Altered skeletal muscle sarco-endoplasmic reticulum Ca²⁺-ATPase calcium transport efficiency after a thermogenic stimulus

Abstract

Exposure to predator threat induces a rapid and robust increase in skeletal muscle thermogenesis in rats. The central nervous system relays threat information to skeletal muscle through activation of the sympathetic nervous system, but muscle mechanisms mediating this thermogenesis remain unidentified. Given the relevance of sarcolipin-mediated futile calcium cycling through the sarco-endoplasmic reticulum Ca²⁺-ATPase (SERCA) pump to mammalian muscle nonshivering thermogenesis, we hypothesized that this plays a role in contextually induced muscle thermogenesis as well. This was assessed by measuring enzymatic activity of SERCA and sarcoplasmic reticulum Ca²⁺ transport, where the apparent coupling ratio (Ca²⁺ uptake rate divided by ATPase activity rate at a standard Ca²⁺ concentration) was predicted to decrease in association with muscle thermogenesis. Sprague–Dawley rats exposed to predator (ferret) odor (PO) showed a rapid decrease in the apparent coupling ratio in the soleus muscle, indicating SERCA uncoupling compared with control-odor-exposed rats. A rat model of high aerobic fitness and elevated muscle thermogenesis also demonstrated soleus muscle SERCA uncoupling relative to their obesity-prone, low-fitness counterparts. Both the high- and low-aerobic fitness rats showed soleus SERCA uncoupling with exposure to PO. Finally, no increase in sarcolipin expression in soleus muscle was detected with PO exposure. This dataset implicates muscle uncoupling of SERCA Ca²⁺ transport and ATP hydrolysis, likely through altered SERCA or sarcolipin function outside of translational regulation, as one contributor to the muscle thermogenesis provoked by exposure to predator threat. These data support the involvement of SERCA uncoupling in both muscle thermogenic induction and enhanced aerobic capacity.

INTRODUCTION

The widespread increase in obesity and associated health sequelae and the difficulty in maintaining weight loss (1, 2) has sharpened interest in approaches to increase energy expenditure. Although the ability of adipose tissue to generate heat has been investigated extensively in preclinical models as well as in humans (3, 4), other sources of thermogenesis have the potential to contribute to energy balance as well (5, 6). Skeletal muscle has a substantial thermogenic capacity that can be exploited to meaningfully impact energy balance in humans (5–7). Muscle is the major determinant of basal metabolic rate as muscle comprises 40% of mass in nonobese humans (8) and consumes nearly 80% of insulin-stimulated glucose uptake (9). We have demonstrated that rats and mice display robust muscle thermogenesis as quickly as 2–15 min after exposure to predator threat (10). In rats, this is associated with a marked increase in caloric expenditure, which can be harnessed to augment weight loss (10).

The central nervous system conveys information regarding predator threats to muscles via the sympathetic nervous system (SNS) (10). At the level of the skeletal myocyte, multiple mechanisms have been proposed to mediate thermogenic processes, but the strongest evidence for the generation of muscle thermogenesis points to altered sarco-endoplasmic reticulum Ca²⁺-ATPase (SERCA) pump efficiency (11–14). After the muscle fiber contraction is induced by Ca²⁺ binding to troponin (15), the SERCA pump relieves the contraction by hydrolyzing ATP to power the transport of Ca²⁺ out of the cytosol and into the sarcoplasmic reticulum (16, 17). Small molecule SERCA inhibitors like sarcolipin can uncouple this process, wherein ATP is hydrolyzed but Ca²⁺ fails to enter the sarcoplasmic reticulum due to “slippage,” generating heat (18–23). Besides pumping during contraction to make relaxation possible, SERCA continues to take up Ca²⁺ into the sarcoendoplasmic reticulum during muscle relaxation, accounting for 30% of resting metabolic rate of muscle in mice (24). Inefficient Ca²⁺ transport by sarcolipin-inhibited SERCA has been proposed to play a predominant role in mammalian muscle thermogenesis, including adaptation to a cold environment (25) and resistance to weight gain on a high-fat diet (26).

