Abstract

Zn(II) complexes of three heteroaryl Schiff bases L¹, HL² and L³ of general formula [ZnL¹Cl₂]∙2H₂O (1), [ZnL²Cl]∙H₂O (2), [ZnL³Cl₂]∙H₂O (3) [L¹ = 1-(quinolin-2-yl)-N-(quinolin-8-yl)methanimine, HL² = 2-((quinolin-8-ylimino)methyl)quinolin-8-ol, L³ = N-(pyridin-2-ylmethyl)-1-(quinolin-2-yl)methanimine] have been synthesised. Based on spectroscopic analyses, five-coordinate geometries were proposed where L¹ and L³ coordinated through the azomethine N and the two quinolines N as neutral NNN-donor modes, while HL² coordinated through the azomethine N, the two quinolines N and the O atom of the OH group as deprotonated with NNNO-donor mode. DFT calculations were used to investigate the electronic properties of the ligands and complexes, and their optimised structures were generated. Complexation and the presence of the OH group enhanced the activity against cercariae of Schistosoma mansoni while L¹ with fused benzene backbone and its complex were inactive compared to L³ with CH₂ backbone from 2-picolylamine derivative. Molecular docking studies revealed that the active compounds displayed higher binding affinity to the active sites of SmCE2a, SER, SmTGR, and SmPNP, with binding energies ranging from −6.1 to −10.1 kcal/mol, indicating that the ligands (HL²), L³ and complex (2) can be potential inhibitors of the four main target proteins of S. mansoni. In silico pharmacokinetics properties of HL², L³ and Zn(II) complex (2) agree with the four drug likeness rules (Lipinski, Ghose, Veber, and Egan rules) and these derivatives can be considered as potential lead candidates for an antischistosomal therapeutic development. Furthermore, the active ligands and complex showed a good bioavailability score of 0.55.

Keywords

Zn(II) Complexes, Heteroaryl Schiff Bases, DFT Studies, In Vitro and in Silico Cercaricidal Potential, Pharmacokinetics Prediction

Share and Cite:

1. Introduction

The chemistry of coordination compounds with heterocyclic ligands containing oxygen and nitrogen as donor atoms has attracted increasing attention because they can exhibit different modes of coordination to a metal centre. Schiff bases are an important class of ligands that have been extensively studied. Aryl Schiff bases are commonly synthesised by the condensation reaction of aromatic aldehydes or ketones with primary amines, resulting in an azomethine moiety with at least one aryl group attached to the carbon or nitrogen atom. The interest in Schiff base ligands is therefore due to their ease of synthesis, selectivity to the central ion [1], synthesis flexibility [2], structural diversity [3], and their broad biological activities [4].

Quinolines are derivatives of pyridine and are found in many biologically active natural and synthetic compounds [5]-[8]. Their Schiff base derivatives have also been reported with various pharmacological activities [9] [10]. Some free Schiff base ligands are known to be active, but literature has shown that the presence of metal in such ligands may increase their activities [11] [12]. Apart from their medicinal properties, Schiff base compounds play an important role as ligands in coordination chemistry. Schiff base ligands that contain multidentate NNN-donor atoms have gained a lot of interest in different fields of coordination chemistry [13]-[16]. Schiff base compounds of 8-hydroxyquinoline are used as chelating agents [17]. Their versatile coordination behaviour is used to prepare various coordination complexes that present interesting properties such as DNA binding ability [18], making them potential targets for anticancer, antimicrobial, antioxidant and bioimaging agents in living cells [19]-[22]. In general, the imine group, which contains an electrophilic carbon and nucleophilic nitrogen, provides excellent binding opportunities that inhibit targeted diseases [23].

Schistosomiasis, a neglected tropical disease, is the second most endemic parasitic worm disease caused by species of the genus Schistosoma, including Schistosoma guineensis, Schistosoma haematobium, Schistosoma intercalatum, Schistosoma japonicum, Schistosoma mansoni, and Schistosoma mekongi [24]. More than 250 million individuals in tropical and subtropical regions of the world are affected by this neglected tropical disease, especially in deprived communities, which lack access to safe drinking water and adequate sanitation [25]-[27]. High prevalence rates have been recorded in Africa, the Caribbean, Latin America, and Asia, with over 90% of cases reported in Sub-Saharan Africa [28]. Considerable efforts to eradicate schistosomiasis from endemic areas through repeated mass preventive chemotherapy, improvement of water, sanitation, and hygiene, snail control, behavioural changes, and environmental management have not been entirely successful [29]. This is mainly due to the high cost involved in treatment campaigns and the inequity of access to preventive chemotherapy [30] [31].

Praziquantel, which is currently the only drug of choice for treatment, has faced the challenge of reduced drug sensitivity to juvenile parasites and the suspected emergence of drug-resistant strains [32]-[34]. While emphasising the need to eliminate schistosomiasis by 2030, the World Health Organisation (WHO) proposed the development of new intervention tools and alternative drugs to Praziquantel [35] [36]. Proteins have been reported as targets in antischistosomal drug discovery [37]. Recent research reported novel antischistosomal drug targets as inhibitors of SmTGR or SmDHODH (the enzyme dihydroorotate dehydrogenase from Schistosoma mansoni) via in silico methods [38] [39]. There is an increasing interest in the use of transition metal compounds as new opportunities to fight against drug-resistant pathogens through their antiparasitic activities [40] [41]. The potential of imine-Cu(II) and Zn(II) metal complexes as new lead compounds against adult schistosomes has been reported [42]. The introduction of metal-based compounds in chemical biology has opened the door to developing novel drug treatment options against neglected tropical diseases [43] [44].

We recently reported on some Schiff base metal complexes with biological activity [45]-[51]. In our search for ligational behaviour of small bioactive molecules towards transition metals and the potential bioactivity of their resulting metal complexes against virus responsible for parasitic diseases, we now report on the synthesis, characterisation, Density Functional Theory studies, in vitro antischistosomal potential, molecular docking and in silico pharmacokinetics predictions of Zn(II) complexes with quinolinyl heteroaryl Schiff bases bearing different backbones derived from quinolin-2-carboxaldehyde, 8-hydroxyquinolin-2-carboxaldehyde with 8-aminoquinoline and 2-picolylamine.

2. Experimental Section

2.1. Materials and Methods

Reagent grade quinolin-2-carboxaldehyde, 8-hydroxyquinolin-2-carboxaldehyde, 8-aminoquinoline, 2-picolylamine and ZnCl₂ were purchased from Sigma Aldrich. Ethanol as a solvent was used as purchased. The melting points were determined with a Stuart SMP11 instrument in sealed capillary and are uncorrected. Infrared spectra were obtained (KBr 400 - 4000 cm⁻¹) on an ALPHA FT-IR Spectrometer from Bruker. UV-visible spectra were carried out with a GENESYS 10 S UV-vis spectrophotometer. A Bruker AV 400 MHz Spectrometer was used for the ¹H and ¹³C NMR analyses.

