Pristimerin ameliorates spasmolytic polypeptide-expressing metaplasia by modulating Cdkn1c (p57)-mediated glycolytic reprogramming

Abstract

BACKGROUND

Spasmolytic polypeptide-expressing metaplasia (SPEM) is a gastric precancerous lesion (GPL) with high malignant potential. The ethyl acetate extract of Celastrus orbiculatus Thunb. effectively ameliorates GPL and gastric cancer progression. Meanwhile, the primary active constituent of this plant, pristimerin, also demonstrates notable antitumor activity.

AIM

To investigate the therapeutic effects of pristimerin on SPEM and its underlying mechanisms.

METHODS

Pristimerin was administered to high-dose tamoxifen-induced SPEM mice to assess its effects on pathological progression, glycolytic reprogramming, and Cdkn1c (p57) expression. Human gastric epithelial (GES-1) cells were treated with tamoxifen and then with pristimerin or 2-deoxy-D-glucose to demonstrate that pristimerin ameliorates SPEM by regulating glycolytic reprogramming. Furthermore, gastric organoids were treated with N-methyl-N’-nitro-N-nitrosoguanidine/Helicobacter pylori, followed by Cdkn1c overexpression or knockdown and then pristimerin, to confirm p57 as the key target through which pristimerin regulates glycolytic reprogramming and reverses SPEM.

RESULTS

Pristimerin effectively ameliorated gastric mucosal damage and oxyntic atrophy induced by high-dose tamoxifen, suppressed the aberrant upregulation of key glycolytic regulators, SPEM-specific markers, and stem cell markers, and upregulated p57 expression. In tamoxifen-induced GES-1 cells, pristimerin exhibited comparable therapeutic effects. Crucially, glycolysis inhibition in GES-1 cells effectively ameliorated tamoxifen-induced SPEM-associated phenotypes. In gastric organoids, Cdkn1c overexpression suppressed glycolytic reprogramming and SPEM phenotype activation, whereas Cdkn1c knockdown attenuated pristimerin-mediated inhibition of glycolysis and amelioration of SPEM.

CONCLUSION

Pristimerin effectively ameliorates gastric mucosal pathological damage and oxyntic atrophy in high-dose tamoxifen-induced SPEM mice, and improves SPEM progression by modulating Cdkn1c (p57)-mediated glycolytic reprogramming.

Key Words: Pristimerin; Spasmolytic polypeptide-expressing metaplasia; Glycolytic reprogramming; Cdkn1c; Gastric organoids; Gastric precancerous lesion

Core Tip: This study systematically investigated the therapeutic effects of the natural compound pristimerin on spasmolytic polypeptide-expressing metaplasia (SPEM) using a high-dose tamoxifen-induced mouse model, tamoxifen-treated human gastric epithelial cells, and N-methyl-N’-nitro-N-nitrosoguanidine/Helicobacter pylori-induced gastric organoids. It elucidated the underlying mechanism by which pristimerin ameliorates SPEM through regulating Cdkn1c (p57) to suppress glycolytic reprogramming. These findings not only provide novel mechanistic insights into pristimerin treatment for SPEM but also offer a crucial theoretical basis for targeting p57 in SPEM therapy.

INTRODUCTION

Based on 2022 global epidemiology statistics, gastric cancer (GC) was the fifth-leading malignancy worldwide in terms of both incidence and mortality[1]. Due to its insidious early symptoms and lack of effective treatments, GC constitutes a substantial public health burden. In recent years, the research focus has gradually shifted from “GC treatment” to “interception of precancerous lesions”. Among these, spasmolytic polypeptide-expressing metaplasia (SPEM) has garnered widespread attention due to its malignant transformation potential. As a secondary lesion of chronic atrophic gastritis (CAG), SPEM exhibits a higher propensity to progress to malignancy compared to CAG[2]. Indeed, studies demonstrate that SPEM possesses genomic features more closely resembling GC than other metaplastic types[3]. Furthermore, SPEM cells derived from chief cells can further develop into multipotent cancer stem cells[4]. Therefore, identifying effective therapeutic targets and developing interventions against SPEM progression are critical for GC prevention.

As a key candidate tumor suppressor within the cyclin-dependent kinase-interacting protein/kinase inhibitory protein (CIP/KIP) family, p57 (Cdkn1c) inhibits tumorigenesis by regulating critical cancer cell processes such as proliferation, migration, and aerobic glycolysis[5]. Downregulation of this factor in advanced-stage tumors leads to loss of its function, while reactivation of p57 expression restores its tumor-suppressive function[6]. Similarly, evidence indicates that p57 is significantly downregulated in injured gastric chief cells. Insufficient p57 expression directly activates chief cell plasticity, driving the acquisition of malignant phenotypes and resulting in aberrantly enhanced proliferative activity. Conversely, restoring p57 expression maintains the long-term “reserve stem cell (RSC)” state in chief cells[7,8]. It is noteworthy that chief cells, as a known source of SPEM cells, undergo transformation into trefoil factor 2 (TFF2)- and mucin 6 (MUC6)-expressing SPEM cells a hallmark feature of this pathological process[9]. Therefore, p57 may represent a promising therapeutic target for maintaining chief cell homeostasis and thereby ameliorating SPEM.

Metabolic reprogramming is a cancer hallmark in which tumor cells preferentially utilize glycolysis for energy under aerobic conditions[10]. Modulation of glycolytic reprogramming has been demonstrated to ameliorate pathological processes in multiple diseases, especially tumors[11,12]. However, recent studies reveal that in gastric precancerous lesion (GPL), gastric mucosal cells remodel glucose metabolism patterns enhanced glycolysis to sustain aberrant proliferation and malignant progression, whereas regulating glycolytic pathways effectively delays precancerous lesion progression[13-15]. Given this evidence, we hypothesize that glycolytic reprogramming similarly drives the progression of SPEM, a specific pathological process, and that intervention targeting this metabolic pathway may represent a novel strategy for ameliorating SPEM.

Celastrus orbiculatus Thunb., a woody vine from the Celastraceae family, has been demonstrated in modern pharmacological studies to significantly inhibit the proliferation, invasion, and metastasis of GC cells through its ethyl acetate extract (COE)[16,17]. Furthermore, COE effectively alleviates inflammatory and metaplastic conditions in GPL by modulating cellular metabolic reprogramming[18], suppressing cancer stem cell markers[19], and restoring the structural and functional damage of gastric chief cells[20]. Pristimerin (Supplementary Figure 1), a triterpenoid compound derived from this plant, serves as its primary and naturally abundant active constituent. It possesses broad-spectrum antitumor properties, inhibiting proliferation and metastasis by triggering multiple cytotoxic mechanisms, including cell cycle arrest, necrosis, apoptosis, and autophagy[21]. Notably, a central anti-tumor mechanism involves direct interaction with and regulation of CIP/KIP proteins, as demonstrated by its modulatory effects on members such as p21 and p27[22,23]. However, whether pristimerin ameliorates SPEM by modulating p57 expression and glycolytic reprogramming remains to be elucidated.

