We conducted this investigation at Anhui University of Chinese Medicine from June 1, 2023, to March 1, 2025.

Isoschaftoside (purity $\ge$98% by HPLC) was purchased from Taizhou Danding Biotechnology Co., Ltd. (No. YJ0871, Taizhou Danding Biotechnology, Taizhou, Jiangsu, China). Mouse IL-6/1$\beta$ (No. JL20268, JL18442), TNF-$\alpha$ (No. JL10484), COX-1 (No. JL13697), and PGE2 (No. JL12640) ELISA kits were acquired from Shanghai Jianglai Biotechnology Co., Ltd. (Shanghai, China). Omeprazole magnesium enteric-coated tablets (No. SAWX) were purchased from AstraZeneca (London, UK).

C57BL/6J mice (6–7 weeks old, male, 20–25 g) were obtained from Hangzhou Ziyuan Experimental Animal Technology Co., Ltd. (Zhejiang, China). Mice were kept under standard specific pathogen-free (SPF) circumstances (humidity 55 $\pm$ 10%, 22 $\pm$ 2 °C, 12-h light/dark cycle) with unrestricted standard rodent chow and water. All experimental procedures received approval from the Experimental Animal Ethics Committee of Anhui University of Chinese Medicine No. AHUCM-mouse-2023169 and were conducted as per the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals. After a week of acclimatization, mice were classified into six groups in a random manner (n = 12 per group): Control, Model, Omeprazole (OME, 20 mg/kg [20]), low-dose isoschaftoside (Is-L, 7.8 mg/kg), medium-dose isoschaftoside (Is-M, 31.2 mg/kg), and high-dose isoschaftoside (Is-H, 93.6 mg/kg). The sample size was detected depending on earlier investigations and power analysis to verify statistical reliability. The Control group administered 0.5% sodium carboxymethyl cellulose (CMC-Na) (No. MB1717, MeilunBio, Dalian, Liaoning, China) solution by oral gavage daily. The remaining groups received aspirin (No. A2093-100G, Sigma, MO, USA) (300 mg/kg in 0.5% CMC-Na [21], by oral gavage each morning) for 14 consecutive days to establish a gastric mucosal injury model. Additionally, the OME group received omeprazole (20 mg/kg, in 0.5% CMC-Na), while the Is-L, Is-M, and Is-H groups were administered isoschaftoside at 7.8, 31.2, and 93.6 mg/kg in 0.5% CMC-Na, respectively, each afternoon. The administration volume was 10 mL/kg body weight.

The mice were anesthetized via intraperitoneal injection of 3% pentobarbital sodium (No. P3761, Sigma, MO, USA) at a dose of 30 mg/kg prior to blood and stomach tissue collection. The procedure was as follows: pentobarbital sodium powder was dissolved in physiological saline to achieve the appropriate concentration. The mice’s abdominal regions were disinfected with alcohol swabs, and the animals were restrained using forceps. A 1 mL syringe was used to draw the prepared pentobarbital sodium solution, which was then injected intraperitoneally into either side of the lower abdomen. Following the injection, mice were monitored for their physiological responses. Anesthetic effects typically appeared within a few minutes, and the mice entered a state of somnolence.

The HE staining experiment was performed according to standard histological techniques [22]. Following euthanasia by cervical dislocation, the stomach was excised, incised along the greater curvature, and rinsed with physiological saline to remove gastric contents. The mucosa was then laid flat, and comparable regions of gastric tissue were collected from each group. Samples underwent fixing in 4% paraformaldehyde (No. 60536ES60, Yeasen, Shanghai, China), dehydrating, embedding in paraffin, sectioning at 5 µm thickness, and staining with HE. Histological evaluation was conducted using Olympus CX23 microscopy (Tokyo, Japan), and images were acquired for analysis.

Immunohistochemistry was performed as previously described [23]. Paraffin-embedded gastric tissue slices were dewaxed and exposed to antigen retrieval by heating in antigen retrieval solution. A 10-min blockage of endogenous peroxidase activity was conducted with 3% H₂O₂ (No. 216763, Sigma, MO, USA) at room temperature. An incubation of the slices was then conducted with anti-MUC2 antibody (ab272692, 1:200, Abcam, Cambridge, UK) overnight at 4 °C, then with horseradish peroxidase (HRP)-conjugated secondary antibody (ZB-2301, 1:20000 dilution, Zsbio, Beijing, China) at 37 °C for 30 min. HRP-conjugated streptavidin was subsequently applied for 30 min at 37 °C. Staining was visualized with diaminobenzidine (DAB) (No. YT8204, ITTABIO, Beijing, China), followed by counterstaining with hematoxylin (No. ZY-6144, ZYBio, Shanghai, China), dehydrating through graded ethanol (No. 10009218, Sinopharm, Shanghai, China), clearing in xylene (No. 10023428, Sinopharm, Shanghai, China), mounting, and examining microscopically. For negative controls, the primary antibody was omitted. Given that MUC2 is well-known for its responsive overexpression to repair the gastric mucosa under injury [13], gastric tissue damaged by aspirin was employed as a positive control. Morphometric analysis was performed by capturing five random fields/section at 200$\times$ magnification. The area of positive DAB staining (brown color) was quantified using ImageJ software (v1.53, NIH, Bethesda, MD, USA) with the “Positive Pixel Count” algorithm.

