Solute Carrier Family 4 (Anion Exchanger), Member 1

Description

Band 3 is the major glycoprotein of the erythrocyte membrane. It mediates exchange of chloride and bicarbonate across the phospholipid bilayer and plays a central role in respiration of carbon dioxide. It is a 93,000-Da protein composed of 2 distinct domains that function independently. The 50,000-Da C-terminal polypeptide codes for the transmembrane domain that is involved in anion transport. The 43,000-Da cytoplasmic domain anchors the membrane cytoskeleton to the membrane through an ankyrin-binding site (band 2.1) and also contains binding sites for hemoglobin and several glycolytic enzymes. Proteins related to red cell band 3 have been identified in several types of nucleated somatic cells (review by Palumbo et al., 1986).

Cloning and Expression

Lux et al. (1989) cloned human band 3 from a fetal liver cDNA library. The deduced 911-amino acid protein is similar in structure to other anion exchangers and is divided into 3 regions: a hydrophobic, cytoplasmic domain that interacts with a variety of membrane and cytoplasmic proteins (residues 1-403); a hydrophobic, transmembrane domain that forms the anion antiporter (residues 404-882); and an acidic, C-terminal domain (residues 883-911). Lux et al. (1989) presented a model in which the protein crosses the membrane 14 times.

Gene Function

Langdon and Holman (1988) concluded that band 3 constitutes the major glucose transporter of human erythrocytes. A monoclonal antibody to band 3 specifically removed band 3 and more than 90% of the reconstitutable glucose transport activity from extracts of erythrocyte membranes; nonimmune serum removed neither. Band 3 is probably a multifunctional transport protein responsible for transport of glucose, anions, and water.

Senescent cell antigen (SCA), an aging antigen, is a protein that appears on old cells and marks them for removal by the immune system. The aging antigen is generated by the degradation of protein band 3. Besides its role in the removal of senescent and damaged cells, SCA also appears to be involved in the removal of erythrocytes in hemolytic anemias and the removal of malaria-infected erythrocytes. Band 3 is found in diverse cell types and tissues besides erythrocytes, including hepatocytes, squamous epithelial cells, lung alveolar cells, lymphocytes, kidney, neurons, and fibroblasts. It is also present in nuclear, Golgi, and mitochondrial membranes. Kay et al. (1990) used synthetic peptides to identify antigenic sites on band 3 recognized by the IgG that binds to old cells.

Tanner (1993) discussed the molecular and cellular biology of the erythrocyte anion exchanger, band 3. It permits the high rate of exchange of chloride ion by bicarbonate ion across the red cell membrane: the efflux of bicarbonate from the cell in exchange for plasma chloride ion in the capillaries of the tissues (the Hamburger shift, or chloride ion shift) and the reverse process in lung capillaries. At least 2 nonerythroid anion exchange genes have been characterized, AE2 (109280) and AE3 (106195), and tentative evidence for a fourth member of the class, AE4 (SLC4A9; 610207), was mentioned. The ability of AE2 and AE3 to mediate anion transport has been confirmed. As outlined by Tanner (1993), it is not strictly accurate to refer to the AE1 gene as being that for the erythroid anion exchanger because the AE1 gene is expressed in some nonerythroid tissues, where it appears to be transcribed from different tissue-specific promoters.

Watts et al. (1996) determined that both ZAP70 (176947) and LCK (153390) can phosphorylate the cytoplasmic fragment of BND3. However, these 2 protein tyrosine kinases act on different sites of the BND3 protein.

Pawloski et al. (2001) demonstrated that in human erythrocytes hemoglobin-derived S-nitrosothiol (SNO), generated from imported nitric acid (NO), is associated predominantly with the red blood cell membrane, and principally with cysteine residues in the hemoglobin-binding cytoplasmic domain of the anion exchanger AE1. Interaction with AE1 promotes the deoxygenated structure in SNO-hemoglobin, which subserves NO group transfer to the membrane. Furthermore, Pawloski et al. (2001) showed that vasodilatory activity is released from this membrane precinct by deoxygenation. Thus, the oxygen-regulated cellular mechanism that couples the synthesis and export of hemoglobin-derived NO bioactivity operates, at least in part, through formation of AE1-SNO at the membrane-cytosol interface.

