Synthesis, structure elucidation, determination of the lipophilicity and identification of antitumour activities in vitro of novel 3-(2-furanyl)-8-aryl- 7,8-dihydroimidazo[2,1-c][1,2,4]triazin-4(6H)-ones with a low cytotoxicity towards normal human skin f

Article abstract of DOI:10.1016/j.bmc.2011.07.027

Eleven novel 3-(2-furanyl)-8-aryl-7,8-dihydroimidazo[2,1-c][1,2,4]triazin- 4(6H)-ones (12-22) were designed and obtained from appropriate 1-aryl-2-hydrazonoimidazolidines (1-11) by condensation reaction with 2-oxo-2-furanacetic acid and subsequent cycloco

Full text of DOI:10.1016/j.bmc.2011.07.027

Bioorganic & Medicinal Chemistry 19 (2011) 5103–5116
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Synthesis, structure elucidation, determination of the lipophilicity and
identification of antitumour activities in vitro of novel 3-(2-furanyl)-8-aryl-7,8-
dihydroimidazo[2,1-c][1,2,4]triazin-4(6H)-ones with a low cytotoxicity towards
normal human skin fibroblast cells
Krzysztof Sztanke a, , Tomasz Tuzimski b, Małgorzata Sztanke a, Jolanta Rzymowska c, Kazimierz Pasternak a
⇑
a
´
Chair and Department of Medical Chemistry, Medical University, 4A Chodzki Street, 20-093 Lublin, Poland
b
c
´
Department of Physical Chemistry, Chair of Chemistry, Faculty of Pharmacy with Medical Analytics Division, Medical University of Lublin, 4A Chodzki Street, 20-093 Lublin, Poland
Department of Biology and Genetics, Medical University, 4A Chodzki Street, 20-093 Lublin, Poland
´
a r t i c l e i n f o
a b s t r a c t
Article history:
Eleven novel 3-(2-furanyl)-8-aryl-7,8-dihydroimidazo[2,1-c][1,2,4]triazin-4(6H)-ones (12–22) were
designed and obtained from appropriate 1-aryl-2-hydrazonoimidazolidines (1–11) by condensation
reaction with 2-oxo-2-furanacetic acid and subsequent cyclocondensation of intermediate chain deriv-
atives. IR, 1H NMR and 13C NMR spectra and elemental analyses confirmed the chemical structure of
all the synthesized compounds. The reversed-phase HPLC method was optimized and proved to be appli-
cable and reliable for the analysis of these unknown small molecules (12–22). These compounds were
chromatographed on octadecyl silica (ODS) stationary phase and their hydrophobic parameters
expressed as the log kw values were determined by RP-HPLC, using mixtures of methanol and water
as mobile phases with different methanol concentrations. Octane-1-sulfonic acid sodium salt (OSA-
Na) and 20% acetate buffer (pH 3.5) was added to the mobile phase (eluent containing 0.01 M/L OSA-
Na in organic modifier (MeOH)—buffered mobile phase). The high values of regression coefficients
(r >0.9841) for all the compounds investigated proved the excellent fit between experimental data
and the Snyder–Soczewin´ ski equation. Results obtained from the reversed-phase HPLC were compared
both with those theoretically calculated and with those obtained from an ALOGPS 2.1. software by the
use of nine different computational methods for estimation of log P. The predicted values of log P by use
of AB log P algorithm revealed the best correlation with the experimental log kw values for the investi-
gated solutes, since a good correlation (r = 0.7760) between these quantities was found. The majority of
novel imidazotriazinones were found to be evidently effective in vitro against human cancerous cells
Received 31 May 2011
Revised 9 July 2011
Accepted 15 July 2011
Available online 23 July 2011
Keywords:
Fused 1,2,4-triazinones
RP-HPLC
Lipophilicity
Comparison of measured log kw and
calculated log P values
Antitumour activity
Selective cytoxicity
(HeLa and T47D) in an effective concentration of 50 lg/mL. Five compounds (13, 15, 16, 18 and 22)
revealed remarkable antiproliferative activities and selective cytotoxicities for cancer cells over normal
HSF cells. Therefore these ones may be considered as a basis for the design of novel useful non-toxic (13,
15 and 16) and low toxic (18 and 22) anticancer agents.
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction
evoke to the normal rapidly dividing cells (i.e., the bone marrow
cells, the epithelial cells in the gastrointestinal tract) and by side
Cancer is still considered to be the largest cause of mortality
(after heart disease) because about 15% of all human deaths in
the developed world are evoked by this severe disease.1 In spite
of meaningful progress in cancer chemotherapy there is still a lack
of clinically useful cytotoxic drugs selective for cancerous cells.
Now then the possibility of application in cancer treatment of
known antiproliferative agents is limited by the damage they also
effects characteristic for given anticancer drugs.2 Therefore, the
development of novel synthetic anticancer agents that show a
selective cytotoxicity for cancer cells over normal cells seems to
be of great urgency. Furthermore, the examination of cytotoxicity
profile still remains especially important in the continuous search
for novel cytotoxic structures.
Our findings suggest that several polynitrogenated small mole-
cules based on the 7,8-dihydroimidazo[2,1-c][1,2,4]triazin-4(6H)-
one template (shown in Table 1) are able to selectively inhibit
the growth of tumour cells, and for this reason in the future these
⇑
Corresponding author. Tel.: +48 81 5357362; fax: +48 81 5357361.
E-mail address: krzysztof.sztanke@umlub.pl (K. Sztanke).
0968-0896/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2011.07.027

5104
K. Sztanke et al. / Bioorg. Med. Chem. 19 (2011) 5103–5116
Table 1
Previously reported3–5 thesis structures based on the 7,8-dihydroimidazo[2,1-c][1,2,4]triazin-4(6H)-one template and a brief summary of the most important assay data
Structure
Cl
A brief summary of the most important assay data
Selective for cancer cells over normal cells. Almost twofold more cytotoxic against cancer cell lines (mostly against LS180, SiHa and T47D
1
cells) than towards normal HSF cells at concentrations of 10 lg/mL (0.031 mM) and 50 lg/mL (0.156 mM) after 72 h of incubation in the 5-
8a
N
7
N
8
bromo-20-deoxyuridine (BrdU) cell proliferation assay. Has been proved to inhibit the growth of tumour cells in a concentration-dependent
N 2
manner.3
N
6
4
5
LD50 i.p. in male Albino-Swiss mice: over 1000 mg/kg.4
3
COOC2H5
O
IT-1
Cl
N
Cl
Reported to be selective for cancer cells over normal cells. Over twofold more cytotoxic against cancer cells (LS180 and SiHa cell lines) than
towards normal HSF cells at concentrations of 10 lg/mL (0.028 mM) and 50 lg/mL (0.141 mM) after 72 h of incubation. Has been shown to
evoke a dose-dependent antitumour effects in the BrdU assay.3 Evidently effective against leukaemic Jurkat cells with the IC50 value of
0.032 mM and low cytotoxic towards the primary cell line of normal human skin fibroblasts after 24 h of incubation in the assay MTT based
(the succinate dehydrogenase inhibition, SDI test)4
N
N
N
COOC2H5
O
IT-2
Reported to be effective against leukaemic Jurkat cells with the IC50 value of 0.085 mM and simultaneously low cytotoxic towards the
primary cell line of normal HSF cells after 24 h of incubation in the MTT assay.4 Completely non-toxic towards normal GMK cells at
N
N
N
concentrations of 10 lg/mL (0.037 mM) and 50 lg/mL (0.184 mM) after 72 h of incubation in a colorimetric BrdU cell proliferation ELISA
N
kit.4LD50 i.p. in male Albino-Swiss mice: over 1000 mg/kg4
CONHNH2
O
IT-3
OCH3
Has been demonstrated remarkable viability decreases (to 57% and 43% in response to treatment with IT-4) in human peripheral blood
myeloma cells (RPMI 8226) both at a concentration of 0.050 and 0.100 mM, respectively after exposure for 24 h. Has also been reported to be
almost 1.7-times more cytotoxic against RPMI 8226 myeloma cells than towards normal HSF cells at concentration of 0.100 mM in the MTT
assay.5 At a concentration of 0.025 mM was able to inhibit migration of human cervical epithelial carcinoma cells (HeLa B) (41% inhibition of
cell migration in response to treatment with IT-4 versus untreated controls) in the monolayer scratch assay.5
LD50 i.p. in male Albino-Swiss mice: over 2000 mg/kg
N
N
N
N
O
IT-4
Cl
Has been proven to evoke viability decreases (to 70% and 60% in response to treatment with IT-5) both at a concentration of 0.050 and
0.100 mM, respectively, in RPMI 8226 myeloma cells after exposure for 24 h. Has been proven to be almost 1.5-times more cytotoxic for
RPMI 8226 cancer cells than towards normal HSF cells at a concentration of 0.100 mM in the MTT assay.5 At a concentration of 0.025 mM was
able to inhibit explicitly migration of HeLa B cancerous cells (48% inhibition of cell migration in response to treatment with IT-5 versus
untreated controls) in the monolayer scratch assay. Has been suggested as a preventive agent against cancer metastasis5
N
N
N
N
O
IT-5
Cl
Selective for cancer cells over normal cells. Has been shown to evoke a dose-dependent viability decreases (to 56% and 36% in response to
treatment with IT-6) both at a concentration of 0.050 and 0.100 mM, respectively in RPMI 8226 myeloma cells after 24-h incubation period.
Completely non-toxic for the primary cell line of normal HSF cells in the MTT assay.5 At a concentration of 0.025 mM was able to significantly
inhibit migration of HeLa B cancer cells (50% inhibition of cell migration in response to treatment with IT-6 versus untreated controls) in the
monolayer scratch assay. Has been proposed as a preventive agent against cancer metastasis.5
N
N
N
N
LD50 i.p. in Albino-Swiss mice: over 2000 mg/kg
O
IT-6
Cl
Cl
Has been reported to evoke viability decreases (to 69 and 58% in response to treatment with IT-7) in human peripheral blood myeloma
N
N
(RPMI 8226) cells both at a concentration of 0.050 and 0.100 mM, respectively after exposure for 24 h in the MTT assay5
N
N
O
IT-7

