Constitutive, Muscle-Specific Orai1 Knockout Results in the Incomplete Assembly of Ca²⁺ Entry Units and a Reduction in the Age-Dependent Formation of Tubular Aggregates

Abstract

Store-operated Ca²⁺ entry (SOCE) is a ubiquitous cellular mechanism that cells use to activate extracellular Ca²⁺ entry when intracellular Ca²⁺ stores are depleted. In skeletal muscle, SOCE occurs within Ca²⁺ entry units (CEUs), intracellular junctions between stacks of SR membranes containing STIM1 and transverse tubules (TTs) containing ORAI1. Gain-of-function mutations in STIM1 and ORAI1 are linked to tubular aggregate (TA) myopathy, a disease characterized by the atypical accumulation of tubes of SR origin. Moreover, SOCE and TAs are increased in the muscles of aged male mice. Here, we assessed the longitudinal effects (from 4–6 months to 10–14 months of age) of constitutive, muscle-specific Orai1 knockout (cOrai1 KO) on skeletal muscle structure, function, and the assembly of TAs and CEUs. The results from these studies indicate that cOrai1 KO mice exhibit a shorter lifespan, reduced body weight, exercise intolerance, decreased muscle-specific force and rate of force production, and an increased number of structurally damaged mitochondria. In addition, electron microscopy analyses revealed (i) the absence of TAs with increasing age and (ii) an increased number of SR stacks without adjacent TTs (i.e., incomplete CEUs) in cOrai1 KO mice. The absence of TAs is consistent with TAs being formed as a result of excessive ORAI1-dependent Ca²⁺ entry.

1. Introduction

Ca²⁺ homeostasis in skeletal muscle fibers is finely regulated through a complex interplay of various mechanisms, including (i) Ca²⁺ release from the sarcoplasmic reticulum (SR) during excitation–contraction (EC) coupling [1]; (ii) Ca²⁺ re-uptake into the SR by sarco-endoplasmic reticulum Ca²⁺ ATP-ase (SERCA) pumps [2,3,4,5,6]; and (iii) store-operated Ca²⁺ entry (SOCE) that promotes the influx of external Ca²⁺ when SR stores are depleted [7]. During repetitive stimulation, loss of intracellular Ca²⁺ due to the extrusion of Ca²⁺ ions from the myoplasm by Na⁺-Ca²⁺ exchangers (NCXs) and plasma membrane Ca²⁺ ATP-ases (PMCAs) could lead to a decrease in the available Ca²⁺ to sequester into the SR, a factor that can contribute to muscle fatigue [1,8]. Consistent with the possibility that this could lead to a reduction in SR Ca²⁺ content and the activation of SOCE, Gissel and Clausen [9] used ⁴⁵Ca²⁺ studies to show that significant Ca²⁺ influx occurs during sustained, repetitive muscle activation.

As in non-excitable cells, SOCE in skeletal muscle is coordinated by two main players [7,10,11]: STIM1 (stromal interaction molecule-1), a luminal endoplasmic reticulum (ER)/SR Ca²⁺ sensor [12,13,14], and ORAI1, a Ca²⁺ release-activated Ca²⁺ (CRAC) channel located in the surface membrane [15,16,17,18]. The activation of SOCE is initiated by Ca²⁺ store depletion, promoting oligomerization and cytoplasmic extension of STIM1 following the dissociation of Ca²⁺ ions from luminal STIM1 EF-hands, well-known Ca²⁺ binding motifs [19]. In turn, extended STIM1 oligomers activate ORAI1 in the plasma membrane, enabling Ca²⁺ entry [20,21].

Experimental evidence collected over the past decade has found that exercise promotes SOCE in the skeletal muscle fibers by promoting the formation of Ca²⁺ entry units (CEUs), dynamic junctions that assemble during exercise and disassemble during recovery [22,23]. CEUs are (i) small and few under sedentary control conditions [22]; (ii) constitutively assembled in fibers that lack calsequestrin 1 (CASQ1) protein expression [24], possibly due to the enhanced susceptibility for Ca²⁺ store depletion [25]; and (iii) absent in the muscles of aged sedentary mice [26], possibly due to the reduced activity-dependent enhancement of SOCE.

Under physiological conditions, ORAI1-dependent SOCE supports maintained force production during repetitive stimulation, promotes muscle fiber growth, and enhances the maintenance of oxidative, fatigue-resistant type I and type IIA fibers [22,23,24,27,28,29,30]. In addition, dysfunction in SOCE activity contributes to multiple pathophysiological conditions: (a) compromised muscle function in aging [26,31,32] and (b) myopathies, including muscular dystrophy (MD), malignant hyperthermia (MH), and tubular aggregate myopathy (TAM) [20,33].

