Mycoplasma hyopneumoniae Inhibits Porcine Beta-Defensin 2 Production by Blocking the Unfolded Protein Response To Facilitate Epithelial Adhesion and Infection

Mycoplasma hyopneumoniae causes the disease porcine enzootic pneumonia, a highly contagious and chronic disease affecting pigs. Understanding the molecular mechanisms of its pathogenicity is critical for developing effective interventions to control this swine respiratory disease. Here, we describe a novel virulence mechanism by which M. hyopneumoniae interferes with the host unfolded protein response (UPR) and eventually facilitates bacterial adhesion and infection.

ABSTRACT

Mycoplasma hyopneumoniae causes the disease porcine enzootic pneumonia, a highly contagious and chronic disease affecting pigs. Understanding the molecular mechanisms of its pathogenicity is critical for developing effective interventions to control this swine respiratory disease. Here, we describe a novel virulence mechanism by which M. hyopneumoniae interferes with the host unfolded protein response (UPR) and eventually facilitates bacterial adhesion and infection. We observed that M. hyopneumoniae infection suppressed the UPR target molecules GRP78 and CHOP by reducing PKR-like endoplasmic reticulum kinase/eukaryotic initiation factor 2 alpha (PERK/eIF2α) phosphorylation, ATF6 cleavage, and X-box binding protein 1 (XBP1) splicing. Interestingly, further analyses revealed that host UPR inhibition subsequently suppressed the NF‐κB pathway, leading to the reduced production of porcine beta-defensin 2 (PBD-2), thus facilitating M. hyopneumoniae adherence and infection. This study provides new insights into the molecular pathogenesis of M. hyopneumoniae and sheds light upon its interactions with the host.

INTRODUCTION

Porcine respiratory diseases negatively impact the economy of the pig rearing industry. Porcine enzootic pneumonia (PEP), a worldwide prevalent chronic respiratory disease, is caused by Mycoplasma hyopneumoniae infection, which is characterized by dry coughing, severe respiratory distress, and growth reduction (1, 2). The infectious process is frequently complicated by coinfections with bacterial and viral pathogens that further aggravate the symptomatic manifestation, clinically referred to as the porcine respiratory disease complex (PRDC) (3, 4). Despite the many efforts to reduce the economic loss caused by M. hyopneumoniae, PEP and PRDC are still major concerns in the swine industry, largely due to unclear pathogenetic processes of M. hyopneumoniae (2, 5).

The endoplasmic reticulum (ER) is a crucial intracellular organelle in eukaryotic cells. It acts not only as a main site of posttranslational modifications, folding, and oligomerization of newly synthesized membrane and secretory proteins but also as a major signal transduction compartment in the cell (6). Many factors, such as bacterial infection, ischemia, hypoxia, heat shock, and increased protein synthesis, impair ER functions, resulting in ER stress and evoking adaptive unfolded protein response (UPR) programs. Three branches of the UPR have been identified in eukaryotic cells, and each branch is named after its major molecular regulator. Thus, PKR-like ER kinase (PERK), inositol-requiring enzyme-1α (IRE1α), and activating transcription factor-6 (ATF6) represent three UPR pathways (7). A hallmark of the UPR is the upregulation of the 78-kDa glucose-regulated protein (GRP78, an ER chaperone protein) and C/EBP homologous protein (CHOP, a proapoptotic factor) (8–10). The UPR is a cytoprotective signaling pathway aimed at restoring cellular homeostasis that also invokes innate immune signaling in response to invading microorganisms (11, 12). The UPR triggers signal transduction events associated with innate immunity and host defense (12). Failure to restore ER functions results in programmed cell death (13). Clearly, the UPR is the first line of defense against pathogens in host immune responses, which plays an important role in host-pathogen interactions (14). Actually, mounting evidence indicates that bacteria such as Brucella abortus (15), Listeria monocytogenes (16), and Yersinia pseudotuberculosis (17) have evolved strategies that mitigate or exploit the UPR to promote their survival, which plays an important role in bacterial pathogenesis (18–23). Despite its critical role in bacterial pathogenesis, the UPR has not been investigated during M. hyopneumoniae infection, raising the question of whether M. hyopneumoniae infection and/or pathogenesis is an ER/UPR-dependent process.

Previous studies have shown that pathogens are able to take advantage of the optimal conditions created by an activated UPR (24). For instance, induction of the UPR following Mycobacterium tuberculosis infection promotes bacterial survival (25). The activation of the transcription factor X-box binding protein 1 (XBP1) pathway of UPR plays a significant role in Salmonella enterica growth and replication (26). It is not surprising that many bacterial pathogens activate and take advantage of the UPR given that the UPR is indeed a stress response (27, 28). Intriguingly, bacterial inhibition of the UPR contributing to intracellular proliferation has recently been described in Simkania negevensis and Legionella pneumophila, which is very rare but is starting to emerge (22, 23). However, how the UPR inhibition cascades to facilitating bacterial infection is still unclear.

Antimicrobial peptides (AMPs) are small (<10-kDa) soluble host defense peptides that play an integral part in the mammalian innate immune response, helping to prevent infection by inhibiting pathogen growth on skin and mucosal surfaces and subsequent dissemination to normally sterile sites (29). Two major classes of AMPs in mammals are defensins and cathelicidins (30). Some bacterial pathogens resist AMP-mediated innate immune clearance to establish infection by interfering with or suppressing host AMP expression levels (29). For example, Shigella flexneri downregulates the expression of cathelicidin LL-37 and β-defensin-1 in intestinal epithelial cells during early infection to promote bacterial survival and colonization and invasion of the gastrointestinal tract (31). Neisseria gonorrhoeae gains a survival advantage in the female genital tract by downregulating cathelicidin LL‐37 expression (32). Although we have gained considerable insights into AMP production modulated by bacterial infection, the mechanisms of how bacteria regulate the expression of AMPs have not been fully elucidated.

In this study, we report that M. hyopneumoniae inhibits the UPR during infection and that this rare modulation promotes its adhesion to and infection of host cells. Interestingly, we demonstrate that M. hyopneumoniae suppresses all three distinct arms of the UPR and cascades repressing NF-κB signaling. Subsequently, this repression of NF-κB signaling reduces the production level of porcine beta-defensin 2 (PBD-2), thereby promoting M. hyopneumoniae adhesion to and infection of host cells. Altogether, our findings reveal a novel virulence mechanism of M. hyopneumoniae, which improves our understanding of the molecular pathogenesis of M. hyopneumoniae and may help exploit new treatments for PEP and PRDC by targeting UPR processes.

RESULTS

M. hyopneumoniae infection alters the UPR.

Bacterial infections often lead to host UPR. We hypothesized that M. hyopneumoniae infection may also induce the UPR innate immune response. To test this hypothesis, we first monitored the mRNA expression of GRP78 and CHOP, the hallmark genes of the UPR, in primary porcine tracheal epithelial cells (PTECs). Interestingly, the mRNA expression of both GRP78 and CHOP was significantly downregulated following M. hyopneumoniae infection, reaching the lowest level at 24 h postinfection (hpi) (Fig. 1A). The downregulation of GRP78 and CHOP expression was confirmed via Western blotting (Fig. 1B and C). The results thus demonstrate that M. hyopneumoniae infection interferes with the UPR in PTECs.

