Proteomic analysis of the action of the Mycobacterium ulcerans toxin mycolactone: targeting host cells cytoskeleton and collagen

Abstract

Buruli ulcer (BU) is a neglected tropical disease caused by Mycobacterium ulcerans. The tissue damage characteristic of BU lesions is known to be driven by the secretion of the potent lipidic exotoxin mycolactone. However, the molecular action of mycolactone on host cell biology mediating cytopathogenesis is not fully understood. Here we applied two-dimensional electrophoresis (2-DE) to identify the mechanisms of mycolactone's cellular action in the L929 mouse fibroblast proteome. This revealed 20 changed spots corresponding to 18 proteins which were clustered mainly into cytoskeleton-related proteins (Dync1i2, Cfl1, Crmp2, Actg1, Stmn1) and collagen biosynthesis enzymes (Plod1, Plod3, P4ha1). In line with cytoskeleton conformational disarrangements that are observed by immunofluorescence, we found several regulators and constituents of both actin- and tubulin-cytoskeleton affected upon exposure to the toxin, providing a novel molecular basis for the effect of mycolactone. Consistent with these cytoskeleton-related alterations, accumulation of autophagosomes as well as an increased protein ubiquitination were observed in mycolactone-treated cells. In vivo analyses in a BU mouse model revealed mycolactone-dependent structural changes in collagen upon infection with M. ulcerans, associated with the reduction of dermal collagen content, which is in line with our proteomic finding of mycolactone-induced down-regulation of several collagen biosynthesis enzymes. Our results unveil the mechanisms of mycolactone-induced molecular cytopathogenesis on exposed host cells, with the toxin compromising cell structure and homeostasis by inducing cytoskeleton alterations, as well as disrupting tissue structure, by impairing the extracellular matrix biosynthesis.

Figure 1. Kinetics of mycolactone cytostatic and cytotoxic effects.

Mouse fibroblasts L929 cells were incubated for 24, 48 or 72(referred to as 48 h+48 h). Cell cycle analysis (A) and annexin-V/PI assay (B) were performed for each-time point. Bars represent the mean + SD (n = 3) from one out of, at least, two independent experiments. Each condition was compared to EtOH-treated samples throughout each time-point (24 h, 48 h and 72 h) by Two-way ANOVA with Bonferroni posttest; statistical differences were represented by *** (P<0.001) for sub-G₀/G₁ and Annexin-V⁺/PI⁻ , and by ^### (P<0.001) for G₀/G₁ Phase and Annexin-V⁺/PI⁺. Each condition at 48 h+48 h time-point was compared with the same condition at the 48 h by Two-way ANOVA with Bonferroni posttest; statistical differences were represented by ^&&& (P<0.001) for sub-G₀/G₁ and Annexin-V⁺/PI⁻, and by ^§§§ (P<0.001) for G₀/G₁ Phase and Annexin-V⁺/PI⁺.

Figure 2. Mycolactone induces cytoskeletal alteration, cell round up and detachment.

Mouse fibroblasts L929 cells seeded on coverslips were incubated for 12, 18 or 24/mL of mycolactone. Cytoskeletal changes were visualized by immunofluorescence microscopy using rhodamine-phalloidin conjugate (red) and a tubulin-specific antibody (green). Nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI). The cellular shape was visualized by differential interference contrast (DIC) microscopy. White horizontal bar represents a 10 µm scale (A). Blue (nucleus) and red (actin) channels were used for 3D remodeling of the respective confocal z stacks (B).

Figure 3. Mycolactone causes time-dependent proteomic changes.

Mouse fibroblasts L929 cells were incubated for 24 or 48(50 ng/mL). Additionally, an assay was performed where cells were incubated for 48 hours in the same conditions followed by a 48 hour incubation period in fresh medium (referred to as 48 h+48 h). Representative silver-stained 2-D gels of total protein extracts (100 µg) are shown for the different exposure times (all analyzed gels are in Supplementary Information). Spots altered upon treatment are represented by numbers (1–20).

Figure 4. Altered proteins upon mycolactone treatment.

Figure shows the proteins identified for each spot (numbers according to figure 3 and figure S1) clustered into five categories. (1) Expression levels show decreased (red), increased (green) or no changes (black) in spot intensities upon mycolactone treatment. (2) MS and MS/MS parameters are presented, including the number of specific matched peptides (P), the sequence coverage (SC) in percentage, as well as the MS scores. Phosphoproteins (*) and putative protein fragments (#) are indicated. Full MS and MS/MS data in Supplementary Information.

Figure 5. Mycolactone causes an up-regulation of the ubiquitin-proteasome pathway and an accumulation of autophagosomes.

In A, B and C, mouse fibroblasts L929 cells were incubated for 24 or 48(−) or mycolactone (50 ng/mL, +). Additionally, an assay was performed where cells were incubated for 48 hours in the same conditions followed by a 48 hour incubation period in fresh medium (referred to as 48 h+48 h). At each time-point, total protein was extracted and Western blot was performed to assess ubiquitinated proteins (A) or LC3 processing (B). Additionally, cytospins were made to assess immunofluorescence LC3 (red). White horizontal bars represent a 10 µm scale (C). In D, mouse fibroblasts L929 cells were incubated for 48 hours either with ethanol or mycolactone (50 ng/mL) and bafilomycin A1 (10 nM, +) or DMSO (vehicle control, −) was added 2 hours prior the end of the assay. Additionally, mouse fibroblast L929 cells were incubated for 48 hours without any stimuli until 2 hours prior the end of the assay, when autophagy was induced with rapamycin (1 µM), and bafilomycin A1 (10 nM, +) or DMSO (vehicle control, −) was added. At the end of the assay, total protein was extracted and Western blot was performed to assess LC3 processing (D).

Figure 6. Mycolactone induces a decrease in dermal collagen fibers.

Mice were infected with virulent mycolactone-secreting (MU98912: dark gray triangles) (n = 5) or avirulent mycolactone-negative (MU5114: light gray squares) (n = 4) strains of M. ulcerans, or injected with PBS (vehicle: white circles) (n = 3) as a control (A, B and C). Alternatively, mice were inoculated with 5 µg of mycolactone (dark gray hexagons) (n = 4) or ethanol (vehicle: white diamonds) (n = 4) as a control (D, E, F). Lesion progression was assessed by measurement of footpad swelling (A and D). Collagen content was assessed by a qualitative blind scoring of the amount of dermal collagen fibers (score 0 = lowest, 4 = highest) visualized in HE-stained sections by polarized light (B and E). The different groups were compared to the control groups (PBS in A and B; ethanol in C and D) by Two-way ANOVA with Bonferroni posttest; statistical differences were represented by ** (P<0.01) or *** (P<0.001). Additionally, mouse footpads exposed to the different stimuli were stained with Masson's trichrome to highlight collagen fibers (green) (C and F). On panel are depicted representative histological image each group. Black horizontal bars represent a 50 µm scale. Histological structures like dermis (D) and the stratum corneum (SC) and stratum spinosum (SS) of the epidermis (ED) are indicated on the upper-left image. Inflammatory infiltrate is indicated by an asterisk.