Reduced osmotic water permeability of the peritoneal barrier in aquaporin-1 knockout mice

Aquaporin-1 (AQP1) water channels are expressed widely in epithelia and cap illary endothelia involved in fluid transport. To test whether AQP1 facilit ates water movement from capillaries into the peritoneal cavity, osmoticall y induced water transport rates mere compared in AQP1 knockout [(-/-)], het erozygous [(+/-)], and wild-type [(+/+)] mice. In (+/+) mice, RT-PCR showed detectable transcripts for AQP1, AQP3, AQP4, AQP7, and AQP8. Immunofluores cence showed AQP1 protein in capillary endothelia and mesangium near the pe ritoneal surface and AQP4 in adherent muscle plasmalemma. For measurement o f water transport, 2 mi of saline containing 300 mM sucrose (600 mosM) were infused rapidly inter the peritoneal cavity via a catheter. Serial fluid s amples (50 mu l) were withdrawn over 60 min, with albumin as a volume marke r. The albumin dilution data showed significantly decreased initial volume influx in AQP1. (-/-) mice: 101 +/- 8, 107 +/- 5, and 42 +/- 4 (SE) mu l/mi n in (+/+), (+/-), and (-/-) mice, respectively [n = 6-10, P < 0.001, (-/-) vs. others]; Volume influx for AQP4 knockout mice was 100 +/- 8 mu l/min. In the absence of an osmotic gradient, (H2O)-H-3 uptake [halftime = 2.3 and 2.2 min in (+/+) and (-/-) mice, respectively], [C-14]urea uptake [half ti me = 7.9 and 7.7 min in (+/+) and (-/-) mice, respectively], and spontaneou s isosmolar fluid absorption from the peritoneal cavity [0.47 +/- 0.05 and 0.46 +/- 0.04 ml/h in (+/+) and (-/-) mice, respectively] were not affected by AQP1 deletion. Therefore, AQP1 provides a major route for osmotically d riven water transport across the peritoneal barrier in peritoneal dialysis.