Overview of autoimmunity

INTRODUCTION — Autoimmune diseases are characterized by a pathologic response to self- or autoantigens; these responses can be categorized as autoimmunity or autoreactivity and underlie a wide range of clinical disorders. These disorders may be generalized, such as systemic rheumatic diseases like systemic lupus erythematosus, vasculitis, and others, or tissue- or organ-specific, such as endocrine and neurologic disorders, including autoimmune thyroiditis and multiple sclerosis, respectively, and other conditions. Autoimmune diseases can be either acute or chronic and can affect essentially all organs and body systems.

An overview of autoimmunity is provided here, including discussions of the features common to and underlying autoimmune disorders and of steps for recognizing autoimmune disease. The clinical manifestations, diagnosis, and management of specific autoimmune conditions are described in detail separately. (See appropriate topic reviews.)

DEFINITION — Autoimmune diseases are characterized by a pathologic state in which an aberrant immune response directed at a normal bodily constituent leads to inflammation, cell injury, or a functional disturbance with clinical manifestations [1,2]. The molecular constituent (ie, protein, carbohydrate, nucleic acid) that is targeted in autoimmunity is called a "self"-antigen or an autoantigen; by contrast, a molecule from an infecting organism that stimulates an immune response is called a foreign antigen or "non-self." An autoimmune disease usually involves both a T and B cell response and can be generalized or tissue- or organ-specific and either acute or chronic.

While autoimmune diseases are a pathologic state, autoimmunity nevertheless derives from the same mechanisms that underlie the normal immune response to foreign antigens. Immune responses can be divided into two broad categories: innate and adaptive. The innate immune response is a rapid and nonspecific response to a challenge, whether it arises from infection, trauma, or stress. By contrast, an adaptive immune response is slow (days to weeks) and involves the production of antigen-specific B or T cells to overcome a foreign challenge.

While innate and adaptive immune responses (along with responsible cells and mediators) can be separated for analysis, an adaptive immune response depends upon the presence of an innate immune response to promote the generation of a specific immune response. Importantly, an adaptive immune response can be persistent and show memory. In this framework, an autoimmune disease results from a specific adaptive autoimmune response to an autoantigen; this response is in violation of the normal function of the immune system in which mechanisms of tolerance prevent "hyper-reactive" autoimmune responses to self-antigens.

In general, treatment of an autoimmune disease requires agents to decrease the activity of the immune system (immunosuppressants) or block the inflammation (antiinflammatories) that lead to tissue injury. The definition of autoimmune disease does not specify the origin of the response nor its induction by a foreign or self-antigen. Depending upon evidence linking autoimmunity to a preceding infection, therapy in some conditions will be directed at a specific infection [3,4]. In some instances, therapy relates to the functional disturbance (eg, insulin replacement therapy in type 1 diabetes).

STEPS FOR RECOGNIZING AUTOIMMUNE DISEASE

Features of autoimmune disease — A number of features characterize autoimmune disorders, and the determination that a disease is autoimmune involves various forms of clinical and laboratory evidence [1,2]. The demonstration of autoantibodies is usually the first step in the diagnosis of an autoimmune disease, although it is not sufficient. The autoantibodies may not be the actual mediators of disease; naturally occurring autoantibodies can occur in immunologically competent people and may even rise nonspecifically during the course of disease or injury. Thus, some clinical manifestation should be evident in addition to the presence of autoantibody. Evidence for inherited susceptibility is often present in families of patients suspected of an autoimmune disease.

Major characteristics include:

●Recognition of autoantibodies and target antigens – The usual first step is the demonstration of autoreactivity, which is most often an autoantibody to a tissue antigen. Even when associated with a clinical condition, autoantibodies may not mediate pathogenesis since naturally occurring autoantibodies can occur in immunologically competent people; levels of these antibodies may rise nonspecifically during the course of infection, cancer, or injury. These antibodies are usually of the immunoglobulin M (IgM) isotype and have low avidity for the self-antigen. Some natural autoantibodies may have beneficial or protective effects, accounting for their production during inflammatory conditions. Thus, the mere presence of a heightened level of autoantibodies does not by itself establish a cause-and-effect relationship.

Although some autoimmune diseases affect a single or limited number of tissues, the target autoantigen may be widely expressed in the body and present in all cells. As such, the autoantigen may not have specific or even apparent relationship to the tissue affected or the clinical manifestations. By contrast, some autoantigens may be selectively expressed in a target tissue and have a clear relationship to pathophysiology (eg, receptor for a hormone or neurotransmitter). Such an antigen is called a tissue-specific antigen (TSA) or a tissue-restricted antigen (TRA). In addition, some autoantibodies to TSAs (eg, thyroid) may not be pathogenic. The differences between ubiquitously expressed molecules and TSAs are relevant for tolerance mechanisms. Finally, some inflammatory processes (eg, formation of neutrophil extracellular traps [NETosis]) appear able incite the production of certain autoantibodies (eg, anti-citrullinated protein antibodies [ACPA], proteinase 3, myeloperoxidase) that are strongly associated with clinical disease states such as rheumatoid arthritis or vasculitis [5].

