FOXM1 Modulation Alleviates Epithelial Remodeling and Inflammation in Eosinophilic Esophagitis

Abstract

Background: Eosinophilic esophagitis (EoE) is a chronic allergic disease characterized by esophageal epithelial remodeling, barrier dysfunction, and inflammation. Despite histologic remission, molecular and structural changes in the epithelium persist, contributing to ongoing symptoms and relapse. The transcription factor FOXM1 has been shown to be a key regulator of epithelial proliferation and inflammation in allergic asthma.

Objective: To investigate the role of FOXM1 in epithelial disruption in EoE and to evaluate the therapeutic potential of FOXM1 inhibition.

Design: FOXM1 expression was analyzed in human esophageal biopsies, patient-derived organoids, and murine EoE models. IL-13 stimulation was used to model EoE in vitro. The effects of FOXM1 inhibition via the small molecule RCM-1 and siRNA-mediated knockdown were assessed by histology, gene expression profiling, organoid formation rates, and barrier integrity assays. RNA sequencing and chromatin immunoprecipitation were performed to elucidate molecular mechanisms.

Results: FOXM1 was significantly upregulated in patients with active EoE and localized to the basal epithelium. IL-13 increased FOXM1 expression, which impaired epithelial differentiation and enhanced basal cell hyperplasia. FOXM1 inhibition restored differentiation markers, reduced basal hyperplasia, and improved barrier function. In murine models, RCM-1 ameliorated epithelial changes and decreased eosinophil infiltration. Mechanistically, FOXM1 directly regulated cell cycle gene, CCNB1, which was upregulated in EoE and downregulated upon FOXM1 inhibition. FOXM1 expression was driven by an IL-13-PI3K/AKT axis.

Conclusion: FOXM1 plays a pivotal role in epithelial disruption in EoE by driving proliferation and impairing differentiation. Targeting FOXM1 restores epithelial homeostasis, mitigates inflammation, and offers a novel therapeutic approach for EoE.

Figure 1.. FOXM1 expression is elevated in patients with active EoE.

(A-C) RNA sequencing analyses on the esophageal biopsy specimens (GSE58640). (A) Top 10 enriched pathways in Pathway Interaction Database (PID) in EoE patients generated by Gene Set Enrichment Analysis. (B) Enrichment plot of FOXM1 pathway in the PID. (C) Reads per kilobase per million (RPKM) values for FOXM1 in control (n = 6) and active EoE (n = 9) subjects. (D) Quantitative RT-PCR for FOXM1 in non-EoE (n = 13), active EoE (n = 15), and inactive EoE (n = 8) biopsies in our cohort. Expressions are shown as relative values against the median expression in non-EoE group. (E) Changes in FOXM1 mRNA expressions in active and inactive phases of the same patient (n = 10). (F and G) Representative images of immunohistochemistry for FOXM1 of non-EoE, active EoE, and inactive EoE biopsies. Scale bar, 100 μm. In the epithelium per high-power field, FOXM1 levels were quantified as number of nucleus-stained cells with FOXM1 divided by the total number of nuclei (n = 7 per group). Data are indicated as means ± SDs. Two-tailed Student’s t-test (C), one-way analysis of variance (D and G), and paired t-test (E) were utilized for statistics. *P <0.05, **P <0.01, ****P <0.0001. FDR, false discovery rate; NES, normalized enrichment score; ns, not significant

Figure 2.. FOXM1 is upregulated by IL-13 stimulation in esophageal epithelium.

(A) Quantitative RT-PCR for FOXM1 of EPC2-hTERT cells stimulated with or without IL-13 (10 ng/ml), IL-4 (10 ng/ml), TNF-α (10 ng/ml), or TGF-β (10 ng/ml) for 24 h in monolayer culture (n = 3). (B) Representative images of immunoblot for FOXM1 of EPC2-hTERT cells stimulated with or without IL-13 (10 ng/ml) for 24 h in monolayer culture. (C and D) EPC2-hTERT or patient derived organoids (PDOs) were cultured for 7 days and then stimulated with or without IL-13 (10 ng/ml) for 4 days. The day 11 organoids were analyzed. (C) Quantitative RT-PCR for FOXM1 of the organoids (n = 3). (D) Representative images of immunohistochemistry for FOXM1 of the organoids. Scale bar, 50 μm. Ratio of nucleus-stained cells with FOXM1 per organoid were counted (n = 10). Data are representative of three independent experiments and indicated as means ± SDs. One-way analysis of variance (A) and two-tailed Student’s t-test (C and D) were utilized for statistics. *P <0.05, **P <0.01, ****P <0.0001. NT, nontreated

Figure 3.. FOXM1 inhibition with RCM-1 abrogates epithelial disruption in the setting of IL-13 stimulation.

