Short-term calorie restriction enhances skeletal muscle stem cell function

Abstract

Calorie restriction (CR) extends life span and ameliorates age-related pathologies in most species studied, yet the mechanisms underlying these effects remain unclear. Using mouse skeletal muscle as a model, we show that CR acts in part by enhancing the function of tissue-specific stem cells. Even short-term CR significantly enhanced stem cell availability and activity in the muscle of young and old animals, in concert with an increase in mitochondrial abundance and induction of conserved metabolic and longevity regulators. Moreover, CR enhanced endogenous muscle repair and CR initiated in either donor or recipient animals improved the contribution of donor cells to regenerating muscle after transplant. These studies indicate that metabolic factors play a critical role in regulating stem cell function and that this regulation can influence the efficacy of recovery from injury and the engraftment of transplanted cells.

Figure 1. Skeletal muscle stem cell frequency and function are enhanced in CR-treated muscle

(A) Representative flow cytometric analysis of satellite cells from young (5 months of age) CR-treated and control C57BL/6 mice. Plots shown depict CXCR4 and β1-integrin staining of Sca1⁻CD45⁻Mac1⁻ cells previously gated also by scatter and vital dyes. Bar chart at right shows the frequency (mean ± SD) of CXCR4⁺β1-integrin⁺ satellite cells among CD45⁻Sca1⁻Mac1⁻ myogenic (non-hematopoietic and non-fibrogenic (Joe et al., 2010; Sherwood et al., 2004; Uezumi et al., 2010)) cells harvested from skeletal muscle. Data compiled from analysis of n=7 CR-treated and n=10 control mice. (B) Average frequency (mean ± SD) of CD45⁻Sca1⁻Mac1⁻CXCR4⁺β1-integrin⁺ satellite cells among total live (propidium iodide-; calcein blue+) mononuclear cells harvested from skeletal muscle was determined by flow cytometry (as in A, data representative of n=7 CR-treated mice and n=4 control mice). (C) Yield of CD45⁻Sca1⁻Mac1⁻CXCR4⁺β1-integrin⁺ satellite cells per gram of muscle following sorting from CR-treated or control muscle. Data compiled from n=5–7 sorts per group. (D) Immunofluorescence staining for Pax7 in freshly sorted CD45⁻Sca1⁻Mac1⁻CXCR4⁺β1-integrin⁺ satellite cells from CR-treated and control mice shows equivalent percentage of positive-staining cells (right). Data represent analysis of 100 cells per group from CR-treated (black bars) or Ctl (grey bars) mice. (E) Frequency (mean ± SD) of Sca-1⁻CD45⁺Mac1⁺ inflammatory cells in muscle was determined by flow cytometry (CR-treated, black bars; Ctl, grey bars; n=8 mice per condition). (F) Immunofluorescence staining of Pax7⁺ satellite cells on single, isolated myofibers prepared from young CR-treated and control mice. Numbers of Pax7⁺ cells per fiber were quantified by counting >20 fibers from each of 4 CR-treated and 4 control mice (total of 80–160 fibers per experimental condition). Data represent mean ± SEM. (G, H) Clonal myogenesis assays of double-sorted satellite cells from young control or CR-treated mice. Data represent mean ± SD and reflect the percent of wells seeded with 1 satellite cell that contained a myogenic colony at day 5–6. Data are compiled from clonal assays of 4–7 individual mice per group, and report comparison of cells from control vs. CR-treated mice in glucose-containing medium (G) or from control mice only cultured in media containing glucose, galactose, or galactose and etoxomir, an inhibitor of mitochondrial energy production (H). (I) Immunofluorescence images showing staining with Mitotracker Green (green) and SOD2 (red). Nuclei were marked by DAPI staining (blue). CR-treated satellite cells showed increased mitochondrial abundance. (J) Oxygen consumption rate (OCR, pMoles/min, top panel) and glycolytic activity, measured by extracellular acidification rate (ECAR, mpH/min, bottom panel), of satellite cells isolated from CR or control mice were measured using a Seahorse Bioscience extracellular flux analyzer (XF24). (K) Immunofluorescence staining for Sirt1, Foxo3a and activated Notch1 (assessed by staining for the Notch intracellular domain (NICD), which is generated by cleavage of Notch receptor following ligand binding) in freshly sorted CD45⁻Sca1⁻Mac1⁻CXCR4⁺β1-integrin⁺ satellite cells from CR-treated and control mice. Percentage of positive-staining cells is shown at right for 100 cells analyzed from CR-treated (black bars) or Ctl (grey bars) mice. (L) Western blot analysis for the mitochondrial proteins cytochrome c and prohibitin in total protein lysate from satellite cells harvested from CR-treated or control mice (n=4 mice per group). GAPDH served as a loading control. See also Figures S1 and S2.

Figure 2. Enhanced muscle repair activity of satellite cells following CR-treatment

Experimental design for each study is shown at left. (A) TA muscles of young CR-treated or control C57BL/6 mice were freeze-injured and harvested for histology 7 days after injury. Muscle regeneration was quantified on hematoxylin/eosin (H&E) tissue sections as the mean number of centrally nucleated myofibers per mm² (n=7 CR and n=6 Ctl, mean ± SD), and indicate an ~15% increase in the number of newly formed fibers in the regenerated tissue of CR mice. These findings are consistent with other studies (McCarthy et al., 2011; Murphy et al., 2011) indicating a correlative, but not necessarily quantitative, relationship between the endogenous content of muscle satellite cells (increased ~2-fold in CR-treated mice, Figure 1C,F) and the number of regenerating myofibers after injury. (B) 8000 double-sorted satellite cells from CR-treated (n=4) or control (n=3) mice were injected intramuscularly into mdx recipients injured 1 day previously by injection of cardiotoxin (CDTX) into the same TA muscle. Transverse cryosections of muscles receiving CR-treated (left) or control (right) satellite cells were stained with anti-dystrophin antibody (red; middle panels). Dystrophin normally is lacking in mdx muscle (Sicinski et al., 1989), and therefore serves as a marker of donor cell contributions to muscle regeneration. Data are presented as the mean (± SD) number of dystrophin⁺ myofibers found in each engrafted muscle. (C) 5000 double-sorted GFP-expressing satellite cells from control mice were injected into the TA of CR- or control-treated C57BL/6 mice (n=4 per group), injured 1 day previously by CTDX injection. Transverse cryosections from muscles harvested 4 weeks after transplant were analyzed for GFP expression by direct epifluorescence. Data are presented as the mean (± SD) number of GFP⁺ myofibers (green; middle panels) found in each engrafted muscle. Data were considered statistically significant at p<0.05 and all p-values were calculated by Student’s t-test. Note that overall levels of engraftment into mdx recipients are superior to those in wild-type recipients (compare (B) to (C)), likely due to differences in endogenous muscle stability, regenerative activity, and host muscle stem cell number (Cerletti et al., 2008a).