https://orcid.org/0000-0002-7579-5106
Figures
Abstract
Trichomonas vaginalis is a prevalent causative agent that causes trichomoniasis leading to uropathogenic inflammation in the host. The crucial role of the actin cytoskeleton in T. vaginalis cytoadherence has been established but the associated signaling has not been fully elucidated. The present study revealed that the T. vaginalis second messenger PIP2 is located in the recurrent flagellum of the less adherent isolate and is more abundant around the cell membrane of the adherent isolates. The T. vaginalis phosphatidylinositol-4-phosphate 5-kinase (TvPI4P5K) with conserved activity phosphorylating PI(4)P to PI(4, 5)P2 was highly expressed in the adherent isolate and partially colocalized with PIP2 on the plasma membrane but with discrete punctate signals in the cytoplasm. Plasma membrane PIP2 degradation by phospholipase C (PLC)-dependent pathway concomitant with increasing intracellular calcium during flagellate-amoeboid morphogenesis. This could be inhibited by Edelfosine or BAPTA simultaneously repressing parasite actin assembly, morphogenesis, and cytoadherence with inhibitory effects similar to the iron-depleted parasite, supporting the significance of PIP2 and iron in T. vaginalis colonization. Intriguingly, iron is required for the optimal expression and cell membrane trafficking of TvPI4P5K for in situ PIP2 production, which was diminished in the iron-depleted parasites. TvPI4P5K-mediated PIP2 signaling may coordinate with iron to modulate T. vaginalis contact-dependent cytolysis to influence host cell viability. These observations provide novel insights into T. vaginalis cytopathogenesis during the host-parasite interaction.
Author summary
Signal transduction in T. vaginalis pathogenesis remains elusive despite elucidation of the distinctive actin-based cytoadherence of the adherent isolates. Herein, we identified a T. vaginalis phosphatidylinositol-4-phosphate 5-kinase (TvPI4P5K) which converts PI(4)P into PI(4, 5)P2. Differential TvPI4P5K expression and specific PIP2 localization in the T. vaginalis isolates with distinct cytoadherence suggest a potential link underlying PIP2 signaling and cytoskeleton behaviors in this parasite. In the parasite transforming from the flagellate into an amoeboid trophozoite, the plasma membrane PIP2 signal may be cleaved by a phospholipase C (PLC)-like pathway, thereafter increasing intracellular calcium, both of which are required for parasite actin assembly, morphogenesis, and cytoadherence. Moreover, environmental iron elicited TvPI4P5K and PIP2 expression and triggered TvPI4P5K translocation to the plasma membrane. PIP2 signaling also modulated the extracellular cytopathic effectors from the parasite to lyse host cells. Our study reveals the biological significance of PIP2 signal transduction in regulating T. vaginalis pathogenesis, which may be a potential therapeutic target.
Citation: Chen Y-J, Wu K-Y, Lin S-F, Huang S-H, Hsu H-C, Hsu H-M (2023) PIP2 regulating calcium signal modulates actin cytoskeleton-dependent cytoadherence and cytolytic capacity in the protozoan parasite Trichomonas vaginalis. PLoS Pathog 19(12): e1011891. https://doi.org/10.1371/journal.ppat.1011891
Editor: Carmen Faso, Universität Bern: Universitat Bern, SWITZERLAND
Received: August 5, 2023; Accepted: December 8, 2023; Published: December 18, 2023
Copyright: © 2023 Chen et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: This study was supported by grants from the National Science and Technology Council (110-2320-B-002-076- and 112-2320-B-002-060-, Grant Recipient: Dr. Hong-Ming Hsu) and National Taiwan University (NTU-JP-112L7226, Grant Recipient: Dr. Hong-Ming Hsu). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Trichomonas vaginalis is a pathogenic protozoan that causes Trichomoniasis, a non-viral sexually transmitted disease prevalent worldwide. This extracellular parasite colonizes the host urogenital tract by adhering to the mucosa, causing uropathogenic inflammation. T. vaginalis interacts with host epithelium cells through numerous surface adhesion molecules [1,2,3,4,5,6], and its surface saccharide moiety is involved in the hemolysis of erythrocytes [7] and phagocytosis of the host cells [8,9]. Trichomonads lyse host cells and phagocytose cell debris for nutrient acquisition, concomitantly damaging the mucosal layer resulting in various cytopathic effects [10,11], including uropathogenic inflammation with worsening symptoms during menstruation. Multiple virulence factors are secreted from or bind to the T. vaginalis cell membrane to modulate parasitic cytoadherence, cytotoxicity, immunoglobulin degradation, and host cell apoptosis [5,12,13,14]. A recent study proposed that T. vaginalis may export extracellular cysteine peptidase through an unconventional lysosomal pathway [15].
Iron is an element essential for growth but an excess is toxic. An abundant element during menstruation, iron multifacetedly regulates T. vaginalis histone modification [16], transcription factor nuclear import [17], morphology [18], the expression of various genes involved in metabolism [16,19], cytoadherence [16,20], and proteolysis [21,22]. The T. vaginalis resistance to complement lysis is attributed to the expression or secretion of multifarious surface proteases with diverse effects on pathogenesis, which are also associated with the environmental iron concentration [21,22,23]. In T. vaginalis signal transduction, iron transiently activates PKA signaling to phosphorylate the Myb3 transcription factor and trigger ubiquitination required for its nuclear translocation and regulation of adhesin expression [17].
The T. vaginalis adherent isolate exhibits remarkable flagellate-amoeboid transition and cytoadherence, which is repressed by actin polymerization inhibitors, to slow down amoeboid migration and reduce cytoadherence, implying the critical roles of the actin cytoskeleton for host colonization. The actin cytoskeleton modulates cell behaviors such as morphological transition, protein trafficking and secretion, membrane adhesion, cell migration, and phagocytosis via diverse pathways between cell types [24,25,26]. Recent studies have demonstrated that the actin-cytoskeleton-mediated amoeboid morphogenesis capacity biologically correlates with cytoadherence and migration crucial for T. vaginalis colonizing host [27,28].
Phosphatidylinositol 4, 5 bisphosphates (PIP2) is a versatile second messenger involved in cytoskeleton organization, vesicle trafficking, and transcription vital to polarized cell growth [29]. Intracellular PIP2 is synthesized by either phosphatidylinositol 4-phosphate 5-kinase (PI4P5K) or phosphatidylinositol 5-phosphate 4-kinase (PI5P4K) using the substrates phosphatidylinositol-4-phosphate [PI(4)P] or phosphatidylinositol-4-phosphate [PI(5)P] respectively [30]. PIP2 on the inner leaflet of the plasma membrane modulates actin cytoskeleton reorganization, including activating assembly accessory protein or branching factors, suppressing disassembly regulator activity, and accelerating the addition of G-actin at the barbed end of growing filaments, to fine-tune cytoskeleton dynamics [31]. By contrast, PIP2 hydrolysis by phospholipase C (PLC) generates inositol 1,4,5-triphosphate (IP3) and diacylglycerol (DAG) as the depolymerization signals. Subsequently, IP3 provokes calcium-dependent signaling to activate severing proteins, such as gelsolins and cofilin, leading to actin meshwork disassembly [32]. The regulation of PIP2 signaling cascades in the peripheral cytoskeleton remodeling differs in various cell types. Nonetheless, PI4P5K on the plasma membrane is essential for yeast cell morphogenesis [33], prompting us to investigate the link between the cytoskeleton and PIP2 signaling in T. vaginalis.
This study identified differential PIP2 and TvPI4P5K expression in adherent and less adherent T. vaginalis isolates and analyzed their functional roles in T. vaginalis cytopathogenicity.
Results
The crosstalk between the complicated regulatory complexes and access signaling around the plasma membrane coordinates cell morphogenesis. Recently, we found that actin cytoskeleton activity is involved in the morphogenesis and cytoadherence of T. vaginalis [27,28], prompting us to investigate the upstream signaling. T1 is a less adherent isolate with only the flagellate form freely swimming by flagellar motoring in suspension. TH17 is an adherent isolate with active flagellate-amoeboid transformation, tightly adhering to the solid surface of culture tubes with locomotion by amoeboid migration [27,28]. The two isolates with different cytoskeleton activities may help link PIP2 signaling with cytoskeleton behaviors in T. vaginalis.
Differential PIP2 expression and localization in T. vaginalis isolates
Phosphatidylinositol (3,4,5)-trisphosphate (PIP3) was undetectable in T. vaginalis, but the PIP2 signal was observed in the cell membrane periphery of the TH17 flagellate trophozoites and the recurrent flagellum of the T1 less adherent isolate (Fig 1A). PIP2 in the cell membrane or recurrent flagellum was also observed in NTU252 and NTU258 adherent isolates or G3 and NTU285 less adherent isolates, respectively (Figs 1B and S1). Overall, PIP2 was more abundant in the adherent isolates than in the less adherent isolates (Fig 1C). In contrast to the long-term cultured experimental T1, G3, and TH17 isolates, the undomesticated short-term cultured NTU252, NTU258, and NTU285 clinical isolates freshly obtained from symptomatic vaginitis patients may preserve the innate parasite nature, suggesting that PIP2 expression and distribution may be associated with parasite cytoadherence or locomotion. According to BLAST in TrichDB (https://trichdb.org/trichdb/app), only TvPI4P5K (TVAG_462290) and TvPI4P5K-2 (TVAG_456620) shared less than 30% sequence identity to human PI4P5K but with a conserved PI4P5K kinase domain (1~ 394 amino acid) and the consensus sequences for binding cell membrane, ATP, PI(4)P, Mg2+, or Mn2+ (S2 Fig) [34,35]. Quantitative PCR (qPCR) showed that TvPI4P5K mRNA was highly expressed in the adherent isolates and positively correlated with PIP2 abundance, whereas TvPI4P5K-2 mRNA expression was equivalent in various isolates (Fig 1D), supporting the TvPI4P5K function in PIP2 production.
(A) The flagellate trophozoites of T1 and TH17 isolates were fixed for IFA detection by anti-PIP2 (upper panel) or PIP3 (lower panel) antibodies. (B) The G3 strain and NTU285, NTU252, and NTU258 clinical isolates were fixed and detected by anti-PIP2 antibody detection for IFA. The arrowheads indicate recurrent flagellum (RF) or plasma membrane (PM). (C) The total lysates were subjected to a dot blot assay for PIP2 and α-tubulin detection. The assay was performed in three biological repeats, and the relative intensity of PIP2 normalized to α-tubulin signal is shown in the bar graph (n = 3, mean ± SD). (D) qPCR was performed to quantify tvpi4p5k and tvpi4p5k-2 gene transcription in various T. vaginalis isolates. The assay was performed in three biological repeats, and the relative gene expression normalized to the β-tubulin signals is shown in the bar graph (n = 3, mean ± SD). For (C) and (D), significant differences were analyzed by Student’s t-tests, with p< 0.05(*) and p< 0.01 (**).
