General and Comparative Endocrinology 195 (2014) 40–46

Contents lists available at ScienceDirect

General and Comparative Endocrinology

journal homepage: www.elsevier.com/locate/ygcen

Changes in plasma angiotensin II, aldosterone, arginine vasotocin, corticosterone, and electrolyte concentrations during acclimation to dry condition and seawater in the crab-eating ⇑ Minoru Uchiyama a, , Sho Maejima a, Marty K.S. Wong b, Narin Preyavichyapugdee c, Chaitip Wanichanon d, Susumu Hyodo b, Yoshio Takei b, Kouhei Matuda a a Department of Life and Environmental Science, Graduate School of Science and Engineering, University of Toyama, Toyama 930-8555, Japan b Department of Marine Biosciences, Atmosphere and Ocean Research Institute, The University of Tokyo, 5-15 Kashiwanoha, Kashiwa, Chiba 277-8564, Japan c Faculty of Sciences and Agricultural Technology, Silpakorn University, Petchaburi IT Campus, Petchaburi 76120, Thailand d Department of Anatomy, Faculty of Science, Mahidol University, Rama VI Road, Bangkok 10400, Thailand article info abstract

Article history: The crab-eating frog Fejervarya cancrivora inhabits mangrove swamps and marshes in Southeast Asia. In Received 26 March 2013 the present study, circulating angiotensin II (Ang II), aldosterone (Aldo), arginine vasotocin (AVT), and Revised 15 October 2013 corticosterone (Cort) concentrations as well as various blood parameters were studied under osmoti- Accepted 17 October 2013 cally stressful conditions. Following acclimation to hyperosmotic seawater and dry condition for 5 days, Available online 30 October 2013 body weight was significantly decreased. Under both conditions, plasma Na+,ClÀ, and urea concentra- tions, hematocrit values (Ht; blood volume indicator), and osmolality were significantly increased. Keywords: Dehydration associated with hypovolemic and hyperosmotic states of body fluids was induced during Renin–angiotensin system acclimation to hyperosmotic seawater and dry condition in the crab-eating . Ang II, Aldo, AVT, and Arginine vasotocin Acclimation to seawater Cort were maintained within relatively narrow concentration ranges in the control frogs; however, in Hypovolemia frogs under dry and hyperosmotic seawater conditions, large variations were observed among individ- Hyperosmolemia uals in each group. Mean plasma Ang II and Aldo concentrations significantly increased in hyperosmotic Crab-eating frog (Fejervarya cancrivora) seawater-acclimated and desiccated frogs. Although mean plasma AVT concentrations in dehydrated frogs of both the groups were approximately 2.0–3.5 times higher than those in the control frogs, the differences were not significant because of the variation. There was a significant correlation between plasma osmolality and AVT as well as Ang II but not Aldo. A significant correlation was also observed between Ht and AVT as well as Ang II. Plasma Ang II was significantly correlated with plasma Aldo. These results indicate that the crab-eating frogs may exhibit similar physiological responses to both seawater-acclimated and dry conditions. It appears that under dehydrated conditions, osmoregu- latory mechanisms participate in stabilization of the situation. The renin–angiotensin system may have pivotal roles in body fluid regulation under volemic and osmotic stress in the Fejervarya species with unique osmoregulation. Ó 2013 Elsevier Inc. All rights reserved.

1. Introduction 1961; Uchiyama et al., 1990). When acclimated to seawater, urea synthesis initially increased in the liver, followed by a significant The crab-eating frog (Fejervarya cancrivora) is the only extant increase in urea concentrations in the plasma, liver, and muscle species that inhabits mangrove swamps along the (Wright et al., 2004). Southeast Asian coast (Gordon et al., 1961; Uchiyama et al., According to previous studies, osmoregulation in is 1990). Tolerance experiments have revealed that F. cancrivora primarily controlled by four hormonal systems, namely the hypo- can tolerate brackish water, and by a stepwise increase in salinity, thalamo–neurohypophysial system, the renin–angiotensin– aldo- survive in 75% or higher seawater concentrations (Gordon et al., sterone system (RAAS), the adrenocorticotropic hormone (ACTH)–corticosteroid system and the natriuretic peptides system; however, details remain unclear (see review by Uchiyama and Abbreviations: Ang II, angiotensin II; Aldo, aldosterone; AVT, arginine vasotocin; Konno, 2006). Although many in vivo and in vitro studies have Cort, corticosterone. ⇑ Corresponding author. Fax: +81 76 445 6549. examined the physiological effects of these hormones on osmoreg- E-mail address: [email protected] (M. Uchiyama). ulation, only a few studies have measured plasma osmoregulatory

