Local differences of thermal preferences in European common frog (Rana temporaria Linnaeus, 1758) tadpoles

Zoologischer Anzeiger 268 (2017) 47–54

Contents lists available at ScienceDirect

Zoologischer Anzeiger journal homepage: www.elsevier.com/locate/jcz

Research paper

Local differences of thermal preferences in European common frog (Rana temporaria Linnaeus, 1758) tadpoles Sanja Drakulic´ a , Heike Feldhaar b , Duje Lisiˇcic´ c , Mia Mioˇc c , Ivan Cizelj d , Michael Seiler b , Theresa Spatz e , Mark-Oliver Rödel a,∗ a

Museum für Naturkunde, Leibniz Institute for Evolution and Biodiversity Science, 10115 Berlin, Germany Department of Animal Ecology I, Bayreuth Center of Ecology and Environmental Research (BayCEER), University of Bayreuth, 95447 Bayreuth, Germany c Department of Animal Physiology, Faculty of Science, University of Zagreb, 10000 Zagreb, Croatia d Zoological Garden of Zagreb, 10000 Zagreb, Croatia e Department of Animal Ecology and Tropical Biology, University of Würzburg, 97074 Würzburg, Germany b

a r t i c l e

i n f o

Article history: Received 18 January 2017 Received in revised form 29 March 2017 Accepted 12 April 2017 Available online 15 April 2017 Corresponding Editor: Alexander Kupfer Keywords: Amphibians Behavioural thermoregulation Preferred temperatures Thermal adaptation

a b s t r a c t Physiological functions of ectotherms and thus their performance depend on environmental temperatures. Many ectotherms are capable of active thermoregulation, e.g. by selecting suitable microhabitats. However, this may be constrained by unavailability of favourable microhabitats or high energetic costs of thermoregulation. Thus, to achieve the optimal performance levels, adaptations to local thermal environments are of great importance. Due to the inability of leaving their aquatic habitat, larval anurans should especially benefit from local thermal adaptations. Rana temporaria is a widely distributed European anuran species, inhabiting a range of microhabitats, which makes it an excellent model to study the potential of local thermal adaptation. We raised R. temporaria tadpoles from Germany and Croatia under respective natural temperature fluctuations and three constant temperatures (15◦ , 20◦ and 25 ◦ C), and tested dependency of their thermal preferences on the developmental stage and temperature regime, and the population origin. Tadpoles of both origins selected higher temperatures towards the end of the developmental period, and their thermal preferences were affected by the developmental temperature. However, regardless of the developmental stage or treatment, tadpoles from warmer Croatia selected higher temperatures than tadpoles from colder Germany. This demonstrates the tendency of adjusting the sensitivity of physiological processes to local thermal conditions. © 2017 Elsevier GmbH. All rights reserved.

1. Introduction Well-adjusted body temperature is crucial for all organisms, since it affects all physiological functions. Ectotherms are not capable of producing significant quantity of body heat, and are thus dependent on the environmental temperature (Angilletta, 2009; Hillman et al., 2009). However, despite their dependence on external heating (or cooling) sources, many ectotherms are capable of actively maintaining their body temperature within a preferred range (Cowles and Bogert, 1944). Mechanisms involved in active thermoregulation may include morphological, physiological and/or

∗ Corresponding author at: Museum für Naturkunde, Leibniz Institute for Evolution and Biodiversity Science, Invalidenstraße 43, 10115 Berlin, Germany. ´ E-mail addresses: [email protected] (S. Drakulic), [email protected] (H. Feldhaar), [email protected] ´ [email protected] (I. Cizelj), [email protected] (M. Seiler), (D. Lisiˇcic), [email protected] (T. Spatz), [email protected] (M.-O. Rödel). http://dx.doi.org/10.1016/j.jcz.2017.04.005 0044-5231/© 2017 Elsevier GmbH. All rights reserved.

behavioural processes, such as selection of favourable microhabitat (Brattstrom, 1979; Huey, 1982; Huey and Pianka, 1977; Hutchison and Maness, 1979). Thermoregulation can enhance performance, and thus, influence fitness (Huey, 1982; Huey and Stevenson, 1979). Therefore, active thermoregulation should be under selection pressure (Angilletta et al., 2002) and organisms usually occur in the environments which best fulfill their temperature needs. However, achieving favourable temperatures is often limited by ecological constrains, such as unavailability of suitable microhabitats. Moreover, active thermoregulation can come at high costs, including energetic budgets, predation pressure, or lost opportunities for feeding or reproduction (Angilletta, 2009; Huey and Slatkin, 1976). These constrains may differ in different environments and, thus, lead to local intra-specific differences in thermal preferences (Angilletta et al., 2002; Huey and Bennett, 1987). Due to their vulnerability to desiccation, thermoregulation of amphibians is especially challenging (Brattstrom, 1979; Tracy,

