After studying this chapter, you should be able to:
Understand the significance of immunity, particularly with respect to defending the body against microbial invaders.
Define the circulating and tissue cell types that contribute to immune and inflammatory responses.
Describe how phagocytes are able to kill internalized bacteria.
Identify the functions of hematopoietic growth factors, cytokines, and chemokines.
Delineate the roles and mechanisms of innate, acquired, humoral, and cellular immunity.
Understand the basis of inflammatory responses and wound healing.
As an open system, the body is continuously called upon to defend itself from potentially harmful invaders such as bacteria, viruses, and other microbes. This is accomplished by the immune system, which is subdivided into innate and adaptive (or acquired) branches. The immune system is composed of specialized effector cells that sense and respond to foreign antigens and other molecular patterns not found in human tissues. Likewise, the immune system clears the body’s own cells that have become senescent or abnormal, such as cancer cells. Finally, normal host tissues occasionally become the subject of inappropriate immune attack, such as in autoimmune diseases or in settings where normal cells are harmed as innocent bystanders when the immune system mounts an inflammatory response to an invader. It is beyond the scope of this volume to provide a full treatment of all aspects of modern immunology. Nevertheless, the student of physiology should have a working knowledge of immune functions and their regulation, due to a growing appreciation for the ways in which the immune system can contribute to normal physiological regulation in a variety of tissues, as well as contributions of immune effectors to pathophysiology.
IMMUNE EFFECTOR CELLS
Many immune effector cells circulate in the blood as the white blood cells. In addition, the blood is the conduit for the precursor cells that eventually develop into the immune cells of the tissues. The circulating immunologic cells include granulocytes (polymorphonuclear leukocytes, PMNs), comprising neutrophils, eosinophils, and basophils; lymphocytes; and monocytes. Immune responses in the tissues are further amplified by these cells following their extravascular migration, as well as tissue macrophages (derived from monocytes) and mast cells (related to basophils). Acting together, these cells provide the body with powerful defenses against tumors and viral, bacterial, and parasitic infections.
All granulocytes have cytoplasmic granules that contain biologically active substances involved in inflammatory and allergic reactions.
The average half-life of a neutrophil in the circulation is 6 h. To maintain the normal circulating blood level, it is therefore necessary to produce over 100 billion neutrophils per day. Many neutrophils enter the tissues, particularly if triggered to do so by an infection or by inflammatory cytokines. They are attracted to the endothelial surface by cell adhesion molecules known as selectins, and they roll along it. They then bind firmly to neutrophil adhesion molecules of the integrin family. They next insinuate themselves through the walls of the capillaries between endothelial cells by a process called diapedesis. Many of those that leave the circulation enter the gastrointestinal tract and are eventually lost from the body.
Invasion of the body by bacteria triggers the inflammatory response. The bone marrow is stimulated to produce and release large numbers of neutrophils. Bacterial products interact with plasma factors and cells to produce agents that attract neutrophils to the infected area (chemotaxis). The chemotactic agents, which are part of a large and expanding family of chemokines (see following text), include a component of the complement system (C5a); leukotrienes; and polypeptides from lymphocytes, mast cells, and basophils. Other plasma factors act on the bacteria to make them “tasty” to the phagocytes (opsonization). The principal opsonins that coat the bacteria are immunoglobulins of a particular class (IgG) and complement proteins (see following text). The coated bacteria then bind to G protein-coupled receptors on the neutrophil cell membrane. This triggers increased motor activity of the cell, exocytosis, and the so-called respiratory burst. The increased motor activity leads to prompt ingestion of the bacteria by endocytosis (phagocytosis). By exocytosis, neutrophil granules discharge their contents into the phagocytic vacuoles containing the bacteria and also into the interstitial space (degranulation). The granules contain various proteases plus antimicrobial proteins called defensins. In addition, the cell membrane-bound enzyme NADPH oxidase is activated, with the production of toxic oxygen metabolites. The combination of the toxic oxygen metabolites and the proteolytic enzymes from the granules makes the neutrophil a very effective killing machine.
Activation of NADPH oxidase is associated with a sharp increase in O2 uptake and metabolism in the neutrophil (the respiratory burst) and generation of by the following reaction:
is a free radical formed by the addition of one electron to O2. Two react with two H+ to form H2O2 in a reaction catalyzed by the cytoplasmic form of superoxide dismutase (SOD-1):
and H2O2 are both oxidants that are effective bactericidal agents, but H2O2 is converted to H2O and O2 by the enzyme catalase. The cytoplasmic form of SOD contains both Zn and Cu. It is found in many parts of the body. It is defective as a result of genetic mutation in a familial form of amyotrophic lateral sclerosis (ALS; see Chapter 15). Therefore, it may be that accumulates in motor neurons and kills them in at least one form of this progressive, fatal disease. Two other forms of SOD encoded by at least one different gene are also found in humans.
Neutrophils also discharge the enzyme myeloperoxidase, which catalyzes the conversion of Cl−, Br−, I−, and SCN− to the corresponding acids (HOCl, HOBr, etc). These acids are also potent oxidants. Because Cl− is present in greatest abundance in body fluids, the principal product is HOCl.
In addition to myeloperoxidase and defensins, neutrophil granules contain elastase, metalloproteinases that attack collagen, and a variety of other proteases that help destroy invading organisms. These enzymes act in a cooperative fashion with , H2O2, and HOCl to produce a killing zone around the activated neutrophil. This zone is effective in killing invading organisms, but in certain diseases (eg, rheumatoid arthritis) the neutrophils may also cause local destruction of host tissue.
Like neutrophils, eosinophils have a short half-life in the circulation, are attracted to the surface of endothelial cells by selectins, bind to integrins that attach them to the vessel wall, and enter the tissues by diapedesis. Like neutrophils, they release proteins, cytokines, and chemokines that produce inflammation but are capable of killing invading organisms. However, eosinophils have some selectivity in the way in which they respond and in the killing molecules they secrete. Their maturation and activation in tissues is particularly stimulated by IL-3, IL-5, and GM-CSF (see below). They are especially abundant in the mucosa of the gastrointestinal tract, where they defend against parasites, and in the mucosa of the respiratory and urinary tracts. Circulating eosinophils are increased in allergic diseases such as asthma and in various other respiratory and gastrointestinal diseases.
