Acta physiol. scand. 1979. 105. 129-136 From the Department of Physiology and Medical Biophysics, University of Uppsala, Sweden
A phenomenologic evaluation of C0,-diffusion restriction in kidney tubules studied in an artificial membrane system BY
MORGANSOHTELL Received 25 April 1978
Abstract SonmxL, M . A phenomenologic eualuation of CO,-d$fusion restriction in kidney tubules studied in an artificial membrane system. Acta physiol. scand. 1979. 105. 129-136. The chemical course in a multi-membrane system with interacting H + and HCO, ions has been described phenomenologically as an analogy of the neutralisation reaction between secreted H+ and filtered HCO, ions i n the proximal tubules of the kidney. It was shown that the produced CO, gave the highest Pco, in the asymmetrically placed reaction centre, which favours a build-up of a high intratubular Pco,. The CO , transport was dependent on the rate-limiting permeation of the reacting ions, and the permeation could be increased by the influence of solutions of macromolecules such as carbonic anhydrase, albumin and dextran.
Hydrogen ions are formed in the proximal tubular cells of the mammalian kidney and transported into the tubular lumen. Somewhere, the hydrogen ions react with the filtered bicarbonate ions, but the exact location of this neutralisation reaction is unknown. It probably occurs in the neighbourhood of the luminal cell membrane. Carbon dioxide is formed and diffuses to the surroundings by its tension gradients. The aim of this investigation was to describe phenomenologically the carbon dioxide diffusion out from the tubular lumen and how a build-up of a carbon dioxide difference between the tubular lumen and the tubular cell might occur. If hydrogen ions and bicarbonate ions are permitted to diffuse towards each other, a reaction will take place as follows: H'
+ HCO, 1 H,CO,
k,
CO, I H,O
k2
Carbonic acid will be formed, but as k, is about 330 times greater than k, (Garg et al. 1972) the equilibrium will be displaced to the right and carbon dioxide will be produced to a greater extent than carbonic acid. From that part of a system where the ions react with each other carbon dioxide will be formed, and carbon dioxide tension gradients will be built up. Rector et al. (1965) and Vieira and Malnic (1968) compared the in vivo p H of the proximal tubular fluid with that obtained for the same fluid in vitro after equilibration t o Pco, of 14 - 195872
129
130
MORGAN SOHTELL
40 mmHg. During control conditions this difference was small. After carbonic anhydrase inhibition, however, these mean differences ranged between 0.41 to 0.85 pH units. The results were interpreted as a chemical nonequilibrium of the bicarbonate buffer system, induced by the enzyme inhibition. Later experiments by Karlmark et al. (1974), however, pointed to the possibility of a higher Pco, in the tubular fluid than in the arterial blood. This was found to be quantitatively equivalent to the chemical non-equilibrium proposed. From direct measurements with a Pco,-microelectrode, Sohtell et al. (1976) obtained results which strongly favoured the assumption that the Pco,in the tubular fluid is higher than in the arterial blood. The investigation was carried out in a multi-membrane system in which the hydrogen ion secretion and bicarbonate reabsorption were simulated by diffusion processes. It was shown that the rate of transport of the carbon dioxide could be altered.
Material and Methods In order to study the reaction between H+ and HCO, coming into a membrane from either side, the membrane was regarded as being divided into nine slices. A multi-chamber system was thus developed, in which the chambers were separated by cellophane membranes. The complete equipment (see Fig. 1) consisted of 10 thermostated chambers. The two end chambers were each connected to a glass container (3 liters, called reservoirs I and I I in Fig. 1) via a punip. The end chamber of the hydrogen side was No. 1 and that of the bicarbonate side No. 10. The membrane between chambers 1 and 2 was called membrane No. I and so forth. Stirring mechanisms driven by electromagnets and silver-silver chloride electrodes for potential difference recordings were situated in the chambers (Fig. 2). In the present investigation the potential differences were recorded so as to be able to follow from outside the chambers the development of a steady state in the reaction system. The electrodes were connected to a mV-meter (PHM26, Radiometer, Copenhagen, Denmark). The experiments were performed at 25°C. The membranes were cut from cellophane tubings (C-75, Dialysis membrane, Union Carbide Corporation, Chicago, Illinois 60638). When studying events taking place in membranes, the influence of the surrounding unstirred layers has to be considered. The effective layer connected to smooth cellophane membranes is not reduced below the order of 0.03 mm, even with vigorous stirring. The thickness of the mem-
Thermostated water bath I
Fig. 1. The complete equipment showing the chambers mounted on to a guide stand and pressed together with wingnuts, the two reservoirs in the thermostated water bath, the pumps circulating thermostated water and end-chamber solutions, respectively, the stirring control of the electromagnets and the mV-meter.
