Aquaporin-9 (AQP9) is an aquaglyceroporin membrane channel shown biophysically to conduct water, glycerol, and other small solutes. Because the physiological role/s of AQP9 remain undefined and the expression sites of AQP9 remain incomplete and conflicting, we generated AQP9 knockout mice. In the absence of physiological stress, knockout mice did not display any visible behavioral or severe physical abnormalities. Immunohistochemical analyses using multiple antibodies revealed AQP9 specific labeling in hepatocytes, epididymis, vas deferens, and in epidermis of wild type mice, but a complete absence of labeling in AQP9 ^−/− mice. In brain, no detectable labeling was observed. Compared with control mice, plasma levels of glycerol and triglycerides were markedly increased in AQP9 ^−/− mice, whereas glucose, urea, free fatty acids, alkaline phosphatase, and cholesterol were not significantly different. Oral administration of glycerol to fasted mice resulted in an acute rise in blood glucose levels in both AQP9 ^−/− and AQP9 ^+/− mice, revealing no defect in utilization of exogenous glycerol as a gluconeogenic substrate and indicating a high gluconeogenic capacity in nonhepatic organs. Obese Lepr ^db / Lepr ^db AQP9 ^−/− and obese Lepr ^db / Lepr ^db AQP9 ^+/− mice showed similar body weight, whereas the glycerol levels in obese Lepr ^db / Lepr ^db AQP9 ^−/− mice were dramatically increased. Consistent with a role of AQP9 in hepatic uptake of glycerol, blood glucose levels were significantly reduced in Lepr ^db / Lepr ^db AQP9 ^−/− mice compared with Lepr ^db / Lepr ^db AQP9 ^+/− in response to 3 h of fasting. Thus, AQP9 is important for hepatic glycerol metabolism and may play a role in glycerol and glucose metabolism in diabetes mellitus.

Aquaporin (AQP7) is expressed in proximal tubules and is involved in glycerol uptake. The cellular expression and physiological function in other organs remain largely undefined. AQP7 knockout (KO) mice were generated and used for immunohistochemical analyses to define the organ and cellular expression of AQP7. AQP7 labeling was found in kidney proximal tubule, heart, skeletal muscle, testis, epididymis, as well as in white and brown adipose tissue (WAT and BAT) of wild-type mice. Importantly, immunoreactivity was completely absent from these tissues in AQP7 KO mice. At the cellular level, the capillary endothelium WAT and BAT displayed prominent staining, whereas AQP7 labeling in adipocyte membranes was undetectable. Double-labeling confocal microscopy revealed coexpression of AQP7 with capillary AQP1 but not with adipocyte GLUT4. Moreover, immunoelectron microscopy and RT-PCR of isolated microvessels confirmed the vascular AQP7 expression. Distinct immunolabeling of the capillary endothelium was also observed in both skeletal and heart muscle with no apparent staining of skeletal or cardiac myocytes. As previously reported, specific immunolabeling was confined to brush border in segment 3 renal proximal tubules and to spermatids and spermatozoa in male reproductive tract. The expression of AQP7 was induced up to 2.2-fold in WAT of mice with streptozotocin-induced diabetes mellitus (S-DM) compared with controls and fasting for 72 h (but not 24 h) induced significant increase in AQP7 expression. In conclusion, AQP7 is expressed in capillary endothelia of adipose tissue (and cardiac and striated muscle) and is upregulated in WAT in response to S-DM supporting its role in glycerol metabolism.

Aquaporin-5 (AQP5), an apical plasma membrane (APM) water channel in salivary glands, lacrimal glands, and airway epithelium, has an important role in fluid secretion. M₃ muscarinic acetylcholine receptor (mAChR)-induced changes in AQP5 localization in rat parotid glands were investigated with immunofluorescence or immunoelectron microscopy, detergent solubility, and gradient density floatation assays. Confocal microscopy revealed AQP5 localization in intracellular vesicles of interlobular duct cells in rat parotid glands and AQP5 trafficking to the APM 10 min after injection of the mAChR agonist cevimeline. Conversely, 60 min after injection, there was a diffuse pattern of AQP5 staining in the cell cytoplasm. The calcium ionophore A-23187 mimicked the effects of cevimeline. Immunoelectron microscopic studies confirmed that cevimeline induced AQP5 trafficking from intracellular structures to APMs in the interlobular duct cells of rat parotid glands. Lipid raft markers flotillin-2 and GM1 colocalized with AQP5 and moved with AQP5 in response to cevimeline. Under control conditions, the majority of AQP5 localized in the Triton X-100-insoluble fraction and floated to the light-density fraction on discontinuous density gradients. After 10-min incubation of parotid tissue slices with cevimeline or A-23187, AQP5 levels decreased in the Triton X-100-insoluble fraction and increased in the Triton X-100-soluble fraction. Thus AQP5 localizes in the intracellular lipid rafts, and M₃ mAChR activation induces AQP5 trafficking to the APM with lipid rafts via intracellular Ca²⁺ signaling and induces AQP5 dissociation from lipid rafts to nonrafts on the APM in the interlobular duct cells of rat parotid glands.

Thirteen mammalian aquaporin (AQP) isoforms have been identified, and they have a unique tissue-specific pattern of expression. AQPs have been documented in the reproductive system of both male and female humans, rats, and mice. However, tissue expression and cellular and subcellular localization of AQPs are unknown in the female reproductive system of pigs. In this study, AQP1 immunoreactivity was detected in the capillary endothelium of the ovary. Distinct immunolabeling of capillary endothelium was also observed in the oviduct and uterus. AQP5 was expressed in flattened follicle cells of primordial follicles, granulosa cells of developing ovarian follicles, and muscle cells of the oviduct and uterus. Staining of AQP5 was also observed in the epithelial cells of the oviduct and uterine epithelium. AQP9 immunoreactivity was observed in granulosa cells of developing follicles. AQP9 was also localized in the luminal epithelial cells of the oviduct and uterine epithelia cells. This is, to our knowledge, the first study that shows tissue expression and cellular and subcellular localization of AQPs in the reproductive system of the female pig. Moreover, these results suggest that several subtypes of the AQPs (AQP1, 5, and 9) are involved in regulation of water homeostasis in the reproductive system of gilts.