PEDV evades MHC-I-related immunity through nsp1-mediated NLRC5 translation inhibition

ABSTRACT

Major histocompatibility complex class I (MHC-I) plays crucial roles against viral infections not only by initiating CD8⁺ T cell immunity but also by modulating natural killer (NK) cell cytotoxicity. Understanding how viruses precisely regulate MHC-I to optimize their infection is important; however, the manipulation of MHC-I molecules by porcine epidemic diarrhea virus (PEDV) remains unclear. In this study, we demonstrate that PEDV infection promotes the transcription of NLRC5, a key transactivator of MHC-I, in several porcine cell lines and in vivo. Paradoxically, no increase in MHC-I expression is observed after PEDV infection both in vitro and in vivo. Mechanistic studies revealed that PEDV infection inhibits the translation of PEDV-elicited NLRC5 mRNA and the expression of downstream MHC-I proteins, without affecting the expression of physiological NLRC5 and MHC-I proteins. Through viral protein screening, we identified PEDV nonstructural protein 1 (nsp1) as the critical antagonist that inhibits NLRC5-mediated upregulation of MHC-I, and the nsp1’s inhibitory effect on MHC-I requires the motif of 15 amino acids at its C-terminus. Notably, our results revealed that the cytotoxic ability of NK cells against PEDV-infected cells is similar to that against healthy cells. Collectively, our findings uncover an immune evasion mechanism by which PEDV-infected cells masquerade as healthy cells to evade NK and T cell immunity. This is achieved by targeting NLRC5, a key MHC-I transcriptional regulator, via nsp1.

IMPORTANCE

Porcine epidemic diarrhea virus (PEDV) is a highly contagious enteric coronavirus that inflicts substantial financial losses on the swine industry. Major histocompatibility complex class I (MHC-I) is a critical factor influencing both CD8⁺ T cell and natural killer (NK) cell immunity. However, how PEDV manipulates MHC-I expression to optimize its infection process remains largely unknown. In this study, we demonstrate that PEDV’s nonstructural protein 1 (nsp1) inhibits virus-mediated induction of MHC-I expression by directly targeting NLRC5, a key MHC-I transactivator. Intriguingly, nsp1 does not reduce physiological NLRC5 and MHC-I expression. This selective inhibition of virus-elicited NLRC5 mRNA translation allows PEDV-infected cells to masquerade as healthy cells, thereby evading CD8⁺ T cell and NK cell cytotoxicity. Our findings provide unique insights into the mechanisms by which PEDV evades CD8⁺ T cell and NK cell immunity.

INTRODUCTION

Porcine epidemic diarrhea virus (PEDV), an alphacoronavirus, is one of the major etiologies that is responsible for the acute, highly contagious disease in the digestive tract of pigs and mainly infects small intestinal villus epithelial cells, leading to a mortality rate of about 80% to 100% in neonatal piglets (1, 2). PEDV was first reported in England and Belgium in the early 1970s with a variant strain emerging in 2010 (3, 4). In the winter of 2010, highly virulent strains of PEDV emerged in China and spread rapidly across the country, resulting in significant economic losses due to its high morbidity and mortality rate (2, 5). PEDV was first reported in the United States in May 2013 and caused the deaths of more than 7 million pigs in the United States during the 2013–2014 outbreak (6). Subsequently, PEDV epidemics have been reported in Peru, Dominica, Colombia, Canada, Ecuador, and other countries. Given the significant threat that PEDV poses to the global pork industry, it is urgent to enhance our understanding of the interaction between the host immune response and PEDV infection.

MHC-I, an essential host molecule that participates in antigen presentation, plays a critical role in presenting intracellular antigens and initializing CD8⁺ T cellular immunity to clear virus-infected cells. MHC-I is composed of human leukocyte antigen (HLA)-A/B/C and β2 microglobulin (β2M) and is expressed on the surface of nucleated cells (7). During viral infection, endogenous viral protein antigens are cleaved into small peptides by proteasomes in the cytosol and transported across the endoplasmic reticulum (ER) membrane with the help of a transporter associated with antigen processing (TAP). Certain peptides, typically 8–11 amino acids in length, are selected to fill the MHC-I binding groove to form the MHC-I-peptide-loading complex and are then presented on the cell surface through a secretory pathway, resulting in the specific recognition of the infected cells by cytotoxic T cells (CTL) (8–10). Many chronic infectious viruses have evolved distinct mechanisms to escape CD8⁺ T cellular immunity by blocking the function of the MHC-I pathway or decreasing MHC-I expression on the cell surface. For instance, herpes simplex virus (HSV), Epstein-Barr virus (EBV), and human cytomegalovirus (HCMV) interfere with antigen presentation by blocking the function of antigen presentation-related genes in the MHC-I pathway (11–13). Human immunodeficiency virus (HIV)−1 Nef and adenovirus E1A lead to the downregulation of MHC-I (14, 15).

MHC-I molecules are critical for the acquisition of functional effector natural killer (NK) cells via their interaction with their specific inhibitory receptors on the NK cells. NK cells, a third major type of lymphocyte, do not express rearranged, antigen-specific receptors and thus are classified as an important component of innate cellular immunity (16). NK cells play substantial roles in controlling virus infection, especially acute virus infections. Although viruses can evade CD8⁺ T cellular immunity by reducing the MHC-I expression, the reduced MHC-I, as an inhibitory ligand for NK cells, promotes the activation of NK cells, resulting in the killing of virus-infected cells (17). We recently reported that PDCoV infection substantially upregulated the expression of porcine MHC-I (18), allowing the virus to escape NK cellular killing. Thus, the precise regulation of MHC-I molecules is critical for viruses to evade NK and CD8⁺ T cellular immunity. However, it remains unclear how PEDV infection modulates the MHC-I pathway.

PEDV is an enveloped virus with a genome size of about 28 kb, consisting of a 5’ cap and a 3’ polyadenylated tail and containing more than seven open reading frames (ORFs), including ORF1a and ORF1b, followed by the spike, ORF3, envelope, membrane, and nucleocapsid (4, 19). ORF1a and ORF1b are processed into 16 nonstructural proteins (nsp1-nsp16) by the proteinase activities of nsp3 and nsp5, and these processed nonstructural proteins promote virus replication in direct or indirect manners. The structure and function of nonstructural proteins are generally relatively conserved among different coronaviruses (CoVs), indicating their crucial roles in the viral life cycle. Among these nonstructural proteins, nsp1, located at the N-terminal of the replicase polyprotein pp1a, is the first protein encoded by the coronavirus genomic RNA. Despite being present exclusively in α-CoVs and β- CoVs, the nsp1 protein of α-CoVs exhibits a low degree of sequence homology compared with that of β-CoVs (20). As a key pathogenicity factor of CoVs, nsp1 hijacks the host translation system to synthesize viral proteins through multiple mechanisms. Inhibiting the expression of antiviral genes by viral infection is crucial for suppressing the host immune response and promoting viral replication (21). It has been reported that the nsp1 of many coronavirus strains can inhibit type I interferon (IFN-I) production and function in various ways (20, 21). SARS-CoV nsp1 impedes IFN induction by inhibiting IRF3 phosphorylation and decreasing the expression of Tyk2 and STAT2 to dampen ISG induction (22). In the case of SARS-CoV-2, the nsp1 mutant strain can modulate the IFN-I response (23). PEDV, an acute porcine diarrheal virus, inhibits IFN production by suppressing NF-κB activation and promoting CREB degradation (24). A recent study showed that the PEDV nsp1 N93/95A mutant strain is more sensitive to IFN but retains viral immunogenicity, making it a potential target for vaccine development (25). Although these studies demonstrate the relationship between PEDV nsp1 and innate immunity, the role of PEDV nsp1 in adaptive immunity remains unclear.

