CCL2 signaling promotes skeletal muscle wasting in non-tumor and breast tumor models

Abstract

Despite advancements in treatment, approximately 25% of patients with breast cancer experience long-term skeletal muscle wasting (SMW), which limits mobility, reduces drug tolerance and adversely impacts survival. By understanding the underlying molecular mechanisms of SMW, we may be able to develop new strategies to alleviate this condition and improve the lives of patients with breast cancer. Chemokines are small soluble factors that regulate homing of immune cells to tissues during inflammation. In breast cancers, overexpression of C-C chemokine ligand 2 (CCL2) correlates with unfavorable prognosis. Elevated levels of CCL2 in peripheral blood indicate possible systemic effects of this chemokine in patients with breast cancer. Here, we investigated the role of CCL2 signaling on SMW in tumor and non-tumor contexts. In vitro, increasing concentrations of CCL2 inhibited myoblast and myotube function through C-C chemokine receptor 2 (CCR2)-dependent mechanisms involving JNK, SMAD3 and AMPK signaling. In healthy mice, delivery of recombinant CCL2 protein promoted SMW in a dose-dependent manner. In vivo knockdown of breast tumor-derived CCL2 partially protected against SMW. Overall, chronic, upregulated CCL2-CCR2 signaling positively regulates SMW, with implications for therapeutic targeting.

Keywords: Breast cancer; CCL2; Cachexia; Chemokine; Skeletal muscle wasting.

Fig. 1.

CCL2 inhibits activity in myoblasts and myotubes at high concentrations. (A,B) C2C12 myoblasts were treated with serum-free (SF) DMEM or DMEM containing 2% horse serum (HS), 40 or 80 ng/ml CCL2, or 4T1 tumor conditioned medium (4T1-CM) for 24 h and analyzed for proliferation by PCNA immunostaining (red) (A), or analyzed for myotube formation through the assessment of myotube fusion index (using an antibody against myosin heavy chain or MYH, red) for up to 96 h (B). (C-E) C2C12 myotubes were treated with SF medium, DMEM containing 2% HS, 40 or 80 ng/ml CCL2, or 4T1-CM. (C) Samples were analyzed for expression of the indicated proteins by immunoblotting. Densitometry was performed on immunoblots. Band intensities of samples were normalized to those of actin and expressed as a percentage relative to those of untreated samples. The arrow indicates the bands analyzed. (D) Intracellular lactate production was measured by biochemical assay. Lactate levels in SF samples were below the limit of detection. (E) The degree of apoptosis was analyzed by cleaved caspase-3 immunostaining. For A,B,E, biomarker expression was quantified by ImageJ (arbitrary units); samples were plated in triplicate and experiments were repeated three times (n=9/group). For C, experiments were repeated five times (n=5/group). For D, samples were plated in triplicate and performed four times (n=12/group). Statistical analysis was performed using one-way ANOVA with Bonferroni’s post hoc comparison (A,B,D,E) or Kruskal–Wallis test with Wilcoxon rank sum post hoc analysis (C). Relevant post hoc comparisons are indicated by lines. Significance was determined by P<05. Means±s.e.m. are shown. Scale bars: 100 μm.

Fig. 2.

CCL2 knockdown prevents mammary carcinoma cells from inhibiting myoblast and myotube activity. (A) 4T1 cells were transfected with control siRNA (4T1.Ctrl) or CCL2 siRNA (4T1.CCL2si) and analyzed for CCL2 levels by ELISA. (B,C) C2C12 myoblasts were treated with serum-free (SF) medium or with conditioned medium from 4T1 cells transfected with control siRNA (4T1.Ctrl-CM) or CCL2 siRNA (4T1.CCL2si-CM) for 24 h and analyzed for cell proliferation by PCNA immunostaining (B) or for differentiation into myotubes through assessment of myotube fusion index (C). (D) C2C12 myotubes were treated with tumor conditioned medium and analyzed for changes in the expression of atrogin-1, MuRF1 and LC3B by immunoblotting. Densitometry was performed on immunoblots. Band intensities of samples were normalized to those of actin and expressed as a percentage relative to those of 4T1.Ctrl-CM samples (D). (E) Intracellular lactate production was measured by biochemical assay. (F) The degree of apoptosis was determined by cleaved caspase-3 immunostaining. For A-C,E,F, biomarker expression was quantified by ImageJ (arbitrary units); samples were plated in triplicate and experiments were performed three times (n=9/group). For D, experiments were repeated five times (n=5/group). Statistical analysis was performed using two-tailed unpaired t-test (A,E,F), one-way ANOVA with Bonferroni's post hoc comparison (B,C) or Wilcoxon rank sum test (D). For P<0.05 determined by ANOVA, relevant post hoc comparisons are indicated by lines. Statistical significance was defined by P<0.05. Means±s.e.m. are shown.

