Potential selection of antimony and methotrexate cross-resistance in Leishmania infantum circulating strains

Background Visceral leishmaniasis (VL) resolution depends on a wide range of factors, including the instauration of an effective treatment coupled to a functional host immune system. Patients with a depressed immune system, like the ones receiving methotrexate (MTX), are at higher risk of developing VL and refusing antileishmanial drugs. Moreover, the alarmingly growing levels of antimicrobial resistance, especially in endemic areas, contribute to the increasing the burden of this complex zoonotic disease. Principal findings To understand the potential links between immunosuppressants and antileishmanial drugs, we have studied the interaction of antimony (Sb) and MTX in a Leishmania infantum reference strain (LiWT) and in two L. infantum clinical strains (LiFS-A and LiFS-B) naturally circulating in non-treated VL dogs in Spain. The LiFS-A strain was isolated before Sb treatment in a case that responded positively to the treatment, while the LiFS-B strain was recovered from a dog before Sb treatment, with the dog later relapsing after the treatment. Our results show that, exposure to Sb or MTX leads to an increase in the production of reactive oxygen species (ROS) in LiWT which correlates with a sensitive phenotype against both drugs in promastigotes and intracellular amastigotes. LiFS-A was sensitive against Sb but resistant against MTX, displaying high levels of protection against ROS when exposed to MTX. LiFS-B was resistant to both drugs. Evaluation of the melting proteomes of the two LiFS, in the presence and absence of Sb and MTX, showed a differential enrichment of direct and indirect targets for both drugs, including common and unique pathways. Conclusion Our results show the potential selection of Sb-MTX cross-resistant parasites in the field, pointing to the possibility to undermine antileishmanial treatment of those patients being treated with immunosuppressant drugs in Leishmania endemic areas.


Introduction
Leishmaniasis is a worldwide infectious disease caused by parasites of the genus Leishmania [1].These parasites have two forms: the extracellular or promastigotes found in the sandfly vector and the intracellular or amastigotes found in the host cells [2].Among the different clinical manifestations, visceral leishmaniasis (VL) is the most severe form of the disease, for which Leishmania infantum is the main causal agent [1].VL is associated with elevated ranges of morbidity and mortality and 300,000 new cases are reported each year, where 95% of them are fatal if untreated [3,4].In the absence of an effective vaccine, control of the disease is based on a very limited pharmacopeia with organic antimonials being one of the key drugs for VL treatment [5].
To bestow their antileishmanial activity, pentavalent antimonials (Sb V ) must enter host infected cells and be reduced into the trivalent antimony (Sb III ) [5,6].Sb III causes oxidative stress by increasing the concentration of reactive oxygen species (ROS), inducing DNA damage that leads apoptosis in the parasite [6,7].Of note, nowadays, VL treatment is hampered, since the use of Sb is compromised due to Leishmania ability to develop and spread antimicrobial resistance, especially in endemic areas of the disease where these drugs have been continuously used in treating both human and canine patients [8,9].Although metal resistance in Leishmania spp. is multifactorial, the main mechanism of Sb detoxification involves the ATPbinding cassette protein MRPA which binds to thiol-conjugated metals and promotes the exocytosis of these complexes outside the parasite [9][10][11][12].In addition to an efficient pharmacological treatment, effective control of VL requires a protective Th1-type immune response by the host [13].Consequently, immunosuppression represents the major individual risk factor to develop severe VL.This has been traditionally reported as an emerging problem in HIV co-and proteomics laboratory equipment was funded by the Canadian Foundation for Innovation (grant no.36706 to FB).FB holds the Canada Research infected patients [14,15].Alarmingly, there is a recent increase in the number of VL cases among patients receiving immunosuppressant treatments to treat autoimmune diseases such as psoriasis, lupus erythematous or rheumatoid arthritis (RA) [16].In these cases, VL treatment becomes more difficult and the risk of relapse increases [14,17].
Methotrexate (MTX) is, for more than 30 years, one of the most successful immunosuppressants for the control of inflammatory conditions (i.e., 60% of RA are currently on or have been on MTX).MTX is an antagonist of folic acid that interferes purine and pyrimidine synthesis by binding to dihydrofolate reductase (DHFR) and pteridine reductase 1 (PTR1) enzymes [18].This results in a rapid depletion of intracellular levels of folates, which impairs DNA synthesis and leads to a decrease in cell proliferation [19].Leishmania as well as other parasites are sensitive to MTX [20], although this drug is not used to treat leishmaniasis.However, as folates and pterins are essential for Leishmania development, these parasites can rapidly evolve resistance to MTX by increasing dhfr-and ptr1-gene dosage [21,22].
