Synthesis and in vitro evaluation of 1,2,4-triazolo[1,5-a][1,3,5]triazine derivatives as thymidine phosphorylase inhibitors

Article abstract of DOI:10.1111/cbdd.12171

In our lead finding program, a series of 1,2,4-triazolo[1,5-a][1,3,5]triazine derivatives were synthesized, and their in vitro thymidine phosphorylase inhibitory potential was explored. Among the different derivatives, compounds having keto group (C = O) at C7 and thioketo group (C = S) at C5 positions showed varying degrees of TP inhibitory activity comparable with positive control, 7-deazaxanthine (7-DX, 2) (IC₅₀ value = 42.63 μm). Enzyme inhibition kinetics study suggested that compound IVn behaved as a mixed-type inhibitor of the enzyme with respect to thymidine (dThd) as a variable substrate. Compound IVn was also found to inhibit PMA-induced MMP-9 expression in MDA-MB-231 cells at sublethal concentrations. Computational docking study was performed to illustrate the enzyme inhibition kinetics and to explore the ligand-enzyme interactions.

Full text of DOI:10.1111/cbdd.12171

Chem Biol Drug Des 2013; 82: 351–360
Research Letter
Synthesis and in vitro Evaluation of 1,2,4-Triazolo[1,5-
a][1,3,5]triazine Derivatives as Thymidine
Phosphorylase Inhibitors
Hriday Bera1,2,*, Anton V. Dolzhenko2,3, Lingyi
respective bases (2). It also acts in the salvage pathway of
pyrimidine nucleotides by transferring the deoxyribosyl
Sun2, Sayan Dutta Gupta1 and Wai-Keung Chui2,*
moiety to form new nucleoside (3). TP is overexpressed in
1Gokaraju Rangaraju College of Pharmacy, Bachupally,
Hyderabad 500090, India
many solid tumors (4), including esophageal, gastric, pan-
2Department of Pharmacy, Faculty of Science, National
creatic, ovarian, breast, renal, bladder and non-small-cell
lung cancers (5,6). It is reported that this enzyme and its
metabolite, 2DDR-1P, stimulate the secretion and/or
expression of the angiogenic agents, including metallopro-
teases and other matrix-degrading enzymes (MMPs), and
induce endothelial cell migration (7–11). In addition,
increased level of TP is associated with protection of the
tumor against apoptosis (12,13) and enhancement of met-
astatic potential (14) that may influence tumor progression
(15). Therefore, inhibition of thymidine phosphorylase may
be viewed as a plausible strategy to overcome its patho-
logical effects. In addition, TP inhibitors may also serve as
adjuvant to enhance the bioavailability and therapeutic effi-
cacy of coadministered anticancer drugs that belong to
the deoxynucleosides (16).
University of Singapore, 18 Science Drive 4, Singapore
117543, Singapore
3School of Pharmacy, Monash University Sunway
Campus, Jalan Lagoon Selatan, Bandar Sunway,
Selangor, WA-46150, Malaysia
*Corresponding authors: Hriday Bera,
hriday.bera@gmail.com; Wai-Keung Chui,
phacwk@nus.edu.sg
In our lead finding program, a series of 1,2,4-triazolo
[1,5-a][1,3,5]triazine derivatives were synthesized, and
their in vitro thymidine phosphorylase inhibitory poten-
tial was explored. Among the different derivatives,
compounds having keto group (C = O) at C7 and thio-
keto group (C = S) at C5 positions showed varying
degrees of TP inhibitory activity comparable with posi-
tive control, 7-deazaxanthine (7-DX, 2) (IC50
value = 42.63 lM). Enzyme inhibition kinetics study
suggested that compound IVn behaved as a mixed-
type inhibitor of the enzyme with respect to thymidine
(dThd) as a variable substrate. Compound IVn was also
found to inhibit PMA-induced MMP-9 expression in
MDA-MB-231 cells at sublethal concentrations. Com-
putational docking study was performed to illustrate
the enzyme inhibition kinetics and to explore the
ligand–enzyme interactions.
Pioneering works in this field generated several potent thy-
midine phosphorylase inhibitors (6,17,18), most of these
inhibitors are derivatives of pyrimidine-2,4-dione with only
a few that are fused bicyclic heterocycles possessing the
homophthalimide moiety, which is believed to be essential
for binding to the active site of this enzyme (19). Fukushi-
ma et al. (20) synthesized 5-chloro-6-[1-(2-iminopyrrolidi-
nyl) methyl] uracil hydrochloride (TPI, 1) (Figure 1), the
most potent human TP inhibitor known so far, which has
an IC50 value of 35 nM and Ki of 20 nM.
