Design, synthesis, molecular simulation, and biological activities of novel quinazolinone-pyrimidine hybrid derivatives as dipeptidyl peptidase-4 inhibitors and anticancer agents

Article abstract of DOI:10.1039/d0nj03774e

Two novel series of quinazolinone-pyrimidine (series a: 9a-9i) and benzyl-pyrimidine hybrids (series b: 12a-12c) were designed, synthesized and characterized by spectroscopic methods. The dipeptidyl peptidase-4 inhibition potencies of these compounds were assessed through a MAK 203 kit. Compound 9e was found to be the most potent agent with an IC50 value of 34.3 ± 3.3 μM. A kinetic study revealed that it acted as a competitive inhibitor. Molecular modeling of these compounds was in agreement with the in vitro results. Due to the crucial role of dipeptidyl peptidase-4 in cancer therapy, the cytotoxic activities of the compounds were also evaluated against three cancerous cell lines (HT-29, SW1116 and A549). Almost all the compounds displayed better antiproliferative activity on colon cancer cell lines (HT-29 and SW1116) compared to a lung cancer cell line (A549). Compounds 9e and 12c exhibited significant activity toward the HT-29 cell line with an IC50 of 10.67 ± 0.3 μM and 27.9 ± 6.5 μM in comparison to sitagliptin and cisplatin as a positive control, respectively. Among the different cells, the compounds showed the best inhibitory effects on HT-29, which was compatible with the greater expression of the dipeptidyl peptidase-4 marker detected by flow cytometry in this cell line. Further studies on the hit compounds (9e and 12c) through cell cycle and apoptosis assays also showed that these compounds could induce cell death by apoptosis or arrest cells in the G2/M phase. Accordingly, the results imply that 9e is a potent inhibitor of dipeptidyl peptidase-4 with efficient anti-cancer activity and could play a role as a cytotoxic agent in colorectal cancer.

Full text of DOI:10.1039/d0nj03774e

NJC
View Article Online
View Journal
PAPER
Design, synthesis, molecular simulation, and
biological activities of novel quinazolinone-
pyrimidine hybrid derivatives as dipeptidyl
peptidase-4 inhibitors and anticancer agents†
Cite this:DOI: 10.1039/d0nj03774e
Leila Emami,a Zahra Faghih,b Amirhossein Sakhteman,a Zahra Rezaei,ac
ac
Zeinab Faghih,c Farnaz Salehia and Soghra Khabnadideh
*
Two novel series of quinazolinone–pyrimidine (series a: 9a–9i) and benzyl-pyrimidine hybrids (series b:
12a–12c) were designed, synthesized and characterized by spectroscopic methods. The dipeptidyl
peptidase-4 inhibition potencies of these compounds were assessed through a MAK 203 kit. Compound
9e was found to be the most potent agent with an IC50 value of 34.3 ꢀ 3.3 mM. A kinetic study revealed
that it acted as a competitive inhibitor. Molecular modeling of these compounds was in agreement with
the in vitro results. Due to the crucial role of dipeptidyl peptidase-4 in cancer therapy, the cytotoxic
activities of the compounds were also evaluated against three cancerous cell lines (HT-29, SW1116 and
A549). Almost all the compounds displayed better antiproliferative activity on colon cancer cell lines
(HT-29 and SW1116) compared to a lung cancer cell line (A549). Compounds 9e and 12c exhibited
significant activity toward the HT-29 cell line with an IC50 of 10.67 ꢀ 0.3 mM and 27.9 ꢀ 6.5 mM in
comparison to sitagliptin and cisplatin as a positive control, respectively. Among the different cells, the
compounds showed the best inhibitory effects on HT-29, which was compatible with the greater
expression of the dipeptidyl peptidase-4 marker detected by flow cytometry in this cell line. Further
studies on the hit compounds (9e and 12c) through cell cycle and apoptosis assays also showed that
these compounds could induce cell death by apoptosis or arrest cells in the G2/M phase. Accordingly,
the results imply that 9e is a potent inhibitor of dipeptidyl peptidase-4 with efficient anti-cancer activity
and could play a role as a cytotoxic agent in colorectal cancer.
Received 27th July 2020,
Accepted 17th October 2020
DOI: 10.1039/d0nj03774e
rsc.li/njc
DPP-4 (CD26) inhibitors have been applied as oral hypoglycemic
drugs with excellent safety profiles and minimal side effects.
1. Introduction
Dipeptidyl peptidase-4 (DPP-4), also identified as Cluster of By inhibiting the DPP-4 transmembrane glycoprotein enzyme,
Differentiation 26 (CD26), is a glycoprotein with enzyme activity these drugs retain glucose homeostasis at the desired value with
expressed in different tissue and cell types. Its exopeptidase an increase in the release of insulin and a decrease in glucagon
function regulates the activity of a variety of substrates, for secretion and gluconeogenesis.4,5 Six FDA approved DPP-4 inhi-
instance cytokines, growth factors, neuropeptides, and incretin bitors including sitagliptin (2006), vildagliptin (2007), saxagliptin
hormones such as glucagon-like peptide-1 (GLP-1) and GLP-2.1–3 (2009), alogliptin (2010), linagliptin (2011) and teneligliptin (2012)
Consequently, DPP-4 (CD26) is currently considered as a target are available currently for DM-2 treatment. On the other hand,
of great importance in the treatment of various disorders in recently, the role of DPP-4 inhibitors in cancer treatment has been
particular diabetes mellitus type 2 (DM-2) and cancer. Recently, a topic of interest in the research area. They play a crucial role in
cancer as tumor suppressors or activators depending on their level
of expression and interaction with the microenvironment and
selected chemokines.6–8 On top of that it has been revealed that
DPP-4 inhibition in colorectal and lung cancers is associated with
improved overall survival of the patients, probably due to inter-
acting with immune cells through chemokines, leading to their
degradation.9 In a mouse model of hepatocellular carcinoma,
DPP-4 inhibitors could also clinically inhibit the progression of
cancer or augment the anti-tumor effects of molecular-targeting
a Department of Medicinal Chemistry, School of Pharmacy, Shiraz University of
Medical Sciences, Shiraz, Iran. E-mail: khabns@sums.ac.ir; Fax: +98-71-32424126;
Tel: +98-71-32424127-8
b Shiraz Institute for Cancer Research, Medical School,
Shiraz University of Medical Sciences, Shiraz, Iran
c Pharmaceutical Sciences Research Center, Shiraz University of Medical Sciences,
Shiraz, Iran
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
d0nj03774e
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
Paper
NJC
Fig. 1 Molecular hybridization design of 3-methyl-6-chloro-uracil-2-chloromethyl-3-substituted phenylquinazolin-4(3H)-one bearing 4-(amino-
methyl) piperidine as novel DPP-4 inhibitory agents.
drugs or immunotherapies.10 It has been also shown that inhibitory activity, we focused on two parameters: firstly, the
sitagliptin limits tumor growth by enhancing chemokine structures of alogliptin (I) (IC50 o 10 nM), linagliptin (II) (IC50
o
mediated eosinophil migration. DPP-4 (CD26) can also regulate 1 nM), and trelagliptin (III) (IC50 o 1 nM) as three main potent
the expression of matrix metalloproteinase, which is responsible and selective DPP-4 inhibitors on the market, and, secondly,
for destroying the extracellular matrix, a necessary process for various quinazolinone and pyrimidine scaffolds, which have
cancer attack and metastasis.11–13
been proved to be useful as DPP-4 inhibitory agents (structures
Quinazoline and quinazolinone are considered as favorable IV–IX) (Fig. 1).33–38 Accordingly, a pharmacophore approach was
and significant medical scaffolds that possess a variety of performed to obtain hybrid structure of aminomethyl
a
pharmacological properties including antidiabetic,14–17 anti- piperidine-pyrimidines with or without quinazolinone. Herein,
cancer,18–22 antifungal,23–25 antibacterial26–28 and antihypertensive in order to increase the DPP-4 inhibition potencies and to give
activity.29 Besides, there is a great deal of evidence showing more diversity, a 3-substituted phenyl moiety was also substi-
pyrimidine analogs as DPP-4 inhibitors.30–32 Thus, in the present tuted on the quinazolinone ring. This substitution was considered
study, for the design of new target compounds with DPP-4 to improve the interactions, i.e., electronic and hydrophobic, with
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
NJC
Paper
the enzyme. Furthermore, we replaced the (3R)-aminopiperidine 13C NMR spectra were recorded using a BRUKER DRX-AVANCE
group of structures V and IX (Fig. 1) with pyrimidine substituted 400 MHz and 100 MHz, respectively, in DMSO-d6. Mass spectra
with a 4-(aminomethyl) piperidine moiety, which is expected to were recorded on a mass instrument using the (M+1) mode from
interact with the DPP-4 enzyme active site through salt bridges Agilent (Agilent Technologies, USA).
as the critical hydrogen bonding interactions. The binding
2.1.1. General procedure for the synthesis of 2-(chloro-
mode of the novel designed compounds was then evaluated methyl)-4H-benzo[d][1,3]oxazin-4-one (3). To a solution of
with computational docking studies on the DPP-4 receptor anthranilic acid (1) (1 mmol) in dichloromethane (10 mL),
(PDB code: 4a5s). Molecular docking simulations were then diisopropylethylamine (DIPEA) (1.5 mmol) was added. Then
performed on two potent compounds (9e and 9f) to reach a full 1.2 mmol of chloroacetyl chloride (2) was added dropwise for
molecular binding mode interacting with the key amino acids 20 minutes at room temperature, and the reaction mixture was
in the active site of the DPP-4 enzyme in the steady state. stirred for 2 hours. The reaction mixture was washed with water
Pharmacological evaluations of the designed compounds as and extracted with ethyl acetate (2 ꢁ 20 mL) and the organic
DPP-4 inhibitors were also performed using a MAK 203 kit. layers dried with anhydrous Na2SO4. Then the remaining
Moreover, the cytotoxic effect of the new compounds with solvents were evaporated in a vacuum to obtain the product
DPP-4 inhibitory activities was assessed against human (Scheme 1, step a).
cancerous cell lines as well as normal cell lines. Additionally,
2.1.2. General procedure for the synthesis of 2-(chloro-
the induction of apoptosis and cell cycle analysis were done to methyl)-3-substituted quinazoline-4(3H)-one (5a–5i). Inter-
investigate the antiproliferative activity and mechanism of the mediate 3 was treated with different substituted anilines
cytotoxic effect on the colon cancer cell lines.
(4a–4i) (1 mmol) and PCl3 (1.5 mmol) in acetonitrile (CH3CN)
at 60 1C for 2 hours. After completion of the reaction, a
saturated NaHCO3 solution was added and extracted with ethyl
acetate (3 ꢁ 20 mL). The organic layers were dried with
anhydrous Na2SO4, and the crude products were purified by
recrystallization with ethanol (Scheme 1, step b).
