Synthesis and anti-antileishmanial profile of novel thiazolidine-4-one derivatives and molecular docking and molecular dynamics simulations studies Authors: Soheila Molaei, Jafar Abbasi Shiran, Arina Esmailzadeh, Soroush Rajabi, Mohammad Reza Rahimi, Yasamin Abedini Zal, Saghi Sepehri Published: 2026 ABSTRACT Leishmaniasis is a severely neglected intracellular protozoan disease caused by infection with parasites of the Leishmania genus, affecting over 98 countries globally. Presently, the infection threatens the health of more than 350 million people globally, resulting in an estimated 60,000 deaths yearly from both visceral and cutaneous forms of the parasite. In present research, a novel series of thiazolidine-4-one analogues (7a-g) was synthesized and evaluated against Leishmania major. In vitro assays demonstrated using Leishmania major promastigotes and amastigotes, while cytotoxicity was assessed on the J774.A1 macrophage cell line. All tested compounds exhibited inhibitory effects against promastigotes, with compounds 7a (IC50 = 34.97 μM), 7c (IC50 = 40.79 μM), and 7b (IC50 = 62.59 μM) showing the highest potency. Against the amastigote form, analogues 7a, 7b, and 7d demonstrated notable efficacy, with IC50 values of 354.53 μM, 665.72 μM, and 746.59 μM, respectively. Flow cytometry analysis revealed that these compounds induce multiple stages of apoptosis, including both early and late apoptosis, as well as necrosis, in promastigotes. Furthermore, some derivatives exhibited moderate cytotoxic effects on macrophages. Compounds 7a and 7b demonstrated the highest selectivity indices (SI) of 6.23 and 3.22, respectively. Overall, the tested compounds showed stronger anti-promastigote activity compared to their anti-amastigote effects. In silico analyses indicated that the thiazolidine-4-one derivatives possess satisfactory drug-likeness characteristics and identified probable metabolic sites via major cytochrome P450 enzymes. Finally, molecular docking and molecular dynamics simulations were performed at the active site of Leishmania major pteridine reductase 1 (LmPTR1). They revealed that the most active compounds, 7a and 7c, adopt favorable binding orientations and maintain stable interactions with key residues of LmPTR1, correlating with their enhanced in vitro activity. Additionally, RMSD, RMSF, hydrogen bonds, Rg, SASA, PCA, and energy analysis during molecular dynamics simulations indicated stable binding of two compounds with LmPTR1. Furthermore, the results from docking and molecular dynamics simulations demonstrated that hydrophobic interactions were critical in the binding of these compounds to LmPTR1.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-026-53762-3. Introduction Neglected tropical diseases (NTDs) involve a different class of infectious diseases caused by protozoa, helminths, viruses, bacteria, fungi, and other pathogens. These diseases are mainly endemic to tropical and subtropical regions, often overlapping with regions of widespread poverty; however, instances of NTDs have also been reported in non-endemic and developed regions. These diseases affect more than 1.5 billion people worldwide and are responsible for over 530,000 deaths every year. In response to the significant health problem of NTDs, research efforts have intensified, concentrating on the development of effective monitoring systems and the improvement of existing treatment strategies to enhance control over their transmission1. NTDs remain prevalent and continue to present a considerable threat to global health and human well-being; however, the rate of drug development to address these conditions remains significantly slow2. Leishmaniasis is an infectious disease classified as an NTD, caused by Leishmania protozoans and spread to humans by infected sandfly bites. Approximately 21 Leishmania species have been identified as pathogenic to humans, with an endemic distribution across Africa, Asia, the Americas, and Southern Europe. The parasite exhibits a biphasic life cycle, alternating between the extracellular, mobile promastigote stage—found in the gut of female sandflies of the Lutzomyia genus (in the Americas) or Phlebotomus genus (in Asia and Africa)—and the intracellular, non-mobile amastigote stage, which replicates inside the phagolysosomal compartments of the host’s macrophages3. It is estimated that more than 6 million people worldwide are infected with leishmania parasites. Up to 30% of these cases may eventually develop severe complications, including cardiac, gastrointestinal, neurological, and other potentially life-threatening manifestations, often appearing years after the initial infection4,5. Three primary clinical appearances of leishmaniasis have been identified in humans: mucocutaneous leishmaniasis (MCL), cutaneous leishmaniasis (CL), and visceral leishmaniasis (VL), also referred to as kala-azar. Among these, VL represents the most serious form and is generally fatal if left untreated. The most prevalent form is CL, which causes skin lesions or ulcers that often lead to permanent scarring or, in some cases, the more disfiguring mucocutaneous form. More than one million new cases of CL are estimated to occur each year. Leishmania major (L. major) is a primary pathogen responsible for CL in North Africa, the Middle East, and Asia6. The effective organization of this disease is difficult due to several factors, including inconsistent and insufficient clinical diagnosis, limited therapeutic options, and the emergence of drug resistance. Thus, the application of new diagnostic tools and identification techniques is crucial to understanding leishmaniasis pathogenesis, eventually facilitating the advancement of more accurate diagnostics and successful treatments6. Moreover, the lack of effective vaccines, challenges in vector control, and the emergence of drug resistance have contributed to the increasing occurrence of leishmaniasis. The most widely employed treatments include pentavalent antimonials, amphotericin B, paromomycin, and pentamidine; however, these drugs are associated with significant side effects, require prolonged administration at high doses, and are typically given parenterally. Developing a safe, effective, and affordable drug would significantly improve the management of leishmaniasis and reduce problems related to current chemotherapy routines7. Nitrogen-containing compounds, including thiazolidinones, account for approximately 20% of all compounds with antiparasitic activity. Thiazolidinone, five-membered heterocyclic ring, is biologically significant and includes a sulfur atom at position 1, a nitrogen atom at position 3, and a carbonyl group at one of the positions—2, 4, or 58. These compounds exhibit considerable pharmacological potential and involve a wide range of drug candidates with diverse activities, including antibiotic, antitumor, anti-inflammatory, antiviral, antimicrobial, antifungal, and antidiabetic effects9. Recently, thiazolidin-2-ones have been investigated as inhibitors of the BRD4 bromodomain, while thiazolidin-5-ones find applications in dye chemistry. The compound 2-thioxo-thiazolidin-4-one, commonly known as rhodanine, serves as a key scaffold in numerous drug-like molecules. Thiazolidine-2,4-dione acts as a potent hypoglycemic agent and is a core component of rosiglitazone, a clinically approved drug for treating type II diabetes. The thiazolidin-4-one moiety, often referred to as a “magic moiety” or “wonder core,” is recognized for its broad spectrum of biological activities, including antimicrobial, antitubercular, antiviral, anti-inflammatory, and antidiabetic effects8,10. Recent studies have primarily focused on investigating the antiparasitic activities of thiazolidinone compounds11. Synthetic chemistry continually demands the development of more efficient and environmentally friendly methods that maximize mass efficiency, minimize intermediate isolation steps, reduce energy consumption, and decreases waste production. Showing organic reactions in a single vessel (one-pot) represents an effective strategy for high yield and they are favored in green chemistry metrics. One-pot synthesis can be classified into three types: domino or cascade reactions, multicomponent reactions (MCRs), and one-pot stepwise synthesis (OPSS). OPSS involves multiple sequential steps in which reactants, reagents, and catalysts are introduced stepwise, allowing changes in reaction conditions at each stage. Examples of OPSS include solid-phase peptide synthesis, organic synthesis employing solid-supported reagents and scavengers, polymer synthesis, and the preparation of certain biomolecules. Numerous biocatalysis-based OPSS methods have also been documented in the literature12. Advanced computational approaches are integral to structure-based drug discovery, enabling the efficient identification of bioactive compounds. Structure-based virtual screening and molecular docking are widely used to predict binding modes within the active site of biological targets. To address the limitations of static docking, molecular dynamics (MD) simulations is employed to account for protein–ligand flexibility and complex stability. MM-PBSA/MM-GBSA binding free energy calculations provides quantitative estimates of ligand-target affinities. Additionally, QSAR modeling and in silico ADMET predictions are applied to correlate molecular features with biological activity and to evaluate pharmacokinetic and toxicity profiles13. In the present study, these complementary computational methods were integrated to support compound prioritization and experimental validation. In this study, a series of derivatives based on the thiazolidine-4-one scaffold were synthesized using a one-pot, two-step, three-component method. These compounds were evaluated against both L. major amastigote and promastigote forms, as well as their cytotoxic effects on macrophage cells. Furthermore, in silico analyses were shown to provide insights into the compounds’ stability and potential for further investigation. Materials and methods Synthesis All chemicals were purchased from Sigma-Aldrich (USA) and used as received, without any additional purification. FT-IR spectra for the compounds were recorded using a Perkin-Elmer spectrometer. The 1H- and 13C-NMR spectra (in DMSO-d6) were acquired on a Bruker Ultrashield instrument operating at 400 and 100 MHz, respectively. Elemental analysis for carbon, hydrogen, and nitrogen was carried out with a Heraeus CHN-O-Rapid analyzer. General procedure for the synthesis of (E)-3-(((E)-N-benzylidene)amino)-2/4-substituted-thiazolidin-4-one (7a-g) A mixture of the corresponding isothiocyanate 1 (1 mmol) and hydrazine derivative 2 (1 mmol) was stirred at room temperature in absolute ethanol (15 mL) and triethylamine (Et3N), leading to the formation of the intermediate N-alkylhydrazinecarbothioamide 3. Subsequently, substituted benzaldehyde 4 (1 mmol) was added to the reaction mixture along with acetic acid and sodium sulfate (Na2SO4) to yield intermediate 5. In the next step, 2-chloroacetyl chloride 6 and Et3N were added, and the mixture was refluxed for 8–12 h (progress was monitored by TLC). Following completion, the reaction mixture was cooled to room temperature and worked up in an ice bath. The resulting precipitates were filtered and recrystallized from a 1:1 (v/v) ethanol/water mixture to afford the pure target compounds (7a-g). The spectral data for the synthesized derivatives are provided below: (E)-3-(((E)-3,4-dimethoxybenzylidene)amino)-2-(methylimino)thiazolidin-4-one (7a) Yellow precipitates, 0.21 gr, yield 75%, m,p.161–163 °C, Rf = 0.5 (n-hexene/ethyl acetate 1:1), FT-IR (KBr): υmax, cm−1 = 3145 (CH, aromatic), 3014 (CH, aliphatic), 1720 (C=O), 1621 (C=N), 1511, 1401 (C=C), 1267 (C–N), 647 (C–S, thiazolidine-4-one). 1H-NMR (400 MHz, DMSO-d6): δ/ppm = 8.62 (1H, s, CH imine), 7.47 (1H, d, J = 1.6 Hz, CH-Ph), 7.44 (1H, d, J = 2.0 Hz, CH-Ph), 7.06 (1H, d, J = 8.4 Hz, CH-Ph), 3.93 (2H, s, CH thiazolidine-4-one), 3.81 (3H, s, OCH3), 3.77 (3H, s, OCH3), 1.04 (3H, s, CH3). 13C-NMR (100 MHz, DMSO-d6): δ/ppm = 176 (C=O, thiazolidine-4-one), 161 (C=N, thiazolidine-4-one), 157 (C, imine), 151 (C-Ph), 150 (C-Ph), 127 (C-Ph), 123 (C-Ph), 111 (C-Ph), 109 (C-Ph), 56 (OCH3), 55 (OCH3), 33 (CH3), 31 (C–S, thiazolidine-4-one). MS m/z (%): 293.01 (M), 278.10, 263.20, 248.15; Anal. calcd for C13H15N3O3S: C 53.23, H 5.15, N 14.32, S 10.93%, found: C 53.61, H 4.92, N 14.01, S 10.75%. N-((E)-3-(((E)-3,4-dimethoxybenzylidene)amino)-4-oxothiazolidin-2-ylidene)nitramide (7b) Yellow precipitates, 0.27 gr, yield 82%, m.p.168–169 °C, Rf = 0.5 (n-hexene/ethyl acetate 1:1), FT-IR (KBr): υmax, cm−1 = 3151 (CH, aromatic), 3002 (CH, aliphatic), 1719 (C=O), 1602 (C=N), 1550, 1402 (C=C), 1232 (C–N), 1137 (OCH3), 640 (C–S, thiazolidine-4-one). 1H-NMR (400 MHz, DMSO-d6): δ/ppm = 11.37 (1H, s, CH imine), 7.95 (1H, s, CH-Ph), 7.43 (1H, d, J = 1.6 Hz, CH-Ph), 6.96 (1H, d, J = 8.0 Hz, CH-Ph), 3.93 (2H, s, CH thiazolidine-4-one), 3.81 (3H, s, OCH3), 3.77 (3H, s, OCH3). 13C-NMR (100 MHz, DMSO-d6): δ/ppm = 178 (C=O, thiazolidine-4-one), 158 (C=N, thiazolidine-4-one), 152 (C, imine), 151 (C-Ph), 149 (C-Ph), 127 (C-Ph), 122 (C-Ph), 111 (C-Ph), 109 (C-Ph), 56 (OCH3), 30 (C–S, thiazolidine-4-one). MS m/z (%): 324.15 (M), 309.10, 294.05, 248.20; Anal. calcd for C12H12N4O5S: C 44.44, H 3.73, N 17.28, S 9.89%, found: C 44.12, H 3.48, N 17.57, S 10.16%. (E)-3-(((E)-3,4-dimethoxybenzylidene)amino)-2-(pyridin-2-ylimino)thiazolidin-4-one (7c) Light yellow precipitates, 0.28 gr, yield 78%, m.p.176–178 °C, Rf = 0.6 (n-hexene/ethyl acetate 1:1). FT-IR (KBr): υmax, cm−1 = 3158 (C-H, aromatic), 3014 (C-H, aliphatic), 1731 (C=O), 1602 (C=N), 1556, 1402 (C=C), 1232 (C–N), 1137 (OCH3), 645 (C–S, thiazolidine-4-one). 1H-NMR (400 MHz, DMSO-d6): δ/ppm = 8.94 (1H, s, CH imine), 8.51 (1H, d, J = 4.8 Hz, CH- 2-pyridine), 8.33 (1H, d, J = 8.4 Hz, CH-2-pyridine), 8.13 (1H, s, CH-2-pyridine), 7.66 (1H, dd, J1 = 8.0 Hz, J2 = 4.8 Hz, CH-Ph), 7.56 (1H, d, J = 4.0 Hz, CH-2-pyridine), 7.30 (1H, dd, J1 = 7.6 Hz, J2 = 4.0 Hz, CH-Ph), 7.01 (1H, d, J = 8.0 Hz, CH-2-pyridine), 4.11 (2H, s, CH-thiazolidine-4-one), 3.82 (3H, s, OCH3), 3.79 (3H, s, OCH3). 13C-NMR (100 MHz, DMSO-d6): δ/ppm = 175 (C=O, thiazolidine-4-one), 159 (C=N, thiazolidine-4-one), 151 (C, 2-pyridine), 149 (C, imine), 145 (C-Ph), 144 (C-Ph), 142 (C, 2-pyridine), 138 (C, 2-pyridine), 135 (C, 2-pyridine), 127 (C-Ph), 125 (C, 2-pyridine), 123 (C, 2-pyridine), 113 (C-Ph), 109 (C-Ph), 56 (OCH3), 30 (C–S, thiazolidine-4-one). MS m/z (%): 356.05 (M), 341.10, 326.05, 248.20; Anal. calcd for C17H16N4O3S: C 57.29, H 4.53, N 15.72, S 9.00%, found: C 56.92, H 4.65, N 15.98, S 9.24%. N-((E)-3-(((E)-4-bromobenzylidene)amino)-4-oxothiazolidin-2-ylidene)nitramide (7d) White precipitates, 0.30 gr, yield 85%, m.p.188–190 °C, Rf = 0.71 (n-hexene/ethyl acetate 1:2). FT-IR (KBr): υmax, cm−1 = 3131 (CH, aromatic), 3002 (CH, aliphatic), 1721 (C=O), 1628 (C=N), 1582, 1401 (C=C), 1309 (C–N), 641 (C–S), 514 (C–Br). 1H-NMR (400 MHz, DMSO-d6): δ/ppm = 8.51 (1H, s, CH imine), 7.72 (4H, d, J = 8.0 Hz, CH-Ph), 3.82 (2H, s, CH-thiazolidine-4-one). 