Upward range shift has been recorded in numerous organisms as a response to climatic change in recent decades since cooler environments at higher altitudes may relieve them from overheating (Parmesan and Yohe, 2003; Molina-Martínez et al., 2016; Freeman et al., 2018). During this process, high-altitude environments may impose other severe physiological stressors to species, which may represent new challenges for their survival and reproduction, including hypoxia and intense UV radiation (Bouverot, 1985; Storz et al., 2010; Qiu et al., 2012). Knowledge on how low-altitude poikilothermic organisms cope with high-altitude environments, however, remains limited, which may impede our understanding of the processes and mechanisms of how species react and evolve under climatic change (Pinsky et al., 2019; Sunday et al., 2019; Feldmeier et al., 2020; Jacobsen, 2020).

Body temperature of poikilothermic animals is largely dependent on environmental temperature (Gracey et al., 2004; Harwood, 2007). Meanwhile, body temperature is a key parameter for poikilothermic animals, and has a significant influence on animal physiological processes including energy metabolism, respiration, acid-base balance, and enzyme activity, of which are all closely correlated (Withers, 1978; Vitt and Caldwell, 2014; Abram et al., 2017; Gangloff et al., 2019; Taylor et al., 2021). Consequently, each poikilothermic species has its own optimal temperature range, and they typically reach their ranges by physiological or behavioral thermoregulation (Vitt and Caldwell, 2014; Rozen-Rechels et al., 2019). Because environmental temperature changes along with altitudinal gradient, poikilothermic animals that live in high-altitudes have adapted to the cold environments by regulating their thermal regimes (Bouverot, 1985; Muir et al., 2014; Domínguez-Godoy et al., 2020). Therefore, understanding the variations of preferred temperature and thermoregulation behavior among high- and low-altitude populations is crucial for apprehending the effects of climatic change on range shift of poikilothermic animals (Trochet et al., 2018).

Hypothermia results in a low temperature preference and can be achieved through thermoregulation behavior in poikilothermic vertebrates, including amphibians, reptiles, and fish (Wood and Gonzales, 1996; Moretti et al., 2018; Jones et al., 2019; Duran et al., 2020; Skandalis et al., 2020). Hypothermia can be induced by exhaustive exercise (Tattersall and Boutilier, 1999), hypoxia (Branco et al., 2014; Skandalis et al., 2020), or other stimulations (Moretti et al., 2018; Duran et al., 2020; Skandalis et al., 2020), and is known as lower body temperature than normal state (Romanovsky et al., 2005). Several physiological mediators of low body temperature in poikilotherms have been uncovered, including acidosis and lactate (Wagner et al., 1999; Nedrow et al., 2001). How animals benefit from hypothermia, however, remains controversial. Decreasing body temperature can shift the oxygen dissociation curve leftward, which increases oxygen loading in lungs, reduces energetic cost of ventilatory and cardiac hyperactivity, and restores acid-base balance in blood and tissue (Withers, 1978; Wang et al., 1998; Petersen et al., 2003). Thus, low body temperature in hypoxia likely represents an adaptive response to protect vital organs and reduce energy expenditure (Wood and Gonzales, 1996; Petersen et al., 2003). Furthermore, hypothermia may be a protective mechanism in hypoxic condition, such as at high-altitudes, which can increase survival rate (Bicego et al., 2007; Branco et al., 2014).

Upward range shift in poikilothermic animals has been a key research topic in recent decades (Gangloff et al., 2019; Domínguez-Godoy et al., 2020). Hence, studies on variations of natural thermal performance and tolerance, and the mechanisms behind the formation of variation are highly desirable, especially thermal biology on variations among individuals and populations within species (Taylor et al., 2021). High-altitude environmental stressor, particularly hypoxia and cold, impose significant constraints on animal physiology. Hypoxia typically has an negative impact on life activities, such as reproduction and development (Souchet et al., 2020). How animals living at different altitudes respond to their environments, and how they may change their thermal physiological reaction norms are fascinating questions. Therefore, intraspecific comparison between high- and low-altitudinal populations of poikilotherms represents an ideal paradigm to explore the variations of thermal biology and enhance understanding of how poikilothermic animals cope with climatic change.

Anurans are typical poikilothermic vertebrates, and are sensitive to rapidly elevating environmental temperature, which makes them excellent models for studying impacts of changing environmental temperature (Bodensteiner et al., 2021; Taylor et al., 2021). The Asiatic Toad (Bufo gargarizans) is a common anuran species that has a wide distribution in east Asia, and a wide altitudinal range from sea level to over 4,000 m above sea level (a.s.l.; AmphibiaChina, 2020). Therefore, we choose Asiatic toads that live in different altitudes to test intraspecific variations of hypothermic performance.

