Polymorphonuclear leukocyte adhesion to articular cartilage is inhibited by cartilage surface macromolecules.

The present studies deal with polymorphonuclear neutrophil (PMN) adhesion inhibitory properties of cartilage surface proteoglycans. Normal human PMN were used in adhesion experiments with bovine cartilage surfaces exposed to neutrophil elastase and reconstituted with fibronectin (Fn) or on plastic-bound Fn. An extract of cartilage surface small proteoglycans (SE) and purified fibromodulin (FM), decorin (DCN), biglycan (BGN), and aggrecan (AGN) on the surface of normal cartilage were used to test for inhibition of Fn-dependent cell adhesion. The PMN did not adhere to intact articular cartilage surfaces, whereas significant adhesion was measured using cartilage explants digested with elastase and reconstituted with Fn. Incubation of elastase-treated, Fn-reconstituted cartilage with 45 microg/ml SE inhibited PMN adhesion by 50.7 +/- 5.8% (P < 0.0001). Addition of 50 microg/ml purified FM to the reconstituted articular surfaces inhibited cell adhesion by 71.2 +/- 13.9% (P < 0.0001). Inhibition of PMN adhesion to plastic-bound Fn was seen with 1.7 microg/ml SE (20.4 +/- 8.0%). Maximal inhibition of 67.4 +/- 14.8% (P < 0.01) was obtained with 17.0 microg/ml SE. With FM, concentrations of 4.3 microg/ml resulted in 34.7 25.2 inhibition (P < 0.001), and maximal inhibition of 66.3 16.2% (P < 0.01) was obtained with 43.0 microg/ml. Similar results were obtained with purified bovine DCN and BGN. The main component of cartilage matrix, AGN, failed to inhibit cell adhesion significantly. The results indicate that macromolecules normally present on articular cartilage surfaces act as a barrier to PMN adhesion. Since cartilage surface proteins are susceptible to breakdown by proteases from synovial fluid inflammatory cells, we postulate that the degradation of this barrier may be responsible for increasing PMN adhesion and subsequent cartilage damage in inflammatory arthritis.

Inflammatory properties of IgG modified by oxygen radicals and peroxynitrite.

In inflammatory arthritis, there is evidence indicating that the affected tissues produce large amounts of oxygen-free radicals and NO. Herein, we examine the biologic effects of exposure of IgG to hypochlorous acid (HOCl) and peroxynitrite (ONOO). The concentrations of IgG modified by chlorination and nitrosation were measured in synovial fluids from inflammatory and noninflammatory arthritis. Human IgG was exposed to increasing concentrations of HOCl and ONOO, and the resulting products were tested for complement component binding; binding to FcgammaRI; activation of polymorphonuclear neutrophils; effect on the Ab-combining site of Abs; and in vivo inflammatory activity in a rabbit model of acute arthritis. Rheumatoid synovial fluids contained significantly greater concentrations of nitrosated and chlorinated IgG compared with ostearthritic specimens. In vitro exposure of human IgG to HOCl and ONOO resulted in a concentration-dependent decrease in C3 and C1q fixation. The decrease in Fc domain-dependent biologic functions was confirmed by competitive binding studies to the FcgammaRI of U937 cells. HOCl-treated IgG monomer was 10 times less effective in competing for binding compared with native IgG, and ONOO-treated IgG was 2.5 times less effective. The modified IgGs were also ineffective in inducing synthesis of H(2)O(2) by human PMN. The Ag-binding domains of IgG also showed a concentration-dependent decrease in binding to Ag. The ability of the modified IgGs to induce acute inflammation in rabbit knees decreased 20-fold as gauged by the intensity of the inflammatory cell exudates. These studies clarify the modulating role of biological oxidants in inflammatory processes in which Ag-autoantibody reactions and immune complex pathogenesis may play an important role.

Macromolecular transport across the superficial layer of articular cartilage.

