The expression and localization of aquaporins (AQP1-AQP5), members of the water channel family, in the developing rat submandibular gland were analysed using RT-PCR, Northern blotting and immunohistochemistry to explore their relation to the development of this salivary gland. RT-PCR analysis revealed unique expression patterns of each AQP. AQP1 was expressed constitutively during prenatal development, whereas the expression of AQP5 became more intense in the course of development from embryonic day 16.5 (E16) to E20. These expression patterns concurred with the results of Northern blot analysis. AQP3 and AQP4 mRNAs in the prenatal development were not detected in Northern blots, although they were detected by RT-PCR. During postnatal development, AQP5 and AQP1 mRNAs were expressed continuously, but no message for AQP3 or AQP4 was detected. AQP2 mRNA was not detected during either prenatal or postnatal development in this tissue. Immunohistochemical studies revealed that AQP5 was first localized at the apical membrane of proacinar cells at E18, and then became clearly distributed at the apical membrane of acinar cells in accordance with the differentiation and establishment of the mature acini. In addition, some vasculature also showed immunoreactivity for AQP5. AQP1 was immunolocalized in the blood vessels, including capillaries, of the gland throughout development. These observations suggest the existence of transcriptional regulation of rat AQP5, which is one of the most probable regulators of saliva production and secretion, during the establishment of the functional submandibular salivary gland.

, an exocrine-type water channel, was detected in the rat duodenum by Western blot analysis, and was localized by immunohistochemistry in the secretory granule membranes as well as in the apical and lateral aspects of the plasma membrane of Brunner's gland cells. Incubation of duodenal slices with vasoactive intestinal polypeptide (VIP) in vitro significantly increased the amount of AQP5 in the apical membrane fraction in a dose-and time-dependent manner with the amount reaching a plateau at 100 nM VIP and becoming near maximal after a 30-s incubation. Protein kinase inhibitors, 1-(5-isoquinolinesulfonyl)-2-methylpiperazine dihydrochloride (H-7, 50 M), and N-[2-(pbromocinnamylamino)ethyl]-5-isoquinolinesulfonamide (H-89; PKAspecific, 1 M) blocked this increase, but PKC-specific inhibitor calphostin C did not, implying the involvement of PKA but not PKC in this cellular event. Intravenous injection with VIP (40 g/kg body wt) provoked dilation of the lumen of the Brunner's gland at 2 and 7 min and increased the staining intensity of AQP5 in the apical and lateral membranes. AQP1 (both nonglycosylated and glycosylated forms) was also found to localize in the apical and basolateral membranes of cells of Brunner's gland. VIP, however, did not provoke any significant change in the AQP1 level in the apical membrane, as judged from the results of the above in vitro and in vivo experiments. These results suggest that VIP induced the exocytosis of granule contents and simultaneously caused translocation of AQP5 but not of AQP1 to the apical membrane in Brunner's gland cells. aquaporin 5; water channel; translocation COMPLEMENTARY DNAS of members of the large family of aquaporins (AQPs) have been cloned recently from a variety of mammalian tissues (29). Proteins of this AQP family selectively transport water and other components, such as urea and glycerol (29). Recent investigations have uncovered the molecular basis of the water transport across the cell membrane; e.g., the AQP5 molecule, one of the exocrine-type AQPs, was demonstrated to be increased in concentration at the apical membrane of the salivary gland in response to the stimulation of muscarinic receptors or to an increase in the intracellular level of calcium ions (10,11,28). AQP1 in rat bile duct cells and AQP2 in rat renal collecting duct cells were reported to traffic from an intracellular location to the apical membrane in response to secretin and vasopressin, respectively (16,21,30).A recent study of ours (22) demonstrated AQP5 to be present in the duodenum, where it was localized in the apical and lateral membranes of the secretory cells of Brunner's gland. In the duodenum, epithelial cells and cells of Brunner's gland secrete bicarbonate and mucin, respectively, into the duodenal lumen, which secretion is thought to play an important role in mucosal protection against gastric acid. Mucin and water are known to combine together to create a viscoelastic gel that becomes infiltrated with bicarbonate, thus forming a physicochemical barrier to hydrogen ions ...