TAM is a relatively rare disorder typically caused by gain-of-function mutations in both STIM1 and ORAI1 [34,35,36,37,38,39,40,41], and to a lesser degree, mutations in genes encoding for CASQ1 and the type-1 ryanodine receptor (RYR1) [34,42,43]. Clinically, TAM presentation is highly variable (from asymptomatic creatine kinase elevation to significant limb weakness, exercise intolerance, cramps, and muscle pain) and may be slowly progressive. However, an increased presence of tubular aggregates (TAs) in the skeletal muscle fibers of affected individuals is a key pathological hallmark observed in all TAM cases. TAs originate from SR membranes [44], as they stain positive for numerous SR proteins, including SERCA, triadin, and CASQ1 [45,46,47]. Also, the accumulation of some sarcolemmal proteins, including the dihydropyridine receptor (DHPR) and ORAI1, has been reported [26,43], though external membranes (i.e., transverse tubules or TTs) are excluded from TAs [45]. TAs are also observed in fast-twitch fibers of the extensor digitorum longus (EDL) muscles in aged male animals [45,46,47]. In mice, TAs are observed under conditions of altered SOCE activity (e.g., aging and TAM), which could contribute to both muscle weakness and increased susceptibility to fatigue. Interestingly, long-term voluntary wheel running exercise significantly reduces the incidence of TAs in aged male mice [26]. However, the precise molecular and morphological mechanisms that underlie the formation and stability of TAs are not fully understood.

Carrell et al. [27] described how slow-twitch soleus muscles from adult skeletal muscle-specific Orai1 knockout (KO) mice (cOrai1 KO) exhibit a significant reduction in myofiber cross-sectional area (CSA) and muscle mass, as well as the replacement of some fatigue-resistant type I fibers with hybrid fibers expressing both type I and type IIA myosins [27]. These findings are consistent with ORAI1 being important for the muscle fibers in the soleus to properly undergo the postnatal transition from fast to slow myosin expression, such that the transitioning fibers become suspended in a hybrid state [48]. In contrast, fast-twitch EDL muscles from these mice exhibited a smaller reduction in muscle mass and CSA without significant fiber type changes. Additionally, ex vivo peak specific force production was reduced in excised muscles and in vivo exercise endurance was compromised in cOrai1 KO mice [27].

In the present study, we assessed the longitudinal effects of skeletal muscle-specific Orai1 KO by comparing the muscle structure and function between 4–6-month-old and 10–14-month-old control and cOrai1 KO mice. The results from these studies indicate that cOrai1 KO mice exhibit a shorter lifespan, reduced body weight, exercise intolerance, decreased muscle-specific force and rate of force production, and an increased number of structurally damaged mitochondria. Unexpectedly, Orai1 gene ablation in the skeletal muscle prevents the formation of TAs with increasing age and increases the presence of SR stacks without the association of TTs, as occurs with CEUs that promote SOCE [22].

2. Materials and Methods

2.1. Animals

cOrai1 KO mice were generated as previously described [27]. Orai1-floxed mice lacking Cre-recombinase were used as the controls. All the mice were from a congenic C57bl/6N background. The animals were housed in microisolator cages at 20 °C with a 12 h light/dark cycle and provided free access to standard chow. The in vivo procedures and experiments were conducted according to the National Committee for the Protection of Animals used for Scientific Purposes (D. lgs n.26/2014) and approved by the Italian Ministry of Health (AN 313/2019-PR) or the University Committee on Animal Resources at the University of Rochester (UCAR2006-114E). Ex vivo experiments were performed on EDL muscles dissected from euthanized control and cOrai1 KO mice. The animals were euthanized by cervical dislocation, as approved by the Italian D. lgs n.26/2014 and the University Committee on Animal Resources at the University of Rochester (UCAR2006-114E).

2.2. Age-Dependent Survival Analysis of Animals Housed under Standard Conditions

The rate of spontaneous mortality of the control and cOrai1 KO male mice housed under standard conditions was monitored throughout their lifespan over an observational period of 30 months. The results were analyzed according to Kaplan–Meier survival curve analysis.

2.3. In Vivo Mouse Weight, Ex Vivo Adipose Tissue Weight, and Grip Strength Analyses

Whole body weight and grip strength were determined in the 4–6-month- and 10–14-month-old control and cOrai1 KO mice. After euthanasia, specific adipose depots (subcutaneous, epididymal, and retroperitoneal) were anatomically dissected and weighed. The tissue weights were normalized by dividing the absolute tissue weight by the body weight for each individual mouse and plotted as the percentage of body weight. The average amount of brown adipose tissue (BAT) was also evaluated and expressed as a percentage of the total adipose tissue (Supplementary Figure S1). The peak force produced by the mice during instinctive grasping (i.e., grip strength) was measured by holding the mice by the tails and lowering them onto metal grating connected to the shaft of a Shimpo Fgv 0.5× force transducer (Metrotec Group, Lezo, Spain). Once the mouse had firmly grabbed the grating, its tail was given a steady, gentle pull [49]. Measurements of the peak force generated by each mouse using both their fore- and hindlimbs were repeated three times with appropriate rest intervals (at least 30 s) to avoid fatigue. The highest value of peak force measured was recorded for each mouse.