For serum collection, the peri-oral hair was trimmed, and the eyeballs were gently enucleated following retro-orbital pressure with ophthalmic forceps. The collection of blood samples was conducted in 1.5 mL microcentrifuge tubes, allowed to clot at room temperature for 30 min, and centrifugation was conducted at 3000 rpm for 10 min at 4 °C for serum separation. ELISA kits were equilibrated to room temperature prior to use, and all procedures (dilution, sample addition, incubation, washing, chromogenic reaction, termination) were conducted as per the manufacturer’s guidelines. Standard curves were created to quantify serum concentrations of IL-6/1$\beta$, COX-1, TNF-$\alpha$, and PGE2.

Protein expression analysis was performed using a standard Western blotting protocol [24]. Processed gastric tissue samples were lysed on ice in radioimmunoprecipitation assay (RIPA) buffer containing phenylmethylsulfonyl fluoride (PMSF) (No. MA0001, Meilunbio, Dalian, Liaoning, China) for 30 min, and protein levels were detected via a bicinchoninic acid (BCA) assay kit (No. B5001, Land Bridge Technology, Beijing, China). Equal protein quantities were exposed to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) (No. P1200-25T, Solarbio Science & Technology, Beijing, China) and transferred to polyvinylidene fluoride (PVDF) membranes (No. IPVH00010, Millipore, Billerica, MA, USA). A 2-h blockage of membranes was conducted with 5% (w/v) non-fat dry milk at 25 °C, then an incubation was conducted with primary antibodies for 2 h, then with HRP-conjugated secondary antibodies for 1 h. The Amersham Imager 600 imaging system (GE Health, Chicago, IL, USA) and enhanced chemiluminescence (ECL) reagent (No. BL520A, Biosharp, Hefei, Anhui, China) were utilized for detection. Densitometric analysis of PI3K, AKT, caspase-9, BCL-2, and BAX protein bands was conducted via ImageJ software (v1.53, NIH, USA). Primary antibodies include Bax (50599-2-1g, 1:2000 dilution, Proteintech, IL, USA), Bcl-2 (ab32124, 1:1000 dilution, abcam, Cambridge, UK), cleaved-casepase-9 (9509, 1:1000 dilution, CST, MA, USA), procasepase-9 (ab32539, 1:5000 dilution, abcam, Cambridge, UK), PI3K (ab86714, 1:1000 dilution, abcam, Cambridge, UK), p-PI3K (ab182651, 1:1000 dilution, abcam, Cambridge, UK), AKT (4691s, 1:1000 dilution, CST, MA, USA), p-AKT (4060s, 1:2000 dilution, CST, MA, USA), and GAPDH (TA-08, 1:2000 dilution, Zsbio, Beijing, China).

SPSS 26.0 (IBM Corp., Armonk, NY, USA) was utilized to conduct statistical analysis, and data are expressed as mean $\pm$ standard deviation. The data distribution normality was verified via the Shapiro-Wilk test. One-way analysis of variance (ANOVA) then Tukey’s post-hoc test made comparisons among multiple groups. A p-value 0.05 was regarded as significant.

HE staining of gastric tissues showed that mice in the Control group exhibited intact gastric mucosa with well-preserved architecture, closely arranged glands, and absence of mucosal bleeding or edema. In contrast, the Model group displayed disorganized and sparse glandular arrangement, partial loss of mucosal epithelial cells, and marked infiltration of inflammatory cells. Compared to the Model group, mice in the OME group demonstrated relatively preserved mucosal structure, more regular glandular organization, and reduced inflammatory cell infiltration. Isoschaftoside treatment produced a dose-dependent improvement, evidenced by increasingly regular and distinct glandular arrangement as the dose increased, along with decreased inflammatory cell infiltration (Fig. 1).

Fig. 1.

Hematoxylin and eosin (HE) staining of mouse gastric mucosal tissue sections. (A) Control, the gastric mucosa is intact with a well-preserved architecture, and the glands are tightly packed and well-aligned. (B) Model, the gastric mucosal structure is severely damaged, accompanied by marked inflammatory cell infiltration. (C) Omeprazole (OME) (20 mg/kg), the gastric mucosal structure is relatively intact with fairly orderly arranged glands and shows only minor inflammatory cell infiltration. (D) Low-dose isoschaftoside (Is-L) (7.8 mg/kg). (E) Medium-dose isoschaftoside (Is-M) (31.2 mg/kg). (F) High-dose isoschaftoside (Is-H) (93.6 mg/kg), with increasing Is concentration, the glands became progressively more organized and distinct, and the inflammatory cell infiltration was reduced. Scale bar, 50 µm. The black arrows indicate areas of inflammatory cell infiltration or injury.

