8-Ethynylxanthines as promising antiproliferative agents, angiogenesis inhibitors, and calcium channel activity modulators

Article abstract of DOI:10.1007/s10593-020-02730-4

[Figure not available: see fulltext.] Synthetic procedures for the preparation of 8-ethynylxanthines by treating 8-bromocaffeine and 8-bromopentoxifylline with terminal acetylenes were elaborated. Certain ethynylxanthine derivatives exhibit high in vitro antiproliferative activity against a panel of cancer cell lines, matrix metalloproteinase and in vitro angiogenesis inhibitory activity. Ca²⁺ channel blocking and agonist activity of the synthesized ethynylxanthines was discussed based on data obtained on the H9C2, SH-SY5Y, and A7R5 cell lines.

Full text of DOI:10.1007/s10593-020-02730-4

DOI 10.1007/s10593-020-02730-4
Chemistry of Heterocyclic Compounds 2020, 56(6), 776–785
8-Ethynylxanthines as promising antiproliferative agents,
angiogenesis inhibitors, and calcium channel activity modulators
Pavel Arsenyan1*, Jelena Vasiljeva1, Ilona Domracheva1, Iveta Kanepe-Lapsa1
1 Latvian Institute of Organic Synthesis,
21 Aizkraukles St., Riga LV-1006, Latvia; e-mail: pavel@osi.lv
Published in Khimiya Geterotsiklicheskikh Soedinenii,
2020, 56(6), 776–785
Submitted October 16, 2019
Accepted after revision April 15, 2020
Synthetic procedures for the preparation of 8-ethynylxanthines by treating 8-bromocaffeine and 8-bromopentoxifylline with terminal
acetylenes were elaborated. Certain ethynylxanthine derivatives exhibit high in vitro antiproliferative activity against a panel of cancer
cell lines, matrix metalloproteinase and in vitro angiogenesis inhibitory activity. Ca2+ channel blocking and agonist activity of the synthe-
sized ethynylxanthines was discussed based on data obtained on the H9C2, SH-SY5Y, and A7R5 cell lines.
Keywords: caffeine, xanthine, angiogenesis, calcium channel, cancer, cytotoxicity.
Cancer is one of the main causes of death in econo-
mically developed countries.1 In 2017, 1.68 million new
cancer cases and 600,920 cancer deaths were projected to
occur only in the US, which has a population of
approximately 324 million people.2,3 In clinical practice,
treatment of cancer involves a wide range of chemotherapy
drugs. However, most of them exhibit various side effects,
high toxicity, and moderate selectivity. Therefore, the
development of a new generation of selective anticancer
agents with low toxicity is one of the main goals for
medicinal chemistry and pharmaceutical industry. Despite
advanced studies on the elaboration of anticancer drugs,
treatment outcome for malignant brain tumors remains
poor. Unfortunately, most of the promising antitumor drugs
used against brain tumors in clinical trials had limited
impact on human survival, which was due to the lack of
efficient drug delivery to the target in the central nervous
system (CNS) at a sufficient concentration.4–6
The increased interest in xanthines stems from the
prevalence of this heterocyclic system in many natural
substances and from the ability of methylxanthine
derivatives to penetrate the blood-brain barrier (BBB).
Xanthines are an important class of compounds with a wide
range of pharmacological effects, including anticancer,
antiHIV, anticoagulant, antispasmodic, and antibacterial
activity. Currently, several anticancer drugs based on the
structure of natural purine analogs are used in clinical
practice (e.g., cladribine, fludarabine, mercaptopurine, thio-
guanine, clofarabine, and nelarabine). These compounds
are the first-line therapeutic agents for malignant hemato-
logic diseases. Agents with 80% cure rate after a course of
monochemotherapy are considered to have a therapeutic
effect of complete remission. Purine analogs, which
include intracellular enzyme inhibitors, act as anti-
metabolites by replacing the natural nucleosides in DNA
and RNA synthesis. Unfortunately, these drugs show a
wide range of side effects, and their high treatment costs
limit their potential use in clinical practice.
To minimize side effects, antitumor drugs based on the
caffeine core are currently being developed. Caffeine
derivatives possess CNS expression as Ca2+ agonists or
antagonists. Recent studies have shown that daily
consumption of caffeinated coffee can lower mouth and
brain cancer formation up to 39%. Additionally, it reduces
the risk of cervical cancer development in women. Caffeine
reduces the possibility of prostate cancer formation in men
by 60% and can prevent breast, colon, and hepatic cancer.7–9
Caffeine also enhances doxorubicin and cisplatin activity in
metastatic carcinomas, lymphomas, and bone and soft
tissue sarcomas.10,11 Additionally, caffeine effectively inhibits
breast cancer resistance protein (BCRP)-mediated multidrug
resistance (MDR) in the MCF-7 and MCF-7/MX100
(mitoxantrone-resistant) cell lines.12 One of the advantages
of caffeine derivatives is their ability to cross the BBB and
0009-3122/20/56(6)-0776©2020 Springer Science+Business Media, LLC
776

Chemistry of Heterocyclic Compounds 2020, 56(6), 776–785
excess N-bromosuccinimide (NBS) in CH2Cl2 in analogy to
a published procedure.15 After 5 days of stirring, the
desired 8-bromocaffeine (1) and 8-bromopentoxifylline (2)
were isolated in almost quantitative yields (99 and 95%,
respectively).
Our synthetic strategy for the introduction of the ethynyl
moiety at position 8 of the xanthine core was based on
Sonogashira-type coupling reactions (Table 1). We found
Table 1. General procedure toward the synthesis
of compounds 3–18
Figure 1. Structures of some drugs used in neuro-oncology.
enter CNS. Consequently, these compounds have potential
to treat malignant diseases in the brain, such as neuro-
blastoma and glioblastoma multiforme.13 By contrast, some
drugs used in neuro-oncology (e.g., Temodar, carmustine
(BCNU), lomustine (CCNU), Fig. 1) have a limited ability
to cross the BBB and are highly toxic.
Caffeine is absorbed rapidly and completely from the
intestinal tract, making it 100% bioavailable. Antitumor
agents have been developed from various caffeine analogs,
including 8-[3-(dimethylamino)propoxy]caffeine (Proxyfeine).
However, introduction of these compounds into the clinic
in the EU and the US has been hindered due to their high
levels of toxicity (LD50 355 mg/kg) and numerous serious
side effects. In Russia and other countries, Proxyfeine is
used in chemotherapy for cancer patients with a high risk
of brain metastases and in the rehabilitation of the
metastatic lesions to the brain, as well as during the early
stages of cancer metastasis treatment.
Based on these considerations, we have contemplated
the design of 8-ethynylxanthines. Here, to address the
question of whether these compounds are promising
candidates for intensive studies as antitumor agents, we
report the elaboration of synthetic procedures for ethynyl-
xanthines and data on the in vitro antiproliferative activity
in various malignant cell lines and on the matrix metallo-
proteinase (MMP) and angiogenesis inhibitory activity
in vitro of these and reference compounds (caffeine,
pentoxifylline, Proxyfeine, and Temodar). Because caffeine
derivatives are known to act as Ca2+ channel modulators,
we present experimental data on this activity induced by
the studied compounds.
The substitution of position 8 in xanthines is a common
way to prevent their oxidation during metabolism. A similar
strategy was used to design the structure of Proxyfeine.
This molecule consists of a caffeine core and a flexible
dimethylaminopropoxy substituent at position 8. By intro-
ducing a triple bond into the linker between the caffeine
moiety and an amine, we attempted to reduce molecular
flexibility while increasing stability to metabolism.
Previously, ethynylxanthines were obtained by reacting
5-chlorouracil with aminoacetonitrile.14 We decided to
utilize the Pd-catalyzed cross-coupling reaction of terminal
acetylenes and 8-bromocaffeine and 8-bromopentoxi-
fylline, as it is the most common procedure for achieving
this goal. Thus, caffeine and pentoxifylline were chosen as
starting materials. Both xanthines were brominated using
Yield, %
Method
Com-
R1
R2
pound
I
42
II
31
III
72
IV
55
Me
Me
C(Me)2OH
3
4
11
49
54
Me
7
8
16
43
48
5
Me
Me
CH2NMe2
4
47
30
58
33
6
7
CH2N(CH2CH2OMe)2
10
Me
52
57
8
Me
35
30
9
Me
34
10
Me
38
44
42
11
12
13
(CH2)4COMe
(CH2)4COMe
C(Me)2OH
47
(CH2)4COMe
45
14
(CH2)4COMe CH2N(CH2CH2OMe)2
(CH2)4COMe
35
41
15
16
(CH2)4COMe
39
36
17
18
Me
Ph
777

Chemistry of Heterocyclic Compounds 2020, 56(6), 776–785
that the product yields in the Pd-catalyzed cross coupling
Proxyfeine, and pentoxifylline were used as reference
compounds. Although antitumor activity of Temodar is
more pronounced in in vivo models, this drug was used as a
reference compound as well. Caffeine and pentoxifylline
exhibited no cytotoxic effect in all the studied cell lines.
