Amides Derived from Vanillic Acid: Coupling Reactions, Antimicrobial Evaluation, and Molecular Docking

A series of amides derived from vanillic acid were obtained by coupling reactions using PyBOP ((Benzotriazol-1-yloxy)tripyrrolidinophosphonium hexafluorophosphate) and DCC (Dicyclohexylcarbodiimide) coupling reagents. These were submitted to biological evaluation for species of Candida, Staphylococcus, and Pseudomonas. The microdilution method in broth was used for the antimicrobial testing to determine the Minimum Inhibitory Concentration (MIC) and to verify the likely mechanism of action for antifungal activity. The ten amides were obtained with yields ranging from 28.81 to 86.44%, and three compounds were novel. In the antibacterial evaluation, the amides (in their greatest concentrations) were bioactive against Staphylococcus aureus strain ATCC 25925. Meanwhile, all of the tested amides presented antifungal activity against at least one strain. The amide with best antifungal profile was compound 7, which featured an MIC of 0.46 μmol/mL, and a mechanism of action involving the plasma membrane and fungal cell wall. The presence of a methyl group in the para position of the aromatic ring is suggested which enhances the activity of the compound against fungi. Docking studies of the ten compounds using the protein 14α-demethylase as a biological target were also performed. The biological results presented good correlation with molecular docking studies demonstrating that a possible site of antifungal action for bioactive amides is the enzyme 14α-demethylase.


Introduction
Microorganisms are routinely associated with diverse serious diseases in a wide range of infections. However, some microorganisms contribute to the organism in terms of maintenance and balance, living in harmony with the host [1,2]. Within the context of infections caused by microorganisms, we highlight the presence of the Gram positive bacterium Staphylococcus aureus in the normal body microbiota, which can act pathogenically in wide ranges, from the most superficial to the most disseminated and severe infections [3,4].
In addition to bacteria, the eukaryotes such as fungi are generally considered to be harmless. However, principally affecting immunocompromised individuals, fungal infections contribute to the increased prevalence of hospitalacquired infections, having high impacts on mortality rates [5]. Treatments of fungal and bacterial infections often use antimicrobial agents from natural or synthetic sources which act on microorganisms by inhibiting growth or causing extermination [6]. In the last decade a considerable increase in the prevalence of antimicrobial resistance associated with prolonged treatment times, greater toxicity, and higher costs has been observed [7]. Antimicrobial resistance itself emphasizes the need for further study to obtain new pharmacologically effective and less toxic substances.
Phenolic compounds are widely distributed in nature, participating in the compositions of vegetables and fruits, and recognized as having antitumor, antimicrobial, and cardiovascular preventive and antidegenerative activities among others [8][9][10]. From this group, there are the phenolic acids (cinnamic and benzoic acid derivatives) and coumarins [11]. The benzoic acid derivatives such as gentisic, vanillic, and p-hydroxybenzoic acid present antimicrobial properties against many bacterial and fungal strains [12,13]. Thus, the objective of this study was to prepare derivative vanillic acid amides and assess their antibacterial and antifungal activities to obtain a derivative with a better antimicrobial profile and as well perform molecular docking of the ten compounds obtained (Figure 1), evidencing potential biological activity when targeting the protein 14 -demethylase.  (1)(2)(3)(4)(5)(6). Vanillic acid (0.59 mmol), triethylamine (0.59 mmol), dimethylformamide (0.59 mmol), and amine (0.59 mmol) were mixed and a solution of PyBOP (0.59 mmol) dissolved in 1.2 mL of dichloromethane (CH 2 Cl 2 ) was added. The reaction was kept in an ice bath for 30 minutes under constant magnetic stirring and then was brought to room temperature for 3 to 7 hours. The solvent was evaporated under reduced pressure and the residue was treated with distilled water (10 mL) and then extracted with ethyl acetate (3 x 10 mL). The organic phase was collected and treated with 1N (10 mL) hydrochloric acid solution, sodium bicarbonate solution 5% (10 mL), and distilled water (10 mL). The organic phase was dried with anhydrous Na 2 SO 4 and the solvent was evaporated under reduced pressure. The residue was purified by silica gel column chromatography (Hexane-EtOAc, 6:4 to 2:8) to obtain the described compounds [14]. (7)(8)(9)(10). Vanillic acid (0.59 mmol), along with 4-dimethylaminopyridine (0.059 mmol), followed by amine (0.59 mmol) were mixed and added to a solution of dicyclohexylcarbodiimide (0.59 mmol) dissolved in 3 mL of dichloromethane (CH 2 Cl 2 ). The solution was stirred for 24 to 48 hours at room temperature. The solvent was evaporated under reduced pressure and the residue was treated with distilled water (10 mL) and extracted with ethyl acetate (3 x 10 mL). The organic phase was collected and treated with 1N hydrochloric acid solution (10 mL), then the sodium bicarbonate solution 5% (10 mL), and distilled water (10 mL). The organic phase was dried with anhydrous Na 2 SO 4 and the solvent was evaporated under reduced pressure. The residue was purified by silica gel column chromatography (Hexane-EtOAc, 6:4 to 2:8) to obtain the described compounds [15].

