Metal-Based Scaffolds of Schiff Bases Derived from Naproxen: Synthesis, Antibacterial Activities, and Molecular Docking Studies

We report here the synthesis, characterization, and antibacterial evaluation of transition metal complexes of Ni, Cu, Co, Mn, Zn, and Cd (6a–f), using a Schiff base ligand (5) derived from naproxen (an anti-inflammatory drug) and 5-bromosalicylaldehyde by a series of reactions. The ligand and the synthesized complexes were characterized by elemental analysis, UV-Visible, FTIR, and XRD techniques. The ligand 5 behaves as a bidentate donor and coordinates with metals in square planar or tetrahedral fashion. In order to evaluate its bioactivity profile, we screened the Schiff base ligand and its metal complexes (6a–f) against different species of bacteria and the complexes were found to exhibit significant antibacterial activity. The complexes showed more potency against Bacillus subtilis as compared to the other species. Moreover, we modeled these complexes’ binding affinity against COX1 protein using computational docking.

The crystal structure data have been deposited at the Cambridge Crystallographic Data Centre under CCDC No. 1556675. These data can be obtained free of charge via http://www.ccdc. cam.ac.uk/conts/retrieving.html (or from the CCDC, 12 Union Road, Cambridge CB2 1EZ, UK; Fax: +44 1223 336033; E-mail: deposit@ccdc.cam.ac.uk).

Synthesis of Target Molecules 6a-f
The Schiff base ligand 5 solution was treated with a corresponding metal salt solution and colored complexes were obtained as precipitates. The color change in the reaction mixture before and after induction of metal ions indicates complex formation. Thin layer chromatography was used to verify the formation of target molecules 6a-f. The FTIR spectral analysis registered the disappearance of the signal corresponding to the phenolic hydroxyl stretching vibration in ligand 5, caused by the chelation with corresponding metal ions. Moreover, the signal due to C=N stretching vibration shifted. Scheme 1 summarizes the synthetic protocol adopted to synthesize the target molecules 6a-f.

Anti-Bacterial Activity
We then screened all of the abovementioned compounds, i.e., Schiff bases and their metal complexes, for their antibacterial activities [37], showing that the Schiff base ligand and its metal complexes exhibited significant antibacterial activities. We selected four stains of different types of bacterial species, Escherichia coli, Streptococcus aureous, Salmonella typhae, and Bacillus subtilis, for use in evaluating antimicrobial potential by the diffusion method (Table 2). Briefly, autoclaved glass plates were used for conducting bacterial activity. Distilled water was used for preparing a solution of agar gel and Ciprofloxacin was used as a reference. Samples of all the derivatives of naproxen were placed in bacterial medium for 24 h. The activity was measured in terms of inhibition zones in mm. All compounds tested showed inhibition to bacteria growth, comparable with that of ciprofloxacin. The complexes with Co, Cu, and Co in general showed improvements over the free ligand across all four bacterial species. Prostaglandin endoperoxide H synthases (PGHSs)-1 and -2 (also called cyclooxygenases (COXs)-1 and -2) catalyze the committed step in prostaglandin biosynthesis. In the pathway, NSAIDs target COX1 and COX2 [38]. Human COX1 enzyme comprises two chains with 553 residues. The docked complexes of synthesized compounds 6a-f bound to COX1 were analyzed on the basis of the lowest binding energy values (kcal/mol) and hydrogen/hydrophobic bonding analyses. The analysis showed that all compounds bind to the enzyme with binding energy values around −11.70 kcal/mol, compared to the reference structure energy value (−8.20 kcal/mol).

Structure Activity Relationship (SAR) Analyses of Synthesized Compounds and Target Protein
The SAR analyses showed that compounds (6a-f) directly interact within the active region of COX1. Results showed that compounds build hydrogen and hydrophobic interactions with active site target residues with appropriate binding distances. In the 6a-docking results, we observed two interactions at different binding residues. The carbonyl oxygen atom of 6a interacts with Asn122 and forms a hydrogen bond having a length of 3.10 Å. Another hydrophobic interaction was observed against Leu112 with bond length 4.26 Å. In 6b-docking complex, we observed two hydrophobic interactions against target protein. In 6b-docking complex, we observed a weak binding interaction at Arg79 and Tyr64 residues having the bond lengths 5.07 and 4.69 Å, respectively. In both docking complexes, the metals containing ligands (6a and 6b) were different; therefore, our model shows little variant binding interaction pattern.
As in 6a-docking, in 6c-docking complex two hydrophobic interactions and one hydrogen bond were observed at different residues (Pro84, Arg79, and Lys532) of target protein. The ligand 6c forms hydrophobic interactions having a bond length of 4.95 and 3.13 Å. Similarly, we show a weak hydrogen bond with bond length 5.07 Å. In 6d-COX1 docking, a couple of hydrogen bonds were seen at Asn122 and Ser126 at a bond length of 2.91 and 2.82 Å, respectively. In both 6c and 6d docking complexes, the hydrogen bonds strengthen the docking complexes. In 6e and 6f complexes, we observed hydrophobic interactions against the target protein residues Thr80, Arg120, Ile89, Tyr64, and Ile46, with distances of 2.96, 3.91, 5.22, 5.06, and 4.37 Å, respectively ( Figure 3). The comparative results showed that the binding residues among standard compounds (5) and other derivatives were similar in 6a and 6d, whereas little fluctuations in residues exist in other docking complexes. A literature study also justified that these binding pocket residues are significant in downstream signaling pathways [39][40][41]. The SAR analyses showed that compounds (6a-f) directly interact within the active region of COX1. Results showed that compounds build hydrogen and hydrophobic interactions with active site target residues with appropriate binding distances. In the 6a-docking results, we observed two interactions at different binding residues. The carbonyl oxygen atom of 6a interacts with Asn122 and forms a hydrogen bond having a length of 3.10 Å. Another hydrophobic interaction was observed against Leu112 with bond length 4.26 Å. In 6b-docking complex, we observed two hydrophobic interactions against target protein. In 6b-docking complex, we observed a weak binding interaction at Arg79 and Tyr64 residues having the bond lengths 5.07 and 4.69 Å, respectively. In both docking complexes, the metals containing ligands (6a and 6b) were different; therefore, our model shows little variant binding interaction pattern.
As in 6a-docking, in 6c-docking complex two hydrophobic interactions and one hydrogen bond were observed at different residues (Pro84, Arg79, and Lys532) of target protein. The ligand 6c forms hydrophobic interactions having a bond length of 4.95 and 3.13 Å. Similarly, we show a weak hydrogen bond with bond length 5.07 Å. In 6d-COX1 docking, a couple of hydrogen bonds were seen at Asn122 and Ser126 at a bond length of 2.91 and 2.82 Å, respectively. In both 6c and 6d docking complexes, the hydrogen bonds strengthen the docking complexes. In 6e and 6f complexes, we observed hydrophobic interactions against the target protein residues Thr80, Arg120, Ile89, Tyr64, and Ile46, with distances of 2.96, 3.91, 5.22, 5.06, and 4.37 Å, respectively ( Figure 3). The comparative results showed that the binding residues among standard compounds (5) and other derivatives were similar in 6a and 6d, whereas little fluctuations in residues exist in other docking complexes. A literature study also justified that these binding pocket residues are significant in downstream signaling pathways [39][40][41].

