The B cell repertoire in rheumatoid arthritis. I. Frequency of EBV-inducible circulating precursors producing autoantibodies.

The frequency of B cell precursors producing antibodies against various autoantigens (Fc fragment of IgG, F(ab')2 fragment of IgG, type II collagen, cytoskeleton filaments and insulin) was determined in patients with rheumatoid arthritis (RA) using immortalization of peripheral blood B cells by the Epstein-Barr virus (EBV) and limiting dilution analysis. Equally large numbers of B cell precursors producing IgM-rheumatoid factors (RFs) were present in the peripheral blood of seronegative and seropositive RA patients and of controls. On average, 1 out of 15,000 B cells could be induced by EBV to secrete IgM-RFs, which represents 0.5-1% of the EBV-induced proliferating clones. By cloning or somatic hetero-hybridization of EBV cell lines derived from patients and controls, we obtained two types of monoclonal RFs: one polyreactive, reacting with Fc but also with the other autoantigens tested, and the other monoreactive, reacting with Fc only and that previously had only been found in the RA B cell repertoire. Moreover, patients and controls had similar numbers of circulating B cell precursors secreting IgM antibodies against other autoantigens that might be regarded as specific targets of RA (F(ab')2 fragment of IgG and type II collagen), and against cytoskeleton filaments that are targets of natural autoantibodies, increased in RA. The frequencies of EBV-induced B cells producing antibodies against all these autoantigens were of the same order of magnitude as the frequency of EBV-induced B cells producing RFs. The patients also possessed a similar number of precursors producing antibodies against insulin, an autoantigen irrelevant to the pathogenesis of the disease, taken as control. These data tend to demonstrate no abnormality in the autoantibody repertoire of B cells activable by EBV in RA, especially those secreting RFs. In vitro spontaneous RF secretion by circulating B cells was observed in seropositive RA patients but not in seronegative patients and in the controls tested. We enumerated the number of B cells spontaneously secreting RFs in seropositive RA patients and found that it correlated with the serum RF titer, but not with the number of RF-secreting B cells activated by EBV. The mean frequency values of B cells secreting RFs either spontaneously or after EBV infection were of the same order of magnitude, showing that the expanded population of in vivo-activated B cells was not (at least partially) infectable by EBV. This raised the possibility that EBV triggers a repertoire which may not reflect the status of B cells secreting autoantibodies in autoimmune diseases.     

Demonstration of anti-idiotypic antibodies directed against IgM rheumatoid factor in the serum of rheumatoid arthritis patients.

We have identified the presence of anti-idiotypic activity against IgMRF in the sera of RA patients. Only patients seropositive for IgMRF had significant levels of anti-idiotypic activity, while seronegative patients and normal volunteers did not. When this anti-idiotypic activity was affinity-purified from a single RA patient, two separate binding activities were identified. IgG antibodies were pepsin-digested to F(ab')2 fragments before affinity-purification to remove the Fc portion capable of binding to IgMRF. Anti-idiotypic F(ab')2 fragments of IgG were eluted from an IgMRF-Sepharose 4B column. These F(ab')2 bound preferentially to IgMRF bearing an idiotype recognized by the anti-idiotypic murine monoclonal 17.109. A second anti-idiotypic F(ab')2 was affinity purified using rabbit anti-human Fc antibody bound to Sepharose 4B. These eluted antibodies behaved as the internal image of IgG, binding five out of seven IgMRF's tested. The binding of both anti-idiotypic F(ab')2 was inhibited with human IgG. The presence of both IgMRF and anti-idiotypic antibodies directed against it in the sera of RA patients suggests that anti-idiotypic antibodies alone are not capable of inhibiting the production of rheumatoid factor.     

The Presence of Anti-Protein A Antibodies in Patients with Rheumatoid Arthritis

Sixteen patients with rheumatoid arthritis (RA) were examined for the presence of anti-protein A antibodies. The F(ab')2 preparations from five RA patients showed significant binding to IgG-free protein A on ELISA. The protein A binding was further examined by immunoblotting. The F(ab')2 preparations of high protein A-binding protein gave a specific reaction with IgG-free protein A on nitrocellulose paper. This demonstrates the presence of anti-protein A antibodies in patients with RA. Those RA patients with anti-protein A antibodies had more active disease as judged by the Lansbury's activity index. The level of serum rheumatoid factor (RAHA) was significantly higher in patients with anti-protein A antibodies than in those without anti-protein A antibodies.     

