Chapter 154. Mechanisms of Autoimmune Disease

Autoimmune diseases including rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE) are relatively common disorders.1 Although the underlying etiologies of these illnesses are still elusive, they arise in the context of a break in the immune tolerance to self.2,3 The mechanisms for abrogation of immune self-tolerance appear to be multifactorial, including genetic and environmental, that work in concert to initiate the eventual hallmarks of disease: unregulated immune activation against self-antigens and subsequent tissue destruction. Immune activation against self-antigens is clinically manifest by the presence of autoantibodies and autoreactive T cells.1,2 Based upon their autoantigenic targets, autoimmune diseases can be classified into organ-specific and systemic autoimmune processes. For example, Grave's disease, with its autoantibodies against the thyroid-stimulating hormone (TSH) receptor, is a typical example of organ-specific autoimmune disease, as is type I diabetes mellitus (DM) with its autoantibodies and autoreactive T cells directed against components of pancreatic β cells, whereas SLE with its characteristic autoantibodies against ubiquitous nuclear antigens is a good example of a systemic autoimmune disease. Although adaptive immune cells such as B cells and T cells recognize self-antigens and hence dominate the phenotype of the patient with autoimmunity, other immune components including antigen-presenting cells (APC) and complement are involved in various steps from initiation of the autoimmune response to tissue destruction.4–6 In this chapter, the contributions of these components to the development of autoimmune diseases will be discussed, focusing on the latest discoveries.

Autoimmune Diseases Are Polygenic

Autoimmune diseases are polygenic, involving both major histocompatibility complex (MHC) and non-MHC genes.7,8 Yet, the concordance rate of autoimmune diseases in monozygotic twins is not 100%, ranging from 10% to 50%, including in RA and SLE,7,9 indicating a significant role for nongenetic factors in the development of these disorders. Nevertheless, genetic influences are strong. For instance, the contribution of alleles of MHC to the development of autoimmunity has been known for more than 30 years. An increased frequency of HLA (human leukocyte antigen)-DR4 has been observed in some patients with RA.8 Using more sophisticated molecular techniques, the sequence of genetic alleles encoding HLA loci have been typed, with the resultant demonstration that the HLA-DRB1 gene is highly polymorphic and such polymorphisms can affect the binding of peptides to HLA molecules and the contacts between the T-cell receptor (TCR) and the HLA molecule.8 Thus, the associations between autoimmune diseases and particular HLA molecules can be explained by a model where disease susceptibility is determined by differences in the ability of different HLA alleles to present autoantigenic peptides to autoreactive T cells.2,8 However, the genetic effect of MHC on disease propensity is broader. As an example, a recent study genotyped a panel of 1,472 single nucleotide polymorphisms (SNPs) across the 3.44 megabase (Mb) classic MHC region in 10,576 DNA samples from patients with autoimmune diseases including SLE, RA, and multiple sclerosis, and from appropriate controls.10 The results of this study showed multiple risk alleles across MHC class I, II, and III in these autoimmune diseases, indicating complex, multilocus effects that span the entire region.

Non-MHC Associations with Autoimmune Diseases

The association of non-MHC genes with autoimmune diseases is now well established,1,7,11 with the recent advent of genome-wide association (GWA) scans that are powerful tools for identifying new genes in autoimmunity.11 These genes include those involved in antigen clearance, apoptosis, cell signaling, and cytokine production as well as in the expression of costimulatory molecules and cytokine receptors.1,2,7,11 For instance, complement components are required for proper clearance of immune complexes, and an increased incidence of homozygous C4 deficiency is found in patients with SLE who have immune complex deposits in damaged organs such as the kidneys.9

Autoimmune Regulator

Recently, a gene called autoimmune regulator (AIRE) gene has been identified as a candidate gene responsible for the development of autoimmune diseases.12,13 AIRE is a transcription factor that regulates the ectopic expression of proteins, normally expressed in peripheral tissues, in the thymus, allowing for thymic expression of the latter and subsequent negative selection of self-reactive thymocytes before they migrate as mature T cells to the secondary lymphoid organs such as the spleen and lymph nodes. An alteration in thymic expression of AIRE can lead to increased generation of autoreactive T cells due to their impaired negative selection.14 Indeed, patients with mutations of the AIRE gene develop a syndrome called autoimmune polyendocrinopathy candidiasis ectodermal dystrophy (APECED) or autoimmune polyglandular syndrome 1 (APS-1) that is characterized by chronic mucocutaneous candidiasis, hypoparathyroidism, and Addison's disease.15 In addition, patients with this condition often have other organ-specific autoimmune diseases including type 1 DM, autoimmune thyroid diseases, gonadal failure, vitiligo, alopecia, dystrophy of dental enamel and nails, and pernicious anemia.15