0016-6480/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ygcen.2013.10.013 M. Uchiyama et al. / General and Comparative Endocrinology 195 (2014) 40–46 41 hormone concentrations in amphibians (see review by Uchiyama It is therefore interesting to determine whether any hormones and Konno, 2006). Elevated plasma arginine vasotocin (AVT) con- are effective and how they act in the osmoregulation of amphibi- centrations were measured in anuran species subjected to hyper- ans inhabiting a brackish water environment. The present study osmotic or hypovolemic stimuli using radioimmunoassay (Konno investigated the effects of desiccation, hyperosmotic and hypoos- et al., 2005; Nouwen and Kühn, 1985). Although we suggested that motic conditions, with a focus on plasma Ang II, AVT, and cortico- there is a significant correlation between plasma osmolality and steroids in the crab-eating frog. AVT concentration in the cane (or marine) toad Bufo marinus (Kon- no et al., 2005), hypernatremia had no influence on the systemic concentration of AVT in the marsh frog ridibunda (Nouwen 2. Materials and methods and Kühn, 1985). Plasma Angiotensin II (Ang II) concentrations did not change significantly in the terrestro–fossorial toad Scaphi- 2.1. opus couchii during intracellular and extracellular dehydration treatments (Johnson and Propper, 2000; Mayer and Propper, Adult male and female crab-eating frogs (F. cancrivora Graven- 2000). In contrast, we found a significant correlation between the horst 1829; formerly known as R. cancrivora) were collected in plasma volume and Ang II concentration in B. marinus (Konno the field at Ban Laem, Thailand, near the Gulf of Thailand. The frogs et al., 2005). It is known that Aldosterone (Aldo) and Corticosterone were identified as the mangrove type of F. cancrivora based on (Cort) increase sodium transport across the skin, urinary bladder, mitochondrial genome data (Kurniawan et al., 2011). They typi- and kidney in amphibians (Stiffler et al., 1986; Van Driessche and cally inhabit brackish water of mangrove forests of Southeast Asia Zeiske, 1985). However, serum Aldo and Cort concentrations did but can survive in fresh water and can also be acclimated to 75% not differ significantly between tap water-acclimated control and seawater or higher. Captured frogs were transported by air in April hyperosmotic-acclimated (solution of 1.2% NaCl, urea, or mannitol) 2010 and May 2011 from Bangkok to the University of Toyama, Ja- Xenopus laevis (Katz and Hanke, 1993). These results indicate that pan, with a capture and export certificate issued by the Thailand AVT, Ang II, and Aldo are involved in osmoregulation of body fluids Department of Fisheries. Thirty-four frogs, weighing 30.5 ± 1.6 g and that there may be interspecies differences in hormonal control [mean ± standard error of the mean (SEM)], were used in the of osmoregulation in amphibians. experiment. Two or three frogs were housed in a plastic container

Fig. 1. Changes in body weight, hematocrit (Ht) value, and plasma components under various environmental conditions. Mean percent changes in body weight (A), changes in Ht (B), plasma osmolality (C), plasma urea (D), and plasma sodium and chloride concentrations (E). Values are means ± SEM. Data were evaluated by one-way ANOVA followed by Bonferroni test (⁄P < 0.05 and ⁄⁄P < 0.01). Different superscripts indicate statistical significance. Numbers in parentheses indicate the number of animals used for measurement of parameters. Abbreviations: Na+, sodium concentration; ClÀ, chloride concentration. Hypersaline and hyposaline indicate hyperosmotic and hypoosmotic seawater groups, respectively. 42 M. Uchiyama et al. / General and Comparative Endocrinology 195 (2014) 40–46