Key words: 1,2,4-triazolo[1,5-a][1,3,5]triazine, computational
docking study, intramolecular heterocyclization, mixed-type
inhibition, thymidine phosphorylase inhibitors
Derivatives of 1,2,4-triazolo[1,5-a][1,3,5]triazine have
shown various bioactivities such as selective adenosine
receptor antagonism (21,22) and xanthine oxidase inhibi-
tion (23). Besides these activities, this scaffold also exhibits
anti-inflammatory, antioxidant, antiproliferative, antibacterial
and antifungal properties (24). However, inhibitory activity
of this fused heterobicyclic system against TP has not yet
been explored. It is hypothesized that structural modifica-
tions of 7-deazaxanthine (7-DX, 2) (Figure 1), a bicyclic
lead compound, by replacing two carbons at positions C9
and C5 with nitrogen moiety and inserting various substitu-
tions at position C6 would demonstrate TP inhibitory activ-
ity. To test this hypothesis, a series of 1,2,4-triazolo[1,5-a]
[1,3,5]triazines (Figure 2) were synthesized via a practical
Received 30 March 2013, revised 15 May 2013 and accepted
for publication 6 June 2013
Introduction
Thymidine phosphorylase (TP, EC 2.4.2.4), also known as
platelet-derived endothelial cell growth factor (PD-ECGF)
and gliostatin (1), catalyses the reversible phosphorolysis
of thymidine or 2’-deoxyuridine and their analogues into 2-
deoxy-alpha-D-ribose-1-phosphate (2DDR-1P) and the
ª 2013 John Wiley & Sons A/S. doi: 10.1111/cbdd.12171
351

Bera et al.
determine potential site of binding and possible interac-
tions of the synthesized compounds with the enzyme.
The synthesis of target compounds IVa-IVo and IIa-IIb
was achieved by intramolecular heterocyclization of N-
(1,2,4-triazol-5(3)-yl)-N’-carbethoxythioureas (4) and 5-
amino-1-carbethoxylthiocarbamoyl-1,2,4-triazole
(5),
Figure 1: Structures of known TP inhibitors.
respectively, in an alkaline medium as illustrated in
Scheme 1. The synthetic methods adopted here are
based on the chemistry developed by Bokaldere and
coworkers (26). This approach afforded products of high
yields and purity. Compounds IIIa-IIIb, corresponding 7-
methylthio derivatives of IIa-IIb, were prepared by treating
II with iodomethane in aqueous alkali. This protocol affor-
ded the products (IIIa-IIIb) with 61–67% yield. The result-
ing methylthio derivatives, IIIa-IIIb, were subsequently
converted to compounds Ia-Ib via reaction of IIIa-IIIb with
a mild oxidizing agent, hydrogen peroxide, in alkaline med-
ium, and this reaction produced highly pure products with
satisfactory yields (54–68%).
Figure 2: Structures of synthesized 1,2,4-triazolo[1,5-a][1,3,5]
triazine.
synthetic approach, and the target compounds were eval-
uated for TP inhibition using an in vitro enzyme assay. To
investigate the pharmacophoric requirements and explore
the chemical space of the active site of the enzyme, posi-
tions C5 and C7 of the triazolo-fused triazine scaffold were
modified to include a carbonyl, a thiocarbonyl and a thiom-
ethyl groups, while the position C2 was modified by insert-
ing different aromatic groups and longer flexible side
chains. Moreover, compound endowed with promising TP
inhibitory activity was tested to investigate its ability to sup-
press the MMP-9 expression, which has been reported to
be upregulated by TP in breast cancer cells (25). In addi-
tion, a computational docking study was performed to
All the synthesized compounds were characterized by
melting points and different spectroscopic techniques (1H
NMR, 13C NMR and MS). The purity of the compounds
was assessed by reverse-phase HPLC method and ele-
mental analysis.a Interestingly, the structures of com-
pounds I–IV were readily distinguished by the use of 13C
NMR spectroscopy. The 13C peak of the thiocarbonyl
(C = S) carbon of IVb appeared at around 175.8 ppm,
while this group showed peaks at about 170.8 ppm for
IIb. The appearance of the peak at about 11.8 ppm
assigned to SMe in the 13C NMR spectrum indicated the
formation of product IIIb. The two signals of the carbonyl
(C = O) groups of compound IIa appeared at about 152.5
Scheme 1: Synthesis of target compounds (I–IV). Reagents and conditions: (A) Ethoxycarbonyl isothiocyanate, acetone, rt, 20 min (B)
Ethoxycarbonyl isothiocyanate, DMF, rt, 4–5 h (C) NaOH, 80% ethanol (aq.), 100 °C, 20 min (D) MeI, NaOH, water, rt, 30 min (E) H2O2,
NaOH, water, 60–70 °C, 4 h.
352
Chem Biol Drug Des 2013; 82: 351–360

1,2,4-Triazolo[1,5-a][1,3,5]triazine Derivatives as TP Inhibitors
and 162.2 ppm in 13C NMR spectra. The purity of all com-
pounds was satisfactory (above 95%).
oxygen atom at C7 with the bioisosteric sulfur resulted in
no improvement in the activity. In addition, introduction of
thiomethyl group at C7 and consequently removal of one
active hydrogen at C6 in compounds IIIa and IIIb showed
no activity, just like in the case of IIa and IIb. Interestingly,
further modification of IIa and IIb by switching the posi-
tions of the carbonyl and thiocarbonyl moieties that affor-
The four groups of compounds (I, II, III and IV) were eval-
uated for in vitro TP inhibitory activity by a spectrophoto-
metric assay that used recombinant human thymidine
phosphorylase, expressed in E. coli (T2807
– Sigma
Aldrich). The original method developed by Krenitsky et al.
