FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W1992578464> ?p ?o ?g. }

• W1992578464 endingPage "25347" @default.
• W1992578464 startingPage "25340" @default.
• W1992578464 abstract "Several membrane channels, like aquaporin-1 (AQP1) and the RhAG protein of the rhesus complex, were hypothesized to be of physiological relevance for CO2 transport. However, the underlying assumption that the lipid matrix imposes a significant barrier to CO2 diffusion was never confirmed experimentally. Here we have monitored transmembrane CO2 flux (JCO2) by imposing a CO2 concentration gradient across planar lipid bilayers and detecting the resulting small pH shift in the immediate membrane vicinity. An analytical model, which accounts for the presence of both carbonic anhydrase and buffer molecules, was fitted to the experimental pH profiles using inverse problems techniques. At pH 7.4, the model revealed that JCO2 was entirely rate-limited by near-membrane unstirred layers (USL), which act as diffusional barriers in series with the membrane. Membrane tightening by sphingomyelin and cholesterol did not alter JCO2 confirming that membrane resistance was comparatively small. In contrast, a pH-induced shift of the CO2 hydration-dehydration equilibrium resulted in a relative membrane contribution of about 15% to the total resistance (pH 9.6). Under these conditions, a membrane CO2 permeability (3.2 ± 1.6 cm/s) was estimated. It indicates that cellular CO2 uptake (pH 7.4) is always USL-limited, because the USL size always exceeds 1 μm. Consequently, facilitation of CO2 transport by AQP1, RhAG, or any other protein is highly unlikely. The conclusion was confirmed by the observation that CO2 permeability of epithelial cell monolayers was always the same whether AQP1 was overexpressed in both the apical and basolateral membranes or not. Several membrane channels, like aquaporin-1 (AQP1) and the RhAG protein of the rhesus complex, were hypothesized to be of physiological relevance for CO2 transport. However, the underlying assumption that the lipid matrix imposes a significant barrier to CO2 diffusion was never confirmed experimentally. Here we have monitored transmembrane CO2 flux (JCO2) by imposing a CO2 concentration gradient across planar lipid bilayers and detecting the resulting small pH shift in the immediate membrane vicinity. An analytical model, which accounts for the presence of both carbonic anhydrase and buffer molecules, was fitted to the experimental pH profiles using inverse problems techniques. At pH 7.4, the model revealed that JCO2 was entirely rate-limited by near-membrane unstirred layers (USL), which act as diffusional barriers in series with the membrane. Membrane tightening by sphingomyelin and cholesterol did not alter JCO2 confirming that membrane resistance was comparatively small. In contrast, a pH-induced shift of the CO2 hydration-dehydration equilibrium resulted in a relative membrane contribution of about 15% to the total resistance (pH 9.6). Under these conditions, a membrane CO2 permeability (3.2 ± 1.6 cm/s) was estimated. It indicates that cellular CO2 uptake (pH 7.4) is always USL-limited, because the USL size always exceeds 1 μm. Consequently, facilitation of CO2 transport by AQP1, RhAG, or any other protein is highly unlikely. The conclusion was confirmed by the observation that CO2 permeability of epithelial cell monolayers was always the same whether AQP1 was overexpressed in both the apical and basolateral membranes or not. The widely accepted model that gases like NH3, CO2, and O2 pass biological membranes by diffusion through the lipid matrix has been recently called into question. For example, the membrane protein channels AmtB and aquaporin-8 have been identified to transport NH3 (1Khademi S. O'Connell III, J. Remis J. Robles-Colmenares Y. Miercke L.J. Stroud R.M. Science. 2004; 305: 1587-1594Crossref PubMed Scopus (548) Google Scholar, 2Saparov S.M. Liu K. Agre P. Pohl P. J. Biol. Chem. 2007; 282: 5296-5301Abstract Full Text Full Text PDF PubMed Scopus (165) Google Scholar). Protein channels such as the human aquaporin-1, the plant aquaporin NtAQP1, and the RhAG protein of the rhesus complex were reported to provide a pathway for CO2 transport (3Endeward V. Musa-Aziz R. Cooper G.J. Chen L.M. Pelletier M.F. Virkki L.V. Supuran C.T. King L.S. Boron W.F. Gros G. FASEB J. 2006; 20: 1974-1981Crossref PubMed Scopus (179) Google Scholar, 4Uehlein N. Lovisolo C. Siefritz F. Kaldenhoff R. Nature. 