Goel et al. (2003) identified a sialic acid-independent host-parasite interaction involved in the Plasmodium falciparum malaria parasite invasion of red blood cells. They showed that 2 nonglycosylated extracellular regions of band 3 function as a crucial host receptor. They identified 2 processing products of merozoite surface protein-1 (MSP1) as major parasite ligands binding to the band 3 receptor.

Bruce et al. (2004) studied the properties of band 3 in red cells lacking glycophorin A (GPA; 617922) and found that sulfate, iodide, and chloride transport were reduced. Increased flexibility of the membrane domain of band 3 was associated with reduced anion transport activity. Bruce et al. (2004) suggested that band 3 in the red cell can take up 2 different structures: one with high anion transport activity when GPA is present and one with lower anion transport activity when GPA is absent.

By yeast 2-hybrid analysis, affinity copurification, coimmunoprecipitation, and fluorescence-based protein fragment complementation, Nuiplot et al. (2015) confirmed direct interaction between TMEM139 and the kidney isoform of AE1 (kAE1). Knockdown of TMEM139 expression in HEK293T cells reduced membrane localization of kAE1. In contrast, overexpression of TMEM139 increased kAE1 surface expression.

Gene Structure

Schofield et al. (1994) demonstrated that the EPB3 gene extends over 18 kb and consists of 20 exons. The cDNA sequence comprises 4,906 nucleotides, excluding the poly(A) tail. They found extensive similarity between the human and mouse genes, although the latter covers 17 kb. The additional length of the human gene is mainly caused by the presence of 6 Alu repetitive units in the human gene between intron 13 and exon 20. Two potential promoter regions are positioned so that they could give rise to the different transcripts found in erythroid cells and in the kidney. The kidney transcript would lack exons 1 through 3 of the erythroid transcript. The translation initiator downstream to the human kidney promoter would give rise to a protein with a 20-amino acid section at the N-terminus that is not present in the erythroid protein. Sahr et al. (1994) concluded that the AE1 gene spans approximately 20 kb and consists of 20 exons separated by 19 introns. Its structure showed close similarity to that of the mouse AE1 gene. Sahr et al. (1994) described the upstream and internal promoter sequences of the human AE1 gene used in erythroid and kidney cells, respectively.

Biochemical Features

Crystal Structure

Arakawa et al. (2015) reported the crystal structure of the band 3 anion exchanger domain (AE1(CTD)) at 3.5 angstroms. The structure is locked in an outward-facing open conformation by an inhibitor. Comparing this structure with a substrate-bound structure of the uracil transporter UraA in an inward-facing conformation allowed Arakawa et al. (2015) to identify the anion-binding position in the AE1(CTD), and to propose a possible transport mechanism that could explain why selected mutations lead to disease.

Mapping

Showe et al. (1987) localized the gene for BND3 to 17q21-qter by Southern blot analysis of DNA from somatic cell hybrids.

Lux et al. (1989) confirmed assignment of the BND3 gene to chromosome 17.

According to HGM10, EPB3 is in the same large restriction fragment as RNU2 (180690), which narrows the localization to 17q21-q22. Using RFLPs of both loci, Stewart et al. (1989) showed that EPB3 is closely linked to NGFR (162010) (maximum lod = 11.40 at theta = 0.00, with a confidence limit of 0.00 to 0.04).

Gross (2018) mapped the SLC4A1 gene to chromosome 17q21.31 based on an alignment of the SLC4A1 sequence (GenBank BC096106) with the genomic sequence (GRCh38).

Molecular Genetics

Mueller and Morrison (1977) and Hsu and Morrison (1985) reported variant forms of band 3 with an elongated N terminus. Both variants are hematologically normal with normal red cell morphologic features; the red cells do not appear to be resistant to invasion by malaria parasites in vitro (Ranney et al., 1990; Schulman et al., 1990).

