Porcine small intestinal epithelial cell line (IPEC-J2 cells, third-generation primary cells) were purchased from Huatuo Biotechnology Co.

IPEC-J2 cells were grown in dulbecco’s modified eEagle’s medium (DMEM, Gibco, China) supplemented with 10% fetal bovine serum (FBS, Gibco, China) under an atmospheric condition of 95% O₂ /5% CO₂ at 37°C in a humidified incubator. The culture medium was refreshed every 2 days during cell growth.

The cell concentration was adjusted to 1.5 × 10⁵ cells/mL, then 1 mL of cell suspension was transferred to the upper chamber of the transwell. When the cell density reached 80%, replaced the upper and lower chambers with medium containing 0.01 mM α-GML (Kevin, China) and incubated the cells for 24, 36, 48 and 60 h, respectively. The upper chambers were then replaced with medium containing 1 mg/mL fluorescein isothiocyanate-dextran (FITC-dextran, Aladdin, China), and the lower chambers were replaced with phosphate buffered saline (PBS). After incubation in a CO₂ incubator protected from light for 1 h, the PBS in the lower chamber was collected, and the relative fluorescence content of the PBS was measured by using an enzyme marker at an excitation wavelength of 493 nm and an emission wavelength of 518 nm.

The cell concentration was adjusted to 4 × 10⁴ cells/mL, add 100 μL of cell suspension to each well of a 96-well plate, and incubate the cells in DMEM complete medium (10% FBS) until the density reaches about 50%, and then replace the DMEM complete medium with DMEM containing 0, 0.05, 0.03, 0.01, 0.008, 0.005, 0.003, and 0.001 mM α-GML and 1% dimethyl Sulfoxide (DMSO, MP Biomedicals, China). Complete medium was used to culture the cells. After culturing the cells in α-GML working solution for 24, 36, 48 and 60 h, 10 μL of cell counting kit-8 (CCK8, NCM Biotech, China) reagent was added to each well, and the cells were incubated in a cell culture incubator protected from light for 1 h. Then the absorbance value was detected at 490 nm using an enzyme marker, and the absorbance value of each well minus the absorbance value of the blank-well medium could represent the cell activity.

At approximately 80% confluence, the cells were exposed to a complete medium supplemented with 0.01 mM α-GML for varying durations, or subjected to various inhibitors across different groups for a continuous period of 48 h. Total RNA was extracted using total RNA isolation kit (NCM Biotech, China), and the cDNA was synthesized using the PrimeScript first-strand cDNA synthesis kit (Novoprotein, China). The qRT-PCR was performed using the SYBR qPCR SuperMix Plus kit (Novoprotein, China) at 95°C for 1 min, followed by 40 cycles of 95°C for 20 s and 60°C for 1 min. Reaction system was listed in Table 1. The levels of gene expression were quantified using the 2^−ΔΔCt method, and the GAPDH gene was used as the reference gene. The primer sequences were listed in Table 2.

When the cell density reached about 80%, cells were incubated in complete medium containing 0.01 mM α-GML for different times, or treated with different inhibitors (groups: control, α-GML, 0.5 nM Staurosporine, α-GML + 0.05 nM Staurosporine, α-GML + 0.5 nM Staurosporine, α-GML + 1 nM Staurosporine, α-GML + 2 nM Staurosporine, 0.5 μM SCH772984, α-GML + 0.1 μM SCH772984, α-GML + 0.5 μM SCH772984, α-GML + 1 μM SCH772984, α-GML + 2 μM SCH772984) for consecutive 48 h. Total protein from cells was extracted using RIPA lysis buffer (NCM Biotech, China). The appropriate amount of protein was separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis, transferred to PVDF membranes (Millipore, USA) which then were blocked with 5% non-fat dried milk at 37°C for 1 h. Then the membranes were incubated with primary antibodies (1:1000, Bioss, China) overnight at 4°C and then incubated with secondary antibodies conjugated with goat anti-rabbit IgG (H&L) HRP (1:2000, Affinity, China) for 1 h at room temperature. ImageJ software was employed for the quantification of protein bands. For each group, the grayscale values of GAPDH were utilized as an internal control to normalize the data. Each experiment was conducted in triplicate to ensure the reliability and reproducibility of the results.

