Preferential utilization of a novel V lambda 3 gene in monoclonal rheumatoid factors derived from the synovial cells of rheumatoid arthritis patients

On the origin of rheumatoid factors: Insights from analyses of variable region sequences

OBJECTIVES

Rheumatoid factors (RFs) are thought to play an important role in rheumatoid arthritis (RA), but are also found in healthy donors (HDs). Previous studies examined variable region sequences of these autoantibodies at a time when knowledge of the human germline repertoire was incomplete. Here we collected and analyzed RF sequence data from the literature to elucidate how RFs develop and whether their characteristics differ between RA patients and HDs.

METHODS

A database was built containing nucleotide sequences of RF heavy and light chain variable domains and characteristics including affinity, isotype and specificity, all collected from published papers. Gene usage and mutation frequencies were analyzed using IMGT/HiV-QUEST. Selection strength was assessed with the BASELINe tool.

RESULTS

Sequences were retrieved for 183 RF clones (87 RA; 67 HDs; 29 other). No biased gene usage was observed for RA and HDs. However, there does appear to be skewed gene usage in RFs from patients with mixed cryoglobulinemia. Mutation frequency varies considerably between RFs, and isotype-switched clones have significantly more mutations. Monospecific RFs carry more mutations than polyspecific RFs; no difference was found for RA- versus HD-derived RFs. Overall, reported affinity is low (median 1 µM), with a non-significant trend toward higher affinity of RA-derived RFs. Mutation frequency and affinity did not appear to be correlated. BASELINe analysis suggests an overall lack of positive selection and less negative selection strength in RA-derived RFs.

CONCLUSIONS

RFs derived from RA patients have similar properties as those derived from HDs. The RF response can be characterized as a moderately matured autoantibody response, with variable levels of somatic hypermutation, but low affinity.

Among B cell non-Hodgkin's lymphomas, MALT lymphomas express a unique antibody repertoire with frequent rheumatoid factor reactivity

We analyzed the structure of antigen receptors of a comprehensive panel of mature B non-Hodgkin's lymphomas (B-NHLs) by comparing, at the amino acid level, their immunoglobulin (Ig)V_H-CDR3s with CDR3 sequences present in GenBank. Follicular lymphomas, diffuse large B cell lymphomas, Burkitt's lymphomas, and myelomas expressed a CDR3 repertoire comparable to that of normal B cells. Mantle cell lymphomas and B cell chronic lymphocytic leukemias (B-CLLs) expressed clearly restricted albeit different CDR3 repertoires. Lymphomas of mucosa-associated lymphoid tissues (MALTs) were unique as 8 out of 45 (18%) of gastric- and 13 out of 32 (41%) of salivary gland-MALT lymphomas expressed B cell antigen receptors with strong CDR3 homology to rheumatoid factors (RFs). Of note, the RF-CDR3 homology without exception included N-region–encoded residues in the hypermutated IgV_H genes, indicating that they were stringently selected for reactivity with auto-IgG. By in vitro binding studies with 10 MALT lymphoma–derived antibodies, we showed that seven of these cases, of which four with RF-CDR3 homology, indeed possessed strong RF reactivity. Of one MALT lymphoma, functional proof for selection of subclones with high RF affinity was obtained. Interestingly, RF-CDR3 homology and t(11;18) appeared to be mutually exclusive features and RF-CDR3 homology was not encountered in any of the 19 pulmonary MALT lymphomas studied.

Immunoglobulin Vlambda light chain gene usage in patients with Sjögren's syndrome

OBJECTIVE

To determine whether patients with Sjögren's syndrome (SS) have abnormalities in Ig Vlambda and Jlambda gene usage, differences in somatic hypermutation, defects in selection, or indications for perturbations of B cell maturation.

METHODS

Individual peripheral B cells from SS patients were analyzed for their Vlambda gene usage by single-cell polymerase chain reaction amplification of genomic DNA and compared with those from normal controls.

