CHAPTER 27 THE RED CELL MEMBRANE

CHAPTER 27 THE RED CELL MEMBRANE
Williams Hematology

CHAPTER 27 THE RED CELL MEMBRANE

PATRICK G. GALLAGHER
BERNARD G. FORGET

Introduction
Composition of the Erythrocyte Membrane

Membrane Lipids

Membrane Proteins
Functions of the Erythrocyte Membrane

Membrane Assembly and Organization

Cellular Deformability and Membrane Stability

Membrane Material Properties

Membrane Biogenesis and Aging

Membrane Permeability
Chapter References

The easy accessibility of the red cell has allowed the human erythrocyte membrane to become the most thoroughly studied biologic membrane. It is composed of three major structural elements: a lipid bilayer primarily composed of phospholipids and cholesterol; integral proteins embedded in the lipid bilayer that span the membrane; and a membrane skeleton on the internal side of the red cell membrane. The erythrocyte membrane has many important functions. The lipid bilayer provides an impermeable barrier between the cytoplasm and the external environment and helps maintain a slippery exterior so that erythrocytes do not adhere to endothelial cells or aggregate and occlude the micro-circulation. The red cell membrane and its skeleton provide the erythrocyte with its unique deformability, durability, and tensile strength to undergo large deformations during repeated passages through narrow microcirculatory channels. The erythrocyte membrane also assembles and organizes the proteins of the lipid bilayer and the underlying skeleton. This allows the red cell to participate in a wide range of functions. These include influencing cellular metabolism by selectively and reversibly binding and inactivating glycolytic enzymes, retaining organic phosphates and other vital compounds, removing metabolic waste, and sequestering the reductants required to prevent corrosion by oxygen. During erythropoiesis, the membrane imports the iron required for the synthesis of hemoglobin. At the level of the organism, the membrane participates in the maintenance of pH homeostasis, participating in the exchange of chloride and bicarbonate. Investigation of disorders of the erythrocyte membrane such as hereditary spherocytosis, elliptocytosis, and pyropoikilocytosis has advanced our understanding of the normal structure/function relationships of the membrane as well as providing us with an understanding of the inheritance and expression of these disorders.

Acronyms and abbreviations that appear in this chapter include: AE1, anion exchanger-1; AQP1, aquaporin-1; ATP, adenosine 5′-triphosphate; HE, hereditary elliptocytosis; LCAT, lecithin-cholesterol acyltransferase; MAGUK, membrane-associated guanylate kinase; PAS, periodic acid-Schiff; PI, phosphatidylinositol; PIP, PI-4-monophosphate; PIP-2, PI-4,5-biphosphate; SDS, sodium dodecyl sulfate.

INTRODUCTION
The erythrocyte membrane accounts for only 1 percent of total weight of the red cell, yet it plays an integral role in the maintenance of erythrocyte integrity. The red cell membrane and its skeleton provide the erythrocyte the flexibility, durability, and tensile strength to undergo large deformations during repeated passages through narrow microcirculatory channels. The red cell membrane maintains a slippery exterior so that erythrocytes do not adhere to endothelial cells or aggregate and occlude the microcirculation. The membrane plays an important role in metabolism by selectively and reversibly binding and inactivating glycolytic enzymes. The membrane retains organic phosphates and other vital compounds and helps remove metabolic waste. The membrane sequesters the reductants required to prevent corrosion by oxygen. During erythropoiesis, the membrane responds to erythropoietin and imports the iron required for the synthesis of hemoglobin. At the level of the organism, the membrane participates in the maintenance of pH homeostasis, participating in the exchange of chloride and bicarbonate.
The easy accessibility of the human erythrocyte membrane has allowed it to become the most thoroughly studied biologic membrane. Erythrocytes are the cells about which the most detailed information is available concerning the normal structure and function of their membrane and about the molecular pathology of disorders due primarily to abnormal membrane or cytoskeletal structure. The erythrocyte membrane remains the paradigm for ongoing studies of other cell types. Although the primary structure (Fig. 27-1) and a number of the important functions of the red cell membrane are known, its study continues to yield important insights into our understanding of membrane structure and function. Genetic investigation of disorders of the erythrocyte membrane has advanced our understanding of the normal structure/function relationships of the membrane as well as provided us with an understanding of the inheritance and expression of these disorders.

FIGURE 27-1 A schematic diagram illustrating the molecular assembly of the major erythrocyte membrane proteins and a model of the principal molecular defect in hereditary spherocytosis (HS), elliptocytosis (HE), and pyropoikilocytosis (HPP). Membrane protein-protein and protein-lipid associations can be divided into two categories: (1) vertical interactions, which are perpendicular to the plane of the membrane and involve spectrin-ankyrin-band 3 interaction, spectrin-protein 4.1-glycophorin C connection, and weak interactions between spectrin and the negatively charged lipids of the inner half of the membrane lipid bilayer, and (2) horizontal interactions, which are parallel to the plane of the membrane.

