Iodinated Salicylhydrazone Derivatives as Potent α-Glucosidase Inhibitors: Synthesis, Enzymatic Activity, Molecular Modeling, and ADMET Profiling

Abstract

Type 2 diabetes mellitus (T2DM) demands safer and more effective therapies to control postprandial hyperglycemia. Here, we report the synthesis and in vitro evaluation of ten salicylic acid-derived Schiff base derivatives (4a–4j) as α-glucosidase inhibitors. Compounds 4e, 4g, 4i, and 4j exhibited potent enzyme inhibition, with IC₅₀ values ranging from 14.86 to 18.05 µM—substantially better than acarbose (IC₅₀ = 45.78 µM). Molecular docking and 500 ns molecular dynamics simulations revealed stable enzyme–ligand complexes driven by π–π stacking, halogen bonding, and hydrophobic interactions. Density Functional Theory (DFT) calculations and molecular electrostatic potential (MEP) maps highlighted key electronic factors, while ADMET analysis confirmed favorable drug-like properties and reduced nephrotoxicity. Structure–activity relationship (SAR) analysis emphasized the importance of halogenation and aromaticity in enhancing bioactivity.

1. Introduction

The increasing prevalence of type 2 diabetes mellitus (T2DM) represents a critical global health challenge, with recent reports from the International Diabetes Federation (IDF) estimating that one in ten adults worldwide will have diabetes by 2030 [1,2,3]. Characterized by chronic hyperglycemia due to insulin resistance or impaired insulin secretion, T2DM is a major risk factor for severe complications, including retinopathy, nephropathy, cardiovascular diseases, and neuropathy [2,4]. According to the World Health Organization (WHO), over 830 million people worldwide have been affected by diabetes as of 2022, with numbers projected to rise significantly [1]. A key strategy in T2DM management is controlling postprandial blood glucose levels, which mitigates long-term complications [4,5]. α-Glucosidase inhibitors (AGIs) have emerged as effective therapeutic agents, as they delay the digestion and absorption of dietary carbohydrates, thereby preventing rapid glucose spikes [6,7,8,9].

Despite their clinical utility, conventional AGIs such as acarbose, voglibose, and miglitol are often associated with gastrointestinal side effects, including flatulence, diarrhea, and abdominal discomfort due to undigested carbohydrates fermenting in the colon. Studies such as those by Hanefeld 1998 [10] and Kageyama 1997 [11] have extensively reviewed these limitations, emphasizing the need for novel AGIs with improved safety profiles. These challenges have driven efforts to develop synthetic derivatives and bioactive scaffolds with improved selectivity and more favorable therapeutic profiles, aiming to balance efficacy with a reduction in adverse effects [12,13].

Among various structural classes, salicylic acid and its derivatives hold significant medicinal importance. As the precursor to aspirin, salicylic acid exhibits unique reactivity due to its ortho-positioned hydroxyl and carboxylic acid groups, which facilitate intramolecular hydrogen bonding and electronic stabilization [14,15]. Recent studies, such as those by Chen 2019 [16] and Aminu 2022 [17], have demonstrated that salicylic acid derivatives exhibit potent α-glucosidase inhibitory activity, highlighting their potential as promising AGI candidates.

Given their versatility, incorporating salicylic acid into hybrid scaffolds has been an effective strategy for enhancing bioactivity. Among various structural classes, Schiff bases have garnered attention for their broad spectrum of biological activities, including antimicrobial, antioxidant, anticancer, anti-inflammatory, and enzyme inhibitory properties [18,19]. Studies, such as those by Afzal 2021 [20] and Sarfaraz 2024 [21], have reported the relevance of Schiff bases in enzyme inhibition, particularly in targeting metabolic enzymes. Schiff bases, formed via the condensation of primary amines with aldehydes or ketones, feature a characteristic imine (–C=N–) linkage, enabling interactions with biological macromolecules, including enzymes and receptors. Their structural simplicity and tunable diversity make Schiff bases attractive candidates for drug discovery [22].

Previous work on Schiff bases derived from salicylic acid scaffolds (Figure 1, top) has demonstrated their biological potential, including antibacterial and enzyme inhibition activities. For instance, compound A [23,24] and compound B [25] exhibit potent antibacterial activity against Gram-positive and Gram-negative bacteria, with MIC values as low as 1.8 µg/mL. Similarly, compound C [26] shows significant antitubercular activity (MIC = 0.39 µg/mL), while compound D [27] effectively inhibits ecKAS III, a key enzyme in bacterial fatty acid biosynthesis. Additionally, compound E [28] targets DNA gyrase and topoisomerase IV, demonstrating dual inhibitory effects. These reports show the versatility of Schiff base derivatives of salicylic acid in medicinal chemistry.

Building on this foundation, Khan 2024 [29] (Figure 1F) demonstrated the potential of Schiff base derivatives of 3,4-dihydroxyphenylacetic acid as α-glucosidase inhibitors. Several of their compounds exhibited superior activity compared to standard drugs like acarbose, underscoring the therapeutic relevance of Schiff bases in T2DM management. These findings support the rationale for further exploration of Schiff base linkages in structurally related scaffolds, such as salicylic acid derivatives, to identify novel AGIs with enhanced pharmacological profiles [30].

Considering the structural diversity and promising bioactivity of Schiff bases, we aimed to design, synthesize, and evaluate a series of salicylic acid-based Schiff base derivatives as α-glucosidase inhibitors. The salicylic acid core was selected due to its ortho-hydroxyl and carboxyl functional groups, which promote intramolecular hydrogen bonding and electronic stabilization. Additionally, iodine substitution at the 5-position of the aromatic ring was introduced to enhance polarizability and facilitate halogen bonding, which can strengthen enzyme interactions. The Schiff base derivatives were synthesized via condensation reactions, where 2-hydroxy-5-iodobenzoic acid hydrazide was coupled with various substituted benzaldehydes. This hydrazone linkage offers hydrogen-bonding and electron-donating capabilities, which are critical for enzyme binding and selectivity.

