Received date: January 19, 2015; Accepted date: February 16, 2015;; Published date: February 18, 2015
Citation: Ravez S, Schifano-Faux N, Barczyk A, Arsenlis S, Castillo Aguilera O, et al. (2015) Synthesis and Antiproliferative Activity of 2-Amino-4-Anilinoquinazoline Derivatives. Med chem 5:067-076. doi:10.4172/2161-0444.1000245
Copyright: © 2015 Ravez S, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Visit for more related articles at Medicinal Chemistry
Recently, we have reported a series of 4-anilino-6,7-dimethoxyquinazolines as tyrosine kinase inhibitors with interesting in vitro IC50 values for the Epidermal Growth Factor Receptor (EGFR) and/or for the Vascular Endothelial Growth Factor Receptor-2 (VEGFR-2). In this paper, we studied the impact of amino group in C-2 position of the quinazoline core on this series. The new synthesized compounds described herein were evaluated for both in vitro EGFR and VEGFR-2 kinase inhibition and antiproliferative activity in different cancer cell lines (PC3, HT29 and MCF7). 2-Aminoquinazolines substituted by a carbamic acid ester group present an interesting antiproliferative activity without tyrosine kinase inhibition. Thus, we drew the present study to explore the potential interaction of these molecules with the double-stranded DNA, the target of many conventional antitumor agents. We evaluated the strength and the way these molecules bind to DNA by UV-visible spectroscopy, circular dichroism, DNA thermal denaturation, and fluorescence measurements.
2-Aminoquinazolines; Cancer; Intercalating agents; PD153035; Kinase inhibition
ATP: Adenosine 5′-triphosphate; BET: Ethidium Bromide; CD: Circular Dichroism; DMSO: Dimethyl Sulfoxide; DNA: DeoxyriboNucleic Acid; EGFR: Epidermal Growth Factor Receptor; EtOAc: Ethyl Acetate; EtOH: Ethanol; FC: Flash Chromatography; MeOH: Methanol; rt: Room Temperature; VEGFR: Vascular Endothelial Growth Factor Receptor
Cancer is one of the most leading causes of mortality in the world, responsible of almost 13% of deaths worldwide. 12.7 million of new cancers were diagnosed in 2008, 7 million of them in developing countries
. There are different cancer treatments whose use depends on the nature and stage of the tumor. Chemotherapy is largely used since cancer cells grow and multiply much faster than most cells in the body. Many different chemotherapy drugs are available. Antineoplastic drugs are divided into five classes including alkylating agents, anti-metabolites, antitumor antibiotics, topoisomerase inhibitors and tubulin-binding drugs
. Unfortunately, these drugs discriminate between cancer cells and normal cells causing several side effects. Targeted therapies focus particular features of cancer cells, almost absent in normal cells. Although they target different receptors, they all interfere in the growth division, repair and/or communication of cancer cells with their environment. Modern targeted therapies include the use of monoclonal antibodiesand kinase inhibitors [3,4]. They mainly target transmembrane receptors like Epidermal Growth Factor Receptor (EGFR), member of the ErbB family and Vascular Endothelial Growth Factor Receptor-2 (VEGFR-2). Numerous studies demonstrated that aberrant expression or mutation of EGFR can lead to several cancers . The VEGF receptors (especially VEGFR-2) are considered as important targets in antiangiogenic therapies in cancer treatments .
Quinazoline derivatives represent an attractive scaffold to design anticancer drugs because of their multiple biological activities, notably as kinase inhibitors . The first generation of 4-anilinoquinazolines such as the compound PD153035 showed a high EGFR affinity but a low in vivo activity . Many studies were performed in our team around this 4-anilinoquinazoline scaffold and led to the determination of EGFR and/or VEGFR-2 tyrosine kinase inhibitors. The enzymatic and cellular results of the most active compounds are presented in Table 1 [9,10].
|Cell growth inhibition
|2||3-Cl,4-F||0.40||5.30||6.6 ±1.70||6.7 ±0.78||4.9 ±0.54|
a The cell growth rate was evaluated by performing the MTS assay. Data are represented as the mean ± SEM of at three experiments performed in triplicate. Higher concentrations were not used to avoid precipitation of the compound in the culture medium
bInhibition of EGFR (purified from human carcinoma A431 cells) tyrosine kinase activity
cInhibition of VEGFR-2 (recombinant protein) tyrosine kinase activity
Table 1: Enzymatic and cell growth inhibition results of quinazoline derivatives.
The halogenated derivatives (PD153035, 1 and 2) showed an important EGFR inhibition. Among these compounds, only compound 2 inhibited the cancer cell growth, with IC50 of 6.6, 6.7 and 4.9 μM on PC3, HT29 and MCF7, respectively. Substitution of the anilino group by a carbamic acid ester moiety leads to an interesting dual inhibition of EGFR and VEGFR-2. However, compounds 3, 4 and 5 showed weak antiproliferative activities on the three used cancer cell lines (IC50>10 μM).
In addition to these results and to the potential applications of the quinazoline core in anticancer drugs, complementary studies were carried out. We studied the impact of an amino group in C-2 position of the quinazoline core for increasing the solubility of compounds in order to improve affinity for EGFR and/or VEGFR-2. For this reason, a series of 2-amino-4-anilinoquinazolines was designed, synthesized and biologically assessed (Figure 1).
