Association of DPP4 with esophageal stricture progression and the Hippo‐YAP pathway Authors: Rui Wu, Chen Wang, Hui‐Min Tao, Jin Huang, Wei‐Qin Li, Zhi‐Ning Fan Published: 2026 ABSTRACT AbstractBackgroundEsophageal stricture (ES), a common complication after endoscopic submucosal dissection (ESD), is the consequence of excessive fibrosis and scar formation. The fibroblast‐to‐myofibroblast differentiation is a key characteristic in the pathogenesis of ES, but the molecular basis remains poorly understood.MethodsAccording to the human transcriptome sequencing analysis, which utilized different expression gene analysis and weighted gene co‐expression network analysis, the potential gene associated with ES was explored. Furthermore, we utilized the rat model of ES and the temporal progression of ES after drug treatment to elucidate the role of DPP4 in ES. Subsequently, we identified a significant association between DPP4 and the progression of ES. Finally, we validated our conclusions through experiments on Bama pigs.ResultsDPP4 expression was consistently elevated in both human ES tissues and the rat ES model. Upon inhibition of DPP4, the proliferative activity of ES primary cells and the expression of fibrosis‐related genes were markedly suppressed. In rats, treatment with a DPP4 inhibitor significantly alleviated ES, which was accompanied by a significant reduction in collagen content and downregulation of fibrosis‐related targets. These effects were observed in conjunction with changes in the Hippo‐YAP pathway, although the data primarily support a parallel association rather than a hierarchical regulatory relationship between DPP4 and this pathway.ConclusionsOur findings suggest that DPP4 may contribute to the remodeling of the regenerative microenvironment in ES, with its expression and function occurring in parallel with alterations in the Hippo‐YAP pathway. Direct mechanistic evidence establishing a linear upstream–downstream relationship between DPP4 and Hippo‐YAP signaling remains to be further elucidated.Keywords: DPP4, esophageal stricture, fibrosis, YAP1 INTRODUCTION Esophageal cancer is ranked as the sixth leading cause of cancer‐related mortality globally. 1 When detected early and treated endoscopically, esophageal cancer survival rates exceed 90% in expert centers. 2 In comparison to esophagectomy, endoscopic submucosal dissection (ESD) shows no significant difference in overall survival (86.4% vs. 81.8%, 95% confidence interval [CI] 0.39–1.11). 3 However, endoscopic treatment is associated with fewer severe complications (15.2% vs. 27.7%, p < 0.001), 4 is better tolerated, incurs lower cost, and is preferred by patients over surgery. Thus, most recent guidelines recommend endoscopic treatment as the first‐line approach for early lesions confined to the mucosa or superficial submucosa. 5 However, esophageal stricture (ES), a common complication after ESD, has hindered the further popularization of this technique. When the resected area exceeds 5 cm in length, the occurrence of ES is common, and in cases of circumferential resections, stricture formation is nearly universal. 6 , 7 , 8 Recent theories behind ES include the formation of granulation tissue, horizontally oriented myofibroblasts, and dense collagen tissue. 9 Up to now, different prophylactic methods have been explored, including steroids, endoscopic balloon dilation, and stent placement. 10 Unfortunately, the curative effects have been variable, and the quest for an effective method is still ongoing. Fibroblast‐to‐myofibroblast differentiation, a key cellular event in fibrosis, is classically mediated by the transforming growth factor β (TGF‐β)/Smad signaling pathway. This canonical pathway has been extensively characterized in various fibrotic diseases, including pulmonary fibrosis, liver cirrhosis, and cardiac fibrosis. However, emerging evidence suggests that additional signaling pathways may contribute to or modulate this process. Recent studies, including those by Shen et al., 11 have revealed an intriguing interplay between the Hippo pathway and TGF‐β/Smad signaling in regulating DPP4+ fibroblast proliferation and adipocyte fibrosis, suggesting that these pathways may cooperate in driving fibrotic responses. Despite these advances in other tissues, the molecular mechanisms underlying ES—a fibrotic complication following endoscopic submucosal dissection—remain poorly understood, largely due to the historical lack of stable and reproducible animal models. Our team has previously established a stable rat model of ES. 12 Leveraging this model, the present study aimed to investigate the involvement of the Hippo‐YAP signaling pathway in ES. Although we acknowledge the well‐established role of TGF‐β/Smad signaling in fibrosis, our focus on Hippo‐YAP was guided by transcriptomic analysis of ES tissues, which revealed significant enrichment of this pathway. We present our findings as evidence of Hippo‐YAP's association with ES, recognizing that this pathway likely operates in concert with, rather than in isolation from, other canonical fibrotic mechanisms. METHODS Animal The study was approved by the Ethics Committee of Nanjing Medical University. All experiments were performed under the supervision of the Ethics Committee of Nanjing Medical University in accordance with the relevant guidelines (IACUC‐2103014, Institutional Animal Care and Use Committee of Nanjing Medical University). All methods are reported in accordance with ARRIVE guidelines. The rats were euthanized with cervical dislocation in accordance with the relevant guidelines. The rats were anesthetized with an intraperitoneal injection of 2% sodium pentobarbital at a dose of 0.2 mL/100 g. All animal experiments were conducted in strict accordance with the approved protocols by the Ethics Committee of Nanjing Medical University. After euthanasia, the rats and pigs employed in our study were carefully collected. White male Sprague Dawley (SD) rats, aged 8–10 weeks and weighing 250 ± 20 g, were procured from the Jiangsu Animal Experimental Center of Medical & Pharmaceutical Research, situated in Nanjing, China. These rats were accommodated in the specific pathogen‐free laboratory facility, each residing in standard cages with three rats per enclosure, all subjected to a constant temperature (25 ± 2°C), humidity (60% ± 10%), and a 12‐h light/dark cycle. For the treatment group rats, we administered a daily oral dose of 2.5 mg/kg 13 of vildagliptin (a DPP4 inhibitor) for 4 weeks before surgery, maintaining the treatment until 2 weeks postoperatively. Conditions of housing and nutrition were identical for normal, ES, and treatment group rats, with ample access to food and water provided in all cases. The conventional‐grade Bama miniature pigs were procured from Jiangsu Langyue Biotechnology Co., Ltd. The experimental pigs were exclusively males, weighing 30 ± 3 kg with an age of 200–250 days. The pigs were housed and maintained at the Standard Experimental Animal Breeding Center (Nanwei Medical Science and Technology Co., Ltd.). For the treatment group pigs, we administered a daily oral dose of 2 mg/kg 14 of vildagliptin for 4 weeks before surgery, maintaining the treatment until 3 weeks postoperatively. Sample sizes (at least n = 5 per group for rat studies; n = 3 per group for pig studies) were determined based on commonly accepted standards in fibrosis research and previous publications using similar animal models. 