++(10)
—
(4-Aminocyclohexyl)
[1-(Hydroxymethyl)propyl]
(4-Aminocyclohexyl)
Hexyl
(3-Hydroxybenzyl)
(3-Hydroxybenzyl)
4.9
10
(4-Hydroxybenzyl)
[(2-Hydroxy-1-phenyl)ethyl]
2.8
>100
20
1.1
[1-(Hydroxymethyl)-2-methylpropyl] (3-Chlorophenyl)
—
—
(4-Aminocyclohexyl)
(4-Hydroxycyclohexyl)
(4-Hydroxycyclohexyl)
(3-Chlorophenyl)
(3-Chlorophenyl)
Cyclohexylmethyl
>100
>100
>167
>167
>167
>167
>167
>167
>167
>167
—
—
Biological activities were tested according to methods described in the Section 4.14 Induction of p53-dependent transcription in Arn8 cells is expressed
here as positive (+), strongly positive (++) or negative (ꢀ), with the concentration exhibiting maximum activity given in brackets in micromolar.
a Values are means of three experiments; SD values did not exceed 15%.
b Roscovitine activity was added as a standard compound for comparison.
CDK2 in a different manner, with C8H shifted towards
Phe80, unable to H-bond to the protein, but still exhib-
iting high affinity to CDK2.24 Similarly, 3,7-disubstitut-
ed pyrazolo[4,3-d]pyrimidines without C8H also
potently inhibit CDKs and probably orient within the
active site in a mode of mentioned guanines.19 The posi-
tion and presence of heteronitrogens in a purine skeleton
thus seems to markedly affect the binding mode of pur-
ine-like ligands. In this respect, synthesis and evaluation
of 8-azapurines was necessary to elucidate SAR of pur-
ine based heterocyclic skeletons.
was also obvious from the in vitro anticancer tests (Ta-
ble 1). However, a 50-fold decrease of CDK inhibitory
potency of all compounds versus parental trisubstituted
purines was unexpectedly not followed by intensive
reduction of cytotoxic properties of the respective 8-
azapurine derivatives (2–5-fold), active in a micromolar
range. Moreover, compound 19 derived from parental
purine CGP 74514 surprisingly exerted even slightly
stronger cytotoxicity on a panel of 17 cell lines from var-
ious human tumor tissues (Table 2).
Table 2. Antiproliferative activity of selected 8-azapurines on cancer
cell lines of different origin
In summary, the following structure–activity relation-
ships of 8-azapurines have been observed: (1) isopropyl
at N9 was the most promising, bulky substituents de-
creased the inhibitory activity to below detectable level
(20 and 21); (2) the most significant structural feature
was the introduction of 2-hydroxybenzylamino (olo-
moucine II) or 3-chloroanilinoamino (purvalanol A)
substituent at C6 position;12,14 (3) [1-(hydroxymeth-
yl)propyl]amino (roscovitine),11 [1-(hydroxymethyl)-2-
methylpropyl]amino (purvalanol)12 and (4-amino-
cyclohexyl)amino (CGP 74514)25 C2-side chains were
also the most effective with 8-azapurines, like in the case
of trisubstituted purines. Although the introduced nitro-
gen decreased the affinity of 8-azapurines to CDK2,
many compounds yet (6, 8–10, 14, 16, 19) retained
IC50 values below olomoucine (7 lM).
Cell line
Compound IC50 (lM)a
8
11
16
19
8.8
HOS
101.7
123.4
75.5
22.1
119.2
16.6
36.9
33.1
110.2
61.5
118.3
29.2
49.2
48.0
60.2
154.6
35.2
33.1
35.3
21.8
20.2
34.7
13.6
21.6
21.3
18.9
47.6
32.5
27.0
11.8
33.0
31.4
32.5
23.1
26.7
HBL100
BT474
MCF7
HT29
CEM
33.2
15.6
18.7
34.6
7.2
9.1
7.6
5.5
11.3
5.1
7.0
5.7
7.8
9.7
7.9
7.8
5.5
7.7
7.6
8.5
6.1
HL60
K562
17.6
24.3
21.4
157.8
26.2
22.9
24.1
38.6
26.8
114.6
25.8
HS913T
A549
A431
SVK14
G361
SKUT-1
HeLa
New 8-azapurine derivatives were also examined for
their cytotoxic properties against MCF7, HOS, G361
and K562 cancer cell lines. A significant drop in anti-
CDK activity, previously shown to correlate with the
antiproliferative effect of 2,6,9-trisubstituted purines,26
T98G
Arn8
a Values are means of three experiments; SD values did not exceed
15%.

5402
L. Havlicek et al. / Bioorg. Med. Chem. 13 (2005) 5399–5407
structures besides CDK2 remains the subject of our fur-
ther work.
4. Experimental
4.1. General experimental procedures
Reagents were from standard commercial sources and of
analytical grade. Thin-layer chromatography (TLC) was
carried out using aluminium sheets with silica gel F254
from Merck. Spots were visualized under UV light
(254 nm). Preparative column chromatography was per-
formed on silica gel (Merck Kieselgel 60, 230–400). Melt-
ing points were determined on a Kofler block and were
uncorrected. The 1H NMR spectra (d, ppm; J, Hz) were
measured on Bruker Avance DRX 500 (500 MHz), Vari-
an VXR-400 (400 MHz) or on Varian Gemini 300
(300 MHz) spectrometers. All spectra were obtained at
25 ꢁC using tetramethylsilane as an internal standard.
Mass spectra were measured on a MS Waters/Micromass,
ZMD-detector, direct inlet, ESI, coin voltage 20 V.
Merck silica gel Kieselgel 60 (230–400 mesh) was used
for column chromatography.
Figure 2. Dose-dependent induction of p53 and p21WAF1 proteins in
MCF7 cells treated with indicated concentrations (lM) of compound 9
for 24 h. The cells were harvested and analyzed on Western blot with
specific antibodies as described in Section 4.
The CDK inhibitor roscovitine has also been shown to
induce nuclear accumulation of active wild-type tumor
suppressor protein p53 in human untransformed and
tumor-derived cells.27,28 Thus, the activation of p53-
dependent transcription after treatment with different
8-azapurine derivatives was also analyzed in the human
melanoma cell line Arn8 expressing b-galactosidase un-
der the control of a p53-responsive promoter.28 Striking
induction of wild-type p53 protein and dramatic
enhancement of p53-dependent transcription, was
observed after treatment with many 8-azapurines
(Table 1). The effect was subsequently verified in 8-azap-
urine 9 treated MCF7 cells, analyzed on western blots
using anti-p53 and anti-p21WAF1 antibodies (Fig. 2).
Interestingly, the total p53 activity induced by new pur-
ine analogs correlates much better with the observed
cytotoxicity in different cancer cell lines. These results
may highlight the therapeutic potential of purine-like
CDK inhibitors as anticancer drugs, especially in tu-
mors retaining a functional wild-type p53 pathway.
However, a recent finding that another CDK inhibitor
flavopiridol exerts its cytotoxic effect via inhibition of
CDK7 and CDK9 suggests that induction of p53 could
be a marker and not a cause of cytotoxicity.29 Despite
that, the most active 8-azapurine 19, that displayed the
highest cytotoxicity on a panel of cell lines, did influence
neither p53 activity in Arn8 cells (Table 1) nor p53 and
p21WAF1 levels in MCF7 cells (data not shown).
