Lead optimization against drug-resistant Leishmania donovani infection: Semi-synthetic derivatives of ethyl linoleate isolated from Indian edible mushroom Meripilus giganteus

Leishmania donovani , Drug-resistant infection, Semi-synthetic derivatives, 50 Trypanothione reductase, Mushroom. 51 52 Introduction. 53 Infectious diseases cause large-scale mortalities despite remarkable discoveries and 54 advancements in modern medical sciences worldwide. Vector -borne infectious diseases like 55 leishmaniasis lack effective vaccines and hence pose a serious threa t of major outbreaks with 56 increasing incidence every year. Visceral leishmaniasis (VL), the deadliest form of 57 leishmaniasis, accounts for an estimated 50000 -90000 new cases annually, worldwide (1). In 58 the year 2023, several new cases of VL were reported to the WHO from India, Ethiopia, 59 Kenya, Somalia, South Sudan, China, Brazil, Eritrea, and Yemen, and this accounted for only 60 25 to 45% of actual cases that occurred across the globe (2). Certain chemotherapeutics hold 61 the gates but lack absolute remission fo r many reasons. Long -term treatment regimens and 62 irregular patient follow -ups lead to the emergence of resistant strains. Several instances of 63 drug resistance have already been reported for sodium antimony gluconate (SAG) from the 64 Indian state of Bihar (3). Drug resistance in Leishmania results from a high rate of drug 65 efflux attributed to the over -expression of membrane -associated proteins like ABC 66 transporter- MRPA or due to the scavenging and pro -parasitic activities of the enzymes of 67 polyamine biosynthetic pathways (4). Liposomal Amph otericin B, the present first-line 68 chemotherapy against VL , is not cost -effective (5). Considering all these drawbacks, we 69 propose naturally occurring purified products and their derivatives as potential candidates for 70 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 4 future drug discovery. Previously, we have shown that a triterpenoid, astrakurkurone, 71 isolated from Astraeus hygrometricus, could inhibit the proliferation of L. donovani and 72 protect the host from experimental VL by inducing immunity (6, 7). A semi -purified 73 carbohydrate fraction from the same mushroom showed an anti -parasitic response by 74 inducing pro -inflammatory cytokines (TNF -α, IL -12, iNOS2) in hosts (8). So, mushrooms 75 can be a good source of potential anti-inflammatory molecules; extracts were previously cited 76 for their anti-viral, anti-bacterial efficacy (9). In the present study, we aimed to isolate active 77 principles from the wild mushroom M. giganteus and synthesise chemical derivatives with 78 the aim of drug development against both the drug-sensitive (10, 11) and the drug-resistant L. 79 donovani infection (12,13). Based on our preliminary observation, a semi -purified ethyl 80 acetate fraction of M. giganteus was found to show a high inhibitory effect against L. 81 donovani (Fig. S1B), it was further purified through HPLC, resulting in six distinct peaks on 82 the HPLC chromatogram (Fig. S1D). Peak 5, characterized as ethyl linoleate, was found most 83 efficient against the promastigotes of drug -sensitive and drug -resistant L. donovani (Fig. 84 S1E). Targeting for better anti-leishmanial efficacy, chemical derivatization of ethyl linoleate 85 (Peak 5) was performed by controlled oxygenation that yielded five derivatives , (EL2-EL6) 86 (Fig. 1). Interestingly, monoepoxides of ethyl linoleate (EL3) were found to inhibit the 87 proliferation of both promastigote and amastigote morphs of drug-sensitive as well as drug -88 resistant L. donovani, in vitro, most significantly (Fig. 2, 3, Table S3, S4, S7A, B). 89 The background study was intended for us to focus on some questions to identify EL3 as a 90 promising anti-leishmanial lead. The first question was the route of administration and the 91 dose selection. The second question was the mechanism of action; does EL3 h ave any 92 specific target on the intracellular parasites in the host? Does it induce the pro -inflammatory 93 responses of the hosts, leading to parasite clearance? The third question was regarding the 94 toxicity towards hosts, if any. Finally, the bioavailability of EL3 in the host system was 95 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 5 investigated, as the longer persistence of drug molecules inside the host system promotes 96 resistance. 97

98 Isolation of active molecules from Meripilus giganteus. 99 Among the different solvent -extracted fractions, t he significant in vitro bioactivity of the 100 ethyl acetate -extracted fraction against the drug -resistant strain of Leishmania (Fig. S1C) 101 prompted us to bioactivity-guided isolation of pure molecules from this fraction. Based on 102 retention times in the optimized mobile phase, six distinct peaks were isolated from the 103 bioactive fraction of the ethyl acetate extract, as depicted in the chromatogram (Fig. S1D), 104 through multiple injections and run sequences using a semi -preparative column. Bioactivity 105 assays against L. donovani promastigotes indicated that fraction 5 exhibited the highest 106 activity. After confirming their bioactivity (Fig. S1E), we proceeded with the structural 107 determination of the active compound (s). Simple ¹H and ¹³C NMR spectral analysis 108 confirmed that, as anticipated, the active fractions consisted of pure single compounds rather 109 than mixtures. 110 The structure of active compounds obtained from peak 5 was determined by thorough 111 analysis of infrared, mass spectr ometric (HRMS), and nuclear magnetic resonance (1D and 112 2D NMR) spectroscopic data. A quartet peak at δ 4.15 ppm for two protons (2H) and a triplet 113 peak for three protons (3H) at δ 0.9 ppm in 1H NMR spectrum of the pure compound 114 obtained from peak 5 suggest ed the presence of an ethyl ester. The carbon peak at δ 174.5 115 ppm in 13C NMR spectra supported the presence of ester carbonyl carbon ( Fig. S2 ). The 116 presence of a strong, sharp carbonyl peak at 1735 cm -1 in the IR spectrum further strongly 117 supports the presence of the ester group. The peak at δ 5.3 ppm in 1H NMR for four protons 118 (4H) suggested the possibility of the presence of multiple similar double bonds with similar 119 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 6 chemical environments, which are probably in coupling with multiple aliphatic protons. Th e 120 appearance of multiple protons in 1H NMR spectrum and multiple methylene (-CH2) peaks at 121 DEPT 90 spectrum indicated the presence of multiple methylene groups in the molecule (14). 122 The correlations among protons in the molecule in correlation spectroscopy (COSY) and C-H 123 correlation data obtained from the HMQC spectrum of the compound suggest that the 124 compound was a long -chain fatty acid ester with two isolated double bonds (15,16). Space 125 integration in NOESY among allylic protons suggests that these double bonds are in cis 126 orientation (17). All the above observations, along with the electrospray ionized mass 127 spectrometric (ESI) molecular ion peak at m/z 308.2725, corresponding to th e molecular 128 formula C 20H36O2, led us to determine the structure of the isolated compound in peak 5 as 129 ethyl linoleate EL1, a known di -unsaturated long -chain fatty acid ester (Fig. S2). The 130 determined structure of the identified compound was further confirmed by comparing it with 131 the reported proton and carbon NMR spectra of ethyl linoleate. 132 Synthesis of a library of polar derivatives of EL1 with increased aqueous solubility. 