Membrane lipid rafts are disturbed in the response of rat skeletal muscle to short-term disuse

Abstract

Marked loss of skeletal muscle mass occurs under various conditions of disuse, but the molecular and cellular mechanisms leading to atrophy are not completely understood. We investigate early molecular events that might play a role in skeletal muscle remodeling during mechanical unloading (disuse). The effects of acute (6–12 h) hindlimb suspension on the soleus muscles from adult rats were examined. The integrity of plasma membrane lipid rafts was tested utilizing cholera toxin B subunit or fluorescent sterols. In addition, resting intracellular Ca²⁺ level was analyzed. Acute disuse disturbed the plasma membrane lipid-ordered phase throughout the sarcolemma and was more pronounced in junctional membrane regions. Ouabain (1 µM), which specifically inhibits the Na-K-ATPase α2 isozyme in rodent skeletal muscles, produced similar lipid raft changes in control muscles but was ineffective in suspended muscles, which showed an initial loss of α2 Na-K-ATPase activity. Lipid rafts were able to recover with cholesterol supplementation, suggesting that disturbance results from cholesterol loss. Repetitive nerve stimulation also restores lipid rafts, specifically in the junctional sarcolemma region. Disuse locally lowered the resting intracellular Ca²⁺ concentration only near the neuromuscular junction of muscle fibers. Our results provide evidence to suggest that the ordering of lipid rafts strongly depends on motor nerve input and may involve interactions with the α2 Na-K-ATPase. Lipid raft disturbance, accompanied by intracellular Ca²⁺ dysregulation, is among the earliest remodeling events induced by skeletal muscle disuse.

Mechanical unloading of human skeletal muscle under conditions of bed rest, joint immobilization, spinal cord injury, weightlessness during space flight, simulated microgravity and other forms of disuse leads to loss of muscle mass, wasting, and functional decline. Skeletal muscle function is essential for health and survival, but the molecular and cellular mechanisms of disuse-induced atrophy are not completely understood (3, 5, 7). Among animal models, hindlimb suspension (HS) of rodents is a well-validated experimental model that mimics many important effects of disuse on human skeletal muscle and has provided insight into the underlying molecular and cellular mechanisms (3, 20, 41, 54, 56).

It is known that HS induces progressive and marked atrophy of the postural skeletal muscles (3, 41, 54). Electromyogram (EMG) activity of the soleus muscle disappears immediately after onset of HS, and reduced motor nerve activity is evident on neurogram (15, 43). Atrophy is evident after 3–7 days of HS, in association with dramatic phenotypic changes that include a slow-to-fast shift in myosin heavy chain expression and genetic reprogramming (3, 5, 7, 54), altered ion channel expression (16, 51), intracellular calcium dysregulation (26, 32, 50, 54), cytoskeletal damage (42), decreased Na-K-ATPase activity and membrane excitability (29, 32), and altered intracellular signaling pathways (7, 40, 54). While these later events are well characterized, much less is known about early molecular events, which precede overt muscle atrophy. Previous studies have shown that protein imbalance in rat soleus muscle is initiated rapidly after only 1 day or less of HS (3, 20, 40, 55). A more complete understanding of the early molecular events induced by skeletal muscle disuse is needed to identify the important signaling pathways leading to this disorder (3).

Na-K-ATPase is essential for excitation and contraction of skeletal muscle (4, 9, 38, 52, 53, 58), which expresses the α1 and α2 subunit isoforms of Na-K-ATPase (44). Some studies suggest that the α2 Na-K-ATPase plays a special role in adaptations to skeletal muscle disuse (6, 29, 30, 48) and that alterations in α2 Na-K-ATPase content and activity are among the earliest detected disuse-induced events (29, 30). It was shown that short-term (1–3 days) HS alters Na-K-ATPase in rat soleus muscle in an isoform-specific manner, suggesting that α2 Na-K-ATPase alterations precede disuse-induced muscle atrophy (29). Moreover, even 6–12 h of HS specifically decreased the α2 Na-K-ATPase activity of rat soleus muscle (30). Both Na-K-ATPase and cholesterol-rich lipid microdomains are essential for a variety of vital cell functions, and reciprocal interactions between cholesterol and Na-K-ATPase are proposed (10, 12, 13, 24). While disuse-induced alterations of Na-K-ATPase function are well documented, the effects of skeletal muscle disuse on plasma membrane lipid rafts are poorly understood.

This study examines our hypothesis that changes in lipid rafts may be an early event accompanying changes in α2 Na-K-ATPase activity in the response of skeletal muscle to disuse. We analyzed the consequences of acute (6–12 h) HS on the stability of the plasma membrane lipid-ordered phase of rat suspended soleus muscle compared with the nonimpaired, continuously contracting diaphragm muscle from the same animals. Also, to test the possibility that reduced Na-K-ATPase activity can disorder membrane lipid rafts, we measured lipid raft distribution during exposure to 1 µM ouabain. This ouabain concentration specifically inhibits the rodent Na-K-ATPase α2 isozyme, the predominant α-isoform in skeletal muscle. Complexed methyl-β-cyclodextrin-cholesterol (MCD-chol) was used to replenish depleted membrane cholesterol. To verify the role of motor activity in the integrity of lipid rafts, experiments with repetitive nerve stimulation of soleus muscle after hindlimb suspension were performed. In addition, we identified early changes in intracellular Ca²⁺ levels that are known to involve a variety of signaling pathways.

METHODS

Animal care and use.