Palatnik et al. (1990) described 3 phenotypes based on the polymorphism of band-3 protein from human red cells. Limited proteolysis of intact red cells from most individuals (homozygotes) yields a peptide of 60 kD, but in some persons (heterozygotes), there is also a 63-kD peptide, and rarely only the single peptide of 63 kD is found. This was the first description of the 63-kD homozygote. The frequency of the p63 allele was estimated to be 0.041 +/- 0.0068 in Caucasoids and 0.125 +/- 0.0121 in Negroids.

Acanthocytosis

Kay et al. (1987, 1988) reported 2 sibs with acanthocytosis whose red cells showed markedly increased anion transport activity. The sibs were clinically normal, the abnormality having been detected through the acanthocytosis found on blood studies for unrelated reasons. Kay et al. (1987, 1988) concluded that the 'disorder' was recessive. Bruce et al. (1993) studied the red cells of one of the sibs reported by Kay et al. (1988) and identified band 3 HT (109270.0032).

Southeast Asian Ovalocytosis

Following up on the demonstration by Liu et al. (1990) that a structurally and functionally abnormal band 3 protein shows absolute linkage with the Southeast Asian Ovalocytosis (SAO; 166900) phenotype, Jarolim et al. (1991) demonstrated that the EPB3 gene in these cases contains a 27-bp deletion, resulting in deletion of 9 amino acids (codons 400-408) in the boundary of cytoplasmic and membrane domains of the band 3 protein (109270.0002). The defect was detected in all 30 ovalocytic subjects from Malaysia, the Philippines, and 2 unrelated coastal regions of Papua New Guinea, whereas it was absent in all 30 controls from Southeast Asia and 20 subjects of different ethnic origin from the United States. The lys56-to-glu mutation (109270.0001) was also found in all SAO subjects; however, it was detected in 5 of 50 control subjects as well, suggesting that it represents a linked polymorphism.

Kidson et al. (1981) found that ovalocytic erythrocytes from Melanesians are resistant to invasion by malaria parasites, thus providing a plausible explanation for the polymorphism (also see Serjeantson et al., 1977). Baer (1988) suggested that Malaysian elliptocytosis may be a balanced polymorphism, i.e., that individuals homozygous for the elliptocytosis allele, not clearly identifiable by any assay, may be differentially susceptible to mortality, whereas the heterozygote is at an advantage. Hadley et al. (1983) showed that Melanesian elliptocytes are highly resistant to invasion by Plasmodium knowlesi and P. falciparum in vitro. This is the only human red cell variant known to be resistant to both.

Coetzer et al. (1996) described a 4-generation South African kindred with dominantly inherited ovalocytosis and hemolytic anemia. All affected subjects exhibited varying degrees of hemolytic anemia. Additionally, there was evidence for independent segregation of the band 3 Memphis I polymorphism (109270.0001) and the 27-bp deletion in BND3 causing SAO. Six SAO subjects and all 3 normal family members were heterozygous for the band 3 Memphis I polymorphism and one SAO subject was homozygous for this mutation.

Spherocytosis Type 4

In a 28-year-old female with congenital spherocytic hemolytic anemia (SPH4; 612653), Jarolim et al. (1991) identified a missense mutation in the SLC4A1 gene (109270.0003).

In a 33-year-old woman with pregnancy-associated hemolytic anemia and spherocytosis, Rybicki et al. (1993) identified a G40K mutation in SLC4A1 (109270.0004).

In a 3-generation Czech family in which 5 affected members exhibited compensated hemolytic disease, Jarolim et al. (1994) identified a 10-bp duplication in the SLC4A1 gene (109270.0005) that segregated with disease.

In affected members of a large Swiss family with spherocytosis, Maillet et al. (1995) identified heterozygosity for an SLC4A1 G771D mutation (109270.0007).

In an 18-year-old French man with moderate hereditary spherocytosis, Alloisio et al. (1996) identified an R150X mutation in SLC4A1 (109270.0009). The proband's mother, who had the same mutation, had a milder clinical presentation. Further investigation revealed a second, paternally inherited SLC4A1 mutation in the proband (109270.0010).

Dhermy et al. (1997) studied 8 kindreds with dominant hereditary spherocytosis and band 3 deficiency mutations. The amount of band 3 appeared to be slightly, but significantly, more reduced in HS patients with missense mutations and presence of the mutant transcripts than in HS patients with premature termination of translation and absence of mutant transcripts, suggesting that SLC4A1 missense mutations may have a dominant-negative effect.

