Skeletal muscle secreted factors prevent glucocorticoid-induced osteocyte apoptosis through activation of β-catenin

Abstract

It is a widely held belief that the sole effect of muscle on bone is through mechanical loading. However, as the two tissues are intimately associated, we hypothesized that muscle myokines may have positive effects on bone. We found that factors produced by muscle will protect osteocytes from undergoing cell death induced by dexamethasone (dex), a glucocorticoid known to induce osteocyte apoptosis thereby compromising their capacity to regulate bone remodeling. Both the trypan blue exclusion assay for cell death and nuclear fragmentation assay for apoptosis were used. MLO-Y4 osteocytes, primary osteocytes, and MC3T3 osteoblastic cells were protected against dex-induced apoptosis by C2C12 myotube conditioned media (MT-CM) or by CM from ex vivo electrically stimulated, intact extensor digitorum longus (EDL) or soleus muscle derived from 4 month-old mice. C2C12 MT-CM, but not undifferentiated myoblast CM prevented dex-induced cell apoptosis and was potent down to 0.1 % CM. The CM from EDL muscle electrically stimulated tetanically at 80 Hz was more potent (10 fold) in prevention of dex-induced osteocyte death than CM from soleus muscle stimulated at the same frequency or CM from EDL stimulated at 1 Hz. This suggests that electrical stimulation increases production of factors that preserve osteocyte viability and that type II fibers are greater producers than type I fibers. The muscle factor(s) appears to protect osteocytes from cell death through activation of the Wnt/β-catenin pathway, as MT-CM induces β-catenin nuclear translocation and β-catenin siRNA abrogated the positive effects of MT-CM on dex-induced apoptosis. We conclude that muscle cells naturally secrete factor(s) that preserve osteocyte viability.

Fig. 1

The effect of skeletal muscle cell conditioned media on MLO-Y4 viability. MLO-Y4 cells were pre-incubated for 1 h with 10 % blank or conditioned media (CM), prior to 6 h incubation with or without dexamethasone (dex). Graphs represent the average percentage of cell death and the standard error of the mean determined by trypan blue assay. Effect of 10 % C2C12 myoblast (MB) and myotube (MT) conditioned media on dexamethasone-induced MLO-Y4 cell death. a: statistical difference compared to all but ‘b’ using an ANOVA and Tukey post-hoc test (p ≤ 0.05; n = 4–6 experiments with 4 replicates each).

Fig. 2

The effect of C2C12 myotube conditioned media on MLO-Y4 apoptosis. MLO-Y4 cells were pre-incubated for 1 h with 10 % blank or conditioned media (CM), prior to 6 h incubation with or without dexamethasone (dex). The effect of 10 % myotube (MT) conditioned media on dexamethasone-induced MLO-Y4 apoptosis was determined by nuclear fragmentation assay. (A) Representative images of DAPI stained cells under treatment conditions – I) control; II) dexamethasone; III) myotube conditioned media; IV) myotube conditioned media + dexamethasone. Scale bar represents 10 μm. (B) Graph shows the average and standard error of the mean of apoptotic cells in each treatment condition. a: statistical difference compared to all using an ANOVA and Tukey post-hoc test (p ≤ 0.05; n = 3 experiments with 4 replicates each).

Fig. 3

The effect of C2C12 myotube conditioned media on viability of primary-derived osteocyte-enriched cells and osteoblastic MC3T3 cells. Primary osteocyte-enriched cells and MC3T3 osteoblastic cells were pre-incubated for 1 h with 10 % blank or conditioned media (CM), prior to 6 h incubation with or without dexamethasone (dex). (A) Graph shows the effect of 10 % C2C12 myotube (MT) conditioned media on dexamethasone-induced apoptosis in primary osteocyte-enriched cells (average and standard deviation). a: statistical difference compared to all using an ANOVA and Tukey post-hoc test (p ≤ 0.05; n = 1 experiment with 6 replicates). (B) Graph shows the effect of 10 % C2C12 myotube conditioned media on dexamethasone-induced cell death in MC3T3 cells (average and standard error of the mean). a: statistical difference compared to all using an ANOVA and Tukey post-hoc test (p ≤ 0.05; n = 3 experiments with 4 replicates each).

