Molecular characterization of a somatically mutated anti-DNA antibody bearing two systemic lupus erythematosus-related idiotypes

Not Dead Yet

I have been a scientific grasshopper throughout my career, moving from question to question within the domain of lupus. This has proven to be immensely gratifying. Scientific exploration is endlessly fascinating, and succeeding in studies you care about with colleagues and trainees leads to strong and lasting bonds. Science isn't easy; being a woman in science presents challenges, but the drive to understand a disease remains strong.

Mechanisms of B cell autoimmunity in SLE

Systemic lupus erythematosus (SLE) is a systemic autoimmune disease that is known to be associated with polyclonal B-cell hyperreactivity. The underlying causes of the diffuse B-cell over-reactivity are unclear, but potential candidates include (a) intrinsic hyper-reactivity leading to polyclonal B-cell activation with disturbed activation thresholds and ineffective negative selection; (b) lack of immunoregulatory functions; (c) secondary effects of an overactive inflammatory environment, such as overactive germinal center and ectopic follicular activity; and/or (d) disturbed cytokine production by non-B immune cells. These mechanisms are not mutually exclusive and may operate to varying extents and at varying times in SLE. Phenotypic and molecular studies as well as the results of recent clinical trials have begun to provide new insights to address these possibilities. Of importance, new information has made it possible to distinguish between the contribution played by abnormalities in central checkpoints that could lead to a pre-immune repertoire enriched in autoreactive B cells, on the one hand, and the possibility that autoimmunity arises in the periphery from somatic hypermutation and abnormal selection during T cell-dependent B-cell responses on the other. There is an intriguing possibility that apoptotic material bound to the surface of follicular dendritic cells positively selects autoreactive B cells that arise from non-autoreactive B-cell precursors as a result of somatic hypermutation and thereby promotes the peripheral emergence of autoimmunity.

Abnormalities of B cell subsets in patients with systemic lupus erythematosus

The prototypic autoimmune disease, SLE, is known to be associated with polyclonal B cell hyperreactivity. Developing an understanding of the complex nature of human B cell differentiation, largely through the application of multiparameter flow cytometry to an analysis of circulating B cells has permitted an assessment of whether specific stages of B cell maturation are affected by the tendency for polyclonal B cell activation. Moreover, the analysis of perturbations of the specific stages of B cell maturation has generated new information on whether abnormalities in B cell differentiation are primarily involved in autoimmune disease immunopathology or, rather, are secondary to the inflammatory environment characteristic of subjects with this autoimmune disease. Multivariant analysis has begun to document abnormalities in B cell maturation that are primarily associated with lupus, or, alternatively related to disease duration, disease activity and concomitant medication. Together, these analyses have provided new insights on the role of B cell over-reactivity in SLE.

Mathematical analysis of antigen selection in somatically mutated immunoglobulin genes associated with autoimmunity

Affinity maturation is a process by which low-affinity antibodies are transformed into highly specific antibodies in germinal centres. This process occurs by hypermutation of immunoglobulin heavy chain variable (IgH V) region genes followed by selection for high-affinity variants. It has been proposed that statistical tests can identify affinity maturation and antigen selection by analysing the frequency of replacement and silent mutations in the complementarity determining regions (CDRs) that contact antigen and the framework regions (FRs) that encode structural integrity. In this study three different methods that have been proposed for detecting selection: the binomial test, the multinomial test and the focused binomial test, have been assessed for their reliability and ability to detect selection in human IgH V genes. We observe first that no statistical test is able to identify selection in the CDR antigen-binding sites, second that tests can reliably detect selection in the FR and third that antibodies from nasal biopsies from patients with Wegener’s granulomatosis and pathogenic antibodies from systemic lupus erythematosus do not appear to be as stringently selected for structural integrity as other groups of functional sequences.

Systematic analysis of sequences of anti-DNA antibodies--relevance to theories of origin and pathogenicity

Sequence analysis of anti-DNA antibodies is important in determining the molecular features which distinguish potentially pathogenic antibodies from those which are less likely to be pathogenic. Previous analysis of murine anti-DNA antibody sequences suggested that particular murine immunoglobulin genes are used preferentially to encode such antibodies and that somatic mutations to arginine, asparagine and lysine may be important in the creation of DNA binding sites. In this paper, a systematic analysis of published human anti-DNA sequences shows no strong evidence for preferential usage of particular human V(H) or V(L) genes in anti-DNA antibodies. Somatic mutations in IgG and IgA antibodies are clustered in the complementarity determining regions (CDRs) due to the effect of antigen drive. This process contributes to an excess of arginine, asparagine and lysine residues in these CDRs, some of which are likely to play an important role in binding to DNA. Computer modeling and in-vitro expression experiments are likely to help define the roles played by these residues in antigen binding and pathogenicity more clearly.

Polyreactivity of disease-associated anti-RBC IgG autoantibodies of patients with warm autoimmune haemolytic anaemia and natural anti-RBC IgG autoantibodies of healthy individuals

Anti-red blood cell (RBC) immunoglobulin G (IgG) autoantibodies are present in patients with warm autoimmune haemolytic anaemia (WAIHA), and, as natural autoantibodies, in healthy individuals. This study investigated whether the feature of polyreactivity discriminates disease-associated from natural anti-RBC IgG autoantibodies. The patterns of reactivity of purified anti-RBC IgG eluted from the RBC of WAIHA patients and from the RBC of healthy individuals were analysed using quantitative immunoblotting on a panel of whole human tissue or bacterial cell extracts as antigen sources. In parallel, the reactivity patterns of IgG purified from plasma were analysed. Anti-RBC IgG of WAIHA patients and of healthy individuals recognized a wide range of self- and nonself-antigens. The reactivity patterns of anti-RBC IgG were homogeneous among patients and controls, did not differ between patients and controls, and were similar to those of IgG purified from plasma in the case of both patients and healthy individuals. The data demonstrate that the anti-RBC IgG autoantibodies of WAIHA patients share extensive similarity with those of healthy individuals. Polyreactivity is a common feature of both disease-associated and natural anti-RBC IgG autoantibodies.

