The animal experiment was approved by the Institutional Animal Care and Use Committee (IACUC) of the Yangzhou University Animal Experiments Ethics Committee (permit number: SYXK (Su) IACUC2012-0029).

The experimental pigs were selected from a farm with an outbreak of porcine epidemic diarrhea. We analyzed the pathogens of PEDV, transmissible gastroenteritis virus (TGEV), porcine rotavirus (PoRV), and porcine deltacoronavirus (PDCoV) of 7-day-old piglets with typical clinical symptoms of epidemic diarrhea and selected four 7-day-old piglets infected with PEDV and four healthy piglets feeding in the same conditions. Our previous findings showed that the PEDV detected in pigs is not specific and belongs to the classic strain CV777 (17). The small intestinal tissues were stored in liquid nitrogen for subsequent use.

Total DNA was extracted from the duodenum, jejunum, and ileum using a DNA extraction kit (Beijing Tiangen Biochemical Co., Ltd., Beijing, China). The concentration and purity of DNA were determined using 2% agarose gel electrophoresis and NanoDrop 1000, respectively. Total RNA was extracted from the intestinal tissues and cells using TRIzol (Invitrogen, USA). Total RNA purity and concentration were assessed using NanoDrop 1000, respectively. RNA samples were stored at −70°C. RNA was reversely transcribed into cDNA using a reverse transcription kit (Vazyme Biotech Co., Ltd., Nanjing, China). The cDNA synthesis reaction mixture (10 μl) contained 5 × PrimeScript Buffer (2 μl), PrimeScript RT Enzyme Mix I (0.5 μl), Oligo dT (0.5 μl), Random hexamers (0.5 μl), RNA (500 ng), and RNase-free H₂O (up to 10 μl total volume). The thermal conditions used were as follows: 25°C for 10 min, 50°C for 30 min, and 85°C for 5 min; the cDNA samples were stored at 4°C.

RT-qPCR was conducted using the AceQ qPCR SYBR Green Master Mix (Vazyme Biotech Co., Ltd., Nanjing, China). The PCR reaction mixture contained 2 μl cDNA, 0.4 μl of each primer (10 μmol/L), 10 μl 2 × SYBR Premix ExTapTMII, 0.4 μl 50 × ROX Reference Dye II, and RNase-free ddH₂O (up to 20 μl total volume). The reaction conditions were as follows: 95°C for 5 min, followed by 40 cycles of 95°C 10 s and 60°C 34 s. The melting curve analysis was performed at 95°C for 10 s, 60°C for 60 s, and 95°C for 15 s.

We rinsed the intestinal tissue with normal saline, randomly cut the tissue, and extracted the total RNA from the tissue. The samples were tested for the presence of PEDV, TGEV, PoRV, and PDCoV using RT-qPCR. Primer information is provided in Supplementary Table 1.

Small intestinal tissues (duodenum, jejunum, and ileum) were fixed with 10% formalin solution, dehydrated with 70, 80, and 90% alcohol, made transparent by treating with xylene solution, embedded with paraffin, sectioned (Leica Rotary Microtome; Leica Biosystems Co., Ltd., Illinois, Buffalo Grove, USA), and stained with hematoxylin-eosin.

IPEC-J2 cells were cultured at 5% CO₂ and 37°C in 12-well plates with Dulbecco's modified Eagle medium (DMEM; Gibco., Grand Island, New York, USA) containing 10% fetal bovine serum. When the cell density reached 80%, the cells were infected with 0.1 MOI PEDV for 12 and 24 h. Non-infected cells were used as the control group. Four replicates were established per group. MUC2 expression in the small intestinal tissue of the diarrhea group and control group and infected cells exposed to different infection periods was detected using RT-qPCR. The fluorescent quantitative primers were designed using Primer 5.0 software; GAPDH and ACTB were used as internal reference genes (Supplementary Table 2).

