Design, Synthesis, and In-Vitro Biological Evaluation of PARP-1 Inhibitors Based on a 4-(Benzylideneamino)-N-(Quinolin-8-yl)Benzamide Scaffold

Abstract Novel poly(ADP-ribose)polymerase (PARP)-1 inhibitors containing, the 4-(benzylideneamino)-N-(quinolin-8-yl)benzamide moiety, were designed and synthesized. The docking study revealed that the designed compounds possess significant to moderate interaction with the targeted enzyme PARP1. Among them compound 3d (−52.04 K/cal) and 3e (−50.234 K/cal) showed similar Glidescore compared to Olaprib (−57.76 K/cal). Some of the synthesized compounds displayed good PARP-1 inhibitory activity, and among them, 3d and 3e were the most potent one. Enzyme inhibitory assay indicated that the compounds 3d, 3e, 3i and 3o exhibited an enzyme inhibitory activity against PARP-1 enzyme similar to that of olaparib. All the synthesized compounds were screened for their in vitro anticancer activity against MCF-7 and MDA-MB-232 cell lines. Among them, compound 3d (60.63 μg/mL and 56.38 μg/mL) and 3e (3 68.03 μg/mL and 54.42 μg/mL) were the most potent ones. In addition, ADMET prediction results indicated that these compounds might be less toxic and display more interesting pharmacokinetic properties.


Introduction
Chambon and colleagues discovered a family of enzymes called poly (ADP-ribose) polymerase in 1963, and which have the ability to catalyze the transfer of ADP ribose to a target protein. 1 This group of enzymes is involved in a variety of biological functions, including transcription, chromatin structure regulation, replication, recombination, and DNA repair. Until now, 18 members of the PARP enzyme family have been identified. 2 PARP1 is the most well-known of them all, with roles in DNA repair, cell proliferation, and cell death. Pharmacological inhibition of PARP has the potential to improve the cytotoxicity of anti-cancer therapies that cause DNA damage, minimize cell necrosis, and down-regulate numerous pathways in inflammatory and damaged tissues. As per various studies, the PARP1 activity is low in the majority of normal human cells, but it is highly unregulated in a number of primary cancer diseases, including lymphomas and cancers of the breast, uterus, and ovaries. 3,4 As a result of its critical role in DNA repair, among other things, the understanding of the function and importance of the PARP1 enzyme has grown in recent years, leading to the development of specific inhibitors in the drug development process for a variety of pathological conditions, including cancer, neurodegenerative disorders, and metabolic disorders. 5 Olaparib, niraparib, rucaparib, and talazoparib are a few PARP-1 inhibitors that have been approved by the FDA. Furthermore, phthalazinones like KU58948 and AZD 2461 have been identified as the most efficient orally accessible PARP inhibitors. According to recent studies, tremendous progress has been made in the field of target-based therapy. Hence this present work aims to developed a novel inhibitor for PARP1 enzyme. [6][7][8] The search for novel PARP-1 inhibitors with high efficacy, considerable water solubility, and adequate oral bioavailability continues. The development of a new family of PARP1 inhibitors is largely focused on boosting NAD þ substrate protein interaction with the enzyme via the nicotinamide moiety of NAD þ . Various investigations have demonstrated that the nicotinamide moiety of NAD þ is used in the design of PARP-1 inhibitors to increase the ligand-protein interaction of NAD þ with PARP-1. 9,10 Aromatic ring and carboxamide moiety, which can form hydrogen bonds and p-stacking interactions in PARP-1 catalytic domain, are common pharmacophoric properties required. S and N with an aromatic ring, carbonyl groups, and carbamide groups are among the characteristics. While interacting with PARP1 enzyme, these substituents are responsible for the creation of hydrogen bonds, stacking bonds, and different favorable interaction(s) that irreversibly restrict its activity. 11 The quinoline nucleus is a structural characteristic found in a wide range of synthesized compounds with a wide range of therapeutic activity. 12 It occurs as natural compounds like cinchona alkaloids and pharmacologically active substances displaying a broad range of biological activities such as anti-malarial, anti-microbial, antinflammatory, anti-convulsant, anti-cancer and antimycobacterial activity. Quinolones have been used in the research of bioorganic and bioorganometallic processes 13 in addition to their therapeutic applications. The quinoline-based compounds were designed as target compounds in this study because they have a broad pharmacological action. Figure 1 shows the various drug molecules containing quinoline moiety. The scaffold quinoline amine was further utilized for our study effort in light of the aforesaid findings and in keeping with our aim in developing novel heterocyclic compounds with anticancer potential. 14 Our research group has created a novel class of PARP1 inhibitors based on these findings, which include 4-(benzylideneamino)-N-(quinolin-8-yl)benzamide derivatives (Scheme 1). The proposed compounds were compared to Olaparib for their ADMET properties. 15 Figure 2 shows the chemical structures of newly designed PARP1 inhibitors.