COe-DIFFUSION IN KIDNEY TUBULES
Middle-Chamber- Block
131
End- Chamber - Block
Fig. 2. The chamber blocks: 1) stopcock, 2 ) Ag/AgCl electrode, 3) electrodeholder, 4) O-rings, 5) a teflon propeller with its spider and built-in rod magnet, 6) iron-core with channels for thermostated water, 7) coil, 8) holes for the guide stand, 9) and 10) inlet and outlet for the circulating end-chamber solution, 1 I ) connecting tube, 12) container for the barium-dihydroxide solution.
brane therefore has to exceed 0.2 to 0.8 mm to minimize the influence of the unstitred layers (Teorell 1936). I n this investigation the membranes were made about 0.7 mm thick (wet thickness), through superimposition of three pieces of the above-mentioned tubings (Teorell 1936). The end chamber concentrations of H+ and HCOT in this model experiment were so selected as to give forces whose relations were approximately comparable to the prevailing conditions in the proximal tubules in the kidney. In order t o evaluate the relative chemical forces (F) of the hydrogen ion secretion and the bicarbonate ion reabsorption, respectively, in the proximal tubules, the following equation and values were used: F
~~
A log %men CCd,
where A is a proportionality constant and C represents the concentration of the ion species. (H+):lumen= M (Waddel et al. 1969). (HCOi):lumen=25 x M 2.0 ,.' lo-' M (Karlmark et al. 1974), cell= 8.9 (a probable concentration of the glomerular filtrate in Bowman's space, without Donnan factor and plasma M (Khuri et al. 1974). This resulted in a water corrections, i.e. the plasma concentration), cell= 11.1 x relative driving force for bicarbonate ions of about 0.35 and for hydrogen ions of -0.35. In the,firsr experiment the following solutions were deposited in the respective chambers: No. 1 : 100 mM HCI, Nos 2-9: 5 mM NaCl and No. 10: 95 m M NaHCO,+ 5 mM NaCI. Chloride was placed in all chambers so that potential changes could be followed from the beginning, I n the secondexperiment the catalytic effect of three kinds of macromolecules on the reaction system was studied. The macromolecules were dissolved in 5 mM NaCl solution to equal final concentrations. The following solutions were studied: 0.05 m M human carbonic anhydrase type B (HCAB),' 0.05 mM human albumin (AB Kabi, Stockholm, Sweden) and 0.05 mM dextran (Dextran T4,,. Pharmacia Fine Chemicals, Uppsala, Sweden). These solutions were deposited in chambers 2-9. The concentration of carbonic anhydrase was about five times that in the kidney cell (Maren 1969).
-_
Kindly received from Dr P. Wistrand, Department of Pharmacology, Uppsala University, Uppsala, Sweden.
FYSIOLOGISKA I NSTl TUTlO NElV I KAROL1rJSKA IUS TI T U W T
132
MORGAN SOHTELL
In the third experiment the solutions were the same as in the first and second expts. The transport of carbon dioxide to chambers 1 and 10 was studied as the rate of carbon dioxide liberation from these chambers. The released carbon dioxide was absorbed and titrated by a barium-dihydroxide solution with bromthymol-blue as an indicator. The solution was kept in a small cup situated in the absorption chamber, shown in Fig. 2. Analyses of about 500 ,ul of the content of each chamber were performed by the following methods:
(H+)
Hydroxyl titration with brom-thymol-blue as indicator, performed in a micro-titration device (Obrink 1955). (C1-) Micro-chloride titration according to the method of Ramsay (Ramsay et a/. 1955). (Na+) Integrating flame photometry (Oberg ef al. 1967). (HCO;) Bicarbonate was analysed as the total carbonic acid content according to a modified Conway unit (Obrink 1955). Pcoi Pco, measurements were performed with a Pco2 electrode (E5036, Radiometer, Copenhagen, Denmark) connected to a mV-meter (PHM26, Radiometer).
Results Fig. 3 shows the ion concentrations and the carbon dioxide tension profiles in the first experiment. The results are considered to have been obtained under a relatively steady state condition (after 5 days). According to the carbon dioxide profiles the neutralisation reaction between H+ and HCO; took place somewhere in the eighth membrane. The quotient between the Pco, gradients from the reaction centre will be equal to the Pco, quotient between the outermost membranes, as the beginnings of the gradients are the same. The Pco, in a membrane was calculated as the mean Pco, of the adjacent chambers. The Pco, in membranes 1 and 9 were 11.0 and 87.2 mmHg, respectively, which will give a quotient of 7.9. In the second experiment, where chambers 2 to 9 contained solutions of HCAB, albumin and dextran, a good steady state was obtained. The Pco, at the reaction centre, the peak
Na+
mM
Tco,
Fig. 3. The results in the first experiment obtained after a diffusion time of five days. Tco, represent the total carbonic acid content.
