*3.5. Relative mRNA and Protein Levels of Key Ca2*<sup>+</sup> *Transport Proteins*/*Channels in Skeletal Muscle Fibers during Di*ff*erent Hibernation Periods*

The mRNA and protein expression levels of several major Ca2<sup>+</sup> transport proteins/channels located in the cytoplasm, SR, and mitochondria, including STIM1, ORAI1, RyR1, LETM1, PMCA3, SERCA1, MCU, MICU1, MICU2, and the free Ca2<sup>+</sup> binding protein CALM, as well as the Pearson correlation coefficients for the co-localization of ORAI1 and STIM1, were detected to explore the potential mechanisms involved in intracellular Ca2<sup>+</sup> fluctuation during hibernation. It should be clarified that, as mRNA expression is very sensitive to both internal and external environmental factors, large variability in the size of the error bars occurred in the mRNA statistical results. Therefore, we set a fold-change of ≥2-fold as the threshold for the biological significance of mRNA expression.

The mRNA and protein expression levels, as well as the Pearson correlation coefficients for the co-localization of ORAI1 and STIM1, were detected to explore the role of the SOCE channel in intracellular Ca2<sup>+</sup> level fluctuation during hibernation. In the PL muscle, compared with that in the SA group, the mRNA expression levels of both STIM1 and ORAI1 increased during IBA (Figure 6A, B). Their protein expression levels showed an increasing (though non-significant) trend (Figure 7B, C). In addition, the Pearson correlation coefficients for the co-localization of ORAI1 and STIM1 were significantly elevated in the IBA group (Figure 8). In the AM muscle, the mRNA and protein expression levels of STIM1 and ORAI1 showed no significant change in the IBA group. However, Pearson's correlation coefficients for the co-localization of ORAI1 and STIM1 exhibited slight increases in the IBA group. Overall, the Pearson's correlation coefficients for the co-localization of ORAI1 and STIM1 in PL and AM muscle increased when ground squirrels aroused from torpor.

**Figure 6.** Changes in the mRNA expression of distinct Ca2<sup>+</sup> transport and binding proteins in PL and AM muscles during different periods. Histograms depicting (**A**) stromal interaction molecule-1 (*STIM1*) mRNA expression, (**B**) *ORAI1* mRNA expression, (**C**) ryanodine receptor 1 (*RyR1*) mRNA expression, (**D**) leucine zipper-EF-hand containing transmembrane protein 1 (*LETM1*) mRNA expression, (**E**) plasma membrane Ca2<sup>+</sup> ATPase (PMCA)3 mRNA expression, (**F**) SR Ca2<sup>+</sup> ATPase 1 (*SERCA1*) mRNA expression, (**G**) mitochondrial calcium uniporter (*MCU*) mRNA expression, (**H**) mitochondrial calcium uptake 1 (*MICU1*) mRNA expression, (**I**) mitochondrial calcium uptake 2 (*MICU2*) mRNA expression, and (**J**) calmodulin (*CALM*) mRNA expression in PL and AM muscles during different periods. PL, plantaris; AM, adductor magnus. SA, summer active group; PRE, pre-hibernation group; LT, late torpor group; IBA, inter-bout arousal group; ET, early torpor group; POST, post-hibernation group. Values are means ± SEM, n = 6–8. \* (*p* < 0.05 and fold change ≥ 2-fold), compared with SA; # (*p* < 0.05 and fold change ≥ 2-fold) compared with PRE; & (*p* < 0.05 and fold change ≥ 2-fold) compared with LT; \$ (*p* < 0.05 and fold change ≥ 2-fold) compared with IBA; + (*p* < 0.05 and fold change ≥ 2-fold) compared with ET.

**Figure 7.** Changes in the protein expression of distinct Ca2<sup>+</sup> transport and binding proteins in PL and AM muscles during different periods. Histograms depicting (**A**) representative Western blot images of STIM1, ORAI1, RyR1, LETM1, PMCA3, SERCA1, MCU, MICU1, and CALM in PL and AM muscles during different periods. Histograms depicting (**B**) STIM1 protein expression, (**C**) ORAI1 protein expression, (**D**) RyR1 protein expression, (**E**) LETM1 protein expression, (**F**) PMCA3 protein expression, (**G**) SERCA1 protein expression, (**H**) MCU protein expression, (**I**) MICU1 protein expression, and (**J**) CALM protein expression in PL and AM muscles during different periods. PL, plantaris; AM, adductor magnus. SA, summer active group; PRE, pre-hibernation group; LT, late torpor group; IBA, inter-bout arousal group; ET, early torpor group; POST, post-hibernation group. Values are means ± SEM, n = 6–8. \* *p* < 0.05, \*\* *p* < 0.01, and \*\*\* *p* < 0.001 compared with SA; # *p* < 0.05, ## *p* < 0.01, and ### *p* < 0.001 compared with PRE; & *p* < 0.05 compared with LT; \$ *p* < 0.05 and \$\$ *p* < 0.01 compared with IBA; ++ *p* < 0.01 compared with ET.

