Evidence against aquaporin-1-dependent CO2 permeability in lung and kidney

Abstract

AQP1-dependent CO2 transport has been suggested from the increased CO2 permeability in Xenopus oocytes expressing AQP1. Potential implications of this finding include AQP1-facilitated CO2 exchange in mammalian lung and HCO3-/CO2 transport in kidney proximal tubule. We reported previously that: (a) CO2 permeability in erythrocytes was not affected by AQP1 deletion, (b) CO2 permeability in liposomes was not affected by AQP1 reconstitution despite a 100-fold increased water permeability, and (c) CO2 blow-off by the lung in living mice was not impaired by AQP1 deletion. We extend these observations by direct measurement of CO2 permeabilities in lung and kidney. CO2 transport across the air-space-capillary barrier in isolated perfused lungs was measured from changes in air-space fluid pH in response to addition/removal of HCO3-/CO2 from the pulmonary artery perfusate. The pH was measured by pleural surface fluorescence of a pH indicator (BCECF-dextran) in the air-space fluid. Air-space fluid pH equilibrated rapidly (t(1/2) approximately 6 s) in response to addition/removal of HCO3-/CO2. However, the kinetics of pH change was not different in lungs of mice lacking AQP1, AQP5 or AQP1/AQP5 together, despite an up to 30-fold reduction in water permeability. CO2 transport across BCECF-loaded apical membrane vesicles from kidney proximal tubule was measured from the kinetics of intravesicular acidification in response to rapid mixing with a HCO3-/CO2 solution. Vesicles rapidly acidified (t(1/2) approximately 10 ms) in response to HCO3-/CO2 addition. However the acidification rate was not different in kidney vesicles from AQP1-null mice despite a 20-fold reduction in water permeability. The results provide direct evidence against physiologically significant transport of CO2 by AQP1 in mammalian lung and kidney.

Figure 1. Air space-capillary CO₂ transport in mouse lung measured by a pleural surface fluorescence method

A, schematic showing an isolated perfused mouse lung. The air-space fluid contains carbonic anhydrase and the fluorescent pH indicator BCECF-dextran. The pulmonary artery perfusate was exchanged between isosmolar CO₂/HCO₃⁻-free and CO₂/HCO₃⁻-containing solutions. CO₂/HCO₃⁻ addition results in acidification of air-space fluid and a decrease in pleural surface BCECF-dextran fluorescence. B, relationship between BCECF fluorescence and pH. Inset, perfusate exchange rate. Time course of pleural surface fluorescence in response to exchange between non-fluorescent and fluorescent (containing FITC-dextran) perfusates. Perfusate flow rate was 3-5 ml min⁻¹, as in subsequent CO₂ transport measurements. C, time course of air-space fluid pH (computed from pleural surface fluorescence) in response to exchange between CO₂/HCO₃⁻-free and CO₂/HCO₃⁻-containing perfusates. Air-space fluid contained carbonic anhydrase (CA) at indicated concentrations. D, same as in C (with 1 mg ml⁻¹ CA) in which the air-space instillate contained indicated amounts of phosphate buffer.

Figure 2. Influence of aquaporin deletion on air-space-capillary CO₂ and osmotic water permeabilities

A, left, CO₂ permeability in lungs from mice of indicated genotype. Time course of air-space fluid pH in response to exchange between CO₂/HCO₃⁻-free and CO₂/HCO₃⁻-containing perfusates measured as in Fig. 1C. The air-space instillate contained 1 mg ml⁻¹ CA and the perfusate contained 0 or 25 mm HCO₃⁻. Right, osmotic water permeability. Time course of air-space fluid osmolality (measured by FITC-dextran fluorescence) in response to exchange between perfusates of osmolality 300 and 500 mosmol kg⁻¹. B, top, representative time course data as in A (left) shown on an expanded time scale. Bottom, summary of rates of CO₂ transport into and out of the air-space fluid. Data are shown as mean and s.e.m. for n = 3.5 lungs per group. Differences not significant.

Figure 3. Osmotic water permeability and CO₂ permeability in apical membrane vesicles from mouse kidney proximal tubule

A, vesicles were subjected to a 250 mm inwardly directed sucrose gradient at 10 °C. Data are plotted using two contiguous times scales to show the complete osmotic equilibration. B, vesicles were loaded with the fluorescent pH indicator BCECF and suspended in an isosmolar buffer at pH 7.4 (not containing CO₂/HCO₃⁻, see Methods). The suspension was mixed in a stopped-flow apparatus with appropriate buffers to give 25 mm CO₂/HCO₃⁻ at pH 7.0 or 8.0. Where indicated, vesicles were incubated with acetazolamide prior to the assay. Curves have been displaced arbitrarily in the y-direction for clarity.

Figure 4. Influence of AQP1 deletion on CO₂ permeability in apical membrane vesicles from proximal tubule

A, vesicles were subjected to a 25 mm CO₂/HCO₃⁻ gradient at a final outside pH of 7.0 as in Fig. 3B. B, averaged P_CO2 computed from experiments as in A. Data are means and s.e.m.