FIG 1.

M. hyopneumoniae infection alters cellular UPR. (A to C) The expression of GRP78 and CHOP was downregulated following M. hyopneumoniae infection in PTECs demonstrated by qPCR (A) and Western blotting (B and C). (D to F) The ER stress inducer Tu did not alter the expression of GRP78 and CHOP in PTECs with a preincubation of M. hyopneumoniae demonstrated by qPCR (D) and Western blotting (E and F). PTECs were incubated with M. hyopneumoniae or an equal volume of PBS for 18 h, followed by the addition of Tu (2 μg/ml) or DMSO and further incubation for 6 h. The data were normalized to the corresponding values in mock-infected cells. All data are presented as the means ± the SDs from three independent experiments, and significance was assessed by one-way ANOVA with Dunnett’s multiple-comparison test relative to the control (0 h) (A to C) or by one-way ANOVA with Tukey’s multiple-comparison test (D to F). *, P < 0.05; **, P < 0.01; ***, P < 0.001. Mhp, M. hyopneumoniae; MW, molecular weight.

To further confirm the UPR interference caused by M. hyopneumoniae infection, we examined the UPR in primary cultured porcine alveolar macrophages (PAMs) following M. hyopneumoniae infection. PAMs are widely used as an in vitro model system to study porcine respiratory pathogen-host interactions and porcine-specific pathogenesis. In M. hyopneumoniae-infected PAMs, a reduction in GRP78 and CHOP mRNA was observed at 0 to 12 hpi, with the lowest level at 12 hpi and UPR induction at 36 hpi (see Fig. S1A in the supplemental material). Consistent with their transcriptional regulation, GRP78 and CHOP protein production was also significantly decreased following M. hyopneumoniae challenge, with the lowest levels reached at 12 hpi (Fig. S1B and C). In contrast, tunicamycin (Tu), a routinely used UPR inducer, caused the opposite changes at both the mRNA and protein levels of GRP78 and CHOP. These data indicate that UPR interference by M. hyopneumoniae infection also occurs in PAM cells.

The above-described results demonstrated that M. hyopneumoniae infection and ER stress oppositely modulated the host UPR, suggesting that M. hyopneumoniae infection would counterbalance the effect of ER stress on the UPR. To test this possibility, we incubated PTECs with M. hyopneumoniae or phosphate-buffered saline (PBS), followed by treatment of Tu to monitor the mRNA expression of GRP78 and CHOP. As expected, GRP78 and CHOP mRNA was upregulated upon Tu treatment alone (P < 0.01); however, they were virtually unchanged after preincubation with M. hyopneumoniae (Fig. 1D). Additionally, no cytotoxicity was observed in Tu treatment (Fig. S1D). Consistent with the mRNA expression results, GRP78 and CHOP proteins were also unaltered in cells with the same treatment (P < 0.05) (Fig. 1E and F).

Taken together, these data suggest that M. hyopneumoniae infection suppresses UPR in porcine cells.

M. hyopneumoniae infection suppresses all three UPR pathways.

The UPR comprises three principal pathways (PERK, ATF6, and IRE1). To gain insights into the impact of M. hyopneumoniae on specific branches of the UPR, we investigated each of the three well-characterized UPR sensors (PERK, ATF6, and XBP1) during M. hyopneumoniae infection. Among the three UPR signaling branches, activation of the PERK pathway results in the phosphorylation of the eukaryotic initiation factor 2 alpha (eIF2α) subunit, leading to the attenuation of protein translation (33). PERK also activates the expression of ATF4, a transcription factor that further activates the proapoptotic genes CHOP and GADD34 (34). In fact, M. hyopneumoniae infection not only reduced PERK phosphorylation but also decreased eIF2α phosphorylation (Fig. 2A). Additionally, PERK-eIF2α repression following M. hyopneumoniae infection resulted in downregulated mRNA levels of the eIF2α downstream ATF4 target gene (Fig. 2B). In response to ER stress, the 90-kDa full-length ATF6 (p90 ATF6) translocates from the ER to the Golgi and is subsequently processed to a 50-kDa active form (35). Western blotting demonstrated that the generation of the active form of ATF6 was decreased in M. hyopneumoniae-infected cells (Fig. 2C and D). These results suggest that M. hyopneumoniae infection suppresses the PERK and ATF6 UPR pathways.

FIG 2.

M. hyopneumoniae inhibits the three UPR branches. (A and B) The inhibition of PERK pathway in PTECs following M. hyopneumoniae infection was confirmed by Western blotting of the phosphorylation level of PERK and eIF2α (A) and/or by qPCR analysis of the ATF4 expression (B). Tu was used as a positive control for UPR activation. (C and D) The downregulation of ATF6 pathway in PTECs with M. hyopneumoniae infection was determined by Western blotting of ATF6 cleavage (p50 ATF6). (E and F) Suppression of the IRE1 pathway in PTECs with M. hyopneumoniae infection was obtained by XBP1 mRNA splicing. (E) Top, the scheme for XBP1 mRNA splicing analysis; bottom, representative gel image of PCR amplification of total XBP1 cDNA digested with PstI after M. hyopneumoniae infection. (F) XBP1 mRNA splicing is demonstrated by the ratio of the band intensities for spliced and total XBP1 DNA. Quantifications were normalized to those of the control (0 h). Data are presented as the means ± SDs of the results from three independent experiments. Significance was assessed by one-way ANOVA with Dunnett’s multiple-comparison test relative to the control (0 h). *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

In response to the accumulation of unfolded proteins in ER, the endoribonuclease IRE1 selectively cleaves a 26-nucleotide (nt) fragment of XBP1 mRNA to generate the active spliced XBP1 (XBP1s) transcription factor, which in turn activates target gene transcription to enhance the ER protein-folding capacity. The XBP1 cDNA was amplified by reverse transcription-PCR (RT-PCR) and digested by PstI, whose recognition site is located within the 26-nt cleaved region (36, 37) (Fig. 2E). We observed a decrease in XBP1s and an increase in unspliced XBP1 (XBP1u) in PTECs upon M. hyopneumoniae infection compared with those of the control cells, indicating that the IRE1 pathway was suppressed (Fig. 2E). Consistent with these findings, XBP1 mRNA splicing (the ratio of spliced XBP1 DNA to total XBP1 DNA) was significantly decreased in PTECs after M. hyopneumoniae infection (Fig. 2F).

To determine whether UPR inhibition by M. hyopneumoniae infection is specific or is a broad effect on many pathways due to cell viability decrease or apoptosis, we assayed the cell viability and apoptosis in PTECs during M. hyopneumoniae infection at 6, 12, and 24 hpi (Fig. S2). We found that M. hyopneumoniae infection neither changed cell viability nor induced apoptosis, suggesting that M. hyopneumoniae-induced UPR inhibition is specific.