Alloisio et al. (1997) reported a V488M mutation in band 3 (109270.0022) that was associated with spherocytosis in heterozygous state. Ribeiro et al. (2000) identified the V488M mutation in homozygosity in a female infant with severe anemia and hydrops, in whom renal tubular acidosis was detected by age 3 months.

In a 29-year-old Japanese man with compensated hemolytic anemia and spherocytosis, Inoue et al. (1998) identified homozygosity for an SLC4A1 G130R mutation (109270.0018).

In a 22-year-old Japanese man who presented with cholelithiasis and hemolysis and had a history of jaundice since early childhood, Iwase et al. (1998) identified a T837A mutation in SLC4A1 (band 3 Tokyo; 109270.0019).

Bruce et al. (2005) identified 11 human pedigrees with dominantly inherited hemolytic anemias, 8 in the hereditary stomatocytosis class (see 'Cryohydrocytosis,' below) and 3 in the spherocytosis class. Affected individuals in these families had an increase in membrane permeability to sodium and potassium ion that was particularly marked at zero degree centigrade. They found that disease in these pedigrees was associated with a series of single amino acid substitutions in the intramembrane domain of the band 3 anion exchanger. Anion movements were reduced in the abnormal red cells. The 'leak' cation fluxes were inhibited by chemically diverse inhibitors of band 3. Expression of the mutated genes in Xenopus laevis oocytes induced abnormal NA and K fluxes in the oocytes, and the induced chloride transport was low. These data were considered consistent with the suggestion that the substitutions convert the protein from an anion exchanger into an unregulated cation channel. All affected individuals were heterozygous for missense mutations in exon 17 of the SLC4A1 gene, including 2 families with spherocytosis who carried the R760Q mutation (109270.0028) that had previously been reported in 2 spherocytosis patients by Jarolim et al. (1995).

Cryohydrocytosis

In 8 unrelated families with cryohydrocytosis (CHC; 185020), Bruce et al. (2005) identified 3 different heterozygous missense mutations in the SLC4A1 gene (109270.0033-109270.0035) that segregated fully with disease in each family.

Choreoacanthocytosis

Tanner (1993) reviewed the evidence that mutations in the AE1 gene can cause choreoacanthocytosis (200150; see Kay, 1991). Kay et al. (1989) reported a band 3 alteration in association with anemia as determined by a reticulocyte count of 20%. The erythrocyte defect was reflected in increased IgG binding, increased breakdown products of band 3, and altered anion- and glucose-transport activity in middle-aged cells. IgG eluted from the red cells of the propositus appeared to have a specificity for senescent cell antigen. This and other studies suggested that band 3 was aging prematurely in erythrocytes of the subject, and that the senescent cell antigen appeared on the middle-aged red cells. Two sibs were affected. Both parents were thought to show 'subtle band 3 changes.' Autosomal recessive inheritance was postulated.

Distal Renal Tubular Acidosis, Autosomal Dominant

Bruce et al. (1997) found that all affected members of 4 families with autosomal dominant familial renal tubular acidosis (RTA; 179800) were heterozygous for mutations in the SLC4A1 gene; these mutations were not found in any of the 9 normal family members studied. In 2 families the mutation was arg589 to his (109270.0012); arg589-to-cys (109270.0013) and ser613-to-phe (109270.0014) changes were found in the other families. Linkage studies confirmed the cosegregation of the disease with a genetic marker close to SLC4A1. Affected individuals with the mutations in arg589 had reduced red cell sulfate transport and altered glycosylation of the red cell band 3 N-glycan chain. The red cells of individuals with the ser613-to-phe mutation had markedly increased red cell sulfate transport but almost normal red cell iodide transport. The erythroid and kidney isoforms of the mutant band 3 protein were expressed in Xenopus oocytes and all showed significant chloride transport activity. Bruce et al. (1997) concluded that dominantly inherited RTA is associated with mutations in band 3; however, both the disease and its autosomal dominant inheritance are not related simply to the anion transport activity of the mutant proteins. Arg589 is located in the cytoplasmic loop between transmembrane segments 6 and 7 of band 3. This arginine is conserved in all known vertebrate sequences of AE1, AE2, and AE3, suggesting that it is functionally important. Arg589 is located in a cluster of basic residues which may form part of the cytoplasmic anion binding site of band 3. The mechanism by which the S613F mutation increases the affinity of the protein for sulfate was not clear. One possibility was that the mutation, which is located near the center of membrane span 7 and results in a substitution of serine by a bulky phenylalanine residue, altered the orientation of membrane span 7 relative to span 6. This may distort the conformation of the cytoplasmic loop between spans 6 and 7 which contains the putative anion binding site so that the clustered basic residues bind sulfate more tightly than the wildtype protein. Bruce et al. (1997) were prompted to undertake this study because of a possible association between dominant RTA and hereditary ovalocytosis (Baehner et al., 1968). Mutations in the families with dominant RTA were different from those affecting band 3 in Southeast Asian ovalocytosis. Complete absence of band 3 was found by Inaba et al. (1996) to result in defective renal acid secretion in cattle.