K. Sztanke et al. / Bioorg. Med. Chem. 19 (2011) 5103–5116
5105
Table 1 (continued)
Structure
A brief summary of the most important assay data
Cl
Cl
N
N
Has been demonstrated remarkable viability decreases (to 43% and 41% in response to treatment with IT-8) in RPMI 8226 human myeloma
cells both at a concentration of 0.050 and 0.100 mM, respectively after 24-h incubation period in the MTT assay5
N
N
O
IT-8
CH3
N
Has been demonstrated anti-migratory effects in vitro. At a single concentration of 0.025 mM was able to inhibit migration ability of human
cervical epithelial cancer cells (HeLa B cells) (32% inhibition of cell migration in response to treatment with IT-9 versus untreated controls) in
the monolayer scratch assay. Has been suggested as a preventive agent against cancer metastasis5
N
N
N
O
IT-9
Cl
N
N
Has been proven to exhibit anti-migratory effects in vitro. At a single concentration of 0.025 mM was able to inhibit explicitly migration of
HeLa B cancerous cells (41% inhibition of cell migration in response to treatment with IT-10 versus untreated controls) in the monolayer
scratch assay. Has been suggested as a preventive agent against cancer metastasis5
N
N
O
IT-10
heterocycles might be powerful drug candidates having a benefi-
cial side-effect profile.3–5 In our recent studies the two successful
derivatives of the 7,8-dihydroimidazo[2,1-c][1,2,4]triazin-4(6H)-
one, possessing remarkable dose-dependent anticancer activities
and distinctly marked lower cytotoxicities towards normal cells,
were discovered. These potent small molecules (namely IT-1 and
IT-2), containing the ethoxycarbonyl formation at the C-3 and a
3-chlorophenyl or a 3,4-dichlorophenyl substituent attached to
the N-8 of the heterobicylic template,3 acquired considerable
interest because of their evidenced profitable cytotoxicity profile.
The 3-carbohydrazide derivative of 7,8-dihydroimidazo[2,
1-c][1,2,4]triazin-4(6H)-one (IT-3), with the unsubstituted phenyl
ring as the best choice of substitution pattern at the N-8, revealing
an antileukaemic activity was also designed in our laboratory and
previously reported.4 This antitumour agent was found to possess a
profitable toxicity profile, since it revealed only low cytotoxicity
towards normal human skin fibroblast (HSF) cells and a complete
lack of cytotoxicity towards normal green monkey kidney (GMK)
cells. In addition, its acute toxicity, assessed in mice according to
Litchfield and Wilcoxon method,6 and expressed as LD50 value,
was found to be relatively low (over 1000 mg kgꢀ1 when injected
intraperitoneally).4 We recently reported several successful 3-phe-
nyl-8-substituted derivatives based on the 7,8-dihydroimi-
dazo[2,1-c][1,2,4]triazin-4(6H)-one template with remarkable
antileukaemic and antimetastatic activities and discussed the pre-
liminary structure–activity relationships in this bioactive set of
substances.5 Substituents at the N-8 of heterobicyclic template
such as: the 4-methoxyphenyl in IT-4, monochloro-substituted
phenyl (mostly meta-chloro- and para-chlorophenyl in IT-5 and
IT-6, respectively) or dichloro-substituted phenyl (mostly 3,4-
dichloro- and 2,5-dichlorophenyl in IT-7 and IT-8, respectively)
proved to be necessary for the antileukaemic potency in this set
of compounds. Furthermore, 3-phenyl substituted derivatives
based on the 7,8-dihydroimidazo[2,1-c][1,2,4]triazin-4(6H)-one
template, containing the 4-methoxyphenyl (IT-4), a 3-methyl-
phenyl (IT-9) and monochloro-substituted phenyl (primarily
ortho-chloro-, meta-chloro- and para-chlorophenyl (IT-10, IT-5
and IT-6, respectively)) at the N-8, revealed significant antimeta-
static properties in human cervix epitheloid carcinoma cells.
Among
them
3-phenyl-8-(4-chlorophenyl)-7,8-dihydroimi-
dazo[2,1-c][1,2,4]triazin-4(6H)-one (IT-6), revealing remarkable
dose dependent viability decreases in human myeloma cells, and
exerting significant antimetastatic properties in human cervical
epithelial carcinoma cells, proved to be completely non-toxic to-
wards normal HSF cells.5
Prompted by these facts and continuing our attempt to achieve
medicinally important agents,3–5,7–12 we report herein an effective
and straightforward route for the synthesis, structure elucidation,
determination of the lipophilicity, comparison with results obtained
by use of hydrophobic substituent constants p, computational algo-
rithms, and anticancer activities in vitro of novel 3-(2-furanyl)-
8-aryl-7,8-dihydroimidazo[2,1-c][1,2,4]triazin-4(6H)-ones (12–22)
with a low cytotoxicity towards normal human skin fibroblast cells.
Cytotoxic effects of these solutes against two cancerous cells: hu-
man cervix epitheloid carcinoma cell line (HeLa) and human breast
cancer cell line (T47D) are discussed considering the substitution
pattern of the compounds, their selectivity and drug-likeness.
2. Results and discussion
2.1. Synthesis of novel 3-(2-furanyl)-8-aryl-7,8-dihydroimidazo
[2,1-c][1,2,4]triazin-4(6H)-ones (12–22)
The reported herein biologically important polyazaheterobicy-
cles are based on the 7,8-dihydroimidazo[2,1-c][1,2,4]triazin-
4(6H)-one template (12–22). The straightforward synthetic route
used in our laboratory that is applicable to a wide range of deriva-
tives is shown in Scheme 1. The tautomeric biologically active 1-ary-
limidazolidin-2-one hydrazones (1-aryl-2-hydrazinoimidazolines)
(1–11)3,7,13 were used as excellent key building blocks for the syn-
thesis of the desired heterobicycles. These ones were prepared in
good yields (62–76%) under mild reaction conditions by
a

5106
K. Sztanke et al. / Bioorg. Med. Chem. 19 (2011) 5103–5116
R
N
N
+
O
NH2
NH
O
1-11
HO
(i), (ii)
O
R
N
N
N
NH
O
HO
O
R
N
N
N
N
O
O
12-22
1,12: R = H; 2,13: R = 2-CH3; 3,14: R = 4-CH3; 4,15: R = 2,3-(CH3)2; 5,16: R = 2-CH3O;
6,17: R = 4-CH3O; 7,18: R = 2-Cl; 8,19: R = 3-Cl, 9,20: R = 4-Cl; 10,21: R = 3,4-Cl2; 11,22:
R = 2,6-Cl2
Scheme 1. Synthesis of 3-(2-furanyl)-8-aryl-7,8-dihydroimidazo[2,1-c][1,2,4]triazin-4(6H)-ones (12–22) considered in the present paper. Reagents and conditions: (i) = (1–
11)ꢁHI, n-BuOH, Et3N, reflux, 15–55 min; (ii) n-BuOH/DMF mixture, reflux, 4–7 h.
nucleophilic displacement of the methylsulfanyl group (attached to
the endocyclic C-2 atom in the 1-arylimidazoline structures) on treat-
ment with hydrazine monohydrate, with concomitant loss of methyl
mercaptan, using patent pending methodology11 (see Scheme 2).
All the reported herein compounds based on the 7,8-dihydroim-
idazo[2,1-c][1,2,4]triazin-4(6H)-one template (12–22) were gener-
ated by the annulation of a 1,2,4-triazine ring to an imidazolidine
core according to [4+2] pattern.14 In this case the N–C–N–N frag-
ment of the 1,2,4-triazine scaffold is derived from the imidazoli-
dine moiety, whereas the C–C fragment is provided by the
2-oxo-2-furanacetic acid in refluxing n-butanol/DMF mixture in the
presence of triethylamine to afford the corresponding novel 3-(2-
furanyl)-8-aryl-7,8-dihydroimidazo[2,1-c][1,2,4]triazin-4(6H)-ones
(12–22). The reaction conditions were established experimentally.
The synthesis proceeded under mild conditions with relatively good
overall yields (50–66%). Reaction of 1-aryl-2-hydrazonoimidazoli-
dine derivatives with 2-oxo-2-furanacetic acid has not been re-
ported in the literature yet. Thus, this reaction represents a novel
method for the synthesis of the compounds based on the 7,8-dihy-
droimidazo[2,1-c][1,2,4]triazin-4(6H)-one template.
second
a
-oxoacid reactant used, that is, 2-oxo-2-furanacetic acid.
The formation of novel 3-(2-furanyl)-8-aryl-7,8-dihydroimi-
dazo[2,1-c][1,2,4]triazin-4(6H)-ones has been proposed to proceed
by a stepwise mechanism. In the first step of reaction, the
Equimolar quantities of the available 1-arylimidazolidin-2-one
hydrazone hydroiodide salts of the type 1–11 were treated with

K. Sztanke et al. / Bioorg. Med. Chem. 19 (2011) 5103–5116
5107
R
R
R
N
N
N
(i)
S
NH
N
CH3
NH2
NH2
N
HI
N HI
NH HI
1-11
1: R = H; 2: R = 2-CH3; 3: R = 4-CH3; 4: R = 2,3-(CH3)2; 5: R = 2-CH3O; 6: R = 4-CH3O; 7:
R = 2-Cl; 8: R = 3-Cl, 9: R = 4-Cl; 10: R = 3,4-Cl2; 11: R = 2,6-Cl2
Scheme 2. Synthetic pathway for the preparation of tautomeric 1-arylimidazolidin-2-one hydrazones (1-aryl-2-hydrazinoimidazolines) (1–11), employed as key building
blocks for the synthesis of the title compounds. Reagents and conditions: (i) = hydrazine hydrate/MeOH, reflux, 24–30 h.
intermediate chain derivatives (depicted on Scheme 1 in the square
bracket) appeared as a result of condensation of the appropriate
1-aryl-2-hydrazonoimidazolidines (1–11) with 2-oxo-2-furanacetic
acid and concomitant loss of water molecule. Because of the presence
of the hydrogen atom on the endocyclic N-3 nitrogen in structures of
these intermediates, their subsequent cyclocondensation followed
with concomitant loss of water molecule and finally afforded
compounds based on the 7,8-dihydroimidazo[2,1-c][1,2,4]triazin-
4(6H)-one template (12–22). The scrutiny of 1H NMR, 13C NMR
and infrared (IR) data confirmed that under the reaction condi-
tions, formation of the 7,8-dihydroimidazo[2,1-c][1,2,4]triazin-
4(6H)-one system of type 12–22 is accompanied with loss of water
molecule. Additionally the concurrent course of cyclization reac-
tion to the 7,8-dihydroimidazo[2,1-c][1,2,4]triazole derivatives
(with concomitant loss of formic acid molecule) was excluded be-
cause no traces of these derivatives were detected.
In view of continuous search for novel and more efficient cyto-
toxic agents that show a profitable toxicity profile, the reported
herein preparation route, leading to the formation of the desired
compounds (12–22), might be considered as a useful synthetic
method because of the affordability of the starting reagents and
reactants, good yields obtained and straightforward product isola-
tion. Furthermore, this efficient synthetic route allows one to ob-
tain a large number of imidazotriazinone analogues.
NMR spectral characteristic of the title compounds (12–22)
revealed in their 1H NMR spectra signals of the H-7 and H-6 (of
both endocyclic methylene groups) as a multiplet in the region
3.98–4.36 ppm (compounds 12–18, 20 and 22), or as a singlet at
d 4.21 ppm (compounds 19 and 21), integrating for four protons
and four protons, respectively. Moreover, in the 1H NMR spectra
of all the investigated imidazotriazinones (12–22), the additional
signals derived from the three-furanyl protons (H-4, H-3, H-5)
were observed. The signals of the H-4 hydrogens of the furanyl ring
were seen at d values in the region 6.61–6.68 ppm, increasing in
the order of substituents: 2,3-(CH3)2 < 2-CH3 = 2-CH3O < 2-Cl < 4-
CH3 = 4-CH3O = 2,6-Cl2 < H < 3-Cl < 4-Cl < 3,4-Cl2. The signals of
the H-3 hydrogens of the furanyl were found to be in the region
7.30–7.43 ppm, increasing in the order of R: 2,3-(CH3)2 = 2-
CH3O < 2-CH3 < 4-CH3O = 2-Cl < 4-CH3 < 2,6-Cl2 < H < 4-Cl < 3-Cl <
3,4-Cl2. In turn, the d values of the H-5 hydrogens of the furanyl
were observed in the region 7.78–7.90 ppm, increasing in the order
of R: 2,3(CH3)2 < 2-CH3 = 2-CH3O < 2-Cl < 2,6-Cl2 < 4-CH3O < 4-
CH3 < H = <3-Cl < 4-Cl < 3,4-Cl2.
In the 13C NMR spectra of the title compounds 12–22, the chem-
ical shift values of the C-7 and C-6 carbon atoms confirmed un-
equal character of both methylene groups as well (d values in the
region 44.8–47.6 ppm for C-7, and d values in the region 40.0–
41.3 ppm for C-6). In DEPT-135 experiment both methylene carbon
atoms were seen as negative signals. The C-4, C-3 and C-5 tertiary
carbons of the furanyl ring were seen at d values ca. 111.2, 112.7
and 143.7 ppm, respectively whereas the C-2 quaternary carbons
of the furanyl were observed at d ca. 146.8 ppm. Additionally, the
endocyclic triazine-C@O (triazine-C-4), triazine-C-3 and C-8a sig-
nals were seen at d values ca. 140.2, 149.8 and 151.0 ppm,
respectively.
2.2. Spectral characteristic of novel 3-(2-furanyl)-8-aryl-7,8-
dihydroimidazo[2,1-c][1,2,4]triazin-4(6H)-ones (12–22)
General structure for unknown imidazotriazinones (12–22)
with atom numbering is shown in Figure 1. The numbering of
the 7,8-dihydroimidazo[2,1-c][1,2,4]triazin-4(6H)-one scaffold
starts with a triazine N-1 nitrogen.
Full spectral characteristic of the title compounds (12–22) is
given in experimental protocols.
In the infrared (IR) spectra of the title compounds 12–22, the
presence of absorption bands in the region 1676–1691 cmꢀ1
and 1548–1557 cmꢀ1, attainable to the triazine-C@O group and
the C@N bond at the ring junction, respectively confirmed the for-
mation of cyclic products, that is bicyclic structures with fully
conjugated system. In the IR spectra of 12–22, the difference
between the frequency of the ring carbonyl group C@O and the
frequency of the C@N bond at the ring junction was found to
be in the range 128–140 cmꢀ1. These data are consistent with
those previously reported by Le Count and Taylor,15 since a sim-
ilar difference between these frequencies in the infrared spectra
of a wide range of fused 1,2,4-triazin-5-ones was observed strictly
for fully conjugated system, where the C@N bond is contained
within the triazinone ring. In addition, in the IR spectra of imi-
dazotriazinones of the type 12–22, the absorption band at around
1219 cmꢀ1 was seen, due to the C–O–CH stretching vibration of
the furanyl ring.
2.3. Determination of the lipophilicity and comparison between
experimental and computational approaches
2.3.1. Lipophilicity of novel 3-(2-furanyl)-8-aryl-7,8-dihydro
imidazo[2,1-c][1,2,4]triazin-4(6H)-ones (12–22), expressed as
their log kw values
Lipophilicity as one of the most important physicochemical
properties of the bioactive molecule is of high significance in the
drug design, quantitative structure–activity relationship
(QSAR)16,17 and quantitative structure retention relationship
(QSRR) studies. This molecular property affects penetration of cell
membranes, blood–brain barrier distribution, drug-receptor inter-
action, drug absorption, excretion, metabolism, toxicity, etc.18–26
The increasing hydrophobicity of a drug-like compound can be
empirically correlated with the increase in biological activity.