TvPI4P5K enzyme activity
To confirm the kinase activity of TvPI4P5K, recombinant His-TvPI4P5K with a wild-type kinase domain (1–414 aa) and a kinase-deficient K136A mutant were purified for the in vitro kinase assay (Fig 2A), revealing that Wt dose-dependently consumed ATP and was less affected by the K136A or BSA (Fig 2B), demonstrating that TvPI4P5K uses ATP to phosphorylate PI(4)P to PI(4,5)P2. Western blotting revealed that endogenous TvPI4P5K expression in the adherent TH17 isolate was higher than in the less adherent T1 isolate (Fig 2C and 2D), in line with the qPCR results (Fig 1D), supporting the potential correlation of TvPI4P5K and PIP2 expression with T. vaginalis cytoadherence.
(A) The bacterial recombinant His-TvPI4P5K wild-type kinase domain (Wt) and kinase-deficient mutant (K136A) were purified for various assays. (B) TvPI4P5K in vitro kinase activity was evaluated using serially-diluted His-TvPI4P5K, K136A mutant, or BSA control reacted with PI(4)P substrate in the presence of ATP. The relative activities of His-TvPI4P5K Wt and K136A mutant were measured at different concentrations as shown in the plot (n = 3, mean ± SD). (C) The total lysates (TL) from T1 and TH17 isolates were subjected to western blotting for TvPI4P5K and α-tubulin detection. The leftward arrowhead indicates the TvPI4P5K signal. The relative TvPI4P5K signal intensity was quantified as shown in (D). (E) T1 and TH17 flagellate trophozoites were double-stained with anti-PIP2 or anti-TvPI4P5K antibodies for IFA. Signal intensity distribution on the yellow line between x and y sites in the representative micrograph (a, b) was analyzed by ImageJ as shown in the plot (a, b). RF indicates recurrent flagellum and PM indicates the plasma membrane boundary. (F) Plasmids with ha-tvpi4p5k Wt or K136A mutant driven by the Fibronectin-like protein (FLP) promoter and neomycin selection marker driven by the α-tubulin promoter, were constructed to overexpress HA-TvPI4P5K and the derived mutant. (G) The protein lysates from the non-transgenic control or HA-TvPI4P5K Wt and K136A transfectants were detected by western blotting using anti-TvPI4P5K or anti-α-tubulin antibody. (H) The protein lysates from (G) were subjected to a dot blot assay to detect PIP2 and α-tubulin. The assay was processed in three biological repeats, and the relative intensities of PIP2 normalized to α-tubulin were quantified as shown in the bar graph (n = 3, mean ± SD). (I) The parasites without or with HA-TvPI4P5K Wt and K136A overexpression were fixed and double-stained with anti-PIP2 and anti-HA antibodies for IFA. Signal intensity distribution on the yellow line between x and y sites in the representative micrograph (a-c) was analyzed as shown in the plot (a-c). PM indicates the plasma membrane boundary. The assay was processed in three biological repeats, and the relative intensities of the overall PIP2 signal were quantified as shown in the bar graph. (n = 3, mean ± SD). For (H) and (I), significant differences were analyzed by Student’s t-tests, with p< 0.05(*) and p< 0.01 (**).
IFA revealed punctate TvPI4P5K staining in the T1 isolate cytoplasm slightly colocalized with the flagellar PIP2, whereas the TH17 isolate had punctate TvPI4P5K staining in the cytoplasm and plasma membrane, with partial colocalization with PIP2 in the plasma membrane, perhaps related to PIP2 production (Figs 2E and S3A). HA-TvPI4P5K Wt and K136A (Fig 2F) were overexpressed five-fold in TH17 trophozoites (Fig 2G), increasing and depleting PIP2 abundance respectively (Fig 2H), consistent with the IFA signals in the plasma membrane (Figs 2I and S3B), indicating that TvPI4P5K is responsible for the plasma membrane PIP2 production. However, K136A was detected in discrete dots in the cytoplasm, implying that PI4P5K kinase activity may have a role in cell membrane translocation. By density gradient ultracentrifugation on T. vaginalis lysates [36], the TvPI4P5K signal was fractionated to the plasma membrane and various membrane compartments (S3C Fig), supporting the presence of TvPI4P5K in the plasma membrane.
PIP2 hydrolysis in T. vaginalis flagellate-amoeboid transition and cytoadherence
Flagellate-amoeboid transition is critical for T. vaginalis cytoadherence [27,28]. TH17 flagellate trophozoites were cultured on a glass slide to trigger contact-dependent amoeboid morphogenesis (Fig 3A, S1 Movie) and after 30 minutes, ~80% of the TH17 non-transfectant or HA-TvPI4P5K transfectant trophozoites transformed into the amoeboid form but only ~40% of the K136A mutant transformed (Fig 3B). When T. vaginalis trophozoites were co-incubated with hVECs, the cytoadherence activity was similar between the non-transgenic control and HA-TvPI4P5K transfectant but reduced by 40% in the K136A mutant (Fig 3C). There was no change in the TvPI4P5K overall signal intensity but the plasma membrane PIP2 signal was depleted in the trophozoites 30 minutes post morphogenesis (Fig 3D). This was accompanied by reduced TvPI4P5K in the cell membrane (Figs 3D, S4A and S4B) and increased IP3 (S5 Fig), linking the potential plasma membrane PIP2 hydrolysis with parasite morphogenesis. The PLC inhibitor, Edelfosine (ET-18-O-CH3, 1-octadecyl-2-O-methyl-glycero-3-phosphocholine), simultaneously maintained the plasma membrane PIP2 level (Figs 3E and S4B), repressed IP3 production (S5 Fig), and reduced amoeboid morphogenesis (Fig 3E). Similar results were observed in the dot blot of parasites undergoing amoeboid transformation (Fig 3F). The overall PIP2 level reduced with morphogenesis was reversed by Edelfosine treatment. Likewise, the overall PIP2 abundance in the HA-TvPI4P5K transfectant was reduced upon morphogenesis but accumulated in the presence of Edelfosine. However, Edelfosine treatment had no effect in the PIP2-depleted K136A mutant (S6 Fig). These results suggest that T. vaginalis morphogenesis may involve PLC-dependent plasma membrane PIP2 degradation. Meanwhile, Edelfosine treatment did not affect parasite vitality, proving its inhibitory efficacy (S7 Fig).
(A) Culturing the parasite on a glass slide for 30 min at 37°C triggers the flagellate-amoeboid transition of TH17 trophozoites in contact with a solid surface, referred to as contact-dependent morphogenesis. (B) The non-transgenic TH17 control and the transfectants overexpressing HA-TvPI4P5K Wt and K136A mutant were cultured in a T25 flask for 30 min to observe the morphology by phase-contrast microscopy. Black and white arrowheads indicate amoeboid and flagellate trophozoites, respectively. The assay was processed in three biological repeats to measure the proportion of the flagellate versus amoeboid trophozoites, as shown in the bar graph (n = 3, mean ± SD). (C) The CFSE-prelabeled parasites were co-cultured with hVECs for 1 hr and bound parasites were detected by a confocal microscope. The assay was processed in three biological repeats, and the relative ratio of the bound parasite was quantified as shown in the bar graph when the non-transgenic control was defined as 100% (n = 3, mean ± SD). (D) The TH17 free flagellate trophozoites before (0′) and after culture on a glass slide for 30 min (30′) were fixed for IFA double staining with anti-PIP2 and anti-TvPI4P5K antibodies. The fluorescence signal was observed by confocal microscopy, and morphology was recorded in the DIC mode to estimate the amoeboid trophozoite ratio, as shown in the parentheses. (E) The TH17 flagellate trophozoites pretreated with DMSO or Edelfosine before (0′) and after adherence to a glass slide for 30 min (30′) were fixed for IFA double-staining with anti-PIP2 and anti-TvPI4P5K antibodies. For (D.) and (E.), signal intensity distribution on the yellow line between x and y sites in the representative micrograph (a, b of D. and a-d of E.) was analyzed by ImageJ as shown in the corresponding plot (a, b of D. and a-d of E.). PM indicates the plasma membrane boundary. (F) The protein lysates extracted from the trophozoites with or without Edelfosine treatment before (0′) and after culture in a T25 flask for 30 min (30′) were subjected to dot blot to detect PIP2 and α-tubulin. The assay was processed in three biological repeats, and the relative PIP2 signal intensity normalized to α-tubulin is shown in the bar graph (n = 3, mean ± SD). For (B), (C), and (F), the significant differences for the paired samples were analyzed by Student’s t-tests, with p< 0.05(*), p< 0.01(**), ns. no significant difference.
PIP2 signaling regulates T. vaginalis intracellular calcium
To validate whether PIP2 hydrolysis influences the T. vaginalis intracellular calcium, a calcium indicator, Calcium Green-1-AM (CG) was preloaded into the parasite for microscopic fluorescence detection. The CG signal intensity was higher in the TH17 adherent isolate (Fig 4A) and further increased in the PIP2-overexpressed HA-TvPI4P5K transfectant but decreased in the PIP2-depleted K136A mutant (Fig 4B), indicating that T. vaginalis PIP2 level was positively associated with the intracellular calcium content. When the TH17 flagellates were cultured on a glass slide, the CG signal was increased in the trophozoites 30-min post-incubation but reduced by Edelfosine treatment and abolished by BAPTA-AM, a cell-permeant calcium chelator can be used to control intracellular calcium level (Fig 4C, S2 Movie). Edelfosine and BAPTA-AM treatment also suppressed parasite morphogenesis (Fig 4D) and cytoadherence (Fig 4E). Simultaneously, the CG signal increased upon morphogenesis and the proportion of amoeboid form reduced in the presence of EGTA, an extracellular calcium chelator (S2 Movie, S8 Fig), suggesting that extracellular calcium influx may be required for T. vaginalis flagellate-amoeboid transition. In summary, the PIP2 hydrolysis followed by increased intracellular calcium is vital for parasite amoeboid morphogenesis and cytoadherence. Edelfosine, BAPTA-AM, or EGTA treatment did not affect parasite viability (S7 Fig), supporting their inhibitory effects.
(A) The T1 and TH17 trophozoites preloaded with Calcium Green (CG) fluorescence dye were fixed for detection by a confocal microscope. (B) The CG-preloaded TH17 non-transgenic flagellate trophozoites and those overexpressing HA-TvPI4P5K Wt and K136A mutant were fixed for confocal microscopic detection. (C) The TH17 flagellates pretreated with DMSO, Edelfosine, and BAPTA-AM were loaded with CG. The trophozoites, before (0′) and after culture on a glass slide for 30 min (30′), were fixed for confocal microscopy. The assay was processed in three biological repeats, and the relative intensity of the CG signal was quantified in 300 trophozoites from five independent microscopic fields as shown in the bar graph (n = 3, mean ± SD). (D) The TH17 flagellates pretreated with DMSO, Edelfosine, and BAPTA-AM were cultured in a T25 flask for 30 min to record the parasite morphology using a phase-contrast microscope. Black and white arrowheads indicate amoeboid and flagellate trophozoites, respectively. The proportion of flagellate versus amoeboid form was quantified in 300 trophozoites within five independent microscopic fields, as shown in the bar graph. The assay was processed in three biological repeats (n = 3, mean ± SD). (E) CFSE-prelabeled TH17 flagellate trophozoites pretreated with DMSO, Edelfosine, or BAPTA-AM were co-cultured with hVECs for 1 hr. After washing, the bound parasites on hVECs were fixed for CFSE signal detection by confocal microscopy. The assay was processed in three biological repeats, and the relative ratio of bound trophozoites was calculated as shown in the bar graph when the DMSO control was defined as 100% (n = 3, mean ± SD). Significant differences were statistically analyzed by Student’s t-tests, with p< 0.05(*) and p< 0.01(**).
