Anti-TB evaluation of novel 2,3-dihydroquinazolin-4(1H)-ones and in silico studies of the active compounds

In vitro anti-tubercular activity of a series of 15 novel 2,3-dihydroquinazolin-4(1H)-one analogues were evaluated against Mycobacterium tuberculosis H37Ra (ATCC 25177 strain). Among the series, seven compounds showed moderate to good anti-TB activity with minimum inhibitory concentration (MIC) values ranging from 25.0–12.5 μg/mL. Further, in silico experiments were carried out to identify the probable ligand-protein interaction. Molecular docking of the target compounds into the active site of enzymes 1DQY Antigen 85C from Mycobacterium Tuberculosis and 2NSD Enoyl Acyl Carrier Protein Reductase reveals notable information on the possible binding interactions.


Introduction
Tuberculosis (TB) a communicable disease, spreads through air droplets containing M. tuberculosis bacilli. TB is treated with a combination of 1st line anti-TB agents, viz. Isoniazid, Rifampicin, Ethambutol and Pyrazinamide. The drug susceptible disease is cured after a 6 months long treatment under DOTs (Directly observed treatment) Program recommended by WHO. The present regimens of TB treatment have many drawbacks such as long treatment duration with drugs that are reported to cause organ toxicity and non-compliance to treatment [1,2]. This may result in the development of drug resistance. To improve the current regimens of TB, medicinal chemists are trying to design and synthesize new drug molecules that may overcome these shortcomings.
Quinazoline is one of the vital N-containing heterocycles bearing a benzene ring and a pyrimidine ring in its structure with chemical formula C 8 H 6 N 2 [3]. Quinazoline and its analogues appear in more than 100 biologically significant naturally occurring alkaloids, most commonly in the form of quinazolin-4(3H)-one moiety [4]. Medicinal chemists synthesized a variety of quinazoline compounds with different biological activities by introducing various active groups to the quinazoline moiety using developed as well as new synthetic protocols. The potential applications of the quinazoline derivatives in the fields of biology, pesticides and medicine have also been explored. One of the important quinazoline derivatives, deoxyvasicinone (2,3-dihydropyrrolo[2,1-b]quinazolin-9(1H)-one) is an alkaloid isolated from the aerial parts of Justicia adhatoda Linn. (Sanskrit-Vasaka), an evergreen sub-herbaceous bush, used extensively as local medicine for cough, cold, bronchitis and asthma [5]. Deoxyvasicinone possesses wide spectrum of biological properties like anti-microbial, antiinflammatory and antidepressant activities [6][7][8][9] as well as very important key intermediate for the synthesis of various natural products such as vasicinone [10], isaindigotone [11] and luotonin A [12]. Because of their diverse biological and pharmacological activities and extensive applications in pharmaceutical, research interest on the synthesis of quinazoline and its derivatives has never faded.

Chemistry
Our aim was to synthesize a series of novel derivatives of 2,3-dihydroquinazolin-4(1H)-one and evaluation of their anti-TB potency. Using our method developed recently [23], we have synthesized ten derivatives of novel 2,3dihydroquinazolin-4(1H)-one. Afterwards, the work was extended for the construction of some novel bis-2,3dihydroquinazolin-4(1H)-one analogues via Glaser coupling reaction using copper acetate in which 2a-2e were used as starting precursors [24]. All the five reactions gave the desired novel bis-2,3-dihydroquinazolin-4(1H)ones in good to excellent yields which was reported in our recent work [23]. From 1 H and 13 C NMR spectroscopic analysis, it was confirmed that all these novel bis-2,3-dihydroquinazolin-4(1H)-one analogues were C 2symmetric. Literature survey reveals that C 2 -symmetric molecules have been employed in a number of catalytic reactions as ligands as they limit different types of side reactions. Therefore, these five C 2 -symmetric analogues of 2,3-dihydroquinazolin-4(1H)-one are also expected to have extensive application in various organic transformations in coming days. All the synthesized 15 novel 2,3-dihydroquinazolin-4(1H)-one compounds are depicted in Scheme 1.

Biological evaluation
Proportion (agar dilution) assay [25] was used for anti-TB activity determination of the synthesized compounds in terms of Minimum Inhibitory Concentrations (MICs) against MTB H 37 Ra (ATCC 25177). Different concentrations of the compounds ranging from 25.0 to 3.125 µg/mL were tested to determine the MICs. From this experiment one compound (2b) was found to display appreciable anti-TB activity with MIC 12.5 µg/mL. Other six compounds (2e, 2h, 2j, 3b, 3d and 3e) showed activity at the concentration of 25.0 µg/mL. The remaining 8 compounds did not show activity up to 25.0 µg/mL, the highest concentration tested. In the present work, Ethambutol, an anti-TB drug, was used as a positive control (Table 1).