(30 Â 30 Â 15 cm) containing moist soil in a room maintained at Plasma osmolality was measured using an osmometer (Vapro 24 ± 2 °C. They were kept on a 12-h light: 12-h dark cycle and osmometer 5520, Wescor Biomedical Systems, UT, USA). Plasma had free access to 20% seawater. They were fed insects, usually sodium and chloride concentrations were measured using an ABL twice a week. 625 analyzer (Radiometer, Copenhagen, Denmark). Plasma urea concentration was measured using the Wako Urea NB test (Wako 2.2. Acclimation experiment Pure Chemical Industries, Japan) in vitro enzymatic colorimetric method. In the control group, the frogs were maintained on moist soil and were allowed free access to 20% seawater. The frogs in the 2.4. Hormone measurements experimental group were divided into three subgroups: the dry group was kept on dry sponges and had no access to water, the Plasma Aldo and Cort concentrations were assayed using the hypotonic seawater group was immersed in 20% seawater, and Aldosterone EIA Kit and Corticosterone EIA Kit, respectively (Cay- the hypertonic seawater group was diurnally acclimated by a step- man Chemical Company, MI, USA), with a microplate reader (Bio- wise increase in salinity from 20%, 40%, and 60% to a final concen- Rad Laboratories, CA, USA). Plasma Ang II and AVT concentrations tration of 80% seawater for 2 days. The frogs in each group were no were measured using a radioimmunoassay, as described previously fed during the 5-day experiment. Acclimated solutions were (Konno et al., 2005). In brief, Ang II and AVT were iodinated with refreshed daily during the treatments, and the osmolality and 125I using the lactoperoxidase method. Plasma samples were ex- sodium concentration of solutions were measured each time using tracted by the addition of an equal volume of cold acidic acetone an osmometer (Osmostat OM-6020, Kyoto Daiichi Kagaku, Japan) (acetone: H2O:1 M HCl = 40:5:1). The extract was dried in a centrif- and an atomic absorption spectrophotometer (Hitachi Instruments ugal concentrator and dissolved in RIA buffer (10 mM phosphate Service, Japan), respectively. buffer, 140 mM NaCl, 0.1% NaN3,10mMe-amino-n-caproic acid, 40 mM 2 K-EDTA and 0.05% Triton X-100, pH 7.4), containing 1% 2.3. Blood sampling and analysis of plasma parameter bovine serum albumin (BSA). After conjugation with bovine thyro- globulin, antisera were raised in rabbit against eel [Asp1–Val5] Ang After 5 days of treatment, the frogs were anesthetized with II, which is a homologous amino acid sequence of Ang II in bullfrog diethyl ether. Blood samples were collected by cardiac puncture (Hasegawa et al., 1983), and arginine vasopressin (AVP). Both anti- either in heparinized 1-mL syringes or hematocrit (Ht) capillaries sera were used at a final dilution of 1:250,000. Plasma samples to measure electrolyte and hormone concentrations. All the sam- (100 lL) were incubated with the antiserum (200 lL) and 0.1 lL ples were immediately placed in ice and then centrifuged at of normal rabbit serum for 20 h at 4 °C. In total, 50 lL of iodinated 2000g for 20 min at 4 °C. A part of the plasma sample was used Ang II and AVT (9500–10,000 cpm) was then added and further to measure osmolality and electrolyte concentrations, and the incubated for 24 h at 4 °C. The bound hormone was precipitated remaining plasma was stored at À30 °C until the analysis of by addition of 100 lL of goat anti-rabbit IgG serum diluted 1:60 hormones. with RIA buffer and 100 lL of 16% polyethylene glycol. After