Whereas the mode of action and mechanism of drug resistance against Sb and MTX have widely explored in Leishmania in the past [6,23], there are no reports available on the potential effects on cross-tolerance or cross-resistance after exposure to any of these two different drugs.
Here we report the first evidence of the potential co-selection of antimicrobial resistance between antimonial drugs and methotrexate in L. infantum circulating strains from untreated, naturally infected dogs.Moreover, the melting proteomes (meltome) of these strains, in the presence and absence of Sb and MTX, has identified differentially enriched direct and indirect targets for both drugs in different genetic backgrounds.This novel knowledge could bring some light into treatment failure and relapses occurring in patients under methotrexate immunosuppression, as well as to be the jumping-off point for tailoring better immunosuppression strategies in leishmaniasis endemic areas.

Drug-response assays in free-living promastigotes and intracellular amastigotes
The antileishmanial activity was assessed by monitoring the growth of non-exposed promastigotes for 72 hours at 25 ˚C in the presence of increasing concentrations of Sb (Potassium antimony tartrate sodium, Sigma) (0, 25, 50, 100, 150, 200, 300, 400 μM) or MTX (methotrexate, Sigma) (0, 10, 50, 100, 1000, 3000, 6000, 10000 nM).The optical density at 600 nm (A600) was measured using a Cytation 5 machine (Agilent, USA).Simultaneously, to investigate if exposure to one drug could induce cross-resistance or tolerance to the other, we subjected LiWT, LiFS-A, and LiFS-B promastigotes to the EC 50 and the EC 90 of either Sb or MTX (administered as single doses) over a period of five days.Following this, we performed a drug-response assay PLOS NEGLECTED TROPICAL DISEASES using the alternate drug (either Sb or MTX) on these 'pre-exposed' promastigotes utilizing the same spectrum of concentrations as previously described.
The intra-macrophage leishmanicidal activity of Sb (sodium stibogluconate, Calbiochem) and MTX was determined through in vitro infections, following our established protocols (9).Briefly, 2.5 × 10 5 BMDM cells were seeded onto Ibidi 12-well chamber slides and maintained in complete DMEM medium.Metacyclic phase promastigotes of LiWT, LiFS-A, and LiFS-B were used at a BMDM to parasite ratio of 1:10 for the infection process.The cells were infected and allowed to incubate for 6 hours at 37˚C with 5% CO2 in drug-free DMEM medium.After a 24-hours drug-free period, the medium was supplemented with increasing concentrations of MTX (0, 20, 50, 100, 200, 500 nM) or Sb (0, 10, 25, 50, 100, 200 μg/mL) for 5 days.To facilitate parasite visualization, the slides were fixed in methanol and stained with Diff-Quick solution.The number of infecting amastigotes per 100 cells was determined by examining 300 macrophages per triplicate assay and normalized to the untreated control.
In all these experiments, either targeting the promastigote or the amastigote stages, EC 50 values were calculated based on dose-response curves analyzed by non-linear regression with GraphPad Prism 10.0 software (GraphPad Software, La Jolla California, USA).An average of at least three independent biological replicates run in triplicate was performed for each determination.

Measurement of Reactive Oxygen Species (ROS) accumulation
Intracellular ROS accumulation was measured using the DCFDA dye (Invitrogen, USA) as previously described [12].Briefly, 5 × 10 7 mid-log LiWT, LiFS-A, and LiFS-B promastigotes were exposed to the EC 90 of the drugs (i.e., Sb or MTX) for 48 hours in M199 medium at 25˚C supplemented with 10% FBS and 5 μg/mL of hemin (pH 7.0).Parasites were washed twice in Hepes-NaCl (21 mM Hepes, 137 mM NaCl, 5 mM KCl, 0.7 mM Na 2 HPO 4 7H 2 O, 6 mM glucose, pH 7.4) and resuspended in 500 μL of Hepes-NaCl containing 25 μg/mL of H 2 DCFDA (Invitrogen, USA).Parasites were then incubated in the dark for 30 min and washed twice with Hepes-NaCl.After washing, 200 μL of the promastigote resuspension was analyzed with a Cytation 5 machine (Agilent, USA) at 485 nm excitation and 535 nm emission wavelengths.Fluorescence was normalized with the number of living parasites determined by propidium iodide (PI) staining and manual counting.Experiments were performed with at least three biological replicates from independent cultures, each of which included three technical replicates.

