Søren Drud-Heydary Nielsen

The existence of water-selective channels has been postulated to explain the high water permeability of erythrocytes and certain epithelial cells. The aquaporin CHIP (channel-forming integral membrane protein of 28 kDa), a molecular water channel, is abundant in erythrocytes and water-permeable segments of the nephron. To determine whether CHIP may mediate transmembrane water movement in other water-permeable epithelia, membranes of multiple organs were studied by immunoblotting, immunohistochemistry, and immunoelectron microscopy using affinity-purified anti-CHIP IgG. The apical membrane of the choroid plexus epithelium was densely stained, implying a role for CHIP in the secretion of cerebrospinal fluid. In the eye, CHIP was abundant in apical and basolateral domains of ciliary epithelium, the site of aqueous humor secretion, and also in lens epithelium and corneal endothelium. CHIP was detected in membranes of hepatic bile ducts and water-resorptive epithelium of gall bladder, suggesting a role in bile secretion and concentration. CHIP was not detected in glandular epithelium of mammary, salivary, or lacrimal glands, suggesting the existence of other water-channel isoforms. CHIP was also not detected within the epithelium of the gastrointestinal mucosa. CHIP was abundant in membranes of intestinal lacteals and continuous capillaries in diverse tissues, including cardiac and skeletal muscle, thus providing a molecular explanation for the known water permeability of certain lymphatics and capillary beds. These studies underscore the hypothesis that CHIP plays a major role in transcellular water movement throughout the body.

Vasopressin (antidiuretic hormone) regulates body water balance by controlling water permeability of the renal collecting ducts. The control mechanisms may involve alterations in the number or unit conductance of water channels in the apical plasma membrane of collecting-duct cells. How this occurs is unknown, but indirect evidence exists for the "shuttle" hypothesis, which states that vasopressin causes exocytic insertion of water channel-laden vesicles from the apical cytosol. To test key aspects of the shuttle hypothesis, we have prepared polyclonal antisera against the recently cloned collecting-duct water channel protein and used the antisera in immunolocalization studies (light and electron microscopic levels) in thin and ultrathin cryosections from rat kidney. Labeling was seen exclusively in collecting-duct principal cells and inner medullary collecting-duct cells. Apical membrane labeling was intense. There was heavy labeling of abundant small subapical vesicles and of membrane structures within multivesicular bodies. In addition, labeling of basolateral plasma membranes in inner medullary collecting ducts was present. Depriving rats of water for 24 or 48 hr markedly increased collecting-duct water-channel protein expression determined by immunoblotting and immunolabeling. These results are compatible with at least two complementary modes of water-channel regulation in collecting-duct cells: (i) control of channel distribution between the apical membrane and a reservoir in subapical vesicles (shuttle hypothesis) and (ii) regulation of the absolute level of expression of water-channel protein.

The aquaporins are a family of water channels expressed in several water-transporting tissues, including the kidney. We have used a peptide-derived, affinity-purified polyclonal antibody to aquaporin-3 (AQP-3) to investigate its localization and regulation in the kidney. Immunoblotting experiments showed expression in both renal cortex and medulla, with greatest expression in the base of the inner medulla. Subcellular fractionation of membranes, using progressively higher centrifugation speeds, revealed that AQP-3 is present predominantly in the 4,000 and 17,000 g pellets and, in contrast to AQP-2, is virtually absent in the high-speed (200,000 g) pellet that contains small intracellular vesicles. Immunocytochemistry and immunofluorescence studies revealed that labeling is restricted to the cortical, outer medullary, and inner medullary collecting ducts. Within the collecting duct, principal cells were labeled, whereas intercalated cells were unlabeled. Consistent with previous immunofluorescence studies (K. Ishibashi, S. Sasaki, K. Fushimi, S. Uchida, M. Kuwahara, H. Saito, T. Furukawa, K. Nakajima, Y. Yamaguchi, T. Gojobori, and F. Marumo. Proc. Natl. Acad. Sci. USA 91: 6269-6273, 1994; T. Ma, A. Frigeri, H. Hasegawa, and A. S. Verkman. J. Biol. Chem. 269: 21845-21849, 1994), the labeling was confined to the basolateral domain. Immunoelectron microscopy, using the immunogold technique in ultrathin cryosections, demonstrated a predominant labeling of the basolateral plasma membranes. In contrast to previous findings with AQP-2, there was only limited AQP-3 labeling of intracellular vesicles, suggesting that this water channel is not regulated acutely through vesicular trafficking. Immunoblotting studies revealed that thirsting of rats for 48 h approximately doubled the amount of AQP-3 protein in the inner medulla. These studies are consistent with a role for AQP-3 in osmotically driven water absorption across the collecting duct epithelium and suggest that the expression of AQP-3 is regulated on a long-term basis.