Introduction

Leishmaniasis is a complex set of diseases caused by protozoan parasites of the genus Leishmania, transmitted to humans through the bite of infected female sandflies1. Leishmania amazonensis is among the most common species associated with patients presenting either with localized cutaneous leishmaniasis or with diffuse cutaneous leishmaniasis (DCL), a rare yet severe form of the disease2.

The biology of Leishmania infection has been broadly defined in macrophages. In vivo, during the insect blood meal, infective promastigotes released from the digestive apparatus of the sandflies are engulfed by the host resident macrophages at the site of the insect bite; after phagocytosis, engulfed parasites transform into replicative amastigotes, which survive and multiply inside vacuolar structures termed parasitophorous vacuoles (PV)3. PV are highly acidic compartments containing multiple lytic enzymes including hydrolases, cathepsins, and ATP-dependent proton pumps (e.g., V-ATPases)4,5 formed by the fusion of trafficking endosomes with lysosomes, and by extensions of the endoplasmic reticulum6. Throughout evolution, in order to subvert such untoward conditions, Leishmania parasites have developed several mechanisms that enable survival and proliferation inside these PV. A classic example is described by parasites of Leishmania mexicana complex (L. amazonensis, L. mexicana, L. pifanoi, L. venezuelensis) that form large PV to attenuate the deleterious effect of inducible nitric oxide synthase and nitric oxide (iNOS/NO) in parasite multiplication7. As a countermeasure, host cells respond to parasite invasion by modulating multiple intracellular pathways to control parasite growth. For instance, macrophages infected with L. amazonensis up-regulate the transcription of the LYST/Beige gene, which encodes the lysosomal trafficking regulator LYST protein, to contain PV formation that restricts parasite growth7,8. Recently, our groups have described another host-parasite molecular mechanism that modulates the immune response by inducing the expression of CD200, an immune checkpoint inhibitor that restricts macrophage activation in vitro and reduces parasite load in vivo9,10,11.

Although macrophages are generally considered the primary host cells, other cells—such as dendritic cells and/or lymphocytes—can be targeted by Leishmania infection12. B-cells have recently gained attention in studies of human DCL due to their elevated levels in skin lesions, which have been postulated to regulate macrophage phenotype to become permissive to parasite proliferation13. Despite this empirical finding, the full contribution of B-cells to the establishment and/or progression of leishmaniasis remains unclear. B-1 cells in particular, which are found mainly in the peritoneal and pleural cavities14,15, comprise an ancient subset of B-cells featuring high expression of natural antibodies (IgM), co-expression of lymphoid and myeloid markers (e.g., CD19 and CD11b), along with the production of large amounts of interleukin (IL)−1016. The importance of B-1 cells, which are divided into two subpopulations, B-1a and B-1b17, has been primarily elucidated in studies using the BALB/XID mouse model, which is deficient in B1-cells, especially the B-1a population18,19.

In vitro, B-1b cells differentiate into adherent mononuclear phagocytes20, also known as B-1 cell-derived phagocytes (henceforth termed B-1P cells here), which can be distinguished from monocytes-derived macrophages by the co-expression of lymphoid and myeloid markers (i.e., IgM+ CD19+F4/80+CD11b+)21.

Previous studies have demonstrated the involvement of B-1 cells in the infection process of various pathogens, including L. amazonensis and other Leishmania species22,23,24. Interestingly, these cells respond to L. amazonensis infection by producing high levels of IL-10 in vitro25,26; however, IL-10 production in infected B-1P cells and the participation of such cells in the progression of leishmaniasis remains largely elusive.

Here we define an as yet unrecognized mechanistic role for B-1P cells in Leishmania infection. In vitro, we show that B-1P cells infected by Leishmania parasites are unable to control parasite proliferation by defective mechanisms of PV maturation associated with impaired recruitment of lysosomes carrying V-ATPase, resulting in the formation of uniquely large and non-acidic vacuoles. These events are independent of the up-regulation of LYST/Beige transcripts or by the production of IL-10. In vivo, we demonstrate that B-1P cells are present at the sites of infection and validate that these cells succumb to the formation of large non-acid compartments harboring parasite multiplication. Together, these results show that B-1P cells have morphologic and functional participation in the progression of L. amazonensis infection, revealing new immunobiological mechanisms of the host-parasite interplay with potential targets for diagnostic application and therapeutic intervention.

Materials and methods

Experimental animals

Female mice of 8 to 12-week-old C57BL/6 strain were used. The animals were kept with no water or dietary restriction, in light/dark cycles under a temperature of 22–25 °C, in the experimental animal facility at the Department of Parasitology, University of São Paulo. All experiments were carried out in accordance with international ARRIVE guidelines and in full agreement with institutional and local regulations [Brazilian Federal Law #11,794, Decree #6,899 and Normative Resolutions published by the National Council for the Control of Animal Experimentation (CONCEA)] and approved by the Institutional Animal Care and Use Committee (IACUC) under protocols #104/2013 and #95/2013 at the Institute of Biomedical Sciences, University of São Paulo, Brazil.

Bone marrow-derived macrophages isolation

Bone marrow-derived macrophages (BMM) were prepared as described10. Briefly, BMM from C57BL/6 mice were obtained after 7 days of bone marrow differentiation in RPMI medium (Vitrocell) supplemented with supplemented with 25 mM HEPES, 1.5 g/L sodium bicarbonate, 1 mM sodium pyruvate, 2 mM L-glutamine, 100 U/mL penicillin, 100 µg/mL streptomycin, 20% (v/v) heat-inactivated fetal bovine serum (iFBS, from Vitrocell) and 20% (v/v) L-929 cell conditioned medium, pH 7.2, at 37 °C in a 5% CO2 standard humidified incubator.

Peritoneal macrophages and B-1P cells isolation

B-1P cells were obtained from C57BL/6 mice by using an adapted two-step differentiation protocol20. Step One: Peritoneal cavity cells were collected by washes with RPMI-1640 medium supplemented with 25 mM HEPES, 1.5 g/L sodium bicarbonate, 1 mM sodium pyruvate, 2 mM L-glutamine, 100 U/mL penicillin, 100 µg/mL streptomycin, without iFBS, pH 7.2. The cells were incubated for 1 h at 37 °C in a 5% CO2 standard humidified incubator. Non-adherent cells were removed, and adherent monolayers were incubated for 5 days in RPMI medium supplemented with 10% (v/v) iFBS. Step Two: After the 5 days of incubation, the adherent cells were used as peritoneal macrophages. For B-1P cells differentiation, the cells present in the supernatant were then incubated in RPMI medium containing 10% (v/v) iFBS and 50% (v/v) L-929 cells conditioned medium for at least 24 h. The number of adhered B-1P cells (37.14% average from the total) was determined by subtracting the total peritoneal cells initially plated minus the number of non-adherent cells counted in the supernatant.