48

S. Drakuli´c et al. / Zoologischer Anzeiger 268 (2017) 47–54

1976). The complex interplay of thermal and hygric requirements and the constraints imposed by the environment have intrigued eco-physiologists for decades (e.g. Hall and Root, 1930; Hillman et al., 2009; Tracy, 1976). The European common frog, Rana temporaria Linnaeus 1758; is an explosive breeder, spawning over only a few days in early-spring. It utilizes a variety of permanent and temporary water bodies, such as ponds, puddles, swamps and creeks for reproduction. It occurs from northern Spain to western Siberia, and from northern Scandinavia to northern Greece (Gollmann et al., 2014), and thus is found in a variety of habitats with differing thermal properties. It has been shown that R. temporaria exhibits thermoregulatory behaviour, such as basking or selection of favourable microhabitats (Sinsch, 1984; Vences et al., 2002). Strikingly, the literature reports very different thermal preferences of R. temporaria adults, in spite of the relatively restricted geographic and climatologically similar central European sites: e.g. 29.6 ± 0.51 ◦ C (Berlin, NE Germany; Strübing, 1954), ca. 10–20 ◦ C (northern Rhineland, NW Germany; Sinsch, 1984), or 19.4 ± 1.7 ◦ C (northern Poland; Köhler et al., 2011). Rühmekorf (1958) reported that temperatures ranging between 21.0 and 26.0 ◦ C are favourable for R. temporaria tadpoles. These differences between populations imply thermal preferences, adjusted to local environmental conditions. Adult frogs can move towards microsites best matching required climatic conditions within its habitat, while larval amphibians can select favourable water temperatures. However, larvae cannot escape their often thermally more homogenous aquatic environments. Thus, adjustments of physiological processes to local thermal environments would be particularly beneficial in the aquatic stage (Seebacher and Franklin, 2005). Herein, we focus on the thermal preferences of R. temporaria tadpoles from two geographically distant populations, one from the north of the Alps, Germany, and another from the south of the Alps, Croatia. These populations could exhibit differences in thermal preferences due to adaptations to local climate regimes or react plastically, as a consequence of acclimation to temperatures experienced in their respective environments. We aimed to test if local adaptation, acclimation, or both, direct the thermal preference of R. temporaria tadpoles. We raised tadpoles from Germany and Croatia under different constant temperatures, as well as under respective local temperature conditions of each population. We then tested their thermal preferences, across different developmental stages. We hypothesized that: (i) tadpoles from the warmer Croatia will choose higher temperatures compared to tadpoles from the cooler Germany, due to local adaptation towards higher temperatures; (ii) preferred temperatures will increase towards the end of the larval period, since final developmental stages are particularly vulnerable, due to e.g. decreased mobility (Dupré and Petranka, 1985; Wassersug and Sperry, 1977), and higher temperatures accelerate development (Ultsch et al., 1999); and (iii) tadpoles developing in higher developmental temperatures will select higher temperatures than tadpoles developing in lower temperatures, due to acclimation.

2. Materials and methods 2.1. Study sites In April 2013, we sampled eggs of 10 freshly laid clutches from a pond (FS06), in Steigerwald, northern Bavaria, Germany (49◦ 55 N, 10◦ 33 E, 409 m asl, see Grözinger et al., 2012, 2014), hereafter referred to as population GER. In March 2014, eggs from 10 clutches were collected from a pond (RJ01) in Medvednica (45◦ 53 N, 16◦ 00 E, 400 m asl), close to Zagreb, Croatia, hereafter referred to as population CRO. Both ponds were permanent and sit-

uated in mixed deciduous forest. The experiments were conducted in the ecological field station of the University of Würzburg in Fabrikschleichach, Germany (April–June 2013) and in the Zagreb Zoo, Croatia (March–May 2014). Both study sites belong to the temperate zone; however their climate differs. The climate in CRO is warmer, with an average annual temperature of 10.8 ◦ C. The coldest month is January (average temperature: 0 ◦ C), the warmest month is July (average temperature: 20.9 ◦ C; Maksimir weather station, 45◦ 49 N, 16◦ 02 E; Croatian Hydrometeorological Service). The mean annual temperature in the GER site is 8.2 ◦ C. January is the coldest (average temp: −0.6 ◦ C) and July the warmest month (average temp: 17.4 ◦ C; Ebrach weather station, 49◦ 51 N, 10◦ 30 E; Deutscher Wetterdienst).

2.2. Experimental procedures 2.2.1. Climate chambers and outdoor treatments After collection, eggs were kept in separate plastic containers filled with original pond water, at a constant temperature of 7–8 ◦ C (similar to the original environmental conditions) for five days. We then assigned them to one of four developmental treatments. Three developmental treatments provided constant temperatures of 15 ◦ C (hereafter referred as T15), 20 ◦ C (T20) and 25 ◦ C (T25), all being within the range of natural temperature variations in both populations. A fourth treatment mirrored the respective local environmental conditions, with natural temperature fluctuations (outdoor treatment, OT). After hatching and reaching developmental stage 25 (after Gosner, 1960: free swimming and feeding), we transferred the tadpoles to the experimental containers. In each of the climate chamber treatments, we placed six tadpoles each in a single plastic container, filled with 1.2 L of mixed deionized/spring water (pH = 6.5–7.0, conductivity = 150–200 ␮S/cm). From each population, we kept a total of 120 tadpoles (6 tadpoles × 20 containers) per climate chamber treatment (T15, T20 and T25); thus 360 tadpoles per population. Light conditions in the climate chambers were set to 16:8 L:D h, corresponding to the natural daylight rhythm. The outdoor treatments provided conditions that mimicked the respective natural environments. Here, ten tadpoles were placed into a plastic container with 5 L of the original pond water (GER: pH = 6.9, conductivity = 110 ␮S/cm, CRO: pH = 7.0, conductivity = 320 ␮S/cm). To buffer the environmental temperature variability, and ensure that temperature variations resemble those in the natural water sources, the volume in the outdoor treatment containers was higher than in the climate chambers’ treatments. Furthermore, the bottom of the container was covered with soil and leaf litter from the original location. Outdoor treatment containers were placed outside in the shade and covered with plastic gauze lids (grid diameter 1 mm) to exclude potential predators. In the outdoor treatments, we kept 300 tadpoles per population (10 tadpoles × 30 containers). In the following, we denote a particular developmental treatment of a particular population using a treatment-population label, e.g. T15GER for eggs/tadpoles developing at 15◦ , originating from Germany. Water quality (temperature, pH, conductivity) was tested regularly. Water in the climate chambers was exchanged two to three times per week, while in the outdoor treatments water was only replaced when necessary (pH out of 6.2–7.5, or CD >370 ␮S/cm; ® AL15MultiMeter Instrument; AquaLytic , Dortmund, Germany). Water temperature in each of the outdoor treatment containers ® was recorded every 3 h, using a thermologger (Theromochron ◦ iButtons© DS1922l, ±0.5 C; Embedded Data Systems, Lawrenceburg, Kentucky, USA). All tadpoles were fed ad libitum with ® commercial fish food (TetraTabiMin ; Tetra, Melle, Germany).

S. Drakuli´c et al. / Zoologischer Anzeiger 268 (2017) 47–54

49

Fig. 1. Aquatic temperature gradient setup for testing thermal preferences of Rana temporaria tadpoles. A thermal gradient (∼12–29 ◦ C) was established by an ice-cooled water flow system and an aquarium heater. The two meters long plastic half-pipe was filled with 2.5 cm of water; for further details see text.

