Synthesis, antitumor and antimicrobial activity of some new 6-methyl-3-phenyl-4(3H)-quinazolinone analogues: in silico studies

Abstract

Some new derivatives of substituted-4(3H)-quinazolinones were synthesized and evaluated for their in vitro antitumor and antimicrobial activities. The results of this study demonstrated that compound 5 yielded selective activities toward NSC Lung Cancer EKVX cell line, Colon Cancer HCT-15 cell line and Breast Cancer MDA-MB-231/ATCC cell line, while NSC Lung Cancer EKVX cell line and CNS Cancer SF-295 cell line were sensitive to compound 8. Additionally, compounds 12 and 13 showed moderate effectiveness toward numerous cell lines belonging to different tumor subpanels. On the other hand, the results of antimicrobial screening revealed that compounds 1, 9 and 14 are the most active against Staphylococcus aureus ATCC 29213 with minimum inhibitory concentration (MIC) of 16, 32 and 32 μg/mL respectively, while compound 14 possessed antimicrobial activities against all tested strains with the lowest MIC compared with other tested compounds. In silico study, ADME-Tox prediction and molecular docking methodology were used to study the antitumor activity and to identify the structural features required for antitumor activity.

• Antimicrobial
• in silico study
• in vitro antitumor
• molecular docking
• quinazolinones

Introduction

Cancer, a continuing to be a major health problem worldwide, is the leading cause of human mortality exceeded only by cardiovascular diseases1. Cancer is a disease characterized by a shift in the controlled mechanisms of cell proliferation and differentiation2. Malignancy is caused by abnormalities in cells, which might be due to inherited genes or outside exposure of the body to chemicals, radiation or even infectious agents3^,4. Several techniques have been adopted for the treatment and eradication of cancerous cells. These techniques involved surgery, radiation, immunotherapy, chemotherapy and chemoprevention. Ideal anticancer drugs would eradicate cancer cells without harming normal tissues. Unfortunately, no currently available agents meet this criterion and the clinical use of drugs involves a weighing of benefits against toxicity in a search of favourable therapeutic index5. On the other hand, the spread of antibiotic resistance among pathogenic bacteria has become a major problem for the clinical management of infectious diseases6. Such medical health problems have encouraged many medicinal chemists to search for novel antibacterial agents other than the analogs of existing antibiotics7.

Quinazolines are frequently used in medicine because of their wide spectrum of biological activities8–26. Based on the good performances of quinazoline derivatives in anticancer application, development of novel quinazoline derivatives as anticancer drugs is a promising field9. FDA has approved several quinazoline derivatives as anticancer drugs, such as Gefitinib, Erlotinib, Lapatinib and Vandetanib9. Moreover, a series of triazol-4-yl-substituted quinazolines and 2-thio-[1,2,4]triazolo[1,5-c]quinazoline derivatives have been identified as potent antimicrobial agents27. Therefore, the development of a novel and efficient anticancer agent that can be used for prophylaxis as well as the treatment of bacterial infections in cancer patients remains an important goal in medicinal chemistry.

Based on aforementioned rational and our studies on quinazoline derivatives as attractive candidates for antitumor9^,23–25 and antimicrobial agents10^,11^,21, we have designed a number of new quinazoline derivatives containing a 3-phenyl substituent attached with different fragments at 2-position to identify the dual antitumor and antimicrobial activities. Moreover, an in silico study, ADME-T prediction and molecular docking studies were used to identify the structural features required for the antitumor properties of the designed compounds.

Materials and methods

Chemistry

Melting points were recorded on Barnstead 9100 Electrothermal melting apparatus. IR spectra (KBr) were recorded on a FT-IR Perkin-Elmer spectrometer (ν cm⁻¹) at Research Center, King Saud University, Saudi Arabia. Nuclear magnetic resonance (¹H and ¹³C NMR) spectra were recorded on Bruker 500 MHz spectrometer (Japan) using DMSO-d₆, C₅D₅N or CDCl₃ as solvents at Research Center, King Saud University, Saudi Arabia. The chemical shifts are expressed in δ ppm using TMS as internal standard. Mass spectra were recorded on an Agilent 6320 Ion Trap mass spectrometers at Research Center, King Saud University, Saudi Arabia. Elemental analysis was carried out for C, H and N at the Research Centre of College of Pharmacy, King Saud University, Saudi Arabia and the results are within ±0.4% of the theoretical values. Solvent evaporation was performed under reduced pressure using Buchan Rotatory Evaporator (Germany). Thin layer chromatography was performed on precoated (0.25 mm) silica gel GF254 plates (E. Merck, Darmstadt, Germany); compounds were detected with 254 nm UV lamp. Silica gel (60–230 mesh) was employed for routine column chromatography separations. Compounds 1–3 and 5 were prepared according to our previous report25, while compounds 4 and 17 were reported by Kennewell et al.26 and Zhao28.

Preparation of compounds 3, 4 and 5

A solution of 2-(ethoxycarbonylmethyl)-thio-3-phenyl-4-oxo-6-methyl-3H-quinazoline (3.0 g, 0.01 mol) and hydrazine hydrate (99%, 8 mL) was heated under reflux for 1 h. The obtained solid was filtered, dried and chromatographed (silica gel, 100 g, elution with EtOAc–Hexan 1:10 v/v) to give compounds 3, 4 and 5 in 6%, 81% and 10%, respectively.

1-Substituted-7-methyl-4-phenyl-[1,2,4]triazolo[4,3-a] quinazolin-5(4H)-one (6–7)

2-Hydrazinyl-6-methyl-3-phenyl-quinazolin-4(3H)-one (5) (265 mg, 1 mmol) and triethylorthoformate or triethylorthoacetate (2.5 mL) were refluxed for 7 h, and the reaction mixture was cooled and poured into ice water. The solid obtained was filtered, washed with water, dried and recrystallized from ethanol.

7-Methyl-4-phenyl-[1,2,4]triazolo[4,3-a]quinazolin-5(4H)-one (6)

Yield 64%, m.p. 270–272 °C; IR (KBr, cm⁻¹) ν: 1716 (C=O); ¹H NMR (DMSO-d₆): δ 8.54 (s, 1H), 7.52–8.31 (m, 3H), 7.64–6.92 (m, 5H), 2.46 (s, 3H, CH₃). 13C NMR (DMSO-d₆): δ 154.9, 151.8, 146.8, 142.4, 135.8, 133.0, 132.9, 129.5, 128.2, 125.9, 125.6, 123.5, 20.6. Ms: [M⁺ 276].

1,7-Dimethyl-4-phenyl-[1,2,4]triazolo[4,3-a]quinazolin-5(4H)-one (7)

Yield 66%, m.p. 280–282 °C; ¹H NMR (C₅D₅N): δ 9.06 (s, 1H), 8.98 (d, 1H, J = 8.0 Hz), 7.93–7.45 (m, 5H), 7.20–7.19 (m, 1H), 2.34 (s, 3H, CH₃), 2.20 (s, 3H, CH₃). 13C NMR, (C₅D₅N): δ 156.7, 151.4, 148.3, 148.1, 133.6, 133.3, 130.7, 130.5, 128.5, 127.2, 126.8, 120.4, 118.9, 21.4, 12.3. Ms: [M⁺ 290].