In this study, we found that PEDV infection upregulates the transcription of NLRC5, a key transactivator of MHC-I, whereas PEDV infection does not induce MHC-I expression both in vitro and in vivo. Mechanistically, we revealed that although NLRC5 transcription is induced during PEDV infection, the translation of NLRC5 protein is dampened by PEDV infection. Further research demonstrated that PEDV nsp1 is the critical antagonist of MHC-I expression by directly targeting NLRC5 translation. Interestingly, PEDV infection does not reduce the expression of physiological NLRC5 and MHC-I proteins, which may allow PEDV-infected cells to masquerade as normal cells, thus evading MHC-I mediated NK and CD8⁺ T cell immune responses. In summary, our findings uncover an immune evasion mechanism utilized by PEDV to escape the MHC-I mediated NK and CD8⁺ T cell immune responses, which may provide a novel potential strategy for coronavirus vaccine design.

RESULTS

PEDV infection promotes the transcription of NLRC5

NOD-like receptor (NLR) family caspase recruitment domain-containing 5 (NLRC5) is a key transactivator that specifically promotes the transcription of MHC-I. Our previous study demonstrated that PDCoV infection can promote the expression of MHC-I molecules by upregulating NLRC5 expression, indicating that NLRC5 also plays an important role in the manipulation of MHC-I expression during coronavirus infection (18). However, whether PEDV, also a porcine diarrhea coronavirus, regulates MHC-I expression is still unknown. First, we revalidated the regulatory effect of NLRC5 on MHC-I expression in porcine small intestinal epithelial cells (IPI-2I). Consistent with our previous data, NLRC5 transfection resulted in significant upregulation of porcine MHC-I (known as SLA-I in porcine) at both the transcription and protein levels (Fig. 1A and B). To determine the expression level of NLRC5 during PEDV infection, high-throughput RNA sequencing (RNA-seq) was performed as previously described(26). The data showed that PEDV infection significantly upregulated the transcription of NLRC5 in IPEC-J2 cells (Fig. 1C) and porcine intestinal enteroids (Fig. 1D). The enhanced transcription of NLRC5 in IPEC-J2 (Fig. 1E), IPI-2I cells, and ileum enteriods (Fig. 1F) after PEDV infection was further confirmed by reverse transcriptase quantitative PCR (RT-qPCR). Compared with the mock infection control, PEDV infection upregulated the mRNA level of NLRC5 in the above-mentioned cell lines, which was consistent with RNA-seq analysis. To further validate whether PEDV infection directly promotes NLRC5 transcription, the luciferase assay was used to monitor the luciferase activity of NLRC5 promoter following PEDV infection in the IPI-2I cell line (Fig. 1G). PEDV infection promoted the luciferase activity of the NLRC5 promoter compared with mock infection control, suggesting that PEDV infection directly enhances NLRC5 transcription. To further confirm whether PEDV infection also upregulates the transcription of NLRC5 in vivo, we subsequently assessed the NLRC5 mRNA in the jejunum, which is the primary target of PEDV infection (Fig. 1H). Consistent with the in vitro data, PEDV infection also upregulated NLRC5 transcription in vivo. Collectively, these data indicate that PEDV infection promotes NLRC5 transcription both in vitro and in vivo.

Fig 1.

PEDV infection promotes NLRC5 transcription. (A and B) IPI-2I cells at 80% confluence were transfected with NLRC5 expression vector or empty vector for 36 h. The relative mRNA expression levels of SLA-1, SLA-2, and SLA-3 were evaluated by RT-qPCR, (A) and the expression level of cell surface SLA-I expression was determined by flow cytometry (B). The mean fluorescence intensity was calculated using FlowJo software. (C and D) Monolayers of IPEC-J2 (C) and jejunum enteroids (D) isolated from specific pathogen-free (SPF) 2-day-old piglets (n = 3 piglets) were infected with PEDV at an MOI of 2. The cells were harvested for RNA-seq at 24 hpi, and the analysis of NLRC5 transcripts was shown as FPKM (fragments per kilobase of transcript per million mapped reads) values. (E and F) IPEC-J2 (E), Ileum enteroids, and IPI-2I cells (F) were infected with PEDV for 24 h, and the relative mRNA expression level of NLRC5 was determined by RT-qPCR. (G) IPI-2I cells at 80% confluence were co-transfected with NLRC5 promoter reporter plasmid and control plasmid pRL-TK for 12 h and infected with PEDV-JMS at an MOI of 1. The cells were harvested for luciferase assay at 24 hpi. (H) SPF piglets were inoculated orally with 1 mL of a solution containing 1 × 10³ TCID₅₀ PEDV and euthanized at 72 hpi. The expression level of NLRC5 in jejunum was determined by RT-qPCR.

PEDV infection does not induce the expression of MHC-I molecules

Our previous studies have demonstrated that PDCoV and transmissible gastroenteritis virus (TGEV) infection sharply promote MHC-I expression in vitro and in vivo (18, 26). To determine whether PEDV infection modulates the MHC-I antigen presentation pathway, we monitored the expression kinetics of porcine MHC-I (known as SLA-I in porcine) during PEDV infection at different time points in IPEC-J2 cells at a multiplicity of infection (MOI) of 1. Unexpectedly, unlike PDCoV and TGEV, PEDV infection did not upregulate the expression of SLA-1, SLA-2, and SLA-3 at any time points, although PEDV infection substantially increased NLRC5 transcription (Fig. 2A through C). Furthermore, we also examined the expression of antigen presentation-related genes TAP1 and β2M after PEDV infection (Fig. 2D and E). Consistent with the pattern of SLA-I expression after PEDV infection, PEDV infection did not upregulate the expressions of TAP1 and β2M. Additionally, the SLA-I expression levels were evaluated in porcine ileum enteroids after PEDV infection, and the results suggested that PEDV infection also did not induce SLA-I expression in enteroids (Fig. 2F through H). The function of MHC-I molecules to mediate CD8⁺ T and NK cellular immunity requires its expression on the cell membrane. Therefore, we next assessed the surface expression of SLA-I at different time points after PEDV infection using flow cytometry. Compared with the mock infection control, PEDV infection did not alter the surface expression of SLA-I at any time points post-infection (Fig. 2I). No alteration of surface SLA-I expression by PEDV infection when compared with the mock infection control is regardless of MOIs (Fig. 2J). Consistent with the in vitro data, PEDV infection in vivo did not upregulate SLA-I expression in jejunal epithelial cells, the primary cellular targets of PEDV via immunohistochemical staining using an antibody against SLA-I (Fig. 2K). Taken together, our results suggest that PEDV infection did not alter SLA-I expression both in vitro and in vivo.

