Chronic alcohol consumption enhances skeletal muscle wasting in mice bearing cachectic cancers: the role of TNFα/myostatin axis

Abstract

Background:

Chronic alcohol consumption enhances cancer-associated cachexia, which is one of the major causes of decreased survival. The precise molecular mechanism of how alcohol consumption enhances cancer-associated cachexia, especially skeletal muscle loss, remains to be elucidated.

Methods:

We used a mouse model of chronic alcohol consumption, in which 20% (w/v) alcohol was provided as sole drinking fluid, and Lewis lung carcinoma to study the underlying mechanisms.

Results:

We found that alcohol consumption up regulated the expression of MAFbx, MuRF-1 and LC3 in skeletal muscle, suggesting that alcohol enhanced ubiquitin-mediated proteolysis and LC3-mediated autophagy. Alcohol consumption enhanced phosphorylation of Smad2/3, p38, and ERK, and decreased the phosphorylation of FOXO1. These are the signaling molecules governing protein degradation pathways. Moreover, alcohol consumption slightly up regulated the expression of insulin receptor substrate-1, did not affect PI3K, but decreased the phosphorylation of Akt and mTOR, and down regulated the expression of Raptor and p70S6K, suggesting that alcohol impaired protein synthesis signaling pathway in skeletal muscle of tumor-bearing mice. Alcohol consumption enhanced the expression of myostatin in skeletal muscle, plasma and tumor, but did not affect the expression of myostatin in non-tumor-bearing mice. In TNFα knockout mice, the effects of alcohol-enhanced expression of myostatin and protein degradation-related signaling molecules, and decreased protein synthesis signaling in skeletal muscle were abolished. Consequently, alcohol consumption neither affected cancer-associated cachexia, nor decreased the survival of TNFα KO mice bearing cachectic cancer.

Conclusions:

Chronic alcohol consumption enhances cancer-associated skeletal muscle loss through suppressing Akt/mTOR-mediated protein synthesis pathway and enhancing protein degradation pathways. This process is initiated by TNFα and mediated by myostatin.

Introduction

Cancer-associated cachexia (CAC) is a multifactorial symptom characterized by progressive loss of body weight, specifically the loss of adipose tissue and skeletal muscle. Approximately 85% of advanced cancer patients experience CAC, which is responsible for nearly 25% of all cancer deaths. CAC is often accompanied by anorexia, which decreases calorie intake leading to the body weight loss. However, CAC is irreversible by increase of caloric intake. Moreover, starvation-induced protein loss is different from CAC-related protein loss. Starvation induces protein loss including protein degradation in both visceral tissue and skeletal muscle, while CAC only induces protein degradation in skeletal muscle (Tisdale, 2009, Argiles et al., 2014).

Ongoing loss of skeletal muscle mass is a key characteristic of CAC (Baracos et al., 2019). Suppressed protein synthesis and enhanced protein degradation play the major roles in CAC-associated skeletal muscle loss (Glass, 2010, Smith and Tisdale, 1993, Argiles et al., 2014). Protein synthesis in skeletal muscle is primarily controlled by insulin-like growth factor-1 (IGF-1) and its receptor (IGFR-1) signaling-induced PI3K/Akt/mTOR signaling pathway (Bodine et al., 2001b, Schiaffino et al., 2013). Activation of IGFR-1 by IGF-1 recruits and activates insulin receptor substrate (IRS), which in turn recruits and activates phosphatidylinositol-3 kinase (PI3K) to generate phosphoinositide-3,4,5-triphosphate (PIP3). PIP3 recruits PDK1 to induce the phosphorylation of Akt. Activated Akt induces the phosphorylation of mammalian target of rapamycin (mTOR) complex 1 (mTORC1), which in turn induces the phosphorylation of p70 ribosomal kinase S6 kinase (p70S6K) and the eukaryotic translation initiation factor-4E binding protein (4E-BP) to initiate the translation of mRNA for protein synthesis. Therefore, Akt/mTOR signaling is a crucial regulator of skeletal muscle atrophy (Bodine et al., 2001b). Impaired Akt/mTOR signaling is commonly found in cachectic cancer host, and is associated with compromised protein synthesis (Fearon et al., 2012, Duval et al., 2018).

The autophagy-lysosome system and ubiquitin-proteasome system are the major pathways that are associated with the skeletal muscle protein degradation and muscle atrophy in cachectic cancer condition (Baracos et al., 2018, Sandri, 2016). Autophagic degradation-associated muscle wasting has been identified in C26 carcinoma-bearing mice, LLC-bearing mice, and Yoshida AH-130 hepatoma-bearing rats (Penna et al., 2013). The ubiquitin-proteasome pathway is the predominant one that controls the degradation of proteins in skeletal muscle in cancer cachectic patients and animals (Khal et al., 2005). Muscle atrophy F-box protein (MAFbx) and muscle RING finger-1 (MuRF1), both are E3 ubiquitin ligases, are the key enzymes that catalyze protein degradation in skeletal muscle (Bodine et al., 2001a, Bodine and Baehr, 2014). These enzymes are up regulated in the skeletal muscle of most cachectic patients and experimental animals (Zhou et al., 2010, Baracos et al., 2018, Johns et al., 2013). The expression of MAFbx and MuRF, and the expression of autophagy-related protein LC3 are regulated by the myostatin/activin A signaling pathway and TNFα/NF-kB pathway (Trendelenburg et al., 2009, Cai et al., 2004, Li et al., 2005, Penna et al., 2013).

