Yiwei Xiaoyu granules mitigate spasmolytic polypeptide-expressing metaplasia-associated gastric mucosal damage by regulating WFDC2 and the miRNA-7/circRNA network

Abstract

BACKGROUND

Gastric cancer develops through a sequence of atrophy, metaplasia, and dysplasia. Spasmolytic polypeptide-expressing metaplasia (SPEM) is a pivotal precancerous lesion associated with inflammation and noncoding RNA dysregulation. In traditional Chinese medicine (TCM), these pathological changes overlap with spleen-stomach deficiency syndrome (SSDS). Yiwei Xiaoyu granules (YWXY), a classical TCM formula, have demonstrated clinical benefit in chronic atrophic gastritis, yet their mechanistic actions in SPEM combined with SSDS remain unclear.

AIM

To investigate the protective effects and mechanisms of YWXY in a tamoxifen-induced SPEM and SSDS composite model.

METHODS

A mouse model of tamoxifen-induced SPEM, with or without SSDS, was established. The effects of YWXY on gastric mucosa were evaluated by histological scoring, quantitative polymerase chain reaction (qPCR), and fluorescence in situ hybridization (FISH). Key molecular endpoints included inflammatory cytokines [interleukin (IL)-1β and tumor necrosis factor (TNF)-α], the miR-7a-5p/Cdr1as axis, and the metaplastic marker WFDC2. Statistical analysis was performed using two-way analysis of variance followed by Tukey’s multiple comparisons.

RESULTS

YWXY exhibited time-dependent modulation of IL-1β, normalizing its expression at day 15 and suppressing persistent elevation at day 30. TNF-α overexpression in SPEM was significantly reduced by YWXY at both time points. Cdr1as downregulation in pathological groups was partially reversed by YWXY, while miR-7a-5p suppression in SSDS and SPEM + SSDS was restored toward baseline. WFDC2 induction in both SPEM and SSDS was significantly attenuated by YWXY, confirmed by qPCR and FISH. These findings indicate coordinated regulation of inflammatory cytokines, ncRNA networks, and metaplastic markers.

CONCLUSION

YWXY alleviates gastric mucosal injury in combined SPEM and SSDS by regulating IL-1β and TNF-α, restoring the miR-7a-5p/Cdr1as axis, and suppressing WFDC2 expression.

Key Words: Spasmolytic polypeptide-expressing metaplasia; Spleen-stomach deficiency syndrome; Yiwei Xiaoyu granules; miR-7a-5p/Cdr1as; WFDC2

Core Tip: Spasmolytic polypeptide-expressing metaplasia (SPEM) represents a key precancerous stage of gastric carcinogenesis and frequently coexists with spleen-stomach deficiency syndrome (SSDS) in clinical practice. Using a combined tamoxifen-induced SPEM and SSDS mouse model, we demonstrate that Yiwei Xiaoyu granules exert sustained protective effects on gastric mucosa. Yiwei Xiaoyu granules bidirectionally regulates inflammatory cytokines, restores the miR-7a-5p/Cdr1as axis, and suppresses WFDC2 induction, thereby linking syndrome-based intervention with molecular mechanisms of gastric precancerous lesions.

INTRODUCTION

Gastric cancer (GC) remains a major public health concern[1], yet the mechanisms that connect early gastric injury to malignant progression are not fully resolved[2-4]. While Helicobacter pylori infection is recognized as a key initiator of chronic gastritis and metaplastic transformation[5], eradication therapy alone does not completely eliminate the long-term risk of cancer[6,7]. Increasing evidence suggests that additional molecular events[8], beyond microbial infection, contribute to the transition from mucosal inflammation to precancerous lesions[9].

Among the different metaplastic changes, spasmolytic polypeptide-expressing metaplasia (SPEM) has attracted attention as a pivotal stage in the remodeling of gastric glands[10-12]. SPEM is driven by the reprogramming of chief cells and is accompanied by alterations in noncoding RNA signaling networks, including miRNAs and circRNAs, which regulate inflammatory and epithelial responses[9]. Of particular interest, the circRNA Cdr1as functions as a sponge for miR-7, as previously demonstrated[13]. Given the tumor-suppressive role of miR-7 and its involvement in gastric carcinogenesis, dysregulation of the Cdr1as/miR-7 axis has been implicated in linking inflammation, metaplasia, and neoplastic progression[14,15].

