Positive selection in autoimmunity: abnormal immune responses to a bacterial dnaJ antigenic determinant in patients with early rheumatoid arthritis

Heterogeneity of rheumatoid arthritis: from phenotypes to genotypes

Rheumatoid arthritis (RA) is now recognized as a multigene disorder with a number of genetic polymorphisms contributing to disease pathogenesis. Here, we propose that the diagnostic category of RA includes multiple subtypes of disease and that the different phenotypes of RA correlate to different genotypes. Support for this concept has come from a reappraisal of the clinical heterogeneity of RA and the observation that HLA-DRB1 polymorphisms are useful in describing genetic heterogeneity of RA phenotypes. A series of HLA-DRB1 genes has been identified as RA associated, and in recent years emphasis has been put on the sequence similarities of these alleles. An alternative view focuses on the amino acid variations found in RA-associated HLA-DRB1 alleles with different alleles being enriched in distinct subtypes of RA. Rheumatoid factor-positive destructive joint disease is predominantly associated with the HLA-DRB1*0401 allele, while HLA-DRB1*0404 and B1*0101 predispose for milder and often seronegative disease. Expression of disease-associated alleles on both haplotypes carries a high risk for extra-articular manifestations. In particular, patients homozygous for HLA-DRB1*0401 frequently develop rheumatoid vasculities on follow-up. Besides HLA gene polymorphisms, abnormalities in the generation and function of CD4 T cells and in inflammatory pathways established in synovial lesions can be used to dissect patient subsets with different variants of RA. Emergence of CD28-deficient CD4 T cells identifies RA patients with extra-articular manifestations. These cells undergo clonal expansion in vivo, produce high amounts of IFN-gamma, and exhibit autoreactivity. Concordance of monozygotic twins for the expression of CD4+ CD28- T cells suggests a role for genetic factors in the generation of these unusual T cells. Evidence for heterogeneity of the synovial component of RA comes from studies describing three distinct patterns of lymphoid organization in the synovium. Based upon the topography of tissue-infiltrating mononuclear cells, diffuse, follicular, and granulomatous variants of rheumatoid synovitis can be distinguished. Each pattern of lymphoid organization correlates with a unique profile of tissue cytokines, demonstrating that several pathways of immune deviation modulate disease expression in RA. A dissection of RA variants would have major implications on how the disease is studied, treated, and managed. Identifying combinations of RA risk genes that correlate with disease variants could, therefore, become an important diagnostic tool.

Perturbation of the T cell repertoire in rheumatoid arthritis

Aberrations in the T cell repertoire with the emergence of oligoclonal populations have been described in patients with rheumatoid arthritis (RA). However, the extent of the repertoire perturbations as well as the underlying mechanisms are not known. We now have examined the diversity of the peripheral CD4 T cell repertoire by determining the frequencies of arbitrarily selected T cell receptor (TCR) beta-chain sequences. Healthy individuals displayed a highly diverse repertoire, with a median frequency of individual TCR beta-chain sequences of 1 in 2.4 x 10(7) CD4 T cells. In RA patients, the median TCR beta-chain frequency was increased 10-fold, indicating marked contraction of the repertoire (P < 0.001). The loss in TCR diversity was not limited to CD4 memory T cells but also involved the compartment of naive T cells, suggesting that it reflected an abnormality in T cell repertoire formation and not a consequence of antigen recognition in the synovium. Also, control patients with chronic inflammatory disease such as hepatitis C expressed a diverse repertoire indistinguishable from that of normals. Telomere length studies indicated an increased replicative history of peripheral CD4 T cells in RA patients, suggesting an enhanced turnover within the CD4 compartment. Compared with age-matched controls, terminal restriction fragment sizes were 1.7 kilobases shorter (P < 0.001). These data demonstrate an altered CD4 T cell homeostasis in RA that may contribute to the autoimmune response as well as to the immunodeficiency in these patients.