Myostatin, a member of TGF-β superfamily, is a negative regulator of skeletal muscle mass (McPherron et al., 1997, Zimmers et al., 2002). Myostatin and activin A are structurally related proteins and have the same ActRIIB receptor (Lee and McPherron, 2001). Blocking ActRIIB dramatically increases skeletal muscle mass without altering satellite cell function (Lee et al., 2012). The binding of myostatin or activin A to ActRIIB activates Smad2/3 via phosphorylation. Phosphorylated Smad2/3 recruits Smad4 to form a Smad complex, which translocate to the nucleus to induce the transcription of muscle atrophy-related genes including MAFbx (Sartori et al., 2009). The activation of ActRIIB also results in the dephosphorylation of Akt, which inhibits Akt/mTOR signaling and impairs protein synthesis and cell proliferation (Sartori et al., 2009, Trendelenburg et al., 2009, Hulmi et al., 2013, Amirouche et al., 2009). The dephosphorylation of Akt also leads to the dephosphorylation of FoxO1/3. Dephosphorylated FoxO1/3 will translocate into the nucleus to turn on the transcription of MAFbx, MuRF1, and LC3B (Sandri et al., 2004, Lee et al., 2011). The expression of myostatin is elevated in cancer patients and experimental animals bearing cachectic cancers, and blocking myostatin signaling significantly alleviates skeletal muscle loss (Nissinen et al., 2018, Toledo et al., 2016, Zhou et al., 2010). Inflammatory cytokines are the major mediator of CAC (Baracos et al., 2018). TNFα and IL-6 up-regulate the expression of MAFbx, MuRF-1, and LC3B through the activation of NF-kB pathway (Zhou et al., 2010, Cai et al., 2004, Penna et al., 2013).

Chronic alcohol consumption induces skeletal muscle atrophy in rats (Lang et al., 1999, Lang et al., 2007, Steiner and Lang, 2015, Gritsyna et al., 2017), enhances muscle wasting in SIV infected macaque monkeys (Molina et al., 2008), and promotes sarcopenia in human alcoholics with alcoholic liver disease (Dasarathy et al., 2017). Chronic alcohol consumption also enhances CAC in mice bearing B16BL6 melanoma (Núñez et al., 2002, Zhang et al., 2015a, Wang et al., 2017). However, the underlying molecular mechanism of how alcohol enhances CAC remains to be elucidated. It is also not known if chronic alcohol consumption only enhances CAC in B16BL6 melanoma-bearing mice, or it has the generality that chronic alcohol consumption enhances CAC in mice-bearing cachectic cancers. Since chronic alcohol consumption increases TNFα and activates NF-kB signaling (Maraslioglu et al., 2014, Iimuro et al., 1997), we hypothesize that chronic alcohol consumption enhances CAC in hosts bearing cachectic cancers through TNFα-mediated muscle atrophy. A mouse model of chronic alcohol consumption and Lewis lung carcinoma (LLC) implantation were used to test this hypothesis. We found that alcohol consumption enhanced CAC, accelerated skeletal muscle loss at the advanced stage of cancer, and decreased the survival of LLC-bearing mice. Alcohol consumption impaired Akt/mTOR signaling pathway, enhanced MAFbx, MuRF-1, and LC3 expression by activating Smad, FOXO and NF-kB signaling pathways. Myostatin was up regulated in the skeletal muscle, peripheral blood and tumor. Depletion of TNFα not only abrogated alcohol-induced up-regulation of myostatin, but also abolished alcohol-enhanced CAC and prolonged the survival of tumor-bearing mice. These results suggest that alcohol consumption enhances skeletal muscle wasting in cachectic tumor-bearing mice through modulating TNFα/myostatin axis.

Material and Methods

Experimental animals and alcohol administration:

TNFα KO mice with C57BL/6 background (B6;129S-Tnf^tm1Gkl/J, Stock No: 003008) were purchased from the Jackson Laboratory (Bar Harbor, ME), and bred at the WSU-Spokane PBS vivarium. Only female mice were used in this research project. Female C57BL/6 mice at 6–7 weeks of age were purchased from Charles River Laboratories (Wilmington, MA) and housed in a specific pathogen-free room at the WSU-Spokane PBS vivarium. Mice were housed in plastic cages with air-filter tops and CareFresh bedding, and allowed free access to Purina 5001 rodent laboratory chow and autoclaved Milli-Q water. After one week of acclimation, mice were randomly divided into two groups: one group was fed with 20% (w/v) alcohol as sole drinking fluid, which was diluted from 190-proof of EverClear (St. Louis, MO) with autoclaved Milli-Q water and filtered through a 0.45-μm filter; the other group was continuously provided with autoclaved Milli-Q water as a control. Both groups had free access to Purina 5001 rodent laboratory chow. After three months of alcohol consumption, mice were used for experiments. C57BL/6 mice are alcohol-preferring mice. In this model, the daily intake of 20% w/v alcohol was about 5.6 ± 1.0 ml/day (Zhang et al., 2012). The blood alcohol concentration was about 0.03 ± 0.004% (Zhang et al., 2012). When the mice in this chronic alcohol model were inoculated with B16BL6 melanoma, we found that there was no significant difference between non-tumor injected mice and tumor-bearing mice in daily intake of alcohol and the blood alcohol concentration (Zhang et al., 2012). This is a well-validated mouse alcohol consumption model (D’Souza El-Guindy et al., 2010). In this model, alcohol consumption does not induce liver injury (Abdallah et al., 1988). All the animal protocols used in this study were approved by the Institutional Animal Care and Use Committee at Washington State University.