2.2.2. Temperature preference experiment We tested the temperature preference of tadpoles using an aquatic thermal gradient setup (Bancroft et al., 2008; Lucas and Reynolds, 1967; Wu et al., 2007). The setup consisted of five twometer long half-pipes (11 cm in diameter) filled with spring water to a depth of approximately 2.5 cm (Fig. 1). The bottoms of the plastic half-pipes were covered with a ∼1 cm layer of aquarium pebbles. Pre-experiments showed that tadpoles behave normally and relaxed with such ground cover and do not search for hiding places. Water was heated with commercial aquarium heaters (50 W; TRIXIE Heimtierbedarf GmbH & Co., Tarp, Germany) at one side, and cooled at the other side of each pipe. The cooling system consisted of a universal aquarium pump (13 W, 1000 L/min; JBL ProFlow 1000, JBL GmbH & Co., Neuhofen, Germany) that pumped ice-cooled water through spiral metal tubes (10 mm in diameter). Thus, a thermal gradient of ∼12–29 ◦ C was established. To protect tadpoles from direct contact with the ice-cold metal spiral and the heaters, respectively, we installed plastic gauze net barriers. Before the start of the experiment, we randomly selected tadpoles from a single developmental treatment, measured their size (snout-vent-length (SVL), in mm), and determined the Gosner stage (Gosner, 1960). Tadpoles that exceeded the pre-defined Gosner stage (see below) were not used. One experimental run consisted of a single tadpole being placed in the central part of a half-pipe, and left for a period of two hours for gradient exploration and thermal preference selection. After two hours, we recorded the position of the tadpole and measured the water temperature at the exact spot (mercury thermometer ± 0.2 ◦ C; T-6000, Miller & Webber, Ridgewood, NY, USA). If the tadpole was swimming at the end of experimental time, or it was positioned less than one cm away from the protective nets (in the “sheltered” area), we excluded it from the analysis. Tadpoles were not fed during the course of the experiment. Tadpoles developing in the outdoor treatment were tested at three different developmental stages – at the early (free swimming and feeding tadpoles, Gosner stage 25–28), the middle (development of hind limbs, Gosner stage 32–35) and the late stage, just before the end of the larval period (development and emergence of the front limbs, Gosner stage 38–43; Gosner, 1960). Tadpoles from the constant temperature treatments (T15, T20 and T25) were tested at the late developmental stage (Gosner 38–43; see further). We wanted to test how population, developmental stage and developmental temperatures influenced the preferred temperatures of R. temporaria tadpoles. To test (i) whether tadpoles demonstrate the local adaptation of thermal preferences, we compared the preferred temperatures of tadpoles in the same Gosner stage of the outdoor treatment (Gosner 25–28, 32–35 or 38–43), and from the same constant temperature treatment (T15, T20 or T25; at Gosner stage 38–43), but originating from different popula-

tion (GER vs. CRO). To test (ii) whether preferred temperatures vary through development, we compared the preferred temperatures of tadpoles from different stages of the outdoor treatment (Gosner stage 25–28, 32–35 and 38–43), within the same population (GER or CRO). To test (iii) whether preferred temperatures are similar to the developmental ones, due to acclimation, we compared preferred temperatures from different developmental treatments (T15, T20 and T25, at Gosner stage 38–43), within the same population (GER or CRO). Furthermore, we wanted to assess whether the thermal preference trend, detected through developmental period in the outdoor treatments of GER population (see Section Results), is a consequence of acclimation to natural temperature fluctuations (Supplementary material, Fig. S1), or is indeed a characteristic of a specific developmental stage. Thus, we tested CRO tadpoles from a single constant temperature treatment (T15), at same three Gosner stages as the outdoor treatment (Gosner 25–28, 32–35 or 38–43). We selected the constant temperature treatment of 15 ◦ C, since this temperature was the most comparable to the mean temperature in the respective outdoor treatment (mean temperature of OT-CRO until G38–43 experiment: 13.9 ◦ C, Fig. S1). To determine the precision of the temperature selection in R. temporaria tadpoles, we followed the approach described by Dupré and Petranka (1985). They used the coefficient of variation (CV = SD/mean*100) of preferred temperatures as a measure of the temperature selection precision of four amphibian species, distinguishing strong (CV < 16.3%) from weak thermoregulators (CV > 22.5%; see also Wu et al., 2007). We calculated CV of preferred temperatures for each experimental treatment (Table 1). We tested all datasets for normality, using Shapiro-Wilks’ normality test. Since some of the datasets deviated from normal distribution, we used the Kruskal-Wallis rank sum test, with pairwise Wilcoxon rank sum post hoc test for the comparison of the preferred temperatures of tadpoles. FDR correction (Benjamini and Hochberg, 1995) was used to correct p-values for multiple comparisons. All calculations and analyses were conducted using R software environment for statistical computing and graphics (R Core Team, 2014; visualisations: ggplot2; Wickham, 2009).

2.3. Ethics statement Animals were cared for in accordance with guidelines on the animal care and use compiled by the American Society of Ichthyologists and Herpetologists (ASIH), The Herpetologists’ League (HL) and the Society for the Study of Amphibians and Reptiles (SSAR). Peter Krämer (higher nature conservation authority of Lower Franconia – “Regierung von Unterfranken”) approved the research in accordance with the Federal Conservation of Nature and Landscape Act (“Bundesnaturschutzgesetz”). In Croatia, R. temporaria is

50

S. Drakuli´c et al. / Zoologischer Anzeiger 268 (2017) 47–54

Table 1 Preferred temperatures of Rana temporaria tadpoles. Tadpoles originated from Germany (GER) and Croatia (CRO), and developed in either A. outdoor treatments (treatment; OT – natural temperature fluctuations) or B. constant temperatures of 15 ◦ C (T15), 20 ◦ C (T20) and 25 ◦ C (T25). Given are developmental stage (stages; Gosner, 1960), sample size (n), median (M), minimum (min) and maximum (max) preferred temperature values (◦ C) and coefficient of variation (CV, %). population A. outdoor treatments GER

CRO

treatment

stages

n

M (◦ C)

min (◦ C)

max (◦ C)

CV (%)