Mucosal modulation of immune responses to heat shock proteins in autoimmune arthritis

Induction of oral tolerance to antigens that are targets of self-reactive immune responses is an attractive approach to antigen-specific immune therapy of autoimmune diseases. Oral tolerization has indeed proven to be safe and effective in amelioration of autoimmune diseases in animal models. In humans, results have been somewhat controversial. The emphasis given to clinical outcome rather than to immunomodulation, and the difficulty in identifying appropriate candidate antigens contribute to the controversy. Heat shock proteins are promising targets for immune intervention. Immune reactivity to heat shock proteins has indeed been correlated with autoimmune arthritis in animal models, and abnormal immune responses to heat shock proteins have been described in human arthritis as well. Despite significant recent progress, little is known at a molecular level regarding the mechanisms which are responsible for a switch from autoimmunity to tolerance in humans. This is particularly true with respect to sequential analysis of several molecular and immunologic markers during both the course and treatment of disease. Novel approaches are currently under way to fill the gaps. We will briefly detail here the experience gained to date, and identify some of the avenues which future research will explore.

Heat shock proteins and arthritis--new readers start here

The possible roles of heat shock proteins in the pathogenesis of inflammatory arthritis have been discussed for a number of years, and investigated intensively in both animal models and human disease. This review surveys evidence which has pointed, on the one hand, to hsp as targets of a pathogenic immune response, and on the other, to an immunoregulatory role for T cell recognition of self hsp. The extent to which findings in experimental animals have led to further insights applicable to human disease is also emphasised.

Isolation and characterization of the dnaKJ operon from Actinobacillus actinomycetemcomitans

The dnaKJ operon of Actinobacillus actinomycetemcomitans Y4 was cloned by the DNA-probing method using synthetic oligonucleotides designed on the basis of two conserved regions in DnaK/hsp70 proteins and sequenced by inverse PCR. The sequenced region was shown to contain two open reading frames coding for proteins analogous to DnaK and DnaJ.

Genetic bias in immune responses to a cassette shared by different microorganisms in patients with rheumatoid arthritis

Rheumatoid arthritis (RA) is an autoimmune disease associated with HLA-DRbeta1 alleles which contain the QKRAA amino acid sequence in their third hypervariable region(s). The QKRAA sequence is also expressed by several human pathogens. We have shown previously that an Escherichia coli peptide encompassing QKRAA is a target of immune responses in RA patients. Here we address two questions: first, whether QKRAA may function as an "immunological cassette" with similar, RA-associated, immunogenic properties when expressed by other common human pathogens; and second, what is the influence of genetic background in the generation of these responses. We find that early RA patients have enhanced humoral and cellular immune responses to Epstein-Barr virus and Brucella ovis and Lactobacillus lactis antigens which contain the QKRAA sequence. These results suggest that the QKRAA sequence is an antigenic epitope on several different microbial proteins, and that RA patients recognize the immunological cassette on different backgrounds. ANOVA of immune responses to "shared epitope" antigens in monozygotic twin couples shows that, despite significantly elevated responses in affected individuals, a similarity between pairs is retained, thus suggesting a role played either by hereditary or shared environmental factors in the genesis or maintenance of these responses.

A function for the QKRAA amino acid motif: mediating binding of DnaJ to DnaK. Implications for the association of rheumatoid arthritis with HLA-DR4

The amino acid motif QKRAA, when expressed on HLA-DRB1, carries susceptibility to develop rheumatoid arthritis. This motif is the basis of strong B and T cell epitopes. Furthermore, it is highly overrepresented in protein databases, suggesting that it carries a function of its own. To identify this function, we used QKRAA peptide affinity columns to screen total protein extracts from Escherichia coli. We found that DnaK, the E. coli 70-kD heat shock protein, binds QKRAA. Of interest, DnaK has a natural ligand, DnaJ, that contains a QKRAA motif. We found that QKRAA-containing peptides inhibit the binding of DnaK to DnaJ. Furthermore, rabbit antibody to the QKRAA motif can inhibit binding of DnaJ to DnaK. These data suggest that QKRAA mediates the binding of E. coli chaperone DnaJ to its partner chaperone DnaK.

