All procedures were approved by the Institutional Animal Care and Use Committee of Hunan Agricultural University.

Newborn piglets (Landrace × Duroc × Yorkshire) were obtained from 4 sows (10 piglets per litter) with the similar parity (3 or 4 parities). The piglets were housed with their own mothers. All 40 piglets were euthanized on postnatal d 0, 1, 3, and 7, with 10 piglets per slaughter. All 30 piglets killed at d 1, 3, and 7 received breast milk except for 10 piglets at d 0. The blood sample was taken from the carotid artery and centrifuged at 3,000 rpm for 15 min at 4°C to obtain serum. Gastric contents (IgG in breast milk), jejunal contents and mucosa were collected and transferred into sterile precooled tubes, and then stored at −80°C until further analysis (Figure 1A).

IPEC-J2 cells were obtained from the Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China). The cells were cultured in Dulbecco’s modified Eagle medium (DMEM) supplemented with 10% fetal bovine serum (FBS) (HyClone, Utah, United States) and 1% Penicillin–Streptomycin (Beyotime Institute of Biotechnology, Shanghai, China). Cells were maintained at 37°C in a humidified incubator under 5% CO₂.

IPEC-J2 cells were seeded in 96-well culture plates and grown to approximately 80% confluence. The cells were incubated with lipopolysaccharide (LPS) (0, 0.1, 0.5, 1, 5, 10 μg/mL) and lipoteichoic acid (LTA) (0, 1, 2.5, 5, 10, 20 μg/mL) in serum-free DMEM for 24 h, respectively, and their viabilities were determined by cell counting kit-8 assay. The maximum concentration that had no effect on cell viability was chosen as the final concentration of LPS and LTA for the following experiments.

IPEC-J2 cells were grown in 12-well transwell system and the changes of trans-epithelial electrical resistance (TEER) were determined using an epithelial voltohmmeter ERS-2 (Merck Millipore, United States). When the epithelial electrical resistance was measured until similar values were recorded on three consecutive measurements, the cells were incubated with LPS and LTA for 24 h, repectively. Then the electrical resistance of each treatment was measured.

Repeating the above procedure until the epithelial electrical resistance was stable. Then, porcine IgG (400 μg/mL) mixed with LPS or LTA or both not was added to the upper chamber of the transwell for 24 h at 37°C. Meanwhile, the opposite chamber was incubated in DMEM serum-free medium. Supernatants from the lower chamber of the transwell were collected at 1, 2, 4, 6, 12, 24 h after incubation. Meanwhile, cells were collected during these time periods.

IPEC-J2 cells were grown at 70–80% confluence treated with or without 5 μM NF-κB inhibitor BAY11-7028 (Sigma-Aldrich, United States) for 1 h and then the cells were stimulated by LPS or LTA or both not for 2 h. Finally, cells were harvested for use.

The IgG levels in serum, gastric contents and cell culture medium supernatants, LPS and LTA levels in jejunal contents were determined using ELISA kits (Enzyme-linked Biotechnology, Shanghai, China) according to the manufacturer’s instructions.

The total DNA from the jejunal content samples was extracted using QIAamp DNA Stool Mini Kit (Qiagen, Germany) following the manufacturer’s instructions. The PCR primer sequences were 338F: 5′-ACTCCTACGGGAGGCAGCA-3′ and 806R: 5′-GGACTACHVGGGTWTCTAAT-3′. The construction of plasmid standard and standard curve drawing were carried out by Paisonol Biotechnology Co., Ltd., Shanghai, China. Gene expression was determined on a real-time quantitative PCR system, using SYBR Green Premix Pro Taq HS qPCR kit (Accurate Biotechnology, Hunan, China). The formula for calculating the copy number is Ct = −KlogX+b. X represents copy number, K represents slope of standard curve, and b represents intercept of standard curve.