When cell density reached approximately 80%, cells were incubated in complete medium containing 0.01 mM α-GML (groups: control, DMSO, α-GML). Cells were then lysed with RIPA lysis buffer and stored at −80°C. The cell samples were then sent to Guangzhou Genedenovo Biotechnology Co., Ltd. for further processing and preparation, and high-throughput, high-precision data acquisition and analysis were conducted on the Omicsmart online platform¹ using Data-Independent Acquisition (DIA) proteomics technology. The proteins with p value<0.05 and absolute value of fold change (FC) greater than 1.2-fold were considered as differentially expressed proteins (DEPs). Finally, functional annotation and pathway analysis of the identified differentially expressed proteins were performed to reveal their potential biological significance and mechanisms of action. Gene Ontology (GO), KEGG and Reactome pathway analysis was used to understand the biological processes and pathways that enriched IPEC-J2 cells after α-GML treatment.

The cell concentration was adjusted to 1.5 × 10⁵ cells/mL and were seeded in a 12-well plate for 12 h. The plasmid of dual luciferase reporter assay system and ATF-2 overexpression was co-transfected into the cells using Lipofectamine 2000 (Invitrogen, USA). After 48 h, the transfected cells with luciferase reporter genes were lysed centrifuged for 3 min. The luciferase activity (firefly catalyzes luciferin at a wavelength of 560 nm and renilla luciferase at a wavelength of 465 nm in the supernatant) was measured with the substrates of dual luciferase reporter assay kit (Yeasen Biotechnology, China) using CLARIOstar microplate readers (BMG LABTECH, Germany).

Statistical analysis was performed using IBM SPSS 26.0 and Graph pad Prism 9.5 analysis software. The LSD-t test was used for two-by-two comparisons between groups, and one-way ANOVA was used for comparisons between multiple groups. Significant differences were found when p < 0.05.

Cells treated with various concentrations of α-GML for different times showed no significant effect on proliferation at 24 h (p > 0.05), but a significant increase was observed at 48 h with 0.01 mM α-GML, which was also effective at enhancing proliferation at 60 h (p < 0.05, p < 0.001) (16). This concentration was chosen for further experiments, where it significantly reduced FITC-dextran flux through IPEC-J2 cells at 36, 48, and 60 h, indicating improved barrier function (p < 0.05 or p < 0.001) (16). qRT-PCR and western blot assays revealed that 0.01 mM α-GML significantly elevated the mRNA and protein levels of ZO-1 and OCLN at 36, 48, and 60 h (p < 0.05, p < 0.01), with no impact on the expression levels of Claudin-1 and Claudin-2 (p > 0.05) (16). Collectively, these results preliminarily suggested that 0.01 mM α-GML significantly up-regulated the expression of ZO-1 and OCLN, and reduced the paracellular permeability of IPEC-J2 cells. This effect is most pronounced after 48 h, thus 48 h has been established as the standard treatment duration for subsequent studies.

Data-independent acquisition (DIA)-based quantitative proteomics was employed to identify and quantify proteins for untargeted proteomics experiments (17). This analysis identified a total of 3,692 differentially expressed proteins (DEPs), with 1,598 proteins upregulated and 2094 proteins downregulated in the α-GML-treated group compare with the DMSO group (Supplementary Figure 1A). The result of GO analysis indicated that α-GML affected biological processes, molecular functions and cellular components in cells. And the top three biological processes were cellular processes, metabolic processes, and biological regulation (Supplementary Figure 1B). Further insights from the KEGG pathway analysis implicated the MAPK signaling pathway as a potential mechanism related to TJs, based on the prevalence of common molecular targets and the significance of the p-values, suggesting that α-GML regulates the expression of TJ proteins through the PKC/MAPK pathway(Figure 1A). The result of Reactome analysis showed that MAPK might phosphorylate activating transcription factor 2 (ATF-2) (Figure 1B). It is reported that ATF-2 had close relevance to intestinal epithelial barrier (18, 19). Therefore, ATF-2 may be a potential transcription factor for up-regulating TJ expression.

To investigate the role of PKC in TJ proteins expression, we treated cells with the PKC inhibitor Staurosporine during culture and assessed changes at both the transcriptional and protein levels. Western blot analysis demonstrated a marked reduction in ZO-1 and OCLN protein expression in cells treated with 0.5 μM Staurosporine, contrasting with a significant increase in the α-GML-treated group compared to the DMSO control (p < 0.01 or p < 0.001). In the Staurosporine and α-GML co-treated groups, protein expression was progressively inhibited with increasing inhibitor concentration (Figures 2A–C, p < 0.05, p < 0.01 or p < 0.001). And the trend of TJ mRNA expression in the experimental group was consistent with the trend of TJ proteins expression (Figures 2D,E, p < 0.05, p < 0.01 or p < 0.001). We further treated the cells with MAPK inhibitor SCH772984, and the results were consistent with the trend observed using the PKC inhibitor (Figures 3A–E, p < 0.05, p < 0.01 or p < 0.001). In addition, the results showed that SCH772984 exposure significantly increased paracellular permeability (p < 0.001), indicating that SCH772984 impaired the cell barrier function. Co-treatment of SCH772984 and α-GML showed that α-GML functionally reversed the increase of the concentration of FITC-dextran induced by SCH772984 (Figure 3F, P < 0.001).