Fig 2.

PEDV infection does not increase the expression of MHC-I in vitro and in vivo. (A–E) One hundred percent of monolayer confluent IPEC-J2 cells was infected with the PEDV-JMS strain at an MOI of 0.5. The cells were harvested at indicated time points, and the relative mRNA expression levels of SLA-1 (A), SLA-2 (B), SLA-3 (C), TAP1 (D), and β2M (E) were determined by RT-qPCR. (F–H) Ileum enteroids were infected with the PEDV-JMS strain at an MOI of 0.5. The cells were harvested at indicated time points, and the relative mRNA expression levels of SLA-1 (F), SLA-2 (G), and SLA-3 (H) were determined by RT-qPCR. (I) IPI-2I cells at 100% confluence were infected with PEDV-JMS strain at an MOI of 0.5, and the cells were harvested at different time points. The surface SLA-I expression level was determined by flow cytometry. (J) IPI-2I cells at 100% confluence were infected with the PEDV-JMS strain at different MOIs, and the cells were harvested at 24 hpi. The surface SLA-I expression level was determined by flow cytometry. (K) SPF piglets were inoculated orally with 1 mL of a solution containing 1 × 10³ TCID₅₀ PEDV and euthanized at 72 hpi. The expression level of SLA-I in jejunum was determined by immunohistochemistry assay.

PEDV infection represses the upregulation of NLRC5 protein expression

PEDV infection promotes NLRC5 expression at the mRNA level, but the expression level of SLA-I remains unchanged. This may be due to the suppression of NLRC5 protein translation or function by PEDV infection. To explore whether PEDV infection affects NLRC5 protein expression, we first evaluated the expression changes of exogenous NLRC5 by indirect immunofluorescence assay (IFA) after PEDV infection in IPI-2I cells transfected with NLRC5 expression vector. Compared with the mock infection control, PEDV infection significantly inhibited exogenous NLRC5 protein expression, which was exhibited in a dose-dependent manner (Fig. 3A). Moreover, in order to examine the inhibitory role of PEDV infection in NLRC5-mediated SLA-I induction, we also measured total cellular and surficial SLA-I expression in IPI-2I cells transfected with the NLRC5 expression vector after PEDV infection at different MOIs. Compared with the empty vector group, transfection with NLRC5 expression vector significantly upregulates total cellular SLA-I expression, approximately 12% of the total cells showed increased SLA-I expression. However, total cellular SLA-I expression showed a significant decrease when NLRC5-transfected cells were infected with PEDV, and the mean fluorescence intensity decreased to the level of the empty vector-transfected group (Fig. 3B and D). Consistent with the results of total cellular SLA-I expression, the reduction of surficial SLA-I expression in NLRC5 transfected cells by PEDV infection displayed a MOI dose-dependent response (Fig. 3C and E). Collectively, our results show that PEDV infection blocked the exogenous NLRC5-mediated induction of SLA-I expression.

Fig 3.

PEDV infection represses the upregulation of NLRC5 protein expression. (A) IPI-2I cells at 80% confluence were transfected with NLRC5 expression vector or empty vector for 12 h and infected with PEDV-JMS at an MOI of 2 and 5. The cells were harvested for an IFA at 24 hpi. The NLRC5 expression level was determined by using an anti-HA antibody (green), and the PEDV was determined by using an anti-PEDV-N antibody (red). Nuclei were stained using DAPI shown in blue. (B–E) IPI-2I cells at 80% confluence were transfected with NLRC5 expression vector or empty vector for 12 h and infected with PEDV-JMS at different MOIs. The cells were harvested at 24 hpi and incubated with anti-SLA-I antibody directly for cell surface SLA-I expression detection (B) or incubated with anti-SLA-I antibody after fixation and permeabilization for total cellular SLA-I expression detection (C). The mean fluorescence intensity of total cellular SLA-I (D) and cell surface SLA-I (E) were calculated using FlowJo.

PEDV nsp1 is the critical antagonist of MHC-I expression

Coronavirus proteins are composed of nonstructural proteins and structural proteins. To investigate which viral proteins mediate the inhibition of NLRC5-induced MHC-I expression after PEDV infection, SLA-I expression was analyzed by flow cytometry after IPI-2I cells were co-transfected with the NLRC5 expression vector, and the recombinant plasmids encoding nonstructural or structure proteins separately, or along with the pCAGGS-HA empty vector (Fig. 4). Among these nonstructural proteins of PEDV-JMS strain, the nsp1 protein exhibited the strongest ability to reduce NLRC5-induced SLA-I expression, whereas nsp14 and nsp15 also inhibited NLRC5-induced SLA-I to a lesser extent. Nsp7 and nsp16 had no effect on SLA-I expression, whereas nsp4, nsp5, nsp6, nsp8, and nsp10 enhanced MHC-I expression to varying degrees (Fig. 4A and B). None of the structural proteins significantly influenced NLRC5-induced MHC-I expression (Fig. 4C and D). The protein expression of constructed recombinant plasmids of PEDV structural and nonstructural proteins was confirmed by IFA (Fig. 4E). These results indicate that PEDV nsp1 is the key antagonist in reducing NLRC5-induced MHC-I expression.

Fig 4.

PEDV nsp1 inhibits NLRC5-induced MHC-I expression. (A and C) IPI-2I cells at 80% confluence were co-transfected with NLRC5 expression vector and PEDV nonstructural proteins (A) or structural proteins (C) expression vectors, respectively, for 36 h. The cells were then permeabilized and incubated with anti-SLA-I antibodies at room temperature for the detection of total cellular SLA-I expression. (B and D) The mean fluorescence intensity of total cellular SLA-I of co-transfected cells was calculated by using FlowJo. The dashed line represents the SLA-I expression level of cells co-transfected with NLRC5 and empty vector. (E) 293T cells at 80% confluence were transfected with PEDV non-structural proteins or structural proteins, respectively, for 36 h. The expression levels of these viral proteins were assessed at 36 hpt by IFA.