1-Substituted-7-methyl-4-phenyl-[1,2,4]triazolo[4,3-a] quinazolin-5(4H)-one (8–9)

A mixture of 2-hydrazinyl-6-methyl-3-phenyl-quinazolin-4(3H)-one (5) (265 mg, 1 mmol) and appropriate carboxylic acid (1 mmol) was refluxed for 8 h in a mixture of DMF (5 mL) and phosphorus oxychloride (5 mL), and the reaction mixture was cooled and poured into ice water. The solid obtained was filtered, washed with water, dried and recrystallized from ethanol.

7-Methyl-4-phenyl-1-styryl-[1,2,4]triazolo[4,3-a]quinazolin-5(4H)-one (8)

Yield 60%, m.p. 291–292 °C; ¹H NMR (C₅D₅N): δ 8.08–7.93 (m, 1H), 7.93–7.45 (m, 5H), 7.83–7.59 (m, 3H), 7.56–7.55 (m, 4H), 7.41–7.10 (m, 6H), 2.42 (d, 3H, J = 6.0 Hz). Ms: [M⁺ 378].

1.1.8.2. 7-Methyl-1,4-diphenyl-[1,2,4]triazolo[4,3-a]quinazolin-5(4H)-one (9)

Yield 65%, m.p. 299–300 °C; ¹H NMR (CDCl₃ & TFA): δ 8.13 (s, 1H), 7.60–7.50 (m, 8H), 8.45 (t, 2H, J = 6.5 Hz), 7.38 (t, 2H, J = 7.5 Hz), 2.46 (s, 3H, CH₃). ¹³C NMR (CDCl₃): δ 156.5, 151.5, 147.0, 135.9, 133.7, 132.8, 131.5, 130.0, 129.7, 129.3, 128.7, 127.8, 126.6, 126.1, 124.7, 117.6, 21.0. Ms: [M⁺ 352].

Ethyl-2-[2-(ethoxycarbonyl)hydrazono]-6-methyl-4-oxo-3-phenyl-3,4-dihydroquinazoline-1(2H)-carboxylate (10)

A mixture of 2-hydrazinyl-6-methyl-3-phenyl-quinazolin-4(3H)-one (5) (265 mg, 1 mmol) and diethyloxalate (150 mg, 1 mmol) was refluxed in ethanol (5 mL) for 12 h, and the reaction mixture was cooled and solvent was evaporated under reduced pressure. The solid obtained was filtered, washed with water, dried and recrystallized from ethanol.

Yield 67%, m.p. 221–222 °C; IR (KBr, cm⁻¹) ν: 1747, 1734, 1710 (C=O), 3288 (NH); ¹H NMR (CDCl₃): δ 8.14 (s, 1H), 7.51–7.39 (m, 7H), 7.55 (t, 2H, J = 7.5 Hz), 4.23 (t, 2H, J = 7.0 Hz), 2.38 (s, 3H, CH₃), 1.55 (t, 2H, J = 7.5 Hz), 1.23 (q, 3H, J = 7.5 Hz), 0.82 (t, 3H, J = 7.5 Hz). ¹³C NMR (CDCl₃): δ 156.2, 147.1, 146.5, 142.5, 136.4, 134.1, 132.4, 132.1, 130.1, 129.7, 127.4, 126.6, 126.3, 117.4, 67.3, 30.2, 21.0, 18.9, 13.5. Ms: [M⁺ 410].

2-Benzylidenehydrazono-6-methyl-3-phenyl-2,3-dihydroquinazolin-4(1H)-one (12)

A mixture of 2-hydrazinyl-6-methyl-3-phenyl-quinazolin-4(3H)-one (5) (265 mg, 1 mmol) and benzaldehyde (106 mg, 1 mmol) was refluxed in methanol (5 mL) for 12 h, and the reaction mixture was cooled and solvent was evaporated under reduced pressure. The solid obtained was filtered, washed with water, dried and recrystallized from ethanol. Yield 80%, m.p. 253–254 °C; IR (KBr, cm⁻¹) ν: 1717 (C=O), 3322 (NH); ¹H NMR (DMSO-d₆): δ 10.73 (s, 1H, NH, D₂O exchangeable), 8.52 (s, 1H), 7.93 (s, 1H), 7.76–7.72 (m, 3H), 7.46–7.00 (m, 9H), 2.35 (s, 3H, CH₃). ¹³C NMR (DMSO-d₆): δ 161.6, 147.8, 146.8, 142.2, 139.0, 138.3, 135.7, 132.4, 129.2, 128.8, 128.5, 126.9, 125.2, 122.3, 119.1, 117.9, 115.2, 20.5. Ms: [M⁺ 354].

7-Methyl-4-phenyl-1-(phenylamino)-[1,2,4]triazolo[4,3-a] quinazolin-5(4H)-one (13)

A mixture of 2-hydrazinyl-6-methyl-3-phenyl-quinazolin-4(3H)-one (5) (265 mg, 1 mmol) and phenylisothiocyanate (135 mg, 1 mmol) was refluxed in methanol (5 mL) for 12 h, and the reaction mixture was cooled and solvent was evaporated under reduced pressure. The solid obtained was filtered, washed with water, dried and recrystallized from ethanol.

Yield 73%, m.p. 300–302 °C; IR (KBr, cm⁻¹) ν: 1716 (C=O), 3327 (NH); ¹H NMR (C₅D₅N): δ 10·41 (s, 1H, NH, D₂O exchangeable), 8.46 (s, 1H), 8.10 (d, 2H, J = 7.0 Hz), 7.70 (d, 2H, J = 6.5 Hz), 7.60 (d, 1H, J = 8.0 Hz), 7.40–7.31 (m, 5H), 7.20–7.17 (m, 1H), 7.08–7.07 (m, 1H), 2.32 (s, 3H, CH3). ¹³C NMR (C₅D₅N): δ 156.5, 147.7, 147.1, 140.5, 133.5, 133.4, 132.2, 130.6, 130.5, 129.9, 129.5, 128.9, 127.0, 126.7, 123.6, 120.6, 119.5, 21.4. Ms: [M⁺ 367].

1-Thioxo-2,4-dihydro-[1,2,4]triazolo[4,3-a]quinazolin-5(1H)- one (14)

To a solution of 2-hydrazinyl-6-methyl-3-phenyl-quinazolin-4(3H)-one (5) (2 mmol, 530 mg) in absolute ethanol (20 mL) containing potassium hydroxide (5 mmol, 280 mg), carbon disulfide (760 mg, 10 mmol) was added and the reaction mixture was stirred at room temperature for 3 h followed by heating under reflux for 6 h. The reaction mixture was cooled; the solvent was evaporated under reduced pressure. The residual solid obtained was dissolved in water, acidified with dilute hydrochloric acid; and the precipitated solid was filtered, washed with water and crystallized from ethanol.

Yield 81%, m.p. 294–296 °C; IR (KBr, cm⁻¹) ν: 1707 (C=O), 3330 (NH); ¹H NMR (DMSO-d₆): δ 12·99 (s, 1H, SH, D₂O exchangeable), 7.75 (s, 1H), 7.61 (d, 1H, J = 7.0 Hz), 7.50–7.47 (m, 2H), 7.42 (d, 1H, J = 7.5 Hz), 7.36 (d, 1H, J = 8.5 Hz), 7.27 (d, 2H, J = 7.5 Hz), 2.37 (s, 3H, CH₃). ¹³C NMR (DMSO-d₆): δ 175.5, 159.8, 150.2, 139.3, 137.6, 136.6, 133.9, 129.0, 128.9, 128.0, 126.7, 125.6, 116.0, 115.7, 20.4. Ms: [M⁺ 308].