Basophils also enter tissues and release proteins and cytokines. They resemble but are not identical to mast cells, and like mast cells they contain histamine (see below). They release histamine and other inflammatory mediators when activated by binding of specific antigens to cell-fixed IgE molecules, and participate in immediate-type hypersensitivity (allergic) reactions. These range from mild urticaria and rhinitis to severe anaphylactic shock. The antigens that trigger IgE formation and basophil (and mast cell) activation are innocuous to most individuals, and are referred to as allergens.
Mast cells are heavily granulated cells of the connective tissue that are abundant in tissues that come into contact with the external environment, such as beneath epithelial surfaces. Their granules contain proteoglycans, histamine, and many proteases. Like basophils, they degranulate when allergens bind to cell-bound IgE molecules directed against them. They are involved in inflammatory responses initiated by immunoglobulins IgE and IgG (see below). The inflammation combats invading parasites. In addition to this involvement in acquired immunity, they release TNF-α in response to bacterial products by an antibody-independent mechanism, thus participating in the nonspecific innate immunity that combats infections prior to the development of an adaptive immune response (see following text). Marked mast cell degranulation produces clinical manifestations of allergy up to and including anaphylaxis.
Monocytes enter the blood from the bone marrow and circulate for about 72 h. They then enter the tissues and become tissue macrophages (Figure 3–1). Their life span in the tissues is unknown, but bone marrow transplantation data in humans suggest that they persist for about 3 months. It appears that they do not reenter the circulation. Some may end up as the multinucleated giant cells seen in chronic inflammatory diseases such as tuberculosis. The tissue macrophages include the Kupffer cells of the liver, pulmonary alveolar macrophages (see Chapter 34), and microglia in the brain, all of which come originally from the circulation. In the past, they have been called the reticuloendothelial system, but the general term tissue macrophage system seems more appropriate.
FIGURE 3–1 Macrophages contacting bacteria and preparing to engulf them. Figure is a colorized version of a scanning electronmicrograph.
Macrophages are activated by cytokines released from T lymphocytes, among others. Activated macrophages migrate in response to chemotactic stimuli and engulf and kill bacteria by processes generally similar to those occurring in neutrophils. They play a key role in innate immunity (see below). They also secrete up to 100 different substances, including factors that affect lymphocytes and other cells, prostaglandins of the E series, and clot-promoting factors.
Lymphocytes are key elements in the production of acquired immunity (see below). After birth, some lymphocytes are formed in the bone marrow. However, most are formed in the lymph nodes (Figure 3–2), thymus, and spleen from precursor cells that originally came from the bone marrow and were processed in the thymus (T cells) or bursal equivalent (B cells, see below). Lymphocytes enter the bloodstream for the most part via the lymphatics. At any given time, only about 2% of the body lymphocytes are in the peripheral blood. Most of the rest are in the lymphoid organs. It has been calculated that in humans, 3.5 × 1010 lymphocytes per day enter the circulation via the thoracic duct alone; however, this count includes cells that reenter the lymphatics and thus traverse the thoracic duct more than once. The effects of adrenocortical hormones on the lymphoid organs, the circulating lymphocytes, and the granulocytes are discussed in Chapter 20.
FIGURE 3–2 Anatomy of a normal lymph node. (After Chandrasoma. Reproduced with permission from McPhee SJ, Lingappa VR, Ganong WF [editors]: Pathophysiology of Disease, 4th ed. McGraw-Hill, 2003.)
During fetal development, and to a much lesser extent during adult life, lymphocyte precursors come from the bone marrow. Those that populate the thymus (Figure 3–3) become transformed by the environment in this organ into T lymphocytes. In birds, the precursors that populate the bursa of Fabricius, a lymphoid structure near the cloaca, become transformed into B lymphocytes. There is no bursa in mammals, and the transformation to B lymphocytes occurs in bursal equivalents, that is, the fetal liver and, after birth, the bone marrow. After residence in the thymus or liver, many of the T and B lymphocytes migrate to the lymph nodes.
FIGURE 3–3 Development of the system mediating acquired immunity.
T and B lymphocytes are morphologically indistinguishable but can be identified by markers on their cell membranes. B cells differentiate into plasma cells and memory B cells. There are three major types of T cells: cytotoxic T cells, helper T cells, and memory T cells. There are two subtypes of helper T cells: T helper 1 (TH1) cells secrete IL-2 and γ-interferon and are concerned primarily with cellular immunity; T helper 2 (TH2) cells secrete IL-4 and IL-5 and interact primarily with B cells in relation to humoral immunity. Cytotoxic T cells destroy transplanted and other foreign cells, with their development aided and directed by helper T cells. Markers on the surface of lymphocytes are assigned CD (clusters of differentiation) numbers on the basis of their reactions to a panel of monoclonal antibodies. Most cytotoxic T cells display the glycoprotein CD8, and helper T cells display the glycoprotein CD4. These proteins are closely associated with the T cell receptors and may function as coreceptors. On the basis of differences in their receptors and functions, cytotoxic T cells are divided into αβ and γδ types (see below). Natural killer (NK) cells (see above) are also cytotoxic lymphocytes, though they are not T cells. Thus, there are three main types of cytotoxic lymphocytes in the body: αβ T cells, γδ T cells, and NK cells.
MEMORY B CELLS & T CELLS
After exposure to a given antigen, a small number of activated B and T cells persist as memory B and T cells. These cells are readily converted to effector cells by a later encounter with the same antigen. This ability to produce an accelerated response to a second exposure to an antigen is a key characteristic of acquired immunity. The ability persists for long periods of time, and in some instances (eg, immunity to measles) it can be life-long.
After activation in lymph nodes, lymphocytes disperse widely throughout the body and are especially plentiful in areas where invading organisms enter the body, for example, the mucosa of the respiratory and gastrointestinal tracts. This puts memory cells close to sites of reinfection and may account in part for the rapidity and strength of their response. Chemokines are involved in guiding activated lymphocytes to these locations.