Comparative meltome analysis using thermal proteomic profiling (TPP)
For TPP analysis, LiFS-A and LiFS-B were prepared following our previously described methods [25].In brief, cultures of LiFS-A and LiFS-B isolates in the mid-log phase underwent multiple centrifugation steps.We conducted experiments with biological triplicates for each isolate.The resulting pellet was washed with PBS 1× (pH 7.4, Gibco, Life Technologies) and then resuspended in 5 mL of lysis buffer.The lysis buffer consisted of 50 mM mono-basic potassium phosphate, 50 mM di-basic potassium phosphate, 0.5 M EDTA, 1 M DTT, 10 mM tosyl-L-lysyl-chloromethane hydrochloride, 0.8% n-octyl-β-D-glucoside, and mini protease inhibitor cocktail (EDTA-free).To obtain sufficient protein, three freeze-thaw cycles were performed, followed by centrifugation at 20,000 g for 20 minutes at 4 ˚C.The protein yield required for the TPP experiment was 4 mg.
Once the lysate was obtained, drug-induced disruption and heat treatment were performed.Each lysate was divided into three subsamples: 100 μM Sb, 100 μM MTX, and a control (vehicle).For each condition, 250 μg of lysate (approximately 100 μL) was added to seven microcentrifuge tubes, with each tube representing a different temperature (37,45,50,55,60,65, and 70 ˚C).The tubes were incubated for three minutes, followed by centrifugation at 20,000 g for 20 minutes at 4 ˚C to recover the soluble protein fraction.The soluble proteins were collected by precipitation using cold acetone and 50 mM tris-HCl, followed by alkylation with 40 mM 2-Iodoacetamide (IAA, Sigma) and digestion with a 1:20 trypsin solution for 24 hours.After incubation, the samples were labeled using a light (test samples) and heavy (internal standard; L. infantum WT maintained at 37 ˚C) dimethyl strategy and mixed for consecutive HPLC-MS/MS analysis using a duplex labeling approach.High-performance liquid chromatography (HPLC) was performed using a Thermo Scientific Vanquish FLEX UHPLC system (San Jose, USA) with gradient elution on a microbore column (particle size: 5 μm, Thermo Biobasic).The mobile phase, a mixture of acetonitrile and water containing 0.1% formic acid, was subjected to a linear gradient shift from 5:95 to 40:60 over a duration of 63 minutes.Detection in the positive ion mode was carried out using a Thermo Scientific Q Exactive Plus Orbitrap Mass Spectrometer, which was integrated with the UHPLC system.The TOP-10 Data Dependent Acquisition method was employed for this purpose.Rather than treating each replicate as an individual sample, we opted to pool the data from these replicates, thereby combining their results to form a single, comprehensive dataset for each condition.The data processing for the study was carried out using Thermo Proteome Discoverer (version 2.4), in combination with SEQUEST.The analysis involved a curated database with FASTA sequences from UniProt specific to L. infantum (TAXON ID 5671).Key settings included an MS 1 tolerance of 10 ppm, MS 2 mass tolerance of 0.02 Da for Orbitrap detection, and trypsin specificity with allowance for two missed cleavages.Fixed modifications included carbamidomethylation of cysteine and dimethylation of lysine and N-terminus, while oxidation of methionine was a variable modification.The minimum peptide length was set at six amino acids, excluding proteins identified by only one peptide.Protein quantification and comparative analysis were based on peak integration, using the average ion intensity of unique peptides to determine protein abundance.For normalization purposes, the protein abundance value at the lowest examined temperature (37˚C) was set as the baseline, represented by a value of 1.The generated melting curves were inspected for a change in melting behavior following the formula described by Franken et al (2015) [26].All melting curves were created using GraphPad Prism 10.The temperature resulting in a 50% of protein denaturalization was defined as the melting temperature (T m ), which was used to calculate the cut-off value (ΔT m = T m drug-T m control).Heat maps were generated through the Heat mapper webserver (www.heatmapper.ca/expression) using its protein expression plugin with average linkage as clustering method applied to rows and Euclidean as distance measurement method.The complete proteomics dataset is available in (S1 Data).