Materials and Methods
The chemicals used were of analytic grades, purchased from Sigma Aldrich, Pakistan, and used without purification except where mentioned. Sodium salt of naproxen was received from Moringa Pharma, Pakistan, as a gift. FTIR Prestige−21 (Shimadzu, Kyoto, Japan) was used for FTIR spectra of the synthesized products. Elemental analyses were carried out with an Exeter Analytical Inc-CE-440 Elemental Analyzer. Melting points were determined by a Gallenkamp digital melting point apparatus (MFB−595−010M). Bruker (KAPPA Apex II) XRD was used for determination of crystalline structure. 1 H-NMR and 13 C-NMR spectra were recorded on a Varian Unity INOVA (300 MHz) spectrometer. 1 H-NMR (300 MHz) and 13 C-NMR (75 MHz) chemical shift values are reported as δ using the residual solvent signal as an internal standard. All NMR measurements were recorded in CDCl 3 as a solvent unless mentioned otherwise. Antibacterial activity was screened by using fresh cultures of different species of bacteria by following the disc diffusion method [42].

Synthesis of Metal Complexes 6a-f from Schiff Base Ligand (5)
Complexes 6a-f were prepared by the same general procedure. The methanolic solution of the ligand (5) and the metal salts were mixed in a 2:1 ratio in the presence of 10% Na 2 CO 3 and refluxed for 1-2 h. The reaction mixture was concentrated and filtered. The solid products were washed and recrystallized with appropriate solvents. FTIR spectra of the compounds can be obtained from the corresponding author free of cost on request.

Anti-Bacterial Testing
Autoclaved glass plates were used for conducting bacterial activity. Distilled water was used for preparing a solution of agar gel and Ciprofloxacin was used as a reference. Samples of all the derivatives of naproxen were placed in bacterial medium for 24 h. The activity was measured in terms of inhibition zones in mm.

Retrieval of COX1 Structure from the Protein Data Bank (PDB)
The three-dimensional (3D) crystal structure of COX1 (PDBID: 3N8Z) was accessed from the PDB (http://www.rcsb.org). The retrieved protein structure was further minimized by using a conjugate gradient algorithm and amber force field in UCSF Chimera 1.10.1 [44].

Designing of Ligands and Molecular Docking
The synthesized ligands (6a-f) were sketched in the ACD/ChemSketch tool and minimized by UCSF Chimera 1.10.1. A docking experiment was used on all synthesized compounds (6a-f) COX1 using a PyRx docking tool [45]. In the docking experiment, grid box parametric dimension values were adjusted as X = −25.6989, Y = 57.7957, and Z = 8.339, respectively. The default exhaustiveness = 8 value was used to obtain the finest binding conformational pose of protein-ligand docked complexes. The docked complexes were evaluated on lowest binding energy (Kcal/mol) values, and the hydrogen/hydrophobic interactions pattern using Discovery Studio (4.1) and UCSF Chimera 1.10.1. To check the validity of our docked complexes, a reference structure (Comp. 5) was also docked to check their binding pattern against COX1.

Conclusions
In this study, we synthesized metal derivatives of Naproxen-based Schiff bases in acidic or basic medium as required. We characterized the prepared compounds, namely naproxen acid from naproxen salt, methyl ester, hydrazide, Schiff base, and metal complexes, by different spectroscopic techniques. Subsequently, we screened these compounds for in vitro antibacterial activity, revealing that all complexes possess such activity. Lastly, we proved antibacterial activity through a computational approach. Our results suggest that these compounds inhibit Cyclooxygenase-1 through specific binding to its active sites. We calculated their individual binding energy against the target protein.
Metal complexes with Co, Cu, and Co showed higher antibacterial activity than ciprofloxacin across all four bacterial species.