Moreover, despite the widespread use of Temodar as a drug
for brain cancer treatment, the in vitro results showed that
the compound had only modest activity against the studied
tumor cell lines (IC50 298 μM on SH-SY5Y neuroblastoma
cells and IC50 31 μM on U937 lymphoma cells). In fact, the
studied malignant cells were more sensitive to Proxyfeine
(IC50 96–283 μM). We observed that 8-ethynylcaffeines
had a strong ability to suppress cancer cell growth. Thus,
caffeine derivatives 8–10 showed a pronounced in vitro
antiproliferative effect on the majority of cancer cells.
Moreover, these compounds were active against neuro-
blastoma and brain tumors (IC50 8.2–14 μM on the SH-SY5Y
and C6 cell lines). Notably, caffeine derivative 5 selectively
inhibits lymphoma U937 cell growth (IC50 5.1 μM), and
derivative 8 inhibits the growth of estrogen-positive human
breast adenocarcinoma MCF-7 cells (IC50 5.7 μM).
Piperidylmethyl derivative of pentoxifylline 17 showed
high cytotoxicity against the studied cancer cells, especially
on the lymphoma U937 (IC50 4.7 μM) and glioma C6
(IC50 7.1 μM) cell lines.
of terminal acetylenes and 8-bromoxanthines strongly
depend on the nature of the catalyst and solvent. Based on
the experimental data, the reaction of terminal acetylenes
with compounds 1 and 2 under routine experimental
conditions (methods I and II) in general led to the
formation of the corresponding 8-ethynylxanthines 3–18 in
very low yields. Surprisingly, we found that a mixture
N-methylpyrrolidine–PhMe, 1:1 (method III) as the
reaction media afforded the desired products in higher
yields. Considering the goal in synthetic chemistry of
developing more eco-friendly procedures, an alternative
method was elaborated (method IV). Performing the
procedure in nontoxic EtOAc and using 2 mol % Pd(PPh3)4
and 2 mol % PdCl2 decreases the cost of the reaction and
simplifies the isolation of the product.16
In vitro antiproliferative activity
After the preparation of a set of 8-ethynylxanthines
according to our designed synthetic approach, their activity
against malignant tumor cell lines was established. The
in vitro cytotoxicity of the 8-ethynylxanthines was tested on
the following monolayer tumor cell lines: MDA-MB-435s
human melanoma, H9C2 rat embryo cardiomyoblasts,
MCF-7 estrogen-positive human breast adenocarcinoma,
HepG2 human hepatocellular carcinoma, SH-SY5Y human
neuroblastoma, C6 rat glioma, U937 human histiocytic
leukemia, and A549 human lung carcinoma. The results of
these experiments are summarized in Table 2. Caffeine,
MMP and angiogenesis inhibition
MMPs, also known as matrixins, hydrolyze components
of the extracellular matrix. These proteinases play a central
role in many biological processes, such as embryogenesis,
normal tissue remodeling, wound healing, and angio-
genesis, and in various diseases, such as atheroma, arthritis,
cancer, and tissue ulceration. MMP-2 (gelatinase A) is very
active in degrading the extracellular matrix. The inhibition
of MMPs by the synthesized 8-ethynylxanthines was
detected using an MMP Inhibitor Fluorimetric Profiling Kit
with NNGH (N-isobutyl-N-(4-methoxyphenylsulfonyl)glycyl
hydroxamic acid) as a standard. According to our data,
reference compounds such as caffeine, Proxyfeine, pentoxi-
fylline, and Temodar did not inhibit any of the studied
human MMPs (Table 3). By contrast, novel caffeine
derivatives 3, 11, and 14 exhibit the ability to selectively
inhibit MMP-2 (gelatinase A) up to 52%. Besides,
(aminocyclohexyl)ethynyl derivative 5 inhibits MMP-10
activity up to 53%. Based on the MMP inhibition data, five
compounds 3, 5, 11, 14, and 18 were chosen for in vitro
angiogenesis inhibition studies. The in vitro experiments
were performed on the Matrigel (BD Biosciences) human
umbilical vein endothelial cell (HUVEC) tube formation
model. According to the obtained experimental results,
ethynylxanthines 3, 5, and 11 activate angiogenesis by 40–
74%, whereas (aminocyclohexyl)ethynyl-substituted pentoxi-
fylline derivative 14 and phenylethynyl-substituted caffeine
derivative 18 exhibit the ability to suppress angiogenesis
by 38 and 100%, respectively, at a concentration of 20 μM
(Fig. 2). There is no direct correlation between angio-
genesis and MMP inhibition. Accordingly, we presume that
the tested compounds modulate the formation of new blood
vessels by another mechanism.
Table 2. In vitro cytotoxicity (IC50*, µM) caused by
MDA-
Compound
H9C2 MCF-7HepG2 SH-SY5Y C6 U937 A549
MB-435s
–***
283
–
Caffeine
Proxyfeine
Pentoxifylline
Temodar
–
–
–
–
135
–
–
–
–
–
111
–
298
83
–
–
–
–
–
–
–
96 97
–
–
–
31
515
119
–
–
–
–
141
–
–
–
–
–
261 79 239
187 95
3
4
–
13
28
–
41
57
–
36
25
–
96
28
–
41
25
–
15 5.1 52
28 9.2 42
5
6
7
–
–
–
8.6
14
8.2
–
–
–
80
7.3
19
75
68
8.2
–
22
–
46
9.8
52
51
5.7
14
8.2
–
169
259
73
4.9
11.8
44
17
60
13.7
–
86
–
96
12
26
34
8.6
14
8.2
–
105
116
64
4.9
9.5
17
11.5 26 8.6
8.5 5.7 17
8.2 5.5 8.2
8
9
10
11
12
13
14
16
17
18
–
124
–
277 83 188
183 85 259
20 87
78
17 12 7.3
7.1 4.7 11.8
17 13.6 85
122
* IC50 – concentration (µM) providing 50% cell killing effect (MTT).
** Tumor cell lines: MDA-MB-435s human melanoma, H9C2 rat
embryo cardiomyoblasts, MCF-7 estrogen-positive human breast adeno-
carcinoma, HepG2 human hepatocellular carcinoma, SH-SY5Y human
neuroblastoma, C6 rat glioma, U937 human histiocytic leukemia, and
A549 human lung carcinoma.
*** ''–'' – No effect.
778

Chemistry of Heterocyclic Compounds 2020, 56(6), 776–785
Table 3. Matrix metalloproteinases and in vitro angiogenesis inhibition (%) caused by 8-ethynylxanthines 3, 5, 11, 14, 18
Compound
MMP-1 MMP-2 MMP-3 MMP-7 MMP-8 MMP-9 MMP-10 MMP-12 MMP-13 MMP-14 In vitro inhibition,** %
NNGH* (1.3 μM)
100
8.0
1.3
0
100
0
100
7.2
2.0
0
7.2
2.3
2.5
1.1
8.5
0
100
3.2
0
100
0
100
0
100
3.2
4.6
3.5
0
100
0
100
9.3
Caffeine
Proxyfeine
0
0.5
0
0
0
7.3
Pentoxifylline
0
0
0
0.6
5.8
0
8.4
Temodar
6.5
0
0
3.8
1.1
0
6.5
9.8
31.0
10.4
3.9
4.7
23.1
3.8
12.0
18.9
8.3
2.3
3.1
6.6
53.3
12.7
0
13.4
20.9
48.1
14.6
31.5
0
47.4
23.7
51.8
39.8
0
2.4
0
–40 11
–74 17
–62 18
3
0
0
0
5
5.9
0
1.1
0
5.6
4.2
1.7
24.0
0
0
11
14
18
2.0
0
38
8
0
20.2
1.8
3.5
100 10
* N-Isobutyl-N-(4-methoxyphenylsulfonyl)glycyl hydroxamic acid.
