免疫学（英文版）：Chapter 20

a73 Organ-Specific Autoimmune Diseases a73 Systemic Autoimmune Diseases a73 Animal Models for Autoimmune Diseases a73 Evidence Implicating the CD4 + T Cell, MHC, and TCR in Autoimmunity a73 Proposed Mechanisms for Induction of Autoimmunity a73 Treatment of Autoimmune Diseases Autoimmunity ???? ?? ??? ???? ???????, ???? ??????? realized that the immune system could go awry and, instead of reacting against foreign antigens, could focus its attack on self-antigens. He termed this con- dition “horror autotoxicus.” We now understand that, while mechanisms of self-tolerance normally protect an individual from potentially self-reactive lymphocytes, there are failures. They result in an inappropriate response of the immune system against self-components termed autoimmunity. In the 1960s, it was believed that all self-reactive lymphocytes were eliminated during their development in the bone mar- row and thymus and that a failure to eliminate these lym- phocytes led to autoimmune consequences. Since the late 1970s, a broad body of experimental evidence has countered that belief, revealing that not all self-reactive lymphocytes are deleted during T-cell and B-cell maturation. Instead, normal healthy individuals have been shown to possess ma- ture, recirculating, self-reactive lymphocytes. Since the pres- ence of these self-reactive lymphocytes in the periphery does not inevitably result in autoimmune reactions, their activity must be regulated in normal individuals through clonal anergy or clonal suppression. A breakdown in this regulation can lead to activation of self-reactive clones of T or B cells, generating humoral or cell-mediated responses against self- antigens. These reactions can cause serious damage to cells and organs, sometimes with fatal consequences. Sometimes the damage to self-cells or organs is caused by antibodies; in other cases, T cells are the culprit. For exam- ple, a common form of autoimmunity is tissue injury by mechanisms similar to type II hypersensitivity reactions. As Chapter 16 showed, type II hypersensitivity reactions in- volve antibody-mediated destruction of cells. Autoimmune hemolytic anemia is an excellent example of such an autoim- mune disease. In this disease, antigens on red blood cells are recognized by auto-antibodies, which results in the destruc- tion of the blood cells, which in turn results in anemia. Auto- antibodies are also the major offender in Hashimoto’s thy- roiditis, in which antibodies reactive with tissue-specific antigens such as thyroid peroxidase and thyroglobulin cause severe tissue destruction. Other autoimmune diseases that involve auto-antibodies are listed in Table 20-1. Many autoimmune diseases are characterized by tissue destruction mediated directly by T cells. A well-known ex- ample is rheumatoid arthritis, in which self-reactive T cells attack the tissue in joints, causing an inflammatory response that results in swelling and tissue destruction. Other exam- ples include insulin-dependent diabetes mellitus and multi- ple sclerosis (see Table 20-1). This chapter describes some common human autoim- mune diseases. These can be divided into two broad cate- gories: organ-specific and systemic autoimmune disease (Table 20-1). Such diseases affect 5%–7% of the human pop- ulation, often causing chronic debilitating illnesses. Several experimental animal models used to study autoimmunity and various mechanisms that may contribute to induction of autoimmune reactions also are described. Finally, current and experimental therapies for treating autoimmune dis- eases are described. Organ-Specific Autoimmune Diseases In an organ-specific autoimmune disease, the immune re- sponse is directed to a target antigen unique to a single organ or gland, so that the manifestations are largely limited to that organ. The cells of the target organs may be damaged di- chapter 20 Kidney Biopsy from Goodpasture’s Syndrone E rectly by humoral or cell-mediated effector mechanisms. Alternatively, the antibodies may overstimulate or block the normal function of the target organ. Some Autoimmune Diseases Are Mediated by Direct Cellular Damage Autoimmune diseases involving direct cellular damage occur when lymphocytes or antibodies bind to cell-membrane an- tigens, causing cellular lysis and/or an inflammatory response in the affected organ. Gradually, the damaged cellular struc- ture is replaced by connective tissue (scar tissue), and the func- tion of the organ declines. This section briefly describes a few examples of this type of autoimmune disease. HASHIMOTO’S THYROIDITIS In Hashimoto’s thyroiditis, which is most frequently seen in middle-aged women, an individual produces auto-antibodies and sensitized T H 1 cells specific for thyroid antigens. The DTH response is characterized by an intense infiltration of the thyroid gland by lymphocytes, macrophages, and plasma cells, which form lymphocytic follicles and germinal centers (Figure 20-1). The ensuing inflammatory response causes a goiter, or visible enlargement of the thyroid gland, a physio- logical response to hypothyroidism. Antibodies are formed to a number of thyroid proteins, including thyroglobulin and thyroid peroxidase, both of which are involved in the uptake of iodine. Binding of the auto-antibodies to these proteins interferes with iodine uptake and leads to decreased produc- tion of thyroid hormones (hypothyroidism). AUTOIMMUNE ANEMIAS Autoimmune anemias include pernicious anemia, autoim- mune hemolytic anemia, and drug-induced hemolytic ane- mia. Pernicious anemia is caused by auto-antibodies to intrin- sic factor, a membrane-bound intestinal protein on gastric parietal cells. Intrinsic factor facilitates uptake of vitamin B 12 Autoimmunity CHAPTER 20 463 TABLE 20-1 Some autoimmune diseases in humans Disease Self-antigen Immune response ORGAN-SPECIFIC AUTOIMMUNE DISEASES Addison’s disease Adrenal cells Auto-antibodies Autoimmune hemolytic anemia RBC membrane proteins Auto-antibodies Goodpasture’s syndrome Renal and lung basement membranes Auto-antibodies Graves’ disease Thyroid-stimulating hormone receptor Auto-antibody (stimulating) Hashimoto’s thyroiditis Thyroid proteins and cells T DTH cells, auto-antibodies Idiopathic thrombocyopenia purpura Platelet membrane proteins Auto-antibodies Insulin-dependent diabetes mellitus Pancreatic beta cells T DTH cells, auto-antibodies Myasthenia gravis Acetylcholine receptors Auto-antibody (blocking) Myocardial infarction Heart Auto-antibodies Pernicious anemia Gastric parietal cells; intrinsic factor Auto-antibody Poststreptococcal glomerulonephritis Kidney Antigen-antibody complexes Spontaneous infertility Sperm Auto-antibodies SYSTEMIC AUTOIMMUNE DISEASES Ankylosing sponkylitis Vertebrae Immune complexes Multiple sclerosis Brain or white matter T H 1 cells and T C cells, auto-antibodies Rheumatoid arthritis Connective tissue, IgG Auto-antibodies, immune complexes Scleroderma Nuclei, heart, lungs, gastrointestinal tract, kidney Auto-antibodies Sjogren’s syndrome Salivary gland, liver, kidney, thyroid Auto-antibodies Systemic lupus erythematosus (SLE) DNA, nuclear protein, RBC and platelet membranes Auto-antibodies, immune complexes from the small intestine. Binding of the auto-antibody to intrinsic factor blocks the intrinsic factor–mediated absorp- tion of vitamin B 12 . In the absence of sufficient vitamin B 12 , which is necessary for proper hematopoiesis, the number of functional mature red blood cells decreases below normal. Pernicious anemia is treated with injections of vitamin B 12 , thus circumventing the defect in its absorption. An individual with autoimmune hemolytic anemia makes auto-antibody to RBC antigens, triggering complement- mediated lysis or antibody-mediated opsonization and phago- cytosis of the red blood cells. One form of autoimmune ane- mia is drug-induced: when certain drugs such as penicillin or the anti-hypertensive agent methyldopa interact with red blood cells, the cells become antigenic. The immunodiag- nostic test for autoimmune hemolytic anemias generally involves a Coombs test, in which the red cells are incubated with an anti–human IgG antiserum. If IgG auto-antibodies are present on the red cells, the cells are agglutinated by the antiserum. GOODPASTURE’S SYNDROME In Goodpasture’s syndrome, auto-antibodies specific for certain basement-membrane antigens bind to the basement membranes of the kidney glomeruli and the alveoli of the lungs. Subsequent complement activation leads to direct cel- lular damage and an ensuing inflammatory response medi- ated by a buildup of complement split products. Damage to the glomerular and alveolar basement membranes leads to progressive kidney damage and pulmonary hemorrhage. Death may ensue within several months of the onset of symptoms. Biopsies from patients with Goodpasture’s syn- drome stained with fluorescent-labeled anti-IgG and anti- C3b reveal linear deposits of IgG and C3b along the base- ment membranes (Figure 20-2). INSULIN-DEPENDENT DIABETES MELLITUS A disease afflicting 0.2% of the population, insulin-dependent diabetes mellitus (IDDM) is caused by an autoimmune attack on the pancreas. The attack is directed against special- ized insulin-producing cells (beta cells) that are located in spherical clusters, called the islets of Langerhans, scattered throughout the pancreas. The autoimmune attack destroys beta cells, resulting in decreased production of insulin and consequently increased levels of blood glucose. Several factors are important in the destruction of beta cells. First, activated CTLs migrate into an islet and begin to attack the insulin- producing cells. Local cytokine production during this 464 PART IV The Immune System in Health and Disease (a) (b) FIGURE 20-1 Photomicrographs of (a) normal thyroid gland show- ing a follicle lined by cuboidal follicular epithelial cells and (b) gland in Hashimoto’s thyroiditis showing intense lymphocyte infiltration. [From Web Path, courtesy of E. C. Klatt, University of Utah.] FIGURE 20-2 Fluorescent anti-IgG staining of a kidney