Cyclosporin A treatment of Leishmania donovani reveals stage-specific functions of cyclophilins in parasite proliferation and viability

Abstract

Background: Cyclosporin A (CsA) has important anti-microbial activity against parasites of the genus Leishmania, suggesting CsA-binding cyclophilins (CyPs) as potential drug targets. However, no information is available on the genetic diversity of this important protein family, and the mechanisms underlying the cytotoxic effects of CsA on intracellular amastigotes are only poorly understood. Here, we performed a first genome-wide analysis of Leishmania CyPs and investigated the effects of CsA on host-free L. donovani amastigotes in order to elucidate the relevance of these parasite proteins for drug development.

Methodology/principal findings: Multiple sequence alignment and cluster analysis identified 17 Leishmania CyPs with significant sequence differences to human CyPs, but with highly conserved functional residues implicated in PPIase function and CsA binding. CsA treatment of promastigotes resulted in a dose-dependent inhibition of cell growth with an IC50 between 15 and 20 microM as demonstrated by proliferation assay and cell cycle analysis. Scanning electron microscopy revealed striking morphological changes in CsA treated promastigotes reminiscent to developing amastigotes, suggesting a role for parasite CyPs in Leishmania differentiation. In contrast to promastigotes, CsA was highly toxic to amastigotes with an IC50 between 5 and 10 microM, revealing for the first time a direct lethal effect of CsA on the pathogenic mammalian stage linked to parasite thermotolerance, independent from host CyPs. Structural modeling, enrichment of CsA-binding proteins from parasite extracts by FPLC, and PPIase activity assays revealed direct interaction of the inhibitor with LmaCyP40, a bifunctional cyclophilin with potential co-chaperone function.

Conclusions/significance: The evolutionary expansion of the Leishmania CyP protein family and the toxicity of CsA on host-free amastigotes suggest important roles of PPIases in parasite biology and implicate Leishmania CyPs in key processes relevant for parasite proliferation and viability. The requirement of Leishmania CyP functions for intracellular parasite survival and their substantial divergence form host CyPs defines these proteins as prime drug targets.

Figure 1. Diagram representing the L. major CyP-like proteins.

17 CyP-like proteins are annotated in the L. major genome database (LmaCyPs). The cyclophilin-like domain (CLD) and other domains were identified with ScanProsite. Most LmaCyPs are characterized by parasite-specific N- and C-terminal extensions. Additional functional domains are identified in two LmaCyPs, LmjF31.0050 (LmaCyP5) and LmjF35.4770 (LmaCyP40), containing a prokaryotic lipid attachment domain (PLD) and tetratricopeptide repeat domains (TPR), respectively. The bar represents 100 amino acids.

Figure 2. Bioinformatics analysis of the L. major CyP-like protein family.

(A) Neighbor-Joining tree (500 bootstrap replicates) of the 118 CyP-like proteins. CyP proteins were identified by PSI-Blasting Human PPIaseA against Human, L. major, L. infantum, L. braziliensis, T. brucei, and T. cruzi genomes. Multiple sequence alignment was performed with T-Coffee and fed into the MEGA4 software package. Numbers on nodes indicate bootstrap support. (B) UPGMA clustering of CyP functional residues. Positions corresponding to the CsA binding sites (as defined on the Human PPIaseA) are displayed on the figure along with the CyP-like protein they originate from. The UPGMA clustering shows groups of putatively identical binding sites.

Figure 3. CsA inhibits L. donovani in vitro growth.

(A) Parasites were treated for up to 48 hours with CsA at the concentrations indicated at 26°C and pH 7.4 for promastigotes, or 37°C and pH 5.5 for amastigotes. Cell density of the samples was estimated using CASY cell counter and expressed in cell density per milliliter. (B) Left panel, growth assessment. Logarithmic promastigotes (pro, □) and axenic amastigotes (ama, ○) were incubated for 48 hours with CsA at the indicated concentrations. Cell density of the samples was estimated using hemocytometer and expressed in % of growth compared to solvent treated controls. Right panel, cell proliferation and viability assay. Promastigotes and axenic amastigotes were treated as detailed in (A). 20 µl of CellTiter-Blue solution was added to the cells after 48 hours of CsA treatment, and the assay was further incubated for 4 hours at 37°C. Resazurin reduction was expressed in % of fluorescence compared to solvent treated cells control. Results are representative of three quadruplicate experiments with mean ± S.D. indicated by the error bars.

Figure 4. CsA affects L. donovani viability and proliferation.

(A) FACS analysis. Logarithmic promastigotes (black bars) and axenic amastigotes (grey bars) were incubated for 48 hours with the indicated CsA concentrations. The cells were then washed once with PBS, stained with 2 µg/ml of PI and analyzed by FACS. Proportion of dead parasites is expressed in % of PI positive (+) stained cells after subtracting the background of solvent treated cells control. The error bars represent the mean±S.D. of four independent experiments. (B) Cell cycle analysis. CsA-treated promastigotes were fixed in cold 90% methanol and stained with propidium iodide for cell cycle analysis. The stained cells were subjected to FACS analysis (exλ = 488 nm/emλ = 617 nm). 10,000 events were analyzed and cell cycle distribution was calculated with the model Dean-Jett-Fox using the FlowJo Software package (Tree Star, Inc.). At least two independent experiments were performed and representative results are shown.