5108
K. Sztanke et al. / Bioorg. Med. Chem. 19 (2011) 5103–5116
Table 2
'
4
Terms of the equation log k = log kw ꢀ S
u for the investigated 3-(2-furanyl)-8-aryl-
'
'
5
3
7,8-dihydroimidazo[2,1-c][1,2,4]triazin-4(6H)-ones (12–22) on an octadecyl silica
column (LC-18) with reversed mobile phases (methanol–water—20% acetate buffer
pH 3.5—octane-1-sulfonic acid sodium salt (0.01 M Lꢀ1))
R
'
'
2
6
Compd
S (slope)
log kw
r
F
s
n
1
N
N
12
13
14
15
16
17
18
19
20
21
22
ꢀ4.01 ( 0.33)
ꢀ3.99 ( 0.38)
ꢀ4.58 ( 0.38)
ꢀ4.47 ( 0.42)
ꢀ4.17 ( 0.43)
ꢀ4.44 ( 0.41)
ꢀ4.30 ( 0.38)
ꢀ4.65 ( 0.24)
ꢀ4.66 ( 0.25)
ꢀ5.03 ( 0.30)
ꢀ4.27 ( 0.25)
2.64 ( 0.17)
2.48 ( 0.20)
3.27 ( 0.20)
3.00 ( 0.22)
2.54 ( 0.23)
2.89 ( 0.21)
2.78 ( 0.20)
3.45 ( 0.14)
3.45 ( 0.14)
4.01 ( 0.19)
2.97 ( 0.15)
0.9898
0.9869
0.9898
0.9869
0.9841
0.9873
0.9883
0.9947
0.9944
0.9946
0.9931
144.57
112.21
144.33
112.04
92.27
115.54
126.26
377.68
355.86
276.65
286.45
0.1055
0.1190
0.1207
0.1337
0.1372
0.1305
0.1209
0.1001
0.1033
0.0957
0.1055
5
5
5
5
5
5
5
6
6
5
6
8a
7
8
N 2
N
6
4
5
2
O
3
5
O
3
4
Figure 1. General structure for 3-(2-furanyl)-8-aryl-7,8-dihydroimidazo[2,1-
c][1,2,4]triazin-4(6H)-ones (12–22) with atom numbering.
50,51
´
data and the Soczewinski–Wachtmeister equation
(expressed
The chromatographic techniques, in particular the reversed
phase-high performance liquid chromatography (RP-HPLC) and
the reversed phase-liquid chromatography (RP-TLC), that simulate
the octanol–water partitioning,27 are widely used for determina-
tion of the lipophilicity of various bioactive molecules.18,20,26–38
These chromatographic approaches correlate the hydrophobicity
of solutes with their retention parameters,19,39,40 and have the ex-
tended measurable lipophilicity range of the partition coefficient
(P) to 106 (in comparison to the traditional shake-flask method,
having a limited measurable lipophilicity range of P from 10ꢀ2 to
104).34 The partition coefficient between an aqueous and a station-
ary phase in reversed-phase liquid chromatography is strictly cor-
related with the lipophilicity of a solute investigated.41,42
In the HPLC series of experiments the newly synthesized 3-
(2-furanyl)-8-aryl-7,8-dihydroimidazo[2,1-c][1,2,4]triazin-4(6H)-
ones (12–22) were chromatographed on the octadecyl silica (ODS)
stationary phase, using mixtures of water (H2O) and the organic
modifier-methanol (MeOH) as mobile phases. Methanol concentra-
tions in the mobile phase were from 30% to 80%. Octane-1-sulfonic
acid sodium salt (OSA-Na) and 20% acetate buffer (pH 3.5) was
added to the mobile phase (eluent containing 0.01 M Lꢀ1 OSA-Na
in organic modifier (MeOH)—buffered mobile phase).
The investigated compounds, such as heterocyclic bases are dif-
ficult for the chromatographic analysis. In the analysis of basic
compounds anionic ion-pairing (IP) reagents, such as octane-1-sul-
fonic acid sodium salt, pentane-1-sulfonic acid sodium salt (PSA-
Na), hexane-1-sulfonic acid (HSA), sodium dodecyl sulfate, alkyl
sulfonates, forming neutral associates are widely used.43–49 The
neutral analyte form possesses higher affinity to the organic com-
ponent of the mobile phase. Better correlation between log kw and
log P values are obtained in systems with ion-pairs, thereupon the
easier movement of neutral associates (ion-pairs) to the organic
phase. In mobile phases of a proper pH with addition of the pair-
creative reagent the ion-pair between analyte ionized forms and
the pair-creative reagent is appeared. Additional advantages of sys-
tems with ion-pairs are found to be both the improvements in peak
shape for analytes and their symmetry. In eluent systems contain-
ing octane-1-sulfonic acid sodium salt, peaks of most of investi-
gated analytes were found to be symmetric.
in the form of the Snyder–Soczewinski relationship52) (Eq. 1):
´
log k ¼ log kw ꢀ S
u
ð1Þ
where the log kw corresponds to the capacity factor k for pure water
as mobile phase, that is, = 0; S denotes the slope of regression plot
and is the volume fraction of organic modifier in the mobile
phase.
u
u
The values of the Fishers statistical significance test were al-
ways higher than the critical value F, thus proving additionally a
good fit of the data with the examined model.
The retention data of presented herein-unknown imidazotriaz-
inone derivatives have not been described in the literature yet and
are the first time presented in this paper. The lipophilicity of all the
solutes (12–22) determined by reversed-phase HPLC, and being
based on extrapolation to
u = 0, is expressed as their log kw values.
The log kw value, defined as the logarithm of capacity factor for
pure water, is widely known as a lipophilicity parameter and is of-
ten used in QSAR studies.37,53 The log kw value proved to be useful
when investigating a series of compounds or drugs covering a wide
lipophilicity range21,35 and helpful in the standardisation of the
retention data. Furthermore, the correlation between the log kw
and the log Poctanol values are considered as good within a series
of closely related compounds.21,54,55
The determined log kw values of novel 3-(2-furanyl)-8-aryl-7,
8-dihydroimidazo[2,1-c][1,2,4]triazin-4(6H)-ones (12–22) were
found to be in the range 2.48–4.01. The relatively high values of
the investigated solutes can be empirically correlated with their
increased permeability through cell membranes, and therefore
can play a crucial role in their antitumour activity.
In this series of closely related solutes (analogues), the lipophil-
icity was dependent both on the character of substituent and on
the position of the substituent on the phenyl ring. The most lipo-
philic imidazotriazinone (21), containing the 3,4-dichloro substitu-
ent as R (with the two electron-withdrawing chlorine atoms), had
the highest log kw value, that is, 4.01. On the contrary, the com-
pound 13 with ortho-CH3 group at the phenyl ring, revealed the
lowest lipophilicity (log kw = 2.48). It was proved that the values
of the log kw for the investigated compounds based on the 7,8-
dihydroimidazo[2,1-c][1,2,4]triazin-4(6H)-one template increased
in the following order of substituents: 2-CH3 < 2-CH3O < H < 2-
Cl < 4-OCH3 < 2,6-Cl2 < 2,3-(CH3)2 < 4CH3 < 3-Cl = 4-Cl < 3,4-Cl2.
Lipophilicity was affected by the chloro substituent on the
phenyl ring, and then the values of the log kw increased in the
order: 2-Cl < 2,6-Cl2 < 3-Cl = 4-Cl < 3,4-Cl2. Lipophilicity was also
exerted by the methyl substituent as R, and then increased in the
order: 2-CH3 < 2,3-(CH3)2 < 4-CH3. In case of the two compounds
The experimental data obtained for reversed-phase HPLC sys-
tem: the intercepts of regression curve (log kw), slopes (S), regres-
sion coefficients (r), F-test values (of the Fisher test of significance),
standard errors of estimate (s) and the number of points of concen-
tration of modifier in eluent systems (n) calculated for the 95% con-
fidence level are listed in Table 2.
The high values of regression coefficients (r >0.9841) for all ele-
ven solutes based on the 7,8-dihydroimidazo[2,1-c][1,2,4]triazin-
4(6H)-one template proved the excellent fit between experimental

