Pathobiology of the Human Erythrocyte and Its Hemoglobins

Essential Features of Red Blood Cell Homeostasis

As discussed in Chapter 27 , the mature RBC is the product of a complex and orderly set of differentiation and maturation steps beginning with the pluripotent stem cell. By complex partially understood mechanisms involving hierarchic networks of cytokines, a portion of these cells becomes committed to differentiate along the erythroid pathway. Commitment to erythropoiesis provokes a progressively increasing sensitivity to the stimulatory actions of the hormone erythropoietin (see Chapters 9 and 27 ). As differentiation proceeds, there is preprogramming of certain genes whose expression at high levels will be required during the maturation phase of erythropoiesis. Genes coding for molecules defining the RBC phenotype (e.g., globin) are poised for activation at later maturation steps.

Intermediate progenitor cells arising during differentiation have been characterized experimentally, including the burst-forming unit-erythroid (BFU-E) and the colony-forming unit-erythroid (CFU-E) stages. BFU-Es are progenitor cells that in culture produce bursts or clusters of erythroid colonies, are relatively less sensitive to erythropoietin, and are more plastic with respect to important gene expression parameters, such as the synthesis of adult or fetal Hb (HbF) by their descendants. CFU-Es produce single colonies, exhibit considerably higher sensitivity to erythropoietin, and appear to be more fixed in their potential to express a particular subset of globin genes. CFU-Es give rise to the first morphologically recognizable erythroid cells, the proerythroblasts. At this “primitive” morphologic stage, the program of erythroid cell expression has already been essentially predetermined. The cell is predestined to undergo only a limited additional number of cell divisions, culminating in the formation of the enucleate reticulocyte. The terminal maturation stages are morphologically recognizable as erythroblasts exhibiting progressive hemoglobinization of the cytoplasm, condensation, and eventual ejection of the nucleus and remodeling of the plasma membrane. The actual expression of the preprogrammed genes occurs during the 5- to 7-day period of erythroblast maturation.

As discussed in Chapters 9 and 27 , the actual reconfiguration of chromatin for activation of the genes and activation itself appears to require the concerted and complex interaction of a diverse but limited group of transcription factors and associated epigenetic regulators. These regulatory proteins recognize a specific array of promoter and enhancer sequences that are embedded as recurrent motifs in and around the appropriate target genes. Even though an enormous amount of information has been gathered about sequences such as the GATA enhancers and their cognate transcription factors (e.g., GATA, FOG, ETS), the precise means by which these sequences and factors cause erythroid differentiation remains under investigation. (The current state of understanding is outlined in Chapter 27 .) At this time, this information is of limited clinical relevance to anemias or polycythemias. Major changes in the balance of post transcriptional regulatory molecules such as pre-mRNA splicing factors, mRNA stabilization factors, and translation factors also occur during maturation. Some of these, notably certain splicing factors, are proving to be relevant to myelodysplastic and myeloproliferative disorders (see Chapters 61 and 70 ). The orderly 14- to 21-day sequence of differentiation and maturation becomes progressively influenced by the levels of erythropoietin available to the progenitor cells, possibly because of increasing density and affinity of erythropoietin receptors on their cell surfaces. Within 24 hours after enucleation, the reticulocyte traverses the bone marrow–blood barrier membrane and enters the circulation as an immature erythrocyte. These cells retain remnants of nucleated precursors in the form of a relatively small number of polyribosomes actively translating messenger ribonucleic acid (RNA) (greater than 90% of which is globin messenger RNA), a cell membrane that retains some molecules and structures reminiscent of its earlier stages of differentiation, and the complement of enzymes, phospholipids, and cytoskeletal proteins that the cell will possess throughout its remaining life span.

During its first 24 hours in the circulation, the reticulocyte spends considerable amounts of time in the spleen, during which its membrane is “polished.” This is a poorly understood remodeling process by which some lipids and proteins, including adhesive molecules such as fibronectin, are removed. The content of polyribosomes and other nucleic acids progressively declines so that stainability with methylene blue is lost by the end of the first day. At this time, the RBC is regarded as a mature erythrocyte, and it circulates relatively unchanged for the remainder of its 120-day life span.