13C-NMR (100 MHz, DMSO-d6): δ/ppm = 172 (C=O, thiazolidine-4-one), 165 (C=N, thiazolidine-4-one), 157 (C, imine), 134 (C-Ph), 132 (C-Ph), 129 (C-Ph), 124 (C-Ph), 32 (C–S, thiazolidine-4-one). MS m/z (%): 343.10 (M), 263.20, 217.15; Anal. calcd for C10H7BrN4O3S: C 35.00, H 2.06, N 16.33, S 9.34%, found: C 35.37, H 1.89, N 16.14, S 9.59%. (E)-3-(((E)-4-hydroxy-3-methoxybenzylidene)amino)-2-(methylimino)thiazolidin-4-one (7e) White precipitates, 0.25 g, yield 88%, m.p.188–189 °C, Rf = 0.71 (n-hexene/ethyl acetate 1:2). FT-IR (KBr): υmax, cm−1 = 3456 (OH), 3139 (CH, aromatic), 2998 (CH, aliphatic), 1709 (C=O), 1604 (C=N), 1511, 1401 (C=C), 1286 (C–N), 1113 (OCH3), 647 (C–S). 1H-NMR (400 MHz, DMSO-d6): δ/ppm = 9.67 (1H, s, OH), 8.34 (1H, s, CH-imine), 7.36 (1H, d, J = 1.6 Hz, CH-Ph), 7.22 (1H, dd, J1 = 8.4 Hz, J2 = 1.6 Hz, CH-Ph), 6.86 (1H, d, J = 8.0 Hz, CH-Ph), 3.94 (2H, s, CH-thiazolidine-4-one), 3.82 (3H, s, OCH3), 3.16 (3H, s, CH3). 13C-NMR (100 MHz, DMSO-d6): δ/ppm = 173 (C=O, thiazolidine-4-one), 163 (C=N, thiazolidine-4-one), 157 (C, imine), 149 (C-Ph), 148 (C-Ph), 126 (C-Ph), 123 (C-Ph), 117 (C-Ph), 116 (C-Ph), 56 (OCH3), 40 (C–S, thiazolidine-4-one), 30 (CH3). MS m/z (%): 279.10 (M), 278.15, 263.20, 248.10; Anal. calcd for C12H13N3O3S: C 51.60, H 4.69, N 15.04, S 11.48%, found: C 51.94, H 4.57, N 15.36, S 11.79%. (E)-3-(((E)-4-bromobenzylidene)amino)-2-(methylimino)thiazolidin-4-one (7f) Yellow precipitates, 0.24 gr, yield 78%, m.p. 197–198 °C, Rf = 0.97 (n-hexene/ethyl acetate 1:2). FT-IR (KBr): υmax, cm−1 = 3138 (CH, aromatic), 3000 (CH, aliphatic), 1717 (C=O), 1623 (C=N), 1576, 1400 (C=C), 1300 (C-N), 641 (C-S), 514 (C–Br). 1H-NMR (400 MHz, DMSO-d6): δ/ppm = 8.43 (1H, s, CH-imine), 7.65 (2H, d, J = 8.4 Hz, CH-Ph), 7.61 (2H, d, J = 8.4 Hz, CH-Ph), 3.90 (2H, s, CH-thiazolidine-4-one), 3.09 (3H, s, CH3). 13C-NMR (100 MHz, DMSO-d6): δ/ppm = 172 (C=O, thiazolidine-4-one), 161 (C=N, thiazolidine-4-one), 156 (C, imine), 133 (C-Ph), 132 (C-Ph), 129 (C-Ph), 124 (C-Ph), 32 (C–S, thiazolidine-4-one), 30 (CH3). MS m/z (%): 312.10 (M), 232.10, 217.20; Anal. calcd for C11H10BrN3OS: C 42.32, H 3.23, N 13.46, S 10.27%, found: C 42.63, H 3.34, N 13.14, S 10.12%. N-((E)-3-(((E)-4-hydroxybenzylidene)amino)-4-oxothiazolidin-2-ylidene)nitramide (7g) White precipitates, 0.22 gr, yield 78%, m.p. 197–198 °C, Rf = 0.97 (n-hexene/ethyl acetate 1:2). FT-IR (KBr): υmax, cm−1 = 3412 (OH), 3171 (CH, aromatic), 2980 (CH, aliphatic), 1684 (C=O), 1614 (C=N), 1584, 1399 (C=C), 1302 (N=O), 1275 (C–N), 654 (C–S). 1H-NMR (400 MHz, DMSO-d6): δ/ppm = 10.04 (1H, s, OH), 8.34 (1H, s, CH-imine), 7.62 (2H, d, J = 8.4 Hz, CH-Ph), 6.85 (2H, d, J = 8.4 Hz, CH-Ph), 3.94 (2H, s, CH-thiazolidine-4-one). 13C-NMR (100 MHz, DMSO-d6): δ/ppm = 172 (C=O, thiazolidine-4-one), 163 (C-Ph), 160 (C=N, thiazolidine-4-one), 157 (C, imine), 130 (C-Ph), 125 (C-Ph), 117 (C-Ph), 32 (C–S, thiazolidine-4-one). MS m/z (%): 280.05 (M), 279.10, 234.20; Anal. calcd for C10H8N4O4S: C 42.86, H 2.88, N 19.99, S 11.44%, found: C 43.22, H 2.75, N 20.19, S 11.29%. Anti-leishmanial activity Herein, variety of standardized reagents and culture media were employed to evaluate the compounds’ activity against L. major. L. major promastigote (MRHO/IR/75/ER) was initially cultured in Novy-MacNeal-Nicolle (NNN) medium, which consists of Brain Heart Infusion Blood Agar and Luria Broth (Merck, Germany). Following cultivation was done in Roswell Park Memorial Institute medium (RPMI-1640), enriched with fetal bovine serum (FBS) and a combination of penicillin (100 U/mL) and streptomycin (100 µg/mL) (Gibco, Germany) to promote growth and prevent microbial contamination14. All reagents were stored according to the manufacturers’ recommended conditions to ensure their stability and reliability throughout the experiments. The synthesized thiazolidine-4-one derivatives (7a-g) were first dissolved in dimethyl sulfoxide (DMSO; Sigma-Aldrich) to preparate stock solutions with a concentration of 5 mg/mL. To minimize the risk of DMSO-related cytotoxicity, the final concentration of DMSO in all biological assays was strictly maintained below 1%, a threshold previously shown to be non-toxic to leishmania parasites14,15. Serial dilutions of the compounds were prepared in phosphate-buffered saline (PBS), resulting in final concentrations ranging from 4.7 to 1000 µg/mL. These dilutions were used to cultures of macrophages, as well as to the amastigote and promastigote forms of the parasite15. To evaluate the cytotoxic and apoptotic effects of the tested compounds, cell viability was assessed by the MTT assay (Sigma-Aldrich). Apoptosis was performed with a FITC Annexin V Apoptosis Detection Kit (Mab Tag, Germany). Amphotericin B (AmpB; Cipla, 50 mg) and meglumine antimoniate (Glucantime®; Sanofi, 1.5 g/5 mL ampoules) were used as reference drugs for assays involving the promastigote and amastigote forms, respectively16. Cultivation of L. major The standard Iranian strain of L. major (MRHO/IR/75/ER) was kindly provided by Dr. Ghaffarifar from the Department of Parasitology, Faculty of Medical Sciences, Tarbiat Modares University, Tehran, Iran. Promastigote forms were initially cultured in NNN medium and then sub-cultured in RPMI-1640 supplemented with 20% heat-inactivated fetal bovine serum (FBS) and 1% Penicillin–Streptomycin solution, containing streptomycin. Cultures were maintained at 18–24 °C under standard laboratory conditions to support optimal parasite growth17. Anti-leishmanial activity Anti-promastigote effects MTT assay for promastigote viability The MTT reduction assay, which relies on the conversion of 2-imino-thiazolidin-4-ones into formazan crystals, was used to evaluate the viability of L. major promastigotes. For the assay, 100 μL of RPMI-1640 medium containing L. major promastigotes at a density of 1 × 106 cells/mL in the logarithmic growth phase was added to each well of a 96-well ELISA plate. The parasites were then treated with different concentrations of the synthesized compounds (4.7–1000 μg/mL). Subsequent steps—including incubation and formazan crystal solubilization—were carried out following established protocols14,18. Following incubation, cell viability was evaluated spectrophotometrically by measuring the absorbance of the formazan product. The percentage of viable promastigotes was determined according to the following equation:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\mathrm{Viability}}\left( \% \right) = {1}00 \times \left[ {\left( {{\mathrm{Abs}}\_{\mathrm{treated}} - {\mathrm{Abs}}\_{\mathrm{blank}}} \right)/\left( {{\mathrm{Abs}}\_{\mathrm{control}} - {\mathrm{Abs}}\_{\mathrm{blank}}} \right)} \right]$$\end{document} Abs_treated shows the absorbance measured from compound-treated cells, Abs_control represents absorbance from untreated cells (negative control), and Abs_blank corresponds to wells containing only the culture medium without cells. Live promastigotes suspended in PBS without drug treatment served as the control group, while cell-free wells were used as the blank (CTRL-). AmB at 1 µg/mL was included as the positive control. The IC50 values were determined as the concentrations of each compound that inhibited promastigote viability by 50% in comparison to untreated controls14,15. Anti-intracellular amastigote activity To evaluate the compounds’ anti-leishmanial activity against intracellular L. major amastigotes, J774A.1 murine macrophages were seeded at a density of 1 × 105 cells per well in 12-well plates containing sterile round glass coverslips. The plates were incubated at 37 °C in a humidified atmosphere with 5% CO2 for 24 h to allow macrophage adhesion. Non-adherent cells were then gently removed by washing with sterile PBS. Metacyclic promastigotes (1 × 106 parasites per well, corresponding to a 10:1 parasite-to-macrophage ratio) were added to the adherent macrophages and incubated for 24 h to start infection. Following this, to remove any non-internalized parasites, wells were washed with PBS. Infected cultures were then treated with numerous concentrations of the compounds (4.7 to 1000 µg/mL) and incubated for an additional 72 h under standard conditions (37 °C, 5% CO2, 95% humidity). After incubation, cells were fixed with methanol and discolored with Giemsa. The number of intracellular amastigotes (Leishman bodies) per 100 macrophages was counted under a light microscope. IC50 values were determined by comparing the reduction in amastigote numbers in treated samples relative to untreated controls. All experiments were performed in triplicate to ensure reproducibility. Glucantime® (GLU) at 100 µg/mL was used as the positive control14,15. Flow cytometry assay for apoptosis detection The induction of apoptosis in L. major promastigotes after treatment with selected compounds was assessed by the Annexin V-FITC Apoptosis Detection Kit (Mab Tag, Germany). Promastigotes in the logarithmic growth phase were treated with the IC50 concentrations of the most potent compounds. After 24 h incubation, the cells were collected, washed twice with cold sterile PBS, and centrifuged at 1000 × g for 15 min. The cell pellets were gently resuspended in 400 μL of binding buffer, and 5 μL of Annexin V-FITC, along with 5 μL of propidium iodide (PI), were subsequently added. The samples were incubated in the dark at room temperature (25 °C) for 15 min to ensure proper staining. Apoptotic and necrotic cell populations were then analyzed using a CyFlow Space flow cytometer (Sysmex-Partec, USA), with data processed by FlowJo™ software (version 10.5.3, Vancouver, BC). Experiments were performed in triplicate, and the apoptotic and necrotic promastigotes’ percentages were calculated for each compound tested14,18. Statistical analysis All experiments were carried out in triplicate, and results were reported as mean ± SD. The IC50 values for Leishmania promastigotes and intracellular amastigotes, as well as the CC50 values for J774A.1 macrophages, were calculated based on the average percentage viability of the respective cell populations. Data analysis, IC50 and CC50 calculations, and graphical presentations were accomplished by GraphPad Prism software (GraphPad Software Inc., San Diego, CA). Changes in promastigote viability over time were analyzed using repeated measures ANOVA in the parasite counting assay. To compare IC50 and CC50 values across different compounds, one-way ANOVA followed by Tukey’s post hoc test was used to identify statistically significant differences, with P-values below 0.05 considered significant. The selectivity index (SI) for each compound was calculated as the ratio of CC50 to the IC50 against amastigotes, providing an indicator of compound selectivity and potential therapeutic window in animal models. Computational studies In silico pharmacokinetic prediction and physicochemical properties The pharmacokinetic prediction and drug-likeness properties of the compounds were evaluated using the ADMETLab 3.0 online platform (https://admetlab3.scbdd.com). The assessed parameters included absorption (GI absorption, P-gp substrate), distribution (blood–brain barrier permeability), metabolism (CYP450 enzyme inhibition), carcinogenicity, drug-induced nephrotoxicity, drug-induced neurotoxicity and drug-likeness based on Lipinski’s rule of five and other physicochemical descriptors. All predictions were performed using the default settings of ADMETLab 3.0. The platform was accessed in Jul. 202419. In silico metabolism of compounds in the presence of cytochrome P450 The potential sites of metabolism for the derivatives were systematically predicted using the SMARTCyp 3.0 online server (https://smartcyp.sund.ku.dk/mol_to_som), which estimates likely cytochrome P450-mediated metabolic hot spots based on molecular structure. The analysis specifically focused on the CYP3A4, CYP2D6, and CYP2C9 isoforms, which are major contributors to drug metabolism in human. Predictions were performed using the default settings of SMARTCyp 3.0. The server was accessed in Sep. 201920. Molecular docking study The interaction of the derivatives with the active site of L. major pteridine reductase 1 (LmPTR1) was investigated through molecular docking using AutoDock 4.221,22. The three-dimensional structure of LmPTR1 (PDB ID: 5L42) was downloaded from the Protein Data Bank (PDB). The docking procedure followed the methodology reported in an earlier study23,24. After removing crystallized water molecules and ligands, Kollman charges and polar hydrogens were added to the enzyme structure. The derivatives’ structures were prepared and energy-minimized using the MM + method in HyperChem 8.0 software. Gasteiger charges were then assigned to the compounds. Finally, the pdbqt files for both the enzyme and the compounds, required for docking, were generated using AutoDock Tools. Docking calculations were performed using the Lamarckian Genetic Algorithm implemented in AutoDock 4.2. For each ligand, multiple docking runs were carried out, and the resulting binding poses were clustered based on an RMSD tolerance of 2.0 Å. RMSD values were calculated relative to the lowest-energy conformation. The most populated cluster with the lowest binding energy was selected as the representative binding mode for further analysis. Molecular dynamics simulations study Molecular dynamics (MD) simulations was carried out using GROMACS version 2022.6, following the protocol described in a previous study25,26. Topology files and force field parameters for the selected compounds were generated using the Acpype tool. Type of box was used cubic and the TIP3P water model was applied to solvate the system, incorporating approximately 11,864 water molecules. To maintain overall charge neutrality, counter-ions were added by replacing water molecules in the simulation box, which contained LmPTR1, the ligand, and water. The Amber99sb force field was applied to describe all bonded and nonbonded interactions. Energy minimization of the protein–ligand complex was performed using steepest descent followed by conjugate gradient algorithms. Following minimization, position restraints were applied during equilibration with NVT and NPT ensembles. During the NVT simulation, the temperature was maintained at 300 K using a coupling constant of 0.1 ps for 500 ps. Once the temperature stabilized, the NPT ensemble was conducted at a constant pressure of 1 bar using the Parrinello-Rahman barostat, with a coupling constant of 5.0 ps, over 1000 ps. The Particle Mesh Ewald method handled long-range electrostatics, the LINCS algorithm maintained covalent bond constraints, and the Lennard–Jones potential was used for van der Waals interactions. Lastly, 150 ns MD simulations was carried out, with equilibration monitored by assessing the stability of energy, system density, temperature, and backbone atom root-mean-square deviations (RMSDs). Principal component analysis and free energy landscape construction Principal component analysis (PCA) was performed on the molecular dynamics (MD) trajectory to characterize the dominant collective motions of the protein–ligand complexes. Prior to PCA, the MD trajectory were aligned to a reference structure in