133 As EL1 (Fig. S2) is a di-unsaturated ester, we could make it comparatively polar by 134 hydrolyzing it to linoleic acid EL2 (Fig. S3), through base-catalyzed hydrolysis using sodium 135 hydroxide solution, without affecting the olefins present in the molecule (18). The successful 136 hydrolysis of EL1 is evident from the disappearance of the characteristic ethoxy ( -OEt) 137 quartet peak at δ 4.12 ppm for two protons (2H) and the triplet peak at δ 0.90 ppm for three 138 protons (3H), in 1H NMR spectra of product EL2. Considering dihydroxylation of the 139 existing double bonds of EL1 could be an effective approach to generate more polar 140 derivatives with enhanced aqueous solubility, we planned to carry out dihydroxylation of 141 both the existing double bonds. However, when w e tried to carry out cis -hydroxylation of 142 both the double bonds of EL1 using osmium tetroxide and N-methyl morpholine oxide, it 143 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 7 failed to react to generate the desired tetrahydroxylated product (19 ). The failure of the direct 144 di-hydroxylation method using osmium tetroxide on EL1 forced us to look for an alternative 145 approach. Our endeavor to epoxidize the olefins of EL1 using 1.5 and 2.5 equiv . of meta -146 chloroperbenzoic acid ( mCPBA) in dichloromethane (DCM) at room temperature, 147 successfully resulted in the formation of mono -epoxy EL3 (Fig. S4) and di-epoxy derivative 148 EL4 (Fig. S5) (20). Confirmation of the formation of monoepoxide EL3 was evident from the 149 reduction in the number of olefinic protons at δ 5.38 ppm from four (in the case of EL1) to 150 two protons and an upfield shift of two protons to δ 2.91 ppm due to the characteristic 151 anisotropic shielding of epoxide. Similarly, the confi rmation of the formation of EL4 was 152 asserted from the disappearance of all four olefinic proton peaks and the upfield appearance 153 of these protons at δ 3.24–3.04 and 2.98 ppm as two multiplets due to the epoxide anisotropy. 154 The diepoxide EL4 was then subjected to acid -catalyzed hydrolysis using dilute sulfuric acid 155 (4%) in an acetonitrile/water (1:1) mixture, yielding the tetrahydroxylated long -chain ethyl 156 ester EL5 (Fig. S6). The formation of EL5 was supported by the expected downfield shift s of 157 the epoxide-associated protons from δ 3.24 –3.04 and 2.98 ppm in EL4 to δ 3.30–4.3 ppm in 158 the ¹H NMR spectrum of EL5. To further enhance polarity, we hydrolyzed the ester group of 159 EL5 using aqueous NaOH, converting it into the corresponding tetrahydrox y acid EL6 (Fig. 160 S7). The successful ester hydrolysis was confirmed by the selective disappearance of the 161 ethoxy peak in EL5, indicating the formation of EL6. 1H and 13C NMR spectra, structure s, 162 and synthesis of all derivatives have been provided in the supplementary file (Fig. S3-S7). 163 Activity-guided selection of the bioactive synthetic derivative against L. donovani 164 promastigotes, in vitro. 165 EL1 (IC50 for AG83 promastigotes, 80.3 ± 2.6 µg/ml, p<0.05 vs. DMSO treated control; IC50 166 for NS2 promastigotes, 176.1 ± 4.4 µg/ml, p<0.05 ), EL2 (IC50 for AG83 promastigotes 79.4 167 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 8 ± 2.9 µg/ml, p<0.05; IC50 for NS2 promastigotes 161.9 ± 5.9 µg/ml, p<0.05 ), and EL3 (IC50 168 for AG83 promastigotes, 77.6 ± 3.2 µg/ml, p<0.05; IC50 for NS2 promastigotes, 99.8 ± 5.7 169 µg/ml, p<0.05 ), vs. DMSO treated control, have been identified among the five derivatives 170 (EL2-EL6) of Peak 5/ EL1 as these synthetic derivatives exhibited potential anti -171 promastigote activity (Fig. 2A, Table S3-S5). Interestingly, the rate of inhibition by EL3 was 172 found higher against the drug-resistant NS2 (49.5±2.8% of inhibition by highest dose, p<0.05 173 vs. control) when compared with the Peak 5 (48.1±1.4%, (p<0.05 vs. control for NS2), 174 originally obtained from M. giganteus, or EL1 (28.5±0.8%, p< 0.05 vs. control for NS2), a 175 synthetic analogue of Peak 5 (Fig. 2A). 176 EL3 showed significant anti-amastigote activity with the least toxicity, in vitro. 177 EL1, EL2, and EL3 were chosen for further experiments for anti-leishmanial screening 178 against the amastigotes, the pathogenic morphs in mammalian hosts. The dose kinetics 179 experiment against the drug-sensitive amastigotes confirmed that EL3 (IC 50: 62.9±2.5 µg/ml, 180 p<0.001 vs. control) wa s the most efficient in inhibiting the amastigotes in infected murine 181 peritoneal macrophages in comparison to Peak 5 (IC 50: 87.7±1.4 µg/ml, p<0.001 vs. control), 182 EL1 (IC50: 90.1±3.3 µg/ml, p<0.001 vs. control) , and EL2 (IC 50: 132.5±1.4 µg/ml, p<0.001 183 vs. control). EL3 was also shown to have the most significant anti-amastigote activity against 184 the drug-resistant parasites (IC50: 43.8±0.6 µg/ml, p<0.001 vs. control) in comparison to Peak 185 5 (IC 50: 87.7±1.3 µg/ml, p<0.001 vs. control), EL1 ( IC50: 94.4 ±1.9 µg/ml, p<0.001 vs. 186 control), and EL2 (IC 50 65.7±5.8 µg/ml, p<0.001) (Fig . 3A and Table S7A, B). As EL3 has 187 been identified as the most promising synthetic derivative , in vitro, we have progressed with 188 further experiments, in vivo, leading to the development of a successful anti-leishmanial lead. 189 EL3 reduced the drug-resistant L. donovani infection in visceral organs more efficiently. 190 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 9 The epoxygenated derivative EL3 inhibited the parasite proliferation of both drug-sensitive 191 and drug-resistant L. donovani infection dose-dependently, in vivo. The highest doses of 500 192 µg/kg body weight and 1mg/kg body weight EL3 were found to be more effective against the 193 drug-resistant L. donovani infection in comparison to infected control and, interestingly, 194 drug-sensitive infection (Fig. 3B and Table S9). The dose of 500 µg/kg body weight inhibited 195 drug-resistant L. donovani infection by 86.7 ± 0.1% (p<0.001 vs. DMSO -treated control) and 196 by 77.3 ± 0.3% (p<0.001 vs. DMSO-treated control) in the spleen and liver, respectively. The 197 higher dose of 1mg/kg body weight was found to be more efficient in inhibiting the parasite 198 proliferation by 90 ± 0.1% (p<0.001 vs. DMSO-treated control) and by 83.3 ± 0.3% (p<0.001 199 vs. DMSO- treated control) in the spleen and liver, respectively (Fig ure 3B and Table S9). 200 This observation opens the possibility of target -specific activity of EL3 on drug -resistant 201 parasites. 202 EL3 effectively downregulates the polyamine biosynthesis pathway of parasites with a 203 higher efficacy against drug-resistant infection. 204 There was a significant decrease in the expression of Leishmania survival enzymes, 205 trypanothione reductase (TR), glutathione s ynthetase (GS), and γ-glutamyl cysteine 206 synthetase (γ -GCS) transcripts for both drug -sensitive and drug -resistant amastigotes in 207 infected spleens with respect to normalized L. donovani 18s rDNA as the housekeeping gene 208 following treatment. More interestingly, EL3 showed a con sistent pattern of higher efficacy 209 in inhibiting the expression of GCS, TR, and GS mRNA of drug -resistant amastigotes 210 when the infected animals were treated in vivo. The fold change of trypanothione reductase at 211 the transcript level was found to be decreased by 50 ± 0.003-fold (p<0.001) in the case of 212 drug-resistant amastigotes, higher than the case of drug-sensitive amastigotes (10 ± 0.06-fold, 213 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 10 p<0.001) (Figure 4A and Table S10). In contrast, EL3 could not dampen the expression of 214 host-specific γ-GCS, GS mRNA expressed in the infected spleen (Figure 4B and Table S11). 215 Three-dimensional structure prediction of 3 potential targets, γ -glutamyl cysteine 216 synthetase (γ-GCS), glutathione synthetase (GS), and trypanothione reductase (TR). 