Elevated mitochondrial Ca²⁺ impairs satellite cell pool expansion in response to skeletal muscle injury

Summary

MICU1 loss-of-function variants in human patients are associated with proximal muscle weakness and myopathy. Mitochondrial Ca²⁺ levels are basally elevated when MICU1, the gatekeeper of the mitochondrial Ca²⁺ uniporter, is absent. The importance of regulating mitochondrial Ca²⁺ in skeletal muscle has generally been studied in mature muscle fibers. How satellite cells are impacted by mitochondrial Ca²⁺ dysregulation is poorly understood. We investigated Micu1 deletion specifically in Pax7+ satellite cells to address this gap in knowledge. Colony-forming activity in vitro was unaffected in Micu1-deficient satellite cells, but colony sizes were smaller. Although satellite cell homeostasis was not significantly affected 1 month following Micu1 deletion, the regenerative response post-injury was significantly impaired. Satellite cell self-renewal from Micu1-deficient donor cells in transplant recipients was also heavily compromised. Our data suggest that properly gating mitochondrial Ca²⁺ import via the uniporter is integral to satellite cell activation from quiescence in response to muscle injury.

Graphical abstract

Highlights

• •Loss of MICU1 in muscle satellite cells elevates mitochondrial Ca²⁺ and impairs respiration
• •MICU1 deletion is associated with delayed first cell division of satellite cells in vitro
• •MICU1 is required for satellite cell expansion after injury, not for basal maintenance
• •MICU1 is necessary for self-renewal of satellite cells upon transplantation

Liu and colleagues show that mitochondrial Ca²⁺ levels in muscle satellite cells rise upon loss of MICU1, the gatekeeping subunit of the mitochondrial Ca²⁺ uniporter. Although basal maintenance of satellite cell numbers is not significantly affected, MICU1 deficiency delays satellite cell division in culture. Correspondingly, MICU1 is required for satellite cell expansion after injury and self-renewal upon transplantation in mice.

Introduction

Calcium—dubbed the “life and death signal” (Berridge et al., 1998)—plays important roles in cell and mitochondrial function, particularly in excitable tissue like muscle. The dynamics of intracellular Ca²⁺ are critical for muscle contraction, and mitochondrial buffering of cytosolic Ca²⁺ also plays a role in bioenergetics. Mitochondrial uptake of Ca²⁺ has been shown to activate multiple matrix dehydrogenases as well as ATP synthase (Glancy et al., 2013), and thus is thought to be important for matching energy output to increased demand. However, excess Ca²⁺ leads to opening of the mitochondrial permeability transition pore (mPTP), mitochondrial swelling, and cell death (Lemasters et al., 2009). Hence, the regulation of Ca²⁺ entry into mitochondria is critically important. Indeed, Ca²⁺ imbalances are seen in a number of muscle disorders, including muscular dystrophies (Vallejo-Illarramendi et al., 2014).

Rapid entry of Ca²⁺ into mitochondria occurs through a multi-protein complex in the inner membrane known as the mitochondrial Ca²⁺ uniporter (mtCU) (DeLuca and Engstrom, 1961; Vasington and Murphy, 1962) comprising the channel-forming protein MCU and multiple regulatory subunits. One mtCU subunit, MICU1, a Ca²⁺-binding, EF-hand protein, inhibits mtCU activity at basal (low) cytosolic Ca²⁺ concentrations (Csordás et al., 2013; Mallilankaraman et al., 2012; Perocchi et al., 2010). Loss of Micu1 and this inhibition permits excess mtCU activity, leading to elevated matrix Ca²⁺ levels. Indeed, increased mitochondrial Ca²⁺ was observed in human patients with loss-of-function mutations in MICU1 (Logan et al., 2014). These individuals presented in childhood with proximal muscle weakness and moderately to grossly elevated serum creatine kinase levels. Other patients with MICU1 mutations were noted to suffer from fatigue and muscle aches after light exercise (Lewis-Smith et al., 2016), and similar clinical symptoms manifested in yet a third group of patients (Musa et al., 2018). Likewise, mitochondrial Ca²⁺ overload and severe neuromuscular impairment were also present in mice with germline Micu1 deletion (Liu et al., 2016). However, mature muscle fiber-specific knockout of Micu1 using MCK-Cre produced a milder phenotype (Debattisti et al., 2019), in which force developed upon tetanic stimulation was reduced, but unlike human patients the mice showed no difference in time to exhaustion on a treadmill. Taken together, the evidence suggests that mitochondrial Ca²⁺ elevation in cell types additional to the fiber contributes to the full myopathic phenotype of MICU1 loss-of-function in humans and mice.

Relatively little is known about the role of mitochondria, or mitochondrial Ca²⁺ in particular, in the resident stem cell population of skeletal muscle, the satellite cells. The contributions of mitochondria to cell fate decisions including activation and self-renewal (Yin et al., 2013), are becoming recognized (Bhattacharya and Scimè, 2020; Hong et al., 2022). Mitochondria are involved in the metabolic reprogramming that occurs during differentiation, with evidence suggesting that quiescent satellite cells are more reliant on fatty acid oxidation, shifting to glycolysis during proliferation (Ryall et al., 2015). However, other studies show that glycolysis upregulation is accompanied by increased fatty acid metabolism and oxidative phosphorylation signatures in activated satellite cells post-injury (Pala et al., 2018; Dell'Orso et al., 2019). Recently, mitochondrial fatty acid oxidation was found to be required to meet metabolic demands for satellite cell proliferation and muscle regeneration (Yue et al., 2025). Furthermore, links between other mitochondrial Ca²⁺-related processes, such as reactive oxygen species signaling and regulation of pyruvate dehydrogenase (PDH) (Hori et al., 2019), have been explored in satellite cell differentiation. Given that Ca²⁺ is an important regulator of mitochondrial bioenergetics and metabolism, it likely plays an as-yet undetermined role in satellite cell fate.

