A chronic high-fat diet exacerbates contractile dysfunction with impaired intracellular Ca²⁺ release capacity in the skeletal muscle of aged mice

Abstract

Obesity and aging reduce skeletal muscle contractile function. However, it remains unclear whether obesity additively promotes muscle contractile dysfunction in the setting of aging. In this study, we investigated skeletal muscle contractile function ex vivo and intracellular Ca²⁺ release in male C57BL/6J mice fed a low-fat diet (LFD) or a high-fat diet (HFD) for 4 or 20 mo. Tetanic force production in the extensor digitorum longus muscle was decreased by aging or HFD feeding, and the further reduction was observed in aged HFD mice. The 20-mo HFD-fed mice, not the 20-mo LFD-fed mice or 4-mo HFD-fed mice, showed reduced intracellular Ca²⁺ peak levels by high concentration of caffeine (25 mM) compared with 4-mo LFD mice. Aging and HFD feeding additively increased intramyocellular lipid (IMCL) levels and were associated with the degree of impaired muscle contractile force and peak Ca²⁺ level. These data suggest that impairment in the contractile force in aged muscle is aggravated by HFD, which may be due, at least in part, to dysfunction in intracellular Ca²⁺ release. The IMCL level may be a marker for impaired muscle contractile force caused by aging and HFD.

NEW & NOTEWORTHY The aim of this study was to examine the effect of high-fat diet (HFD)-induced obesity on contractile function and Ca²⁺ release capacity in aged skeletal muscle. Not only were the force production and peak Ca²⁺ levels decreased by aging and HFD feeding, respectively, but also, these interventions had an additive effect in aged HFD-fed mice. These data suggest that the impairment in the contractile force in aged muscle is aggravated by a HFD, which may be due to synergistic dysfunction in intracellular Ca²⁺ release.

INTRODUCTION

Developed countries are aging rapidly, causing both health and economic problems (5, 42). Physical function declines with age, and it was shown that decreased muscle strength, but not reduced muscle mass, was associated with increased mortality (33). On the other hand, obesity due to overeating and lack of exercise causes major health problems worldwide (15, 48), and thus the number of elderly obese individuals is increasing. To overcome this health problem, we need to understand the pathophysiology related to various aging-related diseases in this population. The influence of obesity on impaired muscle strength caused by aging is an important issue but one that has not yet been fully clarified.

It has been reported that muscle-force production normalized to muscle mass is substantially lower in aged (18) and obese mice (10). Skeletal muscle contractile properties are regulated by the contraction-relaxation cycle that depends on intracellular Ca²⁺ release and uptake into the sarcoplasmic reticulum (SR) (2, 12, 44), and previous studies showed that Ca²⁺ release levels were decreased in aged (3) and obese (7, 45) skeletal muscle. These results suggest that both aging and obesity may impair muscle contractile function through reduced Ca²⁺ release capacity, but the detailed mechanism has not been fully elucidated. It also remains unclear whether obesity further exaggerates impaired muscle contractile force by aging.

Against this background, this study investigated the effects of high-fat diet (HFD)-induced obesity and aging on contractile dysfunction and impaired Ca²⁺ release capacity in skeletal muscle.

MATERIALS AND METHODS

Animals.

Male C57BL/6J mice were housed in cages in a temperature-controlled room under a 12-h light-dark cycle. During a 1-wk adaptation period, all mice were fed standard chow and water ad libitum. After an initial acclimatization period, animals (2 mo of age) were randomly assigned to the following groups: 4-mo low-fat diet (LFD; n = 16), 4-mo HFD (n = 17), 20-mo LFD (n = 18), and 20-mo HFD (n = 10). The HFD mice were fed a diet containing 20% protein, 60% fat, and 20% carbohydrate (D12492; Research Diets, New Brunswick, NJ), and the LFD mice were fed standard chow containing 22.6% protein, 5.6% fat, and 53.8% carbohydrate [a pelleted Charles River Formula (CRF)-1 diet; Charles River, Japan]. After 4 mo (when mice were 6 mo old) or 20 mo (when mice were 22 mo old) of the HFD or standard chow diet, we performed blood and muscle sampling for further analysis as described below. All animal experiments in this study were approved by the Animal Experimental Committee of Juntendo University.