15 , 16 Cell isolation, culture, and transfection Primary rat fibroblasts were isolated using a method previously delineated by Tian et al. 17 In brief, esophageal tissues from six rats with ES were pooled and subjected to digestion with collagenase II (Solar Bio). Following a 30‐min digestion period at 37°C, the remaining microtissues and cells were cultured in 10‐cm culture plates. The culture medium consisted of Dulbecco's modified Eagle medium (DMEM)/F12 supplemented with 10% fetal bovine serum and 1% penicillin–streptomycin, maintained at 37°C in a humidified atmosphere containing 5% CO2. Primary cells were transiently transfected with small interfering RNAs (siRNAs) (GEMA, Shanghai, China) targeting DPP4 messenger RNAs (mRNAs) and their negative controls using the Lipofectamine 3000 Reagent (Invitrogen, L3000015) according to the protocol. To generate DPP4 overexpression plasmids, DPP4 complementary DNAs (cDNAs) were synthesized separately using PrimerSTAR Max DNA Polymerase Mix (Takara, RR036A, Japan) and subcloned into the pENTR vector fused with 3X‐Flag at the C terminus (Vigene Biosciences, Shandong, China). Plasmid transfection was performed using Lipofectamine 3000 (Invitrogen, USA) following the manufacturer's protocol. Participants Ethical approval for all study procedures was granted by the Ethics Committee of the First Affiliated Hospital with Nanjing Medical University (Nanjing, China). This research was performed in accordance with relevant guidelines/regulations, and the informed consent was obtained from all participants and/or their legal guardians. Research involving human research participants has been performed in accordance with the Declaration of Helsinki. Human esophageal tissue comprises a sequencing set and a validation set. The ES tissues in the sequencing set were procured from eight patients (age ranging from 61 to 77 years) who underwent endoscopy at the First Affiliated Hospital with Nanjing Medical University, and the normal esophageal tissues were obtained from 5 cm away from ES tissues in these patients. In the validation set, ES and normal tissues were sourced from 16 patients (10 for immunohistochemistry and 6 for polymerase chain reaction [PCR]) who underwent endoscopy at the same hospital. Detailed sample information is provided in Table S1. Furthermore, we collected tissue samples from the stenotic site and 5 cm adjacent to the stenosis in the esophagus of seven patients who underwent circumferential esophageal ESD on postoperative day 5 and day 14. Immunohistochemical staining was performed on the samples targeting Yes‐associated protein (YAP). For detailed information, refer to Table S2. In addition, a retrospective analysis was conducted. Inclusion criteria included the following parameters: (1) patients who underwent circumferential esophageal ESD with a length > 5 cm; (2) patients diagnosed with diabetes, receiving daily oral DPP4 inhibitors for more than 12 months; (3) patients undergoing endoscopic follow‐up within 3 months postoperatively, without additional therapeutic interventions (such as hormone therapy). A total of eight patients were included in the study, and eight patients from those who underwent circumferential esophageal ESD with a length > 5 cm, without diabetes and not taking DPP4 inhibitors, were randomly selected as the control group (Table S3). Histological evaluation The esophageal samples were promptly fixed in 4% paraformaldehyde upon rat dissection. After fixation, the tissues underwent paraffin embedding and sectioning. Then, the hematoxylin–eosin staining was performed. The paraffin sections were dried at 65°C in an oven, deparaffinized in xylene, and rehydrated in a gradient of alcohols (100%–95%–85%–75%). Subsequently, samples were immersed in hematoxylin for 120 s, followed by rinsing with hydrochloric acid, 75% alcohol, and tap water for 10 s each. The next step involved eosin staining for 5 min, followed by placement in a gradient of alcohols (85%–95%–100%) and xylene for dehydration. In addition, the Masson's trichrome and Sirius Red staining were conducted according to the protocol and instructions of the Masson staining kit (JSENB, Hong Kong, China) and Sirius Red staining kit (Abcam, Shanghai, China). Immunohistochemistry staining Immunohistochemistry (IHC) staining was conducted subsequent to paraffin embedding and sectioning of the samples. The sections underwent a series of treatments including three rounds of 15‐min incubations in xylene, exposure to absolute ethanol thrice, followed by rinsing with deionized water. Antigen retrieval was performed, and the samples were immersed in 3% methanol hydrogen peroxide for 25 min, rinsed with deionized water, and then placed in phosphate‐buffered saline (PBS) for three cycles. The slides were incubated overnight at 4°C with primary antibodies (Rat anti‐DPP4, catalog no. ab187048; Pig and Human anti‐DPP4, catalog no. ab231973; YAP antibody, catalog no. 13584‐1‐AP; α‐smooth muscle actin (α‐SMA) antibody, catalog no. 14395‐1‐AP; TGF‐β1 antibody, catalog no. ab315254; COL1A1 antibody, catalog no. ab279711; integrin β1 antibody, catalog no. ab179471; integrin β3 antibody, catalog no. 18309‐1‐ap). After three washes in PBS, the slides were subjected to a 50‐min incubation at 37°C with secondary antisera (dilution 1:200, Servicebio, GB23303). After three additional PBS washes, 3,3′‐diaminobenzidine tetrahydrochloride (DAB) was applied, and the slides were rinsed in PBS and deionized water. A 3‐min hematoxylin counterstain was performed, and the slides were washed in tap water. Subsequently, the pathology slices were captured using an electron microscope and analyzed using Image J software. Immunofluorescence staining Sections obtained from paraffin embedding were utilized for immunofluorescence staining. After antigen retrieval and blocking procedures, the sections were incubated overnight at 4°C with primary antibodies. Subsequently, the sections underwent a 50‐min incubation at 37°C with the corresponding secondary antisera (dilution 1:200, Servicebio, GB23303). Cell nuclei were stained with 4′,6‐diamidino‐2‐phenylindole (DAPI), and the slices were captured using a confocal laser scanning fluorescence microscope. Polymerase chain reaction Total RNA extraction was carried out using Trizol, and subsequent reverse transcription into cDNA was accomplished using a PrimeScript RT Master Mix Kit. Real‐time polymerase chain reactions (RT‐PCRs) were conducted utilizing SYBR Green on the QuantStudio (TM) 7 Flex System, adhering to the manufacturer's protocols, and the comparative Cт (ΔΔCт) method was employed for result analysis. β‐Actin served as the internal control. 5‐Ethynyl‐2′deoxyuridine assay The fibroblast proliferation was evaluated utilizing the 5‐ethynyl‐2′deoxyuridine assay (EdU) assay kit (RiboBio, Guangzhou, China) in adherence to the manufacturer's guidelines. Briefly, 1 × 104 fibroblasts were seeded into 48‐well plates and cultured for 24 h. Subsequently, the cells were exposed to 50 μmol/L EdU for 2 h, fixed with 4% paraformaldehyde, and stained with Apollo Dye Solution and Hoechst 33342. RNA transcriptome sequencing and bioinformatics analysis This study was the first to perform a transcriptome analysis of ES tissues of humans. Total RNA was extracted from mucosa and submucosal tissue of eight patients (Table S1, ID: S1–S8). Adjusted p < 0.05 and |log2 fold change| > 0 were set as the criteria for selecting significant differentially expressed genes (DEGs). Weighted gene co‐expression network analysis (WGCNA) is a methodology used to reconstruct a gene co‐expression network and concurrently identify modules consisting of highly correlated genes to assess connectivity with external clinical traits. 