5-Fluorouracil-based chemotherapy disrupts autophagy flux and protein synthesis in cultured myotubes: a role for mTORC1 signaling

Abstract

5-Fluorouracil-based chemotherapies, such as FOLFOX (5-fluorouracil, leucovorin, oxaliplatin), are used to treat colon cancer but also induce skeletal muscle toxicities. While FOLFOX can disrupt muscle autophagy signaling in vivo, the upstream mechanisms underlying this effect, as well as the roles of mTORC1 and AMPK signaling, require further investigation. We investigated whether 5-FU, 5-FU + leucovorin, oxaliplatin, or the combined FOLFOX regimen disrupts protein synthesis or autophagy flux through altered mTORC1/AMPK signaling in C2C12 myotubes with or without tumor cell-conditioned media, and whether these effects were readily reversible. We also examined the effects of rapamycin or metformin administration on FOLFOX-induced disruptions to muscle protein synthesis and autophagy. FOLFOX was added to myotube cultures for 24 hours; recovery was assessed by removing FOLFOX from the culture media for an additional 24 hours. Cultured Colon-26 tumor cells (CT26) were used for conditioned media. Autophagy flux was assessed using Bafilomycin A1 and LC3BII/I immunoblotting. FOLFOX and CT26 each decreased myotube diameter and protein synthesis compared to the vehicle. CT26, but not FOLFOX, increased STAT3 phosphorylation. FOLFOX, but not CT26, increased Atrogin-1 protein, while MURF-1 protein was increased by both CT26 and FOLFOX. FOLFOX reduced autophagy flux, AMPK, and ULK1(S555) phosphorylation, which did not recover upon FOLFOX removal. Rapamycin, but not metformin, altered autophagy flux in FOLFOX-treated myotubes and restored suppressed AMPK phosphorylation without increasing protein synthesis. We report that FOLFOX and oxaliplatin disrupt myotube mTORC1/AMPK regulation of autophagy flux, which was not easily reversible. However, rapamycin treatment increased autophagy flux in FOLFOX-treated myotubes.

Graphical Abstract

NEW AND NOTEWORTHY

FOLFOX chemotherapy is a common treatment for colorectal cancer, consisting of a combination of the antimetabolite 5-FU, leucovorin, and oxaliplatin. The disruption of muscle cellular signaling caused by 5-FU-based chemotherapies, which leads to fatigue and weakness, remains poorly understood. Our results provide evidence that FOLFOX can directly suppress myotube AMPK signaling, protein synthesis, and autophagy flux. Importantly, the effects of FOLFOX and the resulting disruptions to muscle protein turnover are long-lasting and not easily reversible.

INTRODUCTION

During the past 25 years, colorectal cancer incidence in the United States has continued to increase in individuals under the age of 50 (1). 5-Fluorouracil (5-FU) based chemotherapy regimens are treatment options that improve colorectal cancer patient survival (2). 5-FU has greater efficacy when combined with other drugs, such as leucovorin, irinotecan, and oxaliplatin (3). Over 70% of late-stage colorectal cancer patients are treated by a combined FOLFOX regimen that contains 5-FU, leucovorin, and oxaliplatin (4). While the combined treatment regimen inhibits colon tumor growth, there are non-specific actions in tissues throughout the body, causing systemic toxicities that contribute to adverse side effects (5). 5-FU and oxaliplatin induce toxicities in skeletal muscle that disrupt protein turnover and metabolic homeostasis in vivo and in vitro (6–8). Given skeletal muscle’s critical role in locomotion, strength, and systemic metabolic regulation, chemotherapy-induced muscle toxicities are associated with decreased health and quality of life in cancer patients after treatment (9).

Cancer and treatment can disrupt several critical skeletal muscle processes involving the regulation of protein turnover and metabolism (2, 10–13). Preclinical studies have demonstrated that chemotherapy treatments (e.g., doxorubicin, cisplatin, FOLFIRI, FOLFOX) alter skeletal muscle mitochondria biogenesis, quality control, and function, and induce ubiquitin-proteosome protein degradation (14–20). Tumor-derived factors also regulate skeletal muscle metabolism and protein turnover in vitro and in vivo (21). Autophagy has a critical role in muscle protein turnover and metabolism regulation by facilitating bulk protein breakdown via lysosomal degradation. These products are a source of recycled components that can be used for cellular metabolic processes (22). Within the tumor, dysregulation of autophagy is associated with growth and resistance to anticancer treatments (23). We have previously shown that FOLFOX administration to cultured myotubes can suppress autophagy flux, protein synthesis, and AMPK (5’AMP-activated protein kinase) phosphorylation (6). However, it is not certain if these direct effects on cultured muscle cells are lasting or readily reversible. Interestingly, we have found that tumor-derived factors from colon-26 cells (CT26) do not suppress myotube markers of autophagy (e.g., LC3B and p62) even though they induce significant atrophy and suppress protein synthesis (24). There is a gap in understanding the combined effects of FOLFOX chemotherapy and colon tumor-derived factors on myotube autophagy flux and protein synthesis regulation.

The mTORC1 (mammalian target of rapamycin complex I) signaling is a crucial component for integrating multiple stimuli, including nutrient or growth factor signaling, to regulate muscle protein turnover and growth (13). Downstream effectors of mTORC1 can suppress autophagy and activate protein synthesis through the activity of ULK1 (Unc-51-like kinase1), rpS6 (ribosomal protein S6), and others (11). AMPK signaling can regulate autophagy through mTORC1 inhibition and ULK1 phosphorylation (25, 26). In preclinical models, cancer and treatment can disrupt the coordinated regulation of muscle protein synthesis and autophagy in vivo and in vitro by disrupting mTORC1 and AMPK signaling (6, 27). Interestingly, mTORC1 hyperactivation is also detrimental to skeletal muscle and has been reported with aging, chronic diseases, and cancer (28). Hyperactivation of mTORC1 signaling in preclinical models of aging is associated with reduced hindlimb skeletal muscle mass (29). In vitro experiments using myotubes co-cultured with colon-26 cells report decreased protein synthesis, autophagy flux, and suppressed mTORC1 signaling transduction, rpS6 phosphorylation (30). Four cycles of FOLFOX administration in vivo can reduce skeletal muscle rpS6, LC3BII protein expression, which coincided with suppressed AMPK phosphorylation and ULK1 (S555) phosphorylation (6). These findings provide indirect in vivo evidence that repeated FOLFOX administration can disrupt skeletal muscle AMPK signaling and markers of autophagy. While there is evidence that 5-FU-based chemotherapies disrupt skeletal muscle autophagy regulation, there are still gaps in understanding the role of disrupted mTORC1 and AMPK signaling in this regulation, and whether these signaling pathways can be targeted for therapeutic benefit. This knowledge is critical to understanding if these signaling pathways can be targeted to prevent cancer and treatment-induced toxicities to skeletal muscle.

