key: cord-0009346-4363h9gx
authors: Vanderlugt, Carol J; Miller, Stephen D
title: Epitope spreading
date: 2002-02-11
journal: Curr Opin Immunol
DOI: 10.1016/s0952-7915(96)80012-4
sha: 84fc9f476857840bb8d9edf56aead2dd908a6e93
doc_id: 9346
cord_uid: 4363h9gx

Epitope (determinant) spreading is the development of immune responses to endogenous epitopes secondary to the release of self antigens during a chronic autoimmune or inflammatory response. The past year has seen considerable advances in our understanding of the contribution of epitope spreading to the chronic pathogenesis of experimental T-cell-mediated and antibody-mediated autoimmune diseases. Most significantly, conclusive functional evidence for a major role for epitope spreading in the chronic pathogenesis of murine relapsing-remitting experimental autoimmune encephalomyelitis, a CD4(+) T-cell-mediated model of multiple sclerosis, was forthcoming.

It has been proposed that autoimmune diseases result directly from dysregulation of the immune system or as a secondary consequence of microbial infection [1, 2] . Because of the promiscuity of TCRs and B cell receptors (BCRs), cross-reactivity between infectious agents and self antigens (molecular mimicry) is not rare, although such activities are better documented at the antibody level than at the T cell level. Recent evidence from animal models of autoimmune disease indicates that disease progression may be due to the activation and recruitment of autoreactive lymphocytes, regardless of the initiating event. These autoreactive lymphocytes are specific for epitopes that are distinct from and non-cross-reactive with the disease-inducing epitope, and resuh from chronic tissue damage (epitope spreading). Support for this hypothesis comes primarily from autoimmune models in which disease is induced with a defined autoepitope and responses to non-cross-reactive T-and/or B-cell epitopes on the same or different self proteins are assessed for pathological potential during disease progression. Induction of chronic tissue pathology by viruses also leads to epitope spreading. Epitope spreading has been demonstrated at the T cell level, particularly in relapsing-remitting experimental autoimmune encephalomyelitis (R-EAE), in antibody-mediated disease models such as systemic lupus erythematosus (SLE), and in viral diseases including Theiler's murine encephalomyelitis virus-induced demyelineating disease.

Murine R-EAE currently constitutes the best characterized chronic CD4 ÷ T-cell-initiated experimental autoimmunc model and has provided much of the initial evidence in support of a major pathological role for epitope spreading in chronic disease due primarily to the precise knowledge of disease-related epitopes on multiple myelin proteins [3,4°] . Early evidence for epitope spreading in murine R-EAE models showed diversification in the number and MHC restriction of myelin epitopes that are recognized following recovery from acute clinical disease [5] [6] [7] [8] [9] . These reports provided intriguing evidence that the T cell repertoire is dynamic, and that recruitment of T cell reactivity to additional specificities occurs during the course of R-EAE.

Our laboratory has examined intramolecular and intermoleeular epitope spreading in the SJL/J mouse in R-EAE that has been induced with the highly immunodominant proteolipid protein (PLP)139-151 epitope or with the weakly encephalitogenic myelin basic protein (MBP)84-104 epitope [10°*]. T cell proliferative and delayed-type hypersensitivity (DTH) responses to the secondary (PLP178-191) epitope, but not to the immunodominant (MBP84-104) epitope, are induced in SJL]J mice following the acute phase of both active and adoptive PLP139-151-induced R-EAE (intramolecular epitope spreading). Central nervous system (CNS) myelin damage is necessary for the initiation of epitope spreading. When induced following disease initiation but prior to the acute clinical episode, tolerance to the disease-inducing PLP139-151 epitope prevented the development of T cell reactivity to the PLP178-191 epitope, and the magnitude of PLP178-191-specific DTH responses correlated with the severity of CNS damage during acute disease [10" ]. In addition, PLP139-151-specific T cell responses were activated in mice following acute tissue damage in MBP84-104-induced R-EAE (intermolecular epitope spreading). Intramolecular and intermolecular T cell epitope spreading has also been reported in SJL/J mice in which R-EAE was induced with a T cell line specific for an MBP exon-2 encoded peptide [11°] , and in B10.RIII mice primed with MBP89-101 [12°]. More recently, Yu et al. [13 °°] reported a predictable sequential epitope spreading cascade in (SWRxSJL)F 1 mice with R-EAE induced by the immunodominant PLP139-151 epitope. During disease progression, responses to PLP249-273 appeared first, followed by proliferation to MBP87-99, and then to PLP173-198.

