Role of PAX5 Chimera in the Pathogenesis of Childhood ALL

PAX5, a powerful transcription factor, which controls B-cell development, is one of the most frequent targets of somatic mutation in B-cell precursor acute lymphoblastic leukemia (BCP-ALL). These mutations comprise deletions, point mutations, and of particular interest, gene rearrangements resulting in fusion transcripts, which encode chimeric proteins. These fusion proteins are potential trans-dominant aberrant transcription factors that antagonize normal PAX5 function. Intriguingly, a multitude of different genes including transcription factors, structural proteins, and the tyrosine kinase JAK2 can serve as PAX5 fusion partners. Currently, studies regarding the function of these PAX5 chimera are mainly focused on the expression analysis of only a few direct PAX5 target genes. However, PAX5 regulates approximately 300 genes, and the extreme functional heterogeneity of the partner proteins suggests that they have an impact on target gene transcription or at least a modulating effect. Moreover, it remains widely elusive whether all PAX5 fusion genes indeed encode aberrant transcription factors, which influence PAX5 target gene expression. Therefore we aim at a detailed functional analysis of PAX5 chimeric proteins to elucidate their contribution to the pathogenesis of BCP-ALL. We propose to study the global impact of PAX5 fusion proteins on the downstream transcriptional network and to dissect whether the different chimera result in distinct or convergent expression signatures. Chromatin immunoprecipitation (ChIP) and electromobility shift assays (EMSA) will be used to determine if all PAX5 chimera are capable to bind to at least a subset of endogenous PAX5 target gene promoter sequences resulting in dysregulation of the respective genes. Further, GeneChip-based global gene expression profiling will serve to analyze whether the moieties provided by the fusion partner protein may have a modulatory effect on PAX5 target gene regulation. In order to verify the relevance of the in vitro data, differentially expressed genes will be validated in the context of human leukemia, namely in primary leukemic cells. Together, these studies will increase our understanding how PAX5 chimeric proteins contribute to the pathogenesis of B-cell precursor acute lymphoblastic leukemia and yield new insights into the function of PAX5 in human B- cell development.

B-cell precursor acute lymphoblastic leukemia (BCP-ALL) is the most common type of cancer in children and young adults. In tumours of the blood system genes that play important roles in blood cell development are frequently genetically altered. PAX5, a powerful transcription factor, which activates a B-cell-specific transcription program and at the same time inactivates lineage-inappropriate genes, thereby controlling the commitment, development, function, and identity of B-cells, is one of the most frequent targets of somatic mutation in BCP-ALL. In 2-3% of the cases genetic rearrangements lead to the expression of fusion transcripts encoding fusion proteins (chimeras), in which the N-terminal region PAX5 is fused to one of a number of different partner proteins, displaying an extreme functional heterogeneity. The main aim of this project was to determine whether all PAX5 chimeras share the same characteristics or whether the partner protein has an impact on the molecular function of the individual fusion protein. On the one hand, we could show that PAX5 chimeras share some features such as their nuclear localization and DNA-binding capacity, both of which are prerequisites for their function as aberrant transcription factors. On the other hand, only some PAX5 chimeras self-interact or interact with the respective wild-type partner protein, suggesting that not only the function of PAX5 but also that of the partner protein may be impaired. Gene expression profiling of primary leukemia samples harbouring different PAX5 fusions provided further evidence that the partner protein has indeed an impact on the mode of function of the individual chimera. Furthermore, we could show that PAX5-JAK2 constitutively activates JAK-STAT signalling and at the same time deregulates wild-type PAX5 target genes, thus, has a dual function. In addition, the mechanism by which PAX5-JAK2 activates the JAK-STAT pathway differs fundamentally from the canonical route by which JAK2 normally regulates this signalling cascade. In this context, we could show that in contrast to JAK2, which resides in the cytoplasm, PAX5-JAK2 exclusively localizes to the nucleus. Importantly, specific inhibitors can efficiently block the activity of PAX5-JAK2, which may open up new avenues for therapeutic approaches. In summary, the research conducted within this project has confirmed our hypothesis that the partner protein is indeed relevant for the function of the individual PAX5 chimera, and, therefore, it is highly conceivable that the respective leukemia(s) differ in terms of their biology and pathogenesis.