From the perspective of traditional Chinese medicine (TCM), precancerous lesions of GC (PLGCs) are frequently associated with spleen-stomach deficiency syndrome (SSDS), which represents an impaired capacity to maintain mucosal and metabolic homeostasis[16,17]. Yiwei Xiaoyu granules (YWXY), a classical herbal prescription, have been clinically applied to treat chronic atrophic gastritis and SSDS, and trials indicate that they can improve intestinal metaplasia scores in affected patients[18]. Our previous experimental work also showed that YWXY can restore miR-7 expression in a tamoxifen-induced SPEM model[19]. To better recapitulate the clinical context-where patients often present with both SSDS and premalignant lesions-we established a combined model of tamoxifen-induced SPEM plus SSDS. This strategy allowed us to evaluate how YWXY modulates inflammatory cytokines, noncoding RNA networks, and mucosal remodeling in an integrated disease-syndrome setting.

MATERIALS AND METHODS

Ethical statement

All animal experiments were reviewed and approved by the Institutional Animal Care and Use Committee of Chongqing Traditional Chinese Medicine Hospital. All procedures were performed in accordance with the Guidelines for the Care and Use of Laboratory Animals and relevant institutional regulations. Efforts were made to minimize animal suffering and to reduce the number of animals used. No unexpected adverse events or animal deaths occurred during the study.

Animal grouping and treatment

Male BALB/c mice (6-8 weeks old, weighing 20-25 g) were obtained from the Institute of Chinese Medicine in Chongqing. Upon arrival, animals were acclimatized under specific pathogen-free conditions with a 12 hours light/dark cycle at 22 ± 2 °C and ad libitum access to food and water. All animals were confirmed to be healthy and had not undergone any prior experimental manipulation. After acclimatization, mice were randomly assigned to experimental groups (n = 10 per group) using a random number table. Investigators were blinded during outcome assessment. Sample size was determined based on previous studies. An overview of the experimental design is provided in Figure 1.

Figure 1 Simplified overview of the experimental design. A spleen-stomach deficiency syndrome associated gastric injury model was established using Xiao Chengqi decoction. Following model induction, mice received therapeutic intervention with Yiwei Xiaoyu granules, with or without tamoxifen administration. Gastric tissues were collected at predefined time points (days 15 and 30) for histological, immunofluorescence, and molecular analyses. Detailed treatment schedules and group allocations are provided in Supplementary Figure 1. SSDS: Spleen-stomach deficiency syndrome; XCQ: Xiao Chengqi decoction; YWXY: Yiwei Xiaoyu granules. This figure was created by http://biorender.comhttps://http://biorender.com/8ozzrx8.

An SSDS-associated gastric injury model was established using Xiao Chengqi decoction (XCQ) administration combined with alternating swimming or fasting, as previously described[20]. Following model induction, mice received therapeutic intervention with YWXY, with or without tamoxifen administration, depending on group assignment.

To induce SPEM, tamoxifen was administered intraperitoneally (3 mg/20 g body weight) on days 12-14, as reported previously[21]. Control animals received corresponding vehicle treatments.

For experimental consistency, the intervention cycle was repeated during days 15-30 following the same procedures applied during the initial phase. Animals were prescheduled for sacrifice in two independent batches [n = 5 mice per group at each time point (days 15 and 30)]. Gastric tissues were collected at these predefined time points for subsequent histological, immunofluorescence (IF), and molecular analyses.

YWXY preparation and quality control followed standardized procedures as reported previously[19,22,23]. XCQ (Rhubarb 12 g, Magnolia officinalis 6 g, Trifoliate 6 g) was supplied by China Resources Sanjiu Medical & Pharmaceutical Co. Ltd. (Chongqing Hospital of TCM). Tamoxifen was dissolved in 10% ethanol/90% sunflower seed oil according to Saenz et al[21].

Group allocation

Mice were allocated into five groups: G1, control group; G2, SSDS model group; G3, SSDS + YWXY treatment group; G4, SSDS + SPEM + YWXY group; and G5, SSDS + SPEM model group. Detailed treatment schedules and timelines for each group are provided in Supplementary Figure 1.

ELISA

Blood samples were collected for biochemical analyses. Serum creatine kinase (CK/CPK), gastrin (GAS), and motilin (MTL) were measured using commercial ELISA kits: Mouse Creatine Kinase (JL18284), Mouse GAS (JL20584), and Mouse MTL (JL10463) (Jianglaibio, Shanghai, China).

