Effect of Terminal Substituents of 2-Aminobenzimidazoles on Non-Covalent Molecular Interactions in Their Transition Metal Coordination Compounds. Evaluation of Their Biological Activity

1. Introduction

Since the pioneering work of cis-platinum and other related platinum compounds [1], biologically active coordination compounds became a new field for bioinorganic chemistry. Platinum is still part of many newly synthetic compounds that attempt to use the covalent mechanism of cis-platinum but using different ligands. [2,3] Although cis-platinum and other metallic compounds are quite useful and effective, their well-known toxicity [4,5] is still a problem to overcome by the design of new compounds with less side effects [6]. There is an interest on focusing in coordination compounds of essential trace elements, as Fe, Co, Ni, Cu and Zn [7-9], commonly found in biological systems as part of metalloproteins or cofactors for many enzymes [10]. The use of trace elements aims to take advantage of the pre-existent metabolic routes for them, so they could be less toxic than heavier metals.

Considerable research has been done to elucidate the biological activity of coordination compounds and how they work inside human cells to achieve its goal. Different mechanisms have been proposed, mainly related to the molecular structure of the compounds, such as redox activity of the compound or its metal center, covalent bonding to DNA and other biomolecules, non-covalent interactions with biomolecules or a combination of the ones mentioned [11]. While redox activity and covalent bonding are more direct and unspecific, a mechanism based on non-covalent interactions may be more difficult to elucidate, but it will be more sensitive to changes in the structure of the coordination compound and the conformation adopted by the biomolecules [12-14].

DNA is an example of the different ways in which a single biomolecule may interact with transition metal coordination compounds. With B-DNA, the complexes may interact through the phosphate backbone of the strands, stablish interactions with the nitrogenated bases at the mayor or the minor grooves, intercalate aromatic moieties between two nucleotides or a combination of these possibilities. All these interactions led to conformational changes of this biomolecule, affecting its stability and those processes in which it is involved [15]. An example of the relevance of the specificity of non-covalent interactions is found with the interaction of coordination compounds and the quadruplex conformation of DNA, observed at the telomeres of chromosomes in guanine rich sections, and that has been related to cell aging and apoptosis [16-20]

The continuous study of non-covalent interactions has been relevant to the design of biologically active compounds [21]. We have been interested on the design of biologically active ligands and their corresponding transition metal coordination compounds, specifically focus on the influence of non-covalent interactions into the biological properties [22].

In a previous work with tinidazole (tnz) and its copper(II) and zinc(II) compounds, it was found that tetrahedral coordination compounds showed excellent antiparasitic activity [23,24]. An important factor for the activity of these compounds was the presence of a bifurcated intramolecular lone pair⋅⋅⋅π interaction (lp), between an O atom of the sulfone group with both imidazolic rings from the coordinated tnz ligands. This interaction stabilized the molecular tetrahedral structure, which was stable enough to be conserved in solution. The biological activity of the tinidazole copper(II) compounds was further investigated. Different counterions were used to generate tnz-based complexes of various geometries to study the influence of the geometry in the biological activity. Cyclic voltammetry and gel electrophoresis experiments were performed to evaluate their oxidative-damaging properties, their redox properties were attributable to both the ligand and the metal ion. Additionally, DNA-interacting ability and cytotoxicity of tnz copper(II) complexes were evaluated. These complexes interact with DNA by means of electrostatic interactions or/and groove binding, and the generation of ROS, in the presence of a reducing agent, induces DNA damage. Cytotoxicity studies with different cancer cell lines revealed that complexes [Cu(tnz)₂(µ-Cl)Cl]₂ and [Cu(tnz)₂Br₂] showed the highest cytotoxicity, while they were moderately toxic to normal cells [25, 26]

Most of the studied cytotoxic coordination compounds have been with chelating ligands, the results with non-chelating ligands, tinidazole [23-26] and clotrimazole [27,28], have shown that weak interactions, such as hydrogen bonding, electrostatic interactions, π stacking, lp⋅⋅⋅π or hydrophobic contacts, as well as geometry and redox properties, have an important role on the biological activity of coordination compounds with monocoordinated ligands.

Based on these results, we were interested to investigate 2-aminobenzimidazole sulfonated derivatives, the presence of the amino group has been found to be of great importance for interactions with or between biomolecules, thanks to the high hydrogen donor character of the group [12]. These interactions have been observed in previous studies of coordination compounds with the unsubstituted 2-aminobenzimidazole ligand, where the -NH₂ gives place to intramolecular hydrogen bonding with the coordinated halides or acetates stabilized the molecular structure [29,30]. In addition, the sulfone group have shown to give place to weak interactions, lp⋅⋅⋅π contacts, or the presence of a phenyl group to π⋅⋅⋅π stacking [31].

Herein we present the structural and spectroscopic characterization, as a non-covalent interactions analysis of the ethyl and phenyl sulfonated ligands, 2-amino-1-(2-phenylsulphonyl)ethylbenzimidazole (sfabz); 2-amino-1-(2-ethylsulfonyl) ethylbenzimidazole (seabz) and their coordination compounds. The antiproliferative activity of the obtained compounds was investigated.

2. Experimental

2.1. Materials and methods

2-aminobenzimidazole, phenylvinyl sulfone and ethylvinyl sulfone were purchased from Sigma-Aldrich and used without further purification. The metal salts K₂CO₃ (99%), CuCl₂·2H₂O (97%), CuBr₂ (98%), NiCl₂·6H₂O (99%), ZnBr₂ (>97%), HgCl₂ (98%) and CdCl₂·2.5H₂O (99%) were purchased from J.T. Baker, the salt ZnCl₂ (>97%), was obtained from Sigma Aldrich and the salt NiBr₂·3H₂O (99%) was purchased from Merck. All solvents were obtained from J.T. Baker. Both salts and solvents were used without further purification.

