Crystallographic, topological, antidiabetic, and docking evaluation of an azo-enamine ligand and its triphenyltin(IV) coordination polymer

Abstract

The ligand 2-((E)-((Z)-3-(((2-hydroxyethyl)amino)methylene)-4-oxocyclohexa-1,5-dien-1-yl)diazenyl)benzoic acid (H₃L) was synthesized and structurally characterized for the first time. Single-crystal X-ray diffraction studies confirmed that H₃L adopts a stable azo-enamine tautomeric form in the solid state stabilized by intramolecular hydrogen bonding. The corresponding polymeric triphenyltin(IV) complex [Ph₃Sn(IV)H₂L] (1) was also synthesized and characterized by FTIR, multinuclear NMR, and X-ray crystallography. Structural analysis revealed a five-coordinate distorted trigonal bipyramidal geometry around the Sn(IV) center. Topological analysis of both structures highlighted distinct hydrogen-bonded and valence-bonded network topologies (2C1 and sql types, respectively), underscoring supramolecular structural features. In vitro α-glucosidase inhibition studies demonstrated that H₃L exhibits strong inhibitory activity, particularly in DMSO (79.36%, IC₅₀ ≈ 26.8 µg mL⁻¹), outperforming the standard inhibitor acarbose, while complex 1 showed negligible inhibition. Molecular docking against α-glucosidase (PDB ID: 5ZCB) supported these findings, showing that H₃L exhibits a more favorable binding affinity (MolDock score: –87.92 kcal mol⁻¹) and a stronger interaction profile than acarbose. These results highlight the potential of H₃L as a promising lead compound for the development of α-glucosidase inhibitors.

---

Introduction

Organotin(IV) carboxylates have garnered considerable attention in recent years due to their structural versatility and promising biological activities.^1–12 These compounds have demonstrated a wide range of bioactivities, including antimicrobial, anticancer, antitumor, antimalarial, urease inhibition, antiproliferative, and antituberculosis effects.^1–12 Structurally, organotin(IV) carboxylates can exist as monomers, dimers, tetramers, oligomeric ladders, hexameric drums, macrocycles, clusters, and cages.^1,3,5,7,12 Their structural diversity arises primarily from the various coordination modes of carboxylate oxygen atoms—monodentate, bidentate bridging, and chelating bidentate. Organotin(IV) carboxylates incorporating azo and imino groups are of particular interest, as both functionalities can serve as spacers, offering a sterically bulky ligand environment. The coordination number of the Sn(IV) center also plays a crucial role in influencing the final structure.

Diabetes, a heritable metabolic disorder, continues to pose a critical global health challenge. Transition metal complexes of vanadium, copper, and zinc are well-documented for their antidiabetic activities.^13–15 In contrast, despite the broad bio-potency of organotin(IV) carboxylates, their antidiabetic potential remains comparatively underexplored, with only a handful of systematic studies available.^3,5,7 In particular, the structural and functional determinants at both the ligand and metal complex levels, including overall architecture, supramolecular assembly, and accessibility of functional groups, that govern antidiabetic activity are poorly defined.

Our earlier studies have shown that fine-tuning the ligand framework strongly influences biological response.^3,5,7 Our group previously reported the X-ray crystal structure and antidiabetic activity of a trimethyltin(IV) complex of 4-(2,4-dihydroxyphenylazo)benzoic acid, which exhibited superior inhibitory activity compared to the standard drug acarbose.⁷ Recently, we also reported a trimethyltin(IV) complex derived from an azo-benzoic acid ligand, 2-[{4-hydroxy-3-[4-hydroxy-3-carboxyphenylimino)methyl]phenylazo}benzoic acid], where both the free ligand (47% α-amylase inhibition) and its Sn(IV) complex (46% inhibition) outperformed acarbose (33%).³ Recent structure–activity relationship analyses emphasized that that the presence and accessibility of phenolic and carboxylic functional groups play a pivotal role in achieving strong α-glucosidase and α-amylase inhibition through hydrogen bonding.^16,17 Nevertheless, systematic efforts to connect structural characteristics and molecular docking insights with antidiabetic performance in organotin(IV) complexes are exceedingly rare.

Recently, we also described the X-ray crystal structure and antibacterial activity of a trimethyltin(IV) complex of an azo-imine-hydroxocarboxylate ligand, 2-((E)-(4-hydroxy-3-((E)-((2-hydroxyethyl)imino)methyl)phenyl)diazenyl)benzoic acid (H₃L). This complex crystallized as a cyclic dimer, with the tin center adopting a distorted trigonal bipyramidal geometry in the solid state.¹² However, neither the antidiabetic potential of the complex nor that of the free ligand H₃L had been previously investigated. Moreover, while H₃L may exist in multiple tautomeric forms (azo-imine, azo-enamine and zwitter ion) (Scheme 1), its solid-state structure had not been reported to date. In the present study, we report for the first time the single-crystal X-ray structure of H₃L in its azo-enamine tautomeric form (Scheme 1). Additionally, we present the synthesis and structural characterization of its triphenyltin(IV) complex, which exhibits a polymeric structure with the tin centers adopting a trigonal bipyramidal geometry in the solid state. The ligand H₃L showed notable α-glucosidase inhibitory activity, with 43.7% inhibition in chloroform and 79.36% in DMSO, compared to acarbose (45.06%), whereas complexation markedly attenuates this activity. To rationalize H₃L's binding affinity at the active site of the α-glucosidase enzyme, a molecular docking study was also carried out and included in this contribution. Collectively, these findings introduce a new lead structure with antidiabetic potential and, more broadly, highlight how functional group accessibility shape bioactivity in organotin(IV) chemistry.

Scheme 1 Possible tautomeric forms of H₃L and synthesis of the polymeric triphenyltin(IV) complex [Ph₃Sn(IV)H₂L] (1).

Experimental

Materials and physical measurements

O-Amino benzoic acid, salicylaldehyde, ethanolamine, triphenyltin(IV) chloride and triethylamine were purchased from MERCK and were used without further purification. Elemental (CHN) analysis was performed on a PerkinElmer 2400 series II instrument. The FTIR spectrum of the complex was obtained using a PerkinElmer FTIR spectrophotometer in the range of 4000–400 cm⁻¹ using a KBr disc. The ¹H and ¹³C NMR spectra of the complex were recorded on a Bruker AMX400 spectrometer at 400.23 and 100.63 MHz, respectively.