PIP2 signaling modulates actin organization in T. vaginalis
Previously, we demonstrated that actin polymerization is crucial in T. vaginalis flagellate-amoeboid transition [27,28], so we further examined whether PIP2 signaling regulates amoeboid morphogenesis through actin cytoskeleton organization. The transfectants equally overexpressing HA-TvPI4P5K Wt and K136A (Fig 5A) were fractionated into the supernatant (G-actin) and pellet (F-actin) to evaluate actin polymerization. Compared to the non-transgenic control parasite, the F-actin content was unchanged in HA-TvPI4P5K Wt with PIP2 overexpression (Fig 5B) but reduced in the PIP2-depleted K136A mutant (Fig 5C), revealing that PIP2 may be involved in actin polymerization. Moreover, F-actin polymerization was inhibited in the parasites treated with Edelfosine (Fig 5D) or BAPTA-AM (Fig 5E). Conversely, a calcium ionophore, A23187, slightly increased CG signal intensity and the F-actin ratio in the parasite (S9 Fig), suggesting that intracellular calcium triggered from PIP2 hydrolysis is required for actin polymerization.
(A) The total lysates from the non-transgenic TH17 trophozoites and those overexpressing HA-TvPI4P5K Wt and K136A were subjected to western blotting detection by anti-HA anti-TvPI4P5K, and anti-α-tubulin antibodies. The total lysates from the TH17 transfectant overexpressing HA-TvPI4P5K Wt (B), K136A mutant (C), or the non-transgenic TH17 parasites treated with DMSO, Edelfosine (EDFO) (D), and BAPTA-AM (E), were fractionated into G-actin containing supernatant (S) and F-actin containing pellet (P) fractions for western blotting. The assay was processed in three biological repeats. The ratio of α-actin signals in the supernatant versus pellet fraction was quantified, as shown in the bar graphs. (n = 3, mean ± SD). Significant differences were statistically measured by Student’s t-tests, with p< 0.05(*), p< 0.01(**), and ns, no significant difference.
Iron is involved in T. vaginalis PIP2 signaling
Iron was previously found to transiently trigger PKA-dependent signaling to regulate the nuclear translocation of the Myb3 transcription factor in T. vaginalis [17], so we investigated whether iron also regulates PIP2 signaling. TH17 trophozoites were cultured overnight in the normal, iron-depleted, and iron-replete growth media for analysis, revealing that the PIP2 (Fig 6A) and TvPI4P5K (Fig 6B) levels were similar in the normal-iron or iron-replete parasites but higher than in the parasites depleted of iron, implying that the iron in the normal growth medium sufficiently affects PIP2 and TvPI4P5K expression in T. vaginalis. IFA double-staining showed that the plasma membrane PIP2 signal intensity was similar in the normal- and iron-replete parasites and more robust than in the iron-depleted parasites (Fig 6C). Punctate TvPI4P5K staining was observed in the cytoplasm of iron-depleted parasites with a lower level of PIP2, whereas numerous TvPI4P5K dots around the cell membrane were partially colocalized with a higher amount of PIP2 in the plasma membrane in the normal- and iron-replete parasites (Figs 6C, S10A and S10B). Iron may increase TvPI4P5K plasma membrane localization for PIP2 production. Since iron did not affect tvpi4p5k gene transcription (S10C Fig), iron may modulate TvPI4P5K expression translationally or post-translationally.
The TH17 trophozoites were incubated in normal-iron, iron-depleted, and iron-repletion medium overnight for further assays. (A) The protein lysates extracted from the trophozoites were subjected to dot blot assay with anti-PIP2 and α-tubulin detection. The blot assay was processed in three biological repeats, and the relative signal intensity of PIP2 normalized to α-tubulin is shown in the bar graph (n = 3, mean ± SD). (B) The total protein lysates (TL) from TH17 trophozoites under various iron conditions were sampled for western blotting using antibodies as indicated. The assay was processed in three biological repeats, and the relative signal intensity is shown in the bar graph (n = 3, mean ± SD). For (A) and (B), significant differences for the paired samples were analyzed by Student’s t-tests, with p< 0.05(*), p< 0.01(**), ns. no significant difference. (C) The TH17 flagellate trophozoites at different iron conditions were fixed for IFA double staining with anti-PIP2 and anti-TvPI4P5K antibodies. The signal intensity distribution on the yellow line between the x and y sites in the representative micrograph (a-c) was analyzed by ImageJ, as shown in the corresponding plots (a-c). PM indicates the plasma membrane boundary.
Iron affects T. vaginalis actin-centric activities
Iron regulates TvPI4P5K expression and plasma membrane localization, thus, we examined whether iron also affects downstream PIP2 signaling cascades and cytoskeleton behaviors in T vaginalis. In contrast to the normal-iron parasites, the levels of intracellular CG signal (Fig 7A), amoeboid morphogenesis (Fig 7B), cytoadherence (Fig 7C), and F-actin polymerization (Fig 7D) were diminished in the iron-depleted parasites but partially restored to the basal level by A23187 challenge, suggesting the potential role of iron in intracellular calcium and actin-based cytoskeleton behaviors of T. vaginalis. However, A23187 had no significant effect on TvPI4P5K plasma membrane translocation and expression or PIP2 level in the iron-depleted TH17 trophozoites, indicating that A23187 only activates PIP2 downstream signals (S11 Fig) and highlighting the significance of intracellular calcium in T. vaginalis cytoskeleton regulation.
TH17 trophozoites were cultured in normal-iron and iron-depleted medium or treated with A23187. The parasites were sampled for CG detection (A), morphogenesis analysis (B), cytoadherence assay (C), and F-actin fractionation (D). For CG detection, the relative signal intensity averaged from 300 trophozoites within five independent microscopic fields was shown in the bar graph. For the morphogenesis assay, the flagellates were cultured in a T25 flask for 30 min to record the parasite morphology using a phase-contrast microscope. Black and white arrowheads indicate amoeboid and flagellate trophozoite, respectively. The proportion of flagellate versus amoeboid form was quantified in 300 trophozoites within five independent microscopic fields, as shown in the bar graph. For cytoadherence assay, the CFSE-labeled trophozoites were co-cultured with hVECs for 1 hr. After washing to remove unbound trophozoites, the bound trophozoites were fixed for CFSE detection, and the relative cytoadherence capacities between samples were measured as shown in the bar graph. For G/F-actin fractionation, the lysates from the trophozoites were fractionated into G-actin-containing supernatant and F-actin-containing pellet fractions for western blotting. The ratio of α-actin signals in the supernatant (S) versus pellet fractions (P) was quantified as shown in the bar graph. All assays were processed in three biological repeats (n = 3, mean ± SD). The significant differences are statistically measured by Student’s t-test, with p<0.05(*), p<0.01(**), and ns, no significant difference.
PIP2 signaling in T. vaginalis cytopathogenic activity
T. vaginalis could lyse multiple types of human cells with contact-dependent disruption or soluble factors [5,10,11]. To clarify the roles of iron and PIP2 signaling in T. vaginalis cytotoxicity, the parasites precultured with or without iron were inoculated into hVECs at different multiplicity of infection (MOI) for the lactate dehydrogenase (LDH) cytotoxicity assay. Since the parasites started to die after 5 hours of culture in keratinocyte serum-free medium under 5% CO2 (S12A Fig), we monitored the early 1-hour cytotoxicity to rule out the influences from the assay, showing that the overall cytotoxicity increased with increasing MOI. In normal-iron parasites at MOI over 4, the cytotoxicity slightly induced in the PIP2-overexpressed HA-TvPI4P5K Wt was reduced in the K136A mutant with PIP2 depletion and those treated with Edelfosine or BAPTA-AM and co-cultured with the protease inhibitor to a similar extent as the iron-depleted parasites (Fig 8A). Additionally, the cytotoxicity of iron-depleted parasites with A23187 activation was restored to the normal-iron parasite level (Fig 8A). In a 4-hr assay, the parasites caused similar cytotoxic tendencies with extents slightly greater (S12B Fig). However, the LDH activity remained low in the spent medium with the parasites only, ruling out the LDH leakage from the parasites in the 1-hr cytotoxicity assay (S12C Fig). PIC did not directly affect hVECs viability (S12D Fig). Regarding parasite contact-dependent cytolysis, the hVECs lysis area per trophozoite for the normal-iron parasites was slightly changed in the HA-TvPI4P5K Wt transfectant, whereas the areas lysed by the K136A mutant and the parasites pretreated with Edelfosine and BAPTA-AM or co-cultured with protease inhibitors were reduced to that like the iron-depleted parasites. Again, the reduced cytolytic area in the parasites with iron depletion was recovered by A23187 to the normal-iron parasite level, supporting the crucial roles of iron and PIP2 signaling cascades in contact-dependent cytolysis (Fig 8B). Notably, the overall parasite cytotoxicity and cytolysis activities were also suppressed by Latrunculin B (LatB), an F-actin assembly inhibitor [28] (Fig 8A and 8B), revealing that the actin cytoskeleton likely contributes to T. vaginalis cytolytic activity. Together, the PIP2-signaling-cascade-mediated actin cytoskeleton may be involved in the contact-dependent cytotoxic effects, associated with extracellular protease activities. Environmental iron is required for T. vaginalis intact cytotoxicity, but whether PIP2 regulates protease surface expression on or release from T. vaginalis needs further elucidation.
The TH17 trophozoites without (Control) or with HA-TvPI4P5K Wt or K136A overexpression pretreated with DMSO, Edelfosine, BAPTA-AM, or LatB were inoculated into hVECs in the presence or absence of a protease inhibitor cocktail (PIC). (A) The parasites were co-cultured with hVECs at different MOI for 1 hr in the LDH cytotoxicity assay. The assay was performed in three biological repeats to measure the relative cytotoxicity (%) as shown in the bar graph (n = 3, mean ± SD). (B) The trophozoites labeled with CMRA were co-cultured with hVECs (MOI = 4) for 1 hr. All specimens were fixed for staining with FITC-conjugated phalloidin. Nuclei were stained with DAPI. The white dashed line indicates the area lysed by the parasite on the hVECs monolayer. The assay was performed in three biological repeats, and the average lysis area per trophozoite is shown in the Box-whisker plot (n = 5, mean ± SD). The significant differences of grouped samples were analyzed by Student’s t-tests with the p< 0.05(*), p< 0.01(**), and ns, no significant difference.
Conclusion
Specific TvPI4P5K and PIP2 expression was observed in the T. vaginalis adherent isolate. During parasite flagellate-amoeboid transition, cell membrane PIP2 was cleaved by a PLC-dependent pathway, increasing intracellular calcium essential for cytoskeleton activities, including actin remodeling, morphogenesis, and cytoadherence, which could be inhibited by Edelfosine or BAPTA and activated by A23187. When the iron content in the normal growth medium sufficiently elicited PIP2 signaling, iron simultaneously regulated the expression and plasma membrane localization of TvPI4P5K for PIP2 production. In the host-parasite interaction, PIP2 signaling cascades modulated parasite morphogenesis and cytoadherence, contributing to the cytopathic effects by extracellular protease-associated cytolysis in an actin cytoskeleton-dependent manner (Fig 9). The PIP2-triggered extracellular cytopathic effectors warrant continued identification but this study provides an insight into a new dimension of T. vaginalis pathogenicity.