Here, we assess the impact of elevated mitochondrial Ca²⁺ in muscle satellite cells by deleting Micu1 using the Pax7-Cre^ERT2 mouse strain (Murphy et al., 2011), which allows temporally regulated (tamoxifen-induced) deletion in Pax7+ cells. Collectively, our in vitro and in vivo data suggest that MICU1 is dispensable for satellite cell homeostasis in the short term but required for mounting a regenerative response following injury.

Results

MICU1 deletion is associated with delayed first cell division of satellite cells in vitro

We combined the Pax7-Cre^ERT2 allele with homozygous floxed alleles of Micu1 (Antony et al., 2016) and a fluorescent reporter for satellite cells, Pax7-ZsGreen (Bosnakovski et al., 2008). When mice were treated with tamoxifen, the mRNA expression level of Micu1 declined significantly in Micu1^MuSC-KO satellite cells compared to Cre-negative Micu1^FL/FL controls, indicating efficient in vivo knockout (Figure S1A). In satellite cells cultured in vitro and passaged once, protein levels of MICU1 also declined (Figures 1B and S1B). While relatively little change occurred in levels of MCU, the pore-forming subunit of the uniporter (Figures 1B and S1C), protein levels of essential MCU regulator (EMRE), another uniporter subunit, trended downward (Figures 1B and S1D), which is a known compensatory response seen in multiple other tissues upon MICU1 depletion (Hasan et al., 2024; Liu et al., 2016; Singh et al., 2022). We next measured mitochondrial Ca²⁺ levels in satellite cells 2 days after sorting, using the Ca²⁺-sensitive, mitochondria-localized fluorescent dye Rhod-2 AM, quantified after thresholding on mitochondria based on MitoView Green signal (Figures 1C and S1E). Based on this microscopy-based assessment, Micu1^MuSC-KO cells exhibited higher mitochondrial Ca²⁺ levels than controls, corroborating that MICU1 deficiency elevates mitochondrial Ca²⁺ (Figure 1D). To determine how excess mitochondrial Ca²⁺ affects mitochondrial function, we performed Seahorse assays on isolated satellite cells 2 days after sorting, and found relatively lower oxygen consumption rates (OCRs) in Micu1^MuSC-KO cells (Figure 1E), demonstrating significantly repressed basal and maximal respiration (Figures S1F and S1G). These data confirm previous studies showing compromised respiratory function in mitochondria with MICU1 deficiency (Tsai et al., 2022; Pakkiriswami et al., 2025), which we hypothesized might impair satellite cell function.

Figure 1.

MICU1 deficiency leads to elevated mitochondrial Ca²⁺, impaired respiration, and delayed first cell division of satellite cells in vitro

See also Figure S1.

(A) Schematic for in vitro experiments showing the timing of tamoxifen (TMX) treatment, then FACS sorting of the ZsGreen+ cells for the CFC assay and time-lapse microscopy.

(B) Representative western blot of cultured satellite cells from Micu1^FL/FL and Micu1^MuSC-KO mice for MICU1, MCU, and EMRE protein levels. VDAC represents a loading control.

(C) Representative microscopy images of Rhod-2 AM and MitoView Green fluorescent staining in Micu1^FL/FL and Micu1^MuSC-KO satellite cells 2 days after isolation and sorting. Note that bright red Rhod-2 AM signal in nucleoli will be removed during image processing to ensure only mitochondrial signal (see Figure S1E). Scale bars represent 5 μm.

(D) Quantification of mitochondrial Rhod-2 AM in Micu1^FL/FL (n = 56) and Micu1^MuSC-KO (n = 46) satellite cells, derived from four mice of each genotype, 2 days after isolation and sorting.

(E) Seahorse XFp analysis of mean OCR in Micu1^FL/FL (n = 8 mice) and Micu1^MuSC-KO (n = 6 mice) satellite cells 2 days after isolation and sorting. OCR was measured under basal conditions and following the addition of oligomycin, carbonyl cyanide 4-(trifluoromethoxy)phenylhydrazone (FCCP), and rotenone and antimycin A.

(F) Cloning efficiency quantified by counting the number of wells containing colonies divided by the number of wells into which a single ZsGreen+ cell was sorted from Micu1^FL/FL (n = 5) and Micu1^MuSC-KO (n = 6) mice.

(G) Quantification of colony sizes derived from ZsGreen+ cells from Micu1^FL/FL (n = 5) and Micu1^MuSC-KO (n = 6) mice as measured by the number of nuclei in each colony.

(H) Quantification of the mean for each mouse (n = 5 Micu1^FL/FL mice and n = 6 Micu1^MuSC-KO mice) of colony sizes derived from its ZsGreen+ cells.

(I) Frequency of ZsGreen+ cells from Micu1^FL/FL (n = 4) and Micu1^MuSC-KO (n = 4) mice that underwent a first cell division as a function of time, detected by time-lapse microscopy. Data are presented as means and SEM; p values were calculated by t test.

We then compared the in vitro proliferation and differentiation ability of Micu1-deficient vs. control satellite cells in a 96-well plate in the colony forming cell (CFC) assay (Figure 1A). The ability of the cells to survive, form colonies and spontaneously differentiate in vitro was minimally affected by Micu1 deletion (Figures 1F, S1H, and S1I). However, mean colony size was ∼20% smaller in the Micu1^MuSC-KO group compared to the control group (Figures 1G and 1H). To further characterize the in vitro proliferation of Micu1-deficient satellite cells, we evaluated activation from quiescence by measuring the time taken for freshly isolated satellite cells to undertake their first cell division. Pax7-ZsGreen+ cells from Micu1^MuSC-KO and Micu1^FL/FL mice were isolated, plated, and tracked by time-lapse microscopy from 18 to 60 h after plating (Figure 1A). Image analysis revealed that Micu1-deficient satellite cells require longer times to complete the first division relative to control satellite cells (Figures 1I and S1J), suggesting that without MICU1, cells exist in a state of deeper quiescence.