Disruptions to skeletal muscle protein turnover, involving aberrant mTORC1 and AMPK signaling, have been reported as critical drivers of cancer and of treatment-induced metabolic dysfunction and cachexia (31–33). The intermittent activation of autophagy, whether by exercise, fasting, or pharmacological compounds (e.g., metformin, rapamycin), can extend lifespan and may even modulate the anti-cancer effects of chemotherapy (34–39). Rapamycin and metformin are FDA-approved drugs that modulate the activity of mTORC1 and AMPK, respectively; however, data are equivocal regarding their roles in regulating cellular autophagy (40–42). The direct effects of FOLFOX chemotherapy on the regulation of muscle protein synthesis and autophagy by mTORC1 and AMPK are not well-established. It is not known whether modulating autophagy or protein synthesis with metformin or rapamycin can attenuate chemotherapy-induced dysregulation of metabolic signaling in muscle. While acute 5-FU-based chemotherapy-induced muscle toxicities have been extensively investigated, there is a limited understanding of whether these effects are readily reversible or lasting. Understanding these knowledge gaps can identify treatment options for muscle wasting and enhancing functional recovery in cancer patients after treatment. The purpose of this study was to investigate whether 5-FU, oxaliplatin, or the combined FOLFOX chemotherapy regimen directly disrupt either protein synthesis or autophagy flux in cultured C2C12 myotubes with or without tumor-conditioned media. We also explored the potential for rapamycin or metformin to improve FOLFOX-induced disruptions to muscle protein synthesis and autophagy in vitro. We hypothesized that the FOLFOX chemotherapy regimen would reduce autophagy flux in cultured myotubes by disrupting AMPK/mTORC1 signaling and the downstream effector ULK1. Furthermore, we hypothesized that either rapamycin or metformin would improve the FOLFOX-induced suppression of autophagy flux in vitro. Our study provides evidence that FOLFOX and the individual component oxaliplatin can directly suppress myotube AMPK/mTORC1 regulation of autophagy flux, and this effect is not readily reversible upon removal of the chemotherapeutic agent. Interestingly, we report that rapamycin administration can increase autophagy flux in myotubes treated with FOLFOX. Metformin did not rescue the FOLFOX suppression of myotube autophagy flux or protein synthesis, despite increasing AMPK phosphorylation.

MATERIALS AND METHODS

C2C12 Cell Culture

C2C12 myoblasts (ATCC CRL-1772) were seeded on type-I collagen-coated 6-well plates in growth media (GM): Dulbecco’s Modified Eagle Media (DMEM; Gibco #11995–065) supplemented with 10% fetal bovine serum (FBS), 50U/mL penicillin, and 50μg/mL streptomycin. Cells were maintained at 37°C, 5% CO2, and used before passage 15. To induce differentiation, myoblasts were switched to differentiation media (DM) at ~90% confluence (DM: DMEM supplemented with 2% horse serum, 50 U/mL penicillin, and 50μg/mL streptomycin). To reduce the effects of proliferating myoblasts, day 3 differentiated myotubes were treated with 20 μM Cytosine β-D-arabinofuranoside (AraC; Sigma #C1768) for 24-hrs as described (43). DM was replenished every 48-hrs and experiments were conducted as described (24).

CT26 conditional media collection and treatment

Colon-26 (CT26) adenocarcinoma (ATCC CRL-2638) cells were cultured in GM. Tumor cell conditioned media (CM), which consists of secreted factors from tumor cells (44), was collected when cells reached ~90% confluence, 48-h post cell seeding. CM was centrifuged at 3000 rpm for 5-min to remove cell debris and then stored in aliquots for one-time use at −20°C. CM was used within 2 months and thawed in a warm water bath at the time of the experiment. GM with no cells was used as media control. It should be noted that myotubes were differentiated up to day 5 in 2% horse serum; thus, upon CM addition, there is a need to account for the higher serum environment (CT26 10% FBS), which can impact cellular signaling (43). At the time of the experiment, GM control or CT26 CM were diluted with 50% serum-free DMEM for a final serum concentration of 5% FBS. 50% CM was chosen as it has been shown to produce significant myotube atrophy (43). Day 5 differentiated myotubes were rinsed with PBS and then switched to GM media control or CT26 CM for 48-h (days 5–7 of differentiation). CM was replenished after 24-h. Myotubes were treated with vehicle or FOLFOX chemotherapy in the last 24-h of GM or CT26 CM incubation.

Chemotherapy treatment and recovery

The effects of FOLFOX chemotherapy and single-agent chemotherapeutics were examined in day 6 differentiated myotube cultures and compared to vehicle (0.1% DMSO) treatment after 24 hours of exposure, as described (15, 19). The FOLFOX chemotherapy cocktail was administered at 2 different doses. High dose FOLFOX consisted of 5-fluorouracil (5-FU, 50 μg/mL), oxaliplatin (OX,3 μg/mL), and leucovorin (Leu, 10 μg/mL), while low dose FOLFOX consisted of 5-FU (25 μg/mL), OX (1.5 μg/mL) and Leu (5 μg/mL). Given that high-dose FOLFOX chemotherapy cocktail showed significant effects, single-agent experiments used the high-dose FOLFOX treatment for 24 hours. It should be noted that leucovorin is used to modulate the activity of 5-FU and, therefore, was not dosed alone. Cell lysates were collected on day 7 of differentiation. To examine the effects of recovery, FOLFOX or single agent chemotherapeutics were dosed as above for 24-h, after which myotubes were rinsed twice with PBS and switched back to DM for 24-h. Cell lysates were collected on day 8 of differentiation for recovery experiments.

Pharmacological Compounds

To assess mTORC1 signaling, Rapamycin (100 nM) was dosed with FOLFOX on day 6 of differentiation for 24-h. Rapamycin forms a complex with FKBP12 to inhibit downstream mTORC1 signaling (45). Metformin (1 mM) was dosed with FOLFOX as above for 24-h used to activate AMPK signaling (40). Autophagy flux was assessed by dosing 100nM Bafilomycin A1 (Sigma #19–148), a specific inhibitor of the lysosomal proton pump V-ATPase, 4-h prior to cell harvest (46). The Surface Sensing of Translation (SUnSET) technique was used to determine protein synthesis rates as described (47). Puromycin (Sigma, #540411) was dosed at 1 μM, 30 minutes prior to cell harvest. Puromycin incorporation into newly synthesized peptides was assessed by Western blot. Myotube lysates not treated with puromycin were used as a negative control.

Myotube Diameter

Myotube diameter was quantified as described (24). Briefly, C2C12 myotube diameter was quantified using ImageJ software (National Institutes of Health, Bethesda, MD, United States). Digital images were captured at ×20 objective brightfield, prior to cell harvest. We performed at least two separate experiments for each condition, and each experimental condition had 3 replicates. For each experimental replicate, there were five non-overlapping images randomly captured. Therefore, each condition had 30 images captured, and over 18 images were analyzed per condition for each myotube diameter measurement. The analysis used unmodified TIFF images accessed in the NIH ImageJ software. A blinded investigator randomly measured the diameters of at least 6 myotubes per image, based on preset inclusion/exclusion criteria: elongated structures with distinct membrane outlines, little to no cellular debris, and no branching points. The average myotube diameter was calculated as the mean of eight measurements taken along the myotube length. For each condition, n = 162–378 myotubes were analyzed.

Western Blotting

Cells were harvested in ice-cold RIPA buffer (Thermo #89901) plus protease phosphatase inhibitor cocktail (Thermo #78442). Cell lysates were centrifuged at 14,000rpm for 15min at 4°C, and the supernatant was collected for downstream protein analyses. Protein concentration was determined using the Bradford method. Lysates were fractionated on SDS-polyacrylamide gels and transferred to PVDF membranes (0.45μM). Ponceau S stain was used to ensure equal loading of each gel. Membranes were blocked for 1-h in 5% milk in Tris-buffered saline with 0.1% Tween-20 (TBST) at room temperature. Primary antibodies (Supplemental Table 1) were purchased from Cell Signaling Technology and diluted 1:1000. Primary antibodies were incubated overnight in 5% milk-TBST. After primary incubation, membranes were washed and incubated in 5% milk-TBST containing anti-rabbit or anti-mouse IgG horseradish-peroxidase conjugated secondary antibodies for 1-h at room temperature (Supplemental Table 1). Enhanced chemiluminescence (Prometheus Pro-Signal Femto, #20–302; Genesee Scientific, San Diego, CA) was used to visualize the antibody-antigen interactions. Immunoblot images were digitally imaged using the iBright 1500 system (Invitrogen, Carlsbad, CA), and blots were quantified by densitometry using imaging software (ImageJ, NIH). Phosphorylated proteins were normalized directly to their respective total protein levels to reflect pathway activation. Total proteins and independent target proteins (e.g., MuRF-1) were measured using equal amounts of protein per lane (10μg). Ponceau S staining was used to visually verify equal loading between lanes. After imaging, membranes were washed and stained with Coomassie Blue or Ponceau S to visualize equal protein loading.