Evidence that human immunoglobulin M rheumatoid factors can Be derived from the natural autoantibody pool and undergo an antigen driven immune response in which somatically mutated rheumatoid factors have lower affinities for immunoglobulin G Fc than their germline counterparts

The question of whether immunoglobulin (Ig)M rheumatoid factors (RF) arise as the result of an abnormal expansion of already existing clones producing natural autoantibodies or emerge as new clones that are somatically mutated owing to an antigen driven immune response has never been conclusively answered. In this study, an inhibition ELISA was utilized to measure the affinities of recombinant antibodies using VH segments reverted back to their closest germline counterparts (germline revertants). In all cases, the somatically mutated parental RFs had a decreased affinity for immunoglobulin (Ig)G Fc compared to the germline revertant, indicating that the antibodies in the germline configuration had the higher affinities. This demonstrates that somatic mutation is not a prerequisite to generate disease associated antibodies. The presence of mutations in the parental IgM RFS suggests that these cells had been involved in a germinal centre reaction. As the germinal centre is the conventional site of the acquisition of mutations during an antigen driven response, these data suggest a role for germinal centres in the generation of the antibody diversity in addition to the selection of higher affinity antibodies. Assuming that only antigen selected cells survive deletion, these data support the hypothesis that IgM RFS can be derived from the natural autoantibody repertoire and result from an antigen driven response. Mechanisms controlling the survival of B cells based on the affinity/avidity of the immunoglobulin receptor are shown to be functional in patients with rheumatoid arthritis.

Autoreactive T cells in murine lupus: origins and roles in autoantibody production

The conventional paradigm to explain systemic lupus erythematosus (SLE) is that disease results from tissue deposition of pathogenic autoantibodies and immune complexes, secondary to activation of autoreactive B cells in the context of help from alphabeta T cells. Recent work in murine lupus has confirmed this notion and demonstrated that autoantigen-specific alphabeta T cells are absolutely required for full penetrance of disease, with such autoreactive alphabeta T cells, even in Fas-intact mice, likely arising from defects in peripheral tolerance. These studies have also revealed a network of regulation that also involves nonclassical pathogenic and downregulatory alphabeta and gammadelta T cells, suggesting that the lupus immune system involves more complex interactions than the conventional paradigm suggests.

Immunoglobulin variable region sequences of human monoclonal anti-DNA antibodies

OBJECTIVE

Anti-DNA antibodies are believed to be important in the pathogenesis of systemic lupus erythematosus (SLE). Antibodies that bind specifically and with high affinity to dsDNA are most closely involved in tissue damage. Analysis of the sequences of the variable regions of human monoclonal anti-DNA antibodies is useful in defining the structural features that give rise to these binding properties. This article systematically reviews the evidence derived from such sequences.

METHOD

Previous reviews of this subject have been hampered by incomplete knowledge of the human immunoglobulin variable region repertoire. In this article, the original sequence data from reports of over 50 human monoclonal antibodies (mAb) are reinterpreted by alignment to the most similar alleles of the most similar germline genes. This allows accurate estimation of the site and nature of somatic mutations.

RESULTS

Human IgG monoclonal anti-DNA antibodies generally carry more mutations than IgM. In many cases these have been selected by an antigen-driven process. In many of the more specific, higher affinity dsDNA binders, there is an accumulation of basic residues in the complementarity determining regions. However, many exceptions to this rule exist, particularly among IgM mAb.

CONCLUSIONS

Unlike murine anti-DNA antibodies, these human mAb show little evidence for preferential use of particular V(H), V(K) and V(lambda) genes or families to encode antibodies of this specificity.

VH usage and somatic hypermutation in peripheral blood B cells of patients with rheumatoid arthritis (RA)

The human antibody repertoire has been demonstrated to have a marked V-gene-dependent bias that is conserved between individuals. In RA patients, certain heavy chain V genes (VH) have been found to be preferentially used for encoding autoantibodies. To determine if such preferential use of VH genes in autoantibodies is associated with a general distortion of the V gene repertoire in RA patients, the VH composition of peripheral blood B cells was analysed among four RA patients and four age- and sex-matched healthy controls. Usage of individual VH genes (eight VH3 and three VH4 genes tested by hybridization with a set of gene-specific oligonucleotide probes) was highly biased among RA patients, but no evidence of a distortion in the bias was observed compared with healthy controls. However, the occurrence of somatic mutations in these VH genes (estimated by differential hybridization with motif-specific oligonucleotide probes targeted to CDR and FR of the tested genes, and by DNA sequence analysis) was strikingly different between patients and healthy subjects. The number of VH3 rearrangements that had accumulated somatic mutations and the number of mutations per rearrangement were significantly elevated in three of the four RA patients. A slight but not significant elevation in mutations among rearranged VH4 genes was also observed in these patients. These data suggest that although usage of individual VH genes among peripheral blood B cells is not affected by the disease, the autoimmune process may involve a significant fraction of the B cell compartment.

Lupus-specific antibodies reveal an altered pattern of somatic mutation

The F4 idiotype is a heavy chain determinant expressed almost exclusively on IgG immunoglobulins and is highly associated with specificity for double-stranded DNA. Since high-titered F4 expression is present predominantly in sera of patients with systemic lupus erythematosus (SLE), we thought F4+ IgG antibodies might constitute a useful subset of immunoglobulins in which to investigate lupus-specific alterations in variable (V) region gene expression or in the process of somatic mutation. This molecular analysis of F4+ B cell lines generated from lupus patients demonstrates that despite the strong association of F4 reactivity with specificity for native DNA, there is no apparent VH gene restriction. Furthermore, VH gene segments encoding these antibodies are also used in protective immune responses. An examination of the process of somatic mutation in F4+ antibodies showed no abnormality in frequency of somatic mutation nor in the distribution of mutations in complementarity-determining regions or framework regions. However, there was a decrease in targeting of mutations to putative mutational hot spots. This subtle difference in mutations present in these antibodies may reflect an intrinsic defect in mutational machinery or, more likely, altered state of B cell activation that affects the mutational process and perhaps also negative selection.

Affinity columns containing anti-DNA Id+ human myeloma proteins adsorb human epibodies from intravenous gamma globulin

OBJECTIVE

To study eluates of intravenous gamma globulin (IVGG) prepared from affinity columns of human cationic IgG myeloma proteins bearing anti-DNA idiotype (Id) markers 16/6, F4, 3I, and 8.12 for possible anti-Id (combining site) blocking activity.

METHODS

Anti-DNA idiotypic antibody activity was studied in 3 preparations of IVGG containing high, medium, and low levels of IgG anti-F(ab')2, and in 4 other commercial IVGG preparations. Affinity-purified IgG anti-DNA (APAD) from systemic lupus erythematosus (SLE) patients was biotinylated, and binding to DNA coated on enzyme-linked immunosorbent assay plates was used to measure anti-DNA antibody activity. IVGG was adsorbed to Sepharose 4B affinity columns linked to a panel of cationic human IgG myeloma proteins positive for anti-DNA Id markers 16/6, F4, 3I, and 8.12. Material adsorbing to such columns was eluted at low pH (2.5) and after neutralization, tested for its ability to inhibit biotinylated APAD reacting with DNA.

