Extracellular serine and glycine are required for mouse and human skeletal muscle stem and progenitor cell function

Abstract

Objective: Skeletal muscle regeneration relies on muscle-specific adult stem cells (MuSCs), MuSC progeny, muscle progenitor cells (MPCs), and a coordinated myogenic program that is influenced by the extracellular environment. Following injury, MPCs undergo a transient and rapid period of population expansion, which is necessary to repair damaged myofibers and restore muscle homeostasis. Certain pathologies (e.g., metabolic diseases and muscle dystrophies) and advanced age are associated with dysregulated muscle regeneration. The availability of serine and glycine, two nutritionally non-essential amino acids, is altered in humans with these pathologies, and these amino acids have been shown to influence the proliferative state of non-muscle cells. Our objective was to determine the role of serine/glycine in MuSC/MPC function.

Methods: Primary human MPCs (hMPCs) were used for in vitro experiments, and young (4-6 mo) and old (>20 mo) mice were used for in vivo experiments. Serine/glycine availability was manipulated using specially formulated media in vitro or dietary restriction in vivo followed by downstream metabolic and cell proliferation analyses.

Results: We identified that serine/glycine are essential for hMPC proliferation. Dietary restriction of serine/glycine in a mouse model of skeletal muscle regeneration lowered the abundance of MuSCs 3 days post-injury. Stable isotope-tracing studies showed that hMPCs rely on extracellular serine/glycine for population expansion because they exhibit a limited capacity for de novo serine/glycine biosynthesis. Restriction of serine/glycine to hMPCs resulted in cell cycle arrest in G0/G1. Extracellular serine/glycine was necessary to support glutathione and global protein synthesis in hMPCs. Using an aged mouse model, we found that reduced serine/glycine availability augmented intermyocellular adipocytes 28 days post-injury.

Conclusions: These studies demonstrated that despite an absolute serine/glycine requirement for MuSC/MPC proliferation, de novo synthesis was inadequate to support these demands, making extracellular serine and glycine conditionally essential for efficient skeletal muscle regeneration.

Keywords: Glycine metabolism; Muscle; Muscle metabolism; Muscle progenitor cell; Muscle regeneration; Muscle stem cell; Proliferation; Protein synthesis; Serine metabolism.

Figure 1

Serine/glycine availability impacted hMPC population expansion in vitro and mouse MuSC abundance in vivo. A) Live cell count was determined by co-staining cells with Hoechst 33342 (to identify all of the cells) and propidium iodide (to identify dead cells) after hMPCs were cultured in media lacking serine/glycine or increasing concentrations of serine/glycine (∗∗indicates significant difference between 100 μM vs -serine/glycine, p < 0.01, ∗∗∗indicates significant difference between 100 μM vs -serine/glycine, p < 0.001, ˆindicates significant difference between 1000 μM vs -serine/glycine, p < 0.05, ˆˆˆindicates significant difference between 1000 μM vs -serine/glycine, and p < 0.001). B) Levels of serine and glycine in hMPCs after 7 days of culture in serine/glycine restricted media as determined by GC–MS. C) Levels of serine and glycine in the plasma of fasted young humans (n = 37) based on GC–MS measurements. D) Schematic of dietary intervention and skeletal muscle injury design as well as subsequent analysis. IHC, immunohistochemistry. E) Serine and glycine concentrations in skeletal muscle tissue after 4 weeks of the control or Ser/Gly_depleted diet in mice (4–6 mo) as measured by GC–MS (n = 3–7 per group). F) Serine and glycine concentrations in plasma after 4 weeks of the control or Ser/Gly_depleted diet in mice (4–6 mo) as measured by GC–MS (n = 3–6 per group). G) Representative images of cross-sections of TA muscles at 3 dpi in mice (4–6 mo) fed either the control or Ser/Gly_depleted diet stained for DAPI (blue), laminin (red), and Pax7 (white). White arrows identify Pax7⁺ staining. H) Quantification of number of Pax7⁺ cells in cross-sections of injured TA muscles 3 dpi in mice (4–6 mo) fed either the control or Ser/Gly_depleted diet (n = 3 per group). All the experiments were repeated with hMPCs derived from the same 5 donors. P values or marks of significance are indicated in the appropriate graphs.

Figure 2

hMPCs exhibited limited capacity for serine/glycine biosynthesis. A) Heatmap of genes involved in serine, glycine, and one-carbon metabolism based on RNA-seq data from hMPCs after 5 days of culture in serine/glycine replete (1000 μM) or restricted conditions. B) Immunoblot for PHGDH, PSAT1, PSPH, SHMT1, and SHMT2 protein normalized to α-TUBULIN for quantification in hMPCs that had been serine/glycine restricted for 5 days. C) Glucose uptake by hMPCs after 5 days of culture in serine/glycine replete or restricted conditions. D) Live cell count of hMPCs cultured in serine/glycine replete media or serine/glycine restricted media and varying doses of glucose. ∗p < 0.05 and ∗∗p < 0.01 relative to serine/glycine containing control. Data are expressed as mean ± SD. E) Percent mass isotopomer distribution of [U–¹³C₆]-glucose-derived serine and glycine in hMPCs cultured in serine/glycine replete (1000 μM) or restricted media for 5 days followed by 48 h in similar media but containing [U–¹³C₆]-glucose. Data are expressed as mean ± SD. ∗∗p < 0.01 and ∗∗∗p < 0.001. All the experiments were repeated with hMPCs derived from the same 5 donors. P values or marks of significance are indicated in the appropriate graphs.