The synthesis of the 4-chloro-2-aminoquinazoline (7) is illustrated in Figure 2. Reaction of commercial methyl 4,5-dimethoxyanthranilate with cyanamide using concentrated HCl as catalyst gave the compound 6 . The quinazolinone(6) was converted to 4-chloro-2-amino-quinazoline (7) with POCl3 in 7% yield . The low yield of this step was probably due to the reaction between 2-amino group and POCl3. Protection of the 2-amino moiety was therefore considered using the pivalic anhydride in the presence of DMAP and Et3N to give compound 8 with good yield . Chlorination with POCl3 of protected quinazolinone (8) led to the key intermediate 4-chloroquinazoline (9) in 75% yield.
As depicted in Figure 3, the synthesis of (aminophenyl)-carbamic acid esters 10-12 was performed by a one-pot reduction procedure of the corresponding nitrophenylisocyanate in the presence of various alcohols in CH2Cl2. This approach has been reported to be efficient and provided good yields and short reaction times . Displacement of the chloro substituent of compound 8 with commercial or previously synthesized anilines in refluxing isopropanol led to the corresponding 4-anilinoquinazolines [15-17]. Deprotection of the 2-pivaloyl group was performed with sodium thiomethoxide in methanol to provide the target compounds (19-24).
The N-(methyl-anilino)-6,7-dimethoxyquinazoline 25 was synthesized in two steps (Figure 4). First, the methylation of protected compound 18 in the presence of iodomethane and sodium hydride in N,N-dimethylformamide allowed to obtain N-(methyl-anilino) quinazoline intermediary. Deprotection of the 2-pivaloyl group was performed with sodium thiomethoxide in methanol to provide the target compound 25.
Biological assays and methods
In vitro kinase assays: Kinase assays were performed in 96-well plates (Multiscreen Durapore. Millipore) using [γ-32P]ATP (Perkin Elmer) and the synthetic polymer poly(Glu4/Tyr) (Sigma Chemicals) as a phosphoracceptor substrate. Tested compounds were dissolved in DMSO. The final concentration of DMSO in assay solutions was 0.1%, which was shown to have no effect on kinase activity.
a) EGFR tyrosine kinase activity: 20 ng of EGFR (purified from human carcinoma A431 cells, Sigma Chemicals) were incubated for 1 h at 28 °C using various concentrations of tested compounds in kinase buffer (HEPES 50 mM pH 7.5, BSA 0.1 mg/mL, MnCl2 10 mM. MgCl2 5 mM, Na3VO4 100 μM, DTT 0.5 mM, poly(Glu4/Tyr) 250 μg/mL, ATP 5 μM. [γ-32P]ATP 0.5 μCi).
b) VEGFR-2 tyrosine kinase activity: 10 ng of VEGFR-2 (Recombinant Human Protein, Invitrogen) were incubated for 1 h at 28°C using various concentrations of tested compounds in kinase buffer (Tris 50 mM pH 7.5, BSA 25 μg/mL, MnCl2 1.5 mM, MgCl2 10 mM, DTT 2.5 mM, Na3VO4 100 μM, ß-glycerophosphate 5 mM, poly(Glu4/Tyr) 250 μg/mL, ATP 5 μM. [γ-32P]ATP 0.5 μCi).
The reaction was stopped by adding 20 μL of trichloroacetic acid, 100%. Wells were washed 10 times with trichloroacetic acid, 10%. Plates were counted in a Top Count (Perkin Elmer) for 1 min per well.
Cells and materials: Human PC3 prostate cancer cells, MCF7 breast cancer cells and HT29 colon cancer cells were obtained from European Collection of Cell Cultures. They were cultured, at 37°C in a CO2 incubator, respectively in RPMI-1640, MEM and DMEM + Glutamax-I medium supplemented with 10% fetal bovine serum, penicillin (100 IU/mL) and streptomycin (100 μg/mL). FBS, medium, penicillin-streptomycin, and other agents used in cell culture studies were purchased from Invitrogen.
Cell proliferation: Briefly, cells were plated at a density of 3×103 cells/well in 96-well plates for 24h. Then the medium was removed, and cells were treated with either DMSO as a control or various concentrations of compounds. The final concentration of DMSO in the medium was <0.1% (v/v). After the cells were incubated for 72h, cell growth was estimated by colorimetric MTS test.
EGF stimulation: Briefly, HT29 cells were plated at a density of 3×103 cells/well in 96-well plates for 24h in DMEM medium supplemented with BSA 1%. Then, cells were treated with either DMSO as a control or compounds (1 μM). The final concentration of DMSO in the medium was <0.1% (v/v). After one hour, cells are stimulated with EGF (30 ng/mL), FBS (10%) or left untreated (BSA 1%). Then, the cells were incubated for 72h and cell growth was estimated by colorimetric MTS test.