OBJECTIVE: To study the role of the superficial layer of articular cartilage in the transport of macromolecular solutes. DESIGN: The articular cartilage of intact bovine carpal bones was incubated with(125)I-labeled bovine serum albumin, human IgG, or horse ferritin for 4 hours. Quadruplicate samples were first incubated with polymorphonuclear neutrophil elastase for 30 minutes to remove the outermost layer covering the articular surface. The rates of exchange of each macromolecule from excised tissue explants in the absence of a concentration gradient were measured at six different time points. The results were expressed as the fraction of radioactive protein exiting the cartilage per mm(2)of tissue, or as picomoles of labeled solute per mm(2). RESULTS: Exchange rates correlated well with molecular mass, and no apparent differences were detected between intact and elastase-treated tissues. However, when the results were expressed in terms of the total number of molecules within the tissue, it was apparent that IgG molecules accumulated in the intact cartilage in larger than expected numbers. This finding was not observed in experiments using elastase-treated tissue. CONCLUSION: These observations suggest that the outermost surface layer does not constitute a barrier to the transport of macromolecules into the deeper zones of the tissue. The higher IgG accumulation observed in intact cartilage suggests that the acidic outer layer of cartilage exhibited attractive interactions, probably ionic in nature, with the cationic fraction of IgG. These observations may relate to our previous work demonstrating that the sequestered immune complexes in the superficial zone of articular cartilage in rheumatoid arthritis, and in the antigen-induced arthritis model, are formed because pre-existing antibody normally present in cartilage irreversibly traps antigen within the tissue.

Systemic lupus erythematosus and guillain-barre syndrome.

Guillain-Barre syndrome (GBS) has been rarely reported as a presenting manifestation of systemic lupus erythematosus (SLE). We describe a 23-year-old patient with SLE who presented with GBS. He developed progressive ascending motor paralysis but also had pulmonary disease, proteinuria, and hypoalbuminemia. Serologic studies revealed ANA and antibodies to double stranded-DNA. Electromyographic and nerve conduction studies were suggestive of acute inflammatory demyelinating polyneuropathy, and a kidney biopsy specimen showed membranous glomerulonephritis. Treatment with corticosteroids and plasmapheresis resulted in clinical improvement of the neurological illness. The renal disease progressed despite prednisone and was eventually treated with hemodialysis. The neurologic disease did not recur.Auto-antibodies reactive with neural tissue and immunological crossreactivity between auto-antibodies in SLE and neural tissue antigens may occur. Complement fixing antibodies to nerve and kidney in patients with GBS and membranous glomerulonephritis, point to a common pathogenic relationship. Such antibodies may be seen in patients with SLE. The increased susceptibility to infection caused by immunosuppression may make patients with SLE susceptible to GBS. Plasmapheresis and corticosteroids, as in our patient, plus i.v. immunoglobulins and cyclophosphamide has brought good results in patients who have SLE with GBS.

Characterization of the macromolecular components of the articular cartilage surface.

The intact surface of articular cartilage is a highly organized structure composed of a variety of macromolecules. The studies reported here deal with a partial characterization of the non-covalently bound components of the outermost layer of articular cartilage. Normal bovine and human cartilage articular surfaces were extracted for 5 min with 4-M guanidine HCl solution. Analysis and quantitation of small proteoglycans in the extract were carried out by PAGE (polyacrylamide gel electrophoresis), Western blot, and radioimmunoassays. The present studies indicate that the major proteins extracted from the articular surface of bovine and human cartilage are the collagen-binding small proteoglycans designated as fibromodulin and albumin. Fibronectin, decorin, and biglycan were also detected in smaller amounts. Immunoblotting of the surface material developed with a monoclonal antibody with keratan sulfate specificity confirmed the presence of fibromodulin coinciding with the major protein band of approximately 70-100-kDa molecular mass. Gel filtration chromatography of the surface material confirmed the previous results. Additional in vitro assays showed that the collagen-binding material extracted from the cartilage surface contained the small proteoglycans. Anti-human fibromodulin antibodies bound in significantly greater amounts to the intact articular surfaces than to cut surfaces of normal human cartilage. It is concluded that small, non-aggregating proteoglycans constitute the major proteoglycan species non-covalently bound to macromolecules at the articular surface of cartilage partially responsible for the interference of anti-collagen type II antibody binding and for the inhibition of cell adhesion to the intact surface.

Covalent cross-linking of immune complexes by oxygen radicals and nitrite.