L. infantum clinical isolates display different sensitivity profiles and enhanced ability to control oxidative stress in the presence of antimony and methotrexate
The drug-resistant profile of current L. infantum strains circulating in dogs presents a significant challenge in the treatment of both canine and human leishmaniasis, especially in pharmacologically immunosuppressed individuals.Several studies have reported an alarming increase in drug resistance, particularly to commonly used antileishmanial drugs such as Sb.In this context, we first evaluated the sensitivity profile of LiFS-A and LiFS-B, two clinical isolates recovered from non-treated, naturally infected dogs in Spain [27].For comparison purposes, we included the L. infantum ITMAP-263 laboratory reference strain (LiWT), which is known to be sensitive to the different antileishmanials.As summarized in Table 1, clinical isolate LiFS-A showed similar levels of Sb sensitivity of those measured for the reference strain (75.35 vs. 68.24μM Sb III in promastigotes; and 68.71 vs. 99.35μg/mL Sb V in amastigotes).Of note, LiFS-B showed a clear resistant profile with EC 50 values > 2.5-fold when compare with the LiWT reference strain (198.2 μM Sb III in promastigotes; and > 200 μg/mL Sb V in amastigotes).While there is no report in the literature of treating leishmaniasis with MTX, this drug is frequently used off label for the treatment of immune-mediated diseases, such as immune-mediated hemolytic anemia and immune-mediated polyarthritis in dogs.As expected, the LiWT strain showed a very sensitive phenotype against this drug in both promastigotes and amastigotes (1.03 and 0.34 μM, respectively).Conversely, both clinical isolates displayed high levels of resistance as both free and intracellular parasites (>500 and > 200 μM, respectively).
Antimicrobial resistance mechanisms in Leishmania can involve alterations in drug targets, decreased drug uptake, increased drug efflux, and enhanced antioxidant defenses.To explore this last feature, we examined the impact of Sb and MTX on the ability of LiWT, LiFS-A and LiFS-B to control reactive oxygen species (ROS) accumulation (Fig 1).In this way, the three strains were exposed to the EC 90 of Sb and MTX.DCFDA fluorescence emission and parasite survival rates were simultaneously measured.Sb and MTX induced major accumulation of ROS in LiWT (up to 892 and 3868 relative fluorescence units (RFU), respectively) after a 48-h exposure to the EC 90 of Sb and MTX.This was coupled with a significant reduction in the presence of viable parasites, which was reduced by more than 53% and 94% when exposed to Sb and MTX, respectively.This is consistent with the antileishmanial effect previously described for these drugs [7,28].Both LiFS-A and LiFS-B displayed similar basal levels of ROS (and similar to LiWT) in the absence of drug pressure.Exposure to Sb led to similar ROS and viability levels in LiFS-A when compared with the reference strain (⁓980 RFU), which is in agreement with its Sb-sensitive profile-as per determined in the drug-response assays.In contrast, LiFS-B exhibited approximately 2-fold lower ROS accumulation compared to LiWT and demonstrated better survival rate when exposed to Sb, providing additional evidence for its Sb-resistant phenotype.Both LiFS-A and LiFS-B exhibited reduced ROS accumulation and enhanced survival compared to LiWT when exposed to MTX.LiFS-A demonstrated approximately 20-fold lower ROS accumulation than LiWT and displayed a similar survival rate to the untreated control (approximately 100%).Conversely, LiFS-B accumulated around 20-fold less ROS than the reference strain but exhibited lower viability than the untreated control (approximately 50%).These findings further support the MTX-resistant phenotype observed in both isolates and suggest enhanced ability to control oxidative stress.

Overexpression of key drug-resistance genes and 'pre-exposure' to Sb or MTX contribute to multidrug-resistance phenotypes
One of the most frequent mechanisms deployed by Leishmania parasites to overcome the action of Sb and MTX is upregulating the expression of drug targets and drug-resistance genes.Overexpression of the gene coding for an ABC-thiol transporter multidrug resistance protein A (mrpA) is frequently reported in Sb-resistant parasites, leading to the intracellular sequestration and subsequent elimination of Sb-thiol conjugates [9].Likewise, MTX-resistance is associated with the overexpression of dihydrofolate reductase (dhfr) and pteridine reductase 1 (ptr1) genes, respectively, encoding the primary and secondary targets of MTX [21].For that reason, we evaluated the expression of mrpA gene, or dhfr and ptr1 genes, associated to Sb or MTX resistance, respectively, in non-exposed promastigotes.Our results illustrated a notable increase in mrpA expression in LiFS-B compared to both LiWT and LiFS-A (Fig 2A ), providing further evidence for its classification as Sb-resistant (Table 1).Regarding MTX genes of resistance, no significant differences in the expression levels of ptr1 were found (Fig 2B).Nonetheless, in LiFS-A, we observed a non-significant trend in the expression of mrpA and ptr1.Both clinical isolates showed a higher expression of the dhfr gene when compared with the reference strain LiWT (Fig 2C ), which could contribute to the survival of these parasites in higher concentrations of MTX as previously reported [12,21].
