Transcriptomic profile of heme-regulated genes in L. major

RNA-seq was used to identify heme-regulated genes in L. major by comparison of the transcriptomic profile of promastigotes grown in the presence or absence of heme. Flow cytometry analysis confirmed the induced ability of parasites to take up porphyrins after heme depletion (around six-fold, Fig 1A). Data analysis revealed that 2061 genes exhibited a FDR (False Discovery Rate) ≤ 0.05 and were considered differentially expressed (DEGs) (S1 Table). Fig 1B shows Volcano plot comparing the fold changes in expression (Log2) and the corresponding adjusted p-values (-log10). Of the 2061 DEGs, 899 were upregulated, whereas 1162 were downregulated. Furthermore, 36% of these DEGs were hypothetical proteins without predicted function, supporting the importance of further studies to characterize these genes. As shown in Fig 1C, RT-qPCR results demonstrated a strong correlation with RNA-seq fold-change data that endorses the robustness and reliability of the RNA-seq findings.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Comparative RNA-seq of L. major promastigotes incubated in the presence or absence of heme.

(A) ZnMP uptake into L. major promastigotes cultured in the presence or absence of heme. L. major promastigotes cultured in RPMI medium supplemented with 20% (v/v) hdFBS (-heme) and hdFBS supplemented with 10 μM hemin (+heme) were incubated with ZnMP, and the intracellular fluorescence was determined by flow cytometry. Results represent the mean±SD of four independent experiments. Statistical analysis was determined using GraphPad Prism9.2 software using the Student’s t-test. ** p < 0.005. (B) Volcano plot from RNA-seq data of heme-depleted parasites. The log₂FC is plotted on the x-axis, and the negative log₁₀FDR (adjusted p-value) is plotted on the y-axis. Red, up-regulated genes; blue, down-regulated genes; gray, genes not significantly differentially expressed. All DEGs are indicated. (C) Comparative analysis of the relative expression levels of selected genes determined by RNA-seq (gray) and validated by RT-qPCR (black) in the absence of heme. Bars represent the Log₂ values of fold-change (FC) expression of the indicated genes. The data were normalized to Cytosolic GAPDH and LmjF.04.0930 expression. The mRNA levels of both genes showed no significant change in the presence or absence of heme. (D) The most relevant DEGs upregulated in the absence of heme in L. major. The analysis was based on an FDR value ≤ 0.05, log₂FC ≥ 1.0.

https://doi.org/10.1371/journal.ppat.1013993.g001

Heme depletion leads to the overexpression of iron transport- and metabolism-related genes

Fig 1C shows all transcripts with more than a two-fold increase in expression in the absence of heme. Interestingly, this list includes genes related to iron transport and metabolism, such as LmjF.30.1610 (LFR1), which encodes a ferric reductase that reduces Fe⁺³ to Fe⁺² [20], and LmjF.31.3060 and LmjF.31.3070 (LIT1–1 and LIT1–2, respectively), which encode the ferrous ion transporter to the cytosol [26]. We can discard an effect due to the reduction in serum iron levels produced during the preparation of hdFBS as we preliminary confirmed that hdFBS contains the iron required to support the growth of heme-depleted parasites incubated with PPIX, to which the available iron is added by ferrochelatase to form heme (Fig 2A). To exclude this possibility definitively, we also analyzed the expression of some DEGs in heme-depleted parasites incubated with PPIX (Fig 2B-C). If the effect on the expression of these genes were due to the reduced iron concentration in hdFBS, PPIX supplementation will not affect their up- or downregulation. On the contrary, if this were exclusively due to the absence of heme, the effect should be reversed as PPIX is converted to heme, fulfilling heme requirements. Therefore, we incubated heme-depleted parasites for 20 hours in medium with 20% hdFBS containing 10 μM of hemin, 10 μM of PPIX, or without supplementation, and analyzed gene expression by RT-qPCR. Fig 2B (upregulated genes) and Fig 2C (downregulated genes) evidenced that changes in gene expression due to heme depletion were reversed when parasites were incubated with PPIX, similar to hemin supplementation. Two illustrative examples are the genes LmjF.31.3070 (LIT1) and LmjF.30.1610 (LFR1); both increased their expression eight-fold in the absence of heme, an effect that was reversed by the single PPIX addition, with none further iron supplementation. This evidenced that PPIX is transformed into heme by the action of Leishmania ferrochelatase using the available iron in hdFBS, which is therefore not limiting for this assay. Consequently, the DEGs observed from RNA-seq were due to the absence of heme and not to lack of iron.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Iron does not drive differential gene expression under heme depletion conditions used.

(A). PPIX supports the growth of heme-depleted L. major promastigotes. Heme-depleted parasites for 48 h were adjusted to an initial concentration of 10⁶ cells/mL and cultured under the indicated conditions. Parasite density was measured every 24 hours for 7 days using the Coulter Z1 (Beckman Coulter). Results represent the mean±SD of three independent experiments. (B and C) Supplementation with PPIX restores the expression levels of DEGs as hemin supplementation. DEGs over-expressed (B) and under-expressed (C) under conditions of heme depletion (black) and supplementation with 10 μM hemin (white) or 10 μM PPIX (gray). Cytosolic GAPDH was used as a normalizing gene, and relative gene expression was set to 1 for parasites cultured in the presence of heme. Statistical analysis was determined using GraphPad Prism9.2 software using the Student’s t-test. ** p < 0.01, *** p < 0.001, **** p < 0.0001.

https://doi.org/10.1371/journal.ppat.1013993.g002

Identification of LHR2, a Leishmania gene upregulated under heme limitation

Among all the DEGs found, we decided to characterize LmjF.29.0868—named hereafter LHR2 for Leishmania Heme Response-2—in depth. LHR2 is the most upregulated gene in the absence of heme (Fig 1D) and whose function has not been previously characterized in Leishmania.

LHR2 encodes a 168-amino acid (aa) protein that between aa 19–63 presents a heme-binding domain (Pfam PF00173: Cytochrome b5-like Heme/Steroid-binding domain) (Fig 3A). This domain, highly conserved in the cytochrome b5 (Cytb5) family of proteins, is involved in cellular processes such as mitochondrial respiration, polyunsaturated fatty acid metabolism, and sterol biosynthesis, among others [7]. A search in BLAST in the non-redundant database (NR) and UniProtKB/Swiss-Prot using LHR2 as the query sequence identified several Cytb5 and Cytb5-like proteins. Additionally, a domain-structure search in Pfam revealed that LHR2 shares not only structural similarity with Cyb5 but also with Dap1p. Dap1p is a hemoprotein encoded by DAP1 with an important heme-dependent function in the ergosterol biosynthetic pathway and also contributes to DNA damage resistance [27]. Moreover, Dap1p belongs to the MAPR (membrane-associated progesterone receptors) family of proteins [28], a subfamily of cytochrome b5-related proteins [29], among which are human PGRMC1 (membrane-associated progesterone receptor component 1) and PGRMC2 (component 2). These proteins, interestingly, play a role in heme trafficking [30–32], and also share identity to LHR2 (21.6 and 22.8%, respectively) (Fig 3B). LHR2 also shares identity to cytochrome b5 (Cyb5) of S. cerevisiae (cytochrome p450 pathway-associated protein) (22.9%), although lower respect to Dap1p (25.9%) (Fig 3B). ScCyb5 is a heme-protein encoded by CYB5 involved in sterol and lipid biosynthesis pathways, that act as an electron donor from NADPH and contributes to the C5-6 desaturation process of sterols [33,34]. Unlike LHR2 and ScDap1p, ScCyb5, HsPGRMC1 and HsPGRMC2 present a theoretical C-terminal transmembrane segment (S1 Fig).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Analysis of the LHR2 sequence.