** In vitro inhibition of HUVEC tube formation.
According to our results (Table 4), the reference
compounds Proxyfeine, pentoxifylline, and Temodar had a
weak effect on Ca2+ channel activity in cardiomyoblast
cells. Caffeine and Temodar exhibit an absence of
antagonist properties in neuroblastoma and smooth muscle
cells. By contrast, Proxyfeine is an antagonist in SH-SY5Y
and A7R5 cells (IC50 0.67 and 0.52 mM, respectively).
Among the caffeine analogs, only caffeine derivative 5
does not block Ca2+ channel activity in the three studied
cell lines. Compound 3 is a selective antagonist in
neuroblastoma cells, whereas compounds 7 and 11 are
selective antagonists in smooth muscle cells. Apparently,
these derivatives predominantly induce depolarization of
Figure 2. Inhibition of HUVEC capillary tube formation. Left –
control, center – treated with compound 14, right – treated with
compound 18.
Ca2+ channel agonist–antagonist modulation
Many signaling processes involved in cancer cell
proliferation and movability depend on Ca2+ signaling. An
important consequence of intracellular signaling is the
increase in intracellular Ca2+ concentration ([Ca2+]i), which
is well known to be a critical signal for gene expression,
motility, differentiation, and survival. Cancer cell
migration depends mainly on actin polymerization and the
intracellular organization of various cytoskeletal proteins,
which are influenced by a variety of actin-binding proteins.
Actin-binding protein activity is regulated by second
messengers such as phosphoinositides and Ca2+.17–19
A change in the concentration of intracellular free Ca2+
([Ca2+]i) triggers a diverse range of cellular processes, such
as muscle contraction, fertilization, secretion, and neuro-
transmitter release.20 As a well-known activator of ryanodine
receptor (RyR), caffeine stimulates the release of Ca2+ from
both the skeletal and cardiac muscle sarco/endoplasmic
reticulum (SR/ER).21
Table 4. Ca2+ channel antagonist effect caused
by 8-ethynylxanthines 3–18 (IC50, mM)*
Compound
Caffeine
H9C2
SH-SY5Y
A7R5
Not tested
>2
–**
0.67 0.02
>2
–
Proxyfeine
0.52 0.03
>2
Pentoxifylline
1.5 0.1
1.5 0.1
–
Temodar
–
–
0.35 0.02
0.057 0.003
–
>2
3
4
0.74 0.03
–
0.25 0.02
–
5
0.56 0.03
–
1.0 0.2
–
>2
6
0.40 0.03
0.41 0.02
0.52 0.02
0.38 0.03
1.0 0.2
0.6 0.01
0.52 0.03
0.13 0.02
0.47 0.01
0.28 0.01
0.29 0.02
7
0.43 0.01
0.58 0.03
0.36 0.01
–
0.52 0.01
0.60 0.01
0.18 0.01
–
8
9
Ca2+ channel blocking activity
10
11
12
13
14
16
17
18
The Ca2+ channel blocking activity induced by the
8-ethynylxanthines was evaluated in rat cardiomyoblast
(H9C2), human neuroblastoma (SH-SY5Y), and smooth
muscle (A7R5) cell lines. Caffeine was used as an agonist
in a cardiomyocyte model for RyR induction of Ca2+
release. Next, KCl was used for the intracellular Ca2+
mobilization through the voltage-gated L-type Ca2+ channel
in A7R5 smooth muscle cells. Carbachol was used in
neuroblastoma SH-SY5Y cells to obtain a carbachol-/IP3-
sensitive Ca2+ store.22
0.89 0.05
–
1.0 0.02
0.57 0.02
0.19 0.02
0.87 0.02
0.35 0.02
1.0 0.2
–
0.30 0.01
0.075 0.005
0.67 0.03
* Data are presented as a mean SD.
** ''–'' – No effect.
779

Chemistry of Heterocyclic Compounds 2020, 56(6), 776–785
the vascular smooth muscle cell membrane, followed by
the opening of the L-type Ca2+ channels; however, they do not
affect receptor-operated Ca2+ channel activity in SH-SY5Y
(stimulated with carbachol) or H9C2 (stimulated with
caffeine) cells. Notably, compound 4 exhibits increased
selective antagonism in neuroblastoma cells (IC50 0.057 mM).
Pentoxifylline analogs 12, 16, and 17 are Ca2+ channel
antagonists in cardiomyocytes; furthermore, the IC50 value
of compound 17 reaches 0.075 mM in cardiomyocytes.
Compound 14 showed an increase in antagonist activity in
neuroblastoma and smooth muscle cells compared to
pentoxifylline and do not block the Ca2+ channels in
cardiomyocytes. Importantly, compounds with higher
antiproliferative activity in vitro than Proxyfeine displayed
Ca2+ channel antagonist properties. The Ca2+ channel
antagonist activity of known antagonist of Ca2+ channels
amlodipine (2-[(2-aminoethoxy)methyl]-4-(2-chlorophenyl)-
6-methyl-1,4-dihydropyridine-3,5-dicarboxylic acid 3-ethyl
5-methyl ester) in the smooth muscle A7R5 cell line
(IC50 0.14 µM)23 indicated that the studied caffeine and
pentoxifylline derivatives 3–18 did not exhibit a strong
ability to influence blood pressure.
Figure 4. 8-Ethynylxanthine-induced Ca2+ influx in SH-SY5Y
cells. Representative traces of changes in [Ca2+]i concentration in
response to application of compounds 5, 7, 11 (500 μM). The
arrow indicates drug injection time. Each point represents the
mean of four values.
intracellular Ca2+. Notably, only compound 5 is an agonist
in aortic cells (A7R5).
Ca2+ channel agonist activity
Compounds that are not Ca2+ channel antagonists show
agonist properties. According to our experimental data, six
compounds induce the release of Ca2+ in the cytoplasm in a
cardiomyocyte model (Figs. 3 and 4). Aminomethylethynyl-
caffeine derivatives 5, 7, and 11 are more potent Ca2+
channel agonists than caffeine. These compounds stimulate
a biphasic Ca2+ response in an early burst (to approximately
2.4 µM [Ca2+]i) of Ca2+ release from intracellular stores,
which is followed by a sustained phase of Ca2+ influx
through the cell membrane. The initial peak is not expressed
in the case of caffeine. The effect of 8-[(1-aminocyclo-
hexyl)ethynyl]caffeine (5) is indicated by its long and slow
increase in intracellular calcium. Compounds 5, 7, and 11
showed agonist properties in neuroblastoma cells, with
compound 5 also exhibiting a prolonged slow increase in
Signaling pathway involved
in intracellular Ca2+ release
Electrically excitable cells can utilize three major Ca2+
signaling pathways, namely, Ca2+ entry through voltage-
dependent Ca2+ channels; release of intracellularly stored
Ca2+ from Ca2+ stores bearing inositol 1,4,5-trisphosphate
receptors (InsP3Rs), which are sensitive to the second
messenger inositol 1,4,5-trisphosphate (InsP3); and release
of Ca2+ from Ca2+ stores bearing ryanodine receptors
(RyRs), which are sensitive to caffeine. These pathways
have some of the best characterized tool compounds.24,25
We studied the possible involvement of the Gi/o/Gq/11
G protein-phospholipase C-InsP3 receptor pathway in the
[Ca2+]i increase in cells induced by the tested compounds.
Pretreatment of cells with 2-aminoethoxydiphenyl borate
(2-APB, an InsP3R inhibitor) at a dose of 100 µM for
15 min did not block the agonist effect of any compounds
in any of the cell lines. This result indicates that the release
of intracellularly stored Ca2+ by the tested compounds is
not associated with the InsP3 receptors.
We further tested the role of RyRs in the compound-
induced Ca2+ response in cells (Table 5). RyRs are responsible
Table 5. Role of ryanodine receptor (RyR)-induced Ca2+ response
Ruthenium red, µM
Com-
pound*
H9C2
SH-SY5Y
A7R5
100 µM 20 µM 100 µM 20 µM 100 µM 20 µM
Caffeine 97
2
6
5
1
2
2
2
80
45
30
65
53
50
65
5
5
5
9
8
2
1
Not tested Not tested Not tested Not tested
Not tested Not tested Not tested Not tested
70
71
74
77
80
64
3
5
7
11
13
14*
62
82
95
2
2
2
17
6
71
5
30
5
Figure 3. 8-Ethynylxanthine-induced Ca2+ influx in H9C2 cells.
Representative traces of changes in [Ca2+]i concentration in
response to application of compounds 5, 7, 11 and caffeine
(500 μM). The arrow indicates drug injection time. Each point
represents the mean of four values.
