Mechanistic Insights into the Antibacterial Activity of Self-Assembled Nickel(II)–Triethanolamine–Cinnamate Complexes

Abstract

Herein, we report the room-temperature isolation of two new structurally similar mononuclear ionic nickel­(II) complexes, [Ni­(H₃ tea)₂]­(2,3-DMCN)₂ (1) and [Ni­(H₃ tea)₂]­(2,4-DMCN)₂(H₂O) (2) (where H₃ tea = 2,2′,2″-nitrilotri­(ethan-1-ol), 2,3-DMCN = 2,3-dimethoxycinnamate, and 2,4-DMCN = 2,4-dimethoxycinnamate). Both the crystalline complexes were structurally characterized by spectroscopic (FT-IR, UV–vis, and XPS) and single-crystal X-ray diffraction (SCXRD) methods. SCXRD unambiguously delineated the Ni­(II) center octahedrally coordinated by strong chelating ligand H₃ tea, yielding complex cations with NiN₂O₄ chromophore, which are counterbalanced by cinnamate anions in both cases. Crystal packing supported by Hirshfeld surface (HS) analysis revealed the role of various hydrogen-bonding interactions in crystal lattice stabilization. The outcomes of theoretical studies using DFT demonstrated notable concordance with the experimentally observed data. In vitro antibacterial studies indicated that both complexes exhibited remarkable selectivity for Gram-negative strains, specifically P. aeruginosa and K. pneumoniae, corroborated with in silico molecular docking results. Mechanistically, the [Ni­(H₃ tea)₂]²⁺ core acts as an electronic shield, which, alongside lipophilic dimethoxycinnamate anions, promotes membrane translocation and subsequent structural disruption. This physical intrusion is complemented by a redox-active mechanism, viz., a high-lying HOMO facilitates interference with the bacterial respiratory chain, triggering intracellular reactive oxygen species generation and lethal oxidative damage to DNA and proteins. Detailed structural analysis of this self-assembled molecular frameworkutilizing both experimental and theoretical toolsprovides deeper mechanistic understanding of the antibacterial effect of Ni­(II)-mixed-ligand complexes, advancing future metallodrug discovery.

1. Introduction

Research interest in the coordination chemistry of divalent transition-metal ions has remained a central pursuit in inorganic chemistry for many decades. − The persistent interest in this field is due to the ability to tune the geometrical, structural, electronic, and optical properties of such coordination systems, delineating potential candidacy for a broad range of applications including hydrogen storage, heterogeneous catalysis, bioinorganic processes, ion exchangers, electroplating, additives in building materials, semiconductor materials, etc. ,− Among the transition-metal complexes, nickel­(II) complexes are prominent hydrogenation catalysts owing to partially filled d-orbitals facilitating the interaction, activation, and hydride formation and thereby impacting the kinetics of hydrogen release. In a biological context, variable oxidation states from Ni­(I)/Ni­(II)/Ni­(III) and the ability to form strong chelates with nitrogen and oxygen donor ligands leading to thermodynamic stability provide further boost. The additional thermodynamic stability of chelates of Ni­(II) facilitates their potential uses in ion exchangers, semiconductor, and electroplating applications. , For potential desired application, structural design, choice of ligands, reaction conditions, and coordination behavior play a significant role.

Despite the significant progress in coordination chemistry, the deliberate design, synthesis of such systems with desirable coordination features which involve sewing of molecules via the interplay of various non-covalent interactions (O–H···O, C–H···O, π···π, etc.) besides control over the experimental conditions (pH, solvent, temperature) for successful isolation is still a challenging aspect in supramolecular chemistry. − Among the divalent metal cations, nickel­(II) complexes have emerged as a class of utmost interest because of their variable coordination geometries, flexible ligand-binding modes, kinetic lability, and their potential roles in catalysis and biology. − The ability of Ni­(II) complexes to adopt different coordination geometries ranging from octahedral to square planar and distorted environments, as governed by ligand structures, steric constraints and nature of coordinating donor atoms, provides an ideal platform for investigating subtle structure–property relationships.In the structural design, and to assess various structural–property relationships, judicious selection of ligands is cardinal besides above-mentioned aspects. Within this context, ligands derived from ethanolamine (amino alcohols) such as triethanolamine (H ₃ tea) have long been recognized for their remarkable chelating ability and structural adaptability besides function as bridging ligands. − The coexistence of N–H and O–H donor sites enables robust coordination to divalent metal centers; besides, pendent groups (O–H and N–H) also facilitate extended hydrogen-bonding interactions networks, leading to the generation of intriguing supramolecular motifs. In addition, aryl carboxylic acids, which displayed various coordination modes toward divalent metal cations, such as chelating, bridging, ionic, etc., led to generation of metal complexes of diverse structural topologies. ,

In this pursuit, cinnamic acid (CA) was chosen as the carboxylate ligand owing to its natural abundance in various plants including cinnamon, Shea butter, balsams, etc. CA and its derivatives are often used in pharmaceutical industry as flavoring agents. , In addition, CA and its derivatives including coordination complexes possess inherent antibacterial, antifungal, antioxidant, and antimalarial activities. , Interestingly, cinnamate derivatives, herein, dimethoxy substituted, which are electronically active (donor effect of the methoxy group) led to a more negative cloud on carboxylate oxygen atoms. Apart from this, substitution patterns on 2,3- and 2,4-positions of cinnamate profoundly impact the electronic and steric features of the isolated coordination complex. In mutual association, using cinnamate and H ₃ tea commonly promotes spontaneous self-assembly under ambient reaction conditions, yielding crystalline complexes stabilized by the interplay of various weak interactions among the interacting groups. There are only a few structural reports on nickel­(II) cinnamate complexes available in the literature with various nitrogen donor ligands such as pyridine, bipyridine, ethylenediamine, salen derivatives, etc. , But interestingly, there is no structural report available where electron-rich cinnamate (2,3-/2,4-methoxycinnamate) in the presence of H ₃ tea was employed and the role of 2,3-/2,4-disubstitution has been explored with the nickel­(II) center. Moreover, we envisaged that the H ₃ tea ligand might show similar structural features as copper­(II) carboxylate complexes reported by our group. , Distinctively, Ni­(II) complexes demonstrated less Jahn–Teller distortion than Cu­(II) complexes where distortion directed weakening of the Cu–O/Cu–N bond to form a bridging coordination mode. On the other hand, strong chelation of H₃ tea toward the Ni­(II) center is highly anticipated owing to the absence of Jahn–Teller distortion and favorable ligand field stabilization energy (LFSE), which might result in interesting structural topologies.

In this endeavor of continuous interest in carboxylate chemistry, the present work reports the detailed structural and mechanistic antimicrobial evaluation of two newly isolated ionic mononuclear dimethoxycinnamate Ni­(II) crystalline complexes: [Ni­(H₃ t ea)₂]­(2,3-DMCN)₂ (1) and [Ni­(H₃ t ea)₂]­(2,4-DMCN)₂(H₂O) (2) at room temperature. The harsh synthesis routes of similar complexes from the literature are enumerated in Table S1. The comprehensive physicochemical analysis, spectroscopic, and single-crystal structural analyses revealed that Ni­(II) center adopt an octahedral coordination stablized by strong chelation of H ₃ tea ligands, while dimethoxycinnamate acts as a counteranion, balancing charge and further stablizing the crystal lattice via mutual interplay of various non-covalent interactions (C–H···O, O–H···O, etc.) in both cases. To probe the electronic structures and global reactivity descriptors, DFT calculations were performed in addition to assessing thermodynamic parameters and bonding vibrations. In addition, antibacterial assays supported by in silico molecular docking demonstrated potent activity, particularly against Gram-negative bacterial strains, i.e., P. aeruginosa (1UIZ) and K. pneumoniae (5XUN). There is less efficacy reported against Gram +ve strains, i.e., S. aureus (5TW8) and B. cereus (1YQH), but comparable to the standard antimicrobial agent, i.e., chloramphenicol.

Driven by the need to develop structurally tunable Ni­(II) carboxylate complexes with enhanced functional properties, this research employs a room-temperature self-assembly approach to bridge existing gaps between molecular architecture and physicochemical performance. The central novelty lies in correlating the positional isomerism of cinnamate derivatives (2,3- vs 2,4-substitution) with distinct supramolecular architectures, governed by a delicate interplay of intermolecular interactions elucidated through single crystal X-ray diffraction and Hirshfeld surface analysis. By seamlessly integrating experimental observations with theoretical modeling, this study establishes a comprehensive structure–activity relationship (SAR) that clarifies the mechanistic basis of the observed antibacterial potency towards Gram -ve bacterial strains. Ultimately, this work advances the rational design of bioactive coordination systems by bridging molecular structure, electronic properties, and biological function within a unified framework, providing a roadmap for the development of structure-guided metallotherapeutics.

