The red blood cell (RBC, erythrocyte) transports oxygen from the lungs to tissues and organs and helps to transport carbon dioxide to the lungs to be exhaled. The mammalian RBC has evolved to do this quite efficiently but has sacrificed greatly for this purpose by losing its nucleus.2,3 The resulting RBC shape facilitates oxygen acquisition and release. However, the cell is left without the ability to synthesize replacement proteins for vital pathways of energy production and membrane integrity, among others. As a result, its life span is limited to about 120 days because these processes inevitably fail. Mechanisms have evolved to maximize oxygen exchange during the life span of the cell and to counter the natural chemical processes that oxidize heme iron, which would leave it nonfunctional for oxygen transport. This chapter examines the energy production pathways of the RBC, discusses mechanisms for maintaining reduced heme iron and facilitating oxygen delivery, and explores the features of the RBC membrane that contribute to effective oxygen delivery for RBCs.

Energy Production—Anaerobic Glycolysis

At the time the RBC ejects its nucleus, the cellular organelles are also packaged and ejected from the cell. Without mitochondria, the RBC is left to rely on glycolysis for energy production. Fortunately, oxygen exchange and oxygen binding to heme iron do not require energy.⁴ Other RBC metabolic processes do require energy, however, and they are listed in Box 9-1. Loss of energy leads to RBC death. In fact, a hereditary deficiency of nearly every enzyme involved in glycolysis has been identified. A common result of these deficiencies is shortened RBC survival, known as hereditary nonspherocytic hemolytic anemia.

Transmembranous cation gradient concentration proteins, called cation pumps, provide a particular example of how energy failure, even in otherwise normal cells, leads to cell death. The pumps maintain high concentrations of intracellular potassium (K⁺), low concentrations of sodium (Na⁺), and very low concentrations of calcium (Ca²⁺) in contrast to extracellular concentrations of low K⁺ and high levels of Na⁺ and Ca²⁺. The pumps consume approximately 15% of RBC adenosine triphosphate (ATP) production as they prevent adverse accumulations of intracellular sodium and calcium. As energy-producing enzymes degrade over time, the pumps fail, Na⁺, Ca²⁺, and water flow in, and the RBC swells and is destroyed. This process is accelerated in hereditary nonspherocytic anemias in which the cells begin their life in the circulation with diminished enzyme concentrations.

Plasma glucose enters the RBC through a facilitated membrane transport system.⁵ Anaerobic glycolysis, the Embden-Meyerhof pathway (EMP; Figure 9-1), requires glucose to generate ATP, a high-energy phosphate source that is the greatest reservoir of energy in the RBC. Energy reserves like glycogen are not available in RBCs, so the RBCs rely mostly on external glucose for glycolysis-generated ATP. Via the EMP, glucose is catabolized to pyruvate, consuming two molecules of ATP per molecule of glucose and maximally generating four molecules of ATP per molecule of glucose, for a net gain of two molecules of ATP.

Figure 9-1 Glucose metabolism in the erythrocyte. *ADP,* adenosine diphosphate; *ATP,* adenosine triphosphate; *G6PD,* glucose-6-phosphate dehydrogenase; *NAD,* nicotinamide adenine dinucleotide; *NADH,* nicotinamide adenine dinucleotide (reduced form); *NADP,* nicotinamide adenine dinucleotide phosphate (oxidized form); *NADPH,* nicotinamide adenine dinucleotide phosphate (reduced form).

The sequential list of biochemical intermediates involved in glucose catabolism, with corresponding enzymes, is given in Figure 9-1. Tables 9-1 through 9-3 organize this information into three phases for better comprehension.