18 In this study, we calculated a weighted adjacency matrix to depict the strength of connections between each pair of genes, and outliers were excluded to ensure the robustness of the co‐expression network. The soft thresholding power was set to 8 to achieve a scale‐free topology network. Subsequently, a hierarchical clustering dendrogram was constructed with distinct branches, each assigned a unique color to represent a module. Finally, these modules were assessed for their association with clinical traits through module‐trait associations based on the module–sample relationship. Additionally, the topological overlap matrix method was employed to verify the correlation characteristics of eigengenes in different modules. Statistics The statistical analysis was conducted using SPSS software version 25.0 (IBM Corp, Armonk, NY, USA). Categorical variables underwent analysis using the chi‐square test or Fisher's exact test, whereas continuous variables were assessed using the t‐test. The significance threshold for statistical analysis was set at p < 0.05. RESULTS Hub genes associated with esophageal stricture A total of 3746 DEGs were identified in human ES tissues compared to normal tissues, of which 2917 genes were upregulated and 829 were downregulated (Figure 1A). The clustering analysis was performed, and genes or samples with similar expression patterns were clustered (Figure S1A). In addition, the WGCNA was conducted, and the power of β = 8 was chosen as the soft‐thresholding parameter to ensure a scale‐free network (Figure S1B,C). A total of 23 modules were identified from 15 872 genes, and each module was assigned a color in hierarchical clustering dendrogram (Figure 1B). The heatmap was plotted to indicate the similarity of co‐expression genes at the network topology level (Figure 1C). The yellow module (containing 1195 genes) was found to have the most prominent module significance (Figure 1D). Consequently, the yellow module was set as the hub module. The heatmap of correlation analysis in all modules is presented in the Figure S1D. Identification of hub genes associated with esophageal stricture. (A) Volcano plot showing differentially expressed genes (DEGs) between stenotic and paired adjacent normal esophageal tissues from eight patients. Red dots indicate significantly upregulated genes (|log2FC| > 0, adjusted p < 0.05), green dots indicate significantly downregulated genes, and blue dots indicate nonsignificant genes. (B) The hierarchical clustering dendrogram of weighted gene co‐expression network analysis (WGCNA). Each branch represents a gene, and each color block represents a co‐expression module. (C) The heatmap of the similarity of co‐expression genes at the network topology level. (D) Heatmap of module–sample relationships, displaying the correlation between each module and clinical traits (stenosis vs. normal). The yellow module showed the highest correlation with esophageal stricture (ES). (E) The Venn diagram showing the intersecting genes between DEGs and key module genes identified using WGCNA. (F) Protein–protein interaction (PPI) network of the hub genes, constructed using the STRING database. Nodes represent proteins, and edges represent interactions. (G, H) Representative immunohistochemistry (IHC) images and quantification of DPP4 expression in normal and ES human tissues. Scale bar = 100 μm. DPP4 expression was significantly higher in ES tissues (n = 10, p < 0.001). (I) Quantitative real‐time polymerase chain reaction (RT‐PCR) analysis of DPP4 messenger (mRNA) expression in normal and ES human tissues (n = 6, p = 0.012). *p < 0.05, ***p < 0.001; error bars represent the standard deviation (SD). Furthermore, we proceeded to intersect the DEGs with the hub module genes identified using WGCNA (Figure 1E). The resulting intersection genes (n = 479) were subjected to a protein–protein interaction (PPI) network analysis, from which we selected the top 100 genes based on the degree (Figure S1E). We posit that these genes exhibit a high correlation with ES. Moreover, these genes were subjected to Kyoto Encyclopedia of Genes and Genomes (KEGG) and Disease Ontology (DO) analyses (Figure S1F,G). The findings indicated a notable enrichment in pathways associated with fibrosis, including RHO GPTase effectors, 19 chemokine receptor binding to chemokines, 20 , 21 , 22 and interferon signaling. 23 , 24 Subsequently, a thorough review of pertinent literature revealed a pronounced association between the DPP4 gene and skin fibrosis, 25 , 26 , 27 , 28 a mechanism analogous to that of ES. Therefore, we contemplate the significant role of DPP4 in the context of ES (Figure 1F). Moreover, significant upregulation of DPP4 in human ES tissues was observed in the validation set (immunohistochemistry [IHC] [n = 10, p < 0.001), PCR [n = 6, p = 0.012]) (Figure 1G–I). This comprehensive validation at the molecular and genetic levels adds robustness to our findings, affirming the pronounced elevation of DPP4 target expression within the pathological context of ES in humans. Constructing the rat model of esophageal stricture Our research team has successfully established a stable rat model of ES in the preliminary stages (Figure 2A). The structural disarray and collagen deposition can be found in the ES tissues (Figure 2B,C). The intricate details can be obtained in our earlier study. 12 We observed a significant elevation of DPP4 in the ES rat model compared to the normal tissues (n = 6, p = 0.006) (Figure 2D) and primarily concentrated in the mucosal and submucosal layers (n = 16, p = 0.005) (Figure 2E,F), mirroring the results obtained from human specimens. This elucidates the significance and universality of DPP4 in ES to a certain extent, underscoring the concordance between the ES rat model and human ES. Construction and validation of a rat model of esophageal stricture (ES). (A) Schematic illustration of the surgical procedure for establishing the rat ES model. After midline abdominal incision and gastrostomy, an electric knife was used to retrogradely injure the esophageal mucosa and submucosa. The incision was sutured after damaging more than two‐thirds of the esophageal circumference. (B) The hematoxylin–eosin (HE) staining of the normal rats and ES rats. Scale bar = 100 μm. (C) The Masson staining of the normal rats and ES rats. Scale bar = 100 μm. (D) The polymerase chain reaction (PCR) of the normal rats and ES rats (n = 6, p = 0.006). (E, F) Representative immunohistochemistry (IHC) images and quantification of DPP4 protein expression in esophageal tissues from normal and ES rats. Scale bar = 100 μm. DPP4 expression was significantly increased in ES rats (n = 16, p = 0.005). **p < 0.01 ; error bars represent the standard deviation (SD). Inhibiting DPP4 significantly alleviates esophageal stricture We isolated primary fibroblasts from ES rats and downregulated DPP4. Then, a significant suppression in the proliferative activity of fibroblasts was observed (p < 0.001) (Figure 3A,B). Simultaneously, key