Here, we examine the potential contribution of altered SERCA Ca²⁺ transport efficiency (17, 27–29) to the acute thermogenesis provoked by predator threat by measuring apparent uncoupling ratio in the presence or absence of predator threat. Next, we investigate how SERCA uncoupling may be related to intrinsic aerobic fitness by comparing muscles of contrasting rat models of high and low aerobic fitness (HCR/LCR) (30, 31) that differ in their muscle thermogenic responses to both activity and predator threat (32–34). These divergent rat models of high and low intrinsic aerobic capacity were derived through artificial selection based on treadmill running endurance (30, 31, 35). The high-fitness HCR are consistently leaner and more active than their low-fitness counterparts (33, 36, 37) and also show enhanced muscle heat generation during activity and exposure to predator odor (33, 34). Probing the mechanistic foundations of muscle thermogenesis in these models may reveal how increased aerobic capacity and metabolic complexity are intertwined with thermogenesis and the evolution of endothermy (38–40). Finally, we determine if the rapid thermogenic response to predator threat and intrinsic phenotypic differences in SERCA transport efficiency are associated with elevated expression of sarcolipin.

MATERIALS AND METHODS

Animals and Experimental Design

All experimental procedures proposed were approved by the Institutional Animal Care and Use Committee (IACUC) of Kent State University and performed in accordance with the National Institutes of Health Guide for the Care and Use and Laboratory Animals. Rats were housed in environmentally controlled rooms with temperatures at 24°C ± 1°C. Lights were on at 7:00 AM and off at 7:00 PM for standard 12:12-h Eastern Standard Time light/dark cycle. Rats had free access to water and standard laboratory chow (5P00 Prolab RMH 3000, LabDiet, St. Louis, MO). All rats were acclimated to the study environment for at least 2 wk. Male Sprague–Dawley rats (n = 16) were obtained from a commercial vendor (Envigo), and female HCR/LCR rats (N = 36, all from generation 41) were obtained from the University of Toledo (31, 41). For HCR and LCR compared for baseline SERCA uncoupling, at the time of phenotyping (at ∼2–3 mo of age), which consisted of five graded treadmill trials (30), body weight did not significantly differ (all unpaired two-tailed t tests, means ± SE: HCR, 176 ± 8 g; LCR, 194 ± 3 g; P = 0.064) but HCR had significantly longer maximal time running on the treadmill (HCR, 74.5 ± 2.7 min; LCR, 16.8 ± 1.5 min), maximal distance (HCR, 2,104 ± 125 m; LCR, 234 ± 28 m), highest speed attained (HCR, 46.7 ± 1.4 m/min; LCR, 18.0 ± 0.7 m/min), and vertical work performed (HCR, 933 ± 47 J; LCR, 114 ± 12 J). For the phenotyping of HCR and LCR exposed to predator or control odor, LCR was significantly heavier (HCR, 167 ± 6 g; LCR, 190 ± 4 g), and HCR had significantly longer maximal running time (HCR, 80.9 ± 1.7 min; LCR, 14.6 ± 0.7 min), maximal distance run (HCR, 2,414 ± 83 m; LCR 193 ± 12 m), maximal speed (50.0 ± 0.8 m/min; LCR, 16.8 ± 0.3 m/min), and vertical work performed (HCR, 1,026 ± 51 J; LCR, 93 ± 5 J).

First, to determine if an acute thermogenic stimulus (exposure to predator threat) decreases SERCA pumping efficiency in muscle, we exposed male Sprague–Dawley rats to either predator odor or control odor (n = 7/exposure group) for 20 min. After predator-odor or control-odor exposure, rats were euthanized, and the soleus muscle was rapidly dissected for later measurement of SERCA enzymatic activity and Ca²⁺ transport. Second, to assess baseline SERCA transport efficiency, unstimulated female HCR and LCR rats (n = 6/phenotype) were euthanized before muscle dissection. Third, to probe the role of altered Ca²⁺ transport efficiency in the differential thermogenic response of HCR and LCR to predator threat (34), female HCR and LCR were exposed to predator odor or control odor (n = 6/phenotype/exposure) for 20 min before isolation of muscle. Finally, in each of these cases, we examined the ability of predator odor exposure to acutely change sarcolipin expression using qPCR to measure sarcolipin mRNA. All analyses utilized soleus muscle, a highly oxidative muscle with abundant expression of the thermogenic SERCA uncoupler sarcolipin (42).

Odor Presentation and Tissue Dissection

Towels that were housed with ferrets (Mustela putorius furo) for 2 wk were shipped frozen, with control towels provided by the same supplier (Marshall BioResources). Towels were cut into 2” × 2” aliquots and refrozen until the day they were used. All animals were habituated to handling and to control (unscented) towel aliquots that were placed into home cages daily for 2 wk before experimental odor exposure by the same investigator (43) presenting the stimulus before muscle dissection. We have previously demonstrated that 3–4 days of habituation suppresses the thermogenic response to the control towel (10).