2. Experimental Section

2.1. Chemicals, Reagents, and Instruments

All the chemicals used in the experimental part were of analytical grade. NiCl₂.6H₂O (CAS RN: 7791-20-0) was purchased from Sigma-Aldrich Ltd. and used with precaution owing to its hygroscopic nature. Trans-2,3-dimethoxycinnamic acid (CAS RN: 7345-82-6) and trans-2,4-dimethoxycinnamic acid (CAS RN: 16909-09-4) were also procured from Sigma-Aldrich Ltd. Triethanolamine (H₃ tea) was procured from TCI Ltd. India (CAS RN: 102-71-6). Organic solvents such as methanol (CAS RN: 67-56-1), chloroform (CAS RN: 67-66-1), and acetone (CAS RN: 67-64-1) were procured from CDH (Central Drug House) Ltd., India. Double distilled water was used for all solution preparations. Melting points were recorded in a capillary tube by using melting point apparatus model MP-D Mitamura Riken Kogyo (Japan). FT-IR spectra for both complexes 1 and 2 were recorded on a Nicolet6700 FT-IR spectrophotometer in the range of 4000–400 cm^–1. Electronic transitions or UV–visible spectra were measured by employing different solvents on a Thermo-Fisher spectrophotometer. XPS spectra delineated surface elemental and chemical state information were recorded on the Thermo-Scientific NEXSA Surface Analysis System.

2.2. Syntheses

2.2.1. Synthesis of Complex 1

In a 100 mL glass beaker, sodium salt of 2,3-dimethoxycinnamic acid was prepared by dissolving 2,3-dimethoxycinnamic acid 2.0 g (9.6 mmol) and sodium hydroxide 0.38 g (9.6 mmol) in minimum amount of water. In a separate beaker, nickel­(II)­chloride hexahydrate 1.140 g (4.8 mmol) was dissolved in 10 mL of deionized water. Two of the above-mentioned solutions were mixed with stirring; on mixing, immediate precipitation (green-colored precipitates) was observed. The precipitates were filtered with the help of fine filter paper and washed with water to remove any excess acid and dried at room temperature. To isolate the crystalline final product, 0.5 g of green-colored precipitates of [Ni­(2,3-cinnamate)₂·(H₂O)₂] were suspended in 10 mL methanol–water (4:1, v/v) mixture and H₃ tea (triethanolamine) was added dropwise until clear greenish solution appeared. The green-colored crystals of complex 1 start appearing after 5 days from solution at the bottom of the beaker. The green crystals from the bottom of the solution were isolated carefully and air-dried and stored in an airtight glass vial for further use. The overall yield of complex 1 was 68% (340 mg), decomposed above 170 °C. Anal. Calcd for C₃₄H₅₂NiN₂O₁₄ (mol wt 771) Cal. C, 52.91; H, 6.74; N, 3.63; Ni, 7.61%; Found: C, 52.71; H, 6.92; N, 3.48; Ni, 7.72%. FT-IR spectrum (cm^–1): 3347 (s), 2938­(m), 2352(s), 1628­(w), 1461(s), 1372(s), 1269(s), 1057(s), 877­(m), 742(s), 587(s), 520(s). UV–visible (MeOH:H₂O): 645 nm (d–d transition). XPS spectra: Ni 2p_3/2 at 855.86 eV and Ni 2p_1/2 at 873.14 eV, O 1s at 532.42 eV and N 1s at 399.93 eV.

2.2.2. Synthesis of Complex 2

A similar procedure as in complex 1 was used to isolate crystalline complex 2 by using 2,4-dimethoxycinnamic acid under similar reaction conditions. The green crystals of complex 2 were observed at the bottom of the reaction vessel, isolated after 6 days. The crystals of complex 2 were then removed and air-dried (yield = 61%, 310 mg). The newly synthesized complex was soluble in water, methanol, and other organic solvents. Complex 2 decomposed above 165 °C. Anal. Calcd for C₃₄H₅₄NiN₂O₁₅ (mol wt 789.5) Cal. C, 52.91; H, 6.74; N, 3.63; Ni, 7.61%; Found: C, 51.95; H, 6.72; N, 3.59; Ni, 7.50%. FT-IR spectrum (cm^–1): 3615­(w), 3335­(br, s), 2809­(w), 1606­(m), 1517(s), 1372(s), 1271(s), 1022­(m), 885­(m), 806(s), 710­(w), 592­(m). UV–visible (MeOH/H₂O): 637 nm (d–d transition). XPS spectra: Ni 2p_3/2 at 855.76 eV and Ni 2p_1/2 at 873.50 eV, O 1s at 532.60 eV and N 1s at 400.04 eV.

2.3. Single-Crystal X-ray Diffraction

XSingle-Crystal X-ray diffraction data to 0.70 Å atomic resolution of green block crystals of complexes 1 and 2 were collected on a Bruker X8 APEXII KAPPA CCD diffractometer at room temperature (296(2) K), using graphite monochromatized Mo Kα radiation (λ = 0.71073 Å). The APEX2 program suite was used to complete data processing including data integration, reduction, multiscan absorption correction, scaling/merging, etc. The intrinsic phasing method implemented in SHELXTL XT was used to solve the crystal structures of 1 and 2, besides refinement using full-matrix least-squares on F ² with SHELXTL XLMP. Anisotropic thermal parameters were used for all non-H atoms. H atoms were positioned and treated using a riding model. H atoms of hydroxyls and water molecule were initially located from difference Fourier electron density maps and were freely refined and constrained with “AFIX 3”, respectively. The pertinent details of data collection and structure refinement parameters of both complexes are given in Table .

1. Summary of Crystal Data and Refinement Details of Complexes 1 and 2 .

identification code	complex 1	complex 2
abbreviated formula	[Ni(H3 tea)2](2,3-DMCN)2	[Ni(H3 tea)2](2,4-DMCN)2(H2O)
empirical formula	C34H52NiN2O14	C34H54NiN2O15
formula weight	771.48	789.50
temperature/K	296(2)	296(2)
crystal system, space group	triclinic, P1̅ (no. 2)	triclinic, P1̅ (no. 2)
a/Å	8.8819(5)	10.8463(5)
b/Å	9.7461(5)	11.0113(5)
c/Å	10.6361(6)	16.0076(7)
α/°	83.120(2)	93.507(2)
β/°	86.598(2)	93.329(2)
γ/°	86.119(2)	96.687(2)
volume/Å3	910.76(9)	1891.15(15)
Z	1	2
ρcalc/g cm–3	1.407	1.386
μ/mm–1	0.603	0.584
F(000)	410	840
crystal size/mm3	0.22 × 0.36 × 0.50	0.12 × 0.20 × 0.24
radiation/Å	Mo Kα (λ = 0.71073)	Mo Kα (λ = 0.71073)
2Θ range for data collection/°	4.218 to 61.634	3.790 to 61.136
resolution/Å	0.70	0.70
index ranges	–12 ≤ h ≤ 12, –13 ≤ k ≤ 14, –15 ≤ l ≤ 15	–15 ≤ h ≤ 15, –15 ≤ k ≤ 15, –22 ≤ l ≤ 22
reflections collected/unique/> 2σ	31087/5641/3182	69845/11584/6964
independent reflections	5641 [R int = 0.0785, R sig = 0.0737]	11584 [R int = 0.0847, R sig = 0.0748]
data/restraints/parameters	3182/0/252	6964/0/497
R 1 , wR 2 (I > 2σ(I))	0.0559, 0.1143	0.0486, 0.0871
R 1, wR 2 (all data), S	0.1253, 0.1520, 1.043	0.1034, 0.1067, 1.004
largest diff. peak/hole/e Å–3	0.56/–0.37	0.36/-0.31
CCDC no.	2490675	2490676

^a,b R = Σ||F _o| – |F _c||/Σ|F _o|; wR = Σ­{w(F _o ² – F _c ²)²/Σw(F _o ²)²}^1/2.

2.4. Quantum Mechanical Calculations

Quantum chemical calculations and the molecular geometry optimization of complexes 1 and 2 were performed with the help of Gaussian09. , The outcomes of the calculations were subsequently visualized and analyzed as molecular electrostatic potential maps (MEP) and Frontier molecular orbitals (FMOs) by GaussView 6.0. Density functional theory (DFT) calculations were performed using the Becke-3-parameter-Lee–Yang–Parr (B3LYP) , functional with a 6-311++G­(d,p) basis set for C, N, and O atoms and the LANL2DZ basis set for Ni atom. The resulting molecular electrostatic potential surface maps enable the assessment of the electronic distribution that provides easy estimation of the chemical reactivity of both complexes. Furthermore, Mulliken population analysis and total energy calculations were performed on the optimized structures using the same B3LYP/6-311++G­(d,p) level of theory.