RESULTS

Molecular differences from controls in Vlambda-Jlambda recombination were identified that were reflected by findings in the nonproductive Vlambda repertoire of the patients, including enhanced rearrangement of Vlambda10A and Jlambda2/3 gene segments. In addition, a number of abnormalities in the productive repertoire were identified, indicating disordered selection. A greater usage of 4 Vlambda genes (2A2, 2B2, 2C, and 7A), representing 56% of all productive Vlambda rearrangements, was observed, suggesting positive selection of these genes. Overutilization of Jlambda2/3 and underutilization of Jlambda7 in both nonproductive and productive Vlambda rearrangements of SS patients compared with controls suggested decreased receptor editing in SS. The mutational frequency did not differ from that in controls, and positive selection of mutations into the productive V gene repertoire was found, similar to that in controls, although mutational targeting toward RGYW/WRCY motifs, typically found in controls, was not found in SS patients.

CONCLUSION

Disturbed regulation of B cell maturation with abnormal selection, defects in editing Ig receptors, and abnormal mutational targeting may contribute to the emergence of autoimmunity in SS.

Perturbations in the impact of mutational activity on Vlambda genes in systemic lupus erythematosus

To assess the impact of somatic hypermutation and selective influences on the Vlambda light chain repertoire in systemic lupus erythematosus (SLE), the frequency and pattern of mutations were analyzed in individual CD19+ B cells from a patient with previously undiagnosed SLE. The mutational frequency of nonproductive and productive rearrangements in the SLE patient was greater (3.1 x 10(-2) vs 3.4 x 10(-2), respectively) than that in normal B cells (1.2 x 10(-2) vs 2.0 x 10(-2), both P < 0.001). The frequencies of mutated rearrangements in both the nonproductive and productive repertoires were significantly higher in the patient with SLE than in normal subjects. Notably, there were no differences in the ratio of replacement to silent (R/S) mutations in the productive and nonproductive repertoires of the SLE patient, whereas the R/S ratio in the framework regions of productive rearrangements of normal subjects was reduced, consistent with active elimination of replacement mutations in this region. The pattern of mutations was abnormal in the SLE patient, with a significant increase in the frequency of G mutations in both the productive and nonproductive repertoires. As in normal subjects, however, mutations were found frequently in specific nucleotide motifs, the RGYW/WRCY sequences, accounting for 34% (nonproductive) and 46% (productive) of all mutations. These data are most consistent with the conclusion that in this SLE patient, the mutational activity was markedly greater than in normal subjects and exhibited some abnormal features. In addition, there was decreased subsequent positive or negative selection of mutations. The enhanced and abnormal mutational activity along with disturbances in selection may play a role in the emergence of autoreactivity in this patient with SLE.

Production and characterization of monoclonal IgM autoantibodies specific for the T-cell receptor

Natural autoantibodies to the T-cell receptor (Tcr) have been identified in all human sera. However, titer, epitope specificity, and isotype vary with physiological conditions, autoimmune diseases, and retroviral infections. The levels of anti-Tcr autoantibodies in rheumatoid arthritis (RA) patients are significantly higher than in normal individuals, and the autoantibodies are typically IgM. To obtain detailed information on these autoantibodies, we generated B-cell heterohybridomas secreting monoclonal IgM autoantibodies (mAAbs) from the synovial tissue and peripheral blood of RA patients. We selected clones secreting mAAbs that bound a major Vbeta epitope defined by a synthetic peptide that contains the CDR1 region of the Vbeta 8.1 gene product. From these we isolated a subset of seven mAAbs that bound a recombinant single-chain Valpha/Vbeta construct containing the peptide epitope and, also to JURKAT cells which express Vbeta 8.1. The mAAbs produced by these clones were distinct from each other in their V-region sequences. However, all the V regions were essentially identical to germline sequences in both the heavy and light chains. Heavy-chain CDR3 segments ranged in length from 17 to 26 residues, did not correspond to any known autoantibodies, and showed extensive N-region diversity in the V(D)J junctions. Five monoclonal autoantibodies use VH 3 genes, while the remaining two utilized VH 4 sequences. Light-chain variable regions used were Vkappa3 (two), Vlambda3 (four), and one Vlambda2. These autoantibodies derived their unique features from their CDR3 segments that could not be aligned with any known sequences.