Antibodies and reagents:

The following antibodies were purchased from Santa Cruz Biotechnology, Inc. (Dallas, TX): Akt (sc-8312), p-Akt (Thr-308) (sc-16646-R), p-Akt (Ser 473) (sc-7985-R), FOXO1 (sc-11350), p-FOXO1 (Thr 24) (sc-16561-R), p-FOXO1 (Ser 256) (sc-101681), p-FOXO1 (Ser 319) (sc-101682), GAPDH (sc-48166), MAFbx (sc-33782), MuRF1 (sc-32920), myostatin (sc-6884), NF-kB p50 (sc-114), (sc-Smad2/3 (FL-425) (sc-8332), p-Smad2/3 (Ser 423/425) (sc-11769), mTOR (sc-8319), p-mTOR (Ser 2448) (sc-101738), p38 (sc-535), p-p38 (Thr180/Tyr182) (sc-17852-R), beta-Tubulin (sc-5247). The following antibodies were purchased from BioLegend (San Diego, CA): Raptor (clone Poly6232), beta-Actin (Clone Poly6221), beta-Actin (Clone 2F1–1), ERK1/2 (Clone W15133B), p-ERK1/2 (Thr202/Thr204) (Clone 6B8B69), NF-kB p65 (Clone Poly6226). The following antibodies was purchased from BD Biosciences (San Diego, CA): PI3K (610045), IRS-1b (611394). p70S6K (Thr389) was purchased from Cell Signaling Technology, Inc. (Danvers, MA). czDa

Tumor cell culture and inoculation:

Lewis lung carcinoma (LLC) (ATCC) cells were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) supplemented with 10% FBS and 1% penicillin/streptomycin and incubated at 37°C, 5% CO2 in a humidified incubator. Cells were harvested at 70–90% confluence. Single cell suspension was made in sterilized PBS at the concentration of 5×10⁶ cells/ml. Tumor cells were inoculated in the right hip area at 10⁶ cells/mouse.

Tumor size and body weight measurement:

For the survival experiments, tumors were measured by a caliper every other day from day 7 when tumor was palpable. Tumor size was calculated with the formula $\text{V}=\frac{\pi }{6}*1.58*{\left(a*b\right)}^{3/2}$, where a is the length of the tumor, b is the width of the tumor (Feldman et al., 2009). Body weight was measured on the days of tumor inoculation (Day 0) and at the 3-week and 4-week time points after tumor inoculation. For survival experiments, body weight on day 0, end-point body weight, with and without tumor, was measured at necropsy.

Skeletal muscle, tumor and blood collection:

At the indicated time points, different cohorts of mice were euthanized. Tibialis anterior (TA), Gastrocnemius (GA), and Vatus lateralus (VA) muscles were collected and the weights were measured. Tumor was also collected and tumor weight was measured. The muscle and tumor samples were immediately stored in liquid nitrogen and then stored to −80°C for future use. Blood samples were collected by cardiac puncture. Plasma was prepared by centrifugation of heparinized blood at 5000 rpm for 3 min, and stored at −80°C for future use.

Western blot:

Skeletal muscle and tumor samples were prepared by using TissueLyzer II (QIAGEN) to homogenize samples in RIPA lysis and extraction buffer supplemented with protease inhibitors and phosphatase inhibitors (Sigma, St. Louis, MO) following the manufacturer’s instruction. The homogenate was incubated at 4°C for 2 h, and then centrifuged at 12000g, 4°C for 15 min. The supernatant was transferred into a fresh tube. Protein concentration was determined by Bradford Protein Assay. Proteins were separated by SDS-PAGE and transferred to a nitrocellulose membrane. The membrane was blocked with 5% BSA blocking buffer for 1 h and rinsed with TBST buffer. The membrane was incubated with primary antibody at 4°C overnight. After 3 washes with TBST buffer, the membrane was incubated with HRP-conjugated secondary antibody at room temperature for 1 h. The membrane was washed with TBST buffer and then developed with SuperSignal™ West Pico PLUS Chemiluminescent Substrate (ThermoFisher Scientific) and analyzed by Bio-Rad Gel Documentation System.

Statistical analysis:

Data were analyzed by Microsoft Excel and GraphPad Prism. Unpaired two-tailed Student’s t-test or two-way ANOVA with Tukey’s multiple comparisons test were used to determine the significance of difference between and among groups as appropriate. The log-rank (Mastel-Cox) test was used to analyze difference in survival between and among experimental groups. Difference between groups was considered significant at p<0.05.

Results

Chronic alcohol consumption enhances CAC, accelerates skeletal muscle loss and decreases the survival of mice bearing LLC tumor.