4-Hydroxy-3-methoxy-N-(4-methoxybenzyl)benzamide (8).
Brown    [16]. The compounds were subjected to microdilution technique in 96-well plates. The samples were dissolved with DMSO and sterile distilled water (to 1.0 mL) obtaining an initial solution of 4000 g/mL. Sabouraud Dextrose Broth (SDB) 100 L was distributed to the 96 wells of the microdilution plates, with U background. Then, 100 L of compound solutions was distributed to the first well line of the plate. In a serial dilution with a ratio of two, it provides final concentrations that ranged from 1000 g/mL to 7.8 g/mL. We then added 100 L of fungal strain inoculum to each well of the plates. The culture sterility medium, the evaluated substances, and microbial growth were executed in parallel. The plates were closed and subjected to a temperature of 35 ± 2 ∘ C for 24 hours. The TTC at 1% (2,3,5-triphenyl tetrazolium chloride, Sigma-Aldrich5) dye was added to each well in order to confirm the presence of viable microorganisms [17]. The MIC was defined as the lowest concentration of the test substance inhibiting visible microbial growth.

Determination of Minimum Fungicide Concentration (MFC) of the Tested Compounds. The Minimum Fungicide
Concentration (MFC) of the compounds was obtained after MIC interpretation. Three (3) 30 L aliquots of supernatant were removed from the wells (where complete fungal growth inhibition was analyzed) and allocated to Petri dishes containing 15 mL of Sabouraud Dextrose. The plates were incubated at 35 ± 2 ∘ C for 24 hours to visual count colony forming units [18,19]. The plates were incubated at 35 ± 2 ∘ C, and readings were taken at 24 hours and 48 hours after incubation. Caspofungin was used as a positive control at an initial concentration of 5 mg/mL [20][21][22].

Antifungal Mechanism of Action for the Amides
(2) Ergosterol Assay. The testing was performed using the microdilution technique, as described earlier, in the presence of exogenous ergosterol (Sigma-Aldrich) in increasing concentrations (100, 200, and 400) g/mL, using Nystatin as a positive control. The plates were incubated at 35 ± 2 ∘ C and the readings were performed at 24 and 48 hours. Tween 80 and 96% ethanol were used for preparation of the ergosterol [20][21][22][23][24].