TvPI4P5K phosphorylates PI(4)P into PI(4,5)P2, which is cleaved by PLC hydrolysis. Although DAG has not yet been identified in T. vaginalis, the induced intracellular calcium regulates actin dynamics essential for morphogenesis and cytoadherence. The cytoskeleton activities were also activated and repressed by A23187 and BAPTA-AM, respectively, indicating the role of calcium in cytoskeleton regulation. The details of how environmental iron induces TvPI4P5K expression and trafficking to the plasma membrane for PIP2 production remain unexplored, but the cell membrane PIP2 signaling cascades are involved in host cell cytolysis in a protease-dependent manner. However, whether PIP2, the actin cytoskeleton, and calcium individually or coordinately regulate T. vaginalis cytotoxicity requires further elucidation. (The illustration was created with BioRender.com).
Discussion
Plasma membrane PIP2
Since PIP3 is undetectable in T. vaginalis, PIP2 may be a significant plasma membrane signaling moiety in this parasite. In mammalian cells, PIP2 is localized to lipid rafts, selectively regulating different cellular responses [37]. In Entamoeba histolytica, cholesterol stimulates PIP2 enrichment in uroid lipid rafts at the trailing edge of a motile polarized trophozoite, increasing intracellular calcium concomitant with increased motility [38]. In T-cell signaling, plasma membrane PIP2 recruits the activated ezrin-radixin-moesin (ERM) binding, spatially regulating actin polymerization. The deactivated ERM is released from the plasma membrane into the cytoplasm, thereby changing membrane fluidity and reshaping the membrane structure necessary for adhesion [39]. In the same way, reduced PIP2 in the plasma membrane of T. vaginalis amoeboid trophozoites may facilitate the membrane reshaping required for morphogenesis.
In the higher eukaryotic organism, PIP2 is cleaved through a well-defined PLC hydrolysis into IP3 and DAG [40] to release intracellular calcium [41] and activated PKC [42], respectively. Also, in mammal sperm cells, PKC regulates F-actin formation [43], and IP3 involves intracellular calcium regulation [44]. Our data showed that PLC-dependent PIP2 cleavage might be conserved in T. vaginalis (Fig 3). The downstream intracellular calcium cascade is essential to F-actin assembly and the extended behaviors, crucial for parasite host colonization [27,28]. However, whether the plasma membrane PIP2 can manipulate peripheral actin polymerization directly or indirectly through a calcium second messenger warrants further investigation. The TrichDB database (https://trichdb.org/trichdb/app) [45] contains one potential phospholipase C precursor (TVAG_180400) with low sequence consensus to mouse PLCZ1 (Q8K4D7) and numerous calcium channels, transporters, and binding proteins, but the biological nature of T. vaginalis PLC remains to be characterized. IP3 receptors are unusual among endoplasmic reticulum proteins functionally expressed at the PM, where very few IP3 receptors contribute to calcium entry [46]. The increasing intracellular calcium level in the transforming T. vaginalis could be effectively reduced in the continuous presence of EGTA (S8 Fig), suggesting that T. vaginalis PIP2 signaling modulates extracellular calcium influx. The detail of how PIP2 regulates calcium flow in T. vaginalis is an intriguing question to be addressed. Although the adherent isolates in this study expressed higher TvPI4P5K and PIP2 (Fig 1), the amoeboid morphogenesis and cytoadherence activities were not significantly changed in the less adherent isolates overexpressing TvPI4P5K (S13 Fig), suggesting that PIP2 might be an effector essential but not sufficient for evoking T. vaginalis actin-based capabilities, and the less adherent isolates may have evolved a divergent regulatory mechanism.
Flagellar PIP2 of T. vaginalis
In many postmitotic cells, the PIP2 level determines the primary cilia length, and shortening from ciliary fission is attributed to constitutively increased ciliary PIP2 and F-actin polymerization [47]. Also, the gelsolin-severing F-actin filaments inhibit flagellar motility in guinea pig sperm [48]. Therefore, whether the specified PIP2 in recurrent flagellar of the less adherent isolate regulates actin activities leading to flagellation or the flagellar locomotion, such as motor stirring driving force, are intriguing questions worthy of further investigation. In contrast to some calcium-activated actin effectors severing the actin meshwork in the higher eukaryotic organisms [32], the increased intracellular calcium promoted F-actin assembly and consequent cytoskeleton behaviors in the amoeboid T. vaginalis trophozoite (Figs 5 and S9), indicating a different calcium function in T. vaginalis actin regulation. Intracellular calcium signaling is essential for the apicomplexan parasites in cell motility, invasion, and egress of host cells [49,50]. However, how upstream signaling mediates the calcium concentration remains unknown in these microorganisms.
TvPI4P5K cell membrane trafficking
Previous studies have reported that mouse Arf6 resides on the endosome and plasma membrane to regulate protein trafficking between the compartments [51,52] and PIP5K endosome-plasma membrane trafficking [53]. Punctate TvPI4P5K staining in the cytoplasm and approaching the plasma membrane in the presence of iron (Figs 6C and S10) suggest that TvPI4P5K may be enclosed in or reside on particular membrane-bound vesicles. Meanwhile, multiple Arf-like proteins found in the TrichDB database supported that TvPI4P5K plasma membrane trafficking mediated with iron is likely through a similar protein cargo delivery system [45].
Iron and signal transduction
Regarding the host-parasite interactions, extracellular calcium may influence trichomonad recognition and binding fibronectin [54] and enhance TvCLP-mediated T. vaginalis aggregation on the host cells [4]. Our study further expands the role of calcium to function as an intracellular second messenger contributing to T. vaginalis pathogenesis.
Iron is an essential nutrient for T. vaginalis and is acquired via a highly specific receptor-mediated mechanism from the host [55]. The absence of iron in the culture medium decreases cell growth and triggers morphological changes from the amoeboid to a rounded flagellate form by an unknown mechanism. Iron exerts versatile roles in parasite growth and virulence, including the expression of genes required for metabolism [19] or virulence [21,22]. Also, iron increases T. vaginalis resistance to complement lysis for evading host immunity [23]. Iron may modulate histone acetylation at H3K27 and tri-methylation at H3K4 within the coding region of iron-responsive ap65-1 and pfo genes, epigenetically regulating cytoadherence [16]. Meanwhile, iron promotes cAMP-dependent TvPKAc activation to phosphorylate the Myb3 transcription factor for rapid nuclear import required for regulating ap65-1 gene expression [17]. In bovine sperm cells, the high PKA activity elicits PIP2 production via PI3K-associated PI4P5K activation, resulting in PLD activation and actin assembly [56]. More investigation is required to elucidate the crosstalk of TvPKAc to TvPI4P5K activity and PIP2 production in T. vaginalis.
Although HA-TvPI4P5K overexpression promoted intracellular PIP2 (Fig 2H), it did not significantly activate actin assembly (Fig 5) or consequent morphogenesis and cytoadherence activities beyond the basal level in the control parasites (Fig 3), implying that the restricted plasma membrane PIP2 pool was adequate for T. vaginalis downstream regulation.
Cytotoxicity
T. vaginalis actin-based phagocytosis damaged and phagocytosed host epithelium cells [9,10,11]. Therefore, PIP2 signaling modulated actin dynamics, and contact-dependent cytolysis may be relevant (Fig 8). Moreover, contact-dependent cytolysis may require extracellular proteases to destroy the host cell membrane integrity. Intriguingly, T. vaginalis has been reported to export cysteine protease via a nonconventional lysosomal pathway [15] or express rhomboid protease on its surface [5]. Given that PIP2 regulates endosomal trafficking and vesicle fusion between intracellular membrane compartments [57], PIP2 signaling cascades may mediate parasite lysosome plasma membrane trafficking to an extracellular dump or present protease on the surface to damage host cells [5]. Furthermore, some specified proteins [14,58] or extracellular vesicles [59,60,61] secreted by the parasites may be uptaken by the host cells to modify the physiology during host-parasite interactions. The intracellular PIP2 and calcium elevations involve rat neuroendocrine cell exocytosis [53] and coordinate with the actin cytoskeleton to control vesicle transportation and secretion [62]. As an actin assembly inhibitor, LatB effectively reduces T. vaginalis F-actin dynamics [28], but whether the cytopathic effects are directly regulated by PIP2 or indirectly attributed to the subsequent signal cascades or actin cytoskeleton warrants further validation. This mechanism may provide new drug targets to restrict T. vaginalis pathogenicity at the initial infection.
Materials and methods
Cell culture
T. vaginalis incubated in TYI medium with 10% bovine serum at 37°C was defined as the normal-iron condition unless specified in the text. For iron-depleted or iron-replete culture conditions, the trophozoites were cultured in TYI growth medium supplemented with 50 μM 2, 2′-dipyridyl, or 250 μM ferrous ammonium sulfate, respectively [17]. For drug treatment, the parasite culture was pretreated with 10 μM of Edelfosine (TargetMol), 20 μM of BAPTA-AM (Abcam), 20 μM of A23187 (Abcam), or 1 μM of Latrunculin B (Sigma-Aldrich) in TYI medium on rotation for 90 min at 37°C. After washing with the medium to remove extracellular drugs, the trophozoites were immediately suspended in the medium or buffer for further assay. EGTA (2 mM) was used to chelate extracellular calcium in the medium. The clinical NTU isolate series were obtained from symptomatic vaginitis patients and maintained in the TYI medium supplemented with 1000 U of penicillin, 1000 μg of streptomycin, and 2.5 μg/ml of amphotericin B, with daily passaging over two weeks to remove microbial contaminants. For T. vaginalis cryopreservation, 2× 107 trophozoites of the clinical isolates were suspended in 1 ml of TYI medium with 7.5% DMSO and kept in the freezing container filled with isopropanol and then placed in a -80°C freezer over 24 hr. All isolates were deposited in the biosafety chamber of our institute for research only (National Taiwan University, College of Medicine). In contrast to the T1, G3, and TH17 experimental isolates [27,28], the fresh clinical isolates cultured short-term for weeks rather than years may preserve more intrinsic virulence. Human vagina epithelium cell (hVECs) was cultured in keratinocyte serum-free medium (Gibco) at 37°C with 5% CO2.
Trypan blue exclusion assay
The parasites were stained with 0.4% trypan blue in PBS to evaluate the viability of the parasites with or without drug treatment. The percentage of viable cells was measured in 300 trophozoites within five independent microscopic fields as follows: .
Plasmid construct
The pFLP-HA-TvCyP2 plasmid [36] driven by the promoter (-786 to +11) of Fibronectin-like protein-1 (FLP) [63] was used to overexpress HA-TvPI4P5K in T. vaginalis. The full-length DNA sequence of the tvpi4p5k gene (TVAG_462290) was amplified from the genomic DNA of the TH17 trophozoites using the primer pair TvPI4P5K-BamHI-5’ and TvPI4P5K-XhoI-3’ (Table 1). The PCR product was gel-purified and subcloned into the pFLP-HA-TvCyP2 backbone vector with BamHI and XhoI restriction enzyme sites to generate the pFLP-HA-TvPI4P5K plasmid. The K136A mutation in the pFLP-HA-TvPI4P5K plasmid was generated by a site-directed mutagenesis kit according to the manufacturer’s instructions (Toyobo) using the primer pair TvPI4P5K (K136A)-5’ and TvPI4P5K (K136A)-3’ (Table 1) to produce the pFLP-HA-TvPI4P5K (K136A) plasmid.
