Plateau pikas inhabiting high-, middle-, and low-altitude regions were live trapped from Maduo, (4,194 m above sea level, n = 24), Guide, (3,663 m, n = 24), and Guinan (3,321 m, n = 24), respectively, in Qinghai Province, in December 2020. Maduo has an annual average temperature of −4°C and an average monthly temperature below −3.0°C, classifying it as an alpine steppe climate. The annual average temperature of Guide is −3.7°C; a plateau continental climate. Guinan does not experience a severely cold winter or an intensely hot summer, and the annual average temperature is 2.3°C; as such, it is also considered a plateau continental climate.

Ten pikas from each altitudinal region were immediately anesthetized and dissected after capture (five females and five males at each altitude). The adipose tissue, liver, and muscle tissue were immediately preserved in liquid nitrogen and stored at −80°C until further RNA extraction and analyses. A further sample of plateau pikas (n = 14) from each altitude were live-transported to the animal laboratory in Xining (2,261 m above sea level, outdoor temperature −2°C, indoor temperature 20°C). They were kept in 545 × 395 × 200 mm polypropylene material cages separately under 12 L: 12 D lighting conditions, and provided with artificial food (Tianjin Tongyu Feed Sales Co. Ltd.) and ad libitum water. Metabolic experiments were conducted within 24 h.

The RMR of plateau pikas were expressed as oxygen consumption per hour per unit body mass [mL O₂/(g·h)] and measured using an 8-channel FMS (Sable Systems International, Henderson, NV, United States) portable respiratory metabolism system. A biochemical incubator was used to control the chamber temperature, and the experimental temperature was set at 27.5°C (which is within the pika thermal neutral zone) with a standard error of 0.5°C. Metabolic measurements were conducted after the pikas had acclimatized in the chamber for 0.5 h and were resting. The RMRs of seven pikas were measured simultaneously, and a blank tube was used as the baseline for carbon dioxide, oxygen, water vapor, and temperature (Tan et al., 2020; Yu et al., 2021). Four rounds of metabolism were measured in 2 h, with each round lasting 30 min. When the chamber temperature was 27.5°C, the average of the lowest metabolic rates of each individual over at least 10 min was selected as the RMR (Boratyński et al., 2017). RMR is the minimum energy requirement for animals to maintain normal physiological activities within a thermally neutral environmental temperature, while at rest (Arnold et al., 2021). Before the experiment, the pikas were fasted for 2–3 h, and their body mass and temperature were measured using an electronic balance and rectal thermometer, respectively. A digital thermometer probe was inserted gently about 2 cm into the rectum; the measurement time did not exceed 30 s.

Noradrenaline (NE) induction is widely used to determine NST because induced heat generation and cold induction are equivalent and the mechanism is the same. The dose was 0.7 mg/kg in reference to the seasonal variation of NST in plateau pika measured by Wang (Wang and wnag, 1990). The 10 plateau pikas at each altitude were brought back to the laboratory and allowed to adapt for 24 h. The pikas were raised in a single cage under 12 L:12 D illumination in the laboratory, fed with sufficient amounts of rabbit pellet feed (Jiangsu Syu Pharmaceutical Biological Engineering Co., Ltd.), provided water ad libitum, and adapted for 2–3 h before the experiment. NST was measured using an 8-channel FMS respiratory metabolic measurement system. NE was injected subcutaneously into the back with a dose equivalent to pika body weight (0.4 mg/kg). The pikas were immediately put back into the respiratory chamber for 30 min. NE was injected with norepinephrine (1 ml containing 2 mg), having been diluted to 0.4 mg/ml by adding normal saline. The NE was produced by Shanghai Wellhope Pharmaceutical Co., Ltd. In general, the peak in metabolic response occurs 10–45 min after the NE injection. A scatterplot of oxygen consumption against determination time was generated, and the average value of 10 consecutive and stable maximum values was taken as the NST value.

qRT-PCR was used to determine the expression of pparα, pgc-1α, CREB, PRDM16, SLN and uncoupling protein 1 (UCP1). The species-specific primer sets and 18s-actin of the genes in plateau pika were designed in accordance with the reference gene sequences in the NCBI (national center for biotechnology information) website (http://www.ncbi.nlm.nih.gov/) for reference North American pika gene sequences. Primer6.0 software was used to design primers.

