Specific and shared idiotypes found on hybridoma anti-DNA autoantibodies derived from rheumatoid arthritis and systemic lupus erythematosus patients

The idiotype determinants found on hybridoma anti-DNA autoantibodies produced from the fusion of peripheral blood lymphocytes from 13 systemic lupus erythematosus (SLE) and five rheumatoid arthritis (RA) patients with the GM 4672 human lymphoblastoid line were analyzed. A total of 47 SLE and 21 RA hybridomas were studied, of which 26 SLE and 10 RA produced anti-DNA autoantibodies. Rabbit antisera, raised to six of the SLE hybridoma anti-DNA IgM antibodies, were rendered idiotype specific by multiple absorptions on human IgM and IgG immunoabsorbent columns. In direct binding radioimmunoassays, all six anti-idiotype antisera reacted specifically with the anti-DNA antibody used as immunogen. In competition studies, five anti-idiotype antisera were able to inhibit the binding of their homologous idiotype to DNA-coated tubes. In addition, DNA and polynucleotides inhibited the binding of the five idiotypes to anti-idiotype-coated tubes, suggesting that these anti-idiotypes react with idiotype determinants located within the antigen-combining sites of the anti-DNA antibody molecules. Shared idiotypes were detected among the 68 hybridoma antibodies by direct binding studies on anti-idiotype-coated tubes. Our results revealed that 58% (21/36) of the anti-DNA antibodies and 16% (5/32) of the non-DNA-binding antibodies reacted with at least one anti-idiotype serum. Five anti-idiotype antisera reacted only with hybridoma anti-DNA antibodies from SLE patients. The other anti-idiotype antiserum reacted with both SLE- and RA-derived hybridoma anti-DNA and non-DNA-binding antibodies. These studies indicate that some anti-idiotype antisera may detect specific idiotypes found only on SLE-derived anti-DNA auto-antibodies, whereas other antisera detect shared idiotypes found on both RA and SLE DNA-binding and non-DNA-binding antibodies.     

A cross-reactive idiotype on anti-DNA antibodies defines a H chain determinant present almost exclusively on IgG antibodies

We have previously reported two anti-idiotypic antibodies, 3I and 8.12, that recognize L chain determinants on anti-DNA antibodies. We have generated a new anti-idiotypic antibody, F4, that recognizes a H chain determinant on cationic anti-DNA antibodies. F4 reactivity is present in high titer in serum of approximately 60% of SLE patients and on 84 of 706 myeloma proteins. It is preferentially associated with 3I reactive L chains. Furthermore, antibodies bearing both the F4 and 3I idiotypic determinants preferentially bind DNA. Amino acid sequencing of H chains isolated from four F4-reactive myeloma proteins suggests that they derive from two currently identified VH gene families. F4 reactivity is restricted almost exclusively to Ig of the IgG isotype suggesting that F4 may recognize either a somatically mutated hypervariable region or a variable region used late in the immune response. F4, therefore, represents a new idiotypic family preferentially associated with auto-Ag specificity and having features of an Ag-driven immune response.     

Idiotypic analysis of myeloma proteins: anti-DNA activity of monoclonal immunoglobulins bearing an SLE idiotype is more common in IgG than IgM antibodies

The anti-idiotype 3I which recognizes a determinant on kappa-chains of anti-DNA antibodies in SLE patients also recognizes a determinant on kappa-chains of 82/706 myeloma proteins tested. Twenty-nine of these 82 proteins bind to double-stranded DNA, including two monoclonal IgM, one monoclonal IgA, and 26 monoclonal IgG proteins. DNA binding is much more frequent in the IgG than in the IgM myeloma proteins (p less than 0.001), and is also associated with cationic antibody charge. Two-dimensional gel electrophoresis reveals markedly increased charge heterogeneity of both heavy and light chains of the monoclonal IgG as compared with the monoclonal IgM proteins. We postulate that the increased charge heterogeneity of the IgG-associated 3I-reactive kappa-light chains may reflect somatic mutation, and that DNA specificity within the 3I idiotype system arises by somatic mutation of germ-line genes found in normal individuals. DNA binding may be associated with those mutations that give rise to a cationic immunoglobulin charge.     