2-Chloro-N′-(6-methyl-4-oxo-3-phenyl-3,4-dihydroquinazolin-2(1H)-ylidene)acetohydrazide (15)

A solution of 2-hydrazinyl-6-methyl-3-phenyl-quinazolin-4(3H)-one (5) (2 mmol, 530 mg) was stirred at room temperature with chloroacetylchloride (2 mmol, 225 mg) in dichloromethane (10 mL) containing triethylamine (5 mmol, 505 mg) for 12 h and the reaction mixture was heated under reflux for 6 h. The reaction mixture was diluted with water and extracted with chloroform, the extract was washed successively with water and brine, dried over anhydrous MgSO₄, evaporated under reduced pressure and solid obtained was crystallized from ethanol. Yield 88%, m.p. 215–216 °C; IR (KBr, cm⁻¹) ν: 1712, 1684 (C=O), 3343, 3318 (NH); ¹H NMR (DMSO-d₆): δ 11·49 (s, 1H, NH, D₂O exchangeable), 9.80 (s, 1H, NH, D₂O exchangeable), 7.80 (s, 1H), 7.68 (d, 2H, J = 7.5 Hz), 7.59–7.52 (m, 2H), 7.43 (t, 1H, J = 7.5 Hz), 7.37 (d, 1H, J = 8.0 Hz), 7.23 (t, 1H, J = 8.0 Hz), 4.25 (s, 2H), 2.40 (s, 3H, CH₃).¹³C NMR (DMSO-d₆): δ 166.7, 158.1, 147.8, 137.0, 136.6, 133.4, 129.6, 128.8, 126.0, 125.0, 123.5, 123.2, 115.3, 41.6, 20.4, Ms: [M + 1 343].

2-Amino-6-methyl-3-phenylquinazolin-4(3H)-one (16), 6-methyl-3-phenylquinazolin-4(3H)-one (17)

A solution of 2-hydrazinyl-6-methyl-3-phenyl-quinazolin-4(3H)-one (5) (2 mmol, 530 mg) in (10 mL) absolute methanol containing triethylamine (4 mmol, 405 mg) was refluxed for 12 h. The reaction mixture was allowed to cool, the solvent was removed under reduced pressure, and the solid obtained was dried and chromatographed (SiO₂, 20 g, elution with CHCl₃–AcOEt, 10:1 v/v) afforded compounds 16 and 17 in 32% and 31% yield, respectively.

2-Amino-6-methyl-3-phenylquinazolin-4(3H)-one (16)

Yield 32%, m.p. 228–230 °C; IR (KBr, cm⁻¹) ν: 1706 (C=O), 3337, 3314 (NH); ¹H NMR (DMSO-d₆): δ 10.80 (s, 1H, NH, D₂O exchangeable), 8.69 (s, 1H, NH, D₂O exchangeable), 7.76 (s, 1H), 7.74 (d, 2H, J = 7.5 Hz), 7.47 (d, 1H, J = 7.0 Hz), 7.36–7.31 (m, 3H), 7.03 (t, 1H, J = 7.0 Hz), 2.36 (s, 3H). ¹³C NMR (DMSO-d₆): δ 161.6, 147.6, 147.8, 146.8, 139.0, 135.7, 132.3, 128.8, 126.4, 122.3, 120.6, 119.1, 118.0, 20.5. Ms: [M⁺ 251].

1-(Chloromethyl)-7-methyl-4-phenyl-[1,2,4]triazolo[4,3-a] quinazolin-5(4H)-one (19)

A solution of 2-hydrazinyl-6-methyl-3-phenyl-quinazolin-4(3H)-one (5) (2 mmol, 530 mg) was stirred at room temperature with (1 M) trifluoroborane (2 mmol, 136 mg) in (10 mL) dry THF for 12 h. The solvent was removed under reduced pressure, and the solid obtained was dried and chromatographed (SiO₂, 20 g, elution with CHCl₃–AcOEt, 10:1 v/v) afforded compound 19.

Yield 88%, m.p. 230–232 °C; IR (KBr, cm⁻¹) ν: 1704 (C=O); ¹H NMR (CDCl₃): δ 8.23 (s, 1H), 7.67–7.59 (m, 6H), 7.50 (d, 1H, J = 8.5 Hz), 4.62 (s, 2H), 2.48 (s, 3H, CH₃).¹³C NMR (CDCl₃): δ 156.3, 149.5, 147.0, 146.5, 136.2, 134.0, 131.2, 130.6, 130.3, 127.6, 126.5, 126.1, 117.5, 33.6, 21.0. Ms: [M⁺ +1 325].

Antitumor screening

A primary anticancer assay was performed for an approximately 60 human tumor cell lines panel derived from nine neoplastic diseases, in accordance with the protocol of the Drug Evaluation Branch, National Cancer Institute, Bethesda, MD29–33.

Antimicrobial screening

The initial screening of antimicrobial activity and determination of minimum inhibitory concentration (MIC) for the tested compounds were performed by cup plate and broth dilution methods, respectively with different strains34. Eighteen different synthesized compounds were screened for their antimicrobial activity against four bacterial standard strains (Staphylococcus aureus ATCC 29213, Bacillus subtilis ATCC 10 400 and Escherichia coli ATCC 25922) and one yeast standard strain, Candida albicans ATCC 2091. The tested compounds were dissolved in dimethyl sulfoxide (DMSO) to obtain stock solution (5120 µg/mL).

Cup plate method

The tested organisms were grown in Cation Adjustment Mueller-Hinton (CAMH) broth (Merck®, Darmstadt, Germany) to mid-log phase. The bacterial suspension was measured spectroscopically using Spectrophotometer (LKB® Ultrospec) at 625 nm to give absorbance 0.1–0.14 (1 × 10⁸ CFU/mL). The bacterial suspension was diluted 1:100 in CAMH broth to obtain 1 × 10⁶ CFU/mL. This suspension was swabbed on a CAMH agar plate (Merck®, Darmstadt, Germany) and allowed to dry completely. Then six cups per plate were made in agar plate using cork borer. A 1 mL of stock solution (5120 µg/mL) was one 2-fold diluted in 1 mL DMSO to obtain 2560 µg/mL. A 100 µL (256 µg) of the tested compound was added into the cup. The plates were kept in refrigerator at 4 °C for 30 min for diffusion. Then, the plates were incubated at 37 °C for 24 h. After the incubation period, the diameter of the inhibition zone (including the diameter cup) was measured and recorded in mm. Ciprofloxacin, Ampicillin (10 µg/cup) Flucytosine and Fluconazole (10 µg/mL) were used as positive controls for antibacterial and antifungal, respectively. The experiment was carried out in duplicate.

Determination of MIC

MIC was determined for the compounds that show antimicrobial activity by cup plate method. Briefly, 2 mL of CAMH broth (Oxoid Chemical Co., UK) was dispensed into screw glass tubes (100 mm × 760 mm). Then the tubes were autoclaved. For each compound, 14 tubes were used. Tubes number 13 and 14 were used as positive growth control (no tested compound) and negative control for medium sterility (no microorganism), respectively. A 1 mL of stock solution (5120 µg/mL) was 10-fold diluted in 9 mL CAMH to obtain 512 µg/mL. A 2 mL of the tested compounds (512 µg/mL) was pipetted into the first tube and mixed well. Then 2 mL was withdrawn from the 1st tube and added to the 2nd tube to make a 2-fold dilution. This procedure was repeated down to 12th tube to reach the concentration of 0.125 µg/mL. A 2 mL was discarded from the 12th tube. A volume of 2 mL of inoculums (1 ×10⁶ CFU/mL) was added to all tubes except tube number 14 to give final 1 × 10⁶ CFU/mL. Ampicillin and Fluconazole were used as positive controls for antibacterial and antifungal, respectively. The inoculated tubes were incubated at 37 °C for 20 h. After the incubation period, the results of MIC were recorded manually and interpreted according to the guidelines of EUCAST.