PEDV nsp1 inhibits the induction of the MHC-I by targeting NLRC5

Previous studies have reported that coronavirus nsp1 assists viruses in escaping the host immune response by inhibiting the protein translation of host antiviral factors (21). It was of interest to investigate whether PEDV nsp1 inhibits the induction of SLA-I by regulating NLRC5 protein expression. In line with the total cellular SLA-I results (Fig. 4), the transient expression of nsp1 almost completely suppressed transient NLRC5-induced cell surface SLA-I (Fig. 5A). To further determine whether the SLA-I inhibition by nsp1 was specific to IPI-2I cells, swine testis (ST) cells were also used to verify this phenomenon. Consistent with IPI-2I cells, overexpression of NLRC5 in ST cells significantly promoted SLA-I expression, whereas co-transfection with nsp1 dramatically reduced SLA-I expression to the level of empty vector control group (Fig. 5B). The transcriptional expression of SLA-I was also determined by RT-qPCR in cells that co-transfected with nsp1 and NLRC5. The results suggest that nsp1 also reduced NLRC5-induced SLA-I expression at the mRNA level, indicating that nsp1 may repress transactivators of SLA-I transcription to inhibit the induction of SLA-I (Fig. 5C). Further research found that co-transfection of NLRC5 with nsp1 significantly attenuated the intensity of green fluorescence, indicating that nsp1 decreased NLRC5 expression at the protein level (Fig. 5D). To investigate whether PEDV nsp1 also inhibits endogenous SLA-I expression, the PEDV nsp1 recombinant expression vector was transfected into IPI-2I cells. As shown in Fig. 5E, nsp1 had no effect on endogenous SLA-I expression when compared with the empty vector control. This is in line with the endogenous NLRC5 expression in the cells transfected with nsp1 (Fig. 5F). The transient overexpression of nsp1 also did not affect physiological NLRC5 expression, indicating that nsp1 only inhibits the “overexpression” of NLRC5, maintaining its expression at a normal state. Meanwhile, we also assessed the expression level of endogenous NLRC5 protein following PEDV infection with different MOIs (Fig. 5G). Unlike NLRC5 transcription (Fig. 1), PEDV infection did not induce NLRC5 protein expression. These results indicated that PEDV infection does not downregulate the expressions of physiological NLRC5 and SLA-I but has an inhibitory effect on exogenously induced SLA-I, thus maintaining the expression of SLA-I at normal level to evade NK and CD8⁺ T cell cytotoxicity.

Fig 5.

PEDV nsp1 inhibits NLRC5-mediated induction of MHC-I expression. (A) IPI-2I cells at 80% confluence were co-transfected with the NLRC5 expression vector and PEDV nsp1 expression vector for 36 h. The cells were then incubated with anti-SLA-I antibody at 4°C to detect cell surface SLA-I expression. The mean fluorescence intensity of cell surface SLA-I expression was calculated by using FlowJo. (B) ST cells at 80% confluence were co-transfected with the NLRC5 expression vector and PEDV nsp1 expression vector for 36 h. The cells were then permeabilized and incubated with anti-SLA-I antibodies at room temperature to detect total cellular SLA-I expression. The mean fluorescence intensity of the total cellular SLA-I expression was calculated by using FlowJo. (C and D) IPI-2I or 293T cells were co-transfected with the NLRC5 expression vector and PEDV nsp1 expression vector for 36 h. The cells were harvested at 36 h post-transfection for detection of SLA-1, SLA-2, and SLA-3 relative mRNA expression by RT-qPCR (C) or detection of NLRC5 expression by IFA (D). The NLRC5 expression level was determined using an anti-HA antibody (green), and the expression level of nsp1 was determined by using an anti-Flag antibody (red). Nuclei were stained using DAPI and are shown in blue. (E) IPI-2I cells were transfected with PEDV nsp1 expression vector or empty vector for 36 h. The cells were permeabilized and incubated with anti-SLA-I antibody at 4°C to detect total cellular SLA-I expression. (F) IPI-2I cells were transfected with nsp1 for 36 h. The endogenous NLRC5 protein expression level was determined by western blotting. (G) IPI-2I cells at 100% confluence were infected with PEDV at different MOIs and were harvested at 24 hpi. The expression level of NLRC5 protein was determined by western blotting.

PEDV nsp1 96-110aa is the critical region for inhibiting the induction of MHC-I

Three motifs (amino acids [aa] 67–71, 78–85, and 103–110) of PEDV nsp1 create a stable functional region critical for inhibiting host protein synthesis, differing considerably from β-CoVs nsp1 (27). To identify which region in PEDV nsp1 is essential for its suppression of NLRC5-induced SLA-I expression, a series of truncated nsp1 plasmids were designed (Fig. 6A). First, NLRC5 expression vector was transfected into IPI-2I cells with an empty vector, nsp1 (1–60 aa), nsp1 (1–85 aa), nsp1 (1–95 aa), nsp1 (1–97 aa), nsp1 (1–102 aa), or nsp1 (1–11 0aa), respectively, and the total cellular MHC-I expression was determined by flow cytometry (Fig. 6B). The results showed that co-transfection of NLRC5 with nsp1 (1–97 aa), nsp1 (1–102 aa), and nsp1 (1–110 aa) attenuated NLRC5-mediated MHC-I expression to varying degrees, whereas nsp1 (1–60 aa), nsp1 (1–85 aa), and nsp1 (1–95 aa) had no effect on NLRC5-mediated MHC-I expression (Fig. 6C). Next, we investigated whether C-terminal motif (96–110 aa) plays a critical role in the inhibition of NLRC5 expression. To this end, NLRC5 expression vector was co-transfected with nsp1 or its truncations in 293T cells, and the expression level of NLRC5 was determined by IFA (Fig. 6D). Co-transfection with nsp1 (1–97 aa) and nsp1 (1–110 aa) significantly attenuated the intensity of green fluorescence, whereas nsp1 (1–95 aa) did not affect NLRC5 expression, indicating that the region from aa 96 to 110 in PEDV nsp1 is the critical motif in decreasing NLRC5 protein expression. Collectively, our data demonstrated that the region from aa 96 to 110 in PEDV nsp1 is responsible for inhibiting the induction of MHC-I by directly targeting NLRC5.

Fig 6.

PEDV nsp1 (amino acids 96–110) is the critical motif in inhibiting the induction of MHC-I. (A) The C-terminal truncations of PEDV nsp1 are shown below the bars. (B–D) IPI-2I cells at 80% confluence were co-transfected with NLRC5 expression vector and nsp1 or truncated plasmids. The cells were harvested at 36 h post-transfection, and the total cellular expression level of SLA-I was determined using flow cytometry (B). The mean fluorescence intensity of total cellular SLA-I expression was calculated using FlowJo (C). Transfected cells were harvested at 36 h post-transfection, and the expression level of NLRC5 was determined by IFA (D). The NLRC5 expression level was determined by using anti-HA antibody (green), and the nsp1 expression level was determined by using anti-Flag antibody (red). Nuclei were stained using DAPI and are shown in blue.