RAW cell line culture

RAW 264.7 macrophage-like cells (American Tissue Type Collection, ATCC) were maintained in RPMI medium supplemented with supplemented with 25 mM HEPES, 1.5 g/L sodium bicarbonate, 1 mM sodium pyruvate, 2 mM L-glutamine, 100 U/mL penicillin, 100 µg/mL streptomycin, 10% (v/v) iFBS, pH 7.2, at 37 °C in a 5% CO2 standard humidified incubator. Subcultures were prepared by washing the cells twice with PBS, followed by the addition of trypsin for 5 min, then dilution in complete medium and thoroughly re-suspension.

Leishmania amazonensis parasites

Leishmania amazonensis (IFLA/BR/67/PH8) parasites were obtained from lesions in C57BL/6 mice and then propagated as promastigotes in M199 medium (Vitrocell) supplemented with 40 mM HEPES, 2.5 µg/mL hemin, 10 mM adenine, 2 mM L-glutamine, 2 µg/mL D-biotin, 100 U/mL penicillin, 100 µg/mL streptomycin and 20% (v/v) iFBS, pH 7.2, at 26 °C in a bio-oxygen demand incubator, as described27,28. Subcultures were made weekly at an initial density of 5 × 105 promastigotes/mL up to six passages. Parasites were washed three times in phosphate-buffered saline (PBS) before use in experiments.

In vitro infection

The cells were seeded at least 24 h prior to infection. For immunofluorescence assays, the cells were seeded on top of glass coverslips in 24-well plates. L. amazonensis promastigotes were added at a multiplicity of infection (MOI) of 2 in RPMI supplemented with 5% (v/v) iFBS and 2% (v/v) L-929 cell supernatant for 2 h at 34 °C in a 5% CO2 standard humidified incubator. The cells were then washed three times with PBS to remove non-internalized parasites, followed by the addition of RPMI supplemented with 10% (v/v) iFBS and 2% (v/v) L-929 cell supernatant and incubated at 34 °C in a 5% CO2 humidified incubator for the indicated periods.

In vivo infection and tissue processing

Wild-type and TLR9−/− mice were injected with 106 L. amazonensis stationary phase promastigotes in the left hind footpad. Lesion size progression was followed by weekly measurements with a caliper by an observer blinded to the experimental group assignments. The parasite load in the lesions was quantified by a limiting dilution assay. Footpad tissue homogenates were dissociated by incubation with collagenase (2 mg/mL, Sigma-Aldrich; St. Louis, MO) in Tyrode buffer (140 mM NaCl, 5 mM KCl, 2.5 mM CaCl2, 10 mM HEPES, 2 mM MgCl2, pH 7.2) for 2 h at 37 °C under mild agitation. The homogenate was filtered through a 70-mm pore-size cell strainer (Falcon®; Corning, NY), centrifuged at 20 x g for 5 min, and washed twice with PBS (230 x g for 10 min). The cellular suspension was diluted 100-fold in complete M199 media, followed by another 10-fold serial dilution in 96-well plates performed in triplicates. After 10 days of incubation at 26 °C, the presence or absence of viable parasites in each well was determined by direct observation under an inverted light microscope. Results are presented in logarithm scale in base 10 (Log10) of the highest dilution in which viable promastigotes were observed. In some experiments, 2.5 × 107 amastigotes were injected in the peritoneal cavity and cells were collected by peritoneal washes with RPMI-1640 medium after 48 h of infection.

In some experiments, footpads of animals were removed and fixed in freshly made 10% formaldehyde (v/v) in PBS, followed by dehydration and paraffin embedding. Serial 5-µm sections were stained by hematoxylin and eosin (H&E, Sigma Aldrich). The histological sections were evaluated and captured under a light microscope (AxioImager M2, Zeiss) coupled with a high-resolution camera (AxioCam HRc, Zeiss).

Phenotypic characterization of B-1 and B-1P cells by flow cytometry

Cells isolated from footpad lesions, peritoneal cells, or the non-adherent cells fraction of five days of culture (B-1 cells) were counted, suspended in PBS containing 2% FBS and incubated with an anti-mouse CD16/CD32 antibodies (BD Pharmingen) to block Fc receptors, followed by staining with fluorochrome-conjugated mAb for surface markers, namely: phycoerythrin (PE)-labeled anti-mouse CD19 (Caltag Medsystems), Pacific Blue (PB)-labeled anti-mouse F4/80 (Thermo Scientific), and fluorescein isothiocyanate (FITC)-labeled anti-mouse CD23 (BD-Pharmingen). Fluorescence-minus-one (FMO) controls were used to determine the cut-off point between background fluorescence and positive populations in multicolor flow cytometry experiments. All the surface markers were incubated for 30 min at 4 °C in the dark. After washes, stained cells were fixed with 4% paraformaldehyde solution for 15 min at room temperature (RT). Cells were then washed, and incubated in PBS containing 2% SFB and then the samples were acquired by using an LSRFortessa™ flow cytometer (BD Bioscience) or an Attune Acoustic Focusing Flow Cytometer (Applied Biosystems) acquiring at least 50,000 events. Data were analyzed by using the FlowJo software, version 10.0.7 (Tree Star Inc. Ashland, OR, USA) as shown (Fig. S1).

Drug treatment in vitro

To evaluate the potential association between the susceptibility to infection with the size of PV, B-1P cells and BMM were pretreated with 1 µM of vacuolin-1 (compound 5114069, Cayman Chemical) for 1 h. The cells were then washed with fresh medium and infected. The phagocytosis capacities of B-1P cells and BMM were evaluated by pretreating the cells with 5 mM latrunculin A (Sigma-Aldrich) for 30 min at 37 °C. After the treatment, the cells were washed with fresh medium and infected on ice for 30 min to allow parasite attachment, followed by incubation of the cells at 34 °C for an additional 30 min.