Most of the patients in the 4 families studied by Bruce et al. (1997) presented clinically with renal stones, and the majority had nephrocalcinosis. One patient in a family with the arg589-to-his mutation had rickets when initially seen at age 10 years and developed osteomalacia at the age of 31 after she stopped taking alkali therapy, but no other patient had bone disease. Eight patients were not acidotic when first seen, and were diagnosed as 'incomplete' dominant RTA because they were unable to excrete a urine more acid than pH 5.3 after oral acute ammonium chloride challenge. Compared with acidotic cases, these patients tended to be younger, with lower plasma creatinines, better preservation of urinary concentrating ability, and less (or no) nephrocalcinosis; over a 10-year period, 2 of the patients spontaneously developed acidosis. Acidotic patients were treated with oral alkalis, usually 6 gm of sodium bicarbonate daily, and had normal acid-base status at the time of the study; nonacidotic patients were not treated.

Karet et al. (1998) screened 26 kindreds with primary distal renal tubular acidosis (dRTA; 179800) for mutations in the AE1 gene. Inheritance was autosomal recessive in 17, autosomal dominant in 1, and uncertain due to unknown parental phenotype or sporadic disease in 8. No mutations in AE1 were detected in any of the autosomal recessive kindreds, and analysis of linkage showed no evidence of linkage of recessive distal RTA to AE1. In contrast, heterozygous mutations in AE1 were identified in the 1 known dominant distal RTA kindred, in 1 sporadic case, and in 1 kindred with 2 affected brothers. In the dominant kindred, an arg589-to-ser mutation (109270.0015) cosegregated with distal RTA in the extended pedigree. In the sporadic case, an arg589-to-his mutation (109270.0012) proved to be a de novo change. In the third kindred, both affected brothers had an intragenic 13-bp duplication resulting in deletion of the last 11 amino acids of AE1 (band 3 Walton; 109270.0025). Parental consanguinity was identified in 14 of the 17 recessive pedigrees. In the recessive kindreds, 19 of 25 patients were diagnosed at 1 year of age or less, and the remainder presented at 6 years or younger. All index cases presented either acutely with vomiting and dehydration, or with failure to thrive or delayed growth. Younger affected sibs were often diagnosed prospectively. All patients with the recessive disease were found to have nephrocalcinosis, nephrolithiasis, or both, and several had rickets. Nine of these patients from 6 families also had bilateral sensorineural deafness confirmed by audiometry; see renal tubular acidosis with progressive nerve deafness (267300). In contrast, in the 1 dominant kindred (with the arg589-to-ser mutation), 2 propositae were diagnosed because of nephrolithiasis at ages 56 and 36 years. Prospective screening identified other affected family members who were all asymptomatic, and most were diagnosed in adulthood. None of the 6 affected members of this family had radiologic evidence of nephrocalcinosis.

The chloride-bicarbonate exchanger AE1, which is mutant in autosomal dominant distal renal tubular acidosis, is normally expressed at the basolateral surface of alpha-intercalated cells in the distal nephron. Devonald et al. (2003) demonstrated that AE1 is aberrantly targeted to the apical surface in this disorder, in contrast with many disorders where mutant membrane proteins are retained intracellularly and degraded.

