Islr KO impairs injury-induced muscle regeneration
The role of Islr in SCs in vivo was assessed using Islr-KO mice (Supplementary Fig. 1a). The results showed that the regenerative capacity of Bacl2-injuried tibialis anterior (TA) muscles was lower in Islr-KO mice than in wild-type (WT) counterparts (Supplementary Fig. 1b), evidenced by a significant reduction in the size of myofibers with centralized nuclei in Islr-KO mice on day 5 after injury (Supplementary Fig. 1c,d). The percentage of regenerating myofibers with two or more centralized nuclei was significantly lower in Islr-KO mice than in control mice (Supplementary Fig. 1e), consistent with a previous study24. Moreover, the number of SCs was marginally lower in Islr-KO mice than in control mice (Supplementary Fig. 1f), suggesting that Islr KO decreases the proliferation of SCs.
To elucidate the role of Islr in SC function, Islr was selectively ablated in Pax7-derived cells by crossing Pax7-CreER mice expressing a Cre-ER recombinase from the Pax7 locus, Pax7CreERT 26 with Islrf/f mice to generate Pax7CreERT/+: Islrf/f mice (Supplementary Fig. 2a). Following the administration of four consecutive doses of tamoxifen (Supplementary Fig. 2a), SCs were isolated by fluorescence-activated cell sorting (FACS) and cultured27, and the Islr protein was efficiently depleted (Supplementary Fig. 2b). Islr KO did not significantly affect animal survival, and cKO animals displayed no noticeable change in phenotype or TA muscle weight (Supplementary Fig. 2c,d). In addition, there was no significant change in the number of SCs on myofibers at 1 week after tamoxifen injection (Supplementary Fig. 2e). Consistent with this finding, the protein concentration of Pax7 was similar between Islr-cKO and control animals (Supplementary Fig. 2f), indicating that Islr loss did not strongly impact the maintenance of the SC pool in the TA muscle.
Islr-cKO limited the proliferation of SCs and shrank the SC pool
To study the role of Islr in SCs during skeletal muscle regeneration, TA muscle injury was induced using 1.2% BaCl228 (Supplementary Fig. 3a). Islr-cKO decreased the regenerative capacity of the TA muscle (Supplementary Fig. 3a). SCs are strongly implicated in skeletal muscle repair29. Therefore, we measured the number of Pax7-positive SCs on days 5 and 7 after injury. The number of SCs was substantially lower in Islr-cKO mice than in control mice on day 5 after injury (Fig. 1a). Similarly, the number of SCs was markedly lower on day 7 after injury (Supplementary Fig. 3b), indicating that Islr helps maintain the stem cell pool after muscle injury.
Next, FACS-isolated SCs were cultured for 5 days and labeled with 5-ethynyl-20-deoxyuridine (EdU) for 2 h. Islr-cKO decreased the number of EdU-positive SCs (Fig. 1b). To elucidate the role of Islr in cell proliferation, myoblasts were immunostained with the cell proliferation marker Ki67. The percentage of Ki67-positive cells was significantly lower in Islr-cKO myoblasts than in control myoblasts (Supplementary Fig. 3c). Moreover, the percentage of MyoD+ Ki67+ cells was significantly lower in the Islr-cKO group (Supplementary Fig. 3d).
The proliferation potential of SCs during muscle regeneration was assessed by Ki67 immunostaining. Proliferating and differentiating cells are usually present in regenerating skeletal muscle30. The number of Ki67-positive cells was significantly lower than in the TA muscles of Islr-cKO mice than in control mice on day 5 after injury (Fig. 1c). Proliferating cells were labeled with EdU for 2 h, and mice were sacrificed on day 3 after injury (Supplementary Fig. 3e). The number of EdU-positive cells was significantly lower in Islr-cKO mice than in control mice on day 3 after injury (Supplementary Fig. 3f). Freshly isolated extensor digitorum longus (EDL) myofiber explants were cultured for 42 h to allow SC activation and proliferation. The results showed that the number of Pax7+ Ki67+ cells was significantly lower in myofibers from Islr-cKO mice than in control mice (Fig. 1d). Moreover, the number of MyoD+Ki67+ cells per myofiber was significantly lower in the Islr-cKO group (Fig. 1e).
Islr KO leads to SC asymmetric division in vivo
To assess the effect of Islr-cKO on satellite cell renewal and differentiation, FACS-isolated SCs were cultured and stained for Pax7 and Ki67. Islr-cKO increase the percentage of self-renewing (Pax7+ Ki67−) cells and decreased the percentage of differentiated (Pax7− Ki67+) cells and the number of Ki67-positive cells (Fig. 2a,b)9. Freshly isolated EDL myofiber explants were cultured for 48 h to allow SCs activation and proliferation. Islr-cKO increased the percentage of Pax7+ Ki67− cells (Fig. 2c,d), suggesting that Islr reduces the ability of SCs to self-renew and return to a quiescent state. Additionally, Islr-cKO and Islr KO increased the number of dividing SCs asymmetrically expressing Pax7 or MyoD in one of paired cells (Fig. 2e,f; Supplementary Fig. 4a,b), indicating that SCs had distinct fates: self-renewal or differentiation31. These results suggest that Islr regulates satellite stem cell division.