biopsy from a patient with Goodpasture’s syndrome reveals linear deposits of auto-antibody along the basement membrane. [From Web Path, courtesy of E. C. Klatt, University of Utah.] response includes IFN-H9253, TNF-H9251, and IL-1. Auto-antibody production can also be a contributing factor in IDDM. The first CTL infiltration and activation of macrophages, fre- quently referred to as insulitis (Figure 20-3), is followed by cytokine release and the presence of auto-antibodies, which leads to a cell-mediated DTH response. The subsequent beta-cell destruction is thought to be mediated by cytokines released during the DTH response and by lytic enzymes released from the activated macrophages. Auto-antibodies to beta cells may contribute to cell destruction by facilitating either antibody-plus-complement lysis or antibody-dependent cell-mediated cytotoxicity (ADCC). The abnormalities in glucose metabolism that are caused by the destruction of islet beta cells result in serious meta- bolic problems that include ketoacidosis and increased urine production. The late stages of the disease are often character- ized by atherosclerotic vascular lesions—which in turn cause gangrene of the extremities due to impeded vascular flow— renal failure, and blindness. If untreated, death can result. The most common therapy for diabetes is daily administra- tion of insulin. This is quite helpful in managing the disease, but, because sporadic doses are not the same as metabolically regulated continuous and controlled release of the hormone, periodically injected doses of insulin do not totally alleviate the problems caused by the disease. Another complicating feature of diabetes is that the disorder can go undetected for several years, allowing irreparable loss of pancreatic tissue to occur before treatment begins. Some Autoimmune Diseases Are Mediated by Stimulating or Blocking Auto-Antibodies In some autoimmune diseases, antibodies act as agonists, binding to hormone receptors in lieu of the normal ligand and stimulating inappropriate activity. This usually leads to an overproduction of mediators or an increase in cell growth. Conversely, auto-antibodies may act as antagonists, binding hormone receptors but blocking receptor function. This gen- erally causes impaired secretion of mediators and gradual atrophy of the affected organ. GRAVES’ DISEASE The production of thyroid hormones is carefully regulated by thyroid-stimulating hormone (TSH), which is produced by the pituitary gland. Binding of TSH to a receptor on thyroid cells activates adenylate cyclase and stimulates the synthesis of two thyroid hormones, thyroxine and triiodothyronine. A patient with Graves’ disease produces auto-antibodies that bind the receptor for TSH and mimic the normal action of TSH, activating adenylate cyclase and resulting in produc- tion of the thyroid hormones. Unlike TSH, however, the auto- antibodies are not regulated, and consequently they over- stimulate the thyroid. For this reason these auto-antibodies are called long-acting thyroid-stimulating (LATS) antibod- ies (Figure 20-4). MYASTHENIA GRAVIS Myasthenia gravis is the prototype autoimmune disease mediated by blocking antibodies. A patient with this disease produces auto-antibodies that bind the acetylcholine recep- tors on the motor end-plates of muscles, blocking the normal binding of acetylcholine and also inducing complement- mediated lysis of the cells. The result is a progressive weaken- ing of the skeletal muscles (Figure 20-5). Ultimately, the anti- bodies destroy the cells bearing the receptors. The early signs of this disease include drooping eyelids and inability to retract the corners of the mouth, which gives the appearance of snarling. Without treatment, progressive weakening of the Autoimmunity CHAPTER 20 465 (a) (b) FIGURE 20-3 Photomicrographs of an islet of Langerhans (a) in pancreas from a normal mouse and (b) one in pancreas from a mouse with a disease resembling insulin-dependent diabetes mellitus. Note the lymphocyte infiltration into the islet (insulitis) in (b). [From M. A. Atkinson and N. K. Maclaren, 1990, Sci. Am. 263(1):62.] muscles can lead to severe impairment of eating as well as problems with movement. However, with appropriate treat- ment, this disease can be managed quite well and afflicted individuals can lead a normal life. Systemic Autoimmune Diseases In systemic autoimmune diseases, the response is directed toward a broad range of target antigens and involves a num- ber of organs and tissues. These diseases reflect a general de- fect in immune regulation that results in hyperactive T cells and B cells. Tissue damage is widespread, both from cell- mediated immune responses and from direct cellular dam- age caused by auto-antibodies or by accumulation of im- mune complexes. Systemic Lupus Erythematosus Attacks Many Tissues One of the best examples of a systemic autoimmune disease is systemic lupus erythematosus (SLE), which typically appears in women between 20 and 40 years of age; the ratio of female to male patients is 10:1. SLE is characterized by fever, weak- ness, arthritis, skin rashes, pleurisy, and kidney dysfunction (Figure 20-6). Lupus is more frequent in African-American and Hispanic women than in Caucasians, although it is not known why this is so. Affected individuals may produce auto- antibodies to a vast array of tissue antigens, such as DNA, his- tones, RBCs, platelets, leukocytes, and clotting factors; inter- action of these auto-antibodies with their specific antigens produces various symptoms. Auto-antibody specific for RBCs and platelets, for example, can lead to complement-mediated lysis, resulting in hemolytic anemia and thrombocytopenia, respectively. When immune complexes of auto-antibodies with various nuclear antigens are deposited along the walls of 466 PART IV The Immune System in Health and Disease STIMULATING AUTO-ANTIBODIES (Graves’ disease) Pituitary gland TSH TSH receptor Thyroid cell Stimulates hormone synthesis Auto-antibody to receptor Regulated production of thyroid hormones Unregulated overproduction of thyroid hormones Negative feedback control Stimulates hormone synthesis BLOCKING AUTO-ANTIBODIES (Myasthenia gravis) Nerve Nerve Acetylcholine Muscle cell AChR Auto-antibody to AChR Muscle activation Muscle activation inhibited FIGURE 20-4 In Graves’ disease, binding of auto-antibodies to the receptor for thyroid-stimulating hormone (TSH) induces unregu- lated activation of the thyroid, leading to overproduction of the thy- roid hormones (purple dots). FIGURE 20-5 In myasthenia gravis, binding of auto-antibodies to the acetylcholine receptor (right) blocks the normal binding of acetyl- choline (burgandy dots) and subsequent muscle activation (left). In addition, the anti-AChR auto-antibody activates complement, which damages the muscle end-plate; the number of acetylcholine receptors declines as the disease progresses. AChR = acetylcholine receptor. small blood vessels, a type III hypersensitive reaction devel- ops. The complexes activate the complement system and generate membrane-attack complexes and complement split products that damage the wall of the blood vessel, resulting in vasculitis and glomerulonephritis. Excessive complement activation in patients with severe SLE produces elevated serum levels of the complement split products C3a and C5a, which may be three to four times higher than normal. C5a induces increased expression of the type 3 complement receptor (CR3) on neutrophils, facilitat- ing neutrophil aggregation and attachment to the vascular endothelium. As neutrophils attach to small blood vessels, the number of circulating neutrophils declines (neutropenia) and various occlusions of the small blood vessels develop (vasculi- tis). These occlusions can lead to widespread tissue damage. Laboratory diagnosis of SLE focuses on the characteristic antinuclear antibodies, which are directed against double- stranded or single-stranded DNA, nucleoprotein, histones, and nucleolar RNA. Indirect immunofluorescent staining with serum from SLE patients produces various characteris- tic nucleus-staining patterns. Multiple Sclerosis Attacks the Central Nervous System Multiple sclerosis (MS) is the most common cause of neuro- logic disability associated with disease in Western countries. The symptoms may be mild, such as numbness in the limbs, or severe, such as paralysis or loss of vision. Most people with MS are diagnosed between the ages of 20 and 40. Individuals with this disease produce autoreactive T cells that participate in the formation of inflammatory lesions along the myelin sheath of nerve fibers. The cerebrospinal fluid of patients with active MS contains activated T lymphocytes, which infiltrate the brain tissue and cause characteristic inflamma- tory lesions, destroying the myelin. Since myelin functions to insulate the nerve fibers, a breakdown in the myelin sheath leads to numerous neurologic dysfunctions. Epidemiological studies indicate that MS is most com- mon in the Northern hemisphere and, interestingly, in the United States. Populations who live north of the 37th parallel have a prevalence of 110–140 cases per 100,000, while those who live south of the 37th parallel show a prevalence of 57–78 per 100,000. And individuals from south of the 37th parallel who move north assume a new risk if the move occurs before 15 years of age. These provocative data suggest that there is an environmental component of the risk of con- tracting MS. This is not the entire story, however, since genetic influences also are important. While the average per- son in the United States has about one chance in 1000 of developing MS, close relatives of people with MS, such as children or siblings, have 1 chance in 50 to 100 of developing MS. The identical twin of a person with MS has a 1 in 3 chance of developing the disease. These data point strongly to the genetic component of the disease. And, as is described in the Clinical Focus of this chapter, MS affects women two to three times more frequently than men. The cause of MS, like