The clonal analysis of anticardiolipin antibodies in a single patient with primary antiphospholipid syndrome reveals an extreme antibody heterogeneity.

The mechanism underlying the prothrombotic state that characterizes the primary antiphospholipid syndrome proves to be difficult to define mainly because of the variety of the phospholipid and protein targets of antiphospholipid antibodies that have been described. Much of the debate is related to the use of polyclonal antibodies during the different antiphospholipid assays. To better describe the antiphospholipid antibodies, a strategy was designed to analyze the reactivity of each one antibody making up the polyclonal anticardiolipin activity, breaking down this reactivity at the clonal level. This was performed in a single patient with primary antiphospholipid syndrome by combining (1) the antigen-specific selection of single cells sorted by flow cytometry using structurally bilayered labeled anionic phospholipids and (2) the cloning of immunoglobulin (Ig) variable (V) region genes originating from individual IgG anticardiolipin-specific B cells by a single-cell polymerase chain reaction technique. The corresponding V regions were cloned in order to express human recombinant antibodies in insect cells by a baculovirus expression system. The molecular analysis, the fine specificity, and the protein cofactor dependency of the first 5 monoclonal IgG anticardiolipins are reported here. This clonal analysis reveals the extreme heterogeneity of these antibodies, which could account for the difficulties in the previous attempts to define the pathogenic antiphospholipid response. This approach should help to unravel the complex antiphospholipid immune response and the mechanism of the prothrombotic state associated with these antibodies, but it could also shed some light on their possible origins. (Blood. 2001;97:3820-3828)

Salivary gland lymphomas in patients with Sjögren's syndrome may frequently develop from rheumatoid factor B cells.

OBJECTIVE: Patients with Sjögren's syndrome (SS) have an increased risk of developing monoclonal B cell non-Hodgkin's lymphomas (MNHL), which frequently occur in the salivary glands (SG). The transition from the benign lymphocyte infiltrate of the gland that characterizes SS to MNHL is not well understood. Previous sequence analyses of the expressed variable (V) region genes have supported the theory that the surface Ig (sIg) plays an important role in the initial expansion of nonmalignant B cell clones and in lymphomagenesis. However, the antigenic specificities of these B cells were unknown. We describe the specificities of the Ig expressed by 2 cases of MNHL that developed in the SG of 2 patients with SS. METHODS: The expressed V genes were amplified by polymerase chain reaction from biopsy specimens, sequenced, and subcloned into eukaryotic expression vectors. The constructs were transfected into P3X63-Ag8.653 cells to obtain 2 monoclonal cell lines, each secreting 1 of the sIg expressed by the MNHL. These IgM were tested by enzyme-linked immunosorbent assay and immunofluorescence against a panel of antigens potentially implicated in SS. RESULTS: Our main finding was that the Ig products of the neoplastic B cells were rheumatoid factors (RF). Contrary to expectations, they did not react with nuclear or cytoplasmic antigens, double-stranded DNA, self antigens commonly bound by natural autoantibodies, or SG tissue. CONCLUSION: Previous analyses of V gene use have provided indirect evidence that SG MNHL may frequently express RF. We demonstrate that this hypothesis is true in the 2 patients we studied. Large-scale studies will be needed to establish the exact frequency of RF specificity among SS-associated MNHL.

Autoantibodies in mice lacking terminal deoxynucleotidyl transferase: evidence for a role of N region addition in the polyreactivity and in the affinities of anti-DNA antibodies.

The generation of terminal deoxynucleotidyl transferase knockout mice (TdT0) has demonstrated that TdT is the only major activity involved in N region addition. This enzyme generates diversity by adding random nucleotides at the V-D-J junctions and by disrupting the formation of repetitive "homology-directed" junctions. Several studies have demonstrated that the Ig heavy chain third complementarity-determining region (H-CDR3) and the N region play a critical role: 1) in distinguishing between polyreactive and monospecific combining sites in natural and Ag-induced Abs; and 2) in the specificity and polyreactivity of natural autoantibodies (autoAbs) and in particular of anti-DNA Abs. To examine the impact of the lack of TdT on the natural autoAb repertoire in adult mice, we have stimulated TdT0 and TdT+ littermates with LPS. Serum studies demonstrate that TdT is not critical for the generation of B cells expressing autoAbs including anti-DNA Abs and rheumatoid factors. However, the generation of a large collection of hybridomas indicates that the frequencies of these cells are reduced in TdT0 mice mainly due to a lower incidence of polyreactivity; also, the lack of N region diversity seems to negatively affect the affinity of anti-DNA Abs. The physiologic relevance of these data is discussed.

A minor group of rheumatoid factors isolated from a patient with rheumatoid arthritis is derived from somatically mutated Vk1 genes further evidence that rheumatoid factors during autoimmune diseases undergo an antigen driven maturation.

