Molecular Docking, ADMET Prediction, DFT, and Biological Studies of Zinc (II) Macrocyclic Complexes

1. Introduction

Nitrogen- containing organic compounds have demonstrated promising bioactivities [1,2,3,4,5]. Among these nitrogen compounds, macrocyclic compounds and their metal complexes have long intrigued chemists [6]. Nature’s macrocycles are well- characterized and serve remarkably diverse roles in biological systems. Synthetic macrocyclic metal complexes have been used for various biological applications, and the synthesis of these frameworks has increased significantly over the past two decades. Examples of natural metal- coordinated macrocyclic complexes include chlorophyll, myoglobin, hemoglobin, and vitamin B 12 [7,8]. In these macrocyclic structures, the metal plays a key role in stabilizing the complexes. These metal- coordinated complexes resist degradation and show enhanced thermal stability [9]. Amide- based macrocyclic complexes are highly interesting due to their unique coordination chemistry, which creates a geometrically constrained environment around a metal ion [10,11,12,13,14]. These large structures also have notable applications as antibacterial and antioxidant agents [15,16]. Zinc is the most essential mineral and is biocompatible with the human body. It is vital for various metabolic and biological processes, playing a crucial role in multiple physiological functions. Zinc also exhibits antimicrobial properties [17]. Zinc ions can effectively penetrate microbial cells and inhibit their growth by disrupting cellular processes [18]. Zinc also acts as an antioxidant by generating reactive oxygen species (ROS), which can damage bacterial cells. Zinc complexes have diverse applications, especially in medicinal chemistry [19]. Therefore, the search for new zinc complexes with strong biological activity is growing, as they could help overcome resistance challenges posed by existing antibiotics. Additionally, chemists are exploring additional applications, which have strengthened our interest and motivated us to synthesize macrocyclic complexes. Building on our previous work, the current aim is to synthesize and characterize novel N4-amide macrocyclic zinc complexes using various diamines and oxalic acid, followed by computational and biological evaluation.

2. Materials and Methods

2.1. Chemicals

All the required chemicals and solvents purchased were of analytical grade. They were used as received without purification. FT-IR spectrophotometer was used to record IR spectra in the range 4000-200 cm-1 using Nujol Mull. The literature method was used to determine the metal contents in the complexes [20]. A digital conductivity meter (HPG System, G-3001) was used for measuring conductivity. Melting points were determined using an electrical melting point apparatus.

2.2. Synthesis

Zinc metal was used as a template for synthesizing the macrocyclic complexes. An ethanolic solution (~20 ml) of zinc chloride (5 mmol) was refluxed with an ethanolic solution (~50 ml) of p-phenylenediamine/ethylenediamine/propylenediamine (10 mmol), respectively, for 30 minutes. Subsequently, to the above mixture solution, 10 mmol oxalic acid in 25 ml of ethanol was added, and refluxed with 1-2 drops of Conc.HCl for 6-8 hours. The condensation reaction yielded the ZnMc1, ZnMc2, and ZnMc3 complexes (Scheme 1). The solution was kept undisturbed at room temperature for 24 hours in desiccators. The colored precipitates of complexes were formed in good yield (~55-60%). They were filtered and washed with solvents such as methanol, ethanol, acetone, and diethyl ether and dried. The solubility test reveals their polar nature as they were found to be soluble in DMF and DMSO but difficult to dissolve in common organic solvents. They were found to decompose above 270 °C.

2.3. Biological Assays

2.3.1. In Vitro Antibacterial Activity

The in vitro antibacterial activity of synthesized zinc macrocyclic complexes was evaluated using the agar well diffusion method [21] with Muller–Hinton agar (MHA) medium. The study focused on Gram-positive bacteria, Bacillus subtilis (MTCC 441). All microbial cultures were adjusted to a 0.5 McFarland standard, corresponding to approximately 1.5 × 10⁸ cfu/mL [22,23]. Each petri dish was poured with 20 ml of MHA medium, and the agar plates were swabbed with 100 μl of each test bacterium’s inoculum, then incubated for 15 min to allow for adsorption. Wells with an 8 mm diameter were punched into the seeded agar plates using a sterile cork borer, and these wells were loaded with 100 μl of each metal complex at concentrations of 50, 75, and 100 mg/ml in 50% DMSO. All plates were incubated at 37 ºC for 24 hours. The antibacterial activity of each synthesized complex was determined by measuring the zone of growth inhibition against the test microorganisms using a zone reader (Hi Antibiotic zone scale). Dimethyl sulfoxide (DMSO) served as a negative control, while Gentamycin was used as a positive control.

