Trichomonas Vaginalis Induces Apoptosis via ROS and ER Stress Through ER-mitochondria Crosstalk in Human Cervical Epithelial Cells

Fei Fei Gao Chungnam National University Juan-Hua Quan A liated Hospital of Guangdong Medical University Min A Lee Chungnam National University Wei Ye A liated Hospital of Guangdong Medical University Jae-Min Yuk Chungnam National University Guang-Ho Cha Chungnam National University In-Wook Choi Chungnam National University Young-Ha Lee (  yhalee@cnu.ac.kr ) Chungnam National University School of Medicine


Experimental designs
SiHa cells were seeded on 96-well plates (for LDH and MTS assay), 12-well coverslips (for immuno uorescence and ROS detection), and 100-mm culture dishes (for western blotting) at various densities and grown to con uence at 37°C in 5% CO 2 .
Live T. vaginalis trophozoites were incubated in SiHa cells using mixed-medium (DMEM/TYM = 2:1) at MOIs of 2 and 5 for 2 and 6 h or T. vaginalis MOI 5 for 6 h at 37°C in 5% CO 2 . And then, the cytotoxicity, apoptosis, ROS production, MMP, induction of ER stress and its mechanism of action, and Bcl-2 family-related protein expressions were evaluated in T. vaginalis-infected SiHa cells by MTS assay, LDH assay, ow cytometry, immunocytochemistry and western blotting.
To determine the role of ROS in T. vaginalis-infected cells, cells were pretreated with the ROS scavenger NAC, and then evaluated for apoptotic features, MMP and ER stress response. To assess the involvement of ER stress in apoptosis induction in T. vaginalis-infected SiHa cells, cells were pre-treated with ER stress inhibitor 4-PBA, NAC or the JNK inhibitor SP600125, and then the apoptotic features, ROS production, MMP, Bcl-2 family proteins and ASK1/JNK pathways were evaluated. Untreated SiHa cells were used as controls. Each experiment was performed at least three times in triplicate.

LDH Assay
The LDH assay was performed for cytotoxicity quanti cation with the CytoTox 96 ® Non-Radioactive Cytotoxicity Assay kit according to the manufacturer's protocol. Brie y, 1×10 4 cells were seeded in 96-well plates and infected with T. vaginalis at various MOI (2,5) for the indicated times (0, 2, 6 h) in an incubator (5% CO 2 , 90% relative humidity, 37°C). Then, 50 µL aliquots from all infected and control wells were transferred into a new 96-well plate, and 50 µL of CytoTox 96 ® reagent was added to each sample aliquot. The plate was covered with foil to protect the samples from light and incubated for 30 min at room temperature. After adding 50 µL of stop solution to each well, the absorbance of the solution was measured immediately at 490 nm using a microplate reader (TECAN, Crailsheim, Germany). LDH release levels in the media were quanti ed and compared to control values according to the kit instructions.

Measurement of ROS Generation by Confocal microscope
SiHa cells (1 × 10 5 cells/well; 12-well plate) seeded on a coverslip were cultured in DMEM supplemented with 10% FBS and the culture medium was replaced when the cells reached 80% con uence. To evaluate the generation of ROS, SiHa cells were infected with live T. vaginalis (MOI 2 and 5) for 2 h, and 6h, and then incubated with 5 M MitoSOX reagent or CellROX reagent and then incubated for 10 min at 37°C, 5% CO 2 in the dark. The stained cells were imaged using an Olympus BX51 uorescence microscope. All experiments were performed on triplicate samples and uorescence intensity was calculated using ImageJ software, and graph was plotted using SigmaPlot 12.5 (Systat Software, San Jose, CA).

ROS-based Flow cytometric Assay
SiHa Cells were cultured in 12-well plates, and infected with T. vaginalis at MOI 2 and 5 for 0, 2 and 6 h with the indicated concentrations of NAC (0, 0.2, 1, 5 mM) pre-treatment or not. Followed by washing with PBS, the cells were removed from the well with 2.5% trypsin-EDTA and re-suspended in PBS with 5 µM MitoSOX reagent or 5 µM CellROX reagent and incubated at 37℃, protected from light, for 30 min. After gently three times washing, the cells were re-suspended in FACS buffer (1% BSA in PBS) and immediately subjected to the acquisition of 10,000 events with FACScan device (BD Biosciences, San Diego, CA, USA). The results are expressed as a histogram emitting the corresponding uorescence, and as a bar graph representing the mean uorescence intensity of all the groups.