To elucidate enzyme–ligand interactions, molecular docking and molecular dynamics simulations were performed, identifying key binding interactions at the α-glucosidase active site. Additionally, Density Functional Theory (DFT) calculations provided insights into molecular orbitals (HOMO-LUMO), charge distribution, and electronic properties, supporting SAR-driven optimization. Finally, ADMET analysis was conducted to assess the pharmacokinetic and toxicity profiles of the designed inhibitors. By integrating experimental and computational methodologies, this study seeks to develop salicylic acid-based Schiff base derivatives as potent α-glucosidase inhibitors with potential applications in T2DM therapy.

2. Materials and Methods

2.1. General Method

All chemicals and reagents were purchased from Sigma-Aldrich (St. Louis, MO, USA) or TCI Chemicals (Portland, OR, USA) and were used without further purification, unless otherwise stated. Solvents, including ethanol (99.5%), were of analytical grade. Reactions were conducted under a dry nitrogen atmosphere using oven-dried glassware. Heating reactions were performed in a paraffin oil bath to ensure uniform temperature control.

Thin-layer chromatography (TLC) on pre-coated silica gel 60 F254 plates (0.25 mm thickness, E. Merck, Darmstadt, Germany) was used to monitor reaction progress and to determine appropriate solvent systems for purification by column chromatography. Spots were visualized under UV light (254 nm and 365 nm) or by staining with a ninhydrin solution. Column chromatography was performed using silica gel (100–200 mesh or 230–400 mesh), with eluent selection guided by TLC mobility.

2.2. Chemistry

The structures of all newly synthesized compounds were confirmed by ¹H NMR, ¹³C NMR, and HRMS analysis. Copies of the corresponding spectra are available in the Supplementary Materials (Section S8).

2.2.1. 5-Iodo-2-Salicylic Ester (2)

5-Iodo-salicylic acid (2 g, 7.57 mmol) was dissolved in ethanol (40 mL) and stirred at room temperature for 10 min. To this solution, 1 M H₂SO₄ (0.56 mL) was added dropwise, and the reaction mixture was refluxed at 90 °C for 18 h. The mixture was then diluted with ethyl acetate (40 mL) and washed successively with aqueous sodium bicarbonate. The organic layer was dried over anhydrous sodium sulfate, concentrated under reduced pressure, and purified by column chromatography (silica gel, 20% ethyl acetate in petroleum ether) to yield 5-iodo-2-salicylic ester as a white solid (1.64 g, 74.14%). ¹H NMR (600 MHz, DMSO-d₆): δ 10.56 (s, 1H), 7.99 (d, J = 2.3 Hz, 1H), 7.77 (dd, J = 8.7, 2.3 Hz, 1H), 6.82 (d, J = 8.7 Hz, 1H), 4.34 (q, J = 7.1 Hz, 2H), 1.33 (t, J = 7.1 Hz, 3H). ¹³C NMR (151 MHz, DMSO-d₆): δ 167.72, 159.99, 143.86, 138.21, 120.59, 116.44, 81.35, 62.08, 14.42. HRMS (TOF) (m/z): [M + H]⁺ calculated for C₉H₉IO₃ 291.9596; found 293.9572.

2.2.2. 5-Iodo-2-Hydroxybenzo-Hydrazide (3)

A solution of 5-iodo-2-salicylic ester (10 mmol) and hydrazine hydrate (20 mmol) in ethanol (50 mL) was heated under reflux for 15 h. The mixture was concentrated and poured onto crushed ice. The resulting solid was filtered, washed with water, dried, and recrystallized from ethanol to yield 5-iodo-2-hydroxybenzo-hydrazide as colorless crystals. ¹H NMR (600 MHz, DMSO-d₆): δ 10.07 (s, 1H), 8.12 (d, J = 2.2 Hz, 1H), 7.65 (dd, J = 8.6, 2.2 Hz, 1H), 6.75 (d, J = 8.6 Hz, 1H). ¹³C NMR (151 MHz, DMSO-d₆): δ 166.62, 159.39, 141.79, 135.99, 120.39, 116.44, 80.98. HRMS (TOF) (m/z): [M + H]⁺ calculated for C₇H₇IN₂O₂ 277.9552; found 277.9557.

2.2.3. General Procedure for Synthesis

5-Iodo-2-hydroxybenzo-hydrazide (10 mmol) and the respective substituted benzaldehydes (10 mmol) were dissolved in ethanol (10 mL) containing a few drops of acetic acid. The reaction mixture was refluxed for 5–9 h. Upon cooling to room temperature, the precipitate was filtered and recrystallized from ethanol, yielding the corresponding derivatives (4a–4j) as crystalline solids [29,31].

(E)-2-hydroxy-5-iodo-N′-(thiazol-2-ylmethylene)benzohydrazide (4a). White solid (68%); ¹H NMR (600 MHz, DMSO-d₆): δ 12.06 (s, 1H), 11.60 (s, 1H), 8.66 (s, 1H), 8.08 (d, J = 2.3 Hz, 1H), 8.00 (d, J = 3.2 Hz, 1H), 7.89 (d, J = 3.2 Hz, 1H), 7.73 (dd, J = 8.6, 2.3 Hz, 1H), 6.84 (d, J = 8.6 Hz, 1H). ¹³C NMR (151 MHz, DMSO-d₆): δ 164.40, 163.48, 158.33, 144.64, 143.41, 142.28, 137.32, 122.85, 120.29, 120.08, 81.45. HRMS (TOF) (m/z): [M + H]⁺ calculated for C₁₁H₈IN₃O₂S 372.9382; found 372.9389.