Collectively, these results demonstrate that M. hyopneumoniae infection suppresses all three UPR signaling branches.

UPR inhibition contributes to M. hyopneumoniae adhesion to host cells.

Previous studies have shown that bacterial pathogens have evolved molecular mechanisms that mitigate the UPR to promote their intracellular survival and proliferation (12). We investigated whether UPR inhibition during infection is beneficial to M. hyopneumoniae survival. Considering that cytoadherence is a crucial step in host colonization by mycoplasmas and subsequent proliferation (38), we examined the number of M. hyopneumoniae cells that adhered to PTECs in response to UPR suppression. PTECs were exposed to different concentrations of 4μ8C (an IRE1-specific inhibitor), 4-benzenesulfonyl fluoride hydrochloride (AEBSF; an ATF6 inhibitor), and GSK2606414 (a PERK-specific inhibitor) to disrupt the three UPR signaling pathways (IRE1, ATF6, and PERK, respectively), followed by M. hyopneumoniae infection. The number of M. hyopneumoniae cells adhered to PTECs was assessed at 12 hpi by TaqMan quantitative PCR (qPCR). The efficiency of 4μ8C inhibition was confirmed by IRE1 protein blotting and XBP1 mRNA excision analyses (Fig. S3A to C). Surprisingly, suppression of IRE1 by 50 and 100 μM 4μ8C significantly increased the number of adhered M. hyopneumoniae cells by 1.78-fold (P < 0.01) and 2.25-fold (P < 0.01), respectively (Fig. 3A). In cells that were treated with 50, 150, and 300 μM AEBSF to repress ATF6, the number of adhered M. hyopneumoniae cells increased by 1.21-, 1.40-, and 1.90-fold, respectively (Fig. 3B). The inhibitory effect of AEBSF on ATF6 was again confirmed by protein blotting (Fig. S3D). Last, compared with the untreated control, PTECs treated with the PERK inhibitor GSK2606414 at 0.5, 1, and 10 μM all significantly increased the number of adhered M. hyopneumoniae cells by 2-fold, 2.44-fold (P < 0.01), and 3.46-fold (P < 0.01), respectively (Fig. 3C) (confirmation of the inhibitory efficiency of GSK2606414 on eIF2α is shown in Fig. S3E).

FIG 3.

(A to C) UPR inhibition boosts M. hyopneumoniae adhesion. M. hyopneumoniae adhesion was promoted in PTECs with IRE1 knockdown by the 4μ8C treatment (A), ATF6 knockdown by the AEBSF treatment (B), or PERK knockdown by the GSK2606414 treatment (C). PTECs were separately treated by 4μ8C, AEBSF, or GSK2606414 for 24 h, following M. hyopneumoniae infection for 12 h, and then cells were washed twice with PBS and harvested to analyze the number of M. hyopneumoniae cells adhered to PTECs by TaqMan qPCR. All assays were performed with three independent experiments, and values represent the means ± SDs. Significance was assessed by one-way ANOVA with Dunnett’s multiple-comparison test relative to the control. **, P < 0.01.

Based on the above-mentioned results, we determined that UPR inhibition played a positive role in M. hyopneumoniae adherence and infection. We reasoned that UPR activation may inhibit M. hyopneumoniae adherence. To test this hypothesis, we investigated the effect of increasing Tu‐induced UPR on M. hyopneumoniae adherence. Compared with the results of the control dimethyl sulfoxide (DMSO) treatment, the Tu (2 μg/ml) treatment significantly decreased the number of M. hyopneumoniae cells adhered to PTECs (Fig. 4A, P < 0.05). In addition, we further assessed the effect of UPR branch activation on the adherence of M. hyopneumoniae in PTECs. We quantified the number of M. hyopneumoniae cells adhered to PTECs in the presence of different concentrations of salubrinal, which activates the PERK UPR branch by increasing eIF2α phosphorylation (Fig. S3F). Activation of the PERK UPR branch reduced the number of M. hyopneumoniae cells adhered to PTECs and exhibited dose-dependent suppression (Fig. 4B, P < 0.01). The IRE1 and ATF6 pathways were activated by transfecting PTECs with the tandem construct hemagglutinin (HA)-tagged XBP1s and HA-tagged p50ATF6, respectively. The expression of XBP1s or p50ATF6 was confirmed by HA or p50ATF6 protein blotting (Fig. S3G and H). Overexpression of XBP1s significantly reduced the number of adhered M. hyopneumoniae cells, as measured by TaqMan qPCR following M. hyopneumoniae infection, relative to that observed in the presence of the empty vector pCAGGS-HA transfection (Fig. 4C, P < 0.05). The transient expression of p50ATF6 also decreased the number of adhered M. hyopneumoniae cells, as shown in Fig. 4D.

FIG 4.

UPR activation is detrimental to M. hyopneumoniae adherence. (A to D) M. hyopneumoniae adherence was inhibited in PTECs with UPR activation by Tu treatment (A), PERK branch activation by salubrinal treatment (B), XBP1 branch activation by XBP1s overexpression (C), or ATF6 branch activation by p50ATF6 overexpression (D). PTECs were transfected with the empty vector PCAGGs-HA, the HA-XBP1s-expressing plasmid, or the HA-p50ATF6-expressing plasmid, or they were treated with DMSO or salubrinal, following M. hyopneumoniae infection for 12 h, after which cells were harvested to analyze M. hyopneumoniae adhesion. The data are presented as the means ± SDs of the results from three independent assays. Significance was assessed by one-way ANOVA with Dunnett’s multiple-comparison test relative to the control. *, P < 0.05; **, P < 0.01.

Thus, the suppression of each examined UPR signaling branch benefits M. hyopneumoniae host adhesion.

UPR inhibition enhances M. hyopneumoniae infection.

As shown in Fig. 3 and 4, UPR inhibition benefited M. hyopneumoniae adherence. It is of interest to ask whether UPR inhibition by M. hyopneumoniae may confer an infection advantage. By performing the same experiments described above using different types of UPR inhibitors, 10,000 cells were subsequently scored and analyzed by a flow cytometer to determine the percentage of infected cells per condition (gating strategy for flow cytometry analysis was shown in Fig. S4), and we observed an increase in M. hyopneumoniae infection of IRE1-inhibited cells. The percentage of infected cells in the presence of 100 μM 4μ8C was 23.55%, whereas this percentage in the untreated group was 14.15% (Fig. 5A and B). Similar results were obtained in the ATF6-inhibited PTECs pretreated with different concentrations of AEBSF. An average of 16.7%, 24.55%, and 30.6% of cells were infected when treated with 50, 150, and 300 μM AEBSF, respectively (Fig. 5A and C). In the GSK2606414-treated groups (PERK-eIF2α was inhibited), 19.55%, 23.2%, and 33.25% of the cells were infected when treated with 0.5, 1, and 10 μM GSK2606414, respectively, in contrast with the untreated control group, which showed 14.15% of the cells being infected (Fig. 5A and D).