Mitochondrial Membrane Potential (MMP) Assay
The MMP was measured using the JC-1 MitoMP detection kit (Dojindo, Kumamoto, Japan). Brie y, SiHa cells were seeded onto coverslips in 12-well plates at a density of 1 × 10 4 cells/well and infected with T. vaginalis at various conditions with or without speci c inhibitors. The cells were then incubated with 4 µM JC-1 uorescence dye at 37°C for 30 min in dark and rinsed three times with HBSS. The stained cells were mounted onto microscope slides in VECTASHIELD HardSet Mounting Medium with DAPI (Vector Laboratories), and images were collected using a laser confocal microscope (Leica, TCS SP8, Wetzlar, Germany). The intensities of green (excitation/emission wavelength= 485/538 nm) and red (excitation/emission wavelength= 485/590 nm) uorescence were analyzed for ≥ 6 microscopic elds in each sample.

Immunocytochemistry
SiHa cells were seeded onto coverslips in 12-well plates at a density of 1 × 10 4 cells/well and incubated for 24 h. The cells were pretreated with or without speci c inhibitors for 2 h and then infected with T. vaginalis at MOI5 for 6 h. The cells were washed with HBSS and xed with freshly prepared 4% paraformaldehyde for 1 h at room temperature. After washing three times with PBS containing 0.3% Triton X-100 (PBS-T) for 5 min, the cells were blocked with 1% BSA in 0.3% PBS-T for 30 min at room temperature and then incubated with CHOP primary antibody for 2 h at room temperature. The cells were washed to remove the excess primary antibody and then incubated with the appropriate uorescently labeled secondary antibody (anti-mouse Alexa Fluor 647) for 2 h at room temperature. After mounting with VECTASHIELD HardSet antifade mounting medium with DAPI (Vector Laboratories, Burlingame, CA, USA) to stain the nucleus, uorescence images were acquired using a Leica confocal software.
Western blotting analysis SDS-PAGE and western blotting analysis were performed to determine numerous proteins expression. SiHa cells were cultured in 100-mm dishes and underwent serum deprivation for 4 h to remove stimulation from serum factors. Then cells were infected by T. vaginalis as indicated. After washing with PBS, proteins were extracted using the PRO-PREP Protein Extraction Solution (iNtRON Biotechnology, Korea). The extract was incubated with the complete protease inhibitor cocktail (Roche, Basel, Switzerland) for 30 min on ice followed by boiling for 10 min, and then centrifuged at 14,000 g for 15 min at 4°C. The supernatant was collected, and equal amounts of protein from each sample were separated by SDS-PAGE and transferred to a polyvinylidene di uoride (PVDF) membrane (Bio-Rad Laboratories, Hercules, CA, USA). The membranes were immersed for blocking at 5% skim milk in Tris-buffered saline (20 mM Tris, 137 mM NaCl, pH 7.6) containing 0.1% Tween-20 (TBST) for 1 h at room temperature. After washing once in TBST, membranes were incubated overnight at 4°C with the primary antibodies diluted in TBST supplemented with 5% BSA. Following three consecutive washes with TBST, membranes were incubated for 90 min with HRP-conjugated anti-mouse or anti-rabbit secondary antibody diluted 1:2000 with 5% skim milk, as described above. The washed membranes were soaked with Immobilon Western Chemiluminescent HRP Substrate (Jackson ImmunoResearch Laboratories), and chemiluminescence was detected with a Fusion Solo System (Vilber Lourmat, Collegien, France). Band intensity was quanti ed using ImageJ software (NIH, Bethesda,MD, USA). These experiments were repeated at least three times.

Statistical analysis
All assays were performed in triplicate, and at least three independent experiments were conducted per test series. Results are presented as the means ± standard deviation (SD). Statistical analysis of the data was performed using unpaired, two-tailed Student's t-tests. A P-value less than 0.05 indicated statistical signi cance.

