Systemic blockade of ACVR2B ligands prevents chemotherapy-induced muscle wasting by restoring muscle protein synthesis without affecting oxidative capacity or atrogenes

Abstract

Doxorubicin is a widely used and effective chemotherapy drug. However, cardiac and skeletal muscle toxicity of doxorubicin limits its use. Inhibiting myostatin/activin signalling can prevent muscle atrophy, but its effects in chemotherapy-induced muscle wasting are unknown. In the present study we investigated the effects of doxorubicin administration alone or combined with activin receptor ligand pathway blockade by soluble activin receptor IIB (sACVR2B-Fc). Doxorubicin administration decreased body mass, muscle size and bone mineral density/content in mice. However, these effects were prevented by sACVR2B-Fc administration. Unlike in many other wasting situations, doxorubicin induced muscle atrophy without markedly increasing typical atrogenes or protein degradation pathways. Instead, doxorubicin decreased muscle protein synthesis which was completely restored by sACVR2B-Fc. Doxorubicin administration also resulted in impaired running performance without effects on skeletal muscle mitochondrial capacity/function or capillary density. Running performance and mitochondrial function were unaltered by sACVR2B-Fc administration. Tumour experiment using Lewis lung carcinoma cells demonstrated that sACVR2B-Fc decreased the cachectic effects of chemotherapy without affecting tumour growth. These results demonstrate that blocking ACVR2B signalling may be a promising strategy to counteract chemotherapy-induced muscle wasting without damage to skeletal muscle oxidative capacity or cancer treatment.

Figure 1. Doxorubicin administration resulted in decreased body and muscle weights that were restored by sACVR2B-Fc treatment.

(a) Body weights during the four-week experiment. Arrows indicate the timing of doxorubicin (orange) and sACVR2B-Fc (blue) injections. Repeated measures ANOVA revealed time x group interaction effect (P < 0.001). Tissue weights of TA (b), gastrocnemius (GA) (c) and soleus (d) muscles and epididymal fat pads (e) relative to tibial length (Supplementary Fig. S1e). Percentage changes in lean (f) and fat (g) mass analysed with DXA. (h) Average feed consumption during and after the treatment with doxorubicin and sACVR2B-Fc. Average feed consumption per mouse was calculated from the pooled feed intake of the whole cage (2–3 mice/cage; N = 3–4 cages/group). N sizes are depicted in the bar graphs. Data are presented as mean ± SEM. In Fig. a: ^#P < 0.05 compared to Ctrl; **P < 0.01 compared to Ctrl and Dox + sACVR2B (Bonferroni). In Figs. b–h: *P < 0.05; **P < 0.01; ***P < 0.001 (Bonferroni).

Figure 2. sACVR2B-Fc administration increased muscle fibre cross-sectional area in doxorubicin-treated mice.

Average fibre CSA (a) and fibre size distribution (b) of the TA muscle at the end of the four-week experiment. (c) Representative immunofluorescence images of dystrophin-stained muscle cryosections. Data are presented as mean ± SEM. **P < 0.01; ***P < 0.001 (Bonferroni).

Figure 3. sACVR2B-Fc treatment improved bone quality in doxorubicin treated mice.

Changes in bone mineral density (a) and content (b) in the four-week experiment. (c) Associations between muscle and lean mass and bone parameters. (d) Associations between muscle size, fat mass and BMD. Correlations between change in lean mass and change in BMD (e, r = 0.70, P < 0.001) and BMC (f, r = 0.72, P < 0.001). Data are presented as mean ± SEM. **P < 0.01; ***P < 0.001 (Bonferroni).

Figure 4. Doxorubicin-treated mice had significantly impaired running capacity with no effect of sACVR2B-Fc.

Distance covered in an incremental treadmill running test until exhaustion. Data are presented as mean ± SEM. *P < 0.05; **P < 0.01 (Bonferroni).

Figure 5. Doxorubicin administration did not cause marked alterations in the markers of ubituitin-proteasome system, autophagy or calpain1 content.

Time-course of protein ubiquitination (a) and LC3-I and -II content (N = 5–9/group) (b) relative to Ctrl and calpain1 content at 20 h relative to Dox (c) in TA muscles. (d) Representative blots of LC3 and calpain1. Data are presented as mean ± SEM. *P < 0.05 (Mann-Whitney U). C = Ctrl; D = Dox; D+A = Dox+sACVR2B.

Figure 6. Doxorubicin administration resulted in decreased muscle protein synthesis that was restored by sACVR2B-Fc.

(a) Muscle protein synthesis relative to Dox analysed with puromycin incorporation method and (b) representative blot (left) with Ponceau S staining (right). (− = negative control for puromycin). (c) Time-course of rpS6 phosphorylation at Ser240/244 relative to Ctrl in TA muscle. p70S6K (Thr389) (d) and eIF2α (Ser51) (e) phosphorylation response 20 hours after a single dose of doxorubicin relative to Dox. (f) Time-course of ERK 1/2 phosphorylation at Thr202/Tyr204 relative to Dox in TA muscle. (g) Representative blots of rpS6, p70S6K, eIF2α and ERK 1/2. (h) REDD1 mRNA expression normalized to 36b4 expression relative to Ctrl in TA muscle 20 hours after a single dose of doxorubicin. Data are presented as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001 (Bonferroni (a;h), Mann-Whitney U -test (c–f)).

Figure 7. Doxorubicin administration did not affect mitochondrial function or markers of mitochondrial content in skeletal muscle.

(a) Mitochondrial respiration in homogenized TA muscle with carbohydrate substrates (N = 7–8/group). Cyt c = cytochrome c; CI/II = complex I/II; ETS = electron transfer system. (b) Citrate synthase (CS) activity measured from TA muscle. Quantification of total content of mitochondrial respiratory chain subunits (c) and cytochrome c (Cyt c) relative to Dox (d) and representative blots (e). Data are presented as mean ± SEM. **P < 0.01 (Bonferroni (a–b), Mann-Whitney U (c–d)).

Figure 8. Doxorubicin administration did not affect skeletal muscle capillary density.

Quantification of the capillary-to-fibre ratio (a) and the number of capillaries per muscle area (b) in TA muscle. (c) Representative immunofluorescence images of CD31/PECAM-1 staining for capillaries. Notice that the representative capillary images are from exactly the same location as the dystrophin staining in Fig. 2c. Data are presented as mean ± SEM. *P < 0.05 (Bonferroni).