Number of chamber
133
CO,-DIFFUSION IN KIDNEY TUBULES
*
Fig. 4. The carbon dioxide profiles obtained in the second experiment with different solutions in chambers 2-9: 0-0 HCAB, O - - - O human Albumin, O . . . . Dextran Ta0and the profile with the lower peak value from the first experiment: 0-0.
Number of chamber
values, were higher than in the first experiment (see Fig. 4). The quotients for the carbon dioxide gradients were 3.7, 5.0 and 4.0 for HCAB, albumin and dextran, respectively. Table I shows the carbon dioxide liberation rates studied in the third experiment. The results have been normalized by setting the liberation rate from chamber No. 1 at 1 when chambers 2 to 9 contained 5 mM NaCl. The table also gives the quotients of the liberation rates from chambers 10 and 1 for the four middle chamber solutions examined. TABLE I. The liberation rates of carbon dioxide and the quotients of the liberation rates between chambers 10 and 1 as obtained from the third experiment. These quotients are to he compared with the quotients for the carbon dioxide tension gradients in the first and second experiments. ~~
~
~~
Solutions
CO, liberation/min
Chamber Nos. 2-9
No. 1
No. 10
Chamber No. lO/l
5 m M NaCl
1 .o 6.0 10.4
9.3 21.7 31.4 20.3
9.3 3.6
5 mM NaCl+albumin 5 m M NaCI k HCAB 5 m M NaCl I dextran
1.5
3.0
2.1
134
MORGAN SOHTELL
Discussion In a system with several chambers in series, Teorell (1936) studied the distribution of ions in a membrane made up of the chamber-separating membranes. The ions were found to be distributed according to their mobilities and concentration differences in the chamber system. He also studied the effect on ion distribution of a chemical reaction within the chamber system. From the distribution and the effects of concentration differences in the considered system the location of the reaction can be calculated. The concentrations of hydrogen and bicarbonate ions are assumed to be equal in either end of the system. The relative mobilities of these ions are 350 and 43.5, respectively, at 25°C (Handbook of Chemistry and Physics 1971). If the length of the system is set at 1.0 and the mobilities of the reacting ions are assumed not to be influenced by other ions, the reaction location will occur 0.11/1 of the distance from the side where bicarbonate emanates. In steady state the concentration profiles are linear to the point of reaction, which then means that the carbon dioxide concentration gradient and thus the rate of carbon dioxide transport towards the bicarbonate side is theoretically 8.0 times greater than that towards the hydrogen side. This value is in accordance with the result of 7.9 obtained in the first experiment, thus also verifying the assymetrically located reaction centre. A number of reports have described a facilitation of carbon dioxide transport through different boundaries. The concentrations of the reacting ions in the different chamber compartments in the present investigation are considered to be homogenous, but not in the membranes and their connecting unstirred layers. Gutknecht et al. (1977) found that carbon dioxide diffusion can be rate-limiting for total carbon dioxide transport in lipid bilayers with unstirred layers, as the neutralisation reaction is too slow to permit even a high bicarbonate concentration to speed up the carbon dioxide production more than 15%. In carbonicanhydrase-catalyzed reactions they found that the bicarbonate ion mobility in itself would be rate-limiting. This supports the finding of a lower peak Pco, in the case of the uncatalyzed reaction as compared with the catalyzed one, which thus could be an effect of a slower bicarbonate permeation through the unstirred layer. Suchdeo ef a!. (1974) found a 3 to 10fold increase of the carbon dioxide flux through a Gelman membrane when adding carbonic anhydrase, and Enns (1967) reported a doubled transport rate through a Millipore filter with washed red-cell ghosts or filtrate from the red-cell solution deposited on one side of the membrane. The relative resistance to permeation of the reacting ions in the unstirred layers can be decreased by co-transport in translational diffusion of macromolecules. Gros et a!. (1974) observed that thin layers of albumin solutions facilitated hydrogen ion diffusion and they also found greater permeation of bicarbonate ions, which was explained as a secondary effect of the increased hydrogen ion diffusion. The facilitating effect of the dextran solution is not fully understood. The results from the second experiment, in which the macromolecular solutions were used, showed a tendency to enhanced transport of carbon dioxide towards the hydrogen and bicarbonate ion side, indicated by the increased Pco,gradients compared with the first experiment. This facilitation was demonstrated by the carbon dioxide liberation study in the third experiment.