In view of the opposite changes between cytoplasmic and SR Ca2<sup>+</sup> during hibernation, and to further explore whether the increased cytoplasmic Ca2<sup>+</sup> during hibernation resulted from SR Ca2<sup>+</sup> release, the mRNA and protein expression levels of RyR1 were measured. As shown in Figure 6C, in the PL muscle, compared with that in the SA group, the mRNA expression of RyR1 was significantly elevated in the IBA group. Its protein expression also increased (47–60%) in the distinct hibernation groups (Figure 7D). In the AM muscle, compared with that in the SA group, no significant change in RyR1 mRNA expression occurred (Figure 6C); however, its protein expression was significantly increased by 32% in the IBA group (Figure 7D). Overall, the protein expression of RyR1 in the PL and AM muscles increased significantly during hibernation (especially when animals aroused from torpor).

To further explore whether increased cytoplasmic Ca2<sup>+</sup> during hibernation resulted from mitochondrial Ca2<sup>+</sup> extrusion, the mRNA and protein expression levels of LETM1 were detected. In the PL muscle, compared with that in the PRE group, the mRNA expression of LETM1 increased significantly in the IBA and ET groups. Its protein expression level also significantly increased (37–40%) in the IBA and ET groups. In the AM muscle, compared with that in the SA group, the mRNA expression of LETM1 was dramatically elevated in the POST group (Figure 6D). The protein expression level was also elevated by (36–48%) in the LT, ET, and POST groups (Figure 7E). Overall, the mRNA and protein expression levels of LETM1 in the PL and AM muscles were elevated during hibernation and after post-hibernation.

**Figure 8.** Changes in Pearson correlation coefficients of ORAI1/STIM1 in PL and AM muscles during different periods. (**A**) Representative fluorescence images in PL and AM muscles during different periods. 800× magnification, scale bar = 100 μm. (**B**) Histogram depicting Pearson correlation coefficients of ORAI1/STIM1 in PL and AM muscles during different periods. PL, plantaris; AM, adductor magnus. SA, summer active group; PRE, pre-hibernation group; LT, late torpor group; IBA, inter-bout arousal group; ET, early torpor group; POST, post-hibernation group. Values are means ± SEM, n = 6–8. \* *p* < 0.05 and \*\*\* *p* < 0.001 compared with SA; # *p* < 0.05 compared with PRE; & *p* < 0.05 compared with LT; \$ *p* < 0.05 and \$\$ *p* < 0.01 compared with IBA.

The results showed that the cytoplasmic Ca2<sup>+</sup> level partially recovered when the ground squirrels re-entered torpor after 12–24 h of IBA, which led us to consider the Ca2<sup>+</sup> level recovery mechanisms involved in this process. The mRNA and protein expression levels of PMCA3 (a major plasma membrane Ca2<sup>+</sup> ATPase in skeletal muscle) were first detected. The results showed that, compared with the SA and PRE groups, the mRNA level of PMCA3 in the PL muscle increased significantly in the IBA and ET groups (Figure 6E). A lower mRNA expression of PMCA3 was observed during hibernation in the AM muscle. In addition, the protein expression level of PMCA3 showed no significant differences among the different groups in either the PL or AM muscles (Figure 7F).

The mRNA and protein expression levels of SERCA1, a highly important Ca2<sup>+</sup> pump located in the SR, were then detected. Compared with that in the SA group, the mRNA expression level of SERCA1 in the PL and AM muscles increased significantly in the IBA group (Figure 6F). Consistent with the changes in mRNA expression levels, the protein level of SERCA1 in the PL and AM muscles also increased significantly (43–58%) in the LT, IBA, and ET groups (Figure 7G). Overall, elevated SERCA1 mRNA and protein expression levels were observed in the PL and AM muscles during hibernation.