Statistical Analysis

Results are reported as mean ± SEM. A minimum of three replicates from at least two independent experiments were used for analysis. All data were normalized to vehicle and showed fold change from vehicle; we anticipated the data to follow a normal distribution. We tested data normality using the Shapiro-Wilk test. We used the Unpaired t-test for the analysis of the difference between vehicle and single agent, One-way ANOVA with Tukey’s multiple comparisons test for the analysis of the difference among vehicle, high, and low doses of FOLFOX, and two-way ANOVA analyses with Tukey’s post hoc multiple comparisons test for 4 or more groups when a significant interaction was present. Specific statistical analyses are indicated in each figure legend. Statistical analysis and figures were prepared using GraphPad (Prism 10 for Mac OS X, La Jolla, CA). Statistical significance was set to p<0.05, and the criteria for considering a trend towards significance were set to p<0.1 for all measures.

RESULTS

Effects of FOLFOX chemotherapy and CT26 conditioned media on myotube diameter, MyHC, STAT3, and E3 Ubiquitin ligase expression in C2C12 myotubes

The direct effects of FOLFOX chemotherapy and CT26-tumor-derived factors were examined in C2C12 myotubes pre-treated with AraC. Briefly, 5-day differentiated myotubes were dosed with GM or CT26 conditioned media (CM). After 24 hours, CM was replenished, and myotubes were dosed with vehicle or FOLFOX for another 24 hours (Figure 1A). While myotube diameter demonstrated a trend (p=.054) for an interaction between CT26 and FOLFOX treatments, there were significant main effects for either CT26 or FOLFOX to reduce myotube diameter (Figure 1B). Myosin heavy chain (MyHC) Type II (fast-twitch/glycolytic) and Type I (slow-twitch/oxidative) protein expression was reduced in myotubes treated with CT26, independent of FOLFOX treatment (Figures 1C–E). CT26-treated myotubes had higher STAT3(Y705) phosphorylation, independent of FOLFOX treatment (Figure 1C and F). CT26 did not alter myotube Atrogin-1 protein expression; however, FOLFOX had higher Atrogin-1 protein expression (Figure 1C and G). MuRF-1 expression demonstrated a significant interaction between CT26 and FOLFOX treatments. Both CT26 and FOLFOX treatments significantly increased MuRF-1 expression. Due to considerable myotube variability in response, the combination of FOLFOX and CT26 treatment did not significantly alter MuRF-1 levels (Figure 1C and H). Ubiquitin protein expression in myotubes was not different across groups (Figure 1I). These data indicate that CT26 and FOLFOX chemotherapy exert distinct and non-additive effects on myotube size, MyHC protein expression, and STAT3 phosphorylation. Specifically, CT26 reduced myotube diameter and MyHC Type II and Type I protein expression, and increased STAT3 phosphorylation. Conversely, FOLFOX reduced myotube diameter without altering MyHC protein expression; importantly, myotube diameter was not further reduced in the combined CT26+FOLFOX condition. Interestingly, FOLFOX did not alter the CT26-induced STAT3 phosphorylation in myotubes.

Figure 1:

Effects of FOLFOX chemotherapy and CT26-conditioned media on myotube diameter, MyHC, STAT3, and E3 Ub ligase expression in C2C12 myotubes. A) Study design: C2C12 myotubes pre-treated with 20μM AraC were switched to GM or CT26 conditioned media (CM) 5-days post differentiation for 24 hours. FOLFOX or vehicle were added to GM or CM 6-days post differentiation for 24 hours. B) Myotube diameter quantification. C) Representative Western blot with Ponceau S protein stain and quantification of D) MyHC Type II (fast-twitch/glycolytic) and E) MyHC Type I (slow-twitch/oxidative) protein expression, F) phosphorylated to total STAT3 (Y705), G) Atrogin-1, and H) MuRF1. I) Representative Western blot with Coomassie blue and quantification of ubiquitin. Samples were collected from two independent experiments (n=6). Data is presented as mean ± SEM. Data is analyzed using Two-way ANOVA with Tukey’s multiple comparisons test. Bold and italic text denotes significance. Statistical significance is p<0.05 and trend is set to p<0.1. Different lowercase letters (a, b, …) denote significant difference between groups.

Effects of FOLFOX chemotherapy and CT26 conditioned media on AMPK/mTORC1 signaling and autophagy flux in C2C12 myotubes

The direct effects of FOLFOX chemotherapy treatment in combination with CT26-tumor-derived factors on myotube AMPK/mTORC1 signaling were examined. FOLFOX suppressed AMPK(T172) phosphorylation independent of CT26 (Figure 2A–B). CT26 or FOLFOX did not alter rpS6(S240/244) phosphorylation (Figure 2A and C). Total rpS6 protein expression was reduced FOLFOX independent of CT26 (Figure 2A and D). FOLFOX suppressed ULK1(S555) phosphorylation independent of CT26 (Figure 2A and E). Both FOLFOX and CT26 reduced ULK1(S757) phosphorylation (Figure 2A and F). Both CT26 and FOLFOX reduced myotube puromycin incorporation (Figure 2G and H).

Figure 2:

Effects of FOLFOX chemotherapy and CT26-conditioned media on AMPK/mTORC1 signaling and autophagy flux in C2C12 myotubes. A) Representative Western blot with Coomassie blue and quantification of B) phosphorylated to total AMPK (T172), C) rpS6 (S240/244), D) total rpS6, E) ULK1 (S555), and F) ULK1 (S757). G) Representative Western blot with Coomassie blue and quantification of H) puromycin incorporation. I) Representative Western blot with Ponceau S protein stain and quantification of J) LACBII/I and K) p62/SQSTM1. Samples were collected from two or three independent experiments (n=6–9). Data is presented as mean ± SEM. Data is analyzed using Two-way ANOVA with Tukey’s multiple comparisons test. Bold and italic text denotes significance. Statistical significance is p<0.05 and trend is set to p<0.1. Different lowercase letters (a, b, c…) denote significant difference between groups.

Autophagy flux was assessed in myotubes treated with bafilomycin A1 in the last 4 hours of CT26 and FOLFOX incubation. Under basal conditions CT26 did not alter the LC3BII/I ratio or p62/SQSTM1 protein expression (Figures 2I–K). p62/SQSTM1 protein expression was higher in FOLFOX treated myotubes, independent of CT26. The LC3BII/I ratio was 139% higher upon autophagy inhibition with bafilomycin in vehicle treated myotubes and this induction was not altered by CT26. The LC3BII/I ratio was 65% higher upon autophagy inhibition with bafilomycin in FOLFOX treated myotubes, which was significantly lower than vehicle controls. CT26 did not alter the FOLFOX suppression of LC3BII/I. p62/SQSTM1 protein expression increased 107% with the addition of bafilomycin in vehicle treated myotubes, and CT26 did not alter this induction. p62SQSTM1 protein expression was higher upon bafilomycin addition in FOLFOX treated myotube cultures (85%, FOLFOX+vehicle vs. FOLFOX+bafilomycin) and CT26 did not affect this induction. These data suggest that FOLFOX inhibits myotube AMPK signaling and autophagy flux regardless of CT26 exposure. Interestingly, CT26 treatment did not affect myotube AMPK phosphorylation or autophagy flux.

Effect of FOLFOX chemotherapy dose, 5-FU, 5-FU+Leucovorin, and oxaliplatin on AMPK/mTORC1 signaling, protein synthesis, and autophagy flux in C2C12 myotubes.