2003; 425: 734-737Crossref PubMed Scopus (506) Google Scholar, 5Endeward V. Cartron J.P. Ripoche P. Gros G. FASEB J. 2007; 22: 1-11Google Scholar). The similarity in the findings for NH3 and CO2 is very surprising because Overtone's rule predicts that their permeabilities, PM, across the lipid phase of biological membranes differ 750-fold. The number was calculated assuming that NH3 and CO2 have comparable membrane diffusivities and that neither one of them belongs to those extremely rare exceptions from Overtone's rule (6Walter A. Gutknecht J. J. Membr. Biol. 1986; 90: 207-217Crossref PubMed Scopus (407) Google Scholar, 7Saparov S.M. Antonenko Y.N. Pohl P. Biophys. J. 2006; 90: L86-L88Abstract Full Text Full Text PDF PubMed Scopus (66) Google Scholar) so that the proportionality between PM and the biphasic partition coefficient (water/organic solvent) applies as shown in Equation 1, PM,CO2=PM,NH3KCO2/KNH3=12cms-1(Eq. 1) where KCO2 ∼ 1.5 (8Simon S.A. Gutknecht J. Biochim. Biophys. Acta. 1980; 596: 352-358Crossref PubMed Scopus (66) Google Scholar), KNH3 ∼ 0.002 (6Walter A. Gutknecht J. J. Membr. Biol. 1986; 90: 207-217Crossref PubMed Scopus (407) Google Scholar), and PM,NH3 = 0.016 cm/s (9Antonenko Y.N. Pohl P. Denisov G.A. Biophys. J. 1997; 72: 2187-2195Abstract Full Text PDF PubMed Scopus (68) Google Scholar). A PM,CO2 of 12 cm/s suggests that the lipid matrix of biological membranes cannot act as a barrier to CO2 diffusion. In fact, a stagnant water layer adjacent to the membrane that has the same thickness (δ) as the membrane would generate the same resistance to CO2 flow as is caused by the membrane itself. Because these so-called unstirred layers (USL) 2The abbreviations used are: USLunstirred layerFITCfluorescein isothiocyanateMDCKMadin-Darby canine kidneyPBSphosphate-buffered salineDPhPCdiphytanoyl-phosphatidylcholineSMsphingomyelinAQP1aquaporin-1CAcarbonic anhydrasehAQP1human AQP1. 2The abbreviations used are: USLunstirred layerFITCfluorescein isothiocyanateMDCKMadin-Darby canine kidneyPBSphosphate-buffered salineDPhPCdiphytanoyl-phosphatidylcholineSMsphingomyelinAQP1aquaporin-1CAcarbonic anhydrasehAQP1human AQP1. are unavoidably present in the vicinity of all biological membranes (10Barry P.H. Diamond J.M. Physiol. Rev. 1984; 64: 763-872Crossref PubMed Scopus (355) Google Scholar) and because their size always exceeds membrane thickness by several orders of magnitude, the contribution of membrane resistance to the overall resistance to CO2 flow is expected to be negligibly small. Thus, lowering membrane resistance by insertion of CO2 conducting channels seems to be of questionable physiological relevance. This is exactly what has been concluded from a study where red blood cell aquaporin-1 has been knocked out without any effect on PM,CO2 (11Yang B.X. Fukuda N. van Hoek A. Matthay M.A. Ma T.H. Verkman A.S. J. Biol. Chem. 2000; 275: 2686-2692Abstract Full Text Full Text PDF PubMed Scopus (164) Google Scholar). This view is further supported by molecular dynamics simulations of CO2 transport through AQP1, which revealed that such a transport process would be physiologically meaningless in phospholipid membranes of common composition (12Hub J.S. de Groot B.L. Biophys. J. 2006; 91: 842-848Abstract Full Text Full Text PDF PubMed Scopus (125) Google Scholar, 13Wang Y. Cohen J. Boron W.F. Schulten K. Tajkhorshid E. J. Struct. Biol. 2007; 157: 534-544Crossref PubMed Scopus (162) Google Scholar). unstirred layer fluorescein isothiocyanate Madin-Darby canine kidney phosphate-buffered saline diphytanoyl-phosphatidylcholine sphingomyelin aquaporin-1 carbonic anhydrase human AQP1. unstirred layer fluorescein isothiocyanate Madin-Darby canine kidney phosphate-buffered saline diphytanoyl-phosphatidylcholine sphingomyelin aquaporin-1 carbonic anhydrase human AQP1. In contrast, Endeward et al. (3Endeward V. Musa-Aziz R. Cooper G.J. Chen L.M. Pelletier M.F. Virkki L.V. Supuran C.T. King L.S. Boron W.F. Gros G. FASEB J. 2006; 20: 1974-1981Crossref PubMed Scopus (179) Google Scholar) reported that AQP1 accounts for over 50% of PM,CO2 in normal red blood cells at physiological pH. Loss of