Tetramethylsilane (TMS) was used as a standard reference at 0.00 ppm chemical shift for both the ¹H and ¹³C NMR spectra. The ¹¹⁹Sn NMR spectrum of the complex was recorded on the Jeol (Delta 6.1) 400 MHz FT NMR spectrometer with tetramethyltin as the standard reference.

Synthesis of 2-((E)-(4-hydroxy-3-((E)-((2-hydroxyethyl)imino)methyl)phenyl)diazenyl)benzoicacid (H₃L)

The ligand was prepared following the reported method¹² and crystallized from anhydrous methanol. The light-yellow crystals were produced from the methanolic solution and some of them were collected for crystallographic analysis.

Synthesis of the triphenyltin(IV) complex, [Ph₃Sn(IV)H₂L]_n (1)

To a hot methanolic solution of 2-((E)-(4-hydroxy-3-((E)-((2-hydroxyethyl)imino)methyl)phenyl)diazenyl)benzoic acid (H₃L) (0.2 g, 0.638 mmol), an alcoholic solution of triethylamine (0.638 mmol) was added dropwise. The reaction mixture was refluxed for half an hour with stirring and then triphenyltin(IV) chloride (0.245 g, 0.638 mmol) was added with continuous stirring and then again refluxed for 5 hours. The reaction mixture was then filtered under hot conditions and kept for slow evaporation. Reddish brown crystals were obtained.

Yield: 0.30 g, 67.5%. m.p.: 186–188 °C. Anal. calcd for C₃₄H₂₉N₃O₄Sn: C, 61.66; H, 4.41; N, 6.34%. Found: C, 61.68; H, 4.39; N, 6.39%. FTIR (KBr, cm⁻¹): 3439 ν(OH), 2927 ν(C–H str. of aliphatic –CH₂), 2677 ν(C–H str. of aromatic –CH), 1653 ν(COO)asy, 1614 ν(C=N), 1473 ν(N=N), 1390 ν(COO)sym, 1179 ν(aliphatic C–O), 1038 ν(N–H), 699 ν(Sn–C). ¹H NMR (CDCl₃, 400.23 MHz) δ_H: 10.06 [s, 1H, (C=NH)], 7.70–7.26 [aromatic–H (both the ligand and tin-phenyl)], 3.07 [s, 1H, aliphatic–OH], 1.22–1.39 [m, aliphatic H] ppm. ¹³C NMR (CDCl₃, 100.63 MHz) δ_C: peaks were observed at 134.29, 130.11, 128.88, 29.56 ppm.¹¹⁹Sn NMR (CDCl₃, 149 MHz): −128.48 ppm.

Crystallographic data collection and structure refinement

The intensity data of H₃L and 1 were collected at room temperature using a Bruker D8 VENTURE diffractometer equipped with a PHOTON II detector with graphite monochromated MoKa radiation (l = 0.71073 Å). Integration and scaling of the intensity data were accomplished using the program SAINT. Absorption corrections were applied using SADABS. The structure of H₃L was solved with the olex2.solve 1.5-dev¹⁸ solution program using iterative methods. The model was refined with olex2.refine 1.5-dev¹⁸ using full matrix least squares minimisation on F². The structure of complex 1 was solved with ShelXT 2018/2 and refined using ShelXL 2019/3.¹⁹ Both structures were processed through the Olex2 graphical interface.²⁰ All non-hydrogen atoms were refined anisotropically. Hydrogen atom positions were calculated geometrically and refined using the riding model. Most hydrogen atom positions were calculated geometrically and refined using the riding model, but some hydrogen atoms were refined freely. The crystal data and structure refinement details are summarized in Table 1.

Table 1 Crystal data and structure refinement for H₃L and 1

Identification code	H3L	1
Empirical formula	C16H15N3O4	C34H29N3O4Sn
Formula weight	313.315	662.29
Temperature/K	298.00	298.15
Crystal system	Orthorhombic	Monoclinic
Space group	P212121	P21/n
a/Å	4.3533(3)	11.5962(2)
b/Å	12.7658(9)	19.0791(3)
c/Å	26.8421(18)	13.6158(2)
α/°	90	90
β/°	90	91.5350(10)
γ/°	90	90
Volume/Å^3	1491.71(18)	3011.35(8)
Z	4	4
ρ calc g/cm^3	1.395	1.461
µ/mm^−1	0.102	7.095
F(000)	656.5	1344.0
Crystal size/mm^3	0.587 × 0.047 × 0.035	0.302 × 0.113 × 0.086
Radiation	Mo Kα (λ = 0.71073)	CuKα (λ = 1.54178)
2θ range for data collection/°	6.56 to 50	7.978 to 136.702
Index ranges	−5 ≤ h ≤ 4	−13 ≤ h ≤ 13
	−16 ≤ k ≤ 16	−23 ≤ k ≤ 22
	−33 ≤ l ≤ 33	−16 ≤ l ≤ 16
Reflections collected	21 625	65 108
Independent reflections	2614 [Rint = 0.0703, Rsigma = 0.0480]	5522 [Rint = 0.0466, Rsigma = 0.0207]
Data/restraints/parameters	2614/0/213	5522/1/385
Goodness-of-fit on F^2	1.086	1.077
Final R indexes [I ≥ 2σ(I)]	R 1 = 0.0711, wR2 = 0.1465	R 1 = 0.0307, wR2 = 0.0710
Final R indexes [all data]	R 1 = 0.0812, wR2 = 0.1517	R 1 = 0.0359, wR2 = 0.0748
Largest diff. peak/hole/e Å^−3	0.27/−0.29	1.01/−0.44
Flack parameter	0.0(9)

---

The topological analysis was performed with the ToposPro program package and the TTD collection of periodic network topologies.²¹ The RCSR three-letter codes²² were used to designate the network topologies.

