BIOLOGICAL EVALUATION, QSAR AND MOLECULAR MODELING STUDIES OF 2,4-DICHLOROBENZOIC ACID DERIVATIVES AS ANTIMICROBIAL AGENTS

Objective: The aim of this study was to evaluate 2,4-dichlorobenzoic acid derivatives as antimicrobial agents through in vitro, QSAR and molecular docking studies. Methods: The compounds were subjected to in vitro antimicrobial screening by test tube dilution method and the structural characteristics governing the antimicrobial potential were studied using QSAR methodology. These compounds were also screened for docking simulation to find out binding confirmation of reported compounds with PDB 1aj0 and 5fsa using AutoDock tools and discovery studio. Results: The antimicrobial evaluation data indicated that compounds 13 and 18 were found to be the most effective against all the bacterial strains and Aspergillus niger while compounds 1 and 14 exhibited more antifungal potential against Candida albicans. QSAR studies confirmed the role of molar refractivity and Balaban index (J) as controlling parameters for antimicrobial potential. Molecular modeling study revealed that compounds interact with the active site of PDB by hydrophobic, hydrogen bonding, and Van der Wall interactions. Conclusion: These test compounds were identified as potent candidates for the control of microbial strains tested, and structural relationship with activity may provide valuable information for further design and synthesis of compounds with antimicrobial potential.


INTRODUCTION
The multidrug resistance associated with life-threatening microbial infections is a re-emerging and alarming microbial threat and amenable for millions of deaths annually due to the inadequacy of efficacious antimicrobial drugs [1]. To improve the current antimicrobial therapies, several studies have been conducted to improve the available treatments, but due to the continuance of the antibiotic resistance, pathogens have become a critical and complicated health issue [2].

Antimicrobial evaluation
The reported compounds were evaluated for their in vitro antimicrobial potential against bacterial and fungal strains through serial dilution technique to get their minimum inhibitory concentration [28]. The weighed amount of nutrient broth and Sabouraud dextrose broth was dissolved in distilled water to prepare nutrient medium for bacteria and fungus, respectively, and 1 ml of nutrient medium was transferred to each test tube. The test sample compound (0.01 g) was dissolved in 10 ml of dry dimethyl sulfoxide to give a stock solution of 100 µg/ml. The 1 ml solution of test compounds was transferred to test tubes having sterilized nutrient medium and diluted serially to get a set of five dilutions of test compounds having concentrations 50, 25, 12.5, 6.25, and 3.125 µg/ml. Minimal inhibitory concentration for each sample was investigated against Gram-positive bacterial strains; Bacillus subtilis (MTCC 441), Staphylococcus aureus (MTCC 7443), and Staphylococcus epidermidis (MTCC 435), two Gram-negative bacterial strains; Escherichia coli (MTCC 1652) and Pseudomonas aeruginosa (MTCC 424), and two fungal strains; Aspergillus niger (MTCC 8189) and Candida albicans (MTCC 227). The freshly cultured strains of each organism were transferred in to test tubes and incubated at 37±1°C for 24 h for bacterial strains, 48 h for C. albicans and 7 days at 25±1°C for A. niger. Ciprofloxacin was considered as a standard for antibacterial and fluconazole for antifungal activity.

RESULTS AND DISCUSSION
In vitro antimicrobial studies All the compounds were evaluated for their in vitro antibacterial activity against S. aureus, B. subtilis, and S. epidermidis (Gram-positive bacteria), and E. coli and P. aeruginosa (Gram-negative bacteria), and antifungal potential against C. albicans and A. niger. The power management integrated circuit (pMIC) values obtained are recorded in Table 1.
Based on results, it could be inferred that compounds 13 and 18 have the highest and broad spectrum of antimicrobial potential while compound 2 exerted the highest activity against S. epidermidis with a pMIC value of 1.785 µM. Compounds 1, 3, and 11 exhibited good antimicrobial activity against all strains. Compounds 1 and 14 exhibited the most potent antifungal activity against C. albicans with a pMIC value of 2.102 µM. All other test compounds showed moderate antimicrobial potential against all bacterial and fungal strains. Compounds 13 and 18 also displayed more inhibition effect against A. niger species. Compounds 1, 3, 11, and 14 exhibited good antifungal potential against A. niger.
It is observed that in this series compound 2 with p-Cl substituent showed selective inhibition against S. epidermidis whereas the activity was diminished when substituted with the o-NO 2 group. Results revealed that chloro substituent on phenyl ring exerted higher activity against all microbial strains as compared to fluoro substituent. Apparently, the compound bearing methoxy group displayed more antimicrobial potential as compared to methyl substituted compound. Replacement of aromatic anilines with aliphatic amines as observed in compounds 15, 16, and 17, afforded a decrease in antimicrobial potential against all strains. A deep insight into the structure of test compounds revealed that their antimicrobial activity seems to be modulated by the effect of the substituent. Above all, it could be inferred that a significant broad spectrum of antimicrobial activity was allied with compounds incorporating electron withdrawing substitution [17].