The pathological contribution of epitope spreading to chronic R-EAE has been verified through the employment of several experimental approaches. Firstly, serial transfer studies have proven the encephalitogenic potential of responses to endogenous myelin epitopes. Splenocytes from mice in remission from PLP139-151induced active or adoptive R-EAE that are activated in vitro with the relapse-associated PLP178-191 epitope transfer R-EAE into naive recipients [10°°]. Similar success in the serial transfer of R-EAE by T cells activated with relapse-associated epitopes has been reported in SJL/J mice in disease that was initiated by T cell lines specific for intact MBP [8] or for an MBP exon-2 encoded peptide [11 °] and in (SWR×SJL)F 1 mice in disease that was initiated by PLP139-151 [13" ]. Secondly, PLP178-191-specific T cells are demonstrable in the CNS of mice that are in remission from PLP139-151-induced R-EAE (CJ Vanderlugt, SD Miller, unpublished data). Most importantly, induction of tolerance to the intact PLP [10 °°] or to the relapse-associated PLP178-191 epitope, but not to the disease-inducing PLP139-151 epitope (Table 1) , protects mice that are in remission from acute PLP139-151-induced R-EAE from renewed disease progression. Similar results have been reported by Yu eta/. [13 °°] in (SWR x SJL)FI mice with PLP139-151-induced R-EAE in which disease progression is blocked by tolerance to MBP87-99. These reports verify our earlier report concerning adoptive MBP-induced R-EAE [14] , wherein during disease remission induction of tolerance to mouse spinal cord homogenate (a crude mixture of Table 1 myelin neuroantigens), but not to the disease-inducing MBP, ameliorated disease relapses. Lastly, we have shown that during remission [15"°] , blockade of the CD28/B7 costimulatory pathway by treatment of SJL/J mice with anti-B7-1 Fab fragments blocked the development of PLP178-191-specific DTH responses and disease progression in PLP139-151-induced R-EAE. Collectively, these studies provide conclusive support for a dominant role of responses to endogenous myelin epitopes in mediating disease relapses.

Samson and Smilek [16] showed, in MBP-induced R-EAE in (PLJxSJL)F 1 mice, that tolerance induced during remission with MBP Acl-ll[Y] protected the mice from relapse. This is in contrast to the preceding data and to the reports by others of epitope spreading in this model [5] [6] [7] . MBP Acl-ll is the dominant encephalitogenic epitope in this system, and MBP Acl-II[Y] is an altered peptide which tolerizes more efficiently than the native epitope. It is possible that new epitopes are not primarily responsible for the first relapse in this model or that the tolerance induced by the altered peptide works primarily via an antigen nonspecific bystander mechanism.

The nonobese diabetic (NOD) mouse serves as a spontaneous murine model of insulin-dependent diabetes mellitus, in which insulitis leads to the destruction of pancreatic 13 cells and consequently to clinical diabetes. T cell and B cell responses to 13 cell antigens arise spontaneously in NOD mice. It is generally believed that islet cell destruction is initiated by a Thl response against the 65 kDa isoform of glutamic acid decarboxylase (GAD)65, which arises within four weeks of age, and, as the animals progress to overt diabetes, that T cell reactivity subsequently spreads to other pancreatic 13 cell antigens such as carboxypeptidase H, insulin, and heat shock protein 65 [17, 18] . This hypothesis is supported by reports that induction of neonatal tolerance to GAD65 via intrathymic [18] or intravenous [17] peptide injections blocked the activation of responses to other ~ cell epitopes and the subsequent development of insulitis and diabetes. More recently, Kaufman and co-workers [19 °°] have shown that intranasal administration of GAD65 induced a Th2 response characterized by an interleukin-5 dominant T cell response and the production of large amounts of GAD65-specific IgG1. More importantly, CD4 + splenic T cells from GAD65-treated mice protected NOD-severe combined immunodeficient mice from diabetes induced by the transfer of NOD T cells (40% late onset compared to 90% incidence in the controls) and also blocked the induction of proliferative responses to heat shock protein and insulin.