Absorption spectroscopy and melting experiments studies: Absorption spectra and melting curves were obtained using an Uvikon 943 spectrophotometer coupled to a Neslab RTE111 cryostat. To perform the melting temperature (Tm) measurements, ctDNA was incubated alone (control Tm), then, each compound was incubated with ctDNA at a drug/ctDNA ratio of 1. Typically, 20 μM of the various drugs were prepared in 1 mL of BPE buffer (6 mM Na2PO4, 2 mM NaH2PO4, 1 mM EDTA, pH 7.1) in the presence of 20 μM of ctDNA and transferred into a quartz cuvette of 10 mm path length. The absorbance at 260 nm was measured every minute over the range 20-100°C with an increment of 1°C per minute and the spectra were recorded from 230 to 500 nm and referenced against a cuvette containing the same DNA concentration in the same buffer. The “melting” temperature Tm, deduced from melting curve, was taken as the midpoint of the hyperchromic transition and corresponds to the temperature for which half of the ctDNA is in single strand form and the other stays in the double strand form. DNA titrations were performed in the same buffer. To 1 mL of drug solution at 20 μM were added aliquots of a concentrated ctDNA solution (0-250 μM).
Circular dichroism: CD spectra were recorded on a Jasco J-810 spectrometer. Solutions of drugs, nucleic acids, and their complexes (1 mL in a 1 mM sodium cacodylate buffer, pH 7.0) were scanned in 10 mm quartz cuvettes. Measurements were made by progressive dilution of drug-DNA complex at a high P/D (phosphate/drug) ratio with a pure ligand solution to yield the desired drug/DNA ratio. Four scans were accumulated and automatically averaged. Then 1 mL of drug solution at 50 μM was successively diluted with increased volumes of DNA at 200 μM.
Fluorescence titration experiments: Fluorescence titration data were recorded at room temperature using SPEX Fluorolog fluorometer. As 2-aminoquinazoline derivatives show weak fluorescence intensity variation with DNA titration, the binding studies were carried out through a competitive displacement fluorometry assay using DNAbound ethidium bromide [18,19]. Excitation was set at 546 nm, and the fluorescence emission was monitored over the range 520-650 nm. Experiments were performed with an ethidium bromide/DNA molar ratio of 1.26:1 and a drug concentration range of 0.1-32 μM in a BPE buffer, pH 7.1. IC50 values for ethidium bromide displacement were calculated using a fitting function incorporated into GraphPad Prism 3.0 software, and the apparent binding constant was calculated as follows: Kapp=(1.26/IC50)Kethidium, with Kethidium=107 M-1.
Melting points were determined with a Büchi 535 capillary melting point apparatus and are uncorrected. Macherey Nagel Polygram® sil /G/UV254 commercial plates were used for analytical TLC as well as UV light and/or with iodine to follow the course of the reaction. Flash chromatography (FC) was performed with silica gel Macherey Nagel Si 60, 0.015−0.040 mm. The structure of each compound was confirmed by IR (Bruker, FT-IR spectrometer ALPHA) and 1H NMR (300 MHz, Bruker AC300P spectrometer). Chemicals shifts (δ) are reported in parts per million downfield from TMS. J values are in hertz, and the splitting patterns are abbreviated as follows: s, singlet; d, doublet; t, triplet; q, quartet; m, multiplet. The purity of the compounds was tested by HPLCfollowed by APCI+ (atmospheric pressure chemical ionization) mass spectral detection on an LC-MS system, ThermoElecton Surveyor MSQ, and was >97%. HRMS experiments were performed on Q ExactiveBenchtop LC-MS (Thermo Scientific).
2-amino-6,7-dimethoxyquinazolin-4-one hydrochloride (6): Concentrated HCl (0.05 mL) was added dropwise to a solution of methyl 2-amino-4,5-dimethoxybenzoate (1 g, 4.73 mmol) and cyanamide (0.32 g, 7.57 mmol) in dioxane (40 mL) at room temperature. The mixture was heated at 80°C for 24 h and then cooled to room temperature. The precipitated was collected by filtration as a white solid (96%); mp>250°C. νmax/cm-1 3193 (NH3 +), 1634 (C=O). 1H NMR (DMSO-d6) δ ppm 3.80 (s, 3H, OCH3), 3.85 (s, 3H, OCH3), 6.82 (s, 1H, ArH), 7.21 (br s, 2H, NH2), 7.28 (s, 1H, ArH), 12.00 (br s, 1H, NH+). LC-MS (APCI+), calcd for C10H11N3O3, m/z: 222 [(M + H)+].
2-amino-4-chloro-6,7-dimethoxyquinazoline (7): A solution of 2-amino-6,7-dimethoxyquinazolin-4-one hydrochloride (6) (1 g, 3.88 mmol) in phosphorus oxychloride (20 mL) was refluxed for 6 h. After removal of the solvent, the residue was dissolved in ice-water (50 mL) and the mixture was neutralized by 10% K2CO3 solution. The precipitate was collected by filtration and dissolved in CH2Cl2 (100 mL). The organic layer was washed with 10% K2CO3 solution, brine and dried over CaCl2, and the solvent was removed under reduced pressure to give (7) as a white solid (7%); mp 194-196°C. νmax/cm-1 3105 (NH2). 1H NMR (CDCl3) δ ppm 4.00 (s, 3H, OCH3), 4.05 (s, 3H, OCH3), 5.10 (br s, 2H, NH2), 6.95 (s, 1H, ArH), 7.25 (s, 1H, ArH). LC-MS (APCI+), calcd for C10H11N3O3, m/z: 240 [(M + H)+for 35Cl] and 242 [(M + H)+ for 37Cl].
