Gastric cancer is the fifth most common malignancy and the fourth leading cause of cancer-related mortality globally.¹ Although the cure rate for early gastric cancer is more than 90%,² the 5-year survival rate for progressive gastric cancer is below 30% even after surgery.³ Understanding the mechanism coordinating gastric cancer formation is important for early detection of the disease. According to the classical theory of Correa cascade,⁴ gastric cancer development involves multiple stages, i.e., originating from healthy gastric mucosa through superficial gastritis, atrophic gastritis, intestinal metaplasia (IM), heterogenous hyperplasia to malignant disease. IM, a precancerous lesion, predicts a high risk of gastric cancer.^5,6 Although histopathological features of IM are extensively studied, the critical factors controlling IM pathogenesis remain uncertain.

The GATA family consists of six members (GATA binding protein 1 [GATA1] through GATA6), which are a class of zinc-finger transcription factors. They play different roles in embryonic development, cell differentiation, and carcinogenesis.^7,8 GATA4, an intestinal transcription factor, is a key factor necessary for endoderm development and intestinal epithelial renewal during embryogenesis.^9,10 GATA4 is mainly distributed in the proximal differentiated epithelial cells of the gastrointestinal tract, but absent in the distal ones.¹¹ GATA4 expression is increased in Barrett's esophagus, IM, and proliferative neuroendocrine cells.¹²

The mucin gene family mainly includes seven members, i.e., MUC1, MUC2, MUC3, MUC4, MUC5A, MUC5B, and MUC6.¹² They confer a protective effect on the mucosal surface.¹³ Among them, MUC2 is specifically expressed in intestinal epithelial cupped cells.¹⁴ MUC2 expression increases significantly when IM occurs, which is manifested as transformation from columnar epithelium to intestinal epithelium.¹⁵ During embryonic development, the expression of MUC2 is regulated by transcription factors associated with intestinal differentiation including GATA4.^16,17 Biochemical studies indicate that GATA4 has the ability to transactivate mouse MUC2 gene through binding to the MUC2 promoter.¹⁸

IM is causally associated with duodenogastric reflux (DGR). Bile acids, a major component of DGR, play an important role in IM formation.^19,20 Chenodeoxycholic acid (CDCA) is commonly used to generate cell models of gastric IM (GIM).^21-23 Caudal-related homeobox transcription factor 2 (CDX2) has been found to regulate intestinal cell growth and differentiation²⁴ and the expression of intestinal markers such as MUC2.²⁵ A recent study has reported that bile acids can upregulate CDX2 and MUC2 expression in normal gastric epithelial cells through activating the farnesoid X receptor (FXR)/nuclear transcription factor-κB (NF-κB) signaling pathway.²³ However, the expression and regulation of GATA4, CDX2, and MUC2 in GIM remain unclear.

In this study, we evaluated the expression of GATA4, CDX2, and MUC2 in a CDCA-induced cell model and clinical specimens of GIM and determined the mechanisms involved in their expression regulation.

1. Cell culture and treatment

Normal human gastric epithelial cells (GES-1) and gastric cancer cells (AGS) were acquired from the American Type Culture Collection (Manassas, VA, USA). GES-1 cells were cultured in Dulbecco’s Modified Eagle Medium (Bioagrio, Mountain View, CA, USA) supplemented with 10% fetal bovine serum (Bioagrio) and AGS cells in RPMI 1640 Medium (Bioagrio) with 10% fetal bovine serum.

GES-1 cells at 50% to 70% confluence were serum starved for 24 hours before being exposed to various concentrations of CDCA (Sigma-Aldrich, Darmstadt, Germany) for an additional 24 hours in order to produce CDCA-induced GIM. For inhibition of NF-κB signaling, CDCA treatment was combined with the NF-κB inhibitor pyrrolidine dithiocarbamate (Selleck, Houston, TX, USA; 50 µM) in GES-1 cells. GES-1 cells were given 24 hours of treatment with 15 µg/µL of betulinic acid (MCE, Shanghai, China) to activate NF-κB signaling.