FIG 5.

UPR inhibition facilitates M. hyopneumoniae infection. (A to D) Flow cytometry analysis showing that UPR inhibition promoted M. hyopneumoniae infection to PTECs. Infected cells were detected by flow cytometry and then gated based on mock-infected cells. (A) Representative fluorescein isothiocyanate (FITC) histograms obtained during the analysis are shown. (B to D) Bar diagrams showing the percentages of M. hyopneumoniae infected cells pretreated with 4μ8C (B), AEBSF (C), or GSK2606414 (D). (E) M. hyopneumoniae was cultured in the presence of DMSO, 4μ8C, AEBSF, or GSK2606414. After incubation for 12 h, M. hyopneumoniae proliferation was assessed by TaqMan qPCR. M. hyopneumoniae from the above-described cultures infected PTECs for 12 h, and M. hyopneumoniae adherence was quantified using TaqMan qPCR. All data are presented as the means ± SDs of the results from three independent experiments, and significance was assessed by one-way ANOVA with Dunnett’s multiple-comparison test relative to the control. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

To exclude the possibility that these inhibitors directly acted upon M. hyopneumoniae to increase its proliferation and adherence, M. hyopneumoniae was cultured for 12 h in the presence of different concentrations of 4μ8C, AEBSF, or GSK2606414, after which the proliferation of M. hyopneumoniae was assessed by TaqMan qPCR. Approximately 1.0 × 10⁷ CFU/ml from the cultures were used to infect PTECs (3.2 × 10⁵ cells/ml) for 12 h, after which the adherence of M. hyopneumoniae was quantified by TaqMan qPCR after washing off the unadhered bacterial cells. No increased proliferation or adherence was observed (Fig. 5E).

Overall, inhibition of UPR signaling promotes M. hyopneumoniae infection of PTECs.

UPR inhibition reduces PBD-2 production.

We next investigated the mechanism by which UPR inhibition promoted M. hyopneumoniae adhesion and infection. Previous studies have suggested that the UPR and other innate immune pathways regulate each other. Parts of the UPR may represent an ancestral innate immune recognition mechanism that is used for the detection of and response to cellular injury by replicating pathogens (39). We speculated that UPR inhibition might regulate immune effectors, thereby promoting M. hyopneumoniae adhesion and infection. To test this idea, we performed qPCR to assess the transcription of different types of immune effectors, including interleukins, interferons, defensins, and tumor necrosis factor alpha (TNF-α), after the suppression of each UPR branch (data not shown). Among these examined immune effectors, the level of PBD-2 mRNA was downregulated in the IRE1-inhibited PTECs (by 4μ8C pretreatment, P < 0.01), and similar results were obtained for ATF6 and PERK inhibition (P < 0.01) (Fig. 6A). In contrast, Tu treatment significantly increased the level of PBD-2 mRNA in PTECs (P < 0.01) (Fig. 6B). These findings suggest that UPR positively regulates PBD-2 production. PBD-2 is one of the beta-defensins secreted by pigs exhibiting high antibacterial activities (40, 41), which is thought to play a significant role in protection against respiratory infections in pig epithelial cells (42, 43). We subsequently monitored the mRNA expression of PBD-2 at different time points after M. hyopneumoniae infection. Interestingly, the transcriptional upregulation of PBD-2 was observed at 6 hpi, followed by transcriptional downregulation at 12 and 24 hpi (P < 0.0001) (Fig. 6C). We concluded that M. hyopneumoniae infection is capable of inducing the release of PBD-2 in an early stage. However, this induction was actively moderated by M. hyopneumoniae, as evidenced by the fact that the transcriptional level of PBD-2 at 12 h was significantly lower than that of the control group (P < 0.0001). Therefore, we speculated that UPR inhibition by M. hyopneumoniae infection participates in reducing the release of PBD-2 in host cells.

FIG 6.

UPR inhibition impairs PBD-2 production. (A and B) UPR positively regulated the expression of PBD-2. PTECs were treated with DMSO, 4μ8C, AEBSF, or GSK2606414 for 24 h (A) or with Tu for 12 h (B), and PBD-2 mRNA expression was quantified by qPCR. (C) PBD-2 mRNA expression following M. hyopneumoniae infection was monitored by qPCR. (D and E) PTECs were treated as described in panels A and B and then incubated with M. hyopneumoniae for 12 h. PBD-2 mRNA levels were detected by qPCR. (F and G) UPR inhibition reduced the expression of PBD-2. (F) PTECs were transfected with siRNAs for 24 h and harvested for qPCR analysis to verify the efficiency of IRE1, ATF6, or eIF2α knockdown separately. NC represents the control siRNA. (G) This was followed by M. hyopneumoniae infection for 12 h, and the cell supernatant was collected for ELISA detection of PBD-2 production. (H) UPR activation increased PBD-2 production. The data represent the means ± SDs of the results from three or six independent experiments. (A to H) Significance was assessed by one-way ANOVA with Dunnett’s multiple-comparison test relative to the control (A, B, D, E, G, and H), by one-way ANOVA with Tukey’s multiple-comparison test (C), or by two-tailed Student's t test (F). *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

To verify whether UPR inhibition by M. hyopneumoniae infection reduced PBD-2 production, we utilized the inhibitors to disrupt each UPR branch separately and noticed that the suppression of each UPR branch results in reduced mRNA expression of PBD-2 in M. hyopneumoniae-infected PTECs (Fig. 6D). However, Tu treatment significantly upregulated the mRNA expression of PBD-2 after M. hyopneumoniae infection (Fig. 6E), suggesting that M. hyopneumoniae antagonized the release of PBD-2 by inhibiting the UPR pathway. In agreement with these results, we also found a decrease in PBD-2 production in M. hyopneumoniae-infected cells in the presence of small interfering RNA (siRNA) duplexes that target the UPR branches through detection by enzyme-linked immunosorbent assay (ELISA) (Fig. 6G). The silencing efficiency of these siRNAs (siIRE1, siATF6, and sieIF-2α) was confirmed by measuring the transcript levels of IRE1, ATF6, and eIF2α, respectively (Fig. 6F). In contrast, transfecting cells with the tandem construct HA-tagged XBP1s or HA-tagged p50ATF6 or treating cells with salubrinal to activate the IRE1, ATF6, or PERK UPR branch, respectively, resulted in significantly increased PBD-2 production in PTECs after M. hyopneumoniae infection (Fig. 6H).

Collectively, these data indicate that UPR inhibition in response to M. hyopneumoniae infection reduces the release of PBD-2.

UPR inhibition reduces PBD-2 production via NF-κB signaling.