Ig λ and Heavy Chain Gene Usage in Early Untreated Systemic Lupus Erythematosus Suggests Intensive B Cell Stimulation

To determine the distribution of Vλ and Jλ as well as VH and JH gene usage in a patient with systemic lupus erythematosus (SLE), productive and nonproductive VJ and V(D)J rearrangements were amplified from individual peripheral CD19+ B cells and were analyzed. No differences in the Vλ and Jλ or the VH and JH gene usage in the nonproductive gene repertoire of this SLE patient were found compared with the distribution of genes found in normal adults, whereas marked skewing of both Vλ and VH was noted among the productive rearrangements. The distribution of productive Vλ rearrangements was skewed, with significantly greater representation of the Jλ distal cluster C Vλ genes and the Vλ distal Jλ7 element, consistent with the possibility that there was receptor editing of the Vλ locus in this patient. Significant bias in VH gene usage was also noted with VH3 family members dominating the peripheral B cell repertoire of the SLE patient (83%) compared with that found in normal subjects (55%; p &lt; 0.001). Notably, a clone of B cells employing the VH3-11 gene for the heavy chain and the Vλ1G segment for the light chain was detected. These data are most consistent with the conclusion that extreme B cell overactivity drives the initial stages of SLE leading to remarkable changes in the peripheral V gene usage that may underlie on fail to prevent the emergence of autoimmunity.

Clonal Analysis of a Human Antibody Response. III. Nucleotide Sequences of Monoclonal IgM, IgG, and IgA to Rabies Virus Reveal Restricted Vκ Gene Utilization, Junctional VκJκ and VλJλ Diversity, and Somatic Hypermutation

In previous work, we generated four IgM, five IgG1, and one IgA1 mAbs to rabies virus using B cells from four subjects vaccinated with inactivated rabies virus, a thymus-dependent (TD) mosaic Ag, and sequenced the mAb VHDJH genes. Here, we have cloned the VκJκ and VλJλ genes to complete the primary structure of the Ag-binding site of these mAbs. While the anti-rabies virus mAb selection of Vλ genes (2e.2.2 twice, DPL11, and DPL23) reflected the representation of the Vλ genes in the human haploid genome (stochastic utilization), that of Vκ genes (O2/O12 twice, O8/O18, A3/A19, A27, and L2) did not (p = 0.0018) (nonstochastic utilization). Furthermore, the selection of both Vκ and Vλ genes by the anti-rabies virus mAbs vastly overlapped with that of 557 assorted VκJκ rearrangements, that of 253 VλJλ rearrangements in λ-type gammopathies, and that of other Abs to thymus-dependent Ags, including 23 anti-HIV mAbs and 51 rheumatoid factors, but differed from that of 43 Abs to Haemophilus influenzae type b polysaccharide, a prototypic thymus-independent (TI) Ag. The anti-rabies virus mAb VκJκ and VλJλ segments displayed variable numbers of somatic mutations, which, in mAb58 and the virus-neutralizing mAb57, entailed a significant concentration of amino acid replacements in the complementarity-determining regions (p = 0.0028 and p = 0.0023, respectively), suggesting a selection by Ag. This Ag-dependent somatic selection process was superimposed on a somatic diversification process that occurred at the stage of B cell receptor for Ag rearrangement, and that entailed V gene 3′ truncation and N nucleotide additions to yield heterogeneous CDR3s.

