Immunopathologic role of proteoglycan antigens in rheumatoid joint disease

Identification of antibodies against extracellular matrix proteins in human osteoarthritis

We investigated the presence of autoantibodies against the extracellular matrix proteins thrombospondin-4 (TSP-4), cartilage oligomeric matrix protein (COMP), C-type lectin domain family 3 member A (CLEC3A), collagen II, collagen VI, matrilin-3, and fibrillin-2 in the serum of osteoarthritis (OA) patients. We compared those results with the presence of such antibodies in rheumatoid arthritis (RA) patients and in healthy donors (HD). Our study examines whether antibodies against extracellular proteins can be used as potential biomarkers to support the clinical diagnosis of OA. 10 OA, 10 RA patients and 10 HD were enrolled in this explorative cross-sectional study. SDS-PAGE and immunoblot were used to investigate the presence of antibodies against extracellular matrix proteins. The serum of 5/10 OA patients but 0/10 HD exhibited TSP-4 IgG isotype antibodies (P = 0.033). The serum of 8/10 OA patients but only 1/10 HD exhibited IgG isotype antibodies against TSP-4 or COMP (P = 0.005). The serum of 9/10 OA patients but only 1/10 HD exhibited IgG isotype antibodies against TSP-4, COMP or CLEC3A (P = 0.005). We found strong evidence for the presence of IgG isotype autoantibodies against the cartilage extracellular matrix proteins TSP-4, COMP and CLEC3A in OA. The detection of IgG isotype autoantibodies against TSP-4, COMP and CLEC3A may support the clinical diagnosis of OA. OA with autoantibodies against cartilage extracellular matrix proteins defines a new OA subgroup suggesting that patients with high concentrations of autoantibodies may benefit from an immune suppressive therapy.

Cytokine-producing B cells: a translational view on their roles in human and mouse autoimmune diseases

B-cell depletion therapy has beneficial effects in autoimmune diseases. This is only partly explained by an elimination of autoantibodies. How does B-cell depletion improve disease? Here, we review preclinical studies showing that B cells can propagate autoimmune disorders through cytokine production. We also highlight clinical observations indicating the relevance of these B-cell functions in human autoimmunity. Abnormalities in B-cell cytokine production have been observed in rheumatoid arthritis, multiple sclerosis, inflammatory bowel disease, and systemic lupus erythematosus. In the first two diseases, B-cell depletion erases these abnormalities, and improves disease progression, suggesting a causative role for defective B-cell cytokine expression in disease pathogenesis. However, in the last two disorders, the pathogenic role of B cells and the effect of B-cell depletion on cytokine-producing B cells remain to be clarified. A better characterization of cytokine-expressing human B-cell subsets, and their modulation by B cell-targeted therapies might help understanding both the successes and failures of current B cell-targeted approaches. This may even lead to the development of novel strategies to deplete or amplify selectively pathogenic or protective subsets, respectively, which might be more effective than global depletion of the B-cell compartment.

Rheumatoid arthritis vaccine therapies: perspectives and lessons from therapeutic ligand epitope antigen presentation system vaccines for models of rheumatoid arthritis

The current status of therapeutic vaccines for autoimmune diseases is reviewed with rheumatoid arthritis as the focus. Therapeutic vaccines for autoimmune diseases must regulate or subdue responses to common self-antigens. Ideally, such a vaccine would initiate an antigen-specific modulation of the T-cell immune response that drives the inflammatory disease. Appropriate animal models and types of T helper cells and signature cytokine responses that drive autoimmune disease are also discussed. Interpretation of these animal models must be done cautiously because the means of initiation, autoantigens, and even the signature cytokine and T helper cell (Th1 or Th17) responses that are involved in the disease may differ significantly from those in humans. We describe ligand epitope antigen presentation system vaccine modulation of T-cell autoimmune responses as a strategy for the design of therapeutic vaccines for rheumatoid arthritis, which may also be effective in other autoimmune conditions.

Intramolecular polyspecificity in CD4 T-cell recognition of Ad-restricted epitopes of proteoglycan aggrecan

T-cell recognition of MHC–peptide complexes shows a high degree of polyspecificity extending to recognition of a large number of structurally unrelated peptides. Examples of polyspecificity reported to date are confined to recognition of epitopes from distinct proteins or synthetic peptide libraries. Here we describe intramolecular polyspecificity of CD4 T cells specific for several epitopes within proteoglycan aggrecan, a structural glycoprotein of cartilage and candidate autoantigen in rheumatoid arthritis. T-cell hybridomas from aggrecan-immunized mice recognized four structurally unrelated epitopes from the G1 domain of aggrecan, but not other aggrecan epitopes or a variety of other peptide epitopes restricted by the same MHC class II allele. We also showed that the hierarchy of cross-reactivity broadly correlated with the strength of peptide binding to MHC class II. Similar polyspecificity was observed in responses of lymph node cells from peptide-immunized mice, suggesting polyspecificity of a significant proportion of the in vivo aggrecan specific T-cell repertoire. Polyspecific recognition of several epitopes within the same autoantigen may provide a novel mechanism to reach the activation threshold of low-affinity autoreactive T cells in the initiation of autoimmune diseases.

Proteoglycan aggrecan conducting T cell activation and apoptosis in a murine model of rheumatoid arthritis

Rheumatoid arthritis (RA) is a systemic autoimmune disease and its targeting of the joints indicates the presence of a candidate autoantigen(s) in synovial joints. Patients with RA show immune responses in their peripheral blood to proteoglycan (PG) aggrecan. One of the most relevant animal models of RA appears to be proteoglycan-induced arthritis (PGIA), and CD4⁺ T cells seem to play a crucial role in the initiation of the disease. In this review, the role of various T cell epitopes of aggrecan in the induction of autoreactive T cell activation and arthritis is discussed. We pay special attention to two critically important arthritogenic epitopes, 5/4E8 and P135H, found in the G1 and G3 domains of PG aggrecan, respectively, in the induction of autoimmune arthritis. Finally, results obtained with the recently developed PG-specific TCR transgenic mice system showed that altered T cell apoptosis, the balance of activation, and apoptosis of autoreactive T cells are critical factors in the development of autoimmunity.

Non-MHC risk alleles in rheumatoid arthritis and in the syntenic chromosome regions of corresponding animal models

Rheumatoid arthritis (RA) is a polygenic autoimmune disease primarily affecting the synovial joints. Numerous animal models show similarities to RA in humans; some of them not only mimic the clinical phenotypes but also demonstrate the involvement of homologous genomic regions in RA. This paper compares corresponding non-MHC genomic regions identified in rodent and human genome-wide association studies (GWAS). To date, over 30 non-MHC RA-associated loci have been identified in humans, and over 100 arthritis-associated loci have been identified in rodent models of RA. The genomic regions associated with the disease are designated by the name(s) of the gene having the most frequent and consistent RA-associated SNPs or a function suggesting their involvement in inflammatory or autoimmune processes. Animal studies on rats and mice preferentially have used single sequence length polymorphism (SSLP) markers to identify disease-associated qualitative and quantitative trait loci (QTLs) in the genome of F2 hybrids of arthritis-susceptible and arthritis-resistant rodent strains. Mouse GWAS appear to be far ahead of rat studies, and significantly more mouse QTLs correspond to human RA risk alleles.

Impaired activation-induced cell death promotes spontaneous arthritis in antigen (cartilage proteoglycan)-specific T cell receptor-transgenic mice

OBJECTIVE

To investigate whether genetic preponderance of a T cell receptor (TCR) recognizing an arthritogenic peptide of human cartilage proteoglycan (PG) is sufficient for development of arthritis.

