The Coronavirus Transmissible Gastroenteritis Virus Evades the Type I Interferon Response through IRE1α-Mediated Manipulation of the MicroRNA miR-30a-5p/SOCS1/3 Axis

TGEV infection downregulates miR-30a-5p expression.

The host miR-30 family (five members, consisting of miR30a to miR30e) plays important roles in cancers and viral infections (30, 34, 36, 37). We recently reported that miR-30a-5p, a member of the miR-30 family, is downregulated and that is expression is inversely correlated with the levels of ER stress in renal cancer, indicating that ER stress might inhibit miR-30a-5p expression (34). To assess whether ER stress suppresses the expression of miR-30a-5p, we initially analyzed the levels of miR-30a-5p in swine testicular (ST) cells following treatment with the ER stress inducer thapsigargin (Tg). Tg treatment substantially diminished the abundance of miR-30a-5p and exhibited dose-dependent suppression (Fig. 1A), indicating that Tg-derived ER stress reduces miR-30a-5p abundance. Our labs and others have shown that, similar to other coronaviral infections, TGEV infection triggers significant ER stress and initiates all three UPR pathways (1, 8). To explore whether miR-30a-5p could be regulated by TGEV infection, we initially monitored miR-30a-5p expression in ST cells after TGEV infection at different multiplicities of infection (MOIs). Compared with mock infection, TGEV infection significantly reduced the levels of miR-30a-5p at 24 h postinfection (hpi) and displayed an MOI-dependent response (Fig. 1B). To determine the stage at which miR-30a-5p suppression by TGEV infection occurs, we analyzed miR-30a-5p expression at different time points after TGEV infection. TGEV infection at an MOI of 1 caused a typical cytopathic effect (CPE), including syncytium formation in ST cells at 24 hpi, and resulted in approximately 35% cell death at 48 hpi. The miR-30a-5p reduction occurred after 12 hpi and then gradually decreased up to 48 hpi (Fig. 1D), indicating that TGEV infection decreases miR-30a-5p abundance at the late stage of infection. TGEV infection in ST cells was confirmed by quantifying the viral genomes (Fig. 1C and E). TGEV primarily infects villous epithelial cells in the small intestine in vivo and causes watery diarrhea. To assess whether TGEV infection also decreases miR-30a-5p expression in vivo, we quantified miR-30a-5p expression in piglet ilea at 48 hpi. TGEV infection resulted in an approximately 5-fold reduction in miR-30a-5p abundance in the ileum in vivo (0.021 ± 0.003) compared with that in uninfected control ileum (0.096 ± 0.016) (P < 0.01) (Fig. 1F). TGEV infection in the ileum was confirmed by quantifying TGEV RNA (Fig. 1G). These results demonstrate that TGEV infection decreases miR-30a-5p expression.

IRE1α-mediated UPR induction inhibits miR-30a-5p expression.