(A) Localization of the heme-binding domain Pfam PF00173 obtained from InterProScan (https://www.ebi.ac.uk/interpro) in LHR2 and the indicated related proteins (B) Identity between LHR2 and the indicated related proteins (Clustal Omega software). Sequence alignments were performed with Clustal Omega software. (C) Amino acid alignment of the Cytb5 heme-binding domain of L. major LHR2, human PGRMC1 and PGRMC2, and S. cerevisiae Dap1p and Cyb5 (Clustal Omega software). The heme-binding motif HPGG and histidines (axial heme-binding ligands) are highlighted with a pink background, whereas the conserved Y and K residues in MAPR proteins were with red and green backgrounds, respectively. Identical amino acids (*), amino acids with very similar properties (:), and amino acids with significant divergent properties (.) are indicated. (D) Molecular docking of the 3D structures of LHR2 (aa 15–109), ScCyb5 (aa 5–88), and ScDap1p (aa 72–179). Axial ligands for LHR2 (H36, H60) and ScCyb5 (H37, H61) are shown in pink, whereas for ScDap1p, axial Y84 (red) and heme coordination residues K137 (green) and Y138 (red) are shown. The heme appeared in yellow in all of them. Molecular docking was performed using the AutoDock Vina tool. (E) LHR2 binds heme through the H of the HPGG domain. Left. The H residue within the HPGG heme-binding motif of the LHR2 protein was mutated, replacing histidine at position 50 with leucine using the GeneArt site-directed mutagenesis system (Invitrogen). Middle. SDS-PAGE analysis with Coomassie Blue staining showed purified Wt and mutated H50L LHR2. Right. UV-visible spectra of purified LHR2 Wt and mutated (LHR2 H50L). The arrow indicates characteristic absorption maxima of heme group in LHR2 (around 415 nm and two additional peaks at 529 and 563 nm). The inset shows the difference in the colour of the bacterial pellet expressing WT and mutated LHR2. (F) Comparison of the 3D structural models of LHR2 with Cyb5 and Dap1p. The 3D structure of LHR2 (green) (aa 1–168) was superimposed with the structure of S. cerevisiae Cyb5 (pink) (aa 1–120) and of of S. cerevisiae Dap1p (yellow) (aa 1–152) using ChimeraX-1.8 software.

https://doi.org/10.1371/journal.ppat.1013993.g003

Multiple alignment of the Cytb5 heme-binding domain of LHR2 with these proteins using Clustal Omega software (Fig 3C) showed that LHR2 contains the highly conserved heme-binding HPGG (His-Pro-Gly-Gly) motif found in Cytb5 proteins, as well as the histidines considered the typical axial ligands of heme coordination in Cytb5 proteins (Fig 3C highlighted by a pink background) [30]. These residues are not conserved in the MAPR family. The human PGRMC1 protein uses the aa Y and K to coordinate heme [30], so Dap1p of S. cerevisiae likely binds heme using the same aa residues (Fig 3C highlighted by a red and green background, respectively). Therefore, LHR2 probably binds to heme via H50 and H74 (H36 and H60 in the 3D structure modelled with I-TASSER aa 15–109) similar to the ScCyb5 that binds heme via H37 and H61 (3D structure aa 5–88), as shown in Fig 3D, whereas ScDap1p does so using the axial Y84 (red) and the heme coordinating residues Y137 (green) and Y138 (red) (3D structure aa 41–152), similar to that described for the PGRMC1 protein of H. sapiens [30]. Indeed, a mutation in H50 of LHR2 prevents heme binding, as shown in Fig 3E. Thus, when heterologously expressed in E. coli, the bacterial pellet expressing WT LHR2 displayed a characteristic brownish color with a faint pink hue, whereas the H50L pellet was whitish, suggesting loss of heme binding. UV–Vis spectra of purified proteins confirm this: WT LHR2 shows absorption at 415 nm and peaks at 529 and 563 nm, absent in the mutant. On the other hand, the structural analysis of these 3D models was compared by their respective RMSD values (Fig 3F), that measures the similarity among three-dimensional structures as it represents the average distance between the corresponding atoms of two superimposed proteins with respect to the number of atoms analyzed: the lower the RMSD, the more similar the two structures [35]. Under this criterion LHR2 (green) is structurally closer to ScCyb5 (pink) using root mean square deviation (RMSD) analysis (the 55-residue RMSD of the two structures is 0.66 Å) than to Dap1p (yellow) (the 22-residue RMSD of the two structures is 0.60 Å), despite the higher sequence similarity to ScDap1p.

LHR2 is essential in the promastigote stage of the parasite

To approach the relevance of LHR2, we used the CRISPR-Cas9 PCR tool-kit [36] to generate LHR2-deficient parasites as described in Fig 4A. Thus, L. major parasites expressing Cas9 and T7RNA polymerase [37] were transfected with a plasmid harboring the sgRNA sequence targeting LHR2 and the donor cassette made of homology regions flanking the neomycin resistance marker (NEO^R) (Fig 4A). Then, transfected parasites were selected on solid medium containing geneticin, and a total of 24 clones were isolated and analyzed by PCR (a representative clone is shown in Fig 4B). All 24 clones were single-LHR2 knockouts (LHR2^+/-) suggesting the essentiality of the gene in L. major promastigotes. Thus, using a forward primer from the 5′ UTR region of the LHR2 gene (Fw5´UTR) and a reverse from the LHR2 ORF region, the expected band of 981 pb was obtained in LHR2^+/− (lanes 1 and 2 in Fig 4B left) and control parasites (lane 4 in Fig 4B left). The right integration of the NEO^R cassette into the LHR2 locus and the replacement of only one of the alleles of the LHR2 gene was endorsed by the amplification of the expected 1231 pb product using primer pair Fw5´UTR-RvNEO in LHR2^+/- parasites (lanes 1 and 2 in Fig 4B right) but absent in control non-transfected parasites (lane 4 in Fig 4B right). Finally, the decrease of LHR2 expression levels in LHR2^+/- respect to control parasites was demonstrated by qPCR (Fig 4C).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. LHR2 is essential in the promastigote stage of the parasite.

(A) Schematic representation of the CRISPR-Cas9 strategy used for LHR2 gene replacement. The two sgRNAs targeting the upstream (5’) and downstream (3’) cleavage sites of the LHR2 coding sequence are indicated by yellow arrows, which guide the DNA cleavage by Cas9 that will be repaired by homologous recombination using the donor DNA containing the resistance gene as a template. The oligonucleotide PCR primers used to confirm LHR2 allele deletion and integration of the geneticin resistance gene (NEO^R) are indicated. (B) Agarose gel analysis of PCR-amplified products with the oligonucleotides Fw5’UTR-RvGene and Fw5’UTR-RvNEO, specific for the LHR2 locus and NEO^R, left and right, respectively. Lane MW shows the molecular mass markers; lanes 1, 2, 3, and 4 correspond to PCR products from genomic DNA from mutants LHR2^+/-, LHR2^+/- AB (add-back), double-LHR2 knockout complemented with episomal LHR2 and wild type, respectively. The 981 bp PCR product of the LHR2 locus (left) and 1231 bp of NEO^R in all clones indicate the deletion of only one allele of LHR2 (right). The essentiality of LHR2 was assessed by the deletion of both alleles only in the presence of episomal LHR2 in all 24 clones tested (a representative clone is shown in lane 3). (C) LHR2^+/- parasites showed decreased LHR2 mRNA levels compared with the control LHR2^+/+ mRNA was obtained from each line, and LHR2 expression levels were analyzed by RT-qPCR. Data represent the mean ± SD of three independent experiments performed in triplicate. Statistically significant differences were determined by GraphPad Prism9.2 software using the Student’s t-test. * p < 0.05. (D) LHR2^+/- promastigotes have a growth defect under low-heme conditions. LHR2^+/+ (Control), LHR2^+/- and LHR2^+/- AB parasites were cultured in medium supplemented with hiFBS (left) or were heme-depleted for 48 h and cultured in medium supplemented with hdFBS and 5 μM hemin (right). Parasite density was determined every 24 hours for 6 days using a Coulter Z1 (Beckman Coulter). Data represent the mean ± SEM from three independent biological replicates. Statistically significant differences were assessed by GraphPad Prism9.2 software using the Student’s t-test. * p < 0.05, ** p < 0.005, *** p < 0.0005.

https://doi.org/10.1371/journal.ppat.1013993.g004

The essential role of LHR2 was confirmed by repetition of the CRISPR-Cas-9 methodology in the presence of episomal LHR2. The elimination of both LHR2 alleles was demonstrated by PCR analysis of genomic DNA from 24 selected clones using the forward primer Fw5´UTR (5′ UTR region of the LHR2 gene) and a reverse primer from the gene (RvGene), (lane 3 in Fig 4B left). Together, these results suggest that LHR2 is essential for the promastigote stage of the parasite, at least in in vitro cultures.