Cytotoxic T-Lymphocyte Antigen-4

An increased incidence of autoimmune diseases has been reported in individuals with a particular variant of genes that affect T-cell activation. These include the cytotoxic T-lymphocyte antigen-4 (CTLA-4) gene and the protein tyrosine phosphatase 22 gene (PTPN22)1,11,16 (see below). CTLA-4 is a key immunoregulatory molecule that restrains T-cell activation. It is expressed on T cells upon TCR stimulation by antigen on APC and competes with CD28, a positive costimulatory molecule also on T cells, for binding to CD80 and CD86 expressed on APC.17 By contrast to CD28 engagement by CD80 and CD86, binding of CTLA-4 to the latter molecules inhibits T-cell activation. The association of SLE and RA with SNPs in the CTLA-4 promoter and coding region has been reported. These polymorphic sites include −1722 position of the CTLA-4 promoter and +49 position of exon-1,18,19 although the functional consequence of such gene polymorphisms is not clear yet. CTLA-4 polymorphisms also have been linked to autoimmune endocrinopathies; for example, a decrease in the alternate splice product of CTLA-4 that is the soluble form of CTLA-4 is found in individuals with CTLA-4 SNP CT60 G/G which is linked with increased susceptibility to Grave's disease, autoimmune hypothyroidism, and type 1 DM.20 This finding suggests a possible functional role for CTLA-4 polymorphisms in autoimmune diseases.

Programmed Cell Death 1

In a manner analogous to CTLA-4, a protein called programmed cell death 1 (PD-1) is expressed on T cells providing an inhibitory signal upon their activation. This molecule is also expressed on other immune cells including B cells and myeloid cells.1,21 PD-1 is encoded by the PDCD1 gene, with a SNP in its fourth intron associated with autoimmunity, including SLE and RA.22–24 Of interest, this disease-associated SNP impairs the binding of the hematopoietic runt-related transcription factor 1 (RUNX1), leading to aberrant expression of PDCD1 gene and subsequently uncontrolled T-cell activation.22 The association of genetic polymorphisms of CTLA-4 and PD-1, molecules involved in regulating T-cell activation, with autoimmune diseases supports the notion that regulation of T-cell activation by these costimulatory molecules is a critical checkpoint for the development of autoimmunity.

Protein Tyrosine Phosphatase 22

The protein tyrosine phosphatase 22 (PTPN22) gene is another recently identified gene that is associated with autoimmune diseases.1 A variant of the PTPN22 gene, which encodes a tryptophan at codon 620 (620W) instead of an arginine (wild-type), has been found to be associated with an increased risk of RA, SLE, type 1 DM, and Grave's disease.1 Such a mutation appears to have functional consequences, since the PTPN22 gene encodes lymphoid tyrosine phosphatase (LYP) that modulates the activation of Lck and other kinases involved in TCR signaling.25 Indeed, the PTPN22 620W allele appears to potentially increase the inhibition of TCR activation by downregulating Lck activation.25 Thus, it still needs to be clarified how this allele promotes the development of autoimmunity.

Tumor Necrosis Factor Receptor Signaling Pathway

Recently, multiple genes encoding molecules involved in tumor necrosis factor (TNF) receptor signaling have been found to be associated with autoimmunity.11 The best known is TNFAIP3 that encodes TNF inducible protein A20 that is a negative regulator of TNF-induced NF-kB signaling pathways. SNP markers in the gene region near the TNFAIP3 locus have been associated with RA, psoriasis, and SLE.26–28 The gene encoding TNFAIP3-interacting protein 1 (TNIP1) that interacts with TNFAIP3 is associated with psoriasis and SLE,27,29 with an association of the TRAF1 (TNF receptor-associated factor 1) gene with RA also reported.30

Cytokine Receptors

Polymorphisms in several cytokine receptors are associated with autoimmunity.11 The IL-23 receptor complex consists of IL-23R and IL-12Rβ1 subunits, with the latter shared with the IL-12 receptor complex that is composed of IL-12Rβ1 and IL-12Rβ2 subunits. The IL23R receptor gene is associated with inflammatory bowel disease and psoriasis.31,32 This is an intriguing point since IL-23 is involved in regulating the development of T helper 17 (Th17) cells that produce the potent proinflammatory cytokine IL-17. In fact, IL-17 is found in psoriatic skin lesions.33 Likewise, the IL12B gene that encodes IL-12β (the p40 subunit of IL-12) has been found to associated with psoriasis.34