Total RNA was extracted from jejunal mucosa and IPEC-J2 cells using RNA rapid extraction kit and transcribed into cDNA using the Evo M-MLV RT Premix kit (Accurate Biotechnology, Hunan, China). The PCR primer sequences utilized for the determination of the gene expression were shown in Table 1. Gene expression was determined on a real-time quantitative PCR system, using SYBR Green Premix Pro Taq HS qPCR kit (Accurate Biotechnology, Hunan, China). Relative quantification of the target gene expression was calculated as previously described methods (Peng et al., 2021).

The total protein of jejunal mucosa and IPEC-J2 cells was extracted with protein extraction reagents (Thermo Fisher Scientific Inc., United States). Proteins were separated by 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and then electrotransferred to polyvinylidene difluoride (PVDF) membranes (BioRad, Hercules, CA, United States). Membranes were blocked and then incubated with the following primary antibodies overnight at 4°C: β-actin (Cell Signaling Technology, #4970), FcRn (Abclonal, A8544), NF-κB P65 (Cell Signaling Technology, #6956), and Phospho-NF-κB P65 (Ser536) (Cell Signaling Technology, #3033). After primary antibody incubation, membranes were washed and then incubated with HRP-conjugated anti-mouse or anti-rabbit IgG antibodies (Beyotime Biotechnology, Shanghai, China). Finally, quantified and digitally analyzed using the image J program (NIH).

The total RNA was extracted from jejunal mucosa using Trizol according to the manufacturer’s instructions. Paired-end (PE) RNA-seq libraries were constructed using the Truseq mRNA-stranded RNA-Seq Library Prep Kit (Illumina, United States). The sequencing of the libraries with a read length of 2 × 150 bp and an insert size of 380 bp was performed on an Illumina HiSeq X Ten sequencing platform. The quality of the raw sequencing data was assessed and filtered with RNA-QC-Chain (Zhou et al., 2018), removing the adaptors, contaminations, and low-quality reads. The filtered reads were aligned to the reference genome of the S. scrofa (NCBI Accession No. GCF_000003025.6) using HISAT2 (v2.1.0) (Kim et al., 2019). The expression level of genes was estimated using fragments per kilobase per million mapped reads (FPKM) by StringTie (v1.3.6) (Pertea et al., 2015). We used DESeq to detect the differentially expressed genes (DEGs), which were defined as genes with|log2(FoldChange)| > 1 and adjusted p < 0.05 between the two groups. The Benjamini-Hochberg (BH) method-based false discovery rate (FDR) multiple test correction was applied to adjust the p value.

The total DNA from the jejunal content samples was extracted using QIAamp DNA Stool Mini Kit (Qiagen, Germany) following the manufacturer’s instructions. After the quality check, DNA was used to construct the PE libraries using NEBNext Ultr DNA Library Prep Kit for Illumina (NEB, United States). A total of 20 samples were successfully sequenced on Illumina HiSeq X Ten platform. The raw sequencing data were evaluated using FastQC and then trimmed by Fastp (v0.20.0) (Brown et al., 2017) to eliminate reads less than 50 bp, adapters, leading or trailing bases with Phred base quality (BQ) scores of <20, and fragments of every five bases with an average BQ score of <25. To obtain the taxonomical composition of the intestinal microbiome for each sample, the high-quality reads were annotated by MetaPhlAn2 (v2.6.0) with default settings (Truong et al., 2015). Alpha diversity was calculated using the QIIME and principal component analysis (PCA) were analyzed by R (v4.0.3) and QIIME. The top 50 taxons with significant differences in relative abundance between groups were clustered by R (v4.0.3).

