Cell Cycle Arrest and Apoptosis Induced by Porphyromonas gingivalis Require Jun N-Terminal Protein Kinase- and p53-Mediated p38 Activation in Human Trophoblasts

ABSTRACT

Porphyromonas gingivalis, a periodontal pathogen, has been implicated as a causative agent of preterm delivery of low-birth-weight infants. We previously reported that P. gingivalis activated cellular DNA damage signaling pathways and ERK1/2 that lead to G₁ arrest and apoptosis in extravillous trophoblast cells (HTR-8 cells) derived from the human placenta. In the present study, we further examined alternative signaling pathways mediating cellular damage caused by P. gingivalis. P. gingivalis infection of HTR-8 cells induced phosphorylation of p38 and Jun N-terminal protein kinase (JNK), while their inhibitors diminished both G₁ arrest and apoptosis. In addition, heat shock protein 27 (HSP27) was phosphorylated through both p38 and JNK, and knockdown of HSP27 with small interfering RNA (siRNA) prevented both G₁ arrest and apoptosis. Furthermore, regulation of G₁ arrest and apoptosis was associated with p21 expression. HTR-8 cells infected with P. gingivalis exhibited upregulation of p21, which was regulated by p53 and HSP27. These results suggest that P. gingivalis induces G₁ arrest and apoptosis via novel molecular pathways that involve p38 and JNK with its downstream effectors in human trophoblasts.

INTRODUCTION

Preterm birth is defined as delivery at less than 37 weeks of gestation and is associated with increased mortality and morbidity (1). Microbial infections likely contribute to preterm delivery of low-birth-weight (PTLBW) infants, particularly in cases of very early preterm delivery (2). Bacterial vaginosis and chorioamnionitis are known to lead to spontaneous preterm birth (3, 4), and intrauterine, lower genital tract, and systemic maternal infections, along with asymptomatic bacteriuria are also linked to PTLBW (2). Microorganisms can invade the amniotic cavity by ascending from the vagina and cervix, as well by hematogenous dissemination through the placenta, causing tissue damage and expulsion of the infected fetus (1).

Various studies have provided evidence for an association of periodontitis and major periodontal pathogens, such as Porphyromonas gingivalis, with PTLBW (5–9). P. gingivalis has been detected in amniotic fluid from women identified as at risk for premature labor (10) as well as in placentas of those with preeclampsia (11). Moreover, P. gingivalis antigens have been detected in placental syncytiotrophoblast, chorionic trophoblast, decidual, and amniotic epithelial cells, as well as vascular cells obtained from women who underwent preterm labor complicated by chorioamnionitis at less than 37 weeks of gestation (12). Some in vivo experimental findings have also suggested that P. gingivalis plays a significant role in pregnancy complications. For example, pregnant rats intravenously infected with P. gingivalis manifested bacterial invasion of the placenta, amniotic fluid, and fetus, along with chorioamnionitis and placentitis (8). Moreover, P. gingivalis was translocated to placental tissues following hematogenous spread, resulting in increased rates of both preterm birth and fetal growth restriction in pregnant mice and rabbits (5, 6). We previously reported that P. gingivalis can invade extravillous trophoblasts (HTR-8 cells) and induced G₁ arrest and apoptosis through ERK1/2 and DNA damage response pathways (9, 13). In addition, P. gingivalis can induce phosphorylation and activation of MEK3 and p38 mitogen-activated protein kinase (MAPK) and can also modulate interleukin 1β (IL-1β) and IL-8 production in HTR-8 cells (14).

Cell cycle arrest and apoptosis are known to be triggered by DNA damage (15), after which DNA double- and single-strand breaks induce activation of ataxia telangiectasia- and Rad3-related proteins (ATR), or ataxia telangiectasia-mutated kinases (ATM). In addition, p38 and Jun N-terminal protein kinase (JNK) pathways are activated when DNA replication and transcription are blocked, resulting in cell cycle progression and apoptosis (15, 16). Phosphorylation of p38 and/or JNK regulates transcription factors such as apoptosis signal-regulating kinase 1 (ASK1), c-jun, HMG box-containing protein 1 (HBP1), activating transcription factor 2 (ATF2), mitogen- and stress-activated protein kinase 1 (MSK1), and heat shock protein 27 (HSP27) (16–19). Also, several pathogenic viruses, such as human immunodeficiency virus type 1, novel pandemic influenza A (H1N1) virus, and Epstein-Barr virus, have been reported to induce cell cycle arrest and/or apoptosis via activation of p38 and JNK in mouse monocytes, human lung carcinoma cells, and human B cells (20–22). On the other hand, Helicobacter pylori activates extracellular signal-regulated kinase (ERK), but not the p38 or JNK pathway, in macrophages, resulting in apoptosis (23). Thus, the pathways responsible for pathogen-induced cell cycle arrest and apoptosis may vary according to cell type and infectious agent.