Mitochondria also plays a critical role in Ca2<sup>+</sup> storage in skeletal muscle. Therefore, the mRNA and protein expression levels of the MCU complex, a major mitochondrial Ca2<sup>+</sup> uptake channel, were determined. In the PL and AM muscles, compared with those in the SA group, no significant changes in mRNA expression levels of MCU, MICU1, or MICU2 were observed during hibernation (Figure 6G,H,I). However, the protein expression levels of MCU and MICU1 were increased significantly (29–51%) in the LT, IBA, and ET groups (Figure 7H,I).

In addition to the Ca2<sup>+</sup> channels or pumps, free Ca2<sup>+</sup> binding protein plays a major role in cytoplasmic Ca2<sup>+</sup> levels. Therefore, we also detected the mRNA and protein expression levels of CALM. As shown in Figure 6J, in the PL and AM muscles, there were no significant differences in the mRNA expression levels of CALM among the different groups. However, the protein expression level was significantly increased during hibernation in the PL muscle, but showed no significant differences in the AM muscle (Figure 7J).

#### **4. Discussion**

A comprehensive time-course investigation was firstly carried out to measure the cytoplasmic, SR, and mitochondrial Ca2<sup>+</sup> levels, as well as clarify the possible Ca2<sup>+</sup> regulatory mechanisms, in the skeletal muscles of Daurian ground squirrels during different hibernation states. Fluctuations in intracellular Ca2<sup>+</sup> levels were observed during the torpor-arousal cycle, with Ca2<sup>+</sup> levels partially recovered when the animals aroused and then re-entered torpor (Figure 9A). Further investigation suggested that the Ca2<sup>+</sup> proteins/channels and free Ca2<sup>+</sup> binding protein located in the cytoplasm, SR, and mitochondria all participated in the fluctuation of intracellular Ca2<sup>+</sup> (Figure 9B).

**Figure 9.** Graphical summary of the study. Our research focused on the potential roles of major Ca2<sup>+</sup> channels and proteins in cytoplasmic Ca2<sup>+</sup> fluctuations in skeletal muscles of Daurian ground squirrels during hibernation. (**A**) Fluctuation of cytoplasmic Ca2<sup>+</sup> levels in skeletal muscle fibers during the torpor-arousal cycle. (**B**) Increased activation probability or protein expression of SOCE, RyR1, and LETM1 may participate in the periodic elevation of cytoplasmic Ca2<sup>+</sup> during hibernation, whereas the increased expression of SERCA1, MCU, and CALM (only for PL) may be potential mechanisms through which hibernators can attenuate cytoplasmic Ca2<sup>+</sup> and restore Ca2<sup>+</sup> homeostasis during hibernation. Compared with that in LT or IBA groups, the decreased activation of SOCE and increased expression of MCU may participate in the partial down-regulation of cytoplasmic Ca2+. PL, plantaris; AM, adductor magnus. SA, summer active group; PRE, pre-hibernation group; LT, late torpor group; IBA, inter-bout arousal group; ET, early torpor group; POST, post-hibernation group.