217 The three-dimensional structures of γ -GCS, GS, and TR from L. donovani were successfully 218 generated using a combination of AlphaFold2 Colab and MODELLER. Validation with 219 PROCHECK confirmed that the modeled structures are within acceptable stereochemical 220 parameters, ensuring that these structural models can be reliably used fo r further in -depth 221 analyses. Post -validation, the AlphaFold2 -generated γ -GCS model and the MODELLER -222 generated models for glutathione synthetase and trypanothione reductase were selected for 223 further studies. The AlphaFold2 model of γ -GCS showed a high confi dence level, indicated 224 by a pLDDT score of 89.9. This score indicates reliable prediction accuracy and structural 225 confidence. In contrast, the MODELLER -generated structures for glutathione synthetase and 226 trypanothione reductase showed significant sequence homology with the known crystal 227 structures 2WYO (Trypanosoma brucei glutathione synthetase) and 2JK6 (trypanothione 228 reductase from L. infantum ), sharing 46.7% and 98.37% sequence identity, respectively. 229 These sequence identities suggest a better alignment of the predicted modeled structures with 230 experimentally validated templates, particularly for trypanothione reductase, which closely 231 resembles its reference structure. The sequence identity and validation results underline the 232 structural integrity of these models, which were leveraged for ligand binding studies, 233 molecular dynamics simulations, and other functional analyses. 234 Binding affinity of EL3 with γ-GCS, GS, and TR. 235 To evaluate the effectiveness and inhibition potential of the EL3 against γ -glutamyl cysteine 236 synthetase (γ -GCS), glutathione synthetase (GS), and trypanothione reductase (TR) , 237 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 11 molecular docking studies were conducted using LeDock. From the 100 docking 238 conformations or poses generated for each protein -EL3 complex, the lowest binding energy 239 pose was selected for intermolecular interaction analysis. The results demonstrated that the 240 EL3 exhibited binding energies of −3.15 kcal/mol, −3.52 kcal/mol, and −4.52 kcal/mol with 241 γ-GCS, GS, and TR, respectively. The negative binding energy values indica te that EL3 can 242 effectively bind within the receptor cavities of γ -GCS, GS, and TR. For γ -GCS, the EL3 243 interacted through hydrophobic contacts with residues Asp470 and Gln473, formed a 244 hydrogen bond with Arg494, and established a salt bridge with Lys483, h ighlighting its 245 affinity for the binding site (Fig. 5A). In GS, the EL3 binding was stabilized primarily by 246 hydrophobic interactions with Ile172, Leu491, and Val587 (Fig . 5B). In TR, hydrophobic 247 interactions were noted with Val34, Val46, Thr51, Thr334, and Ala337, facilitating ligand 248 binding within the binding site (Fig. 5C). 249 MD simulations of three protein-EL3 complexes. 250 The molecular dynamics (MD) simulation results for γ -GCS, GS, and TR over a 100 ns 251 simulation are illustrated in graphical format. The en ergy value indicates that all proteins 252 were stabilized quickly, with GS achieving the lowest energy, suggesting a more stable or 253 lower-energy conformation (Fig. 6A). The root -mean-square deviation (RMSD) plot shows 254 the stability of each protein -EL3 complex, where γ -GCS and TR maintain relatively stable 255 RMSD values of 0.38 nm and 0.91 nm, respectively, indicating minimal structural deviations. 256 However, GS exhibits higher RMSD deviations, reaching an average of 1.91 nm, suggesting 257 greater structural instability in the ligand binding region throughout the simulation time (Fig . 258 6B). The root -mean-square fluctuation (RMSF) aligns with these observations; GS shows 259 more prominent fluctuations across amino acid residues with an average RMSF of 0.22 nm, 260 compared to the lower values observed for γ-GCS (0.09 nm) and TR (0.12 nm) (Fig. 6C). 261 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 12 The radius of gyration (Rg) values reveals the compactness of the proteins, with γ -GCS and 262 TR both averaging 2.48 nm, signifying stable, compact structures, while GS has a slightly 263 higher Rg value of 2.73 nm, indicating a less compact structure (Fig . 6D). The solvent -264 accessible surface area (SASA) results further indicate distinct protein -EL3 exposure to the 265 solvent environment, with glutathione synthetase having the highest surface ar ea value 266 (339.63 nm²), suggesting increased solvent interaction, whereas γ -GCS and TR show lower 267 SASA values of 373.64 nm² and 260.89 nm², respectively (Fig. 6E). 268 Essential dynamics (ED) refers to the application of principal component analysis (PCA) to a 269 protein trajectory, allowing the extraction of essential motions from the movement of the 270 protein molecule. This approach was employed to explore the conformational space of the 271 three target proteins. In the principal component analysis, the trace of covar iance values 272 reflects the overall motion within the complexes. GS shows the largest trace (51.97 nm²), 273 implying extensive conformational changes, whereas γ -GCS and TR have much smaller 274 traces (12.06 and 11.11 nm², respectively), consistent with more restri cted movements (Fig . 275 6F). These results collectively demonstrate that GS exhibits the highest flexibility and 276 structural variation, while γ -GCS and TR display more stable conformations under MD 277 conditions. This variation in structural dynamics could be rel evant to the binding affinities 278 and functional roles of these proteins. 279 Comparative binding free energy study by MM-PBSA. 280 The binding free energy calculations for the EL3 ligand with γ -GCS, GS, and TR yielded 281 values of −12.31±7.92 kcal/mol, −15.41±3.63 kcal/mol, and −29.56±9.28 kcal/mol, 282 respectively, indicating varied binding affinities across the three target proteins. Amon g the 283 three complexes, TR exhibited the most favorable binding energy, suggesting a strong and 284 stable interaction with EL3, while γ -GCS had the least favorable binding energy. The lower 285 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 13 binding energy for TR may be attributed to more optimal binding interactions, likely resulting 286 in a more stable protein -ligand complex relative to the others. The intermediate binding free 287 energy observed for GS reflects a moderate affinity for EL3, which is stronger than that of γ -288 GCS but still less favorable than TR. 289 EL3 did not display any hepato- or nephrotoxicity in the hosts. 290 Evaluation of toxicity, post-treatment in vivo, was assessed through the estimation of serum 291 creatinine and aspartate transaminase (AST), alanine transaminase (ALT), and alkaline 292 phosphatase (AP) l evels. Serum creatinine levels spiked in the infected, untreated control 293 animals (1.3 ± 0.07mg/dL, p<0.001 for AG83 IM and 1.67 ± 0.21mg/dL, p<0.001 for NS2 294 IM) as compared to the uninfected group (0.83 ± 0.02mg/dL). However, creatinine level was 295 reduced greatly and found to align with the normal range - 0.8 ± 0.11 mg/dL for AG83 and 296 0.79 ± 0.07mg/dL for NS2 -infected mice by the 1mg/kg body weight dose. Similar patterns 297 were also observed in AST, AP, and ALP levels in a dose -dependent fashion (Fig. 7 and 298 Table S12). 299 EL3 induced pro-inflammatory responses in the host, in vivo. 300 Besides the disruption of the pro-parasitic thiol synthesis pathway, EL3 also induced the 301 expression of the pro -inflammatory cytokines IL -12, TNF-α, IFN-γ, and IL -6 that instigate 302 and amplify the innate immune responses in the host systems following parasite infection. 303 TNF-α and IFN-γ were found up-regulated when the infected animals were treated with EL3 , 304 which are highly instrumental in generating nitric oxide in macrophages (21). EL3 also 305 elevated the IL -6 dose -dependently, which is essential in resisting drug -resistant infection 306 (Fig. 8) (22). 307 EL3 had a low serum bioavailability with maximum retention at 12 hours. 