2. Experimental
2.1. Chemistry
All chemicals and solvents were obtained from the Sigma
2.1.3. General procedure for the synthesis of 6-chloro-3-
Aldrich, Merck, and Samchun companies without any purification. substituted-1-((4-oxo-3-phenyl-3,4-dihydroquinazoline-2-yl) methyl)
Silica gel pre-coated analytical thin-layer chromatography (TLC) pyrimidine-2,4(1H,3H)-dione derivatives (7a–7i). 6-Chloro-3-methyl
plates were used for monitoring the reaction progress. Melting uracil (6) (2 mmol) was added to the different derivatives of
points were obtained on an Electrothermal 9200 apparatus intermediates 5a–5i (2 mmol) in acetonitrile with the presence of
(Electrothermal, UK), and infrared spectra were determined DIPEA (1.5 mmol). The mixture was stirred under reflux conditions
1
on a VERTEX70 spectrometer (Bruker, Germany). H NMR and for 24 hours. After completion of the reaction, the solvent was
Scheme 1 Synthesis of the amino-quinazolinone–pyrimidine hybrids (9a–9i). Reagents and conditions: (a) DIPEA, DCM, r.t., 2 h. (b) MeCN, PCl3, 60 1C,
2 h. (c) DIPEA, MeCN, reflux, 24 h. (d) i-PrOH, NaHCO3, 65 1C, 24 h.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
Paper
NJC
evaporated and purified by column chromatography using chloro- 125.63, 120.90, 79.13, 66.39, 47.11, 44.15, 38.515, 28.85, 26.40.
form/n-hexane (25/75) as an eluent (Scheme 1, step c). MS (m/z, %): 540.4 (M+, 3.2), 459.2 (1.1), 415.3 (2.2), 332 (6.97),
2.1.4. General procedure for the synthesis of 6-(4-(amino- 303.2 (54.35), 289.1 (38), 234.1 (10.5), 200 (10.38), 145.1 (100),
methyl) piperidine-1-yl)-3-substituted-1-((4-oxo-3-phenyl-3,4-di- 90.1 (62.4), 68.1 (50.39), 50.1 (32.01). Elem. anal. calcd. for
hydroquinazoline-2-yl) methyl) pyrimidine-2,4(1H,3H)-dione
C27H27F3N6O3 (540.5); C, 59.99; H, 5.03; N, 15.55. Found: C,
derivatives (9a–9i). 4-(Aminomethyl) piperidine (8) (0.55 mmol) 59.86; H, 5.06; N, 15.62.
was added to compounds 7a–7i (0.5 mmol) in isopropanol
(10 mL) and sodium bicarbonate (2.02 mmol) at room
2.1.4.3 6-(4-(Aminomethyl) piperidin-1-yl)-1-((3-(4-fluorophenyl)-
temperature. The reaction mixture was warmed to 65 1C and
heated for 24 hours. After completion of the reaction, the
solvent was removed and extracted with dichloromethane
(3 ꢁ 30 mL). Then the solvent was removed and recrystallized
from a mixture of dichloromethane and n-hexane to achieve the
final products (Scheme 1, step d).
4-oxo-3,4-dihydroquinazolin-2-yl) methyl)-3-methyl pyrimidine-2,4
(1H,3H)–dione (9c). White powder; yield 83%; m.p. 156–160 1C;
IR (KBr, cmꢂ1): 3447 (NH2), 3370 (NH2), 2925 (CH), 2854 (CH),
1698 (CQO), 1676 (CQO), 1609 (CQO), 1550 (CQN), 1471
(CQC), 1370 (C–N), 1224–1158 (C–O), 1114 (C–F). 1H NMR
(500 MHz, DMSO-d6) dH (ppm): 8.154 (d, 1H, J = 7.6 Hz, H-5-
ꢂ
ꢂ
quinazolinone), 7.899 (td, 1H, J = 8 Hz, J = 1.2 Hz, H-7-
2.1.4.1 6-(4-(Aminomethyl) piperidin-1-yl)-3-methyl-1-((4-oxo-
3-phenyl-3,4-dihydroquinazolin-2-yl) methyl) pyrimidine-2,4(1H,3H)-
dione (9a). Yellow solid; yield 69%; m.p. 184 1C; IR (KBr, cmꢂ1):
3431 (NH2), 3365 (NH2), 2924 (CH), 2853 (CH), 1686 (CQO), 1660
(CQO), 1611 (CQO), 1593 (CQN), 1473 (CQC), 1333 (C–N),
quinazolinone), 7.719 (d, 1H, J = 8 Hz, H-8-quinazolinone),
ꢂ
7.604 (t, 1H, J = 8 Hz, H-6-quinazolinone), 7.537–7.571 (m, 2H,
ꢂ
aromatic), 7.387 (t, 2H, J = 8.8 Hz, aromatic), 5.138 (s, 2H,
quinazolinone-CH2), 5.049 (s, 1H, uracil), 3.993 (d, 2H, J = 10.8
ꢂ
Hz, NH2), 3.134 (s, 3H, CH3-uracil), 2.654 (t, 2H, J = 12.4 Hz,
1
ꢂ
ꢂ
1271–1193 (C–O). H NMR (400 MHz, DMSO-d6) dH (ppm): 7.907
aliphatic), 2.333 (d, 2H, J = 6.4 Hz, piperidine-CH2-NH2), 1.545–
ꢂ
(d, 1H, J = 7.6 Hz, H-5-quinazolinone), 7.622–7.663 (m, 1H, H-7-
ꢂ
ꢂ
1.606 (m, 2H, aliphatic), 1.339–1.469 (m, 1H, aliphatic), 1.174–
1.30 (m, 1H, aliphatic), 0.835–0.883 (m, 3H, aliphatic), 13C NMR
(125 MHz, DMSO-d6) dC (ppm): 163.36, 162.45, 161.13, 158.89,
154.25, 151.03, 146.66, 134.95, 131.98, 130.77, 130.69, 127.52,
127.33, 126.42, 120.88, 116.46, 116.23, 79.11, 66.15, 47.21, 44.19,
38.69, 28.84, 26.56. MS (m/z, %): 490.4 (M+, 8.9), 473.4 (9.4), 461.3
(5.2), 376.3 (2.1), 361.3 (3.7), 280.2 (3.6), 269.1 (76.3), 254.2
(83.11), 239.2 (100), 212.2 (10.12), 143.2 (19.45), 119.2 (30.46),
95.1 (66.48). Elem. anal. calcd. for C26H27FN6O3 (490.5); C, 63.66;
H, 5.55; N, 17.13. Found: C, 63.61; H, 5.58; N, 17.15.
quinazolinone), 7.459 (d, 1H, J = 8 Hz, H-8-quinazolinone), 7.336–
ꢂ
7.368 (m, 1H, H-6-quinazolinone), 7.273–7.331 (m, 3H, aromatic),
7.229 (d, 2H, J = 6.4 Hz, aromatic), 4.867 (s, 2H, quinazolinone-
CH2), 4.813 (s, 1H, uracil), 3.742 (d, 2H, J = 14 Hz, NH2), 2.881 (s,
ꢂ
ꢂ
ꢂ
3H, CH3-uracil), 2.377–2.436 (m, 2H, aliphatic), 2.198 (d, 2H, J =
6.4 Hz, piperidine-CH2-NH2), 1.273–1.374 (m, 3H, aliphatic),
ꢂ
0.928–1.035 (m, 1H, aliphatic), 0.616–0.679 (m, 3H, aliphatic).
13C NMR (100 MHz, DMSO-d6) dC (ppm): 162.47, 160.99, 158.93,
154.35, 151.05, 146.69, 135.73, 134.91, 129.47, 129.34, 128.36,
127.43, 127.3, 126.42, 120.87, 79.25, 65.98, 45.08, 43.91, 36.71,
28.59, 26.56. MS (m/z, %): 472.5 (M+, 11.25), 442.5 (21.25), 393.5
(5.1), 358.5 (57), 149.1 (11.25), 112.2 (2.5), 80.0 (67.5), 71.1 (51.2),
2.1.4.4 6-(4-(Aminomethyl) piperidin-1-yl)-1-((3-(4-chlorophenyl)-
48.1 (100). Elem. anal. calcd. for C26H28N6O3 (472.5); C, 66.09; 4-oxo-3,4-dihydroquinazolin-2-yl) methyl)-3-methylpyrimidine-
H, 5.97; N, 17.78. Found: C, 66.02; H, 5.62; N, 17.71.
2,4(1H,3H)-dione (9d). Cream powder; yield 76%; m.p. 86–
91 1C; IR (KBr, cmꢂ1): 3445 (NH2), 3387 (NH2), 2922 (CH),
2.1.4.2 6-(4-(Aminomethyl) piperidin-1-yl)-3-methyl-1-((4-oxo- 2850 (CH), 1686 (CQO), 1607 (CQO), 1490 (CQN), 1473
3-(3-(trifluoromethyl) phenyl)-3,4-dihydroquinazolin-2-yl) methyl) (CQC), 1384 (C–N), 1271–1185 (C–O), 865 (C–Cl). 1H NMR
pyrimidine-2,4(1H,3H)-dione (9b). Cream powder; yield 78%; (400 MHz, DMSO-d6) dH (ppm): 8.201 (d, 1H, J = 8 Hz, H-5-
ꢂ
m.p. 173–176 1C; IR (KBr, cmꢂ1): 3442 (NH2), 3386 (NH2), quinazolinone), 7.924 (t, 1H, J = 8 Hz, H-7-quinazolinone),
ꢂ
2927 (CH), 2852 (CH), 1696 (CQO), 1661 (CQO), 1595 7.753 (s, 1H, aromatic), 7.732 (s, 1H, aromatic), 7.611–7.633
(CQO), 1548 (CQN), 1475 (CQC), 1308 (C–N), 1261–1170 (m, 4H, aromatic), 5.331 (s, 1H, uracil), 4.753 (s, 2H,
(C–O), 1071 (C–F). 1H NMR (400 MHz, DMSO-d6) dH (ppm): quinazolinone-CH2), 3.605 (d, 2H, J = 7.2 Hz, NH2), 3.148 (s,
ꢂ
ꢂ
8.168 (d, 1H, J = 8 Hz, H-5-quinazolinone), 7.999 (s, 1H, 3H, CH3-uracil), 2.468 (d, 2H, J = 6.4 Hz, piperidine-H-NH2),
ꢂ
ꢂ
ꢂ
aromatic), 7.924 (td, 1H, J = 8.4 Hz, J = 1.2 Hz, H-7- 1.718 (d, 2H, J = 11.6 Hz aliphatic), 1.480–1.589 (m, 1H,
ꢂ
quinazolinone), 7.864 (s, 2H, aromatic), 7.801 (d, 1H, J = aliphatic), 1.314–1.382 (m, 2H, aliphatic), 1.119–1.172
8 Hz, aromatic), 7.755 (d, 1H, J = 8 Hz, aromatic), 7.627 (t, (m, 3H, aliphatic), 0.916–0.958 (m, 1H, aliphatic). 13C NMR
1H, J = 7.6 Hz, aromatic), 5.071–5.228 (m, 2H, quinazolinone- (100 MHz, DMSO-d6) dC (ppm): 162.15, 161.06, 159.93, 152.05,
CH2), 5.033 (s, 1H, uracil), 3.983 (d, 2H, J = 11.2 Hz, NH2), 3.062 151.91, 146.63, 134.97, 134.68, 134.10, 130.23, 129.76, 127.09,
ꢂ
ꢂ
(s, 3H, CH3-uracil), 2.608–2.669 (m, 2H, aliphatic), 2.355 (d, 2H, 126.41, 121.04, 120.40, 88.66, 50.71, 47.13, 46.98, 38.21, 28.79,
ꢂ
J = 6.4 Hz, piperidine-CH2-NH2), 1.528–1.599 (m, 2H, aliphatic), 27.21. MS (m/z, %): 507.2 (M+, 6.1), 469.4 (44.7), 438.3 (3.5),
ꢂ
1.399–1.467 (m, 1H, aliphatic), 1.174–1.273 (m, 2H, aliphatic), 392.3 (79.45), 335.2 (56.18), 304.1 (42.3), 269.1 (50.9), 255.1
0.868–0.817 (m, 2H, aliphatic). 13C NMR (100 MHz, DMSO-d6) (100), 234.2 (14.44), 165.1 (33.77), 143.1 (39.17), 111.1 (44.15),