OT OT OT OT OT OT

25–28 32–35 38–43 25–28 32–35 38–43

49 44 37 50 42 40

17.6 18.8 22.0 20.4 21.3 24.2

14.8 15.2 16.5 16.6 15.5 17.0

19.6 22.8 26.0 23.0 27.0 27.4

7.51 7.88 10.08 7.69 11.58 11.01

38–43 38–43 38–43 38–43 38–43 38–43

30 20 26 39 33 35

23.0 21.1 19.4 24.4 21.4 21.0

14.0 15.8 16.8 17.8 17.5 15.2

26.6 25.2 24.0 26.8 27.0 24.4

15.74 10.66 11.33 8.65 10.45 8.39

B. constant temperature treatments T15 GER T20 T25 T15 CRO T20 T25

not protected by the law, so no sampling permits were necessary (Ministry of Environmental and Nature Protection, decision class: UP/I-612-07/13-48/107, permit nr. 515-07-1-1-1-14-6; Zagreb, 28 January 2014). Due to the German Protection of Animal Act (“Tierschutzgesetz”, http://www.gesetze-im-internet.de/tierschg/ BJNR012770972.html; §1/§7; 9 December 2010; last accessed on 13 August 2015) and Croatian Regulative on the protection of animals used for scientific purposes (http://narodne-novine.nn.hr/ clanci/sluzbeni/2013 05 55 1129.html; §1/§6f; 08 May 2013; last accessed on 13 August 2015), painless experiments and observations of vertebrates neither require permission nor disclosure. The vertebrates involved, R. temporaria tadpoles, experienced no pain, suffering, complaints or harm. Thus, no Institutional Animal Care and Use Committee (IACUC) or ethics committee approved this study as this was not required by German or Croatian law. 3. Results Temperature preferences of tadpoles were successfully measured with a total of 20–39 tadpoles (constant temperatures) and 37–50 tadpoles (outdoor treatments) per experimental treatment or developmental stage (summarized in Table 1). The tadpoles demonstrated a high precision of temperature selection (low coefficient of variation; CV < 16.3%) over all developmental stages and treatments. The CV of the outdoor treatments (including the three different developmental stages) varied from 7.5 to 10.1% in GER and from 7.7 to 11.6% in CRO. In the constant temperature treatments, the CV of the preferred temperatures ranged from 10.7 to 15.7% in GER tadpoles, and from 8.4 to 10.5% in CRO tadpoles. Regardless of developmental stage or treatment, tadpoles from CRO selected higher temperatures than tadpoles from GER, however without statistical support at T20 (Tables 1 and 2, Figs. 2 and 3). Tadpoles developing in the outdoor treatments selected higher temperatures towards the later developmental stages, in GER and CRO. We found significant differences among all developmental stages of the outdoor treatments (Gosner 25–28, 32–35 and 38–43), in both populations (Tables 1 A and 3 , Fig. 2). Developmental temperatures significantly influenced the thermal preferences of tadpoles (Tables 1 B and 4, Fig. 3). Preferred temperatures were negatively correlated with developmental temperatures of tadpoles developing in constant temperature treatments in both populations (Spearman’s rank correlation, GER: p-value = 0.001, Spearman’s ␳ = −0.36; CRO: p-value < 0.0001, Spearman’s ␳ = −0.56), i.e. tadpoles developing under lower temperatures selected higher temperatures in the gradient. Preferred temperatures of tadpoles from constant developmental temperature treatments in the GER population did not differ between

Table 2 Differences in preferred temperatures of Rana temporaria tadpoles from two different populations (GER vs. CRO). Tadpoles originated from Germany (GER) and Croatia (CRO), and developed in either A. outdoor treatments (treatment; OT) or B. constant temperatures (T15, T20, T25). We compared preferred temperatures of tadpoles from the outdoor treatments, in three different developmental stages (stages; Gosner stages 25–28, 32–35 or 38–43); and of tadpoles from constant temperature treatments, at the end of the developmental period (Gosner stage 38–43). Tadpoles from CRO selected higher temperatures than those from GER, regardless of developmental stage or treatment, however without statistical support in T20. Given are the results of Kruskal-Wallis rank sum test (degrees of freedom (df), chi-squared value (2 ) and p-value). stages

df

2

p-value

A. outdoor treatments 25–28 OT 32–35 OT 38–43 OT

1 1 1

49.13 24.55 7.61

<0.0001 <0.0001 0.006

B. constant temperature treatments T15 38–43 38–43 T20 T25 38–43

1 1 1

5.84 2.36 6.00

0.02 0.12 0.01

treatment

Statistically significant differences (p-values < 0.05) are shown in bold.

T15 and T20, but showed significant differences among other treatments. In the CRO population, temperature preferences differed significantly among all constant developmental temperature treatments (Tables 1B and 4). When comparing tadpoles from the outdoor treatments and constant temperature treatments (at Gosner 38–43), preferred temperatures of OT did not differ from T15, and were higher than T20 and T25, at both GER and CRO (Tables 1 and 4). When comparing CRO tadpoles of the same developmental stage, but developing in different temperature regimes (variable temperatures in OT-CRO vs. constant T15-CRO; Gosner 25–28, 32–35 and 38–43), we found no significant differences (Supplementary material; Tables S1 and S2, Fig. S2). 4. Discussion We hypothesized that tadpoles from different populations will show differences in thermal preferences, due to local adaptation. Indeed, tadpoles originating from warmer Croatia (CRO), selected higher temperatures compared to tadpoles from colder Germany (GER), regardless of developmental stage or treatment. We furthermore predicted that thermal preferences would differ in different developmental phases. Accordingly, tadpoles from both populations selected higher temperatures towards later developmental stages. Finally, we predicted that developmental temperatures will affect the thermal preferences, with tadpoles selecting temperatures similar to developmental ones. However, tadpoles developing

