A transport metabolon. Functional interaction of carbonic anhydrase II and chloride/bicarbonate exchangers.

The cytoplasmic carboxyl-terminal domain of AE1, the plasma membrane chloride/bicarbonate exchanger of erythrocytes, contains a binding site for carbonic anhydrase II (CAII). To examine the physiological role of the AE1/CAII interaction, anion exchange activity of transfected HEK293 cells was monitored by following the changes in intracellular pH associated with AE1-mediated bicarbonate transport. AE1-mediated chloride/bicarbonate exchange was reduced 50-60% by inhibition of endogenous carbonic anhydrase with acetazolamide, which indicates that CAII activity is required for full anion transport activity. AE1 mutants, unable to bind CAII, had significantly lower transport activity than wild-type AE1 (10% of wild-type activity), suggesting that a direct interaction was required. To determine the effect of displacement of endogenous wild-type CAII from its binding site on AE1, AE1-transfected HEK293 cells were co-transfected with cDNA for a functionally inactive CAII mutant, V143Y. AE1 activity was maximally inhibited 61 +/- 4% in the presence of V143Y CAII. A similar effect of V143Y CAII was found for AE2 and AE3cardiac anion exchanger isoforms. We conclude that the binding of CAII to the AE1 carboxyl-terminus potentiates anion transport activity and allows for maximal transport. The interaction of CAII with AE1 forms a transport metabolon, a membrane protein complex involved in regulation of bicarbonate metabolism and transport.

Expression and characterization of the anion transporter homologue YNL275w in Saccharomyces cerevisiae.

A search of the yeast Saccharomyces cerevisiae genome has revealed an open reading frame, YNL275w, which encodes a 576-amino acid protein that shows sequence similarity to the family of mammalian Cl-/HCO3- anion exchangers and Na+/HCO3- cotransporters. This yeast protein also has a very similar hydropathy profile to the mammalian HCO3- transporters, indicating a similar membrane topology and structure. A V5 epitope and His6-tagged version of Ynl275wp was expressed in yeast and was localized to the plasma membrane by subcellular fractionation and immunofluorescence labeling. The protein was purified by nickel affinity chromatography and was found not to be N-glycosylated. The protein's mobility on SDS-PAGE gels was not altered by treatment with N-glycanase F, alpha-mannosidase, or by mutation of each of the five consensus N-glycosylation sites. The protein did not bind to concanavalin A by lectin blotting or lectin affinity chromatography. The expressed protein bound specifically to a stilbene disulfonate inhibitor resin (SITS-Affi-Gel), and this binding could be competed by certain anions (HCO3-, Cl-, NO3-, and I-) but not by others (SO4(2-) and PO4(3-)). These results suggest that the yeast gene YNL275w encodes a nonglycosylated anion transport protein, localized to the plasma membrane.

A membrane metabolon linking carbonic anhydrase with chloride/bicarbonate anion exchangers.

The erythrocyte Cl-/HCO3- anion exchanger (AE1, Band 3) and the enzyme carbonic anhydrase (CA) catalyze interconnected processes involved in bicarbonate metabolism. The high activity form of carbonic anhydrase, CAII binds to an acidic motif located within the carboxyl-terminal tail of anion exchangers via its basic amino-terminal region. CAII is thereby positioned at the cytosolic surface of the membrane, ideally placed to catalyze CO2 hydration and to channel bicarbonate to or from the anion exchanger. This association of a soluble enzyme and a membrane transporter may be an example of a metabolon, a weakly associated complex of sequential metabolic enzymes.

Carbonic anhydrase: in the driver's seat for bicarbonate transport.