dC (ppm): 162.37, 161.11, 158.85, 154.04, 150.53, 146.64, 136.73, 75.1 (30.03). Elem. anal. calcd. for C26H27ClN6O3 (506.1); C,
135.06, 132.95, 130.70, 127.66, 127.39, 126.42, 126.12, 125.67, 61.6; H, 5.37; N, 16.58. Found: C, 61.9; H, 5.31; N, 16.64.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
NJC
Paper
2.1.4.5 6-(4-(Aminomethyl) piperidin-1-yl)-3-methyl-1-((4-oxo- (CQO), 1607 (CQO), 1549 (CQN), 1506 (CQC), 1332 (C–N),
3-(4-phenoxy phenyl)-3,4-dihydro quinazolin-2-yl) methyl) pyrimidine- 1275–1193 (C–O). 1H NMR (400 MHz, DMSO-d6) dH (ppm):
2,4(1H,3H)-dione (9e). Cream powder; yield 87%; m.p. 116– 8.158 (d, 1H, J = 7.2 Hz, H-5-quinazolinone), 7.904 (t, 1H, J =
119 1C; IR (KBr, cmꢂ1): 3441 (NH2), 3385 (NH2), 2922 (CH), 7.2 Hz, H-7-quinazolinone), 7.729 (d, 1H, J = 8 Hz, aromatic),
2850 (CH), 1665 (CQO), 1601 (CQO), 1504 (CQN), 1473 (CQC), 7.610 (t, 1H, J = 7.2 Hz, aromatic), 7.383–7.424 (m, 2H, aro-
1240 (C–N), 1185–1164 (C–O). 1H NMR (400 MHz, DMSO-d6) dH matic), 7.218 (d, 1H, J = 8 Hz, aromatic), 5.196 (s, 2H,
(ppm): 7.92 (dd, 1H, J = 8 Hz, J = 1.2 Hz, H-5-quinazolinone), quinazolinone-CH2), 5.044 (s, 1H, uracil), 4.001 (d, 2H, J =
ꢂ
7.652 (t, 1H, J = 8 Hz, H-7-quinazolinone), 7.477 (d, 1H, J = 8 Hz, 10.4 Hz, NH2), 3.126 (s, 3H, CH3-uracil), 2.654 (t, 2H, J =
ꢂ
ꢂ
H-8-quinazolinone), 7.36 (t, 1H, J = 7.2 Hz, H-6-quinazolinone), 12.4 Hz, aliphatic), 2.355 (d, 2H, J = 6 Hz, piperidine-CH2-
ꢂ
7.186–7.253 (m, 4H, aromatic), 6.973 (t, 1H, J = 7.2 Hz, NH2), 2.263 (s, 3H, CH3-aromatic), 1.556–1.730 (m, 3H,
ꢂ
aromatic), 6.861 (d, 2H, J = 8.8 Hz, aromatic), 6.809 (d, 2H, aliphatic), 1.401–1.424 (m, 1H, aliphatic), 0.785–1.05 (m, 3H,
J = 7.6 Hz, aromatic), 4.932 (s, 2H, quinazolinone-CH2), 4.860 aliphatic). 13C NMR (100 MHz, DMSO-d6) dC (ppm): 162.43,
ꢂ
(s, 1H, uracil), 3.779 (d, 2H, J = 11.2 Hz, NH2), 2.901 (s, 3H, 160.98, 159.21, 158.81, 154.11, 150.93, 146.62, 134.97, 131.96,
ꢂ
CH3-uracil), 2.419 (t, 2H, J = 12 Hz, aliphatic), 2.053 (d, 2H, 131.91, 127.59, 127.35, 126.42, 124.45, 120.92, 115.65, 115.41,
ꢂ
J = 6.4 Hz, piperidine-CH2-NH2), 1.337 (d, 2H, J = 11.2 Hz, 78.97, 66.12, 47.18, 44.07, 28.90, 28.85, 26.49, 13.97. MS (m/z,
ꢂ
aliphatic), 1.120–1.199 (m, 1H, aliphatic), 0.987 (d, 1H, J = 8 Hz, %): 564.4 (M+, 3.1), 346 (8.6), 302.1 (6.4), 267.2 (30.2), 253.2
aliphatic), 0.618–0.574 (m, 3H, aliphatic). 13C NMR (100 MHz, (48.3), 238.3 (43.9), 154.1 (35.8), 82.2 (31.08), 68.1 (100). Elem.
DMSO-d6) dC (ppm): 162.49, 161.17, 158.94, 157.38, 155.76, anal. calcd. for C27H29FN6O3 (504.5); C, 64.27; H, 5.79; N, 16.66.
154.24, 151.19, 146.65, 134.89, 130.49, 130.22, 130.15, 127.49, Found: C, 64.31; H, 5.72; N, 16.65.
127.32, 126.42, 124.15, 120.91, 119.19, 118.58, 79.17, 66.28,
47.19, 44.24, 38.679, 28.84, 26.59. MS (m/z, %): 564.4 (M+, 1.2),
2.1.4.8 6-(4-(Aminomethyl)
piperidin-1-yl)-1-((3-(3-chloro-4-
525.3 (3.1), 362.2 (8.3), 343.2 (48.65), 328.2 (100), 313.2 (52), fluoro phenyl)-4-oxo-3,4-dihydro quinazolin-2-yl) methyl)-3-methyl
237.2 (62.1), 234.1 (30.5), 143.1 (29.1), 115.1 (23.86), 77.1 (70.17). pyrimidine-2,4(1H,3H)-dione (9h). White powder; yield 73%; m.p.
Elem. anal. calcd. for C32H32N6O4 (564.65); C, 68.07; H, 5.71; N, 175–176 1C; IR (KBr, cmꢂ1): 3442 (NH2), 3377 (NH2), 3063–3003
14.88. Found: C, 68.13; H, 5.68; N, 14.89.
(CH), 2927–2853 (CH), 1699 (CQO), 1665 (CQO), 1607 (CQO),
1549 (CQN), 1497 (CQC), 1384 (C–N), 1241–1185 (C–O), 1110
1
2.1.4.6 6-(4-(Amino methyl) piperidin-1-yl)-1-((3-(3,4-dimethyl (C–F), 780 (C-Cl). H NMR (300 MHz, DMSO-d6) dH (ppm): 8.16
phenyl)-4-oxo-3,4-dihydro quinazolin-2-yl) methyl)-3-methyl pyrimidine- (d, 1H, J = 7.2 Hz, H-5-quinazolinone), 7.874–7.942 (m, 2H,
2,4 (1H,3H)-dione (9f). White powder; yield 87%; m.p. 178 1C; aromatic), 7.745 (d, 1H, J = 8.1 Hz, aromatic), 7.582–7.648
IR (KBr, cmꢂ1): 3446 (NH2), 3386 (NH2), 2922 (CH), 2852 (CH), (m, 3H, aromatic), 5.127–5.262 (m, 2H, quinazolinone-CH2),
ꢂ
1687 (CQO), 1661 (CQO), 1598 (CQO), 1548 (CQN), 5.052 (s, 1H, uracil), 4.015 (d, 2H, J = 11.7 Hz, NH2), 3.124
ꢂ
1457 (CQC), 1272 (C–N), 1272–1191 (C–O). 1H NMR (s, 3H, CH3-uracil), 2.662 (t, 2H, J = 12.3 Hz, aliphatic), 2.354
ꢂ
(300 MHz, DMSO-d6) dH (ppm): 8.137 (d, 1H, J = 7.5 Hz, H-5- (d, 2H, J = 6.3 Hz, piperidine-CH2-NH2), 1.565–1.679 (m, 2H,
ꢂ
quinazolinone), 7.872 (t, 1H, J = 7.2 Hz, H-7-quinazolinone), aliphatic), 1.378–1.436 (m, 1H, aliphatic), 1.243–1.298(m, 2H,
7.698 (d, 1H, J = 7.8 Hz, H-8-quinazolinone), 7.579 (t, 1H, J = aliphatic), 0.812–0.876 (m, 2H, aliphatic). 13C NMR (75 MHz,
7.2 Hz, H-6-quinazolinone), 7.260–7.285 (m, 1H, aromatic), DMSO-d6) dC (ppm): 162.42, 161.08, 158.82, 158.69, 156.22,
7.190–7.135 (m, 2H, aromatic), 5.285 (d, 1H, J = 13.8 Hz, uracil), 154.05, 150.63, 146.59, 135.04, 132.84, 132.81, 131.15, 127.40,
4.987–5.036 (m, 2H, quinazolinone-CH2), 3.974 (d, 2H, J = 126.42, 120.91, 117.64, 117.42, 79.09, 66.32, 47.31, 44.19, 38.81,
ꢂ
9.6 Hz, NH2), 3.126 (s, 3H, CH3-uracil), 2.60–2.674 (m, 2H, 30.94, 28.93. MS (m/z, %): 524.3 (M+, 1.95), 493.3 (2.26), 414.2
ꢂ
ꢂ
aliphatic), 2.389 (d, 2H, J = 5.4 Hz, piperidine-CH2-NH2), 2.237 (2.67), 322.1 (5.05), 303.1 (11.37), 273.1 (73), 237.2 (100), 184
ꢂ
(s, 3H, CH3-aromatic), 2.179 (s, 3H, CH3-aromatic), 1.472–1.594 (27.09), 129 (46.08), 68.1 (77.48). Elem. anal. calcd. for
ꢂ
ꢂ
(m, 3H, aliphatic), 1.186–1.270 (m, 1H, aliphatic), 0.866 (s, 3H, C26H26ClFN6O3 (524.98); C, 59.49; H, 4.99; N, 16.01. Found: C,
aliphatic). 13C NMR (75 MHz, DMSO-d6) dC (ppm): 162.99, 59.46; H, 4.52; N, 15.98.
161.55, 159.31, 154.68, 151.80, 147.17, 138.12, 138.06, 135.31,
133.69, 130.81, 129.37, 127.92, 127.79, 126.87, 125.84, 121.40,
2.1.4.9 6-(4-(Amino methyl) piperidin-1-yl)-1-((3-(5-chloro-2-
79.46, 66.52, 47.02, 44.45, 38.27, 29.30, 26.94, 19.66, 19.53. MS methoxy phenyl)-4-oxo-3,4-dihydro quinazolin-2-yl) methyl)-3-
(m/z, %): 369.3 (15.57), 249.2 (1.89), 196.2 (10.92), 173.2 (100), methyl pyrimidine-2,4(1H,3H)-dione (9i). White-cream powder;
103.1 (4.78), 77.2 (28.35), 51.2 (6.24). Elem. anal. calcd. for yield 89%; m.p. 95–100 1C; IR (KBr, cmꢂ1): 3433 (NH2), 3386
C
28H32N6O3 (500.6); C, 67.18; H, 6.44; N, 16.79. Found: C, (NH2), 3064–3006 (CH), 2924–2850 (CH), 1692 (CQO), 1664
67.21; H, 6.12; N, 16.82.
(CQO), 1608 (CQO), 1547 (CQN), 1473 (CQC), 1281 (C–N),
1249–1184 (C–O), 779 (C-Cl). 1H NMR (400 MHz, DMSO-d6)
2.1.4.7 6-(4-(Amino methyl) piperidin-1-yl)-1-((3-(3-fluoro-4- dH (ppm): 8.145 (d, 1H, J = 6.8 Hz, H-5-quinazolinone), 7.91
methyl phenyl)-4-oxo-3,4-dihydro quinazolin-2-yl) methyl)-3- (t, 1H, J = 7.6 Hz, aromatic), 7.715 (d, 1H, J = 8 Hz, aromatic),
methyl pyrimidine-2,4(1H,3H)-dione (9g). White powder; yield 7.680 (s, 1H, aromatic), 7.610 (t, 1H, J = 7.6 Hz, aromatic), 7.555
72%; m.p. 145–150 1C; IR (KBr, cmꢂ1): 3447 (NH2), 3386 (d, 1H, J = 7.2 Hz, aromatic), 7.262 (d, 1H, J = 8.8 Hz, aromatic),
(NH2), 3003–3037 (CH), 2927–2850 (CH), 1690 (CQO), 1661 5.099–5.227 (m, 2H, quinazolinone-CH2), 5.073 (s, 1H, uracil),
ꢂ
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
Paper
NJC
4.021 (d, 2H, J = 1.6 Hz, NH2), 3.71 (s, 3H, CH3-uracil), 3.091 134.43, 128.95, 128.84, 126.53, 126.29, 88.39, 50.82, 47.24,
ꢂ
ꢂ
(s, 3H, CH3-aromatic), 2.668 (t, 2H, J = 12 Hz, aliphatic), 2.403 46.63, 37.59, 28.91, 27.28, 20.62. MS (m/z, %): 342.2 (M+, 12),
ꢂ
(d, 2H, J = 5.2 Hz, piperidine-CH2-NH2), 1.533–1.669 (m, 3H, 237.1 (23), 220.1 (13), 194.1 (10), 163 (8), 113 (11), 105 (100), 96
ꢂ
aliphatic), 1.175–1.275 (m, 2H, aliphatic), 0.818–0.953 (m, 2H, (17), 79 (12). Elem. anal. calcd. for C19H26N4O2 (342.44); C,
aliphatic). 13C NMR (100 MHz, DMSO-d6) dC (ppm): 162.47, 66.64; H, 7.65; N, 16.36. Found: C, 66.64; H, 7.65; N, 16.36.
