Systemic lupus erythematosus is a multisystem autoimmune disease in which the autoantibody response targets a variety of autoantigens of diverse subcellular location. We show here that these autoantigens are clustered in two distinct populations of blebs at the surface of apoptotic cells. The population of smaller blebs contains fragmented endoplasmic reticulum (ER) and ribosomes, as well as the ribonucleoprotein, Ro. The larger blebs (apoptotic bodies) contain nucleosomal DNA, Ro, La, and the small nuclear ribonucleoproteins. These autoantigen clusters have in common their proximity to the ER and nuclear membranes, sites of increased generation of reactive oxygen species in apoptotic cells. Oxidative modification at these sites may be a mechanism that unites this diverse group of molecules together as autoantigens.

The complement system plays a paradoxical role in the development and expression of autoimmunity in humans. The activation of complement in systemic lupus erythematosus (SLE) contributes to tissue injury. In contrast, inherited deficiency of classical pathway components, particularly C1q (ref. 1), is powerfully associated with the development of SLE. This leads to the hypothesis that a physiological action of the early part of the classical pathway protects against the development of SLE (ref. 2) and implies that C1q may play a key role in this respect. C1q-deficient (C1qa-/-) mice were generated by gene targeting and monitored for eight months. C1qa-/- mice had increased mortality and higher titres of autoantibodies, compared with strain-matched controls. Of the C1qa-/- mice, 25% had glomerulonephritis with immune deposits and multiple apoptotic cell bodies. Among mice without glomerulonephritis, there were significantly greater numbers of glomerular apoptotic bodies in C1q-deficient mice compared with controls. The phenotype associated with C1q deficiency was modified by background genes. These findings are compatible with the hypothesis that C1q deficiency causes autoimmunity by impairment of the clearance of apoptotic cells.

Anti-phospholipid (aPL) antibodies that exhibit binding in cardiolipin (CL) ELISA can be purified to greater than 95% purity by sequential phospholipid affinity and ion-exchange chromatography. However, these highly purified aPL antibodies do not bind to the CL antigen when assayed by a modified CL ELISA in which the blocking agent does not contain bovine serum, nor do they bind to phospholipid affinity columns. Binding to the phospholipid antigen will only occur if normal human plasma, human serum, or bovine serum is present, suggesting that the binding of aPL antibodies to CL requires the presence of a plasma/serum cofactor. Using sequential phospholipid affinity, gel-filtration, and ion-exchange chromatography, we have purified this cofactor to homogeneity and shown that the binding of aPL antibodies to CL requires the presence of this cofactor in a dose-dependent manner. N-terminal region sequence analysis of the molecule has identified the cofactor as beta 2-glycoprotein I (beta 2GPI)(apolipoprotein H), a plasma protein known to bind to anionic phospholipids. These findings indicate that the presence of beta 2GPI is an absolute requirement for antibody-phospholipid interaction, suggesting that bound beta 2GPI forms the antigen to which aPL antibodies are directed. Recent evidence indicates that beta 2GPI exerts multiple inhibitory effects on the coagulation pathway and platelet aggregation. Interference with the function of beta 2GPI by aPL antibodies could explain the thrombotic diathesis seen in association with these antibodies.

Immunization with myelin basic protein (MBP) induces experimental allergic encephalomyelitis (EAE), a prototype of CD4+ T-cell mediated autoimmune disease. In rodents, MBP-reactive T-cell clones are specific for a single, dominant determinant on MBP and use a highly restricted number of T-cell receptor genes. Accordingly, EAE has been prevented by various receptor-specific treatments, suggesting similar strategies may be useful for therapy of human autoimmune disease. Here we report that in (SJL x B10.PL)F1 mice, immune dominance of a single determinant, MBP:Ac1-11, is confined to the inductive phase of EAE. In mice with chronic EAE, several additional determinants of MBP in peptides 35-47, 81-100 and 121-140 recall proliferative responses. Most importantly, reactivity to the latter determinants was also detected after induction of EAE with MBP peptide Ac1-11 alone; this demonstrates priming by endogenous MBP determinants. Thus, determinants of MBP that are cryptic after primary immunization can become immunogenic in the course of EAE. Diversification of the autoreactive T-cell repertoire due to 'determinant spreading' has major implications for the pathogenesis of, and the therapeutic approach to, T-cell driven autoimmune disease.