COMPOSITION OF THE ERYTHROCYTE MEMBRANE
The erythrocyte membrane is composed of three major structural elements: a lipid bilayer primarily composed of phospholipids and cholesterol that provides a permeability barrier between the external environment and the red cell cytoplasm; integral proteins embedded in the lipid bilayer that span the membrane; and a membrane skeleton on the internal side of the red cell membrane that provides structural integrity to the cell.
MEMBRANE LIPIDS
COMPOSITION
Lipids comprise 50 to 60 percent of red cell membrane mass. The principal membrane lipids are phospholipids and cholesterol, which are present in nearly equal amount.1 Small amounts of glycolipids, primarily globoside, are also present. The primary phospholipids are phosphatidylcholine (28 percent of total phospholipids), phosphatiylethanolamine (27 percent), sphingomyelin (26 percent), phosphatidylserine (13 percent), and phosphatidylinositol.
Membrane phosphoinositides are phospholipids that contain phosphatidylinositol (PI) or its phosphorylated forms, PI-4-monophosphate and PI-4,5-biphosphate (PIP and PIP-2 respectively). In nucleated cells, phosphoinositides are precursors of important intracellular second messengers such as inositol-1,4,5-triphosphate and diacylglycerol that participate in regulation of many cellular processes. In mature erythrocytes, phosphoinositides represent 2 to 5 percent of total phospholipids, residing largely at the inner membrane surface and undergoing rapid phosphorylation and dephosphorylation. In red cells, they are involved in regulation of calcium transport and interaction of transmembrane and skeletal proteins (e.g., glycophorin C and protein 4.1), and they have been proposed to participate in the control of the discocyte-echinocyte shape transformation.2
In the erythrocyte, cholesterol is present in a free, unesterified form, and it is almost entirely hydrophobic. Its primary role appears to be to control membrane fluidity even under conditions that might lead to phospholipid crystallization and rigidification of the bilayer.
MEMBRANE LIPID DISTRIBUTION
Phospholipids are asymmetrically distributed in the red cell membrane with phosphatidylserine and phosphatidylethanolamine primarily in the inner hemileaflet, while sphingomyelin and phosphatidylcholine are outwardly oriented. This asymmetric distribution of phospholipids is a dynamic system involving a constant exchange (“flip-flop”)3 between the phospholipids of the two-bilayer leaflets. Maintenance of this asymmetry appears to be important in the regulation of hemostasis, as PS on the outer leaflet provides a site for prothrombinase binding, causing the red cell surface to become prothrombotic. Phospholipid flipping may contribute to the occurrence of thromboses in a variety of disorders including sickle cell disease (see Chap. 47) and diabetes.4 The presence of PS on the outer surface of the red cell is one of the earliest changes in apoptosis, and it has been correlated with complement activation and red cell clearance by macrophages.
Enzymes called flippases actively translocate PS and PE to the inner leaflet; floppases catalyze translocation to the outer leaflet. Asymmetry seems to depend on the fact that flipping occurs at a higher rate than flopping. Flippase activity is mediated by a 130-kDa integral membrane protein that is a member of the Mg++-dependent, P-glycoprotein ATPases.5 Floppase activity in red cell membranes appears to be mediated by the multidrug resistance protein 1 (MRP1).6,7 A scramblase activated by elevated intracellular calcium that promotes randomization and loss of asymmetry has been isolated and cloned.8,9 This scramblase mediates redistribution of membrane phospholipids in activated, injured, or apoptotic cells and is activated by calcium.10,11 Derangements within the red cell often raise intracellular calcium by direct or indirect damage to ion channels and pumps. Scott’s syndrome is a congenital bleeding disorder in which red cells and platelets expose subnormal amounts of PS on the outer surface in response to calcium, but it does not appear to be due to scramblase deficiency.12,13
Glycolipids and cholesterol are intercalated between the phospholipids in the bilayer with their long axes perpendicular to the bilayer plane. Red cell glycolipids are located entirely in the external half of the bilayer with their carbohydrate moieties extending into the aqueous phase. They carry several important red cell antigens, including A, B, H, and P, and may serve other important functions. The location of membrane cholesterol is less certain, but it appears that cholesterol is present in about equal proportions on both sides of the bilayer.
LIPID SYNTHESIS AND RENEWAL
The synthesis and assembly of red cell membrane lipids takes place during erythropoiesis. Mature erythrocytes are unable to synthesize fatty acids, phospholipids, or cholesterol de novo and depend on lipid exchange and fatty acid acylation for phospholipid repair and renewal. These renewal pathways, although limited, permit a slow replacement of membrane lipid components.
Lipid exchange rates vary considerably. The exchange of unesterified cholesterol takes place in several hours, while the outer bilayer phospholipid phosphatidylcholine and sphingomyelin exchange with the phospholipids of plasma lipoproteins over a period of days. Because of their inaccessibility, the inner bilayer phospholipids phosphatidylserine and phosphatidylethanolamine are unable to participate in lipid exchange. Unesterified membrane cholesterol exchanges readily with the unesterified cholesterol in plasma lipoproteins where it is partially converted to esterified cholesterol by lecithin-cholesterol acyltransferase (LCAT). Because the newly formed cholesteryl ester cannot return to the red cell membrane, LCAT catalyzes a unidirectional pathway that depletes the membrane of cholesterol and decreases its surface area, and there is virtually no esterified cholesterol in the membrane. This process is reversed when this enzyme is absent or inactive, leading to a net accumulation of free cholesterol in the cells.
In addition to passive exchange, free fatty acids can be incorporated into red cell phospholipids in a two-step reaction requiring lysophospholipid, ATP, magnesium, and coenzyme A. Following the acyl-coenzyme A formation, the fatty acid is incorporated into the lysophospholipid at the inner bilayer leaflet. This pathway also participates in the maintenance of phospholipid asymmetry, as evidenced by a rapid outward translocation of the newly synthesized phosphatidylcholine. Although this pathway consumes a small amount of energy, it may be important for detoxification of naturally formed lysophosphatides in the cells, as evidenced by their gradual accumulation during ATP depletion.
LIPID BILAYER FLUIDITY
Under physiologic conditions, the lipid bilayer is in a liquid state, allowing both the transmembrane proteins and the cell surface molecules (such as surface antigens) to move in the plane of the membrane. Lipid bilayer fluidity is influenced by several factors including: (1) temperature, which determines the phase transition between a liquid state and gel state; (2) free cholesterol content, as the rigid sterol ring of cholesterol decreases lipid bilayer fluidity; and (3) the length and the degree of phospholipid fatty acid saturation. Saturated fatty acids with a relatively rigid backbone resist motion, while the unsaturated fatty acids have relatively unrestricted movements, thereby increasing the fluidity of the lipid bilayer. Because of the differences in the composition of phospholipids between the two-bilayer halves, the bilayer is asymmetric in terms of the fluidity of the two hemileaflets.
MEMBRANE PROTEINS
Several general observations can be made about erythrocyte membrane proteins. Most of these proteins also are present in nonerythroid cells, where they fulfill similar functions. Many of these proteins are members of super families of proteins that are structurally related but genetically distinct. This genetic diversity explains why the clinical expression of many (but not all) red cell membrane protein mutations is confined to the erythroid lineage. Tissue- and developmental stage-specific alternative splicing or the usage of alternate initiation codons or alternate promoters creates multiple isoforms of many of these proteins. Finally, many are large, multifunctional proteins. As a result, mutations within a given region of the protein may lead to distinct differences in abnormalities of function and clinical phenotype.
Membrane proteins are classified according to the ease with which they can be removed from whole red cell membrane preparations in the laboratory. Integral proteins are firmly embedded into or through the lipid bilayer by hydrophobic domains within their amino acid sequences; only harsh reagents such as detergents can extract them. Peripheral proteins are more loosely associated; they are extractable by high- or low-salt or high-pH extraction. Peripheral proteins are attached indirectly to the lipid bilayer by means of covalent or non-covalent binding to the (usually) cytoplasmic domains of embedded or anchored proteins and typically are associated with only one face of the membrane (i.e., exterior or extracellular versus interior or cytoplasmic), whereas many integral proteins often protrude into both spaces. The affinity with which proteins associate with the membrane is not a static property. Rather, proteins can become more or less tightly bound according to their state of phosphorylation, methylation, glycosylation, or lipid modification (myristylation, palmitylation, or farnesylation).2
Fairbanks and colleagues assigned names to the proteins extracted from red cell membranes (Fig. 27-1 and Table 27-1).14 These designations were based on their mobility in a sodium dodecyl sulfate (SDS)-acrylamide gel system; the slowest migrating band was band (or protein) 1, the next slowest band, band 2, etc. Subbands were designated with decimals. After further analysis, some of these proteins, such as bands 1 and 2, alpha and beta spectrin, were renamed. Other proteins, such as protein 4.1, were never renamed.

TABLE 27-1 MAJOR RED CELL MEMBRANE PROTEINS

INTEGRAL MEMBRANE PROTEINS
Band 3 Band 3 (anion exchanger-1, AE1) is an abundant (106 copies per cell) transmembrane glycoprotein with a molecular mass of about 100 kDa. It serves as a regulator of ion content, red cell deformability, intermediary metabolism, and red cell senescence.15,16 The NH2-terminus of the protein encodes a 43-kDa cytoplasmic domain with COOH-terminus of the protein folded into helices and b sheets to form the membrane-spanning domain. The region between the NH2-terminus and the first membrane-spanning segment forms an interhinge domain.
Band 3 is the major anion (chloride-bicarbonate) exchanger of the red cell. It regulates metabolic pathways by sequestering key pathway enzymes, such as the glycolytic enzymes glyceraldehyde-3-phosphate dehydrogenase, phosphoglycerate kinase, and aldolase, as well as carbonic anhydrase II. Band 3 contains important binding sites for interaction with other membrane proteins including ankyrin, protein 4.1, protein 4.2, and possibly spectrin.17,18 Binding of the cytoplasmic domain to ankyrin is a critical mechanism for attachment of the membrane skeleton to the plasma membrane, and the interdomain hinge at this attachment point may be a crucial determinant of the flexibility or rigidity of the erythrocyte.19
The Glycophorins Glycophorins are the most abundant integral membrane glycoproteins in erythrocytes, and, because of their high sialic acid content, they account for more than 95 percent of the periodic acid–Schiff (PAS)-staining capacity of erythrocytes.20 The glycophorins are 0-glycosylated and are composed of a single extracellular hydrophilic NH2-terminal domain, a single membrane-spanning domain, and a COOH-terminal cytoplasmic tail. Characterization of cDNA and genomic clones encoding the glycophorins has revealed that they fall into two distinct subgroups.21 Glycophorins A and B are homologous to each other and are encoded by two closely linked genes. Glycophorins C and D arise from a single locus bearing no particular homology to the genes for glycophorins A and B. Glycophorin D differs from glycophorin C by use of an alternate translation start site created by alternative splicing. Another gene linked in tandem with those for glycophorins A and B, glycophorin E, has been cloned, but no protein product has been identified.22
The functional roles of the glycophorins are beginning to be revealed. Because the glycophorins constitute more than 60 percent of the net negative surface charge of red cells, they may modulate red cell–red cell and red cell–endothelial cell interactions. GPC, which interacts in a complex with protein 4.1 and p55, plays a critical role in regulating the stability, deformability, and shape of the membrane. GPC deficiency leads to elliptocytic erythrocytes that are less stable and less deformable than normal red cells. The glycophorins play important roles in clinical immunohematology, carrying a number of blood group antigens including MN, Ss, Miltenberger V, En(a-), MKMk, and Gerbich (see Chap. 137).
Other Integral Membrane Proteins The red cell membrane contains other integral membrane proteins including the Rh D protein (see Chap. 137) and various ion pumps and channels (see below).
PERIPHERAL MEMBRANE PROTEINS
The major proteins of the erythrocyte membrane skeleton are spectrin; ankyrin; actin; proteins 4.1, 4.2, and 4.9; p55; and the adducins. These proteins form an interlocking network that attaches to the inner face of the membrane primarily by binding to the cytoplasmic domains of band 3 and the glycophorins.
Spectrin Spectrin is the most abundant and largest protein of the erythrocyte membrane skeleton, constituting 75 percent of its mass and present at a concentration of about 200,000 molecules per cell.23 Spectrin is composed of two subunits, a and b, that despite many similarities are structurally distinct and are encoded by separate genes (Fig. 27-2a).24,25 Both a and b spectrin contain homologous 106 amino acid repeats that are folded into a-helical segments containing three antiparallel helices connected by short nonhelical segments.26,27 The presence of spectrin repeats suggests that spectrin evolved from the duplication of a single ancestral gene.28