63 10 Not tested Not tested
57 16 Not tested Not tested
Not tested Not tested Not tested Not tested
Not tested Not tested Not tested Not tested
* Compound concentration 0.5 mM.
780

Chemistry of Heterocyclic Compounds 2020, 56(6), 776–785
for triggering various Ca2+-activated physiological processes
Convenient synthetic protocols for the preparation of
various 8-ethynyl-substituted caffeine and pentoxifylline
derivatives were successfully elaborated. Taken together,
the bioactivity data mentioned above indicates that the
ethynylxanthine pharmacophore is very sensitive to any
structural modifications. Caffeine, pentoxifylline, and even
Temodar (a widely used drug for brain cancer treatment)
exhibit no or slight cytotoxicity against all the studied cell
lines. In fact, the studied malignant cells are more sensitive
to Proxyfeine (IC50 96–283 μM). We have discovered that
8-ethynylxanthines have a strong ability to suppress cancer
cell growth in human breast adenocarcinoma, neuro-
blastoma, and especially glioma and lymphoma cell lines
with IC50 values up to 4.7 μM. These novel caffeine deriva-
tives exhibit the ability to selectively inhibit MMP-2
(gelatinase A) up to 52%. In addition, 8-[(1-aminocyclo-
hexyl)ethynyl]-3,7-dimethyl-1H-purine-2,6(3H,7H)-dione
inhibits MMP-10 activity up to 53%. According to the data
obtained from the in vitro angiogenesis inhibition studies,
8-[(1-aminocyclohexyl)ethynyl]-3,7-dimethyl-1-(5-oxohexyl)-
1H-purine-2,6(3H,7H)-dione and 1,3,7-trimethyl-8-(phenyl-
ethynyl)-1H-purine-2,6(3H,7H)-dione exhibit the ability to
suppress angiogenesis by 38% and 100%, respectively, at a
concentration of 20 μM. There is no direct correlation
between angiogenesis and MMP inhibition, indicating that
the tested compounds modulate the formation of new blood
vessels by another mechanism. The Ca2+ channel activity
induced by the novel ethynylxanthines strongly depends on
the nature of the substituents. Consequently, both agonists
and antagonists of Ca2+ channels were discovered. The
studied compounds were more active antagonists than the
reference compounds (caffeine, pentoxifylline, Temodar,
and Proxyfeine), but, notably, were thousands of times less
active than the well-known drug amlodipine in the smooth
muscle A7R5 cell line. Latter result suggested that the
studied derivatives do not have a strong influence on blood
pressure.
by mediating of Ca2+ efflux from the SR/ER. The excitation–
contraction coupling (ECC) process, which occurs in the
myocytes, is the translation of the electrical impulse into
the physical cell contraction and tissue-specific expression
of RyR1 or RyR2 isoforms in skeletal or cardiac muscles.26
Caffeine increases the activity of RyR channels, albeit not
as selectively as ryanodine, and has proved particularly
useful for the determination of the Ca2+ content in the
intracellular storage organelles, such as the SR of intact
cardiac muscle cells.27
In ruthenium red-pretreated cells, the inhibition of the
compound-induced Ca2+ increase in a dose-dependent
manner was observed. Thus, we concluded that pretreat-
ment of cells with ruthenium red, a specific inhibitor of
RyRs, blocked the compound-induced Ca2+ increase in a
dose-dependent manner in all cell lines. Our findings
suggest that similar to caffeine, the Ca2+ increase induced
by the tested compounds involves the activation of RyRs.
To identify the source of the Ca2+ influx induced by the
tested compounds, we studied the effect of known
antagonist of Ca2+ channels, namely, the classical and
structurally different L-type Ca2+ channel antagonist
amlodipine. We measured the percentage of inhibition of
Ca2+ response induction caused by the studied derivatives
in cells pretreated with amlodipine. Our results showed an
inhibitory effect only on H9C2 cardiomyocyte cells,
indicating that the L-type Ca2+ channels are responsible for
the Ca2+ influx. Moreover, the effect caused by compound
3 was not blocked, which indicates a different mechanism
of action. In A7R5 and neuroblastoma cells, amlodipine
does not have an inhibitory effect on the compound-
induced increase in [Ca2+]i, which led us to conclude that in
this case, the L-type Ca2+ channels are not involved in the
Ca2+ influx (Table 6).
Because the SH-SY5Y cell line possesses both L- and
N-type voltage-dependent Ca2+ channels, these cells were
used to test the potential activation of the N-type channels
by ethynylxanthines. According to our data, none of
compounds 3–18 can compete with ω-conotoxin (a classical
and structurally different N-type Ca2+ channel blocker). In
other words, the N-type Ca2+ channels were not involved in
the Ca2+ influx in neuroblastoma cells.
The Ca2+ channel agonist activity of newly synthesized
compounds results from their influence on receptor-
operated Ca2+ channels, thereby triggering the activation of
RyRs in the studied cell lines. Furthermore, activation of
L-channels was observed in only cardiomyoblasts.
Notably, N-channels were not activated in the nervous
cells. Finally, it can be concluded that the obtained data
provide an opportunity to further modify ethynylxanthenes
in order to find a lead compound with strong antitumor
activity for the treatment of cancer. In addition, these
results support the therapeutic potential of the discovered
compounds to treat diseases with dysregulated Ca2+channel
activity, such as essential tremor, pain, neuropathic pain,
depression, sleep disorders, certain types of cancer, and
sexual dysfunction.
Table 6. Role of L-type Ca2+ channel
Amlodipine, µM
Com-
H9C2
SH-SY5Y
A7R5
pound*
100 µM 20 µM 100 µM 20 µM 100 µM 20 µM
Caffeine 100
4
5
100
2 Not tested Not tested Not tested Not tested
100
–**
59 18 Not tested Not tested Not tested Not tested
3
5
–
–
–
–
–
–
–
–
–
60
75
5
11
13
19
47
3
8
8
8
Not tested Not tested
Not tested Not tested
7
Experimental
7
5
11
13
14
IR spectra were recorded on a Shimadzu Prestige-21
1
FTIR spectrometer in thin films. H and 13C NMR spectra
69
Not tested Not tested Not tested Not tested
Not tested Not tested Not tested Not tested
(400 and 101 MHz, respectively) were recorded on a
Bruker Avance Neo spectrometer at 298 K in CDCl3 or
DMSO-d6. TMS was used as internal standard. Mass
100
2
* Compound concentration 0.5 mM.
** ''–'' – no effect.
781

Chemistry of Heterocyclic Compounds 2020, 56(6), 776–785
spectra were recorded on an Acquity UPLC (ESI) instru-
(containing 0.1% HCl) – MeCN (5–70%) as eluent to give
ment. HRMS (ESI) were recorded on a Waters Synapt GII
Q-ToF UPLC/MS system. Elemental analyses were
performed on a Carlo Erba 1108 apparatus. Melting points
were measured on an Optimelt apparatus and were not
corrected. Thin-layer chromatography (TLC) was per-
formed using Merck Silica gel 60 F254 plates and visualized
by UV (254 nm) fluorescence. ZEOCHEM silica gel
(ZEOprep 60/35–70 microns – SI23501) was used for
column chromatography.
the title compound
Method II. Dry NMP or AcNMe2 (40 ml) was added to a
mixture of Pd(OAc)2 (100 mg, 0.446 mmol), CuI (169 mg,
0.868 mmol), and Ph3P (233 mg, 0.890 mmol). The
reaction mixture was stirred for 15 min at 40°C with
simultaneous purge with argon. Then a solution of
8-bromo-1,3,7-trimethyl-3,7-dihydro-1H-purine-2,6-dione
(1) (1.21 g, 4.46 mmol) and the appropriate alkyne
(0.65 ml, 6.69 mmol) in dry DIEA (4.0 ml) was added and
stirring was continued at 55°C for 24 h. Isolation of the
product as in method I.
Method III. Dry NMP (10 ml) was added to a mixture of
Pd(OAc)2 (100 mg, 0.446 mmol), CuI (169 mg, 0.868 mmol),
and Ph3P (233 mg, 0.890 mmol). The reaction mixture was
stirred for 15 min at 40°C with simultaneous purge with
argon. Then a solution of 8-bromo-1,3,7-trimethyl-3,7-dihydro-
1H-purine-2,6-dione (1) or 8-bromo-3,7-dimethyl-1-(5-oxo-
hexyl)-3,7-dihydro-1H-purine-2,6-dione (2) (4.46 mmol)
and the appropriate alkyne (6.69 mmol) and dry DIEA
(4.0 ml) in NMP (10 ml) and PhMe (10 ml) was added and
stirring was continued at 50°C for 24 h. Isolation of the
product as in method I.