(27) was modified and adopted. In this assay, the
decrease in absorbance of thymidine (dThd) at 290 nm
due to enzymatic reaction was monitored after 4, 8, 12,
16 and 20 min. Preliminary screening of enzyme inhibition
was carried out at 100 lM of each compound. The inhibi-
tory potency of each compound was calculated using
GraphPad Prism version 4.0, and the results obtained
were expressed in terms of IC50 values and were com-
pared with that of 7-DX (Table 1).
ded
5-thioxo-5,6-dihydro-4H-[1,2,4]triazole[1,5-a][1,3,5]
triazin-7-one analogues (IVa and IVb) led to significant
improvement in inhibition profile. Therefore, from the preli-
minary screening of different derivatives, it was revealed
that a particular orientation of the C(=S)NHC(=O) moiety
would be required to confer the inhibitory potency. Addi-
tional variation in substitution at C2 of IVa and IVb was
carried out, and the resultant compounds IVa-IVo demon-
strated varying degrees of TP inhibition with IC50 values
ranging between 13 and 60 lM. The IC50 values of the test
compounds were compared with the positive control, 7-
DX, that exhibited IC50 value of 42.63 lM in the same
experimental conditions, which was found to be in accor-
dance with the previous report (5). Compound IVn, 2-(4-
bromo-3-methyl-phenyl)-5-thioxo-5,6-dihydro-4H-[1,2,4]
triazolo[1,5-a][1,3,5]triazin-7-one, was observed to have
best anti-TP activity (IC50 = 13.09 lM) among the four
groups of compounds evaluated in this study. Compounds
Unexpectedly, 1,2,4-triazolo[1,5-a][1,3,5]triazin-5,7-dione
analogues Ia and Ib having the homophthalimide moiety
did not display any inhibitory potential at 100 lM. This
result was not consistent with the previous report, indicat-
ing that the homophthalimide group was a pharmaco-
phoric requirement to impart inhibitory activity against TP
(19). Modification of these compounds by replacing the
Table 1: Thymidine phosphorylase inhibitory activity of the synthesized compounds (I-IV)
Anti-TP activitya
Entry
Compounds
R
IC50 (lM)b
1
Ia
SCH3
>100
2
3
Ib
IIa
Ph
SCH3
>100
>100
4
IIb
Ph
>100
5
6
7
8
IIIa
IIIb
IVa
IVb
IVc
IVd
IVe
IVf
SCH3
Ph
SCH3
Ph
>100
>100
51.67 Æ 5.92
39.56 Æ 1.78
55.38 Æ 4.19
59.88 Æ 4.79
59.25 Æ 3.25
43.32 Æ 5.03
31.54 Æ 5.15
30.95 Æ 4.20
24.53 Æ 4.91
31.58 Æ 3.10
22.06 Æ 4.42
34.42 Æ 3.21
31.78 Æ 3.56
13.09 Æ 1.85
58.13 Æ 2.61
42.63 Æ 5.25
9
SCH2Ph
SCH2CH2Ph
SCH2CH2CH2Ph
CH2Ph
2-furyl
2-thienyl
2-pyridyl
4-MeC6H4
4-BrC6H4
3-MeC6H4
3-BrC6H4
4-Br,3-MeC6H3
H
10
11
12
13
14
15
16
17
18
19
20
21
22
IVg
IVh
IVi
IVj
IVk
IVl
IVm
IVn
IVo
7-DX
–
aValues are means of three experiments.
bValues are presented as means Æ SD.
Chem Biol Drug Des 2013; 82: 351–360
353

Bera et al.
IVa and IVb were also modified by inserting longer flexible
side chain and higher spacer length at position C2; how-
ever, these efforts resulted in a decrease in activity, as evi-
dent in compounds IVc, IVd, IVe and IVf. In contrast,
replacement of phenyl ring with other aromatic rings like
furan, thiophene and pyridyl that yielded compounds IVg,
IVh and IVi led to an improvement in binding affinity. In
other words, the IC50 values of compounds IVg, IVh and
IVi were 31.54 lM, 30.95 lM and 24.53 lM, respectively,
that were lower to the IC50 value of IVb. The phenyl ring
was modified by inserting electron-withdrawing (Br) and
electron-donating (Me) substituents at both para- and
meta-positions. Although a little difference in binding affin-
ity was observed in metal-substituted compounds (IVl and
IVm), compound IVk (logP – 2.29) having bromo-group at
para-position displayed significantly higher activity com-
pared with compound IVj (logP – 2.03) carrying methyl
moiety. Therefore, it was evident that electronic effect of
substitutions might have played a pivotal role in binding
interactions. Moreover, it was believed that electron-with-
drawing (Br) group may impart higher hydrophobicity to
this compound, resulting in the enhancement of the TP
inhibitory activity. As shown in the Table 1, with few
exceptions, compounds (IV) with aromatic or aliphatic
substituents at position C2 exhibited better inhibition pro-
file than the unsubstituted compound IVo, which is con-
sidered as most structurally similar to the lead compound,
7-DX.
Based on the results of TP inhibition analysis, we selected
compound IVn for additional tests. To discriminate
between reversible and irreversible inhibition, TP in a 1-mL
reaction mixture (concentrations varying from 0.05 to
0.0075 U/mL) was treated with compound IVn at the con-
centration of 0 and 30 lM. The reaction mixture was incu-
bated for 30 min at room temperature to monitor the
conversion of dThd to thymine. The results of this experi-
ment suggested that compound IVn behaved as a revers-
ible inhibitor of the enzyme because two enzyme activity
curves converge to the intersection (zero) point of the abs-
issa/ordinate graph (Figure 3). Reversible inhibitors bind to
enzymes non-covalently, and the enzyme activity is recov-
ered immediately after removal of the inhibitors and thus
provide a potentially superior tolerability and safety profile
as compared to irreversible inhibitors. Balzarini et al. and
Hussain et al. (28,29) reported similar type of studies on
the inhibition of TP with some TP inhibitors, belonging to
diverse chemical classes. They also observed a similar
inhibition type.