A conserved anti-DNA antibody idiotype associated with nephritis in murine and human systemic lupus erythematosus

In order to identify unique structural features of pathogenic autoantibodies to DNA in SLE, a murine anti-anti-DNA (anti-Id) mAb (mAb 1C7) was produced in response to immunization of lupus mice with a syngeneic anti-DNA mAb (mAb 3E10). Immunization of lupus mice with mAb 3E10 inhibited production of native anti-DNA antibodies, suppressed development of lupus kidney disease (nephritis), and induced production of anti-anti-DNA (anti-Id) antibodies. mAb 1C7 bound F(ab')2 fragments of mAb 3E10, and it bound other murine anti-DNA mAb, but not murine mAb or polyclonal serum antibodies unreactive with DNA. Moreover, binding of mAb 1C7 anti-Id to mAb 3E10 was inhibited by DNA, suggesting anti-Id binding within or near the binding site for DNA. Furthermore, mAb 1C7 bound serum IgG immunoglobulins from 9/12 patients with lupus nephritis and serum anti-DNA antibodies compared to only 3/12 SLE patients with comparable serum levels of anti-DNA antibodies, but without nephritis (p = 0.04), and only 1/53 SLE patients without serum anti-DNA antibodies, 0/49 patients with rheumatoid arthritis, and 1/47 healthy subjects (p less than 0.001). These results provide evidence that mAb 1C7 identifies a conserved Id associated with anti-DNA antibodies in murine and human SLE and may be useful as a structural probe to characterize pathogenic anti-DNA antibodies in SLE.     

Detection of masked anti-DNA antibodies in lupus sera by a monoclonal anti-idiotype

We report here the use of a monoclonal anti-idiotype 3I to human anti-DNA antibodies to detect in serum idiotype-positive antigen-binding antibodies lacking DNA-binding activity as measured by conventional antigen binding assays. We studied paired serum samples from 13 patients with systemic lupus obtained at two times in the course of their disease: in each patient, one serum sample has anti-DNA activity and the second serum sample has no anti-dsDNA activity detectable by Millipore filter, ELISA, or Crithidia assay. Reactivity with 3I as detected with a radioimmunoassay (RIA) was present in all 13 sera with anti-dsDNA activity. Six patients showed a decrease in 3I reactivity to normal levels in the second serum sample, in which anti-dsDNA antibodies were not detectable by conventional antigen-binding assays. The other seven patients' second serum sample continued to show elevated 3I reactivity by RIA even though no anti-dsDNA activity was apparent. When the 3I-reactive antibodies from these latter patients' sera were eluted from a 3I affinity column, they revealed DNA-binding activity. Furthermore, dsDNA binding by these sera was apparent when they were displayed on Western blots of isoelectric focusing gels run in 8 M urea and incubated with radiolabeled dsDNA. These results indicate that the 3I anti-idiotype can detect anti-DNA antibodies in some sera of SLE patients that lack anti-DNA activity by ordinary assays. These antibodies may be inhibited in binding dsDNA by excess antigen or autologous anti-idiotype, and their DNA binding activity can be unmasked by procedures promoting immune complex dissociation.     

Preferential expression of the systemic lupus erythematosus-associated idiotype 8.12 in sera containing monoclonal immunoglobulins

The 8.12 idiotype defines a population of anti-DNA antibodies present in the serum of patients with systemic lupus erythematosus. As part of our studies to elucidate the genetic origin and structural features of anti-DNA antibodies, we examined monoclonal immunoglobulin (Ig)-containing sera from 706 patients for expression of the 8.12 idiotype. We found 41 such sera to have significant 8.12 reactivity (greater than 4 SD above the mean of normal controls) and demonstrated that in 24 of these sera (8 IgM, 14 IgG, and 2 IgA) this reactivity could be localized to the monoclonal protein. In addition, 12 of the 8.12-reactive monoclonal Ig (11 IgG and 1 IgA) bind dsDNA. In the other 17 sera, the 8.12 reactivity could be attributed to polyclonal antibody. These findings provide further evidence that the serum monoclonal Ig frequently express the antigenic and idiotypic reactivities of autoantibodies. Furthermore, these data support the contention that anti-DNA specificity may result from somatic diversification of germ-line Ig gene sequences.     

