The effects of testosterone deprivation and supplementation on proteasomal and autophagy activity in the skeletal muscle of the male mouse: differential effects on high-androgen responder and low-androgen responder muscle groups

Abstract

Men with prostate cancer who receive androgen deprivation therapy show profound skeletal muscle loss. We hypothesized that the androgen deficiency activates not only the ubiquitin-proteasome systems but also the autophagy and affects key aspects of the molecular cross talk between protein synthesis and degradation. Here, 2-month-old male mice were castrated and treated with either testosterone (T) propionate or vehicle for 7 days (short term) or 43 days (long term), and with and without hydroxyflutamide. Castrated mice showed rapid and profound atrophy of the levator ani muscle (high androgen responder) at short term and lesser atrophy of the triceps muscle (low androgen responder) at long term. Levator ani and triceps muscles of castrated mice showed increased level of autophagy markers and lysosome enzymatic activity; only the levator ani showed increased proteasomal enzymatic activity. The levator ani muscle of the castrated mice showed increased level and activation of forkhead box protein O3A, the inhibition of mechanistic target of rapamicyn, and the activation of tuberous sclerosis complex protein 2 and 5'-AMP-activated protein kinase. Similar results were obtained in the triceps muscle of castrated mice. T rescued the loss of muscle mass after orchiectomy and inhibited lysosome and proteasome pathways dose dependently and in a seemingly IGF-I-dependent manner. Hydroxyflutamide attenuated the effect of T in the levator ani muscle of castrated mice. In conclusion, androgen deprivation in adult mice induces muscle atrophy associated with proteasomal and lysosomal activity. T optimizes muscle protein balance by modulating the equilibrium between mechanistic target of rapamicyn and 5'-AMP-activated protein kinase pathways.

Figure 1.

Castration is associated with loss of muscle mass and increased proteasome enzymatic activity in the levator ani muscle. (A) The mice were Cx and treated after 7 days with either vehicle or 3 mg/(kg × d) T (T3) for an additional 7 days. Sham-operated mice served as controls. (B) Castration for 7 or 14 days reduced levator ani (LA) muscle mass, decreased AR protein level (C and D), and induced 20S proteasome enzymatic activity (E) and MuRF1 and MAFbx mRNA expression measured by qPCR (F). T administration restored these changes to that seen in sham-operated mice. Results are mean ± SEM; n = 4–5/group. *, P < .05; **, P < .01; ***, P < .001 vs sham mice; #, P < .05; ##, P < .01 vs Cx 7d mice; ^, P < .05; ^^^, P < .001 vs Cx 14d mice. Cx 7d, mice castrated for 7 days; Cx 14d, mice castrated for 14 days; Cx7d/ T 7d, mice castrated for 7 days and then treated with T for 7 days starting on day 8.

Figure 2.

Castration is associated with increased cathepsin L activity along with increased expression levels and activity of ALP markers in the levator ani muscle. (A) Cathepsin L enzymatic activity. mRNA expression level was measured using qPCR for LC3B and cathepsin L (B), Bnip3 and Beclin1 (C), and Klf15 and Tfeb (D). Castration increased the level of LC3BI (E and F) and LC3BII (E) as well as increased the ratio of LC3BII to LC3BI (G). Results are mean ± SEM; n = 4–5/group. *, P < .05; **, P < .01; ***, P < .001 vs sham mice; #, P < .05; ##, P < .01; ###, P < .001 vs Cx 7d mice; ^, P < .05; ^^, P < .01; ^^^, P < .001 vs Cx 14d mice.

Figure 3.

Effect of castration and T supplementation on FoxOs gene expression and activation in the levator ani muscle. Increased mRNA level of FoxO1, FoxO3, and FoxO4 (A) assessed by qPCR. Decreased phosphorylation of FoxO3A on serine 318/321 and increased level of total FoxO3A (B–D) measured by Western blotting. There were no significant changes in total and serine 256 phosphorylation of FoxO1 (B, E, and F). The changes in total FoxO4 protein level were inconsistent in castrated mice (B, G, and H). Results are mean ± SEM; n = 4–5/group. *, P < .05; **, P < .01; ***, P < .001 vs sham mice; #, P < .05; ##, P < .01 vs Cx 7d mice; ^^, P < .01; ^^^, P < .001 vs Cx 14d mice.

Figure 4.

T deprivation is associated with mTOR inhibition in the levator ani muscle. Western blotting showing reduced mTOR activation (A and B), increased level of total TSC2 (A and C), reduced Akt-dependent TSC2 phosphorylation on serine 1462 (A and E), increased AMPKα-dependent Raptor phosphorylation on serine 792 (A and F), and numerical increase in AMPKα-dependent TSC2 phosphorylation on serine 1387 (A and G) in castrated mice. (D) qPCR showing mRNA expression level of TSC2. Results are mean ± SEM; n = 4–5/group. **P < .01; ***P < .001 vs sham mice; #, P < .05; ###, P < .001 vs Cx 7d mice; ^, P < .05; ^^, P < .01; ^^^, P < .001 vs Cx 14d mice.

Figure 5.

Castration activates AMPKα and PPARδ in the levator ani muscle. Castration was associated with increased AMPKα phosphorylation on threonine 172 (A and B), increased Lkb1 gene expression (C), and increased PPARδ protein (A and D) and mRNA (E) content and Pdk4 gene expression (E). Results are mean ± SEM; n = 4–5/group. *, P < .05; **, P < .01; ***, P < .001 vs sham mice; ##, P < .01; ###, P < .001 vs Cx 7d mice; ^^, P < .01; ^^^, P < .001 vs Cx 14d mice.

Figure 6.

Castration is associated with loss of muscle mass and increased cathepsin L enzymatic activity in the triceps muscle. (A) The mice were Cx on day 0 and treated 7 days after castration with either vehicle or T for an additional 43 days. Castration reduced triceps muscle mass (B) but did not induce proteasome activity (C). (D) qPCR for MuRF1 and MAFbx. Castration induced cathepsin L activity (E) and increased LC3B, cathepsin L, Klf15, and Tfeb (F and H), but not Bnip3 and Beclin1 (G), gene expression. LC3B level measured by Western blotting (I–K). Results are mean ± SEM; n = 4–5/group. *, P < .05; **, P < .01; ***, P < .001 vs sham mice; ^, P < .05; ^^, P < .01; ^^^, P < .001 vs Cx mice.

Figure 7.

Castration activates FoxO3 in the triceps muscle. T deprivation significantly increased FoxO1, FoxO3, and FoxO4 gene expression (A) and activated FoxO3A (B–D). Castrated mice showed a trend of induction of both FoxO1 and FoxO4 (B and E–H). Results are mean ± SEM; n = 4–5/group. *, P < .05; **, P < .01; ***, P < .001 vs sham mice; ^, P < .05; ^^, P < .01; ^^^, P < .001 vs Cx mice.

Figure 8.

Castration inhibits mTOR and activates AMPK in the triceps muscle. (A and B) Western blotting for mTOR. Castration increased TSC2 activation and level (A, C, and D) but did not change significantly the AMPKα-dependent phosphorylation of TSC2 (F). (E) qPCR of TSC2. T deprivation induced Raptor inhibition by AMPKα (A and G). (H and I) Western blotting showing that castration induces, and T represses, AMPKα activation. (J) qPCR of Lkb1. (K) Castration reduced, and T increased, PPARδ protein level. Results are mean ± SEM; n = 4–5/group. *, P < .05; **, P < .01; ***, P < .001 vs sham mice; ^, P < .05; ^^, P < .01; ^^^, P < .001 vs Cx mice.