Total RNA was extracted from tissues using the Uniq-10 Column Trizol Total RNA Extraction Kit (B511321) in accordance with the kit instructions. The primer was set for 18s and six transforming genes were designed for quantitative real-time PCR. Quantitative real-time PCR was completed using the 2SG Fast qPCR Master Mix (B639271, BBI, Roche) in the LightCycler480 II type fluorescent quantitative PCR instrument (Roche, Rotkreuz, Switzerland). qRT-PCR was carried out in a 10-μL reaction system, which was composed of 5 μL of 2 SybrGreen qPCR Master Mix, 1 μL of cDNA and 0.2 μL of each primer (10 μM/L), and 3.6 μL of ddH₂O. All PCR reactions were repeated. The thermal cycling conditions were as follows: 95°C for 3 min, 45 cycles at 95°C for 5 s, and 60°C for 30 s. The melting curve analysis revealed genes and 18s-amplified single-PCR and final products. We constructed a standard curve for each gene by diluting the cDNA sequence fivefold. The standard curve analysis of target genes and 18 s showed that they had similar amplification efficiency, which ensured the effectiveness of the comparative quantification method. Gene expression was calculated using the 2^−ΔΔCt method and expressed as relative quantities. The nucleotide sequences of primers used for qPCR are shown in Supplementary Table S1.

All data analyses were conducted using R 3.4.3 software. Body mass, RMR, and metabolism-related gene expression levels were analyzed using two-way analysis of variance. Prior to all statistical analyses, data were examined for assumptions of normality and homogeneity of variance by using Shapiro–Wilk and Levene tests, respectively. Differences among groups were detected using Duncan’s multiple range test. Results were presented as mean ± 0.5 standard error (SE); n is the sample size. p < 0.05 was considered statistically significant.

The body masses of plateau pikas from the high- and middle-altitude regions were significantly higher than those from low-altitude regions (F = 7.16, p < 0.05; Figure 1A). The mass-corrected RMRs of plateau pikas were 1.55 ± 0.18, 1.52 ± 0.33, and 1.39 ± 0.17 ml/(g·h) in high-, middle-, and low-altitude regions, respectively. The RMRs of plateau pikas from high- and middle-altitude regions were significantly higher than those from the low-altitude region (F = 3.49, p < 0.05; Figure 1B).

To explore molecular signatures of the thermogenesis of WAT and BAT, we performed profiling of gene expression in the two fat tissues from the three elevation groups. No significant difference was observed in gene expression between the sexes (F = 0.36, p > 0.05), whereas significant differences were detected between the three altitudes (F = 15.56, p < 0.05). UCP1, PRDM16, and PGC-1a are the key transcriptional regulators associated with browning and BAT, they were the transcriptional co-activator that is involved in browning and mitochondrial biogenesis. The expression levels of the UCP1 protein in BAT from high- and medium-altitude regions were higher than those from the low-altitude region, which indicates that BAT is specialized for NST and energy dissipation through the action of UCP1. Evidence for an increase in NST is provided in the supplementary material (Supplementary Figure S1). Similar results were obtained for pgc-1α (Figures 2A,C). The expression levels of pgc-1α, pparα, and PRDM16 genes in the WAT and BAT of plateau pikas from high- and medium-altitude regions were significantly higher than those from the low-altitude region (p < 0.05; Figures 2A,C).

Liver is a metabolic organ, and its metabolic function is controlled by metabolic hormones such as insulin. To explore the molecular characteristics of hepatic gluconeogenesis in liver, we detected the differential expression of energy metabolism molecules in three plateau pika samples at different altitudes. No significant difference was observed between the sexes (F = 1.587, p > 0.05), whereas a significant difference in gene expression levels was found between altitudes (F = 13.59, p < 0.05). The main transcription factors inducing gluconeogenesis include CREB, FoxO1, and several nuclear receptors. PGC-1α is a key transcriptional coactivator for FoxO1 in hepatic gluconeogenesis, which plays a key role in maintaining long-term gluconeogenesis under conditions of scarce food resource. The expression levels of pgc-1α and pparα genes in the liver of plateau pikas from the high-and medium-altitude regions were significantly higher than those from the low-altitude region, but the expression levels of the CREB gene were not significantly different between the three regions (p > 0.05, Figure 2B).