(E)-2-hydroxy-5-iodo-N′-(pyridin-3-ylmethylene)benzohydrazide (4b). White solid (78%); ¹H NMR (600 MHz, DMSO-d₆): δ 11.96 (s, 1H), 11.80 (s, 1H), 8.88 (d, J = 2.2 Hz, 1H), 8.64 (dd, J = 4.8, 1.7 Hz, 1H), 8.50 (s, 1H), 8.20–8.12 (m, 2H), 7.73 (dd, J = 8.7, 2.2 Hz, 1H), 7.51 (dd, J = 8.0, 4.8 Hz, 1H), 6.84 (d, J = 8.7 Hz, 1H). ¹³C NMR (151 MHz, DMSO-d₆): δ 163.60, 158.68, 151.45, 149.40, 146.70, 142.24, 137.21, 134.09, 130.44, 124.55, 120.36, 119.64, 81.42. HRMS (TOF) (m/z): [M + H]⁺ calculated for C₁₃H₁₀IN₃O₂ 366.9818; found 366.9836.

(E)-2-hydroxy-N′-(4-hydroxy-3-nitrobenzylidene)-5-iodobenzohydrazide (4c). Yellow solid (75%); ¹H NMR (600 MHz, DMSO-d₆): δ 11.86 (s, 2H), 11.59 (s, 1H), 8.40 (s, 1H), 8.22 (d, J = 2.3 Hz, 1H), 8.15 (d, J = 2.3 Hz, 1H), 7.99–7.90 (m, 1H), 7.76–7.67 (m, 1H), 7.23 (d, J = 8.5 Hz, 1H), 6.83 (d, J = 8.5 Hz, 1H). ¹³C NMR (151 MHz, DMSO-d₆): δ 163.53, 158.78, 154.04, 147.41, 142.19, 137.55, 137.10, 133.49, 125.95, 124.80, 120.35, 120.23, 119.48, 81.39. HRMS (TOF) (m/z): [M + H]⁺ calculated for C₁₄H₁₀IN₃O₅ 426.9665; found 426.9654.

(E)-2-hydroxy-5-iodo-N′-(4-nitrobenzylidene)benzohydrazide (4d). Yellow solid (78%); ¹H NMR (600 MHz, DMSO-d₆): δ 12.04 (s, 1H), 11.73 (s, 1H), 8.54 (s, 1H), 8.31 (d, J = 8.6 Hz, 2H), 8.13 (d, J = 2.3 Hz, 1H), 8.00 (d, J = 8.5 Hz, 2H), 7.73 (dd, J = 8.6, 2.3 Hz, 1H), 6.84 (d, J = 8.7 Hz, 1H). ¹³C NMR (151 MHz, DMSO-d₆): δ 163.64, 158.50, 148.47, 146.80, 142.29, 140.80, 137.38, 128.66, 124.58, 120.32, 119.83, 81.47. HRMS (TOF) (m/z): [M + H]⁺ calculated for C₁₄H₁₀IN₃O₄ 410.9716; found 410.9721.

(E)-N′-(3-bromobenzylidene)-2-hydroxy-5-iodobenzohydrazide (4e). White solid (82%); ¹H NMR (600 MHz, DMSO-d₆): δ 11.94 (s, 1H), 11.83 (s, 1H), 8.41 (s, 1H), 8.15 (d, J = 2.3 Hz, 1H), 7.94 (t, J = 1.8 Hz, 1H), 7.76–7.70 (m, 2H), 7.68–7.63 (m, 1H), 7.44 (t, J = 7.9 Hz, 1H), 6.84 (d, J = 8.6 Hz, 1H). ¹³C NMR (151 MHz, DMSO-d₆): δ 163.65, 158.71, 147.60, 142.24, 137.20, 136.96, 133.33, 131.53, 129.81, 126.85, 122.67, 120.35, 119.58, 81.41. HRMS (TOF) (m/z): [M + H]⁺ calculated for C₁₄H₁₀IN₂O₂ 443.8970; found 443.8959.

(E)-2-hydroxy-N′-(4-hydroxybenzylidene)-5-iodobenzohydrazide (4f). White solid (76%); ¹H NMR (600 MHz, DMSO-d₆): δ 12.07 (s, 1H), 11.70 (s, 1H), 10.00 (s, 1H), 8.34 (s, 1H), 8.18 (d, J = 2.2 Hz, 1H), 7.71 (dd, J = 8.6, 2.2 Hz, 1H), 7.59 (d, J = 8.3 Hz, 2H), 6.84 (dd, J = 24.2, 8.5 Hz, 3H). ¹³C NMR (151 MHz, DMSO-d₆): δ 163.57, 160.21, 159.22, 149.95, 142.14, 136.77, 129.62, 125.40, 120.42, 119.08, 116.25, 81.26. HRMS (TOF) (m/z): [M + H]⁺ calculated for C₁₄H₁₁IN₂O₃ 381.9814; found 381.9822.

(E)-N′-(2-fluorobenzylidene)-2-hydroxy-5-iodobenzohydrazide (4g). White solid (80%); ¹H NMR (600 MHz, DMSO-d₆): δ 11.98 (s, 1H), 11.85 (s, 1H), 8.69 (s, 1H), 8.16 (d, J = 2.3 Hz, 1H), 7.96 (td, J = 7.6, 1.7 Hz, 1H), 7.73 (dd, J = 8.7, 2.2 Hz, 1H), 7.56–7.51 (m, 1H), 7.37–7.30 (m, 2H), 6.84 (d, J = 8.7 Hz, 1H). ¹³C NMR (151 MHz, DMSO-d₆): δ 163.92, 162.22, 160.56, 159.09, 142.32, 142.24, 142.21, 136.91, 132.88, 132.82, 126.94, 125.51, 122.09, 122.02, 120.42, 119.28, 116.63, 116.49, 81.31. HRMS (TOF) (m/z) [M + H]⁺ calculated for C₁₄H₁₀FIN₂O₂ 383.9771; found 383.9770.

