A frog in hot water: the effect of temperature elevation on the adrenal stress response of an African amphibian

Introduction

Temperature stands out as a crucial factor influencing the development and survival of herptiles (Tattersall et al., 2012; White, Alton & Frappell, 2012). The physiological and biological functions of reptiles and amphibians have evolved in close correlation with temperatures specific to their habitats (Khelifa et al., 2019). While ectotherms thrive in specific, often elevated temperatures (referred to as the “optimal temperature”), exposure to temperature extremes beyond an individual’s tolerance range can induce substantial alterations in biochemical processes, individual fitness, and overall health (Gunderson & Stillman, 2015; Huey & Kingsolver, 1989). Among vertebrates, amphibians emerge as particularly susceptible to temperature variability. This vulnerability is attributed to their distinctive physiology, specific habitat requirements, and a limited capacity to migrate over substantial distances in pursuit of new habitats (Baldwin, Calhoun & De Maynadier, 2006; Carvalho, Navas & Pereira, 2010; Foden et al., 2013). Consequently, amphibians exhibit a diminished capacity to respond as efficiently to drastic temperature fluctuations compared to other vertebrate wildlife (migration/physiological adaptation/phenotypic plasticity; Bernardo & Spotila, 2006; Kearney, Shine & Porter, 2009). These factors, along with the low levels of heat tolerance inherent in amphibians (Nowakowski et al., 2018), means a sudden increase in ambient temperature can result in distinct changes in physiological and biochemical processes (Carey & Alexander, 2003). In addition to this, elevated temperatures can also lead to dehydration and lowered survival rates of amphibian species (Lertzman-Lepofsky et al., 2020).

Numerous studies have attempted to elucidate the impact of temperature variability on free-ranging amphibian populations through the application of direct monitoring or modelling tools (Bartelt & Peterson, 2005; Enriquez-Urzelai et al., 2019; Winter et al., 2016). Despite the efficacy of these tools, their implementation in the captive environment has been limited. This is strange, as by 2020 more than 540 amphibian species were captively housed, totalling 7% of the 7,658 known amphibian species (Jacken, Rödder & Ziegler, 2020). While many amphibians are kept legally in captive environments such as zoos for display purposes, it is important that housing conditions are sufficient to ensure individual health and welfare (Nash, 2005). Furthermore, despite an ever-growing call to enhance the ability to monitor the health and welfare of captive animals (Ferrie et al., 2014), very few facilities housing amphibians have incorporated robust behavioural and physiological monitoring programs to ensure captive individuals will be able to survive current threats (Harding, Griffiths & Pavajeau, 2016). Ensuring the effective and optimal temperature regime within the captive environment is a key factor which needs to be determined for amphibians (Carrillo, Johnson & Mendelson III, 2015). Importantly, this includes an understanding of a species’ temperature tolerance range, as temperatures reaching the upper or lower limit of an organism may result in increased physiological stress. As an elevation in temperature can activate the physiological stress response in many wildlife species (Alfonso, Gesto & Sadoul, 2021; Frigerio et al., 2004), therefore implementing endocrine monitoring may be an ideal tool for monitoring the response of amphibians to management practices, such as temperature determination, in captivity.

The physiology stress response includes the hyperactivation of the hypothalamic-pituitary-interrenal (HPI) axis and an increase in secreted glucocorticoids (GC). An acute elevation in GC levels is adaptive in nature, ensuring an individual can survive within a challenging environment by altering metabolic function and behaviour (Sapolsky, Romero & Munck, 2000; Wack et al., 2012). However, chronically elevated GC levels can be detrimental to amphibians, leading to a decrease in adaptability and coping capacity against natural and anthropogenic stressors (Baker, Gobush & Vynne, 2013; Boonstra et al., 2001; Narayan & Hero, 2014; Phillips & Puschendorf, 2013). As a result of the importance of GCs in amphibian welfare, the quantification of adrenocortical hormones in amphibians under thermal stress, in both controlled and field conditions, can provide valuable information on the “sublethal” effects of temperature fluctuations in these animals (Narayan, 2016).