Reverse transcription quantitative polymerase chain reaction

Total RNA was extracted from gastric tissue using the LS1040 kit (Promega, Shanghai, China). cDNA was synthesized and gene expression was analyzed by real-time reverse transcription quantitative polymerase chain reaction (RT-qPCR). Relative mRNA expression levels were normalized to β-actin and calculated using the 2^–ΔΔCt method. Primer sequences used for RT-qPCR are provided in Supplementary Table 1.

IF

Paraffin-embedded gastric tissue sections were deparaffinized and subjected to antigen retrieval with EDTA buffer (pH 8.0) using a microwave heating cycle (8 minutes medium heat, 8 minutes pause, 7 minutes low heat), and cooled to room temperature. Sections were washed three times in phosphate-buffered saline (PBS) (pH 7.4, 5 minutes each), circled with a hydrophobic barrier pen, and incubated in 3% H₂O₂ for 25 minutes in the dark. After PBS washes, primary antibodies were applied overnight at 4 °C: Anti-TFF2 (1:100, ab203237; Abcam), anti-Clusterin (1:1000, 53417-1; SAB). The next day, horseradish peroxidase-conjugated secondary antibodies were applied for 50 minutes at room temperature, followed by TSA amplification. After antigen retrieval and PBS washes, a second primary/secondary antibody set was added. Nuclei were counterstained with 4′,6-diamidino-2-phenylindole for 10 minutes, washed, and mounted with antifade medium. Images were captured under a fluorescence microscope.

Fluorescence in situ hybridization

Fluorescence in situ hybridization (FISH) analysis for miR-7a-5p and WFDC2 was performed as described previously[19]. Locked nucleic acid probes are listed in Supplementary Table 2.

Image quantification

Fluorescence images from IF and FISH were analyzed using ImageJ. For each sample, representative fields were selected under identical exposure settings. The percentage of positive area was quantified by applying a consistent threshold to binarize the fluorescence signal, followed by calculating the ratio of positive-stained pixels to the total tissue area. All image processing steps were performed in a blinded manner to minimize observer bias, and the averaged values from multiple fields were used for statistical analysis.

Statistical analysis

Primary outcomes included IF and ELISA measurements; secondary outcomes included PCR and FISH analyses. Data are presented as mean ± SD unless otherwise indicated. Statistical analyses were performed using GraphPad Prism 9.0 (San Diego, CA, United States). Group differences were assessed using one-way analysis of variance (ANOVA) with Tukey’s post hoc test, two-way ANOVA for time- and treatment-dependent effects, or student’s t test. For single-time-point comparisons, one-way ANOVA was applied, whereas two-way ANOVA was used for analyses involving both time and treatment factors. Nonparametric tests were applied when assumptions of normality were not met. Two-tailed P < 0.05 was considered statistically significant. The specific statistical methods applied to each experiment are detailed in the corresponding figure legends. No animals or data points were excluded from the analysis.

RESULTS

Glandular relocation of clusterin/TFF2 and altered CPK/GAS/MTL levels indicate successful establishment of SPEM and SSDS models

Under physiological conditions, clusterin is mainly localized primarily within oxyntic glands, particularly in parietal cells, whereas TFF2 expression is weak and restricted to mucous neck and antral glands. Following tamoxifen induction of SPEM, TFF2 expression markedly increased in the basal gland regions, indicative of metaplastic transformation, and clusterin was concurrently upregulated in these areas, suggesting a potential role in metaplasia-associated protective or remodeling processes[24,25]. To verify successful establishment of SPEM and SSDS models, both IF and ELISA were conducted. No significant differences in baseline characteristics were observed among groups, and no adverse events occurred during the study. IF double staining for clusterin and TFF2 confirmed SPEM development. Colocalization of clusterin (green) and TFF2 (red) in G5 mice was predominantly detected in the glandular neck and basal regions (Figure 2A). By contrast, G1 controls exhibited TFF2 expression mainly in the surface mucous layer and gastric pits, with minimal clusterin signal. These findings confirmed successful SPEM induction.