Published research, including ours, indicated that chronic alcohol consumption enhances CAC in mice bearing B16BL6 melanoma (Wang et al., 2017, Zhang et al., 2015a, Núñez et al., 2002). To determine the generality of alcohol consumption enhancing CAC, we examined if chronic alcohol consumption enhances CAC in mice bearing LLC tumor, which is a well-known cachectic cancer (Puppa et al., 2014). The body weight was measured before tumor inoculation (day 0), and on day 21 and day 28 after tumor inoculation for the muscle sample harvesting groups, and at necropsy for survival groups. Results indicated that chronic alcohol consumption did not affect body weight in mice without tumor (day 0) (Fig. 1A), which is consistent with our previous results (Zhang et al., 2015a). At necropsy the body weight without tumor was significantly lower than the body weight before tumor inoculation in both groups of mice (Fig. 1A). The body weight at necropsy was significantly lower in alcohol-consuming mice than in water drinking mice both with and without tumor (Fig. 1A, 1B). There was no difference in tumor growth between the two groups of mice from day 7 to day 19 after tumor inoculation (Fig. 1C). The tumor weight at necropsy was significantly higher in water-drinking mice than in alcohol-consuming mice (Fig. 1D). This could result from the longer survival time of watering-drinking mice compared to alcohol-consuming mice. Compared to the body weight before tumor inoculation, the body weight without tumor at necropsy decreased 12.5% and 22.2% in water-drinking and alcohol-consuming mice, respectively. The body weights on day 21 and day 28 after tumor inoculation were not significantly different from the body weight before tumor inoculation in water-drinking mice. In alcohol-consuming mice, body weight decreased 5.4% on day 21 compared to day 0, but it was not statistically significant; while the body weight decreased 14% on day 28 compared to day 0, and it was statistically significant (Fig. 1E). There was no difference in the weight of skeletal muscle GA, TA and VA on day 21 between water-drinking mice and alcohol-consuming mice (Fig. 1F). The weight of GA and TA on day 28 was significantly lower in alcohol-consuming mice compared to water-drinking mice, but there was no difference in VA between the two groups of mice (Fig. 1G). The survival was significantly decreased in alcohol-consuming and tumor-bearing mice compared to water-drinking and tumor-bearing mice (Fig. 1H). The median survival time was 29 and 37 days in alcohol-consuming tumor-bearing mice and water-drinking and tumor-bearing mice, respectively (Fig. 1H). Collectively, these data indicated that alcohol consumption did not affect body weight or muscle mass in non-tumor injected mice, however, LLC tumor induced CAC in mice, chronic alcohol consumption significantly enhanced CAC and decreased the survival of mice bearing LLC tumor.

Figure 1. Effects of chronic alcohol consumption on body weight, skeletal muscle and survival of mice bearing LLC.

After 3 months of alcohol consumption, mice were inoculated subcutaneously with LLC cancer cells. A. Showing body weight on the day of tumor inoculation (Day 0) and at necropsy with tumor removed. Two-way ANOVA with Tukey’s multiple comparison test was used to determine the difference between groups. **p<0.01, ***p<0.001, ****p<0.0001. B. Showing body weight with tumor at necropsy. Two tailed Student’s t-test was used to determine the difference between the two groups of mice. ***p<0.001. C. Tumor size at the indicated days after tumor inoculation. D. Tumor weight at necropsy. Two tailed Student’s t-test was used to determine the difference between the two groups of mice. **p<0.01. E. Showing body weight without tumor at the indicated dates after tumor inoculation. Difference between groups was analyzed by Two-ANOVA with Tukey’s multiple comparison. *p<0.05. F and G. Showing weight of the indicated skeletal muscle at the 3-week and 4-week points after tumor inoculation, respectively. Student’s T-test was used to determine the significance of difference between groups. *p<0.05, **p<0.01. H. Showing the survival curves of two groups of mice. Log-rank (Mantel-Cox) test was used to determine the difference between the two groups. The number in parentheses is the median survival. Data=Mean±SD. Each group contained 10–12 mice. Water=Water-drinking mice. EtOH=Alcohol-consuming mice.

Chronic alcohol consumption does not affect the expression of PI3K, but inhibits the phosphorylation and expression of Akt, mTOR, Raptor and p70S6K signaling molecules in GA of LLC tumor-bearing mice.

Since chronic alcohol consumption enhanced CAC and accelerated skeletal muscle loss, we next sought to determine if this alcohol-enhanced muscle loss results from impaired the signaling pathways related to protein synthesis in skeletal muscle of LLC tumor-bearing mice. We used western blot to determine the expression and phosphorylation of signaling molecules governing protein synthesis in skeletal muscle GA at 4 weeks after tumor inoculation. Results indicated that alcohol consumption slightly increased IRS-1b expression, but did not affect P13K expression in GA muscle of LLC tumor-bearing mice (Fig. 2A, 2B). Chronic alcohol consumption significantly inhibited the phosphorylation of Akt (Ser 473) and mTOR (Ser 2448), and down-regulated the expression of Raptor and p70S6K (Fig. 2A, 2B). These results suggested that alcohol consumption impaired Akt/mTOR/Raptor/p70S6K signaling pathway, which could compromise protein synthesis in GA of tumor-bearing mice. However, alcohol consumption did not affect the upstream PI3K signaling in LLC tumor-bearing mice.

Figure 2. Chronic alcohol consumption inhibited Akt/mTOR signaling pathway in the skeletal muscle of mice bearing LLC.