Synthetic work
Based on SAR studies of previously available PARP-1 inhibitors such as Olaparib and Veliparib, the compounds 3a-3o were designed for synthesis. 4-nitro-N-(quinolin-8-yl) benzamide (1) was synthesized from the reaction between 8-aminoquinoline and 4-nitrobenzoylchloride in the presence of triethylamine and THF, which acted as a solvent for the reaction. The obtained product of compound 1 was treated with Fe powder and acetic acid to give the corresponding amine derivative of 4-amino-N-(quinolin-8-yl) benzamide (2). Further, compound 2 was reacted with different aromatic aldehydes in the presence of methanol. It gave corresponding 4-(benzylideneamino)-N-(quinolin-8-yl) benzamide derivatives (3a-3o, Scheme 1). All the compounds and intermediates were purified by successive recrystallization from ethanol. The IR spectrum of the final synthesized compounds showed absorption bands around 3300-3156 cm À1 for amide NH, while the distinguishing broad absorption peaks C ¼ O for CONH were observed in the range of 1720-1690 cm À1 , 3350-3157 cm À1 for NH, 1489-1464 cm À1 for CH 2 , 1379-1344 cm À1 for CH 3 , and 800-700 cm À1 for aromatic rings. These compounds also exhibited appropriate peaks at corresponding ppm in their 1 H NMR spectra. The 1H NMR spectra of the synthesized compounds revealed a singlet signal at 9.3-9.7 for H of NH, a singlet signal at 8-8.5 for H of ethene, and a signal at 7.5-8.5 for H of aromatic ring. The 13 C NMR spectra of synthesized compounds revealed a signal at 160-175 for carbonyl carbon, a signal at 150-160 for ethene carbon, and a signal at 120-145 for aromatic carbon. The corresponding molecular ion peaks in the LC-MS spectra were in conformity with the assigned structures. All the synthesized compounds were subjected to short-term in vitro cytotoxicity studies using breast cancer cell lines and docking studies.

Molecular docking
The in-silico docking study of the designed molecules (3a-3o) to the enzyme's active sites was performed by the Glide module of Schrodinger suit-2018-2 to determine the binding affinities of the ligands. The designed compounds were docked toward PARP1 (PDB ID: 4ZZZ) in order to   ascertain their PARP1 inhibition activity against breast cancer. The compounds 3a-3o exhibited good affinity for the receptor when compared with olaparib with PARP1 inhibitory activity as an anti-breast cancer agent. The Glide scores of docking studies against PARP1 inhibitor (PDB ID: 4ZZZ) are shown in Table 1. From the in-silico docking results, it is evident that the interactions are mainly lipophilic factors due to the presence of aromatic heterocyclic rings. Among the docked compounds, compound 3d possesses a similar glide score-52.04 K/cal compared to the standard drug olaparib-57.76 K/cal. The compound 3d shows one hydrogen bond between the hydroxyl group of the ligand and the amino acid GCL 2017 and it also shows pi-pi intraction between Tyr 907 and the quinoline moiety. The compound 3e shows a significant glide score of 50.234 K/cal along with 2 hydrogen bonds with amino acids Asp 770 and Ser 864. It also shows Pi-Pi intraction with his 862 and quinoline moiety. Following compound 3e, the compound 3i shows a good glide score of 50.408 K/cal along with one hydrogen bond with amino acid Asp 770. The remaining docked compound shows a glide score range from 35 to 48 K/cal along with one or two hydrogen bond interactions. Figures 3-7 show the docking pose of compounds 3d, 3e, 3i, and 3j and standard olaparib.

Insilico ADMET prediction
The insilico ADMET properties of the designed ligands (3a-3o) were determined by the Qikprop module of the Schr€ odinger suite 2018-2. The molecular weight of the designed compounds ranged between 341 and 450. The estimated number of hydrogen bond donors was in the range of 1-2. The estimated number of hydrogen bond acceptors was between 2.5-4.5. The predicted octanol/water partition co-efficient was in the range of 4.1 to 5.1 and the number of likely metabolic reactions was between 1-3. The number of violations of Lipinski's rule of five was 0-2. All  Table 2.