Rh protein complex reduced PM,CO2 from initially 0.15 to 0.07 cm/s (5Endeward V. Cartron J.P. Ripoche P. Gros G. FASEB J. 2007; 22: 1-11Google Scholar). Because the PM,CO2 of 0.07 cm/s is in sharp contrast to the predicted value of 12 cm/s, this study aims to elucidate whether CO2 is an exception from Overtone's rule. In case of applicability of Overtone's rule to CO2 transport, neither AQP1 nor Rh proteins would be able to facilitate CO2 membrane diffusion. Planar Lipid Membranes—Planar bilayer lipid membranes were formed from the following: (i) pure diphytanoyl-phosphatidylcholine (DPhPC), (ii) from a 3:2:1 mixture of cholesterol, DPhPC, and egg sphingomyelin (SM, 84% palmitoyl-SM, 6% stearoyl-SM, 10% longer saturated acyl chain SM), or (iii) from a mixture that mimicked the lipid composition of the red cell plasma membrane. It was composed of egg phosphatidylcholine, egg phosphatidylethanolamine, brain phosphatidylserine, cholesterol, and SM (all Avanti Polar Lipids, Alabaster, AL) (14Devaux P.F. Seigneuret M. Biochim. Biophys. Acta. 1985; 822: 63-125Crossref PubMed Scopus (243) Google Scholar). The lipids were dissolved at 20 mg/ml in n-decane (15Mueller P. Rudin D.O. Tien H.T. Wescott W.C. J. Phys. Chem. 1963; 67: 534-535Crossref Scopus (387) Google Scholar) and were spread across a 250–300-μm circular hole in a septum separating two aqueous compartments (each 1.4 ml) of a Teflon chamber. Bilayer capacitance and conductance were continuously monitored by a picoamperometer (model VA10, NPI Electronic GmbH, Tamm, Germany) connected via Ag/AgCl electrodes situated at both sides of the membrane. The aqueous salt solutions contained 66 mm NaCl (Merck). Depending on pH, they were either buffered with 30 mm HEPES (Fluka, Vienna, Austria) or 30 mm CAPSO (Sigma). After a stable planar membrane was formed NaHCO3, Na2CO3, and carbonic anhydrase (Merck) were added to one or both sides of the membrane (see figure legends), and the chamber was covered with a Teflon lid. Magnetic stirrer bars continuously agitated the solutions. Belonging to the class of weak acids, CO2 transport across membranes includes several steps (Fig. 1) as follows: (a) the diffusion of HCO3- (pKHCO3 = 6.1) and CO32- (pKCO3 = 10) to the planar membrane; (b) proton uptake; (c) diffusion of CO2 through the bilayer; (d) dissociation, and (e) diffusion of HCO3- and CO32- into the bulk. The buffering solution provides a proton source and a proton sink in the cis and trans USL, respectively. The chemical reactions were catalyzed by the presence of carbonic anhydrase (CA). Cell Culture—Stably aquaporin-1-overexpressing Madin-Darby canine kidney (MDCK-AQP1) cells (16Deen P.M. Nielsen S. Bindels R.J. van Os C.H. Pfluegers Arch. 1997; 433: 780-787Crossref PubMed Scopus (45) Google Scholar) and MDCK cells were cultured in Dulbecco's modified Eagle's medium supplemented with 110 mg/liter sodium pyruvate, 584 mg/liter l-glutamine, nonessential amino acids, 5% fetal calf serum, 20 mm HEPES, 0.1% NaHCO3, and penicillin/streptomycin at 37 °C in 8.5% CO2. The AQP1 expression vector includes the hygromycin B resistance gene. It allowed maintenance of selective pressure by adding 75 μg/ml hygromycin B to the AQP1-MDCK cell medium (16Deen P.M. Nielsen S. Bindels R.J. van Os C.H. Pfluegers Arch. 1997; 433: 780-787Crossref PubMed Scopus (45) Google Scholar). For microelectrode measurements, the cells were seeded 1:1 onto semipermeable supports (Transwell, Costar) with a surface area of 0.33 cm2 and cultured again until the electrical resistance reached >3 kΩ indicating a tight monolayer (usually after 3–4 days). All cell experiments were carried out in HBBS buffer (HBBS buffer = 118 mm NaCl, 4.6 mm KCl, 10 mm glucose, and 20 mm HEPES, pH 7.4) at 37 °C. Immunoblots—Growing cells were harvested by scraping, washed twice in cold PBS, and lysed in RIPAII (500 mm NaCl; 50 mm Tris/HCl, pH 7.4; 0.1% SDS; 1% Nonidet P-40; 0.5% sodium deoxycholate; 0.05% NaN3) for 30 min on ice. Lysates were sonified, and the cell debris was pelleted with 12,000 × g. The total protein concentration of the supernatant was determined by Bradford protein assay (Pierce). A quantity of 20 μg (plus one additional lane with 10 μg for the MDCK-AQP1 sample) of total protein per lane was separated by 12% SDS-PAGE on Mini Protean III (Bio-Rad) and transferred to polyvinylidene difluoride membranes (Pall). Membranes were blocked with 5% milk powder in 0.05% Tween/PBS, incubated with anti-hAQP1 antibody (Alpha Diagnostics), and the secondary antibodies Goat Anti-Rabbit IgG (Sigma) diluted in 1% milk powder/Tween/PBS. The target proteins on the membrane were detected by chemiluminescence using Bio-Rad Universal Hood II. Immunofluorescence—Stably transfected MDCK-hAQP1 or nontransfected MDCK cells were seeded on coverslips and cultivated for at least 48 h before fixation. Cells were washed twice with PBS, fixed with 2% formaldehyde/PBS for 20 min at room temperature, washed twice with PBS, permeabilized with 0.2% Triton X-100/PBS for 20 min, and saturated with 0.2% fish gelatin. The cells were incubated with anti-hAQP1 antibody (10 μg/ml) followed by FITC-labeled secondary antibody. The coverslips were fixed on glass slides with DAKO fluorescent mounting medium. Cells were analyzed and pictured by a Zeiss confocal microscope (LSM 510). Microelectrode Measurements—Because all kinetic measurements of CO2 flux are compromised by USL effects, we monitored the accompanying pH changes in the steady state. A scanning pH-sensitive microelectrode was moved by a hydraulic microdrive manipulator (Narishige, Tokyo, Japan) within the stagnant water layer with a velocity of 2 μm/s toward or away from the membrane (17Antonenko Y.N. Denisov G.A. Pohl P. Biophys. J. 1993; 64: 1701-1710Abstract Full Text PDF PubMed Scopus (55) Google Scholar). Addition of CA did not affect the sensitivity of the electrode. Voltage recordings were performed every second with an electrometer (model 6514, Keithley Instruments), connected via an IEEE-488 interface to a personal computer. The electrodes were manufactured from borosilicate glass that was pulled to a tip size of 2–4 μm, silanized (bis(dimethylamino)dimethylsilane, Fluka), and filled with a proton-sensitive mixture (Hydrogen Ionophore II mixture A, Selectophore, Fluka). Analytical Model—Similar to our previous weak acid/base transport model (9Antonenko Y.N. Pohl P. Denisov G.A. Biophys. J. 1997; 72: 2187-2195Abstract Full Text PDF PubMed Scopus (68) Google Scholar, 17Antonenko Y.N. Denisov G.A. Pohl P. Biophys. J. 1993; 64: 1701-1710Abstract Full Text PDF PubMed Scopus (55) Google Scholar), all relevant proton-transfer reactions and diffusion processes were taken into account. Using Fick's 1st and 2nd laws, a set of coupled differential Equations 2 and 3 was written, Jicis(x)=-Didcicis(x)dx,dJicis(x)dx=Ri(ccis(x)),ccis=(c1cis,...,c8cis),xin(0,δ),i=1,...,8,(Eq. 2) Jitrans(x)=-Didcitrans(x)dx,dJitrans(x)dx=Ri(ctrans(x))ctrans=(c1trans,...,c8trans),xin(δ,2δ)(Eq. 3) where Jicis, cicis, Jitrans, and citrans are, respectively, the fluxes and the concentrations of the ith species in the membrane vicinity in the cis and trans compartments. The index i denotes the following species: 1 = H+,2 = OH-, 3 = A-, 4 = AH, 5 = CO2, 6 = HCO3-, 7 = CO32-, 8 = H2O. A- and AH are, respectively, deprotonated and protonated buffer molecules. Di and Ri symbolize the corresponding diffusion coefficients and the specific local rates of expenditure as shown in Equation 4, R1(c)=(k+1c8-k-1c1c2)+(k-2c5-k+2c1c6)+(k-3c6-k+3c1c7)+(k+5c4-k-5c1c3)R2(c)=(k+1c8-k-1c1c2)+(k+4c6-k-4c5c2)R3(c)=-R4(c)=(k+5c4-k-5c1c3)R5(c)=(k+2c1c6-k-2c5)+(k+4c6-k-4c2c5)R6(c)=(k-2c5-k+2c1c6)+(k+3c1c7-k-3c6)+(k-4c5c2-k+4c6)R7(c)=(k-3c6-k+3c1c7)R8(c)=(k-1c1c2-k+1c8)(Eq. 4) valid for both c = ccis and c = ctrans. The notations of kinetic rates and equilibrium constants are depicted in Fig. 1. The corresponding values are listed in Table 1.TABLE 1List of parameters used in our computations The dissociation constant is expressed as K = 10-pK = kd/ka. Parameters and parameter ranges, respectively, are taken from Refs. 7Saparov S.M. Antonenko Y.N. Pohl P. Biophys. J. 2006; 90: L86-L88Abstract Full Text Full Text PDF PubMed Scopus (66) Google Scholar, 12Hub J.S. de Groot B.L. Biophys. J. 2006; 91: 842-848Abstract Full Text Full Text PDF PubMed Scopus (125) Google Scholar, 17Antonenko Y.N. Denisov G.A. Pohl P. Biophys. J. 1993; 64: 1701-1710Abstract Full Text PDF PubMed Scopus (55) Google Scholar, 18Gutknecht J. Bisson M.A. Tosteson D.C. J. Gen. Physiol. 