K. Sztanke et al. / Bioorg. Med. Chem. 19 (2011) 5103–5116
5109
(16 and 17), with the electron-donating methoxy group attached to
the phenyl ring, the analysis of RP-HPLC results confirmed the
higher lipophilicity of the para-methoxy derivative (log kw = 2.89),
comparatively to the ortho-methoxy analogue (log kw = 2.54).
Generally, the determined values of the log kw of the para- and
the meta-substituted solutes were found to be higher than those of
the ortho-substituted derivatives. Thus, the decreased lipophilicity
values of the ortho-substituted compounds can be attributed to the
steric effect. Presumably, the substituents in the ortho position at
the phenyl ring display the steric hindrance forcing-out-of-plane
free rotation, which decreases lipophilicity53–56 of molecules (13,
15, 16, 18 and 22).
The examined solutes: 19, 20 and 21, with the electron-with-
drawing monochloro and dichloro substituents, revealed the larg-
est log kw values: 3.45 for both meta-chloro- and para-chloro-
substituted derivatives and 4.01 for the 3,4-dichloro-substituted
compound.
It was proved that adding the second chloro substituent to the
meta-chloro- or the para-chloro-substituted derivative (19 or 20)
resulted in noticeable lipophilicity increase. It is seen for the 3,4-
dichloro compound (21), having the highest lipophilicity value
(log kw = 4.01). However, adding the second ortho-chloro group to
the ortho-chloro-substituted derivative (18) resulted only in slight
increase in lipophilicity. It is clearly seen in the case of the 2,6-di-
chloro-substituted analogue (22), that has subtly increased lipo-
philicity (log kw value = 2.97) in comparison to the ortho-chloro
analogue (log kw = 2.78), supposedly due to the steric hindrance
forcing-out-of-plane rotation of molecule, which decreases
lipophilicity.
In comparing the influence of the 3,4-dichloro with a 2,6-di-
chloro substitution on the lipophilicity, the two ortho-substitutions
decreased significantly the log kw value. It has been proved for the
2,6-dichloro compound (22), revealing the decrease in lipophilicity
by 1.04 (log kw = 2.97), in comparison to the 3,4-dichloro analogue
(21), with the log kw of 4.01. The observed remarkable decrease in
the lipophilicity may be explained due to the steric effect forcing
out-of-plane rotation,56–59 that is caused by the ortho–ortho substi-
tution of the bulky halogen atoms in compound 22.
the significant lipophilicity increase in case of the para-CH3O com-
pound (17) (log kw = 2.89), in relation to the parent compound (12),
having a log kw value of 2.64.
2.3.2. Comparison of determined log kw values and calculated
(log kw (theoretical)) values
In view of Hansch and Fujita approach, based on the concept of
hydrophobic substituent constants p, the trial was carried out with
the purpose to derive the theoretical log kw values for imidazotri-
azinone analogues that have substituents on the phenyl ring to
compare them with those determined experimentally.
The above-mentioned Hansch and Fujita approach, allows one
to calculate the partition coefficient for a structural analogue of
the parent compound theoretically by knowing the substituent
hydrophobicity constants p 16,56,60–63
that is, the contribution that
,
a given substituent makes to hydrophobicity, relative to hydrogen
and having the determined experimentally value of the partition
coefficient for the parent compound without substituent.62 So cal-
culated value of the partition coefficient for an analogue that have
a given substituent can be mostly considered as the first approxi-
mation of that determined experimentally. For example, the calcu-
lated value of the logarithm of partition coefficient for such a
simple structure as para-nitrotoluene is 2.41, whereas the experi-
mentally determined log P value for this compound is found to
be 2.37. The theoretical log P value for para-nitrotoluene is the
sum of log P for benzene (being the parent compound) and the sub-
stituent constant values for CH3 (0.56) and NO2 (ꢀ0.28).56
We were able to calculate the theoretical log kw values for the
following substituted analogues of the examined imidazotriazi-
nones: 13, 14, 15, 16, 17, 18, 19, 20, 21, 22 (see Table 3), having
the determined log kw values for the parent compound, that is,
the phenyl analogue (12) and known values of hydrophobicity con-
stants
p for CH3, CH3O and Cl for aromatic substituents (0.56,
ꢀ0.02 and 0.7156, respectively), using the Eq. 2:
log kwðtheoreticalÞ ¼ log kwðparent compoundÞ þ pðsubstituentÞ
ð2Þ
where the log kw (theoretical) is the theoretically calculated logarithm
of partition coefficient for the compound with a substituent,
Analysing the impact of the electron-donating methyl group on
the retention indices, we found that ortho-methyl substitution
(compound 13) evoked only slight drop in lipophilicity
(log kw = 2.48), whereas para- methyl substitution (solute 14) re-
sulted in significant lipophilicity increase (log kw = 3.27) in relation
to parent compound 12, revealing the log kw of 2.64. Introducing
the additional methyl substituent to the ortho-methyl derivative
(13), having the log kw of 2.48, resulted in noticeable lipophilicity
increase. It is clearly seen for the 2,3-dimethyl compound (15),
having the log kw = 3.00.
log kw
is the logarithm of partition coefficient for
(parent compound)
the parent compound and p(substituent) denotes the hydrophobicity
constant for a methyl, methoxy and chloro group (attached to
p
phenyl ring).
Attempts were made to correlate both determined and theoret-
ically calculated log kw values.
Relationships between determined log kw values and those the-
oretically calculated are shown in Figures 2 and 3. We were unable
to derive a good correlation, in respect thereof all the substituted
compounds (13–22), since then the regression coefficient (r) value
is equal 0.5387 (see Fig. 2), due to the presence of four outliers
Introducing the strong electron-donating methoxy substituent
to the phenyl ring resulted in slight lipophilicity decrease in case
of the ortho-CH3O analogue (16) (log kw = 2.54), whereas caused
Table 3
Comparison of determined log kw values and those theoretically calculated (log kw
values)
(theoretical)
Parent compd
Compd with substituent
log kw for parent compd
log kw for compd with substituent
Theoretical log kw for compd with substituent
12
12
12
12
12
12
12
12
12
12
13
14
15
16
17
18
19
20
21
22
2.64
2.64
2.64
2.64
2.64
2.64
2.64
2.64
2.64
2.64
2.48
3.27
3.00
2.54
2.89
2.78
3.45
3.45
4.01
2.97
2.64 + 0.56 = 3.20
2.64 + 0.56 = 3.20
2.48 + 2 ꢂ 0.56 = 3.76
2.64 + (ꢀ0.02) = 2.62
2.64 + (ꢀ0.02) = 2.62
2.64 + 0.71 = 3.35
2.64 + 0.71 = 3.35
2.64 + 0.71 = 3.35
2.64 + 2 ꢂ 0.71 = 4.06
2.64 + 2 ꢂ 0.71 = 4.06

5110
K. Sztanke et al. / Bioorg. Med. Chem. 19 (2011) 5103–5116
w(horetical)
w(horetical)
k
k
log k w
log k w
Figure 2. Correlation between measured log kw values and those theoretically
calculated (log kw (theoretical)) for all the imidazotriazinone analogues that have
substituents at the phenyl ring.
Figure 3. Correlation between experimental log kw values and those theoretically
calculated (log kw for substituted imidazotriazinone analogues after
omission of four outliers (ortho-substituted derivatives: 13, 15, 18 and 22 for which
the ortho-effect was distinctly marked).
)
(theoretical)
(ortho-substituted derivatives: 13, 15, 18 and 22 for which the
ortho-effect was distinctly marked). For these ortho-substituted
compounds the determined log kw values were found to be dis-
tinctly lower than those theoretically calculated using the substitu-
ent constants
p
of Hansch and Fujita.56 Previously published
examples suggest that substituents in the ortho position hinder
the rotational freedom of the aromatic phenyl ring, which may de-
crease lipophilicity.37,56–59 The regression equation obtained after
P
omission of these four outliers was found to be: log kw
(theoretical)
= 1.0345 log kw (experimental) ꢀ 0.1811 and then the correlation coef-
ficient was very good, r = 0.9732, that is illustrated in Figure 3.
2.3.3. Correlation between log kw values (for methanol–water
mobile phases on the octadecyl silica column (LC-18)) for the
investigated imidazotriazinones and calculated log P values
The relationships between log kw and log P values were also
studied. Log P values were obtained from an ALOGPS 2.1. soft-
ware64–66 by the use of nine different computational methods for
estimation of log P: MLOGP, ALOGPs, AClogP, ABlogP, miLogP,
ALOGP, KOWWIN, XLOGP2 and XLOGP3. The graphs are drawn
by plotting log P calculated values versus log kw determined val-
ues. The correlation coefficients (r) are provided in Figure 4 and
in Supplementary data (see Figs. 1–8). Results obtained from the
reversed-phase HPLC were found to be the most different from
those obtained by use of MLOGP (see Fig. 8 in Supplementary data).
The predicted values of log P by use of AB log P algorithm revealed
the best correlation with the experimental log kw values for the
investigated solutes, since a good correlation (r = 0.7760) between
these quantities was found and shown in Figure 4. Lower correla-
tions were observed between log P predicted by other computa-
tional methods and the chromatographically determined
hydrophobicity parameter, log kw. All these correlations are shown
in Figures 1–8 in Supplementary data.
log P = 0.8348 log k w - 0.1606
r = 0.7760
log k w
Figure 4. Correlation between log kw values for reversed mobile phases (methanol–
water—20% acetate buffer pH 3.5—octane-1-sulfonic acid sodium salt (0.01 M Lꢀ1))
on the octadecyl silica column (LC-18)) and calculated log P (ABlogP) values for the
investigated imidazotriazinones.
determined log kw values were higher by 0.15, 0.94 and 0.26,
respectively than those from ABlogP method of calculation. In turn,
for compounds 16 and 17, with the ortho-CH3O and the para-CH3O
substituents, the determined log kw values were higher by 1.04 and
0.97, respectively, than those predicted on the basis of ABlog P
method of calculation. The parent compound (12) that has unsub-
stituted phenyl ring revealed the log kw value higher by 0.78 than
that predicted on the basis of ABlogP method of calculation.
The lipophilicity of drug candidates should be determined
experimentally, because of the imperfections in lipophilicity pre-
diction.26 A comparison between experimental log kw values and
those obtained by use of computational algorithms showed that
none computational method was able to predict correctly the log P
of these basic compounds. Therefore the presented herein
reversed-phase HPLC technique can be regarded as the better
Both retention data and software calculated retention parame-
ters indicated the 3,4-dichloro-substituted compound (21) as sol-
ute of the highest lipophilicity. For compounds 19, 20, 21 and 22,
that have 3-Cl, 4-Cl, 3,4-Cl2 and 2,6-Cl2 substituents, the deter-
mined log kw values were found to be higher by 0.78, 0.88, 0.73
and 0.21, respectively, than those predicted on the basis of ABlogP
method of calculation. Besides, for compounds 13, 14 and 15, with
2-CH3, 4-CH3 and 2,3-(CH3)2 substituents at the phenyl ring, the

K. Sztanke et al. / Bioorg. Med. Chem. 19 (2011) 5103–5116
5111
Table 4
Growth inhibition (GI) ratio of normal and tumour cells by the examined compounds in an effective concentration of 50 l
g mLꢀ1
R
N
N
N
N
O
O
Compound
R
Incubation time (h)
Growth inhibition (%) in cell lines
HeLa
HSF
T47D
12
13
14
15
16
17
18
19
20
21
22
H
24
48
72
24
48
72
24
48
72
24
48
72
24
48
72
24
48
72
24
48
72
24
48
72
24
48
72
24
48
72
24
48
72
50
50
60
0
0
0
30
50
50
0
0
0
50
75
100
25
25
50
50
50
80
60
75
90
25
35
80
80
80
95
80
80
100
80
80
95
50
75
90
50
50
95
80
80
95
50
85
95
25
50
75
40
60
80
0
15
25
0
30
70
50
70
80
5
15
25
20
50
80
0
15
15
10
20
40
10
20
25
2-CH3
4-CH3
2,3-(CH3)2
2-OCH3
4-OCH3
2-Cl
0
0
5
10
50
60
10
20
20
25
60
75
50
80
90
0
3-Cl
4-Cl
3,4-Cl2
2,6-Cl2
50
80
10
15
15
Normal cell line: HSF, human skin fibroblast cells.
Cancer cell lines: HeLa (ECACC 93021013), human Negroid cervix epitheloid carcinoma cells; T47D (ECACC 85102201), human breast carcinoma cells.
An effective concentration of 50 l
g mLꢀ1 corresponds to following concentration values: 0.18 mM (12), 0.17 mM (13, 14), 0.16 mM (15), 0.16 mM (16, 17), 0.16 mM (18–20),
0.14 mM (21, 22).
alternative in comparison to computational methods of lipophilic-
ity prediction. The determined log kw values for biologically active
compounds based on the 7,8-dihydroimidazo[2,1-c][1,2,4]triazin-
4(6H)-one template can be regarded as their hydrophobic parame-
ters. These ones are the first time presented and can be used in
further QSAR studies.
mate the selectivity profile of tested compounds. The obtained re-
sults are presented as growth inhibition (GI) ratio of normal and
tumour cells by the examined compounds given in an effective
concentration of 50 l
g mLꢀ1 after 24-, 48- and 72-h periods of
incubation (see Table 4).
The majority of novel 3-(2-furanyl)-8-aryl-7,8-dihydroimi-
dazo[2,1-c][1,2,4]triazin-4(6H)-ones were found to be evidently
effective in vitro against human cervix epitheloid carcinoma (HeLa)
and human breast carcinoma (T47D) cells.
2.4. Anticancer activities in vitro of novel 3-(2-furanyl)-8-aryl-
7,8-dihydroimidazo[2,1-c][1,2,4]triazin-4(6H)-ones (12–22)
The highest cytotoxic potency against HeLa cancer cells was
found for compound 18, bearing the electron-withdrawing chloro
substituent at ortho- position of the phenyl ring. This imidazotriaz-
inone showed 80%, 80% and 100% growth inhibition against HeLa
cells after 24, 48 and 72 h of incubation, respectively.
Simultaneously, this evidently effective compound (18) revealed
the distinctly marked lower cytotoxicity towards normal cell
line—human skin fibroblast (HSF) cells and several times higher
The newly synthesized compounds based on the 7,8-dihydro-
imidazo[2,1-c][1,2,4]triazin-4(6H)-one template (12–22) were
screened for their cytotoxicity against two tumour cell lines: HeLa
(ECACC 93021013, human Negroid cervix epitheloid carcinoma
cells) and T47D (ECACC 85102201, human breast carcinoma cells)
in vitro. Furthermore, the primary cell line of normal human skin
fibroblast (HSF) cells was included in the cytotoxicity study to esti-