Since MAPK is downstream of PKC, cells are treated with PKC inhibitor Staurosporine to detect the phosphorylation level of MAPK protein by western blot assay. The results indicated that Staurosporine significantly decreased the expression of p-MAPK (p < 0.05 or p < 0.01). However, co-treatment with α-GML reversed the reduction in p-MAPK expression induced by Staurosporine (p < 0.01), with no significant differences observed in total MAPK expression across all groups (Figures 3G–K, p > 0.05). These findings suggest that α-GML’s upregulation of TJ expression is indeed associated with the PKC/MAPK pathway.

According to the above results, we proposed the mechanisms by which α-GML might promote the expression of TJ proteins by affecting MAPK/ATF-2 signaling pathway. The results showed that the treatment of α-GML significantly up-regulated the phosphorylation level of ATF compared with DMSO group (p < 0.05). Co-treatment of MAPK inhibitor SCH772984 and α-GML showed that SCH772984 functionally reversed the increase of the expression of p-ATF-2 induced by α-GML (p < 0.01). Expression of ATF-2 protein was not obviously different in all groups (Figures 4A–C, p > 0.05). To investigate the role of ATF-2 in modulating the expression of TJ proteins, we transfected cells with the plasmid designed for ATF-2 overexpression. The results showed that the expression level of ATF-2 and p-ATF-2 was significantly up-regulated (Figures 4D–G, p < 0.01, p < 0.001). Similarly, the expression level of TJ proteins was up-regulated after overexpressed with ATF-2 in cells (Figures 4H–L, p < 0.05, p < 0.01, p < 0.001).

Given that ATF-2 is a transcription factor, we postulated that ATF-2 may activate TJ transcription by directly binding to the promoter, thereby enhancing protein expression. We predicted the presence of ATF-2 binding sites upstream of the ZO-1 and OCLN transcription start sites (TSS) utilizing the JASPAR database available at https://jaspar.elixir.no/ (Figure 4M). Our findings identified a consensus ATF-2 binding site within the ZO-1 TSS, located between −565 bp and −577 bp (CATGACCTCATCA) with a relative profile score threshold 98.54%. What’s more, the OCLN promoter sequence harbors two distinct ATF-2 binding sites, located at −4,078 bp to −4,087 bp and −4,732 bp to −4,741 bp (CTGAGGTCAG), respectively, with a relative profile score threshold of 89.72%. To further validate the role of the ATF-2 binding sequence in ZO-1 promoter activity, a site-directed mutagenesis was performed to mutate the ATF-2 binding site (CATGACCTCATCA to CGTCAAGACGAC) within the ZO-1 promoter region. The resulting plasmid, along with the ATF-2 overexpression plasmid, was co-transfected into the cells using a dual luciferase reporter assay system. We found that the ATF-2 overexpression plasmid caused a significant increase in ZO-1 promoter activity (p < 0.001). The mutation of the ATF-2 binding site inhibited the increase in ZO-1 promoter activity mediated by ATF-2 overexpression (Figures 4N,O, p > 0.05).

To delineate the mechanism of TJ expression mediated by ATF-2, we investigated the interplay between MAPK and ATF-2. Cells were then pretreated with SCH772984 and ATF-2 overexpression plasmid, western blot assay and qRT-PCR was used to detect TJ and ATF-2 expression. The findings indicated that SCH772984 markedly reduced the protein expression of p-ATF-2, ZO-1, and OCLN (p < 0.05 or p < 0.01), whereas the expression level of ATF-2 remained unchanged in comparison to the control group (p > 0.05). Concurrent treatment with SCH772984 and OE-ATF-2 demonstrated that OE-ATF-2 was able to functionally reverse the decrease in p-ATF-2 and TJ proteins expression induced by SCH772984 in IPEC-J2 cells (Figures 5A–E, p < 0.05, p < 0.01), yet the expression level of ATF-2 still did not exhibit a significant change (p > 0.05). The mRNA expression trends for ATF-2 and TJ were consistent with the protein expression patterns (Figures 5F–H, p < 0.05, p < 0.01 or p < 0.001). The results confirmed that α-GML treatment elevated the phosphorylation status of the transcription factor ATF-2 by activating MAPK, subsequently increasing the expression of TJ proteins.