Finally, we tested whether LHR2^+/- promastigotes grow at the same rate as wild type in normal medium and whether they display a phenotype in low-heme conditions. As shown in Fig 4D, this is indeed the case, supporting the notion that LHR2 is involved in the heme response.

Heterozygous deletion of LHR2 prevents L. major virulence

In the next step, we studied the importance of LHR2 for the clinically relevant amastigote stage, analyzing the ability of LHR2^+/- parasites in vitro to infect host macrophages and replicate as intracellular amastigotes.

BMDM macrophages were infected with stationary-phase promastigotes from LHR2^+/+ (control), LHR2^+/−, and LHR2^+/− AB parasites with a multiplicity of infection (MOI) of 1:10 (macrophages: parasites). After staining with CellMask DeepRed and DAPI, the infection index (parasites/100 macrophages) was quantified at 24- and 96-hours post-infection (hpi) from fluorescence microscopy images using the automatic tool FiCRoN. Fig 5A shows a representative picture of the infection of BMDM macrophages with control and LHR2^+/− stationary-phase L. major promastigotes at 96 hpi. According to Fig 5B, at 24 hpi, infection index among the different parasites did not differ significantly, but at at 96 hpi, control parasites doubled the infection rate, while most of the LHR2^+/- parasites were dead, with an infection rate of only 4% (p < 0.0005) compared with the control line (Fig 5A and 5B). This decrease was completely reversed in LHR2^+/-parasites complemented with episomal LHR2 (LHR2^+/-AB) (p < 0.0001), with an infection rate similar to that of the control line (Fig 5B). Next, the role of the previously discussed, putative heme-binding HPGG motif in LHR2 functionality was evaluated profiting from the phenotype LHR2^+/-AB parasites. Thus, LHR2^+/− parasites were complemented with an episomal mutated LHR2, in which H50 of the HPGG motif was changed to leucine (LHR2^+/−AB H50L) by site-directed mutagenesis, to infect macrophages. The results showed that LHR2^+/−AB H50L failed to reverse the phenotype as the parasites harboring the mutated gene did not multiply as intracellular amastigotes (Fig 5B), underlying the importance of heme binding in LHR2 functionality.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. LHR2 is required for the intracellular replication of amastigotes and development of cutaneous leishmaniasis.

(A) Representative picture of the infection of MBMD macrophages with control and LHR2^+/− stationary-phase L. major promastigotes at 96 hpi. Infected cells were fixed and stained with CellMask DeepRed, and DAPI for visualization by confocal fluorescence microscopy, using a Leica SP8 spectral microscope. The cytoplasm (mc) and nuclei (mn) of the macrophages, and the kinetoplast of intracellular amastigotes (k) were indicated. DAPI is shown in grayscale to increase image contrast and improve parasite visualization. Magnification bar = 5 μm. (B) Inability of LHR2^+/− to replicate as intracellular amastigotes. BMDM macrophages were infected with LHR2^+/+, LHR2^+/−, LHR2^+/− AB, and LHR2^+/− AB H50L stationary-phase promastigotes. At 24 hpi, non-phagocytized parasites were washed, and at 24 and 96 hpi, infected cells were fixed and stained with CellMask DeepRed, and DAPI. The infection index (parasites/100 macrophages) was automatically calculated using the FiCRoN tool (n = 300 macrophages/group). The results were expressed as the mean±SEM of two independent experiments performed in triplicate. (C) BALB/c mice were infected with 1 × 10⁴ metacyclic-phase L. major promastigotes of the LHR2^+/+, LHR2^+/−, LHR2^+/− AB, and LHR2^+/− AB H50L lines in the right hind footpad. Inflammation progression (difference between inoculated and contralateral uninfected footpad) was recorded weekly. The values represent the mean±SEM of six mice. Statistically significant differences were determined by GraphPad Prism9.2 software using the Student’s t-test. **** p < 0.0001, *** p < 0.0005, ** p < 0.005. (D) Images show representative pictures of footpad inflammation at eight weeks post-infection.

https://doi.org/10.1371/journal.ppat.1013993.g005

The close to nil intracellular replication of LHR2^+/− parasites in vitro, prompted us to analyze their pathogenicity in an in vivo murine model of cutaneous leishmaniasis. BALB/c mice were infected with 1x10⁴ LHR2^+/+ (control), LHR2^+/−, LHR2^+/−AB, and LHR2^+/−AB H50L metacyclic promastigotes by right hind footpad inoculation, and progression of inflammation was monitored weekly for eight weeks (Fig 5C). Mice infected with control parasites started to develop a progressive inflammatory reaction. In contrast, mice infected with LHR2^+/− parasites did not show any sign of disease during these eight weeks. As observed with in vitro infections, LHR2^+/−AB parasites rescued the virulence phenotype, encompassing inflammation and progression of the disease similar to the control line. In contrast, mice infected with LHR2^+/−AB H50L parasites did not develop cutaneous leishmaniasis.

LHR2 is located in the parasite mitochondrion

We next mapped the intracellular localization of the LHR2 protein using the same CRISPR/Cas9 PCR tool-kit [36] to endogenously label LHR2 with the fluorescent marker mNG, as described in Fig 6A. Fig 6B shows that the LHR2-mNG protein was located at the parasite mitochondrion as deduced from its co-localization with the mitochondrial marker MitoTracker Red, with a Pearson’s co-localization index of 0.92 ± 0.008.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 6. LHR2 is located in the mitochondrion of the parasite.

(A) Schematic representation of the CRISPR/Cas9 strategy used for C-terminal in situ tagging of the LHR2 protein. Promastigotes of the L. major Cas9/T7 line were cotransfected with template DNA for in vivo gRNA transcription, which hybridizes near the stop codon of LHR2, and donor DNA containing the fluorescent protein mNeonGreen (mNG) and the puromycin resistance gene (PURO^R) flanked by 30nt homology regions identical to the final coding sequence and 3´UTR of the LHR2 locus. Transfected parasites were subsequently selected with 30µg/mL puromycin for seven days and observed by fluorescence microscopy. (B) Mitochondrial localization of LHR2-mNG. Log-phase promastigotes expressing mNG-tagged LHR2 protein were incubated with 50 nM of the mitochondrial marker MitoTracker Red (Mit-Red) at 28°C for 30 min and subsequently observed by fluorescence microscopy to determine colocalization of both tags (Merge). The figure shows a representative parasite of a total parasite population with a similar fluorescence pattern. The differential interference contrast (DIC) image is shown on the right side. Magnfication bar = 5 μm.

https://doi.org/10.1371/journal.ppat.1013993.g006

Single-LHR2 knockout parasites have increased mitochondrial ROS levels at the promastigote stage

The fact that LHR2 locates to the parasite mitochondrion, mutations in yeast cytochrome b proteins have been associated with impaired respiratory function [38], and PGRMC2 knockout cells show severely reduced respiratory capacity [31], prompted us to use respiration and mitochondrial membrane potential (ΔΨm) as possible parameters of LHR2 functionality. Therefore, we studied the impact of deleting a single LHR2 allele on basal mitochondrial respiration and ΔΨm in L. major by polarographic quantitation of the OCR (Oxygen Consumption Rate) and Rh123 accumulation, respectively, for Wt and LHR2^+/- L. major promastigotes. Our results showed no significant differences in the OCR (Fig 7A) or Rh123 accumulation (Fig 7B) between both lines, suggesting that that deletion of one allele of the LHR2 gene does not affect the respiration of L. major promastigotes.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 7. Effect of heterozygous deletion of LHR2 on mitochondrial oxygen consumption,ΔΨm and ROS levels in promastigotes.