Perhaps the most remarkable feature of the human RBC is its durability, given that it is an enucleated cell devoid of organelles that are critical for the survival and function of most other cell types. The RBC has no mitochondria available for efficient oxidative metabolism; no ribosomes for regeneration of lost or damaged proteins; a very limited metabolic repertoire that largely precludes de novo synthesis of lipids; and no nucleus to direct regenerative processes, adaptation to circulatory stresses, or cell division to replenish itself. Given these handicaps, the 120-day survival of these cells is even more striking considering the multiple and often exceedingly hostile environments they must traverse. Mechanical stresses of the circulation include high hydrostatic pressure and turbulence and the shear stresses inherent in a microcirculation networked with many capillaries having diameters only one-third to one-half that of the normal RBC. Biochemical stresses include osmotic and redox fluxes associated with travel through the collecting system of the kidney; the sluggish vascular beds of the spleen, muscle, and bone; and the rapid changes in ambient oxygen pressures occurring in the lungs. All conspire to damage RBCs. Their 4-month survival is truly remarkable.

The ability of the RBC to persist in the circulation depends on its simple but exquisitely adaptive membrane structures; its pathways of intermediary energy metabolism and redox regulation; and its ability to maintain its largest cytoplasmic component, Hb, in a soluble and non-oxidized state. The membrane and enzymes of the RBC appear to be exquisitely crafted to protect the cell from the external ravages of the circulation and the potential internal assaults of the massive amount of iron-rich and potentially oxidizing protein represented by its complement of Hb molecules. For these reasons, a few basic features of these membrane and enzyme systems merit comment before considering the Hb molecule itself.

Major Features of the Red Blood Cell Membrane

Chapter 46 describes the RBC membrane in considerable detail. Only a few major aspects of that discussion bear repeating for the purposes of this chapter. The RBC membrane and its underlying cytoskeleton have evolved to provide mechanical strength and the necessary pliability and resilience to withstand the mechanical, osmotic, and chemical stresses of the circulation. Because a lipid bilayer membrane essentially has the physical properties of a soap bubble, it would rapidly be emulsified in the circulation. Strength and order are provided to the lipid bilayer by the hexagonal arrays of the highly helical protein spectrin, which forms a latticework underlying the membrane.

The spectrin meshwork is held together by adaptor molecules, such as protein 4.1, adducin, p55, and ankyrin, arrayed at defined points along the highly coiled, rod-like structure of the spectrin oligomers (see figures in Chapter 46 ). These protein–protein interactions appear to be critical for holding the latticework together in what has been described as the “horizontal” dimension that permits resistance to shear stress. The involvement of intermediate-length actin fibers and the variability of binding affinities governed by phosphorylation and other biochemical alterations appear to provide some flexibility and pliability at these points of interaction. Strength in the “vertical” dimension is provided by additional molecules or additional binding functions of the same molecule, whereby the latticework is attached to the lipid bilayer. For the most part, the physiologically important attachments appear to be indirect. Linkage is mediated through the interaction of the adaptor proteins, such as ankyrin and protein 4.1, with both the cytoplasmic domains of abundant transmembrane proteins and specific domains within the spectrin molecule. These proteins traverse and are embedded in the lipid bilayer, providing a firm anchor. The two most critical of these molecules appear to be band 3 (i.e., the anion transport channel) and a glycophorin, probably glycophorin C/D. A possible additional stabilizing role for the Rh protein complex has been suggested. The construction of these attachments by multiple “hinge” or coupling molecules appears to provide for the flexibility and distensibility of the RBC membrane, a property essential to its ability to flow through small capillaries.