MICU1 is dispensable for basal satellite cell maintenance within one month

To understand whether MICU1 plays a role in the homeostatic maintenance of the satellite cell pool, we induced deletion of Micu1 in vivo under static conditions. Micu1^MuSC-KO and Micu1^FL/FL Pax7-ZsGreen+ mice were injected with tamoxifen IP for 5 consecutive days, then maintained on tamoxifen diet for 1 month to suppress any escapers (von Maltzahn et al., 2013) (Figure 2A). There was no difference in hindlimb muscle mass between controls and Micu1-deficient mice (Figures 2C and S2A–S2C). When satellite cells were quantified by FACS using the Pax7-ZsGreen reporter, we observed a modest reduction, but it was not statistically significant (Figures 2B and 2D). This non-statistically significant trend of modest decline was replicated in mice lacking the ZsGreen reporter using an alternative approach, sorting on Itga7 and VCAM1, to identify satellite cells (Figures S2D–S2G).

Figure 2.

MICU1 is dispensable for basal satellite cell maintenance within 1 month

See also Figure S2.

(A) Schematic for homeostatic satellite cell experiments showing the number of tamoxifen (TMX) injections, the duration of tamoxifen diet, and time of muscle harvest for FACS analysis.

(B) Representative FACS plots showing the total number of ZsGreen+ cells from Micu1^FL/FL and Micu1^MuSC-KO mice.

(C) Quantification of TA mass from Micu1^FL/FL (n = 8) and Micu1^MuSC-KO (n = 10) mice.

(D) Quantification of total number of ZsGreen+ cells in the TAs from Micu1^FL/FL (n = 4) and Micu1^MuSC-KO (n = 5) mice. Data are presented as means and SEM; p value was calculated by t test.

MICU1 is required for satellite cell expansion after injury

We then evaluated the necessity of MICU1 in the context of injury. We again treated Micu1^MuSC-KO and Micu1^FL/FL Pax7-ZsGreen+ mice with tamoxifen IP for 5 consecutive days, but at day 4 of the treatment regimen we unilaterally injured one tibialis anterior (TA) with cardiotoxin (CTX). Mice were then maintained on tamoxifen diet to suppress escapers and allowed to recover for 2 weeks (Figure 3A). TA mass after CTX injury tended to be increased relative to the contralateral uninjured TA (Figure 3B), an effect that was more evident in the control than the Micu1^MuSC-KO muscle, although neither increase reached statistical significance. We quantified the number of Pax7-ZsGreen+ cells by FACS in the injured animals and found a significant failure of the satellite cells from Micu1^MuSC-KO muscle to expand in response to injury, unlike the control which responded to CTX injury by nearly doubling the number of ZsGreen+ cells in the injured limb compared to the uninjured (Figure 3C, p = 0.0014). We independently replicated this experiment on mice lacking the ZsGreen reporter, using Itga7/VCAM1 to identify satellite cells (Chan et al., 2013). Again, TA mass post-injury was not different between Micu1^MuSC-KO and Micu1^FL/FL mice (Figures S3A and S3B), but Itga7+/VCAM1+ satellite cell number was significantly lower in Micu1^MuSC-KO and Micu1^FL/FL TAs post-injury (Figures S3C and S3D). In parallel experiments, we also stained frozen sections with anti-Pax7 antibody to corroborate using a different method the lower numbers of Pax7+ cells in injured Micu1^MuSC-KO TAs (Figures 3D and 3E). We also stained frozen sections with laminin to visualize basement membranes, which revealed a higher frequency of small myofibers and fewer large fibers in Micu1^MuSC-KO TAs compared with control (Figures 3F and 3G), another indication of impaired regeneration. Collectively, these data suggest that MICU1 is necessary in satellite cells for an optimal regenerative response.

Figure 3.

MICU1 is required for satellite cell expansion after injury, see also Figure S3

(A) Schematic for the injury experiment showing the time of cardiotoxin injury to TA muscles followed by maintenance on tamoxifen diet. Unilateral injury was performed on ZsGreen+ mice, and 14 days post-injury ZsGreen+ cells from injured versus contralateral TAs were analyzed by FACS. In a parallel experiment in ZsGreen-mice, both TAs underwent injury, then 14 days post-injury muscles were frozen in OCT and cryosectioned for immunofluorescence staining.

(B) Quantification of muscle mass of injured and contralateral TAs from Micu1^FL/FL (n = 9) and Micu1^MuSC-KO (n = 10) mice.

(C) Quantification of the total number of ZsGreen+ cells in injured and contralateral uninjured TA from Micu1^FL/FL (n = 7) and Micu1^MuSC-KO (n = 8) mice.

(D) Quantification based on immunofluorescence staining of the number of Pax7+ cells in TA cross sections from Micu1^FL/FL (n = 5) and Micu1^MuSC-KO (n = 5) mice 14 days post-injury, per cross section (left) and per mm² (right).

(E) Representative immunofluorescence images of injured TA muscles from Micu1^FL/FL and Micu1^MuSC-KO mice stained with PAX7 antibody. Scale bars represent 100 μm.

(F) Representative immunofluorescence images of injured TA muscles from Micu1^FL/FL and Micu1^MuSC-KO mice stained with laminin antibody. Scale bars represent 100 μm.

(G) Myofiber cross-sectional area distribution in injured TA muscles of Micu1^FL/FL (n = 4) and Micu1^MuSC-KO (n = 4) mice. Data are presented as means and SEM, p value was calculated by t test and two-way ANOVA, ^∗p < 0.05, ^∗∗p < 0.01.

MICU1 is necessary for self-renewal of satellite cells upon transplantation

Transplantation of low numbers of satellite cells is an extremely sensitive assay for the self-renewal and differentiation potential of cells in question. We used our previously developed transplantation-based assay in which 300 Pax7-ZsGreen+ cells are transplanted into CTX-injured, irradiated TA muscles of syngeneic C57BL/6 recipients. One month after transplant, one TA muscle was digested and run entirely through FACS to quantify contribution to the satellite cell pool. We adapted this assay to test the role of MICU1 in satellite cell pool regeneration by using 300 donor-derived Pax7-ZsGreen+ cells from either Micu1^FL/FL or Micu1^MuSC-KO mice. Donor mice were treated with five doses of IP tamoxifen and starting on the day of transplantation, tamoxifen was provided in the diet of the recipients to suppress escapers (Figure 4A).