2-pivaloylamino-6,7-dimethoxyquinazolin-4-one (8): To a solution of 2-amino-6,7-dimethoxy-quinazolin-4-one hydrochloride (6) (2 g, 9.04 mmol) in DMF (40 mL) were added pivalic anhydride (5.05 g, 27.1 mmol), DMAP (0.05 g, 0.45 mmol), Et3N (4.57g, 45.2 mmol). The mixture was stirred at 60°C for 2 h and the solvent was removed under reduced pressure. The residue was dissolved in water and the mixture was neutralized by 10% K2CO3. The precipitate was collected by filtration and dried in vacuo to give (8) as a white solid (92%); mp 159-161°C. νmax/cm-1 3050 (NH), 1672 (C=O). 1H NMR (DMSO-d6) δ ppm 1.30 (s, 9H, t-Bu), 3.80 (s, 3H, OCH3), 3.82 (s, 3H, OCH3), 6.90 (s, 1H, ArH), 7.40 (s, 1H, ArH), 11.00 (br s, 1H, NH), 12.00 (br s, 1H, NH). LC-MS (APCI+), calcd for C15H19N3O4, m/z: 306 [(M + H)+].
2-pivaloylamino-4-chloro-6,7-dimethoxyquinazoline (9): A solution of 2-pivaloylamino-6,7-dimethoxyquinazolin-4-one (8) (2.55 g, 8.35 mmol) in phosphorus oxychloride (20 mL) was refluxed for 6 h. After removal of the solvent, the residue was dissolved in ice-water (50 mL) and the mixture was neutralized by 10% K2CO3 solution. The precipitate was collected by filtration and dissolved in CH2Cl2 (100 mL). The organic layer was washed with 10% K2CO3 solution, brine and dried over CaCl2, and the solvent was removed under reduced pressure to give (9) as a white solid (75%); mp 139-141°C. νmax/cm-1 3042 (NH), 1675 (C=O). 1H NMR (DMSO-d6) δ ppm 1.30 (s, 9H, t-Bu), 4.00 (s, 3H, OCH3), 4.05 (s, 3H, OCH3), 7.20 (s, 1H, ArH), 7.40 (s, 1H, ArH), 10.20 (br s, 1H, NH). LC-MS (APCI+), calcd for C15H18ClN3O3, m/z: 324 [(M + H)+ for 35Cl] and 326 [(M + H)+ for 37Cl].
General procedure for (aminophenyl)carbamic acid esters (10-12): To a solution of corresponding commercial p-nitrophenylisocyanate (1 g) in 50 mL of a mixture of alcohol (MeOH or EtOH) and dichloromethane (5:5), was added Raney® nickel. After 18 h at room temperature in hydrogen atmosphere, the mixture was filtered. The solvant was removed under reduced pressure and the residue was purified by FC (CH2Cl2 /AcOEt, 9:1).
N-(4-aminophenyl)carbamic acid methyl ester (10): Beige solid (50%); mp 81-83°C. νmax/cm-1 3385 (NH2), 1711 (C=O), 1265 (NH).1H NMR (DMSO-d6) δ ppm 3.60 (s, 3H, CH3), 4.70 (br s, 2H, NH2), 6.50 (d, 2H, J=8.2 Hz, 2 ArH), 7.00 (d, 2H, J=8.2 Hz, 2 ArH), 9.00 (br s, 1H, NH). LC-MS (APCI+), calcd for C8H10N2O2, m/z: 167 [(M + H)+].
N-(3-methyl-4-aminophenyl)carbamic acid ethyl ester (11): Brown solid (97%); mp 90-93°C.νmax/cm-13382 (NH2), 1710 (C=O), 1265 (NH).1H NMR (DMSO-d6) δ ppm 1.22 (t, 3H, J=7.3 Hz, CH3), 2.19 (s, 3H, CH3), 4.11 (q, 2H, J=7.3 Hz, CH2), 5.33 (br s, 2H, NH2), 7.12 (m, 2H, 2 ArH), 7.45 (d, 1H, J=9.0 Hz, ArH), 8.97 (br s, 1H, NH). LC-MS (APCI+), calcd for C10H14N2O2, m/z: 195 [(M + H)+].
N-(3-chloro-4-aminophenyl)carbamic acid ethyl ester (12): Beige solid (84%); mp 109-111°C.νmax/cm-13385 (NH2), 1711 (C=O), 1264 (NH).1H NMR (DMSO-d6) δ ppm 1.28 (t, 3H, J=7.3 Hz, CH3), 4.05 (q, 2H, J=7.3 Hz, CH2), 5.30 (br s, 2H, NH2), 6.46 (dd, 1H, J=2.3 and 8.7 Hz, ArH), 6.61 (d, 1H, J=2.3 Hz, ArH), 6.97 (d, 1H, J=8.7 Hz, ArH), 8.52 (br s, 1H, NH). LC-MS (APCI+), calcd for C9H11ClN2O2, m/z: 215 [(M + H)+ for 35Cl] and 217 [(M + H)+ for 37Cl].