As described in Chapter 46 , the complex structure of the membrane is exquisitely sensitive to perturbations impinging on any of its components. In particular, the membrane cytoskeleton and phospholipid structures are each highly susceptible to oxidation, particularly by partially proteolyzed molecules of Hb, which denature to form highly toxic compounds called hemopyrroles . This interaction of denatured Hb with the RBC membrane is clinically important, as illustrated by its impact on the pathophysiology of sickle cell anemia (see Chapters 42 and 43 ) or of oxidized and precipitated globin inclusion bodies in thalassemia (see Chapter 41 ) and unstable Hb disorders (see Chapter 44 ). In this chapter, it is sufficient to note that alterations of the proteins of the RBC membrane/cytoskeleton complex can contribute to shortening the life span of the RBC. Damage can result from intrinsic defects in the cytoskeletal proteins themselves or from the susceptibility of these proteins to direct oxidation or attack by oxidized or denatured Hb molecules. Readers are referred to Chapter 44, Chapter 45, Chapter 46, Chapter 47, Chapter 48 for detailed descriptions of the relevant phenomena.

Enzymes of Red Blood Cell Intermediary Metabolism

Mammalian erythrocytes possess a highly specialized but remarkably simplified set of metabolic pathways. As discussed in Chapter 45 , there are essentially three relevant sets of pathways. The first two are interconnected by the enzyme glucose-6-phosphate dehydrogenase (G6PD). Glucose entering the RBC is metabolized by an anaerobic pathway, the Embden–Meyerhof pathway, which terminates with the enzyme lactic dehydrogenase, forming lactate. Despite its inefficiency (a net of only two adenosine triphosphate [ATP]/glucose molecules generated per glucose molecule consumed), this pathway is the sole source of usable ATP in the cell. Moreover, the pathway generates reduced nicotinamide adenine dinucleotide (NADH), a molecule necessary for driving the reduction of methemoglobin to Hb (see Chapter 45, Chapter 48 ). A shunt within this pathway, the Rapoport–Luebering shunt, generates the compound 2,3-bisphosphoglycerate (bis[phosphoglyceric acid]) (2,3-BPG), an important cofactor that, when bound to Hb, reduces the affinity of Hb for oxygen (see Hemoglobin Function). The ATP generated is necessary for kinase reactions controlling phosphorylation of membrane and signaling components, for fueling ion pumps and channels, and for maintaining phospholipid levels.

The anaerobic metabolic pathway generates, as one of its intermediates, glucose-6 phosphate, which is the substrate for G6PD. G6PD appears to be the rate-limiting enzyme for a linked pathway called the oxidative hexose monophosphate shunt . This pathway involves a cascade of reactions culminating in the reduction of oxidized glutathione to reduced glutathione. Reduced glutathione is used to reverse the oxidation of critical structures, including Hb, cytoskeletal proteins, and membrane lipids. Anaerobic glycolysis generates NADH for methemoglobin reduction, 2,3-BPG for modulation of Hb–oxygen affinity, and ATP for metabolic energy requirements. Its end product is lactate. The oxidative hexose monophosphate shunt generates NADH phosphate (NADPH) and reduced glutathione for use as the major erythrocyte antioxidant.

During the past decade, most of the enzymes (or at least the erythroid isoforms of these enzymes) involved in RBC intermediary metabolism have been characterized at the molecular level by cloning of their cDNAs, genomic loci, or both. Some of the more relevant information arising from this progress is discussed in Chapter 45 . The erythrocyte possesses membrane-based signaling receptors and cytoplasmic signal transduction elements similar, although perhaps less elaborate, than those of nucleated cells. The relevance of these systems to the pathophysiology of RBC disorders is just becoming apparent.

Red Blood Cell Senescence and Destruction

Erythrocytes, despite their impressive adaptations to circulatory stresses, eventually wear out and are destroyed. RBC survival in humans appears to be remarkably uniform under normal circumstances, spanning approximately 120 days from release of the reticulocyte into circulation to sequestration of the senescent RBC in the reticuloendothelial cells of the liver and spleen. The precise signal, or signals, marking RBCs for destruction remain incompletely understood, as does the underlying pathophysiology within the RBC or on its surface. However, several interrelated theories have emerged; these are discussed only briefly because they are mentioned in other chapters.