METHODS

We performed a longitudinal study using BALB/c mice expressing a TCR that recognizes the arthritogenic ATEGRVRVNSAYQDK peptide of human cartilage PG. PG-specific TCR-transgenic (PG-TCR-Tg) mice were inspected weekly for peripheral arthritis until 12 months of age. Peripheral joints were examined histologically, and T cell responses, T cell activation markers, serum cytokines, and autoantibodies were measured. Apoptosis and signaling studies were performed in vitro on T cells from aged PG-TCR-Tg mice.

RESULTS

Spontaneous arthritis developed as early as 5-6 months of age, and the incidence increased to 40-50% by 12 months of age. Progressive inflammation began with cartilage and bone erosions in the interphalangeal joints, and later expanded to the proximal joints of the front and hind paws. Spontaneous arthritis was associated with a high proportion of activated CD4+ T cells, enhanced interferon-γ and interleukin-17 (IL-17) production, and elevated levels of serum autoantibodies. PG-TCR-Tg mice lacking IL-4 developed arthritis earlier and at a higher incidence than IL-4-sufficient mice. Antigen-specific activation-induced cell death was diminished in vitro in CD4+ T cells of PG-TCR-Tg mice with spontaneous arthritis, especially in those lacking IL-4.

CONCLUSION

The presence of CD4+ T cells expressing a TCR specific for an arthritogenic PG epitope is sufficient to trigger spontaneous autoimmune inflammation in the joints of BALB/c mice. IL-4 appears to be a negative regulator of this disease, through attenuation of activation-induced cell death.

Pan-DR-binding Hsp60 self epitopes induce an interleukin-10-mediated immune response in rheumatoid arthritis

OBJECTIVE

Human Hsp60 is expressed in the joints of patients with rheumatoid arthritis (RA) and can elicit a regulatory T cell response in the peripheral blood and synovial fluid. However, Hsp60 can also trigger strong proinflammatory pathways. Thus, to understand the nature of these Hsp60-directed responses in RA, it is necessary to study such responses at the molecular, epitope-specific level. This study was undertaken to characterize the disease specificity and function of pan-DR-binding Hsp60-derived epitopes as possible modulators of autoimmune inflammation in RA.

METHODS

Lymphocyte proliferation assays (using (3)H-thymidine incorporation and carboxyfluorescein diacetate succinimidyl ester [CFSE] staining) and measurement of cytokine production (using multiplex immunoassay and intracellular staining) were performed after in vitro activation of peripheral blood mononuclear cells from patients with RA, compared with healthy controls.

RESULTS

A disease (RA)-specific immune recognition, characterized by T cell proliferation as well as increased production of tumor necrosis factor alpha (TNFalpha), interleukin-1beta (IL-1beta), and IL-10, was found for 3 of the 8 selected peptides in patients with RA as compared with healthy controls (P < 0.05). Intracellular cytokine staining and CFSE labeling showed that CD4+ T cells were the subset primarily responsible for both the T cell proliferation and the cytokine production in RA. Interestingly, the human peptides had a remarkably different phenotype, with a 5-10-fold higher IL-10:TNFalpha ratio, compared with that of the microbial peptides.

CONCLUSION

These results suggest a disease-specific immune-modulatory role of epitope-specific T cells in the inflammatory processes of RA. Therefore, these pan-DR-binding epitopes could be used as a tool to study the autoreactive T cell response in RA and might be suitable candidates for use in immunotherapy.

The value of animal models in predicting genetic susceptibility to complex diseases such as rheumatoid arthritis

For a long time, genetic studies of complex diseases were most successfully conducted in animal models. However, the field of genetics is now rapidly evolving, and human genetics has also started to produce strong candidate genes for complex diseases. This raises the question of how to continue gene-finding attempts in animals and how to use animal models to enhance our understanding of gene function. In this review we summarize the uses and advantages of animal studies in identification of disease susceptibility genes, focusing on rheumatoid arthritis. We are convinced that animal genetics will remain a valuable tool for the identification and investigation of pathways that lead to disease, well into the future.

The enthesis organ concept and its relevance to the spondyloarthropathies

A characteristic feature of the spondyloarthropathies is inflammation at tendon or ligament attachment sites. This has traditionally been viewed as a focal abnormality, even though the inflammatory reaction intrinsic to enthesitis may be quite extensive. We argue that the diffuse nature of the pathology is best understood in the context of an 'enthesis organ concept'. This highlights the fact that stress concentration at an insertion site involves not only the enthesis itself, but neighbouring tissues as well. The archetypal enthesis organ is that of the Achilles tendon where intermittent contact between tendon and bone immediately proximal to the enthesis leads to the formation of fibrocartilages on the deep surface of the tendon and on the opposing calcaneal tuberosity, but similar functional modifications are widespread throughout the skeleton. Many entheses have bursae and fat near the insertion site and both of these serve to promote frictionless movement. Collectively, the fibrocartilages, bursa, fat pad and the enthesis itself constitute the enthesis organ. However, it also includes both the immediately adjacent trabecular bone networks and in some cases deep fascia. The concept of a synovio-entheseal complex (SEC) and of a 'functional enthesis' are complimentary to that of an enthesis organ and also have important implications for understanding spondyloarthropathy. The SEC concept emphasizes the interdependence between synovial membrane and entheses within enthesis organs. It draws attention to the fact that one component (the enthesis) is prone to microdamage and the other (the synovium) to inflammation. If an enthesis is damaged, any ensuing inflammatory reaction is likely to occur in the synovium. The concept of a 'functional enthesis' serves to emphasise anatomical, biomechanical and pathological features that are shared between true fibrocartilaginous entheses and regions proximal to the attachment sites themselves where tendons or ligaments wrap around bony pulleys. Such'wrap-around regions' are well documented sites of pathology in SpA-for tenosynovitis is a recognized feature. Stress concentration at the enthesis itself is dissipated at many sites by fibrous connections between one tendon or ligament and another, close to the insertion site. At a microscopic level, enthesis fibrocartilage is of paramount importance in ensuring that fibre bending of the tendon or ligament is not focused at the hard tissue interface. Normal enthesis organs are avascular in their fibrocartilaginous regions, but tissue microdamage to entheses is common and appears to be associated with tissue repair responses and vessel ingrowth. This makes the enthesis organ a site where adjuvant molecules derived from bacteria may be preferentially deposited. This microdamage and propensity for bacterial molecule deposition in the context of genetic factors such as HLA-B27 appears to lead to the characteristic inflammatory changes of AS. Understanding the enthesis organ concept helps to explain synovitis and osteitis in spondyloarthropathy. An appreciation of the complex anatomy of 'articular enthesis organs' (e.g., that associated with the distal interphalangeal joints) is helpful in understanding disease patterns in psoriatic arthritis. In this chapter, we review the extent and types ofenthesis organs and show how a patho-anatomic appreciation of these structures leads to a new platform for understanding the pathogenesis of SpA.

Increased arthritis susceptibility in cartilage proteoglycan–specific T cell receptor–transgenic mice

OBJECTIVE

To better understand the role of antigen (arthritogenic epitope)-specific T cells in the development of autoimmune arthritis.