160.35, 158.97, 154.24, 153.61, 151.01, 146.63, 135.10, 130.74,
129.52, 127.65, 127.35, 126.45, 124.97, 124.14, 120.68, 113.86,
79.22, 66.09, 56.21, 46.27, 44.06, 37.46, 28.72, 26.42. MS (m/z,
%): 537.02 (M+, 1.6), 437.1 (1.5), 328.2 (3.6), 315.2 (20.3), 300.1
(27.5), 284.1 (30.3), 271.1 (46.07), 269.1 (100), 255.1 (23.5), 238.2
(14.8), 208.2 (9.5), 154.1 (16.8), 102.1 (19.7), 76.1 (45.5). Elem.
anal. calcd. for C27H29ClN6O4 (537.02); C, 60.39; H, 5.44; N,
15.65. Found: C, 60.42; H, 5.41; N, 15.63.
6-(4-(Aminomethyl) piperidin-1-yl)-1-(4-bromo benzyl)-3-methyl
pyrimidine-2,4(1H,3H)-dione (12c). White powder; yield 86%;
m.p. 93–96 1C; IR (KBr, cmꢂ1): 3423 (NH2), 3386 (NH2), 2920
(CH), 2849 (CH), 1699 (CQO), 1652 (CQO), 1542 (CQN), 1445
1
(CQC), 1377 (C–N), 680 (C-Br). H NMR (400 MHz, DMSO-d6)
dH (ppm): 7.294 (d, 2H, J = 8 Hz, aromatic), 6.948 (d, 2H, J = 8
Hz, aromatic), 5.016-5.092 (m, 1H, uracil), 4.685–4.718 (m, 2H,
2.1.5. General procedure for the synthesis of substituted
uracil-CH2), 2.891–2.853 (m, 5H, CH3-uracil, NH2), 2.642–2.745
ꢂ
ꢂ
1-benzyl-6-chloro-3-methyl
pyrimidine-2,4(1H,3H)-dione
(m, 1H, aromatic), 2.171–2.384 (m, 4H, aliphatic), 1.485–1.514
(m, 1H, aliphatic), 1.286–1.397 (m, 1H, aliphatic), 0.947–1.170
(m, 3H, aliphatic), 0.635–0.755 (m, 1H, aliphatic), 13C NMR
(100 MHz, DMSO-d6) dC (ppm): 162.13, 160.01, 152.17, 136.96,
131.29, 131.18, 128.81, 128.84, 120.08, 88.74, 50.68, 46.92,
45.62, 36.26, 28.73, 27.32. MS (m/z, %): 408.1 (M+, 12), 237.1
(75), 220.1 (42), 194.1 (34), 168.9 (100), 112.9 (28), 95.9 (73), 81.9
(53), 67.9 (26), 56 (28), 44.1 (39). Elem. anal. calcd. for
(11a–11c). To a stirred solution of different benzyl bromides
(10a–10c) (1 mmol) in THF (10 mL), 3-methyl-6-chlorouracil (6)
(0.46 mmol) and DIPEA (0.5 mmol) were added at room
temperature and left to stir for 6 hours at 40 1C and then
cooled to room temperature. Water (10 mL) was added and the
reaction mixture was stirred for 2 hours. The solids were
filtered, washed with isopropanol (3 ꢁ 40 mL), and dried at
60 1C to afford the products (11a–11c).
C
18H23BrN4O2 (407.31); C, 53.08; H, 5.69; N, 13.76. Found: C,
2.1.6. General procedure for the synthesis of 6-(4-(amino-
methyl) piperidine-1-yl)-substituted-1-benzyl-3-methyl pyrimidine-
2,4(1H,3H)-dione (12a–12c). The same as step d in series a (9a–9i).
53.08; H, 5.69; N, 13.76.
2.2. Docking procedure
The 3D X-ray crystal structure of the DPP-4 receptor (PDB code:
4a5s) was selected from the Protein Data Bank (http://www.
rcsb.org) based on the similarity of its co-crystal ligand to our
designed compounds. The structure of the synthesized com-
pounds was generated, minimized, and converted to pdbqt
format. For the preparation of the pdbqt format of the protein,
we removed the cognate ligand and water molecules, added
missing hydrogen atoms, and finally merged non-polar hydro-
gens according to their corresponding carbons. All preparation
was performed using the AutoDock Tools package (1.5.6). The
docking procedure was done in a grid box with a size of 30 ꢁ
30 ꢁ 30 and a center of x = 20.05, y = 33.12, z = 55.75 using
AutoDock Vina (1.1.2) using an in-house batch script
(DOCKFACE).39,40 The exhaustiveness was set to 100, and other
docking parameters were set as default. The binding interactions
of the docked compounds and the receptor were analyzed using
PLIP (fully automated protein–ligand interaction profiler).
6-(4-(Aminomethyl) piperidin-1-yl)-1-benzyl-3-methyl pyrimidine-
2,4(1H,3H)-dione (12a). Yellow powder; yield 75%; m.p. 107 1C;
IR (KBr, cmꢂ1): 3363 (NH2), 3343 (NH2), 3028–2935 (CH), 2911–
2824 (CH), 1706 (CQO), 1648 (CQO), 1601 (CQN), 1451 (CQC),
1345–1311 (C–N), 1H NMR (300 MHz, DMSO-d6) dH (ppm): 7.305
(d, 2H, J = 6 Hz, aromatic), 7.194–7.257 (m, 3H, J = 10 Hz,
aromatic), 5.293 (s, 1H, uracil), 4.971 (s, 2H, uracil-CH2), 3.111
ꢂ
(s, 5H, CH3-uracil, NH2), 2.823 (s, 1H, aliphatic), 2.507–2.567
ꢂ
(m, 2H, aliphatic), 2.438–2.449 (m, 2H, piperidine-CH2-NH2),
ꢂ
1.665–1.704 (m, 2H, aliphatic), 1.362–1.448 (m, 1H, aliphatic),
1.123–1.158 (m, 3H, aliphatic), 13C NMR (75 MHz, DMSO-d6) dC
(ppm): 162.66, 160.72, 152.71, 137.99, 128.99, 127.53, 126.95,
89.005, 51.34, 47.92, 47.29, 38.31, 29.39, 27.80. MS (m/z, %):
328.2 (M+, 15), 237.1 (31), 220.1 (22), 194.1 (14), 163.1 (16), 113.1
(17), 91 (100). Elem. anal. calcd. for C18H24N4O2 (328.4); C, 65.83;
H, 7.37; N, 17.06. Found: C, 65.83; H, 7.37; N, 17.06.
6-(4-(Aminomethyl) piperidin-1-yl)-3-methyl-1-(4-methyl benzyl)
pyrimidine-2,4(1H,3H)-dione (12b). Cream powder; yield 71%;
2.3. Molecular simulation
m.p. 95–98 1C; IR (KBr, cmꢂ1): 3363 (NH2), 3294 (NH2), 2926 Gromacs v5.0.4 was utilized as a platform to carry out MD
(CH), 2854 (CH), 1704 (CQO), 1646 (CQO), 1515 (CQN), 1446 simulation, and VMD v1.8.7 was employed as a visualization
(CQC), 1377 (C–N), 1H NMR (400 MHz, DMSO-d6) dH (ppm): tool. The protein structure file was downloaded from www.rcsb.
7.109–7.129 (m, 2H, aromatic), 7.066–7.086 (m, 2H, aromatic), org (PDB ID: 4a5s). The topology file for the ligand was
5.278 (s, 2H, uracil-CH2), 4.923 (s, 1H, uracil), 3.106–3.122 (m, generated using the PRODRG server, and the force field GRO-
ꢂ
5H, CH3-uracil, NH2), 2.571 (t, 2H, J = 10.8 Hz, aliphatic), 2.463 MOS96 53a6 was used during the MD simulation. The dimen-
ꢂ
(d, 2H, J = 6.4 Hz, piperidine-CH2-NH2), 2.257 (s, 3H, CH3), sions of the simulation box were defined as (12 12 12), and
ꢂ
ꢂ
1.708 (d, 2H, J = 11.6 Hz, piperidine-CH2-NH2), 1.374–1.410 solvent spc216 was added to the whole system. After that the
ꢂ
(m, 1H, aliphatic), 1.123–1.252 (m, 4H, aliphatic), 13C NMR system was neutralized, and the physiological concentration of
(100 MHz, DMSO-d6) dC (ppm): 162.16, 160.22, 152.19, 136.18, NaCl (0.15 M) was added to the system.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
NJC
Paper
The equilibration procedure contained 3 phases, including to the DPP-4 enzyme. The plots were drawn using a weighted
restraining the ligand (force 1000 in three dimensions), using least-squares analysis that assumed the variance of velocity (v)
the NVT ensemble for 100 ps (temperature coupling algorithm: to be a constant percentage of v for the entire data set. The
v-rescale; modified Berendsen thermostat) and finally incorpor- slopes of these Lineweaver–Burk plots were then drawn against
ating the NPT ensemble for 100 ps (pressure coupling algo- the concentration of 9e in the weighted analysis, and Ki
rithm: Berendsen).
was determined as the intercept on the negative X-axis. Data
Afterward, final MD for 100 ns was performed on each analysis was performed with GraphPad Prism 8.0 software
system (temperature coupling algorithm: v-rescale, and pres- (GraphPad Software Inc.).
sure coupling algorithm: Parrinello–Rahman). After that the
trajectories of the simulations were loaded into VMD for further
analyses and visualization of the binding mode.
2.6. Statistical analysis
The data were expressed as the mean ꢀ SD for each analysis.
GraphPad Prism software was used to perform one-way
ANOVA statistical analyses, followed by Tukey’s multiple
comparison test.
2.4. In vitro DPP-4 assay
The DPP-4 inhibitory activities of the designed compounds
were measured using a MAK 203 kit (Sigma-Aldrich, Germany),
which works based on cleaving the non-fluorescent substrate
(H-Gly-Pro-AMC) to generate fluorescent product 7-amino-4-
methyl coumarin (AMC). AMC can emit fluorescence with a
wavelength of 460 nm at an excitation wavelength of 360 nm.