Experiments were performed on male Wistar rats (180–230 g). Animals were housed in a temperature- and humidity-controlled room with food and water ad libitum. All procedures involving rats were performed in accordance with the recommendations for the Guide for the Care and Use of Laboratory Animals (https://www.nap.edu/openbook.php?isbn=0309053773). The experimental protocol met the requirements of the EU Directive 2010/63/EU for animal experiments and was approved by the Bioethics Committee of Kazan State Medical University. The animals were subjected to HS individually in custom cages for 6 or 12 h, as described previously (41); control animals were not suspended. Soleus and diaphragm muscles were removed from euthanized animals. After removal, soleus and semi-diaphragm muscles with nerve stump were placed in a chamber with physiological solution containing (in mM) 137 NaCl, 5 KCl, 2 CaCl₂, 2 MgCl₂, 24 NaHCO₃, 1 NaH₂PO₄, 11 glucose, pH 7.4. The solution was continuously bubbled with 95% O₂ and 5% CO₂ and maintained at room temperature. These muscles were used immediately for Ca²⁺ imaging or for lipid raft staining, including experiments with electrical nerve stimulation.

Staining of lipid rafts.

Lipid rafts were stained with subunit B from cholera toxin (CTxB) conjugated to Alexa Fluor 488 (18). CTxB selectively interacts with patches of GM1 ganglioside, predominately located in lipid rafts (37). To identify the junctional membrane region, tetramethylrhodamine-α-bungarotoxin (αBtx, Biotium), a fluorescent-labeled specific ligand of nicotinic acetylcholine receptors (nAChRs), was used. The neuromuscular preparation was incubated for 15 min in physiological solution with CTxB (1 µg/ml) and αBtx (1 μM). The muscle was then incubated for 20 min in physiological solution with anti-CTxB antibody. Anti-CTxB antibody was added to cross-link the CTxB in lipid rafts into distinct patches on the plasma membrane. These patches were visualized by fluorescence microscopy. Subsequently, the preparation was washed for 30 min and visualized.

CTxB/αBtx fluorescence was excited by light of 488/555 nm wavelength, and emission was recorded using band-pass filters of 505–545/610–650 nm. Fluorescence was defined as the averaged fluorescence intensity in arbitrary units from pixels in junctional and extrajunctional membrane regions. Analysis of junctional fluorescence was performed in the region defined by αBtx staining. Extrajunctional fluorescence was calculated for an area (~200 μm²) of muscle membrane outside of the αBtx-positive region. Images were acquired using an Olympus BX51WI microscope with a confocal attachment Disk Speed Unit and LumPlanPF 100xw objective and captured with a DP71 (Olympus) CCD camera. Image analysis was performed using Cell^P (Olympus) and ImagePro software (Media Cybernetics, Bethesda, MD). Multiple Z-axis optical sections were taken using a focus stepper (ECO-MOT).

Methyl-β-cyclodextrin (MCD, 0.1 mM) was used to deplete plasma membrane cholesterol. Complexed MCD-cholesterol (MCD-chol) was used to replenish membrane cholesterol (27, 62). Muscles were exposed to 5 mM MCD-chol (dissolved in physiological solution) for 15 min. This time of application and concentration are sufficient to reach saturation with membrane cholesterol (27). After exposure to MCD-chol, the muscles were briefly washed and then labeled by incubation with CTxB (1 µg/ml) for 15 min. Before image acquisition, the stained muscles were rinsed 15–20 min in physiological solution.

Plasma membranes were also stained with fluorescent sterols. 22-NBD-cholesterol (22-(N-(7-Nitrobenz-2-Oxa-1,3-Diazol-4-yl)Amino)-23,24-Bisnor-5-Cholen-3β-Ol, Molecular Probes) was used as an environment-sensitive probe that localizes in the membrane's interior. 22-NBD-cholesterol fluorescence intensity increases in response to phase changes of membrane lipids from a raft to nonraft fractions; and its distribution in phosphatidylcholine/cholesterol bilayers is opposite to that of cholesterol (36, 45). The spectral properties of 22-NBD-cholesterol are almost independent of the phase state of the bilayer (45). A stock solution of 22-NBD-cholesterol (10 mg/0.1 ml in ethanol) was prepared and used at a final concentration of 0.2 μM in physiological saline. Muscles were incubated for 20 min.

Dehydroergosterol [ergosta-5,7,9(11),22-tetraen-3β-ol, DHE], the fluorescent sterol most structurally and functionally similar to cholesterol, has been proven to be a useful probe for real-time imaging of the sterol environment and intracellular sterol trafficking in living cells. DHE is a naturally occurring, fluorescent sterol analog that mimics many properties of cholesterol; DHE preferentially associates with lipid rafts and is used as a probe for cholesterol distribution (19, 39). A stock solution of DHE (5 mM, Sigma Aldrich) was made in ethanol and stored under nitrogen at –80°C. For staining, DHE labeling solution containing DHE-methyl-β-cyclodextrin was prepared as described (23). The final concentration of DHE used for labeling muscles in physiological saline was 60 µM (incubation for 40 min). DHE reaches a stable distribution after 30–60 min that remains virtually unchanged during chase periods up to 18 h in CHO cells (23).The final concentration of ethanol in physiological saline containing 22-NBD-cholesterol or DHE did not exceed 0.001%.

Freshly isolated muscles were incubated with physiological saline (bubbled with 95% O₂ and 5% CO₂) containing 1 μM αBtx and 0.2 μM 22-NBD-cholesterol (or 60 μM DHE) for 20 min (or 40 min), respectively, followed by a 20 min wash. Superficial regions of the muscle were imaged with a ×40, 1.3 numerical aperture objective using a Leica TCS SP5 confocal system configured for concurrent viewing of rhodamine and 22-NBD-cholesterol or DHE fluorescence.