QSAR study
In this present study, dataset of 18 compounds (1-18) was used to establish a quantitative relationship between antimicrobial activity and structural descriptors of substituted sulfonamide derivatives. The structural descriptors coding for lipophilic, steric, electronic, and molecular connectivity were used in this study, and the corresponding value for each molecular descriptor is presented in Table 2. A correlation matrix was constructed to find out the inter-relationship of different calculated molecular descriptors and antimicrobial activity as well as interrelationship of different molecular descriptors with each other (Table 3).
It was observed from correlation matrix that highest correlation existed between Randic parameter (R) and first-order molecular connectivity index ( 1 χ, r=1.000) and between Weiner index (W) and zero-order molecular connectivity index ( 0 χ, r=0.995). The least interrelationship was observed for the energy of highest occupied molecular orbital (HOMO) and energy of lowest unoccupied molecular orbital (r=0.050) and energy of HOMO and third-order alpha shape indices (kα 3, r=−0.063). In general high inter-relationship was observed for different calculated molecular descriptors with each other. The correlation of molecular descriptors with respective antibacterial and antifungal activity is presented in Table 4.
The different molecular descriptors were subjected to linear free energy regression analysis with antibacterial activity against S. aureus and a mathematical model (Eq. 1) illustrating the importance of molar refractivity (MR) was obtained. was used to predict the antibacterial activity of test compounds against S. aureus and a comparison of observed, predicted and residual antibacterial activity is presented in Table 5.
The results presented in Table 5  The r value in Equation 4 was 0.643 which was less than the required limit (r≥0.7) and the outcomes of Equation 4 were studied in detail to The QSAR model represented by Equation 5 has been built up using 13 compounds and it has got high r, r 2 , q 2 , F values, and low S values, which supported the fact that model represented by Equation 5 was valid and it can be used for prediction of antifungal activity of target compounds against C. albicans, and the predicted antimicrobial activity is presented in Table 5. Valid QSAR models were not obtained for antibacterial activity of test compounds against S. epidermidis.

Docking study
In silico molecular docking studies were executed to investigate the possible potential binding modes for these sulfonamide derivatives against microbial targets. In the present study, H-bonding, hydrophobic interactions, and free binding energy say, and dock score were considered for the analysis (Table 6).
In the case of antibacterial activity, all the docked compounds showed binding energy from -6.8 kcal/mol to -8.6 kcal/mol. In the binding mode of topmost active, compound 13 potently bound to 1aj0 through four hydrogen bonding and hydrophobic interactions (Fig. 1). In the case of antifungal activity, all the compounds displayed binding energy from −6.6 kcal/mol to −8.4 kcal/mol. The foremost active compound 1 formed two hydrogen bonding, hydrophobic and Van der Wall interactions with active binding site (Fig. 2)     In a comparison of a compound containing NO 2 substitution, compounds 1, 3, and 14, amide-pi-stacked interaction was only found in compound 1, substituted with p-NO 2 group. Compound 2 (p-Cl) formed a similar type of interactions as compared to compound 12 (p-F) with two more

Thakral and Singh
interactions such as alkyl and pi-alkyl interaction with Ile:379 and Phe:105 amino acid residues, respectively, that resulted in an increase in antifungal potential. Compounds with aliphatic substitutions such as 15, 16, and 17, amide-pi-stacked, and pi-pi stacked interactions were not observed which might have contributed for the decrease in antifungal potential.

CONCLUSION
The set of 18 test compounds was subjected to antimicrobial evaluation and results depicted that compounds with NO 2 substitution were most active and para-substituted nitro group effectively increase the antimicrobial potential as compared to substitution at ortho and meta positions. The compounds with chloro substitution were found to be better in activity in comparison to other halogen substituents. The QSAR studies depicted the involvement of lipophilic parameter MR and Balaban index (J) as governors of antimicrobial potential in case of antibacterial and antifungal activity, respectively. Docking studies correlated the activity with structural features by depicting various interaction modes through which the test compounds were interacting with microbial targets such as hydrophobic, hydrogen bonding, and Van der Wall interactions.

ACKNOWLEDGMENTS
The author (VS) gratefully acknowledge the financial support as a minor project for purchase of chemicals and Junior Research Fellow award to Ms. Samridhi Thakral by Dr. A. P. J. Abdul Kalam Central Instrumentation laboratory, G. J. U. S. and T., Hisar under DST-PURSE Programme. The authors are thankful to Chairman, Department of Pharmaceutical Sciences, G. J. U. S. and T., Hisar for providing necessary facilities to carry out this research work.