Synthesis of benzylazide,31 4-amino-1-benzyl-5-carbam-
oyl-1,2,3-triazole (Ia)32 (yield 75%, mp 231–234 ꢁC) and
9-benzyl-2,6-dihydroxy-8-azapurine (IIa)22 (yield 58%,
mp 290–291 ꢁC) were prepared according to the indicat-
ed references. The isopropylazide was prepared from
isopropylbromide according to the common procedure
and distilled at 60 ꢁC/850 mbar.33
4.2. Prepared compounds
4.2.1. 9-Benzyl-2,6-dichloro-8-azapurine (IIIa). 2,6-Luti-
dine (0.7 mL) was added to the cooled (ꢀ20 ꢁC) mixture
of 9-benzyl-2,6-dihydroxy-8-azapurine IIa (1.11 mmol)
and phosphoryl chloride (3 mL).22 The reaction mixture
was heated for 1 h to reach 110 ꢁC and this temperature
was maintained for 9 h. After concentration in vacuo
(temperature up to 80 ꢁC) the residue was decomposed
by ice–water mixture (0 ꢁC). The product was extracted
with chloroform, the solution was decolorized by acti-
vated charcoal, evaporated and the residue was crystal-
lized from abs. Et2O (85%; mp 92–94 ꢁC). 1H NMR
(500 MHz, CDCl3): 5.03 (2H, s, J = 6.7, CH2); 7.34–
7.39 (3H, m, Ph); 7.47 (2H, m, Ph).
3. Conclusions
In summary, the prepared trisubstituted 8-azapurines
(1,2,3-triazolo[4,5-d]pyrimidines) proved the negative
action of extra nitrogen in the purine ring on biological
activity. Previously reported correlation coefficients be-
tween CDK inhibition by roscovitine-like inhibitors
and their antiproliferative activity demonstrated an
obvious link.26 On the other hand, the values were not
very high, giving us the space to speculate that CDKs,
once believed to be specifically inhibited by 2,6,9-trisub-
stituted purines, seem not to be the one and only targets
responsible for the antiproliferative activity of purine
CDK inhibitors. Moreover, several other enzymes/pro-
teins were recently identified to interact with purines
as well as other types of CDK inhibitors.30 In the light
of the above findings the possibility of targeting other
4.2.2. 4-Amino-5-carbamoyl-1-isopropyl-1,2,3-triazole (I).
Isopropylazide (62 g, 0.73 mol) and 2-cyanoacetamide
(63 g, 0.75 mol) were added to a solution of sodium ethox-
ide (from 17.3 g Na0) in 1 L absolute ethanol. The mixture
was heated for 1 h under reflux and gentle reflux was
maintained for 20 h. The reaction mixture was concen-
trated in vacuo (up to 40 ꢁC) and the residue was dissolved
in water. The product was extracted with ethyl acetate, the
organic layer was evaporated away and the product was
crystallized from water and re-crystallized from methanol
(yield 36%, 44 g, mp 209–215 ꢁC), MS ESI+170.2 (100,
M+H+), 192.1 (30, M+Na+). 1H NMR (500 MHz,
D2O): 1.29 (6H, d, J = 6.7, (CH3)2CH); 4.29 (1H, sept,
J = 6.7, (CH3)2CH). IR (KBr): 3440 (H2NCO), 3306

L. Havlicek et al. / Bioorg. Med. Chem. 13 (2005) 5399–5407
5403
(H2NCO), 3188 (NH2), 1666, 1645, 1560, 1304, 1253,
1243, 1079, 1031,1012. Anal. (C6H11N5O) C, H, N.
4.2.7. 9-Benzyl-2-chloro-6-(3-chlorophenyl)amino-8-azap-
urine. Reaction in 1-butanol (25 ꢁC, 3 days), 80% yield
after column chromatography (toluene:chloroform,
7:3), amorphous, MS ESI+371.0 (100, M+H+). Anal.
(C17H12N6Cl2) C, H, N.
4.2.3. 2,6-Dihydroxy-9-isopropyl-8-azapurine (IIb). A
mixture of sodium ethoxide (obtained from 12 g Na0
and absolute 1.4 L EtOH) and 4-amino-5-carbamoyl-1-
isopropyl-1,2,3-triazole Ia (22 g, 0.12 mol) was warmed
to reflux. Diethyl carbonate (22 mL, 0.18 mol) was added
over 40 min and it was refluxed for another 4 h. A power-
ful mechanical stirrer was necessary for appropriate stir-
ring of the viscous mixture. After cooling, 0.4 L of water
was added and pH was adjusted to 6. The reaction mixture
was then concentrated to volume approximately 0.4 L in
vacuo and the precipitated product was filtered. Recrys-
tallization from water and then from methanol afforded
the product in 32% yield (7.2 g, mp 303–306 ꢁC), MS
ESI+218.1 (100, M+Na+), 413.2 (20, 2 M+Na+),
4.2.8.
6-Benzylamino-2-chloro-9-isopropyl-8-azapurine
(IVb). Reaction in 1-butanol (4–8 ꢁC), precipitated in
83% yield, used without column chromatography, mp
184–185 ꢁC, MS ESI+303.2 (100, M+H+). 1H NMR
(500 MHz, CDCl3 + 20% CD3OD): 1.68 (6H, d, J = 6.7,
(CH3)2CH); 4.85 (2H, s, CH2Ph); 5.10 (1H, sept,
J = 5.8, CH(CH3)2); 5.33 (1H, s, NH); 7.30–7.45 (5H,
m, Ph). Anal. (C14H15N6Cl) C, H, N.
4.2.9. 2-Chloro-6-(3-hydroxy-4-methoxybenzyl)amino-9-
isopropyl-8-azapurine (IVc). Reaction in MeOH (4–
8 ꢁC), 65% yield after column chromatography (tolu-
ene:EtOAc with a trace of NH4OH, 7:3), crystallization
from CHCl3/Et2O, mp 177–179 ꢁC, MS ESI+349.1 (100,
M+H+), ESIꢀ347.1 (100, MꢀH+). 1H NMR (500 MHz,
CDCl3 + 20% CD3OD): 1.68 (6H, d, J = 6.7, (CH3)2CH);
3.89 (3H, s, OCH3), 4.74 (2H, s, CH2Ar); 5.10 (1H, sept,
J = 6.7, (CH3)2CH); 5.23 (1H, s, NH); 6.83 (1H, d,
J = 8.0, Ar); 6.89 (1H, d, J = 8.2, Ar); 6.95 (1H, s, Ar);
7.28 (1H, s, OH). Anal. (C15H17N6ClO2) C, H, N.
1
ESIꢀ194.2 (100, MꢀH+). H NMR (500 MHz, D2O):
1.33 (6H, d, J = 6.8, (CH3)2CH); 4.56 (1H, sept, J = 6.8,
(CH3)2CH). IR (KBr): 3412 vb and 3170 sh (HO associ-
ates), 1680, 1609, 1571, 1515, 1386, 1274. Anal.
(C7H9N5O2) C, H, N.