genes associated with collagen and fibrosis (including COL1A1, COL3A1, acta2, TGF‐β1, and YAP1) were markedly inhibited (Figure 3C). We administered vildagliptin (a DPP4 inhibitor) orally at 2.5 mg/kg for ES rats. After 2 weeks postoperatively, the markedly elevated levels of DPP4 in ES rats exhibited a significant reduction after the oral administration of vildagliptin (Figure S2A). Simultaneously, cholangioscopy examination revealed a significant inhibition of ES in the rats (Figure S2B). Subsequent dissection of the rats confirmed a marked reduction in the esophageal constriction ratio (Figure S2C). Inhibiting DPP4 significantly alleviates esophageal stricture. (A, B) Representative images and quantification of 5‐ethynyl‐2′deoxyuridine assay (EdU) incorporation assay in primary esophageal fibroblasts transfected with control small interfering RNA (siRNA) (si‐NC) or DPP4‐targeting siRNA (si‐DPP4). EdU‐positive cells (red) indicate proliferating cells; nuclei were counterstained with 4′,6‐diamidino‐2‐phenylindole (DAPI) (blue). Scale bar = 20 μm. The percentage of EdU‐positive cells was significantly reduced in si‐DPP4 group (n = 6, p < 0.001). (C) The polymerase chain reaction (PCR) of COL1A1, COL3A1, acta2, TGF‐β1, and YAP1 on normal and CD26‐knockdown primary fibroblasts. (D) The hematoxylin–eosin (HE) staining of normal rats (n = 7), esophageal stricture rats (n = 7), and rats treated with DPP4 inhibition (n = 7). Scale bar = 100 μm. (E) The Masson staining of normal rats, esophageal stricture rats, and rats treated with DPP4 inhibition. Scale bar = 100 μm. (F) The immunohistochemistry (IHC) of the COL1A1 in normal rats, esophageal stricture rats, and rats with DPP4 inhibition. Scale bar = 100 μm. (G) The IHC of the tumor growth factor β (TGF‐β) in normal rats, esophageal stricture rats, and rats with DPP4 inhibition. Scale bar = 100 μm. (H) The immunofluorescence staining of the DPP4 and α‐smooth muscle Actin (α‐SMA) in esophageal stricture rats (a. stricture tissues, b. paraesophageal stricture tissues) and rats with DPP4 inhibition (c. stricture tissues, d. paraesophageal stricture tissues). (I–K) The integrated optical density (IOD) of DPP4, esophageal constriction ratio, and collagen area ratio in normal rats, esophageal stricture rats, and rats with DPP4 inhibition. *p < 0.05, **p < 0.01, ***p < 0.001; error bars represent the standard deviation (SD). Each group has seven replicates. Esophageal epithelial cells of ES rats undergo destruction and damage, followed by a process of reconstruction utilizing collagen in our previous research. 12 In the present study, we observed a similar phenomenon in ES rats, whereas the treated group of rats exhibited an undisturbed esophageal structure. There was an absence of submucosal fibrotic thickening, and the pathological architecture closely resembled that of normal rats (Figures 3D,E and S2D). IHC staining indicated a significant suppression of collagen‐related target (COL1A1) (Figures 3F and S2F) and fibrosis‐related target (TGF‐β1) (Figures 3G and S2G). In addition, the expression of myofibroblast marker (α‐SMA) significantly decreased with the inhibition of DPP4 (Figures 3H and S2E,H). The integrated optical density (IOD) of DPP4, esophageal constriction ratio, and collagen area ratio in normal rats, esophageal stricture rats, and rats with DPP4 inhibition were shown in Figure 3I‐K.This suggests the reversal of the fibrotic process and the attenuation of scar thickening, leading to the alleviation of ES. Pathological progression in esophageal stricture and treatment group rats Furthermore, we dissected the esophagus of rats from the ES and treatment groups on postoperative days 3, 7, 10, and 14 to observe the pathological progression. In the rats with ES, the levels of DPP4 gradually increased, whereas in the treatment group, DPP4 was maintained at relatively lower levels (Figure 4A). Pathological progression in esophageal stricture and treatment group rats. (A) Representative immunohistochemistry (IHC) staining of DPP4 in esophageal tissues from esophageal stricture (ES) rats and DPP4 inhibition‐treated rats at postoperative days 3, 7, 10, and 14 (n = 5 per group per time point). Scale bar = 100 μm. (B) Representative hematoxylin–eosin (HE) staining of esophageal tissues from ES rats and DPP4 inhibition‐treated rats at the indicated time points. Scale bar = 100 μm. (C) Representative Masson's trichrome staining of esophageal tissues from ES rats and DPP4 inhibition‐treated rats at the indicated time points. Scale bar = 100 μm. (D–F) The integrated optical density (IOD) of DPP4, esophageal constriction ratio, and collagen area ratio of ES rats and rats with DPP4 inhibition on postoperative days 3, 7, 10, and 14. *p < 0.05, **p < 0.01, ***p < 0.001; error bars represent the standard deviation (SD). Each group has five replicates at each time point. Macroscopic images from the outer layer and longitudinal sections of the esophagus indicate that both the stenotic and treated esophagus in the inflammatory phase (Figure S3A, day 3) exhibit relatively pronounced inflammatory reactions. The pathological examination indicates the disruption of the mucosal layer in both the ES and the treatment groups. The inflammatory cell infiltration and structural disarray are exhibited in the ES rats, with no collagen proliferation observed yet. Similar occurrences are also present in the treatment group, albeit to a relatively milder degree (Figures 4B,C and S3B, day 3). By the early proliferative phase (Figure S3A, day 7), there is notable edema present in both, accompanied by a slight increase in the esophageal diameter. The pathological examination indicates a substantial recruitment of fibroblasts and the initiation of collagen tissue formation in the ES rats. However, after the inhibition of DPP4, cellular density in the fibrotic area was reduced, and collagen deposition was less pronounced (Figures 4B,C and S3B, day 7), suggesting the potential initiation of ES reversal. As the late proliferative phase ensues (Figure S3A, day 10), the esophagus in the stenosis group undergoes significant constriction. Histologically, the inflammatory response had essentially resolved, whereas marked collagen deposition was observed in the submucosal layer. Conversely, in DPP4‐inhibited rats, Sirius Red staining revealed markedly less collagen deposition, correlating with the cessation of the ES process (Figures 4B,C and S3B, day 10). By the remodeling period, the emergence of a typical narrow morphology persists in the stenosis group (Figure S3A, day 14). There is a significant deposition of collagen and remodeling of esophageal mucosa and submucosa layers, whereas in the DPP4 inhibition group, collagen deposition is less pronounced, leading to an esophageal structure that closely resembles normal anatomy (Figures 4B,C and S3B, day 14). In summary, the levels of DPP4 increased in the rats with ES (Figure 4D). The inhibition of DPP4 leads to a gradual alleviation of esophageal narrowing, a reduction in collagen area, and a reversal of the ES process (Figure 4E,F). Association of DPP4 with esophageal stricture and the hippo‐YAP pathway For rats euthanized on days 3, 7, 10, and 14, transcriptome sequencings were conducted (Figure S3C), and KEGG analysis revealed a significant enrichment of the Hippo signaling pathway on the 14th day (Figure 5A). Due to ethical