Proteome and morphological analysis show unexpected differences between promastigotes of Leishmania amazonensis PH8 and LV79 strains

Abstract

Background: Leishmaniases are diseases caused by Leishmania protozoans that affect around 12 million people. Leishmania promastigotes are transmitted to vertebrates by female phlebotomine flies during their blood meal. Parasites attach to phagocytic cells, are phagocytosed and differentiate into amastigotes. We previously showed that PH8 and LV79 strains of Leishmania amazonensis have different virulence in mice and that their amastigotes differ in their proteomes. In this work, we compare promastigotes' infectivity in macrophages, their proteomes and morphologies.

Methods/principal findings: Phagocytosis assays showed that promastigotes adhesion to and phagocytosis by macrophages is higher in PH8 than LV79. To identify proteins that differ between the two strains and that may eventually contribute for these differences we used a label-free proteomic approach to compare promastigote´s membrane-enriched fractions. Proteomic analysis enabled precise discrimination of PH8 and LV79 protein profiles and the identification of several differentially abundant proteins. The proteins more abundant in LV79 promastigotes participate mainly in translation and amino acid and nucleotide metabolism, while the more abundant in PH8 are involved in carbohydrate metabolism, cytoskeleton composition and vesicle/membrane trafficking. Interestingly, although the virulence factor GP63 was more abundant in the less virulent LV79 strain, zymography suggests a higher protease activity in PH8. Enolase, which may be related to virulence, was more abundant in PH8 promastigotes. Unexpectedly, flow cytometry and morphometric analysis indicate higher abundance of metacyclics in LV79.

Conclusions/significance: Proteome comparison of PH8 and LV79 promastigotes generated a list of differential proteins, some of which may be further prospected to affect the infectivity of promastigotes. Although proteomic profile of PH8 includes more proteins characteristic of metacyclics, flow cytometry and morphometric analysis indicate a higher abundance of metacyclics in LV79 cultures. These results shed light to the gaps in our knowledge of metacyclogenesis in L. amazonensis, and to proteins that should be studied in the context of infection by this species.

Fig 1. In vitro phagocytosis assay with PH8 and LV79 promastigotes.

A. Number of promastigotes of PH8 and LV79 strains adhered to 500 macrophages. B. Number of promastigotes of PH8 and LV79 strains phagocytosed by 500 macrophages. Phagocytosis assays were performed with murine bone marrow-derived macrophages and a MOI of 10:1. Data represent mean ± SD of three technical replicates. Statistical analysis by Student’s t-test, *: p < .05, **: p < .01, ***: p < .001. Representative results of two experiments with similar profile.

Fig 2. Validation of sodium carbonate extraction with LV79 promastigotes.

A. 20 μg of protein of total cell (T), cytoplasmic (C) and membrane-enriched (M) fractions were analyzed by SDS-PAGE. B. Abundance of GP63 and TXNPx in 15 μg of total (T), cytoplasmic (C) and membrane-enriched (M) fractions were compared by Western blot.

Fig 3. Comparison of membrane-enriched proteomes of PH8 and LV79 promastigotes.

A. Venn diagram showing the number of proteins identified exclusively in PH8, in LV79 and identified in both strains. B. Clustering of the six samples (three biological samples (S1, S2 and S3) for each strain) by principal component analysis of all proteins identified. C. Heat map constructed based on the hierarchical clustering of the 6 samples (three samples (S1, S2 and S3) for each strain) based on Z-scores calculated from the log2 of LFQ Intensity values of differentially abundant proteins identified after Student’s t test with Benjamini-Hochberg correction and FDR = 0.05.

Fig 4. Classification according to subcellular localization of all proteins identified in the membrane-enriched proteome of PH8 and LV79 promastigotes.

A. Circle chart showing percentage of proteins identified in proteomic analysis belonging to each subcellular localization. B. Circle chart showing percentage soluble and integral membrane proteins C. Circle chart showing percentage of integral membrane proteins identified in proteomic analysis belonging to each subcellular localization.