2.2. Synthesis of the ligands.

2.2.1. Synthesis of 2-amino-1-(2-phenylsulfonyl)ethylbenzimidazole (sfabz)

The ligand was synthesized by mixing phenylvinyl sulfone (3.3182 mmol, 0.5582 g), 2-aminobenzimidazole (3.3182 mmol, 0.4418 g), and K₂CO₃ (1.6591 mmol, 0.2293 g) in 10 mL of acetonitrile. The mixture was stirred under reflux for ten minutes, left to stand at room temperature and then filtered. The precipitate was washed with a concentrated solution of NH₄Cl and then with distilled water. Crystals suitable for single crystal X-ray diffraction were obtained by slow evaporation of a solution of the ligand in methanol. Yield: 92%. Anal. Calculated for C₁₅H₁₅N₃O₂S: C, 59.78%; H, 5.02%; N, 13.95%; S, 10.62%. Experimental: C, 59.58%; H, 4.82%; N, 14.19%; S, 10.76%. IR (ν cm^-1 ): 3434 ν_as(NH₂), 3341 ν_s(NH₂), 1664 ν(C=C)+ρ(NH₂)+ν(C2-N3), 1552 ν(C=N)+ρ(NH₂)+ν(C2-N10), 1286 ν_as(SO₂), 1140 ν_s(SO₂). RMN: δ 7.86 (H17, d, J= 7.4 Hz, 2H), 7.71 (H19, t, J= 7.4 Hz, 1H), 7.58 (H18, t, J= 7.4 Hz, 2H), 7.05 (H4, d, J= 7.5 Hz, 1H), 6.89 (H5, t, J= 7.5, 1H), 6.79 (H6, t, J= 7.5 Hz, 1H), 6.74 (H7, d, J= 7.5 Hz, 1H), 6.43 (H10, s, 2H), 4.28 (H11, t, J= 7.0 Hz, 2H), 3.72 (H12, t, J= 7.0 Hz, 2H).

2.2.2. Synthesis of 2-amino-1-(2-ethylsulfonyl)ethylbenzimidazole (seabz)

The ligand was prepared by mixing ethylvinyl sulfone (4.768 mmol, 0.5755 g) and 2-aminobenzimidazole (4.768 mmol, 0.6376 g) in 10 mL of acetonitrile. This mixture was stirred under reflux for 30 minutes, then was left to stand at room temperature. The precipitate was filtered and washed with 5 mL of ethylacetate. Crystals suitable for single crystal X-ray diffraction were obtained by slow evaporation of a solution of the ligand in methanol. Yield: 66%. Anal. Calculated for C₁₁H₁₅N₃O₂S: C, 51.24%; H, 6.06%; N, 16.30%; S, 12.44%. Experimental: C, 51.27%; H, 5.93%; N, 16.35%; S, 12.50%. IR (ν cm^-1 ): 3460 ν_as(NH₂), 3373 ν_s(NH₂), 1639 ν(C=C)+ρ(NH₂)+ν(C2-N3), 1549 ν(C=N)+ρ(NH₂)+ν(C2-N10), 1281 ν_as(SO₂), 1132 ν_s(SO₂). RMN: δ 7.13 y 7.15 (H4 y H7, dd, J= 7.4, 1.2 Hz, 2H), 6.94 (H6, td, J= 7.4, 1.2 Hz, 1H), 6.88 (H5, td, J= 7.4, 1.2 Hz, 2H), 6.48 (H10, s, 2H), 4.39 (H11, d, J= 7.2 Hz, 2H), 3.46 (H12, t, J= 7.2 Hz, 2H), 3.08 (H16, q, J= 7.4 Hz, 2H), 1.15 (H17, t, J= 7.4 Hz, 2H).

2.3. Synthesis of the coordination compounds

Coordination compounds of sfabz and seabz were synthesized by similar procedures. A mixture of the ligand and the metal salt in a methanol solution (10 mL) was stirred under reflux for 30 minutes. The reaction mixture was left to stand at room temperature, the obtained products were filtered and washed with cold ethanol. Details of the reaction conditions are discussed below.

2.3.1. [Ni(sfabz)₂Cl₂] (1)

A solution of the ligand (0.1507 g, 0.5 mmol) and NiCl₂·6H₂O (0.0594 g, 0.25 mmol) in 15 mL of ethanol was stirred under reflux for 30 minutes. The reaction mixture was left to stand at room temperature. The obtained product was filtered and washed with 10 mL of cold ethanol. Crystals suitable for single crystal X-ray diffraction were obtained preparing a solution in acetone of the solid obtained and leaving it to slow evaporation. Yield: 80%. Anal. Calculated for NiC₃₀H₃₂N₆O₅S₂Cl₂: C, 48.02%; H, 4.30%; N, 11.20%; S, 8.55%. Experimental: C, 47.62%; H, 3.56%; N, 10.80%; S, 8.01%. IR (ν cm^-1 ): 3387 ν_as(NH₂), 3305 ν_s(NH₂), 1645 ν(C=C)+ρ(NH₂)+ν(C2-N3), 1552 ν(C=N)+ρ(NH₂)+ν(C2-N10), 1292 ν_as(SO₂), 1141 ν_s(SO₂).

2.3.2. [Ni(sfabz)₂Br₂] (2)

A solution of the ligand (0.1507 g, 0.5 mmol) and NiBr₂·3H₂O (0.0681 g, 0.25 mmol) in 15 mL of ethanol was stirred under reflux for 30 minutes. The reaction mixture was left to stand at room temperature. The obtained product was filtered and washed with 10 mL of cold ethanol. The compound was dissolved in acetone and crystals suitable for single crystal X-ray diffraction were obtained by slow evaporation of the solvent. Yield: 71%. Anal. Calculated for NiC₃₀H₃₀N₆O₄S₂Br₂: C, 43.88%; H, 3.68%; N, 10.23%; S, 7.81%. Experimental: C, 43.56%; H, 3.20%; N, 10.10%; S, 7.66%. IR (ν cm^-1 ): 3418 ν_as(NH₂), 3307 ν_s(NH₂), 1643 ν(C=C)+ρ(NH₂)+ν(C2-N3), 1549 ν(C=N)+ρ(NH₂)+ν(C2-N10), 1290 ν_as(SO₂), 1146 ν_s(SO₂).