Distal Renal Tubular Acidosis with Hemolytic Anemia

Tanphaichitr et al. (1998) described novel AE1 mutations in a Thai family with a recessive syndrome of dRTA and hemolytic anemia in which red cell anion transport was normal (611590). A brother and sister were triply homozygous for 2 benign mutations, M31T and K56E (109270.0001), and for a loss-of-function mutation, G701D (109270.0016). The AE1 G701D loss-of-function mutation was accompanied by impaired trafficking to the Xenopus oocyte surface. Coexpression of the erythroid AE1 chaperonin, glycophorin A, along with the AE1 G701D mutation, rescued both AE1-mediated chloride ion transport and AE1 surface expression in oocytes. The genetic and functional data suggested that the homozygous AE1 G701D mutation causes recessively transmitted dRTA in this kindred with apparently normal erythroid anion transport.

Bruce et al. (2000) studied 3 Malaysian and 6 Papua New Guinean families with dRTA and Southeast Asian ovalocytosis (SAO). The SAO deletion mutation (109270.0002) occurred in many of the families but did not itself result in distal renal tubular acidosis. Compound heterozygotes of each of the 3 dRTA mutations (G701D, 109270.0016; A858D, 109270.0020; delV850 109270.0021) with SAO all had dRTA, evidence of hemolytic anemia, and abnormal red cell properties. The A858D mutation showed dominant inheritance and the recessive delV850 and G701D mutations showed a pseudodominant phenotype when the transport-inactive SAO allele was also present. Red cell and Xenopus oocyte expression studies showed that the delV850 and A858D mutant proteins had greatly decreased anion transport when present as compound heterozygotes with each other or with SAO. Red cells with A858D/SAO had only 3% of the sulfite ion efflux of normal cells, the lowest anion transport activity reported for human red cells to that time. Bruce et al. (2000) confirmed that the G701D mutant protein has an absolute requirement for glycophorin A for movement to the cell surface.

In a female infant with severe anemia and hydrops, in whom renal tubular acidosis was detected by age 3 months, Ribeiro et al. (2000) identified homozygosity for a V488M mutation (109270.0022), which had previously been reported in association with spherocytosis in heterozygous state by Alloisio et al. (1997).

Sritippayawan et al. (2004) reported 2 Thai families with recessive dRTA due to different compound heterozygous mutations of the SLC4A1 gene. In the first family, the patient with dRTA had compound heterozygous G701D/S773P (109270.0026) mutations. In the second family, the patient and his sister had dRTA and SAO, and were compound heterozygotes for the SAO deletion mutation and an R602H mutation (109270.0027). Sritippayawan et al. (2004) noted that the second patient had a severe form of dRTA whereas his sister had only mild metabolic acidosis, indicating that other modifying factors or genes might play a role in governing the severity of the disease.

Kittanakom et al. (2004) transiently transfected human embryonic kidney HEK293 cells with the renal isoform of SLC4A1 containing the S773P mutation, alone or in combination with wildtype SLC4A1 or with the G701D mutant. The S773P mutant was expressed at a 3-fold lower level than wildtype, had a 2-fold decrease in its half-life, and was targeted for degradation by the proteasome. Both S773P and G701D exhibited defective trafficking to the plasma membrane, providing an explanation for the dysfunction found in dRTA.

Blood Groups

Diego blood group (110500) Di(a) is a low-incidence blood group antigen in Caucasians that is antithetical to Di(b). Prevalence of Di(a) is much higher in American Indians, reaching up to 54% in some groups of South American Indians. Bruce et al. (1994) demonstrated that the Diego blood group polymorphism is the result of a single amino acid substitution at position 854 of the AE1 molecule, with proline of the wildtype band 3 corresponding to the Di(b) antigen and leucine to the Di(a) antigen. Subsequently, Bruce et al. (1995) mapped the low-incidence blood group antigen Wr(a) (109270.0011) to the C-terminal end of the fourth ectoplasmic loop and defined a single amino acid substitution in Wr(b) (109270.0006). Jarolim et al. (1998) studied the molecular basis of 7 low-incidence blood group antigens that likewise are due to variation in AE1.