We then aimed to define the signaling pathway involved in UPR-mediated PBD-2 production. Here, we investigated NF-κB signaling in PTECs treated with the aforementioned inhibitors following M. hyopneumoniae infection by phospho-p65 and IκBα protein blot analyses, given that the UPR and NF-κB interact on multiple levels (44). We observed significantly decreased phospho-p65 expression and increased IκBα expression, indicating that the suppression of the UPR branch inhibited NF-κB signaling in M. hyopneumoniae-infected PTECs. In contrast, cells treated with Tu exhibited a significant increase in p65 phosphorylation and a marked reduction in IκBα expression (Fig. 7A). Consistent with the results of Tu treatment and the addition of inhibitors in M. hyopneumoniae-infected PTECs, the activation of each of the three UPR branches (by transfecting cells with the tandem construct HA-tagged XBP1s or HA-tagged p50ATF6 or salubrinal treatment) resulted in NF-κB signaling activation (Fig. 7B and C), whereas the addition of siRNAs targeting the UPR branches led to NF-κB signaling inhibition in M. hyopneumoniae-infected cells (Fig. 7D). These results demonstrated that UPR inhibition downregulated NF-κB signaling.

FIG 7.

UPR inhibition decreases PBD-2 production via NF-κB signaling. (A to D) UPR positively regulated NF-κB signaling, as demonstrated by the measurement of p-P65, P65, IκBα, or β-actin protein. (E and F) PDTC reduced PBD-2 production in PTEC-activated UPR by inhibiting NF-κB signaling. PTECs were pretreated with PDTC for 1 h and then incubated with Tu for 12 h, following M. hyopneumoniae infection. PBD-2 expression was determined by qPCR (E) and ELISA (F). M. hyopneumoniae inhibited NF-κB signaling at 12 hpi, as demonstrated by Western blotting (G). All assays were performed with three independent experiments, and the values represent the means ± SDs of the results. Significance was assessed by one-way ANOVA with Dunnett’s multiple-comparison test relative to the control. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Furthermore, NF-κB has also been shown to be a critical transcription factor for PBD-2 expression (45). We thus investigated whether NF-κB signaling affected PBD-2 production upon UPR inhibition. We treated PTECs with different concentrations of an NF-κB inhibitor, pyrrolidine dithiocarbamate (PDTC), followed by incubation with 2 μg/ml Tu. We found that PDTC pretreatment prevented the expected upregulation of PBD-2 mRNA expression in PTECs with an inhibited UPR (Fig. 7E). PDTC treatment (at a concentration of ≥100 μM) even downregulated PBD-2 mRNA to a level lower than that of untreated PTECs (P < 0.05). The result was further confirmed by ELISA (P < 0.05) (Fig. 7F). It had already been confirmed that UPR activation resulted in the upregulation expression of PBD-2 mRNA and protein expression by using recombinant plasmids or the specific activator in M. hyopneumoniae-infected PTECs. However, if these cells were pretreated with PDTC, the mRNA expression or the production of UPR-induced PBD-2 was significantly diminished (Fig. S5), indicating that the specific suppression of NF-κB signaling decreased PBD-2 production. This result was further testified by the detection of the NF-κB signaling level at 12 h after M. hyopneumoniae infection (at 12 hpi, we have confirmed that M. hyopneumoniae reduced PBD-2 production). We observed a significant decrease in phospho-p65 and an increase in IκBα, compared with those levels in uninfected PTECs, suggesting that NF-κB signaling was inhibited (Fig. 7G).

Taken together, our data demonstrate that UPR inhibition-induced reduction in PBD-2 might occur by suppressing NF-κB activation.

The reduction in PBD-2 release facilitates M. hyopneumoniae adhesion to host cells.

To confirm whether PBD-2 impacts M. hyopneumoniae adherence, HA-tagged PBD-2 cDNA (HA-PBD-2) was transfected into PTECs, and the number of M. hyopneumoniae cells adhered to PTECs was assessed using TaqMan qPCR. We found that M. hyopneumoniae cell adherence to host cells was reduced in a dose-dependent manner after HA-PBD-2 transfection (P < 0.05) (Fig. 8A). To directly verify whether PBD-2 impairs M. hyopneumoniae adherence, we stimulated M. hyopneumoniae-infected PTECs with different concentrations of recombinant PBD-2 protein for 6 h and evaluated the proliferation and adherence of M. hyopneumoniae by TaqMan qPCR. Compared with the results of the untreated control cells, the qPCR results showed a significant reduction in M. hyopneumoniae proliferation and adherence in a dose-dependent manner in PBD-2-treated PTECs. Importantly, PBD-2 had a greater impact on the adherence than the proliferation of M. hyopneumoniae (P < 0.05, Fig. 8B and C). These results were further confirmed by an immunofluorescence assay (IFA) (Fig. 8D).

FIG 8.

The decrease in PBD-2 release enhances M. hyopneumoniae adherence. (A) PBD-2 overexpression suppressed M. hyopneumoniae adherence to PTECs demonstrated by TaqMan qPCR. (B and C) M. hyopneumoniae proliferation (B) and adherence (C) were inhibited by the addition of recombinant PBD-2 in PTECs for 6 h. (D) An M. hyopneumoniae P46 protein IFA with mouse anti-M. hyopneumoniae monoclonal antibody 3G11 (green) was utilized to detect M. hyopneumoniae adherence to PTECs in the presence of recombinant PBD-2 protein for 12 h. DAPI was used to stain cellular nuclei (blue). Controls on uninfected cells with or without PBD-2 protein treatment were shown in Fig. S6. (E) PBD-2 suppressed the expression of M. hyopneumoniae adhesins P97 and P116. The data used for statistical analysis represent the means ± SDs of the results from three independent experiments. Significance was assessed by one-way ANOVA with Dunnett’s multiple-comparison test relative to the control. *, P < 0.05; **, P < 0.01; ***, P < 0.001. GFP, green fluorescent protein.

In addition, M. hyopneumoniae encodes several adhesins. P97 (46) and P116 (47) have both been reported to mediate the adherence of M. hyopneumoniae to swine cilia. We reckoned that PBD-2 decreased M. hyopneumoniae adhesion to host cells by affecting adhesin expression. To verify our hypothesis that PBD-2 hinders the adhesin expression, we assessed the mRNA expression of P97 and P116 of M. hyopneumoniae in the presence of 500 pg/ml recombinant PBD-2 protein in the culture medium. Compared with the PBS-treated control, PBD-2 treatment significantly suppressed the mRNA expression of P97 and P116 (P < 0.01), suggesting that PBD-2 was capable of inhibiting the expression of adhesins and thereby decreasing M. hyopneumoniae adhesion to host cells.

Together, these results suggest that the reduction in PBD-2 production promotes M. hyopneumoniae adhesion to host cells.