GRANULOCYTE & MACROPHAGE COLONY-STIMULATING FACTORS
The production of white blood cells is regulated with great precision in healthy individuals, and the production of granulocytes is rapidly and dramatically increased in infections. The proliferation and self-renewal of hematopoietic stem cells (HSCs) depends on stem cell factor (SCF). Other factors specify particular lineages. The proliferation and maturation of the cells that enter the blood from the marrow are regulated by growth factors that cause cells in one or more of the committed cell lines to proliferate and mature (Table 3–1). The regulation of erythrocyte production by erythropoietin is discussed in Chapter 38. Three additional factors are called colony-stimulating factors (CSFs), because they cause appropriate single stem cells to proliferate in soft agar, forming colonies. The factors stimulating the production of committed stem cells include granulocyte–macrophage CSF (GM-CSF), granulocyte CSF (G-CSF), and macrophage CSF (M-CSF). Interleukins IL-1 and IL-6 followed by IL-3 (Table 3–1) act in sequence to convert pluripotential uncommitted stem cells to committed progenitor cells. IL-3 is also known as multi-CSF. Each of the CSFs has a predominant action, but all the CSFs and interleukins also have other overlapping actions. In addition, they activate and sustain mature blood cells. It is interesting in this regard that the genes for many of these factors are located together on the long arm of chromosome 5 and may have originated by duplication of an ancestral gene. It is also interesting that basal hematopoiesis is normal in mice in which the GM-CSF gene is knocked out indicating that loss of one factor can be compensated for by others. On the other hand, the absence of GM-CSF causes accumulation of surfactant in the lungs (see Chapter 34).
TABLE 3–1 Hematopoietic growth factors.
As noted in Chapter 38, erythropoietin is produced in part by kidney cells and is a circulating hormone. The other factors are produced by macrophages, activated T cells, fibroblasts, and endothelial cells. For the most part, the factors act locally in the bone marrow (Clinical Box 3–1).
CLINICAL BOX 3–1
Disorders of Phagocytic Function
More than 15 primary defects in neutrophil function have been described, along with at least 30 other conditions in which there is a secondary depression of the function of neutrophils. Patients with these diseases are prone to infections that are relatively mild when only the neutrophil system is involved, but which can be severe when the monocyte-tissue macrophage system is also involved. In one syndrome (neutrophil hypomotility), actin in the neutrophils does not polymerize normally, and the neutrophils move slowly. In another, there is a congenital deficiency of leukocyte integrins. In a more serious disease (chronic granulomatous disease), there is a failure to generate in both neutrophils and monocytes and consequent inability to kill many phagocytosed bacteria. In severe congenital glucose 6-phosphate dehydrogenase deficiency, there are multiple infections because of failure to generate the NADPH necessary for production. In congenital myeloperoxidase deficiency, microbial killing power is reduced because hypochlorous acid is not formed.
The cornerstones of treatment in disorders of phagocytic function include scrupulous efforts to avoid exposure to infectious agents, and antibiotic and antifungal prophylaxis. Antimicrobial therapies must also be implemented aggressively if infections occur. Sometimes, surgery is needed to excise and/or drain abscesses and relieve obstructions. Bone marrow transplantation may offer the hope of a definitive cure for severe conditions, such as chronic granulomatous disease. Sufferers of this condition have a significantly reduced life expectancy due to recurrent infections and their complications, and so the risks of bone marrow transplantation may be deemed acceptable. Gene therapy, on the other hand, remains a distant goal.
Insects and other invertebrates have only innate immunity. This system is triggered by receptors that bind sequences of sugars, lipids, amino acids, or nucleic acids that are common on bacteria and other microorganisms, but are not found in eukaryotic cells. These receptors, in turn, activate various defense mechanisms. The receptors are coded in the germ line, and their fundamental structure is not modified by exposure to antigen. The activated defenses include, in various species, release of interferons, phagocytosis, production of antibacterial peptides, activation of the complement system, and several proteolytic cascades. Even plants release antibacterial peptides in response to infection. This primitive immune system is also important in vertebrates, particularly in the early response to infection. However, in vertebrates, innate immunity is also complemented by adaptive or acquired immunity, a system in which T and B lymphocytes are activated by specific antigens. T cells bear receptors related to antibody molecules, but which remain cell-bound. When these receptors encounter their cognate antigen, the T cell is stimulated to proliferate and produce cytokines that orchestrate the immune response, including that of B cells. Activated B lymphocytes form clones that produce secreted antibodies, which attack foreign proteins. After the invasion is repelled, small numbers of lymphocytes persist as memory cells so that a second exposure to the same antigen provokes a prompt and magnified immune attack. The genetic event that led to acquired immunity occurred 450 million years ago in the ancestors of jawed vertebrates and was probably insertion of a transposon into the genome in a way that made possible the generation of the immense repertoire of T cell receptors and antibodies that can be produced by the body.
In vertebrates, including humans, innate immunity provides the first line of defense against infections, but it also triggers the slower but more specific acquired immune response (Figure 3–4). In vertebrates, natural and acquired immune mechanisms also attack tumors and tissue transplanted from other animals.
FIGURE 3–4 How bacteria, viruses, and tumors trigger innate immunity and initiate the acquired immune response. Arrows indicate mediators/cytokines that act on the target cell shown and/or pathways of differentiation. APC, antigen-presenting cell; M, monocyte; N, neutrophil; TH1 and TH2, helper T cells type 1 and type 2, respectively.
Once activated, immune cells communicate by means of cytokines and chemokines. They kill viruses, bacteria, and other foreign cells by secreting other cytokines and activating the complement system.
Cytokines are hormone-like molecules that act—generally in a paracrine fashion—to regulate immune responses. They are secreted not only by lymphocytes and macrophages but by endothelial cells, neurons, glial cells, and other types of cells. Most of the cytokines were initially named for their actions, for example, B cell-differentiating factor, or B cell-stimulating factor 2. However, the nomenclature has since been rationalized by international agreement to that of the interleukins. For example, the name of B cell-differentiating factor was changed to interleukin-4. A number of cytokines selected for their biological and clinical relevance are listed in Table 3–2, but it would be beyond the scope of this text to list all cytokines, which now number more than 100.
TABLE 3–2 Examples of cytokines and their clinical relevance.
Many of the receptors for cytokines and hematopoietic growth factors (see above), as well as the receptors for prolactin (see Chapter 22), and growth hormone (see Chapter 18) are members of a cytokine-receptor superfamily that has three subfamilies (Figure 3–5). The members of subfamily 1, which includes the receptors for IL-4 and IL-7, are homodimers. The members of subfamily 2, which includes the receptors for IL-3, IL-5, and IL-6, are heterodimers. The receptor for IL-2 (and several other cytokines) consists of a heterodimer plus an unrelated protein, the so-called Tac antigen. The other members of subfamily 3 have the same γ chain as IL-2R. The extracellular domain of the homodimer and heterodimer subunits all contain four conserved cysteine residues plus a conserved Trp-Ser-X-Trp-Ser domain, and although the intracellular portions do not contain tyrosine kinase catalytic domains, they activate cytoplasmic tyrosine kinases when ligand binds to the receptors.