After 3 months of alcohol consumption, mice were inoculated subcutaneously with LLC cancer cells. Skeletal muscle samples were collected after 4 weeks of tumor inoculation. The expression and phosphorylation of the indicated proteins were determined by western blotting. Beta-actin and tubulin were used as loading control. A. Representative western blot shows the expression and phosphorylation of the indicated protein in GA muscle. B. Relative expression of the indicated protein to internal control or the phosphorylation of the indicated protein to their respective general protein expression. *p<0.05. Data=Mean±SD. Data are from one representative experiment of two independent experiments. In each independent experiment, each group contained 5 mice. Water=Water-drinking mice. EtOH=Alcohol-consuming mice.

The expression of MuRF-1, MAFbx, and LC3-I/II is up-regulated in the skeletal muscle of alcohol-consuming and tumor-bearing mice.

One of the major causes of muscle wasting in cancer cachexia is enhanced protein degradation. MuRF-1 and MAFbx are the key enzymes in skeletal muscle that govern ubiquitin-mediated proteolysis (Bodine and Baehr, 2014). Enhanced muscle fiber autophagy is another cause of muscle wasting in cancer-cachexia patients, and elevated LC3, specifically LC3-II expression, is the biomarker of enhanced muscle autophagy (Aversa et al., 2016). We next determined if alcohol consumption affects MuRF-1, MAFbx, and LC3-I/II expression in skeletal muscle GA of LLC tumor-bearing mice. Results indicated that the expression of MuRF-1, MAFbx, and LC3-I/II, especially LC3-II, were significantly increased in GA muscle 4 weeks after tumor inoculation (Fig. 3A, B, C). These results suggest that alcohol consumption enhances the signaling governing ubiquitin-mediated proteolysis and LC3-mediated muscle autophagy in LLC tumor-bearing mice.

Figure 3. Chronic alcohol consumption enhanced the expression of MuRF-1, MAFbx, and LC3-I/II in GA muscle after 4-week of LLC tumor inoculation.

A. Representative western blot shows the expression of the indicated protein in GA. Tubulin was used as a loading control. B. Relative expression of MuRF-1 and MAFbx to tubulin in GA. C. Relative expression of LC3II to tubulin in GA. *p<0.05. Data=Mean±SD. Data are from one representative experiment of two independent experiments. In each independent experiment, each group contained 5 mice. Water=Water-drinking mice. EtOH=Alcohol-consuming mice.

Chronic alcohol consumption activates FOXO and Smad signaling pathways.

The expression of MuRF-1, MAFbx, and LC3-I/II in skeletal muscle is tightly regulated by transcription factors FOXO and Smad. FOXO transcriptional activity plays critical roles in muscle atrophy during cachexia (Reed et al., 2012). The dephosphorylation of FOXO1 induces the translocation of FOXO1 from cytoplasm to nucleus to activate the transcription of MuRF-1, MAFbx, and LC3-I/II (Mammucari et al., 2007, Sandri et al., 2004, Sartori et al., 2009). The phosphorylation of Smad2/3 facilitates the formation of Smad2/3/4 complex. This complex will translocate from cytoplasm to nucleus to direct the transcription of MAFbx and LC3-I/II (Egerman and Glass, 2014). The phosphorylation of Smad2/3 also inhibits the phosphorylation of Akt, which in turn inhibits the phosphorylation of FOXO1/3 and enhances the transcription of MuRF-1, MAFbx and LC3-I/II (Sartori et al., 2009, Sandri et al., 2004, Trendelenburg et al., 2009). We next determined if alcohol consumption affected the phosphorylation of FOXO1 and Smad 2/3. Results indicated that alcohol consumption enhanced Smad2/3 (Ser423/425) phosphorylation (Fig. 4A, 4B) and inhibited the phosphorylation of FOXO1 (Thr24) (Fig. 4A, 4C) in GA muscle 4 weeks after LLC tumor inoculation.

Figure 4. Effects of chronic alcohol consumption on the phosphorylation of FOXO1 and Smad2/3 in GA muscle after 4-week of LLC tumor inoculation.

A. Western blot shows phosphorylated FOXO1 and Smad2/3 in GA. Actin was used as loading control. B. Ratio of phosphorylated Smad2/3 to total Smad2/3. C. Ratio of phosphorylated FOXO1 to actin. Data=Mean±SD. Data are from one representative experiment of two independent experiments. In each independent experiment, each group contained 5 mice. Water=Water-drinking mice. EtOH=Alcohol-consuming mice.

Chronic alcohol consumption enhances the expression of myostatin in skeletal muscle, tumor, and blood of tumor-bearing mice.

Myostatin plays a key role in regulating protein synthesis in skeletal muscle by modulating the Akt/mTOR/p70S6K pathway, and modulating protein degradation by regulating Smad/FOXO/MuRF-1/MAFbx/LC3 signaling pathways (Trendelenburg et al., 2009, Egerman and Glass, 2014). Skeletal muscle wasting in CAC is tightly associated with the up-regulation of myostatin (Costelli et al., 2008, Smith et al., 2015, Gallot et al., 2014). We next determined if the expression of myostatin was altered by alcohol consumption in LLC tumor-bearing mice. Results indicated that alcohol consumption significantly increased the expression of myostatin in GA muscle, tumor, and blood (Fig. 5A, 5B). In non-tumor-injected mice alcohol consumption did not affect myostatin expression in the skeletal muscle (Figure 5C), which is consistent with results that alcohol consumption did not affect the body weight of non-tumor-injected mice (Fig. 1A).

Figure 5. Chronic alcohol consumption up-regulated the expression of myostatin in tumor, skeletal muscle and plasma.