The mechanisms responsible for G₁ arrest and apoptosis in trophoblasts induced by P. gingivalis are not well understood. The present results show that p38 and JNK are activated together with their downstream signaling molecules, such as HSP27 and p21, leading to G₁ arrest and apoptosis in P. gingivalis-infected trophoblasts.

RESULTS

p38 and JNK are associated with P. gingivalis-mediated G₁ arrest and apoptosis.

We previously reported that G₁ arrest and apoptosis occur following P. gingivalis infection at a multiplicity of infection (MOI) of 200, but not at MOIs of 10 and 100 under the same experimental conditions, as adopted in the present study (9). Additionally, multiple signaling pathways were activated by P. gingivalis from 24 to 48 h after infection (13). Therefore, we first examined the activation status of p38 and JNK in HTR-8 cells infected with P. gingivalis at an MOI of 200. Following P. gingivalis infection, p38 phosphorylation was induced over 24 to 48 h, while JNK2 phosphorylation also occurred, with a peak at 48 h (Fig. 1). Next, we examined the involvement of activated p38 and JNK in G₁ arrest and apoptosis. Pretreatment of HTR-8 cells with SB202190 (p38 inhibitor) or SP600125 (JNK inhibitor) reduced the level of G₁ arrest and apoptosis induced by P. gingivalis, whereas their effects were negligible in terms of cell necrosis (Table 1). Gingipains of P. gingivalis can modulate on apoptosis and apoptosis-related molecules in trophobalsts as a result of release from intracellular P. gingivalis (13). On the other hand, gingipains can be released into medium in a soluble form (24). To determine the role of exogenous gingipains in cell death and activation of apoptosis-related molecules, apoptosis and p38/JNK pathways were examined using a gingipain fraction and KDP136 (Rgp/Kgp-null mutant). Apoptosis was not markedly induced by either (see Fig. S1 in the supplemental material), supporting a role for intracellular gingipains in apoptosis. Inhibition of p38 or JNK abrogated caspase 3 activity (Fig. 2), suggesting that G₁ arrest and apoptosis induced by P. gingivalis in HTR-8 cells are regulated through the p38 and JNK pathways. It has been reported that p38 and/or JNK mediate G₁ arrest and apoptosis together with their downstream molecules, such as ASK1, ATF2, HBP1, and HSP27 (17–19). Therefore, we also examined phosphorylation of those molecules up to 48 h after infection in HTR-8 cells (Fig. 3). P. gingivalis infection was found to induce phosphorylation of HSP27 (Thr82), with peaks at 12 and 48 h, whereas ASK1 (Thr845), c-Jun (Ser63), HBP1 (Ser302), MSK1 (Thr581), and ATF2 (Thr71) were negligibly activated. In addition, the effects of gingipains, as an exogenous stimulus, were faint in regard to phosphorylation of p38 and HSP27, in comparison to infection of living cells (Fig. S2). These findings suggest that G₁ arrest and apoptosis by P. gingivalis are regulated by HSP27 phosphorylation via the p38 and JNK pathways. Furthermore, we speculate that invasive P. gingivalis activates p38, JNK, and HSP27 to cause G₁ arrest and apoptosis, possibly through intracellular release of gingipains.

FIG 1.

p38 and JNK phosphorylation upregulated in HTR-8 cells infected with P. gingivalis. (A) HTR-8 cells were infected with P. gingivalis at an MOI of 200 for the indicated times. The expression profiles of p38 and JNK were examined by immunoblotting. Tubulin was included as a loading control. (B) Densitometric analysis of blots showing the phosphorylation and total proteins, expressed in arbitrary units. Data are representative of results from three independent experiments.

TABLE 1.