4.2.4. 2,6-Dichloro-9-isopropyl-8-azapurine (IIIb). 2,6-
Lutidine (0.7 mL) was added to the cooled (ꢀ20 ꢁC)
mixture of 2,6-dihydroxy-9-isopropyl-8-azapurine (IIb,
1.11 mmol) and phosphoryl chloride (3 mL). The reac-
tion mixture was heated for 3 h at 120 ꢁC. After evapo-
ration in vacuo (temperature up to 80 ꢁC) the residue
was decomposed by ice–water mixture (0 ꢁC) with simul-
taneous extraction of product in benzene. The solution
was immediately dried with MgSO4, evaporated and
immediately used in reaction with appropriate amines.
4.2.10. 2-Chloro-6-(2-hydroxybenzyl)amino-9-isopropyl-
8-azapurine (IVd). Reaction in MeOH (4–8 ꢁC), precipi-
tated in 60% yield, recrystallization from MeOH, mp
198–206 ꢁC, MS ESI+319.2 (100, M+H+), ESIꢀ317.2
1
(100, MꢀH+). H NMR (500 MHz, CDCl3): 1.75 (6H,
d, J = 6.7, (CH3)2CH); 4.85 (2H, d, J = 6.5, CH2Ar);
5.18 (1H, sept, J = 6.7, (CH3)2CH); 6.85 (1H, dd, J =
7.3, J = 7.3, Ar), 7.00 (1H, d, J = 8.0, Ar), 7.24 (1H, dd,
J = 7.3, J = 8.6, Ar); 7.43 (1H, d, J = 7.5, Ar); 9.28 (1H,
bt, NH); 9.58 (1H, s, OH). Anal. (C14H15N6ClO) C, H, N.
4.2.5. General preparation of 6-alkylamino-2-chloro-9-
isopropyl(or benzyl)-8-azapurine (IV). 2,6-Dichloro
derivative IIIa,b was dissolved in a minimum amount
of abs. MeOH (benzylamine or its derivatives) or 1-bu-
tanol (3-chloroaniline), 2 eq. of triethylamine were add-
ed and the mixture was cooled to 5–10 ꢁC. After
addition of 1 alkyl amine eq. the course of the reaction
was monitored by TLC chromatography. The reaction
with less reactive amines (chloroanilines) was carried
out at room temperature. Some products precipitated
but all derivatives IV were still purified by flash chroma-
tography on Kieselgel. Details are given below, the given
yields refer to the weight of crude evaporated residue of
dichloroderivative III.
4.2.11. 2-Chloro-6-(3-hydroxybenzyl)amino-9-isopropyl-
8-azapurine (IVe). Reaction in MeOH (4–8 ꢁC), 65%
yield is after column chromatography (0–1% MeOH in
chloroform), crystallization from Et2O, mp 189–
193 ꢁC, MS ESI+319.2 (100, M+H+), ESIꢀ317.2 (100,
1
MꢀH+). H NMR (500 MHz, CDCl3 + 20% CD3OD):
1.68 (6H, d, J = 6.6, (CH3)2CH); 4.77 (2H, s, CH2Ar);
5.09 (1H, sept, J = 6.6, (CH3)2CH); 6.77 (1H, dd,
J = 7.0, J = 3.0, Ar); 6.87 (2H, m, Ar); 7.18 (1H, dd,
J = 7.8, J = 7.8, Ar). Anal. (C14H15N6ClO) C, H, N.
4.2.6. 9-Benzyl-2-chloro-6-(cyclohexylmethyl)amino-8-
azapurine (IVa). Reaction in MeOH (25 ꢁC, 16 h), 85%
yield after column chromatography in chloroform, crys-
tallization from abs. Et2O, mp 142–147 ꢁC, MS
4.2.12. 2-Chloro-6-(4-hydroxybenzyl)amino-9-isopropyl-
8-azapurine (IVf). Reaction in MeOH (4–8 ꢁC), 70%
yield after column chromatography (0–2% MeOH in
chloroform), crystallization from CHCl3/Et2O, mp
206–209 ꢁC, MS ESI+319.2 (100, M+H+), ESIꢀ317.2
(100, MꢀH+).1H NMR (500 MHz, CDCl3 + 20%
CD3OD): 1.68 (6H, d, J = 7.2, (CH3)2CH); 4.73 (2H,
s, CH2Ar); 5.09 (1H, sept, J = 7.2, (CH3)2CH); 6.81
(2H, d, J = 7.9, Ar); 7.27 (2H, d, J = 7.9); 7.30 (1H, s,
OH). Anal. (C14H15N6ClO) C, H, N.
1
ESI+357.0 (100, M+H+). H NMR (500 MHz, CDCl3):
1.05 (1H, m, C6H11); 1.24 (4H, m, C6H11); 1.62–1.87
(6H, m, C6H11); 3.54 (1.36H, dd, J = 6.8, 6.0, NHCH2);
3.97 (0.64H, dd, J = 6.8, 6.0, NHCH2); 5.69 (2H, s,
CH2Ph); 6.23 (0.32H, t, J = 6.6, NH); 6.94 (0.68H, t,
J = 6.0, NH); 7.29–7.37 (3H, meta + para Ph); 7.43
(1.36H, m, ortho Ph); 7.45 (0.64H, m, ortho Ph). The
two compounds (ratio 0.68:0.32) are rotamers (exchange
peaks in ROSY between NH and between CH2 of both
species). Anal. (C18H21N6Cl) C, H, N.
4.2.13.
2-Chloro-9-isopropyl-6-(1-phenyl-2-hydroxyeth-
yl)amino-8-azapurine (IVg). Reaction in MeOH (4–8 ꢁC),
65% yield after column chromatography (chloroform),

5404
L. Havlicek et al. / Bioorg. Med. Chem. 13 (2005) 5399–5407
crystallization from Et2O, mp 180–186 ꢁC, MS ESI+333.2
(100, M+H+). 1H NMR (500 MHz, CDCl3): 1.67 (6H, dd,
J = 7.8, J = 7.8, (CH3)2CH); 4.07 (2H, m, CH2); 5.08 (1H,
sept, J = 7.7, (CH3)2CH); 5.58 (1H, dd, J = 5.1, J = 5.1
CHPh), 7.28–7.49 (5H, m, Ph). 13C NMR (125 MHz,
CDCl3): 22.01q; 50.96d; 56.65d; 66.50t; 123.80s; 126.94d;
128.10d; 128.87d, 138.22s; 149.30s; 158.44s. Anal.
(C15H17N6ClO) C, H, N.
4.89 (1H, sept, J = 6.9, CH(CH3)2); 5.14 (1H, br s, NH);
6.54 (1H, br s, NH); 7.25–7.38 (5H, m, Ph). Anal.
(C18H25N7O) C, H, N.
