Design and synthesis of tetrahydrophthalimide derivatives as inhibitors of HIV-1 reverse transcriptase

Background Non-nucleoside reverse transcriptase inhibitors (NNRTIs) are one of the key components in highly active anti-retroviral therapy because of their high specificity and less toxicity. NNRTIs inhibit reverse transcriptase enzyme by binding to the allosteric site, which is 10Å away from the active site. Rapid emergence of resistance is the major problem with all anti-HIV agents. Hence, there is continuous need to develop novel anti-HIV agents active against both drug sensitive and resistance strains. Results All the 16 synthesized 2-(1,3-dioxo-3a,4-dihydro-1H-isoindol-2(3H,7H,7aH)-yl)-N-(substitutedphenyl) acetamide 4(a-p) analogs were characterized by Fourier transform infrared spectroscopy, proton nuclear magnetic resonance spectroscopy, mass spectroscopy, and elemental analysis. Lipinski rule of five parameters and molecular parameters like solubility, drug likeness, and drug score were derived for designed analogs using online servers like Molinspiration and Osiris property explorer. Synthesized compounds were evaluated for their HIV-1 reverse transcriptase inhibitor activity by HIV-1 RNA-dependent DNA polymerase activity assay at 2 and 20 μM concentrations. Conclusions Among the 16 synthesized compounds, 4a, 4b, 4f, 4g, 4k, and 4l showed weak reverse transcriptase inhibitor activity at 20 μM concentration. For the designed compounds, there was no correlation observed between molecular modeling and in vitro studies.


Background
Acquired immune deficiency syndrome (AIDS) is the advance stage of infection caused by the human immunodeficiency virus (HIV-1). AIDS and AIDS-ailed infections are major leading causes of death. According to UNAIDS-2012 report, 33 million people are living with AIDS and 1.7 million people died in the year 2011 [1]. Highly active antiretroviral therapy (HAART), a combination of two nucleotide or nucleoside reverse transcriptase inhibitor (NRTIs) and one protease inhibitor (PI), is generally used for AIDS. Alternative combinations like two NRTIs and one nonnucleoside reverse transcriptase inhibitor (NNRTI) or two NRTIs and one integrase inhibitor are used. NNRTIs are the key components in HAART because of their high potency, selectivity, and less toxicity when compared to NRTIs and PIs [2,3]. Currently five NNRTIs are approved by United States Food and Drug Administration. Among them, nevirapine, delavirdine, and efavirenz are first generation, which already got resistance. Etravirine and rilpivirine are potent and currently using second generation NNRTIs. However, the occurrence of the high mutation rate of the virus and the resulting emergence of resistance make the researchers run a never-ending marathon to keep developing new drugs active against both drug-sensitive and drug-resistance strain with better therapeutic profile [4,5].
Compounds having phthalimide scaffold exhibit antiinflammatory [8], anticancer [9], antibacterial [10], HIV-1 RT [11,12], and HIV-1 integrase inhibitory [13] activities. An extensive perusal of literature revealed that little work has been done on phthalimides and tetrahydrophthalimide as NNRTIs. In view of these facts and our interest on the development of novel NNRTIs, we have chosen tetrahydrophthalimide scaffold as one of the hydrophobic wings in butterfly-shape pharmacophore. All the newly synthesized compounds were designed based on the derived pharmacophoric model with acetamide moiety as hydrophilic body, and tetrahydrophthalimide and substituted aromatic amines as hydrophobic wings (general structure shown in Figure 1).

Molecular docking study
The docking studies of all the derivatives 4(a-p) were performed using molecular modeling software Autodock 4.2 (The Scripps Research Institute, CA, USA) [14] installed on a single machine running on a 3.4-GHz Pentium processor with Windows XP SP2 as the operating system. HIV-1 RT enzyme (pdb code 1rt2 (shown in Figure 2)) was taken from the RCSB, used as target protein [7]. Target protein pdb was further refined by removal of water molecules and by adding polar hydrogens and Kollmancharges. For the docking, a grid spacing of 0.375 Å and 63 × 63 × 63 number of points were used. The grid was centered on the active site. The auto grid program generated separate grid maps for all atom types of the ligand structures and one for electrostatic interactions. PRODRG online server was used to generate the energy minimized conformations of the ligands in pdb format [15]. Energy minimized conformation of ligands was subjected to calculation of Gasteiger-Huckel charges and saved in default format of Autodock. Autodock generated 50 possible binding conformations, i.e., 50 runs for each docking by using LGA search. Default protocol was applied, with initial population of 150 randomly placed individuals, a maximum number of 2.5 × 10 5 energy evaluations and 2.7 × 10 4 generations. A mutation rate of 0.02 and a crossover rate of 0.8 were used.

