Protochlamydia Induces Apoptosis of Human HEp-2 Cells through Mitochondrial Dysfunction Mediated by Chlamydial Protease-Like Activity Factor

Obligate amoebal endosymbiotic bacterium Protochlamydia with ancestral pathogenic chlamydial features evolved to survive within protist hosts, such as Acanthamoba, 0.7–1.4 billion years ago, but not within vertebrates including humans. This observation raises the possibility that interactions between Protochlamydia and human cells may result in a novel cytopathic effect, leading to new insights into host-parasite relationships. Previously, we reported that Protochlamydia induces apoptosis of the immortalized human cell line, HEp-2. In this study, we attempted to elucidate the molecular mechanism underlying this apoptosis. We first confirmed that, upon stimulation with the bacteria, poly (ADP-ribose) polymerase (PARP) was cleaved at an early stage in HEp-2 cells, which was dependent on the amount of bacteria. A pan-caspase inhibitor and both caspase-3 and -9 inhibitors similarly inhibited the apoptosis of HEp-2 cells. A decrease of the mitochondrial membrane potential was also confirmed. Furthermore, lactacystin, an inhibitor of chlamydial protease-like activity factor (CPAF), blocked the apoptosis. Cytochalasin D also inhibited the apoptosis, which was dependent on the drug concentration, indicating that bacterial entry into cells was required to induce apoptosis. Interestingly, Yersinia type III inhibitors (ME0052, ME0053, and ME0054) did not have any effect on the apoptosis. We also confirmed that the Protochlamydia used in this study possessed a homologue of the cpaf gene and that two critical residues, histidine-101 and serine-499 of C. trachomatis CPAF in the active center, were conserved. Thus, our results indicate that after entry, Protochlamydia-secreted CPAF induces mitochondrial dysfunction with a decrease of the membrane potential, followed by caspase-9, caspase-3 and PARP cleavages for apoptosis. More interestingly, because C. trachomatis infection can block the apoptosis, our finding implies unique features of CPAF between pathogenic and primitive chlamydiae.


Introduction
Members of the order Chlamydiales are obligate intracellular bacteria that were discovered about a century ago. Although ancient chlamydiae diverged into pathogenic and primitive chlamydiae 0.7-1.4 billion years ago, all pathogenic chlamydiae species have co-evolved with their vertebrate hosts and so-called primitive chlamydiae have evolved as endosymbionts of lower eukaryotes, namely free-living amoebae (Acanthamoeba) complete with a powerful bacterial killing mechanism [1][2][3][4]. While pathogenic chlamydiae, including Chlamydia trachomatis and C. pneumoniae, are well-known human pathogens and the major cause of preventable blindness, as well as sexually transmitted and respiratory diseases [5][6][7], primitive chlamydiae, including Parachlamydia acanthamoebae, Neochlamydia hartmanellae and Protochlamydia amoebophila, are also likely to be implicated in human respiratory diseases and abortion [8][9][10][11]. Pathogenic chlamydiae have developed through a decrease in genome size and loss of redundant genes, which may be a strategy to evade the host immune network, resulting in a shift to parasitic energy and metabolic requirements, and genomes of approximately 1.0-1.2 Mb [2,3]. However, the genome of representative primitive chlamydia (Protochlamydia UWE25) is not in the process of becoming smaller and has stabilized at 2.4 Mb [4]. This observation implies the possibility that, to overcome stressful conditions, primitive chlamydiae still possess certain molecules that pathogenic chlamydiae have already lost. Thus, comparison of the two chlamydiae, which have evolved separately through different paths and inhabiting niches, is extremely intriguing and may lead to new insights into host-parasite relationships.
The complicated manipulation mechanism of pathogenic chlamydiae, which occur in host cells, is becoming more obvious. It is the striking view that chlamydial type III effector proteins, which are inclusion membrane proteins (Incs), are primarily responsible for the process of inclusion biogenesis [12][13][14][15][16]. Furthermore, pathogenic chlamydiae also possess chlamydial protease-like activity factor (CPAF) that causes two significant modifications of cellular function [17][18][19][20][21]. One of the functions is responsible for inclusion maturation through cellular matrix degradation of the inclusion membrane backbone, thereby providing flexibility to mature inclusion bodies depending on the bacterial amounts [19]. More importantly, the other function contributes to prevention of apoptosis of infected cells through degradation of BH3-only proteins, which is a switch signal, followed by accumulation of Bax proteins, which induce pore formation on mitochondria, cytochrome c release, caspase-9 and -3 activation, and then poly (ADP-ribose) polymerase (PARP) cleavage for direct apoptosis induction [19,20]. Furthermore, contrary to most T3SS effectors such as Incs, which exhibits little conservation at sequence level among chlamydial members, CPAF is a highly conserved protease, indicating a critical role to achieve pathogenic chlamydial survival in mammalian cells including human cells [17][18][19][20][21].
Thus far, in contrast to pathogenic chlamydiae, we have found an interesting feature of primitive chlamydia in which Protochlamydia, but not the killed bacteria, induce apoptosis of immortalized human HEp-2 cells [22], suggesting an unknown molecular mechanism of pathogenic chlamydial manipulation in cells. Therefore, we attempted to elucidate the molecular mechanism of primitive chlamydial apoptosis induction by either connecting type III effector or CPAF activities, which are critical regulators to achieve adaptation in mammalian cells.

