Fragile X mental retardation protein regulates skeletal muscle stem cell activity by regulating the stability of Myf5 mRNA

Abstract

Background: Regeneration of adult tissues relies on adult stem cells that are primed to enter a differentiation program, while typically remaining quiescent. In mouse skeletal muscle, these features are reconciled by multiple translational control mechanisms that ensure primed muscle stem cells (MuSCs) are not activated. In quiescent MuSCs, this concept is illustrated by reversible microRNA silencing of Myf5 translation, mediated by microRNA-31 and fragile X mental retardation protein (FMRP).

Methods: In this work, we take advantage of FMRP knockout (Fmr1 ^-/- ) mice to support the role for FMRP in maintaining stem cell properties of the MuSC. We compare the activity of MuSCs in vivo after acute injury and engraftment, as well as ex vivo during culture. We use RNA immunoprecipitation and 3'UTR poly-adenine (poly(A)) length assays to assess the impact of FMRP on the stability of transcripts for myogenic regulatory factors.

Results: We show that RNA-binding FMRP is required to maintain the MuSC pool. More specifically, FMRP is required for stem cell properties of muscle stem cells, which include MuSC capacity to prime the myogenic program, their self-renewal, and their capacity to efficiently regenerate muscle. We provide evidence that FMRP regulation of MuSC activity occurs in part by the capacity of FMRP to directly bind Myf5 transcripts and impact rates of Myf5 deadenylation.

Conclusions: Our results provide further evidence supporting a role for post-transcriptional silencing platforms by RNA-binding proteins in maintaining stemness properties of adult stem cells. In addition, deregulated MuSC activity in the absence of Fmr1 may have implications for fragile X syndrome, which is associated with muscle hypotonia during infancy.

Keywords: Fragile X mental retardation protein; Muscle stem cell; Myogenic regulatory factor Myf5; Satellite cell.

Fig. 1

Skeletal muscle of Fmr1 ^−/− mice have reduced numbers of satellite cells. a Immunostaining of PAX7 (green) on newly isolated single EDL myofibers isolated from adult wild-type (upper panel) and Fmr1 ^−/− (lower panel) mice. b Numbers of PAX7-positive satellite cells per EDL myofiber isolated from wild-type (white) and Fmr1 ^−/− (gray) mice. c Immunostaining of PAX7 (red) and bungarotoxin (Btx) labeling of neuromuscular junctions (NMJ) on single EDL myofibers isolated from wild-type and Fmr1 ^−/− mice. d Numbers of NMJs per EDL myofiber. e Levels of Pax7, Myf5, and MyoD transcripts, relative to Actb, in satellite cells isolated from adult muscle of wild-type (wt, black) and Fmr1 ^−/− (red) mice. f Fraction of PAX7(+) satellite cells that immunostain for MYF5 on single EDL myofibers isolated from muscle of wild-type (wt) and Fmr1 ^−/− mice. g Fraction of PAX7(+) satellite cells that immunostain for MYOD on single EDL myofibers isolated from muscle of wild-type (wt) and Fmr1 ^−/− mice. All values indicate mean (n ≥ 3) ± s.e.m. *p < 0.05, ***p < 0.001. ns not significant

Fig. 2

FMRP is required for efficient priming of muscle stem cells to rapidly activate the myogenic program, self-renewal, and differentiation. a Fraction of PAX7(+) satellite cells accumulating MYF5 (left) and MYOD (right) after 1 h culture of single EDL myofibers. b Representative immunostaining of PAX7 (green) and MYOD (red) on single EDL myofibers isolated from adult wild-type and Fmr1 ^−/− mice after 48 h culture. c Numbers of satellite cells undergoing self-renewal (PAX7+MYOD−), activating the myogenic program (PAX7+MYOD+) and differentiating (PAX7−MYOD+) on single EDL myofibers isolated from adult wild-type (white) and Fmr1 ^−/− (gray) mice after 2-day culture. d Levels of MyoD, Myog, and Myh3 transcripts, relative to Actb, in satellite cells isolated from adult muscle of wild-type (wt, black) and Fmr1 ^−/− (gray) mice, after 4-day culture. e Levels of MYOG protein, relative to ACTB, from cell lysates of 4-day cultured satellite cells isolated from muscle of wt (black) and Fmr1 ^−/− (white) mice. Representative immunoblot is shown. MYOG levels are calculated by densitometry of three immunoblots. f Levels of Pax7 and Myf5 transcripts, relative to Actb, in satellite cells isolated from adult muscle of wild-type (wt, black) and Fmr1 ^−/− (gray) mice, 2 days after ctx injury. g Representative images of EdU(+) MyoD(+) satellite cells present on transverse sections of TA muscle, 3 days after ctx injury. White arrows indicate MYOD(+) EdU(+) satellite cells. Scale bar, 50 μm. h Fraction of MYOD(+) satellite cells that are EdU(+), as indicated in g. All values indicate mean (n ≥ 3) ± s.e.m. *p < 0.05, ***p < 0.001