TABLE 9-1

Glucose Catabolism: First Phase

Substrates	Enzyme	Products
Glucose, ATP	Hexokinase	G6P, ADP
G6P	Glucose phosphate isomerase	F6P
F6P, ATP	Phosphofructokinase	F-1,6-P; ADP
F-1,6-P	Fructodiphosphate adolase	DHAP, G3P

ADP, Adenosine diphosphate; ATP, adenosine triphosphate; DHAP, dihydroxyacetone phosphate; F-1,6-P, fructose 1,6-bisphosphate; F6P, fructose 6-phosphate; G3P, glucose 3-phosphate; G6P, glucose 6-phosphate.

TABLE 9-2

Glucose Catabolism: Second Phase

Substrates	Enzyme	Product
G3P	Glyceraldehyde-3-phosphate dehydrogenase	1,3-BPG
1,3-BPG, ADP	Phosphoglycerate kinase	3-PG, ATP
1,3-BPG	Bisphosphoglyceromutase	2,3-BPG
2,3-BPG	Bisphosphoglycerate phosphatase	3-PG

1,3-BPG, 1,3 Bisphosphoglycerate; 2,3-BPG, 2,3 bisphosphoglycerate; 3-PG, 3-phosphoglycerate; ADP, adenosine diphosphate; ATP, adenosine triphosphate; G3P, glucose 3-phosphate.

TABLE 9-3

Glucose Catabolism: Third Phase

Substrates	Enzyme	Product
3-PG	Monophosphoglyceromutase	2-PG
2-PG	Phosphopyruvate hydratase (enolase)	PEP
PEP, ADP	Pyruvate kinase	Pyruvate, ATP

2-PG, 2-Phosphoglycerate; 3-PG, 3-phosphoglycerate; ADP, adenosine diphosphate; ATP, adenosine triphosphate; PEP, phosphoenolpyruvate.

The first phase of glucose catabolism involves glucose phosphorylation, isomerization, and diphosphorylation to yield fructose 1,6-bisphosphate (F-1,6-P). F-1,6-P serves as the substrate for aldolase cleavage for the final product of phase 1 glycolysis: glyceraldehyde 3-phosphate (G3P; see Figure 9-1 and Table 9-1). Hexokinase and phosphofructokinase are rate-limiting in steady-state anaerobic glycolysis, even though hexokinase has the lowest activity of all the glycolytic enzymes.

The second phase of glucose catabolism converts G3P to 3-phosphoglycerate (3-PG). The substrates, enzymes, and products for this phase of glycolytic metabolism are summarized in Table 9-2. During the first reaction step, G3P is phosphorylated with a high-energy phosphate and oxidized to 1,3-bisphosphoglycerate (1,3-BPG) through the action of glyceraldehyde-3-phosphate dehydrogenase (G3PD). 1,3-BPG is dephosphorylated by phosphoglycerate kinase, which generates ATP and 3-PG.

The third phase of anaerobic glucose catabolism converts 3-PG to pyruvate with the generation of ATP. Substrates, enzymes, and products are listed in Table 9-3. The product 3-PG is isomerized by phosphoglyceromutase to 2-phosphoglycerate (2-PG). Enolase then converts 2-PG to phosphoenolpyruvate (PEP). Pyruvate kinase (PK) splits off the phosphates, forming ATP and pyruvate. PK activity is allosterically modulated by increased concentrations of F-1,6-P, which enhances the affinity of PK for PEP.⁵ Thus when the F-1,6-P is plentiful, increased activity of PK favors pyruvate production. Pyruvic acid may diffuse from the erythrocyte or may become a substrate for lactate dehydrogenase with regeneration of the oxidixed form of nicotinamide adenine dinucleotide (NAD⁺). The ratio of NAD⁺ to the reduced form (NADH) may modify the activity of this enzyme.

Glycolysis Diversion Pathways (Shunts)

Without organelles, RBCs take advantage of diversions or shunts from the glycolytic pathway to generate products that maximize oxygen delivery from several vantage points. The three diversions are the hexose monophosphate pathway (HMP) or aerobic glycolysis, the methemoglobin reductase pathway, and the Rapoport-Luebering pathway.