RESULTS

Only 0.05-0.9% of IVGGs bound firmly to Id affinity columns. These IVGGs were then eluted, using pH 2.5 glycine-saline and eluates neutralized to pH 7.4. Column flowthrough and eluate fractions were compared for their ability to block SLE APAD reacting with DNA. Significant inhibition of SLE APAD combining sites was observed with eluates from anti-DNA Id affinity columns; however, no correlation between IVGG anti-F(ab')2 activity and true anti-Id blocking of APAD was apparent. No residual anti-Id activity remained in column flowthrough fractions. No anti-Id blocking activity was recorded for IVGG eluates from human cationic myeloma columns devoid of the 4 anti-DNA Id markers. DNase treatment of IVGG or Id column eluates did not affect anti-Id blocking activity. Thus, all detectable anti-DNA idiotypic antibody capable of blocking SLE anti-DNA combining sites bound to Id+ affinity columns. Column eluates also showed some relative concentration of IgG anti-DNA activity, which was of lower affinity for DNA than antibodies also present in eluates which blocked anti-DNA combining sites.

CONCLUSION

The presence of both anti-DNA and antiidiotypic (anti-combining site) activity in human anti-DNA Id column eluates indicates that epibodies from IVGG are relatively concentrated when this strategy is used. This approach may lead to a new strategy for treatment of SLE nephritis.

VH gene use by HIV type 1-associated lymphoproliferations

The pathogenesis of polyclonal HIV-associated lymphomas lacking traditional B cell cofactors (i.e., Epstein-Barr virus [EBV] infection, c-myc translocations) is poorly understood. A multistep pathogenesis model has been proposed in which polyclonal lymphomas represent an earlier stage in HIV-associated lymphomagenesis before the emergence of a dominant malignant clone. Chronically present antigens have been proposed as a likely stimulus for polyclonal B cell proliferation; if so, polyclonal lymphoma-associated immunoglobulins (Igs) should have molecular evidence of somatic hypermutation, a process by which antibody affinity maturation in response to chronic antigenic stimulation occurs. Molecular analyses of Ig heavy chain variable (V(H)) gene use by B cells in a polyclonal HIV-associated large cell lymphoma lacking EBV and c-myc rearrangement was undertaken. Eighteen randomly selected clones generated from RT-PCR yielded 15 unique V(H) sequences, all of which were most homologous to only three previously identified germline V(H)1 genes. Two sets of clones (consisting of three and two clones, respectively) had identical V(H) gene sequences, and one pair of clones had identical third complementarity determining regions (CDR3s) but different V(H) gene sequences; eight clones were <95% homologous to their most related germline V(H)1 genes. We compared these results with Ig V(H)1 gene use by B cells present in a reactive hyperplastic lymph node obtained from an HIV-1-infected individual. Fifteen clones randomly selected from RT-PCRs yielded 15 unique V(H)1 sequences, all of which were most homologous to 5 previously identified germline V(H)1 genes; 10 clones were <95% homologous to their most related germline gene. Binomial probability analysis revealed that only 1 of the 15 unique V(H)1 sequences derived from the polyclonal lymphoma (i.e., 7%), as compared with 5 of 15 unique V(H)1 sequences derived from the reactive lymph node (i.e., 33%), had a low probability of occurrence by random chance (p < 0.05). These data provide molecular evidence of polyclonality in an HIV-associated polyclonal lymphoma, demonstrate a qualitative difference in somatic hypermutations of Ig V(H) genes associated with malignant versus reactive B cell lymphoproliferations, and support an antigen-mediated multistep pathogenesis model of HIV-1-associated lymphomagenesis.

Molecular, metabolic and immune evidence suggest that systemic autoimmune disease is antigen-mediated

Patients with systemic lupus erythematosus generate a sustained immune response against self. The tools of modern molecular biology have been applied to cell activities and elements/signals of the immune system, but a structural or regulatory defect has not been found. When deoxyribonucleic acids for autoantibodies were cloned and sequenced, they were like other autoantibody DNA sequences; when genetic materials for autoantibodies were inserted into transgenic mice, cells secreting the antibodies were subject to normal control mechanisms and eliminated. A failure to clear self-reactive antibody producing thymocytes has not been demonstrated in human systemic lupus erythematosus. Molecular analyses of the efferent side of the immune response have been largely normal in systemic lupus erythematosus. The structure of autoantibodies suggests that they have been generated by selection pressures and the presence of endogenous antigens. If the immune system attack on self was secondary, structural changes and metabolic reactions capable of generating antigens should be found in systemic lupus erythematosus cells. Structural changes have been found in deoxyribonucleic acid from phytohaemagglutinin-stimulated systemic lupus erythematosus lymphocytes in the form of S1 nuclease-sensitive deoxyribonucleic acid breaks. Altered cellular macromolecules could result from endogenous metabolic processes, particularly oxygen free radicals and arachidonic acid metabolites. Excess free-radical species, generating positive nitroblue tetrazolium-reacting material and positive chemiluminescence, have been found in most but not all phytohaemagglutinin-stimulated lupus lymphocyte samples. If endogenous metabolic processes act on endogenous deoxyribonucleic acid, endogenous cell DNA breakdown may lead to low molecular weight deoxyribonucleic acids and deoxyribonucleic acid/immune complexes in systemic lupus erythematosus sera that are potentially immunogenic. These combined findings suggest that the exaggerated immune responses of systemic lupus erythematosus may be a normal response to protect the host from a perceived antigenic threat.

Repertoires of antigen receptors in Tdt congenitally deficient mice

Tdt deficient mice show lack of N region in V(D)J junctions of immunoglobulin and T cell receptor genes and revealed that "immature recombinase" in fetal stage would boil down to no more than a lack of Tdt. Although particular junctions which are thought to be created by homology-mediated joining are frequently observed, one fourth of junctions lacked even one bp of overlap, indicating the existence of a V(D)J joining pathway that is homology independent. Lymphocyte repertoire which express VH81X gene without N region is negatively selected, which shows that the repertoire of Tdt deficient mice is not a truly fetal repertoire. Positive selection of thymocytes is more efficient in Tdt deficient mice. Furthermore Tdt-/- mice produce significant amounts of anti-dsDNA antibodies as Tdt+/+ mice, indicating that increased diversity of the third complementarity-determining region (CDR3) by Tdt is not essential for the expansion of precursor B cells programmed to produce anti-DNA antibodies.