C02-DIFFUSION IN KIDNEY TUBULES
135
Recently Wistrand et al. (1977) showed the existence of an enzyme that is kinetically similar to HCAB in the brush border and peritubular membranes of the rat kidney. The membrane enzyme was found to have 3 % of the total cell enzyme activity. If the distribution of proteins inside the cell is equal to that in the membranes, this should lead to a somewhat higher concentration in the membranes. According t o Maren (1969), however, only 0.1 '% of the enzyme supply in the tubular tissue is utilized. The amount of carbonic anhydrase in the present investigation is therefore regarded as being in a suitable excess. The increasing effect of carbonic anhydrase o n the reaction rate is, however, of minor importance in the present system. As shown in Fig. 4,all the macromolecules studied give about the same carbon dioxide profile. This means that the macromolecular reduction of the resistance to permeation of the reacting ions is of the same order. The permeation of the ions thus seems t o be rate-limiting for carbon dioxide transport. The effect of the carbonic anhydrase on the reaction rate would probably have been more pronounced if thecarbon dioxide transport had been rate-limiting in this membrane system. The driving force for carbon dioxide diffusion through a membrane should be increased by an enzymatic conversion of carbon dioxide into bicarbonate. The phenomenologically described chemical course in this investigation on interacting hydrogen and bicarbonate ions could be analogous to the situation in the proximal tubules. The finding of the elevated Pcozin the proximal tubules (Sohtell et al. 1976) may be a result of a neutralisation reaction taking place adjacent to the luminal cell membrane. The elevated Pco, in the reaction centre would create an increased luminal fluid Pco, and a build-up of a gradient for carbon dioxide transport out from the tubular lumen. The results also indicate that the net carbon dioxide transport is enhanced by carbonic anhydrase, by its increasing effect on the permeation of hydrogen and bicarbonate ions.
References ENNS,T., Facilitation by carbonic anhydrase of carbon dioxide transport. Science 1967. 155. 44-47. GARG,L. C. and T. H. MAREN, The rates of hydration of carbon dioxide and dehydration of carbonic acid a t 37°C. Biochem. biuphy.7. Acta (Amst.) 1972. 261. 70-76. GROS,G. and W. MOLL, Facilitated diffusion of CO, across albumin solutions. J . gen. Phvsiol. 1974. 64. 356-371, GUTKNECHT, J., M. A. BISON and F. C. TOSTESON, Diffusion of carbon dioxide through lipid bilayer membranes. Effects of carbonic anhydrase, bicarbonate and unstirred layers. f. gen. Physiol. 1977. 69. 779-794. Handbook of Chemistry and Physics, 51 Ed., 1971. The Chemical Rubber Co., Cleveland, Ohio. KARLMARK, B. and B. G. DANIELSON, Titrable acid, Pco,, bicarbonate and ammonium ions along the rat proximal tubule. Acta physiol. w a n d 1974. 91. 243-258. KHURI,R. N., S. K. AQULIAN, K. BOGAARIAN, R. NASSARand W. WISE,Intracellular bicarbonate in single cells of necturus kidney proximal tubules. Pfliigers Arch. ges. Physiol. 1974. 349. 295-299. MAREN, T. H., Renal carbonic anhydrase and the pharmacology of sulfonaniide inhibitors. Handb. E x p . Pharm. 1969. 24. 195-256. BERG, A., H . R . ULFENDAHL and G. WALLIN, An integrating flame photometer for simultaneous microanalysis of sodium and potassium in biological fluids. Analyt. Biochem. 1967. 18. 543-558. OBRINK. K . J., A modified Conway unit for microdiffusion analysis. Biuchem. J . 1955. 59: 1. 134. RAMSAY, J. A., R. H. J. BROWN and P. C. CROGHAN, Electrometric titration of chloride in small volumes. f. exp. Biol. 1955. 32. 822-829. RECTOR,F. C., N . W . CARTER and D. W. SELDIN, with the technical assistance of A. C. NUNN,The mechan-
136
MORGAN SOHTELL
ism of bicarbonate reabsorption in the proximal and distal tubules of the kidney. J. clin. Inuesf. 1965. 44: 2. 278-290. SOHTELL, M. and B. KARLMARK, In vivo micropuncture PCO,measurements. Pflugers Arch. ges. Physiol. 1976. 363. 179-180. SUCHDEO, S. R. and J. S. SCHULTZ, Mass transfer of CO, across membranes: facilitation in the presence of bicarbonate ion and of the enzyme carbonic anhydrase. Biochem. biuphys. Acta (Amst.) 1974. 352. 412440. TEORELL, T., A method of studying conditions within diffusion layers. J . biol. Chem. 1936. 113. 735-748. WADDEL,W. J. and R. G. BATES,Intracellular pH. Physiol. Reu. 1969. 49. 285-329. VIEIRA,F. C. and G. MALNIC, Hydrogen ion secretion by rat renal cortical tubules as studied by an antimony microelectrode. Amer. J . Physiul. 1968. 214. 710-718. WISTRAND, P. J. and R. KINNE,Carbonic anhydrase activity of isolated brush border and basal-lateral membranes of renal tubular cells. Pftugers Arch. ges. Phvsiul. 1972. 370. 121-126.