Rheumatoid factors in immune complexes of patients with rheumatoid arthritis

Conclusions Substantial evidence indicates that humoral immunity through antigen-antibody complex formation contributes to the pathogenesis of RA. Several studies have examined the composition of immune complexes present in the synovial fluid of patients with RA. The constituents of these immune complexes have been immunoglobulins and complement components. Only a few polypeptides in these immune complexes have not been identified. These unidentified components account for only a few percent of total proteins present. These results suggest that rheumatoid factors and their antigens are quantitatively the important ingredients of immune complexes in synovial fluid of patients with RA. The immune deposits in the superficial layers of articular cartilage from patients with RA contain RFs and antibodies to type II collagen.IgG RFs are unique autoantibodies in that they can form immune complexes with pathogenic potential by self-association and without the presence of separate antigen molecules. Self-association becomes possible when the antigenic specificity of IgG RFs is directed to antigenic determinants present on the Fc region of the same molecule. Self-association of IgG RFs is enhanced in a milieu of high concentration of these antibodies and in relative paucity of normal IgG — i. e., conditions that exist in the synovial tissue where IgG RFs are synthesized.Relatively little attention has been devoted to antigenic specificity of RFs in relation to disease. The self-associating IgG RFs illustrate how the antibody specificity and presence of antigenic determinants can contribute to formation of unique immune complexes. One of the major antigenic determinants for IgM RFs is in the C2-C3 domain interface and involves some of the same amino acids that constitute the site for binding SPA. The specificity for self-associating IgG RFs is directed to the same antigenic determinant. Fc binding proteins from Streptococci and the Fc binding proteins induced by herpes type 1 virus bind to the same or overlapping areas of human IgG. The reasons for this similarity of binding to IgG between microbial Fc binding proteins and RFs is not apparent. This relationship, however, may stimulate the synthesis of some RFs by an internal image autoantiidiotype mechanism.     

Rheumatoid factors

Rheumatoid factors (FRs) are auto antibodies which react with different antigenic sites on the Fc piece of human IgG. They appear in large amounts in serum and synovial fluid of patients with rheumatoid arthritis (RA) but they are not specific to this disease. Different methods of detection have been proposed but they induce a low threshold of sensibility and regularly give rise to a low specificity for RA diagnosis. Those auto antibodies have a weak affinity for their antigen. However, despite this weak affinity, they contribute to the formation of large and stable articular immune complexes which could induce inflammatory synovial processes during RA. Occurrence of RFs during various infectious diseases is an argument for a protective role of RFs. This protective role is supported by the findings that RFs structural genes are largely distributed in normal populations and, in part, inherited, and that normal peripheral blood lymphocytes are able to make IgM FR when in vitro polyclonally stimulated.     

Identification of IgG rheumatoid factors by a novel method utilizing immunoblotting.

Monoclonal IgG rheumatoid factors (RFs) are identified using a novel immunoblot assay which detects RFs that bind to SDS-denatured Fc on nitrocellulose. Recognition of these self-associating antibodies is by the use of F(ab')2 fragments of light chain-specific antisera. In this way, IgG RFs can be easily identified and the precise binding characteristics to different isotypes of IgG, or other antigens, further specified. The assay can also be used to detect other classes of RFs such as IgM RFs. Although less sensitive than the standard ELISA, the use of this immunoblot RF assay (IRFA) will identify IgG RFs and their target antigen with precision.     

Human IgG anti-F(ab'')2 antibodies possess rheumatoid factor activity.