T. vaginalis induced cytotoxicity and apoptosis in SiHa cells in a parasite burden-and infection time-dependent manner
To assess the cytotoxicity of live T. vaginalis trophozoites in cervical epithelial SiHa cells, the cells were infected with T. vaginalis at different multiplicity of infection (MOI) and infection time before conducting the LDH assay. The cytotoxicity of SiHa cells infected with T. vaginalis at MOI 2 was 13.5 ± 1.1%, 26.9 ± 2.7%, and 46.9 ± 1.7% at 0, 2, and 6 h post-infection, respectively. Further, we observed that the toxicity of T. vaginalis in the SiHa cells was parasite burden-dependent, as it signi cantly increased to 48.9 ± 2.8% and 83.2 ± 5.0% at 2 and 6 h post-infection of T. vaginalis (MOI 5), respectively (Fig. 1A).
To determine whether this cytotoxicity was associated with apoptosis, we evaluated the apoptosis indicator proteins in the T. vaginalis-infected SiHa cells. The expression levels of cleaved PARP and caspase 3, as observed by western blotting, signi cantly increased from 2 h after T. vaginalis infection at MOI 2 in a parasite burden-and infection time-dependent manner (Fig. 1B).
Together, these results suggested that T. vaginalis induces cytotoxicity and apoptosis in SiHa cells in a parasite burden-and infection time-dependent manner.

T. vaginalis led to the generation of mitochondrial ROS in SiHa cells in parasite burden-and infection time-dependent manner
Oxidative stress is an important factor for cytotoxicity associated with apoptosis [17]. To investigate ROS production in T. vaginalis-infected SiHa cells, cellular and mitochondrial ROS levels were measured following infection. Cellular ROS levels signi cantly increased in the T. vaginalis-infected SiHa cells (MOI 2) from 2 h after infection and increased further at 6 h post-infection. Cellular ROS levels were signi cantly higher in the cells infected with T. vaginalis at MOI 5 than in cells infected with T. vaginalis at MOI 2 ( Fig. 2A).
We further investigated mitochondrial ROS production using the MitoSOX Red staining technique and found that ROS production signi cantly increased from 2 h after T. vaginalis infection in direct proportion to parasite burden and infection time in the SiHa cells, which was similar to that in case of cellular ROS production (Fig. 2B).
These ndings suggested that T. vaginalis induces mitochondrial ROS production in SiHa cells in a parasite burden-and infection time-dependent manner.

T. vaginalis induced mitochondrial apoptosis through ROS in SiHa cells
To determine whether ROS are involved in the induction of apoptosis in T. vaginalis-infected SiHa cells, the cells were pretreated with the ROS scavenger N-Acetyl cysteine (NAC) and the expression of apoptosis-related proteins and MMP was evaluated. We rst determined the effects of NAC on cell viability using the MTS assay. Treatment of the SiHa cells with 0.2-1 mM NAC for 6 h elicited no signi cant differences in cell viability compared with those in the medium-treated group; however, the SiHa cells treated with 5 mM NAC for 6 h showed slightly reduced viability (see Additional le 1: Fig.  S1A). Both cellular and mitochondrial ROS production were signi cantly suppressed in the T. vaginalisinfected SiHa cells following NAC pretreatment in a concentration-dependent manner (see Additional le 2: Fig. S2A, B).
As shown in Fig. 3A, in the NAC-pretreated SiHa cells, T. vaginalis-induced cleavage of PARP and caspase-3 was signi cantly suppressed in a dose-dependent manner. The permeabilization of the mitochondrial outer membrane is a crucial step in the progression of apoptosis [6,7]; thus, we also investigated whether the T. vaginalis-induced ROS-dependent apoptosis in SiHa cells was related to the mitochondrial apoptotic pathways. JC-1 staining, an indicator of MMP. T. vaginalis infection, revealed a notable decrease in MMP levels, as indicated by the reduction in red uorescence signal and an increase in green JC-1 uorescence signal, in a parasite burden-and infection time-dependent manner compared with the signals in untreated control cells (Fig. 3B). However, these changes in JC-1 dye uorescence were suppressed by NAC pretreatment in a dose-dependent manner (Fig. 3C).
These results suggested that T. vaginalis induces mitochondrial apoptosis through ROS, and ROS act upstream of PARP and caspase-3 during apoptosis induction.
T. vaginalis induced ER stress response and ER stressmediated apoptosis in SiHa cells.
ER stress is known to contribute to apoptosis [18,19,21]. First, to check whether T. vaginalis causes ER stress in SiHa cells, we investigated the expression of ER stress-related proteins in T. vaginalis-infected SiHa cells. As shown in Fig. 4A, Ero1-Lα, p-IRE1α, CHOP, p-PERK, p-elF2α, ATF4, and cleaved ATF6 protein levels were signi cantly increased in the T. vaginalis-infected SiHa cells in a parasite burden-and infection time-dependent manner, whereas procaspase-12 levels were decreased as compared with those in the untreated control group (Fig. 4A).
Next, to con rm whether apoptosis is induced by ER stress in T. vaginalis-infected SiHa cells, the cells were pretreated with the ER stress inhibitor 4-phenylbutyric acid (4-PBA), and the expression of ER stressand apoptosis-related proteins was evaluated. The SiHa cells treated with 5 mM 4-PBA for 6 h did not show a signi cant difference in viability compared with that of the untreated control group (see Additional le 1: Fig. S1B). The expression of ER stress-related proteins was reversed in the 4-PBApretreated T. vaginalis-infected SiHa cells in a dose-dependent manner as compared with that in the untreated T. vaginalis-infected cells (Fig. 4B). In addition, 4-PBA pretreatment attenuated the levels of cleaved PARP and caspase-3 in T. vaginalis-infected SiHa cells in a dose-dependent manner (Fig. 4C). These results demonstrated that T. vaginalis induces ER stress response as well as ER stress-mediated apoptosis in SiHa cells.