1977; 69: 779-794Crossref PubMed Scopus (292) Google Scholar, 39Gutman M. Nachliel E. Biochim. Biophys. Acta. 1995; 1231: 123-138Crossref Scopus (82) Google Scholar.ParameterValueD19.31 × 10–9 (m2 s–1)D25.26 × 10–9 (m2 s–1)D3,45.1 × 10–10 (m2 s–1)D52.9 × 10–9 (m2 s–1)D6,72 × 10–9 (m2 s–1)D82.5 × 10–9 (m2 s–1)pKa15.74pKb6.1pKc10kd1.26 × 10–8pKe9.6k–15 × 1011 (m3 s–1kmol–1)k+22 × 106–2 × 107 (m3 s–1kmol–1)k+31 × 1010–2 × 1010 (m3 s–1kmol–1)k-41 × 108–8.5 × 109 (m3 s–1kmol–1)k-51 × 1010–2 × 1010 (m3 s–1kmol–1) Open table in a new tab At x = δ the fluxes of all species are required to be equal to zero except for CO2, where we have Equations 5, 6, 7. J5cis=PM,CO2(c5cis-c5trans)(Eq. 5) J5trans=-PM,CO2(c5cis-c5trans)(Eq. 6) with JM,CO2=|J5cis|=|J5trans|(Eq. 7) If the transmembrane CO2 flux, JM,CO2, is determined by the resistance of near-membrane USLs, the concentration difference c5cis-c5trans does not depend on PM,CO2. Consequently, only a lower boundary of PM,CO2 can be determined from JM,CO2 measurements. Above that boundary the analytical model (see Equations 2, 3, 4, 5, 6, 7) fit the experimental pH profiles perfectly well for any arbitrary chosen value of PM,CO2. To determine the lower boundary, a penalty term for PM,CO2 was introduced into the fitting procedure (see “Appendix”) that disfavors large PM,CO2 values. Finally, the experimental bulk concentrations are used as boundary conditions for the concentrations at x = 0 and x = 2δ. We started our experiments at nonphysiological high pH values because Gutknecht et al. (18Gutknecht J. Bisson M.A. Tosteson D.C. J. Gen. Physiol. 1977; 69: 779-794Crossref PubMed Scopus (292) Google Scholar) reported that for pH > 9 and saturating CA concentrations, USL effects are negligible. To test whether the lipid bilayer constitutes the main diffusion barrier for CO2 transport under these conditions, we imposed a CO2 concentration gradient across the membrane and measured resulting pH shifts in the immediate membrane vicinity on the trans side (the CO2 source) at various CO2 concentrations (Fig. 2). pH was detected by scanning microelectrodes that were moved stepwise in the direction normal to the planar bilayer (17Antonenko Y.N. Denisov G.A. Pohl P. Biophys. J. 1993; 64: 1701-1710Abstract Full Text PDF PubMed Scopus (55) Google Scholar, 19Pohl P. Antonenko Y.N. Rosenfeld E.H. Biochim. Biophys. Acta. 1993; 1152: 155-160Crossref PubMed Scopus (39) Google Scholar). Increasing concentrations of CO2 resulted in increasing pH shifts from bulk pH of 9.6. Within a distance of ±50 μm from the membrane surface, the analytical model (Equations 2, 3, 4, 5, 6, 7) was fitted to the experimental pH profiles (Fig. 2d, gray lines) using inverse problem techniques (see “Appendix”). A PM,CO2 value of 3.2 cm/s was determined. The system converged with a zero penalty term (see “Experimental Procedures”). Because this term penalizes large PM,CO2 values, which may be computed in case flux limitations by USLs, the observation indicated that bilayer resistance to CO2 flux (JCO2) was not negligible. We subsequently increased the penalty term and found that the lower boundary of PM,CO2 that described the measured profile within the experimental error was equal to 1.6 cm/s. To demonstrate the validity of Overtone's law, a theoretical profile was calculated assuming that PM,CO2 = PM,NH3 = 0.016 cm/s (e). Its discrepancy with our experiment clearly demonstrates that PM,CO2 >> PM,NH3. From the experimental pH profiles (Fig. 2a) the corresponding theoretical CO2 profiles were calculated (Fig. 3). They revealed that most of the resistance (∼85%) to CO2 movement is because of its diffusion through the USL. The observation that membrane permeation is not rate-limiting suggests that the calculated PM,CO2 of 3.2 ± 1.6 cm/s is likely to be an underestimation (compare Refs. 9Antonenko Y.N. Pohl P. Denisov G.A. Biophys. J. 1997; 72: 2187-2195Abstract Full Text PDF PubMed Scopus (68) Google Scholar, 17Antonenko Y.N. Denisov G.A. Pohl P. Biophys. J. 1993; 64: 1701-1710Abstract Full Text PDF PubMed Scopus (55) Google Scholar). It is important to note that within the first few micrometers the protonation-deprotonation reactions of the weak acid are not in equilibrium. Similar reaction layers have previously been observed in the case of acetaldehyde transport at low alcohol dehydrogenase concentrations (20Antonenko Y.N. Pohl P. Rosenfeld E.H. Arch. Biochem. Biophys. 