(E)-2-hydroxy-N′-(2-hydroxy-4-methoxybenzylidene)-5-iodobenzohydrazide (4h). White solid (70%); ¹H NMR (600 MHz, DMSO-d₆): δ 11.94 (s, 2H), 11.42 (s, 1H), 8.58 (s, 1H), 8.17 (d, J = 2.2 Hz, 1H), 7.72 (dd, J = 8.7, 2.2 Hz, 1H), 7.46 (d, J = 8.6 Hz, 1H), 6.83 (d, J = 8.7 Hz, 1H), 6.54 (dd, J = 8.6, 2.4 Hz, 1H), 6.51 (d, J = 2.5 Hz, 1H), 3.79 (s, 3H). ¹³C NMR (151 MHz, DMSO-d₆): δ 163.29, 162.83, 159.94, 159.10, 150.14, 142.29, 136.85, 131.48, 120.42, 118.88, 112.16, 107.11, 101.62, 81.32, 55.82. HRMS (TOF) (m/z): [M + H]⁺ calculated for C₁₅H₁₃IN₂O₄ 411.9920; found 411.9912.

(E)-2-hydroxy-5-iodo-N′-(naphthalen-2-ylmethylene)benzohydrazide (4i). White solid (71%); ¹H NMR (600 MHz, DMSO-d₆): δ 11.93 (d, J = 14.5 Hz, 2H), 8.61 (s, 1H), 8.19 (d, J = 2.1 Hz, 2H), 8.05–8.02 (m, 1H), 8.00 (d, J = 1.2 Hz, 2H), 7.98–7.95 (m, 1H), 7.74 (dd, J = 8.6, 2.2 Hz, 1H), 7.60–7.57 (m, 2H), 6.85 (d, J = 8.6 Hz, 1H). ¹³C NMR (151 MHz, DMSO-d₆): δ 163.70, 158.91, 149.42, 142.21, 137.05, 134.35, 133.30, 132.24, 129.59, 129.06, 128.90, 128.29, 127.79, 127.31, 123.18, 120.39, 119.55, 81.38. HRMS (TOF) (m/z): [M + H]⁺ calculated for C₁₈H₁₃IN₂O₂ 416.0022; found 416.0022.

(E)-N′-benzylidene-2-hydroxy-5-iodobenzohydrazide (4j). White solid (76%); ¹H NMR (600 MHz, DMSO-d₆): δ 11.91 (s, 1H), 11.85 (s, 1H), 8.45 (s, 1H), 8.17 (d, J = 2.2 Hz, 1H), 7.82–7.67 (m, 3H), 7.48 (td, J = 5.1, 4.7, 3.0 Hz, 3H), 6.84 (d, J = 8.6 Hz, 1H). ¹³C NMR (151 MHz, DMSO-d₆): δ 163.67, 158.93, 149.51, 142.21, 137.04, 134.48, 130.88, 129.37, 127.77, 120.39, 119.41, 81.36. HRMS (TOF) (m/z): [M + H]⁺ calculated for C₁₄H₁₁IN₂O₂ 366.9865; found 366.9856.

2.3. Biological Assay

2.3.1. Screening for α-Glucosidase Inhibitory Activity

The α-glucosidase inhibitory activity of 4a–4j was evaluated using p-nitrophenyl-α-D-glucopyranoside (pNPG) as the substrate. Test samples were prepared as 0.1 mM solutions in dimethyl sulfoxide (DMSO) [24]. In a 96-well microplate, 10 µL of each sample was mixed with 70 µL of 100 mM phosphate buffer (pH 6.8) and 10 µL of α-glucosidase enzyme solution (0.05 U/mL final concentration). The α-glucosidase enzyme was generously provided by the Complex Carbohydrate Research Center (University of Georgia, USA). The reaction mixture was incubated at room temperature for 15 min, after which 10 µL of the substrate solution (5 mM pNPG in phosphate buffer) was added to each well.

After an additional 1 h incubation, the release of p-nitrophenol was monitored spectrophotometrically at 450 nm using a microplate reader. Individual blanks were prepared by replacing the substrate with phosphate buffer to correct for background absorbance. Acarbose, a commercially available α-glucosidase inhibitor, was used as a positive control, while DMSO served as the vehicle control [31]. All screening experiments were performed in triplicate.

2.3.2. Determination of IC₅₀ Values

For the most active compounds, the IC₅₀ values (concentration required to inhibit 50% of enzyme activity) were determined. Samples were prepared at varying concentrations (100, 50, 25, 12.5, 6.25, 3.125, and 1.5625 µM), and each concentration was assayed in triplicate. The assay was performed as described above, and the percentage inhibition at each concentration was calculated using the following formula:

$$\mathrm{\%}Inhibition=\left(1-{\displaystyle \frac{{Absorption}_{sample}}{{Absorption}_{control}}}\right)\times 100$$

The percentage inhibition values were plotted against the log-transformed concentrations of the compounds. Nonlinear regression analysis using a four-parameter logistic model was performed in Python 3.10 (utilizing the SciPy 1.11 library) to determine the IC₅₀ values.

2.4. Theoretical Study

To complement the experimental findings, comprehensive theoretical investigations were carried out to elucidate the electronic, structural, and pharmacokinetic properties of 4a–4j, as well as their interactions with the α-glucosidase enzyme. The following approaches were employed:

2.4.1. Density Functional Theory (DFT) and Molecular Electrostatic Potential (MEP)

The electronic properties of 4a–4j were evaluated using DFT calculations. Geometry optimizations were carried out using the Gaussian09 software package at the ωB97X-D/def2-tzvpp level of theory [32,33]. No symmetry constraints were applied during the optimization. ΔE values were calculated to assess the electronic reactivity of the compounds. The MEP maps were generated to visualize the charge distribution over the molecule’s surface, highlighting the electrophilic and nucleophilic regions.