Stimulation protocol.

In some experiments, staining with CTxB was performed immediately following repetitive stimulation of the soleus muscle after hindlimb suspension. A freshly isolated muscle with nerve stump was placed in the chamber and continuously perfused with physiological solution. The muscle was stimulated via the nerve to produce evoked contractions for 15 min (cyclic stimulation 15 Hz for 1 min/1.5 Hz for 1 min). Stimulation was achieved by delivering suprathreshold 0.2-ms voltage pulses (DS3 Digitimer) via platinum electrodes placed near visually identified motor nerve branches.

Measurement of resting intracellular Ca²⁺ levels.

The high-affinity cell-permeant Ca²⁺ indicator Fluo-4AM was dissolved in DMSO and stored frozen for up to 1 wk in the dark. Prior to the experiment, Pluronic F-127 was added into an aliquot of Fluo-4AM to facilitate dissolving Fluo-4AM in physiological solution. The final concentration of Fluo-4AM was 1 μM, and the contents of DMSO and Pluronic F-127 were <0.001%. The isolated intact muscle was stained with 1 μM Fluo-4AM for 10 min at room temperature. Afterwards, the specimen was perfused with a physiological solution for 40 min, and fluorescence measurements in junctional and extrajunctional membrane regions were performed. The time elapsed between skeletal muscles removal and measurement of intracellular Ca²⁺ concentration was ~60 min.

Fluo-4AM was excited using light of 488 nm wavelengths and its emission was recorded using a band-pass 505–550 nm filter. The fluorescence signal was calibrated using a Ca²⁺ calibration kit (57), consisting of Ca²⁺-free (10 mM K₂EGTA, 100 mM KCl, and 10 mM MOPS, pH 7.2) and 40 µM Ca²⁺-containing (10 mM CaEGTA, 100 mM KCl, and 10 mM MOPS, pH 7.2) solutions. Calibrations were performed by exposing each loaded muscle to controlled Ca²⁺ buffers in the presence of an ionophore (10 µM ionomycin) to equilibrate extra- and intracellular Ca²⁺ levels. The following equation was used to obtain [Ca²⁺]_free for experimental samples from measured values of F: [Ca²⁺]_free = K_d * [F − F_min]/[F_max − F], where F_min is fluorescence of Fluo-4AM in the absence of Ca²⁺, F_max is fluorescence of Ca²⁺ saturated Fluo-4AM, and F is fluorescence at intermediate Ca²⁺ levels. Note that, initially, the same equation was used to determine the ion dissociation constant (K_d = 345 nM).

Images were acquired using an Olympus BX51WI microscope with a confocal attachment Disk Speed Unit and UPLANSapo 60хw objective and captured with a CCD camera (Orca R2, Hamamatsu). Image analysis was performed using Cell^P and ImagePro software.

Statistical analysis.

Data are given as means ± SE. Statistical significance of the difference between group means was evaluated using a Student’s t-test (ORIGIN 6.1. software) and one-way ANOVA followed by the Bonferroni post hoc test (ORIGIN Pro 8 software). A value of P < 0.05 was considered significant. The sample sizes (number of measurements) achieved a statistical power > 80%.

RESULTS

Acute hindlimb suspension modulates membrane lipid distribution.

Previous investigations have demonstrated a direct molecular interaction between cholesterol and nAChR. Cholesterol and lipid rafts serve as a signaling platform for nAChR clustering (34, 61) and also contribute to maintaining endplate electrogenesis (31). Therefore, we investigated whether HS may interfere with plasma membrane lipids at the neuromuscular junction (NMJ) and extrajunctional membrane regions using a specific ligand of nAChRs (rhodamine-conjugated α-bungarotoxin, αBtx), a specific marker of lipid rafts, CTxB, and the lipid environment-sensitive probes, 22-NBD-cholesterol and dehydroergosterol (DHE) (see methods).

As expected, both the αBtx signal, which identifies the junctional membrane region (nAChRs), and the CTxB signal identifying lipid rafts were localized at the NMJ of control soleus muscles. The localization remained unchanged after 6 h of HS (Fig. 1A). HS did not affect the fluorescence from nAChRs; however, CTxB fluorescence decreased by 39% (P < 0.01) (Fig. 1B). In extrajunctional membrane regions, the initial CTxB fluorescence was threefold lower than in junctional regions; however, it also decreased by 28% (P < 0.01) after HS (Fig. 1B). The decrease in CTxB fluorescence may be caused by “outflow” of GM1 molecules from lipid raft domains due to their disorganization. This result suggests that acute HS induces a disturbance of lipid rafts that is more pronounced in junctional membrane regions.

Fig. 1.

Hindlimb suspension (HS) disturbs the distribution of membrane lipid rafts. A, top: control. The nicotinic acetylcholine receptors and lipid rafts localize at the muscle endplate. A rat soleus muscle was dual-labeled with rhodamine-conjugated α-bungarotoxin (red channel) and CTxB (green channel). Overlap (orange channel). A, bottom: same labeling of an endplate from a rat soleus muscle after 6 h of HS. B: averaged fluorescence (in arbitrary units) from control rats (4 muscles, 20 junctional and 30 extrajunctional membrane regions) and rats after 6 h of HS (4 muscles, 20 junctional and 39 extrajunctional membrane regions). Black bars, junctional; white bars, extrajunctional membrane regions of muscle fibers. **P < 0.01 compared with corresponding control. Scale bars, 10 μm.