S. Drakuli´c et al. / Zoologischer Anzeiger 268 (2017) 47–54

51

Table 3 Differences in preferred temperatures among different developmental stages of Rana temporaria tadpoles, developing under natural temperature variations in the outdoor treatments. We compared preferred temperatures of tadpoles developing in outdoor treatments (OT) of the same population (either Germany: OT-GER or Croatia: OT-CRO), at three different developmental stages (Gosner stage 25–28, 32–35 and 38–43). Preferred temperatures increased towards later developmental stages in both populations. Given are results of Kruskal-Wallis rank sum test (degrees of freedom (df), chi-squared value (2 ) and p-value); and p-values after pairwise Wilcoxon rank sum post hoc test, with FDR p-value adjustment method for multiple comparisons. OT-GER Kruskal-Wallis rank sum test 2 = 72.63, df = 2, p-value < 0.0001 pairwise Wilcoxon rank sum post hoc test, p-values Gosner 25–28 Gosner 32–35 <0.0001 <0.0001 Gosner 38–43

OT-CRO 2 = 43.15, df = 2, p-value < 0.0001 Gosner 32–35 – <0.0001

Gosner 25–28 0.004 <0.0001

Gosner 32–35 Gosner 38–43

Gosner 32–35 – <0.0001

Statistically significant differences (p-values < 0.05) are shown in bold.

Table 4 Differences in preferred temperatures of Rana temporaria tadpoles developing under different treatments, at Gosner stage 38–43. We compared preferred temperatures of tadpoles originating from the same population (either Germany: GER or Croatia: CRO), but developing in different developmental treatments – in the outdoor treatments (OT) and under constant temperatures (T15, T20, T25). Given are results of Kruskal-Wallis rank sum test (degrees of freedom (df), chi-squared value (2 ) and p-value); and p-values after pairwise Wilcoxon rank sum post hoc test, with FDR p-value adjustment method for multiple comparisons. GER Kruskal-Wallis rank sum test 2 = 20.35, df = 3, p-value = 0.0001 pairwise Wilcoxon rank sum post hoc test, p-values T15 OT 0.895 – T15 0.038 0.106 T20 <0.0001 0.022 T25

CRO 2 = 42.40, df = 3, p-value < 0.0001 T20 – – 0.038

T15 T20 T25

OT 0.810 0.004 <0.0001

T15 – 0.004 <0.0001

T20 – – 0.004

Statistically significant differences (p-values < 0.05) are shown in bold.

Fig. 2. Preferred temperatures of Rana temporaria tadpoles in different stages (Gosner stages 25–28, 32–35 and 38–43) developing under naturally fluctuating temperatures (outdoor treatments, OT). Tadpoles originated from Croatia (CRO, light grey) and Germany (GER, dark grey). Significant differences between populations are indicated with an asterisk (Kruskal-Wallis rank sum test, p < 0.05; compare Tables 1 A, 2 A and 3).

at higher temperatures had lower thermal preferences, compared to tadpoles developing at lower temperatures, and vice versa. Animals are usually adapted to the temperatures they experience in their environment (Keller and Seehausen, 2012; Levinton, 1983). Higher temperatures selected by CRO-tadpoles are likely

Fig. 3. Preferred temperatures of advanced (Gosner stages 38–43) Rana temporaria tadpoles developing in constant temperature treatments of 15 ◦ C (T15), 20 ◦ C (T20) and 25 ◦ C (T25). Tadpoles originated from Croatia (CRO, light grey) and Germany (GER, dark grey). Significant differences between populations are indicated with an asterisk (Kruskal-Wallis rank sum test, p < 0.05; compare Tables 1 B, 2 B and 4).

a consequence of physiological adaptations to the local thermal conditions, since environmental temperatures in CRO are higher. Similar effect was demonstrated studding six populations of a frog Quasipaa spinosa (David, 1875), where preferred temperatures correlated with the environmental ones, along the latitudinal gradient (Zheng and Liu, 2010). Furthermore, aquatic ectotherms

52

S. Drakuli´c et al. / Zoologischer Anzeiger 268 (2017) 47–54

living under higher environmental temperatures may face higher desiccation risk. This is particularly important for species using temporary or semi-permanent environments, such as many R. temporaria breeding sites, where desiccation can cause complete mortality of a year’s tadpole cohort. It is known that R. temporaria tadpoles can exhibit higher developmental rates in environments with increased desiccation risks (Lind et al., 2010; Lind and Johannson, 2007). Due to higher temperatures, especially in the summer months, desiccation risk is generally higher for CRO population, and a shorter developmental period should be favoured. Since higher temperatures lead to increased development of anuran larvae (Atkinson, 1994; Ultsch et al., 1999), including the species and populations tested herein (see Drakulic´ et al., 2016 and Table S3), the selection of higher temperatures may thus be a mechanism of accelerating development. Consistency of higher thermal preferences of CRO in common environments (constant temperature treatments) implies a genetic source of this effect and thus local adaptation. Thermal adaptations can increase the performance, and consequentially fitness, in a respective environment (Angilletta et al., 2002). For example, box terrapins (Terrapene ornata (Agassiz, 1857)) from higher latitudes (colder environment) had lower thermal preference than conspecifics from lower latitudes, which allowed a prolongation of the daily activity period in the colder environment (Ellner and Karasov, 1993). We predicted that not only origin, but likewise developmental stage will influence the thermal preference of tadpoles. A developmental shift in thermal preferences was previously shown in other amphibian species, e.g. Lithobates catesbeianus (Shaw, 1802) (Hutchison and Hill, 1978), Rana cascadae Slater, 1939 (Wollmuth et al., 1987), Anaxyrus americanus (Holbrook, 1836) and Ambystoma texanum (Matthes, 1855) (Dupré and Petranka, 1985). We tested tadpoles from the outdoor treatment at early, middle and late developmental stage. Tadpoles selected the lowest temperatures at the earliest stage, while the highest temperatures were selected at the late stage, at both GER and CRO. It is possible that tadpoles choose higher temperatures towards the later developmental stages in response to respective natural temperature variation (Fig. S1). However, when the effect of environmental temperature increment was removed, by raising tadpoles under constant temperature treatment (at T15-CRO), the pattern of an increase in preferred temperatures with development persisted. What could be the possible sources of variation in thermal preferences through development? Tadpoles are practically feeding machines; adapted for the efficient exploitation of temporarily available rich nutrient sources (Altig et al., 2007; Wassersug, 1984). Lower temperatures usually lead to decreased developmental rates of ectotherms, but result in a bigger size (Atkinson, 1994, 1995; Ultsch et al., 1999). Therefore, lower temperature selection in the early stages would decelerate the development, allowing tadpoles to invest time and resources in growth and reserves deposition. Approaching the completion of metamorphosis, anuran larvae undergo radical changes of the body (Brown and Cai, 2007). Hind limb development in the later stages decreases hydrodynamic properties of the body, affecting the swimming performance of tadpoles and making them especially prone to predation (Wassersug, 1997; Wassersug and Sperry, 1977). Therefore, it is beneficial to accelerate the development during this period (Dupré and Petranka, 1985; Wollmuth et al., 1987). Moreover, intense metamorphic changes, including e.g. remodelling of organs systems, formatting new organs or restructuring the skeleton (Brown and Cai, 2007), likely require increased metabolic activity (Hutchison and Hill, 1978). Therefore, increased thermal preferences could be the behavioural response to increased vulnerability and metabolic requirements in the later developmental stages. We further hypothesised that developmental temperatures will influence the thermal preferences of tadpoles, with higher temper-