DISCUSSION

The host UPR is responsible for responding to ER stress and is a critical component of protein homeostasis. Bacterial modulation of the host UPR can be a double-edged sword, in that bacteria may modulate the UPR to their advantage during infection, for example, promoting survival and infection; on the other hand, UPR modulation may aid the host innate immune system in recognizing bacterial infection and mounting the appropriate defense against bacterial infection (12). Reports indicating that bacterial infection can modulate this response are very rare but are starting to emerge. Intracellular bacterial pathogens such as Legionella pneumophila, Simkania negevensis, and Brucella spp. are able to subvert the host UPR to benefit their intracellular proliferation (22, 23, 48). In addition, bacterial toxins, including listeriolysin O (LLO) of Listeria monocytogenes and Shiga toxin produced by Shiga toxin-producing Escherichia coli (STEC), exemplify their abilities to induce the host UPR (20, 21). Given the multitude of virulence factors expressed by different bacterial pathogens, it is very likely that novel mechanisms by which bacterial pathogens manipulate the host UPR to promote their replication, survival, and spread remain to be identified. Here, we showed that M. hyopneumoniae inhibits the host UPR, modulates the NF-κB pathway, and eventually reduces PBD-2 production, facilitating its adherence to and infection of the host cells (Fig. 9). This finding represents a novel virulence mechanism of M. hyopneumoniae.

FIG 9.

Model depicting the promotion of epithelial adhesion and infection by UPR inhibition during M. hyopneumoniae infection. M. hyopneumoniae suppresses all three UPR signaling pathways (ATF6, IRE1, and PERK). UPR inhibition induced by M. hyopneumoniae diminishes NF-κB activation and then reduces PBD-2 production, leading to the promotion of M. hyopneumoniae epithelial adhesion and infection. The arrows indicate activation, and the blunt-ended lines indicate inhibition. Molecules and cellular effects modulated by M. hyopneumoniae are highlighted in red squares.

Our results showed that M. hyopneumoniae infection suppressed the host UPR through the inhibition of all three pathways mediated by PERK, ATF6, and IRE1 (Fig. 1 and 2). It has been demonstrated that Legionella spp. inhibit the host UPR by suppressing the IRE1 pathway, one of the three UPR pathways, thus blocking the splicing of XBP1u mRNA to spliced XBP1 (XBP1s). However, the ATF6 and PERK pathways remained unaffected (22). S. negevensis infection induced an ER stress response, which was subsequently downregulated, suggesting that it might be able to inhibit the host UPR (23); however, it is not yet known which pathway(s) is responsible for the suppression of the host UPR during S. negevensis infection. To the best of our knowledge, this is the first report showing that bacterial infection inhibited the host UPR by suppressing all three branches mediated by PERK, ATF6, and IRE1. In contrast, it has been reported that more bacterial pathogens are able to induce, but not suppress, the host UPR through the differential regulation of the three branches mediated by PERK, ATF6, and IRE1. Observation from an RNA interference (RNAi) screen searching for factors required for bacterial growth indicated that Brucella spp. might selectively induce the host UPR by activating the IRE1 pathway of the tripartite UPR signaling cascade but not PERK or ATF6 (15). Similarly, Shiga toxin from E. coli, cholera toxin from Vibrio cholerae, and various pore-forming toxins selectively induced the IRE1 pathway or IRE1 and ATF6 pathways that mediated the host UPR (20, 49, 50).

NF-κB, as a master transcription regulator, plays a central role in the host response to infections by microbial pathogens, orchestrating the innate and acquired host immune responses (51). Many different pathogenic bacteria have evolved diverse strategies to manipulate this pathway for their own benefit. The activation of the NF-κB pathway is frequently associated with infections by many bacterial pathogens, and their activation of the NF-κB pathway through induction of the host UPR has been well studied (52, 53). Conversely, some other bacterial pathogens are able to inhibit the host NF-κB pathway. The Gram-negative extracellular bacteria Yersinia pestis, Yersinia pseudotuberculosis, and Yersinia enterocolitica use a type III secretion system to disrupt NF-κB activation (53), while uropathogenic E. coli (UPEC) increases the stability of IκBα, thus preventing its degradation and blocking the NF-κB-dependent pathway (54). However, it is even not clear whether these bacterial pathogens inhibit the host NF-κB pathway by suppressing the UPR, not to mention the involved mechanism by which host NF-κB is inhibited. Our results showed that M. hyopneumoniae infection in PTECs and PAMs significantly suppressed the NF-κB pathway by inhibiting the host UPR, and it was further demonstrated that suppressing any one of the three UPR branches via the inhibitor of PERK, ATF6, or IRE1α (GSK2606414, AEBSF, or 4μ8C) significantly blocked the activation of the NF-κB pathway (Fig. 7). Our study, for the first time, elucidates the mechanisms by which M. hyopneumoniae inhibits the NF-κB pathway through suppression of the host UPR by inhibiting the three branches mediated by PERK, ATF6, and IRE1.

It has been reported that bacterial pathogens usually manipulate the UPR and thus modulate the NF-κB signaling pathway to trigger the host innate immune response, including the induction of proinflammatory cytokines, such as interleukin 6 (IL-6), TNF-α, and IL-1β (12). Antimicrobial peptides are an important component of the host innate immune (55). The most interesting finding in our study is that the suppression of NF-κB signaling mediated by UPR inhibition led to a decrease in PBD-2 production (Fig. 6 and 7). PBD-2 is a member of β-defensin, and only β-defensins have been found in pigs to date (56), being widely expressed in the skin, mucosal surfaces of airways, digestive tract, and urogenital tract in pigs (42). Previous studies have revealed that PBD-2 is an effective antibacterial agent against various bacteria in vitro (43, 56, 57). Our results showed that the decrease in PBD-2 production resulted in enhanced adherence and infection of M. hyopneumoniae in PTECs (Fig. 8A to D). Our in vitro assays also demonstrated a significant decrease in M. hyopneumoniae proliferation and adherence to PTECs by overexpression of PBD-2 or with recombinant PBD-2 protein treatment, further confirming the role of PBD-2 in adhesion and infection by M. hyopneumoniae. More interestingly, we found that PBD-2 could significantly reduce the expression of adhesins P97 and P116 in M. hyopneumoniae cells and played a more important role in suppressing adherence than in inhibiting the proliferation of M. hyopneumoniae (Fig. 8B, C, and E). Therefore, it is very likely that M. hyopneumoniae infection inhibits the host UPR and NF-κB pathway, resulting in the reduced production of PBD-2, leading not only to enhanced viability and/or growth of M. hyopneumoniae but also increased the expression of adhesins P97 and P116, thus facilitating bacterial adherence and infection.

In summary, we identified a previously unrecognized virulence mechanism in which M. hyopneumoniae decreases the production of the PBD-2 by interfering with the UPR/NF-κB axis, thus facilitating M. hyopneumoniae adherence and infection. Our findings lead to a better understanding of the interaction between M. hyopneumoniae and the host, providing new insights into the molecular pathogenesis of M. hyopneumoniae. We hope that this and future studies may lead to the development of new therapies against diseases caused by M. hyopneumoniae.

MATERIALS AND METHODS

Ethics statements.

Primary porcine tracheal epithelial cell (PTEC) and porcine alveolar macrophage (PAM) collection was approved by the Committee on the Ethics of Animal Experiments of the Harbin Veterinary Research Institute of the Chinese Academy of Agricultural Sciences and performed according to the approved animal care guidelines and protocols.

Mycoplasma strain, cells, and culture conditions.