most autoimmune diseases, is not well understood. However, there are some suggestions that infection by certain viruses may predispose a person to MS. Certainly some viruses can cause demyelinating diseases, and it is tempting to speculate that virus infection plays a signifi- cant role in MS, but at present there is no definitive data implicating a particular virus. Rheumatoid Arthritis Attacks Joints Rheumatoid arthritis is a common autoimmune disorder, most often affecting women from 40 to 60 years old. The major symptom is chronic inflammation of the joints, although the hematologic, cardiovascular, and respiratory systems are also frequently affected. Many individuals with rheumatoid arthritis produce a group of auto-antibodies called rheumatoid factors that are reactive with determi- nants in the Fc region of IgG. The classic rheumatoid factor is an IgM antibody with that reactivity. Such auto-antibodies bind to normal circulating IgG, forming IgM-IgG complexes that are deposited in the joints. These immune complexes can activate the complement cascade, resulting in a type III hypersensitive reaction, which leads to chronic inflammation of the joints. Animal Models for Autoimmune Diseases Animal models for autoimmune diseases have contributed valuable insights into the mechanism of autoimmunity, to Autoimmunity CHAPTER 20 467 FIGURE 20-6 Characteristic “butterfly” rash over the cheeks of a young girl with systemic lupus erythematosus. [From L. Steinman, 1993, Sci. Am. 269(3):80.] our understanding of autoimmunity in humans, and to potential treatments. Autoimmunity develops spontaneously in certain inbred strains of animals and can also be induced by certain experimental manipulations (Table 20-2). Autoimmunity Can Develop Spontaneously in Animals A number of autoimmune diseases that develop sponta- neously in animals exhibit important clinical and pathologic similarities to certain autoimmune diseases in humans. Cer- tain inbred mouse strains have been particularly valuable models for illuminating the immunologic defects involved in the development of autoimmunity. New Zealand Black (NZB) mice and F 1 hybrids of NZB and New Zealand White (NZW) mice spontaneously develop autoimmune diseases that closely resemble systemic lupus ery- thematosus. NZB mice spontaneously develop autoimmune hemolytic anemia between 2 and 4 months of age, at which time various auto-antibodies can be detected, including anti- bodies to erythrocytes, nuclear proteins, DNA, and T lym- phocytes. F 1 hybrid animals develop glomerulonephritis from immune-complex deposits in the kidney and die prematurely by 18 months. As in human SLE, the incidence of autoimmu- nity in the (NZB H11003 NZW)F 1 hybrids is greater in females. An accelerated and severe form of systemic autoimmune disease resembling systemic lupus erythematosus develops in a mouse strain called MRL/lpr/lpr. These mice are homozy- gous for a gene called lpr, which has been identified as a defective fas gene. The fas-gene product is a cell-surface pro- tein belonging to the TNF family of cysteine-rich membrane receptors (see Figure 12-6d). When the normal Fas protein interacts with its ligand, it transduces a signal that leads to apoptotic death of the Fas-bearing cells. This mechanism may operate in destruction of target cells by some CTLs (see Figure 14-9). Fas is known also to be essential in the death of hyperactivated peripheral CD4 + cells. Normally, when ma- ture peripheral T cells become activated, they are induced to express both Fas antigen and Fas ligand, When Fas-bearing cells come into contact with a neighboring activated cell bear- ing Fas ligand, the Fas-bearing cell is induced to die. It is also possible that Fas ligand can engage Fas from the same cell, inducing a cellular suicide. In the absence of Fas, mature peripheral T cells do not die, and these activated cells con- tinue to proliferate and produce cytokines that result in grossly enlarged lymph nodes and spleen. Defects in fas ex- pression similar to that found in the lpr mouse are observed in humans, and these can have severe consequences. However there is no link between fas expression and SLE in humans, which suggests that the lpr mouse may not be a true model for SLE. Another important animal model is the nonobese dia- betic (NOD) mouse, which spontaneously develops a form of diabetes that resembles human insulin-dependent dia- 468 PART IV The Immune System in Health and Disease TABLE 20-2 Experimental animal models of autoimmune diseases Disease Possible human transferred Animal model disease counterpart Inducing antigen by T cells SPONTANEOUS AUTOIMMUNE DISEASES Nonobese diabetic (NOD) Insulin-dependent diabetes Unknown Yes mouse mellitus (IDDM) (NZB H11003 NZW) F 1 mouse Systemic lupus erythematosus (SLE) Unknown Yes Obese-strain chicken Hashimoto’s thyroiditis Thyroglobulin Yes EXPERIMENTALLY INDUCED AUTOIMMUNE DISEASES* Experimental autoimmune Myasthenia gravis Acetylcholine receptor Yes myasthenia gravis (EAMG) Experimental autoimmune Multiple sclerosis (MS) Myelin basic protein (MBP); Yes encephalomyelitis (EAE) proteolipid protien (PLP) Autoimmune arthritis (AA) Rheumatoid arthritis M. tuberculosis (proteoglycans) Yes Experimental autoimmune Hashimoto’s thyroiditis Thyroglobulin Yes thyroiditis (EAT) * These diseases can be induced by injecting appropriate animals with the indicated antigen in complete Freund’s adjuvant. Except for autoimmune arthritis, the antigens used correspond to the self-antigens associated with the human-disease counterpart. Rheumatoid arthritis involves reaction to proteoglycans, which are self-antigens associated with connective tissue. betes mellitus (IDDM). Like the human disease, the NOD mouse disease begins with lymphocytic infiltration into the islets of the pancreas. Also, as in IDDM, there is a strong asso- ciation between certain MHC alleles and the development of diabetes in these mice. Experiments have shown that T cells from diabetic mice can transfer diabetes to nondiabetic re- cipients. For example, when the immune system of normal mice is destroyed by lethal doses of x-rays and then is recon- stituted with an injection of bone-marrow cells from NOD mice, the reconstituted mice develop diabetes. Conversely, when the immune system of still healthy NOD mice is de- stroyed by x-irradiation and then reconstituted with normal bone-marrow cells, the NOD mice do not develop diabetes. Various studies have demonstrated a pivotal role for CD4 + T cells in the NOD mouse, and recent evidence implicates the T H 1 subset in disease development. Several other spontaneous autoimmune diseases have been discovered in animals that have served as models for similar human diseases. Among these are Obese-strain chickens, which develop both humoral and cell-mediated reactivity to thyroglobulin resembling that seen in Hashimoto’s thyroiditis. Autoimmunity Can Be Induced Experimentally in Animals Autoimmune dysfunctions similar to certain human autoim- mune diseases can be induced experimentally in some ani- mals (see Table 20-2). One of the first such animal models was discovered serendipitously in 1973 when rabbits were immunized with acetylcholine receptors purified from elec- tric eels. The animals soon developed muscular weakness similar to that seen in myasthenia gravis. This experimental autoimmune myasthenia gravis (EAMG) was shown to result when antibodies to the acetylcholine receptor blocked mus- cle stimulation by acetylcholine in the synapse. Within a year, this animal model had proved its value with the discovery that auto-antibodies to the acetylcholine receptor were the cause of myasthenia gravis in humans. Experimental autoimmune encephalomyelitis (EAE) is another animal model that has greatly improved under- standing of autoimmunity. This is one of the best-studied models of autoimmune disease. EAE is mediated solely by T cells and can be induced in a variety of species by immu- nization with myelin basic protein (MBP) or proteolipid protein (PLP) in complete Freund’s adjuvant (Figure 20-7). Within 2–3 weeks the animals develop cellular infiltration of the myelin sheaths of the central nervous system, resulting in demyelination and paralysis. Most of the animals die, but others have milder symptoms, and some animals develop a chronic form of the disease that resembles chronic relapsing and remitting MS in humans. Those that recover are resistant to the development of disease from a subsequent injection of MBP and adjuvant. The mouse EAE model provides a system for testing treatments for human MS. For example, because MBP- or PLP-specific T-cell clones are found in the periphery, it is assumed that these clones must have escaped negative selec- tion in the thymus. Recent mouse experiments have suggested that orally administered MBP may make these antigen-specific peripheral T-cell clones self-tolerant. These studies have paved the way for clinical trials in MS patients. Experimental autoimmune thyroiditis (EAT) can be induced in a number of animals by immunizing with thy- roglobulin in complete Freund’s adjuvant. Both humoral anti- bodies and T H 1 cells directed against the thyroglobulin de- velop, resulting in thyroid inflammation. EAT appears to best mimic Hashimoto’s thyroiditis. In contrast to both EAE and EAT, which are induced by immunizing with self-antigens, autoimmune arthritis (AA) is induced by immunizing rats with Mycobacterium tuberculosis in complete Freund’s adju- vant. These animals develop an arthritis whose features are similar to those of rheumatoid arthritis in humans. Evidence Implicating the CD4 + T Cell, MHC, and TCR in Autoimmunity The inappropriate response to self-antigens that character- izes all autoimmune diseases can involve either the humoral or cell-mediated branches of the immune system. Identifying the defects underlying human autoimmune diseases has been difficult; more success has been achieved in characteriz- ing the immune defects in the various animal models. Each of the animal models has implicated the CD4 + T cell as the