To better understand the structural basis for rheumatoid factor [RF] activity and the origin of autoantibodies in human autoimmune diseases, we isolated the RF producing B cells from the peripheral blood and from the synovial fluid of a patient suffering from rheumatoid arthritis [RA]. We previously demonstrated that a significant fraction of these RF were derived from three V kappa III genes known to encode most of the monoclonal RF light chain variable regions. To get more insight into the actual repertoire of RF-V kappa genes during RA, we analyzed the nucleotide sequences of RF light chain variable regions of other V kappa families. Using two sets of polymerase chain reactions in order to amplify the cDNA derived from RF producing cells from the same patient KRA, we isolated only three different rearranged V kappa-J kappa complexes: slkv5, slkv7 and bkv42, all derived from V kappa I germ-line genes not previously known to be associated with RF activity; this suggests that the repertoire of VL genes coding for RF during RA is more diverse than the one involved in the generation of paraprotein RF during monoclonal lymphoid proliferations, although there remains a possible bias in favor of the V kappa III family. Moreover, each of these genes is somatically mutated with a pattern suggesting a selective pressure of the antigen. Particularly interesting is the additional proline residue at the V kappa-J kappa junction of bkv42, an unorthodox feature that we found previously in more than 50% of RF V kappa III-J kappa gene complexes. Finally, the homogeneity of some non conservative mutations suggests the existence of a restricted set of pathogenic epitopes driving the production of RF during RA.

Molecular analysis of the V kappa III-J kappa junctional diversity of polyclonal rheumatoid factors during rheumatoid arthritis frequently reveals N addition.

Much interest was stirred in recent years by the evidence that rheumatoid factors (RF) variable regions are encoded by a restricted set of V genes, with little or no somatic mutations, that are often overexpressed in the fetal repertoire. This is reminiscent of what has been observed for natural autoantibodies. However, these data come from studies of monoclonal RF (mRF) isolated from patients with lymphoproliferative disorders who usually do not present autoimmune symptoms. The molecular characterization of RF during autoimmune diseases such as rheumatoid arthritis (RA) has been hampered for some time because of their polyclonality; recently using the polymerase chain reaction method, we have demonstrated that RF kappa variable regions from a patient with RA were encoded by V kappa III genes known to code for mRF but that these genes had undergone somatic mutations with a pattern suggesting an antigen-driven maturation. Because an important role of the light chain third complementarity-determining region (CDR3) in anti-IgG reactivity and idiotype expression has already been suspected for RF, we now report the molecular characterization of the junction regions of these rearranged V kappa gens. Surprisingly, our data show that in 55% of the cases there is addition of a proline and/or glycine amino acid residue at the recombination site between V kappa and J kappa. The sequence analysis of our patients' germ-line Vg and J kappa 4 genes segments and their flanking regions demonstrates that the additional codons are not readily explicable by recombination between germ-line sequences and probably result from an N addition process. Since we could not find such an additional codon in 15 previously published mRF kappa chains we suggest that "pathogenic" RF during RA and mRF derive from different, although overlapping, B cell subsets. Moreover, since additional codons at the recombination site of V kappa and J kappa seem exceptional in expressed human kappa chains and because the resulting amino acid residue is a proline in most cases, we think that RF kappa chain CDR3 is under a very strong selective pressure during RA.

The majority of peripheral blood monoclonal IgM secreting cells are CD5 negative in three patients with mixed cryoglobulinemia.

The mixed cryoglobulinemia is considered to be a nonmalignant human B-cell proliferation that frequently produces a monoclonal IgM with anti-IgG activity (rheumatoid factor). Using murine monoclonal anti-idiotypic antibodies specific for private or minor idiotopes on monoclonal IgM from three patients suffering from nonmalignant mixed cryoglobulinemia, we investigated the presence of the CD5 antigen on the monoclonal IgM producing cells in these patients. It is shown by two-color cytofluorometric analysis that the majority of the peripheral blood monoclonal IgM rheumatoid factor secreting cells is CD5 negative in these three patients. One of the monoclonal rheumatoid factor K variable regions was sequenced at the protein level and belongs to the human VK III group, as a high proportion of monoclonal rheumatoid factors and some B-cell chronic lymphocytic leukemia (CLL) membrane bound Igs. Thus, despite the preferential use of similar VK genes and the absence of somatic mutation affecting these variable regions in both malignant B-cell CLL and nonmalignant mixed cryoglobulinemia, these proliferating B cells differ in the CD5 membrane expression.

Monoclonal rheumatoid factor-secreting cells in a patient with mixed cryoglobulinemia. Homogeneity and stability of the idiotypic production and in vitro idiotypic suppression.

The potential therapeutic value of anti-idiotypic antibodies during B cell proliferations largely depends on the stability of the target Ig idiotopes. We investigated this stability in a clinical condition of so called nonmalignant monoclonal B cell proliferation, mixed cryoglobulinemia. The idiotypic profile of a single IgM kappa monoclonal auto-antibody with anti-IgG activity (rheumatoid factor (RF] which originated from a patient suffering from a nonmalignant mixed cryoglobulinemia was followed over a period of 3 yr. As judged from the reactivity of a panel of five different mouse monoclonal anti-idiotypic antibodies mapping the RF variable regions, there was no idiotypic change in the serum IgM RF. At a cellular level, in vitro stimulation of the patient's PBL gives rise to IgM kappa auto-antibodies that were shown to bear the same idiotypic determinants as the serum IgM kappa. We then investigated the effects of the anti-idiotypic antibodies on the in vitro IgM kappa production. When stimulated with PWM and in the presence of anti-idiotypic antibodies (10 micrograms/ml), the patient's PBL produced less IgM RF (18 to 62% inhibition). The same inhibition of IgM RF production was observed after EBV infection of the patient's PBL (from 19 to 90% inhibition). In both cases, the remaining IgM RF production was idiotypically indistinguishable from the serum IgM RF. The implications of the idiotypic stability and of the results of in vitro idiotypic manipulation could be important in view of both the understanding of nonmalignant cryoglobulinemia and of the possible therapeutic use of anti-idiotypic antibodies in diseases.