2.4. The Treadmill Endurance Exercise Task

Ten-to eleven-month-old control and cOrai1 KO old mice were pre-trained on a 6-lane treadmill (Columbus Instruments, Columbus, OH, USA) at a modest treadmill speed of 5 m/min for 5 min at a 0° incline over 3 consecutive days. On the fourth day, the mice were subject to a 1 h endurance run on the treadmill (with a 700 m total distance), starting at 5 m/min for 10 min, followed by 25 min at a speed of 10 m/min, then 20 min at 15 m/min, and finally 5 min at 20 m/min. Continued running was encouraged by delivering brief (<1 s) sprays of air to the mouse’s backside using a Whoosh Duster™ (Control company, Houston, TX, USA). The number of rests during each 5 min window of time was recorded for each mouse. Exhaustion was defined as the inability of the mouse to re-engage on the treadmill after 3 consecutive <1 s sprays of air, as described previously [27]. The accumulative number of rests and the total running distance were recorded for each mouse.

2.5. Ex Vivo Muscle-Specific Force Measurements

Ex vivo assessments of muscle-specific force, the kinetics of force production/relaxation, and susceptibility to fatigue during repetitive high-frequency stimulation were made in the excised EDL muscles. Briefly, mice were anesthetized by intra-peritoneal injection of an anesthetic cocktail, as described previously [30]. EDL muscles from 10–11-month-old control and cOrai1 KO mice were isolated, tied using 4-0 surgical sutures, carefully excised, attached to a servo motor and force transducer (1200 A, Aurora Scientific, Aurora, ON, Canada), and placed between two platinum electrode plates in a chamber continuously perfused with oxygenated Ringer solution, containing 137 mM NaCl, 5 mM KCl, 1.2 mM NaH₂PO₄, 1 mM MgSO₄, 2 mM CaCl₂, 10 mM glucose, and 24 mM NaHCO₃, pH = 7.4. Before starting each experiment, the optimal stimulation intensity and muscle length (L_o) were determined using a series of 1 Hz twitch stimulation trains to guide stretching the muscle to the length that generated the maximal force (F_o). After establishing L_o, the muscles were first equilibrated using three 500 ms 150 Hz tetani delivered at 1 min intervals. The EDL muscles were then subjected to a force–frequency stimulation protocol (from 1 to 250 Hz for EDL muscles). To assess muscle fatigability, the EDL muscles were subjected to a repetitive, high-frequency stimulation protocol (60 stimulus trains of 50 Hz and 500 ms in duration delivered every 2.5 s). Muscle force was recorded using dynamic muscle control software v5.415 (Aurora Scientific, Aurora, ON, Canada) and analyzed using dynamic muscle analysis software v5.200 (Aurora Scientific, Aurora, ON, Canada). Muscle physiological CSA and specific force were calculated as described previously [50].

2.6. Sample Preparation for Histology and Electron Microscopy (EM)

The EDL muscles were quickly dissected from euthanized 4–6-month- and 10–14-month-old control and cOrai1 KO mice, pinned on Sylgard dishes, fixed at room temperature with 3.5% glutaraldehyde in 0.1 M NaCaCO buffer (pH 7.2), and stored in the fixative at 4 °C until embedding. The fixed muscles were then postfixed in osmium tetroxide (OsO₄), stained en bloc, dehydrated, and embedded as previously described [51,52]. For histological examination by light microscopy, semithin sections (~700 nm) were cut using a Leica Ultracut R microtome (Leica Microsystem, Vienna, Austria) with a diamond knife (Diatome, Biel, Switzerland) and stained in a solution containing 1% toluidine blue O and 1% sodium borate (tetra) in distilled water for 3 min on a hot plate at 55–60 °C. After washing and drying them, the sections were mounted with DPX media for histology (Sigma–Aldrich, Milan, Italy) and observed with a Leica DMLB light microscope connected to a DFC450 camera equipped with Application Suite v 4.13.0 (Leica Microsystem, CMS GmbH, Vienna, Austria) for Windows 11 pro v 23H2 (Microsoft, Seattle, WA, USA). For the EM, ultrathin sections (~50 nm) were cut using an Ultracut R microtome (Leica Microsystem, Vienna, Austria) with a diamond knife (Diatome, Biel, Switzerland) and double-stained with a uranyl acetate replacement and lead citrate. The sections were viewed in an FP 505 Morgagni Series 268D electron microscope (FEI Company, Brno, Czechia) equipped with a Megaview III digital camera and the Soft Imaging System v 3.2 (Olympus Soft Imaging Solutions, Munster, Germany) at 60 kV.