A challenge in deciding whether a condition is autoimmune relates to the identification of the target antigen and the establishment of an immunoassay to detect the corresponding autoantibodies. In some instances, an assay can allow detection of a broad range of autoantibodies to cellular antigens and have utility as a general screen in patient evaluation. The antinuclear antibody (ANA) assay is the prime example of this kind of screening assay; ANA assays can detect antibodies to many different nuclear and cytoplasmic antigens [6]. Assays of this kind can be informative in many different clinical settings, although, by their nature, these assays are not specific.

Another limitation of the ANA assay, in particular, relates to the detection of autoantibodies in the otherwise healthy population [7]. The frequency of ANA positivity can approach 10 to 20 percent; this frequency may be increasing, suggesting some environmental influence leading to a general increase in autoreactivity [8]. This situation provides further evidence that autoantibody production can occur in otherwise healthy individuals and that autoantibody production is not sufficient for clinical disease.

Because of advances in molecular biology, the identity of many self-antigens has been determined, and assays are now available for routine testing. These assays can help assess the role of autoimmunity in a particular condition and result from intensive searches for the presence of autoantibodies in conditions of previously unknown etiology. Such searches, for example, have demonstrated the presence of autoantibodies in paraneoplastic neurologic syndromes [9]. Assays for autoantibodies can also allow the definition of new syndromes, as well as the subdivision of conditions into subtypes that may differ in clinical course and treatment response. Thus, discoveries of new autoantibodies in sera from patients with optic neuritis allow the definition of syndromes distinguished by reactivity to aquaporin 4 (AQP4) or myelin oligodendrocyte glycoprotein (MOG) [10].

For conditions characterized by a functional disturbance, an in vitro assay with cells from a tissue or a tissue culture line can be used to assess agonist or antagonist actions of a serum or purified antibody.

●Clinical autoimmunity – An autoimmune disease becomes clinically recognized at the time that the patient experiences disease manifestations in terms of signs and symptoms. This phase of disease can be called clinical autoimmunity. For some conditions, evidence of autoimmunity may predate clinical findings in terms of autoantibody production or abnormal immune reactivity such as increased cytokine production [11,12]. Indeed, autoantibodies may be present many years before the diagnosis of a disease in the clinic such as systemic lupus erythematosus, rheumatoid arthritis, antiphospholipid syndrome, and type 1 diabetes mellitus.

Pre-autoimmunity defines the period in which serologic and other immunologic findings are present but clinical manifestations are not yet apparent. Combined with genetic information or family history, the presence of elevated levels of autoantibodies may be highly predictive of the later onset of an autoimmune disorder and underpin efforts at prevention in individuals at high risk of disease [13,14].

●Inherited susceptibility – Genetic studies of families and of large populations of patients have demonstrated the role of inherited susceptibility factors in many autoimmune diseases. While genes of the major histocompatibility complex (MHC) are among the most prominent susceptibility factors, the genetics of disease in humans are complex, involving multiple genes whose array may differ among patients.

For some autoimmune diseases, over 100 genes have already been identified [15]. These findings suggest that genetic factors may affect the poise or balance of the immune system, predisposing an affected individual to generating an autoimmune response to an antigenic challenge that would be without consequence in other individuals. While gene polymorphisms may influence the signaling thresholds of individual immune cells, several or more of such polymorphisms may be needed for a significant increase in autoreactivity.

The use of genetic markers to predict disease depends on the frequency of the disease in a population, although targeted screening of family members who are a high risk (eg, siblings, identical twins) can be more productive.

Formal demonstration that a disease is autoimmune in origin requires several kinds of evidence that have analogy to Koch postulates for showing that a disease arises from infection. For autoimmune disease, the term "Witebsky postulates" has sometimes been used to categorize the different types of evidence [16,17]. (See 'Direct evidence' below and 'Indirect evidence' below and 'Circumstantial evidence' below.)

Direct evidence — Direct evidence of causality, a stringent level of evidence for autoimmunity, requires that an autoimmune response produces the characteristic pathologic features of the disease. This usually involves reproducing the disease totally or, in part, by transfer of autoantibody from a patient to a healthy recipient. This is typically demonstrated experimentally using transfer systems involving animals, although transfer of blood from a patient to an otherwise healthy subject has rarely occurred, and in vitro or model test systems have also been used. An important void in studies of human autoimmune disease is the lack of methods that directly show the pathogenic effects of T cells by adoptive transfer.

Examples of direct evidence include:

●Antibody transfer from human patient to animal model – One striking example of antibody transfer was the reproduction of pemphigus by injection of patient serum into a neonatal mouse [18]. This procedure reproduced the essential pathologic features of the disease.