Carbonic anhydrases are a widely expressed family of enzymes that catalyze the reversible reaction: CO(2) + H(2)O <=> HCO(3)(-) + H(+). These enzymes therefore both produce HCO(3)(-) for transport across membranes and consume HCO(3)(-) that has been transported across membranes. Thus these enzymes could be expected to have a key role in driving the transport of HCO(3)(-) across cells and epithelial layers. Plasma membrane anion exchange proteins (AE) transport chloride and bicarbonate across most mammalian membranes in a one-for-one exchange reaction and act as a model for our understanding of HCO(3)(-) transport processes. Recently it was shown that AE1, found in erythrocytes and kidney, binds carbonic anhydrase II (CAII) via the cytosolic C-terminal tail of AE1. To examine the physiological consequences of the interaction between CAII and AE1, we characterized Cl(-)/HCO(3)(-) exchange activity in transfected HEK293 cells. Treatment of AE1-transfected cells with acetazolamide, a CAII inhibitor, almost fully inhibited anion exchange activity, indicating that endogenous CAII activity is essential for transport. Further experiments to examine the role of the AE1/CAII interaction will include measurements of the transport activity of AE1 following mutation of the CAII binding site. In a second approach a functionally inactive CA mutant, V143Y, will be co-expressed with AE1 in HEK293 cells. Since over expression of V143Y CAII would displace endogenous wild-type CAII from AE1, a loss of transport activity would be observed if binding to the AE1 C-terminus is required for transport.

Localization of the Cl-/HCO3- anion exchanger binding site to the amino-terminal region of carbonic anhydrase II.

Human carbonic anhydrase II (CAII) possesses a binding site for an acidic motif (D887ADD) within the carboxyl-terminal region (Ct) of the human erythrocyte chloride/bicarbonate anion exchanger, AE1. In this study, the amino acid sequence comprising this AE1 binding site was localized to the first 17 residues of CAII, which form a basic patch on the surface of the protein. Truncation of the amino terminal of CAII by five residues resulted in a 3-fold reduction in the apparent affinity of the interaction with a GST fusion protein of the Ct of AE1 (GST-Ct) measured by a sensitive microtiter plate binding assay. Further amino-terminal truncation of CAII by 17 or 24 residues caused a loss of binding. The homologous isoform CAI does not bind AE1, despite having 60% sequence identity to CAII. One major difference between the two CA isoforms, within the amino-terminal region, is a high content of histidine residues in CAII (His3, -4, -10, -15, -17) not found in CAI. Mutation of pairs of these histidines (and one lysine) in CAII to the analogous residues in CAI (H3P/H4D or K9D/H10K or H15Q/H17S), or combinations of these various double mutants, did not greatly affect binding between GST-Ct and the mutant CAII. However, when all six of the targeted CAII residues were mutated to the corresponding sequence in CAI, binding of GST-Ct was lost. These results indicate that the AE1 binding site is located within the first 17 residues of CAII, and that the interaction is mediated by electrostatic interactions involving histidine and/or lysine residues. Further specificity for the interaction of AE1 and CAII is provided by a conserved leucine residue (L886) in AE1 that, when mutated to alanine, resulted in loss of GST-Ct binding to immobilized CAII. The binding of the basic amino-terminal region of CAII to an acidic Ct in AE1 provides a structural basis for linking bicarbonate transport across the cell membrane to intracellular bicarbonate metabolism.

Processing of N-linked oligosaccharide depends on its location in the anion exchanger, AE1, membrane glycoprotein.

The human erythrocyte anion exchanger (AE)1 (Band 3) contains a single complex N-linked oligosaccharide that is attached to Asn(642) in the fourth extracellular loop of this polytopic membrane protein, while other isoforms (AE2, AE3 and trout AE1) are N-glycosylated on the preceding extracellular loop. Human AE1 expressed in transfected human embryonic kidney (HEK)-293 or COS-7 cells contained a high-mannose oligosaccharide. The lack of oligosaccharide processing was not due to retention of AE1 in the endoplasmic reticulum since biotinylation assays showed that approx. 30% of the protein was expressed at the cell surface. Moving the N-glycosylation site to the preceding extracellular loop in an AE1 glycosylation mutant (N555) resulted in processing of the oligosaccharide and production of a complex form of AE1. A double N-glycosylation mutant (N555/N642) contained both a high-mannose and a complex oligosaccharide chain. The complex form of the N555 mutant could be biotinylated showing that this form of the glycoprotein was at the cell surface. Pulse-chase experiments showed that the N555 mutant was efficiently converted from a high-mannose to a complex oligosaccharide with a half-time of approx. 4 h, which reflected the time course of trafficking of AE1 from the endoplasmic reticulum to the plasma membrane. The turnover of the complex form of the N555 mutant occurred with a half-life of approx. 15 h. The results show that the oligosaccharide attached to the endogenous site in extracellular loop 4 in human AE1 is not processed in HEK-293 or COS-7 cells, while the oligosaccharide attached to the preceding loop is converted into the complex form.