1996; 333: 225-232Crossref PubMed Scopus (17) Google Scholar). For H+, HCO3-, and CO2 to be in equilibrium, the deprotonation rate of HCO3- would have to be in the range of 1013–1014 m-1 s-1, which is clearly nonphysiological. If the permeability value close to that of ammonia is assumed, the steep gradients in the immediate membrane vicinity disappear, and the equilibrium is maintained throughout the USL (Fig. 3B). To study CO2 permeability under physiological conditions, bulk pH was decreased to 7.5 (Fig. 4A). The corresponding spatial CO2 distribution revealed the lack of a transmembrane CO2 concentration difference (Fig. 4B), indicating that at pH 7.5, CO2 permeability is completely USL-limited. Removal of CA from the cis side of the membrane decreased the flux 2–3-fold. Because of the decrease of the reaction rate, the reaction layer further extended into the solution and was reflected by the pH profile as a deviation from the simple exponential decay. Consequently, the size of the USL, δ, defined in terms of the pH gradient at the membrane water interface increased from 160 μm to more than 500 μm (Fig. 4A). In line with a previous report by Gutknecht et al. (18Gutknecht J. Bisson M.A. Tosteson D.C. J. Gen. Physiol. 1977; 69: 779-794Crossref PubMed Scopus (292) Google Scholar), total lack of CA in both compartments resulted in a large decrease of the CO2 flux and thus in an undetectable small interfacial pH shift (data not shown). If our conclusion was correct and CO2 diffusion through the membrane is not rate-limiting, changes of membrane composition are not expected to result in alterations of CO2 membrane permeability. Thus, in contrast to membrane permeabilities of water (21Hill W.G. Zeidel M.L. J. Biol. Chem. 2000; 275: 30176-30185Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar, 22Krylov A.V. Pohl P. Zeidel M.L. Hill W.G. J. Gen. Physiol. 2001; 118: 333-340Crossref PubMed Scopus (71) Google Scholar) and ammonia (2Saparov S.M. Liu K. Agre P. Pohl P. J. Biol. Chem. 2007; 282: 5296-5301Abstract Full Text Full Text PDF PubMed Scopus (165) Google Scholar, 9Antonenko Y.N. Pohl P. Denisov G.A. Biophys. J. 1997; 72: 2187-2195Abstract Full Text PDF PubMed Scopus (68) Google Scholar, 23Lande M.B. Donovan J.M. Zeidel M.L. J. Gen. Physiol. 1995; 106: 67-84Crossref PubMed Scopus (306) Google Scholar), membrane tightening by cholesterol and SM should not alter the apparent CO2 permeability of 3.2 cm/s. At least three effects contribute to the reduction in membrane microviscosity (24Smaby J.M. Brockman H.L. Brown R.E. Biochemistry. 1994; 33: 9135-9142Crossref PubMed Scopus (168) Google Scholar, 25Huster D. Arnold K. Gawrisch K. Biochemistry. 1998; 37: 17299-17308Crossref PubMed Scopus (240) Google Scholar). (i) Cholesterol increases the packing density of lipids. (ii) Sphingomyelin molecules form strong hydrogen bonds with each other and with cholesterol. (iii) The fully saturated SM lowers the bilayer content of unsaturated fatty acids. In line with our anticipation, we did not observe significant differences in the CO2-coupled proton fluxes adjacent to (i) pure DPhPC membranes, (ii) DPhPC membranes tightened by cholesterol and SM, and (iii) bilayers mimicking the composition of red blood cell membranes (Fig. 5). As the presence of cholesterol and SM is associated with both decreased membrane fluidity and decreased membrane diffusion coefficients of water and solutes (23Lande M.B. Donovan J.M. Zeidel M.L. J. Gen. Physiol. 1995; 106: 67-84Crossref PubMed Scopus (306) Google Scholar, 26Finkelstein A. Zimmerberg J. Cohen F.S. Annu. Rev. Physiol. 1986; 48: 163-174Crossref PubMed Google Scholar), the CO2 flux invariability indicates that the lipid bilayer does not act as a barrier to CO2 diffusion. Because USL adjacent to biological membranes are about 2 orders of magnitude smaller than those adjacent to planar bilayers, we calculated the contribution a 1-μm-thick USL would make to CO2 membrane resistance. Solving the inverse problem with all model parameters but δ unaltered (Table 1, pH 7.5), we found that such an USL still imposes ∼95% of the total resistance to CO2 flow (Fig. 6). Because, for example, red blood cells have a USL that exceeds 1 μm in size (27Huxley V.H. Kutchai H. J. Physiol. (Lond.). 