●Transplacental human-to-human autoantibody transfer – Human-to-human transfer of autoantibody may result from transplacental transmission of the disease. Examples of maternal-fetal transmission have been well documented in cases of Graves' disease [19], myasthenia gravis [20], and the complete heart block and other cardiac abnormalities in neonatal lupus associated with maternal lupus and Sjögren's disease [21]. Most of the clinical manifestations in the offspring are temporary because the autoantibody in these cases is provided through passive transfer of serum from the mother. An exception is the congenital heart block and other cardiac abnormalities of neonatal lupus, which are persistent and potentially life threatening. In this situation, the heart block may be the outcome of intense local inflammation damaging cells of the conduction system [22,23]. (See "Pathogenesis of Sjögren’s disease" and "Pregnancy in women with systemic lupus erythematosus" and "Neonatal lupus: Epidemiology, pathogenesis, clinical manifestations, and diagnosis".)

●Autoantibody effects in vitro – There are situations in which the pathologic effects of antibody can be reproduced in an in vitro model system. Examples are found in diseases of the blood, such as hemolytic anemia, leukopenia, or thrombocytopenia (see "Cold agglutinin disease", section on 'Pathogenesis' and "Warm autoimmune hemolytic anemia (AIHA) in adults", section on 'Direct antiglobulin (Coombs) testing'). Antiphospholipid antibodies, another example, are autoantibodies that may affect blood clotting in laboratory tests [24]. (See "Diagnosis of antiphospholipid syndrome", section on 'Patients on an anticoagulant'.)

An important gap in studying autoimmunity relates to methods to show the pathogenic effects of T cells. Efforts have been made to adapt immunodeficient animals as "living test tubes" for the demonstration of autoimmune-induced pathology. Experiments designed to show the effects of cytotoxic lymphocytes against human tissue cells in the test tube require that the cells be cultured in vitro while retaining their native antigens.

Indirect evidence — A second type of evidence to show causality comes from the simulation of disease in an experimental animal. Different animal models are utilized for this purpose: reproduction of disease in animals via immunization with the appropriate self-antigen, naturally occurring disease in animals that resembles its human counterpart, and disease resulting from manipulation of the immune system. (See 'Immunization with self-antigen' below and 'Naturally occurring disease in animals' below and 'Immune system manipulation' below.)

Immunization with self-antigen — Two examples illustrate the use of immunization with self-antigen to mimic human autoimmune disorders:

●Autoimmune thyroiditis – A good example of an experimental model is autoimmune thyroiditis in the mouse [25]. Here, the experimental antigen, thyroglobulin, is a major target of autoantibody production in patients. In genetically susceptible strains of mice, immunization with thyroglobulin produces a pathologic picture of chronic thyroiditis that closely resembles the human disease (see "Pathogenesis of Hashimoto's thyroiditis (chronic autoimmune thyroiditis)"). Another useful example is myocarditis, which can be reproduced by immunization of susceptible mice with murine myosin [26]. (See "Myocarditis: Causes and pathogenesis".)

●Experimental autoimmune encephalomyelitis – One of the most widely studied models of autoimmune disease induced by immunization is experimental autoimmune encephalomyelitis (EAE) [27,28]. The demyelinating changes in the experimental disease show similarities to multiple sclerosis in humans. Three antigens can induce this disease in rodents: myelin basic protein (MBP), proteolipid protein (PLP), and myelin oligodendrocyte glycoprotein (MOG). Immunization with MOG produces axonal loss as well as demyelination and has a relapsing and remitting course similar to multiple sclerosis in humans [29]. Manipulation of components of the immune system, such as the introduction of a T cell receptor for a disease-producing antigen into an otherwise disease-resistant strain, can further define steps in the autoimmune process [30].

Naturally occurring disease in animals — Another type of disease model is derived from naturally occurring disease in animals that resembles its human counterpart. As examples:

●Mouse models of systemic lupus erythematosus – Diseases with features of systemic lupus erythematosus occur in particular inbred strains of mice. These strains include the NZB/NZW F1 hybrid, MRL/lpr, and BXSB [31]. These mice express autoantibodies to deoxyribonucleic acid (DNA; anti-DNA), a characteristic antibody of systemic lupus erythematosus, along with renal disease that is mediated by immune complexes. Cell-transfer experiments clearly establish these diseases as autoimmune in origin [32,33]. Development of autoimmunity in these mice is spontaneous and does not require any immunization, although environmental factors may affect the course of disease. Among these factors, the microbiome may have an important role in determining the development of disease [34]. (See "Epidemiology and pathogenesis of systemic lupus erythematosus".)

●Nonobese diabetic mouse model of type 1 (autoimmune) diabetes – Another important spontaneous model is the nonobese diabetic (NOD) mouse, which develops a disease closely resembling type 1 (autoimmune) diabetes [35]. These mice have been particularly useful for delineating the major genetic traits that increase susceptibility to the disease, some of which also appear to operate in human type 1 diabetes and other autoimmune diseases [36].