General procedure for 2-amino-4-anilino-6,7- dimethoxyquinazoline derivatives (19-24): To a solution of 2-pivaloylamino-4-chloro-6,7-dimethoxyquinazoline(8) (1 eq) in isopropanol, was added corresponding anilino derivatives (1,2 eq). After 2 h at reflux, the mixture was filtered. The obtained residue was stirred in methanol with sodium thiomethoxide at reflux for 2 h. The solvant was removed under reduced pressure. The residue was dissolved in water and the precipitate was collected by filtration and recristallized to give desired compound.
2-amino-4-(3-bromoanilino)-6,7-dimethoxyquinazoline (19): Crystallization from acetonitrile gave pure 19 as white solid (91%); mp> 250°C. νmax/cm-1 3190 (NH2). 1H NMR (DMSO-d6) δ ppm 3.82 (s, 3H, OCH3), 3.88 (s, 3H, OCH3), 6.05 (br s, 2H, NH2), 6.71 (s, 1H, ArH), 7.35 (m, 2H, 2 ArH), 7.63 (s, 1H, ArH), 8.05 (m, 2H, 2 ArH), 9.12 (br s, 1H, NH). LC-MS (APCI+), calcd for C16H15BrN4O2, m/z: 375 [(M + H)+ for 79Br] and 377 [(M + H)+ for 81Br]. HRMS (ESI (M+H)+ m/z) calcd for C16H15BrN4O2 375.0451 found 375.0446.
2-amino-4-(3-bromo-4-methylanilino)-6,7- dimethoxyquinazoline (20): Crystallization from ethanol gave pure 20 as white solid (25%); mp> 250°C; νmax/cm-1 3193 (NH2). 1H NMR (DMSO-d6) δ ppm 2.30 (s, 3H, CH3), 3.80 (s, 3H, OCH3), 3.90 (s, 3H, OCH3), 6.01 (br s, 2H, NH2), 6.75 (s, 1H, ArH), 7.35 (d, 1H, J=6.1 Hz, ArH), 7.66 (s, 1H, ArH), 8.01 (m, 2H, 2 ArH), 9.10 (br s, 1H, NH). LC-MS (APCI+ ), calcd for C17H17BrN4O2, m/z: 389 [(M + H)+for 79Br] and 391 [(M + H)+ for 81Br]. HRMS (ESI (M+H)+ m/z) calcd for C17H17BrN4O2 389.0608 found 389.0606.
2-amino-4-(3-chloro-4-fluoroanilino)-6,7-dimethoxyquinazoline (21): Crystallization from acetonitrile gave pure 21 as white solid (17%); mp>250°C. νmax/cm-1 3202 (NH2). 1H NMR (DMSO-d6) δ ppm 3.82 (s, 3H, OCH3), 3.83 (s, 3H, OCH3), 6.08 (br s, 2H, NH2), 6.75 (s, 1H, ArH), 7.38 (m, 1H, ArH), 7.62 (s, 1H, ArH), 7.91 (m, 1H, ArH), 8.12 (m, H, ArH), 9.18 (br s, 1H, NH). LC-MS (APCI+), calcd for C16H14ClFN4O2, m/z: 349 [(M + H)+for 35Cl] and 351 [(M + H)+ for 37Cl]. HRMS (ESI (M+H)+ m/z) calcd for C16H14ClFN4O2 349.0862 found 349.0859.
N-[4-(2-amino-6,7-dimethoxyquinazolin-4-ylamino)phenyl] carbamic acid methyl ester (22): Crystallization from ethanol gave pure 22 as white solid (90%); mp 197-199°C. νmax/cm-1 3356 (NH2), 3304 (NH), 1710 (C=O). 1H NMR (DMSO-d6) δ ppm 3.45 (s, 3H, CH3), 3.95 (s, 3H, OCH3), 4.00 (s, 3H, OCH3), 6.01 (br s, 2H, NH2), 7.50 (d, 2H, J=8.9 Hz, 2 ArH), 7.55 (s, 1H, ArH), 7.65 (d, 2H, J=8.9 Hz, 2 ArH), 8.20 (s, 1H, ArH), 9.10 (br s, 1H, NH), 9.50 (br s, 1H, NH). LC-MS (APCI+), calcd for C18H19N5O4, m/z: 370 [(M + H)+]. HRMS (ESI (M+H)+ m/z) calcd for C18H19N5O4 370.1510 found 370.1509.