Figure 4.

MICU1 is necessary for self-renewal of satellite cells upon transplantation

(A) Schematic for the transplantation experiment showing that donor mice were treated for 5 consecutive days with IP tamoxifen, then ZsGreen+ cells were harvested and transplanted into the TA muscle of irradiated and injured recipient mice, then the recipients were maintained on tamoxifen diet until TA muscles were collected and analyzed by FACS.

(B) Representative FACS plots showing the total number of ZsGreen+ cells in the TA muscle of the recipient mice after transplantation of Micu1^FL/FL or Micu1^MuSC-KO satellite cells.

(C) Quantification of the total number of ZsGreen+ cells in the TA of the recipient mice after transplant of Micu1^FL/FL cells (n = 8 recipient mice) and Micu1^MuSC-KO (n = 8 recipient mice). Data are presented as means and SEM, p value was calculated by t test.

FACS analysis revealed that Micu1 deletion led to a significant inhibition of satellite cell self-renewal, evident by a 50% drop in the number of Pax7-ZsGreen+ cells in the recipient TA muscles of the Micu1^MuSC-KO group (Figures 4B and 4C). These data reveal an impaired self-renewal phenotype of the satellite cells in the Micu1^MuSC-KO group that was observed after CTX injury.

Discussion

Mitochondrial Ca²⁺ perturbations lead to profound impacts in skeletal muscle (Huo and Molkentin, 2024) as has been shown in multiple studies via deletion of the gene encoding the pore-forming subunit of the uniporter, Mcu (Gherardi et al., 2019; Kwong et al., 2018; Mammucari et al., 2015). Inhibiting mitochondrial Ca²⁺ uptake by skeletal muscle-specific Mcu deletion or silencing led to metabolic switching toward fatty acid oxidation (Gherardi et al., 2019; Kwong et al., 2018) and impinged on pathways regulating muscle size (Mammucari et al., 2015). Correspondingly, overexpression of the MCU inhibitor gene Mcub, in skeletal muscle also increased fatty acid usage, and deletion of Mcub had the opposite effect (Huo and Molkentin, 2024). These data underscore the importance of well-regulated mitochondrial Ca²⁺ handling in skeletal muscle. Strikingly, mitochondrial Ca²⁺ overload caused by loss-of-function MICU1 mutations results in persistent weakness and myopathy in human patients (Logan et al., 2014; Musa et al., 2018), which is echoed in mice with global Micu1 deletion (Liu et al., 2016). As the severity of this phenotype was not fully recapitulated in mice with muscle fiber-specific Micu1 deletion driven by MCK-Cre (Debattisti et al., 2019), we sought to distinguish the contribution of mitochondrial Ca²⁺ disruption due to Micu1 deletion specifically in satellite cells.

Limitations of our approach include the inherent scarcity and small size of these cells, which precluded standard functional assays performed using larger quantities of isolated mitochondria, such as the Ca²⁺ retention assay. Indeed, satellite cells have few mitochondria, correlating with low metabolic activity (Montarras et al., 2013). However, using a live cell imaging approach, we were able to show that MICU1 loss and subsequent “leakiness” of the mtCU led to elevated mitochondrial Ca²⁺ in isolated satellite cells, as has been observed across all studied cell and tissue types to date (Antony et al., 2016; Singh et al., 2022; Hasan et al., 2024; Liu et al., 2016; Pakkiriswami et al., 2025). Furthermore, as has been observed in other tissues (Hasan et al., 2024; Liu et al., 2016; Singh et al., 2022), elevated mitochondrial Ca²⁺ induced by MICU1 deficiency led toward compensatory decreased levels of the essential mtCU subunit EMRE, as seen in Figure 1B. This compensation is thought to re-normalize basal mitochondrial Ca²⁺ levels; however, mtCU-mediated rapid mitochondrial Ca²⁺ uptake in response to stimulation, which requires EMRE, may become blunted. Because mitochondrial Ca²⁺ import through the mtCU signals a surge in demand for ATP to induce a corresponding rapid increase in ATP production, losing this responsiveness could potentially disrupt cell signaling. Mitochondrial Ca²⁺ signaling may be linked to known roles for cytosolic Ca²⁺ in muscle regeneration (Tu et al., 2016). For instance, abolishing Ca²⁺ transients reduced the number of activated muscle satellite cells in regenerating tadpoles (Tu and Borodinsky, 2014) and inhibiting stretch-mediated Ca²⁺ influx prevented rat muscle satellite cell activation (Hara et al., 2012). Interference with cytosolic-mitochondrial Ca²⁺ signaling via basal elevation of mitochondrial Ca²⁺ or impaired mitochondrial Ca²⁺ uptake could form a barrier to metabolic activation required for satellite cells to proliferate upon muscle injury.

We have shown here that in the absence of MICU1, satellite cells reside in a deeper state of quiescence, taking significantly longer to divide upon activation. When muscle is subjected to severe injury, we find a corresponding delay in new fiber growth, greater frequency of very small fibers and fewer larger fibers, as well as a defect in the expansion of the satellite cell population. In the steady state, satellite cell self-renewal is very modest, and correspondingly, in unperturbed mice, we noted only a modest trend that was not statistically significant toward decline in satellite cell number one month post-deletion of Micu1. It is possible that longer-term follow up would reveal a continued decline. However, transplantation of low numbers of satellite cells requires significant self-renewal, which profoundly stretches their regenerative potential. The transplantation of 300 satellite cells revealed that MICU1-deficient satellite cells suffer an impaired ability to undergo self-renewal to properly repopulate the satellite cell compartment. These data support the idea that well-regulated mitochondrial Ca²⁺ is required for activation of satellite cells from quiescence, and that elevated mitochondrial Ca²⁺ impairs satellite cell proliferation and reduces self-renewal potential. This leads to impaired regeneration and a depleted satellite cell pool after major injury events.