SLE is a systemic rheumatic disease characterized by anti-DNA and antispliceosomal complex antibodies (anti-SM and anti-nribonucleoprotein). James et al. [20" ] used SmB peptide epitopes, immunodominant in human SLE, to induce an SLE-like syndrome including the development of antinuclear antibodies anti-DNA antibodies, thrombocytopenia, seizures and proteinuria in mice. These animals also developed autoantibodies that bound to other non-cross-reactive spliceosomal proteins. The authors hypothesize a role for B cell epitope spreading in which B cells directed against disease-initiating epitopes mediate antigen-specific uptake of large complexes leading to the efficient presentation of self epitopes from distinct proteins that make up the complexes. Autoantibodies to both components of the La/Ro ribonucleoprotein complex are seen in SLE and are a prominent feature of Sj6grens syndrome. Intramolecular epitope spreading was demonstrated by the induction of at, toantibodies to multiple nonoverlapping regions of La after immunization with the La A subfragment and intermolecular spreading was demonstrated by the appearance of anti-60kDa Ro IgG antibodies in mice immunized with mouse or human La [21°°]. Thus, the development of autoantibodies to multiple components of the La/Ro ribonucleoprotein complex follows the initiation of an immune response to a single component and epitope spreading is likely to account for the appearance of mixed autoantibody patterns in systemic autoimmune diseases.

Transient production of autoantibodies commonly occurs in acute and chronic viral infections in animals and man. Most of these responses are probably not pathological and may even be immunoregulatory. Self-specific T cell responses during viral infection are less well documented. In genetically susceptible individuals, infection-induced autoimmunity has been postulated to ensue from the activation of autoreactive T cells secondary to an encounter with a pathogen by molecular mimicry [22] , by the release of sequestered myelin antigens secondary to virus-specific T-cell-initiated myelin damage, that is, epitope spreading [3,4°,10°'], and/or by the nonspecific stimulation of autoreactive T cells by microbial-encoded superantigens [23] .

Epitope spreading has been reported in animal models of virus-induced demyelination. Splenic T cells from Lewis rats infected with measles virus or routine coronavirus, strain JHM, have been reported to proliferate in response to MBP and to transfer R-EAE to naive syngeneic rats when activated in vitro with MBP. No cross-reactivity between MBP and measles virus was demonstrable, however, [24-261. Measles virus infection also enhanced the susceptibility of Lewis rats to a normally nonencephalitogenic MBP. Proliferation and Thl-type cytokine production in response to MBP has also been reported in SJL/J and B6 mice infected with Semliki Forest virus [27] and in SJL/J mice infected with Sindbis virus [28] .

Infection of SJL/J mice with Theiler's murine encephalomyelitis virus induces a chronic progressive CD4 + T-cell-mediated demyelinating disease initiated by virusspecific Thl cells targeting virus that is persisting in the CNS [29] . T cell responses to virus epitopes arise within 7-10 days postinfection, preceding the development of clinical disease which begins 25-35 days postinfection. T cell proliferative and DTH responses to the immunodominant PLP139-151 epitopc are demonstrable 3-4 weeks after disease onset [4" ] and DTH responses to additional encephalitogenic myelin peptides (PLP178-191, PLP56-70 and MOG92-106) are apparent 157-207 days postinfection (SD Miller et al., unpublished data). Significantly, no cross-reactivity between these autoepitopes and Theiler's murine encephalomyelitis virus epitopes has been found.

Patients with infectious mononucleosis develop IgM autoantibodies to hematopoietic cell antigens p542 and p554. Anti-p542 antibodies cross-react with a shared region of Epstein-Barr virus nuclear antigen-1. Anti-P554 antibodies do not cross-react with Epstein-Barr virus peptides, however, and appear to arise via epitope spreading. Significantly, patients with progressive systemic sclerosis, SLE, ulcerative colitis, Sj6grens syndrome, rheumatoid arthritis, ankylosing spondylitis, and Crohn's disease develop high titers of IgG anti-p542 specific for non-cross-reactive detcrminants [30, 31] .

Scrcarz and co-workers [32, 33] have proposed a comprehensive model in which dominant epitopes are defined as those epitopes to which an animal initially responds when primed to a protein or an infectious agent. Responses to other epitopes on that molecule, which arise later or upon hyperimmunization, are termed secondary or cryptic. According to this hypothesis, an immune response targeting one or two dominant epitopes on a infectious agent is not sufficient. The immune system has evolved a mechanism, therefore, for increasing the number of epitopes targeted during an infection. This results in the appearance of a response against cryptic epitopes, but can also lead to autoimmunity via the activation of autoreactive T and/or B cells which would not have undergone central or peripheral tolerance. This hypothesis has concentrated mainly on the diversification of the immune response after priming with intact proteins, for example, MBP in R-EAE, wherein responses to the cryptic epitopes arise only following initial tissue damage.