Receptors for Immunoglobulin G Fc Portion (FcγR)

In addition to genes expressed by T cells, those involved in humoral immune responses also appear to be associated with autoimmunity. The receptors for immunoglobulin G (IgG) Fc portion (FcγR) are expressed on B cells, macrophages, monocytes, granulocytes, and dendritic cells (DCs). There are eight genes identified for the human FcγR: FCGRIA, FCGRIB, FCGRIC, FCGRIIA, FCGRIIB, FCGRIIC, FCGRIIIA, and FCGRIIIB.35 The binding of FcγRs to IgG, except for that to FcγRIIb, results in a broad spectrum of activating cellular responses including phagocytosis, cytolysis, cytokine production, and degranulation that lead to inflammation. By contrast, FcγRIIb physiologically serves as an inhibitory receptor that downregulates the effector function of cells.35 An increased frequency of some variants of FCGR genes has been reported in humans with autoimmune diseases including SLE and RA.35–39 FcγRIIa with an arginine at position 131 has greater signaling in response to IgG1 engagement than FcγRIIa with a histidine at the same position, suggesting that FcγRIIa gene polymorphisms have physiologic relevance.39 Of interest, FcγRIIa with an arginine has been reported as a susceptibility factor for SLE in some ethnic groups.35,40 Similarly, a variant of FcγRIIIa with a phenylalanine at position 158 may be a susceptibility factor for SLE and RA.36,38 In addition, genetic polymorphisms of the inhibitory FcγRIIb have been reported in patients with SLE.37,41,42 The best known such polymorphism is a substitution of an isoleucine at position 232 (187 by counting from the N-terminus of the mature protein) with a threonine.37,41,42 The latter genotype is more commonly reported in patients with SLE. While the biologic implication of such an alteration on the development of SLE is as yet unclear, functioning inhibitory FcR are necessary for dampening autoimmunity in murine lupus.43

Overall, genetic studies on MHC genes and non-MHC genes support the notion that some patients with autoimmune diseases have an increased genetic risk for the development of autoimmune diseases secondary to genetic alterations affecting immune cell function. However, such alterations are not always observed in patients with autoimmune diseases and ethnic differences in genetic susceptibility are not uncommon. These findings support the idea that autoimmune diseases are polygenic and that environment factors are also involved in their development (Fig. 154-1).

Central Tolerance

Intact mechanisms of self-tolerance are necessary to prevent the development of autoimmune diseases.2 Self-tolerance is generated and maintained through the capacity of the immune system to distinguish self-reactive B and T cells from nonself-reactive cells. These processes begin during the development of lymphocytes in the bone marrow and thymus, respectively.3 Such selective processes are controlled by the binding affinity of antigen receptors to self-antigens during lymphocyte development. B and T cells expressing antigen receptors with affinity for self-antigens are altered, either becoming nonreactive to antigenic stimuli (through a process called anergy induction) or changing their antigen receptors so that they no longer bind self-antigens (a process called receptor editing, operative in self-reactive B cells), or if they bind too strongly to self, they are simply killed via apoptosis. These interventions lead to central tolerance to self-antigens.3 A defect in central tolerance can cause autoimmune diseases as evidenced by the development of APS-1 in humans with an alteration in the AIRE gene (see Section “Genetics”) that is involved in negatively selecting autoreactive T cells.44,45

Peripheral Tolerance

Despite removing or modifying autoreactive lymphocytes in the thymus and bone marrow by selective processes, some lymphocytes reactive to self-antigens still mature and enter the peripheral immune system; i.e., negative selection is not perfect, probably because if it were, potentially beneficial T and B cells would never make it past central selection in the thymus and bone marrow, respectively. The autoreactive T and B cells that escape central selection can potentially be activated upon recognition of self-antigens; thus, peripheral tolerance mechanisms need to be available to avoid the development of autoimmunity. Such mechanisms are several, and involve minimizing contacts of autoreactive cells with self-antigens (ignorance of the self-reactive lymphocyte), not providing appropriate signals for their activation, terminating activated autoreactive cells by regulatory molecules such as PD-1 or CTLA4, and/or actively suppressing autoimmune responses by other T cells.2 In critical organs like the brain, eyes, and gonads, immunological ignorance is a principal mechanism for avoidance of autoreactivity via the provision of anatomical barriers separating the tissue and lymphocytes. Activation of the latter can be further prevented by the relative paucity of APCs in these organs and the minimal expression of MHC molecules in these tissues.46,47 These sites are termed immunologically privileged since even tissue grafts do not elicit immune responses therein. Immunologically privileged sites have additional mechanisms that prevent immune activation, including production of TGF-β, which can suppress immune responses such as in the anterior chamber of the eyes, and constitutive expression of Fas ligand that induces apoptosis of infiltrating activated lymphocytes.46,47