N-[2-methyl-4-(2-amino-6,7-dimethoxyquinazolin-4-ylamino) phenyl]carbamic acid ethyl ester (23): Crystallization from acetonitrile gave pure 23 as white solid (90%); mp> 250°C. νmax/cm-1 3384 (NH2), 3305 (NH), 1713 (C=O).1H NMR (DMSO-d6) δ ppm 1.25 (t, 3H, J=7.6 Hz, CH3), 2.20 (s, 3H, CH3), 3.90 (s, 3H, OCH3), 3.95 (s, 3H, OCH3), 4.15 (q, 2H, J=7.6 Hz, CH2), 6.02 (br s, 2H, NH2), 7.00 (s, 1H, ArH), 7.45 (Dd, 1H, J=3.2 and 9.0 Hz, ArH), 7.50 (D, 1H, J=9.0 Hz, ArH), 8.10 (s, 1H, ArH), 8.90 (d, 1H, J=3.2 Hz, ArH), 10.60 (br s, 1H, NH), 12.75 (br s, 1H, NH). LC-MS (APCI+), calcd for C20H23N5>O4, m/z: 398 [(M + H)+]. HRMS (ESI (M+H)+ m/z) calcd for C20H23N5>O4 398.1823 found 398.1821.
N-[2-chloro-4-(2-amino-6,7-dimethoxyquinazolin-4-ylamino) phenyl]carbamic acid ethyl ester (24): Crystallization from acetonitrile gave pure 24 as white solid (65%); mp> 250°C. νmax/cm-1 3380 (NH2), 3304 (NH), 1707 (C=O).1H NMR (DMSO-d6) δ ppm 1.20 (t, 3H, J=7.6 Hz, CH3), 3.80 (s, 3H, OCH3), 3.85 (s, 3H, OCH3), 4.10 (q, 2H, J=7.6 Hz, CH2, ArH), 6.00 (br s, 2H, NH2), 6.70 (s, 1H, ArH), 7.40 (D, 1H, J=8.9 Hz, ArH), 7.60 (s, 1H, ArH), 7.90 (Dd, 1H, J=3.1 and 8.5 Hz), 8.00 (d, 1H, J=2.9 Hz, ArH), 8.90 (br s, 1H, NH), 9.15 (br s, 1H, NH). LC-MS (APCI+), calcd for C19H20ClN5O4, m/z: 418 [(M + H)+ for 35Cl] and 420 [(M + H)+ for 37Cl]. HRMS (ESI (M+H)+ m/z) calcd for C19H20ClN5O4 418.1277 found 418.1274.
N-[2-chloro-4-(2-amino-6,7-dimethoxyquinazolin-4-N- (methyl)ylamino)phenyl]carbamic acid ethyl ester (25): A mixture of compound 18 (1 eq) and sodium hydride (60% in oil) (2 eq) in N,Ndimethylformamide (10 mL) was stirred 5 h at room temperature in a nitrogen atmosphere. Iodomethane (2 eq) was added, and the mixture was stirred for 16 h. The reaction was quenched by water, and then the aqueous solution was extracted with EtOAc, washed with a solution of K2CO3 10%, and dried over MgSO4. The solvent was removed under reduced pressure. The obtained residue was stirred in methanol with sodium thiomethoxide at reflux for 2 h. The solvant was removed under reduced pressure. The residue was dissolved in water and the precipitate was collected by filtration and recristallized in methanol to give 25 as a white solid (47%); mp>250°C. νmax/cm-1 3321 (NH2), 3310 (NH), 1792 (C=O).1H NMR (DMSO-d6) δ ppm 1.22 (t, 3H, J=7.6 Hz, CH3), 3.80 (s, 3H, OCH3), 3.84 (s, 3H, OCH3), 4.12 (q, 2H, J=7.6 Hz, CH2), 6.03 (br s, 2H, NH), 6.73 (s, 1H, ArH), 7.88 (m, 1H, ArH), 7.68 (s, 1H, ArH), 8.02 (m, 2H, 2 ArH), 9.33 (br s, 1H, NH). LC-MS (APCI+), calcd for C20H22ClN5O4, m/z: 432 [(M + H)+ for 35Cl] and 434 [(M + H)+ for 37Cl]. HRMS (ESI (M+H)+ m/z) calcd forC20H22ClN5O4 432.1433 found 432.1433.
Enzymatic inhibition and in vitro antiproliferative activity
Inhibitory activity against EGFR and VEGFR-2 tyrosine kinases of synthesized compounds was determined by measuring the in vitro levels of phosphorylation of the tyrosine-specific peptides (poly(Glu- Tyr)substrate) using [γ-32P]ATP. The compounds were also evaluated for their antiproliferative activities towards three cancer cell lines: the hormone-independent PC3 prostate cancer cells, the MCF7 breast cancer cells and the HT29 colon cancer cells, by MTS assay during 72 hours. Antiproliferative activities of compounds were compared to vandetanib, a dual EGFR/VEGFR-2 inhibitor. The data are summarized in Table 2.
|Cell growth inhibitory
|19||NH2||CH3||1.80||>10||>10||3.7 ±0.78||6.0 ±1.67|
|20||NH2||CH3||2.64||>10||4.8 ±1.20||3.1 ±0.89||3.6 ±0.62|
|2||3-Cl,4-F||H||CH3||0.40||5.30||6.6 ±1.70||6.7 ±0.78||4.9 ±0.54|
|21||NH2||CH3||0.13||>10||>10||6.2 ±0.28||4.6 ±0.55|
|22||NH2||CH3||>10||>10||7.6 ±0.57||7.1 ±0.61||8.6 ±0.95|
|23||NH2||CH3||>10||>10||>10||7.2 ±0.71||7.0 ±0.91|
|24||NH2||CH3||>10||>10||6.0 ±1.25||3.6 ±2.51||2.0±0.30|
aThe cell growth rate was evaluated by performing the MTS assay. Data are represented as the mean ± SEM of at three experiments performed in triplicate. Higher concentrations were not used to avoid precipitation of the compound in the culture medium
bInhibition of EGFR (purified from human arcinoma A431 cells) tyrosine kinase activity
cInhibition of VEGFR-2 (recombinant protein) tyrosine kinase activity
Table 2: Enzymatic and cell growth inhibition results of 2-aminoquinazoline derivatives.