2.2. General Procedure for the Synthesis of L¹ and L³

The Schiff base ligands L¹ and L³ used in the synthesis of the Zn(II) complexes were obtained from the condensation reaction of quinolin-2-carboxaldehyde with 8-aminoquinoline and 2-picolylamine, as indicated in Scheme 1. HL² was synthesised as reported in the literature [52].

2.2.1. Synthesis of L¹

(1 g, 6.8 mmol) of quinoline-2-carbaldehyde in 50 mL of ethanol was added dropwise with stirring to 8-aminoquinoline (0.97 g, 6.8 mmol) in 50 mL of ethanol in a round-bottom flask. Three drops of glacial acetic acid were added to the mixture. The reaction mixture was then heated to reflux for 4 hours, and the precipitate obtained after cooling overnight was filtered, washed twice with ethanol and dried (1.73 g, 90%). Melting point: 182˚C. FTIR (max/cm⁻¹): 3335 vw, 3048 vw, 1681 vw, 1593.8 (C=N), 1567 w, 1503 s, 1476 w, 1426 w, 1373 m, 1140 w, 820 m, 750 vs, 620 w, 497 m, 423 w. NMR: δH (400 MHz; DMSO): 8.83 (1H, s), 8.72 (1H, dd, J = 4 Hz), 8.50 (1H, dd, J = 8 Hz), 8.29 (1H, dd, J = 8 Hz), 8.16 (1H, d, J= 8 Hz), 8.12 (1H, d, J = 4 Hz), 7.91 (1H, t, J = 8 Hz, J = 12 Hz), 7.84 (1H, d, J =4 Hz), 7.70 (1H, t, J =8 Hz, J = 12 Hz), 7.45 (1H, dd, J = 4 Hz), 7.30 (1H, dd, J = 4 Hz), 7.06 (1H, d, J = 8 Hz), 6.9 (1H, d, J = 4 Hz), 3.41 (H₂O). δC (100 MHz; DMSO): 160.2, 151.0, 147.4, 147.0, 139.0, 138.6, 137.4, 137.2, 136.3, 130.1, 128.9, 128.41, 128.4, 128.3, 127.9, 121.9, 121.3, 114.1, 109.1. UV-vis: λ_max (DMSO/nm): 245, 280, 385.

2.2.2. Synthesis of L³

(1.46 g, 10 mmol) of quinoline-2-carbaldehyde in 50 mL of ethanol was added dropwise with stirring to 2-picolylamine (1 mL, 10 mmol) in 50 mL of ethanol in a round-bottom flask. Three drops of glacial acetic acid were added to the mixture. The reaction mixture was then heated to reflux for 4 hours, and the precipitate obtained after cooling overnight was filtered, washed twice with ethanol and dried (1.65 g, 67%). Melting point: 262˚C. FTIR (max/cm⁻¹): 3300 br, 3054 w, 2926 w, 2855 w, 1673 w, 1588 s (C=N), 1566 w, 1496 w, 1461 s, 1425 s, 1144 m, 1110 w, 1090 w, 1044 w, 1014 w, 995 m, 836 m, 744 vs, 694 w, 617 m, 476 m, 425 w, 401 m. NMR: δH (400 MHz; DMSO): 8.56 (1H, t, J = 8 Hz, J = 12 Hz), 8.35 (1H, d, J = 8 Hz), 8.30 (1H, s), 8.07 (1H, d, J = 8 Hz), 7.95 (1H, t, J= 8 Hz, J = 12 Hz), 7.87 (1H, d, J = 8 Hz), 7.72 (1H, d, J = 8 Hz), 7.49 (1H, t, J = 4 Hz, J = 12 Hz), 7.39 (1H, t, J = 8 Hz, J = 12 Hz), 7.28 (1H, t, J = 4 Hz, J = 8 Hz), 7.09 (1H, t, J = 4 Hz, J = 8 Hz), 6.83 (1H, d, J = 4 Hz), 4.72 (2H, d, J= 4 Hz), 3.45 (H₂O). δC (100 MHz; DMSO): 166.7, 162.2, 156.8, 150.9, 149.0, 137.3, 136.9, 130.9, 128.7, 128.6, 128.5, 126.0, 123.4, 122.4, 119.1, 62.0. UV-vis: λ_max (DMSO/nm): 250, 300, 390.

Scheme 1. Synthesis of L¹, HL², L³ ligand and their 1, 2, and 3 Zn(II) complexes.

2.3. General Procedure for the Synthesis of Zn(II) Complexes

2.3.1. Synthesis of [ZnL¹Cl₂]∙2H₂O (1)

A 20 mL ethanol solution of ZnCl₂ (0.136 g, 1 mmol) was added dropwise to a 30 mL hot ethanol solution of L¹ (0.283 g, 1 mmol) while stirring. The resulting mixture was refluxed for 5 hours, and the resulting precipitate obtained on cooling overnight was filtered, washed with ethanol and dried (0.34 g, 83%). Melting point: >285˚C. FTIR (max/cm⁻¹): 3393 w, 3059 w, 1594.3 s (C=N), 1505 s, 1450 w, 1429 m, 1379 m, 1343 w, 1218 w, 1146 w, 1071 w, 833 s, 759 vs, 620 w, 477 m, 449 w, 417 w. NMR: δH (400 MHz; DMSO): 9.17(1H, dd, J = 4 Hz), 8.91 (1H, s), 8.73 (1H, d, J = 8 Hz), 8.53 (1H, t, J = 8 Hz, J = 4 Hz), 8.17 (1H_, t, J= 4 Hz, J = 8 Hz), 7.98 (1H, d, J = 8 Hz), 7.89 (1H, dd, J₌ 4 Hz), 7.80 (1H, t, J =4 Hz, J = 8 Hz), 7.73 (1H, dd, J = 4 Hz, J = 8 Hz), 7.66 (1H, d, J = 4 Hz), 7.30 (1H, t,, J = 4 Hz, J = 8 Hz), 7.18 (1H, brs), 7.02 (1H, d, J = 8 Hz), 3.39 (H₂O). ¹³C NMR was not obtained due to low resolution. UV-vis: λ_max (DMSO/nm): 255, 283, 585. Molar Conductivity: 15.1 S∙cm²∙mol⁻¹.

