Norman A. Staines

The Human Endoplasmic Reticulum Molecular Chaperone BiP Is an Autoantigen for Rheumatoid Arthritis and Prevents the Induction of Experimental Arthritis

Rheumatoid arthritis (RA) is the most common, crippling human autoimmune disease. Using Western blotting and tandem mass spectroscopy, we have identified the endoplasmic reticulum chaperone BiP, a 78-kDa glucose-regulated protein, as a possible autoantigen. It preferentially stimulated increased proliferation of synovial T cells from patients with RA but not from patients with other arthritides. Mice with established collagen- or pristane-induced arthritis developed IgG Abs to BiP. Although BiP injected in CFA failed to induce arthritis in several strains of rats and mice, including HLA-DR4+/−- and HLA-DR1+/+-transgenic animals, it completely inhibited the development of arthritis when given i.v. 1 wk before the injection of type II collagen arthritis. Preimmunization with BiP suppressed the development of adjuvant arthritis in Lewis rats in a similar manner. This is the first report of a mammalian chaperone that is an autoantigen in human RA and in experimental arthritis and that can also prevent the induction of experimental arthritis. These findings may stimulate the development of new immunotherapies for the treatment of RA.

Mucosal tolerance and suppression of collagen-induced arthritis (CIA) induced by nasal inhalation of synthetic peptide 184-198 of bovine type II collagen (CII) expressing a dominant T cell epitope

The purpose of the study was to map the dominant T cell epitope of the CB11 sequence of CII in RT1u haplotype rats and to determine if, when used as a synthetic peptide, it would induce tolerance to protect against CIA. A dominant epitope corresponding to residues 184‐198 included in the sequence of the CB11 fragment of bovine CII was identified in proliferation assay using peptides in an epitope scanning system using synthetic peptides of 15 amino acids, overlapping by 12 amino acids. This epitope is bovine‐specific, but cross‐reacts with the corresponding rat peptide. Minor epitopes in the bovine CB11 sequence were also autoantigenic. Use of independently synthesized and purified 184‐198 peptide confirmed its dominance in the T cell responses of arthritic rats. The peptide itself was not arthritogenic. Cells from lymph nodes draining arthritic feet were particularly responsive to the dominant peptide sequence, and showed evidence of epitope spreading to include reactions to at least four subdominant epitopes. Mucosal tolerance was successfully induced by instilling CII into the nose of rats before induction of CIA; this was found to delay the onset of disease, reduce mean disease severity, shift the anti‐CII antibody response to favour antibodies of the IgGl, rather than the IgG2b isiotype, and to reduce T cell reactivity to both CII and to the 184‐198 peptide. The dominant 184‐198 peptide itself had the same tolerogenic effects when given nasally to rats daily, on the 4 days immediately preceding the induction of CIA. Two forms of CIA with acute and delayed disease onset were each modified by pre‐treatment with the peptide. This study demonstrates that mucosal tolerance to CII can be induced by delivering it nasally in a way similar to that achieved previously by oral delivery, and that the use of an immunodominant epitope contained in a synthetic peptide will also suppress the immunologic and arthritic responses to collagen.

Control of rheumatoid arthritis by oral tolerance.

OBJECTIVE Previous randomized controlled trials for treatment of rheumatoid arthritis (RA) with acid-soluble chicken and bovine type II collagen (CII) have produced conflicting results. This randomized, double-blind, controlled trial examined the therapeutic effect of bovine CII tablets in RA. METHODS CII tablets were prepared by adsorption onto a lactose base. Patients with a duration of RA of > or = 2 years and who had failed treatment with at least 1 slow-acting drug were recruited, provided that they had active arthritis. Patients were randomly assigned to receive either 0.05 mg, 0.5 mg, or 5 mg of CII or placebo daily for 6 months. All slow-acting drugs were stopped at least 4 weeks before starting CII, although prednisolone was permitted at dosages < 10 mg/day. Clinical assessments were performed at screening and at 0, 1, 4, 8, 12, 16, 20, and 24 weeks of treatment. RESULTS Fifty-five patients were recruited. Initially, there were no significant differences in mean Disease Activity Scores between groups. At 24 weeks, there was a significant difference (P = 0.041, by Kruskal-Wallis analysis of variance); the major components of this difference were attributable to relatively large decreases in the 0.5 mg CII group (19% of initial values) and to minimal decreases in patients receiving placebo (3% of initial values). Twenty patients had American College of Rheumatology 20% responses; 11 of these were in the 0.5 mg CII group and 3 were in each of the other groups, a significant difference (chi2 = 14.6, P = 0.002). There was no significant difference in any clinical measure between the placebo, 0.05 mg CII, and 5 mg CII groups. There were no side effects associated with CII treatment. CONCLUSION Treatment with 0.5 mg/day of bovine CII is well tolerated and produces small, but significant, disease improvement in RA. However, the therapeutic window is narrow. The difference between our results and those of other trials may relate to the dose, species, and formulation of the CII.

