Wnt antagonism without TGFβ induces rapid MSC chondrogenesis via increasing AJ interactions and restricting lineage commitment

Abstract

Human mesenchymal stem cells (MSCs) remain one of the best cell sources for cartilage, a tissue without regenerative capacity. However, MSC chondrogenesis is commonly induced through TGFβ, a pleomorphic growth factor without specificity for this lineage. Using tissue- and induced pluripotent stem cell-derived MSCs, we demonstrate an efficient and precise approach to induce chondrogenesis through Wnt/β-catenin antagonism alone without TGFβ. Compared to TGFβ, Wnt/β-catenin antagonism more rapidly induced MSC chondrogenesis without eliciting off-target lineage specification toward smooth muscle or hypertrophy; this was mediated through increasing N-cadherin levels and β-catenin interactions-key components of the adherens junctions (AJ)-and increasing cytoskeleton-mediated condensation. Validation with transcriptomic analysis of human chondrocytes compared to MSCs and osteoblasts showed significant downregulation of Wnt/β-catenin and TGFβ signaling along with upregulation of α-catenin as an upstream regulator. Our findings underscore the importance of understanding developmental pathways and structural modifications in achieving efficient MSC chondrogenesis for translational application.

Keywords: Bioengineering; Biological sciences; Molecular medicine; Tissue engineering.

Graphical abstract

Figure 1

Wnt/β-catenin antagonism enhances human mesenchymal stem cell (MSC) chondrogenesis while agonism suppress chondrogenesis and upregulate master osteogenic transcription factor RUNX2 (A) Alcian blue staining (left panel) of 3D pellet-cultured human induced pluripotent stem cell-derived MSCs (iPSC-MSCs), embryonic stem cell-derived MSCs (ESC-MSCs), and bone marrow-MSCs (BMMSCs) treated with 10 μM of either the Wnt/β-catenin antagonist XAV939 (XAV) or the agonist CHIR99021 (CHIR) in complete chondrogenic medium containing TGFβ3 (ChM) for 20 days. Quantification of Alcian blue staining (right graph) was performed, with comparisons of Wnt antagonism or agonism to ChM for each MSC type. Scale bar, 500 μm. (B) Alcian blue staining (left panel) and quantification (right panel) of pellet-cultured iPSC-MSCs treated with either XAV or CHIR at the indicated concentrations in ChM for 20 days. Scale bar, 500 μm. (C–F) Gene expression levels of the chondrogenic genes (C) SOX9, (D) collagen 2A1 (COL2A1), and (E) aggrecan (ACAN), as well as (F) RUNX2, the osteogenic master transcription factor, in 3D pellet-cultured iPSC-MSCs treated with either XAV or CHIR in ChM cultured for 3 days as analyzed by qPCR. Data are represented as mean ± SD One-way ANOVA: ∗, p< 0.05, ∗∗, p< 0.01, ∗∗∗, p< 0.001.

Figure 2

Gene set of signaling by TGFβ family are enriched in human smooth muscle cells compared to MSCs, and TGFβ rapidly increases α-smooth muscle actin (αSMA) and RUNX2 expression in MSCs during chondrogenic induction (A) GSEA enrichment plot of signaling by TGFβ family members in human BM-MSCs (MSC; GSE128949) compared to smooth muscle cells (SMC; GSE109859). The normalized enrichment score (NES) and nominal p values are shown. (B) Relative changes of mean expression levels of selected relevant genes and the p value in human primary SMCs compared to BM-MSCs (SMC versus MSC). Genes with significant upregulation (p < 0.05, fold-change>1) are colored in red. (C and D) Confocal immunofluorescence detection (scale bar, 20 μm) and (D) quantification of αSMA on Day 3 of pellet-cultured iPSC-MSCs in ChBM alone or with addition of TGFβ1 or TGFβ3; nuclei are labeled by 4′,6-diamidino-2-phenylindole (DAPI) and αSMA fluorescent intensity was normalized to nuclear staining (DAPI). (E) Gene expression levels of RUNX2 as quantified by qPCR on Day 3 of pellet-cultured iPSC-MSCs in ChBM alone or with addition of TGFβ1 or TGFβ3. Data are represented as mean ± SD. One-way ANOVA: ∗, p< 0.05, ∗∗, p< 0.01, ∗∗∗, p< 0.001.

Figure 3

Wnt/β-catenin antagonism alone induced more rapid MSC chondrogenesis than TGFβ in vitro and in vivo (A) Alcian blue staining of pellet-cultured iPSC-MSCs treated with the indicated modulators (10 ng/mL TGFβ3, 10 μM CHIR, or 10 μM XAV) at Day 20 and Day 10 (left top and bottom panels, respectively), with absorbance quantification for Day 10 results (right panel). Scale bar, 500 μm. (B) Alcian blue staining (left panel) and absorbance quantification (right panel) of micromass-cultured iPSC-MSCs treated with the indicated modulators for 10 days. Scale bar, 5 mm. (C and D) Expression levels of chondrogenic gene (C) COL2A1 and (D) ACAN in 3-day pellet-cultured MSCs treated with the indicated modulators for 3 days as quantified by qPCR. (E) Schematic procedure of in vivo experimentation. Mouse BM-MSCs were cultured as 3D pellets in ChBM first then transplanted subcutaneously into wildtype mouse. Modulators (10 μM of CHIR, XAV, or 10 ng/mL of TGFβ3) were injected locally every 3 days until harvest at Day 20. (F) Alcian blue staining (left panel) and absorbance quantification (right panel) of harvested tissue sections from transplanted MSCs treated with the indicated modulators at Day 20. Scale bar, 500 μm. Data are represented as mean ± SD One-way ANOVA: ∗, p < 0.05, ∗∗, p < 0.01, ∗∗∗, p < 0.001.