constraints in obtaining full‐thickness human esophageal specimens, particularly from patients with established strictures, we performed an exploratory analysis of YAP expression using available epithelial biopsies (Figures 5B and S4A). Although this approach has inherent limitations—notably the absence of lamina propria—it provided an initial observation of YAP upregulation in human ES tissues. Human tissue samples indicate that, on the 14th day after ESD, YAP levels are significantly elevated in the stenotic tissue compared to adjacent nonstenotic tissue. However, on the 5th day after ESD, YAP levels were normal. DPP4 promotes esophageal stricture (ES) via the Hippo‐YAP pathway. (A) Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis of ES rats and the rats with DPP4 inhibition on the 14th day. The top 20 enriched pathways are shown. The Hippo signaling pathway was significantly enriched. The data are sourced from KEGG, and we have obtained copyright permission from KEGG, ensuring full compliance with their guidelines. (B) The immunohistochemistry (IHC) of the YAP in stenotic site and 5 cm adjacent to the stenosis in the esophagus of seven patients who underwent circumferential esophageal endoscopic submucosal dissection (ESD) on postoperative days 5 and 14. Scale bar = 100 μm. (C) The fragments per kilobase of transcript per million mapped reads (FPKM) expression levels of Hippo pathway‐related genes (LATS1, TEAD4, TGFBR1, SMAD3) in ES rats and DPP4i‐treated rats at postoperative days 3, 7, 10, and 14. (D, E) The IHC of YAP in ES rats and rats with DPP4 inhibition on postoperative days 3, 7, 10, and 14. Scale bar = 100 μm. *p < 0.05, **p < 0.01, ***p < 0.001; error bars represent the standard deviation (SD). Further analysis revealed that genes associated with fibrosis within the Hippo signaling pathway—including LATS1, TEAD4, TGF‐βR1, and Smad3—exhibited time‐dependent upregulation in ES tissues. Notably, this upregulation was attenuated following DPP4 inhibition (Figure 5C). In addition, fibrotic and inflammatory markers exhibited a downregulation trend after the inhibition of DPP4 (Figure S4B–J). This suggests that DPP4 inhibitors may suppress the progression of inflammation, consistent with previous research findings. 29 , 30 In addition, a gradual increase in YAP1 expression in rats with ES was observed, and after DPP4 inhibition, YAP1 levels gradually decreased, showing a significant difference between days 10 and 14 (Figure 5D,E). These observations point to a close association between DPP4 and YAP accumulation in the context of ES. Specifically, DPP4 inhibition was accompanied by reduced YAP1 expression and nuclear translocation, suggesting that DPP4 may influence ES progression, at least in part, through its interaction with the Hippo‐YAP signaling axis. Validation in large animals We strictly adhered to clinical operating standards for performing ESD on Bama pigs, excising approximately 3 cm of circumferential mucosa. Postoperative examination after 3 weeks revealed the development of typical ES (Figures 6A and S5A). We observed a significant alleviation of ES in Bama pigs with the administration of an appropriate dose of the DPP4 inhibitor (2 mg/kg) (Figures 6A and S5B). However, excessive dosages (25 mg/kg) led to suboptimal mucosal healing (Figure S5C). Validation in large animals and retrospective analysis of human samples. (A) Endoscopy of normal pigs, esophageal stricture pigs, and pigs with DPP4 inhibition. (B) The hematoxylin–eosin (HE) staining of normal pigs, esophageal stricture pigs, and pigs with DPP4 inhibition (10×, scale bar = 100 μm). (C) The HE staining of normal pigs, esophageal stricture pigs, and pigs with DPP4 inhibition (40×, scale bar = 100 μm). (D) The Masson staining of normal pigs, esophageal stricture pigs, and pigs with DPP4 inhibition (10×, scale bar = 100 μm). (E) The Masson staining of normal pigs, esophageal stricture pigs, and pigs with DPP4 inhibition (40×, scale bar = 100 μm). (F–H) The immunofluorescence staining of the DPP4, α‐smooth muscle Actin (α‐SMA), and YAP in normal pigs, esophageal stricture pigs, and pigs with DPP4 inhibition. (I, J) The immunohistochemistry (IHC) of integrated optical density (IOD) of DPP4 and YAP in normal pigs, esophageal stricture pigs, and pigs with DPP4 inhibition. **p < 0.01, ***p < 0.001; error bars represent the standard deviation (SD). HE and Masson's analyses reveal that, akin to the rat model, the pig model of ES manifests a substantial accumulation of collagen in the submucosal layer. Moreover, there is a transition in the collagen structure from a typical mesh‐like distribution to a parallel arrangement. This alteration suggests noticeable scar formation and a reduction in esophageal elasticity. After treatment, there is a notable reduction in the quantity of esophageal collagen. Additionally, the collagen reverts to a mesh‐like distribution, suggesting the inhibition of scar formation (Figure 6B–E). Similar to the rat model, after ES, DPP4, YAP1, and the myofibroblast marker acta2 all show a significant increase. Inhibition of DPP4 results in a decrease in YAP1 and acta2, leading to the alleviation of ES (Figures 6F–J and S6A,B). To explore the potential clinical relevance of our experimental findings, we performed an exploratory retrospective analysis. These patients appeared to exhibit a lower incidence of ES compared to those who did not receive DPP4 inhibitors (Supplement 7; Table S3). DISCUSSION Benign ES is a challenging clinical condition characterized by excessive fibrotic repair following tissue injury. With the development of ESD techniques in recent years, the incidence of ES has been on the rise. 31 , 32 Symptoms of benign ES mainly manifest as dysphagia, aspiration, malnutrition, and even cachexia, causing significant distress to patients. A considerable proportion of refractory stricture patients cannot achieve long‐term relief and require repeated or even lifelong dilation treatments. 33 This significantly impacts patients' quality of life and contributes to economic and societal burdens. Nevertheless, the molecular processes governing fibrosis after esophageal injury remain poorly understood. Therefore, in‐depth research into the pathogenic mechanisms of fibrosis after esophageal injury holds significant theoretical and clinical implications for the diagnosis and treatment of benign ES (Figure 7). DPP4 regulation of Hippo‐YAP signaling reshapes the regenerative microenvironment in esophageal stricture (ES). Our sequencing results suggest that DPP4 is among the genes of interest in ES, warranting further functional investigation. DPP4 is a cell surface serine exopeptidase that cleaves X‐prolyl dipeptides from the N‐terminus of polypeptides. 34 , 35 In 2015, Rinkevich et al. 25 first reported that DPP4+ fibroblasts are the predominant cell population in skin fibrosis. In 2020, the Worthen et al. 36 discovered that DPP4+ fibroblasts in human skin tissue exhibit enhanced differentiation and secretion capabilities, producing abundant collagen and promoting skin scar formation. In the same year, Soare et al. 37 revealed that elevated DPP4 expression is a hallmark of fibroblast differentiation into myofibroblasts, and pharmacological or genetic interference inhibiting DPP4 can suppress TGF‐β1‐induced fibroblast‐to‐myofibroblast differentiation. Furthermore, in the liver fibrosis models, DPP4 inhibition was shown to reduce liver fibrosis by inhibiting the phenotypic transformation of hepatic stellate cells and reducing collagen synthesis. 