Could specific oral tolerance be a therapy for autoimmune disease

Several experimental autoimmune diseases have been successfully suppressed by the induction of specific oral tolerance. Here, Stephen Thompson and Norman Staines review the nature and mechanisms of this form of tolerance and discuss its possible applications in the control of human autoimmune diseases.

Different mechanisms by which anti-DNA MoAbs bind to human endothelial cells and glomerular mesangial cells.

The mechanisms by which anti‐DNA MoAbs derived from MRL‐lpr/lpr mice, bind to human umbilical vein endothelial cells (HUVEC) and glomerular mesangial cells were studied using a cellular ELISA. DNAse‐treatment of either the MoAb or HUVEC followed by reconstitution with DNA and/or histones was performed to determine whether DNA and histones mediated such binding. It was found that Mo Ab 410 bound to H U VEC and mesangial cells in the form of preformed DNA/anti‐DNA immune complex, and such binding was facilitated by histones. In contrast, MoAb 152 bound directly to cell membrane‐associated DNA, and adding DNA to MoAb 152 reduced its cellular binding. DNA binds endothelial cell surface and histones enhance the binding of both MoAb 410 and MoAb 152 to HUVEC by increasing cell membrane‐associated DNA. Finally, the degree of MoAb binding to HUVEC is critically influenced by the relative concentrations of antibody, DNA, and histones.

Suppression of collagen induced arthritis by oral administration of type II collagen: changes in immune and arthritic responses mediated by active peripheral suppression.

The oral administration of CII by gavage to WA/KIR rats before a conventional arthritogenic challenge with bovine CII in FIA reduced the incidence (by 23%) and delayed the onset of collagen-induced arthritis in about 50% of the animals. Selective changes in B cell and T cell responses to CII in animals treated this way are interpreted to indicate a state of tolerance or hyporesponsiveness to CII. Tolerant animals made less serum antibody, to bovine and rat CII, of the IgG2b isotype and more of the IgG1 isotype. Phenotypic and functional analysis of peripheral lymph node cells showed that those from tolerized animals expressed less MHC Class II, proliferated less and secreted less IgG2b anti-CII antibody in response to stimulation in vitro with CII when compared with cells from non-tolerant animals. However, this depression of the immune responses to CII seen in vitro was overcome when the cells were incubated with increasing amounts of CII. Tolerance could be transferred to normal animals. Spleen cells, and nylon wool-filtered splenic T cells (but not mesenteric lymph node cells) adoptively transferred hyporesponsiveness to normal recipients which were then less susceptible to collagen-induced arthritis. Transfer of serum from gavaged animals did not modify the susceptibility of normal recipients to arthritis. Spleen cells from gavaged animals suppressed proliferative and antibody responses in co-cultures in vitro with lymph node cells from animals immunized with CII in FIA. The suppressive spleen cell population contained more cells expressing MHC Class II, in both the CD8+ and CD4+ populations. These studies show that the oral administration of CII alters the subsequent immune response to the arthritogenic challenge and indicate that this oral tolerance of CII is due, not to clonal deletion or anergy, but rather to an antigen-driven active suppression mechanism that affects both T cells and B cells, most likely through the action of regulatory cytokines IL-4, IL-10 and TGF beta.

Histone Peptide-Induced Nasal Tolerance: Suppression of Murine Lupus

Induced mucosal tolerance has been shown to be beneficial in preventing or treating a number of murine and human autoimmune disorders. However, this particular form of therapy has not been thoroughly tested in systemic lupus erythematosus. In this study, we investigated the conditions for induction of nasal tolerance using a histone peptide named H471 expressing a dominant T cell epitope in the histone protein H4 of mononucleosome in lupus-prone SNF1 female mice. We also tested the effect of chronic peptide nasal treatment on the development of autoimmune reactivities in these mice. Results demonstrated that a dose-dependent nasal tolerance to peptide H471 can be achieved before or after peptide sensitization in SNF1 mice. In addition, tolerance to mononucleosomes was induced by nasal instillation of SNF1 mice with H471. This was accompanied by an increase in IL-10 and suppression of IFN-γ production by lymph node cells. Suppression of Th1-type cytokines was also observed in SNF1 mice that were nasally administered with H471 before intradermal injection with the peptide. Finally, chronic nasal instillation of mice with the H471 peptide not only suppressed the development of autoantibodies, but also altered the severity of glomerulonephritis in lupus-prone SNF1 mice.

Comparison of DNA antibody idiotypes in human sera: an international collaborative study of 19 idiotypes from 11 different laboratories.