2.3.3. [Ni(seabz)₂Cl₂] (3)

A solution of the ligand (0.1267 g, 0.5 mmol) and NiCl₂·6H₂O (0.0594 g, 0.25 mmol) in 15 mL of ethanol was stirred under reflux for 30 minutes. The reaction mixture was left to stand at room temperature. The obtained product was filtered and washed with 10 mL of cold ethanol. The compound was dissolved in acetone and crystals suitable for single crystal X-ray diffraction were obtained by slow evaporation of the solvent. Yield: 89%. Anal. Calculated for NiC₂₂H₃₄N₆O₆S₂Cl₂: C, 39.31%; H, 5.10%; N, 12.50%; S, 9.54%. Experimental: C, 39.69%; H, 4.82%; N, 12.56%; S, 8.80%. IR (ν cm^-1 ): 3407 ν_as(NH₂), 3321 ν_s(NH₂), 1646 ν(C=C)+ρ(NH₂)+ν(C2-N3), 1559 ν(C=N)+ρ(NH₂)+ν(C2-N10), 1295 ν_as(SO₂), 1128 ν_s(SO₂).

2.3.4. [Ni(seabz)₂Br₂] (4)

A solution of the ligand (0.1267 g, 0.5 mmol) and NiBr₂·3H₂O (0.0681 g, 0.25 mmol) in 15 mL of ethanol was stirred under reflux for 30 minutes. The reaction mixture was left to stand at room temperature. The obtained product was filtered and washed with 10 mL of cold ethanol. Yield: 69%. Anal. Calculated for NiC₂₂H₃₅N₆O_6.5S₂Br₂: C, 34.31%; H, 4.58%; N, 10.91%; S, 8.37%. Experimental: C, 33.97%; H, 4.11%; N, 11.25%; S, 7.84%. IR (ν cm^-1 ): 3387 ν_as(NH₂), 3311 ν_s(NH₂), 1645 ν(C=C)+ρ(NH₂)+ν(C2-N3), 1550 ν(C=N)+ρ(NH₂)+ν(C2-N10), 1294 ν_as(SO₂), 1126 ν_s(SO₂).

2.3.5. [Cu(sfabz)₂Cl₂] (5)

A solution of 0.0426 g (2.5 mmol) of CuCl₂·2H₂O and 0.2260 g (7.5 mmol) of the ligand sfabz in 15 mL of ethanol was stirred under reflux for one hour. The reaction mixture was left to stand at room temperature. The obtained product was filtered and washed with 10 mL of cold ethanol. Crystals suitable for single crystal X-ray diffraction were obtained by slow evaporation of the remanent ethanolic reaction mixture. Yield: 65%. Anal. Calculated for CuC₃₀H₃₁N₆O_4.5S₂Cl₂: C, 48.29%; H, 4.19%; N, 11.26%; S, 8.60%. Experimental: C, 48.10%; H, 3.94%; N, 11.36%; S, 8.95%. IR (ν cm^-1 ): 3440 ν_as(NH₂), 3372 ν_s(NH₂), 1638 ν(C=C)+ρ(NH₂)+ν(C2-N3), 1546 ν(C=N)+ρ(NH₂)+ν(C2-N10), 1289 ν_as(SO₂), 1138 ν_s(SO₂).

2.3.6. [Cu(sfabz)₂Br₂] (6)

A solution of the ligand (0.1507 g, 0.5 mmol) and CuBr₂ (0.0558 g, 0.25 mmol) in 15 mL of ethanol was stirred under reflux for 30 minutes. The reaction mixture was left to stand at room temperature. The obtained product was filtered and washed with 10 mL of cold ethanol. Yield: 97%. Anal. Calculated for CuC₃₀H₃₂N₆O₅S₂Br₂: C, 42.69%; H, 3.82%; N, 9.96%; S, 7.60%. Experimental: C, 42.41%; H, 3.49%; N, 9.87%; S, 7.31%. IR (ν cm^-1 ): 3525 ν_as(NH₂), 3318 ν_s(NH₂), 1643 ν(C=C)+ρ(NH₂)+ν(C2-N3), 1557 ν(C=N)+ρ(NH₂)+ν(C2-N10), 1294 ν_as(SO₂), 1141 ν_s(SO₂).

2.3.7. [Cu(seabz)₂Cl₂] (7)

A solution of 0.0426 g (2.5x10^-4 mol) of CuCl₂·2H₂O and 0.1267 g (5x10^-4 mol) of the ligand seabz in 15 mL of ethanol was stirred under reflux for 30 minutes. The solvent was evaporated under heating and the precipitate was filtered and washed with 10 mL of cold ethanol. Yield: 80%. Anal. Calculated for CuC₂₂H₃₄N₆O₆S₂Cl₂: C, 39.11%; H, 5.08%; N, 12.45%; S, 9.47%. Experimental: C, 39.17%; H, 5.44%; N, 12.68%; S, 7.97%. IR (ν cm^-1 ): 3397 ν_as(NH₂), 3308 ν_s(NH₂), 1645 ν(C=C)+ρ(NH₂)+ν(C2-N3), 1558 ν(C=N)+ρ(NH₂)+ν(C2-N10), 1294 ν_as(SO₂), 1125 ν_s(SO₂).