Normal human cord blood B cells can produce high affinity IgG antibodies to dsDNA that are recognized by cord blood-derived anti-idiotypic antibodies

It is possible to identify, in Epstein-Barr virus-transformed normal human cord blood B cell populations, cells present at a low frequency that produce IgG antibodies specific for dsDNA. By cloning out these B cells as immortalized monoclonal cell lines, it could be shown that the antibodies were the products of CD5 positive B cells. Two monoclonal anti-dsDNA antibodies were derived from cell lines T52 and A7 and these were further characterized as anionic (pI approximately 6.4) IgG4 kappa antibodies that bound with affinities of 7.18 x 10(9) l/mol and 3.28 x 10(9) l/mol, respectively, to dsDNA but did not bind to ssDNA. These affinities were similar to those of polyclonal IgG anti-dsDNA antibodies from lupus patients, which ranged from 1 x 10(9) -8.9 x 10(10) l/mol. Both T52 and A7 monoclonal anti-dsDNA antibodies were recognized by cord blood-derived IgM antibodies. These IgM antibodies were not rheumatoid factors but bound to the F(ab')2 of A7 and T52 while failing to recognize T50, which is an autologous IgG4 kappa monoclonal antibody without specificity for dsDNA. A cloned B cell line A24 generated from the same cord blood sample as A7 produced an IgM monoclonal antibody that bound to the heavy chains of T52 and A7, but not T50 on Western blot and inhibited the binding of these antibodies to dsDNA. A7 and T52 competitively inhibited each other in their binding to the anti-idiotype A24, and A24 inhibited the binding to dsDNA of some polyclonal IgG anti-dsDNA antibodies purified from sera of lupus patients. The level of inhibition of binding of these antibodies to dsDNA was directly proportional to the levels of expression of the idiotype recognized by A24 on these antibodies. The normal human cord blood, therefore, may contain cells that form an idiotype/anti-idiotype network in which the idiotype is expressed on IgG antibodies with specificity for dsDNA and the anti-idiotype is an IgM antibody that binds to a heavy chain idiotope in such a way as to interfere with its interaction with dsDNA. The presence of a similar idiotype on some polyclonal anti-dsDNA antibodies in lupus that are similarly inhibitable by the cord blood-derived anti-idiotype raises the possibility that this network may persist in later life and perhaps become dysfunctional in systemic lupus erythematosus.

Repertoire cloning of a human IgG inhibitor of alpha IIB beta 3 function. The OG idiotype

A patient (OG) with Glanzmann thrombasthenia became refractory to platelet transfusion following the production of IgG antibodies (Ab1) specific for the integrin subunit beta 3. We generated recombinant VH and VL cDNA libraries using IgG-specific mRNA isolated from OG peripheral blood B-lymphocytes that had been selected for binding to antigen (alpha IIb beta 3 adsorbed to plastic dishes). These antigen-specific B-lymphocytes contain rearranged VH DNA segments that belong exclusively to the VH4 gene family. Recombinant Fab were expressed on the surface of filamentous phage coinfected with VH and VL segments cloned into the phagemid pHEN1 or the phage fd-tet-DOG1. To facilitate selection of the desired recombinant Ab1 Fab, we developed a rabbit polyclonal antibody specific for affinity-purified OG anti-beta 3 Fab (Ab2). Ab2 reacts specifically with Ab1, and this interaction is inhibited by purified alpha IIb beta 3. Following three rounds of phage selection on Ab2 adsorbed to plastic dishes and random reassociation of heavy and light chains, we isolated Ab1 Fab and tested their binding to alpha IIb beta 3. Five Id-positive Fab were selected for further characterization. These Fab use one of two VH genes (H21 or H23) complexed with one of three V lambda genes. Subsequent sequence data demonstrated that all three lambda genes are the same clone L22 which uses a germline V lambda gene segment. Fab using H23 bind to alpha IIb beta 3, while those using H21 do not. Based on sequence homology, both H21 and H23 use VH gene segments belonging to the VH4 gene family. Thus, the idiotype OG is restricted to the VH4 gene family and is the first sequenced prototype of human antibodies that bind close to or at a functional epitope(s) of alpha IIb beta 3.

Tropical rheumatology. Immunological aspects

The presence of auto-antibodies in infectious diseases continues to puzzle and provoke. It is hoped that sequencing studies in particular will yield further clues as to the role and mechanism of production of autoantibodies in infectious diseases. This, in turn, may also provide further insights into the role of auto-antibodies in auto-immune diseases. From a practical clinical viewpoint, the search for improved auto-antibody tests and new diagnostic markers with improved sensitivity and specificity must continue in the tropics. Until this is achieved, the results of auto-antibody tests in persons living in the tropics, persons from the tropics or patients with tropical infections, must be interpreted with caution.

Analysis of the V kappa I family: germline genes from an SLE patient and expressed autoantibodies

Our studies of anti-DNA antibodies in systemic lupus erythematosus have demonstrated a preferential use of the V kappa I family to encode light chains of antibodies that express the anti-DNA associated 3I idiotype. This idiotype is present on a high percentage of anti-DNA antibodies in approximately 80% of SLE patients1,2. In this study, we employed PCR to obtain V kappa I germline genes from a lupus patient in order to address the following questions: Do the V kappa I germline genes of an individual with autoimmune disease differ from those of healthy individuals? What V kappa I genes are used to encode autoantibodies and are they used to encode protective antibodies also? Does the V kappa I gene family display peculiarities in V gene segment rearrangement or somatic mutation? Our analysis shows that the coding region sequences of germline genes of an autoimmune individual are highly homologous to those of non-autoimmune individuals. In addition, the same germline genes can be utilized to encode antibodies to both exogenous and self antigens. While rearranged V kappa genes are ordinarily derived from the J kappa proximal region of the V kappa locus, V kappa I genes encoding autoantibodies derive primarily from the J kappa distal region. It is not yet clear if this applies equally to V kappa I encoded antibodies directed to foreign antigen.

