Identification of T Cell Epitope of ADAMTS13 in Thrombotic Thrombocytopenic Purpura Patients

Introduction:

Thrombotic Thrombocytopenic Purpura (TTP) results from the development of auto-antibodies directed against A Disintegrin And Metalloproteinase with Thrombospondin type 1 repeats, 13th member (A13). The implication of CD4+ T-cells in the pathogenesis of the disease is suggested by the existence of a restriction to HLA DRB1*11 allele and by the isotype switch of the antibodies. However, T-cell autoimmune response to A13 and the properties of CD4+ T-cells from TTP patients have never been studied. Here, we determined the immunodominant T-cell epitope of A13 in TTP patients.

Methods:

Using the IEDB website, we predicted in silico the immunodominant peptides of A13 based on their binding capacity to HLA DR11 haplotype. Subsequently, these peptides were synthesized and validated in vitro for their binding capacity to purified HLA-DR11 molecules using an ELISA competitive assay. The peptides that bound with the best capacity to HLA-DRB1*11 molecule were then tested for their recognition by human CD4+ T-cells from HLA DRB1*11 healthy donors and patients, at diagnosis or in remission. To this end, CD4+ T-cells were repetitively stimulated with HLA-DRB1*11 monocyte-derived dendritic cells loaded with the peptides and T-cell line were generated after amplification of interferon-γ secreting cells selected upon stimulation. The effect of individual peptide on activation of the established CD4+ T-cell line was assessed by interferon-γ (IFNγ) ELISPOT. Next, we evaluated the promiscuous HLA-binding capacity of the DRB1*11 identified peptides using the same method in HLA DRB1*01 TTP patients. Finally, in order to validate the involvement of these peptides in an immune response toward A13 in vivo, we immunized a humanized HLA DRB1*01-transgenic H-2 class I-/class II-knockout mouse with full length recombinant human A13 (rhA13). We then generated A13-specific T-cell hybridomas restricted to human HLA DRB1*01 and investigated whether the peptides previously identified were recognized by the hybridomas.

Results

A first list of 48 peptides with reliable predicted binding scores was elaborated through IEDB analysis. Of these, twenty-one peptides demonstrated a high binding capacity to HLA DRB1*11 molecules on ELISA competitive assay. These were selected to stimulate human CD4+ T-cells and we generated CD4+ T-cell lines from HLA DRB1*11 healthy donors and patients (n=5). Six A13 derived peptides were able to activate CD4+ T-cell lines, as revealed by IFNγ secretion by ELISPOT. The peptides were identified to be located within different domains of the protein but more particularly in the spacer and CUB2 domains. Interestingly, two of the identified peptides demonstrated promiscuity based on their ability to activate a CD4+ T-cell line we generated from a HLA DRB1*01 TTP patient. In parallel studies, using HLA DRB1*01 transgenic mice immunized with rhA13, we generated A13-specific T-cell hybridomas. The screening of their specificity allowed us to identify only one A13 derived peptide. The sequence of the peptide, located within the CUB2 domain, was precisely determined, it is promiscuous between DRB1*01 and DRB1*11 haplotype and represents the immunodominant CD4+ T-cell epitope of ADAMTS13.

Conclusion:

We identified several undescribed CD4+T-cell epitopes of A13 in HLA DRB1*1101 patients. They are located in different domains of the protein, particularly in the spacer and CUB2 domains. One of them, located in the CUB2 domain, is promiscuous to HLA DRB1*0101 and responsible for the immunodominant response to A13. The results we obtained, lead us to generate the tools to study the specific cells involved in the origin of the physiopathological process of the disease.