2.3.2. Synthesis of [ZnL²Cl]∙H₂O (2)

A 20 mL hot ethanol solution of ZnCl₂ (0.136 g, 1 mmol) was added dropwise to a 30 mL hot ethanol solution of HL² (0.299 g, 1 mmol) while stirring. The resulting mixture was refluxed for 5 hours, and the precipitate obtained on cooling overnight was filtered, washed with ethanol and dried (0.32 g, 81%). Melting point: >339˚C. FTIR (max/cm⁻¹): 3336 br, 3055 w, 1597 w (C=N), 1573 s, 1505 s, 1435 m, 1376 m, 1330 m, 1241 m, 1196 w, 1110 m, 1087 m, 820 s, 787 s, 750 vs, 721 w, 539 w, 498 w, 463 m, 434 w, 419 m. NMR: δH (400 MHz; DMSO): 8.84 (1H, s), 8.72 (1H, dd, J = 4 Hz), 8.48 (1H, d, J = 4 Hz), 8.17 (1H, dd, J= 8 Hz), 7.86 (1H d, J = 4 Hz), 7.72 (1H, dd, J =4 Hz, J = 8 Hz), 7.54 (1H, t, J = 4 Hz, J = 8 Hz), 7.45 (1H, dd, J = 4 Hz), 7.28 (1H, dd, J = 4 Hz, J = 8 Hz), 7.05 (1H, d, J = 4 Hz), 6.88 (1H, d, J = 4 Hz), 6.78 (1H, d, J = 8 Hz), 3.38 (H₂O). δC (100 MHz; DMSO): 193.8, 156.5, 154.3, 151.8, 147.7, 147.4, 145.7, 137.9, 136.3, 134.8, 129.0, 128.1, 124.5, 121.9, 121.4, 118.3, 114.2, 109.1, 106.4. UV-vis: λ_max (DMSO/nm): 275, 330, 590. Molar Conductivity: 35.9 S∙cm²∙mol⁻¹.

2.3.3. Synthesis of [ZnL³Cl₂]∙H₂O (3)

A 20 mL hot ethanol solution of (0.136 g, 1 mmol) ZnCl₂ was added dropwise to a 30 mL hot ethanol solution of L³ (0.247 g, 1 mmol). The resulting mixture was refluxed for 5 hours and the precipitate obtained after cooling overnight was filtered, washed with ethanol and dried (0.28 g, 73%). Melting point: 281˚C. FTIR (max/cm⁻¹): 3392 br, 3075 w, 1596 s (C=N), 1568 w, 1509 m, 1477 m, 1434 m, 1378 w, 1343 w, 1294 m, 1216 m, 1148 w, 1051 w, 1016 w, 832 m, 756 vs, 637 w, 483 w, 410 w. NMR: δH (400 MHz, DMSO): 9.02 (1 H, s), 8.78 (1H, dd, J = 8 Hz) 8.56 (1H, d, J = 4 Hz), 8.45 (1H, t, J = 4 Hz, J = 8 Hz), 8.33 (1H, t, J = 4 Hz, J = 8 Hz), 8.17 (1H, t, J = 4 Hz, J = 8 Hz), 8.07 (1H, t, J = 4 Hz, J = 8 Hz), 7.96 (1H, dd, J =4 Hz), 7.79 (1H, dd, J =4 Hz, J = 8 Hz), 7.72 (1H, dd, J = 4 Hz), 7.61 (1H, dd, J = 4 Hz), 4.81 (2H, s), 3.44 (H₂O). δC (100 MHz, DMSO): 166.5, 164.7, 158.5, 156.9, 155.6, 149.4, 147.5, 146.5, 144.2, 141.3, 138.2, 137.2, 130.9, 129.8, 128.5, 124.2, 121.7, 60.4. UV-vis: λ_max (DMSO/nm): 265, 330, 395. Molar conductivity: 17.8 S∙cm²∙mol⁻¹.

2.4. Biological Methodology

2.4.1. Cercaricidal Activity

The cercaricidal activity was assessed according to the protocol described by Dongmo Nguepi et al. [53] with some modifications. Briefly, stock solutions of compounds dissolved in DMSO (20 mg/mL) were diluted using distilled water to obtain a concentration of 2 mg/mL and to bring down the concentration of DMSO to at most 1% in the test plate. Cercariae suspensions corresponding to 20 cercariae per well were inoculated in the wells of a 96-well plate, and solution was added to make final concentrations of 200, 40, 8, 1.6, 0.32, and 0.064 µg/mL. The independent biological repeats were done in duplicates and incubated for 3 hours. Triplicates were used, and controls: Niclosamide 1% as a positive control and distilled water as a negative control. The cercaricidal activity of the compounds was read microscopically from the mortality of cercariae.

2.4.2. Cytotoxicity Test

The biosafety assessment was conducted using African green monkey Vero cells (ATCC-CRL-1586). The assay was performed according to the protocol described by Balbaied and Moore (2020) [54] with some slight modifications. Briefly, 100 µL of cell suspension was distributed in each well of a 96-well plate at a charge of 1 × 10⁴ cells/well and followed by an incubation of the plate at 37˚C, 5% CO₂ for 18 hours for cell seeding. After this, the culture medium was gently removed with the aid of a pipette and replaced by 90 µL of fresh DMEM medium pre-heated at 37˚C and 10 µL of each concentration for all the samples was added to the plate. The wells with only cells and culture medium constituted the negative control while those containing cells and culture medium supplemented with 10 µL of podophyllotoxin at 50 µM represented the positive control and the plate was incubated at 37˚C, 5% CO₂ for 48 hours. After the treatment of cells with the different compounds, 10 µL of resazurin solution (0.15 mg/mL) was added and the fluorescence signal was allowed to develop for 3 - 4 hours, after which the optical density was read using an infinite microplate reader (Tecan) at a wavelength of excitation and emission of 530 and 590 nm, respectively. The selectivity index (SI) is defined as the ratio of the Vero cells toxicity to the cercaricidal activity and was determined by dividing the CC₅₀ values by the LC₅₀ values of the active compounds. It reflects the effectiveness against target parasites while minimising toxicity to host cells.

2.5. Methodology for DFT Analysis

Theoretical analysis of structural and electronic properties of the synthesised ligands (L¹, HL² and L³) and their Zn(II) complexes was carried out using the hybrid meta exchange correlational functional m06 [55] associated with 6 - 31++g (d, p) basis set [56]. This basis set contains polarised functions that improve the flexibility of the basis set and diffuse functions, which are appropriate to handle anions and high electronegative elements such as chlorine. These functions are essential to accurately determine the geometry, the energy and non-covalent interactions of the investigated system. Although the m06 functional already includes long-range correction, and additional dispersion correction was included through Grimme’s D3 model [57] to improve on the optimised geometry. All calculations were carried out in the vacuum with the aid of Gaussian 16, Revision C.01 package [58] while GaussView 6.0.16 [59] visualisation program was used to build the structures and to analyse the obtained results. The starting molecular geometries for the optimisation process were proposed based on information obtained from spectroscopic and physicochemical characterisation of the compounds. The molecules were optimised using tight criteria and the values of the frequencies of the normal modes of vibration of the optimised structures were determined at the same levels of theory to ensure no imaginary frequency. The studied molecules were considered neutral and were optimized following the restricted scheme as they are all closed shells. Bader’s quantum theory of atom in molecules [60], reduced density gradient method [61] and Wiberg bond index implemented in Multiwfn 3.8 were used to characterise metal-ligand interactions. The distribution of HOMO and LUMO and the molecular electrostatic potential (MEP) map [62] [63] at the surface of optimised molecules were used to investigate their electrophilic and nucleophilic regions. HOMO-LUMO gap ($\Delta E_g=E_{LUMO}-E_{HOMO}$ ), chemical potential ($\mu =\frac{E_{LUMO}+E_{HOMO}}{2}$ ), electronegativity ($\chi =-\mu$ ), ionisation potential ($IP=-E_{HOMO}$ ), electron affinity ($EA=-E_{LUMO}$ ), chemical hardness ($\eta =\frac{E_{LUMO}-E_{HOMO}}{2}$ ), chemical softness ($S=\frac{1}{\eta }$ ), electrophilic index ($\omega =\frac{{\mu }^2}{2\eta }$ ) and the maximum electron density ($\Delta N_{max}=-\frac{\mu }{\eta }$ ) were used to predict the interaction of the study’s compounds with their environment [64]-[66].