After 3 months of alcohol consumption, mice were inoculated subcutaneously with LLC cancer cells. Skeletal muscle, tumor and blood samples were collected after 4 weeks of tumor inoculation. A. Representative western blot shows the expression of myostatin in tumor (upper panel), skeletal muscle (middle panel) and plasma (lower panel). Actin and tubulin were used as loading control in tumor and muscle, transferrin was used as loading control in plasma. B. Relative expression of myostatin in the indicated tissues. C. Expression of myostatin in skeletal muscle of mice without tumor. *p<0.05, ***p<0.001. Data=Mean±SD. Data are from one representative experiment of two independent experiments. In each independent experiment, each group contained 5 mice. WT=Water-drinking and tumor-bearing mice. ET=Alcohol-consuming and tumor-bearing mice. WC=Water-drinking and non-tumor-injected mice. EC= Alcohol-consuming and non-tumor-injected mice.

Chronic alcohol consumption activates ERK, p38, and NF-kB signaling pathways.

The expression of MuRF-1, MAFbx and LC3-I/II can also be regulated by the inflammatory cytokine-activated MAPK and NF-kB pathways (Li et al., 2005, Schakman et al., 2012). We next determined if the ERK, p38 and NF-kB signaling pathways were enhanced by chronic alcohol consumption. Results indicated that the expression of NF-kB, and the phosphorylation of p38 (Thr180/Tyr182) and ERK (Thr202/Tyr204) were significantly enhanced in the GA muscle of alcohol-consuming mice compared to water-drinking mice 4 weeks after tumor inoculation (Fig. 6A, 6B). These results suggested that alcohol consumption enhanced the activation of NF-kB and MAPK pathways in skeletal muscle of LLC tumor-bearing mice.

Figure 6. Chronic alcohol consumption increased the expression and phosphorylation of ERK, p38 and NF-kB in skeletal muscle 4 weeks after LLC tumor inoculation.

A. Representative western blot shows expression and phosphorylation of ERK, p38 and NF-kB in GA. Actin was used as internal control. B. Relative expression of phosphorylated p38 and ERK to their respective total protein and NF-kB to actin. **p<0.01, ***p<0.001. Data=Mean±SD. Data are from one representative experiment of two independent experiments. In each independent experiment, each group contained 5 mice. Water=Water-drinking mice. EtOH=Alcohol-consuming mice.

Depletion of TNFα abrogates alcohol consumption-enhanced myostatin and MAFbx expression and the enhanced-phosphorylation of Akt, FOXO1 and p38 in skeletal muscle of LLC tumor-bearing mice.

Inflammatory cytokines, especially TNFα, play the key role in the induction of CAC. We next used TNFα KO mice to determine if depletion of TNFα affects the expression and activation of key signaling molecules associated with CAC. Results indicated that depletion of TNFα abrogated the effects of alcohol consumption-enhanced myostatin and MAFbx expression, and alcohol-modulated Akt, Smad, FOXO1, p38 phosphorylation in the GA of LLC-bearing mice after four weeks of tumor inoculation (Fig. 7A, 7B). Consistent with these signaling pathway changes, alcohol consumption-enhanced skeletal muscle (GA and TA) loss after four weeks of tumor inoculation was abrogated in TNFα KO mice (Fig. 7C, 7D). Alcohol consumption did not affect TNFα KO mice development, which is evidenced by the fact that there was no difference in the length of tibia between the two groups of mice (Fig. 7E). These results suggest that TNFα is the cytokine that dominates alcohol consumption-induced myostatin-mediated skeletal muscle wasting in mice bearing LLC tumor.

Figure 7. Effects of chronic alcohol consumption on signaling proteins related to CAC in GA muscle of TNFα KO mice bearing LLC tumor.

After 3 months of alcohol consumption, TNFα KO mice were inoculated subcutaneously with LLC cancer cells. Skeletal muscle samples were collected after 4 weeks of tumor inoculation. A. Representative western blot shows the expression and phosphorylation of the indicated proteins in the skeletal muscle GA. GAPDH and tubulin were used as loading control. B. Relative expression and phosphorylation of the indicated protein. C. Indicated muscle weight after 4 weeks of tumor inoculation. D. ratio of indicated muscle weight to tibia length after 4 weeks of tumor inoculation. E. Tibia length. Data=Mean±SD. Data are from one representative experiment of two independent experiments. In each independent experiment, each group contained 5 mice. Water=Water-drinking mice. EtOH=Alcohol-consuming mice.

Depletion of TNFα abolishes alcohol consumption-enhanced body weight loss and significantly increases the survival time of mice bearing LLC tumor.

We further determined if depletion of TNFα affected body weight and survival of LLC tumor bearing mice. Results indicated that chronic alcohol consumption neither affected the body weight of mice before tumor inoculation, nor affected the body weight of mice at necropsy. Although the body weight of mice at necropsy was lower than before tumor inoculation in both water-drinking mice and alcohol-consuming mice, it was not statistically significantly (Fig. 8A). The magnitude of body weight loss of TNFα KO mice bearing LLC tumor was significantly lower than their wild-type counterparts of water-drinking or alcohol-consuming mice, which was more pronounced in the alcohol-consuming mice (Fig. 8B). Tumor growth tended to be slower in alcohol-consuming TNFα KO mice compared to their water-drinking counterparts, but it was not statistically significant (Fig. 8C). The survival time of alcohol-consuming tumor-bearing TNFα KO mice also tended to be longer than their water-drinking counterparts, but it was also not statistically significant (p=0.06) (Fig. 8D). Depletion of TNFα did not significantly increase the survival of water-drinking and tumor-bearing mice compared to their wild-type counterparts (Fig. 8D). However, in alcohol-consuming mice it significantly increased the survival of TNFα KO mice compared to their wild-type counterparts (Fig. 8D). These results suggest that depletion of TNFα abrogates alcohol consumption-enhanced CAC and increases the survival of tumor-bearing mice.