2.3.2. Anticancer Activity

The human cancer cell lines used in this study were HCT-15, cultured in Roswell Park Memorial Institute (RPMI, Sigma-Aldrich, United States) supplemented with 10% fetal bovine serum (Sigma-Aldrich, United States) in tissue culture flasks at a 5% CO₂ atmosphere and 37 °C [24]. The cells were seeded at 3 × 10⁴ cells per well in 100 µL of Dulbecco’s Modified Eagle’s Medium (DMEM) [for HCT-15 cells, from Roswell Park Memorial Institute (RPMI)] containing 10% fetal bovine serum (FBS) in a 96-well tissue culture plate and incubated for 72 hours at 37 °C in a CO₂ incubator maintaining 5% CO₂ and 90% relative humidity. Afterwards, the cells were exposed to different concentrations of zinc complexes prepared in DMEM/RPMI, specifically 100, 50, 25, and 12.5 µM, for 24 hours. Subsequently, the cells were incubated with 100 µL of MTT dye (0.5 mg/mL) in a CO₂ incubator for an additional 4 hours. After incubation, the purple formazan produced by the cells appeared as dark crystals at the bottom of the wells. The culture medium was carefully removed from each well to avoid disrupting the monolayer. Then, 100 µL of dimethyl sulfoxide (DMSO) was added to each well. The plates were thoroughly mixed to dissolve the formazan crystals, resulting in a purple solution. The absorbance of the 96-well plates was measured at 570 nm using a Labsystems Multiskan EX ELISA reader against a reagent blank. The results were calculated as the micromolar concentration of 50% cell growth inhibition (IC₅₀) for each complex. The MTT assay was performed in three independent experiments, each conducted in triplicate.

2.4. Computational Studies

2.4.1. DFT Studies

An electronic structure of atoms, molecules, and solids can be computed with an effective DFT method [25]. To interpret the reactivity and chemical structure, electronic structures are calculated. Ground state of synthesized complexes was optimized. To find the optimized geometry first-principles calculations based on density functional theory within a generalized gradient approximation (GGA) were employed. For greater reliability in our results, we use the Perdew–Burke–Ernzerhof [26] form of the GGA with the Quantum Espresso package [27] based on ultrasoft pseudopotentials [28] and a plane wave basis. Local minima of the energy landscape are obtained through structural relaxation starting with different initial structures using Hellman–Feynman forces and stresses [29]. The kinetic energy cutoff for wave functions was 50 Ry, and a gamma-only k-mesh was used in sampling Brillouin zones. Around 20 Å vacuum space was chosen to avoid any interaction between the periodic images.

2.4.2. Molecular Docking Studies

Molecular docking is a computational approach used to predict the orientation and binding poses of a ligand when it interacts with a target receptor, generally a protein or nucleic acid [30]. Established docking techniques assume rigid or semi-flexible receptor and ligand representations, employing standardized search algorithms such as genetic algorithms, Monte Carlo simulations, and progressive construction methods, integrated with experimental, force-field-based, or knowledge-based scoring functions to evaluate binding affinity [31]. It is widely known that these traditional methods are computationally capable and extensively used for virtual screening; they usually strive to perfectly capture protein flexibility, solvent effects, and induced-fit conformational transformations [32]. Current advancements in molecular docking have addressed these drawbacks by combining receptor ensemble docking, flexible docking, and hybrid docking-molecular dynamics approaches that more accurately account for conformational variations [33]. Moreover, the integration of Python-related scripts has remarkably exceeded scoring functions, pose predictions, and ranking efficacy by integrating complex non-linear interaction patterns from mass structural datasets [34]. Cloud computing, GPU acceleration, and AI-derived docking platforms have also made large-scale, high-throughput docking operations possible with improved speed and accuracy, making modern docking a high-powered tool in structure-based drug development and drug repurposing [35]. In the current study, the two receptor protein targets with PDB ID (6COX, 7AHL) were used, respectively, structures obtained from the PDB database for the docking process. The protein preparation was done using multiple techniques, such as the binding site evaluation, which was done using the built-in Python script, and then these x, y, z values were applied in AutoDock tools to set the center and size of the grid box. For receptor protein 6COX- the size of grid was – 68, 84, 74 and center was- 27.732, 29.435, 40.186, and for receptor protein 7AHL- the size of the grid was- 93, 73, 87, and center values were- 54.711, 7.741, 35.624. From homomer chains of the target protein structure, the A chains were selected for docking. Two molecules ZnMc2 and ZnMc3 were selected and prepared as ligands [36]. The docking operation was run through AutoDock Vina in Anaconda Prompt.