We have shown that polymorphonuclear neutrophils mediate the covalent cross-linking of immune complexes (ICs) using H2O2 and myeloperoxidase (MPO). Moreover, activated superficial chondrocytes produce large amounts of nitric oxide (NO), suggesting that high concentrations of these radicals may interact at the cartilage surface in rheumatoid arthritis. We describe the effects of the interaction of NO and its decay product, NO2, with H2O2 and MPO on IC cross-linking. Cross-linking was measured by resistance to the guanidine extraction of plastic-bound ICs. The combination of H2O2, MPO, and NO in the absence of O2 did not alter the magnitude of cross-linking. The addition of O2 resulted in a significant enhancement of cross-linking (p < 0.004), suggesting that nitrite was responsible for the increase observed. Indeed, NaNO2 greatly increased H2O2-dependent cross-linking (control: 29.2+/-3.8; 1 mM NaNO2: 58.4+/-9.9; 10 mM: 60.4+/-4.2% cross-linking,p < 0.0002). Sodium azide, which is an inhibitor of MPO, completely inhibited cross-linking. These results indicated that the product of interaction of H2O2 and NO2 mediated by MPO may be responsible for the increase in cross-linking. The generation of nitrotyrosine was demonstrated when NO2 was added to the cross-linking system. Cross-linking was also shown with an O2--generating system and NO. Peroxynitrite alone mediated cross-linking (100 microM ONOO-: 40.3+/-1.9% cross-linking; p < 0.002), and the addition of MPO significantly enhanced this effect (100 microM: 57.7+/-6.0%; p < 0.0002 with respect to no nitrite control). Oxygen radicals and NO are likely to interact at the cartilage surface in inflammatory arthritis, resulting in an increase in oxidative damage within the joint cavity.

Juvenile rheumatoid arthritis and common variable hypogammaglobulinemia.

We describe a 9-year-old white boy with systemic juvenile rheumatoid arthritis (JRA) who developed pancytopenia and hypersplenism at the age of 13 years. He underwent splenectomy and 3 years later he developed Coombs' positive hemolytic anemia, alopecia, juvenile warts, and multiple bacterial infections. At that time, investigations were compatible with severe hypogammaglobulinemia associated with common variable immunodeficiency. Concomitantly with this condition he experienced complete remission of his inflammatory arthritis. Immunologic studies of B and T lymphocyte function showed that the number of circulating T and B lymphocytes were normal, while T cell function was depressed, as evidenced by markedly reduced proliferative responses to mitogens and antigens, and ability to mediate B cell help. In addition, his circulating B cells were unable to secrete IgM or IgG. He also exhibited anergy to intradermal challenge with a battery of common antigens. The literature dealing with this clinical association is reviewed, and possible immunologic mechanisms involved are discussed.

Isolation and differential secretion of metalloproteinase by superficial chondrocytes in articular cartilage.

Chondrocytes from superficial layers of articular cartilage have distinct phenotypic properties which are different from those of cells obtained from the deeper areas. We describe a method that isolates highly purified articular cartilage chondrocytes from the superficial layers. When the superficial cells are stimulated in vitro with a source of cytokines, they secrete greater amounts of metalloproteinase compared to chondrocytes obtained from a deeper area.

Nitric oxide production by superficial and deep articular chondrocytes.

OBJECTIVE: Chondrocytes have been shown to produce large amounts of nitric oxide (NO) when appropriately stimulated with proinflammatory cytokines or bacterial lipopolysaccharide (LPS). In view of recent observations underscoring profound phenotypic differences between superficial and deep articular chondrocytes, these studies investigated NO production, inducible NO synthase (iNOS) activity, and messenger RNA (mRNA) expression of superficial and deep cartilage explants and cells. METHODS: Superficial and deep bovine and human articular cartilage explants and isolated bovine chondrocytes were cultured in the presence of stimulating cytokines or LPS. NO was measured by the Griess reagent. Inducible NOS activity was quantitated by conversion of L-14 C-arginine to L-14C-citrulline. Inducible NOS mRNA expression was quantitated by reverse transcription-polymerase chain reaction (RT-PCR) and in situ hybridization. RESULTS: Superficial bovine cartilage explants stimulated with interleukin-1 alpha, LPS, or tumor necrosis factor alpha for 24 and 48 hours produced significantly more NO than did deep explants with all stimulants and at both times. Similar results were obtained with stimulated isolated superficial and deep cells. NO synthase activity, measured by the conversion of L-14C-arginine to L-14C-citrulline, paralleled NO production. Comparable results were obtained using explants from a normal human donor. Semiquantitation of iNOS mRNA by RT-PCR showed significantly larger amounts of PCR products in superficial cells and superficial explants. These results were confirmed by in situ hybridization of explants and isolated cells. CONCLUSION: Increased NO production at the cartilage surface-synovial fluid interface may play an important role in the modulation of cartilage damage in inflammatory arthritis.