Three siRNA sequences (si-MUC2-1, si-MUC2-2, and si-MUC2-3) and negative control sequence (si-MUC2-NC; Supplementary Table 3) were designed and synthesized by Gene Pharma Co., Ltd. (Shanghai, China). The siRNAs and negative control group (NC) were transfected into IPEC-J2 cells using Lipofectamine 2000, with three replicates per treatment group. Cells were incubated overnight and analyzed for expression of fluorescein-labeled siRNAs after 24 h. RNA interference efficiency was determined using RT-qPCR, and the treatment group with the highest interference efficiency was selected for lipofection; NC and Control were used as control groups. Cells were infected with 0.1 MOI PEDV, and the DMEM complete medium was changed after 2 h. The cell morphology was observed under the fluorescence microscope (MoticAE31 inverted fluorescence microscope; Motic China Group Co., Ltd., Xiamen, China). The mRNA expression level of PEDV M gene was determined using RT-qPCR after 24 h.

The promoter region (2,000 bp) of pig MUC2 was determined using NCBI (https://www.ncbi.nlm.nih.gov/) and Ensembl (http://asia.ensembl.org/index.html) databases. Methprimer software (http://www.urogene.org/cgi-bin/me-thprimer/methprimer.cgi) was used to predict the CpG island in the promoter region of the MUC2 gene, and primers used in BSP-PCR were designed [Supplementary Table 4, MUC2 (BSP)]. The Alibaba2 software (http://gene-regulation.com/pub/programs/alibaba2/index.html) was used to predict the transcriptional elements in CpG islands. The genomic DNA was transformed according to the manufacturer instructions of the D5006 bisulfite conversion kit (Beijing Tianmo Technology Development Co., Ltd., Beijing, China). DNA was amplified using the BSP-PCR primers; the PCR reaction mixture (25 μl) included 1 μl of each primer (10 μmol/L), ZYMO Taq Premix (12.5 μl), DNA (2 μl), and ddH₂O (up to 25 μl total volume). The reaction conditions were as follows: 95°C for 10 min; 38 cycles of 95°C for 30 s, 52°C annealing for 30 s, and 72°C extension for 35 s; and final extension for 10 min. The PCR product was stored at 4°C. The purified PCR product was ligated into the pMD-19T vector (Bao Bioengineering Dalian Co., Ltd., Dalian, China) overnight at 16°C. The ligation product was transformed into Escherichia coli DH-5α competent cells (Beijing Tiangen Biochemical Co., Ltd., Beijing, China) and plated on ampicillin-containing agar plates at 37°C. After 12-16 h, 20 positive monoclonal colonies were picked out per plate and sent to the company (Songon Biotech Co., Ltd., Shanghai, China) for sequencing. The software QUMA (http://quma.cdb.riken.jp/) was used to determine the methylation ratio of each CpG site, and the positive sequencing results were compared with the reference genome.

We constructed a recombinant plasmid containing the MUC2 promoter, and primers to amplify the promoter were designed (Supplementary Table 4, MUC2-1). The pGL3-basic vector and PCR products were digested using SpeI and NcoI, and the purified products were ligated and transformed into Escherichia coli DH-5α competent cells. Monoclonal colonies were picked out for sequencing, and it was verified whether the recombinant plasmid was successfully constructed. The recombinant plasmid pGL3-MUC2 was methylated with methyltransferase M.SssI, and the product was purified and recovered using the TIANquick Midi purification kit (Beijing Tiangen Biochemical Co., Ltd., Beijing, China). Next, 100 ng M.SssI-methylated plasmid pGL3-MUC2, untreated pGL3-MUC2 plasmid, empty plasmid pGL3-basic, and 2 ng Renilla luciferase reporter plasmid (pRL-TK) were transfected into IPEC-J2 cells cultured in 12-well plates at 37°C for 48 h, and the fluorescence activity was detected using the dual-luciferase reporter gene system.