Next, to further understand the potential effect of a pharmacological immunosuppression (i.e., induced by MTX) on the outcome of L. infantum treatment (i.e., induced by Sb), we evaluated the impact of a single-dose exposure ('pre-exposure' to EC 50 or EC 90 for 5 days) to MTX or Sb prior to characterizing these parasites in drug-response assays.As depicted in Fig 3A, 'pre-exposure' to Sb EC 90 led to a significant reduction in MTX sensitivity in LiWT parasites (2.65-fold).Markedly, this phenomenon was bidirectional, and 'pre-exposure' to either the EC 50 or EC 90 of MTX resulted in a great decrease in the sensitivity of the LiWT reference strain to Sb (Fig 3B).The effect was maximal when exposing LiWT to MTX EC 90 (Fig 3A ), probably due to the rapid emergence of a subset of the population carrying amplifications of the H locus which contains ptr1 but also mrpA [29].This aspect was further investigated by evaluating the expression levels of ptr1 and mrpA in LiWT pre-exposed to EC 90 of MTX, along with argininosuccinate synthase (ass), a third gene within the H locus.As anticipated, the population that was recovered exhibited a significant increase in the mRNA expression levels of these three

Clinical isolates' melting proteomics reveals different protein interactions with the drugs pointing to different mechanisms of drug resistance
The mechanisms underlying antimony resistance have been extensively described in laboratory lines of Leishmania.However, there may be differences between these mechanisms and  B (E-F) promastigotes to the EC 50 and EC 90 (previously calculated; Table 1) of Sb or MTX, parasites were submitted to increasing concentrations of MTX and Sb to evaluate potential changes in their phenotype against these drugs.EC 50 values were calculated from concentration-response curves performed with biological triplicates after nonlinear fitting with GraphPad Prism 10 software.https://doi.org/10.1371/journal.pntd.0012015.g003those operating in Leishmania circulating isolates [30].To map Sb and MTX potential targets (both direct and indirect) in LiFS-A and LiFS-B clinical isolates, we used a powerful multiplexed, quantitative mass spectrometry-based proteomics approach named Thermal Proteomic Profiling (TPP), which enables monitoring the melting profile of thousands of expressed soluble proteins in drug-sensitive and drug resistant L. infantum parasites, in the presence (or absence) of any antileishmanial drug [25].As previously described, in our TPP approach, we used a fixed concentration of drug (100 μM SB or MTX) for the induction of drug-driven disruption and seven different temperatures for the temperature range (37-70 ˚C) [25].
We first measured the impact of Sb and MTX on the thermal stability of the soluble proteins of LiFS-A strain (Sb-sensitive and MTX-resistant; Table 1).We obtained and analyzed quantitative data to determine the thermal stability of 1147 soluble proteins (S1 Data).Out of these proteins, 118 exhibited variations in their thermal stability and their ΔT m was positive (ΔT m > 0) (Table 2).As depicted in Fig 4A and 4B, these proteins demonstrated enhanced stability at lower temperatures, specifically between 37 and 50 ˚C.Noteworthy proteins in this category included a mitochondrial elongation factor (E9AGQ3), a putative iron-sulfur reiske protein (A4IB55), a calmodulin-like protein (A4IBS7), as well as various ribosomal proteins (L6, L12, L13, L23, S20, L34, S6, S13, S4, L18, and S9).Conversely, in our investigation of the interaction between the LiFS-A strain and MTX, we identified 1158 soluble proteins.Among these proteins, we obtained melting curve profiles in the presence of MTX for 84 of them (Fig 4C and  4D).Table 2 provides a summary of the proteins identified in this analysis, including PTR1 (A4I067), two ribosomal proteins (A4HS42 and A0A6L0XG31), an amidohydrolase (A4I5G9), the cytochrome C1 mitochondrial protein (A4HT63), and an oligopeptidase b (A4HTZ8), among others.
