Restricted Immunglobulin Gene Usage of Spleen-Derived Specific Anti-ADAMTS13 B-Cells In Relapsing TTP Treated with or without Rituximab.

Thrombotic Thrombocytopenic Purpura (TTP) is a severe, life-threatening disease and results from an impaired regulation of the von Willebrand factor multimer size due to a deficiency of its cleaving protease ADAMTS13. In case of acquired TTP, inhibitory autoantibodies neutralize and/or accelerate ADAMTS13 clearance. To better understand the role of these autoantibodies in the pathogenesis of TTP and to develop more specific therapies, molecular analysis of the pathological relevant isotype IgG₄ is necessary. Rituximab, a B-cell depleting anti-CD20 mAb is used to treat various autoimmune diseases with variable success. Increasing evidence points at the spleen, rather than the bone marrow as the major reservoir of memory B cells, which are thus accessible to Rituximab. The analysis of the humoral immune response of splenic B-cells towards ADAMTS13 in TTP patients treated with/without Rituximab offers a unique opportunity to address resistance to Rituximab.

Two patients suffering from acute TTP with high titers of anti-ADAMTS13 antibodies were splenectomized because of frequent relapses, which occurred in one case even after several courses of Rituximab (patient B). Splenic mononuclear cells were isolated and used to clone the patient's entire IgG₄ repertoire using phage display technology. Two IgGk-IgGλFab libraries of ∼2 × 10¹⁰ members each were constructed and selected against recombinant ADAMTS13 coated on ELISA plates. Additionally an existing naïve IgM Fab library constructed from cord blood RNA was screened identically. To analyze the memory repertoire of Rituximab-resistant and non-resistant anti-ADAMTS13 IgG, the subset of switched memory B cells (CD19, CD27 IgG) was isolated by magnetic beads (Miltenyi Biotec) from the splenic mononuclear cells of both patients prior to immortalization by Epstein-Barr virus (EBV) transformation to generate monoclonal antibodies.

Phage display. After 5 rounds of selection a 10⁵-fold enrichment in eluted phages was observed. An ELISA screening after the fourth panning round showed 16/34 clones displayed a strong binding to ADAMTS13. Sequencing of DNA encoding the heavy chain variable region (IgVH) of strong binding clones revealed that the Fabs of patient A were most homologous to germline genes IgVH1-3*01 (7/11) and IgVH4-28*01 (4/11). The usage of these germline genes was also observed in patient B (1/5 and 2/5 clones IgVH1 and IgVH4, respectively), in addition 2/5 clones used IgVH1-69*01 germline gene, which had been previously observed in other acquired TTP patients. Six out of 14 analyzed anti-ADAMTS13 Fabs of the naïve IgM Fab library showed the same restricted gene usage (VH1-2, VH1-69 and VH4-28), whereas 4/14 clones were found to comprise heavy chains encoded by IgVH6-1*02, 2/14 were encoded by IgVH3-48*02 and 2/14 by IgVH3-24*04. The homology to the closest germline gene for all VH genes was in the range of 81.6–99%, with the lowest mutation rate found in cord blood anti-ADAMTS13 IgM Fabs (range 1–7.7%) followed by the anti-ADAMTS13 Fabs of patient B (after Rituximab; range 2.9–13.6%, average 10%) and finally patient A (range 8.3–18.6%, average 15%), the mutation rate being in the range of an antigen-driven immune response. EBV transformation. Screening of the supernatant of single clones of EBV transformed memory B-cells for the presence of anti-ADAMTS13 IgG antibodies was positive in 5/125 (4%) of clones of patient A and in 8/109 (7.3%) of patient B. Sequencing of IgVH encoding DNA of 3 single clones to date revealed gene usage of IgVH1-3*01 in 2 clones (one patient A and B) with a mutation rate of 15.4% and one (patient A) being homolog to IgVH3-30*022 with a lower mutation rate of 8.2%, in the upper limit of the naïve repertoire.

VH gene usage of specific anti-ADAMTS13 B-cells, including Rituximab-resistant specific anti-ADAMTS13 B-cells is genetically restricted to VH1, VH3 and VH4. Rituximab may deplete circulating specific anti-ADAMTS13 B cells. However, newly differentiated specific anti-ADAMTS13 B cells characterized by a lower degree of somatic mutation may evolve from the memory cell pool giving rise to pathological anti-ADAMTS13 antibody. Functional characterization (inhibitory capacity, affinity and epitope mapping) of these anti-ADAMTS13 Fabs are underway.

No relevant conflicts of interest to declare.