Figure 3

Serine/glycine restriction caused cell cycle arrest in G0/G1 in hMPCs. A) Percentage of dead cells was quantified by dividing all the hMPCs that stained positive for propidium iodide by all of the hMPCs that stained positive for Hoechst 33342 after 5 days of culture in media with varying concentrations of serine/glycine. B) Propidium iodide staining and analysis via flow cytometry was used to determine the DNA content of hMPCs after 5 days of serine/glycine restriction. ∗p < 0.05. C) BrdU incorporation after a 24-h pulse in hMPCs undergoing serine/glycine restriction for 5 days. Data are expressed as mean ± SD. D) Immunoblot for PAX7, MYOD, and MYOGENIN protein normalized to α-TUBULIN for quantification in hMPCs that had been serine/glycine restricted for 5 days. ∗∗p < 0.01 and ∗∗∗p < 0.001. N.D., not detectable. Data are expressed as mean ± SD. E) The number of live cells after culture in serine/glycine restricted followed by serine/glycine replete media. All the experiments were repeated with hMPCs derived from the same 5 donors.

Figure 4

Serine/glycine restriction reduced intracellular glutathione levels that are necessary for hMPC proliferation. A) The glutathione synthesis pathway with genes upregulated after serine/glycine restriction according to RNA-seq is in blue. CBS, cystathionine β-synthase; CTH, cystathionine gamma-lyase; GCLC, glutamate-cysteine ligase catalytic subunit; GSS, glutathione synthetase. B) Total intracellular glutathione levels in hMPCs after 5 days in serine/glycine restricted or replete media. C) Ratio of reduced to oxidized intracellular glutathione levels in hMPCs after 5 days in serine/glycine restricted or replete media. GSH, reduced glutathione; GSSG, oxidized glutathione. D) ROS in hMPCs after 5 days in serine/glycine restricted or replete media. E) Total intracellular glutathione levels in hMPCs after 5 days of serine/glycine restriction with or without 10 μM of cell permeable glutathione ethyl ester (GSHee). F) ROS in hMPCs after 5 days of serine/glycine restriction with or without 10 μM of GSHee. G) Live cell count after 5 days of serine/glycine restriction with or without GSHee (10 μM). H) Total intracellular glutathione levels in hMPCs after 5 days of serine/glycine restriction or 1000 μM of serine/glycine and 100 μM of glutathione inhibitor l-buthionine sulfoximine (GSHi). I) Live cell count after 5 days of hMPCs cultured in serine/glycine replete media, serine/glycine restricted media, or a combination of serine/glycine replete media and 100 μM of GSHi. ∗p = 0.05 and ∗∗p < 0.01. All the experiments were repeated with hMPCs derived from the same 5 donors. P values or marks of significance are indicated in the appropriate graphs.

Figure 5

Serine/glycine restriction depressed global protein synthesis and hMPC proliferation in a p-EIFα-dependent manner. A) Expression of known ATF4 targets based on RNA sequencing of hMPCs cultured in serine/glycine restricted media or serine/glycine replete media for 5 days. Average log fold change of transcripts in serine/glycine restricted samples vs serine/glycine replete samples. Dark blue indicates log fold change < −1 and light blue indicates log fold change >1. B) Top, protein levels of ATF4, p-eIF2α, and total eIF2α determined by immunoblot hMPCs cultured in serine/glycine replete or restricted media. Expression was initially normalized to α-TUBULIN expression and p-eIF2α normalized to total eIF2α expression. Bottom, quantification of protein expression. C) Intracellular amino acid abundance normalized to the total ion count in hMPCs cultured in serine/glycine replete or restricted media for 5 days. D) Protein synthesis as determined by O-propargyl-puromycin incorporation of hMPCs cultured in serine/glycine replete or restricted media. E) Live cell count after 5 days of serine/glycine restriction with or without varying doses of ISRIB. ∗p < 0.05 and ∗∗p < 0.01. All the experiments were repeated with hMPCs derived from the same 5 donors. P values are indicated in the appropriate graphs.

Figure 6

Dietary restriction of serine/glycine disrupted cell numbers and impaired muscle regeneration in the aged mice. A) Skeletal muscle serine levels negatively correlated with age (n = 58, r = −.44, and p < 0.0001) based on Pearson's correlation coefficient. B) Schematic of dietary intervention and skeletal muscle injury design as well as subsequent analysis. IHC, immunohistochemistry. C) Serine and glycine concentrations in skeletal muscle tissue after 4 weeks of the control or Ser/Gly_depleted diet in the mice (>20 mo) as measured by GC–MS (n = 3–7 per group). D) Serine and glycine concentrations in plasma after 4 weeks of the control or Ser/Gly_depleted diet in the mice (>20 mo) as measured by GC–MS (n = 3–7 per group). E) Representative images of cross-sections of TA muscles at 3 dpi in the mice (>20 mo) fed either the control or Ser/Gly_depleted diet stained for DAPI (blue), laminin (red), and Pax7 (white). White arrows identify Pax7⁺ staining. F) Quantification of the number of Pax7⁺ cells in cross-sections of injured TA muscles at 3 dpi in the mice (>20 mo) fed either the control or Ser/Gly_depleted diet (n = 3 per group). G) Representative images of hematoxylin and eosin (H & E) stained TA muscles at 28 dpi in the old mice fed the control or Ser/Gly_depleted diet. H) Quantification of the distribution of myofiber size of TA muscles at 28 dpi in the old mice fed the control or Ser/Gly_depleted diet (n = 3 per group). I) Representative images of cross-sections of TA muscles at 28 dpi in the old mice fed either the control or Ser/Gly_depleted diet stained for DAPI (blue) and perlipin-1 (green).