2.5. Hirshfeld Surface and Energy Framework Analyses

The HS analysis, 2D fingerprint plots, and 3D energy frameworks calculations were carried out with the help of the Crystal Explorer 17 program by using SCXRD data (CIF files) of complexes 1 and 2. Molecular Hirshfeld surface (HS) analysis was performed, from an effort to define the space occupied by a molecule in a crystal in order to partition the electron density of a crystal into molecular fragments of electron densities. 2D fingerprint plots were generated to provide a summary of the intermolecular interactions that are present in the crystal structure. To generate the 2D fingerprint plots, two distances d_e and d_i were measured; d_e is the distance from HS to the nearest atom located externally to the surface and d_i is the distance from HS to the nearest atom located internally to the surface. Energy framework analysis provides distinctive quantitative understanding of intermolecular interaction energies and supramolecular architecture within the crystal structure. A crystal of surrounding molecules within a 3.8 Å radius was generated around a central molecule of both the complexes. The energy framework was constructed by using the B3LYP/6-31G­(d,p) method and the electron densities with applied scale factors of k _ele = 1.057, k _pol = 0.740, k _disp = 0.817, and k _rep = 0.618 for the electrostatic, polarization, dispersion, and repulsion energy components, respectively. , Void analysis is important for evaluating the mechanical stabilty of the packed crystal. Void analysis was performed based on the cumulative electron densities of the spherically symmetric atoms present in the asymmetric units.

2.6. In Vitro Antibacterial Study

To investigate the biological potential, the synthesized complexes were examined for their antibacterial activity against a range of bacterial strains. In the present study, two Gram +ve bacterial strains, namely, S. aureus and B. cereus, and two Gram −ve bacterial strains, namely, P. aeruginosa and K. pneumoniae, were employed for antibacterial evaluation. In addition, in vitro antibacterial activities of complexes 1 and 2 were compared with the reference standard antibacterial agent, chloramphenicol. The minimum inhibitory concentrations (MICs) of complexes 1 and 2 were determined using a standard 96-well microdilution technique.

2.7. In Silico Antibacterial Study by Molecular Docking Simulations

To evaluate the inhibitory potential of the complexes at the molecular level, in silico antibacterial analysis was conducted through molecular docking calculations. This analysis identifies the most favorable binding pose of the ligand within the target protein, establishing the structural basis for a stable inhibitory complex. We obtained the atomic coordinates of the target proteins from the RCSB Protein Data Bank (PDB) to conduct the docking simulations using AutoDock Vina. All parameters required for protein–ligand interaction analysis, including grid calculations, were generated using AutoDock tools. The aim of the present work was to investigate the protein target binding of complexes 1 and 2 embracing various organic ligands. The optimal binding site based on ligand–protein interaction energy was evaluated, and the conformation exhibiting the lowest binding energy was chosen. To determine the optimum binding conformation between the ligand and protein, the Lamarckian genetic algorithm (LGA) was used. The grid boxes that represent the coordinates corresponding to the binding sites of the target proteins for the bacterial enzymes are as follows: 1UIZ: 25.287, 26.493, 120.740; 5XUN: 23.857, −21.119, 37.921; 5TW8: 28.039, −53.755, 16.079; and 1YQH: 27.612, 49.354, 47.363 for the respective x, y, and z dimensions. The predicted active sites of the target proteins are provided in Figure S1. By using BIOVIA Discovery Studio, we can visualize intermolecular interactions of eight different protein–ligand complexes . ,

3. Results and Discussion

3.1. Synthesis

Crystals of complexes 1 and 2 were isolated by using appropriate precursors at room temperature in a methanol–water solvent mixture (4:1, v/v). The schematic representation of the synthesis of nickel­(II) complexes 1 and 2 is given in Scheme S1 (Supporting Information). The complexes might have two types of coordination possibilities: either H₃ tea acts as a bridge among the two Ni­(II) centers by undergoing one deprotonation to form H₂ tea or H₃ tea ligand acts as a chelating ligand and thus displaces the cinnamate group as a counteranion. Furthermore, on the basis of solubility behavior of complexes 1 and 2 (soluble in water, methanol, and ethanol), it suggested ionic formulation for both cases. Based on the known coordintaion chemistry of nickel­(II) (d⁸) as reported in the existing literature, H₃ tea might behave as a strong chelating ligand toward the nickel­(II) center and thus lead to the formation of ionic complexes. Owing to the kinetic lability of the nickel­(II) center, the strong coordination affinity of the nickel­(II) center toward H₃ tea, and the second-sphere coordination of flexible cinnamate ligands, this is an arduous task. Strongly interacting functional groups endowed on ligands (C–H, O–H) facilitate the intensive hydrogen-bonding interactions. The exact structures, coordination geometries, and ionic nature of complexes 1 and 2 have been unambiguously confirmed with the help of single-crystal X-ray structure determination of both complexes.

3.2. Single-Crystal Structure Determination

3.2.1. Molecular Descriptions of Complexes 1 and 2

Both complexes 1 and 2 crystallized in the triclinic space group P1̅ (no. 2) with the unit cell volume of 2 doubled that of 1, indicating that the unit cells, respectively, contained two and one formula units (Table ). In case of complex 1, Ni­(II) is sitting on the special position at the edge center of triclinic unit cell, overall delineated complex cation [Ni­(H₃ tea)₂]²⁺ that are counterbalanced by two 2,3-DMCN anions. The complex cation exhibited a distorted octahedral geometry, in which four hydroxyl groups (OH) from H₃ tea form a square plane (Ni–O1 = 2.069 Å, Ni–O2 = 2.078); besides, two N atoms of inversion-related H₃ tea molecules occupy apical positions at a distance (Ni–N1) of 2.098 Å. Moreover, one hydroxyl ethyl arm of H₃ tea remains free susceptible to hydrogen bonding with counter 2,3-DMCN anions (Figure ).

1.

Molecular structures of complexes 1 (a) and 2 (b); non-H atoms shown as thermal ellipsoids at 50% probability level and hydrogens shown as ball–stick models. Atom numbering for complex 1 is only given to the asymmetric unit.

In case of complex 2, the asymmetric unit also showed a similar complex molecular unit as in complex 1 in which the NiN₂O₄ chromophore constitutes a complex cation that is charge-balanced by two 2,4-DMCN anions. The carboxylate groups of 2,3-DMCN and 2,4-DMCN counteranions in both complexeslie in the aromatic plane owing to less steric repulsion. To minimize steric crowding and proper packing, the characteristic difference in methoxy-group orientation in 2,3-DMCN of complex 1 is more pronounced than 2,4-DMCN of complex 2. The characteristic bond distances and bond angles in complexes 1 and 2 are given in Table S2.

3.2.2. Packing Analysis

Crystal lattices of both complexes were stabilized via the mutual interplay of C–H···O and O–H···O hydrogen-bonding interactions. The magnitude of overall hydrogen-bonding interaction is larger in complex 2 than in 1. In complex 1, the C1–H1B···O3­(−x + 1, −y + 1, −z) interaction with the bond distance and bond angle of 2.58 Å and 167.0° played a key role in making a layered pattern among the complex cations and forming R₂ ² (12) rings, extending along one direction (Figure ). Apart from the complex cation, anions showed stabilization through the interplay of O–H···O H-bond R⁴ ₄(18) rings, forming anionic layers. Interestingly, the cooperation of O–H···O H-bonds, i.e., O1–H1···O7 = 1.65(4) Å and O2–H2···O6 = 1.65(4) Å in the horizonal direction and O3–H3···O6 = 1.95(4) Å in the vertical direction, yielded the ladder-like H-bond network (Figure S2).

2.

R₂ ² (12) ring network via C–H···O interactions forming layered patterns of complex cations in complex 1.

Furthermore, lattice stabilization in complex 2 is also facilitated by the interplay of C–H···O and O–H···O hydrogen-bonding interactions. In comparison to complex 1, the crystal lattice of complex 2 was additionally stabilized by lattice water molecules, which led to various O–H···O interactions with the complex cation and anions (O1W–H1W1···O12 = 2.10 Å, O1W–H2W1···O10 = 2.13 Å). The complex cation units in complex 2 showed a chain-like pattern (C² ₃(12)) to form a layered structure of the hydrogen-bonding network (Figure ).

3.

Complex cation units of complex 2 forming endless C² ₃(12) chains of H-bonds.

Additionally, the various hydrogen-bonding interactions (C–H···O and O–H···O) play a significant role in lattice stabilization. Note that no C–H···π and π···π bonding was observed in both cases, which might be attributed to the steric bulkiness of methoxy groups in cinnamate moieties at 2,3- or 2,4-positions in respective complexes 1 and 2. The detailed lattice packing of complexes 1 and 2 with ample hydrogen-bonding interactions is illustrated in Figure S3.

Steric crowding among 2,4-DMCN in complex 2 led to larger vacant spaces inside the lattice besides the random packing arrangement of 2,4-DMCN moieties as compared to 2,3-DMCN in complex 1, which showed an alternate orientation of 2,3-DMCN in the lattice. In complex 2, the Ni­(II)-based complex cations partially occupy structural voids, facilitating the formation of robust hydrogen-bonding networks that stabilize the characteristic layered ribbon-like architectures. The pictorial representation of the arrangement of complex cationic and anionic layers in complexes 1 and 2 is illustrated in Figure . Various hydrogen-bonding parameters used to explore the packing analysis are given in Table S3. Furthermore, a comprehensive comparison of the supramolecular networks and non-covalent interactions of complexes 1 and 2 with similar systems − is detailed in Table S4.