order to remove overall translational and rotational motions. The analysis was carried out on the Cartesian coordinates of selected atoms (e.g., Cα atoms of the protein backbone) extracted from the equilibrated portion of the simulation. The covariance matrix of atomic positional fluctuations was constructed according to:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\mathrm{C}}_{{{\mathrm{ij}}}} = \left\langle {\left( {{\mathrm{X}}_{{\mathrm{i}}} - \left\langle {{\mathrm{X}}_{{\mathrm{i}}} } \right\rangle } \right)\left( {{\mathrm{X}}_{{\mathrm{j}}} - \left\langle {{\mathrm{X}}_{{\mathrm{j}}} } \right\rangle } \right)} \right\rangle$$\end{document}where Xi and Xj represent the Cartesian coordinates of atoms i and j, respectively, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left\langle . \right\rangle$$\end{document} denotes the time average over the MD trajectory. Diagonalization of the covariance matrix yielded a set of eigenvalues and corresponding eigenvectors, with the eigenvectors defining the principal components (PCs) and the eigenvalues representing the amplitude of motion along each PC. The first two principal components (PC1 and PC2), which accounted for the largest fraction of the total variance, were selected to describe the essential dynamics of the system. The MD trajectory was projected onto these principal components to obtain a two-dimensional conformational space. The free energy landscape (FEL) was then constructed by calculating the probability distribution P(PC1, PC2) of the projected conformations and applying the Boltzmann relation:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\upkappa}_{{\mathrm{B}}} {\mathrm{TlnP}}\left( {{\mathrm{PC1}},{\text{ PC2}}} \right) - = {\mathrm{G}}\left( {{\mathrm{PC1}},{\text{ PC2}}} \right)$$\end{document}where κB is the Boltzmann constant and T is the absolute temperature of the simulation. The resulting FEL provides a thermodynamic representation of the conformational space, allowing the identification of low-energy basins corresponding to the most stable and frequently sampled binding conformations of the protein–ligand complexes. Binding free energy analyses based on the MM/PBSA method Binding free energy calculations were performed to quantitatively evaluate the binding affinity of the selected compounds using the g_mmpbsa tool based on the MM/PBSA method. The binding free energy was calculated as:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\Delta {\mathrm{G}}_{{{\mathrm{binding}}}} = \Delta {\mathrm{G}}_{{{\mathrm{complex}}}} {-}\left( {\Delta {\mathrm{G}}_{{{\mathrm{protein}}}} + \Delta {\mathrm{G}}_{{{\mathrm{ligand}}}} } \right)$$\end{document} According to the MM/PBSA approach, the binding free energy is defined as:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\Delta {\mathrm{G}}_{{{\mathrm{binding}}}} = \Delta {\mathrm{G}}_{{{\mathrm{MM}}}} + \Delta {\mathrm{G}}_{{{\mathrm{PB}}}} + \Delta {\mathrm{G}}_{{{\mathrm{SA}}}} - {\mathrm{T}}\Delta {\mathrm{S}}$$\end{document} There ∆GMM includes van der Waals and electrostatic interactions, ∆GPB and ∆GSA represent polar and nonpolar solvation energies, respectively, and − T∆S corresponds to the entropic contribution. MM/PBSA calculations were carried out for the two selected protein–ligand complexes using snapshots extracted from the final stable 10 ns of the MD trajectory. Results and discussions Rationale and design Thiazolidinones are carbonyl-containing derivatives of thiazolidine, the saturated counterpart of thiazole. Structurally, they consist of a five-membered, non-aromatic heterocyclic ring featuring both sulfur and nitrogen atoms. Compounds with a thiazolidinone core display a wide spectrum of pharmacological effects. Due to the vital role of thiazolidinones in the mechanism of many therapeutic agents, they have gained significant attention from medicinal chemists as a valuable scaffold for designing novel drug candidates. Various thiazolidinone derivatives have demonstrated bioactivities like antibacterial, antifungal, anticancer, antioxidant, antimalarial, and antitubercular effects27. Numerous studies have investigated the anti-leishmanial properties of thiazolidinones. For example, Haroon et al. evaluated these compounds against Leishmania amazonensis and Leishmania infantum. Based on a structure–activity relationship (SAR) analysis of thiazolidin-4-one, the presence of a C5-ketone group in combination with a C4-ethyl acetate moiety significantly enhanced leishmanicidal activity compared to miltefosine (Compound 1; Fig. 1)28. Bhat et al. synthesized some of the quinoline-thiazolidin-4-one hybrids and assessed their inhibitory effects against L. donovani. Among them, Compound 2 exhibited notable anti-leishmanial activity. Additionally, cytotoxicity assessment using the MTT assay on mouse embryonic fibroblast cells indicated that compound 2 had low toxicity (Fig. 1)29. Schadich et al. evaluated a series of thiazolidine-based hybrids against L. major (Compound 3; Fig. 1). The most potent candidates demonstrated low cytotoxicity toward normal human fibroblasts and exhibited a favorable therapeutic index. SAR analysis suggested that the lack of substitution at position 3 of the thiazolidinone ring, along with the lack of bulky groups at the five-membered ring positions, contributed to the enhanced anti-leishmanial activity11. Neri et al. synthesized thiazolidine-2,4-dione derivatives with potential anti-leishmanial activity. Among them, the most active derivative against L. infantum (Compound 4; Fig. 1) had a halogen substituent at the 3-position of the phenyl ring. However, introducing halogens at the 2- or 6-positions resulted in reduced activity. A major limitation of this series was their cytotoxicity, although compound 4 demonstrated a SI greater than one. Additionally, compound 5 (Fig. 1) showed strong binding affinity to LmPTR1. Further analysis revealed that small hydrophobic substituents attached to the meta/para positions are essential for activity against LmPTR1. Whereas lipophilic groups at the ortho-position negatively influenced the compound’s efficacy against L. infantum30. Leite et al. utilized the thiazolidinone ring as a bioisosteric substitute for the pteridine/purine core in the design of PTR1 inhibitors. Among the thiazolidine-2,4-dione derivatives investigated, compound 6 (Fig. 1) was introduced as the most promising lead candidate31. Filho et al. applied the concept of bioisosterism to transform a potent antiparasitic aryl thiosemicarbazone, known for its cruzain-inhibitory activity, into thiazolidine-4-one derivatives. Evaluation of antiparasitic activity and cytotoxicity of compound 7 (Fig. 1) against Trypanosoma cruzi Y-strain trypomastigotes and J774 macrophages showed high potency and selectivity. Notably, a phenyl group at the N3 position of the heterocyclic ring of the compound 7 highlighted this substituent as a key structural determinant. Overall, the study demonstrated that converting thiosemicarbazones into thiazolidine-4-ones preserves antiparasitic efficacy while improving selectivity32.Fig. 1Design plan of thiazolidinone analogues (7a-g) against the leishmanial parasite. Design plan of thiazolidinone analogues (7a-g) against the leishmanial parasite. According to previous studies, in this study, a series of novel thiazolidine-4-one derivatives featuring electron-withdrawing groups (EWG), electron-donating groups (EDG), and heteroaromatic rings connected to the NH position of the thiazolidine-4-one scaffold, along with various substitutions at the para/meta positions of the benzylidene ring was synthesized. These compounds were tested for activity against L. major. Furthermore, in silico analyses were performed to guide and validate the structure-based design approach. The design strategy of this work was grounded in findings from previous studies (Fig. 1). Chemistry In current research, (E)-3-(((E)-N-benzylidene)amino)-2/4-substituted-thiazolidin-4-ones (7a-g) were synthesized via a one-pot, three-step method. The synthetic route for these compounds (7a-g) is illustrated in Scheme 1 and Fig. 233,34. The reaction between hydrazine, various benzaldehydes, and suitable isothiocyanates with 2-chloroacetyl chloride was led to the formation of substituted (E)-3-(((E)-N-benzylidene)amino)-2/4-substituted-thiazolidin-4-one derivatives (7a-g). All compounds were efficiently separated using simple filtration and further purified using a water/ethanol (1:1 v/v) mixture. This method provided the analogues in good yields. Various spectroscopic techniques have been employed to confirm the structures of all synthesized compounds, including FT-IR, 1H-NMR, 13C-NMR, as well as mass spectrometry and elemental analysis. The FT-IR spectra of (E)-3-(((E)-N-benzylidene)amino)-2/4-substituted-thiazolidin-4-one derivatives (7a-g) exhibited characteristic C=N stretching vibrations in the range of 1602–1628 cm−1. Aromatic C=C stretching bands were observed between 1511–1584 cm−1 and 1399–1402 cm−1. The C–N and C–S stretching vibrations were appeared at 1232–1309 cm−1 and 640–654 cm−1, respectively. In the 1H-NMR spectra, the imine proton was observed between 11.37 and 8.34 ppm, while aromatic protons were detected from 8.33 to 6.85 ppm. The proton of the thiazolidine-4-one ring was found at 3.94–3.82 ppm. Signals corresponding to methoxy and methyl protons appeared at 3.82–3.77 ppm and 3.16–1.04 ppm, respectively. The 13C-NMR spectra showed the carbonyl carbon (C=O) of the thiazolidine-4-one ring between 172 and 179 ppm. The imine carbons of the ring and the imine group were appeared at 159–165 ppm and 152–157 ppm, respectively. Aromatic carbons were observed in the range of 151–111 ppm, while methoxy and methyl carbons were detected between 56 and 30 ppm.Scheme 1Synthesis of (E)-3-(((E)-N-benzylidene)amino)-2/4-substituted-thiazolidin-4-one (7a-g).Fig. 2Structural synthesized (E)-3-(((E)-N-benzylidene)amino)-2/4-substituted-thiazolidin-4-one (7a-g). Synthesis of (E)-3-(((E)-N-benzylidene)amino)-2/4-substituted-thiazolidin-4-one (7a-g). Structural synthesized (E)-3-(((E)-N-benzylidene)amino)-2/4-substituted-thiazolidin-4-one (7a-g). The proposed reaction mechanism of (E)-3-(((E)-N-benzylidene)amino)-2/4-substituted-thiazolidin-4-ones (7a-g) is illustrated in Scheme 2. Initially, the substituted isothiocyanate undergoes a reaction with hydrazine, resulting in the formation of a hydrazinecarbothioamide intermediate. Subsequently, this intermediate condenses with benzaldehyde derivatives to provide intermediate 5. Next, triethylamine (Et3N) acts as a base catalyst, and the hydrazone NH group in intermediate 5 deprotonates, a process driven by the relatively high acidity of this proton. Following this, the nitrogen atom attacks the carbonyl carbon of 2-chloroacetyl chloride. In the presence of excess Et3N, condensation and cyclization reactions occur, resulting in the formation of the final compounds (7a-g).Scheme 2The probable reaction mechanism for the one-pot, three-steps, and four-component synthesis of (E)-3-(((E)-N-benzylidene)amino)-2/4-substituted-thiazolidin-4-one (7a-g). The probable reaction mechanism for the one-pot, three-steps, and four-component synthesis of (E)-3-(((E)-N-benzylidene)amino)-2/4-substituted-thiazolidin-4-one (7a-g). Anti-leishmanial activity Promastigote and amastigote cytotoxic assay The anti-leishmanial activity and cytotoxicity of the synthesized derivatives were assessed against L. major using the MTT assay. Figure 3 illustrates the percentage of promastigote viability after 72 h of exposure to varying concentrations of the compounds. Among them, compounds 7a and 7c demonstrated the strongest inhibitory effects on promastigote growth, with IC50 values of 34.97 ± 1.80 µM and 40.79 ± 1.54 µM, respectively. On the other hand, compound 7e revealed the lowest inhibition, with an IC50 of 1375.1 ± 3.37 µM. A summary of the inhibitory effects of all compounds against L. major promastigotes is presented in Table 1. Statistical analysis showed that none of the compounds had a significantly different inhibitory effect compared to the positive control, Amphotericin B (AmB) (p > 0.05) (Table 1).Fig. 3Log concentration (µM) vs % viability of the promastigotes against (E)-3-(((E)-N-benzylidene)amino)-2/4-substituted-thiazolidin-4-one (7a-g) at different concentrations by MTT assay.Table 1IC50 values (µM) of synthesized thiazolidine-4-one derivatives were evaluated against both promastigote and amastigote forms of L. major. Cytotoxicity (CC50) was assessed on the J774A.1 cell line, and the selectivity index (SI) was calculated based on the amastigote results. Data are presented as mean ± SD (n = 3).CompoundIC50 (µM) (L. major), (MRHO/IR/75/ER)CC50 (µM) (J774 A.1 cell line)P-value vs. GLUSI (CC50/IC50 of amastigotes)PromastigotesP-value vs. AmBAmastigotesP-value vs. GLU7a34.97 ± 1.80p > 0.05354.53 ± 3.03p > 0.052209.38 ± 2.86p > 0.056.237b62.59 ± 1.94p > 0.05665.72 ± 2.77p > 0.052145 ± 2.83p > 0.053.227c40.79 ± 1.54p > 0.05991.58 ± 2.47p > 0.052856.30 ± 0.89p > 0.052.887d283.22 ± 2.60p > 0.05746.59 ± 2.01p > 0.05976.20 ± 2.42P < 0.051.37e1375.10 ± 3.37p > 0.051598.22 ± 2.04p > 0.052126.30 ± 2.43p > 0.051.337f.959 ± 27.54p > 0.052344.72 ± 2.88p > 0.055061.02 ± 2.33p > 0.052.157g984.10 ± 3.68p > 0.051417.25 ± 2.64p > 0.052119.10 ± 2.42p > 0.051.49AmB0.96 ± 0.93NR21.5 ± 1.2NR1595 ± 3.9NR74.2GLU790 ± 1.90NR211.88 ± 0.68NR1728.49 ± 4.7NR8.15 Log concentration (µM) vs % viability of the promastigotes against (E)-3-(((E)-N-benzylidene)amino)-2/4-substituted-thiazolidin-4-one (7a-g) at different concentrations by MTT assay. Table 1IC50 values (µM) of synthesized thiazolidine-4-one derivatives were evaluated against both promastigote and amastigote forms of L. major. Cytotoxicity (CC50) was assessed on the J774A.1 cell line, and the selectivity index (SI) was calculated based on the amastigote results. Data are presented as mean ± SD (n = 3).CompoundIC50 (µM) (L. major), (MRHO/IR/75/ER)CC50 (µM) (J774 A.1 cell line)P-value vs. GLUSI (CC50/IC50 of amastigotes)PromastigotesP-value vs. AmBAmastigotesP-value vs. GLU7a34.97 ± 1.80p > 0.05354.53 ± 3.03p > 0.052209.38 ± 2.86p > 0.056.237b62.59 ± 1.94p > 0.05665.72 ± 2.77p > 0.052145 ± 2.83p > 0.053.227c40.79 ± 1.54p > 0.05991.58 ± 2.47p > 0.052856.30 ± 0.89p > 0.052.887d283.22 ± 2.60p > 0.05746.59 ± 2.01p > 0.05976.20 ± 2.42P < 0.051.37e1375.10 ± 3.37p > 0.051598.22 ± 2.04p > 0.052126.30 ± 2.43p > 0.051.337f.959 ± 27.54p > 0.052344.72 ± 2.88p > 0.055061.02 ± 2.33p > 0.052.157g984.10 ± 3.68p > 0.051417.25 ± 2.64p > 0.052119.10 ± 2.42p > 0.051.49AmB0.96 ± 0.93NR21.5 ± 1.2NR1595 ± 3.9NR74.2GLU790 ± 1.90NR211.88 ± 0.68NR1728.49 ± 4.7NR8.15 IC50 values (µM) of synthesized thiazolidine-4-one derivatives were evaluated against both promastigote and amastigote forms of L. major. Cytotoxicity (CC50) was assessed on the J774A.1 cell line, and the selectivity index (SI) was calculated based on the amastigote results. Data are presented as mean ± SD (n = 3). To assess the anti-parasitic activity of the synthesized compounds against the amastigote form of L. major, the murine macrophage cell line J774.A1 was employed. These cells