Next, we conducted an evaluation of the proteomic profile in the antimony-resistant clinical isolate LiFS-B, both in the presence and absence of Sb.Among the 1193 soluble proteins identified (S1 Data), 42 exhibited a pattern that allowed us to calculate melting curves (Fig 5A and 5B).Compared to LiFS-A, the Sb-resistant strain showed a general decrease in proteinthermal stabilization, with 76 fewer proteins exhibiting temperature variations suitable for melting curve calculations (118 in LiFS-A versus 42 in LiFS-B), which further confirms a decreased interaction of Sb with its proteome.Within these 42 proteins, particularly those exhibiting the highest ΔTm, we identified several key proteins: an alanine-tRNA ligase (A4I013), two ribosomal proteins-the 60S acidic ribosomal protein P0 and the 40S ribosomal protein S24 (A4I2U1 and A4ID74, respectively)-, a putative 60S ribosomal protein (L10A), a putative ATP synthase F1 subunit protein (A4HZI3), a eukaryotic translation initiation factor (A4I5Y5), and an uncharacterized protein (A4HZ42) (Table 3).To elucidate the potential function of the uncharacterized protein and its involvement in Sb resistance, we employed databases such as PantherDB (http://www.pantherdb.org/,accessed on 20 June 2023), InterPro (https://www.ebi.ac.uk/interpro/, accessed on 20 June 2023), and Uniprot (https://www.uniprot.org/,accessed on 30 January 2024).This search revealed a possible orthologous relationship (92.33% identity) between our uncharacterized protein and a NTF2 (Nuclear transport factor 2) domain-containing protein found in the same chromosome of L. major (LMJF_21_0430, accession number Q4QCH42), potentially belonging to Ras-GTPase-activating protein-binding.
Finally, we evaluated the proteomic meltome profile in the LiFS-B strain in the absence and presence of MTX.Among the 1112 soluble proteins identified (S1 Data), melting curves were determined for 55 of them.Like the other drug analyzed, we observed protein stabilization at low temperatures, both with and without the presence of MTX (Fig 5C and 5D).Table 3 summarizes the proteins identified, including ATP synthase subunit beta (A4I1G1), an activator of Hsp90 ATPase (A4HXP7), a SNF1-related protein kinase (A4I088), a putative long-chain fatty PLOS NEGLECTED TROPICAL DISEASES  PLOS NEGLECTED TROPICAL DISEASES PLOS NEGLECTED TROPICAL DISEASES PLOS NEGLECTED TROPICAL DISEASES acid (A4HRH2), a GDP-mannose pyrophosphorylase (A4I048), two conserved hypothetical proteins (A4HXB7 and A4I5W4), and two ribosomal proteins (L10 and S10).

Discussion
The primary challenge in immunosuppressed patients with VL remains the inadequate effectiveness of antileishmanial treatments and the heightened risk of relapses [31,32].This predicament is further exacerbated by the escalating emergence of drug-resistant strains in Leishmania parasites [33].A deficient immune response can largely permit the outgrowth of persisters or other Leishmania variants that exhibit intermediate resistance levels.Of note, due to its genomic plasticity-coupled to the shared use of antileishmanials in animals and humans-, for many Leishmania infections, drug resistant parasites are likely present by the time chemotherapy starts [34].Current research on drug resistance in Leishmania during treatment has mainly focused on the interaction between the parasite and antileishmanials, frequently ignoring the direct impact of other drugs such as immunosuppressants on Leishmania evolution.This study uncovers, for the first time, cross-resistance between MTX and Sb in clinical L. infantum isolates.This finding is significant as it may compromise treatment efficacy in immunosuppressed patients and contribute to the spread of drug-resistant parasites.First, we assessed the responsiveness of both isolates to Sb and MTX, focusing on their capability to regulate ROS levels.LiFS-A exhibited a susceptibility to Sb and demonstrated an  Color range depicts the relative protein abundance of the soluble fractions at different temperatures.Heat maps were generated through the Heat mapper webserver (www.heatmapper.ca/expression)using its protein expression plugin with average linkage as clustering method applied to rows and Euclidean as distance measurement method.Of note, within the two field strains, 20 proteins were identified as shared following exposure to Sb.These shared proteins encompass ribosomal proteins, elongation increased accumulation of ROS when exposed to this drug.This could be explained by the fact that the trypanothione/trypanothione reductase (TR) system, essential for the parasite's oxidoreductive balance, is disrupted by trivalent Sb.This disruption causes a rapid efflux of T [SH]2 and glutathione and leads to apoptosis by increasing ROS and intracellular Ca 2+ levels [35,36].On the other hand, isolate