Figure 2 Validation of spasmolytic polypeptide-expressing metaplasia and spleen-stomach deficiency syndrome models through immunofluorescence and ELISA at day 15. A: Gastric mucosal immunofluorescence showing translocation of TFF2 (red) and clusterin (green) from pit regions to glandular bases in spasmolytic polypeptide-expressing metaplasia + spleen-stomach deficiency syndrome co-induction (G5) vs normal architecture (G1). Nuclei counterstained with 4’, 6-diamidino-2-phenylindole (blue). Scale bar = 200 μm; B: Creatine phosphokinase; C: Gastrin; D: Motilin. ^aP < 0.0001 vs G2, ^bP < 0.001 vs G5, ^cP < 0.05 vs G2, ^dP < 0.05 vs G3, ^eP < 0.05 vs G2, ^fP < 0.001 vs G5, ^gP < 0.01 vs G5. Data are presented as mean ± SD (n = 5 per group). Statistical analysis was performed using one-way analysis of variance followed by Tukey’s multiple-comparisons test. CPK: Creatine phosphokinase; MTL: Motilin; GAS: Gastrin.

In SSDS models, characteristic physiological alterations include elevated serum CPK, suggestive of systemic energy metabolism disturbance, reduced GAS levels, and decreased MTL, indicative of dysregulated gastrointestinal motility[26]. Clinically, reduced MTL has also been validated as an independent predictor in spleen-deficiency functional dyspepsia (area under the curve = 0.9615), confirming its diagnostic utility[27]. Compared with G1 controls, serum CPK levels were significantly elevated in G2 model mice (ELISA, P < 0.0001) (Figure 2B), suggesting impaired skeletal muscle energy metabolism in the SSDS model. Treatment with YWXY granules (G3) partially attenuated this elevation, although the reduction did not reach significance (P = 0.2562 vs G2). CPK levels in G4 and G5 were restored to values comparable with G1 (P > 0.05). However, untreated G5 mice still displayed higher expression than G1 (P < 0.001), indicating that YWXY exerts a normalizing effect on CPK expression.

Regarding GAS, G2 mice exhibited significantly lower concentrations compared with G1 (P < 0.05) (Figure 2C), whereas no significant difference was observed between G5 and G1. Although YWXY treatment increased GAS levels, this trend did not achieve significance, suggesting that tamoxifen administration was associated with partial normalization of GAS levels.

In terms of MTL, the SSDS group (G2) showed significantly decreased serum concentrations compared with G1 (P < 0.05) (Figure 2D). YWXY treatment improved gastric motility, mitochondrial function, and glandular secretory activity. Thus, the combined profile of elevated CPK, reduced GAS, and decreased MTL provides robust evidence confirming the successful establishment of the SSDS model.

YWXY alleviates epithelial injury and restores gastric histology at early and late stages of SSDS and SPEM models

On day 15 of the experiment, gastric mucosa and blood samples were collected to confirm successful model establishment. Hematoxylin and eosin staining demonstrated that the SSDS group (G2) exhibited marked inflammatory cell infiltration in the lamina propria and epithelial erosion, while the combined SSDS + SPEM group (G5) displayed more severe pathological alterations, including irregular deep glandular structures with basally located ovoid nuclei and a foamy cytoplasmic appearance (Figure 3A). Consistent morphological changes at day 30 are shown in Figure 3B. In contrast, treatment with YWXY (G3 and G4) restored glandular histological architecture toward normal, with an increased number of chief cells (Figure 3C and D). Quantitative analysis confirmed that, at day 15, G1 had significantly higher chief cell counts than G2 (P = 0.0070) and G5 (P = 0.0052), while G4 treatment markedly rescued chief cell numbers compared with G2 (P = 0.0043) and G5 (P = 0.0032). Similarly, histological gastritis scores at day 15 were elevated in G2 and G5 relative to controls (P < 0.0001) but were significantly reduced in YWXY-treated groups (G3 vs G2, P = 0.0044; G4 vs G5, P < 0.0001) (Figure 3D).