Individual preparations of affinity purified anti-F(ab')2 antibodies and anti-Fc antibodies isolated from the sera of patients with rheumatoid arthritis (RA), were examined for reactivity with the Fab and Fc fragments of human IgG. Western blot assays demonstrated specific interaction of affinity-purified anti-Fab antibodies with both Fab and Fc molecules. Approximately one-half of the anti-Fab antibody preparations studied contained IgG antibodies reactive with Fab and Fc fragments in ELISA, suggesting the existence of naturally occurring epibody-like autoantibodies in these patients. Thirteen of 14 affinity-purified anti-Fc antibody preparations contained IgG cross-reactive with Fab molecules in ELISA. Double-adsorption assays on affinity columns demonstrated that a minimum of 14%, and possibly as much as 50%, of the IgG anti-Fab antibodies reacted with the Fc of IgG. Conversely, a minimum of 12%, and possibly as much as 70%, of the IgG anti-Fc antibodies reacted with IgG Fab molecules. Anti-Fab antibodies isolated from non-RA individuals also exhibited anti-Fc reactivity in ELISA, demonstrating the presence of these dual-reactive antibodies in other autoimmune and normal individuals. These studies establish the presence of naturally occurring IgG autoantibodies reactive with both the Fab and Fc fragments of human IgG. Their existence emphasizes the potential of anti-immunoglobulin antibodies to recognize a multiplicity of antigens, possibly including other members of the immunoglobulin supergene family.     

Composition of immune deposits present in glomeruli of NZB/W F1 mice

This paper describes the results of experiments designed to investigate the composition of immune complexes present, in the form of immune deposits, in glomeruli of NZB/NZW F1 mice. Granular deposits of mouse IgG were present along the glomerular capillary walls of 6- to 12-month-old mice. Disappearance of mouse IgG from glomerular deposits, indicating a dissociation of immune complexes, was observed following incubation of kidney sections with an excess of mouse IgG, mouse Fc fragments, rat IgG, and rat Fc fragments, but not with human and rabbit Cohn fraction-II (FII), DNA, nucleohistone, and PBS. Antinuclear antibody activity in mouse sera or in glomerular eluates was removed by absorption with mouse IgG or mouse Fc fragments, rat IgG or rat Fc fragments, DNA, and nucleo-histone, but not by absorption with human or rabbit FII. These results suggest that the IgG antinuclear antibodies present in the sera and in glomerular deposits possess rheumatoid factor (RF) activity. In other experiments, kidney sections were incubated with various concentrations of pepsin, which digests the Fc portion of the IgG. After digestion, the sections were washed and stained for mouse IgG, IgG F(ab')2, and IgG Fc. At concentration of 10 micrograms/ml, pepsin completely removed IgG and IgG Fc, whereas faint IgG F(ab')2 deposits persisted in glomerular deposits. At the concentration of 1 microgram/ml, deposits of mouse IgG, F(ab')2, and Fc persisted, while F(ab')2 was observed bound to nuclei of glomerular cells. At the pepsin concentration of 0.1 microgram/ml or 0.01 microgram/ml, IgG F(ab')2 was bound to the nuclei of glomerular and tubular cells, indicating that the digestion of the Fc portion of IgG had released F(ab')2 with nuclear reactivity from glomerular deposits. The solubilization of mouse IgG from glomerular immune deposits with mouse IgG and the demonstration that pepsin digestion releases mouse F(ab')2 with nuclear reactivity are consistent with the interpretation that the immune deposits present in glomeruli of NZB/NZW F1 mice contain complexes formed by antinuclear IgG and IgG RF. These two antibodies probably cross-react and form multilayer aggregates which contribute to the formation of immune deposits.     

IgA Rheumatoid Factor and IgG Dietary Protein Antibodies are Associated in Rheumatoid Arthritis

This study sought to determine whether patients with rheumatoid arthritis (RA) were immunologically sensitised to dietary protein (DP). Using an enzyme linked immunosorbent assay (ELISA), antibodies to milk and wheat proteins were measured in 93 unselected out-patients with classical or definite RA. Of these 93, 53 had raised levels of IgG antibodies to one or both dietary proteins (DP). In the DP antibody positive group, 48 patients (90%) also had raised levels of IgA rheumatoid factor (measured by ELISA) while only 7 (17%) of the 40 DP antibody negative patients had detectable IgA RF; P<0.02. There was no association between IgM rheumatoid factor and dietary protein antibodies. These results demonstrate that in RA, raised levels of IgA RF are associated with an increased IgG response to antigens which enter the body through the gastrointestinal tract. A breakdown in gastrointestinal tolerance to dietary antigens may play a role in the immunopathogenesis of RA in these patients who might therefore benefit from dietary manipulation.     