To elucidate the protective action of isoschaftoside on the gastric mucosa, MUC2 expression was examined by immunohistochemistry. The outcomes illustrated that MUC2 expression in the gastric tissue of mice in the Model group was significantly higher than the Control group (p 0.001). Following isoschaftoside treatment, MUC2 expression was reduced dose-dependently manner compared to the Model group (Is-M, p 0.01; Is-H, p 0.001). The level of MUC2 in the Is-H group was not significantly different from that in the OME group (p $>$ 0.05) (Fig. 2).

Fig. 2.

Detection of mucin 2 (MUC2) expression in mouse gastric mucosa by immunohistochemistry. (A) Control, shows weak positive staining. (B) Model, exhibits strong positive staining for MUC2, indicating a significant upregulation. (C) OME (20 mg/kg), displays moderate positive staining for MUC2. (D) Is-L (7.8 mg/kg), strong positive staining. (E) Is-M (31.2 mg/kg) and (F) Is-H (93.6 mg/kg), moderate positive staining, suggesting a partial reversal of MUC2 overexpression induced in the Model group. Con represent Control group. ^#⁢#⁢#p 0.001 vs Con. *p 0.05, and **p 0.01 vs Model. Bars signify mean $\pm$ SD (n = 3). Scale bar, 100 µm.

To assess systemic inflammatory responses, serum cytokine levels were measured in mice. Compared with the Control group, mice in the Model group exhibited significantly elevated concentrations of the pro-inflammatory cytokines IL-6, IL-1$\beta$, and TNF-$\alpha$ (p 0.001), and significantly reduced levels of the anti-inflammatory mediators COX-1 and PGE2 (p 0.001). Administration of isoschaftoside resulted in a dose-dependent attenuation of the increases in IL-6, IL-1$\beta$, and TNF-$\alpha$ (p 0.001), accompanied by significant elevation of COX-1 and PGE2 levels relative to the Model group (p 0.001) (Fig. 3). The effects of high-dose isoschaftoside on these cytokines were not significantly different from those of omeprazole (p $>$ 0.05).

Fig. 3.

ELISA detection of IL-6, IL-1$\beta$, TNF-$\alpha$, COX-1, and PGE2 levels in mouse serum. Con represent Control group. ^#⁢#⁢#p 0.001 vs Con. ***p 0.001 vs Model. Bars represent mean $\pm$ SD (n = 3). PGE2, prostaglandin E2.

The levels of the anti-apoptotic protein Bcl-2 and the pro-apoptotic proteins Bax and caspase-9 were investigated. Following treatment with isoschaftoside, it was observed that this compound could counteract the aspirin-induced decrease in Bcl-2 expression (Is-M, p 0.01 and Is-H, p 0.001 vs Model) and the increase in cleaved-caspase-9 (Is-M, Is-H, p 0.001 vs Model) and Bax expression (Is-M, Is-H, p 0.001 vs Model) dose-dependently manner, and Bax/Bcl-2 level in Is-H group were increased than Model group (p 0.001) (Fig. 4). The Is-H effect on these proteins was comparable to that of OME (p $>$ 0.05). This indicates that isoschaftoside resistance to gastric mucosal injury is related to apoptosis.

Fig. 4.

The isoschaftoside effect on the expression of correlated proteins. (A) Western blot analysis. (B–G) Protein expression levels. Con represent Control group. **p 0.01 and ***p 0.001 vs Con. ^#⁢#p 0.01, and ^#⁢#⁢#p 0.001 vs Model. Bars signify mean $\pm$ SD (n = 3). Bcl-2, B-cell lymphoma 2; Bax, Bcl-2-associated X protein; p-AKT, phospho-protein kinase B.

The regulatory effects of isoschaftoside on PI3K/AKT pathway-correlated proteins p-PI3K and p-AKT were examined. The outcomes indicated that the p-PI3K and p-AKT levels increased following aspirin-induced injury (p 0.001 vs Control). However, after treatment with isoschaftoside, the levels of p-PI3K and p-AKT were significantly lower than those in the Model group (p-PI3K: Is-M p 0.01, Is-H p 0.001; p-AKT: Ia-M and Is-H p 0.001) (Fig. 4). The suppression of PI3K/AKT phosphorylation by Is-H was not significantly different from that by OME (p $>$ 0.05). This suggests that isoschaftoside may exert anti-apoptotic effects by suppressing p-PI3K and p-AKT, thereby protecting gastric mucosal cells from damage.