2. Clinical specimens

To investigate the expression of GATA4 and MUC2 in IM specimens, 40 paraffin-embedded gastric IM tissues and 40 paraffin-embedded gastritis tissues were obtained from patients (45 males and 35 females; age, 30 to 84 years) who had undergone endoscopic biopsy at the Pathology Department of Chongqing Emergency Medical Center (Chongqing, China) between 2020 and 2021. Based on the percentage of the stomach glands being superseded by metaplastic tissues, gastric IM specimens were divided into mild IM, middle IM, and severe IM. Additionally, 10 human gastric IM tissues and 10 gastritis tissues were collected from the Gastrointestinal Endoscopy Room of Chongqing Emergency Medical Center (Chongqing, China). All tissues were verified by at least two pathological experts (Li Peng and Jing Chen) based on hematoxylin and eosin staining results. All of the patients signed the informed consent forms. Our research was granted approval by the Institutional Ethics Committee of Chongqing Emergency Medical Center (Ethics Review Committee Clinical Trial Approval No. 46, 2022).

3. Total RNA extraction and qRT-PCR

Total RNA from cell lines and tissue was extracted using RNAiso Plus reagent (Takara, Kusatsu, Japan) following the manufacturer’s guidelines. RNA was reverse-transcribed into complementary DNA using PrimeScript^TM RT reagent kit with gDNA Eraser (Takara). Quantitative real-time polymerase chain reaction (qRT-PCR) was conducted using TB Green Premix Ex Taq II (Takara) in a final volume of 10 µL on a CFX96^TM Real-Time PCR Detection system (Bio-Rad Laboratories, Hercules, CA, USA). The target messenger RNA expression was calculated using the 2^-∆∆Ct method and normalized to glyceraldehyde 3-phosphate dehydrogenase. The sequences of PCR primers are shown in Supplementary Table 1.

4. Total protein extraction and Western blotting analysis

Total protein was extracted using radioimmunoprecipitation assay buffer comprising protease and phosphatase inhibitors (Beyotime Biotechnology, Shanghai, China). Nuclear and cytoplasmic proteins were extracted with Nuclear and Cytoplasmic Protein Extraction Kit (KeyGen Biotech, Nanjing, China). Proteins were quantified using the bicinchoninic acid method based on the manufacturer’s agreement (Beyotime Biotechnology). The following primary antibodies were used in this study: anti-CDX2 (1:1,000; Cell Signaling Technology, #12306), anti-MUC2 (1:1,000; Abcam, ab134119), anti-KLF4 (1:1,000; Abcam, ab215036), anti-GATA4 (1:100; Santa Cruz Biotechnology, sc-25310), anti-NF-κB p105/p50 (1:1,000; Abcam, ab32360), anti-NF-κB p65 (1:1,000; Cell Signaling Technology, #4764), anti-phospho-NF-κB p50 (1:1,000; Abmart, PA1798), anti-phospho-NF-κB p65 (1:1,000; Cell Signaling Technology, #3033) and anti-Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (1:1,000; ZSGB-BIO, TA08). Goat anti-rabbit IgG and anti-mouse IgG (Biosharp) was used as secondary antibodies. Protein bands were visualized using chemifluorescence.

5. Immunohistochemistry

Formalin‐fixed paraffin‐embedded tissue slices were processed for deparaffination and rehydration. Slices were then processed with primary antibodies against human GATA4 (1:100; Santa Cruz Biotechnology), human MUC2 (1:15,000; Abcam), and mouse Gata4 (1:100; Abmart, T56814). Following treatment with secondary antibodies conjugated to horseradish peroxidase (Dako, Glostrup, Denmark) for 30 min, the slices were developed with 3,3’-diaminobenzidine solution (Dako). Results were scored independently by at least two professional pathologists. The immunoreactivity score was computed by multiplying the intensity of the staining by the proportion of positive cells. The intensity of the staining was measured as indicated below: 0, negative; 1, weak; 2, moderate; 3, strong. The positive rate of staining was scored as follows: 0, <10%; 1, 10% to 25%; 2, 26% to 50%; 3, 51% to 75%; 4, >75%.