Effects of p38 and JNK inhibitors on G₁ arrest and apoptosis^a

Treatment	G1 phase in cell cycle (%)	S phase in cell cycle (%)	G2/M phase in cell cycle (%)	Early apoptosis (%)	Necrosis (%)
No infection	53.15 ± 1.14	38.15 ± 0.50	8.69 ± 0.81	5.25 ± 0.71	0.09 ± 0.03
SB202190	50.78 ± 1.32	40.27 ± 1.83	8.96 ± 0.94	7.58 ± 1.03	0.32 ± 0.16
SP600125	50.02 ± 1.29	31.60 ± 0.61	18.38 ± 0.94	11.55 ± 0.95	0.09 ± 0.08
P. gingivalis	71.12 ± 0.31**	19.22 ± 1.16**	9.67 ± 0.90	58.23 ± 2.60**	17.59 ± 1.99**
SB202190 + P. gingivalis	64.33 ± 2.22*	23.94 ± 1.02*	11.74 ± 4.22	45.99 ± 2.30*	16.56 ± 3.25
SP600125 + P. gingivalis	63.79 ± 3.07*	22.09 ± 4.07	14.12 ± 1.71*	45.01 ± 2.58*	18.19 ± 1.80

^aSB202190 is a p38 inhibitor, and SP600125 is a JNK inhibitor. Cells in which G₁ arrest was assessed (G₁ phase, S phase, and G₂/M phase) were fixed and stained with PI. Cells in early apoptosis and necrosis were not fixed and were stained with annexin V and PI. *, significant difference compared with P. gingivalis-infected cells (P < 0.05); **, significant difference compared with uninfected cells (P < 0.01).

FIG 2.

Effects of p38 and JNK inhibitors on caspase 3 activity. HTR-8 cells were infected with P. gingivalis at an MOI of 200 for 48 h. p38 and JNK inhibitors were added for 2 h prior to infection, and caspase 3 activity was measured. Fold changes were calculated relative to results for uninfected cells without the inhibitors. *, P < 0.01 (Student t test) compared to P. gingivalis-infected cells with inhibitors.

FIG 3.

Activation of p38 and JNK downstream signaling molecules in HTR-8 cells infected with P. gingivalis. (A) HTR-8 cells were infected with P. gingivalis at an MOI of 200 for the indicated times, and then lysates of infected and uninfected cells were immunoblotted with antibodies to p38 or JNK downstream signaling molecules. Tubulin was included as a loading control. (B) Densitometric analysis of blots showing the phosphorylation and total proteins, expressed in arbitrary units. Data shown are representative of results from three independent experiments.

P. gingivalis induces p21 expression via p38 and JNK.

In small-cell lung carcinoma cell lines, activation of p38 and JNK results in phosphorylation of p53 and consequently apoptosis (18). Therefore, we investigated whether p38 and JNK activate p53 in HTR-8 cells infected with P. gingivalis. As shown in Fig. 4, p38 and JNK inhibitors had no effect on p53 phosphorylation or accumulation. However, inhibition of p38 or JNK prevented HSP27 phosphorylation and p21 expression following P. gingivalis infection (Fig. 4). While the results indicate that neither p38 nor JNK activates p53, other studies with colon carcinoma and multiple myeloma cell lines have reported that p53 induces the phosphorylation of p38 and/or JNK, leading to apoptosis (25, 26). Therefore, we transcriptionally silenced p53 expression with small interfering RNA (siRNA) (Fig. 5A) in order to examine its involvement in p38 and JNK phosphorylation. As shown in Fig. 5B and C, knockdown of p53 did not affect activation of JNK by P. gingivalis. On the other hand, the levels of phosphorylation of p38 and HSP27 induced by P. gingivalis were decreased following p53 knockdown (Fig. 5B and C). These results suggest that the p38/HSP27 pathway is controlled by p53 phosphorylation following P. gingivalis infection in HTR-8 cells.

FIG 4.

Effects of p38 and JNK inhibitors on p53 downstream signaling molecule activation. HTR-8 cells were infected with P. gingivalis at an MOI of 200 for 48 h. p38 and JNK inhibitors were added for 2 h prior to infection, and then the expression profiles of signaling molecules were examined by immunoblotting. Tubulin was included as a loading control. Also depicted is densitometric analysis of blots showing phosphorylation and total proteins, expressed in arbitrary units. Data are representative of results from three independent experiments. (A) SB202190 treatment (p38 inhibitor); (B) densitometric analysis following SB202190 treatment; (C) SP600125 treatment (JNK inhibitor); (D) densitometric analysis following SP600125 treatment.