Vacuole pH detection with acridine orange

Infected cells were incubated for 15 min at 34 °C with RPMI medium containing acridine orange (AO) [2 µg/mL 3,6-bis-(dimethylamino) acridine; Sigma-Aldrich] in 5% CO2. After washes with PBS to remove excess AO, the cells were incubated in RPMI with 5% FBS medium supplemented with 2% L-929 conditioned media, and allowed to settle at 37 °C in a 5% CO2 humidified incubator for 30 min before imaging.

Immunofluorescence and image analyses

For the immunofluorescence assays, the samples were processed and stained using two strategies. For staining and counting, coverslips containing the infected cells were washed and fixed with 100% ice-cold methanol for 5 min. After fixation, the samples were treated with blocking and permeabilization solution [1% BSA, 0.1% saponin, and 0.1% sodium azide in 25 mM Tris-buffered saline (TBS)] for 30 min. The coverslips were incubated with anti-LAMP-1 polyclonal antibody (Developmental Studies Hybridoma Bank) followed by anti-rat IgG antibody Alexa Fluor 568-conjugated (Thermo Scientific); anti-Leishmania serum followed by anti-rabbit IgG antibody Alexa Fluor 488-conjugated (Thermo Scientific); PE-labeled anti-mouse CD19 (BD Bioscience); FITC-labeled anti-mouse IgM (BD Bioscience); anti-EEA-1 (BD Bioscience) polyclonal antibody followed by anti-mouse IgG antibody Alexa Fluor 488 (Thermo Scientific); and anti-V-ATPase polyclonal antibody (Anti-V-ATP6V0D2, Sigma Aldrich) followed by anti-rabbit IgG antibody Alexa Fluor 488 (Thermo Scientific). Cells were incubated with 10 µg/mL DAPI (Sigma-Aldrich) to detect the nuclei of parasites and host cells. For the assessment of phagocytosis, the samples were processed without permeabilization (staining only OUT parasites) and incubated with anti-Leishmania polyclonal antibody followed by anti-rabbit IgG antibody Alexa Fluor 488 and 10 µg/mL propidium iodide (PI; Sigma-Aldrich). The images were randomly acquired (an observer blinded to the experimental group assignments) in a fluorescence microscope (Leica DMI6000B/AF6000) coupled to a digital camera system (DFC365FX) moving through visual fields in parallel rows across each coverslip, and analyzed by using ImageJ. An appropriate filter set was used depending on the sample fluorescence labeling. The regions of interest (ROIs) were delineated manually to measure diameter (Fig. 3) or fluorescence intensity. The number of events analyzed is annotated in corresponding figure legends.

Fig. 1
figure 1

B-1P cells are highly permissive to L. amazonensis proliferation. (A) Infection of B-1P cells and BMM for 2 h with L. amazonensis stationary-phase promastigotes (MOI = 2). The results are expressed as Leishmania per 100 infected cells and correspond to the mean ± standard deviation (SD) of three independent biological assays. No statistical significance (n.s.) was observed (Student’s t-test). See also Fig. S1 for additional data. (B) Leishmania proliferation in B-1P cells and BMM after 48 h. The results are expressed as Leishmania per infected cell and correspond to the mean ± SD of triplicates. **, p < 0.003 (Student’s t-test). (C) Representative images of the in vitro infection assay (48 h) are presented in (D) and (E). By immunofluorescence, parasites are visualized in green, PV in red (LAMP-1), and the nuclei of the cells/parasites in blue (DAPI). All the channels were merged with phase contrast. The high number of parasites contained in a PV in B-1P cells is indicated (white arrow). Scale bar, 25 μm. (D) Quantification of the total of infected cells (% of infection) is shown in (C). Data correspond to the mean ± SD of biological triplicate assays. *, p < 0.0389 (Student’s t-test). (E) B-1P cells and (F) BMM were pre-treated with latrunculin A (Lat A), washed, and incubated with L. amazonensis promastigotes for the phagocytosis assay. The IN/OUT parasite assays were determined and expressed by the percentage of Leishmania in each cell. Results correspond to the mean ± SD of three independent biological assays. ***, p < 0.0001 (two-way ANOVA). See also Fig. S2 for additional data. (G) Quantification of IL-10 and (H) TNF-α in the culture supernatant of B-1P cells and BMM 24 h after infection with L. amazonensis. Results correspond to the mean ± SD of three independent biological assays. No statistical significance (n.s., p = 0.3071); ***, p < 0.001 (Student’s t-test), respectively.

Full size image

Additionally, vacuole pH detection with AO images was acquired in the InCell Analyzer High Content Imaging System (GE), version 2200, by using a 20x objective. The images from the Cell Analyzer High Content Imaging System were analyzed by CellProfiler software, providing the following outputs per image. Total number of cells, total number of infected cells, number of parasites per infected cell, the ratio of infected cells to the total number of cells, and intensity of Cy3 fluorescence channel per vacuole (Fig. S5).

Infection quantification

The infection quantification was performed by examining stained slides on a fluorescence microscope with an objective lens under immersion oil (Leica DMI6000B/AF6000; 100x lens). Quantification was performed on a regular basis by an investigator blinded to the group assignments. Moving through visual fields in parallel rows across each coverslip, the number of macrophages and the number of intracellular parasites was quantified with a manual click-counter. At least 300 cells per coverslip were counted. Parasite load was determined by counting the number of intracellular parasites in at least 100 infected cells. The infection index was obtained by multiplying the percentage of infection per the average number of intracellular parasites per cell. In the phagocytic capacity assay, the attached parasites (OUT) were visualized by microscopy in green and red (anti-Leishmania and Nuclei PI-staining, respectively), while internalized (IN) parasites were visualized only in red (Propidium iodide, PI).

Cytokine quantification

Cell culture supernatants were used for cytokines and/or nitrite quantification. IL-10 and TNF-α were assayed by using the Cytometric Bead Array (CBA) kit for Mouse inflammation (BD Biosciences). The sample acquisitions were carried out on FACSCanto II (BD Biosciences) and the data analysis was performed with the FACSDiva™ software.