Apoptosis of HEp-2 Cells is Dependent on the Multiplicity of Infection (MOI) of Protochlamydia at an Early Stage
We first determined whether apoptosis induction was dependent on bacterial load or timing. As shown in Figure 1A and B, DAPI staining revealed that Protochlamydia obviously induced apoptosis of HEp-2 cells and, as expected, was dependent on bacterial MOI as demonstrated previously [22]. We also confirmed this feature by western blot analysis using PARP cleavage as a marker of apoptosis, which is located downstream of the apoptosis pathway [23], indicating maximum induction of apoptosis at an MOI of 100 ( Figure 1C), possibly by the presence of unknown physical limitation on chlamydial adhesion to cells. We next determined the timing of HEp-2 cell apoptosis after incubation with the bacteria. As a result, PARP cleavage began at 8 h after incubation ( Figure 1D). Taken together, the data revealed that some effector molecules might be involved in the apoptosis of HEp-2 cells.
Apoptosis and Mitochondrial Dysfunction Followed by Caspase-9 and 23, and then PARP Cleavages Pathogenic chlamydial CPAF directly contributes to the prevention of apoptosis of infected cells through degradation of BH3-only proteins to maintain infected host cells [19][20][21], which is a possible evolutionary path of pathogenic chlamydiae, revolving mitochondrial dysfunction. We therefore determined whether Protochlamydia could modulate mitochondrial function with activation of caspases and PARP cleavages. Using a DAPI staining assay, we found that a pan-caspase inhibitor obviously blocked Protochlamydia-induced apoptosis, and staurosporine, a stimulator that induces caspase-dependent apoptosis (Figure 2A and B). It was also confirmed by western blotting that the inhibitor blocked PARP cleavage ( Figure 2C). Using several specific caspase inhibitors, we next determined which caspase molecule was involved in the induction of apoptosis. The results clearly indicated that caspase-3 and 29 inhibitors, but not caspase-1 and 28 inhibitors, blocked the apoptosis ( Figure 3A), suggesting that apoptosis occurred through the mitochondrial pathway and was triggered by mitochondrial dysfunction. We also confirmed a decrease of the mitochondrial membrane potential in HEp-2 cells incubated with Protochlamydia ( Figure 3B), suggesting mitochondrial dysfunction. Thus, taken together, we clearly observed that Protochlamydia induces apoptosis by mitochondrial dysfunction followed by activations of caspase-9 and 23, and PARP cleavage.

Bacterial Entry into Cells is Required to Induce Apoptosis
We assessed whether the apoptosis induced by Protochlamydia was required for bacterial entry into cells using cytochalasin D, an inhibitor that blocks actin remodeling. As a result, the number of dead cells was significantly decreased by treatment with cytochalasin D, which was dependent on the drug concentration ( Figure 4A). We also confirmed that the amount of cleaved PARP was clearly decreased by cytochalasin D treatment ( Figure 4B). Taken together, the results indicated that bacterial entry into cells is required to induce apoptosis on HEp-2 cells, suggesting that effector molecules secreted into the cytoplasm by bacteria may be involved in the apoptosis.

