Michael R. Toon

The aqueous pore in the red cell membrane: band 3 as a channel for anions, cations, nonelectrolytes, and water.

This article develops arguments for the existence of an aqueous pore in the red cell membrane as the principal route for passive flux of ions, water, and small nonelectrolytes and proposes a molecular model for the pore. In principle, such an aqueous pore would provide easy passage into and out of the cell for all solutes small enough to enter the channel. The red cell membrane, however, regulates the fluxes of cations and anions closely and discriminates carefully among other small solutes. These constraints have been incorporated into the model, which visualizes the channel and its associated regulatory system as governing passive transport of ions of either sign, as well as water and small nonelectrolytes into and out of the cell. The model, which was formulated to consolidate a number of observations already in the literature, has caused us to look for new interrelations between inhibitors specific to cation, anion, and nonelectrolyte transport. The results of these experiments, presented below, demonstrate that interrelations d o exist and provide evidence that supports the view that a common aqueous channel provides primary access to the red cell cytoplasm.

Interaction between red cell membrane band 3 and cytosolic carbonic anhydrase

We have previously proposed that a membrane transport complex, centered on the human red cell anion transport protein, band 3, links the transport of anions, cations and glucose. Since band 3 is specialized for HCO3−/Cl− exchange, we thought there might also be a linkage with carbonic anhydrase (CA) which hydrates CO2 to HCO3−. CA is a cytosolic enzyme which is not present in the red cell membrane. The rate of reaction of CA with the fluorescent inhibitor, dansylsulfonamide (DNSA) can be measured by stopped-flow spectrofluorimetry and used to characterize the normal CA configuration. If a perturbation applied to a membrane protein alters DNSA/CA binding kinetics, we conclude that the perturbation has changed the CA configuration by either direct or allosteric means. Our experiments show that covalent reaction of the specific stilbene anion exchange inhibitor, DIDS, with the red cell membrane, significantly alters DNSA/CA binding kinetics. Another specific anion exchange inhibitor, benzene sulfonate (BSate), which has been shown to bind to the DIDS site causes a larger change in DNSA/CA binding kinetics; DIDS reverses the BSate effect. These experiments show that there is a linkage between band 3 and CA, consistent with CA interaction with the cytosolic pole of band 3.This work was supported in part by a grant-in-aid from the American Heart Association, by the Squibb Institute for Medical Research and by The Council for Tobacco Research.

Osmotic properties of human red cells

SummaryWhen an osmotic pressure gradient is applied to human red cells, the volume changes anomalously, as if there were a significant fraction of “nonosmotic water” which could not serve as solvent for the cell solutes, a finding which has been discussed widely in the literature. In 1968, Gary-Bobo and Solomon (J. Gen. Physiol.52:825) concluded that the anomalies could not be entirely explained by the colligative properties of hemoglobin (Hb) and proposed that there was an additional concentration dependence of the Hb charge (zHb). A number of investigators, particularly Freedman and Hoffman (1979,J. Gen. Physiol.74:157) have been unable to confirm Gary-Bobo and Solomons experimental evidence for this concentration dependence of zHb and we now report that we are also unable to repeat the earlier experiments. Nonetheless, there still remains a significant anomaly which amounts to 12.5±0.8% of the total isosmotic cell water (P≪0.0005,t test), even after taking account of the concentration dependence of the Hb osmotic coefficient and all the other known physical chemical constraints, ideal and nonideal. It is suggested that the anomalies at high Hb concentration in shrunken cells may arise from the ionic strength dependence of the Hb osmotic coefficient. In swollen red cells at low ionic strength, solute binding to membrane and intracellular proteins is increased and it is suggested that this factor may account, in part, for the anomalous behavior of these cells.