5112
K. Sztanke et al. / Bioorg. Med. Chem. 19 (2011) 5103–5116
against HeLa cancer cells. Therefore, imidazotriazinone of the type
18 might be considered as a basis for the design of novel low toxic
anticancer agents.
compound 12, bearing an unsubstituted, electron-rich phenyl ring,
which seemed to be beneficial for activity. This potent small mole-
cule proved to be effective both against T47D cancer cells (50%, 85%
and 95% growth inhibition) and against HeLa cancer cells (50%, 75%
and 100% growth inhibition) after 24, 48 and 72 h of incubation,
respectively.
In the case of the most potent 2-Cl derivative (18), adding a fur-
ther chloro substituent at position 6 of the phenyl resulted in still
evidently active and selective 2,6-Cl2 compound (22). This one re-
vealed the remarkable antitumour activity against HeLa cancer
cells (80%, 80% and 95% growth inhibition after 24, 48 and 72 h
of incubation, respectively), slightly improved activity against
T47D cancer cells and slightly improved selectivity for cancer cells
over normal cells, comparatively to 18. In conclusion, the two
promising cytotoxic agents of the type 18 and 22 demonstrated
selective cytotoxicity for HeLa cancer cells over normal HSF cells.
Therefore these potent compounds can be viewed as the lead
structures for the design of novel low-toxic anticancer drugs.
Replacing the 2,6-Cl2 substituent with a 3,4-Cl2 group decreased
the activity against HeLa cancer cells (50%, 50% and 95% growth
inhibition after 24, 48 and 72 h of incubation, respectively) and
resulted in increase of the cytotoxicity against T47D cancer cells
after 72 h of incubation. It is clearly seen that this structural
change was also responsible for the unwanted cytotoxicity
increase towards normal HSF cells after 48- and 72-h periods of
incubation. Thus, it was demonstrated that the two-chloro groups
had to be retained in ortho–ortho positions of the phenyl ring
rather than in meta–para positions in order to retain the selective
action of the structure. Therefore, in this case, the 2,6-dichloro
substitution in comparison to 3,4-dichloro substitution is found
to be the better substitution pattern for the phenyl ring in this
set of compounds.
Moving the chloro substituent to the meta position of the phe-
nyl ring resulted in still evidently effective compound (19) against
HeLa carcinoma cells (80%, 80%, 95% growth inhibition after 24, 48,
72 h of incubation, respectively). Furthermore, this structural
change led to an increase in cytotoxicity for T47D cancer cells
(20%, 50%, 80% growth inhibition after 24, 48, 72 h of incubation,
respectively), comparatively to the 2-Cl analogue (18). Unfortu-
nately, this structural change led to a fall in selectivity, demon-
strating a significant cytotoxicity increase towards normal HSF
cells. A chloro substituent para to the phenyl ring led to a slight
drop in the potency against HeLa cancer cells (50%, 75%, 90%
growth inhibition after 24-, 48-, 72-h periods of incubation, respec-
tively) and to the cytotoxicity decrease against T47D cancer cells
but also evoked a significant cytotoxicity increase towards normal
HSF cells, as evidenced by the para-Cl compound (20), compara-
tively to 19. In conclusion, it was demonstrated that the mono-
chloro group had to be retained in the ortho position of the aro-
matic phenyl ring rather than meta or para positions in order to re-
tain the selective action of the structure.
Compound 14 revealed noticeable inhibitory activities both in
T47D cancer cells (40%, 60% and 80% growth inhibition) and in
HeLa cancer cells (50%, 50% and 80% growth inhibition) after 24-,
48- and 72-h periods of incubation, respectively. Placing a methyl
group ortho to the phenyl ring significantly decreased the potency
against HeLa cancer cells but fortunately resulted in the com-
pletely loss of cytotoxicity towards normal HSF cells as evidenced
by the ortho-CH3 derivative (13). Introducing a further methyl sub-
stituent to the phenyl ring to give the 2,3-(CH3)2 derivative (15)
was also tried. This resulted in a remarkable increase of the po-
tency against HeLa cancer cells (60%, 75% and 90% growth inhibi-
tion after 24, 48 and 72 h of incubation, respectively) and
simultaneously led to decrease of the activity against T47D cancer
cells, comparatively to the 2-CH3 analogue (13). It is worth noting
that this structural change had no influence on the selectivity pro-
file because compound 15 (so as 13) was found to be completely
non-toxic towards normal HSF cells.
Structure–activity relationships in the bioactive set of the
compounds based on the 7,8-dihydroimidazo[2,1-c][1,2,4]triazin-
4(6H)-one template, led to heterocycles with remarkable antitu-
mour activities in vitro. It is noteworthy that 2-chloro and 2,6-di-
chloro substitution was exactly the best choice of substitution
pattern in the series of mono- and dichloro-substituted-phenyl
analogues of 3-(2-furanyl)-7,8-dihydroimidazo[2,1-c][1,2,4]tria-
zin-4(6H)-one, as evidenced by the highly potent and low toxic
compounds (18 and 22). In addition, it was proved that 2-methyl,
2,3-dimethyl and 2-methoxy substitution at the phenyl ring was
essential for the selectivity of the examined structural analogues,
as evidenced by imidazotriazinones of the type 13, 15 and 16.
Two bioisosteric replacements of the methyl group that is
placed both in the ortho and in the para position of the phenyl ring
(compounds 13 and 14, respectively) with a chloro substituent, of
comparable size and lipophilicity, in place of the methyl, led to the
active bioisosters (18 and 20, respectively) in vitro. The best bal-
ance of cytotoxicity versus more selectivity was achiewed with a
methyl group. The para-methyl and the ortho-methyl substituents
on the phenyl ring were found to be better than the para-chloro
and the ortho-chloro groups for activity against T47D cells and
Table 5
The IC50 values against HeLa and T47D cancer cells for the most active imidazotri-
azinone 18 and for the parent compound 12, calculated from the dose–response
curves obtained after 24, 48 and 72 h of incubation
Three compounds: 17, 19 and 22 proved to be evidently effec-
tive in vitro. All these derivatives revealed the same strong antipro-
liferative effects (80%, 80% and 95% growth inhibition against HeLa
cancer cells after 24, 48 and 72 h of incubation, respectively). Fur-
thermore the para-CH3O derivative (17) and the meta-Cl analogue
(19) exerted remarkable cytotoxic activities against T47D cancer
cells (50%, 70%, 80% growth inhibition, and 20%, 50%, 80% growth
inhibition after 24-, 48-, 72-h periods of incubation, respectively).
Placing the methoxy group ortho to the phenyl ring led to decrease
of the antiproliferative activity against HeLa and T47D cancer cells
(25%, 35%, 80% growth inhibition and 0%, 30%, 70% growth inhibi-
tion after 24, 48, 72 h of incubation, respectively) but also resulted
in a nearly complete loss of the cytotoxicity towards normal HSF
cells as evidenced by the 2-CH3O compound (16), comparatively
to the 4-CH3O analogue (17).
R
N
N
N
N
O
O
Compound
R
Incubation time (h) IC50 values (in mM) in tumour cell
lines
HeLa
T47D
12
18
H
24
48
72
0.18
0.13
0.10
0.11
0.11
0.09
0.18
0.11
0.10
>0.16
>0.16
>0.16
2-Cl 24
In turn, the highest cytotoxicity against human breast carcinoma
(T47D) cells in the set of the tested novel 3-(2-furanyl) -8-aryl-7,8-
dihydroimidazo[2,1-c][1,2,4]triazin-4(6H)-ones was found for
48
72

K. Sztanke et al. / Bioorg. Med. Chem. 19 (2011) 5103–5116
5113
Table 6
Chemical structures and drug-likeness68 of novel 3-(2-furanyl)-8-(phenyl or substituted-phenyl)-7,8-dihydroimidazo[2,1-c][1,2,4]triazin-4(6H)-ones (12–22) those lipophilicities
were determined
R
N
N
N
N
O
O
a
Compd
R
Formula
15H12N4O2
C16H14N4O2
C16H14N4O2
C17H16N4O2
C16H14N4O3
C16H14N4O3
C15H11ClN4O2
log kw (<5)
H-bond donorsb (<5)
H-bond acceptorsc (<10)
MWT (<500)
12
13
14
15
16
17
18
19
20
21
22
H
C
2.64
2.48
3.27
3.00
2.54
2.89
2.78
3.45
3.45
4.01
2.97
0
0
0
0
0
0
0
0
0
0
0
6
6
6
6
7
7
6
6
6
6
6
280.28
294.31
294.31
308.34
310.31
310.31
314.73
314.73
314.73
349.17
349.17
2-CH3
4-CH3
2,3-(CH3)2
2-OCH3
4-OCH3
2-Cl
3-Cl
4-Cl
3,4-Cl2
2,6-Cl2
C
C
15H11ClN4O2
15H11ClN4O2
C15H10Cl2N4O2
C15H10Cl2N4O2
a
log kw values (HPLC) determined on an octadecyl silica column (LC-18) with mobile phase of the type MeOH–H2O according to the procedure reported in this paper (see
Table 2).
b
H-bond donors expressed as the sum of NHs and OHs.
H-bond acceptors expressed as the sum of Ns and Os.
c
for decreased cytoxicity towards HSF cells. However, the para-
chloro and the ortho-chloro substituents on the phenyl ring were
better than a methyl group for activity against HeLa cells. From
the point of view of further studies, the chloro substituent can be
considered as resistant to oxidative biotransformation in vivo,
whereas a methyl group as liable to oxidation.67
The IC50 values against HeLa and T47D carcinoma cells for the
most active imidazotriazinone 18 and for the parent compound
12 calculated from the dose-response curves obtained after 24,
48 and 72 h of incubation from 5-bromo-20-deoxyuridine (BrdU)
assay were determined. These are listed in Table 5.
In conclusion, in particular, five novel compounds based on the
7,8-dihydroimidazo[2,1-c][1,2,4]triazin-4(6H)-one template such
as 13, 15, 16, 18 and 22 demonstrate remarkable antiproliferative
activities and selectivities, and therefore these may be considered
as lead structures for the design of novel useful non-toxic (13, 15
and 16) and low toxic (18 and 22) antitumour agents.
with selective toxicity for cancer cells over normal HSF cells were
assembled. In view of continuous search for novel heterocycles with
selective cytotoxicity for cancer cells over normal cells, the synthetic
approach leading to the formation of novel 3-(2-furanyl)-8-aryl-7,
8-dihydroimidazo[2,1-c][1,2,4]triazin-4(6H)-ones, described in this
paper, might be considered as a useful method for the preparation of
these biologically active compounds because of the affordability of
the starting materials, straightforward product isolation and good
yields obtained. Furthermore, this efficient synthetic route allows
one to produce a large number of imidazotriazinone analogues.
The reversed-phase HPLC method was optimized and proved to be
applicable and reliable for the analysis of eleven novel compounds
based on the 7,8-dihydroimidazo[2,1-c][1,2,4]triazin-4(6H)-one
template (12–22). The high values of regression coefficients
(r >0.9841) proved the excellent fit between experimental data
and the equation: log k = log kw ꢀ Su. The obtained log kw values
can be regarded as hydrophobic parameters of the investigated sol-
utes and these were foundto be in the range of 2.48–4.01. The major-
ity of novel 3-(2-furanyl)-8-aryl-7,8-dihydroimidazo[2,1-c][1,
2,4]triazin-4(6H)-ones were found to be evidently effective
in vitro against cancerous cells (HeLa and T47D). Among them, com-
pounds 13, 15, 16, 18 and 22 demonstrate the remarkable antiprolif-
erative activity and selective cytotoxicity for cancer cells over
normal human skin fibroblast cells. Therefore these ones may be
considered as lead structures for the design of novel useful non-toxic
(13, 15 and 16) and low toxic (18 and 22) antitumour agents. The
drug-likeness of all the compounds investigated was assessed on
the basis of their structural properties by applying the Lipniski’s rule
of five. Fortunately, all the solutes investigated have all four param-
eters important for the favourable bioavailability.
2.5. Drug-likeness of the investigated 3-(2-furanyl)-8-aryl-7,8-
dihydroimidazo[2,1-c][1,2,4]triazin-4(6H)-ones (12–22)
Based on Lipinski’s rule of five,68 derived from an analysis of
drug-like compounds from the World Drug Index database,69 all
the reported herein unknown small molecules based on the 7,8-
dihydroimidazo[2,1-c][1,2,4]triazin-4(6H)-one template (12–22)
have all four parameters important for the favourable oral bioavail-
ability: the determined log kw values are less than 5, the molecular
weight of all the compounds is under 500 Da, all the solutes have
not more than 5 H-bond donors (expressed as the sum of NHs
and OHs) and not more than 10 H-bond acceptors (expressed as
the number of Ns and Os) (see Table 6), that would make them
likely to be orally active agents. It is worth noting that all eleven
solutes studied fulfil all the criteria described by rule of five. There-
by, the increased chances of good bioavailability in this unknown
set of compounds make them promising biologically active agents.
4. Experimental protocols
4.1. Instrumentations and general materials
Chemicals (2-oxo-2-furanacetic acid, triethylamine) were pur-
chased from Fluka as ‘for synthesis grade’ and used without further
purification. Solvents (n-butanol, N,N-dimethylformamide) were
purchased from E. Merck (Darmstadt, Germany). Melting points
(mp) were determined on a Boetius apparatus and given uncor-
3. Conclusion
In conclusion, by annulation of a 1,2,4-triazine ring to an imidaz-
olidine moiety according to [4+2] pattern novel antitumour agents