A phenomenologic evaluation of C0,-diffusion restriction in kidney tubules studied in an artificial membrane system BY
MORGANSOHTELL Received 25 April 1978
Abstract SonmxL, M . A phenomenologic eualuation of CO,-d$fusion restriction in kidney tubules studied in an artificial membrane system. Acta physiol. scand. 1979. 105. 129-136. The chemical course in a multi-membrane system with interacting H + and HCO, ions has been described phenomenologically as an analogy of the neutralisation reaction between secreted H+ and filtered HCO, ions i n the proximal tubules of the kidney. It was shown that the produced CO, gave the highest Pco, in the asymmetrically placed reaction centre, which favours a build-up of a high intratubular Pco,. The CO , transport was dependent on the rate-limiting permeation of the reacting ions, and the permeation could be increased by the influence of solutions of macromolecules such as carbonic anhydrase, albumin and dextran.
Hydrogen ions are formed in the proximal tubular cells of the mammalian kidney and transported into the tubular lumen. Somewhere, the hydrogen ions react with the filtered bicarbonate ions, but the exact location of this neutralisation reaction is unknown. It probably occurs in the neighbourhood of the luminal cell membrane. Carbon dioxide is formed and diffuses to the surroundings by its tension gradients. The aim of this investigation was to describe phenomenologically the carbon dioxide diffusion out from the tubular lumen and how a build-up of a carbon dioxide difference between the tubular lumen and the tubular cell might occur. If hydrogen ions and bicarbonate ions are permitted to diffuse towards each other, a reaction will take place as follows: H'
+ HCO, 1 H,CO,
k,
CO, I H,O
k2
Carbonic acid will be formed, but as k, is about 330 times greater than k, (Garg et al. 1972) the equilibrium will be displaced to the right and carbon dioxide will be produced to a greater extent than carbonic acid. From that part of a system where the ions react with each other carbon dioxide will be formed, and carbon dioxide tension gradients will be built up. Rector et al. (1965) and Vieira and Malnic (1968) compared the in vivo p H of the proximal tubular fluid with that obtained for the same fluid in vitro after equilibration t o Pco, of 14 - 195872
129
130
MORGAN SOHTELL
40 mmHg. During control conditions this difference was small. After carbonic anhydrase inhibition, however, these mean differences ranged between 0.41 to 0.85 pH units. The results were interpreted as a chemical nonequilibrium of the bicarbonate buffer system, induced by the enzyme inhibition. Later experiments by Karlmark et al. (1974), however, pointed to the possibility of a higher Pco, in the tubular fluid than in the arterial blood. This was found to be quantitatively equivalent to the chemical non-equilibrium proposed. From direct measurements with a Pco,-microelectrode, Sohtell et al. (1976) obtained results which strongly favoured the assumption that the Pco,in the tubular fluid is higher than in the arterial blood. The investigation was carried out in a multi-membrane system in which the hydrogen ion secretion and bicarbonate reabsorption were simulated by diffusion processes. It was shown that the rate of transport of the carbon dioxide could be altered.
Material and Methods In order to study the reaction between H+ and HCO, coming into a membrane from either side, the membrane was regarded as being divided into nine slices. A multi-chamber system was thus developed, in which the chambers were separated by cellophane membranes. The complete equipment (see Fig. 1) consisted of 10 thermostated chambers. The two end chambers were each connected to a glass container (3 liters, called reservoirs I and I I in Fig. 1) via a punip. The end chamber of the hydrogen side was No. 1 and that of the bicarbonate side No. 10. The membrane between chambers 1 and 2 was called membrane No. I and so forth. Stirring mechanisms driven by electromagnets and silver-silver chloride electrodes for potential difference recordings were situated in the chambers (Fig. 2). In the present investigation the potential differences were recorded so as to be able to follow from outside the chambers the development of a steady state in the reaction system. The electrodes were connected to a mV-meter (PHM26, Radiometer, Copenhagen, Denmark). The experiments were performed at 25°C. The membranes were cut from cellophane tubings (C-75, Dialysis membrane, Union Carbide Corporation, Chicago, Illinois 60638). When studying events taking place in membranes, the influence of the surrounding unstirred layers has to be considered. The effective layer connected to smooth cellophane membranes is not reduced below the order of 0.03 mm, even with vigorous stirring. The thickness of the mem-
Thermostated water bath I
Fig. 1. The complete equipment showing the chambers mounted on to a guide stand and pressed together with wingnuts, the two reservoirs in the thermostated water bath, the pumps circulating thermostated water and end-chamber solutions, respectively, the stirring control of the electromagnets and the mV-meter.