Fig. 3.

CCR2 inhibition by INCB3344 treatment rescues myoblast proliferation, myotube formation and myotube function. (A,B) C2C12 myoblasts were treated with 80 ng/ml CCL2 with increasing concentrations of INCB3344 for 24 h and analyzed for cell proliferation by PCNA immunostaining (A) or for myotube formation through the assessment of myotube fusion index (B). (C) C2C12 myotube cultures were treated with CCL2 with or without INC3344 and analyzed for the expression of atrogin-1, MuRF1 or LC3B by immunoblotting. Densitometry was performed on immunoblots. Band intensities of samples were normalized to those of actin and expressed as a percentage relative to those of untreated cells. (D) Intracellular lactate production was measured by biochemical assay. (E) The degree of apoptosis was determined by cleaved caspase-3 immunostaining. Biomarker expression was quantified by ImageJ (arbitrary units). For A,B,D,E, samples were plated in triplicate and experiments were performed three times (n=9/group). For C, experiments were repeated five times (n=5/group). Statistical analysis was measured using one-way ANOVA with Bonferroni’s post hoc comparison (A,B,D,E) or Kruskal–Wallis test with Wilcoxon rank sum post hoc analysis (C). Statistical significance was defined as P<0.05. For P>0.05 determined by ANOVA, the P-value is shown in the top left corner of the graph. For P<0.05 determined by ANOVA, relevant post hoc comparisons are indicated by lines. Means±s.e.m. are shown.

Fig. 4.

CCL2 negatively regulates myoblast proliferation, myotube formation and myotube activity through JNK-, SMAD3- and AMPK-dependent mechanisms. (A,B) C2C12 myoblasts were treated with 80 ng/ml CCL2 in the presence or absence of 10 mM SP600125 (JNK inhibitor), 50 nM SIS3 (SMAD3 inhibitor) or 1.5 mM BAY3827 (AMPK inhibitor) and analyzed for changes in proliferation by PCNA immunostaining (A) or for myotube formation through the assessment of myotube fusion index (B). (C) C2C12 myotubes were treated with 80 ng/ml CCL2 in the presence or absence of SP600126, SIS3 or BAY3827 and analyzed for expression of the indicated proteins by immunoblotting. Densitometry was performed on immunoblots. Band intensities of samples were normalized to those of actin and expressed as a percentage relative to those of untreated cells. (D) Intracellular lactate production was measured by biochemical assay. (E) The degree of apoptosis was determined by cleaved caspase-3 immunostaining. Biomarker expression was quantified by ImageJ (arbitrary units). For A,B,D,E, samples were plated in triplicate and experiments were performed three times (n=9/group). For C, experiments were repeated five times (n=5/group). Statistical analysis was performed using one-way ANOVA with Bonferroni's post hoc comparison (A,B,D,E) or Kruskal–Wallis test with Wilcoxon rank sum post hoc analysis (C). Relevant post hoc comparisons are indicated by lines. Statistical significance was determined by P<0.05. Means±s.e.m. are shown.

Fig. 5.

In vivo delivery of high-dose CCL2 reduces grip strength and fat mass. (A) Daily schedule of CCL2 delivery and assessment of muscle strength and body mass composition. Healthy, tumor-free, female BALB/c mice received intramuscular (IM) injections of recombinant murine CCL2 (10 or 50 ng) or saline vehicle control twice a week (n=9/group). Mice were then subjected to grip strength tests and EchoMRI once a week, the day after the second intramuscular injection. Baseline measurements were taken at week 0. Mice were assessed during weeks 1-4 and then euthanized for tissue harvest. (B) Grip strength normalized to body mass is shown for all four limbs (left graph) and hindlimbs (right graph). Grip strength units are defined by ounce-force (ozf). (C,D) Lean mass (C) and fat mass (D) were assessed by EchoMRI. Lean and fat masses relative to body weight were normalized to baseline measurements (week 0). (E-H) Quadricep muscle tissues were immunostained for CCL2 (E), atrogin-1 (F), MuRF1 (G) or LC3B (H). Longitudinal sections are shown. (I) Analysis of cross-sectional area of quadricep muscle tissues stained with Hematoxylin and Eosin (H&E). Biomarker expression or cross-sectional area was quantified using ImageJ (arbitrary units). Statistical analysis was performed using one-way ANOVA with Bonferroni's post hoc comparison. Statistical significance was defined by P<0.05. For B-D, the P-value determined by ANOVA for weeks 1 and 4 is shown on the top left corner of each line graph. For E-I, the P-value determined by ANOVA is shown on the top left corner of each bar graph. Means±s.e.m. are shown. Scale bars: 100 μm.