4.

Pictorial representation of the arrangement of complex cationic and anionic counterparts in the lattice of complexes 1 (a) and 2 (b).

3.3. Fourier Transform Infrared (FT-IR) and Electronic Spectroscopies (UV–Vis)

The infrared spectra, i.e., stretching and bending vibrations, for the synthesized complexes 1 and 2 were recorded over the spectral range of 4000–400 cm^–1 and specific band assignments were made with reference to previously reported literature. , The corresponding spectra with band assignments are shown in Figure S4. Furthermore, the theoretically derived FT-IR spectra obtained from DFT calculations for complexes 1 and 2 are also provided in Figure S5. The results show a strong agreement between the theoretically predicted data and experimental findings (Table S5). The absorption bands in the region 3650–3000 cm^–1 were attributed to O–H stretching vibration. In case of complex 1, the sharp band at 3347 cm^–1 represents the O–H stretching vibration, whereas in case of complex 2, slight band broadening at 3335 cm^–1 significantly indicates O–H stretching vibration involving stronger hydrogen bonding, possibly due to the presence of a coordinated water molecule. The weak intensity bands in the region 3000–2250 cm^–1 are attributed to methylene (−CH₂−) stretching vibrations associated with H₃ tea and sp² C–H stretching vibrations associated with the dimethoxy ligand. Other weak intensity bands observed in the region 1650–1480 cm^–1 correspond to sp² CHCH stretching vibrations of cinnamate ligand. In case of complex 1, the bands at 1461 cm^–1 and 1372 cm^–1 were assigned to υ_as(COO) asymmetric and υ_S(COO) symmetric stretching vibrations of the carboxylate moiety of 2,3-DMCN, respectively. The bands at 1269 cm^–1 and 1057 cm^–1 were assigned as υ_as(C–O–C) asymmetric and υ_S(C–O–C) symmetric stretching vibrations of the methoxy groups of 2,3-DMCN. The weak bands in the range 700–400 cm^–1 referred to the Ni–N and Ni–O stretching vibrations. Similarly for complex 2, the bands observed at 1372 cm^–1 and 1271 cm^–1 were assigned as υ_as(COO) asymmetric and υ_s(COO) symmetric stretching vibrations of the carboxylate moiety of 2,4-DMCN, respectively. The bands at 1271 cm^–1 and 1022 cm^–1 were assigned as υ_as(C–O–C) asymmetric and υ_S(C–O–C) symmetric stretching vibrations of the methoxy groups of 2,4-DMCN. The weak bands in the range 700–400 cm^–1 referred to the Ni–N and Ni–O stretching vibrations as for complex 1.

To gain insights into electronic transitions, UV–Visible spectra for both complexes were recorded using a solvent mixture of methanol:water (4:1, v/v) and a concentration of 10 mg/mL. The absorbance bands in the visible region at 645 and 637 nm for respective complexes 1 and 2 delineated the d–d transition for the NiN₂O₄ chromophore. Additionally, absorbance peaks in the region 389–383 and 246–230 nm for complexes 1 and 2 corresponded to the extended π–π* and n–π* transitions, respectively (Figure S6).

3.4. X-ray Photoelectron Spectroscopy (XPS)

XPS analysis was employed to determine the valence states of the primary components that are present within the complexes. The XPS survey spectra of complexes 1 and 2, as depicted in Figure S7, revealed distinct peaks corresponding to the presence of nickel, oxygen, nitrogen, and carbon in both complexes. The peak deconvolution for nickel, oxygen, and nitrogen in Figure S8 indicated that for complex 1, the Ni­(II) species were observed at 855.86 eV (Ni 2p_3/2) and 873.14 eV (Ni 2p_1/2) and for complex 2, the Ni­(II) species were observed at 855.76 eV (Ni 2p_3/2) and 873.50 eV (Ni 2p_1/2) along with broad satellite peaks. The observed peaks at 532.42 eV (complex 1) and 532.60 eV (complex 2) corresponded to O 1s, ascribed to the coordinated oxygen (metal–oxygen bond). The observed peaks at 399.93 eV (complex 1) and 400.04 eV (complex 2) corresponded to N 1s, attributed to the coordinated nitrogen (metal–nitrogen bond). In brief, XPS clearly delineated the coordination feature of Ni­(II) with ligand N- and O-donors. The quantitative assessment of both complexes from XPS is illustrated in Table S6.

3.5. Quantum Mechanical Calculations

Quantum mechanical calculations were performed using density functional theory (DFT) to provide the favorably optimized structures and to inspect various electronic properties of the complexes using the B3LYP/6-311G­(d,p) level of theory. The molecular structures of complexes 1 and 2 were visualized with the help of Avogadro software. HOMO-LUMO analysis was used to derive the molecular properties of the complexes, such as the electron affinity (EA), ionization potential (IP), chemical potential (μ), electronegativity (X), chemical hardness (η), softness (S), and electronegativity index (ω).

3.5.1. Energy and Geometry Optimization

To determine the most favorable geometry of the synthesized complex 1, quantum mechanical calculations were performed using the B3LYP/6-311G­(d,p) level of theory. Figure S9 shows the optimized geometry of complexes 1 and 2, and comparison of the bond lengths and bond angles obtained from SC-XRD data and theoretically predicted by DFT calculations showed the similarities (Tables S7 and S8). The bond lengths and Mulliken charges were calculated for the optimized geometry of complex 1, corresponding to the lowest-energy configuration.

3.5.2. Frontier Molecular Orbitals

The chemical stability of the molecule is effectively described via HOMO LUMO analysis by utilizing the frontier molecular orbital (FMO) theory. The energetic behavior and electron distribution of the complex can be explained by analyzing the HOMO and LUMO energies. Ionization potential (IP) and electron affinity (EA) are defined as IP = −E _HOMO and EA = −E _LUMO. The chemical reactivity and kinetic stability of the complexes can be determined by the energy gap, i.e., E _LUMO – E _HOMO. − The energy gap of complex 2 was found to be relatively small (0.7527 eV), which signifies its high reactivity and polarizable nature. − Figure represents the HOMO–LUMO energy diagram of complexes 1 and 2 with their characteristic band gaps. Based on the calculated HOMO–LUMO energy gaps, complex 2 exhibited a smaller gap compared to complex 1. Therefore, the reactivity follows the order: complex 2 > complex 1.

5.

HOMO-LUMO energy diagrams of complexes 1 and 2.

The quantum chemical calculations utilize various global reactivity descriptors: IP, EA, electronegativity (X), global electrophilicity index (ω), hardness (η), softness (S), and chemical potential (μ) to establish the correlations among energy, structure, and reactivity. These can be derived from the energies corresponding to FMOs, specifically the HOMO and LUMO energies. Koopman’s theorem was employed to evaluate the global reactivity descriptors of complexes 1 and 2, which are defined by the mathematical expressions listed in Table . The chemical potential (μ) provides insights into charge transfer that occuring within the complexes in their ground state. The electrophilicity index (ω) indicates the electrophilic nature of the complexes, describing the chemical reactivity. Electronegativity (X) defines the ability to attract electron pairs shared in a covalent bond, and the global softness describes the chemical reactivity of the complexes. Among the two, complex 1 showed the highest hardness (η) value of 1.0084 eV, indicating a relatively lower chemical softness and consequently reduced propensity toward chemical reactivity. In contrast, complex 2 displayed the lowest η value at 0.3763 eV, reflecting greater softness and enhanced tendency for chemical reactivity. These observations are consistent with the disparities noted in the band gaps of complexes 1 and 2. Notably, literature precedents suggest that higher HOMO energies enhance electron-donating capability toward biological receptors, while lower LUMO energies facilitate electron acceptance; both electronic traits are pivotal for stabilizing ligand–protein interactions. This dual electronic character is prominently exhibited by the complexes reported herein.