COe-DIFFUSION IN KIDNEY TUBULES
Middle-Chamber- Block
131
End- Chamber - Block
Fig. 2. The chamber blocks: 1) stopcock, 2 ) Ag/AgCl electrode, 3) electrodeholder, 4) O-rings, 5) a teflon propeller with its spider and built-in rod magnet, 6) iron-core with channels for thermostated water, 7) coil, 8) holes for the guide stand, 9) and 10) inlet and outlet for the circulating end-chamber solution, 1 I ) connecting tube, 12) container for the barium-dihydroxide solution.
brane therefore has to exceed 0.2 to 0.8 mm to minimize the influence of the unstitred layers (Teorell 1936). I n this investigation the membranes were made about 0.7 mm thick (wet thickness), through superimposition of three pieces of the above-mentioned tubings (Teorell 1936). The end chamber concentrations of H+ and HCOT in this model experiment were so selected as to give forces whose relations were approximately comparable to the prevailing conditions in the proximal tubules in the kidney. In order t o evaluate the relative chemical forces (F) of the hydrogen ion secretion and the bicarbonate ion reabsorption, respectively, in the proximal tubules, the following equation and values were used: F
~~
A log %men CCd,
where A is a proportionality constant and C represents the concentration of the ion species. (H+):lumen= M (Waddel et al. 1969). (HCOi):lumen=25 x M 2.0 ,.' lo-' M (Karlmark et al. 1974), cell= 8.9 (a probable concentration of the glomerular filtrate in Bowman's space, without Donnan factor and plasma M (Khuri et al. 1974). This resulted in a water corrections, i.e. the plasma concentration), cell= 11.1 x relative driving force for bicarbonate ions of about 0.35 and for hydrogen ions of -0.35. In the,firsr experiment the following solutions were deposited in the respective chambers: No. 1 : 100 mM HCI, Nos 2-9: 5 mM NaCl and No. 10: 95 m M NaHCO,+ 5 mM NaCI. Chloride was placed in all chambers so that potential changes could be followed from the beginning, I n the secondexperiment the catalytic effect of three kinds of macromolecules on the reaction system was studied. The macromolecules were dissolved in 5 mM NaCl solution to equal final concentrations. The following solutions were studied: 0.05 m M human carbonic anhydrase type B (HCAB),' 0.05 mM human albumin (AB Kabi, Stockholm, Sweden) and 0.05 mM dextran (Dextran T4,,. Pharmacia Fine Chemicals, Uppsala, Sweden). These solutions were deposited in chambers 2-9. The concentration of carbonic anhydrase was about five times that in the kidney cell (Maren 1969).
-_
Kindly received from Dr P. Wistrand, Department of Pharmacology, Uppsala University, Uppsala, Sweden.
FYSIOLOGISKA I NSTl TUTlO NElV I KAROL1rJSKA IUS TI T U W T
132
MORGAN SOHTELL
In the third experiment the solutions were the same as in the first and second expts. The transport of carbon dioxide to chambers 1 and 10 was studied as the rate of carbon dioxide liberation from these chambers. The released carbon dioxide was absorbed and titrated by a barium-dihydroxide solution with bromthymol-blue as an indicator. The solution was kept in a small cup situated in the absorption chamber, shown in Fig. 2. Analyses of about 500 ,ul of the content of each chamber were performed by the following methods:
(H+)
Hydroxyl titration with brom-thymol-blue as indicator, performed in a micro-titration device (Obrink 1955). (C1-) Micro-chloride titration according to the method of Ramsay (Ramsay et a/. 1955). (Na+) Integrating flame photometry (Oberg ef al. 1967). (HCO;) Bicarbonate was analysed as the total carbonic acid content according to a modified Conway unit (Obrink 1955). Pcoi Pco, measurements were performed with a Pco2 electrode (E5036, Radiometer, Copenhagen, Denmark) connected to a mV-meter (PHM26, Radiometer).
Results Fig. 3 shows the ion concentrations and the carbon dioxide tension profiles in the first experiment. The results are considered to have been obtained under a relatively steady state condition (after 5 days). According to the carbon dioxide profiles the neutralisation reaction between H+ and HCO; took place somewhere in the eighth membrane. The quotient between the Pco, gradients from the reaction centre will be equal to the Pco, quotient between the outermost membranes, as the beginnings of the gradients are the same. The Pco, in a membrane was calculated as the mean Pco, of the adjacent chambers. The Pco, in membranes 1 and 9 were 11.0 and 87.2 mmHg, respectively, which will give a quotient of 7.9. In the second experiment, where chambers 2 to 9 contained solutions of HCAB, albumin and dextran, a good steady state was obtained. The Pco, at the reaction centre, the peak
Na+
mM
Tco,
Fig. 3. The results in the first experiment obtained after a diffusion time of five days. Tco, represent the total carbonic acid content.