Nevertheless, since autoantigens are expressed on some tissues and autoreactive T (and B) lymphocytes are present in the circulation, there is still a reasonable chance for the latter to recognize autoantigens expressed by MHC molecules on APC. However, such an autoantigenic signal through potentially self-reactive antigen receptors is not enough to activate lymphocytes, as additional signals (second signals) provided by APCs, such as CD80 and CD86, to CD28 on T cells are required for proper activation.17,48 When proper costimulation is not provided by antigen presenting cells, autoreactive T cells that are only signaled through the TCR undergo apoptosis or become anergized. As CD80 and CD86 upregulation on APCs requires an inflammatory stimulus provided to the APC such as that provided by a pathogen during infection, presentation of self-antigens in the absence of inflammation does not lead to autoreactive T cell activation. By contrast, peptides of pathogens lead to cellular activation as they are presented to the immune system in the appropriate inflammatory context. Of course, this regulatory control on autoimmunity may be bypassed if in a similar manner self-antigens are presented at a site of inflammation. Fortunately, this appears to be a relatively unusual event, and autoimmunity invoked by this mechanism is typically avoided.

Autoreactive lymphocytes that manage to get activated in the periphery can be also deleted by apoptosis (programmed cell death) and/or their activation terminated by inhibitory molecules expressed on autoreactive lymphocytes. Fas ligand, or Fas, expressed on T cells is largely responsible for the former.49,50 The latter process can be achieved through CTLA-4 and PD-1 expressed on activated T cells that inhibit and downregulate T cell activation.17,21 The potential roles for CTLA-4 and PD-1 in controlling autoimmunity have been demonstrated by mouse studies where these genes have been genetically eliminated, for example, with resultant autoimmunity.51,52 In addition, genetic polymorphisms in these molecules have been reported to be associated with autoimmune diseases including SLE and RA (see Section “Genetics”). More importantly, there have been attempts to treat autoimmune diseases by modulating CTLA-4 and PD-1 inhibitory pathways.53,54 Indeed, abatacept, a recombinant fusion protein comprising the extracellular domain of human CTLA4 and a fragment of the Fc domain of human IgG1, is available for the treatment of RA.53 This drug, like CTLA4, competes with CD28 for CD80 and CD86 binding; however, in contrast to CTLA4 engagement that leads to downregulation of T cell activation, it blocks T cell activation and thus selectively modulates T cell activation.

The activation of autoreactive lymphocytes is also actively suppressed through other mechanisms. The most appealing such mechanism is immune suppression through subsets of T cells called regulatory T cells (Tregs), discussed below.

Regulatory T Cells

The concept that certain T cells can regulate immune responses dates back to early 1970s. These cells, that Gershon initially called “suppressor cells,”55 remained largely undefined and their mechanism of suppression not identified. After their original discovery, they remained out of vogue until over a decade later when others including Sakaguchi reported modulation of immune responses by CD4+ T cells expressing CD25 (IL-2 receptor alpha chain)56 as well as by T cells producing TGF-β and IL-10.57–59 Cells with immune regulatory functions are now called regulatory T cells (Tregs). Several subsets of Tregs have been identified thus far, including: (1) naturally occurring CD4+ CD25+ Tregs expressing the transcription factor protein forkhead box P3 (Foxp3); (2) CD4+ T cells producing IL-10 called type 1 regulatory T cells (Tr1); (3) CD4+ T cells producing TGF-β named T helper 3 cells (Th3); and (4) CD8+ T cells producing IL-10 or TGF-β.60,61