2.7. Quantitative Analyses of the Histological and EM Images

Histological sections of the EDL muscles from the control and cOrai1 KO mice stained in a solution containing 1% toluidine blue O and 1% sodium borate (tetra) in distilled water were analyzed with a Leica DMLB light microscope connected to a DFC450 camera (Leica Microsystem, Vienna, Austria) equipped with Application Suite v 4.13.0 (Leica Microsystem, Vienna, Austria) for Windows 11 pro v 23H2 (Microsoft, Seattle, WA, USA) to quantify the following:

−

Muscle fiber CSA: CSA was measured in light microscopy images from semithin transverse sections of whole EDL muscles from the 4–6-month- and 10–14-month-old control (n = 3–5) and cOrai1 KO (n = 3–5) mice by manually tracing the individual fibers, followed by automatic calculation of the fiber CSA with Application Suite software v 4.13.0 (Leica Microsystem, Vienna, Austria) for Windows 11 pro v 23H2 (Microsoft, Seattle, WA, USA).

−

Percentage of fibers with TAs: For quantitative analyses of TAs, light microscopy images of histological semithin transverse sections of the whole EDL muscles from the 10–14-month-old control (n = 3) and cOrai1 KO (n = 3) mice were taken. The percentage of muscle fibers containing TAs was evaluated in 1756 EDL fibers from the control mice and 1300 EDL fibers from the cOrai1 KO mice by counting the number of fibers presenting areas within the sarcoplasm with abnormal dark stained material.

EM images of the EDL muscles fibers from the control and cOrai1 KO mice were taken with a FP 505 Morgagni Series 268D electron microscope (FEI Company, Brno, Czechia) equipped with a Megaview III digital camera and the Soft Imaging System v 3.2 (Olympus Soft Imaging Solutions, Munster, Germany) and used for the following quantitative analyses:

−

Mitochondrial number and damage: The number of mitochondria/100 µm² and the percentage of severely altered mitochondria were evaluated from electron micrographs of non-overlapping regions randomly collected from longitudinal sections at 14,000× magnification. For each specimen, 10 fibers and 5 micrographs/fiber were analyzed. The number of severely altered mitochondria, classified as previously described [51], was reported as a percentage value of the total number of mitochondria.

−

TA number/fiber and TA average size: For quantitative analyses of the TA number/fiber and their average size, electron micrographs from transverse sections of whole EDL muscles collected from 10–14-month-old control (n = 3) and cOrai1 KO (n = 3) mice were taken.

−

Presence of SR stacks. The percentage of fibers presenting SR stacks and the number of SR stacks in 100 µm² were determined in electron micrographs of non-overlapping regions randomly collected from transverse sections at 28,000× magnification. For each specimen, 10–15 representative fibers and 5 micrographs/fiber were analyzed.

−

Total TT length at the I band. Extensions of the TT network into the I band region of the sarcomere (i.e., total TT length) were measured in electron micrographs of non-overlapping regions randomly collected from transverse sections at 28,000× magnification and reported as length (µm)/100 µm². The total TT network included both triadic TTs, identified as the emerging tubules between the two SR terminal cisternae in the transverse or longitudinal junctions, and non-triadic TTs, characterized by a narrow profile and lacking an electron-dense content. For each specimen, 10–15 representative fibers and 5 micrographs/fiber were analyzed. Quantitative analyses of both the SR stacks and TT network extensions were conducted using Analy-SIS software v 3.2 (Olympus Soft Imaging Solutions, Munster, Germany).

2.8. Statistical Analyses

Statistical significance was determined using either an unpaired, two-tailed Student’s t-test (when comparing means between two groups) or two-way ANOVA, followed by the post hoc Tukey test for repeated measures (when comparing across more than two groups) using Prism 9 software (GraphPad Software, Boston, MA, USA). The normal distribution of the data was checked with the Shapiro–Wilk normality test using Prism 9 software (GraphPad Software, Boston, MA, USA). When a Gaussian distribution was not confirmed, a non-parametric t-test (Mann–Whitney U test) was used. Survival data were evaluated by Kaplan–Meier analysis, and statistical significance was assessed using a Log-rank Mantel–Cox test. In all cases, differences were considered statistically significant at * p < 0.05. All the data are presented as means ± SEM.