RBCs accumulate surface blemishes during their lives in circulation. These appear to result in part from the accumulation of small amounts of oxygen damage to membrane structures. The altered regions are sensed by the reticuloendothelial cells during passage of the erythrocytes through the liver and spleen. Removal or pitting of these damaged regions from RBC membranes can be documented microscopically; small amounts of the normal membrane are also lost during the process.

The biconcave disk shape of the RBC, so important to its distensibility, depends on a high ratio of surface area to volume. This requires a redundant membrane surface area. The membrane surface area of the normal biconcave disk is approximately 140 μm ² . To enclose a sphere containing a normal RBC volume (≈90 fL), only approximately 95 μm ² would be needed. Progressive loss of membrane surface by means of the pitting phenomenon ultimately causes the aging erythrocyte to assume a more rigid spherical shape. A sphere is inevitably far less distensible and far less capable of “squeezing” through small apertures than a disk, especially in the sluggish and tortuous circulation of the spleen. The entrapped RBCs become engulfed by macrophages and catabolized. This geometric mechanism contributes to the eventual destruction of RBCs.

RBCs progressively lose some of the critical enzymes needed for intermediary metabolism and antioxidant capacity. G6PD levels, for example, progressively decline during the circulating life span, as do levels of several other enzymes. The decline of certain enzymes can be used as a crude means of estimating the relative age of different RBC populations. The biochemical or oxidative mechanism of destruction postulates that aged RBCs are eventually depleted of critical enzymes needed for the maintenance of redox status. Oxidation of critical membrane proteins, lipids, and Hb would then ensue, causing distortion and rigidity of the RBC membrane, with accelerated loss as previously described. The end product would be spherocytes incapable of traversing the splenic vascular bed and escaping engulfment by the reticuloendothelial cell.

It has been proposed that an immune mechanism can contribute to normal and pathologic RBC senescence. This hypothesis is based on the observation that oxidative damage, regardless of cause, promotes a clustering, or capping , of oligomers of band 3 on the RBC surface. Under normal circumstances, band 3 molecules form monomers, dimers, or tetramers. Higher-order aggregates appear to be recognized by an endogenous isoantibody possessed by most people. Any RBC accumulating oxidative damage from wear and tear in the circulation, from the depletion of enzymes, or from internal pathologic processes such as denaturation of Hb in certain hemoglobinopathies can accumulate these aggregates. The aggregates would then be bound by antibodies and be removed by the reticuloendothelial cells as antigen–antibody complexes using the same means used by reticuloendothelial cells to recognize any immune complex. This mechanism could also provide for the pitting or polishing of damaged RBC membranes. All three of the proposed mechanisms are interrelated by their inception with oxidative damage.

Other membrane-related changes might influence RBC destruction. Bcl-X _L , a suppressor of apoptosis, is present in erythrocyte membranes, and its antagonization may promote cell death. This may be mediated by calcium accumulation and phosphatidylserine exposure. Cholesterol and fatty acids accumulate on the aging RBC membrane and might be targets for oxidation induced by reactive oxygen species.

RBCs are removed from the circulation by splenic macrophages, probably by several mechanisms. SHPS-1, a surface glycoprotein and a member of the immunoglobulin superfamily that interacts with RBC membrane CD47, is abundant in macrophages. Studies using mice expressing a mutant SHPS-1 suggested that this molecule might negatively regulate phagocytosis, influencing cell life span. Increasing phosphatidylserine exposure and reduced aminophospholipid translocase activity during aging might induce oxidative damage to the cell. It is probable that these mechanisms leading to cell destruction are not mutually exclusive, that no single effect predominates, and that these events occur at different times at different sites of RBC damage.

Regardless of the mechanism(s) fostering eventual senescence and destruction of RBCs, the process itself involves components clinically useful for the assessment of anemias associated with accelerated destruction. Chief among these is the generation of indirect or unconjugated bilirubin, the byproduct of heme catabolism occurring within the reticuloendothelial cells. In markedly accelerated states of RBC destruction, hypertrophy of the liver and spleen can also occur, providing a useful physical indicator of hemolytic anemia. These indirect clinical features, coupled with the reticulocyte count remain more useful for detecting clinical hemolysis than complicated studies of RBC kinetics.