2.6. Docking Computing Details

Docking experiments were conducted to assess the binding affinity between the active compounds and four important protein drug targets of schistosomes using the command prompt of the web server CB-DOCK 2 (https://cadd.labshare.cn/cb-dock2/php/blinddock.php). CB-Dock is a protein-ligand blind computational biology docking server which employs a protein-surface-curvature-based cavity detection approach to identify cavities or binding sites on the surface of given proteins, to guide the molecular docking with Auto Dock Vina [67].

The 3D structure of the two different proteins of S. mansoni Sm thioredoxin glutathione reductase SmTGR (PDB code: 2X8H) and Sm purine nucleoside phosphorylase SmPNP (PDB code: 3FAZ) were received from the Protein Data Bank (PDB) (https://www.rcsb.org/); while Schistosoma epidermal growth factor receptor SER (Uniprotkb code: Q26569) and Schistosoma mansoni cercarial elastase 2a SmCE2a (Uniprotkb code: Q8MUV8) protein sequences were extracted from Uniprotkb, and their 3D structures were determined using the I-TASSER online server (https://zhanggroup.org/I-TASSER/). I-TASSER server implements the I-TASSER-based algorithms for protein structure and function predictions. It detects structure templates from the Protein Data Bank by a technique called fold recognition. To evaluate the docking of each Schistosoma protein and the different compounds with the CB-Dock server, the PDB format of the protein and the PDB file of the query compound were uploaded into the CB-Dock server and the “Auto-blind docking” parameter was selected. CB-Dock2 improves protein-ligand blind docking by integrating cavity detection, docking and homologous template fitting. This server utilises fitdock doc-king method and AutoDock Vina for template-independent blind docking and the results are given as vina score (ranges from −3.8 to −18) [67]. The visualisation and analysis present the results, which can be visualised and analysed by interactive NGL Viewers for 3D structures and 2D sequences, together with abundant information about binding sites, template structures, binding scores, contact residues, docking centre, and cavity volume.

3. Results and Discussion

3.1. Spectroscopic Characterisation of the Schiff Base Ligands and Their Zinc(II) Complexes

The IR spectra provide very interesting information about the nature of functional groups present in the ligand, which can be attached to the central metal atom. Selected IR frequencies of ligands and complexes are presented in Table 1. Figure S1 (see supplementary images) shows the IR spectra of ligands and Zinc(II) complexes. The spectra of the ligands showed characteristic prominent bands at 1593.8, 1596 [52] and 1588 cm⁻¹ due to the azomethine band (C=N) respectively for L¹, HL² and L³ and the O-H stretching and bending vibrations in HL² were assigned to bands at 3368 and 1235 cm⁻¹ respectively. C=N imine vibrations in some reported quinolines Schiff base derivatives were assigned in the range of 1550 - 1650 cm⁻¹ [20], 1594 cm⁻¹ [68] and the aromatic O-H stretching and bending vibrations were reported in the range 3367 - 3371 cm⁻¹ [69], 3377 cm⁻¹ [52] and 1240 cm⁻¹ [52], 1282 cm⁻¹ [70] or 1300 cm⁻¹ [71] respectively. Upon coordination, the imine bonds of the Zn(II) complexes 1, 2 and 3 shifted to the higher vibrations of 1594.3, 1597 and 1596 cm⁻¹ respectively, indicating the participation of that group in coordination. In the spectra of each ligand, two mid-intensity bands appear at (1140, 750), (1085, 747) and (1144, 744) cm⁻¹, which are assigned to the pyridine or quinoline ring bending and outside the plan deformation respectively [72]. These bands are shifted to the higher values by 2 - 12 cm⁻¹ in the spectra of the Zn(II) complexes, indicating that the quinoline and pyridine nitrogen are involved in coordination with the metal ion. Displacements to higher values of 1241 cm⁻¹ in the zinc complex of HL² also appear for the C-O bending of the hydroxyl group, indicating coordination of the oxygen to the metal through deprotonation [70]. All the IR data support the idea that in (1) and (3) complexes, the ligands L¹ and L³ works as tridentate NNN, while in (2), HL² acts as a tetradentate NNNO, being coordinated through the azomethine nitrogen, the pyridine and quinoline nitrogen and the hydroxyl oxygen. New weak intensities, non-ligand bands are observed in the region 417, 410 cm⁻¹ and 419, 498 cm⁻¹ in (1), (3) and (2) respectively, indicating the formation of M-N and M-O bonds [70] [73].

Table 1. Selected IR frequencies (cm⁻¹) of ligands and Zn(II) complexes.

The ¹H NMR spectra of the ligands and their Zn(II) complexes were determined. In the free ligand, the distinct peak of the azomethine HC=N was observed at 8.83, 8.80 [52] and 8.30 ppm, for L¹, HL² and L³ respectively. A significant downfield shift of this proton to 8.90, 8.84 and 9.00 ppm was detected in complexes (1), (2) and (3). These results indicated the coordination of the nitrogen of the azomethine to the metal ion. The chemical shift at 10.23 ppm, corresponding to the proton of the hydroxyl group of HL², disappeared upon complex formation. The deprotonation of this proton leads to the coordination of the oxygen atom toward the Zn(II) ion in complex 2. Moreover, chemical shifts in the range of 6.76 - 8.72 (6.93 - 8.25 ppm) [52] and 6.83 - 8.56 ppm assigned to protons of quinoline rings of ligands L¹, HL² and quinoline and pyridine rings of L³, also presented significant shifts in the range 6.78 - 9.17 ppm and 7.61 - 8.78 ppm upon the formation of complexes 1, 2 and 3 respectively. This result could affirm the binding of the nitrogen donors of this heterocyclic aromatic and pyridine rings to the zinc metal ion.

