Molecular mechanisms of water and solute transport across archaebacterial lipid membranes.

Archaebacteria thrive in environments characterized by anaeobiosis, saturated salt, and both high and low extremes of temperature and pH. The bulk of their membrane lipids are polar, characterized by the archaeal structural features typified by ether linkage of the glycerol backbone to isoprenoid chains of constant length, often fully saturated, and with sn-2,3 stereochemistry opposite that of glycerolipids of Bacteria and Eukarya. Also unique to these bacteria are macrocyclic archaeol and membrane spanning caldarchaeol lipids that are found in some extreme thermophiles and methanogens. To define the barrier function of archaebacterial membranes and to examine the effects of these unique structural features on permeabilities, we investigated the water, solute (urea and glycerol), proton, and ammonia permeability of liposomes formed by these lipids. Both the macrocyclic archaeol and caldarchaeol lipids reduced the water, ammonia, urea, and glycerol permeability of liposomes significantly (6-120-fold) compared with diphytanylphosphatidylcholine liposomes. The presence of the ether bond and phytanyl chains did not significantly affect these permeabilities. However, the apparent proton permeability was reduced 3-fold by the presence of an ether bond. The presence of macrocyclic archaeol and caldarchaeol structures further reduced apparent proton permeabilities by 10-17-fold. These results indicate that the limiting mobility of the midplane hydrocarbon region of the membranes formed by macrocyclic archaeol and caldarchaeol lipids play a significant role in reducing the permeability properties of the lipid membrane. In addition, it appears that substituting ether for ester bonds presents an additional barrier to proton flux.

Water transport across yeast vacuolar and plasma membrane-targeted secretory vesicles occurs by passive diffusion.

To determine whether solute transport across yeast membranes was facilitated, we measured the water and solute permeations of vacuole-derived and late secretory vesicles in Saccharomyces cerevisiae; all permeations were consistent with passive diffusive flow. We also overexpressed Fps1p, the putative glycerol facilitator in S. cerevisiae, in secretory vesicles but observed no effect on water, glycerol, formamide, or urea permeations. However, spheroplasts prepared from the strain overexpressing Fps1p showed enhanced glycerol uptake, suggesting that Fps1p becomes active only upon insertion in the plasma membrane.

Hourglass pore-forming domains restrict aquaporin-1 tetramer assembly.

The AQP1 water channel protein is a homotetramer with 28 kDa subunits containing six transmembrane domains. The sequence-related loops B (cytoplasmic) and E (extracellular) were predicted to overlap within the membrane, forming an aqueous pore ("the hourglass") flanked by the corresponding B and E residues 73 and 189. Cryoelectron microscopy of AQP1 previously revealed the central hourglass structure surrounded by six transmembrane helices which provide contact points between subunits. Several mutants in loop B and E residues were nonfunctional when expressed in X. laevis oocytes, but their ability to form tetramers is unknown. To explore the possible functional dependence of hourglass domains in adjacent subunits, we prepared a series of tandem dimers as single 55 kDa polypeptides containing different combinations of wild-type (AQP1) or mutant subunits (A73M or C189M). In oocytes, AQP1-AQP1 exhibited high osmotic water permeability, and AQP1-C189M exhibited half activity. Dimer polypeptides with A73M were nonfunctional or not expressed. In yeast secretory vesicles, AQP1-AQP1 exhibited high water permeability, AQP1-C189M exhibited half activity, and both were inhibited by pCMBS. Although expressed, the dimer polypeptides with A73M were all nonfunctional. Tetramer formation was investigated by detergent solubilization and velocity sedimentation through sucrose gradients. Dimer polypeptides containing one A73M subunit or two C189M subunits migrated with slower velocity (s < 3.5 S). In contrast, dimer polypeptides with one C189M subunit migrated with velocity similar to native AQP1 tetramers (s approximately 6 S). Thus, although hourglass pore-forming domains are not points of subunit-subunit contact, the structure of loop B is important to normal tetramer assembly.