McManus et al. (2000) demonstrated that the Froese blood group polymorphism (601551) is the result of change in the SLC4A1 gene (109270.0029).

Zelinski et al. (2000) demonstrated that the Swann blood group (601550) is due to molecular changes in the SLC4A1 gene (109270.0030).

Animal Model

Inaba et al. (1996) studied a moderately uncompensated bovine anemia associated with spherocytosis inherited in an autosomal incompletely dominant mode and retarded growth. Using biochemical methods they showed that the bovine red cells lacked the band 3 protein completely. Sequence analysis of EPB3 cDNA and genomic DNA showed a C-to-T transition resulting in a missense mutation: CGA-to-TGA; arg646-to-ter. The location of the mutation was at the position corresponding to codon 646 in human EPB3 cDNA. The animal red cells were deficient in spectrin, ankyrin, actin (see 102630), and protein 4.2 (177070), resulting in a distorted and disrupted membrane skeletal network with decreased density. Therefore, the animal's red cell membranes were extremely unstable and showed the loss of surface area in several distinct ways such as invagination, vesiculation, and extrusion of microvesicles, leading to the formation of spherocytes. Inaba et al. (1996) also found that total deficiency of bovine band 3 also resulted in defective chloride/bicarbonate exchange, causing mild acidosis with decreases in bicarbonate concentration and total CO(2) in the animal's blood. The results demonstrated to the authors that bovine band 3 contributes to red cell membrane stability, CO(2) transport, and acid-base homeostasis, but is not always essential to the survival of this mammal.

Erythroid band 3 (AE1) is one of 3 anion exchanges that are encoded by separate genes. The AE1 gene is transcribed by 2 promoters: the upstream promoter is used for erythroid band 3, whereas the downstream promoter initiates transcription of the band 3 isoform in kidney. To assess the biologic consequences of band 3 deficiency, Southgate et al. (1996) selectively inactivated erythroid but not kidney band 3 by gene targeting in mice. Although no death in utero occurred, most homozygous mice died within 2 weeks after birth. The erythroid band 3-null mice showed retarded growth, spherocytic red blood cell morphology, and severe hemolytic anemia. Remarkably, the band 3 -/- red blood cells assembled normal membrane skeleton, thus challenging the notion that the presence of band 3 is required for stable biogenesis of the membrane skeleton. Similarly, Peters et al. (1996) used targeted mutagenesis in the mouse to assess AE1 function in vivo. RBCs lacking AE1 spontaneously shed membrane vesicles and tubules, leading to severe spherocytosis and hemolysis, but the levels of the major skeleton components, the synthesis of spectrin in mutant erythroblasts, and skeletal architecture were normal or nearly normal. Their results indicated that AE1 does not regulate RBC membrane skeleton assembly in vivo but is essential for membrane stability. Peters et al. (1996) postulated that stabilization is achieved through AE1-lipid interactions and that loss of these interactions is a key pathogenic event in hereditary spherocytosis. Jay (1996) reviewed the role of band 3 in red cell homeostasis and cell shape.

Paw et al. (2003) characterized a zebrafish mutant called retsina (ret) that exhibits an erythroid-specific defect in cell division with marked dyserythropoiesis similar to human congenital dyserythropoietic anemia (see 224100). Erythroblasts from ret fish show binuclearity and undergo apoptosis due to a failure in the completion of chromosome segregation and cytokinesis. Through positional cloning, Paw et al. (2003) demonstrated that the ret mutation is in the Slc4a1 gene, encoding the anion exchanger-1 (also called band 3 and AE1), an erythroid-specific cytoskeletal protein. They further showed an association between deficiency in Slc4a1 and mitotic defects in the mouse. Rescue experiments in ret zebrafish embryos expressing transgenic Slc4a1 with a variety of mutations showed that the requirement for band 3 in normal erythroid mitosis is mediated through its protein 4.1R-binding domains. Paw et al. (2003) concluded that their report established an evolutionarily conserved role for band 3 in erythroid-specific cell division and illustrated the concept of cell-specific adaptation for mitosis.