5114
K. Sztanke et al. / Bioorg. Med. Chem. 19 (2011) 5103–5116
rected. The purity of each compound was established by thin-layer
chromatography. TLC was carried out on commercial Merck SiO2
60 F254 plates having fluorescence indicator. The spots were visual-
ized in UV light k = 254 nm and 355 nm. 1H NMR and 13C NMR
spectra for compounds 12–22 were recorded on a Bruker Avance
300 MHz spectrometer in DMSO-d6 with TMS as an external stan-
dard at 295 K. The infrared (IR) spectra for compounds 12–22 were
determined in KBr disk using a Nicolet 8700 A FT-IR spectrometer
(Thermo Scientific).
(CH), 126.6 (CH), 127.6 (CH), 130.5 (CH), 135.6 (C), 136.2 (C),
139.6 (C-4), 143.4 (CH, furanyl C-5), 147.0 (C, furanyl C-2), 150.0
(C-3), 151.6 (C-8a).
4.1.2.3. 3-(2-Furanyl)-8-(4-methylphenyl)-7,8-dihydroimidazo
[2,1-c][1,2,4]triazin-4(6H)-one
(14). Recrystallization
from
DMF/MeOH (5:1) mixture; yield 64%, mp 322–323 °C. Anal. Calcd
for C16H14N4O2: C, 65.30; H, 4.79; N, 19.04. Found: C, 65.21; H,
4.80; N, 18.98. IR (KBr)
(m,
cmꢀ1): 3152.29–3012.16 (CH),
2956.38–2858.23 (CH3, 2CH2), 1687.08 (triazine –C@O), 1614.71,
1584.72, 1519.29, 1502.19, (aromatic skeleton), 1552.65 (C@N),
1479.40–1379.71 (2CH2, CH3), 1215.18 (furanyl C–O–C), 819.58
(1,4-disubstituted benzene ring); 1H NMR (d, ppm, DMSO-d6,
TMS, 300 MHz): 2.32 (s, 3H, CH3), 4.13–4.24 (m, 4H, 2CH2), 6.64
(dd, J4,5 = 1.8 Hz, J3,4 = 3.38 Hz, 1H, furanyl H-4), 7.26 (dd,
J = 8.5 Hz, 2H, phenyl H-20 and H-60), 7.34 (dd, J3,4 = 3.4 Hz,
J3,5 = 0.7 Hz, 1H, furanyl H-3), 7.75 (dd, J = 8.6 Hz, 2H phenyl H-30
and H-50), 7.84 (dd, J4,5 = 1.8 Hz, J3,5 = 0.8 Hz, 1H, furanyl H-5), 13C
NMR (d, ppm, DMSO-d6, TMS, 300 MHz): 19.8 (CH3), 39.9 (C-6,
CH2), 44.9 (C-7, CH2), 111.2 (CH, furanyl C-4), 112.9 (CH, furanyl
C-3), 118.8 (2CH), 128.8 (2CH), 132.6 (C), 136.0 (C), 140.1 (C-4),
143.7 (CH, furanyl C-5), 146.8 (C, furanyl C-2), 149.7 (C-3), 150.5
(C-8a).
4.1.1. Synthesis of 1-aryl-2-hydrazinoimidazoline (1-aryl-2-
hydrazonoimidazolidine) hydroiodides (1–11)
Compounds 2 and 4 have been synthesized and characterized,
as previously described.13 According to the same synthetic proce-
dure compound 7 has been prepared that spectral characterization
has been previously reported.3 Applying the same synthetic ap-
proach compounds 1, 3, 5, 6, 8–11 have also been obtained and
characterized as earlier described.7
4.1.2. General procedure for synthesis of 3-(2-furanyl)-8-aryl-
7,8-dihydroimidazo[2,1-c][1,2,4]triazin-4(6H)-ones (12–22)
A
mixture of 1-aryl-2-hydrazonoimidazolidine hydroiodide
(0.01 mol), 2-oxo-2-furanacetic acid (1.4 g, 0.01 mol) and triethyl-
amine (1.4 mL) in n-butanol (20 mL) was heated under reflux with
vigorously stirring for 15–55 minutes. After that time the precipi-
tate appeared, while the solution was still boiling. Then appropri-
ate amount of DMF was added until complete solid dissolution
was achieved and the liquid reaction solution was still refluxed
with stirring for 4–7 h. The reaction mixture was concentrated to
ca. two-thirds its original volume by evaporation under reduced
pressure. After cooling the mixture was refrigerated overnight,
and during that time precipitation of the solid started. Then the
formed crude product was filtered off, washed with cold methanol
and finally purified by recrystallization from DMF/methanol mix-
ture in the proportion indicated.
4.1.2.4. 3-(2-Furanyl)-8-(2,3-dimethylphenyl)-7,8-dihydroimi-
dazo[2,1-c][1,2,4]triazin-4(6H)-one (15). Recrystallization from
DMF/MeOH (3:1) mixture; yield 57%, mp 236–238 °C. Anal. Calcd
for C17H16N4O2: C, 66.22; H, 5.23; N, 18.17. Found: C, 66.13; H,
5.21; N, 18.23. IR (KBr)
(m,
cmꢀ1): 3143.50–3012.04 (CH),
2952.53–2849.75 (2CH3, 2CH2), 1675.95 (triazine –C@O),
1598.31, 1499.60 (aromatic skeleton), 1548.48 (C@N), 1475.03–
1384.67 (2CH2, CH3), 1221.54 (furanyl C–O–C), 779.04, 764.37,
692.55 (1,2,3-trisubstituted benzene ring); 1H NMR (d, ppm,
DMSO-d6, TMS, 300 MHz): 2.16 (s, 3H, CH3), 2.31 (s, 3H, CH3),
3.98–4.33 (m, 4H, 2CH2), 6.61 (dd, J4,5 = 1.8 Hz, J3,4 = 3.4 Hz, 1H,
furanyl H-4), 7.17–7.28 (m, 3H, phenyl H), 7.30 (dd, J3,4 = 3.4 Hz,
J3,5 = 0.7 Hz, 1H, furanyl H-3), 7.78 (dd, J4,5 = 1.8 Hz, J3,5 = 0.8 Hz,
1H, furanyl H-5); 13C NMR (d, ppm, DMSO-d6, TMS, 300 MHz)
13.5 (CH3), 19.2 (CH3), 40.9 (C-6, CH2), 47.6 (C-7, CH2), 111.1 (CH,
furanyl C-4), 112.2 (CH, furanyl C-3), 124.2 (CH), 125.7 (CH),
128.9 (CH), 134.3 (C), 136.3 (C), 137.5 (C), 139.6 (C-4), 143.3 (CH,
furanyl C-5), 147.1 (C, furanyl C-2), 150.0 (C-3), 151.8 (C-8a).
4.1.2.1. 3-(2-Furanyl)-8-phenyl-7,8-dihydroimidazo[2,1-c][1,2,4
]triazin-4(6H)-one (12). Recrystallization from DMF/MeOH (4:1)
mixture; yield 66%, mp 259–261 °C. Anal. Calcd for C15H12N4O2: C,
64.28; H, 4.32; N, 19.99. Found: C, 64.35; H, 4.30; N, 19.91. IR (KBr)
(m,
cmꢀ1): 3135.23–3007.32 (CH), 2972.65–2888.72 (2CH2),
1685.51 (triazine–C@O), 1602.25, 1584.28, 1504.76, 1494.32 (aro-
matic skeleton), 1552.57 (C@N), 1481.71–1453.94 (2CH2), 1219.17
(furanyl C–O–C), 889.67, 753.82, 688.13 (monosubstituted ben-
zene ring); 1H NMR (d, ppm, DMSO-d6, TMS, 300 MHz): 4.18–
4.25 (m, 4H, 2CH2), 6.65, 7.36, 7.85 (3H, furanyl H-4, H-3 and H-
5), 7.14–7.89 (m, 5H, phenyl H); 13C NMR (d, ppm, DMSO-d6,
TMS, 300 MHz): 40.0 (C-6, CH2), 44.8 (C-7, CH2), 111.2 (CH, furanyl
C-4), 113.1 (CH, furanyl C-3), 118.7 (CH), 123.2 (2CH), 128.3 (2CH),
138.5 (C), 140.4 (C-4), 143.8 (CH, furanyl C-5), 146.8 (C, furanyl C-
2), 149.7 (C-3), 150.5 (C-8a).
4.1.2.5. 3-(2-Furanyl)-8-(2-methoxyphenyl)-7,8-dihydroimidazo
[2,1-c][1,2,4]triazin-4(6H)-one
(16). Recrystallization
from
DMF/MeOH (2:1) mixture; yield 58%, mp 248–250 °C. Anal. Calcd
for C16H14N4O3: C, 61.93; H, 4.55; N, 18.06. Found: C, 61.80; H,
4.53; N, 18.12. IR (KBr)
(m,
cmꢀ1): 3145.41–2979.62 (CH),
2943.53–2834.73 (CH3O, 2CH2), 1679.60 (triazine –C@O),
1602.79, 1507.62 (aromatic skeleton), 1549.14 (C@N), 1464.15–
1435.41 (2CH2, CH3O), 1223.35 (furanyl C–O–C), 746.91 (1,2-
disubstituted benzene ring); 1H NMR (d, ppm, DMSO-d6, TMS,
300 MHz): 3.83 (s, 3H, OCH3), 4.00–4.30 (m, 4H, 2CH2), 6.62 (dd,
J4,5 = 1.8 Hz, J3,4 = 3.4 Hz, 1H, furanyl H-4), 7.01–7.52 (m, 4H, phe-
nyl H), 7.30 (dd, J3,4 = 3.4 Hz, J3,5 = 0.7 Hz, 1H, furanyl H-3), 7.79
(dd, J4,5 = 1.8 Hz, J3,5 = 0.8 Hz, 1H, furanyl H-5); 13C NMR (d, ppm,
DMSO-d6, TMS, 300 MHz) 40.9 (C-6, CH2), 46.7 (C-7, CH2), 55.6
(CH3O), 111.1 (CH, furanyl C-4), 112.3 (CH, furanyl C-3), 112.8
(CH), 120.3 (CH), 126.1 (C), 127.9 (CH), 128.5 (CH), 139.8 (C-4),
143.3 (CH, furanyl C-5), 147.1 (C, furanyl C-2), 149.9 (C-3), 152.0
(C-8a), 154.6 (C).
4.1.2.2. 3-(2-Furanyl)-8-(2-methylphenyl)-7,8-dihydroimidazo
[2,1-c][1,2,4]triazin-4(6H)-one
(13). Recrystallization
from
DMF/MeOH (4:1) mixture; yield 61%, mp 281–282 °C. Anal. Calcd
for C16H14N4O2: C, 65.30; H, 4.79; N, 19.04. Found: C, 65.40; H,
4.77; N, 19.11. IR (KBr)
(m,
cmꢀ1): 3148.40–2988.51 (CH),
2958.93–2853.69 (CH3, 2CH2), 1686.06 (triazine –C@O), 1604.43,
1499.69 (aromatic skeleton), 1552.47 (C@N), 1458.10–1380.52
(2CH2, CH3), 1218.84 (furanyl C–O–C), 761.69 (1,2-disubstituted
benzene ring); 1H NMR (d, ppm, DMSO-d6, TMS, 300 MHz): 2.28
(s, 3H, CH3), 4.02–4.33 (m, 4H, 2CH2), 6.62, 7.31, 7.79 (m, 3H, fura-
nyl H-4, H-3 and H-5), 7.34–7.46 (m, 4H, phenyl H); 13C NMR (d,
ppm, DMSO-d6, TMS, 300 MHz): 17.0 (CH3), 41.0 (C-6, CH2), 47.3
(C-7, CH2), 111.1 (CH, furanyl C-4), 112.3 (CH, furanyl C-3), 126.4
4.1.2.6. 3-(2-Furanyl)-8-(4-methoxyphenyl)-7,8-dihydroimidazo
[2,1-c][1,2,4]triazin-4(6H)-one
(17). Recrystallization
from
DMF/MeOH (5:1) mixture; yield 60%, mp 275–277 °C. Anal. Calcd
for C16H14N4O3: C, 61.93; H, 4.55; N, 18.06. Found: C, 61.85; H,