primary mediator of autoimmune disease. For example, the evidence is quite strong that, in mice, EAE is caused by CD4 + T H 1 cells specific for the immunizing antigen. The disease can be transferred from one animal into another by T cells from animals immunized with either MBP or PLP or by Autoimmunity CHAPTER 20 469 Normal rat MBP + CFA Lymph-node cells + MBP EAE rat (paralysis) Normal ratEAE rat (most die; some recover) MBP-specific T-cell clones FIGURE 20-7 Experimental autoimmune encephalomyelitis (EAE) can be induced in rats by injecting them with myelin basic protein (MBP) in complete Freud’s adjuvant (CFA). MBP-specific T-cell clones can be generated by culturing lymph-node cells from EAE rats with MBP. When these T cells are injected into normal animals, most develop EAE and die, although a few recover. cloned T-cell lines from such animals. It also has been shown that disease can be prevented by treating animals with anti- CD4 antibodies. These data are compelling evidence for the involvement of CD4 in the establishment of EAE. T-cell recognition of antigen, of course, involves a trimol- ecular complex of the T-cell receptor, an MHC molecule, and antigenic peptide (see Figure 9-16). Thus, an individual sus- ceptible to autoimmunity must possess MHC molecules and T-cell receptors capable of binding self-antigens. CD4 + T Cells and T H 1/T H 2 Balance Plays an Important Role in Autoimmunity in Some Animal Models Autoimmune T-cell clones have been obtained from all of the animal models listed in Table 20-2 by culturing lymphocytes from the autoimmune animals in the presence of various T-cell growth factors and by inducing proliferation of spe- cific autoimmune clones with the various autoantigens. For example, when lymph-node cells from EAE rats are cultured in vitro with myelin basic protein (MBP), clones of activated T cells emerge. When sufficient numbers of these MBP- specific T-cell clones are injected intravenously into normal syngeneic animals, the cells cross the blood-brain barrier and induce demyelination; EAE develops very quickly, within 5 days (see Figure 20-7). A similar experimental protocol has been used to isolate T-cell clones specific for thyroglobulin and for M. tuberculosis from EAT and AA animals, respectively. In each case, the T-cell clone induces the experimental autoimmune disease in nor- mal animals. Examination of these T cells has revealed that they bear the CD4 membrane marker. In a number of animal models for autoimmune diseases it has been possible to reverse the autoimmunity by depleting the T-cell population with antibody directed against CD4. For example, weekly injections of anti-CD4 monoclonal antibody abolished the autoimmune symptoms in (NZB H11003 NZW) F 1 mice and in mice with EAE. Most cases of organ-specific autoimmune disease develop as a consequence of self-reactive CD4 + T cells.Analysis of these cells has revealed that the T H 1/T H 2 balance can affect whether autoimmunity develops. T H 1 cells have been implicated in the development of autoimmunity, whereas, in a number of cases, T H 2 cells not only protect against the induction of disease but also against progression of established disease. In EAE, for example, immunohistologic studies revealed the presence of T H 1 cytokines (IL-2, TNF-H9251, and IFN-H9253) in the central ner- vous system tissues at the height of the disease. In addition, the MBP-specific CD4 + T-cell clones generated from animals with EAE, as shown in Figure 20-7, can be separated into T H 1 and T H 2 clones. Experiments have shown that only the T H 1 clones transfer EAE to normal healthy mice, whereas the T H 2 clones not only do not transfer EAE to normal healthy mice but also protect the mice against induction of EAE by subsequent immunization with MBP plus adjuvant. Experiments that assessed the role of various cytokines or cytokine inhibitors on the development of EAE have provided further evidence for the different roles of T H 1 and T H 2 cells in autoimmunity. When mice were injected with IL-4 at the time of immunization with MBP plus adjuvant, the development of EAE was inhibited, whereas administration of IL-12 had the opposite effect, promoting the development of EAE. As noted in Chapter 12, IL-4 promotes development of T H 2 cells and IFN-H9253, in addition to other cytokines such as IL-12, promotes development of T H 1 cells (see Figure 12-12). Thus, the ob- served effects of IL-4 and IL-12 on EAE development are con- sistent with a role for T H 1 cells in the genesis of autoimmunity. Autoimmunity Can Be Associated with the MHC or with Particular T-Cell Receptors Several types of studies have supported an association be- tween expression of a particular MHC allele and susceptibil- ity to autoimmunity, an issue covered in detail in Chapter 7. The strongest association between an HLA allele and an autoimmune disease is seen in ankylosing spondylitis, an inflammatory disease of vertebral joints. Individuals who have HLA-B27 have a 90 times greater likelihood of develop- ing ankylosing spondylitis than individuals with a different HLA-B allele. However, the existence of such an association should not be interpreted to imply that the expression of a particular MHC allele has caused the disease, because the relationship between MHC alleles and development of auto- immune disease is complex. It is interesting to note that, unlike many other autoimmune diseases, 90% of the cases of ankylosing spondylitis are male. The presence of T-cell receptors containing particular V H9251 and V H9252 domains also has been linked to a number of auto- immune diseases, including experimental EAE and its human counterpart, multiple sclerosis. In one approach, T cells spe- cific for various encephalitogenic peptides of MBP were cloned and their T-cell receptors analyzed. For example, T-cell clones were obtained from PL/J mice by culturing their T cells with the acetylated amino-terminal nonapeptide of MBP presented in association with a class II IA u MHC mole- cule. Analysis of the T-cell receptors on these clones revealed a restricted repertoire of V H9251 and V H9252 domains: 100% of the T-cell clones expressed V H9251 4.3, and 80% of the T-cell clones expressed V H9252 8.2. In human autoimmune diseases, evidence for restricted TCR expression has been obtained for both mul- tiple sclerosis and myasthenia gravis. The preferential expres- sion of TCR variable-region genes in these autoimmune T-cell clones suggests that a single epitope might induce the clonal expansion of a small number of pathogenic T cells. Proposed Mechanisms for Induction of Autoimmunity A variety of mechanisms have been proposed to account for the T-cell–mediated generation of autoimmune diseases (Figure 20-8). Evidence exists for each of these mechanisms, 470 PART IV The Immune System in Health and Disease and it is likely that autoimmunity does not develop from a single event but rather from a number of different events. In addition, susceptibility to many autoimmune diseases differs between the two sexes. As noted earlier, Hashimoto’s thyroiditis, systemic lupus erythematosus, multiple sclerosis, rheumatoid arthritis, and scleroderma preferentially affect women. Factors that have been proposed to account for this preferential susceptibility, such as hormonal differences be- tween the sexes and the potential effects of fetal cells in the maternal circulation during pregnancy, are discussed in the Clinical Focus. Release of Sequestered Antigens Can Induce Autoimmune Disease As discussed in Chapter 10, the induction of self-tolerance in T cells results from exposure of immature thymocytes to self- antigens and the subsequent clonal deletion of those that are self-reactive. Any tissue antigens that are sequestered from the circulation, and are therefore not seen by the developing T cells in the thymus, will not induce self-tolerance. Exposure of mature T cells to such normally sequestered antigens at a later time might result in their activation. Myelin basic protein (MBP) is an example of an antigen normally sequestered from the immune system, in this case by the blood-brain barrier. In the EAE model, animals are in- jected directly with MBP, together with adjuvant, under con- ditions that maximize immune exposure. In this type of ani- mal model, the immune system is exposed to sequestered self- antigens under nonphysiologic conditions; however, trauma to tissues following either an accident or a viral or bacterial infection might also release sequestered antigens into the cir- culation. A few tissue antigens are known to fall into this cate- gory. For example, sperm arise late in development and are sequestered from the circulation. However, after a vasectomy, some sperm antigens are released into the circulation and can induce auto-antibody formation in some men. Similarly, the release of lens protein after eye damage or of heart-muscle antigens after myocardial infarction has been shown to lead on occasion to the formation of auto-antibodies. Autoimmunity CHAPTER 20 471 VISUALIZING CONCEPTS T H cell Tissue damage Inflammation and local DTH Ab to self-antigens Activated macrophage Target tissue epithelium IFN-γ T H cell Activated T H cell T H cell CTL T H cell Help Plasma cell B cell T C cell Polyclonal activation T H cell B cell T H cell IL-2 Class II MHC Release of sequestered antigen Inappropriate MHC expression on non-APCs APC with cross-reacting Ag (molecular mimicry) FIGURE 20-8 Proposed mechanisms for inducing autoim- mune responses. Normal thymic selection appears to generate some self-reactive T H cells; abnormalities in this process may generate even more self-reactive T H cells. Activation of these self- reactive T cells in various ways, as well as polyclonal activation of B cells, is thought to induce an autoimmune response, in this case resulting in tissue damage. In all likelihood, several mecha- nisms are involved in each autoimmune disease. [Adapted from V. Kumar et al., 1989, Annu. Rev. Immunol. 