Docking methodology

Molecular modeling studies were performed with MOE 2008.10, software available from Chemical Computing Group Inc., 1010 Sherbrooke Street West, Suite 910, Montreal, Canada H3A2R7.

Selection of protein crystal structures

Ligand-bound crystallographic structure of epidermal growth factor receptor (EGFR) kinase is available in the Protein Data Bank. In this study, EGFR kinase crystal structure 1M17 was evaluated and selected for docking. The errors of the protein were corrected by the structure preparation process in MOE. The first step in the generation of suitable protein structures is the assignment of hydrogen positions on the basis of default rules. Water molecules contained in the PDB file have been removed. Finally, partial charges (Gasteiger methodology) were calculated and the active site of the ensemble has been defined as the collection of residues within 10.0 Å of the bound inhibitor and comprised the union of all ligands of the ensemble. All atoms located less than 10.0 Å from any ligand atom were considered.

Preparation of the ligand

The ligand coordinates were built using the builder tool of the MOE program. Next, the correct atom types (including hybridization states) and correct bond types were defined, hydrogen atoms were added, charges were assigned to each atom and finally the structures were energy-minimized (MMFF94x, gradient: 0.01).

Docking experiment

The docking experiment on EGFR kinase was carried out by superimposing the energy minimized ligand on erlotinib in the PDB file 1M17, after which erlotinib was deleted. The default triangle matcher placement method was used for docking. GBVI/WSA dG scoring function which estimates the free energy of binding of the ligand from a given pose was used to rank the final poses. The ligand–enzyme complex with lowest S score was selected.

Results and discussion

Chemistry

Synthesis of 2-mercapto-6-methyl-3-phenylquinazolin-4(3H)-one as starting material in 90% yield was achieved by the reaction of 5-methylanthranilic acid with 2-phenylisothiocyanate in absolute ethanol25. 2-Mercapto-6-methyl-3-phenylquinazolin-4(3H)-one was reacted with ethyl 2-bromoacetate in dry acetone in the presence of potassium carbonate at room temperature to afford ethyl 2-[(6-methyl-4-oxo-3-phenyl-3,4-dihydroquinazolin-2-yl)thio]acetate (1) in 95% yield, and the latter compound was heated with hydrazine hydrate in ethanol to furnish 2-[(6-methyl-4-oxo-3-phenyl-3,4-dihydroquinazolin-2-yl)thio]acetohydrazid (2) in 90% yield. Furthermore, heating of the ester 1 with hydrazine hydrate gave 2-hydrazino-3-(2-methylphenyl)-3H-quinazolin-4-one (5) as a key intermediate in 81% yield, in addition to 2-hydrazinyl-6-methylquinazolin-4(3H)-one (3) and 3-amino-2-hydrazinyl-6-methylquinazolin-4(3H)-one (4) as by-products in 6% and 10% yield, respectively (Scheme 1).

Scheme 1. Synthesis of 2-hydrazinyl-6-methyl-3-phenyl-quinazolin-4(3H)-one (5) as a key intermediate.

2-Hydrazino-3-(2-methylphenyl)-3H-quinazolin-4-one (5) was reacted with various one carbon donors, such as triethylorthoformate, triethyortholacetate, cinnamic acid and benzoic acid, to produce 1-substituted-7-methyl-4-phenyl-[1,2,4]triazolo[4,3-a]quinazolin-5(4H)-one derivatives (6–9) in 60–66% yields. Ethyl 2-(2-(ethoxycarbonyl)hydrazono)-6-methyl-4-oxo-3-phenyl-3,4-dihydroquinazoline-1(2H) carboxylate (10) was obtained in 67% yield by the reaction of compound 5 with diethyloxalate (Scheme 2).

Scheme 2. Reactions of compound 5 with orthoesters, carboxylic acids and diethyloxalate.

Additionally, the synthesis of [(benzylidene)hydrazono]-6-methyl-3-phenyl-2,3-dihydroquinazolin-4(1H)-one (12) in 80% yield and 7-methyl-4-phenyl-1-(phenylamino)-[1,2,4]triazolo[4,3-a]quinazolin-5(4H)-one (13) in 73% yield were achieved by boiling of compound 5 with benzaldehyde or phenylisothiocyanate in methanol. Moreover, compound 5 was treated with carbon disulfide in ethanolic potassium hydroxide to afford 7-methyl-4-phenyl-1-thioxo-2,4-dihydro-[1,2,4]triazolo[4,3-a]quinazolin-5(1H)-one (14) in 81% yield (Scheme 3).

Scheme 3. Reactions of compound 5 with benzaldehyde, phenylisothiocyanate and carbondisulfide.

On the other hand, 2-chloro-N′-(6-methyl-4-oxo-3-phenyl-3,4-dihydroquinazolin-2(1H)-ylidene)acetohydrazide (15) was obtained by heating compound 5 with chloroacetylchloride in dichloromethane. An attempt to cyclize compound 15 by boiling in methanol containing triethylamine was unsuccessful and a mixture of 2-amino-6-methyl-3-phenylquinazolin-4(3H)-one (16) and 6-methyl-3-phenylquinazolin-4(3H)-one (17) was obtained rather than 1-(chloromethyl)-7-methyl-4-phenyl-[1,2,4]triazolo[4,3-a]quinazolin-5(4H)-one (18) in 32% and 31% yields, respectively. Finally, compound 15 was cyclized to the corresponding 1-(chloromethyl)-7-methyl-4-phenyl-[1,2,4]triazolo[4,3-a]quinazolin-5(4H)-one (19) by stirring the reaction mixture in tetrahydrofuran containing Lewis acid, such as borontriflouride, in 88% yield (Scheme 4).

Scheme 4. Synthesis of 1-(chloromethyl)-7-methyl-4-phenyl-[1,2,4]triazolo[4,3-a]quinazolin-5(4H)-one (19).

Antitumor activity

In vitro antitumor evaluations of seven compounds indicated in Table 1 were selected by National Cancer Institute, Bethesda, Maryland, USA on the basis of degree of the structure variation and computer modeling techniques for evaluation of their anticancer activity.

Table 1. Antitumor activity of quinazoline derivatives 5–9, 12 and 13 presented as growth inhibition percentages (GI%) over 60 subpanel tumor cell lines.