METHODS

A transgenic (Tg) mouse expressing the T cell receptor (TCR) Valpha1.1 and V(beta)4 chains specific for a dominant arthritogenic epitope (designated 5/4E8) of human cartilage proteoglycan (HuPG) aggrecan was generated. This TCR-Tg mouse strain was backcrossed into the PG-induced arthritis (PGIA)-susceptible BALB/c strain and tested for arthritis incidence and severity.

RESULTS

CD4+ TCR-Tg T cells carried functionally active TCR specific for a dominant arthritogenic epitope of HuPG (5/4E8). T cells of naive TCR-Tg mice were in an activated stage, since the in vitro response to HuPG or to peptide stimulation induced interferon-gamma and interleukin-4 production. TCR-Tg mice uniformly, without exception, developed severe and progressive polyarthritis, even without adjuvant. Inflamed joints showed extensive cartilage degradation and bone erosions, similar to that seen in the arthritic joints of wild-type BALB/c mice with PGIA. Spleen cells from both naive and HuPG-immunized arthritic TCR-Tg mice could adoptively transfer arthritis when injected into syngeneic BALB/c.SCID recipient mice.

CONCLUSION

TCR-Tg BALB/c mice display increased arthritis susceptibility and develop aggravated disease upon in vivo antigen stimulation. This model using TCR-Tg mice is a novel and valuable research tool for studying mechanisms of antigen (arthritogenic epitope)-driven regulation of arthritis and understanding how T cells recognize autoantigen in the joints. This type of mouse could also be used to develop new immunomodulatory strategies in T cell-mediated autoimmune diseases.

Molecular manipulation with the arthritogenic epitopes of the G1 domain of human cartilage proteoglycan aggrecan

Systemic immunization of BALB/c mice with human cartilage proteoglycan (PG) aggrecan induces progressive polyarthritis. The G1 domain of the PG aggrecan molecule contains most of the T cell epitopes, including three immunodominant ('arthritogenic') and at least six subdominant T cell epitopes. The three dominant T cell epitopes (P49, P70 and P155) were deleted individually or in combination by site directed mutagenesis, and the recombinant human G1 (rhG1) domain (wild type and mutated) proteins were used for immunization. Close to 100% of BALB/c mice immunized with the wild-type (nonmutated) rhG1 domain developed severe arthritis, which was 75% in the absence of P70 (5/4E8) epitope, and very low (< 10% incidence) when all three dominant T cell epitopes were deleted. The onset was delayed and the severity of arthritis reduced in animals when dominant T cell epitopes were missing from the immunizing rhG1 domain. The lack of T cell response to the deleted epitope(s) was specific, but the overall immune response against the wild-type rhG1 domain of human PG was not significantly affected. This study helped us to understand the dynamics and immune-regulatory mechanisms of arthritis, and supported the hypothesis that the development of autoimmune arthritis requires a concerted T cell response to multiple epitopes, rather than the immune response to a single arthritogenic structure.

Naive transgenic T cells expressing cartilage proteoglycan-specific TCR induce arthritis upon in vivo activation

Proteoglycan (PG)-induced arthritis (PGIA), a murine model for rheumatoid arthritis (RA), is driven by antigen (PG)-specific T and B cell activation. In order to analyze the pathogenic role of antigen-specific T cells in the development of autoimmune arthritis, we have generated a transgenic (Tg) mouse. The CD4(+) T cells of this TCR-5/4E8-Tg line express a functional T cell receptor (TCR) composed of the Valpha1.1 and Vbeta4 chains with specificity for the dominant arthritogenic T cell epitope of human cartilage PG. Adoptive transfer of naive TCR-5/4E8-Tg cells induced arthritis with severe clinical symptoms in syngeneic immunodeficient BALB/c.RAG2(-/-) mice. In vivo activation of TCR-5/4E8-Tg CD4(+)Vbeta4(+) cells with cartilage PG seemed to be critical for arthritis induction. Arthritis never developed after transfer of naive wild-type cells. The arthritis was characterized as a chronic progressive disease with intermittent spontaneous exacerbations and remissions. Inflamed joints showed extensive cartilage damage and bone erosions leading to massive ankylosis in peripheral joints. These PG epitope-specific TCR-5/4E8-Tg mice can be valuable research tools for studying antigen-driven T cell regulation in arthritis, and migration of T cells to the joints. In addition the model may be used for the development of immune modulating strategies in T cell-mediated autoimmune diseases.

T-cell recognition of differentially tolerated epitopes of cartilage proteoglycan aggrecan in arthritis

Proteoglycan (PG) aggrecan, a major macromolecular component of cartilage, is highly immunogenic; it induces arthritis in genetically susceptible BALB/c mice. The present study maps the T-cell epitope repertoire of cartilage PG by identifying a total of 27 distinct T-cell epitopes. An epitope hierarchy, accounting for the different effector functions of PG-specific T cells, and determinant spreading, has been found. T-cell responses to four epitopes were associated with arthritis induction. Some of the T-cell epitopes were full T-cell activators, whereas a number of subdominant and cryptic epitopes proved to be partial activators in vitro, inducing either cytokine secretion or T-cell proliferation, but not both. A few T-cell epitopes of the core protein of cartilage PG were clearly recognized by T cells in PG-immunized arthritic animals, but the corresponding peptides did not induce T-cell responses when injected into naive BALB/c mice; thus these T-cell epitopes were designated as "conditionally immunogenic."

A longitudinal study on an autoimmune murine model of ankylosing spondylitis

BACKGROUND

Proteoglycan aggrecan (PG)-induced arthritis (PGIA) is the only systemic autoimmune murine model which affects the axial skeleton, but no studies have been performed characterising the progression of spine involvement.

OBJECTIVES

To follow pathological events in experimental spondylitis, and underline its clinical, radiographic, and histological similarities to human ankylosing spondylitis (AS); and to determine whether the spondyloarthropathy is a shared phenomenon with PGIA, or an "independent" disease.

METHODS

Arthritis/spondylitis susceptible BALB/c and resistant DBA/2 mice, and their F1 and F2 hybrids were immunised with cartilage PG, and radiographic and histological studies were performed before onset and weekly during the progression of spondylitis.

RESULTS

About 70% of the PG immunised BALB/c mice develop spondyloarthropathy (proteoglycan-induced spondylitis (PGISp), and the progression of the disease is very similar to human AS. It begins with inflammation in the sacroiliac joints and with enthesitis, and then progresses upwards, affecting multiple intervertebral disks. In F2 hybrids of arthritis/spondylitis susceptible BALB/c and resistant DBA/2 mice the incidence of arthritis was 43.5%, whereas the incidence of spondylitis was >60%. Some arthritic F2 hybrid mice had no spondylitis, whereas others developed spondylitis in the absence of peripheral arthritis.

CONCLUSIONS

The PGISp model provides a valuable tool for studying autoimmune reactions in spondylitis, and identifying genetic loci associated with spondyloarthropathy.

Analysis of the CD8+ T cell response to the G1 domain of aggrecan in ankylosing spondylitis

BACKGROUND

CD4+ T cell responses to the G1 domain of aggrecan in patients with ankylosing spondylitis (AS) were recently reported. Whether such an immune response can be seen in the CD8+ subpopulation has not yet been determined.

OBJECTIVE

To determine if HLA-B27 restricted G1-specific CD8+ T cells are present in AS and to analyse immunodominant CD8+ T cell epitopes.