Since the rats’ muscle temperature begins to increase essentially immediately after predator odor exposure begins (10), timing of odor presentation was staggered. The odor stimulus (i.e., towel) was placed into a rat’s cage by briefly lifting and replacing the microisolator lid. After a total of 20 min of exposure, rats were anesthetized with 5% isoflurane through inhalation and rapidly decapitated. Soleus muscle was rapidly dissected, homogenized in buffer, aliquoted into microtubes, and snap-frozen in liquid nitrogen within 10 min of decapitation to preserve Ca²⁺ uptake and SERCA ATPase activity (44). As described by Duhamel et al., in vivo transport efficiency is represented in homogenate samples isolated rapidly at a controlled temperature (45). Although the intracellular conditions in which sarcoplasmic reticulum Ca²⁺ transport and ATPase occur are not maintainable indefinitely in vitro, measurement is possible in freshly thawed samples that were rapidly collected and processed to temporarily preserve in vivo conditions. A maximum of 12 rats were exposed per day within approximately a 3-h timespan, at the midpoint of the light phase when the rats normally rest, as described previously (10).

SERCA Transport Efficiency

The Indo-1 (Biotium, Fremont, CA) assay uses the difference in fluorescence emission between Ca²⁺-bound Indo-1 and Ca²⁺-free Indo-1 to measure change in Ca²⁺ concentration. As Ca²⁺ is taken up into the sarcoplasmic reticulum, the concentration of free Ca²⁺ in the cytosol decreases, causing Ca²⁺ to unbind from Indo-1, increasing the ratio of Ca²⁺-free Indo-1 to Ca²⁺-bound Indo-1 and changing fluorescence emission wavelength, detected using a spectrofluorometer with dual-emission monochromators (Biotek Cytation 5). Ca²⁺ uptake was measured in soleus homogenates at a starting free Ca²⁺ concentration of 1,500 nM. Ca²⁺ uptake rate is calculated in units of micromoles of Ca²⁺ ions per gram of protein in homogenate per minute (μmol of Ca²⁺/g of protein/min).

Oxalate and phosphate were used to increase rates of Ca²⁺ uptake because they precipitate Ca²⁺ inside the sarcoplasmic reticulum, preventing the increasing Ca²⁺ ion levels inside the lumen from inhibiting further Ca²⁺ uptake from the cytosol (46). Ca²⁺ ionophore A23187 (Sigma C7522) was also used to prevent buildup of an inhibitory ionic gradient across the sarcoplasmic reticulum membrane, making the membrane permeable and eliminating variability secondary to Ca²⁺ leak or release. Maximal SERCA activity was measured directly and SERCA activity-pCa curves were generated using GraphPad Prism by nonlinear regression curve fitting, using the equation for the general cooperative model for substrate activation. pCa is Ca²⁺ concentration, represented using –log10. ATPase activity rate was measured as the rate of oxidation of reduced nicotinamide adenine dinucleotide + hydrogen (NADH) to NAD⁺ in a 30-min kinetic assay of absorbance, as decreasing NADH causes decreased absorbance at 340 nm. Units of ATPase activity rate are expressed as micromoles of Ca²⁺ ions per gram of protein in homogenate per minute (μmol of Ca²⁺/g of protein/min) (47).

As previously validated and described elsewhere (44, 48, 49), ATPase activity and Ca²⁺ uptake were measured at the same concentration of free Ca²⁺ in the cytosol, [Ca²⁺]_f = 1,000 nM, normalized for protein concentration, and used to calculate apparent coupling ratio (ACR) with the following equation:

$$\text{ACR}=\frac{\text{Uptake\hspace{0.17em}rate\hspace{0.17em}}\left({\text{μmol\hspace{0.17em}of\hspace{0.17em}Ca}}^{2+}/\text{g\hspace{0.17em}of\hspace{0.17em}protein}/\text{min}\right)}{\text{ATPase\hspace{0.17em}activity\hspace{0.17em}}({\text{μmol\hspace{0.17em}of\hspace{0.17em}Ca}}^{2+}/\text{g\hspace{0.17em}of\hspace{0.17em}protein}/\text{min})}$$

Using the same Ca²⁺ concentration for both uptake and ATPase activity assays is necessary due to the changes in concentration in the cytosol during both ATPase activity and Ca²⁺ uptake. Ca²⁺ uptake and ATPase activity are both measured using kinetic assays and over a range of concentrations to accurately capture their rates at the exact Ca²⁺ concentration of interest (29, 50, 51).