Our model for epitope spreading in chronic CD4+ T-cell-mediated autoimmune disease is shown in Figure 1 [3] . In brief, T cells specific for the initiating self or viral epitope induce the inflammatory cascade in the target organ, resulting in tissue damage. Tissue debris is taken up by macrophages and/or antigen-specific B cells and presented to naive tissue-specific T cells which, once activated, perpetuate the inflammatory response. This hypothesis is consistent with that of Elson et al. [34] , who propose that epitope spreading is a Thl-linked phenomenon linked to altered processing of self epitopes and with the hypothesis of Mamula and Janeway [35] , who suggest that antigen-specific B cells (I0 000-fold more efficient than other APCs) may be responsible for the diversification of autoimmunc responses. Model of epitope spreading in chronic T-cell-mediated autoimmune disease. (a) Upon priming with the initiating self (1") epitope in an appropriate adjuvant, Thl cells are activated in the peripheral lymphoid tissue. These Thl cells proliferate, upregulate appropriate homing receptors, traffic via the bloodstream to the blood-tissue barrier, and transmigrate (adhere and penetrate) into the target organ (b). The autoreactive Thl cells encounter antigen presented in the context of MHC class il by tissue-specific APCs triggering local secretion of chemokines and cytokines (¢). These soluble mediators attract and activate peripheral monocytes (Mono) and macrophages (Me) and tissue-resident mononuclear inflammatory cells. The phagocytic activity of these activated macrophages along with the secretion of cytotoxic and proinflammatory molecules like TNF-o~, nitric oxide (NO), and 0 2 radicals cause self-tissue destruction (d) resulting in the expression of clinical symptoms. Tissue debris phagocytized by macrophages and/or B cells could be presented to naive T cells, which can traverse the compromised blood-tissue barrier, in the target organ and/or in the peripheral lymphoid tissue (e), leading to the activation and expansion of pathological T and/or B cells specific for self epitopes (2") other than the initial disease-inducing (1") epitope which mediate a second wave of tissue destruction. LT, leukotriene; MIP, macrophage inflammatry protein; TNF, tumor necrosis factor.

Evidence that autoimmune responses are dynamic, with specificities evolving over time, has been demonstrated in a variety of experimentally induced, virally induced, and spontaneous animal models of autoimmunity at both the T cell and B cell levels. Most significantly, evidence supporting a major functional role for epitope spreading in chronic autoimmune disease has emerged from the study of murine R-EAE models wherein disease progression can be ameliorated using either antigen-specific tolerance to the relapse-associated epitopes or blockade of the CD28/B7 costimulatory pathway. Future areas of interest include: elucidation of the factors which govern epitope dominance and the sequential pattern of epitope spreading; determination of the site of priming of T cells specific for endogenous self epitopes and study of the role of B cells and tissue-resident APCs and costimulatory requirements involved in this process; and determination of the relative efficiencies of and mechanisms by which various antigen-specific and nonspecific immunoregulatory strategies control the epitope spreading process.

Papers of particular interest, published within the annual period of the review, have been highlighted as:

• of special interest • • of outstanding interest

Molecular mimicry and autoimmune disease

Viruses, ¢ytokines, antigens, and autoimmunity

The immunopathogenesis and regulation of T-cell mediated demyelinating diseases

Evolution of the T cell repertoire during the course of experimental autoimmune encephalomyelitis

The authors review evidence for epitope spreading to endogenous myelin epitopes in peptide-induced R-EAE models in the SJL/J mouse and following infection with Theilers' murine encephalomyelitis virus

Kinetics and specificity of T and B cell responses in relapsing experimental allergic encephalomyelitis

Alterations in T cell antigen specificity and class II restriction during the course of chronic relapsing experimental allergic encephalomyelitis

T cell sensitization to proteolipid protein in myelin basic protein-induced experimental allergic encephalomyelitis

Development of reactivity to new myelin antigens during chronic relapsing autoimmune demyelination

Spreading of T-cell autoimmunity to cryptic determinants of an autoantigen

Functional • • evidence for epitope spreading in the relapsing pathology of EAE in the SJL/J mouse