The αBtx signal also localizes with the 22-NBD-cholesterol signal (green channel) at the NMJ of control soleus muscles. Again, the localization remained unchanged after 6 h of HS (Fig. 2A); however 22-NBD-cholesterol fluorescence increased by 82 ± 12% (P < 0.01) (Fig. 3A). The fluorescence from nAChRs was not changed after HS (not shown). Therefore, 6 h of HS did not alter localization of nAChRs at the cholesterol rich NMJ, but it significantly increased 22-NBD-cholesterol fluorescence at the NMJ. In extrajunctional membrane regions, the initial 22-NBD-cholesterol fluorescence was ~2.5-fold lower than at junctional regions; however, it also increased to the same extent by 96 ± 22% (P < 0.01) after HS (Fig. 3A).

Fig. 2.

Staining of plasma membrane with fluorescent sterols. A, top: control. A rat soleus muscle was dual-labeled with rhodamine-conjugated α-bungarotoxin (red channel) and 22-NBD-cholesterol (green channel). A, bottom: same labeling after 6 h of HS. B, top: control. A rat soleus muscle was dual-labeled with rhodamine-conjugated α-bungarotoxin (red channel) and DHE (blue channel). B, bottom: same labeling after 6 h of HS. Scale bars, 10 μm.

Fig. 3.

Acute hindlimb suspension disrupts the distribution of plasma membrane lipids. Averaged fluorescence (in arbitrary units) from control muscles and muscles after 6 h of HS labeled with 22-NBD-cholesterol (A) or DHE (B). A: control, 4/4 muscles, 25/27 junctional and 27/27 extrajunctional membrane regions; after 6 h of HS, 5/5 muscles, 42/24 junctional and 44/24 extrajunctional membrane regions. B: control, 4/4 muscles, 34/26 junctional and 34/27 extrajunctional membrane regions; after 6 h of HS, 5/5 muscles, 49/34 junctional and 49/34 extrajunctional membrane regions. Numbers for soleus/diaphragm muscles are indicated above. Black bars, junctional; white bars, extrajunctional membrane regions of muscle fibers. *P < 0.05; **P < 0.01 compared with corresponding control.

22-NBD-cholesterol strongly associates with lipid-disordered membrane regions, and its fluorescence decreases upon lipid-ordering phase formation (36, 45). In the mouse neuromuscular junction, interventions that decrease lipid raft formation (marked with CTxB) strongly increase 22-NBD-cholesterol fluorescence (27). This suggests that the increase in 22-NBD-cholesterol fluorescence observed both in junctional and in extrajunctional membranes after 6 h of HS may reflect a reduction in lipid-ordered regions, decreased raft stability, and expansion of lipid-disordered membrane phases.

The observed decrease in DHE fluorescence at both junctional and extrajunctional membrane regions (Figs. 2B and 3B) further supports this interpretation. In model and biological membranes, DHE codistributes with cholesterol, with a strong preference for lipid rafts (19). Therefore, both the DHE and the 22-NBD-cholesterol measurements suggest that acute HS disrupts lipid rafts.

In addition, we compared the DHE and the 22-NBD-cholesterol signals in inactive soleus and continuously active diaphragm muscle from the same animal. The diaphragm and soleus signals changed in the same direction (Fig. 3, A and B). However, the changes were much less pronounced in the diaphragm, suggesting that lipid raft organization follows the level of muscle use. This finding suggests that lipid raft integrity is sensitive to the level of muscle use per se, independent of any other factors related to HS.

Reduced Na-K-ATPase activity disordered membrane lipid rafts.

The dependence of Na-K-ATPase activity on cholesterol has previously been shown, and reciprocal cholesterol/Na-K-ATPase interactions are proposed (10, 12, 13, 24). It has been shown that 6 h of HS specifically reduced the α2 Na-K-ATPase activity of rat soleus muscle, evident as a loss of electrogenic potential contributed by the α2 pump in both junctional and extrajunctional membrane regions, with the greatest depolarization occurring at the junctional membrane (30).

It is possible that Na-K-ATPase activity is required to maintain ordered membrane lipid rafts. To test this possibility, we measured CTxB fluorescence in soleus muscles exposed to 1 µM ouabain. In rat skeletal muscle, 1 µM ouabain specifically inhibits the ouabain-sensitive Na-K-ATPase α2 isozyme, the major isoform of skeletal muscle, without effect on the minor, ouabain-resistant α1 isozyme (25, 33). In control muscles, a 15-min exposure to 1 µM ouabain decreases CTxB fluorescence in junctional and extrajunctional membrane regions by ~30% and ~20% (P < 0.05), respectively (Fig. 4 A, C, and D; see also Fig. 6). This result suggests that inhibition of the Na-K-ATPase α2 isozyme induces a disturbance of lipid rafts that is similar to the effect of HS alone (Fig. 1) and is also more pronounced in junctional membrane regions. Conversely, after 6 h of HS, CTxB fluorescence did not change during ouabain addition (Fig. 4, B, E, and F; see also Fig. 6). This observation is consistent with our earlier findings that 6 h of HS itself specifically inhibits the α2 Na-K-ATPase activity (30); accordingly, 1 µM ouabain is ineffective after HS.

Fig. 4.