Electronic spectral analysis of ligands exhibited three major absorption bands in the range 245 - 250 nm, 280 - 300 nm and 380 - 390 nm, that may be attributed to the π-π* transitions of the aromatic system, π-π* transitions of the azomethine and n-π* transition due to the lone pair electrons of the azomethine N [68]. Electronic spectra of Zn(II) complexes showed transition bands in the ranges of 255 - 275 nm, 283 - 330 nm, confirming their intra-ligand charge transfer transitions. Additional new bands at 585, 590 and 395 nm in complexes (1), (2) and (3) respectively were attributed to metal-ligand charge transfer (CT) [69]. The observed shift associated with the decrease in intensity of the absorption bands may be attributed to the coordination of the ligand heteroatoms to the metal centre. Figure S2 (see supplementary images) shows the electronic spectra of ligands and Zn(II) complexes. The lower values of 15.1, 35.9 and 17.8 S∙cm²∙mol⁻¹ of molar conductivity in DMSO for (1), (2) and (3) respectively, indicated that the complexes are non-electrolytes in solution [74]. Based on spectroscopic analyses, L¹ and L³ ligands can be considered as neutral tridentate with NNN-donor site in their Zn(II) complexes (1) and (3), giving the proposed five coordinate geometries. On the other hand, HL² ligand in its Zn(II) complex (3) shows a monobasic tetradentate NNNO-donor site, giving also a proposed five coordinate geometry.

3.2. Thermogravimetric Analyses of Zn(II) Metal Complexes

The Zn(II) complex (1) was thermally decomposed in two steps. The first decomposition step gave an observed mass loss of 7.98% (calc. 7.94%), which occurred within the range from 25 to 180˚C. This step might be attributed to the loss of two molecules of crystallization water. The second decomposition step occurred over the range 200˚C - 560˚C with two DTA exothermic peaks in the 410˚C - 460˚C and 470˚C - 510˚C, in which the complex lost the organic fragments, Cl₂ and NO₂ with an observed mass loss of 85.06% (calc. 85.23%). The horizontal line in the TGA beyond 590˚C indicated the formation of ZnO residue as the final product of decomposition.

The Zn(II) complexes (2) and (3) gave a decomposition pattern of three stages. The first step of decomposition between 25˚C - 140˚C and {25˚C - 180˚C} with weight loss observed of 3.65% (calc. 4.34%) and {4.31% (calc. 4.51%)} respectively, can be attributed to the loss of one water molecule of hydration. The second step of decomposition was between 140˚C - 290˚C and {200˚C - 470˚C} with an observed weight loss of 10.09% (calc. 10.62%) and {20.36% (calc. 20.42%} respectively, which can be attributed to the loss of one molecule of CO₂ and {1/2 chlorine and NO₂}. The third step of decomposition between 300˚C - 600˚C and {480˚C - 600˚C,} with an observed weight of 82.77% (calc. 82.00%) with two exothermic DTA peaks in the range 460˚C - 520 C and 570˚C - 600˚C and {64.54% (calc. 64.31%) with an exothermic DTA in the rage of 580˚C - 680˚C}, that may correspond to the loss of NO, Cl₂ and two C₁₀H₇N molecules and {the loss of chlorine and complete decomposition of the ligand} respectively. No further mass loss was observed beyond 600˚C and 700˚C respectively in 2 and 3 where the presence of a horizontal line indicated that zinc oxide may be the residual final product. The TGA analyses generally show the presence of water molecules of crystallisation and confirm the proposed general formula of the Zn complexes. Figure S3 (see supplementary images) shows the TGA graph of Zn(II) complexes.

3.3. Biological Evaluation

3.3.1. Cercaricidal Activity

The anticercarial activity was examined for the ligands and their metal complexes against the S. mansoni cercariae obtained from Biomphalaria pfeifferi snails. The lethal concentration LC₅₀ and the median cytotoxic concentration CC₅₀ of the active compounds are presented in Table 2. The results of anticercarial activity show that the ligands HL² and L³ show a very good to good activity with an LC₅₀ of 1.23 and 19.92 µg/ml [75].

Table 2. Median lethal concentrations (LC₅₀), median cytotoxic concentration (CC₅₀) and selectivity index (SI).

The significant difference in potency between the two ligands suggests that specific structural features or mechanisms of action may be responsible for the observed cercaricidal activity. This difference may be due to the presence of one hydroxyl group in HL², which increases its lipophilicity as it has been reported that large heteroaromatic moieties are generally more effective against the larval stage of S. mansoni [76]. Upon complexation, 2 with LC₅₀ of 1.14 µg/ml shows increasing activity compared to the free ligand (1.23 µg/ml). Among the tested compounds, HL² and its Zn(II) complex 2 show higher activity and can be considered as lead compound for antischistosomal activity. L¹ and its complex, even at higher concentrations, were inactive compared to L³ with 2-picolylamine derivative. This increase in activity with the CH backbone was also observed in the catalytic activity of Ni(II) complex compared to the phenyl backbone [16].

3.3.2. Cytotoxicity Activity

To evaluate the safety profile of the compounds, the cytotoxic assay of the active compound was performed on Vero cells. According to the US National Cancer Institute guidelines, a median CC₅₀ > 30 µg/mL is non-cytotoxic and <30 µg/mL is cytotoxic [77]. Compounds are generally considered to have significant cytotoxic potential when the CC₅₀ is ≤30 µg/ml [78]. Since the observed CC₅₀ exceeds this threshold, the compounds do not meet the NCI’s criterion for strong cytotoxic activity. Also, it can be noticed that the complexation with Zn(II) metal decreases the cytotoxicity CC₅₀ of the ligand. The selectivity index of HL² of 53.10 was higher than that of complex 2 of 44.83, but both are in the similar range of non-cytotoxicity, indicating that HL² and its Zn(II) complex 2 have the best SI among the tested compounds and can be potential candidates for therapeutic development against schistosomiasis.

3.4. DFT Studies

3.4.1. Geometry Optimisation

The optimised geometry of the ligands and their coordination compounds at 6 - 31++g (d, p)/m06-D3 level of theory (Figure 1) shows that L¹ and L³ act as a tridentate NNN-donor ligand and HL² acts as a tetradentate NNNO-donor ligand. In all these complexes, Zn²⁺ ions are penta-coordinated, and it is found in trigonal-bipyramidal geometry for 1 and 3 and in square-pyramidal geometry in 2. These results are confirmed by the molecular graphs of the studied compounds (Figure 1) derived from Bader’s quantum theory of atom in molecules [60], which show the presence of bond critical point (yellow balls) and bond path (orange lines) between nitrogen/oxygen/chlorine atoms and Zn(II) ions. The properties of the electron density function at the BCP of various coordination bonds were used to characterise metal-ligand bonds in the study compounds. Therefore, the electronic density, (𝜌(𝑟)), the Laplacian of the electronic density function, (∇²𝜌(𝑟)), the total electronic density, (𝐻(𝑟)) and the negative ratio of 𝐺(𝑟) and 𝑉(𝑟)) were calculated at the various BCPs and the results are reported in supplementary Table S1. It was found that for all coordination bonds in these compounds: $0.039076\le \rho \left(r\right)\le 0.082841,$ ${\nabla }^2\rho \left(r\right)>0$ , $H\left(r\right)<0$ , $0.5<-\frac{G\left(r\right)}{V\left(r\right)}<1$ which

Figure 1. Optimised geometries of ligands and Zn(II) complexes (a) and their molecular graphs associated to their NCIs iso-surface (b) (L2 = HL2; L1Zn = 1; L2Zn = 2; L3Zn = 3).

are characteristic of intermediate interactions that is partial covalent bonds. The 3D isosurface of non-covalent interactions of the various compounds derived from the reduced density gradient method combination [79], combined with molecular graphs (Figure 1), reveals the existence of weak van der Waals interactions (green surfaces) such as H---H in all the investigated compounds and C-H---Cl in all the Zn(II) complexes. Strong O-H---N hydrogen bond is found in HL² while steric repulsive interactions are observed at the centre of the various rings present in the ligands as well as in the complexes.