Conclusions

In summary, these studies demonstrate that satellite cells show tremendous dependence on properly regulated mitochondrial Ca²⁺ during self-renewal and regeneration when muscle is perturbed by injury.

Resource availability

Lead contact

Further information and requests for resources and reagents should be directed to and will be provided by the lead contact, Julia C. Liu, (julialiu@umn.edu).

Materials availability

Materials are available upon reasonable request to the lead contact.

Data and code availability

Flow cytometry, western blot, qPCR, and inverted microscopy raw data files reported in this study will be shared by the lead contact upon request. Any additional information required to reanalyze the data reported in this study is available by the lead contact upon request.

Acknowledgments

We are deeply grateful to Dr. György Hajnóczky at Thomas Jefferson University for generously sharing the Micu1^FL/FL mice. We would like to thank all members of the Liu and Kyba labs for helpful discussions, and in particular Dr. Ulas Ozkurede and members of Dr. Emilyn Alejandro’s lab, Dr. Seokwon Jo and Dr. V Pszczolkowski, for help with preparing Seahorse reagents and setting up the Seahorse XFp Analyzer. We would also like to thank Cynthia Faraday for assistance with the graphical abstract. This research was funded by grants R01GM149666 to E.B., R01AR055685 and R01AG062899 to M.K., and K22HL137901 and R01HL164491 to J.C.L. from the National Institutes of Health.

Author contributions

Investigation, A.S.S., L.S.D., M.K.S., L.L., J.H.S., S.P., S.J.C., S.I., A.M.B., M.K., and J.C.L.; analysis, I.R.A., A.S.S., E.B., M.K., and J.C.L.; writing – original draft preparation, A.S.S., M.K., and J.C.L.; review and editing, E.B., M.K., and J.C.L. All authors have read and agreed to the final version of the manuscript.

Declaration of interests

The authors declare no competing interests.

STAR★Methods

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
PE-Cy7 rat anti-mouse CD31 clone 390	BD Biosciences	561410; RRID: AB_1061200
PE-Cy7 rat anti-mouse CD45 clone 30-F11	BD Biosciences	552848; RRID:A B_394489
Biotin rat anti-mouse CD106 clone 429	BD Biosciences	553331; RRID: AB_394787
PE Streptavidin	BD Biosciences	554061; RRID: AB_10053328
α-7 integrin 647 clone R2F2	AbLab	67-0010-05; RRID: AB_2890939
MICU1	Sigma Aldrich	HPA037480; RRID: AB_10696934
MCU	Cell Signaling	14997S; RRID: AB_2721812
EMRE	New England Peptide	custom antibody
VDAC1	Santa Cruz Biotechnology	sc-390996; RRID: AB_2750920
IRDye 800CW anti-rabbit secondary	LI-COR	926–32211; RRID: AB_621843
IRDye 680LT anti-mouse secondary	LI-COR	926–68020; RRID: AB_10706161
HRP-conjugated anti-rabbit secondary	Cell Signaling Technology	7074; RRID: AB_2099233
anti-Pax7 mouse IgG1	Developmental Studies Hybridoma Bank	PAX7; RRID: AB_2299243
MYH1E myosin heavy chain	Developmental Studies Hybridoma Bank	MF20; RRID: AB_2147781
goat anti-mouse biotin-conjugated secondary antibody	Jackson Immuno Research Laboratories Inc	115-065-205; RRID: AB_2338571
monoclonal antibody to laminin clone LAM-89	Sigma Aldrich	L8271; RRID: AB_477162
Alexa Fluor 488 goat anti-mouse IgG antibody	Life Technologies	A11029; RRID: AB_2534088
Chemicals, peptides, and recombinant proteins
tamoxifen diet	Envigo	TD.140251
tamoxifen	Sigma	T5648-1G
cardiotoxin	Sigma	217503
collagenase type II	Gibco	17101–015
dispase	Gibco	17105–041
RIPA buffer	Thermo Scientific	89900
Halt Protease and Phosphatase Inhibitor Cocktail	Thermo Scientific	78440
Bolt LDS sample buffer	Invitrogen	B0007
Bolt Bis-Tris 4–12% gels	Invitrogen	NW04120BOX
TBS blocking buffer	LI-COR	927–60001
Clarity Western ECL Substrate	BioRad	1705060
oligomycin	Sigma	O4876
carbonyl cyanide 4-(trifluoromethoxy)phenylhydrazone	Sigma	C2920
rotenone	Sigma	R8875
antimycin-a	Sigma	A8674
Rhod-2 AM	Invitrogen	R1245MP
MitoView Green	Biotium	70054
QIAzol lysis reagent	Qiagen	79306
O.C.T. compound cryostat embedding medium	Scigen	4586
2-methylbutane	Thermo Scientific	019387
Triton X-100	Sigma	X100
ProLong Gold Antifade Mountant with DAPI	Invitrogen	P36931
Immu-Mount	Epredia	9990402
Gelatin	Sigma	G2500
Fetal Bovine Serum	Gibco	16000044
DMEM/F12 medium without L-glutamine	Cell Gro	15-090-CV
horse serum	Gibco	26050–088
human basic fibroblast growth factor	Peprotech	100–18
Glutamax	Gibco	35050–061
chick embryonic extract	US Biological	C3999
penicillin/streptomycin	Gibco	15140–122
Ham’s F10	HyClone	MT10070CV
4-hydroxytamoxifen	Sigma	H7904
Critical commercial assays
TSA cyanine 3 kit	Akoya Biosciences	NEL744001KT
Vectastain ABC reagent	Vector Laboratories	PK-6100
Akoya TNB blocking reagent	Akoya Biosciences	FP1020
Luna Universal One-Step RT-qPCR kit	New England Biolabs	E3005S
Experimental models: Organisms/strains
Mouse: C57BL/6J	The Jackson Laboratory	JAX: 000664
Mouse: B6.Cg-Pax7tm1(cre/ERT2)Gaka/J	The Jackson Laboratory	JAX: 017763
Mouse: Micu1FL/FL mice	Antony et al. (2016)	N/A
Mouse: Pax7-ZsGreen mice	Bosnakovski et al. (2008)	N/A
Oligonucleotides
Micu1 primers, Forward 5′-CAA-CGA-ACC-TGG-TGA-AAC-CG-3′ Reverse 5′-GTC-TGC-CAG-AAC-TGG-TTC-C-3′	This paper	N/A
eukaryotic translation initiation factor EIF35S primers, Forward 5′-CTG-AGG-ATG-TGC-TGT-CTG-GGA-A-3′ Reverse 5′-CCT-TTG-CCT-CCA-CTT-CGG-TC-3′	This paper	N/A
ribosomal protein 36B4 primers, Forward 5′-GGC-CCT-GCA-CTC-TCG-CTT-TC-3′ Reverse 5′-TGC-CAG-GAC-GCG-CTT-GT-3′	This paper	N/A
Software and algorithms
ImageJ	Schneider et al., 2012	https://imagej.nih.gov/ij/
GraphPad Prism 8.0	GraphPad Software Inc	N/A