Method IV. A vial was charged with Pd(PPh3)4 (346 mg,
0.3 mmol), PdCl2 (51 mg, 0.3 mmol), Ph3P (157 mg,
0.6 mmol), CuI (58 mg, 0.3 mmol), 8-bromo-1,3,7-
trimethyl-3,7-dihydro-1H-purine-2,6-dione (1) or 8-bromo-
3,7-dimethyl-1-(5-oxohexyl)-3,7-dihydro-1H-purine-2,6-
dione (2) (15.0 mmol), the appropriate alkyne (21.0 mmol),
DIEA (5.0 ml), and EtOAc (70 ml). The reaction mixture
was stirred for 15 min at 40°C with simultaneous purge
with argon. Then reaction mixture was stirred for
additional 2–4 h. After cooling to room temperature, the
reaction mixture was filtered through a silica gel pad and
washed with EtOAc (200 ml). Then solvent was evaporated
under reduced pressure. The crude product was purified by
flash chromatography on silica gel using mixture of H2O
(containing 0.1% HCl) – MeCN (5–70%).
All reagents and solvents were purchased from
commercial suppliers and used without further purification,
unless otherwise stated.
8-Bromo-1,3,7-trimethyl-3,7-dihydro-1H-purine-2,6-
dione (1).15 Caffeine (10.0 g, 0.05 mol) and NBS (17.6 g,
0.10 mol) were added to freshly distilled CH2Cl2 (150 ml)
charged into a round-bottom flask. When the solids dissolved
in the solvent, H2O (50 ml) was added and the reaction
mixture was stirred for 5 days. Then cold 2 M aq NaOH
(30 ml) was added and the mixture was shaken till decolo-
rization. The organic layer was separated, washed with
H2O (2×200 ml), dried over Na2SO4, filtered, and
evaporated under reduced pressure to give the title
compound as colorless solid. Yield 13.5 g (99%). 1H NMR
spectrum (CDCl3), δ, ppm: 3.41 (3H, s, 1-NCH3); 3.57 (3H,
s, 3-NCH3); 3.97 (3H, s, 7-NCH3).
8-Bromo-3,7-dimethyl-1-(5-oxohexyl)-3,7-dihydro-
1H-purine-2,6-dione (2).15 Pentoxifylline (13.4 g, 0.05 mol)
and NBS (17.6 g, 0.10 mol) were added into a round-
bottom flask containing freshly distilled CH2Cl2 (150 ml).
When the solids dissolved, H2O (50 ml) was added and the
reaction mixture was stirred for 5 days. Then cold 2 M aq
NaOH (30 ml) was added and the mixture was shaken till
decolorization. The organic layer was separated, washed
with H2O (2×200 ml), dried over Na2SO4, filtered, and
evaporated under reduced pressure to give the title
compound as colorless solid. Yield 16.95 g (95%).
1H NMR spectrum (CDCl3), δ, ppm (J, Hz): 1.61–1.67 (4H,
m, C(O)CH2(CH2)2); 2.14 (3H, s, CH3C(O)); 2.49 (2H, t,
J = 7.0, CH3C(O)CH2); 3.54 (3H, s, 3-NCH3); 3.95 (3H, s,
7-NCH3); 3.99 (2H, t, J = 7.0, (CH2)3CH2N).
Synthesis of 8-ethynylxanthines 3–18 (General proce-
dure). Method I. Dry N-methylpyrrolidine (NMP) or DMF
(40 ml) was added to a mixture of PdCl2 (113 mg,
0.637 mmol), CuI (242 mg, 1.27 mmol), and Ph3P (333 mg,
1.27 mmol). The reaction mixture was stirred at 40°C for
15 min with simultaneous purge with argon. Then a
solution of 8-bromo-1,3,7-trimethyl-3,7-dihydro-1H-purine-
2,6-dione (1) (1.73 g, 6.37 mmol) and the appropriate
alkyne (9.56 mmol) in dry DIEA (3.6 ml) was added and
stirring was continued at 50°C for 24 h. After cooling to
room temperature, the reaction mixture was poured into
EtOAc (300 ml), washed with brine (80 ml) containing
aq NH3 (0.5 ml) and stirred at room temperature for the
additional 30 min. The aqueous phase was separated, and
the organic phase was washed with brine (3×80 ml), dried
over anhydrous Na2SO4, and concentrated under reduced
pressure. The crude product was purified by flash
chromatography on silica gel using a mixture of H2O
8-(3-Hydroxy-3-methylbut-1-yn-1-yl)-1,3,7-trimethyl-
3,7-dihydro-1H-purine-2,6-dione (3).28 Yield 0.73 g (42%,
method I), 0.38 g (31%, method II), 0.88 g (72%, method
III), 2.26 g (55%, method IV), colorless solid, mp 180–
1
182°C. H NMR spectrum (CDCl3), δ, ppm: 1.63 (6H, s,
2CH3); 3.36 (3H, s, 1-NCH3); 3.51 (3H, s, 3-NCH3); 3.52
(1H, br. s, OH); 3.90 (3H, s, 7-NCH3). 13C NMR spectrum
(CDCl3), δ, ppm: 28.0; 29.8; 30.8; 32.9; 65.1; 69.6; 102.5;
107.5; 135.0; 147.4; 151.5; 154.6.
8-[(1-Hydroxycyclohexyl)ethynyl]-1,3,7-trimethyl-3,7-
dihydro-1H-purine-2,6-dione (4).28 Yield 0.19 g (11%,
method II), 0.85 g (49%, method III), 0.94 g (54%, method
1
IV), colorless solid, mp 194–196°C. H NMR spectrum
(CDCl3), δ, ppm: 1.24–1.33 (1H, m, CH2 cyclohexane); 1.48–
1.57 (3H, m, CH2 cyclohexane); 1.66–1.75 (4H, m, CH2
cyclohexane); 1.94–2.02 (2H, m, CH2 cyclohexane); 3.30
(1H, br. s, OH); 3.34 (3H, s, 1-NCH3); 3.49 (3H, s, 3-NCH3);
3.89 (3H, s, 7-NCH3). 13C NMR spectrum (CDCl3), δ, ppm:
22.9; 24.8; 27.9; 29.7; 32.9; 39.2; 68.6; 71.5; 102.0; 107.4;
135.1; 147.3; 151.4; 154.5. Mass spectrum, m/z: 317 [M+H]+.
Found, m/z: 317.1621 [M+H]+. C16H21N4O3. Calculated, m/z:
782

Chemistry of Heterocyclic Compounds 2020, 56(6), 776–785
317.1535. Found: C 60.71; H 6.35; N 17.30. C16H20N4O3.
28.0; 29.8; 48.3; 53.4; 73.2; 94.2; 107.6; 135.5; 147.6;
151.5; 154.8. Mass spectrum, m/z: 316 [M+H]+. Found, m/z:
316.1779 [M+H]+. C16H22N5O2. Calculated, m/z: 316.1695.
8-[3-(Azepan-1-yl)prop-1-yn-1-yl]-1,3,7-trimethyl-3,7-
dihydro-1H-purine-2,6-dione hydrate (10). Yield 0.61 g
(34%, method III), brownish solid, mp 99–101°C. 1H NMR
spectrum (CDCl3), δ, ppm (J, Hz): 1.61–1.64 (4H, m,
CH2CH2CH2CH2); 1.70–1.76 (4H, m, CH2CH2CH2CH2);
2.81 (4H, t, J = 6.7, 2NCH2); 3.39 (3H, s, 1-NCH3); 3.55
(3H, s, 3-NCH3); 3.74 (2H, s, CH2N); 4.01 (3H, s, 7-NCH3).
13C NMR spectrum (CDCl3), δ, ppm: 26.6; 27.9; 28.0;
29.7; 33.1; 48.7; 55.3; 72.9; 93.2; 107.6; 135.4; 147.6;
151.5; 154.8. Mass spectrum, m/z: 330 [M+H]+. Found, m/z:
330.1933 [M+H]+. C17H24N5O2. Calculated, m/z: 330.1852.
Found, %: C 60.35; H 6.95; N 20.31. C17H23N5O2·0.5H2O.
Calculated, %: C 60.64; H 7.15; N 20.69.
Calculated, %: C 60.75; H 6.37; N 17.71.
8-[(1-Aminocyclohexyl)ethynyl]-1,3,7-trimethyl-3,7-
dihydro-1H-purine-2,6-dione (5). Yield 0.12 g (7%,
method I), 0.27g (16%, method II), 0.74 g (43%, method
III), 0.83g (48%, method IV), colorless solid, mp 187–189°C.