Transport parameters in the human red cell membrane: solute-membrane interactions of hydrophilic alcohols and their effect on permeation☆

A systematic study has been made of the three coefficients that describe the human red cell membrane transport of a series of short straight-chain hydrophilic alcohols: the permeability coefficient, omega i, the reflection coefficient, sigma i, and the hydraulic conductivity, Lp. Ethylene glycol transport is saturable with Km = 220 +/- 50 mM; there is a second, low-affinity, ethylene glycol site which inhibits water transport (K = 570 +/- 140 mM, max. inhib. = 90 +/- 10%). sigma eth gly = 0.71 +/- 0.04 which is significantly less than 1 (n = 6, P less than 0.001), as are sigma i for six other alcohols (n = 23), thus providing strong thermodynamic evidence that water and these alcohols cross the red cell membrane, at least in part, in an aqueous channel. The solute/membrane frictional coefficient, fsm, for all seven alcohols has been determined and found to decrease monotonically as membrane permeability increases. The red cell membrane has been perturbed by treatments with phenylglyoxal and BS3 (bis(succinimidyl suberate]; these treatments are accompanied by correlated modulation of both ethylene glycol and urea permeability. In one set of experiments in control cells, urea permeability is correlated with water permeability; and, in another set, ethylene glycol permeability is correlated with water permeability. All of these observations support the proposition that the urea class of solutes, the ethylene glycol class of solutes and water all cross the membrane through the same aqueous pore. A schematic model of the red cell pore, consistent with the experimental observations, is presented.

PERMEABILITY AND REFLECTION COEFFICIENTS OF UREA AND SMALL AMIDES IN THE HUMAN RED CELL

Abstract. Measurement of the transport parameters that govern the passage of urea and amides across the red cell membrane leads to important questions about transport of water. It had initially been thought that small protein channels, permeable to water and small solutes, traversed the membrane (see Solomon, 1987). Recently, however, very strong evidence has been presented that the 28 kDa protein, CHIP28, found in the red cell membrane, is the locus of the water channel (see Agre et al., 1993). CHIP28 transports water very rapidly but does not transport small nonelectrolytes such as urea.The irreversible thermodynamic parameter, σi, the reflection coefficient, is a measure of the relationship between the permeability of the solute and that of water. If a solute permeates by dissolution in the membrane, σi= 1.0; if it permeates by passage through an aqueous channel, σi < 1.0. For urea, Goldstein and Solomon (1960) found that σurea= 0.62 ± 0.03 which meant that urea crosses the red cell membrane in a water-filled channel. This result and many subsequent observations that showed that σurea < 1.0 are at variance with the observation that CHIP28 is impermeable to urea.In view of this problem, we have made a new series of measurements of σi for urea and other small solutes by a different method, which obviates many of the criticisms Macey and Karan (1993) have made of our earlier method. The new method (Chen et al., 1988), which relies upon fluorescence of the intracellular dye, fluorescein sulfonate, leads to the corrected value, σurea,corr= 0.64 ± 0.03 for ghosts, in good agreement with earlier data for red cells. Thus, the conclusion on irreversible thermodynamic and other grounds that urea and water share a common channel is in disagreement with the view that CHIP28 provides the sole channel for water entrance into the cell.

Modulation of water and urea transport in human red cells: Effects of pH and phloretin

SummaryIt has previously been shown by Macey and Farmer (Biochim. Biophys. Acta211:104–106, 1970) that phloretin inhibits urea transport across the human red cell membrane yet has no effect on water transport. Jennings and Solomon (J. Gen. Physiol.67:381–397, 1976) have shown that there are separate lipid and protein binding sites for phloretin on the red cell membrane. We have now found that urea transport is inhibited by phloretin binding to the lipids with aK1 of 25±8 μm in reason-able agreement with theKD of 54±5 μm for lipid binding. These experiments show that lipid/protein interactions can alter the conformational state of the urea transport protein. Phloretin binding to the protein site also modulates red cell urea transport, but the modulation is opposed by the specific stilbene anion transport inhibitor, DIDS (4,4′-diisothiocyano-2,2′-stilbene disulfonate), suggesting a linkage between the urea transport protein and band 3. Neither the lipid nor the protein phloretin binding site has any significant effect on water transport. Water transport is, however, inhibited by up to 30% in a pH-dependent manner by DIDS binding, which suggests that the DIDS/band 3 complex can modulate water transport.