Incorporation of an amino group in C-2 position of 4-anilinoquinazolines substituted by one or two halogens increased antiproliferative activities of compounds despite a slight decrease of EGFR inhibition. Indeed, the halogenated compounds (19, 20 and 21) showed a good affinity for EGF receptor and an interesting inhibition of proliferation for the three studied cancer cell lines. A complementary study was performed to show the correlation between antiproliferative activity of the halogenated compound 21 (EGFR IC50=0.13 μM) and the EGFR inhibition. Indeed, the proliferation of HT29 cells was studied on several batches of cells: (a) control cells, (b) cells stimulated by Fetal Bovine Serum (FBS) containing many growth factors including EGF and (c) cells stimulated by growth factor EGF (Figure 5).
Figure 5: Percentage of HT29 cells untreated or treated by inhibitor (PD153035 and compound 21) under different conditions (FBS or EGF stimulation). The cell growth rate was evaluated by performing the MTS assay. Data are represented as the mean ± SEM of at three experiments performed in triplicate (***p<0.001).
As shown in Figure 5, the EGF inhibitor PD153035, used as reference compound, and the compound 21 inhibited only the EGF-mediated cellular proliferation. As observed in Figure 5a, cell proliferation was not reduced by these inhibitors without FBS or EGF stimulation. So, the inhibition of cellular proliferation by the halogenated compound 21 could be due to its potential EGFR inhibition.
The introduction of an amino group in C-2 position of the quinazoline core in compounds 22, 23 and 24 led to a decrease of EGFR and VEGFR-2 inhibition (IC50>10 μM) and an increase of cell growth inhibition of three cancer cell lines. The 2-amino-4-anilinoquinazoline 24, with an ethyl carbamate in para and a chlorine in meta position on aniline, showed the best antiproliferative activities on PC3 (6.0 μM), HT29 (3.6 μM) and MCF7 (2.0 μM). So, we also explored several studies to explain the antiproliferative effect of compound 24. These experiments are described in the following section.
DNA binding properties
A previous study, performed in our team demonstrated that 4-(N-methyl)-anilinoquinazoline derivatives bound to ctDNA (calf thymus DNA) by an intercalative process [9,16]. Similarly we studied the capacity of the 2-aminoquinazoline 24 to bind to the DNA. To evaluate the role of the amino group in C-2 position in the interaction with ctDNA, we chose to test the 4-anilinoquinazoline 5 without amino group in C-2 position as reference. As shown previously , N-methylation increased interaction with ctDNA, so we also assessed the N-methyl analogue of 24 (compound 25), and an intercalator agent (EBE-A22), synthesized by us .
Therefore, we engaged different methods to study the binding mode to ctDNA for the newly synthesized compounds: DNA thermal denaturation, UV-visible spectroscopy, circular dichroism and fluorescence measurements.
DNA thermal denaturation experiments: The ability of the drugs to protect ctDNA against thermal denaturation was used as an indicator of their capacity to bind and to stabilize the DNA double helix. The ΔTm values (ΔTm=TmDrug-DNAcomplex– TmDNAalone) are presented in Table 3.
|24||NH2||H||3-Cl-4-NHCOOC2H5||8.1 ± 0.8|
|25||NH2||CH3||3-Cl-4-NHCOOC2H5||6.3 ± 0.6|
|EBE-A22||H||CH3||3-Br||8.1 ± 0.8|
Variations in melting temperature (ΔTm=TmDrug-DNAcomplex– TmDNAalone) at drug/ctDNA ratio of 1.
Means values are calculated from at least two separate experiments with typical standard deviations of less than ± 10%.
Table 3: DNA thermal denaturation results.
The compound 24 show a ΔTm value of 8.1°C which indicates a stabilization of DNA against heat denaturation. This compound displays the same value that the reference intercalator compound EBE-A22. The comparison of results obtained for compounds 5 and 24 shows that incorporation of an amino group in C-2 position on 4-anilinoquinazolines is considered as a positive element to observe DNA interaction, probably due to additional electrostatic interactions involving DNA phosphodiesters groups. Contrary to previous results , the N-methylation in 4-position of the 2-amino-4- anilinoquinazoline did not improve the DNA interaction (25 vs. 24).
UV-Visible spectroscopy: The binding of four compounds was studied by UV-visible spectroscopy. Figure 6 displays UV absorption measurements recorded upon ctDNA titration into a bufferedaqueous solution.
Figure 6: UV-visible spectra of drug at the concentration of 20 µM upon addition of ctDNA (0-250 µM) in BPE buffer (pH 7.1). Figure depicting the fitting plots obtained between 300 and 420 nm. Arrows correspond to hypochromic (↓) and bathochromic (→) effects, respectively. Red rings localize the isobestic point.