Antidiabetic methodology

The antidiabetic activities of the two compounds, (H₃L) and 1, were evaluated using a 96-well microplate-based α-glucosidase assay.²³ Samples for the enzyme assay were prepared by dissolving 1 mg of each compound in 20 µL chloroform (C), ethylacetate (B), methanol (A), and DMSO (D) and diluted to 1000 µL in a 2 mL Eppendorf tube with sterile water (Milli-Q). The final concentration of each organic solvent in the assay mixture was maintained below 2% (v/v), a level confirmed not to interfere with the enzymatic activity, ensuring observed inhibition reflects ligand–enzyme interactions.^24,25 Solvent controls (≤2% v/v) showed no detectable effect on enzyme activity; identical solvent conditions were therefore used for all assays. The α-glucosidase assay was conducted in a 75 µL reaction volume using a 96-well (flat bottom) microplate. A volume of 25 µL sample solution was combined with 25 µL of α-glucosidase enzyme (0.5U) and pre-incubated at 37 °C ± 2 °C for 10 minutes. After pre-incubation, 25 µL of substrate solution (0.5 mM, p-nitrophenyl α-D glucopyranoside, PNPG, in phosphate buffer (pH 6.9)) was added to the mixture and incubated at 37 °C ± 2 °C for 30 minutes. Acarbose (ACB), a standard α-glucosidase inhibitor drug was used as a reference. The reaction was stopped by adding 100 µL of 0.2M sodium carbonate solution. The quantity of p-nitrophenol (yellow colour) released from PNPG was measured at 405 nm using a UV-visible spectrophotometer (Mustikan Sky High, Thermo Scientific) in a 96-well microplate. Appropriate blanks for each sample, standard drug and control reactions were also included in the assay. All experiments were performed in five replicates. The percentage of α-glucosidase inhibition was calculated using the formula:

α-glucosidase inhibition (%) = [(control OD − sample OD)/control OD] × 100

[Control OD = absorbance (OD) of the control reaction (without test sample or inhibitor) − blank OD; sample OD = sample OD − sample blank OD].

Dose-effect analysis of H₃L in chloroform (H₃LC) and DMSO (H₃LD)

Based on the result of a screening test of antidiabetic activities, H₃L dissolved in chloroform (H₃LC) and DMSO (H₃LD) showed significant antidiabetic activities and were further analysed for their dose-effect relationship by calculating IC₅₀. For dose-effect analysis of antidiabetic activity, an in vitro α-glucosidase assay was performed as described above in a 75 µL reaction volume using a 96-well microplate. Sample solutions of (H₃L) (1 mg mL⁻¹) i.e., 5, 10, 15, 20, and 25 µL (∼67, 133, 266, and 333 µg mL⁻¹) were mixed with 25 µL of α-glucosidase enzyme (0.5U), and the volume was adjusted to 75 µL with assay buffer. The mixture was pre-incubated at 37 °C ± 2 °C for 10 minutes. Subsequently, 25 µL of substrate (0.5 mM, PNPG) was added, and the reaction mixture was incubated at 37 °C ± 2 °C for 30 minutes. Acarbose as the standard inhibitor of the enzyme was also included with similar concentrations in the assay. The reaction was stopped by adding 100 µL of 0.2 M sodium carbonate solution. The quantity of p-nitrophenol liberated from PNPG was measured at 405 nm using a UV-visible spectrophotometer (Multiskan Sky High, Thermo Scientific) in a 96-well microplate. Suitable blanks of test samples, acarbose and control reactions were included for each treatment in the assay. All experiments were conducted in five replicates. The percentage enzyme inhibition by different doses of test samples and acarbose was calculated using the same method described above for the screening tests.

Molecular docking methodology

Molecular docking was conducted to investigate the interactions of H₃L and ACB with α-glucosidase (PDB ID: 5ZCB). Key methodological details, including protein preparation, active site characterization, definition of binding sites and grid dimensions, docking protocol parameters, and binding affinity estimation, are provided in the SI. Briefly, docking simulations in Molegro virtual docker (MVD) employed an optimized structure of the target, flexible ligand and protein side chains, and evaluation with the PLP (piecewise linear potential) scoring functions. The best binding poses and interaction energies were selected for further analysis.

Results and discussion

Synthesis

Tryphenyltin(IV) complex of 2-((E)-(4-hydroxy-3-((E)-((2-hydroxyethyl)imino)methyl)phenyl)diazenyl)benzoic acid (H₃L) was synthesized by refluxing triphenyltin(IV) chloride with the ligand (H₃L) in basic medium in anhydrous methanol solution maintaining a 1 : 1 metal–ligand ratio (Scheme 1). The synthesis and characterization of the ligand (H₃L) was reported in our earlier report,¹² but its X-ray crystal structure was not explored previously. The ligand may exist in tautomeric or zwitterionic forms, as shown in Scheme 1. Moreover, single crystals of both the ligand and the complex were obtained by slow evaporation of their alcoholic solution. The complex was obtained in good yield and was found to be soluble in almost all organic solvents.

FTIR spectroscopy

The free ligand (H₃L) showed asymmetric stretching frequencies, [ν_asy(COO)] at 1698 cm⁻¹,¹² while after coordination to Sn(IV) this band shifted to 1653 cm⁻¹ in the complex, as shown in Fig. S1 (SI). The shifting of the band to a lower frequency from the ligand to the complex indicated carboxylate coordination of the ligand to the tin atom.⁷ Moreover, the mode of carboxylate coordination was also confirmed by the difference between the ν_asy(COO) and ν_sym(COO) stretching frequencies, Δν[Δν = ν_asy(COO) − ν_sym(COO)]. It was observed that the difference between the ν_asy(COO) and ν_sym(COO) stretching frequencies was 263 cm⁻¹ which was greater than 200 cm⁻¹ indicating the monodentate mode of carboxylate coordination.^2,7 Again, phenolic–OH stretching frequency of the uncoordinated ligand, ν(OH) appeared at 3450 cm⁻¹,¹² while in the tin(IV) complex this band appeared at 3439 cm⁻¹. A decrease in the stretching frequency of the phenolic–OH group also provided information about the coordination of the phenolic oxygen atoms to the tin centre.³ Stretching frequency of the aliphatic –CH₂ group in the free ligand and the complex was observed at 2929 and 2927 cm⁻¹, respectively. The aromatic –CH stretching frequency of the complex was observed at 2677 cm⁻¹. Thus, a comparison between the IR spectra of the free ligand and the complex gave significant information regarding the coordination of the ligand to the tin centre through both the carboxylic oxygen (monodentate) and phenolic oxygen atom which was further confirmed by the single crystal X-ray structure of the complex.