FIGURE 27-2 pectrin, ankyrin, and protein 4.1. (a) a and b spectrin. Both proteins are composed of multiple homologous triple helical repetitive segments, numbered starting from the NH2-terminus. a spectrin and b spectrin are shown in antiparallel orientation, their configuration in the spectrin heterodimer. Stippled regions represent nonhomologous segments. The aI domain (a tryptic peptide of a spectrin involved in association of a and b spectrin), the spectrin nucleation site, and the ankyrin, actin, and protein 4.1 protein-binding sites are shown. In the head region of spectrin, a and b spectrin interact to form either a heterodimer (SpD) or tetramer (SpT). The contact site between the a and b chains of a spectrin heterodimer or the opposed a and b chains of the tetramer is formed by a combined ab triple helical segment (insert). (b) Ankyrin. The three major functional and structural domains, as defined by limited proteolytic digestion, are shown. The band 3 and spectrin-binding regions are shaded. The regulatory domain is subject to extensive alternative splicing, including the band 2.2 splice, which produces an activated form of ankyrin. (c) Protein 4.1. The four major functional and structural domains, as defined by limited proteolytic digestion, are shown. The regions where the 4.1 protein binds to other membrane proteins are shaded. The protein 4.1a isoform is derived from the 4.1b isoform by deamidation of aspartic acid 508 (see text for details).