Identification of the carbonic anhydrase II binding site in the Cl(-)/HCO(3)(-) anion exchanger AE1.

The human Cl(-)/HCO(3)(-) anion exchanger (AE1) possesses a binding site within its 33 residue carboxyl-terminal region (Ct) for carbonic anhydrase II (CAII). The amino acid sequence comprising this CAII binding site was determined by peptide competition and by testing the ability of truncation and point mutants of the Ct sequence to bind CAII with a sensitive microtiter plate binding assay. A synthetic peptide consisting of the entire 33 residues of the Ct (residues 879-911) could compete with a GST fusion protein of the Ct (GST-Ct) for binding to immobilized CAII, while a peptide consisting of the last 16 residues (896-911) could not. A series of truncation mutants of the GST-Ct showed that the terminal 21 residues of AE1 were not required for binding CAII. Removal of four additional residues (887-890) from the Ct resulted in loss of CAII binding. Acidic residues in this region (D887ADD) were critical for binding since mutating this sequence in the GST-Ct to DAAA, AAAA, or NANN caused loss of CAII binding. A GST-Ct construct mutated to D887ANE, the homologous sequence in AE2, could bind CAII. AE2 is a widely expressed anion exchanger and has a homologous Ct region with 60% sequence identity to AE1. A GST fusion protein of the 33 residue Ct of AE2 could bind to CAII similarly to the Ct of AE1. Tethering of CAII to an acidic motif within the Ct of anion exchangers may be a general mechanism for promoting bicarbonate transport across cell membranes.

Resolution of glycoproteins by a lectin gel-shift assay.

Gel-shift assays previously described in the literature are based on protein-protein or protein-DNA interactions. We show that carbohydrate-lectin interactions can be successfully used to alter the electrophoretic mobility of glycosylated, but not nonglycosylated, protein species in SDS-polyacrylamide gels. We were able to separate the two closely migrating mono- (95 kDa) and nonglycosylated (92 kDa) forms of a polytopic membrane protein, anion exchanger 1 (AE1), synthesized by cell-free translation or in transfected HEK293 cells. Concanavalin A was selected as the lectin due to the high mannose content of the oligosaccharide chain on AE1. Concanavalin A was either added to the samples prior to loading or copolymerized in a top layer of the separating gel, the latter being the method of choice. The presence of concanavalin A resulted in slower mobility of the monoglycosylated protein while the mobility of the nonglycosylated form was not altered. The shift in mobility was dependent on concentration of concanavalin A and the length of separating gel containing copolymerized concanavalin A. When a diglycosylated mutant of AE1 was tested, good separation was achieved at lower concentrations of concanavalin A. This lectin gel-shift assay allows the separation of N-glycosylated and nonglycosylated forms of the protein.

Trafficking and folding defects in hereditary spherocytosis mutants of the human red cell anion exchanger.

Hereditary spherocytosis (HS) is a common inherited hemolytic anemia caused by mutations in erythrocyte proteins including the anion exchanger, AE1 (band 3). This study examined seven missense mutations (L707P, R760Q, R760W, R808C, H834P, T837M, and R870W) located in the membrane domain of the human AE1 that are associated with this disease. The HS mutants, constructed in full-length AE1 cDNA, could be transiently expressed to similar levels in HEK 293 cells. Immunofluorescence, cell surface biotinylation, and pulse chase labeling showed that the HS mutants all exhibited defective cellular trafficking from the endoplasmic reticulum to the plasma membrane. Impaired binding to an inhibitor affinity matrix indicated that the mutant proteins had non-native structures and may be misfolded. Further characterization of the HS R760Q mutant showed no change in its oligomeric structure or turnover (half-life = 15 h) compared to wild-type AE1, suggesting the mutant was not aggregated or targeted for rapid degradation via the proteasome. Intracellular retention of HS mutant AE1 would lead to destruction of the protein during erythroid development and would account for the lack of HS mutant AE1 in the plasma membrane of the mature red cell.