M. hyopneumoniae ATCC 25095 was purchased from the American Type Culture Collection (ATCC) and cultivated in Mycoplasma medium (Basal Media, Shanghai, China) at 37°C. To estimate the numbers of CFU in the cultures, serial dilutions were plated on modified pleuropneumonia-like organism (PPLO) medium containing 1.5% agarose (V2111; Promega) and incubated at 37°C. CFU were counted 7 to 10 days later using a microscope (58). M. hyopneumoniae was pelleted by centrifugation at 10,000 × g for 10 min and resuspended at 1.0 × 10⁷ CFU/ml in phosphate-buffered saline (PBS).

PTECs were prepared from the tracheas of two 5-week-old specific-pathogen-free (SPF) piglets using previously described protocols (59) and were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% heat-inactivated fetal bovine serum (FBS), 1% antibiotics (100 μg/ml streptomycin and 100 U/ml penicillin; Gibco), and 10 mM HEPES (Invitrogen). The cells were incubated at 37°C in 5% CO₂. PAMs were obtained from lavage fluid samples from the lungs of two 5-week-old SPF piglets and cultured in RPMI 1640 (Gibco) supplemented with 10% FBS and 1% antibiotics.

M. hyopneumoniae infection and chemical treatments.

PTECs (3.2 × 10⁵ cells/ml) were plated in 12-well plates. Upon reaching 90% confluence, PTECs were pretreated with different concentrations of chemicals or the same volume of dimethyl sulfoxide (DMSO) for 24 h, followed by inoculation with M. hyopneumoniae (1.0 × 10⁷ CFU/ml). After incubation for 12 h, the cells were harvested and then subjected to Western blotting.

Tu and 4μ8C were purchased from Sigma-Aldrich. GSK2606414, AEBSF, salubrinal, and PDTC were purchased from Selleckchem and dissolved in DMSO. The recombinant PBD-2 protein was purchased from USCN Life Science (Wuhan, China).

Real-time quantitative RT-PCR and Xbp1 splicing.

To detect UPR induction, PTECs (3.2 × 10⁵ cells/ml) and PAMs (6.0 × 10⁵ cells/ml) were infected with M. hyopneumoniae (1.0 × 10⁷ CFU/ml), and the cells were collected at 0, 6, 12, 24, 36, and 48 hpi. Total RNA was extracted from the cells using an RNeasy minikit (Qiagen Sciences, Hilden, Germany), according to the manufacturer’s instructions. RNA was reverse transcribed using a Transcriptor first-strand cDNA synthesis kit (Roche Diagnostics, Indianapolis, IN, USA).

qPCR was performed in triplicate using FastStart Universal SYBR green master (Rox; Roche Diagnostics, Indianapolis, IN, USA). All data were acquired using a QuantStudio 3 real-time PCR system (Applied Biosystems, Carlsbad, CA, USA). The expression value of each gene was normalized to that of glyceraldehyde-3-phosphate dehydrogenase (GAPDH), and final values were calculated using the ΔΔCT method. The results were analyzed using the QuantStudio design and analysis software v1.4 (Applied Biosystems). All of the primers used in this study are summarized in Table 1.

TABLE 1.

Primers used in this study

Target by usea	Primer direction	Sequence (5′–3′)	Restriction enzyme(s)
RT-qPCR
GRP78	Forward	TGGTGTCTTTGAAGTCGTGG
	Reverse	TCCGAACATCTTTGCCAGTC
CHOP	Forward	GCCTTTCTCCTTTGGGACACTGTCCAGC
	Reverse	CTCGGCGAGTCGCCTCTACTTCCC
ATF4	Forward	CCCTTTACGTTCTTGCAAACTC
	Reverse	GCTTCCTATCTCCTTCCGAGA
IRE1α	Forward	TCCCCGAGTGTTTCAGTTTC
	Reverse	ATGGCGATGTACTGGAACTG
ATF6	Forward	GCCGTCCCAGATATTAGTCAC
	Reverse	AATCCCAATCTTCATCCGACC
eIF2α	Forward	TTCCAAGTGATGTACCCAGTG
	Reverse	CGGTTTCTCATTTCCTGGTTG
PBD-2	Forward	ACCTGCTTACGGGTCTTG
	Reverse	CTCTGCTGTGGCTTCTGG
GAPDH	Forward	TCGGAGTGAACGGATTTGG
	Reverse	TGGGTGGAATCATACTGGAAC
P97	Forward	TTGGGTGGCTAAGTTTCT
	Reverse	AGTTCATCAAAGCGAGTA
P116	Forward	CTCAGATGGAAACGGTCTT
	Reverse	TCGGGTTTGATAAAGGTAAT
PPIA	Forward	GACTGAGTGGTTGGATGG
	Reverse	TGATCTTCTTGCTGGTCTT
Amplification of XBP1s and p50ATF6 genes
XBP1s	Forward	GGGGTACCATGGTGGTGGTGGCAGCTGCGCAGAG	KpnI
	Reverse	CCGCTCGAGTCACTTCATTAATGGCTTCCAGCTTGGC	XhoI
P50ATF6	Forward	CGAGCTCGGATGGAGTCGCCTTTTAG	SacI
	Reverse	CCGCTCGAGCTAACAGACAGCTCTTCGCTTTGG	XhoI

^aRT-qPCR, reverse transcription-quantitative PCR.

To amplify xbp1 mRNA, PCR was performed for 30 cycles (94°C for 30 s, 58°C for 30 s, and 72°C for 1 min [10 min in the final cycle]) using the xbp1 primers 5′-AAACAGAGTAGCAGCGCAGACTGC-3′ and 5′-GGATCTCTAAAACTAGAGGCTTGGTG-3′, as well as the GAPDH primers 5′-ACCCAGAAGACTGTGGATGG-3′ and 5′-CCCTGTTGCTGTAGCCAAAT-3′. The XBP1 PCR product was digested with the restriction endonuclease PstI, with PstI cleavage of XBP1u yielding two fragments (285 and 310 bp).

Western blotting.

PTECs and PAMs were infected with M. hyopneumoniae and harvested at 0, 6, 12, 24, 36, and 48 hpi. Equal numbers of cells were lysed with the cell lysis buffer (1 protease inhibitor cocktail tablet [Roche] per 10 ml NP-40 buffer) for 30 min. The protein concentration was determined using a bicinchoninic acid (BCA) protein assay kit (Beyotime, Nantong, China). Equal amounts of total cell lysates were separated by SDS-PAGE. The proteins in the gel were transferred onto nitrocellulose membranes (GE Healthcare Life Science, Piscataway, NJ, USA), which were then blocked with 5% skim milk in Tris-buffered saline with Tween 20 (TBST; Solarbio, Beijing, China) at 4°C overnight and then incubated for 2 h with different primary antibodies at room temperature. Antibodies against GRP78 (ab21685), ATF6 (ab37149), and CHOP (9C8) (ab11419) were purchased from Abcam (Cambridge, MA, USA). Antibodies against total PERK (D11A8) (5683S), p-PERK (16F8) (3179S), total eIF2α (L57A5) (2103S), p-eIF2α (D9G8) (3398S), IRE1α (14C10) (3294S), IκBα (L35A5), phospho-NF-κB P65 (93H1), and NF-κB P65 (L8F6) were purchased from Cell Signaling Technology (Beverly, MA, USA). An antibody against β-actin (TA-09) was purchased from Zhongshan Goldenbridge-Bio (Beijing, China). DyLight 800-labeled goat anti-mouse IgG (H+L) or DyLight 800-labeled goat anti-rabbit IgG (H+L, 1:10,000; Kirkegaard & Perry Laboratories, Gaithersburg, MD, USA) was used for detection. The membrane was scanned using an Odyssey infrared imaging system, and the fluorescence intensity of each band was measured using Odyssey 2.1 software (Li-Cor Biosciences).