Addition of ctDNA induced few marked changes of the absorption spectrum of the compound 5 considered as a negative control. The spectrum of the 2-amino analogue (24) varied with DNA titration as illustrated by bathochromic and hypochromic shifts and an absorption maximum red-shifted from 350 to 370 nm. An isobestic point at 382 nm was detected pointing to the existence of a single binding mode. The UV-visible spectrum of compound 25 showed also bathochromic and hypochromic shifts but no clear isobestic point was spotted.
Circular dichroism (CD): Preliminary evaluations showed a potential interaction between the compounds 24 or 25 and ctDNA. To confirm these results and to define more precisely the binding process, a spectroscopic method utilizing polarized light and CD measurements was performed. The CD spectra of ctDNA or studied compounds are characterized by no change of ellipticity (O) between 320 and 420 nm (data not shown). The intensity changes of ellipticity are usually the result of the interaction of drugs with ctDNA. Solutions of DNA in the presence of drugs were analyzed by CD (Figure 7).
As depicted in Figure 7, no variation of ellipticity was observed on the spectrum of the compound 5 in the presence of DNA, pointing no interaction between this compound and ctDNA. However, for compounds 24, 25 or EBE-A22, variations of ellipticity were observed in the spectral range 320-420 nm. CD measurements of EBE-A22 and 24 showed a similar negative band centered at 352 and 362 nm respectively. This typical negative CD is consistent with an intercalative DNA binding mode. The N-methylation of 24 induced a failover of CD signal. Indeed, a positive band centered at 375 nm was observed on the spectrum of 25. This intensity change could be the result of the interaction of 25 with the DNA minor groove .
Fluorescence measurements: DNA binding affinity of compound 24 was quantified by means of fluorescence Kapp (Figure 8). As a weak fluorescence was observed upon DNA titration, an indirect method was privileged.18,19 We used the conventional fluorescence quenching assay based on DNA binding competition between the intercalating drug ethidium bromide (BET) and the tested molecule. Indeed, BET is an efficient DNA intercalator and its interaction with DNA causes highly fluorescence emission. The binding ability of drugs to form a drug-DNA complex can be determined from the displacement of BET, causing a quenching of the fluorescence signal due to the dissolved oxygen in aqueous solution.
The fluorescence quenching shown in Figure 8 revealed that the compound 24 was able to displace BET by competitive binding in the same DNA intercalation site. Fluorescence variations induced by addition of drug allowed the determination of the apparent binding constant (Kapp) of 24 using nonlinear least-squares analysis (GraphPad Prism software): Kapp =1.39×106 ± 0.07×106 M-1. This data attests that 2-aminoquinazoline 24 is a potent DNA intercalating agent. For the compounds 5 and 25, fluorescence variation was too weak to calculate the apparent binding constant (Supplementary file).
We have studied the binding mode of several compounds by DNA thermal denaturation, UV-visible spectroscopy, circular dichroism and fluorescence measurements. The obtained results revealed that the compound 24 presented a good affinity for DNA and led to a duplex stabilization (Table of Content).
In conclusion, a series of 2-aminoquinazolines substituted by different anilino groups in C-4 position was designed and synthesized. Most of the synthesized compounds inhibited the proliferation of the three studied cancer cell lines (PC3, prostate cancer; MCF7, breast cancer and HT29, colon cancer). Several studies were conducted to explain the antiproliferative effect. 2-Aminoquinazolines substituted in C-4 position by halogeno-aniline (19, 20 and 21) showed an EGFR tyrosine kinase inhibition with IC50 values at micromolar range, in correlation with their antiproliferative activities. Substitution of anilino group by a bulky group like carbamate triggered a loss of tyrosine kinase inhibition. However, the antiproliferative activities of carbamate derivatives could be explained by their interaction with DNA highlighted by several studies: DNA thermal denaturation, UV-visible spectroscopy, circular dichroism and fluorescence measurements. The various spectra revealed that compound 24 is a DNA intercalating agent. However, the affinity of the 2-amino-4-anilinoquinazoline 24 for DNA is lower compared to classical potent intercalators such ethidium bromide. Nevertheless, the DNA-binding studies confirmed that the 2-amino-4-anilinoquinazoline core is a suitable skeleton, allowing an effective DNA intercalative process. Further development for this series should include a series of 4-anilino-benzo-(1,2,3)-triazines or 2-aminomethyl-4-anilinoquinazolines.
Here, we report the highlighting of 2-aminoquinazoline scaffold as DNA intercalating agents for cancer therapy
Kinase assays were performed on the Binding-Platform of ICPAL, University of Lille 2, supported by PRIM (Po^le de Recherche Interdisciplinaire pour le Me´dicament). We thank the LaRMN, Faculty of Pharmacy, University of Lille 2. The authors thank Céline Lenglart and Nathalie Duhal from the CUMA (Centre Universitaire de Mesures et d’Analyses), University of Lille 2, for mass spectrometry analysis.
Make the best use of Scientific Research and information from our 700 + peer reviewed, Open Access Journals