Heterologous expression in Tritrichomonas foetus of functional Trichomonas vaginalis AP65 adhesin

Abstract

Background: Trichomonosis, caused by Trichomonas vaginalis, is the number one, nonviral sexually transmitted infection that has adverse consequences for the health of women and children. The interaction of T. vaginalis with vaginal epithelial cells (VECs), a step preparatory to infection, is mediated in part by the prominent surface protein AP65. The bovine trichomonad, Tritrichomonas foetus, adheres poorly to human VECs. Thus, we established a transfection system for heterologous expression of the T. vaginalis AP65 in T. foetus, as an alternative approach to confirm adhesin function for this virulence factor.

Results: In this study, we show stable transfection and expression of the T. vaginalis ap65 gene in T. foetus from an episomal pBS-ap65-neo plasmid. Expression of the gene and protein was confirmed by RT-PCR and immunoblots, respectively. AP65 in transformed T. foetus bound to host cells. Specific mAbs revealed episomally-expressed AP65 targeted to the parasite surface and hydrogenosome organelles. Importantly, surface-expression of AP65 in T. foetus paralleled increased levels of adherence of transfected bovine trichomonads to human VECs.

Conclusion: The T. vaginalis AP65 adhesin was stably expressed in T. foetus, and the data obtained using this heterologous system strongly supports the role of AP65 as a prominent adhesin for T. vaginalis. In addition, the heterologous expression in T. foetus of a T. vaginalis gene offers an important, new approach for confirming and characterizing virulence factors.

Figure 1

Tritrichomonas foetus has lower levels of adherence to HeLa epithelial cells and immortalized human MS-74 VECs and has no detectable ap65 adhesin gene. (A) Bar graph showing the mean level of adherence by T. foetus (shaded bars) derived from four independent adherence experiments, each of which was performed with quadruplicate samples. Adherence levels of T. vaginalis (open bars) were normalized to 100% for comparative purposes. (B) Northern analysis to detect ap65 mRNA amounts was performed using 10 μg of total RNA per lane separated on 1.2% agarose-formaldehyde gels and transferred to Hybond-P membranes. The blot was probed with DIG-labelled ap65, which hybridized to an ~2-kb ap65 transcript in T. vaginalis, as before [16]. No similar signal was detected in T. foetus and the drug resistant MR100 T. vaginalis used as a negative control [15]. (C) Ethidium bromide (EtBr)-stained RNA gel showing the rRNA bands. Equal loading of RNA in all the lanes is evident by similar amounts of the rRNA bands, which served as internal controls for the Northern blot in part B.

Figure 2

Transfection and episomal expression of ap65 in T. foetus. (A) Agarose gel stained with EtBr shows results of a PCR reaction to amplify the 795-bp neo gene from transfected T. foetus total genomic DNA. The neo gene PCR product is detected only in T. foetus transfected with pBS-ap65-neo (lane 3) and pBS-neo (lane 4), but not the wild type T. foetus (Tf; lane 2) and control T. vaginalis (Tv; lane 1). (B) Agarose gel stained with EtBr shows RT-PCR products for the ap65 transcript in wt T. foetus (lane 1), T. vaginalis (lane 2), T. foetus transfected with pBS-ap65-neo (lane 3), and T. foetus transfected with the pBS-neo plasmid as a negative control. RT-PCR was performed using primers to amplify a 580-bp region of the ap65 gene and a 650-bp region of the a-tubulin gene used as an internal control.

Figure 3

Monoclonal antibody (mAb) detects AP65 expressed in pBS-ap65-neo (Parts A and B) and pBS-ap65-HA-neo (Part C) transfected T. foetus parasites. (A) Total proteins from 10⁷ parasites were separated on 10% SDS-PAGE and blotted onto Hybond-P membranes for probing with mAb DM116 specific for AP65, as before (Garcia et al., 2003). The AP65 protein was readily detected in total protein blots of transfected T. foetus (lane 2) and T. vaginalis (lane 4), but not in wt T. foetus (lane 1) and in T. foetus transfected with the control pBS-neo plasmid (lane 3). The migration of detected AP65 was ~65-kDa as expected based on molecular weight standards. (B) A ligand assay showing AP65 binds to immortalized human MS-74 VECs, as before (Garcia et al., 2003). Parasites bound to VECs were solubilized and electrophoresed as above for blotting onto Hybond-P membranes. The DM116 mAb was used to detect AP65 in transfected T. foetus (lane 2) and T. vaginalis (lane 4). No protein was detected in wt T. foetus (lane 1) and T. foetus transfected with the control plasmid (lane 3).