Molecular characterization of monoclonal IgM derived from human B cell lines expressing the 4C9 rheumatoid factor associated idiotype

Ten human monoclonal B cell lines that express the RF associated Id 4C9 were analyzed using an immunogenetic approach. Five of eight tested lines were also strongly positive for the 6B6.6 Id. We found that all the 4C9/6B6.6 positive lines expressed VkIIIa light chain genes. In contrast, 4C9 reactivity was also found on a cell line expressing a VkIIIb light chain gene and on a line expressing a V light chain gene. The two anti-Ids recognized a linear light chain determinant on Humkv328 encoded light chains but also a conformational determinant on Vg encoded light chains that appeared to be dependent on the presence of a heavy chain. Idiotypic reactivity occurred on both RF positive and RF negative antibodies. Within this idiotypic system, the basis for idiotypic reactivity and RF reactivity is complex, subject to both heavy and light chain gene usage and sensitive to small numbers of somatic mutations.

How unique are pathogenic anti-DNA autoantibody V regions?

Unique characteristics of the molecular structure of V regions encoding pathogenic anti-DNA autoantibodies have become apparent upon comparison of a large number of nucleotide sequences encoding this autospecificity. Moreover, the generation of transgenic animals expressing V regions encoding anti-DNA autoantibodies has shed light on the tolerizing mechanisms that regulate B cells producing antibodies against DNA.

Mutational analysis of an autoantibody: differential binding and pathogenicity

We have used site-directed mutagenesis to change amino acid residues in the heavy chain of the pathogenic R4A anti-double-stranded DNA (dsDNA) antibody and have looked for resultant alterations in DNA binding and in pathogenicity. The data demonstrate that single amino acid substitutions in both complementarity determining and framework regions alter antigen binding. Changes in only a few amino acids entirely ablate DNA specificity or cause a 10-fold increase in relative binding. In vivo studies in mice of the pathogenicity of the mutated antibodies show that a single amino acid substitution leading to a loss of dsDNA binding leads also to a loss of glomerular sequestration. Amino acid substitutions that increase relative affinity for dsDNA cause a change in localization of immunoglobulin deposition from glomeruli to renal tubules. These studies demonstrate that small numbers of amino acid substitutions can dramatically alter antigen binding and pathogenicity, and that the pathogenicity of anti-DNA antibodies does not strictly correlate with affinity for DNA.

Autoantigen-specific T cell proliferation induced by the ribosomal P2 protein in patients with systemic lupus erythematosus

A role for helper T cells in the induction of pathogenic lupus autoantibodies is increasingly supported by data from studies of murine lupus and patients with systemic lupus erythematosus (SLE). However, the poor in vitro function of SLE T cells has hampered the identification and characterization of autoantigen-specific T cells. We used recombinant fusion proteins to study the T cell proliferative response of 31 lupus patients and 27 healthy subjects to a well-characterized SLE autoantigen, the ribosomal P2 protein. Although PBMC from SLE patients showed marked impairment in the proliferative response to the common recall antigen tetanus toxoid when compared with normal subjects, a significantly greater proportion of SLE patients (32%) than normal individuals (0%) showed a T cell response to a recombinant P2 fusion protein. When the SLE patients were subgrouped according to the presence of serum anti-P autoantibody, 7 of 10 anti-P antibody-positive patients, but 0 of 20 anti-P antibody-negative SLE patients, demonstrated > 2,000 cpm [3H]thymidine incorporation and a P2 stimulation index > 5. The specificity of the T cell proliferative response for the P2 protein was confirmed by studies using a second recombinant human P2 fusion protein and by the specific activation of P2-primed T cells by recombinant P2 in secondary cultures. Moreover, the T cell proliferative response to the P2 autoantigen was mediated by CD4-positive T cells and was inhibited by anti-MHC class II antibodies. These data demonstrate the presence of autoantigen-specific T helper cells in patients with SLE and suggest that these T cells drive the production of autoantibodies by B lymphocytes.

Structure of the VH and VL segments of monoreactive and polyreactive IgA autoantibodies to DNA in patients with systemic lupus erythematosus

Anti-DNA IgA autoantibodies play an important immunopathologic role in SLE patients. To analyze the cellular origin and the VH and VL structure of anti-DNA IgA autoantibodies, we generated five IgA1 mAbs to DNA using B lymphocytes from three SLE patients. Two mAbs bound to ssDNA only and one to both ssDNA and dsDNA (monoreactive antibodies). The remaining two mAbs bound to DNA (one to ssDNA and the other to both ssDNA and dsDNA) and to other self and foreign Ag (polyreactive antibodies). The IgA mAb relative avidity for DNA ranged from 7.5 x 10(-8) to 8.0 x 10(-10) g/microliters. The anti-DNA IgA mAb used VH segments of the VHI(VI-3b), VHII (VH2-MC2), VHIII (WHG16G and VH26c), and VHIV (V71-2) families in conjunction with V kappa I, V kappa IIIb, or V lambda I segments. All IgA mAb VH segments were juxtaposed with JH4b segments. The heavy chain CDR3 sequences were divergent in composition and length. When compared with those of the closest reported germ line genes, the IgA mAb VH and VL gene sequences displayed a number of differences. That these differences represented somatic point mutations was formally proved in both the monoreactive IgA mAb 412.67.F1.3 and the polyreactive IgA mAb 412.66.F1 VH segments by differential PCR amplification and cloning and sequencing of genomic DNA from the mAb-producing cell lines and autologous polymorphonuclear cells. The sequences of the germ line genes that putatively gave rise to the mAb 412.67.F1.3 and mAb 412.66.F1 VH segments were identical with those of the WHG16G and VH26c genes, respectively. In not only the monoreactive mAb 412.67.F1.3 but also the polyreactive mAb 412.66.F1 and mAb 448.9G.F1 VH segments, the higher concentration of replacement (R) mutations and the higher R:S (silent) mutation ratios in the complementarity-determining region (infinity; 19:0) than in the framework region (1.0) (p = 0.00001, chi 2 test) were highly consistent with selection by Ag. In the five IgA mAb VH and VL segments, the putative and verified somatic point mutations yielded 68 amino acid replacements, of which 38 were nonconserved. Twenty of these yielded positively charged or polar residues that play a major role in DNA binding, including seven Arg, five Lys, three Tyr, two Gln, two His, and a Thr. The conserved amino acid changes included seven Asn. These findings suggest that anti-DNA IgA autoantibodies use a broad selection of VH and VL genes and enhance their fit for Ag by undergoing somatic hypermutation and Ag selection.(ABSTRACT TRUNCATED AT 400 WORDS)

Germline-encoded IgG antibodies bind mouse cartilage in vivo: epitope- and idiotype-specific binding and inhibition