Experimental model details

Mice

All procedures were performed in accordance with a protocol approved by the Institutional Animal Care and Use Committees at the University of Minnesota. All experiments were conducted on adult male and female mice 2–4 months of age. Micu1^FL/FL mice (C57BL/6J) were generously shared by Dr. György Hajnóczky at Thomas Jefferson University and were crossed in house with Pax7-Cre^ERT2 and Pax7-ZsGreen mice to obtain Micu1^FL/FL; Pax7-Cre^ERT2; Pax7-ZsGreen (experimental, abbreviated Micu1^MuSC-KO), and Micu1^FL/FL; Pax7-ZsGreen (control, lacking Cre) mice were generated in-house. The Pax7-Cre^ERT2 mouse line is the Pax7-Cre^ERT2(GAKA) from Gabrielle Kardon’s lab, which preserves the endogenous Pax7 locus, and heterozygosity of this transgene was maintained throughout all breeding. Other experiments used the same strains lacking Pax7-ZsGreen expression. Transplant recipients were C57BL/6J mice purchased from Jackson Laboratory (strain # 000664) (Bosnakovski et al., 2008).

To induce Micu1 deletion, all mice were treated for five consecutive days with 80 mg/kg intraperitoneal tamoxifen (Sigma Aldrich, T5648-1G) dissolved at a concentration 12.75 mg in 1 mL corn oil, after which mice were then maintained on tamoxifen diet (Envigo, TD.140251) for the remaining duration of each experiment. For the transplantation experiments, donor mice were treated with the IP tamoxifen then the recipient mice were maintained on tamoxifen food until the time of euthanasia. This knockout protocol overcomes the ability of any escapers to reconstitute the satellite cell pool (von Maltzahn et al., 2013; Shams et al., 2022).

Method details

Flow cytometry/fluorescence-activated cell sorting (FACS)

Flow cytometric isolations of satellite cells from bulk hindlimb or individual muscle digests were performed as described in detail previously (Arpke et al., 2021). Briefly, muscles were dissected, minced in parallel with muscle fibers, and digested with 0.2% collagenase type II in DMEM and dispase (17101-015 and 17105-041, resp.; Gibco, Grand Island, NY). To obtain satellite cells from Pax7-ZsGreen mice, mononuclear cells were isolated as described above and were incubated in FACS staining medium containing PI (Arpke et al., 2013). Absolute satellite cell counts by FACS were confirmed through gating ZsGreen+ cells. Investigators were blinded to groups during FACS analysis.

For isolation of satellite cells from ZsGreen– mice, mononuclear cells were stained using an antibody mixture of 1 μL PE-Cy7 rat anti-mouse CD31 (clone 390; 561410; BD Biosciences, San Diego, CA), 1 μL PE-Cy7 rat anti-mouse CD45 (clone 30-F11; 552848; BD Biosciences), 1 μL Biotin rat anti-mouse CD106 (clone 429 (MVCAM.A); 553331; BD Biosciences), 1 μL PE Streptavidin (554061; BD Biosciences), and 2 μL α-7 integrin 647 (clone R2F2; 67-0010-05 AbLab; Vancouver, B.C., Canada). Samples were incubated with this antibody cocktail, washed, and resuspended with FACS staining medium (2% Fetal Bovine Serum [FBS; 16000044; Gibco] in phosphate-buffered saline [PBS]) containing 0.5 μg/mL propidium iodide (PI) for analysis on a FACSAriaII (BD Biosciences, San Diego, CA). Total satellite cells (Lin–, [i.e., PE-Cy7–]; double-positive, i.e., VCAM1+, α7 integrin+ cells) were analyzed from the entire TA muscle sample.

Colony forming cell assay

Satellite cells, identified by FACS as ZsGreen+ cells, were single cell-sorted into 96-well plates in 200 μL mouse myogenic medium (MMM): DMEM/F12 medium without L-glutamine (Cell Gro, Manassas, VA; 15-090-CV) containing 20% FBS (HyClone), 10% horse serum (Gibco, 26050-088), 50 ng/μL human basic fibroblast growth factor (Peprotech, Rocky Hill, NJ; 100-18), 1% penicillin/streptomycin (Gibco, 15140-122), 1% Glutamax (Gibco, 35050-061), and 0.5% chick embryonic extract (US Biological, Swampscott, MA; C3999), containing 5 mM 4-hydroxy tamoxifen (4OHT) (Sigma Aldrich, H7904) at the time of plating. Plates were maintained at 37°C at 5% CO₂, 5% O₂, 90% N₂) for 8 days. Colonies were identified and fixed with 4% paraformaldehyde for 20 min at room temperature, stained with MF20 myosin heavy chain antibody (Developmental Studies Hybridoma Bank, University of Iowa) and counterstained with DAPI and imaged on a Zeiss AxioObserver Z1 inverted microscope with an AxioCamMR3 camera (Thornwood, NY, United States) (Shams and Kyba, 2023; Ippolito et al., 2012).