The values of selected bond length and bond angles of the optimised geometry of ligands and Zn(II) complexes are reported in supplementary Table S2. From these values, 2 has the shortest N4-Zn²⁺, N4’-Zn²⁺ and Cl-Zn²⁺ bonds, whereas it has the longest values of the N1-Zn²⁺ bond. This result is attributed to the fact that HL² is deprotonated in 2, which enhances the ligand-metal charge transfer in the concerned complex. The formation of 2 is governed by the formation of the O-Zn²⁺ bond as the O atom is negatively charged; therefore, it is formed in such a way to maximise the interactions between O and N4 atoms and the Zn(II) ion. This explains the following decreasing order of the coordination bonds in 2: N1-Zn > N4’-Zn > N4-Zn > O-Zn. The results of the bond length also show that, apart from the N1-Zn bond, N-Zn bonds are shorter in 1 compared to 3 and Cl-Zn bonds are similar in the two complexes. The trigonal-bipyramidal and square-pyramidal geometries around the Zn(II) ions are all distorted due to the constraint of the formation of the five members’ ring.

From the data presented in Table S2, it can also be concluded that, although the various rings are planar, the three ligands are not planar, neither in the isolated form nor in the complexes. The calculated values of bonds length and angles within the coordination sphere were closer to experimental and theoretical results of coordination compounds with a similar environment found in literature [80]. The strength of the various metal-ligand bonds was investigated using the Wiberg bond index (WBI) calculated as implemented in the Multiwfn 3.8 software. The variation of the Wiberg bond index of the coordination bonds of the different complexes is shown in supplementary Table S1. From these values, Zn-Cl bonds have the highest value of WBI (which increases in the order L¹Zn < L³Zn < L²Zn), followed by Zn-O bonds. In general, L¹Zn has the lowest values of WBI of Zn-X (with X = N, O and Cl) while L²Zn has the highest values. These results suggest that metal-ligand bonds are stronger in L2Zn, followed by L³Zn; the interaction between L¹ and Zn is less intense due to the small ligand-metal charge transfer in this complex. This observation may explain the fact that Zn(II) ion has the highest atomic charge (1.333) in L¹Zn compared to 1.057 and 0.851 in L²Zn and L³Zn respectively. The analysis of the figure suggests that the strength of Zn-N increases in the order Zn-N1 < Zn-N4 < Zn-N4’ in L¹Zn and L²Zn and Zn-N4 < Zn-N4’ < Zn-N1 in L³Zn.

3.4.2. Frontier Molecular Orbitals (FMOs)

Frontier molecular orbitals in general and particularly the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) play a key role during chemical reactivity and their properties are usually used to predict the chemical reactivity of molecules and to some extent their biological activities. The distribution of HOMO and LUMO on the molecule is related to the regions of the molecule with electron excess and electron deficiency respectively. The energy of the HOMO and LUMO is generally associated with the ability of the molecule to donate and to accept electrons respectively, whereas the energy difference between these orbitals mainly characterises the easiest with which a molecule interacts with its environment. The distribution of the FMOs of both ligands and their Zn(II) coordination compounds is presented in Figure 2. It is observed that both HOMO and LUMO of the L³ ligand are concentrated on the surface of the aldehyde rings, while for L¹, the LUMO is concentrated on the aldehyde ring and the HOMO on the aminoquinoline ring. For HL², the LUMO is concentrated on the aldehyde ring and the HOMO on the surface of the two quinoline rings. For L¹Zn and L³Zn, the HOMO is totally carried by the Cl atoms, while in L²Zn, it is carried by the quinoline ring linked to the deprotonated O atom. The LUMO of the three Zn(II) complexes is distributed on the overall surface of the ligands.

The values of HOMO-LUMO energy gap ($\Delta E_g$ ) of the investigated compounds are also reported in Figure 2. From these values, HL² is the most reactive of the three ligands, while L³ is the least reactive. Therefore, the addition of the OH group in L¹ has a destabilising effect although it favours the formation of a strong hydrogen bond with the nitrogen atom of the same ring. The high instability of L¹ and HL² compared to L³ is attributed to electrostatic repulsion between the electron lone pairs of the various heterogeneous atoms, which are found in the same region of space in those ligands. For L³, thus to the tetrahedral geometry of Csp³, only two (2) of these nitrogen atoms are found in the same region (N1-C6’-C5’-N4’ = 90$$ \mathit{˚}$$). The resulting Zn(II) complexes are less stable than their isolated ligands. However, their stability follows the same order of isolated ligands, i.e., L²-Zn is the least reactive coordination compound, followed by L¹-Zn.

Figure 2. HOMO and LUMO distribution of ligands and their Zn(II) complexes.

Global reactive descriptors of the studied compounds were determined using the conceptual DFT based on the Kopman’s approximation. Table 3 summarised the values of the ionisation potential (IP), electron affinity (EA), electronegativity (χ), chemical potential (µ), hardness (η), softness (S) and electrophilicity (ω) index and the maximum number of electrons (ΔN_max) that each compound can acquire. The calculated values of the chemical potential and the hardness show that complexes are less stable than their corresponding isolated ligands, as predicted by the HOMO-LUMO energy gap.

The hardness of these compounds increases as follows: L²Zn < L¹Zn < L³Zn < HL² < L¹ < L³. The electrophilicity index reveals that complexes have higher electrophilic power compared to ligand and acquire more electron density than isolated ligands. Electrophilic power and the maximum number of electrons each of these compounds can acquire increase according to the following order: L³ < L¹ < HL² < L³Zn < L¹Zn < L²Zn. For the investigated compounds, the reactivity varies in the same order as the electrophilic power. The electron affinity of the studied compounds increases in the same order as the electrophilic index, except that the position of L¹Zn and L²Zn is interchanged. The ionisation potential increases as follows in the series: L²Zn < L¹ = HL² < L¹Zn < L³Zn < L³. From these results, L²Zn is the softest of the studied molecules; it has the lowest value of IP and the highest value of ω, while L³ is the hardest compound with the highest IP and lowest ω.