We examined the acute effects of FOLFOX dose by treating myotubes with either a high or a low dose of FOLFOX for 24 hours (Figure S1A). Myotube diameter was approximately 16% lower in myotubes treated with low and high dose FOLFOX compared to vehicle (Figure S1B). MyHC Type II (fast-twitch/glycolytic) and Type I (slow-twitch/oxidative) protein expressions were 29% and 42% lower, respectively, by the FOLFOX high dose compared to vehicle (Figure S1C). There were no differences in MyHC Type II (fast-twitch/glycolytic) and Type I (slow-twitch/oxidative) protein expression for low-dose FOLFOX treatment compared to vehicle (Figure S1C). The phosphorylated—to-total Akt(S473) ratio was 44% lower in low dose and 67% lower in high-dose FOLFOX-treated myotubes compared to vehicle (Figure S1D). p70S6K(T389) was 40% lower in high-dose FOLFOX and not altered by low-dose FOLFOX compared to vehicle (Figure S1D). Atrogin-1 protein expression was not altered by low or high dose FOLFOX compared to vehicle (Figure S1E). MuRF1 protein expression was 83% higher in high-dose FOLFOX compared to vehicle and not altered by low-dose FOLFOX (Figure S1E). The phosphorylated to total AMPK (T172) ratio was approximately 45% lower in both low and high dose FOLFOX compared to the vehicle (Figure S2A-B). The phosphorylated to total rpS6 (S240/244) ratio was approximately 24% lower in both low and high dose FOLFOX compared to the vehicle (Figure S2A and C). Total rpS6 protein, a component of the 40S small ribosomal subunit (48), was approximately 16% lower in low and high dose FOLFOX compared to the vehicle (Figure S2A and D). The phosphorylated to total ULK1(S555) ratio was 46% lower in high-dose FOLFOX compared to low-dose and vehicle (Figure S2A and E). There were no differences between low-dose FOLFOX and vehicle for ULK1(S555) phosphorylation (Figure S2A and E). The phosphorylated to total ULK1(S757) ratio was not altered by low or high dose FOLFOX (Figure S2A and F). Protein synthesis was reduced in a dose-dependent manner; specifically, puromycin incorporation was 30% lower in low-dose and 53% lower in high-dose FOLFOX compared to vehicle (Figure S2G). Puromycin incorporation for myotubes treated with high-dose FOLFOX was 33% lower compared to low-dose FOLFOX (Figure S2G). LC3B mRNA expression was 70% higher in low-dose and 118% higher in high-dose FOLFOX compared to vehicle (Figure S2H). p62/SQSTM1 mRNA expression was 108% higher in low dose and 171% higher in high dose FOLFOX compared to vehicle (Figure S2 I). As expected, Bafilomycin had higher LC3BII/I and p62/SQSTM1 protein expression in vehicle and FOLFOX-treated myotubes compared to non-bafilomycin-treated myotubes (Figures S2J-L). FOLFOX treatment caused lower LC3BII/I and higher p62/SQSTM1 protein levels compared with vehicle, indicative of suppressed autophagy flux. These data demonstrate that 24-hour high-dose FOLFOX chemotherapy treatment directly suppresses myotube AMPK and mTORC1 signaling, affecting protein synthesis and autophagy flux. FOLFOX chemotherapy represents a combined cocktail of chemotherapeutic agents.

We next examined the effects of individual components in the FOLFOX regimen on AMPK/mTORC1 signaling, protein synthesis, and autophagy flux. Leucovorin is a folic acid analog used to enhance the efficacy of 5-FU; therefore, it was not dosed alone. Six-day differentiated myotube cultures (pre-treated with AraC) were treated with 5-FU, 5-FU+Leucovorin (5-FU+Leu), or oxaliplatin (OX) for 24 hours. The phosphorylated to total AMPK(T172) protein ratio was 48% lower in 5-FU and 37% lower in 5-FU+Leu treatments compared to vehicle (Figures 3A and C–D). There was a trend (p=0.059) for OX to have lower AMPK(T172) phosphorylation than vehicle (Figures 3A and E). The rpS6(S240/244) phosphorylation was 54% higher in 5-FU-treated myotubes compared to vehicle. There were no effects for 5-FU+Leu or OX to alter rpS6(S240/244) phosphorylation (Figures 3A and C–E). Total rpS6 protein expression was lower in 5-FU (−44%), 5-FU+Leu (−46%), and OX (−34%) compared to vehicle (Figures 3A and C–E). The phosphorylated-to-total ULK1(S555) ratio was 44% lower in OX than in vehicle (Figures 3A and E). ULK1(S555) phosphorylation was not altered by 5-FU or 5-FU+Leu (Figures 3A, C–D). ULK1(S757) phosphorylation was lower in 5-FU (−49%), 5-FU+Leu (−46%), and OX (−42%) compared to vehicle (Figures 3A and C–E). Puromycin incorporation did not differ across groups (Figures 3B and C–E). Bafilomycin increased the LC3BII/I ratio and p62/SQSTM1 protein expression independent of 5-FU and 5-FU+Leu (Figure 3F–J). Bafilomycin increased the LC3BII/I ratio in vehicle and OX-treated myotubes (Figure 3F and K). OX-treated myotubes had lower LC3BII/I ratio and a higher p62/SQSTM1 protein expression, suggesting a suppression of autophagy flux (Figure 3F and K–L). Both bafilomycin and OX increased p62/SQSTM1 protein expression (Figure 3F and L).

Figure 3:

Effect of 5-FU, 5-FU+Leucovorin, and oxaliplatin on AMPK/mTORC1 signaling, protein synthesis, and autophagy flux in C2C12 myotubes. A) and B) Representative Western blot with Coomassie blue and quantified protein expression of AMPKa(T172), rpS6(S240/244), total rpS6, ULK1(S555), ULK1(S757) and puromycin incorporation for C) 5-FU, D) 5-FU+LEC and E) OX. F) Representative Western blot with Coomassie blue and quantification of LC3B II/I and p62SQSTM1 for 5-FU, I-J) 5-FU + Leu, and K-L) OX. 100nM Bafilomycin A1 was used to assess autophagy flux 4-h prior to cell harvest. Samples were collected from two or three independent experiments (n=6–9). Data is presented as Mean±SEM. Data is analyzed using unpaired t test (C-E) or Two-Way ANOVA with Tukey’s multiple comparisons test (G-L). Statistical significance is set to p<0.05 and trend is set to p<0.1. Bold and italic values denote significance. * p<0.05, ** p<0.01, *** p<0.001.

In summary, 5-FU and 5-FU+Leu treatment for 24 hours were sufficient to reduce myotube AMPK(T172) phosphorylation compared with vehicle; however, this did not correspond to changes in the downstream AMPK phosphorylation target, ULK1(S555). Conversely, OX demonstrated a trend toward reduced AMPK phosphorylation compared with vehicle, and OX treatment suppressed ULK1(S555) phosphorylation. All components of the FOLFOX regimen decreased total rpS6 protein expression and ULK1(S757) phosphorylation when administered as single agents. There were no effects for 5-FU, 5-FU+Leucovorin, or oxaliplatin to alter protein synthesis. Autophagy flux was not altered by 5-FU or 5-FU+Leucovorin, while oxaliplatin suppressed the LC3BII/I ratio and increased p62/SQSTM1 protein expression following bafilomycin treatment. These data suggest that the combined FOLFOX regimen exerts more consistent effects in suppressing AMPK-mediated autophagy signaling. Interestingly, oxaliplatin alone was sufficient to reduce autophagy flux in myotubes, whereas treatments with 5-FU and 5-FU+Leu reduced AMPK phosphorylation but did not affect autophagy flux.

Lasting effects of FOLFOX on AMPK/mTORC1 signaling, protein synthesis, and autophagy flux in C2C12 myotubes.