This study explored the therapeutic efficacy of pristimerin in SPEM and its underlying mechanisms. The results demonstrated that pristimerin significantly alleviated gastric mucosal pathological damage and oxyntic atrophy in high-dose tamoxifen (HDT)-induced SPEM mice. More importantly, this study provides the first mechanistic evidence that pristimerin ameliorates SPEM by modulating Cdkn1c (p57)-mediated glycolytic reprogramming. These findings not only provide novel mechanistic insights into pristimerin treatment for SPEM but also establish p57 as a promising therapeutic target for this condition.

MATERIALS AND METHODS

Key reagents and resources

Pristimerin (B20098; purity: 99.3%) was obtained from Shanghai Yuanye Biotechnology Co., Ltd. (China). Hematoxylin (G1004; Servicebio, China) and eosin (71014544; Sinopharm, China) were used for histological staining, supplemented with an alcian blue/periodic acid schiff (AB/PAS) kit (60534ES50; Yeasen, China). Antibodies targeting MUC6 (ABIN6263435; Antibodies Online, Germany) and pyruvate kinase M2 (PKM2) (ABIN744773; Antibodies Online, Germany) were employed. Additional antibodies and kits from Abcam (United Kingdom) included: Anti-TFF2 (ab49526), anti-Ki67 (ab15580), anti-hypoxia-inducible factor-1α (HIF-1α) (ab228649), anti-lactate dehydrogenase A (LDHA) (ab52488), glycolysis/mitochondrial stress test kits (ab222946/ab232857), and anti-β-actin (ab8226). Pyruvate and lactate assay kits (A019-2-1/A081-1-1) were purchased from the Nanjing Jiancheng Bioengineering Institute. Other critical reagents comprised anti-hydrogen-potassium-adenosine-triphosphatase (H⁺/K⁺ ATPase) (A274; Merck KGaA, Germany), anti-gastric intrinsic factor (GIF) (PA5-87282; Invitrogen, United States), anti-Mist (sc-166181; Santa Cruz Biotechnology, United States), anti-whey-acidic-protein four-disulfide core domain protein 2 (WFDC2) (LS-B7194; LSBio, United States), anti-leucine-rich repeat-containing G-protein coupled receptor 5 (Lgr5) (A10545; ABclonal, China), anti-tumor necrosis factor receptor superfamily member 19 (Troy) (sc-515473; Santa Cruz, United States), anti-p57 (sc-71823; Santa Cruz, United States), 4’,6-diamidino-2-phenylindole (DAPI) nuclear stain (EZ3412B05; Beyotime, China), the carcinogen N-methyl-N’-nitro-N-nitrosoguanidine (MNNG) (ZG4T1-FP; Tokyo Chemical Industry, Japan), Helicobacter pylori (H. pylori) (B84182; Mingzhou Biotech, China), human gastric epithelial cells (GES-1) (GuYan Biotech, China), tamoxifen (TAM) (54965-24-1), 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) (M405849; Aladdin, China), 5-Ethynyl-2’-deoxyuridine (EdU) proliferation kit (A598377; Aladdin, China), Lipofectamine^TM 3000 (L3000001; Thermo Fisher, United States), RNAiMAX (13778075; Thermo Fisher, United States), and Matrigel basement membrane matrix (356231; Corning, United States).

In vivo animal modeling and therapeutic intervention

Following a 7-day acclimation period, thirty female C57BL/6J mice (6-week-old, specific pathogen free, body weight range 16.0-18.0 g) were divided into five weight-matched groups (n = 6 per group): Control, model, and three pristimerin treatment groups. Mice were obtained from Jiangsu GemPharmatech Co., Ltd. and maintained at 24 °C on a 12-hour light/dark cycle, and provided free access to food and water. This study was conducted in accordance with the approved animal ethics protocol. The sample size (n = 6 per group) was determined to balance statistical requirements with the “Reduction” principle of animal ethics, and this protocol was approved by the Institutional Animal Care and Use Committee of Anhui Medical University (No. LLSC20220294).

Mice in the pristimerin-treated (Pri) groups received daily intraperitoneal injections of 0.05 mg/kg/day [low-dose pristimerin (Pri-L)], 0.1 mg/kg/day [medium-dose pristimerin (Pri-M)], or 0.2 mg/kg/day [high-dose pristimerin (Pri-H)] pristimerin for 21 consecutive days, while those in the model and control groups were administered an equivalent volume of saline. Following an 8-day intervention period, the pristimerin and model groups underwent 3-days of HDT administration (250 mg/kg, intraperitoneal injection) dissolved in a vehicle of 10% ethanol and 90% sunflower oil. Control animals received only the vehicle solution. Body weight measurements were systematically performed at 7-day intervals throughout the study period. Upon completion of the treatment regimen, mice were humanely euthanized under terminal anesthesia to harvest gastric tissues.

All procedures were carried out in a specific pathogen free facility. No specific measures, such as randomization of procedural order or counterbalancing of cage positions, were taken to control for these potential confounders. This study implemented full blinding. The personnel responsible for group allocation did not participate in subsequent experimental procedures. By concealing the drug administration process, all researchers remained unaware of the group assignments. The allocation list was unblinded only after the final data analysis was completed.

Tissue staining and histopathological analysis

Gastric corpus specimens were longitudinally incised, immersion-fixed in 10% neutral buffered formalin, and then processed via graded ethanol-xylene dehydration, paraffin embedding, and microtomy (3-4 μm). Dewaxed sections underwent histochemical staining with hematoxylin and eosin (HE) or AB/PAS. Blinded evaluation (Nikon Eclipse E100) assessed mucosal injury, glandular atrophy, and metaplasia. ImageJ was used to analyze histopathological alterations in HE-stained sections (nuclei: Blue/purple; cytoplasm: Pink) and to quantify mucin-positive areas (acidic: Blue; neutral: Magenta) in AB/PAS-stained tissues.

In vitro cellular modeling and intervention

GES-1 cells were routinely propagated in RPMI-1640 supplemented with heat-inactivated fetal bovine serum (FBS) and antibiotic-antimycotic solution (penicillin 100 U/mL, streptomycin 100 μg/mL). To induce SPEM-like phenotypes, cells underwent serum starvation in medium containing 0.1% FBS for 24 hours, followed by incubation with 10 μmol/L TAM in complete medium for an additional 24 hours. For the therapeutic intervention, SPEM cells were exposed to different concentrations of pristimerin or the glycolysis inhibitor 2-deoxy-D-glucose (2-DG) (5 mmol/L) for 24 hours, and controls received equivalent volumes of dimethyl sulfoxide (DMSO).