Fig. 2. Changes in plasma Ang II (A), AVT (B), Aldo (C), and Cort (D) concentrations under various environmental conditions. Values are means ± SEM. Data were analyzed by Kruskal–Wallis test followed by Steel test (⁄P < 0.05). ⁄P < 0.05 indicates statistical significance compared with the control group. Numbers in parentheses indicate the number of animals used for measurement of parameters. Abbreviations: Ang II, angiotensin II; AVT, arginine vasotocin; Hyper, hyperosmotic seawater group; and Hypo, hypoosmotic seawater group. M. Uchiyama et al. / General and Comparative Endocrinology 195 (2014) 40–46 43

Fig. 3. Correlations between plasma osmolality and plasma Ang II (A), Aldo (B), AVT (C), and Cort (D) concentrations. Correlations between Ht value and plasma Ang II (E), Aldo (F), AVT (G), and Cort (H) concentrations. Data are expressed as correlation coefficients (r). Control (s), dry (j), hyperosmotic ( ), and hypoosmotic ( ) treatment groups are shown. See the legend of Fig. 2 for abbreviations. centrifugation at 3500 rpm for 1 h, radioactivity of the pellet was calculated from premeasured standards (1–104 pM) in four-param- counted in a c-counter (RIA Star 5420, Packard, Meriden, CT, eter logistic analysis of the data by KyPlot software (KyensLab Inc., USA). Plasma concentrations of total Ang II and AVT were Tokyo, Japan). 44 M. Uchiyama et al. / General and Comparative Endocrinology 195 (2014) 40–46

Fig. 4. Correlations between plasma concentrations of hormones Ang II and Aldo (A) and AVT (B) and between Aldo and AVT (C) and Cort (D). Data are expressed as correlation coefficients (r). Control (s), dry (j), hyperosmotic ( ), and hypoosmotic ( ) treatment groups are shown. See the legend of Fig. 2 for abbreviations.

2.5. Statistical analysis plasma osmolality in the hypoosmotic-acclimated frogs were maintained at the same concentrations as those in the control Data are represented as means ± SEM were analyzed using frogs. Plasma Na+,ClÀ, and urea concentrations in the frogs under parametric and nonparametric tests. Statistical significance was dry and hyperosmotic conditions were significantly higher than assessed by one-way analysis of variance (ANOVA), followed by those under hypoosmotic conditions (P < 0.01). These results indi- Bonferroni multiple comparison test, Kruskal–Walis test, and Steel cate that frogs were dehydrated under hyperosmotic and dry con- test, a nonparametric multiple comparison procedure. Correlations ditions. In contrast, there were no significant changes in body were calculated by Spearman’s correlation analysis. P values of weight, Ht, and plasma osmolality between the control and <0.05 were considered statistically significant. Statistical analysis hypoosmotic-acclimated frogs. was performed using a personal computer with the program Excel Statistics (Esumi Co., Ltd., Tokyo, Japan). 3.2. Plasma Ang II, AVT, Aldo, and Cort concentrations