FIG 5.

siRNA knockdown of p53 suppresses p38 and HSP27 activation. HTR-8 cells were transfected with siRNA for p53 or a nontarget control (siNT). (A) Cells were lysed and immunoblotted with anti-p53 or tubulin antibodies at 48 h after transfection. (B) HTR-8 cells were infected with P. gingivalis at an MOI of 200 for the indicated times, and then lysates of infected and uninfected cells were immunoblotted with antibodies to p38 or JNK downstream signaling molecules. Tubulin was included as a loading control. (C) Densitometric analysis of blots showing the phosphorylation and total proteins, expressed in arbitrary units. Data are representative of results from three independent experiments.

HSP27 is associated with P. gingivalis-mediated G₁ arrest and apoptosis.

HSP27 phosphorylation enhances p53 function along with DNA damage response in human embryonic kidney cells (27). To evaluate the effects of HSP27 activation on cell cycle arrest and apoptosis, we silenced HSP27 expression with siRNA (Fig. 6A). In the control cells, P. gingivalis infection enhanced p53 phosphorylation and accumulation, as well as p21 expression. In contrast, no induction of p21 was observed in HSP27 knockdown cells (Fig. 6B). Moreover, G₁ arrest was reduced, whereas cell cycle progression through the G₂/M and S phases proceeded normally, in the HSP27 knockdown cells (Fig. 6C). Additionally, HSP27 knockdown abrogated apoptosis induction and reduced caspase 3 activation (Fig. 6D and E), whereas necrotic responses did not differ between the control and HSP27 knockdown cells (Fig. 6D). These results suggest that P. gingivalis diverts HSP27 phosphorylation events via p53/p38 and JNK, which causes cell cycle disturbance and eventually apoptosis.

FIG 6.

siRNA knockdown of HSP27 suppresses P. gingivalis-mediated G₁ arrest and apoptosis. HTR-8 cells were transfected with siRNA for HSP27 or a nontarget control (siNT). (A) Immunoblotting was performed with anti-HSP27 or tubulin antibodies at 48 h after transfection. (B) siRNA knockdown cells were infected with P. gingivalis at an MOI of 200 for 48 h, and then expression profiles of signaling molecules were examined by immunoblotting. Tubulin was included as a loading control. Also depicted is densitometric analysis of blots showing the phosphorylation and total proteins, expressed in arbitrary units. Data are representative of results from three independent experiments. (C) siRNA knockdown cells were infected with P. gingivalis as described for panel B, and the cell cycle distribution was determined using flow cytometry. (D) siRNA knockdown cells were infected with P. gingivalis as described for panel B, followed by staining with annexin V and propidium iodide (PI), and analyzed using flow cytometry. (E) siRNA knockdown cells were infected with P. gingivalis for 48 h as described for panel B, and then caspase 3 activity was measured. Fold changes were calculated as described for panel B and are shown relative to P. gingivalis-infected NT siRNA. *, P < 0.05; **, P < 0.01 (Student's t test) compared with P. gingivalis-infected NT siRNA.

Involvement of gingipains in p38 and JNK pathways.

Arginine-specific gingipains (Rgp) and lysine-specific gingipain (Kgp) are cysteine proteases produced by P. gingivalis (24). In this study, we examined the effects of gingipains on expression and phosphorylation status of p38, JNK, ASK1, c-jun, HBP1, ATF2, and HSP27 using the treated KDP136 strain, which does not produce gingipains but expresses fimbriae on the cell surface (13). The treated KDP136 strain did not cause an increase in phosphorylation of HSP27, p38, or JNK (Fig. 7). For other molecules tested, the effects of treated KDP136 were similar to those of the wild type (Fig. 3 and 7C and D). These results suggest that P. gingivalis gingipains have specific effects on HSP27 phosphorylation via p38, JNK, and p53 and modulate the activity of multiple signaling pathways, resulting in both cell cycle arrest and cell death in HTR-8 cells.

FIG 7.