Multinuclear (¹H, ¹³C, ¹¹⁹Sn) NMR spectroscopy

The ¹H, ¹³C and ¹¹⁹Sn NMR study of the complex was carried out in deuterated chloroform. All spectra are provided in the SI (Fig. S2–S5). A singlet was observed at 10.06 ppm in the ¹H NMR spectrum (Fig. S2 and S3 in the SI) of the complex, indicating an imine(C=N) proton. Signals for aromatic protons of both the ligand and the tin phenyl group appeared in the range of 7.26–7.70 ppm. A sharp singlet was observed at 3.07 ppm due to aliphatic hydroxy (OH) protons present in the ligand. Also, a multiplate was observed at 1.22–1.39 ppm signifying the appearance of aliphatic (–CH₂–CH₂–) protons in the ligand. The signal for phenolic proton was not observed as coordination of the ligand to the tin centre occurred through the phenolic oxygen besides the carboxylic oxygen atom. In the ¹³C NMR spectrum (Fig. S4, SI) the signals observed at 134.29, 130.11, 128.88 and 29.56 ppm may be assigned due to the presence of aromatic ring carbons. One sharp resonance peak at −128.48 ppm was observed in the ¹¹⁹Sn NMR spectrum of the complex (Fig. S5, SI) which indicated five coordinated trigonal bipyramidal geometry in the solution state.^3,5 Five coordinated trigonal bipyramidal geometry of the complex in the solid state established by single crystal X-ray crystallography retained its identity in solution state.

Crystal structure description and topological studies of H₃L

The molecular framework of the ligand H₃L is displayed in Fig. 1. The ligand adopts a trans configuration across the azo linkage, with an N=N bond length of 1.267(2) Å. The benzene ring A makes a dihedral angle of 1.60(15)° with the benzene ring B. All bond lengths and angles within the aromatic systems (Ring A and Ring B) fall within expected ranges, showing no unusual deviations. Notably, the C–O bond (connected to Ring A) measures 1.276(4) Å, characteristic of a C=O double bond. The exocyclic C–C bond (1.425(5) Å) connected to Ring A is intermediate between standard single and double bond lengths, indicative of a partial double bond characteristic. These measurements collectively suggest that the structure adopts an azo-enamine tautomeric form in the solid state (see Scheme 1), rather than an alternative tautomeric or zwitterionic structure.

Fig. 1 Molecular structure of H₃L.

To quantitatively validate the aromatic character of the rings and reinforce the tautomeric assignment, we employed the harmonic oscillator model of aromaticity (HOMA).^26,27 The HOMA index is calculated as follows:

Here, n represents the number of C–C bonds within the ring, α is a constant (257.7), and R_opt is the optimal bond length for ideal aromatic C–C bonds, taken as 1.388 Å. A HOMA value close to 1 suggests strong aromaticity, while values near 0 reflect non-aromatic character. Generally, a range of 0.900–0.990 indicates aromatic behavior, whereas 0.500–0.800 suggests partial aromaticity or deviation from aromaticity. The rings A and B yielded HOMA values of 0.667 and 0.970, respectively. These data show that ring B maintains strong aromatic character, while ring A exhibits diminished aromatic character – likely due to conjugation effects linked to the azo-enamine tautomerism. This observation is consistent with similar reported azo-enamine structures in the literature.^28,29 The crystal packing is stabilized by intramolecular N–H⋯O and O–H⋯N hydrogen bonds, along with intermolecular C–H⋯O and C–H⋯N interactions (Fig. 2a).

Fig. 2 (a) View of the crystal packing of H₃L; (b) the underlying net in the standard representation of the H-bonded molecular structure of H₃L. Blue spheres correspond to each molecule.

For topological analysis, the standard description³⁰ of a hydrogen-bonded crystal includes a simplification procedure, i.e., representation of the molecular network in terms of a graph-theory approach, taking into account hydrogen bonds between molecules. The simplification procedure consists of representing the molecule by its center of mass, keeping the connectivity of the molecule with its neighbors; all intermolecular contacts between a given pair of molecules transform to the same edge between the molecular centers of mass in the simplified net. Thus, this description characterizes the way molecules are hydrogen-bonded in crystals. The standard representation of the structure resulted in an underlying net of the 2C1 topological type (Fig. 2b). If we consider the structure in standard representation of the Coulomb or vdW-bonded molecular structures, where the centers of mass of the organic ligands are nodes of the underlying net in the representation, then there is a 14-c net of the tcg-x topological type (Fig. 3)

Fig. 3 The initial structure (top) and the underlying net in the standard representation of Coulomb or vdW-bonded molecular structures (bottom). Blue spheres correspond to molecules.

Crystal structure description and topological studies of [Ph₃Sn(IV)H₂L]_n (1)

Compound 1 features a polymeric R₃SnO₂ structural motif (Fig. 4), wherein the Sn center adopts a five-coordinate geometry. The coordination sphere comprises three equatorial phenyl groups and two axial oxygen atoms—one from a carboxylate and the other from a carbonyl group—each originating from two different ligand molecules. The coordination environment around the Sn center can be further evaluated using the Addison parameter (τ),³¹ which quantifies the degree of distortion between ideal square pyramidal (τ = 0) and trigonal bipyramidal (τ = 1) geometries. It is calculated using the formula τ = (β − α)/60°, where β and α (in °) represent the two largest coordination angles. In this case, a τ value of 0.76 indicates a geometry that is significantly closer to trigonal bipyramidal (TBP). Intramolecular N–H⋯O and intermolecular O–H⋯O hydrogen bonds, along with π–π stacking interactions contribute to crystal packing. The crystal packing consists of chains. Chains are connected to each other by hydrogen bonds and form layers which are connected to each other by van der Waals bonds (Fig. 5).