Calnexin interaction with N-glycosylation mutants of a polytopic membrane glycoprotein, the human erythrocyte anion exchanger 1 (band 3).

The interaction of the endoplasmic reticulum chaperone calnexin with N-glycosylation mutants of a polytopic membrane glycoprotein, the human erythrocyte anion exchanger (AE1), was characterized by cell-free translation and in transfected HEK293 cells, followed by co-immunoprecipitation using anti-calnexin antibody. AE1 contains 12-14 transmembrane segments and has a single site of N-glycosylation at Asn-642 in the fourth extracytosolic loop. This site was mutated (N642D) to create a nonglycosylated protein. Calnexin showed a preferential interaction with N-glycosylated AE1 relative to nonglycosylated AE1 both in vitro and in vivo. This interaction could be blocked by inhibition of glucosidases I and II with castanospermine. Calnexin had access to novel N-glycosylated sites created in other extracytosolic loops in AE1 by site-directed or insertional mutagenesis. The interaction with AE1 was enhanced when multiple sites were introduced into the same loop or into two different loops. An association of calnexin with truncated versions of N-glycosylated AE1 was detected after release of the nascent chains from ribosomes with puromycin. The results show that the interaction of calnexin with the polytopic membrane glycoprotein AE1 was dependent on the presence but not the location of the oligosaccharide. Furthermore, calnexin was associated with AE1 after release of AE1 from the translocation machinery.

Transmembrane folding of the human erythrocyte anion exchanger (AE1, Band 3) determined by scanning and insertional N-glycosylation mutagenesis.

The human erythrocyte anion exchanger (AE1, Band 3) contains up to 14 transmembrane segments, with a single site of N-glycosylation at Asn642 in extracellular (EC) loop 4. Scanning and insertional N-glycosylation mutagenesis were used to determine the folding pattern of AE1 in the membrane. Full-length AE1, when expressed in transfected human embryonic kidney (HEK)-293 or COS-7 cells, retained a high-mannose oligosaccharide structure. Scanning N-glycosylation mutagenesis of EC loop 4 showed that N-glycosylation acceptor sites (Asn-Xaa-Ser/Thr) spaced 12 residues from the ends of adjacent transmembrane segments could be N-glycosylated. An acceptor site introduced at position 743 in intracellular (IC) loop 5 that could be N-glycosylated in a cell-free translation system was not N-glycosylated in transfected cells. Mutations designed to disrupt the folding of this loop enhanced the level of N-glycosylation at Asn743 in vitro. The results suggest that this loop might be transiently exposed to the lumen of the endoplasmic reticulum during biosynthesis but normally folds rapidly, precluding N-glycosylation. EC loop 4 insertions into positions 428, 484, 754 and 854 in EC loops 1, 2, 6 and 7 respectively were efficiently N-glycosylated, showing that these regions were extracellular. EC loop 4 insertions into positions 731 or 785 were poorly N-glycosylated, which was inconsistent with an extracellular disposition for these regions of AE1. Insertion of EC loop 4 into positions 599 and 820 in IC loops 3 and 6 respectively were not N-glycosylated in cells, which was consistent with a cytosolic disposition for these loops. Inhibitor-affinity chromatography with 4-acetamido-4'-isothiocyanostilbene-2,2'-disulphonate (SITS)-Affi-Gel was used to assess whether the AE1 mutants were in a native state. Mutants with insertions at positions 428, 484, 599, 731 and 785 showed impaired inhibitor binding, whereas insertions at positions 754, 820 and 854 retained binding. The results indicate that the folding of the C-terminal region of AE1 is more complex than originally proposed and that this region of the transporter might have a dynamic aspect.

Identification of the synthetic surfactant nonylphenol ethoxylate: a P-glycoprotein substrate in human urine.