To produce the recombinant proteins of TvPI4P5K kinase domain (1 to 414 amino acid) wild type and K136A. DNA fragments amplified from the pFLP-HA-TvPI4P5K or pFLP-HA-TvPI4P5K (K136A) template plasmids using the primer pair TvPI4P5K-BamHI-5 and TvPI4P5K-414-XhoI-3’ (Table 1) were separated in a 1.5% agarose gel and recovered using the gel extraction kit. The purified DNA fragments pre-digested with BamHI/XhoI were ligated into the pET28a expression vector to generate the pET28-His-TvPI4P5K (1–414) and pET28-His-TvPI4P5K (1–414) (K136A) plasmids.
The plasmid DNA was transfected into T. vaginalis by electroporation (BTX D DNA delivery system) to establish stable transfectant clones selected by paromomycin (200 μg/ml).
Recombinant protein production
The plasmid was transformed into Escherichia coli (E. coli, BL21) for recombinant protein production as previously described [17,36]. E. coli culture at OD260 0.6 was induced with 1 mM IPTG and then incubated at 30°C for 3 hrs. The bacteria were recovered by 3000× g centrifugation and washed once with PBS followed by sonication. The bacterial lysate was centrifuged at 15000× g at 4°C for 15 min to remove cell debris and insoluble proteins before purification using a 1-ml Ni-NTA column as suggested by the supplier (Qiagen).
RT-PCR and qPCR
RNA was extracted from T. vaginalis with TRIZOL reagent (Invitrogen) and reversely transcribed by Superscript III transcriptase (Invitrogen) using Oligo (dT) to produce the first-strand cDNA.
For qPCR, the relative expressions of tvpi4p5k and tvpi4p5k-2 genes normalized to β-tubulin were quantified from the first-strand cDNA by QuantStudio 5 Real-Time PCR System (ThermoFisher Scientific) using a qPCRBIO SyGreen Blue Mix Lo-ROX kit (PCRBIOSYSTEMS). The reaction was performed with an initial denaturation at 95°C for 3 min, followed by 40 cycles of 94°C for 3 sec and 40 s at 60°C each. The tvpi4p5k, tvpi4p5k-2, and β-tubulin genes were amplified by the primer pairs of TvPI4P5K-qPCR-5’ /TvPI4P5K-qPCR-3’, TvPI4P5K-2-qPCR-5’/TvPI4P5K-2-qPCR-3’, and Tub-qPCR-5’/Tub-qPCR-3’, respectively (Table 1).
Calcium Green intracellular calcium detection
The parasites were incubated in PBS containing 1% BSA and 5 μM Calcium Green-1-AM (CG, Invitrogen) at 37°C for 20 min. After removing the excess dye by washing with 1 ml PBS once, the parasites were cultured on the glass slide at 37°C for the specified time frame. After washing away the unbound parasites with 37°C pre-warmed PBS, the sample was fixed with 4% formaldehyde, then air-dried and mounted in anti-fade medium with DAPI. The fluorescence signal was observed by confocal microscopy (Zeiss, LSM-780) with an excitation wavelength of 506 nm and an emission wavelength of 531 nm. The signal intensity of 300 trophozoites within five microscopic fields was measured by ImageJ v.1.53k software (National Institutes of Health). Dynamics of the intracellular CG signal intensity in parasites upon amoeboid morphogenesis was recorded by time-lapse fluorescence microscopy with a capturing rate of 40 sec per one frame for 20 min.
Immunofluorescence assay
The parasite trophozoites adhering to the glass slides were fixed with 4% formaldehyde, followed by the permeabilization with 0.5% saponin in TBS. After triple washes with TBS, the samples were incubated in TBS with 10% goat serum at 37°C for 20 min. After three washes with TBS to remove previous reactants, the sample was incubated with primary mouse anti-PIP2 (400×, Echelon Biosciences), mouse anti-PIP3 (200×, Echelon Biosciences), mouse anti-HA (200×, Sigma-Aldrich), rat anti-HA (100×, Roche) and rabbit anti-TvPI4P5K (400×, Lab-made) antibodies diluted in TBS containing 1% BSA, at 4°C overnight. Then, the sample was washed three times with TBS and reacted with FITC or Cy3-conjugated goat anti-mouse IgM or IgG secondary antibodies prepared in TBS containing 1% BSA, at 37°C for 1 hr. The slide was washed three times in TBS and then air-dried at room temperature for 20 min. The sample mounted in an anti-fade medium with DAPI (Vector Labs) was observed by a confocal microscope (Objective: Plan-Apochromat 100/1.40 Oil Ph3, LSM-780, Zeiss) and the fluorescent images were captured in a single Z-slice for figure construction.
Dot blot assay
Approximately 1× 107 trophozoites were vigorously vortexed in 200 μl of lysis buffer (1% Triton X-100, 100 μg/ml TLCK, 1× Protease inhibitor cocktail, 1× phosphatase inhibitor cocktail, 5 mM EDTA in TBS) at 4°C for 20 min to prepare the total protein lysate. The serial-diluted protein lysates were blotted on the nitrocellulose (NC) membrane (ThermoFisher Scientific) by a 96-well microfiltration apparatus (BIO-RAD), then the NC membrane was blocked in blocking buffer (5% nonfat milk in TBS with 0.1% Tween-20) with gentle agitation at 37°C for 1 hr. After triple washes with TBS containing 0.1% Tween-20 (TBST), the membrane was incubated with the primary mouse anti-PIP2 (4000×, Echelon Biosciences) or mouse anti-α-tubulin (Sigma Aldrich, DM-1A) antibodies at 4°C overnight. Then, the membrane was washed with TBST three times, followed by incubation with HRP-conjugated goat anti-mouse IgM or IgG secondary antibodies at 37°C for 1 hr. The signal was detected by an enhanced chemiluminescence substrate (ThermoFisher Scientific) and imaged by a UVP imaging system (Analytik Jena Company).
Western blotting
The protein sample (30μg/lane) denatured in 1× SDS sample buffer was separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) in a 12% gel and blotted onto the polyvinylidene difluoride (PVDF) membrane by a tank transfer system (BIO-RAD). The membrane was incubated in the blocking buffer (5% nonfat milk in TBST) at 37°C for 1 hr and then incubated with primary antibodies, mouse anti-HA (5000×, Sigma-Aldrich), rabbit anti-TvPI4P5K (15000×), mouse anti-α-actin (10000×, Genetex), mouse anti-GAPDH (5000×), mouse anti-α-tubulin (10000×, Sigma-Aldrich), and rat anti-TvCyP2 (3000×) [36] diluted in blocking buffer, at 4°C overnight. After optimal washing by TBST, the membrane was reacted with HRP-conjugated goat anti-mouse, rat, or rabbit IgG secondary antibody (5000×, Jackson ImmunoResearch) prepared in the blocking buffer, at 37°C for 1 hr. The signal was detected by LimiFlash Ultima Chemiluminescent Substrate (VISUAL PROTEIN). To detect endogenous TvPI4P5K expression, 80 μg of protein lysate and a high-sensitive Femto Chemiluminescent substrate (Visual Protein) were used for western blotting detection.
In vitro phospholipid kinase assay for TvPI4P5K
The in vitro PI4P5K activity assay kit (Echelon Biosciences, K-5700) is an ATP depletion assay that quantifies the remaining ATP in the sample following the kinase reaction. The PIP5K activity was measured according to the manufacturer’s instructions (Echelon Biosciences). Briefly, 10 μl of the different concentrations of recombinant His-TvPI4P5K (1–414 amino acid) wild type or K136A protein in 1× KBZ buffer (Echelon Biosciences) was mixed with 10 μl of 4× PI(4)P substrate solution [400 μM PI(4)P] (Echelon Biosciences) before the addition of 20 μl of 2× ATP solution (20 μM ATP) (Echelon Biosciences). The sample was reacted at 37°C for 2 hrs, then 40 μl of ATP detector was added to each well for 20 min at room temperature in the dark, and the luminescent signal was detected by a spectrophotometer at a filter wavelength of 550 nm [35,64].
In vivo G-actin/ F-actin fractionation
A commercial in vivo assay biochem kit (Cytoskeleton Inc) was used to fractionate G- and F-actin, according to the operating manual, with minor modifications. Briefly, approximately 3× 107 trophozoites were incubated in cell lysis buffer (Cytoskeleton Inc) with vigorous agitation at 4°C for 30 min and homogenized using a 23-gauge needle on a 5-ml syringe. Next, the total lysate was centrifuged at 1000× g to remove the cell debris, followed by ultracentrifugation at 100000× g for 1 hr to recover the insoluble F-actin in the pellet from soluble G-actin in the supernatant. In western blotting, α-tubulin and TvCyP2 were detected as purity markers for the supernatant and pellet fractions, respectively. The α-actin signal intensity in the supernatant and pellet fractions were first normalized to TvCyP2 and α-tubulin, respectively, and then the α-actin signal ratio between supernatant and pellet fractions was calculated to evaluate actin assembly [28].
Morphogenesis analysis
The flagellate trophozoites were cultured in a T25 flask at 37°C for 30 min, and the parasite morphology was observed by microscopy at phase-contrast mode (Olympus, CKX31). The flagellate trophozoites had a solid spherical shape and diameter under 10 μm, whereas the amoeboid trophozoites had a diameter over 10 μm and a distinctive irregular appearance or flat round disk form laying on the glass surface. The percentage of flagellate or amoeboid forms was measured in 300 trophozoites within five random microscopic fields. Dynamics of parasite amoeboid morphogenesis were recorded by time-lapse microscopy with a capturing rate of 30 sec per frame.
Cytoadherence assay
The trophozoites were incubated with 5 μM CFSE (ThermoFisher Scientific) in PBS with 1% BSA on a gentle rotation at 37°C for 30 min. After washing three times with PBS to remove the excess dye, the parasites were suspended in keratinocyte serum-free medium (Gibco) for analysis. The CFSE-labeled parasites were co-cultured with hVECs at the multiplicity of infection (MOI) of 2 in a minimal thin layer of keratinocyte serum-free medium for 1 hr in 5% CO2. After washing the unbound trophozoites away with pre-warmed PBS, the CFSE signal was detected by inverted fluorescence microscopy (CFSE Ex/Em = 492/517) (Axiovert 200M, Zeiss). For each assay, the signal intensities quantified by ImageJ v.1.53k software (National Institutes of Health) in five independent microscopic fields were averaged to evaluate the relative parasite cytoadherence. The signal from the control parasites was defined at 100%.
