Fig 1.
A platform to study the interactions between T. vaginalis trophozoites and human epithelial cells.
A semi-quantitative binding assay in conjunction with real-time imaging and scanning microscopy (SEM) were applied to study the interactions between T. vaginalis and human urogenital tract epithelial cells. For the binding assay or real-time microscopy, trophozoites were labeled with CFSE or Hoechst 33342, respectively, while non-labeled trophozoites were explored for SEM. At defined intervals, unbound trophozoites were removed and the adherent trophozoites were fixed for microscopic examination.
Fig 2.
The kinetics of T. vaginalis trophozoites binding to hVECs.
The trophozoites were labeled with CFSE and co-cultured with hVECs. A. trophozoites at a moi of 3 to 1, 1 to 1 or 1 to 3 were incubated with hVECs for 30 min. B. T1 or TH17 trophozoite were co-incubated with hVECs (moi 1:3) for 1, 15, 30 or 60 min. C. trophozoites were co-incubated with hVECs (moi 1:3) in the presence of 250 mM sucrose or lactose for 30 min and after the removal of unbound trophozoites, cell cultures were fixed for imaging by fluorescence and phase-contrast microscopy. PI, post-infection. The clustered foci with over three trophozoites are circled by white-dashed lines. The experiments were performed in triplicate and the average number of binding trophozoite or clustered foci per 1,000 hVECs were measured as shown in the bar graphs (mean ± SD). The statistical difference was analyzed by the Student’s t-test with P<0.05 (*), P< 0.01(**), and ns, not significant.
Fig 3.
Involvement of galectins in the host-parasite interaction.
A and B. Surface expression or secretion of galectins. CFSE-labeled TH17 trophozoites (green) were co-incubated with hVECs (moi 1:3) were sampled at intervals for the detection of galectin-1 (red) and galectin-3 (cyan) by IFA (A), while these galectins in the conditioned medium were detected by ELISA (B). PI, post-infection. Adherent trophozoites in A are indicated by white triangles. C, D, and E. T1 or TH17 trophozoites were incubated in fresh or conditioned KSFM (D) from hVECs cultures for 30 min. Cells were fixed for detection of galectin-1 or -3 by IFA without permeation for (C) or with permeation for (D and E) before immune detection. E. IFA co-staining with anti-galectin-3 antibody and LysoTracker. F and G. T1 or TH17 trophozoites were incubated with FITC-conjugated recombinant galectin-1 or galectin-3 and analyzed by flow cytometry (F) or fluorescence microscopy (G). The bar graph is shown by mean ± SD. All assays were repeated three times and the representative data are shown here. Differences were statistically analyzed by Student’s t-test, with P<0.01(**) and P<0.05(*).
Fig 4.
Morphological transformation of T. vaginalis on exposure to hVECs.
A. TH17 trophozoites (a) were co-incubated with hVECs for 15 (b. and c.), 30 (d, e, f), or 60 min (g and h) for SEM. White triangles point to the flagellum (Fg), axostyle (Ax), or membrane protrusions (Pt). The representative morphology was observed from at least ~50 trophozoites at stages as specified in the infection course. B. The trophozoites treated with DMSO or CytD were cultured in T25 flasks for 1 hr and the morphology was monitored by phase-contrast microscopy. The black and white arrowheads respectively indicate the amoeboid and flagellate forms of T. vaginalis. The percentage of flagellate against amoeboid trophozoites was measured in 600 trophozoites within 6 microscopic fields as shown in the bar graph. C. TH17 trophozoites treated with DMSO or CytD were fractionated into F-actin and G-actin-containing fractions for western blotting. The ratio of the α-actin signal in G-actin versus F-actin was quantified as shown in the bar graph. α-Tubulin (α-Tub) and TvCyP2 were detected as the purity markers of F-actin and G-actin fractions, respectively. D. The trophozoites pretreated with DMSO or CytD were co-cultured with hVECs (moi 1:3) for 1 hr in cytoadherence binding assay. The number of trophozoites bound per 1000 hVECs was measured as shown in the bar graph (mean ± SD). All experiments were repeated three times. Significant differences were statistically analyzed by the Student’s t-tests, with P<0.01(**) and P<0.05(*).
Fig 5.
Real-time imaging of the host-parasite interactions.