3. Results

3.1. cOrai1 KO Mice Exhibit Reduced Survival

cOrai1 KO mice (i.e., mice specifically lacking the expression of the ORAI1 Ca²⁺ release-activated Ca²⁺ (CRAC) channel in the skeletal muscle) were generated using the cre-loxP system (crossing Orai1-floxed mice with MCK-cre mice) in the laboratory of Robert T. Dirksen (the University of Rochester), as described previously [27]. Orai1-floxed mice lacking Cre-recombinase were used as the controls. The rate of spontaneous mortality of the control and cOrai1 KO mice under standard housing conditions was determined during an observational period of 30 months. The survival curves of the control and cOrai1 KO mice are shown in Figure 1A. Interestingly, the mortality rate of the cOrai1 KO mice (50% survival < 500 days) was significantly higher (p < 0.01) than that of the control mice (50% survival < 700 days). Of note, no cOrai1 KO mouse survived beyond 600 days, while some of the control mice survived for up to 800 days. Moreover, the appearance of the cOrai1 KO mice at 14 months of age was visibly different from that of the age-matched control mice, including frizzy and faded fur, a smaller size and ocular opacity, hunched backs in a few cases, etc. (Figure 1B,C).

3.2. cOrai1 KO Mice Exhibit a Reduced Body Weight, EDL Mass and Cross-Sectional Area (CSA), and Grip Strength

We measured body weight, EDL mass, and grip strength in young adult (4–6 months) and older adult (10–14 months) control and cOrai1 KO mice (Figure 2). Similar to the results reported previously for younger cOrai1 KO mice [27], the average body weight (Figure 2A) and EDL mass (Figure 2B) were significantly reduced in the 4–6- and 10–14-month-old cOrai1 KO mice; thus, our results confirm the prior findings but across two distinct age ranges (4–6 and 10–14 months of age). This reduction in body weight between the two groups of mice was more pronounced at 10–14 months of age, as the control mice exhibited an increase in body weight between 4–6 and 10–14 months of age that was absent in the cOrai1 KO mice (Figure 2A). Moreover, different adipose deposits (e.g., cervical, interscapular, axillary–thoracic, mesenteric, abdominal–pelvic, retroperitoneal, inguinal) were anatomically dissected (Supplementary Figure S1) and weighed for both groups of mice at 4–6 (Supplementary Figure S1A,B) and 10–14 months of age and reported as percentages of total body fat (Supplementary Figure S1C). The brown adipose tissue (BAT) mass was also evaluated (Supplementary Figure S1D). The percentages of total body fat and BAT were unchanged between the 4–6-month-old control and cOrai1 KO mice. The total body fat content was increased in the 10–14-month-old mice despite a reduction in their body weight, suggesting the role of a reduction in muscle mass. Finally, BAT seemed to be differently affected in the two groups of mice at the older age (Supplementary Figure S1D). Carrell et al. [27] found that a reduction in the EDL mass of young adult cOrai1 KO mice was due in part to a modest reduction in the fiber CSA. Thus, we assessed the fiber CSA in histological images taken from transverse sections of the EDL muscles from both the 4–6-month- and 10–14-month-old control and cOrai1 KO mice (Figure 3). Consistent with prior results, analysis of the distribution frequency of the myofiber CSA revealed a modest shift to smaller areas in the cOrai1 KO mice at 4–6 months of age (Figure 3C). This shift to lower myofiber CSA values was even greater in the 10–14-month-old cOrai1 KO mice (Figure 3D), a feature that can explain the significantly reduced EDL mass observed in the cOrai1 KO mice at 10–14 months of age when compared to that observed at 4–6 months of age (see Figure 2B).

Moreover, consistent with the observed reduction in muscle mass (Figure 2B), the in vivo grip strength (normalized to body weight) was also significantly reduced in both the 4–6-month-old and 10–14-month-old cOrai1 KO mice (Figure 2C).

3.3. Ten- to Eleven-Month-Old cOrai1 KO Mice Exhibit Impaired In Vivo and Ex Vivo Skeletal Muscle Function

We compared the in vivo exercise endurance (Figure 4) and ex vivo EDL contractile function (Figure 5) of the 10–11-month-old control and cOrai1 KO mice (images of representative 10.5-month-old mice are shown in Figure 4A), an age just prior to the increased incidence of spontaneous death observed for the cOrai1 KO mice (see Figure 1A). In vivo exercise endurance was evaluated using a treadmill endurance run task in which the mice were encouraged to run for 1 h on a flat (0°) treadmill with a slowly increasing speed (see Materials and Methods for details). The total distance traveled was 700 m for the mice that completed the task. The 10–11-month-old control mice typically rested often, and ~30% refused to run at some point during the 1 h task (Figure 4B, black). Meanwhile, the 10–11-month-old cOrai1 KO mice rested even more often, and none of these mice completed the entire 1 h task (Figure 4B, red). As a result, the cOrai1 KO mice exhibited a significant reduction in their average total distance run during the task, consistent with greater exercise intolerance.