METHODS

Peripheral blood mononuclear cells of 45 patients with AS were stimulated with overlapping 18-mer peptides covering the whole G1 protein. Results were compared with those for patients with rheumatoid arthritis (RA) and healthy controls. For epitope analysis, G1-specific interferon gamma positive (IFNgamma+) T cells were isolated by magnetic activated cell sorting. After in vitro expansion, CD8+ T cells were restimulated with 14 subpools of G1 peptides. T cells responding to G1 peptide subpools were quantified by flow cytometry according to IFNgamma secretion. Predicted peptides were subsequently confirmed by stimulation with single peptides.

RESULTS

G1-specific CD8+ T cell responses were found in 29/45 (64%) patients with AS, 18/35 (51%) patients with RA, but not in healthy controls. Five CD8+ T cell epitopes were identified as immunodominant in five patients. However, the T cell response was not HLA-B27 restricted. Nonamer peptides with an HLA-B27 binding motif did not induce a T cell response.

CONCLUSION

A G1 peptide-specific CD8+ T cell response is present in AS but also in patients with RA. It does not seem to be HLA-B27 restricted. Whether such a response has a role in the pathogenesis of AS needs clarification.

Induction of arthritis in HLA-DR4-humanized and HLA-DQ8-humanized mice by human cartilage proteoglycan aggrecan but only in the presence of an appropriate (non-MHC) genetic background

OBJECTIVE

To determine whether the rheumatoid arthritis (RA)-predisposing class II molecules of the major histocompatibility complex (MHC) can present cartilage proteoglycan (PG) aggrecan, and if so, to determine the epitope repertoire of the human cartilage PG in HLA-transgenic mice and determine whether HLA-transgenic mice develop arthritis in response to immunization with human cartilage PG.

METHODS

Mice transgenic for HLA-DR2.Ab(0), DR3.Ab(0), DR4.Ab(0), and DQ8.Ab(0), lacking their own (mouse) class II antigens (Ab(0)), on the original (arthritis-resistant) and the arthritis-susceptible BALB/c backgrounds, were immunized with human cartilage PG. The T cell epitope repertoire presented by these class II MHC alleles was determined using a synthetic peptide library (143 peptides of the core protein of human cartilage PG), and arthritis development was monitored and compared in wild-type and HLA-transgenic/congenic BALB/c mice.

RESULTS

Mice of the 4 HLA-transgenic lines, either on the original mixed, arthritis-resistant background or DR4.Ab(0)- and DQ8.Ab(0)-transgenic/congenic mice on the arthritis-susceptible BALB/c genetic background, responded well to PG immunization (as assessed by T cell responses and antibody and cytokine production), and a number of T cell epitopes along the core protein of human cartilage PG were identified. DR4.Ab(0)- and DQ8.Ab(0)-transgenic mice immunized with human cartilage PG developed arthritis, but only when these class II MHC molecules were present on the arthritis-susceptible (BALB/c) genetic background.

CONCLUSION

A number of human cartilage PG epitopes can be presented by HLA alleles that predispose to the development of RA, but the epitopes of the cartilage PG presented by HLA-DR4 or HLA-DQ8 can induce arthritis only in the presence of an appropriate genetic (non-MHC) background.

Induction of arthritis in SCID mice by T cells specific for the “shared epitope” sequence in the G3 domain of human cartilage proteoglycan

OBJECTIVE

To study the immunologic function and determine the fine epitope structure of a synthetic peptide p135H ((2373)TTYKRRLQKRSSRHP) of the G3 domain of human cartilage proteoglycan (aggrecan), which contains a highly homologous sequence motif of the shared epitope (QKRAA), the most common sequence motif in HLA-DR4 alleles, which predispose humans to the development of rheumatoid arthritis (RA).

METHODS

Synthetic p135 peptides with altered sequences were used for (hyper)immunization of arthritis-susceptible BALB/c mice and then challenged with a single dose of cartilage proteoglycan. Human p135 (p135H) and mouse p135 (p135M) synthetic peptides of the G3 domain of aggrecan were used to prime lymphocytes, which were then used for adoptive transfer of arthritis into "presensitized" SCID mice, determining cross-reactivity among p135 peptides and their analogous sequences, and generating T cell hybridomas. T cell hybridomas were also used for arthritis transfer into SCID mice and for characterizing the fine epitope structure of T cell receptor (TCR) and major histo-compatibility complex (MHC) binding sites of the immunogenic/arthritogenic p135H sequence.

RESULTS

While p135H peptide-(hyper)immunized mice became sensitized, they developed arthritis only after injection of a single dose of cartilage proteoglycan aggrecan. An altered peptide sequence (p135H-AA) carrying the shared epitope motif (QKRAA) was as effective as the natural peptide p135H sequence for inducing arthritis. Mouse p135M-specific lymphocytes induced arthritis with a lower incidence, but synthetic peptides to Escherichia coli heat-shock protein (DnaJ) or HLA-DR4 allele (both having the shared epitope sequence with different flanking regions) were also positive. Fine epitope sequence recognition of an arthritogenic T cell hybridoma derived from p135H-primed lymphocyte population was determined. Interestingly, in the most central position, a basic amino acid triplet of p135H peptide was found to be the MHC-binding motif, whereas the flanking amino acids bound to the TCR.

CONCLUSION

Peptide p135H, corresponding to the peptide sequence in the G3 domain of human cartilage proteoglycan aggrecan, is immunogenic/arthritogenic in BALB/c mice. Peptide p135H includes a highly homologous motif of the shared epitope, a sequence that is overrepresented in bacterial heat-shock proteins, envelope protein of human JC polyomavirus, and numerous HLA-DR4 alleles. Since the G3 domain of cartilage proteoglycan aggrecan with the p135 sequence is "lost" during the normal metabolic turnover of cartilage proteoglycan or in pathologic conditions, an antigenoriented T cell migration into joints of presensitized (susceptible) individuals may contribute to the organ-specificity of RA.

Aggrecan: A Target Molecule of Autoimmune Reactions

Aggrecan in cartilage forms aggregates with hyaluronan and link protein, embedded in a collagen network. It accounts for the compressive stiffness and resilience of the hyaline cartilage. Many forms of inflammatory arthritis were shown to be accompanied with aggrecan degradation and loss from the cartilage. The loss of this major component of cartilage renders the tissue more vulnerable when exposed to abrasive forces. Therefore, aggrecan degradation may significantly contribute to cartilage destruction in arthritis. Furthermore, fragments of degraded aggrecan are released during joint inflammation. Thus, molecules of an avascular, immune-privileged tissue (hyaline cartilage) may become accessible to the cells of the immune system. Similarly, there is a "leakage" of aggrecan fragments from cartilage during aging and after joint injury, which may also lead to autosensibilisation. Autoimmune reactivity to aggrecan can be detected in human joint diseases, as well as in animal models of arthritis. The epitopes involved in these processes are currently being identified. Recent data from work with mice suggest a strong immune response focused to the N-terminal G1 domain of aggrecan that leads to arthritis and spondylitis.