Figure 3: Reversible inhibition of TP by IVn was determined by
incubating 0.05–0.0075 U/mL of TP in the presence of 30 lM of
compound IVn. Control experiments were carried out using
DMSO. Results are presented as means Æ SD; SD denoted by
error bars (Experiments carried out in triplicate).
Figure 4: Lineweaver-Burk plots of TP inhibition by IVn, in the
presence of variable concentrations of dThd demonstrating mixed
type enzyme inhibition. Results are presented as means Æ SD;
SD denoted by error bars (Experiments carried out in triplicate).
Figure 5: Response of kinetic parameters Km and Vmax with
increasing concentration of IVn in variable concentrations of
dThd. Results are presented as means Æ SD; SD denoted by
error bars (Experiments carried out in triplicate).
354
Chem Biol Drug Des 2013; 82: 351–360

1,2,4-Triazolo[1,5-a][1,3,5]triazine Derivatives as TP Inhibitors
Table 2: A summary of the kinetic parameters of IVn
Compound Substrate Ki value (lM)a,b
by the fact that increasing inhibitor concentration was
accompanied by an increase in Km values, whereas the
Vmax values decreased gradually (Figure 5). Therefore,
compound IVn can directly interact with enzyme as well
as enzyme–substrate complex, which inactivates the
enzyme but does not affect the binding of substrates. To
explain the mixed-type inhibition, we adopted the model of
Scheme 2. The inhibition constants (ki and aKi) for com-
pound IVn were evaluated from replotting the slopes and
intercepts of the double reciprocal plot against the inhibitor
concentrations (Table 2). The values of inhibition constants
(Ki and aKi) (Table 2) indicated that the inhibitor has stron-
ger affinity toward the free enzyme than enzyme–substrate
complex (a > 1). Moreover, the enzyme–inhibitor complex
would exhibit lower affinity for substrate compared with
free enzyme (a > 1) (Scheme 2). These results suggested
that compound IVn might interact with the enzyme differ-
ently compared with 7-DX, which behaved as a competi-
tive or mixed-type inhibitor in the presence of variable
concentrations of thymine (28).
a
Type of inhibition
IVn
dThd
24.21 Æ 0.53 2.48 Mixed type
aValues are means of three experiments.
bValues are presented as means Æ SD.
Scheme 2: Enzyme kinetic model for mixed type inhibition.
To investigate the ability of compound IVn to inhibit PMA-
induced MNP-9 expression, gelatine zymography (30) was
performed in MDA-MB-231 cells. The cells were treated
with increasing doses of compound (50–200 lM) in the
presence of PMA (80 nM). Interestingly, a decrease in band
intensity compared with vehicle control was detected for
compound IVn at 200 lM (Figure 6 A), indicating an atten-
uation of proteolytic activity of MMP-9 (band corresponded
to the MW of 92 kDa) at the 200 lM concentration. Quan-
tification of the band intensities in the zymograms using
Image Gauge 4.0 software conferred that compound IVn
significantly inhibited the MMP-9 activity at 100 and
200 lM concentrations (p < 0.05). The results also demon-
strated that a PMA-induced increase in MMP-9 expression
To elucidate the pattern of reversible inhibition, a brief
kinetic study was attempted using compound IVn at dif-
ferent inhibitor concentrations (0, 2, 15, 20, 30 lM). The
analysis was carried out by monitoring enzyme inhibition at
varying thymidine (dThd) concentrations (1000, 500, 300,
200, 100 lM) and maintaining phosphate (KPi) concentra-
tion constant at 25 mM. The data obtained were analyzed
by the double reciprocal plot method (Figure 4). The
results implied that compound IVn was a mixed-type
inhibitor in the presence of thymidine as a variable sub-
strate because the straight lines corresponding to different
concentrations of the inhibitor intersected in the second
quadrant of the reciprocal plot. This was further confirmed
A
B
Figure 6: Effect of compound
IVn and 7-DX on PMA-induced
MMP-9 expression and cell
growth in MDA-MB-231 cells: (A)
suppressive effects of IVn and 7-
DX on MMP-9 expression and
their dose response; (B)
C
densitometry analysis of MMP-9
expression. (C) cell viability (%) of
cells after 72 h treatment of
compound IVn and 7-DX. Results
are presented as means Æ SD;
SD denoted by error bars
(Experiments carried out in
triplicate). *p < 0.05.
Chem Biol Drug Des 2013; 82: 351–360
355

Bera et al.
A
B
D
F
C
E
Figure 7: Graphical
representation of interactions
between TP and its inhibitors
analyzed by computational
docking: (A) hydrogen bonding of
7-DX (colored ball and stick) with
Lys221, Thr118 (green colored
ball and stick); (B) hydrophobic
interaction of 7-DX (colored ball
and stick) with amino acid (orange
colored sphere); (C) hydrogen
bonding of IVb (colored ball and
stick) with Thr118 (green colored
ball and stick); (D) hydrophobic
interaction of IVb (colored ball
and stick) with amino acid (orange
colored sphere); (E) hydrogen
bonding of IVn (colored ball and
stick) with Lys221 and Thr118
(green colored ball and stick); (F)
hydrophobic interaction of IVn
(colored ball and stick) with amino
acid (orange colored sphere); (G)
an overlay of TPI and IVn bound
to the active site of TP (TPI, in
pink and IVn, in green).