6. Immunofluorescence staining

GES-1 cells treated with or without CDCA were seeded in 24-well plates and immunofluorescence staining was performed after 24 hours. After fixation, permeability and sealing, cell was incubated with a primary anti-NF-κB p65 (1:200; Cell Signaling Technology, #4764) overnight at 4°C. After washing with phosphate-buffered saline, it was incubated with secondary antibody, and then nucleated with DAPI (Beyotime). Confocal laser-scanning microscope (Leica TCS SP8, Wetzlar, Germany) was used to capture the images.

7. Cell transfection

Lipofectamine 3000 (Invitrogen, Waltham, MA, USA) was used to transfect cells at 50% to 70% confluence with the relevant constructs following the manufacturer’s procedure. After 48 to 72 hours, transfected cells were collected for gene expression analysis. GATA4 and CDX2 overexpression plasmids were purchased from Youbio (Changsha, China). Small interfering RNAs targeting GATA4, CDX2, and p65 were purchased from Oligobio (Beijing, China).

8. Luciferase reporter assay

PCR was used to acquire segments of the human MUC2, CDX2, and GATA4 promoters, which were then inserted into the pGL3-basic vector (Oligobio). Lipofectamine 3000 transfection reagent was used to transfect HEK293 cells planted in 24-well plates with the reporter plasmids. To test transfection effectiveness, a Renilla luciferase-expressing plasmid was cotransfected. After 36 hours, cells were collected and lysed to perform luciferase activity detection with the Dual-Luciferase Reporter Assay Kit (Promega, Madison, WI, USA).

9. Chromatin immunoprecipitation

The cells were cross-linked with 4% formaldehyde for 10 minutes. DNA was sheared by sonication. Sonicated samples were processed with anti-GATA4 antibody (Santa Cruz Biotechnology) or control immunoglobulin G. Protein G beads were used to capture the immunocomplexes. DNA was extracted and analyzed by qRT-PCR analysis. The sequences of qRT-PCR primers are shown in Supplementary Table 1.

10. Animal model

An animal model of GIM was established as described previously.²³ Male C57BL/6J mice (age, 7 weeks; weight, ~20 g) were acquired from GemPharmatech (Jiangsu, China) and fed in the Laboratory Animal Center of Chongqing Medical University. The mice received administration of CDCA (10 mM) or phosphate-buffered saline to the stomach through a plastic feeding tube for 50 days, 2 times per day. Each treatment group had six mice. The mice were sacrificed, then gastric mucosa tissues were obtained for gene expression detection. The animal experimental procedures received approval from the Chongqing Medical University Animal Care and Use Committee (Ethics Review Committee Clinical Trial Approval No. 46, 2022).

11. Statistical analysis

All statistical analyses were conducted using the SPSS software version 19.0 (IBM Corp., Armonk, NY, USA). The data are presented as the mean±standard error of the mean. The differences between the two groups were compared using the Student t-test. The relationship between GATA4 and MUC2 protein expression levels was evaluated using linear correlation coefficients (Pearson). When p-values were below 0.05, the results were regarded as statistically significant.

1. GATA4 expression is upregulated in CDCA-induced gastric epithelial cells and gastric IM tissues

Analysis of data from The Cancer Genome Atlas (https://portal.gdc.cancer.gov/) revealed that GATA4 was expressed at higher levels in gastric cancer than in normal gastric tissues (Supplementary Fig. 1A and B). A previous study has shown that GATA4 is also elevated in IM.¹¹ These results suggest that GATA4 may play an important role in gastric pathogenesis. CDCA has been reported to increase the expression of IM markers in GES-1 cells.^23,26,27 Consistently, our data showed that CDCA treatment of GES-1 cells upregulated intestinal markers CDX2, KLF4, and MUC2 in a dose-dependent manner (Fig. 1A). Western blotting results validated the upregulation of these intestinal markers by CDCA (Fig. 1B). Most interestingly, CDCA treatment also resulted in a dose-dependent increase in both the messenger RNA and protein levels of GATA4 in GES-1 cells (Fig. 1C and D). Next, we detected the expression of GATA4 in clinical gastritis and IM specimens. It was revealed that the messenger RNA and protein levels of GATA4 were remarkably increased in IM tissues compared to gastritis tissues (Fig. 1E and F).