Absence of peripheral blood T cell responses to "shared epitope' containing peptides in recent onset rheumatoid arthritis

OBJECTIVES

To determine if peptides containing the 'shared epitope' sequence, QKRAA, from either endogenous, HLA-DR beta 1 (0401), or exogenous, Escherichia coli dnaJ, sources activate T cells in recent onset rheumatoid arthritis (RA).

METHODS

Peripheral blood mononuclear cell (PBMC) proliferative and whole blood cytokine responses to shared epitope containing peptides from DR beta 1 (0401) and E coli dnaJ, to control peptides from DR beta 1 (0402) and hsp40 and to the recall antigen, tetanus toxoid, were tested in 20 untreated, recent onset RA subjects, 20 HLA, age, and sex matched healthy controls and 18 other subjects with inflammatory arthritis. PBMC proliferative responses to a second E coli dnaJ peptide (with the shared epitope at the N-terminus) and two peptides from type II collagen with high affinity for DR4(0401) were tested in a further 16 recent onset RA and 17 control subjects.

RESULTS

PBMC proliferation and whole blood interferon gamma or interleukin 10 production in response to the shared epitope containing and control peptides were not different between the disease and control groups. On the other hand, compared with controls, RA subjects had significantly higher proliferation to a collagen II (aa 1307-1319) peptide, but significantly lower proliferation and interferon gamma production to tetanus toxoid.

CONCLUSION

Recent onset RA subjects had no demonstrable increase in peripheral blood T cell reactivity to shared epitope containing peptides. However, a proportion had increased T cell reactivity to a peptide of similar length from a candidate RA autoantigen, collagen type II. Their impaired responses to tetanus are in keeping with evidence for general T cell hyporesponsiveness in RA.

Molecular mimicry: can epitope mimicry induce autoimmune disease?

Mimicry of host antigens by infectious agents may induce cross-reactive autoimmune responses to epitopes within host proteins which, in susceptible individuals, may tip the balance of immunological response versus tolerance toward response and subsequently lead to autoimmune disease. Epitope mimicry may indeed be involved in the pathogenesis of several diseases such as post-viral myocarditis or Chagas disease, but for many other diseases in which it has been implicated, such as insulin-dependent diabetes mellitis or rheumatoid arthritis, convincing evidence is still lacking. Even if an epitope mimic can support a cross-reactive T or B cell response in vitro, its ability to induce an autoimmune disease in vivo will depend upon the appropriate presentation of the mimicked host antigen in the target tissue and, in the case of T cell mimics, the ability of the mimicking epitope to induce a proliferative rather than anergizing response upon engagement of the MHC-peptide complex with the T cell receptor. B cell presentation of mimicking foreign antigen to T cells is a possible mechanism for instigating an autoimmune response to self antigens that in turn can lead to autoimmune disease under particular conditions of antigen presentation, secondary signalling and effector cell repertoire. In this review evidence in support of epitope mimicry is examined in the light of the necessary immunological considerations of the theory.

