Tight-skin mouse autoantibody repertoire: analysis of VH and VK gene usage

Innate murine B cells produce anti-disialosyl antibodies reactive with Campylobacter jejuni LPS and gangliosides that are polyreactive and encoded by a restricted set of unmutated V genes

In Guillain-Barré syndrome following Campylobacter enteritis, anti-lipopolysaccharide antibodies cross-react with neural gangliosides, thereby precipitating autoimmune neuropathy. We examined the properties of 15 murine anti-LPS/ganglioside mAbs specific for NeuAc(alpha2-8)NeuAc-Gal disialosyl epitopes. Many mAbs displayed features of an innate B cell origin including polyreactivity (13/15), hybridoma CD5 mRNA expression (5/15), predominance of IgM (9/15) or IgG3 (3/6) isotype, low affinity, and utilisation of unmutated VH and VL VDJ rearrangements. Antibody specificity resided in highly selective V gene usage, with 6/15 mAbs being encoded by the VH7183.3b gene. These data indicate that neuropathogenic antiganglioside autoantibodies can arise from the natural autoantibody repertoire.

Murine animal models of systemic sclerosis

Animal models of systemic connective tissue diseases have provided valuable insights into the causative mechanisms and the pathogenesis of these diseases, and have provided the means to test potentially useful therapeutic interventions. Although numerous animal models for systemic sclerosis (SSc) have been described, the most extensively studied are murine. One advantage of murine animal models is the large body of genetic information available for the mouse that is not available for other species. No animal model described to date reproduces precisely all manifestations of SSc. However, all animal models display tissue fibrotic changes similar to those present in SSc. The prudent interpretation of the results obtained from the study of animal models has provided substantial and valuable information about the pathogenesis of the human disease.

V lambda-light chain genes reconstitute immune responses to defined carbohydrate antigens or haptens by utilizing different VH genes

The contribution of the lambda-light chain to the development of peripheral B cell repertoire and generation of specific antibodies to haptens and polysaccharide antigens was studied in genetically manipulated kappa-deficient and lambda 2-transgenic mice. The results clearly demonstrate a non-stoichiometric VH gene family expression in the absence of k-light chain and suggest a non-stochastic pairing between VH and V lambda genes, expressed in the peripheral B cell repertoire. A shift in VH gene utilization in the case of VI lambda + antibodies was evident in response to beta 2-6 fructosan and TNP hapten. These observations demonstrate the availability of compensatory mechanisms in the absence of VK genes and are consistent with the hypothesis that VH gene family expression is controlled by genetic factors from inside the VH locus. Furthermore, genetic factors from outside the VH locus, namely restricted available light chain diversity, may lead to a shift in VH gene utilization in the peripheral B cell repertoire.

The primary antibody repertoire of normal, immunodeficient and autoimmune mice is characterized by differences in V gene expression

During the last decade, the structure and organization of the immunoglobulin heavy and light chain locis have been defined in mice and humans. Studies on VH gene expression at different stages of development, in different organs and disease states have provided useful insight into the construction of a primary antibody repertoire in mice. Clearly, 3'VH genes 7183, Q52 and Vh11, which are conserved during evolution, are preferentially expressed during early development of the B-lymphocyte repertoire. A preferential use for the V kappa 4 gene family is evident during early B-cell development. The initial development of the primary antibody repertoire is therefore influenced by a restricted set of VH and V kappa gene elements. The restricted B-cell repertoire is subsequently normalized in the periphery, as revealed by stochastic VH gene expression, as a result of exposure to environmental antigens. Obviously, the peripheral B-cell pool characterized by stochastic VH gene expression is selectively replenished by newly generated B cells in bone marrow that preferentially expresses 3'VH genes. The V kappa genes are, however, expressed in a non-random manner in the neonatal and adult B-lymphocyte repertoire that is probably related to VH and V kappa association dynamics and/or positive or negative selection. Interestingly, these characteristics of neonatal and adult primary repertoire are noted in both B1 and B2 lymphocytes. No remarkable age-related differences are evident for VH and V kappa gene expression. In healthy mice, both the mitogen responsive (available) and unstimulated (expressed) B-cell repertoire show similar VH gene expression. Interestingly, VH gene expression varies in different organs which may reflect, or occur as a result of, the specialized function of each organ. For example, J558 gene expression is higher in the peripheral LN where B cells continuously encounter exogenous antigens. The skewed VH and V kappa gene expression noted in immunodeficient and autoimmune lupus-prone mice reflects the impairment of the primary antibody repertoire associated with immunodeficiency and autoimmune disorders.