HTR-8 cells infected with a P. gingivalis fimbriated gingipain-null mutant strain do not activate the p38 or JNK pathway. Cells were infected with P. gingivalis-treated KDP136 (fimbriated gingipain-null mutant strain) at an MOI of 200 for the indicated times, and then lysates of infected and uninfected cells were analyzed by immunoblotting. (A) The expression profiles of p38 and JNK were examined by immunoblotting. Tubulin was included as a loading control. (B) Densitometric analysis of blots showing the phosphorylation and total proteins, with values expressed in arbitrary units. Data are representative of results from three independent experiments. (C) The expression profiles of p38 or JNK downstream signaling molecules were examined by immunoblotting. Tubulin was included as a loading control. (D) Densitometric analysis of blots showing the phosphorylation and total proteins, with values expressed in arbitrary units. Data are representative of results from three independent experiments.

DISCUSSION

Placental development requires intracellular coordination between the fetal and maternal uterine compartments, including fetal trophoblasts, maternal stromal cells, and immune cells (28). Progenitor cytotrophoblasts differentiate into syncytium trophoblasts and extravillous trophoblasts (EVT), with EVT notably playing an important role in fetal growth and stability (28). Apoptosis is important for normal placental developmental processes, including trophoblast attachment, as well as invasion and differentiation (29). However, excessive trophoblast apoptosis has been observed in pregnancies associated with fetal growth restriction and preterm birth (30), which may be due to microbial infection (31). Several pathogens, such as Listeria monocytogenes, human herpesvirus 8, and Zika virus, have been reported to invade trophoblasts, resulting in apoptosis of placental cells (32–34). P. gingivalis infection of HTR-8 cells can cause cytopathic effects likely mediated by DNA damage response and ERK1/2 activation (9, 13). In addition, expression of more than 2,000 genes is regulated by P. gingivalis infection of HTR-8 cells (14). p38 is activated following P. gingivalis infection of HTR-8 cells (14), and JNK phosphorylation is responsible for inducing cell cycle arrest and apoptosis in HTR-8 cells exposed to hyperosmolar stress (35). In the present study, we examined signaling pathways that regulate cell cycle disruption and apoptosis caused by P. gingivalis infection and found evidence of involvement by p53, p21, and HSP27. Based on these findings, we propose that P. gingivalis initiates the signaling cascade shown in Fig. 8.

FIG 8.

Proposed schematic model for cell cycle regulation and induction of apoptosis in HTR-8 cells infected with P. gingivalis. P. gingivalis invasion leads to activation of p53 and JNK. Upon activation, p53 directly induces p21 expression, leading to G₁ arrest and apoptosis (18). Moreover, p53 phosphorylates p38, resulting in progressive p38 activation. Phosphorylation of p38 and JNK causes p21 activation, which, in turn, induces G₁ arrest and apoptosis.

The p38 and/or the JNK pathway is activated by various types of cellular stress and control cell proliferation, differentiation, survival, migration, cell cycle progression, and apoptosis (16, 36). Hence, our results showing activation of p38 and JNK are consistent with our findings of G₁ arrest and apoptosis mediated by P. gingivalis (Fig. 1 and Table 1). p38 generally phosphorylates p53, which can lead to induction of the G₂/M checkpoint induction (37), while p53, in turn, enhances p38 phosphorylation and p21 expression, which causes a G₁ cell cycle block and apoptosis (38). Indeed, multiple pathways induced by P. gingivalis converge on G₁ arrest and apoptosis, as P. gingivalis also activates cellular DNA damage signaling pathways that act through an ATR/Chk2/p53 pathway to cause G₁ arrest and apoptosis (13). In the present study, knockdown of p53 inhibited phosphorylation of p38 but not of JNK (Fig. 5), suggesting that the DNA damage response induces phosphorylation of p38 through p53 activation in trophoblasts. These findings suggest that p38 phosphorylation is controlled by p53 following P. gingivalis infection.