2.4.2. Molecular Docking (MD) and Molecular Dynamics Simulation (MDS)

Molecular docking studies were performed to identify the binding interactions between the Schiff base derivatives and the active site of α-glucosidase. An ensemble docking–virtual screening pipeline was employed using 10 different conformations of α-glucosidase obtained from 500 ns of molecular dynamics simulations. The binding free energies (ΔG) were calculated using GNINA’s convolutional neural network (CNN)-based scoring method [34,35,36].

For the top-ranked compounds, 4e and 4j, molecular dynamics simulations were conducted for 500 ns in triplicate to evaluate the stability and intermolecular interactions of the enzyme–ligand complexes. Key parameters, including root mean square deviations (RMSDs), root mean square fluctuations (RMSFs), and ligand residence time, were analyzed to confirm the stability and binding efficiency of the compounds.

2.4.3. ADMET Predictions

The pharmacokinetic and toxicity profiles of the synthesized Schiff base derivatives were evaluated using ADMETLab v3.0 (https://admetlab3.scbdd.com/server/evaluationCal, accessed on 7 December 2024) [37]. The ADMETLab v3.0 platform was selected for this analysis due to its comprehensive nature, its use of validated machine learning models trained on large, curated datasets, and its established use in the field, ensuring reliable in silico predictions. A comprehensive profile was calculated to assess drug-likeness and potential safety. This included key physicochemical descriptors, pharmacokinetic parameters covering Absorption, Distribution, Metabolism, and Excretion (ADME), and critical Toxicity endpoints (including hepatotoxicity, cardiotoxicity, mutagenicity, and nephrotoxicity).

2.4.4. SAR and Statistical Analysis

The SAR analysis was conducted to identify the structural features of the Schiff base derivatives that contribute to their α-glucosidase inhibitory activity. Key descriptors, including electronic properties, hydrophobicity, and steric factors, were evaluated to correlate structural modifications with biological activity.

All statistical analyses, including analysis of variance (ANOVA) and post hoc Tukey’s HSD tests, were performed using R programming. Nonlinear regression models were utilized to determine IC₅₀ values and their confidence intervals.

All graphs and plots, including dose–response curves, molecular property distributions, and correlation analyses, were generated using Python. Libraries such as Matplotlib 3.7 and Seaborn 0.12 were employed to ensure high-quality visual representations of the data.

3. Results and Discussion

3.1. Synthesis and Characterization

The synthesis of Schiff base derivatives 4a–4j was accomplished via a three-step synthetic strategy, as depicted in Scheme 1. The starting material, 5-iodo-salicylic acid 1, underwent esterification using ethanol and catalytic sulfuric acid, yielding 5-iodo-2-salicylic ester 2 in high yield. Subsequent hydrazinolysis of compound 2 with hydrazine hydrate under reflux conditions afforded the key intermediate, 5-iodo-2-hydroxybenzo-hydrazide 3 [26].

The final step involved the condensation of the hydrazide derivative 3 with various substituted benzaldehydes in the presence of catalytic acetic acid in refluxing ethanol, leading to the formation of 4a–4j. The reaction proceeded smoothly with excellent yields ranging from 68% to 82%, confirming the efficiency and reproducibility of the synthetic route [31].

The structures of the synthesized compounds 4a–4j were confirmed using ¹H NMR, ¹³C NMR, and high-resolution mass spectrometry (HRMS). The ¹H NMR spectra confirmed Schiff base formation by the presence of characteristic imine proton signals (–CH=N–) at δ 8.34–8.88 ppm, along with downfield shifts of the hydrazide NH and aromatic protons, indicative of imine formation. The ¹³C NMR spectra exhibited characteristic signals for the imine carbon (C=N) in the range of δ 158–164 ppm, further supporting the formation of the Schiff base linkage. Additionally, HRMS analysis provided molecular ion peaks consistent with the expected molecular formulas, confirming the structural integrity of the synthesized compounds.

3.2. Density Functional Theory (DFT) Study

To investigate the electronic properties influencing the reactivity and potential inhibitory activity of compounds 4a–4j, DFT calculations were performed using the ωB97X-D/def2-tzvpp level of theory [32,33]. The HOMO-LUMO energy gaps (ΔE) for the compounds ranged from 7.24 to 8.09 eV, reflecting differences in their electronic reactivity. Compounds with lower ΔE values, such as 4c (7.24 eV) and 4h (7.50 eV), exhibited higher theoretical reactivity, while higher ΔE values, such as in 4b (8.09 eV), indicated reduced electronic reactivity.

Despite these variations, no direct correlation was observed between ΔE values and IC₅₀ values. For instance, 4e, which exhibited the lowest IC₅₀ value (14.86 ± 0.24 µM), had a relatively high ΔE (8.03 eV), while 4g and 4j, with IC₅₀ values of 15.58 ± 0.30 µM and 17.56 ± 0.39 µM, respectively, also had higher ΔE values (8.04 eV and 8.01 eV, respectively). Meanwhile, 4h, which displayed minimal % inhibition (14.33 ± 7.47%) and had one of the smallest ΔE values (7.50 eV), was not tested for IC₅₀ due to its weak inhibition. These results imply that factors beyond electronic properties, such as steric effects, solvation, and binding dynamics, play a dominant role in determining biological activity.

To complement these findings, molecular electrostatic potential (MEP) maps were generated to visualize charge distribution and identify regions of electrophilicity and nucleophilicity. The MEP analysis revealed that compounds with balanced charge distributions, such as 4e, 4i, and 4j, demonstrated higher inhibition, highlighting the critical role of well-positioned electrophilic and nucleophilic regions in enzyme binding. Conversely, uneven or poorly localized charge distributions, as seen in compounds like 4h, resulted in significantly lower inhibitory activity, which proves that optimized charge balance is essential for effective binding interactions (detailed MEP data are presented in Sections S4.3 and S7.9 of the Supplementary Materials).