We next investigated the mechanisms responsible for the suppression of miR-30a-5p by TGEV infection. In recent studies, IRE1α, a highly conserved ER stress sensor comprising a protein kinase and RNase, has the ability to degrade miRNAs, in addition to degrading mRNA under ER stress (35, 38, 39). IRE1α activation by TGEV infection was assessed by real-time PCR (RT-PCR) and PstI digestion, which showed a recognition site located within the 26-nt region of XBP1 cDNA removed by IRE1α-mediated splicing, as previously described (12, 40). Consistent with previous results for the coronavirus mouse hepatitis virus (MHV) (41), TGEV infection caused substantial cytoplasmic cleavage of the XBP1u transcript into the XBP1s transcription factor starting at 24 hpi, indicating that IRE1α is activated by TGEV infection (Fig. 2A). IRE1α activation was further confirmed by analyzing the expression of the XBP1s downstream target gene, namely, ER-localized DnaJ homologue 4 (ERdj4) (Fig. 2B). The activity of IRE1α (the ratio of spliced XBP1 DNA to total XBP1u DNA) was inversely correlated with the levels of miR-30a-5p expression (Fig. 2A and 1D) (R = 0.933, P < 0.01). The decrease in miR-30a-5p expression primarily occurred within 24 to 48 hpi, the period in which significant IRE1α activation was triggered by TGEV infection (Fig. 1D and 2A). These findings suggest that TGEV-induced IRE1α activation may involve decreased miR-30a-5p expression during TGEV infection. To demonstrate that activated IRE1α is responsible for the downregulation of miR-30a-5p, we monitored the expression of miR-30a-5p in TGEV-infected or Tg-treated cells after inhibiting the IRE1α function with 4μ8c, a highly specific and selective inhibitor of the RNase activity of IRE1α (14). The effective blockage of the IRE1α RNase function by 4μ8c was confirmed by PCR analysis of XBP1 cleavage (Fig. 2C). The inhibition of IRE1α RNase by 4μ8c almost completely abolished the suppression of miR-30a-5p by TGEV or Tg (Fig. 2D and E). To further verify the contribution of IRE1α to miR-30a-5p expression, we knocked down IRE1α expression by specific small interfering RNAs (siRNAs), and the efficiency of IRE1α knockdown was confirmed by Western blotting (Fig. 2F). The silencing of IRE1α by siRNAs significantly rescued the decreased miR-30a-5p induced by TGEV infection (Fig. 2G) or Tg treatment (Fig. 2H) (P < 0.05). In addition, the efficiency of miR-30a-5p restored by IRE1α siRNAs (siIRE1α#1 to siIRE1α#3) was correlated with the knockdown efficiency of IRE1α siRNAs (Fig. 2F to H). Taken together, these results show that activated IRE1α reduces miR-30a-5p expression.

miR-30a-5p suppresses TGEV propagation.

In agreement with our previous results, compared with the results of control dimethyl sulfoxide (DMSO) treatment, blocking IRE1α with 4μ8c significantly reduced TGEV replication, as revealed by measuring the quantities of viral genomes (Fig. 3A) and infectious virions (Fig. 3B) (8). Consistent with 4μ8c treatment results, knockdown of IRE1α by siIRE1α#3, which was the most efficient at silencing IRE1α, significantly reduced the TGEV genome quantity and titer. Both siIRE1α#1 and siIRE1α#2 also exhibited a tendency to decrease TGEV propagation in ST cells but not significantly because of the silencing efficiency (Fig. 3C and D). The observed levels of TGEV reduction were correlated with the knockdown efficiency of IRE1α siRNAs (Fig. 2F and 3 and 3D). These data demonstrate that the IRE1α pathway promotes TGEV replication.

Since TGEV replication was reduced after blocking IRE1α signaling cascades by 4μ8c or IRE1α-specific siIRE1α#3, which rescued the IRE1α-mediated downregulation of miR-30a-5p expression, we reasoned that miR-30a-5p may inhibit TGEV infection. To investigate the role of miR-30a-5p in TGEV propagation, we monitored TGEV infection in ST cells after transfecting miR-30a-5p mimics at 24 h prior to infection. Compared with negative control (NC) mimics, the miR-30a-5p mimics decreased the TGEV genome quantity and progeny viral titer by up to 17-fold. The inhibition of TGEV by preexisting miR-30a-5p mimics was dose dependent, and TGEV infection (MOI = 0.01) was substantially reduced by 80 nM and 160 nM miR-30a-5p mimics (P < 0.001) (Fig. 3E and F). The miR-30a-5p mimics also significantly decreased TGEV infection when an MOI of 1 was used (data not shown). The suppression of TGEV infection by miR-30a-5p was further confirmed by a TGEV nucleocapsid protein immunofluorescence assay (IFA) (Fig. 3G). To determine the phase at which preexisting miR-30a-5p overexpression suppresses TGEV infection, we measured TGEV infection at different time points in the presence of miR-30a-5p overexpression. miR-30a-5p significantly suppressed TGEV replication at 24 and 36 hpi, indicating that the miR-30a-5p-mediated reduction in TGEV replication occurs at the late stage of TGEV infection (Fig. 3H). Furthermore, the specific suppression of endogenous miR-30a-5p in ST cells by the miR-30a-5p inhibitor boosted TGEV infection compared with that achieved with the NC inhibitor (NC-i) (Fig. 3I). These results demonstrate that miR-30a-5p suppresses TGEV infection. To verify whether IRE1α facilitates TGEV replication by manipulating miR-30a-5p expression, we analyzed TGEV production in ST cells transfected with a miR-30a-5p inhibitor in the presence of 4μ8c at a TGEV MOI of 1.0. The suppression of endogenous miR-30a-5p by the miR-30a-5p inhibitor significantly abrogated 4μ8c-mediated TGEV suppression compared with that achieved with the mock (DMSO)-treated control (Fig. 3I), indicating that IRE1α facilitates TGEV infection by manipulating miR-30a-5p expression. Taken together, our data indicate that IRE1α enhances TGEV replication by decreasing miR-30a-5p abundance.