Glucose and insulin tolerance tests.

Intraperitoneal glucose tolerance tests (GTTs) were performed after mice had fasted overnight (16–18 h; 0.5 g/kg body weight). Intraperitoneal insulin tolerance tests (ITTs) were performed (5–6 h) with regular human insulin (0.75 U/kg body weight; Eli Lilly). Glucose and insulin were injected intraperitoneally, and blood glucose levels were assessed by tail bleeds at 0, 15, 30, 60, 90, 120, and 150 min. The fasting plasma insulin levels were measured using an ELISA kit (Morinaga Co., Kanagawa, Japan), according to the manufacturer’s protocol. The homeostasis model assessment of insulin resistance (HOMA-IR) was calculated as previously described (31).

Muscle preparation.

Mice were anesthetized with an intraperitoneal injection of sodium pentobarbital (70 mg/kg body weight). Once a surgical level of anesthesia was reached, the muscles were dissected, and the abdominal visceral fat was collected and weighed (41). The extensor digitorum longus (EDL), soleus (SOL), and flexor digitorum brevis (FDB) muscles were used.

Muscle contraction measurement.

The force-frequency relationship was assessed in intact EDL and SOL muscle samples, as previously described (13, 14). Briefly, isolated EDL and SOL muscle preparations were mounted on a force transducer (UL-100; Minebea, Tokyo, Japan), and the hock was fixed in a chamber containing Krebs solution [in mM: 120 NaCl, 5 KCl, 2 CaCl₂, 1 magnesium chloride (MgCl₂), 1 NaH₂PO₄, 25 NaHCO₃, and 11 glucose], bubbled with 95% O₂-5% CO₂. At the end of the 5-min equilibration, the isolated muscle was stimulated with 500 ms trains of current pulses at 1, 10, 20, 30, 40, 50, 70, 100, and 150 Hz at 1-min intervals, and the contractile force was measured at an acquisition rate of 1 kHz. The optimal muscle length was determined by the maximum twitch force. After muscle length was measured, the muscles were removed from the chamber, the tendons were dissected out, and muscle weight was measured. Specific force was calculated from the absolute force, muscle weight, and muscle length, assuming a density of 1.056 g/mL. At the end of the experiments, the EDL muscles were frozen rapidly in isopentane, cooled in liquid nitrogen, and subsequently used for histochemical staining.

Intracellular Ca²⁺ measurement.

Intracellular Ca²⁺ measurement was performed in intact FDB muscle samples, as previously described (14). The excised, intact FDB muscles were digested under gentle agitation for 2 h at 37°C in a rodent Krebs solution, supplemented with 2 mg/mL collagenase type II (CLS-2; Worthington Biochemical, Lakewood, NJ). The muscles were then suspended in Ringer’s without collagenase type II and triturated repeatedly using Pasteur pipettes in which their tips had been fire polished to produce openings of a successively smaller size. Single FDB fibers were then plated on 35 mm glass-bottomed dishes and allowed to settle for >30 min before experimentation. The fibers were used within 7 h of isolation. Single FDB fibers were loaded with 5 µM Fluo-4 AM (Invitrogen/Molecular Probes, Waltham, MA) and 5 µM Fura-2 AM (Dojindo, Kumamoto, Japan) in the dark for 30 min at room temperature (22–25°C), followed by a washing with dye-free Krebs solution (in mM: 142 NaCl, 5 KCl, 2 CaCl₂, 1 MgCl₂, 5 HEPES, and 10 glucose, pH 7.4). Then, single FDB fibers were washed in dye-free solution, supplemented with 25 μM N-benzyl-p-toluene sulfonamide (BTS; AAJ64910MA; Fisher Scientific) for 10 min to block contractions. Fluorescence images of the Fluo-4- and/or Fura-2-loaded fibers were captured via an objective lens [CFI Super Fluor× 20, numerical aperture (NA) 0.75; Nikon, Tokyo, Japan] and a cooled charge-coupled device (CCD) camera (ORCA-ER; Hamamatsu Photonics, Shizuoka, Japan) at an acquisition rate of 200 Hz (Fluo-4) and 1.42–10 Hz (Fura-2). The fluorescence images were processed using AQUA-COSMOS/ratio software (Hamamatsu Photonics). Ca²⁺ concentration ([Ca²⁺]) levels were assessed using Fluo-4 during electrical stimulation with individual 500 ms trains at 1 and or 100 Hz. The peak fluorescence intensity value (F) was measured during contraction, and the quantification data were plotted as changes from the precontraction level (F0).

