An Atypical F-Actin Capping Protein Modulates Cytoskeleton Behaviors Crucial for Trichomonas vaginalis Colonization

  1. Kai-Hsuan Wang
  2. Jing-Yang Chang
  3. Fu-An Li
  4. Kuan-Yi Wu
  5. Shu-Hao Hsu
  6. Yen-Ju Chen
  7. Tse-Ling Chu
  8. Jessica Lin
  9. Hong-Ming Hsu

  • Curated by eLife

    eLife logo

    Evaluation Summary:

    Trichomonas vaginalis is a sexually transmitted protist that colonizes its host by transitioning from flagellar locomotion to an adherent ameboid movement. In this manuscript, Wang and coauthors use a wide range of experimental approaches to investigate the function of a novel actin capping protein in T. vaginalis cytoadherence and cell motility. The work provides an intriguing example of how an unusual capping protein may impact cytoskeletal organization and cell behavior.

    (This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #2 and Reviewer #3 agreed to share their names with the authors.)

Reviewed by

Read the full article

Listed in

Log in to save this article

Abstract

Trichomoniasis is one of the most prevalent nonviral sexually transmitted diseases. T. vaginalis cytoadherence to urogenital epithelium cells is the first step in the colonization of the host. However, studies on the mechanisms of cytoadherence have focused mainly on the role of adhesion molecules, and their effects are limited when analyzed by loss- or gain-of-function assays.

Article activity feed

  1. Version published to 10.1128/spectrum.00596-23
  2. eLife

    Evaluation Summary:

    Trichomonas vaginalis is a sexually transmitted protist that colonizes its host by transitioning from flagellar locomotion to an adherent ameboid movement. In this manuscript, Wang and coauthors use a wide range of experimental approaches to investigate the function of a novel actin capping protein in T. vaginalis cytoadherence and cell motility. The work provides an intriguing example of how an unusual capping protein may impact cytoskeletal organization and cell behavior.

    (This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #2 and Reviewer #3 agreed to share their names with the authors.)

  3. eLife

    Reviewer #1 (Public Review):

    The transition from flagellar motility to ameboid movement enables Trichomonas vaginalis to interact more intimately with the cells of the urogenital tract to colonize a host. This transition is characterized by a profound morphological shift that allows the parasite to adopt a more ameboid type of movement on the colonized epithelium. Over the past several years some of the molecules on the surface of Trichomonas have been characterized, but little is known about the mechanisms of cytoskeletal rearrangement that mediate the transition to cytoadherence and ameboid movement. The present study capitalizes on T. vaginalis isolates that retain a non-adherent (T1) or adherent (TH1) phenotype to identify differences in the actin cytoskeleton that are correlated with cytoadherence, uncovering a new protein (TvFACPα) that appears to regulate the process.

    The authors first establish that the amounts of actin and actinin are correlated with greater cytoadherence. This result is somewhat expected for ameboid movement; nevertheless, the conclusion is supported by treatment with Latrunculin B. Immunoprecipitating T. vaginalis actin, the authors find a non-canonical homolog of the cap-binding protein alpha subunit (TvFACPα). The effects of TvFACPα on cytoadherence, morphogenesis, and wound-closing assays are compelling. In particular, the use of several mutants affecting actin binding (Δ237) and phosphorylation (S2A/D) provides some information about the structure-function relationship of the protein. Based on the known phosphorylation of human CPα by casein kinase 2 (CKII), the authors use an antibody against the known phosphorylation motif of CKII (pS/pTDXE) which partially matches the suspected phosphorylation motif on TvFACPα: pSESE. The S2A mutant is not recognized by the antibody (suggesting that it indeed recognizes the intended PTM) and the signal is diminished by treatment with the CKII inhibitor TBB. Perhaps the most compelling evidence for the effect of TvFACPα phosphorylation is the impact of TBB on ameboid morphogenesis, which is overcome by the expression of the S2D mutant of TvFACPα.

    Several experiments also focus on the biochemical activity of TvFACPα to claim that it functions as a typical capping protein. The evidence presented relies on steady-state experiments and co-immunoprecipitation, which are ill-suited to determine the function of actin-binding proteins. Calling into further question the validity of their biochemical assays, binding assays showed relatively modest differences between G and F actin binding by WT TvFACPα (less than 2-fold), and a similarly modest decrease in binding when the putative actin-binding domain of TvFACPα was removed. The micromolar affinities calculated are also much higher than the typical nanomolar affinities of cap-binding proteins. Other cap-binding proteins occur as heterodimers, so the lack of a β subunit in the IPs calls into question the true identity of the novel factor. While the studies with live parasites do support the importance of TvFACPα in Trichomonas morphogenesis and cytoadherence, more detailed studies will be necessary to define its biochemical function.

  4. eLife

    Reviewer #2 (Public Review):

    Here Wang et al have studied the role of the actin cytoskeleton in Trichomonas vaginalis pathogenicity with a focus on understanding the role of the actin cytoskeleton in transitioning from fast-moving flagellates to adherent ameboid cells. Intriguingly they determined that the adherent TH17 strain contains more actin and the actin-bundling protein alpha-actinin than the less adherent T1 strain. The adherent strain more readily morphs into the adherent ameboid form which has more polymerized F-actin than the flagellated form. Disruption of actin polymerization with LatB prevents ameboid morphogenesis and blocks adhesion. Although LatB treatment interferes with adhesion the authors show that their relatively short LatB treatments do not alter the distribution of adhesive molecules such as AP, PFO, and CLP on the cell surface.