In F. cancrivora, the mean plasma Ang II, AVT, Aldo, and Cort con- 3. Results centrations were 12.4 fmol/mL (n = 6), 29.8 fmol/mL (n = 6), 8.9 pmol/mL (n = 3), and 7.2 pmol/mL (n = 3) respectively. Plasma 3.1. Change in body weight, Ht, plasma osmolality, urea concentration, Ang II, AVT, Aldo and Cort concentrations in the frogs under various and electrolyte concentrations osmotic conditions are shown in Fig. 2. Concentrations of each hor- mone were maintained within a relatively narrow range in the frogs Body weights of frogs were 28.6 ± 2.0 (n = 6), 29.8 ± 3.4 (n = 10), under control and hypoosmotic conditions; plasma Ang II, AVT, and 32.6 ± 3.5 (n = 11) and 29.7 ± 3.3 (n = 7) in the control, dry, hyper- Aldo concentrations in the hypoosmotic-acclimated frogs did not osmotic seawater, and hypoosmotic seawater groups, respectively. differ significantly from those in the control fogs (Fig. 2A–C). On The frogs kept under dry and hyperosmotic conditions showed a the other hand, large inter-individual variations in Ang II, AVT, Aldo, significant reduction in body weight (percent change based on and Cort concentrations were observed in dehydrated (desiccated the initial body weight) after 5 days of treatment. The frogs under and hyperosmotic-acclimated) frogs. Plasma Ang II concentrations control and hypoosmotic seawater conditions showed a 4% reduc- in these frogs were significantly higher than those in the control tion in body weight during the experiment (Fig. 1A). The body frogs (P < 0.05, Steel test; Fig. 2A). Plasma AVT, Aldo, and Cort con- À1 weight of frogs placed in 80% seawater (800 mosmol kg H2O) de- centrations in the desiccated and hyperosmotic-acclimated frogs creased to approximately 72% of their initial body weight, which were 2–3.5, 2–3, and 7–8 times higher than those in the control or was comparable to the body weight of frogs under dry conditions hypoosmotic-acclimated frogs, respectively. There was a significant at the end of experiment. Ht, plasma osmolality, and Na+,ClÀ, difference in the AVT, Aldo, and Cort concentrations among the and urea concentrations are presented in Fig. 1B–E. Increases in groups (AVT, Aldo, and Cort, P < 0.05; Kruskal–Walis test); however, Ht and plasma osmolality in the desiccated and hyperosmotic- the increases in AVT, Aldo, and Cort concentrations in dehydrated acclimated frogs significantly differed from those in control and frogs were not statistically different from those in control or hypoosmotic-acclimated frogs (P < 0.01). In contrast, Ht and hypoosmotic frogs (P > 0.05, Steel test; Fig. 2B–D). M. Uchiyama et al. / General and Comparative Endocrinology 195 (2014) 40–46 45