Figure 8. Depletion of TNFα abrogates alcohol consumption-enhanced CAC and increases the survival of LLC tumor-bearing mice.

A. Body weight before tumor inoculation and at necropsy of TNFα KO mice. B. Percentage of body weight change at necropsy to body weight before tumor inoculation of wild type mice and TNFα KO mice. C. Tumor size of LLC tumor in TNFα KO mice on the indicated days after tumor inoculation. D. Survival of wild type type mice and TNFα KO mice. Log-rank (Mantel-Cox) test was used to determine the difference between the two groups. The number in parentheses is the median survival. Data=Mean±SD. Each group contained 8–12 mice. Water=Water-drinking mice. EtOH=Alcohol-consuming mice.

Discussion

Our present research clearly indicated that chronic alcohol consumption enhanced CAC in LLC-bearing mice. These results combined with the previously published research that alcohol consumption also enhances CAC in B16BL6 melanoma-bearing mice suggest that chronic alcohol consumption have the generality of enhancing CAC in cachectic tumor-bearing mice. Alcohol-enhanced cancer-associated skeletal muscle loss in LLC-bearing mice could mainly result from impaired protein synthesis and enhanced protein degradation in skeletal muscle. This is based on the following facts: 1) alcohol consumption impaired Akt/mTOR/p70S6K signaling, which is the pathway that is responsible for protein synthesis; 2) alcohol consumption up regulates the expression of MAFbx, MRUF-1 and LC3, which are the key molecules that governing protein degradation through ubiquitin-proteasome proteolytic pathway and autophagic pathway in skeletal muscle. The dysregulation of protein synthesis and degradation signaling pathways in the skeletal muscle of alcohol-consuming and LLC-tumor-bearing mice was associated with the overexpression of myostatin. TNFα was the key mediator of alcohol-enhanced skeletal muscle loss in LLC-bearing mice, which was evidenced by the fact that depletion of TNFα abrogated alcohol-induced overexpression of myostatin and prevented the body weight loss of LLC tumor-bearing mice. Therefore, alcohol consumption enhanced cancer-associated skeletal muscle loss through modulating the TNFα/myostatin axis (Fig. 9).

Figure 9.

Mechanistic scheme shows how chronic alcohol consumption modulates signaling pathways governing protein synthesis and degradation to enhance CAC-associated skeletal muscle loss through TNFα and myostatin axis.

The homeostasis of skeletal muscle mass is mainly balanced by skeletal muscle protein synthesis and degradation. Inhibition in protein synthesis and/or increase in protein degradation lead to muscle mass loss. The major signaling pathway that controls skeletal muscle protein synthesis is Akt/mTOR signaling. The upstream signaling of Akt/mTOR is IGFR-1, IRS and PI3K (Schiaffino and Mammucari, 2011). In the present study, we found that alcohol consumption did not alter the expression of PI3K but even slightly increased the expression of IRS in skeletal muscle after 4 weeks of tumor inoculation (Fig. 2). However, the expression and phosphorylation of Akt, mTOR, Raptor and p70S6K were significantly inhibited, which suggested that inhibition by alcohol on Akt/mTOR signaling in the skeletal muscle of LLC tumor-bearing mice was not through IGFR-1/IRS/PI3K. It is well documented that the activation of myostatin/ActRII signaling could inhibit Akt/mTOR signaling through the phosphorylation of Smad2/3 (Trendelenburg et al., 2009). Indeed, we found that the protein level of myostatin in tumor, blood, and skeletal muscle was significantly higher in the alcohol-consuming and tumor-bearing mice than in their water-drinking counterparts (Fig. 5). Taken together, these results suggested that chronic alcohol consumption inhibited Akt/mTOR/Raptor/p70S6K likely through enhancing the activation of myostatin/Smad2/3 signaling pathway.

The activation of myostatin signaling not only inhibits protein synthesis through suppressing Akt/mTOR signaling, but also enhances protein degradation through activating ubiquitin-mediated proteolytic degradation and LC3-associated autophagy (Lee et al., 2011). Indeed, the expression of MuRF-1, MAFbx, and LC3-II in GA of alcohol-consuming and tumor-bearing mice was significantly increased. The expression of MAFbx and LC3 is regulated by FOXO family transcription factors (Sengupta et al., 2009). The dephosphorylation of FOXO1/3 facilitates the translocation of these transcription factors from cytosol to nucleus to initiate target gene including MAFbx expression. Akt plays a key role in controlling FOXO1/3 phosphorylation (Tzivion et al., 2011). Since alcohol consumption inhibited Akt phosphorylation, it would also inhibit FOXO1/3 phosphorylation. This was what we observed in the GA of alcohol-consuming and tumor-bearing mice. Therefore, alcohol enhanced the activation of myostatin/Smad2/3/Akt/FOXO signaling pathway to accelerate protein degradation through enhancing MAFbx-mediated proteolysis and LC3II-associated autophagy (Fig. 9).