Quantitative real-time PCR

The RNA extraction and purification were performed by using the Quick-RNA™ MiniPrep kit (Zymo Research). Quantification was carried out in a NanoDrop instrument (Thermo Scientific) and cDNA synthesis was performed by using a kit (SuperScript III Reverse Transcriptase, Invitrogen). Quantitative PCR reaction was performed on the equipment StepOnePlus™ Real-Time PCR System (Applied Biosystems) by using the Maxima SYBR Green/ROX qPCR Master Mix (Thermo Scientific). Oligonucleotide primers, forward: 5’-AGC AGA AGG TGA TAG ACC AGA A-3’ and reverse: 5’-CCC ACA CTT GGA TCA TCA ATG C-3’ were used to amplify a portion of LYST/Beige cDNA; and forward: 5’-TCA GTC AAC GGG GGA CAT AAA-3’ and reverse: 5’-GGG GTC GTA CTG CTT AAC CAG-3’ to amplify the housekeeping gene (Hprt1). The reaction was incubated for 10 min at 95 °C and then for 40 cycles of 15 s at 95 °C, followed by 30 s at 55 °C and 30 s at 72 °C. Fluorescence was detected at each annealing step. Technical triplicates were performed for each reaction and negative controls were included. The data were presented as relative quantification normalized by Hprt1 expression levels calculated through the 2-ΔΔCt methodology29.

Western blot

A total of 20 µg of protein/well were separated under reducing conditions on a 12% sodium dodecyl sulfate (SDS) polyacrylamide gel and blotted onto nitrocellulose membranes with a transfer system (BioRad Laboratories, Hercules; CA). The membranes were probed with an anti-LAMP-1 polyclonal antibody followed by peroxidase-conjugated anti-rat IgG mAb (Imuny-VBP Biotecnologia) or anti-V-ATPase polyclonal antibody (Anti-V-ATP6V0D2, Sigma Aldrich) followed by peroxidase-conjugated anti-rabbit IgG mAb (Imuny-VBP Biotecnologia). Immunoblots were developed by using the Supersignal West Pico Chemiluminescent Substrate (Thermo Scientific) and detected with a ChemiDoc Imaging System (BioRad) on the ImageLab (BioRad) software.

Statistical analysis

Data were analyzed with Prism 6.0 (GraphPad Software, San Diego, CA). Statistical significance was determined by the Student’s t-test (two-tailed), two-way analysis of variance (ANOVA), or ANOVA one-way for data with Gaussian distribution and similar variation between groups. After the test for normal and log non-normal distribution, non-Gaussian data were analyzed by Mann-Whitney or Kruskal-Wallis, or Wilcoxon test, as indicated. Statistically significant differences were defined as * when p-values were < 0.05, ** p < 0.01 and *** p < 0.001. Results represent means ± standard deviation (SD) or standard error of the mean (SEM) as indicated. The number of independent experiments, and technical and biological replicates are indicated in corresponding figure legends. Biological replicates mean different biological samples (e.g., different mice, different cell cultures, different protein preparations). Technical replicates mean different measurements with the same biological sample. Independent experiments are experiments performed on different days with different biological samples. The number of mice per group is annotated in the corresponding figure legends.

Results

B-1P cells are permissive to L. amazonensis infection in vitro

B-1 cells were isolated from the peritoneal cavity of mice as described20. Here we adapted an experimental two-step cell adhesion protocol (Fig. S1, A-C) to enrich the population of B-1 phagocyte precursors (B-1b cells) for the differentiation into B-1P cells in vitro (Fig. S1 D). B-1 enriched cell populations were confirmed by flow cytometry assays with specific cell surface markers (Fig. S1 E). In these defined experimental conditions, we first established the baseline infection rate and parasite multiplicity in B-1P cells, side-by-side with bone marrow-derived macrophages (BMM), the gold-standard model of infection. We observed that despite no difference in the percentage of infection between B-1P cells and BMM after 2 h of infection (Fig. 1A), parasite multiplicity in B-1P cells was almost three times higher than BMM after 48 h of infection (Fig. 1B). Immunofluorescence staining assays with specific antibodies against intracellular amastigotes or lysosomal-associated membrane protein-1 (LAMP-1) and 4′,6-diamidino-2-phenylindole (DAPI) show that B-1P cells harbor a high number of amastigotes inside of large vacuolar structures (Fig. 1C) relative to BMM (Fig. 1D).

Fig. 2
figure 2

Large Leishmania-PV in B-1P cells is independent of the LYST/Beige gene. B-1P cells and BMM were infected with stationary-phase promastigotes of L. amazonensis for 2 h, washed, and incubated for either 24–96 h. (A) Phase contrast images showing higher magnifications of Leishmania-containing PV in B-1P cells or BMM. Schematic representation of Leishmania-PV diameter measurement. Scale bar, 10 μm. (B) Quantification of the diameter of Leishmania-PV in B-1P cells and BMM, after 96 h. The values represent the mean of 80 independent PV measurements. Results correspond to the mean ± SD of three independent biological assays. ***, p < 0.001. (C) Level of LYST/Beige transcripts in B-1P cells and BMM analyzed by qPCR. Results correspond to the mean ± SD of three independent biological assays. ***, p < 0.001 (Student’s t-test). (D) Representative phase contrast images showing the effect of vacuolin-1 (1 µM) in the Leishmania-PV (white arrows) from B-1P cells and BMM, when compared to images in (A). Scale bar, 10 μm. (E) Quantification of Leishmania-PV diameter in B-1P cells and BMM, untreated (ctr) or pretreated cells with vacuolin-1 (vac-1) and measured after 96 h of infection. The values represent the mean of 50 independent PV measurements, in each condition. Results correspond to the mean ± SD of three independent biological assays. *, p < 0.05; ***, p < 0.05 (one-way ANOVA). (F) Quantification of the infection index from the assay in (D). Data correspond to the mean ± SD of three independent biological assays. *, p < 0.05 (one-way ANOVA).