Endocrine monitoring has primarily been conducted through the collection and analysis of blood or excreted material such as urine and faeces (Sheriff et al., 2011). However, both techniques have considerable flaws when implemented in amphibian research. Firstly, to collect blood samples, animals need to be captured and restrained for prolonged periods at a time, resulting in additional stress for the animal and a possible GC feedback (Romero & Reed, 2005; Tylan et al., 2020). In addition to this, the small body size of most amphibians results in limited venipuncture sites (Heatley & Johnson, 2009; Tapley, Acosta-Galvis & Lopez, 2011). Although urine and faecal collection, and the analysis of hormone metabolites therein, is considered non-invasive in nature (Hodges, Brown & Heistermann, 2010), the small body size inherent in many amphibian species can result in the collection of insufficient sample mass; furthermore, the extended gut passage time in many species (>33 h; Jiang & Claussen, 1993) inhibits the ability to collect frequently from monitored populations (De la Navarre, 2006). A recently validated method for monitoring GC levels in the dermal secretion of amphibians was developed by Santymire, Manjerovic & Sacerdote-Velat (2018). Although the technique is semi-invasive in nature, requiring the brief capture and restraint of study animals, the process ensures longitudinal sampling, while avoiding the effect of GC feedback. Similar to urine and faeces, dermal GC (dGC) concentrations represent an integrative measure of GCs across a prolonged period, e.g., “pooled” (Scheun et al., 2019). Despite the numerous advantages of using mucous as a monitoring matrix, no study to date has used it to monitor the physiological stress response in any amphibian species in response to elevated ambient temperatures. However, studies monitoring amphibian urinary glucocorticoid metabolite (uGCM) concentrations in response to temperature elevation found a significant increase in uGCM levels following temperature alterations (Jessop et al., 2018).

To monitor the effect of elevated temperature on the adrenocortical response in amphibians, we used the edible bullfrog (Pyxicephalus edulis) as a study species. The edible bullfrog is an African amphibian found throughout sub-Saharan Africa, where the species lives underground until habitat conditions improve following seasonally dependent rainfall (Jacobsen, 1989; Vlok et al., 2013). The genus Pyxichephalus are known to enter belowground sites and hibernate to avoid unfavourable conditions (Thomas et al., 2014); individuals will emerge from their belowground sites when rainfall occurs, usually in spring or summer (Channing, Du Preez & Passmore, 1994). The commonly held believe that species capable of retreating underground to escape unfavourable conditions are thereby better equipped to withstand such events may not be entirely accurate. As extreme weather events become more intense and frequent (Stott, 2016), initial models suggest that temperatures below ground could rise by 5 °C by 2065 (Petrie et al., 2020). Further projections indicate that an increase of more than 10 °C at depths of 0–20 cm below the surface can is expected beyond 2065. Thus, underground behaviours displayed by amphibian species will likely not protect individuals from climate change. Except for the validation of mucous and urine as robust matrices for monitoring GCs and their metabolites in the species (Scheun et al., 2019), no further endocrine research has been conducted on the species. As such, the current study aimed to increase our understanding of the physiological responses through the HPI axis and body mass in an African amphibian to elevated temperature change over prolonged periods (5 weeks).

We hypothesised that an increase in environmental temperatures would lead to the hyperactivation of the HPI axis, resulting in a significant increase in GC secretion in dermal glucocorticoid concentrations. Similarly, we hypothesised that an increase in environmental temperature will lead to a decrease in body mass.