K. Sztanke et al. / Bioorg. Med. Chem. 19 (2011) 5103–5116
5115
4.57; N, 17.99. IR (KBr) (
m
, cmꢀ1): 3119.32–2985.17 (CH), 2934.32–
4.1.2.10. 3-(2-Furanyl)-8-(3,4-dichlorophenyl)-7,8-dihydroimi-
dazo[2,1-c][1,2,4]triazin-4(6H)-one (21). Recrystallization from
DMF/MeOH (8:1) mixture; yield 63%, mp 343–344 °C. Analysis
Calcd for C15H10Cl2N4O2: C, 51.60; H, 2.89; Cl, 20.31; N, 16.05.
2835.58 (CH3O, 2CH2), 1690.83 (triazine –C@O), 1614.86, 1589.27,
1513.05 (aromatic skeleton), 1556.07 (C@N), 1481.55–1438.09
(2CH2, CH3O), 1219.27 (furanyl C–O–C), 832.40 (1,4-disubstituted
benzene ring); 1H NMR (d, ppm, DMSO-d6, TMS, 300 MHz): 3.78
(s, 3H, OCH3), 4.12–4.25 (m, 4H, 2CH2), 6.64, 7.33, 7.83 (3H, furanyl
H-4, H-3 and H-5), 7.02 (dd, J = 9.1 Hz, 2H, phenyl H-20 and H-60,
7.75 (dd, J = 9.1 Hz, 2H, phenyl H-30 and H-50); 13C NMR (d, ppm,
DMSO-d6, TMS, 300 MHz) 40.0 (C-6, CH2), 45.4 (C-7, CH2), 55.0
(CH3O), 111.2 (CH, furanyl C-4), 112.7 (CH, furanyl C-3), 113.9
(2CH), 120.9 (2CH), 131.6 (C), 140.0 (C-4), 143.6 (CH, furanyl C-
5), 146.9 (C, furanyl C-2), 149.8 (C-3), 150.6 (C-8a), 155.7 (C).
Found: C, 51.46; H, 2.90; Cl, 20.23; N, 15.99. IR (KBr) (m
, cmꢀ1):
3145.86–2961.62 (CH), 2923.95–2852.49 (2CH2), 1690.99 (triazine
–C@O), 1595.01, 1570.83, 1497.49 (aromatic skeleton), 1551.49
(C@N), 1481.29–1395.47 (2CH2), 1219.64 (furanyl C–O–C),
1072.87 (aromatic C–Cl), 884.54, 838.02, 698.65 (1,2,4-trisubsti-
tuted benzene ring); 1H NMR (d, ppm, DMSO-d6, TMS, 300 MHz):
4.21 (s, 4H, 2CH2), 6.68 (dd, J4,5 = 1.8 Hz, J3,4 = 3.4 Hz, 1H, H-4 fura-
nyl), 7.40 (dd, J3,4 = 3.4 Hz, J3,5 = 0.8 Hz, 1H, H-3 furanyl), 7.89 (dd,
J4,5 = 1.8 Hz, J3,5 = 0.8 Hz, 1H, H-5 furanyl), 7.70–8.36 (m, 3H, phe-
nyl H); 13C NMR (d, ppm, DMSO-d6, TMS, 300 MHz) 40.0 (C-6, CH2),
44.7 (C-7, CH2), 111.3 (CH, furanyl C-4), 113.7 (CH, furanyl C-3),
118.3 (C), 120.0 (CH), 124.9 (CH), 130.1 (CH), 131.0 (C), 138.5 (C),
140.9 (C-4), 144.1 (CH, furanyl C-5), 146.6 (C, furanyl C-2), 149.5
(C-3), 150.2 (C-8a).
4.1.2.7. 3-(2-Furanyl)-8-(2-chlorophenyl)-7,8-dihydroimidazo
[2,1-c][1,2,4]triazin-4(6H)-one
DMF/MeOH (6:1) mixture; yield 59%, mp 286–288 °C. Anal. Calcd
(18). Recrystallization
from
for C15H11ClN4O2: C, 57.24; H, 3.52; Cl, 11.26; N, 17.80. Found: C,
57.15; H, 3.50; Cl, 11.31; N, 17.88. IR (KBr) (m
, cmꢀ1): 3148.12–
2993.92 (CH), 2962.16–2904.93 (2CH2), 1686.07 (triazine –C@O),
1591.45, 1499.86 (aromatic skeleton), 1552.54 (C@N), 1458.15–
1441.63 (2CH2), 1217.75 (furanyl C–O–C), 1080.57 (aromatic C–
Cl), 760.51 (1,2-disubstituted benzene ring); 1H NMR (d, ppm,
DMSO-d6, TMS, 300 MHz): 4.05–4.36 (m, 4H, 2CH2), 6.63, 7.33,
7.81 (3H, furanyl H-4, H-3 and H-5), 7.43–7.70 (m, 4H, phenyl H;
13C NMR (d, ppm, DMSO-d6, TMS, 300 MHz) 41.1 (C-6), 47.0 (C-
7), 111.2 (CH, furanyl C-4), 112.7 (CH, furanyl C-3), 127.8 (CH),
129.3 (CH), 129.4 (CH), 129.8 (CH), 131.2 (C), 135.0 (C), 140.2 (C-
4), 143.6 (CH, furanyl C-5), 146.8 (C, furanyl C-2), 149.8 (C-3),
151.7 (C-8a).
4.1.2.11. 3-(2-Furanyl)-8-(2,6-dichlorophenyl)-7,8-dihydroimi-
dazo[2,1-c][1,2,4]triazin-4(6H)-one (22). Recrystallization from
DMF/MeOH (4:1) mixture; yield 50%, mp 279–281 °C. Anal. Calcd
for C15H10Cl2N4O2: C, 51.60; H, 2.89; Cl, 20.31; N, 16.05. Found:
C, 51.68; H, 2.87; Cl, 20.26; N, 16.11. IR (KBr) (m
, cmꢀ1): 3154.64–
3016.26 (CH), 2981.97–2895.54 (2CH2), 1686.61 (triazine –C@O),
1590.05, 1504.37 (aromatic skeleton), 1552.58 (C@N), 1481.04–
1456.43 (2CH2), 1220.02 (furanyl C–O–C), 1102.40 (aromatic C–
Cl), 782.46, 758.15, 697.19 (1,2,3-trisubstituted benzene ring); 1H
NMR (d, ppm, DMSO-d6, TMS, 300 MHz): 4.06–4.35 (m, 4H,
2CH2), 6.64 (dd, J4,5 = 1.8 Hz, J3,4 = 3.4 Hz, 1H, furanyl H-4), 7.35
(dd, J3,4 = 3.4 Hz, J3,5 = 0.70 Hz, 1H, furanyl H-3), 7.51–7.85 (m,
3H: phenyl 3H); 7.82 (dd, J4,5 = 1.8 Hz, J3,5 = 0.7 Hz, 1H, furanyl H-
5);13C NMR (d, ppm, DMSO-d6, TMS, 300 MHz) 41.3 (C-6, CH2),
46.9 (C-7, CH2), 111.2 (CH, furanyl C-4), 113.0 (CH, furanyl C-3),
129.0 (CH), 129.2 (CH), 129.9 (C), 131.2 (CH), 131.7 (C), 136.4 (C),
140.5 (C-4), 143.7 (CH, furanyl C-5), 146.8 (C, furanyl C-2), 149.7
(C-3), 151.7 (C-8a).
4.1.2.8. 3-(2-Furanyl)-8-(3-chlorophenyl)-7,8-dihydroimidazo
[2,1-c][1,2,4]triazin-4(6H)-one
(19). Recrystallization
from
DMF/MeOH (4:1) mixture; yield 54%, mp 277–279 °C. Anal. Calcd
for C15H11ClN4O2: C, 57.24; H, 3.52; Cl, 11.26; N, 17.80. Found: C,
57.39; H, 3.48; Cl, 11.22; N, 17.73. IR (KBr) (m
, cmꢀ1): 3159.92–
3001.80 (CH), 2968.44–2899.90 (2CH2), 1694.03 (triazine –C@O),
1596.57, 1576.00, 1496.66 (aromatic skeleton), 1556.69 (C@N),
1445.80–1421.13 (2CH2), 1219.14 (furanyl C–O–C), 1096.49 (aro-
matic C–Cl), 883.90, 742.72, 698.37 (1,3-disubstituted benzene
ring); 1H NMR (d, ppm, DMSO-d6, TMS, 300 MHz): 4.21 (s, 4H,
2CH2), 6.66, 7.39, 7.87 (3H, furanyl H-4, H-3 and H-5), 7.19- 8.19
(m, 4H, phenyl H); 13C NMR (d, ppm, DMSO-d6, TMS, 300 MHz)
40.0 (C-6, CH2), 44.7 (C-7, CH2), 111.3 (CH, furanyl C-4), 113.5
(CH, furanyl C-3), 116.7 (CH), 118.2 (CH), 122.7 (CH), 129.9 (CH),
133.0 (C), 139.8 (C), 140.7 (C-4), 144.0 (CH, furanyl C-5), 146.6
(C, furanyl C-2), 149.6 (C-3), 150.3 (C-8a).
4.1.3. HPLC experimental
Novel 3-(2-furanyl)-8-aryl-7,8-dihydroimidazo[2,1-c][1,2,4]tri
azin-4(6H)-ones (12–22) were analyzed with methanol (from
30% to 80%)—20% acetate buffer pH 3.5—octane-1-sulfonic acid so-
dium salt (0.01 M Lꢀ1) (v/v/v) as mobile phases at 22 °C, using an
Agilent Technologies 1200 series chromatograph equipped with a
quaternary gradient pump with a degasser set at a flow rate of
1 mL/min and with a diode array detector (DAD). Eluates were
injected into the eluent with a Rheodyne 20
apparatus was equipped with ZORBAX Eclipse XDB-C18
150 mm ꢂ 4.6 mm column, 5- m particle size (Agilent Technolo-
lL injector. The HPLC
4.1.2.9. 3-(2-Furanyl)-8-(4-chlorophenyl)-7,8-dihydroimidazo
a
[2,1-c][1,2,4]triazin-4(6H)-one
(20). Recrystallization
from
l
DMF/MeOH (5:1) mixture; yield 60%, mp 319–320 °C. Anal. Calcd
gies, Wilmington, DE, USA). Methanol used as the mobile phase
modifier was of HPLC grade from C. Roth (Karlsruhe, Germany).
Octane-1-sulfonic acid sodium salt (OSA-Na) and acetate buffer
were from Merck (Darmstadt, Germany). Double distilled water
purified with the Millipore system was used. The pH of buffer used
in experiments, that is, 0.2 M acetate was measured for the aque-
ous solution.
for C15H11ClN4O2: C, 57.24; H, 3.52; Cl, 11.26; N, 17.80. Found: C,
57.44; H, 3.54; Cl, 11.29; N, 17.77. IR (KBr) (m
, cmꢀ1): 3157.09–
2996.01 (CH), 2926.80–2852.55 (2CH2), 1690.33 (triazine –C@O),
1581.33, 1497.17 (aromatic skeleton), 1551.61 (C@N), 1414.43–
1395.80 (2CH2), 1217.66 (furanyl C–O–C), 1091.23 (aromatic C–
Cl), 829.03 (1,4-disubstituted benzene ring); 1H NMR (d, ppm,
DMSO-d6, TMS, 300 MHz): 4.17–4.25 (m, 4H, 2CH2), 6.67 (dd,
J4,5 = 1.8 Hz, J3,4 = 3.4 Hz, 1H, H-4 furanyl), 7.38 (dd, J3,4 = 3.4 Hz,
J3,5 = 0.8 Hz, 1H, H-3 furanyl), 7.52 (dd, J = 9.1 Hz, 2H, ar.: H-20
and H-60), 7.88 (dd, J4,5 = 1.8 Hz, J3,5 = 0.8 Hz, 1H, H-5 furanyl),
7.93 (dd, J = 9.1 Hz, 2H ar: H-30 and H-50); 13C NMR (d, ppm,
DMSO-d6, TMS, 300 MHz) 40.0 (C-6, CH2), 44.8 (C-7, CH2), 111.3
(CH, furanyl C-4), 113.3 (CH, furanyl C-3), 120.2 (2CH), 127.2 (C),
128.2 (2CH), 137.4 (C), 140.6 (C-4), 143.9 (CH, furanyl C-5), 146.7
(C, furanyl C-2), 149.6 (C-3), 150.3 (C-8a).
4.1.4. Inhibition of tumour cell growth assay
All newly synthesized imidazotriazinones of the type 12–22,
were evaluated for their cytotoxic properties in vitro against two
human tumour cell lines: HeLa (ECACC 93021013, human Negroid
cervix epitheloid carcinoma cells) and T47D (ECACC 85102201, hu-
man breast carcinoma cells) in vitro. Furthermore, the primary cell
line of normal human skin fibroblasts was included in the cytotox-
icity study.