(A) Basal respiration, determined as the OCR, was measured in real time in Control (LHR2^+/+) and LHR2^+/- L. major promastigotes using the Oxygraph + /Oxytherm system (Oxygraph System. V.2.20, Hansatech). Data represent the mean ± SD of two independent experiments. (B) ΔΨm was measured by assessing the accumulation of Rh123 (0.5 µM for 10 min) in Control (LHR2^+/+) and LHR2^+/- L. major promastigotes by flow cytometry. Measurements are expressed as mean intracellular fluorescence (RFU) ± SD from three independent experiments. (C) Mitochondrial ROS was measured by assessing the accumulation of MitoSOX (3 µM, 30 min) in Control (LHR2^+/+) and LHR2^+/- L. major promastigotes by flow cytometry. Measurements are expressed mean intracellular fluorescence (RFU) ± SD from three independent experiments. Significant differences versus the control were determined by Student’s t test (*** p < 0.0005).

https://doi.org/10.1371/journal.ppat.1013993.g007

On the other hand, since mitochondria represent the main source of oxidants in eukaryotic cells, we employed MitoSOX Red dye (a mitochondrial superoxide indicator) to assess mitochondrial superoxide accumulation, which reflects intracellular ROS levels. As shown in Fig 7C, LHR2^+/- promastigotes exhibited a ~ 70% increase in mitochondrial ROS compared to the Wt line, suggesting a regulatory or protective role for LHR2 in mitochondrial redox homeostasis.

LHR2 rescue the phenotype of S. cerevisiae Dap1p

As previously described, LHR2 showed a slightly higher identity to Dap1p (25.9%) than to Cyb5 (22.9%) in S. cerevisiae. However, in both 3D structure and form of heme-binding, LHR2 is closer to Cyb5 than to Dap1p. To test whether LHR2 might have a similar function in either of these proteins, functional complementation assays were conducted in dap1Δ and cyb5Δ yeast. First, as LHR2 is located in Leishmania mitochondria, whereas in yeast Cyb5 is localized to the ER membrane [33,39] and Dap1p has a perinuclear location (probably corresponding to ER), as well as in lipid particles [40,41], the location of LRH2 expressed in yeast was analyzed. As shown in Fig 8A, LHR2-eGFP localizes to the perinuclear ER of yeast as deduced from its co-localization with the marker ER-Red (Merge), while it does not colocalize with mitochondrial markers (S2 Fig).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 8. LHR2 and ScDap1p reciprocally complement each other’s function in yeast and Leishmania.

(A) LHR2-eGFP was expressed and localized in wild-type (WT) S. cerevisiae yeast. Yeast with the LHR2-eGFP construct were incubated with 1 μM of the ER-Tracker Red marker (ER-RED) at RT for 30 min and then 0.1μg/mL of DAPI at RT for 10 min. Fluorescence microscopy images reveal blue nuclei (DAPI) and the overlapping between eGFP and marker ER-RED (Merge), suggesting that LHR2 localizes to the perinuclear ER of yeast. The figure shows representative cells from a total population with a similar fluorescence pattern. Magnification bar = 5 μm, at the right, DIC image. (B-C) LHR2 allowed the normal growth of dap1Δ yeast in medium with 50 μM fluconazole (B) and 0.02% MMS (C). dap1Δ yeast were transformed with the pADH-LEU vector containing or not LHR2. Strains were grown overnight at 30°C, adjusted to an OD₆₀₀ of 4.0, five-fold serial diluted, spotted on YPD-glucose medium plates containing or not 50 μM fluconazole (B) or 0.02% MMS (C), and incubated at 30°C for 3 days. For the streaked plates shown on the right, yeasts at OD₆₀₀ = 0.032 were used. (D) Localization of S. cerevisiae Dap1p expressed in L. major. Parasites with the Dap1-GFP construct were incubated with 50 nM of the mitochondrial marker MitoTracker Deep Red (Mit-Red) at 28°C for 30 min to determine the co-localization of both markers (Merge) by fluorescence microscopy. The figure shows a representative parasite from a total population with a similar fluorescence pattern. Magnification bar = 5 μm. The differential interference contrast (DIC) image is seen on the right. (E) BMDM macrophages were infected with stationary phase promastigotes of the different lines, and the resulting intracellular amastigotes were visualized by fluorescence microscopy at 24 and 96 hpi, as described in Fig 5. The infection index was automatically calculated using the FiCRoN tool, counting a minimum of 300 macrophages per group. The results represent the mean±SEM of two independent experiments performed in triplicate. * p < 0.05 ** p < 0.005.

https://doi.org/10.1371/journal.ppat.1013993.g008

Yeast Cyb5 is a cytosol-targeted ER membrane-bound hemoprotein that acts as an electron transporter for several membrane oxygenases [33,34] and participates in sterol and sphingolipid biosynthesis [42]. To evaluate the effect of LHR2 on the growth of cyb5Δ yeast, we performed LHR2 expression in S. cerevisiae cyb5Δ cells. H₂O₂ resistance associated to cytochrome b5 (CybE) deficiency was described for the pathogenic fungus Aspergillus fumigatus [43]; thus, we explored the ability of H₂O₂ to inhibit the growth of control yeast while allowing cyb5Δ proliferation, an effect obtained at 4.0 mM (S3A Fig). S3A Fig shows that cyb5Δ yeast expressing LHR2 grew in 4.0 mM H₂O₂-containing plates, similar to the growth of cyb5Δ cells transfected with the empty plasmid. In contrast, wild-type yeast cells were unable to multiply; thus, LHR2 failed to complement ScCyb5p function.

We then evaluated the possible functional relationship of LHR2 with S. cerevisiae Dap1p. ScDap1p is a hemoprotein important for cell survival and involved in several functions, such as cell cycle progression, regulation of mitochondrial stability, and ergosterol synthesis. Deletion of the DAP1 gene causes a drastic growth defect in yeast cells in the presence of fluconazole, a sterol synthesis inhibitor, and the methylating agent MMS that affects cellular DNA. Thus, we transfected dap1Δ yeast with the yeast expression plasmid pADH-LEU containing or not LHR2. dap1Δ yeast cells transfected with the empty plasmid did not grow in 50 μM fluconazole (Fig 8B) or 0.02% (w/v) MMS (Fig 8C), as previously described [28,44]. Contrarily, dap1Δ cells expressing LHR2 grew normally, similar to wild-type in the presence of both drugs (Fig 8B and 8C), hence, LHR2 may functionally complement ScDap1p. Taken together, these results suggest that LHR2 is functionally related to the S. cerevisiae Dap1 protein but not to Cyb5.

S. cerevisiae Dap1p rescues the function of LHR2 in Leishmania

The next step was to evaluate whether ScDap1p complements the role of LHR2 in Leishmania. First, we expressed a GFP-tagged version of ScDap1 in L. major parasites. As shown in Fig 8D, ScDAP1-GFP has a mitochondrial localization in L. major according to its colocalization with the mitochondrial marker MitoTracker Red, (Pearson’s colocalization index = 0.95 ± 0.008), similar to the endogenous localization of LHR2 (Fig 6). Furthermore, we were able to generate double-LHR2 knockout promastigotes using CRISPR-Cas9 after complementation of the parasites with episomal ScDap1p (S4A-B Fig). These results support that ScDap1p complements LHR2 function in L. major promastigotes.

