All animals were obtained from the Shanghai Slaccas Laboratory Animal Co. Ltd. (Shanghai, China). All experimental protocols were approved by the Institutional Animal Care and Use Committee at Fudan University School of Pharmacy and were performed in strict accordance with the guidelines.

Female C57BL/6J mice were maintained at 20–25°C with 12/12-h light/dark cycles and ad libitum access to food and water. The experiments began when the mice were 10 weeks old and weighed approximately 20 g. Mice were distributed as follows (n = 6 in each group): control group, PD group, 5/6 Nx group, 5/6 Nx group + PD group, PD + AS IV group, 5/6 Nx + PD + AS IV group. Mice in all 5/6Nx-related groups underwent 5/6 nephrectomy to induce uremia (Tan et al., 2019). The nephrectomy involved the removal of two-thirds of the left kidney and the complete removal of the right kidney, with a 1-week interval between the two resections. After day 0 (before 5/6 nephrectomy) and day 21, serum urea and creatinine levels were measured to confirm the successful modeling of uremia. The mice were then exposed to 2 ml of standard PDF (4.25% g/dL dextrose) daily for a period of 6 weeks in all PD groups (Ferrantelli et al., 2016; Yang et al., 2021). Mice in the PD + AS IV group and the 5/6 Nx + PDF + AS IV group were intragastrically treated with AS IV (10 mg/kg/d, purity ≥98%, Adamas, Shanghai, China) (Jiang et al., 2017; Gao et al., 2020) for 6 weeks.

Male Sprague-Dawley (SD) rats weighing 250 g were used. After 1 week of adaption, 5/6 nephrectomy was performed to induce uremia, then PDF was performed daily for 6 weeks, as described previously. All rats were randomly divided into the following groups (n = 3 in each group): control group, PD group, 5/6 Nx group, 5/6Nx + PD group. Health conditions of animals were checked daily.

After harvesting the animals, the colon was collected for pathological analysis. Colon tissues were fixed in 4% paraformaldehyde overnight and embedded in paraffin. Subsequently, colonic sections were sectioned at a thickness of 4-μm thickness and stained with hematoxylin and eosin (H&E) and periodic acid-Schiff (PAS).

For TEM studies, colonic tissues were cut into small blocks and fixed in a cold buffered solution containing 2% glutardialdehyde and 2% paraformaldehyde. Specimens were cut into semi-thin sections (0.5 μm) and ultrathin sections (70–90 nm) from the interested regions and examined using a transmission electron microscope (TEM 109, Zeiss, Jena, Germany).

For IF assays, the colon sections were deparaffinized with xylene and rehydrated with consecutive ethanol washes. The samples were incubated with anti-occludin antibody (Abcam, Cambridge, MA, USA) and washed with PBS. T84 cells were incubated with antibody against β-catenin (Proteintech, Rosemount, IL, USA), and nuclei were stained with 40,6-diamidino-2-phenylindole (DAPI). For IHC staining, sections were subjected to antigen retrieval using citric acid buffer and then incubated with primary antibody against β-catenin (Proteintech, Rosemount, IL, USA) at 4°C overnight. After three washes with Tris-buffered saline, the sections were incubated with horseradish peroxidase-conjugated Affinipure goat anti-rabbit IgG. The immunoreactivity was visualized by treatment using the Dako Envision kit HRP (K4006, Dako). The images were captured by a confocal microscope.

T84 cells were obtained from the Culture Collection of the Chinese Academy of Sciences, Shanghai, China. T84 cells were cultured in DMEM/F-12 (BIOAGRIO, Shanghai, China) medium supplemented with 10% fetal bovine serum (Gibco, Gaithersburg, MD, USA) at 37 °C with 5% humidified CO₂. After 3-5 passages, T84 cells were seeded into 6-well plates, and experiments were performed when a 50–60% confluence was reached. The cells were exposed to 4.25% dextrose-containing PDF and AS IV (1, 5, and 10 μM) for various time periods. Subsequently, MK-2206 (AKT inhibitor, MedChemExpress, Princeton, NJ, USA) dissolved in DMSO at a concentration of 5 μM was added to the AS IV (5 μM) adminstration group to identify the role of AKT in this setting.