Fig. 4

Conditioned media from NIH3T3s, HeLa and CRL-1927 cells did not protect against dexamethasone-induced MLO-Y4 cell death. NIH3T3 fibroblast, HeLa epithelial cell, and CRL-1927 mesangial cell conditioned media (CM) were used at 10 % for pre-incubation of MLO-Y4 cells prior to 6 h incubation with or without dexamethasone (dex). The graph shows that only pre-incubation with myotube (MT) conditioned media significantly blocked by dexamethasone-induced cell death in MLO-Y4 cells. a: statistical difference compared to all but ‘b’ using ANOVA and Tukey post-hoc test (p ≤ 0.05; n = 3–5 experiments with 4 replicates each).

Fig. 5

Dose response curve of C2C12 myotube conditioned media to prevent dexamethasone-induced cell death in MLO-Y4 cells. MLO-Y4 cells were pre-incubated for 1 h with dilutions as indicated of myotube conditioned media (MT-CM), prior to 6 h incubation with or without dexamethasone (dex). Graph shows the average cell death and standard error of the mean of MLO-Y4 cell death under the treatment conditions. Dilution down to 0.1 % myotube conditioned media is still effective in preventing dex-induced cell death. a: statistical difference compared to all but ‘b’ using an ANOVA and Tukey post-hoc test (p ≤ 0.05; n = 3 experiments with 4 replicates each).

Fig. 6

Effect of skeletal muscle cell conditioned media derived during a high and low frequency loading regime on MLO-Y4 viability. Extensior digitorum longus (EDL) and soleus (SOL) from 5 months old C57BL6 were contracted at 1 Hz or 80 Hz to generate conditioned media (CM). MLO-Y4 cells were pre-incubated for 1 h with 10 % blank or conditioned media (A) or 0.1 % blank or conditioned media (B), prior to 6 h incubation with or without dexamethasone (dex). Graphs represent the average percentage of cell death and the standard error of the mean as determined by trypan blue assay. (A) a: statistical difference compared to all using an ANOVA and Tukey post-hoc test (p ≤ 0.05; n = 3 experiments with 4 replicates each). (B) a: dex-treated groups are statistical indifferent to each other apart from ‘b’ using an ANOVA and Tukey post-hoc test (p ≤ 0.05; n = 3 experiments with 4 replicates each).

Fig. 7

C2C12 myotube conditioned media induces β-catenin translocation to the nucleus in MLO-Y4 cells. (A) Immunofluorescent staining of β-catenin in MLO-Y4 cells exposed to 10 % blank or myotube conditioned media (MT-CM) for 2 h. Scale bar represents 50 μm. (B) Graph shows the average and standard error of the mean of nuclear to cytoplasmic ratio of the intensity of the staining. a: statistical difference compared to blank media using t-test (p ≤ 0.05; n = 1 experiment with 40 cells each).

Fig. 8

β-Catenin silencing in MLO-Y4 osteocytes abrogates the effect of C2C12 myotube conditioned media. (A) Representative western blot of total β-catenin in non-transfected, vehicle-treated, control siRNA transfected, or β-catenin transfected MLO-Y4 cells. (B) MLO-Y4 osteocytes were treated with siRNA to β-catenin, control (Risc-free) siRNA, Oligofectamin (vehicle) or non-transfected 48 h prior to 1 h pre-incubation with myotube conditioned media (MT-CM) and 6 h treatment with or without dexamethasone (dex). Graph shows the average and standard deviation of the percentage of apoptotic cells as determined by nuclear fragmentation for representative experiment. a: statistical difference compared to all but ‘b’ using ANOVA and LSD post-hoc test (p ≤ 0.05; n = 1 experiments with 4 replicates).