According to the protocol, the assay was performed by mixing
25 mL of 4ꢁ compounds with various concentrations and 50 mL
of inhibition reaction mix (49 mL of DPP4 assay buffer with 1 mL
of DPP4 enzyme) in each well. A blank control (a well without
the enzyme) for all concentrations of compounds and an
enzymatic control (without a compound but with the same
volume) were also included. Then, the plate was incubated at
37 1C for 10 minutes. Afterward, 25 mL of the enzymatic reaction
mix containing 23 mL DPP4 assay buffer and DPP4 substrate
was added to each reaction well (tests, blank and enzymatic
control). Assays were performed in 96-well black plates using a
plate reader (POLARstar Omega) (Germany). The fluorescence
(FLU, lex = 360/lem = 460 nm) was measured on a microplate
reader in kinetic mode for 30 minutes at 37 1C. The slope
between the two times T1 and T2 in the linear range of the
fluorescence plot (DFLU/minute) was then obtained. The slope
of all tested compounds was subtracted from the relevant
sample blank to get the corrected measurement. The following
formula was used to measure the percentage of relative inhibi-
tion for each compound:
2.7. Biological assay
2.7.1. Cell lines and cell culture. Human colorectal carci-
nomas (HT-29 and SW1116), non-small cell lung carcinoma
(A549), and normal lung cells (MRC-5) were purchased from the
National Cell Bank of Iran (NCBI, Pasteur Institute, Tehran,
Iran). HT-29 was cultured in DMEM high glucose (Bio Idea,
Iran) in the presence of 2% L-glutamine (Gibco, USA). For
SW1116 and A549, RPMI 1640 medium (Bio Idea, Iran), and
for MRC-5, DMEM/Ham’s F12 (Bio Idea, Iran) were used.
All media were also supplemented with 10% fetal bovine serum
(FBS; Gibco, USA) and 1% penicillin–streptomycin (Biosera,
France) and kept at 37 1C in a humidified CO2 incubator.
The antiproliferative activity of all designed compounds
(9a–9i and 12a–12c) was determined by standard 3-(4,5dimethyl-
thiazol-yl)-2,5-diphenyl-tetrazolium bromide (MTT) assay according
to reported protocols.41 The cells were harvested from culture
using trypsin/EDTA 0.5% solution (Gibco/USA) and then were
seeded in 96-well microplates at a density of 1 ꢁ 104 cells per
well for the HT-29 and 15 ꢁ 104 cells per well for the MRC-5 cell
line and 8 ꢁ 103 cells per well for the A549 and SW1116 cell lines
in 100 mL of complete culture medium. After 24 h, each cell line
was treated with five different concentrations of the designed
compounds, and cisplatin and sitagliptin as positive controls
(5 to 500 mM) in a triplicate manner. We tried to maintain the
final DMSO concentration, the solvent, at less than 0.1% to avoid
its cytotoxic effect. Three untreated wells were considered as
negative controls. After 72 hours, the media were removed and
replaced with 100 mL fresh media containing 0.5 mg mLꢂ1 MTT
solution. They were incubated for 4 hours at 37 1C in the
incubator to create formazan purple crystals. The media were
then replaced with 150 mL of DMSO and incubated at 37 1C in the
dark for 10 minutes to dissolve the crystals. The absorbance of
individual wells was read at 490 nm using a microplate ELISA
reader. To estimate and analyze the data, Excel 2016 and Curve
Expert 1.4. were used. A plot of the percentage inhibition versus
ðSlope EC ꢂ Slope SMÞ
% Relative Inhibition ¼
ꢁ 100
Slope EC
Slope SM = Slope of sample inhibitor
Slope EC = Slope of enzyme control
The IC50 values, which show the concentration of the
compound with a 50% inhibitory effect on the enzyme, were
determined for all compounds using Curve Expert 1.4. software
and Excel 2016.
2.5. Kinetic study of DPP-4 inhibition
The kinetic mechanism of DPP-4 inhibition was determined by concentration was delineated and IC50, demonstrating the
Elman’s method in four different concentrations (100, 50, 10, concentration with 50% growth inhibition, was obtained. The
and 1 mM) of one of the most potent compounds, 9e. A double data were presented as the mean ꢀ SD for each analysis.
reciprocal plot was drawn as 1/[velocity] against 1/[substrate] at Statistical analyses were performed by one-way ANOVA and
different concentration ratios of the substrate (H-Gly-Pro-AMC) Tukey’s multiple comparison tests using Graph Pad Prism 8.0
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
Paper
NJC
software (GraphPad Software Inc.). Values were considered signi- steps is displayed in Scheme 1. In the first step, anthranilic
ficant when P o 0.05. acid (1) was converted to benzoxazine-4-one (3) with chloro
2.7.2. Surface and intracellular staining of DPP-4 (CD26). acetyl chloride (2) in the presence of a catalytic amount
For surface staining, the cells were harvested and aliquoted at of diisopropyl ethyl amine (DIPEA) in dichloromethane
a density of 25 ꢁ 104 cells in each tube and incubated with PE- (DCM) for 2 hours at room temperature. Next, a nucleophilic
conjugated anti-CD26 (Biolegend, USA) for 30 minutes in the attack occurred between 3 and nine different substituted
dark. The unstained tube was used as the negative control for anilines (4a–4i) to give 2-(chloromethyl)-3-aromatic substituted
each cell line. The cells were then washed twice and stained quinazoline-4(3H)-one derivatives 5a–5i under acidic condi-
with 2 mL of 7-AAD (BD Biosciences, USA) to distinguish viable tions at 60 1C with excellent yields (68–92%).44 The third step
from nonviable cells. For intracellular staining, the harvested of the reaction was done by replacing the chlorine atom at the
cells were first fixed with 1% paraformaldehyde (Sigma-Aldrich, side chain with 6-chloro-3-methyl-uracil (6) to achieve inter-
Germany), washed and then permeabilized with 1ꢁ BD mediates 7a–7i using optimized conditions. For this purpose,
Perm/Wash buffer (BD Biosciences) for 1 hour at 4 1C. The different polar protic (EtOH and MeOH) and aprotic (DMF,
permeabilized cells were then stained with PE-conjugated anti- MeCN, DCM, and THF) solvents were tested together with
CD26 (Biolegend) for 30 minutes. The cells were washed 2 times different temperatures, varied bases (K2CO3, NaH, DIPEA, and
with 1ꢁ BD Perm/Wash buffer, re-suspended in Phosphate Cs2CO3) and various time points. The optimized conditions
Buffered Saline (PBS 1ꢁ) buffer, and studied immediately on were found to be in acetonitrile (MeCN) as a solvent under basic
a FACSCalibur flow cytometer (BD Biosciences). The data were conditions of DIPEA and reflux for 24 hours. In the case of
analyzed with FlowJo software.
by-product creation, the compound was purified on a chroma-
2.7.3. Apoptosis analysis. To describe the ability of com- tography plate, using silica gel and 25% chloroform in n-hexane
pound 9e and 12c to induce apoptosis in the HT-29 cell line, a as an eluent. To achieve the final target compounds (9a–9i),
PE Annexin V apoptosis detection kit with 7-AAD (Biolegend) 4-(aminomethyl) piperidine (8) was reacted with intermediates
was used. According to a previous protocol,42,43 100 ꢁ 103 cells 7a–7i through nucleophilic substitution under a moderate
per 500 mM of complete culture medium were seeded in a basic catalyst (K2CO3) and iso-propyl alcohol as a solvent in
24-well culture plate. After 24 hours, three different concentra- high yields (69–89%). The chemical structures of all com-
tions of 9e (6, 12, 24 mM) and 12c (15, 30, 60 mM) were used to pounds were confirmed by IR, 1H-NMR, 13C-NMR, and mass
treat the cells for a period of 24 and 72 hours. An untreated spectroscopies. In IR analysis, the stretching frequencyy of N–H
sample was considered as a negative control. Then, the treated bond in primary amine was observed at 3363–3447 cmꢂ1
and non-treated cells were trypsinized and washed two times and 3387–3294 cmꢂ1. The significant feature of the 1H NMR
with cold PBS 1ꢁ. The pellets were dissolved in 50 mL binding spectrum of compounds 9a–9i was a doublet peak at 3.605–
buffer and transferred to polystyrene tubes (BD Biosciences, 4.021 ppm related to the NH2 proton. The two protons of CH2
USA), stained with 2 mL of PE-conjugated AV and 2 mL of 7-AAD which are between quinazolinone and the uracil ring appeared
and incubated at room temperature for 15 minutes in the dark. as a singlet for all compounds except 9b, 9f, 9h, and 9i. The
Finally, 300 mL binding buffer was added and the sample was peaks appearing as a doublet of doublets can be attributed to
directly studied using a four-color FACSCalibur flow cytometer the existence of CF3, 3,4-di-Me, 3Cl-4F, and 5Cl-2OMe groups
(BD Biosciences). The data were analyzed with FlowJo software. on the phenyl ring, which can lead to nonequivalent protons in
2.7.4. Cell cycle analysis. 100 ꢁ 103 HT-29 cells per 500 mM the compounds. The proton peak of uracil was indicated as a
were seeded in a 24-well plate and treated with three different singlet at 4.813–5.33 ppm for all compounds except 9f. The
concentrations of 9e (6, 12, 24 mM) and 12c (15, 30, 60 mM) for important piece of the 13C NMR spectrum of the compounds
24 and 72 hours. Then the cells were harvested and washed was a singlet peak related to C5 of the uracil scaffold, which
with cold PBS 1ꢁ, stabilized slowly in cold 70% ethanol, and appeared in the range of 78.97–88.74 ppm depending on the
incubated at 4 1C for 7 days. The fixed cells were washed twice other substitutions. The carbon of the carbonyl group in the
in PBS 1X and centrifuged at 400 ꢁ g at 4 1C. Afterward, the cells quinazoline ring was displayed in the range of 162.15–
were resuspended with 50 mL ribonuclease A (100 mg mLꢂ1), 162.99 ppm.
Sigma-Aldrich, Germany) to remove RNA. Finally, 200 mL propi-
dium iodide (PI, 50 mg mLꢂ1 solution) (Sigma-Aldrich, Germany)
3.2. Synthesis and Characterization of the amino-benzyl-
pyrimidine hybrids (12a–12c)
was added to stain DNA. The cells were analyzed on a four-color
FACSCalibur flow cytometer (BD Biosciences) with the appropriate The benzyl-pyrimidine hybrids (12a–12c) (series b) were pre-
settings. Finally, the data were analyzed with FlowJo software.
pared by known literature methods.44,45 Firstly, 6-chloro-3-
methyl-uracil (6) was reacted with three substituted benzyl
bromides (10a–10c) in THF using DIPEA to achieve intermedi-
ates 11a–11c. Then, 4-(aminomethyl) piperidine (8) was reacted
with 11a–11c under the optimized conditions described for step
3 in series a to get the final compounds (Scheme 2). The
3. Results and discussion
3.1. Synthesis and Characterization of the
amino-quinazolinone–pyrimidine hybrids (9a–9i)
The synthetic route for the preparation of the quinazoline- chemical structures of all compounds were confirmed by IR,
pyrimidine hybrids (9a–9i) (series a) through four different 1H-NMR, 13C-NMR, and mass spectroscopies. In IR analysis, the
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
NJC
Paper
compound 9e can be summarized by pi–pi interactions with
0
Trp 629 located in the S2 site and the interaction of the
carbonyl group at the C-4 position of the quinazolinone ring
with Ser 630 of the catalytic triad in the S2 pocket. The
4-phenoxy substituent occupying the S1 subsite is aligned with
the residues Tyr 662, Tyr 666, and Tyr 631 and might take a role
in its enhanced DPP-4 activity compared to the other synthe-
sized compounds. There are also hydrogen bond and pi–cation
interactions through the carbonyl group and the N-3 position of
uracil with Lys 554.
Scheme 2 Synthesis of the amino-benzyl-pyrimidine hybrids (12a–12c).
Reagents and conditions: (e) DIPEA, THF, 40 1C, 6 h. (f) 4-(Aminomethyl)
piperidine, i-PrOH, NaHCO3, MeCN, 65 1C, 24 h.
In compound 12c, the 4-(aminomethyl) piperidine group
formed salt bridges with Glu 206 and the 4-bromobenzyl group
filled the S1 pocket of the enzyme with Tyr 666 and Tyr 662.