T. vaginalis induced ROS-dependent ER stress responses in SiHa cells.
ER and mitochondria are closely related, both structurally and functionally [18,19,22,23]. Thus, to evaluate the effects of mitochondrial ROS in the induction of ER stress response in T. vaginalis-infected SiHa cells, the SiHa cells were pretreated with NAC, and the expression of ER stress-related proteins was investigated. Pretreatment with NAC signi cantly reduced the expression of Ero1-Lα, p-IRE1α, CHOP, p-PERK, p-elF2α, ATF4, and cleaved ATF6 in the T. vaginalis-infected SiHa cells in a dose-dependent manner, whereas the expression level of procaspase-12 was increased in a dose-dependent manner (Fig. 5A). Confocal microscopy imaging also con rmed that NAC pretreatment reduced T. vaginalisinduced upregulation of CHOP protein expression in a dose-dependent manner (Fig. 5B).
These results indicated that T. vaginalis induces ER stress response through mitochondrial ROS in SiHa cells, which leads to apoptosis.

T. vaginalis induced ER stress-mediated mitochondrial dysfunction in SiHa cells.
We also investigated whether ER stress affects mitochondrial functions in T. vaginalis-infected SiHa cells. Members of the Bcl-2 family are major regulators of mitochondrial integrity and mitochondria-dependent caspase activation, and MMP is the key indicator of mitochondrial apoptosis [6,7]. Therefore, we investigated mitochondrial functions, including alternation of MMP and Bcl-2 family members, in T. vaginalis-infected SiHa cells. JC-1 dye staining revealed that T. vaginalis induced MMP depolarization in the SiHa cells; however, these changes in uorescence were suppressed by 4-PBA pretreatment (Fig. 6A).
The expression of p-Bad (Ser112), Bax, Bak, Bik, and Puma was signi cantly increased in the T. vaginalisinfected SiHa cells (Fig. 6B), whereas the Bcl-xL and Mcl-1 levels were decreased in a parasite burdendependent manner (Fig. 6C). The expression levels of p-Bad (Ser 112), Bax, Bak, Bik, and Puma were signi cantly decreased in the 4-PBA-pretreated T. vaginalis-infected SiHa cells in a dose-dependent manner (Fig. 6D), whereas the levels of Bcl-xL and Mcl-1 proteins were increased (Fig. 6E). These results clearly indicated that T. vaginalis induces mitochondrial dysfunction via ER stress in SiHa cells, leading to apoptosis.
Previous studies have reported that the IRE1/ASK1/JNK signaling cascade is one of the main downstream targets for the regulation of ER stress-induced cell apoptosis [18,19,[21][22][23]. Therefore, we investigated whether T. vaginalis activates ER stress-dependent apoptosis through the ASK1/JNK pathway in SiHa cells. We found a substantial increase in the phosphorylation levels of both ASK1 (Ser83, Ser967, and Thr845) and JNK in the T. vaginalis-infected SiHa cells in a parasite burden-and infection time-dependent manner (Fig. 7A). Next, we investigated whether ASK1/JNK activation was related to ER stress in the T. vaginalis-infected SiHa cells. Western blotting analysis revealed that 4-PBA pretreatment attenuated the T. vaginalis-induced elevation of ASK1 and JNK phosphorylation levels in a dose-dependent manner (Fig. 7B). These data indicated that the T. vaginalis-induced ER stress activates the ASK1/JNK pathway in SiHa cells.
To evaluate whether the ASK1/JNK pathway is associated with mitochondrial apoptosis in T. vaginalisinfected SiHa cells, the cells were preincubated with the JNK inhibitor SP600125, and expression of MMP, Bcl-2 family proteins, and apoptosis-related proteins was investigated. The SiHa cells treated with 30 µM SP600125 for 6 h did not display a signi cant difference in their viability as compared with those of the untreated control group (see Additional le 1: Fig. S1C). Western blotting analysis revealed that Bak, Bax, cleaved PARP, and caspase-3 levels decreased in a dose-dependent manner in the SP600125-pretreated T. vaginalis-infected SiHa cells (Fig. 7C). In contrast, Bcl-xL and Mcl-1 levels were increased by pretreatment with SP600125 in the T. vaginalis-infected SiHa cells in a dose-dependent manner (Fig. 7D). In addition, SP600125 pretreatment substantially increased MMP levels in the T. vaginalis-infected SiHa cells, as demonstrated by increased red and reduced green JC-1 uorescence, in a dose-dependent manner (Fig. 7E). These observations provided evidence that T. vaginalis induces ER stress-mediated mitochondrial apoptosis via the IRE1/ASK1/JNK/Bcl-2 family protein signaling pathways in SiHa cells.