The Pierce Agarose ChIP Kit (Thermo Fisher Scientific, USA) was used for the chromatin immunoprecipitation (ChIP)-PCR assay. Approximately 80 mg of jejunum tissue samples were cut and placed in pre-chilled phosphate-buffered saline; the tissue samples were then fixed with formaldehyde to lyse the tissue and cross-link protein to DNA. The chromatin was enzymatically digested and incubated with specific anti-YY1 antibodies (Proteintech Group, Inc., USA). The protein-DNA complexes were precipitated, eluted, and de-crosslinked; proteins were digested, and enriched DNA was purified and finally precipitated and eluted for PCR amplification. ChIP-PCR was used to detect the enriched precipitated DNA. The positive control using amplified the DNA precipitated by anti-RNA Polymerase II antibody precipitation obtained DNA as a template, the negative control using rabbit Anti-IgG antibody as a template, whereas the input test group amplified MUC2 gene using unprecipitated DNA as template. The antibodies used in the positive and negative controls were from the Pierce Agarose ChIP Kit. The primers designed for ChIP-PCR are presented in Supplementary Table 4 [MUC2 (CHIP-PCR)]. The reaction mixture (20 μl) contained 1 μl of each primer (10 μmol/L), 1 μl DNA, 10 μl 2 × Taq Master Mix, and 7 μl ddH₂O. The reaction procedure was as follows: 95°C for 5 min; 38 cycles of 95°C for 30 s, 60°C annealing for 30 s, and 72°C for 30 s; and 72°C for 10 min. PCR products were detected by 2% agarose gel electrophoresis and compared with the target sequence on the basis of the sequencing results.

Three siRNA sequences and one negative control sequence were designed for the transcription factor Yin Yang 1 (Supplementary Table 3). Primers were designed and synthesized by Gene Pharma Co. Ltd. (Shanghai, China). si-YY1-1, si-YY1-2, si-YY1-3, and si-YY1-NC were transfected into IPEC-J2 cells, with three replicates in each treatment group, and the expression of fluorescein-labeled siRNAs in the cells was assessed after 24 h. The relative expression of the YY1 gene was detected by RT-qPCR, and the RNA interference efficiency was determined. The treatment group with the highest interference efficiency was selected for further analysis.

The RT-qPCR results were analyzed using the 2^−ΔΔCt method; SPSS 18.0 software was used to analyze the data. The relative quantitative results were analyzed by using independent sample t-test for the three replicates of each set; the results are presented as mean ± standard deviation (Mean ± SD). The correlation between promoter methylation and mRNA expression levels was analyzed using Pearson's correlation test. The relative mRNA expression was analyzed by t-test using SPSS 18.0 software.

In this study, PCR was used to ensure that the piglets in the normal group were not infected with PEDV and the piglets in the diarrhea group were only infected with PEDV. The results showed that the bands in the diarrhea group were consistent with the expected product size (216 bp), while the normal piglet samples showed no obvious bands (Supplementary Figure 1A). In order to eliminate the interference of other pathogens, we employed agarose gel electrophoresis to detect the other common viral diarrhea pathogens (TGEV, PoRV, PDCoV), and the result showed no obvious bands (Supplementary Figure 1B). These results suggested that the piglets in the diarrhea group were only infected with PEDV, whereas the piglets in the healthy group were not infected with PEDV. Therefore, the samples were deemed suitable to use in the subsequent experiments.

PEDV can cause damage to the intestinal barrier, in the present study, we want to verify whether PEDV can also damage the intestinal barrier in pigs. So, we observed the intestinal damage caused by PEDV infection under microscope. As shown in Figure 1A, the intestinal villi of the diarrhea group piglets were structurally damaged: the villous density and height were decreased, the villi become sparse, the lamina propria was exposed, and the intestinal glands were atrophic. In contrast, the intestinal mucosal barrier of healthy piglet was intact, and the intestinal mucosal epithelial cells were well-organized, clear in outline, and regular in arrangement.

At the same time, we observed the damage caused by PEDV to IPEC-J2 cells using optical microscopy. The result showed that, after 24 h of PEDV infection, cell morphology began to change, and numbers of atrophied cell increased (Figure 1B-III), whereas after 48 h of PEDV infection, cell lesions occurred, normal cell morphology was completely lost, and extensive cell death occurred (Figure 1B-IV). These results showed that PEDV could cause damage to IPEC-J2 cells and small intestine tissues in piglets.