Since performance in behavioral assays such as the treadmill endurance task is subject to multiple variables in addition to intrinsic muscle function (e.g., mouse weight, motivation, motor unit complexity, synaptic transmission), we also compared the ex vivo contractile function of isolated EDL muscles from these same 10–11-month-old control (n = 7) and cOrai1 KO (n = 5) mice (Figure 5), which is a finding not measured or reported previously. The maximal magnitude of the electrically evoked specific force was reduced in the EDL muscles from the 10–11-month-old cOrai1 KO mice compared to that of the age-matched control mice (Figure 5A,B). A reduction in the peak specific force was observed across all stimulation frequencies, though statistical significance in this limited cohort was only observed at the lower stimulation frequencies. In addition, the maximal rate of specific force production, but not the maximum rate of relaxation, during both twitch and tetanic (150 Hz) stimulation was significantly reduced in the EDL muscles from the 10–11-month-old cOrai1 KO mice (Figure 5C). Finally, peak specific force production during repetitive, high-frequency stimulation was also significantly reduced in the EDL muscles from the 10–11-month-old cOrai1 KO mice (Figure 5D). Together, the results in Figure 4 and Figure 5 demonstrate that skeletal muscle function is significantly compromised in cOrai1 KO mice at an age immediately prior to the observed increase in mortality compared to that of the control mice.

3.4. Increased Percentage of Damaged Mitochondria in EDL Fibers from cOrai1 KO Mice

Mitochondrial loss, damage, and mislocalization are phenomena that are widely described in the literature, both in aging and in muscle diseases [51,52,53,54,55,56,57,58,59,60]. Thus, we quantified both the total number of mitochondria per 100 µm² and the percentage of damaged mitochondria in EM images of EDL muscle fibers from the 4–6- and 10–14-month-old control and cOrai1 KO mice (Figure 6). Most of the mitochondria in the EDL fibers of the control mice were not damaged, as they exhibited a dark matrix, with the inner and outer membranes clearly visible, and they were properly positioned within the I band of the sarcomere (Figure 6A). On the other hand, in some areas of the fibers from the cOrai1 KO animals, the mitochondrial ultrastructure was compromised (Figure 6B). Quantitative analyses revealed that the total number of mitochondria per 100 µm² was significantly reduced in the EDL muscles of the cOrai1 KO mice (Figure 6C), while the percentage of damaged mitochondria was significantly increased (Figure 6D), both at 4–6 months and 10–14 months of age, compared to that observed for age-matched control samples.

3.5. Tubular Aggregates (TAs) Do Not Form in EDL Muscles of Aging cOrai1 KO Mice

TAs are tightly packed straight tubes of SR membrane origin (see enlarged details in Figure 7A). TAs represent an age-related remodeling of the SR in the muscle fibers of male mice that is associated with dysfunction in SOCE [26,45]. TAs are also the main histopathological hallmark in the skeletal muscle fibers of patients affected by TAM, a disease associated with gain-of-function mutations in the human STIM1 and ORAI1 genes [34,35,36,37,38,39,40,41] and, more recently, with some mutations in the CASQ1 and RYR1 genes [34,42,43]. Here, we quantified the presence of TAs in histological transverse sections of the EDL muscles from 10–14-month-old control and cOrai1 KO male mice. Surprisingly, TAs were only found in EDL muscles from the control mice (Figure 7A, empty arrows) but not in samples from the cOrai1 KO mice (Figure 7B). Specifically, when they were present, we quantified the (i) percentage of fibers with TAs (Figure 7C); (ii) the average number of TAs per fiber (Figure 7D); and (iii) the average TA size (Figure 7E). The data obtained from these quantitative analyses of the TAs in the control mice and the evidence of their absence in the cOrai1 KO mice suggest that ORAI1 is crucial to the formation of TAs during aging.

3.6. EDL Fibers from cOrai1 KO Mice Exhibit an Increase in SR Stacks without TT Extensions

We previously demonstrated that exercise induces a remodeling of the SR to form flat parallel stacks in association with extensions of the TTs in the I band, which results in the formation of SR-TT junctions [22]. These SR-TT junctions within the I band are referred to as CEUs since they are associated with an increase in STIM1-ORAI1 co-localization [22], SOCE activity, SR Ca²⁺ store refilling, Ca²⁺ release, and force production during repetitive stimulation [23,24]. Here, we quantified the SR stacks and TTs in the I band in EM images of transverse sections from the EDL muscles of both young (4–6 months) and older adult (10–14 months) control (Figure 8A) and cOrai1 KO (Figure 8B) mice. These analyses revealed an increase in the percentage of fibers presenting SR stacks in the EDL fibers from the 10–14-month-old cOrai1 KO mice when compared to those of the age-matched control mice, while no such differences were observed in the EDL muscles from the younger mice (Figure 8C). In contrast, the number of SR stacks per unit area (100 µm²) was significantly increased in the EDL muscles at both ages (Figure 8D). We also quantified the total TT length within the I band, which is the second element required for a functional CEU. This analysis did not reveal any statistically significant change in the TT length in the I band in the EDL muscles of the cOrai1 KO mice (Figure 8E). The increased presence of SR stacks in the I band without a corresponding increase in associated TTs indicates that the number of functional CEUs was not increased in the EDL muscles of the cOrai1 KO mice.