Fig. 6.

Tumoral CCL2 siRNA knockdown rescues lean mass but not grip strength. (A) Daily schedule of Ca-TAT/siRNA delivery and assessment of muscle strength and body mass composition. BALB/c female mice bearing 4T1 mammary carcinomas received intra-tumoral injections of 10 µg Ca-TAT peptides complexed to control (Ctrl, n=13) or CCL2 (CCL2si, n=13) siRNAs, once a day for 3 days. Mice were subject to grip strength tests and EchoMRI one day after the last injection of Ca-TAT/siRNA complexes. Tumors were established at week 0. Intra-tumoral injections were performed during weeks 1 to 3, and then mice were euthanized for tissue harvest. Healthy normal BALB/c mice were used as a baseline control (n=14). (B) Grip strength normalized to body mass is shown for all four limbs (left) and hindlimbs (right). Grip strength units are defined by ounce-force (ozf). (C,D) Lean mass (C) and fat mass (D) were assessed by EchoMRI. The percentages of lean mass and fat mass were calculated by dividing the lean or fat mass by total body mass and multiplying by 100. (E,F) Total body mass (E) or tumor mass (F) at endpoint is shown for the indicated groups. Healthy, untreated mice were used as a baseline control (n=13). Statistical analysis was performed using one-way ANOVA with Bonferroni's post hoc analysis (B-E) or two-tailed unpaired t-test (F). For B-F, the P-value determined by ANOVA is shown on the top left corner of each graph. Statistical significance was defined by P<0.05. Means±s.e.m. are shown.

Fig. 7.

Intra-tumoral knockdown of CCL2 reduces expression of biomarkers associated with skeletal muscle atrophy. (A) Blood serum levels of CCL2 from mice at endpoint were measured by ELISA. (B) Immunostaining was performed for CCL2 in mammary epithelial tissues. (C-F) Immunostaining was performed for CCL2 (C), atrogin-1 (D), MuRF-1 (E) or LC3B (F) in quadricep muscle tissues. Longitudinal sections are shown. (G) The cross-sectional area was analyzed for quadricep muscle stained with H&E. Representative images from eight mice per group are shown. Biomarker expression or cross-sectional area was quantified using ImageJ (arbitrary units). Statistical analysis was measured using two-tailed unpaired t-test (A) or one-way ANOVA with Bonferroni's post hoc comparison (B-G). Statistical significance was defined P<0.05. For A,C-G, the P-value determined by ANOVA is shown on the top left corner of each graph. Means±s.e.m. are shown. Scale bars: 100 μm.

Fig. 8.

Tumoral knockdown of CCL2 increases neutrophils and T helper cells in skeletal muscle tissues. 4T1 tumor-bearing mice were treated with Ca-TAT peptides complexed to control (Ctrl) or CCL2 (CCL2si) siRNAs. At endpoint, gastrocnemius muscle tissues were harvested, digested into single-cell suspensions and analyzed by flow cytometry for the presence of (A) macrophages (F4/80⁺), neutrophils (F4/80⁻/Ly6G⁺) and T cells (CD3⁺); (B) macrophage subpopulations – M1 (F4/80⁺ CD11b⁺ Ly6C⁺ CD11c⁺ or F4/80⁺ CD11b⁺ Ly6C⁺ CD80⁺) or M2 (F4/80⁺ CD11b⁺ Ly6C⁺ CD206⁺); (C) neutrophil subpopulations – N1 (F4/80⁻ Ly6G⁺ CD11b⁺ CD11c⁺ or F4/80⁻ Ly6G⁺ CD11b⁺ CD80⁺) or N2 (F4/80⁻/Ly6G/CD11b/CD206); and (D) T cell subpopulations – T helper cells (CD3⁺ CD4⁺ CD25⁻), T regs (CD3⁺ CD4⁺ CD25⁺) or cytotoxic T lymphocytes (CTLs) (CD3⁺ CD8a⁺). n=8/group. Skeletal muscle tissues from healthy, tumor free mice were used as a baseline control. Statistical analysis was performed using one-way ANOVA with Bonferroni's post hoc comparison. Statistical significance was determined by P<0.05. For A-D, the P-value determined by ANOVA is shown for each immune group comparison. Means±s.e.m. are shown.