Hexose Monophosphate Pathway

Aerobic or oxidative glycolysis occurs through a diversion of glucose catabolism into the HMP, also known as the pentose phosphate shunt (see Figure 9-1). The HMP detoxifies accumulated peroxide, an agent that oxidizes heme iron, proteins, and lipids, especially those containing thiol groups. Peroxide arises spontaneously from the reduction of oxygen in the cell’s aqueous environment.⁵ By detoxifying peroxide, the HMP extends the functional life span of the RBC.

The HMP diverts glucose 6-phosphate (G6P) to pentose phosphate by the action of glucose-6-phosphate dehydrogenase (G6PD). In the process, oxidized nicotinamide adenine dinucleotide phosphate (NADP⁺) is converted to the reduced form (NADPH). NADPH is then available to reduce the oxidized form of glutathione (GSSG) to its reduced form (GSH) using glutathione reductase. As the “thio” indicates, glutathione is a sulfur-containing compound and is, in fact, a tripeptide containing cysteine. The designation GSH highlights the sulfur moiety. Once glutathione is reduced, it is available to reduce peroxide to water and oxygen via glutathione peroxidase.

During steady-state glycolysis, 5% to 10% of G6P is diverted to the HMP. After oxidative challenge, the activity of the HMP may increase twentyfold to thirtyfold.⁶ The HMP catabolizes G6P to ribulose 5-phosphate and carbon dioxide by oxidizing G6P at carbon 1. The substrates, enzymes, and products of the HMP are listed in Table 9-4.

TABLE 9-4

Glucose Catabolism: Hexose Monophosphate Pathway

Substrates	Enzyme	Product
G6P	Glucose-6-phosphate dehydrogenase	6-PG
6-PG	Phosphogluconolactonase Phosphogluconate dehydrogenase

R5P

6-PG, 6-Phosphogluconate; G6P, glucose 6-phosphate; R5P, ribulose 5-phosphate.

G6PD provides the only means of generating NADPH for glutathione reduction, and in its absence erythrocytes are particularly vulnerable to oxidative damage.⁷ Normal G6PD activity is able to detoxify oxidative compounds and safeguard hemoglobin, sulfhydryl-containing enzymes, and membrane thiols, allowing normally functioning RBCs to carry enormous quantities of oxygen safely. However, G6PD deficiency is the most common human RBC enzyme deficiency worldwide resulting in hereditary nonspherocytic anemia (see Chapter 23).

Methemoglobin Reductase Pathway

Heme iron is constantly exposed to oxygen, an oxidizing agent.⁸ In addition, the accumulation of peroxide oxidizes heme iron from the ferrous to the ferric state, forming methemoglobin. Although the HMP is able to prevent some hemoglobin oxidation by reducing peroxide, it is not able to reduce methemoglobin once it forms. NADPH is able to do so, but only slowly. The reduction of methemoglobin by NADPH is far more efficient in the presence of methemoglobin reductase, more properly called cytochrome b₅ reductase (cytob5r). Using H⁺ from NADH formed when G3P is converted to 1,3-BPG, soluble cytochrome b₅ reductase acts as an intermediate electron carrier, swiftly returning the oxidized iron to its ferrous, oxygen-carrying state. Cytochrome b₅ reductase accounts for more than 65% of the methemoglobin-reducing capacity within the RBC.⁸

Rapoport-Luebering Pathway

A third metabolic shunt generates 2,3-bisphosphoglycerate (2,3-BPG; also called 2,3-diphosphoglycerate or 2,3-DPG). 1,3-BPG is diverted by bisphosphoglycerate mutase to form 2,3-BPG. 2,3-BPG regulates oxygen delivery to tissues by essentially competing with oxygen for hemoglobin. When 2,3-BPG binds, oxygen is released, which enhances delivery of oxygen to the tissues.