3.3. Correlations between plasma components and hormones ity (hyperosmolemia). There was a significant correlation between Ht as an indicator of plasma volume and Ang II as well as AVT con- The results of correlations between plasma components and centration. Thus, it seems that dehydration (hyperosmolemia and hormones are shown in Figs. 3 and 4. There was a significant cor- hypovolemia) may stimulate the secretion of these osmoregulatory relation between plasma osmolality and Ang II, AVT, or Cort con- hormones under hyperosmotic conditions in F. cancrivora. Plasma centration (Ang II vs. osmolality, P < 0.01; AVT vs. osmolality, Ang II concentration was significantly higher in hyperosmotic- P < 0.01; Cort vs. osmolality, P < 0.05); however, no correlation acclimated and desiccated frogs. In hyperosmotic-acclimated frogs, was observed between plasma osmolality and Aldo concentration the plasma AVT concentration was not statistically higher; how- (Fig. 3A–D). Similarly, there were significant correlations between ever, the mean values were approximately 2–3.5 times those in plasma Na+ concentration and Ang II and AVT concentrations but control frogs. Furthermore, Aldo and Cort concentrations signifi- not the Aldo concentration (data not shown). On the other hand, cantly correlated with Ang II but not AVT concentration under var- there was a significant correlation between Ht and Ang II or AVT; ious osmotic stress conditions. Thus, it appears that Ang II plays however, there was no correlation between Ht and Aldo or Cort pivotal roles in body fluid stabilization under dehydrated concentration (Fig. 3E–H). Changes in body weight significantly condition. correlated with plasma Ang II (r = 0.46, P < 0.05), AVT (r = 0.53, It is generally accepted that dehydration under dry and hyper- P < 0.05), Aldo (r = 0.446, P < 0.05), and Cort (r = 0.724, P < 0.05) osmotic conditions induces an increase in plasma urea concentra- concentrations. Plasma Ang II concentration was significantly cor- tion (e.g., Balinsky, 1981; Jørgensen, 1997). It is assumed that the related with plasma Aldo and AVT concentrations (Fig. 4A and B). accumulation of urea in the body fluids enables some amphibians There was no correlation between plasma Aldo and AVT concentra- to tolerate high salinities and dry conditions (Katz, 1989; Withers tion (Fig. 4C); however, there was a significant correlation between and Guppy, 1996). It is also suggested that the increase in urea con- plasma Aldo and Cort concentration (Fig. 4D). centration associated with salinity tolerance is due to both urea retention and increased urea synthesis. Wright et al. (2004) ob- served that carbamoyl phosphate synthetase I (a hepatic ornithine 4. Discussion urea cycle enzyme) increased, and that plasma and muscle urea concentrations increased in seawater-acclimated F. cancrivora. Till In the present study, hypertonic dehydration [i.e., loss of mass date, there is no evidence for the acceleration of urea synthesis by (total body weight, TBW) and a significant increase in Ht and plas- Ang II or AVT in amphibians. However, several hormones, including ma osmolality] was observed under dry and hyperosmotic seawa- Ang II and AVP, stimulate urea synthesis in mammals (Corvera and ter environments after 5 days of acclimation in F. cancrivora. The García-Sáinz, 1983; Luther et al., 2012). increase in plasma osmolality mainly resulted from the increase During the past two decades, molecular approaches have re- in Na+,ClÀ, and urea concentrations in both the groups. Although sulted in the cloning of several urea transporter cDNA isoforms de- salt acclimation and water restriction are both osmotically stress- rived from two gene families: urea transporter-A type and -B type ful conditions, the two situations are quite different in their nature. (Sands, 2003). In rats and mice, Ang II and AVP regulate urea trans- In the hyperosmotic environment, there is an abundance of Na+ port and accumulation through urea transporter proteins in the and ClÀ and water availability depends on the osmotic gradient. kidney (Kato et al., 2000; Klein et al., 2006, 2012). We previously In contrast, under dry conditions, water, Na+, and ClÀ are quite re- reported that AVT stimulated urea transport and expression of urea stricted. Nevertheless, there was no significant difference between transporter protein and mRNA in the kidney and urinary bladder of plasma constituents and hormone concentrations in desiccated B. marinus (Konno et al., 2007). Although the urinary bladder of F. and hyperosmotic seawater-acclimated frogs. Thus, an osmoregu- cancrivora was less responsive to mammalian neurohypophysial latory system may stabilize large fluctuations in body fluids con- hormones than that of R. temporaria and B. melanostictus (Dicker stituents rather than maintaining their constant concentrations and Elliott, 1970), urea transport was significantly increased when in frogs under such stressful conditions. an isolated urinary bladder was exposed in vitro to AVT (Chew In F. cancrivora, mean plasma Ang II, AVT, and Aldo concentra- et al., 1972). Thus, Ang II and/or AVT may stimulate the expression tions are similar to those observed in amphibians in previous stud- of urea transporter protein and accumulation of urea in F. cancrivo- ies (Konno et al., 2005; Mayer and Propper, 2000; Nouwen and ra under dry and hyperosmotic environments. Kühn, 1985; Wright et al., 2003). It is conceivable that changes in In conclusions, our data demonstrate that concentrations of Ang body fluid content reflect changes in hormone concentrations II and other plasma components were increased in dehydrated and vice versa. In aquatic environments, water and solutes of body frogs under dry and hyperosmotic environments. Hyperosmolality fluids, particularly Na+ and ClÀ, are continuously exchanged across of plasma was induced by increases in Na+,ClÀ, and urea concen- the integument and excreted in dilute urine. In the present study, trations. These results indicate that crab-eating frogs may exhibit plasma osmolality and Ht values in hypoosmotic-acclimated frogs similar physiological responses to both seawater-acclimated and were similar to those in control frogs, which could freely migrate dry conditions. Dehydration (hyperosmolemia and hypovolemia) between 20% seawater and land. According to in vitro experiments, may stimulate renin–angiotensin system in F. cancrivora. On the Aldo, Ang II, and AVT actively increased Na+ transport across the other hand, there was a significant correlation between plasma abdominal skin and urinary bladder (Asher and Garty, 1988; Niel- AVT concentration and plasma osmolality as well as Ht; however, sen, 1997; Proto et al., 1983). Thus, increases in concentrations of there was a large individual variation, and plasma AVT concentra- those hormones would be expected in hypoosmotic-acclimated tion in dehydrated frogs did not increase compared with that in frogs. However, plasma Ang II, Aldo, and AVT concentrations in control frogs. Therefore, the renin–angiotensin system may play hypoosmotic-acclimated frogs were very similar to those in control pivotal roles in body fluid regulation under volemic and osmotic frogs. This suggests that relatively stable body fluid homeostasis is stress in the crab-eating frog species with unique osmoregulation. maintained without any special hyperactivity of osmotic hormones under hypoosmotic conditions in F. cancrivora. In contrast, plasma Acknowledgments concentrations of osmoregulatory hormones (Ang II, AVT, Aldo, and Cort) showed large variations among individuals under hyperos- This work was supported in part by grants to MU from the JSPS motic and dry conditions. Three hormones (Ang II, AVT, and Cort) (No. 22570062) and by the Cooperative Program (No. 005, 2011) of significantly increased in relation to an increase in plasma osmolal- Atmosphere and Ocean Research Institute, the University of Tokyo. 46 M. Uchiyama et al. / General and Comparative Endocrinology 195 (2014) 40–46