IR spectrum, ν, cm–1: 1042, 1224, 1429, 1544, 1662, 1704,
1
2228, 2856, 2933. H NMR spectrum (CDCl3), δ, ppm:
1.20–1.29 (1H, m, CH2 cyclohexane); 1.48–1.67 (5H, m,
CH2 cyclohexane); 1.72–1.77 (2H, m, CH2 cyclohexane);
1.81 (2H, br. s, NH2); 1.93–1.97 (2H, m, CH2
cyclohexane); 3.39 (3H, s, 1-NCH3); 3.55 (3H, s, 3-NCH3);
3.98 (3H, s, 7-NCH3). 13C NMR spectrum (CDCl3), δ, ppm:
23.2; 25.1; 27.9; 29.7; 33.0; 39.6; 50.4; 70.9; 104.5; 107.6;
135.8; 147.6; 151.5; 154.8. Found, m/z: 316.1779 [M+H]+.
C16H22N5O2. Calculated, m/z: 316.1768.
8-[3-(Dimethylamino)prop-1-yn-1-yl]-1,3,7-trimethyl-
3,7-dihydro-1H-purine-2,6-dione hydrochloride (6). Yield
0.06 g (4%, method II), 0.71 g (47%, method III), 0.87 g
(58%, method IV), colorless solid, mp 197–199°C
1,3,7-Trimethyl-8-[3-(morpholin-4-yl)prop-1-yn-1-yl]-
3,7-dihydro-1H-purine-2,6-dione (11). Yield 0.66 g (38%,
method III), colorless solid, mp 88–190°C. 1H NMR
spectrum (CDCl3), δ, ppm (J, Hz): 2.64 (4H, t, J = 4.8,
2NCH2); 3.40 (3H, s, 1-NCH3); 3.56 (3H, s, 3-NCH3); 3.64
(2H, s, CH2N); 3.76 (4H, t, J = 4.8, 2CH2O); 4.01 (3H, s,
7-NCH3). 13C NMR spectrum (CDCl3), δ, ppm: 28.0; 29.7;
33.2; 47.8; 52.3; 66.7; 73.7; 93.1; 107.7; 135.2; 147.6;
151.5; 154.8. Mass spectrum, m/z: 318 [M+H]+.
1
(decomp.). H NMR spectrum (DMSO-d6), δ, ppm: 2.88
(6H, s, N(CH3)2); 3.21 (3H, s, 1-NCH3); 3.38 (3H, s, 3-NCH3);
3.96 (3H, s, 7-NCH3); 4.50 (2H, s, CH2N). 13C NMR
spectrum (DMSO-d6), δ, ppm: 27.6; 29.3; 33.0; 41.6; 45.7;
76.4; 87.1; 107.6; 132.9; 146.8; 150.7; 154.0. Mass spectrum,
m/z: 277 [M+H]+. Found, m/z: 276.1463 [M+H]+. C13H18N5O2.
Calculated, m/z: 276.1382. Found, %: C 47.37; H 5.64;
N 20.96. C13H17N5O2·1.5HCl. Calculated, %: C 47.32;
H 5.65; N 21.22.
8-(3-Hydroxy-3-methylbut-1-yn-1-yl)-3,7-dimethyl-
1-(5-oxohexyl)-3,7-dihydro-1H-purine-2,6-dione
(12).
Yield 0.66 g (44%, method III), 0.71 g (47%, method IV),
brownish solid, mp 96–98°C. IR spectrum, ν, cm–1: 761,
969, 1171, 1217, 1428, 1545, 1601, 1661, 1703, 2238, 2946,
8-{3-[Bis(2-methoxyethyl)amino]prop-1-yn-1-yl}-1,3,7-
trimethyl-3,7-dihydro-1H-purine-2,6-dione (7). Yield
0.2 g (10%, method II), 0.59 g (30%, method III), 0.65 g
1
3416. H NMR spectrum (CDCl3), δ, ppm (J, Hz): 1.62–
1.66 (4H, m, CH2(CH2)2CH2N); 1.65 (6H, s, 2CH3); 2.12
(3H, s, CH3CO); 2.48 (2H, t, J = 7.1, CH3C(O)CH2); 2.72 (1H,
br. s, OH); 3.53 (3H, s, 3-NCH3); 3.96 (3H, s, 7-NCH3);
3.97 (2H, t, J = 7.1, (CH2)3CH2N). 13C NMR spectrum
(CDCl3), δ, ppm: 20.9; 27.3; 29.7; 29.9; 30.9; 33.0; 40.9;
43.1; 65.4; 70.0; 102.2; 107.7; 135.1; 147.6; 151.2; 154.6;
208.7. Mass spectrum, m/z: 361 [M+H]+. Found, m/z:
361.1884 [M+H]+. C18H25N4O4. Calculated, m/z: 361.1798.
8-[(1-Hydroxycyclohexyl)ethynyl]-3,7-dimethyl-1-(5-oxo-
hexyl)-3,7-dihydro-1H-purine-2,6-dione (13). Yield 0.70 g
(42%, method III), colorless solid, mp 126–128°C.
IR spectrum, ν, cm–1: 761, 969, 1076, 1170, 1222, 1339,
1428, 1486, 1546, 1601, 1661, 1708, 2231, 2858, 2936,
1
(33%, method IV), brownish foam. H NMR spectrum
(CDCl3), δ, ppm (J, Hz): 2.79 (4H, t, J = 5.5,
2NCH2CH2O); 3.31 (6H, s, 2OCH3); 3.34 (3H, s, 1-NCH3);
3.48 (4H, t, J = 5.5, 2NCH2CH2O); 3.51 (3H, s, 3-NCH3);
3.82 (2H, s, CH2N); 3.96 (3H, s, 7-NCH3). 13C NMR
spectrum (CDCl3), δ, ppm: 27.8; 29.6; 33.0; 44.1; 53.5;
58.7; 70.9; 73.2; 94.1; 107.5; 135.4; 147.5; 151.4; 154.7.
Mass spectrum, m/z: 364 [M+H]+.
1,3,7-Trimethyl-8-[3-(pyrrolidin-1-yl)prop-1-yn-1-yl]-
3,7-dihydro-1H-purine-2,6-dione (8). Yield 0.13 g (8%,
method I), 0.86 g (52%, method III), 0.94 g (57%, method
1
IV), colorless solid, mp 149–150°C. H NMR spectrum
1
(CDCl3), δ, ppm: 1.83–1.86 (4H, m, 2CH2CH2N); 2.68–
2.72 (4H, m, 2CH2CH2N); 3.39 (3H, s, 1-NCH3); 3.56 (3H,
s, 3-NCH3); 3.75 (2H, s, CH2N); 4.00 (3H, s, 7-NCH3).
13C NMR spectrum (CDCl3), δ, ppm: 23.8; 27.9; 29.7;
33.1; 43.6; 52.7; 72.6; 94.5; 107.6; 135.4; 147.6; 151.5;
154.8. Mass spectrum, m/z: 302 [M+H]+. Found, m/z:
302.1623 [M+H]+. C15H20N5O2. Calculated, m/z: 302.1539.
1,3,7-Trimethyl-8-[3-(piperidin-1-yl)prop-1-yn-1-yl]-
3,7-dihydro-1H-purine-2,6-dione (9). Yield 0.60 g (35%,
method III), 0.52 g (30%, method IV), colorless solid,
mp 137–138°C. 1H NMR spectrum (CDCl3), δ, ppm: 1.40–1.48
(2H, m, CH2CH2CH2); 1.60–1.67 (4H, m, CH2CH2CH2);
2.54–2.60 (4H, m, 2NCH2CH2); 3.39 (3H, s, 1-NCH3); 3.55
(3H, s, 3-NCH3); 3.60 (2H, s, CH2N); 4.00 (3H, s, 7-NCH3).
13C NMR spectrum (CDCl3), δ, ppm: 23.6; 25.8; 27.9;
3431. H NMR spectrum (CDCl3), δ, ppm (J, Hz): 1.27–
1.37 (1H, m, CH2 cyclohexane); 1.54–1.80 (11H, m, CH2
cyclohexane, CH2(CH2)2CH2N); 2.02–2.08 (2H, m, CH2
cyclohexane); 2.13 (3H, s, CH3CO); 2.48 (2H, t, J = 7.1,
CH3C(O)CH2); 3.54 (3H, s, 3-NCH3); 3.97 (3H, s, 7-NCH3);
3.99 (2H, t, J = 7.1, (CH2)3CH2N). 13C NMR spectrum
(CDCl3), δ, ppm: 20.9; 23.0; 24.9; 27.4; 29.7; 29.9; 33.1;
39.3; 40.9; 43.1; 69.0; 72.0; 101.5; 107.8; 135.3; 147.6;
151.2; 154.6; 208.6. Mass spectrum, m/z: 401 [M+H]+.
