Exploring central‐pore amino‐acid residues important for CO2 permeation through human aquaporin‐5 (AQP5)

Based on molecular‐dynamics (MD) simulations of bovine AQP1 (bAQP1), the major barrier to gas permeation through the extracellular side of the central pore does not correspond to the region of maximum pore constriction (Val52) but to a region above (further from the membrane's center) that has a dense layer of water near Asp50 (Wang et al, JSB 2007). We made all possible single‐point mutations of human AQP5 at Thr41 (~bAQP1‐Asp50) and Leu43 (~bAQP1‐Val52). We used microelectrodes to monitor the maximum rise in extracellular surface pH (ΔpH S ) caused by CO 2 influx (index of CO 2 permeability) into Xenopus oocytes, and monitored cell swelling under hypotonic conditions to compute osmotic water permeability (P f ). Of all Thr41 mutants, so far only T41F—when compared to wild type (wt)—causes a statistically significant fall in channel‐dependent ΔpH S (to value not different from 0). No Thr41 mutants affect P f . Furthermore, no Leu43 mutants produce differences in either ΔpH S or P f . Our results are consistent with the MD data for bAQP1 in that a mutation of Thr41 can block CO 2 permeability but has no effect on water permeability; whereas mutations of Leu43 do not affect either CO 2 or water permeability. Thus, the major extracellular‐side barrier for CO 2 through the central pore could be near AQP5‐Thr41 (more outside central pore), rather than Leu43 (just inside the central pore). Funded by AHA 09POST2060873 and ONR N00014 ‐08‐10532