Full size image

To evaluate whether the effect in parasite multiplicity is associated with the process of invasion, either B-1P cells or BMM were first pretreated with latrunculin A (Lat A), which inhibits phagocytosis by preventing actin polymerization; then the number of intracellular (IN) versus extracellular parasites (OUT) was determined by using immunofluorescence staining in non-permeabilized cells (Fig. S2 A). Lat A treatment inhibited parasite entry in > 80% of B-1P cells (Fig. 1E) and BMM (Fig. 1F), thereby suggesting that B-1P cells (similarly to macrophages) promoted Leishmania phagocytosis. Other similarities between B-1P cells and BMM were observed regarding the polarity of parasite entry during phagocytosis (Fig. S2, B and C) and early recruitment of intracellular vesicles, such as endosomes and lysosomes, in the biogenesis of PV (Fig. S2, D-F)30,31,32,33. Finally, since B-1P cells originate from B-1 cells, one of the main sources of the anti-inflammatory cytokine interleukin-10 (IL-10)34, we also evaluated the levels of anti- or pro-inflammatory cytokines in the supernatant of B-1P cells or BMM infected cells. No detectable differences in the levels of IL-10 were observed in B-1P cells or BMM infected with L. amazonensis (Fig. 1G), data indicative that the susceptibility to infection of B-1P cells associated with the large PV is apparently independent of IL-10 production. In contrast, the levels of TNF-α, a pro-inflammatory cytokine involved in the defense against Leishmania infection35, were markedly low in B-1P cells relative to BMM (Fig. 1H), suggesting that these cells are either unable or inefficient in their molecular response against the parasite. Other cytokines such as MCP-1 and IL-6 were also evaluated but remained at background levels (data not shown). These results indicate that B-1P cells functionally behave as conventional macrophages in the process of Leishmania infection; however, these cells are highly susceptible to parasite proliferation, independently of IL-10, suggesting that other mechanisms associated with the intracellular trafficking of vesicles may be involved in the biogenesis of these large Leishmania-PV.

PV expansion is independent of the LYST/Beige gene in B-1P cells

To evaluate the anatomical and functional attributes of the Leishmania-PV in B-1P cells, we performed a series of in vitro infection assays for microscopy and imaging analysis. First, we used a well-established method to determine the size of Leishmania-PV by measuring the largest diameter of the vacuole7,36. After 96 h of infection, the Leishmania-PV (arrows) grew abnormally large in B-1P cells when compared in side-by-side experiments with BMM (Fig. 2A, note schematic representation at the right side of each corresponding image), with some vacuoles reaching over 25 μm in diameter (average, 15.3 ± 0.3 μm). In contrast, the Leishmania-PV of infected BMM grew smaller than B-1P cells with only a few cells containing vacuoles of ~ 20 μm in diameter (average, 12.8 ± 0.2 μm) (Fig. 2B).

Fig. 3
figure 3

Impaired recruitment of V-ATPase to the Leishmania-PV in B-1P cells. B-1P and RAW cells were infected with stationary-phase promastigotes and stained with acridine orange (AO) and Hoechst for DNA staining in live microscope imaging assays. (A) Representative images of B-1P cells containing Leishmania-PV and stained with AO and Hoechst (nuclei of cells) showing neutral/basic pH (low-AO signal) or acidic pH (high-AO signal). The panel at right represents a merged image of the AO red channel merged (at left) including the phase contrast image. White arrows and punctuated circles show Leishmania-PV. Scale bar, 10 μm. (B) Percentage of Leishmania-PV containing low/high pH identified in (A). Results correspond to the mean ± SD of three independent biological assays. ***, p < 0.0004 (two-way ANOVA). (C) Number of Leishmania-PV per 100 infected B-1P or RAW cells after 48 h of infection. No statistical significance (n.s). (D) Representative images of Leishmania-PV (PV) in B-1P cells (top images) and BMM (bottom images) showing V-ATPase (green) staining marker. In the inset, the phase contrast is shown with white arrows indicating the PV. V-ATPase image was merged with DAPI in the panel on the right. Punctuated yellow circles indicate Leishmania-PV; Scale bar, 5 μm. See also Fig. S3 for additional data. (E) Representative image of the fluorescence intensity analysis in the selected area in B-1P cells (yellow circles) corresponding to the Leishmania-PV, and are represented in arbitrary units (A.U.). (F) Quantification of the recruitment of V-ATPase or (G) LAMP-1 marker to the Leishmania-PV in B-1P cells and BMM. Results correspond to the mean ± SD of three independent biological assays. *, p < 0.0427 (Wilcoxon test); no statistical significance, (n.s.).

Full size image

Next, to begin to evaluate the molecular mechanisms associated with these empiric observations, we analyzed the expression of the LYST/Beige gene, which encodes for the LYST protein responsible for regulating the size of lysosomes37. In macrophages, the gain- and loss-of-function of the LYST/Beige gene have been associated with the control of Leishmania-PV expansion; indeed, LYST/Beige upregulation markedly reduces the size of lysosomes, limiting the expansion of the Leishmania-PV and controlling parasite growth7. Thus, to investigate whether LYST/Beige gene is regulated in B-1P cells during L. amazonensis infection, total mRNA obtained from B-1P cells or BMM infected with L. amazonensis for 96 h was analyzed by quantitative PCR (qPCR). Surprisingly, the relative levels of LYST/Beige transcripts in B-1P cells (~ 2.8-fold) were nearly double those levels in BMM (~ 1.5-fold) (Fig. 2C), results indicating that B-1P cells —unlike BMM— are unable to control the expansion of Leishmania-PV via LYST. Moreover, to evaluate the potential association with LYST and intracellular trafficking of lysosomes, we next tested the expansion of Leishmania-PV in B-1P cells or BMM pretreated with vacuolin-1, a cell-permeable inhibitor of lysosomal exocytosis that causes rapid expansion of late endosomes/lysosomes38,39 and the formation of large Leishmania-PV7. Notably, vacuolin-1 barely affected the diameter of Leishmania-PV in B-1P cells, from 18.4 ± 4.4 μm (in non-treated cells) to 20.4 ± 4.0 μm (in treated cells) (Fig. 2D and E) with no detectable differences in the infection index (Fig. 2F); however, vacuolin-1 clearly increased the diameter of Leishmania-PV in BMM (from 12.6 ± 2.4 μm in non-treated cells to 18.3 ± 5.2 μm in treated cells) and the infection index (Fig. 2E and F), data indicating that expansion of Leishmania-PV facilitates parasite multiplication. These results suggest that in B-1P cells inherent inhibitory trafficking processes, rather than the size of lysosomes, induce the formation of abnormally large Leishmania-PV. Taken together, the results establish that B-1P cells are unable to control the expansion of large Leishmania-PV independently of the up-regulation of the LYST/Beige gene and confirm that the formation of large vacuolar structures accounts for parasite proliferation in B-1P cells.

