Shifts in environmental conditions caused by different climate events can have profound effects on an organism’s phenotype, fitness, reproductive success and life span. To manage these changes, the organism needs to optimize the trade-offs in time and resource allocation (Metcalfe and Alonso-Alvarez, 2010). Amphibian larvae are often challenged by fluctuations in the environment (Gavrilović et al., 2021; Gavrić et al., 2021). Increased global temperatures and anthropogenic habitat destruction augment the pond desiccation process, which alters larval development (Burraco et al., 2020a). A low water volume limits larval movement, food sources, increases density and predator exposure (Kohli et al., 2019). Under such unfavorable conditions, larvae activate mechanisms of phenotypic plasticity response [the hypothalamic−pituitary−interrenal (HPI) axis and increased thyroid hormone levels] and accelerate the developmental process that allows them to leave the aquatic environment (Burraco et al., 2017; Ruthsatz et al., 2020). Phenotypic-developmental plasticity is particularly critical for amphibians and other species with complex life cycles as it allows the organism to react to specific environmental inputs (Simon et al., 2015; Agudelo-Cantero and Navas, 2019). The shortening of hydroperiods was shown to be an environmental factor that can result in adaptive phenotypic plasticity in amphibian larvae development (Newman, 1989; Denver, 1997; Richter-Boix et al., 2011). Even though at the moment plastic response is beneficial and allows for a niche shift, it affects the timing of metamorphosis, upsets the sensitive energy balance (between growth, development, storage and maintenance) and changes the body-size threshold, which is crucial for the survival of postmetamorphic individuals. The potential carry-over effects on postmetamorphs due to the allocation of energy sources toward premature metamorphosis can limit the larval energy required for growth, the antioxidant and immune systems (Gervasi and Foufopoulos, 2007; Prokić et al., 2019; Sinsch et al., 2020). The most studied trade-off between the aquatic and terrestrial life stages is between body size and developmental time (Newman, 1989; Denver, 1997). Accelerated development is associated with a smaller size at metamorphosis (Nicieza et al., 2006; Lind and Johansson, 2011; Székely et al., 2017, 2020). The body size and energy reserves of metamorphosed individuals entering a novel habitat are generally considered to be positively correlated with fitness, survival of the first year of life, hibernation and size at maturity (Álvarez and Nicieza, 2002; Sinsch et al., 2020). These traits are especially crucial when juveniles face the uncertainty of finding food due to a lack of foraging experience and adaptation to new food sources (Richter-Boix et al., 2006). If the environmental conditions are favorable, smaller juveniles could try to achieve optimal body size through compensatory growth. But if they are faced with the energy limitations imposed by food deprivation, juveniles need to switch to the mode of using internal energy stores, to change energy metabolism and to downsize all processes that are non-essential for survival (Moreira et al., 2018; Prokić et al., 2021). Operating a biological system under a food-deprived state results in physiological modifications in mitochondria activity and increased levels of oxidative stress (Salin et al., 2018; Metcalfe and Olsson, 2021). The state of oxidative stress is associated with damage of macromolecules through lipid peroxidation, protein denaturation, nucleic acid damage and impaired cell function, with consequences on subsequent life events (Halliwell and Gutteridge, 2015). Therefore, the response of an antioxidant system (AOS) is a pivotal process in preventing oxidative stress damage (Metcalfe and Alonso-Alvarez, 2010). The price of operating and maintaining this system is related to the greater use of nutritional reserves (Eikenaar et al., 2018; Janssens and Stoks, 2018; Prokić et al., 2018). With this in mind and according to the life history theory, juveniles with a smaller body should experience a greater trade-off in oxidative stress under conditions of limited energy such as fasting. This is to be expected because of the lower capability of smaller individuals to invest in antioxidant protection in comparison to larger organisms. However, as the abrupt ontogenetic switch points caused by pond drying conditions can additionally affect a set of modifications in oxidative metabolism necessary for the metamorphosis (Allen, 1991; Guxens et al., 2012; Moreira et al., 2021) this may reflect on juveniles AOS in response to fasting conditions. Adaptive reaction to stressful conditions can in some cases enhance cellular response, leading to a hormetic response (Oliveira et al., 2018).

The aims of this study were to investigate (i) the effects of exposure to an environmental stressor (desiccation) during anuran larval development on body size and the oxidative status of metamorphosed individuals, and (ii) the possible trade-off in body size (a smaller body size) induced by desiccation reflected on fitness—on the oxidative stress of juvenile individuals under the energy condition imposed by fasting. To test this, we manipulated the water levels, imitating pond-drying conditions that affected the growth of yellow belly toad (Bombina variegata) larvae. B. variegata was chosen as a model organism for examining the carry-over effects because of the considerable variations in body size of juveniles in natural populations (Biancardi and Di Cerbo, 2010; Schäfer et al., 2018) and the ability to reproduce and develop in ponds with an increased risk of desiccation before metamorphosis (Hartel et al., 2007). Starting from the results of our previous studies and the above mentioned, we assumed that (i) pre-exposure to desiccation would result in a lower body size and modification of the oxidative status of juvenile toads, and that (ii) food deprivation would induce oxidative stress in juvenile individuals, resulting in greater oxidative damage in individuals undergoing developmental desiccation stress.