308 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 14 The bioavailability of this compound in the serum of the host system was also checked at 309 different time points, i.e. , 1, 3, 6, 12, 24, 48, and 72 hours, post single intramuscular 310 administration of 1 mg/kg body weight of the drug to naïve BALB/c mice. During the 311 standardization process, the retention time for EL3 was initial ly set at 8.4 minutes at a 312 wavelength of 275 nm. The resulting chromatogram showed no rise at the early 1 -hour time 313 point, but the concentration gradually increased; an initial concentration was detected at 3 314 hrs. post -injection, of 1.41 ± 0.1µg/ml (p<0.00 1 vs. 1 hr. mice serum) and 12 hours to a 315 maximum of 6.8±0.2 µg/ml (p<0.001 vs. 1 hr. mice serum) post -intramuscular injection. The 316 concentration then decreased to 1.3±0.1µg/ml (p<0.001 vs. untreated mice serum) at 48 hours 317 for the untreated mice. No detection of EL3 was found at 1 hr and 72 hr. serum samples (Fig . 318 9, Table S13). 319 Discussion. 320 The structure-activity relationship (SAR) investigation reveals that even minor modifications 321 to the ethyl linoleate analogue (EL1) can significantly influence anti-leishmanial efficacy and 322 pharmacokinetic behaviour. While increasing polarity by introducing additional hydroxyl 323 groups in EL4, EL5, and EL6 enhanced aqueous solubility, this modification unexpectedly 324 reduced bioactivity, likely due to impaired membrane permeability and diminished 325 intracellular accumulation. In contrast, the mono-epoxygenated derivative EL3, which retains 326 a balanced hydrophobic –hydrophilic character with fewer hydroxyl groups, exhibited the 327 most promising in vitro and in vivo activity against both drug-sensitive and drug-resistant L. 328 donovani strains. This enhanced efficacy is supported by its potent inhibition of 329 promastigotes and amastigotes, low host toxicity, and its ability to downregulate key parasite 330 survival enzymes involved in polyamine biosynthesis, particularly trypanothione reductase. 331 Molecular docking and dynamic simulations further corroborate that EL3 forms stable 332 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 15 hydrophobic interactions within the binding pockets of these enzymes, resulting in a higher 333 binding affinity than its more polar counterparts. These findings, coupled w ith EL3’s 334 favourable pharmacokinetic profile and its capacity to elicit a pro-inflammatory host 335 response, underscore the importance of achieving an optimal balance between aqueous 336 solubility and lipophilicity to maximize anti-leishmanial activity. 337 Drug design and therapeutic strategies against L. donovani infection neglect the difference in 338 the pro -parasitic mechanisms that aid in escape from the host immune pathways, such as 339 generating inflammatory cytokines or superoxides as ROS or NOS. This limits their use in 340 various cases. Also, prevalent drugs in practice are reported to have severe toxicity and side 341 effects on host metabolism. Longer persistence following a long -term treatment can lead to 342 the emergence of drug -resistant strains that cause a higher rat e of treatment failure. In this 343 study, we isolated a bio -active ethyl linoleate from the ethyl acetate extract with anti -344 leishmanial activity and then further achieved a single epoxygenated derivative (EL3) from 345 its synthetic analog (EL1), which had signif icant anti -proliferative properties against 346 promastigotes and amastigotes of both the drug -resistant (NS2) as well as drug -sensitive 347 (AG83) L. donovani parasites. Interestingly, the molecule was more effective against the 348 resistant strain than the sensitiv e strain, both in vitro and in vivo. The inhibitory effect was 349 due to the differential rates of gene expression of the thiol pathway between the virulent 350 strains of the drug-resistant and drug-sensitive L. donovani. The enzymes γ-GCS, GS, and TR 351 are needed in higher concentrations in resistant strains, and hence, lower expressivity of these 352 enzymes marks the exceptional role of EL3 as an anti -leishmanial lead in therapeutics and 353 drug design. Moreover, a secondary mode of action involving the induction of pr o-354 inflammatory cytokine gene expression in macrophages also adds to the efficacy of the 355 compound. 356 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 16

and methods. 357 Collection of Meripilus giganteus. 358 The wild edible mushroom M. giganteus (Pers.) P. Karst, a Basidiomycota fungus, is found 359 on stumps of freshly fallen trees and at the base of standing trees; often apparently growing 360 from the ground, but always in contact with wood, widely distributed in the Northeastern and 361 Eastern parts of the Himalayas (23). The basidiocarps were collected from different areas of 362 Darjeeling during May to September, 2012 -2015 (24). Information on edibility was gathered 363 through discussions and direct interviews with local people and by direct observation of how 364 this mushroom was collected and used. The damaged, infected, and very young fruit body of 365 this mushroom was avoided, and the sample was collected preciously. The morphological and 366 ecological features were noted, and color photographs of the materials were taken during 367 field trips. After the specimens were brought to the laboratory, the macroscopic and 368 microscopic properties were determined. Then the specimen was identified according to the 369 previously described methods (23, 25, 26) . The voucher specimen of the mushroom was 370 deposited with the accession code CHU AM266 in the Mycological Herbarium of the 371 Department of Botany, University of Calcutta (24). 372 Isolation of active molecule and synthesis of derivatives. 373 After the collection of the raw edible M. giganteus from the local market of Gangtok, India , 374 the samples were dried in a hot air oven at 30°C for 48 hours. The dried mushroom was then 375 ground into a fine powder using a mixer grinder to facilitate optimal metabolite extraction 376 with different solvents. For bioassay -guided isolation of active compound(s) , different crude 377 extract was prepared using solvents of varying polarities to extract metabolites selectively. 378 Initially, 20 g of finely powdered M. giganteus was soaked in 200 mL of petroleum ether in a 379 round-bottom flask and gently stirred at room temperature for 48 hours to extract non -polar 380 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 17 metabolites. The mixture was then filtered using vacuum filtration, and the solvent was 381 evaporated at room temperature using a rotary evaporator, yielding a greasy brownish extract 382 residue for bioassay. The remaining residue was subsequently soaked in 200 mL of 383 chloroform under gentle stirring for 48 hours. Vacuum filtration followed by evaporation of 384 solvent from the filtrate gives a yellowish, greasy material that has been used fo r bio-activity 385 evaluation. The residue was subsequently treated with ethyl acetate (EtOAc) and methanol 386 using a similar procedure, and the resulting solvent -extracted residues were prepared for 387 bioactivity assays. These extracted residues were then evaluat ed for their bioactivity against 388 drug-sensitive and drug-resistant promastigotes. Among them, the ethyl acetate extract of M. 389 giganteus exhibited notably intriguing bioactivity against the drug -resistant strain of L. 390 donovani. The significant in vitro bioactivity of the ethyl acetate-extracted residue against the 391 drug-resistant strain of Leishmania prompted us to bioactivity -guided isolation of pure 392 molecules from this extract. 