The effects of AS IV on T84 cell viability were assessed with the Cell Counting Kit-8 (CCK8, Beyotime Biotechnology, Beijing, China). In detail, 5 ×10³ cells/well were seeded in 96-well plates and exposed to various concentrations of AS IV for 24 or 48 h. At the end of the experiment, 10 μl of CCK8 solution was added and co-cultured at 37 C for 2 h, then the absorbance value was calculated at 450 nm.

T84 cells were seeded in 12-well plates and scratch wounds were made with a sterile 200 μL pipette tip. The migration of cells into the wound area was observed and photographed at 0 and 48 h using a microscope. The migration distance was measured by ImageJ (Rawak Software Inc. Germany).

Cell and colon tissues were lysed with RIPA buffer containing protease and phosphatase inhibitors and proteins were collected by centrifuging. The proteins were separated by SDS-PAGE and transferred onto a polyvinylidene difluoride (PVDF) membrane (Millipore Corporation, Billerica, MA, USA). The PVDF membranes were then incubated with the corresponding primary antibodies of anti-GAPDH, anti-β-catenin and anti-phosphorylated (p)-GSK3β^Ser9 (Proteintech, Rosemount, IL, USA); anti-ZO-1, anti-occludin, and anti-p-AKT^S473 (Abcam, Cambridge, MA, USA); at 4 °C overnight. The membranes were washed 3 times and horseradish peroxidase-conjugated secondary antibodies were added. Finally, specific immunoblots were detected by chemiluminescence. The grey density of the protein bands was normalized to the respective control according to the manufacturer’s instructions.

Total RNA from T84 cells was extracted by Trizol reagent (Takara, Japan) according to the manufacturer’s instructions. The cDNA was synthesized using PrimeScript TM RT Master Mix (Perfect Real Time, Takara, Japan), and then each sample was analyzed using a BIO-RAD CFX Connect Real-time PCR system with SYBR Premix Ex Taq TM (Takara, Japan) and the specific primers. Relative quantification of gene expression was performed using the 2^-△△Cq as fold changes. The designed gene-specific primers are shown in Table 1.

All data were analyzed by one-way analysis of variance (ANOVA) or unpaired two-tail Student’s t-test with GraphPad Prism (Version 8.0; GraphPad Software, San Diego, CA, USA), and a p-value of <0.05 was considered as statistically significant.

We adopted two different mice PD models with uremic mice and normal mice. The prolonged exposure to PDF caused colon length atrophy in the PD group and in the 5/6Nx + PD group (Figures 1B,C). Furthermore, in the PD-related groups, H&E staining showed the infiltration of inflammatory cells, the defects of crypts and the irregular arrangement of the crypts in the colon (Figure 1D) and the PAS staining results revealed the lower number of goblet cells (Figure 1E). These results demonstrated that long-term PDF treatment would cause pathological damage to the colon in normal or uremic mice, which was also confirmed in rat PD models (Supplementary Figure S1). Interestingly, AS IV (Figure 1A) significantly alleviated colon length atrophy, improved crypt morphology, prevented inflammatory infiltration, and increased expression of expression of goblet cells expression (Figures 1D,E). Furthermore, we detected an ultrastructural alteration in the intestinal mucosal epithelium before and after AS IV treatment. TEM of the intestinal epithelium showed intact tight junctions (TJ), orderly microvilli, and assembled desmosome in the control group. After long-term PDF administration, we observed loss of the TJ membrane and desmosome and abnormal appearance of microvilli in the PD group and in the 5/6Nx + PD group, while treatment with AS IV alleviated these distortions (Figure 1F).