Quantitative Polymerase Chain Reaction for Analysis of Muscle mRNA

Muscle (∼0.05 g soleus) was homogenized by Bullet Blender 24 Gold (Next Advance), and mRNA was isolated using TRIzol Reagent (Thermo Fisher Scientific); bromochloropropane (BCP) was added and incubated for 5 min at room temperature to facilitate phase separation, then samples were centrifuged at 12,000 RCF for 10 min at 4°C. The aqueous RNA phase (∼250 μL) was pipetted into a new microtube and mixed with 100% ethanol (150 μL) to precipitate mRNA. The mixture was then pipetted on Ribopure (Invitrogen) silica-based column membrane, and samples were centrifuged at 12,000 RCF for 30 s at room temperature. Samples were washed according to kit instructions, and RNA was eluted from the filter with 100-μL kit elution buffer by centrifuging at 12,000 RCF for 30 s at room temperature. The purity and quantity of mRNA were measured using NanoDrop spectrophotometer (Thermo Fisher Scientific), resulting in A260/280 ratios ranging from 1.8 to 2.1 with 50–110 ng/μL mRNA concentration. Approximately 250 ng of isolated RNA was reverse transcribed using a TaqMan Reverse Transcription Kit (Invitrogen). The target cDNA was amplified by PCR at 37°C for 60 min, and 95°C for 5 min. All qPCR assays used the Brilliant III Ultra-Fast QPCR Master Mix (Agilent Technologies) and PrimeTime Gene Expression Probes (IDT DNA Technologies). The assay identification numbers for each gene are as follows: Gapdh (Rn.PT.39a.11180736.g) and Sln (Rn.PT.58.8785990). Relative expression for each gene was calculated using the housekeeping gene Gapdh as a reference and the 2^−ΔCt method (52) as described previously (53).

Statistical Analyses

Statistics were performed in IBM SPSS 24/26. Data are means ± SE. Apparent coupling ratio (ACR) data were not normally distributed. To ameliorate this violation of assumptions for t tests and analysis of variance (ANOVA), apparent coupling ratio percentages for all studies were transformed using arcsine transformation (also called arcsine square root transformation, or the angular transformation), calculated as 2 × the arcsine of the square root of the proportion = arcsine transformed apparent coupling ratio (AACR, a unitless ratio) (54).

$$\text{AACR}=2\times \text{\hspace{0.17em}arcsine}\sqrt{\frac{\text{Uptake\hspace{0.17em}rate\hspace{0.17em}}\left({\text{μmol\hspace{0.17em}of\hspace{0.17em}Ca}}^{2+}/\text{g\hspace{0.17em}of\hspace{0.17em}protein}/\text{min}\right)}{\text{ATPase\hspace{0.17em}activity\hspace{0.17em}}({\text{μmol\hspace{0.17em}of\hspace{0.17em}Ca}}^{2+}/\text{g\hspace{0.17em}of\hspace{0.17em}protein}/\text{min})}}$$

Outliers were identified by inspection of boxplots (i.e., outside the bounds of median ± 1.5 × interquartile range). For the Sprague-Dawley rats, one outlier was identified outside these bounds. Since this odor-exposed arcsine transformed ACR was a high outlier, including it in the analysis would bias the results in the opposite direction we predicted, and no specific cause for excluding it was identified. Therefore, this outlier was included in the analysis. There was heterogeneity of variances for control and predator odor-exposed Sprague-Dawley rats’ AACR, as assessed by Levene’s test for equality of variances (P = 0.003). Control and predator odor-exposed AACR were compared using a Welch’s t test.

For the unstimulated high- and low-capacity runner rats, there were no outliers by inspection of AACR boxplots (i.e., beyond the bounds of median ± 1.5 × interquartile range). AACR was normally distributed, as assessed by Shapiro– Wilk’s test (P > 0.05). Levene’s test for equality of variance (P = 0.026) contravened the assumption of variance homogeneity, so Welch’s t test was used to determine if there were differences in AACR between low- and high-capacity runner rats.