2,3-BPG forms 3-PG by the action of bisphosphoglycerate phosphatase. This diversion of 1,3-BPG to form 2,3-BPG immediately sacrifices the production of two ATP molecules via glycolysis. However, there is the loss of another two molecules of ATP at the level of PK, because fewer molecules of PEP are formed. Because two ATP molecules were used to generate 1,3-BPG and production of 2,3-BPG eliminates the production of four molecules, the cell is put in ATP deficit by this diversion. The cell maintains a delicate balance between ATP generation to support the energy requirements of cell metabolism and the need to maintain the appropriate oxygenation/deoxygenation status of hemoglobin. Acidic pH and low concentrations of 3-PG and 2-PG inhibit the activity of bisphosphoglyceromutase, thus inhibiting the shunt and retaining 1,3-BPG in the EMP. These conditions and decreased ATP activate bisphosphoglycerate phosphatase, which returns 2,3-BPG to the mainstream of glycolysis. In summary, these conditions favor generation of ATP by causing the conversion of more 1,3-BPG directly to 3-PG and also returning some 2,3 BPG to 3-PG for ATP generation downstream by PK.

Rbc Membrane

RBC Deformability

RBCs are biconcave and average 90 fL in volume.⁹ Their average surface area is 140 µm², a 40% excess of surface area compared with a 90-fL sphere. This excess surface-to-volume ratio enables RBCs to stretch undamaged up to 2.5 times their resting diameter as they pass through narrow capillaries and through splenic pores 2 µm in diameter, a property called RBC deformability.¹⁰ The RBC plasma membrane, which is 5 µm thick, is 100 times more elastic than a comparable latex membrane, yet has tensile (lateral) strength greater than that of steel. The deformable RBC membrane provides the broad surface area and close tissue contact necessary to support the delivery of oxygen from lungs to body tissue and carbon dioxide from body tissue to lungs.

RBC deformability depends not only on RBC geometry but also on relative cytoplasmic (hemoglobin) viscosity. The normal mean cell hemoglobin concentration (MCHC) ranges from 32% to 36% (see Chapter 14 and inside front cover), and as hemoglobin concentration rises, viscosity rises.¹¹ Hemoglobin concentrations above 36% compromise deformability and shorten the RBC life span, because the more viscous cells cannot accommodate to narrow capillaries or splenic pores. As RBCs age, they lose membrane surface area while retaining hemoglobin. The hemoglobin becomes more and more concentrated, and eventually the RBC, unable to pass through the splenic pores, is destroyed by splenic macrophages. Refer to Chapter 8 for a more complete discussion of RBC senescence theories.

RBC Membrane Lipids

Membrane elasticity (pliancy) is the third property that contributes to deformability. The RBC membrane consists of approximately 8% carbohydrates, 52% proteins, and 40% lipids.¹² The lipid portion, equal parts of cholesterol and phospholipids, forms a bilayer universal to all animal cells (see Figure 13-9). Four main phospholipids form an impenetrable fluid barrier as their hydrophilic polar head groups are arrayed upon the membrane’s surfaces, oriented toward both the aqueous plasma and the cytoplasm, respectively.¹³ Their hydrophobic nonpolar acyl tails arrange themselves to form a central layer dynamically sequestered from the aqueous plasma and cytoplasm. The membrane maintains extreme differences in osmotic pressure, cation concentrations, and gas concentrations between the plasma external to the cell and the internal cytoplasm.¹⁴ Phospholipids reseal rapidly when the membrane is torn.

Cholesterol, esterified and largely hydrophobic, resides parallel to the acyl tails of the phospholipids, equally distributed between the outer and inner layers, and evenly dispersed within each layer, approximately one molecule per phospholipid molecule. Cholesterol’s β hydroxyl group, the only hydrophilic portion of the molecule, anchors within the polar head groups, while the rest of the molecule intercalates among and parallel to the acyl tails. Cholesterol confers tensile strength to the lipid bilayer.¹⁵

The ratio of cholesterol to phospholipids remains relatively constant and balances the need for deformability and strength. Membrane enzymes maintain the cholesterol concentration by regularly exchanging membrane and plasma cholesterol. Deficiencies in these enzymes are associated with membrane abnormalities such as acanthocytosis as the membrane loses tensile strength. Conversely, as cholesterol concentration rises, the membrane gains strength but loses elasticity.