Validation of docking
Initially, the receptor was docked with extracted ligand TNK 651 in order to validate the docking calculations, reliability, and reproducibility of the docking parameters for the study. It was evident that the docked pose of the re-docked ligand was almost superimposed with that of the co-crystallized ligand ( Figure 3) with RMSD value of 0.5. Then, docking was performed with the standard drug efavirenz with 1tr2 for validation, and the mode of interaction was shown in Figure 4.
The binding free energies (docking score) and predicted inhibitory constant (Ki) values of the designed analogs were compared with binding free energies and inhibitory constants of the co-crystallized ligand TNK-651 and standard drug efavirenz. Binding free energies and predicted inhibitory constant values of TNK-651, efavirenz, and designed analogs were given in Table 1. Docking studies of   Figure 6.

Molecular parameters
Lipinski rule of five parameters like ClogP, molecular weight, number of hydrogen bond acceptors (HBA), number of hydrogen bond donors (HBD), solubility, drug likeness, and drug score were derived through online servers Molinspiration (Molinspiration Cheminformatics, Nova Ulica, Slovak Republic) and OSIRIS (Organic Chemistry, Switzerland) property calculator [16,17]. All the calculated values were given in Table 2.

Results and discussion
Chemistry Designed analogs were synthesized using a synthetic protocol shown in Scheme 1. In the first step, 2-chloro-N-(substituted phenyl) acetamide (2a-p) analogs were synthe sized by treating substituted anilines with 2-chloroacetyl chloride in dichloromethane and triethylamine as base. 2chloro-N-(substituted phenyl) acetamide (2a-p) intermediates were then treated with tetrahydrophthalimide (3)   in presence of base potassium carbonate and acetonitrile as solvent to yield titled compounds as final products 4(a-p) [18,19]. Synthesized compounds were isolated as pure and characterized by IR, 1 H NMR, mass, and elemental analysis data. In general, the IR spectra of the synthesized compounds showed N-H stretching at around 3,408 to 3,259 cm −1 , C = O (amide) absorption band at around 1,703 to 1,682 cm −1 , C = O (phthalimide) absorption band at around 1,786 to 1,768 and 1,712 to 1,702 cm −1 , C-O-C absorption (methoxy) band at around 1,249 to 1,234 cm −1 , and C-Cl absorption band at around 697 to 678 cm −1 . The 1 H NMR spectrum of the product 4c (see 'Experimental' section) showed two characteristic singlets at δ 4.27 and δ 2.30 because of COCH 2 -N and CH 3 , respectively. One broad singlet at δ 7.36 indicates the presence of NH, two doublets at δ 7.32, and δ 7.10 confirms the presence of para-substituted benzene ring. Besides these, the aliphatic region also showed the characteristic multiplet peaks due to CH = CH, CH-CH, and =CH-CH 2 at δ 5.96 to 5.97, δ 3.21 to 3.23,δ 2.63 to 2.69, and δ 2.26 to 2.28, respectively. Mass spectral analysis of the compounds4a and 4c showing the molecular ion peak at 285.6 and 299.6 (M + 1), respectively, confirms the molecular weight of the desired compounds.

In vitro HIV-1 RT inhibitory activity
All the synthesized compounds 4(a-p) were evaluated for HIV-1 RT inhibitory activity at concentrations 2 and 20 μM by using HIV-1 RT RNA-dependent DNA polymerase activity assay [20]. HIV-1 RT inhibitory activity results are shown in Table 3. Rilpivirine was used as standard drug in the assay.

Experimental
All solvents and reagents purchased from Sigma (Bangalore, India) or Merck (NJ, USA) companies are used as received without further purification. Solvent system used throughout experimental work for running thin layer chromatography was ethyl acetate and hexane mixture (30:70) in order to monitor the reaction.