In vitro three-dimensional organoid modeling and intervention

Murine gastric organoids were constructed as described by Shibata et al[24]. Following microdissection into 1 mm² fragments, gastric corpus tissues underwent sequential processing: Three ice-cold phosphate-buffered saline (PBS) washes, followed by 3-hour incubation in 10 mmol/L ethylenediaminetetraacetic acid-supplemented PBS under chilled conditions (4 °C). After centrifugation, pellets were mechanically dissociated under microscopy to release gastric glands. The glands were then filtered, centrifuged, and embedded in Matrigel. Organoids were cultured in advanced Dulbecco’s modified eagle’s medium (DMEM)/F12 medium supplemented with B27, N2, N-acetylcysteine, gastrin, epidermal growth factor, R-spondin1, noggin, fibroblast growth factor 10, Wnt3A, and penicillin/streptomycin (37 °C, 5% carbon dioxide, medium refreshed every 48 hours). Organoids > 400 μm were passaged, and third-passage organoids were used. For SPEM induction, organoids were co-cultured with H. pylori (multiplicity of infection: 2 × 10⁵) in 0.02 μg/mL MNNG-containing medium for 24 hours. Following Cdkn1c overexpression/knockdown, gastric organoids were treated with 0.2 μmol/L pristimerin for 24 hours.

Plasmid transfection

The coding sequence of the Cdkn1c gene was cloned into the pcDNA3.1 vector via double digestion with HindIII/EcoRI to construct the overexpression plasmid. SPEM-derived organoids were dissociated into single-cell suspensions by enzymatic digestion and filtration. Lipofectamine^TM 3000 mediated transfection of Cdkn1c-overexpressing plasmid or empty vector constructs into cells, followed by selection with 500 μg/mL G418 to establish stable cell lines. Transfected cells were re-embedded in Matrigel for three-dimensional (3D) expansion under standard organoid culture conditions.

Small interfering RNA transfection

Negative control small interfering RNA (siRNA) and Cdkn1c-targeting siRNA (sequences listed in Supplementary Table 1) were synthesized by GenePharma (Shanghai, China). SPEM organoids were dissociated into single-cell suspensions, and siRNA-lipid complexes were prepared by mixing siRNA with Lipofectamine^TM RNAiMAX at a 1:1 (v/v) ratio. Following 6-hour incubation with the complexes, cells were replenished with complete medium. After 48 hours, transfected cells were re-suspended in Matrigel for 3D expansion.

Immunohistochemistry

Following dewaxing, antigen retrieval was performed in citrate buffer. Subsequently, quenching with 3% hydrogen peroxide and blocking with 10% goat serum were performed. Primary antibodies were applied overnight at 4 °C, with subsequent incubation of horseradish peroxidase (HRP)-conjugated secondary antibodies. Chromogenic development employed diaminobenzidine substrate followed by hematoxylin counterstaining. Whole-slide imaging was conducted on an Olympus BX53 platform. ImageJ quantified positively stained areas in immunohistochemically stained sections.

Immunofluorescence

Following antigen retrieval, tissue sections underwent blocking with 5% FBS prior to overnight exposure to primary antibodies at 4 °C. After three PBS washes, secondary antibody incubation with fluorophore conjugates was performed for 1 hour. Nuclear counterstaining was achieved using DAPI. Fluorescence was visualized using an Olympus BX53 microscope. ImageJ was utilized to quantify immunofluorescence signals.

Western blot

Proteins extracted with radio immunoprecipitation assay buffer (protease/phosphatase inhibitors) were quantified by the bicinchoninic acid assay. Protein electrophoresis utilized sodium dodecyl sulfate-polyacrylamide gel electrophoresis, followed by polyvinylidene difluoride membrane electroblotting. Post-blocking (5% bovine serum albumin), membranes underwent primary antibody incubation, rigorous tris buffered saline tween washes, and HRP-secondary antibody probing. Immunoreactivity was visualized via electrochemiluminescence reagent exposure and signal capture. Target protein expression was determined by ImageJ analysis, normalized to β-actin, and presented as fold changes vs controls.

Real-time quantitative polymerase chain reaction

Complementary DNA synthesis from total RNA employed Moloney murine leukemia virus reverse transcriptase. Polymerase chain reaction amplification was performed with primers listed in Supplementary Table 2 (5-6 replicates/group). Products were electrophoresed on 1% agarose gels and analyzed by GelStain imaging. Relative gene expression was calculated using the 2^-ΔΔCT method, normalized to a housekeeping gene, and expressed as fold change relative to the control group.

MTT assay

After treatment with pristimerin or 2-DG, the culture medium in the plates was aspirated, followed by the addition of 20 μL MTT reagent to each well and incubation for 4 hours. Subsequently, 150 μL DMSO was added, and the plates were gently shaken in the dark for 15 minutes. Cell viability was assessed by 490-nm optical density measurements using a microplate reader.

EdU staining

Cells (5 × 10³/well, 24-well plate) were treated with 10 μmol/L EdU for 4 hours, followed by sequential fixation (4% paraformaldehyde solution, 15 minutes) and permeabilization (0.5% Triton X-100 solution, 20 minutes) at room temperature. Subsequently, EdU was labeled using the click reaction mixture under light-protected conditions at room temperature for 30 minutes, and nuclei were counterstained with DAPI (1 μg/mL) for 10 minutes. Final images were acquired using a Nikon Eclipse Ti2 inverted fluorescence microscope.

Ultrastructural observation and analysis

Following dual fixation (2.5% glutaraldehyde and 2% osmium tetroxide), gastric specimens underwent ethanol gradient dehydration and epoxy resin embedding. Uranyl acetate and lead citrate were applied to counterstain ultrathin sections. Cellular ultrastructure was imaged using a transmission electron microscope (Hitachi HT7800).

Pyruvate and lactate content assays

Lactate and pyruvate concentrations in the cell culture medium were determined with their respective assay kits, in strict adherence to the manufacturer’s protocols. Briefly, the conditioned medium was collected and centrifuged to remove cellular debris, and the clarified supernatant was taken for detection. For lactate quantification, the enzyme working solution and chromogenic agent were combined with 20 μL of the supernatant. After incubation at 37 °C for 10 minutes, the stop solution was added, and the absorbance was measured at a wavelength of 530 nm. For pyruvate quantification, 0.1 mL of supernatant was subjected to a 10-minute, 37 °C incubation with the chromogenic agent. Subsequently, following the addition of alkaline solution, the reaction mixture was kept at room temperature for 5 minutes before the final absorbance measurement at 505 nm. Standard curves were generated using serial concentrations of standards, and the metabolite concentrations were calculated accordingly.