In the pharmacological experiment in intact FDB muscle, resting [Ca²⁺] levels were assessed by changes in the Fura-2 signal intensity in response to caffeine (C0750; Sigma-Aldrich, St. Louis, MO) and 4-chloro-m-cresol (4-CmC; C55402; Sigma-Aldrich) (36). We superfused caffeine or 4-CmC with a Ca²⁺-free solution, and peak [Ca²⁺] levels were assessed using Fura-2 during caffeine (120 s) or 4-CmC (30 s) stimulation. Single FDB fibers were preincubated in Krebs-Henseleit buffer (KHB) solution, supplemented with 30 mM magnesium chloride (MgCl₂; M1028; Sigma-Aldrich) for 10 min before 4-CmC stimulation. Adenosine triphosphate (ATP) was used from Sigma-Aldrich (A2383). The peak fluorescence intensity value (R; ratiometric images) was measured during stimulation, and the ratiometric quantification data (ΔR/R0 ratio) are expressed as changes from the prestimulation levels (R0). In each experiment, the solutions were superfused at a rate of 2 mL/min. We placed a plastic ring wall around the glass-bottom dishes to limit the perfusion volume less than ~300 μL.

Western blotting analysis.

The expression levels of skeletal muscle Ca²⁺ regulatory proteins were analyzed by Western blotting, as previously described (13). The following primary antibodies were used: anti-type 1 ryanodine receptor (RyR) antibody 34C (MA3-925; Thermo Fisher Scientific, Waltham, MA), anti-dihydropyridine receptor (DHPR) antibody 20A (ab2864; Abcam, Cambridge, MA), anti-calsequestrin (CSQ) antibody VIIID12 (MA3-913; Thermo Scientific), anti-sarcalumenin (SAR) antibody XIIC4 (sc-58845; Santa Cruz Biotechnology, Dallas, TX), anti-SR Ca²⁺ ATPase 1 (SERCA1) antibody IIH11 (MA3-911; Thermo Scientific), anti-parvalbumin (PV) antibody (ab32895; Abcam), anti-FK506 binding protein 12 (FKBP12) antibody (ab2918; Abcam), and anti-glyceraldehyde 3-phosphate dehydrogenase (GAPDH) antibody 14C10 (Number 2118; Cell Signaling Technology, Danvers, MA). The protein levels were normalized to the GAPDH expression level and are expressed relative to the levels in 4-mo LFD mice.

Immunofluorescence microscopy.

Histological sections of mouse muscle fibers were examined and analyzed as previously described (14). The lipid droplets in muscle sections were measured using LipiDye (Lipid Droplet Green; Funakoshi, Tokyo, Japan). In brief, sections fixed with 4% formaldehyde were incubated with 1 μM LipiDye for 30 min at 4°C. Then, the sections were washed two times with phosphate-buffered saline (PBS). Images (1,024 × 1,024 pixels) were acquired using the laser lines of an LSM780 confocal laser-scanning microscope [40×/1.4 NA oil objective lens (PlanApo; Carl Zeiss MicroImaging, Inc., Oberkochen, Germany)]. The image was analyzed using either Spectralyzer (custom-designed) or ImageJ software. The fluorescence percentage of individual fibers was calculated with an ImageJ macro application. The quantitative fluorescence analyses were performed by counting ~39 random muscle fiber sections in each group.

Electron microscopy.

Sample preparation and electron microscopy quantitative analysis were performed, as previously described (12). In brief, the EDL muscles were fixed with 2.5% glutaraldehyde (TAAB) in 0.1 M phosphate buffer (pH 7.4), followed by postfixation with 1% OsO₄ in the same buffer. Fixed specimens were dehydrated with a graded series of ethanol and embedded in Epok 812 (Oken Shoji, Tokyo, Japan). Ultrathin sections were cut and stained with uranyl acetate and lead citrate. The specimens were examined with an HT7700 transmission electron microscope (Hitachi, Tokyo, Japan). Next, electron micrographs were taken at high magnification using the Soft Imaging System. Individual lipid droplets from three mice per group were manually traced in longitudinal orientations using ImageJ software. Quantitative analyses were performed using ~114 random lipid droplets in a longitudinal muscle area from mice in each group.