Fig. 3

Delayed regeneration after acute injury of muscle in Fmr1 ^−/− mice. a Immunostaining laminin (Lam, green), embryonic myosin heavy chain (embMHC, red) on transverse sections of TA muscle isolated from wild-type (left), and Fmr1 ^−/− (right) mice 10 days after injury. Scale bar, white 50 μm. b Mean myofiber cross-section area (CSA) of wild-type (white) and Fmr1 ^−/− TA muscle 10 days after injury. Numbers of myofibers with greater area than the indicated bin label are shown. c Immunostaining laminin (Lam, red), embryonic myosin heavy chain (embMHC, green) on transverse sections of TA muscle isolated from wild-type (left), and Fmr1 ^−/− (right) mice 21 days after injury. Scale bar, white 50 μm. d Mean myofiber cross-section area (CSA) of wild-type (white) and Fmr1 ^−/− TA muscle 21 days after injury. Numbers of myofibers with greater area than the indicated bin label are shown. e Immunostaining PAX7 (green) laminin (red) on transverse sections of TA muscle isolated from wild-type (upper panel) or Fmr1 ^−/− (lower panel) 21 days after cardiotoxin injury. f Number of PAX7+ satellite cells present underneath the basal lamina, 21 days after injury of the TA muscle of wild-type (white) and Fmr1 ^−/− (gray) mice. Values indicate mean (n ≥ 3) ± s.e.m. *p < 0.05

Fig. 4

Satellite cells isolated from Fmr1 ^−/− mice have reduced the capacity to self-renew and contribute to the satellite cell pool after intramuscular engraftment into a mouse model of Duchenne muscular dystrophy. a Schematic representation of cell engraftment. b Immunostaining dystrophin (green) and PAX7 (red) on transverse sections 21 days after engraftment of satellite cells that had been newly isolated from muscle of Pax3 ^GFP/+ (wild type; left) and Pax3 ^GFP/+ ; Fmr1 ^−/− (Fmr1 ^−/−; right) mice. Arrowheads indicate location of satellite cells. Dotted lines indicate position of dystrophin-negative myofibers. Scale bars (white) indicate 50 μm. c Numbers of dystrophin-positive myofibers 21 days after engraftment of 10,000 satellite cells isolated from Pax3 ^GFP/+ (wt) and Pax3 ^GFP/+ ; Fmr1 ^−/− (Fmr1 ^−/−) mice. d Immunostaining GFP (green) and PAX7 (red) on transverse sections 21 days after engraftment of satellite cells from muscle of Pax3 ^GFP/+ (wild-type; left) and Pax3 ^GFP/+ ; Fmr1 ^−/− (Fmr1 ^−/−; right) mice. Arrowheads indicate location of GFP(+) satellite cells of donor origin. e Numbers of PAX7-positive and GFP-positive satellite cells of donor origin, per 100 dystrophin-positive myofibers 21 days after engraftment of 10,000 satellite cells isolated from Pax3 ^GFP/+ (wt) and Pax3 ^GFP/+ ; Fmr1 ^−/− (Fmr1 ^−/−) mice. Values indicate mean (n ≥ 3) ± s.e.m. **p < 0.01. ns not significant

Fig. 5

FMRP binds Myf5 transcripts and affects rates of deadenylation. a Illustration of Myf5 transcripts (Mus musculus) with FMRP binding WGGA (UGGA, AGGA) consensus sites indicated. Consensus sites which are within close proximity to each other and form potential FMRP binding G-quadraplex sites (gray) are indicated. One such G-quadraplex is in proximity to the microRNA-31 (miR-31) binding site on the 3’UTR of Myf5. b Immunoprecipitation (IP) of FMRP-mRNA complexes from C2C12 cells transfected with plasmids encoding GFP and FLAG-hFMRP. Immunoprecipitating antibodies are against FLAG-hFMRP (anti-FLAG, left) and GFP (anti-GFP, right). Amplification of Myf5 (upper panels) and Pax7 (lower panels) transcripts by semi-quantitative RT-PCR are shown, with thermocycle numbers indicated. c Poly(A) tail lengths of Myf5 (left) and Pax7 (right) transcripts amplified from total RNA isolated from 3-day cultures of satellite cells isolated from muscle of adult Pax3 ^GFP/+ (wild-type, wt) and Fmr1 ^−/− mice. Estimated polyA lengths are indicated adjacent to each acrylamide gel. Gene-specific primers (GSP) against Myf5 and Pax7, with product size indicated, are shown in the lower panels. A red dotted line indicates Myf5 polyA lengths < 200 nucleotides present in wild-type but not in Fmr1 ^−/− MuSCs