Cell viability measurements.

Cell viability was determined using a Cell Counting Kit-8 (CCK-8), according to the manufacturer’s protocol (Vazyme, Nanjing, China).

Adherence assessment.

To evaluate the effect of UPR induction in PTECs on the adherence of M. hyopneumoniae, qPCR was performed. PTECs (3.2 × 10⁵ cells/ml) were pretreated with 4μ8C, AEBSF, or GSK2606414, followed by inoculation with M. hyopneumoniae (1.0 × 10⁷ CFU/ml) for 12 h. Afterwards, the cells were washed twice with PBS and harvested using a cell scraper. Total bacterial genomic DNA was extracted using an E.Z.N.A. bacterial DNA kit (Omega BioTek, Norcross, GA, USA), following the manufacturer’s protocol. The M. hyopneumoniae copy number was quantitated in cell samples by qPCR using a specific TaqMan probe assay (60). The forward primer (F, 5′-CCAGAACCAAATTCCTTCGCTG-3′), reverse primer (R, 5′-ACTGGCTGAACTTCATCTGGGCTA-3′), and TaqMan probe (5′-FAM-AGCAGATCTTAGTCAAAGTGCCCGTG-BHQ-3′) (FAM, 6-carboxyfluorescein; BHQ, black hole quencher) were designed based on the conserved regions of the M. hyopneumoniae P97 gene (GenBank accession no. CP002274.1) and synthesized by Comate Bioscience (Jilin, China).

The PCR product amplified from the M. hyopneumoniae genome template DNA with the F and R primers was used to construct the plasmid pMD18-T-P97, which was used as a standard to calculate the copy number of M. hyopneumoniae in the qPCR assay. The plasmid was sequenced and serially diluted 10-fold, and different copy numbers of DNA (10⁵ to 10¹¹ copies/μl) were used as standard templates for the qPCR assay. Standard curves of quantification cycle (C_q) values were plotted against the logarithm of plasmid copy numbers/reaction. The composition of the qPCR assay mixture is described in the manufacturer’s protocol (Premix Ex Taq [probe qPCR]; TaKaRa, Dalian, China). Amplification was performed using a QuantStudio 3 real-time PCR system (Applied Biosystems, Carlsbad, CA, USA).

Flow cytometry.

PTECs (3.2 × 10⁵) were pretreated with DMSO or different concentrations of drugs in growth medium at 37°C, followed by infection with M. hyopneumoniae (1.0 × 10⁷ CFU/ml). At 12 hpi, the cells were harvested and then incubated with the mouse anti-M. hyopneumoniae monoclonal antibody 3G11 or IgG2a mouse kappa isotype control monoclonal antibodies (MAbs; ab18413; Abcam) for 1 h at 37°C. Subsequently, the cells were incubated with goat anti-mouse IgG-Alexa Fluor 488 (Invitrogen) for 1 h at 37°C in the dark. Finally, the cell samples were fixed with 4% paraformaldehyde (PFA) for 30 min. To determine the percentage of infected cells per condition, 10,000 cells were scored and analyzed using an Accuri C6 Plus flow cytometer. In addition, the isotype control was used to determine background staining. All flow cytometry data were analyzed using the FlowJo software (version X10.0; FlowJo, LLC, Ashland, OR).

Plasmid construction, siRNAs, and transfection.

The PBD-2 cDNA with KpnI and XhoI sites was prepared by the Beijing Genomics Institute (BGI, Beijing, China) and cloned into the pCAGGS-HA vector (Clontech).

For the generation of pHA-XBP1s, XBP1s cDNA was PCR amplified from PTECs and inserted between the KpnI and XhoI sites of pCAGGS-HA. For the construction of pHA-p50ATF6, the p50ATF6 sequence was cloned into the vector pCAGGS-HA using the SacI and XhoI restriction sites.

All siRNAs listed in Table 2 were designed and synthesized by GenePharma (China).

TABLE 2.

Sequences of siRNA used to ablate ATF6, eIF2α, and IRE1 protein expression in PTECs

Target	Sequence direction	Sequence (5′–3′)
siATF6	Forward	CCAGAAGUUAUCAAGACUUTT
	Reverse	AAGUCUUGAUAACUUCUGGTT
sieIF-2α	Forward	CCGGUUUAGUCCCGAGAAATT
	Reverse	UUUCUCGGGACUAAACCGGTT
siIRE1	Forward	GCACAGACCUGAAGUUCAATT
	Reverse	UUGAACUUCAGGUCUGUGCTT
Control siRNA	Forward	UUCUCCGAACGUGUCACGUTT
	Reverse	ACGUGACACGUUCGGAGAATT

PTECs were transfected with the recombinant plasmids or siRNAs using TransIT-X2 reagent (Mirus Bio) for 24 h.

ELISA.

Cell culture supernatants were collected to quantify the secretion levels of PBD-2 using commercial ELISA kits (USCN Life Science, Inc., Wuhan, China), according to the manufacturer’s instructions. All samples were assayed in triplicate.

Immunofluorescence assay.

PTECs (3.2 × 10⁵ cells/ml) were incubated with recombinant porcine PBD-2 protein (500 or 800 pg/ml) and M. hyopneumoniae (1.0 × 10⁷ CFU/ml) infection for 12 h. The cells were washed three times in PBS and fixed with 4% PFA for 30 min at room temperature. After blocking with blocking buffer (PBS with 5% FBS and 5% skim milk in PBS) for 30 min at 37°C, PTECs were incubated with a mouse anti-M. hyopneumoniae monoclonal antibody 3G11 overnight at 4°C and stained with goat anti-mouse IgG-Alexa Fluor 488 (1:500; Invitrogen) for 60 min at 37°C in the dark. DAPI (4′,6-diamidino-2-phenylindole; 1:100; Sigma, USA) was used to stain cellular nuclei. The stained cells were visualized using an Evos FL Auto 2 fluorescence microscope.

Statistical analysis.

The Prism software (version 7.0 for Mac; GraphPad Software, Inc.) was used for all statistical analyses. Data obtained from several experiments are reported as the mean ± standard deviation (SD). The significance of differences between two groups was determined with a two-tailed Student's t test. One-way analysis of variance (ANOVA) with Dunnett’s or Tukey’s test was employed for multigroup comparisons. For all analyses, a P value of <0.05 was considered statistically significant.