*Cells* **2020**, *9*, 42

During prolonged hibernation, hibernators experience various stressful conditions, including prolonged inactivity, hypoxia, fasting, and repeated ischemia-reperfusion from the torpor-arousal cycle. Here, after several months of hindlimb inactivity during hibernation, slight decreases in muscle mass and single muscle fiber CSA (15–20%) were observed, suggesting that only limited skeletal muscle loss occurs in the PL and AM muscles, consistent with our previous report on PL and gastrocnemius muscles [36]. Obviously, hibernators can be considered good anti-atrophy models, and their unique skeletal muscle preservation mechanisms deserve further exploration. In view of the critical role of intracellular Ca2<sup>+</sup> homeostasis in skeletal muscle maintenance, the present study focused on changes in Ca2<sup>+</sup> levels in the different compartments (including cytoplasm, SR, and mitochondria) of skeletal muscle fibers during different hibernation stages. The results showed that the cytoplasmic Ca2<sup>+</sup> levels in the PL and AM muscles were elevated to varying degrees during hibernation, suggesting that, similar to the phenomenon found in non-hibernators, prolonged skeletal muscle disuse during hibernation was accompanied by elevated cytoplasmic Ca2<sup>+</sup> levels [37–40]. However, fluctuations in cytoplasmic Ca2<sup>+</sup> levels were observed during the torpor-arousal cycle, with an increase during LT and partial recovery when the squirrels re-entered torpor after transient arousal. Therefore, periodic arousal from torpor may be a strategy employed by hibernators to adjust and stabilize cytoplasmic Ca2<sup>+</sup> levels to effectively avoid excessive Ca2+-induced skeletal muscle loss or damage. Our previous study found that, compared with pre-hibernation levels, serum Ca2<sup>+</sup> concentrations in ground squirrels increased significantly during hibernation and recovered after post-hibernation [41]. This raises the question of whether increased serum Ca2<sup>+</sup> concentrations during hibernation influence cytoplasmic Ca2<sup>+</sup> levels. SOCE is the main channel of extracellular Ca2<sup>+</sup> influx. Higher co-localization coefficients of its two major components, i.e., STIM1 and ORAI1, represent a higher activation probability of SOCE [35]. Our results showed that the co-localization coefficients of STIM1 and ORAI1 increased significantly during LT and IBA, suggesting that the activation probability of SOCE increased during these two stages. Furthermore, the extracellular Ca2<sup>+</sup> influx mediated by SOCE may be one reason for the increased cytoplasmic Ca2<sup>+</sup> level during LT and IBA. In addition to extracellular Ca2<sup>+</sup> influx, Ca2<sup>+</sup> release from intracellular Ca2<sup>+</sup> storage can also cause elevated cytoplasmic Ca2<sup>+</sup> levels. RyR1 is the main Ca2<sup>+</sup> release channel located in the SR. The results showed that its protein expression level was up-regulated to varying degrees during different periods of hibernation. Combined with the phenomenon that the Ca2<sup>+</sup> concentration in the SR showed the opposite change to that of cytoplasmic Ca2<sup>+</sup> during hibernation, the increased release of SR Ca2<sup>+</sup> was likely a major reason for the increased Ca2<sup>+</sup> level in skeletal muscles of ground squirrels during hibernation. In addition to SR Ca2<sup>+</sup> release, several proteins also mediated Ca2<sup>+</sup> efflux from the mitochondria to the cytoplasm. The present study showed that the protein expression of LETM1, a major channel that mediates Ca2<sup>+</sup> efflux in the mitochondrial membrane, was increased to varying degrees during different periods of hibernation, intimating that Ca2<sup>+</sup> release mediated by LETM1 in the mitochondria may also be involved in the increase in cytoplasmic Ca2<sup>+</sup> levels during hibernation.

As a Ca2<sup>+</sup> pump located in the cell membrane, PMCA3 can pump intracellular Ca2<sup>+</sup> out of the cell. In the current study, however, the PMCA3 protein expression levels showed no significant differences during hibernation. Therefore, PMCA3 did not appear to play a substantial role in the fluctuation of cytoplasmic Ca2<sup>+</sup> in hibernation. In contrast, the protein expression levels of SERCA1, a major Ca2<sup>+</sup> uptake channel in the SR, were significantly up-regulated during different stages of hibernation, contrary to the decrease in SERCA activity found in non-hibernating animals under disuse conditions [42]. Our previous study showed that the protein content of SERCA2 in the soleus and extensor digitorum longus muscles of Daurian ground squirrels increased significantly during hibernation and IBA [7]. Other studies have also reported that SERCA activity is more resistant to temperature reduction in hibernating cardiac muscle than that in non-hibernating rats [43,44]. Such evidence indicates that, although faced with various stresses during hibernation, including a low temperature, low metabolism, and prolonged skeletal muscle disuse, SERCA1 still maintains high activity and can pump more cytoplasmic Ca2<sup>+</sup> into the SR, thus avoiding an excessive increase in

cytoplasmic Ca2<sup>+</sup> and related skeletal muscle injury. In addition, the protein expression of MCU, a major Ca2<sup>+</sup> absorption channel located in the mitochondrial membrane, was elevated during hibernation, and may therefore be another important mechanism for the absorption of Ca2<sup>+</sup> into the mitochondria and alleviation of the Ca2<sup>+</sup> level in the cytoplasm. However, we know that when the concentration of mitochondrial Ca2<sup>+</sup> reaches a certain threshold, mitochondrial depolarization is triggered and, as a consequence, the pro-apoptotic protein Bax is activated, translocated, and inserted into the outer membrane via Bax/Bax-homo-oligomerization [45]. This is followed by the formation and opening of a mitochondrial permeability transition pore (mPTP), through which cytochrome C (a mitochondria-residing apoptogenic factor) is released into the cytosol, leading to the cleavage of nuclear DNA and cell apoptosis [3,46]. In the present study, the level of mitochondrial Ca2<sup>+</sup> only increased slightly during hibernation. Therefore, the simultaneously higher expression of MCU and LETM1 may be a strategy employed by hibernating ground squirrels to control the balance between cytoplasmic and mitochondrial Ca2<sup>+</sup>, and thereby avoid mitochondrial Ca2<sup>+</sup> concentration-induced skeletal muscle damage, such as cell apoptosis. As a free Ca2<sup>+</sup> binding protein, CALM can relieve the elevation of cytoplasmic Ca2<sup>+</sup> by binding to free Ca2<sup>+</sup> in the cytoplasm. Our results showed that the protein expression of CALM in the PL muscle increased significantly, and the enhanced free Ca2<sup>+</sup> binding capacity mediated by the elevated protein expression of CALM may thus be another important mechanism in hibernating ground squirrels to alleviate cytoplasmic Ca2<sup>+</sup> levels during hibernation.