To compare Ca²⁺ transport efficiency in HCR and LCR exposed to the predator or control odors, a two-way ANOVA was conducted to examine the effects of line (HCR or LCR) and exposure (to control odor or predator odor) on arcsine transformed ACR. Residual analysis was performed to test for the assumptions of the two-way ANOVA, and an analysis of simple main effects was performed with Bonferroni adjustment causing significance to be accepted at the P < 0.025 level. All pairwise comparisons were examined for each simple main effect with reported 95% confidence intervals in parentheses and Bonferroni-adjusted significance within each simple main effect P < 0.025. One outlier was identified by inspection of boxplots (i.e., outside of median ± 1.5 × interquartile range). This control-exposed HCR outlier had a low AACR near zero. Briefly, it was concluded that this data point was erroneous due to the known effect of too-low ACR resulting from a delay during any of the time-sensitive steps in the uptake or activity assays. To test the validity of this decision, ANOVA was performed both including and excluding this outlier; similar outcomes were obtained in both cases, and the ANOVA excluding the outlier is reported below. Normality was assessed using Shapiro–Wilk’s test for each cell of the design; residuals were normally distributed (P > 0.05). Homogeneity of variances was assessed by Levene’s test; based on median and with adjusted degrees of freedom, variances were homogenous (P = 0.064).

RESULTS

Predator Odor Exposure Increased Muscle SERCA Uncoupling

Soleus AACR was significantly lower in rats exposed to predator odor compared with the control stimulus (P < 0.05; Fig. 1). This reflects SERCA uncoupling and is consistent with elevated skeletal muscle thermogenesis in response to PO exposure (10).

Figure 1.

Exposure to predator odor significantly decreased SERCA Ca²⁺ transport efficiency in rat soleus muscle. Male Sprague-Dawley rats (n = 7 rats/group) were exposed to either ferret odor or a control stimulus for 20 min before measuring apparent uncoupling (Ca²⁺ uptake rate-to-SERCA ATPase activity). *Significantly less efficient than control-exposed rats, Welch’s t test of arcsine-square root normalized data, P < 0.05, means ± SE. SERCA, sarco-endoplasmic reticulum Ca²⁺-ATPase.

High-Fitness Rats Show Relatively More Muscle SERCA Uncoupling Relative to Low-Fitness Rats

The t test revealed that soleus from HCR showed significantly lower AACR compared with LCR (P < 0.001; Fig. 2). This reflects higher SERCA efficiency in LCR than in the high-fitness HCR, consistent with the muscle thermogenic capacity seen in these models (32, 33).

Figure 2.

Low-capacity runners (LCR) had significantly higher soleus muscle SERCA Ca²⁺ transport efficiency compared with high-capacity runners (HCR). Soleus muscles from adult female rats artificially selected for high (HCR) and low (LCR) intrinsic aerobic fitness were measured for apparent uncoupling (Ca²⁺ uptake rate-to-SERCA ATPase activity) without any experimental manipulation. * Significantly less efficient than LCR (n = 6 rats/phenotype), Welch’s t test of arcsine-square root normalized data, P < 0.001, means ±SE. SERCA, sarco-endoplasmic reticulum Ca²⁺-ATPase.

Rats with High and Low Aerobic Fitness Both Increased Muscle SERCA Uncoupling after Exposure to Predator Odor

As shown in Fig. 3, the 2 × 2 ANOVA revealed a significant interaction between selected line (HCR or LCR) and predator-odor exposure on AACR (P = 0.019), as well as main effects of selected line (HCR vs. LCR) and predator odor. An analysis of simple main effects was performed with Bonferroni adjustment causing significance to be accepted at the P < 0.025 level. There was a significant difference where AACR was lower in HCR than LCR rats exposed to the control odor (P < 0.001; partial η² = 0.810) consistent with the previous experiment. There was also a significant difference where AACR remained lower in HCR than LCR rats exposed to predator odor (P < 0.001; partial η² = 0.650). There was a significant difference where AACR was lower in control LCR than in predator odor-exposed LCR (P < 0.001; partial η² = 0.632); soleus AACR was not significantly different between control and predator-odor exposure for HCR, however.

Figure 3.

Exposure to predator odor (PO) significantly decreased soleus muscle SERCA Ca²⁺ transport efficiency in low-capacity runners (LCR) but not high-capacity runners (HCR). Adult female HCR and LCR were exposed to either ferret odor or a control stimulus (n = 6 rats/group, total N = 24) for 20 min before measuring soleus apparent coupling ratio (Ca²⁺ uptake rate-to-SERCA ATPase activity). A two-way analysis of variance of the arcsine-square root normalized apparent coupling ratio revealed a main effect where HCR soleus had significantly lower transport efficiency than LCR, a main effect where PO-exposed rats showed significantly lower Ca²⁺ transport efficiency, and a significant interaction (*p = 0.019), where lower SERCA transport efficiency was seen with PO exposure in LCR but not HCR after Bonferroni adjustment (P < 0.025). ^aApparent coupling ratio was significantly lower in control-exposed HCR compared with control-exposed LCR, ^bPO-exposed LCR compared with control-exposed LCR, and ^cPO-exposed HCR compared with PO-exposed LCR (all P < 0.001). Data represented as means ± SE. SERCA, sarco-endoplasmic reticulum Ca²⁺-ATPase.