We studied the lasting effects of FOLFOX (recovery) on myotubes treated with FOLFOX for 24 hours, then switched back to standard culture media (differentiation media, 2% horse serum) for another 24 hours (Figure 4A). The phosphorylated to total AMPK (T172) protein ratio was 49% lower in the low dose group and 47% lower in the high dose FOLFOX group during recovery (Figure 4B–C). The rpS6(S240/244) phosphorylation was 26% higher in low-dose and 43% higher in high-dose FOLFOX compared to vehicle at recovery (Figure 4B and D). Total rpS6 protein expression was 32% lower in low-dose and 35% lower in high-dose FOLFOX at recovery compared to vehicle (Figure 4B and E). ULK1(S555) phosphorylation was not altered by low or high-dose FOLFOX at recovery (Figure 4B and F). ULK1(S757) phosphorylation was 28% lower in low-dose and 35% lower in high-dose FOLFOX compared to vehicle at recovery (Figure 4B and G). Puromycin incorporation was 26% lower in low-dose and 38% lower in high-dose FOLFOX compared to vehicle at recovery (Figure 4H–I). Bafilomycin increased LC3BII/I and p62/SQSTM1 protein expression in vehicle and FOLFOX-treated myotubes at recovery (Figures 4J–L). FOLFOX had lower LC3BII/I and higher p62/SQSTM1 protein expression at recovery compared to vehicles. These data indicate that FOLFOX induces lasting effects to reduce myotube rpS6 protein expression and protein synthesis. Additionally, FOLFOX suppression of autophagy flux was not reversible and remained reduced relative to vehicle controls after 24 hours of recovery.

Figure 4:

Lasting effects of FOLFOX on AMPK/mTORC1 signaling, protein synthesis, and autophagy flux in C2C12 myotubes. A) Study Design: C2C12 myotubes pre-treated with 20μM AraC were dosed 6-d post differentiation with vehicle (0.1% DMSO) or FOLFOX (5-FU: 50μg/mL, Leu: 10μg/mL, OX: 3μg/mL) for 24-hrs and then switched back to DM for 24-hrs. B) Representative Western blot with Coomassie blue and quantified phosphorylated to total protein expression of C) AMPKa(T172), D) rpS6 (S240/244), F) ULK1(S555), G) ULK1(S757) and E) quantification of total rpS6. H) Representative Western blot with Coomassie blue and I) quantification of puromycin incorporation. Puromycin (1μM) was added 30-min prior to cell harvest, NC denotes negative control sample not treated with puromycin. J) Representative Western blot with Coomassie blue and quantification of K) LC3B II/I and L) p62/SQSTM1. BafilomycinA1 (100nM) was used to assess autophagy flux 4-h prior to cell harvest. Samples were collected from two independent experiments (n=8). Data is presented as Mean±SEM. Data is analyzed using One-Way ANOVA (C-I) or Two-Way ANOVA (J-L) with Tukey’s multiple comparisons test. Statistical significance is set to p<0.05 and trend is set to p<0.1. Bold and italic values denote significance. Different lowercase letters (a, b, c…) denote significant difference between groups.

Lasting effects of 5-FU, 5-FU+Leucovorin, and oxaliplatin on AMPK/mTORC1 signaling, protein synthesis, and autophagy flux in C2C12 myotubes.

The lasting effects (recovery) of single chemotherapeutic agents in the FOLFOX cocktail were examined in myotubes treated with 5-FU, 5-FU+leucovorin, or oxaliplatin for 24 hours and switched back to standard culture media for another 24 hours. There were no effects of 5-FU, 5-FU+Leu, or OX on AMPK(T172) phosphorylation compared to vehicle during recovery (Figure 5A and C–E). The rpS6(S240/244) phosphorylation was 31% higher in 5-FU-treated myotubes compared to vehicle (Figure 5A and C–E). There were no effects for 5-FU+Leu or OX to alter rpS6 phosphorylation at recovery (Figure 5A and C–E). Total rpS6 protein expression was lower in 5-FU (−44%) and OX (−35%) compared to vehicle at recovery (Figure 5A and C–E). ULK1(S555) phosphorylation was not different between vehicle and 5-FU, 5-FU+Leu, or OX (Figure 5A and C–E). The phosphorylated to total ULK1(S757) ratio was lower in 5-FU+Leu (−24%) and OX (−41%) compared to vehicle at recovery (Figure 5A and C–E). At recovery, puromycin incorporation was lower in myotubes treated with 5-FU (−51%) and OX (−54%) compared to vehicle (Figure 5B and C–E). Puromycin incorporation was not different between vehicle and 5-FU+Leu at recovery (Figure 5B and C–E). Bafilomycin increased the LC3BII/I ratio and p62/SQSTM1 protein expression independent of 5-FU, 5-FU+Leu, or OX treatments at recovery (Figure 5F–L). 5-FU or 5-FU+Leu did not alter LC3BII/I or p62/SQSTM1 protein expression at recovery (Figure 5F–J). OX had lower LC3BII/I and p62/SQSTM1protein expression compared to vehicle at recovery (Figure 5F and K–L).

Figure 5:

Lasting effects of 5-FU, 5-FU+Leucovorin, and oxaliplatin on AMPK/mTORC1 signaling, protein synthesis, and autophagy flux in C2C12 myotubes. A) and B) Representative Western blot with Coomassie blue and quantified phosphorylated to total protein expression of A) AMPKa(T172), B) rpS6(S240/244), C) total rpS6, D) ULK1(S555), and E) ULK1(S757) and puromycin incorporation for C) 5-FU, D) 5-FU+LEC and E) OX. Puromycin (1μM) was added 30-min prior to cell harvest, NC denotes negative control sample not treated with puromycin. F) Representative Western blot with Coomassie blue and quantification of LC3B II/I and p62SQSTM1 for G-H) 5-FU, I-J) 5-FU + Leu, and K-L) OX. 100nM Bafilomycin A1 was used to assess autophagy flux 4-h prior to cell harvest. Samples were collected from 2–3 independent experiments (n=6–9). Data is presented as Mean±SEM. Data is analyzed using unpaired t test (C-E) or Two-way ANOVA with Tukey’s multiple comparisons test (G-L). Statistical significance is set to p<0.05 and trend is set to p<0.1. Bold and italic values denote significance. * p<0.05, ** p<0.01, *** p<0.001.

In summary, these data show that after removing 5-FU, 5-FU+leucovorin, or oxaliplatin from the cell culture media (i.e., during recovery), the suppression of AMPK(T172) phosphorylation was no longer different from the vehicle control, indicating complete recovery. However, 5-FU and OX treatments had lasting effects on suppressing total rpS6 protein expression. There were ongoing effects of 5-FU and oxaliplatin on inhibiting myotube protein synthesis, and oxaliplatin's suppression of autophagy flux did not recover within 24 hours. Effect of Rapamycin administration and mTORC1 regulation of protein synthesis and autophagy flux in C2C12 myotubes treated with FOLFOX.

Effect of Rapamycin administration and mTORC1 regulation of protein synthesis and autophagy flux in C2C12 myotubes treated with FOLFOX.