FIGURE 3–5 Members of one of the cytokine receptor superfamilies, showing shared structural elements. Note that all the subunits except the α subunit in subfamily 3 have four conserved cysteine residues (open boxes at top) and a Trp-Ser-X-Trp-Ser motif (pink). Many subunits also contain a critical regulatory domain in their cytoplasmic portions (green). CNTF, ciliary neurotrophic factor; LIF, leukemia inhibitory factor; OSM, oncostatin M; PRL, prolactin. (Modified from D’Andrea AD: Cytokine receptors in congenital hematopoietic disease. N Engl J Med 1994;330:839.)
The effects of the principal cytokines are listed in Table 3–2. Some of them have systemic as well as local paracrine effects. For example, IL-1, IL-6, and tumor necrosis factor α cause fever, and IL-1 increases slow-wave sleep and reduces appetite.
Another superfamily of cytokines is the chemokine family. Chemokines are substances that attract neutrophils (see previous text) and other white blood cells to areas of inflammation or immune response. Over 40 have now been identified, and it is clear that they also play a role in the regulation of cell growth and angiogenesis. The chemokine receptors are G protein-coupled receptors that cause, among other things, extension of pseudopodia with migration of the cell toward the source of the chemokine.
THE COMPLEMENT SYSTEM
The cell-killing effects of innate and acquired immunity are mediated in part by a system of more than 30 plasma proteins originally named the complement system because they “complemented” the effects of antibodies. Three different pathways or enzyme cascades activate the system: the classic pathway, triggered by immune complexes; the mannose-binding lectin pathway, triggered when this lectin binds mannose groups in bacteria; and the alternative or properdin pathway, triggered by contact with various viruses, bacteria, fungi, and tumor cells. The proteins that are produced have three functions: they help kill invading organisms by opsonization, chemotaxis, and eventual lysis of the cells; they serve in part as a bridge from innate to acquired immunity by activating B cells and aiding immune memory; and they help dispose of waste products after apoptosis. Cell lysis, one of the principal ways the complement system kills cells, is brought about by inserting proteins called perforins into their cell membranes. These create holes, which permit free flow of ions and thus disruption of membrane polarity.
The cells that mediate innate immunity include neutrophils, macrophages, and natural killer cells, large cytotoxic lymphocytes distinct from both T and B cells. All these cells respond to molecular patterns produced by bacteria and to other substances characteristic of viruses, tumor, and transplant cells. Many cells that are not professional immunocytes may nevertheless also contribute to innate immune responses, such as endothelial and epithelial cells. The activated cells produce their effects via the release of cytokines, as well as, in some cases, complement and other systems.
Innate immunity in Drosophila centers around a receptor protein named toll, which binds fungal antigens and triggers activation of genes coding for antifungal proteins. An expanding list of toll-like receptors (TLRs) have now been identified in humans and other vertebrates. One of these, TLR4, binds bacterial lipopolysaccharide and a protein called CD14, and this initiates intracellular events that activate transcription of genes for a variety of proteins involved in innate immune responses. This is important because bacterial lipopolysaccharide produced by gram-negative organisms is the cause of septic shock. TLR2 mediates the response to microbial lipoproteins, TLR6 cooperates with TLR2 in recognizing certain peptidoglycans, TLR5 recognizes a molecule known as flagellin in bacterial flagellae, and TLR9 recognizes bacterial DNA. TLRs are referred to as pattern recognition receptors (PRRs), because they recognize and respond to the molecular patterns expressed by pathogens. Other PRRs may be intracellular, such as the so-called NOD proteins. One NOD protein, NOD2, has received attention as a candidate gene leading to the intestinal inflammatory condition, Crohn’s disease (Clinical Box 3–2).
CLINICAL BOX 3–2
Crohn’s disease is a chronic, relapsing, and remitting disease that involves transmural inflammation of the intestine that can occur at any point along the length of the gastrointestinal tract, but most commonly is confined to the distal small intestine and colon. Patients with this condition suffer from changes in bowel habits, bloody diarrhea, severe abdominal pain, weight loss, and malnutrition. Evidence is accumulating that the disease reflects a failure to down-regulate inflammatory responses to the normal gut commensal microbiota. In genetically-susceptible individuals, mutations in genes controlling innate immune responses (eg, NOD2) or regulators of acquired immunity appear to predispose to disease when individuals are exposed to appropriate environmental factors, which can include a change in the microbiota or stress
During flares of Crohn’s disease, the mainstay of treatment remains high-dose corticosteroids to suppress inflammation nonspecifically. Surgery is often required to treat complications such as strictures, fistulas, and abscesses. Some patients with severe disease also benefit from ongoing treatment with immunosuppressive drugs, or from treatment with antibodies targeted against tumor necrosis factor-α. Probiotics, therapeutic microorganisms designed to restore a “healthy” microbiota, may have some role in prophylaxis. The pathogenesis of Crohn’s disease, as well as the related inflammatory bowel disease, ulcerative colitis, remains the subject of intense investigation, and therapies that target specific facets of the inflammatory cascade that may be selectively implicated in individual patients with differing genetic backgrounds are under development.
As noted previously, the key to acquired immunity is the ability of lymphocytes to produce antibodies (in the case of B cells) or cell-surface receptors (in the case of T cells) that are specific for one of the many millions of foreign agents that may invade the body. The antigens stimulating production of T cell receptors or antibodies are usually proteins and polypeptides, but antibodies can also be formed against nucleic acids and lipids if these are presented as nucleoproteins and lipoproteins. Antibodies to small molecules can also be produced experimentally if the molecules are bound to protein. Acquired immunity has two components: humoral immunity and cellular immunity. Humoral immunity is mediated by circulating immunoglobulin antibodies in the γ-globulin fraction of the plasma proteins. Immunoglobulins are produced by differentiated forms of B lymphocytes known as plasma cells, and they activate the complement system and attack and neutralize antigens. Humoral immunity is a major defense against bacterial infections. Cellular immunity is mediated by T lymphocytes. It is responsible for delayed allergic reactions and rejection of transplants of foreign tissue. Cytotoxic T cells attack and destroy cells bearing the antigen that activated them. They kill by inserting perforins (see above) and by initiating apoptosis. Cellular immunity constitutes a major defense against infections due to viruses, fungi, and a few bacteria such as the tubercle bacillus. It also helps defend against tumors.