Table 3. Values (in eV) of some descriptors of the global reactivity of the studied compounds.

3.4.3. Molecular Electrostatic Potential (MEP)

The molecular electrostatic potential map of the studied compounds was used to determine their regions with positive and negative electrostatic potential and therefore predict their interactions with other molecules during chemical reactions and biological activity [81]. The MEP map of the studied compounds is presented in Figure 3.

It is observed that for all the investigated compounds, nitrogen, oxygen and chlorine atoms carry the negative electrostatic potential while C and H carry the positive electrostatic potential. N1 atom carries the most negative electronegative potential of the three ligands, followed by N4’ in L1 and L2. In L3, N4 is more nucleophilic than N4’, probably due to electron delocalisation. The results suggest that the reaction between the three ligands and the Zn²⁺ starts with the interaction between N1 and Zn²⁺, which clearly explains why N1-Zn²⁺ is the shortest bond, followed by N4’-Zn²⁺ in L1Zn and L3Zn. The trend of coordination bond length in the L2Zn is attributed to the fact that HL² is deprotonated. In the complexes, the highly negative potential is carried by Cl atoms. However, it is important to note that the blue colour of the surface around H atoms is more intense in the complexes

Figure 3. MEP of ligands and their Zn(II) complexes.

than in the isolated ligand, which suggests that the electrophilicity of the concerned increases upon chelation. The H atom of the imine function is the most acidic proton of all the studied compounds, and it is more acidic in the complexes due to the Ligand-metal charge transfer that occurs during the chelation process.

3.5. Molecular Docking Simulations

To assess the molecular interactions and the binding affinity of the active compounds with 4 important target proteins of schistosomes parasites: Schistosoma mansoni Cercarial Elastase 2a (SmCE2a), Schistosoma Epidermal growth factor Receptor (SER), Schistosoma mansoni Thioredoxin Glutathione Reductase (SmTGR), Schistosoma mansoni Purine Nucleoside Phosphorylase (SmPNP), molecular docking predictions were performed, and the results are presented in Table 4 and Figure 4. The results of docking with the active compounds were compared to the results obtained with the known inhibitors of the targeted proteins (PMSF, TyrphostinAG1478, Auranofin, Neolignan).

Table 4. Binding energies (kcal/mol) for the molecular docking simulation of the active compounds with the schistosomes drug targets proteins.

The results in Table 4 suggest that all the active compounds have good binding affinities with the active site of the target proteins, highlighted by their low binding energies ranging from −6.1 kcal/mol to −10.1 kcal/mol. Moreover, compound 2 emerged as the best compound with strong binding affinities with very low binding energies (−8.3 (SmCE2a), −7.9 (SER), −7.9 (SmTGR), −10.1 (SmPNP) kcal/mol) across targets as compared to the two ligands. Also, SmPNP shows strong interactions with all three compounds −8.5 kcal/mol (L³), −9.5 kcal/mol (HL²), −10.1 kcal/mol (2). Indeed, the binding affinities of these three compounds were comparable to or stronger than that of the known inhibitors of the four targets: −4.1, −6.7, −7.2, −8.7 kcal/mol for SmCE2a-PMSF, SER-TyrphostinAG1478, SmTGR-Auranofin and SmPNP-Neolignan respectively.

The contact residues with compounds and known inhibitors in the active site of the proteins are presented in supplementary Table S3. The two ligands formed interactions with SmTGR through several contact residues, including CYS 154 and CYS 159, which are important amino acid residues in the active site of SmTGR responsible for the enzyme activity [82]. Furthermore, compounds L³, HL² and 2 formed 3, 5 and 1 hydrogen bonds respectively with THR 158 TYR 138 ALA 256 (L³), CYS 159 THR 442 LEU 441 ASP 443 THR 442 (HL²) and VAL 469 (2). The greater the number of hydrogen bonds, the higher the binding efficiency and inhibition [83]. It has been demonstrated that treatment of S. mansoni with a TGR inhibitor resulted in 100% mortality of the parasite in vitro, implying that these compounds can be used against schistosomiasis [84]. The primary amino acids involved during Ligand-SmPNP interaction are CYS 33, ARG 86, HIS 88, TYR 90, ALA 118, ALA 119, PRO 200, TYR 202, GLU 203, VAL 219, MET 221, THR 244, ASN 245, PRO 257, and HIS 259. Similar interacting residues were reported [85]. 4, 4 and 2 hydrogen bond interactions with the amino acids residues MET 221, ALA 118, VAL 262, SER 35 (L³), VAL 262, SER 35, LEU 263, HIS 259 (HL²) and MET 221, HIS 259 (2) were formed in the active site while they have been reported to be part of important amino residues in the active site of SmPNP responsible for the enzyme activity [86]. Therefore, due to all these facts, all the active compounds may be good inhibitors of SmPNP, leading to a stop or a decrease in purine base synthesis necessary for nucleic acid synthesis and consequently the parasite death [85].

SmCE2a is another important target, an enzyme responsible for degrading skin host tissue during cercariae penetration to facilitate invasion [87]. All the compounds docked to this enzyme had lower binding energies −6.3 kcal/mol, −7.7 kcal/mol, −8.3 kcal/mol than its known inhibitor PMSF, highlighting their strong binding affinities to SmCE2a better than PMSF, which was reported to inhibit cercarial elastase in vitro [88]. The binding modes and interactions of ligands-SmCE2a show that the active compounds interact with the enzyme through important bond formation, including 3 hydrogen bonds with HIS 67, TYR 104, SER 217 for L³-SmCE2a, 5 with HIS 67, TYR 107, ASP 125 for HL²-SmCE2a and 1 with SER 217 for 2-SmCE2a, thereby contributing to the docked complexes stability and reinforcing the binding affinity.

Schistosoma epidermal growth factor receptor SER are receptor tyrosine kinases (RTK) that represent key molecules for the development and reproductive activity processes in schistosome worms, and targeting RTK, including SER, can affect the survival of S. mansoni [89]. From the docking results of this enzyme, the low binding energies as compared to the known inhibitor Tyrphostin AG1478 suggest that the active compounds bind strongly to SER, added to the hydrogen bonds they formed with GLN 147, CYS 270, GLY 280, GLY 268, ASP 90, MET 277, GLY 280, that stabilise the complexes, thus could be potential inhibitors of SER. Figure 4 illustrates the binding modes and interactions identified in Table S3.

Figure 4. Docked compounds HL² and L³, 2 and inhibitors in the active sites of (a) SmCE2a, (b)SER, (c) SmTGR, and (d) SmPNP.

3.6. In Silico Physicochemical and Pharmacokinetics Predictions

After evaluating the binding affinities of the compounds with the schistosome targets, the three active compounds and Praziquantel (the drug used against schistosomiasis) were subjected to pharmacokinetic property prediction. This involved an examination of their physicochemical properties and drug-likeness characteristics. The predicted properties of active compounds and Praziquantel calculated using the SwissADME web server (http://www.swissadme.ch) and presented in Table 5 provide insights into their potential physicochemical properties, lipophilicity, water solubility, pharmacokinetics, drug likeness and medicinal chemistry profiles [90].