Molecular interactions between human IgG, IgM rheumatoid factor and streptococcal IgG Fc receptors

Group A streptococci type M15 were previously shown to bind both human IgG via the Fc component and a purified monoclonal IgM kappa rheumatoid factor (IgM RF). Using 125I-labelled IgG and 125I-labelled IgM RF, the present study gave association constants of 2.2 x 10(7) and 2.9 x 10(8) M-1, respectively. The binding of 125I-IgG to the streptococci was inhibited by unlabelled IgG, IgG Fc and fragment D of staphylococcal protein A but not by the IgM RF or F(ab')2 of anti-idiotype antibodies to RF (anti-Id RF). Inversely, unlabelled IgM RF and anti-Id RF inhibited the binding of 125I-IgM RF markedly and unlabelled human IgG and IgG Fc only slightly or moderately, respectively. Thus, group A streptococci type M15 showed different binding sites for IgG Fc and the antibody combining sites of a human monoclonal RF. The findings were still more complex on a background of previous reports showing that streptococcal IgG Fc receptors and RFs bind to the same amino acids on the Fc molecule. This complex pattern may play a role in the pathogenesis of rheumatoid arthritis.     

A serum antibody in patients with rheumatoid arthritis stimulates cathepsin B activity in peritoneal mouse macrophages.

Mouse peritoneal macrophages were stimulated by sera from patients with active rheumatoid arthritis (RA) to increased intracellular cathepsin B activity. By gel filtration of three RA sera, the stimulatory activity was found in the IgG and to a lesser extent in the IgM containing fraction. The DEAE-cellulose purified IgG preparations of five additional RA patients stimulated intracellular cathepsin B activity significantly above IgG from healthy controls. IgG and IgM antibodies to macrophages were detected in sera from RA patients but not from controls by indirect immunofluorescence (IIF) technique. Pepsin F (ab')2 fragments of IgG from the RA patients also gave clearcut membrane fluorescent staining of the macrophages which demonstrated the antibody nature of the binding. A good correlation between the cathepsin B assay and the IIF was found when serial dilutions of serum were compared.     

IgG with a Deviant Conformation in Serum and Synovial Fluid from Rheumatoid Arthritis Patients

Specific rabbit antisera were prepared against an IgG with a special conformation (IgG spec.) previously detected in some sera from patients with rheumatoid arthritis. The antibodies had no affinity to normal human IgG and were not anti-idiotypic to human rheumatoid factor. The affinity of IgG spec. to the antibodies could not be explained by an antiglobulin activity to rabbit IgG. The amount of protein with affinity to immobilized specific IgG F(ab')2 of the antibodies was determined in serum and synovial fluid from patients with various joint diseases. A relationship between the content of IgG spec. and the diagnosis of seropositive rheumatoid arthritis was found on analysis of serum samples. IgG spec. also occurred in synovial fluid from some individuals with seropositive rheumatoid arthritis. Differences in the serum content of IgG spec. could not be explained by differences in the normal IgG content. Circular dichroism analysis of isolated IgG spec. showed that in the region(s) close to tyrosine residue(s) this polyclonal protein had similarities to heat-aggregated IgG.     

Interaction between herpes simplex type 1-induced Fc receptor and human and rabbit immunoglobulin G (IgG) domains.

Cells infected with herpes simplex virus type 1 (HSV-1) express a cell surface receptor able to bind to the Fc region of immunoglobulin G (IgG). The ability of HSV-1-infected cells to bind 125I-labelled human and rabbit IgG and IgG fragments was studied to localize the site of interaction to the C gamma 2 or C gamma 3 domains of IgG. 125I-labelled IgG and IgG Fc fragments consisting of C gamma 2 and C gamma 3 domains bound strongly to HSV-infected cells and did not bind to uninfected cells. In contrast, 125I-labelled F(ab')2, Facb [consisting of F(ab')2 and C gamma 2 domains] and pFc' (consisting of C gamma 3 domains) fragments did not bind to any of these cells. Unlabelled IgG and IgG Fc fragments inhibited the interaction between 125I-labelled rabbit IgG Fc and the HSV Fc receptor, whereas F(ab')2, Facb and pFc' fragments failed to inhibit this interaction. These data indicate that the HSV Fc receptor requires both the C gamma 2 and C gamma 3 domains for interaction with the IgG molecule analogous to the known interaction of protein A of Staphylococcus aureus, the Fc binding proteins of Group A, C and G streptococci, and certain human rheumatoid factors.     