Oxygen consumption and extracellular acidification rate assays

Mitochondrial respiration was assessed using the Seahorse XF Cell Mito Stress Test Kit. In brief, cells were seeded in XF24 plates at a density of 7 × 10⁴ cells/well and cultured overnight. On the day of assay, cells were equilibrated in XF DMEM assay medium (supplemented with 1 mmol/L pyruvate, 2 mmol/L glutamine, and 10 mmol/L glucose, potential of hydrogen = 7.4) for 60 minutes in a non-carbon dioxide incubator. During the assay, oligomycin (15 μmol), trifluoromethoxy carbonylcyanide phenylhydrazone (2 μmol/L), and rotenone/antimycin A (5 μmol/L) were sequentially injected. Data were normalized to cell number and analyzed using Wave software (Agilent, United States). Glycolytic function was evaluated using the Seahorse XF glycolysis stress test kit. Cells were plated as described above and incubated in XF DMEM medium containing 2 mmol/L glutamine. After equilibration, glucose (100 mmol/L), oligomycin (10 μmol/L), and 2-DG (500 mmol/L) were injected sequentially. All measurements were performed on a Seahorse XFe24 Analyzer (Agilent, United States), and data were processed using the associated glycolysis report generator.

Statistical analysis

Statistical analysis was conducted on data from a minimum of five independent biological replicates using GraphPad Prism 9.5, with all results expressed as mean ± SD and significance defined as P ≤ 0.05. Following verification of parametric test assumptions, an unpaired Student’s t-test (for two groups) or one-way analysis of variance (for multiple groups) was used, with Tukey’s or Dunnett’s post-hoc test applied for all pairwise or model group comparisons, respectively.

RESULTS

Pristimerin alleviates HDT-induced gastric mucosal pathological damage in SPEM mice

A murine SPEM model was induced by HDT to assess the therapeutic potential of pristimerin[25]. The treatment groups received pristimerin at varying doses via intraperitoneal injection (Figure 1A). During the intervention period, HDT induction significantly attenuated body weight gain in mice, whereas pristimerin treatment effectively maintained growth trajectory (Figure 1B). Macroscopic evaluation revealed that gastric mucosa in the model group exhibited pallor/duskiness, increased firmness, rugae flattening or disappearance, accompanied by focal nodular/ulcer-like lesions. Conversely, pristimerin substantially ameliorated these pathological phenotypes (Figure 1C). Histopathological analysis of the gastric mucosa in model mice revealed severely disrupted glandular architecture, a disorganized cellular arrangement with morphological abnormalities, a substantial depletion of parietal and chief cells, alongside evident inflammatory infiltration. Pristimerin improved these pathological features (Figure 1D). AB/PAS staining further confirmed abnormally expanded acid mucin-positive areas in the gastric mucosa of the model group, while pristimerin treatment substantially reversed this pathological secretion (Figure 1E). Collectively, this evidence demonstrates that pristimerin significantly alleviates gastric mucosal pathological damage and modulates mucus secretion in HDT-induced SPEM mice.

Figure 1 Pristimerin alleviates high-dose tamoxifen-induced gastric mucosal pathological damage in spasmolytic polypeptide-expressing metaplasia mice. A: Induction of spasmolytic polypeptide-expressing metaplasia mouse model and pristimerin therapeutic regimen; B: Body weight changes in mice; C: Gross morphological observation of gastric tissues; D: Gastric tissue histology by hematoxylin-eosin staining. Scale bar, 100 μm; E: Gastric mucins detected by alcian blue/periodic acid Schiff staining. Scale bar, 100 μm. ^dP < 0.0001. Values represent the mean ± SD (n = 6). HDT: High-dose tamoxifen; Pri-L: Low-dose pristimerin; Pri-M: Medium-dose pristimerin; Pri-H: High-dose pristimerin.

Pristimerin alleviates SPEM-related oxyntic atrophy in mice

Oxyntic atrophy, a landmark pathological alteration in SPEM, is characterized by progressive depletion and functional impairment of chief cells (enzyme-producing) and parietal cells (acid-secreting)[26]. To elucidate the impact of pristimerin on oxyntic atrophy in SPEM mice, we focused on the expression changes of parietal cell marker H⁺/K⁺ ATPase, chief cell marker GIF, and its maturation marker mist. The results demonstrated that the gastric mucosa of HDT-induced SPEM mice displayed a notable downregulation of H⁺/K⁺ ATPase, GIF, and mist, while pristimerin treatment effectively reversed these reductions (Figure 2A-C), consistent with histopathological findings. Ki67 detection further revealed abnormally enhanced proliferative activity in the gastric mucosa of the model group, which was suppressed by pristimerin treatment (Figure 2D). These results demonstrate that pristimerin effectively preserves chief and parietal cells, attenuates HDT-induced oxyntic atrophy in the gastric mucosa, and suppresses aberrant cellular proliferation.

Figure 2 Pristimerin alleviates spasmolytic polypeptide-expressing metaplasia-related oxyntic atrophy in mice. A-C: Immunofluorescence and reverse transcription quantitative polymerase chain reaction detection of hydrogen-potassium-adenosine-triphosphatase (parietal cells), gastric intrinsic factor and mist (chief cells) expression in gastric mucosa of high-dose tamoxifen-induced spasmolytic polypeptide-expressing metaplasia mice treated with pristimerin. Scale bar, 20 μm; D: Inhibitory effect of pristimerin on gastric mucosal cell proliferation in mice. Scale bar, 20 μm. ^aP < 0.05; ^bP < 0.01; ^cP < 0.001; ^dP < 0.0001. Values represent the mean ± SD (n = 6). DAPI: 4’,6-diamidino-2-phenylindole; H⁺/K⁺ ATPase: Hydrogen-potassium-adenosine-triphosphatase; GIF: Gastric intrinsic factor; Pri-L: Low-dose pristimerin; Pri-M: Medium-dose pristimerin; Pri-H: High-dose pristimerin; IOD: Integrated optical density.

Pristimerin suppresses aberrant expression of SPEM-specific and stem cell markers

SPEM is a metaplastic lesion arising during gastric mucosal carcinogenesis, characterized by chief cell trans differentiation into TFF2/MUC6-aberrant metaplastic cells localized at gastric corpus gland bases[9,27]. WFDC2 is another important specific biomarker in SPEM, closely associated with malignant evolution[28]. To further elucidate the therapeutic potential of pristimerin against SPEM, we first analyzed the expression patterns of these characteristic markers in SPEM lesions. The results demonstrated downregulation of GIF concomitant with upregulation of TFF2/MUC6 in the model group vs controls. Pristimerin treatment significantly reversed this trend, with GIF expression gradually recovering and TFF2/MUC6 expression being reduced as pristimerin concentration increased (Figure 3A). WFDC2 exhibited concordant expression trends with TFF2/MUC6 (Figure 3B).