Interrelation of ethylene glycol, urea and water transport in the red cell

The reflection coefficient, sigma j, which measures the coupling between the jth solute and water transport across a semipermeable membrane, varies between 0 and 1.0. Values of sigma j significantly less than 1.0 provide irreversible thermodynamic proof that there is coupling between the transport of solute and solvent and thus that they share a common pathway. We have developed an improved method for measuring sigma and have used it to determine that sigma ethylene glycol = 0.71 +/- 0.03 and sigma urea = 0.65 +/- 0.03, in agreement with many, but not all, previous determinations. Since both of these values are significantly lower than 1.0, they show that there is a common ethylene glycol/water pathway and a common urea/water pathway. Addition of first one and then two methyl groups to urea increases sigma to 0.89 +/- 0.04 for methylurea and 0.98 +/- 0.4 for 1,3-dimethylurea, consistent with passage through an aqueous pore with a sharp cutoff in the 6-7 A region.

Site of red cell cation leak induced by mercurial sulfhydryl reagents

It has been suggested that the human red cell anion transport protein, band 3, is the site not only of the cation leak induced in human red cells by treatment with the sulfhydryl reagent pCMBS (p-chloromercuribenzene sulfonate) but is also the site for the inhibition of water flux induced by the same reagent. Our experiments indicate that N-ethylmaleimide, a sulfhydryl reagent that does not inhibit water transport, also does not induce a cation leak. We have found that the profile of inhibition of water transport by mercurial sulfhydryl reagents is closely mirrored by the effect of these same reagents on the induction of the cation leak. In order to determine whether these effects are caused by band 3 we have reconstituted phosphatidylcholine vesicles containing only purified band 3. Control experiments indicate that these band 3 vesicles do not contain (Na+ + K+)-ATPase as measured by ATP dephosphorylation. pCMBS treatment caused a significant increase in the cation leak in this preparation, consistent with the view that the pCMBS-induced cation leak in whole red cells is mediated by band 3.

Modulation of water transport in human red cells: effect of urea

We have studied the effect of urea on water flux in the human red cell and have found that 500 mosmolal urea inhibits osmotic water transport by 39%. The Ki for urea inhibition of water flux is 550 +/- 80 mosmolal, higher than, but comparable with, the Km of urea transport into the red cell of 220-330 mM given by Mayrand and Levitt (J. Gen. Physiol. 55 (1983) 427) and Brahm (J. Gen. Physiol. 82 (1983) 1). Other amides, such as propionamide and valeramide, as well as methyl-substituted ureas, have similar effects, although an indifferent molecule, such as 0.5 M creatinine, has no effect. Urea can be washed off the inhibition site with buffer, and the effects of urea concentrations as high as 1.2 osmolal are entirely reversible. 500 mosmolal urea also significantly increases the reflection coefficient for ethylene glycol, sigma eth gly, from 0.71 +/- 0.03 in control experiments to 0.86 +/- 0.04 (P less than 0.0005, t-test), and propionamide has a similar effect on sigma eth gly. These results show that urea can modulate ethylene glycol transport through the red cell membrane, and are consistent with, but not proof of, the presence of a single class of aqueous channels through which both ethylene glycol and urea enter the red cell. It is suggested that the physiological purpose of these low-affinity urea sites is to modulate water flow out of the red cell during passage through the regions of 0.5-0.6 M urea in the kidney.

Is an intact cytoskeleton required for red cell urea and water transport

In order to determine the membrane protein(s) responsible for urea and water transport across the human red cell membrane, we planned to reconstitute purified membrane proteins into phosphatidylcholine vesicles. In preparatory experiments, we reconstituted a mixture of all of the red cell integral membrane proteins into phosphatidylcholine vesicles, but found that p-chloromercuribenzenesulfonate (pCMBS), which normally inhibits osmotic water permeability by approximately 90%, has no effect on this preparation. The preparation was also unable to transport urea at the high rates found in red cells, though glucose transport was normal. White ghosts, washed free of hemoglobin and resealed, also did not preserve normal urea and pCMBS-inhibitable water transport. One-step ghosts, prepared in Hepes buffer in a single-step procedure, without washing, retained normal urea and pCMBS-inhibitable water transport. Perturbations of the cytoskeleton in one-step ghosts, by removal of tropomyosin, or by severing the ankyrin link which binds band 3 to spectrin, caused the loss of urea and pCMBS-inhibitable water transport. These experiments suggest that an unperturbed cytoskeleton may be required for normal urea and pCMBS-inhibitable water transport. They also show that the pCMBS inhibition of water transport is dissociable from the water transport process and suggest a linkage between the pCMBS water transport inhibition site and the urea transport protein.