Human IgG Rheumatoid Factors and RF-Like Immune Complexes Induce IgGl Rheumatoid Factor Production in Mice

The synovial fluid of patients with rheumatoid arthritis (RA) was found to contain IgG and /or IgG-containing immune complexes (ICs) that stimulated an intense antibody formation when injected into mice of certain strains, notably of NZ background. The response was characterized by high and sustained levels of IgG1 antibodies with rheumatoid factor (RF) activity. In the study described, we investigated whether it is the antibodies with RF activity in the synovial fluid, that are responsible for stimulation of mouse RF in vivo. Different mouse strains were injected with synovial fluid from a seropositive RA patient (RA-SF), with human monoclonal antibodies with RF activity, with a human non-RF monoclonal antibody or with different preformed RF-like antibody antibody (Ab-Ab) ICs. The experimental mice were monitored subsequently for IgGl RF production. IgGl RF antibodies were found in all strains (NZB, BALB/c and CBA) injected with Ab-Ab ICs formed at equivalence, but only in NZB using RA-SF or human monoclonal antibodies with RF activity. Optimal production of IgG I RF by Ab-Ab ICs required the integrity of Fc and F(ab)'₂ portions respectively of the antibodies; soluble and truncated ICs were less effective. Further studies demonstrated that the IgGl RF response was not simply the result of a specific immune response against human IgG, since humoral immunity against human IgG was induced only when combined with an efficient adjuvant. During a typical adjuvant-associated primary response specific antibodies of IgM, IgG1 and IgG2a isotypes were found, i. e. quite different from the selective IgGl response induced by RF-like containing immune complexes. This conclusion is substantiated further by the clear differences in responses to IgG containing fraction obtained from RA-SF in NZ mice compared to other strains. Our findings argue for a different type of reaction leading to the selective IgGl response and might aid in elucidating the mechanisms for chronic production of antibodies with RF activity in patients with RA.     

Binding site and subclass specificity of the herpes simplex virus type 1-induced Fc receptor

Immunoglobulin Fc-binding activity was detected by indirect immunofluorescence employing fluorochrome conjugated F(ab')2 antibody fragments on acetone-fixed cell cultures infected with herpes simplex virus type 1 (HSV-1). Using this method the Fc receptor-like activity seemed to be restricted to the IgG class of human immunoglobulins. While IgG1, IgG2, and IgG4 myeloma proteins bind to this putative Fc gamma receptor at a concentration of 0.002 mg/ml, IgG3 myeloma proteins were without activity at 0.1 mg/ml. The binding activity was associated with the Fc fragments of IgG, while the pFc' fragments of IgG appeared to be unable to bind in this assay system. The reactivity and specificity of the HSV-1 Fc receptor was independent of both the type of tissue culture cells used and the strain of HSV-1 inducing the Fc receptor-like activity. The HSV-1-induced Fc receptor has a similar specificity for human immunoglobulin class and subclasses as staphylococcal Protein A. However, these two Fc receptors exhibit at least one striking difference. The IgG3 G3m(st) protein which binds to Protein A does not bind to HSV-1-induced Fc receptor. A possible reaction site for the HSV-1 Fc receptor on IgG could be at or near Asp 276.     

Binding of monoclonal IgM rheumatoid factor to streptococci via the antibody combining site

Radiolabelled monoclonal IgM rheumatoid factors, four from patients with type II essential cryoglobulinaemia and one originating from a patient with rheumatoid arthritis, were tested for binding to group A, B, C and G streptococci and Escherichia coli. Two of the preparations exhibited different binding patterns for the streptococci, whereas the remaining three were not reactive. The binding of one of the factors to group A streptococci type 15 was inhibited by F(ab')2-fragments of anti-idiotypic antibodies but not anti-IgM Fc or F(ab')2 of normal rabbit IgG, indicating that the reaction was mediated via the antibody combining sites.     