Whether induced or spontaneously occurring, animal models are also important for testing interventions including pharmacologic agents that may reveal mechanisms or provide proof-of-principle data for preclinical drug development. While these animal model systems resemble features of human disease, there are nevertheless important differences in underlying mechanisms that lead to caution in extrapolating information to the human situation.

Immune system manipulation — Studies with genetically manipulated mice indicate that a wide range of genes involving the immune system can contribute to autoimmunity [37,38]. These studies employ molecular biologic techniques that enable the creation of mice in which a particular gene can have increased expression (eg, transgenic mice) or decreased expression (eg, gene deletion or "knockout" mice). With a change in expression of a single gene, disease with features of systemic lupus erythematosus, for example, can occur, but the occurrence of disease can differ among inbred strains, pointing to epistatic effects, by which the phenotypic effect of the change depends upon variants in one or more other genes.

The most important genetic trait in determining susceptibility to autoimmune disease is the human class II MHC human leukocyte antigen (HLA). Replacing the MHC of a mouse with a human susceptibility HLA haplotype can render the mouse susceptible to spontaneous autoimmune disease [39]. For instance, simply inserting the susceptible HLA haplotype can induce spontaneous myocarditis in previously resistant mice [40].

As a further example of a genetic immune manipulation that results in autoimmune disease, models of inflammatory bowel disease have been described in animals in which particular cytokines such as interleukin (IL) 2 and IL-10 have been eliminated [41,42]. Another example is in mice deficient in programmed cell death (PD) 1 immunoinhibitory coreceptor that develop autoimmune dilated cardiomyopathy with production of high titers of autoantibodies against cardiac troponin-1 [43].

It has become possible to relate these findings on immune system manipulation in animal experiments to human subjects receiving immune checkpoint blockade as cancer immunotherapy. Some of the patients develop autoimmune disease that resembles the spontaneous condition [44,45]. (See "Toxicities associated with immune checkpoint inhibitors" and "Rheumatologic complications of checkpoint inhibitor immunotherapy".)

Circumstantial evidence — The lowest level of proof, circumstantial evidence, is the one most commonly available to attribute an enigmatic human disease to autoimmunity. Such evidence may include systemic inflammation, the presence of autoantibodies, clustering of diseases within an individual or family, bias to certain HLA haplotypes, a sex bias, and a response to immunosuppressive therapy.

●Systemic inflammation and autoantibodies – Circumstantial evidence can suggest that an enigmatic human disease results from autoimmunity. The first piece of suggestive evidence is frequently evidence of systemic inflammation (eg, increased sedimentation rate, increased levels of C-reactive protein), along with findings of immune activation such as increased levels of immunoglobulins. The presence of autoantibodies also suggests an autoimmune process. While there are now assays for a very large number of autoantibodies, initial screening may involve testing for ANAs. The limitation of using this kind of evidence for positing a disease as autoimmune is the high frequency of false-positive results. Natural autoantibodies are very common and may rise nonspecifically during a disease process [46].

●Familial clustering and genetic factors – A second kind of circumstantial evidence comes from the finding that autoimmune diseases tend to cluster, probably because they share some genetic susceptibility traits. As examples, a single individual may have more than one autoimmune disease, and members of the family share the same or other, related autoimmune diseases. The association of one disease of unclear etiology with another of authentic autoimmune etiology strengthens the possibility that the former is also an autoimmune disorder [47,48].

As stated above, the best-studied autoimmune diseases show a particular bias to certain HLA haplotypes, usually the class II category. Since the class II MHC genes are critical in antigen presentation and disease induction, they can modulate the autoimmune response [49]. Next, a large number of regulatory genes contribute to normal immunologic homeostasis. Allelic differences in these genes may contribute to a greater or lesser susceptibility to various autoimmune diseases. These genes include such regulatory genes as CTLA4 and PTPN22 [50,51]. Yet, in the aggregate, all of these heritable traits, including polymorphisms of MHC class II and regulatory genes, contribute less than half of the susceptibility to autoimmune diseases, even in genetically identical twins [52]. As an example, a retrospective cohort study from Sweden found that monozygotic identical twin concordance for ACPA with or without rheumatoid arthritis was less than 5 percent [53].

In addition, postgenomic epigenetic effects add to incomplete concordance of monozygotic twins. Epigenetic changes in gene expression occur by mechanisms other than alterations in the DNA sequence. These changes include methylation of DNA nucleotides and histone post-translational modifications affecting chromatin structure. These and many other epigenetic changes account for the fact that even monozygotic twins do not have identical immune responses [54].

Thus, the majority of patients with an autoimmune disease present without a clear family history. In part, this paradox underlines the key role of environmental factors as triggers of autoimmune disease in genetically more susceptible individuals. (See 'Inductive mechanisms' below and 'Pathogenetic mechanisms' below.)