Sarcolipin mRNA expression was not elevated with SERCA uncoupling. qPCR for sarcolipin mRNA expression showed no significant difference between control odor-exposed Sprague–Dawley rats and PO-exposed Sprague-Dawley rats (P > 0.05; Fig. 4A). qPCR for sarcolipin mRNA expression did not differ between LCR and HCR (P > 0.05; Fig. 4B). Similarly, the two-way ANOVA revealed no interaction of line by exposure and no significant main effects for HCR and LCR exposed to predator odor or control odor (P > 0.05; Fig. 4C).

Figure 4.

Soleus muscle sarcolipin mRNA did not differ with exposure to predator odor (PO) or between high- and low-fitness rats. A: no change in sarcolipin mRNA was detected in PO-exposed Sprague-Dawley rats compared with control odor-exposed rats (n = 7 rats/group; two-tailed t test comparing 2^−dCT). B: no difference in sarcolipin mRNA was detected between soleus from high-capacity runners (HCR, n = 6 rats/group) and low-capacity runners (LCR; n = 6; two-tailed t test). C: a two-way analysis of variance found no significant main effect or interaction when comparing sarcolipin mRNA in HCR and LCR exposed to either PO or control odor (n = 6 rats/group). Data represented as means ± SE.

DISCUSSION

The relevance of muscle thermogenesis to energy balance is increasingly recognized, including in humans (5). Strong evidence points toward uncoupling of SERCA transport as the most relevant generator of heat in muscle, aside from contraction (12, 13, 25, 26, 55). Here, we measured apparent coupling between SERCA enzymatic activity and Ca²⁺ transport to assess the potential of a strong thermogenic stimulator (exposure to predator threat) to alter muscle SERCA Ca²⁺ transport efficiency. Exposing rats to ferret odor for just 20 min resulted in lower muscle SERCA transport efficiency to less than half that of control-exposed rats. This strongly implicates SERCA uncoupling as the predominant mechanism underlying acute muscle thermogenesis provoked by exposure to predator threat (Fig. 1). The ability of predator odor to uncouple SERCA Ca²⁺ transport was rapid, mirroring the increase in muscle temperature (10). Moreover, suppressed baseline muscle SERCA coupling was also observed in rats with high aerobic fitness that show enhanced muscle thermogenic potential (32, 34). Rats with both high and low aerobic fitness displayed muscle SERCA uncoupling after exposure to predator odor, but higher baseline muscle thermogenic capacity did not coincide with greater predator odor-associated SERCA uncoupling (Figs. 2 and 3). Overall, these findings strongly point to muscle SERCA uncoupling in the ability of predator odor to rapidly provoke thermogenesis and also suggest that SERCA transport efficiency may also underlie the link between muscle thermogenesis and aerobic capacity.

The pathway that allows for an environmental stimulus, namely, predator threat, to stimulate muscle thermogenesis likely involves the hypothalamus and the SNS. Our previous data implicate ventromedial hypothalamus (VMH) in the central modulation of muscle thermogenesis (32, 53, 56). This region is also critical to the ability to display a behavioral response to predator threat, including predator odor (57–59). The ability of predator odor to fully evoke downstream muscle thermogenesis in the gastrocnemius muscle group is also reliant on an intact ipsilateral lumbar sympathetic nerve as well as β-adrenergic receptors (10). Here, we implicate SERCA Ca²⁺ transport as the likely principal mediator of this thermogenic response at the level of the myocyte in soleus muscle. Taken together, this evidence suggests a pathway whereby predator threat increases VMH, including steroidogenic factor 1 (SF-1) (57, 59–61), cell activity to promote SNS outflow to muscle. The resulting activation of adrenergic receptors (10, 62) increases skeletal muscle sarcolipin uncoupling of SERCA (26), where SERCA transports fewer calcium ions into the sarcoendoplasmic lumen while freeing energy by continuing to hydrolyze ATP; the unused energy is released as heat (9, 12). This hypothesis is limited by the lack of a direct connection between β-adrenergic receptor activation and sarcolipin interaction with SERCA, in contrast to the direct signaling outlined in cardiomyocytes (63). The β-adrenergic agonist fenoterol increases the maximal activity of SERCA in gastrocnemius muscle (64), and SERCA inefficiency has been proposed as the primary mechanism of skeletal muscle nonshivering thermogenesis induced by cold exposure (40) or a high-calorie diet (25).