The desiccation risk in some anuran species affects the larval period (Newman, 1989; Denver, 1997; Lind and Johansson, 2011; Richter-Boix et al., 2011; Székely et al., 2017). In most cases, escape from an unfavorable aquatic environment reduces larval periods and the quality of metamorph phenotypes (resulting in a smaller size and a lower body condition index) (Gomez-Mestre et al., 2013; Burraco et al., 2017; Kohli et al., 2019; Székely et al., 2020; Petrović et al., 2021). In B. variegata, pond drying leads to a significantly lower body size (SVL, BM and BCI) in comparison to individuals developing in constant water availability, but with no acceleration in developmental rate. Our results, in conjunction with previous studies conducted on the same species under desiccation-simulated scenarios, point to the presence of a trade-off in body size at metamorphosis (Sinsch et al., 2020; Petrović et al., 2021). The lower body size of juveniles is associated with a lower resistance to food absence due to low lipid reserves (Bouchard et al., 2016), postponed ossification (Arendt and Wilson, 2000), and decreased immunological function (Kohli et al., 2019). Since larval conditions predict the juvenile phenotype, we examined the carry-over effects of desiccation stress experienced by tadpoles on the oxidative status of juveniles under fasting and feeding conditions. Studying the impact of stressful conditions during development on the ability of juveniles to respond to subsequent stress is important in understanding the challenges that amphibians face in an unpredictable environment.

Under favorable feeding conditions, the oxidative status of B. variegata juveniles from the desiccation regime showed a significant increase in AOS components and the absence of any increased lipid oxidative damage as compared to individuals that developed under constant water availability. A buffered AOS (GSH-Px and GR) and the absence of oxidative damage in response to pond drying were also reported in the case of freshly metamorphosed Pelobates cultripes (Burraco et al., 2017). The observed antioxidant response of juveniles reared under decreasing water levels can be triggered by an increased metabolic rate as the consequence of compensatory growth. Even though juveniles that experienced desiccation had smaller body sizes (SVL, body mass and BCI) after metamorphosis, 2 weeks of feeding lowered these changes, especially in body mass. These individuals gained more body mass, almost doubling by the end of the experiment, indicating greater feeding activity in comparison to juveniles that were exposed to non-stressed conditions throughout the experiment. An increase in mass was expected as fat storage is crucial for survival of the upcoming hibernation process (Reading, 2007; Scott et al., 2007). Székely et al. (2020) confirmed the ability of smaller-sized individuals to compensate for their initial disadvantages. Compensatory growth could be achieved through changes in behavior, increased and longer foraging and changed digestive activity (Tarvin et al., 2015; Buhaciuc et al., 2017), or by modification of the feeding apparatus that allowed for digestion of larger prey and/or a greater amount of food (Bouchard et al., 2016). All of these activities are related to increased metabolic and mitochondrial activities, which lead to higher production of ROS (Metcalfe and Monaghan, 2001; Metcalfe and Alonso-Alvarez, 2010; Burraco et al., 2020b). However, the disturbed redox balance of desiccation experienced by juveniles and caused by a higher metabolic rate appeared to be managed by the AOS. We observed higher activity of SOD, an enzyme that protects against the superoxide anion, the primary radical produced by the mitochondria. Aside from SOD, the activation of the GSH system parameters (total GSH, GR, and GST) supported by the increased levels of thiol groups in juveniles overwhelmed the increase in ROS concentrations and prevented serious oxidative damage. The GSH system is an important part of the cell redox system that scavenges ROS directly through GSH or indirectly as a cofactor for GSH-dependent enzymes (Halliwell and Gutteridge, 2015). Compensatory strategies for delayed hatching of Rana temporaria larvae were also followed by activation of the antioxidant machinery and lower LPO levels, which indicate that low lipid peroxidation was favored by increased antioxidant activities (Burraco et al., 2020b).