Autoantibodies specific for type-II collagen (CII) occur in mice and rats with collagen-induced arthritis (CIA). The binding in vitro and in vivo of mouse monoclonal antibodies (MoAbs) specific for separate epitopes in CII have been investigated. Two-day-old mice were injected intraperitoneally (i.p.) with the anti-CII antibody CIID3 in both unlabelled and biotinylated form. It was found that antibodies binding to the same epitope in CII in vivo can inhibit others from binding in an epitope-specific fashion. The binding in vivo and in vitro of anti-CII antibodies could be inhibited also by an anti-idiotypic rat antiserum produced against the D3 antibody. The anti-idiotypic antiserum inhibited the binding of the antibody D3 and the idiotypically related antibody C2. The cDNA's of anti-CII antibodies D3, C2, and F4 were sequenced and found to contain germline encoded V-genes, apparently without somatic mutations. The variable heavy chain of D3 and C2 both expressed the same VH rearrangement, confirming that they share idiotypes. This report demonstrates that CII-specific germline-encoded IgG autoantibodies bind specifically to normal cartilage in vivo via their combining site.

Rheumatoid factors from the peripheral blood of two patients with rheumatoid arthritis are genetically heterogeneous and somatically mutated

We report the DNA sequences of the heavy and light chain immunoglobulin genes of 11 monoclonal rheumatoid factor (RF)-secreting lines derived from the peripheral blood of two patients with rheumatoid arthritis (RA). It is evident from immunogenetic analysis of these lines that RA-associated RF activity can arise from a wide variety of heavy and light chain genes and gene combinations. Although the RF response from our two patients shows a bias in gene usage toward those genes used to encode monoclonal RF, particularly VkIII, relatively few of these RFs are reactive with the monoclonal antiidiotypes 6B6.6 and 17.109 that define VkIII germline-encoded light chains and the loss of this idiotypic reactivity is clearly related to somatic mutation. Finally, RFs derived from peripheral blood of RA patients show a similar heterogeneity of epitope binding to Fc as that seen for synovium-derived RF and some are clearly different in binding specificity from the restricted RF population found in patients with B cell malignancies. Somatic mutations as well as different VH/VL combinations contribute to the heterogeneity in the binding patterns of these RA-derived RF.

On the biological origin of anti-double-stranded (ds) DNA antibodies: systemic lupus erythematosus-related anti-dsDNA antibodies are induced by polyomavirus BK in lupus-prone (NZBxNZW) F1 hybrids, but not in normal mice

We have recently demonstrated that polyomavirus BK and isolated BK double-stranded (ds)DNA have a strong potential for induction of anti-dsDNA antibodies. Here, data are presented that demonstrate that normal mice (a term used in this report for mice not predisposed to a lupus-like syndrome) of four different strains responded to both BK virus and BK dsDNA by producing transient antibodies binding preferentially to the viral dsDNA itself. These antibodies did not bind in the Crithidia luciliae assay, and did not seem to be of pathogenic significance, as neither signs of proteinuria nor immunochemical signs of glomerulonephritis developed in these mice. In contrast, 5-week-old (NZBxNZW)F1 mice developed strong and persistent anti-dsDNA antibodies in response to BK virus and BK dsDNA, with similar features to those of anti-dsDNA antibodies from individuals with systemic lupus erythematosus: they reacted strongly in the Crithidia luciliae assay and cross-reacted with viral as well as with mammalian dsDNA. Furthermore, persistent proteinuria and glomerulonephritis, with demonstrable heavy mesangial deposits of immune complexes containing IgG anti-dsDNA antibodies, developed 2-3 months earlier than in spontaneously autoimmune control mice. The relevance of these observations to a viral origin of anti-dsDNA antibodies in lupus is discussed.

The preferential expression of the anti-DNA associated 8.12 idiotype in lupus is not genetically controlled

Anti-DNA antibodies are autoantibodies unique to systemic lupus erythematosus. Studies of their structure have demonstrated cross reactive idiotypes present in genetically unrelated individuals. Despite much research, it is still not clear what triggers their production and what governs the presence of particular idiotypic determinants in their structure. To study the role of genetic and environmental factors in the expression of idiotype, we analyzed sera of SLE patients, their family members and nonautoimmune individuals vaccinated with pneumococcal polysaccharide, for the presence of the 8.12 idiotype, which is present on lambda light chains of anti-DNA antibodies. Elevated titers of the 8.12 idiotype was found in the serum of 57% of SLE patients. Elevated titers were present in only 9% of family members, and always associated with the presence of high levels of IgG anti-DNA antibodies. Following vaccination with pneumococcal polysaccharide, 8.12 reactive anti-pneumococcal antibodies were produced by 7 of 10 non-autoimmune individuals and 8.12 reactive anti-DNA antibodies by one. These results suggest that 8.12 reactive antibodies are antigen driven and bind structurally related antigens, but there is no evidence that expression of this idiotype is genetically controlled.

Expansion of a B-lymphocyte clone producing IgM auto-antibodies encoded by a somatically mutated VHI gene in the spleen of an autoimmune patient

Seventy-six human B-cell hybridomas were obtained by fusing B lymphocytes from the spleen of a patient with chronic idiopathic thrombocytopenia (ITP). Two independent hybridoma clones producing IgM autoantibodies reacting with platelets and other antigens from both the internal and the external environments were established from this fusion experiment. The IgM autoantibodies produced by the two hybridoma clones were found to be encoded by identical VHDJH and VLJL genes. The comparison of the VHI gene expressed in both hybridomas with the germline equivalent cloned from the patient's DNA showed two somatic mutations in the complementarity-determining regions CDR1 and CDR2 resulting in amino acid replacements. These data suggest the selection and expansion of an autoantibody-producing B-cell clone in the spleen of an ITP patient, probably as a result of (auto)antigen-driven stimulation.

Antibodies to dsDNA are produced during primary BK virus infection in man, indicating that anti-dsDNA antibodies may be related to virus replication in vivo

Experimental immunizations with both the Polyomavirus BK and with the isolated viral genomic dsDNA regularly induce antibodies with a relative affinity for BK virus dsDNA. In the present study we demonstrate that the anti-dsDNA responses to BK virus in experimental animals also appear during natural BK virus infection in man. Fifty-nine children were examined over time for serological signs of primary BK virus infection. Of eight children found to undergo primary infection with BK virus, anti-BK dsDNA antibodies appeared in all. In 4 of the 8 patients the antibodies cross-reacted significantly with mammalian dsDNA, and weak cross-reactions were also noted in at least three other patients. The antibodies resembled those induced in the experimental model with regard to their relative affinity for BK dsDNA. In contrast, most, but not all, anti-dsDNA antibodies from 10 SLE patients cross-reacted extensively with dsDNA from viral and mammalian origin. Thus, a dsDNA virus like BK virus may provoke immunological intolerance to dsDNA, but, with qualities different from those produced during SLE. The present observations demonstrate that induction of anti-dsDNA antibodies is not restricted to experimental immunization of animals, but does also take place in humans during naturally acquired BK virus infection. The relevance of this model for the spontaneous production of anti-dsDNA antibodies is discussed.