4.2.19. 6-Benzylamino-2-[1-(hydroxymethyl)-2-methyl-
propyl]amino-9-isopropyl-8-azapurine (5). Reaction in
NMP (110 ꢁC, 20 h, 10 amine eq.), 64% yield after col-
umn chromatography (0–1% MeOH in chloroform),
crystallization from abs. Et2O, mp 137–142 ꢁC, MS
1
4.2.14. 2-Chloro-6-(3-chlorophenyl)amino-9-isopropyl-8-
azapurine (IVh). Reaction in 1-butanol (25 ꢁC, 3 days),
80% yield after column chromatography (toluene:chloro-
form, 7:3), amorphous, MS ESI+323.0 (100, M+H+),
ESI+370.2 (100, M+H+). H NMR (500 MHz, CDCl3):
1.01 (6H, dd, J = 5.8, J = 5.8, (CH3)2CHCH); 1.64 (6H,
dd, J = 6.7, J = 4.0, (CH3)2CHN); 1.69 (1H, br s, (CH3)2
CHCH); 1.97 (1H, s, OH); 3.70 (1H, br s, CHHOH);
3.86–3.89 (2H, bm, CHHOH + CHCH2OH); 4.77 (2H,
bm, CH2Ph); 4.89 (1H, sept, J = 6.7, NCH(CH3)2); 5.15
(1H, br s, NH); 6.47 (1H, br s, NH); 7.26–7.38 (5H, m,
Ph). Anal. (C19H27N7O) C, H, N.
1
325.0 (63, M+H+). H NMR (500 MHz, CDCl3): 1.70
(6H, d, J = 6.8, (CH3)2CH); 4.95 (1H, sept, J = 6.8,
(CH3)2CH); 5.10 (1H, br s, NH); 7.10 (1H, bd, J = 15.0,
Ar); 7.27–7.39 (3H, m, Ar). Anal. (C13H12N6Cl2) C, H, N.
4.2.15. General procedure for preparation of 2-alkylami-
no-6-alkylamino-9-isopropyl(or benzyl)-8-azapurines (V).
The appropriate 2-chloro derivative (IVa–h) was dis-
solved in 1-butanol or N-methyl-2-pyrrolidon (NMP)
(usually 0.1 g in 1.5 mL) and 5 eq. of appropriate alkyl
amine R2NH2 (unless otherwise stated) were added.
Neat amine was used in the case of less reactive 2-ami-
no-1-butanol and amino-3-methyl-1-butanol. After
heating (as specified below) the reaction mixture was
evaporated in vacuo and the residue was purified by
flash chromatography on Kieselgel. Details of individual
compounds are given below.
4.2.20.6-(3-Hydroxy-4-methoxybenzyl)amino-2-[1-(hy-
droxymethyl)propyl]amino-9-isopropyl-8-azapurine (6).
Reactionin1-butanol(100 ꢁC, 5amineeq., 2 h), 68%yield
after column chromatography (0–2% MeOH in chloro-
form), crystallization from CHCl3/abs. Et2O, mp 121–
123 ꢁC, MS ESI+402.3 (100, M+H+). 1H NMR
(300 MHz, CDCl3): 1.02 (3H, t, J = 7.3 Hz, CH3CH2);
1.68 (6H, d, J = 6.8, (CH3)2CH); 1.57 (2H, m, CH2CH3);
3.65(1H, mCHHOH); 3.81(1H, mCHHOH);3.89(3H, s,
CH3O); 4.77 (2H, br s, CH2Ar); 4.92 (1H, sept, J = 6.8,
CH(CH3)2); 5.23 (1H, br s, NH); 6.83–6.89 (2H, m, Ar);
6.95(1H,s,Ar).Anal.(C19H27N7O3)C,H,N.
4.2.16. 2-(trans-4-Aminocyclohexyl)amino-6-benzylami-
no-9-isopropyl-8-azapurine (1). Reaction in NMP
(100 ꢁC, 1 h), 72% yield after column chromatography
(0–10% MeOH in toluene with a trace of NH4OH), crys-
tallization from CHCl3/Et2O, mp 143–150 ꢁC, MS
ESI+381.3 (100, M+H+). 1H NMR (500 MHz,
CD3OD): 1.29–1.50 (4H, m, C6H11); 1.61 (6H, d,
J = 6.6, CH3)2CH; 2.00–2.20 (4H, m, C6H11); 3.00
(1H, bm, CHNH2); 3.81 (1H, bm, CHNH); 4.61 (2H,
s, CH2Ph); 7.21–7.39 (5H, m, Ph). Signal of CH(CH3)
is hidden in solvent signal. Anal. (C20H28N8) C, H, N.
4.2.21. 2-(cis-2-Aminocyclohexyl)amino-6-(3-hydroxy-4-
methoxybenzyl)amino-9-isopropyl-8-azapurine (7). Reac-
tion in 1-butanol (100 ꢁC, 20 amine eq., 2 h), 56% yield
after column chromatography (0–5% MeOH in chloro-
form with a trace of NH4OH), crystallization from abs.
Et2O, mp 118–121 ꢁC, MS ESI+427.4 (100, M+H+). 1H
NMR (400 MHz, CDCl3): 1H NMR (500 MHz,
CD3OD): 1.45–2.12 (8H, m, C6H11); 1.61 (3H, d,
J = 6.8, (CH3)CH); 1.62 (3H, d, J = 6.8, CH(CH3)); 3.56
(1H, br s, CHNH2); 3.89 (1H, s, CH3O); 4.45 (1H, br s,
CHNH); 4.74 (2H, s, CH2Ar); 4.96 (1H, m, (CH3)2CH);
6.83 (1H, d, J = 8.0, Ar); 6.89 (1H, d, J = 8.2, Ar); 6.95
(1H, d, J = 8.6, Ar). Anal. (C21H30N8O2) C, H, N.
4.2.17. 6-Benzylamino-2-(3-hydroxypropyl)amino-9-iso-
propyl-8-azapurine (3). Reaction in 1-butanol (100 ꢁC,
1.5 h), 85% yield after column chromatography (0–2%
MeOH in chloroform), crystallization from CHCl3/
Et2O, mp 123–125 ꢁC, MS ESI+342.3 (100, M+H+). 1H
NMR (300 MHz, CDCl3): 1.65 (6H, d, J = 7.0, (CH3)2
CH); 1.80 (2H, m, CH2CH2CH2); 3.65 (2H, m,
CH2CH2CH2); 4.76 (2H, br s, CH2Ph); 4.88 (1H, sept,
J = 6.6, CH(CH3)2); 5.27 (1H, br s, NH); 6.30 (1H, br s,
NH); 7.30–7.40 (5H, m, Ph). Anal. (C17H23N7O) C, H, N.
4.2.22. 6-(2-Hydroxybenzyl)amino-2-[1-(hydroxymeth-
yl)propyl]amino-9-isopropyl-8-azapurine (8). Reaction in
NMP (105 ꢁC, 10 amine eq., 2 h), 73% yield after column
chromatography (0–1% MeOH in chloroform), amor-
phous, MS ESI+372.3 (100, M+H+), ESIꢀ370.2 (100,
MꢀH+). 1H NMR (500 MHz, CDCl3): 1.07 (3H, t,
J = 7.3, CH3CH2); 1.68 (6H, dd, J = 7.6, J = 7.6,
(CH3)2CH); 1.73 (2H, m, CH3CH2,); 3.74 (1H, m,
CHHOH); 3.94–4.09 (2H, m, CHHOH + CHN); 4.47
(1H, br s, CHHPh); 4.58 (1H, br s, CHHPh); 4.96 (1H,
sept, J = 7.3, (CH3)2CH); 5.70 (1H, br s, NH); 6.79 (1H,
dd, J = 7.3, J = 7.3, Ar); 6.89 (1H, d, J = 8.0, Ar); 7.17
(1H, dd, J = 7.3, J = 7.3, Ar); 7.39 (1H, d, J = 8.0, Ar).