Figure 3 Histopathological restoration of gastric mucosa by Yiwei Xiaoyu granules treatment at days 15 and 30. A: Representative hematoxylin and eosin (H&E) staining images at day 15 showing restoration of glandular architecture in spleen-stomach deficiency syndrome, spasmolytic polypeptide-expressing metaplasia, and combined models after Yiwei Xiaoyu granules (YWXY) treatment; B: Representative H&E staining images at day 30 demonstrating attenuation of glandular atrophy, inflammatory infiltration, and epithelial damage in YWXY-treated groups. Scale bar = 400 μm; C: Quantification of chief cell numbers at days 15 and 30; D: Quantification of gastritis histology index at days 15 and 30. Data are presented as mean ± SD (n = 5 per group). Statistical analysis was performed using two-way analysis of variance followed by Tukey’s multiple-comparisons test. ^aP < 0.05 vs G2, ^bP < 0.05 vs G5, ^cP < 0.05 vs G5, ^dP < 0.05 vs G5, ^eP < 0.01 vs G4, ^fP < 0.0001 vs G5, ^gP < 0.01 vs day 15 in G4, ^hP < 0.001 vs G2, ⁱP < 0.01 vs G3, ^jP < 0.0001 vs G5, ^kP < 0.0001 vs G5, ^lP < 0.05 vs G2, ^mP < 0.01 vs G4, ⁿP < 0.0001 vs G5, ^oP < 0.0001 vs G5.

By day 30, when prior studies have reported that tamoxifen-induced SPEM lesions spontaneously regress within 2 weeks, our data indicated that SSDS pathology persisted. In the SSDS group (G2), the gastric mucosa still showed epithelial damage and inflammatory infiltration, while the combined induction group (G5) exhibited aggravated mucosal injury with prominent glandular atrophy (Figure 3B). YWXY treatment in G3 and G4 markedly reduced inflammatory pathology and restored mucosal morphology toward normal. Chief cell analysis at day 30 revealed significant increases in G4 compared with G1 (P = 0.0153), G5 (P < 0.0001) and G2 (P = 0.0065). Likewise, gastritis scores at day 30 were markedly elevated in untreated groups (e.g., G1 vs G5, P < 0.0001; G2 vs G5, P < 0.0001), but YWXY intervention significantly attenuated pathology, as shown by improvements in G4 vs G5 (P < 0.0001) and G3 vs G5 (P < 0.0001) (Figure 3D). Importantly, longitudinal two-way ANOVA revealed a significant difference between day 15 and day 30 in chief cell counts in G4 (P = 0.0049), indicating that YWXY treatment exerted sustained benefits over time.

Collectively, these results demonstrated that YWXY consistently ameliorated epithelial damage, reduces inflammatory infiltration, and restores gastric glandular architecture in both SSDS and SPEM models, with significant improvements in chief cell numbers and gastritis histology index observed at both early (day 15) and late (day 30) stages.

YWXY modulates inflammatory cytokines, noncoding RNAs, and downstream effectors in SPEM and SSDS models

Biphasic modulation of interleukin-1β expression by YWXY intervention: On day 15, interleukin (IL)-1β mRNA levels were markedly reduced in SSDS (G2) and in tamoxifen-SPEM + SSDS mice (G5) compared with healthy controls (G1) (G1 vs G2: -4.858, P < 0.0001; G1 vs G5: -1.295, P = 0.0059). YWXY treatment restored IL-1β expression toward normal, with significant increases relative to untreated counterparts (G3 vs G2: +5.470, P < 0.0001; G4 vs G5: -2.089, P < 0.0001). By day 30, IL-1β expression was elevated in pathological groups (e.g., G1 vs G5: +2.644, P < 0.0001), while YWXY significantly reduced IL-1β levels (G3 vs G2: -1.371, P = 0.0032; G4 vs G5: -1.981, P < 0.0001), suggesting a biphasic effect: Early restoration followed by suppression of persistent cytokine activation (Figure 4A).