Intestinal epithelial cells and TJs constitute the intestinal epithelial barrier, which controls the transport of molecules across cells and regulates intestinal permeability (Kumar et al., 2021; Kuo et al., 2021). Here, we performed western blotting and RT-PCR to assess altered levels of TJ proteins, including ZO-1 and occludin, after continuous PDF exposure or AS IV administration. Interestingly, after long-term PDF treatment, the protein expression of ZO-1 and occludin decreased in the PD group and in the 5/6Nx + PD group, but was significantly reversed by AS IV (Figures 2A,B). Furthermore, the RT-PCR results showed a decrease in ZO-1 and occludin expression during PD compared to the control group, but mRNA expression levels were up-regulated after AS IV administration (Figure 2C). Together, these data demonstrated that AS IV could promote the repair of the intestinal epithelial barrier in vivo.

We explored the effects of PDF and AS IV on T84 cells in vitro. First, we found that AS IV itself could promote T84 cell proliferation and improve cell viability (Figure 3A). As shown in Figure 3B, cell viability decreased dramatically with exposure to PDF, while AS IV (especially at 5 μM) significantly promoted recovery of cell viability in T84 cells. Wound healing is important for the repair of the damaged intestinal epithelial barrier. When PDF was applied to T84 cells for 48 h, cell growth and migration were severely inhibited. Supplementation with AS IV improved the closure of the scratch wound at various concentrations, and maximum efficacy was observed at 5 μM (Figures 3C,D). These results suggested that AS IV promoted intestinal mucosal barrier repair by improving cell activity and promoting wound healing.

Consistent with the results in vivo, the deleterious effect of PDF on TJs was also observed in vitro. As shown in Figures 4A,B, after T84 cells were exposed to PDF for 72 h, the expression of TJ proteins, including ZO-1 and occludin, was severely down-regulated and the expression decreased in a dose- and time-dependent manner after PDF exposure (Supplementary Figure S2A, B). Various doses of AS IV were applied for treatment, the expression of the ZO-1 and occludin proteins was significantly up-regulated, which indicated that AS IV could improve intestinal permeability in vitro (Figures 4A,B). Furthermore, AS IV also increased the levels of ZO-1 and occludin mRNA in vitro (Figure 4C).

To clarify the underlying mechanism whereby AS IV protected T84 cells against PDF-related intestinal epithelial barrier damage, we explored the AKT/GSK3β/β-catenin pathway. As shown in Figure 5A, exposure to PDF reduced the phosphorylation levels of AKT (Ser473) and GSK-3β (Ser9), and concomitantly increased β-catenin degradation. Interestingly, compared to the PDF-treated group, supplementation with AS IV increased AKT phosphorylation levels (Ser473), its activated form, then phosphorylated its downstream GSK-3β (Ser9), the inactivated form. Furthermore, combining the results of western blotting and IF nuclear localization, AS IV subsequently increased the accumulation and nuclear translocation of β-catenin (Figures 5A–C). Therefore, these data demonstrated that AS IV may exert its activity through the AKT/GSK3β/β-catenin pathway. To further verify the underlying pathway, we added 5 μM MK-2206 to the AS IV group. As expected, the addition of MK2206 significantly inhibited the efficacy of AS IV, resulting in a decrease in cell viability (Figure 5D) and downregulated the expression of ZO-1 and occludin (Figures 5E,F). Altogether, these results suggested that AS IV exerted therapeutic effects by improving cell viability and promoting the expression of TJ proteins through the AKT/GSK3β/β-catenin pathway.

To validate the underlying mechanism of AS IV, we explored the AKT/GSK-3β/β-catenin pathway in vivo. In line with the above findings, western blotting and IHC analyses showed decreased levels of p-AKT (Ser473) and GSK-3β (Ser 9), which was accompanied by a lower expression of β-catenin in the PD group and the 5/6Nx + PD group. However, for the PD + AS IV group and 5/6Nx + PD + AS IV group, we found that administration of AS IV could activate the biological activity of AKT (Ser473), and then inhibit the activity of GSK-3β (Ser 9), ultimately leading to the increased accumulation of β-catenin (Figures 6A–C). These data further confirmed that AS IV promoted the repair of the intestinal mucosal epithelial barrier repair via the AKT/GSK3β/β-catenin signaling axis.