Alcohol consumption not only up-regulated the expression of MAFbx in the skeletal muscle of LLC-bearing mice, but also up-regulated the expression of MuRF-1 in GA muscle. Both MAFbx and MuRF-1 belong to the E-ubiquitin ligase family and play important roles in ubiquitin-mediated proteolytic degradation of skeletal muscle protein (Bodine and Baehr, 2014). The expression of MAFbx is regulated by myostatin/Smad signaling. However, it was reported that neither myostatin nor phosphorylation of Smad3 could directly regulate MuRF-1 expression (McFarlane et al., 2006, Goodman et al., 2013). The expression of MuRF-1 and MAFbx can be regulated by TNFα, MAPK and NF-kB pathways (Li et al., 2005). Alcohol consumption did enhance the activation of MAPK and NF-kB signaling which was evidenced by the enhanced phosphorylation of p38 and ERK1/2 and the up-regulation of NF-kB.

Inflammatory cytokines, such as TNFα and IL-6, are the key mediator of CAC (Patel and Patel, 2017, Carson and Baltgalvis, 2010). We previously found that chronic alcohol consumption significantly increased TNFα and IL-6 expression in B16BL6 melanoma-bearing mice (Wang et al., 2017, Zhang et al., 2015c). These cytokines could be the major mediator of alcohol-enhanced CAC in LLC tumor-bearing mice. Because alcohol consumption enhanced the expression of myostatin, and myostatin modulates protein synthesis and degradation in skeletal muscle as discussed above, we would like to determine if the upregulation of myostatin in alcohol-consuming and tumor-bearing mice was regulated by these inflammatory cytokines. It was reported that TNFα, but not IL-6, could induce the expression of myostatin through activation of NF-kB (Zhang et al., 2011). Myostatin induces IL-6 expression (Zhang et al., 2011). Therefore, we used TNFα KO mice to determine if depletion of TNFα could abrogate the up-regulation of myostatin and mitigate alcohol-enhanced CAC. Indeed, depletion of TNFα not only abrogated alcohol-enhanced up-regulation of myostatin and CAC-related signaling pathways, but also abolished CAC. More importantly, depletion of TNFα significantly increased survival of alcohol consuming and tumor-bearing mice compared to their wild type counterparts.

It should be noted that in the water-drinking mice depletion of TNFα mitigated CAC, but did not significantly increase survival compared to the wild type mice (median survival: 44 vs 37, p=0.13). However, in alcohol-consuming mice, depletion of TNFα not only significantly inhibited CAC, but also significantly increased survival compared to the wild type mice (median survival: 53 vs 29, p<0.0001). For TNFα KO mice, the survival time of alcohol-consuming LLC-bearing mice tended to be longer than their water-drinking counterparts were (55 vs 44, p=0.06). Although the underlying mechanism is not fully understood now, we believe it is at least partly associated with the immune system. We previously found that chronic alcohol consumption activated the immune system in the steady state, which in turn inhibited melanoma growth at the early stage of tumor progression (Zhang and Meadows, 2005, Zhang et al., 2015b, Zhang et al., 2015a). However, with tumor progression alcohol consumption accelerated immune dysfunction (Zhang and Meadows, 2010). Immunization could enhance antitumor immunity, however, accelerated immune dysfunction and enhanced CAC still decreased survival of alcohol-consuming and tumor-bearing mice (Zhang et al., 2015a). Multiple factors, such as increased myeloid-derived suppressor cells and skewed iNKT cell cytokine profile, are involved in alcohol-induced immune dysfunction (Zhang and Meadows, 2010, Zhang et al., 2015c); the precise mechanism remains to be elucidated. It was reported that TGF-β family members, such as activin A, inhibit NK cell antitumor immunity through activating Smad signaling pathway in NK cells (Robson et al., 2009, Gao et al., 2017). Myostatin and activin share activin receptor type IIB (ActRIIB). Both play important role in CAC through activating ActRIIB/Smad signaling pathway (Zhou et al., 2010). In the present study we found that alcohol consumption increased myostatin in skeletal muscle, tumor, and blood (Fig. 5A, 5B), but did not alter activin expression (data not shown). It would be reasonable to speculate that the increased myostatin could also activate Smad signaling pathway in NK cells to accelerate NK cell dysfunction in tumor-bearing mice. Depletion of TNFα will not only mitigate CAC, but also alleviate the dysfunction of the immune system. With the advantage of alcohol consumption activating the immune system, alcohol consumption would benefit antitumor immunity and the survival of tumor-bearing mice as we observed in the TNFα KO mice. This hypothesis will be investigated in the future.

In summary, chronic alcohol consumption accelerates CAC, especially skeletal muscle loss in mice-bearing cachectic cancers. Alcohol-enhanced cancer-associated skeletal muscle loss is mediated by TNFα. TNFα up-regulates myostatin expression, which in turn inhibits Akt/mTOR signaling to impair protein synthesis, and activates Samd2/3 and FOXOs to enhance MAFbx-mediated proteolysis and LC3-associated autophagy (Fig. 9). Therefore, TNFα-myostatin axis plays a key role in alcohol-enhanced cancer-associated skeletal muscle loss. Depletion of TNFα can abrogate CAC and increase the survival of alcohol-consuming and tumor-bearing mice.