Figure 5. CsA-treated L. donovani promastigotes show altered morphology.

Promastigotes were incubated with 0.15% ethanol or 15 µM (B, C) or 20 µM (A) CsA at 26°C, pH 7.4 for 72 hours. Axenic amastigotes were prepared as described in experimental procedure. 10⁷ cells were fixed with either methanol for Giemsa staining (A), or 2.5% glutaraldehyde for scanning electron microscopy (B). The bar corresponds to 1 µm (B) and 5 µm (A). Two independent experiments were performed and representative fields are shown. (C) Flagellum length measurement. CsA-treated and solvent treated cells L. donovani promastigotes were fixed in methanol and stained with anti-tubulin monoclonal antibody. Flagellum length was measured from a total of 180 cells each for control and CsA-treated samples. Only cells with a single flagellum that was completely visible and fully in focus were taken into account. Samples were observed with a DMR Leica microscope and images were captured with a Cool Snap HQ camera (Roper Scientific). Images were analysed using the IPLab Spectrum 3.9 software (Scanalytics & BD Biosciences) and flagellum length was measured using ImageJ (NIH). (D) Immunoblot analysis of CsA treated parasites. Parasites were treated with solvent or 15 µM CsA for 72 hours, lysed in 1× Laemmli buffer, and lysates equal to 2×10⁷ cells were analyzed by immunoblotting. Promastigote specific marker LPG (upper), amastigote specific marker A2 (middle) and α-tubulin (lower) were analyzed. Two independent experiments which gave identical results were performed.

Figure 6. The stage-specific effects of CsA occur through distinct mechanisms.

Parasites were treated for 48 hours with either CsA or FK506 at the concentrations indicated at 26°C and pH 7.4 for promastigotes, or 37°C and pH 5.5 for amastigotes. (A) Cell density of the samples was estimated using CASY cell counter and expressed in cell number per milliliter, after 24 to 48 hours treatment. (B) Cell proliferation was measured using CellTiter-Blue cell viability assay by following resazurin reduction, which is expressed in % of fluorescence compared to solvent treated cells control. (C) Cell death was measured by propidium iodide staining and FACS analysis as detailed in legend of Fig. 3. Results of (A) are representative of three quadruplicate experiments with mean ± S.D represented by the error bars. Three independent experiments were performed for (B) with the error bars representing ±S.D.

Figure 7. CsA affects L. donovani thermotolerance.

Axenic amastigotes and promastigotes were either treated with solvent or 15 µM CsA, and incubated at either 26°C or 37°C. At the time points indicated, aliquots of the cells were stained with propidium iodide and analyzed by FACS. The proportion of dead parasites is expressed in % of PI positive (+) cells. Three independent experiments were performed, and one representative triplicate experiment is shown. The error bars represent ±S.D.

Figure 8. Leishmania cyclophilin 40 is a target for CsA.

(A) Structural modelling. Cyclosporin (CsA) binding pockets of L. donovani cyclophilin CyP2 (PDB code: 3eov, left upper panel) and L. major CyP40 in the presence of the CsA ligand (model complex, right upper panel) are shown. Cyclophilin residues are coloured in yellow and CsA is coloured by atom type. H-bonds are displayed as green dotted lines. The lower panel shows a multiple sequence alignment of the CsA binding regions of L. donovani CyP2 (PDB code: 3eov) and the six L. major cyclophilins analyzed. The residues in close proximity to the ligand are shown in bold. Yellow and orange filling identifies residues forming respectively one or two hydrogen bonds with CsA atoms. (B) Affinity chromatography and Western blotting. Total protein extracts obtained from logarithmic L. donovani promastigote cultures were incubated with resin alone (−), or resin coupled with CsA (+), bound proteins were analyzed by SDS-PAGE and SyproRuby staining, and identified by MS analysis (left panel, CyP2), or Western blotting using polyclonal anti-LmaCyP40 antibody (right panel). In, input; Ft, flow through; El, eluate. (C) PPIase assay. The cis/trans isomerization activity of recombinant Leishmania major GST::Strep::CyP40 was performed using Abz-Ala-Ala-Pro-Phe-pNA as substrate. Determination of k_cat/K_m (upper panel) was performed by evaluation of the linear dependency of k_enz from the concentration of GST::Strep::CyP40. Each data point represents the mean of three independent measurements. The lower panel shows inhibition of Leishmania major GST::Strep::CyP40 peptidyl prolyl cis/trans isomerase activity by CsA. PPIase activity was measured at increasing amounts of CsA using the substrate Abz-Ala-Ala-Pro-Phe-pNA.