The phospholipids are asymmetrically distributed. Phosphatidylcholine and sphingomyelin predominate in the outer layer; phosphatidylserine (PS) and phosphatidylethanolamine form most of the inner layer. Distribution of the phospholipids is energy dependent, relying on a number of membrane-associated enzymes, called flippases, floppases, and scramblases, for their position.¹⁶ When phospholipid distribution is disrupted, as in sickle cell anemia and thalassemia or in RBCs that have reached the end of their 120-day life span, PS, the only negatively charged phospholipid, redistributes (flips) to the outer layer. Splenic macrophages possess receptors that bind PS and destroy senescent RBCs. Refer to Chapter 8 for a complete discussion of RBC senescence theories.

Membrane phospholipids and cholesterol may also redistribute laterally so that the RBC membrane may respond to stresses and deform within 100 milliseconds of being challenged by the presence of a narrow passage, such as when arriving at a capillary. Redistribution becomes limited as the proportion of cholesterol increases. Plasma bile salt concentration also affects cholesterol exchange.¹⁷ In liver disease with low bile salt concentration, membrane cholesterol concentration becomes reduced. As a result, the more elastic cell membrane shows a “target cell” appearance when the RBCs are spread on a slide (see Figure 18-1).

Glycolipids (sugar-bearing lipids) make up 5% of the external half of the RBC membrane.¹⁸ They associate in clumps or rafts and support carbohydrate side chains that extend into the aqueous plasma to help form the glycocalyx. The glycocalyx is a layer of carbohydrates whose net negative charge prevents microbial attack and protects the RBC from mechanical damage caused by adhesion to neighboring RBCs or to the endothelium. Glycolipids may bear a few copies of carbohydrate-based blood group antigens, for example, antigens of the ABH and the Lewis blood group systems.

RBC Membrane Proteins

Although cholesterol and phospholipids constitute the principal RBC membrane structure, transmembranous (integral) and skeletal (cytoskeletal, peripheral) proteins make up 52% of the membrane structure by mass.¹⁹ A proteomic study reveals there are at least 300 RBC membrane proteins, including 105 transmembranous proteins. Of the purported 300 membrane proteins, about 50 have been characterized and named, some with a few hundred copies per cell, others with over a million copies per cell.²⁰

Transmembranous Proteins

The transmembranous proteins serve a number of RBC functions, as listed in Table 9-5.²¹ Through glycosylation they support surface carbohydrates, which join with glycolipids to make up the protective glycocalyx.²² They serve as transport and adhesion sites and signaling receptors. Any disruption in transport protein function changes the osmotic tension of the cytoplasm, which leads to a rise in viscosity and loss of deformability. Any change affecting adhesion proteins permits RBCs to adhere to each other and to the vessel walls, promoting fragmentation (vesiculation), reducing membrane flexibility, and shortening the RBC life span. Signaling receptors bind plasma ligands and trigger energy activation of submembranous G proteins that then initiate various energy-dependent cellular activities, a process called signal transduction.