Apoptosis induced by an anticancer drug, such as N-(4-hydroxyphenyl) retinamide or apigenin, is triggered by an increase in reactive oxygen species, leading to phosphorylation of HSP27 through p38 and JNK and subsequent apoptosis (39). In this study, we noted that HSP27 was phosphorylated at threonine 82 via both the p38 and JNK pathways, which in turn led to upregulation of p21 expression (Fig. 3 and 4). Thus, it is suggested that HSP27 phosphorylation occurs under the control of the p53/p38 and JNK pathways following P. gingivalis infection. In addition, it has been shown that HSP27 phosphorylated by p38 can bind to p53, resulting in an increase in the level of p21 expression, cell cycle arrest, and apoptosis in rat myoblast cell lines (40). We found that HSP27 phosphorylation was decreased in p53-silenced cells following P. gingivalis infection, while the levels of p53 were not diminished in HSP27 knockdown cells (Fig. 5 and 6). Additionally, HSP27 siRNA silencing diminished p21 expression and abrogated both G₁ arrest and apoptosis (Fig. 6). Those findings are consistent with reports showing that p38 activated by p53 phosphorylation can lead to p21 induction and cell cycle arrest (41) and that HSP27 expression induced by p53 is associated with apoptosis (42). In another study, we showed that p21 was under the control of the ERK1/2-Ets1 and p53 pathways in HTR-8 cells infected with P. gingivalis (13). The present findings suggest that P. gingivalis exploits the p53/p38/HSP27 and JNK/HSP27 signaling pathways, as well as the ERK1/2-Ets1 pathway, to upregulate p21 expression levels, thus contributing to G₁ cell cycle arrest and apoptosis. Although neither p38 nor JNK phosphorylation occurred in response to treated KDP136 (Fig. 7A), we previously showed that p53 phosphorylation and p21 induction were directly affected by P. gingivalis internalized in HTR-8 cells with or without gingipains (13). Together, these results suggest that HSP27 phosphorylation via the p38 and JNK pathways requires the presence of gingipains.

Caspase 3 was not activated in HSP27 knockdown cells (Fig. 6E), whereas necrotic changes were observed in cells treated with HSP27 siRNA (Fig. 6D) as well as with MAPK inhibitors (p38 and JNK) (Table 1). In the classical distinction, the differences between apoptosis and necrosis are characterized as programmed versus accidental forms of cell death. However, a recent study showed that necrosis can also be programmed, though in a manner distinguishable from that of apoptosis (43). Programmed necrosis is a caspase-independent cell death that occurs as a backup strategy for apoptosis or when caspases are inactivated (44). Thus, necrosis might be induced in both HSP27-silenced cells and cells treated with MAPK inhibitors (p38 and JNK) following P. gingivalis infection. The ratio of apoptosis in HSP27 knockdown cells did not parallel the reduction in caspase 3 activity (Fig. 6D and E). In HTR-8 cells infected with P. gingivalis, caspase 3 activity is affected by multiple signaling pathways, such as DNA damage signaling pathways, p38, JNK, and ERK1/2-Ets1 (9, 13). As shown in Fig. 6E, even though HSP27 expression was silenced, the effect on apoptosis was limited. Therefore, decreases in caspase 3 activation may not completely recapitulate the effect of treatment with p38 or JNK inhibitors.

In conclusion, P. gingivalis exploits the JNK and p53/p38 pathways in HTR-8 cells, leading to cellular damage. We propose that activation of the complex signaling network, such as the HSP27/p21 pathway via p53/p38 and JNK, following P. gingivalis invasion of trophoblasts is an important factor for negative control of cell cycle progression and subsequent promotion of apoptosis, with the potential for a negative impact on the maintenance of pregnancy.

MATERIALS AND METHODS

Bacterial and cell cultures.

The bacterial strains used were P. gingivalis ATCC 33277 and related mutants, including a long fimbria-null (fimA) mutant (KDP150) (45) and Rgp/Kgp-null (deficient in rgpA, rgpB, and kgp) triple mutant (KDP136) (46). Bacterial cells were grown in Trypticase soy broth supplemented with yeast extract (1 mg/ml), menadione (1 μg/ml), and hemin (5 μg/ml), as described previously (9). KDP136 was fimbriated as described previously (47). Briefly, the supernatant obtained from a fimbria-null (KDP150) culture was filtered through a 0.2 μm-pore-size filter (Asahi Glass Co., Ltd., Tokyo, Japan), and then a membrane vesicle-depleted supernatant (VDS) containing soluble gingipains was obtained by centrifugation at 100,000 × g for 50 min. The gingipain-null mutant (KDP136) was inoculated into fresh culture medium containing 30% VDS, and gingipain activity allowed processing of surface fimbrial structures. The HTR-8/SVneo trophoblast line (referred to here as HTR-8 cells) was provided by Charles Graham (Kingston, Ontario, Canada). HTR-8 cells originated from human first-trimester (8 to 10 weeks of gestation) placenta explant cultures and were immortalized by the simian virus 40 (SV40) large T antigen (48). Cells were cultured in RPMI 1640 medium (Sigma-Aldrich, St. Louis, MO) supplemented with 5% fetal bovine serum at 37°C in 5% CO₂.