4. Discussion

4.1. Main Findings

In a previous publication, cOrai1 KO mice were studied from 3 to 12 months of age, with most of the mice being 3 to 6 months old [27]. In order to assess the potential long-term effects of Orai1 ablation, here, we compared results obtained in cOrai1 KO mice across two age ranges: 4–6 months and 10–14 months of age. The older group range (i.e., 10–14 months of age) was chosen based on the reduced survival of the cOrai1 KO mice compared to that of the control mice (Figure 1A). During this period, the cOrai1 KO mice exhibit a reduced body weight, skeletal muscle mass, and grip strength (Figure 2), as well as a decreased fiber CSA (Figure 3), exercise intolerance (Figure 4), and reduced EDL-specific force production (Figure 5). We also detected increased mitochondrial damage (Figure 6), an alteration that could contribute to muscle dysfunction and that is commonly found in aging and several muscle disorders [51,53,56,58,59,60]. Overall, these results are consistent with, but more severe than, those previously reported for a younger mouse cohort [27]. Thus, Orai1 ablation results in a slowly progressive and increasing impairment of muscle function, consistent with ORAI1-dependent SOCE not only being important for muscle growth [11,27,30] but also for maintenance of muscle mass, force production, and exercise tolerance later in life.

Importantly, we also found that (a) EDL muscle fibers from aging cOrai1 KO mice do not develop TAs, as muscle fibers from age-matched control mice do (Figure 7), and (b) an increased number of SR stacks at both 4–6 months and 10–14 months of age (Figure 8), which, however, was not accompanied by elongation of the TTs in the I band, the second element required for full assembly of functional CEUs (Figure 8).

4.2. Our Present Results in Relation to Previous Findings

We previously suggested that the deficits in maximal force generation and exercise endurance observed in cOrai1 KO mice are due primarily to a reduced oxidative fatigue-resistant fiber content and calcium store capacity, while acute (1-month) deletion of Orai1 in adult muscle (using tamoxifen-inducible, muscle-specific Orai1 KO mice generated by crossing Orai1-floxed mice with muscle-specific, inducible human skeletal actin (HSA)-mutated estrogen receptor (Mer)-Cre-Mer (MCM) mice) did not significantly alter the muscle fiber type distribution, CSA, or force production [27]. However, Michelucci et al. [23] subsequently reported a marked reduction in sustained force production during repetitive, high-frequency stimulation in EDL muscles after the acute knockout of Orai1 in adult mice sufficient to abolish SOCE. Thus, ORAI1 function is clearly important for sustained muscle force production during repetitive stimulation. Unfortunately, neither our study nor Michelucci et al. [23] investigated the impact of long-term Orai1 knockout in adult mice (e.g., from 4 months of age to 12 months of age). Thus, the effect of long-term, tamoxifen-induced Orai1 knockout in adult mice on sustained muscle force production during repetitive, high-frequency stimulation and the age-dependent assembly of CEUs and formation of TAs remain unknown.

4.3. Lack of Orai1 Prevents the Assembly of Tubular Aggregates (TAs)

An important new finding of this study is the absence of TAs in the EDL muscles of 10–14-month-old male cOrai1 KO mice (Figure 7). TAs are peculiar aggregations of straight SR tubes found in muscle biopsies of patients with TAM, which are linked to gain-of-function mutations in ORAI1 and STIM1 [34,35,36,37,38,39,40,41,47]. The development of TAs resulting from mutations in ORAI1 and STIM1 could be the result of excessive constitutive Ca²⁺ entry, which could lead to chronically elevated myoplasmic Ca²⁺ levels. In this scenario, TAs may reflect a compensation for the inability of the SR/plasma membrane to clear excessive Ca²⁺. In addition, TAs are also observed in fast-twitch muscle fibers from aged male mice [26,45,46]. Some authors have suggested that with age, RYRs becomes leaky [61], and TAs serve as an additional Ca²⁺ buffer in cases of excessive Ca²⁺ accumulation in the cytoplasm. The addition of voluntary wheel running exercise prevents the formation of TAs in male mice during aging [26], possibly by limiting age-dependent enhancement of RYR Ca²⁺ leakage, as observed in sedentary mice.

The results presented in this study suggest that ORAI1-dependent Ca²⁺ entry is required for the formation of TAs in male mice during the aging process. This finding supports the idea that TAs may serve as an adaptive mechanism of the skeletal muscle designed to limit the damaging effects of excessive accumulation of Ca²⁺ in the myoplasm during aging and TAM (due to either excessive Ca²⁺ entry or SR leakage).