Reconstitution of water channel function of aquaporins 1 and 2 by expression in yeast secretory vesicles.

Aquaporins 1 (AQP1) and 2 (AQP2) were expressed in the yeast secretory mutant sec6-4. The mutant accumulates post-Golgi, plasma membrane-targeted vesicles and may be used to produce large quantities of membrane proteins. AQP1 or AQP2 were inducibly expressed in yeast and were localized within isolated sec6-4 vesicles by immunoblot analysis. Secretory vesicles containing AQP1 and AQP2 exhibited high water permeabilities and low activation energies for water flow, indicating expression of functional AQP1 and AQP2. AQP1 solubilized from secretory vesicles was successfully reconstituted into proteoliposomes, demonstrating the ability to use the yeast system to express aquaporins for reconstitution studies. The AQP2-containing secretory vesicles showed no increased permeability toward formamide, urea, glycerol, or protons compared with control vesicles, demonstrating that AQP2 is highly selective for water over these other substances. We conclude that the expression of aquaporins in yeast sec6 vesicles is a valid system to further study mammalian water channel function.

Functional analysis of aquaporin-1 deficient red cells. The Colton-null phenotype.

The aquaporin-1 (AQP1) water transport protein contains a polymorphism corresponding to the Colton red blood cell antigens. To define the fraction of membrane water permeability mediated by AQP1, red cells were obtained from human kindreds with the rare Colton-null phenotype. Homozygosity or heterozygosity for deletion of exon I in AQP1 correlated with total or partial deficiency of AQP1 protein. Homozygote red cell morphology appeared normal, but clinical laboratory studies revealed slightly reduced red cell life span in vivo; deformability studies revealed a slight reduction in membrane surface area. Diffusional water permeability (Pd) was measured under isotonic conditions by pulsed field gradient NMR. Osmotic water permeability (Pf) was measured by change in light scattering after rapid exposure of red cells to increased extracellular osmolality. AQP1 contributes approximately 64% (Pd = 1.5 x 10(-3) cm/s) of the total diffusional water permeability pathway, and lipid permeation apparently comprises approximately 23%. In contrast, AQP1 contributes > 85% (Pf = 19 x 10(-3) cm/s) of the total osmotic water permeability pathway, and lipid permeation apparently comprises only approximately 10%. The ratio of AQP1-mediated Pf to Pd predicts the length of the aqueous pore to be 36 A.

Stretch sensitivity of transmembrane mobility of hydrogen peroxide through voids in the bilayer. Role of cardiolipin.

Availability of voids for diffusion of quinone in the membrane was shown to be the rate-limiting step in electron transport in mitochondria and chloroplasts (Mathai, J. C., Sauna, Z. E., John, O., and Sitaramam, V (1993) J. Biol. Chem. 268, 15442-15454). The primary role of voids in these diffusion-controlled reactions required a more rigorous documentation of the role of diffusion in membranes by independent measurements. The transbilayer diffusion of hydrogen peroxide as monitored by occluded catalase activity was developed as a kinetically valid probe to specifically address this question. This in turn led to unique results on the mechanistic basis of stretch (= hypo-osmotic) activation of hydrogen peroxide permeation via such voids. The rate of peroxide permeation is shown to be markedly stretch sensitive in some cells/organelles (e.g. peroxisomes) and insensitive in others (e.g. erythrocytes); this was equally true of liposomes prepared from lipids extracted from the corresponding cells/organelles. The molecular basis of stretch sensitivity was uncovered using specific binary mixtures of lipids: while pure phosphatidyl choline liposomes were stretch insensitive, these became sensitive when doped only with specific lipids, viz. cardiolipin and cerebrosides. Cholesterol abolished this stretch sensitivity in ternary mixtures. Induction of stretch sensitivity by cardiolipin was marked by lowering of activation energy for peroxide diffusion, a negative temperature coefficient for glucose permeation while further addition of cholesterol reversed these phenomena. The steady state fluorescence polarization studies revealed intimate correlations between anisotropy, hydrogen peroxide diffusion, and stretch sensitivity consistent with presence of voids in these binary mixtures.