2. Global Reactivity Descriptors of Complexes 1 and 2 .

molecular descriptors	mathematical expression	complex 1	complex 2
HOMO (E HOMO) (eV)	–	0.8383	0.0435
LUMO (E LUMO) (eV)	–	2.8550	0.7962
energy gap (E g) (eV)	E (LUMO) – E (HOMO)	2.0167	0.7527
ionization potential (IP)	IP = −E (HOMO)	–0.8383	–0.0435
electron affinity (EA)	EA = −E (LUMO)	–2.8550	–0.7962
electronegativity (X)	X = $\frac{\mathrm{I}\mathrm{P}+\mathrm{E}\mathrm{A}}{2}$	–1.8460	–0.4198
chemical potential (μ)	μ = −( $\frac{\mathrm{I}\mathrm{P}+\mathrm{E}\mathrm{A}}{2}$ )	1.8460	0.4198
hardness (η)	η = $\frac{\mathrm{I}\mathrm{P}-\mathrm{E}\mathrm{A}}{2}$	1.0084	0.3763
softness (σ)	σ = 1/η	0.9917	2.6570
global softness (S)	S = 1/2η	0.4958	1.3280
global electrophilicity index (ω)	ω = μ2/2η	1.6897	0.2340
total energy (a.u.)	–	–1932.057	–2658.357

3.5.3. Mulliken Population Analysis

Calculation of the Mulliken population analysis , delves into the partitioning of total charge density and provides insights on some of the molecular properties such as dipole moment and polarizability. This analysis gives an estimated partial atomic charge on each atom, which is essential for developing an understanding of how molecules interact with one another. In both complexes, the oxygen and nitrogen donor atoms of the triethanolamine ligand possess significantly negative electrostatic potential (negative net atomic charge), as shown by the Mulliken charge distribution in Table S9. These negatively charged sites serve as active coordination centers for the Ni­(II) cation, thereby facilitating charge transfer within the complex. The Mulliken charge analysis revealed significant electron delocalization across the ligand framework and a partially positive Ni­(II) cation in both complexes.

3.5.4. Molecular Electrostatic Potential

MEP analysis revealed the electron density around each atom, providing useful information about the electrophilic and nucleophilic recognition sites and depicting the intermolecular interaction (hydrogen -bonding) sites. The MEPs of complexes 1 and 2 are presented in Figure , the negative areas representing high electron density (red color) were linked to electrophilic reactivity, and the positive areas representing low electron density (blue color) were related to nucleophilic reactivity. The MEP maps illustrate a prominent electropositive region (blue) localized on the carbon skeleton of the triethanolamine moiety, in contrast to a high negative potential (red) centered on the carboxylate oxygens of the cinnamate ligands. This distinct charge separation delineates well-defined electrophilic and nucleophilic domains within the molecular architecture. Consequently, the integrated MEP and FMO analyses provide a coherent electronic-level rationale for the observed antibacterial activity, demonstrating that the spatial distribution of electrostatic potentials, alongside frontier orbital energies, critically governs the interactions between these complexes and their biological targets.

6.

MEPs of complexes 1 (a) and 2 (b).

3.6. Hirshfeld Surface (HS) Analysis

HS analysis is an illustrative tool to quantify and assess the nature and role of weak interactions including hydrogen-bonding interactions, C–H···π interactions, π···π stacking, etc. The d _e and d _i properties in HS examinations help in describing various non-covalent interactions. The characteristic ranges of these interactions, denoted by different colors, scale ranging from red to blue: 0.615–2.401 Å and 0.614–2.571 Å for complex 1 and 0.635–2.531 Å and 0.636–2.649 Å for complex 2 gave a clear illustration about different electron density regions (Figure S10). Color coding acts as markers for identifying different kinds of interactions. For example, red spots on the HS plots correspond to electronegative regions (typically associated with oxygen or nitrogen lone pairs), whereas blue areas highlight electropositive regions, such as those surrounding the Ni­(II) centers or hydrogen atoms. Apart from this, the van der Waals interaction region is marked as white, indicating very weak interactions (intra- or intermolecular). The triangular red and blue spots on the HS plotted over the shape index indicated the intermolecular π···π stacking interactions in the lattice. The various 3D patterns such as d _i, d _e, d _norm, shape index, curvedness, and fragment patch for complexes 1 and 2 are presented in Figure S10.

Moreover, 2D fingerprint plots were also generated to gain quantitative estimation of non-covalent intermolecular interactions in the crystal structure. The 2D fingerprint plots of all the intermolecular contacts are shown in Figure S11. . The 2D fingerprint plots (d _e vs d _i) were employed to visualize interatomic contacts, with the characteristic spikes delineate specific interatomic contacts that contribute significantly to the stabilization of crystal packing. Additionally, 2D plots signify the quantification of the overall non-covalent interaction percentage in addition to individual non-covalent interactions. For better understanding, the percentage contribution of each interaction within the HS to atoms present in the vicinity of the HS was analysed and represented as pie charts (Figure ). In the case of complex 1, the maximum contributions to the HS were provided from H···H (57.8%), O···H (14.6%), and H···O (17.4%). The minimum contributions are observed from C···H (1.9%), H···C (5.4%), C···C (1.0%), O···O (0.1%), C···O (0.9%), and O···C (0.9%). Apart from this, in the case of complex 2, the maximum contributions to the HS are also from H···H (52.8%), O···H (14.9%), and H···O (13.3%), whereas C···H (10.0%), H···C (8.0%), C···C (0.3%), O···O (0.3%), C···O (0.2%), and O···C (0.2%) contribute very little. Further details such as general surface information, surface properties information, and the information about the surface area of nearby interactions between atoms within and outside the surface are presented in Tables S10–S12.

7.

Percentage contributions of the non-covalent interactions of the atoms inside the HS to an atom outside the HS for complexes 1 and 2.

Additionally, void analysis was also performed, a step toward mechanical stability assessment in crystal packing. The void surface analysis corresponded to an isosurface of the procrystal electron density and was computed for the entire unit cell. Enclosed volumes are defined at the interface between void surface and unit cell boundaries, including the regions bounded by the capping faces introduced into these intersections. The voids in the crystal packing of complexes 1 and 2 are illustrated in Figure S12.

The calculated void volume of 94.48 ų (10.37%) and 227.21 ų (12.02%) corresponded to complexes 1 and 2, respectively. Void surface analysis of both complexes 1 and 2 clearly delineated that there are no significant empty spaces within the packed crystal, inferring close packing of molecules in the crystal lattice. The surface area of the void, globularity, and asphericity of complex 1 are 300.26 Ų, 0.334, and 0.024, respectively. The respective corresponding values for complex 2 are 695.08 Ų, 0.259, and 0.098.

3.7. Energy Framework Calculations

The total intermolecular interaction energies (E _tot) in kJ/mol include different energy components: electrostatic (E _ele), polarization (E _pol), dispersion (E _dis), and repulsion (E _rep),as listed in Tables and S13 for complexes 1 and 2, respectively, along with corresponding symmetry operations. The molecular pairs that are involved in the evaluation of the intermolecular interaction energies along the c-axis for complex 1 are illustrated in Figure a. The Coulombic interaction energy and dispersion energy for complex 1, visualized along the c axis, are illustrated in Figure b,c using red and green color-coding schemes, respectively.

3. Interaction Energies (kJ/mol) in the Crystal Structure of Complex 1, Calculated Using the B3LYP/6-31G­(d,p) Energy Model .

^a N indicates the number of molecules with a molecular centroid-to-centroid distance of R (Å). The color code aligns with that used in Figure a.

8.

Representation of the interaction energies of complex 1 along the c axis, (a) electrostatic potential forces, (b) Coulomb interaction energy, and (c) dispersion energy.

Cylindrical representations in the energy framework depict the relative strength of the intermolecular interactions, providing insight into the anisotropy and directional preferences of molecular packing within the crystal structure. For complex 1, the molecule highlighted in red, located 10.20 Å from the centroid of the reference molecule, exhibits a total interaction energy of −4.9 kJ/mol, while the molecule highlighted in green located 8.67 Å from the centroid of the reference molecule exhibits the total interaction energy of −77.9 kJ/mol.

Similarly, the molecular pairs that are involved in the evaluation of the intermolecular interaction energies in complex 2 are depicted in Figure a–c. For complex 2, the molecule highlighted in red, located 7.09 Å from the centroid of the reference molecule, exhibits a total interaction energy of −23.8 kJ/mol, while the molecule highlighted in green located 11.01 Å from the centroid of the reference molecule exhibits a total interaction energy of −9.4 kJ/mol. The dominant interaction types for both complexes are electrostatic or Coulomb interaction energies, as indicated by the width of the cylinders (Figures b and b).

9.

Representation of the interaction energies of complex 2 along the c axis, (a) electrostatic potential forces, (b) Coulomb interaction energy, and (c) dispersion energy.

3.8. In Vitro Antibacterial Study

To determine the minimum inhibitory concentration (MIC) of the synthesized complexes 1 and 2, the 96-well plate broth microdilution method was adopted. The stock solution of 1 mg mL^–1 in dimethyl sulfoxide (DMSO) of both complexes was prepared. A column containing 12 wells was specified for the analysis, with the last two columns serving as positive control having Muller–Hinton broth (MHB), different bacterial strains, resazurin dye, and reference drug chloramphenicol and negative control having MHB, different bacterial strains, resazurin dye, and DMSO. In each row, the first well received 200 μL of MHB containing complexes 1 and 2 in DMSO at a concentration of 500 μg mL^–1 (representing the highest concentration); after this, in all wells, the remaining 100 μL of MHB was added. From the first well, 100 μL was transferred into the subsequent well using a sterile tip, thereby reducing the concentration to half of the original. The procedure was repeated sequentially, resulting in a two-fold serial dilution, generating concentration ranging from 500 to 0.244 μg mL^–1. In each well, 10 μL of bacterial culture and 10 μL of resazurin dye were added, and the contents were properly mixed. After this, the microtiter plate was incubated at 35 °C for 24 h under controlled conditions. Following incubation, the bacterial growth was assessed to determine the MIC (Table ).