Lupus serum and normal human serum contain anti-DNA antibodies with the same idiotypic marker

We have previously shown that serum from patients with active SLE contain high levels of Id-16/6 and anti-DNA antibodies. In this study we investigated whether serum Id 16/6 is related to anti-DNA antibodies. Sera from 12 patients with active SLE were absorbed individually with poly(dT) cellulose (to purify anti-DNA antibodies) and rabbit (R) anti-Id-16/6 Sepharose (to purify Id 16/6 Ig). Removal of all anti-DNA activity removed most of the Id-16/6. Conversely, removal of all Id 16/6 removed most of the anti-DNA activity. Although there was no measurable anti-DNA antibody activity in normal serum, such antibodies were isolated by absorption with poly(dT) cellulose. The eluted immunoglobulins also had Id 16/6 activity. Similarly, Id 16/6 with anti-DNA activity were isolated from normal serum by absorption with R anti-Id 16/6 Sepharose. We conclude that a large fraction of anti-DNA antibodies in SLE serum are Id-16/6+, and that most Id 16/6 immunoglobulins in lupus serum have anti-DNA activity. Our observations suggest that lupus anti-DNA antibodies result from an overproduction of autoantibodies that are present in normal people.     

Hybridoma anti-DNA autoantibodies from patients with rheumatoid arthritis and systemic lupus erythematosus demonstrate similar nucleic acid binding characteristics

Hybridoma anti-DNA antibodies have been generated from the fusion of the GM 4672 lymphoblastoid line with peripheral blood lymphocytes from four normal subjects, nine patients with rheumatoid arthritis (RA), and 13 patients with systemic lupus erythematosus (SLE). A total of 441 hybridoma clones were obtained, of which 37 secreted anti-DNA autoantibodies. The nucleic acid binding characteristics of the anti-DNA antibodies produced by two hybridomas from normal subjects, nine hybridomas from RA patients, and 18 hybridomas from SLE patients are reported. The hybridoma anti-DNA antibodies from all three groups showed similar antigen-binding characteristics for denatured DNA (dDNA), native DNA (nDNA), poly(I), poly(dT), and cardiolipin, by both direct binding and competitive binding analyses. One difference noted between normal-derived anti-DNA antibodies and autoimmune-derived antibodies was the inability of the former to react with z-DNA. However, this requires further substantiation with larger numbers of normal-derived clones. The broad overlap of reactivity to nucleic acid antigens among individual anti-DNA autoantibodies found in two clinically different autoimmune diseases, namely RA and SLE, suggests that the pathogenicity of anti-DNA autoantibodies may bear no relationship to their nucleic acid antigen-binding characteristics.     

Use of anti-idiotypic antibodies to explore genetic mechanisms of production of anti-DNA antibodies

Systemic lupus erythematosus (SLE) is characterized by the production of autoantibodies with a broad range of antigenic specificities, including specificity for double-stranded DNA. Analysis of the idiotypic profile of anti-DNA antibodies both in humans and mice has demonstrated presence of cross-reactive idiotypes, suggesting that they arise from a restricted number of germline genes. Our laboratory has previously reported the generation of 3I, a monoclonal anti-idiotypic antibody which recognizes a cross-reactive idiotype on anti-DNA antibodies in a majority of unrelated humans with SLE. We have recently studied the expression of 3I in sera of three human kindreds with familial SLE. We found 6 of 8 SLE patients and 15 of 19 unaffected family members had elevated 3I reactivity. Eleven of these family members had no anti-DNA activity despite elevated 3I reactivity, suggesting that expression of this idiotype in certain individuals is part of the normal immune response. In another set of experiments using an in vitro culture system we examined somatic mutants of the S107 mouse myeloma cell line. This line makes an antibody which bears the T15 idiotype, a common idiotype on antibodies to the bacterial antigen phosphoryl choline (PC). U4, a mutant, makes an immunoglobulin which varies by one amino acid from the parent protein, retains the T15 idiotype, but loses reactivity with PC and acquires reactivity with DNA. We have found that some anti-DNA antibodies in mice with spontaneous lupus and in mice immunologically induced to make anti-DNA antibodies bear the T15 idiotype and may represent somatic mutants arising in vivo.     