Relation between the heavy chain complementarity region 3 characteristics and rheumatoid factor binding properties

Among the rheumatoid factors (RFs), monospecific and polyspecific types can be distinguished. However the molecular basis responsible for their different specificity is not well understood. In a previous report, we have shown that the binding of the majority of the polyspecific antibodies is salt-sensitive. No binding to IgG was observed under high ionic strength (0.3-0.5 M NaCl). This salt-sensitivity was only observed for 18% of the monospecific RFs. Here, we have analyzed 14 RFs representing the 3 different groups (6 salt-insensitive monospecific, 4 salt-sensitive monospecific and 4 salt-sensitive polyspecific RFs). By analysis of the amino acid composition and the distribution of polar and non-polar residues of their heavy chain complementarity-determining region 3 (H-CDR3) in relation to mono/polyspecificity, salt-sensitivity and reactivity against human IgG subclasses, we have identified common structural features responsible for their different binding properties. Salt-sensitive RFs (mono as well as polyspecific antibodies) were characterized by long H-CDR3's (15.3+/-2.7) that contained large numbers of hydrophilic residues such as arginine and serine, while salt-insensitive RFs had more hydrophobic H-CDR3's of smaller length (11.3+/-2.4). In addition, for the monospecific RFs, remarkably similar hydrophilicity H-CDR3 profiles were found that were correlated with their specificity for IgG subclasses. These observations confirm the importance of the H-CDR3 for the binding of RFs to IgG. Furthermore, on the basis of their shorter H-CDR3's and their rather unique H-CDR3 hydrophilicity profiles, it is likely that the majority of the monospecific RFs should be considered as a group of RFs that is independent of the polyspecific RF repertoire.

Analysis of the molecular basis of synovial rheumatoid factors in rheumatoid arthritis

The objective of this study was to better understand the molecular basis of IgM rheumatoid factor in rheumatoid arthritis (RA). We recently generated 10 different monoclonal IgM RF (mRF) molecules isolated from the synovium of a single patient with RA. The heavy (H) and light chain (L) variable region (V) genes of these 10 mRFs were cloned and sequenced. Six mRFs used kappa light chains and 4 mRFs used lambda light chains. Of particular interest, 8 of 10 heavy chains used the JH4 joining region gene, and all five VH4 heavy chains used the DK4 diversity region gene with the JH4. Four of the VH4 clones used the same germline gene, likely representing a novel but closely related germline gene to VH4.18, and may be clonally related because of the extensive homology in their heavy chain sequence. Two VH4 clones shared the same light chain gene, VkappaIIIb kv325 (99% homology) and the same JK4 joining region gene, while three VH4 clones used two different light chain genes, an uncommon Vkappa4 and a Vlambda4 gene, respectively. In this RA patient, there was recurrent utilization of VH4-DK4-21/10-JH4 genes and a recurring association with gene elements Vkappa3 and Vlambda4. Recurring usage of Vkappa3 (kv325) and Vlambda4 (lv418) gene elements may result from a light chain editing process whereby immature autoreactive B cells encountering self-antigen attempt, and often succeed, in altering their specificities through secondary Ig light chain gene rearrangement. Moreover, the oligoclonality of these RFs suggest clonal relatedness secondary to an antigen-driven response.

Rheumatoid factor autoantibodies in health and disease

Recent advances in molecular biological and human cell hybridization technology have significantly advanced the knowledge of mechanisms that underlie human rheumatoid factor (RF) production. These advances have provided insight into the etiopathogenesis of synovial inflammation and lymphocyte recruitment in rheumatoid arthritis (RA) joints. We have examined the mechanisms that lead to RF production in RA patients and those that regulate RF production in normals. The studies revealed structural features that distinguish RF produced in normals from those produced in RA synovial tissue. There are significant differences in the use of VL and VH genes between the two RF populations. Furthermore, IgV genes encoding synovial RF in RA have extensive evidence for nucleotide changes, leading to amino acid replacement in the complementarity determining regions (CDRs). In addition, RF produced in RA synovia show evidence for affinity maturation, isotype switch to IgG RF, and repertoire shift indicative of a continued recruitment of B cells. Together with computer modeling and crystallographic studies, our data suggest that the mechanisms that operate on RF selection in RA synovia are similar to immune responses to exogenous antigens. In contrast, RF established from human immunized donors (HID) are characterized by a very low ratio of replacement to silent (R:S) nucleotide changes in the CDR1+2. In addition, there is little increase in affinity with increasing numbers of mutations. There is thus evidence for regulatory mechanisms that limit affinity maturation of RF in normals.

