This study was approved and monitored by the institutional review board for ethical conduct of research involving human subjects of the Centro Internacional de Entrenamiento e Investigaciones Médicas - CIDEIM, in accordance with national (resolution 008430, República de Colombia, Ministry of Health, 1993) and international (Declaration of Helsinki and amendments, World Medical Association, Fortaleza, Brazil, October 2013) guidelines. All individuals voluntarily participated in the study and written informed consent was obtained from each participant.

Additive-free meglumine antimoniate (MA) (Walter Reed 214975AK; lot no. BLO918690-278-1A1W601) was kindly provided by the Walter Reed Army Institute, Silver Spring, MD, USA. Phorbol-12-myristate 13-acetate was purchased from Sigma–Aldrich and zinc acetate dihydrate from J.T. Baker.

Six adult patients, 18 to 65 years of age, with parasitological diagnosis of CL and time of lesion evolution <6 months, without apparent immune deficiencies (negative HIV test, no evidence of immunological disorder nor treatment with medication having immunomodulating effects), participated in this study. For in vitro primary macrophage differentiation, peripheral blood mononuclear cells (PBMCs) were obtained from study participants by separation using a Ficoll-Hypaque (Sigma-Aldrich) gradient.

The human pro-monocytic cell line THP-1 and derived lines were maintained at 1 x 10⁶ (Firdous et al., 2009) cells/mL in RPMI 1640 supplemented with 10% heat inactivated FBS, 100 μg/mL streptomycin, 100 U/mL penicillin, 5 µg/mL puromycin (only for maintenance of transfected cells lines), at 37°C and 5% CO₂. THP-1 monocytes were differentiated with 250 ng/mL of PMA for 3 hours, washed twice with D-PBS and cultured 24h in 6 well plates. Human PBMC-derived monocytes were differentiated to macrophages by adherence to cell culture plastic-ware as previously described (Dohmen et al., 2016).

Antimony susceptible L. (V.) panamensis promastigotes (MHOM/CO/2002/3594) were kept at 25°C in RPMI supplemented with 10% heat-inactivated FBS, 100 μg/mL streptomycin, 100 U/mL penicillin. Primary human macrophages and differentiated THP-1 cells were infected with human AB+ serum-opsonized stationary phase promastigotes at 10:1 Leishmania-macrophage ratio for 2h, washed twice with D-PBS and incubated for 24h at 34°C, 5% CO₂. After infection was established, cells were exposed for 24h or 48h to MA (8, 16, and 32 μg/mL), or left untreated as a control. Intracellular parasite survival was measured by RT-qPCR as previously described (Romero et al., 2010) (primers used in Supplementary Table S1 ).

Total RNA was isolated with Trizol from uninfected, infected, and drug-treated macrophages. RNA quality was assessed with an Agilent 2100 Bioanalyzer using RNAnano chips (Agilent). RNA Integrity Number (RIN) ≥ 7 was considered acceptable to continue with library construction. Poly-A enriched libraries were generated from macrophage RNA extracts using the Illumina TruSeq Standard mRNA preparation kit and checked via the Bioanalyzer and quantitative PCR (KAPA Biosystems). Paired-end reads (100 nt) were obtained using the Illumina HiSeq 1500 (BioProject ID PRJNA633893, Supplementary Table S2 ). Fastqc [Anon, (2020)] was used to evaluate sequencing quality; Trimmomatic (Bolger et al., 2014) filtered low-quality reads and trimmed bases when the mean quality score fell below a threshold phred score of 20. Reads were mapped against the human (Miga et al., 2014) (hg38) and L. V. panamensis (Aslett et al., 2010) genomes (v36ish) using tophat (Trapnell et al., 2012). HTSeq (Anders et al., 2015) was used to count reads mapping to each gene feature. The count tables were restricted to the set of protein coding genes and filtered to remove non-expressed and very weakly expressed genes. The remaining genes were assessed for significant outliers and batch effects by visualizations of normalized data. Library sizes and count densities were calculated on non-normalized data; pairwise correlation and distances, outlier detection, and principal component analysis (PCA) were performed on log₂, cpm, quantile normalized data with and without accounting for batch in the model or surrogate estimation with sva (using svaseq or combat) (Leek et al., 2012). DESeq2 (Love et al., 2014) was used to perform differential expression analyses alongside a statistically uninformed basic method as a negative control ( Supplementary Table S3 ). Differentially expressed genes were contrasted between control vs. infected and MA-treated primary human macrophages. Genes deemed significantly different according to DESeq2 (|log₂FC| > 1.0 and a FDR adjusted p-value < 0.05) were passed to various ontology tools. Enrichment PPI network analysis was carried out using STRING 10.

A lentivirus-based system was used for shRNA-mediated gene silencing in THP-1 monocytes as previously described (Zhou et al., 2012). At least two independent sets of oligonucleotide pairs for gene knockdown of human MTs and the transcription factor MTF-1 ( Supplementary Table S4 ) were synthesized and cloned into the pLKO.1-TCR vector (Addgene, Cambridge, MA, USA); the same vector was also used as an empty vector control. Lentiviral particles were generated by co-transfection of endotoxin-free hairpin-containing pLKO.1-TCR, psPAX2 and MD2.G (Addgene) into HEK-293T cells. FuGENE HD (Roche) was used as the transfection reagent. Lentivirus-containing cell supernatant was collected 4 days after transfection and subsequently used to transduce THP-1 monocytes in medium containing 10 mg/mL polybrene in a proportion 1:1. Transduced cells were selected under puromycin pressure (5 µg/mL) for a minimum of 5 days. Gene knockdown was confirmed by RT–qPCR using SYBR green (Applied Biosystems) and TaqMan® Gene Expression Assays (Applied Biosystems, Supplementary Table 1 ). shRNA transduction was confirmed by DNA sequencing.