Effect of reduced Na-K-ATPase activity on the distribution of membrane lipid rafts. A and B: a rat soleus muscle was dual-labeled with rhodamine-conjugated α-bungarotoxin (red channel) and CTxB (green channel) (overlap, orange channel) to identify lipid rafts at the neuromuscular junction. A: control muscle. B: same labeling after 6 h of HS. Top: 0 min, before 1 µM ouabain addition. Bottom: 15 min, same membrane regions after 15 min of ouabain action. Scale bars, 10 μm. C–F: averaged CTxB fluorescence (in arbitrary units). C and D: control rats (8 muscles from 6 rats). E and F: rats after 6 h of HS (8 muscles from 6 rats). C and E: no change in CTxB fluorescence up to 15 min in untreated muscle in normal physiological solution. D and F: changes in CTxB fluorescence after addition of 1 µM ouabain (arrows). Closed symbols, junctional; open symbols, extrajunctional membrane regions of muscle fibers. *P < 0.05 compared with initial CTxB fluorescence before ouabain addition (indicated as 1.0).

Fig. 6.

Replenishment of membrane cholesterol restores lipid rafts disturbed by ouabain and hindlimb suspension. A: rat soleus muscles were labeled with CTxB (green channel). Left, control muscle; right, same labeling in muscle after 6 h of HS. Fluorescent images after initial CTxB staining (CTxB), after addition of 1 µM ouabain for 15 min (Oua), after subsequent addition of MCD-chol for 15 min (Oua+Chol), and after a second staining with CTxB (Oua+CTxB) are shown. Scale bars, 10 μm. B and C: averaged CTxB fluorescence (in arbitrary units) after initial CTxB staining (white bars), after 15-min exposure to 1 µM ouabain (light gray bars), after 15 min subsequent addition of MCD-chol (dark gray bars), and after a second staining with CTxB (black bars). Each bar corresponds to the respective images in A. B and C: junctional (B) and extrajunctional membrane regions (C). *P < 0.05, **P < 0.01, ***P < 0.001 compared with corresponding initial CTxB staining (white bars). Bars represent mean data from 4 muscles obtained from 4 rats for each group.

The fluorescence from nAChRs during ouabain action did not change either in control or in suspended muscles (see Fig. 4, A and B; statistical bars not shown).

These findings demonstrate that ouabain alone, acting on the Na-K-ATPase α2 isozyme, is able to produce a specific disturbance of lipid rafts similar to that produced by acute HS. In other words, inhibition of Na-K-ATPase α2 activity either by HS alone or by 1 µM ouabain induces the same disturbance of lipid rafts. Moreover, these observations indicate that the effects of ouabain and acute HS are not additive.

Complexed MCD-chol was used to test whether the observed disturbance of lipid rafts results from cholesterol loss (see methods). This compound enriches membranes with cholesterol and restores membrane lipid rafts destroyed by MCD-induced cholesterol depletion (27, 62). In control soleus muscles, a 15-min exposure to 0.1 mM MCD decreased CTxB fluorescence in the junctional membrane by ~30% (P < 0.05), similar to ouabain (Fig. 5A). After washout with normal physiological solution for 15 min followed by a second staining with CTxB, CTxB fluorescence was unchanged (Fig. 5A). Because newly formed lipid rafts are not expected to be labeled, the second staining with CTxB confirmed that the MCD-induced destruction of lipid rafts is irreversible. However, if MCD treatment is followed by treatment with complexed MCD-chol, a second CTxB staining returned CTxB fluorescence to initial levels (Fig. 5B), indicating restoration of lipid rafts.

Fig. 5.

Methyl-β-cyclodextrin-cholesterol (MCD-chol) restores membrane lipid rafts destroyed after MCD-induced cholesterol depletion. A and B: junctional membrane regions of rat soleus muscle. Muscles were initially stained with a marker for lipid rafts (CТхВ, open bars), then exposed to 0.1 mM MCD for 15 min. CTxB fluorescence decreased significantly in MCD (MCD, light gray bars). Subsequently, muscles were incubated with normal physiological solution (NS, dark gray bar, A) or MCD-chol (MCD-chol, dark gray bar, B) followed by a second staining with CTxB (CTxB, black bars, A and B). Fluorescence did not return to initial levels during MCD washout (A); however, it was completely restored after replenishment of membrane cholesterol with MCD-chol (B). y-Axis, averaged CTxB fluorescence (in arbitrary units) after initial CTxB staining. *P < 0.05 compared with corresponding initial CTxB staining (white bars). Bars represent mean data from 4 muscles obtained from 3 rats in both A and B experiments.

This protocol was further used to test whether replenishment of membrane cholesterol might restore lipid rafts disturbed by ouabain and HS. As shown above, a 15-min exposure to 1 µM ouabain decreased CTxB fluorescence in junctional membrane regions in control muscles but was ineffective in muscles after 6 h of HS (Fig. 6, A and B). The subsequent addition of complexed MCD-chol (in the presence of ouabain) did not change CTxB fluorescence in either control or suspended muscles (Fig. 6, A and B). However, a second staining with CTxB returned CTxB fluorescence to initial (before ouabain addition) levels both in control muscles and after 6 h of HS (Fig. 6, A and B). Notably, the initial CTxB fluorescence in suspended muscles was twofold lower than in control muscles; however, it also increased up to initial control fluorescence (Fig. 6B). Similar but less pronounced changes were observed in extrajunctional membrane regions (Fig. 6C). These data indicate that enrichment of membranes with exogenous cholesterol can restore lipid rafts disturbed either by ouabain or by acute HS and suggest that the disturbance results from cholesterol loss.