Induction of arthritis in BALB/c mice by cartilage link protein: involvement of distinct regions recognized by T and B lymphocytes

Both type II collagen and the proteoglycan aggrecan are capable of inducing an erosive inflammatory polyarthritis in mice. In this study we provide the first demonstration that link protein (LP), purified from bovine cartilage, can produce a persistent, erosive, inflammatory polyarthritis when injected repeatedly intraperitoneally into BALB/c mice. We discovered a single T-cell epitope, located within residues 266 to 290 of bovine LP (NDGAQIAKVGQIFAAWKLLGYDRCD), which is recognized by bovine LP-specific T lymphocytes. We also identified three immunogenic regions in bovine LP that contain epitopes recognized by antibodies in hyperimmunized sera. One of these B-cell regions is found in the most species-variable domain of LP (residues 1 to 36), whereas the other epitopes are located in the most conserved regions (residues 186 to 230 and 286 to 310). The latter two regions contain an AGWLSDGSVQYP motif shared by the G1 globulin domain of aggrecan core protein, versican, neurocan, glial hyaluronan-binding protein, and the hyaluronan receptor CD44. Our data reveal that the induction of arthritis is associated with antibody reactivities to B-cell epitopes located at residues 1 to 19. Together, these observations show that another cartilage protein, LP, like type II collagen and the proteoglycan aggrecan, is capable of inducing an erosive inflammatory arthritis in mice and that the immunity to LP involves recognition of both T- and B-cell epitopes. This immunity may be of importance in the pathogenesis of inflammatory joint diseases, such as juvenile rheumatoid arthritis, in which cellular immunity to LP has been demonstrated.

Presence of antibodies to native G1 domain of aggrecan core protein in synovial fluids from patients with various joint diseases

OBJECTIVE

To investigate the occurrence of IgG antibodies to aggrecan in synovial fluids (SF) from patients with arthritis and various articular diseases, and to determine the nature of epitopes present within aggrecan that react with these antibodies.

METHODS

SF samples were reacted with native aggrecan, reduced and alkylated aggrecan, chondroitin sulfate, and keratan sulfate, using dot-blots and a novel enzyme-linked immunosorbent assay (ELISA). The nature of the epitopes present on aggrecan was elucidated using Western blots and a competitive inhibition ELISA.

RESULTS

IgG antibodies to aggrecan were found in > 50% of the SF samples tested. No IgG antibody reactivity was observed in serum from the same patients. The antibodies appeared to react predominantly with native aggrecan, and there was no disease specificity. It was shown that the epitopes to these antibodies were located within the N-terminal region of the core protein.

CONCLUSION

This study demonstrates the frequent occurrence of IgG antibodies to aggrecan in human SF. The major epitope is located in the G1 domain of the aggrecan core protein. These IgG antibodies appear to be produced locally within the synovial cavity, probably in response to various articular diseases, resulting in the loss of native aggrecan from articular cartilage.

B- and T-cell autoantigens in pristane-induced arthritis

Pristane-induced arthritis (PIA) is a murine disease resembling rheumatoid arthritis (RA) which is characterized by autoimmune responses to joint tissues. To identify the range of potential antigens targeted in PIA, proteins from arthritic or normal joint extracts were fractionated by sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) and systematically screened for the ability to react with either serum IgG, or cultured splenic T cells, obtained from healthy or arthritic mice. Extracts from both normal and arthritic animals contained multiple proteins that were capable of reacting with murine serum IgG in immunoblotting experiments. In healthy controls, more bands were identified in extracts prepared from 30-week-old mice than from 8-week-old animals, but the widest range of proteins bound were derived from arthritic joints. Furthermore, the sera from PIA-positive mice reacted with more bands from each of the extracts than did normal sera. Fractionated extracts prepared from healthy joints failed to stimulate the in vitro proliferation of splenic T cells from either normal or arthritic animals. When arthritic joint components were screened, T cells from healthy mice responded weakly to some fractions, but multiple fractions elicited strong proliferation by T cells from mice with PIA. A band of apparent molecular mass 60000 was the protein most commonly bound by serum IgG from arthritic mice, and the corresponding fraction stimulated the highest responses by T cells from PIA-positive animals. These results are consistent with the notion that the 60,000 MW mammalian heat-shock protein is an important antigen in PIA, but that the autoimmune response diversifies with the development of arthritis to target multiple joint components.

HLA-DQ8 transgenic mice are highly susceptible to collagen-induced arthritis: a novel model for human polyarthritis

Genetic studies have indicated that susceptibility to rheumatoid arthritis (RA) maps to the HLA-DR locus of the major histocompatibility complex. Strong linkage disequilibrium between certain HLA-DQ genes and HLA-DR genes associated with RA, however, suggests that HLA-DQ molecules may also play a role in RA susceptibility. To examine the role of HLA-DQ molecules in arthritis, we generated transgenic mice expressing the DQA1*0301 and DQB1*0302 genes from an RA predisposing haplotype (DQ8/DR4Dw4). The transgenes were introduced into mouse class II-deficient H-2Ab0 mice, and their susceptibility to experimental collagen-induced arthritis was evaluated. The HLA-DQ8+,H-2Ab0 mice displayed good expression of the DQ8 molecule, while no surface expression of endogenous murine class II molecules could be detected. The DQ8 molecule also induced the selection of CD4+ T cells expressing a normal repertoire of V beta T cell receptors. Immunization of HLA-DQ8+,H-2Ab0 mice with bovine type II collagen (CII) induced a strong antibody response that was cross-reactive to homologous mouse CII. Also, in vitro proliferative responses against bovine CII, which were blocked in the presence of an antibody specific for HLA-DQ and mouse CD4, were detected. Finally, a severe polyarthritis developed in a majority of HLA-DQ8+,H-2Ab0 mice, which was indistinguishable from the disease observed in arthritis susceptible B10.T(6R) (H-2Aq) controls. In contrast, HLA-DQ8-,H-2Ab0 fullsibs did not generate CII antibody and were completely resistant to arthritis. Therefore, these results strongly suggest that HLA-DQ8 molecules contribute to genetic susceptibility to arthritis and also establish a novel animal model for the study of human arthritis.

Cellular immunity to cartilage link protein in patients with inflammatory arthritis and non-arthritic controls

OBJECTIVES

To determine if increased T cell responses to articular cartilage link protein have any correlation with rheumatoid arthritis (RA), and if RA patients with increased responses to link protein also respond to a 17 amino acid peptide covering the 'arthritogenic' epitope in mycobacterial hsp65 which is homologous with link protein.

METHODS

The reactivity of T cells from both peripheral blood and synovial fluid, to highly purified human cartilage link protein, hsp65, the 17 amino acid peptide, and bovine type II collagen was determined in patients with RA and nonarthritic controls, by measuring the rate of mononuclear cell proliferation in the presence and absence of antigen.

RESULTS

Using peripheral blood mononuclear cells (PBMC), significant reactivity (stimulation index (SI) > 1.5) to link protein was found in 12 of 46 RA patients (26%), but in only four of 44 controls (9%). A greater proportion of RA patients (eight of 17:47%) were reactive to link protein when mononuclear cells from synovial fluid were tested. SI values, however, were generally low (0.5-3.1) and only one patient showed a PBMC response above a reference range of values calculated from the logarithmic values of the normal control population. No reactivity was observed against a 17 amino acid synthetic peptide including the arthritogenic epitope from the mycobacterial hsp65 to which T cell clones isolated from rats in the adjuvant arthritis model react. However, eight of nine RA patients and all of seven controls reacted to the intact hsp65.

CONCLUSION

It remains unclear if T cell responses to link protein are involved in the pathogenesis of RA, but it is unlikely that T cells specific for the sequence homologous with the arthritogenic epitope in hsp65 are present in RA patients.