Statistical analysis.

Values are expressed as means ± SE. Statistical analyses were performed using Prism version 7.0 (GraphPad Software, San Diego, CA). To compare two nonpaired groups, Student’s t test was applied. For comparisons of more than two groups, a two-way ANOVA was applied, followed by Bonferroni post hoc tests. In all tests, P < 0.05 was considered statistically significant.

RESULTS

Effects of HFD on physical characteristics and metabolic parameters in aged mice.

Metabolic parameters were measured after LFD or HFD administration for 4 mo (when mice were 6 mo old) or 20 mo (when mice were 22 mo old). Compared with LFD, HFD feeding resulted in a significant increase in body weight, abdominal visceral fat, fasting glucose levels, insulin level, and HOMA-IR in both the 4-mo and 20-mo groups (Table 1). Similarly, the HFD significantly deteriorated glucose tolerance at both time points (Fig. 1, A and B). In addition, ITT revealed that insulin sensitivity in the 4-mo HFD and 20-mo LFD groups was trending to be lower, whereas significantly deteriorating in the 20-mo HFD group (Fig. 1, C and D). The weights of the EDL muscles were lower in the 20-mo LFD group but unchanged in the 4-mo HFD or 20-mo HFD group compared with the 4-mo LFD group (Table 1).

Table 1.

Physiological parameters

	4-mo LFD	4-mo HFD	20-mo LFD	20-mo HFD
Body weight, g	30.8 ± 0.5	49.4 ± 0.7**	34.4 ± 0.7	56.8 ± 4.9**#††
EDL weight, mg	13.1 ± 0.4	13.0 ± 0.4	11.3 ± 0.8*	13.0 ± 0.2
SOL weight, mg	10.0 ± 0.3	11.4 ± 0.4	10.5 ± 0.2	10.8 ± 0.5
Abdominal visceral fat, g	0.48 ± 0.3	2.53 ± 0.2**	0.99 ± 1.0	3.47 ± 0.6**#††
Fasting glucose, mg/dL	83.9 ± 2.7	128.1 ± 12.0**	97.9 ± 3.1	178.4 ± 10.2**#††
Fasting insulin, ng/mL	0.23 ± 0.04	1.07 ± 0.18**	0.28 ± 0.02	0.89 ± 0.08**
HOMA-IR, IU/L × mg/dL/405	1.20 ± 0.16	8.63 ± 1.61**	1.76 ± 0.10	10.38 ± 1.36**

Values are the means ± SE; n = 5–18 per group. EDL, extensor digitorum longus; HFD, high-fat diet; HOMA-IR, homeostasis model assessment of insulin resistance; LFD, low-fat diet; SOL, soleus.

^*P < 0.05 and

^**P < 0.01 vs. 4-mo LFD mice.

^#P < 0.01 vs. 4-mo HFD mice.

^††P < 0.01 vs. 20-mo LFD mice.

Fig. 1.

Parameters of insulin sensitivity. A: glucose tolerance test (GTT). B: area under the curve (AUC) quantification of GTT results; n = 5–10 per group. C: insulin tolerance test (ITT). D: AUC quantification of ITT results; n = 5–9 per group. HFD, high-fat diet. The data are presented as the means ± SE. *P < 0.05 vs. 4-mo low-fat diet (LFD).

Effects of HFD on muscle contractile function in aged mice.

Next, we measured muscle contractile function in EDL and SOL muscles in all groups. The specific force at stimulation frequencies ranging from 50 to 150 Hz in the EDL muscles was lower in the 4-mo HFD and 20-mo LFD groups compared with the 4-mo LFD group. Furthermore, the contractile force was further reduced in the 20-mo HFD group (Fig. 2A). Similar findings were seen in absolute force data (Fig. 2B). Indeed, the time-to-peak (TTP) force and the half-relaxation time (HRT) during twitch stimulation were prolonged in EDL muscle in the 20-mo HFD group (Fig. 2C). On the other hand, the specific force, the absolute force, and TTP and HRT in SOL muscle were comparable among groups (Fig. 2, D–F). These data suggest that HFD and aging additively impair muscle contractile force in the EDL muscle but not the SOL muscle. In addition, prolonged TTP and HRT in EDL muscle samples in the 20-mo HFD group suggest the presence of impaired calcium regulation in the muscle.