Figure 4 Yiwei Xiaoyu granules regulate inflammatory cytokines and Cdr1as expression in spleen-stomach deficiency syndrome and tamoxifen-induced spasmolytic polypeptide-expressing metaplasia models. A: Fold-change of interleukin-1β mRNA levels on days 15 and 30. At day 15: ^aP < 0.0001 vs G2; ^bP < 0.01 vs G5; ^cP < 0.0001 vs G3; ^dP < 0.001 vs G5. At day 30: ^eP < 0.001 vs G2; ^fP < 0.01 vs G3; ^gP < 0.0001, ^hP < 0.0001 vs G5. In the right panel, ⁱP < 0.0001, ^jP < 0.0001, ^kP < 0.0001 vs day 15 within the same group; B: Tumor necrosis factor-α mRNA expression on days 15 and 30. At day 15: ⁱP < 0.0001 vs G4; ^mP < 0.0001 vs G5; ⁿP < 0.0001 vs G5. At day 30: ^oP < 0.05 vs G2, ^pP < 0.001 vs G3, ^qP < 0.0001 vs G5, ^rP < 0.0001 vs G5. In the right panel, ^sP < 0.0001, ^tP < 0.0001, ^uP < 0.0001, ^vP < 0.0001 vs day 15 within the same group; C: Cdr1as expression across experimental groups at the indicated time points. At day 15: ^wP < 0.01 vs G2, ^xP < 0.0001 vs G5, ^yP < 0.0001 vs G5. At day 30: ^zP < 0.001 vs G3, P < 0.01 vs G4, P < 0.0001 vs G5. In the right panel, P < 0.0001, P < 0.0001 and P < 0.01 vs day 15 within the same group. Data are presented as mean ± SD (n = 5 per group). Statistical analysis was performed using two-way analysis of variance followed by Tukey’s multiple-comparisons test. IL: Interleukin; TNF-α: Tumor necrosis factor-α.

Tumor necrosis factor-α, Cdr1as, and miR-7a-5p regulation: Tumor necrosis factor (TNF)-α was persistently elevated in tamoxifen–SPEM groups at both time points. YWXY significantly attenuated this increase (e.g., G4 vs G5: -1.350 on day 15, P < 0.0001; -3.289 on day 30, P < 0.0001) (Figure 4B).

Cdr1as expression was decreased in several pathological groups on day 15, with partial rescue by YWXY in the tamoxifen–SPEM + SSDS background (G4 vs G5: -5.367, P < 0.0001). By day 30, expression declined across all groups without significant YWXY-associated differences (Figure 4C).

miR-7a-5p FISH revealed marked reductions in G2 and G5 compared to controls (both P < 0.001). YWXY restored miR-7a-5p expression in G3 and G4, reaching levels comparable to G1 (G2 vs G3: -3.305, P = 0.0013; G2 vs G4: -3.381, P = 0.0010). Notably, untreated G5 remained significantly lower than controls (G1 vs G5: +3.674, P = 0.0004), whereas YWXY-treated groups were normalized (Figure 5).

Figure 5 Yiwei Xiaoyu granules (YWXY) restore miR-7a-5p expression in gastric mucosa. A: Representative fluorescence in situ hybridization (FISH) images showing miR-7a-5p expression in the gastric mucosa on day 30; B: Quantification of miR-7a-5p-positive area (% area). ^aP < 0.001 vs G2, ^bP < 0.01 vs G3, ^cP < 0.01 vs G5, ^dP < 0.001 vs G5. Scale bar = 100 µm. Data are shown as mean ± SD (n = 5/group).

WFDC2 expression dynamics validated by polymerase chain reaction and FISH: Consistent with noncoding RNA changes, WFDC2 expression was strongly induced in pathological states. FISH results showed robust WFDC2 induction in G5, with significant increases compared to all other groups (e.g., G1 vs G5: -5.431, P = 0.0001). YWXY treatment markedly attenuated this increase, particularly in the tamoxifen–SPEM background (G4 vs G5: -5.403, P = 0.0001).

Polymerase chain reaction analysis further confirmed these findings. On day 15, WFDC2 mRNA was significantly upregulated in all disease groups relative to G1, with the highest expression in G4 and G5 (G1 vs G4: +2.984, P < 0.0001; G1 vs G5: +1.956, P < 0.0001). YWXY partially reduced WFDC2 expression, although levels in G4 remained elevated. By day 30, WFDC2 remained elevated in G5 compared to all other groups (G1 vs G5: +3.505, P < 0.0001), while YWXY normalized expression in G3 and attenuated it in G4 (G4 vs G5: -2.337, P < 0.0001) (Figure 6).

Figure 6 WFDC2 expression in gastric mucosa at days 15 and 30. A: Relative WFDC2 mRNA expression levels at days 15 and 30 in different experimental groups. At day 15: ^aP < 0.01 vs G2, ^bP < 0.0001 vs G3, ^cP < 0.01 vs G5, ^dP < 0.0001 vs G4, ^eP < 0.0001 vs G5. At day 30: ^fP < 0.001 vs G2, ^gP < 0.0001 vs G5, ^hP < 0.0001 vs G5; B: Representative fluorescence in situ hybridization (FISH) images showing WFDC2 expression in gastric mucosa across experimental groups. Scale bar = 100 µm; C: Quantification of WFDC2-positive area (%Area) across experimental groups. ^mP < 0.001, ⁿP < 0.001 vs G5. Data are presented as mean ± SD (n = 5 per group). Statistical analysis was performed using two-way analysis of variance followed by Tukey’s multiple-comparisons test.