As the largest organ, skeletal muscle is also a major contributor to metabolic rate and can significantly affect metabolism and body weight by increasing muscle energy expenditure through non-shivering thermogenesis. We studied the differential expression of thermogenic molecules in the skeletal muscle of pikas at the three altitudes. No significant difference was observed in gene expression between the sexes (F = 0.78, p > 0.05), but gene expressions in the three regions were significantly different (F = 11.56, p < 0.05). The expression levels of pgc-1α, pparα, and SLN genes in the muscle of plateau pikas from the high-altitude region were significantly higher than those from the medium-altitude region (F = 15.49, p < 0.05). The expression levels of pgc-1α, pparα, and SLN genes of plateau pikas from the medium-altitude region were significantly higher than those from the low-altitude region (F = 14.91, p < 0.05; Figure 2D).

Selective pressures affecting metabolism are complex and can influence metabolic rate through multiple pathways (Zheng et al., 2014b) such as body size, climate, activity, and habits (Killen et al., 2016). Body mass is the most direct indicator of animal energy reserves (Swanson et al., 2017). For example, statistical tests found that the metabolic rate of 533 species of birds was positively correlated with body mass (McNab, 2009). In the present study, the body masses of plateau pikas from high- and middle-altitudes were significantly higher than those from the low altitude. Significant correlations between body mass and metabolic rate were found in the hamster subfamily (Bozinovic, 1992). According to Bergmann’s law, the increase in body mass decreases the surface-to-volume ratio, thereby reducing heat loss and living costs (Zheng et al., 2014a). Many variables associated with physiology are correlated with latitude, indicating that climate is an important factor for the evolution of life-history traits (Tószögyová, 2020). Meta-analyses about the metabolic rate of 69 species of tropical birds and 59 species of temperate birds found that tropical migrants in temperate habitats have lower metabolic rates than do temperate residents (Wiersma et al., 2007). Compared with Cricetulus barabensis kept at room temperature, the energy intake of individuals adapted to a low temperature was higher while the energy intake of individuals adapted to a high temperature was lower (Zhou et al., 2015). The RMRs of plateau pikas in the current study inhabiting high altitude are significantly higher than those of pikas inhabiting low altitude. High metabolic rates may be caused by the biochemical activities of several tissues including the liver, BAT, and skeletal muscle, which all have high mitochondrial oxidative phosphorylation rates (Burton et al., 2011; Selman et al., 2013). Increased metabolism plays an important role in thermal regulation in animals living in an extremely cold environment (Gordon, 2012; McKie et al., 2019).

Cold and hypoxia are defining features of the Qinghai–Tibet Plateau environment, and plateau pika have developed tolerances to this harsh environment (Xie et al., 2014; Wei et al., 2016). Consistent with previous studies that demonstrated that pika have tolerance to hypoxia and low-temperatures (Yang et al., 2006; Zhu et al., 2018), a previous study showed that plateau pika can effectively endure extremely cold environments (Li et al., 2001). Earlier studies found that plateau pika have high NST to cope with the cold environment on the plateau in comparison to Ochotonidae from other regions (Wang et al., 2006; Luo et al., 2008). NST is related to tissue heat production, especially adipose tissue, which is important in regulating body temperature and energy homeostasis in cold environments (Zhu et al., 2017b). As two major types of adipose tissue, WAT is involved in energy storage and BAT is involved in energy expenditure and thermogenesis. BAT and WAT can be conditionally interconverted in response to neuroendocrinal factors, β-3-adrenergic stimulation, and cold stress exposure. WAT responds quickly to environmental changes under cold conditions and takes on the characteristics of BAT. When animals inhabit a cold environment, WAT may possibly transform into beige and brown adipocytes to increase NST in order to adapt to cold conditions (Nedergaard et al., 2007) (Figure 3A).