(A) Representative images of immunoblot for FOXM1. Monolayer-cultured EPC2-hTERT cells were treated with DMSO or RCM-1 (10 μM or 20 μM) for 24 h. (B) Representative images of immunoblot for TP63 and IVL. EPC2-hTERT cells were treated with or without IL-13 (10 ng/ml) and RCM-1 (20 μM) in the setting of high-calcium KSFM (1.8 mM Ca²⁺). (C-F) Patient derived organoids (PDO1: C and D; PDO2: E and F) were cultured for 7 days and then treated with or without IL-13 (10 ng/ml) and RCM-1 (10 μM) for 4 days. Day 11 organoids were subjected to quantitative RT-PCR (n = 3), hematoxylin and eosin (H&E) staining, immunohistochemistry for TP63, and immunofluorescence staining for IVL (red) and FLG (green) of the organoids. Representative images are shown. DAPI (blue). Scale bar, 50 μm. (G and H) Representative phase contrast images of EPC2-hTERT organoids. Organoids were treated with or without IL-13 (10 ng/ml) and RCM-1 (10 μM) from day 7 to day 11 and then passaged. Organoid formation rate (OFR) was assessed at day 11 (passage 1). Scale bar, 1000 μm. OFR was defined as the number of organoids (≥50 μm) divided by the total seeded cells (n = 6). Data are representative of three independent experiments and indicated as means ± SDs. One-way analysis of variance (C and E) and two-tailed Student’s t-test (H) were utilized for statistics. *P <0.05, **P <0.01, ****P <0.0001. NT, nontreated

Figure 4.. FOXM1 inhibition restores epithelial barrier integrity disruptions caused by IL-13 stimulation.

(A) Transepithelial electrical resistance (TEER) of the EPC2-hTERT ALI-cultures (n = 3). EPC2-hTERT cells were kept in low-calcium (0.09 mM Ca²⁺) KSFM for 3 days, followed by high-calcium KSFM (1.8 mM Ca²⁺) for 5 days, and then remove media at day 8. Air-liquid interface (ALI)-cultured cells were stimulated with or without IL-13 (10 ng/ml) and RCM-1 (20 μM) from day 9 to 15. (B) Quantitative RT-PCR for TP63, SOX2, IVL, and FLG of the day 15 EPC2-hTERT ALI cultures (n = 3). (C) Representative images of hematoxylin and eosin (H&E), immunohistochemistry for TP63, and immunofluorescence staining for IVL (red), and FLG (green) of the day 15 EPC2-hTERT ALI-cultures. DAPI (blue). Scale bar, 50 μm. Data are representative of two independent experiments and indicated as means ± SDs. One-way analysis of variance (A and B) was utilized for statistics. *P <0.05, ***P <0.001, ****P <0.0001

Figure 5.. FOXM1 inhibition leads to decreases epithelial disruption and reduces inflammation in murine EoE model

(A) Schematic of murine EoE model and RCM-1 treatment. (B) Representative images of hematoxylin and eosin (H&E), immunohistochemistry for FOXM1, TP63, and Ki-67 of the murine esophagi. Arrowheads show infiltrating eosinophils. Scale bar, 50 μm. (C) Number of eosinophils per high-power field (hpf) in the esophagus. Number of FOXM1, TP63, or Ki-67-stained cells in the epithelium per hpf (n = 10). (D) Representative flow cytometry images of eosinophils in the murine esophagi. (E) Frequencies of eosinophils divided by CD45⁺ cells in the murine esophagi as measured by flow cytometry (n = 5). Data are indicated as means ± SDs. One-way analysis of variance (C and E) was utilized for statistics. *P <0.05, **P <0.01, ***P <0.001, ****P<0.0001. OVA, ovalbumin; i.p., intraperitoneal injection; i.g., intragastric gavage administration; ns, not significant

Figure 6.. RCM-1 suppresses eosinophil chemotaxis in EoE.