During the torpor-arousal cycle, the intracellular Ca2<sup>+</sup> level showed increased-decreased fluctuation. Compared with that during LT and IBA, the cytoplasmic Ca2<sup>+</sup> level in the skeletal muscles of the Daurian ground squirrels was down-regulated to varying degrees after re-entry into torpor (ET). Compared with levels in LT, the co-localization coefficients of STIM1 and ORAI1 decreased to varying degrees, representing decreased Ca2<sup>+</sup> influx mediated by SOCE, which may be another critical mechanism used to avoid the intracellular Ca2<sup>+</sup> increase caused by persistent Ca2<sup>+</sup> influx during hibernation. In addition, we found that the higher protein expression of Ca2<sup>+</sup> uptake channel MCU in ET may be another vital mechanism through which PL muscle can relieve cytoplasmic Ca2<sup>+</sup> overload. It is worth noting that, after arousal from hibernation, the intracellular Ca2<sup>+</sup> levels (i.e., cytoplasmic, SR, and mitochondrial) returned to the levels found in the SA and PRE groups, thus reflecting a strong and rapid Ca2<sup>+</sup> recovery ability of these ground squirrels. We further found that the expression of Ca2<sup>+</sup> transport proteins/channels or free Ca2<sup>+</sup> binding protein also returned to the levels observed in the SA and PRE groups. This may be due to the sampling time of the POST group, which occurred 3 d after the squirrels aroused from torpor, and all physiological states had returned to normal.

By comparing and analyzing the mRNA and protein expression levels of each Ca2<sup>+</sup> transporter protein, we found that the levels were not always consistent in the same stage. In particular, during LT, the mRNA levels of most Ca2<sup>+</sup> transport proteins/channels were decreased or showed no significant change, whereas the corresponding protein expression levels were significantly increased. In view of these results, we checked the transcription and proteomic analysis results of the skeletal muscles during different hibernation stages (unpublished data) and found that the predicted results were consistent with our experimental findings. Previous studies have demonstrated that transcriptional elongation and initiation are essentially arrested during hibernation [47,48], which may be one of the main reasons for the lower mRNA expression of Ca2<sup>+</sup> transport proteins/channels in skeletal muscles of Daurian ground squirrels during LT. In other words, the inconsistent changes in mRNA and protein expression levels may be the result of total inhibition at the transcriptional level during LT, although the corresponding protein expression was not affected during hibernation.

In summary, despite experiencing stresses such as a low temperature, low metabolism, and prolonged hindlimb inactivity during hibernation, hibernating ground squirrels still possess a strong Ca2<sup>+</sup> operation ability. Here, by regulating the activity and protein expression of Ca2<sup>+</sup> pumps, Ca2<sup>+</sup> channels, and Ca2<sup>+</sup> binding proteins in the cytoplasm, SR, and mitochondrial membrane, the dynamic balance of intracellular Ca2<sup>+</sup> homeostasis was well-maintained during hibernation. Therefore, maintaining intracellular Ca2<sup>+</sup> homeostasis and avoiding skeletal muscle injury caused by

its disturbance appear to be priority strategies employed by hibernating squirrels to cope with the various stresses induced during the torpor-arousal cycle.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2073-4409/9/1/42/s1, Figure S1: Represent image of the complete SDS-PAGE lane for each antibody used in present study.

**Author Contributions:** J.Z. and X.L. participated in the study design and carried out the molecular laboratory work, acquisition of data, data analysis and interpretation, and drafting of the manuscript. F.I., S.X., Z.W., C.Y., and X.P. helped draft the manuscript. Y.G. contributed to the conception and design of the study and final approval of the submitted version. H.C. and H.W. critically revised the paper for important intellectual content. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was supported by funds from the National Natural Science Foundation of China (No. 31772459), Shaanxi Province Natural Science Basic Research Program (2018JM3015), and Key Project of North Minzu University (No. 2017KJ28, 2016skky02).

**Conflicts of Interest:** The authors declare no conflicts of interest.