7:657.] Data indicate that injection of normally sequestered anti- gens directly into the thymus can reverse the development of tissue-specific autoimmune disease in animal models. For instance, intrathymic injection of pancreatic islet beta cells prevented development of autoimmunity in NOD mice. Moreover, EAE was prevented in susceptible rats by prior injection of MBP directly into the thymus. In these experi- ments, exposure of immature T cells to self-antigens that normally are not present in the thymus presumably led to tolerance to these antigens. 472 PART IV The Immune System in Health and Disease young females. Women tend to have higher levels of CD4 + T cells and signifi- cantly higher levels of serum IgM. In mice, whose gender differences are easier to study, there is a large body of lit- erature documenting gender differences in immune responses. Female mice are much more likely than male mice to de- velop T H 1 responses and, in infections to which pro-inflammatory T H 1 responses are beneficial, are more likely to be re- sistant to the infection. An excellent example is infection by viruses such as vesicular stomatitis virus (VSV), herpes simplex virus (HSV), and Theiler’s mu- rine encephalomyelitis virus (TMEV). Clearance of these viruses is enhanced by T H 1 responses. In some cases, however, a pro-inflammatory response can be deleterious. For example, a T H 1 response to lymphocytic choriomeningitis virus (LCMV) correlates with more severe dis- ease and significant pathology. Thus, female mice are more likely to succumb to infection with LCMV. The fact that gen- der is important in LCMV infection is underscored by experiments demonstrat- ing that castrated male mice behave like females and are more likely to succumb to infection than their un-castrated male littermates. Another disease in which gender plays a role is infection by coxsackie virus Type B-3 (CVB-3), an etiological agent of immune myocarditis. Male mice are much more susceptible to this disease than females. CVB-3 induces a predominant T H 1 response in males, while females, contrary to the situations described above, respond by mounting a protective T H 2 response. The response by females can be altered by injecting them with testosterone, which makes them susceptible to the disease. Ad- ditionally, the male response can be al- tered by injecting them with estradiol, which makes them resistant to the virus. These data in mice are consistent with the possibility that basic differences may well exist between men and women in their responses to pathogens. We must stress, however, that the particular gen- der differences observed in mice may not extend to human populations. How do these gender differences arise? The evidence cited above that estradiol or testosterone can alter the outcome of infection by CVB-3 suggests a critical role for sex hormones. In hu- mans it appears that estrogen on its own does not play a significant role in the eti- ology of either RA or MS, but there are indications that it may be important in SLE. This is suggested by data indicat- ing that estrogen can stimulate auto- antibody production in SLE-prone mice and these effects can be modulated by an anti-estrogenic compound. Such data imply that, at least in mice, estrogen is capable of triggering SLE-like autoimmu- nity. Additionally, androgens such as testosterone clearly play an important role in some autoimmune diseases. Female NOD mice are much more sus- ceptible to spontaneous diabetes, and castration significantly increases the susceptibility of male NOD mice. Fe- male SJL mice are more likely to be sus- ceptible to EAE, a mouse MS-like dis- ease. This indicates that testosterone may well be effective in ameliorating some autoimmune responses and so Of the nearly 9 million individuals in the United States with autoimmune disease, approxi- mately 6.7 million are women. This pre- disposition to autoimmunity is more apparent in some diseases than others. For example, the female:male ratio of individuals who suffer from diseases such as multiple sclerosis (MS) or rheumatoid arthritis (RA) is approxi- mately two or three females to one male, and there are nine women for every one male afflicted with systemic lupus ery- thematosus (SLE). However, these sta- tistics do not tell the entire story, since, in some diseases, MS for example, the severity of the disease can be worse in men than in women. The fact that women are more susceptible to autoim- mune disease has been recognized for several years but the reasons for this in- creased risk are not entirely understood. Here some of the possible explanations are considered. Although it may seem unlikely, con- siderable evidence suggests there are significant gender differences in im- mune responses in both humans and mice. Immunization studies in both spe- cies suggest that females produce a higher titer of antibodies than males. In fact, females in general tend to mount more vigorous immune responses. In humans, this is particularly apparent in CLINICAL FOCUS Why Are Women More Susceptible Than Men to Autoimmunity? Gender Differences in Autoimmune Disease Molecular Mimicry May Contribute to Autoimmune Disease For several reasons, the notion that microbial or viral agents might play a role in autoimmunity is very attractive. It is well accepted that migrant human populations acquire the dis- eases of the area to which they move and that the incidence of autoimmunity has increased dramatically as populations have become more mobile. This, coupled with the fact that a number of viruses and bacteria have been shown to possess Autoimmunity CHAPTER 20 473 can have a profound influence on im- mune responses, as demonstrated in mice by removal of the anterior pituitary: this results in a severe immunosuppres- sion, which can be entirely reversed by treatment with exogenous prolactin. The presence of prolactin receptors on peri- pheral T and B cells in humans is further evidence that this hormone may play a role in regulating immune responses. In fact, some evidence suggests that pro- lactin may tend to turn cells towards T H 1-dominated immune responses. Pregnancy may give us a clue to how sex plays a role in regulating immune response. It is clear that, while women normally mount a normal response to foreign antigens, during pregnancy it is critical that the mother tolerate the fetus (which is, in fact, a foreign graft). This makes it very likely that the female immune system undergoes important modifications during pregnancy. Recall that women normally tend to mount more T H 1-like responses than T H 2 re- sponses. During pregnancy, however, wo- men mount more T H 2-like responses. It is thought that pregnancy-associated lev- els of sex steroids may promote an anti- inflammatory environment. In this re- gard, it is notable that diseases enhanced by T H 2-like responses, such as SLE, which has a strong antibody-mediated component, can be exacerbated during pregnancy, while diseases that involve inflammatory responses, such as RA and MS, sometimes are ameliorated in preg- nant women. Another effect of pregnancy is the presence of fetal cells in the maternal cir- culation (see the description of sclero- derma on page 000). It is known that fetal cells can persist in the maternal cir- culation for decades, so these long-lived fetal cells may play a significant role in the development of autoimmune dis- ease. Furthermore, the exchange of cells during pregnancy is bi-directional (cells of the mother may also appear in the fetal circulatory system), and this has led some to postulate that the presence of mother’s cells in the male circulation could be a contributing factor in autoim- mune disease. In summary, women and men differ significantly in their ability to mount an immune response. Women mount more robust immune responses, and these responses tend to be more T H 1-like. It has been reported that estrogen is im- munostimulatory; this may be due, in part, to the ability of the hormone to reg- ulate specific gene expression through the estrogen receptor. Furthermore, the incidence of autoimmune diseases is sharply higher in women than in men. These observations have generated the compelling hypothesis that the tendency of females to mount more T H 1-like re- sponses may, in part, explain differences in susceptibility to autoimmunity. Since this type of response is pro-inflammatory, it may enhance the development of autoimmunity. Whether the bias towards a T H 1 response is due to differences in sex steroids between males and females is less certain, but surely, in the next sev- eral years, experiments that explore this idea are likely to be pursued vigorously. NOTE: The data discussed in this Clinical Focus were extracted from a let- ter to Science (C. C. Whitacre, S. C. Rein- gold, and P. A. O’Looney, 1999, Science 283:1277) from the Task Force on Gender, MS, and Autoimmunity, a group convened by the National Multiple Scle- rosis Society to begin a dialog on issues of gender and autoimmune disease. Science also has established a Web site (http://www.sciencemag.org/feature/ data/983519.shl) that contains more de- tailed data concerning autoimmunity and gender. may be protective against several auto- immune diseases, including MS, dia- betes, SLE, and Sjogren’s syndrome. Why do sex steroids affect immune responses? This is not well understood, but it is likely that these hormones, which circulate throughout the body, alter immune responses by altering patterns of gene expression. The sex steroids, a highly lipophilic group of compounds, function by passing through the cell membrane and binding a cytoplasmic receptor. Each hormone has a cognate receptor and binding of hormone to re- ceptor leads to the activation or, in some instances, repression of gene expression. This is mediated by the binding of the receptor/hormone complex receptor to a specific DNA sequence. Thus, estro- gen enters a cell, binds to the estrogen receptor, and induces the binding of the estrogen receptor to a specific DNA sequence, which in turn results in the modulation of transcription. Therefore, in cells that contain