closely resemble human macrophages, making them a suitable in vitro model for simulating the parasite’s intracellular environment within the host35. This in vitro system provides a controlled environment to accurately evaluate how the synthesized compounds influence macrophage activity and function—key factors in the host’s defense against Leishmania parasites in vivo. Anti-parasitic assays were shown using Glucantime (GLU) as the positive control and appropriate negative controls. All tests were performed in triplicate over a 72 h incubation period. According to the results (Fig. 4), compound 7a exhibited the highest inhibitory activity, with an IC50 value of 354.53 ± 3.03 µM, while compound 7f displayed minimal inhibition, with an IC50 of 2344.72 ± 2.88 µM against the amastigote form. A summary of the compounds’ inhibitory effects against L. major amastigotes is presented in Table 1. Statistical analysis showed no significant difference in efficacy observed between any of the tested compounds and the positive control (GLU) (p > 0.05).Fig. 4Log concentration (µM) vs % viability of the amastigotes against (E)-3-(((E)-N-benzylidene)amino)-2/4-substituted-thiazolidin-4-one (7a-g) at different concentrations by MTT assay. Log concentration (µM) vs % viability of the amastigotes against (E)-3-(((E)-N-benzylidene)amino)-2/4-substituted-thiazolidin-4-one (7a-g) at different concentrations by MTT assay. Evaluation of cytotoxic effects of compounds on macrophages (J774.A1) To evaluate the cytotoxic effect of the synthesized compounds on macrophages, the MTT colorimetric assay was utilized. In this assay, peritoneal macrophages were exposed to different compounds, and their viability was compared to that of the GLU (Fig. 5). Analysis of the data showed that compound 7d was the most cytotoxicity compound, with a CC50 value of 976.20 ± 42.0 µM. In contrast, compound 7f exhibited the least cytotoxicity, with a CC50 of 5061.02 ± 33.20 µM (Table 1). These results suggested that compound 7d poses a higher risk of toxicity to macrophages, however compound 7f may offer greater therapeutic safety due to its lower cytotoxicity.Fig. 5Log concentration (µM) vs % viability of the J774.A1 macrophages against (E)-3-(((E)-N-benzylidene)amino)-2/4-substituted-thiazolidin-4-one (7a-g) at different concentrations by MTT assay. Log concentration (µM) vs % viability of the J774.A1 macrophages against (E)-3-(((E)-N-benzylidene)amino)-2/4-substituted-thiazolidin-4-one (7a-g) at different concentrations by MTT assay. Evaluating drug safety plays an essential role in developing novel therapeutic compounds, as it determines whether a compound can deliver its intended effects without harming host (mammalian) cells. To evaluate this, researchers often use a parameter called the Selectivity Index (SI), which measures a compound’s ability to target parasitic cells while minimizing toxicity to normal mammalian cells. The SI is calculated by dividing the CC50 (the concentration that reduces host cell viability by 50%) by the IC50 (the concentration that inhibits 50% of parasite activity). A higher SI value indicates greater selectivity and suggests a stronger potential for safe and effective therapeutic use. Analysis of the SI values showed that compound 7a, with an SI of 6.23, had higher selectivity for L. major over mammalian cells. The SI values for all tested compounds have been listed in Table 1. These findings emphasize the relatively low cytotoxicity of compound 7a to mammalian cells, alongside its notable antiparasitic activity. Furthermore, the MTT assay results showed that all derivatives—except for 7d—did not show statistically significant cytotoxicity toward uninfected macrophages compared to the GLU control (p > 0.05). In contrast, compound 7d significantly reduced macrophage viability (p < 0.05), suggesting a higher risk of host cell toxicity. Apoptosis profile The IC50 concentrations of the more potent derivatives (compounds 7a-d) were used to evaluate their capability to induce apoptosis in L. major promastigotes. The findings revealed that these analogues triggered varying levels of early and late apoptosis, as well as necrosis. Specifically, the apoptosis’ percentages at IC50 for compounds 7a-d and GLU were 61.1%, 43.1%, 35.7%, 5.73%, and 28.73%, respectively. Compounds 7a and 7c induced significantly higher apoptotic effects compared to GLU (P < 0.01). Notably, compound 7d exhibited a markedly greater necrotic effect (27.5%) than both the others and GLU (p < 0.001) (Fig. 6).Fig. 6The apoptotic and necrotic profiles of L. major promastigotes were evaluated at IC50 concentrations of various compounds and compared to GLU, a standard drug used in Leishmaniasis treatment. The apoptotic and necrotic profiles of L. major promastigotes were evaluated at IC50 concentrations of various compounds and compared to GLU, a standard drug used in Leishmaniasis treatment. Structure–activity relationship (SAR) studies The thiazolidine-4-one derivatives (7a-g) were tested against both the L. major promastigote and amastigote forms. The IC50S of compounds were measured and compared to those of GLU and AmB (Table 1). Among them, compounds 7a-c exhibited the strongest anti-leishmanial activity (Table 1). A closer look at these three compounds shows that they share the same benzylidene moiety but differ in the substituents attached to the N-thiazolidine-4-one ring. Structurally, compound 7b demonstrated weaker anti-promastigote activity compared to 7a and 7c. This may be attributed to the presence of a nitramide group (-NH-NO2) in 7b, which is a strong EWG. This group can reduce the molecule’s electronic stability and hinder its interaction with biological targets. Additionally, the high polarity of the nitro group may limit its ability to cross membranes, and it could also be chemically unstable or metabolically reactive. Compound 7c showed moderate anti-promastigote activity. It contains an imino group (C=N) linked to a pyridine ring, which can be protonated at physiological pH, reducing cellular uptake. The relatively large size and polarity of this structure may further restrict cell penetration or interact with intracellular targets. In contrast, compound 7a features a smaller, less polar methylimino group (= N–CH3), which can increase lipophilicity and membrane permeability, facilitating more efficient cell entry. The absence of acidic or basic groups at this position (due to the methyl group) also may improve chemical stability under physiological conditions, resulting in superior anti-promastigote activity. Based on these observations, compound 7a appears to be more potent than 7b and 7c. In summary, differences in anti-promastigote activity among these compounds can be explained by: (I) Lipophilicity and size—compound 7a’s smaller, more lipophilic methylimino group promotes better membrane penetration; (II) Polarity and stability—compound 7b’s highly polar nitro group may lead to rapid degradation or metabolism; and (III) Electronic characteristics—compound 7c’s pyridine moiety may reduce bioavailability due to protonation and limited target interaction. Regarding activity against the amastigote form, the efficacy ranking is 7a > 7b > 7c. The relatively higher anti-amastigote activity of compound 7b compared to 7c might be due to the nitro group enhancing aqueous solubility or optimizing the compound’s lipophilic-hydrophilic balance, which facilitates better penetration and intracellular accumulation in amastigotes. Conversely, the aromatic and bulkier pyridinyloimino group in 7c may reduce solubility and hinder membrane permeation. Since the promastigote and amastigote forms differ in membrane composition and metabolic pathways, compounds that can efficiently penetrate promastigotes or selectively interact with their enzymes and biological targets tend to show greater effectiveness against this stage36–38. In compound 7c, the pyridinyloimino group may enhance lipophilicity or improve membrane permeability in promastigotes compared to compound 7b, which could facilitate better access to biological targets. On the other hand, the nitro group in compound 7b might reduce permeability in promastigotes or be subject to metabolic inactivation by the parasite. When comparing the anti-leishmanial activity of compounds 7a, 7e, and 7f, all of which share the thiazolidine-4-one core but differ in their aromatic ring substituents (benzylidene moiety). Compound 7a was the most potent compound. It features two methoxy groups at the meta- and para-positions of the benzylidene ring and showed stronger activity than 7e and 7f (Table 1). Methoxy groups are strong EDGs, which can raise electron density on the benzylidene ring and might enhance interactions with biological targets. Additionally, improving lipophilicity helps the compound penetrate membranes more effectively. This combination of electron donation and increased hydrophobicity likely contributes to its superior biological activity. Compound 7f, with a bromine atom at the para-benzylidene position, showed moderate anti-promastigote activity. Bromine is a moderate EWG by induction but can act slightly as an electron donor through resonance. It can increase lipophilicity, aiding membrane permeability; however, its electron-withdrawing effect might weaken electronic interactions with target molecules. Compound 7e, which contains both methoxy and hydroxyl groups at the meta- and para-positions, was less active against promastigotes compared to 7a and 7f. The hydroxyl group is polar and hydrophilic, which can reduce membrane permeability. As a result, compound 7e is overall more polar and less lipophilic. Moreover, the presence of hydroxy and methoxy groups may form intramolecular hydrogen bonding, potentially causing conformational changes that affect activity. In summary, the nature and position of substituents on the aromatic benzylidene ring play a crucial role in determining the anti-leishmanial potency of these compounds11,28–32. EDGs such as methoxy significantly improve activity by enhancing both membrane penetration and interactions with molecular targets. In contrast, polar groups such as the hydroxyl group tend to reduce membrane permeability and may interfere with optimal binding, resulting in decreased activity. Interestingly, compound 7e showed higher anti-amastigote activity compared to 7f, which contrasts with the trend observed in promastigotes, where 7f was more active than 7e. This difference might be clarified by the fact that 7e’s hydroxyl and methoxy groups can form stronger hydrogen bonds with biological targets inside amastigotes, aiding penetration and binding to intracellular proteins and enzymes, thus improving its effectiveness. Meanwhile, 7f contains a bromine atom, which is hydrophobic and relatively large, properties that may enhance its membrane penetration or specific interactions in promastigotes. Overall, since amastigote and promastigote forms differ in membrane composition and metabolism, compounds may vary in effectiveness depending on these differences. Additionally, promastigotes may be less sensitive to polar groups or metabolize such compounds faster, while amastigotes could retain and absorb them more efficiently. Comparing compounds 7b, 7d, and 7g, which share the thiazolidin-4-one core but differ in substituents on the benzylidene ring. Compound 7d showed moderate activity. The bromine atom at the para-position in 7d is a moderate EWG by induction and a slight electron donor by resonance, increasing lipophilicity but not as much as methoxy groups. Compound 7b, with two methoxy groups at meta- and para-positions, enhances electron density on the ring, promotes π-π stacking and hydrogen bonding with targets, and improves membrane permeability, making it the most potent among the three. On the other hand, 7g, bearing a hydroxyl group, is highly polar and hydrophilic, which reduces lipophilicity and membrane penetration. The hydroxyl group can also form hydrogen bonds that may interfere with target binding or compound stability. Overall, EDGs and lipophilic moieties, such as a methoxy group, might facilitate membrane penetration and target interaction, while polar groups, such as hydroxyl, may reduce bioavailability and activity. Though all these compounds contain nitro groups, their effects on activity are minor and consistent, with most differences attributed to the aromatic substituents. The anti-leishmanial activities of these compounds are consistent across both promastigote and amastigote stages. Looking at compounds 7d and 7f, both have a common benzylidene group but differ in substituents on the thiazolidin-4-one core. Compound 7d exhibited stronger anti-promastigote and anti-amastigote effects than 7f. This is likely because 7d contains a nitroimino group, a strong EWG that increases the electrophilicity of the active site, enhancing interactions with parasitic enzymes. This group can improve cellular uptake and may disrupt critical parasite enzymes. In contrast, 7f’s methylimino group is an EDG and may not interact as effectively with biological targets; it could also negatively impact compound stability or distribution. Both compounds possess a para-bromine substituent on the benzylidene ring, which enhances lipophilicity and facilitates membrane penetration; however, since this feature is common to both, it does not account for the observed differences in their biological activity. Again, their activities against both parasite forms follow a consistent pattern. From evaluating all tested compounds against both L. major amastigote and promastigote forms, several SARs emerged: (I) Most compounds showed similar activity patterns against both forms, suggesting a shared mechanism or target. (II) Strong EWGs like nitroimino groups or halogenated rings enhance activity, likely by increasing electrophilicity and strengthening interactions with parasite biomolecules. (III) The thiazolidin-4-one ring is a key scaffold to activity, probably through hydrogen bonding or π-π stacking with biological targets. (IV) Side chain differences greatly influence activity; bulky or highly polar groups tend to reduce permeability, especially in intracellular amastigotes. (V) Moderate lipophilicity correlates with better activity, notably against promastigotes, as overly lipophilic compounds may have solubility issues, while very hydrophilic ones struggle to penetrate membranes. (VI) Nitroimino-containing compounds outperform methylimino-substituted ones, highlighting the importance of EWGs for improving cellular uptake and target engagement. Computational studies In silico assessment of pharmacokinetic profiles and physicochemical parameters Lipinski’s rule of five provides criteria for evaluating drug-likeness, stating that a compound should have a molecular weight (MW) of 500 or less, no more than 10 hydrogen bond acceptors, no more than 5 hydrogen bond donors, and a LogP value not exceeding 5. Compounds that violate these parameters may suffer from poor oral bioavailability20. As presented in Table 2, all analogues met the criteria for MW and LogP, with values under 500 and LogP below 