4.4. Lack of ORAI1 Results in Incomplete Assembly of Ca²⁺ Entry Units (CEUs)

The intracellular sites of STIM1-ORAI1 interaction in the skeletal fibers remained elusive for many years after SOCE was first identified in adult muscle [62,63]. However, experimental evidence collected over the past decade suggests that (i) STIM1 is mainly enriched in the SR present in the I band [22,30], (ii) ORAI1 is present within the TTs [22,30], and (iii) STIM1-ORAI1 co-localization in the I band increases during exercise due to the translocation of ORAI1-containing TTs from the triad to the I band [22]. Increased interaction between STIM1 and ORAI1 in the I band reflects the formation of CEUs, dynamic junctions between SR stacks and extended TTs [22]. Here, we found the incidence of SR stacks to be increased in the EDL muscles of the cOrai1 KO mice (Figure 8), though without associated TTs. The observation that this increase in SR stacks is not accompanied by extensions of the TTs indicates incomplete assembly of the CEUs and is consistent with the absence of ORAI1-dependent Ca²⁺ entry into the muscle fibers of cOrai1 KO mice [27]. The detailed molecular mechanisms that underlie the increased assembly of SR stacks in the skeletal muscle of cORAI1 KO mice are unclear and deserve further investigation. We know that the total releasable Ca²⁺ store content is significantly reduced in muscle fibers from adult cOrai1 KO mice but not in younger adolescent (4–6-week-old) cOrai1 KO mice (see Figure 1H in [27]). Hence, the increased presence of SR stacks in older cOrai1 KO mice could represent a futile attempt of the fibers to enhance Ca²⁺ entry and store content, which is doomed to fail due to the constitutive lack of ORAI1. As it has been shown that STIM1 interacts with and modulates the function of SERCA pumps [64], the formation of SR stacks may also reflect an attempt to augment SERCA function in order to sequester cytoplasmic Ca²⁺ better in older muscle fibers.

4.5. Lifespan Reduction and Premature Aging in Mice Lacking ORAI1

Aging is a physiological process in which cell division and tissue repair are not as efficient as in adulthood [65,66,67]. In skeletal muscle, aging is characterized by a loss of muscle mass (sarcopenia), reduced specific force production, and enhanced susceptibility to fatigue. Loss of muscle mass likely involves a combination of reduced fiber CSAs and the loss of fast-twitch motor units, with some being cross-innervated with slow-twitch motor neurons [68,69]. Ca²⁺ plays a crucial role as a second messenger in many skeletal muscle functions, including growth, contraction, and gene transduction. The lack of ORAI1-mediated Ca²⁺ entry into the skeletal muscle of cOrai1 KO mice alters muscle growth and fiber type determination [27]. Some of the changes observed in the muscles of 10–14-month-old cOrai1 KO mice resemble a premature aging phenotype [51,52], including reduced muscle mass (Figure 2), decreased specific force production (Figure 5), and increased mitochondrial damage (Figure 6), as does their visual appearance at the end stage (i.e., smaller body size, eye opacity, frizzy and faded fur, hunched backs in a few cases, etc.) (Figure 1C). In contrast, other findings are not consistent with a premature aging phenotype [22,26,45], including an increased incidence of SR stacks (Figure 8) and a reduction in TAs (Figure 7). These age-dependent changes could reflect muscle adaptations designed to counteract the lack of ORAI1-dependent SOCE, possibly to limit the degree of premature muscle aging. In any event, the decreased lifespan and reduced muscle mass/function of cOrai1 KO mice clearly indicate that ORAI1-dependent SOCE plays an important role in muscle maintenance throughout life.

5. Conclusions

SOCE was first measured in adult skeletal muscle in 2001 and was subsequently shown to be coordinated by STIM1 and ORAI1 in 2008 [10,11]. The first mutations in STIM1 and ORAI1 linked to TAM were identified several years later [35,36,39]. Finally, intracellular sites of exercise-induced interactions between ORAI1 and STIM1 in skeletal muscle fibers (i.e., CEUs) were first described in 2017 [22]. Nevertheless, a comprehensive understanding of STIM1-ORAI1 SOCE and its role in skeletal muscle health and disease have yet to be fully elucidated. Here, we show that muscle-specific Orai1 ablation in mice leads to a progressive impairment of skeletal muscle function that ultimately leads to a reduced lifespan.

The overall picture emerging from this work is that ORAI1 function is important for skeletal muscle growth, force production, exercise tolerance, and long-term maintenance of muscle mass. Importantly, we also show that the absence of Orai1 results in (a) absence of the age-related aggregation of TAs, consistent with TAs being formed as a result of excessive ORAI1-dependent Ca²⁺ entry, and (b) incomplete assembly of the CEUs, i.e., the TTs do not elongate to create contact with SR stacks, suggesting that ORAI1 is required for the elongation of the TTs and the formation of junctions with SR stacks in the I band. One interesting point that would be crucial to address in future experiments is whether the acute deletion of Orai1 in adult muscle (instead of constitutive knockout) would impact age-related TA formation and CEU assembly.