miR-30a-5p enhances IFN-I antiviral signaling cascades rather than IFN-I induction.

Next, we explored the mechanisms responsible for miR-30a-5p-mediated TGEV inhibition. TGEV efficiently replicates in cells, despite significant amounts of endogenous IFN-I production after infection (25, 26), indicating that some underlying mechanisms are exploited by TGEV to escape IFN-I-induced antiviral responses. Given that miR-30a-5p inhibited TGEV replication at the late stage of infection (Fig. 3H), we hypothesized that miR-30a-5p suppresses TGEV replication possibly by enhancing IFN-I antiviral signaling cascades rather than by promoting the production of IFN-I. To exclude the possibility that miR-30a-5p enhances the production of IFN-I, we initially analyzed IFN-β expression in ST cells following TGEV infection when overexpressing miR-30a-5p. Consistent with the findings of previous studies (8, 25), TGEV infection elicited a substantial amount of IFN-β production at 24 hpi (Fig. 4A). Overexpression of miR-30a-5p did not significantly increase IFN-β production, as measured by quantifying IFN-β protein levels following TGEV infection relative to those observed in the presence of NC mimics (Fig. 4A), indicating that miR-30a-5p does not modulate TGEV-induced IFN-I production.

The binding of IFN-I to the IFN-I receptor primarily activates the JAK-STAT1 signaling pathway and induces hundreds of ISGs to inhibit viral infection. To verify our hypothesis that miR-30a-5p hinders TGEV replication largely by enhancing IFN-I antiviral signaling cascades, we initially monitored the kinetic profiles of interferon-stimulated gene 15 (ISG15) expression after TGEV infection (Fig. 4B). The expression of ISG15 peaked at 24 hpi and then significantly diminished, although IFN-β was continuously elevated during the period from 24 to 48 hpi (Fig. 4B) (P < 0.05). Consistent with the kinetic profile of ISG15, STAT1 blotting results demonstrated that STAT1 phosphorylation peaked at 24 hpi and then decreased (Fig. 4C). These results suggest that IFN-I antiviral signaling cascades are undermined during the later stage of TGEV infection, in contrast to IFN-β production. To directly verify if TGEV infection impairs IFN-I antiviral signaling at the later stage of infection, we stimulated TGEV-infected ST cells at 36 hpi by IFN-β for 30 min and evaluated STAT1 activation. Compared with uninfected control cells, TGEV-infected cells exhibited a significant reduction in IFN-β-elicited STAT1 phosphorylation (p-STAT1) (Fig. 4D), indicating that TGEV infection impairs IFN-I antiviral signaling at a later stage. Furthermore, the time frame of the observed decrease in ISG15 expression matched that of the observed significant reduction in miR-30a-5p abundance during TGEV infection (Fig. 4B), suggesting that miR-30a-5p may be relevant for the suppression of the IFN-I antiviral response at the later stage of infection. To validate the role of miR-30a-5p in modulating IFN-I antiviral signaling, we initially monitored the activity of the IFN-stimulated response element (ISRE) with a luciferase assay following IFN-β stimulation. Compared with NC mimics, MiR-30a-5p mimics significantly enhanced the activity of ISRE stimulated by IFN-β; in contrast, compared with the NC inhibitor (NC-i), the miR-30a-5p inhibitor suppressed IFN-β-derived ISRE activity (Fig. 4E). We next measured the expression of antiviral ISG genes (ISG15, 2′-5′-oligoadenylate synthetase-like [OASL], and myxovirus resistance protein 1 [MxA]) in the presence of miR-30a-5p overexpression during TGEV infection (Fig. 4F) or IFN-β stimulation (Fig. 4G). Compared to the NC mimic control, miR-30a-5p overexpression greatly upregulated the expression of antiviral ISG genes in response to either TGEV infection or IFN-β stimulation (Fig. 4F and G). Consistent with the results of ISRE luciferase and ISG expression assays, the overexpression of miR-30a-5p significantly promoted p-STAT1, whereas the inhibition of endogenous miR-30a-5p decreased p-STAT1 levels following IFN-β stimulation or TGEV infection (Fig. 4H). These data demonstrate that miR-30a-5p promotes IFN-I signaling. Overall, we conclude that miR-30a-5p reinforces IFN-I antiviral signaling rather than IFN-I production.