We then analyzed this functional complementation in intracellular amastigotes. BMDM macrophages were infected with: LHR2^+/+ (Control), LHR2^+/-, LHR2^-/- + LHR2, and LHR2^-/- + ScDAP1, and intracellular amastigotes were determined at 24 and 96 hpi. As expected, no statistically significant differences between the lines were observed at 24 hpi (Fig 8E). At 96 hpi, LHR2^+/- parasites were eliminated, while control (LHR^+/+), LHR2^-/- + LHR2, and LHR2^-/- + ScDAP1 parasites replicated as amastigotes (Fig 8E). This ability of ScDap1 to fully rescue the deleterious phenotype caused by the absence of LHR2 in intracellular amastigotes underline that ScDap1p performs the function of LHR2 in Leishmania.

Additionally, we evaluated whether ScCyb5 could complement the role of LHR2 in Leishmania. Similar to ScDap1p, the ScCyb5-GFP fusion protein exhibited mitochondrial localization in L. major, as demonstrated by its colocalization with the mitochondrial marker MitoTracker Red (Pearson’s colocalization coefficient = 0.97 ± 0.003) (S3B Fig). Furthermore, as shown in S3C Fig, in the presence of ScCYB5 we successfully disrupted both alleles of the LHR2 gene locus, but the LHR2 coding sequence remained detectable by amplification with LHR2-internal primers, suggesting the gene was relocated to another genomic site. Nevertheless, ScCYB5 did not restore intracellular amastigote growth (S3D Fig), indicating that ScCYB5 does not rescue the function of LHR2 in Leishmania.

Heterozygous deletion of LHR2 does not alters ergosterol levels in promastigotes

Deletion of DAP1 in S. cerevisiae results in reduced ergosterol levels, consistent with the role of Dap1p in supporting cytochrome P450–dependent steps of the sterol biosynthetic pathway [28,44] To investigate the role of LHR2 at this level, we first assessed the sensitivity of L. major promastigotes to amphotericin B (AMB) as a function of LHR2 expression. Sensitivity to AMB correlates with cellular ergosterol levels, as reduced ergosterol content diminishes the drug’s ability to bind and form membrane-disrupting complexes [45,46]. As shown in S2 Table, no significant differences were observed among the different lines, suggesting that LHR2 does not affect parasite ergosterol levels. This finding was confirmed using filipin, a robust in situ marker of plasma-membrane ergosterol with demonstrated efficacy and specificity [47,48]. As shown in S5 Fig filipin successfully labels the Leishmania plasma membrane (left panel), but we detected no differences between LHR2^+/+ and LHR2^+/- parasites (right panel). Altogether, these data indicate that deletion of a single LHR2 allele does not impact plasma-membrane ergosterol content in L. major promastigotes.

In silico analysis of the therapeutic potential of LHR2

The fact that the deletion of a single LHR2 allele prevents amastigote replication within macrophages, as well as disease development in a murine model of cutaneous leishmaniasis, suggests that LHR2 could be an interesting therapeutic target, as it would not require total LHR2 inhibition to achieve a therapeutic effect. For this, it will be essential a fine specificity for the inhibitor, in order to not affect the function of its human related proteins. We therefore analyzed the structural differences of LHR2 to similar human proteins. As mentioned above, the domain Cyt-b5 is highly conserved in the cytochrome b5-like family, which includes a diverse range of proteins, such as steroid-binding proteins, progesterone receptors, some chitin synthases, etc. [49–51]. Using the MEGA11 program [52], we first performed a phylogenetic analysis of sequences of LHR2 and human proteins containing the Cyt-b5 domain with a length of ≤ 200 residues, including the MAPR family proteins, PGRMC1 (PGRC1) and PGRMC2 (PGRC2) (O00264 and O15173, respectively, UniProtKB/Swiss-Prot database). Fig 9A shows that LHR2 is evolutionarily closer to the 150–amino acid human CYB5B protein (O43169, UniProtKB/Swiss-Prot), with which it shares 27% identity.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 9. The structure of LHR2 reveals a parasite-selective pocket as a feasible design for inhibitors.

(A) Phylogenetic tree of human proteins containing a Cyt-b5 domain with a length of ≤ 200 residues, ScDap1p and LHR2, was performed using the MEGA11 program. The HsO43169 human protein (HsCYB5B) (UniProtKB/Swiss-Prot database) (rectangle) was identified as having the highest identity (27%) with respect to LHR2. (B-above) 3D models of LHR2 (aa 1-168) (light gray) and HsCYB5B (PDB 3NER, aa 1-150) (purple). Images were generated with the ChimeraX-1.8 program. The dashed circle indicates the structural prediction between aa 99-168 of LHR2. The prediction of a pocket of LHR2 for the design of potential inhibitor molecules is shown in the yellow region. Heme-binding histidines (H) are also shown in the two protein structures. (B-below) Structural overlay of the LHR2 (light gray) and HsCYB5B (purple) models. (C) Molecular surface representation of LHR2, HsCYB5B, and HsPGRMC1, colored according to the electrostatic potential (Coulombic), ranging from red for negative potential, white for neutral, and blue for positive potential.

https://doi.org/10.1371/journal.ppat.1013993.g009

Next, we compared the 3D structure of LHR2 and HsCYB5B. As LHR2 contains a region between aa 99–168 that scapes 3D structural prediction from homology modeling and alphafold, a prediction model was constructed with the I-TASSER server [53], that predicted several helices in this region (Fig 9B, dashed line). We used ChimeraX-1.8 software [54] to compare this 3D LHR2 model with the crystal structure of CYB5B from H. sapiens (Fig 9B). We identified a common sequence stretch between the two proteins that contains the Cyt-b5 domain and, interestingly, a 69-aa region of LHR2 near the C-terminus (aa 99–168), specific for the parasite protein (Fig 9B structural overlay). In fact, a general BLAST using this LHR2 region yielded no candidates, emphasizing its relevance for the design of specific inhibitors. The modelling of LHR2 displayed a unique cavity or pocket only in the parasite protein (Fig 9B, yellow region, and S6 Fig). This pocket could potentially be explored to design possible inhibitors specifically active against LHR2. Additionally, to evaluate the specificity of the possible inhibitor-binding pocket of LHR2 with respect to human MAPR family proteins, we superimposed the 3D LHR2 model with the crystals of human PGRMC1 (homologous to Dap1p). We also performed representations of the modeling of the two proteins from different views using the ChimeraX-1.8 program (S7 Fig). The analyses also show that the predicted LHR2 cavity for possible inhibitor binding does not overlap with human PGRMC1 protein crystals.

Likewise, strong differences in the distribution of the Coulombic electrostatic potential were observed between the surfaces of the parasite and human proteins. Fig 9C (ChimeraX, v1.7) shows the prevalence of positively charged residues (blue), including the possible inhibitor binding cavity for the LHR2 protein, in contrast with the predominance of anionic residues (red) in human CYB5B and PGRMC1.

Altogether, this analysis underlies significant structural differences between Leishmania LHR2 and related human proteins as a new avenue for the design of specific inhibitors against the parasite.

Ethics statement

In vivo infection studies in mice were performed according to the National/European Union Guidelines for the Care and Use of Laboratory Animals in Research and with the approval of the Ethics Committee of the Spanish National Research Council.

Parasite culture conditions

Parasites of L. major (MHOM/IL/ 80/Friedlin) were kept in a virulent state by passage in BALB/c mice, and promastigote forms were cultured in RPMI-1640 medium (Invitrogen) supplemented with 10% heat-inactivated fetal bovine serum (hiFBS, Invitrogen) and 5 μM biopterin at 28°C. For the transcriptomic assay, promastigote forms were cultured as described above at an initial concentration of 1x10⁶ parasites/mL for 48 hours. Mid-log phase promastigotes were then harvested by centrifugation (1000 × g for 10 min at 4°C), washed twice with medium without hiFBS, and resuspended at 5x10⁶ parasites/mL in RPMI-1640 medium supplemented with 5 μM biopterin and 20% heme-depleted FBS (hdFBS), prepared as previously described [17,37,55]. Parasites were incubated at 28ºC for 24 hours in the presence or absence of 10 μM hemin or PPIX when indicated) [37]. Under these conditions, parasites with and without supplementation grow similarly. Following this incubation period, cultures were split into two aliquots for i) RNA extraction to carry out subsequent RNA-seq analysis and ii) Zn (II) Mesoporphyrin IX (ZnMP) (Frontier Scientific) uptake experiments. Four independent samples of each condition—promastigotes grown in the presence or absence of heme—were obtained.