Fig. 4 Molecular structure of 1 showing the TBP structure around tin atoms in the polymeric chain of [Ph₃Sn(IV)H₂L]_n.

Fig. 5 (a) Initial chair-like structure; (b) chains (highlighted in color for clarity) forming layers through hydrogen bonding; (c) layers (highlighted in color for clarity) held together by van der Waals interactions.

From a topological viewpoint, the framework can be simplified to its underlying net in the so-called standard representation of the valence-bonded MOFs.³² All metal atoms and the centers of mass of the organic ligands are nodes of the underlying net in the representation. This is a 2-c net of the 2C1 topological type (Fig. S6 in the SI). If we consider the structure in standard representation of H-bonded or specific-bonded molecular structures and thus consider the packing of chains in the structure, then there is a 4-c net of the sql topological type (Fig. 6). Overall, both systems display notable topological transitions: while H₃L transforms from a simple hydrogen-bonded 2C1 net to a more complex 14-connected tcg-x network upon considering all intermolecular interactions, the Sn(IV) polymer (1) similarly evolves from a basic 2-connected 2C1 net in its valence-bonded representation to a 4-connected sql-type net when extended hydrogen-bonding and van der Waals interactions are included. A comparative summary of the structural features and supramolecular synthons observed in [Ph₃Sn(IV)H₂L] (1) and related tin(IV) complexes is provided in Table S1 (see the SI). This table shows our findings within the broader landscape of organotin crystal engineering, illuminating both common features and key advances achieved in this work.

Fig. 6 The underlying net in the standard representation of the H-bonded MOFs before removing 1-c and 2-c nodes (top) and after removing 1-c nodes (bottom). Pink spheres correspond to Sn atoms, blue spheres correspond to phenyl ligands, and green spheres correspond to H₂L ligands.

The topological divergence between H₃L and 1 has direct implications for functional-group accessibility. In H₃L, the hydrogen-bonded 2C1 → tcg-x networks preserve conformational flexibility and expose key carboxyl, phenolic, and azo-enamine moieties for enzyme interactions. In contrast, incorporation of these donor sites into the rigid sql-type framework in 1 restricts their accessibility and increases structural rigidity, thereby providing a clear structural basis for the diminished α-glucosidase inhibitory activity (vide infra).

Antidiabetic activity screening and dose-effect analysis

The results of the screening of the antidiabetic activities of the two compounds (H₃L and 1) in four different solvents are shown in Table 2 and Fig. S7 (in the SI). The H₃L compound exhibited significant antidiabetic activity as revealed by inhibitions of α-glucosidase activity in chloroform (43.7%) and DMSO (79.36%) as compared to the standard drug, acarbose (45.06%). The samples dissolved in ethyl acetate and methanol showed lower inhibition activities than chloroform, DMSO and acarbose. The compound 1 in different solvents showed very low enzyme inhibition activities as compared to H₃L and acarbose.

Table 2 Screening test for α-glucosidase inhibition of the two compounds in different solvents in the presence of acarbose (ACB)

Sample ID	Solvents	α-Glucosidase inhibition (%)	SD (±)
H3LC	Chloroform	43.70	9.17
H3LB	Ethyl acetate	15.14	3.6
H3LA	Methanol	16.12	6.58
H3LD	DMSO	79.36	6.85
ACB	Water	45.06	0.92
1C	Chloroform	2.24	1.53
1B	Ethyl acetate	1.61	1.02
1A	Methanol	7.23	0.85
1D	DMSO	0.00	0.00

---

Solvent choice in α-glucosidase inhibition assays critically affects both the measured potency and mechanism of inhibition.³³ Polar aprotic solvents such as DMSO are widely used for their exceptional solubilizing ability, ensuring that small molecule inhibitors remain fully available and conformationally flexible for enzymatic binding. Enhanced solubility in DMSO likely enables H₃L to retain a conformation that exposes donor functionalities (hydroxyl, carboxyl, and azo groups), crucial for forming H-bonds and π-stacking interactions with active site residues, as confirmed by parallel molecular docking analyses. In contrast, decreased solubility and possible aggregation in less polar solvents like methanol or chloroform may reduce the effective free inhibitor concentration, yielding lower observed inhibition despite preserved intrinsic activity.

While organic solvents can modulate enzyme conformation or stability, especially at high concentrations, the literature and our solvent control experiments confirm that α-glucosidase retains its native, catalytically competent structure at or below 2% (v/v) DMSO or similar solvents.^24,25 Under these conditions, there is minimal disruption of hydrogen bonding networks and the enzyme's electronic environment. Therefore, inhibition readouts reliably reflect ligand–enzyme interactions rather than solvent artifacts. This is further supported by many published studies noting solvent-dependent variation in inhibitor activity,³³ which can differ by orders of magnitude due to differences in solubility, aggregation, or protein–ligand–solvent microenvironments. The robustness of these findings is ensured by the inclusion of parallel solvent controls and analysis across multiple solvent systems.

The results of the dose-effect relationship analysis are shown in Table 3 and Fig. S8 (see the SI). H₃L in chloroform and DMSO exhibited IC₅₀ values of 25.29 µg mL⁻¹ and 26.79 µg mL⁻¹ different from that of the standard drug acarbose (11.26 µg mL⁻¹). Although there was higher α-glucosidase inhibition activity of DMSO during the screening test, the dose-effect analysis revealed that there is no significant variation in the IC₅₀ value between the two solvents.