Number of chamber
133
CO,-DIFFUSION IN KIDNEY TUBULES
*
Fig. 4. The carbon dioxide profiles obtained in the second experiment with different solutions in chambers 2-9: 0-0 HCAB, O - - - O human Albumin, O . . . . Dextran Ta0and the profile with the lower peak value from the first experiment: 0-0.
Number of chamber
values, were higher than in the first experiment (see Fig. 4). The quotients for the carbon dioxide gradients were 3.7, 5.0 and 4.0 for HCAB, albumin and dextran, respectively. Table I shows the carbon dioxide liberation rates studied in the third experiment. The results have been normalized by setting the liberation rate from chamber No. 1 at 1 when chambers 2 to 9 contained 5 mM NaCl. The table also gives the quotients of the liberation rates from chambers 10 and 1 for the four middle chamber solutions examined. TABLE I. The liberation rates of carbon dioxide and the quotients of the liberation rates between chambers 10 and 1 as obtained from the third experiment. These quotients are to he compared with the quotients for the carbon dioxide tension gradients in the first and second experiments. ~~
~
~~
Solutions
CO, liberation/min
Chamber Nos. 2-9
No. 1
No. 10
Chamber No. lO/l
5 m M NaCl
1 .o 6.0 10.4
9.3 21.7 31.4 20.3
9.3 3.6
5 mM NaCl+albumin 5 m M NaCI k HCAB 5 m M NaCl I dextran
1.5
3.0
2.1
134
MORGAN SOHTELL
Discussion In a system with several chambers in series, Teorell (1936) studied the distribution of ions in a membrane made up of the chamber-separating membranes. The ions were found to be distributed according to their mobilities and concentration differences in the chamber system. He also studied the effect on ion distribution of a chemical reaction within the chamber system. From the distribution and the effects of concentration differences in the considered system the location of the reaction can be calculated. The concentrations of hydrogen and bicarbonate ions are assumed to be equal in either end of the system. The relative mobilities of these ions are 350 and 43.5, respectively, at 25°C (Handbook of Chemistry and Physics 1971). If the length of the system is set at 1.0 and the mobilities of the reacting ions are assumed not to be influenced by other ions, the reaction location will occur 0.11/1 of the distance from the side where bicarbonate emanates. In steady state the concentration profiles are linear to the point of reaction, which then means that the carbon dioxide concentration gradient and thus the rate of carbon dioxide transport towards the bicarbonate side is theoretically 8.0 times greater than that towards the hydrogen side. This value is in accordance with the result of 7.9 obtained in the first experiment, thus also verifying the assymetrically located reaction centre. A number of reports have described a facilitation of carbon dioxide transport through different boundaries. The concentrations of the reacting ions in the different chamber compartments in the present investigation are considered to be homogenous, but not in the membranes and their connecting unstirred layers. Gutknecht et al. (1977) found that carbon dioxide diffusion can be rate-limiting for total carbon dioxide transport in lipid bilayers with unstirred layers, as the neutralisation reaction is too slow to permit even a high bicarbonate concentration to speed up the carbon dioxide production more than 15%. In carbonicanhydrase-catalyzed reactions they found that the bicarbonate ion mobility in itself would be rate-limiting. This supports the finding of a lower peak Pco, in the case of the uncatalyzed reaction as compared with the catalyzed one, which thus could be an effect of a slower bicarbonate permeation through the unstirred layer. Suchdeo ef a!. (1974) found a 3 to 10fold increase of the carbon dioxide flux through a Gelman membrane when adding carbonic anhydrase, and Enns (1967) reported a doubled transport rate through a Millipore filter with washed red-cell ghosts or filtrate from the red-cell solution deposited on one side of the membrane. The relative resistance to permeation of the reacting ions in the unstirred layers can be decreased by co-transport in translational diffusion of macromolecules. Gros et a!. (1974) observed that thin layers of albumin solutions facilitated hydrogen ion diffusion and they also found greater permeation of bicarbonate ions, which was explained as a secondary effect of the increased hydrogen ion diffusion. The facilitating effect of the dextran solution is not fully understood. The results from the second experiment, in which the macromolecular solutions were used, showed a tendency to enhanced transport of carbon dioxide towards the hydrogen and bicarbonate ion side, indicated by the increased Pco,gradients compared with the first experiment. This facilitation was demonstrated by the carbon dioxide liberation study in the third experiment.
