Mechanical activation of β-catenin regulates phenotype in adult murine marrow-derived mesenchymal stem cells

Abstract

Regulation of skeletal remodeling appears to influence the differentiation of multipotent mesenchymal stem cells (MSC) resident in the bone marrow. As murine marrow cultures are contaminated with hematopoietic cells, they are problematic for studying direct effects of mechanical input. Here we use a modified technique to isolate marrow-derived MSC (mdMSC) from adult mice, yielding a population able to differentiate into adipogenic and osteogenic phenotypes that is devoid of hematopoietic cells. In pure mdMSC populations, a daily strain regimen inhibited adipogenic differentiation, suppressing expression of PPARγ and adiponectin. Strain increased β-catenin and inhibition of adipogenesis required this effect. Under osteogenic conditions, strain activated β-catenin signaling and increased expression of WISP1 and COX2. mdMSC were also generated from mice lacking caveolin-1, a protein known to sequester β-catenin: caveolin-1((-/-)) mdMSC exhibited retarded differentiation along both adipogenic and osteogenic lineages but retained mechanical responses that involved β-catenin activation. Interestingly, caveolin-1((-/-)) mdMSC failed to express bone sialoprotein and did not form mineralized nodules. In summary, mdMSC from adult mice respond to both soluble factors and mechanical input, with mechanical activation of β-catenin influencing phenotype. As such, these cells offer a useful model for studies of direct mechanical regulation of MSC differentiation and function.

Fig. 1. Murine mdMSC differentiate in response to culture conditions

mdMSC were cultured in osteogenic (OSTEO) or adipogenic (ADIPO) medium. (A) Designated mRNA was amplified by real-time RT-PCR. “a” shows significant difference from day 0, p < 0.01. (B) Alkaline phosphatase staining of day 5 cultures. Scale bar = 200 μm. (C) PPARγ mRNA was amplified by real-time RT-PCR. “a” shows significant difference from day 0, p < 0.01. (D) Oil red O staining of day 5 cultures. Scale bar = 200 μm. (E) Total cellular proteins were immunoblotted for designated proteins, with β-tubulin used as a loading control.

Fig. 2. Mechanical strain inhibits adipogenesis

(A) mdMSC in adipogenic medium were exposed to daily strain regimen × 3 d, and designated mRNA was amplified by real-time RT-PCR. “a” shows significant difference from unstrained control, p < 0.01. (B) Total cellular proteins were analyzed by immunoblotting after treatment as in (A) with daily strain × 5 d. CTL = control culture; STR = strained culture. Densitometric analysis of active β-catenin was performed (n = 4 experiments). “a” shows significant difference from unstrained control, p < 0.05. (C) Proteins from nuclear and cytoplasmic fractionates were immunoblotted for active β-catenin and adiponectin, respectively, along with loading controls, after treatment as in (A) with daily strain × 4 d and an undifferentiated (UND) mdMSC control sample was included for comparison. (D) Immunoblots of total cellular proteins from mdMSC treated with nonsense siRNA (− siCat) or siRNA targeting β-catenin (+ siCat), then cultured in adipogenic medium ± daily strain. Densitometric analyses of PPARγ and adiponectin were performed (n = 3 experiments). “a” shows significant difference from unstrained control, p < 0.05. (E) The GSK3β inhibitor SB415286 (SB415, 20 μM) was added to mdMSC in adipogenic medium and proteins were analyzed by immunoblotting on day 4. (F) Oil red O staining after treatment as in (D) for 5 d. Scale bar = 200 μm.

Fig. 3. Strain activates β-catenin signaling in osteogenic cultures

mdMSC were cultured in osteogenic medium for 5 d and strain applied for 30 min to 6 h. (A) The cytoplasmic protein fractionate was analyzed for β-catenin and LDH (loading control). (B) Total cellular proteins were immunoblotted for phosphorylated GSK3β and Akt. (C) WISP1 and COX2 mRNA were amplified by real-time RT-PCR in cultures strained for 6 h. “a” shows significant difference from unstrained control, p < 0.01.

Fig. 4. Absence of caveolin-1 does not block strain effects on adipogenesis

mdMSC were cultured in adipogenic medium. (A) Oil red O staining of wild-type (WT) and Cav-1^(−/−) mdMSC. Scale bar = 200 μm. (B) Total cellular proteins were immunoblotted for designated proteins. (C) Total cellular proteins were immunoblotted for caveolin-1 and actin (loading control). (D) Daily strain regimen × 7 d was applied to Cav-1^(−/−) mdMSC, and designated proteins were analyzed by immunoblotting. CTL = control culture; STR = strained culture.

Fig. 5. Osteogenic differentiation is disrupted in mdMSC lacking caveolin-1

mdMSC were cultured in osteogenic medium. (A) Designated mRNA from Cav-1^(−/−) mdMSC was amplified by real-time RT-PCR. “a” shows significant difference from day 0, p < 0.01. (B) Alizarin red S staining of three-week wild-type (WT) and Cav-1^(−/−) mdMSC cultures. (C) Strain was applied to day 5 Cav-1^(−/−) mdMSC cultures for 6 h. WISP1 and COX2 mRNA were amplified by real-time RT-PCR. “a” shows significant difference from unstrained control, p < 0.01.