The fundamental structure of the spectrin molecule is that of ab heterodimers that align and intertwine with each other in antiparallel fashion with respect to their NH2-termini to form flexible, rodlike molecules (Fig. 27-2a).26,29 These dimers further self-associate to form tetramers and higher-order oligomers. These tetramers, composed of multiple repeats, provide a strong, elastic, rodlike filament that associates into multimolecular complexes capable of lending shape and resiliency to the overlying plasma membrane via formation of a lattice-like meshwork linked to integral membrane proteins.30 Direct interactions of a weaker nature may also occur between spectrin filaments and the lipid bilayer itself. The side-to-side assembly of a- and b- spectrin chains in a zipper-like fashion begins at a defined nucleation site composed of four repeats from each chain, a19 to a22 and b1 to b4 respectively.31,32 After tight association of complementary nucleation sites, a conformational change is initiated that promotes pairing of the remainder of the two chains. A common a-spectrin variant, aLELY, interferes with normal nucleation and decreases the synthesis of functionally competent spectrin chains and may influence clinical expression of spectrin mutations (see Chap. 43).33
The NH2-terminus of a spectrin and the COOH-terminus of b spectrin are the regions involved in ab heterodimer self-association.29 Spectrin also binds to actin and protein 4.1 via the NH2-terminus of b spectrin and ankyrin via sites in repeats b15 and b16 near the COOH-terminus.34,35 and 36 Other nonrepeat sequences in spectrin provide the recognition sites for binding to other modifiers, including kinases and calmodulin.
The functions of spectrin are to maintain cellular shape, regulate the lateral mobility of integral membrane proteins, and provide structural support for the lipid bilayer.23 Defects in the ab self-association site are associated with hereditary elliptocytosis and hereditary pyropoikilocytosis (see Chap. 43). Compound heterozygosity or homozgosity for defects outside the ab self-association site are associated with severe, recessively inherited spherocytosis.
Ankyrin Ankyrin is an asymmetric polar protein that can be separated into three functional domains by mild proteolysis: an NH2-terminal membrane-binding domain that contains sites for band 3 and other ligands, a central domain that contains sites for spectrin binding, and a COOH-terminal “regulatory” domain that influences ankyrin-protein interactions (Fig. 27-2b).23,37,38 The membrane-binding domain contains 24 tandem repeats called cdc10/ankyrin repeats that contain multiple protein-binding sites.39 Ankyrin repeats are highly conserved, L-shaped structures composed of a pair of a-helices that form an antiparallel coiled-coil, followed by an extended loop perpendicular to the helices and a b hairpin.40 These repeats have been found in proteins with a wide variety of functions.41,42 The regulatory domain consists of multiple isoforms generated by alternative splicing.43,44 One of these isoforms (ankyrin 2.2) enhances ankyrin binding to band 3 and spectrin.43
Ankyrin provides the primary linkage between the membrane skeleton via spectrin binding and the lipid bilayer via band 3 binding (Fig. 27-1). Disruption of either of these linkages significantly decreases membrane stability. Ankyrin also appears to be involved in the local segregation of integral membrane proteins within function domains on the plasma membrane. The importance of ankyrin in the maintenance of membrane stability is underscored by the observation that abnormalities of ankyrin are the most common cause of typical hereditary spherocytosis (see Chap. 43).
Protein 4.1 Protein 4.1 is a phosphoprotein that can be separated by mild chymotryptic digestion into four proteolytic domains: 30 kDa, 16 kDa, 10 kDa, and 22 to 24 kDa (Fig. 27-2c). In red cells, two molecular weight forms are found, protein 4.1a and protein 4.1b, with protein 4.1a predominating in older erythrocytes. Protein 4.1a is derived from protein 4.1b by the gradual deamidation of two Asn residues in a nonenzymatic, age-dependent manner.45 Alternative splicing leads to the production of a large number of tissue- and developmental stage-specific protein 4.1 isoforms,46,47,48,49 and 50 e.g., alternatively spliced isoforms of the 10-kDa domain contain the spectrin-actin-binding site and provide erythroid and stage-specific specificity.47,48 and 49 Protein 4.1 utilizes two different initiation codons. The upstream initiation codon encodes a protein of 135 kDa found in most nonerythroid cells.50 The downstream initiation codon encodes the 85-kDa protein found primarily in erythrocytes.
The primary role of protein 4.1 is in the linkage of the spectrin-actin membrane skeleton to the lipid bilayer by facilitating complex formation between spectrin-actin fibers, the cytoplasmic domain of band 3, and p55/GPC (Fig. 27-1).51 Qualitative or quantitative defects of protein 4.1 lead to hereditary elliptocytosis (HE) with concomitant GPC and p55 deficiency (see Chap. 43). HE-related protein 4.1 mutations have included variants that affect protein 4.1 alternative splicing and initiation codon usage. Interestingly, mice with targeted disruption of the protein 4.1 gene demonstrate, in addition to hematologic effects, subtle neurologic abnormalities.52 The applicability of this observation to humans with defects of protein 4.1 is unknown.
Protein 4.2 Protein 4.2 is a member of the transglutaminase family of proteins.53 However, protein 4.2 does not possess transglutaminase activity as it lacks a critical residue in the active transglutaminase site. There are at least four isoforms of protein 4.2 created by alternative splicing; the functional significance of these is not known.54 Protein 4.2 binds to several proteins, including band 3, protein 4.1, ankyrin, and ankyrin-protein 3 complexes. The major function of protein 4.2 is to stabilize spectrin-actin-ankyrin association with band 3.55 It may also protect the membrane skeleton from premature aging by binding calcium and other cofactors that normally activate red cell transglutaminases, as these transglutaminases would otherwise cross-link proteins and lead to their inactivation. Deficiency of protein 4.2 has been associated with recessively inherited hereditary spherocytosis (see Chap. 43). Erythrocytes from mice with targeted inactivation of the protein 4.2 gene are dehydrated spherocytes with altered cation content (increased K+/decreased Na+).56
p55 Protein p55 is a phosphoprotein member of the MAGUK (membrane-associated guanylate kinase) family of proteins.57 Homologues of p55 include signal transduction proteins, tumor suppressor genes, and proteins important in cell-cell interactions. p55 binds to protein 4.1 through a binding motif in the COOH-terminal MAGUK domain and to GPC via a PDZ motif.58 A primary deficiency state for p55 has not been described, possibly because it is a widely expressed protein, and it may play a critical role in protein-protein interactions in other tissues. Deficiency of protein 4.1 or GPC lead to concomitant p55 deficiency. Studies of this interesting protein may shed important light on mechanisms whereby the erythrocyte membrane influences other cellular processes.
Adducin Adducin, a calcium/calmodulin-binding phosphoprotein located at the spectrin-actin junctional complex, is composed of ab adducin heterodimers.59 a and b adducin are structurally similar proteins encoded by separate genes.60 Adducin contains a “MARCKS” phosphorylation domain that regulates calcium/calmodulin-regulated capping and bundling of actin filaments.61,62 Adducin promotes the interaction of spectrin and actin and binds and bundles actin filaments.63,64 A primary deficiency of adducin in human disease has not been described. Mice with targeted inactivation of b adducin suffer from compensated spherocytic anemia, suggesting that the adducins may be candidate genes for recessively inherited spherocytosis.65
Other Peripheral Membrane Proteins Dematin (protein 4.9), tropomyosin, proteins related to troponin, and other proteins associated with actin in nonerythroid cells are found in erythrocytes. The functional roles of these proteins are now being revealed. For example, the amount of dematin present in the erythrocyte declines dramatically during erythrocyte maturation suggesting that it may play an important role in cellular maturation.
FUNCTIONS OF THE ERYTHROCYTE MEMBRANE
The roles of the erythrocyte membrane include assembling and organizing proteins of the lipid bilayer and the underlying skeleton, providing the red cell with its unique deformability and stability, participating in membrane biogenesis and aging, and providing an impermeable barrier between the erythrocyte cytoplasm and the external environment.
MEMBRANE ASSEMBLY AND ORGANIZATION
Membrane organization arises from interactions between integral membrane proteins and other molecules contacting the hydrophilic faces of the membrane and by protein-protein or protein-lipid interactions within the bilayer or the underlying membrane skeleton. The avidity of these interactions is modulated by posttranslational modifications of the participating proteins. By utilizing the cytoplasmic domains of embedded proteins as attachment points, the membrane skeleton not only affixes itself to the lipid bilayer but also provides a means to order the topological arrangement of transmembrane proteins.66 This attachment constrains motion along the transverse plane.
In the intact erythrocyte membrane, the membrane skeleton appears as a lattice-like network, with about 60 percent of the lipid bilayer directly laminated to the underlying membrane skeleton.67 When skeletal preparations are stretched, the individual skeletal proteins can be visualized as a highly ordered lattice of hexagons. The corners of each hexagon are globular structures called the junctional complex composed of complexes of F-actin, along with dematin, adducin, and protein 4.1.68 Spectrin tetramers form the arms of the hexagons, cross-bridging individual junctional complexes. Spectrin cross-bridges are largely formed by spectrin tetramers, with occasional double tetramers or hexamers. Each spectrin tetramer is composed of two ab heterodimers assembled at their “head” regions into tetramers. At their tails, the tetramers bind to junctional complexes of actin, with the aid of protein 4.1 and adducin. The above horizontal protein contacts are important in the maintenance of the structural integrity of the cell, accounting for the high tensile strength of the erythrocyte.
The skeleton is affixed to the integral proteins of the membrane by several protein-protein interactions (Fig. 27-1).23,68,69 Spectrin tetramers are connected to ankyrin, the major skeleton/membrane linkage protein via an interaction site in b spectrin. Ankyrin links the underlying spectrin skeleton to tetramers of band 3, the major transmembrane protein of the red cell. At the distal ends of spectrin tetramers, spectrin binds to the membrane via linkage to protein 4.1, which binds GPC and protein p55. In addition, both spectrin and protein 4.1 bind weakly to phosphatidylserine, which preferentially is located at the inner leaflet of the lipid bilayer. These vertical protein-protein and protein-lipid interactions are critical in the stabilization of the lipid bilayer, precluding its loss from the cells.
As discussed in Chap. 43, hereditary spherocytosis is characterized by defects of vertical interactions, which lead to uncoupling of the lipid bilayer from the skeleton and a release of membrane microvesicles.70 In contrast, the principal defects in hereditary elliptocytosis and pyropoikilocytosis involve horizontal interactions of membrane skeletal proteins that maintain the two-dimensional integrity of the skeleton.
Red cell membrane proteins are subject to a variety of posttranslational modifications or other regulatory effects including phosphorylation, fatty acid acylation, methylation, glycosylation, deamidation, oxidation, and limited proteolytic cleavage.2 With the exception of membrane protein phosphorylation, such modifications are relatively static and irreversible. In contrast, membrane protein phosphorylation represents a highly dynamic system of multiple protein kinases and phosphatases that constantly phosphorylate and dephosphorylate serine, threonine, and tyrosine residues, often in an amino-acid-specific and protein-site-specific manner, thereby tightly regulating association of membrane proteins. Additionally, membrane protein associations are influenced by a variety of intracellular factors including calcium and calmodulin, phosphoinositides, and polyanions such as 2,3-bis-phosphoglycerate.
The red cell surface is negatively charged, primarily because of a high concentration of neuraminic acid residues. Ninety percent of these residues reside on glycophorin A with the remaining shared by the other glycophorins and band 3. Alterations in erythrocyte surface charge appear to have deleterious effects on the cell. For example, in sickle red cells, surface charge clustering may play a role in the adhesion of these cells to the surface of endothelial cells.
CELLULAR DEFORMABILITY AND MEMBRANE STABILITY
The most important property of red cells required for normal survival is cellular deformability.71 Deformability refers to the ability of the erythrocyte to undergo distortions and deformations and then to resume its normal shape without fragmentation or loss of integrity. This is best exemplified in the wall of the splenic sinus where red cells squeeze through narrow slits among the endothelial cells that line the splenic sinus wall. The cellular deformability of erythrocytes is determined by three factors: (1) cell geometry (biconcave disc shape); (2) cytoplasmic viscosity, principally determined by the properties and the concentration of hemoglobin in the cells; and (3) intrinsic viscoelastic properties of the red cell membrane (or membrane deformability).72 Among these factors, cell geometry as determined by the contribution of the surface-to-volume ratio is the most important, as exemplified by the cellular lesion of hereditary spherocytes. On the other hand, the intrinsic viscoelastic properties of the red cell are likely to have a relatively small effect on red cell survival. Southeast Asian ovalocytes are very rigid, yet they have a normal survival in vivo.
The cellular geometry, i.e., the biconcave disc shape of red cells, is critical for their survival. This cell surface shape provides a high ratio of surface area to cellular volume. The normal volume of the erythrocyte is about 90 µm3; the minimum surface area that could encase this volume would be a sphere of about 98 µm3. The surface area of a biconcave disc enclosing this volume is about 140 µm3. Thus, shape alone provides the red cell with a considerable amount of redundant membrane and cytoskeleton. This feature provides the extra membrane surface area needed when red cells swell. More importantly, this geometric arrangement allows red cells to be stretched as they undergo deformation and distortion in response to the mechanical stress of the circulation. Loss of membrane by partial phagocytosis in immune hemolytic anemias or by fragmentation of bits of membrane from the cell in patients with cytoskeletal defects leads to elliptocytic or spherocytic shapes having greatly reduced surface area and, therefore, much less deformability.73 The consequent reduction in tolerance of these cells to osmotic stress explains why anemias due to membrane defects are often accompanied by osmotic fragility, the basis for the clinical laboratory test. Conversely, if erythrocytes are engorged with water, they become macrospherocytic and less deformable.
Thus it is obvious that the organization of the membrane skeleton and its attachment to the plasma membrane influence the stability and deformability of the red cell. In the resting state, the folded helical segments of spectrin are highly coiled. Membrane deformation is accompanied by a rearrangement of the spectrin-actin-based membrane skeleton network with some spectrin molecules becoming uncoiled and extended, whereas others become more compressed and folded, resulting in no net change in surface area. Thus, shape changes but surface area does not. The extent to which this stretching and compression are possible determines the extent of deformability. Mutations or acquired alterations in membrane proteins that influence the spectrin-actin-based lattice of proteins leads to membrane loss with a concomitant decrease in surface area and a change in cell geometry.
Red cell viscosity is largely determined by hemoglobin content.73 At normal intracellular concentrations (27–35g/dl), viscosity contributes very little to cellular deformability. When erythrocytes become dehydrated, the effective intracellular hemoglobin concentration rises, and viscosity increases exponentially. Membrane pumps and channels normally maintain intracellular volumes that hold hemoglobin concentrations below the level at which cytoplasmic viscosity has an impact on deformability. Inherited anomalies of pumps or channels (e.g., hereditary xerocytosis) or derangements caused by polymerized or crystallized hemoglobin (e.g., sickle cell anemia or HbC disease), lead to cellular dehydration and greatly increased red cell viscosity.
MEMBRANE MATERIAL PROPERTIES
The material properties of the membrane reflect the properties of both the lipid bilayer and the skeleton. During deformation, the membrane undergoes bending, which is restricted by the incompressibility of the lipid bilayer. It has been proposed that such bending is facilitated by the rapid translocation of cholesterol from the inner to the outer hemileaflet (Fig. 27-3). When red cells are suspended in hypotonic solutions, such as during osmotic fragility testing (see Chap. 43), they swell, reaching a nearly spherical shape because the bilayer membrane cannot expand its surface area more than 3 to 4 percent. Further lowering of the osmotic pressure results in membrane rupture, and intracellular hemoglobin is discharged into the supernatant.