TABLE 9-5

Names and Properties of Selected Transmembranous RBC Proteins

Transmembranous Protein	Band	Molecular Weight (D)	Copies per Cell (×10³)	% of Total Protein	Function
Aquaporin 1					Water transporter
Band 3 (anion exchanger 1)	3	90,000-102,000	1200	27%	Anion transporter, supports ABH antigens
Ca²⁺ ATPase					Ca²⁺ transporter
Duffy		35,000-43,000			G protein–coupled receptor, supports Duffy antigens
Glut-1	4.5	45,000-75,000		5%	Glucose transporter, supports ABH antigens
Glycophorin A	PAS-1	36,000	500-1000	85% of glycophorins	Transports negatively charged sialic acid, supports MN determinants
Glycophorin B	PAS-4	20,000	100-300	10% of glycophorins	Transports negatively charged sialic acid, supports Ss determinants
Glycophorin C	PAS-2	14,000-32,000	50-100	4% of glycophorins	Transport negatively charged sialic acid, supports Gerbich system determinants
ICAM-4					Integrin adhesion
K⁺-Cl^– cotransporter
Kell		93,000			Zn²⁺-binding endopeptidase, Kell antigens
Kidd		46,000-50,000			Urea transporter
Na⁺,K⁺-ATPase
Na⁺-K⁺-2Cl^– cotransporter
Na⁺-Cl^– cotransporter
Na⁺-K⁺ cotransporter
Rh		30,000-45,500			D and CcEe antigens
RhAG		50,000			Necessary for expression of D and CcEe antigens; gas transporter, probably of CO₂

ATPase, Adenosine triphosphatase; Duffy, Duffy blood group system protein; ICAM, intracellular adhesion molecule; Kell, Kell blood group system protein; PAS, periodic acid–Schiff dye; RBC, red blood cell; Rh, Rh blood group system protein; RhAG, Rh antigen expression protein.

The transmembranous proteins assemble into one of two macromolecular complexes named by their respective skeletal anchorages, ankyrin or protein 4.1 (Figure 9-2). These complexes and their anchorages provide RBC membrane structural integrity, because the membrane relies on the cytoplasmic skeletal proteins positioned immediately within (underneath) the lipid bilayer for its ability to retain (and return to) its biconcave shape despite deformability. The transmembranous proteins provide vertical membrane structure.

Figure 9-2 Representation of the human red blood cell membrane. The transmembranous proteins assemble in one of two complexes defined by their anchorage to skeletal proteins ankyrin and 4.1. Band 3 is the most abundant transmembranous protein. In the ankyrin complex band 3 and protein 4.2 anchor to ankyrin, which is bound to the spectrin backbone. In the 4.1 complex, band 3, Rh, and other transmembranous proteins bind the complex of dematin, adducin, actin, tropomyosin, and tropomodulin through protein 4.1. *CD47,* signaling receptor; *Duffy,* Duffy blood group system protein; *GPA,* glycophorin A; *GPC,* glycophorin C; *Kell,* Kell blood group system protein; *LW,* Landsteiner-Weiner blood group system protein; *Rh,* Rh blood group system protein; *RhAG,* Rh antigen expression protein; *XK,* X-linked Kell antigen expression protein.

Transmembranous proteins support carbohydrate-defined blood group antigens.²³ For instance, band 3 (anion transport) and Glut-1 (glucose transport) support the majority of ABH system carbohydrate determinants by virtue of their high copy numbers.²⁴ ABH system determinants are also found on several low copy number transmembranous proteins. Certain transmembranous proteins provide peptide epitopes. For instance, glycophorin A carries the peptide-defined M and N determinants, and glycophorin B carries the Ss determinants, which together comprise the MNSs system.^25,26

The Rh system employs two multipass transmembranous lipoproteins and a multipass glycoprotein that each cross the membrane 12 times.27–30 The two lipoproteins present the D and CcEe epitopes, respectively, but expression of the D and CcEe antigens requires the separately inherited glycoprotein RhAG, which localizes near the Rh lipoproteins in the ankyrin complex. Loss of the RhAG glycoprotein prevents expression of both the D and CcEd antigens (Rh-null) and is associated with morphologic RBC abnormalities. Additional blood group antigens localize to the 4.1 complex or specialized proteins.^28–30

A few copies of phosphatidylinositol (PI), not mentioned in the RBC Membrane Lipids section, reside in the outer, plasma-side portion of the lipid bilayer. These serve as anchors for two surface proteins, decay-accelerating factor (DAF, or CD55) and membrane inhibitor of reactive lysis (MIRL, or CD59).31,32 These proteins appear to float on the surface of the membrane as they link to PI through a glycan core consisting of multiple sugars. CD55 and CD59 are linked to PI by phosphatidylinositol glycan class A (PIG-A). In paroxysmal nocturnal hemoglobinuria (see Chapter 23), PIG-A acquires a mutation, CD55 and CD59 become deficient, and the cell is susceptible to complement-mediated hemolysis.

