WNT signaling contributes to the extrahepatic bile duct proliferative response to obstruction in mice

Abstract

Biliary obstruction and cholangiocyte hyperproliferation are important features of cholangiopathies affecting the large extrahepatic bile duct (EHBD). The mechanisms underlying obstruction-induced cholangiocyte proliferation in the EHBD remain poorly understood. Developmental pathways, including WNT signaling, are implicated in regulating injury responses in many tissues, including the liver. To investigate the contribution of WNT signaling to obstruction-induced cholangiocyte proliferation in the EHBD, we used complementary in vivo and in vitro models with pharmacologic interventions and transcriptomic analyses. To model obstruction, we used bile duct ligation (BDL) in mice. Human and mouse biliary organoids and mouse biliary explants were used to investigate the effects of WNT activation and inhibition in vitro. We observed an upregulation of WNT ligand expression associated with increased biliary proliferation following obstruction. Cholangiocytes were identified as both WNT ligand-expressing and WNT-responsive cells. Inhibition of WNT signaling decreased cholangiocyte proliferation in vivo and in vitro, while activation increased proliferation. WNT effects on cholangiocyte proliferation were β-catenin dependent, and we showed a direct effect of WNT7B on cholangiocyte growth. Our studies suggested that cholangiocyte-derived WNT ligands can activate WNT signaling to induce proliferation after obstructive injury. These findings implicate the WNT pathway in injury-induced cholangiocyte proliferation within the EHBD.

Keywords: Cell cycle; Growth factors; Hepatology; Mouse models.

Figure 1. Acute injury induces epithelial cell proliferation and upregulation of cell cycle–related genes.

(A) Schematic of sample collection for histological and transcriptomic assessment of the EHBD after BDL. (B) Ki67 (green) and DAPI (blue) immunofluorescence staining 24 hours after BDL. Asterisks indicate lumen, arrows indicate Ki67⁺ cells. Scale bar: 20 μm. (C) Morphometric analysis of cholangiocyte proliferation in sham/BDL EHBDs. (D) Principal component analysis (PCA) from bulk RNA-Seq of sham/BDL EHBDs 24 hours after surgery. (E) Volcano plot depicting DEGs, cyclins, and their kinases associated with cell cycle as well as functional cholangiocyte genes (indicated by asterisks). (F) Normalized counts for proliferation markers Pcna, Mki67, Top2a, and Aurka from sham/BDL EHBDs. (G) Ontology of the top 500 upregulated genes from sham/BDL EHBDs. Unpaired Student’s t test for morphometric analysis of proliferation. Statistical significance of bulk RNA-Seq was assessed using DESeq2. ****P < 0.0001. n = 4–5 mice/group.

Figure 2. WNT signaling is upregulated following 24-hour EHBD injury.

(A) Bulk RNA-Seq analysis of transcription differences among the WNT ligands expressed in the EHBD. (B) qPCR of the EHBD dominant WNT ligand Wnt7b (left y axis) and cholangiocyte proliferation (right y axis) in sham (24 hour) and ligated mice at 12, 24, and 48 hours after BDL (15). (C) Bulk RNA-Seq analysis of WNT target gene expression changes after BDL. One-way ANOVA with Bonferroni’s multiple-comparison test was used to assess qPCR data. Statistical significance of targets identified by the bulk RNA-Seq analysis was assessed using DESeq2. ***P < 0.001, ****P < 0.0001, ^####P < 0.0001, 1-way ANOVA with Bonferroni’s multiple-comparison test was used to assess proliferation rates. n = 3–13 mice/group.

Figure 3. Cholangiocytes express WNT ligands.

(A) UMAP with epithelial (blue), fibroblast (red), and immune (green) cell clusters shown. Distribution of cells from the combined sham/BDL samples (left), sham alone (middle), and BDL (right). (B) Changes in cluster compositions following BDL based on cluster proportions. Epi, epithelial; Fib, fibroblast; Imm, immune cell clusters. (C) Expression of the predominant WNT ligands in the EHBD. n = 2 samples/treatment (1 male, 1 female), n = 5 mice/sample. (D) In situ hybridization for Wnt7b in the EHBD of sham/BDL mice. Asterisks indicate lumen, line separates epithelial and stromal cell compartments, lower panels show magnification of area indicated by dashed black box, and arrows indicate Wnt7b⁺ cells. Scale bar: 50 μm. n = 3 mice/group.