Taken together, these results demonstrated that YWXY regulated early and late inflammatory signals (IL-1β and TNF-α), restored noncoding RNA homeostasis (Cdr1as and miR-7a-5p), and suppressed pathological WFDC2 induction, thereby ameliorating gastric mucosal injury in SPEM and SSDS models.

DISCUSSION

The inflammatory microenvironment is recognized as a critical driver in the progression of chronic atrophic gastritis, and the concept of inflammation-to-cancer transformation has become a central research focus in TCM strategies for gastric carcinogenesis[28]. In this context, inflammatory stress is increasingly viewed not merely as a pathological trigger but as a dynamic disruption of immune–metabolic homeostasis, which underlies the differentiation of cold and hot syndromes in gastric disorders[29]. Consequently, a major challenge in TCM research on PLGC is integrating modern disease models with syndrome-based pathophysiological frameworks.

In recent years, considerable attention has been directed toward gastric homeostasis and repair using SPEM models[30-34]. Among them, the tamoxifen-induced SPEM model[21], with its rapid onset and well-established methodology, has emerged as a robust experimental platform and an ideal carrier for investigating disease-syndrome integration in TCM. In the present study, we developed a composite murine model combining tamoxifen-induced SPEM with SSDS, thereby more closely simulating the clinical context in which PLGC patients often present with concomitant SSDS. Histological and biochemical assessments confirmed the reliability of this model (Figure 2). Specifically, double IF staining for clusterin and TFF2 revealed their aberrant localization from the normal gastric niche to the glandular neck and basal regions, consistent with SPEM transformation. In parallel, altered serum levels of CPK, GAS, and MTL confirmed the SSDS phenotype, aligning with previous findings of metabolic and gastrointestinal dysregulation in SSDS[26,27]. Collectively, this integrated model provides a biologically and clinically relevant foundation for elucidating PLGC pathogenesis and evaluating the therapeutic effects of YWXY.

Using this composite model, we demonstrated that YWXY conferred significant protective effects on the gastric mucosa through histopathological and molecular mechanisms. IL-1β and TNF-α are key proinflammatory cytokines implicated in parietal cell loss, SPEM initiation, and progression toward dysplasia when persistently activated[35]. By evaluating their expression at distinct time points, we identified biphasic regulation of IL-1β in response to YWXY intervention. Specifically, IL-1β expression was restored from pathological suppression during the early injury phase (day 15), while excessive activation was attenuated at the later stage (day 30) (Figure 4A). This temporal modulation suggests that YWXY promotes immune homeostasis rather than indiscriminate cytokine suppression, a feature that distinguishes multi-component TCM formulas from single-target anti-inflammatory agents such as NSAIDs. Consistently, YWXY also reduced sustained TNF-α overexpression in tamoxifen-induced SPEM (Figure 4B). Such bidirectional regulation of inflammatory responses aligns with the immunomodulatory characteristics reported for other classical TCM prescriptions[36].

Our integrated analysis extended beyond inflammatory mediators to encompass regulatory noncoding RNAs and secreted proteins. Previous studies have shown that the tumor-suppressive activity of miR-7 can be antagonized by its circRNA sponge, CDR1as, and dysregulation of this axis contributes to malignant progression[37-40]. It should be emphasized that the present study was not designed to construct a comprehensive competing endogenous RNA (ceRNA) network encompassing all potential miRNAs, circRNAs, lncRNAs, and downstream targets. Instead, we adopted a hypothesis-driven strategy, focusing on the miR-7a-5p/Cdr1as axis as a biologically representative regulatory module at the intersection of inflammation, epithelial differentiation, and metaplastic remodeling (Figure 4C and Figure 5).