393 The initial thin layer chromatographic (TLC) assessment of the residue of the ethyl acetate 394 extract displayed the presence of multiple closely spaced spots under different staining agents 395 (Fig. S1A). Due to the presence of various clusters of closely spaced spots in the TLC plate, 396 we found it convenient to identify the bioactive clu ster through initial preparative thin layer 397 chromatographic fractionation of EtOAc extract. The fractionation was carried out using 398 silica gel GF254 (Merk) on a 20 cm x 20 cm glass plate using Chloroform:Di -ethylEther: Pet 399 Ether (10:2:1) as the optimized s olvent system. Finally, we used an HPLC system equipped 400 with reverse phase semi -preparative column (C18, Agilent, Column: Zorbax, 9 × 250 mm, 401 particle size 5 μm, flow rate: 0.6 mL/min) and acetonitrile: water (7:3) as the optimized 402 mobile phase to isolate the active compound from this bioactive fraction, obtaining the 403 following chromatogram. 404 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 18 1H, 13C {1H}, NMR spectra were collected using Bruker Avance III 400 ( 1H: 400 MHz, 13C 405 {1H}: 100 MHz) and were referenced to the resonances of the solvent used with TMS as 406 internal standard. The chemical shifts were recorded in parts per million (ppm, δ) relative to 407 CDCl3 (7.28 ppm for 1H and 77.00 for 13C) and DMSO d6 (2.49 for 1H and 40.09 for 13C), 408 and coupling constants ( J) are reported in Hertz (Hz). Coupling pat terns are indicated as: br 409 (broad), s (singlet), d (doublet), t (triplet), q (quartet), p (pentet “quintet”), dd (doublet of 410 doublet), td (triplet of doublets) , or m (multiplet). Mass spectra were recorded on Accurate 411 Mass Q -TOF LC/MS (Agilent Technologies Singapore, G6520B). Thin layer 412 chromatography (TLC) was carried out to monitor the progress of the reactions using Merck 413 pre-coated TLC plates (silica gel 60 F254 0.25 mm) and visualized by 254 nm and 366 nm 414 UV light. Preparative thin -layer chromatography was carried out using silica gel GF254 415 (Merk) on a 20 cm x 20 cm glass plate. For isolation of active compound, we used HPLC 416 system (Agilent, Column: Zorbax, 9 X 250 mm of particle size 5μm. Column 417 chromatography was performed using silica gel (particle size 60-120 mesh) and eluted with a 418 petroleum ether and ethyl acetate mixture. The structure and NMR spectra of the isolated 419 peak 5 and synthetic derivatives have been provided (Fig. S2 - S7). 420 Parasites and animals. 421 Drug-sensitive AG83 parasites were origi nally obtained from CSIR -Indian Institute of 422 Chemical Biology, Kolkata Jadavpur (10, 11), and drug-resistant NS2 parasites were kindly 423 provided by Professor Mitali Chatterjee , Institute of PG Medical Education & Research, 424 Kolkata, West Bengal, India (12, 13). L. donovani promastigotes were transformed from 425 splenic aspirates of infected BALB/c mice as described previously (11). Male 4-5 weeks old 426 BALB/c were procured from Centre for Laboratory Animal Research and Training, Kalyani, 427 West Bengal, and allowed with rodent pellet diet, ad libidum, with a 12 -hour cycle of light 428 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 19 and dark as per CPSEA guidelines (File no. IAEC -1394/2015-16/5, dated 16.12.2015) 429 following ARRIVED guidelines of the Institutional Animal Ethics Committee, WBSU, 430 Barasat (13). 431 Anti-promastigote activity and determination of inhibitory concentration, in vitro. 432 The anti-promastigote activity of ethyl linoleate (Peak 5) and its synthetic derivatives (EL1 to 433 EL6) was evaluated against the promastigotes of drug -sensitive and drug -resistant L. 434 donovani, using the modified MTT assay, using the conventional tetrazolium MTT salt (27). 435 Briefly, 3×104 promastigotes of both drug -sensitive and drug -resistant strains were cultured 436 in each well of 96 -well plates (purchased from Genaxy Scientific Pvt. Ltd), in 100 µl 437 medium. After being treated with respective drugs for 48 hours, MTT was added at a 5 mg/ml 438 concentration and allowed to form the formazan crystals, which were subsequently dissolved, 439 and OD was taken using a multiplate reader (Bio-rad, USA). 440 Anti-amastigote assay and evaluation of cytotoxicity in infected peritoneal 441 macrophages, in vitro. 442 Macrophages were obtained from 4% thioglycolate-stimulated peritoneal exudates of 443 BALB/c mice, cul tured for 48 hours for adhesion and extension. Further, the peritoneal 444 macrophages were infected with Leishmania promastigotes (cells: parasite = 1:10) in 10% 445 FBS supplemented RPMI 1640 (28). Infected macrophages were treated with 25, 50, and 75 446 µg/ml of isolated ethyl linoleate, synthetic ethyl linoleate (EL1), and its synthetic derivatives, 447 EL2 and EL3, for 48 hours. Giemsa -stained micrographs were observed under a microscope 448 (Carl Zeiss, Axioscope) to count the amastigotes/ 100 macrophages (6,7). 449 In vivo efficacy and host toxicity of EL3. 450 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 20 Male BALB/c mice (4-6 weeks, 5 mice per group) were infected with 2×10 7 parasites of both 451 AG83 and NS2 strains via intravenous (IV) route and treated with, 0.25 mg, 0.35 mg, 0.5 mg, 452 and 1 mg/kg body weight of EL3 (one-month post-infection) via intramuscular (IM) route for 453 an alternative 5 days, and the animals were sacrificed at one and half month of the post -454 infected period (scheme of the in vivo experiment illustrated in Fig. S8). Parasite survival 455 and proliferation in the spleen and liver were calculated from the Giemsa -stained 456 micrographs and expressed by Stauber's formula (7). 457 Nephro- and hepato-toxicity assessment. 458 Experimental animals were anesthetized before sacrifice using diethyl ether. Blood was 459 collected from the tail vein, and tubes were placed in a slanting position while the blood was 460 allowed to form a clot for 45 min. Tubes were centrifuged at 1500Xg for 30 minutes, and the 461 sera were collected from the supernatant without hemolysis and stored at -20ºC for further 462 use. Serum parameters for drug toxicity, such as creatinine, AST, AP, and ALT levels, were 463 measured using Autospan kinetic assay kits, as per the manufacturer’s protocol (29). 464 Bioavailability of EL3 in serum, in vivo. 465 Blood was collected from naïve BALB/c mice (4-5 weeks) treated with a single 466 intramuscular dose of 1 mg/kg body weight of EL3 and the serum was isolated at various 467 time points (1, 3, 6, 12, 24, 48, and 72 hours post -treatment) were subjected to protein 468 precipitation using acetonitrile (ACN) followed by centrifugation at 14000g for 15 minutes at 469 4°C. The clear supernatant from each sa mple was filtered using a 0.22 µm syringe filter 470 before HPLC. 471 The presence of EL3 in the sera was detected in a C18 reverse phase column- Spherisorb 472 (Waters) with dimensions of 4.6 x 150 mm and a 5 µm particle size. The separation was 473 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 21 carried out using a mobile phase consisting of (a) HPLC-grade water and (b) HPLC-grade 474 ACN in a gradient over 25 minutes at a flow rate of 1 ml/min. Dual absorbance detectors 475 were used for detection, with one set at 275 nm and the other at 254 nm. Samples were 476 injected using a Hamilton micro-syringe into the Shimadzu UFLC system (29). 477 Analysis of intracellular L. donovani enzymatic pathway and host-specific pro-478 inflammatory cytokines, in vivo. 479 4-6 weeks male BALB/c mice were divided into groups of uninfected, infected control (both 480 AG83 and NS2), infected with AG83 and treated with 0.5 mg/kg body weight dose, infected 481 with AG83 and treated with 1 mg/kg body weight dose, infected with NS2 and treated with 482 0.5 mg/kg body weight dose, infected with NS2 and treated with 1 mg/kg body weight dose 483 of EL3. All doses administered intramuscularly, on each alternate day, up to 14 14 -day dose 484 regimen. The total mRNA was isolated from whole spleen tissue for real-time PCR of γ-GCS, 485 GS, and TR transcripts from the intracellular parasites in a Bio -Rad thermo cycler using 486 iTaqTM SYBR green master mix. The analyses we re performed using Bio -Rad CFX-maestro 487 software. The relative mRNA expression (fold changes) was calculated using the 2 -(ΔΔCT) 488