Immune responses are initiated by HLA-DR+ cells, which present antigen to T cells. Observations that HLA-DR may be experimentally induced on thyroid epithelium and that HLA-DR occurs on thyrocytes in autoimmune thyroid diseases suggest a mechanism of autoimmunity with special relevance to organ-specific diseases. This involves the local aberrant expression of HLA-DR antigens by epithelial cells and their subsequent capacity to present autoantigens occurring on their surfaces to T lymphocytes. For autoantigens which T cells recognise infrequently because of their restricted tissue location and low concentration in the circulation, T-cell tolerance is unlikely, and so induction of autoreactive T cells would occur. Because interferon is the best known inducer of DR antigen expression and viral infections may predate endocrine autoimmunity, the following sequence seems likely: local viral infection which causes interferon production, or other local environmental factors which would induce DR expression, presentation of autoantigens, and subsequent autoimmune T-cell induction. These T cells would activate effector B and T cells. Whether the initial induction of autoimmune T cells leads to autoimmune disease would depend on factors such as abnormalities of the suppressor T-cell pathway, reported to coexist with autoimmunity and necessary to induce autoimmune disease in mice. This mechanism of autoimmune disease induction explains vague associations with viral infections and long latency periods before disease becomes manifest and gives a simple explanation for the well-documented association between HLA-DR and autoimmune diseases in man.

Insulin-dependent diabetes mellitus (IDDM) in non-obese diabetic (NOD) mice results from the T-lymphocyte-mediated destruction of the insulin-producing pancreatic beta-cells and serves as a model for human IDDM. Whereas a number of autoantibodies are associated with IDDM, it is unclear when and to what beta-cell antigens pathogenic T cells become activated during the disease process. We report here that a T-helper-1 (Th1) response to glutamate decarboxylase develops in NOD mice at the same time as the onset of insulitis. This response is initially limited to a confined region of glutamate decarboxylase, but later spreads intramolecularly to additional determinants. Subsequently, T-cell reactivity arises to other beta-cell antigens, consistent with intermolecular diversification of the response. Prevention of the spontaneous anti-glutamate decarboxylase response, by tolerization of glutamate decarboxylase-reactive T cells, blocks the development of T-cell autoimmunity to other beta-cell antigens, as well as insulitis and diabetes. Our data suggest that (1) glutamate decarboxylase is a key target antigen in the induction of murine IDDM;(2) autoimmunity to glutamate decarboxylase triggers T-cell responses to other beta-cell antigens, and (3) spontaneous autoimmune disease can be prevented by tolerization to the initiating target antigen.

Celiac disease is characterized by small intestinal damage with loss of absorptive villi and hyperplasia of the crypts, typically leading to malabsorption. In addition to nutrient deficiencies, prolonged celiac disease is associated with an increased risk for malignancy, especially intestinal T-cell lymphoma. Celiac disease is precipitated by ingestion of the protein gliadin, a component of wheat gluten, and usually resolves on its withdrawal. Gliadin initiates mucosal damage which involves an immunological process in individuals with a genetic predisposition. However, the mechanism responsible for the small intestinal damage characteristic of celiac disease is still under debate. Small intestinal biopsy with the demonstration of a flat mucosa which is reversed on a gluten-free diet is considered the main approach for diagnosis of classical celiac disease. In addition, IgA antibodies against gliadin and endomysium, a structure of the smooth muscle connective tissue, are valuable tools for the detection of patients with celiac disease and for therapy control. Incidence rates of childhood celiac disease range from 1:300 in Western Ireland to 1:4700 in other European countries, and subclinical cases detected by serological screening revealed prevalences of 3.3 and 4 per 1000 in Italy and the USA, respectively. IgA antibodies to endomysium are particularly specific indicators of celiac disease, suggesting that this structure contains one or more target autoantigens that play a role in the pathogenesis of the disease. However, the identification of the endomysial autoantigen(s) has remained elusive. We identified tissue transglutaminase as the unknown endomysial autoantigen. Interestingly, gliadin is a preferred substrate for this enzyme, giving rise to novel antigenic epitopes.

Knowing the autoantigen target(s) in an organ-specific autoimmune disease is essential to understanding its pathogenesis. Insulin-dependent diabetes mellitus (IDDM) is an autoimmune disease characterized by lymphocytic infiltration of the islets of Langerhans (insulitis) and destruction of insulin-secreting pancreatic beta-cells. Several beta-cell proteins have been identified as autoantigens, but their importance in the diabetogenic process is not known. The non-obese diabetic (NOD) mouse is a murine model for spontaneous IDDM. Here we determine the temporal sequence of T-cell and antibody responses in NOD mice to a panel of five murine beta-cell antigens and find that antibody and T-cell responses specific for the two isoforms of glutamic acid decarboxylase (GAD) are first detected in 4-week-old NOD mice. This GAD-specific reactivity coincides with the earliest detectable response to an islet extract, and with the onset of insulitis. Furthermore, NOD mice receiving intrathymic injections of GAD65 exhibit markedly reduced T-cell proliferative responses to GAD and to the rest of the panel, in addition to remaining free of diabetes. These results indicate that the spontaneous response to beta-cell antigens arises very early in life and that the anti-GAD immune response has a critical role in the disease process during this period.