	60 cell lines assay in one dose 10.0 µM concentration: GI%
Compd No.	MGI%	PCE	Most sensitive cell lines
5	3	9/55	Leukemia (MOLT-4: 14.0%, SR: 22.0%), NSC Lung Cancer (EKVX 27.0%), Colon Cancer (HCT-15: 25%), CNS Cancer (SF-268: 11%, SF-295: 21%), Melanoma (MDA-MB-435: 13%), Breast Cancer (MDA-MB-231/ATCC: 60%, BT-549: 11%)
6	3	2/44	Leukemia (NCI-H522: 10%), Melanoma (UACC-62: 10%)
7	0	4/55	CNS Cancer (SF-295: 10%), Breast Cancer (MDA-MB-231/ATCC: 17%, BT-549: 10%, MDA-MB-468: 13%)
8	5	12/54	Leukemia (K-562: 16%, MOLT-4: 13%, SR: 22%), NSC Lung Cancer (EKVX: 52%), Colon Cancer (HCT-15: 12%), CNS Cancer (SF-295: 36%, SF-539: 15%), Melanoma (MDA-MB-435: 18%), Ovarian Cancer (OVCAR-5: 11%), Renal Cancer (CAKI-1: 17%), Breast Cancer (BT-549: 17%, T-47D: 10%)
9	4	11/55	Leukemia (MOLT-4: 10%, SR: 26%), NSC Lung Cancer (NCI-H460: 11%), Colon Cancer (HCT-116: 16%), CNS Cancer (SF-539: 12%), Melanoma (MDA-MB-435: 14%), Ovarian Cancer (IGROV1: 19%, NCI/ADR-RES: 11%), Renal Cancer (UO-31: 16%), Breast Cancer (MDA-MB-231/ATCC: 13%, BT-549: 26%)
12	12	23/55	Leukemia (CCRF-CEM: 32%, K-562: 43%, MOLT-4: 22%, SR: 51%), NSC Lung Cancer (NCI-H226: 11%, NCI-H460: 44%), Colon Cancer (HCT-116: 35%, HCT-15: 57%), CNS Cancer (SF-268: 12%, SF-295: 25%, SNB-75: 11%), Melanoma (LOX IMVI: 30%, SK-MEL-5: 32%, UACC-62: 20%), Ovarian Cancer (OVCAR-4: 21%, OVCAR-5: 11%, NCI/ADR-RES: 14%), Renal Cancer (786-0: 12%, CAKI-1: 12%), Prostate Cancer (PC-3: 30%), Breast Cancer (MCF7: 14%, MDA-MB-231/ATCC: 17%, MDA-MB-549: 11%)
13	11	25/55	Leukemia (K-562: 16%, MOLT-4: 30%, SR: 25%), NSC Lung Cancer (EKVX: 16%, NCI-H226: 11%, NCI-H23: 16%), CNS Cancer (SF-295: 14%, SF-539: 15%, SNB-19: 15%, SNB-75: 18%), Melanoma (MALME-3M: 17%, M14: 12%, MDA-MB-435: 25%, SK-MEL-5: 13%, UACC-62: 10%), Ovarian Cancer (OVCAR-4: 21%, NCI/ADR-RES: 18%), Renal Cancer (ACHN: 13%, CAKI-1: 31%, RXF 393: 12%, UO-31: 37%), Prostate Cancer (PC-3: 16%), Breast Cancer (MDA-MB-231/ATCC: 35%, BT-549: 30%, MDA-MB-468 19%)

PCE: positive cytotoxic effect, which is the ratio between number of cell lines with percentage growth inhibition >10% and total number of cell lines. MGI%: mean growth inhibition percentage.

The selected compounds 5–9, 12 and 13 were subjected to the National Cancer Institute (NCI) in vitro disease-oriented human cells screening panel assay for in vitro antitumor activity. A single dose (10 µM) of the test compounds is used in the full NCI 60 cell lines panel assay including nine tumor subpanels namely; Leukemia, Non-small cell lung, Colon, CNS, Melanoma, Ovarian, Renal, Prostate and Breast cancer cells29–33. The data are reported as mean-graph of the percent growth of the treated cells, and presented as percentage growth inhibition (GI%) caused by the test compounds (Table 1).

The tested compounds 5–9, 12 and 13 displayed moderate activities in the in vitro screening on the tested cell lines in 10 µM concentration with positive cytotoxic effect (PCE) of 2/44, 4/55, 9/55, 12/54, 11/55, 23/55 and 25/55 respectively (Table 1). Compounds 5, 8, 9, 12 and 13 showed a cytotoxic effect on some of the cancer cell lines with mean growth inhibition percentages (MGI%) of 3%, 5%, 4%, 12% and 11%, respectively (Table 1).

Comparison of the antitumor activities of compounds 5, 8, 12 and 13 with GI% over the most sensitive tumor cell line against 5-flurouracil showed that compounds 5, 8, 12 and 13 (Table 2) possess activities almost equal to or higher than those of 5-flurouracil against most cell lines (Figure 1).

Figure 1. Antitumor activity of compounds 5, 8, 12 and 13 with GI% over the most sensitive tumour cell line.

Table 2. Antitumor activity of compounds 5, 8, 12 and 13 with GI% over the most sensitive tumour cell line.

	% Growth Inhibition (GI%)*
Subpanel tumor cell lines	5	8	12	13	5-Flu
Leukemia
CCRF-CEM	–	–	32	–	57.1
K-562	–	–	43	–	42.3
SR	–	–	51	25	25
MOLT-4	–	–	–	30	43
NSC Lung Cancer
EKVX	27	52	–	–	39.0
NCI-H460	–	–	44	–	13
Colon Cancer
HCT-116	–	–	35	–	17.8
HCT-15	25	–	57	–	26
CNS Cancer
SF-295	–	36	–	–	69.1
Melanoma
LOX IMVI	–	–	30	–	30.4
MDA-MB-435	–	–	–	25	36.6
SK-MEL-5	–	–	32	–	95.5
Renal Cancer
CAKI-1	–	–	–	31	39.4
UO-31	–	–	–	37	41.3
Prostate Cancer
PC-3	–	–	30	–	58.2
Breast Cancer
MDA-MB 231/ATCC	60	–	–	35	78.1
BT-549	–	26	–	30	Nt

5-Flu: 5-fluorouracil, Nt: not tested.

*Only GI% higher than 25% are shown.

Regarding the activity toward individual cell lines; compounds 5, 8, 9, 12 and 13 showed selective activities against MOLT-4 and SR leukemia cell lines with GI values of 14%, 13%, 10%, 22%, 30%, 22%, 22%, 26%, 51% and 25%, respectively. Additionally, compounds 8, 12 and 13 disclosed activity against K-562 leukemia cell line with GI values of 16%, 43% and 16%, respectively, whereas NCI-H522 and CCRF-CEM leukemia cell lines were sensitive to compounds 6 and 12, respectively.

Non-small cell lung; EKVX cell line proved to be selectively sensitive to compounds 5, 8 and 13 with GI values of 27%, 52% and 16%, respectively. In addition, compounds 9 and 12 proved to be susceptible to the NCI-H460 cell line with GI values of 11% and 44%, respectively. Compounds 12 and 13 have dedicated activity against NCI-H226 cell line with GI values of 11% of each; meanwhile, compound 13 showed certain activity against NCI-H322 cell line in 13%.

Activity against colon cancer; compounds 5, 8 and 12 showed GI values of 23%, 12% and 22% with colon HCT-15 cell line, respectively, while compounds 9 and 12 demonstrated moderate activities against HCT-116 cancer cell line with GI values of 16% and 35%, respectively.

Concerning CNS cancer; compounds 5, 7, 8, 9, 12 and 13 showed moderate potency against CNS cancer SF-295 cell line with GI values of 21%, 10%, 36% and 25%, respectively. Compounds 8, 9 and 13 showed GI values of 15%, 12% and 15% to CNS cancer SF-539 cell line, respectively. CNS cancer SF-268 proved to be sensitive to compounds 5 and 12 with GI values of 11% and 12%, while compounds 12 and 13 showed certain activity to SNB-75 cancer cell with GI values of 11% and 18%, respectively. Additionally, compound 13 displayed modest activity against SNB-19 cancer cell with GI value of 15%.