Figure 3 Inhibitory effect of pristimerin on aberrantly expressed spasmolytic polypeptide-expressing metaplasia-specific markers and stem cell markers. A: Immunofluorescence co-localization analysis of chief cell marker gastric intrinsic factor with spasmolytic polypeptide-expressing metaplasia (SPEM)-specific markers trefoil factor 2 (TFF2)/mucin 6 (MUC6) in mouse gastric tissues, and detection of TFF2/MUC6 expression by reverse transcription quantitative polymerase chain reaction in mouse stomach. Scale bar, 20 μm; B: Expression of SPEM-specific marker whey-acidic-protein four-disulfide core domain protein 2 (WFDC2) in mouse gastric tissues. Scale bar, 100 μm; C: 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay of human gastric epithelial (GES-1) cells viability treated with varying concentrations of tamoxifen or pristimerin; D: Western blot analysis of SPEM-specific markers (TFF2/MUC6/WFDC2) in GES-1 cells with β-actin-normalized quantification (vs control); E and F: Inhibitory effect of pristimerin on stem cell markers leucine-rich repeat-containing G-protein coupled receptor 5 and tumor necrosis factor receptor superfamily member 19 in gastric mucosa of SPEM mice (E) and GES-1 cells (F). Scale bar, 20 μm. ^aP < 0.05; ^bP < 0.01; ^cP < 0.001; ^dP < 0.0001. Values represent the mean ± SD (A, B and E: n = 6; C, D and F: n = 5). DAPI: 4’,6-diamidino-2-phenylindole; GIF: Gastric intrinsic factor; Pri-L: Low-dose pristimerin; Pri-M: Medium-dose pristimerin; Pri-H: High-dose pristimerin; IOD: Integrated optical density; MUC6: Mucin 6; TAM: Tamoxifen; TFF2: Trefoil factor 2; WFDC2: Whey-acidic-protein four-disulfide core domain protein 2; MOD: Mean optical density; Lgr5: Leucine-rich repeat-containing G-protein coupled receptor 5; Troy: Tumor necrosis factor receptor superfamily member 19.

To validate this mechanism in vitro, GES-1 cells were treated with varying concentrations of TAM, and the half-maximal inhibitory concentration (IC₅₀) was determined. The results showed IC₅₀ values of 25.90 μmol/L and 12.57 μmol/L after 24 hours and 48 hours of treatment, respectively. Notably, treatment with 10 μmol/L TAM for 24 hours significantly enhanced cell viability (Figure 3C), a phenotype consistent with the aberrant proliferative characteristics of SPEM[26]. The significantly upregulated expression of SPEM characteristic marker genes (TFF2, MUC6, and WFDC2; Supplementary Figure 2A) further confirmed the reliability of the model. In addition, cytotoxicity assessment of pristimerin revealed IC₅₀ values of 0.65 μmol/L and 0.38 μmol/L after 24 hours and 48 hours of treatment in GES-1 cells, respectively. Following 24 hours of intervention, cell viability remained above 60% when the concentration was below 0.4 μmol/L (Figure 3C). Therefore, 0.1 μmol/L (Pri-L), 0.2 μmol/L (Pri-M), and 0.4 μmol/L (Pri-H) were selected as the low-, medium-, and high-intervention concentrations for subsequent experiments. Western blotting revealed significantly elevated expression of TFF2, MUC6, and WFDC2 in TAM-induced GES-1 cells; pristimerin treatment suppressed this effect. In vitro experiments further validated pristimerin’s regulatory capacity on these markers (Figure 3D).

Lgr5 and Troy, recognized as stem cell markers, are broadly expressed in the gastric pylorus and fundus, predominantly localized to the basal and isthmus regions of the gastric glands. Studies have demonstrated that aberrant overexpression of these two markers in gastric body glands strongly correlates with gastric mucosal malignant progression[29]. Therefore, we further analyzed the expression profiles of Lgr5 and Troy. Significant upregulation of both markers was observed in gastric body tissues of SPEM mouse models. Notably, pristimerin treatment significantly suppressed their expression (Figure 3E). Consistent results were obtained in vitro (Figure 3F). Collectively, these findings establish pristimerin’s potent suppression of aberrant SPEM-specific and stem cell markers, demonstrating molecular-level evidence for its therapeutic efficacy in SPEM.

Pristimerin regulates glycolytic reprogramming in gastric mucosal cells

Glycolytic reprogramming is the key metabolic basis promoting the malignant transformation of gastric mucosal cells and disease progression. Analysis of key glycolytic regulators (HIF-1α, LDHA, and PKM2) demonstrated a notable elevation of their expression in the HDT-induced mice. Pristimerin treatment suppressed the expression of these factors (Figure 4A). Ultrastructural observation revealed that gastric mucosal tissue cells in SPEM model mice exhibited dilated endoplasmic reticulum, swollen mitochondria, accompanied by a significant decrease in the number of secretory granules and autophagic vesicles. In contrast, the Pri group demonstrated a denser endoplasmic reticulum structure, well-preserved mitochondrial morphology, and a significant increase in secretory granules and autophagic vesicles (Figure 4B).

Figure 4 Pristimerin regulates glycolytic reprogramming in gastric mucosal cells. A: Inhibitory effects of pristimerin on key glycolytic regulators (hypoxia-inducible factor-1α, lactate dehydrogenase A, pyruvate kinase M2) in the gastric mucosa of mice. Scale bar, 100 μm; B: Representative transmission electron microscopy images revealing ultrastructural features in mouse gastric mucosal cells. Scale bar, 2 μm; C: Analysis of cellular energy metabolism by Seahorse XF technology in human gastric epithelial (GES-1) cells following treatment with pristimerin; D: Quantitative analysis of pyruvate and lactate production in GES-1 cells. ^aP < 0.05. ^bP < 0.01. ^cP < 0.001. ^dP < 0.0001. Values represent the mean ± SD (A: n = 6; C and D: n = 5). HIF-1α: Hypoxia-inducible factor-1α; LDHA: Lactate dehydrogenase A; PKM2: Pyruvate kinase M2; Pri-L: Low-dose pristimerin; Pri-M: Medium-dose pristimerin; Pri-H: High-dose pristimerin; MOD: Mean optical density; FCCP: Trifluoromethoxy carbonylcyanide phenylhydrazone; 2-DG: 2-deoxy-D-glucose; OCR: Oxygen consumption rate; ECAR: Extracellular acidification rate.