FIGURE 27-3 Material properties of the red cell membrane. (a) Membrane bending. The degree of membrane bending is restricted by the limited compressibility of the lipid bilayer. The rapid translocation of cholesterol (shaded diamonds) from the inner to the outer leaflet reduces the compression of the inner bilayer leaflet, thereby facilitating bending. (b) Skeletal deformation. While hydrophobicity of the red cell membrane lipid bilayer precludes the increase in its surface area without rupture, the membrane can undergo a large deformation under a constant surface area because of the viscoelastic properties of the membrane skeleton. During uniaxial extension, the skeleton undergoes stretching (top rectangle). After a cessation of an external force, a square surface area is resumed because the protein connections within this elastic skeletal network remain intact. Extensive or prolonged uniaxial extension leads to a rearrangement of the skeletal network because of a disruption of existing skeletal protein connections and a formation of new protein contacts. This leads to a permanent plastic deformation (bottom rectangle). (c) Bilayer couple hypothesis and the stomatocyte-discocyte-echinocyte transformation. Red cell shape reflects the ratio of the surface areas of the two hemileaflets of the lipid bilayer. The compounds (black triangles) that preferentially intercalate into the outer hemileaflet of the lipid bilayer produce its expansion followed by red cell crenation (echinocytosis or acanthocytosis). In contrast, expansion of the inner lipid bilayer leaflet produces a cup shape (stomatocytosis) and surface invaginations.

The membrane skeleton determines both the solid and semisolid properties of the membrane. The solid properties are exemplified by an elastic extension of cells that completely restores their normal shape after the applied force has been removed. An example is a cell that has been deformed when passing through fenestrations of the splenic sinus wall. This elastic recovery of the normal shape is facilitated by the unique molecular anatomy of the skeletal lattice. Here the individual hexagons are in a compact, unextended configuration with the junctional complexes close to each other and the cross-linking arms of spectrin tetramers folded between them, thus allowing large unidirectional extensions without disruption of the lattice (see Fig. 27-3). The skeleton remains unperturbed during such deformation. On the other hand, application of large or prolonged forces allows the skeletal elements to reorganize into a new configuration; this produces a permanent plastic deformation. When the force is excessive, membrane fragmentation ensues. An example is vessels damaged when red cells are trapped by fibrin strands; after release from this site, the erythrocytes either are permanently deformed or are fragmented.
MEMBRANE BIOGENESIS AND AGING
Membrane protein biosynthesis occurs asynchronously during erythropoiesis. Early in erythroid development, the major proteins of the membrane skeleton (spectrin, ankyrin, and the 4.1 protein) are synthesized.74,75 However, they turn over rapidly and do not assemble into a permanent network. At the proerythroblast stage, the synthesis of band 3 is initiated and, together with the synthesis of protein 4.1, increases up to the late erythroblast stage. During this time, mRNA levels and synthesis of spectrin and ankyrin protein decline. In contrast, the fraction of newly assembled spectrin and ankyrin protein on the membrane progressively increases, and the turnover of these proteins on the membrane declines.
Increased recruitment and stabilization of spectrin and ankyrin on the membrane in spite of the declining synthesis of these proteins is temporally related to a progressive increase in the synthesis of band 3 and protein 4.1, the principal bilayer anchors of the membrane skeleton.76 Thus early studies suggested that the early steps of red cell membrane assembly were controlled by band 3 production where, after insertion into the membrane, it directed the assembly of stable macromolecular complexes from presynthesized pools of other proteins.77,78 The role of band 3 in membrane assembly has been questioned by the following recent findings: (1) The organization of preformed pools of cytoskeletal elements induced by band 3 synthesis is not seen in nontransformed cells; and (2) band 3 knock-out mice exhibit normal membrane biogenesis even though their red cell membranes are unstable in the circulation.79,80
The biosynthesis and assembly of spectrin subunits is complex. b-spectrin biosynthesis exceeds that of a spectrin in the early erythroblasts derived from both embryonic (yolk sac) and fetal/adult (liver/spleen) origins. This ratio is preserved during later stages of erythropoiesis in embryonic cells, but not in fetal/adult-derived late erythroblasts and reticulocytes. In these latter cells, a-spectrin gene expression increases, whereas b-spectrin gene expression remains constant, resulting in a predominance of a-spectrin mRNA and protein during the late stages, when active assembly of the actual membrane is occurring most rapidly. ab-spectrin subunits are incorporated into the membrane in a 1:1 stoichiometric ratio, regardless of their rates of synthesis.74,81,82 This point is important in the analysis of inherited hemolytic anemias. Human a-spectrin synthesis exceeds that of b-spectrin by 2:1 during the later stages of erythropoiesis, when, presumably, membrane assembly is proceeding rapidly. The availability of b-spectrin subunits therefore determines the maximum rate and amount of stable spectrin assembly. Thus, mutations reducing steady-state levels of newly synthesized b spectrin should have a far greater phenotypic impact than do mutations causing comparable decreases in a-spectrin biosynthesis. Analyses of patients with hereditary hemolytic anemias support this prediction (see Chap. 43).
At the stage of orthochromatic erythroblast, when membrane biogenesis is nearly completed, the cell membrane undergoes a series of critical remodeling steps.83,84 The membrane surrounding the nucleus contains an actin ring that likely participates in the expulsion of the nucleus from the erythroblast.85 At the same time, the spectrin skeleton segregates into the region of the incipient reticulocyte, while some surface receptors cluster in membrane regions surrounding the extruded nucleus.
Some synthesis of spectrin, band 3, protein 4.1, and GPC continues in the newly enucleated reticulocyte, but most membrane remodeling occurs after translation. The reticulocyte is multilobular and motile; it possesses mitochondria, polyribosomes, and numerous membrane proteins that are either absent or much less abundant in mature red cells. In addition, phospholipid composition and inside-outside lipid distribution are different. Reticulocytes are far less deformable and considerably more unstable mechanically than are mature erythrocytes. Maturation begins in the bone marrow and lasts for 2 or 3 days. It is completed in the circulation and perhaps in the spleen where it has been termed splenic polishing. Reticulocytes first become cup-shaped before acquiring their final biconcave disc shape. This process involves major reorganization of both membrane phospholipids and cytoskeletal and embedded proteins, as well as the loss of lipids and proteins, including receptors for transferrin, insulin, and fibronectin.
RED CELL AGING
The mechanism of red cell aging is discussed in Chap. 29.
FETAL RED CELLS
Fetal erythrocytes differ in a number of respects including activity of both glycolytic and nonglycolytic enzymes, altered ATP and phosphate metabolism, differences in methemoglobin content and oxygen affinity, and altered storage characteristics (reviewed in Gallagher86). These erythrocytes exhibit increased rigidity, increased mechanical fragility, and decreased life span (average 45 to 70 days) compared to adult red cells.
There are also differences in the membranes of fetal and adult erythrocytes. ABO and I antigens and the receptors for the adsorbed serum antigens of the Lewis system are incompletely expressed. Fetal membranes are more permeable to monovalent cations and contain less Na+-K+-ATPase activity. They contain more phospholipid and cholesterol per cell and, as a consequence, have a larger surface-to-volume ratio and are slightly more osmotically resistant than adult cells. The ratio of sphingomyelin to phosphatidylcholine is increased in fetal membranes and differences in fatty acid composition exist, but these changes evidently tend to balance each other, as membrane fluidity is normal. The protein composition of fetal red cell membrane is quantitatively normal.
MEMBRANE PERMEABILITY
The normal red cell membrane is nearly impermeable to monovalent and divalent cations, thereby maintaining a high potassium, low sodium, and very low calcium content. In contrast, the red cell is highly permeable to water and anions, which are readily exchanged, and as a result erythrocytes behave as nearly perfect osmometers. Water and ion transport pathways in the red cell membrane (Fig. 27-4) include energy-driven membrane pumps, gradient-driven systems, and various channels.87,88 An important feature of the normal red cell is its ability to maintain a constant volume. The mechanisms by which red cells “sense” changes in cell volume and activate appropriate volume regulatory pathways are unknown. Glucose is transported without the expenditure of energy utilizing a transporter, while larger charged molecules, such as ATP and related compounds, do not cross the normal red cell membrane, although phosphoenolpyruvate is an exception to this rule (Chap. 140).