Fig. 2.

Force-frequency curve in fast- and slow-twitch skeletal muscle. Specific force-frequency relationship (A) and absolute force-frequency relationship (B) in the extensor digitorum longus (EDL) muscle. C: time to peak (TTP) and half-relaxation time (HRT) of muscle twitches in the EDL. Specific force-frequency relationship (D) and absolute force-frequency relationship (E) in the soleus (SOL) muscle. F: TTP and HRT of muscle twitches in the SOL muscle; n = 5–7 per group. The data are presented as the means ± SE. *P < 0.05 vs. 4-mo low-fat diet (LFD); †P < 0.05 vs. 4-mo high-fat diet (HFD) and 20-mo LFD.

Effect of HFD on [Ca²⁺] release in aged mice.

Next, we measured Ca²⁺ flux in intact FDB muscles during electrically stimulated contraction and pharmacological stimulation. Contraction-stimulated Ca²⁺ peak levels in the 4-mo HFD, 20-mo LFD, and 20-mo HFD groups were lower compared with the 4-mo LFD group (Fig. 3, A and B). In particular, a substantial reduction of Ca²⁺ peak levels was observed during tetanic stimulation in the 20-mo HFD group compared with the 4-mo LFD and HFD groups. For the pharmacological experiments, we first used a direct activator of RyR, 4-CmC (36). The Ca²⁺ peak levels stimulated by 4-CmC were reduced in the 20-mo LFD and HFD groups compared with the 4-mo LFD group (Fig. 3, C and D). In addition, 4-CmC significantly reduced the Ca²⁺ peak levels in the 20-mo HFD group compared with the 4-mo HFD group. Of note, the Ca²⁺ peak levels following 4-CmC administration were dramatically reduced in all groups by the presence of MgCl₂, which suppresses RyR-dependent Ca²⁺ release (28) (Fig. 3, C and D). Intriguingly, with a high concentration of caffeine (25 mM), which promoted greater cytosolic calcium, Ca²⁺ peak levels only in the 20-mo HFD group were significantly decreased compared with other groups (Fig. 3, E and F). Next, we investigated intracellular Ca²⁺ levels following the administration of ATP and caffeine, which are known to stimulate the opening and activation of the RyR1 receptor (11). We found that 10 mM caffeine, with or without ATP stimulation, reduced Ca²⁺ peak levels in the 4-mo HFD, 20-mo LFD, and 20-mo HFD groups compared with the 4-mo LFD group (Fig. 3G). Together, these data suggest that both aging and HFD cause skeletal muscle calcium dysregulation. Whereas the degree of the impairments in each group was variable, the most severe impairment for the calcium regulation consistently occurred in the 20-mo HFD group.

Fig. 3.

Ca²⁺ flux during electrically stimulated contraction and pharmacological ryanodine receptor (RyR) stimulation. A: representative twitch (1 Hz) and tetanus (100 Hz) stimulation. B: average peak Ca²⁺ responses (Fluo-4 fluorescence) during twitch and tetanus stimulation. C: representative 1 mM 4-chloro-m-cresol (4-CmC) stimulation, with or without 30 mM MgCl₂. D: average peak Ca²⁺ responses (Fura-2 fluorescence) during 4-CmC, with or without 30 mM MgCl₂ stimulation. E: representative 10 mM or 25 mM caffeine stimulation. F: average peak Ca²⁺ responses (Fura-2 fluorescence) during 10 mM and 25 mM caffeine. G: average peak Ca²⁺ responses (Fura-2 fluorescence) during 10 mM caffeine, with or without 2 mM adenosine triphosphate (ATP). The peak fluorescence intensity (F or R, ratiometric images) during the stimulation was measured, and the quantified data are shown in a graph as changes (ΔF or ΔR) from the prestimulation level (F0 or R0). Intracellular Ca²⁺ concentration ([Ca²⁺]_i) is expressed as the Δ ratio (ΔF/F0 or ΔR/R0); n = 7–20 fibers and 3–4 mice per group. The data are presented as the means ± SE. *P < 0.05 vs. 4-mo low-fat diet (LFD); #P < 0.05 vs. 4-mo high-fat diet (HFD); †P < 0.05 vs. 4-mo HFD and 20-mo LFD.