In vitro, treatment of TAM-induced GES-1 cells with varying concentrations of pristimerin, followed by assessment using the Seahorse XF Metabolic Analyzer, revealed that TAM induction significantly enhanced the glycolytic rate and glycolytic capacity of GES-1 cells, while markedly reducing both the basal oxygen consumption rate (OCR) and maximal OCR. Pristimerin treatment rescued this metabolic imbalance (Figure 4C). Metabolite analysis further demonstrated that pristimerin effectively suppressed the TAM-induced elevation in pyruvate and lactate levels (Figure 4D). Collectively, these results confirm that gastric mucosal cells in SPEM lesions exhibit a metabolic disorder characterized by aberrantly enhanced glycolysis and suppressed oxidative phosphorylation. Pristimerin restores metabolic homeostasis by reprogramming glycolytic metabolism.

Modulation of glycolytic reprogramming ameliorates TAM-induced SPEM phenotypes in GES-1 cells

To further investigate the impact of glycolytic reprogramming on SPEM and its role in pristimerin-mediated amelioration of SPEM pathology, we treated TAM-induced GES-1 cells with medium-concentration pristimerin or 2-DG (Figure 5A). The results revealed that 24-hour treatment with 10 μmol/L TAM induced abnormally increased cell density, disorganized growth, and disordered arrangement in GES-1 cells, with visible cell aggregation in localized areas. Both 2-DG and pristimerin treatments effectively ameliorated these pathological phenotypes (Figure 5A). Analysis of key glycolytic regulators and metabolites demonstrated that TAM induction significantly upregulated the expression of HIF-1α, LDHA, and PKM2, while promoting the accumulation of pyruvate and lactate. Notably, interventions with either 2-DG or pristimerin significantly suppressed these aberrant pathological alterations. Crucially, the combined intervention exhibited significantly stronger inhibitory effects than 2-DG monotherapy (Figure 5B and C). These findings indicate that pristimerin not only effectively counteracts the aberrantly enhanced glycolysis in TAM-induced GES-1 cells but may also potentiate glycolytic suppression through a mechanism independent of 2-DG.

Figure 5 Pristimerin ameliorates tamoxifen-induced spasmolytic polypeptide-expressing metaplasia phenotype in human gastric epithelial cells via suppression of glycolysis. A: Representative images of tamoxifen-induced human gastric epithelial (GES-1) cells following pristimerin or 2-deoxy-D-glucose (2-DG) intervention. Scale bar, 100 μm; B: Expression of key glycolytic regulators in GES-1 cells following pristimerin or 2-DG intervention; C: Quantitative analysis of pyruvic acid and lactate production in GES-1 cells; D: 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay for GES-1 cell viability; E: 5-Ethynyl-2’-deoxyuridine assay for detecting proliferation activity in GES-1 cells. Scale bar, 20 μm; F: Inhibitory effects of pristimerin or 2-DG on spasmolytic polypeptide-expressing metaplasia-specific marker expression in GES-1 cells (Western blot; β-actin-normalized relative expression vs control); G: Expression of stem cell markers in GES-1 cells following pristimerin or 2-DG treatment. Scale bar, 20 μm. ^aP < 0.05. ^bP < 0.01. ^cP < 0.001. ^dP < 0.0001. Values represent the mean ± SD (n = 5). 2-DG: 2-deoxy-D-glucose; Pri: Pristimerin; HIF-1α: Hypoxia-inducible factor-1α; LDHA: Lactate dehydrogenase A; PKM2: Pyruvate kinase M2; TAM: Tamoxifen; DAPI: 4’,6-diamidino-2-phenylindole; EdU: 5-Ethynyl-2’-deoxyuridine; MUC6: Mucin 6; TFF2: Trefoil factor 2; WFDC2: Whey-acidic-protein four-disulfide core domain protein 2; Lgr5: Leucine-rich repeat-containing G-protein coupled receptor 5; Troy: Tumor necrosis factor receptor superfamily member 19; IOD: Integrated optical density.

Further investigation revealed that 2-DG and pristimerin treatment differentially suppressed the TAM-induced pathological enhancement of cell viability and proliferative activity in GES-1 cells (Figure 5D and E). More importantly, both agents significantly suppressed the aberrant upregulation of SPEM-specific markers (TFF2, MUC6, WFDC2) and stem cell markers (Lgr5, Troy). Furthermore, the combined treatment of pristimerin and 2-DG exerted a markedly stronger inhibitory effect on these markers than 2-DG monotherapy (Figure 5F and G). Collectively, these results indicate that pristimerin ameliorates the TAM-induced SPEM phenotype in GES-1 cells by modulating glycolytic reprogramming. The synergistic effect observed when combined with 2-DG may arise from their profound combined inhibition of the glycolytic pathway or bypass mechanisms beyond the classical glycolytic route.

Pristimerin upregulates p57 to modulate glycolytic reprogramming and ameliorate the SPEM phenotype in gastric organoids

To investigate pristimerin’s regulation of p57, we assessed its expression in mouse gastric tissues via immunofluorescence. HDT-induced SPEM mice exhibited dramatically reduced p57 expression, while pristimerin treatment significantly upregulated it, with positive signals predominantly localized at the glandular base (Figure 6A). To further clarify the role of p57 in pristimerin-mediated glycolytic reprogramming and SPEM amelioration, we established an SPEM gastric organoid model induced by MNNG/H. pylori (Supplementary Figure 2B-D) and generated a p57-overexpressing model (Cdkn1c-OE) via plasmid transfection (Supplementary Figure 3A). Organoids were randomized into control, model, Cdkn1c-OE, Pri, and Pri + Cdkn1c-OE groups, with corresponding interventions applied.

Figure 6 Pristimerin upregulates p57 to modulate glycolytic reprogramming and ameliorate the spasmolytic polypeptide-expressing metaplasia phenotype in gastric organoids. A: Pristimerin upregulates gastric mucosal p57 in spasmolytic polypeptide-expressing metaplasia (SPEM) mice. Scale bar, 20 μm; B: Representative images of normal gastric organoids, N-methyl-N’-nitro-N-nitrosoguanidine/Helicobacter pylori-induced SPEM gastric organoids, and those following Cdkn1c overexpression or pristimerin intervention, along with quantitative analyses of organoid number and diameter. Scale bar, 100 μm; C: P57 expression in SPEM gastric organoids following Cdkn1c overexpression or pristimerin treatment. Scale bar, 10 μm; D-F: Both Cdkn1c overexpression and pristimerin treatment suppressed cell proliferation [reverse transcription quantitative polymerase chain reaction (RT-qPCR); Western blot, β-actin-normalized relative expression vs control] in SPEM gastric organoids, downregulated key glycolytic regulators (RT-qPCR; Western blot), stem cell markers (immunofluorescence: Scale bar, 10 μm), and SPEM-specific markers (RT-qPCR; Western blot), while promoting gastric intrinsic factor (RT-qPCR; Western blot) expression with pristimerin exhibiting stronger effects. ^aP < 0.05. ^bP < 0.01. ^cP < 0.001. ^dP < 0.0001. Values represent the mean ± SD (A: n = 6; B-F: n = 5). NS: Not significant; Cdkn1c-OE: Cdkn1c-overexpressing spasmolytic polypeptide-expressing metaplasia gastric organoids; DAPI: 4’,6-diamidino-2-phenylindole; GIF: Gastric intrinsic factor; Pri-L: Low-dose pristimerin; Pri-M: Medium-dose pristimerin; Pri-H: High-dose pristimerin; IOD: Integrated optical density; HIF-1α: Hypoxia-inducible factor-1α; LDHA: Lactate dehydrogenase A; PKM2: Pyruvate kinase M2; MUC6: Mucin 6; TFF2: Trefoil factor 2; WFDC2: Whey-acidic-protein four-disulfide core domain protein 2; Lgr5: Leucine-rich repeat-containing G-protein coupled receptor 5; Troy: Tumor necrosis factor receptor superfamily member 19.

