Analysis of anti-lymphocyte autoimmune antibody responses

 
Principal Investigator :  Rahul Pal

Co-Investigator
Rohini Handa, AIIMS, New Delhi

Project Associates / Assistants
Roopali Gandhi

Ph D Students
Joy Das

Systemic lupus erythematosus (SLE) is a prototypic non-organ specific autoimmune disease. Immune dysfunction is a hallmark of the disease, and abnormalities are observed in cells belonging to both the B and T cell lineages. A generalized hyper-gammaglobulinemia is observed and is often associated with the presence of specific humoral responses to several self-proteins and to double-stranded DNA. Lymphocytes are also often the targets of autoantibody responses; such responses are believed to contribute towards pathogenesis. This study aims to investigate the nature of the anti-lymphocytic autoimmune responses observed in SLE and to draw correlations with antibody-mediated biological effects. Monoclonal anti-lymphocyte antibodies would be generated towards this effort; cell-type and antigenic specificity will be determined. The effects of monoclonal antibodies on lymphocyte function will be analyzed in vitro and molecular characterization of reactive antigens will be carried out.

The long-term objectives of the project would include investigations aimed at understanding the observed temporal relationships between anti-lymphocyte antibodies and autoantibodies of diverse specificities. Cross-reactivities, as well as idiotypic connectivity, with known and well-characterized autoantigens would be investigated. Using appropriate animal models, the postulate that anti-lymphocyte antibodies could arise via epitope spreading would be studied.

Two human monoclonal antibodies PR5 and RN86, generated from peripheral B cells of SLE patients, were subjected to further biological and genetic analysis. The antibodies were purified from culture supernatant in order to permit unambiguous characterization. In all subsequent assays, human IgM similarly purified from peripheral blood was used as the negative control.

In addition to previously documented reactivity on T cells and monocytes, PR5 was observed to bind granulocytes and erythrocytes as well. It thus appears that the antibody exhibits very wide polyreactivity, binding to several cell types present in the blood. PR5 induced a dose dependent decrease in antigen-induced proliferation of human PBMC, and a variable degree of complement-mediated toxicity was demonstrated against erythrocytes and granulocytes, confirming and extending work reported last year. Immunoprecipitations and Western blots carried out to ascertain the nature of reactive antigen(s) upon the surface of erythrocytes revealed dominant reactivity towards a 30KDa membrane protein, the identity of which is being ascertained.

Genetic analysis of the heavy and light chain variable regions of PR5 was completed. The heavy chain demonstrated closest homology with the 08 germline (GL) gene, belonging to the VH1 family; a six base deletion was observed at the 3’ end of the gene (see below). Heavy chain sequences revealed very few mutations, and these were not sequestered in the CDR1 or CDR2 regions. The antibody demonstrated an essentially germline configuration. However, extensive non-templated nucleotide modifications were noted at both V-D and D-J junctions, as diagrammed below. For the J chain, complete identity was observed beyond the region shown. In comparison with closest germline segments (top line), bases in bold typeface refer to deletions and the underlined bases to non-templated additions. The arrow indicates a point mutation:

08(VH1) GL

D1-26 GL

JH4 GL

...CTGTGCGAGAGA GGTATAGTGGGAGCTACTACT    

ACTACTT..

PR5 Heavy Chain
...CTGTGC     CAAAAGCCCAAGG AGTGGGAACTACTACT

TCGGTCCT

TT..

In the case of the PR5 light chain, the entire L6 germline gene (VkIII family) was used (not shown). The Jk2 segment was employed with a single base pair deletion at the 5’ end and a point mutation, as indicated by the arrow

Jk2 GL
GTACACTTTTGGCCAGGGGACCAAGCTGGAGATCAAAC

PR5 Light Chain
   GTACACTTTTGGCCAGGGGACCAAGCTGGAGATCCAAC

Near germline-encoded (but heavy and light chain CDR3 region-modified) polyreactive antibodies directed at ubiquitous antigens present on the cell surface of healthy cells can therefore mediate immune dysfunction in multiple ways. Establishing the identity of reactive antigens could be important in explaining some aspects of disease etiology.