MUC2 protects the gut from pathogenic microorganisms, so does MUC2 play a role in PEDV infestation? To investigate that, the mRNA level of MUC2 was detected at the cellular and organizational levels after PEDV infection. The results showed that the MUC2 expression level in the small intestine (duodenum, jejunum, and ileum) of piglets from the diarrhea group was significantly higher than that from the healthy group (P < 0.01; Figure 2A). In addition, the results of PEDV infection of IPEC-J2 cells at different time periods showed that MUC2 expression was significantly up-regulated at 24 h (P < 0.05; Figure 2B). Overall, these results suggested that MUC2 may play a significant role in PEDV infection.

To further investigate the role of MUC2 in PEDV infection, we used RNA interference to down regulate MUC2 expression. The fluorescein markers, used to monitor siRNA transfection, were highly expressed after 24 h, indicating that siRNAs had been successfully transfected (Figure 3A). The treatment group with the highest interference efficiency (si-MUC2-1) was selected for the subsequent experiments (Figure 3B). Then we detected the PEDV M gene to investigated the effect of MUC2 down regulation on PEDV replication. The result showed that after 24 h of PEDV infection, M gene expression in the MUC2-RNAi group was significantly higher than that in the MUC2-NC group (P < 0.05), and significantly higher than that in the Control group (P < 0.01; Figure 3C). In general, these results showed that down regulation of MUC2 expression may contribute to PEDV replication.

To investigate whether DNA methylation is regulating the expression of MUC2, we predicted the CpG island in the MUC2 promoter and the results showed that there was a CpG island ~208 bp in size (−324 to −116 bp; Supplementary Figure 2A). The CpG island was amplified by PCR, and the PCR product was detected by 1% agarose gel electrophoresis. The result showed a single bright band, and the size was as expected (Supplementary Figure 2B).

In order to further explore the effect of DNA methylation on MUC2 expression. We analyzed the correlation between CpG island methylation in MUC2 promoter and mRNA expression. The results of the sequencing showed that there were 16 CG sites in the CpG island of the MUC2 promoter region, all of which showed different degrees of methylation (Figure 4A); the methylation level in the diarrhea group was lower than that in the healthy group (P > 0.05; Figure 4B). The methylation levels of most sites were negatively correlated with mRNA expression, except mC-15 and mC-16; among these sites, the methylation level of mC-5 was significantly correlated with mRNA expression (P < 0.05; Figure 4C).

To further prove that DNA methylation can affect MUC2 expression. We detected the effect of promoter methylation of MUC2 gene on promoter activity using Dual-luciferase reporter assay. The results showed that the methylation of the CpG island of the MUC2 promoter significantly reduced the luciferase activity (P < 0.01; Figure 5), suggesting that DNA methylation inhibited the transcriptional activity of MUC2.

We next predicted transcription factors in CpG island of MUC2 promoter region to investigate how the mC-5 site inhibits MUC2 expression. The result showed that the CpG island of MUC2 promoter has 14 transcription factors. A Pearson correlation analysis suggested that the methylation level of the mC-5 site in the binding domain of the transcription factor YY1 was significantly negatively correlated with gene expression (P < 0.05; Supplementary Figure 3).

Then, we used CHIP-PCR to investigate whether transcription factor YY1 binds to the promoter region of MUC2. The result showed a single bright band (104 bp), which was consistent with the target fragment in size and with the sequencing results (Figure 6). The results suggested that transcription factor YY1 specifically binds to the mC-5 site of MUC2 gene promoter region.

To explore the effect of transcription factor YY1 on MUC2 gene expression, we constructed the RNA interference vector of YY1 and transfected it into IPEC-J2 cells. The fluorescein markers were detected to be highly expressed after 24 h, indicating that siRNAs were successfully transfected (Figure 7A). The treatment group with the highest interference efficiency (si-YY1-1) was used to interfere MUC2 expression. (Figure 7B) RT-qPCR results showed that the expression level of MUC2 gene decreased significantly (P < 0.05; Figure 7C) suggesting that interfering transcription factor YY1 can down regulate MUC2 gene expression.