Deficient recruitment of V-ATPase impairs acidification of Leishmania-PV in B-1P cells

During Leishmania-PV biogenesis, fusion with lysosomal vesicles promotes its acidification and activation of lytic enzymes to either disable or destroy intracellular parasites40. To investigate the acidification of Leishmania-PV in B-1P cells, we used the permeable fluorescent probe acridine orange (AO), which emits red fluorescence once sequestered/trapped into acidic compartments. Representative images of infected B-1P cells clearly show Leishmania-PV (white arrow and punctate yellow circle) with low (Fig. 3A, top panel) or high (Fig. 3A, bottom panel) AO red fluorescence signal concentrated in the PV. Under these experimental conditions, we determined the acidification of Leishmania-PV in B-1P cells in comparison with the murine macrophage RAW cell line after 48 h of infection. Notably, nearly half of the Leishmania-PV in B-1P cells were non-acidic (low AO signal), as opposed to most (~ 90%) of the highly acidic (high AO signal) Leishmania-PV in RAW cells (Fig. 3B). No difference in the number of PV per infected cell was observed (Fig. 3C). These results indicate that the process of acidification and maturation of the Leishmania-PV is even more defective and/or delayed in infected B-1P cells than previously reported41. This striking functional feature represents the first example of an entirely new mechanism of parasite resistance for long periods of infection, which has not as yet been reported for any other host cells, including but not limited to macrophages.

To further investigate the defect in the acidification of Leishmania-PV in B-1P cells, we next evaluated the presence of the ATP-dependent proton pump V-ATPase and the structural lysosome protein LAMP-1, a positive control for the lysosomal recruitment and acidification in Leishmania-PV by immunofluorescence microscopy. In these experiments, B-1P cells or BMM were infected with L. amazonensis for 96 h, fixed, and stained with specific antibodies to show V-ATPase recruitment to the membrane of the Leishmania-PV (white arrows, Fig. 3D and Fig. S3 A). In B-1P cells, V-ATPase was observed mostly spread out through the cell cytoplasm with only a few concentrated structures surrounding Leishmania-PV (Fig. 3D; top image, yellow arrow). In contrast, V-ATPase was clearly observed and concentrated in spots at the membrane of Leishmania-PV in BMM (Fig. 3D; bottom image). Given that expression of V-ATPase or LAMP-1 remained unaltered in either B-1P cells or BMM (infected or uninfected cells; Fig. S3, B and C), these data suggest that defective recruitment of V-ATPase impairs Leishmania-PV acidification. To demonstrate the defective recruitment of V-ATPase to the formed PV, we measured the intensity of the immunofluorescence signal of V-ATPase and LAMP-1 at the Leishmania-PV membrane through an imaging-based software (Fig. 3E). The defect in V-ATPase recruitment in the Leishmania-PV of B-1P cells was confirmed by the low-intensity fluorescence signal, compared to BMM (Fig. 3F) and to LAMP-1 (Fig. 3G), which has remained unaltered in both cells. Together the results show that the large Leishmania-PV in B-1P cells are likely non-acidic or less acidic compartments due to the impaired recruitment of V-ATPase, which is the main source of acidification in BMM. These data suggest that this highly favorable microenvironment poses no selective pressure against the parasite, thereby allowing or enabling the uncontrolled proliferation of new replicative forms of L. amazonensis.

Large non-acidic Leishmania-PV in B-1P cells favors parasitic infection and proliferation in vivo

To evaluate the contribution of B-1P cells in the progression of L. amazonensis cutaneous lesions in vivo, we infected C57BL/6 mice footpads and, after five weeks post-infection, the lesions were dissected, and the tissue processed for histology and flow cytometry. Representative images of the H&E-stained section of mice footpad lesions show the number of infected cells containing parasites inside large Leishmania-PV (Fig. 4A). To identify the surface phenotype of the infected cells, the footpad lesions were processed and stained for flow cytometry. The cells were analyzed with specific antibodies and subclassified into two main populations: B-1P cells (CD45+F4/80+CD19+) or macrophages (CD45+F4/80+CD19−) from the absolute number of cells obtained from the lesions (Fig. S4 A). As expected, most of the cells (~ 75%) of the cells obtained from the footpad were classified as macrophages (F4/80+CD19−); however, ~ 25% of the cells were classified as F4/80+ CD19+, an indication of the presence of B-1P cells in the cutaneous lesion of L. amazonensis (Fig. 4B). These results are further supported by the identification of F4/80+ CD19+ B-1P cells from the footpads of infected mice that are genetically deficient in Toll-like receptor-9 (TLR9−/−) (Fig. 4C), which are known to control the recruitment of macrophages to the site of lesions10, but still develop small inflammatory lesions. The presence of B-1P cells in TLR9−/− infected mice suggests that the migration of these cells to sites of infection may be independent of the recruitment of monocyte-derived macrophages; thus, we reasoned that B-1P cells may be important for parasite multiplication together with infected macrophages.

Fig. 4
figure 4

B-1P cells are recruited to cutaneous lesions in vivo. (A) Representative image of the lesions and a histological section of an infected footpad during the course of C57BL/6 mice infection with L. amazonensis. The inset image shows large Leishmania-PV of different cells within the infected tissue. (B) Normalized percentage of CD45+F4/80+ populations presenting B-1P cells (CD19+ cells) or macrophages (CD19− cells) from infected wild-type (WT) mice with L. amazonensis. Different cell populations obtained from footpad lesions were analyzed by flow cytometry and represented as the percentage of total cells. Results correspond to the mean ± SD (n = 4). ***, p < 0.001 (Student’s t-test). (C) Normalized percentage of CD45+F4/80+ populations presenting B-1P cells (CD19+ cells) or macrophages (CD19− cells) from infected TLR9−/− mice with L. amazonensis. Results correspond to mean ± SD (n = 4). ***, p < 0.001 (Student’s t-test). (D) Total cell fluorescence intensity (CTCF) for the presence B-1P cells (IgM+) or PM (IgM−) cell populations in comparison with non-infected (NI) or infected (I) samples from the peritoneal cavity. At least 50 cells were analyzed by using Image J software. Results correspond to the mean ± SD. ****, p < 0.0001 (Student’s t-test). (E) Flow cytometry analysis of the B-1P cells (F4/80+CD19+) cell population 48 h after Leishmania infection in the peritoneal cavity of C57BL/6 mice. (F) Peritoneal cells were stained with AO, IgM-FITC, and Hoechst. B-1P cells (IgM+), macrophages (IgM−), and lymphocytes (IgM− and small nuclei) were visualized in the representative image. Scale bar, 10 μm. (G) B-1P cells and (H) Ex-vivo analysis obtained from the peritoneal cavity of infected mice. The cells were stained and analyzed in Image J software measuring the corrected total cell fluorescence intensity (CTCF) and AO in different cell populations. At least 50 cells were analyzed. Results correspond to the mean ± SD. ***, p = 0.0003 (Student’s t-test). See also Fig. S4 for additional data.