Vasculitis and lymphoproliferative diseases.

Vasculitis as a complication of lymphoproliferative diseases is relatively uncommon. Cutaneous vasculitis is most commonly necrotizing and leukocytoclastic. Granulomatous vasculitis occurs rarely with lymphoproliferative diseases, and even less commonly T-cell lymphocytic vasculitis with eosinophilia. The most common systemic vasculitis is caused by cryoglobulinemia, due to either lymphocytic lymphoma or Waldenström's macroglobulinemia. Other unusual associations involving systemic vasculitides include polyarteritis nodosa and hairy cell leukemia, Wegener's granulomatosis and Hodgkin's disease, granulomatous angiitis of the central nervous system and lymphoma, temporal arteritis and lymphoma, and Henoch-Schönlein purpura and lymphoma. The vasculitis may predate the diagnosis of the lymphoproliferative disease, and patients with vasculitis should be screened and monitored for lymphoproliferative diseases.

Mixed cryoglobulinemia and vasculitis: a novel pathogenic mechanism.

We describe a patient with lymphoma and mixed cryoglobulinemia who developed vasculitis and membranoproliferative glomerulonephritis concomitant with severe decreases in serum IgG and complement levels. At the time of her demise, her serum cryoglobulin solubility had decreased compared to a specimen obtained when she was asymptomatic. This suggests that a change in antigen-antibody ratio of her cryoglobulin towards immune equivalence, due to decreased available antigen (IgG), was responsible for the development of immune complex vasculitis.

Fibronectin synthesis in superficial and deep layers of normal articular cartilage.

OBJECTIVE: To study the distribution and synthesis of fibronectin (FN) in superficial and deep layers of normal articular cartilage. METHODS: Superficial and deep bovine and human articular cartilage slices were used to extract and quantitate FN by radioimmunoassay. Chondrocytes were also isolated by collagenase digestion for FN extraction and culture. Superficial and deep cartilage explants were cultured with and without stimulation by cytokines. Quantitation of newly synthesized FN was carried out by incubation with 35S-methionine. FN was purified on gelatin-agarose columns and further characterized by polyacrylamide gel electrophoresis. FN messenger RNA (mRNA) was quantitated by Northern blot analysis. RESULTS: Freshly isolated bovine chondrocytes from deep cartilage contained 2.3 +/- 0.2 times more FN than was found in superficial cells (P < 0.025). Deep cartilage explants contained 1.2 times more FN than was found in superficial tissue. Explants obtained from deep cartilage synthesized 2.4 times more FN per cell than did superficial tissues (P < 0.01). FN synthesis as a fraction of total protein synthesis was significantly greater in deep explants (P < 0.01) compared with superficial tissues. Isolated deep chondrocytes in culture synthesized 1.89 +/- 0.33-fold more FN than did superficial cells (P < 0.05). Cytokine-stimulated superficial cartilage explants failed to respond in terms of FN synthesis. FN mRNA quantitation showed no significant differences between superficial and deep populations. CONCLUSION: Since FN plays a major role in cell adhesion to damaged cartilage surfaces, our results suggest that modulation of FN synthesis near the articular surface of cartilage may be one of the factors that impede pannus invasion following an inflammatory insult to the joint.

Inhibition of human fibroblast adhesion by cartilage surface proteoglycans.

OBJECTIVE: Recent studies from our laboratory have identified the nonaggregating, collagen-binding proteoglycans, fibromodulin (FM) and decorin, and fibronectin (Fn) and albumin, noncovalently bound at the articular surface of cartilage. The present studies were designed to investigate the interactions between these cartilage macromolecules and the underlying collagen matrix and their role as a barrier to cell adhesion in intact articular cartilage. METHODS: Cell adhesion studies were carried out with human skin fibroblasts incubated on the articular surface of bovine cartilage explants and on collagen-coated and/or Fn-coated plastic surfaces. Interactions of collagen and Fn with either FM or decorin were studied by radioimmunoassay of the same surfaces, using specific antibodies. RESULTS: The present studies show that 1) Fn is immunologically detectable at the intact articular surface of cartilage; 2) fibroblast adhesion to Fn is inhibited by cartilage surface extract proteins and by purified FM, but not by purified decorin; 3) FM has binding affinity for Fn; 4) FM interferes with the binding of a monoclonal antibody specific for the cell-binding domain of Fn; and 5) FM and decorin inhibit collagen-dependent fibroblast adhesion. CONCLUSION: These results indicate that the small proteoglycans at the normal articular surface may act as a barrier to cell adhesion. Since protective cartilage surface proteins break down readily after the induction of acute arthritis in experimental animals, and in rheumatoid cartilage specimens, it is postulated that proteolytic degradation of the surface proteoglycans may be responsible for increasing cell adhesion to, and subsequent pannus invasion of, articular cartilage in inflammatory arthritis.