G
356
Chem Biol Drug Des 2013; 82: 351–360

1,2,4-Triazolo[1,5-a][1,3,5]triazine Derivatives as TP Inhibitors
was suppressed dose dependently by the compound IVn
(Figure 6 B). However, reference inhibitor, 7-DX, did not
display any substantial inhibition of MMP-9 secretion even
at the 200 lM concentration.
closely matching with the reported crystallographic study
ꢀ
(33) with a RMS deviation of 0.5 A (Figure S1 a, b in
Appendix S1).
The analysis of molecular docking between TP and refer-
ence compound, 7-DX, suggested that the carbonyl group
at C4 and NH-7 of 7-DX formed strong hydrogen-bonding
contacts with Lys221 and Tyr118, respectively, located in
the active site of enzyme (33) (Figure 7 A). The NH group
adjacent to the thiocarbonyl moiety in identified inhibitor
(IVb) demonstrated hydrogen-bonding interaction with
Tyr118, whereas the thiocarbonyl moiety placed itself into
a hydrophobic space created by some amino acid resi-
dues like Ile-218, Ile-214, Val-208, Val-241, Val-208 and
Leu-148 (Figure 7 C, D). As a result, the compound IVb
exhibited inhibitory activity with an IC50 value in micromolar
range being comparable with 7-DX (Table 1). On the other
hand, compounds Ib, IIb and IIIb were found to be inac-
tive, although they displayed the higher total score com-
pared with IVb (Table 3). The compound Ib could not
produce tight hydrophobic contact into the hydrophobic
pocket as compared to compound IVb due to the pres-
ence of carbonyl moiety at C5 that has smaller atomic size
as well as high polarity, which led to misalignment of the
entire compound (Figure S1 c, d in Appendix S1). More-
To ascertain that the inhibition of MMP-9 expression by
the compound was not the consequence of its cytotoxic
effect, the inhibitory effect of the test compounds on
MDA-MB-231 cell growth was determined by means of
MTT assay (31). As depicted in Figure 5 C, compounds
IVn and 7-DX exhibited cell viability at around 80–100% at
the tested doses after 72 h of exposure as compared to
vehicle control (Figure 6 C).
To explore the plausible binding mode and molecular
interactions responsible for enzyme inhibition of the syn-
thesized compounds, geometry-optimized structure of
compounds was allowed to dock at the active site of TP
(PDB code: 2WK6) (32) using SYBYL-X 1.3. The lowest
energy conformation of the compounds was adopted to
analyze the docking results. Before docking, the docking
protocol was validated by comparing the binding interac-
tion of TPI in the active site of TP with the previously
reported results. The predicted binding mode and interac-
tion pattern of TPI generated by our software were
Table 3: Comparison of docking scores (kcal/mol) of compounds (I–IV)
Cpd
Total Scorea
Crash Scoreb
Polar Scorec
G Scored
PMF Scoree
D Scoref
Chem Scoreg
C Scoreh
Ia
4.43
4.75
4.57
3.98
4.61
3.75
2.93
3.46
3.96
3.46
3.04
4.48
2.17
2.41
2.85
3.48
2.39
4.24
3.09
3.19
3.73
4.98
À0.92
À1.22
À0.76
À2.71
À2.33
À3.63
À0.51
À1.07
À2.97
À3.65
À5.23
À1.45
À1.50
À1.64
À0.99
À1.99
À1.84
À1.85
À1.76
À1.87
À0.63
À0.66
1.14
1.72
1.11
1.18
1.61
1.57
1.10
0.83
1.36
0.76
1.02
0.64
0.28
0.34
0.72
0.64
2.03
0.79
0.62
0.85
1.13
2.16
96.08
101.44
96.08
95.31
88.66
87.70
86.51
83.74
102.97
89.55
97.43
72.01
54.70
70.15
80.92
84.70
58.17
82.98
70.74
83.45
98.63
108.26
135.53
143.96
122.07
120.39
138.93
136.40
122.02
93.10
134.32
137.87
134.68
110.27
73.54
À8.47
À13.14
À18.89
À20.01
À21.65
À22.57
À17.13
À15.30
À11.05
À9.87
-3.20
1
1
1
2
1
1
1
3
0
0
0
2
2
2
3
2
3
2
3
3
0
1
Ib
IIa
À6.38
À3.18
À6.28
À3.33
À4.72
À3.49
À8.22
À3.34
À3.42
À3.63
À7.27
À6.95
À6.12
À8.32
À5.99
À7.83
À8.12
À8.10
À5.91
À5.52
À8.25
IIb
IIIa
IIIb
IVa
IVb
IVc
IVd
IVe
IVf
À11.02
À14.03
À10.97
À11.25
À14.90
À14.62
À13.18
À14.17
À11.47
À15.50
À6.58
IVg
IVh
IVi
60.23
92.26
115.63
63.30
93.96
IVj
IVk
IVl
IVm
IVn
IVo
7-DX
95.68
102.24
126.32
142.09
À8.52
aTotal score reports total output of all the scores.
bCrash score indicating the ability of the compound to penetrate the active site of the protein.
cPolar score showing the polar interaction between the ligand and the protein.
dG score revealing hydrogen bonding, complex (ligand-protein) and internal (ligand–ligand) interaction.
eMF score demonstrates the free energies of interactions for protein–ligand atom pairs.
fD score is based on van der Waals interaction between protein and ligand.
gChem score includes terms for hydrogen bonding, metal–ligand interaction, lipophilic contact and rotational entropy, along with an inter-
cept term.
hC score: the consensus score: It exhibits the sum of the number of ‘good’ results for each ligand in each scoring function.