38 In our study, DPP4 was significantly upregulated in ES patients and animal models. Inhibition of DPP4 in primary ES fibroblasts resulted in a significant decrease in proliferative activity, accompanied by a significant reduction in fibrosis‐related genes. In our experimental design, vildagliptin was administered prophylactically—starting 4 weeks prior to ESD and continuing for 2 weeks postoperatively—which significantly alleviated ES in rat models. It is important to note that this design tests preventive efficacy rather than therapeutic intervention on established strictures. Whether DPP4 inhibition confers therapeutic benefit when initiated after stricture formation remains unknown. The expression of myofibroblast marker (α‐SMA), TGF‐β, and collagen marker (COL1A1) was significantly reduced, leading to a substantial decrease in collagen deposition. Furthermore, in the study of the process of ES and relief, we found that after DPP4 inhibition, there was a certain alleviation of the inflammatory response, a significant reversal of the recruitment and proliferation of fibroblasts, and a reduction in collagen deposition, providing a pathological explanation for the mechanism of action of the drug. In summary, our findings are consistent with the possibility that DPP4 may serve as a characteristic biomarker of esophageal stricture and may contribute to its progression. However, definitive conclusions regarding causality require additional mechanistic studies. The differentiation of fibroblasts into myofibroblasts is the key effector way in fibrosis, and the activated myofibroblasts exhibit significantly enhanced abilities in proliferation, migration, and secretion of cytokines and extracellular matrix. Their contractile force contributes to physiological tissue remodeling, wound healing, and repair. However, excessive activation can lead to abnormal collagen deposition and scar formation, disrupting the residual normal tissue and altering its biochemical and biophysical microenvironment, causing dysregulation in the function of nearby normal cells. 39 The Hippo signaling pathway has been thoroughly characterized as a crucial mediator of sustained fibroblast‐to‐myofibroblast differentiation in fibrotic diseases. 40 Zhang et al. 41 suggested that the positive feedback through NUAK1 can activate YAP/TAZ signaling, promoting the fibroblast‐to‐myofibroblast differentiation and process of skin fibrosis. Haak et al. 42 proposed that inhibiting YAP/TAZ signaling with a fibroblast‐specific dopamine receptor D1 antagonist effectively alleviates pulmonary fibrosis, which inhibits the fibroblast‐to‐myofibroblast differentiation. Xiao et al. 43 found that the activation of the Hippo signaling pathway induces the transformation of cardiac fibroblasts into myofibroblasts. Lowering the levels of YAP/TAZ effectors in Lats1/2 mutant hearts can mitigate postinfarction fibrotic phenotypes. Consistent with these observations, our study identified a parallel association between DPP4 and the Hippo‐YAP axis in the context of esophageal fibrosis. In DPP4‐positive fibroblasts, enrichment of Hippo pathway–related genes was observed alongside elevated YAP expression. Pharmacological inhibition of DPP4 in primary fibroblasts corresponded with a significant reduction in YAP levels, coinciding with decreased fibroblast proliferative activity. It is important to note, however, that these correlative data do not distinguish whether DPP4 acts upstream of Hippo‐YAP, downstream, or via a parallel pathway. As such, we interpret the relationship as a concurrent phenomenon rather than a proven mechanistic cascade. In ES tissues, concurrent upregulation of DPP4 and YAP was noted, and this molecular profile was spatially and temporally correlated with the presence of fibroblast‐to‐myofibroblast differentiation. Sustained DPP4 inhibition in experimental models was accompanied by persistent downregulation of YAP, a marked reduction in the abundance of myofibroblasts, and a structural reorganization of collagen from a parallel‐aligned pattern characteristic of ES to a reticular mesh‐like architecture. Furthermore, DPP4 inhibition was associated with altered subcellular localization of YAP, characterized by increased cytoplasmic retention of phosphorylated YAP, which correlated with a scarless esophageal healing phenotype. Additionally, the inhibition of DPP4 is associated with YAP nuclear translocation, wherein phosphorylated YAP enters the cytoplasm, mediating a scarless esophageal environment, which aligns with the findings of Brewer et al. 44 Whether this association reflects a direct regulatory relationship remains to be determined. The translational relevance of our findings was further validated in a Bama miniature pig model, which anatomically and physiologically more closely resembles the human esophagus than rodents. This large animal model not only enables clinically relevant endoscopic assessment but also serves as a robust platform for evaluating therapeutic efficacy against refractory strictures. The consistent antifibrotic effects observed across both rodent and porcine models, under a prophylactic administration regimen, indicate that DPP4 inhibition can prevent the development of esophageal stricture when given prior to injury. These findings support DPP4 as a potential target for preventive intervention; however, they do not directly inform its efficacy as a treatment for established strictures. Furthermore, we conducted a safety assessment of the vildagliptin, which is a frontline antidiabetic medication in clinical practice. At therapeutic doses, vildagliptin does not significantly induce apoptosis in primary fibroblasts (Figure S8A). Additionally, therapeutic doses of vildagliptin have no apparent impact on liver and kidney function, as well as blood glucose and lipid levels in ES rats (Figure S8B). The pathology of the model pigs also showed normal liver and kidney function (Figure S8C). These findings collectively indicate that, at therapeutic doses, the drug does not exhibit notable hepatorenal or metabolic effects. However, we strongly advise caution in considering vildagliptin as a therapeutic intervention for ES in human patients. Any potential clinical application would require rigorous safety studies and thorough evaluation. In addition, YAP‐specific knockdown or overexpression experiments are shown in Figure S9. We recognize that our study did not experimentally dissect the hierarchical or interactive relationship between Hippo‐YAP and TGF‐β signaling in ES, nor did we evaluate the potential involvement of other fibrotic pathways. This represents an important limitation and a direction for future investigation. Studies employing pathway‐specific inhibitors, genetic ablation models, and comparative transcriptomic analyses will be necessary to fully elucidate the complex molecular circuitry driving esophageal fibrosis and to identify optimal therapeutic strategies. In addition, although we used vildagliptin as a pharmacological inhibitor of DPP4, pharmacological inhibition alone cannot establish DPP4‐specific action, particularly in the complex wound‐healing environment of esophageal stricture, which involves inflammation, fibroblast activation, and tissue remodeling. The