Results and Discussion
Vanillic acid was used as the starting material for the preparation of a collection of ten amides that were obtained through coupling reaction with PyBOP (1-6) and through coupling reaction with DCC (7-10) using the amines isobutylamine, pyrrolidine, cyclohexylamine, aniline, benzylamine, 4-fluorobenzilamine, 4-methylbenzylamine, 4-methoxybenzyl-am-ine, 3.4-dimethoxybenzylamine, and piperonylamine (Scheme 1). In 1 H NMR analysis, the guiding signals in common for all of the amides derived from vanillic acid ( Figure 2 In this study ten amides were tested against strains of Candida (Table 1). In accordance with the MFC/MIC relation [30] and as contained in Table 1, all of the tested compounds presented fungicidal activity against the tested strains. The amides, 1 3, 5, and 8, were active against the strain of Candida albicans at the highest concentration tested, while the amides 4 and 6 were bioactives, presenting respective MICs of 2.17 and 1.81 mol/mL. These data are in agreement with a previous study in which it tested several bioactive halogenated amides against bacterial and fungal strains, among them the vanillic amide N-(4-fluorobenzyl)-4-hydroxy-3-methoxybenzamide, which presented a MIC of 1.81 mol/mL against a Candida albicans strain [13]. The other amides showed no activity against this strain. It can be suggested that for amide 2 in relation to amide 3 the    albicans [31]. The data indicated that fluoride contributes to compound bioactivity. In the test using the Candida glabrata strain, the bioactive amides were 4, 6, and 7. The amides 6 and 7 presented activity in the highest concentration tested (MIC = 3.62 and 3.68 mol/mL) yet not presenting differences in potency with the changes of the para position substituent on the benzene ring. Compound 4 presented an MIC = 2.17 mol/mL, showing that the presence of the benzene ring linked directly to the nitrogen contributes to improvement of activity, being the same as with the C. albicans strain. In the test using Candida krusei, only amides 4 and 6 were bioactive presenting MICs for the concentrated tested. We note that the presence of substituents on both amides contributed to bioactivity against the four strains tested. Another strain tested was Candida guilliermondii, which presented sensitivity to a greater number of compounds in relation to the other strains. The amides 1-5 presented MICs of from 2.25 mol/mL to 1.94 mol/mL. These results suggest that changing the substituent group on these amides did not cause significant changes in bioactivity against the C. guilliermondii strain. The amides 9 and 10 presented respective MICs of 1.66 mol/mL and 1.57 mol/mL, indicating that varying substituents did not potentiate compound bioactivity. Amide 7 was bioactive with an MIC value of 0.46 mol/mL, thus presenting the greatest activity in relation to the other amides. One can again suggest that the presence of an electron donor group attached to the aromatic ring in the para position (such as a methyl) can generate an increase in antifungal potency against the C. guilliermondii strain. A study developed by Reddy, Ravinder, and Kanjinal (2012) has demonstrated that vanillic derivatives have antifungal activity against species of the genus Candida [32].
The most outstanding amide was compound 7 with an MIC = 0.46 mol/mL against the Candida strains. This compound was then subjected to mechanism of action testing against the C. guilliermondii strain via two pharmacological strategies, respectively, using ergosterol and sorbitol to determining the likely activity in the plasma membrane or cell wall (Table 2). Sterols participate in the constitution of all types of fungal cells. Ergosterol is the principal sterol and acts by modulating membrane fluidity, growth, and cell proliferation [33,34]. Testing to detect the biological target of the amide 7 was performed by adding more ergosterol to the medium containing the fungus and the compound. There was an increase in the MIC of the compound to 3.68 mol/mL, indicating that the fungus used the added ergosterol to reproduce, which demonstrates that the compound acts possibly by inhibiting synthesis or through binding directly to ergosterol. Azoles and polyenes are antifungal drug classes that act on ergosterol to treat fungal infections [4]. In the test using sorbitol, an osmotic protector, which acts by preventing changes in the fungal cell wall, the MIC of amide 7 increased, permitting growth at all concentrations, indicating that the substance acts by interfering with cell wall synthesis as well. Without the osmotic protector, the compound would cause lysis of the cells in the cell wall, yet in the presence of sorbitol, the fungus is protected and continues to reproduce. The substance thus acts through interference with the permeability of the cell membrane, with action on ergosterol, and in modulating cell wall function [35]. All tested compounds presented antibacterial activity against the Gram-positive lineage strain Staphylococcus aureus ATCC 25925, as evidenced by the minimum inhibitory concentration (MIC), the highest concentration tested for all the amides prevailed (Table 3). Amides 8, 9, and 10 presented better antibacterial activities (with MIC values of between 3.14-3.48 mol/mL), as compared to the other amides tested, which presented no bioactivity with changing of substituents. Thus, it can be suggested that the presence of one and/or two methoxyl group substituents in the meta and para positions or a dioxymethylene group in the compound enhances its antibacterial action.
Top conformers, based on docking scores (Figure 3(a)), were further selected for ligand-protein interaction analysis in the active cavity of the target protein. From the MD study, it was observed that, for all compounds, the pi terminal electron systems of the aromatic rings presented molecular interactions with the active site residues. All of the synthesized compounds were categorized based on MIC values. Category-1 included only one, the most active or potent compound, amide 7 (Figure 3(b)); category-4 included the nonactive amide 8 (Figure 3(c)). Comparing the modes of interaction of both 7 and 8, five interactions with active site residues were adopt by compound 7. Tyr131 and Ile379, respectively, form pi-pi and pi-H interactions with the terminal aromatic rings of the compound. Similarly, Ile377 and Trp 239 established pihydrogen interactions. Met487 forms a hydrogen bond (2.1Å) with the compound's formamide oxygen. For compound 8, no interactions were observed with the active site residues reflecting its nonactive nature. This might be caused by polar groups of the compound (attached to the terminal phenyl rings) which restrict the compound from molecular interaction. Category-2 included amides 10, 9, 6, 5, and 3.
All of the compounds in this category adopted various pi-pi and hydrogen interactions. Most of the interactions observed were pi-pi and hydrogen bonds with the compound's side benzo[d] [1,3] dioxol, methoxyphenol, and formamide moiety oxygen. Amide 10 pi-pi interactions were observed with the active site residues His236, Trp 239, and Tyr131 as seen in Figure 3(d). For compounds 9, 8, and 5 as presented in Figures 3(e), 3(f), and 3(g), similar interaction modes were observed, with the exception of His489 that was found to be involved in hydrogen bonding (with compounds 9 and 6) and a pi-H bond with compound 5. All the blue colored label residues indicate active site residues with no interactions. Category-3 includes amides 4, 1, and 2, which were generally observed making only hydrophobic interactions (data not shown), which stabilized the compounds in the active site cavity and make these compounds active against the target protein. On the whole for these compounds, the docking study revealed good correlations with the biological study.

Conclusion
All of the amides tested presented antifungal activity in against most of the strains tested. Amide 7 presented the best activity (lowest MIC) against C. guilliermondii strain 207, suggesting that the presence of a methyl group in the para position of the aromatic ring helps to potentiate antifungal activity. The remaining compounds presented low to moderate to bioactivity against the tested strains. The activity of amide 7 in fungal species suggests modification of cell wall functions and the plasma membrane. In the antibacterial tests, only Staphylococcus aureus strain ATCC 25925 presented sensitivity to all of the compounds. From molecular docking carried out with the ten derived compounds, a better interaction for compound 7 with 14 -demethylase was observed, suggesting it as a biological target. Thus, by presenting promising new prototypes, the results explained in this work serve as a basis for further research involving more powerful antimicrobials using vanillic acid derived amides.

Data Availability
The article data used to support the findings of this study have been deposited in the Federal University of Paraíba repository https://repositorio.ufpb.br/jspui/handle/123456789/13651.

Conflicts of Interest
The authors declare that there are no conflicts of interest.