C02-DIFFUSION IN KIDNEY TUBULES
135
Recently Wistrand et al. (1977) showed the existence of an enzyme that is kinetically similar to HCAB in the brush border and peritubular membranes of the rat kidney. The membrane enzyme was found to have 3 % of the total cell enzyme activity. If the distribution of proteins inside the cell is equal to that in the membranes, this should lead to a somewhat higher concentration in the membranes. According t o Maren (1969), however, only 0.1 '% of the enzyme supply in the tubular tissue is utilized. The amount of carbonic anhydrase in the present investigation is therefore regarded as being in a suitable excess. The increasing effect of carbonic anhydrase o n the reaction rate is, however, of minor importance in the present system. As shown in Fig. 4,all the macromolecules studied give about the same carbon dioxide profile. This means that the macromolecular reduction of the resistance to permeation of the reacting ions is of the same order. The permeation of the ions thus seems t o be rate-limiting for carbon dioxide transport. The effect of the carbonic anhydrase on the reaction rate would probably have been more pronounced if thecarbon dioxide transport had been rate-limiting in this membrane system. The driving force for carbon dioxide diffusion through a membrane should be increased by an enzymatic conversion of carbon dioxide into bicarbonate. The phenomenologically described chemical course in this investigation on interacting hydrogen and bicarbonate ions could be analogous to the situation in the proximal tubules. The finding of the elevated Pcozin the proximal tubules (Sohtell et al. 1976) may be a result of a neutralisation reaction taking place adjacent to the luminal cell membrane. The elevated Pco, in the reaction centre would create an increased luminal fluid Pco, and a build-up of a gradient for carbon dioxide transport out from the tubular lumen. The results also indicate that the net carbon dioxide transport is enhanced by carbonic anhydrase, by its increasing effect on the permeation of hydrogen and bicarbonate ions.
References ENNS,T., Facilitation by carbonic anhydrase of carbon dioxide transport. Science 1967. 155. 44-47. GARG,L. C. and T. H. MAREN, The rates of hydration of carbon dioxide and dehydration of carbonic acid a t 37°C. Biochem. biuphy.7. Acta (Amst.) 1972. 261. 70-76. GROS,G. and W. MOLL, Facilitated diffusion of CO, across albumin solutions. J . gen. Phvsiol. 1974. 64. 356-371, GUTKNECHT, J., M. A. BISON and F. C. TOSTESON, Diffusion of carbon dioxide through lipid bilayer membranes. Effects of carbonic anhydrase, bicarbonate and unstirred layers. f. gen. Physiol. 1977. 69. 779-794. Handbook of Chemistry and Physics, 51 Ed., 1971. The Chemical Rubber Co., Cleveland, Ohio. KARLMARK, B. and B. G. DANIELSON, Titrable acid, Pco,, bicarbonate and ammonium ions along the rat proximal tubule. Acta physiol. w a n d 1974. 91. 243-258. KHURI,R. N., S. K. AQULIAN, K. BOGAARIAN, R. NASSARand W. WISE,Intracellular bicarbonate in single cells of necturus kidney proximal tubules. Pfliigers Arch. ges. Physiol. 1974. 349. 295-299. MAREN, T. H., Renal carbonic anhydrase and the pharmacology of sulfonaniide inhibitors. Handb. E x p . Pharm. 1969. 24. 195-256. BERG, A., H . R . ULFENDAHL and G. WALLIN, An integrating flame photometer for simultaneous microanalysis of sodium and potassium in biological fluids. Analyt. Biochem. 1967. 18. 543-558. OBRINK. K . J., A modified Conway unit for microdiffusion analysis. Biuchem. J . 1955. 59: 1. 134. RAMSAY, J. A., R. H. J. BROWN and P. C. CROGHAN, Electrometric titration of chloride in small volumes. f. exp. Biol. 1955. 32. 822-829. RECTOR,F. C., N . W . CARTER and D. W. SELDIN, with the technical assistance of A. C. NUNN,The mechan-
136
MORGAN SOHTELL
ism of bicarbonate reabsorption in the proximal and distal tubules of the kidney. J. clin. Inuesf. 1965. 44: 2. 278-290. SOHTELL, M. and B. KARLMARK, In vivo micropuncture PCO,measurements. Pflugers Arch. ges. Physiol. 1976. 363. 179-180. SUCHDEO, S. R. and J. S. SCHULTZ, Mass transfer of CO, across membranes: facilitation in the presence of bicarbonate ion and of the enzyme carbonic anhydrase. Biochem. biuphys. Acta (Amst.) 1974. 352. 412440. TEORELL, T., A method of studying conditions within diffusion layers. J . biol. Chem. 1936. 113. 735-748. WADDEL,W. J. and R. G. BATES,Intracellular pH. Physiol. Reu. 1969. 49. 285-329. VIEIRA,F. C. and G. MALNIC, Hydrogen ion secretion by rat renal cortical tubules as studied by an antimony microelectrode. Amer. J . Physiul. 1968. 214. 710-718. WISTRAND, P. J. and R. KINNE,Carbonic anhydrase activity of isolated brush border and basal-lateral membranes of renal tubular cells. Pftugers Arch. ges. Phvsiul. 1972. 370. 121-126.