Molecular docking study
In silico experiments were carried out to identify the ligand-protein interaction. In this work, LibDock was used to evaluate the binding affinities between the active compounds and the enzymes 2NSD Enoyl Acyl Carrier Protein Reductase and 1DQY Antigen 85C from M. tuberculosis. Among all the compounds tested against the Enoyl Acyl Carrier Protein Reductase, five compounds exhibited good docking scores ( Table 2). Among the five compounds, compound 3b showed the lowest CDOCKER interaction energy score of −54.5525 kcal/ mol. Apart from the binding energy, it has formed one conventional H-bond with the residue Phe41. Conventional H-bonds are the major contributors for the stability in binding of protein and compounds. Apart from Hbond, it also formed some other interaction such as C-H bond, unfavorable donor-donor, pi-donor H-bond, pisigma, pi-pi T-shaped and pi-alkyl. The 2D and 3D structures of the docked complex is depicted in Fig. 1.
Similarly an in silico experiment was executed for these seven compounds on 1DQY Antigen 85C from M. tuberculosis for exploring the anti-TB potency of the synthesized novel analogues. Only two compounds tested against the Antigen 85C showed good docking scores (Table 3). Compound 2b showed the lowest CDOCKER interaction energy score of −37.0648 kcal/mol. Apart from the binding energy it has formed 2 conventional Hbonds with the residues Asp38 and Arg41 which are lying within the active pocket of the protein. Apart from Hbonds, it also formed some other interaction such as C-H bond, pi-pi T-shaped and pi-alkyl. The 3D and 2D structures of the docked complex is shown in Fig. 2.

Conclusion
In summary, we designed and successfully executed synthesis of some novel derivatives of 2,3-dihydroquinazolin-4(1H)-one including five C 2 symmetric bis-2,3-dihydroquinazolin-4(1H)-ones. The newly synthesized compounds were evaluated for their potent in vitro anti-tubercular activity. Among the 15 novel compounds, seven were identified as potent anti-TB agents with MIC in the range of 25.0-12.5 µg/mL. The most potent compound, 2b inhibited the growth of MTB Scheme 1 Novel analogues of 2,3-dihydroquinazolin-4(1H)one

Evaluation of in vitro anti-tubercular activity of compounds
Initially all the synthesized compounds were screened against M. tuberculosis H 37 Ra (ATCC 25177 strain) at the single concentration of 25 µg/mL. The active compounds from the above screening were further tested to determine MIC using Agar Proportion assay. To make stocks (5 mg/ mL), the compounds were dissolved in DMSO. Serial two fold dilutions were also made in DMSO from the stocks. 0.1 mL of compound or DMSO (negative control) or anti-TB drugs (positive controls) was added to 1.9 mL MB7H10 agar medium (in tubes, temperature 45-50°C, with OADC supplement, final concentration 10%). Then the contents were mixed and allowed to solidify as slants. Three week old culture of M. tuberculosis H 37 Ra was harvested from L-J medium and its suspension (1 mg/mL, equivalent to approximately 10 8 bacilli) was prepared in normal saline containing 0.05% Tween-80. 10 μL of 1:10 dilution of this suspension (~10 5 bacilli) was inoculated into each tube for 4 weeks at the temperature of 37°C. The lowest concentration of a compound up to which there was no visible growth of bacilli was its Minimal Inhibitory Concentration (MIC). Ethambutol was used as reference drug.

CDocker algorithim in Discovery studio
The Dock Ligands (CDOCKER) protocol is an implementation of the CDOCKER algorithim. It allows running a refinement docking of any number of ligands with a single protein receptor. CDOCKER is a grid based molecular docking method that employs CHARMm. The receptor is held rigid while the ligands are allowed to flex during the refinement. For predocked ligands, prior knowledge of binding site is not required. It is possible, however, to specify the ligand placement in the active site using a binding site sphere. Random ligand conformations are generated from the initial ligand structure through high temperature molecular dynamics, followed by random rotations. The random conformations are refined by grid based (GRID 1) simulated annealing and a final grid based or full force field minimization. CDOCKER uses a detailed atomic force field that is comprised of accuracy of automated MD docking with a soft-core potential over Monte Carlo (MC) Simulations and Genetic Alogorithim (GA) in searching a large configuration space.

Force-field and grid
The grid origin is situated at the center of the active sites of the protein. A + 1 point charge probe is used to map electrostatic interactions to the grid. The grid's vdW interactions are generated in one of two ways. The energies and forces are generated for the soft-core potentials in the heatingcooling stages and with normal non bond potentials in the final minimization step. The soft-core is generally approximated by the below function: where E* ij is the energy of regular non bond (vdW or electrostatic) potential. The coefficients a and b were extracted from two equations that express equality of regular and soft potential and forces at the switching distance.