Pathogenesis of rheumatoid arthritis

Chronicity and destructive potential are characteristic features of the inflammatory response in the synovial membrane typical for RA. The dominant paradigm has proposed that an exogenous antigen, likely an infectious organism, targets the synovia and elicits a chronic immune response. Support for this disease model has come from describing the cellular components of the inflammatory lesions, which are composed of macrophages, T cells, and B cells. The observation that HLA molecules function by specifically binding antigenic peptides and presenting them to T cells has boosted the concept of an antigen-driven response. The last decade in RA research has been dominated by a shift from premolecular to molecular techniques. A major effort has been made to determine which cytokines and inflammatory mediators are produced at the site of disease. Tissue residing and infiltrating cells secrete proinflammatory cytokines in situ, which likely have a critical role in amplifying and maintaining the inflammation. We are beginning to understand that migration of inflammatory cells into the tissue is an important component of disease, specifically because adhesion molecules not only facilitate tissue infiltration, but also affect cell activation and cell-cell and cell-matrix interactions. The paradigm that RA is an antigen-driven and thus T cell-mediated disease has brought attempts to use T cell-depleting reagents as therapeutics. Although T cells could be eliminated in the peripheral blood, overall therapeutic benefits have been minimal and accompanied by major side effects. The lack of therapeutic efficacy has been demonstrated to be combined with the persistence and the selective proliferation of T cells in the joint, reemphasizing the role of tissue-infiltrating T cells in the disease. Studies of the composition of the T cell infiltrate have demonstrated heterogeneity, indicating that disease-relevant T cells are probably low in frequency. A new perspective on the role of T cells in RA has been opened by studies establishing that RA patients select a unique repertoire of T cells in the thymus and that clonal expansion of CD4 T cells is a frequent event in RA patients. Pathology of T cell function might be much more systemic than suspected so far. RFs remain the major autoantibodies in RA patients. In the last 10 years, it has become clear that they are not exclusively built under pathologic conditions but that RF-expressing B cells are an important element of normal immune responses. All immunoglobulin isotypes are represented among RF molecules. Some of them have accumulated somatic mutations, suggesting the influence of antigen recognition and T cell help. T cell control of RF production may explain the observation that RF positivity is an HLA-dependent phenomenon. Major progress in understanding pathologic events leading to RA can be expected by abandoning single hit models, which are too simplified and underestimate the complexity of the disease. In particular, taking into account that nonimmune tissues and mechanisms might be equally important in pathogenesis will open new avenues of conceptual approaches. Cross-fertilization will likely come from genetic studies aimed at detecting underlying genetic risk factors in common genetic disease. Emerging data indicate that several genetic risk determinants, each of which is nonpathologic if occurring alone, can add up to confer disease risk. One of these genetic elements in RA has been mapped to the HLA region. A sequence polymorphism in the HLA-DR B1 gene appears to be a strong genetic risk factor in several ethnic groups. Correlation of clinical presentation of RA and the inheritance of the RA risk gene suggests that the gene product is not necessary in disease initiation but functions by modulating disease pattern and severity. The next decade in RA research will be dedicated toward unraveling how genetic determinants can introduce pathology (e.g., how HLA genes can function as progre

Virus-induced autoimmune disease

The breaking of tolerance or unresponsiveness to self-antigens, involving the activation of autoreactive lymphocytes, is a critical event leading to autoimmune diseases. The precise mechanisms by which this can occur are mostly unknown. Viruses have been implicated in this process, among other etiological factors, such as genetic predisposition and cytokine activity. Several ways have been proposed by which a viral infection might break tolerance to self and trigger an autoreactive cascade that ultimately leads to the destruction of a specific cell type or an entire organ. The process termed "molecular mimicry' and the use of transgenic models in which viral and host genes can be manipulated to analyze their effects in causing autoimmunity have been particular focuses for research. For example, there is a transgenic murine model of virus-induced autoimmune disease, in which a known viral gene is selectively expressed as a self-antigen in beta cells of the pancreas. In these mice, insulin-dependent diabetes develops after either a viral infection, the release of a cytokine such as IFN-gamma, or the expression of the costimulatory molecule B7.1 in the islets of Langerhans. Recent studies using this model have contributed to the understanding of the pathogenesis of virus-induced autoimmune disease and have furthered the design and testing of novel immunotherapeutic approaches.