●Sex bias – Most, but not all, autoimmune diseases are more common in women than men. A sex bias, therefore, provides increased circumstantial evidence of an autoimmune etiology. The sex-based differences in autoimmune diseases have provided new information on the important role of sex hormones in the pathogenesis of autoimmune disease [48,55,56]. Genes on the X-chromosome involved with immunoregulation can also contribute to susceptibility to autoimmunity, especially if these genes are overexpressed because of disturbances of X-chromosome inactivation, the mechanism by which one X-chromosome is inactivated in each cell [57,58].

●Response to immunosuppressive therapy – Finally, the favorable response of a disease to immunosuppressive treatment is often an initial clinical hint of a possible autoimmune etiology. Of these treatments, glucocorticoids have both antiinflammatory and immunosuppressive actions that can reduce symptoms of many diseases, including infections, suggesting caution in inferring autoimmunity from a pharmacologic effect.

INDUCTIVE MECHANISMS — The presence of B and/or T cell autoreactivity is the defining characteristic of an autoimmune disease. While B and/or T cell autoreactivity may be the proximate cause of disease, an autoimmune response results from the complex interplay of multiple cell populations including antigen-presenting cells, T helper (Th) cells, B cells, and regulatory cells, among others, with cytokines influencing the magnitude of these responses at multiple levels.

An autoimmune response can be initiated by an autologous (self) or foreign (non-self) antigen; induction by a foreign antigen suggests structural similarity with an autoantigen, with this cross-reactivity denoted as molecular mimicry. Identification of a molecular mimic can involve a search for shared amino acid sequences between a foreign and self-antigen. Thus, the binding of antibodies to both a foreign and self-antigen provides further evidence for molecular mimicry. Studies of this kind have provided evidence for molecular mimicry in diseases such as rheumatoid arthritis [59].

Like all immune responses, autoimmunity appears under genetic control of antigen recognition, cellular interactions, and eventual outcomes. In addition, environmental agents may promote autoimmune responses [60]. Infections, for example, can provide the requisite antigen by mimicking or altering of self-antigens or heightening the overall level of immune reactivity or by causing antigen "spillage" [61,62].

Even without an infection, microorganisms may also influence both the initiation and progression of an autoimmune response. In individuals, populations of commensal microorganisms ("microbiota") can profoundly affect the induction of such diseases as inflammatory bowel diseases [63]. The gut or gastrointestinal tract is the largest source of microorganisms in the body including bacteria, viruses, and fungi, but microorganisms also inhabit the skin and mucus membranes. Changes in the composition of the microbiota can occur in autoimmune disease producing a state known as dysbiosis [64,65]. As an example, the increased levels of a specific gut organism (Ruminococcus gnavus) are associated with lupus nephritis in humans [66].

Organisms in the microbiome may also enter the blood because of impaired gut permeability and directly contact cells of the immune system. Thus, studies indicate that Enterococcus gallinarum can be found in the liver of patients with systemic lupus erythematosus, with cross-reactivity contributing an autoimmune response to the Ro antigen [67]. In addition, infection can provide the inflammatory context that favors immune responses, through activation of the innate immune responses [68]. Thus, attention has focused on the Epstein-Barr virus as an instigating agent in different autoimmune diseases, multiple sclerosis, and systemic lupus erythematosus [69,70].

PATHOGENETIC MECHANISMS — The development of autoimmune disease depends upon an imbalance between pathogenic factors generated by autoreactive T and B cells and the regulatory factors that normally control the immune response. Despite their diverse etiologies, autoimmune diseases display certain pathogenetic mechanisms in common. With few exceptions, development of these diseases requires the presence of self-reactive CD4-positive T lymphocytes to promote antigen-specific responses. Diseases in which there is a chronic inflammatory response but no evidence of adaptive immunity in the form of self-reactive T cells may reflect increased activation of innate immunity and can represent autoinflammatory diseases [71]. (See 'Autoinflammatory diseases' below.)

The following regulatory and pathogenic factors contribute to the pathogenesis of autoimmune disease:

●Breakdown or defects in immune tolerance – Autoimmune disease represents a breakdown in the normal mechanisms that prevent autoreactivity of both T and B cells. These mechanisms affect the development and regulation of multiple cell populations and act both centrally as well as peripherally. For T cells, the thymus is the locale for the establishment of central tolerance, while the bone marrow is the locale for B cells. Central tolerance may not be complete, however, and autoreactive cells can emerge in the periphery following somatic mutation, especially among B cells. Thus, in addition to deletion and elimination of cells in the thymus or bone marrow, other regulatory interactions are necessary to restrain autoreactivity [72-74].

