Trichomonas vaginalis lipophosphoglycan mutants have reduced adherence and cytotoxicity to human ectocervical cells

Abstract

The extracellular human pathogen Trichomonas vaginalis is covered by a dense glycocalyx thought to play a role in host-parasite interactions. The main component of the glycocalyx is lipophosphoglycan (LPG), a polysaccharide anchored in the plasma membrane by inositol phosphoceramide. To study the role of LPG in trichomonads, we produced T. vaginalis LPG mutants by chemical mutagenesis and lectin selection and characterized them using morphological, biochemical, and functional assays. Two independently selected LPG mutants, with growth rates comparable to that of the wild-type (parent) strain, lost the ability to bind the lectins Ricinnus comunis agglutinin I (RCA120) and wheat germ agglutinin, indicating alterations in surface galactose and glucosamine residues. LPG isolated from mutants migrated faster than parent strain LPG on sodium dodecyl sulfate-polyacrylamide gel electrophoresis, suggesting the mutants had shorter LPG molecules. Dionex high-performance anion exchange chromatography with pulsed amperometric detection analyses revealed galactosamine, glucosamine, galactose, glucose, mannose/xylose, and rhamnose as the main monosaccharides of T. vaginalis parent strain LPG. LPG from both mutants showed a reduction of galactose and glucosamine, corresponding with the reduced size of their LPG molecules and inability to bind the lectins RCA120 and wheat germ agglutinin. Mutant parasites were defective in attachment to plastic, a characteristic associated with avirulent strains of T. vaginalis. Moreover, the mutants were less adherent and less cytotoxic to human vaginal ectocervical cells in vitro than the parental strain. Finally, while parent strain LPG could inhibit the attachment of parent strain parasites to vaginal cells, LPG from either mutant could not inhibit attachment. These combined results demonstrate that T. vaginalis adherence to host cells is LPG mediated and that an altered LPG leads to reduced adherence and cytotoxicity of this parasite.

FIG. 1.

Fluorescent lectin labeling of parent (PA) and LPG mutant parasites (2E2 and 4-12). Parasites were fixed with 5% formalin, washed with PBS, and labeled with the fluorescently labeled lectins fluorescein isothiocyanate-RCA120 (FITC-RCA₁₂₀) or rhodamine-wheat germ agglutinin (Rho-WGA), as indicated.

FIG. 2.

Growth curve comparing T. vaginalis parent strain with LPG chemical mutants 2E2 and 4-12. Parasites were resuspended to 10⁴ parasites/ml and suspensions were aliquoted in 15-ml conical tubes at 37°C. At the points indicated, triplicates of each sample were incubated on ice for 10 min to release parasites that had adhered to culture tube and cells were counted in a hemacytometer. The graph represents data from three independent experiments (mean ± standard deviation).

FIG. 3.

SDS-PAGE of extracted LPG from parent (PA) and mutant parasites. LPG was prepared by sonication of parasites, DNase and RNase treatment, followed by proteinase K digestion and phenol extraction. The aqueous phase was concentrated and extracted with solvent E (H₂O, ethanol, diethylether, pyridine, NH₄OH, 15:15:5:1:0.02) as previously described (48). Purified LPG was run on a 12% SDS-PAGE gel under reducing conditions and stained with polysaccharide-specific silver stain.

FIG. 4.

Dionex HPAEC-PAD rhamnose ramp. Chromatogram from HPLC analysis of monosaccharides using CarboPac PA-1 column. Monosaccharides from purified parent LPG (a) or standards (b) were resolved by Dionex HPAEC-PAD to confirm the presence of rhamnose in T. vaginalis LPG. The retention time for rhamnose in the standards coincides with the rhamnose peak in the LPG sample.

FIG. 5.

Adherence to plastic by parent and LPG mutant parasites. Parent (PA) and LPG mutant parasites were grown to mid-log phase, washed, and resuspended in TYM medium, and 5 × 10⁵ cells were distributed in triplicate in the wells of 24-well tissue culture plates. After 4 h incubation at 37°C and 5% CO₂, wells were washed three times and attached parasites were stained with crystal violet. Absorbance was then measured at 570 nm after dissolving the dye in SDS-ethanol. The graph depicts data from a representative experiment with standard deviations.

FIG. 6.

Adherence of parent (PA) and LPG mutant parasites to human ectocervical cell monolayers. Mid-log-phase parasites were labeled with Cell Tracker Blue (Molecular Probes). Labeled parasites were then incubated for 30 min with ectocervical cell monolayers grown on coverslips in 24-well plates at 37°C and 5% CO₂. Coverslips were washed to remove nonadherent parasites, mounted, and visualized by fluorescence microscopy (a). Data are from three experiments showing the average number of parasites counted per coverslip with standard deviations (b).

FIG. 7.

Adherence of T. vaginalis parent strain to human vaginal ectocervical inhibited by parent (PA) but not mutant LPG. Adherence assays were performed as described in Materials and Methods except ectocervical cells were incubated with 150 μg of purified LPG for 1 h before addition of parasites. The average number of parasites counted per coverslip is shown with standard deviations.

FIG. 8.

Cytotoxicity of parent (PA) and LPG mutant parasites to ectocervical cells. Parasites were washed and incubated in 48-well tissue culture plates with human ectocervical cell monolayers for 8 h at 37°C and 5% CO₂. Release of lactate dehydrogenase in the supernatants from the mammalian cells was determined with the Cytotox ONE kit (Promega), following the manufacturer's instructions. Data from a representative experiment are shown, expressed as percent cytotoxicity calculated from the maximum release of lactate dehydrogenase after total lysis of ectocervical cells.