Genetic control of T cell receptor BJ gene expression in peripheral lymphocytes of normal and rheumatoid arthritis monozygotic twins

The amino acids encoded at the junctions of T cell receptor (TCR) V and J genes directly interact with MHC bound peptides. However, the regulation of the human TCRBJ gene repertoire has been difficult to analyze, because of the potentially complex number of BJ gene rearrangements. To overcome this problem, we developed a PCR-ELISA method to study BJ gene expression, and compared peripheral T lymphocytes from 12 pairs of monozygotic twins, including 6 rheumatoid arthritis (RA) discordant pairs, and 5 normals. Analyses of the TCRBV5, 13 and 17 gene families, which have been reported to be increased in RA patients, showed: (a) the three TCRBV transcripts have common features of BJ gene usage; (b) TCR transcripts from each TCRBV family display a distinctive BJ gene profile, which is displayed better by CD4+ than CD8+ lymphocytes; (c) the BJ gene repertoires of monozygotic twins are more similar than those of unrelated individuals; and (d) the inflammation of RA does not induce specific changes in the genetically determined pattern of BJ expression. These results indicate that the frequency of expression particular TCRBV-TCRBJ recombinants in human lymphocytes is controlled genetically, and is maintained despite the presence of a chronic inflammatory disease.

Using transgenic mouse models to dissect the pathogenesis of virus-induced autoimmune disorders of the islets of Langerhans and the central nervous system

Viruses have often been associated with autoimmune diseases. One mechanism by which self-destruction can be triggered is molecular mimicry. Many examples of cross-reactive immune responses between pathogens and self-antigens have been described. This review presents two transgenic models of autoimmune disease induced by a virus through activation of anti-self lymphocytes. Viral antigens are expressed as transgenes either in beta-cells of the pancreas or in the oligodendrocytes of the CNS. Infection by a virus encoding the same gene activated autoreactive T cells that cleared the viral infection, and as a consequence of transgene expression resulted in organ-specific autoimmune disease. In both transgenic mouse models, autoreactive lymphocytes that escaped thymic negative selection were present in the periphery. Several factors are described that play a role in the regulation of the self-reactive process precipitated by a viral infection. These include the quantity of activated autoreactive T cells, the affinity of these T cells, the number of memory T cells generated following primary infection, costimulation by accessory molecules, and the types and locations of cytokines produced. In addition, unique barriers exist in target tissues that prevent or suppress autoreactive responses and define to a large extent the outcome of disease. Restimulation of autoreactive memory lymphocytes may be required to bypass these barriers and enhance autoimmune disease. Therapy directed at modifying these factors can reduce and even prevent autoimmune disease after it has been initiated.

Association of tumor necrosis factor microsatellite polymorphisms with HLA-DRB1*04-bearing haplotypes in rheumatoid arthritis patients

OBJECTIVE

To investigate 1) tumor necrosis factor (TNF) microsatellite allele frequencies in rheumatoid arthritis (RA), and 2) associations between TNF microsatellites and RA-associated HLA specificities in order to build up extended HLA haplotypes.

METHODS

Eighty-five caucasoid patients with RA and 109 healthy caucasoid controls were typed for TNF microsatellites a-d using fluorescent-labeled primers and semiautomated genotyping. A further 56 RA patients who were selected for having certain HLA-DRB1 types were also typed for these TNF microsatellites. Linkage disequilibria between TNF and HLA alleles were calculated, and extended haplotypes were established.

RESULTS

The TNFa6 allele frequency was significantly increased in the RA patients compared with the controls (P = 0.0019, odds ratio [OR] 2.5, 95% confidence interval [95% CI] 1.3-4.6), an increase that was further evident in patients who were HLA-DRB1*0401 homozygous (P = 0.0003, OR 7.3, 95% CI 2.2-24.4). This increase was found to be due to association with HLA-DRB1*0401. No TNF microsatellite allele was found to be associated with HLA-DRB1*0404. Three HLA extended haplotypes were identified in the RA group: 1) HLA-DRB1*0401;TNFd4;TNFa6;TNFb5;HLA-B44; HLA-Cw5;HLA-A2, 2) HLA-DRB1*0301;TNFd2; TNFa2;TNFb3;HLA-B8;HLA-Cw7;HLA-A1, and 3) TNFd5;TNFc2;TNFa2;TNFb1;HLA-B62;HLA-Cw3.