Cytolysis by fluorescence microscopy
The trophozoites were incubated with 5 μM Orange-CMRA (ThermoFisher Scientific) in PBS with 1% BSA on a gentle rotation at 37°C for 30 min. After washing three times with PBS to remove the excess dye, the parasites were suspended in keratinocyte serum-free medium (Gibco) for analysis. The Orange-CMRA-labeled TH17 trophozoites and hVECs were co-cultured at MOI of 4 on a coverslip placed in a culture microplate with keratinocyte serum-free medium. The microplate was centrifuged at 200× g for 5 min to sediment trophozoites for contact with the hVECs monolayer. After removing unbound trophozoites by three washes with medium, samples were incubated in keratinocyte serum-free medium under the atmosphere with 5% CO2 at 37°C for 1 hr. The specimens fixed with 4% formaldehyde, then permeabilized with PBS containing 0.2% Triton X-100, were stained with FITC-conjugated Phalloidin (1000×, Abcam, ab235137) at room temperature for 1 hr. The samples were washed three times with PBS and air-dried for 20 min. The coverslips were mounted in an anti-fade medium with DAPI and inverted on a glass slide for observation by confocal microscopy (FITC Ex/Em = 492/518, Orange-CMRA Ex/Em = 548/576) (LSM700, Zeiss). The clear lytic area was measured by AxioVision software (Rel.4.8, Zeiss). For each assay, the lysis area per trophozoite was averaged from five independent microscopic fields to evaluate parasite cytolysis activity.
Lactate dehydrogenase (LDH) cytotoxicity assay
The host cells incubated in a 96-well microplate at 90% confluence were inoculated with the parasite in keratinocyte serum-free medium in 5% CO2 for 1 or 4 hrs. The spent media collected at different time points post-infection were centrifuged at 1000× g to remove the cells or parasites before analysis. The supernatant was assayed using the LDH cytotoxicity assay kit (Biochain). Briefly, 45 μl of Assay Mixture was added to 100 μl of test sample for 30 min at room temperature, then 50 μl of Stop Solution was added, and the colorimetric signal was detected at OD490 on a spectrophotometer (SpectraMax190, Molecular Devices). For each assay, the samples collected from parasite-free host cells treated with or without Lysis Solution were detected as high- and low-level controls, respectively. The cytotoxicity (%) was measured as follows: .
IP3 detection
Approximately 3× 107 trophozoites suspended or scratched in 500 μl of 4°C PBS were lysed by ultrasonication on ice. The cell lysates were pelleted by 15000× g centrifugation to remove cell debris and insoluble particles, then diluted in Sample Diluent (Abcam) for colorimetric quantification with an IP3 ELISA kit (Abcam) at OD450.
Subcellular fractionation by differential and gradient centrifugation
Organelle fractions were purified from 250 ml of cells for biochemical characterizations by differential and gradient centrifugation as previously described with some modifications [36]. Briefly, the postnuclear lysate was processed by differential centrifugation into crude membrane fractions, P15 and P100, and the soluble S100 fraction. The P100 pellet was re-suspended in 0.5 ml of TBS by sonication, mixed with 0.1 ml of 60% OptiPrep, and layered onto an OptiPrep gradient gel (12~30%), which was formed by a step-wise 2% increase in each layer. Samples were centrifuged at 3.53×105 g at 4°C for 4 hr (Beckman SW60 rotor). The sample was fractionated into 200-μl fractions starting from the top of the gradient.
Supporting information
S1 Fig. Cytoadherence of the clinical isolates.
The CFSE-labeled trophozoites from G3 and NTU252, NTU258, and NTU285 clinical isolates, were co-cultured with hVECs for cytoadherence assay. The CFSE signal was recorded by fluorescence microscopy and quantified by calculating the number of adherent parasites to 1000 hVECs. The assay was processed in three biological repeats (n = 3, mean ± SD). The significant differences for the paired samples were analyzed by Student’s t-tests, with p< 0.05(*), p< 0.01(**), ns. no significant difference.
https://doi.org/10.1371/journal.ppat.1011891.s001
(TIF)
S2 Fig. Protein sequence alignment for TvPI4P5K.
The alignment of human hPI4P5K (Q99755), T. vaginalis TvPI4P5K (TVAG_462290), and TvPI4P5K-2 (TVAG_456620), with identical or similar amino acids highlighted. The putative sites for ATP interaction (*), membrane binding (+), PI4P substrate binding (†), Mn2+ and Mg2+ binding (⬤), and the kinase-activity-essential K136 residue (▼) are indicated at the top of sequences.
https://doi.org/10.1371/journal.ppat.1011891.s002
(TIF)
S3 Fig. The partial colocalization of PIP2 and TvPI4P5K at the plasma membrane.
The IFA images from Fig 2E and 2I were viewed by 2.5D view of ZEN software as shown in (A) and (B), respectively. The intensities in the two-dimensional image were converted into a height map and are represented by the extension in the Z-direction. RF indicates recurrent flagellum and PM indicates plasma membrane. The yellow arrowheads indicate colocalization (yellow) of TvPI4P5K (red) and PIP2 (green) in particular plasma membrane regions. (C) P100 was fractionated by OptiPrep density gradient ultracentrifugation (left panel) and 200-μl aliquots were collected from the top of each gradient for western blotting using antibodies as indicated. Myb1, Myb3IPhmw, and TvGα protein were detected as membrane compartment markers for P100/Myb1, P100/Myb3IPhmw, and the plasma membrane (PM), respectively.
https://doi.org/10.1371/journal.ppat.1011891.s003
(TIF)
S4 Fig. The partial plasma membrane colocalization of PIP2 and TvPI4P5K.
The IFA images from Fig 3D and 3E were viewed by 2.5D view of ZEN software as shown in (A) and (B), respectively. The intensities in a two-dimensional image were converted into a height map and represented by the extension in the Z-direction. The yellow arrowheads indicate colocalization (yellow) of TvPI4P5K (red) and PIP2 (green) at specific plasma membrane regions.
https://doi.org/10.1371/journal.ppat.1011891.s004
(TIF)
S5 Fig. IP3 production upon T. vaginalis morphogenesis.
The cell lysates from trophozoites before (0′) and after amoeboid transition (30′) were detected with a commercial IP3 ELISA kit, and the relative colorimetric signal was analyzed by a spectrophotometer at OD450. The assay was processed in three biological repeats (n = 3, mean ± SD). The significant differences for the paired samples were analyzed by Student’s t-tests, with p< 0.05(*), p< 0.01(**), ns. no significant difference.
https://doi.org/10.1371/journal.ppat.1011891.s005
(TIF)
S6 Fig. Edelfosine inhibits PIP2 reduction in the parasites with amoeboid morphogenesis.
The protein lysates extracted from the non-transgenic and transgenic trophozoites with or without Edelfosine treatment before (0′) and after culture in a T25 flask for 30 min (30′) were subjected to dot blot for PIP2 and α-tubulin detection. The assay was processed in three biological repeats, and the relative PIP2 signal intensity normalized to α-tubulin is shown in the bar graph (n = 3, mean ± SD). The significant differences for the paired conditional samples were analyzed by Student’s t-tests, with p< 0.05(*), p< 0.01(**), ns. no significant difference.
https://doi.org/10.1371/journal.ppat.1011891.s006
(TIF)
S7 Fig. Drug effects on T. vaginalis viability.
The viability of T. vaginalis trophozoites treated with DMSO, Edelfosine, BAPTA-AM, or EGTA was analyzed using the Trypan blue exclusion assay. The assay was processed in three biological repeats (n = 3, mean ± SD). The significant differences for the paired samples were analyzed by Student’s t-tests, with p< 0.05(*), p< 0.01(**), ns. no significant difference.
https://doi.org/10.1371/journal.ppat.1011891.s007
(TIF)
S8 Fig. Extracellular calcium involves PIP2-dependent T. vaginalis morphogenesis.
The CG-preloaded TH17 flagellates were inoculated in the medium with or without EGTA and incubated on a glass slide for 30 min. The parasites before (0′) and after (30′) morphogenesis were fixed for CG detection (A) or morphogenesis assay (B). (A) The relative CG signal intensity was quantified as shown in the bar graph. (B) The percentage of flagellate versus amoeboid trophozoites was measured as shown in the bar graph. Black and white arrowheads mark amoeboid and flagellate trophozoites, respectively. The assays were processed in three biological repeats (n = 3, mean ± SD). Significant differences were statistically measured by Student’s t-tests, with p< 0.05(*), p< 0.01(**), and ns, no significant difference.
https://doi.org/10.1371/journal.ppat.1011891.s008
(TIF)
S9 Fig. A23187 induces T. vaginalis intracellular calcium and actin polymerization.
(A) The TH17 trophozoites treated with DMSO or A23187 were loaded with CG for confocal microscopy. The relative intensity of the CG signal was quantified in 300 trophozoites from five independent microscopic fields as shown in the bar graph (n = 3, mean ± SD). (B) The TH17 trophozoites pretreated with DMSO or A23187 were fractionated into supernatant and pellet fractions for western blotting. The ratio of α-actin signals in the supernatant (S) versus pellet (P) fractions was quantified, as shown in the bar graphs. The assays were processed in three biological repeats (n = 3, mean ± SD). Significant differences were statistically measured by Student’s t-tests with p< 0.05(*), p< 0.01(**), and ns, no significant difference.
https://doi.org/10.1371/journal.ppat.1011891.s009
(TIF)
S10 Fig. Iron triggered TvPI4P5K plasma membrane localization.
(A) The TH17 trophozoites cultured in normal-iron, iron-depleted, and iron-repletion medium overnight were fixed for IFA double staining for TvPI4P5K (red) and PIP2 (green) detection. The nuclei were stained with DAPI. The images boxed are magnified as shown in plots (a-c). Scale bar: 2 μm. White arrowheads indicate the parasite plasma membrane border (PM), and C indicates the cytoplasm. (B) The IFA images from (A) were viewed by 2.5D view of ZEN software. The yellow arrowheads indicate the partially colocalized signals (yellow) of PIP2 (green) and TvPI4P5K (red) in specific plasma membrane areas. (C) Transcription of the tvpi4p5k gene in T. vaginalis from various iron conditions was quantified by qPCR. The relative gene expression normalized to β-tubulin was shown in the bar graph (n = 3, mean ± SD). Significant differences were analyzed by Student’s t-tests, with p< 0.05(*) and p< 0.01 (**).
https://doi.org/10.1371/journal.ppat.1011891.s010
(TIF)
S11 Fig. A23187 does not affect TvPI4P5K-dependent PIP2 production.
The iron-depleted TH17 trophozoites with or without the A23187 challenge were fixed for IFA double staining with anti-TvPI4P5K and anti-PIP2 antibodies. Nuclei were stained with DAPI. Scale bar: 2 μm. The signal intensity distribution on the yellow line between x and y sites in the representative micrograph (a, b) was analyzed by ImageJ as shown in the corresponding plots (a, b). PM indicates the plasma membrane boundary.
https://doi.org/10.1371/journal.ppat.1011891.s011
(TIF)
S12 Fig. Culture conditions for parasites or hVECs viabilities.