(30). 489 Splenocytes from a similar set of animals were isolated and pulsed with 25 µg crude soluble 490 antigen and maintained in RPMI 1640 medium with 5% CO 2 concentration, at 37ºC for 6 491 hours. The cDNAs were subsequently amplified with specific primers for IL -10, TGF -β, 492 TNF-α, IFN -γ, IL -12, and IL -6 (Table S13) using a semi -quantitative reverse transcriptase 493 PCR in a thermal cycler (Eppendorf, Germany). In all cases, the specific expressions of the 494 cytokines were normalized against murine GAPDH using the housekeeping gene (22). 495 Three-dimensional structure prediction of γ -glutamyl cysteine synthetase, glutathione 496 synthetase, and trypanothione reductase from Leishmania donovani. 497 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 22 The sequences of the three target proteins , γ -GCS, GS, and TR from L. donovani were 498 retrieved from the U niProt database using the specific IDs Q67BG3, E9BBX9, and P39050, 499 respectively. The three -dimensional structures of these proteins were generated using 500 MODELLER 9v10 and AlphaFold2 Colab (31, 32). In MODELLER, a total of 100 models 501 were generated, and the best one was chosen based on the DOPE score (33). AlphaFold2, on 502 the other hand, computes the pLDDT and pTM scores to assess the accuracy of its 503 predictions, with the top-ranked prediction by pLDDT used for further analysis (34, 35). The 504 stereochemical qualities of the three protein models were validated by Ramachandran plot 505 using PROCHECK (36). The two -dimensional (2D) chemical structure of the synthesized 506 ligand was sketched using ChemDraw and was converted into the corresponding standard 507 three-dimensional (3D) structure by using open Babel (37, 38). 508 Molecular docking of EL3 with the three target proteins. 509 To assess the binding efficiency of the EL3 molecule in predicting its interaction with the 510 three target proteins of L. donovani , molecular docking was employed using LeDock 511 software (39). All proteins and EL3 molecules were energy minimized before docking. The 512 three target proteins were arranged in a cubic box with a grid point spacing of 0.3750 Å. The 513 100 docking conformations were generated for each protein -ligand complex. The binding 514 conformations were assessed based on their binding energy (kcal/mol), and the conformation 515 with the lowest binding energy (indicating the strongest binding affin ity) for each protein -516 EL3 complex was selected for further analysis. The molecular interactions of protein -ligand 517 complexes were generated by using PLIP (Protein-Ligand Interaction Profiler) (40). 518 Molecular dynamics simulation of protein-EL3 complexes. 519 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 23 In this study, molecular dynamics (MD) simulations of three L. donovani protein -EL3 520 complexes, γ-GCS, GS, and TR were performed to investigate their stability and molecular 521 interaction mechanisms. GROMACS v 2023 and AMBER 99SB force field were used for the 522 MD simulations study (41, 42) . EL3 topology was generated using ACPYPE 523 (AnteChamberPYthon Parser interface) for GROMACS compatibility (43). Each protein-EL3 524 system was solvated in a cubic water box with a TIP3P water model by maintaining periodic 525 boundary conditions (PBC) (44, 45) . All three protein -ligand complexes were neu tralized 526 with sodium and chloride ions. Energy minimization of each system was performed using the 527 steepest descent until the maximum force was smaller than 1000 kJ/mol/nm (46). Each 528 protein-EL3 complex was equilibrated with a 200ps isothermal -isochoric en semble, NVT 529 followed by a 200ps isothermal -isobaric ensemble NPT. These two equilibration methods 530 stabilized systems at 310K and 1 bar pressure. Temperature and pressure coupling were 531 controlled using the Berendsen thermostat and Parrinello -Rahman methods, respectively (47, 532 48). The Particle Mesh Ewald (PME) method was used to calculate the long -range 533 electrostatic interactions, with cut -off radii of 0.9 nm set for both Van der Waals and short -534 range Coulombic interactions (49). The Linear Constraint Solver (LINCS) algorithm was 535 used to fix the peptide bond lengths and angles, ensuring simulation stability and accurate 536 dynamics of the protein -EL3 complexes (50). Production simulations were conducted for 537 100ns, with outputs saved at 10ps intervals. Principal Component Analysis (PCA) or essential 538 dynamics was conducted on backbone atoms of three target proteins to identify primary 539 motion patterns, with covariance matrices generated to examine system flexibility (51). For 540 binding affinity assessments, the MM -PBSA (Molecular Mechanics Poisson -Boltzmann 541 Surface Area) method was employed on selected trajectory frames of each system, 542 calculating interactions like electrostatics and van der Waals forces, yielding detailed free 543 energy profiles across the three protein -EL3 complexes (52). Stability metrics such as 544 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 24 RMSD, RMSF, Rg, and SASA were analyzed to validate the reliability of the simulation 545 outcomes. 546 Estimation of binding free energy. 547 To calculate the binding free energy of EL3 with the three target proteins, γ -GCS, GS, and 548 TR from L. donovani, we employed the MM-PBSA method (52). Using the GROMACS tool, 549 binding free energy calculations were estimated over the last 50 ns trajectory frames, 550 ensuring the system had achieved equilibrium. A total of 100 snapshots were extracted from 551 the trajectory at 0.5 ns intervals, providing representative conformations of the protein -EL3 552 complexes for reliable energy estimation. 553 Statistical analysis. 554 All results shown are representative of three different biological replicates of each experiment. The 555 obtained data were analysed using Sigma Plot 11 software. Statistical analyse s were done using 556 ANOVA, and significance testing of each result was done by Tukey’s test. Results were represented 557 as mean± SEM. 558 Acknowledgments: We thank the Vice Chancellor of West Bengal State University for 559 providing the research infrastructure required for this work. We acknowledge the DST-FIST, 560 Govt. of India (Ref. SR/FST/LS1-001/2014), and DBT-BOOST, Govt. of West Bengal (Ref. 561 49[11]/BT [Estt]/1P-4/2013 [Part-1]), for providing funds for the real-time PCR facility in the 562 Department of Zoology, WBSU, Barasat. We also acknowledge the iSTEM facility of the 563 Bose Institute, Kolkata, for providing us with the HPLC facility. SC acknowledges CSIR-564 IICB for infrastructure support. AC acknowledges ICMR for research associateship 565 [BMI/11(55)2022]. 566 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 25 Funding: This work was supported by the Department of Biotechnology, Government of 567 India [Ref. BT/PR16064/NER /95/88/2015, dated 09/01/2017 and BT/PR16064/NER /95/60/ 568 2015, dated 09/01/2017]. 569 Conflict of interest: The authors declare no commercial or financial conflict of interest or 570 personal relationships that could have appeared to influence the present work 571 APPENDIX 572 TLC, thin layer chromatography, HPLC - High performance liquid chromatography, DMSO - 573 Dimethyl sulphoxide, MTT -3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium 574 bromide, IL- Interleukin, IFN- Interferon, 575 576