CONCLUSION

TNF microsatellites found to be associated with RA do not appear to be independent of class II HLA associations.

Role of enteric bacteria in the pathogenesis of rheumatoid arthritis: evidence for antibodies to enterobacterial common antigens in rheumatoid sera and synovial fluids

OBJECTIVE

To study antibodies to Escherichia coli O:14, which expresses large amounts of enterobacterial common antigen (ECA), and their corresponding antigen molecules in serum and synovial fluid samples from patients with rheumatoid arthritis (RA).

METHODS

Enzyme linked immunosorbent assay (ELISA) was used to measure antibodies to heat killed E coli O:14 in serum and synovial fluid samples from patients with RA and control subjects including healthy donors and patients with osteoarthritis. ELISA was also used to perform absorption analyses of antibodies to E coli O:14 with several enteric bacteria and their lipopolysaccharide (LPS). In addition, antigenic molecules reacting with E coli O:14 antibodies from patients with RA were examined using immunoblot analysis and N-terminal amino acid analysis.

RESULTS

Compared with control subjects, patients with RA showed significantly increased titres of antibodies against heat killed E coli O:14 in 33 of 83 serum samples (39.8%) and 38 of 58 joint fluid samples (65.5%). Absorption analyses with enteric bacteria and their LPS resulted in the reduction of antibody titres to heat killed E coli O:14 in serum and synovial fluid samples from the RA patients. In addition, immunoblot analysis of the samples from RA patients revealed not only a ladder-like banding pattern equivalent to ECA associated with LPS, but also two clear bands of bacterial outer membrane proteins of 35 kDa (Omp A) and 38 kDa (Omp C), having amino acid sequence homology with those of other Enterobacteriaceae.

CONCLUSION

These results suggest that some patients with RA are sensitised to antigens common to Enterobacteriaceae, and this may prove relevant to the future development of immunotherapy for RA. Furthermore, this sensitisation to antigens found commonly in Enterobacteriaceae may have a key role in the pathogenesis of human RA similar to that described previously in our animal model.

HLA-DR4 and HLA-DR10 motifs that carry susceptibility to rheumatoid arthritis bind 70-kD heat shock proteins

Most patients with rheumatoid arthritis express particular HLA-DR alleles. The DRbeta1 chains of these alleles share a highly homologous amino acid motif, in their third hypervariable (HV3) region, and this motif seems to help the development of rheumatoid arthritis via unknown mechanisms. In an attempt to identify a ligand of this motif, we screened bacterial proteins. HV3 peptides from HLA-DRB1 alleles containing a QKRAA or RRRAA motif bound the 70-kD heat shock protein (HSP) from Escherichia coli, dnaK. In lymphoblastoid cells homozygous for these same HLA-DRB1 alleles the constitutive 70-kD HSP, HSP73, that targets selected proteins to lysosomes coprecipitated with HLA-DR. Thus the QKRAA and RRRAA amino acid motifs of HLA-DR mediate binding of HLA-DR to HSP73. This property may influence the intracellular route, processing or peptide associations of the HLA-DRbeta1 chain in these two rheumatoid arthritis-associated alleles.

Molecular mimicry in the MHC: hidden clues to autoimmunity?

The term 'molecular mimicry' has been used to describe a spectrum of antigenic crossreactivities thought to underlie autoimmune disease. For T-cell crossreactivities to occur, appropriate T-cell clones must be available. Here, Harold Baum, Huw Davies and Mark Peakman speculate that an important source of self-peptides that govern thymic selection of such clones are MHC molecules themselves.