Effect of HFD on calcium-related proteins in aged mice.

To identify the molecule mechanisms responsible for the impaired Ca²⁺ regulation by aging and HFD, we investigated the protein abundance of these enzymes. As shown in Fig. 4, the expressions of RyR, DHPR, FKBP, SERCA1, PV, and CSQ proteins were comparable among the groups. Instead, the SAR protein levels in the 20-mo LFD and 20-mo HFD groups were decreased compared with those in the 4-mo LFD group. These data indicate that skeletal muscle SAR protein is reduced with aging but not with HFD feeding.

Fig. 4.

Abundance of calcium-regulated proteins. Representative Western blots are shown of ryanodine receptor (RyR), dihydropyridine receptor (DHPR), FK506 binding proteins (FKBP), sarcoplasmic reticulum calcium ATPase type 1 (SERCA1), parvalbumin (PV), calsequestrin (CSQ), sarcalumenin (SAR), and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) expression in extensor digitorum longus muscle. HFD, high-fat diet. The data are presented as the means ± SE; n = 5–7 per group. *P < 0.05 vs. 4-mo low-fat diet (LFD).

Effect of HFD on lipid accumulation in aged mice.

Intramyocellular lipid (IMCL) accumulation is associated with muscle weakness (4, 9, 13). Thus we assessed IMCL accumulation in EDL muscle in each group by histochemical analyses. The quantification of lipid accumulation by LipiDye intensity showed that lipid droplet volume was increased in the 4-mo HFD and 20-mo LFD groups compared with that in the 4-mo LFD group (Fig. 5, A and B). Furthermore, this effect was additive in the 20-mo HFD group (Fig. 5B). We also assessed the IMCL droplet size with electron microscopy. Similar to the LipiDye results, IMCL droplet size was also greater in the 4-mo HFD and 20-mo LFD groups compared with that in the 4-mo LFD group (Fig. 5, C–E), and this effect was additive in the 20-mo HFD group (Fig. 5E).

Fig. 5.

Quantification of intramyocellular lipid by histocytochemical analysis and electron microscopy analyses. A: transverse sections of the extensor digitorum longus (EDL) muscle stained with LipiDye. Scale bar, 10 μm. B: quantification of the fluorescence intensity per individual fiber (n = 30–39 fibers from 4 to 6 mice). Representative transmission electron microscope images at low (C; scale bar, 1 µm) and high (D; scale bar, 500 nm) magnification from EDL muscles. E: quantification of lipid droplet size to determine the intermyofibrillar area of EDL muscle samples (samples comprise 60–114 pictures/10 fibers in 3 animals). The data are presented as the means ± SE. *P < 0.05 vs. 4-mo low-fat diet (LFD); †P < 0.05 vs. 4-mo high-fat diet (HFD) and 20-mo LFD.

Next, with the use of the mean values in each group, we investigated the relationship between the IMCL droplet size and the specific forces during twitch stimulation and tetanic stimulation, as well as Ca²⁺ peak levels and TTP during twitch stimulation. As shown in Fig. 6, increased IMCL droplet size by aging and HFD seems to be associated with lowered specific forces (Fig. 6, A and B), calcium peak levels (Fig. 6C), and delayed TTP (Fig. 6D), respectively. These data suggest that aging and HFD additively increased IMCL level, and these IMCL levels were associated with the degree of impaired muscle contractile force in each group.

Fig. 6.

Relationship between Ca²⁺ release capacity and lipid droplet size. Relationship between twitch force (A) and tetanic force (B) in the extensor digitorum longus (EDL) muscle. C: average peak Ca²⁺ responses (Fluo-4 fluorescence) during twitch forces. D: time to peak (TTP) and lipid morphology size. Assessed by transmission electron microscopy (lipid droplet size). ΔF/F0, ratio of change in peak fluorescence intensity/prestimulation level; HFD, high-fat diet; LFD, low-fat diet.