The number of different antigens recognized by lymphocytes in the body is extremely large. The repertoire develops initially without exposure to the antigen. Stem cells differentiate into many million different T and B lymphocytes, each with the ability to respond to a particular antigen. When the antigen first enters the body, it can bind directly to the appropriate receptors on B cells. However, a full antibody response requires that the B cells contact helper T cells. In the case of T cells, the antigen is taken up by an antigen-presenting cell (APC) and partially digested. A peptide fragment of it is presented to the appropriate receptors on T cells. In either case, the cells are stimulated to divide, forming clones of cells that respond to this antigen (clonal selection). Effector cells are also subject to negative selection, during which lymphocyte precursors that are reactive with self-antigens are normally deleted. This results in immune tolerance. It is this latter process that presumably goes awry in autoimmune diseases, where the body reacts to and destroys cells expressing normal proteins, with accompanying inflammation that may lead to tissue destruction.
APCs include specialized cells called dendritic cells in the lymph nodes and spleen and the Langerhans dendritic cells in the skin. Macrophages and B cells themselves, and likely many other cell types, can also function as APCs. For example, in the intestine, the epithelial cells that line the tract are likely important in presenting antigens derived from commensal bacteria. In APCs, polypeptide products of antigen digestion are coupled to protein products of the major histocompatibility complex (MHC) genes and presented on the surface of the cell. The products of the MHC genes are called human leukocyte antigens (HLA).
The genes of the MHC, which are located on the short arm of human chromosome 6, encode glycoproteins and are divided into two classes on the basis of structure and function. Class I antigens are composed of a 45-kDa heavy chain associated noncovalently with β2-microglobulin encoded by a gene outside the MHC (Figure 3–6). They are found on all nucleated cells. Class II antigens are heterodimers made up of a 29–34-kDa α chain associated noncovalently with a 25–28-kDa β chain. They are present in “professional” APCs, including B cells, and in activated T cells.
FIGURE 3–6 Structure of human histocompatibility antigen HLA-A2. The antigen-binding pocket is at the top and is formed by the α1 and α2 parts of the molecule. The α3 portion and the associated β2-microglobulin (β2m) are close to the membrane. The extension of the C terminal from α 3 that provides the transmembrane domain and the small cytoplasmic portion of the molecule have been omitted. (Reproduced with permission from Bjorkman PJ, et al: Structure of the human histocompatibility antigen HLA-A2. Nature 1987;329:506.)
The class I MHC proteins (MHC-I proteins) are coupled primarily to peptide fragments generated from proteins synthesized within cells. Peptides to which the host is not tolerant (eg, those from mutant or viral proteins) are recognized by T cells. The digestion of these proteins occurs in complexes of proteolytic enzymes known as proteasomes, and the peptide fragments bind to MHC proteins in the endoplasmic reticulum. The class II MHC proteins (MHC-II proteins) are concerned primarily with peptide products of extracellular antigens, such as bacteria, that enter the cell by endocytosis and are digested in the late endosomes.
T CELL RECEPTORS
The MHC protein–peptide complexes on the surface of the APCs bind to appropriate T cells. Therefore, receptors on the T cells must recognize a very wide variety of complexes. Most of the receptors on circulating T cells are made up of two polypeptide units designated α and β. They form heterodimers that recognize the MHC proteins and the antigen fragments with which they are combined (Figure 3–7). These cells are called αβ T cells. On the other hand, about 10% of circulating T cells have two different polypeptides designated γ and δ in their receptors, and they are called γδ T cells. These T cells are prominent in the mucosa of the gastrointestinal tract, and there is evidence that they form a link between the innate and acquired immune systems by way of the cytokines they secrete (Figure 3–3).
FIGURE 3–7 Interaction between antigen-presenting cell (top) and αβ T lymphocyte (bottom). The MHC proteins (in this case, MHC-I) and their peptide antigen fragment bind to the α and β units that combine to form the T cell receptor.
CD8 occurs on the surface of cytotoxic T cells that bind MHC-I proteins, and CD4 occurs on the surface of helper T cells that bind MHC-II proteins (Figure 3–8). The CD8 and CD4 proteins facilitate the binding of the MHC proteins to the T cell receptors, and they also foster lymphocyte development. The activated CD8 cytotoxic T cells kill their targets directly, whereas the activated CD4 helper T cells secrete cytokines that activate other lymphocytes.
FIGURE 3–8 Diagrammatic summary of the structure of CD4 and CD8, and their relation to MHC-I and MHC-II proteins. Note that CD4 is a single protein, whereas CD8 is a heterodimer.
The T cell receptors are surrounded by adhesion molecules and proteins that bind to complementary proteins in the APC when the two cells transiently join to form the “immunologic synapse” that permits T cell activation to occur (Figure 3–7). It is now generally accepted that two signals are necessary to produce activation. One is produced by the binding of the digested antigen to the T cell receptor. The other is produced by the joining of the surrounding proteins in the “synapse.” If the first signal occurs but the second does not, the T cell is inactivated and becomes unresponsive.
As noted above, B cells can bind antigens directly, but they must contact helper T cells to produce full activation and antibody formation. It is the TH2 subtype that is mainly involved. Helper T cells develop along the TH2 lineage in response to IL-4 (see below). On the other hand, IL-12 promotes the TH1 phenotype. IL-2 acts in an autocrine fashion to cause activated T cells to proliferate. The role of various cytokines in B cell and T cell activation is summarized in Figure 3–9.
FIGURE 3–9 Summary of acquired immunity. (1) An antigen-presenting cell ingests and partially digests an antigen, then presents part of the antigen along with MHC peptides (in this case, MHC II peptides on the cell surface). (2) An “immune synapse” forms with a naive CD4 T cell, which is activated to produce IL-2. (3) IL-2 acts in an auto-crine fashion to cause the cell to multiply, forming a clone. (4) The activated CD4 cell may promote B cell activation and production of plasma cells or it may activate a cytotoxic CD8 cell. The CD8 cell can also be activated by forming a synapse with an MCH I antigen-presenting cell. (Reproduced with permission from McPhee SJ, Lingappa VR, Ganong WF [editors]: Pathophysiology of Disease, 6th ed. McGraw-Hill, 2010.)
The activated B cells proliferate and transform into memory B cells (see above) and plasma cells. The plasma cells secrete large quantities of antibodies into the general circulation. The antibodies circulate in the globulin fraction of the plasma and, like antibodies elsewhere, are called immunoglobulins. The immunoglobulins are actually the secreted form of antigen-binding receptors on the B cell membrane.