Hemin was obtained from Sigma. A hemin stock solution was prepared by dissolving 10 mg of hemin in 0.1 N KOH, followed by dilution in 900 µL of 50 mM potassium phosphate (KPi) buffer at pH 7.2 containing 100 mM NaCl. The solution was centrifuged at 10,000 × g for 3 minutes. The supernatant was collected and further diluted 1.5-fold in KPi buffer pH 7.2 with 100 mM NaCl. The concentration was determined by diluting the sample in 5 mM NaOH and measuring absorbance at 385 nm, using an extinction coefficient (ε) of 58.4 mM ⁻ ¹·cm ⁻ ¹. ZnMP and PPIX-Na were purchased from Frontier Scientific and used freshly prepared according to [64]. Around 2 mg of porphyrin were dissolved in 500 µL of 10% ethanolamine. Then, 100 mg of BSA were added, and the volume was adjusted to 10 mL with water, resulting in a solution containing 1% ethanolamine and 1% BSA. The pH was adjusted to 7.4 using 1 N HCl. The concentration was determined by diluting a sample in 2.7 N HCl and measuring absorbance at of 402 (ZnMP) and 409 (PPIX) nm using an extinction coefficient (ε) of 404.9 (ZnMP) and 218.12 (PPIX) mM ⁻ ¹·cm ⁻ ¹.

Uptake of ZnMP in Leishmania promastigotes

Porphyrin uptake in Leishmania promastigotes was performed as previously described [17]. Briefly, washed L. major promastigotes were suspended in HPMI medium (20 mM HEPES pH 7.2, 132 mM NaCl, 3.5 mM KCl, 0.5 mM MgCl₂, 1 mM CaCl₂, 5 mM D- glucose) and incubated with 10 µM ZnMP at 28°C for 10 min. Parasites were then washed three times with ice-cold PBS containing 5% BSA. After washing with cold PBS and labeling the dead cells with Sytox Green (20nM) (Invitrogen), parasites were fixed with 2% paraformaldehyde in PBS. Intracellular protoporphyrin fluorescence was measured by flow cytometry (excitation at 405 nm and emission between 575 and 585 nm) employing a FACSAria III cytometer (Becton Dickinson, USA).

RNA isolation and RNA-Seq library preparation

Total RNA extraction from 5x10⁷ promastigotes was performed using the RNA Isolation Kit (Roche Biochemicals), which was subsequently treated with DNase (Promega) to eliminate contaminating DNA, according to the manufacturer’s instructions.

RNA-Seq was performed in the genomics facility at the Pfizer-University of Granada-Junta de Andalucía Centre for Genomics and Oncological Research Centre (GENYO). Standard libraries for massive sequencing were generated using the TruSeq RNA Sample Prep Kit (Illumina). Briefly, poly-A⁺RNA was selected by oligo-dT beads, and RNA fragmentation was achieved using divalent cations under high temperature. Subsequently, these fragments were used to build a cDNA library. Library fragments were run on an agarose gel; a 300–400 bp region was excised, and cDNA purified. Two cDNA libraries were constructed: first strand synthesis of one of them was initiated with only random hexadeoxynucleotide primers (Illumina standard protocol); however, for the first strand synthesis of the second library, as an additional component, we introduced the 5’-T15VN-3’ oligonucleotide, together with the random hexamer primers present in the kit. Later, the second strand of the cDNA was produced. The cDNA ends were repaired and adenylated; afterward, adapters were added at both ends. Finally, the library was enriched in ligated fragments by limited PCR amplification. Sequencing was carried out in a Nextseq 500 Illumina system, and paired reads of 75 nucleotides were obtained for each of the four replicates of promastigotes grown with or without heme.

RNA-Seq pre-processing, mapping, and quantification

Raw reads were analyzed using the miARma-Seq pipeline v1.7 [65], which includes quality filtering, read alignment, and differential expression analyses. Sequences were quality-filtered using the standard Illumina process with the FASTQC tool [66]. Reads were mapped to version 31 of the L. major genome, obtained from the TriTrypDB database (www.tritrypdb.org) using BWA [67], Hisat [68], and STAR [69] as included in the latest miARma-Seq version. The abundance of reads mapping to each gene feature in the TriTrypDB L. major annotation (v 31.0) was determined using featureCounts as included in the latest version of miARma-Seq [65].

RNA-Seq expression analysis

Non-expressed and weakly expressed genes, defined as having < 1 read per million in at least four samples, were removed before differential expression analysis. edgeR [70] was applied to identify differentially expressed (DE) genes, thus reads were normalized using the TMM method (trimmed mean of M-values) and exact tests to calculate differences between two groups of negative-binomial counts were performed to gather the log₂ fold change (log₂FC) and the Benjamini and Hochberg false discovery rate (FDR) for each gene. Common genes with an FDR ≤ 0.05 in BWA, Hisat, and STAR analyses were considered significantly and differentially expressed.

qRT-PCR validation of gene expression changes

RNA samples used for the experimental validation of gene expression changes using quantitative reverse transcriptase PCR (qRT-PCR) were the same as those used for the RNA-Seq assays. cDNA was synthesized from 1 µg of total RNA using the qScript cDNA Synthesis Kit (Quanta Biosciences, Inc.), according to the manufacturer’s instructions. Lack of genomic DNA contamination was confirmed by PCR amplification of RNA samples without prior cDNA synthesis. Quantitative PCR was performed in a CFX96 cycler (BioRad); each 10 µL reaction contained 5 µL of PerfeCta SYBR Green SuperMix (Quanta Biosciences), 400 nM of the specific primer pair (S3 Table), and 2 µL of a 1:4 dilution of the synthetized cDNA. The primer pairs used to amplify cDNA were designed with Primer3 software [71], and standard curves for each primer pair were generated with 10-fold serial dilutions of L. major genomic DNA to determine primer efficiency, which showed amplification efficiency values ranging from 91% to 108% and no primer dimers were detected. All reactions were performed in triplicate, a negative control with no template was included, and each product was verified by melting curve analysis. The relative gene expression was calculated using the critical threshold (ΔΔCt) method [72] and using Cytosolic GAPDH (LmjF.36.2350) and LmjF.04.0930 gene expression as internal controls to normalize the data. The RNA sequencing data revealed no significant change in the mRNA levels of both genes between the normal and heme-depleted growth conditions, confirming they are stable and reliable as internal controls.

CRISPR-Cas9 mediated deletion of LHR2

LHR2 (LmjF.29.0868) was deleted from the L. major genome using the CRISPR-Cas9 tool kit as described previously [36] and kindly provided by Dr. Eva Gluenz (University of Oxford, UK). Log-phase promastigotes of L. major Cas9/T7 [37] with or without episomal LHR2 expression (genetic complementation, see below) were co-transfected with a repair DNA template (pTNeo_v1 LeishGEdit) encoding a neomycin resistance marker (NEO R) flanked by 30-nt homology arms corresponding to 5′UTR and 3′ UTR of LHR2 gene, together with two single-guide RNA (sgRNA) templates targeting LHR2, using program V-033 of the Amaxa Nucleofector System (Lonza). Deleted LHR2 was replaced by donor DNAs containing the neomycin resistance cassette flanked by 30-nt homology arms from the 5′UTR and 3′UTR specific sequence, which were amplified from the pTNeo plasmid, respectively. sgRNA templates were amplified using the common reverse primer G00 (sgRNA scaffold) and primers forward 5´GAAATTAATACGACTCACTATAGGCAATGTGAGAGAATTGCGTAGTTTTAGAGCTAGAAATAGC3´ for 5′ sgRNA and 5´GAAATTAATACGACTCACTATAGGGAGAGTGACGCGTTACGCCTGTTTTAGAGCTAGAAATAGC3´for 3′ sgRNA (GJ3Tag). Primers were designed using an online resource (http://www.leishgedit.net). Following transfection and clonal selection onto semi-solid culture under antibiotic pressure (40 µg/mL neomycin) (Sigma-Aldrich), 24 clones were selected and propagated in a liquid culture medium. LHR2 deletion and resistance cassette integration were confirmed by PCR analysis using the forward common primer 5´CGTATTCGCTCTACTACCGTC3´ and reverse specific primers 5´GAACAGTCTCAGAGCAACCTCAG3´ and 5´GCTGCCTCGTCCTGCAGTTC3´, amplifying fragments of the LHR2 (981 bp) and neomycin-resistance gene (1231 bp), respectively. Furthermore, LHR2 expression levels were measured by qRT-PCR in mutant lines (knockout and complemented add-back (AB) parasites), using L. major Cas9/T7 as a control and Cytosolic GAPDH as a constitutive normalizer. Total RNA from log-phase promastigotes was isolated with the RNeasy kit (Qiagen, USA), according to the manufacturer’s instructions. cDNA synthesis and quantitative PCR were performed as previously described.