There were two groups (n = 3 each group) of IPEC-J2 cells that were used for the phosphoproteomic analysis, one is LPS treatment group, treated with 1 μg/mL LPS for 2 h, and the other is LTA treatment group, treated with 2.5 μg/mL LTA for 2 h. SDT (4%SDS, 100 mM Tris–HCl, 1 mM DTT, pH7.6) buffer was used for sample lysis and protein extraction. The amount of protein was quantified with the BCA Protein Assay Kit (Bio-Rad, United States) following the manufacturer’s instructions. Protein digestion and phosphopeptides enrichment were performed as described previously (Wiśniewski et al., 2009). The enriched phosphopeptides was concentrated in vacuum and redissolved in 20 μL 0.1% formic acid solution for mass spectrometry analysis. Liquid chromatography–tandem mass spectrometry (LC–MS/MS) analysis was carried out by Applied Protein Technology Co., Ltd., Shanghai, China. The MS raw data for each sample were combined and searched using the Maxquant (1.5.2.8) software for identification and quantitation analysis. Phosphopeptides with the Fold Change >2 (up-regulated) or Fold Change <0.5 (down-regulated) were further applied to Student’s t-test at a univariate level to measure the significance, the p-values <0.05 were considered statistically significant (Figure 2A).

Data are presented as means ± standard deviation (SD). A two-sided unpaired Student’s t-test with Benjamini–Hochberg correction was used to compare the two groups. The differences among the three groups were analyzed using one-way analysis of variance (ANOVA) followed by Duncan’s test. Significance was set at p < 0.05.

Piglets which did not receive milk after parturition were recorded as d 0, and 24 h after suckling were recorded as d 1. As shown in Figure 1B, with the growth of piglets, the IgG levels in the stomach milk of piglets on d 1 were significantly higher than that on d 3 and d 7 (p < 0.05). On the contrary, the serum IgG content of piglets continued to increase from the d 1 of suckling to the peak on the d 7. The IgG content in serum of piglets on d 3 and d 7 was significantly higher than that on d 1 (p < 0.05). It is known that IgG in newborn piglets is derived entirely from breast milk, as shown in Figure 1C, the serum of piglets at d 0 contains no IgG. And FcRn is the only receptor for transporting IgG. We detected the mRNA expression and protein expression of FcRn in jejunal mucosa of piglets (Figures 1D,E). The trend of FcRn mRNA expression and protein expression was consistent, with the highest expression on the d 1 and then down-regulated. There was a significant difference between d 1 and d 3 and d 7 in FcRn mRNA expression and protein expression (p < 0.05), but the difference between d 3 and d 7 was not significant (p > 0.05). The mRNA expression of FcRn in jejunal mucosa of piglets was significantly different between d 0 and d 1 (p < 0.05), but no significant difference in protein expression (p > 0.05).

We performed absolute quantitative analysis of jejunal microbiota in piglets at four time points (d 0, 1, 3, and 7). As shown in Figure 3A, the amount of jejunal bacteria at d 0 was significantly lower than that at the other three time points (d 1, 3, and 7) (p < 0.05), while there was no significant difference among the other three time points (p > 0.05). The results of LPS content and LTA content in jejunum were consistent (Figures 3B,C). From d 0 to d 7, the contents of LPS and LTA in jejunum of piglets increased gradually. The contents of LPS and LTA in jejunum of piglets were not significantly different between d 0 and d 1 (p > 0.05), but d 0 and d 1 were significantly different from d 3 and d 7 (p < 0.05), respectively, and there were also significant differences between d 3 and d 7 (p < 0.05). To further clarify the colonization rules of intestinal microbiota in piglets, metagenomic sequencing was performed on intestinal microbiota at four time points. The Chao 1 and ACE indices that estimate microbial richness, and the Shannon and Simpson diversity indices which reflects species biodiversity, were calculated to evaluate the alpha diversity (Figure 3D). The species richness (Chao 1 and ACE) of d 0 was significantly lower than that of other three time points (d 1, 3, and 7) (p < 0.05). The results of PCA showed that jejunal microbiota of piglets at the four time points did not completely overlap, indicating that the jejunum of piglets was colonized with different bacteria at the four time points (Figure 3E). Besides, we analyzed the taxonomic composition and top 50 differential bacteria composition of jejunum flora of piglets at four time points (Figures 3F,G). At species level, the dominant bacteria in jejunum of piglets was Clostridioides difficile both at four time points. Except for Lactobacillus mucosae, Lactobacillus delbrueckii, and Lactobacillus amylovorus, there was no significant difference in the relative abundance of other bacteria in jejunum of piglets at four time points. The analysis of top 50 differential bacteria composition also reflects the colonization rule of piglets jejunal microbiota at four time points. By the 7th day after birth, Lactobacillus gradually became the dominant bacteria in jejunal of piglets. In addition, we calculated the proportion of Gram-positive bacteria and Gram-negative bacteria in the top 20 species levels of jejunum of piglets (Figure 3H). The results showed that the relative abundance of Gram-positive bacteria in jejunum significantly decreased and the relative abundance of Gram-negative bacteria significantly increased with the increase of age of piglets (p < 0.05). And the proportion of Gram-positive bacteria in jejunum of piglets was always higher than that of Gram-negative bacteria at four time points (p < 0.05).