2.3.8. [Cu(seabz)₂Br₂] (8)

A solution of 0.0558 g (2.5x10^-4 mol) of CuBr₂ and 0.1267 g (5x10^-4 mol) of the ligand in 15 mL of ethanol was stirred under reflux for 30 minutes. The solution of the reaction mixture was evaporated under heating. The precipitate was filtered and washed with 10 mL of cold ethanol. Yield: 81%. Anal. Calculated for CuC₂₂H₃₀N₆O₄S₂Br₂: C, 36.20%; H, 4.14%; N, 11.51%; S, 8.79%. Experimental: C, 36.40%; H, 4.46%; N, 11.66%; S, 8.18%. IR (ν cm^-1 ): 3388 ν_as(NH₂), 3321 ν_s(NH₂), 1639 ν(C=C)+ρ(NH₂)+ν(C2-N3), 1549 ν(C=N)+ρ(NH₂)+ν(C2-N10), 1289 ν_as(SO₂), 1123 ν_s(SO₂).

2.3.9. [Zn(sfabz)₂Cl₂] (9)

A solution of the ligand (0.1507 g, 0.5 mmol) and ZnCl₂ (0.0341 g, 0.25 mmol) in 15 mL of ethanol was stirred under reflux for 30 minutes. The reaction mixture was left to stand at room temperature. The obtained product was filtered and washed with 10 mL of cold ethanol. The compound was dissolved in acetone. Crystals suitable for X-ray diffraction were obtained by slow evaporation of the solvent. Yield: 84%. Anal. Calculated for ZnC₃₀H₃₁N₆O_4.5S₂Cl₂: C, 48.17%; H, 4.18%; N, 11.23%; S, 8.57%. Experimental: C, 48.17%; H, 3.35%; N, 11.28%; S, 8.39%. IR (ν cm^-1 ): 3390 ν_as(NH₂), 3315 ν_s(NH₂), 1646 ν(C=C)+ρ(NH₂)+ν(C2-N3), 1556 ν(C=N)+ρ(NH₂)+ν(C2-N10), 1293 ν_as(SO₂), 1142 ν_s(SO₂).

2.3.10. [Zn(sfabz)₂Br₂] (10)

A mixture of the ligand (0.1507 g, 0.5 mmol) and ZnBr₂ (0.0563 g, 0.25 mmol) in 15 mL of ethanol was stirred under reflux for 30 minutes. The reaction mixture was left to stand at room temperature. The reaction mixture was left to stand at room temperature. The obtained product was filtered and washed with 10 mL of cold ethanol. The compound was dissolved in acetone. Crystals suitable for X-ray diffraction were obtained by slow evaporation of the solvent. Yield: 78%. Anal. Calculated for ZnC₃₀H₃₀N₆O₄S₂Br₂: C, 43.52%; H, 3.65%; N, 10.15%; S, 7.28%. Experimental: C, 43.28%; H, 3.30%; N, 10.11%; S, 7.28%. IR (ν cm^-1 ): 3415 ν_as(NH₂), 3312 ν_s(NH₂), 1627 ν(C=C)+ρ(NH₂)+ν(C2-N3), 1552 ν(C=N)+ρ(NH₂)+ν(C2-N10), 1290 ν_as(SO₂), 1136 ν_s(SO₂).

2.3.11. [Zn(seabz)₂Cl₂] (11)

A solution of the ligand (0.1267 g, 0.5 mmol) and ZnCl₂ (0.0341 g, 0.25 mmol) in 15 mL of ethanol was stirred under reflux for 30 minutes. The reaction mixture was left to stand at room temperature. The obtained product was filtered and washed with 10 mL of cold ethanol. The reaction mixture was left to stand at room temperature. The obtained product was filtered and washed with 10 mL of cold ethanol. The compound was dissolved in acetonitrile. Crystals suitable for X-ray diffraction were obtained by slow evaporation of the solvent. Yield: 79%. Anal. Calculated for ZnC₂₂H₃₂N₆O₅S₂Cl₂: C, 39.98%; H, 4.88%; N, 12.72%; S, 9.70%. Experimental: C, 40.02%; H, 5.46%; N, 13.12%; S, 9.08%. IR (ν cm^-1 ): 3411 ν_as(NH₂), 3328 ν_s(NH₂), 1650 ν(C=C)+ρ(NH₂)+ν(C2-N3), 1562 ν(C=N)+ρ(NH₂)+ν(C2-N10), 1295 ν_as(SO₂), 1128 ν_s(SO₂).

2.3.12. [Zn(seabz)₂Br₂] (12)

A solution of the ligand (0.1267 g, 0.5 mmol) and ZnBr₂ (0.0563 g, 0.25 mmol) in 15 mL of ethanol was stirred under reflux for 30 minutes. The reaction mixture was left to stand at room temperature. The obtained product was filtered and washed with 10 mL of cold ethanol. Yield: 92%. Anal. Calculated for ZnC₂₂H₃₀N₆O₄S₂Br₂: C, 36.11%; H, 4.13%; N, 11.48%; S, 8.76%. Experimental: C, 35.87%; H, 3.86%; N, 11.90%; S, 7.62%. IR (ν cm^-1 ): 3400 and 3365 ν_as(NH₂), 3330 and 3306 ν_s(NH₂), 1640 ν(C=C)+ρ(NH₂)+ν(C2-N3), 1550 ν(C=N)+ρ(NH₂)+ν(C2-N10), 1290 ν_as(SO₂), 1124 ν_s(SO₂).

2.3.13. [Cd(sfabz)₂Cl₂] (13)

A solution of the ligand (0.1507 g, 0.5 mmol) and CdCl₂·2.5H₂O (0.0572 g, 0.25 mmol) in 15 mL of ethanol was stirred under reflux for 30 minutes. The reaction mixture was left to stand at room temperature. The obtained product was filtered and washed with 10 mL of cold ethanol. Yield: 90%. Anal. Calculated for CdC₃₀H₃₀N₆O₄S₂Cl₂: C, 45.84%; H, 3.85%; N, 10.69%; S, 8.16%. Experimental: C, 46.10%; H, 3.88%; N, 10.82%; S, 7.21%. IR (ν cm^-1 ): 3419 ν_as(NH₂), 3356 ν_s(NH₂), 1655 ν(C=C)+ρ(NH₂)+ν(C2-N3), 1568 ν(C=N)+ρ(NH₂)+ν(C2-N10), 1289 ν_as(SO₂), 1144 ν_s(SO₂).