The effects of mTORC1 inhibition on the regulation of protein synthesis and autophagy flux were assessed in myotube cultures treated with FOLFOX and 100nM rapamycin for 24 hours (Figure 6A). Protein expressions of rpS6 (S240/244), AMPK (T172) and ULK1 (S555) demonstrated a significant interaction between rapamycin and FOLFOX treatments (Figure 6B–F). Rapamycin reduced the phosphorylated to total rpS6 protein ratio by approximately −96% in vehicle and FOLFOX-treated myotubes (Figure 6B–C). The phosphorylated to total rpS6 ratio was 68% higher in FOLFOX compared to vehicle (Figure 6B–C). However, this increase was largely driven by a reduction in total rpS6 protein expression, which was 34% lower in FOLFOX compared to vehicle (Figure 6B and D). Rapamycin reduced total rpS6 expression −58% in vehicle and −48% in FOLFOX-treated myotubes (Figure 6B and D). The phosphorylated to total AMPK(T172) protein ratio was reduced by rapamycin (−38%) and FOLFOX (−59%) compared to vehicle-treated myotubes (Figure 6B and E). Interestingly, rapamycin treatment restored the FOLFOX suppression of AMPK(T172) phosphorylation, which was not different from vehicle-treated myotubes (Figure 6B and E). AMPK(T172) phosphorylation was 154% higher in FOLFOX+rapamycin-treated myotubes compared to FOLFOX alone (Figure 6B and E). ULK1(S555) phosphorylation was 49% lower in FOLFOX compared to vehicle; there were no differences between vehicle, rapamycin, or FOLFOX+rapamycin for ULK1(S555) phosphorylation (Figure 6B and F). Rapamycin had lower ULK1(S757) phosphorylation independent of FOLFOX (Figure 6B and G). Rapamycin had lower puromycin incorporation independent of FOLFOX (Figure 6H–I).

Figure 6:

Effect of Rapamycin administration and mTORC1 regulation of protein synthesis and autophagy flux in C2C12 myotubes treated with FOLFOX. A) Study Design: C2C12 myotubes pre-treated with 20μM AraC were dosed with vehicle (0.1% DMSO) or FOLFOX (5-FU: 50μg/mL; Leucovorin 10μg/mL; Oxaliplatin 3μg/mL) plus 100nM Rapamycin or vehicle for 24-hrs. B) Representative Western blot with Ponceau S protein stain and quantified phosphorylated to total protein expression of C) rpS6(S240/244) and D) total rpS6, E) AMPKa (T172) F) ULK1(S555), G) ULK1(S757). H) Representative Western blot with Coomassie blue and I) quantification of puromycin incorporation for protein synthesis. Puromycin (1μM) was added 30-min prior to cell harvest, NC denotes negative control sample not treated with puromycin. J) Representative Western blot with Coomassie blue and quantification of K) LC3B II/I and L) p62/SQSTM1 protein expression. 100nM Bafilomycin A1 was used to assess autophagy flux 4-h prior to cell harvest. Samples were collected from 2–3 independent experiments (n=6–9). Data is presented as Mean±SEM. Data is analyzed using Two-Way ANOVA with Tukey’s multiple comparisons test. Statistical significance is set to p<0.05 and trend is set to p<0.1. Bold and italic values denote significance. Different lowercase letters (a, b, c…) denote significant difference between groups.

Next, the effect of rapamycin on the FOLFOX suppression of autophagy flux was assessed in myotube cultures treated with bafilomycin (Figure 6J–L). The LC3BII/I ratio was 36% higher in myotubes treated with rapamycin alone compared to the vehicle. In vehicle-treated myotubes dosed with bafilomycin, the LC3BII/I ratio was higher with rapamycin treatment (65%, bafilomycin vs. rapamycin+bafilomycin). Interestingly, similar effects of rapamycin were observed in FOLFOX-treated myotubes, although to a lesser extent compared to the vehicle. The LC3BII/I ratio was 63% higher in rapamycin+FOLFOX-treated myotubes without bafilomycin. Upon autophagy inhibition, the LC3BII/I protein ratio remained higher in rapamycin+FOLFOX-treated myotubes. (60%, bafilomycin vs. rapamycin+bafilomycin) (Figure 6J–K). In vehicle-treated myotubes, rapamycin reduced the bafilomycin induction of p62/SQSTM1 protein expression (−90%, bafilomycin vs. rapamycin+bafilomycin). p62/SQSTM1 protein expression was 75% higher in FOLFOX compared to vehicle myotubes not treated with bafilomycin, while rapamycin treated myotubes had lower (−69%) p62/SQSTM1 protein expression compared to vehicle. Upon autophagy inhibition, p62/SQSTM1 protein expression was 84% higher in FOLFOX-treated myotubes, and this increase was attenuated by rapamycin (−83%, bafilomycin vs. rapamycin+bafilomycin)(Figure 6L).

In summary, these data demonstrate that the inhibition of mTORC1 by rapamycin was sufficient to partially rescue the FOLFOX suppression of autophagy flux. Interestingly, rapamycin treatment blocked the FOLFOX suppression of AMPK(T172) phosphorylation.

Effect of Metformin-induced AMPK activation on the regulation of protein synthesis and autophagy flux in C2C12 myotubes treated with FOLFOX.

The effects of AMPK activation on the regulation of myotube protein synthesis and autophagy flux were assessed in myotube cultures treated with FOLFOX and metformin for 24 hours (Figure 7A). Metformin induced AMPK(T172) phosphorylation in vehicle and FOLFOX-treated myotubes (Figure 7B–C). However, FOLFOX-treated myotubes had lower AMPK(T172) phosphorylation independent of metformin treatment (Figure 7B–C). Acetyl-CoA carboxylase (ACC) is an enzyme involved in lipid metabolism and a direct downstream phosphorylation target of AMPK (49). Metformin induced ACC(S79) in vehicle and FOLFOX-treated myotubes (Figure 7B and D). However, ACC(S79) was lower in FOLFOX-treated myotubes compared to vehicle (Figure 7B and D). FOLFOX reduced total ACC protein expression independent of metformin (Figure 7B and E). Protein expressions of rpS6 (S240/244) demonstrated a significant interaction between metformin and FOLFOX treatments. The phosphorylated to total rpS6 protein ratio was not altered by metformin in vehicle-treated myotubes (Figure 7B and F). In FOLFOX-treated myotubes, rpS6(S240/244) phosphorylation was 59% lower with metformin treatment compared to FOLFOX alone (Figure 7B and F). Both FOLFOX and metformin reduced total rpS6 protein expression (Figure 7B and G). Protein expressions of ULK1(S555) and ULK1(S757) demonstrated a significant interaction between metformin and FOLFOX treatments (Figure 7B and H–I). In vehicle-treated myotubes, the phosphorylation to total ULK1(S555) protein ratio was 77% higher with metformin (Figure 7B and H). ULK1(S555) phosphorylation was 35% lower in FOLFOX compared to vehicle, and metformin did not alter the FOLFOX suppression of ULK(S555) phosphorylation (Figure 7B and H). Metformin treatment did not alter ULK1(S757) phosphorylation in vehicle-treated myotubes (Figure 7B and I). ULK1(S757) phosphorylation was 24% lower in FOLFOX, and metformin further reduced ULK1(S757) phosphorylation in FOLFOX treated myotubes (−25% vs. FOLFOX vehicle) (Figure 7B and I). Both FOLFOX and metformin reduced puromycin incorporation (Figure 7J and K).

Figure 7:

Effect of Metformin-induced AMPK activation on the regulation of protein synthesis and autophagy flux in C2C12 myotubes treated with FOLFOX. A) Study Design: C2C12 myotubes pre-treated with 20μM AraC were dosed with vehicle (0.1% DMSO) or FOLFOX (5-FU: 50μg/mL; Leucovorin 10μg/mL; Oxaliplatin 3μg/mL) plus 1mM Metformin or vehicle for 24-hrs. B) Representative Western blot with Coomassie blue and quantified phosphorylated to total protein expression of C) AMPK (T172), D) ACC (S79), E) total rpS6, F) rpS6 (S240/244), G) total rpS6, H) ULK1 (S555), and I) ULK1 (S757). J) Representative Western blot with Coomassie blue and K) quantification of puromycin incorporation for protein synthesis. Puromycin (1μM) was added 30-min prior to cell harvest, NC denotes negative control sample not treated with puromycin. L) Representative Western blot with Coomassie blue and quantification of M) LC3B II/I and N) p62/SQSTM1 protein expression. 100nM Bafilomycin A1 was used to assess autophagy flux 4-hrs prior to cell harvest. Samples were collected from 2 independent experiments (n=6). Data is presented as Mean±SEM. Data is analyzed using Two-Way ANOVA with Tukey’s post hoc. Statistical significance is set to p<0.05 and trend is set to p<0.1. Bold and italic values denote significance. Different letters (a, b, c…) denote significant difference between groups.