1981; 316: 75-83Crossref Scopus (49) Google Scholar, 28Coin J.T. Olson J.S. J. Biol. Chem. 1979; 254: 1178-1190Abstract Full Text PDF PubMed Google Scholar, 29Vandegriff K.D. Olson J.S. J. Biol. Chem. 1984; 259: 12609-12618Abstract Full Text PDF PubMed Google Scholar), an AQP1-facilitated CO2 diffusion across red blood cell membranes is extremely unlikely. A meaningful physiological contribution of protein channels to CO2 transport remains doubtful even if they function in tight epithelial membranes. The prediction was confirmed by studying CO2 transport through monolayers of AQP1 overexpressing MDCK cells (Fig. 7). Although the presence of AQP1 in both the apical and basolateral membranes (Fig. 7, A and B) mediated a 3-fold increase in water flux (Fig. 7C), the pH profiles measured in the presence of a transepithelial CO2 gradient (Fig. 7D) indicated that the CO2 fluxes across AQP1 expressing and nonexpressing cells were identical (Fig. 7D).FIGURE 7Water and CO2 transport through monolayers formed by MDCK cells and by AQP1 overexpressing MDCK cells (MDCK-AQP1). A, immunoblot of cell lysates probed with AQP1 antibody. Both nonglycosylated (*) and glycosylated (**) forms of AQP1 were observed (compare also 38). B, immunofluorescence of MDCK-AQP1 cells. An antibody against human AQP1 and a FITC-labeled secondary antibody were used. Nontransfected MDCK cells as well as MDCK-AQP1 cells treated solely with FITC-labeled secondary antibody did not show immunofluorescence (data not shown). C, water permeability was measured by imposing an osmotic gradient of 380 mm (500 mm sorbitol added) over the epithelial monolayer and detecting the dilution of Mg2+ ions close to the basolateral membrane surface with a scanning ion-sensitive microelectrode (left panel). Seven independent runs of the experiment (n = 7) indicated that AQP1 accounted for 2/3 of the total water flux (inset). D, representative pH profiles measured in the presence of the indicated HCO3- gradients across monolayers formed by MDCK and MDCK-AQP1 cells. The slope at the membrane water interface is crucial for flux calculation. Differences in the profile course between MDCK and MDCK-AQP1 cells are solely because of variations in the stirring conditions. E, CO2 fluxes across AQP1 expressing and nonexpressing cells were identical (n = 10).View Large Image Figure ViewerDownload Hi-res image Download (PPT) As predicted by Overtone's rule, the lipid matrix of biological membranes does not represent a barrier to CO2 transport. Measurements of pH shifts in the immediate vicinity of planar bilayers and epithelial monolayers demonstrate that near-membrane USLs generate the main resistance to CO2 flow. Model simulations confirm that the result is valid for USLs less than 1-μm-thick. Because membrane tightening by cholesterol and SM does not reduce membrane CO2 permeability below 3.2 cm/s, facilitation of CO2 membrane transport by proteinaceous molecules is virtually impossible. Here, CO2 transport differs fundamentally from ammonia and water transport, which, in the absence of specific membrane channels, exhibit permeabilities that are orders of magnitude smaller (2Saparov S.M. Liu K. Agre P. Pohl P. J. Biol. Chem. 2007; 282: 5296-5301Abstract Full Text Full Text PDF PubMed Scopus (165) Google Scholar, 9Antonenko Y.N. Pohl P. Denisov G.A. Biophys. J. 1997; 72: 2187-2195Abstract Full Text PDF PubMed Scopus (68) Google Scholar, 22Krylov A.V. Pohl P. Zeidel M.L. Hill W.G. J. Gen. Physiol. 2001; 118: 333-340Crossref PubMed Scopus (71) Google Scholar). The lack of a facilitating effect of aquaporins on CO2 transport has been demonstrated before using knock-out mice (11Yang B.X. Fukuda N. van Hoek A. Matthay M.A. Ma T.H. Verkman A.S. J. Biol. Chem. 2000; 275: 2686-2692Abstract Full Text Full Text PDF PubMed Scopus (164) Google Scholar). However, the conclusion of this study" @default.
• W1992578464 created "2016-06-24" @default.
• W1992578464 creator A5002582958 @default.
• W1992578464 creator A5009442181 @default.
• W1992578464 creator A5027230438 @default.
• W1992578464 creator A5046446582 @default.
• W1992578464 creator A5046446817 @default.
• W1992578464 creator A5082440110 @default.
• W1992578464 creator A5082601284 @default.
• W1992578464 date "2008-09-01" @default.
• W1992578464 modified "2023-10-16" @default.