2.3.14. [Cd(seabz)₂Cl₂] (14)

A solution of the ligand (0.1267 g, 0.5 mmol) and CdCl₂·2.5H₂O (0.0572 g, 0.25 mmol) in 15 mL of ethanol was stirred under reflux for 30 minutes. The reaction mixture was left to stand at room temperature. The obtained product was filtered and washed with 10 mL of cold ethanol. The compound was dissolved in acetonitrile. Crystals suitable for X-ray diffraction were obtained by slow evaporation of the solvent. Yield: 89%. Anal. Calculated for CdC₂₂H₃₁N₆O_4.5S₂Cl₂: C, 37.80%; H, 4.47%; N, 12.02%; S, 9.18%. Experimental: C, 37.96%; H, 4.77%; N, 11.73%; S, 8.13%. IR (ν cm^-1 ): 3400 ν_as(NH₂), 3332 ν_s(NH₂), 1648 ν(C=C)+ρ(NH₂)+ν(C2-N3), 1558 ν(C=N)+ρ(NH₂)+ν(C2-N10), 1295 ν_as(SO₂), 1127 ν_s(SO₂).

2.3.15. [Hg(sfabz)₂Cl₂] (15)

A solution of the ligand (0.1507 g, 0.5 mmol) and HgCl₂ (0.0679 g, 0.25 mmol) in 15 mL of ethanol was stirred under reflux for 30 minutes. The reaction mixture was left to stand at room temperature. The obtained product was filtered and washed with 10 mL of cold ethanol. Yield: 93%. Anal. Calculated for HgC₃₀H₃₀N₆O₄S₂Cl₂: C, 41.22%; H, 3.46%; N, 9.61%; S, 7.34%. Experimental: C, 41.10%; H, 3.35%; N, 9.91%; S, 6.40%. IR (ν cm^-1 ): 3387 ν_as(NH₂), 3311 ν_s(NH₂), 1645 ν(C=C)+ρ(NH₂)+ν(C2-N3), 1552 ν(C=N)+ρ(NH₂)+ν(C2-N10), 1292 ν_as(SO₂), 1141 ν_s(SO₂).

2.3.16. [Hg(seabz)₂Cl₂] (16)

A solution of the ligand (0.1267 g, 0.5 mmol) and HgCl₂ (0.0678 g, 0.25 mmol) in 15 mL of ethanol was stirred under reflux for 30 minutes. The reaction mixture was left to stand at room temperature. The obtained product was filtered and washed with 10 mL of cold ethanol. The compound was dissolved in acetonitrile. Crystals suitable for X-ray diffraction were obtained by slow evaporation of the solvent. Yield: 68%. Anal. Calculated for HgC₂₂H₃₀N₆O₄S₂Cl₂: C, 33.96%; H, 3.89%; N, 10.80%; S, 8.24%. Experimental: C, 33.95%; H, 4.28%; N, 10.73%; S, 7.08%. IR (ν cm^-1 ): 3403 ν_as(NH₂), 3331 ν_s(NH₂), 1647 ν(C=C)+ρ(NH₂)+ν(C2-N3), 1559 ν(C=N)+ρ(NH₂)+ν(C2-N10), 1295 ν_as(SO₂), 1125 ν_s(SO₂).

2.4. Physical Measurements

FT IR spectra were recorded with an FT-IR/FT-FIR Spectrum 400 spectrophotometer using a universal ATR accessory Perkin-Elmer (4000-400 cm^-1). The UV-Vis-NIR spectra (diffuse reflectance, 40000-5000 cm^-1) were recorded on a Cary-5000 (Varian) spectrophotometer, spectra recorded in solution for the copper(II) compounds was obtained from a 1x10^-3 M solution of the compounds in DMSO. Elemental analyses were carried in a Fisons EA 1180 analyzer. The ¹H and ¹³C NMR spectra were recorded with a Varian Unity Inova spectrometer with a frequency of 400 MHz for ¹H and 100 MHz for ¹³C, using DMSO-d⁶ as solvent, chemical shifts (δ) are reported in ppm referred to tetramethylsilane (TMS).

2.5. Solution studies

In order to study the stability of the copper(II) compounds in solution their spectra were recorded in DMSO (1x10^-3M) on a Cary-5000 (Varian) spectrophotometer for 24 h. For the zinc(II) compounds their 1H-NMR spectra was obtained using a Varian Unity Inova spectrometer with a frequency of 400 MHz, using DMSO-d6 as solvent, chemical shifts (δ) are reported in ppm referred to tetramethylsilane (TMS).

2.6. X-ray Crystallography

X-ray diffraction data for the ligands and the compounds 1, 2, 3 and 10 were obtained using standard procedures on an Oxford Diffraction Gemini “A” instrument with a CCD area detector using graphite-monochromated Mo(Kα) radiation for both ligands and the compounds 1, 2 and 3, and Cu(Kα) for compound 10. Data for both ligands and the compounds 1, 2, 3 and 10 were obtained at 130 K. Intensities were measured using φ + ω scans.

Diffraction data for compounds 5 and 9 were obtained on a Bruker D8 Venture\k-geometry diffractometer equipped with a CCD detector using graphite-monochromated Mo(Kα) radiation at 293.15 K for compound 5, and 150 K for compound 9. All structures were solved using direct methods, using the package SHELXS-2012 and refined with an anisotropic approach for non-hydrogen atoms using the SHELXL-2014/7 program. All hydrogen atoms that couldn’t be detected were positioned geometrically as riding on their parent atoms, with C–H = 0.93–0.99 A and Uiso(H) = −1.2Ueq(C) for aromatic and methylene groups [32-34]. All crystallographic data can be found in Tables S1-S3.