Western immunoblotting.

HTR-8 cells were solubilized in cell lysis/extraction reagent (Sigma-Aldrich) containing a protease and phosphatase inhibitor cocktail (Thermo Scientific, Rockford, IL) and gingipain-specific inhibitors (KYT-1 and KYT-36; Peptide Institute, Osaka, Japan). Immunoblotting was performed as previously described (9). Blots were probed at 4°C overnight with the following primary antibodies: anti-phospho-p38 (Thr180/Tyr182), 1:1,000; anti-p38, 1:1,000; anti-phospho-JNK (Thr183/Tyr185), 1:1,000; anti-JNK, 1:1,000; anti-phospho-ASK1 (Thr845), 1:1,000; anti-ASK, 1:1,000; anti-phospho-c-Jun (Ser63), 1:1,000; anti-c-Jun, 1:1,000; anti-phospho-HBP1 (Ser302), 1:1,000; anti-HBP, 1:1,000; anti-phospho-MSK1 (Thr581), 1:1,000; anti-MSK, 1:1,000; anti-phospho-ATF2 (Thr71), 1:1,000; anti-ATF2, 1:1,000; anti-phospho-HSP27 (Thr82), 1:1,000; anti-HSP27, 1:1,000; anti-phospho-p53 (Ser15), 1:1,000, anti-p53, 1:1,000; and anti-p21, 1:1,000 (Cell Signaling Technology, Beverly, MA). Proteins and phosphorylated proteins were detected using Pierce ECL substrate (Thermo Scientific). Blots were stripped and probed with antitubulin antibody (Sigma-Aldrich) as a loading control. Densitometric analysis of bands was performed using ImageJ software.

RNA interference.

Small interfering RNA (siRNA) for HSP27 and a control siRNA were purchased from Santa Cruz Biotechnology (Santa Cruz, CA), while that for p53 (5′-GCAUGAACCGGAGGCCCAUTT-3′) (49) was purchased from Sigma Genosys Japan (Ishikari, Japan). The siRNAs were introduced into HTR-8 cells using Lipofectamine 2000 reagent (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions. At 24 h after transfection, the medium was replaced and cells were incubated for a further 24 h prior to infection.

Chemicals.

SB202190, a p38-MAPK specific inhibitor, and SP600125, a JNK-MAPK specific inhibitor, were purchased from Sigma-Aldrich. The solvents used and final concentrations were as follows: SB202190, 10 μM in dimethyl sulfoxide (DMSO), and SP600125, 10 μM in DMSO. HTR-8 cells were preincubated with the inhibitors for 2 h prior to infection.

Flow cytometry. (i) Cell cycle analysis.

Infected or control HTR-8 cells were trypsinized, washed with cold phosphate-buffered saline (PBS), and then fixed in 70% ethanol at −20°C overnight. Ethanol-fixed cells were washed with PBS and incubated in 1 ml of 0.1 mg/ml RNase A solution at 37°C for 30 min. The cells were stained with 50 μg/ml of propidium iodide (PI; Sigma-Aldrich). Cell cycle analysis of 30,000 cells per sample was performed with excitation at 488 nm in a flow cytometer (FACScan, Becton-Dickinson, San Jose, CA). Data were analyzed with Cell-Quest software (BD Biosciences, Bedford, MA) and ModFit LT 3.0 (Verity Software, La Jolla, CA). The values were in agreement with those obtained with Cell-Quest software when the S-phase events were divided between the G₁ and G₂/M phases, based on the peak channels of fluorescence intensity. The percentages of cells in the G₁, S, and G₂/M phase were calculated using ModFit LT 3.0 software.

(ii) Apoptosis.

For annexin V staining, cells were harvested and stained with an annexin V-fluorescein isothiocyanate (FITC) apoptosis detection kit (BioVision, Mountain View, CA) according to the manufacturer's protocol, and flow cytometric analysis was performed.

Caspase 3 activity.

Caspase 3 activity was measured using a caspase 3/CPP32 colorimetric assay kit (BioVision) according to the manufacturer's instructions on a microplate reader at 405 nm (Bio-Rad; model 680).

Statistical analyses.

Quantitative data are presented as means ± standard deviations (SDs). Statistical analyses were performed using an unpaired Student t test.