3.3. ADMET Profiling

To evaluate the drug-likeness and potential safety of the synthesized derivatives, a comprehensive in silico ADMET (Absorption, Distribution, Metabolism, Excretion, and Toxicity) profile was predicted for all compounds (4a–4j) and the reference drug, acarbose. The results are summarized in

Table 1. Predicted ADMET profile of salicylhydrazone derivatives (4a–4j) and acarbose.

Parameters	4a	4b	4c	4d	4e	4f	4g	4h	4i	4j	Acarbose
MW (g/mol) 1	371.94	366.98	426.97	410.97	443.9	381.98	383.98	411.99	416	365.99	645.25
logP 2	3.59	3.42	4.11	4.09	4.56	3.88	4.19	4.04	4.95	4.25	−4.48
logS 3	−5.01	−5.2	−5.67	−5.96	−6	−5.58	−5.58	−5.65	−5.79	−5.47	0.53
TPSA (Å 2) 4	62.55	74.58	125.06	104.83	61.69	81.92	61.69	91.15	61.69	61.69	321.17
Lipinski’s Rule 5	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✗
PAINS 6	No	No	Yes	No	No	Yes	No	Yes	No	No	No
HIA 7	Low	Low	Low	Low	Low	Low	Low	Low	Low	Low	Low
Caco-2 Perm. 8	−4.77	−4.71	−4.99	−5	−4.93	−4.86	−4.59	−5.07	−4.74	−4.75	−7.29
Pgp Substrate 9	No	Yes	No	No	Yes	Yes	Yes	Yes	Yes	Yes	Yes
PPB (%) 10	98.69	98.12	99.46	98.92	99	98.62	98.67	98.25	99.15	98.83	15.22
BBB Penetration 11	No	No	No	No	No	No	No	No	No	No	No
CYP Inhibition 12	1A2, 2C9, 2C19	1A2, 2C9	1A2, 2C9	1A2, 2C9	1A2, 2C9	1A2, 2C9	1A2, 2C9	1A2, 2C9	1A2, 2C9	1A2, 2C9	None
CL Plasma (ml/min/kg) 13	5.08	3.68	3.38	3.73	4.04	4.45	4.28	4.66	4.42	4.42	0.14
T1/2 (h) 14	0.66	0.77	0.92	0.76	0.74	0.82	0.72	0.75	0.66	0.69	3.64
hERG Blockade 15	Low	Low	Low	High	Low	Low	Low	Low	High	Low	Low
Hepatotoxicity 15	Low	Low	Low	Low	Low	Low	Low	Low	Low	Low	Low
AMES Mut. 15	Low	Low	High	High	Low	Low	Low	Low	High	Low	High
Nephrotoxicity 15	High	Low	Low	Low	Low	Low	High	High	High	Low	High

¹ Molecular Weight. ² Predicted octanol/water partition coefficient. ³ Predicted aqueous solubility. ⁴ Topological polar surface area. ⁵ Adherence to Lipinski’s Rule of Five. ⁶ Pan-Assay Interference Compounds alert. ⁷ Predicted human intestinal absorption. ⁸ Predicted Caco-2 cell permeability (log nm/s). ⁹ Predicted P-glycoprotein substrate (probability > 0.5). ¹⁰ Predicted plasma protein binding percentage. ¹¹ Predicted blood–brain barrier penetration (probability > 0.5). ¹² Major CYP isoforms (1A2, 2C9, 2C19, 2D6, 3A4) with a predicted inhibition probability > 0.5 are listed. ¹³ Predicted plasma clearance. ¹⁴ Predicted elimination half-life. ¹⁵ Predicted toxicity risk is classified as low (probability < 0.5) or high (probability ≥ 0.5).

. Overall, all synthesized compounds (4a–4j) exhibited favorable physicochemical properties, adhering to Lipinski’s Rule of Five, which suggests good potential for oral bioavailability. The predicted logP values (3.42 to 4.95) indicate good lipophilicity, while TPSA values are all well within the favorable range for cell permeability. Notably, compounds 4c, 4f, and 4h raised a PAINS alert, flagging them as potential assay-interfering compounds.

In terms of pharmacokinetics, all derivatives were predicted to have low human intestinal absorption (HIA) but good Caco-2 permeability, suggesting that while passive absorption might be limited, the compounds can effectively permeate cell membranes once absorbed. All synthesized compounds showed very high plasma protein binding (>98%), which could lead to a longer half-life but reduced free drug concentration compared to acarbose (15.2%). Encouragingly, none of the compounds were predicted to penetrate the blood–brain barrier, indicating a low risk of CNS side effects.

A significant finding from the metabolism analysis is the potential for drug–drug interactions. All ten synthesized compounds (4a–4j) are predicted to be inhibitors of both CYP1A2 and CYP2C9. This contrasts sharply with acarbose, which showed no inhibition potential for the five major isoforms.

The toxicity profiles revealed varied risks. Most compounds were predicted to have a low risk for hERG blockade and hepatotoxicity. However, several compounds were flagged for potential AMES mutagenicity (4c, 4d, 4i) and nephrotoxicity (4a, 4g, 4h, 4i), highlighting specific safety concerns that would need to be addressed in further studies. In comparison, acarbose also presents a high risk for AMES mutagenicity and nephrotoxicity, suggesting that the new derivatives are not necessarily at a disadvantage in these specific areas.

These in silico findings align with trends observed for other salicylhydrazone scaffolds in the literature which often display favorable lipophilicity and cell permeability but can present challenges with high plasma protein binding and specific CYP450 isoform interactions [23,24,25,26,27,28,29,30]. The predicted inhibition of CYP1A2 and CYP2C9 across our series, for instance, highlights a common liability for this chemotype that requires consideration during lead optimization. However, the favorable safety predictions for most of our compounds, particularly the low risk of cardiotoxicity (hERG blockade) and mutagenicity, position this series as a promising starting point for the development of safer α-glucosidase inhibitors.