Post-intervention analysis revealed that MNNG/H. pylori treatment significantly reduced both the number and diameter of organoids, whereas both Cdkn1c overexpression and pristimerin intervention effectively ameliorated this phenotype (Figure 6B). Immunofluorescence confirmed that both plasmid transfection and pristimerin treatment upregulated p57 expression in gastric organoids (Figure 6C) while concurrently suppressing Ki67, key glycolytic regulators (HIF-1α/PKM2/LDHA; Figure 6D), stem cell markers (Lgr5/Troy; Figure 6E), and SPEM-specific markers (TFF2/MUC6/WFDC2; Figure 6F). These results demonstrate that downregulated p57 expression in SPEM lesions is closely associated with aberrant glycolytic reprogramming and activation of malignant phenotypes, and that restoration of p57 expression effectively suppresses this pathological mechanism, while pristimerin inhibits glycolytic pathways and SPEM phenotypes by upregulating p57.

Cdkn1c knockdown attenuates pristimerin’s regulation of glycolytic reprogramming and amelioration of SPEM

To validate the regulatory mechanism by which pristimerin targets p57, we constructed a p57-knockdown model (Cdkn1c-KD) in SPEM gastric organoids using Cdkn1c siRNA (Supplementary Figure 3B) and established five groups, including control, model, Cdkn1c-KD, Pri, and Pri + Cdkn1c-KD. The results demonstrated that Cdkn1c knockdown significantly inhibited organoid growth, with both number and diameter markedly lower than those in the model group. Notably, pristimerin’s ability to ameliorate this growth inhibition was substantially attenuated under p57-deficient conditions (Figure 7A). Mechanistically, knockdown of Cdkn1c led to a significant down-regulation of its messenger RNA and the encoded p57 protein, to levels even lower than those in the model. Although pristimerin up-regulated the expression of both in both SPEM and Cdkn1c-knockdown gastric organoids, this effect was significantly attenuated in the knockdown models (Figure 7B). Crucially, compared to the model group, Cdkn1c knockdown triggered multi-pathological cascades, further upregulating Ki67, key glycolytic regulators, stem cell markers, and SPEM-specific markers (Figure 7C-E). These findings establish p57 as a critical negative regulator in glycolytic reprogramming and SPEM progression. Corresponding intervention experiments demonstrated that pristimerin significantly suppressed glycolytic reprogramming and the SPEM phenotype in gastric organoids; however, this inhibitory effect was markedly attenuated upon Cdkn1c knockdown (Figure 7C-E). These findings indicate that p57 is an indispensable mediator essential for pristimerin to exert its ameliorative effects on SPEM.

Figure 7 Cdkn1c knockdown attenuates pristimerin’s regulation of glycolytic reprogramming and amelioration of spasmolytic polypeptide-expressing metaplasia. A: Representative images of normal gastric organoids, N-methyl-N’-nitro-N-nitrosoguanidine/Helicobacter pylori-induced spasmolytic polypeptide-expressing metaplasia (SPEM) gastric organoids, and those following Cdkn1c knockdown or pristimerin intervention, along with quantitative analyses of organoid number and diameter. Scale bar, 100 μm; B: Effects of Cdkn1c knockdown or pristimerin treatment on p57 expression levels in gastric organoids. Scale bar, 10 μm; C-E: Cdkn1c knockdown promotes cell proliferation [reverse transcription quantitative polymerase chain reaction (RT-qPCR); Western blot, β-actin-normalized relative expression vs control] and enhances the expression of key glycolytic regulators (RT-qPCR; Western blot), stem cell markers (immunofluorescence: Scale bar, 10 μm), and SPEM-specific markers (RT-qPCR; Western blot) in SPEM gastric organoids, while suppressing gastric intrinsic factor (RT-qPCR; Western blot) expression. It also attenuates the regulatory effects of pristimerin on these phenotypes. ^aP < 0.05. ^bP < 0.01. ^cP < 0.001. ^dP < 0.0001. Values represent the mean ± SD (n = 5). Cdkn1c-KD: Cdkn1c-knockdown spasmolytic polypeptide-expressing metaplasia organoids; DAPI: 4’,6-diamidino-2-phenylindole; GIF: Gastric intrinsic factor; Pri: Pristimerin; IOD: Integrated optical density; HIF-1α: Hypoxia-inducible factor-1α; LDHA: Lactate dehydrogenase A; PKM2: Pyruvate kinase M2; MUC6: Mucin 6; TFF2: Trefoil factor 2; WFDC2: Whey-acidic-protein four-disulfide core domain protein 2; Lgr5: Leucine-rich repeat-containing G-protein coupled receptor 5; Troy: Tumor necrosis factor receptor superfamily member 19.

DISCUSSION

According to the Correa cascade hypothesis, the normal gastric mucosa must undergo a series of pathological stages to develop into GC, primarily including CAG, intestinal metaplasia (IM), and dysplasia. Among these stages, the transition into metaplasia represents the pivotal point at which the gastric mucosa acquires malignant transformation potential[30]. Recent studies have found that SPEM is not only an important precursor lesion of IM, but also a key link in the malignant progression of gastric mucosa, and its persistent existence will significantly increase the risk of mucosal carcinogenesis[31,32]. Furthermore, clinicopathological analysis further shows that SPEM cells are found in 90% of resected GC tissues and exhibit high microsatellite instability and genome-wide genetic instability[33]. This further demonstrates the central role of SPEM in the development process of GC. Therefore, early intervention targeting SPEM is of great significance for blocking the malignant progression of gastric mucosa.