Mapping of arthritogenic/autoimmune epitopes of cartilage aggrecans in proteoglycan-induced arthritis

Immunization of BALB/c mice with chondroitin sulfate-depleted proteoglycan (aggrecan) of fetal human cartilage produces progressive polyarthritis and ankylosing spondylitis. The development of the disease in genetically susceptible BALB/c mice is dependent upon the expression of both cell-mediated and humoral immune responses against the host mouse cartilage proteoglycan (PG). Although cartilage PGs from various species have many biochemical and immunological similarities, only a select group of PGs from fetal and newborn human, fetal pig and canine articular cartilages, human osteophytes and human chondrosarcomas are able to induce arthritis in BALB/c mice. Arthritis develops only in mice that also develop autoantibodies to self-cartilage PGs, although autoantibodies occasionally are present in non-arthritic animals as well. The protease-sensitive auto/arthritogenic epitope(s) is located in, or close to, the chondroitin sulfate (CS) attachment region of the PG molecule. The primary structure of the core protein is responsible for the autoimmune/arthritogenic effect of this select group of PGs, whereas the core protein epitopes are masked by glycosaminoglycan (GAG)-side chains. The CS side chains seem to inhibit antigen recognition in all aggrecans with arthritogenic potential, whereas a similar effect with keratan sulfate (KS) appears only in PGs of aging cartilages.

High-density proteoglycan induces specific suppression of adjuvant-induced arthritis in rats

In vitro data support the view that T cells in adjuvant-induced arthritis (AIA) respond to the proteoglycan (PG) component of articular cartilage; however, an in vivo role for PG in AIA has yet to be shown. To do so, we examined the effects of pretreatment with bovine cartilage high density PG (HDPG) on AIA induced by heat-killed Mycobacterium butyricum in Lewis rats. Purified bovine cartilage HDPG emulsified in Freund's incomplete adjuvant (FIA) was injected intradermally into rats 7 days before challenge with Myco. butyricum. The severity of arthritis was significantly suppressed in rats pretreated with as little as 0.75 mg of HDPG, and the arthritis was completely suppressed in rats pretreated with 3.0 mg of HDPG. This suppression was specific, as the same treatment did not protect against type II collagen-induced arthritis. Suppression of AIA is primarily a property of the HDPG, as suppression of the arthritis was significantly less with pretreatment with 3.0 mg of middle density fractions of PG, and no suppression was observed with pretreatment with the lowest density fraction of PG. Thus we report that pretreatment with cartilage HDPG, but not lower density PG, can induce specific suppression of AIA. These in vivo results support the view that immunity to cartilage HDPG plays a major role in the pathogenesis of AIA, and can induce specific tolerance to this type of arthritis.

Cellular responses to human chondrocytes: absence of allogeneic responses in the presence of HLA-DR and ICAM-1

To assess the accessory cell function of human articular chondrocytes, we assessed the ability of human chondrocytes to stimulate allogeneic peripheral blood mononuclear cells (PBMC) and to support phytohaemagglutinin (PHA)-induced proliferation of highly purified T cells. We also examined the surface expression of HLA-DR and ICAM-1 on the chondrocytes both unstimulated and stimulated with cytokines in vitro. Chondrocytes failed to stimulate allogeneic PBMC despite the constitutive expression of MHC class I molecules and the cytokine-induced expression of class II molecules but were able to support T cell proliferation to PHA, IFN-gamma and to a limited extent, IL-1 beta, induced class II expression on chondrocytes. ICAM-1 was present on 94-99% of freshly isolated cells; this declined with culture (17-59%; P < 0.005) but was readily induced by IFN-gamma, IL-1 beta, and tumour necrosis factor-alpha. Alloreactivity and, presumably, autoreactivity to chondrocytes requires factors in addition to the surface expression of DR and ICAM-1. However the presence of these molecules suggests a capacity for cell-cell interactions in inflammatory sites such as the cartilage pannus junction.

Dependence of proteoglycan induced arthritis in BALB/c mice on the development of autoantibodies to high density proteoglycans

BALB/c mice were immunised with high or low density native human cartilage proteoglycans, or the respective core proteins obtained from chondroitin ABC lyase digestion. Mice injected with high density native proteoglycans developed arthritis whereas mice injected with low density proteoglycans or with core proteins did not. Analysis of the immune response by enzyme linked immunosorbent assay (ELISA) and western blot showed a stronger and more polyspecific response in animals injected with low density proteoglycans compared with mice with arthritis which had been injected with high density proteoglycans. Autoantibodies to mouse high density proteoglycans were only present in mice injected with native human high density proteoglycans, however. The data suggest that an arthritogenic epitope lies within the glycosaminoglycan rich region of the native proteoglycan molecule, which may induce an autoantibody response and subsequently arthritis in BALB/c mice.

Cellular immunity to cartilage proteoglycans: relevance to the pathogenesis of ankylosing spondylitis

Cellular immunity to cartilage proteoglycans may be responsible for sustaining chronic inflammation in ankylosing spondylitis. This hypothesis was examined by measuring peripheral blood and synovial fluid mononuclear cell proliferation in five preparations of human cartilage proteoglycan monomer in vitro. Peripheral blood mononuclear cells from 25 patients and synovial fluid mononuclear cells from five patients were compared with those from normal and disease control subjects matched for age. No significant differences were found between the three groups. This suggests that autoimmune responses to cartilage proteoglycans are unlikely to play a significant part in the pathogenesis of ankylosing spondylitis.

A proteolytic fragment from human link protein is taken up and processed by monocytes and B cells

Mild digestion of 125I-labelled human proteoglycan aggregates with trypsin or stromelysin produced specific peptides that were taken up rapidly by THP-1 monocytes. SDS/PAGE of undigested aggregate showed that the three components of molecular mass 48, 44 and 41 kDa, corresponding to isoforms of link protein originally present, had been converted into a single component of 41 kDa by trypsin treatment, and that fragments of 6-12 kDa were present in fractions containing the high-uptake peptide. Separate proteolysis of isolated proteoglycan monomer and link protein confirmed that the specific high-uptake fragment was derived from link protein. Uptake of the link fragment was rapid, reaching a maximum after 5 min, and specific, since it was blocked by metabolic or serine proteinase inhibitors and at 4 degrees C. After uptake the cleaved fragment was processed further, with 50% of the radiolabel being released as degraded peptides within 5 min. In contrast, accumulation of whole aggregate reached a maximum after 45 min and only 50% had been released after 2 h. Uptake of aggregate was less affected by inhibitors or at low temperature, suggesting that a separate mechanism existed for its turnover. The aggregate was transported to lysosomes after uptake, although the link fragment did not sediment with either lysosomes or plasma membranes, suggesting that it was present in the cytoplasm or in very labile vesicles. However, the mode of handling of the peptide by the cells remains unclear. The link fragment was taken up by several different monocytic and B cell lines, but not by mouse fibroblasts or peritoneal macrophages. These data suggest that a surface serine proteinase on monocytes and B cells enables them to process and take up a fragment of link protein derived by extracellular proteolysis.

Comparative analysis of murine T lymphocyte responses to cartilage proteoglycans

Cartilage proteoglycans are large molecules consisting of several sub-regions each of which comprises homologous repeating subunits. Comparisons of murine primed popliteal lymph node responses to human cartilage proteoglycans in BALB and B10 congenic mice showed that the major histocompatibility complex (MHC) influences T cell responsiveness to this antigen. H-2k and H-2d were higher responders than H-2b. Responses were MHC class II-restricted, and human cartilage proteoglycans were cross-reactive with mouse cartilage proteoglycans for a BALB/c T cell line. The proportion of proteoglycan-specific T lymphocytes in BALB/c primed popliteal lymph nodes was about 45% lower in females than males. These results show that in mice both MHC haplotype and sex can determine T lymphocyte responsiveness to cartilage proteoglycans. If the same mechanisms apply in humans they could be important in determining the HLA-DR haplotype associations and the predilection of rheumatoid arthritis for females.