We would like to thank Dr. Masayuki Sumida of Hiroshima Univer- Konno, N., Hyodo, S., Takei, Y., Matsuda, K., Uchiyama, M., 2005. Plasma aldosterone, sity for his help with frog species identification based on mtDNA angiotensin II, and arginine vasotocin concentrations in the toad, Bufo marinus, following osmotic treatments. Gen. Comp. Endocrinol. 140, 86–93. gene sequences. We thank students of the Faculty of Animal Sci- Konno, N., Hyodo, S., Matsuda, K., Uchiyama, M., 2007. Arginine vasotocin promotes ences and Agricultural Technology, Silpakorn University, for cap- urea permeability through urea transporter expressed in the toad urinary turing the crab-eating frogs and the students of the Faculty of bladder cells. Gen. Comp. Endocrinol. 152, 281–285. Kurniawan, N., Djong, T.H., Islam, M.M., Nishizawa, T., Belabut, D.M., Sen, Y.H., Science, University of Toyama, for rearing the frogs. Wanichanon, R., Yasri, I., Sumida, M., 2011. Taxonomic status of three types of Fejervarya cancrivora from and other Asian countries based on References morphological observations and crossing experiments. Zool. Sci. 28, 12–24. Luther, J.M., Luo, P., Wang, Z., Cohen, S.E., Kim, H.-S., Fogo, A.B., Brown, N.J., 2012.