P-glycoprotein (Mdr1p) is an ATP-dependent drug efflux pump that is overexpressed in multidrug-resistant cells and some cancers. Mdr1p is also expressed in normal tissues like the kidney, where it can mediate transepithelial drug transport. A human urinary compound that reverses multidrug resistance and blocks [3H]azidopine photolabeling of P-glycoprotein was purified to homogeneity and identified by 1H-NMR and mass spectrometry as the synthetic surfactant nonylphenol ethoxylate (NPE). Multidrug-resistant Chinese hamster ovary (CHO) C5 cells accumulated less [3H]NPE than parental drug-sensitive Aux-B1 cells, and Mdr1p substrates, verapamil and cyclosporin A, increased this surfactant's accumulation in C5 cells. NPE blocked the net transepithelial transport (basolateral to apical) of [3H]cyclosporin A in epithelia formed by Madin-Darby canine kidney (MDCK) cells. Net transepithelial transport (basal to apical) of [3H]NPE was demonstrated in MDCK cells and was inhibited by cyclosporin A. These findings show NPE is a Mdr1p substrate excreted into urine by kidney P-glycoprotein. NPE is a widely used surfactant and a known hormone disrupter that is readily absorbed orally or topically. The current findings indicate the function of kidney Mdr1p may be to eliminate exogenous compounds from the body.

Two glycoprotein populations of band 3 dimers are present in human erythrocytes.

The human erythrocyte Band 3 anion exchanger contains a single site of N-glycosylation that contains either a short complex oligosaccharide or an extended polylactosaminyl oligosaccharide. Approximately equal amounts of the different glycosylated forms of Band 3 are found in human red cells. As Band 3 exists predominantly as dimers, they may be uniform and consist of a subunit containing an extended oligosaccharide paired with a subunit containing a short oligosaccharide chain. Alternatively, Band 3 dimers may be comprised of subunits that either contain polylactosaminyl or short oligosaccharide chains. To distinguish between these two extremes, the ability of Band 3 membrane domain dimers to bind to immobilized tomato lectin, which specifically binds polylactosaminyl oligosaccharide, was tested. The dimeric membrane domain of Band 3 could be resolved into two fractions by tomato lectin chromatography. This shows that Band 3 dimers are not homogeneous and that two pools exist in red cells, some with a long polylactosaminyl oligosaccharide and the other with a short complex type. The amount of short chain form recovered in the unbound fraction was higher than expected for a random distribution of oligosaccharide chains on Band 3 dimers. Detergent extraction experiments showed that Band 3 glycoforms did not display a differential interaction with the cytoskeleton. The ability to separate Band 3 dimers into two glycoform populations suggests that subunit exchange between dimers does not occur in the membrane or in detergent solution. Furthermore, the results show that while one population of Band 3 dimers is processed to contain polylactosaminyl oligosaccharide, the other largely escapes this processing step.

Carbonic anhydrase II binds to the carboxyl terminus of human band 3, the erythrocyte C1-/HCO3- exchanger.

In this study, we provide evidence that the 33-residue carboxyl-terminal (Ct) region of the human erythrocyte chloride/bicarbonate exchanger, band 3, binds carbonic anhydrase II (CAII). Immunofluorescence showed that tomato lectin-mediated clustering of band 3 in ghost membranes caused a similar clustering of CAII, indicating an in situ association. CAII cosolubilized and coimmunoprecipitated with band 3, suggesting that the two proteins form a complex. Band 3 (K1/2 = 70 nM) or the membrane domain of band 3 (K1/2 = 100 nM) bound saturably to immobilized CAII in a solid phase binding assay. The interaction with CAII was specifically blocked by an antibody to the Ct of band 3. Affinity blotting showed that a glutathione S-transferase (GST)-fusion protein (GST-Ct) containing the last 33 residues of human band 3 bound to CAII. The solid phase binding assay showed that binding of GST-Ct to immobilized CAII was saturable (K1/2 = 20 nM). The binding rate was slow (t1/2 = 12 h) at physiological ionic strength and pH but was enhanced at low ionic strength or acidic pH. Intact band 3 (Ki = 15 nM), the membrane domain of band 3 (Ki = 100 nM), or antibodies to the Ct of band 3 were able to block GST-Ct binding to CAII, confirming the specificity of the interaction. Affinity chromatography showed that CAII bound to immobilized GST-Ct with a 1:1 stoichiometry. This work indicates that CAII, the bicarbonate supplier, is directly coupled to band 3, the chloride/bicarbonate exchanger in red blood cells.