Consistent with previously published results (42, 43), we demonstrated that IFN-β substantially reduced TGEV infection and exhibited anti-TGEV activity in vitro (Fig. 4I). In agreement with the result that miR-30a-5p enhanced IFN-I antiviral signaling compared with NC mimics, overexpression of miR-30a-5p reduced TGEV infection and increased the efficiency of IFN-β in inhibiting TGEV infection by more than 100-fold (Fig. 4I) (P < 0.01). The inhibition of endogenous miR-30a-5p by the inhibitor significantly increased TGEV infection and decreased the efficiency of IFN-β in suppressing TGEV infection (Fig. 4I). Therefore, these data demonstrate that miR-30a-5p inhibits TGEV replication by enhancing IFN-I antiviral signaling rather than by modulating IFN-I production.

miR-30a-5p directly targets SOCS1 and SOCS3.

To elucidate the underlying mechanisms by which miR-30a-5p enhances IFN-I signaling, we performed computational analysis by using the TargetScan prediction program to identify the potential target genes of miR-30a-5p. Since miR-30a-5p promoted IFN-I signaling rather than IFN-I induction (Fig. 4), we primarily focused on the target genes of miR-30a-5p that enhance IFN signaling pathways. Computational analysis showed that miR-30a-5p could potentially target SOCS1 and SOCS3 through a 3′ UTR site that is conserved in mammals (Fig. 5A). The members of the SOCS family of proteins are endogenous potent negative regulators of JAK-STAT signaling transduction (19, 44). Next, we explored whether miR-30a-5p directly targets SOCS proteins and enhances IFN-I signaling by downregulating SOCS expression. To verify whether miR-30a-5p directly targets SOCS1 and SOCS3, we cloned the predicted target sites in the 3′ UTRs of SOCS1 or SOCS3 into a firefly luciferase reporter vector. In addition, a mutant vector was constructed to eliminate possible recognition by replacing five seed nucleotides (in which the underlined nucleotides in the sequence UGUUUAC were changed, resulting in the sequence UAGGGUC) as previously described (Fig. 5A) (30). Compared with the NC mimic treatment, overexpression of miR-30a-5p in ST cells decreased the activity of the luciferase reporter containing the SOCS1 or SOCS3 wild-type target sequence but not that of the luciferase reporter containing the mutant target site of SOCS1 or SOCS3. In contrast, compared with the NC inhibitor, the miR-30a-5p inhibitor increased the activity of the luciferase reporter containing the SOCS1 or SOCS3 wild-type target sequence but not that of the luciferase reporter containing the mutant target site (Fig. 5B). These results confirm that miR-30a-5p directly targets the 3′ UTRs of SOCS1 and SOCS3. To further verify that SOCS1 and SOCS3 are direct targets of miR-30a-5p, we examined the expression of SOCS1 and SOCS3 in ST cells transfected with the miR-30a-5p mimics or inhibitor. As expected, the miR-30a-5p mimics significantly decreased the transcript levels of SOCS1 and SOCS3 in ST cells (Fig. 5C). The diminished expression of SOCS1 and SOCS3 induced by miR-30a-5p overexpression was verified by Western blotting (Fig. 5D). Conversely, compared with the NC inhibitor, the miR-30a-5p inhibitor increased the expression of SOCS1 and SOCS3 in ST cells (Fig. 5C and D). The modulation of miR-30a-5p abundance exerted an effect on the expression of SOCS1 and SOCS3 in TGEV-infected ST cells similar to that in TGEV-uninfected ST cells (Fig. 5D). Altogether, these data demonstrate that miR-30a-5p downregulates the expression of SOCS1 and SOCS3 by directly targeting their 3′ UTRs.