Regarding Melanoma; compounds 5, 9 and 13 were active against MDA-MB-435 cell line with GI values of 13%, 14% and 25%, respectively. Melanoma UACC-62 and LOX IMVI cell lines were sensitive to compounds 6 and 12 in 10% and 30%, respectively.

Compound 13 showed moderate activities against MALME-3 M and M14 cell lines in 17% and 12%, while compounds 12 and 13 possessed activities against UACC-62 and SK-MEL-5 cell lines with GI values of 20%, 10%, 32% and 13%, respectively.

Investigation of activity toward ovarian cancer; compounds 12 and 13 showed moderate activities against Ovarian NCI/ADR-RES and OVCAR-4 cell lines with GI values of 14%, 18%, 21%, and 21%, respectively. OVCAR-5 cell line was perceptive to compounds 8 and 12 with GI values of 11%, while Ovarian IGROV1 and NCI/ADR-RES cell lines were responsive to compound 9 with GI values of 19% and11%, respectively.

Relating to renal cancer; compounds 8 and 12 were active against renal CAKI-1 cell line with GI values of 17% and 12%, respectively, whereas ACHN, RXF 393, UO-31 and CAKI-1 cell lines were sensitive to compound 13 with GI values of 13%, 12%, 37% and 31%, respectively. Renal UO-31and 786-0 cell lines were responsive to compounds 9 and 12 with GI values of 16% and 12%, respectively. Prostate PC-3 cell line proved to be selectively sensitive to compounds 12 and 8 with GI value of 30% and 16%, respectively.

Pertaining to breast cancer; breast MDA-MB-231/ATCC and BT-549 cell lines possessed convinced response to compounds 5, 7, 9, 12 and 13 with GI values of 60%, 17%, 13%, 17%, 35%, 11%, 10%, 26%, 11%, and 30%, respectively. Compounds 7 and 13 showed GI effectiveness against breast MDA-MB-468 cell line with values of 13% and 19%, respectively. Additionally, compounds 8 and 12 have selective activities against breast T-47D and MCF7cell lines with GI values of 10% and 14%, respectively.

Antibacterial and antifungal activities

Compounds 1–19 in addition to the reference Ciprofloxacin and Ampicillin were tested for their in vitro antibacterial activity against Gram negative (E. coli ATCC 25922), Gram-positive (S. aureus ATCC 29213 and B. subtilis ATCC 10400) microorganisms. Also compounds 1–19 in addition to the references Fluconazole and Flucytosine were tested for their in vitro antifungal activity against Candida albicans ATCC2091. The minimal inhibitory concentrations (MIC in µg/mL) or the lowest drug concentrations that prevent visible growth of bacteria (Table 3) were determined by a standard broth micro-dilution technique using the European Committee for Antimicrobial Susceptibility Testing (EUCAST) and Laboratory Standards method34.

Table 3. Antibacterial and antifungal activity of compounds 1–19 with MIC µg/mL.

	Antimicrobial activities
	Staphylococcus aureus ATCC 29213		Bacillus subtilis ATCC 10400		Escherichia coli ATCC 25922		Candida albicans ATCC2091
Compound No.	Cup plate (mm)	MIC (µg/mL)	Cup plate (mm)	MIC (µg/mL)	Cup plate (mm)	MIC (µg/mL)	Cup plate (mm)	MIC (µg/mL)
1	19	16	Nil	ND	12	64	10	256
2	Nil	ND	Nil	ND	Nil	ND	Nil	ND
3	Nil	ND	15	128	Nil	ND	Nil	ND
4	11	128	14	128	11	64	Nil	ND
5	Nil	ND	12	256	Nil	ND	15	64
6	Nil	ND	Nil	ND	Nil	ND	Nil	ND
7	Nil	ND	14	128	13	64	Nil	ND
8	Nil	ND	Nil	ND	Nil	ND	Nil	ND
9	17	32	15	128	Nil	ND	14	64
10	12	128	11	256	Nil	ND	Nil	ND
12	Nil	ND	12	256	13	64	Nil	ND
13	10	256	Nil	ND	15	32	16	64
14	17	32	15	128	19	32	15	64
15	11	256	Nil	ND	Nil	ND	10	256
16	10	256	Nil	ND	Nil	ND	Nil	ND
17	Nil	ND	Nil	ND	Nil	ND	Nil	ND
19	12	128	12	128	Nil	ND	11	256
CI	27	1	25	1	29	1	ND	ND
AM	20	16	18	16	22	16	ND	ND
FLU	ND	ND	ND	ND	ND	ND	28	16
FL	ND	ND	ND	ND	ND	ND	18	16

FLU: Fluconazole; FL: Flucytosine; AM: Ampicillin; CI: Ciprofloxacin; MIC = minimum inhibitory concentration; ND = not determined.

The results of MIC tests against Gram-positive and Gram-negative bacteria revealed that compounds 2, 6, 8 and 17 were inactive against all tested organisms either bacteria or fungi. On the other hand, the tested compounds were more active against Gram-positive bacteria (S. auerus, B. subtilis) than Gram-negative bacteria (E. coli).

Compounds 1, 4, 9, 10, 13, 14, 15, 16 and 19 exhibited selective activity against Gram-positive S. aureus ATCC 29213 with MIC; 16, 128, 32, 128, 256, 32, 256, 256, 128 µg/mL, respectively.

Additionally, compounds 3, 4, 5, 7, 9, 10, 12, 14 and 19 possessed activities against Gram-positive B. subtilis ATCC 10400 with MIC; 128, 128, 256, 128, 128, 256, 256, 128, 128 µg/mL, respectively. Compounds 4, 9, 10, 14 and 19 have broad spectrum activities against both B. subtilis ATCC 10400 and S. aureus ATCC 29213. Compounds 1, 4, 7, 12, 13 and 14 were sensitive to Gram-negative E. coli ATCC 25922 in comparison with MIC; 64, 64, 64, 64, 32 and 32 µg/mL, respectively. Compounds 4 and 14 show extended spectrum activity against S. aureus ATCC 29213, B. subtilis ATCC 10400 and E. coli ATCC 25922.

Regarding antifungal activities, compounds 1, 5, 9, 13, 14, 15 and 19 possessed activities against C. albicans ATCC 2091 with MIC; 256, 64, 64, 64, 64, 256 and 256 µg/mL, respectively. The results revealed that compounds 9 and 14 were found to be most active against S. auerus ATCC 29213 with MIC of 32 μg/mL and inhibition zone of 17 mm, while compound 14 had activity against all tested strains either bacterial or C. albicans strain with lowest MIC against all the tested strains (Figure 2).

Figure 2. Antibacterial and antifungal activity of the active compounds against the reference drugs with MIC µg/mL.

In silico studies of antitumor activity

Lipinski’s rule of five (the effect of lipophilic and steric parameters)

As a part of our study, the compliance of compounds to the Lipinski’s rule of five was evaluated35. In addition, the polar surface area (PSA) of the tested compounds was also calculated (Table 4), since it is another key property that has been linked to drug bioavailability, where passively absorbed compounds with a PSA >140 Ų are thought to have low oral bioavailability36. The results disclosed in Table 4 show that all of the synthesized compounds comply with these rules (molecular weight = 266–378, Clog P = 1.46–5.0, nON = 5–6 and nOHNH = 0–3). Hence, theoretically, all of these compounds should present good passive oral absorption and the differences in their bioactivity cannot be attributed to this property24^,35–37.