Found, m/z: 401.2196 [M+H]+. C21H29N4O4. Calculated, m/z:
401.2111.
8-[(1-Aminocyclohexyl)ethynyl]-3,7-dimethyl-1-(5-oxo-
hexyl)-3,7-dihydro-1H-purine-2,6-dione (14). Yield 0.75 g
1
(45%, method III), colorless solid, mp > 200°C. H NMR
spectrum (CDCl3), δ, ppm: 1.24–1.34 (1H, m, CH2 cyclohexane);
783

Chemistry of Heterocyclic Compounds 2020, 56(6), 776–785
1.61–1.72 (7H, m, CH2 cyclohexane); 1.81–1.84 (2H, m,
7-NCH3); 7.39–7.46 (3H, m, H Ph); 7.61–7.63 (2H, m,
H Ph).
In vitro cytotoxicity assay. Monolayer tumor cell lines:
CH2(CH2)2CH2N); 1.95–2.01 (2H, m, CH2(CH2)2CH2N);
2.13 (3H, s, CH3CO); 2.34–2.38 (2H, m, CH2 cyclo-
hexane); 2.48 (2H, t, J = 6.8, CH3C(O)CH2); 3.49 (3H, s,
3-NCH3); 3.95 (2H, t, J = 6.8, (CH2)3CH2N); 4.01 (3H, s,
7-NCH3); 9.29 (2H, br. s, NH2). 13C NMR spectrum
(CDCl3), δ, ppm: 20.9; 22.5; 24.1; 27.3; 25.1; 29.7; 29.9;
33.5; 35.9; 41.0; 43.1; 53.6; 75.7; 93.9; 108.0; 133.8; 147.4;
150.9; 154.4; 208.6. Mass spectrum, m/z: 400 [M+H]+.
Found, m/z: 401.2198 [M+2]+. C21H31N5O3. Calculated, m/z:
401.2270.
MDA-MB-435s (human melanoma), H9C2 (rat embrio
cardiomyoblasts), MCF-7 (human breast adenocarcinoma,
estrogen-positive), HepG2 (human hepatocellular carcinoma),
SH-SY5Y (human neuroblastoma), C6 (rat glioma), U937
(human histiocytic leukemia), and A549 (human lung
carcinoma). The cells were seeded in 96-well plates in
DMEM and cultivated for 72 h while exposed to different
concentrations of the compounds. Cell viability was
measured via MTT assay.30 Briefly, after incubation with
the compounds, the culture medium was removed, and
fresh medium containing 0.2 mg/ml MTT was added. After
incubation (3 h, 37°C, 5% CO2), the MTT-containing medium
was removed, and 200 μl of DMSO was immediately
added to each sample. The samples were assessed at
540 nm on a Tecan Infinite1000 microplate reader. The
half-maximal inhibitory concentration (IC50) of each com-
pound was calculated using GraphPad Prism 3.0.
8-{3-[Bis(2-methoxyethyl)amino]prop-1-yn-1-yl}-3,7-
dimethyl-1-(5-oxohexyl)-3,7-dihydro-1H-purine-2,6-dione
(15). Yield 0.65 g (35%, method III), brownish foam.
IR spectrum, ν, cm–1: 1118, 1438, 1547, 1661, 1702, 2950,
1
3415. H NMR spectrum (CDCl3), δ, ppm (J, Hz): 1.60–
1.65 (4H, m, CH2(CH2)2CH2N); 2.11 (3H, s, CH3CO); 2.47
(2H, t, J = 6.9, CH3C(O)CH2); 3.38 (9H, s, 3-NCH3,
2OCH3); 3.43–3.56 (4H, m, 2NCH2CH2O); 3.56 (3H, s,
7-NCH3); 3.92–3.97 (4H, m, 2NCH2CH2O); 3.97 (2H, t,
J = 6.9, (CH2)3CH2N); 4.11 (2H, s, CH2N). 13C NMR
spectrum (CDCl3), δ, ppm: 20.9; 27.4; 29.8; 31.9; 40.7;
43.1; 54.2; 58.8; 63.0; 66.5; 71.0; 76.9; 107.2; 110.3; 143.1;
147.3; 148.1; 151.3; 155.1; 208.6. Mass spectrum, m/z: 448
[M+H]+. Found, m/z: 448.2552 [M+H]+. C22H34N5O5.
Calculated, m/z: 448.2482.
MMP assay. Inhibitors of matrix metalloproteinase
enzymes were detected with the use of MMP Inhibitor
Fluorimetric Profiling kit (Biomol, USA) accordingly to
manufacturer's instructions. MMP activity assays were
performed in 96-well plates using the recombinant human
MMP-1–10, MMP-12–14 catalytic domains and
OmniMMPTM fluorogenic substrate Mca-Pro-Leu-Gly-Leu-
Dpa-Ala-Arg-NH2. The test compounds (20 µM) were
dissolved in DMSO. NNGH (N-isobutyl-N-(4-methoxy-
phenylsulfonyl)glycyl hydroxamic acid) was used as a
prototypic control inhibitor. The rate of substrate hydro-
lysis was determined by fluorescence intensity measuring
for 10 min at 37°C temperature using a fluorescence plate
reader (Tecan infinite M1000, Austria) with excitation at
328 nm and emission at 420 nm. Data analysis was
performed using program GraphPad Prism 3.0.
3,7-Dimethyl-1-(5-oxohexyl)-8-[3-(pyrrolidin-1-yl)prop-
1-yn-1-yl]-3,7-dihydro-1H-purine-2,6-dione (16). Yield
0.66 g (41%, method III), brownish foam. IR spectrum,
ν, cm–1: 769, 1223, 1436, 1539, 1653, 1700, 2803, 2961.
1H NMR spectrum (CDCl3), δ, ppm (J, Hz): 1.61–1.70 (4H,
m, CH2(CH2)2CH2N); 1.82–1.86 (4H, m, CH2 pyrrolidine);
2.13 (3H, s, CH3CO); 2.49 (2H, t, J = 7.0, CH3C(O)CH2);
2.64–2.68 (4H, m, CH2 pyrrolidine); 3.48 (2H, s, CH2N);
3.58 (3H, s, 3-NCH3); 3.93 (3H, s, 7-NCH3); 4.01 (2H, t,
J = 7.0, (CH2)3CH2N). 13C NMR spectrum (CDCl3),
δ, ppm: 20.8; 23.9; 27.3; 29.6; 29.9; 33.6; 40.9; 43.0; 43.3;
52.7; 77.6; 85.1; 108.4; 133.0; 147.4; 151.0; 154.5; 208.5.
Mass spectrum, m/z: 386 [M+H]+. Found, m/z: 386.2194
[M+H]+. C20H28N5O3. Calculated, m/z: 386.2114.
In vitro angiogenesis assay. An in vitro Matrigel assay
was performed according to the literature methodology.31
HUVEC cells were grown in Vascular Cell Basal medium
with 2% fetal bovine serum, bovine brain extract, hydro-
cortisone, human endothelial growth factor, and
gentamicin/amphotericin B (ATCC) in incubator with 5%
CO2 and at 37°C. Matrigel was dispensed into 96-well
plates (50 µl/well). Plates were incubated at 37°C for 1 h to
allow polymerization of matrigel. HUVEC cultures were
added into each matrigel-coated well at a final concen-
tration of 10 000 cells/well. After 8 h incubation at 37°C
cells were labeled with calcein-AM (Molecular Probes) and
photographed at 4-fold magnification using an inverted
fluorescence microscope (Nicon) calculating the length of
capillary tubes.
3,7-Dimethyl-1-(5-oxohexyl)-8-[3-(piperidin-1-yl)prop-
1-yn-1-yl]-3,7-dihydro-1H-purine-2,6-dione (17). Yield
0.65 g (39%, method III), colorless solid, mp 182–184°C.