Cartilage proteoglycan-induced arthritis in BALB/c mice. Antibodies that recognize human and mouse cartilage proteoglycan and can cause depletion of cartilage proteoglycan with little or no synovitis

Human fetal cartilage proteoglycan (PG) induces the development of an erosive polyarthritis and spondylitis in BALB/c mice. We have examined the properties of 3 monoclonal antibodies (MAb) to human fetal cartilage PG isolated from immunized mice that cross-react with mouse cartilage PG. Compared with sera from arthritic mice, which contain antibodies reactive with keratan sulfate, MAb 202 (IgG1) reacted only with a protein-related epitope that is distributed on both hyaluronic acid-binding and chondroitin sulfate-attachment regions. MAb 813 (IgG1) reacted with the same fragments and recognized an epitope with the immunologic characteristics of keratan sulfate. MAb 945 (IgM) remains to be further characterized. Introduction of hybridomas secreting MAb 202 and MAb 945 into irradiated mice resulted in the loss of PG from articular cartilage and from growth plate cartilage (with MAb 202 only), as revealed by a loss of staining with toluidine blue. There was no synovial hyperplasia with MAb 202, but some hyperplasia and mononuclear cell infiltration was seen with MAb 945. This was accompanied by the binding of immunoglobulins to articular cartilage, as demonstrated by immunofluorescence. The hybridoma secreting MAb 813 produced no cartilage changes or synovitis, and there was no immunoglobulin binding to cartilage. Polymorphonuclear leukocyte infiltration was never observed with these antibodies. These studies indicate that MAb reactive with mouse cartilage PG can cause the depletion of PG from hyaline cartilage by mechanisms that may be both complement dependent and complement independent. Antibodies may serve to release and expose PG antigen to immune cells, as well as causing a loss of the mechanical properties of cartilage that are PG dependent.

Proteoglycan-induced polyarthritis and spondylitis adoptively transferred to naive (nonimmunized) BALB/c mice

Mononuclear cells from BALB/c mice with progressive polyarthritis and spondylitis induced by injection of fetal human articular cartilage proteoglycan (PG) were used to transfer arthritis by intravenous injection into irradiated, nonimmunized syngeneic mice. Successful transfer of arthritis to BALB/c mice required the injection of lymphocytes from mice with arthritis, along with 50 micrograms of human fetal PG, or lymphocytes stimulated in vitro with either fetal human PG or with mouse cartilage PG. In addition, interleukin-2 or immune sera from animals with arthritis significantly reduced the time to onset of transferred disease. The onset of adoptively transferred arthritis, using cells and antigen, from the time of the first injection (38.2 +/- 18.2 days, mean +/- SD) was shortened if lymphocytes from mice with transferred arthritis were reinjected (retransferred) into other, irradiated syngeneic mice (6.1 +/- 2.6 days). The appearance of autoreactive antibodies to mouse cartilage PG in the sera of mice with adoptively transferred arthritis (secondary or tertiary) preceded the appearance of the first clinical symptoms by a few days. The transfer of arthritis was blocked by pretreatment of donor (arthritic) lymphocytes with either anti-T cell or anti-B cell antibodies and complement. Exposure of mononuclear cells from mice with arthritis to PG, and its removal prior to transfer, also resulted in transfer of the arthritis. PG-induced arthritis was not transferred to nonirradiated mice, nor to irradiated mice injected with lymphocytes from animals with primary arthritis without chondroitinase ABC-digested fetal human PG. Arthritis never developed after injection of immune sera from mice with arthritis (without cells), nor when cells of nonarthritic animals were used with chondroitinase ABC-digested fetal human PG, with or without interleukin-2.

Lymphocyte extracellular matrix interactions. Induction of interferon by connective tissue components

Limiting dilution analysis was performed in the presence of interleukin 2 (IL-2) on lymphocytes isolated from the synovial fluid (SF) and peripheral blood (PB) of patients with rheumatoid arthritis (RA) and PB of normal donors. Clones of these 'spontaneously' IL-2-responsive cells from PB and SF were compared for their reactivity with components of the extracellular matrix (i.e. native or denatured type I or type II collagen and proteoglycan). It was determined that all clones from both PB and SF were activated to produce interferon (IFN) in the presence of any of the connective tissue components (CTC). Clones derived from normal PB behaved in a similar fashion but produced lower IFN-gamma levels. There was a synergy between the CTC and serum or plasma fibronectin, which was more apparent when soluble CTC were used as the stimuli rather than immobilized CTC. The fibronectin alone was unable to induce IFN-gamma production under any of the conditions tested (i.e. soluble or immobilized). These results demonstrate that clones of IL-2-responsive T cells can be activated by interactions with connective tissue components to produce IFN-gamma.

Interleukin 2 responsive T cell clones from rheumatoid and normal subjects: proliferative responses to connective tissue elements

In vivo-activated interleukin 2 responsive T cell clones were generated from peripheral blood (PB) and synovial fluid (SF) of rheumatoid arthritis (RA) patients and from normal control PB. The specificity of these clones was assessed by measuring proliferation induced by the connective tissue elements (CTE) collagen types I and II, native and denatured, proteoglycans, and irrelevant control antigens. The cloned T cells from RA patients but not from normal subjects responded in vitro with proliferation to all CTE but not to control antigens purified protein derivative, ovalbumin, or lysozyme. Proliferation occurred in the presence and absence of accessory cells (AC), but the responses were consistently higher in the presence of AC. Antibodies to HLA-DR abrogated the proliferative response to CTE suggesting that DR expression was necessary for the induction of proliferation. These findings demonstrate the existence of clonable T cells responsive to CTE in PB and SF of RA patients. Expression of reactivity to CTE may contribute to the chronicity of the inflammation in RA.

Enzyme-linked immunosorbent assay for hyaluronate using cartilage proteoglycan and an antibody to keratan sulfate

An enzyme-linked immunosorbent assay has been developed to measure hyaluronate concentrations in biological samples. The assay is based on the aggregation of hyaluronate with cartilage proteoglycan monomer, followed by binding of a monoclonal antibody to the keratan sulfate on the proteoglycan. The sensitivity of the assay is 10 ng hyaluronate/ml of either serum or conditioned cell culture medium. The coefficient of variability was between 10 and 20%. Hyaluronate added to samples was recovered quantitatively and digestion of cell culture medium with protease did not affect the concentration of hyaluronate. Hyaluronate polysaccharides as small as a decasaccharide can be measured. This sensitive and convenient assay can be used for measuring large numbers of biological samples from a variety of animal and tissue sources.

Humoral immunity to link protein in patients with inflammatory joint disease, osteoarthritis, and in non-arthritic controls

Cartilage link protein of high purity was prepared and used in an enzyme linked immunosorbent assay (ELISA). Antibodies to link protein were sought in the sera of 98 patients with rheumatic disorders; 38 with rheumatoid arthritis (RA), 29 with osteoarthritis (OA), 13 with psoriatic arthritis (PA), nine with ankylosing spondylitis (AS), nine with systemic lupus erythematosus (SLE), and in 83 healthy controls. Antibodies were detected in all groups with the following prevalences: 21/83 normals, 9/38 RA, 7/29 OA, 7/13 PA, 3/9 AS, and 4/9 SLE. No statistically significant differences existed between the groups with regard to either prevalence or mean titre of anti-link antibodies. Serum antibodies to proteoglycan link protein appear to be no more common in patients with rheumatic disorders than in healthy controls.