In humans and laboratory animals, fearful stimuli induce thermogenesis in a similar manner as exposure to predator threat; this has been alternately referred to as fear- or stress-induced hyperthermia or psychogenic fever (65–69); social stimuli also induce hyperthermia, even in the absence of aggression or defeat (70). Though we have demonstrated that the muscle thermogenic response to ferret odor is distinct from other stressful or aversive stimuli and even other predator odors (10), it is likely that stress-induced hyperthermia may evoke muscle SERCA uncoupling as well. Though brown adipose tissue activation has been identified in stress-induced and social hyperthermia in mice and rats (69–73), the hyperthermia induced through an acute bout of conditioned fear in rats was not reliant on brown adipose thermogenesis (62). These data also do not extend to the potential mechanistic interaction between predator odor-induced thermogenesis and other energetic demands that induce thermogenesis, like cold or exercise (74). Finally, additional mechanisms that produce thermogenesis in muscle could also be contributing to the heat generated during predator-odor exposure (75); in these rats, shivering was not observed but neither was it excluded.

Lower SERCA Ca²⁺ transport efficiency is also seen in soleus of rats with high intrinsic aerobic capacity or fitness (i.e., HCR; Fig. 2). This is consistent with the elevated thermogenic capacity of this phenotype (33, 34). HCR show enhanced thermogenesis during low-to-moderate intensity activity (e.g., walking on a treadmill) compared with LCR (32, 33). We also previously reported higher relative SERCA1 and 2 mRNA in HCR compared with LCR muscles (32). The actions of SERCA uncoupling in muscle could potentially affect not only muscle heat generation but may also contribute to other aspects of the metabolic phenotype of the aerobically fit rats (76). Physical activity in rats with high aerobic capacity is more energetically costly, such that their activity is less efficient during walking (33, 36). Mechanistic overlap may link muscle function and performance to energetic inefficiency. For example, muscle overexpression of the SERCA uncoupler sarcolipin alters muscle contractile properties, with muscles from mice overexpressing skeletal muscle sarcolipin producing higher twitch force in EDL and less fatigue in both EDL and soleus muscles [however, see Tupling et al. (77)]; the sarcolipin-overexpressing mice also exhibited less running fatigue (42). The elevated cytosolic Ca²⁺ transients resulting from sarcolipin-induced SERCA uncoupling also promoted mitochondrial biogenesis and oxidative capacity, connecting muscle performance to another thermogenic process (14). Indeed, cold adaptation, improved performance or endurance, and fuel inefficiency cooccur in multiple species (78–80), suggesting that these traits could be mechanistically intertwined at the level of skeletal muscle.

Predator threat acutely suppressed SERCA Ca²⁺ transport efficiency in the contrasting rat model of high and low aerobic fitness (Fig. 3). The aerobically fit HCR show an enhanced muscle thermogenic response to predator odor (34). This led us to predict that the high-fitness rats would also show a greater change in SERCA Ca²⁺ transport efficiency with exposure to predator odor. Contrary to this prediction, although the LCR showed a significant increase in SERCA uncoupling with predator-odor exposure, the difference in SERCA apparent coupling ratio between predator odor- and control odor-exposed soleus did not reach significance after adjustment for multiple comparisons (Fig. 3). This could be a result of a floor effect where apparent coupling ratio was sufficiently low in the HCR that further uncoupling capacity is minimal, as the SERCA coupling ratios in HCR soleus were not only low but also uniform relative to other conditions (Fig. 3). Also, although the link between SERCA uncoupling and heat generation is established (12, 13), the magnitude of heat generated may not be linear with respect to SERCA uncoupling. Finally, additional mechanisms (75, 81) could be differentially altered in HCR and LCR that could contribute to muscle heat generation with predator threat.