We analyzed the correlation between serum IgG content and jejunal mucosal FcRn mRNA expression with total jejunal bacteria, jejunal LPS content, jejunal LTA content and jejunal microbiota alpha diversity index (Chao1 and ACE) at 4 time points, respectively. As shown in Figure 3I, the content of serum IgG had significantly positive correlation with total jejunal bacteria, jejunal LPS content, jejunal LTA content and jejunal microbiota alpha diversity index (Chao1 and ACE) (p < 0.05), and the expression of FcRn mRNA in jejunum had significantly negative correlation with total jejunal bacteria, jejunal LPS content and jejunal LTA content (p < 0.05). Obviously, the content of LPS and LTA in jejunum had the strongest correlation with serum IgG content and jejunal FcRn mRNA expression (p < 0.01). Moreover, we performed redundancy analysis (RDA) between the top 50 jejunal bacteria in genus level and serum IgG content, jejunal LPS content, jejunal LTA content, total jejunal bacteria, and mRAN expression level of FcRn in jejunal mucosal tissue. The results of RDA showed that the microflora composition in jejunum of piglets at four time points had no significant difference in genus level. Furthermore, the interaction between jejunum microbial composition and serum IgG content, jejunal LPS content, jejunal LTA content, total jejunal bacteria, and mRAN expression level of FcRn in jejunal mucosal tissue was highly significant (p = 0.001) (Figure 3J).

The results of PCA based on the gene expression of four time points reflected the degree of differences in the genetic profiles of piglets at d 0, 1, 3, and 7 (Figure 4A). Through the traditional t-test, the differences between the combinations of any two time periods were found (Figure 4B), and finally the list of 6,525 DEGs was combined. Then, after performing unsupervised hierarchical clustering based on the identified DEGs (Figure 4C), we found that the samples at the four time points (d0, d1, d3, d7) could be roughly clustered into three independent groups (d 0 as one group, d 1 as one group, d 3 and d 7 as one group). This suggested that d 1 may be the key point of intestinal function transformation in piglets. Similarly, d 1 is a key time point for intestinal flora colonization. Therefore, we conducted Spearman correlation analysis between the top 50 species level differential bacteria screened in the metagenomic results and 6,525 DEGs (Figure 4D). We screened 6,525 DEGs with |correlation coefficient| > 0.6 and adjusted p < 0.05, and performed KEGG pathway enrichment analysis on the eligible genes (Figure 4E). The results showed that genes strongly associated with differential bacteria were enriched into three categories: Environmental Information Processing, Cellular Processes and Human Diseases, including the NF-κB signaling pathway. Finally, we verified the expression levels of some genes of the NF-κB signaling pathway (Figures 4F–H). The gene expression of TLR2, TLR4 and P65 in the jejunum of piglets were significantly upregulated at d 1 when compared with d 0 (p < 0.05). Besides, all three genes were downregulated from d 1 to d 7, and all three genes mRNA expression were significantly inhibited at d 7 versus the d 1 (p < 0.05).