Rate-limiting step in electron transport. Osmotically sensitive diffusion of quinones through voids in the bilayer.

Respiration in mitochondria and photosynthesis in chloroplasts varied with the osmotic stretch of the membrane such that these processes were uniformly inhibited at higher osmolalities. A systematic evaluation of segmental electron transport in these intact particles showed that no individual complex exhibited osmotic sensitivity, whereas osmotic sensitivity appeared wherever the assay involved crossing over the corresponding quinone in the electron transport chain. The evidence was consistent with the rate-limiting step in electron transport being the availability of voids for quinone migration rather than any of the components of electron transport chain per se. Evidence based on quinone reconstitution in mitochondria depleted of quinone by acetone treatment clearly distinguished the kinetic control in the hypotonic domain and diffusive control via availability of voids in the hypertonic domain. Influence as well as the presence of voids was further confirmed in quinone-depleted mitochondria reconstituted with quinone as well as cholesterol. Decrease in lateral diffusion of the fluorescent probe, 12-(9-anthroyl)stearic acid, on osmotic compression of the bilayer is consistent with a change in void size distribution on osmotic compression of the bilayer. A direct correlation between succinate cytochrome c oxidoreductase activity and diffusivity of fluorescent probe 12-(9-anthroyl)stearic acid confirmed the availability of voids as the rate-limiting step in electron transport.

Hydrogen peroxide permeation across liposomal membranes: a novel method to assess structural flaws in liposomes.

Hydrogen peroxide permeation across large multilamellar vesicles of defined and complex lipid composition was shown to obey precise kinetic relationships for the activity of the occluded catalase. Careful assay conditions precluded simultaneous peroxidative damage to the lipids. The kinetic data was consistent with a barrier role for the bilayer for hydrogen peroxide permeation. More interestingly, hydrogen peroxide permeation across liposomes of complex lipid mixtures exhibited osmotic inhibition of permeation of hydrogen peroxide. On the other hand, purified egg lecithin vesicles did not exhibit any effect of external osmolality on hydrogen peroxide permeation in an experimentally defined non-lytic zone of external osmolarity. These results argue in favour of a heterogeneous, heteroporous structure of bilayers with complex lipid composition.

Variable porosity of the mitochondrial inner membrane induced by energization.

The empirically observed relationship between the activity of membrane-bound enzyme systems and transporters and the external osmotic pressure offered a direct method to assess the reflection coefficients to polyols in respiring mitochondria. These osmotically modulated reaction rates varied with the molecular mass of the external polyol similar to volume and solute fluxes across dialysis membranes. The equivalent pore radii of mitochondria were shown to increase with respiration (and temperature) and decrease on addition of the uncoupler, 2,4-dinitrophenol. The magnitude of the induced porosity in the inner membrane was large enough to render the chemiosmotic mechanism inoperable in well-coupled rat liver mitochondria.

Differential effects of osmotic pressure on mitochondrial respiratory chain and indices of oxidative phosphorylation.

Oxidative phosphorylation was critically evaluated in terms of activities which are sensitive and insensitive to variations in external osmotic pressure in mitochondria. Integrity of mitochondria was determined in terms of a variety of parameters, including the latency of the occluded enzymes, by careful titrations as a function of external osmotic pressure as well as detergent concentrations. The evidence indicated that the rate-limiting step in respiratory states 2 and 4 would be osmotically insensitive, as opposed to the osmotically sensitive respiration of states 1 and 3 and uncoupler-stimulated respiration with glutamate + malate and succinate. Cytochrome oxidase activity in mitochondria as well as in purified reconstituted systems exhibited osmotic insensitivity but marked sensitivity to ionic strength, offering an interesting model to study the osmotically insensitive respiration. Cytochrome oxidase activity led to permeation of mannitol across the mitochondrial inner membrane. Stimulation of cytochrome oxidase activity by uncouplers did not require an intact membrane.