Table 5. Pharmacokinetic properties predictions of the active compounds.

The selection of new drugs and formulations is based on the physicochemical properties, which include the molecular weight, solubility, molar refractivity, topological polar surface area and hydrogen bonding capacity. These properties are essential factors for finding new drug candidates and should agree with different rule-based filters such as drug-likeness rules [90]. There are some drug-likeness rules [91] which can be described as Lipinski’s rule (MW ≤ 500, Mlog P ≤ 5, H-acc ≤ 10, H-don ≤ 5); Ghose’s filter (160 ≤ MW ≤ 480, −0.4 ≤ Wlog P ≤ 5.6, 40 ≤ MR ≤ 130, and 20 ≤ atoms ≤ 70); Veber’s filter (R-bonds ≤ 10, and TPSA ≤ 140), and Egan’s filter (Wlog P ≤ 5.88, and TPSA ≤ 131.6). Based on these rules, we found that the three active compounds agree with the four drug-likeness rules (Lipinski, Ghose, Veber, and Egan rules) and these derivatives can be accepted as lead compounds for the development of a new drug candidate [92]. A compound is like a drug if the drug-likeness score is valued positively, or the compound is non-drug if the drug-likeness score is valued negatively [93].

The drug-likeness score analysis is based on the positive and negative values of the target compounds, where the Oral drugs have fewer H-bond acceptors, donors, and rotatable bonds [94] and in the same way, compounds with a TPSA of >140 Ų would be poorly absorbed with less than 10% fractional absorption. Thus, compounds L³, HL², 2 with rotatable bonds 3, 2, 0 respectively and hydrogen bond donors and acceptors < 4 fulfil all the required criteria to be well absorbed. The consensus log P o/w of the three predicted compounds of 2.77, 3.47, and 2.41 for L³, HL², and 2 respectively, indicated that the compounds are lipophilic, with HL² being the most lipophilic, given that a higher log Po/w value indicates higher lipophilicity and can be easily absorbed through barriers because of the lipid nature of targets [95]. Furthermore, aqueous solubility is an important parameter influencing absorption and facilitating the ease of drug handling and formulation [96]. Compound L³ was classified as soluble, while HL² and 2 exhibited moderate solubility from their consensus Log S values. Interestingly, the three investigated compounds had high gastrointestinal absorptions and were blood-brain barrier permeant, meaning that they have a good absorption.

CYP enzymes and Permeability glycoprotein (P-gp) can process small molecules synergistically to improve the protection of tissues and organisms [97]. P-gp expressed in the intestinal epithelium pumps xenobiotics such as drugs back into the intestinal lumen and in the capillary endothelial cells composing the blood-brain barrier, where it pumps them back into the capillaries. While CYP P450 enzymes are responsible for drug metabolism, which can lead to drug toxicities and reduced pharmacological effect [98]. From the results, the three compounds were P-gp substrates, meaning that these compounds interact with P-gp, limiting intracellular drug accumulation in the organism. On the other hand, the compounds were either inhibitors or noninhibitors of the different isoforms of CYP P450, though it was reported that only CYP2D6 and CYP3A4 are responsible for drug metabolism [96]. Only 2 is not an inhibitor of CYP2D6, while the rest, including the reference Praziquantel, inhibited it and all the three compounds, including Praziquantel, were inhibitors of CYP3A4. Overall, SwissADME has computational filters that include Lipinski, Ghose, Egan, Veber, and were developed by leading pharmaceutical companies and computational chemist to evaluate the drug-likeness of small molecules.

Generally, drug-likeness is predicted using Lipinski (rule of five) and Veber’s rules, consequently, a compound that can be used as orally active drug should not present more than one violation [99]. Lipinski’s Rule of five is the pioneer filter that is associated with 90% of the oral drugs that have reached phase II clinical trials [93]. This Rule of 5 predicts that poor absorption or permeation is more likely when there are more than 5 H-bond donors, 10 H-bond acceptors, molecular weight greater than 500, and the calculated Log P (CLog P) greater than 5 [100]. From the prediction, all the compounds complied with the four rules-based filters, and no one violated any of the rules, implying these compounds could be promising lead compounds for the development of good oral drugs. Moreover, a bioavailability score of 0.55 predicts the probability of these compounds to have at least 10% oral bioavailability in rat [101]. Additionally, the medicinal chemistry shows that these compounds can be easily synthesized step by step in the laboratory, highlighted by their low synthetic accessibility scores of 2.76 (L3), 2.73 (HL²), and 3.99 (2).

4. Conclusion

Three quinolinyl Schiff base ligands, L¹, HL² and L³, were synthesized, and they acted as tridentate NNN- or tetradentate NNNO-donors. Their Zn(II) metal complexes (1, 2 and 3) showed trigonal bipyramidal and square pyramidal proposed geometry based on the spectroscopic analyses. Density Functional Theory used to optimise the structure of ligands and their metal complexes was discussed. L³, HL² and 2 displaced very good antischistosomal activity, with compound 2 being the most active. In general, Zn(II) complex 2 shows slightly higher activity (1.14 µg/mL) than its ligand HL² (1.23 µg/mL), therefore they show more promising results as lead compounds against S. mansoni parasite. All the active compounds were also non-cytotoxic to the Vero cell line and both HL² and its Zn(II) complex 2 are in similar range with good selectivity index. Furthermore, molecular docking studies clearly indicated that the active compounds exhibited good docking scores against S. mansoni main target enzymes, with values comparable or better than their known inhibitors. Finally, the drug-likeness prediction showed that two active Schiff bases (L³, HL²) and the Zn(II) complex (2) agree with the four drug-likeness rules (Lipinski, Ghose, Veber, and Egan rules) and these derivatives can be considered as potential lead compounds for therapeutic development against S. mansoni parasite.

Acknowledgements

FMM is grateful to the FAIRE programme provided by the Cambridge Crystallographic Data Centre (CCDC) for the opportunity to use the Cambridge Structural Database (CSD) and associated software.

Supplementary Information

Figure S1. IR spectrum of ligands and Zn(II) complexes.

Figure S2. Electronic spectra of ligands and Zn(II) complexes.

TGA of L2Zn

TGA of L1Zn

TGA of L3Zn

Figure S3. Thermograms of Zn(II) complexes.

Table S1. Value of Wiberg bond index (WBI) and of some properties of the electronic function at BCPs of coordination bonds in the studied complexes.

Table S2. Values of selected parameters of the optimized geometry of ligands and Zn(II) complexes.

Table S3. Contact residues with compounds in the active sites of the proteins.

Conflicts of Interest

The authors declare no conflicts of interest regarding the publication of this paper.

References