Table 4. Calculated Lipinski’s rule of five for the designed antitumor compounds.

			Parameter
Comp. No.	PCE*	MR†	Log P‡	PSA¶	MW§	nOHNH||	nON#	Nviolations**
5	9/55	76.67	1.46	70.72	266.3	3	5	0
6	2/44	80.77	2.9	50.50	276.29	0	5	0
7	4/55	85.19	3.13	51.02	290.32	0	5	0
8	12/54	115.09	5.0	51.02	378.43	0	5	0
9	11/55	105.87	4.54	51.02	352.39	0	5	0
12	23/55	107.52	4.68	57.06	354.4	1	5	0
13	25/55	108.93	4.8	62.53	367.4	1	6	0
5-Flu	–	25.9	−0.84	58.2	130.0	2	4	0
Erlotinib	–	110.06	3.07	74.73	393.0	1	7	0

*Data taken from Table 1.

†Molar refractometry.

‡Calculated lipophilicity.

¶Polar surface area (Ų).

§Molecular weight.

||Number of hydrogen bond donor.

#Number of hydrogen bond acceptor.

**Number of violation from Lipinski’s rule of five.

The introduction of cyclic ring fragments incorporating the quinazoline core and the variation of the substituents on these fragments have allowed us to evaluate the influence of lipophilicity and steric parameters at the pharmacophoric part of the molecules. Table 4 gathers cytotoxic effect as well as values of Clog P (lipophilic factor), molar refractometry and PSA (steric factors) for each compound, determined by using online molinspiration program (http://www.molinspiration.com/cgi-bin/properties) and iLab2 online program (https://ilab.acdlabs.com/iLab2/index.php). It is clear that the cytotoxic effect increases with the increase in molar refractometry from 76–85 cm³/mol (5–7) to 105–115 cm³/mol (8, 9, 12 and 13). The optimal refractometry for the active compounds was found to lie in the range of 105.87–108.93 cm³/mol (Table 4). Although, lipophilicity does not exert a significant effect on activity, an increase in potency was observed in compounds 8, 9, 12 and 13 with log P values >3.0 and ≤5.0. The results shown in Table 4 have almost similar values with erlotinib which indicated the structural similarity of the designed molecules with erlotinib.

ADME-Tox evaluation

To estimate the prospect of the designed compounds as antitumor agents compared with the reported antitumor agents erlotinib and 5-Flu; their drug-likeness were calculated according to absorption, distribution, metabolism, elimination and toxicity (ADME-T) program (Table 5), and defined human intestinal absorption (HIA) model37^,38. It was predicted that the examined compounds could be transported across the intestinal epithelium, and they can cross the blood–brain barrier (BBB) and are of medium aqueous soluble. The values of HIA, protein binding, BBB crossing and solubility prediction for all compounds are presented in Table 5 using iLab2 online program (https://ilab.acdlabs.com/iLab2/index.php). In general, all compounds presented some advantages and disadvantages when compared to each other and the results were compared with erlotinib and 5-Flu as reference drugs. No marked differences, in human oral bioavailability (30%–>70%), human intestinal absorption (98–100%), human jejunum permeability and health effects in rodent toxicity profiles, were observed among the designed compounds. However, the solubility related parameters call for attention, since the promising compounds were calculated to be at least as soluble as the reported compounds, and are predicted to have oral bioavailability and absorption significantly higher than that of the reported antitumor agents 5-Flu and similar to erlotinib. It can be deduced from these results that the pharmacokinetic profile of the designed compounds is affected and modified by the presence of cyclic triazole and the substituents connected to quinazoline scaffold.

Table 5. The predicted ADME-Tox and solubility parameters of the tested compounds and reported antitumor agents.

ADME-Tox	5	6	7	8	9	12	13	Erlotinib*	5-Flu*
Solubility (Log S)	−2.8 (pH 1.7); −3.72 (pH 8.0)	−4.29 (pH 1.7); −4.83 (pH 8.0)	−4.39 (pH 1.7); −5.32 (pH 8.0)	−6.41 (pH 1.7); −6.88 (pH 8.0)	−6.52 (pH 1.7); −6.78 (pH 8.0)	−4.69 (pH 1.7); −5.42 (pH 8.0)	−5.97 (pH 1.7); −6.53 (pH 8.0)	−2.31 (pH 1.7); −4.78 (pH 8.0)	−0.79 (pH 1.7); 0.67 (pH 8.0)
LogS w  (pure water)	−4.45	−3.50	−3.73	−4.64	−5.13	−5.90	−4.65	−5.00	−1.56
F (%)†	>70% (0.77)	>70% (0.53)	>70% (0.53)	30–70% (0.52)	30–70% (0.52)	30–70% (0.52)	30–70% (0.52)	30–70% (0.50)	>70% (0.92)
HIA (%)‡	98%	100%	100%	100%	100%	100%	100%	100%	98%
Pe (cm/s)¶	1.36 × 10^−4	8.32 × 10^−4	8.11 × 10^−4	7.49 × 10^−4	7.71 × 10^−4	7.45 × 10^−4	7.61 × 10^−4	7.35 × 10^−4	1.5 × 10^−4
Vd (L/kg)§	2.32	1.67	1.92	2.82	2.41	6.45	2.43	1.52	0.56
%PPB||	85.66%	96.53%	96.50%	98.7%	99.21%	98.97%	99.34%	98.99%	15%
Log BB#  (Log PS)**	−0.43 (−2.7)	−0.98 (−1.6)	−0.90 (−1.6)	−0.69 (−1.2)	−0.78 (−1.2)	−0.28 (−1.5)	−0.64 (−1.3)	−0.74 (−1.4)	−0.05 (−3.0)
LD50 mouse  (mg kg^−1, oral)	150	610	600	700	560	760	820	810	860
LD50 mouse  (mg kg^−1,  intraperitioneal)	280	590	550	610	760	260	680	270	500
LD50 mouse  (mg kg^−1,  intravenous)	52	130	120	80	100	37	98	58	120
LD50 mouse  (mg kg^−1,  subcutaneous)	38	260	250	220	180	46	160	280	400
Ames test	0.88	0.36	0.38	0.19	0.47	0.24	0.21	0.15	0.7

*5-Flurouracil.

†Human oral bioavailability (Probability).

‡Human intestinal absorption.

¶Permeability (Human Jejunum).

§Volume of distribution.

||Plasma protein binding.

#Blood brain barrier.

**Rate of brain penetration.

Drug score, drug-likeness and toxicities profiles

Currently there are many approaches that assess a compound drug-likeness based on topological descriptors, fingerprints of molecular drug-likeness structure keys or other properties39. In the Osiris program (http://www.organic-chemistry.org/prog/peo) the occurrence frequency of each fragment is determined within the collection created by shredding 3300 traded drugs as well as 15 000 commercially available chemicals (Fluka) yielding a complete list of all available fragments. In this work, we used the Osiris program for calculating the fragment based drug-likeness of the antitumor compounds also comparing them with 5-Flu and erlotinib (Figure 3). Interestingly, the derivatives 5, 6, 7, 8, 9, 12 and 13 presented good drug-likeness values (−0.40 to 4.63) compared with 5-Flu and erlotinib (−4.50 and −6.73, respectively). In this study, we also verified the drug-Score40 as the theoretical data showed that compounds 5, 6, 7, 8, 9, 12 and 13 (0.33–0.52) presented values once again higher than 5-Flu and erlotinib (0.06 and 0.38, respectively). Moreover, we used the Osiris program to predict the overall toxicity of the most active derivatives as it may point to the presence of some fragments generally responsible for the irritant, mutagenic, tumorigenic or reproductive effects in these molecules. Interestingly, all of the active derivatives presented a low in silico toxicity risk profile, better than 5-Flu and similar to erlotinib (Figure 3). These theoretical data reinforced the cytotoxicity experimental data described in this work pointing these compounds as lead compounds for further study.