Protochlamydia-induced Apoptosis
Genome sequencing revealed genes encoding putative type III machinery in the genomes of primitive (Protochlamydia UWE25) [4] and pathogenic chlamydiae [13,14]. Interestingly, it appears likely that the features of the chlamydial type III gene cluster are closely  related to those of Yersinia type III gene clusters based on high BLAST hit scores [4,13,14], indicating that Yersinia type III effector inhibitors may function in apoptosis induction. In fact, type III effector inhibitors block the growth of Waddlia in host cells, which is one of the primitive chlamydiae [24]. We therefore examined the effect of type III inhibitors (ME0052, ME0053, and ME0054) on the apoptosis using DAPI staining and western blotting. As a result, in contrast to our expectation, no effect of the inhibitors on the apoptosis was observed ( Figure S1).

Protochlamydia-secreted CPAF is Involved in the Apoptosis
We next assessed whether CPAF is involved in the apoptosis. CPAF was widely conserved among chlamydiae including Protochlamydia R18 used in this study, although similarity scores differed among chlamydiae ( Figure 5A). Interestingly, while the similarity scores among pathogenic chlamydiae were very high (% of sequence similarity; 47.4-100), the scores among primitive chlamydiae were surprisingly low (% of sequence similarity; 29-100) ( Figure 5A). However, histidine-101 and serine-499 in CPAF (C. trachomatis; DUW_3CX), which are two critical residues in the active center [25], were well conserved between primitive and pathogenic chlamydial CPAFs ( Figure 5B). Thus, the data suggested that primitive chlamydial CPAF from Protochlamydia is still active and has a critical role in amoebal and mammalian cells. It is therefore expected that a CPAF inhibitor, lactacystin, which directly binds to and blocks pathological chlamydial CPAF activiy [25], could similarly work for inhibition of primitive chlamydial CPAF activity as well, possibly preventing apoptosis. As a result, lactacystin caused a decrease in the number of dead cells ( Figure 6A) and blocked PARP cleavage, which was dependent on drug concentration ( Figure 6B), indicating that CPAF was involved in the induction of HEp-2 cell apoptosis. In addition, effect of lactacystin itself on inhibition of apoptosis is minimal because of more like lactacystin acting as an accelerator on apoptosis [26].