Full size image

To further validate our observations, we adopted another experimental model of infection in which Leishmania parasites were injected directly into the peritoneal cavity, the main site of resident B-1 cells in mice15. The rationale for this approach was to enrich the population of B-1P cells targeted by the parasites and to evaluate the development of the infection in these cells ex vivo. Indeed, we observed an increase in the number of B-1P cells upon infection, with no difference in the number of peritoneal macrophages (PM) (Fig. 4D), therefore confirming that B-1P cells are either recruited or differentiated from the precursor B-1 cells at the site of infection in vivo (as previously observed in the context of the footpad infection model).

After 48 h of infection, cells from the peritoneal cavity were recovered and analyzed by flow cytometry. Three main cell populations were identified according to the expression of CD19 and F4/80 surface markers: F4/80+CD19+ (B-1P cells), F4/80+CD19− (macrophages), and F4/80−CD19+ (B-1 cells) (Fig. 4E). To further analyze the B-1P cells and macrophage population ex vivo, the cells were subsequently stained with AO and analyzed by fluorescence microscopy. The cells were also co-stained with anti-IgM antibody, which is an exclusive B cell marker and clearly differentiates B-1P cells from PM (Fig. 4F). Next, we compared the acidification profile of B-1P cells or PM, either infected (I) or not-infected (NI) by the parasites injected in the peritoneal cavity, by calculating the intensity of the corrected total cell fluorescence (CTCF) coefficients of AO in each cell population (Fig. S4, B and C). Notably, no changes in acidity were observed in B-1P cells either infected or non-infected by the parasite (Fig. 4G), as opposed to the increased acidification observed in PM infected by the parasite (Fig. 4H). Moreover, we also used an independent automated imaging software analysis tool (Operetta High Content Imaging System; Fig. S5) to evaluate whether the infection index (assessed by the staining the nucleus of the cell and the parasites – green fluorescence FITC channel) in either B-1P cells or BMM was associated to Leishmania-PV acidification (assessed by the AO dye – red fluorescence Cy3 channel) (Fig. 5A). By using a similar approach, except by infecting the differentiated cells obtained from the peritoneal cavity in vitro, we observed that the number of parasites per infected cell (Fig. 5B) and the infection rate (Fig. 5C) were higher in B-1P cells, which presented less acidification (fluorescence intensity) in the Leishmania-PV, when compared to BMM (Fig. 5D). Together, these results show that B-1P cells containing less acidic PV favor high Leishmania infection rate, presumably by the lack or defective mechanisms of intracellular trafficking.

Fig. 5
figure 5

Impaired recruitment of V-ATPase is associated with non-acidic Leishmania-PV in B-1P cells in vivo. (A) Representative images of either B-1P cells or peritoneal macrophages (PM) infected with Leishmania amastigotes and stained with AO. The images were acquired in the InCell Analyzer High Content Imaging System by using two fluorescence channels: FITC (DNA and RNA) and Cy3 (acidic compartments). The merge channel images are shown. N: cell nucleus; white arrows: Leishmania nucleus. Scale bar, 10 μm. See also Fig. S5 for additional data. (B) The scatter plot shows the mean number of parasites/infected cells. Each circle represents an individual field, and bars are mean ± SD of technical replicates. ***, p < 0.0001 (Mann-Whitney test). (C) The scatter plot shows the percentage of infection. Each circle represents an individual field, and bars are the mean ± SD of technical replicates. ***, p < 0.0001 (Mann-Whitney test). (D) The scatter plot shows vacuole mean fluorescence intensity (A.U.) in the Cy3 fluorescence channel. Each circle represents an individual vacuole, and bars are the mean ± SD of technical replicates. ***, p < 0.0001 (Mann-Whitney test). (E) Representative images showing infected cells containing Leishmania-PV (white circle) stained for V-ATPase (green) and for IgM (purple signal). Parasites (white arrows) inside PV (white circle) by phase contrast and immunofluorescence staining merge (including LAMP-1 staining in red) showing infected B-1P cells (IgM+) and PM (IgM−) from the same representative image is showed. Scale bar, 10 μm. (F) Leishmania-PV areas were selected in a random and blinded fashion for analysis of the ratio of the fluorescence intensity for V-ATPase/LAMP-1. Results correspond to the mean ± SD of replicates. ***, p < 0.0001 (Student’s t-test).

Full size image

Finally, we used the ex-vivo infected cells to evaluate the recruitment of V-ATPase and LAMP-1 to the Leishmania-PV by using immunofluorescence microscopy. As previously observed in vitro, the V-ATPase signal in Leishmania-PV (white circle) is markedly reduced in infected B-1P cells (IgM+ staining) and in sharp contrast with the representative peritoneal macrophages (PM) displayed in the same image (IgM− staining) (Fig. 5E). The relative quantification of the V-ATPase/LAMP-1 ratio indicates low co-localization of V-ATPase with LAMP-1 confirming the in vitro observations that B-1P cells are defective in recruiting V-ATPase to newly formed Leishmania-PV. In summary, our in vivo data indicate that (i) B-1P cells are indeed recruited to differential sites of infection regardless of anatomical location (e.g., footpad or peritoneal cavity); that (ii) these cells are infected by the parasite, and that (iii) the parasites reside and proliferate in large non-acidic vacuoles that are formed upon the deficiency in recruiting V-ATPase. These functional results in vivo are entirely consistent and strongly supported by our in vitro analysis showing that infection of B-1P cells by L. amazonensis favors parasite replication and survival and therefore has a mechanistic role in the progression of cutaneous lesions in vivo.

Discussion

While B cells play an essential role during infection by different pathogens, including Leishmania25,42, the specific role of B-1P cells, an exclusive subtype of B cells, is still poorly understood. Here, we show for the first time that B-1P cells are highly susceptible to parasite infection by their endogenous inability to regulate the size and acidification of Leishmania amazonensis-PV. The deficiency in reducing intracellular vesicle fusion via the LYST/Beige gene —combined with the impaired recruitment of lysosome containing V-ATPase— results in the formation of large and non-acid Leishmania-PV, which favors parasitic survival and proliferation. Even though for many other species of Leishmania, the role of B-1P cells remains an open question to be addressed in future work, this functional discovery sheds light on an as yet unknown mechanism of cellular susceptibility to intracellular pathogens in mammalian hosts.