Molecular characterization of a cross-reactive idiotope on human immunoglobulins utilizing the VH4-21 gene segment

The anti-idiotypic (anti-Id) antibody (Ab) 9G4 binds a cross-reactive idiotope (CRI) present in a select group of human autoantibodies. This Id has been localized to the portion of immunoglobulin (Ig) heavy (H) chains encoded by the VH4-21 gene segment, a member of the human VH4 family. This gene segment is utilized by essentially all cold agglutinin (CA) Abs with I/i specificity isolated from patients with CA disease stemming from chronic lymphoproliferative disorders. In this study, mutational analysis of a CA has been used to determine the structural basis for 9G4 binding to Abs utilizing the VH4-21 gene segment. Recombinant CA H chain mutants were produced and their 9G4 reactivity determined. Mutants were generated by exchanging VH4-21 sequences in the FR1, CDR1, and CDR2 with corresponding sequences from a closely related gene segment V71-2, a VH4 family member that is associated neither with Abs having CA activity nor with Abs that react with 9G4. The results indicate that the motif AVY at amino acid positions 23-25 in FR1 defines the 9G4 idiotope. Reaction of these recombinant Abs with a polyclonal rabbit anti-CA antiserum absorbed to render it specific for a CA CRI also maps predominantly to FR1. These findings indicate that the solvent-exposed FR1 plays an important role in eliciting an immune response to Igs.

Evidence for antigen driven selection in two monoclonal auto-antibodies derived from nonobese diabetic mice

The nonobese diabetic (NOD) mouse is a model of human type I diabetes. This diabetes is due to massive infiltration of the pancreatic beta cell of islets by autoreactive T cells (insulitis) followed by the destruction of insulin-producing cells. Circulating autoantibodies are also detected, notably against glutamic acid decarboxylase, peripherin and insulin. Two monoclonal autoantibodies directed against insulin and peripherin were obtained by fusing NOD spleen and myeloma cells. We report here the nucleotide sequence of the genes encoding for the V regions of these two antibodies. Somatic mutations were identified by comparing the light chain nucleotide sequence of one of these autoantibodies with its germline counterpart precursor established from NOD mice after PCR gene amplification. The other one displays N additions on both sides of the D region. These results strongly suggest that both autoantibodies have undergone diversification, either N additions or somatic mutations, and therefore present structural features of antibodies derived from animals immunized against exogenous antigens.

Cloning of a human immunoglobulin gene fragment containing both VH-D and D-JH rearrangements: implication for VH-D as an intermediate to VH-D-JH formation

In an Epstein-Barr virus-transformed human B cell line we found an unusual immunoglobulin heavy chain gene rearrangement. Restriction mapping and sequencing analysis led us to conclude that VH-D and D-JH recombination took place in a single allele. Both VH-D and D-JH complexes still had their recombination signal sequences adjacent and the DNA sandwiched by these two complexes retained a germ-line configuration, suggesting the potential for a secondary rearrangement resulting in a VH-D(-D)-JH formation. With this finding, we propose a novel pathway, in which the VH-D complex is an intermediate in the formation of a functional VH exon.

Polyreactivity is a property of natural and disease-associated human autoantibodies

Polyreactivity was earlier recognized as a feature of naturally expressed autoantibodies in serum. In the present study, we have compared the reactivity on a panel of self antigens of affinity-purified anti-DNA and anti-thyroglobulin (TG) IgG autoantibodies from the serum of patients with systemic lupus erythematosus (SLE) and autoimmune thyroiditis with their affinity-purified counterparts isolated from the serum of healthy individuals. Anti-DNA autoantibodies exhibited a similar degree of polyreactivity whether originating from patients or from healthy adults. Natural anti-TG autoantibodies were also found to be polyreactive. Anti-TG autoantibodies from patients with Hashimoto's thyroiditis showed little or no polyreactivity. Natural anti-TG autoantibodies were equally polyreactive whether or not they belonged to a fraction of normal IgG that is connected through V regions with other IgG molecules from the same source. These results indicate that polyreactivity of autoantibodies is a feature that does not allow one to distinguish between natural and disease-associated autoantibodies as well as between V-region-connected and unconnected autoantibodies.

Variable region sequences of autoantibodies from mice with experimental systemic lupus erythematosus

We have sequenced nine monoclonal antibodies (mAb) derived from C3H.SW mice in which experimental systemic lupus erythematosus (SLE) was induced. The hybridomas were selected for binding to DNA or to HeLa nuclear extract (NE). Three mAb were found to bind DNA, and are shown to exhibit sequence characteristics of pathogenic anti-DNA antibodies. One, mAb 2C4C2, is shown to use a heavy chain V region gene (VH) identical to the VH of anti-DNA mAb isolated from other lupus-prone mice, namely (NZB x NZW)F1. The light chain V region gene (VL) of mAb 2C4C2 is 98% homologous to the VL of another anti-DNA mAb, also isolated from (NZB x NZW)F1 mice. The other two anti-DNA mAb, 5G12-4 and 5G12-6, share 93% of their VH sequences with that of mAb 2C4C2. Six mAb bound proteins of HeLa NE. Four of these six antibodies were found to use the VH124 VH and V-L7 VL. The nine mAb use a total of five VH and four VL germ-line genes, demonstrating that the autoantibodies induced in mice with experimental SLE do not originate from one B cell clone. Three of these nine VH and VL were identical in sequence to germ-line genes, while at least three others had somatic mutations. The latter suggests that the above autoantibodies arise in mice by both usage of existing (pre-immune) B cells, and through an antigen-driven process. Furthermore, it appears that autoantibodies found in mice with experimental SLE use genetic elements similar to those used by mAb that were isolated from mouse strains which develop lupus spontaneously.