(A) The viability of TH17 trophozoites in a microplate with keratinocyte serum-free medium at 37°C in 5% CO2 was evaluated at different time points by the trypan blue exclusion assay. (B) The hVECs were co-cultured with the parasites with conditions as in Fig 8A at different MOI for 4 hr. The spent medium supernatant was collected for LDH cytotoxicity assay. (C) The parasites were pretreated as in Fig 8A. Different amounts of trophozoites (equivalent to the trophozoites number used in Fig 8A) were inoculated in keratinocyte serum-free medium for 1 hr at 37°C under 5% CO2. The spent medium supernatant was collected for LDH cytotoxicity assay. (D) The spent media from the parasite-free hVECs culture treated with DMSO or PIC for 1 or 4 hr were collected for LDH cytotoxicity assay. The assays were processed in three biological repeats (n = 3, mean ± SD), and significant differences were analyzed by Student’s t-tests, with p< 0.05(*) and p< 0.01 (**).
https://doi.org/10.1371/journal.ppat.1011891.s012
(TIF)
S13 Fig. Effects of HA-TvPI4P5K overexpression in the less adherent isolate.
T1 trophozoites with or without HA-TvPI4P5K overexpression were sampled for IFA double staining with anti-HA or anti-PIP2 (A), morphogenesis assay (B), and cytoadherence assay (C). For (B), white arrowheads label flagellate trophozoites. The assay was processed in three biological repeats to measure the proportion of the flagellate versus amoeboid trophozoites, as shown in the bar graph (n = 3, mean ± SD). For (C), the CFSE-prelabeled parasites were co-cultured with hVECs for 1 hr. After washing, the bound parasites were detected by a confocal microscope. The assay was processed in three biological repeats, and the relative ratio of the bound parasite was quantified with the non-transgenic control defined as 1 (n = 3, mean ± SD). The significant differences were analyzed by Student’s t-tests, with p< 0.05(*), p< 0.01(**), ns. no significant difference.
https://doi.org/10.1371/journal.ppat.1011891.s013
(TIF)
S1 Movie. Dynamics of T. vaginalis amoeboid morphogenesis.
The TH17 flagellate trophozoites were cultured in a glass slide at 37°C. Dynamics of flagellate-amoeboid transition was recorded by time-lapse microscopy with a capturing rate of 30 sec per one frame for 30 min.
https://doi.org/10.1371/journal.ppat.1011891.s014
(MP4)
S2 Movie. Dynamics of CG signal in the parasite with amoeboid transformation.
The DMSO- or Edelfosine-pretreated TH17 flagellate trophozoites were loaded with CG and inoculated in the medium with or without EGTA. The conditional samples were cultured on a glass slide at 37°C. Dynamics of CG signal intensity in the transforming trophozoites were recorded by time-lapse fluorescence microscopy with a capturing rate of 40 sec per one frame for 20 min. Scale bar: 10 μm.
https://doi.org/10.1371/journal.ppat.1011891.s015
(MP4)
S1 Data. Raw Blots.
This file includes raw image data of Figs 1C, 2A, 2C, 2G, 2H, 3F, 5A, 5B, 5C, 5D, 5E, 6A, 6B, 7D, S3C, S6 and S9B.
https://doi.org/10.1371/journal.ppat.1011891.s016
(PDF)
S2 Data. Statistical Analysis.
This file includes raw data of statistical analysis in Figs 1C, 1D, 2B, 2D, 2E, 2H, 2I, 3B, 3C, 3D, 3E, 3F, 4A, 4B, 4C, 4D, 4E, 5B, 5C, 5D, 5E, 6A, 6B, 6C, 7A, 7B, 7C, 7D, 8A, 8B, S1, S5, S6, S7, S8A, S9A, S9B, S10C, S11, S12A, S12B, S12C, S12D, S13B and S13C.
https://doi.org/10.1371/journal.ppat.1011891.s017
(XLSX)
Acknowledgments
We are grateful to Dr. Jung-Hsiang Tai (Institute of Biomedical Sciences, Academia Sinica, Taiwan) for providing the T. vaginalis T1 isolate. We thank the imaging core at the First Core Labs, National Taiwan University College of Medicine, for the technical support in image acquisition and analysis.
References
- 1. de Miguel N, Lustig G, Twu O, Chattopadhyay A, Wohlschlegel JA, Johnson PJ. Proteome analysis of the surface of Trichomonas vaginalis reveals novel proteins and strain-dependent differential expression. Mol Cell Proteomics. 2010 Jul;9(7):1554–66. pmid:20467041
- 2. Pachano T, Nievas YR, Lizarraga A, Johnson PJ, Strobl-Mazzulla PH, de Miguel N. Epigenetics regulates transcription and pathogenesis in the parasite Trichomonas vaginalis. Cell Microbiol. 2017 Jun;19(6). pmid:28054438
- 3. Handrich MR, Garg SG, Sommerville EW, Hirt RP, Gould SB. Characterization of the BspA and Pmp protein family of trichomonads. Parasit Vectors. 2019 Aug 19;12(1):406. pmid:31426868
- 4. Chen YP, Riestra AM, Rai AK, Johnson PJ. A Novel Cadherin-like Protein Mediates Adherence to and Killing of Host Cells by the Parasite Trichomonas vaginalis. mBio. 2019 May 14;10(3). pii: e00720–19. pmid:31088924
- 5. Riestra AM, Gandhi S, Sweredoski MJ, Moradian A, Hess S, Urban S, Johnson PJ. A Trichomonas vaginalis Rhomboid Protease and Its Substrate Modulate Parasite Attachment and Cytolysis of Host Cells. PLoS Pathog. 2015 Dec 18;11(12):e1005294. pmid:26684303
- 6. Hirt RP, Noel CJ, Sicheritz-Ponten T, Tachezy J, Fiori PL. Trichomonas vaginalis surface proteins: a view from the genome. Trends Parasitol. 2007 Nov;23(11):540–7. pmid:17962075
- 7. De Carli G, Brasseur P, Rott M, da Silva A, Wendorff A. Hemolytic activity of various strains and clones of Trichomonas vaginalis and Tritrichomonas suis. Rev Latinoam Microbiol. 1994 Jul-Sep;36(3):207–11. Spanish. pmid:7709097.
- 8. Benchimol M, de Souza W. Carbohydrate involvement in the association of a prokaryotic cell with Trichomonas vaginalis and Tritrichomonas foetus. Parasitol Res. 1995;81(6):459–64. pmid:7567902.
- 9. Pereira-Neves A, Benchimol M. Phagocytosis by Trichomonas vaginalis: new insights. Biol Cell. 2007 Feb;99(2):87–101. pmid:17029588.
- 10. Midlej V, Benchimol M. Trichomonas vaginalis kills and eats-evidence for phagocytic activity as a cytopathic effect. Parasitology. 2010 Jan;137(1):65–76. Epub 2009 Sep 2. pmid:19723359.
- 11. Lustig G, Ryan CM, Secor WE, Johnson PJ. Trichomonas vaginalis contact-dependent cytolysis of epithelial cells. Infect Immun. 2013 May;81(5):1411–9. Epub 2013 Feb 19. pmid:23429535; PMCID: PMC3648012.
- 12. Hernández HM, Marcet R, Sarracent J. Biological roles of cysteine proteinases in the pathogenesis of Trichomonas vaginalis. Parasite. 2014;21:54. Epub 2014 Oct 28. pmid:25348828; PMCID: PMC4209856.
- 13. Štáfková J, Rada P, Meloni D, Žárský V, Smutná T, Zimmann N, et al. Dynamic secretome of Trichomonas vaginalis: Case study of β-amylases. Mol Cell Proteomics. 2018 Feb;17(2):304–320. Epub 2017 Dec 12. pmid:29233912; PMCID: PMC5795393.
- 14. Singh M, Beri D, Nageshan RK, Chavaan L, Gadara D, Poojary M, et al. A secreted Heat shock protein 90 of Trichomonas vaginalis. PLoS Negl Trop Dis. 2018 May 16;12(5):e0006493. pmid:29768419; PMCID: PMC5973626.
- 15. Zimmann N, Rada P, Žárský V, Smutná T, Záhonová K, Dacks J, et al. Proteomic Analysis of Trichomonas vaginalis Phagolysosome, Lysosomal Targeting, and Unconventional Secretion of Cysteine Peptidases. Mol Cell Proteomics. 2022 Jan;21(1):100174. Epub 2021 Nov 8. pmid:34763061; PMCID: PMC8717582.
- 16. Song MJ, Kim M, Choi Y, Yi MH, Kim J, Park SJ, et al. Epigenome mapping highlights chromatin-mediated gene regulation in the protozoan parasite Trichomonas vaginalis. Sci Rep. 2017 Mar 27;7:45365. pmid:28345651; PMCID: PMC5366954.
- 17. Hsu HM, Lee Y, Hsu PH, Liu HW, Chu CH, Chou YW et al. Signal transduction triggered by iron to induce the nuclear importation of a Myb3 transcription factor in the parasitic protozoan Trichomonas vaginalis. J Biol Chem. 2014 Oct 17;289(42):29334–49. Epub 2014 Sep 2. pmid:25183012; PMCID: PMC4200283.
- 18. Dias-Lopes G, Saboia-Vahia L, Margotti ET, Fernandes NS, Castro CLF, Oliveira FO Junior, et al. Morphologic study of the effect of iron on pseudocyst formation in Trichomonas vaginalis and its interaction with human epithelial cells. Mem Inst Oswaldo Cruz. 2017 Oct;112(10):664–673. pmid:28953994; PMCID: PMC5607515.
- 19. Beltrán NC, Horváthová L, Jedelský PL, Sedinová M, Rada P, Marcinčiková M, et al. Iron-induced changes in the proteome of Trichomonas vaginalis hydrogenosomes. PLoS One. 2013 May 31;8(5):e65148. pmid:23741475; PMCID: PMC3669245.
- 20. Moreno-Brito V, Yáñez-Gómez C, Meza-Cervantez P, Avila-González L, Rodríguez MA, Ortega-López J, et al. A Trichomonas vaginalis 120 kDa protein with identity to hydrogenosome pyruvate:ferredoxin oxidoreductase is a surface adhesin induced by iron. Cell Microbiol. 2005 Feb;7(2):245–58. pmid:15659068.
- 21. Arroyo R, Cárdenas-Guerra RE, Figueroa-Angulo EE, Puente-Rivera J, Zamudio-Prieto O, Ortega-López J. Trichomonas vaginalis Cysteine Proteinases: Iron Response in Gene Expression and Proteolytic Activity. Biomed Res Int. 2015;2015:946787. Epub 2015 May 18. pmid:26090464; PMCID: PMC4450334.
- 22. Rivera-Rivas LA, Lorenzo-Benito S, Sánchez-Rodríguez DB, Miranda-Ozuna JF, Euceda-Padilla EA, Ortega-López J, Chávez-Munguía B, Lagunes-Guillén A, Velázquez-Valassi B, Jasso-Villazul L, Arroyo R. The effect of iron on Trichomonas vaginalis TvCP2: a cysteine proteinase found in vaginal secretions of trichomoniasis patients. Parasitology. 2020 Jun;147(7):760–774. Epub 2020 Mar 16. pmid:32174285.
- 23. Alderete JF, Provenzano D, Lehker MW. Iron mediates Trichomonas vaginalis resistance to complement lysis. Microb Pathog. 1995 Aug;19(2):93–103. pmid:8577239.
- 24. Tang DD, Gerlach BD. The roles and regulation of the actin cytoskeleton, intermediate filaments and microtubules in smooth muscle cell migration. Respir Res. 2017 Apr 8;18(1):54. pmid:28390425; PMCID: PMC5385055.