Rheumatoid factor and glomerulonephritis

It is presently unknown whether rheumatoid factors have a pathogenic role in the development of various types of glomerulonephritis with immune deposits. Three isotypes of rheumatoid factors (RFs), which are autoantibodies to IgG, were measured using the solid-phase fluorescence immunoassay in sera from patients with diffuse proliferative lupus nephritis (DPLN), membranous lupus nephritis (MLN), IgA nephropathy (IgAN) and idiopathic membranous nephropathy (MN). RF activity of immunoglobulins deposited in the glomeruli from these patients was also studied by examining the binding of the FITC-conjugated human IgG and Fc portion of IgG to the glomeruli of renal biopsy specimens. IgG, IgA and IgM RFs were significantly increased in sera from patients with DPLN, and the increase was significantly lower in patients with MLN, IgAN and MN. Human IgG bound to immunoglobulin on the glomeruli only in DPLN, but not in MLN, IgAN or MN. The Fc portion of IgG was demonstrated to be involved in this reaction. It was suggested that RFs and IgG may play a major role in immune deposits on the glomeruli in DPLN and may be involved in the development of DPLN; however, this is not likely in MLN, IgAN or MN.     

Antibodies to Human Immunoglobulin Detected by Hemolysis of Human Immunoglobulin-Coated Red Blood Cells

A hemolysis assay was developed to detect alloantibodies to human immunoglobulin. A total of 1035 serum samples was tested. Anti-IgM antibodies were found in 8% of 59 normal persons and in 13% of 439 multiparous women, with the highest incidence of 67% in 341 dialysis patients. Although the anti-IgM antibodies were inhibited by both IgM and IgG, it appeared that they were also inhibited by F(ab')₂ but not by Fc. Anti-IgG antibodies were more strongly inhibited by Fc than F(ab')₂. These results suggest that anti-IgM antibodies might be analogous to antiidiotypic antibodies directed to F(ab')₂, whereas anti-IgG antibodies tend to have greater reactivity to Fc.     

The Distinct Subgroup of Patients with Rheumatoid Arthritis Shown by IgG3-Reactive Rheumatoid Factor

The reactivity of rheumatoid factor (RF) with immunoglobulins of the IgG3 subclass was examined in 49 patients with rheumatoid arthritis (RA) using two types of IgG3 myeloma (routine and IgG3m-15 allotype). Among 49 patients, serum from eight cases showed positive reactivity with both types of IgG3 myeloma by radio-immunoassay (RIA). The isotype of IgG3-reactive RF was not specific; it belonged to the IgM class as well as the IgG subclasses IgG1, IgG2 and IgG4. The patients with IgG3-reactive RF belonged to the clinically-severe classification of RA, having a high erythrocyte sedimentation rate (ESR), high titre in the RA hemagglutination (RAHA) test, and above all they had low levels of complement. Generally, it is concluded that patients with IgG3-reactive RF have serious arthritis and that IgG3-reactive RF might play an important role in the inflammatory process. Furthermore, it was also shown that the RF-reactive site was not associated with the protein-A binding site of IgG3, since RF reacting with IgG3m-15 reacted similarly with routine IgG3, regardless of the difference of the protein-A binding activity. This was confirmed by adding protein-A to the reaction of RF and IgG3m-15 which binds with protein-A. ‘This suggests that the actual reactive site of RF is different to the site that binds protein-A.     

Anti-fibronectin autoantibodies in systemic lupus erythematosus, rheumatoid polyarthritis, and various viral or bacterial infectious diseases

A micro-enzyme linked immunosorbent assay (ELISA) aimed at detecting anti-fibronectin (anti-Fn) antibodies has been developed and standardized. Fifty sera from systemic lupus erythematosus (SLE), 50 from rheumatoid arthritis (RA) patients, as well as 200 sera from patients with bacterial or viral infections were assayed for the presence of anti-Fn autoantibodies. The IgG fractions of three representative positive sera (1 SLE, 1 RA and 1 streptococcal endocarditis) were digested with pepsin and the resulting F(ab')2 fragment assayed in the test. The presence of the anti-Fn activity in these fragments as well as lack of correlation in individual sera between the level of anti-Fn (as determined by ELISA) and that of Ig or immune complexes, suggest that our anti-Fn autoantibodies are indeed detected in our assay. The meaning of these antibodies, which were also found with bacterial and viral infections is discussed within the frame of the fibronectin biological properties.