Numerical naming—for instance, band 3, protein 4.1, and protein 4.2—derives from historical (preproteomics) protein identification techniques that distinguished 15 membrane proteins using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), as illustrated in Figure 9-3.³³ Bands migrate through the gel, with their velocity a property of their molecular weight and net charge, and are identified using Coomassie blue dye. The glycophorins, with abundant carbohydrate side chains, are stained using periodic acid–Schiff (PAS) dye.

Figure 9-3 Sodium dodecyl sulfate-polyacrylamide gel electrophoresis of RBC membrane proteins stained with Coomassie blue bye. Lane A illustrates numerical band naming based on migration. Lane B names and illustrates the positions of some of the major proteins. (Adapted from Costa FF, Agre P, Watkins PC, et al: Linkage of dominant hereditary spherocytosis to the gene for the erythrocyte membrane-skeleton protein ankyrin, *N Engl J Med* 323:1046, 1990.)

Band 3, protein 4.2, and RhAG, members of the ankyrin complex, link their associated proteins and the bilayer membrane to the skeletal proteins through ankyrin.34–36 Likewise, glycophorin C, Rh, and blood group Duffy link the 4.1 complex through protein 4.1.³⁷ The 4.1 anchorage also includes the more recently defined proteins adducin and dematin, which link with band 3 and Glut-1, respectively.³⁸

Skeletal Proteins

The principal skeletal proteins are the filamentous α-spectrin and β-spectrin (Table 9-6), which assemble to form an antiparallel heterodimer held together with a series of lateral bonds.³⁹ Antiparallel means that the carboxyl (COOH) end of one strand associates with the amino (NH₃) end of the other. The spectrins form a hexagonal lattice (Figure 9-4) that is immediately adjacent to the cytoplasmic membrane lipid layer and provides lateral membrane stability.⁴⁰ Because the skeletal proteins do not penetrate the bilayer, they are also called peripheral proteins.

TABLE 9-6

Names and Properties of Selected Skeletal RBC Proteins

Skeletal Protein	Band	Molecular Weight (D)	Copies per Cell (×10³)	% of Total Protein	Function
α-spectrin	1	240,000-280,000	240	16%	Filamentous antiparallel heterodimer, primary cell skeleton
β-spectrin	2	220,000-246,000	240	14%	Filamentous antiparallel heterodimer, primary cell skeleton
Adducin	2.9	80,000-103,000	60	4%	Caps actin filament
Ankyrin	2.1	206,000-210,000	120	4.5%	Anchors band 3 and protein 4.2
Dematin	4.9	43,000-52,000	40	1%	Actin bundling protein
F-actin	5	42,000-43,000	400-500	5.5%	Binds β-spectrin
G3PD	6	35,000-37,000	500	3.5%
Protein 4.1	4.1	66,000-80,000	200	5%	Anchors 4.1R complex
Protein 4.2 (protein kinase)	4.2	72,000-77,000	200	5%	Anchors ankyrin complex
Tropomodulin	5	41,000-43,000	30	—	Caps actin filament
Tropomyosin	7	27,000-38,000	80	1%	Regulates actin polymerization

G3PD, Glucose-3-phosphate dehydrogenase glyceraldehyde; RBC, red blood cell.

