Yuji Ogushi

Water Adaptation Strategy in Anuran Amphibians: Molecular Diversity of Aquaporin

Most adult anuran amphibians except for the aquatic species absorb water across the ventral pelvic skin and reabsorb it from urine in the urinary bladder. Many terrestrial and arboreal species use a region in the posterior or pelvic region of the ventral skin that is specialized for rapid rehydration from shallow water sources or moist substrates. Periods of terrestrial activity can be prolonged by reabsorption of dilute urine from the urinary bladder. Aquaporin (AQP), a water channel protein, plays a fundamental role in these water absorption/reabsorption processes, which are regulated by antidiuretic hormone. Characterization of AQPs from various anurans revealed that the unique water homeostasis is basically mediated by two types of anuran-specific AQPs, i.e. ventral pelvic skin and urinary bladder type, respectively. The bladder-type AQP is further expressed in the pelvic skin of terrestrial and arboreal species, together with the pelvic skin-type AQP. In contrast, the pelvic skin-type AQP (AQP-x3) of the aquatic Xenopus has lost the ability of efficient protein production. The extra C-terminal tail in AQP-x3 consisting of 33 nucleotides within the coding region appears to participate in the posttranscriptional regulation of AQP-x3 gene expression by attenuating protein expression. The positive transcriptional regulation of bladder-type AQP in the pelvic skin and negative posttranscriptional regulation of pelvic skin-type AQP provide flexibility in the water regulation mechanisms, which might have contributed to the evolutionary adaptation of anurans to a wide variety of water environments.

Correlation between aquaporin and water permeability in response to vasotocin, hydrin and {beta}-adrenergic effectors in the ventral pelvic skin of the tree frog Hyla japonica.

SUMMARY The ventral pelvic skin of the tree frog Hyla japonica expresses two kinds of arginine vasotocin (AVT)-stimulated aquaporins (AQP-h2 and AQP-h3), which affect the capacity of the frogs skin to absorb water. As such, it can be used as a model system for analyzing the molecular mechanisms of water permeability. We investigated AQP dynamics and water permeability in the pelvic skin of H. japonica following challenge with AVT, hydrins (intermediate peptides of pro-AVT) and β-adrenergic effectors. In the in vivo experiment, both AQP-h2 and AQP-h3 proteins were translocated to the apical plasma membrane in the principal cells of the first-reacting cell (FRC) layer in the pelvic skin following challenge with AVT, hydrin 1 and hydrin 2, thereby increasing the water permeability of the pelvic skin. The β-adrenergic receptor agonist isoproterenol (IP) and its anatagonist propranolol (PP) in combination with AVT or hydrins were used as challenge in the in vitro experiment. IP increased water permeability whereas PP inhibited it, and both events were well correlated with the translocation of the AQPs to the apical membrane. In the PP+AVT-treated skins, labels for AQP-h2 and AQP-h3 were differentially visible among the principal cells; the apical plasma membrane of some cells was labeled while others were not, indicating that the response of PP or AVT is different from cell to cell. These results provide morphological evidence that the principal cells of the FRC layers may have two kinds of receptors: a V2 receptor and β-adrenergic receptor.

The water-absorption region of ventral skin of several semiterrestrial and aquatic anuran amphibians identified by aquaporins