Repair characteristics of the articular cartilage surface following acute inflammatory arthritis.

OBJECTIVE: To investigate the repair characteristics of the surface protein in an acute arthritis of short duration. METHODS: Knee arthritis in rabbits was induced by intraarticular (ia) injection of 20 micrograms E. coli endotoxin into one knee of 12 rabbits. The contralateral knee served as control. Four animals were killed 3, 6, and 10 days after ia injection. The articular cartilage surface was probed by quantitation of anticollagen type II (anti-CII) antibody binding. The suprapatellar bursae were washed, and the recovered cells counted. RESULTS: A significant increase in anti-CII antibody binding compared to control was measured 3 days after ia injection, coinciding with evidence of acute arthritis (injected joint: 1.1 x 10(7), control: 4.2 x 10(4) cells/joint; percent increase in anti-CII binding: 36.7 +/- 10.7; p < 0.02). Six days after ia injection, the acute arthritis showed a major decrease in intensity whereas anti-CII binding was still abnormal (injected joint: 1.7 x 10(6) cells/joint; percent increase in anti-CII binding: 24.7 +/- 19.6; p < 0.05). On Day 10, there was minimal evidence of arthritis in 3 of 4 rabbits, and anti-CII antibody binding returned to normal (injected joint: 1.9 x 10(5) cells/joint; percent anti-CII binding: -5.8 +/- 3.9). There was a strong positive correlation between individual synovial fluid cell counts and the percent increase in anti-CII binding (R = 0.72, p < 0.02). CONCLUSION: The magnitude of binding of anti-CII antibodies to the articular cartilage surface constitutes a sensitive probe for the detection of early damage following a transient inflammatory insult. Our studies indicate that after acute injury, the cartilage surface may show evidence of damage for at least 6 days when probed with anti-CII antibodies.

Binding characteristics of antitype II collagen antibody to the surface of diseased human cartilage as a probe for tissue damage.

OBJECTIVE: Studies have shown that collagen type II (CII) on the intact articular surface of cartilage is partially protected from binding to anti-CII antibodies by material proteinaceous in nature, not present in synovial fluid, synthesized by resident chondrocytes, and exquisitely sensitive to polymorphonuclear (PMN) attack and neutrophil elastase digestion. Thus, anti-CII antibody differential binding to articular cartilage surfaces before and after brief neutrophil elastase digestion may be used as a sensitive marker of cartilage damage. METHODS: We measured binding of anti-CII antibodies to the pannus-free articular surfaces of 4 normal, 11 rheumatoid (RA), and 10 osteoarthritic (OA) cartilage specimens, before and after brief digestion with PMN elastase. In addition, antibody binding was quantitated with an antiserum against a 4 M guanidine extract of human cartilage surface. RESULTS: Whereas anti-CII binding increased 59.0% +/- 2.8 after 1 h incubation of normal cartilage with elastase, both the RA and OA specimens failed to show significant increases (RA: 1.0 +/- 0.1; p < 0.001; OA: 27.2% +/- 1.6, p < 0.05). Moreover, anti-CII antibody binding to untreated cartilage specimens was highest for the RA group (Normal: 189.2.1 +/- 38.7 pg anti-lg/mg tissue; RA: 407.5 +/- 80.6, p < 0.05; OA: 243.6 +/- 50.6, NS). Concomitant binding studies with antiserum against cartilage surface material showed greater antibody binding to the articular surfaces than to the cut cartilage surfaces in normal and OA specimens. RA cartilage samples exhibited somewhat smaller antibody binding to the articular surfaces. CONCLUSION: Our studies suggest that in human inflammatory and noninflammatory arthritides the articular cartilage surface undergoes alterations that can be detected by differential binding with anti-CII antibodies, before and after brief digestion with PMN elastase.