Measuring time to first division

Satellite cells were sorted from Micu1^MuSC-KO and Micu^FL/FL Pax7-ZsGreen mice by FACS and plated for live-cell imaging into 0.1% gelatin-coated 96-well plates (Mattek, P96G-1.5-5-F) (7,000 cells/well) containing growth medium. Cells from the two sources were treated with myoblast growth medium, MGM: Ham’s F10 medium with L-glutamine (HyClone, MT10070CV) containing 20% FBS (HyClone), 1% Glutamax (Gibco, 35050-061), 1% Pen/Strep (Gibco), and 10 ng/mL hb-FGF (PeproTech, Cranbury, NJ, United States), containing 5 mM 4-hydroxy tamoxifen (4OHT) (Sigma Aldrich, H7904) at the time of plating and provided fresh medium 18 h after plating. Time-lapse imaging was performed from 18 to 60 h after plating with a Nikon Eclipse Ti-inverted fluorescence microscope equipped with an automated stage (Prior) and a custom chamber to maintain a constant 37°C temperature, high humidity, and 5% CO₂. Multiple positions were analyzed per group with phase contrast images acquired every 10 min. Images were collected using a 20X CFI Plan Apochromat Lambda (NA = 0.75) objective (Nikon). For each condition, at least 100 individual cells were tracked. Following imaging, data were exported as individual TIFFs for each position and time point. ImageJ was used to stitch TIFF images from each location, and time to first division was determined for each as previously described (Larson et al., 2022; Shams et al., 2022).

Immunoblotting

After sorting (see above), 36,000 ZsGreen+ satellite cells were plated in MMM media on 0.1% gelatin-coated 24-well tissue culture plates and passaged once to 0.1% gelatin-coated 6-well plates to expand cells to ensure sufficient material. Cells were lysed in RIPA buffer (Thermo Scientific, 89900) with Halt Protease and Phosphatase Inhibitor Cocktail (Thermo Scientific, 78440), and protein concentrations were determined by BCA assay (Thermo Scientific Pierce, 23227). Protein samples (15 μg) were prepared with Bolt LDS sample buffer (Invitrogen, B0007) with 10% β-mercaptoethanol and run on Bolt Bis-Tris 4–12% gels (Invitrogen, NW04120BOX), then transferred to nitrocellulose membranes using a Bio-Rad Transblot Turbo device. Total protein was quantified from Ponceau staining, then blots were washed and blocked in Intercept TBS blocking buffer (LI-COR, 927–60001). Blots were incubated overnight at 4°C with the following primary antibodies: MICU1 (1:200; Sigma Aldrich, HPA037480), MCU (1:500; Cell Signaling, 14997S), EMRE (1:100, custom antibody from New England Peptide), and VDAC1 (1:1000; Santa Cruz Biotechnology, sc-390996). LI-COR IRDye secondary antibodies were used (LI-COR, 926–32211 and 926–68020), and detection was performed using an LI-COR Odyssey system, except for probing MICU1, for which HRP-conjugated anti-rabbit secondary (Cell Signaling Technology, 7074) was used. In this case, Clarity Western ECL Substrate (BioRad, 1705060) was used, and the blot was imaged on an Azure 300 Imaging System.

Seahorse analysis

After sorting (see above), 30,000 or 40,000 ZsGreen+ satellite cells were plated in MMM media on 0.1% gelatin-coated Seahorse XFp plates. After 2 days, media was replaced with Seahorse XF DMEM medium, pH 7.4 (Agilent, 103575-100) for 1 h in a 37°C non-CO2 incubator. Then the plate was transferred to the Seahorse XFp Analyzer (Agilent) for analysis. Cells underwent basal oxygen consumption rate (OCR) measurements followed by successive treatments with 1.5 μM oligomycin (Sigma, O4876), 2 μM carbonyl cyanide 4-(trifluoromethoxy)phenylhydrazone (FCCP; Sigma, C2920), and rot/ant solution with 1 μM rotenone (Sigma, R8875) and 1 μM antimycin-a (Sigma, A8674). OCR measurements were normalized to cell counts based on brightfield microscopy post-Seahorse.

Microscopy-based estimation of mitochondrial Ca²⁺

After sorting (see above), at least 10,000 ZsGreen+ satellite cells were plated in MMM media on 0.1% gelatin-coated glass-bottom dishes (Mattek, P35G-1.5-14-C). After 2 days, media was replaced with normal Tyrode’s (NT) solution (140 mM NaCl, 5.4 mM KCl, 0.53 mM MgCl₂, 0.33 mM NaH₂PO₄, 20 mM HEPES, 5.5 mM glucose, and 1.2 mM CaCl₂; pH 7.4) with the cell-permeable mitochondrial-localized Ca²⁺ indicator Rhod-2 AM (2.5 μM; R1245MP, Invitrogen) for 1 h at room temperature to ensure mitochondrial loading. Cells were then washed with NT solution and further incubated with the fluorescent mitochondrial dye MitoView Green (1.5 μM; 70054, Biotium) in NT solution for 30 min in a 37°C non-CO₂ incubator, which also served as a washout period for Rhod-2 AM. Following replacement of fresh NT solution, cells were imaged with a confocal microscope (LSM 900 with Airyscan 2, Zeiss) using 561 and 488 nm laser excitation with a 63× objective. During analysis in ImageJ software, a threshold was applied for the MitoView Green signal to create a binary mitochondrial mask of pixels corresponding to mitochondria, enabling the quantification of Rhod-2 AM fluorescence of the pixels contained within the mitochondrial mask.