Furthermore, the carbonyl group at position 3 of uracil formed
hydrogen bond interactions with Ser 630 of the catalytic triad
and Arg 125 in the S2 pocket. To explain the binding mode of
sitagliptin, in brief, it should be announced that the three
fluorophenyl moieties of sitagliptin were occupying the S1
pocket via Tyr 662 and Tyr 666. In addition, a hydrogen bond
between the NH group and His 740, Arg 125, and Ser 630 was
observed (Fig. 2).
stretch bond of the primary amine was observed at 3363–
3423 cmꢂ1 and 3386–3294 cmꢂ1. The two protons of CH2 which
are between aryl and the uracil ring appeared as a singlet in the
range of 4.72–4.97 ppm. The proton peak of position 5 in uracil
appeared as a singlet except in the case of 12c. The significant
features of the 13C NMR spectrum were a singlet peak related
to C5 of the uracil scaffold, which appeared in the range of
88.39–89.00 ppm, and the peak of the carbonyl group at the
uracil moiety, observed in the range of 162.13–162.65 ppm.
3.3. Molecular docking study
Several methods were introduced to verify the accuracy of
Molecular docking was performed to delve into the interac- the docking results,47,48 including a re-docking simulation of
tions, free binding energy, and molecular binding mode of the the co-crystal ligand (N7F) in the active site of DPP4, which
synthesized compounds against the DPP-4 enzyme. The range revealed a root mean square deviation (RMSD) of 0.4 Å.
of free binding energy values was observed between –7.5 (10b) Meanwhile, the receiver operating characteristic (ROC) and
and –9.3 (6e), as displayed in Table 2. The binding mode of 9a, enrichment factor (EFmax)49 were also studied. Firstly, a bench-
9e, 12c, and sitagliptin in the active site of the DPP-4 enzyme is mark set of compounds with assay data against DPP-4 was
provided in Fig. 2. As depicted in Fig. 2, the phenyl ring of the retrieved from the ChEMBL database in the SMILES format.50
quinazolinone moiety in compound 9a binds to the S2
The compounds were then converted to 3D mol2 by DOCK-
0
subsite46 via pi–pi interactions with Trp 629. Also, N-1 of the FACE, in-house batch script software. The ligands were classi-
quinazolinone ring interacts through a hydrogen bond with Ser fied according to their experimental data (IC50) as active and
630, and some hydrophobic interactions with the residues Tyr inactive (decoy). The resulting ligands were subjected to docking
547, Lys 554, and Tyr 631 were observed. The binding mode for simulations as described for the co-crystal ligand. Finally, the
binding energy of the best pose for each ligand was used to
calculate the two metrics of the area under the curve of the ROC
(AUC-ROC) curves and the enrichment factor (EFmax).
The plots and results of ROC and EFmax are shown in Fig. 3.
The AUC-ROC of 0.922 and EFmax of 3.2 verified that the
docking scores could distinguish between active and inactive
structures.
3.4. A molecular dynamics simulation study
Molecular dynamics simulations (MD) for the 9e and 9f com-
pounds were performed to cover the space of moderate and
high affinities against the DPP-4 enzyme. The protein–ligand
complexes and individual root mean square deviations
(RMSDs) of compounds 9e and 9f were achieved after 100 ns
MD simulation, as seen in Fig. 4a. The plateau at the terminal
part of the curves for the two compounds showed that the
system is in the equilibrated state and has stability during the
simulation.
According to Fig. 4b (the plot of the RMSD), a more stable
pattern was observed for compound 9e concerning 9f. The
energy values for ligands 9e and 9f at different time levels
Fig. 2 Interactions of 9a, 9e, 12c and sitagliptin with the residues in the
binding site of the DPP-4 receptor (4a5s).
during the 100 ns simulation are displayed in Table 1. The van
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
Paper
NJC
Fig. 3 Enrichment factor (left) and ROC (right) diagrams for the DPP-4 (4a5s) receptor.
Fig. 4 (a) Root mean square deviation (RMSD) values of 4a5s-9e and 4a5s-9f as red and black, respectively, at different time levels during molecular
dynamics (MD) simulation studies, and (b) RMSD values of the 9e (red) and 9f (black) ligands.
Table 1 Different types of intramolecular forces of the 9e and 9f complexes through MD simulation
van der Waals energy
Electrostatic energy
Polar solvation energy
SASA energy
Binding energy
Entry
(kJ molꢂ1
)
(kJ molꢂ1
)
(kJ molꢂ1
)
(kJ molꢂ1
)
(kJ molꢂ1
)
9e
9f
ꢂ167.690
ꢂ115.754
ꢂ74.239
ꢂ35.990
125.773
75.484
ꢂ26.715
ꢂ12.604
ꢂ132.872
ꢂ88.864
3.5. Determination of DPP-4 inhibition and structure–activity
relationship studies
der Waals energy for compounds 9e and 9f was ꢂ167.690
and ꢂ115.754, respectively. This difference in energy levels
indicates a more specific molecular shape for 9e in comparison
with 9f. The same pattern can be observed while comparing
other components of intramolecular forces in 9e compared to
9f. This result is suggestive of the more stable complex of 9e in
the active site of the enzyme.
According to the fluctuations of the amino acids displayed
in the RMSF plot (Fig. 5), the domains covering residues 200–
210, 550–700 and 910–920 were found to be the parts of the
active site with the highest fluctuation during the simulation
(Fig. 5b). In summary, both protein–ligand complexes have
been stabilized under NPT conditions; however, 9e has a more
favorable effect on the DPP-4 enzyme.
The DPP-4 enzymatic activity of the synthesized compounds
(9a–9i and 12a–12c) was assayed by cleavage of H-Gly-
Pro-amino methyl coumarin (H-Gly-Pro-AMC) as a florigenic
substrate. The percentages of DPP-4 enzyme inhibition at
concentrations of 100 mM and IC50 for each compound and
sitagliptin (as a standard drug) are presented in Table 2.
Generally, six derivatives (9c, 9d, 9e, 9g, 12a, and 12c) had
more than 50% inhibition at 100 mM. 9e with an IC50 value of
34.3 ꢀ 3.3 mM was found to be the most active hit compound.
The structure–activity relationship for compounds 9a–9i indi-
cated that substitution of 4-phenoxy on ring A significantly
increased the DPP-4 inhibitory activity. One possible reason for
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
NJC
Paper
Fig. 5 (a) Potential energies of 6e-4a5s and 6f-4a5s interactions during 10ꢂ5 ps MD simulation. (b) The root mean square fluctuation (RMSF) of the 9e
and 9f complexes.
this observation could be the better accommodation of the and an electron-releasing group at position 3 and 4 of ring A,
4-phenoxy phenyl ring despite its greater surface area into the respectively, showed higher DPP-4 inhibitory activity than the
S1 pocket of the enzyme. This result is also in accord with the others. On the other hand, the 9i derivative tethering a 5-chloro-
binding energies obtained from docking experiments. In addition, 2-methoxy group displayed lower inhibitory activities than the
having a higher hydrophilicity/lipophilicity balance of 9e might other compounds. The results collectively indicated that the
increase its solubility (log P = 3.4) and so its penetration into mono-substituted compounds showed higher DPP-4 inhibitory
the target cells. Among those compounds containing mono- activities than the disubstituted derivatives. In addition, the
substituted groups (9b, 9c, and 9d), compounds 9c and 9d with combination of electron-withdrawing and releasing groups
electron-withdrawing groups at position 4 of ring A showed higher influenced the DPP-4 inhibitory activity of the compounds.
DPP-4 inhibitory activities compared to 9b bearing CF3 at position Moreover, the quinazoline-pyrimidine hybrid with a bulky
3 (in the order of F 4 Cl). The results also showed that removing substituent i.e., 9e, had higher activity, probably due to complete
the substitution on ring A in 9a even led to a greater decrease in accommodation into the S1 pocket of the enzyme.
the inhibitory activity. In the case of the disubstituted compounds
A comparison between the amino-benzyl-pyrimidine hybrids
(9f, 9g, and 9h), compound 9g, which has an electron-withdrawing (12a–12c) also indicated that 12a with no substituent on ring A
Table 2 DPP-4 inhibitory activity, binding energy (kcal molꢂ1) and log P of the synthesized compounds
%DPP-4 Enzyme
Binding energy
inhibition at 100 mMa
Entry
R
IC50 (mM)
log P
(kcal molꢂ1
)
9a
9b
9c
9d
9e
9f
9g
9h
9i
12a
12b
12c
H
3-CF3
4-F
27.1 ꢀ 0.67
23.6 ꢀ 6.2
62.1 ꢀ 1.0
53.5 ꢀ 1.1
78.3 ꢀ 1.7
30.5 ꢀ 8.2
50.9 ꢀ 0.84
31.5 ꢀ 0.95
30.8 ꢀ 6.5
63.8 ꢀ 3.8
25.8 ꢀ 1.3
51.5 ꢀ 1.3
0.1 mM
249.0 ꢀ 9
1.9
2.8
2.0
2.4
3.4
2.8
2.5
2.6
2.3
1.1
1.6
2.0
2.2
ꢂ8.8
ꢂ8.9
ꢂ8.2
ꢂ8.7
ꢂ9.3
ꢂ8.1
ꢂ8.6
ꢂ8.5
ꢂ8.8
ꢂ8.5
ꢂ7.5
ꢂ8.6
ꢂ9.3
197.0 ꢀ 11
62.6 ꢀ 2.1
87.7 ꢀ 2.7
34.3 ꢀ 3.3
157.0 ꢀ 4.5
100.1 ꢀ 0.3
195.0 ꢀ 5
4-Cl
4-Phenoxy
3,4-diMe
3F-4Me
3Cl-4F
5Cl-2OMe
H
4-Me
4-Br
—
210.0 ꢀ 1.2
85.0 ꢀ 2.6
281.0 ꢀ 0.5
98.3 ꢀ 1.2
0.021 ꢀ 0.03
Sitagliptin
1 mM
91.0 ꢀ 2.8
67.0 ꢀ 1.4
a
Values represent mean ꢀ standard deviation SD of three independent experiments.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
Paper
NJC
Fig. 6 Kinetics of DPP-4 inhibition of compound 9e: (a) Lineweaver–Burk plot, and (b) Lineweaver–Burk secondary plot.
showed higher inhibitory activity against DPP-4 than 12b and The molecular weight (MW) of most compounds was in the
12c, in the order of H 4 Br 4 Me.
range of 328–566. The log P values displayed that all of the
compounds have reasonable lipophilicity. The hydrogen bond
properties (as donors or acceptors), total polar surface area
(TPSA), and rotatable bond number of all designed compounds
are within the acceptable limit. Generally, the data indicated
that all of the compounds followed Lipinski’s rule of five except
the MW in only some cases.
The drug-likeness and Absorbance Distribution Metabolism
and Excretion (ADME) properties of all compounds and sita-
gliptin were derived from the preADMET online server (http://
preadmet.bmdrc.org/). The ADME properties, including Human
Intestinal Absorption (HIA), in vitro Caucasian colon adenocarci-
noma cell permeability (Caco-2), skin permeability, in vitro Plasma
Protein Binding (PPB) and in vivo Blood–Brain Barrier penetration
(BBB) are shown in Table 4. HIA analysis indicated that all
compounds displayed good human intestinal absorption (HIA),
which causes quick absorption from the intestine to the blood-
stream. The Caco-2 permeability parameters suggest that these
3.6. Kinetic study of DPP-4 inhibition
The pharmacokinetic profile of 9e, the most potent DPP-4
inhibitor compound, was determined by the analysis of the
Lineweaver–Burk (double-reciprocal) plot. Plots of 1/V versus
1/[S] for DPP-4 inhibition displayed a competitive inhibitory
mode by the same Y-intercept, and different slopes and X-inter-
cept at increasing concentration of the inhibitor, 0, 1, 10, 50
and 100 mM (Fig. 6a). The Ki value (inhibition constant value)
was found to be 33.24 mM for 9e as the most potent compound
using the Michaelis–Menten equation (Fig. 6b).
3.7. In silico Physicochemical parameter (ADME) prediction
The physicochemical properties of all synthesized compounds
and sitagliptin were calculated using the tool at http://www.
swissadme.ch/, and the results are revealed in Table 3.