Table 3 IC₅₀ values of H₃L in chloroform and DMSO and acarbose (ACB)

Samples	IC50 (µg mL^−1)	SD (±)
H3LC	25.29	14.02
H3LD	26.79	18.39
ACB (standard)	11.26	0.45

---

Structure and topology influence on activity

The pronounced α-glucosidase inhibition exhibited by the free ligand H₃L, in contrast to its triphenyltin(IV) complex, can be rationalized by multiple structural factors. In H₃L, the carboxylic acid group remains uncoordinated and thus fully accessible, allowing it to participate in productive hydrogen bonding interactions with the enzyme active site. Although H₃L exists as an azo-enamine tautomer in the solid state, where a phenolic OH is not present, the tautomeric equilibrium in solution may allow transient presentation of donor functionalities, further supporting effective enzyme binding. Conversely, in the Sn(IV) complex, both the carboxylate and enol/phenol oxygen atoms are coordinated to the tin center, significantly reducing their ability to interact with the enzyme. Literature clearly supports that the presence and accessibility of these groups are crucial for potent α-glucosidase inhibition and antidiabetic activity.^16,17 In addition, the triphenyltin(IV) substituent introduces considerable steric bulk and rigidity, likely hindering the molecular flexibility and dynamic fit required for optimal inhibition. These results further suggest that the molecular topology of H₃L, characterized by its accessible hydrogen-bonding network and flexible supramolecular arrangement, plays a pivotal role in its strong antidiabetic activity. In contrast, the rigid, extended sql-type topology and more crowded environment of the polymeric Sn(IV) complex likely restrict essential molecular interactions with the enzyme target, resulting in negligible inhibition. Collectively, these observations highlight a broader design principle relevant to α-glucosidase inhibitor development: molecular scaffolds that preserve conformational flexibility and accessible polar hydrogen-bond donors/acceptors (as in the discrete, hydrogen-bonded networks of H₃L) facilitate effective enzyme recognition, whereas coordination-induced rigidification and donor sequestration (as in the rigid sql-type polymeric framework of complex 1) markedly attenuate inhibitory potency. This structure–activity insight aligns with contemporary studies emphasising the role of topological features and functional-group accessibility in biological performance.^34,35

It is noteworthy that under the mild, near-neutral assay conditions employed in this study (phosphate buffer, pH ≈ 6.9, 37 °C ± 2 °C), the polymeric triphenyltin(IV) complex 1 exhibits negligible α-glucosidase inhibition in all tested solvents, in contrast to the pronounced activity of the free ligand H₃L. This behaviour, together with the well-documented structural robustness of triphenyltin(IV) carboxylates under short-term biological assay conditions,^1,8,36 suggests that significant leaching of tin ions is unlikely in the present system. Decomposition of organotin compounds is typically associated with prolonged environmental exposure involving air, UV radiation, microorganisms, or soil matrices, rather than buffered in vitro enzymatic assays.³⁷ Accordingly, the biologically active species identified here is the Sn-free ligand H₃L.

Molecular docking study

Molecular docking simulations were performed to assess the binding interactions of the H₃L ligand and the standard inhibitor acarbose (ACB) with α-glucosidase (PDB ID: 5ZCB, Fig. S9 in the SI). The results revealed distinct differences in their interaction profiles, which may account for variations in their observed biological activities. The H₃L ligand exhibited a notably better docking performance (MolDock score: –87.92 kcal mol⁻¹, Re-rank score: –79.84 kcal mol⁻¹, interaction energy: –112.71 kcal mol⁻¹ and binding affinity: –33.0 kcal mol⁻¹), all significantly more favorable than those recorded for ACB (–49.41, –54.81, –96.45, and –24.04 kcal mol⁻¹, respectively; Table S2 in the SI). This suggests that H₃L displayed a more negative overall binding energy, indicating a thermodynamically stronger and more stable interaction with the enzyme's active site. The binding pose of the H₃L ligand (Fig. 7) confirmed its appropriate orientation within the catalytic pocket of α-glucosidase. The 3D binding modes and corresponding 2D interaction diagram (Fig. 7; see also Fig. S10–S12 in the SI), consistent with representations in related enzyme-inhibitor studies,^38,39 illustrate a balanced profile of van der Waals contacts, hydrogen bond interactions, and electrostatic complementarity, even though the number of discrete hydrogen bonds was modest. This suggests that the ligand H₃L achieves stability within the active site through a combination of favorable steric fit and dispersed polar and non-polar interactions, rather than relying on a single strong interaction.

Fig. 7 Molecular docking of the H₃L ligand at the active site of 5ZCB (α-glucosidase): (a) overall 3D ribbon representation of 5ZCB showing ligand binding; (b) enlarged view of the docked ligand at the binding pocket highlighting multiple hydrogen-bond interactions with key residues such as Glu141, Ser142, Ser145, and Ser224, with interaction distances ranging from 2.93 to 3.28 Å; (c) 2D interaction map illustrating hydrogen bonds (dashed arrows) and hydrophobic contacts (arcs/lines); distances are given in Å.

In contrast, the interaction analysis of acarbose (ACB) indicated limited engagement with the enzyme, forming only one hydrogen bond (with Lys395) and two hydrophobic interactions (with Ile143 and Tyr221) (Fig. S13–S16 in the SI). The restricted number and nature of these interactions suggest a less stable and less specific binding mode for ACB under the same docking conditions. To complement the docking discussion, all specific amino-acid interactions for H₃L and ACB have been tabulated in Table S3 (see the SI).

It may be noted that molecular docking provides a static, qualitative prediction of possible binding modes and is influenced by scoring functions, grid parameters, and the absence of explicit solvent and entropy effects. As a result, it may overemphasize interactions from certain fragments while underrepresenting dynamic conformational changes or solvent-modulated behavior observed experimentally. This helps explain why H₃L exhibits strong in vitro α-glucosidase inhibition, particularly in DMSO, likely due to improved solubility and better exposure of key polar moieties, even though docking highlights one dominant fragment. These limitations emphasize that docking should be interpreted cautiously as a supportive tool, with experimental IC₅₀ values serving as the primary measure of activity.

Taken together, these results provide a strong computational rationale for the experimentally observed superior inhibitory activity of ligand H₃L compared to acarbose. While ACB is an established α-glucosidase inhibitor, the docking data indicate that the H₃L ligand may offer enhanced binding affinity and better spatial accommodation within the active site, supporting its potential as a more effective or complementary therapeutic candidate. Computational approaches such as molecular docking are increasingly recognized as valuable tools for mechanistic interpretation and the rational design of enzyme inhibitors. The present findings are in line with recent literature,^40,41 where docking-based analyses have played a crucial role in elucidating ligand–enzyme interactions and in guiding the rational development of new bioactive frameworks. Further experimental validation, including enzyme inhibition kinetics and quantitative structure–activity relationship (QSAR) studies, would strengthen and extend these observations.