PEDV-infected cells masquerade as healthy cells to avoid NK cell immunity

NK cells are cytotoxic innate-like lymphocytes and play critical roles in controlling virus infection in the host, especially acute virus infection. A key feature of NK cell self or non-self recognition is to recognize the expression of MHC-I molecules on target cells. NK cells preferentially kill MHC I-deficient virus-infected cells, whereas the high levels of MHC-I expression are better protected from NK cell killing. Although the expression of the inhibitory ligand MHC-I remains at physiological levels during PEDV infection, it is still uncertain whether PEDV-infected cells can avoid NK cell-mediated killing. To investigate the impact of PEDV infection on the cytotoxic ability of NK cells toward target cells, we conducted an NK cell-mediated cytotoxicity assay following methodologies outlined in prior studies (28). Peripheral blood mononuclear cells (PBMCs) were stimulated with porcine IL-2 for 16 h to activate NK cells, which were subsequently co-cultured with the target cells. The death of the target cells was assessed by flow cytometry after 4–6 h of co-culture (Fig. 7A). The results showed that the killing rate of effector cells against PEDV-infected target cells ranged from 8% to 10%, which was comparable with the killing rate of mock-infected cells, with no statistically significant difference (Fig. 7B). This result suggests that PEDV-infected target cells can masquerade as healthy cells and avoid NK cell-mediated killing by inhibiting the upregulation of MHC-I.

Fig 7.

PEDV-infected cells masquerade as healthy cells to avoid NK cell immunity. (A and B) PBMCs isolated from healthy pigs were stimulated with porcine IL-2 for 16 h to activate NK cells, which were subsequently co-cultured with the target cells. After 4–6 h of co-culture, flow cytometry was performed to detect the mortality of target cells (A), and NK cell-mediated cytotoxicity was calculated as described in Materials and Methods (B).

DISCUSSION

PEDV infection leads to considerable mortality for newborn piglets and huge economic losses for the pork industry (2). MHC-I molecules, which actively participate in CD8⁺ T and NK cellular immunity, are critical for host antiviral immune response. Recent studies have found that SARS-CoV-2 and Middle East Respiratory Syndrome CoV (MERS-CoV) decrease the expression of MHC-I and antigen presentation-related genes, thus impeding CD8⁺ T cellular immunity and facilitating the production of progeny viruses without being eliminated (29, 30). However, whether PEDV infection modulates the expression of MHC-I molecules is still elusive and poorly understood. In this study, we found that the induction of MHC-I in porcine intestinal epithelium cells was impaired during PEDV infection due to the inhibition of the translation of NLRC5 mRNA. Mechanistically, we found that the nsp1 of PEDV inhibits the induction of MHC-I expression by directly targeting NLRC5 while not reducing physiological NLRC5 and MHC-I expression, thereby disguising it as healthy cells and avoiding NK and CD8⁺ T cell cytotoxicity. Our results reveal an unappreciated mechanism employed by PEDV to avoid the MHC-I mediated immune response by directly targeting NLRC5, via nsp1.

MHC-I, being responsible for the presentation of intracellular antigens, plays a critical role in initializing CD8⁺ T cellular immunity to kill virus-infected cells (10). To impair the induction of the antiviral CD8⁺ T immunity, many viruses have evolved multiple strategies to interfere with MHC-I antigen presentation, primarily through three pathways: downregulating the MHC-I expression, blocking the assembly of antigen peptides with MHC-I molecules, and preventing the display of antigen peptides on the cell membrane surface (31). The Epstein-Barr virus (EBV) utilizes the BILF3 protein to escape immune recognition by inducing ubiquitination and downregulation of MHC-I and MHC-II (32). Additionally, the adenovirus E19 protein can directly interact with MHC-I molecules, resulting in MHC-I retention in the ER lumen (33). Moreover, some viruses specifically target MHC-I internalization and degradation to achieve immune escape. Kaposi’s sarcoma-associated herpesvirus (KSHV) encodes K3 and K5 zinc finger membrane proteins, which promotes MHC-I disappearance from the cell membrane surface by inducing endocytosis, without affecting the transport and expression of MHC-I (34). Regarding coronavirus infection, some studies have shown that SARS-CoV-2 infection downregulates MHC-I expression, whereas MHV infection upregulates the expression of MHC-I molecules in astrocytic cells. Our previous studies demonstrated that PDCoV infection robustly upregulates MHC-I expression by promoting NLRC5 expression, which is activated by the RIG-I-mediated IFN signaling pathway and IRF1 expression (18). Meanwhile, another swine enteric virus, TGEV, also sharply promotes MHC-I expression, although the detailed mechanism remains unclear (26). However, PEDV, which is also an α-CoV like TGEV, did not modulate MHC-I expression both in vivo and in vitro (Fig. 2). Typically, an increase in MHC-I expression enhances antigen presentation, thereby facilitating improved T cell recognition and activation. Therefore, the unchanged MHC-I expression during PEDV infection potentially benefits virus infection not to further augment T cell activation. This may be the reason why PEDV outbreaks are common. However, whether the impaired MHC-I induction by PEDV infection contributes to a weak CD8⁺ T cell immune response during PEDV infection remains unclear and merits further investigation.

On the other hand, the expression of MHC-I molecules, which are the major ligands of NK cell’s inhibitory receptors, plays an important role in manipulating NK cell activation (16). NK cells play critical roles in the first-line defense against virus infection. Unlike T cells, NK cells can directly kill virus-infected cells lacking MHC-I expression, a phenomenon known as “missing self-recognition,” and do not need specific antigen presentation. Some acute infection viruses, like the ZIKA virus and influenza virus, upregulate the MHC-I expression to evade NK cellular immunity (35, 36). Meanwhile, our previous results demonstrated that PDCoV infection robustly upregulates MHC-I expression, and our recent research found that PDCoV infection can also escape from NK cellular immunity via the NLRC5-MHC-I axis (data not shown) (18). These results suggest that modulation of MHC-I expression seems like a “double-edged sword” for viral infection. Excessive upregulation of MHC-I expression may enhance CD8⁺ T cellular immunity, whereas excessive downregulation may activate NK cellular immunity. Our study revealed that PEDV infection did not alter physiological MHC-I expression, which makes PEDV-infected cells masquerade as normal cells, thereby avoiding the activation of CD8⁺ T and NK cells and thus maximizing virus infection in the host (Fig. 8).

Fig 8.

Model depicting PEDV’s evasion of the MHC-I-mediated immune response by nsp1-mediated direct inhibition of virus-induced NLRC5 mRNA translation. PEDV benefits its infection best by disguising PEDV-infected cells as “normal” cells by maintaining no alteration of MHC-I expression. Mechanistically, the PEDV nsp1 protein dampens the increase of MHC-I by inhibiting the translation of virus-elicited NLRC5 mRNA, thereby disguising PEDV-infected cells as healthy cells and evading CD8⁺ T and NK cell cytotoxicity. The arrows indicate activation, and the blunt-ended lines indicate inhibition.