Materials & Methods

Results

The results of the linear mixed model indicated that the “Elevated” state was significantly higher than the reference state (z = 3.299, p = 0.001, Table 1). The model included an intercept for individual, reflecting the variability in baseline dGC_log levels across individual animals. In addition to this, the model provided a good fit for the dGC_log data used, as evident by the log-likelihood value of 155.043. The subsequent post-hoc analysis conducted in response to the model findings showed that there was a significant difference in dGC_log levels between the “Elevated” state and all other states used (p < 0.001, Table 1). The largest difference in Dgc_log levels was observed between the baselineEXP and Elevated periods (Table 1). The graphical analysis supported the findings of the linear mixed model (Fig. 1). While the median values for the “Elevated” state was significantly higher than all other states, it should be noted that the standard deviation observed was considerable for all states (“BaselineCtr” = 0.075, “Baseline Ctr 2” = 0.085, “BaselineEXP” = 0.064. “Elevated” = 0.077; Fig. 1); such a high degree of deviation indicates a high level of individual variation within each state.

The linear mixed model analysis revealed a significant effect of group on mass change (z =  − 6.96, p < 0.001). The fixed effect of group indicated that the experimental group (average change: −9.89%, Stdev = 4.02, Fig. 2) had a significantly different mass change compared to the control group (average change: 2.56%, Stdev = 3,78). The variance of the random effect was estimated to be 7.618, indicating individual variability in percentage mass change.

Discussion

This study clearly demonstrates the effect of elevated temperature on the physiological response of P. edulis, including significant increases in observed dGC secretion and a notable decrease in individual weight. More importantly, this research highlighted the usefulness of using dGC monitoring to determine the effect of environmental temperature on amphibian welfare.

The experimental group of this study showed a significant increase in dGC concentrations due to higher environmental temperatures. Such temperature increases have consistently shown adverse impacts on ectotherms, highlighting a universal stress response across diverse species. Notably, research on various reptile species, including the garter snake, Thamnophis elegans (Gangloff et al., 2016), and the northern alligator, Elgaria coerulea (Telemeco & Addis, 2014), support these findings. This hyperactivation of the HPI axis in response to elevated temperature has also been observed in several amphibian species, such as the red-backed salamander, Plethodon cinereu s (Novarro et al., 2018), but also in in the cane toad, Rhinella marina (Jessop et al., 2018; Narayan & Hero, 2014). As such, the current study aligns with broader findings that illustrate an increase in GC secretion as a common physiological response to thermal stressors among ectothermic taxa. The level of dGC increase observed in the current study is similar to that found during an ACTH-challenge conducted on the species (Scheun et al., 2019). This further supports the physiological sensitivity of amphibians to environmental temperatures (Duarte et al., 2012; Novarro et al., 2018).

Although the dGC levels of control animals were significantly lower than experimental animals, several individuals showed higher than expected dGC concentration during Baseline 2. Research has demonstrated the presence of variations in both the physiological stress response and baseline levels of glucocorticoids among individuals and between different sexes (Bourke, Harrell & Neigh, 2012; Eliason et al., 2020; Koolhaas et al., 2010). Furthermore, capture events might have resulted in elevated dGC levels as the study proceeded (Meijer, Sommer & Spriuijt, 2007; Narayan, Cockrem & Hero, 2013). However, additional research needs to be conducted to determine which of the discussed factors might have been responsible for the elevated dGC levels in certain animals of the control group.

All animals within the experimental group showed a significant decrease in weight compared to the control animals. Ectotherm metabolism, like many aspects of their physiology, has evolved under specific environmental temperatures to ensure optimal physiological function (Navas, Gomes & Carvalho, 2008). The use of an external energy source to regulate internal temperatures, though energetically efficient, does make ectotherms vulnerable to any change in environmental temperatures (Li, Cohen & Rohr, 2013). An increase in environmental temperature will lead to greater energetic demands that will likely exceed the resources available to amphibians (Rollins-Smith & Le Sage, 2023). This might well explain the loss in body weight of the experimental animals; as experimental temperatures were increased during the study, food availability remained constant, creating a deficit between energy requirements and availability. While this deficit is easily removed within the captive environment, free-ranging amphibians will have to increase their foraging efforts to meet daily energy requirements (Rohr & Palmer, 2013). Engaging in further activity is expected to further increase body temperature and exacerbate water loss across numerous species, thereby intensifying the issue (Li, Cohen & Rohr, 2013; Navas, Gomes & Carvalho, 2008). This clearly highlights the complexity of the threat climate change poses to free-ranging individuals. A holistic approach should be considered when studying the effect of climate change on any amphibian species, with aspects such as behavioural and physiological responses included in the analysis.