Figure 3. In silico toxicity risks (upper panel), drug-Likeness (lower right panel) and drug-Score (lower left panel) of the active antitumor quinazoline derivatives compared with reference drugs 5-Flu and Erlotinib (M, mutagenic; T, tumorigenic; I, irritant; R, reproductive).

Predicted bioactivity score (molinspiration calculations)

The molinspiration bioactivity prediction (http://www.molinspiration.com/docu/miscreen/druglikeness.html) is fast (100 000 molecules may be screened in about 30 min) and therefore allows processing of very large molecular libraries. Validation tests performed on various target classes (including of GPCR ligand, ion channel modulator, nuclear receptor legend, kinase inhibitor, protease inhibitor, enzyme inhibitor) show 10–20-fold increases in hit rate in comparison with standard/random selection of molecules for screening. Calculated drug likeness score of each compounds were compared with the specific activity of other compounds and the results were compared with standard drug. For organic molecules, the probability is if the bioactivity score is (>0), then it is active, if (−5.0 to 0.0) then moderately active, if (<−5.0) then inactive. The calculated value of the predicted bioactivity for compounds 5, 6, 7, 8, 9, 12, 13, 5-Flu and erlotinib is given in Table 6. The results indicated that the synthesized quinazoline derivatives have a predicted kinase inhibiting activity almost similar to erlotinib.

Table 6. In silico predicted bioactivity of the designed antitumor compounds compared with reference drugs 5-Flu and Erlotinib.

Compounds	Protease inhibitor	Enzyme inhibitor	GPCR ligand	Kinase inhibitor	Ion channel modulator	Nuclear receptor ligand
5	−0.69	−0.25	−0.34	−0.22	−0.62	−0.80
6	−0.91	−0.52	−0.52	−0.28	−0.48	−1.07
7	−0.96	−0.65	−0.47	−0.33	−0.55	−1.09
8	−0.71	−0.49	−0.30	−0.22	−0.39	−0.76
9	−0.75	−0.56	−0.41	−0.18	−0.47	−0.82
12	−0.72	−0.37	−0.40	−0.33	−0.66	−0.71
13	−0.61	−0.47	−0.38	−0.04	−0.45	−0.94
Erlotinib	−0.16	0.18	0.13	0.68	0.11	−0.10
5−Flu	−3.15	−1.56	−2.60	−2.61	−1.95	−3.04

Molecular docking study

The level of the predicted activity against kinase enzyme and antitumor activities of the proposed compounds over breast cancer cell, colon cancer cell, lung cancer cell and renal cancer cell (EGFR is highly expressed) prompted us to perform molecular docking into the ATP binding site of EGFR to predict if these compounds have analogous binding mode to the EGFR inhibitors24^,41–43. Compounds 5, 12 and 13 were used for docking study as representative examples of good predicted kinase inhibiting activity (Table 6) and promising antitumor activity (Tables 1 and 2, and Figure 4). All the calculations were performed using MOE 2008.10 software44 installed on 2.4G Core™ i7. The crystal structure of EGFR with erlotinib (Tarceva™) (PDB code: 1M17) was obtained from PDB45. The automated docking program of MOE 2008.10 was used to dock compounds 5, 12, 13 and erlotinib into ATP binding site of EGFR (Figure 4). The complexes were energy-minimized with a MMFF94 force field46 till the gradient convergence 0.01 kcal/mol was reached. The binding energies of −8.73, −15.21, −12.89 and −27.03 kcal/mol were obtained for 5, 12, 13 and erlotinib, respectively (Figure 4). These docking studies have revealed that the quinazoline ring binds to a narrow hydrophobic pocket in the N-terminal domain of EGFR-TK where N-1 of the quinazoline ring interacts with the backbone NH of Met-769 via a hydrogen bond, and similarly, a water (HOH-10) molecule-mediated hydrogen bonding interaction with Thr-830 and Thr-766 side chain. These interactions revealed the importance of nitrogen atoms for binding and the subsequent inhibitory capacity.

Figure 4. Docking of compounds 5 (upper right panel), 12 (lower left panel) and 13 (lower right panel) into the active site of epidermal growth factor receptor. Upper left panel showed the erlotinib inhibitor/EGFR complex. Hydrogen bonds are shown in red.

EGFR–TK complex with compounds 12 and 13 showed the occurrence of three hydrogen bonds with Met-769 (3.22 and 3.26 Å respectively), HOH-10 (3.49, 2.88, 3.61 and 3.09 Å, respectively) mediated hydrogen bonding interaction with Thr-830 and Thr-766 side chain. Moreover, additional non-classical hydrogen bonds are formed between aromatic moiety at 3-position of quinazoline core and the side chain of Gln-767. The lower interaction energy observed for 5 (Figure 4, upper right panel) rationalizes the insufficient binding into the EGFR-TK active site than that of compounds 12 and 13 (Figure 4, lower panels). The insufficient binding can be explained in terms of the occurrence of only two hydrogen bonds between the N-1 and carbonyl of quinazoline ring system with Met-769 (3.28 Å) and HOH-10 mediated hydrogen bonding interaction with Thr-830 and Thr-766 side chain (2.98 Å). In short, Figure 4 demonstrates binding models of quinazoline in the ATP binding site and the results of this molecular docking could support the postulation that our active compounds may act on the same enzyme target where EGFR inhibitor acts confirming the molecular design of the reported class of antitumor agents.

Conclusion

New derivatives of substituted-4(3H)-quinazolinones were synthesized and evaluated for their in vitro antitumor and antimicrobial activity. A single dose (10 µM) of the test compounds was used in the NCI 60 cell lines panel assay. The results of this study demonstrated that compounds 5, 8, 12 and 13 yielded selective activities toward numerous cell lines belonging to different tumor subpanels. They are compatible with Lipinski’s rule of five (molecular weight, Clog P, hydrogen bond-donating and accepting capabilities). Molecular docking studies further supported the antitumor activity of compounds 5, 12, 13 compared with EGFR kinase inhibitors and further helped in understanding the various interactions between the ligands and enzyme active sites in detail and thereby helped to design novel potent derivatives. On the other hand, the results of antimicrobial screening revealed that compounds 1, 9 and 14 were found to be most active against S. auerus ATCC 29213 with MIC of 16, 32 and 32 μg/mL and inhibition zone of 19, 17 and 17 mm, respectively, while compound 14 had activity against all tested strains either bacterial or C. albicans strain with lowest MIC against all the tested strains.

Supplementary material available online

Supplementary Information

Declaration of interest

The authors extend their appreciation to the Deanship of Scientific Research at King Saud University for funding the work through the research group project No. RG-1435-046. The authors would like to express their gratitude and thanks to the National Cancer Institute (NCI), Bethesda Maryland, USA, http://dtp.cancer.gov/ for doing the antitumor testing of the new compounds.