Anion exchangers in the red cell and beyond.

The anion exchanger genes (AE1-3) encode a family of transport proteins responsible for the electroneutral exchange of bicarbonate and chloride across membranes. These transporters are important in processes such as pH regulation and bicarbonate metabolism. This article reviews recent progress in this field based on presentations made at a satellite workshop on anion exchangers held in conjunction with the 8th Fisher Winternational Symposium on Cellular and Molecular Biology entitled "Membrane proteins in health and disease." The transmembrane topology of AE1 has been refined using various combinations of protein chemistry and site-directed mutagenesis. The use of specific inhibitors and novel expression systems continues to reveal fundamental features of the anion exchanger mechanism and its regulation. The importance of anion exchangers in blood and kidney diseases is underscored by the identification and characterization of a plethora of novel mutations in the AE1 gene. Investigations of anion exchanger structure and function have moved beyond studies of the red cell protein into the larger arenas of cellular and molecular biology.

Structural and functional consequences of an N-glycosylation mutation (HEMPAS) affecting human erythrocyte membrane glycoproteins.

Band 3, the human erythrocyte anion exchanger (AE1), and the glucose transporter (GLUT1) proteins each contain a single site of N-glycosylation that is heterogeneously glycosylated. Lectin binding and enzymatic deglycosylation assays showed that the polylactosaminyl oligosaccharide structure of these glycoproteins was altered to a high mannose or hybrid glycan form in three patients with hereditary erythroblastic multinuclearity, with a positive acidified-serum lysis test (HEMPAS). Offspring from one of the HEMPAS patients had intermediate levels of polylactosaminyl oligosaccharide associated with AE1 and GLUT1, suggesting they may have been heterozygous for the genetic defect. The array of polylactosaminyl-containing glycoproteins present in EBV-transformed lymphoblasts derived from fresh blood of HEMPAS patients was similar to control lymphoblasts. HEMPAS lymphoblasts do not therefore express the defect in polylactosamine synthesis found in erythroid cells, indicating that lymphoid cells are not deficient in the processing enzymes or contain an alternative oligosaccharide processing pathway. Purified HEMPAS band 3 had an unaltered oligomeric structure but dimers aggregated more rapidly in detergent solution than normal band 3. The altered oligosaccharide structure did not affect the sensitivity of band 3 to proteolytic digestion in intact red cells but a greater amount of HEMPAS band 3 was associated with the cytoskeleton. The transport activities of AE1 and GLUT1 in HEMPAS erythrocytes were similar to those in normal controls. This shows that the HEMPAS glycosylation defect does not impair the functional accumulation of these two important erythrocyte membrane transporters even though it produces subtle structural changes in band 3 that result in its increased cytoskeletal interaction and self association in detergent solution.

Mapping the ends of transmembrane segments in a polytopic membrane protein. Scanning N-glycosylation mutagenesis of extracytosolic loops in the anion exchanger, band 3.

Band 3, the anion exchanger of human erythrocytes, contains up to 14 transmembrane (TM) segments and has a single endogenous site of N-glycosylation at Asn642 in extracellular (EC) loop 4. The requirements for N-glycosylation of EC loops and the topology of this polytopic membrane protein were determined by scanning N-glycosylation mutagenesis and cell-free translation in a reticulocyte lysate supplemented with microsomal membranes. The endogenous and novel acceptor sites located near the middle of the 35 residue EC loop 4 were efficiently N-glycosylated; however, no N-glycosylation occurred at sites located within sharply defined regions close to the adjacent TM segments. Acceptor sites located in the center of EC loop 3, which contains 25 residues, were poorly N-glycosylated. Expansion of this loop with a 4-residue insert containing an acceptor site increased N-glycosylation. Acceptor sites located in short (<10 residues) loops (putative EC loops 1, 2, 6, and 7) were not N-glycosylated; however, insertion of EC loop 4 into EC loops 1, 2, or 7, but not 6, resulted in efficient N-glycosylation. Acceptor sites in putative intracellular (IC) loop 5 exhibited a similar pattern of N-glycosylation as EC loop 4, indicating a lumenal disposition during biosynthesis. To be efficiently N-glycosylated, EC loops in polytopic membrane proteins must be larger than 25 residues in size, with acceptor sites located greater than 12 residues away from the preceding TM segment and greater than 14 residues away from the following TM segment. Application of this requirement allowed a significant refinement of the topology of Band 3 including a more accurate mapping of the ends of TM segments. The strict distance dependence for N-glycosylation of loops suggests that TM segments in polytopic membrane proteins are held quite precisely within the translocation machinery during the N-glycosylation process.