For the LHR2 genetic complementation of L. major LHR2^+/- mutants, the gene was PCR amplified using forward 5´GGGGGATCCATGCTTAACGATTTGCTATTTCTTAG3´ (FwGJBa) and reverse 5´CCCGGATCCTCAGCGTACCCGCTGTG3´ (RvGJBa) primers (restriction sites are underlined), cloned into BamHI sites of the Leishmania expression vector pXG-BSD, kindly provided by Dr. Stephen M. Beverly (Washington University School of Medicine, St. Louis, MO, USA), and sequenced. To evaluate the relevance of LHR2 heme binding in intracellular parasite replication, site-directed mutagenesis (Quik-Change II-XL Site-Directed Mutagenesis; Agilent Technologies, USA; forward 5´GTGCTAGGGTCGCTTCCAGGCGGTGAG3´ and reverse 5´CTCACCGCCTGGAAGCGACCCTAGCAC3´ primers) was performed to replace adenine (A) with thymine (T) within the HPGG heme-binding domain of LHR2, producing an H50L mutation. Log-phase L. major LHR2^+/- promastigotes were transfected with either the pXG-LHR2 or pXG-LHR2H50L construct and selected with increasing concentrations between 6–50 µg/mL of the antibiotic blasticidin (BSD) (Sigma-Aldrich), to generate stable cell lines.

LHR2 cellular localization by fluorescence microscopy

LHR2 was endogenously tagged at the C-terminus with the fluorescent marker mNeon Green (mNG) using the CRISPR/Cas9 gene-editing tool, as previously described [36]. Briefly, donor DNA was amplified from the pPLOT mNG-puro plasmid using the forward 5´GATTTCTCACCATGCCCACAGCGGGTACGCGGTTCTGGTAGTGGTTCCG3´ and reverse 5´AGCAACAGAAGGCGCTAGTAGAAGGAATGTCCAATTTGAGAGACCTGTGC3´ primers. The 3´sgRNA template was amplified using the forward GJ3Tag (described above) and G00 (sgRNA scaffold) primers designed by the LeishGEdit online resource. Log-phase L. major Cas9/T7 promastigotes [37] were co-transfected with donor DNA and sgRNA templates using program V-033 of the Amaxa Nucelofector System (Lonza) and selected in the presence of 30 µg/ml of puromycin. To study colocalization in the mitochondrion, promastigotes were incubated at 28°C with 50 nM of MitoTracker Red (ThermoFisher, Invitrogen, USA) for 30 min. Images were acquired with a fluorescence microscope (Leica Widefield Microscope DMi8 with and Orca-Flash 4.0 LT Digital CMOS camera) and deconvolved using the Iterative Deconvolve 3D ImageJ plugin (Fiji).

To analyze LHR2 localization in S. cerevisiae, BY4741 wild-type yeast expressing LHR2-eGFP were used in the exponential growth phase. Cells were washed twice with phosphate-buffered saline (PBS), fixed with 4% paraformaldehyde (PFA) for 10 min, and incubated with 1 μM of the endoplasmic reticulum (ER) marker ER-Tracker Red (ER-RED) (Invitrogen) at room temperature (RT) for 30 min and then with 0.1μg/mL of DAPI at RT for 10 min. Next, cells were washed twice in cold PBS and processed with a fluorescence microscope as described above.

Basal mitochondrial respiration assays in promastigotes of L. major

Promastigotes from LHR2^+/+ (Control) and mutant lines LHR2^+/- and LHR2^+/- AB were cultured at an initial concentration of 1x10⁶ parasites/mL in modified RPMI-1640 medium supplemented with 10% hiFBS at 28ºC for 48 hours. To assess basal respiration, parasites in log-phase of growth were adjusted to 50x10⁶ cell/mL in cell respiration buffer (10 mM Tris-HCl pH 7.2, 125 mM sucrose, 65 mM KCl, 1 mM MgCl₂, 25 mM NaH₂PO₄, 0.3 mM EGTA, 5 mM sodium succinate) in a final volume of 600 μL. Finally, the OCR (Oxygen Consumption Rate) was determined in real time under basal conditions, using the Oxygraph + /Oxytherm System equipment (Oxygraph System. V.2.20, Oxygraph + Oxygen Electrode System, Hansatech Instruments).

Measurement of mitochondrial membrane potential (∆ψm)

∆ψm was measured by determining Rh123 accumulation by flow cytometry, as described previously [48]. (Parasites (1 x 10⁶ promastigotes) were incubated with 0.5 µM Rh123 for 10 min at 28ºC. Parasites fully depolarized by incubation in 10 µM FCCP for 10 min at 28ºC, were used as controls. Next, the samples were washed twice with PBS and analyzed by flow cytometry (excitation: 455–507; emission: 529 nm).

Measurement of reactive oxygen species (ROS) production

The generation of intracellular ROS was measured using the cell-permeable fluorogenic probe MitoSOX red, which selectively targets mitochondria, as described in Manzano et al [73]. Thus, 1 x10⁶ promastigotes of the different L. major lines were loaded with 3 µM MitoSOX for 30 min at 28°C, and then washed and resuspended in PBS. Next, the fluorescence of oxidized MitoSOX red was measured by flow cytometry (excitation: 510 nm; emission: 580 nm).

L. major promastigotes sensitivity to AmB

The inhibitory concentration 50 (IC₅₀) of AmB in L. major promastigotes was determined using the MTT assay as previously described [74]. Briefly, promastigotes (4 × 10⁵ cells) were cultured in freshly prepared RPMI medium supplemented with 10% hiFBS, in the presence or absence of increasing concentrations of AmB (0–70 μg/mL), for 72 h at 28°C. After incubation, MTT solution (5 mg/mL) was added to each well, and samples were incubated for an additional 4 h at 28°C. During this period, the yellow tetrazolium MTT dye was reduced to insoluble purple formazan crystals by metabolically active cells. The resulting formazan crystals were solubilized with 20% (v/v) SDS and further incubated at 37°C for 4 h. Promastigotes cultured without AmB were used as untreated controls, while culture medium without parasites or AmB served as the blank. Absorbance was measured at 570 nm using a SpectraMax plate reader (Avantor) with SoftMax software. IC₅₀ values were calculated from dose–response curves generated using SigmaPlot software version 14.5. Three independent experiments were performed.

Measurement of Ergosterol levels

To assess ergosterol levels in the plasma membrane, filipin staining was employed [47,48]. (LHR2^+/+ (control) and LHR2^+/– promastigotes were cultured in RPMI-1640 medium (Invitrogen) supplemented with 10% heat-inactivated fetal bovine serum (hiFBS, Invitrogen) at 28 °C, starting at an initial density of 1 × 10⁶ parasites/mL. Parasites in mid-log phase (1 × 10⁷ cells/mL) were harvested by centrifugation at 1000 × g for 10 min at 4 °C and washed twice with 1 × PBS. The parasites were then incubated on ice with 2 μM filipin for 10 min in culture medium without hiFBS, washed twice with PBS, and analyzed by flow cytometry to determine sample fluorescence (excitation: 360 nm; emission: 480 nm).