The intermittent cold exposure experiment demonstrated that plateau pika kept in warm temperatures have little classical brown fat, but the “browning” of WATs is detected rapidly upon cold exposure. The expression of several brown fat differentiation markers, including UCP1, increases simultaneously. The increase in UCP1 expression enhances adaptive thermogenesis (Bai et al., 2015). The study about Tupaia belangeri shows that RMR and the expression levels of pparα, pgc-1α, and PRDM16 increase significantly under cold acclimation, suggesting that browning may appear in WAT (Hou et al., 2020). A genomic, proteomic, and morphological study of energy metabolism in highland pikas and Tupaia belangeri in summer and winter studies revealed that subcutaneous WAT in winter show BAT morphological and histological features (Li et al., 2019). Furthermore, BAT-specific genes, such as UCP1, Cox4, and pgc-1α, are highly expressed in WAT in winter (Li J. et al., 2018). These results suggest that plateau pika adapt to a cold environment by browning scarfskin WAT and adding BAT to increase heat production. Our results suggest that a high expression of pgc-1α may be involved in the critical adaptation mechanisms in pika to cope with the harsh environment of the Qinghai–Tibet plateau. pgc-1α is essential for brown fat thermogenesis and complementary mitochondrial biogenesis, and is also involved in the browning of WAT (Finck and Kelly, 2006). High mRNA expression levels of pgc-1α are observed in high-altitude groups, suggesting high levels of thermogenesis within the tissues. Overall, our study indicated that plateau pikas inhabiting the high altitudes of the Qinghai–Tibet Plateau can regulate their relative gene expression in adipose tissue to, in turn, regulate metabolic level and thermogenic-related physiological performance.

The skeletal muscle is a major determinant of basal metabolic rate (Maurya et al., 2018). Skeletal muscle also plays a central role in temperature homeostasis and can be recruited to produce heat through NST (Nowack et al., 2017). As an uncoupler of the sarcoplasmic reticulum calcium ATPase (SERCA) pump, SLN can enhance futile cycling and increase ATP hydrolysis, thereby creating chronic energy demand (Sahoo et al., 2013). The SLN/SERCA interaction plays a dual role: it creates energy demands in muscle and activates Ca²⁺-dependent signaling, such as pgc-1α and pparα, to increase ATP production through increased mitochondrial biogenesis (Shaikh et al., 2016) (Figure 3B). In genetically engineered SLN mouse models, SLN knockout mice have reduced cold adaptive thermogenesis (Bal et al., 2018). The loss of SLN predisposes mice to diet-induced obesity, indicating that SLN may regulate their energy balance (Bal et al., 2012). Compared to SLN gene-lacking mice, the overexpression of the SLN gene leads to a loss of body mass and increases in the depletion of fat deposits (Rotter et al., 2018). The Rotter et al. (2018) study was conducted at thermoneutrality, which can minimize the contribution to metabolic rate of thermogenic mechanisms. Thus, a high energy consumption may be due to SLN-mediated energy expenditure (Maurya et al., 2015). In the present study, the SLN gene expression level of plateau pika from the high-altitude region was significantly higher than that of plateau pika from the middle-altitude region, which in turn was significantly higher than the SLN gene expression of pika from the low-altitude region. This result suggests that the SLN gene is important in regulating the heat production of plateau pikas at different altitudes.

The liver is an essential metabolic organ, and its metabolic function is regulated by insulin and other metabolic hormones. Numerous transcription factors and coactivators, including CREB, pparg, and pgc-1 regulate the expression of enzymes that catalyze key steps of metabolic pathways, thus managing the energy metabolism of liver (Rui, 2014). When food is scarce, the hepatic gluconeogenesis pathway is enhanced by decreasing the concentration of insulin and increasing the concentration of insulin counter-regulatory hormones, such as glucagon (Han et al., 2016). pgc-1α, CREB/CRTC2, and FoxO1 genes are critical in coordinating the fasting-mediated activation of gluconeogenesis in the liver (Oh et al., 2013) (Figure 3C). In our study, the pgc-1α and pparα in the liver of plateau pika from the high-altitude region were significantly higher than those of pika from the low-altitude region, whereas no significant difference in the CREB gene was detected among the three regions, suggesting that ATP depletion is due to activity-induced energy demands and the storage of fatty acids, cholesterol, glycogen, and proteins, especially in liver (Ke et al., 2018).

In conclusion, the RMR, and the expression of skeletal muscle thermogenic genes and lipid transcription factor genes in plateau pika increases with rising altitude on the Qinghai–Tibet Plateau. Therefore, plateau pikas inhabiting high-altitude environments can survive extreme environments by increasing their metabolic rate, and gene expression of skeletal muscle thermogenesis and adipose tissue. Browning increases the expression of UCPI to promote BAT cell differentiation, thermogenesis, and metabolism. These physiological and gene expression changes confer plateau pika the ability to survive in an extreme environment.