FIGURE 27-4 Principal ion transport pathways of the human erythrocyte. AE-1: band 3, the anion exchanger; AQP1: the water channel aquaporin 1; KCC-1: KCl cotransport system of the family of chloride-cation cotransporters; NKCC2: basolateral molecular form of Na-K-Cl cotransport; SK: small conductance potassium channel. Reprinted with permission from Brugnara.87

The effects of disruption of the red cell permeability barrier are illustrated by complement-mediated hemolysis. Complement activation on the red cell surface leads to formation of the membrane attack complex, composed of terminal complement components embedded in the lipid bilayer. This multimolecular complex acts as a cation channel, allowing passive movements of sodium, potassium, and calcium across the membrane according to their concentration gradients. Attracted by fixed anions, such as hemoglobin, ATP, and 2,3-BPG, sodium accumulates in the cell in excess of potassium loss and in excess of the compensatory efforts of the Na+/K+-pump. The resulting increase in intracellular monovalent cations and water is followed by cell swelling and ultimately colloid osmotic hemolysis.
ENERGY-DRIVEN MEMBRANE PUMPS
In the red cell, two ion-motive ATPase-dependent cation pumps maintain low intracellular sodium and calcium and high potassium.87 The ouabain inhibitable Na+-K+-ATPase (the sodium pump) extrudes sodium in exchange for potassium in a 3:2 stoichiometry. Ca++-ATPase is a calmodulin-activated pump that extrudes calcium from the red cell and maintains a very low intracellular calcium concentration, thus protecting cells from multiple deleterious effects of calcium. Examples of such deleterious effects include echinocytosis, membrane vesiculation, calpain activation, membrane proteolysis, and cellular dehydration. Elevated intracellular calcium plays an important role in the pathophysiology of sickle cell disease, as increased levels of intracellular calcium observed during sickling are due to an increase in Ca++ flux and reduced activity of the Ca++-ATPase. The membrane also contains an ATP-driven GSSG transporter (Chap. 26) and amino acid transport systems.
GRADIENT-DRIVEN SYSTEMS
The Na+/K+ gradient established by the sodium pump is used by several passive, gradient-driven systems to move ions across the red cell membrane.87 These include the K+Cl+-cotransporter, band 3 (see above), the Na+-K+Cl-cotransporter, and the Na+-H+-exchanger. The Na+-K+Cl-cotransporter plays only a minor role in the red cell. The Na+-H+-exchanger appears to play a role primarily in early erythrocyte maturation. The K+Cl+-cotransporter is a typical carrier-mediated cotransporter, which is particularly active in reticulocytes.89,90 It is activated by cell swelling, acidification, depletion of intracellular magnesium and thiol oxidation.
CHANNELS
Channels of the red cell include voltage-gated channels (mediated via Na+K+-ATPase), water channels (the aquaporins), and the Ca++-activated K+-channel. The Ca++-activated K+-channel, also called the Gardos channel after its discoverer Dr. George Gardos, causes selective loss of K+ in response to an increase in intracellular Ca.++91,92 In sickle cells, increased activity of both the Gardos channel and the K+-Cl-cotransporter leads to a net loss of K+ and water, leading to cellular dehydration and the formation of intermediate and hyperdense erythrocytes.93,94 Recently, pharmacologic manipulation of these two channels has been tried in attempts to improve cellular hydration of the red cell and ameliorate the clinical course of patients with sickle cell disease.95,96
The aquaporins are membrane channel proteins that serve as selective pores through which water crosses the plasma membrane.97,98 Aquaporin-1, AQP1, which is expressed in many tissues including erythrocytes, contributes to the ability of the red cell to adjust rapidly to changes in osmolality. AQP1 contains the epitope for the Colton blood group system. The genetic basis of the rare Colton null phenotype has been identified as a mutation of the highly conserved NPA motif of AQP1 essential for channel function.99 Colton null individuals exhibit no obvious clinical phenotype, although mice with targeted inactivation of AQP1 become hyperosmolar after fluid restriction.100 Recently, evidence for the presence of AQP3 in erythrocytes has been presented.101
CHAPTER REFERENCES

1.
Ways P, Hanahan DJ: Characterization and quantitation of red cell lipids in normal man. J Lipid Res 5:318, 1964.

2.
Cohen CM, Gascard P: Regulation and post-translational modification of erythrocyte membrane and membrane-skeletal proteins. Semin Hematol 29:244, 1992.

3.
Bevers EM, Comfurius P, Dekkers DW, Zwaal RF: Lipid translocation across the plasma membrane of mammalian cells. Biochim Biophys Acta 1439:317, 1999.

4.
Andrews DA, Low PS: Role of red blood cells in thrombosis. Curr Opin Hematol 6:76, 1999.

5.
Tang X, Halleck MS, Schlegel RA, Williamson P: A subfamily of P-type ATPases with aminophospholipid transporting activity. Science 272:1495, 1996.

6.
Dekkers DW, Comfurius P, Schroit AJ, et al: Transbilayer movement of NBD-labeled phospholipids in red blood cell membranes: outward-directed transport by the multidrug resistance protein 1 (MRP1). Biochemistry 37:14833, 1998.

7.
Kamp D, Haest CW: Evidence for a role of the multidrug resistance protein (MRP) in the outward translocation of NBD-phospholipids in the erythrocyte membrane. Biochim Biophys Acta 1372:91, 1998.

8.
Basse F, Stout JG, Sims PJ, Wiedmer T: Isolation of an erythrocyte membrane protein that mediates Ca2+-dependent transbilayer movement of phospholipid. J Biol Chem 271:17205, 1996.

9.
Zhou Q, Zhao J, Stout JG, et al: Molecular cloning of human plasma membrane phospholipid scramblase. A protein mediating transbilayer movement of plasma membrane phospholipids. J Biol Chem 272:18240, 1997.

10.
Zhao J, Zhou Q, Wiedmer T, Sims PJ: Level of expression of phospholipid scramblase regulates induced movement of phosphatidylserine to the cell surface. J Biol Chem 273:6603, 1998.

11.
Zhou Q, Sims PJ, Wiedmer T: Identity of a conserved motif in phospholipid scramblase that is required for Ca2+-accelerated transbilayer movement of membrane phospholipids. Biochemistry 37:2356, 1998.

12.
Dekkers DW, Comfurius P, Vuist WM, et al: Impaired Ca2+-induced tyrosine phosphorylation and defective lipid scrambling in erythrocytes from a patient with Scott syndrome: a study using an inhibitor for scramblase that mimics the defect in Scott syndrome. Blood 91:2133, 1998.

13.
Stout JG, Basse F, Luhm RA, et al: Scott syndrome erythrocytes contain a membrane protein capable of mediating Ca2+-dependent transbilayer migration of membrane phospholipids. J Clin Invest 99:2232, 1997.

14.
Fairbanks G, Steck TL, Wallach DFH: Electrophoretic analysis of the major polypeptides of the human erythrocyte membrane. Biochemistry 10:2606, 1971.