4. Representing the Antibacterial Potency in Terms of MIC (μg/mL) of Complexes 1 and 2 against Gram-Negative and Gram-Positive Bacterial Strains.

	MIC (μg/mL)
	Gram-negative bacteria		Gram-positive bacteria
	P. aeruginosa	K. pneumoniae	S. aureus	B. cereus
complex 1	0.48	0.48	125	250
complex 2	0.48	0.97	125	250
chloramphenicol	7.81	3.10	3.78	3.78

Based on the literature, an evaluation was carried out to determine the effectiveness of the synthesized Ni­(II) complexes 1 and 2 against a range of bacterial strains, with comparisons made to previously reported data (Table ). From the assessment that is relied on the MIC values, it was observed that complexes 1 and 2 exhibited significant antibacterial activity. Both complexes demonstrated higher antibacterial efficacy against Gram-negative bacterial strains, P. aeruginosa and K. pneumoniae, with an equal MIC value of 0.48 μg/mL, compared to the Gram-positive bacterial strains, S. aureus and B. cereus, with MIC values of 125 and 250 μg/mL, respectively. The main component of the cell wall of Gram-positive and Gram-negative bacteria is peptidoglycan. The Gram-negative bacteria have a thin layer of peptidoglycan, whereas the Gram-positive bacteria have a thick layer of peptidoglycan. This enhanced activity can be rationalized based on chelation theory and its influence on biological activity. The partial sharing of metal ion’s positive charge with the ligand donor groups, chelation significantly lowers the polarizability of metal ion, in addition to this chelation also increases the lipophilicity of the chelate complexes. These physiochemical changes facilitate the interaction of the metal complex with bacterial cell walls and cause interference with normal bacterial cell functions and inhibit bacterial growth. ,,

5. Comparison of the Antibacterial Potency (MIC) of Complexes 1 and 2 with Previously Reported Ni­(II) Complexes against Different Bacterial Strains.

Ni(II) complexes	bacterial strain	MIC (μg/mL)	refs
thiosemicarbazone	P. aeruginosa	500
tridentate N3 ligand	S. aureus	140
amidodithiophosphonato	S. aureus	100
Schiff bases	P. aeruginosa	256
new sulfonamide derivatives	S. aureus	220
polyhydroxybenzaldehyde N4-thiosemicarbazones	P. aeruginosa	31.2
carbothioamide derivatives	P. aeruginosa	7.34
imine-based tridentate ligand	P. aeruginosa	15.63
[Ni(H3tea)2](2,3-DMCN)2 (1)	S. aureus	125	this work
[Ni(H3tea)2](2,4-DMCN)2(H2O) (2)	P. aeruginosa	0.48	this work

It is noteworthy to mention here that both Ni­(II) complexes demonstrated remarkable Gram-negative selectivity, outperforming many conventional antibiotics that typically fail to penetrate the restrictive outer membrane. This 500-fold potency differential (0.5 vs 250 μg/mL) is driven by membrane-selective translocation; the [Ni­(H₃ tea)₂]²⁺ core and lipophilic methoxycinnamate anions facilitate efficient partitioning into the Gram-negative lipid bilayer. Conversely, the dense Gram-positive peptidoglycan mesh acts as a physical sieve, sterically entrapping these bulky supramolecular assemblies and precluding cellular entry. ,

3.9. In Silico Antibacterial Study

To gain deeper insight into the interactions between synthesized complexes 1 and 2 and target proteins: P. aeruginosa with PDB ID: 1UIZ and K. pneumoniae with PDB ID: 5XUN (Gram-negative) and S. aureus with 5TW8 and B. cereus with 1YQH (Gram-positive), we performed molecular docking studies. Each complex unveiled substantial interactions with a pair of Gram-negative (P. aeruginosaandK. pneumoniae) and Gram-positive bacteria (S. aureus and B. cereus) receptors via hydrogen -bonding, electrostatic, π–π, π–σ, π–alkyl, π–cation, and π–anion interactions. It was evident from the binding free energy data that complex 1 exhibited the most significant binding energy and most reassuring inhibition constant. Accordingly, the interactions of complexes 1 and 2 with the target proteins: P. aeruginosa,K. pneumoniae,S. aureus, and B. cereus were examined in detail.

Interactions between complex 1 and P. aeruginosa (PDB ID: 1UIZ) include conventional hydrogen-bonding interaction, carbon–hydrogen bonding, T-shaped π–π* stacking, and π–alkyl interactions (Figure a₁2D and Figure a₂3D binding mode). Two conventional hydrogen-bonding interactions involving the oxygen atoms of 2,3-dimethoxycinnamic acid, one with TYR15 having a bond length of 2.10 Å and the other with ASN47, measuring a bond length of 2.67 Å, were observed. The carbon–hydrogen bonding involves the interaction of the C–H group of the 2,3-dimethoxycinnamic acid and triethanolamine ligands with a polar atom of the TYR15 amino acid residue. The T-shaped π–π* stacking involves the π-electron cloud of the 2,3-dimethoxycinnamic acid and the π-electron cloud of the TYR15, perpendicular to each other, and the other π–alkyl interaction also involves the π-electron cloud of the ligand interaction with the hydrophobic part (alkyl side chains) of the TYR15 amino acid residue. Complex 1 exhibits a binding free energy of −8.1 kcal mol^–1 and a calculated inhibition constant (K _i) of 1.142 μM.

10.

Molecular docking analysis of complexes 1 against (a) P. aeruginosa, (b) K. pneumoniae, (c) S. aureus, and (d) B. cereus. Subscript 1 represents 2D and 3D binding modes.

Complex 1 shows captivating interactions with K. pneumoniae (PDB ID: 5XUN), involving conventional hydrogen bonding, carbon–hydrogen bonding, π–donor hydrogen bond, T-shaped π–π*, and π–alkyl interactions with different amino acid residues (Figure b₁2D and b₂3D binding mode). Four traditional hydrogen bonds were observed between the oxygen atoms of the ligands: 2,3-dimethoxycinnamic acid and triethanolamine and ASN88 (bond length of 2.70 Å), GLN87 (2.73 Å), SER68 (3.37 Å), and ARG96 (2.44 Å). The carbon–hydrogen bonding involves the carbon–hydrogen bond of the ligands and THR93 (2.76 Å) and ASN88 (2.70 Å) amino acid residues; the π–donor hydrogen bonding involves the hydrogen attached to the electronegative atom of the triethanolamine ligand and the PHE133 (2.92 Å) amino acid residue. The T-shaped π–π* stacking was observed between the π-electron cloud of the 2,3-dimethoxycinnamic acid and the π-electron cloud of TYR86 (4.87 Å) and a π–alkyl interaction is observed with the TYR 86 amino acid residue. The binding free energy of −7.7 kcal mol^–1 and inhibition constant of 2.245 μM were observed in case of complex 1 against the K. pneumoniae bacterial strain.

Interactions of complex 1 with S. aureus (PDB ID: 5TW8) involve three conventional hydrogen bonding, carbon–hydrogen bonding, π–anion bonding, and π–hydrogen bonding (Figure c₁2D and Figure c₂3D binding mode). The oxygen atoms of the ligands and the serine residues: SER75 (2.32 Å), SER262 (2.57 Å), and SER116 (2.07 Å) of the amino acids are involved in the hydrogen bonding; the interaction distance is shown in the bracket. The carbon–hydrogen-bonding, π–anion-bonding, and π–donor hydrogen-bonding interactions involve ASP264 (3.16 Å), GLU114 (4.32 Å), and ASP264 (3.01 Å) residues, respectively. Complex 1 exhibited a binding free energy of −7.9 kcal mol^–1 and inhibition constant of 1.601 μM against the S. aureus bacterial strain.

Complex 1 exhibited hydrogen bonding, carbon–hydrogen bonding, attractive charges, π–alkyl bonding, and unfavorable negative–negative interactions against the B. cereus (PDB ID: 1YQH) bacterial strain (Figure d₁2D and Figure d₂3D binding mode). The conventional hydrogen bonding involves the hydrogen atom attached to the electronegative oxygen atoms of the ligands and GLU29 (2.01 Å) amino acid residue. The carbon–hydrogen bonding is formed between the carbon–hydrogen bond of the ligand and GLN33 (3.49 Å) residue. Electrostatic attraction involves the positively and negatively charged residue of the ligands and the GLU29 (2.01 Å), ASP67 (4.76 Å), and LYS26 (5.39 Å) residues. The unfavorable negative–negative interaction (electrostatic repulsion) involves the negatively charged atom of the ligands and ASP67 (4.07 Å) amino acid residue and π–alkyl interactions involves the π-electron cloud of the ligand and the ARG60 (3.97 Å) amino acid residue. The binding free energy of −5.4 kcal mol^–1 and inhibition constant of 109.28 μM were calculated in case of complex 1 against the B. cereus bacterial strain.