NLRC5, the largest member of the nod-like receptor family, plays a vital role in the expression of MHC-I genes and related antigen presentation genes by directly binding to the regulatory elements in the promoter of MHC-I molecules in the nucleus, thereby manipulating CD8⁺ T cellular immunity (37). Due to its vital role in the modulation of MHC-I expression, NLRC5 has become an attractive target for many viruses. It has been reported that SARS-CoV-2 infection dampens MHC-I induction by viral protein ORF6 by inhibiting NLRC5 expression at the transcription level and suppressing the CITA function of NLRC5 (30). This research shares some similarities with our findings, but the specific inhibition mechanisms appear to be different. Both SARS-CoV-2 and PEDV infections impede the upregulation of MHC-I by directly targeting NLRC5 expression. Nevertheless, the SARS-CoV-2 infection suppresses NLRC5 expression at the transcription level and inhibits its nuclear transport through ORF6. In contrast, PEDV infection upregulates NLRC5 expression at the transcription level but dampens the induction of NLRC5 protein expression due to the inhibitory function of nsp1.

Nsp1 protein, unique in α-CoVs and β-CoVs, consists of about 110 amino acids (α-CoVs) or 180 amino acids (β-CoVs) with a unique hydrophobic β-barrel amino acids motif and flexible N terminal and C terminal tails (21). The nsp1 of α-CoVs and β-CoVs share a low degree of sequence homology, but the core domains are highly conserved and possess the biological function to inhibit host gene expression through multi-pronged strategies (27, 38, 39). In the case of α-CoVs, the nsp1 of HCoV-229E and HCoV-NL63 has been reported to induce degradation of host mRNA and also bind to the small ribosome 40S subunit, hijacking the host translation system to increase the synthesis of viral proteins for replication (40). TGEV nsp1 has also been reported to inhibit host protein translation, but this repression is not achieved by binding to the ribosome. Additionally, TGEV nsp1 does not induce mRNA degradation, which differs from HCoV-229E and HCoV-NL63 (41). The detailed mechanism for coronavirus nsp1-mediated suppression of host proteins is variable and needs further investigation. Shen et al. reported that PEDV nsp1 may induce host mRNA decay to inhibit host gene expression, and PEDV nsp1 does not interact with ribosomal S6, which is a part of 40S, indicating that PEDV nsp1 may need the help of other host proteins to achieve host gene inhibition (27). During PEDV infection, although the mRNA levels of NLRC5 increase, the presence of nsp1 inhibits the translation of NLRC5 mRNA. The nsp1 of coronavirus primarily suppresses the synthesis of newly produced proteins. Physiological NLRC5, which may have been produced before infection, has a long half-life (exceeding 30 h, predicted by using ProtParam) and stable structure, which may be similar to actin and tubulin, and may degrade at a slower rate. As a result, nsp1 only inhibits the translation of freshly synthesized NLRC5 mRNA without reducing the level of pre-existing physiological NLRC5 protein. The C-terminal of coronavirus nsp1 is essential for protein translation inhibition, such as aa 277–309 of MHV and aa 160–173 of SARS-CoV (42, 43). Shen et al. reported that three regions (aa 67–71, aa 78–85, and aa 103–110) of PEDV nsp1 are responsible for protein translation inhibition, differing from β-CoVs. Our results demonstrated that deletion of the C-terminal (aa 96–110) of nsp1 abolished the inhibitory effect of nsp1 on NLRC5-induced MHC-I expression, indicating that aa 96–110 is the key region in the process of MHC-I inhibition. These results suggest that PEDV nsp1 may utilize different mechanisms to inhibit MHC-I, and further research is required to understand the detailed function of nsp1 on the NLRC5-MHC-I axis. The secondary structure of PEDV nsp1 consists of two α-helices (α1 and α2) and six β-strands (β1–β6), with the β-strands forming a β-barrel fold (27). This barrel-like structure plays a crucial role in the protein, particularly in binding sites and catalytic centers. Using the Protein Data Bank (PDB) and AlphaFold 3 to analyze the structure of PEDV nsp1 (ID: 5XBC), we found that the amino acids (Tyr and Phe) at positions 96 and 97 are located at the start of the β6-strand. Their deletion results in the complete loss of the β6-strand structure, potentially destabilizing the β-barrel fold and thus leading to a loss of protein activity or function. In addition, the inhibition of host proteins by PEDV nsp1 is not totally universal because PEDV nsp1 has a pronounced inhibitory effect on the expression of exogenous porcine aminopeptidase N (APN), whereas it has no effect on porcine melanoma differentiation-associated gene 5 (MDA5) (data not shown). The detailed mechanism of nsp1 inhibition on host proteins remains elusive and needs further investigation.

MATERIALS AND METHODS

Cell cultures and virus

IPI-2I and ST were cultured in Dulbecco-modified Eagle medium (Gibco, USA) with 10% fetal bovine serum (Gibco, USA) and 1% penicillin-streptomycin (Beyotime Biotechnology, China) and maintained in a humidified atmosphere at 37°C with 5% CO₂. The PEDV-JMS strain (GenBank No. PP461398.1) was isolated and maintained in our lab.

Cloning and construction of plasmids

The porcine NLRC5 expression vector and NLRC5 reporter plasmid were reserved in our lab and described previously (18). PEDV viral protein expression vectors were amplified from PEDV cDNA using the ClonExpress Ultra one-step cloning kit (Vazyme, China) with specific primers (Table 1) and cloned into pCAGGS vector fused with an HA tag or pCDNA3.1(+) vector fused with a Flag tag at the carboxyl terminus. Additionally, to determine the functional region of PEDV nsp1, nsp1 truncations (aa 1–60, aa 1–85, aa 1–95, aa 1–97, aa 1–102, and aa 1–110) were constructed into the pCAGGS vector using EcoR I and Xho I restriction sites. All constructs were validated by DNA sequencing.

TABLE 1.