15.
Alper SL: The band 3-related anion exchanger (AE) gene family. Annu Rev Physiol 53:549, 1991.

16.
Tanner MJ: Molecular and cellular biology of the erythrocyte anion exchanger (AE1). Semin Hematol 30:34, 1993.

17.
Low PS: Structure and function of the cytoplasmic domain of band 3: center of erythrocyte membrane-peripheral protein interactions. Biochim Biophys Acta 864:145, 1986.

18.
Van Dort HM, Moriyama R, Low PS: Effect of band 3 subunit equilibrium on the kinetics and affinity of ankyrin binding to erythrocyte membrane vesicles. J Biol Chem 273:14819, 1998.

19.
Mohandas N, Winardi R, Knowles D, et al: Molecular basis for membrane rigidity of hereditary ovalocytosis. A novel mechanism involving the cytoplasmic domain of band 3. J Clin Invest 89:686, 1992.

20.
Chasis JA, Mohandas N: Red blood cell glycophorins. Blood 80:1869, 1992.

21.
Cartron JP, Rahuel C: Human erythrocyte glycophorins: protein and gene structure analyses. Transfus Med Rev 6:63, 1992.

22.
Kudo S, Fukuda M: Identification of a novel human glycophorin, glycophorin E, by isolation of genomic clones and complementary DNA clones utilizing polymerase chain reaction. J Biol Chem 265:1102, 1990.

23.
Morrow JS, Rimm DL, Kennedy SP, et al: Of membrane stability and mosaics: the spectrin cytoskeleton, in Handbook of Physiology, edited by J Hoffman, J Jamieson, p 485. Oxford, London, 1997.

24.
Gallagher PG, Forget BG: Spectrin genes in health and disease. Semin Hematol 30:4, 1993.

25.
Winkelmann JC, Forget BG: Erythroid and nonerythroid spectrins. Blood 81:373, 1993.

26.
Bennett V, Lambert S: The spectrin skeleton: from red cells to brain. J Clin Invest 87:1483, 1991.

27.
Speicher DW, Marchesi VT: Erythrocyte spectrin is comprised of many homologous triple helical segments. Nature 311:177, 1984.

28.
Thomas GH, Newbern EC, Korte CC, et al: Intragenic duplication and divergence in the spectrin superfamily of proteins. Mol Biol Evol 14:1285, 1997.

29.
Cherry L, Menhart N, Fung LW: Interactions of the a-spectrin Nterminal region with b spectrin. J Biol Chem 274:2077, 1999.

30.
Grum VL, Li DN, MacDonald RI, Mondragon A: Structures of two repeats of spectrin suggest models of flexibility. Cell 98:523, 1999.

31.
Ursitti JA, Kotula L, DeSilva TM, et al: Mapping the human erythrocyte beta-spectrin dimer initiation site using recombinant peptides and correlation of its phasing with the alpha-actinin dimer site. J Biol Chem 271:6636, 1996.

32.
Viel A, Gee MS, Tomooka L, Branton D: Motifs involved in interchain binding at the tail-end of spectrin. Biochim Biophys Acta 1384:396, 1998.

33.
Alloisio N, Morle L, Marechal J, et al: Sp alpha V/41: a common spectrin polymorphism at the alpha IV-alpha V domain junction. Relevance to the expression level of hereditary elliptocytosis due to alpha-spectrin variants located in trans. J Clin Invest 87:2169, 1991.

34.
Tanaka T, Kadowski K, Lazarides E, Sobue K: Ca2+- dependent regulation of the spectrin actin interaction by calmodulin and protein 4.1. J Biol Chem 266:1134, 1991.

35.
Becker PS, Schwartz MA, Morrow JS, Lux SE: Radiolabel-transfer cross-linking demonstrates that protein 4.1 binds to the N-terminal region of beta spectrin and to actin in binary interactions. Eur J Biochem 193:827, 1990.

36.
Kennedy SP, Warren SL, Forget BG, Morrow JS: Ankyrin binds to the 15th repetitive unit of erythroid and nonerythroid b spectrin. J Cell Biol 115:267, 1991.

37.
Bennett V: Ankyrins. Adaptors between diverse plasma membrane proteins and the cytoplasm. J Biol Chem 267:8703, 1992.

38.
Peters LL, Lux SE: Ankyrins: structure and function in normal cells and hereditary spherocytes. Semin Hematol 30:85, 1993.

39.
Michaely P, Bennett V: The membrane-binding domain of ankyrin contains four independently folded subdomains, each comprised of six ankyrin repeats. J Biol Chem 268:22703, 1993.

40.
Batchelor AH, Piper DE, de la Brousse FC, et al: The structure of GABPa/b: an ETS domain-ankyrin repeat heterodimer bound to DNA. Science 279:1037, 1998.

41.
Bork P: Hundreds of ankyrin-like repeats in functionally diverse proteins: mobile modules that cross phyla horizontally? Proteins 17:363, 1993.

42.
Sedgwick SG, Smerdon SJ: The ankyrin repeat: a diversity of interactions on a common structural framework. Trends Biochem Sci 24:311, 1999.

43.
Davis LH, Davis JQ, Bennett V: Ankyrin regulation: an alternatively spliced segment of the regulatory domain functions as an intramolecular modulator. J Biol Chem 267:18966, 1992.

44.
Gallagher PG, Tse WT, Scarpa AL, et al: Structure and organization of the human ankyrin-1 gene: basis for complexity of pre-mRNA processing. J Biol Chem 272:19220, 1997.

45.
Inaba N, Gupta KC, Kuwabara M, et al: Deamidation of human erythrocyte protein 4.1: possible role in aging. Blood 79:3355, 1992.

46.
Conboy J: The role of alternative pre-mRNA splicing in regulating the structure and function of skeletal protein 4.1. Proc Soc Exper Biol Med 220:73, 1999.

47.
Gascard P, Lee G, Coulombel L, et al: Characterization of multiple isoforms of protein 4.1r expressed during erythroid terminal differentiation. Blood 92:4404, 1998.

48.
Huang JP, Tang CJ, Kou GH, et al: Genomic structure of the locus encoding protein 4.1. Structural basis for complex combinational patterns of tissue-specific alternative RNA splicing. J Biol Chem 268:3758, 1993.

49.
Tang TK, Qin Z, Leto T, et al: Heterogeneity of mRNA and protein products arising from the protein 4.1 gene in erythroid and nonerythroid tissues. J Cell Biol 110:617, 1990.

50.
Chasis JA, Coulombel L, Conboy J, et al: Differentiation-associated switches in protein 4.1 expression. Synthesis of multiple structural isoforms during normal human erythropoiesis. J Clin Invest 91:329, 1993.

51.
Marfatia SM, Leu RA, Branton D, Chishti AH: Identification of the protein 4.1 binding interface on glycophorin C and p55, a homologue of the Drosophila discs-large tumor suppressor protein. J Biol Chem 270:715, 1995.

52.
Shi ZT, Afzal V, Coller B, et al: Protein 4.1R-deficient mice are viable but have erythroid membrane skeleton abnormalities. J Clin Invest 103:331, 1999.

53.
Yawata Y: Red cell membrane protein band 4.2: phenotypic, genetic and electron microscopic aspects. Biochim Biophys Acta 1204:131, 1994.

54.
Wada H, Kanzaki A, Yawata A, et al: Late expression of red cell membrane protein 4.2 in normal human erythroid maturation with seven isoforms of the protein 4.2 gene. Exp Hematol 27:54, 1999.

55.
Rybicki AC, Schwartz RS, Hustedt EJ, Cobb CE: Increased rotational mobility and extractability of band 3 from protein 4.2-deficient erythrocyte membranes: evidence of a role for protein 4.2 in strengthening the band 3-cytoskeleton linkage. Blood 88:2745, 1996.

56.
Peters LL, Jindel HK, Gwynn B, et al: Mild spherocytosis and altered red cell ion transport in protein 4.2-null mice. J Clin Invest 103:1527, 1999.

57.
Chishti AH: Function of p55 and its nonerythroid homologues. Curr Opin Hematol 5:116, 1998.

58.
Marfatia SM, Morais-Cabral JH, Kim AC, et al: The PDZ domain of human erythrocyte p55 mediates its binding to the cytoplasmic carboxyl terminus of glycophorin C. Analysis of the binding interface by in vitro mutagenesis. J Biol Chem 272:24191, 1997.

59.
Li X, Bennett V: Identification of the spectrin subunit and domains required for formation of spectrin/adducin/actin complexes. J Biol Chem 271:15695, 1996.