Interactions of complex 2 with P. aeruginosa (PDB ID: 1UIZ) involve hydrogen-bonding, carbon–hydrogen-bonding, alkyl, and π–alkyl interactions with different amino acid residues (Figure a₁2D and Figure a₂3D binding mode). The conventional hydrogen bonding involves the electronegative oxygen atom of the ligand: 2,4-dimethoxycinnamic acid and ASN47 (2.51 Å) amino acid residue; the bond distance of the interaction is shown in the bracket. The carbon–hydrogen bonding, alkyl, and π–alkyl interactions involve PRO16 (3.27 Å), ARG14 (4.35 Å), PRO51 (4.94 Å), LEU60 (3.82 Å), PHE45 (4.86 Å), and PHE45 (5.34 Å), respectively. Complex 2 exhibited the binding free energy of −9.5 kcal mol^–1 and the inhibition constant of 0.107 μM against the P. aeruginosa bacterial strain.

11.

Molecular docking analysis of complexes 2 against (a) P. aeruginosa, (b) K. pneumoniae, (c) S. aureus, and (d) B. cereus. Subscript 2 represents 2D and 3D binding modes.

Complex 2 displays robust interactions with different amino acid residues of K. pneumoniae (PDB ID: 5XUN), involving conventional hydrogen-bonding, carbon–hydrogen-bonding, π–anion, π–π* stacking, alkyl, and π–alkyl interactions (Figure b₁2D and Figure b₂3D binding mode). The hydrogen bonding involves the interactions of the oxygen atoms of the ligands: 2,4-dimethoxycinnamic acid and triethanolamine with ASN88 (2.48 Å), GLN87 (2.87 Å), and ASN88 (2.66 Å) amino acid residues. The carbon–hydrogen bonding involves interactions between carbon–hydrogen bonds of the ligands and TYR86 (3.86 Å) and GLY67 (3.53 Å). The π–π* stacking involves the π-electron cloud of the ligands and the π-electron cloud of the TYR86 (3.86 Å) amino acid residue. The π–anion, alkyl, and π–alkyl interactions were also observed with GLU135 (3.99 Å), LEU45 (4.0 Å), and PRO90 (3.72 Å), respectively. Complex 2 exhibited the binding free energy of −7.6 kcal mol^–1 and inhibition constant of 2.658 μM against the K. pneumoniae bacterial strain.

Interactions of complex 2 with the amino acid residues of the S. aureus (PDB ID: 5TW8) involve five conventional hydrogen bonding, carbon–hydrogen bonding, electrostatic attraction, π–anion, π–sigma, alkyl, π–alkyl interactions (Figure c₁2D and Figure c₂3D binding mode). The hydrogen bonding involves the interactions of the electronegative oxygen atoms of the ligands 2,4-dimethoxycinnamic acid and triethanolamine with the amino acid residues: GLN228 (2.28 Å), ASN267 (2.26 Å), LYS221 (1.97 Å), ASP97 (3.20 Å), and THR265 (3.36 Å). The carbon–hydrogen bonding involves the interactions of the carbon–hydrogen bond of the ligands with the ASP290 (3.79 Å), ASP224 (3.46 Å), and THR69 (3.51 Å) amino acid residues. The electrostatic attractive interaction was observed between the positively and negatively charged species of the ligand and the amino acid residues: ASP290 (3.79 Å) and ASP224 (3.46 Å). The π–anion, π–sigma, alkyl, and π–alkyl interaction involves interactions with ASP290 (3.07 Å), THR50 (3.68 Å), VAL202 (3.89 Å), and LYS70 (5.33 Å) amino acid residues, respectively. Complex 2 exhibited the binding free energy of −9.2 kcal mol^–1 and inhibition constant of 0.178 μM against the S. aureus bacterial strain.

Complex 2 shows various interactions with the amino acid residue of B. cereus (PDB ID: 1YQH), including hydrogen bonding, carbon–hydrogen bonding, electrostatic attractive and electrostatic repulsive, alkyl, and π–alkyl interactions (Figure d₁2D and Figure d₂3D binding mode). The conventional hydrogen bonding involves the oxygen atom of the ligand and LYS26 (5.19 Å) amino acid residue. The carbon–hydrogen bonding involves the interaction of the carbon–hydrogen bond of the ligand with the ASP67 (3.30 Å) amino acid residue. The electrostatic attractive interaction was observed between the positively and negatively charged species of ligands and amino acid residues GLU29 (4.48 Å), ASP67 (3.30 Å), and LYS26 (5.19 Å). The unfavorable negative–negative interaction that represents the electrostatic repulsive interaction was observed between the like charged species of the ligand and ASP67 (3.23 Å) amino acid residue. The alkyl and π–alkyl interactions were also observed with ALA68 and LYS26 amino acid residues. The binding free energy of −4.7 kcal mol^–1 and inhibition constant of 356.53 μM were calculated for complex 2 against the B. cereus bacterial strain.

The binding scores and corresponding inhibition constants of complexes 1, 2, and reference standard (chloramphenicol) with target bacterial proteins are listed in Table . Moreover, various non-covalent interactions between the amino acid residues and complex binding sites are provided in Table . The radar plots of MIC values obtained by in vitro antibacterial study and binding free energy (docking score) derived by in silico antibacterial study are displayed in Figure . The MIC values are lower for P. aeruginosa and K. pneumoniae (Gram-negative) compared to the other tested bacterial strains. In addition, the binding free energy values are also more negative for both, indicating a stronger and stable interaction between the complex and the targeted protein. This shows better agreement of the in vitro and in silico antibacterial studies.

6. Binding Free Energies and Inhibition Constants of Complexes 1, 2, and Reference Standard (Chloramphenicol) with Target Bacterial Proteins.

complexes	target ID	cells	binding free energy (kcal mol–1)	inhibition constant (K i) (μM)
complex 1	1UIZ	P. aeruginosa	–8.1	1.142
	5XUN	K. pneumoniae	–7.7	2.245
	5TW8	S. aureus	–7.9	1.601
	1YQH	B. cereus	–5.4	109.28
complex 2	1UIZ	P. aeruginosa	–9.5	0.107
	5XUN	K. pneumoniae	–7.6	2.658
	5TW8	S. aureus	–9.2	0.178
	1YQH	B. cereus	–4.7	356.53
chloramphenicol	1UIZ	P. aeruginosa	–6.6	10.27
	5XUN	K. pneumoniae	–7.1	6.025
	5TW8	S. aureus	–6.5	17.04
	1YQH	B. cereus	–4.6	422.09

7. Binding Free Energies, Inhibition Constants (K _i), and the Interaction Distances with Corresponding Amino Acids of Complexes 1 and 2 against P. aeruginosa (PDB ID: 1UIZ), K. pneumoniae (PDB ID: 5XUN), S. aureus (PDB ID: 5TW8), and B. cereus (PDB ID: 1YQH).

complex ID	target ID	binding free energy (kcal mol–1)	inhibition constant (K i) (μM)	amino acids residues	type of interactions	distance (Å)
complex 1	1UIZ	–8.1	1.142	ASN47, TYR15, TYR15, TYR15	hydrogen bonding, carbon–hydrogen bonding, π–π* stacking, π–alkyl	2.67, 3.54, 3.46, 2.67
	5XUN	–7.7	2.245	SER68, ASN88	hydrogen bonding, carbon–hydrogen bonding, π–donor hydrogen bond, π–π* stacking, π–alkyl	3.37, 2.70, 2.92, 4.87, 4.87
				PHE133, TYR86
				TYR86
	5TW8	–7.9	1.601	SER75, ASP264, GLU114, ASP264	hydrogen bonding, carbon–hydrogen bonding, π–anion, π–donor hydrogen bond	2.32, 3.26, 4.32, 3.10
	1YQH	–5.4	109.28	GLU29, GLN33, ASP67, ARG60, ASP67	hydrogen bonding, carbon–hydrogen bonding, attractive charges, π–alkyl, unfavorable negative–negative	2.01, 3.49, 4.76, 3.97, 4.07
complex 2	1UIZ	–9.5	0.107	ASN47, PRO16, ARG14, LEU60	hydrogen bonding, carbon–hydrogen bonding, alkyl, π–alkyl	2.15, 3.27, 4.35, 3.82
	5XUN	–7.6	2.658	GLN87, GLY67, GLU135, TYR86, LEU45, PRO90	hydrogen bonding, carbon–hydrogen bonding, π–anion, π–π* stacking, alkyl, π–alkyl	2.87, 3.53, 3.99, 3.86, 4.0, 3.72
	5TW8	–9.2	0.178	GLN228, ASP290, ASP224, ASP290, THR50, VAL202, LYS70	hydrogen bonding, carbon–hydrogen bonding, attractive charges, π–anion, π–sigma, alkyl, π–alkyl	2.28, 3.79, 3.46, 3.07, 3.68, 3.89, 5.31
	1YQH	–4.7	356.53	LYS26, ASP67, GLU29, ASP67, ALA68, LYS26	hydrogen bonding, carbon–hydrogen bonding, attractive charges, unfavorable negative–negative, alkyl, π–alkyl	5.19, 3.30, 4.48, 3.23, 3.90, 5.19

12.