577 1. Ghosh S, Nath S, Roy K, Karmakar S, Pal C. 2023. Leishmaniasis: tissue tropism in 578 relation to the species diversity, p 133-153. In Mukherjee B, Bhattacharya A, 579 Mukhopadhyay R, Guedes B, Aguiar A (ed), Pathobiology of Parasitic Protozoa: 580 Dynamics and Dimensions Springer Nature, Singapore. 581 2. Fact sheets: Details on Leishmaniasis. World Health Organization, 582 https://www.who.int/news-room/fact-sheets/detail/leishmaniasis, (accessed 2024-02-04). 583 3. Sundar S, More DK, Singh MK, Singh VP, Sharma S, Makharia A, Kumar PCK, Murray 584 HW. 2000. Failure of Pentavalent Antimony in Visceral Leishmaniasis in India: Report 585 from the Center of the Indian Epidemic. Clinical Infectious Diseases, 31:1104–1107. 586 4. Ponte-Sucre A, Gamarro F, Dujardin J-C, Barrett MP, López-Vélez R, García-Hernández 587 R, Pountain AW, Mwenechanya R, Papadopoulou B. 2017. Drug resistance and treatment 588 failure in leishmaniasis: A 21st century challenge. PLoSNegl Trop Dis 11:e0006052. 589 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 26 5. Wijnant GJ, Dumetz F, Dirkx L, Bulté D, Cuypers B, Van Bocxlaer K, Hendrickx S. 590 2022. Tackling Drug Resistance and Other Causes of Treatment Failure in Leishmaniasis. 591 Front Trop Dis 3:e837460. 592 6. Mallick S, Dey S , Mandal S , Dutta A, Mukherjee D, Biswas G , Chatterjee S , Mallick 593 S , Lai TK, Acharya K , and Pal C. 2015. A novel Triterpene from Astraeus 594 hygrometricus Induces Reactive Oxygen Species Leading to death in Leishmania 595 donovani. Future Microbiol 10:763–789. 596 7. Mallick S, Dutta A, Chaudhuri A, Mukherjee D, Dey S, Halder S, Ghosh J, Mukherjee D, 597 Sultana SS, Biswas G, Lai TK, Patra P, Sarkar I, Chakraborty S, Saha B, Acharya K, Pal 598 C. 2016. Successful Therapy of Murine Visceral Leishmaniasis with Astrakurkurone, a 599 Triterpene Isolated from the Mushroom Astraeus hygrometricus, Involves the Induction 600 of Protective Cell-Mediated Immunity and TLR9. Antimicrob Agents Chemother 601 60:2696–2708. 602 8. Hussain A, Ghosh S, Roy K, Nath S, Sarkar B, Dutta A, Maji P, Basu S, Paul S, Dey S, 603 Chakraborty K, Raychaudhury B, Acharya K, Ganguly J, Pal C. 2021. A mushroom 604 derived ‘carbohydrate -fraction’ reinstates host -immunity and protects from Leishmania 605 donovani infection. Parasite Immunol 43:e12806. 606 9. Banik SP, Khowala S, Pal C, Mukherjee S. 2015. Proteomic Approaches to Identify 607 Novel Therapeutics and Nutraceuticals from Filamentous Fungi, p. 265 –295. In 608 Genomics, Proteomics and Metabolomics in Nutraceuticals and Functional Foods. John 609 Wiley & Sons, Ltd. 610 10. Pal C, Raha M, Basu A, Roy KC, Gupta A, Ghosh M, Sahu NP, Banerjee S, Mandal NB, 611 Bandyopadhyay S. 2002. Combination Therapy with Indolylquinoline Derivative and 612 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 27 Sodium Antimony Gluconate Cures Established Visceral Leishmaniasis in Hamsters. 613 Antimicrob Agents Chemother 46: 259–261. 614 11. Mallick S, Halder S, Dutta A, Dey S, Paul K, Maiti S, Bandyopadhyay C, Saha B, Pal C. 615 2013. Chromone linked nitrone derivative induces the expression of iNOS2 and Th1 616 cytokines but reduces the Th2 response in experimental visceral leishmaniasis. 617 IntImmunopharmacol 15:772–779. 618 12. Maharjan M, Singh S, Chatterjee M, Madhubala R. 2008. Role of aquaglyceroporin 619 (AQP1) gene and drug uptake in antimony-resistant clinical isolates of Leishmania 620 donovani. 1. Am J Trop Med Hyg 79:69–75. 621 13. Roy K, Ghosh S, Karan M, Karmakar S, Nath S, Das B, Paul S, Mandal P, Ray M, Das 622 M, Mukherjee S, Dey S, Pal C. 2024. Activation of neutrophils excels the therapeutic 623 potential of Mycobacterium indicus pranii and heat-induced promastigotes against 624 antimony-resistant Leishmania donovani infection. Scand J Immunol 99:e13350. 625 14. Doddrell DM, Pegg DT, Bendall MR. 1982. Distortionless enhancement of NMR signals 626 by polarization transfer. J MagnReson 1969 48:323–327. 627 15. Aue WP, Bartholdi E, Ernst RR. 1976. Two‐dimensiona l spectroscopy. Application to 628 nuclear magnetic resonance. J ChemPhys 64:2229–2246. 629 16. Castañar L, Parella T. 2015. Chapter Four - Recent Advances in Small Molecule NMR: 630 Improved HSQC and HSQMBC Experiments, p. 163 –232. In Webb, GA (ed), Annual 631 Reports on NMR Spectroscopy. Academic Press. 632 17. Noggle JH, Schirmer RE. 1971. The Nuclear Overhauser Effect: Chemical Applications, 633 Academic Press, New York and London. ISBN 9780125206501. 634 18. McMurry J. 1996. Organic Chemistry, 4th ed., Brooks/Cole, Pacific Grove, CA. 635 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 28 19. VanRheenen V, Kelly RC, Cha DY. 1976. An improved catalytic OsO4 oxidation of 636 olefins to cis-1, 2-glycols using tertiary amine oxides as the oxidant. Tetrahedron Letters 637 17:1973-1976. 638 20. Li JJ, Corey EJ. 2007. Name Reactions of Functional Group Transformations. John Wiley 639 & Sons. 640 21. Salim T, Sershen CL, May EE. 2016. Investigating the Role of TNF- α and IFN -γ 641 Activation on the Dynamics of iNOS Gene Expression in LPS Stimulated Macrophages. 642 PLOS ONE 11:e0153289. 643 22. Dey S, Mukherjee D, Sultana SS, Mallick S, Dutta A, Ghosh J, Hussain A, Sarkar B, 644 Mandal S, Patra P, Saha B, Pal C. 2020. Combination of Mycobacterium indicuspranii 645 and Heat-Induced Promastigotes Cures Drug-Resistant Leishmania Infection : Critical 646 Role of Interleukin-6-Producing Classical Dendritic Cells. Infect Immun 88:e00222-19. 647 23. Das K. 2009. Mushrooms of Sikkim; Sikkim Biodiversity Board, Sikkim, India. 648 24. Maity P, Nandi AK, Manna DK,Pattanayak M,Sen IK,Bhanja SK,Samanta S, Panda BC, 649 PaloiS, Acharya K, Islam SS. 2016. Structural characterization and antioxidant activity of 650 a glucan from Meripilus giganteus. Carbohydrate Polymers 157:1237–1245. 651 25. Ramsbottom J. 1923. A handbook of the larger British fungi. Order of the Trustees of the 652 British Museum. 653 26. Singer R. 1962. The Agaricales in modern taxonomy. 2 nd ed.Hafner Publishing Co., New 654 York. 655 27. Yousuf M, Mukherjee D, Dey S, Chatterjee S, Pal A, Sarkar B, Pal C, Adhikari S. 2018. 656 Synthesis and biological evaluation of polyhydroxylatedoxindole derivatives as potential 657 antileishmanial agent. Bioorg Med ChemLett 28:1056–1062. 658 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 29 28. Mallick S, Dutta A, Ghosh J, Maiti S, Mandal AK, Banerjee R, Bandyopadhyay C, Pal C. 659 2011. Protective Therapy with Novel Chromone Derivative against Leishmania donovani 660 Infection Induces Th1 Response in vivo. Chemotherapy 57:388–393. 661 29. Mukherjee D, Yousuf M, Dey S, Chakraborty S, Chaudhuri A, Kumar V, Sarkar B, Nath 662 S, Hussain A, Dutta A, Mishra T, Roy BG, Singh S, Chakraborty S, Adhikari S, Pal C. 663 2020. Targeting the Trypanothione Reductase of Tissue-Residing Leishmania in Hosts’ 664 Reticuloendothelial System: A Flexible Water-Soluble Ferrocenylquinoline-Based 665 Preclinical Drug Candidate. J Med Chem 63:15621–15638. 666 30. Roy K, Ghosh S, Karmakar S, Mandal P, Hussain A, Dutta A, Pal C. 2025. Inverse 667 correlation between Leishmania-induced TLR1/2 and TGF- β different ially regulates 668 parasite persistence in bone marrow during the chronic phase of infection. Cytokine 669 185:156811. 670 31. Eswar N, Eramian D, Webb B, Shen M-Y, Sali A. 2008. Protein Structure Modeling with 671 MODELLER, p. 145–159. In Kobe, B, Guss, M, Huber, T (eds.), Structural Proteomics: 672 High-Throughput Methods. Humana Press, Totowa, NJ. 673 32. Jumper J, Evans R, Pritzel A, Green T, Figurnov M, Ronneberger O, Tunyasuvunakool 674 K, Bates R, Žídek A, Potapenko A, Bridgland A, Meyer C, Kohl SAA, Ballard AJ, Cowie 675 A, Romera-Paredes B, Nikolov S, Jain R, Adler J, Back T, Petersen S, Reiman D, Clancy 676 E, Zielinski M, Steinegger M, Pacholska M, Berghammer T, Bodenstein S, Silver D, 677 Vinyals O, Senior AW, Kavukcuoglu K, Kohli P, Hassabis D. 2021. Highly accurate 678 protein structure prediction with AlphaFold. Nature 596:583–589. 