In-vitro cytotoxicity
The results of in vitro cytotocity activity of the compounds were expressed as IC 50 values, which were determined by plotting the percentage of cell viability versus concentration of sample on a logarithmic graph and reading the control. The experiments were performed in triplicates, and then, the final IC 50 values were calculated by taking an average of triplicate experimental results. The study was compared with the PARP-1 inhibitor, olaparib. The in vitro cytotoxicity study was carried out by the MTT assay method with cell lines (MDA-MB-231 and MCF-7). All the tested compounds displayed an IC 50 less than 250 lg/mL at a concentration range of 30-250 lg/mL. Among the tested compounds, compound substituted with bromo-salicyclic acid ((3d 5 60.63 lg/ mL and 56.38 lg/mL) and (3e 5 68.03 lg/mL and 54.42 lg/mL)) showed significant activity against both tested cell lines, MDA-MB-231 and MCF-7 respectively. The results of in vitro cytotoxicity studies are shown in Table 3.

PARP enzyme inhibition assay
The target compounds (3a-3o) were evaluated for their inhibitory activity against the PARP1 enzyme. Among these new derivatives, 3d (21.21 ± 4.1 nM) was found to inhibit PARP-1 catalytic domain most significantly, and then compound 3e (22.18 ± 4.2 nM) possessed significant PARP-1 inhibition. Compound 3d and 3e have more inhibitory activity than the remaining tested compounds. This may be due to the presence of more electronegative atoms (Br and OH) in the substitution. Of these compounds, compound 3i (33.44 ± 6.4 nM) and compound 3o (35.64 ± 5.4 nM) have moderate inhibitory activity against PARP1. The remaining tested compounds have moderate to good inhibitory activity against the PARP-1 enzyme. The result is shown in Table 4.

Conclusion
The physicochemical and spectroscopic data confirmed the structural integrity of the newly synthesized compounds. The investigated molecules displayed a similar manner to protein binding to the active site of PARP1 protein (PDB ID: 4ZZZ) in molecular docking studies. The calculated docking energies indicated that its interaction with PARP1 is favorable, but only to a limited extent. The enzyme inhibitory assay indicated that the enzyme inhibitory activities of 3d, 3e, 3i, and 3o against PARP-1 enzyme were similar to those of olaparib. All the synthesized compounds were screened for their in vitro cytotoxicity against MCF-7 and MDA-MB-232 cell lines. Compounds 3d and 3e emerged to be the most active compounds against both the breast cancer cell lines. In addition, ADMET prediction results indicated that these compounds might possess less toxicity and pharmacokinetic properties. The study thus serves as an attempt to progress toward the discovery of novel anticancer drugs. Additional derivatives may be prepared and further extended in-depth investigations into other cancer cells would be implemented to establish a SAR (Structural activity relationship) for rational study. From the present investigation, it may be concluded that 4-(benzylideneamino)-N-(quinolin-8-yl) benzamide derivatives need to undergo further investigation to develop as a potential candidate drug for cancer.

Materials
All the solvents (AR grade) and reagents were dried and purified. The reactions were performed on oven dried glassware. All reagents and solvents were obtained from the supplier or recrystallized/redistilled unless otherwise noted. The purity of the synthesized compounds was monitored by thin layer chromatography (TLC). The reaction progress was monitored by TLC using precoated silica 60 F254, 0.25 mm aluminum plates (Merck) and n-hexane: ethyl acetate (5:5) used as the mobile phase. The developed chromatogram was visualized by iodine vapor and melting points were determined in open capillary tubes. The melting points were recorded using Veego VMP-1 melting point apparatus and were uncorrected. Infrared (IR) spectra were recorded on a Perkin-Elmer Fourier Transform-Infrared Spectrometer using KBr pellets. Elemental analyses (C, H, and N) were done with a Shimadzu analyzer (Mumbai, Maharashtra, India) and all analyses  were consistent (within 0.4%) with theoretical values. The 1 H NMR and 13 C NMR spectra of synthesized compounds were recorded on a Bruker NMR Spectrometer (Billerica, MA, USA) at 400 MHz frequency in deuterated DMSO and CDCl 3 and using TMS as an internal standard (chemical shift in ppm). The mass spectra of some compounds were scanned on Shimadzu LC-MS/MS.