Figure 4

Immunoblots with specific mAb 12G4 detecting AP65 immunoprecipitated from extracts of total trichomonad protein preparations (A) and of proteins after the ligand assay (B) derived from purified hydrogenosomes (lanes 1 through 3) and membrane fractions (lanes 4 through 6). AP65 was immunoprecipitated with mAb from protein extracts from hydrogenosomes and membrane fractions of T. foetus (Tf; lanes 1 and 4), T. vaginalis (Tv; lanes 2 and 5), and transfected T. foetus (Tf-pBS-ap65-neo; lanes 3 and 6). Immunoprecipitated AP65 was prepared as described in the Experimental design section and used in a ligand assay to monitor the amount of AP65 adhesin bound to MS-74 VECs as shown above in Figure 3.

Figure 5

Immunofluorescence and corresponding brightfield microscopy showing pBS-ap65-neo transfected T. foetus (Tf-pBS-ap65-neo) expressing AP65 on the surface of non-permeabilized trichomonads (A2) and in hydrogenosomes in permeabilized parasites (B2). The distinct patterns were obtained with the mAbs 12G4 for surface fluorescence and F11 for hydrogenosome fluorescence shown recently to detect the protein in respective compartments (Garcia et al., 2003). T. vaginalis fluorescence patterns (Tv; A4 and B4) were positive controls, and results are identical to those obtained recently (Garcia et al., 2003). No evidence of fluorescence was seen for wild type T. foetus (Tf) and for T. foetus transfected with control plasmid (Tf-pBS-neo). Control hybridoma supernatant lacking anti-AP65 mAb was unreactive with trichomonads.

Figure 6

Immunoblot detection of AP65-HA in T. foetus transfected with pBS-ap65-HA-neo, and immunofluorescence microscopy detecting AP65-HA protein on the surface (non-permeabilized) and in hydrogenosomes (permeabilized). (A) Duplicate immunoblots prepared from total proteins as above were probed with 12G4 mAb to AP65 (lanes 1 through 3) and anti-HA mAb (lanes 4 through 6). AP65 was readily detected by 12G4 in blots of T. vaginalis (lane 1) and T. foetus transfected with pBS-ap65-HA-neo (lane 2) but not T. foetus with control plasmid (lane 3). The anti-HA mAb detected the AP65-HA protein only in T. foetus transfected with pBS-ap65-HA-neo (lane 5). No protein was detectable by anti-HA mAb in blots of control T. vaginalis (lane 4) and T. foetus with the pBS-neo plasmid (lane 6). As an additional negative control for all immunoblots, no protein was detected in the absence of primary antibody or using control hybridoma supernatant. (B) Fluorescence is visualized using anti-HA mAb only in the T. foetus organisms transfected with the pBS-ap65-HA-neo plasmid but not the pBS-neo plasmid. Not unexpectedly, no fluorescence was seen using anti-HA mAb with T. foetus and with the use of control hybridoma supernatant lacking mAb to AP65. Finally, no fluorescence with anti-HA mAb was obtained using T. vaginalis organisms under identical experimental conditions.

Figure 7

T. foetus transfected with pBS-ap65-neo (Tf-pBS-ap65-neo) displays enhanced levels of adherence to immortalized human MS-74 VECs compared to T. foetus (Tf) parasites. The percent level of adherence was adjusted with that seen for T. vaginalis (Tv). T. foetus with control plasmid gave levels of adherence similar to T. foetus (data not shown). The enhanced adherence obtained with transfected T. foetus (Tf-pBS-ap65-neo) was inhibited by anti-AP65 IgG (hatched middle bar). The solid gray bar (right) represents inhibition by normal rabbit serum (NRS) control. The results are the average from four different experiments, and each experiment was carried out using quadruplicate samples. The statistical significance of the results is indicated by the asterisk above the bar graph.