In this study, we used an exhaustive-exercise induced hypothermia and compared hypothermic performance between high- and low-altitudinal populations of Asiatic toads, both at their original habitat and in a common garden environment. We hypothesized that the high-altitude toads would have a stronger hypothermic reaction (lower body temperature) to exhaustive exercise than the low-altitude toads do, because cold temperature and hypoxia at high-altitudes may interactively affect their thermal physiology and high-altitudes toads may have adapted or acclimated to their environments. Besides, modifications of their thermal physiology could be fixed (evolutionary adaptation) or plastic (phenotypic plasticity).

The average moved distance of the origin-L population in rounds 1 and 2 (total trial n = 16) was 44.87 ± 37.73 m (mean ± sd), and in rounds 3 and 4 (total trial n = 16) was 61.33 ± 33.40 m. The same measurements of the origin-H population were 14.35 ± 10.51 m and 17.08 ± 10.06 m for rounds 1 and 2 and for rounds 3 and 4, respectively. The ΔT_S of the origin L population in round 1 and round 2 was –1.41 ± 0.76°C, and in round 3 and round 4 was –3.01 ± 1.05. The same measurements of the origin H population were 0.11 ± 0.93°C and –3.05 ± 0.71°C for rounds 1 and 2 and for rounds 3 and 4, respectively.

In rounds 5 and 6, the ΔT_S was –1.61 ± 0.61°C in the experimental group, and was 0.06 ± 0.31°C in the control group. The ΔT_C in the two groups were –1.17 ± 0.58°C and 0.08 ± 0.17°C, respectively. The control group were not allowed to move in trials, so the moved distance was zero, and the distance of the experimental group was 24.65 ± 28.52 m.

Furthermore, the adjusted R² from the GLM between ΔT_S and ΔT_C was 0.90. After removing three outliers, the adjusted R² increased to 0.93 (Supplementary Table 2). Therefore, variations between the two measurements were within an acceptable level (estimate ± se = 1.363 ± 0.074; p < 0.01; Table 2 and Figure 2A), and ΔT_S was a valid substitute for ΔT_C.

Model analysis indicated a non-linear correlation between quadratic moved distance and ΔT_S (0.254 ± 0.068, p < 0.01; –0.807 ± 0.142, p < 0.01; Table 2 and Figure 2B). The results clearly demonstrated that exhaustive exercise caused hypothermia. Body temperature reduction increased with prolonged moved distance, but the slope of changing temperature gradually became flat along with the distance.

The constructed linear model, which contained two interactive effects, supported a significant effect of all independent variables on ΔT_S. The total moved distance was negatively correlated with ΔT_S (–0.731 ± 0.299, p < 0.01; Table 3 and Figure 3A), which was concordant with the polynomial regression. Round was positively correlated with ΔT_S (0.377 ± 0.187, p < 0.05; Table 3 and Figure 3B), which meant the ΔT_S was significantly increased in round 2 and round 4, compared to round 1 and round 3, except at the Kangding location (Figure 3B). When compared to the origin-H population, the origin-L population had a positive relationship with ΔT_S (0.579 ± 0.223, p < 0.05; Table 3 and Figure 3A). When compared to the Chengdu location, the Kangding location had a significantly positive correlation with ΔT_S (4.804 ± 0.787, p < 0.01; Table 3 and Figure 3B), but the Shimian location was not (0.853 ± 0.784, p < 0.05; Table 3).

A better way to understand the results was through the interactive effects. Interactive effect of distance and origin was significantly correlated with ΔT_S (distance × Shimian: 0.632 ± 0.312, p < 0.05; Table 3), which meant when moving the same distance, a toad from the origin-L group had a larger ΔT_S than a toad from the origin-H group (Figure 3A). Another set of significantly interactive variables was round and location (round × Kangding: –1.344 ± 0.328, p < 0.01; Table 3), which predicted a negative relationship between round and ΔT_S when the racetrack trials were conducted in Kangding, and the slope was neither similar to slope in Shimian nor similar to slope in Chengdu (Figure 3B).

The origin-H toads had lower body temperature before exercise than origin-L toads did in round 1 and round 2 (round 1: W = 0, p < 0.001; round 2: W = 11, p < 0.05; Supplementary Table 5 and Supplementary Figure 1A). However, origin-H toads had higher body temperature than origin-L toads did in rounds 3 and 4 (round 3: T = 2.147, p < 0.05; round 4: W = 56, p < 0.05; Supplementary Table 5 and Supplementary Figure 1A). As for the post-exercise body temperature, there was no significant difference between the two population (Supplementary Table 5 and Supplementary Figure 1B).

Furthermore, origin had significant correlation with moved distance from the linear model (origin-L: 1.289 ± 0.287, p < 0.01; Supplementary Table 6 and Supplementary Figure 2).

Our results showed that high-altitude population of Asiatic toads had a stronger hypothermic reaction to exhaustive exercise than low-altitude conspecific toads did. The high-altitude population had lower body temperature at round 2 than round 1 at its original habitat, which was different from toads of low-altitude. When the effect of moved distance was controlled, the high-altitude population had larger body temperature reduction than low-altitude population at both original habitats and common garden location. In other words, high-altitude population did not change the extent of hypothermic reaction to exhaustive exercise after acclimation.