atures chosen by the animals developing in higher temperature treatments, and vice versa, due to thermal acclimation. A similar effect was previously reported for other larval anurans (e.g. Lithobates catesbeianus, Hutchison and Hill, 1978; Wells, 2007). However, temperatures selected by the tadpoles from constant temperature treatments (T15, T20, T25, at the late Gosner stage) were negatively correlated with the developmental ones. While the thermal preferences of T20 tadpoles were comparable with developmental temperature, in both populations, preferences of T15 were higher, and of T25 were lower than their developmental temperatures. As already outlined, advanced tadpoles stages are especially prone to predation. Thus, tadpoles developing at lower temperatures (T15, as well as in the outdoor treatments; mean temperatures during development: CRO: 13.9 ◦ C, GER: 12.4 ◦ C; Fig. S1) selected relatively high temperatures, probably to shorten the period of increased vulnerability. However, increased developmental temperature of amphibian larvae usually results in a smaller final size (Atkinson, 1994, 1995; Ultsch et al., 1999) and can lead to reduced body condition (Dittrich et al., 2016). Furthermore, due to higher intensity of cellular processes (e.g. transcription and translation), it may cause developmental errors, resulting in e.g. tumors initiation, weaker immune response or reduced stress tolerance (Arendt, 1997). Another potential limitation of increased temperatures is decreased oxygen availability. Higher temperatures lead to reduced levels of oxygen dissolved in water, while at the same time they increase metabolic rates, and thus oxygen requirements of tadpoles (Gatten et al., 1992; Wells, 2007). The selection of lower temperatures after acclimation to the temperatures above their thermal preferences was previously demonstrated in e.g. iguanid lizards (Mueller, 1970; Wilhoft and Anderson, 1960), or other anurans (e.g. Lithobates catesbeianus; Hutchison and Hill, 1978; Lillywhite, 1971). Since environmental temperatures as high as 25◦ are not commonly available during the earliest developmental stages in our populations, the potential of acclimation to temperatures this high might be limited. Thus, it is probable that the physiological rates of tadpoles developing under constantly high temperatures (T25) were simply too high, and tadpoles selected lower temperatures to decrease them. Therefore, temperature selection of tadpoles from our experiment demonstrates the tendency of developmental rate optimization. A similar effect was shown in the study of Kealoha Freidenburg and Skelly (2004) comparing thermal preferences of Lithobates sylvaticus (Le Conte, 1825) tadpoles from differing habitats, on a microgeographical scale. Tadpoles originating from shaded, cooler ponds selected higher temperatures (20.4 ± 0.3 ◦ C), with higher precision of temperature selection, compared the ones from warmer, sun-exposed ponds (17.6 ± 1.2 ◦ C). Since temperatures in the shaded ponds were lower, the tadpoles from this environment presumably selected higher temperatures, seizing the opportunity to maximize their growth and developmental rates, while those from sun-exposed ponds did not demonstrate precise behavioural thermoregulation. It is not probable that natural selection influences life-history traits independently. More likely, selective pressure acts on trait complexes, leading to an overall increased performance, and consequently, fitness of organisms, resulting in thermal coadaptation (Angilletta, 2009). Thus, thermally coadapted organisms should exhibit similarities in, for example, field body temperatures, preferred temperatures and thermal optima for e.g. locomotion (Blouin-Demers et al., 2003). A study conducted on the same R. temporaria populations, focusing on the duration of development and jumping performance of metamorphs (Drakulic´ et al., 2016), revealed that tadpoles from the colder environment (GER) have a longer developmental period than those from the warmer (CRO), even when tadpoles are raised in common environments (constant temperature treatments), indicating a genetic source of this variation. Furthermore, thermal sensitivity of jumping performance