5. Moreover, their total polar surface area (TPSA) extended from 45.03 to 108.4 Å2, and rotatable bonds was between 2 and 5. Both parameters are within the commonly recognized limits for effective oral bioavailability (rotatable bonds ≤ 10; TPSA ≤ 140 Å2). The compounds also featured between 4 and 9 hydrogen bond acceptors and 0 to 1 hydrogen bond donors, aligning well with Lipinski’s rule.Table 2Prediction of drug-likeness and physicochemical properties of analogues (7a-g) was performed using the ADMETLab3.0 online platform.CompoundMWnHAnHDnRotTPSALogSLogP7a293.086.00.04.063.49-3.871.407b324.059.00.05.0106.63-4.141.367c356.097.00.05.076.38-3.551.267d341.947.00.03.088.17-4.352.037e279.076.01.03.074.49-3.690.917f310.974.00.02.045.03-4.222.167g280.038.01.03.0108.40-3.480.78 Table 2Prediction of drug-likeness and physicochemical properties of analogues (7a-g) was performed using the ADMETLab3.0 online platform.CompoundMWnHAnHDnRotTPSALogSLogP7a293.086.00.04.063.49-3.871.407b324.059.00.05.0106.63-4.141.367c356.097.00.05.076.38-3.551.267d341.947.00.03.088.17-4.352.037e279.076.01.03.074.49-3.690.917f310.974.00.02.045.03-4.222.167g280.038.01.03.0108.40-3.480.78 Prediction of drug-likeness and physicochemical properties of analogues (7a-g) was performed using the ADMETLab3.0 online platform. P-glycoprotein (P-gp), an ATP-binding cassette (ABC) transporter family member, functions as a transmembrane efflux pump and is strongly expressed across biological barriers, including the small intestine, liver, kidneys, and the BBB. By actively exporting drugs out of cells, P-gp can significantly impact intestinal drug absorption. Predictive models indicated that no analogue serves as a substrate for P-gp, suggesting a lower likelihood of P-gp-mediated efflux and, consequently, the possibility of enhanced oral uptake. Crossing the BBB is a key pharmacokinetic feature for drugs targeting the central nervous system (CNS), as only compounds that can penetrate this barrier are capable of exerting effects within the brain20. According to the data, compounds 7a, 7e, and 7f were predicted to cross the BBB, implying some degree of CNS exposure (Table 3).Table 3Pharmacokinetic predictions for compounds (7a-g) were performed using the ADMETLab3.0 online platform.CompoundPgp substrateHIABBBCarcinogenicityDrug-induced nephrotoxicityDrug-induced neurotoxicityCYP2C9 inhibitorCYP2D6 inhibitorCYP3A4 inhibitor7aNoYesYesMediumMediumMediumYesNoYes7bNoYesNoYesYesNoYesNoNo7cNoYesNoMediumMediumMediumYesNoYes7dNoYesNoYesYesNoYesNoNo7eNoYesYesMediumMediumMediumYesYesYes7fNoYesYesMediumMediumMediumYesNoNo7gNoYesNoYesMediumNoNoNoNo Table 3Pharmacokinetic predictions for compounds (7a-g) were performed using the ADMETLab3.0 online platform.CompoundPgp substrateHIABBBCarcinogenicityDrug-induced nephrotoxicityDrug-induced neurotoxicityCYP2C9 inhibitorCYP2D6 inhibitorCYP3A4 inhibitor7aNoYesYesMediumMediumMediumYesNoYes7bNoYesNoYesYesNoYesNoNo7cNoYesNoMediumMediumMediumYesNoYes7dNoYesNoYesYesNoYesNoNo7eNoYesYesMediumMediumMediumYesYesYes7fNoYesYesMediumMediumMediumYesNoNo7gNoYesNoYesMediumNoNoNoNo Pharmacokinetic predictions for compounds (7a-g) were performed using the ADMETLab3.0 online platform. Cytochrome P450 enzymes play an important role in the metabolism of a wide range of drugs, with several major isoforms. The compounds were evaluated for their potential to inhibit CYP3A4, CYP2D6, and CYP2C9. All compounds except 7e showed no inhibitory activity toward CYP2D6. Derivatives 7a, 7c, and 7e were predicted to inhibit CYP3A4, while all analogues, except 7g, demonstrated inhibitory activity against CYP2C9 (Table 3). Carcinogenicity remains a major safety concern in drug development, often requiring extensive long-term studies due to its typically delayed onset. Prediction models classified all tested compounds as potentially carcinogenic. In addition, all analogues showed signs of possible drug-induced nephrotoxicity. However, compounds 7b, 7d, and 7g were not predicted to be neurotoxic, indicating a lower risk of CNS toxicity (Table 3). In silico prediction of metabolic processes involving cytochrome P450 Drug-drug interactions may occur when xenobiotics—primarily drugs and their metabolites—alter the absorption, distribution, metabolism, or excretion (ADME) of concurrently administered drugs. Such interactions may result in adverse effects, either because of simultaneous administration of multiple drugs or via interactions between a drug and its metabolites20. Importantly, three key CYP450 enzymes are responsible for the metabolism of more than half of all marketed drugs: CYP2C9, CYP3A4, and CYP2D6. Tables 4, 5 and 6 highlight the top three atomic sites within each compound that are most susceptible to metabolic transformation. According to SMARTCyp 3.0 predictions, the three atomic sites in each compound with the lowest reactivity scores were considered as the most probable sites of metabolic modification. Specifically, for CYP2C9, the primary metabolic hotspot in compounds 7a-c is at atomic site S14; in compounds 7d, 7f, and 7g, it is at site S4; and in compound 7e, it is at site S13 (Table 4). Regarding CYP3A4, the main site for adduct formation remains S14 in derivatives 7a-c, while S4 is dominant in derivatives 7d, 7f, and 7g. For compound 7e, S13 is identified as the key metabolic site (Table 5). Similarly, for CYP2D6, atomic sites S14 (in 7a-c), S4 (in 7d, 7f, and 7g), and S13 (in 7e) represent the primary locations of predicted metabolic activity (Table 6).Table 4Regarding the CYP2C9 isoform, the top three metabolically prioritized atoms are in the analogues.CompoundCYP2C92C9 RankingAtom2C9 ScoreEnergy2D SASASpan2endCOO-DistSimilarity7a1S.1438.634.545.5100.32C.1746.248.557.1000.33C.159.662.264.3001.07b1S.1444.534.543.9200.32C.159.662.264.3001.03C.665.562.264.3101.07c1S.1456.434.543.9400.32C.159.662.264.3001.03C.665.562.264.3101.07d1S.444.534.543.9200.32C.368.257.733.5200.33C.1594.884.128.3200.77e1S.1338.634.545.5100.32C.1646.248.557.1000.33C.559.662.264.3001.07f1S.438.634.545.5100.32C.746.248.557.1000.33C.362.357.733.5100.37g1S.444.534.543.9200.32C.368.257.733.5200.33C.1587.777.231.3200.7Table 5Regarding the CYP3A4 isoform, the top three metabolically prioritized atoms are in the analogues.CompoundCYP3A43A4 RankingAtom3A4 ScoreEnergy2D SASASpan2endRelative SpanSimilarity7a1S.1425.434.545.510.90.32C.1738.248.557.101.00.33C.1349.157.733.510.90.37b1S.1426.134.543.920.80.32C.1349.757.733.520.80.33C.151.662.264.301.01.07c1S.1427.034.543.940.70.32C.1350.657.733.540.70.33C.151.662.264.301.01.07d1S.426.234.543.920.80.32C.349.857.733.520.80.33C.1473.880.828.630.70.77e1S.1325.534.545.510.90.32C.1638.248.557.101.00.33C.1249.257.733.510.90.37f1S.425.534.545.510.90.32C.738.248.557.101.00.33C.349.257.733.510.90.37g1S.426.234.543.920.80.32C.349.857.733.520.80.33C.1569.477.231.320.80.7Table 6Regarding the CYP2D6 isoform, the top three metabolically prioritized atoms are in the analogues.CompoundCYP2D62D6 RankingAtom2D6 ScoreEnergy2D SASASpan2endN + DistSimilarity7a1S.1439.434.545.5100.32C.1746.248.557.1000.33C.159.662.264.3001.07b1S.1446.134.543.9200.32C.159.662.264.3001.03C.666.362.264.3101.07c1S.1459.634.543.9400.32C.159.662.264.3001.03C.666.362.264.3101.07d1S.446.134.543.9200.32C.369.857.733.5200.33C.1596.484.128.3200.77e1S.1339.434.545.5100.32C.1646.248.557.1000.33C.559.662.264.3001.07f1S.439.434.545.5100.32C.746.248.557.1000.33C.363.157.733.5100.37g1S.446.134.543.9200.32C.369.857.733.5200.33C.1589.377.231.3200.7 Table 4Regarding the CYP2C9 isoform, the top three metabolically prioritized atoms are in the analogues.CompoundCYP2C92C9 RankingAtom2C9 ScoreEnergy2D SASASpan2endCOO-DistSimilarity7a1S.1438.634.545.5100.32C.1746.248.557.1000.33C.159.662.264.3001.07b1S.1444.534.543.9200.32C.159.662.264.3001.03C.665.562.264.3101.07c1S.1456.434.543.9400.32C.159.662.264.3001.03C.665.562.264.3101.07d1S.444.534.543.9200.32C.368.257.733.5200.33C.1594.884.128.3200.77e1S.1338.634.545.5100.32C.1646.248.557.1000.33C.559.662.264.3001.07f1S.438.634.545.5100.32C.746.248.557.1000.33C.362.357.733.5100.37g1S.444.534.543.9200.32C.368.257.733.5200.33C.1587.777.231.3200.7 Regarding the CYP2C9 isoform, the top three metabolically prioritized atoms are in the analogues. Table 5Regarding the CYP3A4 isoform, the top three metabolically prioritized atoms are in the analogues.CompoundCYP3A43A4 RankingAtom3A4 ScoreEnergy2D SASASpan2endRelative SpanSimilarity7a1S.1425.434.545.510.90.32C.1738.248.557.101.00.33C.1349.157.733.510.90.37b1S.1426.134.543.920.80.32C.1349.757.733.520.80.33C.151.662.264.301.01.07c1S.1427.034.543.940.70.32C.1350.657.733.540.70.33C.151.662.264.301.01.07d1S.426.234.543.920.80.32C.349.857.733.520.80.33C.1473.880.828.630.70.77e1S.1325.534.545.510.90.32C.1638.248.557.101.00.33C.1249.257.733.510.90.37f1S.425.534.545.510.90.32C.738.248.557.101.00.33C.349.257.733.510.90.37g1S.426.234.543.920.80.32C.349.857.733.520.80.33C.1569.477.231.320.80.7 Regarding the CYP3A4 isoform, the top three metabolically prioritized atoms are in the analogues. Table 6Regarding the CYP2D6 isoform, the top three metabolically prioritized atoms are in the analogues.CompoundCYP2D62D6 RankingAtom2D6 ScoreEnergy2D SASASpan2endN + DistSimilarity7a1S.1439.434.545.5100.32C.1746.248.557.1000.33C.159.662.264.3001.07b1S.1446.134.543.9200.32C.159.662.264.3001.03C.666.362.264.3101.07c1S.1459.634.543.9400.32C.159.662.264.3001.03C.666.362.264.3101.07d1S.446.134.543.9200.32C.369.857.733.5200.33C.1596.484.128.3200.77e1S.1339.434.545.5100.32C.1646.248.557.1000.33C.559.662.264.3001.07f1S.439.434.545.5100.32C.746.248.557.1000.33C.363.157.733.5100.37g1S.446.134.543.9200.32C.369.857.733.5200.33C.1589.377.231.3200.7 Regarding the CYP2D6 isoform, the top three metabolically prioritized atoms are in the analogues. Molecular docking study Computational methods such as molecular docking serve as powerful tools for exploring interactions between enzymes and small molecules. They also play a crucial role in recognizing key pharmacophoric features, thereby guiding the design of potential therapeutically active compounds39. This section aims to provide a rationale for the observed anti-leishmanial activity by focusing on a specific enzymatic target within the parasite. The enzyme pteridine reductase 1 (LmPTR1) from L. major was investigated as a possible target. PTR1, an NADPH-dependent short-chain oxidoreductase, is crucial for the pterin salvage pathway in the parasite40. Leishmania parasites uniquely possess the ability to salvage pterins from the host, a capability absent in the host itself due to a lack of PTR1 activity and the formation of pterin analogues derived from GTP. Consequently, the PTR1 gene represents a significant therapeutic target for drug development21. This enzyme catalyzes the conversion of biopterin into its reduced forms, H2-biopterin and H4-biopterin, while also participating in the degradation of folate derivatives such as 7,8-dihydrofolate and tetrahydrofolate. Disruption of the PTR1 gene leads to the death of promastigotes during the insect stage of the parasite’s life cycle. However, this fatal effect can be partially reversed by supplementation with reduced pterins—but not folates—highlighting the essential role of H4-biopterin. Current studies propose that H4-biopterin plays a regulatory role in parasite differentiation. PTR1-deficient mutants exhibit significantly reduced levels of H4-biopterin, which triggers premature differentiation into the highly infectious metacyclic promastigote form within the sand fly vector41. PTR1 shows a much broader range of activity, catalyzing successive reductions of conjugated (folate) and unconjugated (biopterin) pterins. Under physiological conditions, PTR1 is responsible of reduction of 10% of the folic acid required by the cell, but when classical antifolate drugs inhibit. dihydrofolate reductase-thymidylate synthase (DHFR-TS), PTR1 can be overexpressed, allowing for a significant reduction of the necessary amounts of folates to ensure parasite survival. This compensatory mechanism suggests that treatment of trypanosomatid infections could be achieved through inhibition of both DHFR and PTR1 by a single drug or a combination of molecules that are specific inhibitors of the two targets. This evidence supports the importance of PTR1 for parasitic survival and indicates that it is a promising drug target for the treatment of Leishmania42. Nare, Luba, and Beverly reported that upon DHFR-TS inhibition, PTR1 provides enough folate to guarantee the parasite survival43. Furthermore, this enzyme is not present in mammalian hosts rendering it a promising target for drug development31. In some studies, the PTR1 inhibitory activity of thiazolidines and thiazolidinones was investigated. Linciano et al. reported the synthesis of 2-amino-1,3,4-thiadiazoles and evaluated their anti-leishmanial activity. Also, the most potent TbPTR1 inhibitors were assayed44. Leite et al. showed that thiazolidine-2,4-dione derivatives are micromolar inhibitors of LmPTR1 that display competitive behavior with the enzyme cofactor (NADPH)31. Neri et al. synthesized a series of thiazolidine-2,4-dione derivatives and evaluated against LmPTR130. Given the critical role of the PTR1 enzyme in the life cycle of the Leishmania parasite, as well as the history of previous studies that have shown the modulatory activity of thiazolidinone compounds against related targets, in this study, we investigated the stability and molecular interaction patterns of thiazolidinone derivatives with this key enzyme in leishmania. The catalytic site of PTR1 comprises a triad or tetrad of essential amino acid residues: tyrosine, asparagine, serine, and lysine residues. Among these, aspartate (such as ASP181) is noted to be particularly stable, similar to serine. ARG17 interacts with the pyrophosphate group of the cofactor, which contributes to the stabilization of the cofactor-enzyme complex. Moreover, PHE113 is crucial for ligand stabilization in the active site through hydrophobic interactions that anchor the substrate or inhibitor45,46. The presence of cofactor NADPH is essential for the binding of inhibitors or substrates to PTR147. PHE113 engages in hydrophobic interactions with the nicotinamide ring of NADPH, playing a key role in facilitating substrate recognition and catalytic activity. High conservation of the catalytic amino acids TYR191, ASP181, LYS198, and TYR194 indicates a consistent catalytic mechanism of PTR1 across multiple leishmania species22,45,46,48. To validate the docking approach, the co-crystallized ligand was re-docked into the PTR1 binding site, successfully reproducing the original pose from the PDB structure with the highest clustering a root-mean-square deviation (RMSD) of 1.89 Å46. All (E)-3-(((E)-N-benzylidene)amino)-2/4-substituted-thiazolidin-4-one derivatives (7a-7g) were docked into the LmPTR1 active site to determine their optimal binding conformations. Among these, compounds 7a and 7c exhibited the strongest activity and lowest free binding energies, which corresponded well with their in vitro results. As shown in Fig. 7, ligand 7a formed hydrogen bonds with TYR194 and the cofactor NADPH, interactions that