CD4+ CD25+ Regulatory T Cells

Among the best characterized Tregs are naturally occurring CD4+ CD25+ Foxp3+ cells (FOXP3+ Treg) that are produced in the thymus. The potential role for these Tregs in controlling the development of autoimmunity was initially demonstrated by studies where mouse T cell suspensions were transferred into congenic athymic nude mice. If the latter animals received such suspensions depleted of CD4+ CD25+ T cells, they developed an autoimmune syndrome including thyroiditis, gastritis, insulitis, sialoadenitis, adrenalitis, oophoritis, glomerulonephritis, and polyarthritis.56 By contrast, athymic nude mice that received whole T-cell suspensions did not develop autoimmunity, indicating a role for CD4+ CD25+ T cells in preventing autoimmune diseases.56 Subsequently, the inhibitory effects of CD4+ CD25+ Tregs in different types of animal models of autoimmune diseases including collagen-induced arthritis (CIA) and autoimmune diabetes were documented.62,63 Yet, not all CD4+ CD25+ T cells have immunoregulatory functions as CD25 is upregulated upon T-cell activation. Thus, the search for other markers that specifically identify these cells has been extensive, and has led to the identification of CD39, CD103, CTLA-4, glucocorticoid-induced TNF receptor (GITR), lymphocyte activation gene 3 (LAG-3), interleukin 7 receptor, and Foxp364,65 as potential markers. The most promising of this group is Foxp3. Disruption of its gene in mice results in autoimmune diseases,66,67 whereas overexpression of Foxp3 in T cells induces expansion of CD4+ T cells with immunosuppressive function.66–68 Furthermore, CD4+ CD25− T cells transfected with FoxP3 acquire immunosuppressive function.66,67 Although it was originally thought that FoxP+ Tregs develop only in the thymus, studies indicate that CD25− CD4+ T cells in the periphery can become FoxP3+ Treg-like cells with immune regulatory function in the presence of T-cell receptor triggering, IL-2, and TGF-β.69 The latter cells are now called adaptive or inducible FoxP3+ Treg (FOXP3+ iTreg) whereas the former cells are called naturally occurring FoxP3 Treg (nTreg).69

Although the exact mechanism(s) by which naturally occurring FoxP3+ Tregs suppress autoreactive T cells is not fully understood, cell-to-cell contact is necessary and CTLA-4 appears to be required.70,71 CTLA-4 expressed on Tregs interacts with CD80 and CD86 expressed on APC as well as on the target T cells72,73; while CD80 and CD86, ligands for CTLA-4, are conventionally expressed on APC as noted above, they have also been found on activated CD4+ T cells.72,73 These molecules likely interact with CTLA-4 expressed on FoxP3+ Treg, providing “outside-in” signaling leading to immune suppression. IL-2 appears to be involved in inducing and maintaining these cells.71 IL-2 can upregulate FoxP3 expression through activating STAT5.74,75 Of interest, CD25, the IL-2 receptor α chain, could be involved as a suppressive mechanism of FoxP3+ Tregs by absorbing IL-2 and subsequently reducing the availability of this cytokine to other T cells.76 These Tregs also express CD39 (ectonucleoside triphosphate diphosphohydrolase 1) and CD73 (ecto-5-nucleotidase). These molecules generate pericellular adenosine with immunosuppressive activity by catalyzing extracellular nucleotides.77 FoxP3+ Tregs can suppress T cell function by inducing the production of the enzyme indoleamine 2,3-dioxygenase (IDO) from DCs. IDO catabolizes conversion of the essential amino acid tryptophan to kynurenine that is harmful to T cells.71,78 Overall, FoxP3 Tregs likely employ multiple mechanisms in suppressing other T cell function.

By contrast to animal studies, the data on CD4+ CD25+ Tregs in human autoimmune diseases is relatively minimal. In patients with type 1 DM, a typical example of an organ-specific autoimmune disease, the frequency of CD4+ CD25+ Tregs has been shown to be significantly lower than in healthy controls and patients with type 2 DM, suggesting a role for these cells in disease development.79 However, in separate studies, no difference in the frequency of CD4+ CD25+ Tregs and FOXP3 gene allelic variation has been reported in patients with type 1 DM compared to healthy controls,80,81 although the inhibitory function of such cells was impaired.81 Similarly, there was no difference in the frequency of CD4+ CD25+ T cells and their functional markers including Foxp3 among patients with APS-II characterized by Addison's disease, type I diabetes and autoimmune thyroid disease as well as control patients with single autoimmune endocrinopathies and normal healthy donors.82 However, CD4+ CD25+ Tregs from APS-II patients were defective in their suppressive capacity.82 These findings suggest that patients with autoimmune endocrinopathies including type 1 DM and APS-II may have an alteration in the suppressive function of CD4+ CD25+ Tregs.