In general, mammalian hosts have evolved to resist parasite infection through multiple mechanisms such as nitric oxide (NO) production, secretion of inflammatory cytokines, and the formation of acidic and hydrolytic vacuolar structures that control parasite proliferation43,44,45. On the other hand, parasites such as L. amazonensis have co-evolved to escape these defense mechanisms by expanding these structures into large vacuoles that attenuate the deleterious effect of hydrolysis and low pH in parasite survival. Another sophisticated example of this host-parasite functional interplay is a mechanism adopted by Leishmania to inhibit the microbicidal response of macrophages via modulation of the CD200/CD200R signaling pathways9,10. In B-1P cells, the impairment of Leishmania-PV acidification allows for parasite multiplication, thus suggesting that the potential encounter of parasites with these cells would favor disease progression, perhaps even in the most severe forms of leishmaniasis. This concept is further supported by recent findings showing infiltrating CD19+ B cells13 in lesions derived from human patients with diffuse cutaneous leishmaniasis (DLC).

Although differences in PV maturation have been documented in other cells such as neutrophils, DCs, or monocytes46, the defective attributes of B-1P cells, exploited by the L. amazonensis to facilitate survival and replication, appear to serve as an as yet unrecognized mechanism of parasite evasion. A potential non-mutually exclusive explanation is that the phagocytosis in these cells might represent an ancient cellular mechanism, which—for unknown evolutionary reasons—has not developed toward host protection. During the B-1 transition to phagocytes, the cells acquire genetic and morphological macrophage-like signatures such as F4/80. Therefore, it is plausible that these cells may also express other important ligands (e.g., MerTK, or CD36), which are known to be associated with phagocytosis47,48. If confirmed, these receptors could be partially responsible for the lack or blockage in signaling pathways molecules that connect with the PV maturation machinery49. The characterization of surface cell markers for the B-1P cells may unravel the connection with a defective PV maturation on this cell subset and its role in other pathogen infections. Recently, it has been shown that the severe acute respiratory syndrome coronavirus 2 (SARS-COV-2) inhibits autophagosome-lysosome fusion50 and egress from cells exploiting unconventional de-acidified lysosomes with inactive degradation enzymes51. Nevertheless, further studies will shed light on the function of B-1P cells as phagocytes by identifying membrane receptors, which are critical for the biological function of these cells.

While changes in acidity conditions within the PV are critical for pathogen killing and antigen presentation via major histocompatibility complex (MHC) in macrophages52, both features are paradoxically either unknown or absent in this specific B cell subtype. Indeed, the LYST/Beige gene up-regulation might suggest that B-1P cells, like macrophages, can control intracellular trafficking pathways to contain Leishmania-PV expansion. However, these cells were not only unable to control the formation of abnormally large vacuoles but also inefficient in promoting the recruitment and fusion of acidic vesicles containing V-ATPase to newly formed Leishmania-PV. Given that B-1P cells originate from residual B-1 cells, it is tempting to speculate that modifications in the cytoskeleton network during the transition into phagocytic-like cells could interfere with the intracellular trafficking of vesicles, especially the lysosomes containing V-ATPase, which are required for acidification53. Another plausible explanation would be the contribution of recycling endosome/lysosomes or Golgi-derived vesicles, which could facilitate the formation of large and underdeveloped phagosome54,55,56.

Another intriguing aspect in Leishmania biology is the different phenotypes of the PV sizes from certain Leishmania species such as L. amazonensis compared to L. chagasi, the latter associated with visceral leishmaniasis (also called “kala-azar”), an often lethal form of the disease if left untreated. More relevant is the fact that L. chagasi proliferates inside phagocytic cells in individual PV, suggesting differential mechanisms of PV biogenesis57 when compared to L. amazonensis PV. Previous data show that B-1 cells play an essential role in the host susceptibility to L. chagasi infection22, with an essential function for IL-10 in this process. However, it remains unclear whether IL-10 production is sustained after transitioning to B-1P cells and upon infection with the parasite. Notably, the in vitro data presented here suggest that these cells are likely similar to macrophages than B cells.

Finally, to obtain mechanistic insights and to explore the potential role of B-1P cells in the infection in vivo, we used two models of infection where we unequivocally show that these cells are indeed recruited to the site of lesion and targeted by the parasite, therefore suggesting a central, if as yet unappreciated, role of B-1P cells in disease progression. Although the presence of B-1 cells in inflammatory sites has been demonstrated during infection58, it is the first time that a mechanistic role of deficiency in PV maturation has been proposed to explain the susceptibility of these cells to infection and permissiveness to parasite growth.

Other aspects merit further discussion. First, this experimental evidence is also supported by an infection model in BALB/XID mice, which are naturally deficient in B-1 cells and therefore are resistant to infection by different Leishmania species22,23. Second, one should note that in cutaneous leishmaniasis, T-lymphocytes, and activated macrophages are responsible for controlling the infection59; however, in DCL, the lesions are constituted by unresponsive T-lymphocytes surrounded by a much higher number of infected macrophages60, and a marked plasma cell infiltration61. Because we show that B-1P cells migrate to inflammatory sites where they would trigger an anti-inflammatory response62,63,64, one might also speculate that the results reported here could represent a novel role of these cells, not only by allowing parasite infection and proliferation but also perhaps by contributing to the overall spread of the disease. On another line of investigation, our group has previously reported that the presence of B-1 cells is associated with malignant melanoma biology, mediated by the cell-to-cell adhesion through the melanoma cell adhesion molecule (MCAM, also known as MUC18 or CD146), by which the presence of this specific cell population promotes tumor progression and metastasis in murine models and human patients65, data attesting to the functional relevance of B-1 cells not only in human parasitology but also in cancer biology.

In conclusion, here we show that B-1P cells create an internal environment with reduced acidification and impaired recruitment of V-ATPase. We highlight a heretofore-unrecognized function of B-1P cells as alternative susceptible target cells for the intracellular parasite L. amazonensis, with differential morphologic and functional attributes from resident macrophages, which is bound to have translational implications in mechanism-based therapeutic strategies and for medical parasitology at large. Future studies shall determine the full biological and clinical relevance of B-1 lymphocytes or B-1P cells in ulcerative or nodular lesions from unfortunate human patients affected by leishmaniasis worldwide.