Regions of specialization for water absorption across the skin of Bufonid and Ranid anurans were identified by immunohistochemistry and Western blot analysis, using antibodies raised against arginine vasotocin (AVT)-stimulated aquaporins (AQPs) that are specific to absorbing regions of Hyla japonica. In Bufo marinus, labeling for Hyla urinary bladder-type AQP (AQP-h2), which is also localized in the urinary bladder, occurred in the ventral surface of the hindlimb, pelvic, and pectoral regions. AQP-h2 was not detected in any skin regions of Rana catesbeiana, Rana japonica, or Rana nigromaculata. Hyla ventral skin-type AQP (AQP-h3), which is found in the ventral skin but not the bladder of H. japonica, was localized in the hindlimb, pelvic, and pectoral skins of Bufo marinus, in addition to AQP-h2. AQP-h3 was also localized in ventral skin of the hindlimb of all three Rana species and also in the pelvic region of R. catesbiana. Messenger RNA for AQP-x3, a homolog of AQP-h3, could be identified by RT-PCR from the hindlimb, pectoral, and pelvic regions of the ventral skin of Xenopus laevis, although AVT had no effect on water permeability. In contrast, 10(-8) M AVT-stimulated water permeability and translocation of AQP-h2 and AQP-h3 into the apical membrane of epithelial cells in regions of the skin of species where they had been localized by immunohistochemistry and Western blot analysis. Finally, water permeability of the hindlimb skin of B. marinus and all the Rana species was stimulated by hydrins 1 and 2 to a similar level as seen for AVT. The present data demonstrate species differences in the occurrence, distribution, and regulation of AQPs in regions of skin specialized for rapid water absorption that can be associated with habitat and also phylogeny.

Novel Vasotocin-Regulated Aquaporins Expressed in the Ventral Skin of Semiaquatic Anuran Amphibians: Evolution of Cutaneous Water-Absorbing Mechanisms

Until now, it was believed that only one form of arginine vasotocin (AVT)-regulated aquaporin (AQP) existed to control water absorption from the ventral skin of semiaquatic anuran amphibians, eg, AQP-rj3(a) in Rana japonica. In the present study, we have identified a novel form of ventral skin-type AQP, AQP-rj3b, in R. japonica by cDNA cloning. The oocyte swelling assay confirmed that AQP-rj3b can facilitate water permeability. Both AQP-rj3a and AQP-rj3b were expressed abundantly in the ventral hindlimb skin and weakly in the ventral pelvic skin. For the hindlimb skin, water permeability was increased in response to AVT, although the hydroosmotic response was not statistically significant in the pelvic skin. Isoproterenol augmented water permeability of the hindlimb skin, and the response was inhibited by propranolol. These events were well correlated with the intracellular trafficking of the AQPs. Immunohistochemistry showed that both AQP-rj3 proteins were translocated from the cytoplasmic pool to the apical membrane of principal cells in the first-reacting cell layer of the hindlimb skin after stimulation with AVT and/or isoproterenol. The type-b AQP was also found in R. (Lithobates) catesbeiana and R. (Pelophylax) nigromaculata. Molecular phylogenetic analysis indicated that the type-a is closely related to ventral skin-type AQPs from aquatic Xenopus, whereas the type-b is closer to the AQPs from terrestrial Bufo and Hyla, suggesting that the AQPs from terrestrial species are not the orthologue of the AQPs from aquatic species. Based on these results, we propose a model for the evolution of cutaneous water-absorbing mechanisms in association with AQPs.

Expression of a mammalian aquaporin 3 homolog in the anterior pituitary gonadotrophs of the tree frog, Hyla japonica

Aquaporins (AQPs) are a family of water channel proteins that play a major role in maintaining water homeostasis in various organisms. Several AQPs have been identified in the tree frog, Hyla japonica. Of these, AQP-h3BL, which is expressed in the basolateral membrane of the epithelial cells, is a homolog of mammalian AQP3. Using immunohistochemistry and in situ RT-PCR, we have demonstrated that AQP-h3BL is expressed in the anterior pituitary gonadotrophs of the tree frog but not in the other hormone-producing cells of the anterior pituitary. In gonadotrophs labeled for luteinizing hormone subunit-β (LHβ), AQP-h3BL protein was found to reside in the plasma membrane, the nuclear membrane and the cytoplasm. Double-labeling of AQP-h3BL mRNA and LHβ protein revealed that AQP-h3BL mRNA is expressed in the gonadotrophs. Following stimulation by gonadotropin-releasing hormone (GnRH), the label for AQP-h3BL localized in the plasma membrane became more intense, concomitant with the transport of LHβ-positive materials to the plasma membrane. These developments coincided with a decrease in the labeling density in the cytoplasm and near the nuclear membrane, suggesting that the latter localizations may function as “storage area“ for AQP-h3BL. Immunoelectron microscopy also confirmed these localizations of AQP-h3BL protein. Based on these results, we suggest that AQP-h3BL protein in the frog gonadotrophs is involved in the formation of secretory granules, the swelling and increase in the volume of the granules and exocytosis.