LiFS-B was able to control ROS levels following MTX and Sb exposure, which correlated with a drug-resistant phenotype against both drugs.Of note, it has been observed that Leishmania strains recovered from immunosuppressed patients exhibit decreased sensitivity to Sb-based treatments [37].Notably, most cases of secondary treatment failure with Sb occur in immunosuppressed patients due to their diminished immune response, which promotes parasite multiplication and hampers the efficacy of antimonial drugs, facilitating resistance development and subsequent relapses in leishmaniasis after treatment [37,38].Next, we evaluated the expression of key genes involved in resistance against Sb and MTX.One of the most significant findings of this study is the clear evidence that pre-exposure to Sb results in a notable increase in the EC 50 against MTX, and conversely, pre-exposure to MTX leads to a similar increase in the EC 50 against Sb.Previous in vitro studies have demonstrated that exposure of parasites to Sb can lead to the co-amplification of the mrpA and ptr1 coding genes.These genes are in proximity (chromosome 23), and their amplification can occur through rearrangements within the same intergenic regions [34].Our analyses of field isolates did not reveal overexpression of either of these two genes.However, we were able to identify a clear upregulation of dhfr in both MTX-resistant isolates.This finding is consistent with the previous study conducted by Rastrojo and colleagues, where they demonstrated the overexpression of the dhfr transcript gene in a Sb-resistant strain of L. donovani [39].While some studies have reported the upregulation of dhfr in Sb-resistant strains, indicating its potential involvement in resistance, other studies have not observed significant changes in dhfr expression levels.It is important to note that drug resistance in Leishmania (as illustrated in this work) is multifactorial and can involve a combination of mechanisms, including alterations in drug transporters, drug metabolism, DNA repair mechanisms, and drug target modifications.Further research and investigation are required to fully elucidate the potential role of DHFR in Sb resistance and to better understand its significance in the overall resistance mechanism.
Finally, to better understand the interactions of Sb and MTX with Leishmania proteins in these two clinical isolates-as well as the potential mechanisms of cross-resistance-, we used a TPP-TR recently implemented for Leishmania parasites [25].TPP analysis conducted in the presence of Sb indicated that proteins in the Sb-sensitive strain LiFS-A were associated with the activation of the mitochondrial respiratory chain.Within this group of proteins, we identified an NADH oxidase protein that it is known to increase its expression following exposure to Sb, resulting in an excessive production of superoxide [40].Additionally, a mitochondrial cytochrome C1 protein was discovered, which, upon binding to Sb, could disrupt ATP synthesis, ultimately leading to the elimination of the parasites.Furthermore, we observed a highly thermally stable iron-sulfur protein that is potentially associated with the inhibition of thiol metabolism.On the other hand, the analysis of the meltome of the Sb-resistant isolate (LiFS-B) revealed significant stabilization of an alanine-tRNA ligase in the presence of Sb, along with three ribosomal proteins.These four proteins play a crucial role in ribosomal biogenesis and protein synthesis.During the late stage of promastigote differentiation, there is typically a factors, and heat shock proteins.Additionally, 14 proteins were recognized as common to both strains subsequent to their interaction with MTX, highlighting a prevalence of ribosomal proteins and those associated with the parasite's cellular respiration.https://doi.org/10.1371/journal.pntd.0012015.g005PLOS NEGLECTED TROPICAL DISEASES PLOS NEGLECTED TROPICAL DISEASES PLOS NEGLECTED TROPICAL DISEASES metabolic stabilization accompanied by a decrease in the abundance of ribosomal proteins and tRNA synthetases [41].However, variations in protein abundance could indicate a heightened metabolic activity and functional adaptation to external factors [42], such as drug pressure.Additionally, ribosomal proteins, along with translational proteins and others, contribute to the proliferation of promastigotes, allowing them to evade the host immune response and increasing their virulence [43], which is supported by findings showing that Sb-resistant field strains display increased virulence [44].It is important to note that using the TPP-TR approach we cannot identify any direct interaction of MRPA with Sb, as this must be conjugated with thiols to bid the ABC transporter [9].However, we were able to prove that mrpA expression levels are