The effect of AMPK activation on autophagy flux was assessed in myotubes treated with FOLFOX and metformin for 24 hours (Figure 7L–M). In vehicle-treated myotubes, there was no effect of metformin to alter the LC3BII/I ratio (Figure 7L–M). There was no effect of metformin to alter the bafilomycin-induced LC3BII/I changes in vehicle-treated myotubes (Figure 7M). In FOLFOX-treated myotubes, metformin (without bafilomycin) did not alter the LC3BII/I ratio (Figure 7M). There were no effects for metformin to alter the FOLFOX suppression of LC3BII/I (Figure 7M). Metformin did not alter p62/SQSTM1 protein expression in vehicle-treated myotubes (Figure 7N). There was a main effect for bafilomycin to increase p62/SQSTM1 and a significant effect for metformin to reduce p62/SQSTM1 in vehicle-treated myotubes. In FOLFOX-treated myotubes, autophagy inhibition with bafilomycin resulted in a 70% increase in p62/SQSTM1 protein expression (FOLFOX+vehicle vs. FOLFOX+bafilomycin), and this induction was 38% lower with metformin treatment (Figure 7K and M).

In summary, although metformin increased AMPK (T172) phosphorylation in FOLFOX-treated myotubes, the increase was less pronounced compared to myotubes not treated with FOLFOX. Metformin treatment did not sufficiently increase ULK1 (S555) phosphorylation in FOLFOX-treated myotubes. Metformin also further decreased total rpS6 protein levels and protein synthesis in these myotubes. Metformin did not affect autophagy flux in either vehicle or FOLFOX-treated myotubes, although it activated AMPK and downstream ULK1 signaling in vehicle-treated myotubes. Interestingly, metformin decreased the FOLFOX-induced expression of p62/SQSTM1 protein.

DISCUSSION

FOLFOX chemotherapy is a common treatment for colorectal cancer, consisting of a cocktail containing the antimetabolite 5-FU, folinic acid leucovorin, and the platinum-based alkylating agent oxaliplatin (3). Preclinical studies have demonstrated that a variety of 5-FU-based chemotherapeutic regimens cause skeletal muscle toxicities leading to atrophy, weakness, and mitochondrial dysfunction (14, 50–52). However, how 5-FU-based chemotherapies directly disrupt muscle cellular signaling, leading to fatigue and weakness, remains poorly understood. To this end, our study investigated the direct effects of 5-FU, oxaliplatin, or the FOLFOX chemotherapy regimen on myotube protein synthesis and autophagy flux. Our results provide evidence that FOLFOX can directly suppress myotube AMPK signaling, protein synthesis, and autophagy flux. Importantly, the effects of FOFOX and the disruptions to muscle protein turnover were long-lasting and not easily reversible, indicating that the toxicity extends beyond the acute disruption of signaling.

We also investigated the combined effects of CT26 tumor cell conditioned media and FOLFOX. Interestingly, both CT26 and FOLFOX were sufficient to reduce myotube diameter and protein synthesis, but they differentially affected STAT3 phosphorylation and fast- and slow-type myosin heavy chain expression. CT26 media, but not FOLFOX, induced STAT3 phosphorylation. Therefore, the effects of FOLFOX, including the suppression of AMPK phosphorylation, autophagy flux, and the induction of Atrogin-1 protein expression, appear independent of STAT3 signaling. Although Atrogin-1 and MuRF-1 are both atrogenes, they may have distinct transcriptional regulation. FOLFOX can induce cellular stress, which could lead to independent, widespread activation of FoxO transcription factors, thereby increasing Atrogin-1. In contrast, MuRF-1 is highly responsive to inflammatory cytokines (e.g., TNF-a) in the CT26-CM via STAT3 pathways. Our findings suggest that FOLFOX-induced growth decrements in myotubes are mediated by alternative signaling pathways compared to CT26 stimuli, but ultimately promote proteolytic activation and reduced myotube diameter. We also observed similar findings on STAT3 protein expression in tumor-bearing mice treated with FOLFOX in BALB/c mice (53). We next examined whether manipulating AMPK and mTORC1 signaling with metformin or rapamycin could rescue FOLFOX-induced suppression of myotube autophagy flux and protein synthesis. Interestingly, we report that rapamycin administration partially restored the FOLFOX suppression of autophagy flux and rescued AMPK signaling. Metformin treatment did not rescue the FOLFOX suppression of myotube autophagy flux or protein synthesis, despite activating AMPK signaling. The potential benefits of rapamycin treatment for offsetting chemotherapy-related myopathies, and the mechanisms by which it indirectly affects disrupted muscle signaling pathways beyond mTORC1, such as AMPK, remain to be determined and warrant further investigation.

An important aspect of our study was that oxaliplatin, a component of FOLFOX, was sufficient to suppress autophagy flux, but not protein synthesis. Not surprisingly, our data provides evidence that the combined FOLFOX drug cocktail exerts more severe disruption to myotube AMPK signaling, protein synthesis, and autophagy flux relative to each drug dosed alone. While all components of the FOLFOX regimen reduced rpS6 protein expression and ULK1(S757) phosphorylation when dosed as single agents, there were no effects to alter protein synthesis acutely. Interestingly, 5-FU and oxaliplatin only reduced protein synthesis during recovery, demonstrating that these single-component chemotherapy drugs can exert lasting toxic effects on myotube protein metabolism. Oxaliplatin is associated with severe neurotoxicity in patients, and in some cases, oxaliplatin is removed from patients receiving the FOLFOX regimen due to adverse side effects (54). Notably, studies have suggested that long-term side effects of oxaliplatin may result from long-lasting (8–75 months after diagnosis) platinum retention in tissues and plasma samples (55). Preclinical studies have demonstrated that 5-FU treatment alone causes many acute toxicities that can alter resident immune cells, suppress regeneration (8). While 5-FU can suppress Akt(S473) phosphorylation and increase markers of cell stress (17), we report that oxaliplatin directly dysregulates autophagy flux signaling in myotubes. Our findings reveal that oxaliplatin can directly disrupt myotube cell signaling, which is important because platinum-based chemotherapies like oxaliplatin are often linked to muscle toxicities involving nerve damage and disruption of the neuromuscular junction (56). It is interesting to speculate on the cause of the lasting suppression of muscle protein synthesis following FOLFOX treatment, which could result from platinum accumulation. However, it is interesting to speculate that this lasting suppression is driven by non-genomic mechanisms, such as the direct incorporation of 5-FU into ribosomal RNA (rRNA) and transfer RNA (tRNA). Since ribosome function is critical to muscle protein synthesis (48), disruptions in ribosome number and function caused by 5-FU could impair translational capacity long after the drug has been cleared systemically. Therefore, further research is needed to determine how 5-FU affects ribosome regulation in muscle during and after treatment, an area that is not well understood. In addition, chemotherapy-induced DNA damage and ribosomal dysfunction can drive muscle stem cells towards cellular senescence, which inhibits protein synthesis. Interestingly, we found that individual chemotherapeutic agents in the FOLFOX cocktail did not acutely inhibit myotube protein synthesis but did suppress protein synthesis during the recovery phase. It is worth considering that this delay may be due to altered ribosomal turnover kinetics. Mature ribosomes have a relatively long half-life and can temporarily support translation during acute drug exposure. However, 5-FU and other agents disrupt the synthesis of new rRNA and tRNA templates. Therefore, as the original pool of healthy ribosomes naturally turns over during recovery, the muscle will be left with dysfunctional translational machinery, which would negatively affect protein synthesis. Additionally, acute exposure to chemotherapeutic agents could cause sublethal mitochondrial damage or stable transcriptional changes that do not immediately disrupt homeostasis but gradually accumulate over time, resulting in a chronic energy deficit and a lasting downregulation of anabolic signaling during recovery.