3. Results and Discussion

3.1. Analytical Studies

The analytical data for macrocyclic complexes indicate their formula as: [Zn(CxHyN4O4)Cl2], where x=16,10 or 8; y= 12 or 16. The zinc complexes were soluble in DMF and DMSO but insoluble in water. They are thermally stable to 250 °C and decompose without melting. The chloride ion test was found to be positive after decomposing the complexes. Attempts to crystallize the complexes using mixtures of solvents, slow diffusion, or low-temperature crystallization were unsuccessful. However, analytical, spectroscopic, and magnetic data enable us to predict the structures of the complexes. All the complexes give satisfactory elemental analyses, as given in Table 1. A distorted octahedral geometry is confirmed for these complexes.

Conductivity in DMSO indicated them to be non-electrolytes (15–20 ohm^-1 cm² mol^-1)[37]. The molar conductance value indicates the formation of a covalent bond between the chloride ion and Zn(II). These complexes are non-electrolytes in nature, indicating the presence of a chloride ion inside the coordination sphere. The characteristic asymmetric and symmetric stretching vibrations band due to the amino group (NH2) in the IR spectra of diamines was absent in all the amide complexes. The band due to the OH group of dicarboxylic acids was also absent in the synthesized complexes. The absence of these characteristic bands indicates the formation of a macrocyclic framework. The IR absorption frequencies of the amide group occur at 1670-1680, 1565-1585, 1240-1270, and 650-682 cm^-1 for ZnMc1-ZnMc3 complex [13,15]. The lower values in the amide bands in the range 3240–3260 cm⁻¹, confirm the coordination of N–H with zinc metal [38]. The Far IR spectra show the band at 450-485 cm⁻¹ corresponding to (M–N) stretching vibrations, highlighting the coordination of amide nitrogen with zinc metal [39]. The stretching vibration band in the range 300–330 cm⁻¹is attributed to M-Cl bond [40,41]. The absorption band in the range 1445-1480 cm⁻¹, 1040-1075 cm⁻¹ and 725-760 cm⁻¹ may be assigned to phenyl ring vibrations in ZnMc1.

3.2. Pharmacological Studies

3.2.1. Antibacterial Activity

In this study, all the synthesized macrocyclic complexes were evaluated against Gram-positive bacteria, Bacillus subtilis. All three synthesized macrocyclic complexes were screened for their antibacterial activities. All the tested complexes show variable antibacterial activities against B. subtilis at 75 and 100 mg/mL concentrations. Standard antibiotic, specifically gentamycin, was used to compare the antibacterial activities exhibited by the synthesized complexes (Table 2). Zinc complexes (ZnMc1-ZnMc3) were tested for in vitro cytotoxicity against HCT-15 (human colon cancer) using MTT assay. The reduction in cell viability obtained by a specific increase in concentration of complexes against a fixed number of cancer cells is shown in Table 3. IC50 values (µM) of the complexes are presented in Table 3. ZnMc1 exhibits the highest activity in inhibiting cell proliferation.

3.3. Computational Studies

3.3.1. DFT Studies

The relaxed geometry of the molecules is shown in Figure 1. DFT calculations indicate the formation of one primary (P) and one secondary (S) bond between Zinc and Nitrogen atom. As the molecules contain more than one bond between the same elements, the maximum and minimum bond lengths are given in Table 4. Total energy per atom for each molecule is also presented in Table 4. Furthermore, the energies of HOMO and LUMO are found (Figure 2), and the band gap Eg is calculated as the difference between HOMO and LUMO. Electronegativity and electronic affinity can be calculated, respectively, from the two relationships (2) and (3) [42,43].

$$E_g=E_{HOMO}-E_{LUMO}$$

(1)

$$X={\displaystyle \frac{E_{HOMO}-E_{LUMO}}{2}}$$

(2)

For synthesized molecules, the electronic states, i.e., E_HOMO, E_LUMO, and ${\mathrm{E}}_{\mathrm{g}}$ are given in eV in Table 5. The Electron affinity (E.A), Ionization potentials (I.P), and electronegativity (X) are also given for the studied molecules

Our results show that the HOMO energies follow the trend of increasing energy for compounds ZnMc1 >ZnMc3>ZnMc2

whereas the LUMO energies for compounds can be arranged as follows: ZnMc1 >ZnMc3>ZnMc2

The band gap for the complexes increases in order ZnMc1 >ZnMc3>ZnMc2, and electronegativity decreases as ZnMc1 >ZnMc3>ZnMc2

According to Koopman’s theory [44], the following equations (3) and (4) can express the ionization potential and electron affinity, as shown in the Table

I.P = - E_HOMO

(3)

E.A = E_LUMO

(4)

Our results show the highest I.P for ZnMc2, whereas the lowest value is shown by ZnMc1, from eq (3) it is clear that E.A will follow the same trend as shown by E_LUMO.