• W1992578464 title "Carbon Dioxide Transport through Membranes" @default.
• W1992578464 cites W1479762169 @default.
• W1992578464 cites W1584104642 @default.
• W1992578464 cites W1595402887 @default.
• W1992578464 cites W1964474152 @default.
• W1992578464 cites W1965927041 @default.
• W1992578464 cites W1975902868 @default.
• W1992578464 cites W1994914508 @default.
• W1992578464 cites W1995087505 @default.
• W1992578464 cites W1995093903 @default.
• W1992578464 cites W1999068193 @default.
• W1992578464 cites W2005279807 @default.
• W1992578464 cites W2007443819 @default.
• W1992578464 cites W2021181817 @default.
• W1992578464 cites W2029109058 @default.
• W1992578464 cites W2040969251 @default.
• W1992578464 cites W2045804403 @default.
• W1992578464 cites W2052491625 @default.
• W1992578464 cites W2054669438 @default.
• W1992578464 cites W2058577011 @default.
• W1992578464 cites W2059166957 @default.
• W1992578464 cites W2061124406 @default.
• W1992578464 cites W2064451603 @default.
• W1992578464 cites W2067008989 @default.
• W1992578464 cites W2072292027 @default.
• W1992578464 cites W2076936663 @default.
• W1992578464 cites W2113881348 @default.
• W1992578464 cites W2119861313 @default.
• W1992578464 cites W2130392636 @default.
• W1992578464 cites W2131720489 @default.
• W1992578464 cites W2151155243 @default.
• W1992578464 cites W2154756937 @default.
• W1992578464 cites W2166974541 @default.
• W1992578464 cites W2177637821 @default.
• W1992578464 cites W2200560911 @default.
• W1992578464 cites W2264495525 @default.
• W1992578464 cites W2288983040 @default.
• W1992578464 cites W2406179563 @default.
• W1992578464 cites W72955407 @default.
• W1992578464 doi "https://doi.org/10.1074/jbc.m800096200" @default.
• W1992578464 hasPubMedCentralId "https://www.ncbi.nlm.nih.gov/pmc/articles/2533081" @default.
• W1992578464 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/18617525" @default.
• W1992578464 hasPublicationYear "2008" @default.
• W1992578464 type Work @default.
• W1992578464 sameAs 1992578464 @default.
• W1992578464 citedByCount "144" @default.
• W1992578464 countsByYear W19925784642012 @default.
• W1992578464 countsByYear W19925784642013 @default.
• W1992578464 countsByYear W19925784642014 @default.
• W1992578464 countsByYear W19925784642015 @default.
• W1992578464 countsByYear W19925784642016 @default.
• W1992578464 countsByYear W19925784642017 @default.
• W1992578464 countsByYear W19925784642018 @default.
• W1992578464 countsByYear W19925784642019 @default.
• W1992578464 countsByYear W19925784642020 @default.
• W1992578464 countsByYear W19925784642021 @default.
• W1992578464 countsByYear W19925784642022 @default.
• W1992578464 countsByYear W19925784642023 @default.
• W1992578464 crossrefType "journal-article" @default.
• W1992578464 hasAuthorship W1992578464A5002582958 @default.
• W1992578464 hasAuthorship W1992578464A5009442181 @default.
• W1992578464 hasAuthorship W1992578464A5027230438 @default.
• W1992578464 hasAuthorship W1992578464A5046446582 @default.
• W1992578464 hasAuthorship W1992578464A5046446817 @default.
• W1992578464 hasAuthorship W1992578464A5082440110 @default.
• W1992578464 hasAuthorship W1992578464A5082601284 @default.
• W1992578464 hasBestOaLocation W19925784641 @default.
• W1992578464 hasConcept C12554922 @default.
• W1992578464 hasConcept C127413603 @default.
• W1992578464 hasConcept C178790620 @default.
• W1992578464 hasConcept C185592680 @default.
• W1992578464 hasConcept C41625074 @default.
• W1992578464 hasConcept C42360764 @default.
• W1992578464 hasConcept C530467964 @default.
• W1992578464 hasConcept C55493867 @default.
• W1992578464 hasConcept C86803240 @default.
• W1992578464 hasConceptScore W1992578464C12554922 @default.
• W1992578464 hasConceptScore W1992578464C127413603 @default.
• W1992578464 hasConceptScore W1992578464C178790620 @default.
• W1992578464 hasConceptScore W1992578464C185592680 @default.
• W1992578464 hasConceptScore W1992578464C41625074 @default.
• W1992578464 hasConceptScore W1992578464C42360764 @default.
• W1992578464 hasConceptScore W1992578464C530467964 @default.
• W1992578464 hasConceptScore W1992578464C55493867 @default.
• W1992578464 hasConceptScore W1992578464C86803240 @default.
• W1992578464 hasIssue "37" @default.
• W1992578464 hasLocation W19925784641 @default.

• next