Discussion
It is well known that apoptosis occurs in human cells through two distinct pathways, namely extrinsic and intrinsic pathways [27][28][29][30]. The extrinsic pathway is induced by physical or environmental stimuli such as TNFa through the cellular death receptor, followed by caspase-8 and 23 activations [28,29]. The intrinsic pathway begins internal stresses such as accumulation of miss-folded proteins or effector molecules secreted by pathogens, followed by mitochondrial disability, and caspase-9 and 23 activations [30]. Both pathways require PARP cleavage for complete activation, which is located downstream of the apoptosis pathway, and is directly linked to nuclease activation [27][28][29][30]. Because caspase-9 and 23 activations occur with PARP cleavage, it is clear that the apoptosis pathway induced by Protochlamydia is the intrinsic pathway. The finding that another primitive chlamydial strain, Protochlamydia UWE25, similarly induces apoptosis of the insect cell line S2 cells by DNA fragmentation [31] supports our data.
As described previously, heat-or UV-killed Protochlamydia do not induce apoptosis [22]. In this study, we also confirmed that apoptosis begun at an early time point until almost 8 h after incubation. Moreover, we found that cytochalasin D blocked the apoptosis, indicating a requirement for bacterial entry into cells. It is well known that pathogenic chlamydial effectors manipulate the host response to create an optimal cellular environment [12][13][14][15][16][17][18][19][20][21]. It is also known that effector proteins are synthesized at a late stage of the developmental cycle [32][33][34], possibly accumulating in the infective Protochlamydia progeny to infect host cells as a secondary infection. Type III effectors, which are pathogenic chlamydial Incs, have been well investigated, and are responsible for the process of inclusion biogenesis [35][36][37]. Thus far, primitive chlamydial Incs with similar features to those of pathogenic chlamydiae have been identified [16]. In the present study, Yersinia type III inhibitors could not block the apoptosis. However, possible role of type III effectors in such apoptosis still remains unclear. It has been reported that Protochlamydia UWE25 possesses approximately 100 Kb of inserted island genes consisting of 72 distinct genes in the genome, which possibly encode type IV secretion machinery [38]. Although our genome sequencing data of Protochlamydia R18 also revealed the presence of Protochlamydia R18 ORFs against the UWE25 type IV gene cluster with a minimal BLAST hit score (data not shown), possible role of type IV effectors in the apoptosis also remains unknown. Furthermore, pathogenic chlamydiae possess CPAF as mentioned above, which is responsible for inclusion maturation through cellular matrix degradation of the inclusion membrane backbone, and is possibly secreted by type II secretion machinery. This process provides flexibility to mature inclusion bodies, dependent on the bacterial amount, and prevent apoptosis of infected cells through degradation of BH3-only proteins [17][18][19][20][21]. Meanwhile, pathogenic chlamydial CPAF appears to have a broad range of substrate specificity and thus it could not deny that the specificity had changed through evolution from ancestral chlamydiae. We therefore focused on type II secretion effector CPAF as an intriguing candidate in the apoptosis induction. In fact, we clearly demonstrated that Protochlamydia CPAF is required for the apoptosis induction in HEp-2 cells.
While the amino acid sequence of Protochlamydia R18 CPAF was found to have 99.1% (% of sequence similarity) with that of UWE25 CPAF, the R18 CPAF sequence did not resemble pathogenic chlamydial CPAFs [about 31.4% (% of sequence similarity)]. This finding suggests that ancestral chlamydial CPAF drastically changed after diverging into primitive and pathogenic chlamydiae. Moreover, histidine-101 and serine-499 of C. trachomatis CPAF, which are in the active center of activity [25], were obviously conserved among chlamydiae, regardless of  primitive chlamydiae. This observation also suggests that CPAF activity is very important to achieve adaptation in host cells, and has a critical role in manipulation of host cells.
Interestingly, pathogenic chlamydial infection (C. trachomatis L2) blocked the apoptosis (data not shown). It is well known that inhibition of pathogenic chlamydia-induced apoptosis occurs through degradation of BH3-only proteins (Puma, Bim, and Bik) by CPAF secreted into infected cells [17][18][19][20][21]. However, other substrates of CPAF may be involved in the apoptosis [39]. Thus, it is possible that pathogenic chlamydial CPAF interacts with primitive chlamydial CPAF itself to block Protochlamydia-induced apoptosis, implying a critical role of pathogenic chlamydial CPAF with a wide range of substrate specificities obtained through evolution.
In conclusion, we elucidated the molecular mechanism of HEp-2 cell apoptosis induced by Protochlamydia R18, showing that after entry, Protochlamydia-secreted CPAF induces mitochondrial dysfunction with a decrease of the membrane potential, followed by caspase-9, caspase-3 and PARP cleavages for apoptosis (a hypothetical scenario, Figure 7). The finding suggests a possible evolutionary path in which pathogenic chlamydiae may begin to eliminate the strong cytopathic effect of ancestral chlamydia on mammalian cells, and a serine-protease enzyme, CPAF, may be a key molecule that determines the survival of host cell.

Bacterial Stocks and Assessment of Numbers
Protochlamydia R18 used in this study was originally isolated from a river in Sapporo, Japan, as an endosymbiont found in environmental Acanthamoeba [40]. We also confirmed that the Protochmaydia 16S rRNA sequence was identical to that of Protochlamydia UWE25 (AB506679) with a 99.2% identity score [40]. The bacterial stock was prepared as follows. Bacteria were maintained within infected amoebae. Briefly, infected cells were harvested and disrupted by freeze/thawing. After centrifugation at 1806 g for 5 min to remove cell debris, bacteria were concentrated by high-speed centrifugation at 8006 g for 30 min. The bacterial pellet was resuspended in sucrose-phosphate-glutamic acid buffer consisting of 0.2 M sucrose, 3.8 mM KH 2 PO 4 , 6.7 mM Na 2 HPO 4 and 5 mM L-glutamic acid (pH 7.4), and then stored at -80uC until needed. The number of EB was determined by an amoeba-infectious unit (AIU) assay, using a coculture of the amoebae, established previously [41].