As shown in Supplementary Figure S1, when the concentration of LPS and LTA were 1 μg/mL and 2.5 μg/mL, respectively, the cell viability and cell barrier function were not affected by 24 h incubation with cells. Thus, we incubated IPEC-J2 with 1 μg/mL LPS and 2.5 μg/mL LTA for 24 h, respectively, to explore the effects of cells treated with LPS or LTA on IgG transmembrane transport (Figures 5A,B). Obviously, from the second hour onwards, the IgG content in transwell lower chamber and IgG transmembrane transport rate at each time point (2, 4, 6, 12, 24 h) in LPS treated group were always significantly lower than those in CON group (p < 0.05). In contrast, the LTA treated group significantly increased the content of IgG in transwell lower chamber transported from the upper chamber compared to CON group at whole time (1, 2, 4, 6, 12, 24 h) (p < 0.05). As shown in Figure 5C, with the accumulation of incubation time, LPS treatment significantly inhibited the mRNA expression of FcRn (p < 0.001), while LTA treatment significantly enhanced the mRNA expression of FcRn (p < 0.001). Apparently, the mRNA expression of FcRn in LTA group was significantly up-regulated versus the LPS group at whole time (1, 2, 4, 6, 12, 24 h) (p < 0.001). Moreover, Western blot analysis showed that LPS and LTA treatment both activated NF-κB signaling pathway and promoted the phosphorylation of P65 (p < 0.001). The activation of NF-κB signaling pathway by LTA was significantly stronger than that by LPS (p < 0.01). Consistent with the results of qPCR, LPS treatment remarkably inhibited the protein level of FcRn expression (p < 0.001) while LTA treatment markedly promoted the protein level of FcRn expression (p = 0.007) compared with the cells without any treatment. Apparently, the protein level of FcRn expression in LTA group was significantly increased versus the LPS group at whole time (1, 2, 4, 6, 12, 24 h; p < 0.05; Figure 5D). Therefore, to explore whether the down-regulation and up-regulation of FcRn induced by LPS and LTA were related to the NF-κB signaling pathway, NF-κB inhibitor BAY11-7082 was used to pretreat IPEC-J2 cells before stimulation with 1 μg/mL LPS or 2.5 μg/mL LTA. We found that blocking NF-κB signaling pathway decreased mRNA and protein expressions of FcRn induced by LTA, but it did not affect the down-regulation of mRNA and protein expressions of FcRn induced by LPS (Figures 5E–H).

Both LPS and LTA treated cells activated the NF-κB pathway, but LPS inhibited the expression of FcRn while LTA promoted the expression of FcRn. In order to explore the reason for the difference between LPS and LTA treatment groups, phosphorylated proteome was detected in cells treated with LPS or LTA (Figure 2). A total of 742 differentially modified peptides were identified by LPS and LTA treatment groups. Compared with LPS treatment group, 304 peptides were up-regulated and 438 peptides were down-regulated in LTA treatment group (p < 0.05; Figure 2B). Significant differences in phosphorylated modified peptides between LPS and LTA treatment groups were visualized in the form of a volcano plot (Figure 2C). Hierarchical clustering algorithm was used to group and classify the differentially expressed phosphorylated peptides in LPS and LTA treatment groups. The proteins corresponding to the significantly differentially expressed phosphorylated peptides obtained can effectively separate the comparison groups, indicating that the screening of differentially expressed phosphorylated peptides can represent the influence of biological treatment on samples (Figure 2D). Next, we performed KEGG pathway enrichment analysis on the proteins corresponding to the differentially expressed phosphorylated peptides screened, and the results showed that significant changes occurred in important pathways such as calcium signaling pathway, AMPK signaling pathway, alcoholic liver disease, TNF signaling pathway, longevity regulating pathway (Figure 2E). In addition, KEGG signaling pathway enrichment of differentially expressed phosphorylated peptides showed that NF-κB signaling pathway was significantly up-regulated in LTA treatment group compared with LPS treatment group (p < 0.05; Figure 2F).