Circulating antibodies protect their host by binding to and neutralizing some protein toxins, by blocking the attachment of some viruses and bacteria to cells, by opsonizing bacteria (see above), and by activating complement. Five general types of immunoglobulin antibodies are produced by plasma cells. The basic component of each is a symmetric unit containing four polypeptide chains (Figure 3–10). The two long chains are called heavy chains, whereas the two short chains are called light chains. There are two types of light chains, k and λ, and eight types of heavy chains. The chains are joined by disulfide bridges that permit mobility, and there are intrachain disulfide bridges as well. In addition, the heavy chains are flexible in a region called the hinge. Each heavy chain has a variable (V) segment in which the amino acid sequence is highly variable, a diversity (D) segment in which the amino acid segment is also highly variable, a joining (J) segment in which the sequence is moderately variable, and a constant (C) segment in which the sequence is constant. Each light chain has a V, J, and C segment. The V segments form part of the antigen-binding sites (Fab portion of the molecule [Figure 3–10]). The Fc portion of the molecule is the effector portion, which mediates the reactions initiated by antibodies.
FIGURE 3–10 Typical immunoglobulin G molecule. Fab, portion of the molecule that is concerned with antigen binding; Fc, effector portion of the molecule. The constant regions are pink and purple, and the variable regions are orange. The constant segment of the heavy chain is subdivided into CH1, CH2, and CH3. SS lines indicate intersegmental disulfide bonds. On the right side, the C labels are omitted to show regions JH, D, and JL.
Two of the classes of immunoglobulins contain additional polypeptide components (Table 3–3). In IgM, five of the basic immunoglobulin units join around a polypeptide called the J chain to form a pentamer. In IgA, the secretory immunoglobulin, the immunoglobulin units form dimers and trimers around a J chain and a polypeptide that comes from epithelial cells, the secretory component (SC).
TABLE 3–3 Human immunoglobulins.a
In the intestine, bacterial and viral antigens are taken up by M cells (see Chapter 26) and passed on to underlying aggregates of lymphoid tissue (Peyer’s patches), where they activate naive T cells. These lymphocytes then form B cells that infiltrate mucosa of the gastrointestinal, respiratory, genitourinary, and female reproductive tracts and the breast. There they secrete large amounts of IgA when exposed again to the antigen that initially stimulated them. The epithelial cells produce the SC, which acts as a receptor for, and binds to, IgA. The resulting secretory immunoglobulin passes through the epithelial cell and is secreted by exocytosis. This system of secretory immunity is an important and effective defense mechanism at all mucosal surfaces. It also accounts for the immune protection that is conferred by breast feeding of infants whose immune systems are otherwise immature, because IgA is secreted into the breast milk.
GENETIC BASIS OF DIVERSITY IN THE IMMUNE SYSTEM
The genetic mechanism for the production of the immensely large number of different configurations of immunoglobulins produced by human B cells, as well as T cell receptors, is a fascinating biologic problem. Diversity is brought about in part by the fact that in immunoglobulin molecules there are two kinds of light chains and eight kinds of heavy chains. As noted previously, there are areas of great variability (hypervariable regions) in each chain. The variable portion of the heavy chains consists of the V, D, and J segments. In the gene family responsible for this region, there are several hundred different coding regions for the V segment, about 20 for the D segment, and four for the J segment. During B cell development, one V coding region, one D coding region, and one J coding region are selected at random and recombined to form the gene that produces that particular variable portion. A similar variable recombination takes place in the coding regions responsible for the two variable segments (V and J) in the light chain. In addition, the J segments are variable because the gene segments join in an imprecise and variable fashion (junctional site diversity) and nucleotides are sometimes added (junctional insertion diversity). It has been calculated that these mechanisms permit the production of about 1015 different immunoglobulin molecules. Additional variability is added by somatic mutation.
Similar gene rearrangement and joining mechanisms operate to produce the diversity in T cell receptors. In humans, the α subunit has a V region encoded by 1 of about 50 different genes and a J region encoded by 1 of another 50 different genes. The β subunits have a V region encoded by 1 of about 50 genes, a D region encoded by 1 of 2 genes, and a J region encoded by 1 of 13 genes. These variable regions permit the generation of up to an estimated 1015different T cell receptors (Clinical Box 3–3 and Clinical Box 3–4).
CLINICAL BOX 3–3
Sometimes the processes that eliminate antibodies against self-antigens fail and a variety of different autoimmune diseases are produced. These can be B cell- or T cell-mediated and can be organ-specific or systemic. They include type 1 diabetes mellitus (antibodies against pancreatic islet B cells), myasthenia gravis (antibodies against nicotinic cholinergic receptors), and multiple sclerosis (antibodies against myelin basic protein and several other components of myelin). In some instances, the antibodies are against receptors and are capable of activating those receptors; for example, antibodies against TSH receptors increase thyroid activity and cause Graves’ disease (see Chapter 19). Other conditions are due to the production of antibodies against invading organisms that cross-react with normal body constituents (molecular mimicry). An example is rheumatic fever following a streptococcal infection; a portion of cardiac myosin resembles a portion of the streptococcal M protein, and antibodies induced by the latter attack the former and damage the heart. Some conditions may be due to bystander effects, in which inflammation sensitizes T cells in the neighborhood, causing them to become activated when otherwise they would not respond.
The therapy of autoimmune disorders rests on efforts to replace or restore the damaged function (eg, provision of exogenous insulin in type 1 diabetes) as well as nonspecific efforts to reduce inflammation (using corticosteroids) or to suppress immunity. Recently, agents that deplete or dampen the function of B cells have been shown to have some efficacy in a range of autoimmune disorders, including rheumatoid arthritis, most likely by interrupting the production of autoantibodies that contribute to disease pathogenesis.
CLINICAL BOX 3–4
The T lymphocyte system is responsible for the rejection of transplanted tissue. When tissues such as skin and kidneys are transplanted from a donor to a recipient of the same species, the transplants “take” and function for a while but then become necrotic and are “rejected” because the recipient develops an immune response to the transplanted tissue. This is generally true even if the donor and recipient are close relatives, and the only transplants that are never rejected are those from an identical twin. Nevertheless, organ transplantation remains the only viable option in a number of end stage diseases.