DISCUSSION

The present study found that skeletal muscle-force production was decreased with aging and HFD feeding and that this effect was additive in the aged HFD-fed mice. Consistent with these results, electrically stimulated Ca²⁺ peak levels were also reduced by aging and HFD feeding, with additive effects in the aged HFD-fed mice. Experiments with a high concentration of caffeine also support this notion. Furthermore, increased IMCL droplet size was associated with impaired muscle contractile function or lowered Ca²⁺ release capacity. Finally, SAR, a calcium-binding protein in the SR that contributes to SR Ca²⁺ content, was decreased in the skeletal muscle of aged mice.

Many studies have demonstrated skeletal muscle contractile dysfunction in mice with HFD-induced obesity (1, 10, 13, 44) and in aged mice (3, 6, 30). Regarding muscle strength in aged mice with HFD-induced obesity, a previous study demonstrated that 14-mo-old mice (middle aged) that consumed a HFD for 5 mo exhibited decreased skeletal muscle strength compared with control mice (29). Similarly, Abrigo et al. (1) showed that the specific force in isolated fast-twitch skeletal muscle was lower in 50-wk-old mice that had consumed a HFD for 30 wk. In terms of the fiber type that is affected by aging and HFD, recent studies, including ours, have shown that only fast-twitch muscle, not slow-twitch muscle, exhibits a reduction in specific force in mice with HFD-induced obesity (10, 13, 19, 44). Consistent with these data, in this study, we demonstrated that force production in fast-twitch skeletal muscle was impaired by aging and HFD-induced obesity, whereas slow-twitch skeletal muscle was barely affected.

Consistent with earlier reports (3, 39), we found reduced peak Ca²⁺ levels during muscle contraction in aged mice (20-mo LFD group) and young mice with HFD-induced obesity (4-mo HFD group). Since Marks and colleagues (11) demonstrated that caffeine and ATP augment RyR1 opening, we investigated peak Ca²⁺ levels using RyR agonists (4-CmC, caffeine, and caffeine/ATP) in muscle samples from HFD mice, aged mice, and aged HFD mice compared with young LFD mice. Interestingly, only 20-mo HFD mice showed reduced Ca²⁺ peak levels following stimulation with high-concentration caffeine (Fig. 3F). A previous study demonstrated that the amount of releasable Ca²⁺ content from the SR was decreased in the skeletal muscle of elderly humans (27), and we confirmed that relative to the 4-mo LFD group, TTP (which reflects Ca²⁺ release capacity during muscle-twitch stimulation) was delayed only in the 20-mo HFD group (Fig. 2C). Taken together, our data suggest that aging and HFD-induced obesity additively impair intracellular Ca²⁺ release in muscle.

Transient increases of intracellular Ca²⁺ in myocytes are initiated by sarcolemmal and transverse tubule depolarization, triggering Ca²⁺ release from the SR via RyR. Ca²⁺ release from RyR is activated through charge-dependent conformational changes in the DHPR (43). CSQ is located in the SR lumen and modulates the sensitivity of RyR (47). FKBP12 is involved in both orthograde and retrograde coupling between the L-type Ca²⁺ channel and RyR1, affecting SR Ca²⁺ release (3, 46). SAR is a Ca²⁺-binding protein localized in the SR of the intracellular Ca²⁺ store (35, 49, 50). Previous studies have shown that the expression levels of these Ca²⁺ release proteins in skeletal muscle are similar in aged and young mice (3, 37). In contrast, SAR is reduced in aged rodent muscle (25, 35) and muscle in elderly women (20). Among the proteins that we investigated, we only found a consistent reduction in SAR proteins in the skeletal muscles of aged LFD-fed and HFD-fed mice (Fig. 4). We recently demonstrated that 6 wk of exercise training improved contractile dysfunction and the failure of Ca²⁺ release and increased protein expression of SAR in the muscle in type 2 diabetic mice (14). Intriguingly, a recent study showed that 9 wk of training by electrical stimulation increased SAR in old muscles (32). In addition, a previous study suggested that SAR is involved in exercise ability after endurance exercise training (23). It is expected that exercise intervention may increase SAR and improve calcium handling and contractile dysfunction in the muscle. However, previous studies demonstrated that muscle contractile force was not reduced in SAR-knockout mice (49). Thus it is still unclear whether decreased SAR protein levels contribute to the reduced muscle contraction caused by aging and HFD, and further investigations are required to clarify the mechanisms.