TGEV infection downregulated miR-30a-5p expression through activation of IRE1α (Fig. 1 and 2). Reasonably, TGEV infection would upregulate the expression of SOCS1 and SOCS3. As expected, TGEV infection substantially elevated the expression of SOCS1 and SOCS3 in ST cells at 24 hpi and exhibited a dose-dependent induction of MOIs (Fig. 5E). Furthermore, the increased expression of SOCS1 and SOCS3 began at 12 hpi, and expression was substantially elevated at 24 hpi following TGEV infection, as observed by measuring both mRNA and protein levels, which were inversely correlated with the kinetic expression profiles of miR-30a-5p (Fig. 5F and G). To further validate the role of miR-30a-5p in manipulating SOCS1 and SOCS3 expression via IRE1α, we assessed the expression of SOCS1 and SOCS3 after TGEV infection in the presence of 100 μM 4μ8c pretreatment. Compared with the DMSO-pretreated mock-treated control, 4μ8c pretreatment strongly decreased the expression of SOCS1 and SOCS3 in ST cells following TGEV infection, and the reduction in SOCS1 and SOCS3 expression by 4μ8c was counteracted by the miR-30a-5p inhibitor (Fig. 5H), indicating that TGEV upregulates the expression of SOCS1 and SOCS3 through IRE1α-mediated modulation of miR-30a-5p expression. Consistent with the in vitro SOCS expression results, the expression of both SOCS1 (Fig. 5I) and SOCS3 (Fig. 5J) in the TGEV-infected ileum was also upregulated more than 10-fold. Taken together, these data demonstrate that TGEV infection upregulates the expression of SOCS1 and SOCS3 via the IRE1α-mediated modulation of miR-30a-5p abundance.

Increased expression of SOCS1 or SOCS3 disrupts the IFN-I antiviral response and facilitates TGEV replication.

Because miR-30a-5p hindered TGEV infection (Fig. 3) and enhanced IFN-I signaling (Fig. 4), next, we explored whether TGEV-derived suppression of miR-30a-5p facilitates TGEV infection via modulation of IFN-I signaling by SOCS1 and SOCS3. Initially, we analyzed the activation of the JAK-STAT1 pathway in response to IFN-I stimulation or TGEV infection of ST cells transfected with pCAGGS-SOCS1, pCAGGS-SOCS3, or the empty vector. The expression of SOCS1 or SOCS3 was confirmed by an IFA (Fig. 6A). As expected, the overexpression of SOCS1 or SOCS3 reduced p-STAT1 and the ratio of p-STAT1 to total STAT1 in response to either IFN-β stimulation or TGEV infection (Fig. 6B), indicating that the elevated expression of SOCS1 or SOCS3 suppresses the p-STAT1 induced by IFN-β or TGEV infection. Consistent with the disruption of IFN-I-activated JAK-STAT1 signaling by either SOCS1 or SOCS3, transient expression of SOCS1 substantially dampened the anti-TGEV activity of IFN-β and enhanced TGEV infection, as indicated by quantification of the TGEV genomes and the viral titers, and the transient expression of SOCS3 also dampened the anti-TGEV activity of IFN-β but significantly elevated the levels of TGEV genomes only, despite the trend of an increase in TGEV titers (Fig. 6C). Importantly, the overexpression of either SOCS1 or SOCS3 enhanced TGEV infection under physiological infection conditions in the absence of IFN-β stimulation (Fig. 6D). Overexpression of SOCS1 suppressed p-STAT1, IFN-β antiviral activity, and TGEV infection more efficiently than did overexpression of SOCS3 (Fig. 6B to D). These results demonstrate that the increased expression of SOCS1 and SOCS3 hinders the JAK-STAT1 signaling elicited by IFN-I or TGEV and enhances TGEV infection. To further verify the roles of SOCS1 and SOCS3 in IFN-I anti-TGEV activity, we silenced the expression of endogenous SOCS1 and SOCS3 using siRNAs. The knockdown efficiency of SOCS1 siRNAs (siSOCS1s) and SOCS3 siRNAs was confirmed by Western blotting (Fig. 6E). In contrast, knockdown of endogenous SOCS1 or SOCS3 by specific siRNAs in ST cells substantially enhanced the anti-TGEV activity of IFN-β (Fig. 6F) and the expression of IFN-I-induced ISGs (Fig. 6G). Overall, the silencing of endogenous SOCS1 and SOCS3 expression greatly decreased TGEV titers by up to 10-fold (Fig. 6H) and enhanced the expression of IFN-induced ISGs (Fig. 6I) in non-IFN-β-stimulated ST cells compared with those in siRNA NC-transfected cells. In agreement with SOCS1 and SOCS3 overexpression results, the silencing of SOCS1 enhanced the IFN-β-mediated antiviral response and hindered TGEV infection more efficiently than did the knockdown of SOCS3 (Fig. 6E to I), suggesting that SOCS1 has a more important role in IFN-I signaling than does SOCS3. Thus, these data collectively indicate that TGEV infection upregulates the expression of SOCS1 and SOCS3, which undermines JAK-STAT1 signaling and facilitates TGEV infection.

