Aquaporin-2 (AQP2) water channels are shuttled from intracellular vesicles to the apical plasma membrane of renal collecting duct principal cells in response to vasopressin (AVP). This shuttling response can be modulated by a variety of local and systemic factors, including prostaglandins. Previous reports have suggested that prostaglandins may increase collecting duct water permeability in their own right, but inhibit the response to AVP, and there has been considerable debate about the receptors involved (Veis et al. 1990; Hebert et al. 1993). In the present study, we have investigated the ability of PGE₂ and PGI₂ to cause shuttling of AQP2, alone and in conjunction with AVP.

For each experiment, two male Wistar rats (200-250 g) were anaesthetised with pentobarbitone sodium (240 mg kg^-1 I.P.) and killed by cervical dislocation. The kidneys were rapidly removed and inner medullary tubule suspensions prepared as previously described (Shaw & Marples, 2002). The pooled suspension was divided into six aliquots, and incubated at 37°C for 20 min with: control, AVP (1 nM), PGE₂ (280 nM), PGI₂ (280 nM), PGE₂ + AVP, and PGI₂ + AVP. The tissue was then homogenised, and plasma membrane- (PM) and intracellular vesicle- (ICV) enriched fractions prepared for Western blotting with an antibody against the c-terminus of AQP2. After visualisation by enzyme chemiluminescence (ECL), the PM:ICV ratio of the control for each experiment was set to 100 %, and other values expressed relative to this. Thus an increase in the PM:ICV ratio demonstrates an increase in the fraction of AQP2 present in the plasma membrane. Data, expressed as a percentage of control, are shown as means ± S.E.M., and were analysed using Student’s unpaired t test, and considered significant if P < 0.05.

In all experiments, AVP caused a shift in the PM:ICV ratio (189 ± 13 % of control, n = 7, P < 0.001). Both PGE₂ (133 ± 11 %, P < 0.05) and PGI₂ (160 ± 20 %, P < 0.05) induced a modest but significant shift on their own. In neither case did the addition of AVP cause a significant further shift (PGE₂ +AVP 161 ± 19 % n.s., PGI₂ +AVP 260 ± 45 % n.s.), although it is also the case that the increase in ratio caused by AVP is not significantly diminished by PGI₂.

These results indicate that either PGE₂ and PGI₂ alone can induce shuttling of AQP2 to the apical plasma membrane, while PGE₂, but probably not PGI₂, impairs the ability of AVP to induce such shuttling. It remains to be determined which receptor types mediate the various effects of these prostaglandins.

This work was funded by the MRC.