The behavior of T1 or TH17 trophozoites in the presence (A) or absence (B) of hVECs was recorded by real time-imaging (S4, S5 and S6 Videos). A representative touchdown scenario of a T1 or TH17 trophozoite on the glass surface via the axostyle is captured as shown in (A). Anchor of a Hoechst 33342-labeled TH17 trophozoite on a hVECs over 4 min with contiguous bulging and shrinking of its axostyle is shown in (B). Notably, the duration of the anchor varied in the individuals. Real-time images were captured by DIC at the rate of 1 frame per 30 sec over 30 min. White triangles indicate where the parasite axostyle (Ax) or human cells (hVECs) are located.
Fig 6.
The involvement of the axostyle in the cytoadherence of the TH17 isolate.
The trophozoites of TH17 isolate were pretreated with DMSO or TPI. A. the trophozoites cultured on a glass slide were examined by IFA using anti-α-tubulin or anti-α-actin antibodies. B. Axostyles in the trophozoites co-cultured with hVECs were observed by SEM. The arrowheads indicate axostyle (Ax); anterior flagellum (AF); recurrent flagellum (RF). C. Protein lysates from treated TH17 trophozoites were subjected to western blotting using anti-α-tubulin (α-Tub) or anti-GAPDH antibodies. D. CFSE-prelabelled trophozoites were co-cultured with hVECs (moi 1:3) for 30 min and the number of trophozoites bound per 1,000 hVECs was calculated as shown in the bar graph (mean ± SD). All assays were repeated three times. Significant differences were statistically analyzed by the Student’s t-tests with P< 0.01(**), P<0.05 (*), and ns, not significant.
Fig 7.
Motility of nonadherent versus adherent T. vaginalis isolates.
The tract of a single TH17 or T1 trophozoite migrating on a glass slide for 1 min is depicted in A. The migratory paths from ~50 trophozoites are shown in B, with the average track displacement length (TDL). The average velocities with standard deviations were statistically analyzed by Student’s t-tests as shown in C. The relative velocity of T1 in D. or TH17 in E., the trophozoites cultured in KSFM or with hVECs over 15 min was normalized by the average velocity of those in 1 min post-incubation in KSFM. The assays were repeated three times. For D. and E., the statistical analysis was measured by Bonferroni post hoc tests (n = 10 to 30 for each group). The error bars represent standard deviations, P<0.01(**) and P<0.05(*).
Fig 8.
Epigenetic divergence of adherent versus nonadherent T. vaginalis isolates.
Site-specific histone acetylation in T. vaginalis was examined by IFA using the antibodies as indicated on the top of each panel. The cell morphology was observed by microscopy at DIC mode. Scale bar represents 5 μm. The relative intensities of nuclear signals were quantified as shown in the bar graphs (n = 3, mean ± SD). Significant differences were statistically analyzed by Student’s t-test with P< 0.01(**) or P< 0.05 (*).
Fig 9.
Mycoplasma symbiosis detection in the TH17 isolate.
A. DNA extracted from T1, TH17, and PM1 isolates were amplified by PCR using specific primers for T. vaginalis 18s rRNA or Mycoplasma spp. 16s rRNA genes. The PCR products were separated in a 1% agarose gel. B. The trophozoites from T1, TH17, and PM1 isolates were fixed on a glass slide and stained with DAPI for confocal microscopy. The cell morphology was visualized by phase-contrast microscopy. The T. vaginalis nuclei are indicated by the letter N, and the Mycoplasma DNA puncta are indicated by white arrowheads. Scale bar represents 5 μm.
Fig 10.
The differential host-parasite interaction of T. vaginalis.
On infection, T. vaginalis may trigger the surface expression and secretion of galectins from hVECs. The extracellular galectins may initially bind on the parasite surface, then galectin-3 is internalized and enclosed in the lysosomes inside the trophozoites of adherent isolate. The extracellular galectins bound to the parasite might trigger the aggregation of trophozoites on hVECs, as this is diminished in the presence of lactose. In the adherent isolate, the predominant flagellate-amoeboid transition, as well as cytoadherence, were simultaneously suppressed by CytD, speculating that actin cytoskeleton-based behavior may mediate cytoadherence. When axostyle microtubule assembly was disturbed by TPI, the axostyle anchoring was abolished with the reduced cytoadherence, suggesting that axostyle anchoring might be an early event required for cytoadherence. For T. vaginalis, it is speculated that the cytoskeleton and cell surface adhesion molecules may coordinate their cytoadherence, so these factors should be considered in the study of host-parasite interactions and provide new insights into the host colonization of T. vaginalis.