are key to stabilizing and ensuring the specificity of binding within the active site. The ASP181 residue interacted with the ligand’s benzylidene group, likely through electrostatic or π-anion interactions. ASP residues often contribute to proton transfer and ionic stabilization during binding. Additionally, the sulfur atom of the thiazolidin-4-one ring engaged with NADPH. Residues such as LEU188, MET183, and TYR194 interacted with the aromatic portion of the ligand via alkyl and π-alkyl contacts, further strengthening the ligand–protein complex. Other residues near the binding site—including TYR191, THR184, ARG17, PRO187, SER111, LYS198, GLY225, GLN186, THR195, and PHE113—may also contribute through indirect, hydrophobic, or weak polar interactions.Fig. 7(A) 3D and (B) 2D interactions illustrating the binding of compound 7a to the active site residues of LmPTR1. (A) 3D and (B) 2D interactions illustrating the binding of compound 7a to the active site residues of LmPTR1. Compound 7c, which is the second most potent against the promastigote form after 7a and ranks third against the amastigote form following compound 7b, formed three hydrogen bonds with TYR194, LEU188, and the cofactor NADPH. LEU188 also engaged in a π-sigma interaction with the benzylidene moiety. Similar to compound 7a, the ASP181 residue interacted with the benzylidene group, likely through electrostatic or π-anion interactions. Additionally, the cofactor NADPH interacted with the pyridine ring via a π-lone pair interaction. Several residues—including LEU188, MET183, LEU226, LEU229, and TYR194—made alkyl and π-alkyl contacts with the aromatic portion of the ligand. Other residues near the binding site, such as THR184, ARG17, PRO187, SER111, LYS198, GLN186, THR195, SER227, VAL230, VAL228, and PHE113, may contribute to ligand binding through hydrophobic or weak polar interactions (Fig. 8).Fig. 8(A) 3D and (B) 2D interactions illustrating the binding of compound 7c to the active site residues of LmPTR1. (A) 3D and (B) 2D interactions illustrating the binding of compound 7c to the active site residues of LmPTR1. Compound 7b was the most potent compound after compound 7c against the promastigote form and ranked second following 7a against the amastigote form. TYR194 and the cofactor NADPH formed hydrogen bonds with compound 7b, specifically interacting with the methoxy group located at the meta position of the benzylidene moiety. This differs from compounds 7a and 7c, where these interactions occur with the carbonyl group of the thiazolidin-4-one ring. Additionally, the ASP181 residue interacted with the nitro group of the ligand, likely through an electrostatic interaction. Both NADPH and TYR194 were also engaged in alkyl and π-alkyl interactions with the ligand’s aromatic portion. Other nearby residues—including TYR191, THR184, ARG17, PRO187, SER111, LYS198, GLY225, GLN186, THR195, and MET183—may contribute to the binding environment. Notably, PHE113 interacted with the benzylidene moiety via π-π stacking (Fig. 9).Fig. 9(A) 3D and (B) 2D interactions illustrating the binding of compound 7b to the active site residues of LmPTR1. (A) 3D and (B) 2D interactions illustrating the binding of compound 7b to the active site residues of LmPTR1. Based on the detailed interaction data and supported by biological activity trends, compounds 7a-c exhibited distinct binding profiles that correlated with their anti-leishmanial activity. Compound 7a demonstrated the highest activity against the amastigote form, followed by 7b. While 7c showed superior activity primarily against promastigotes. Both 7a and 7c formed more specific hydrogen bonds with the core thiazolidin-4-one scaffold, whereas 7b’s interactions occurred mainly through its peripheral meta-methoxy substitution, which may result in weaker anchoring within the binding site. Compound 7c also formed an additional hydrogen bond with LEU188; however, its location on a ring substituent rather than the core scaffold might reduce its binding strength. All three compounds interacted with ASP181, but 7a and 7c engaged through the core aromatic benzylidene group. However, 7b relied on its nitro group, which may be less optimally positioned within the binding pocket. Notably, 7a exhibited a unique sulfur-NADPH interaction that might enhance complex stability, whereas 7c formed a π-lone pair interaction with NADPH, potentially a weaker interaction. In contrast, 7b’s interaction with NADPH is more peripheral, possibly contributing to its comparatively lower potency. All compounds showed π-alkyl or alkyl interactions with residues such as LEU188, MET183, and TYR194. Specifically, 7a formed π-alkyl interactions with LEU188 and TYR194, 7c showed a π-σ interaction with LEU188, and 7b uniquely included π-π stacking with PHE113. It may improve stability, but it is insufficient to surpass the stronger interactions observed in 7a and 7c. Finally, all three ligands are further stabilized by a network of neighboring residues, including TYR191, THR184, ARG17, PRO187, SER111, LYS198, GLN186, THR195, and PHE113, which likely contribute through indirect hydrophobic and weak polar interactions. Compound 7c, in particular, showed additional interactions with residues such as LEU226, LEU229, SER227, VAL230, and VAL228, potentially improving the characteristics of the binding site. The comprehensive analysis of compounds 7e and 7f focused on their binding interactions and corresponding biological activity. Compound 7e formed several interactions with key residues, but its overall structure appeared more dispersed compared to 7f. For example, portions like the aromatic ring are located farther from the core of the active site, which may lead to weaker binding energy and reduced stability of the ligand-enzyme complex. In contrast, compound 7f is fitted more compactly and cohesively within the binding pocket. Its aromatic ring is positioned closer to ASP181, likely allowing for π-anion interactions or even indirect hydrogen bonding mediated by water molecules. This better spatial arrangement enhances binding strength and increases the ligand’s residence time, which correlates with its higher observed anti-leishmanial activity. Both compounds interacted with ASP181, a negatively charged residue known to be catalytically or structurally important in leishmania enzymes. However, 7f engaged more closely and effectively with this residue, stabilizing the ligand within the active site. This stronger interaction likely reduced ligand mobility and prolonged binding duration, thereby enhancing the compound’s inhibitory potential49,50. Both compounds formed hydrogen bonds with the cofactor NADPH, but in compound 7f, these interactions were more central and stronger, anchoring the ligand more firmly within the binding pocket. Since disrupting cofactor binding is a known mechanism for enzyme inhibition, this enhanced interaction with NADPH likely contributes directly to the compound’s inhibitory potency. Additionally, compound 7f interacted effectively through hydrophobic interactions with residues like PHE113 and LEU188. Its aromatic structure facilitates π-π stacking and van der Waals interactions, further stabilizing the complex. In contrast, compound 7e lacked the optimal spatial arrangement for these interactions, which diminished its overall effectiveness (Fig. 10).Fig. 10(A) Conformations of 7e (A) and 7f (B) with LmPRT1 residues active site. (A) Conformations of 7e (A) and 7f (B) with LmPRT1 residues active site. A comparison of how compounds 7a, 7f, and 7e interacted at the molecular level with the LmPRT1 enzyme’s active site showed clear differences in their binding behavior. Compound 7a is notable for the most numerous and diverse interactions. These include hydrogen bonds with important residues such as the cofactor NADPH, TYR194, and SER111, which help increase binding specificity. It also engaged in electrostatic and π-π interactions with ASP181 and LEU188, crucial for inhibiting the enzyme and stabilizing its shape. Additionally, 7a fits more fully and complementarily within the binding pocket. In comparison, compound 7f kept some key interactions, especially with ASP181, LEU188, and NADPH, but it didn’t form hydrogen bonds with SER111 or TYR194. This limits both its binding strength and flexibility in orientation. Compound 7e showed the fewest interactions, mainly only bonding with NADPH and ASP181. It lacked stabilizing interactions such as SER111 or PHE113, which weakened its anchoring and reduced its anti-leishmanial activity. Three compounds formed interactions with the key catalytic residue Asp181. However, in 7a, this interaction was strengthened by additional supportive interactions, making the binding more stable. In 7f, the interaction is present but may be less stabilizing, while in 7e, the ASP181 interaction may be more stabilizing, diminishing its effect on ligand binding and activity. Furthermore, 7a covers the binding site most comprehensively, stabilizing both hydrophilic and hydrophobic areas. Compound 7f exhibited partial occupancy of the binding site, missing some hydrophilic interactions, and 7e accepted a more compact shape that engages fewer important residues. Overall, the binding mode of 7a aligns well with its stronger anti-leishmanial activity, as its interactions securely anchor and properly orient the molecule in the enzyme’s active site. Compound 7f showed moderate interactions and activity, while 7e’s limited binding corresponds with the weakest biological effect among the three. In what follows, the data will be analyzed for compounds 7d and 7g. Compound 7d showed greater anti-leishmanial activity compared to 7g. According to docking results, the ASP181 residue formed an electrostatic interaction with the nitro group of 7d. Additionally, TYR194 and LEU188 engaged in hydrophobic interactions with the compound’s central core, helping to strengthen its binding. PHE113 and the cofactor NADPH further stabilized the benzylidene moiety of 7d through π-π stacking and π-alkyl interactions (Fig. 11). Notably, 7d formed no hydrogen bonds with the enzyme residues. In contrast, compound 7g maintained π-anion and electrostatic interactions with ASP181. TYR194 and MET183 contribute via hydrogen bonding and van der Waals interactions. PHE113 formed π-anion and π-sulfur interactions, while NADPH interacted with 7g through both hydrogen bonding and π-cation interactions. The hydroxyl group on the para-benzylidene moiety of 7g formed a hydrogen bond with THR195, though this had a relatively minor effect on catalytic activity or strong ligand binding (Fig. 11). Overall, the superior anti-leishmanial effect of compound 7d can be attributed to its broader and stronger interactions with key catalytic residues, especially ASP181, NDP302, PHE113, and TYR194, which raise its binding affinity and inhibitory activity.Fig. 11(A) Conformations of 7d (A) and 7g (B) with LmPRT1 residues active site. (A) Conformations of 7d (A) and 7g (B) with LmPRT1 residues active site. Three structurally related compounds—7b, 7d, and 7g—were studied based on how they bind to the LmPRT1 active site. Their anti-leishmanial activities ranked as follows: 7b showed the strongest effect, followed by 7d, and 7g was the weakest. Compound 7b, which had the highest anti-leishmanial activity, formed a wide and varied network of interactions. These included strong π-π stacking with PHE113, π-anion interactions with ASP181, and several hydrogen bonds with key residues like TYR194, THR184, and the cofactor NADPH. It fit deeply and precisely within the binding pocket, indicating an optimal orientation and strong stabilization, likely explaining its potent inhibitory effect. In comparison, compound 7d showed moderate activity, exhibited fewer and spatially dispersed interactions mainly involving PHE113 and ASP181. It lacked significant π-π stacking or hydrogen bonds with important residues such as TYR194 and NADPH. The compound was partially positioned outside the active site, suggesting a less ideal fit and weaker binding affinity, which may account for its lower anti-leishmanial activity. Compound 7g displayed the weakest interaction profile. While PHE113 formed π-cation and π-sulfur interactions with 7g, this residue engaged in π-π stacking in both 7b and 7d. The hydroxyl group on 7g’s para-benzylidene moiety formed a hydrogen bond with THR195, though this interaction likely plays a minor role in catalysis or ligand binding. THR195 mostly participates through hydrophobic interactions with other compounds. Regarding the cofactor NADPH, it formed hydrogen bonds and π-alkyl interactions with 7b, π-alkyl interactions with 7d, and both hydrogen bond and π-anion interactions with 7g. Favorable hydrogen bonds and hydrophobic interactions were observed between the compound and important residues like Asp181, Phe113, Leu188, Gly225, Arg17, Tyr194, and Ser111, in addition to the cofactor NADPH. These interactions highlighted the strong potential of compounds 7a and 7c to bond to the catalytic activity of LmPTR146,51. From a spatial perspective, the inhibitor effectively mimics the pterin major group found in the standard antifolate agent methotrexate. It binds within the LmPTR1 active site through a sandwiched hydrophobic stacking arrangement. Additionally, its planar aromatic ring engaged in a π-π stacking interaction with the aromatic side chain of Phe113 that plays a key role in maintaining the ligand’s position within the enzyme’s active site46,48. TYR194, Asp181, and the cofactor NADPH play key roles in forming hydrogen bond interactions. In contrast, hydrophobic interactions primarily involve residues such as THR195, LEU229, PHE113, TYR191, MET183, and LEU188. Notably, PHE113 is also crucial for establishing a π-π stacking interaction, which plays a crucial role in stabilizing the ligand within the enzyme’s catalytic site40,45,46,52,53. Overall, the molecular docking results and anti-leishmanial activity of the compounds were correlated and confirmed each other. The integrated in vitro and in silico results provide multiple SAR insights for this series of compounds. Comparative analysis of compounds 7a-g clearly demonstrated that both the nature of the substituent on the thiazolidine-4-one nitrogen atom and the electronic properties of the benzylidene moiety play critical roles in determining anti-leishmanial activity. Among the tested derivatives, compounds 7a-c exhibited superior inhibitory effects against L. major promastigotes and amastigotes, with compound 7a showing the highest potency and selectivity index. Structurally, these compounds share a common benzylidene framework, while differences in the N-substituent markedly influence lipophilicity, polarity, and overall molecular stability. The enhanced in vitro anti-leishmanial activity of compound 7a can be attributed to the presence of a small and less polar methylimino group, which improves membrane permeability and facilitates efficient access to intracellular targets. These experimental findings are in strong agreement with the molecular docking studies, which revealed that the most active compounds adopt favorable binding