Thymic selection is critical for normal T cell development and is dependent upon interaction with antigen-presenting cells in the thymus for steps called positive and negative selection. The interactions and the cells responsible for antigen presentation differ for ubiquitously expressed antigens and those that are tissue selective or restricted. The autoimmune regulator (AIRE) gene is an autosomal recessive gene that is responsible for intrathymic presentation of autologous antigen tissue-restricted antigens (TRAs) whose expression would otherwise be insufficient to allow the establishment of tolerance. Mutations in the AIRE gene cause several combinations of autoimmune endocrine diseases, such as autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy (APECED), because the appropriate self-antigens are not properly presented in the thymus [75]. (See "Chronic mucocutaneous candidiasis", section on 'Clinical features of AIRE deficiency/APECED'.)

Although there is experimental evidence that deletion of self-reactive lymphocytes occurs in the thymus, this mechanism seems to mostly affect prominent systemically expressed antigens, such as those of the major blood groups and histocompatibility complex [76]. By comparison, in the case of most other self-antigens, deletion of self-reactive T cells in the thymus is lacking or incomplete. Fortunately, mechanisms in the periphery can retard the activation of self-reactive T cells that escape deletion in the thymus.

Another mechanism of peripheral self-tolerance involves immunologic ignorance, that is, the lack of a productive encounter between the T cell and its corresponding peptide/major histocompatibility complex (MHC) complex on an antigen-presenting cell. Ignorance can be overcome by changes in antigen availability, such as presentation by an infecting microorganism [77]. These events can cause autoimmunity and possibly give rise to autoimmune disease.

●Defects in active regulation and control of autoreactivity – Finally, the autoimmune response can be kept in check by active regulation [78]. Regulatory T cells (Tregs) are diverse in properties and origin; some arise in the thymus as an alternative fate of antigen-specific cells rather than deletion. This fate may depend on the strength of interaction of developing T cells with antigen-presenting cells in the thymus [79,80].

Among the major regulatory cells are a subset of T cells bearing the markers CD4, CD25, and forkhead box protein P3 (FOXP3) [81]. Two different subpopulations of CD4+ Tregs can be distinguished. Naturally recurring Tregs exert their suppressive effects by cell-to-cell contact of membrane-bound molecules, such as CTLA-4. Induced Tregs, in contrast, are cell-contact-independent and operate mainly through soluble suppressive cytokines, such as interleukin (IL) 10 and transforming growth factor (TGF)-beta [82]. Treg function and phenotype can exhibit plasticity, and cells with diminished suppressive capacity may be seen in some autoimmune disorders [83,84]. Studies in animals show that antigen-specific Tregs are more effective in suppressing autoimmune disease when compared with polyclonal Tregs [85]. Clinical trials of antigen-specific Tregs are underway in organ-specific autoimmune diseases [86]. Other specialized populations of T cells such as some that express the natural killer (NK) receptor (called NK-T cells) can also regulate autoimmune disease [78,87].

A heritable defect in the suppressive effects of CD4+ CD25+ FOXP3 Tregs appears to be the mechanism underlying the rare lethal autoimmune human disease "immune dysregulation, polyendocrinopathy, enteropathy, X-linked" (IPEX syndrome, MIM 304790). Mutations in the FOXP3 located on the X chromosome have been noted in most of the affected patients [88]. In a mouse model, deficiency of FOXP3 results in an absence of CD4+ CD25+ T cells. In humans with IPEX, the number of Tregs appears to be normal, but their ability to suppress autologous effector T cells is markedly impaired [89]. (See "IPEX: Immune dysregulation, polyendocrinopathy, enteropathy, X-linked".)

●Defects in regulation of autoimmune B cell responses – Like the regulation of T cells, regulation of B cell responses occurs at multiple levels, both centrally and peripherally. As the B cell repertoire develops, autoreactive cells are eliminated at various steps, called checkpoints, by mechanisms that include deletion, anergy, and receptor editing. Receptor editing involves a secondary gene rearrangement to replace an autoreactive immunoglobulin receptor with one of another specificity. Since B cell responses involve the generation of somatic mutants to increase antibody affinity, regulatory interactions are key throughout an immune response since a somatic mutant generated to a foreign antigen could inadvertently bind a self-antigen. Studies on patients with systemic lupus erythematosus indicate disturbances in immune checkpoints, with an increased presence of autoreactive B cells in the preimmune repertoire [90-92].

●Targeting of cell surface and soluble antigens and immune complex formation – Whether clinical manifestations of autoimmune disease will follow from the presence of B or T cell autoreactivity depends upon both the quality of the immune response and the availability of the corresponding antigen. Antigens on the surface of circulating cells such as blood cells are readily available to circulating antibody and, therefore, such cells may be damaged or eliminated by autoantibody acting with complement, cytotoxic T cells, or phagocytes [93]. The receptors on cell surfaces such as the thyroid-stimulating hormone (TSH) receptor on the thyroid cell or the acetylcholine receptor at neuromuscular junctions may also be directly bound by autoantibody [94]. This interaction may result in stimulating the receptor in Graves' disease or in blocking the neuromuscular transmission as observed in myasthenia gravis.