3.3.2. Molecular Docking

The docking operation was run through AutoDock Vina in Anaconda Prompt. The binding affinity with both proteins was studied for two complexes. Macrocyclic complex, ZnMc2 and ZnMc3 show docking score of -6.5k/Cal and -6.4 k/Cal with PDB ID- 6COX and -5.30 k/Cal and -5.4 k/Cal with protein ID- 7AHL(Table 6). These hit candidates were further checked on BioVia for interaction analysis, and they also showed van der Waals and Pi-Pi interactions with the respective target proteins [45]. (Figure 3 and Figure 4) Eventually, this study gives a justification for these molecules to take them into consideration for further validation through a molecular dynamics approach.

3.4. ADMET Analysis

SwissADME [46] and pkCSM [47] were used for analyzing several physicochemical properties ( Table 7, Table 8, Table 9 and Table 10). The three macrocyclic complexes have molecular weights less than 500 and do not violate the criteria for hydrogen bond acceptors (HBA less than 10), hydrogen bond donors (HBD less than 5), and rotatable bonds (RB less than 5). Also, they do not exceed the maximum limit of LogP value of 5 [48]. Table 8 depicts the drug-likeness rules (Lipinski’s rule of 5, Ghose rule, Veber’s rule, Egan’s rule, and Muegge’s rule) and bioavailability. These rules ensure the drug-like properties of test macrocyclic complexes. Violation of more than one parameter reduces the drug’s ability (Figure 5). All three complexes meet all the parameters. Complex ZnMc2 and ZnMc3 did not follow the Ghose rule. The bioavailability of the test macrocyclic complexes was found to be 0.55. Since Veber‘s rule is obeyed by these complexes, this suggests good oral bioavailability of all the complexes.The parameters of ADMET, like log S (water solubility), TPSA (topological polar surface area), GI absorption, Blood Brain Barrier permeability, and P-glycoprotein substrates, were studied. ZnMc1 is found to inhibit the CYP1A2 inhibitor and the CYP3A4 inhibitor. so, can slow down the activity of Cytochrome P450 1A2 (CYP1A2) and Cytochrome P450 3A4 (CYP3A4) enzyme (Table 10) and act as a substrate for P glycoprotein.

Toxicity is an important parameter that dictates the efficiency of a drug as a medicine. Toxicity is an important parameter that dictates the efficiency of a drug as a medicine. The complexes were found to be non- AMES toxic (non-mutagenic), non-hepatotoxic, non-skin sensitizer, and non-hERG inhibitor (Table 10). Oral rat acute and chronic toxicity (LD50 and LOAEL) values predicted are given in the table and are in safe range. These parameters suggest that, in the future, these macrocyclic complexes might act as a good drug candidate

4. Conclusions

This study reports the synthesis of amide macrocyclic complexes of zinc using the template method with oxalic acid and different diamines (p-phenylenediamine, ethylenediamine, and propylenediamine). Spectroscopic studies predicted an octahedral geometry. These complexes demonstrate activity against bacterial strains. In these zinc complexes, nitrogen coordinates to the zinc ion and shares its charge, reducing polarity and increasing lipophilicity. This lipophilicity helps the macrocyclic complexes cross bacterial membranes easily and exhibit antibacterial activity. The structures of these zinc (II) complexes were also optimized theoretically, and their electronic and thermodynamic properties were analyzed using density functional theory (DFT). The energy gap between HOMO and LUMO, as well as electronegativity and electron affinity, were calculated with DFT, in silico molecular docking studies of the ZnMc2 macrocyclic complex with Cyclooxygenase-2 (Prostaglandin Synthase-2) and alpha-hemolysin from Staphylococcus revealed various interactions. ADMET parameters indicate that these macrocyclic complexes could serve as promising drug candidates in the future, though further research is necessary to develop potent inhibitors.