IRE1α facilitates TGEV infection by modulating the miR-30a-5p/SOCS axis.

We demonstrated above that TGEV-activated IRE1α decreased the levels of miR-30a-5p (Fig. 2) and that miR-30a-5p directly targeted SOCS family proteins and downregulated their expression, which enhanced STAT1 activation and hindered TGEV infection (Fig. 3 to 6). To further verify whether TGEV escapes the IFN-I response by disrupting the JAK-STAT1 pathway through IRE1α-mediated modulation of the miR-30a-5p/SOCS axis, we initially monitored TGEV infection after inhibiting STAT1 with fludarabine (a STAT1-specific inhibitor) in ST cells treated with 4μ8c. The suppression of IRE1α by 4μ8c promoted STAT1 activation, resulting in TGEV inhibition (Fig. 7A). The disruption of 4μ8c-mediated enhancement of STAT1 signaling by fludarabine significantly rescued TGEV suppression by 4μ8c (Fig. 7A). In agreement with these results, the overexpression of miR-30a-5p enhanced STAT1 activation and hindered TGEV infection, and fludarabine-mediated blockade of STAT1 activation significantly rescued the suppression of TGEV infection by miR-30a-5p (Fig. 7C). The inhibition of STAT1 activation by fludarabine was confirmed by Western blotting for p-STAT1 and total STAT1 (Fig. 7B and D). The rescue of 4μ8c- or miR-30a-5p-mediated TGEV inhibition by fludarabine treatment was not due to cellular cytotoxicity since treatment with 4μ8c, fludarabine, or 4μ8c plus fludarabine did not cause cellular cytotoxicity at the concentrations used in the study (Fig. 7E). These results suggest that IRE1α enhances TGEV infection through miR-30a-5p/SOCS-mediated disruption of the JAK-STAT1 pathway. Finally, to directly verify whether the effect of the miR-30a-5p/SOCS axis on TGEV infection is dependent on the IFN-I antiviral response, we assessed the effect of the suppression of miR-30a-5p on TGEV in ST cells after knockdown of the IFN-I receptor IFNAR1. The silencing efficacy of IFNAR1 siRNAs (siIFNAR1) was verified by measuring the transcripts of IFNAR (Fig. 7F). Among the 3 tested siRNAs, only siIFNAR1#1 substantially silenced IFNAR1 expression (Fig. 7F). The silencing of IFNAR1 by siIFNAR1#1 compared with that by the siRNA control was further confirmed by the reduction in ISG expression (ISG15, OASL, and MxA) after IFN-β stimulation (Fig. 7G). Unlike the siRNA control, no inhibitory effect of miR-30a-5p on TGEV was observed in IFNAR1-slienced cells (Fig. 7H). In line with the TGEV titer results, miR-30a-5p overexpression did not result in the enhancement of STAT1 activation in IFNAR1-silenced ST cells, whereas miR-30a-5p overexpression did in wild-type ST cells (Fig. 7I). These data demonstrate that miR-30a-5p overexpression suppresses TGEV infection via the IFN-I response. In summary, IRE1α enhances TGEV infection through miR-30a-5p/SOCS axis-mediated escape from the IFN-I response.

Cell culture, viruses, and virus infection.