Differences in mutational patterns between rheumatoid factors in health and disease are related to variable heavy chain family and germ-line gene usage

The sequences of the heavy chain variable (V(H)) segment and dissociation constants (Kd) of 14 IgM rheumatoid factors (RF) derived from 11 different germline gene segments from five healthy immunized donors (HID) are described. We extend a previous analysis of two clones from one donor using only the germline segment DP-10. In the present study, the mutation patterns of these new RF and the two earlier reported HID RF clones are analyzed in relation to V(H) family, germ-line origin, and Kd. The panel of HID RF is further compared with 33 previously described IgM RF from patients with rheumatoid arthritis (RA). There is a high rate of mutation in the panel of HID RF (mean of ten mutations/V(H)). RF originating in RA patients have a comparable mutation rate (mean of 11 mutations/V(H)), suggesting that hypermutation of IgM RF is not disease related. The HID RF have, however, a significantly lower affinity for IgG than the RA RF. We found that the structural basis of the differences between HID and RA RF is related to V(H) family usage. RF of the V(H)1 family use very similar germ-line genes in HID and RA patients. HID RF of the V(H)1 family have, however, a low ratio of replacement-to-silent (R:S) mutations of only 0.41 in the heavy chain complementarity region (CDR(H))1 and 2. This is statistically significantly lower than the corresponding ratio of 3.14 in the V(H)1 RA RF. In contrast, RF of the V(H)3 family from HID and RA patients have very similar R:S ratios of 1.75 and 1.71 in CDR(H)1 and 2, respectively. The V(H)3 RA RF are, however, predominantly encoded by genes not encoding any HID RF. Thus, both repertoire differences and hypermutation resulting in significantly lower R:S ratios can be observed in RF from HID compared with RA RF.

Repertoire cloning of lupus anti-DNA autoantibodies

To investigate the autoantibody repertoire associated with SLE, we have created phage display IgG Fab libraries from two clinically active SLE patients and from the healthy identical twin of one of these patients. The libraries from the lupus discordant twins were found to both include unusually large representations of the V(H)5 gene family. By panning with DNA, the SLE libraries each yielded IgG anti-double-stranded (ds) DNA autoantibodies, which are characteristic of lupus disease. These included a V(H)5 autoantibody from the affected twin, that has a targeted cluster of mutations that potentially improves binding affinity. The recovered IgG anti-dsDNA autoantibodies expressed the same idiotypes associated with the in vivo IgG anti-dsDNA response of the respective SLE donor. Heavy-light chain shuffling experiments demonstrated a case in which the in vitro creation of anti-dsDNA binding activity required restrictive pairing of a heavy chain with Vlambda light chains similar to those in circulating anti-dsDNA autoantibodies. By contrast, IgG anti-ds autoantibodies could not be recovered from the library from the healthy twin, or from shuffled libraries with heavy chains from the healthy twin. These repertoire analyses illustrate how inheritance and somatic processes interplay to produce lupus-associated IgG autoantibodies.

Comparison of rheumatoid factors of rheumatoid arthritis patients, of individuals with mycobacterial infections and of normal controls: evidence for maturation in the absence of an autoimmune response

We analyzed the rheumatoid factors (RF) produced by Epstein-Barr virus-transformed monoclonal B cells established from four patients with rheumatoid arthritis (RA), three individuals with a history of Mycobacterium tuberculosis (TB) and four normal controls (NI). Fifty-eight RF were analyzed for specific activity (international units-RF/microgram) for the Fc part of IgG and their interaction with tetanus toxoid (TT) and DNA (polyspecificity). Furthermore, we sequenced the V-D-J heavy chain region of 16 (9TB-/7RA-) RF. Significant differences were observed between the NI-RF and the TB- and RA-RF. While the RF repertoire of normal individuals comprised of low-avidity RF of which the majority (15/17) were polyspecific, more than half of the TB- and RA-RF were monoreactive. Furthermore, the monospecific TB- and RA-RF were of significantly higher avidity than the NI-RF (RA > TB > > NI). With respect to polyspecificity specificity, the RF in the three groups were comparable: the interaction with DNA, TT as well as with Fc was inhibited either by an increase of the ionic strength to 0.3-0.5 M NaCl or by addition of the polyanion dextran sulfate, indicating that the antibodies interacted with similar anionic epitopes shared by the three antigens. Analysis of the V-D-J heavy chain regions showed significant differences between the respective RF. The salt-sensitive binding was highly correlated with the presence of arginine in the complementarity-determining region 3 (CDR3). Furthermore, whereas the polyspecific RF consisted predominantly of germ-line encoded antibodies, the genes of the monospecific RA/TB-RF were somatically mutated (RA > TB). It is therefore likely that maturation of RF can be initiated by chronic infections and that monospecific, somatically mutated RF are not a unique characteristic of autoimmune diseases.