Figure 1. GATA4 expression is upregulated in CDCA-induced gastric IM cell models and gastric IM tissues. (A) Quantitative real-time polymerase chain reaction was used to detect CDX2, KLF4, and MUC2 mRNA expression. (B) Western blotting analysis of CDX2, KLF4, and MUC2 protein expression. (C) The mRNA and (D) protein levels of GATA4 were detected in GES-1 cells treated with or without or CDCA. (E) The mRNA and (F) protein expression of GATA4 was measured in clinical specimen tissues. Data are shown as mean±standard error of the mean (n=3). DMSO, Dimethyl sulfoxide; CDX2, caudal-related homeobox 2; KLF4, kruppel-like factor 4; MUC2, mucin 2; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; GATA4, GATA binding protein 4; CDCA, chenodeoxycholic acid; IM, intestinal metaplasia; mRNA, messenger RNA; GES-1, gastric epithelial cell; NS, not significant. *p<0.05; ^†p<0.01; ^‡p<0.001.

2. Correlation between GATA4 and MUC2 in gastric IM tissues

In order to elucidate the relationship between GATA4 and MUC2 in gastric IM, we detected GATA4 and MUC2 expression in 40 paraffin-embedded gastric IM tissues and 40 paraffin-embedded gastritis tissues by immunohistochemistry. Representative GATA4 and MUC2 staining images in gastritis and gastric IM tissues are shown in Fig. 2A and Supplementary Fig. 1C. Gastric IM tissues had higher immunoreactivity scores for GATA4 and MUC2 than gastritis tissues (Fig. 2B and C). Moreover, both GATA4 and MUC2 immunoreactivity scores increased with the severity of IM (Fig. 2D and E). Additionally, there was a significant, positive relationship between the expression of GATA4 and MUC2 in gastric IM tissues (p<0.01) (Fig. 2F).

Figure 2. Correlation between GATA4 and MUC2 in GIM tissues. (A) Immunohistochemistry staining of GATA4 and MUC2 in gastritis and GIM tissues. Scale bar, 50 µm. (B, C) GIM tissues had higher IRS scores for (B) GATA4 and (C) MUC2 than gastritis tissues. (D) GATA4 and (E) MUC2 IRS scores increased with the severity of IM. (F) Correlation between GATA4 and MUC2 in GIM tissues. GATA4, GATA binding protein 4; MUC2, mucin 2; GIM, gastric intestinal metaplasia; IRS, immunoreactivity score; IM, intestinal metaplasia. *p<0.05; ^†p<0.01; ^‡p<0.001.

3. CDCA induces the expression of GATA4 and MUC2 through activation of NF-^κB signaling

Having identified the positive relationship between GATA4 and MUC2 in gastric IM samples, we tested whether GATA4 could transactivate MUC2. When GATA4 was overexpressed, MUC2 expression was markedly enhanced in GES-1 cells (Fig. 3A). On the contrary, MUC2 expression was significantly reduced in AGS gastric epithelial cells when GATA4 was silenced by specific GATA4-targeting small interfering RNAs (Fig. 3B). In a cell model of CDCA-induced GIM, knockdown of GATA4 also inhibited the expression of MUC2 (Fig. 3C). Based on JASPAR program (https://jaspar.genereg.net/), we predicted putative binding sites for GATA4 in the MUC2 promoter (Fig. 3D). Luciferase reporter gene analysis demonstrated that GATA4 overexpression significantly increased the expression of luciferase reporters harboring different MUC2 promoter regions, especially the region (–1000/–505) (Fig. 3E). Chromatin immunoprecipitation assay further confirmed GATA4 binding to the MUC2 promoter, which was strengthened by CDCA treatment (Fig. 3F and G).