Chem Biol Drug Des 2013; 82: 351–360
357

Bera et al.
over, the energy required for hydrogen bonding, metal–
ligand interaction, lipophilic contact and rotational entropy
for IVb is less compared with Ib that is reflected in their G
score, PMF score, D score and Chem score (Table 3).
The compounds IIb and IIIb were having inappropriate
penetration (Crash score À2.71, À3.63 kcal/mol, respec-
tively) into the binding site, resulting in a decrease in inter-
action forces with the amino acids of binding site. A
similar trend was observed for compounds bearing thiom-
ethyl moiety (Ia, IIa, IIIa, IVa) (Table 3).
from these efforts and structure–activity relationships will
be reported in the due course.
Acknowledgments
The authors are thankful to the National Medical
Research Council grant, Singapore (R-148-000-102-275),
and Academic Research Funds of National University of
Singapore for their financial assistance. Thanks are also
extended to Gokaraju Rangaraju Educational Society for
providing essential laboratory facilities and support for this
project. There are no conflict of interests to be men-
tioned.
It was also evident from the docking score (Table 3) that
insertion of flexible chains into the lead structure (IVb)
resulted in inappropriate penetration and diminished
inhibitory potential, as found in compounds IVc, IVd, IVe
and IVf (Crash score À2.97, À3.65, À5.23, and
À1.45 kcal/mol, respectively). The total score, D score,
Chem score are in favor of compound IVi, and thus, it
exhibited higher inhibitory response (IC50 = 24.53 lM)
among all other heterocyclic derivatives (IVg, IVh and
IVi). The most active compound IVn showed an improve-
ment in interactions with respect to hydrogen bond for-
mation and hydrophobic contacts compared with
compound IVb (Figure 7 E,F). Moreover, compound IVn
exhibited better polar interaction, hydrogen bonding,
complex (ligand–protein) and internal (ligand–ligand) ener-
gies, free energies of interactions and van der Waals
interaction than positive control, 7-DX (Table 3). The anal-
ysis of molecular docking between TP and IVn sug-
gested that although compound IVn interacted with
amino acid residues located at or adjacent to the thymi-
dine binding site (33), the binding orientation of IVn was
different compared with TPI (competitive inhibitor), and
References
1. Asai K., Nakanishi K., Isobe I., Eksioglu Y.Z., Hirano A.,
Hama K., Miyamoto T., Kato T. (1992) Neurotrophic
action of gliostatin on cortical neurons. Identity of
gliostatin and platelet-derived endothelial cell growth
factor. J Biol Chem;267:20311–20316.
2. Iltzsch M.H., El Kouni M.H., Cha S. (1985) Kinetic stud-
ies of thymidine phosphorylase from mouse liver. Bio-
chemistry;24:6799–6807.
3. Schwartz M. (1971) Thymidine phosphorylase from
Escherichia coli. Properties and kinetics. Eur J Bio-
chem;21:191–198.
4. Takebayashi Y., Yamada K., Miyadera K., Sumizawa
T., Furukawa T., Kinoshita F., Aoki D., Okumura H.,
Yamada Y., Akiyama S., Aikou T. (1996) The activity
and expression of thymidine phosphorylase in human
solid tumours. Eur J Cancer; Pt A32:1227–1232.
5. Pocher F., Spadari S. (2001) Thymidine phosphorylase:
a two-face Janus in anticancer chemotherapy. Curr
Cancer Drug Targets;1:141–153.
consequently, it showed
a mixed-type inhibition with
respect to thymidine (Figure 7 G).
In conclusion, the intramolecular heterocyclization reaction
was used to generate the target 1,2,4-triazolo[1,5-a][1,3,5]
triazine derivatives. The in vitro evaluation of anti-TP effect
revealed that among the different derivatives of 1,2,4-triaz-
olo[1,5-a][1,3,5]triazine, compounds having thiocarbonyl
group and carbonyl group in positions C5 and C7, respec-
tively, viz., 5-thioxo-5,6-dihydro-4H-[1,2,4]triazolo[1,5-a]
[1,3,5]triazin-7-one and its analogues, showed inhibitory
activity against TP. The representative compound IVn was
found to exhibit mixed-type inhibition toward the enzyme
in the presence of variable concentrations of thymidine,
which was consistent with the results derived from the
molecular docking experiment. Compound IVn also inhib-
ited MMP-9 expression in MDA-MB-231 cells at sublethal
concentrations. Moreover, the docking study gave a justi-
fied binding mode of the active as well as inactive com-
pounds in the binding site of TP. Therefore, these
compounds would serve as leads for further design and
optimization of fused-ring TP inhibitors. Following these
promising results, further studies are in progress with
diverse structural modifications on that scaffolds for
improved biological activities, and the results obtained
6. Perez-Perez M.J., Priego E.M., Hernandez A.I., Camar-
asa M.J., Balzarini J., Liekens S. (2005) Thymidine
phosphorylase inhibitors: recent developments and
potential therapeutic applications. Mini Rev Med
Chem;5:1113–1123.