Self-association of Band 3, the human erythrocyte anion exchanger, in detergent solution.

Dimeric Band 3 purified in n-dodecyl octaethyleneglycol (C12E8) underwent an irreversible, temperature-dependent association, resulting in a complex with a Stokes radius slightly larger than a native tetramer, before forming a higher molecular weight aggregate. Self-association occurred with a half-time of about 1 h at 37 degrees C but did not occur at 0 degrees C after several days. No change in the secondary structure of Band 3, as observed by circular dichroism, occurred during the association process. However, self-association of Band 3 was accompanied by loss of the stilbene disulfonate inhibitor binding site. No association or loss of inhibitor binding occurred with the dimeric membrane domain under similar incubation conditions. The membrane domain dimer was also stable over a wide range of pH (5.5-9.5) and buffer conditions, while Band 3 aggregated below pH 6.5. Inhibitors of anion transport, which stabilize the membrane domain, slowed the association. Band 3, depleted of phospholipids by extensive washing of resin-bound protein with detergent or, incubated with excess detergent, was more prone to aggregation. The membrane domain also showed some aggregation when depleted of lipids. Preparations could be stabilized by adding dimyristoylphosphatidylcholine (DMPC) prior to the 37 degrees C incubation. The effect of inhibitors and DMPC was additive, with a combination of 1 mM 4,4'-dinitrostilbene-2,2'-disulfonate (DNDS) and 1:1 (wt/wt) DMPC:Band 3 stabilizing 90% of the protein to a 24-h incubation at 37 degrees C. The results suggest that self-association of Band 3 dimers is promoted by the cytoplasmic domain but results in alterations to the membrane domain involving the loss of essential phospholipids. Addition of phospholipid or inhibitors to Band 3 results in a stable preparation of the intact protein that may be suitable for crystallization studies.

Random mutagenesis reveals a novel site involved in inhibitor interaction within the fourth transmembrane segment of the Na+/H+ exchanger-1.

We constructed and expressed human Na+/H+ exchanger (NHE-1 isoform) cDNAs randomly mutagenized within the sequence encoding the transmembrane region of the exchanger. Using acute intracellular acidifications in the presence of the NHE-1 inhibitor amiloride (300 microM), we selected a clone expressing a NHE-1 protein exhibiting a 3.3-fold increase in K(i) for amiloride (10 microM instead of 3 microM). Sequencing its cDNA revealed one point mutation resulting in a Gly174Ser substitution near the carboxy-terminal end of the putative fourth transmembrane domain of NHE-1. The introduction of this mutation into the wild-type NHE-1 cDNA and its expression reproduced the features of the mutant. Site-directed Gly174Ala and Gly174Asp substitutions resulted, respectively, in no change and in an approximately 4-fold decrease in the amiloride affinity. An additional mutation (Leu163Phe) in transmembrane segment four has previously been shown to result in a decreased sensitivity to amiloride and its derivatives. The Leu163Phe/Gly174Ser double mutant possesses a strongly reduced affinity for various inhibitors (17 microM for amiloride, 2 microM for MPA, and 20 microM for HOE694) and also a decreased affinity (28 mM instead of 14 mM) for sodium. Although distant in the transmembrane segment, Leu163 and Gly174 residues are both not hydrogen-bonded, being one helix turn from proline residues, and are therefore located in highly flexible regions of the protein. This flexibility and the availability of free carbonyls may play an important role in the interaction with the inhibitors and transported cations.