Radar plots of (a) MIC (μg mL^–1), (b) binding free energy (kcal mol^–1), and (c) inhibition constant (K _i) (μM) of complexes 1, 2, and reference chloramphenicol. PDB codes: 1UIZ, P. aeruginosa; 5XUN, K. pneumoniae; 5TW8, S. aureus; and 1YQH, B. cereus.

Among both complexes, complex 2 exhibited comparatively more negative binding free energy (docking score) than complex 1, suggesting a stronger predicted affinity toward the biological target. This observation correlates with its higher chemical reactivity and smaller band gap, indicating that the enhanced electronic flexibility of complex 2 contributes to its improved interactions with the target system.

3.10. In Silico Toxicity Prediction

To predict toxicity according to the Globally Harmonized System (GHS) of the synthesized complexes 1 and 2, we performed an in silico toxicity prediction analysis by using the ProTox-II online Web server. This tool was employed to assess different aspects of complex toxicity such as cytotoxicity, ecotoxicity, clinical toxicity and mutagenicity, and the LD₅₀ (lethal dose 50) values; the outcomes are listed in Table S14. Complexes 1 and 2 exhibited the LD₅₀ values of 3900 and 1750 mg kg^–1 which refers to GHS toxicity class IV. The toxicity prediction shows no activity in terms of cytotoxicity, ecotoxicity, clinical toxicity, and mutagenicity, supporting their potential to progress in future metallodrug design. ,

3.11. Mechanistic Insights into Antibacterial Action

The findings of this study provide a comprehensive structural and electronic rationale for the enhanced antibacterial potency of the room-temperature self-assembled [Ni­(H₃ tea)₂]­(DMCN)₂ complexes. SCXRD confirmed the formation of a stable mononuclear cationic core protected by a voluminous hydrogen-bonded outer sphere of dimethoxycinnamates. Mechanistically, the DFT-calculated MEP maps reveal that the H₃ tea octahedral cage effectively delocalizes the Ni­(II) central charge, leading to reduced polarity of the Ni­(II) center, which significantly boosts the overall lipophilicity of the complex. This electronic masking, coupled with the lipophilic methoxy groups of the cinnamate anions, facilitates the complex penetrating the hydrophobic bacterial lipid bilayer, disrupting the membrane structure, inducing depolarization, and leading to cell death. Notably, both Ni­(II) complexes exhibited marked selectivity toward Gram-negative bacteria, surpassing the efficacy of many conventional antibiotics that are often hindered by limited permeability across the outer membrane. This enhanced activity can be attributed to membrane-selective translocation mechanisms. Specifically, the [Ni­(H₃ tea)₂]⁺² core, in conjugation with the lipophilic dimethoxycinnamate anions, facilitates efficient partitioning into the Gram-negative lipid bilayer. In contrast, the thick and highly cross-linked peptidoglycan layer characteristic of Gram-positive bacteria imposes a significant steric barrier, restricting the diffusion of these bulky supramolecular assemblies and thereby limiting cellular uptake. ,

Furthermore, to validate the above-mentioned mechanism, DFT calculations suggested that both complexes with narrow HOMO–LUMO energy gaps possess high chemical softness and reactivity. This electronic configuration facilitates easier interaction with biological nucleophiles. We propose that the antimicrobial activity is driven by a redox-mediated mechanism, where the high-lying HOMO allows the complex to interfere with the bacterial respiratory chain, leading to the intracellular generation of reactive oxygen species (ROS). ROS induces severe oxidative stress in bacteria, damaging DNA, proteins, and cell membranes and giving rise to cell death. Furthermore, molecular docking studies provided the final piece of the mechanistic puzzle, showing that the 2,4-dimethoxy isomer (in 2) achieves superior binding affinity compared to the 2,3-dimethoxy isomer (in 1). This is attributed to the optimized spatial alignment of the para-methoxy groups within the protein’s hydrophobic pockets, stabilized by a synergistic network of π–π stacking and H-bonding. In summary, the transition from a simple metal salt to a self-assembled mixed-ligand complex transforms the nickel center from a biologically sequestered ion into a membrane-permeable, electronically active “Trojan Horse” capable of targeted bacterial disruption. , The pictorial representation of the structure-disrupting mechanism by nickel complexes is illustrated in Figure .

13.

Explanation of the structure–activity relationship (SAR) of complexes 1 and 2.

4. Conclusions

Two self-assembled mononuclear ionic Ni­(II) complexes, [Ni­(H₃ tea)₂]­(2,3-DMCN)₂ (1) and [Ni­(H₃ tea)₂]­(2,4-DMCN)₂(H₂O) (2), have been successfully synthesized at room temperature under ambient conditions after crystallization by a slow evaporation process. Complexes 1 and 2 were spectroscopically and crystallographically characterized. Single-crystal X-ray diffraction revealed the distorted octahedral of both complexes and ionic coordination of dimethoxy-substituted cinnamate ligands. XPS analysis delineated that in both complexes, Ni­(II) is coordinated to nitrogen and oxygen donor ligands. Quantum mechanical calculations further substantiated the quantitative evaluation of interaction energies and band gaps in HOMO–LUMO energy levels, providing insights into chemical reactivity and kinetic stability of both complexes. The computational reactivity ranking (2 > 1) correlates well with the experimental antibacterial efficacy of the complexes. Hirshfeld surface and packing analyses revealed crystal lattice stabilization via the interplay of non-covalent interactions.

In vitro antibacterial studies indicated that both complexes exhibited remarkable antibacterial activity against gram –ve bacterial strains (P. aeruginosa and K. pneumoniae) with corresponding minimum inhibitory concentration (MIC) values of 0.48, 0.48, 0.48, and 0.97 μg/mL, respectively. Furthermore, experimental antibacterial resistance of these complexes has also been corroborated with in silico molecular docking with the P. aeruginosa bacterial strain (PDB ID: 1UIZ). The corresponding binding constants and docking scores are 1.142 and 0.107 μM and −8.1 and −9.5 kcal mol^–1 for the respective 1-1UIZ and 2-1UIZ adducts, which are better than the standard drug analogue, chloramphenicol. The experimental results unequivocally showed corroboration with the MEP and FMO prediction.

Ultimately, the synergistic analysis of the supramolecular architecture, non-covalent interaction networks, and integrated computational modeling provides a comprehensive understanding of the stability and therapeutic potential of these Ni­(II) complexes. This multidimensional approach offers a robust framework for the rational design of next-generation bioactive metallotherapeutics with precisely tuned biological profiles.

Glossary

Abbreviations

SCXRD

single-crystal X-ray diffraction

XPS

X-ray photoelectron spectroscopy

DFT

density functional theory

HS

Hirshfeld surface

H₃ tea

2,2′,2″-nitrilotri­(ethan-1-ol)

2,3-DMCN

2,3-dimethoxycinnamate

2,4-DMCN

2,4-dimethoxycinnamate

MIC

minimum inhibitory concentration

There are no data used in this research article.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsphyschemau.6c00030.

• Crystallographic data (CIFs) of complexes 1 and 2 (ZIP)

• Comparison of synthetic conditions (Table S1); selected bond distances and bond angles and hydrogen bonding parameters of complexes 1 and 2 (Tables S2 and S3); comparison of supramolecular assemblies with the literature (Table S4); comparison of experimental and theoretically derived FT-IR spectra of complexes 1 and 2 (Table S5); peak deconvolution by XPS (Table S6); comparison of structural parameters of complexes 1 and 2 from SCXRD and DFT calculation (Tables S7 and S8); Mulliken charges on atoms of complexes 1 and 2 (Table S9); surface properties from HS analysis (Tables S10–S12); energy framework color coding (Table S13); predicted toxicity results (Table S14); synthetic routes of complexes 1 and 2 (Scheme S1); active sites of bacterial proteins for complexes 1 and 2 (Figure S1); H bonding of complexes 1 and 2 (Figures S2 and S3); FT-IR spectra of complexes 1 and 2 from experiment and DFT calculation (Figures S4 and S5); electronic spectral details and XPS values (Figures S6–S8); and DFT-optimized structures, 3D Hirshfeld surfaces, 2D fingerprint plots, and voids in the crystals of complexes 1 and 2 (Figures S9–S12) (PDF)

Sunidhi: experimental, writing, data interpretation, and data analysis; Chetan Chauhan: formal analysis; Rajesh Kumar: formal analyses, data interpretation and writing, and reviewing; Thammarat Aree: X-ray data collection, analysis, editing, review, supervision, writing, and critical analysis; Santosh Kumar: conceptualization, methodology, data interpretation, writing the manuscript, supervision, and funding and mentoring.

The authors declare no competing financial interest.