Early life exposure to a stressor, in this case desiccation, will exert a negative or positive effect on the oxidative status during other life stages depending on the severity of the stressor and the environment during formative life (Costantini et al., 2012, 2014). Desiccation stress activates in larvae a physiological response that is mediated by the HPI axis and the AOS (Gomez-Mestre et al., 2013; Burraco et al., 2017; Petrović et al., 2021). In our previous study, the larvae of B. variegata under decreasing water exposure experienced higher levels of oxidative stress during metamorphic climax than individuals developing under constant water availability (Petrović et al., 2021). Individuals at Gosner’s stage 46 from the desiccation treatment group exhibited increased LPO and lower values of GSH, CAT, GR, GST and SH groups in comparison to the control. Based on the ability of desiccation to alter the oxidative status in tadpoles, the increase in AOS in juveniles that developed under desiccation could be explained as a consequence of prior exposure to stress (Oliveira et al., 2018). Costantini et al. (2014) suggested that the response to stress in early life may be beneficial for an organism’s fitness when the organism encounters the stressor again, whereas phenotypic adjustments resulting in a stress response could present a certain cost if there is no subsequent exposure to stress in adulthood. This raises the question of whether boosting the AOS in pre-desiccation juveniles provides oxidative protection against other stressors or whether it represents a cost since maintenance of the AOS is demanding energy-wise. With this in mind, yellow belly toad juveniles were subjected to food deprivation stress in order to determine how the oxidative status is modified depending on the body size and the developmental water regime. Our initial assumption was that the stress experienced during larval development would negatively affect the oxidative status of juveniles. Effects were expected due to the capability of desiccation to reduce body size and the energy reserves in juveniles, thereby limiting the energy necessary for antioxidant protection from fasting-induced stress. Fasting in anuran species is common, especially in freshly metamorphosed individuals in the wild (Geise and Linsenmair, 1988; Scott et al., 2007). At this life stage, it is caused by adaptation to the new terrestrial environment when individuals rely on their energy reserves (Nicieza et al., 2006; Bouchard et al., 2016). Fasting is associated with metabolic depression, reduced mitochondrial respiratory capacities and oxygen fluctuations, which induce increased ROS production and oxidative stress (McCue, 2010; Salin et al., 2018; Ensminger et al., 2021; Metcalfe and Olsson, 2021). Food deprivation or low intake of endogenous antioxidants is also reflected in the ability of organisms to equilibrate their oxidative status (De Block and Stoks, 2008). Cabrera-Guzmán et al. (2020) suggested a role for anuran diet composition on the antioxidant and immune responses to stressors. The functioning of some antioxidants is dependent on components that are driven directly from nutrients (GSH from cysteine precursor amino acids, GSH-Px from selenium, and SOD from copper and zinc) (Isaksson et al., 2011). Our results showed that short-term fasting in yellow belly toad juveniles induced oxidative stress when compared to fed individuals, regardless of the water regime. However, a comparison of smaller-body juveniles exposed to pre-desiccation and individuals exposed to a constant water regime but under food absence revealed no significant differences in the examined parameters, suggesting the lack of hormesis in oxidative stress response. This was opposite to our results observed on pre-desiccated juveniles (in comparison to non-desiccated), that exhibited better antioxidant response to following acute chemical stressor (pesticide deltamethrin) (Radovanović et al., 2021).

Food-deprived individuals from both the desiccation and constant water availability groups displayed a diminished antioxidant response, reduced mitochondrial antioxidant protection (SOD and GSH-Px) and second-line defense (GSH, GR, SH groups, and GST), as well as increased susceptibility to lipid oxidative damage. These results are in agreement with our study conducted on fasting larvae of crested newts (Triturus sp.) (Prokić et al., 2021). Two weeks of food deprivation also led to oxidative damage of lipids with a similar pattern of AOS change (SOD, GSH-Px, CAT, GST, and GSH) as observed in B. variegata juveniles, confirming that oxidative stress can be considered as the cost of short-term energetic benefits of reduced metabolism in response to fasting. A similar pattern of changes in oxidative status to impose energy-limited conditions regardless of body size, stress pre-exposure and in both anuran and tailed amphibian species suggests that fasting in amphibians activates a strong and specific response.

In conclusion, exposure to pond drying-out during larval development reduces the body size of juveniles and induces compensatory growth under feeding conditions. In terms of oxidative stress, the carry-over effects of development under a decreasing water level could be observed as a boosted AOS in juveniles under benign conditions. Changes in water level and body size did not produce a substantial effect on the response to fasting-induced stress during the juvenile stage. This implies an ability of individuals with lower fitness to invest sufficient energy to maintain levels of oxidative stress that match those of larger-bodied individuals exposed to food deprivation. Overall, this study provides evidence that stressful conditions experienced during tadpole development can to some extent reflect on the fitness and oxidative stress of juveniles. Therefore, the hypothesis that metamorphic body size influences later life stages should be considered with caution. The questions that remain to be studied are: could the changes in the body size and oxidative stress of juveniles exposed to desiccation developmental stress seen in the laboratory have repercussions in harsher and more dynamic natural conditions; and the possible long-term or delayed adverse effects.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The animal study was reviewed and approved by the Ministry of Environmental Protection of the Republic of Serbia (Permit No. 353-01-2876/2019-04), Veterinary Directorate, Department of Animal Welfare, Ministry of Agriculture, Forestry and Water Management of the Republic of Serbia (Approval No. 323-07-08393/2020-05/4).

MP, TP, and NT contributed to the conception and design of the study. MP organized the database and performed the statistical analysis. MP and TP wrote the first draft of the manuscript. MP, TP, BG, SD, JG, AK, NT, TV, and TR performed the investigation and experimental procedures. AK, TV, and NT performed the resources and experimental setup. MP and NT supervised the work. All authors contributed to writing, reviewing and editing the final version of the manuscript and approved the submitted version.

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.