Anal. (C18H25N7O2) C, H, N.
4.2.18. 6-Benzylamino-2-[1-(hydroxymethyl)propyl]ami-
no-9-isopropyl-8-azapurine (4). Reaction in neat amine
(100 ꢁC, 50 amine eq., 20 h), 64% yield after column
chromatography (0–2% MeOH in chloroform), crystalli-
zation from CHCl3/Et2O, mp 122–127 ꢁC, MS
1
ESI+356.3 (100, M+H+). H NMR (300 MHz, CDCl3):
1.02 (3H, t, J = 7.4, CH3CH2); 1.56 (2H, m, CH2CH3);
1.64 (6H, d, J = 6,9, (CH3)2CH); 3.64(1H, m, CHNH);
3.77–4.02 (2H, m, AB, CH2OH); 4.77 (2H, br s, CH2Ph);
4.2.23. 6-(2-Hydroxybenzyl)amino-2-[1-(hydroxymethyl)-
2-methylpropyl]amino-9-isopropyl-8-azapurine (9). Reac-
tion in 1-butanol (120 ꢁC, 5 amine eq., 22 h), 88% yield

L. Havlicek et al. / Bioorg. Med. Chem. 13 (2005) 5399–5407
5405
after column chromatography (chloroform), crystalliza-
tion from abs. Et2O, mp 77–81 ꢁC, MS ESI+386.3 (100,
M+H+), ESIꢀ384.2 (100, MꢀH+). 1H NMR (500 MHz,
CDCl3): 1.07 (6H, d, J = 6.6, (CH3)2CHCH); 1.66 (3H,
d, J = 7.6, (CH3)CHN); 1.67 (3H, d, J = 7.6,
CH(CH3)N); 2.06 (1H, m, br s, (CH3)2CHCH); 3.71–
3.99 (3H, m AB + m, CH2HOH + CHNH); 4.60 (2H,
bd, CH2Ar); 4.93 (1H, sept, J = 6.6, NCH(CH3)2);
5.52 (1H, br s, OH); 6.81 (1H, t, J = 7.3, Ar); 6.91
(1H, d, J = 8.0, Ar); 7.19 (1H, t, J = 7.4, Ar); 7.36
(1H, d, J = 8.0, Ar); 8.50 (1H, br s, NHCH2). The pro-
ton 2D-COSY experiments were used for the assignment
of signals. Anal. (C19H27N7O2) C, H, N.
sept, J = 6.6, NCH(CH3)2); 5.41 (1H, br s, OH); 6.78
(1H, dd, J = 8.0, J = 3.0, Ar); 6.88 (2H, m, Ar); 7.18
(1H, m, Ar). Anal. (C19H27N7O2) C, H, N.
4.2.28. 2-(trans-4-Aminocyclohexyl)amino-6-(3-hydroxy-
benzyl)amino-9-isopropyl-8-azapurine (14). Reaction in
1-butanol (120 ꢁC, 20 h, 6 amine eq.), 77% yield after
column chromatography (0–10% MeOH in chloroform
with a trace of NH4OH), crystallization from abs.
Et2O, mp 109–114 ꢁC, MS ESI+397.2 (100, M+H+).
1H NMR (500 MHz, CD3OD): 1.25–1.53 (4H, m,
C6H10); 1.60 (6H, d, J = 6.6, (CH3)2CH); 1.97–2.22
(4H, m, C6H10); 3.00 (1H, m, CHNH2); 3.83 (1H, m,
CHNH); 4.69 (2H, s, CH2Ar); 5.13 (1H, m, (CH3)2CH);
6.67 (1H, m, Ar); 6.82 (2H, m, Ar); 7.12 (1H, dd, J = 7.8,
J = 7.8, Ar). Anal. (C20H28N8O) C, H, N.
4.2.24.
2-(cis-2-Aminocyclohexyl)amino-6-(2-hydroxy-
benzyl)amino-9-isopropyl-8-azapurine (10). Reaction in
1-butanol (115 ꢁC, 6 amine eq., 2 h), 72% yield after col-
umn chromatography (0–5% MeOH in chloroform with
a trace of NH4OH), amorphous, MS ESI+397.4 (100,
M+H+), MS ESIꢀ395.4 (100, MꢀH+). 1H NMR
(500 MHz, CD3OD): 1.62 (6H, dd, J = 6.7, J = 2.7,
(CH3)2CH); 1.45–2.10 (8H, m, C6H10); 3.55 (1H, br s,
CHNH2); 4.44 (1H, br s, CHNH); 4.75 (2H, s, CH2Ar);
4.96 (1H, m, (CH3)2CH); 6.81 (2H, m, Ar), 7.11 (1H, dd,
J = 7.6, J = 7.6, Ar); 7.22 (1H, d, J = 7.2, Ar). Anal.
(C20H28N8O) C, H, N.
4.2.29. 6-(3-Hydroxybenzyl)amino-2-[1-(hydroxymethyl)
propyl]amino-9-isopropyl-8-azapurine (15). Reaction in
1-butanol (100 ꢁC, 8 h, 15 amine eq.), 73% yield after col-
umn chromatography (0–3% MeOH in chloroform), crys-
tallization from abs. Et2O/pentane, mp 105–109 ꢁC, MS
1
ESI+371.2 (100, M+H+). H NMR (300 MHz, CDCl3):
1.03 (3H, t, J = 7.3 Hz, CH3CH2); 1.56 (2H, bm,
CH2CH3); 1.64 (6H, d, J = 6,9 Hz, (CH3)2CH); 3.65
(1H, m, CHNH); 3.77–4.02 (2H, bm, AB, CH2OH);
4.77 (2H, br s, CH2Ar); 4.82 (1H, sept, J = 6.9 Hz,
CH(CH3)2); 5.30 (1H, br s, NH); 6.60 (1H, br s, NH);
6.78–6.88 (3H, m, Ar); 7.30 (1H, s, Ar). Anal.
(C18H25N7O2) C, H, N.
4.2.25. 2-(trans-4-Aminocyclohexyl)amino-6-(2-hydroxy-
benzyl)amino-9-isopropyl-8-azapurine (11). Reaction in
neat amine (115 ꢁC, 4 h, 35 amine eq.), 84% yield after
column chromatography (0–8% MeOH in chloroform
with a trace of NH4OH), crystallization from CHCl3/
abs. Et2O, mp 120–123 ꢁC, MS ESI+397.4 (100,
M+H+), MS ESIꢀ395.4 (100, MꢀH+). 1H NMR
(500 MHz, CD3OD): 1.29–1.50 (4H, m, C6H10); 1.61
(6H, d, J = 6.6, CH3)2CH; 2.00–2.25 (4H, m, C6H10);
3.00 (1H, bt, CHNH2); 3.81 (1H, bm, CHNH); 4.62
(2H, s, CH2Ar); 4.90 (1H, sept, J = 6.6, (CH3)2CH);
6.81 (2H, m, Ar), 7.11 (1H, dd, J = 7.6, J = 7.6, Ar);
7.225 (1H, d, J = 7.2, Ar). Anal. (C20H28N8O) C, H, N.