Our results confirm the exhaustive-exercise induced hypothermia in the Asiatic Toad, a phenomenon that has previously been reported for a few other poikilothermic animals (Petersen et al., 2003). Behavioral hypothermia has been proposed as a protective strategy to reduce metabolic demands during hypoxia or after exercise (Wagner et al., 1999; Hicks and Wang, 2004; Branco et al., 2006). Poikilothermic animals in hypothermic state may reduce oxygen consumption, increase affinity of hemoglobin for oxygen, and reduce energetically costly responses (Steiner and Branco, 2002). In turtles (Chrysemys picta), for instance, the oxygen-hemoglobin dissociation curve shifts rightward at high temperatures, meaning that oxygen saturation changes to a lower level at the same PO₂ at high temperature than at low temperature (da Silva et al., 2013). Besides, glucose metabolism is positively correlated with oxygen availability (Gullino et al., 1968; Storz and McClelland, 2017), and experiment on vitro muscle of the American Bullfrog (Lithobates catesbeianus) suggests that net rate of glucose metabolism at low temperature is reduced compared to that at high temperature (Petersen and Gleeson, 2009). Therefore, hypothermia may increase survival rate of poikilothermic animals under various metabolism interruptions such as hypoxia or exhaustion (Wood and Gonzales, 1996; Petersen et al., 2003; Morris, 2004).

An interesting difference between high- and low-altitude toads is the predicted slope between moved distance and body temperature reduction, which indicates a lower body temperature of high-altitude toads than that of low-altitude conspecifics when moving the same distance. We postulate that this variation may represent an adaptative shift of hypothermic performance in high-altitude environments. For poikilothermic animals living at high altitudes, the cost of allocating energy to thermoregulation to achieve a high body temperature as a response to stress (e.g., infection) may offset the benefits of the high body temperature in cold and hypoxic environments (Duran et al., 2020). Thus, these poikilothermic animals tend to lower their body temperature to save energy for other more important life activities (Duran et al., 2020). On the other hand, oxygen uptake at resting state changes along with body temperature in most animals (Deluen et al., 2022), and animals in a hypoxic environment may suffer from hyperthermia (Wood and Gonzales, 1996). In turn, hyperthermia favors higher metabolism and promotes body recovery but also demands higher oxygen uptake. Amphibians like Bufo marinus have a low temperature preference during hypoxic exposure, and this behavioral hypothermia increases arterial saturation and reduces oxygen uptake (Wood and Malvin, 1991). For poikilothermic vertebrates, low temperature also eases ventilatory responses in hypoxic conditions (da Silva et al., 2013). Another possible explanation is the metabolic cold adaptation (MCA) hypothesis, which suggests poikilotherms in cold environments would have a high standard metabolic rate to ensure life activity at low temperatures (Gaston et al., 2009; Deluen et al., 2022). High-altitude populations of the Pyrenean Brook Newts have a higher baseline of the relationship between standard metabolic rate and body temperature compared to low-altitude populations, which supports the MCA (Deluen et al., 2022). Thus, the high-altitude toads may have a metabolic level at a low body temperature similar to that of low-altitude toads at a high body temperature. Hence, more exploration on the physiological significance of hypothermic performance in high-altitudinal poikilotherms is needed.

The observed variations of hypothermic reaction may predict a potential changing direction for low-altitude populations when facing warming climate. Warming temperature affects important physiological processes such as metabolism, thus increased temperature may push poikilothermic organisms to higher altitudes or latitudes for cooler environments (da Silva et al., 2013; Gangloff et al., 2019; Feldmeier et al., 2020; Jacobsen, 2020). Low-altitude toads may change their hypothermic performance to match their high-altitude conspecifics when they shift to high-altitude environments.

Another key challenge to poikilotherms is whether thermal trait modifications can keep pace with the warming climate (Bodensteiner et al., 2021). Poikilotherms have a relatively low plasticity in thermal tolerance, which means physiological adjustments of thermoregulation may not be sufficient for poikilotherms when facing rapid climate warming (Gunderson and Stillman, 2015; Gangloff et al., 2019). According to the common garden comparison of toad populations, both high- and low-altitude toads retain their own extents of hypothermic reaction. Thus, coping with climate change relying on plasticity of hypothermic performance may not be optimistic for the Asiatic Toad.

Exhausting-exercise induces a hypothermic response in both high- and low-altitude toads, and this reaction leads to measurable reduction in body temperature. Besides, high-altitude toads have lower body temperature than low-altitude conspecifics do when moving the same distance. However, the common garden comparison suggests minimum capacity on plasticity of hypothermic performance for both high- and low-altitude toad populations.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

The animal study was reviewed and approved by the Animal Care and Use Committee at the Chengdu Institute of Biology, Chinese Academy of Sciences.

KH: conceptualization and data curation. ZY and KH: methodology. ZY: software, formal analysis, writing – original draft, and visualization. YQ: investigation, supervision, and project administration. ZY and YQ: writing – review and editing. All authors read, edited, and approved the final submitted version of the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.