Recent studies on Traditional Chinese medicine compound formulas in SPEM treatment mechanisms have achieved preliminary progress. For instance, Weiwei decoction can improve SPEM in both H. pylori-infected mice and Atp4a gene knockout mice by inhibiting intestinal-type gastric adenocarcinoma-associated transcripts, restoring the expression of Atp4a and pepsinogen II, and reducing M2 macrophage infiltration[34]. Similarly, Zuojin capsule alleviates TAM-induced SPEM by modulating cyclins and the phosphatidylinositol 3-kinase/protein kinase B signaling axis[26]. It is noteworthy that, compared to complex herbal preparations, their bioactive compounds generally exhibit superior cell membrane permeability and more targeted regulatory capacity for intracellular signaling[35]. The COE of Celastrus orbiculatus Thunb. has been demonstrated to effectively inhibit epithelial-mesenchymal transition in gastric mucosa, suppress malignant proliferation and invasion of GC cells, as well as reverse GPL[16,17,20]. As a major bioactive component of this plant, the anti-tumor potential of pristimerin has been confirmed by numerous studies. Building upon this foundation, the present study further delineates the therapeutic efficacy of pristimerin against SPEM and provides initial insights into its mechanism of action.

In this study, utilizing an HDT-induced SPEM mouse model, we demonstrated that pristimerin significantly alleviates gastric mucosal pathological damage, including improvement of oxyntic atrophy and suppression of metaplastic lesions. This robustly confirms pristimerin’s therapeutic potential in reversing SPEM. At the molecular level, pristimerin effectively suppressed SPEM characteristics in gastric mucosal cells by downregulating specific markers TFF2 and MUC6. WFDC2 a small secretory protein overexpressed in numerous cancers and correlated with adverse prognosis[36,37] was recently shown to promote SPEM progression via interleukin-33 upregulation[27]. Our findings reveal that pristimerin significantly downregulates WFDC2 in gastric tissues of SPEM mice. Furthermore, we observed aberrantly elevated expression of the stem cell markers Lgr5 and Troy in the gastric corpus of SPEM mice. As reported, Lgr5⁺/Troy⁺ stem cells serve as critical origins of GC cells, where their dysregulated overexpression drives tumorigenesis, progression, and metastasis[29,38,39]. Consequently, the pathological upregulation of Lgr5 and Troy in SPEM mice may suggest that these tissues have already acquired features predisposing them to malignant transformation. Importantly, pristimerin treatment not only suppressed these SPEM-specific markers but also significantly downregulated Lgr5 and Troy expression. Collectively, these results indicate that pristimerin synergistically reverses core molecular features of SPEM through multi-target regulation, thereby inhibiting disease progression.

Mechanistically, our data demonstrate that the ameliorative effect of pristimerin on SPEM is mediated through its modulation of glycolytic reprogramming. Glycolytic reprogramming is an adaptive change in energy metabolism for cells coping with pathological microenvironments, by regulating gene expression, enzyme activity, and signaling pathways to satisfy demands for proliferation, stress response, or differentiation[40]. This phenomenon is extensively documented in tumors, where its functional scope extends beyond rapid energy provision. Studies demonstrate that tumor cells competitively deplete nutrients to impair immunometabolic fitness, thereby suppressing antitumor immunity[41,42]. Critical rate-limiting enzymes (e.g., HIF-1α, PKM2, LDHA) and metabolites (e.g., lactate) within this metabolic cascade directly drive malignant progression[43-46]. Recent studies have shown that gastric mucosal cells in GPL enhance glycolysis to adapt to the chronic inflammatory and hypoxic microenvironment, whereas inhibition of glycolysis reverses GPL progression[13-15]. These findings confirm that glycolytic reprogramming serves as a core mechanism throughout the entire process from precancerous lesions to malignant tumors. Its metabolites and cascading effects constitute pivotal drivers of disease progression, establishing metabolic intervention as a vital therapeutic strategy[47]. Our study revealed that during SPEM progression, the expression of key glycolytic regulators (HIF-1α, PKM2, LDHA) is upregulated, glycolytic efficiency is increased, and metabolites accumulate. It is noteworthy that inhibition of glycolysis significantly alleviates disease progression, while pristimerin can improve SPEM by modulating glycolytic reprogramming.

Further investigation identified Cdkn1c/p57, a member of the CIP/KIP family, as the key molecular target through which pristimerin regulates glycolytic reprogramming and ameliorates SPEM phenotypes. Previous studies have established that p57 not only inhibits tumor growth through regulating cell cycle progression, proliferation, metastasis, and apoptosis[48-50], but also suppresses disease progression in melanoma by inhibiting glycolysis and promoting oxidative phosphorylation[51]. In regulating gastric mucosal homeostasis, sustained high expression of p57 is essential for maintaining the “RSC” state of chief cells, while its downregulation triggers aberrant proliferation and malignant phenotypes[7]. Our findings in both SPEM mouse and organoid models further validated this mechanism, demonstrating that p57 downregulation consequently led to the loss of chief cell characteristics, emergence of the SPEM phenotype, and abnormal gastric mucosal cell proliferation, whereas its restoration effectively reversed these pathological alterations. Notably, pristimerin exerted its therapeutic effects specifically through restoring p57 expression, and its efficacy was significantly attenuated under p57-deficient conditions. Furthermore, our study reveals for the first time in SPEM models the crucial regulatory role of p57 in glycolytic reprogramming, demonstrating that p57 deficiency induced upregulated expression of key glycolytic regulators including HIF-1α, LDHA, and PKM2, while p57 restoration effectively corrected this aberrant activation.

Finally, it is imperative to conscientiously acknowledge the limitations of this study. First, as a preclinical investigation, the findings are entirely derived from animal, cell, and organoid models, and lack human pharmacokinetic, bioavailability, and toxicity data, which poses challenges for clinical translation. Second, the experimental sample size was relatively small, and the in vivo studies did not include mechanistically well-defined positive control groups. Third, real-time metabolic analysis of glycolytic reprogramming was lacking. Fourth, the upstream regulatory mechanisms of pristimerin on Cdkn1c/p57 remain to be fully elucidated. Lastly, it is noteworthy that due to issues such as toxicity concerns, species differences, and the absence of comprehensive preclinical safety data, the effective dosage and delivery route of pristimerin established in the current mouse model cannot be directly extrapolated to humans. Before considering its clinical relevance, comprehensive pharmacological optimization of pristimerin is necessary. Future studies should enlarge the sample size, incorporate mechanistically defined positive controls to validate and extend the current findings, and systematically conduct preclinical pharmacokinetic and safety evaluations to provide a solid foundation for clinical translation.

CONCLUSION

In summary, this study provides the first demonstration of pristimerin’s significant ameliorative effect on SPEM and elucidates its core mechanism in ameliorating disease progression through regulation of Cdkn1c (p57)-mediated glycolytic reprogramming (Supplementary Figure 4). This finding not only establishes p57 as a critical target for pristimerin-mediated amelioration of SPEM, but also provides an important theoretical foundation for further exploration of the molecular mechanisms in gastric mucosal lesions and the development of therapeutic strategies targeting the glycolytic pathway.

ACKNOWLEDGEMENTS

The authors sincerely thank the laboratory colleagues for their excellent technical assistance.