ST cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Gibco, USA) supplemented with 10% heat-inactivated fetal bovine serum (FBS; Gibco, USA), 100 U/ml penicillin, and 100 μg/ml streptomycin in an incubator with 5% CO₂ at 37°C. TGEV strain H87, derived from the virulent strain H16 (GenBank accession number FJ755618), was propagated in ST cells. A TGEV stock was prepared and titrated as previously described (56). For TGEV infection, ST cells were infected with TGEV H87 at the desired MOI or mock infected with DMEM. After a 2-h incubation at 37°C, cells were washed and cultured in DMEM supplemented with 1% DMSO and 0.3% trypsin (0.25%; Gibco, USA) until harvested.

Synthetic miRNAs, siRNAs, and transfection.

All miRNA mimics, miRNA inhibitors, and siRNAs listed in Table 1 were designed and synthesized by GenePharma (China). Cells were seeded for 16 h before transfection with plasmid DNA or synthetic oligonucleotides using the Lipofectamine 2000 reagent (Invitrogen, USA) or the Lipofectamine RNAiMAX reagent (Invitrogen) according to the manufacturer's protocol. The cells were infected with TGEV or stimulated by IFN-β at 24 h after transfection. The cells were harvested for real-time reverse transcription (RT)-quantitative PCR (qPCR) or Western blotting after a 24-h incubation.

Chemical treatments.

Thapsigargin (Tg; a potent inducer of ER stress) and 4μ8c (a specific inhibitor of IRE1α) (14) were purchased from Sigma-Aldrich, and fludarabine (a STAT1-specific inhibitor) was purchased from Selleck. ST cells were pretreated with different concentrations of chemicals or DMSO for 2 or 24 h, followed by inoculation with TGEV. After a 2-h incubation, the supernatant was removed and replaced with culture medium containing different doses of chemicals. The samples were harvested for RT-qPCR and viral titration at 24 hpi.

RNA extraction, reverse transcription, and qPCR.

Total cellular RNA was extracted using the TRIzol reagent (Thermo Fisher Scientific, USA) according to the manufacturer's protocol. cDNA was prepared with a PrimeScript II 1st-strand cDNA synthesis kit (TaKaRa, Japan). The expression pattern of each gene was analyzed by RT-qPCR using a LightCycler 480 II system (Roche, Swiss) as previously described (56). For miRNA analyses, RNA was reverse transcribed by using an miRNA first-strand cDNA synthesis kit (Sangon Biotech, China), and miRNA expression was assessed by RT-qPCR. PCR amplification was performed in triplicate using the following conditions: 95°C for 10 min, followed by 40 cycles of two steps (95°C for 15 s and 60°C for 40 s). The relative expression levels of miRNAs were normalized to those of U6. The RNA expression results are presented as the means ± standard errors of the means (SEM) from 3 independent experiments. The sequences of the RT-qPCR primers are listed in Table 2.

microRNA target prediction and plasmid construction.

The TargetScan web server (version 7.1; http://www.targetscan.org) was used to predict SOCS1 and SOCS3 3′ UTRs as potential targets of miR-30a-5p. The 3′ UTR of porcine SOCS1 (GenBank accession number GQ421919.1) or SOCS3 (GenBank accession number AY785557.1) was amplified and inserted into the pmirGLO luciferase reporter vector (Promega, USA) using the NheI and XbaI restriction sites. The mutant types of SOCS1 or SOCS3 3′ UTR vectors were constructed by mutating five seed nucleotides using a site-directed mutagenesis kit (Stratagene, USA) according to the manufacturer's instructions. To construct the SOCS1 and SOCS3 expression vectors, we amplified the full-length coding DNA sequence regions by specific primers, and the amplicons were cloned into the vector pCAGGS-HA (Addgene) using the EcoRI and KpnI restriction sites. The porcine ISRE reporter plasmid was kindly provided by Wenhai Feng from China Agricultural University and was previously described (30). All primers are listed in Table 2.

Dual-luciferase reporter assays.