Identification of a Lipid Peroxidation Product as the Source of Oxidation-specific Epitopes Recognized by Anti-DNA Autoantibodies

Lipid peroxidation in tissue and in tissue fractions represents a degradative process, which is the consequence of the production and the propagation of free radical reactions primarily involving membrane polyunsaturated fatty acids, and has been implicated in the pathogenesis of numerous diseases, including systemic lupus erythematosus (SLE). We have found that bovine serum albumin incubated with peroxidized polyunsaturated fatty acids significantly cross-reacted with the sera from MRL-lpr mice, a representative murine model of SLE. To identify the active substances responsible for the generation of autoantigenic epitopes recognized by the SLE sera, we performed the activity-guiding separation of a principal source from 13-hydroperoxy-9Z,11E-octadecadienoic acid and identified 4-oxo-2-nonenal (ONE), a highly reactive aldehyde originating from the peroxidation of ω6 polyunsaturated fatty acids, as the source of the autoantigenic epitopes. When the age-dependent change in the antibody titer against the ONE-modified protein was measured in the sera from MRL-lpr mice and control MRL-MpJ mice, all of the MRL-lpr mice developed an anti-ONE titer, which was comparable with the anti-DNA titer. Strikingly, a subset of the anti-DNA monoclonal antibodies generated from the SLE mice showing recognition specificity toward DNA cross-reacted with the ONE-specific epitopes. Furthermore, these dual-specific antibodies rapidly bound and internalized into living cells. These findings raised the possibility that the enhanced lipid peroxidation followed by the generation of ONE may be involved in the pathogenesis of autoimmune disorders.     

The role of the idiotypic network in the induction of experimental systemic lupus erythematosus

Systemic lupus erythematosus (SLE) has been induced in C3H.SW mice by their immunization with a human monoclonal anti-DNA antibody that bears a common idiotype-16/6 Id. Following immunization, high levels of murine anti-16/6 and anti-anti-16/6 antibodies were detected in the sera of the immunized mice. Elevated titers of autoantibodies reacting with ssDNA, dsDNA, poly(I), poly(G), RNP, Ro, and La were also observed. The serological findings were associated with significant proteinuria, leukopenia, and elevated erythrocyte sedimentation rate. Immune complex deposition in the glomerular mesangium and sclerosis of the glomeruli were demonstrated. To study whether or not anti-idiotype antibodies are involved in the induction of the disease, a murine monoclonal antibody against the 16/6 Id was prepared and injected into C3H.SW mice. The anti-16/6 Id antibody induced experimental SLE similarly to the 16/6 Id with an accelerated kidney pathology. A study performed on different mouse strains indicated that the susceptibility to the induction of SLE by the 16/6 Id is strain dependent and directly correlates to their ability to produce anti-16/6 Id specific antibodies.     

Expression of anti-idiotypic clones against auto-anti-DNA antibodies in normal individuals

A serum pool from 280 blood donors and individual samples from blood donors were assayed for anti-idiotypic activity to auto-anti-DNA antibodies by competitive radioimmunoassays. We found that serum from several normal blood donors inhibited the binding activity of anti-DNA antibodies affinity purified from the sera of patients with systemic lupus erythematosus. This inhibition was due to immunoglobulin molecules but was not due to rheumatoid factor activity. Antiallotypic antibodies were not responsible for the anti-anti-DNA activity detected. Because this inhibition was blocked by DNA molecules, the observed reactivity was probably caused by idiotype-anti-idiotype interactions. These results provide evidence that anti-idiotype antibodies against anti-DNA autoantibodies are present in certain normal human sera. Anti-anti-DNA antibodies could play a role in the regulation of autoimmunity to DNA.     

Protein-bound 4-hydroxy-2-nonenal: an endogenous triggering antigen of antI-DNA response

Several lines of evidence indicate that the nonenzymatic oxidative modification of proteins and the subsequent accumulation of the modified proteins have been found in cells during aging and oxidative stress and in various pathological states, including premature diseases, muscular dystrophy, rheumatoid arthritis, and atherosclerosis. Our previous work suggested the existence of molecular mimicry between antibodies raised against hydroxy-2-nonenal (HNE)-modified protein and anti-DNA autoantibodies, a serologic hallmark of systemic lupus erythematosus (SLE). In the present study, we investigated the possible involvement of HNE-modified proteins as the endogenous source of the anti-DNA antibodies. Accumulation of the antigen recognized by the antibody against the HNE-modified protein was observed in the nucleus of almost all of the epidermal cells from patients with autoimmune diseases, including SLE. The SLE patients also showed significantly higher serum levels of the anti-HNE titer than healthy individuals. To determine if a specific anti-DNA response could be initiated by the HNE-derived epitopes, we immunized BALB/c mice with the HNE-modified protein and observed a progressive increase in the anti-DNA response. Moreover, we generated the monoclonal antibodies, showing recognition specificity toward DNA, and found that they can bind to two structurally distinct antigens (i.e. the native DNA and protein-bound 4-oxo-2-nonenal). The findings in this study provide evidence to suspect an etiologic role for lipid peroxidation in autoimmune diseases.     