orientations within the active site of LmPTR1. In particular, compounds 7a and 7c displayed the best binding energy values and formed stable hydrogen bonding and hydrophobic interactions with key amino acid residues involved in enzymatic activity. The docking poses indicated that EDG substituents on the benzylidene ring, such as methoxy groups, enhance π-π stacking and van der Waals interactions, thereby strengthening ligand–protein binding. Conversely, compounds bearing highly polar or bulky substituents showed weaker docking scores and less optimal interactions, which is consistent with their reduced in vitro anti-leishmanial efficacy. Overall, the strong correlation between anti-leishmanial activity and molecular docking predictions highlights the reliability of the docking model and confirms that rational structural modifications directly influence target engagement and antiparasitic potency. These results collectively support the proposed SAR model and provide a solid foundation for the future design of more potent thiazolidine-4-one-based anti-leishmanial agents. A summary, a suitable molecule for the enzyme includes hydrogen bond acceptors, a hydrophobic group, and an aromatic ring system. We can deduce certain aspects of the link between in vitro anti-leishmanial assay and molecular docking results for series: (I) favorable electronic substituents such as methoxy or bromine groups in the para-/meta-positions of the benzylidene ring, which improve π-π stacking and optimize electron distribution; (II) N-substituents like pyridine-3-yl that may form hydrogen bonds or coordinate with metal ions, enhancing binding affinity and target specificity; (III) Structural suitability that promotes efficient entry into cells and binding to intracellular molecules like enzymes or nucleic acids; (IV) an optimal balance between lipophilicity and hydrophilicity, promoting both solubility and membrane permeability. In contrast, compounds with bulkier groups, potent EWG (such as nitro), or N-substituents with significant steric hindrance were generally less active—likely due to reduced target accessibility or weaker binding interactions (Fig. 12).Fig. 12The pharmacophore features recognized as critical for complex interactions within the LmPRT1 active site. The pharmacophore features recognized as critical for complex interactions within the LmPRT1 active site. Molecular dynamics simulations study Molecular dynamics (MD) simulations was carried out to assess the stability and dynamic flexibility of the docking complexes over time in a water-based system. This strategy enables the improvement of docking results by accounting for the dynamic behavior of both the compound and the protein, and for a more accurate evaluation of ligand binding affinities. The MD simulations supported the docking outcomes, confirming the involvement of specific amino acids in key hydrogen and hydrophobic bonding interactions. Compounds 7a and 7c bound to LmPTR1, along with the unbound (apo) form of LmPTR1, were selected for simulation. Each system underwent 150 ns MD simulations. The obtained trajectories were subjected to further analysis using several parameters54. System stability and flexibility (root mean square deviation (RMSD)) Backbone RMSD values were analyzed over 150 ns MD simulations to assess the stability of compound binding within the LmPRT1 active site, based on predicted affinities. RMSD was used to monitor structural deviations and conformational changes of the complexes relative to their initial configurations. As shown in Fig. 13, throughout the 150 ns simulation, the apo structure of LmPRT1 remained stable, showing a consistent average RMSD of 2.16 Å. Both ligand-enzyme complexes also exhibited stable behavior. Notably, the 7a-LmPRT1 complex maintained the greatest structural stability, with an average RMSD of 1.96 Å, compared to the 7c-LmPRT1 complex, which showed a marginally higher RMSD of 2.25 Å. These findings indicated that the 7a-enzyme complex is more conformationally stable compared to both the apo protein and the 7c complex. The relatively small fluctuations observed for 7a-LmPRT1 indicated a stable binding conformation. In general, RMSD deviations for small globular proteins within the 1–3 Å range are typically considered to reflect stable conformations. Deviations beyond this range classically signify substantial changes in the protein’s conformation structure55. The candidate complexes demonstrated good structural stability, as shown by their average RMSD values, suggesting that the enzyme’s overall conformation remained consistent throughout the simulations. The results also confirmed that the ligands remained stably bound within the active site, with no significant displacement or dissociation from their original binding positions.Fig. 13RMSD plot of LmPRT1 with molecules 7a (Green), 7c (Red), and apo protein (Blue) throughout the simulation time. RMSD plot of LmPRT1 with molecules 7a (Green), 7c (Red), and apo protein (Blue) throughout the simulation time. As shown in Fig. 13, the RMSD plot for the 7a-LmPRT1 complex indicates that structural stability was achieved within the first 11 ns of the simulation. In comparison, the 7c-LmPRT1 complex reached a stable conformation slightly later, around 18 ns, followed by a modest increase in RMSD values between 106 and 150 ns. Importantly, the presence of the candidate compounds did not lead to any significant conformational changes in the enzyme. Furthermore, compound 7c exhibited RMSD values similar to the apo form of LmPRT1, while compound 7a consistently showed lower RMSD values than 7c, indicating greater structural stability. Both compounds exhibited stable interaction within the active site throughout the simulation period (Fig. 13). These observations align well with anti-leishmanial assay results, as lower RMSD values are often indicative of better ligand accommodation and stronger binding within the active site. Overall, the MD simulations data confirm the stability of the ligand–protein complexes. Similarly, the RMSD values of the ligands themselves were analyzed separately (Fig. 14). Both compounds 7a and 7c displayed stable RMSD profiles within the LmPRT1 active site, achieving stability after approximately 10 ns. These results indicated that the ligands maintained a stable orientation within the binding pocket during the simulation, maintaining consistent interactions with both the key residues and cofactor. Two complexes demonstrated low RMSD values with minimal fluctuations, further supporting their stable accommodation within the active site (Fig. 14).Fig. 14Overlay of RMSD profiles for ligands 7a (green) and 7c (red) bound to the LmPRT1 enzyme. Overlay of RMSD profiles for ligands 7a (green) and 7c (red) bound to the LmPRT1 enzyme. Fluctuations in the residual components (root mean square fluctuations (RMSF)) Root Mean Square Fluctuation (RMSF) analysis is used to identify flexible and rigid regions within an enzyme, offering insights into the dynamic behavior and stability of compound-enzyme complexes. This method reveals how atomic positions fluctuate over time, particularly in flexible regions of the enzyme during ligand binding56. This was assessed by analyzing RMSF values, which measure the deviation of atomic positions from their original orientations and reflect the extent of movement in the enzyme’s residues57. RMSF calculations for the complexes were based on the C-α atoms. The fluctuation levels for most residues remained below 3.0 Å, except for certain regions corresponding to loops or turns within the enzyme58. Regions such as loops, turns, and coils exhibit higher RMSF values compared to more rigid secondary structures like β-sheets and α-helices59. Higher RMSF values indicate greater flexibility and reduced stability in the binding region56. Figure 15 illustrates the RMSF profiles of the LmPRT1 complexes with ligands 7a and 7c, along with the apo protein. During the MD simulations, all complexes demonstrated structural stability and showed comparable fluctuation profiles. The mean RMSF values for all complexes remained below 1.5 Å. While certain regions showed higher fluctuations, others remained relatively stable. Notably, residues within the active site displayed RMSF values lower than 0.4 nm, representing that this region is densely packed, rigid, and maintains stability in the dynamic environment60. These results indicated that all complexes maintained stable and active interactions with the enzyme throughout the simulation. Both the apo protein and the ligand-LmPRT1 complexes exhibited higher fluctuations about residue regions 60–70, 100–120, 200–220, and 245–26446. The loop regions of LmPTR1 exhibited pronounced fluctuations during the simulations. Specifically, residues 60 to 70 showed higher fluctuations in all ligand-LmPRT1 complexes compared to the apo form. In the region covering residues 100 to 130, fluctuations were comparable across both complexes and the apo protein. Between residues 200 and 220, all complexes displayed lower fluctuations than the apo enzyme. Notably, at 245 to 250 amino acids, the 7a complex demonstrated the lowest fluctuations, whereas residues 250 to 264 exhibited the most fluctuations in the 7c complex. These observations suggest that ligand binding influences the mobility of specific amino acids.Fig. 15RMSF profiles of LmPRT1 residues over a 150 ns MD simulation for the apo form (blue) and complexes with ligands 7a (green) and 7c (red), with the apo LmPRT1 reference shown in purple. RMSF profiles of LmPRT1 residues over a 150 ns MD simulation for the apo form (blue) and complexes with ligands 7a (green) and 7c (red), with the apo LmPRT1 reference shown in purple. Importantly, the regions with significant fluctuations include key active site residues—TYR194, ASP181, LEU188, TYR191, and PHE113—which maintained low RMSF values (< 1.8 Å) and exhibited similar fluctuation patterns across all complexes. This indicates that both 7a and 7c bind effectively, causing minimal conformational changes relative to the apo form (Fig. 15). Overall, the enzyme in the apo state was more flexible than when bound to the compound. Furthermore, these fluctuating regions were consistently observed across all derivatives, highlighting their importance for ligand binding. The RMSF results align well with previous structural studies, including crystallographic data of LmPRT1 and its ligand-LmPRT1 forms45,46,53,61,62. Assessment of structural compactness (radius of gyration (Rg)) The system’s compactness was examined through analysis of the radius of gyration (Rg), which exhibits structural characteristics and folding behavior of the complexes during the entire simulation period. Higher Rg values are typically observed in expanded or unfolded conformations, whereas lower values are consistent with tightly packed and folded molecular states. Thus, Rg serves as a measure of structural packing and flexibility over time, as shown in Fig. 16. During the simulations, the complexes exhibited Rg values ranging between 1.8 and 1.9 nm. The average Rg values for the 7a-LmPRT1, 7c-LmPRT1 complexes, and the apo protein were 1.868 nm, 1.874 nm, and 1.856 nm, respectively. Throughout the MD runs, the Rg values remained relatively stable, showing no significant fluctuations. A slight decrease in Rg was observed for the ligand-LmPRT1 complexes, suggesting increased compactness and structural rigidity. Overall, these findings indicate that both the apo protein and the compound-enzyme complexes maintained stable conformations, with no notable differences in Rg fluctuations between the 7a and 7c complexes compared to the apo form.Fig. 16Backbone Rg profiles throughout the MD simulation for the apo protein (blue) and complexes with 7a (green) and 7c (red). Backbone Rg profiles throughout the MD simulation for the apo protein (blue) and complexes with 7a (green) and 7c (red). For compound 7a, a low Rg was observed, indicating a relatively compact ligand and confirming its stability when binding to the LmPRT1 protein. Compared to the apo form of the protein, both compounds 7a and 7c caused less enzyme compression during MD simulations. Compound 7a exhibited a stable Rg beginning at approximately 30 ns, followed by a gradual decrease by the end of the simulation period, with a slight fluctuation observed from 120 to 130 ns. Overall, throughout the MD simulations, 7a showed a greater Rg than the apo form of the protein. Similarly, 7c displayed a nearly constant Rg up to 45 ns, after which it decreased until the end of the simulation (Fig. 16). The Rg variation pattern closely followed the RMSD trends, indicating that these ligands cause minimal changes in the enzyme’s structural compactness. The small fluctuations in Rg values further support the stability of these complexes. The data indicate that the compounds exhibited firm interactions at the active site, as demonstrated by the relatively stable folded structures of all tested complexes during the final 80 ns of the simulation. Consequently, all complexes showed consistently strong Rg values, reflecting a tightly folded enzyme-compound structure. This suggests that the complexes enhance the enzyme’s rigidity and compactness, thereby improving its overall stability. Importantly, the Rg values for all complexes were not significantly different from those of the apo protein (Fig. 16), indicating that ligand binding does not disrupt the overall folding of LmPTR1. These observations align well with the assay results reported by Hardy46,63. The active site’s rigidity stops LmPTR1 from experiencing major conformational changes when a ligand binds. The RMSF and Rg results show that the active site remains stable, and interactions with the molecules don’t cause significant fluctuations in amino acids64. Solvent-accessible surface area (SASA) analysis The solvent-accessible surface area (SASA) was used to examine how solvent exposure, influenced by flexible packing, affects the stability of the complexes. SASA quantifies the extent of interaction between the solvent and the complexes, offering insights into how the complexes interact with surrounding solvent molecules. Additionally, SASA data evaluate the degree of conformational changes throughout the binding process. For example, when a foreign substance, such as a drug binds, it can increase the exposure of hydrophobic residues to the solvent. Overall, the average Rg and SASA values from the study indicated that the enzyme, when bound to the ligands, experienced only slight conformational changes—likely because the ligands occupied its active site65. The solvent molecule’s strong binding affinity to the ligand provides valuable insights into the dynamic flexibility and significant conformational transitions during the enzyme-ligand interaction66. A high SASA value indicates an enhanced enzyme surface area, while a lower SASA value reflects a more compact structure or reduction in surface exposure59. Fluctuations in SASA for all complexes are depicted in Fig. 17. The average SASA values for 7a and 7c bound to LmPRT1, as well as for the apo protein, were 136.64 nm2, 138.28 