A number of treatments have been developed to overcome the rejection of transplanted organs in humans. The goal of treatment is to stop rejection without leaving the patient vulnerable to massive infections. One approach is to kill T lymphocytes by killing all rapidly dividing cells with drugs such as azathioprine, a purine antimetabolite, but this makes patients susceptible to infections and cancer. Another is to administer glucocorticoids, which inhibit cytotoxic T cell proliferation by inhibiting production of IL-2, but these cause osteoporosis, mental changes, and the other facets of Cushing syndrome (see Chapter 20). More recently, immunosuppressive drugs such as cyclosporine or tacrolimus (FK-506) have found favor. Activation of the T cell receptor normally increases intracellular Ca2+, which acts via calmodulin to activate calcineurin. Calcineurin dephosphorylates the transcription factor NF-AT, which moves to the nucleus and increases the activity of genes coding for IL-2 and related stimulatory cytokines. Cyclosporine and tacrolimus prevent the dephosphorylation of NF-AT. However, these drugs inhibit all T cell-mediated immune responses, and cyclosporine causes kidney damage and cancer. A new and promising approach to transplant rejection is the production of T cell unresponsiveness by using drugs that block the costimulation that is required for normal activation (see text). Clinically effective drugs that act in this fashion could be of great value to transplant surgeons.
A variety of immunodeficiency states can arise from defects in these various stages of B and T lymphocyte maturation. These are summarized in Figure 3–11.
FIGURE 3–11 Sites of congenital blockade of B and T lymphocyte maturation in various immunodeficiency states. SCID, severe combined immune deficiency. (Modified from Rosen FS, Cooper MD, Wedgwood RJP: The primary immunodeficiencies. N Engl J Med 1995;333:431.)
Platelets are circulating cells that are important mediators of hemostasis. While not immune cells, per se, they often participate in the response to tissue injury in cooperation with inflammatory cell types (see below). They have a ring of microtubules around their periphery and an extensively invaginated membrane with an intricate canalicular system in contact with the ECF. Their membranes contain receptors for collagen, ADP, vessel wall von Willebrand factor (see below), and fibrinogen. Their cytoplasm contains actin, myosin, glycogen, lysosomes, and two types of granules: (1) dense granules, which contain the nonprotein substances that are secreted in response to platelet activation, including serotonin, ADP, and other adenine nucleotides; and (2) α-granules, which contain secreted proteins. These proteins include clotting factors and platelet-derived growth factor (PDGF). PDGF is also produced by macrophages and endothelial cells. It is a dimer made up of A and B subunit polypeptides. Homodimers (AA and BB), as well as the heterodimer (AB), are produced. PDGF stimulates wound healing and is a potent mitogen for vascular smooth muscle. Blood vessel walls as well as platelets contain von Willebrand factor, which, in addition to its role in adhesion, regulates circulating levels of factor VIII (see below).
When a blood vessel wall is injured, platelets adhere to the exposed collagen and von Willebrand factor in the wall via receptors on the platelet membrane. Von Willebrand factor is a very large circulating molecule that is produced by endothelial cells. Binding produces platelet activations, which release the contents of their granules. The released ADP acts on the ADP receptors in the platelet membranes to produce further accumulation of more platelets (platelet aggregation). Humans have at least three different types of platelet ADP receptors: P2Y1, P2Y2, and P2X1. These are obviously attractive targets for drug development, and several new inhibitors have shown promise in the prevention of heart attacks and strokes. Aggregation is also fostered by platelet-activating factor (PAF), a cytokine secreted by neutrophils and monocytes as well as platelets. This compound also has inflammatory activity. It is an ether phospholipid, 1-alkyl-2-acetylglyceryl-3-phosphorylcho-line, which is produced from membrane lipids. It acts via a G protein-coupled receptor to increase the production of arachidonic acid derivatives, including thromboxane A2. The role of this compound in the balance between clotting and anticlotting activity at the site of vascular injury is discussed in Chapter 31.
Platelet production is regulated by the CSFs that control the production of the platelet precursors in the bone marrow, known as megakaryocytes, plus thrombopoietin, a circulating protein factor. This factor, which facilitates megakaryocyte maturation, is produced constitutively by the liver and kidneys, and there are thrombopoietin receptors on platelets. Consequently, when the number of platelets is low, less is bound and more is available to stimulate production of platelets. Conversely, when the number of platelets is high, more is bound and less is available, producing a form of feedback control of platelet production. The amino terminal portion of the thrombopoietin molecule has the platelet-stimulating activity, whereas the carboxyl terminal portion contains many carbohydrate residues and is concerned with the bioavailability of the molecule.
When the platelet count is low, clot retraction is deficient and there is poor constriction of ruptured vessels. The resulting clinical syndrome (thrombocytopenic purpura) is characterized by easy bruisability and multiple subcutaneous hemorrhages. Purpura may also occur when the platelet count is normal, and in some of these cases, the circulating platelets are abnormal (thrombasthenic purpura). Individuals with thrombocytosis are predisposed to thrombotic events.
INFLAMMATION & WOUND HEALING
Inflammation is a complex localized response to foreign substances such as bacteria or in some instances to internally produced substances. It includes a sequence of reactions initially involving cytokines, neutrophils, adhesion molecules, complement, and IgG. PAF, an agent with potent inflammatory effects, also plays a role. Later, monocytes and lymphocytes are involved. Arterioles in the inflamed area dilate, and capillary permeability is increased (see Chapters 32 and 33). When the inflammation occurs in or just under the skin (Figure 3–12), it is characterized by redness, swelling, tenderness, and pain. Elsewhere, it is a key component of asthma, ulcerative colitis, Crohn’s disease, rheumatoid arthritis, and many other diseases (Clinical Box 3–2).
FIGURE 3–12 Cutaneous wound 3 days after injury, showing the multiple cytokines and growth factors affecting the repair process. VEGF, vascular endothelial growth factor. For other abbreviations, see Appendix. Note the epidermis growing down under the fibrin clot, restoring skin continuity. (Modified from Singer AJ, Clark RAF: Cutaneous wound healing. N Engl J Med 1999;341:738.)
Evidence is accumulating that a transcription factor, nuclear factor-κB, plays a key role in the inflammatory response. NF-κB is a heterodimer that normally exists in the cytoplasm of cells bound to IκBα, which renders it inactive. Stimuli such as cytokines, viruses, and oxidants induce signals that allow NF-κB to dissociate from IκBα, which is then degraded. NF-κB moves to the nucleus, where it binds to the DNA of the genes for numerous inflammatory mediators, resulting in their increased production and secretion. Glucocorticoids inhibit the activation of NF-κB by increasing the production of IκBα, and this is probably the main basis of their anti-inflammatory action (see Chapter 20).