Islr prevents the activation of the p-ERK1/2 pathway in SCs
To investigate the molecular mechanisms underlying the effect of Islr on SC division, we performed RNA sequencing (RNA-seq) to interrogate the transcriptomic changes caused by Islr-cKO (Supplementary Fig. 5a). Kyoto Encyclopedia of Genes and Genomes and Gene Ontology enrichment analyses demonstrated that Islr was strongly implicated in the MAPK signaling pathway and that Islr upregulated several genes from ERK1/2 signaling pathway (Fig. 3a,b).
The concentration of phosphorylation ERK1/2 proteins increased in proliferating SCs lacking Islr (Fig. 3c). To confirm these findings, SCs in EDL myofibers were treated with the p-ERK1/2 inhibitor PD98059 or PD18435232 (Supplementary Fig. 5b). PD98059 decreased the rate of asymmetric division (Fig. 3d), the number of Pax7+ MyoD− cells per myofiber at 48 h after treatment (Fig. 3e), and the number of Pax7+ Ki67− cells in WT mice (Fig. 3f). PD184352 treatment yielded similar results (Supplementary Fig. 5c,d).
We found that p-ERK1/2 was expressed asymmetrically in dividing SCs (Supplementary Fig. 6b), as was the case with phosphorylated p3833.
Islr binds to and stabilizes SPARC
To elucidate the role of Islr in the p-ERK1/2 pathway, p-ERK1/2 pathway proteins that interact with Islr were predicted using the STRING website34. Co-immunoprecipitation assays showed that exogenous Islr interacted with SPARC in 293T cells (Fig. 4a). In addition, Islr and SPARC were co-localized in the cytoplasm of 293T cells (Fig. 4b). Immunoblot analysis showed that Islr-cKO significantly increased SPAC protein concentrations (Fig. 4c) but not its mRNA levels (Fig. 4c), indicating that Islr decreased SPARC protein concentrations in activated SCs. Islr stabilizes protein through the proteasome pathway24. Therefore, 293T cells were transfected with Islr-GFP and the SPARC-Flag plasmid to determine whether Islr regulates proteasome-dependent degradation of SPARC. Cycloheximide chase assay and treatment with the proteasome inhibitor MG132 revealed that Islr regulated SPARC via the proteasomal degradation (Fig. 4d,e), demonstrating that Islr contributes to the degradation of SPARC. However, Islr stabilizes SPARC via a mechanism other than the ubiquitin-proteasome pathway (Fig. 4f). These results suggest that Islr directly binds to SPARC and causes its proteasome-mediated degradation in SCs.
Islr KO promotes p-ERK1/2 activity through SPARC to stimulate asymmetric division
Although the concentrations of SPARC and p-ERK1/2 proteins increased in Islr-deficient SCs (Supplementary Fig. 3c,4b), whether SPARC regulates p-ERK1/2 activity via an SC-independent mechanism is unknown. To address this question, we transfected proliferating WT SCs with a SPARC siRNA and found that this siRNA decreased the concentration of p-ERK1/2 proteins and the number of Pax7+ Ki67– cells (Fig. 5a-c). In addition, PD98059 treatment significantly reduced the number of Pax7+ Ki67–SCs and asymmetric division in EDL myofibers of Islr-cKO mice (Fig. 5d-f). These results indicate that Islr regulates SC division by activating SPARC/p-ERK1/2 signaling.
Islr KO aggravates the dystrophic phenotype in mdx mice
The impact of Islr KO on injury-induced muscle regeneration was assessed in an mdx mouse model, in which mutations in the dystrophin gene leads to long-term myofibers degeneration and regeneration accompanied by the activation and turnover of SCs, mimicking the phenotype of human Duchene muscular dystrophy (DMD)35. Islr-cKO mice were crossed with mdx mice to generate Islr cKO-mdx mice. Two-months-old mdx mice were treated with four consecutive doses of tamoxifen to delete Islr (Fig. 6a). After 7 months, relative TA muscle weight was significantly lower in Islr cKO-mdx mice than in mdx counterparts (Fig. 6b).
Histological analysis revealed that TA muscle fibers and the muscle cross-sectional area were significantly smaller in Islr cKO-mdx mice than in mdx controls (Fig. 6c). These findings suggested that Islr KO in SCs aggravates the dystrophic phenotypes in mdx mice. In addition, the number of Pax7-positive cells decreased remarkably in Islr cKO-mdx mice (Fig. 6d), suggesting that Islr KO impairs SC expansion during muscle regeneration. These results indicate that Islr is essential for replenishing SC pools and stimulating muscle regeneration.