S. Drakuli´c et al. / Zoologischer Anzeiger 268 (2017) 47–54

differed between populations. Froglets from colder GER were able to achieve the maximal jumping performance under lower environmental temperatures (20 ◦ C) compared to froglets from warmer CRO (25 ◦ C). Therefore, the association of preferred temperatures, developmental rates and thermal sensitivity of jumping performance implies that thermal physiology and behavioural thermoregulation possibly adapted in the same direction, in a regard to the respective thermal conditions of these populations (Angilletta, 2009; Huey and Bennet, 1987). 5. Conclusion Our results showed the active behavioural modification of body temperature in Rana temporaria tadpoles. Furthermore, we demonstrated a significant effect of population origin, developmental stage and developmental temperature on the thermal preferences of tadpoles. These results indicate that the two populations are locally adapted to different thermal environments. Thus, differing strategies, including local adaptation and/or behavioural compensation, may be developed as the result of the complex interactions of species-specific requirements and the ecological constrains imposed by the respective environments. Acknowledgements We thank Carolin Dittrich, Juliane Huster, Julian Rieß, Mar´ Dean Karaica, Boris Lauˇs and Zˇ eljka vin Schäfer, Mladen Drakulic, Komorski for their help with field work and experiments. HansJoachim Poethke and the staff of the Ecological Field Station Fabrikschleichach, as well as the Zoological garden of Zagreb, gave access to the research infrastructure under their care. Many thanks to Vedran Franke and Petar Glaˇzar for help with statistical analysis. We thank Bayerische Staatsforsten and Public Institution Nature Park Medvednica for permission to work in the forests under their administration. We are grateful to two anonymous reviewers for their constructive and helpful comments. German Academic Exchange Service – DAAD supported this study, awarding Research Grant for Doctoral Candidates and Young Academics and Scientists to S.D. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jcz.2017.04.005. References Agassiz, L., 1857. Contributions to the Natural History of the United States of America. First Monograph. Volume I, Part 2. North American Testudinidae. Little, Brown & Co., Boston, pp. 233–452. Altig, R., Whiles, M., Taylor, C., 2007. What do tadpoles really eat? Assessing the trophic status of an understudied and imperiled group of consumers in freshwater habitats. Freshw. Biol. 52, 386–395. Angilletta, M.J., Niewiarowski, P.H., Navas, C.S., 2002. The evolution of thermal physiology in ectotherms. J. Therm. Biol. 27, 249–268. Angilletta, M.J., 2009. Thermal Adaptation: A Theoretical and Empirical Synthesis. Oxford University Press, Oxford. Arendt, J., 1997. Adaptive intrinsic growth rates: an integration across taxa. Q. Rev. Biol. 72, 149–177. Atkinson, D., 1994. Temperature and organism size: a biological law for ectotherms? Adv. Ecol. Res. 25, 1–58. Atkinson, D., 1995. Effects of temperature on the size of aquatic ectotherms: exceptions to the general rule. J. Therm. Biol. 20, 61–74. Bancroft, B.A., Baker, N.J., Searle, C.L., Garcia, T.S., Blaustein, A.R., 2008. Larval amphibians seek warm temperatures and do not avoid harmful UVB radiation. Behav. Ecol. 19, 879–886. Benjamini, Y., Hochberg, Y., 1995. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. R. Stat. Soc. Ser. B (Stat. Methodol.) 57, 289–300.

53

Blouin-Demers, G., Weatherhead, P.J., McCracken, H.A., 2003. A test of the thermal coadaptation hypothesis with black rat snakes (Elaphe obsoleta) and northern water snakes (Nerodia sipedon). J. Therm. Biol. 28, 331–340. Brattstrom, B., 1979. Amphibian temperature regulation studies in the field and laboratory. Am. Zool. 356, 345–356. Brown, D., Cai, L., 2007. Amphibian metamorphosis. Dev. Biol. 306, 20–33. Cowles, R., Bogert, C., 1944. A preliminary study of the thermal requirements of desert reptiles. Bull. Am. Mus. Nat. Hist. 83, 261–296. Croatian Meteorological and Hydrological Service. http://meteo.hr/index en.php (Accessed 17 January 2017). David, A., 1875. Journal de mon troisième voyage d’exploration dans l’empire Chinoise, Volume II. Hachette et cie, Paris. Deutscher Wetterdienst. http://www.dwd.de/DE/Home/home node.html (Accessed 17 January 2017). ´ S., Schellenberg, M., Thein, J., Rödel, M.-O., 2016. Some like it Dittrich, C., Drakulic, hot? Developmental differences in yellow-bellied toad tadpoles (Bombina variegata) from geographically close but different habitats. Can. J. Zool. 94, 69–77. ´ S., Feldhaar, H., Lisiˇcic, ´ D., Mioˇc, M., Cizelj, I., Seiler, M., Spatz, T., Rödel, Drakulic, M.-O., 2016. Population-specific effects of developmental temperature on body condition and jumping performance of a wide-spread European frog. Ecol. Evol. 6, 3115–3128. Dupré, R., Petranka, J., 1985. Ontogeny of temperature selection in larval amphibians. Copeia 1985, 462–467. Ellner, L., Karasov, W., 1993. Latitudinal variation in the thermal biology of ornate box turtles. Copeia 1993, 447–455. Gatten, R.E., Miller, K., Full, R.J., Feder, M.E., Burggren, W.W., 1992. Energetics of amphibians at rest and during locomotion. In: Feder, M.E., Burggren, W.W. (Eds.), Environmental Physiology of the Amphibians. University of Chicago Press, Chicago, pp. 314–377. Gollmann, B., Borkin, L., Grossenbacher, K., Weddeling, K., 2014. Rana temporaria Linnaeus 1758 – Grassfrosch. In: Grossenbacher, K. (Ed.), Handbuch der Reptilien und Amphibien Europas. Aula Verlag, Wiebelsheim, pp. 304–437. Gosner, K.L., 1960. A simplified table for staging anuran embryos and larvae with notes on identification. Herpetologica 16, 183–190. Grözinger, F., Wertz, A., Thein, J., Feldhaar, H., Rödel, M.-O., 2012. Environmental factors fail to explain oviposition site use in the European common frog. J. Zool. 288, 103–111. Grözinger, F., Thein, J., Feldhaar, H., Rödel, M.-O., 2014. Giants, dwarfs and the environment – metamorphic trait plasticity in the common frog. PLoS One 9 (3), e89982. Hall, F.G., Root, R.W., 1930. The influence of humidity on the body temperature of certain poikilotherms. Biol. Bull. 58, 52–58. Hillman, S.S., Withers, P.C., Drewes, R.C., Hillyard, S.D., 2009. Ecological and environmental physiology of amphibians. In: Bruggem, W. (Ed.), Ecological and Environmental Physiology Series. Oxford University Press, Oxford. Holbrook, J.E., 1836. North American Herpetology; Or Description of the Reptiles Inhabiting the United States, Volume I. J. Dobson, Philadelphia. Huey, R.B., Bennett, A., 1987. Phylogenetic studies of coadaptation: preferred temperatures versus optimal performance temperatures of lizards. Evolution 41, 1098–1115. Huey, R.B., Pianka, E., 1977. Seasonal variation in thermoregulatory behavior and body temperature of diurnal Kalahari lizards. Ecology 58, 1066–1075. Huey, R.B., Slatkin, M., 1976. Cost and benefits of lizard thermoregulation. Q. Rev. Biol. 51, 363–384. Huey, R.B., Stevenson, R., 1979. Integrating thermal physiology and ecology of ectotherms: a discussion of approaches. Am. Zool. 19, 357–366. Huey, R.B., 1982. Temperature, physiology, and the ecology of reptiles. Biol. Reptiles 12, 25–91. Hutchison, V., Hill, L., 1978. Thermal selection of bullfrog tadpoles (Rana catesbeiana) at different stages of development and acclimation tempeatures. J. Therm. Biol. 3, 57–60. Hutchison, V., Maness, J., 1979. The role of behavior in temperature acclimation and tolerance in ectotherms. Am. Zool. 384, 367–384. Köhler, A., Sadowska, J., Olszewska, J., Trzeciak, P., Berger-Tal, O., Tracy, C., 2011. Staying warm or moist? Operative temperature and thermal preferences of common frogs (Rana temporaria), and effects on locomotion. Herpetol. J. 21, 17–26. Kealoha Freidenburg, L., Skelly, D., 2004. Microgeographical variation in thermal preference by an amphibian. Ecol. Lett. 7, 369–373. Keller, I., Seehausen, O., 2012. Thermal adaptation and ecological speciation. Mol. Ecol. 21, 782–799. Le Conte, J.E., 1825. Remarks on the American species of the genera Hyla and Rana. Ann. Lyceum Nat. Hist. N. Y. 1, 278–282. Levinton, J., 1983. The latitudinal compensation hypothesis: growth data and a model of latitudinal growth differentiation based upon energy budgets. I. Interspecific comparison of Ophryotrocha (Polychaeta: Dorvilleidae). Biol. Bull. 165, 686–698. Lillywhite, H., 1971. Temperature selection by the bullfrog, Rana catesbeiana. Comp. Biochem. Physiol. A 40, 213–227. Lind, M.I., Johannson, F., 2007. The degree of phenotypic plasticity is correlated with the spatial environmental heterogeniety experienced by island populations of Rana temporaria. J. Evol. Biol. 20, 1288–1297. Lind, M.I., Ingvarsson, P.K., Johansson, H., Hall, D., Johansson, F., 2010. Gene flow and selection on phenotypic plasticity in an island system of Rana temporaria. Evolution 65, 684–697.