Synthesis of 4-nitro-N-(quinolin-8-yl)benzamide (1)
A solution of 8-aminoquinoline (1 mmol) and triethylamine (2 mmol) in dry THF (20 mL) was stirred for 15 min. To the reaction mixture, a solution of 4-nitrobenzoylchloride (2 mmol) in dry THF (20 mL) was added. The resulting mixture was stirred for 5 h at room temperature. The completion of the reaction is monitored by Thin Layer Chromatography (TLC) using ethyl acetate and n-hexane as mobile phases. After completion of the reaction, the reaction mixture was cooled at RT and filtered to afford crystals of 4-nitro-N-(quinolin-8-yl) benzamide.

Synthesis of 4-amino-N-(quinolin-8-yl)benzamide (2)
To a solution of 4-nitro-N-(quinolin-8-yl)benzamide (2.8 mmol) in ethanol (40 mL) was added Fe O (11.5 mmol) and acetic acid (2.6 mL). the resulting mixture was refluxed for about 13 hrs to produce a dark brown suspension. The completion of reaction is monitored by Thin Layer Chromatography (TLC) using ethyl acetate and n-hexane as mobile phase. After completion of reaction, the suspension was filtered through a pad of Celite. The filtrate was concentrated in vacuum and purified by column chromatography using ethyl acetate and n-hexane (3:7) as mobile phase.
5.1.3. General procedure for the synthesis of 4-(benzylideneamino)-N-(quinolin-8-yl)benzamide derivatives (3a-3o) A mixture of compound 2 (0.1 mmol) and the required aromatic aldehyde (0.1 mmol) in methanol (30 mL) was refluxed at 90 C for 6 h. The completion of the reaction was monitored by TLC using ethyl acetate: n-hexane (4:6) as the mobile phase. The resulting mixture was poured on crushed ice. The resulting solid was filtered then recrystallized using ethanol. . Crystallographic molecules of water have been removed, which includes all the water molecules without any interaction with the ligand and are more than 5 away from the ligand. The atoms missing in the side loop of the protein structure were added using Prime (Schrodinger 2018-2). For all amino acid residues, hydrogen bonds were added at pH 7.0 after taking the least ionization states into account. To alleviate the steric hindrance, the energy minimization was carried out till 0.30 value of Root Mean Square Deviation was attained using the OPLS-3 force field. The coordinates of co-crystals already present in the protein were used as the active site and the binding pocket was generated using a centroid grid box. The size of the pocket created is proportional to the size of the co-crystal. Compounds after being optimized using LigPrep were docked into the generated grid box using the 'extra-precision mode' of Glide (Schrodinger Suite 2018-2), avoiding all constraints. Then compounds were selected by analyzing Glide Scores. 14  with 10% FBS, penicillin (100 U/mL), and streptomycin (100 lg/mL). The cells were grown at 37 C with 5% CO 2 .

MTT assay
Based on the principle for MTT assay is the cleavage of the tetrazolium salt of MTT (3-(4,5 dimethyl thiazole-2yl)-2,5-diphenyl tetrazolium bromide). The number of viable cells is proportional to their ability to reduce the tetrazolium salt to formazan. DMEM media with 10% FBS was used to adjust the cell culture to 1.0105 cells/mL. 100 L of distilled cell suspension (about 10,000 cells/well) was added to each well of a 96 well flat bottom micro titer plate. After the cell population was determined to be sufficient after 24 hours, the cells were centrifuged, and the pellets were suspended in 100 L of various test sample concentrations made in maintenance media. The plates were incubated for 48 hours at 37 Cin a 5% CO 2 environment, with observations recorded every 24 hours. MTT (2 mg/mL) in MEM-PR (MEM without phenol red) was added after 48 hours. The plates were incubated at 37 C for 2 hours (5% CO 2 atmosphere). The 100 mL of DMSO was added and the plates were shaken to solubilize the formed formazan. The absorbance was measured using a microplate reader at a wavelength of 540 nm. The percentage of cell viability was calculated using the formula. 17,18 %Cell viability ¼ Mean OD of individual sample Â 100 Mean OD of control

PARP1 enzyme inhibition assay
According to the manufacturer's instructions, an enzyme assay was performed using a highly sensitive fluorescence test (HT-F Homogeneous Inhibition Assay; Trevigen). Damaged DNA was combined with escalating levels of NADþ (0-100 nM) and then incubated with recombinant human PARP-1 enzyme (expressed as a GST fusion protein in E. coli cells) for 30 minutes in the presence or absence of successive dilutions (dose-response curve) of examined drugs. The reactions were stopped by adding a resazurin stock solution, and after 30 minutes of incubation, the resazurin-dependent fluorescence was measured using a multiplate reader (Victor; Perkin Elmer).
Within each assay, a NAD standard curve was created. The ALLFIT programme was used to calculate inhibitor IC 50 values. 19