Autophagy is critical for several skeletal muscle processes involving metabolism, growth, and atrophy, and cachectic skeletal muscles from tumor-bearing mice have altered autophagy flux levels. (27, 57, 58). Importantly, autophagy requires the coordinated regulation of several signaling pathways, including AMPK, Akt, mTORC1, and FOXO signaling pathways (59), which have all been shown to be disrupted in cachectic muscle (60–62). mTORC1 activation has a well-described role in regulating muscle protein synthesis (11, 13) and autophagy initiation (11). Rapamycin is a potent mTORC1 inhibitor and is used clinically as an immunosuppressant during organ transplantation (63). Recently, rapamycin has acquired significant attention for its potential anti-aging and anti-cancer effects (63, 64). Rapamycin administration can increase autophagy in mouse skeletal muscle, which could benefit muscle functionality and metabolic health (65, 66). Our results provide evidence that rapamycin can directly regulate autophagy flux in myotubes and enhance autophagy flux in the presence of FOLFOX. As expected, rapamycin reduced protein synthesis in vehicle-treated myotubes and did not rescue FOLFOX suppression of protein synthesis. Further research is warranted to examine whether the improvement of muscle autophagy flux without rescuing protein synthesis will benefit muscle metabolic health. Interestingly, our results show that rapamycin treatment can inhibit AMPK phosphorylation under baseline conditions; it also rescues the FOLFOX-induced suppression of p-AMPK. As AMPK is not thought to be a direct target of rapamycin (67), it is interesting to speculate that the improvement in AMPK was an indirect effect involving the overall enhancement of muscle ATP production, possibly through greater mitochondrial function or reduced reactive oxygen species formation. We speculate that rapamycin treatment indirectly enhances AMPK phosphorylation by alleviating other FOLFOX-induced metabolic disruptions in the myotubes. However, further research is necessary to test this hypothesis. Additionally, restoring AMPK phosphorylation could help cells better coordinate survival and autophagic clearance pathways during FOLFOX treatment. Since ULK1 can integrate mTORC1 and AMPK signaling to regulate autophagy initiation (68), we also report the interesting finding that FOLFOX alone can suppress ULK1(S555), which was rescued by rapamycin. Overall, our findings suggest that rapamycin may have therapeutic benefits for chemotherapy-induced metabolic disruptions in skeletal muscle, possibly by improving autophagy regulation.

Metformin, a biguanide derivative approved by the FDA to treat type 2 diabetes (68), can increase AMPK phosphorylation in skeletal muscle and liver cells (40, 69). Metformin also reduces colorectal cancer risk in type 2 diabetes patients and is associated with better prognosis in patients diagnosed with CRC (38, 70, 71). Metformin has also been reported to attenuate 5-FU oral tissue toxicities in mice (Sun et al., 2022). While we examined metformin’s ability to rescue FOLFOX suppression of myotube AMPK signaling, metformin’s mechanism of action for activating AMPK remains under investigation (72). Metformin is thought to increase AMPK activity by inhibiting complex I in the mitochondrial electron transport chain, thereby increasing the AMP/ATP ratio and promoting uncoupled respiration (73). However, metformin has been shown to increase AMPK phosphorylation independent of changes in cellular AMP/ATP through lysosomal v-ATPase inhibition (74). Other potential mechanisms for the metformin induction of AMPK include glucagon-like peptide 1 (GLP-1) secretion (75) and mTORC1 inhibition (76). In our investigation, we cannot rule out the possibility that metformin inhibition of complex I could also result in additional undesirable effects in myotubes over the 24-h incubation period. Interestingly, metformin's induction of myotube AMPK and downstream ACC and ULK1 signaling did not lead to changes in autophagy flux. Additionally, FOLFOX treatment inhibited the metformin-induced activation of ULK1. We speculate that the metabolic stress caused by FOLFOX could impair later stages of the autophagy pathway—such as lysosomal acidification or autophagosome-lysosome fusion—resulting in a situation where upstream signaling is active, but actual clearance (flux) cannot occur. Regarding ULK1 regulation, the activating phosphorylation by AMPK at Ser555 can be hierarchically overridden by inhibitory phosphorylation from mTORC1 at Ser757, disconnecting signaling from the actual functional flux. Evidence for the metformin regulation of autophagy is equivocal and seems to be dependent on cell type. Metformin can increase autophagy in tumor cells (77, 78) and activate liver and intestine cell autophagy signaling (79, 80). Metformin dosed in combination with rapamycin has been reported to activate myotube AMPK, but not indices of autophagy (81). Our current results suggest that metformin-induced AMPK activation was insufficient to regulate autophagy in cultured myotubes. While AMPK signaling is involved in many aspects of cellular metabolism, our data suggest that metformin does not rescue FOLFOX-induced deficits in protein synthesis or autophagy flux, indicating that its therapeutic potential for mitigating 5-FU-associated muscle wasting may be limited in this context. However, due to metformin’s many beneficial metabolic effects, further in vivo studies are needed to explore how metformin can impact chemotherapeutic disruptions to systemic and muscle metabolism.

The present study, examining the effects of FOLFOX and CT26 on cultured myotube autophagy flux and protein synthesis regulation, has several limitations and delimitations that should be considered when interpreting the results. First, this study’s outcomes are limited to the mechanistic investigation of cultured myotubes, which allows the determination of the direct effects of CT26 tumor-derived factors and FOLFOX on myotube cellular regulation. Further investigation is required to determine the effects of CT26 and FOLFOX on regulation in the complex in vivo skeletal muscle environment, where direct and indirect effects of chemotherapy on other muscle cell types and systemic regulation can impact these processes. The effect of CT26 and FOLFOX on other critical cell types such as endothelial cells, immune cells, and fibroblasts warrants further investigation. There are also aspects of the in vitro study design that could impact the outcomes, including the tumor type, duration of defined media exposure, and dose of the chemotherapy agents used. To serve as a baseline for the current study, we used, wherever possible, previously published cell culture study designs involving tumor-conditioned media and chemotherapy dosing and exposure (6, 24). Further investigation is also needed to evaluate the in vivo effects of metformin and rapamycin, which could have important systemic interactions with the cancer and treatment environment that impact skeletal muscle.

In conclusion, our study provides evidence that FOLFOX can directly suppress AMPK and mTORC1 regulation of autophagy flux independent of STAT3 activation, and this effect is not readily reversible upon removal of the chemotherapeutic agent. Importantly, oxaliplatin administration alone was sufficient to suppress autophagy flux acutely and after recovery. However, both 5-FU and oxaliplatin treatments led to long-term suppression of protein synthesis after recovery. We also report that rapamycin administration partially restored FOLFOX-induced suppression of myotube autophagy flux. While metformin was sufficient to activate AMPK in FOLFOX-treated cells, it did not rescue the FOLFOX suppression of myotube autophagy flux or protein synthesis, despite increasing AMPK phosphorylation. Neither metformin nor rapamycin treatments were sufficient to rescue suppressed protein synthesis. Therefore, further research is necessary to assess whether FOLFOX-treated myotubes respond to anabolic agents (e.g., amino acids, IGF-1, insulin, osteocalcin) that enhance protein synthesis and help maintain protein homeostasis. Overall, additional studies are needed to evaluate whether rapamycin has therapeutic potential to counteract long-term metabolic toxicities in skeletal muscle caused by FOLFOX chemotherapy.

SUPPLEMENTARY MATERIAL

Supplemental Figures. 1–2. Link: https://figshare.com/s/b835a70e6eee840beb95

Supplemental Table. 1. Link: https://figshare.com/s/ef386f09d05ff4857fa7

DATA AVAILABILITY

The datasets for this study are available from the corresponding author upon reasonable request.