7. Brown N.S., Jones A., Fujiyama C., Harris A.L., Bic-
knell R. (2000) Thymidine phosphorylase induces carci-
noma cell oxidative stress and promotes secretion of
angiogenic factors. Cancer Res;60:6298–6302.
8. Gunningham S.P., Currie M.J., Morrin H.R., Tan E.Y.,
Turley H., Dachs G.U., Watson A.I., Frampton C., Rob-
inson B.A., Fox S.B. (2007) The angiogenic factor thy-
midine phosphorylase up-regulates the cell adhesion
molecule P-selectin in human vascular endothelial cells
and is associated with P-selectin expression in breast
cancers. J Pathol;212:335–344.
9. Muro H., Waguri-Nagaya Y., Mukofujiwara Y., Iwahashi
T., Otsuka T., Matsui N., Moriyama A., Asai K., Kato T.
(1999) Autocrine induction of gliostatin/platelet-derived
endothelial cell growth factor (GLS/PD-ECGF) and
GLS-induced expression of matrix metalloproteinases
358
Chem Biol Drug Des 2013; 82: 351–360

1,2,4-Triazolo[1,5-a][1,3,5]triazine Derivatives as TP Inhibitors
in rheumatoid arthritis synoviocytes. Rheumatology 38:
1195–1202.
21. Poucher S.M., Keddie J.R., Brooks R., Shaw G.R.,
McKillop D. (1996) Pharmacodynamics of ZM 241385,
a potent A(2a) adenosine receptor antagonist, after
enteric administration in rat, cat and dog. J Pharm
Pharmacol;48:601–606.
22. Keddie J.R., Poucher S.M., Shaw G.R., Brooks R.,
Collis M.G. (1996) In vivo characterisation of ZM
241385, a selective adenosine A2A receptor antago-
nist. Eur J Pharmacol;301:107–113.
10. Pula G., Mayr U., Evans C., Prokopi M., Vara D.S., Yin
X., Astroulakis Z., Xiao Q., Hill J., Xu Q., Mayr M.
(2009) Proteomics identifies thymidine phosphorylase
as a key regulator of the angiogenic potential of colony
forming units and endothelial progenitor cell cultures.
Circ Res;104:32–40.
11. Bijnsdorp I.V., Capriotti F., Kruyt F.A.E., Losekoot N.,
Fukushima M., Griffioen A.W., Thijssen V.L., Peters
G.J. (2011) Thymidine phosphorylase in cancer cells
stimulates human endothelial cell migration and inva-
sion by the secretion of angiogenic factors. Br J Can-
cer;104:1185–1192.
23. Robins R.K., Revankar G.R., O’Brien D.E. (1985) Pur-
ine analog inhibitors of xanthine oxidase - Structure
activity relationships and proposed binding of the
molybdenum
cofactor.
J
Heterocycl
Chem;22:601–634.
12. Ikeda R., Furukawa T., Mitsuo R., Noguchi T., Kitazono
M., Okumura H., Sumizawa T., et al. (2003) Thymidine
phosphorylase inhibits apoptosis induced by cisplatin.
Biochem Biophys Res Commun;301:358–363.
13. Kitazono M., Takebayashi Y., Ishitsuka K., Takao S.,
Tani A., Furukawa T., Miyadera K., Yamada Y., Aikou
T., Akiyama S. (1998) Prevention of hypoxia-induced
apoptosis by the angiogenic factor thymidine phos-
phorylase. Biochem Biophys Res Commun;253:797–
803.
14. Rofstad E.K., Halsor E.F. (2000) Vascular endothelial
growth factor, interleukin 8, platelet-derived endothelial
cell growth factor, and basic fibroblast growth factor
promote angiogenesis and metastasis in human mela-
noma xenografts. Cancer Res;60:4932–4938.
15. Bronckaers A., Gago F., Balzarini J., Liekens S. (2009)
The dual role of thymidine phosphorylase in cancer
24. Dolzhenko A.V., Dolzhenko A.V., Chui W.K. (2006)
1,2,4-triazolo[1,5-
synthesis and
cles;68:1723–1759.
a][1,3,5]triazines
biological
(5-azapurines):
Heterocy-
activity.
25. Kurizaki T., Toi M., Tominaga T. (1998) Relationship
between matrix metalloproteinase expression and
tumor angiogenesis in human breast carcinoma. Oncol
Rep;5:673–677.
26. Bokaldere R., Liepina A. (1973) s-Triazolo[1,5-a]-s-tri-
azinonethiones. Synthesis and reactions. Khim Geter-
otsikl Soedin;2:276–280.
27. Krenitsky T.A., Bushby S.R.M. (1979) U.S. Patent, 4,
178,212, 1-8, Research Triangle Park, NC: Burroughs
Welcome Co., 1979.
28. Balzarini J., Degreve B., Esteban-Gamboa A., Esnouf
R., De Clercq E., Engelborghs Y., Camarasa M.J.,
ꢂ
ꢂ
Perez-Perez M.J. (2000) Kinetic analysis of novel multi-
substrate analogue inhibitors of thymidine phosphory-
lase. FEBS Lett;483:181–185.
development
and
chemotherapy.