3.4. α-Glucosidase Inhibition Assay

The α-glucosidase inhibitory activity of compounds 4a–4j was evaluated to identify potent enzyme inhibitors [29,31]. The percentage inhibition at a fixed concentration of 100 µM was determined, and the results are summarized in

Table 2. α-Glucosidase inhibitory activity of compounds 4a–4j at 100 µM concentration, IC₅₀ values, HOMO-LUMO energy gap, binding free energy (BFE), and CNN-based binding affinity (AFF).

Entry	Compounds	Ar	% Inhibition 1	IC50 (µM)	ΔE 2 (eV)	BFE 3 (Kcal mol−1)	Affinity 4 (pK Units)
1	4a	thiazole	66.71 ± 4.27	nd 5	7.89	−6.46 ± 0.62	4.74 ± 0.33
2	4b	pyridine	51.65 ± 4.72	nd	8.09	−5.83 ± 0.58	4.78 ± 0.35
3	4c	3-NO2-4-OH-Ph	45.51 ± 7.45	nd	7.24	−6.62 ± 0.66	4.60 ± 0.27
4	4d	4-NO2-Ph	59.86 ± 2.60	nd	7.60	−6.61 ± 0.78	4.69 ± 0.27
5	4e	3-Br-Ph	92.35 ± 2.52 *	14.86 ± 0.24 **	8.03	−7.23 ± 0.64	4.63 ± 0.26
6	4f	4-OH-Ph	55.75 ± 1.91	nd	7.73	−6.87 ± 0.60	4.77 ± 0.26
7	4g	2-F-Ph	85.69 ± 1.59	15.58 ± 0.30 **	8.04	−7.09 ± 0.63	4.71 ± 0.23
8	4h	2-OH-4-OCH3-Ph	14.33 ± 7.47	nd	7.50	−6.59 ± 0.59	4.64 ± 0.24
9	4i	Naph	88.64 ± 0.78 **	18.05 ± 0.92 **	7.62	−7.10 ± 0.88	4.67 ± 0.20
10	4j	Ph	93.84 ± 1.49 **	17.56 ± 0.39 **	8.01	−7.38 ± 0.82	4.78 ± 0.31
11	Acarbose	-	84.66 ± 0.71	45.78 ± 1.95	-	−7.41 ± 0.79	5.00 ± 0.22

¹ % Inhibition values were calculated based on absorbance at 450 nm at a fixed concentration of 100 µM and converted to % inhibition using the standard formula; ² ΔE (eV): HOMO-LUMO energy gap calculated at the ωB97X-D/def2-tzvpp level, indicating electronic reactivity. ³ Binding free energy and ⁴ CNN-based calculated affinity in pK units obtained from molecular dynamic simulations; ⁵ nd: IC₅₀ not determined. * p < 0.05 and ** p < 0.01 compared to acarbose control.

and illustrated in Figure 2. Among the tested compounds, 4e, 4j, 4i, and 4g exhibited the highest inhibitory activities, with % inhibition values of 92.4 ± 2.52%, 93.8 ± 1.49%, 88.6 ± 0.78%, and 85.7 ± 1.59%, respectively. These values significantly exceed the inhibition of the standard drug acarbose (84.7 ± 0.71%), (p < 0.01). Compound 4g also showed high activity, though it was not statistically different from acarbose (p > 0.05). In contrast, some compounds, such as 4h, demonstrated minimal activity (14.3 ± 7.47%), while others displayed moderate activity, with % inhibition values ranging from 45.5 ± 7.45% (4c) to 66.7 ± 4.27% (4a).

To further quantify the inhibitory potency, IC₅₀ values were determined for the most active compounds (4e, 4g, 4i, and 4j) using a concentration–response assay [20,23]. The IC₅₀ value of 4e was found to be 14.86 ± 0.24 µM, while 4g, 4i, and 4j exhibited IC₅₀ values of 15.58 ± 0.30 µM, 18.05 ± 0.92 µM, and 17.56 ± 0.39 µM, respectively (Table 2). All four compounds demonstrated significantly greater potency (p < 0.01) than acarbose (45.78 ± 1.95 µM), confirming the greater potency of these derivatives.

Statistical analysis of the % inhibition data, including one-way ANOVA followed by Tukey’s HSD test, confirmed that the differences between inhibitory activities of the compounds were statistically significant (p < 0.05). The statistical validation indicated that compounds 4e, 4j, and 4i performed significantly better than weaker inhibitors like 4h, reinforcing the observed trends in inhibition.

3.5. Structure–Activity Relationship (SAR) Analysis

The SAR analysis of derivatives 4a–4j reveals critical structural features that influence α-glucosidase inhibitory activity. The most potent inhibitors, 4j, 4e, 4i, and 4g, exhibited the highest % inhibition values (93.8%, 92.4%, 88.6%, and 85.5%, respectively), suggesting that specific substituents and molecular architectures significantly impact enzyme binding and activity. The planar aromatic systems in 4i (naphthalene) and 4j (phenyl) likely enhance π–π stacking interactions within the enzyme’s active site, thereby contributing to their superior inhibitory effects. Similarly, halogenated derivatives such as 4e (3-Br) and 4g (2-F) demonstrated strong inhibition, suggesting that electron-withdrawing groups (EWGs) optimize electronic distribution and hydrophobic interactions, facilitating stable enzyme–ligand interactions.

Conversely, compounds bearing electron-donating groups (EDGs), such as hydroxyl (-OH) and methoxy (-OCH₃) substituents, exhibited reduced activity. For example, 4c (3-NO₂-4-OH) and 4h (2-OH-4-OCH₃) demonstrated only 45.5% and 14.3% inhibition, respectively, indicating that steric hindrance and uneven charge distribution may disrupt the optimal binding conformation at the enzyme’s active site. These findings emphasize that substituent effects, particularly the balance between EWGs and EDGs, are pivotal in modulating inhibitory activity.