The authors show the involvement of intermolecular epitope spreading to PLP139-151 in MBP84-104-induced R-EAE and intramolecular epitope spreading to PLP178-191 in PLP139-151-induced R-EAE. The extent of activation of relapse-associated epitopes was shown to correlate with the degree of tissue damage during acute disease. Splenic T cells specific for the relapse-associated epitopes could transfer disease to naive mice and tolerance to the intact PLP protein during disease remission

The immune response to a subdominant • epitope in myelin basic protein exon-2 results in immunity to intra-and intermolecular dominant epitopes

The authors show that R-EAE induced by transfer of a T cell line specific for the product exon-2 of the MBP gene leads to R-EAE in the SJL/J mouse and the subsequent development of splenic proliferative responses to MBPAcl-11, MBP43-88, MBP89-101, and PLP139-151

• Spreading of the immune response to different myelin basic protein peptides in chronic experimental autoimmune encephalomyelitis in B10.RIII mice

The authors show that induction of R-EAE in B10.RIll mice immunized with MBP89-101 eventually results in the development of interferon-producing T cells to peptides outside of this region

A predictable sequential • . determinant spreading cascade invariably accompanies progression of experimental autoimmune encephalomyelitis: a basis for peptide-specific therapy after onset of clinical disease

The authors determine the sequential pattern of epitope spreading in PLP139-151 -induced R-EAE in the (SWRxSJL)F1 mouse and show that tolerance to the relapse-associated MBP87-99 epitope blocks disease progression

Successful treatment of paralytic relapses in adoptive experimental autoimmune encephalomyelitis via neuroantigen-specific tolerance

Blockade of CD28/B7-1 interaction prevents epitope spreading and clinical relapses of murine EAE Immunity

The authors show that treatment of SJL/J mice during remission from active PLP139-151-induced R-EAE with anti-BT-1 Fab fragments blocks the activation of T cell specific for the relapse-associated PLP178-191 epitope and inhibits expression of clinical relapses

Spontaneous loss of T cell tolerance to glutamic acid decarboxylase in murine insulin-dependent diabetes

Immune response to glutamic acid decarboxylase correlates with insulitis in non-obese diabetic mice

The authors show that intranasal tolerance to the proposed disease inducing epitope on GAD65 results in activation of Th2 cells which reduce the incidence of diabetes in NOD mice and blocks the activation of autoreactive

Immunoglobulin • -epitope spreading and autoimmune disease after peptide immunization: Sm B/B'-derived PPPGMRPP and PPPGIRGP induce splice•some autoimmunity

The authors show in an experimental model of SLE that rabbits primed to Sm BIB" peptides develop an SLE-like syndrome along with autoantibody responses to spliceosomal proteins other than Sm BIB

Intra-and intermolecular • • spreading of autoimmunity involving the nuclear self-antigens La (SS-B) and R• (SS-A)

The authors show that, upon immunization of mice with a fragment of the La protein, autoantibody responses to other non-cross-reactive components of the La/Ro ribonucleoprotein complex are generated

Amino acid homology between the encephalitogenic site of myelin basic protein and virus: mechanism for autoimmunity

Superantigens: bacterial and viral proteins that manipulate the immune system

Ter Meulen V: Induction of autoimmune reactions to myelin basic protein in measles virus encephalitis in Lewis rats

Ter Meulen V: Characterization of measles virus-induced cellular autoimmune reactions against myelin basic protein in Lewis rats

Autoimmune reactions against myelin basic protein induced by corona and measles viruses

Immune responses, and autoimmune outcome, during virus infection of the central nervous system

Role of the immune response in Sindbis virus-induced paralysis of SJL/J mice

Immunologic aspects of Theiler's murine encephalomyelitis virus (TMEV)-induced demyelinating disease

Epstein-Barr virus-induced autoimmune responses. I. |mmunoglobulin M autoantibodies to proteins mimicking and not mimicking Epstein-Barr virus nuclear antigen-I

Epstein-Barr virus-induced autoimmune responses. II. Immunoglobulin G autoantibodies to mimicking and nonmimicking epitopes. Presence in autoimmune disease

Dominance and crypticity of T cell antigenic determinants

Determinant spreading and the dynamics of the autoimmune T-cell repertoire

Immunologically ignorant autoreactive T cells, epitope spreading and repertoire limitation

Do B cells drive the diversification of immune responses