Notably, both miR-7a-5p and Cdr1as were downregulated in the SSDS and combined SPEM + SSDS models. This result may appear to contradict the classical circRNA “sponge” paradigm, which predicts inverse expression relationships. However, accumulating evidence indicates that circRNA-miRNA interactions are highly context dependent. Under conditions of acute injury, inflammation, or epithelial stress, global transcriptional repression, altered RNA stability, and shifts in cellular composition may lead to parallel downregulation of circRNAs and their cognate miRNAs, rather than reciprocal changes[41]. In this context, the observed co-downregulation likely reflects injury-associated collapse of noncoding RNA regulatory homeostasis, while the partial restoration of miR-7a-5p by YWXY suggests that the formula acts through multi-layered regulatory mechanisms beyond circRNA sponging alone. These findings highlight the limitations of applying static ceRNA models in dynamic injury settings and underscore the need for contextual interpretation of noncoding RNA networks.

WFDC2, also known as human epididymis protein 4, is a small secretory protein increasingly recognized as a mediator of epithelial injury responses. While WFDC2 has been reported as a biomarker in fibrotic disorders and multiple malignancies, including GC[31,40,42], recent evidence indicates that WFDC2 is actively involved in SPEM development and parietal cell loss-associated gastric remodeling, functioning downstream of inflammatory cues such as IL-33 signaling[31]. From this perspective, WFDC2 represents a biologically plausible effector linking inflammation to metaplastic transformation, rather than a mere cancer-associated marker. In our study, marked induction of WFDC2 at both the transcriptional and spatial levels in SPEM and SSDS models, and its robust suppression by YWXY (Figure 5), suggest that YWXY may interfere with early pro-metaplastic signaling cascades. Clinically, WFDC2 may therefore serve as a candidate biomarker reflecting active injury-driven epithelial remodeling in SSDS-associated PLGC.

Taken together, our findings indicate that gastric mucosal responses to the dual insult of SPEM and SSDS involve temporally coordinated regulation of inflammatory cytokines, noncoding RNA axes, and secreted effector proteins (Figure 7). YWXY intervention not only improved histological injury but also normalized or suppressed key molecular alterations, highlighting its therapeutic potential for patients with SSDS-associated PLGC. This integrated perspective provides new insight into the mechanistic underpinnings of YWXY action, bridging traditional syndrome-based models with molecular pathology.

Figure 7 Proposed schematic model of Yiwei Xiaoyu granules-mediated modulation of the miR-7a-5p-Cdr1as-WFDC2-cytokine axis in gastric mucosa. Arrow (↑): Activation or increase, blunt line (┤): Inhibition or decrease. YWXY: Yiwei Xiaoyu granules; Tam: Tamoxifen; SPEM: Spasmolytic polypeptide-expressing metaplasia. This figure was created by BioRender.com. https://BioRender.com/yezh7od.

From a translational standpoint, YWXY is clinically prescribed for patients with SSDS, particularly those at elevated risk of PLGC[18]. By modeling combined SSDS and tamoxifen-induced SPEM, our study more closely reflected the patient population where YWXY is clinically applied, bridging the gap between experimental systems and real-world therapeutic use. The observed restoration of motility-related peptides (GAS and MTL), normalization of skeletal muscle energy metabolism (CPK), and modulation of inflammatory and regulatory RNAs suggests that YWXY exerts both systemic and local protective effects on gastric homeostasis.

Several limitations of this study should be acknowledged. First, although the composite model enhances clinical relevance, it does not fully recapitulate the chronicity and heterogeneity of human PLGC progression. Second, our conclusions are primarily based on transcriptomic and protein expression analyses, which demonstrate strong associations but do not establish definitive causality. Functional rescue experiments (e.g., miR-7 modulation) and comprehensive ceRNA network reconstruction were beyond the scope of the present study and should be prioritized in future investigations. Finally, YWXY is a multi-component formulation, and further fractionation and target validation studies are required to identify its key bioactive constituents. Despite these limitations, our study provides a coherent mechanistic framework and a strong rationale for future in-depth investigations. In addition, all experimental procedures were designed to adhere to the principles of replacement, reduction, and refinement (3Rs) to minimize animal use and suffering.

CONCLUSION

In conclusion, YWXY alleviates gastric mucosal injury in a composite SSDS + tamoxifen-induced SPEM model by coordinating inflammatory and noncoding RNA responses and suppressing WFDC2 induction. These results link syndrome-based TCM intervention to molecular mechanisms of injury-driven metaplastic remodeling and provide a rationale for further translational investigation of YWXY in SSDS-associated PLGC.