5116
K. Sztanke et al. / Bioorg. Med. Chem. 19 (2011) 5103–5116
In the current protocol each cell line was inoculated at 104 cells
per mL density on microtiter plates. Test agents were then added at
the examined concentration 50
g mLꢀ1 and cultures were incu-
bated at standard conditions (37 °C, 5% CO2, 90% humidity) for
24, 48 and 72 h. End-point determinations were made with 5-bro-
mo-20-deoxyuridine (BrdU) labeling and detection kit III70–73 on
Elisa reader (BIO-TEC Instruments USA).
26. Komsta, Ł.; Sztanke, K.; Ma˛czka, P.; Ucherek, M.; Skibin´ ski, R.; Gumieniczek, A.
J. Planar Chromatogr. 2009, 22, 327.
27. Giaginis, C.; Tsantili-Kakoulidou, A. J. Planar Chromatogr. 2009, 22, 217.
28. Kaliszan, R.; Nasal, A.; Markuszewski, M. J. Anal. Bioanal. Chem. 2003, 377,
803.
l
´
29. Bajda, M.; Bucki, A.; Szle˛k, J.; Szwaczkiewicz, M.; Swierczek, M.; Malawska, B.
Biomed. Chromatogr. 2008, 22, 428.
30. Pehourcq, F.; Matoga, M.; Jarry, C.; Bannawarth, B. J. Liq. Chromatogr. Related
Technol. 2001, 24, 2177.
31. Cimpan, G.; Hadaruga, M.; Miclaus, V. J. Chromatogr., A 2000, 869, 49.
32. Matysiak, J. Eur. J. Med. Chem. 2007, 42, 940.
33. Krauth, F.; Dahse, H.-M.; Rüttinger, H.-H.; Frohberg, P. Bioorg. Med. Chem. 2010,
18, 1816.
The growth percentage of normal and carcinoma cells was
evaluated spectrophotometrically versus untreated controls.
Results for each spectrophotometrical measure were noticed as
per cent of growth inhibition. All the investigated compounds
were dissolved in DMSO prior to dilution into the biological
assay. The solvent (DMSO) in the highest concentration used in
the bioassay did not reveal any cytotoxic activity. The obtained
results are presented as growth inhibition (GI) ratio of normal
and tumour cells by the examined compounds given in an
34. Brain-Isasi, S.; Quezada, C.; Pessoa, H.; Morello, A.; Kogan, M. J.; Alvarez-Lueje,
A. Bioorg. Med. Chem. 2008, 16, 7622.
´
35. Sztanke, K.; Markowski, W.; Swieboda, R.; Polak, B. Eur. J. Med. Chem. 2010, 45,
2644.
36. Sztanke, K.; Tuzimski, T.; Rzymowska, J.; Pasternak, K.; Kandefer-Szerszen´ , M.
Eur. J. Med. Chem. 2008, 43, 404.
37. Tuzimski, T.; Sztanke, K. J. Planar Chromatogr. 2005, 18, 274.
38. Sławik, T.; Kowalski, C. J. Chromatogr., A 2002, 952, 295.
39. Kaliszan, R. In Advances in Chromatography; Brown, P. R., Grushka, E., Eds.;
Dekker M.: New York, 1993; p 147.
40. Valko, K.; Plass, M.; Bevan, C.; Reynolds, D.; Abraham, M. H. J. Chromatogr., A
1998, 797, 41.
effective concentration of 50 l
g mLꢀ1 after 24-, 48- and
72-h periods of incubation. All experiments were done in
triplicates.
41. Valko, K. J. Chromatogr., A 2004, 1037, 299.
42. Gocan, S.; Cimpan, G.; Comer, J. Adv. Chromatogr. 2006, 44, 79.
43. Petruczynik, A. J. AOAC Int. 2011, 94, 77.
Supplementary data
44. Petruczynik, A.; Waksmundzka-Hajnos, M.; Hajnos, M. L. J. Chromatogr. Sci.
2005, 43, 183.
45. Flieger, J.; Swieboda, R. J. Chromatogr. A 2008, 1192, 218.
46. Ciolino, L. A.; Turner, J. A.; McCauley, H. A.; Smallwood, A. W.; Yi, T. Y. J.
Chromatogr. 1999, 852, 451.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2011.07.027.
´
47. Mroczek, T.; Glowniak, K.; Wlaszczyk, A. J. Chromatogr., A 2002, 949, 249.
48. Putzbach, K.; Rimmer, C. A.; Sharpless, K. E.; Sander, L. C. J. Chromatogr., A 2007,
1156, 304.
References and notes
1. Patrick, G. L. An Introduction to Medicinal Chemistry; Oxford, 2009. p. 519.
2. Denny, W. A. Eur. J. Med. Chem. 2001, 36, 577.
3. Sztanke, K.; Rzymowska, J.; Niemczyk, M.; Dybała, I.; Kozioł, A. E. Eur. J. Med.
Chem. 2006, 41, 539.
4. Sztanke, K.; Pasternak, K.; Rzymowska, J.; Sztanke, M.; Kandefer-Szerszen´ , M.
Eur. J. Med. Chem. 2008, 43, 1085.
49. Bidlingmeyer, B. A. J. Chromatogr. Sci. 1980, 18, 525.
´
50. Soczewinski, E.; Wachtmeister, C. A. J. Chromatogr. 1962, 7, 311.
51. Soczewin´ ski, E. Quantitative Retention-Eluent Composition Relationships in
Partition and Adsorption Chromatography. In A Century of Separation Science;
Issaq, H., Ed.; Dekker M.: New York, 2002; p 179. Chapter 11.
52. Snyder, L. R.; Dolan, J. W.; Gant, J. R. J. Chromatogr. 1979, 165, 3.
53. Hsieh, M. M.; Dorsey, J. G. J. Chromatogr. 1993, 631, 63.
54. Van de Waterbeemd, H.; Kansy, M.; Wagner, B. In Fischer in Drug Action and
Toxicology; Pliska, V., Testa, B., Van de Waterbeemd, H., Eds.; VCH, 1996; p 73.
55. Lambert, W. J. J. Chromatogr. 1993, 656, 469.
56. Carrupt, P. A.; Testa, B.; Gaillard, P. Computational Approaches to Lipophilicity:
Methods and Applications In Reviews in Computational Chemistry; Lipkowitz, K.
B., Boyd, D. B., Eds.; Wiley-VCH, Wiley J. and Sons: New York, 1997; Vol. 11,.
Chapter 5.
5. Sztanke, K.; Pasternak, K.; Sztanke, M.; Kandefer-Szerszen´ , M.; Kozioł, A. E.;
Dybała, I. Bioorg. Med. Chem. Lett. 2009, 19, 5095.
6. Litchfield, L. T.; Wilcoxon, F. J. Pharmacol. Exp. Ther. 1949, 96, 99.
7. Sztanke, K.; Rzymowska, J.; Niemczyk, M.; Dybała, I.; Kozioł, A. E. Eur. J. Med.
Chem. 2006, 41, 1373.
8. Sztanke, K.; Pasternak, K.; Rzymowska, J.; Sztanke, M.; Kandefer-Szerszen´ , M.
Bioorg. Med. Chem. 2007, 15, 2837.
9. Sztanke, K.; Pasternak, K.; Sidor-Wójtowicz, A.; Truchlinska, J.; Józwiak, K.
Bioorg. Med. Chem. 2006, 14, 3635.
10. Sztanke, K.; Pasternak, K.; Rajtar, B.; Sztanke, M.; Majek, M.; Polz-Dacewicz, M.
Bioorg. Med. Chem. 2007, 15, 5480.
11. Sztanke, K. PL Patent, Appl. No. 373877, 2005.
´
´
57. Fujita, T. Prog. Phys. Org. Chem. 1983, 14, 75.
58. Fujita, T.; Nishioka, T. Prog. Phys. Org. Chem. 1976, 12, 49.
59. Wermuth, C. G. In The practice of Medicinal Chemistry; Wermuth, C. G., Ed.;
Academic Press, 1996; p 333.
60. Hansch, C.; Maloney, P. P.; Fujita, T.; Muir, R. M. Nature 1962, 194, 178.
61. Fujita, T.; Iwasa, J.; Hansch, C. J. Am. Chem. Soc. 1964, 86, 5175.
62. Hansch, C.; Muir, R. M.; Fujita, T.; Maloney, P. P.; Geiger, F.; Streich, M. J. Am.
Chem. Soc. 1963, 85, 2817.
12. Sztanke, K.; Fidecka, S.; Ke˛dzierska, E.; Karczmarzyk, Z.; Pihlaja, K.; Matosiuk,
D. Eur. J. Med. Chem. 2005, 40, 127.
´
´
13. Tkaczynski, T.; Janocha, R.; Szacon, E.; Sztanke, K. Acta Pol. Pharm.-Drug Res.
1995, 52, 39.
14. Rusinov, V. L.; Ulomskii, E. N.; Chupakhin, O. N.; Charushkin, V. N. Russ. Chem.
Bull., Int. Ed. 2008, 57, 985.
15. Le Count, D. J.; Taylor, P. J. Tetrahedron 1975, 31, 433.
16. Hansch, C.; Leo, A. Substituent Constants for Correlation Analysis in Chemistry and
Biology; Willey Interscience: New York, 1979.
17. Van de Waterbeemd, H.; Testa, B. In Advances in Drug Research; Testa, B., Ed.;
Academic Press: London, 1987; Vol. 16, p 87.
18. Kulig, K.; Malawska, B. Biomed. Chromatogr. 2003, 17, 318.
19. Hollosy, F.; Seprödi, J.; Örfi, L.; Erös, D.; Keri, G.; Idei, M. J. Chromatogr., B 2002,
780, 355.
20. Gulyaeva, N.; Zaslavsky, A.; Lechner, P.; Chlenov, M.; McConnell, O.; Chait, A.;
Kipnis, V.; Zaslavsky, B. Eur. J. Med. Chem. 2003, 38, 391.
21. Liu, X.; Tanaka, H.; Yamauchi, A.; Testa, B.; Chuman, H. Helv. Chim. Acta 2004,
87, 2866.
22. Buur, A.; Trier, L.; Magnusson, C.; Artursson, P. Int. J. Pharm. 1996, 129, 223.
23. Artursson, P. J. Pharm. Sci. 1990, 79, 476.
63. Patrick, G. L. An Introduction to Medicinal Chemistry; Oxford University Press,
2009. p 381.
64. On-line software ALOGPS 2.1. (http://www.vcclab.org/).
65. Tetko, I. V.; Tanchuk, V. Y. J. Chem. Inf. Comput. Sci. 2002, 42, 1136.
66. Tetko, I. V.; Gasteiger, J.; Todeschini, R.; Mauri, A.; Livingstone, D.; Ertl, P.;
Palyulin, V. A.; Radchenko, E. V.; Zefirov, N. S.; Makarenko, A. S.; Tanchuk, V. Y.;
Prokopenko, V. V. J. Comput. Aid. Mol. Des. 2005, 19, 453.
67. Patrick, G. L. An Introduction to Medicinal Chemistry; Oxford University Press,
2009. p 553.
68. Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J. Adv. Drug Delivery Rev.
2001, 46, 3.
69. Patrick, G. L. An Introduction to Medicinal Chemistry; Oxford University Press,
2009. p 153.
70. Ellwart, J.; Dormer, P. Cytometry 1985, 6, 513.
71. Magaud, J. P.; Sargent, I.; Mason, D. Y. J. Immunol. Methods 1988, 106, 95.
72. Muir, D.; Varon, S.; Manthorpe, M. Anal. Biochem. 1990, 185, 377.
73. Huong, P. L. T. J. Immunol. Methods 1991, 140, 243.
24. Levin, V. A. J. Med. Chem. 1980, 23, 682.
25. Taylor, D. C.; Pownall, R.; Burke, W. J. Pharm. Pharmacol. 1985, 37, 280.