Isolation of proteoglycan-specific T lymphocytes from patients with ankylosing spondylitis

Three T-cell lines and clones of the OKT4 phenotype have been isolated from the peripheral blood of three patients with ankylosing spondylitis. Antigen specificities of T cells were determined with purified protein derivative-(PPD) and cartilage-derived antigens, namely proteoglycans from human articular cartilage and intervertebral disc, bovine nasal cartilage, and rat chondrosarcoma and human type II collagen from cartilage. A cell line from one patient reacted with proteoglycans from human articular cartilage and human intervertebral disc, but the other two cell lines (each from a different patient) and four clones from one of the latter two lines proved to be highly specific for the human articular cartilage proteoglycan. From a study of four proteoglycan specific clones isolated from one patient, it is clear that removal of chondroitin sulfate had no effect on immunoreactivity but digestion of proteoglycan with pronase or alkali/sodium borohydride treatment abolished all reactivity. A OKT4-positive T-cell clone isolated from a healthy adult which was reactive to PPD was used to compare the antigen specificity of cells: this clone showed no reactivity to any of the other putative antigens listed above.

Evidence that a humoral immune response to autologous cartilage proteoglycan can participate in the induction of cartilage pathology

We examined antiproteoglycan antibodies as an autoimmune response for induction of synovitis. This hypothesis was studied by monitoring humoral antiproteoglycan antibody following IgG induction of experimental immune synovitis, localization in the articular cartilage of an autologous immune response, and loss of proteoglycan from cartilage following intravenous administration of antiproteoglycan monoclonal antibodies. The data support the hypothesis that autoimmunity to cartilage macromolecules may play a role in the etiopathogenesis of arthritis.

Ultrastructural changes in scleral proteoglycans precede destruction of the collagen fibril matrix in necrotizing scleritis

Proteoglycans were localized using cuprolinic blue and visualized by electron microscopy in sclera from the eye of a 58-year-old woman with necrotizing scleritis. Samples of sclera taken from clinically normal anterior and posterior sclera and from sites involved in anterior necrotizing scleritis with ulceration and posterior scleritis showed markedly different patterns of proteoglycan distribution. Proteoglycans appeared as stain-positive filaments associated regularly with the 'd/e' bands of the collagen fibrils in normal anterior sclera, but were absent from intact collagen fibrils in the zone of anterior ulcerative scleritis. In posterior sclera including a site of posterior scleritis, proteoglycans were reduced in number with more heterogeneous associations with collagen fibrils. Depletion of scleral proteoglycans therefore precedes the degradation of scleral collagen in necrotizing scleritis, which may also involve the synthesis of atypical proteoglycan types in pathological sclera.

Intact proteoglycan is a polyclonal activator of murine B-lymphocytes

Destruction of articular cartilage is the hallmark of both inflammatory and degenerative arthritides. Since degradation of cartilage results in the release of proteoglycan (PG) monomers and fragments into the synovial fluid, the present study was initiated to determine whether hyaline cartilage PG can induce a cellular immune response. Nonimmune spleen cells obtained from A/St and C57Bl/6 mice were cultured with a crude extract of bovine nasal cartilage, a PG aggregate fraction, PG monomer or degraded PG monomer for varying time periods. Only intact PG monomer induced a proliferative response which peaked at day 2. The responding cell was a B-lymphocyte which did not require T-helper cell activity. Our results suggest that intact PG monomer is a polyclonal activator of B-lymphocytes.

Immunity to cartilage proteoglycans in BALB/c mice with progressive polyarthritis and ankylosing spondylitis induced by injection of human cartilage proteoglycan

Intraperitoneal injection of human fetal cartilage proteoglycan (depleted of chondroitin sulfate) in Freund's complete or incomplete adjuvant induces a chronic erosive polyarthritis and spondylitis in all female BALB/c mice. This occurrence is strain-specific but not haplotype-specific, and it is sex-related. The development of the arthritis is associated with the natural presence of cellular immunity to the immunizing antigen and to chondroitinase ABC-treated mouse cartilage proteoglycan. In addition, relatively more antibody to the immunizing proteoglycan is elicited in arthritic mice, and antibodies are produced that cross-react with native mouse proteoglycan. This combination of immune responses is not observed in mice that do not develop arthritis. Associated with the arthritis is the development of cytotoxicity to mouse chondrocytes and, in some animals, of rheumatoid factor, immune deposits in joint tissues and kidneys, and the production of autoantibodies to mouse type II collagen. These observations might be related to our earlier demonstration that immunity to human cartilage proteoglycan is observed in some patients with ankylosing spondylitis.

Proteoglycan-induced arthritis in BALB/c mice. Clinical features and histopathology

Immunization with chondroitinase ABC-digested fetal human cartilage proteoglycan and Freund's complete adjuvant induced polyarthritis and ankylosing spondylitis in female BALB/c mice. The initial external symptoms of the joint inflammation were swelling and redness. This was associated with edema of the synovium and periarticular tissues and gross proliferation of cells, which reached a peak during weeks 7-9 of the experiment. Mononuclear cell infiltration, with perivascular concentration and occlusion of small vessels, was common. Synovitis increased in severity, villous pannus developed, and erosions of bone, articular cartilage, and occasionally, growth plate were observed. The lumbar spine and the proximal intervertebral discs of the tail also exhibited inflammatory and degenerative changes. As the arthritis progressed, sometimes with acute inflammatory exacerbations, more joints became involved and, by the sixteenth to the twentieth weeks of the experiment, a progressive polyarthritis, with gross joint deformities and restricted function, developed in the majority of the limb joints. Clinical and morphologic features of the disease correlated well with radiologic analysis and with an increased deposition of 99mTc-methylene diphosphonate (determined by radionuclide imaging). The development of this arthritis was accompanied by the expression of cell-mediated and humoral immunity to the immunizing antigen. However, this immunity was also observed, although it was generally less well developed, in mice that received the intact or digested proteoglycan without adjuvant. These mice did not usually develop arthritis. Control mice that received only adjuvant did not develop arthritis.

Epitopes of proteoglycans eliciting an anti-proteoglycan response in chronic immune synovitis

This study details the immune response to cartilage proteoglycan in experimental chronic IgG-induced immune synovitis. Antibodies reactive with purified rabbit proteoglycan monomer were observed in nine of nine rabbits with immune synovitis. IgG-immunized but nonsynovitic control animals with no pathology showed no antibody response. A panel of murine monoclonal antibodies with defined specificity towards rabbit proteoglycan were utilized to characterize the epitope specificity of the immune synovitis polyclonal anti-proteoglycan response. One murine monoclonal antibody, 6C11, inhibited the binding of the polyclonal antisera to proteoglycan in all nine animals with significant (greater than 40%) inhibition in six of nine rabbits. Further inhibition studies utilizing DEAE-cellulose-resolved proteoglycan tryptic peptides revealed that peptides poor in chondroitin sulfate were strong inhibitors of binding of the polyclonal antibodies to the proteoglycan substrate. In particular, keratan sulfate-containing tryptic peptides were most inhibitory on a per weight basis. These results indicate that, in chronic IgG-induced immune synovitis, anti-proteoglycan antibodies elicited are heterogeneous with regard to specificity, but a relatively large proportion predominantly recognized a portion of the proteoglycan molecule containing core protein and associated keratan sulfate.