Aerobic Exercise and Pharmacological Treatments Counteract Cachexia by Modulating Autophagy in Colon Cancer

Abstract

Recent studies have correlated physical activity with a better prognosis in cachectic patients, although the underlying mechanisms are not yet understood. In order to identify the pathways involved in the physical activity-mediated rescue of skeletal muscle mass and function, we investigated the effects of voluntary exercise on cachexia in colon carcinoma (C26)-bearing mice. Voluntary exercise prevented loss of muscle mass and function, ultimately increasing survival of C26-bearing mice. We found that the autophagic flux is overloaded in skeletal muscle of both colon carcinoma murine models and patients, but not in running C26-bearing mice, thus suggesting that exercise may release the autophagic flux and ultimately rescue muscle homeostasis. Treatment of C26-bearing mice with either AICAR or rapamycin, two drugs that trigger the autophagic flux, also rescued muscle mass and prevented atrogene induction. Similar effects were reproduced on myotubes in vitro, which displayed atrophy following exposure to C26-conditioned medium, a phenomenon that was rescued by AICAR or rapamycin treatment and relies on autophagosome-lysosome fusion (inhibited by chloroquine). Since AICAR, rapamycin and exercise equally affect the autophagic system and counteract cachexia, we believe autophagy-triggering drugs may be exploited to treat cachexia in conditions in which exercise cannot be prescribed.

Figure 1. Voluntary wheel running counteracts cancer cachexia and prolongs survival of tumor-bearing mice.

(a) TA muscle weight of C26-bearing mice in the absence (—) or presence (WR) of voluntary wheel running for 19 days. WR rescues muscle mass of C26-bearing mice. Data are shown as mean ± SEM; 6 < n < 9 for each group, *p < 0.05 by Student’s t-test. (b) Representative immunostaining for laminin on TA muscle of C26-bearing mice in the absence (—) or presence (WR) of wheel running for 19 days, showing that WR increases fiber size and improves basement membrane morphology in C26-bearing mice. Scale bar = 100 microns. (c) Median ± SEM of glycolytic TA myofiber cross-sectional area of C26-bearing mice in the absence (—) or presence (WR) of wheel running for 19 days. WR rescues fiber size of C26-bearing mice. n = 5 for each group, *p < 0.05 by Student’s t-test. (d) Fatigue time of EDL muscle of control or C26-bearing mice in the absence (—) or presence (WR) of wheel running for 19 days. Data are shown as mean ± SEM. Two-way ANOVA (F = 5.77; df 1; p = 0.025) shows an interaction between the negative effect of C26 tumor and wheel running, which restores fatigue time to control levels; data are shown as mean ± SEM; 6 < n < 9 for each group, *p < 0.05 by Tukey’s HSD test. (e) Survival curves derived from Cox model for statistical analysis of C26-bearing mice in the absence (C26, circles) or presence (C26 WR, triangles) of wheel running. Wheel running significantly (p < 0.041) increased survival of C26-bearing mice. n = 9 for each group. (f) Linear correlation between running distance (Km per day) and life span (days of survival) in C26-bearing mice. From the curve equation, one can infer that each km/day of wheel running corresponds to an increase of approximately 4 days in life span.

Figure 2. Wheel running counteracts tumor-induced atrogene expression and restores the autophagic marker levels, which are altered in both murine and human skeletal muscle in cancer cachexia.

(a) Expression of Atrogin 1 and MuRF1 genes by real-time RT-PCR, in mice in the absence (—) or presence (WR) of wheel running, 19 days after transplantation in C26-bearing mice and in control mice (CTR). Values were normalized to Gapdh expression. Data are presented as mean ± SEM. Kruskal-Wallis test revealed a global significant difference between the three groups, followed by post-hoc test *p < 0.05. (b) Representative immunostaining for p62 on TA muscle of C26-bearing mice in the absence (—) or presence (WR) of wheel running for 19 days, showing an increase of p62 induction in C26-bearing mice. Scale bar = 20 microns. (c) Western blot analysis of LC3bI, LC3bII and p62 expression in control mice (CTR), or mice in the absence (—) or presence (WR) of wheel running, 19 days after C26 transplantation. Gapdh was used as a loading control. Densitometric analyses of Western blot showing LC3bII/LC3bI ratio (d) and p62/Gapdh (e) ratio in mice in the absence (—) or presence (WR) of wheel running, 19 days after C26 transplantation, compared with control mice (CTR). Data are presented as mean ± SEM. F = 4.63; df 2; p < 0.05 and F = 14.18; df 2; p < 0.02 by ANOVA; *p < 0.005 and **p < 0.005 by Tukey’s HSD test. (f) Expression of LC3b and p62 by Western blot analyses in skeletal muscle biopsies from colon carcinoma patients and healthy (CTR) subjects. Gapdh was used as a loading control. The lanes were run on the same type of gel but were non-contiguous, as indicated by the thin black lines. Densitometric analyses of Western blot showing significantly higher expression of LC3bII/LC3BI ratio (g) and p62/Gapdh ratio (h) in skeletal muscle biopsies from colon carcinoma patients (CARC), compared with healthy subjects (CTR). Data are presented as mean ± SEM. *p < 0.05 by Student’s t-test.