The distribution of and relationships between 18 anti-DNA antibody idiotypes and one anti-acetylcholine receptor antibody idiotype have been tested in an international collaborative study of human sera from 180 individuals. The main finding is that the serum levels of many of these idiotypes, whether of murine or human origin, show a high degree of statistical correlation. The studies in a wide range of autoimmune rheumatic diseases confirm that none of the idiotypes tested is disease specific, but 13 of 15 (87%) whose levels were recorded as OD units or cpm correlated strongly with anti-ssDNA antibody levels and 11 of 15 (73%) with total serum IgM. Expression of several idiotypes was found to fluctuate in parallel with disease activity in SLE; levels of others were also elevated in the healthy relatives of lupus patients whilst a few were also raised in the spouses of these patients. The data support the notion that there may be only a few groups of related DNA antibody idiotypes. The correlations between the idiotypes with regard to their quantities, association with disease activity, and wide distribution in different diseases and healthy individuals suggest at least two explanations. First, all of these idiotypes may be present in normal immunoglobulin repertoires and simply increase in response to poly- or oligoclonal B-cell activation in autoimmune diseases. Secondly, these idiotypes may be structurally linked to each other, so that their behaviour under conditions of specific antigenic stimulation is similar. Genetic and structural studies will be required to distinguish between these possibilities.

Importance of dose of type II collagen in suppression of collagen-induced arthritis by nasal tolerance

OBJECTIVE To determine the influence of the dose of collagen given nasally on the induction of specific mucosal tolerance in collagen-induced arthritis. METHODS The severity of clinical arthritis induced in DBA/1 mice was studied after the nasal administration (before disease induction) of 1 of 4 doses (across a 2-log range) of bovine type II collagen (CII). Parameters of immunity included lymphocyte proliferation and cytokine production in vitro in response to antigen stimulation, and the production of anticollagen IgG antibody subclasses. RESULTS The 3 highest doses (20, 80, and 320 microg) ameliorated disease severity, whereas the lowest dose (5 microg) aggravated disease. These findings correlated well with antigen-specific T cell proliferation and cytokine and antibody production. T cell proliferation was suppressed by the higher doses of CII, whereas the low dose enhanced T cell proliferation, indicating it primed the T cells. Suppression of T cell proliferation could be overcome by the addition of exogenous interleukin-2 (IL-2) to these cultures. Decreased T cell proliferation was associated with suppression of both Th1 (interferon-gamma [IFNgamma]) and Th2 (IL-4) cytokines and all the subclasses of anticollagen IgG in mice receiving 20, 80, or 320 microg of collagen. Overall, the highest dose of collagen (320 microg) was less effective at suppressing the immune response and disease than the 20-microg or 80-microg doses. There was an increased production of antibodies of all IgG isotypes, and of the Th1-associated cytokines IFNgamma and IL-2, in animals that had received the lowest dose of 5 microg collagen nasally. CONCLUSION Nasal administration of antigens is effective in inducing tolerance and reducing disease severity, but the effects are dose dependent. Low doses can prime the immune system and aggravate disease; high doses may not suppress disease. Suppression of the immune response, which correlates with suppression of disease, is not obviously associated with a type I to type II T cell switch, but rather with an overall suppression of both forms of T cell response, with a potential role for anergy of T cells in this process.

A role for DNA in anti-DNA antibodies binding to endothelial cells

Vascular injury and microvascular thrombosis are prominent features of systemic lupus erythematosus, as are circulating DNA-binding antibodies (DNAb). Experimental glomerulonephritis can be induced by anti-endothelial cell antibodies, and polyreactive DNAb might be pathogenetic by binding to endothelial cells, perhaps influencing their non-thrombogenic nature. To test this hypothesis, eight monoclonal antibodies (mAb) that bind to DNA derived from (NZB x NZW)F1 or MRL/Mp-lpr/lpr mice, were tested for their ability to bind to human umbilical vein endothelial cells (HUVEC). Binding was assessed using flow cytometry, fluorescence microscopy and cellular ELISA. Three of the eight mAb, at concentrations employed in this study, bound to HUVEC and dermal fibroblasts. Of these three mAb, one bound also to platelets. Two of the three demonstrated strong binding to (1) freshly isolated, collagenase-digested HUVEC, (2) 2nd passage HUVEC in suspension after trypsinization and, (3) 2nd passage HUVEC growing on plastic plates. To determine whether DNA itself acted as a ligand in this binding, prior treatment with DNAase was studied. Treatment of the endothelial cells with DNAase had no effect on the binding of one mAb, but DNAase treatment of this monoclonal itself resulted in a 60% reduction in binding to HUVEC, suggesting that the binding might be mediated through DNA in the form of a DNA/anti-DNA immune complex. In contrast, DNAase digestion of the endothelial cells caused a 40% reduction in the binding of the other two monoclonal antibodies. Furthermore, one of the two mAb bound 30% more to HUVEC after themselves being subjected to DNAase treatment. These two monoclonals may therefore be binding directly to HUVEC, possibly to DNA associated with the membrane. Prior DNAase digestion of dermal fibroblasts had a more profound effect on the binding of all three autoantibodies compared to HUVEC after similar treatment. Therefore, DNA can bind independently to either antibody or cell, thus supporting build up of complexes and capture of preformed complexes. Functionally, the binding of mAb to HUVEC did not influence thrombin-induced prostacyclin synthesis, in contrast to a control monoclonal anti-endothelial cell antibody EN4, which did.