54

S. Drakuli´c et al. / Zoologischer Anzeiger 268 (2017) 47–54

Linnaeus, C., 1758. Systema naturae per regna tria naturae: secundum classes, ordines, genera, species, cum characteribus, differentiis, synonymis, locis, Volume I. L. Salvii, Stockholm. Lucas, E.A., Reynolds, W.A., 1967. Temperature selection by amphibian larvae. Physiol. Zool. 40, 159–171. Matthes, B., 1855. Die Hemibatrachier im Allgemeinen und die Hemibatrachier von Nord-Amerika im Speciellen. Allg. Deutsche Nat. Zeit. NS 1, 249–280. Mueller, C., 1970. Temperature acclimation in two species of Sceloporus. Herpetologica 26, 83–85. R Core Team, 2014. R: A Language and Environment for Statistical Computing. http://www.r-project.org/ (Accessed 17 January 2017). Rühmekorf, E., 1958. Beiträge zur Ökologie Mitteleuropäaischer Salientia II: Temperaturwahl der Larven. Z. Morphol. Ökol. Tiere (Zoomorphol.) 47, 20–36. Seebacher, F., Franklin, C.E., 2005. Physiological mechanisms of thermoregulation in reptiles: a review. J. Comp. Physiol. B Biochem. Syst. Environ. Physiol. 175, 533–541. Shaw, G., 1802. General zoology or systematic natural history. Part 1. Amphibia, Volume III. T. Davison, London. Sinsch, U., 1984. Thermal influences on the habitat preference and the diurnal activity in three European Rana species. Oecologia 64, 125–131. Slater, J.R., 1939. Description and life-history of a new Rana from Washington. Herpetologica 1, 145–149. Strübing, H., 1954. Über Vorzugstemperaturen von Amphibien. Z. Morphol. Ökol. Tiere (Zoomorphol.) 43, 357–386. Tracy, C., 1976. A model of the dynamic exchanges of water and energy between a terrestrial amphibian and its environment. Ecol. Monogr. 46, 293–326. Ultsch, G.R., Bradford, D.F., Freda, J., 1999. Physiology: coping with the environment. In: McDiarmid, R.W., Altig, R. (Eds.), Tadpoles: The Biology of Anuran Larvae. The University of Chicago Press, Chicago and London, pp. 189–214.

Vences, M., Galán, P., Vieites, D., Puente, M., Oetter, K., Wanke, S., 2002. Field body temperatures and heating rates in a montane frog population: the importance of black dorsal pattern for thermoregulation. Ann. Zool. Fenn. 39, 209–220. Wassersug, R., Sperry, D., 1977. The relationships of locomotion to differential predation on Pseudacris triseriata (Anura: Hylidae). Ecology 58, 830–839. Wassersug, R., 1984. Why tadpoles love fast-food? Nat. Hist. 93, 60–69. Wassersug, R., 1997. Where the tadpole meets the world – observations and speculations on biomechanical and biochemical factors that influence metamorphosis in anurans. Am. Zool. 136, 124–136. Wells, K.D., 2007. The Ecology and Behavior of Amphibians. Oxford University Press, Oxford. Wickham, H., 2009. ggplot2: Elegant Graphics for Data analysis. Springer-Verlag, New York, New York. Wilhoft, D., Anderson, J., 1960. Effect of acclimation on the preferred body temperature of the lizard, Sceloporus occidentalis. Science 131, 959–1960. Wollmuth, L., Crawshaw, L., Forbes, R., Grahn, D., 1987. Temperature selection during development in a montane anuran species, Rana cascadae. Physiol. Zool. 60, 472–480. Wu, H.-J., Yen, C.-F., Kam, Y.-C., 2007. Metabolic compensation and behavioral thermoregulation of subtropical rhacophorid (Polypedates megacephalus) tadpoles in container habitats. Comp. Biochem. Physiol. B 146, 101–106. Zheng, R.-Q., Liu, C.-T., 2010. Giant spiny-frog (Paa spinosa) from different populations differ in thermal preference but not in thermal tolerance. Aquat. Ecol. 44, 723–729.