It is known that cells undergoing apoptosis are preferentially recognized and internalized by phagocytic cells; these latter cells have recently been shown to express the IgM receptor. We therefore determined the influence the apoptotic-cell specific antibody RN86 on this process. MOLT-4 cells were labeled with 3H-thymidine and apoptosis was induced by treatment with Camptothecin, a DNA topoisomerase I inhibitor. It was observed that the presence of RN86 in the culture caused an enhancement in the engulfment of apoptotic cells. This data assumes significance, since some specificities present in normal human IgM have recently been shown to bind to apoptotic cells. Coupled with the fact that serum IgM knockout animals display features of systemic autoimmune disease, these results may indicate a disease-ameliorating role for RN86-like antibodies.

Last year, studies to investigate the possible role of the idiotypic network in the phenomenon of epitope spreading were initiated, since idiotypic connectivity has been previously demonstrated in SLE-related autoreactive specificities, and one of the RN86-reactive antigens (Ro60) exists in complex with La, another ribonucleoprotein. Antibodies generated in rats against RN86 were capable of preventing the interaction of the monoclonal antibody with its cellular target(s). To assess if epitope spreading had indeed occurred, the reactivity of these antibodies was assessed on the panel of autoantigens (Ro60, Ro52, La, SmB, SmD, the U1RNP A protein and the 70kDa protein). In consonance with our hypothesis, the antibodies recognized La (a ribonucleoprotein associated with Ro60), reactivity not apparent in the original antibody. Interestingly, antisera also recognized Ro60, as did the immunogen RN86; this data was reminiscent of results with the 16/6 anti-DNA idiotype described in the literature. Efforts are underway to determine if discrete or cross-reactive antibodies are responsible for this recognition.

Sequencing of the RN86 heavy and light chain variable regions was also carried out. A comparison of the RN86 heavy chain gene (Accession Number: AJ548507) with the NCBI database revealed closest homology with the V3-53 heavy chain germline (GL) gene, belonging to the VH3 family. The D3-10 and JH6 segments appeared to be employed; both were extensively modified. Heavy chain sequences revealed very few mutations, and these were not sequestered in the CDR1 or CDR2 regions. The antibody demonstrated an essentially germline configuration. The entire V3-53 gene was employed (not shown), without 3’ deletions. Once again, for the J chain, complete identity was observed beyond the region shown; with respect to respective germline segments, bases in bold indicate deletions, while underlined bases indicate non-templated additions:

D3-10 GL

    

JH6 GL
GTATTACTAT GGTTCGGGGAGTTATTATAAC

ATTA

CT..

RN86 Heavy Chain
TTTG GGTTCGGGGAGTTATTATA

TGGG

CT..

The RN86 light chain gene (Accession Number: AJ548508) demonstrated closest homology with the 02 germline gene of the Vk1 family. Again, insignificant mutations were observed in the gene; there was however, a three base deletion at the 3’ end of the 02 gene. The arrow indicates a
T    A mutation

02(Vk1) GL

Jk1 GL

...TACAGTACCCCTCC

GTGG..

RN86 Light Chain
...TACAGTACCCC

GAGG..

These results indicate that near germline-encoded (but heavy and light chain CDR3 region-modified) immune responses to apoptosis-related antigens may influence the subsequent course of autoimmune disease by at least two distinct mechanisms; by influencing the immunological availability of potential autoantigens; and by stimulation of the idiotypic network to induce intermolecular epitope spreading.

Publications

Original peer-reviewed articles

1.     *Kathuria S, Sriraman R, Nath R, Sack M, Pal R, Artsaenko O, Talwar GP, Fischer R and Finnern R (2002) Efficacy of plant produced recombinant antibodies against hCG. Human Reprod 17:2054-2061 (*in press last year, since published).

2.     Kathuria S, Nath R, Pal R, Om Singh, Fischer R, Lohiya NK and Talwar GP (2002) Functional recombinant antibodies against human chorionic gonadotropin expressed in plants. Curr Sci 82:1452-1457.