Figure 3. AICAR or rapamycin treatment counteracts muscle wasting in cancer cachexia.

(a) Western blot showing total and phosphorylated AMPK levels, as well as ACC activation (p-ACC) in AICAR-treated mice, and (b) total and phosphorylated mTOR levels in rapamycin-treated mice (RAPA), compared with vehicle-treated mice (—) as well as untreated mice (CTR). Treatments consisted in daily IP injection of vehicle or [250 mg/Kg] AICAR (AICAR) or [2 mg/Kg] rapamycin (RAPA) for 19 days. (c) TA muscle weight of C26-bearing mice in the absence (—) or presence of treatment as indicated. ANOVA (F = 9.47; df 2; p < 0.0005) revealed a significant effect of the treatments on TA weight. (d) Representative immunostaining for laminin on TA muscle of C26-bearing mice in the presence of the treatments indicated, showing increased fiber size in treated animals. Scale bar = 100 microns. (e) Median ± SEM of TA myofiber cross-sectional area of C26-bearing mice in the presence of the treatments indicated. ANOVA (F = 6.86; df 2; p < 0.008) revealed a significant effect of treatments on muscle fiber CSA (F, see results). Data are shown as mean ± SEM; 12 < n < 26 for each group, * p < 0.05 by Tukey’s HSD test.

Figure 4. AICAR or rapamycin treatment counteracts the induction of atrogene expression and modulates autophagic markers in cancer cachexia.

(a) Expression of Atrogin 1 and MuRF1 genes as shown by real-time RT-PCR, in mice that received daily IP treatment with vehicle (—), AICAR (AICAR) or rapamycin (RAPA), for 19 days after C26 implant, compared with control mice (CTR). Values were normalized to Gapdh. ANOVA (F = 10.08; df 3; p < 0.01;F = 7.90; df 3; p < 0.01; for Atrogin 1 and MuRF1, respectively) showed a significant effect of the treatments on ubiquitine-ligase relative expression. (b) Representative immunostaining for p62 on TA muscle of C26-bearing mice in the absence (—) or presence of the treatments indicated, showing an increase in p62 induction in C26-bearing mice. Scale bar = 20 microns. (c) Western blot analysis of LC3bI, LC3bII, p62 and Gapdh expression in the presence of the treatments indicated. The lanes separated by the thin black lines were run at the same time on parallel gels and were, therefore, non-contiguous. (d) Densitometric analyses of Western blot showing LC3bII/LC3bI ratio in mice in the presence of the treatments indicated, compared with control mice (CTR). ANOVA (F = 6.26; df 3; p < 0.01) revealed a significant effect of the treatments on the LC3bII/LC3bI ratio. (e) Densitometric analyses of Western blot showing p62 expression in mice in the presence of the treatments indicated, compared with control mice (CTR). Values were normalized to Gapdh. ANOVA (F = 5.92; df 3; p < 0.008) revealed a significant effect of treatments on p62 protein expression. Data are presented as mean ± SEM. *p < 0.05; **p < 0.005 by Tukey’s HSD test.

Figure 5. Both AICAR and rapamycin treatments counteract C2C12 myotube atrophy induced by C26 conditioned medium.

(a) Representative immunostaining for myosin heavy chain on C2C12 myotubes cultured for 48 hours in the absence (CTR) or presence of C26-conditioned medium (C26), in the absence (—) or presence of 1 milliM AICAR, or 500 nanoM rapamycin, without (−CQ) or with (+CQ) 50 microM chloroquine. Scale bar = 20 microns. (b) Quantification of C2C12 myotube diameters in all the aforementioned conditions. n = 5. ^$p < 0.0001vs -CQ; **p < 0.001; ***p < 0.0001.

Figure 6. Exercise, AICAR and rapamycin counteract muscle wasting in cancer cachexia and modulate an adequate autophagic flux level.

(A) In physiological conditions, a balance between autophagosome production and clearance maintains an adequate autophagic flux (red arrow) in the muscles, mirrored by basal levels of LC3bII (black dots) and p62 (green dots) expression. In cachectic muscles from C26-bearing mice (B), a strong accumulation of both LC3bII and p62 proteins points to an unbalanced autophagosome production/clearance ratio. This correlates with the pathophysiological features observed in cancer-related muscle wasting, including overexpression of Atrogin1 and Murf1 genes, as well as a decline in body weight (BD), muscle weight (MW), fiber size and muscle function. Spontaneous wheel running, or treatment with AICAR or rapamycin (C) counteracts cancer-related muscle wasting in tumor-bearing mice and induces a decrease in both LC3bII and p62 accumulation. Atrogin1 and Murf1 gene expression was restored to the basal levels observed in healthy muscles. This is associated with increased body and muscle weight and improved muscle function.