- 25. Parsons JT, Horwitz AR, Schwartz MA. Cell adhesion: integrating cytoskeletal dynamics and cellular tension. Nat Rev Mol Cell Biol. 2010 Sep;11(9):633–43. pmid:20729930; PMCID: PMC2992881.
- 26. Svitkina T. The Actin Cytoskeleton and Actin-Based Motility. Cold Spring Harb Perspect Biol. 2018 Jan 2;10(1):a018267. pmid:29295889; PMCID: PMC5749151.
- 27. Hsu HM, Yang YY, Huang YH, Chu CH, Tu TJ, Wu YT, et al. Distinct features of the host-parasite interactions between nonadherent and adherent Trichomonas vaginalis isolates. PLoS Negl Trop Dis. 2023 Jan 3;17(1):e0011016. pmid:36595499; PMCID: PMC9810166.
- 28. Wang KH, Chang JY, Li FA, Wu KY, Hsu SH, Chen YJ, et al. An Atypical F-Actin Capping Protein Modulates Cytoskeleton Behaviors Crucial for Trichomonas vaginalis Colonization. Microbiol Spectr. 2023 Jun 13:e0059623. pmid:37310229.
- 29. Czech MP. PIP2 and PIP3. Cell. 2000 Mar 17; 100: 603–606. pmid:10761925
- 30. Shulga YV, Anderson RA, Topham MK, Epand RM. Phosphatidylinositol-4-phosphate 5-kinase isoforms exhibit acyl chain selectivity for both substrate and lipid activator. J Biol Chem. 2012 Oct 19;287(43):35953–63. Epub 2012 Sep 1. pmid:22942276; PMCID: PMC3476263.
- 31. Janmey PA, Bucki R, Radhakrishnan R. Regulation of actin assembly by PI(4,5)P2 and other inositol phospholipids: An update on possible mechanisms. Biochem Biophys Res Commun. 2018 Nov 25;506(2):307–314. Epub 2018 Aug 13. pmid:30139519; PMCID: PMC6269227.
- 32. Logan MR, Mandato CA. Regulation of the actin cytoskeleton by PIP2 in cytokinesis. Biol Cell. 2006 Jun;98(6):377–88. pmid:16704377.
- 33. Homma K, Terui S, Minemura M, Qadota H, Anraku Y, Kanaho Y, et al. Phosphatidylinositol-4-phosphate 5-kinase localized on the plasma membrane is essential for yeast cell morphogenesis. J Biol Chem. 1998 Jun 19;273(25):15779–86. pmid:9624177.
- 34. Rao VD, Misra S, Boronenkov IV, Anderson RA, Hurley JH. Structure of type IIbeta phosphatidylinositol phosphate kinase: a protein kinase fold flattened for interfacial phosphorylation. Cell. 1998;94(6):829–839. pmid:9753329
- 35. Ishihara H, Shibasaki Y, Kizuki N, Wada T, Yazaki Y, Asano T, et al. Type I phosphatidylinositol-4-phosphate 5-kinases. Cloning of the third isoform and deletion/substitution analysis of members of this novel lipid kinase family. J Biol Chem. 1998 Apr 10;273(15):8741–8. pmid:9535851.
- 36. Hsu HM, Huang YH, Aryal S, Liu HW, Chen C, Chen SH, et al. Endomembrane Protein Trafficking Regulated by a TvCyP2 Cyclophilin in the Protozoan Parasite, Trichomonas vaginalis. Sci Rep. 2020 Jan 27;10(1):1275. pmid:31988345; PMCID: PMC6985235.
- 37. Van Rheenen J, Achame EM, Janssen H, Calafat J, Jalink K. PIP2 signaling in lipid domains: a critical re-evaluation. EMBO J. 2005 May 4;24(9):1664–73. Epub 2005 Apr 21. pmid:15861130; PMCID: PMC1142585.
- 38. Koushik AB, Powell RR, Temesvari LA. Localization of phosphatidylinositol 4,5-bisphosphate to lipid rafts and uroids in the human protozoan parasite Entamoeba histolytica. Infect Immun. 2013 Jun;81(6):2145–55. Epub 2013 Apr 1. pmid:23545298; PMCID: PMC3676028.
- 39. Johnson CM, Rodgers W. Spatial Segregation of Phosphatidylinositol 4,5-Bisphosphate (PIP(2)) Signaling in Immune Cell Functions. Immunol Endocr Metab Agents Med Chem. 2008 Dec 1;8(4):349–357. pmid:19956793; PMCID: PMC2771939.
- 40. Berridge MJ. Inositol trisphosphate and diacylglycerol: two interacting second messengers. Annu Rev Biochem. 1987;56:159–93. pmid:3304132.
- 41. Berridge MJ, Irvine RF. Inositol trisphosphate, a novel second messenger in cellular signal transduction. Nature 1984; 312:315–321. pmid:6095092
- 42. Nishizuka Y. The role of protein kinase C in cell surface signal transduction and tumour promotion. Nature. 1984 Apr 19–25;308(5961):693–8. pmid:6232463.
- 43. Cohen G, Rubinstein S, Gur Y, Breitbart H. Crosstalk between protein kinase A and C regulates phospholipase D and F-actin formation during sperm capacitation. Dev Biol. 2004 Mar 1;267(1):230–41. pmid:14975729.
- 44. Walensky LD, Snyder SH. Inositol 1,4,5-trisphosphate receptors selectively localized to the acrosomes of mammalian sperm. J Cell Biol. 1995 Aug;130(4):857–69. pmid:7642703; PMCID: PMC2199962.
- 45. Carlton JM, Hirt RP, Silva JC, Delcher AL, Schatz M, Zhao Q, et al. Draft genome sequence of the sexually transmitted pathogen Trichomonas vaginalis. Science. 2007 Jan 12;315(5809):207–12. pmid:17218520; PMCID: PMC2080659.
- 46. Dellis O, Dedos SG, Tovey SC, Taufiq-Ur-Rahman, Dubel SJ, Taylor CW. Ca2+ entry through plasma membrane IP3 receptors. Science. 2006 Jul 14;313(5784):229–33. pmid:16840702.
- 47. Stilling S, Kalliakoudas T, Benninghoven-Frey H, Inoue T, Falkenburger BH. PIP2 determines length and stability of primary cilia by balancing membrane turnovers. Commun Biol. 2022 Jan 25;5(1):93. pmid:35079141; PMCID: PMC8789910.
- 48. Azamar Y, Uribe S, Mújica A. F-actin involvement in guinea pig sperm motility. Mol Reprod Dev. 2007 Mar;74(3):312–20. pmid:16998842.
- 49. Moreno SN, Ayong L, Pace DA. Calcium storage and function in apicomplexan parasites. Essays Biochem. 2011;51:97–110. pmid:22023444; PMCID: PMC3488345.
- 50. Scarpelli PH, Pecenin MF, Garcia CRS. Intracellular Ca2+ Signaling in Protozoan Parasites: An Overview with a Focus on Mitochondria. Int J Mol Sci. 2021 Jan 5;22(1):469. pmid:33466510; PMCID: PMC7796463.
- 51. D’Souza-Schorey C, Li G, Colombo MI, Stahl PD. A regulatory role for ARF6 in receptor-mediated endocytosis. Science. 1995 Feb 24;267(5201):1175–8. pmid:7855600.
- 52. Peters PJ, Hsu VW, Ooi CE, Finazzi D, Teal SB, Oorschot V, Donaldson JG, Klausner RD. Overexpression of wild-type and mutant ARF1 and ARF6: distinct perturbations of nonoverlapping membrane compartments. J Cell Biol. 1995 Mar;128(6):1003–17. pmid:7896867; PMCID: PMC2120412.
- 53. Aikawa Y, Martin TF. ARF6 regulates a plasma membrane pool of phosphatidylinositol(4,5)bisphosphate required for regulated exocytosis. J Cell Biol. 2003 Aug 18;162(4):647–59. pmid:12925709; PMCID: PMC2173784.
- 54. Crouch ML, Benchimol M, Alderete JF. Binding of fibronectin by Trichomonas vaginalis is influenced by iron and calcium. Microb Pathog. 2001 Sep;31(3):131–44. pmid:11500098.
- 55. Lehker MW, Alderete JF. Iron regulates growth of Trichomonas vaginalis and the expression of immunogenic trichomonad proteins. Mol Microbiol. 1992 Jan;6(1):123–32. pmid:1310792.
- 56. Etkovitz N, Rubinstein S, Daniel L, Breitbart H. Role of PI3-kinase and PI4-kinase in actin polymerization during bovine sperm capacitation. Biol Reprod. 2007 Aug;77(2):263–73. Epub 2007 May 9. pmid:17494916.
- 57. Tan X, Thapa N, Choi S, Anderson RA. Emerging roles of PtdIns(4,5)P2—beyond the plasma membrane. J Cell Sci. 2015 Nov 15;128(22):4047–56. pmid:26574506; PMCID: PMC4712784.
- 58. Twu O, Dessí D, Vu A, Mercer F, Stevens GC, de Miguel N, et al. Trichomonas vaginalis homolog of macrophage migration inhibitory factor induces prostate cell growth, invasiveness, and inflammatory responses. Proc Natl Acad Sci U S A. 2014 Jun 3;111(22):8179–84. Epub 2014 May 19. pmid:24843155; PMCID: PMC4050605.
- 59. Nievas YR, Coceres VM, Midlej V, de Souza W, Benchimol M, Pereira-Neves A et al. Membrane-shed vesicles from the parasite Trichomonas vaginalis: characterization and their association with cell interaction. Cell Mol Life Sci. 2018 Jun;75(12):2211–2226. Epub 2017 Dec 8. pmid:29222644.
- 60. Twu O, de Miguel N, Lustig G, Stevens GC, Vashisht AA, Wohlschlegel JA et al. Trichomonas vaginalis exosomes deliver cargo to host cells and mediate host∶parasite interactions. PLoS Pathog. 2013;9(7):e1003482. Epub 2013 Jul 11. pmid:23853596; PMCID: PMC3708881.
- 61. Rai AK, Johnson PJ. Trichomonas vaginalis extracellular vesicles are internalized by host cells using proteoglycans and caveolin-dependent endocytosis. Proc Natl Acad Sci U S A. 2019 Oct 22;116(43):21354–21360. Epub 2019 Oct 10. pmid:31601738; PMCID: PMC6815132.
- 62. Zhang R, Xu Y, Yi R, Shen J, Huang S. Actin Cytoskeleton in the Control of Vesicle Transport, Cytoplasmic Organization and Pollen Tube Tip Growth. Plant Physiol. 2023 Apr 1:kiad203. pmid:37002825.
- 63. Crouch MV, Alderete JF. Trichomonas vaginalis has two fibronectin-like iron-regulated genes. Arch Med Res. 2001 Mar-Apr;32(2):102–7. pmid:11343805.
- 64. Liu L, Zhang S, Wang Y, Bao W, Zhou Y, Dang W et al. BIG1 controls macrophage pro-inflammatory responses through ARF3-mediated PI(4,5)P2 synthesis. Cell Death Dis. 2020 May 15;11(5):374. pmid:32415087; PMCID: PMC7229175.