Figure 4

Wnt/β-catenin antagonism but not TGFβ agonism during MSC chondrogenesis decreased canonical Wnt/β-catenin transcriptional activity including RUNX2 expression (A) Immunofluorescent staining of β-catenin in iPSC-MSC pellets cultured in ChBM only or with addition of TGFβ3 (10 ng/mL), CHIR (10 μM), or XAV (10 μM) at 1 day. Dotted line indicates nuclear borders (stained with DAPI). Scale bar, 10 μm. (B) Quantification of nuclear β-catenin intensity. (C and D) Gene expression levels of (C) AXIN2, (D) TCF7 and (E) RUNX2 in iPSC-MSC pellets cultured in ChBM only or with addition of TGFβ3 (10 ng/mL), CHIR (10 μM), or XAV (10 μM) for 3 days as quantified by qPCR. #, compared to all other groups. Data are represented as mean ± SD One-way ANOVA: ∗, p< 0.05, ∗∗, p< 0.01, ∗∗∗, p< 0.001.

Figure 5

Wnt/β-catenin antagonism but not TGFβ agonism increases N-cadherin expression, and interactions with β-catenin as well as actin cytoskeleton-mediated condensation (A and B) Representative confocal immunofluorescence microscopy images and (B) quantification of N-cadherin expression in iPSC-MSCs cultured as micromass in ChBM only or with addition of TGFβ3 (10 ng/mL), CHIR (10 μM), or XAV (10 μM) for 1 day. Nuclei are labeled by DAPI. Scale bar, 5 μm. (C and D) Proximity ligation assay (PLA) for N-cadherin-β-catenin (Ncad-βcat) interaction and (D) quantification of signal counts per field of iPSC-MSCs cultured as micromass in the indicated chondrogenic conditions for 1 day. Nuclei are labeled by DAPI. Yellow scale bar, 20 μm. White scale bar, 5 μm. (E) Relationship of core elements in N-cadherin pathway derived from the Pathway Interaction Database. PID_NCADHERIN_PATHWAY filtered by first neighborhood of CTNNB1 and CDH2 are presented. (F and G) Images of phase contrast and Alcian blue staining images (top and bottom panel respectively) and (G) absorbance quantification of pellet-cultured iPSC-MSCs treated with the indicated modulators (0.25 μM cytochalasin D (CytoD), 10 μM XAV) at Day 10. Data are represented as mean ± SD One-way ANOVA: ∗, p< 0.05, ∗∗, p< 0.01, ∗∗∗, p< 0.001.

Figure 6

Wnt/β-catenin and TGFβ-related pathways are significantly downregulated in human primary chondrocytes compared to MSCs and osteoblasts (A and B) Principal component analysis (PCA) based on transcriptomic data and (B) interpreted relationship between human primary chondrocytes (Chondro), osteoblasts (Ostb) and BM-MSCs (MSC). (C) Enrichment analysis of major developmental pathways in human primary chondrocytes compared to MSCs (Chondro versus MSC), osteoblasts compared to MSCs (Ostb versus MSC) and chondrocytes compared to osteoblasts (Chondro versus Ostb) by Gene Set Enrichment Analysis (GSEA). Pathways in blue and orange are significantly enriched (p<0.05) with negative and positive normalized enrichment score respectively, non-significant pathways (p>0.05) were colored in gray. (D) GSEA enrichment plot of signaling by Wnt (stable Identifier: R-HSA-195721) and TGFβ family members (stable Identifier: R-HSA-9006936) in human primary chondrocytes compared to MSCs (left 2 panels) and osteoblasts (right 2 panels). The normalized enrichment score (NES) and nominal p values are shown. (E) Heatmap showing the Robust Multi-array Average (RMA)-normalized expression levels of selected genes relevant to osteogenesis (Osteo), hypertrophic cartilage (Hypertrophy), chondrogenesis (Chondro), inhibition of Wnt/β-catenin signaling (Wnt/β-catenin inhibition) and activation of Wnt/β-catenin signaling (Wnt/β-catenin activation) in samples of human primary chondrocytes, BMMSCs and osteoblasts. (F) Relative changes of mean expression levels of selected gene sets and the p value in human primary chondrocytes compared to MSCs and chondrocytes compared to osteoblasts. Genes with significant upregulation (p < 0.05, fold-change>1) are colored in red, and with significant downregulation (p < 0.05, fold-change<−1) are colored in green. (G) Upstream analysis for differential gene expression of human primary chondrocytes compared to MSCs (Chondro versus MSC, top panel) and compared to osteoblasts (Chondro versus Ostb, bottom panel) generated by Ingenuity Pathway Analysis (IPA). Wnt-, TGFβ- and adherens junction-related factors with significance (p < 0.05) are presented.

Figure 7

Wnt/β-catenin antagonism induce robust and specific MSC chondrogenesis Wnt/β-catenin antagonism induces rapid chondrogenic differentiation by restricting osteogenic lineage commitment and enhancing pellet condensation through increasing N-cadherin expression and N-cadherin/β-catenin interactions at the AJ. The removal of TGFβ further avoids off-target specification toward osteogenesis/hypertrophy, fibrosis, and smooth muscle differentiation.