1H NMR spectrum (CDCl3), δ, ppm (J, Hz): 1.34–1.48 (1H,
m, CH2 piperidine); 1.58–1.66 (4H, m, CH2(CH2)2CH2N);
1.90–1.93 (3H, m, CH2 piperidine); 2.11 (3H, s, CH3CO);
2.21–2.29 (2H, m, CH2 piperidine); 2.47 (2H, t, J = 7.0,
CH3C(O)CH2); 2.94–2.97 (2H, m, CH2 piperidine); 3.51
(3H, s, 3-NCH3); 3.60–3.63 (2H, m, CH2 piperidine); 3.97
(2H, t, J = 7.0, (CH2)3CH2N); 4.07 (3H, s, 7-NCH3); 4.22
(2H, s, CH2N). 13C NMR spectrum (CDCl3), δ, ppm: 20.8;
21.5; 22.7; 27.3; 29.6; 29.9; 33.7; 40.9; 43.0; 46.7; 52.6;
78.7; 84.6; 108.5; 133.1; 147.5; 151.0; 154.5; 208.5. Mass
spectrum, m/z: 400 [M+H]+. Found, m/z: 400.2357 [M+H]+.
C21H29N5O3. Calculated, m/z: 400.2270.
Intracellular Ca2+ measurements. Changes in intra-
cellular [Ca2+]i concentration were studied using Fluo-4
NW Calcium Assay kit (Invitrogen, Sweden) accordingly
to manufacturer's protocol. The сells A7R5, H9C2 or
SH-SY5Y were seeded into 96-well plate at 10 000 cells
per well and incubated for 72 h. Intracellular free Ca2+
concentration [Ca2+]i was measured in confluent
monolayers of cells with the Ca2+-sensitive fluorescent
1,3,7-Trimethyl-8-(phenylethynyl)-3,7-dihydro-1H-
purine-2,6-dione (18).29 Yield 0.58 g (36%, method III),
1
colorless solid. H NMR spectrum (CDCl3), δ, ppm: 3.43
(3H, s, 1-NCH3); 3.61 (3H, s, 3-NCH3); 4.10 (3H, s,
784

Chemistry of Heterocyclic Compounds 2020, 56(6), 776–785
indicator Fluo 4 NW. The model for investigation of Ca2+
channel blocking activity caused by 8-ethynylxanthine
derivatives was based on the analysis of the effect of
caffeine on intracellular Ca2+ mobilization in H9C2 cells;
accordingly, carbachol was used in case of SH-SY5Y cells
and KCl effect on intracellular Ca2+ mobilization in A7R5
cells. The cells were preincubated with tested compounds
at concentrations from 0.01 to 2 mM in the dark for 15 min.
Application of caffeine (1.2 mM) for H9C2, carbachol
(30 µM) for SH-SY5Y and KCl (50 mM) for A7R5 to Fluo-
4NW loaded cells stimulated intracellular Ca2+ mobiliza-
tion. In case of Ca2+ channel agonist activity studies Fura-
4NW loaded cells were stimulated by the addition of tested
compounds. Changes in [Ca2+]i were measured with the
fluorescence emitted at 516 nm due to alternate excitation
7. Wilson, K. M.; Kasperzyk, J. L.; Rider, J. R.; Kenfield, S.;
van Dam, R. M.; Stampfer, M. J.; Giovannucci, E.; Mucci, L. A.
J. Nat. Cancer Inst. 2010, 103, 876.
8. Geybels, M. S.; Neuhouser, M. L.; Wright, J. L.; Stott-Miller, M.;
Stanford, J. L. Cancer, Causes Control 2013, 24, 1947.
9. Bravi, F.; Bosetti, C.; Tavani, A.; Bagnardi, V.; Gallus, S.;
Negri, E.; Franceschi, S.; La Vecchia, C. Hepatology 2007,
46, 430.
10. Yamamoto, N.; Tsuchiya, H. J. Caffeine Res. 2011, 1, 163.
11. Hayashi, M.; Tsuchiya, H.; Yamamoto, N.; Karita, M.; Shirai, T.;
Nishida, H.; Takeuchi, A.; Tomita, K. Anticancer Res. 2005,
25, 2399.
12. Ding, R.; Shi, J.; Pabon, K.; Scotto, K. W. Mol. Pharmacol.
2012, 81, 328.
13. Vartanyan, L. P.; Kolesova, M. B.; Gornaeva, G. F.;
Pustovalov, Yu. I. Psychopharmacol. Biol. Narcol. [In
Russian] 2005, 5, 1093.
at 494 nm using
a fluorescence spectrophotometer
14. Rybár, A.; Pfleiderer, W. Collect. Czeh. Chem. Commun.
1987, 52, 2730.
(TECAN, Infinite M1000). Intracellular Ca2+ concentration
([Ca2+]i) was calculated from a standard curve with known
amounts of free Ca2+ using the standard Ca2+-EGTA
buffering system (Calcium Calibration Buffer kit, cat.
No. C-3008, Life Technologies). IC50 values for the tested
compounds were calculated using the GraphPad Prism 3.0
software. All data are representative as the mean of four
independent experiments.
15. Soltani Rad, M. N.; Behrouz, S.; Nekoei, A.-R. Synlett 2012, 1191.
16. Arsenyan, P.; Rubina, K.; Vasiljeva, J.; Belyakov, S.
Tetrahedron Lett. 2013, 54, 6524.
17. Schafer, B.; Gschwind, A.; Ullrich, A. Oncogene 2004, 23, 991.
18. Yamazaki, D.; Kurisu, S.; Takenawa, T. Cancer Sci. 2005, 96,
379.
19. Ridley, A. J.; Schwartz, M. A.; Burridge, K.; Firtel, R. A.;
Ginsberg, M. H.; Borisy, G.; Parsons, J. T.; Horwitz, A. R.
Science 2003, 302, 1704.
20. Berridge, M. J.; Bootman, M. D.; Lipp, P. Nature 1998, 395, 645.
21. Rousseau, E.; Ladine, J.; Liu, Q.-Y.; Meissner, G. Arch.
Biochem. Biophys. 1988, 267, 75.
Detection the role of InsP3R in the Ca2+ response was
investigated by the InsP3Rs inhibitor 2-APB. The role of
ryanodine receptor in Ca2+ response was investigated by
ruthenium red as a potent RyR inhibitor. For detection of
the role of L-type calcium channel in Ca2+ response,
amlodipine was used as L-type Ca2+ channel blocker;
besides, ω-conotoxin was used as N-type Ca2+ channel
blocker. Cells were pretreated for 15 min with the
corresponding inhibitor and then stimulated intracellular
Ca2+ mobilization with tested compound.
22. Garavito-Aguilar, Z. V.; Recio-Pinto, E.; Corrales, A. V.;
Zhang, J.; Blanck, T. J. J.; Xu, F. Brain Res. 2004, 1011, 177.
23. Rucins, M.; Gosteva, M.; Domracheva, I.; Kanepe-Lapsa, I.;
Belyakov, S.; Plotniece, M.; Pajuste, K.; Cekavicus, B.;
Jekabsone, M.; Sobolev, A.; Shestakova, I.; Plotniece, А.
Chem. Heterocycl. Compd. 2015, 50, 1432. [Khim. Geterotsikl.
Soedin. 2014, 1557.]
24. Berridge, M. J. Neuron 1998, 21, 13.
References
25. Verkhratsky, A. Physiol. Rev. 2005, 85, 201.
26. Fleischer, S.; Inui, M. Prog. Clin. Biol. Res. 1988, 273, 435.
27. Thomas, N. L.; Williams, A. J. Wiley Interdiscip. Rev.:
Membr. Transp. Signaling 2012, 1, 383.
1. Jemal, A.; Bray, F.; Center, M. M.; Ferlay, J.; Ward, E.;
Forman, D. Ca–Cancer J. Clin. 2011, 61, 69.
2. Jemal, A.; Siegel, R.; Xu, J.; Ward, E. Ca–Cancer J. Clin.
2010, 60, 277.
3. Siegel, R. L.; Miller, K. D.; Jemal, A. Ca–Cancer J. Clin.
2017, 67, 7.
4. Siegal, T. J. Neuro-Oncology 2013, 15, 656.
5. Lampson, L. A. Drug Discovery Today 2009, 14, 185.
6. Huse, J. T.; Holland, E. C. Nat. Rev. Cancer 2010, 10, 319.
28. Arsenyan, P.; Vasiljeva, J.; Ivanova, A.; Belyakov, S.
Mendeleev Commun. 2019, 29, 96.
29. Schwarz Pharma AG. EP Patent EP1939197 A1, 2008.
30. Mosmann, T. J. Immunol. Methods 1983, 65, 55.
31. Ponce, M. L. In Angiogenesis Protocols; Murray, J. C., Ed.;
Humana Press, 2001, p. 205.
785