An additional experiment with short-term (15 min) 1 µM ouabain treatment followed by washout with physiological solution without complexed MCD-chol was performed (Fig. 7). These experiments demonstrate that, in control muscles, the disturbance of lipid rafts induced by short-term binding of 1 µM ouabain to Na-K-ATPase α2 isozyme is reversible, even without cholesterol replenishment (Fig. 7A). A number of studies have reported that Na-K-ATPase undergoes internalization upon ouabain binding, and this process occurs in a time scale of hours (11, 28). The duration of ouabain treatment in the current study (15 min) is too short to elicit longer-term processes such as Na-K-ATPase internalization, translocation, or protein turnover. This finding demonstrates that short-term ouabain binding to α2 Na-K-ATPase is sufficient to induce lipid raft disruption; and conversely, this effect is reversible after ouabain removal. On the other hand, no spontaneous recovery of lipid rafts is observed after 6 h of HS (Fig. 7B), and replenishment of membrane cholesterol is required (Fig. 6).

Fig. 7.

Disruption of lipid raft stability by ouabain is reversible; disordering after hindlimb suspension is stable. Averaged CTxB fluorescence (in arbitrary units) after initial CTxB staining (white bars), after 15 min exposure to 1 µM ouabain (light gray bars), after 15 min subsequent washout with normal physiological solution (dark gray bars), and after a second staining with CTxB (black bars) is shown. Junctional membrane regions of rat soleus muscle. A: control muscles. B: muscles after 6 h of HS. **P < 0.01 compared with corresponding initial CTxB staining (white bars). Bars represent mean data from 4 muscles obtained from 4 rats for each group.

In summary, both HS- and ouabain-induced disturbance of lipid rafts can be reversed by replenishment of membrane cholesterol, suggesting that the disordering in both cases results from cholesterol loss. However, the disordering actions of acute HS and ouabain on lipid raft stability are not additive and differ in their ability to spontaneously recover.

Motor activity specifically restores lipid rafts at junctional membrane regions.

As recently demonstrated, low-intensity muscle workload specifically restores electrogenic activity of the α2 Na-K-ATPase in acutely suspended rat soleus muscle (30). To test whether lipid rafts are also restored, CTxB fluorescence was measured in HS-suspended soleus muscles immediately after nerve stimulation producing 15-min contractions (see methods for details).

In addition, in these experiments, HS was prolonged up to 12 h to test whether the disuse-induced lipid rafts changes are stable. After 12 h of HS, CTxB fluorescence was approximately twofold lower (P < 0.01) compared with control both in junctional and in extrajunctional membrane regions (Fig. 8, A and B). These results confirm that HS-induced lipid raft disturbances are not temporary and reversible. Repetitive nerve stimulation was able to specifically restore (P < 0.01) CTxB fluorescence in junctional but not extrajunctional membrane regions (Fig. 8), providing further evidence that lipid raft integrity depends strongly on soleus muscle use.

Fig. 8.

Repetitive stimulation of rat soleus muscle after 12 h of hindlimb suspension specifically restores lipid rafts in junctional sarcolemma region of muscle fibers. CTxB fluorescence was measured in junctional (A) and extrajunctional (B) membrane regions in control muscles (white bars), after 12 h of HS (gray bars), and after 12 h of HS followed by stimulation via nerve for 15 min (cyclic stimulation 15 Hz for 1 min/1.5 Hz for 1 min, see methods) (black bars). **P < 0.01, ***P < 0.001, 4 muscles from 4 rats for each group.

Hindlimb suspension locally decreases resting intracellular Ca²⁺ levels near the NMJ.

Calcium is an essential second messenger for a range of muscle functions and is involved in a variety of signaling pathways including gene activation (14). Different studies have reported that resting intracellular calcium concentration ([Ca²⁺]_i) in soleus muscle fibers both increases (26, 32, 54) and decreases during prolonged HS (50). However, [Ca²⁺]_i changes during the initial hours of HS have not been examined. In this study, we tested whether the resting [Ca²⁺]_i of soleus muscle fibers is altered during 6 h of HS. We also investigated whether a gradient of [Ca²⁺]_i may exist between junctional and extrajunctional regions.

We found that the initial [Ca²⁺]_i in control muscles was twofold higher in junctional (~150 nM) compared with extrajunctional (~75 nM) membrane regions (Fig. 9, P < 0.01). This difference might result from residual Ca²⁺ resulting from entry through the endplate nAChRs, which is known to produce a local increase in [Ca²⁺]_i (2). After 6 h of HS, resting [Ca²⁺]_i near junctional membranes decreased twofold (P < 0.01), while [Ca²⁺]_i in extrajunctional regions remained unchanged; consequently, the intracellular Ca²⁺ gradient from junctional to extrajunctional regions decreased and the [Ca²⁺]_i throughout the whole muscle fibers became homogeneous (Fig. 9).

Fig. 9.

Resting cytosolic calcium concentration, [Ca²⁺]_i, in rat soleus muscles in control and after 6 h of hindlimb suspension. Black bars, junctional; white bars, extrajunctional membrane regions of muscle fibers. Four muscles of each group. **P < 0.01 between junctional and extrajunctional regions. After 6 h of HS, the normally enhanced Ca²⁺ levels near the NMJ disappeared and [Ca²⁺]_i at junctional and extrajunctional regions equalized.

It was previously shown that ouabain binding triggers slow [Ca²⁺]_i oscillations that result from a complex interplay between several Ca²⁺-dependent signaling pathways (1, 17). Similar ouabain-induced [Ca²⁺]_i oscillations were observed in our experiments both in control soleus muscles and after 6 h of HS; however, [Ca²⁺]_i changes were significantly (P < 0.05) different in suspended compared with control muscles at the same time points (Fig. 10). This finding suggests that the effects of acute HS may involve changes in intracellular Ca²⁺ handling.