This conclusion is further supported by the DFT results (discussed in Section 3.3), which revealed no direct correlation between electronic reactivity (ΔE gap) and biological activity, reinforcing that steric and hydrophobic interactions are the dominant drivers of the SAR for this series. MEP maps further confirmed that balanced charge distributions in 4e, 4i, and 4j contribute to favorable enzyme binding, reinforcing their high inhibitory potential. This indicates that optimal electronic properties, combined with steric accessibility and favorable binding interactions, are essential for enhancing α-glucosidase inhibition.

The SAR analysis underscores the multifaceted nature of enzyme inhibition, where electronic, structural, and steric factors collectively determine activity. Among the synthesized Schiff base derivatives, 4e and 4j emerge as the most promising candidates, demonstrating high inhibitory activity and favorable molecular properties. (Detailed SAR analysis is provided in the Supplementary Materials).

3.6. Molecular Docking and Binding Free Energy Analysis

An in silico virtual screening campaign consisting of molecular docking, binding free energy calculations, conventional molecular dynamics simulations (cMDSs), and drug-likeness analysis was conducted to evaluate the binding potential of Schiff base derivatives toward human α-glucosidase (α-GLU) and to identify potential drug candidates [34].

To improve the accuracy of molecular docking predictions, an ensemble docking–virtual screening pipeline was implemented. This approach utilized ten distinct α-GLU conformations, extracted from 500 ns of cMDSs, to account for conformational flexibility and enhance the robustness of molecular docking evaluations [35,36]. The binding free energies and docking scores (kcal mol⁻¹) were averaged and further corrected using GNINA’s convolutional neural network (CNN) method for more reliable affinity predictions [38,39].

As shown in Figure 3, Schiff base derivatives exhibited favorable binding free energies, comparable to or better than those of acarbose, a well-known commercial α-GLU inhibitor. Among the tested compounds, 4j displayed the lowest energy value (−7.38 kcal mol⁻¹), followed by 4e (−7.23 kcal mol⁻¹), and 4i and 4g (both ~−7.10 kcal mol⁻¹). In comparison, acarbose exhibited a binding free energy of −7.41 kcal mol⁻¹, suggesting that some Schiff base derivatives have comparable or stronger binding potential than the reference inhibitor. The observed differences in binding free energies may be attributed to variations in binding modes, functional group interactions, and molecular flexibility.

3.7. Molecular Dynamics Simulation for 4e and 4j

To further assess binding stability, the top-performing derivatives, 4e and 4j, were selected for 500 ns molecular dynamics simulations. The selection was based on a multi-parameter evaluation. While compounds 4e, 4g, 4i, and 4j all demonstrated excellent IC₅₀ values, significantly better than those of acarbose, 4e and 4j were ultimately chosen based on a combination of (1) their superior performance in the initial inhibition screen (92.4% and 93.8%, respectively), (2) their leading binding free energy scores (−7.23 and −7.38 kcal/mol, respectively), and (3) their favorable physicochemical properties, such as good solubility, which was a limiting factor for compound 4i. This comprehensive assessment identified 4e and 4j as the most promising overall candidates for in-depth stability analysis. These simulations aimed to evaluate intermolecular interactions, conformational stability, and binding mode persistence within the α-glucosidase (α-GLU) active site. The root mean square deviation (RMSD), root mean square fluctuation (RMSF), and key intermolecular interactions were analyzed to quantify the stability and molecular adaptability of the enzyme–inhibitor complexes.

As illustrated in Figure 4A, the overall conformational stability of the 4e/α-GLU and 4j/α-GLU complexes exhibited minimal structural deviations (<0.5 Å), comparable to the control acarbose-bound enzyme simulation. Notably, 4e induced fewer atomic motions than acarbose, suggesting higher stability in the binding pocket. Both compounds also exhibited similar fluctuations in α-GLU side chains, particularly around catalytic residues, further confirming binding site accommodation and retention (Figure 4B).

Analysis of active site flexibility revealed that general acid–base residues within the α-GLU catalytic region underwent restricted motion (<1.4 Å RMSF values) in the presence of 4e and 4j, implying that these inhibitors successfully stabilized the enzyme’s active conformation. The presence of π-stacking interactions between the iodine-modified phenyl moiety of Schiff bases and Trp406 significantly contributed to binding affinity and stability. Additionally, halogen moieties facilitated hydrogen bond formation with adjacent water molecules, reinforcing ligand stability in the enzyme binding pocket.

Interestingly, center-to-edge π-stacking interactions with Tyr299 played a pivotal role in enhancing ligand residence time, measured at 88% for 4e and 92% for 4j during each simulation replica (Figure 4C,D). This interaction promoted a single, dominant binding mode, maintaining proximity to catalytically relevant residues. However, despite these stabilizing forces, the formation of hydrogen bonds with surrounding water molecules may increase the ΔG of desolvation, potentially reducing the efficiency of intermolecular interactions and affecting the overall drugability of these Schiff base derivatives.

4. Conclusions

In this study, a series of iodinated salicylhydrazone derivatives were synthesized and identified as potent α-glucosidase inhibitors, with compounds 4e, 4g, 4i, and 4j exhibiting IC₅₀ values superior to those of the standard drug, acarbose. These findings position them as promising leads for developing agents with a postprandial antihyperglycemic effect for managing type 2 diabetes. Integrated computational analyses, including molecular docking, dynamics simulations, and DFT studies, revealed that the enhanced potency is likely driven by stable binding in the catalytic active site through key interactions like π–π stacking and halogen bonding, which is suggestive of a competitive inhibition mechanism. While promising, this work is limited to in vitro and in silico models. Future efforts should therefore focus on structural optimization, in vivo validation, and detailed enzyme kinetic studies to confirm the mechanism of action, advancing these compounds toward preclinical evaluation.