4.2.30. 2-(trans-4-Aminocyclohexyl)amino-6-(4-hydroxy-
benzyl)amino-9-isopropyl-8-azapurine (16). Reaction in
neat amine (115 ꢁC, 4 h, 35 amine eq.), 84% yield after col-
umn chromatography (0–8% MeOH in chloroform with a
trace of NH4OH), crystallization from abs. Et2O, mp
118–122 ꢁC, MS ESI+397.2 (100, M+H+). 1H NMR
(500 MHz, CD3OD): 1.25–1.55 (4H, m, C6H10); 1.62
(6H, d, J = 6.8, (CH3)2CH); 2.00–2.20 (4H, m, C6H10);
3.08 (1H, m, CHNH2); 3.85 (1H, m, CHNH), 4.64 (2H,
s, CH2Ar); 5.10 (1H, m, (CH3)2CH); 5.43 (1H, s, NH);
6.750 (2H, d, J = 8.2, Ar); 7.210 (2H, d, J = 8.3, Ar). Anal.
(C20H28N8O) C, H, N.
4.2.26. 2,6-Bis[(2-Hydroxybenzyl)amino]-9-isopropyl-8-
azapurine (12). Reaction in NMP (100 ꢁC, 2 h), 68% yield
after column chromatography (chloroform), crystalliza-
tion from chloroform/abs. Et2O, mp 215–217 ꢁC, MS
4.2.31. 2-Hexylamino-9-isopropyl-6-(1-phenyl-2-hydroxy-
ethyl)amino-8-azapurine (17). Reaction in 1-butanol
(100 ꢁC, 2 h), 85% yield after column chromatography
(chloroform), amorphous, MS ESI+398.4 (100, M+H+).
1H NMR (400 MHz, CDCl3): 0.90 (3H, t, J = 6.7,
CH3CH2), 1.31 (6H, m, (CH2)3), 1.63 (2H, m,
CH2CH2N), 1.67 (6H, dd, J = 7.8, J = 7.8, (CH3)2CH);
3.32 (2H, dt, J = 6.6, J = 6.6, NHCH2); 4.07 (2H, m,
CH2OH); 5.00 (1H, sept, J = 7.8, (CH3)2CH); 5.58 (1H,
dd, J = 5.1, J = 5.1 CHPh); 7.28–7.49 (5H, m, Ph). Anal.
(C21H31N7O) C, H, N.
1
ESI+406.3 (100, M+H+), ESIꢀ404.3 (100, MꢀH+). H
NMR (500 MHz, CDCl3 + 20% CD3OD): 1.68 (6H, d,
J = 6.6, (CH3)2CH); 4.57 (2H, d, CH2Ar); 4.62 (2H, s,
CH2Ar); 4.98 (1H, sept, J = 6.7, (CH3)2CH); 6.83–6.92
(4H, m, Ar), 7.19 (2H, dd, J = 7.4, J = 7.4, Ar); 7.27
(2H, d, J = 6.8, Ar). Anal. (C21H23N7O2) C, H, N.
4.2.27. 6-(3-Hydroxybenzyl)amino-2-[1-(hydroxymethyl)-
2-methylpropyl]amino-9-isopropyl-8-azapurine
(13).
Reaction in 1-butanol (100 ꢁC, 22 h, 10 amine eq.),
67% yield after column chromatography (0–3% MeOH
in chloroform), crystallization from abs. EtOAc, mp
150–152 ꢁC, MS ESI+386.2 (100, M+H+), ESIꢀ384.2
4.2.32. 6-(3-Chlorophenyl)amino-2-[1-(hydroxymethyl)-2-
methylpropyl]amino-9-isopropyl-8-azapurine (18). Reac-
tion in 1-butanol (100 ꢁC, 22 h, 10 amine eq.), yield 56%
after column chromatography in chloroform, crystalliza-
tion from abs. Et2O, mp 160–163 ꢁC, MS ESI+390.2 (100,
M+H+). 1H NMR (300 MHz, CDCl3): 1.03 (3H, d,
J = 7.0, (CH3)2CHCH); 1.05 (3H, d, J = 7.0,
1
(100, MꢀH+). H NMR (500 MHz, CDCl3): 1.07 (6H,
d, J = 6.6, (CH3)2CHCH); 1.66 (3H, d, J = 7.6,
(CH3)CHN); 1.67 (3H, d, J = 7.7, CH(CH3)N); 2.06
(1H, m, br s, (CH3)2CHCH); 3.70–3.96 (3H, m AB+m,
CH2HOH + CHNH); 4.60 (2H, bd, CH2Ph); 4.93 (1H,

5406
L. Havlicek et al. / Bioorg. Med. Chem. 13 (2005) 5399–5407
(CH3)2CHCH); 1.64 (6H, d, J = 6.7, (CH3)2CHN); 1.78
(1H, br s, OH); 2.01 (1H, m, J = 7.0, CHCH(CH3)2);
3.68–4.02 (3H, m, HOCH2CH); 4.87 (1H, sept, J = 6.7,
(CH3)2CHN); 5.30 (1H, br s, NH); 7.09 (1H, d, J = 7.7,
Ar); 7.23–7.25 (2H, m, Ar); 7.48 (1H, m, Ar); 8.00
(1H, br s, NH). Anal. (C18H24N7ClO) C, H, N.
9-isopropyl-2-methyl-8-azapurine (1.2 mmol, 0.23 g)
were suspended in chloroform (1.3 mL) and heated un-
der reflux for 90 min (the mixture became clear nearly
immediately). After concentration in vacuo (tempera-
ture up to 80 ꢁC) the residue was decomposed by ice–
water mixture (0 ꢁC) with simultaneous extraction of
the product into benzene. The benzene solution was
dried with MgSO4, evaporated and dissolved in 1-buta-
nol (3 mL). Benzylamine (3 mmol, 0.3 mL) was added
and the reaction mixture was heated to 100 ꢁC for
90 min and then concentrated in vacuo. Product 2 was
obtained after flash chromatography on Kieselgel (1%
MeOH in chloroform), yield 29%, crystallization from
Et2O, mp 133–139 ꢁC, MS 283.1 (100, M+H+). 1H
NMR 1.69 (6H, d, J = 6.6, (CH3)2CH); 2.63 (3H, s,
CH3); 4.88 (2H, br s, CH2Ph); 5.12 (1H, sept, J = 6.6,
CH(CH3)2); 7.30–7.43 m (5H, Ph). Anal. (C15H18N6)
C, H, N.
4.2.33. 2-(trans-4-Aminocyclohexyl)amino-6-(3-chloro-
phenyl)amino-9-isopropyl-8-azapurine (19). Reaction in
neat amine (115 ꢁC, 4 h, 35 amine eq.), yield 89% after
column chromatography (0–8% MeOH in chloroform
with a trace of NH4OH), crystallization from abs.
EtOAc/ abs. Et2O, mp 221–223 ꢁC, MS ESI+401.3
1
(100, M+H+). H NMR (300 MHz, CDCl3): 1.20–1.36
(4H, m, C6H10); 1.66d (6H, d, J = 6.7, (CH3)2CH);
1.93 (2H, m, C6H10); 2.20 (2H, m, C6H10); 2.72 (1H,
m, CHNH2); 3.85 (1H, m, CHNH); 4.93 (1H, sept,
J = 6.7, (CH3)2CH); 5.08 (1H, d, NH); 7.09 (1H, d,
J = 7.8, Ar); 7.23–7.31 (2H, m, Ar); 7.48 (1H, m, Ar).