A role for CD4+ CD25+ Tregs in the development of SLE and RA, typical systemic autoimmune diseases, has also been explored. Most studies on CD4+ CD25+ Tregs in human lupus reported a decreased frequency of this cell subset,83 suggesting their potential role in the development of SLE. In RA, CD4+ CD25+ Tregs from both peripheral blood and synovial fluid have been studied.84–86 The frequency of CD25+ CD4+ Tregs in the latter fluid was higher than that in peripheral blood,85,86 although there was no significant difference in the numbers of CD4+ CD25+ Tregs in peripheral blood between patients and controls.85,86 Of interest, compromised function of CD4+ CD25+ Treg cells in suppressing inflammatory cytokines from CD4+ CD25− T cells was reported in RA.87 Furthermore, such impairment was reversed after treating patients with anti-TNF-α therapy. These findings suggest that patients with RA have a functional, but not a numerical, defect in CD4+ CD25+ Tregs.

Other Regulatory T Cells

In contrast to CD4+ CD25+ Tregs that naturally occur, other regulatory T cells can be induced in vivo and in vitro.61 These cells include Tr1, Th3 (CD4+ T cells producing large amounts of TGF-β) and CD8+ CD28− T cells that develop in the setting of immune stimulation such as that initiated by chronic infections. For instance, humans chronically infected with hepatitis C or Epstein Barr viruses, or with M. tuberculosis or the nematode Onchocerca volvulus, may develop pathogen-specific Tr1 cells that produce large amounts of IL-10 that can suppress antigen-specific or nonspecific T-cell responses.88–91 These cells can also be generated after ingestion of a foreign antigen via the oral route.57 The immunosuppressive mechanisms employed by Tr1 and Th3 cells are dependent on IL-10 and TGF-β production rather than direct cell-to-cell contact as in natural Tregs.61

CD8+ T cells may also have immunosuppressive activity.60 Mice deficient in CD8+ T cells develop experimental autoimmune encephalomyelitis (EAE), an animal model of multiple sclerosis.92 However, EAE in the same animals was milder when they received adoptive transfers of CD8+ CD28− T cells. Moreover, CD8+ CD28−, but not CD8+ CD28+, T cells suppressed in vitro IFN-γ production by CD4+ T cells specific for myelin oligodendrocyte glycoprotein, an autoantigen in EAE. Such suppression required cell-to-cell contact and was mediated by inhibition of upregulation of costimulatory molecules on APCs that led to decreased costimulation of CD4+ T cells. Of interest, a recent study suggested a potential role for CD8+ T cells with suppressor activity in improving lupus-like disease in mice.93 Decreased levels of pathogenic anti-dsDNA antibodies with a resultant increase in survival were found in lupus-prone (NZB × NZW)F1 mice when injected with an artificial peptide called pConsensus (pCons) that was developed based on T cell stimulatory VH sequences found in (NZB × NZW)F1 anti-DNA antibodies. This protection depended in part on the generation of peripheral TGFβ− and Foxp3-expressing inhibitory CD8+ T cells, and suggested that CD8+ T cells with regulatory activity could be a potential way to modulate autoimmune diseases.

T Helper 17 (Th17) Cells

IL-17 (IL-17A), a potent pro-inflammatory cytokine, has an important role in defending the mammalian host against challenge with extracellular bacteria and fungi. IL-17 is made primarily by a unique CD4+ T cell subset called T helper 17 (Th17) cells, although it is also produced by CD8+ and γδ T cells94–96 as well as by CD3+CD4−CD8− (double negative or DN) T cells and NK cells.97,98 Cytokines including IL-1b, IL-6, TGF-β, IL-21, and IL-23 are critical for the development and expansion of Th17 cells.99–102 In fact, studies reported that CD4+ T cells expressing IL-1 receptor I or IL-23 receptor had the strong capacity to produce IL-17.103,104 The development of Th17 cells is regulated by the transcription factors RORγt (retinoic-acid related orphan receptor) (the human ortholog is RORC), RORα,105 and IRF4.106 In addition to IL-17, Th17 cells produce IL-17F, IL-22, and IL-26 as well as chemokine CCL20.104,107,108 Th17 cells characteristically express CD161, a c-type lectin receptor associated with natural killer cells, as well as chemokine receptors CCR4 and CCR6.109 These molecules have been used to purify Th17 cells.