Sequences of primers used for PCR^a

Primer name	Sequence (5′−3′)
Nsp1-F	GTTCCAGATTACGCTATGGCTAGCAACCATGTCACATT
Nsp1-R	CTCGAGGCATGCCCGTCAACCACCACGACGACCAAAAG
Nsp4-F	GTTCCAGATTACGCTATGGGTCTTCCTAGTTTTTCAAAGG
Nsp4-R	CTCGAGGCATGCCCGTCACTGTAGAGTTGAATTGTAACTCACAGTGG
Nsp5-F	GTTCCAGATTACGCTATGGCTGGCTTGCGTAAGATG
Nsp5-R	CTCGAGGCATGCCCGTCACTGAAGATTAACGCCATACATTTGA
Nsp6-F	GTTCCAGATTACGCTATGGGTGGCTATGTGTCACGC
Nsp6-R	CTCGAGGCATGCCCGTCACTGAACGGAAGAAATCTTAATATTCCG
Nsp7-F	GTTCCAGATTACGCTATGTCTAAACTGACTGATATTAAGT
Nsp7-R	CTCGAGGCATGCCCGCTACTGCAACATACTATTGTCATTA
Nsp8-F	GTTCCAGATTACGCTATGAGTGTTGCATCTACTTATGTAGGTTT
Nsp8-R	CTCGAGGCATGCCCGTCACTGGAGCTTAACAATACGCTCACA
Nsp9-F	GTTCCAGATTACGCTATGAATAATGAGATTATTCCTGGTAAGC
Nsp9-R	CTCGAGGCATGCCCGTCACTGCAAGCGTACAGTGGCAC
Nsp10-F	GTTCCAGATTACGCTATGGCTGGTAAACAAACAGAACAG
Nsp10-R	CTCGAGGCATGCCCGTCATTGCATAATGGATCTGTCACAAGTG
Nsp13-F	GTTCCAGATTACGCTATGTCTGCAGGGCTTTGTGTTG
Nsp13-R	CTCGAGGCATGCCCGTCACTGCAAATCAGACAATTTAAGCTCA
Nsp14-F	GTTCCAGATTACGCTATGGCTAATGAGGGTTGTGGTC
Nsp14-R	CTCGAGGCATGCCCGTCATTGCAAATTGTTACTAAATGTCTGCC
Nsp15-F	GTTCCAGATTACGCTATGGGTCTTGAGAACATTGCTTTC
Nsp15-R	CTCGAGGCATGCCCGTCATTGAAGTTGTGGATAAAATGTCTGG
Nsp16-F	GTTCCAGATTACGCTATGGCCAGTGAATGGAAGTGTG
Nsp16-R	CTCGAGGCATGCCCGTCATTTGTTTACGTTGACCAAATGATTAG
S-F	CGGGGTACC ATGAGGTCTTTAATTTACTTCTGG
S-R	CCGCTCGAGCTACTTGTCATCGTCGTCCTTGTAATCGCCACTGCCCTGCACGTGGACCTTTT
E-F	GTTCCAGATTACGCTATGCTACAATTAGTGAATGATAAT
E-R	CTCGAGGCATGCCCGTTATACGTCAATAACAGTACTGGG
M-F	GTTCCAGATTACGCT ATGTCTAACGGTTCTATTCC
M-R	CTCGAGGCATGCCCGTTAGACTAAATGAAGCACTTTCT
N-F	GTTCCAGATTACGCTATGGCTTCTGTCAGCTTTCAG
N-R	CTCGAGGCATGCCCGTTAATTTCCTGTATCGAAGATCTCGT
ORF3-F	GTTCCAGATTACGCTATGTTTCTTGGACTTTTTCAATACA
ORF3-R	CTCGAGGCATGCCCGTCATTCACTAATTGTAGCATACTCG
Nsp1(1–60)-F	GTTCCAGATTACGCTATGGCTAGCAACCATGTCACATT
Nsp1(1–60)-R	CCGCTCGAGCTAGTCTTCGGGAAGCAATCC
Nsp1(1–85)-F	GTTCCAGATTACGCTATGGCTAGCAACCATGTCACATT
Nsp1(1–85)-R	CCGCTCGAGCTAGATGTTTCTGGGGCGGCTAC
Nsp1(1–95)-F	GTTCCAGATTACGCTATGGCTAGCAACCATGTCACATT
Nsp1(1–95)-R	CCGCTCGAGCTAATTACAGTTAGAAAATAATAGCCAACCAC
Nsp1(1–97)-F	GTTCCAGATTACGCTATGGCTAGCAACCATGTCACATT
Nsp1(1–97)-R	CCGCTCGAGCTAGAAGTAATTACAGTTAGAAAAT
Nsp1(1–102)-F	GTTCCAGATTACGCTATGGCTAGCAACCATGTCACATT
Nsp1(1–102)-R	CCGCTCGAGCTACTCTAACTCTTCGAGGAAGTAATTACAGTT

^a F: forward primer; R: reverse primer.

Virus infection and transfection

IPEC-J2 or IPI-2I cells at 100% confluence were infected with PEDV at the desired MOI or mock-infected with DMEM for 2 h. The cells were then washed three times with DMEM to remove unbound virus and cultured at 37°C until harvested.

For co-transfection of NLRC5 and PEDV viral protein constructs, IPI-2I or ST cells were seeded in 24-well plates at 80% confluence and then transfected with NLRC5 (200 ng) and non-structural constructs (600 ng) using Lipofectamine 2000 reagent (Invitrogen, USA) according to the manufacturer’s guidelines. After 36 h of transfection, cells were harvested for RT-qPCR, IFA, or flow cytometry analyses.

RNA extraction and RT-qPCR analysis

Total cellular RNA was extracted using the Simply P total RNA extraction kit (BioFlux, China) according to the manufacturer’s instructions at indicated time points after infection or transfection. cDNAs were synthesized from RNA using PrimeScript II first-strand cDNA synthesis kit (TaKaRa, Japan), and the relative qPCR was performed using SYBR Green PCR Master Mix (Roche, Switzerland) with gene-specific primers for target genes and host gene GAPDH as control. Data from three independent experiments were used for analysis. All primers used for RT-qPCR are shown in Table 2.

TABLE 2.

Sequences of primers used for RT-qPCR

Primer name	Sequence (5′−3′)
NLRC5-qPCR-F	TCCAAACAAGTGCGATGA
NLRC5-qPCR-R	TCCAAACAAGTGCGATGA
SLA-1-qPCR-F	GTGGCTGGAGTTGTGATC
SLA-1-qPCR -R	ACCCTTGGTAAGGGACAC
SLA-2-qPCR -F	AGGGAGAGAGGAGCTACC
SLA-2-qPCR -R	ATGTGTCTTTGGAGGCTC
SLA-3-qPCR -F	CACAGACTTTCCGAGTGA
SLA-3-qPCR -R	TAGGCGTCCTGACTGTAC
β2M-qPCR -F	CCTGTCTTTCAGCAAGGA
β2M-qPCR -R	CGGTTAGTGGTCTCGATC
TAP1-qPCR-F	ACGGGGACTGTGTCTCTT
TAP1-qPCR-R	GAGATTCCTGCACCTGTG
GAPDH-qPCR-F	CCTTCCGTGTCCCTACTGCCAAC
GAPDH-qPCR-R	GACGCCTGCTTCACCACCTTCT

Dual-luciferase reporter assay

The luciferase assay was performed as described previously (18). Briefly, IPI-2I cells were seeded in 48-well cell plates at 80% confluence and then transfected with pGL3-NLRC5 promoter expression vector and internal control pRL-TK for 12 h, followed by infection with PEDV for 24 h before harvesting. After rinsing twice with PBS, cells were gently rocked for 15 min with passive lysis buffer (PLB). Firefly luciferase activity and Renilla luciferase activity were measured with LAR II and Stop&Glo Reagent on PE Envision (PerkinElmer, USA). Data are normalized for transfection efficiency by dividing Firefly luciferase activity by Renilla luciferase activity.

