Hypoxia promotes satellite cell self-renewal and enhances the efficiency of myoblast transplantation

Abstract

Microenvironmental oxygen (O(2)) regulates stem cell activity, and a hypoxic niche with low oxygen levels has been reported in multiple stem cell types. Satellite cells are muscle-resident stem cells that maintain the homeostasis and mediate the regeneration of skeletal muscles. We demonstrate here that hypoxic culture conditions favor the quiescence of satellite cell-derived primary myoblasts by upregulating Pax7, a key regulator of satellite cell self-renewal, and downregulating MyoD and myogenin. During myoblast division, hypoxia promotes asymmetric self-renewal divisions and inhibits asymmetric differentiation divisions without affecting the overall rate of proliferation. Mechanistic studies reveal that hypoxia activates the Notch signaling pathway, which subsequently represses the expression of miR-1 and miR-206 through canonical Hes/Hey proteins, leading to increased levels of Pax7. More importantly, hypoxia conditioning enhances the efficiency of myoblast transplantation and the self-renewal of implanted cells. Given the robust effects of hypoxia on maintaining the quiescence and promoting the self-renewal of cultured myoblasts, we predict that oxygen levels in the satellite cell niche play a central role in precisely balancing quiescence versus activation, and self-renewal versus differentiation, in muscle stem cells in vivo.

Fig. 1.

Hypoxia favors the quiescent state in cultured myoblasts. (A) Representative images of primary skeletal myoblasts cultured under 21% O₂ (normoxia) or 1% O₂ (hypoxia) for 48 hours and labeled with Pax7 (red) and MyoD (green). Nuclei were counterstained with Hoechst33342 (blue). (B) Percentage of quiescent (Pax7⁺MyoD^–), proliferating (Pax7⁺MyoD⁺) and differentiating (Pax7^–MyoD⁺) cells. n=3 independent experiments, with five different areas (at least 150 cells/area) analyzed in each experiment. Error bars represent s.e.m. ^**P<0.01. n.s., not significant.

Fig. 2.

Hypoxia promotes asymmetric self-renewal division in cultured myoblasts. (A) The strategy for synchronizing cultured myoblasts in late telophase stage. (B) Primary myoblasts were synchronized at the telophase stage and stained for Pax7 (red) and MyoD (green). Doublets of Pax7⁺MyoD⁺:Pax7⁺MyoD^– sister cells (8-10) represent asymmetric self-renewal division; Pax7⁺MyoD⁺:Pax7⁺MyoD⁺ (1-4, 7, 12) represent symmetric proliferative division; and Pax7⁺MyoD⁺:Pax7^–MyoD⁺ (5, 6, 11) represent asymmetric differentiation division. Scale bar: 25 μm. (C) Percentages of asymmetric self-renewal, symmetric proliferative and asymmetric differentiation divisions under exposure of 21% O₂ or 1% O₂. n=3 independent experiments, with at least 60 doublets analyzed in each experiment. ^**P<0.01.

Fig. 3.

Hypoxia upregulates Pax7 by downregulating miR-1/206. Primary myoblasts were cultured under normoxia (N; 21% O₂) or hypoxia (H; 1% O₂). (A) qPCR for relative expression of Pax7 mRNA normalized to 18s. (B) Western blot analysis for Pax7, MyoD and α-tubulin at the indicated time points. (C) Hypoxia inhibits miR-1 and miR-206 expression. Myoblasts were cultured under normoxia or hypoxia for 48 hours and qRT-PCR results were normalized to 18s and U6. (D) Efficient knockdown of miR-1/206 by LNAs. Myoblasts were transfected with 50 nM control LNA or LNAs against miR-1/206 and cultured for 48 hours. (E,F) Pax7 protein (E) and Pax7 mRNA (F) levels under hypoxia and normoxia, in the presence or absence of LNA against miR-1/206. Data are from five independent experiments. ^*P<0.05, ^**P<0.01. n.s., not significant.

Fig. 4.