IL-17 induces cytokines (IL-6, IL-8, GM-CSF, and G-CSF), chemokines (CXCL1 and CXCL10), and metalloproteinases from target cells including epithelial cells and fibroblasts.110 IL-17 potently recruits and activates neutrophils through inducing GM-CSF and IL-8 production,110 leading to neutrophil-mediated inflammatory responses. A potential role for IL-17 in autoimmunity has been demonstrated through mouse studies of EAE and CIA, models for multiple sclerosis and RA, respectively,111–113 as well as murine lupus models.114,115 Increased levels of IL-17 in blood and tissues as well as an increased frequency of peripheral Th17 cells were reported in patients with psoriasis, inflammatory bowel disease, and SLE,97,116–121 suggesting a pathogenic role in human inflammatory diseases. A recent study reported that IL-17 could promote the survival and proliferation of human B cells in conjunction with B-cell activating factor (BAFF).122 This can be an additional mechanism of how IL-17 promotes inflammatory autoimmune diseases like lupus and MS where autoantibodies have a pathogenic role.

Dendritic Cells and Toll-Like Receptors

Various cells of the innate immune system appear to be involved in the regulation of autoimmunity, including DCs, natural killer cells, γδ T cells, natural killer T cells, and even mast cells.123–128 Among the best characterized are DCs, or professional APCs.129 DCs present antigens to T cells for their activation, but their requirement to provide costimulation via CD80 and CD86 limits autoreactive T-cell activation in the absence of inflammatory signals. Interaction of CD80 on DCs and CD28 on T cells polarizes TCR-triggered naive T cells into T helper 1 cells producing IFN-γ. In addition, activated DCs produce multiple cytokines including IL-12, TNF-α, and IL-6 that can promote inflammation. DCs express Toll-like receptors (TLRs), or pattern recognition receptors that recognize various pathogen-associated molecular structures including lipopolysaccharide, dsRNA, ssRNA, and hypomethylated cytosine and guanosine sequence (CpG) DNA.130 Thus, microorganisms can potentially promote autoimmunity and inflammation through these receptors by activating DCs.129 Activated DC can also affect CD4+ CD25+ Treg function. Microbial activation of TLRs blocks the suppressive effect of CD4+ CD25+ Treg cells, allowing activation of pathogen-specific adaptive immune responses.131 This block of suppressor activity is dependent in part on interleukin-6, which was induced by TLRs upon recognition of microbial products.

DCs and TLRs are also likely involved in the development of SLE. Rönnblom and Alm have demonstrated that immune complexes in lupus serum containing autoantibodies associated with apoptotic fragments of cells including DNA or RNA can stimulate plasmacytoid DCs (pDCs) to produce interferon-α (IFN-α) with resultant autoimmune inflammation.5,132 Such events can be mediated by TLRs.133,134 pDCs express TLR9 that recognize CpG DNA130; thus, immune complexes in lupus serum containing CpG DNA and anti-DNA antibodies can stimulate pDC to produce IFN-α, which promotes autoimmunity (see below). This notion is supported by two studies showing that engagement of B cell receptors binding self-Ig (rheumatoid factor) complexes with DNA or DNA per se with their cognate ligand activates B cells by subsequent delivery of the nucleic acid TLR9 in endosomes.135,136 A similar concept has been raised by recent studies demonstrating the role of TLR7 and TLR8 in cellular responses to ssRNA137,138 since antibodies against small nuclear RNA and protein complexes are often found in lupus patients.133,134

Drugs and Autoimmunity

Certain drugs can induce autoimmune diseases, with potentially protean clinical manifestations.139 Among the best-known inducers of autoimmunity are procainamide and hydralazine that can lead to drug-induced lupus in a minority of patients who take these agents. Although the exact mechanisms that promote procainamide- and hydralazine-induced lupus are not fully understood, recent studies suggest that alterations in DNA methylation, a process critically involved in gene regulation, by these drugs could be responsible for the development of lupus.140,141 Procainamide and hydralazine can decrease DNA methylation in T cells, resulting in increased expression of genes such as lymphocyte function-associated antigen (LFA-1) which can potentially promote autoimmunity.142,143 Evidence from mice also suggests that procainamide may alter central (thymic) tolerance in T-cell selection, leading to escape of autoreactive T cells into the periphery.144

Interferon-α and Autoimmunity

Ironically drugs that modulate the immune system, for example IFN-α, can also induce autoimmunity. In patients treated with IFN-α for hepatitis C or for malignancies, drug-induced lupus with nephritis and antineutrophil cytoplasmic antibodies (c-ANCA)-associated vasculitis have been reported.133,134,139 In fact, the potential role of IFN-α in the pathogenesis of lupus was suggested in late 1970s by a study reporting increased serum levels of IFN-α in lupus patients.145 This notion has been recently revived by several groups that have demonstrated increased expression of interferon-induced genes in peripheral blood cells from lupus patients, using microarray analyses.146,147