The green fluorescent protein (GFP) of the jellyfish, Aeqorea victoria, was used as an autofluorescent tag to track the trafficking of aquaporin 5 (AQP5), an exocrine gland-type water channel. Two groups of chimeric proteins were constructed; one in which GFP was fused to the amino-terminus of AQP5 (GFP-AQP5) and the other, in which it was fused to the carboxyl terminus of it (AQP5-GFP). In each group, 2 chimeras were produced, a wild-type AQP5 with its normal sequence and a mutant AQP5 having a mutated amino acid at 259, i.e., GFP-AQP5-T259A and AQP5-GFP-T259A. They were used to transfect Madin-Darby canine kidney (MDCK) cells. The GFP-AQP5 chimera was localized in the intracellular vesicles, which trafficked to the plasma membrane in response to N(6), 2'-O-dibutyryladenosine 3', 5'-cyclic monophosphate (dbcAMP). Membrane trafficking was inhibited by N-[2-(p-bromocinnamylamino)ethyl]-5-isoquimolinesulfonamide (H-89) but not by palmitoyl-dl-carnitine chloride (PCC). In contrast, the AQP5-GFP chimera expressed in MDCK cells was localized constitutively on the plasma membrane. The cellular localization of the latter chimera was not affected by stimulation with dbcAMP in the presence or absence of H-89 or PCC. Replacement of Thr-259 with Ala-259 did not affect the dbcAMP-induced translocation of the chimeric protein, suggesting that phosphorylation of Thr-259 was not necessary for AQP5 trafficking under the present experimental conditions. Thus, the GFP-AQP5 chimera will be a useful tool to study AQP5 trafficking in vitro, whereas the constitutive membrane localization of the AQP5-GFP chimera suggests the importance of the carboxyl terminus of the AQP5 protein for its sorting, whether it is translocated to intracellular vesicles or to the plasma membrane.

SummaryThe intraperitoneal injection of lipopolysaccharide (LPS) (400 lg/kg body weight) induced the expression of mRNAs of inflammatory cytokines such as interleukin (IL)-1b, IL-6 and tumour necrosis factor (TNF)-a in the submandibular gland (SMG) of C3H/HeN mice but not that of C3H/HeJ mice, a mutant strain for Toll-like receptor-4 (TLR-4 -mutant). The mRNA levels of these cytokines in the SMG of the wild-type mice increased as early as 3 hr after injection, peaked at 3-6 hr, and had decreased again by 24 hr. In this study, we particularly focused on IL-1b, and induction by this endotoxin was investigated in detail. Denervation of the superior cervical trunk and chorda tympani nerve did not diminish the LPS-induced elevation of IL-1b mRNA in the SMG, indicating the irrelevance of the central nervous system in this induction. TLR-4 mRNA and protein were shown to be strongly expressed in the SMG, suggesting the direct action of LPS on this gland. IL-1b proteins were localized in the secretory granules of granular convoluted tubular (GCT) cells, and their molecular weights in the gland were 17Á5 and 20 kDa. IL-1b of the same size appeared in the saliva 6 hr after LPS injection in C3H/HeN but not in C3H/HeJ mice. The present study thus suggests that IL-1b, an inflammation cytokine, is induced and secreted into the saliva in response to endotoxin injected intraperitoneally.

By Western blot analysis, the expression level of aquaporin (AQP) 5 in the submandibular gland (SMG) was found to be different among individual rats of the Sprague-Dawley (SD) strain. Such differences were observed for AQP5 but not for AQP1 and consequently the SD strain was divided into two groups, one expressing a high level of AQP5 and the other a low one. The difference in average intensity of expression between the two groups was more than twofold. Immunohistochemical analysis of the SMG demonstrated that the AQP5 protein was localized in the basal and apical/lateral plasma membrane of acinar cells in rats expressing the high level of AQP5. In the rat expressing the low level, however, this channel protein was localized strongly in the apical/lateral plasma membrane, but only very weakly in the basal membrane of the acinar cells. Such a diverse localization of AQP5 was confirmed by Western blotting as well. Breeding between brother and sister was repeated for two times within high expressers and low expressers to obtain the third generation progenies (F2); the AQP5 level of the SMG in the third generation (F2 rats) from high expressers was significantly higher than the F2 from low expressers. Our present study suggests the existence of genetic variation in the expression of a water channel protein, AQP5, in rats.