Figure 4. Cholangiocytes are WNT target cells.

(A) UMAP of combined cells from sham/BDL scRNA-Seq datasets. (B) WNT target genes localize primarily to epithelial cells clusters in all scRNA-Seq samples. n = 2 samples/treatment (1 male, 1 female), 5 mice/sample. (C and D) BIRC5 (green) and CD44 (red) immunofluorescence staining with DAPI (blue) in sham/BDL mice. Asterisks indicate lumen, lower panels show increased magnification of dotted white box, arrows indicate BIRC5⁺ and CD44⁺ cells, respectively. Scale bar: 50 μM. n = 3 mice/group.

Figure 5. Canonical WNT activation promotes cholangiocyte proliferation in mouse EHBD explants and mouse and human EHBD organoids.

(A and B) Diagram of EHBD explant and organoid culturing protocols. (C) Timeline of EHBD explant experiment with 10 μM CHIR treatment. (D) EdU (green), KRT19 (red), and DAPI (blue) staining in explants. (E) Morphometric analysis of proliferation in biliary explants. (F and G) BIRC5 (green), KRT19 (red), and DAPI (blue) staining in explants and quantification. Asterisks indicate lumen, arrows indicate EdU⁺ and BIRC5⁺ cells. Scale bar: 50 μm. n = 3–5 mice/group. (H) Timeline of the experiment treating mouse and human biliary organoids with 3 μM CHIR. (I–K) Mouse organoid images (I), size (J), and viability (K) by measurement of culture ATP levels in vehicle- and CHIR-treated organoids.(L–N) Human organoid images (L), size (M), and viability (N) measurements in vehicle- and CHIR-treated organoids. n = 3 biological replicates. Scale bar: 500 μm. Size and viability measurements are normalized the vehicle controls. Two-way ANOVA with Bonferroni’s multiple-comparison test was used for morphometric analysis of EdU and BIRC5. Unpaired Student’s t test was used for organoid experiments. *P < 0.05, ****P < 0.0001.

Figure 6. Endogenous WNT ligands promote organoid growth through activation of canonical WNT signaling.

(A) qPCR from mouse EHBD organoids for the 4 most abundant WNT ligands of the EHBD. (B) Images from mouse organoid cultures 1, 2, and 3 passages after being grown in WNT-free media. (C) Bulk RNA-Seq analysis of human organoids for WNT ligands. (D) Timeline for C59 administration to inhibit WNT secretion by mouse organoids. (E–G) Mouse organoid images (E), size (F), and viability (G) measurements after treatment with 10 μM C59. (H) Timeline for IWR-1 administration to inhibit β-catenin in mouse organoids. (I–K) Mouse organoid images (I), size (J), and viability (K) following 5 μM IWR-1 treatment. (L) Timeline for C59 administration in human organoids. (M–O) Human organoid images (M), size (N), viability (O) after 10 μM C59 treatment. (P) Timeline for IWR-1 treatment in human organoids. (Q–S) Human organoid images (Q), size (R), and viability (S) following 5 μM IWR-1 treatment. Size and viability measurements are normalized to the vehicle controls. Scale bar: 500 μm. One-sample t test. *P < 0.05, **P < 0.01, ***P < 0.001, n = 3–5 biological replicates.

Figure 7. Inhibition of WNT ligand secretion leads to decreased cholangiocyte proliferation in the injured mouse EHBD.

(A) Timeline of C59 administration during sham and BDL surgeries. (B) H&E staining of EHBDs of sham/BDL mice treated with either vehicle (Veh) or C59. (C) EdU (green), KRT19 (red), and DAPI (blue) immunofluorescence staining of Veh/C59-treated sham/BDL mice. (D) Morphometric analysis of cholangiocyte proliferation using EdU marker. Asterisks indicate lumen, and arrows indicate EdU⁺ cholangiocytes. Scale bar: 50 μm. Two-way ANOVA was used to compare sham and BDL samples with Bonferroni’s multiple-comparison test. *P < 0.05. n = 3–5 mice/group.