Oncostatin M signaling drives cancer-associated skeletal muscle wasting

Abstract

Progressive weakness and muscle loss are associated with multiple chronic conditions, including muscular dystrophy and cancer. Cancer-associated cachexia, characterized by dramatic weight loss and fatigue, leads to reduced quality of life and poor survival. Inflammatory cytokines have been implicated in muscle atrophy; however, available anticytokine therapies failed to prevent muscle wasting in cancer patients. Here, we show that oncostatin M (OSM) is a potent inducer of muscle atrophy. OSM triggers cellular atrophy in primary myotubes using the JAK/STAT3 pathway. Identification of OSM targets by RNA sequencing reveals the induction of various muscle atrophy-related genes, including Atrogin1. OSM overexpression in mice causes muscle wasting, whereas muscle-specific deletion of the OSM receptor (OSMR) and the neutralization of circulating OSM preserves muscle mass and function in tumor-bearing mice. Our results indicate that activated OSM/OSMR signaling drives muscle atrophy, and the therapeutic targeting of this pathway may be useful in preventing muscle wasting.

Keywords: JAK/STAT3 signaling; cancer cachexia; oncostatin M; skeletal muscle atrophy.

Graphical abstract

Figure 1

OSM promotes cellular atrophy in cultured primary myotubes (A and B) C57BL/6 mice inoculated with LLC (A) or B16 (B) cells were sacrificed 16 or 14 days later, respectively, and changes in gene expression of gastrocnemius muscle were determined by RT-qPCR (n = 5 for the control groups, n = 7 for the LLC group, and n = 5 for the B16 group). (C) Mouse primary myotubes were treated with recombinant OSM (250 ng/mL) for 48 h. Changes in gene expression were determined by RT-qPCR (n = 3 for each group). (D and E) Mouse primary myotubes were transduced with a GFP adenovirus. Cells were treated with recombinant OSM, IL-6, or LIF (each 250 ng/mL) for 48 h and then visualized under the fluorescence microscope. Scale bar, 50 μm (D). Average myotube diameter was measured (n = 4 for each group) (E). (F and G) Mouse primary myotubes were treated with recombinant OSM, IL-6, or LIF (each 250 ng/mL) for 48 h. Gene expression profiles were analyzed by RNA sequencing. The heatmap of significant genes up- or downregulated more than 2-fold is shown (n = 2 for each group) (F). Changes in gene expression were determined by RT-qPCR (n = 3 for each group) (G). Data are presented as mean ± SEM. Statistical analysis was conducted using 2-tailed t test (A–C) and 1-way ANOVA with Tukey’s post hoc test (E and G).

Figure 2

OSM uses JAK/STAT3 signaling to elicit its effects in myotubes (A) Mouse primary myotubes were treated with recombinant OSM, IL-6, or LIF (each 250 ng/mL) for 10 min. Protein levels were determined by western blotting. Blots are representative of 3 independent experiments. (B) Mouse primary myotubes were treated with Rux (2 μM) for 30 min, and then recombinant OSM (250 ng/mL) was added for 10 min. Protein levels were determined by western blotting. Blots are representative of 3 independent experiments. (C) Mouse primary myotubes were treated with Rux (2 μM) and recombinant OSM (250 ng/mL) for 48 h mRNA levels were determined by RT-qPCR (n = 3 for each group). (D–G) Mouse primary myotubes were transduced with LacZ or Stat3-Y705F expressing adenoviruses and treated with recombinant OSM (250 ng/mL) for 48 h. Protein levels were determined by western blotting (D). Blots are representative of 3 independent experiments. mRNA levels were tested by RT-qPCR (n = 3 for each group) (E). Cells were also transduced with a GFP adenovirus for fluorescence imaging. The average myotube diameter was measured (n = 4 for each group) (F). Myotubes were visualized under the fluorescence microscope. Scale bar, 50 μm (G). Data are presented as mean ± SEM. Statistical analysis was conducted using 1-way ANOVA with Tukey’s post hoc test.

Figure 3

OSM overexpression causes muscle atrophy in mice TA muscles of C57BL/6 mice were transduced with LacZ or OSM expressing adenoviruses. Mice were sacrificed 7 days later (n = 6 for each group). (A and B)Tissues were weighed (A) and H&E stained (B). Scale bar, 100 μm. (C and D) Muscle fiber CSA (C) and the fiber frequency distribution were determined (D) (n = 3 for each group). (E–G) Changes in gene expression were determined by RT-qPCR (n = 6 for each group) (E). Protein levels were tested by western blotting. Asterisk indicates nonspecific band (F). Band intensities were quantified (G) (n = 6 for each group). Data are presented as individual measurements (points) and mean ± SEM. Statistical analysis was conducted using 2-tailed t test.

Figure 4

Muscle-specific depletion of OSMR protects from muscle loss Mice inoculated with B16 cells were sacrificed 14 days later (n = 5 for the KO group and n = 7 for other groups). (A–C) Muscle tissues were weighed (A). Forelimb grip strength (B) and inverted screen hanging performance (C) were measured before the sacrifice (n = 6 for the WT group, n = 5 for the KO group, and n = 7 for other groups). (D–F) Gastrocnemius muscle cross-sections were H&E stained (D), and CSA (E) and the fiber frequency distribution (F) were measured (n = 4 for each group). Scale bar, 100 μm. (G and H) Protein levels were tested by western blotting (G), and the band intensities were quantified (H) (n = 3 for each group). Data are presented as individual measurements (points) and mean ± SEM. Statistical analysis was conducted using 2-way ANOVA with Tukey’s post hoc test.

Figure 5

Neutralization of OSM ameliorates tumor-induced muscle wasting (A) Mouse primary myotubes were treated with recombinant OSM (250 ng/mL) and IgG or anti-OSM antibodies (10 μg/mL) for 48 h. Changes in gene expression were determined by RT-qPCR (n = 3 for each group). (B–E) Mice inoculated with LLC cells received IgG or anti-OSM antibody injections and were sacrificed 16 days posttumor inoculation. Tumor (C) and muscle tissues (D) were weighed. Forelimb grip strength was measured before the sacrifice (n = 7 for each group) (E). (F–H) Gastrocnemius muscle cross-sections were H&E stained (G), CSA (F), and the fiber frequency distribution (H) were measured (n = 3 for each group). Scale bar, 100 μm. (I and J) Gastrocnemius muscle protein levels were determined by western blotting. Asterisk indicates the nonspecific band (I). Band intensities were quantified (J) (n = 3 for each group). (K and L) OSMR expression values of normal subjects and PDAC patients were analyzed by GEO2R (GSE130563) (n = 16 for noncancer controls, n = 5 for noncachectic PDAC patients, and n = 17 for cachectic PDAC patients) (K). GSEA of the top 200 OSM target genes was performed by comparing cachectic PDAC patients with noncancer controls and noncachectic PDAC patients (GEO: GSE130563) (L). Data are presented as individual measurements (points) and mean ± SEM. Statistical analysis was conducted using 1-way ANOVA with Tukey’s post hoc test (A, C–F, H, and J) and GEO2R with adjustments for multiple tests (K and L).