60.
Joshi R, Gilligan DM, Otto E, et al: Primary structure and domain organization of human alpha and beta adducin. J Cell Biol 115:665, 1991.

61.
Kuhlman PA, Fowler VM: Purification and characterization of an alpha 1 beta 2 isoform of CapZ from human erythrocytes: cytosolic location and inability to bind to Mg2+ ghosts suggest that erythrocyte actin filaments are capped by adducin. Biochemistry 36:13461, 1997.

62.
Kuhlman PA, Hughes CA, Bennett V, Fowler VM: A new function for adducin. Calcium/calmodulin-regulated capping of the barbed ends of actin filaments. J Biol Chem 271:7986, 1996.

63.
Li X, Matsuoka Y, Bennett V: Adducin preferentially recruits spectrin to the fast growing ends of actin filaments in a complex requiring the MARCKS-related domain and a newly defined oligomerization domain. J Biol Chem 273:19329, 1998.

64.
Matsuoka Y, Li X, Bennett V: Adducin is an in vivo substrate for protein kinase C: phosphorylation in the MARCKS-related domain inhibits activity in promoting spectrin-actin complexes and occurs in many cells, including dendritic spines of neurons. J Cell Biol 142:485, 1998.

65.
Gilligan DM, Lozovatsky L, Mohandas N, et al: Targeted disruption of the beta adducin gene (Add2) causes red blood cell spherocytosis in mice. Proc Natl Acad Sci USA 96:10717, 1999.

66.
De Matteis MA, Morrow JS: The role of ankyrin and spectrin in membrane transport and domain formation. Curr Opin Cell Biol 10:542, 1998.

67.
Liu SC, Derick LH, Palek J: Visualization of the hexagonal lattice in the erythrocyte membrane skeleton. J Cell Biol 104:527, 1987.

68.
Gilligan DM, Bennett V: The junctional complex of the membrane skeleton. Semin Hematol 30:74, 1993.

69.
Workman RF, Low PS: Biochemical analysis of potential sites for protein 4.1-mediated anchoring of the spectrin-actin skeleton to the erythrocyte membrane. J Biol Chem 273:6171, 1998.

70.
Tse WT, Lux SE: Red blood cell membrane disorders. Br J Haematol 104:2, 1999.

71.
Mohandas N, Chasis JA: Red blood cell deformability, membrane material properties and shape: regulation by transmembrane, skeletal and cytosolic proteins and lipids. Semin Hematol 30:171, 1993.

72.
Hochmuth RM, Waugh RE: Erythrocyte membrane elasticity and viscosity. Annu Rev Physiol 49:209, 1987.

73.
Narla M, Chasis JA, Shohet SB: The influence of membrane skeleton on red cell deformability, membrane material properties, and shape. Semin Hematol 20:225, 1983.

74.
Hanspal M, Palek J: Synthesis and assembly of membrane skeletal proteins in mammalian red cell precursors. J Cell Biol 105:147, 1987.

75.
Lazarides E: From genes to structural morphogenesis: the genesis and epigenesis of a red blood cell. Cell 51:345, 1987.

76.
Hanspal M, Palek J: Biogenesis of normal and abnormal red blood cell membrane skeleton. Semin Hematol 29:305, 1992.

77.
Hanspal M, Hanspal JS, Kalraiya R, Palek J: The expression and synthesis of the band 3 protein initiates the formation of a stable membrane skeleton in murine Rauscher-transformed erythroid cells. Eur J Cell Biol 58:313, 1992.

78.
Woods CM, Boyer B, Vogt PK, Lazarides E: Control of erythroid differentiation: asynchronous expression of the anion transporter and the peripheral components of the membrane skeleton in AEV- and S13-transformed cells. J Cell Biol 103:1789, 1986.

79.
Peters LL, Shivdasani RA, Liu SC, et al: Anion exchanger 1 (band 3) is required to prevent erythrocyte membrane surface loss but not to form the membrane skeleton. Cell 86:917, 1996.

80.
Southgate CD, Chishti AH, Mitchell B, et al: Targeted disruption of the murine erythroid band 3 gene results in spherocytosis and severe haemolytic anaemia despite a normal membrane skeleton. Nat Genet 14:227, 1996.

81.
Bodine DM, Birkenmeier CS, Barker JE: Spectrin deficient inherited hemolytic anemias in the mouse: characterization by spectrin synthesis and mRNA activity in reticulocytes. Cell 37:721, 1984.

82.
Peters LL, White RA, Birkenmeier CS, et al: Changing in cytoskeletal mRNA expression and protein synthesis patterns during murine erythropoiesis in vivo. Proc Natl Acad Sci USA 89:5749, 1992.

83.
Koury MJ, Bondurant MC, Rana SS: Changes in erythroid membrane proteins during erythropoietin-mediated terminal differentiation. J Cell Physiol 133:438, 1987.

84.
Chasis JA, Prenant M, Leung A, Mohandas N: Membrane assembly and remodeling during reticulocyte maturation. Blood 74:1112, 1989.

85.
Takano-Ohmuro H, Mukaida M, Morioka K: Distribution of actin, myosin, and spectrin during enucleation in erythroid cells of hamster embryo. Cell Motil Cytoskeleton 34:95, 1996.

86.
Gallagher PG: Disorders of erythrocyte metabolism and shape, in Hematologic Problems in the Neonate, edited by RD Christensen, p 209. Saunders, Philadephia, 1999.

87.
Brugnara C: Erythrocyte membrane transport physiology. Curr Opin Hematol 4:122, 1997.

88.
Ellory JC, Gibson JS, Stewart GW: Pathophysiology of abnormal cell volume in human red cells. Contrib Nephrol 123:220, 1998.

89.
Lauf PK, Bauer J, Adragna NC, et al: Erythrocyte K-Cl cotransport: properties and regulation. Am J Physiol 263:C917, 1992.

90.
Gillen CM, Brill S, Payne JA, Forbush B: Molecular cloning and functional expression of the K-Cl cotransporter from rabbit, rat, and human. A new member of the cation-chloride cotransporter family. J Biol Chem 271:16237, 1996.

91.
Ishii TM, Silvia C, Hirschberg B, et al: A human intermediate conductance calcium-activated potassium channel. Proc Natl Acad Sci USA 94:11651, 1997.

92.
Vandorpe DH, Shmukler BE, Jiang LW, et al: cDNA cloning and functional characterization of the mouse Ca++-gated K+ channel, mIK1. Roles in regulatory volume decrease and erythroid differentiation. J Biol Chem 273:21553, 1998.

93.
Brugnara C: Erythrocyte dehydration in pathophysiology and treatment of sickle cell disease. Curr Opin Hematol 2:132, 1995.

94.
Schwartz RS, Musto S, Fabry ME, Nagel RL: Two distinct pathways mediate the formation of intermediate density cells and hyperdense cells from normal density sickle red blood cells. Blood 92:4844, 1998.

95.
Brugnara C, Gee B, Armsby CC, et al: Therapy with oral clotrimazole induces inhibition of the Gardos channel and reduction of erythrocyte dehydration in patients with sickle cell disease. J Clin Invest 97:1227, 1996.

96.
De Franceschi L Bachir D, Galacteros F, et al: Oral magnesium supplements reduce erythrocyte dehydration in patients with sickle cell disease. J Clin Invest 100:1847, 1997.

97.
Lee MD, King LS, Agre P: The aquaporin family of water channel proteins in clinical medicine. Medicine (Baltimore) 76:141, 1997.

98.
Agre P, Bonhivers M, Borgnia MJ: The aquaporins, blueprints for cellular plumbing systems. J Biol Chem 273:14659, 1998.

99.
Chretien S, de Figueiredo M, Cartron JP: A single mutation inside the NPA motif of aquaporin 1 found in a Colton-null phenotype. Blood 92:5a, 1998.

100.
Ma TH, Yang BX, Gillespie A, et al: Severely impaired urinary concentrating ability in transgenic mice lacking aquaporin-1 water channels. J Biol Chem 273:4296, 1998.

101.
Roudier N, Verbavatz JM, Maurel C, et al: Evidence for the presence of aquaporin-3 in human red blood cells. J Biol Chem 273:8407, 1998.
Books@Ovid
Copyright © 2001 McGraw-Hill
Ernest Beutler, Marshall A. Lichtman, Barry S. Coller, Thomas J. Kipps, and Uri Seligsohn
Williams Hematology