Flow cytometry assay

To analyze surface SLA-I expression level, cells were washed twice with PBS supplemented with 0.1% bovine serum albumin (BSA) and digested with trypsin at the indicated time point after transfection or infection. After blocking with an isotype control antibody, the cells were incubated with FITC-conjugated anti-SLA-I antibody (Bio-Rad, USA) for 1 h at 4°C. Finally, the cells were washed three times and passed through a 35-µm filter. Flow cytometry data acquisition and analysis were performed using BD Canto II (BD Biosciences, USA) and FlowJo software (Tree Star, USA).

To analyze the total cellular SLA-I expression level, the cells were digested and resuspended with fixation/permeabilization solution (BD Biosciences, USA) for 20 min at 4°C. Following this, cells were washed twice with BD Perm/Wash buffer and incubated with fluorescein isothiocyanate-conjugated anti-SLA-I antibody (Bio-Rad, USA) for 30 min at room temperature. Finally, the cells were washed three times and passed through a 35-µm filter. Flow cytometry data acquisition and analysis were performed on BD Canto II (BD Biosciences, USA) using FlowJo software (Tree Star, USA).

Indirect immunofluorescence assay

To investigate the effect of PEDV infection or nsp1 on NLRC5 protein expression, the IFA was performed as described previously (26). Briefly, after transfection or infection, IPI-2I cells were fixed with 4% formaldehyde and permeabilized with 0.2% Triton X-100 for 15 min. PBS containing 5% FBS and 5% skim milk (Sigma-Aldrich, USA) was used to block the cells to reduce background fluorescence and false positives. After blocking, the cells were incubated with the HA-tag specific antibody (Abcam, USA) and the PEDV-specific monoclonal antibody (produced in our laboratory) for 1.5 h at 37°C. For NLRC5 and nsp1 co-transfected cells, the blocked cells were incubated with HA-tag specific antibody (Abcam, USA) and Flag-tag specific antibody (CST, USA). The cells were then washed and incubated with Alexa Fluor 546 goat anti-mouse IgG antibody (Thermo Fisher Scientific, USA) and Alexa Fluor 488 donkey anti-rabbit IgG antibody (Thermo Fisher Scientific, USA) for 1 h at 37°C. Finally, all the cells were stained with DAPI (4’,6-diamidino-2-phenylindole) and then visualized using a Nikon Ti2-E fluorescence microscope.

SDS-PAGE and western blotting

IPI-2I cells seeded in 6-well plates were harvested in Nonidet P-40 (NP-40) supplemented with protease inhibitor cocktail (Roche, Switzerland) and phenylmethylsulfonyl fluoride (PMSF) for 30 min on ice. After centrifugation for 10 min at 12,000 g, the supernatants were collected and boiled with sodium dodecyl sulfate (SDS) sample buffer for 10 min. For immunoblotting, equal amounts of extracts were separated by SDS–polyacrylamide gel electrophoresis (PAGE) and then were transferred onto nitrocellulose membranes (GE Healthcare, USA). For immunoblot analysis, the membranes were blocked with 5% skim milk in Tris buffered saline with Tween-20 (TBST) and incubated with appropriately diluted anti-NLRC5 primary antibody (prepared by our lab) overnight at 4°C. After incubation with HRP-conjugated secondary antibody (Abcam, USA), blots were measured using a Tanon 4800 (Tanon, China).

Cytotoxicity assay

NK cell-mediated cytotoxicity assay was conducted following methodologies outlined in prior studies (28). IPI cells were infected with PEDV at different MOIs at 36 hpi and the IPI cells were labeled with CellTrace CFSE (Invitrogen, USA). Following staining, the target cells were washed and plated at a density of 1 × 10⁵ cells/mL. PBMCs were isolated from the blood of healthy pigs and subsequently treated with IL-2 overnight to activate NK cells. Activated PBMCs (2 × 10⁶ cells/mL) were then co-cultured with IPI cells for approximately 6 h. Post-co-culture, the cells were stained with eBioscience Fixable Viability Dye eFluor (Thermo, USA) to identify dead cells. IPI cells treated with Triton X-100 served as the positive control. Data acquisition was performed using a BD Canto II (BD Biosciences, USA) and analyzed with FlowJo software (Tree Star, USA). Percent cytotoxicity was calculated as: Percent cytotoxicity = 100 × [(Mortality rate of target cells in the co-culture group) － (Mortality rate of target cells in the solo-culture group)] ÷ [(Mortality rate of triton X-100 treated cells) － (Mortality rate of target cells in the solo-culture group)].

Experimental infection of piglets and immunohistochemistry

Six 5-day-old SPF piglets from a swine herd were confirmed negative for PDCoV, porcine rotavirus (PoRV), PEDV, and TGEV by qRT-PCR and randomly divided into two groups. SPF piglets in group 1 were inoculated orally with 1 mL of solution containing 1 × 10³ 50% tissue culture infective doses (TCID₅₀) of PEDV. The remaining three piglets in the control group were sham-infected with DMEM. The clinical signs of vomiting and diarrhea were monitored and scored on each test day. All the piglets were euthanized at 72 hpi. Tissue samples were collected in 4% formaldehyde for further histopathological examination.

Intestinal samples were cut from the 4% formaldehyde-fixed jejunum tissues and embedded in paraffin blocks following standard methods. IHC staining of SLA-I was performed by using monoclonal antibodies purchased from Bio-Rad. After rinsing with PBS three times, slides were incubated with horseradish peroxidase (HRP)-conjugated goat anti-mouse secondary antibody (Abcam, USA) at room temperature for 40 min. The slides were subsequently incubated with diaminobenzidine (DAB) without light for 5–10 min and stained with hematoxylin. After dehydration, clearing, and mounting, immunoreactivity was visualized using an optical microscope.

Statistical analysis

All data were presented as the mean ± the standard error of the mean (SEM) from three independent experiments and analyzed by using Prism 7 (GraphPad). P-values were calculated using an unpaired two-tailed t-test and considered statistically significant if P＜0.05.

ETHICS APPROVAL

The animal study was reviewed and approved by the Laboratory Animal Ethical Committee of China Agricultural University, Beijing. Written informed consent was obtained from the owners for the participation of their animals in this study.

DATA AVAILABILITY

The data that support the findings of this study are available on request from the corresponding author.