A subgroup of human VH3 germline genes that encode a high-avidity synovial rheumatoid factor

We have previously derived and identified a highly avid monoclonal IgM rheumatoid factor (mRF), C6, from unstimulated rheumatoid synovial cells (RSC). At the time, the closet VH germline gene, VH26, demonstrated only 88% homology with C6. To identify the germline counterpart of C6, genomic DNA from the same rheumatoid arthritis (RA) patient from whom C6 was derived was used in the polymerase chain reaction (PCR). Four of the six closely related germline genes that we sequenced had exonic regions that were identical with the VH region of C6 cDNA. These six germline sequences differed in their intronic regions, suggesting that they were distinct, but closely related genomic sequences. To further evaluate the extent of these related genes we identified nine additional germline genes having VH-encoding exons that were 86-97% identical to the C6 cDNA sequence. Furthermore, we examined the polymorphic nature of the C6 VH gene using single strand conformation polymorphism (SSCP), and identified two peaks, confirming the existence of highly homologous genes. The sequence and polymorphism data suggest that: (1) the VH region of the high avidity mRF C6 was derived from an unmutated germline gene; (2) C6 was encoded by a VH gene belonging to a set of homologous genes within the larger VH3 family; and (3) in addition to somatic rearrangements of B-cell genes and antigen-driven somatic mutation, gene duplication and conversion events of germline genes could be important in generating diversity and polyclonality among high-affinity pathogenic autoantibodies.

IgH and L chain variable region gene sequence analyses of twelve synovial tissue-derived B cell lines producing IgA, IgG, and IgM rheumatoid factors structure/function comparisons of antigenic specificity, V gene sequence, and Ig isotype

In the present study, the complete sequences of the Ig H and L chain variable region genes of twelve RF+ B cell lines from two patients with RA were analyzed. Seven of the RF-producing B cells used VH3 family genes, four used VH4 genes, and one a VH1 gene. All but two of the cell lines expressing VH3 genes utilized different family members; among the VH4-expressing cells, a more restricted pattern was noted. V kappa gene use was restricted to the V kappa I and III families; V lambda gene use was more diverse, involving five different families. Computer comparisons of the expressed VH genes with their presumed germline progenitors indicated significant differences in every instance; eight of the corresponding VL genes also were significantly different. In many cases, assignment of the germline D segment(s) incorporated into the rearranged VH genes was impossible. These differences from the germline gene segments indicated the extensive changes induced by rearrangement, enzymatic activities, and somatic mutation. In hopes of defining a structural reason for the disparate antigen specificities of these cells, the CDR3 amino acid sequences of the multi- vs. the mono-reactive RF-producers were compared. Although CDR3 length was not appreciably different between these two sets of mAb, a greater than two-fold increase in charged amino acids was found in the H chain CDR3 of the multireactive RF. This relationship did not exist for the L chain CDR3. Thus, these sequence data indicate the use of a broad base of Ig V gene segments that have undergone extensive diversification. Based on the localization of R substitutions in the CDR of most of the V genes studied, the diversification appears to be antigen driven and selected. The significance of these findings for the evolution of these B cell clones into isotype-switched producers that are heterogeneous for antigen specificity (mono- vs. multi-reactivity) is discussed.