Analysis of anti-lymphocyte autoimmune antibody responses

Co-Investigator
Rohini Handa, AIIMS, New Delhi

Project Associate
Roopali Gandhi

PhD Student
Joy Das

The long-term objectives of the project 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.

Efforts were undertaken to procure, by recombinant means, ribonucleoproteins that are targets of the humoral immune response in SLE. These reagents would prove useful in subsequent characterization of anti-lymphocyte antibody responses. Qiagen-based plasmids encoding a panel of antigens were obtained from Dr Shu Man Fu at the University of Virginia. Transformation of M15 E. coli cells was conducted using standard protocols. Recombinant E. coli were induced by 1mM IPTG. Whole cell lysates or inclusion body preparations were analyzed for protein expression by SDS-PAGE and western blots. In the latter instance, anti-histidine-HRPO or Ni-HRPO conjugates were employed as revealing reagents, taking advantage of the presence of the 6X-Histidine at the either amino or carboxyl terminal of the constructs. Ro60, Ro52, SmD and SmB were successfully expressed. Nickel-NTA affinity column chromatography was conducted to purify the antigens. Optimization studies, including scale-up to the fermentor stage, are under way.

Characterization of human monoclonal anti-lymphocyte antibodies was extended. Previous work had established cell surface reactivity of antibody Pr5 (IgMk) towards a variety of cell lines, as opposed to the poor recognition of normal, healthy cells by RN86. In addition (and again in contrast to RN86), the antibody demonstrated no reactivity towards the panel of recombinant antigens outlined above. Studies on normal human PBMC revealed further differences between the antibodies. RN86 recognized PBMC poorly, unless cells had been previously been activated by PMA and Ionomycin. Analysis using anti-CD69 antibodies and Annexin-V revealed this to be an apoptosis-specific, rather than an activation-specific, recognition. Pr5 on the other hand, recognized normal, non-activated PBMC very efficiently. Two-colour flow cytometry elicited interesting data on the relative abundance of the antigen on different cell types; CD4 T cells were the most weakly stained (and in some instances, not stained at all), followed by CD8 T cells. Monocytes appeared to be recognized to the maximum extent.

Significant recognition of normal human PBMC, and the fact that Pr5 was of the IgM isotype, prompted the conduction of a complement fixation test. Briefly, PBMCs were incubated with different concentrations of the antibody for one hour at 40C. Titrated rabbit serum was added as a source of complement. After another incubation for three hours at room temperature, the extent of cell death was quantified by trypan blue exclusion and analyzed microscopically, or by propidium iodide and analyzed by FACS. Expectedly, RN86 (used as an isotype- and species-matched negative control because of the absence of PBMC recognition, as outlined above) was ineffective in causing complement-mediated lysis. Under these conditions, W6/32 (a complement-fixing antibody directed against monomorphic determinants on the human MHC Class I molecule) caused death in @ 80% of the cells. Pr5 induced death in about @ 25% of the cells, and initial observations indicate a specific reduction in the monocyte population. It would be interesting to attempt correlate the selective killing of certain cell types with the immune dysfunction that characterizes the disease.

Preliminary experiments reveal a significant inhibitory influence of Pr5 upon a mixed lymphocyte reaction. These experiments involved incubations over a 5-day period, using de-complemented fetal calf serum. If confirmed, these results indicate that a single cell-surface reactive antibody can mediate immune dysfunction by two very distinct mechanisms – cell death and cell stasis.

Genetic characterization of these antibodies was initiated. Briefly, subcloned hybridoma lines were selected for antibody high secretion and total RNA was isolated using TRIZOL. Reverse transcription yielded cDNA; PCR amplification reactions were then carried out, using degenerate Ig primers that were either light chain or heavy chain specific. Amplification products in the region of 500-600 base pairs were considered putative Ig sequences, and were ligated into the pGEM-T Easy vector. DH5a E. coli cells were transformed, and selection was carried out using X-gal, IPTG and ampicillin. Plasmid DNA was isolated from several white colonies after appropriate expansion; restriction digestion was conducted to confirm the presence of an insert of expected size. Sequencing was carried out on an automated sequencer.

The sequence for the Pr5 k light chain has currently been compiled. Comparison with the NCBI Ig sequence database reveals a 99% sequence identity with the kappa germline gene L6. The only mutation appears to be a G® A change in the CDR1 region. This data reveals that a potentially pathogenic autoantibody can carry an essentially unmutated light chain sequence. Heavy chain analysis would no doubt be revealing, especially since the CDR3 region frequently contains non-encoded bases. Work upon the apoptotic-cell specific antibody, RN86, has also been initiated.

Stimulation of the idiotypic network by antibodies that recognize a molecule forming part of a supramolecular complex (such as Ro60/La) may lead to additional diversification of the humoral response. Probing experiments were conducted with RN86, since the recognized antigens were known; in addition, its polyreactive nature may increase the chances and the extent of diversity. The antibody was purified by ammonium sulfate precipitation, followed by affinity chromatography and rabbits and rats were immunized with this preparation. Antisera could prevent the binding of RN86 to permeabilized cells, whereas control antisera was ineffective. Western blot analysis revealed the recognition of antigenic moities distinct from those recognized by RN86. The nature of these antigens is under investigation.

Publications

Original peer-reviewed articles

1.   *Pal R and Om Singh (2001) Absence of corpus luteum rescue by chorionic gonadotropin in women immunized with a contraceptive vaccine. Fertil Steril 76:332-336 (*in press last year, since published).

2.   *Gupta A, Pal R, Ahlawat S, Bhatia P and Om Singh (2001) Enhanced immunogenecity of a contraceptive vaccine using diverse synthetic carriers. Vaccine 19:3384-3389 (*in press last year, since published).

3.   *Gupta A, Chandrasekhar S, Pal R, Ahlawat S and Om Singh (2001) High expression of human chorionic gonadotropin b-subunit using a synthetic vaccinia virus promoter. J Mol Endocrinol 26:281-287 (*in press last year, since published).

4.     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 human chorionic gonadotropin. Hum Reprod (in press).

Patents

1.     Panda AK, Pal R, Venkatesan N and Om Singh. A process for improving the immunogenicity of a weak antigen. Indian patent application No. 794/Del/2001 filed in Jul 2001.