hormone receptors, sex hormones can regulate gene expres- sion, and it is highly likely that sex steroids play an important role in the immune system through their receptors. Whether various cells of the immune sys- tem contain hormone receptors is not known at present; to understand how sex hormones mediate immune responses, clearly we must determine which cells express which hormone receptors. Hormonal effects on immune re- sponses may not be limited to steroidal sex hormones. Prolactin, a hormone that is expressed in higher levels in women than in men, is not a member of the lipophilic sex steroid family that includes estrogen, progesterone, and testoster- one. But prolactin secretion (by the ante- rior pituitary) is stimulated by estrogen, thus explaining the higher levels of pro- lactin in women and the very high levels observed during pregnancy. Prolactin A U : s upply 5e page r e f ( or delet e this) antigenic determinants that are identical or similar to normal host-cell components led Michael Oldstone to propose that a pathogen may express a region of protein that resembles a particular self-component in conformation or primary se- quence. Such molecular mimicry appears in a wide variety of organisms (Table 20-3). In one study, 600 different mono- clonal antibodies specific for 11 different viruses were tested to evaluate their reactivity with normal tissue antigens. More than 3% of the virus-specific antibodies tested also bound to normal tissue, suggesting that molecular mimicry is a fairly common phenomenon. Molecular mimicry has been suggested as one mecha- nism that leads to autoimmunity. One of the best examples of this type of autoimmune reaction is post-rabies encephalitis, which used to develop in some individuals who had received the rabies vaccine. In the past, the rabies virus was grown in rabbit brain-cell cultures, and preparations of the vaccine included antigens derived from the rabbit brain cells. In a vaccinated person, these rabbit brain-cell antigens could induce formation of antibodies and activated T cells, which could cross-react with the recipient’s own brain cells, leading to encephalitis. Cross-reacting antibodies are also thought to be the cause of heart damage in rheumatic fever, which can sometimes develop after a Streptococcus infection. In this case, the antibodies are to streptococcal antigens, but they cross-react with the heart muscle. There Is Evidence for Mimicry Between MBP and Viral Peptides Since the encephalitogenic MBP peptides are known, the ex- tent to which they are mimicked by proteins from other organ- isms can be assessed. For example, one MBP peptide (amino acid residues 61–69) is highly homologous with a peptide in the P3 protein of the measles virus (see Table 20-3). In one study, the sequence of another encephalitogenic MBP peptide (66–75) was compared with the known sequences of a large number of viral proteins. This computer analysis revealed sequence homologies between this MBP peptide and a num- ber of peptides from animal viruses, including influenza, poly- oma, adenovirus, Rous sarcoma, Abelson leukemia, polio- myelitis, Epstein-Barr, and hepatitis B viruses. 474 PART IV The Immune System in Health and Disease TABLE 20-3 Molecular mimicry between proteins of infectious organisms and human host proteins Protein* Residue ? Sequence ? Human cytomegalovirus IE279PDPLGRPDED HLA-DR molecule 60 VTELGRPDAE Poliovirus VP 0STTKESRGTT Acetylcholine receptor 176 TV IKESRGTK Papilloma virus E276SLHLESLKDS Insulin receptor 66 VYGLESLKDL Rabies virus glycoprotein 147 T KESLVI IS Insulin receptor 764 N KESLVISE Klebsiella pneumoniae nitrogenase 186 SRQTDREDE HLA-B27 molecule 70 KAQTDREDL Adenovirus 12 E1B 384 L RRGMFRPSQCN H9251-Gliadin 206 L GQGSFRPSQQN Human immunodeficiency virus p24 160 GVETTTPS Human IgG constant region 466 GVETTTPS Measles virus P313LEC IRAL K Corticotropin 18 LEC IRAC K Measles virus P331EISDNLGQE Myelin basic protein 61 EISFKLGQE *In each pair, the human protein is listed second. The proteins in each pair have been shown to exhibit immunologic cross-reactivity. ? Each number indicates the position on the intact protein of the amino-terminal amino acid in the listed sequence. ? Amino acid residues are indicated by single-letter code. Identical residues are shown in blue. SOURCE: Adapted from M. B. A. Oldstone, 1987, Cell 50:819. One peptide from the polymerase enzyme of the hepati- tis B virus was particularly striking, exhibiting 60% homol- ogy with a sequence in the encephalitogenic MBP peptide. To test the hypothesis that molecular mimicry can generate autoimmunity, rabbits were immunized with this hepatitis B virus peptide. The peptide was shown to induce both the for- mation of antibody and the proliferation of T cells that cross- reacted with MBP; in addition, central nervous system tissue from the immunized rabbits showed cellular infiltration characteristic of EAE. These findings suggest that infection with certain viruses expressing epitopes that mimic sequestered self-components, such as myelin basic protein, may induce autoimmunity to those components. Susceptibility to this type of autoimmunity may also be influenced by the MHC haplotype of the individ- ual, since certain class I and class II MHC molecules may be more effective than others in presenting the homologous pep- tide for T-cell activation. Another particularly compelling example of molecular mimicry comes from studies of herpes stromal keratinitis (HSK). In these studies, investigators showed that prior in- fection of mice with herpes simplex virus Type 1 leads to a disease known as herpes stromal keratinitis (HSK), an auto- immune-like disease in which T cells specific for a particular viral peptide attack corneal tissue, thus causing blindness. These data demonstrated very clearly that a particular epi- tope of HSV-1 is responsible for the disease and that mutant strains of HSV-1 that lack this epitope do not cause HSK. The data provide strong evidence for molecular mimicry in the development of a particular autoimmune disease. Inappropriate Expression of Class II MHC Molecules Can Sensitize Autoreactive T Cells The pancreatic beta cells of individuals with insulin-dependent diabetes mellitus (IDDM) express high levels of both class I and class II MHC molecules, whereas healthy beta cells ex- press lower levels of class I and do not express class II at all. Similarly, thyroid acinar cells from those with Graves’ disease have been shown to express class II MHC molecules on their membranes. This inappropriate expression of class II MHC molecules, which are normally expressed only on antigen- presenting cells, may serve to sensitize T H cells to peptides derived from the beta cells or thyroid cells, allowing activa- tion of B cells or T C cells or sensitization of T H 1 cells against self-antigens. Other evidence suggests that certain agents can induce some cells that should not express class II MHC molecules to express them. For example, the T-cell mitogen phytohemag- glutinin (PHA) has been shown to induce thyroid cells to express class II molecules. In vitro studies reveal that IFN-H9253 also induces increases in class II MHC molecules on a wide variety of cells, including pancreatic beta cells, intestinal epithelial cells, melanoma cells, and thyroid acinar cells. It was hypothesized that trauma or viral infection in an organ may induce a localized inflammatory response and thus in- creased concentrations of IFN-H9253 in the affected organ. If IFN- H9253induces class II MHC expression on non-antigen-presenting cells, inappropriate T H -cell activation might follow, with autoimmune consequences. It is noteworthy that SLE pa- tients with active disease have higher serum titers of IFN-H9253 than patients with inactive disease. These data suggested that the increase in IFN-H9253 in these patients may lead to inappro- priate expression of class II MHC molecules and thus to T-cell activation against a variety of autoantigens. An interesting transgenic mouse system implicates IFN-H9253 and inappropriate class II MHC expression in autoim- munity. In this system, an IFN-H9253 transgene was genetically engineered with the insulin promoter, so that the transgenic mice secreted IFN-H9253 from their pancreatic beta cells (Figure 20-9a). Since IFN-H9253 up-regulates class II MHC expression, these transgenic mice also expressed class II MHC molecules on their pancreatic beta cells. The mice developed diabetes, which was associated with cellular infiltration of lympho- cytes and inflammatory cells like the infiltration seen in auto- immune NOD mice and in patients with insulin-dependent diabetes mellitus (Figure 20-9b). Although inappropriate class II MHC expression on pan- creatic beta cells may be involved in the autoimmune reaction in these transgenic mice, other factors also may play a role. For example, IFN-H9253 is known to induce production of several other cytokines, including IL-1 and TNF. Therefore, the devel- opment of autoimmunity in this transgenic system may in- volve antigen presentation by class II MHC molecules on pan- creatic beta cells together with a co-stimulatory signal, such as IL-1, that may activate self-reactive T cells. There is also some evidence to suggest that IL-1, IFN-H9253, and TNF may directly impair the secretory function of human beta cells. Polyclonal B-Cell Activation Can Lead to Autoimmune Disease A number of viruses and bacteria can induce nonspecific poly- clonal B-cell activation. Gram-negative bacteria, cytomega- lovirus, and Epstein-Barr virus (EBV) are all known to be such polyclonal activators, inducing the proliferation of numer- ous clones of B cells that express IgM in the absence of T H cells. If B cells reactive to self-antigens are activated by this mechanism, auto-antibodies can appear. For instance, during infectious mononucleosis, which is caused by EBV, a variety of auto-antibodies are produced, including auto- antibodies reactive to T and B cells, rheumatoid factors, and antinuclear antibodies. Similarly, lymphocytes from