Another aspect to consider when assessing the physiological stress and weight loss in our study animal, is the level of evaporative water loss (EWL) that might occur. Unlike reptiles, amphibians have highly permeable skin that they use to control their internal water balance, but also to assist in regulating body temperature when environmental temperatures exceed the thermoneutral zone (Bickford et al., 2010; Lertzman-Lepofsky et al., 2020; Sannolo, Barroso & Carretero, 2018). As water balance maintenance is crucial for growth, performance, and survival within amphibian species, EWL is often employed by amphibians to survive environmental extremes (Peterman, Locke & Semlitsch, 2013). However, recent climate change models covering large regions of eastern and southern Africa, in which the current study species can be found, predict an increase in average annual temperatures, while fluctuations in precipitation is also expected (Hulme et al., 2001; Nyong & Niang-Diop, 2006). Such extreme or prolonged events will result in considerable EWL in the majority of amphibian species (Lertzman-Lepofsky et al., 2020). This will in turn result in a loss in weight as well as an increase in physiological stress experienced. For captive amphibians such as P. edulis this might not be a lethal aspect to overcome. However, considerable EWL in free-ranging amphibian populations, and the inability to meet the resulting water requirements, may result in several lethal and sublethal effect (Blaustein et al., 2010). However, to further enhance our understanding of EWL in the study species, additional research on captive and free-ranging populations are required.

While the strain produced by both elevated metabolic rates and EWL may have been responsible for the observed increase in dGC concentrations and the decrease in weight of experimental animals, it must be noted that a complex relationship exists between temperature, metabolic rate, water regulations and the HPI axis. Several studies have shown the role of GC in driving metabolic rates in mammalian and herptile species (Crespi & Denver, 2005; De Bruijn & Romero, 2018; Wack et al., 2012). Hence, the observed rise in GC levels due to heightened ambient temperatures could potentially have led to an increase in metabolic activity, resulting in a discernible loss of mass among the animals in the experimental group. Similary, dehydration in amphibians will increase HPI-responsiveness, leading to an increase in secreted GCs (Madelaire et al., 2020). As such, an increase in environmental temperature, a decrease in water availability and higher levels of EWL are responsible for elevated GC levels, which in turn will affect metabolic rates. This complexity, along with the behavioural response inherent in each species, must be considered when conducting climate change related research on captive and free-ranging populations.

Conclusions

While the impacts of extreme environmental conditions on amphibians are widely recognized, there is limited understanding of the direct physiological and physical effects, particularly concerning African species. This study was conducted to rectify this limitation, focusing on a prevalent African amphibian species and employing mucous as a matrix. While an increase in secreted GCs was expected, the scale thereof was surprising. Peak dGC levels not only aligned with prior ACTH results from the species but also highlighted a lack of decline throughout the experiment, underscoring the inability of the study species to adapt and/or acclimatize to extreme conditions. While the 5 °C increase used during this study seems excessive, it does stress the response of the edible bullfrog under extreme, long-term environmental changes. Furthermore, it suggests that beyond a specific threshold, amphibians struggle to regulate biochemical processes, even in the presence of a consistent supply of food, water, and shelter. In such instances, merely improving the habitat for endangered species grappling with climate change will prove inadequate for the effective conservation of their populations in the wild. The findings of this study contribute to our understanding of how African amphibians could react to climate change; however, it should be regarded as a preliminary step, encouraging researchers to undertake more intricate investigations involving this species and others. Research must be conducted to investigate the significance of factors such as temperature, humidity, water and food availability, as well as their combined effects, on the physiological responses in amphibians.

Supplemental Information