Fig 5. Classification of differentially abundant proteins based on biological process.

A. Circle chart showing percentage of the proteins more abundant in PH8 promastigotes belonging to each biological process. B. Circle chart showing percentage of the proteins more abundant in LV79 promastigotes belonging to each biological process. C. Bar diagram showing the percentage of differential proteins from each biological process detected as more abundant in LV79 and in PH8 strains.

Fig 6. GP63 abundance and activity in L. amazonensis promastigotes of LV79 and PH8 strains.

A. GP63 abundance (non-log transformed LFQ intensities) in proteomes of membrane-enriched fractions of LV79 and PH8 promastigotes. B. GP63 (upper figure) and α- tubulin (lower figure) abundances in total extracts (left) and membrane-enriched fractions (right) (three biological samples for each strain) of LV79 and PH8 by Western blot. For membrane-enriched fractions, GP63 is shown together with tubulin labeling in the upper figure. Graphs show normalized values (GP63/ α- tubulin). Graph C. GP63 proteolytic activity in total extracts (three biological samples for each strain) was measured by zymography (one experiment representative of three). A non-reducing Western blot for GP63 is shown in the right. Data in A and B represent means and SD of three biological samples (S1, S2 and S3) for each strain. For A, statistical analysis was performed by Student’s t test with Benjamini-Hochberg correction (FDR = 0.05) and the resulting q-value is shown in graph. For B, statistical analysis was performed by Student’s t-test, *: p < .05, **: p < .01, ***: p < .001.

Fig 7. Enolase abundance and activity in L. amazonensis promastigotes of LV79 and PH8 strains.

A. Enolase abundance (non-log transformed LFQ intensities) in proteomes of membrane-enriched fractions of LV79 and PH8. B. Enolase abundance in total extracts (left) and membrane-enriched fractions (right) (three biological samples (S1, S2 and S3) for each strain) of LV79 and PH8 by western blot. C. Enolase activity in total extracts was monitored by the PEP conversion to 2-PGA, which was measured spectrophotometrically at 240 nm. Data in A, B and C represent means and SD of three, three and five biological replicates, respectively. For A, statistical analysis was performed by Student’s t test with Benjamini-Hochberg correction (FDR = 0.05) and the resulting q-value is shown in graph. For B and C, statistical analysis was performed by Student’s t-test, *: p < .05, **: p < .01, ***: p < .001.

Fig 8. Frequency of metacyclic promastigote forms in L. amazonensis LV79 and PH8 strains.

L. amazonensis promastigotes from LV79 and PH8 strains at early stationary phase (day 4) were analyzed by flow cytometry. A. Dot plots of SSC and FSC features representative of Leishmania gating strategy in FSC^low and SSC, excluding debris, analyzed using BD LSR Fortessa Cell Analyser (Becton Dickinson). B. Frequency of metacyclic promastigotes in three independent experiments containing 5 replicates of each strain. Each bar represents the mean of the percent ±SD of metacyclics. Statistical analysis by Student’s t-test, ****: p < 0.0001.

Fig 9. Proportion of procyclics, nectomonads, leptomonads and metacyclics in log phase (day 2) and early stationary phase (day 4) cultures of L. (L.) amazonensis LV79 (A) and PH8 (B) strains.

Flagellum and body length from 150 parasites were measured using ImageJ and parasites were classified according to [21].

Fig 10. Relative expression of SHERP and META1 in LV79 and PH8 cultures.

Synchronized cultures from day 2 (D2) and day 4 (D4) were collected for comparative analysis of transcripts of SHERP, META1 and GAPDH. Data obtained by RealTime RT-PCR of three independent cultures for each strain, using GAPDH as the reference gene and mean values of LV79 D2 for normalization. Statistical analysis by ANOVA with Tukey posttest.