Figure 3. NF-κB is involved in CDCA-induced upregulation of GATA4 and MUC2. (A) Western blotting analysis of MUC2 protein levels in GES-1 cells overexpressing GATA4. (B) Western blotting analysis of MUC2 protein levels in AGS cells transfected with indicated small interfering RNAs. (C) Knockdown of GATA4 inhibited CDCA-induced upregulation of MUC2. (D) Based on the JASPAR program, putative binding sites for GATA4 in the MUC2 promoter were predicted. (E) Luciferase reporter assay. GATA4 overexpression increased the expression of luciferase reporters harboring different MUC2 promoter regions. (F, G) Chromatin immunoprecipitation assay showing GATA4 binding to the MUC2 promoter. (H) Analysis of p50, p65, phosphorylated p50 (p-p50) and phosphorylated p65 (p-p65) protein levels in CDCA-treated GES-1 cells. (I) Immunofluorescence staining for p65 (red). Cell nuclei were counterstained with DAPI (blue). Scar bar: 20 µm. (J) Western blotting analysis showed increased protein levels of nuclear p50 and p65 in CDCA-treated GES-1 cells. (K) Knockdown of p65 inhibited CDCA-induced upregulation of GATA4. (L) Betulinic acid enhanced the expression of GATA4 and MUC2. (M) Inhibition of NF-κB using pyrrolidine dithiocarbamate diminished CDCA-induced GATA4 and MUC2 protein expression. Data are shown as mean±standard error of the mean (n=3). NF-κB, nuclear transcription factor-κB; CDCA, chenodeoxycholic acid; GATA4, GATA binding protein 4; MUC2, mucin 2; DMSO, Dimethyl sulfoxide; GES-1, gastric epithelial cell; AGS, human gastric adenocarcinoma cell line; NC, negative control; siRNA, small interfering RNA; M, marker; IgG, immunoglobulin G; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; p65, NF-κB p65; p50, NF-κB p50; p-p65, phospho-NF-κB p65; p-p50, phospho-NF-κB p50; BetA, betulinic acid; PDTC, pyrrolidine dithiocarbamate. *p<0.01; ^†p<0.001.

As bile acid-induced NF-κB signaling plays an essential role in gastric carcinogenesis,^28,29 we checked whether NF-κB signaling could be implicated in CDCA-induced upregulation of GATA4 and MUC2. In agreement with a previous study,²³ CDCA treatment of GES-1 cells led to an increase in total and phosphorylated p50 and p65 protein levels (Fig. 3H). Next, we examined the nuclear p50 and p65 distribution. CDCA treatment induced the nuclear translocation of p50 and p65 (Fig. 3I and J). In a cell model of CDCA-induced GIM, knockdown of p65 inhibited the expression of GATA4 (Fig. 3K). Similarly, betulinic acid, a potent activator of NF-κB signaling,³⁰ promoted the expression of GATA4 and MUC2 (Fig. 3L). In addition, an NF-κB inhibitor pyrrolidine dithiocarbamate diminished CDCA-induced GATA4 and MUC2 protein expression (Fig. 3M). These results suggest that NF-κB signaling is implicated in CDCA-induced expression of GATA4 and MUC2.

4. GATA4 and CDX2 can regulate each other in CDCA-induced GIM

It has been reported that gastric IM can be promoted by bile acids through increasing CDX2 and MUC2 expression via the FXR/NF-κB signaling pathway.²³ Since our data showed that MUC2 can be transactivated by GATA4, we sought to investigate the relationship between CDX2 and GATA4. As shown in Fig. 4A, overexpression of GATA4 induced CDX2 protein expression in GES-1 cells. Depletion of GATA4 suppressed CDX2 expression in AGS cells (Fig. 4B) and attenuated CDCA-induced CDX2 expression (Fig. 4C). When CDX2 was overexpressed, the protein expression of GATA4 was potentiated (Fig. 4D). Moreover, silencing of CDX2 blocked the expression of GATA4 in AGS cells (Fig. 4E) and CDCA-treated GES-1 cells (Fig. 4F). It was predicated that the CDX2 promoter had potential GATA4 binding sites and the GATA4 promoter had CDX2 binding sites (Fig. 4G). Luciferase reporter assays confirmed that GATA4 induced the expression of CDX2 promoter-driven reporters (Fig. 4H) and CDX2 stimulated the expression of GATA4 promoter-driven reporters (Fig. 4I). Taken together, these results suggest that GATA4 and CDX2 can positively regulate each other in CDCA-induced GIM.