In other diseases, autoantibodies are produced to particular enzymes. As an example, antibodies to the P450 enzymes are prominent in autoimmune hepatitis and primary biliary cholangitis (previously referred to as primary biliary cirrhosis) [95,96]. These enzymes are widely distributed in the body, raising the questions of why the diseases are organ-specific. Evidence suggests that the enzymes may be accessible to antibodies or T cells primarily in reactivated locations.

●Immune complexes – An important mechanism for autoimmune disease is the formation of immune complexes between autoantibodies and the corresponding autoantigen present in the circulation and/or on cell surfaces. For systemic lupus erythematosus, immune complexes of anti-DNA antibodies with DNA can deposit in the kidney and activate complement to produce nephritis [97]. These immune complexes can also enter cells of the innate immune system to stimulate the production of cytokines, most prominently type 1 interferon, by interaction with internal nucleic acid sensors. Autoantibodies to ribonucleic acid (RNA)-binding proteins (RBPs) can also stimulate the production of cytokines since the RNA associated with RBPs can interact with these sensors. The formation of immune complexes depends on the amount of antibody, its properties (eg, avidity), and the availability of self-antigen [98].

The ability of immune complexes to stimulate cytokine production demonstrates an important feature of nucleic acids as targets of autoantibodies. Unlike many autoantigens, DNA and RNA have intrinsic immunostimulatory activity, likely related to the role of internal nucleic acid sensors in host defense; these sensors can detect DNA and RNA in the cytoplasm of cells and, therefore, can respond to intracellular infection from bacteria or viruses [99]. Deficiency of the nucleases that degrade DNA and RNA either intracellularly or extracellularly can also lead to autoimmunity in both patients and animal models. The production of anti-DNA antibodies is a notable consequence of deficiency of deoxyribonuclease (DNase) enzymes, with impaired degradation increasing the amount of self-DNA antigen available to stimulate autoantibody production [100].

●Effector T cell-mediated injury – While many other autoimmune diseases result from direct effects of autoantibodies, some diseases are associated with T cell-mediated immune responses. Sometimes, cytotoxic effector T cells may be generated that can damage their respective target cell [101]. In other cases, cytokines are produced that are harmful to surrounding tissue cells [102]. Finally, T cells may activate macrophages, which can produce a tissue injury through their soluble products, including cytokines and reactive oxygen intermediates. A functionally significant polymorphism of a lymphocyte specific phosphatase known as the protein tyrosine phosphatase non-receptor 22 gene (PTPN22) has been linked to T cell hyperresponsiveness. Carriage of the 1858T variant of PTPN22 is associated with several autoimmune conditions, including rheumatoid arthritis, systemic lupus erythematosus, Graves' disease, and type 1 diabetes mellitus [103].

●Innate immune mechanisms – High-throughput genetic and genomic studies have also focused attention on innate mechanisms in autoimmunity [104]. The innate immune system uses sets of molecules known as pattern-recognition receptors that have been selected through evolution to recognize molecular patterns found in microorganisms. Families of pattern-recognition receptors include the toll-like receptors, the nucleotide oligomerization domain-like receptors, and the NACHT leucine-rich-repeat proteins (NALPs) [105,106]. (See "An overview of the innate immune system".)

Both NALP1 and NALP3 comprise part of cytoplasmic complexes called inflammasomes that regulate the activation of caspase 1, which converts ILs into their active forms [107]. Variants of NALP1 are associated with autoimmune disorders that cluster with vitiligo (eg, autoimmune thyroid disease, latent autoimmune diabetes mellitus in adults, rheumatoid arthritis, psoriasis, pernicious anemia, and Addison disease) [108].

●Roles of T cell subsets – In human disease, it is frequently impossible to clearly separate the injury that is from antibody-mediated in contrast to cell-mediated reactions. In general, however, the T cell-mediated responses are associated with the T helper (Th) 1 subset of CD4+ T cells [109]. Antibody-mediated mechanisms, in general, associate with Th2 responses. Thus, some autoimmune disorders may benefit by a shift from a Th1 to a Th2 response, which can be mediated therapeutically.

This original simplistic view of the Th1/Th2 dichotomy is not easily applied to human disease in which both Th1 and Th2 responses are generally evident. In fact, some autoimmune diseases are actually downregulated by administering interferon-gamma, the prototypic Th1 cytokine [110]. It is now evident that a third T cell subset producing IL-17 (termed Th17) plays a critical role in several autoimmune conditions. Interestingly, there is a reciprocal relationship between Th17 and Tregs that may dictate the final outcome of an autoimmune reaction. Inflammatory cytokines TGF-gamma and interferon-gamma play central roles in setting the balance [111].