679 33. Shen M, Sali A. 2006. Statistical potential for assessment and prediction of protein 680 structures. Protein Sci 15: 2507–2524. 681 34. Holm L. 2022. Dali server: structural unification of protein families. Nucleic Acids Res 682 50: W210–W215. 683 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 30 35. Yin R, Feng BY, Varshney A, Pierce BG. 2022. Benchmarking AlphaFold for protein 684 complex modeling reveals accuracy determinants. Protein Sci 31:e4379. 685 36. Laskowski RA, MacArthur MW, Moss DS, Thornton JM. 1993. PROCHECK: a program 686 to check the stereochemical quality of protein structures. J ApplCrystallogr 26:283–291. 687 37. Mendelsohn LD. 2004. ChemDraw 8 Ultra, Windows and Macintosh Versions. J 688 ChemInfComputSci 44:2225–2226. 689 38. O’Boyle NM, Banck M, James CA, Morley C, Vandermeersch T, Hutchison GR. 2011. 690 Open Babel: An open chemical toolbox. J Cheminform 3:1-14. 691 39. Liu N, Xu Z. 2019. Using LeDock as a docking tool for computational drug design. IOP 692 Conference Series Earth and Environmental Science 218: 012143. 693 40. Salentin S, Schreiber S, Haupt VJ, Adasme MF, Schroeder M. 2015. PLIP: fully 694 automated protein–ligand interaction profiler. Nucleic Acids Res 43:W443–W447. 695 41. Van Der Spoel D, Lindahl E, Hess B, Groenhof G, Mark AE, Berendsen HJ. 2005. 696 GROMACS: Fast, flexible, and free. J ComputChem 26:1701–1718. 697 42. Wang J, Wolf RM, Caldwell JW, Kollman PA, Case DA. 2004. Development and testing 698 of a general amber force field. J ComputChem 25:1157–1174. 699 43. Da Silva AWS, Vranken WF. 2012. ACPYPE - AnteChamberPYthon Parser InterFaCE. 700 BMC Res Notes 5:1-8. 701 44. Mark P, Nilsson L. 2001. Structure and Dynamics of the TIP3P, SPC, and SPC/E Water 702 Models at 298 K. J PhysChem A 105:9954–9960. 703 45. Makov G, Payne MC. 1995. Periodic boundary conditions in ab initio calculations. 704 Physical Review. B, Condensed Matter 51: 4014–4022. 705 46. Meza JC. 2010. Steepest descent. Wiley Interdisciplinary Reviews: Computational 706 Statistics 2:719-22. 707 47. Lemak AS, Balabaev NK. 1994. On The Berendsen Thermostat. MolSimul 13:177–187. 708 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 31 48. Martoňák R, Lai o A, Parrinello M. 2003. Predicting Crystal Structures: The Parrinello-709 Rahman Method Revisited. Phys Rev Lett 90:075503. 710 49. Petersen HG. 1995. Accuracy and efficiency of the particle mesh Ewald method. J 711 ChemPhys 103:3668–3679. 712 50. Hess B, Bekker H, Berendsen HJC, Fraaije JGEM. 1997. LINCS: A linear constraint 713 solver for molecular simulations. J ComputChem 18:1463–1472. 714 51. Abdi H, Williams LJ. 2010. Principal component analysis. Wiley Interdisciplinary 715 Reviews Computational Statistics 2:433–459. 716 52. Genheden S, Ryde U. 2015. The MM/PBSA and MM/GBSA methods to estimate ligand-717 binding affinities. Expert Opinion on Drug Discovery 10:449–461. 718 719 Figure Legends: 720 Figure 1: Synthesis scheme of the derivatives (EL2- EL6) from Peak 5/ EL1. 721 Figure 2: Anti-promastigote activity of ethyl linoleate and synthetic derivatives, in vitro. (A) 722 Original Peak 5 compound and EL3 reduced the proliferation dose dependently as observed 723 by MTT assay. Results are shown as mean ± SEM of three different experiments performed 724 in triplicate; *p<0.05 vs. DMSO-treated control. (B) The IC50 doses of reference drugs were 725 tested on both AG83 and NS2 parasites, **p<0.001 vs. control. 726 Figure 3: Anti-amastigote activity of ethyl linoleate and synthetic derivatives, in vitro and in 727 vivo.(A) Anti-amastigote activity of EL1, EL2, and EL3 against intracellular amastigotes in 728 infected macrophages, in vitro . (B) The anti-amastigote effect of EL3, in vivo . Data 729 represented as the mean ± SEM of three different experiments performed against DMSO -730 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 32 treated control, and significance was calculated by ANOVA using GraphPad Prism (Version 731 8), *p<0.001, #p<0.05. 732 Figure 4: Relative mRNA expression of the polyamine biosynthesis pathway key enzymes of 733 the parasite and the murine host. (A) The relative γ -GCS, GS, and TR ( Leishmania specific) 734 mRNA expressions in the spleens of experimental animals, *p<0.05. (B) The relative mRN A 735 expressions of γ -GCS and GS (host-specific) in the spleens of experimental animals, 736 **p<0.001. Data represented as mean± SEM of two experiments, with 5 mice in each group. 737 Figure 5: Molecular interactions of EL-3 with three target proteins γ-glutamyl cyst eine 738 synthetase, glutathione synthetase, and trypanothione reductase. Each protein is depicted in a 739 combination of surface and ribbon structures, while EL3 is shown in stick representation. A 740 red box highlights a close -up view of the binding site, which is visualized as a yellow area 741 within a 5 Å radius from EL3. (A) γ -GCS- EL3 complex. Blue, yellow, and grey lines (solid 742 and dotted) represent hydrogen bonds, salt bridges, and hydrophobic interactions. (B) GS - 743 EL3 complex. Grey dotted lines represent hydr ophobic interactions. (C) TR - EL3 complex. 744 Grey dotted lines represent hydrophobic interactions. 745 Figure 6: MD simulation trajectory parameters of three protein-ligand complexes. (A) 746 Energy; (B) RMSD; (C) RMSF; (D) Rg; (E) SASA; (F) Trace of covariance matrix (PCA). 747 RMSD: Root mean square deviation; RMSF: Root mean square fluctuation; Rg: Radius of 748 gyration; SASA: Solvent- accessible surface area. γ -GCS, GS, and TR were represented as 749 blue, red, and yellow lines, respectively. 750 Figure 7: Estimation of the serum biomarkers (A) Aspartate transaminase , (B) Creatinine, 751 (C) Alkaline phosphatase, and (D) Alanine transaminase levels from the in vivo experimental 752 animals. All values are compared to the uninfected control group (*p<0.001, **p<0.004). 753 Data represented as mean± SEM of two experiments, with 5 mice in each group. 754 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint 33 Figure 8: Induction of pro-inflammatory cytokines in the host, in vivo . (A) Induction of 755 cytokines at the mRNA level. (B) Densitometry of the cumulative data, *p<0.001.All 756 experiments were carried in duplicate, with 5 mice in each group. 757 Figure 9: Bioavailability of EL3 from serum samples detected by HPLC at various time 758 points (from 1h to 48h) post intramuscular injection of EL3 1 mg/kg b.w. dose in vivo . 759 Significance was tested with respect to untreated mouse sera for various time point groups 760 (*p<0.001). Data represented as mean± SEM of three different experiments. 761 762 CRediT authorship contribution statement: 763 Supriya Nath: Investigation, Methodology, Formal analysis, Validation, Writin g – original 764 draft; Karan Chhetri: Chemical synthesis, Methodology, Formal analysis; Aabid Hussain: 765 Investigation, Methodology, Formal analysis, Ankur Chaudhuri: Investigation, 766 Methodology, In silico Validation, Formal analysis; Joydip Ghosh: Investigation; Sondipon 767 Chakraborty: Investigation, Methodology, Formal analysis, Validation; Debarati 768 Mukherjee: Investigation; Mintu Karan: Investigation; Bikramjit Raychaudhury: 769 Methodology; Krishnendu Acharya: Methodology; Saikat Chakrabarti: Writing-review & 770 editing, Data curation, Conceptualization , Bioinformatics; Biswajit Gopal Roy: Writing- 771 review & editing, Data curation, Conceptualization, Chemical synthesis, Supervision ; 772 Chiranjib Pal: Writing- review & editing (Final draft), Validation, Project administration, 773 Fund acquisition, Data curation, Conceptualization, Supervision. 774 775 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 7, 2025. ; https://doi.org/10.1101/2025.07.03.663049doi: bioRxiv preprint