Table 3 Physicochemical properties of the synthesized compounds
Table 4 In silico ADME of the synthesized compounds
TPSA
Lipinski
Entry
MW
log Pb HBDc HBAd (A2)e
n-RBf violation
a
Absorption
In vitro
Distribution
% In vitro
plasma
9a
9b
9c
9d
9e
9f
9g
9h
9i
12a
12b
12c
472.5 1.9
540.5 2.8
490.5 2.0
507.0 2.4
566.6 3.4
500.1 2.8
504.6 2.5
524.9 2.6
537.0 2.3
328.4 1.1
342.4 1.6
407.3 2.0
1
1
1
1
1
1
1
1
1
1
1
1
1
5
8
6
5
7
5
6
6
6
108.1
108.1
108.1
108.1
114.0
108.1
108.1
108.1
117.4
73.3
73.3
73.3
77.0
r140 r10
5
6
5
5
7
5
5
5
6
4
4
4
6
0
1
0
1
1
1
1
1
1
0
caco-2 cell
In vitro skin
%
permeability permeability
protein
bonding
Entry
HIAa (nm sꢂ1
)
(log Kp, cm hꢂ1
)
%BBBb
9a
9b
9c
9d
9e
9f
9g
9h
9i
12a
12b
12c
98.14 20.87
97.95 21.5
98.12 21.64
97.46 22.7
97.65 24.43
98.24 21.55
97.97 21.18
97.45 22.55
97.99 23.03
97.19 21.16
97.19 21.14
96.73 24.31
ꢂ3.7
ꢂ2.5
ꢂ4.0
ꢂ3.8
ꢂ2.7
ꢂ3.6
ꢂ4.0
ꢂ4.0
ꢂ3.9
ꢂ3.8
ꢂ3.8
ꢂ3.8
ꢂ2.86
54.7
73.4
59.5
71.0
80.9
50.9
66.0
74.25
65.91
35.71
41.05
57.72
64.09
0.17
0.32
0.25
0.31
0.29
0.15
0.17
0.35
0.24
0.03
0.08
0.08
0.02
3
3
3
0
0
r1
Sitagliptin 407.3 2.2
10
r10
Rule of
Lipinski
a
r500 r5
r5
b
Molecular weight (MW). Logarithm of the partition coefficient
c
between n-octanol and water (log P). Number of hydrogen bond donors
(HBD). Number of hydrogen bond acceptors (HBA). Topological polar
surface area (TPSA). Number of rotatable bonds (nRB).
Sitagliptin 96.90 30.6
d
e
f
a
b
Human intestinal absorption. In vivo blood–brain barrier penetration.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
NJC
Paper
Fig. 7 Expression profile of DPP-4 (CD26)-PE in HT-29, SW1116 and A549. The stained diagrams (green) were overlaid on the corresponding unstained
diagrams (blue) in both the surface (a) and intracellular (b) regions.
compounds have moderate permeability for penetration of biolo- that DPP-4 (CD26) expression is associated with immune
gical membranes. Compounds 9b and 9e have useful skin pene- regulation, which is mainly exerted through enzymatic cleavage
tration (Kp = ꢂ2.5 and ꢂ2.76 cm hꢂ1) compared to sitagliptin (Kp = of cytokines and interfering with T cell activation and cell
ꢂ2.86 cm hꢂ1). Also, most of the compounds bound moderately migration.51,52 In this study, to investigate the activity of the
to plasma protein and so can diffuse or transport across the cell designed compounds as DPP-4 (CD26) inhibitors, we first
membranes. Moreover, the designed compounds showed low determined the expression of the DPP-4 (CD26) marker in the
blood–brain barrier percentages and probably will not exert a investigated cells in both the surface and intracellular regions
considerable neurotoxic effect.
(Fig. 7). For surface-staining, harvested cells were also stained
with 7-AAD vital dye to efficiently exclude nonspecific binding
usually observed in nonviable cells; the frequency of DPP-4
(CD26) positive cells was then determined in 7-AAD neg viable
cells. The maximum surface and intracellular expressions of
3.8. Flow cytometric analysis of DPP-4 (CD26) expression
DPP-4 (CD26) is a transmembrane glycoprotein expressed in
different cell types, including tumor cells. It has been shown
Table 5 In vitro cytotoxic activity of the novel designed compounds on the studied cancerous cell lines
(IC50 ꢀ SD)a mM
Entry
R
HT-29
SW1116
A549
MRC-5
9a
9b
9c
9d
9e
9f
9g
9h
H
3-CF3
4-F
88.6 ꢀ 11.4
103.15 ꢀ 3.1
48.4 ꢀ 3
85.4 ꢀ 5
229 ꢀ 1
316 ꢀ 6.8
292.5 ꢀ 10.3
213.7 ꢀ 4.5
143 ꢀ 5.6
85.2 ꢀ 3.8
306 ꢀ 10.8
217.5 ꢀ 5.1
324 ꢀ 5.6
196 ꢀ 2.3
417 ꢀ 11.8
359 ꢀ 6.2
142 ꢀ 5.8
163.2 ꢀ 3.75
54.0 ꢀ 1
256.45 ꢀ 4.95
192 ꢀ 3
4-Cl
36.1 ꢀ 2.1
10.67 ꢀ 0.3
69.15 ꢀ 2.6
62.1 ꢀ 1.1
69.1 ꢀ 2.4
67.9 ꢀ 2.2
129.5 ꢀ 11.5
78.3 ꢀ 7.3
27.9 ꢀ 6.5
35 ꢀ 1.2
58.3 ꢀ 3.35
19.03 ꢀ 0.03
129.4 ꢀ 2.5
63.5 ꢀ 0.5
70.4 ꢀ 1.94
103 ꢀ 7
125.9 (ꢀ3.9)
60.5 ꢀ 2.7
4-Phenoxy
3,4-diMe
3F-4Me
3Cl-4F
5Cl-2OMe
H
4-Me
4-Br
—
243.5 ꢀ 4.5
168.25 ꢀ 0.05
227.65 ꢀ 1.35
182.25 ꢀ 2.75
277.5 ꢀ 14.5
291.25 ꢀ 7.25
120.5 ꢀ 16.5
4500
9i
12a
12b
12c
Sitagliptin
Cisplatin
160.7 ꢀ 2.25
190.9 ꢀ 1.95
49.2 ꢀ 1.2
389 ꢀ 2.5
—
107.1 ꢀ 2.3
41.69 ꢀ 1.32
5.7 ꢀ 1.6
13.5 ꢀ 1.5
a
Measured in two or three independent experiments.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
Paper
NJC
minimum percentage of both surface and intracellular DPP-4
(CD26) expression (2.59% and 6.72%). These observations
partly explain the better cytotoxic activities of the designed
compounds as DPP-4 inhibitors in the HT-29 cell line. Surface
and intracellular expression of DPP-4 (CD26) were reported in
other studies on cancer53,54 as well as at https://www.proteina
tlas.org/.
3.9. Cytotoxic activity studies
The tumor growth inhibition of the designed compounds was
evaluated against four human cell lines including A549 (non-
small cell lung carcinoma), SW1116 (colon carcinoma), HT-29
(colorectal adenocarcinoma) and MRC-5 (normal cells isolated
from human lung tissue). The results were compared to cisplatin
and sitagliptin as the standard drugs. As summarized in Table 5,
all compounds showed desirable inhibitory effects on both the
colorectal and lung carcinoma cell lines compared to sitagliptin as
Fig. 8 Cytotoxic effects of compounds 9a–9i, 12a–12c, sitagliptin and
cisplatin on HT-29, SW1116, and A549 carcinoma and MRC-5 normal lines.
DPP-4 (CD26) were observed in colorectal carcinoma cell lines
HT-29 (24.9% and 63%) followed by SW1116 (2.22% and
8.48%), while the lung cancer cell line (A549) showed the
Fig. 9 Flow cytometric analysis of the apoptotic effect of 9e and 12c. HT-29 cells for 72 h with 9e (C) and 12c (D). Untreated cells (A) were mostly viable
as they were negative for both Annexin V and 7-AAD. Ctrl (+) heated at 56 1C (B).
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
NJC
Paper
the standard drug. The best cytotoxic effect was observed on
On the other hand, the comparison among 9f, 9g, 9h, and 9i
HT-29, followed by SW1116 and less on A549, probably related bearing disubstituted groups indicated that electronic balance
to higher expression of DPP-4 (CD26) in the HT-29 cell line between electron-withdrawing and releasing substituents
than the others. All compounds also showed a lesser cytotoxic improves the anti-tumor activity of the compound. In series b
effect on the normal cells (MRC-5) in comparison to the (12a–12c), it is clear that the presence of a halogen group (Br) at
carcinoma cell lines, which showed their relative selectivity the para position of ring A significantly increases their anti-
between tumorigenic and non-tumorigenic cell lines. Almost tumor activities. There was good agreement between the DPP-4
all compounds (except 12a) showed significantly higher cyto- inhibitory activity and anticancer activity of this series.
toxic activities than cisplatin and sitagliptin on the HT-29 cell
3.10. Determining the apoptotic effect of 9e and 12c on the
HT-29 cell line
line (Table S1, ESI†). Among the two series, compounds 9e
with an IC50 of 10.67 ꢀ 0.3 mM and 12c with an IC50 of 27.9 ꢀ
6.5 mM were the most potent compounds compared to sita- Apoptosis was measured using an Annexin V (AV)-7-amino-
gliptin (IC50 = 35 ꢀ 1.2 mM) and cisplatin (IC50 = 107.1 ꢀ actinomycin D (7-AAD) apoptosis detection kit by the flow
2.3 mM), Fig. 8. Similarly, these two compounds (IC50 = 19.03 ꢀ cytometry method. This technique allows us to detect the
0.03 mM for 9e and 49.2 ꢀ 1.2 for 12c) showed the best viable cells (AVneg/7-AADneg), early phase apoptotic cells
inhibitory effects on SW1116, another colorectal cell line. (AVpos/7-AADneg), late phase apoptotic cells (AVpos/7-AADpos),
Meanwhile, in the case of A549, all compounds presented and necrotic cells (AVneg/7-AADpos). In this method,
significantly lower activities IC50 in the range of 60–291 mM fluorochrome-labeled AV can especially detect phosphatidylser-
(Table 5 and Table S3, ESI†). The structure–activity relationship ine that translocates to the outer membrane during apoptosis.
also indicated that in series 9a–9i, among the mono-substituted 7-AAD is a DNA specific dye that distinguishes live and dead
groups, it seems that having bulky (4-phenoxy) and halogen groups cells. The apoptotic effect of 9e and 12c on HT-29 cells was
in the order of Cl 4 F 4 CF3 on ring A increases the antiproli- evaluated in three different concentrations (6, 12, and 24 mM
ferative activity, which is compatible with the results obtained with for 9e and 15, 30, and 60 mM for 12c) after 24 and 72 hours. The
the DPP-4 inhibitory assay. 9e with the best DPP-4 inhibitory results showed that during the first 24 hours, neither of the 9e
activity (IC50 = 34.3 mM) also had higher cytotoxic activity.
and 12c compounds could induce apoptosis in the HT-29 cells
Fig. 10 The effect of 9e and 12c on the cell cycle in the HT-29 cell line after 72 hours of treatment.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
Paper
NJC
(Fig. S58, ESI†), but the percentages of AV positive cells potent candidate in colon cancer treatment, yet they need
increased from 18% to 44%, 67% and 77% for 9e and 50%, further optimized modifications.
64% and 70% for 12c after 72 hours of treatment (Fig. 9). These
results implied that the apoptotic effects of 9e and 12c in the
HT-29 cell line are dose and time-dependent.
Conflicts of interest
There are no conflicts to declare.