TA injury and transplantation

Adult (2–4 months old) Micu1^MuSC-KO and Micu1^FL/FL Pax7-ZsGreen mice were anesthetized with ketamine and xylazine, both hind limbs were shaved and sterilized using surgical betadine solution, the skin over the TA was opened with a scissor, and both TA muscles were exposed. For injury-only experiments, 20 μL of cardiotoxin (10 μM in PBS, Sigma, 217503) was injected with a Hamilton syringe and the skin was closed using absorbable suture. For transplantation experiments, 48 h prior to transplantation of cells, C57BL/6J recipient mice were anesthetized with ketamine and xylazine and both hind limbs were subjected to 1,200 cGy irradiation using an RS 2000 Biological Research Irradiator (Rad Source Technologies, Inc., Suwanee, GA) with lead shields protecting the body and forelimbs. 24 h prior to transplant. 15 μL cardiotoxin was then injected into the TA under anesthesia as above. 24 h later, 300 ZsGreen+ satellite cells that had been collected by FACS from donor mice were transplanted in a volume of 10 μL PBS into each TA. 4 weeks after transplantation, transplanted TAs were prepared for FACS analysis as described above to count the number of self-renewed ZsGreen+ cells (Sullivan et al., 2025; Shams et al., 2025).

RNA isolation, reverse transcription and RT-qPCR

Total RNA was extracted using Qiazol reagent (Qiagen, 79306) as described in the manufacturer’s protocol. Isolated mRNA was directly used for RT-qPCR with Luna Universal One-Step RT-qPCR kit (New England Biolabs, E3005S) on the CFX Connect Real-time System (Bio-Rad, CA). Relative quantification of Micu1 to the housekeeping gene expression was determined using the following primer pairs: Micu1, 5′-CAA-CGA-ACC-TGG-TGA-AAC-CG-3′ and 5′-GTC-TGC-CAG-AAC-TGG-TTC-C-3’; eukaryotic translation initiation factor EIF35S, 5′-CTG-AGG-ATG-TGC-TGT-CTG-GGA-A-3′ and 5′-CCT-TTG-CCT-CCA-CTT-CGG-TC-3’; ribosomal protein 36B4, 5′-GGC-CCT-GCA-CTC-TCG-CTT-TC-3′ and 5′-TGC-CAG-GAC-GCG-CTT-GT-3’.

Histology

TA muscles were removed and placed in OCT Compound (Scigen Scientific 4586, Gardena, CA, United States), frozen in liquid nitrogen-cooled 2-methylbutane (Thermo Scientific, 019387) and stored at −80°C. 10 μm cryosections were cut on a Leica CM3050 S cryostat (Leica Microsystems, Buffalo Grove, IL, United States). Cryosections were air dried, then processed for immunohistochemistry.

Pax7 staining

Sections were fixed in 4% paraformaldehyde (PFA), washed with PBS, and boiled in heat-induced antigen retrieval buffer (1.8 mM citric acid and 8.2 mM sodium citrate in water) for 30 min using an Instant Pot pressure cooker (Instant Appliances). Sections were incubated for 10 min in H₂O₂ to block endogenous peroxidase activity and then blocked for nonspecific binding in 0.5% Akoya TNB blocking reagent [0.1 M Tris-HCl, pH 7.5; 0.15 M NaCl; 0.5% tyramide signal amplification (TSA) blocking reagent, FP1020] for 1 h at room temperature. Following blocking, sections were incubated with anti-Pax7 mouse IgG1 primary antibody (PAX7, Developmental Studies Hybridoma Bank, 1:10) and in TNB blocking buffer overnight at 4°C. After washing with PBS, sections were incubated with goat anti-mouse biotin-conjugated secondary antibody (115-065-205; Jackson Immuno Research Laboratories Inc, West Grove, PA; 1:1,000) in TNB blocking buffer for 2 h at room temperature. Visualization of the Pax7 primary antibody was achieved by incubation of the sections with the Vectastain ABC reagent (PK-6100; Vector Laboratories, Burlingame, CA) for 3 h and incubation in the dark with TSA cyanine 3 kit (NEL744001KT; Akoya Biosciences, Marlborough, MA; 1:50) in diluent buffer for 5 min. The sections then were mounted with ProLong Gold Antifade Mountant with 4,6-diamidino-2-phenylindole (DAPI) (Invitrogen, P36931).

Laminin staining

Briefly, slides were fixed with 4% paraformaldehyde, washed 3 times with PBS, permeabilized with 0.3% Triton X-100 (Sigma, X100), and blocked for 1 h with 3% BSA in permeabilization solution. Slides were incubated overnight at 4°C 1:500 with a mouse monoclonal antibody to laminin (Sigma, clone LAM-71, L8271), then sections were incubated for 90 min at RT with Alexa Fluor 488 goat anti-mouse IgG antibodies (Life Technologies, A11029). Coverslips were mounted with Immu-Mount (Epredia, 9990402). Slides were imaged with a Zeiss Axio Imager M2 with an AxioCam MRm camera (Carl Zeiss Microscopy, LLC, Thornwood, NY, United States). The cross-sectional areas of muscle fibers was done by analyzing the whole TA section using FIJI. All channels were split, and the laminin channel was used for further analysis. The brightness, contrast, and threshold were adjusted and applied in the same manner to all images. Images were converted to a mask and inverted. Images were then counted for the number of fibers and Feret diameter (Shams et al., 2022).

Quantification and statistical analysis

Data were analyzed with two-tailed unpaired Student’s t test for determining significant differences between two groups or two-way ANOVA with Tukey’s post hoc test for determining significant differences between three or more groups. Data were tested for normality using the Kolmogorov-Smirnov test. Differences were considered significant at p ≤ 0.05. Data are presented as mean ± standard error of the mean (SEM). All statistical testing was performed using GraphPad Prism 8.0 (GraphPad Software Inc., San Diego, CA, United States).

Published: June 25, 2026