(A) Quantitative RT-PCR for CCL26. EPC2-hTERT or patient derived organoids (PDOs) were cultured for 7 days and then stimulated with or without IL-13 (10 ng/ml) and RCM-1 (10 μM) for 4 days. Day 11 organoids were harvested (n = 3). Data are representative of three independent experiments and indicated as means ± SDs. (B and C) Correlation analyses between FOXM1 and eosinophils and CCL26 on esophageal biopsies (n = 34). FOXM1 and CCL26 mRNA expressions were measured by quantitative RT-PCR and number of eosinophils per high-power field (hpf) were evaluated in the biopsies. One-way analysis of variance (A) and Pearson correlation coefficient (B and C) were utilized for statistics. *P <0.05, **P <0.01, ****P <0.0001

Figure 7.. FOXM1 alters cell cycle progression via transcriptional regulation of CCNB1 in esophageal epithelium.

(A) Top 5 enriched and depleted terms on differentially expressed genes (DEGs) overlapping in siFOXM1_1 and siFOXM1_3 cells based on Gene Ontology analysis. After the transfection, EPC2-hTERT cells were cultured in high-calcium (1.8 mM Ca²⁺) KSFM for 3 days and submitted to RNA sequencing. (B) Venn diagrams and heatmap generated by DEGs overlapping in the siFOXM1 and the EoE biopsy RNA sequencing (GSE58640). (C) Reads per kilobase per million (RPKM) values for CCNB1 in the EoE biopsy RNA sequencing (GSE58640) (control: n= 6, active EoE: n = 9). (D and E) Quantitative RT-PCR (n = 3) and representative images of immunoblot for FOXM1 in siFOXM1-transfected EPC2-hTERT cells in monolayer culture. The day following the transfection, cells were cultured in high-calcium (1.8 mM Ca²⁺) KSFM for 2 days along with or without 1 day of IL-13 (10 ng/ml) stimulation. (F) WST-1 assay for siFOXM1-transfected EPC2-hTERT cells (n = 3) (G) Cell cycle assay for siFOXM1-transfected EPC2-hTERT cells as measured by flow cytometry (n = 3). The day following the transfection, cells were cultured in high-calcium KSFM (1.8 mM Ca²⁺) with or without IL-13 (10 ng/ml) for 2 days. (H) Schematic of FOXM1 binding site identified on the promoter of CCNB1 and ChIP-qPCR for the binding site. Results were represented by fold enrichment method against IgG (n = 3; two samples of IL-13 IgG were not detected by qRT-PCR). (I) Schematic of the mechanism by which FOXM1 regulates the epithelial proliferation-differentiation gradient in esophagus. Data are representative of three independent experiments and indicated as means ± SDs. One-way analysis of variance (D, F, and G) and two-tailed Student’s t-test (C) were utilized for statistics. *P <0.05, **P <0.01, ***P <0.001, ****P <0.0001. FDR, false discovery rate; NC, negative control; NT, nontreated

Figure 8:. PI3K-AKT activation drives FOXM1 expression in EoE.

(A and B) Representative immunohistochemical images of phospho-PI3K (p-PI3K) in biopsies from patients with non-EoE, active EoE, and inactive EoE. Scale bar, 100 μm. In the epithelium per high-power field, p-PI3K levels were quantified along with the following immunohistochemical score. Stained-nucleic intensity and stained-cytoplasmic area were score as 1, 2, 3, or 4, and then each value was added (n = 7 per group). Data are indicated as means ± SDs. (C) Representative images of immunoblot for p-PI3K, PI3K, p-AKT, and AKT in EPC2-hTERT cells stimulated with IL-13 (10ng/mL) for 30 min. (D) Representative images of immunoblot for FOXM1 in EPC2-hTERT cells treated with LY294002 (10μM) for 24 h. (E) Schematic of upstream mechanism of FOXM1 in esophageal epithelium. One-way analysis of variance (B) was utilized for statistics. *P <0.05. NT, nontreated; ns, not significant