Figure 9-4 Spectrin-based cytoskeleton on the cytoplasmic side of the human red blood cell membrane. A, Junctional complex composed of actin filaments containing 14 to 16 actin monomers, band 4.1, adducin, and tropomyosin, which probably determines the length of the actin filaments. B, Spectrin dimers form a lattice that binds band 3 and protein 4.2 (not shown) via ankyrin and band 3, Glut-1 and Duffy (not shown), and glycophorin via protein 4.1.

The secondary structure of both α- and β-spectrin features triple-helical repeats of 106 amino acids each; 20 such repeats make up α-spectrin and 16 make up β-spectrin.⁴¹ Essential to the cytoskeleton are the previously mentioned ankyrin, protein 4.1, adducin and dematin, and, in addition, actin, tropomyosin, and tropomodulin (see Figure 9-4).³⁵ A single helix at the amino terminus of α-spectrin consistently binds a pair of helices at the carboxy terminus of the β-spectrin chain, forming a stable triple helix that holds together the ends of the heterodimers.⁴² Joining these ends are actin and protein 4.1. Actin forms short filaments of 14 to 16 monomers whose length is regulated by tropomyosin. Adducin and tropomodulin cap the ends of actin, and dematin appears to stabilize actin in a manner that is the subject of current investigation.⁴³

Spectrin dimer bonds that appear along the length of the molecules disassociate and reassociate (open and close) during RBC deformation.⁴⁴ Likewise, the 20 α-spectrin and 16 β-spectrin repeated helices unfold and refold. These flexible interactions plus the spectrin–actin–protein 4.1 junctions are key regulators of membrane elasticity and mechanical stability, and abnormalities in any of these proteins result in deformation-induced membrane fragmentation. For instance, autosomal dominant mutations that affect the integrity of band 3, RhAG, ankyrin, protein 4.1, or spectrin are associated with hereditary spherocytosis (see Chapter 23).^45,46 In these cases there are too few vertical anchorages to maintain membrane stability. The lipid membrane peels off in small blebs called vesicles, whereas the cytoplasmic volume remains intact. This generates spherocytes with a reduced membrane-to-cytoplasm ratio. Conversely, hereditary elliptocytosis (ovalocytosis) arises from one of several autosomal dominant mutations affecting spectrin dimer-to-dimer lateral bonds or the spectrin–ankyrin–protein 4.1 junction.⁴⁷ In hereditary elliptocytosis, the membrane fails to rebound from deformation, and RBCs progressively elongate to form visible elliptocytes, which causes a mild to severe hemolytic anemia.⁴⁸

Osmotic Balance and Permeability

The RBC membrane is impermeable to Na⁺, K⁺, and Ca²⁺. It is permeable to water and the anions bicarbonate (HCO₃^–) and chloride (Cl^–), which freely exchange between plasma and RBC cytoplasm.⁴⁹ Aquaporin 1 is a transmembranous protein that forms pores or channels whose surface charges create inward flow of water in response to internal osmotic changes. Glucose is transported without energy expenditure by the transmembranous protein Glut-1.

Adenosine triphosphatase (ATPase)–dependent (energy-dependent) cation pumps (see Table 9-5) regulate the concentrations of Na⁺ and K⁺, maintaining intracellular-to-extracellular ratios of 1:12 and 25:1, respectively.^50,51 These enzyme-based pump mechanisms, in addition to aquaporin, maintain osmotic balance. Pump mechanism damage permits the influx of Na⁺, with water following osmotically. The cell swells, becomes spheroid, and eventually ruptures, spilling hemoglobin. This phenomenon is called colloid osmotic hemolysis.

Ca²⁺ ATPase extrudes calcium, maintaining exceptionally low intracellular levels of 5 to 10 µmol/L. Calmodulin, a cytoplasmic Ca²⁺-binding protein, controls the function of Ca²⁺ ATPase.⁵²

Sickle cell disease provides an example of increased cation permeability. When crystallized sickle hemoglobin deforms the cells, internal levels of Na⁺, K⁺, and especially Ca²⁺ rise, which results in hemolysis.⁵³