Amoebae and Human Cells
Free-living amoebae, A. castellanii C3 (ATCC 50739), were purchased from the ATCC and used to assess bacterial infectious progenies by an AIU assay. As mentioned above, amoebae harboring Protochlamydia were used to prepare bacterial stocks. Both amoebae were maintained in PYG broth (0.75% (w/v) peptone, 0.75% (w/v) yeast extract and 1.5% (w/v) glucose) at 30uC [22]. Immortalized epithelial cell line HEp-2 purchased from the Reken Cell Bank (Ibaraki, Japan) was used for assessment of cell death.  MN). All drugs except lactacystin were dissolved in DMSO according to the manufacturer's instructions, and stored at 220uC until use; lactacystin was also dissolved in water and stored at 220uC. Yersinea type III inhibitors (ME0052, ME0053, and ME0054) were kindly provided by Dr. Mikael Elofsson (Umeå University, Sweden), which were also dissolved in DMSO and stored at room temperature.

Cell Cultures with Bacteria
Immortalized HEp-2 cells (2610 5 ) were cultured with or without bacteria adjusted to an MOI of 10-100 or with staurosporine (10 mM) (Sigma), as a positive control for induction of apoptosis, for up to 24 h at 37uC with 5% CO 2 in DMEM containing 10% heat-inactivated fetal calf serum. Cells were also cultured in the presence or absence of bacteria with or without caspase inhibitors (100 mM) (See above), cytochalasin D (0.5-2 mM) or lactacystin (2, 10 mM). No cytotoxicity of these drugs at working concentrations in the cells was confirmed.

Assessment of Cell Death
Cell death was estimated by changes of nuclear morphology using DAPI staining according to our previous study [22]. Experiments were performed independently at least three times. Data were expressed as the mean 6 standard deviation (SD).

Western Blot Analysis
Cells collected from each culture were boiled for 5 min at 100uC in a reducing sample buffer containing 2-mercaptoethanol. Then, sample was loaded and separated by 10% (w/v) SDS-PAGE (20 mA, 80 min). Separated proteins were transferred to a polyvinylidene difluoride membrane by semi-dry electroblotting. Membranes were blocked with 5% (w/v) skim milk in TBS-T and then incubated with an anti-PARP (Roche Diagnostics, Indianapolis, IN) for 1 h at room temperature, followed by a HRPconjugated goat anti-rabbit IgG for 1 h at room temperature. Labeled proteins were visualized with Pierce ECL western blotting substrate (Thermo Scientific).

Assessment of Mitochondrial Membrane Integrity
HEp-2 cells were stimulated with Protochlamydia for 8 h, and then incubated with 100 nM MitoTracker Red CMXRos (Invitrogen, Grand Island, NY) for 30 min at 37uC according to the manufacturer's instructions. After fixation with 4% paraformaldehyde, the cells were observed under a confocal laser scanning microscope.

Statistical Analysis
Comparisons of bacterial numbers were assessed using an unpaired t-test. A value of p,0.05 was considered significant. Figure S1 Effect of type III inhibitors on apoptosis induced by Protochlamydia. A) Cells were cultured with bacteria (MOI 100) in the presence or absence of each type III inhibitor (ME52, ME53, and ME54) for 24 h. The number of dead cells was estimated by DAPI staining. Data are the means 6 SD from at least three independent experiments performed in triplicate. B) Representative western blot showing changes of PARP cleavage induced by Protochlamydia in the presence of each type III inhibitor (ME52, ME53, and ME54). Cells stimulated with bacteria were collected at 8 h after incubation and then subjected to western blotting with an antibody against PARP. DMSO; solvent control. a-tubulin; internal control. (TIF)