No difference in soleus sarcolipin mRNA was detected in association with SERCA uncoupling, either after exposure to predator odor or in association with the high-fitness phenotype (Fig. 4). The lack of change in sarcolipin transcript is perhaps not surprising given the timing of the predator-odor stimulus; SERCA uncoupling was measured in rats exposed to the odor for just 20 min. Even so, soleus sarcolipin expression did not differ between high- and low-fitness rats under baseline conditions. These findings counter the supposition that activation of this muscle thermogenic process relies on, or even involves, transcriptional activation of sarcolipin in muscle. This does not negate the possibility that sarcolipin is recruited through a mechanism other than transcriptional regulation, however, to alter SERCA function. Similarly, the process of termination of the sarcolipin-SERCA interaction and therefore cessation of heat generation is also unknown. Although sarcolipin can alter SERCA function, and the overexpression of muscle sarcolipin produced meaningful metabolic changes (25, 42), endogenous posttranslational modification of sarcolipin may be key to transducing this thermogenic stimulus by altering sarcolipin’s interaction with SERCA. Downstream effects on cellular metabolism may stem from altered cytosolic Ca²⁺ and Ca²⁺ signaling (74), which are also impacted by membrane Ca²⁺ leak or release, for example through ryanodine receptor 1 (82, 83). Since the temperature increase of nonshivering thermogenesis starts immediately, it is likely that there is a ready pool of sarcolipin in the cytosol available to be quickly modified to bind to SERCA when dephosphorylated by protein kinase A (PKA). PKA is one intracellular messenger that is involved in sarcolipin’s regulation (84), in addition to calcium concentration (85) also affecting sarcolipin-SERCA binding.

Part of the interest in SERCA Ca²⁺ transport efficiency stems from the possibility of exploiting this mechanism to increase muscle energy expenditure in the service of combating weight gain and obesity (5, 14, 26). Overexpression of the SERCA inhibitory peptide sarcolipin induced weight loss, including on a high-fat diet, increasing fatty acid oxidation and energy expenditure as well as muscle glucose uptake (14, 26). Deletion of this small SERCA inhibitor results in weight gain (26) and worsens high-fat diet-induced adipose remodeling and inflammation in mice (86). Here, we show that, even without significantly less sarcolipin, low-fitness rats prone to obesity have more efficient SERCA Ca²⁺ transport (Figs. 2–4). Despite this, muscles from these low-fitness rats more readily alter SERCA Ca²⁺ transport efficiency with a thermogenic stimulus (Fig. 3), suggesting the potential to reverse this aspect of muscle function. Muscle nonshivering thermogenesis may be particularly relevant to thermoregulation in larger mammals that lack significant brown adipose tissue (6, 40, 87). The thermogenic potential of the large mass of muscle in larger mammals including humans could lead to considerable caloric expenditure when activated.

Perspectives and Significance

Here, we establish for the first time that a strong, contextual thermogenic stimulus (exposure to predator threat in the form of predator odor) rapidly reduces SERCA Ca²⁺ transport efficiency. This inefficient Ca²⁺ transport likely underlies most of the muscle heat generation in this context (10). Moreover, this is relevant to the obese phenotype seen in rats with low aerobic fitness, which have relatively high SERCA Ca²⁺ transport efficiency. This may have implications regarding mechanisms connecting enhanced muscle performance (e.g., strength, endurance, fatigue resistance), increased caloric expenditure of muscle (e.g., cost of transport or muscle work efficiency), and heat generation (33, 36, 42, 78–80). For example, cold adaptation increases running endurance (80); conversely, high aerobic fitness is associated with elevated muscle thermogenic capacity and lower fuel economy of activity (33, 34, 36). Overexpression of muscle sarcolipin confers resistance to fatigue (42). Altogether, this leads us to speculate that the ability to induce SERCA uncoupling and futile Ca²⁺ cycling may underlie each of these traits. Thus, it is possible that the evolution of endothermy involved challenges other than adaptation to cold, for example, as a byproduct of energy balance regulation (38) or survival in a predator-rich environment. At the level of the individual, muscle futile calcium cycling is a driver of the ability of the predator-threat context to induce thermogenesis and amplify energy expenditure. This highlights the potential of exploiting this brain-SNS-muscle pathway for thermogenic control to modulate energy balance.

DATA AVAILABILITY

The data that support this study are available upon request from the corresponding author.

GRANTS

This work was supported by grants from the National Institute of Diabetes and Digestive and Kidney Diseases Grant R15DK121246 (to C.M.N.), by the Kent State University Graduate Student Senate Research Award (to L.A.H), and by P40OD021331 (to L.G.K. and S.L.B.).

DISCLAIMERS

The sponsors had no involvement in research design, data collection, analysis, or report writing for submission.

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

L.A.H., and C.M.N. conceived and designed research; L.A.H. performed experiments; L.A.H., C.M.N. analyzed data; L.A.H., L.G.K., S.L.B., and C.M.N. interpreted results of experiments; L.A.H. prepared figures; L.A.H. and C.M.N drafted manuscript; L.A.H., L.G.K., S.L.B., and C.M.N. edited and revised manuscript; L.A.H., L.G.K., S.L.B., and C.M.N. approved final version of manuscript.