nm2, and 133.58 nm2, respectively (Fig. 17). These results indicated that all complexes exhibited SASA values closely resembling those of the apo protein. The consistently high and stable SASA values suggest that the compounds remain accessible to solvent molecules while the overall enzyme structure stays unchanged67.Fig. 17Backbone SASA profiles throughout the MD in the presence of ligands 7a (green), 5c (red), and the apo protein (blue). Backbone SASA profiles throughout the MD in the presence of ligands 7a (green), 5c (red), and the apo protein (blue). During the MD simulations, the SASA values for the apo protein and the complexes showed similar patterns. A comparable trend was observed for the apo protein and the 7a- and 7c-LmPRT1 complexes, with stability achieved after about 5 ns. In summary, decreased SASA values show reduced solvent-exposed surface area, which limits structural flexibility and leads to a more stable and compact configuration68. Based on the average SASA values calculated for the various complexes, the stability ranking is apo protein > 7a > 7c. The consistently low SASA values across all complexes suggest they remained stable throughout the simulation. Hydrogen bond analysis of complexes The structural integrity of the complex is preserved through various interactions, including hydrogen bonding, hydrophobic contacts, and electrostatic interactions. Among these, the contribution of hydrogen bonds is critical in stabilizing the structural conformation by supporting multiple types of interactions59. Intermolecular hydrogen bonding constitutes a fundamental interaction in the field of drug discovery. These hydrogen bonds play a crucial role in stabilizing ligand binding. The number of hydrogen bonds formed between molecules largely determines the rigidity and specificity of the protein–ligand interaction61. A higher prevalence of hydrogen bonding interactions typically translates to a stronger binding affinity between the enzyme and its ligand56. During MD simulations, hydrogen bonding patterns within each complex were systematically analyzed. Some hydrogen bonds formed, broke, and reformed over time. Figure 18 provides a visual representation of the hydrogen bond counts across the analyzed complexes. These transient hydrogen bonds, caused by the movement of both the enzyme and the compound, may contribute slightly to the compound’s affinity for LmPRT1. Compared to the 7a-LmPRT1 complex, the 7c-LmPRT1 complex showed higher hydrogen bonds. Overall, the findings displayed that both complexes form hydrogen bonds with LmPRT1, contributing to their stability. On average, the 7a and 7c complexes formed 0.51 and 1.07 hydrogen bonds, respectively. Moreover, the hydrogen bonds in both complexes remained stable throughout the simulation, indicating strong interactions between the enzyme and the compounds, which result in a very stable conformation.Fig. 18Hydrogen bond formation between the compound and the enzyme was monitored during the simulation, in the presence of 7a (green) and 7c (blue). Hydrogen bond formation between the compound and the enzyme was monitored during the simulation, in the presence of 7a (green) and 7c (blue). Principal component analysis (PCA) study Principal Component Analysis (PCA) is a widely used multivariate statistical technique for investigating the collective motions of biomolecular systems, particularly by capturing correlated fluctuations of Cα atoms during MD simulations. By reducing the dimensionality of the trajectory data, PCA allows the identification of dominant, large-scale motions that govern the overall conformational behavior of proteins and protein–ligand complexes69. In the present study, PCA was applied to the MD trajectories of the 7a-LmPRT1 and 7c-LmPRT1 complexes, as well as the apo form of LmPRT1, to evaluate their conformational sampling and dynamic stability throughout the simulation period. The eigenvalue spectrum revealed that the first few principal components (PCs) accounted for the majority of the total variance in the system, indicating that the essential dynamics of the protein can be effectively described by a limited number of collective motions. This observation suggests that the conformational space explored during the simulations is governed primarily by a small set of dominant structural fluctuations. The eigenvector projections for the 7a-LmPRT1 complex, 7c-LmPRT1 complex, and the apo protein are presented in Fig. 19. Comparative analysis of these projections demonstrates that the 7c-LmPRT1 complex undergoes substantially larger conformational rearrangements, reflecting enhanced structural flexibility. In contrast, the 7a-LmPRT1 complex exhibits the smallest amplitude of motion, indicative of a more rigid and conformationally stable protein–ligand assembly. Notably, the apo protein displays the highest degree of conformational variability, highlighting the stabilizing effect exerted by ligand binding on the overall protein structure.Fig. 19PCA results showing the eigenvector indices for 7a, 7c, and the apo protein throughout molecular dynamics simulations. PCA results showing the eigenvector indices for 7a, 7c, and the apo protein throughout molecular dynamics simulations. The two-dimensional (2D) and three-dimensional (3D) free energy landscapes (FELs) derived from principal component analysis provide valuable insight into the conformational stability and dynamic behavior of the LmPRT1 protein in complex with compounds 7a and 7c. In these representations, the first two principal components (PC1 and PC2) describe the dominant collective motions of the system, while the free energy surface reflects the relative stability of the sampled conformational states. For the 7a-LmPRT1 complex, the 2D FEL shows two well-defined and deep low-energy basins, characterized by compact purple regions surrounded by smooth contour lines. These deep minima indicate the presence of highly stable and thermodynamically favorable conformations. The relatively confined distribution of conformational states along PC1 and PC2 suggests that compound 7a restricts large-scale protein motions, leading to a structurally stable and rigid complex. The smooth energy contours between neighboring minima imply that transitions between these states occur through low-energy pathways, consistent with limited but controlled conformational flexibility. In contrast, the 7c-LmPRT1 complex exhibits a broader and more elongated low-energy region in the 2D FEL. Although a dominant minimum is still present, it is noticeably shallower and more diffuse compared to that observed for 7a. This distribution indicates that the protein samples a wider range of conformations when bound to compound 7c. The extended shape of the energy basin along PC1 suggests enhanced collective motion and greater conformational plasticity, allowing frequent transitions between multiple nearby metastable states. The 3D FELs further emphasize these differences. In the 7a-LmPRT1, the free energy surface displays steep walls surrounding deep and narrow valleys, reflecting strong energetic constraints that confine the system to a limited number of stable conformations. This topography is characteristic of a highly stabilized protein–ligand complex, where ligand binding reduces structural fluctuations and favors a native-like folded state. Conversely, the 7c-LmPRT1 shows a rugged but wider energy surface with multiple shallow depressions and gentler slopes. This indicates a multistable conformational ensemble, in which the protein frequently explores intermediate or partially rearranged states. The presence of multiple shallow minima suggests lower energy barriers between states, enabling faster and more frequent conformational transitions. Taken together, the FEL analysis clearly demonstrates that compound 7a induces greater structural stability and conformational restriction in LmPRT1, as evidenced by deep, localized energy minima and limited conformational sampling. In contrast, compound 7c promotes enhanced flexibility and dynamic heterogeneity, reflected by broader energy basins and a more dispersed conformational landscape. These differences have important functional implications. The increased rigidity observed for the 7a-LmPRT1 complex may favor stronger binding affinity and longer residence time, while the higher flexibility associated with the 7c-LmPRT1 complex may facilitate dynamic rearrangements, potentially affecting binding kinetics, allosteric communication, or ligand dissociation. Overall, the FEL profiles highlight distinct thermodynamic and kinetic behaviors induced by the two compounds, underscoring their different modes of interaction with LmPRT1 (Figs. 20 and 21).Fig. 20Two-dimensional free energy landscape (FEL) projected onto the first and second principal components (PCA1 and PCA2) of the system’s conformational ensemble.Fig. 21Three-dimensional free energy landscape (FEL) projected onto the first and second principal components (PCA1 and PCA2) of the system’s conformational ensemble. Two-dimensional free energy landscape (FEL) projected onto the first and second principal components (PCA1 and PCA2) of the system’s conformational ensemble. Three-dimensional free energy landscape (FEL) projected onto the first and second principal components (PCA1 and PCA2) of the system’s conformational ensemble. The Ramachandran plot of complexes The Ramachandran plot serves as a visual tool to map the φ (phi) and ψ (psi) dihedral angles of amino acid residues in the enzyme’s conformational framework. These angles describe how the amino acids in a polypeptide chain rotate around the bonds connecting them. The plot helps identify allowed and disallowed regions for these angles based on steric hindrance and other structural constraints. Assessing the stereochemical quality of an enzyme’s structure relies heavily on the Ramachandran plot. The absence or scarcity of points in certain regions signifies conformations that are energetically prohibited due to steric clashes or unfavorable torsional strain. Densely populated areas correspond to energetically favorable conformations of the enzyme backbone. The plot is divided into three main regions: (I) Most preferred area: This region includes the most energetically favorable combinations of ψ (psi) and φ (phi) angles, typically associated with common secondary structures like beta sheets and alpha helices. (II) Acceptable area: Torsion angles in this area are energetically allowed but occur less commonly than those within the preferred conformational zone. (III) Outlier area: This area characterizes energetically unfavorable angle combinations. Points here may indicate potential structural problems, including steric hindrance or distorted bond angles and bond lengths70,71. The Ramachandran plot displayed different shaded regions on the ψ-ϕ plot for the complexes (Fig. 22). The maximum favored areas (A–C) accounted for over 90% of the residues, while the permitted areas (a-d) made up about 8%46.Fig. 22More than 90% of amino acids fall within the core region of the active site in LmPRT1, as shown by the Ramachandran plots of the complexes and apo protein. More than 90% of amino acids fall within the core region of the active site in LmPRT1, as shown by the Ramachandran plots of the complexes and apo protein. Binding free energy analyses using the MM/PBSA method Binding free energy calculations based on molecular simulations represent one of the most reliable approaches for validating ligand–protein interactions with favorable thermodynamic profiles. In this study, the binding free energies of selected compounds in complex with LmPTR1 were calculated using the MM/PBSA method based on the last 10 ns of the MD trajectories (Table 7). The investigated ligands included 7a and 7c. All compounds exhibited negative binding free energy values, indicating favorable binding to LmPTR1.Table 7Binding free energy (KJ/mol) for selected two compounds.ComplexΔUeleΔUvdWΔSASA/nm2ΔGbinding7a− 6.47− 166.03− 15.78− 121.067c− 6.95− 132.35− 16.69− 86.09 Table 7Binding free energy (KJ/mol) for selected two compounds.ComplexΔUeleΔUvdWΔSASA/nm2ΔGbinding7a− 6.47− 166.03− 15.78− 121.067c− 6.95− 132.35− 16.69− 86.09 Binding free energy (KJ/mol) for selected two compounds. The total binding free energy of 7a was calculated as ∆Gbinding = − 121.06 kJ/mol in contrast, the compound 7c showed ∆Gbinding = − 86.09 kJ/mol. 7a indicated significantly stronger binding affinities compared to 7c. The enhanced binding affinity of 7a is primarily attributed to its strong van der Waals (− 166.03 kJ/mol) and slightly electrostatic (− 6.47 kJ/mol) interaction energies, resulting in a tightly bound complex (Table 7). Based on these results, the 7a-LmPRT1 complex exhibited substantially higher stability than the 7c-LmPRT1 complex. Furthermore, favorable contributions from electrostatic interactions, van der Waals forces, and nonpolar solvation energy, along with a lower polar solvation penalty and a more negative nonpolar desolvation contribution, play critical roles in strengthening the interaction between 7a and LmPRT1. These findings confirm that 7a forms a stable and energetically favorable binding mode within the active site, interacting effectively with key residues. Additionally, two compounds exhibited higher SASA than electrostatic interactions upon binding to LmPRT1, indicating that larger portions of this ligand are buried within the hydrophobic pocket of the binding site. Analysis of the free energy components reveals that van der Waals interactions are the dominant contributors to ligand binding. Consistent with molecular docking and MD simulations results, the binding pocket is predominantly hydrophobic, as further supported by the MM/PBSA calculations, where the van der Waals energy (∆E_vdw) exceeds the electrostatic contribution (∆E_ele). Overall, the MM/PBSA binding free energy analysis is in good agreement with the molecular docking and molecular dynamics simulation results. Conclusion In this study, a novel one-pot, two-step, three-component approach was developed to synthesize a series of thiazolidin-4-one analogues with anti-leishmanial activity. Biological evaluation identified compounds 7a and 7c as the most promising derivatives, showing strong activity against L. major and low cytotoxicity toward host macrophages, while 7a and 7b were particularly effective against the amastigote stage. SAR analysis highlighted the importance of electronic effects and appropriate N-substitution in enhancing anti-leishmanial potency. Apoptosis assays, together with in silico ADME predictions, supported a targeted mechanism of action and favorable drug-likeness. Molecular docking and molecular dynamics simulations further confirmed stable interactions of the lead compounds with Lm PTR1. Overall, these findings introduce thiazolidin-4-one derivatives as novel lead scaffolds for anti-leishmanial drug development. Future work will focus on structural optimization, in vivo validation, and mechanistic confirmation to advance these compounds toward therapeutic application. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1