Molecular Machinery for Vasotocin-Dependent Transepithelial Water Movement in Amphibians: Aquaporins and Evolution

Amphibians represent the first vertebrates to adapt to terrestrial environments, and are successfully distributed around the world. The ventral skin, kidney, and urinary bladder are important osmoregulatory organs for adult anuran amphibians. Water channel proteins, called aquaporins (AQPs), play key roles in transepithelial water absorption/reabsorption in these organs. At least 43 types of AQPs were identified in anurans; a recent phylogenetic analysis categorized anuran AQPs among 16 classes (AQP0-14, 16). Anuran-specific AQPa2 was assigned to AQP6, then was further subdivided into the ventral skin-type (AQP6vs; AQPa2S), whose expression is confined to the ventral skin, and the urinary bladder-type (AQP6ub; AQPa2U), which is basically expressed in the urinary bladder. For the osmoregulatory organs, AQP3 is constitutively located in the basolateral plasma membrane of tight-junctioned epithelial cells. AQP6vs, AQP2 and/or AQP6ub are also expressed in these epithelial cells and are translocated to the apical membrane in response to arginine vasotocin, thereby regulating water absorption/reabsorption. It was suggested recently that two subtypes of AQP6vs contribute to cutaneous water absorption in Ranid species. In addition, AQP5 (AQP5a) and AQP5L (AQP5b) were identified from Xenopus tropicalis Gray, 1864, and AQP5 was localized to the apical membrane of luminal epithelial cells of the urinary bladder in dehydrated Xenopus. This finding suggested that AQP5 may be involved in water reabsorption from this organ under dehydration. Based on the hitherto reported information, we propose models for the evolution of water-absorbing/reabsorbing mechanisms in anuran osmoregulatory organs in association with AQPs.

Molecular and cellular characterization of urinary bladder-type aquaporin in Xenopus laevis

In contrast to many anuran amphibians, water is not reabsorbed from the urinary bladder in aquatic Xenopus, thereby helping to prevent excessive water influx. However, little is known about the molecular mechanisms for this process. In the present study, we have identified urinary bladder-type aquaporin, AQP-x2, in Xenopus laevis by cDNA cloning. The predicted amino acid sequence contained six putative transmembrane domains and the two conserved Asn-Pro-Ala motifs, characteristic of AQPs. The sequence also contained a putative N-glycosylation site and phosphorylation motifs for protein kinase A and protein kinase C. The oocyte swelling assay showed that AQP-x2 facilitated water permeability. Reverse transcription-PCR analysis indicated that AQP-x2 mRNA was expressed in the urinary bladder and lung, and faintly in the kidney. Immunomicroscopical study further localized AQP-x2 protein to the cytoplasm of granular cells in the luminal epithelium of the urinary bladder whilst AQP3 was observed along the basolateral side of these cells. In vitro stimulation of the urinary bladder with 10(-8)M vasotocin (AVT), 10(-8)M hydrin 1, or 10(-8)M hydrin 2 had no clear effect on the subcellular distribution of AQP-x2. When the AVT concentration was increased to 10(-6)M, however, AQP-x2 was partially transferred to the apical plasma membrane. The treatment with hydrin 1 or hydrin 2 at the same concentration failed to induce the translocation to the apical membrane. On the other hand, AQP3 remained along the basolateral side even after the treatment with vasotocin or hydrins. The results suggest that the poor responsiveness of AQP-x2 to neurohypophyseal peptides may be a main cause for the little water permeability of the urinary bladder of X. laevis.