[Investigation of the possible use of B cells for the diagnosis of acute Lyme disease]. (Doctoral thesis)

Misdiagnosis of acute Lyme disease still remains a huge problem. Also many debates are going on about the definition and underlying causes of the chronic course of this disease. Due to the scarcity of Borrelia in patient samples, direct detection of the pathogen is challenging. Diagnostic laboratories need to work with indirect approaches to prove the presence of Borrelia inside patients. The best and currently only accepted test to support the diagnosis of acute Lyme disease infection is the detection of Borrelia-reactive antibodies. Although extensive optimization has led to the development of specific and quite reliable tests to prove the presence of these molecules, they have the limitation that they do not enable to directly distinguish an ongoing from a previous infection. Since B cells are the cells of our immune system that react most specifically to immune challenges, they are very interesting candidates to be explored for biomarker research. As opposed to the antibodies, which can stay in peripheral blood for prolonged times even after clearance of the infection, Borrelia-reactive B cells should only be elevated when the pathogen is present. The main goal of this study was to prove the presence of Borrelia-reactive B cells in peripheral blood of acute Lyme disease patients and to assess to what extent expanded clones overlap between patients. Our results are in line with other B cell repertoire studies, indicating that individuals that have been challenged with the same immune stimulus have a higher chance to present overlapping clones as compared to other randomly selected donors. Since we were not able to use precise timepoints in the case of acute disease, it was quite surprising to find an increased overlap of B cell clones in our patients. Our data indicate, that overlapping clones appear at heterogeneous levels between donors. Rather patient unique B cell clones are predominating individual repertoires. This is in accordance with the rather dynamic nature of B cell immune responses. Since the occurrence of identical B cell receptor sequences in different patients was a rare event, we explored whether sequence clustering using less stringent parameters would allow to extract B cell clones reacting towards the same epitope. Differences in sequences between clones were however too large to allow reliable grouping of sequences according to epitope-reactivity. For this purpose, more sophisticated bioinformatics clustering tools need to be developed. Those need to take mutations at key residues and three dimensional structures of antibody molecules into account. From our findings, we conclude, that B cell repertoire signatures might be more complex than initially imagined. We will probably not be able to isolate a few key CDR3 amino acid signature candidates, but we will end up with a rather large list of Borrelia-reactive B cell clones that can be distinguished from B cells reactive towards other antigens by the presence of single key mutations. The polyspecific and dynamic nature of the B cell immune response will probably require to prove the presence of certain combinations of clones. Matching of B cell clones with their corresponding antigens and analysis of mutation levels, isotypes and phylogenetic relationships among sequences from these clones has the potential to allow to distinguish ongoing (primary and secondary) from past immune responses and autoimmune diseases. This might help to solve some of the controversies about post treatment Lyme disease.