PMA-differentiated THP-1 cells and derived cell lines were exposed to a dose range (8 μg/mL – 256 μg/mL) of MA for 72 hours. Cell viability was assessed by MTT assay (ATCC) ( Supplementary Figure S1 ).

Based on the distribution of the data, differences in gene expression were tested with one-way analysis of variance and Tukey’s multiple comparisons test. Differences in variance for the remaining experiments were analyzed with unpaired t test. A significance level of p ≤ 0.05 was used for all statistical tests. Statistical analyses were performed using GraphPad Prism software (version 6).

Peripheral blood mononuclear cell (PBMC)-derived macrophages from CL patients (n=6) were infected ex vivo with L.V. panamensis and exposed to Sb^V (32µg-Sb/mL, as meglumine antimoniate - MA). Following RNA-seq data collection, a total of 16,841 transcripts were detected. Principal Component Analysis (PCA) showed separation between uninfected/untreated control macrophages and those infected with L.V. panamensis and exposed to MA ( Figure 1A ). After filtering the differential expression (DE) data by |logFC| ≥ 2 and p ≤ 0.05, a set of 217 transcripts remained, of which 111 were up-regulated and 106 down-regulated ( Supplementary Table 3 ). Interestingly, among the top twenty up-regulated transcripts, seven were metallothionein genes: MT1-E, MT1-F, MT1-G, MT1-H, MT1-M, MT1-X and MT2-A ( Figure 1B ; Supplementary Table 3 ).

To discern how MTs participate in the activity of antimonials, THP-1 cells were exposed to Sb^V and the expression of MT1-E, MT1-F, MT1-X, MT2-A, MT3 and MT4 was quantified (primer sequences available in Supplementary Table 1 ). Expression of MT1-X was the highest, peaking at 8-fold induction over untreated cells; followed by MT1-E, MT2-A and MT1-F ( Figure 2A ). MT3 and MT4 transcripts were not detected in macrophages.

We evaluated the effect of maximal induction of MTs in intracellular Leishmania survival. THP-1 cells were exposed to a dose range of 25 µM to 400 µM Zn acetate for 24h, and peak expression of MTs (>100 fold) was observed with 400 µM Zn acetate (Namdarghanbari et al., 2011) ( Figure 2B ). Therefore, THP-1 cells were pre-treated for 24h with 400 µM Zn acetate, followed by L. V. panamensis infection for additional 24h, and exposed to increasing and non-cytotoxic concentrations of MA ( Supplementary Figure 1 ). Pre-treatment with Zn acetate increased over 40% the Sb-dependent intracellular elimination of L.V. panamensis ( Figure 2C ), supporting the contribution of MTs to this phenotype.

An initial assessment of the participation of MT2-A gene silencing on intracellular survival of Leishmania did not show any significant effect ( Supplementary Figure S2 ). This led us to hypothesize that a compensatory effect of other MTs maybe operating in our system. Taking advantage of the high sequence similarity of MT genes, an shRNA was constructed which targets the tandem knockdown -KD- (MT_tandem^KD) of MT1 and MT2 family member genes ( Figure 3A ). Expression of MT1-E, 1F, 1X and MT2-A was efficiently silenced as shown in Figure 3B . To explore the phenotypic effects of MTs KD on intracellular parasite survival, MT_tandem^KD cells were infected with L. V. panamensis and exposed to Sb^V. Despite effective KD of MT genes, intracellular parasite survival remained unchanged compared to empty-vector transfected control cells ( Figure 3C ). Considering that expression of MT genes is strongly induced by Leishmania and Sb^V, we questioned whether MTs knockdown was maintained during the experimental conditions (infection and drug exposure). Despite the efficient tandem KD of MTs at basal conditions, their strong transcriptional up-regulation during Leishmania and Sb^V exposure overcame the shRNA-silencing effect ( Figure 3D ), explaining why parasite survival was similar in MT_tandem^KD and empty-vector control cells.

Metal transcription factor-1 (MTF-1) is the principal transcription factor regulating expression of MTs genes during metal-induced stress (Heuchel et al., 1994). We explored whether knockdown of MTF-1 could limit the transcriptional induction of MT genes in our experimental conditions. Using shRNA, the steady-state level of MTF-1 was knocked down by 50% ( Figure 4A ). Expression of MTs and slc30a1 (ZnT1) in MTF-1^KD cells was evaluated under basal conditions and upon infection with L.V. panamensis and exposure to Sb^V. A slight reduction (ranging from 5% to 40%) of MT1 and MT2 genes expression was observed in uninfected and unstimulated MTF-1^KD cells ( Figure 4B ). However, induction of MT1-E, F, X and MT2-A expression was completely repressed in L.V. panamensis infected cells subsequently exposed to Sb ( Figure 4C ). In the case of slc30a1 expression, we observed no significant differences in its expression ( Supplementary Figure S3 ), suggesting that the effect of MTF-1^KD on MT expression responds from a more specific mechanisms triggered by the presence of xenobiotic metals such as aSb.

MTF-1^KD and empty vector control THP-1 cells were infected and exposed to Sb^V (8 - 32 µg/mL), and intracellular parasite survival measured by RT-qPCR. Sb-dependent parasite killing was efficient in empty-vector control cells, where parasite survival was below 50% for all Sb^V doses tested ( Figure 4D ). In contrast, Leishmania survival after drug exposure significantly increased in MTF-1^KD cells compared to empty vector control cells ( Figure 4D ), and was above 50% for all evaluated doses and up to 75% in the 8 µg/mL Sb^V dose. These data suggest that MTF-1^KD favors intracellular survival of Leishmania by impairing MTs gene expression.