Fig. 10.

Effect of ouabain on [Ca²⁺]_i in soleus muscles. Open symbols, control muscles; filled symbols, muscles after 6 h of HS. A: no change in [Ca²⁺]_i during 25 min of preincubation in the absence of ouabain. B and C: addition of 1 µM and 500 µM ouabain (arrows) induces intracellular Ca²⁺ oscillations in both control and suspended muscles at junctional (B) and extrajunctional (C) regions. Four muscles of each group. *P < 0.05 difference between suspended and control muscles at corresponding time points. y-Axis, initial [Ca²⁺]_i level before ouabain addition is set as 1.00.

DISCUSSION

Acute disuse induces changes in membrane lipid integrity.

A normal segregation of plasma membrane lipid phases is essential to maintain membrane fluidity, curvature, ion channel and transporter functions, and compartmentalization. In particular, cholesterol-rich membrane domains (lipid rafts and caveolae) form scaffolds for assembly of multiple signaling complexes (13, 34, 47). A number of neurodegenerative diseases are associated with disturbances in lipid raft integrity (49). Our results demonstrate that acute disuse disrupts plasma membrane lipid-ordered rafts in all membrane regions of soleus muscles, with the greatest disturbance at the NMJ. These findings suggest that the stability of lipid rafts is subject to regulation by skeletal muscle motor activity.

The molecular mechanisms by which muscle disuse disrupts lipid rafts are not known. The submembranous actin-based cytoskeleton has been shown to play an essential role in lipid raft stability (46). Accordingly, acute (6–12 h) HS is reported to alter the cytoskeleton and transverse stiffness of rat soleus muscle via an effect on nonmuscle α-actinins (42).

In addition, reciprocal interactions between Na-K-ATPase and cholesterol have been proposed (10, 12, 13, 24, 31) and, respectively, our results demonstrate a close relationship between the Na-K-ATPase α2 isozyme and lipid raft stability. Acute disuse decreases α2 Na-K-ATPase pump electrogenic activity and depolarizes membrane potential in soleus muscles (30), accompanied by disturbance of lipid rafts (this study). Conversely, repetitive nerve stimulation restored both Na-K-ATPase α2 isozyme electrogenic activity (30) and lipid rafts in suspended soleus muscles (this study). MCD-induced lipid rafts disordering was shown to be associated with specific inhibition of α2 Na-K-ATPase isozyme electrogenic activity (31). Conversely, acute inhibition of Na-K-ATPase α2 activity by ouabain in control, nonsuspended soleus muscles induces a disturbance of lipid rafts similar to that induced by acute disuse alone (this study). Finally, both ouabain- and disuse-induced disruption of lipid rafts can be reversed by replenishing membrane cholesterol, suggesting that the disturbance results from cholesterol loss.

Collectively, these findings indicate reciprocal cholesterol/α2 Na-K-ATPase interactions. However, the question remains open whether HS-induced Na-K-ATPase loss induces lipid raft disturbance or vice versa. The molecular mechanisms of Na-K-ATPase regulation by surrounding lipids is well described (13, 22). Conversely, we can only speculate regarding the possibility that α2 Na-K-ATPase may regulate lipid stability. Acute HS inhibits α2 Na-K-ATPase activity presumably via a FXYD1-dependent regulatory mechanism accompanied by increased FXYD1 abundance (30). Such regulation is expected to stabilize the enzyme in the plasma membrane (24), consistent with our finding that HS does not change α2 Na-K-ATPase membrane localization (30). Therefore, the role of α2 Na-K-ATPase in lipid raft stability may be mediated by its ouabain receptor site. Notably, ouabain inhibits Na-K-ATPase transport by stabilizing the enzyme in an E2, nontransporting conformation. Three specific lipid/Na-K-ATPase interactions are proposed that either stabilize the protein or stimulate or inhibit Na-K-ATPase activity, with separate binding sites and distinct kinetic mechanisms. Both stimulatory and inhibitory lipid interactions poise the conformational equilibrium toward the E2 state (22). These findings together with results of this study lead us to propose that reciprocal interactions between cholesterol and α2 Na-K-ATPase are more favored in the E2 enzyme conformation.

Intriguingly, it was recently shown that plasma membrane depolarization itself can induce nanoscale reorganization of membrane phospholipids (60). Therefore, the possibility exists that acute HS may alter membrane lipid phase ordering due to loss in α2 Na-K-ATPase isozyme activity and sarcolemma depolarization.

Acute disuse modulates intracellular Ca²⁺ oscillations induced by ouabain binding to α2 Na-K-ATPase.

Calcium is an essential second messenger for a range of muscle functions, including actin microfilaments and the cytoskeletal scaffold (14), and caveolae/lipid rafts are involved in Ca²⁺-signaling pathway regulation (47). Na-K-ATPase, acting through its extracellular ouabain-specific binding site, functions as a membrane receptor for multiple intracellular signaling pathways (35). Our study further shows that ouabain binding to the α2 Na-K-ATPase subunit induces slow intracellular Ca²⁺ oscillations in control rat soleus muscle that are similar to the ouabain-induced Ca²⁺ oscillations reported in other cell types (1, 17). These oscillations result from a complex interplay between Ca²⁺ transporters, various Ca²⁺ channels, Ca²⁺-binding proteins, and Ca²⁺ efflux initiated in direct response to ouabain binding (1, 17). Acute HS modulates these Ca²⁺ oscillations, suggesting disuse-induced changes in intracellular Ca²⁺ handling.

Acute disuse locally decreases resting intracellular Ca²⁺ levels near the NMJ.