IMCL accumulates in obese (24) and elderly people (38) and is associated not only with insulin resistance (24, 38) but also with decreased muscle contractile force (9). We (13) and others (4) also demonstrated that a HFD increased IMCL and impaired muscle contractile force in an animal model. In the present study, we also identified a relationship between increased IMCL and Ca²⁺ release capacity, the latter expressed as the specific force and TTP during twitch stimulation in skeletal muscle (Fig. 6). Previous studies demonstrated that higher IMCL levels were associated with decreased muscle-specific force and impaired physical function in obese older adults (9). In skeletal muscle, muscle contraction and relaxation speeds are critically dependent on effective SR Ca²⁺ handling (2). A previous study demonstrated that IMCL was increased, and TTP was delayed in skeletal muscle in a streptozocin-induced type 1 diabetic mouse model (26). Here, we showed that IMCL was correlated with impairment of muscle contractile force by aging and HFD, and therefore, the IMCL level may serve as a good marker for impaired muscle contractility in this setting. Although in this study, we only measured the amount of IMCL, several studies suggested that altered IMCL composition may cause changes in skeletal muscle contractile function via calcium kinetics. One previous study showed that sphingosine blocked calcium release from the RyR and reduced the activity of RyR channels reconstituted into planar lipid bilayers (40); indeed, sphingolipids, such as ceramide, sphingosine, and sphingosine-1-phosphate, promote skeletal muscle weakness (34). In addition, studies using fatty acid synthase knockout mice (17) or choline/ethanolamine phosphotransferase 1 knockout mice (16) showed that decreased intramyocellular phosphatidylethanolamine may impair SERCA activity and cause muscle weakness. Thus changes in IMCL levels caused by aging and a HFD may be accompanied by alterations in IMCL species, which in turn, may impair Ca²⁺ release capacity in skeletal muscle.

Although our study demonstrated that HFD-induced obesity results in reduced contractile function in aged skeletal muscle, Hill et al. (21) showed that the short term (9 wk) of high fat diet-induced obesity did not cause muscle contractile dysfunction in aged (79-wk) female mice. On the other hand, Hurst et al. (22) suggested that 8-wk HFD, but not 4-wk HFD, reduces muscle contractile function at 20 wk in female mice. In addition, contractile properties in skeletal muscle in female mice are different from those in male mice (8). Thus the effect of HFD on contractile function might be influenced by sex, age, and duration of HFD. Since we provided the HFD to male mice for 20 mo, one should be careful when generalizing the results in the present study.

In conclusion, our study demonstrates that HFD-induced obesity results in significantly reduced contractile function and Ca²⁺ release capacity in aged skeletal muscle. These changes may be driven by a diet- and age-induced lipidomic milieu, which interferes with SR Ca²⁺ cycling, in skeletal muscle. The data confirm the idea that diet plays an important role in the maintenance of skeletal muscle contractile property and demonstrate for the first time that these diet-dependent effects are equally relevant with aging. The synergistic effects of obesity and aging on muscle function may exacerbate morbidity and mortality. Further studies will investigate the potential role of the lipid stress signal that may interfere with muscle contractile function with diet and aging.

GRANTS

This work was supported, in part, by the Strategic Research Foundation at Private Universities and KAKENHI (Grants 16J11193 and 15H06597) from the Ministry of Education, Culture, Sports, Science and Technology of Japan and Suzuken Memorial Foundation.

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

H.E. conceived and designed research; H.E., R. Kakigi, and R.H. performed experiments; H.E. analyzed data; H.E., Y.T., S.K., R. Kakigi, R.H., K.F., R. Kawamori, and H.W. interpreted results of experiments; H.E. prepared figures; H.E. and Y.T. drafted manuscript; H.E., Y.T., S.K., R. Kakigi, R.H., K.F., R. Kawamori, and H.W. edited and revised manuscript; H.E., Y.T., S.K., R. Kakigi, R.H., K.F., R. Kawamori, and H.W. approved final version of manuscript.