Figure 4. GATA4 and CDX2 reciprocally regulate each other in CDCA-induced GIM. (A) Western blotting analysis of CDX2 expression in GES-1 cells overexpressing GATA4. (B) Western blotting analysis of CDX2 expression in AGS cells transfected with indicated siRNAs. (C) Depletion of GATA4 blocked CDCA-induced CDX2 expression in GES-1 cells. (D) Western blotting analysis of GATA4 expression in GES-1 cells overexpressing CDX2. (E) Western blotting analysis of GATA4 expression in AGS cells transfected with indicated siRNAs. (F) Depletion of CDX2 blocked CDCA-induced GATA4 expression in GES-1 cells. (G) Based on the JASPAR program, the CDX2 promoter was predicted to have potential GATA4 binding sites, while the GATA4 promoter was predicted to contain CDX2 binding sites. (H) Luciferase reporter assays showing that GATA4 induced the expression of CDX2 promoter-driven reporters. (I) Luciferase reporter assays showing that CDX2 stimulated the expression of GATA4 promoter-driven reporters. Data are shown as mean±standard error of the mean (n=3). GATA4, GATA binding protein 4; CDX2, caudal-related homeobox 2; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; CDCA, chenodeoxycholic acid; GIM, gastric intestinal metaplasia; GES-1, gastric epithelial cell; AGS, human gastric adenocarcinoma cell line; siRNA, small interfering RNA; NC, negative control. *p<0.01; ^†p<0.001.

5. Bile acids induce alterations of molecules associated with premalignant lesions and GATA4 in the mouse gastric mucosa

To further study the influence of bile acids on IM formation in the stomach, an animal model of DGR was established by injecting bile acids into the stomach of mice.²³ Immunohistochemistry showed that GATA4 expression was increased in the stomach mucosa from mice treated with CDCA, compared to those treated with phosphate-buffered saline alone (Fig. 5A). Western blotting analysis further demonstrated that the protein levels of MUC2, CDX2, GATA4, p50, and p65 were markedly elevated in the stomach mucosa of mice exposed to CDCA (Fig. 5B-G). Animal studies suggest that gastric IM formation involves alteration of GATA4, CDX2, and MUC2.

Figure 5. Bile acids induce alterations in molecules associated with premalignant lesions and GATA4 in the mouse gastric mucosa. (A) Immunohistochemistry staining of GATA4 in gastric tissues collected from mice treated with or without CDCA. Scale bar, 50 µm. (B-G) Western blotting analysis of indicated proteins in the gastric mucosa collected from mice treated with or without CDCA. GATA4, GATA binding protein 4; CDCA, chenodeoxycholic acid; MUC2, mucin 2; CDX2, caudal-related homeobox 2; p50, NF-κB p50; p65, NF-κB p65; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; PBS, phosphate-buffered saline. *p<0.01; ^†p<0.001.

In this study, our data show that reciprocal transactivation between GATA4 and CDX2 occurs in CDCA-induced GIM. NF-κB signaling is implicated in the triggering of GATA4 and CDX2 upregulation. Both GATA4 and CDX2 promote the expression of MUC2, consequently facilitating the formation of GIM.

DGR is a critical risk factor for gastric cancer.³¹ Massive production of bile acids, a hallmark feature of DGR, is associated with the development of tumors at the esophagus, pharynx, stomach, small intestine, and colon.³² There is evidence that exposure to bile acids can lead to Barrett's esophagus^33,34 and GIM.¹⁹ A multicenter, large-scale cross-sectional study from Japan³⁵ revealed that a high concentration of bile acids is associated with an elevated risk of GIM in the stomach. These results suggest that bile acids play an essential role in the formation of GIM. Therefore, in this study, we establish a GIM cell model by treating gastric epithelial cells with CDCA.