Identification and sequence of the VH gene elements encoding a human anti-DNA antibody

Antibodies to DNA similar to those found in patients with systemic lupus erythematosus (SLE) and autoimmune mice can be derived from the lymphocytes of normal individuals. It is not known whether these normal derived anti-DNA antibodies are made from the same VH gene elements as the anti-DNA antibodies made by SLE patients. To begin to answer this question, we examined mu chain cDNA clones from human hybrid clone C6B2 producing anti-DNA antibodies. The sequence of the 500 base pair restriction fragment containing the variable region (5' terminus) was determined and was sequenced. This antibody uses a VHII heavy chain subgroup gene, a J3 joining segment, a hitherto unknown D segment, and a previously reported leader sequence. Significant homology was found to a mouse anti-DNA antibody sequence in the use of VH subgroup in J3, and in the hypervariable regions with a shared Ser-Tyr construction in CDR1 and an identical five amino acid residue stretch in CDR2. Comparison with the limited sequence data of published SLE monoclonal anti-DNA antibodies, both human and mouse, suggests that this shared Ser-Tyr may be important in some but not all antibodies to DNA. Comparison of C6B2 antibody is made with other known antibody sequences with identification of those residues likely to be part of the antigen binding site.     

The generation of monoclonal anti-idiotype antibodies to human B cell-derived leukemias and lymphomas

We developed murine anti-idiotype monoclonal antibodies for each of four patients with B cell-derived leukemias and lymphomas. Idiotypic immunoglobulin was isolated from mouse X human tumor-cell hybridomas or from patients' serum and was used to immunize mice for the development of murine anti-idiotype monoclonal antibodies. Each patient's anti-idiotype antibodies demonstrated reactivity restricted to the immunizing immunoglobulin, thereby limiting their therapeutic utility to a single individual. In addition, we isolated isotype switch variants of hybridomas producing monoclonal anti-idiotypic antibody. The restricted specificity of these antibodies was found to be of value for the analysis of the extent of malignant B cell infiltration in a variety of tissues from several patients. Large populations of idiotype-bearing cells were detectable in biopsy specimens from patients K.T. and L.H. In contrast, although bone marrow specimens from patient G.D. were apparently devoid of morphologically abnormal cells, a small, highly fluorescent population of cells was demonstrable underscoring the potential utility of these antibodies for posttreatment evaluation as well as for therapy. In a fourth patient, H.M., anti-idiotype antibodies developed against the circulating macroglobulin isolated from his plasma failed to react with either his circulating or bone marrow hairy cell leukemia cells. However, examination of an enlarged inguinal lymph node revealed the presence of a large number of idiotype-bearing cells. Thus, the presence of two distinct malignant B cell clones were discovered in this individual through the use of anti-idiotype monoclonal antibodies. Anti-idiotype antibodies, therefore, represent a highly specific tool for the evaluation and potential therapy of B cell malignancies in individual patients.     

Characterization of the human Ig V lambda II gene family and analysis of V lambda II and C lambda polymorphism in systemic lupus erythematosus.

We report the cDNA sequence of an expressed human V lambda II gene and present an RFLP analysis of the Ig gene family defined by this clone. This V lambda II gene was expressed in a monoclonal B cell line generated from a patient with SLE by transformation with EBV. The encoded lambda L chain displays the 8.12 Id, an Id common to anti-DNA antibodies from patients with SLE. Using a coding region probe we estimate from Southern blot analysis that the germline V lambda II gene family contains at least 15 members. Many of the V lambda II restriction fragments are polymorphic both in SLE patients and in nonautoimmune individuals. EcoRI, HindIII, and TaqI RFLP analyses of the V lambda II gene family and EcoRI analysis of the C lambda gene family reveal no polymorphisms specific to SLE. Observed V lambda II and C lambda allele frequencies are the same among SLE patients and nonautoimmune individuals, and show no evidence of linkage disequilibrium between the two loci.     