2.7. Cell growth inhibition

2.7.1. Cell culture

HeLa (cervix-uterine) MCF-7 (breast), HCT-15 (colorectal) and A549 (lung) human carcinoma cell lines and L929 mouse fibroblast, were acquired from ATCC (American Tissue Culture Collection) and maintained in incubation at 310 K and 5% CO2 with RPMI (GIBCO®, Invitrogen corporation) supplemented with 10% BFS (GIBCO®, Invitrogen corporation), 1% L-glutamine and 1% penicillin/streptomycin. Experiments were performed with cells within at least 5 passages from each other. All cells were split when around 80–95% confluence was reached using 0.25% trypsin/EDTA.

2.7.2. In vitro growth inhibition assay

After plating 2 × 10⁴ cells/well in 96-well microplate (Costar®) with 300 μL capacity and allowed to attach incubating at 310 K for 48 h, HeLa (cervix-uterine) MCF-7 (breast), HCT-15 (colorectal) and A549 (lung) human carcinoma cell lines and L929 mouse fibroblast were treated with sfabz, seabz and their Cu(II), and Zn(II) coordination compounds. The test metal compounds were made up in 5% DMSO and saline to give a 1 mM stock solution by initial dissolution in DMSO followed by dilution with saline. Sonication was sometimes used to facilitate complete dissolution. Serial dilutions were carried out to give final screening concentrations of ligands and the coordination compounds of 400, 200, and 20 μM (final concentration of DMSO of 0.5% (v/v)). Aliquots of 50 μL of these solutions were added to the wells (in triplicate) already containing 150 μL of media, so that the final concentrations were 100, 50 and 5 μM (final concentration of DMSO of 0.125% (v/v)). The cells were exposed to the complex for 24 h, which then was removed and the cells washed with washing media followed by the addition of 200 μL of fresh RPMI media. Then the cells were incubated for 72 h of recovery time. The remaining biomass was then estimated by the sulforhodamine B assay [35] (SRBassay). The three screening concentrations were used in an initial test of activity. The selected complexes were then tested for half maximal inhibitory concentration (IC50) value determination. The previously described assay was then repeated but using six different concentrations of complex instead, ranging from 0.1 to 100 μM. Each assay was done in triplicate. IC50 values were obtained from plots of % cell survival against log of the drug concentration.

3. Results and Discussion

3.1. Spectroscopic Characterization and magnetic susceptibility

Chlorido and bromido, Ni^II, Cu^II, Zn^II, Cd^II and Hg^II coordination compounds of 2-amino-1-(2-phenylsulfonyl)ethylbenzimidazole (sfabz) and 2-amino-1-(2-ethylsulfonyl)ethylbenzimidazole (seabz), were obtained. Their general structures were proposed based on spectroscopical data as well as elemental analyses. When single crystals were obtained the proposed structure was confirmed by the X-ray diffraction structure. The magnetic moments were also determined.

3.1.1. IR spectra.

The phenylsulfonated ligand (sfabz) presented the ν_as(NH₂) and the ν_s(NH₂) vibrations in 3434 and 3341 cm^-1 respectively, it also presented a band at 1664 cm^-1 that was assigned as the contributions from the ν(C=C), the ρ(NH₂) and the ν(C2-N3) vibrations; in a similar way, the band at 1552 cm^-1 was assigned as the sum of the contributions from the ν(C=N), the ρ(NH₂) and the ν(C2-N10) vibrations. Finally, the spectra presented bands at 1286 and 1140 cm^-1, which were assigned to the ν_as(SO₂) and ν_s(SO₂) vibrations respectively. Benzimidazolic bands were assigned as proposed by Sudha and coworkers [36]. Upon coordination of sfabz through the N3, the band centered in 1664 cm^-1 was shifted to lower energy (1656-1627 cm^-1) and the band at 1552 cm^-1 showed shifts to higher or lower energy depending on the metal (1571-1546 cm^-1). Sulfone bands ν_as(SO₂) and ν_s(SO₂) were shifted to higher energy (1294-1289 cm^-1 and 1144-1141 cm^-1 respectively). Only in compounds 5 and 10 the ν_s(SO₂) band was shifted to lower energy (1138-1136 cm^-1).

On the other hand, the ethylsulfonated ligand (seabz) presented bands at 3460 and 3373 cm^-1, assigned to the ν_as(NH₂) and ν_s(NH₂) vibrations respectively. Also, the bands at 1639 and 1549 cm^-1 were assigned in the same composed way as in the sfabz ligand. Finally, bands at 1281 and 1132 cm^-1, attributed to the ν_as(SO₂) and ν_s(SO₂) and both vibrations of the amino group (3414-3365 cm^-1 and 3332-3306 cm^-1, respectively). were shifted upon coordination.

3.1.2. Electronic spectroscopy and magnetic susceptibility.

For all nickel(II) and copper(II) compounds, the effective magnetic moment was determined and the UV-Vis-NIR was recorded. All the nickel(II) compounds were assigned to a tetrahedral geometry and because of that it was possible to calculate the ν₁ transition according to the graphical method described by Lever [37], the assigned transitions, as well as the effective magnetic moment are shown in Table 1.

The electronic transitions agree with the expected values for a nickel(II) (d⁸) tetrahedral compounds. The effective magnetic moment for these complexes is well within the range of 3.2-4.1 B.M. for nickel(II) showing this geometry [38]. Furthermore, the experimental results presented here are supported by their X-ray structure (vide infra).

Similarly, the diffuse reflectance electronic spectra for copper(II) compounds 5-8, show the d-d transition ca. 10000 cm^-1. Previously reported distorted tetrahedral copper(II) compounds have shown d-d transitions around these values [39, 40]. To further assess the stability of these compounds in solution, the spectra in a DMSO solution was obtained. Table 1 shows the values of the d-d electronic transitions at similar values to those in solid state, suggesting the conservation of the ligands in solution (Figure S1, S2.). Compounds 5-8 depict a µ_eff within the expected range of 1.8-2.2 B.M. [38].