Discussion
Trichomoniasis is a highly prevalent STI; however, its pathophysiology remains unclear. In this study, we evaluated the role of ER stress in apoptosis induction in T. vaginalis-infected human cervical epithelial SiHa cells. We observed that T. vaginalis induces apoptosis through mitochondrial ROS production associated with mitochondrial dysfunction and ER stress response in a parasite burden-and infection time-dependent manner. Also T. vaginalis induces ROS-dependent ER stress response and ER stressmediated mitochondrial apoptosis in SiHa cells. To the best of our knowledge, this is the rst study on ER stress involvement in apoptosis induction of T. vaginalis-infected mammalian cells via the ROS-ER stress-mitochondrial apoptosis pathways.
ROS generation is one of the most critical factors for apoptosis induction in T. vaginalis-infected mammalian cells [8,9]. The present ndings showed that T. vaginalis induces mitochondrial ROS production in SiHa cells in a parasite burden-and infection time-dependent manner, and also induces apoptosis and ER stress response. Thus, we investigated the effects of ROS on ER stress induction in T. vaginalis-infected SiHa cells. Perturbations in the physiological status of the ER trigger a speci c response known as the ER stress response or unfolded protein response (UPR), which induces premature apoptosis [18,19,21]. ROS induce ER stress-mediated apoptosis during some viral and bacterial infections [24][25][26]. Previous studies have shown that ER stress induces apoptosis of cells infected with Mycobacterium [24], Toxoplasma [28], Leishmania [29], etc. In mammalian cells, ER stress signaling is typically sensed by three major ER-resident transmembrane molecules-IRE1, PERK, and ATF6 [18,21,27]. In this study, T. vaginalis infection upregulated p-IRE1α, p-PERK, ATF4, cleaved ATF6, and CHOP levels and downregulated the levels of procaspase-4 and − 12 in SiHa cells in a parasite burden-and infection time-dependent manner. However, 4-PBA and NAC pretreatment signi cantly reversed the expression of ER stress-related proteins in T. vaginalis-infected SiHa cells in a dose-dependent manner. Our results indicated that T. vaginalis induces a direct ER stress response as well as via an ROS-dependent ER stress response in SiHa cells. Furthermore, 4-PBA or NAC pretreatment apparently suppressed the apoptosis and mitochondrial ROS production in T. vaginalis-infected SiHa cells. These results indicated that T. vaginalis induces ER stress-mediated apoptosis in SiHa cells besides ROS-dependent apoptosis; thus, ER stress is responsible for apoptosis induction in T. vaginalis-infected SiHa cells. ER stress-mediated apoptosis has also been observed in T. gondii-infected neural cells [28] and Leishmania infantum-infected macrophages [29].
ER and mitochondria communicate with each other via calcium and ROS [18,19,22]. Mitochondria are important in the regulation and transmission of apoptotic signals, which are regulated by a balance between the Bcl-2-family proteins [7,22]. In our study, T. vaginalis induced MMP depolarization, which led to an increase in Bak and Bax levels and a decrease in Bcl-xL and Mcl-1 levels; this caused mitochondrial dysfunction, which could have stimulated the release of cytochrome c into the cytosol of the T. vaginalisinfected cells [12][13][14]. However, pretreatment with NAC or 4-PBA effectively reversed MMP depolarization and the expression pattern of Bcl-2 family proteins in a dose-dependent manner. These data suggested that T. vaginalis induces mitochondrial apoptosis in SiHa cells through ROS as well as T. vaginalisinduced ER stress. These ndings are in accordance with those reported in previous studies, wherein M. avium induced apoptosis through ROS-dependent ER stress response in macrophages [24] and silibinin, a biologically active compound of milk thistle, induced mitochondrial ROS-dependent apoptosis through ER stress response and disruption of Ca 2+ homeostasis in prostate cell lines [30].
The apoptotic crosstalk between the ER and the mitochondria is regulated by Bcl-2 family proteins, and ER stress regulates the expression of the Bcl-2 family proteins, ultimately leading to the mitochondriamediated apoptotic caspase cascade [18,19,22,23]. During severe and sustained ER stress conditions, the intrinsic apoptotic pathway can be activated by IRE1-mediated activation of tumor necrosis factor receptor-associated factor 2 (TRAF2), which stimulates the ASK1/JNK kinase cascade [21,25,27]. Thus, we investigated the involvement of the IRE1/ASK1/JNK pathways during ER stress-induced apoptosis and found that T. vaginalis induced ASK1 and JNK phosphorylation in SiHa cells. However, pretreatment with 4-PBA or SP600125 attenuated the expression levels of IRE1α, cleaved caspase-3 and PARP, p-ASK1, and p-JNK in the T. vaginalis-infected SiHa cells. In addition, we investigated the involvement of Bcl-2 family members, which act downstream of the IRE1/ASK1/JNK pathways, during ER stress-induced apoptosis. Pretreatment with 4-PBA or SP600125 increased the expression of MMP, Bcl-xl, and Mcl-1, whereas attenuated the expression of Bax and Bak in the T. vaginalis-infected SiHa cells in a dosedependent manner. These results indicated that ER stress and Bcl-2 family proteins communicate with each other during apoptosis induction, and T. vaginalis induces apoptosis via ER stress-mediated mitochondrial dysfunction in SiHa cells through the IRE1/ASK1/JNK apoptotic pathways. These ndings could be explained by previous reports suggesting that Bax and Bak localize to both the mitochondria and ER and their over-expression promotes apoptotic calcium mobilization during ER-mediated mitochondrial cell death [31,32]. A similar apoptotic mechanism was reported in silica nanoparticletreated endothelial cells, wherein apoptosis was induced through the ER stress-associated IRE1/JNK, CHOP, and caspase-12 down to the mitochondria via Bcl-2 family members [33].

Conclusions
Our results showed that T. vaginalis induced apoptosis through mitochondrial ROS generation and ER stress response as well as induced direct and ROS-dependent ER stress response in SiHa cells. T. vaginalis also induced ER stress-mediated mitochondrial dysfunction and nally led to apoptosis of SiHa cells through the IRE1/ASK1/JNK/Bcl-2 family protein pathways (Fig. 8). Thus, T. vaginalis infection induces apoptosis via ROS-ER stress-mitochondrial apoptosis pathways by interacting with the mitochondria and ER in human cervical epithelial SiHa cells. These ndings may contribute to expand our understanding of the molecular mechanisms underlying the pathogenesis of human trichomoniasis.