Anal. (C19H25N8Cl) C, H, N.
4.3. Kinase inhibition assay
4.2.34.
9-Benzyl-6-(3-chlorophenyl)amino-2-(trans-4-
Human 6· His-tagged cyclin E/Cdk2 complex was pro-
duced in Sf9 insect cells co-infected with appropriate
baculoviral constructs. The cells were harvested 70 h
post infection in lysis buffer (50 mM Tris pH 7.4,
150 mM NaCl, 5 mM EDTA, 20 mM NaF, 1% Tween
20, protease inhibitors) for 30 min on ice and the soluble
fraction was recovered by centrifugation at 14.000g for
10 min. The kinase was purified on Ni–NTA column
(Qiagen) according to manufacturerÕs instructions,
stored at 4 ꢁC and used within a week. To carry out
experiments on kinetics under linear conditions, the final
point test system for kinase activity measurement was
used. The assay mixture contained 1 mg/mL histone
(Sigma Type III-S), 15 lM ATP, 0,2 lCi [c-33P]ATP
and tested compound in a final volume of 10 lL, all in
reaction buffer: 50 mM Hepes 7,4 pH, 10 mM MgCl2,
5 mM EGTA, 10 mM 2-glycerolphosphate, 1 mM
NaF, 1 mM DTT and protease inhibitors. After
10 min, the incubations were stopped by adding 5%
H3PO4 and spotted on P81 phosphocellulose paper
(Whatman). After washing in 5% H3PO4, the kinase
activity was measured by digital imaging analyzer
BAS-1800 (Fujifilm). The kinase activity was expressed
as a percentage of maximum activity, the IC50 values
were determined by graphic analysis.
hydroxycyclohexyl)amino-8-azapurine (20). Reaction in
NMP (120 ꢁC, 22 h), yield 66% after column chromato-
graphy (0–2% MeOH in chloroform), crystallization
from EtOH, mp 209 ꢁC, MS ESI+450.3 (100, M+H+).
1H NMR (500 MHz, CDCl3 + 20% CD3OD): [1.21–
1.40 (2H, m); 1.43–1.60 (2H, bs); 2.00–2.10 (2H, bd);
2.14–2.23 (2H, bd); C6H10]; 3.65 (1H, m, CHOH); 4.80
(1H, m, CHNH); 5.58 (2H, s, CH2Ar); 7.10 (1H, d,
J = 7.5, Ar), 7.28–7.40 (7H, m, Ar), 7.53 (1H, bm, Ar).
Anal. (C23H24N7Cl) C, H, N.
4.2.35. 9-Benzyl-6-(cyclohexylmethyl)amino-2-(trans-4-
hydroxycyclohexyl)amino-8-azapurine (21). Reaction in
NMP (120 ꢁC, 22 h), yield 65% after column chromato-
graphy (0–2% MeOH in chloroform), crystallization
from chloroform, mp 154–156 ꢁC, MS ESI+436.3 (100,
M+H+). H NMR: (500 MHz, CDCl3 + 20% CD3OD):
1
[0.98–1.10 (2H, m); 113–1.30 (6H, m); 1.37–1.50 (2H,
m); 1.60–1.85 (4H, m); 1.94–2.20 (5H, m) C6H10,
C6H11]; 3.39 (2H, s, CH2C6H11); 3.64 (1H, m, CHOH);
3.82 (1H, m, CHNH); 5.53 (2H, s, CH2Ph); 7.26–7.40
(5H, m, Ph). Anal. (C24H33N7O) C, H, N.
4.2.36. 2-Methyl-6-hydroxy-9-isopropyl-8-azapurine. Iso-
propylazide (13 mmol, 1.1 g,) and 2-cyanoacetamide
(10 mmol, 0.85 g) were added to a solution of sodium
ethoxide (from 0.28 g Na0, 24 mmol) in 1 L abs. EtOH.
The mixture was warmed for 1 h under reflux and gentle
reflux was maintained for 2 h. Anhydrous EtOAc was
added and the reflux was continued for other 9 h. The
reaction mixture was then concentrated in vacuo (up
to 40 ꢁC), the residue was dissolved in water and was
adjusted to pH 6. The product was extracted with chlo-
roform, solution was dried and concentrated in vacuo.
Crystallization from methanol, yield 31% (0.59 g), mp
205–213 ꢁC, MS ESIꢀ192.2 (100, MꢀH+). Anal.
(C8H11N5O) C, H, N.
4.4. Anticancer activity in vitro
The tumor cells (purchased from The American Type
Culture Collection) were grown in DMEM medium
(Gibco-BRL) supplemented with 10% (v/v) fetal bovine
serum and L-glutamine (0.3 g/L) and maintained at
37 ꢁC in a humidified atmosphere with 5% CO2. For
anticancer cytotoxicity estimation, 104 cells were seeded
into each well of a 96-well plate, allowed to stabilize for
at least 2 h and the tested inhibitors were added at var-
ious concentrations in triplicate. Tested compounds
were dissolved in DMSO to achieve 100 mM solution
and concentration series were added to cells. Maximum
concentration of DMSO in the assay never exceeded
0.1%. Three days after drugs, a Calcein AM (Molecular
Probes) solution was added and allowed to enter the
4.2.37. 6-Benzylamino-9-isopropyl-2-methyl-8-azapurine
(2). Thionylchloride (10 mmol, 0.8 mL), abs. N,N-di-
methylformamide (1.5 mmol, 0.12 mL) and 6-hydroxy-

L. Havlicek et al. / Bioorg. Med. Chem. 13 (2005) 5399–5407
5407
8. Fischer, P. M. Curr. Opin. Drug Discov. Dev. 2001, 4, 623.
9. Sielecki, T. M.; Boyan, J. F.; Benfield, P. A.; Trainor, G.
L. J. Med. Chem. 2000, 43, 1.
cells for 1 h. Fluorescence of viable cells was quantified
on Fluoroskan Ascent (Microsystems) and cytotoxic
effective concentrations were expressed as GI50Õs sub-
tracted from dose-response curves.
10. Vesely, J.; Havlicek, L.; Strnad, M.; Blow, J. J.; Donella-
Deana, A.; Pinna, L.; Letham, D. S.; Kato, J.; Detivaud,
L.; Leclerc, S.; Meijer, L. Eur. J. Biochem. 1994, 224, 771.
11. Havlicek, L.; Hanus, J.; Vesely, J.; Leclerc, S.; Meijer, L.;
Shaw, G.; Strnad, M. J. Med. Chem. 1997, 40, 408.
12. Gray, N. S.; Wodicka, L.; Thunnissen, A. M.; Norman, T.
C.; Kwon, S.; Espinoza, F. H.; Morgan, D. O.; Barnes, G.;
LeClerc, S.; Meijer, L.; Kim, S. H.; Lockhart, D. J.;
Schultz, P. G. Science 1998, 281, 533.