SYSTEMIC RESPONSE TO INJURY
Cytokines produced in response to inflammation and other injuries, as well as disseminated infection, also produce systemic responses. These include alterations in plasma acute phase proteins, defined as proteins whose concentration is increased or decreased by at least 25% following injury. Many of the proteins are of hepatic origin. A number of them are shown in Figure 3–13. The causes of the changes in concentration are incompletely understood, but it can be said that many of the changes make homeostatic sense. Thus, for example, an increase in C-reactive protein activates monocytes and causes further production of cytokines. Other changes that occur in response to injury include somnolence, negative nitrogen balance, and fever.
FIGURE 3–13 Time course of changes in some major acute phase proteins. C3, C3 component of complement. (Modified and reproduced with permission from McAdam KP, Elin RJ, Sipe JD, Wolff SM: Changes in human serum amyloid A and C-reactive protein after etiocholanolone-induced inflammation. J Clin Invest, 1978 Feb;61(2):390–394.)
When tissue is damaged, platelets adhere to exposed matrix via integrins that bind to collagen and laminin (Figure 3–12). Blood coagulation produces thrombin, which promotes platelet aggregation and granule release. The platelet granules generate an inflammatory response. White blood cells are attracted by selectins and bind to integrins on endothelial cells, leading to their extravasation through the blood vessel walls. Cytokines released by the white blood cells and platelets up-regulate integrins on macrophages, which migrate to the area of injury, and on fibroblasts and epithelial cells, which mediate wound healing and scar formation. Plasmin aids healing by removing excess fibrin. This aids the migration of keratinocytes into the wound to restore the epithelium under the scab. Collagensynthesis is upregulated, producing the scar. Wounds gain 20% of their ultimate strength in 3 weeks and later gain more strength, but they never reach more than about 70% of the strength of normal skin.
Immune and inflammatory responses are mediated by several different cell types—granulocytes, lymphocytes, monocytes, mast cells, tissue macrophages, and antigen-presenting cells—that arise predominantly from the bone marrow and may circulate or reside in connective tissues.
Granulocytes mount phagocytic responses that engulf and destroy bacteria. These are accompanied by the release of reactive oxygen species and other mediators into adjacent tissues that may cause tissue injury.
Mast cells and basophils underpin allergic reactions to substances that would be treated as innocuous by nonallergic individuals.
A variety of soluble mediators orchestrate the development of immunologic effector cells and their subsequent immune and inflammatory reactions.
Innate immunity represents an evolutionarily conserved, primitive response to stereotypical microbial components.
Acquired immunity is slower to develop than innate immunity, but long-lasting and more effective.
Genetic rearrangements endow B and T lymphocytes with a vast array of receptors capable of recognizing billions of foreign antigens.
Self-reactive lymphocytes are normally deleted; a failure of this process leads to autoimmune disease. Disease can also result from abnormal function or development of granulocytes and lymphocytes. In these latter cases, deficient immune responses to microbial threats usually result.
Inflammatory responses occur in response to infection or injury, and serve to resolve the threat, although they may cause damage to otherwise healthy tissue. A number of chronic diseases reflect excessive inflammatory responses that persist even once the threat is controlled, or are triggered by stimuli that healthy individuals would not respond to.
For all questions, select the single best answer unless otherwise directed.
1. In an experiment, a scientist treats a group of mice with an antiserum that substantially depletes the number of circulating neutrophils. Compared with untreated control animals, the mice with reduced numbers of neutrophils were found to be significantly more susceptible to death induced by bacterial inoculation. The increased mortality can be ascribed to a relative deficit in which of the following?
A. Acquired immunity
D. Granulocyte/macrophage colony stimulating factor (GM-CSF)
2. A 20-year-old college student comes to the student health center in April complaining of runny nose and congestion, itchy eyes, and wheezing. She reports that similar symptoms have occurred at the same time each year, and that she obtains some relief from over-the-counter antihistamine drugs, although they make her too drowsy to study. Her symptoms can most likely be attributed to inappropriate synthesis of which of the following antibodies specific for tree pollen?
3. If a nasal biopsy were performed on the patient described in Question 2 while symptomatic, histologic examination of the tissue would most likely reveal degranulation of which of the following cell types?
A. Dendritic cells
E. Mast cells
4. A biotechnology company is working to design a new therapeutic strategy for cancer that involves triggering an enhanced immune response to cellular proteins that are mutated in the disease. Which of the following immune cells or processes will most likely not be required for a successful therapy?
A. Cytotoxic T cells
B. Antigen presentation in the context of MHC-II
C. Proteosomal degradation
D. Gene rearrangements producing T cell receptors
E. The immune synapse
5. The ability of the blood to phagocytose pathogens and mount a respiratory burst is increased by
A. interleukin-2 (IL-2)
B. granulocyte colony-stimulating factor (G-CSF)
D. interleukin-4 (IL-4)
E. interleukin-5 (IL-5)
6. Cells responsible for innate immunity are activated most commonly by
C. carbohydrate sequences in bacterial cell walls
7. A patient suffering from an acute flare in his rheumatoid arthritis undergoes a procedure where fluid is removed from his swollen and inflamed knee joint. Biochemical analysis of the inflammatory cells recovered from the removed fluid would most likely reveal a decrease in which of the following proteins?
A. Interleukin 1
B. Tumor necrosis factor-α
C. Nuclear factor-κB
E. von Willbrand factor
Delibro G: The Robin Hood of antigen presentation. Science 2004;302:485.
Delves PJ, Roitt IM: The immune system. (Two parts.) N Engl J Med 2000;343:37,108.
Dhainaut J-K, Thijs LG, Park G (editors): Septic Shock. WB Saunders, 2000.
Ganz T: Defensins and host defense. Science 1999;286:420.
Karin M, Ben-Neriah Y: Phosphorylation meets ubiquitination: the control of NF-κB activity. Annu Rev Immunol 2000; 18:621.
Samstein B, Emond JC: Liver transplant from living related donors. Annu Rev Med 2001;52:147.
Singer AJ, Clark RAF: Cutaneous wound healing. N Engl J Med 1999;341:738
Tedder TF, Steeber DA, Chen A, et al: The selectins: Vascular adhesion molecules. FASEB J 1995;9:866.
Tilney NL: Transplant: From Myth to Reality. Yale University Press, 2003.
Walport MJ: Complement. (Two parts.) N Engl J Med 2001;344:1058, 1140.