Med
Res
Rev;29:903–953.
16. Yano S., Kazuno H., Suzuki N., Emura T., Wierzba K.,
Yamashita J.I., Tada Y., Yamada Y., Fukushima M.,
Asao T. (2004) Synthesis and evaluation of 6-methy-
lene-bridged uracil derivatives. Part 1: discovery of
novel orally active inhibitors of human thymidine phos-
phorylase. Bioorg Med Chem;12:3431–3441.
29. Hussain S., Gaffney J., Ahmed N., Slevin M., Iqbal
Choudhary M., Ahmad V.U., Qasmi Z., Abbasi M.A.
(2009) An investigation of the kinetic and anti-angio-
genic properties of plant glycoside inhibitors of thymi-
dine phosphorylase. J Asian Nat Prod Res;11:159–
167.
17. Diab S.A., De Schutter C., Muzard M., Plantier-Royon
R., Pfund E., Lequeux T. (2012) Fluorophosphonylated
nucleoside derivatives as new series of thymidine
phosphorylase multisubstrate inhibitors. J Med Chem
55: 2758–2768.
30. Liotta L.A., Stetler-Stevenson W.G. (1990) Metallopro-
teinases and cancer invasion. Semin Cancer
Biol;1:99–106.
31. Alley M.C., Scudiero D.A., Monks A., Hursey Czerwin-
ski M.L.M.J., Fine D.L., Abbott B.J., Mayo J.G., Shoe-
maker R.H., Boyd M.R. (1988) Feasibility of drug
screening with panels of human tumor cell lines using
a microculture tetrazolium assay. Cancer Res;48:589–
601.
32. Mitsiki E., Papageorgiou A.C., Iyer S., Thiyagarajan N.,
Prior S.H., Sleep D., Finnis C., Acharya K.R. (2009)
Structures of native human thymidine phosphorylase
and in complex with 5- iodouracil. Biochem Biophys
Res Commun;386:666–670.
ꢁ
ꢁ
18. Pomeisl K., Horska K., Pohl R., Blazek J., Krecmerova M.
(2012) Syntheses of 1-[2 (phosphonomethoxy)alkyl]thy-
mine monophosphates and an evaluation of their inhibi-
tory activity toward human thymidine phosphorylase.
Nucleosides, Nucleotides Nucleic Acids;31:159–171.
19. Baker B.R., Kelley J.L. (1970) Irreversible enzyme
inhibitors. CLXXI. Inhibition of FUDR phosphorylase
from Walker 256 rat tumor by 5-substituted uracils. J
Med Chem;13:461–466.
20. Fukushima M., Suzuki N., Emura T., Yano S., Kazuno
H., Tada Y., Yamada Y., Asao T. (2000) Structure and
activity of specific inhibitors of thymidine phosphorylase
to potentiate the function of antitumor 2’-deoxyribonu-
cleosides. Biochem Pharmacol;59:1227–1236.
33. Norman R.A., Barry S.T., Bate M., Breed J., Colls
J.G., Ernill R.J., Luke R.W., Minshull C.A., McAlister
M.S., McCall E.J., McMiken H.H., Paterson D.S.,
Timms D., Tucker J.A., Pauptit R.A. (2004) Crystal
Structure of Human Thymidine Phosphorylase in
Chem Biol Drug Des 2013; 82: 351–360
359

Bera et al.
Complex with
ture;12:75–84.
a Small Molecule Inhibitor. Struc-
Supporting Information
Additional Supporting Information may be found in the
online version of this article:
Note
aAll final compounds were characterized by 1H NMR, 13C
NMR, ESI-MS and elemental analysis. Representative
examples, 2-phenyl-5-thioxo-5,6-dihydro-4H-[1,2,4]triazolo
[1,5-a][1,3,5]triazin-7-one (IVb). Yield 78%; mp 258–
259 °C (EtOH); ESI-MS m/z 244.1 (M-1)+; purity > 95%;
1H NMR (300 MHz, DMSO-d6): d 7.53–7.55 (m, 3H, H-3’,
H-4’ and H-5’), 8.04–8.07 (m, 2H, H-2’ and H-6’), 13.12
(s, 1H, NH), 14.31 (br. s, 1H, NH); 13C NMR (75 MHz,
DMSO-d6): d 127.1 (C-2’ and C-6’), 129.5 (C-3’ and C-
5’), 129.7 (C-4’), 131.3 (C-1’), 141.7 (C-2), 151.9 (C-9),
162.4 (C-7), 175.8 (C-5); Anal. calcd. for C10H7N5OS: C,
48.97; H, 2.88; N, 28.55. Found: C, 48.28; H, 2.71; N,
28.20.
Appendix S1. Full experimental detail, 1H and 13C NMR
spectra.
Figure S1. Molecular interactions of TPI and Ib with
enzyme, thymidine phosphorylase: (a) hydrogen bonding
of TPI (colored ball and stick) with Ser217, Arg202,
Lys221, Thr151 and His116 (green colored ball and stick);
(b) hydrophobic interaction of TPI (colored ball and stick)
with amino acid ( orange colored sphere); (c) hydrogen
bonding of Ib (colored ball and stick) with Ser217, Arg202
and Lys221 (green colored ball and stick); (d) hydrophobic
interaction of Ib (colored ball and stick) with amino acid
(orange colored sphere).
360
Chem Biol Drug Des 2013; 82: 351–360