13. Dreyer, M. K.; Borcherding, D. R.; Dumont, J. A.; Peet,
N. P.; Tsay, J. T.; Wright, P. S.; Bitonti, A. J.; Shen, J.;
Kim, S. H. J. Med. Chem. 2001, 44, 524.
14. Krystof, V.; Lenobel, R.; Havlicek, L.; Kuzma, M.;
Strnad, M. Bioorg. Med. Chem. Lett. 2002, 12, 3283.
15. Shum, P. W.; Peet, N. P.; Weintraub, P. M.; Le, T. B.;
Zhao, Z.; Barbone, F.; Cashman, B.; Tsay, J.; Dwyer, S.;
Loos, P. C.; Powers, E. A.; Kropp, K.; Wright, P. S.;
Bitonti, A.; Dumont, J.; Borcherding, D. R. Nucleos.
Nucleot. Nucl. Acids 2001, 20, 1067.
4.5. p53-dependent transcriptional activity
In order to measure p53-dependent transcriptional activ-
ity, b-galactosidase activity was determined for the hu-
man melanoma cell line Arn8, which was established
using stable transfection of cell line A375 with p53-
responsive reporter construct pRGCDfoslacZ.28 After
24 h incubation with compounds in a 96-well microtitre
plate, the cells were fixed with 2% formaldehyde and
0.2% glutaraldehyde, washed with PBS and developed
in X-gal solution (0.2 mg/mL in PBS) overnight. Positive
cells were scored under a light microscope.
4.6. SDS–polyacrylamide gel electrophoresis and
immunoblotting
16. OÕConnor, D. S.; Wall, N. R.; Porter, A. C.; Altieri, D. C.
Cancer Cell 2002, 2, 43.
17. McClue, S. J.; Blake, D.; Clarke, R.; Cowan, A.; Cum-
mings, L.; Fischer, P. M.; MacKenzie, M.; Melville, J.;
Stewart, K.; Wang, S.; Zhelev, N.; Zheleva, D.; Lane, D.
P. Int. J. Cancer 2002, 102, 463.
18. Gibson, A. E.; Arris, C. E.; Bentley, J.; Boyle, F. T.;
Curtin, N. J.; Davies, T. G.; Endicott, J. A.; Golding, B.
T.; Grant, S.; Griffin, R. J.; Jewsbury, P.; Johnson, L. N.;
Mesguiche, V.; Newell, D. R.; Noble, M. E.; Tucker, J. A.;
Whitfield, H. J. J. Med. Chem. 2002, 45, 3381.
19. Moravcova, D.; Krystof, V.; Havlicek, L.; Moravec, J.;
Lenobel, R.; Strnad, M. Bioorg. Med. Chem. Lett. 2003,
13, 2989.
For direct immunoblotting, total cellular protein lysates
were prepared by harvesting cells in hot electrophoresis
sample buffer. Proteins were then separated by SDS–poly-
acrylamide gel electrophoresis on a 12.5% gel and trans-
ferred onto a nitrocellulose membrane. The membrane
was blocked in 5% low fat milk and 0.1% Tween 20 in
PBS for 2 h and probed overnight with monoclonal anti-
bodies against p53 (clone DO-1), PCNA (clone PC-10)
and p21WAF1/CIP1 (clone 118). After washing three times
in PBS plus 0.1% Tween 20, peroxidase conjugated rabbit
antimouse immunoglobulin antiserum was applied as the
secondary antibody. To visualize peroxidase activity,
ECL+ (Amersham Biosciences) reagents were used
according to the manufacturerÕs instructions.
20. Capek, P.; Otmar, M.; Masojidkova, M.; Votruba, I.;
Holy, A. Collect. Czech. Chem. Commun. 2003, 68, 779.
21. Robins, R. K.; Dille, K. L.; Christensen, B. E. J. Org.
Chem. 1954, 19, 930.
22. Bredereck, H.; Baumann, W. Liebigs Ann. Chem. 1967,
701, 143.
Acknowledgments
23. Biagi, G.; Giorgi, I.; Livi, O.; Scartoni, V.; Breschi, C.;
Martiny, C.; Scatizzi, R. Farmaco 1995, 50, 659.
24. Arris, C. E.; Boyle, F. T.; Calvert, A. H.; Curtin, N. J.;
Endicott, J. A.; Garman, E. F.; Gibson, A. E.; Golding, B.
T.; Grant, S.; Griffin, R. J.; Jewsbury, P.; Johnson, L. N.;
Lawrie, A. M.; Newell, D. R.; Noble, M. E. M.; Sausville,
E. A.; Schultz, R.; Yu, W. J. Med. Chem. 2000, 43, 2797.
25. Imbach, P.; Carparo, H. G.; Furet, P.; Mett, H.; Meyer,
T.; Zimmerman, J. Bioorg. Med. Chem. Lett. 1999, 9, 91.
26. Vermeulen, K.; Strnad, M.; Krystof, V.; Havlicek, L.; Van
der Aa, A.; Lenjou, M.; Nijs, G.; Rodrigus, I.; Stockman,
B.; Van Onckelen, H.; Van Bockstaele, D. R.; Berneman,
Z. N. Leukemia 2002, 16, 299.
We thank D.P. Lane for baculoviruses, R. Hrabal for
NMR analyses, K. Faksova and E. Friesnerova for tech-
nical assistance and A. Oulton for English corrections.
The work was supported by grant No. B6137301 of the
Grant Agency of the ASCR, by Grant No. 301/05/0418
of the Czech Grant Agency (L.H.), and by grant No.
MSM 6198959216 of the Czech Ministry of Education.
References and notes
1. Sausville, E. A. Trends Mol. Med. 2002, 8, S32.
2. Chen, Y.-N. P.; Sharma, S. K.; Ramsey, T. M.; Jiang, L.;
Martin, M. S.; Baker, K.; Adams, P. D.; Bair, K. W.;
Kaelin, W. G. Proc. Natl. Acad. Sci. U.S.A 1999, 96, 4325.
3. Lees, J. A.; Weinberg, R. A. Proc. Natl. Acad. Sci. U.S.A
1999, 96, 4421.
4. Tetsu, O.; McCormick, F. Cancer Cell 2003, 3, 233.
5. Hinds, P. W. Cancer Cell 2003, 3, 305.
6. Moore, J. D.; Kirk, J. A.; Hunt, T. Science 2003, 300, 987.
7. Napolitano, G.; Majello, B.; Lania, L. Int. J. Oncol. 2002,
21, 171.
27. David-Pfeuty, T. Oncogene 1999, 18, 7409.
28. Kotala, V.; Uldrijan, S.; Horky, M.; Trbusek, M.; Strnad,
M.; Vojtesek, B. Cell. Mol. Life Sci. 2001, 58, 1333.
29. Demidenko, Z. N.; Blagosklonny, M. V. Cancer Res.
2004, 64, 3653.
30. Knockaert, M.; Meijer, L. Biochem. Pharmacol. 2002, 64,
819.
31. Curtius, T.; Ehrhart, G. Chem. Ber. 1922, 55, 1565.
32. Hoover, J. R. E.; Day, A. R. J. Am. Chem. Soc. 1956, 78,
5832.
33. Boyer, J. H.; Hamer, J. J. Am. Chem. Soc. 1955, 77, 951.