IFN-α can affect innate and adaptive immune cell function. While it has antiproliferative effects on T cells,148 at the same time it enhances immunoglobulin isotype switching by stimulating DCs.149 IFN-α also induces maturation of DCs and drives monocytes to become more effective in stimulating T cells, including those that are potentially autoreactive.150–152 The results of studies in mice also provide evidence for a role of this cytokine in the development of lupus. Administration of poly(inosine-cytosine), a synthetic dsRNA and strong IFN-α inducer, accelerated development of disease in lupus-prone (NZB × NZW)F1 mice.153 Lupus-prone mice lacking receptors for type I IFN including IFN-α had reduced anti-dsDNA antibodies and disease activity.154

Toxins and Autoimmunity

Environmental chemicals and toxins also have been reported as a potential cause for autoimmune diseases.155 Probably the best-known chemical is crystalline silica that has been reported to be associated with the development of systemic sclerosis.155 In addition to this condition, there are patients with SLE and RA who were exposed to crystalline silica,155 although such epidemiologic data is difficult to provide evidence for causation. On the other hand, lupus-prone (NZB × NZW)F1 mice exposed to silica have increased levels of antinuclear antibodies, with decreased survival compared to mice not exposed to this chemical.156 Although the exact mechanism for the development of autoimmune diseases after crystalline silica exposure is unknown, this chemical could serve as an adjuvant promoting immune responses with an increase in the production of proinflammatory cytokines such as TNF-α.157 Although there are case reports of autoimmune diseases associated with the exposure to toxic chemicals as well as animals with autoimmune diseases aggravated by such chemicals, additional studies are needed to define the causal relationship between chemical exposure and autoimmunity.

Infections and Autoimmunity

Infection has been suggested as a possible causative factor for autoimmunity. Postulated mechanisms include: (1) release of inflammatory cytokines such as IFN-α by APC that can modulate immune responses; (2) production of cross-reactive antibodies or T cells that can recognize both autoantigens and foreign antigens (molecular mimicry); and (3) polyclonal activation of autoreactive T cells by superantigens.2 However, these possible mechanisms are not yet proven. Nevertheless, a variety of infectious organisms including virus and bacteria have been linked to autoimmune diseases,158 including Epstein Barr virus (EBV) as a trigger for SLE. The prevalence of EBV infection as determined by measuring anti-EBV antibodies is higher in patients with SLE compared to healthy controls,159,160 and patients with SLE have increased EBV viral loads in peripheral blood and altered T-cell immune responses to EBV compared to healthy controls.161 These findings suggest that the immune control of EBV is altered in SLE although it is not clear whether such an alteration is the cause or the consequence of SLE. Of interest, a recent study showed that EBV-encoded small RNA (EBER) and EBV double-stranded DNA (dsDNA) could induce IFN-α production from human plasmacytoid DC via binding to TLR7 and 9, respectively, suggesting a potential pathologic role for increased viral loads of EBV in human lupus.162 An intriguing question is why only a small fraction of individuals develop autoimmunity even though infections that may be associated with autoimmunity such as EBV are very common in the population. One likely answer is diversity in the genetic background among individuals, which can lead to different immune responses to the same infectious organisms and possible development of autoimmune diseases.

The mechanisms involved in tissue damages in autoimmune diseases are similar to those that are used in eradicating invading foreign organisms. Autoantibodies can directly bind to autoantigens that are expressed on cells, which results in destruction of cells (type II antibody-mediated hypersensitivity reaction).2 Autoimmune hemolytic anemia and autoimmune thrombocytopenia, which can be seen in SLE, are good examples of cell destruction by this mechanism.9 Some autoantibodies can bind autoantigens and activate the complement cascade, which leads to the recruitment of inflammatory cells and subsequent tissue damage (type III immune complex-mediated hypersensitivity reaction).163 For instance, autoantibodies against chromatin and ribonucleoproteins appear to have a pathogenic role in lupus.164 The deposition of immune complexes of these autoantibodies with their respective autoantigens in target organs, such as in the kidney, leads to activation of complement and Fc receptor binding with subsequent tissue injury.165,166 Although autoreactive T cells are required for the development of autoimmune diseases and autoantibodies,167,168 it is still unclear whether autoreactive T cells are directly involved in damaging tissues in systemic autoimmune syndromes such as SLE, whereas their role in directly promoting tissue injury in organ-specific autoimmune syndromes, such as type 1 DM and multiple sclerosis, are much better established.