patients with SLE produce large quantities of IgM in culture, suggest- ing that they have been polyclonally activated. Many AIDS patients also show high levels of nonspecific antibody and auto-antibodies to RBCs and platelets. These patients are often coinfected with other viruses such as EBV and cyto- megalovirus, which may induce the polyclonal B-cell activa- tion that results in auto-antibody production. Autoimmunity CHAPTER 20 475 Treatment of Autoimmune Diseases Ideally, treatment for autoimmune diseases should be aimed at reducing only the autoimmune response while leaving the rest of the immune system intact. To date, this ideal has not been reached. Current therapies for autoimmune diseases are not cures but merely palliatives, aimed at reducing symptoms to provide the patient with an acceptable quality of life. For the most part, these treatments provide nonspecific suppression of the immune system and thus do not distinguish between a patho- logic autoimmune response and a protective immune re- sponse. Immunosuppressive drugs (e.g., corticosteroids, aza- thioprine, and cyclophosphamide) are often given with the intent of slowing proliferation of lymphocytes. By depressing the immune response in general, such drugs can reduce the severity of autoimmune symptoms. The general reduction in immune responsiveness, however, puts the patient at greater risk for infection or the development of cancer. A somewhat more selective approach employs cyclosporin A or FK506 to treat autoimmunity. These agents block signal transduc- tion mediated by the T-cell receptor; thus, they inhibit only antigen-activated T cells while sparing nonactivated ones. Another therapeutic approach that has produced positive results in some cases of myasthenia gravis is removal of the thymus. Because patients with this disease often have thymic abnormalities (e.g., thymic hyperplasia or thymomas), adult thymectomy often increases the likelihood of remission of symptoms. Patients with Graves’ disease, myasthenia gravis, rheumatoid arthritis, or systemic lupus erythematosus may experience short-term benefit from plasmapheresis. In this process, plasma is removed from a patient’s blood by continuous-flow centrifugation. The blood cells are then re- suspended in a suitable medium and returned to the patient. Plasmapheresis has been beneficial to patients with autoim- mune diseases involving antigen-antibody complexes, which are removed with the plasma. Removal of the complexes, although only temporary, can result in a short-term reduction in symptoms. On the positive side, studies with experimental autoim- mune animal models have provided evidence that it is indeed possible to induce specific immunity to the development of autoimmunity. Several of these approaches are described below and outlined in Figure 20-10. T-Cell Vaccination Is a Possible Therapy The basis for T-cell vaccination as a therapy for some auto- immune diseases came from experiments with the EAE ani- mal model. When rats were injected with low doses (<10 –4 ) of cloned T cells specific for MBP, they did not develop 476 PART IV The Immune System in Health and Disease 3'5' PI/IFN-γ transgene PancreasIFN-γ Cellular infiltration Transgenic mouse developed IDDM (a) (b) IFN-γ gene IFN-γ gene Insulin gene promoter Insulin gene terminator region 3' 5' Insulin gene promoter (PI) Insulin gene terminator region Insulin structural gene Insulin structural gene Poly A FIGURE 20-9 Insulin-dependent diabetes mellitus (IDDM) in transgenic mice. (a) Production of transgenic mice contaIning an IFN-H9253 transgene linked to the insulin promoter (PI). The transgenics, which expressed the PI/IFN-H9253 transgene only in the pancreas, devel- oped symptoms characteristic of IDDM. (b) Pancreatic islets of Langerhans from a normal BALB/c mouse (left) and from PI/IFN-H9253 transgenics at 3 weeks (right) showing infiltration of inflammatory cells. [Part (b) from N. Sarvetnick, 1988, Cell 52:773.] symptoms of EAE. Instead they became resistant to the development of EAE when later challenged with a lethal dose of activated MBP-specific T cells or MBP in adjuvant. Later findings revealed that the efficacy of these autoimmune T-cell clones as a vaccine could be enhanced by crosslinking the cell-membrane components with formaldehyde or glu- taraldehyde. When crosslinked T cells were injected into ani- mals with active EAE, permanent remission of symptoms was observed. The crosslinked T cells apparently elicit regulatory T cells specific for TCR variable-region determinants of the autoimmune clones. Presumably these regulatory T cells act to suppress the autoimmune T cells that mediate EAE. Peptide Blockade of MHC Molecules Can Modulate Autoimmune Responses Identification and sequencing of various autoantigens has led to the development of new approaches to modulate auto- immune T-cell activity. In EAE, for example, the encephalito- genic peptides of MBP have been well characterized. Syn- thetic peptides differing by only one amino acid from their MBP counterpart have been shown to bind to the appropri- ate MHC molecule. Moreover, when sufficient amounts of such a peptide were administered along with the correspond- ing encephalitogenic MBP peptide, the clinical development of EAE was blocked. Presumably, the synthetic peptide acts as a competitor, occupying the antigen-binding cleft on MHC molecules and thus preventing binding of the MBP peptide. In other studies, blocking peptides complexed to soluble class II MHC molecules reversed the clinical progression of EAE in mice, presumably by inducing a state of clonal anergy in the autoimmune T cells. Monoclonal Antibodies May Be Used to Treat Autoimmunity Monoclonal antibodies have been used successfully to treat autoimmune disease in several animal models. For example, a high percentage of (NZB H11003 NZW) F 1 mice given weekly injections of high doses of monoclonal antibody specific for the CD4 membrane molecule recovered from their autoim- mune lupus-like symptoms (Figure 20-11). Similar positive results were observed in NOD mice, in which treatment with an anti-CD4 monoclonal antibody led to disappearance of the lymphocytic infiltration and diabetic symptoms. Because anti-CD4 monoclonal antibodies block or deplete all T H cells, regardless of their specificity, they can threaten the overall immune responsiveness of the recipient. One remedy for this disadvantage is to try to block antigen- activated T H cells only, since these cells are involved in the autoimmune state. To do this, researchers have used mono- clonal antibody directed against the H9251 subunit of the high- affinity IL-2 receptor, which is expressed only by antigen- activated T H cells. Because the IL-2R H9251 subunit is expressed at higher levels on autoimmune T cells, monoclonal anti- body to the H9251 subunit (anti-TAC) might preferentially block autoreactive T cells. This approach was tested in adult rats injected with activated MBP-specific T cells in the presence or absence of anti-TAC. All the control rats died of EAE, whereas six of the nine treated with anti-TAC had no symp- toms, and the symptoms in the other three were mild. Autoimmunity CHAPTER 20 477 IL-2R IL-2 toxin or Anti-IL-2R CD4 Anti-CD4 Anti- MHC Anti- TCR Peptide blockade APC Autoreactive T cell FIGURE 20-10 Some experimental agents for immunointerven- tion in autoimmune disease. 5 0 20 40 Survival, % 60 Treatment started Anti-CD4 Saline control Age, months 80 100 67891011213141518 FIGURE 20-11 Weekly injections of anti-CD4 monoclonal anti- body into (NZB H11003 NZW) F 1 mice exhibiting autoimmune lupus-like symptoms significantly increased their survival rate. [Adapted from D. Wofsy, 1988, Prog. Allergy 45:106.] The association of autoimmune disease with restricted TCR expression in a number of animal models has prompted researchers to see if blockage of the preferred receptors with monoclonal antibody might be therapeutic. Injection of PL/J mice with monoclonal antibody specific for the V H9252 8.2 T-cell receptor prevented induction of EAE by MBP in adjuvant. Even more promising was the finding that the V H9252 8.2 mono- clonal antibody could also reverse the symptoms of auto- immunity in mice manifesting induced EAE (Figure 20-12) and that these mice manifested long-term remission. Clearly, the use of monoclonal antibodies as a treatment for human autoimmune diseases presents exciting possibilities. Similarly, the association of various MHC alleles with auto- immunity (see Table 7-4), as well as the evidence for increased or inappropriate MHC expression in some autoimmune dis- ease, offers the possibility that monoclonal antibodies against appropriate MHC molecules might retard development of autoimmunity. Moreover, since antigen-presenting cells ex- press many different class II MHC molecules, it should theo- retically be possible to selectively block an MHC molecule that is associated with autoimmunity while sparing the others. In one study, injecting mice with monoclonal antibodies to class II MHC molecules before injecting MBP blocked the develop- ment of EAE. If, instead, the antibody was given after the injec- tion of MBP, development of EAE was delayed but not pre- vented. In nonhuman primates, monoclonal antibodies to HLA-DR and HLA-DQ have been shown to reverse EAE. Oral Antigens Can Induce Tolerance When antigens are administered orally, they tend to induce the state of immunologic unresponsiveness called tolerance. For example, as mentioned earlier in this chapter, mice fed MBP do not develop EAE after subsequent injection of MBP. This finding led to a double-blind pilot trial in which 30 indi- viduals with multiple sclerosis were fed either a placebo or 300 mg of bovine myelin every day for a year. The results of this study revealed that T cells specific for MBP were reduced in the myelin-fed group; there also was some suggestion that MS symptoms were reduced in the male recipients (although the reduction fell short of statistical significance) but not in the female recipients. While the results of oral