VH gene structure predicts a large potential anti-insulin repertoire

The majority of insulin antibodies derived from immunization are IgG antibodies that cross-react extensively with the autologous hormone. To examine the relationship between VH genes expressed by such self-reactive antibodies and their germline (non-rearranged) counterparts, we used the polymerase chain reaction (PCR) to amplify and isolate the germline progenitors of anti-insulin VH genes derived from BALB/c mice immunized with beef or human insulin. Results indicate that two anti-insulin mAbs (123 and 124) express VH genes which arise from a small subset of the J558 gene family and are highly homologous to the VH gene used by the murine CD5 + B-cell tumor, BCL1. The anti-insulin IgG mAb 127 belongs to the VH-VIII (Vgam 3.2) family and the amplification and isolation of germline VH genes from this small family precisely identified only two somatic mutation events in the CDRH2 of mAb 127. Another anti-insulin mAb, 133, also shows two replacement substitutions in the CDRHs when compared to the germline encoded anti-dextran antibody 19.1.2. These findings indicate that the IgG response to this small self-protein uses multiple VH genes which are largely germline encoded with only a low level of somatic mutation in their CDRHs. Additionally, analysis of N-segment additions in CDRH3s indicates anti-insulin B cells may originate from both early (fetal) and adult repertoires. These data are consistent with the concept that the mechanisms of clonal anergy or deletion do not regulate anti-insulin B cells and indicate that there is a large potential VH gene repertoire for insulin.

Autoantibodies in scleroderma and tightskin mice

There is much evidence to suggest that scleroderma in human patients is caused by a fundamental defect in the immune system. In tightskin mice, the scleroderma syndrome is associated with autoimmunity, particularly autoantibodies interacting with scleroderma target antigens.

Molecular characterization of J558 genes encoding tight-skin mouse autoantibodies: identical heavy-chain variable genes code for antibodies with different specificities

Tight-skin mouse, a mutant strain with a single gene defect, develops cutaneous hyperplasia and specific autoantibodies, like humans affected by scleroderma. The autoantibodies produced in the tight-skin mouse are encoded primarily by heavy-chain variable (VH) genes from the J558 family. To understand the genetic basis of production of autoantibodies, we have analyzed the structure of J558 genes encoding these autoantibodies. The results showed that J558 genes encoding these antibodies were not derived from a selected germ-line gene(s) or a single subfamily but were derived from genes belonging to diverse J558 subfamilies. However, two prototype VH genes representing two new subfamilies were found to be repeatedly expressed in their germ-line form in eight independent clones. Autoantibodies with distinct specificities appear to be generated by pairing of similar/identical VH genes with different V kappa genes derived from the same or different families. Fourteen of 18 autoantibodies shared a conserved heptapeptide sequence motif, YNEKFKG, in the second complementarity-determining region of heavy chains. Usage of germ-line genes from diverse J558 subfamilies bearing a common motif to encode autoantibodies suggests a regulatory role for this motif. Thus, selection and expansion of the autoreactive B-cell repertoire in the tight-skin mouse appear to be VH-gene mediated. The frequency of N nucleotide addition at diversity-joining (D-JH) junctions was lower, whereas the frequency of usage of the DFL16 segment was higher. Finally, in contrast to normal and other autoimmune mouse strains, the frequencies of D-D fusions and D inversions were higher in tight-skin mouse total immunoglobulin as well as autoantibody repertoires.

Restricted VH gene usage by murine hybridomas directed against the human N, but not M, blood group antigen

The M and N human blood group antigens are complex glycopeptide determinants at the amino terminus of the red blood cell membrane glycoprotein, glycophorin A. The heavy and light chain variable region cDNA sequences were determined for seven murine monoclonal antibodies recognizing glycophorin A. Three of the antibodies were anti-M and four were anti-N. Each of the anti-M antibodies was composed of VH and VL regions derived from distinct germline gene families (VH1 (J558), VH4 (X24), VH5 (7183), VK5, VK8, and VK19). In contrast, all four anti-N heavy chains were composed of VH regions derived from the VH2 (Q52) germline gene family and all used the same J4 gene segment. In addition, two of the anti-N light chains were composed of VK regions from the VK8 germline gene family and used the J1 gene segment. Since each anti-N hybridoma was derived from different mice immunized by different protocols, these results suggest that the murine immune response to the N, but not the M, human blood group antigen is restricted.