3.1.3. NMR studies.

¹H, ¹³C and HSQC spectra were obtained for both ligands and compounds 9, 11, 13, 14, 15 and 16. The ¹H-NMR and ¹³C-NMR signals were assigned, according to Figure 1a, and are in agreement with the HSQC spectra (Tables S4 and S5). Similarly, Figure 1.b, depicts the assignation for the ligand seafz, corroborated through HSQC (Tables S6 and S7).

For both series of coordination compounds, signals of the ligands, in both ¹H and ¹³C spectra, were shifted upon coordination (Δδ=ligand-complex, Δδ > 0.05 ppm for ¹H and Δδ>0.1 ppm for ¹³C). Table 2 and Table 3 resume the effect of coordination in the ¹H and ¹³C signals of the ligands except for their ethyl and phenyl substituents.

According to the ¹H spectra, it is possible to differentiate the behavior of the three groups of protons: the benzimidazolic protons, the amino group, and the aliphatic chain protons. For the benzimidazolic protons of both series of compounds, the deshielding effect is more pronounced in H6 and H7 and smaller in H5. Comparing the effect of different metals, the deshielding at H5-H7 follows the trend Hg=Zn<Cd, while at H4 all metals cause almost similar displacements in the sfabz series and Cd causes the bigger effect in the seabz series.

In both ligands’ series the amino protons get deshielded upon coordination to the metal ion. This effect is more predominant with Zn, and lower with Cd. Similarly, the aliphatic chain shows a significant displacement with same tendency seen in the amino group, namely, Zn>Hg>Cd.

Comparing the effect of the metal in the ¹³C spectra, the shielding and deshielding are more pronounced in the Zn compounds and less prominent in the Hg and Cd compounds. The ¹³C chemical shifts of the aliphatic chain showed a slight deshielding at C11, and a shielding effect on the C12, an effect that can be attributed to the sulfone group in the chain. The terminal chain is also affected by the sulfone, were the aliphatic ethyl chain shows no significant changes in its chemical shifts for both ¹H and ¹³C data, while the phenyl group shows significant changes at the C16 and C17 positions (Δδ C16_máx= 0.5 ppm, Δδ C17_máx= 0.3 ppm).

3.2. X-ray structures of the ligands and their coordination compounds

3.2.1. Crystal structure of the 2-aminobenzimidazolic ligands.

Despite the difference in the terminal substituent, both the seabz and sfabz ligands crystallize in a P 2₁/n space group, within a monoclinic crystal system. However, these two ligands show different intramolecular interactions. For the seabz ligand, one of the oxygen atoms in the sulfone group is orientated towards the benzimidazolic ring, due to a lone pair···π_bz interaction at 3.775 Å (Figure 2a) Instead, the sfabz ligand depicts the sulfone group away from the benzimidazolic ring, thus generating a weak H···π_phe at 3.791 Å, between a benzimidazolic proton and the terminal phenyl ring (Figure 2b).

Another difference found in the crystal structures of both ligands are the type of intermolecular interactions, mainly the hydrogen bonding involving the N3 of the benzimidazolic ring. The sfabz ligand depicts two strong hydrogen bonds with the amino group acting as the donor and the N3, the acceptor. This interaction is at 2.041 Å with an angle of 174.58º. Two neighboring molecules show one of these hydrogen bonds each, forming a dimer as depicted in Figure 3a. On the other hand, the seabz ligand also shows a hydrogen bond with the N3 as the acceptor. However, this is a weak hydrogen bond given the donor is a –CH₂– group at 2.603 Å and 157.15 º. Alternatively, what seems to be a major stabilizing interaction, between four seabz molecules, is a series of hydrogen bonds with the amino group acting as the donor, with the S=O accepting two protons form different molecules. Their angles and distances (2.088 Å, 2.127 Å and 171.03º, 164.33º) indicate that these interactions are moderate, Figure 3b depicts these hydrogen bonds.

3.2.2. Crystal structure of the coordination compounds of seabz and sfabz.

Herein there were synthesized coordination compounds with both alkylsulfonated benzimidazole ligands to be able to compare the effect of the substituents in the interactions found in the crystal structures of their coordination compounds. As mentioned above, all compounds show the formula [ML₂X₂] (M²⁺ = Ni, Cu, Zn, Cd, Hg; L= seabz, sfabz; X= Cl^-, Br^-). For some of the seabz complexes, adequate crystals for X-ray diffraction were not obtain. This could be due to the fact that the ethyl group will present far less non-covalent interactions than the phenyl ring does (vide infra). Because of this, for compounds 11, 14, and 16, only basic structural features and connectivity are discussed.

Compounds 1, 2 (L = sfabz) and 3 (L = seabz) depict a Ni center with either a Cl^- or Br^- anion. Compounds 1 and 2 both are obtained in a P-1 space group, within a triclinic crystal system, regardless of the acetone molecule from the solvent in compound 2. Alternatively, compound 3 is obtained in a P 2₁/c space group and a monoclinic system. The bond lengths and angles around the metal ion for these three compounds are shown in Table 4. From the result shown in this table, it is noticeable that the terminal substituent in the alkylsulfonated chain does not have a major effect in the bonds around the metal ion.