higher in the LiFS-B isolate.In TPP experiments with MTX, we observed a significant increase in various proteins involved in mitochondrial processes in both field isolates.This heightened activity could help in detoxifying the effects of MTX, maintaining cellular homeostasis, and potentially activating compensatory pathways that mitigate MTX's impact.In this way, LiFS-A's meltome in the presence of MTX displayed an enrichment in both cytochrome c reductase activity and translation.By increasing cytochrome c reductase activity, Leishmania might improve the efficiency of its mitochondrial electron transport chain, maintaining ATP production and protecting the mitochondria from damage caused by ROS [45].Enhanced translation in Leishmania in response to MTX exposure might be a compensatory mechanism to increase the production of proteins necessary for DNA repair, detoxification, and stress response.On the other hand, MTX induced enrichment in carboxylic acid metabolism proteins in LiFS-B's meltome.This observation aligns with various prior studies which have identified that modifications in carboxylic acid metabolism, through either gene amplification or changes in enzyme functionality, can contribute to the development of MTX resistance in Leishmania [46,47].Our analyses also pinpointed the interaction between PTR1 and MTX in LiFS-A.MTX competitively inhibits DHFR, which is responsible for converting dihydrofolate (DHF) to tetrahydrofolate (THF), an essential cofactor in the synthesis of nucleotides.However, PTR1 can convert DHF back to THF, thus bypassing the inhibitory action of MTX on DHFR [22,48].In addition, we observed higher thermal stabilization of other important proteins, including an ATP synthase subunit beta and an activator of HSP90 ATPase.These findings lead us to hypothesize that there is an increase in the proton gradient during the entry of MTX into the parasites via FBT transport, which may explain the increased stabilization of ATP synthase in at least one of the resistant isolates.Markedly, the observed increase in these chaperones may be linked to an adaptation to stress by promoting protein folding and stability of certain target proteins involved in MTX metabolism, transport, or detoxification pathways.
In summary, our study highlights the potential risk of Sb-MTX cross-resistance selection when L. infantum parasites are exposed to either of these drugs.This finding may help explain the relapses of visceral leishmaniasis observed in immunosuppressed patients treated with MTX.Importantly, this new knowledge has the potential to inform the development of more tailored immunosuppression regimens, thereby reducing the risk of selecting and spreading drug-resistant parasites, particularly in endemic areas.Additionally, we have provided a comprehensive list of potential Sb-and MTX-interacting proteins and pathways that could be further explored as targets for therapeutic interventions and as biomarkers of drug resistance in future studies.

Fig 3 .
Fig 3. Phenotypic characterization of L. infantum WT reference strain and LiFS-A and LiFS-B clinical isolates after 'pre-exposure' to Sb and MTX.Five days after 'pre-exposing' LiWT (A-B), LiFS-A (C-D), LiFS-B (E-F) promastigotes to the EC 50 and EC 90 (previously calculated; Table1) of Sb or MTX, parasites were submitted to increasing concentrations of MTX and Sb to evaluate potential changes in their phenotype against these drugs.EC 50 values were calculated from concentration-response curves performed with biological triplicates after nonlinear fitting with GraphPad Prism 10 software.

Fig 4 .Fig 5 .
Fig 4. Heat map representation (row Z-score) of the general thermal stability of LiFS-A soluble protein cell extracts.Normalized protein abundance of LiFS-A proteins for which full melting curves were acquired in the absence (A) or in the presence (B) of 100 μM Sb (118 proteins) and in the absence (C) or in the presence (D) of 100 μM MTX (84 proteins).Color range depicts the relative protein abundance of the soluble fractions at different temperatures.Heat maps were generated through the Heat mapper webserver (www.heatmapper.ca/expression)using its protein expression plugin with average linkage as clustering method applied to rows and Euclidean as distance measurement method.https://doi.org/10.1371/journal.pntd.0012015.g004

Table 2 . Summary of proteins identified in Sb-treated and MTX-treated LiFS-A, demonstrating a positive temperature shift.
Proteins that are common between the strains LiFS-A and LiFS-B are highlighted in bold for easy identification.

Table 3 . Summary of proteins identified in Sb-treated and MTX-treated LiFS-B, demonstrating a positive temperature shift.
Proteins that are common between the strains LiFS-B and LiFS-A are highlighted in bold for easy identification.