Human B-cell clones expressing lupus nephritis-associated anti-DNA idiotypes are preferentially expanded without somatic mutation

Human monoclonal anti-single/double-stranded (ss/ds) DNA antibodies (NE-1 and NE-13) expressed cross-reactive idiotypes (Id), NE-1 Id, which have been detected on the lupus glomeruli-deposited anti-DNA antibodies. The nucleotide sequences of the variable regions of NE-1 and NE-13 clones were analogous except for one nucleotide difference in the Vk region. The VH and Vk gene segments of NE-13 clone were identical with germline genes VH4.21 and Vb (or Vb'), respectively. CDR3s of NE-1 and NE-13 heavy chains were arginine rich and CDR1s contained an amino acid stretch, SGYY, the inverted sequence of YYGS, which was shared among CDR3s of several anti-DNA antibodies. Clonal frequency analysis using a limiting dilution method revealed that NE-1 Id-positive clones at precursor cell level increased in lupus patients. These findings suggest that some IgM anti-DNA clones which express NE-1 Id associated with lupus nephritis use germline genes without mutation and they may be preferentially expanded at the precursor cell levels as well as at the mature cell level.

Immunization of BALB/c mice with a monoclonal anti-DNA antibody induces an anti-idiotypic antibody reactive with a cell-surface DNA binding protein

DNA binds to cell-surface proteins on human and murine leukocytes and induces secretion of the cytokine interleukin 6 (IL-6). Cell-surface DNA binding molecules have been shown to serve as target antigens for the production of autoantibodies in patients with systemic lupus erythematosus (SLE), and in lupus-prone mice. Recent studies have demonstrated that a subset of anti-anti-DNA antibodies, isolated from patients with SLE, are idiotypically related to antibodies reactive with a cell-surface DNA binding molecule. We now report that immunization of normal mice with a murine monoclonal anti-DNA antibody induces an anti-idiotypic response which has reactivity with a cell-surface DNA binding molecule. An anti-idiotypic anti-DNA monoclonal antibody (LB17) was isolated from the spleen of an immunized mouse. This monoclonal antibody blocked the binding of DNA to murine splenocytes and mimicked the functional effect of DNA by stimulating the secretion of IL-6. These experiments provide further evidence for an idiotypic connectivity between antibodies to cell-surface DNA binding proteins and anti-DNA antibodies. It is hypothesized that this idiotypic system is part of the network of natural autoantibodies and that its perturbation may give rise to pathogenic antibodies.

Lupus-derived human monoclonal IgM anti-DNA antibody displays monospecificity, high affinity and private idiotype specificity

A human monoclonal IgM k anti-DNA antibody, designated 2F7, was prepared by somatic hybridization of peripheral blood lymphocytes from a lupus patient with a human-mouse heterohybridoma cell line, K6H6/B5. 2F7 was tested for its antigen binding and idiotypic specificity by direct binding and inhibition enzyme-linked immunosorbent assays. 2F7 had a high binding activity to single-stranded DNA (ssDNA) but not to double-stranded DNA. It cross-reacted with single-stranded homopolymers with pyrimidine bases and double-stranded polynucleotides containing those homopolymers, suggesting that 2F7 recognizes a conformational determinant made up of both deoxyribose-phosphate backbone and specific nucleotide base. 2F7 did not cross-react with eight structurally unrelated self-antigens. Dissociation constant (Kd) of 2F7 for sonicated ssDNA was approximately 4.5 x 10(-9) M, indicating its relatively high affinity. Idiotypic characterization with rabbit anti-idiotype raised against 2F7 suggested that 2F7 expressed an idiotype at or near its antigen-binding sites that was not detected in sera from 20 unrelated lupus patients, 10 lupus family members and 10 normal individuals. These results suggest that certain IgM class anti-DNA antibodies in human systemic lupus erythematosus may arise by antigen stimulation and not simply by polyclonal B-cell activation.

Sequence analyses of three immunoglobulin G anti-virus antibodies reveal their utilization of autoantibody-related immunoglobulin Vh genes, but not V lambda genes

Accumulated sequence analyses of the antibody repertoire have revealed that most autoantibodies and developmentally regulated antibodies share a small set of germline Ig-variable region (V) genes. The findings have prompted speculation that certain autoantibodies are of developmental importance and may be instrumental in maintaining homeostasis of the adult antibody repertoire. In order to evaluate this hypothesis critically, it is first necessary to determine the V gene usage in human antibodies against foreign substances. Unfortunately, only a few such antibodies have had their heavy and light chains characterized. To rectify the situation, we adapted the anchored polymerase chain reaction to clone and analyze rapidly the expressed V genes for three anti-virus IgG antibodies. The results show that all three heavy chain V (Vh) genes are highly homologous to the known autoantibody-related Vh genes. In contrast, two light chain V (VL) genes of the V lambda 1 subgroup are similar to a non-autoantibody-related germline V lambda 1 gene. Taken together with the reported Vh and VL sequences of several antibodies against viruses and bacteria, the data show that many antipathogen antibodies may use the same small set of Vh genes that encode autoantibodies, but diverse VL genes that are distinct from autoantibody-related VL genes. Thus, only a small portion of the potentially functional germline Vh genes are used recurrently to generate most antibodies in a normal antibody repertoire, regardless of their reactivities with either self or non-self.

Antibodies to eukaryotic, including autologous, native DNA are produced during BK virus infection, but not after immunization with non-infectious BK DNA

The contemporary view concerning the origin of anti-dsDNA antibodies is that eukaryotic dsDNA is not immunogenic. Results presented here, however, show (1) that inoculation of rabbits with BK virus elicits antibodies to eukaryotic, including autologous, dsDNA, (2) that the transition from a non-immunogenic to an immunogenic state of autologous dsDNA depends on productive infection with BK virus, and (3) that inoculation with protein-free circular BK dsDNA initiates both infection in vivo and production of antibodies to autologous dsDNA. Non-infectious linearized BK dsDNA did not elicit any anti-dsDNA antibodies, while the same DNA molecule, when complexed with methylated bovine serum albumin, elicited anti-dsDNA antibodies solely recognizing BK dsDNA. Neither of the two linearized BK dsDNA preparations initiated infection. Using two different techniques, we could demonstrate that two separate sets of anti-dsDNA antibodies were produced during viral infection; one recognizing BK dsDNA, and the other recognizing autologous dsDNA. Thus, in contrast to previous assumptions, autologous dsDNA may be immunogenic. Based on the present results, we propose that autologous dsDNA can be rendered immunogenic through complex formation with viral DNA binding protein(s) such as the structural protein VP1 or the tumour antigen T. Such DNA-protein complexes may bypass a putative T-cell tolerance to autologous dsDNA.