AUTOINFLAMMATORY DISEASES — The term "autoinflammatory disease" refers to a group of mostly inherited disorders that resemble in some respects the traditional autoimmune diseases in terms of the presence of inflammation as well as multisystem manifestations [112]. These disorders are distinguished by the absence of a defined adaptive immune response. The prototypic disorders were in a group of "hereditary periodic fever syndromes" that commonly share involvement of certain specific cytokines such as interleukin (IL) 1-beta and tumor necrosis factor (TNF). Study of these disorders have revealed a prominent role for the innate immune system. The category of autoinflammatory diseases has gradually expanded to include a group of disorders sharing an abnormally increased inflammatory response mediated predominantly by cells and molecules of the innate immune system with a significant host predisposition. (See "The autoinflammatory diseases: An overview".)

Mutant NACHT leucine-rich-repeat protein 3 (NALP3), also known as cryopyrin, is linked to three inflammatory disorders associated with autosomal dominant inheritance: familial cold urticaria syndrome, Muckle-Wells syndrome, and neonatal-onset multisystem inflammatory disease (see "Cryopyrin-associated periodic syndromes and related disorders"). A related group of diseases is characterized by increased levels of interferon. These diseases are termed interferonopathies and result from mutations in a variety of genes [113]. (See "Autoinflammatory diseases mediated by interferon production and signaling (interferonopathies)".)

VACCINES — The protective value of vaccines in patients with autoimmune disease is well established based on sound epidemiologic data, while the possibility that vaccines may induce or exacerbate autoimmune disease remains speculative [114,115].

While immunization is a powerful approach to reduce the occurrence of infectious diseases, concerns about safety have been raised especially in the lay press. An important component of certain vaccines is a substance known as an adjuvant of which alum is commonly used. An adjuvant can trigger the innate immune system to promote the immune response to the vaccinating antigen, raising the possibility that a response to a self-antigen can also be induced. In general, patients with an autoimmune disease have increased immune reactivity, but there are no well-designed epidemiologic studies showing a significant increase in any autoimmune disease following vaccination of these patients including coronavirus disease 2019 (COVID-19) RNA vaccination [116].

Other studies fail to show severe systemic adverse effects, including autoimmune disorders, with immunization using adjuvant compared with nonadjuvanted vaccines. Indeed, many patients with autoimmune disease are relatively poor immune responders to vaccine antigens and may require additional boosters. (See "Immunizations in autoimmune inflammatory rheumatic disease in adults".)

SUMMARY

●Definition – Autoimmune disease is a pathologic condition caused by an adaptive autoimmune response directed against an antigen within the body of the host, termed a self-antigen. The response may be induced by a foreign or self-antigen. It usually involves both a T and B cell response. (See 'Definition' above.)

●Features of autoimmune disease – The demonstration of autoantibodies is usually the first step in the diagnosis of an autoimmune disease, although it is not sufficient. The antibodies may not be the actual mediators of disease. Naturally occurring autoantibodies can occur in immunologically competent people and may even rise nonspecifically during the course of disease or injury. (See 'Features of autoimmune disease' above.)

●Evidence of causality – There are various levels of evidence in studies of the causality of autoimmune disease. Direct evidence requires demonstration that autoimmune response can produce the disease. Indirect evidence typically involves manipulation of the immune system or study of naturally occurring disease in animal models. Finally, circumstantial evidence is most common and includes studies on the presence of autoantibodies, clustering of diseases within an individual or family, bias to certain human leukocyte antigen (HLA) haplotypes, a sex bias, and a response to immunosuppressive therapy. (See 'Direct evidence' above and 'Indirect evidence' above and 'Circumstantial evidence' above.)

●Inductive mechanisms – An autoimmune response can be initiated by autologous or foreign antigens. Like all immune responses, autoimmunity appears under strict genetic control of antigen recognition, cellular interactions, and eventual outcomes. (See 'Inductive mechanisms' above.)

●Pathogenic mechanisms – The development of autoimmune disease depends upon an imbalance between pathogenic factors generated by autoreactive T and B cells and the regulatory factors that normally control the immune response. For example, the balance of activities between various effector T cells and regulatory T cells (Tregs) helps to determine whether an individual maintains normal self-tolerance or progresses to autoimmune disease. (See 'Pathogenetic mechanisms' above.)

●Autoinflammatory diseases – Diseases in which there is chronic inflammation but no evidence of autoreactive T or B cells, termed autoinflammatory diseases, are associated with the innate immune response. (See 'Autoinflammatory diseases' above.)

●Vaccines – The protective value of vaccines in patients with autoimmune disease is well established based on sound epidemiologic data, while the possibility that vaccines may induce or exacerbate autoimmune disease remains speculative. In general, patients with an autoimmune disease have increased immune reactivity, but there are no well-designed epidemiologic studies showing a significant increase in any autoimmune disease following vaccination of these patients. (See 'Vaccines' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Noel R Rose, MD, PhD, now deceased, who contributed to an earlier version of this topic review.