Respiratory aquaporins in lung inflammation: the night is young.

From the nasopharynx to the pleural space, alterations in fluid transport are central to a wide range of pathophysiologic processes in the respiratory tract. Despite intense investigation, details of the pathways for water transport in the respiratory tract remain unclear. Insight into molecular mechanisms of membrane water permeability was provided by the discovery of the aquaporins, a growing family of water-specific membrane channel proteins (1). In this issue, Towne and colleagues (2) provide a rigorous description of changes in aquaporin expression in the respiratory tract by using a murine model of pulmonary inflammation, the adenovirus infection. The existence of water-specific membrane channel proteins was postulated by a small group of physiologists for several decades, but the molecular identity of such proteins remained enigmatic until recently. Aquaporin-1 (AQP1) was serendipitously discovered during studies of erythrocyte membrane proteins (3). Several features suggested that AQP1 could be the long-sought water channel, a suspicion that was confirmed by expression in Xenopus laevis oocytes. Transfer of AQP1-expressing oocytes from isotonic to hypotonic medium resulted in rapid swelling and rupture, whereas control oocytes changed little in volume (4). Subsequent investigations in oocytes, as well as proteoliposomes reconstituted with purified AQP1 protein, demonstrated that the channel is specifically permeable to water, but not to other small molecules, including protons, ions, urea, and glycerol (4, 5). More recently, it has been suggested that AQP1 is also permeated by CO 2 (6); however, the magnitude of this permeability may be much lower than that of water, and the physiologic relevance is debated. Four aquaporins have been identified in the respiratory tract: AQP1, AQP3, AQP4, and AQP5, each with a unique distribution suggesting distinct physiologic roles (7, 8). AQP1 is expressed in the apical and basolateral membrane of the microvascular endothelium (Figure 1A), as well as in the visceral pleura. AQP3 is expressed in the basolateral membrane of basal cells found in the tracheal and nasopharyngeal epithelium. AQP4 is present in the basolateral membrane of ciliated columnar cells in bronchial, tracheal, and nasopharyngeal epithelium. AQP5 is expressed in the apical membrane of type I pneumocytes (Figure 1B), as well as in the apical membrane of acinar cells in submucosal glands of the airways and nasopharynx. The specificity of AQP5 for type I pneumocytes has also been demonstrated in studies of alveolar epithelial cell differentiation (9). The ontogeny of each of the aquaporins in the lung is likewise distinct. AQP1 is expressed from late gestation in rat lung and is induced by corticosteroids in both fetal and adult animals (7). AQP5 is expressed shortly after birth and is not steroid responsive (10). Both AQP1 and AQP5 are expressed at high levels in adult animals. AQP4 expression increases transiently after birth (10, 11) and is enhanced by corticosteroids and b -agonists (11). The complex distribution and ontogeny of aquaporins in the respiratory tract suggests potential involvement in a variety of pathophysiologic processes. The first examples of rate-limiting aquaporin expression were derived from studies of the kidney. AQP2 is the vasopressin-responsive water channel that confers high water permeability on the collecting duct (12). Deen and colleagues (13) demonstrated that mutations in the AQP2 gene produce nephrogenic diabetes insipidus in humans. This observation has now been expanded by the demonstration that acquired forms of nephrogenic diabetes insipidus, including lithium therapy, chronic hypokalemia, and ureteral obstruction, all result from downregulation of AQP2 expression (12). More recently, upregulation of AQP2 has been shown in fluid retention states, including congestive heart failure, pregnancy, and cirrhosis (1). The physiologic relevance of AQP1 in the proximal tubule was recently confirmed in AQP1-null mice, in which a profound urinary concentrating defect became evident after water deprivation (14). Another example of rate-limiting aquaporin expression has recently been demonstrated in the salivary gland, where AQP5 is expressed in the apical membrane of acinar cells (8, 15). AQP5-null mice have a marked reduction in saliva formation (16), and transfection of an aquaporin gene into radiation-damaged salivary glands partially restores function (17). Several lines of evidence suggest a physiologic role for aquaporins in the respiratory tract. Studies of in situ perfused sheep lungs (18) and perfused distal airway seg( Received in original form November 3, 1999 )