observed therapeutic benefits may reflect indirect anti‐inflammatory or systemic effects rather than a direct antifibrotic mechanism mediated specifically by DPP4. Although siRNA‐mediated knockdown in primary fibroblasts partially addresses DPP4 specificity in vitro, in vivo evidence remains limited. Future studies employing genetic loss‐of‐function approaches (e.g., fibroblast‐specific Dpp4 knockout) or independent DPP4‐targeting strategies (e.g., neutralizing antibodies or alternative small‐molecule inhibitors with distinct scaffolds) are warranted to confirm target specificity. Besides, limitations include small sample sizes, a permissive DEG threshold, and the exploratory application of WGCNA and PPI on a modest dataset; thus, our findings are hypothesis‐generating and require validation in larger independent cohorts. In all, we found that DPP4 is identified as a contributor to the remodeling of the regenerative microenvironment in ES, and its functional changes occur in parallel with alterations in the activity of the Hippo‐YAP pathway. These parallel observations do not yet establish a direct causal or hierarchical link between DPP4 and Hippo‐YAP signaling, which awaits future mechanistic dissection. Importantly, the current evidence is limited to prophylactic administration of a DPP4 inhibitor prior to esophageal injury; therapeutic efficacy against established strictures has not been demonstrated. Future studies employing postinjury treatment initiation will be required to determine whether DPP4 inhibition has broader therapeutic applicability. AUTHOR CONTRIBUTIONS Rui Wu: Conceptualization; data curation; formal analysis; funding acquisition; investigation; methodology; project administration; resources; software; supervision. Chen Wang: Conceptualization; data curation; formal analysis; software; validation; visualization. Hui‐Min Tao: Conceptualization; formal analysis; funding acquisition; validation; visualization. Jin Huang: Conceptualization; data curation; project administration; resources; writing – original draft. Wei‐Qin Li: Conceptualization; data curation; formal analysis; methodology; project administration; resources; supervision; validation. Zhi‐Ning Fan: Conceptualization; data curation; formal analysis; methodology; project administration; resources; supervision; validation. FUNDING INFORMATION This work was supported by the Postgraduate Research & Practice Innovation Program of Jiangsu Province(SJCX25_0071), the key projects of the Changzhou Medical Center of Nanjing Medical University (CMCM202310), Eastern Theater Command General Hospital of the Chinese People's Liberation Army In‐Hospital Research Project (2024JCYJQN126). CONFLICT OF INTEREST STATEMENT The authors declare that they have no competing interests. ETHICS STATEMENT The study was approved by the Ethics Committee of Jiangsu Province Hospital. All experiments were performed under the supervision of the Ethics Committee of Nanjing Medical University in accordance with the relevant guidelines (approval number: IACUC‐2103014, Institutional Animal Care and Use Committee of Nanjing Medical University; approval number: 2024DZGJJ‐139, Ethics Committee of the Eastern Theater Command General Hospital). All methods are reported in accordance with ARRIVE guidelines. All participants provided written consent. Supporting information Supplementary Figure 1. Bioinformatic analysis of transcriptome sequencing. (A) The clustering analysis of transcriptome sequencing and genes and samples with analogous expression patterns are clustered. (B, C) The soft‐thresholding parameter of Weighted gene co‐expression network analysis (WGCNA) was chosen. (D) The heatmap of correlation analysis in all modules. (E) The protein–protein interaction (PPI) network analysis of the top 100 genes. (F, G) The Kyoto Encyclopedia of Genes and Genomes (KEGG) and Disease Ontology (DO) analyses of the top 100 genes. Supplementary Figure 2. Inhibition of DPP4 significantly alleviates esophageal stricture in rats. (A) The immunohistochemistry (IHC) of DPP4 in the normal rats, esophageal stricture rats, and rats with DPP4 inhibition. (B) The cholangioscopy examination of the normal rats, esophageal stricture rats, and rats with DPP4 inhibition. (C) The anatomical view of esophagus in the normal rats, esophageal stricture rats, and rats with DPP4 inhibition. (D) The Picrosirius red staining of the normal rats, esophageal stricture rats, and rats with DPP4 inhibition. (E) The IHC of α‐smooth muscle actin (α‐SMA) in the normal rats, esophageal stricture rats, and rats with DPP4 inhibition. (F–H) The integrated optical density (IOD) of COL1A1, tumor growth factor β (TGF‐β), and α‐SMA in the normal rats, esophageal stricture rats, and rats with DPP4 inhibition. *p < 0.05, **p < 0.01, ***p < 0.001; error bars represent the standard deviation (SD). Each group has seven replicates. Supplementary Figure 3. Progress of esophageal stricture and its treatments in rats. (A) The anatomical view of esophagus in esophageal stricture rats and rats with DPP4 inhibition on postoperative days 3, 7, 10, and 14. (B) The Picrosirius red staining in esophageal stricture rats and rats with DPP4 inhibition on postoperative days 3, 7, 10, and 14. (C) The transcriptome sequencings in esophageal stricture rats and rats with DPP4 inhibition on postoperative days 3, 7, 10, and 14. Error bars represent the standard deviation (SD). Each group has three replicates at each time point. Supplementary Figure 4. Inhibition of DPP4 downregulates the genes associated with Hippo pathway, fibrosis, and inflammation. (A) The YAP expression in stenotic site/5 cm adjacent to the stenosis was assessed on postoperative days 5 and 14. (B–F) The fragments per kilobase of transcript per million mapped reads (FPKM) expression level of integrin β1, tumor necrosis factor (TNF), CXCR4, CD80, and CD68 in esophageal stricture rats and rats with DPP4 inhibition on postoperative days 3, 7, 10, and 14. (G–J) The immunohistochemistry (IHC) of integrin β1 and integrin β3 in esophageal stricture rats and rats with DPP4 inhibition on postoperative days 3, 7, 10, and 14. Error bars represent the standard deviation (SD). Each group has five replicates at each time point. Supplementary Figure 5. Postoperative anatomical view of the Bama pigs at 3 weeks. (A) After removal of circumferential 3‐cm esophageal mucosa in Bama pigs, significant esophageal stricture developed 3 weeks postoperatively. (B) After continuous oral administration of DPP4 inhibitor (2 mg/kg) from 4 weeks before surgery to 3 weeks after surgery, esophageal stricture was significantly alleviated. (C) After continuous overdosed oral administration of DPP4 inhibitor (25 mg/kg) from 4 weeks before surgery to 3 weeks after surgery, poor healing of the esophageal mucosa occurred. Supplementary Figure 6. Esophageal stricture surgery and postoperative period in Bama pigs. (A) We strictly adhered to clinical operating standards for performing endoscopic submucosal dissection (ESD) on Bama pigs, including localization, creating an esophageal tunnel, mucosal resection, and securing the specimen. (B) The immunohistochemistry (IHC) of the DPP4 and YAP1 in normal pigs, esophageal stricture pigs (5× and 30×), and pigs with DPP4 inhibition (5× and 30×). Supplementary Figure 7. The patients who underwent esophageal circumferential endoscopic submucosal dissection (ESD) and were treated with DPP4 inhibitors clinically. Supplementary Figure 8. Safety assessment of the vildagliptin. (A) The apoptosis assay in primary fibroblasts was conducted with different doses of vildagliptin. (B) Biochemical tests, including liver and kidney function, as well as blood glucose and lipid levels, were monitored in both pre‐ and postmedicine rats. (C) The pathology of liver and kidney was conducted in the normal and treatment pigs. Error bars represent the standard deviation (SD). Each group has three replicates. Supplementary Figure 9. Changes in fibrosis‐related genes following YAP‐specific knockout and overexpression. (A) Changes in COL1A1 following YAP knockout and overexpression. (B) Changes in α‐SMA following YAP knockout and overexpression. Supplementary Table 1. The human esophageal tissue in the sequencing set and the validation set. Supplementary Table 2. The esophageal tissue of seven patients who underwent circumferential esophageal endoscopic submucosal dissection (ESD) on postoperative days 5 and 14. Supplementary Table 3. The esophageal tissue of 16 patients who underwent circumferential esophageal endoscopic submucosal dissection (ESD). ACKNOWLEDGMENTS The authors have nothing to report. Wu R, Wang C, Tao H‐M, Huang J, Li W‐Q, Fan Z‐N. Association of DPP4 with esophageal stricture progression and the Hippo‐YAP pathway. Anim Models Exp Med. 2026;00:1‐15. doi: 10.1002/ame2.70239 Rui Wu and Chen Wang are considered as first authors. Contributor Information Jin Huang, Email: hj042153@hotmail.com. Wei‐Qin Li, Email: ctgchina@medbit.cn. Zhi‐Ning Fan, Email: fanzhining@njmu.edu.cn. DATA AVAILABILITY STATEMENT The datasets generated and/or analyzed during the current study are available in the OMIX database, China National Center for Bioinformation (https://ngdc.cncb.ac.cn/omix/preview/h1fH4IWI, accession no. OMIX009325). REFERENCES Associated Data Supplementary Materials Supplementary Figure 1. Bioinformatic analysis of transcriptome sequencing. (A) The clustering analysis of transcriptome sequencing and genes and samples with analogous expression patterns are clustered. (B, C) The soft‐thresholding parameter of Weighted gene co‐expression network analysis (WGCNA) was chosen. (D) The heatmap of correlation analysis in all modules. (E) The protein–protein interaction (PPI) network analysis of the top 100 genes. (F, G) The Kyoto Encyclopedia of Genes and Genomes (KEGG) and Disease Ontology (DO) analyses of the top 100 genes. Supplementary Figure 2. Inhibition of DPP4 significantly alleviates esophageal stricture in rats. (A) The immunohistochemistry (IHC) of DPP4 in the normal rats, esophageal stricture rats, and rats with DPP4 inhibition. (B) The cholangioscopy examination of the normal rats, esophageal stricture rats, and rats with DPP4 inhibition. (C) The anatomical view of esophagus in the normal rats, esophageal stricture rats, and rats with DPP4 inhibition. (D) The Picrosirius red staining of the normal rats, esophageal stricture rats, and rats with DPP4 inhibition. (E) The IHC of α‐smooth muscle actin (α‐SMA) in the normal rats, esophageal stricture rats, and rats with DPP4 inhibition. (F–H) The integrated optical density (IOD) of COL1A1, tumor growth factor β (TGF‐β), and α‐SMA in the normal rats, esophageal stricture rats, and rats with DPP4 inhibition. *p < 0.05, **p < 0.01, ***p < 0.001; error bars represent the standard deviation (SD). Each group has seven replicates. Supplementary Figure 3. Progress of esophageal stricture and its treatments in rats. (A) The anatomical view of esophagus in esophageal stricture rats and rats with DPP4 inhibition on postoperative days 3, 7, 10, and 14. (B) The Picrosirius red staining in esophageal stricture rats and rats with DPP4 inhibition on postoperative days 3, 7, 10, and 14. (C) The transcriptome sequencings in esophageal stricture rats and rats with DPP4 inhibition on postoperative days 3, 7, 10, and 14. Error bars represent the standard deviation (SD). Each group has three replicates at each time point. Supplementary Figure 4. Inhibition of DPP4 downregulates the genes associated with Hippo pathway, fibrosis, and inflammation. (A) The YAP expression in stenotic site/5 cm adjacent to the stenosis was assessed on postoperative days 5 and 14. (B–F) The fragments per kilobase of transcript per million mapped reads (FPKM) expression level of integrin β1, tumor necrosis factor (TNF), CXCR4, CD80, and CD68 in esophageal stricture rats and rats with DPP4 inhibition on postoperative days 3, 7, 10, and 14. (G–J) The immunohistochemistry (IHC) of integrin β1 and integrin β3 in esophageal stricture rats and rats with DPP4 inhibition on postoperative days 3, 7, 10, and 14. Error bars represent the standard deviation (SD). Each group has five replicates at each time point. Supplementary Figure 5. Postoperative anatomical view of the Bama pigs at 3 weeks. (A) After removal of circumferential 3‐cm esophageal mucosa in Bama pigs, significant esophageal stricture developed 3 weeks postoperatively. (B) After continuous oral administration of DPP4 inhibitor (2 mg/kg) from 4 weeks before surgery to 3 weeks after surgery, esophageal stricture was significantly alleviated. (C) After continuous overdosed oral administration of DPP4 inhibitor (25 mg/kg) from 4 weeks before surgery to 3 weeks after surgery, poor healing of the esophageal mucosa occurred. Supplementary Figure 6. Esophageal stricture surgery and postoperative period in Bama pigs. (A) We strictly adhered to clinical operating standards for performing endoscopic submucosal dissection (ESD) on Bama pigs, including localization, creating an esophageal tunnel, mucosal resection, and securing the specimen. (B) The immunohistochemistry (IHC) of the DPP4 and YAP1 in normal pigs, esophageal stricture pigs (5× and 30×), and pigs with DPP4 inhibition (5× and 30×). Supplementary Figure 7. The patients who underwent esophageal circumferential endoscopic submucosal dissection (ESD) and were treated with DPP4 inhibitors clinically. Supplementary Figure 8. Safety assessment of the vildagliptin. (A) The apoptosis assay in primary fibroblasts was conducted with different doses of vildagliptin. (B) Biochemical tests, including liver and kidney function, as well as blood glucose and lipid levels, were monitored in both pre‐ and postmedicine rats. (C) The pathology of liver and kidney was conducted in the normal and treatment pigs. Error bars represent the standard deviation (SD). Each group has three replicates. Supplementary Figure 9. Changes in fibrosis‐related genes following YAP‐specific knockout and overexpression. (A) Changes in COL1A1 following YAP knockout and overexpression. (B) Changes in α‐SMA following YAP knockout and overexpression. Supplementary Table 1. The human esophageal tissue in the sequencing set and the validation set. Supplementary Table 2. The esophageal tissue of seven patients who underwent circumferential esophageal endoscopic submucosal dissection (ESD) on postoperative days 5 and 14. Supplementary Table 3. The esophageal tissue of 16 patients who underwent circumferential esophageal endoscopic submucosal dissection (ESD). Data Availability Statement The datasets generated and/or analyzed during the current study are available in the OMIX database, China National Center for Bioinformation (https://ngdc.cncb.ac.cn/omix/preview/h1fH4IWI, accession no. OMIX009325).