It has been suggested that GATA4 may be associated with gastric carcinogenesis.³⁶ GATA4 can interact with KLF5 and GATA6 to promote gastric cancer progression.³⁷ GATA4 is also involved in intestinal epithelial morphogenesis and differentiation.³⁸ Given the biological function of GATA4 in regulating epithelial development and malignant transformation, we hypothesized that GATA4 might play a pivotal role in GIM. In agreement with this hypothesis, our data showed that GATA4 is significantly increased in the CDCA-induced cell model of GIM. It has been documented that MUC2 is transcriptionally regulated by GATA4 transcription factor in the mouse intestine.¹⁸ Analysis of the MUC2 promoter reveals that within the proximal promoter region there are binding sites for multiple transcription factors including GATA4 and CDX2.³⁹ Our data show that GATA4 is enriched at the promoter of MUC2 and can trigger the transcription of MUC2 promoter-driven luciferase reporter. Consistent with the induction of GATA4, CDCA-treated gastric epithelial cells also exhibit a significant elevation of MUC2. Most interestingly, GATA4 expression was increased in IM tissues compared to gastritis tissues. Moreover, GATA4 expression is positively correlated with MUC2 expression in GIM tissues. Taken together, these findings suggest GATA4 may participate in GIM through transactivation of MUC2. Bile acid, as a signal regulator, usually regulates downstream signal transduction by activating FXR and G-protein-coupled bile acid receptor 1.⁴⁰ In recent years, more and more studies have proved that FXR plays a pivotal role in the formation of IM induced by bile acid.^23,27,41 This study shows that GATA4 expression is increased in bile acid-induced GIM. Therefore, we speculate that there is a potential regulatory relationship between FXR and GATA4. This issue will be addressed in future work.

NF-κB signaling has important roles in inflammation, innate immunity, cell proliferation, apoptosis, and differentiation.^42,43 Consistent with previous studies,²³ our data demonstrate that NF-κB is activated in CDCA-induced GIM. It has been reported that GATA4 can induce mesenchymal-epithelial transformation and cellular senescence in hepatocellular carcinoma by the NF-κB signaling pathway.⁴⁴ In the current study, we show that NF-κB signaling is involved in the upregulation of GATA4 and MUC2 in CDCA-induced GIM. When NF-κB signaling was blocked, GATA4 and MUC2 upregulation in response to CDCA was impaired. A recent study has indicated that CDCA can affect the expression of CDX2 and MUC2 through NF-κB signaling pathway.²³ Given our data showing that CDX2 can transactivate GATA4, we suggest that CDCA might induce the activation of NF-κB signaling to promote CDX2 expression, which in turn trigger transactivation of GATA4 and MUC2. It has been shown that CDX2 and GATA4 have synergistic effects on the intestinal cell differentiation program by regulating intestinal genes.^45,46 We also explored the relationship between CDX2 and GATA4 in CDCA-induced GIM. Most interestingly, GATA4 and CDX2 can transactivate each other. Taken together, we propose a model in which activation of NF-κB signaling by CDCA promotes the expression of GATA4 and CDX2 to transactivate MUC2 in gastric epithelial cells (Fig. 6).

Figure 6. A schematic model showing that induction of NF-κB signaling via CDCA promotes the expression of GATA4 and CDX2 to transactivate MUC2 in gastric epithelial cells, consequently contributing to GIM. NF-κB, nuclear transcription factor-κB; CDCA, chenodeoxycholic acid; GATA4, GATA binding protein 4; CDX2, caudal-related homeobox 2; MUC2, mucin 2; GIM, gastric intestinal metaplasia.

In summary, GATA4 is aberrantly upregulated and positively correlated with MUC2 in GIM. GATA4 and CDX2 form a positive feedback loop and promote MUC2 transcription. The roles of GATA4, CDX2, and MUC2 in the pathogenesis of GIM warrant further investigation.

Supplementary materials can be accessed at https://doi.org/10.5009/gnl220394.

This work was supported by grants from the Chongqing Medical Scientific Research Project (Joint project of Chongqing Health Commission and Science and Technology Bureau, 2020FYYX084 to J.T.), Chongqing Natural Science Foundation (cstc2019jscx-msxmX0202 to X.W., CSTB2022NSCQ-MSX1257 to L.R, cstc2018jcyjAX0006 to J.G.).

This work was supported by two professional pathologists, Li Peng and Jing Chen (Department of Gastroenterology, Chongqing Emergency Medical Center, Chongqing University Central Hospital, Chongqing University School of Medicine).

No potential conflict of interest relevant to this article was reported.

Study concept and design: J.G. Data acquisition: X.Y., X.X., T.Y. Data analysis and interpretation: X.Y. Drafting of the manuscript: X.Y. Critical revision of the manuscript for important intellectual content: L.R., J.T., H.P. Statistical analysis: X.W., T.Y., X.Y. Obtained funding: J.T., X.W., J.G. Administrative, technical, or material support; study supervision: J.G. Approval of final manuscript: all authors.