In vitro and in vivo infection assays

Primary bone marrow cells isolated from mice were differentiated into macrophages for seven days in high-glucose DMEM medium (Gibco, Life Technologies) supplemented with macrophage colony-stimulating factor (M-CSF) (Sigma-Aldrich) on coverslips in 24-well plates [75]. These bone marrow-derived macrophages (BMDM) were then infected with stationary phase promastigotes of L. major lines, as previously described [37], staining and acquisition of fluorescence microscopy images was performed as described [76]. Briefly, infected cells were treated with CellMask Deep Red (Invitrogen, ThermoFisher, Waltham, USA) at 37 °C for 10 min according to the manufacturer’s instructions for macrophage labeling and 2 µg/mL of DAPI (Thermo Scientific ThermoFisher, Waltham, USA) for 5 min for macrophage and intracellular parasite DNA labeling, and visualized by fluorescence microscopy. Images were acquired with a fluorescence microscope (Widefield Microscope DMi8 with Orca-Flash 4.0 LT Digital CMOS camera Leica), using the Leica Application Suite X software platform (LAS X) with acquisition conditions of 385 nm filter (LED_405) (15%) excitation intensity for DAPI and 635 nm filter (DFTCY5) (15%) excitation intensity for CellMask Deep Red, 0.45 numerical aperture, magnification of 63x/1.40 and exposure times between 2–15 ms. The infection index (the percentage of infected macrophages multiplied by the average number of parasites per macrophage) was determined by FiCRoN, a recently developed deep learning quantification algorithm [77]. Experiments were performed twice in triplicate. Statistical analysis was performed using GraphPad Prism7 software using the Student’s t-test.

Briefly, seven-week-old male BALB/c mice were purchased from the Charles River Breeding Laboratories and maintained in our Animal Facility Service under pathogen-free conditions. Mice, divided into three groups of six animals, were injected subcutaneously in their right hind footpads with 10⁴ metacyclic promastigotes of LHR2^+/+, LHR2^+/-, LHR2^+/- AB (add-back), or LHR2^+/- H50L lines, resuspended in 40 ml of Dulbecco’s PBS w/Ca²⁺ and w/Mg²⁺ (Biowest, Nuaillé, France), Metacyclic promastigotes were purified from stationary-phase cultures by negative selection with peanut agglutinin (PNA) as previously described [78]. Inflammation progression (difference between inoculated and contralateral uninfected footpads) was recorded weekly using a Digimatic Caliper (Mitutoyo, Kawasaki, Japan).

LHR2 protein expression and purification

The genes encoding LHR2 and the mutant variant LHR2 H50L, in which histidine at position 50 was substituted with leucine as above described, were cloned into the pET21b+ vector (Novagen) for the expression of recombinant proteins carrying a C-terminal hexahistidine tag in Escherichia coli BL21 (DE3). Protein purification was performed under native conditions using Ni-NTA (nitrilotriacetic acid) agarose affinity chromatography, following the manufacturer’s instructions (Qiagen GmbH, Hilden, Germany).

Heterologous expression of LHR2 in yeast and drug susceptibility testing

S. cerevisiae BY4741 wild type, DAP1∆ (BY4741; MATa; his3Δ1; leu2Δ0; met15Δ0; ura3Δ0; YPL170w::kanMX4), and CYB5∆ (BY4741; MATa; his3Δ1; leu2Δ0; met15Δ0; ura3Δ0; YNL111c::kanMX4), obtained from EUROSCARF (European S, cerevisiae Archive for Functional Analysis), were grown at 30°C in YPD (yeast extract–peptone–dextrose). LHR2 was amplified by PCR using the primer pair forward: 5´GGCCCGGGATGCTTAACGATTTGCTATTTCTTAG3´ (restriction sites are underlined), or FwGJBam and reverse RvGJBa (described above), cloned into the XmaI and BamHI sites of the yeast expression vector pADH-LEU and the BamHI site of the yeast vector eGFP_ADH1_pESC-URA (eGFP-tagged version of LHR2), and sequenced. Then, yeast was transformed using lithium acetate medium (Gietz, 2014),and transformants were selected on 2% w/v glucose SC (− Leu) and SC (− Ura, LHR2-eGFP yeast) plates. The spot growth assay was performed as described in [80] using yeast in the exponential growth phase. Briefly, cells were grown overnight at 30°C with constant shaking at 200 rpm in 5-ml cultures. Yeast concentration was determined by hemocytometer and diluted using serial 1:5 dilutions from an optical density at 600 nm (OD₆₀₀) of 4.0. Five μl of each dilution was spotted on the medium to be tested. For the streak method, yeasts with OD₆₀₀ = 0.032 were used. Plates were incubated at 30 °C for 3–4 days. These assays were performed in duplicate. Fluconazole, methyl methanesulfonate (MMS), and H₂O₂ were obtained from Sigma-Aldrich (USA).

Heterologous expression of ScDAP1p and ScCYB5 in L. major

The forward 5´GGGGGATCCATGTCCTTCATTAAAAACTTGTTATTTGG3´ and reverse 5´CCCGGATCCTCATACGTTCACGCCAGGCTC-3´ primer pair were used to PCR amplify the ORF of yeast DAP1 and the forward 5´ GGGGGATCCATGCCTAAAGTTTACAGTTACC3´ and reverse 5´ CCCGGATCCTTATTCGTTCAACAAATAATAAGCAAC3´ primer pair were used to PCR amplify the ORF of yeast CYB5. The restriction sites are underlined in each sequence. Next, the amplified portion was cloned into the BamHI site of the pXG-SAT vector kindly provided by Dr. Stephen M. Beverly (Washington University School of Medicine, St. Louis, MO, USA), which was transfected in log-phase promastigotes of L. major by electroporation using program V-033 of the Amaxa Nucelofector System (Lonza). Overexpressed cells were maintained at 50 µg/ml nourseothricin (Sigma-Aldrich). LHR2 was then eliminated by CRISPR-Cas9 as above described.

Domain-structural analysis

The LHR2 amino acid sequences were used to perform a domain-structural analysis in the Pfam database. Three relevant entries were identified: The Cytochrome b5-like heme/steroid-binding domain superfamily (IPR036400), the Cytochrome b5-like heme/steroid-binding domain (IPR001199), and Cytochrome b5 (IPR050668). The accession number IPR001199 was subsequently queried in the AlphaFold cross-reference section of Pfam to identify proteins containing this domain in Saccharomyces cerevisiae (strain ATCC 204508/ S288c). Among the six proteins retrieved, ScCYB5 and ScDAP1 were confirmed as domain-containing candidates and were of a similar size to LHR2.

Protein structure molecular model and molecular docking

The 3D structure for protein LHR2 (Uniprot ID E9ADU2) was generated using the I-TASSER server. For the proteins ScCyb5 (Uniprot ID P40312) and ScDap1 (Uniprot ID Q12091), the 3D structures were obtained from the AlphaFold database. Molecular docking was performed with Autodock Vina 1.2.5 from the AutoDock suit [81] with the flexible docking option. The proteins were prepared for docking using the mk_prepare_receptor.py script, and the heme group was prepared with the mk_prepare_ligand.py from the suite. The flexible residues included HIS36 and HIS60 (LHR2), HIS37 and HIS61 (ScCyb5), and TYR84, TYR138, and LYS137 (ScDap1p). All proteins and complex visualizations were performed in ChimeraX-1.8 [54]. The electrostatic option was used for Coulombic electrostatic potential representation, and structural alignments were generated using Matchmaker, included in ChimeraX-1.8.

Statistical analysis

All data are presented as the mean and standard deviation (SD) or standard error (SEM). Statistical significance was determined using the Student’s t-test. Graphical illustrations and statistical analyses were performed using GraphPad Software, Version 7.