tolerance induction in mice were promising, the data from humans do not appear to be as beneficial. However, the human clinical trials are in the early stages, and it may be that the peptides used so far were not the most effective, or perhaps the doses were not correct. Because of the promise of this approach as shown in animal studies, it is likely that more clinical trials will be conducted over the next few years. SUMMARY a73 Human autoimmune diseases can be divided into organ- specific and systemic diseases. The organ-specific diseases involve an autoimmune response directed primarily against a single organ or gland. The systemic diseases are directed against a broad spectrum of tissues and have manifestations in a variety of organs resulting from cell-mediated responses and cellular damage caused by auto-antibodies or immune complexes. a73 There are both spontaneous and experimental animal mod- els for autoimmune diseases. Spontaneous autoimmune di- seases resembling systemic lupus erythematosus occur in NZB and (NZB H11003 NZW) F 1 mice and in MRL/lpr/lpr mice, which have a defective fas gene. Several experimental animal models have been developed by immunizing animals with self-antigens in the presence of adjuvant. a73 Studies with experimental autoimmune animal models have revealed a central role for CD4 + T H cells in the devel- opment of autoimmunity. In each of the experimentally induced autoimmune diseases, autoimmune T-cell clones can be isolated that induce the autoimmune disease in nor- mal animals. The relative number of T H 1 and T H 2 cells appears to play a pivotal role in determining whether auto- immunity develops: T H 1 cells promote the development of autoimmunity, whereas T H 2 cells appear to block develop- ment of autoimmune disease and also block the progres- sion of the disease once it is established. The MHC haplo- type of the experimental animal determines the ability to present various autoantigens to T H cells. a73 A variety of mechanisms have been proposed for induction of autoimmunity, including release of sequestered antigens, molecular mimicry, inappropriate class II MHC expression on cells (in some cases stimulated by IFN-H9253, and polyclonal B-cell activation. Evidence exists for each of these mecha- 478 PART IV The Immune System in Health and Disease EAE severity 3 2 1 0 01020 30 40 Anti-V β 8.2 Control Time after antibody injection, days FIGURE 20-12 Injection of monoclonal antibody to the V H9252 8.2 T-cell receptor into PL/J mice exhibiting EAE symptoms produced nearly complete remission of symptoms. EAE was induced by inject- ing mice with MBP-specific T-cell clones. EAE severity scale: 3 = total paralysis of lower limbs; 2 = partial paralysis of lower limbs; 1 = limb tail; 0 = normal (no symptoms). [Adapted from H. Acha-Orbea et al., 1989, Annu. Rev. Immunol. 7:371.] nisms, reflecting the many different pathways leading to autoimmune reactions. a73 Current therapies for autoimmune diseases include treat- ment with immunosuppressive drugs, thymectomy, and plasmapheresis for diseases involving immune complexes. Other therapies include vaccination with T cells specific for a given autoantigen, administration of synthetic block- ing peptides that compete with autoantigen for binding to MHC molecules, treatment with monoclonal antibodies that react with some component specifically involved in an autoimmune reaction, and induction of tolerance to autoantigens by administering them orally. References Benoist, C., and D. Matis. 2001. Autoimmunity provoked by infection: how good is the case for T cell epitope mimicry? Nat Immunol. 2:797. Charlton, B., and K. J. Lafferty. 1995. The T H 1/T H 2 balance in autoimmunity. Curr. Opin. Immunol. 7:793. Erikson, J., et al. 1998. Self-reactive B cells in nonautoimmune and autoimmune mice. Immunol. Res. 17:49. Hausmann, S., and K. W. Wucherpfennig. 1997. Activation of autoreactive T cells by peptides from human pathogens. Curr. Opin. Immunol. 9:831. Horwitz, M. S., et al. 1998. Diabetes induced by Coxsackie virus: initiation by bystander damage and not molecular mimicry. Nature Med. 4:781. King, C., and N. Sarvetnick. 1997. Organ specific autoimmunity. Curr. Opin. Immunol. 9:863. Levin, M. C., et al. 2002. Autoimmunity due to molecular mim- icry as a cause of neurological disease. Nat Med. 8:509. Liblau, R. S., S. M. Singer, and H. O. McDevitt. 1995. T H 1 and T H 2 CD4 + T cells in the pathogenesis of organ-specific auto- immune diseases. Immunol. Today 16:34. Lockshin, M. D. 1998. Why women? J. Am. Med. Womens Assoc. 53:4. O’Garra, A., L. Steinman, and K. Gijbels. 1997. CD4 + T-cell sub- sets in autoimmunity. Curr. Opin. Immunol. 9:872. Rose, N. R. 1998. The role of infection in the pathogenesis of autoimmune disease. Semin. Immunol. 10:5. Silverstein, A. M., and N. R. Rose. 1997. On the mystique of the immunological self. Immunol. Rev. 159:197. Steinman, L., J. R. Oskenberg, and C. C. A. Bernard. 1992. Asso- ciation of susceptibility to multiple sclerosis with TCR genes. Immunol. Today 13:49. Streilein, J. W., M. R. Dana, and B. R. Ksander. 1997. Immunity causing blindness: five different paths to herpes stromal kerati- tis. Immunol. Today 18:443. Theofilopoulos, A. N. 1995. The basis of autoimmunity. Part I: Mechanisms of aberrant self-recognition. Immunol. Today 16:90. Theofilopoulos, A. N. 1995. The basis of autoimmunity. Part II: Genetic predisposition. Immunol. Today 16:150. Vyse, T. J., and B. L. Kotzin. 1998. Genetic susceptibility to sys- temic lupus erythematosus. Annu. Rev. Immunol. 16:261. Weiner, H. L., et al. 1993. Double-blind pilot trial of oral toler- ization with myelin antigens in multiple sclerosis. Science 259:1321. Wilder, R. L. 1995. Neuroendocrine-immune system interac- tions and autoimmunity. Annu. Rev. Immunol. 13:307. Wucherpfennig, K. W. 2001.Mechanisms for the induction of autoimmunity by infectious agents. J Clin. Invest. 108:1097. Zhao, Z-S, et al. 1998. Molecular mimicry by herpes simplex virus type-1: autoimmune disease after viral infection. Science 279:1344. USEFUL WEB SITES http://www.lupus.org/index.html The site for the Lupus Foundation of America contains valu- able information for patients and family members as well as current information about research in this area. http://www.nih.gov/niams/ Home page for the National Institute for Arthritis and Mus- culoskeletal and Skin Diseases. This site contains links to other arthritis sites. http://www.niddk.nih.gov/ Home page for the National Institute for Diabetes and Diges- tive and Kidney Diseases. This site contains an exhaustive list of links to other diabetes health-related sites. http://www.sciencemag.org/feature/data/983519.shl Link to a Web site that provides specific information concern- ing the role of gender in autoimmune disease. Study Questions CLINICAL FOCUS QUESTION What are some of the possible reasons why females are more susceptible to autoimmune diseases than males? 1. For each of the following autoimmune diseases (a–m), select the most appropriate characteristic (1–13) listed below. Diseases a. Experimental autoimmune encephalitis (EAE) b. Goodpasture’s syndrome c. Graves’ disease d. Systemic lupus erythematosus (SLE) d. Insulin-dependent diabetes mellitus (IDDM) Autoimmunity CHAPTER 20 479 Go to www.whfreeman.com/immunology Self-Test Review and quiz of key terms f. Rheumatoid arthritis g. Hashimoto’s thyroiditis h. Experimental autoimmune myasthenia gravis (EAMG) i. Myasthenia gravis j. Pernicious anemia k. Multiple sclerosis l. Autoimmune hemolytic anemia Characteristics (1) Auto-antibodies to intrinsic factor block vitamin B 12 absorption (2) Auto-antibodies to acetylcholine receptor (3) T H 1-cell reaction to thyroid antigens (4) Auto-antibodies to RBC antigens (5) T-cell response to myelin (6) Induced by injection of myelin basic protein plus com- plete Freund’s adjuvant (7) Auto-antibody to IgG (8) Auto-antibodies to basement membrane (9) Auto-antibodies to DNA and DNA-associated protein (10) Auto-antibodies to receptor for thyroid-stimulating hormone (11) Induced by injection of acetylcholine receptors (12) T H 1-cell response to pancreatic beta cells 2. Experimental autoimmune encephalitis (EAE) has proved to be a useful animal model of autoimmune disorders. a. Describe how this animal model is made. b. What is unusual about the animals that recover from EAE? c. How has this animal model indicated a role for T cells in the development of autoimmunity? 3. Molecular mimicry is one mechanism proposed to account for the development of autoimmunity. How has induction of EAE with myelin basic protein contributed to the understanding of molecular mimicry in autoimmune disease? 4. Describe at least three different mechanisms by which a localized viral infection might contribute to the development of an organ-specific autoimmune disease. 5. Transgenic mice expressing the IFN-H9253 transgene linked to the insulin promoter developed diabetes. a. Why was the insulin promoter used? b. What is the evidence that the diabetes in these mice is due to autoimmune damage? c. What is unusual about MHC expression in this system? d. How might this system mimic events that might be caused by a localized viral infection in the pancreas? 6. Monoclonal antibodies have been administered for therapy in various autoimmune animal models. Which monoclonal anti- bodies have been used and what is the rationale for these approaches? 7. Indicate whether each of the following statements is true or false. If you think a statement is false, explain why. a. T H 1 cells have been associated with development of autoimmunity. b. Immunization of mice with IL-12 prevents induction of EAE by injection of myelin basic protein plus adjuvant. c. The presence of the HLA B27 allele is diagnostic for ankylosing spondylitis, an autoimmune disease affecting the vertebrae. d. Individuals with pernicious anemia produce antibodies to intrinsic factor. e. A defect in the gene encoding Fas can reduce pro- grammed cell death by apoptosis. 480 PART IV The Immune System in Health and Disease