Inhibition of miR-1/206 mimics the effect of hypoxia on the quiescence and self-renewal of myoblasts. Myoblasts were transfected with 50 nM control LNA or LNAs against miR-1/206, cultured under 21% O₂ (N, normoxia) or 1% O₂ (H, hypoxia) for 48 hours. (A) Relative abundance of quiescent (Pax7⁺MyoD^–), proliferating (Pax7⁺MyoD⁺) and differentiating (Pax7^–MyoD⁺) cells. n=3. (B) Sister cells were synchronized at telophase stage and their cell fates were analyzed based on Pax7 (red) and MyoD (green). The graph shows the frequency of asymmetric self-renewal, symmetric proliferative and asymmetric differentiation divisions. n=3. ^*P<0.05, ^**P<0.01. n.s., not significant.

Fig. 5.

Notch signaling represses miR-1/206. (A,B) Myoblasts were cultured under 21% O₂ and 1% O₂ for 48 hours and cells were collected for western blot analysis of N1ICD (A) and qPCR for Notch targets Hes1 and Hey1 (B) expression. (C) Myoblasts derived from Rosa26-N1ICD mice were infected with adenovirus-GFP or adenovirus-Cre. Cells were collected after 48 hours of infection for qPCR analysis of miR-1/206. (D-F) Notch1^fl/fl knockout mice, in which the first exon of Notch1 is floxed and can be deleted by Cre, leading to mutation of Notch1. Myoblasts derived from Notch1^fl/fl mice were infected with adenovirus-GFP and adenovirus-Cre. Cells were collected after 48 hours of infection for qPCR detection of miR-1/206 expression (D) and Notch1 and Pax7 expression (E), and western blot analysis for Notch1, Pax7, myogenin and GAPDH expression (F). (G-I) Wild-type myoblasts were treated with 10 μM DAPT, a pharmacological inhibitor of γ-secretase that cleaves Notch receptor and releases functional NICD. Cells were collected after 48 hours of DAPT treatment for qPCR detections of miR-1/206 (G) and Pax7 mRNAs (H), and western blot analysis for N1ICD, Pax7 and GAPDH proteins (I). (J,K) Wild-type myoblasts were treated with 10 μM DAPT 12 hours before exposure to 21% O₂ or 1% O₂ for an additional 48 hours, and analyzed for miR-1/206 expression by qPCR (J) and Pax7 expression by western blot (K). (L) MyoD^iCre/iCre (MyoD^–/–) myoblasts were electroporated with the pEF1a-N1ICD plasmid expressing N1ICD or with the empty vector pEF1a, cultured for 48 hours, and analyzed for miR-206 expression by qPCR. (M) MyoD^–/– myoblasts were treated with 10 μM DAPT for 48 hours, and analyzed for miR-1/206 expression by qPCR. (N) Wild-type myoblasts were electroporated with the plasmids expressing Hes1 and Hey1 or with the empty vector, cultured for 48 hours, and analyzed for miR-1/206 expression by qPCR. Expression levels of mRNAs were normalized to 18s and miRNAs were normalized to 18s and U6. n=3-6. ^*P<0.05, ^**P<0.01.

Fig. 6.

Enhanced transplantation efficiency of hypoxia conditioned myoblasts. mTmG myoblasts expressing membrane-targeted tandem dimer Tomato (mT, a red fluorescent protein), cultured under normoxia or hypoxia for 48 hours, were transplanted into the left and right TA muscles of MDX mice, respectively. (A-D) Representative images of mT⁺ (A,B) and dystrophin⁺ fibers (C,D) at 3 weeks post injection (wpi). (E) The relative number of mT⁺ fibers after 1-4 dpi, 1-2 wpi and 3-4 wpi. N represents the number of mouse analyzed, with five sections counted in each muscle. (F) The relative number of dystrophin⁺ myofibers (n=5) after 2-3 wpi. (G) Relative mTmG genomic DNA (n=4) from samples of 3-4 wpi. (H) M-Cadherin⁺ RFP⁺ satellite cell numbers per microscopic region. Scale bar: 50 μm. ^*P<0.05, ^**P<0.01.