Some studies have shown that resting intracellular calcium concentration ([Ca²⁺]_i) increases in soleus muscle fibers within the first few days of HS (26, 32, 54). An increase in [Ca²⁺]_i might be a key trigger to downstream signaling events leading to muscle atrophy. In particular, it was proposed that accumulated Ca²⁺ might trigger activation of calpains, further promoting muscle atrophy (54, 59). Conversely, some studies have reported a decrease of resting [Ca²⁺]_i in soleus muscle fibers during prolonged HS (50). If [Ca²⁺]_i changes are time dependent, as is likely, these diverse results may simply reflect changes in [Ca²⁺]_i at different durations of HS.

These divergent results are explained in part by our present finding that acute (6 h) HS alters [Ca²⁺]_i differently in distinct sarcolemma regions; decreased [Ca²⁺]_i is observed only near the NMJ. In our experiments, the initial [Ca²⁺]_i in control muscles was twofold higher in junctional compared with extrajunctional membrane regions. This significant difference might result from residual Ca²⁺ entry through endplate nAChRs from quantal and nonquantal acetylcholine, which is known to produce a local increase in [Ca²⁺]_i (2). HS-induced local decrease in [Ca²⁺]_i at the NMJ may result in part from the reduced synaptic input that is evident on neurogram immediately at the onset of HS (15, 43), which is expected to reduce Ca²⁺ entry through the open nAChRs.

Based on the observation that [Ca²⁺]_i is higher in the slow-twitch soleus muscle compared fast-twitch extensor digitorum longus muscle, prolonged HS is expected to decrease [Ca²⁺]_i during a shift of soleus muscle toward a faster phenotype (50). However, the acute, local decrease in intracellular Ca²⁺ level at the junctional regions does not follow the expected pattern of a global Ca²⁺ decrease during fiber phenotype switching.

Our finding suggests that a disuse-induced local decrease in [Ca²⁺]_i at the NMJ is an early step in the intracellular Ca²⁺ signaling pathways involved in the development of atrophy. These findings are important because reduced nerve activity on neurogram (15, 43) is one of the earliest detectable changes produced by HS. The junctional nuclei (3–6 per NMJ) represent a distinct subset of muscle nuclei (8) which are stably anchored beneath the post-synaptic membrane by Syne proteins (21). The presence of junction-localized nuclei is expected to render the NMJ exquisitely sensitive to remodeling in response to altered patterns of nerve input. Because [Ca²⁺]_i is a key signaling molecule for gene transcription, decreased local [Ca²⁺]_i at the NMJ is expected to drive remodeling of the NMJ, which, in turn, may be a primary event that initiates global remodeling of the muscle fiber.

Conclusion.

This is the first demonstration that membrane lipid rafts are subject to regulation by skeletal muscle motor activity and that disruption of lipid rafts is an early event in the disuse-induced progression to atrophy. The finding that even short-term nerve stimulation can specifically restore lipid rafts in junctional membrane regions of disused muscles provides further evidence that motor nerve input is a major driver for lipid raft integrity. Because lipid rafts serve as platforms for assembly of multiple signaling complexes, their disruption may drive many early changes in intracellular signaling pathways. The finding that acute disuse differently alters resting [Ca²⁺]_i in distinct sarcolemma regions of the rat soleus muscle is novel and was unexpected. It suggests that the NMJ, and subjunctional myonuclei, may play a leading role in the molecular remodeling that progresses to atrophy. A local decrease in junctional [Ca²⁺]_i following the loss of nerve input is expected to remodel the NMJ and influence global muscle fiber remodeling.

In addition, we demonstrate that a reciprocal relationship exists between lipid raft stability and the Na-K-ATPase α2 isozyme. The nature of this interaction is not completely understood. A simple hypothesis is that α2 Na-K-ATPase/cholesterol interactions are important for membrane lipid raft ordering. In other words, lipid rafts are essential for Na-K-ATPase targeting and function and, conversely, the presence of Na-K-ATPase α2 in lipid rafts helps maintain their structure and stability.

Finally, we should note that the conclusions of this study are subject to the limitations of pharmacological manipulations and fluorescence microscopy. Future studies using superresolution microscopy and expression of recombinant proteins will be required to identify the precise molecular interactions between skeletal muscle motor activity and sarcolemma lipid raft integrity.

GRANTS

This work was supported by Russian Foundation for Basic Research Grant 16-04-00562; by Russian Science Foundation Grant 16-15-10220; by St. Petersburg State University Research Grant 1.50.1621.2013; by Swedish Research Council Grant K2013-55X-14191-12-3, Strategic Research Program in Diabetes at Karolinska Institutet; by the Novo Nordisk Foundation Grants 10559 and NNF14OC0012731; and by National Institute of Arthritis and Musculoskeletal and Skin Diseases Grant R01 AR063710.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

A.M.P., J.A.H., and I.I.K. conceived and designed the research; A.M.P., V.V.K., and A.N.V. performed experiments; A.M.P., V.V.K., A.N.V., A.V.C., and I.I.K. analyzed data; A.M.P., A.L.Z., A.V.C., J.A.H., and I.I.K. interpreted results of experiments; A.M.P., A.N.V., and I.I.K. prepared figures; A.M.P., J.A.H., and I.I.K. drafted manuscript; A.M.P., V.V.M., A.V.C., J.A.H., and I.I.K. edited and revised manuscript; A.M.P., V.V.K., V.V.M., A.N.V., A.L.Z., A.V.C., J.A.H., and I.I.K. approved final version of manuscript.