Water and Gas Transport through Aquaporins

Dozens of papers in high impact journals assign a CO2 transporting function to some of the aquaporin family members. The non-gas-channel portion of the aquaporin containing membranes would have to have a CO2 permeability (PCO2) which is from 200-fold to 1000-fold smaller than PCO2 of lipid bilayers of comparable composition to be physiologically relevant. Yet, the molecular mechanism which might prevent a neutral molecule with a hexadecane:water partition coefficient of ~1 from freely passing biological membranes is unknown. Thus, we will investigate the hypothesis that PCO2 of the mammalian epithelia is decreased by a protein barrier (e.g uroplakin) and/or regulated by carbonic anhydrase. In continuing our search for lipid constituents which may decrease the gas permeability of lipid bilayers, we are going to reconstitute archaeal lipids and extend the study to include the investigation of H2S transport. If archaeal lipids represent a barrier to H2S, the function of aquaporins in thermophilic archae may be resolved. Moreover, we are going to perform the first stoichiometric study of ammonia and ammonium transport through plant aquaporins. Since there is general disagreement between molecular dynamics simulations and experiments in uncovering which single channel water permeability pf is larger, that of orthodox aquaporins or glyceroaquaporins, we will measure the water flux through planar membranes reconstituted with the purified AqpZ, GlpF and AQP1 proteins and simultaneously determine protein abundance by fluorescence correlation spectroscopy. If pf of GlpF is higher, the doubts raised against the applicability of the water models for molecular dynamics simulations in confined geometries do not appear to be justified. Finally, we will test the prediction made by molecular dynamics simulations whereby conserved arginine in the aquaporin pore is susceptible to transmembrane voltage, i.e. we will provide the first experimental test to determine whether human aquaporins are gated.

Membranes restrict the undesirable loss of genetic material, proteins, and nutrients from cells. This is mandatory to maintain an individual microclimate within the cell that ensures optimal conditions for the specific cell tasks within the various tissues and organs of organisms. However, the membrane must allow nutrients and metabolites to enter and leave the cell, respectively. For example, H2O and CO2 must be permanently shuttled in and out of the cell. Although both molecules are able to diffuse through the lipid part of the membrane, specialized channels, so-called aquaporins, are thought to regulate their flux.Aquaporins have been shown to conduct water in a single-file fashion. That is, the channel is so narrow that water molecules cannot pass each other within the channel. If aquaporins were to function like normal water pipes, friction of the water molecules against the channel wall would result in very slow water movement. This is obviously not the case. However, a clear undisputed microscopic picture of how fast the single-file water moves through the channel is not available. We have filled the gap by developing two different assays: one for assessing the single channel water permeability of aquaporins in their native environment of the cell membrane and a second one for measuring the single channel water permeability in a pure system that contains only the purified channel and the lipid bilayer in which it is embedded. We have counted the channels per unit of membrane area by using single molecule techniques, i.e. we attached a dye to each channel and monitored its fluorescence and we used an atomic force microscope that like a miniature gramophone detected the small height differences between the protein and the surrounding lipid. We derived the water flux from either the dilution of small molecules close to the channel mouth or from the change in volume that was enclosed into aquaporin-containing vesicular membranes. We observed that there is virtually no friction with the channel wall. In contrast, the water molecules are three to ten times more mobile than their bulk counterparts. Our assays have great potential for detecting the effect of pharmacological compounds on the single channel permeability and may thus be very useful for drug development.CO2 is fast enough on its own in permeating the membrane and does not likely require further acceleration by an aquaporin. We have therefore tested whether uroplakin may restrict membrane CO2 diffusion. Although uroplakin plaques on top of epithelial cells reduce the permeability of the urinary bladder and ducts for ammonia and water, they failed to produce similar effects on CO2. Instead CO2 transport is regulated by an enzyme: carbonic anhydrase. It catalyses the transformation of bicarbonate into CO2. If present in low amounts, the reaction is very slow - such as in soda water where it takes hours for the CO2 bubbles to form and leave the water. If present in large quantities, CO2 is formed instantaneously and leaves the cell without being hindered by the membrane and without being assisted by aquaporins.