Human immune response to multiple injections of murine monoclonal IgG

Murine monoclonal antibody infusions in humans should induce a human anti-mouse immunoglobulin (mIgG) immune response, especially if multiple infusions over an extended period of time are necessary for therapeutic efficacy. We have administered multiple infusions of the murine monoclonal antibody T101 to patients with cutaneous T cell lymphoma (CTCL) or chronic lymphocytic leukemia (CLL). Five of 10 CTCL patients, compared with zero of six CLL patients, developed antibodies to mIgG. In those CTCL patients who did not demonstrate anti-mIgG antibodies, we were unable to correlate the lack of response to any of a large number of clinical parameters. Anti-mIgG antibodies were of both the mu and gamma isotypes and were detectable 14 days after the first infusion. Multiple infusions were associated with elevated titers. The anti-idiotypic portion of the anti-mIgG titer steadily increased with each infusion until eventually, in one patient receiving eight weekly infusions, well over one-half the serum anti-mIgG recognized only T101 and not four other murine IgG2AK antibodies tested. To increase our confidence in these findings, four separate assay systems were used to make these determinations. The identification of anti-idiotype antibodies as the dominant species of the immune response to multiple infusions of murine monoclonal antibody has major implications for future work with monoclonal antibodies. Although it has been suggested that human monoclonal antibodies would obviate an immune response, our work implies that such antibodies might still induce anti-idiotype antibodies if multiple infusions are administered.     

Detection of cross reactive anti-DNA antibody idiotypes on tissue-bound immunoglobulins from skin biopsies of lupus patients

Cross-reactive anti-DNA antibody idiotypes have been identified on tissue-bound immunoglobulins from skin biopsies of patients with systemic lupus erythematosus (SLE) and discoid lupus erythematosus (DLE). Four polyclonal and two monoclonal anti-idiotypic reagents were used to screen biopsies from 24 patients with SLE, 23 patients with DLE, and 15 other patients with IgM-positive skin biopsies. Up to 46% of the SLE patients and 30% of the DLE patients were found to share idiotypes present on immunoglobulins deposited at the dermal-epidermal junction. Inhibition studies in four patients indicated that the idiotypes were on anti-DNA antibodies. In contrast, none of the anti-idiotypic antibodies bound to any of the control biopsies. These findings imply that some tissue-bound autoantibodies are derived from related families of high-frequency germ-line genes that are expressed in both SLE and DLE.     

Anti-RNP monoclonal antibodies derived from a mouse strain with lupus-like autoimmunity

systemic lupus erythematosus (SLE) and related rheumatic and connective-tissue diseases are often associated with the production of antibodies directed against a variety of specific cellular components. Recent evidence indicates that two such autoantigens, the Sm and RNP antigens recognized by SLE sera, exist in small ribonucleoprotein complexes found in the nuclei of higher eukaryotes. Studies of the structure and function of these autoantigenic particles with human sera used as probes have been limited because of the multiplicity of autoantibodies often found in an individual serum. Through this communication, we report that MRL/Mp-+/+ (MRL/n) mice, which spontaneously develop a disease exhibiting many of the characteristics of human SLE, possess anti-RNP antibodies in addition to anti-Sm and anti-DNA as previously reported. Spleen cells from one such autoimmune mouse were used to produce a stable hybridoma secreting antibodies that react simultaneously with a protein of Mr 40,000 and a doublet of approximately 70,000, a pattern of reactivity identical to and characteristic of human SLE anti-RNP autoantibodies.     

Pathogenic autoantibodies in systemic lupus erythematosus are derived from both self-reactive and non-self-reactive B cells

Previous studies have shown that both murine and human anti-double-stranded DNA (anti-dsDNA) antibodies can develop from non-DNA-reactive B cells and suggest a crucial role for somatic mutation in dsDNA binding. However, since only a limited number of human anti-dsDNA antibodies have been analyzed previously, we could not exclude other mechanisms for the generation of anti-dsDNA antibodies in patients with systemic lupus erythematosus (SLE). Therefore, we isolated IgM anti-dsDNA antibodies from peripheral blood B cells of a patient with SLE. Three somatically mutated IgM anti-DNA antibodies with pathogenic potential (glomerular binding) were reverted to their germline configuration. Although all three IgM anti-dsDNA antibodies came from the same lupus patient, they displayed different profiles. Reversion to the germline sequence of autoantibodies A9 and B5 resulted in decreased dsDNA binding. In contrast, the germline form of G3-recognized dsDNA as well as the mutated counterpart. These results suggest that mutated IgM anti-dsDNA antibodies may develop from both DNA- and non-DNA-reactive B cells. The implications are that B cell activation occurs in response to self and non-self antigens, while selection after activation may be mediated by self antigen in SLE. Moreover, ineffective tolerance checkpoints may exist before and after antigen activation in SLE.