For miRNA target verification, wild-type or mutant SOCS1 and SOCS3 3′ UTR luciferase reporter vectors were cotransfected with miR-30a-5p mimics, NC mimic, miR-30a-5p inhibitor, or NC inhibitor (NC-i) into ST cells for 36 h. Furthermore, the ISRE reporter plasmid and pRL-TK were cotransfected with either miR-30a-5p mimics, NC mimic, inhibitor, or NC inhibitor for 24 h, followed by incubation with porcine IFN-β (100 ng/ml) for 12 h. Next, the cells were collected for reporter activity testing with a dual-luciferase reporter assay system (Promega, USA) according to the manufacturer's protocol. All reporter assays were independently repeated at least three times.

IFAs.

Immunofluorescence assay (IFAs) were performed as previously described (57). Briefly, ST cells were fixed with 4% paraformaldehyde and permeabilized with 0.2% Triton X-100 and then blocked with blocking buffer for 2 h at 37°C. Cells were incubated with an anti-TGEV nucleocapsid monoclonal antibody (1:1,000) stocked in the State Key Laboratory of Veterinary Biotechnology or an antihemagglutinin (anti-HA) monoclonal antibody (1:5,000; Sigma-Aldrich,) at 37°C for 2 h. The cells were then labeled with an Alexa Fluor 546 goat anti-mouse immunoglobulin G (IgG) antibody (1:500; Thermo Fisher Scientific, USA) for 1 h at 37°C. Cellular nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI; 1:100). Stained cells were visualized using an AMG EVOS F1 florescence microscope.

ELISA.

Culture supernatants were collected from treated cells at the times indicated above and stored at −80°C until testing. IFN-β protein levels in the culture supernatants were determined by the use of porcine IFN-β enzyme-linked immunosorbent assay (ELISA) kits (Bio-Swamp, China) according to the manufacturer's instructions.

Protein extraction and Western blotting.

To analyze the levels of host IRE1α, phospho-STAT1, STAT1, SOCS1, and SOCS3 proteins, we analyzed the cellular lysate by Western blotting as previously described (56). Briefly, ST cells were lysed with Nonidet P-40 (NP-40) lysis buffer (Beyotime, China) supplemented with a protease inhibitor cocktail (Roche). The protein concentrations of the supernatant fraction were measured with a bicinchoninic acid (BCA) assay kit (Beyotime, China) and were equalized with the extraction reagent. Equal amounts of proteins were separated on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gels and transferred onto a nitrocellulose membrane (GE Healthcare, USA). Membranes were blocked and incubated with the corresponding primary antibody at 4°C overnight. The primary antibodies used were as follows: β-actin (1:5,000; Sigma-Aldrich), SOCS1 (1:500; Sigma-Aldrich), SOCS3 (1:1,000; Cell Signaling Technology), phospho-STAT1 (1:1,000; Abcam), STAT1 (1:1,000; Abcam), and IRE1α (1:500; Santa Cruz Biotechnology). After washing, the membrane was incubated with horseradish peroxidase (HRP)-conjugated anti-mouse IgG or anti-rabbit IgG for 1 h at room temperature. Then, immunolabeled proteins were visualized using an electrochemiluminescence (ECL) reagent (Thermo Fisher Scientific). The intensity of each band was analyzed using ImageJ software.

TGEV infection of piglets.

Twelve 2-day-old specific-pathogen-free (SPF) piglets were randomly divided into two groups. The piglets in group 1 were orally inoculated with 5 ml 1 × 10⁵ 50% tissue culture infective doses (TCID₅₀) of the TGEV H87 strain. After viral infection, clinical signs were recorded on a daily basis. All piglets were euthanized at 48 hpi. Piglet small intestine samples were collected for RT-qPCR analyses. The TGEV infection experiment was approved by the Animal Care and Ethics Committee of the Harbin Veterinary Research Institute.

Cellular cytotoxicity assay.

ST cells were cultured in 96-well plates for 24 h, followed by incubation with Tg, 4μ8c, fludarabine, 4μ8c plus fludarabine, or the same volume of DMSO for 24 h or 48 h. Next, cellular cytotoxicity assays were performed using a cell counting kit 8 (CCK-8) assay as previously described (56).

Statistical analysis.

All results in the figures are presented, where appropriate, as the means ± SEM from three independent experiments and were analyzed in GraphPad Prism software (GraphPad Software, Inc.). Differences were considered significant if the P value was <0.05.