However, the ligands’ terminal group does affect the interactions that each of this compounds present. Even between the two sfabz Ni coordination compounds, different intramolecular contacts are found. The crystal structure for compound 1 (Fig 4a) depicts both ligands in different conformations, one being extended and the other one with both aromatic rings facing each other. In doing so, a H···π_bz contact can be found at 2.954 Å. In contrast to the H···π_phe interaction observed in the free ligand (vide supra), in the nickel compound this interaction is between two different ligands, where the aromatic rings, acting as donor and acceptor, are reversed giving a H···π_bz contact. Compound 2, also a sfabz derivative, shows a similar interaction, between benzimidazolic moieties. The H··· π_bz contact (3.577 Å) is shown in Figure 4b. Finally, compound 3, rather than depicting a H···π contact, it depicts a lone pair···π intramolecular interaction between the sulfone group and the benzimidazolic ring (Fig 4c). This interaction is at 3.416 Å and with an angle centroid-N-O of 85.51º, indicative of a strong non-covalent interaction that stabilizes the crystal structure. This emphasizes the importance of the terminal substituent in this type of ligands, when the substituent is a phenyl ring, primarily depicts H···π contacts. Whereas, when it is an ethyl group, this interaction is no longer present giving place to a different interaction, namely, lp···π.

Regarding the intermolecular interactions of these nickel(II) complexes, both sfabz compounds show very different contacts in their crystal structure. Compound 2 depicts a displaced π···π stacking interaction between two benzimidazolic rings at 3.673 Å (centroid-centroid) (Figure 5b). Alternatively, compound 1 shows hydrogen bonds with the amino and a –CH₂– group acting as the donor, and an oxygen of the sulfone group as the acceptor (Figure 5a). These differences in intermolecular interactions between these two compounds can be attributed, mainly, to the fact that compound 2 crystalizes with an acetone molecule. Doing a similar analysis with compound 3 with seabz, what seems to be the most important contact is a lone pair···π interaction between S=O and the benzimidazolic ring at 3.347Å and 89.59º, directed towards the center of the imidazolic ring of this molecule (Figure 5c). Interestingly, the other oxygen of the same sulfone group is the one showing the intramolecular lone pair···π interaction (vide supra). This is relevant because in our previous work with alkylsulfonated ligands [24-26], when one oxygen of the sulfone group is depicting such interactions, the other oxygen does not show any other contact, not even weak hydrogen bonds.

As mentioned above, adequate crystals for X-ray diffraction were easier to obtain with the sfabz ligand. However, the copper(II) complex (compound 5) shows great disorder in one of the sfabz ligands. Regardless, it is still possible to see its connectivity and geometrical features. Compound 5 crystalizes in a P-1 space group, within a triclinic system. Although this compound still stabilizes a tetrahedral geometry, it is more distorted that the nickel and zinc coordination compounds, as its noticeable for its larger N-Cu-N’ angle of 135.54º (Figure 6). No relevant intramolecular contacts could be assigned in this structure, due to the structural disorder.

There were obtained two compounds with Zn(II) and sfabz that crystallized in a triclinic system and a P-1 space group. Angles and bond distance around the metal ion are summarize in Table 5. Although compound 10 depicts an acetone molecule in the crystal structure, this doesn´t seem to affect the intramolecular interactions, as both Zn(II) sfabz complexes depict a H···π contact with one benzimidazolic ring acting as the donor, and the other one as de acceptor, with a distance of 3.404 Å for compound 9 and 3.630 Å for compound 10. In contrast to the NiX₂ derivatives discussed above, where the crystal structure was either stabilized through either hydrogen bonds with the sulfone or through π stacking between the benzimidazolic rings, compounds 9 and 10, depict both of these interactions at the same time with one neighboring molecule. Figure 7 shows the intermolecular interactions for compound 9, as an example with the relevant angle and distances. The corresponding values for compound 10 can be found in parenthesis in the same Figure.

The connectivity and general structure features for compounds 11, 14 and 16 (Zn, Cd and Hg, respectively) are depicted in Figure 8. As the obtained crystals were not adequate to properly obtain their X-ray structure, only general aspects of the compounds can be assessed. Namely, all three compounds depict two alkylsulfone ligands and two halogens, yielding tetracoordinated compounds, a distorted tetrahedral geometry, as seen for all the crystal structures depicted herein. It is noteworthy that, for these three compounds, an acetonitrile molecule is present in the crystal structure. This highlights the importance of the solvent being used for crystallization as, even though some crystal structures depict acetone molecules, this does not affect the quality of the X-ray diffractions obtained. Whereas, using acetontitrile as a solvent introduces disorder and lower-quality crystals are obtained.

3.3. Stability in solution and antiproliferative activity.

The stability in solution of the copper(II) compounds was assessed as described in the methods section of this article. In all cases, the same electronic transitions described above were observed without significant change after a day in solution, indicating that no changes in the geometry or coordination were observed. As the Cu(II) compounds, zinc(II) compounds showed no significant variation in their NMR spectra after the same time period. Both series of compounds were suitable for the biological activity studies.

To determine their antiproliferative activity, a cell-viability assay was carried out with the aforementioned copper(II) and zinc(II) compounds against HeLa (cervix carcinoma), HCT-15 (colorectal adenocarcinoma), MCF-7 (breast adenocarcinoma), A549 (lung adenocarcinoma) and L929 (healthy connective mice tissue).

The IC₅₀ of the compounds were determined and it is presented below (Table 6). As observed in the table, IC₅₀ for both ligands were the highest of all compounds for every cancer cell line, indicating that the ligands alone were not active. In fact, most of the coordination compounds were significantly less active than cis-platinum in most cancer cell lines. Only copper(II) sfabz coordination compounds were active enough to be compared with cis-platinum when tested in the HeLa cell line, being [Cu(sfabz)₂Br₂] (6) slightly better than the Pt reference compound.

Although only few compounds showed IC₅₀ comparable to cisplatin, there is a visible pattern for the antiproliferative activity of the compounds. Despite the substituent of the benzimidazolic ligand, copper(II) compounds showed higher activity than their zinc(II) homologues. Aditionally, bromo containing compounds were typically more active than their chloro counterparts. Comparing both series of complexes, seabz compounds were substantially less active than the sfabz compounds.

Finally, all compounds were generally less active than cis-platinum towards healthy mice tissue. Giving the selectivity of copper(II) sfabz compounds on cancer cell lines, they are worth further investigation.