Conclusions

In summary, we have synthesized and structurally characterized an azo-enamine-type benzene hydroxocarboxylic acid ligand (H₃L) and its polymeric triphenyltin(IV) complex (1). The crystal structure of H₃L confirmed the presence of a stabilized azo-enamine tautomer, while the Sn(IV) center in complex 1 adopted a trigonal bipyramidal geometry. Topological analyses provided further insight into the extended packing, revealing a 2C1 net in H₃L and sql-type connectivity in the polymeric complex, emphasizing the role of hydrogen bonding and coordination in the supramolecular architecture. Biologically, H₃L demonstrated significant α-glucosidase inhibition and superior docking performance compared to acarbose, whereas complex 1 exhibited negligible enzyme inhibition. These findings establish H₃L as a structurally and biologically promising scaffold for antidiabetic drug development.

Although this study did not incorporate quantitative topology descriptors or graph-theoretical QSAR methods, qualitative correlations between molecular structure, topological accessibility, and biological function were established. Incorporation of formal topology-based models and descriptor analysis in future studies would enable a deeper quantitative framework for predicting and optimizing antidiabetic activity within this class of compounds. Extension to quantitative QSAR or topology-based descriptors (e.g., Zagreb indices, Wiener connectivity) could enable predictive ranking, virtual screening, and systematic optimization of inhibitory potency (IC₅₀ prediction) beyond the qualitative insights established here.

Author contributions

Anisha Das: methodology, investigation, writing – original draft, formal analysis. Debabrata Nama: methodology, formal analysis. Subhadip Roy: software, investigation, formal analysis, writing – original draft. Chinmoy Majumder: formal analysis. S. Sureshkumar Singh: formal analysis. Alka Sandham: formal analysis. Salam Pradeep Singh: software, formal analysis. Manojit Roy: supervision, conceptualization, writing – review & editing. Tarun Kumar Misra: supervision, conceptualization, writing – review & editing.

Conflicts of interest

There are no conflicts to declare.

Data availability

The authors confirm that the data supporting the findings of this study are provided within the article and its supplementary information (SI). Supplementary information: experimental procedures including the full molecular docking methodology and the IR spectrum of the triphenyltin(IV) complex [Ph₃Sn(IV)H₂L] (1) are provided in Fig. S1, and its multinuclear NMR spectra (¹H, expanded ¹H, ¹³C, and ¹¹⁹Sn) are presented in Fig. S2–S5. Fig. S6 shows the topological representation of the valence-bonded framework of 1. α-Glucosidase inhibition profiles and IC₅₀ plots of H₃L and acarbose (ACB) are provided in Fig. S7 and S8. Fig. S9 illustrates the binding cavity of α-glucosidase (PDB ID: 5ZCB). Fig. S10–S12 depict the molecular, steric, energy-map, and electrostatic interaction analyses of H₃L at the enzyme active site, while Fig. S13–S16 present the corresponding interaction analyses for ACB. Tables S1–S3 summarize crystallographic features and supramolecular synthons, docking scores and energy components, and detailed residue-level interaction parameters for both H₃L and ACB. See DOI: https://doi.org/10.1039/d5nj03575a.

Raw data are available from the corresponding author upon reasonable request.

2483977 (1) and CCDC 2483978 (H₃L) contain the supplementary crystallographic data for this paper.^42a,b

The three-dimensional (3D) structure of α-glucosidase (PDB ID: 5ZCB) was retrieved from the Protein Data Bank.⁴³

Acknowledgements

The authors gratefully acknowledge SAIF, IIT Madras for the X-ray crystallographic study, IISc. Bangalore for the ¹H, ¹³C NMR facility and IIT Delhi for the ¹¹⁹Sn NMR spectral studies. The authors also wish to acknowledge the Department of Botany, Manipur University, for the antidiabetic study. Special thanks are due to the Department of Chemistry, NIT Agartala and authorities of The ICFAI University Tripura for providing essential research facilities.