    To understand how actin is being regulated in T. vaginalis the authors pulldown HA-tagged Tvactin and identify interacting partners with mass spectroscopy. They identify TVAG_47023 as a potential protein of interest due to homology with capping protein and name this protein TvFACP. The authors demonstrate that TvFACP can IP actin from cell extracts. They then proceed to purify His-TvFACP and GST-Tvactin to demonstrate direct interaction. Their approach of using bacterial-expressed actin is non-conventional since it is well known that eukaryotic actin requires several chaperones not present in prokaryotes for proper folding. As expected Tvactin was insoluble and was found in inclusion bodies. The authors used urea to solubilize the protein some of which re-folded into soluble protein after buffer exchange. A major concern with this experiment is that the authors did not use any other assay to confirm that their bacterially expressed actin behaved as expected. They should have verified filament formation by negative EM staining or labeled some of the actin for TIRF microscopy assays. Additionally, the 26 kD GST tag has the potential to interfere with actin dynamics which is why most studies remove affinity tags. Alternatively, the authors could have tested binding with commercially available actin. The authors then proceeded to use the purified components to determine the Kd for Tvactin and TvFACP. Intriguingly the authors determine the Kd is lower for G-actin than F-actin indicating that TvFACP preferentially binds to non-filamentous G-actin which is in contrast to canonical capping protein. However, since the quality of this actin is not verified it is not clear that the assay results can be trusted. Despite the preferential association of TvFACP for G-actin in the in vitro assays, localization studies indicate that TvFACP is associated with phalloidin stained structures which indicates that there is an association of TvFACP with filamentous actin structures. Overexpression of TvFACP reduces the ratio of polymerized to unpolymerized actin.

    The authors then explore whether TvFACP might have a role in regulating the transition between flagellated trophozoites and ameboid trophozoites. It was determined that TvFACP has a role in preventing F-actin formation in flagellates. The authors then determined that Ser2 is a phosphosite that regulates the association of TvFACP with actin. An S2A mutant that cannot be phosphorylated associates with actin and prevents ameboid morphogenesis while an S2D mutant does not associate with actin or alter morphogenesis. Since TvFACP S2 is a predicted Casein kinase II (CKII) phosphorylation site, the authors tested the ability of the CKII inhibitor TBB to alter phosphorylation and the association of TvFACP with actin. They found that TBB inhibited phosphorylation and increased the association of TvFACP with actin consistent with the S2A point mutation. Although the result is consistent with the alanine and phosphomimetic mutants, the authors used 250uM of inhibitor which could certainly result in off-target results. As a point of comparison, the IC50 for TBB is reported to be 0.5 uM so here the inhibitor was used at 500x the reported IC50, and at this high level the reduction in phosphorylation may be non-specific for CKII.

    Overall, the results of the manuscript align to support a role for TvFACP in regulating morphogenesis between fast swimming flagellated trophozoites and slow crawling adherent ameboid trophozoites and points toward a potential signaling pathway that regulates this transition. However, in addition to the two technical issues raised above, the relationship between TvFACP and its binding to F- and G- actin remains incompletely resolved. To determine if Tv F-actin capping protein truly binds F-actin the authors should perform TIRF microscopy to determine if TvFACP would be found at the end of filaments and also reveal the extent to which TvFACP alters actin organization and dynamics.

  5. eLife

    Reviewer #3 (Public Review):

    The manuscript by Wang et al. investigates the role of actin and an associated capping protein in cytoadherence and motility of T. vaginalis and represents a substantial amount of work. The authors first demonstrate the adherent lines and clinical isolates express high levels of actin than non-adherent lines, and that a higher percentage of actin is found in the filamentous form in these isolates. FACP was subsequently identified as an actin-binding protein in immunoprecipitation experiments. Overexpression of FACP-WT, but not overexpression of FACP lacking a putative actin-binding domain, resulted in a decreased amount of F-actin in cells, suggesting a role for FACP in limiting actin polymerization by presumably capping the barbed (+) end of filaments. Phosphorylation of FACP at serine 2, mitigates this effect demonstrating that phosphorylation is important for the actin-binding ability of FACP. Phosphorylation also leads to lower adherence to epithelial cells.

    However, a major conclusion of this paper, namely that FACP acts via a novel mechanism and binds both G and F-actin, is not supported by the data. This conclusion is based on experiments with recombinant TvActin expressed in bacteria and co-immunoprecipitation of FACP with actin. The execution of these experiments is problematic for a number of reasons:

    1. The authors state in the methods that the majority of GST-actin is found in inclusion bodies in E. coli. The protein was solubilized in 8M urea, which will denature the protein and the authors then attempted to refold actin by dialysis in G-buffer. F-actin buffer was then added to induce polymerization. The authors provide no evidence that actin folds correctly upon renaturation with G-buffer. It is quite possible that the proteins that pellet upon the addition of the F-buffer are not filaments but insoluble aggregates. I say this because (1) the assay is done at 80 picomoles, which is well below the critical concentration for most actins (typically the Cc is ~0.1-0.5uM range), and (2) the authors provide no evidence by EM or light microscopy to demonstrate that actin filaments are formed under these conditions. Inclusion of these controls in the manuscript is critical to the interpretation of all experiments which utilized the recombinant actin, including the elisa-based assay which is offered as evidence for an interaction with G-actin.

    2. In a number of experiments, the authors performed His-tagged immunoprecipitation of FACP to identify interacting proteins. Actin is found to co-IP with FACP, however, it is not clear if the immunoprecipitated actin represents an interaction with FACP with the F or G isoform. The interpretation of this data is critical for the conclusions of the paper, where the authors argue that FACP has an "atypical" mode of action (title) and the authors' conclusion (line 608) that FACP binds directly to G or F-actin.

  10. Version published to 10.1101/2022.06.09.495492v1 on bioRxiv