+ Aldosterone deficiency and mineralocorticoid receptor antagonism prevent Asher, C., Garty, H., 1988. Aldosterone increases the apical Na permeability of toad angiotensin II-induced cardiac, renal, and vascular injury. Kidney Int. 82, 643– bladder by two different mechanisms. Proc. Natl. Acad. Sci. 85, 7413–7417. 651. Balinsky, J.B., 1981. Adaptation of nitrogen metabolism to hyperosmotic Mayer, L.P., Propper, C.R., 2000. Intra- and extracellular dehydration has no effect on environment in amphibia. J. Exp. Zool. 215, 335–350. plasma levels of angiotensin II in an amphibian. J. Exp. Zool. 286, 343–349. Chew, M.-M., Elliott, A.B., Wong, H.Y., 1972. Permeability of urinary bladder of Rana Nielsen, R., 1997. Correlation between transepithelial Na+ transport and cancrivora to urea in the presence of oxytocin. J. Physiol. 223, 757–772. transepithelial water movement across isolated frog skin (Rana esculenta). J. Corvera, S., García-Sáinz, J.A., 1983. Hormonal stimulation of mitochondrial Membrane Biol. 159, 61–69. glutaminase. Effects of vasopressin, angiotensin II, adrenaline and glucagon. Nouwen, E.J., Kühn, E.R., 1985. Volumetric control of arginine vasotocin and Biochem. J. 210, 957–960. mesotocin release in the frog (Rana ridibunda). J. Endocrinol. 105, 371–377. Dicker, S.E., Elliott, A.B., 1970. Effect of neurohypophysial hormones on fluid Proto, M.C., Coviello, A., Khosla, M.C., Bumpus, F.M., 1983. In: Effects of frog-skin movement across isolated bladder of Rana cancrivora, Rana temporaria and Bufo angiotensin II in amphibians. Hypertension 5, V-101–V-104. melanostictus. J. Physiol. 210, 137–149. Sands, J.M., 2003. Topical review, Molecular mechanisms of urea transport. J. Gordon, M.S., Schmidt-Nielsen, K., Kelly, H.M., 1961. Osmotic regulation in the crab- Membrane Biol. 191, 149–163. eating frog (Rana cancrivora). J. Exp. Biol. 38, 659–678. Stiffler, D.F., De Ruyter, M.L., Hanson, P.B., Marshall, M., 1986. Interrenal function in Hasegawa, Y., Watanabe, T.X., Sokabe, H., Nakajima, T., 1983. Chemical structure of larval Ambystoma tigrinum: I. Responses to alterations in external electrolyte angiotensin in the bullfrog Rana catesbeiana. Gen. Comp. Endocrinol. 50, 75–80. concentrations. Gen. Comp. Endocrinol. 62, 290–297. Johnson, W.E., Propper, C.R., 2000. Effects of dehydration on plasma osmolality, Uchiyama, M., Konno, N., 2006. Review: hormonal regulation of ion and water thirst-related behavior, and plasma and brain angiotensin concentrations in transport in anuran amphibians. Gen. Comp. Endocrinol. 147, 54–61. Couch’s spadefoot toad, Scaphiopus couchii. J. Exp. Zool. 286, 572–584. Uchiyama, M., Ogasawara, T., Hirano, T., Kikuyama, S., Sasayama, Y., Oguro, C., 1990. Jørgensen, C.B., 1997. Review: urea and amphibian water economy. Comp. Biochem. Serum and urine osmolyte concentrations during acclimation to various Physiol. Part A: Physiol. 117, 161–170. dilutions of seawater in the crab-eating frog, Rana cancrivora. Zool. Sci. 7, Kato, A., Klein, J.D., Zhang, C., Sands, J.M., 2000. Angiotensin II increases vasopressin- 967–971. stimulated facilitated urea permeability in rat terminal IMCDs. Am. J. Physiol. Van Driessche, W., Zeiske, W., 1985. Ionic channels in epithelial cell membranes. Renal Physiol. 279, F835–F840. Physiol. Rev. 65, 833–903. Katz, U., 1989. Mini review. Strategies of adaptation to osmotic stress in anuran Withers, P.C., Guppy, M., 1996. Do Australian desert frogs co-accumulate Amphibia under salt and burrowing conditions. Comp. Biochem. Physiol. Part A: counteracting solutes with urea during aestivation? J. Exp. Biol. 199, 1809– Physiol. 93A, 499–503. 1816. Katz, U., Hanke, W., 1993. Mechanisms of hyperosmotic acclimation in Xenopus Wright, M.L., Guertin, C.J., Duffy, J.L., Szatkowski, M.C., Visconti, R.F., Alves, C.D., laevis (salt, urea or mannitol). J. Comp. Physiol. B. 163, 189–195. 2003. Developmental and diel profiles of plasma corticosteroids in the bullfrog, Klein, J.D., Fröhlich, O., Blount, M.A., Martin, C.F., Smith, T.D., Sands, J.M., 2006. Rana catesbeiana. Comp. Biochem. Physiol. Part A: Mol. Integ. Physiol. 135, 585– Vasopressin increases plasma membrane accumulation of urea transporter UT- 595. A1 in rat inner medullary collecting ducts. J. Am. Soc. Nephrol. 17, 2680–2686. Wright, P., Anderson, P., Weng, L., Frick, N., Wong, W.P., Ip, Y.K., 2004. The crab- Klein, J.D., Martin, C.F., Kent, K.J., Sands, J.M., 2012. Protein kinase C-a mediates eating frog, Rana cancrivora, up-regulates hepatic carbamoyl phosphate hypertonicity-stimulated increase in urea transporter phosphorylation in the synthetase I activity and tissue osmolyte levels in response to increased inner medullary collecting duct. Am. J. Physiol. Renal Physiol. 302, F1098– salinity. J. Exp. Zool. 301A, 559–568. F1130.