References

1. E. R. T. Tiekink, Appl. Organomet. Chem., 1991, 5, 1.
2. M. Roy, S. S. Devi, S. Roy, C. B. Singh and K. S. Singh, Inorg. Chim. Acta, 2015, 426, 89.
3. P. Debnath, P. Debnath, S. Roy, M. B. Devi, M. M. Devi, K. Sarangthem, S. S. Singh, M. Roy, A. S. Novikov and T. K. Misra, Inorg. Chim. Acta, 2024, 559, 121805.
4. T. S. Basu Baul, D. Dutta, A. Duthie, N. Guchhait, G. M. R. Bruno, M. F. C. Guedes da Silva, R. B. Mokhamatam, N. Raviprakash and S. K. Manna, J. Inorg. Biochem., 2017, 166, 34.
5. M. Roy, S. Roy, K. S. Singh, J. Kalita and S. S. Singh, New J. Chem., 2016, 40, 1471.
6. P. Debnath, P. Debnath, T. S. Devi, S. S. Singh, A. Bhattacharya, K. S. Singh, M. Roy and T. K. Misra, J. Coord. Chem., 2022, 75, 3015.
7. M. Roy, S. Roy, K. S. Singh, J. Kalita and S. S. Singh, Inorg. Chim. Acta, 2016, 439, 164.
8. M. K. Amir, S. Khan, Z. Rehman, A. Shah and I. S. Butler, Inorg. Chim. Acta, 2014, 423, 14–25.
9. M. Roy, S. Roy, N. M. Devi, C. B. Singh and K. S. Singh, J. Mol. Struct., 2016, 1119, 64.
10. K. S. Singh, M. Roy, S. Roy, B. Ghosh, N. M. Devi, C. B. Singh and L. K. Mun, J. Coord. Chem., 2017, 70, 361.
11. T. S. Basu Baul, K. S. Singh, A. Lyčka, A. Linden, X. Song, A. Zapata and G. Eng, Appl. Organomet. Chem., 2006, 20, 788.
12. P. Debnath, P. Debnath, M. Roy, L. Sieron, W. Maniukiewicz, T. Aktar, D. Maiti, A. S. Novikov and T. K. Misra, Crystals, 2022, 12, 1582.
13. I. Z. Gundhla, R. S. Walmsley, V. Ugirinema, N. O. Mnonopi, E. Hosten, R. Betz, C. L. Frost and Z. R. Tshentu, J. Inorg. Biochem., 2015, 145, 11–18.
14. M. E. López-Viseras, B. Fernández, S. Hilfiker, C. S. González, J. L. González, A. J. Calahorro, E. Colacio and A. Rodríguez-Diéguez, J. Inorg. Biochem., 2014, 131, 64–67.
15. J. Vančo, J. Marek, Z. Trávníček, E. Račanská, J. Muselík and O. G. Švajlenová, J. Inorg. Biochem., 2008, 102, 595–605.
16. A. Aleixandre, J. V. Gil, J. Sineiro and C. M. Rosell, Food Chem., 2022, 372, 131231.
17. H. X. Nguyen, T. M. Le, T. H. Le, T. Q. Truong, B. Q. H. Nguyen, P. T. Nguyen, K. M. Le, T. N. Van Do, M. T. T. Nguyen, M. H. Nguyen and N. T. Nguyen, RSC Adv., 2025, 15, 26444–26454.
18. L. J. Bourhis, O. V. Dolomanov, R. J. Gildea, J. A. K. Howard and H. Puschmann, Acta Crystallogr., Sect. A: Found. Adv., 2015, 71, 59–75.
19. G. M. Sheldrick, Acta Crystallogr., Sect. C: Struct, Chem., 2015, 71, 3–8.
20. O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. A. K. Howard and H. Puschmann, J. Appl. Crystallogr., 2009, 42, 339–341.
21. V. A. Blatov, A. P. Shevchenko and D. M. Proserpio, Cryst. Growth Des., 2014, 14, 3576–3586.
22. M. O’Keeffe, M. A. Peskov, S. J. Ramsden and O. M. Yaghi, Acc. Chem. Res., 2008, 41, 1782–1789.
23. D. Kumar, H. Kumar, J. R. Vedasiromoni and B. C. Pal, Phytochem. Anal., 2012, 23, 421–425.
24. D. Kumar, N. Gupta, R. Ghosh, R. H. Gaonkar and B. C. Pal, J. Funct. Foods, 2013, 5, 211–218.
25. D. Kumar, H. Kumar, J. R. Vedasiromoni and B. C. Pal, Phytochem. Anal., 2012, 23, 421–425.
26. J. Kruszewski and T. M. Krygowski, Tetrahedron Lett., 1972, 13, 3839–3842.
27. Ç. Albayrak, G. Kaştaş, M. Odabaşoğlu and O. Büyükgüngör, J. Mol. Struct., 2011, 1000, 162–170.
28. A. Gözel, M. Kose, D. Karakaş, H. Atabey, V. McKee and M. Kurtoglu, J. Mol. Struct., 2014, 1074, 449–456.
29. M. Sarigul, A. Sari, M. Kose, V. McKee, M. Elmastas, I. Demirtas and M. Kurtoglu, Inorg. Chim. Acta, 2016, 444, 166–175.
30. E. V. Alexandrov, V. A. Blatov, A. V. Kochetkov and D. M. Proserpio, CrystEngComm, 2011, 13, 3947–3958.
31. A. W. Addison, T. N. Rao, J. Reedijk, J. van Rijn and G. C. Verschoor, J. Chem. Soc., Dalton Trans., 1984, 1349–1356.
32. E. V. Alexandrov, V. A. Blatov, A. V. Kochetkov and D. M. Proserpio, CrystEngComm, 2011, 13, 3947–3958.
33. Y.-T. Lin, H.-R. Lin, C.-S. Yang, C.-C. Liaw, P.-J. Sung, Y.-H. Kuo, M.-J. Cheng and J.-J. Chen, Antioxidants, 2022, 11, 320.
34. C. Calabuig, G. M. Antón-Fos, J. Gálvez and R. García-Doménech, Int. J. Pharm., 2004, 278, 111–118.
35. M. Popescu, A. Velea, C. Mihai and S. Tivadar, Dig. J. Nanomater. Bios., 2010, 5, 629–633.
36. S. Tikhonov, P. Ostroverkhov, N. Suvorov, A. Mironov, Y. Efimova, A. Plutinskaya, A. Pankratov, A. Ignatova, A. Feofanov, E. Diachkova, Y. Vasil’ev and M. Grin, Int. J. Mol. Sci., 2021, 22, 13563.
37. A. Olushola Sunday, B. Abdullahi Alafara and O. Godwin Oladele, Chem. Speciation Bioavailability, 2012, 24, 216–226.
38. G. Biswas, M. Afzal, U. C. Halder and N. Sepay, J. Mol. Struct., 2025, 1322, 140358.
39. T. Huang, J. Sun, Q. Wang, J. Gao and Y. Liu, Molecules, 2015, 20, 16221–16234.
40. D. Majumdar, A. Dubey, A. Tufail, D. Sutradhar and S. Roy, Heliyon, 2023, 9, e16057.
41. D. Majumdar, A. Chatterjee, M. Feizi-Dehnayebi, N. S. Kiran, B. Tuzun and D. Mishra, Heliyon, 2024, 10, e35591.
42. (a) CCDC 2483978: Experimental Crystal Structure Determination, 2026 DOI:10.5517/ccdc.csd.cc2pcsb6; (b) CCDC 2483977: Experimental Crystal Structure Determination, 2026 DOI:10.5517/ccdc.csd.cc2pcs95.
43. W. Auiewiriyanukul, W. Saburi, K. Kato, M. Yao and H. Mori, Function and structure of GH13_31 alpha-glucosidase with high alpha-(1→4)-glucosidic linkage specificity and transglucosylation activity, FEBS Lett., 2018, 592, 2268–2281 DOI:10.1002/1873-3468.13126.

---

Footnote

† Dedicated to Professor R. N. Dutta Purkayastha (Tripura University) on the occasion of his retirement and for his contribution to inorganic chemistry.

---