3.11. The potential effect of 9e and 12c on the HT-29 cells’ cell
cycle
A flow cytometry-based method was performed to investigate
the cell cycle distribution of the most potent compounds of
both series (9e and 12c) on the HT-29 cell line. For this purpose,
the HT-29 cells were treated with three different concentrations
of 9e (6 mM, 12 mM, and 24 mM) and 12c (15 mM, 30 mM, and
60 mM) compared to untreated cells for 24 and 72 hours. The
cell cycle profiles after 24 hours and 72 hours are depicted in
Fig. 10 and Fig. S59 (ESI†), respectively. As could be seen, after
72 hours of incubation, the cells treated with 6 mM and 12 mM
of 9e showed accumulation in the G2/M phase compared with
the untreated cells, 10.5% and 17.4% vs. 7.77%, respectively.
Similar results were also observed for 12c with 15 mM (5.62%),
30 mM (9.64%), and 60 mM (29.4%) (Fig. 12, ESI†). The results
indicated that compounds 9e and 12c could induce partial cell
cycle arrest in G2/M phases.
Acknowledgements
Financial assistance from the Shiraz University of Medical
Sciences by way of grant number 97-01-05-17869 is gratefully
acknowledged.
References
1 G. Ghersi, Q. Zhao, M. Salamone, Y. Yeh, S. Zucker and
W.-T. Chen, Cancer Res., 2006, 66, 4652–4661.
2 M. Gonzalez-Gronow, H. E. Grenett, M. R. Weber, G. Gawdi
and S. V. Pizzo, Biochem. J., 2001, 355, 397–407.
3 P. A. Havre, L. H. Dang, K. Ohnuma, S. Iwata, C. Morimoto
and N. H. Dang, BMC Cancer, 2013, 13, 517.
4 A. J. Scheen, Expert Opin Drug Discovery, 2015, 14, 505–524.
5 S. H. Tella and M. S. Rendell, Expert Opin Drug Discovery,
2015, 14, 127–140.
6 A. Beckenkamp, S. Davies, J. B. Willig and A. Buffon, Tumor
Biol., 2016, 37, 7059–7073.
7 P. A. Havre, M. Abe, Y. Urasaki, K. Ohnuma, C. Morimoto
and N. H. Dang, Front. Biosci., 2008, 13, 1634–1645.
8 B. Pro and N. H. Dang, Histol. Histopathol., 2004, 4,
1345–1351.
4. Conclusion
In the present research, certain quinazolinone–pyrimidine and
benzyl-pyrimidine hybrid derivatives were designed, synthe-
sized, and evaluated for their DPP-4 inhibitory activity and
antiproliferative potential. Our results illustrated that com-
pound 9e, containing a 4-phenoxy moiety on the phenyl ring
at position 3 of the quinazolinone scaffold, was the most potent
derivative with a DPP-4 inhibitory function and cytotoxic effect.
9 R. Bishnoi, Y. R. Hong, C. Shah, A. Ali, W. P. Skelton IV and
J. Huo, et al., Cancer Med., 2019, 8, 3918–3927.
Furthermore, inhibition kinetics analysis indicated that 9e 10 S. Nishina, A. Yamauchi, T. Kawaguchi, K. Kaku, M. Goto
acted in a competitive inhibitory mode. The SAR studies
reported that the substitution of a bulky substituent or an
and K. Sasaki, et al., Cell Mol. Gastroenterol Hepatol., 2019, 7,
115–134.
electron-withdrawing with an electron-donating group on the 11 C. Hollande, J. Boussier, J. Ziai, T. Nozawa, V. Bondet and
phenyl ring increased the DPP-4 inhibitory activity. The highest W. Phung, et al., Nat. Immunol., 2019, 20, 257–264.
anticancer activity belonged to compound 9e, particularly on 12 M. Kraman, P. J. Bambrough, J. N. Arnold, E. W. Roberts,
colorectal cancer HT-29 and SW1116 cell lines. The greater
cytotoxic activity of the studied compounds could be attributed
L. Magiera and J. O. Jones, et al., Science, 2010, 330,
827–830.
to the greater expression of DPP-4 (CD26) in both the surface 13 K. Ohnuma, R. Hatano and C. Morimoto, Nat. Immunol.,
and intracellular compartments in the colorectal cancer cell 2015, 16, 791–792.
lines in comparison with the lung cell lines as demonstrated 14 T. Iino, Y. Sasaki, M. Bamba, M. Mitsuya, A. Ohno and
with the flow cytometry method.
K. Kamata, et al., Bioorg. Med. Chem. Lett., 2009, 19,
A good correlation was observed between the CD26 expres-
5531–5538.
sion and the anticancer activity as well as the DPP-4 inhibitory 15 K. Lin, Z. Cai, F. Wang, W. Zhang and W. Zhou, Chem.
effect for compounds 9e and 12c (as the most potent com- Pharm. Bull., 2013, c12-01046.
pounds in each series). Not only could these compounds 16 A. Padmaja, C. Rajasekhar, A. Muralikrishna and
induce apoptosis in the HT-29 cell line in a dose and time- V. Padmavathi, Eur. J. Med. Chem., 2011, 46, 5034–5038.
dependent manner, but also they could arrest cells in the G2/M 17 H. Xie, L. Zeng, S. Zeng, X. Lu, X. Zhao and G. Zhang, et al.,
phase. The physicochemical and drug-likeness properties of Eur. J. Med. Chem., 2013, 68, 312–320.
the designed compounds indicated that the compounds are 18 P. M. Chandrika, T. Yakaiah, A. R. R. Rao, B. Narsaiah,
compatible with the Lipinski rule, and 9e is a moderate and
selective candidate as a DPP-4 inhibitor and also acted as
N. C. Reddy and V. Sridhar, et al., Eur. J. Med. Chem., 2008,
43, 846–852.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
NJC
Paper
19 M. Dupuy, F. Pinguet, O. Chavignon, J.-M. Chezal, J.-C. 37 M. B. Wallace, J. Feng, Z. Zhang, R. J. Skene, L. Shi and
Teulade and J.-P. Chapat, et al., Chem. Pharm. Bull., 2001,
49, 1061–1065.
C. L. Caster, et al., Bioorg. Med. Chem. Lett., 2008, 18,
2362–2367.
20 Y. Takase, T. Saeki, N. Watanabe, H. Adachi, S. Souda and 38 Z. Zhang, M. B. Wallace, J. Feng, J. A. Stafford, R. J. Skene
I. Saito, J. Med. Chem., 1994, 37, 2106–2111. and L. Shi, et al., J. Med. Chem., 2011, 54, 510–524.
21 N. Rahmannejadi, S. Khabnadideh and I. Yavari, Monatsh. 39 L. Emami, R. Sabet, A. Sakhteman and M. Khoshneviszadeh,
Chem., 2018, 149, 2085–2092. J. Pharm. Res. Int., 2019, 1–15.
22 N. Rahmannejadi, I. Yavari and S. Khabnadideh, J. Heterocycl. 40 M. Fereidoonnezhad, Z. Faghih, A. Mojaddami, Z. Rezaei
Chem., 2020, 57, 978–982. and A. Sakhteman, Iran. J. Pharm. Res., 2017, 16, 981.
23 R. A. Castaldo, D. W. Gump and J. J. McCormack, Antimicrob. 41 M. Fereidoonnezhad, H. A. Mirsadeghi, S. Abedanzadeh,
Agents Chemother., 1979, 15, 81–86.
24 J. H. Chan, J. S. Hong, L. F. Kuyper, D. P. Baccanari,
A. Yazdani, A. Alamdarlou and M. Babaghasabha, et al., New
J. Chem., 2019, 43, 13173–13182.
S. S. Joyner and R. L. Tansik, et al., J. Med. Chem., 1995, 42 Z. Faghih, N. Rahmannejadi, R. Sabet, K. Zomorodian,
38, 3608–3616.
M. Asad and S. Khabnadideh, Results Pharm Sci., 2019,
14, 115.
25 N. Rahmannejadi, K. Zomorodian, Z. Faghih, Z. Faghih and
S. Khabnadideh, J. Pharm. Res. Int., 2017, 1–11.
26 J. Hanusek, Chem. Listy, 2001, 95, 811–813.
27 L. Kubicova, P. Kudelova, H. Dostal and K. Waisser, Folia
Pharm. Univ. Carol., 2000, 25, 81–87.
28 M. Divar, K. Zomorodian, S. Bastan, S. Yazdanpanah and
S. Khabnadideh, Information Management, 2018, 15,
1457–1466.
29 M. Yen, J. Sheu, I. Peng, Y. Lee and J. Chern, J. Pharm.
Pharmacol., 1996, 48, 90–95.
30 B. Lam, Z. Zhang, J. A. Stafford, R. J. Skene, L. Shi and S. L.
Gwaltney II, Bioorg. Med. Chem. Lett., 2012, 22, 6628–6631.
31 J.-U. Peters, S. Weber, S. Kritter, P. Weiss, A. Wallier and
´
43 H. R. Shahsavari, N. Gimenez, E. Lalinde, M. T. Moreno,
M. Fereidoonnezhad and R. Babadi Aghakhanpour, et al.,
Eur. J. Inorg. Chem., 2019, 1360–1373.
44 P. Ramesh, D. Srinivasulu, D. Subbarao, K. Hemakumar,
D. Rajasekhar and C. Nagaraju, Pharm. Lett., 2012, 4,
1344–1351.
45 S. Xu, Q. Hao, H. Li, Z. Liu and W. Zhou, Org. Process Res.
Dev., 2017, 21, 585–589.
46 W. A. Weihofen, J. Liu, W. Reutter, W. Saenger and H. Fan,
J. Biol. Chem., 2005, 280, 14911–14917.
47 D. K. Yadav, F. Khan and A. S. Negi, J. Mol. Model., 2012, 18,
2513–2525.
M. Boehringer, et al., Bioorg. Med. Chem. Lett., 2004, 14, 48 D. K. Yadav, S. Kumar, H. S. Saloni, M.-H. Kim, P. Sharma
1491–1493. and S. Misra, et al., Drug Des., Dev. Ther, 2017, 11, 1859.
32 S. R. Sagar, J. K. Agarwal, D. H. Pandya, R. P. Dash, 49 C. Granchi, A. Capecchi, G. Del Frate, A. Martinelli, M. Macchia
M. Nivsarkar and K. K. Vasu, Bioorg. Med. Chem. Lett., 2015,
25, 4428–4433.
33 M. Boehringer, H. Fischer, M. Hennig, D. Hunziker,
and F. Minutolo, et al., Molecules, 2015, 20, 8772–8790.
50 P. Chen, C. G. Caldwell, W. Ashton, J. K. Wu, H. He and
K. A. Lyons, et al., Bioorg. Med. Chem. Lett., 2011, 21, 1880–1886.
J. Huwyler and B. Kuhn, et al., Bioorg. Med. Chem. Lett., 51 S. Iwata and C. Morimoto, J. Exp. Med., 1999, 190, 301–306.
2010, 20, 1106–1108.
52 K. Ohnuma, N. H. Dang and C. Morimoto, Trends Immunol.,
2008, 29, 295–301.
53 D. Darmoul, M. Lacasa, L. Baricault, D. Marguet, C. Sapin
and P. Trotot, et al., J. Biol. Chem., 1992, 267, 4824–4833.
34 Y. Miyamoto, Y. Banno, T. Yamashita, T. Fujimoto, S. Oi and
Y. Moritoh, et al., J. Med. Chem., 2011, 54, 831–850.
35 K. V. Parsa and M. Pal, Expert Opin. Drug Discovery, 2011, 6,
855–869.
´
´
54 L. Vazquez-Iglesias, L. Barcia-Castro, M. Rodrıguez-Quiroga,
´
36 J. Peters, S. Weber, S. Kritter, P. Weiss and A. Wallier, Bioorg.
Med. Chem. Lett., 2004, 14, 1491–1493.
M. P. de la Cadena, J. Rodrıguez-Berrocal and O. J. Cordero,
Biol. Open, 2019, 8, bio041673.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020