Final Activity Report Summary - CELLULAR SIGNALING (Comparative functional genomics of cellular signalling pathways) Signalling pathways play an important role in development, cell communication and homeostasis and their activity has to be tightly controlled. Cancer and many other human diseases are often caused by aberrations in cellular signalling pathways. The analysis of cellular signalling systems is therefore of key importance both for the understanding of principle regulatory mechanisms and for the development of targeted therapeutic approaches to revert pathophysiological aberrations. Studies in model organisms such as Drosophila, Zebrafish and mice will contribute significantly to the understanding of cellular signalling in vivo. The Wnt signalling pathway plays an important role in Drosophila and human during development and has been implicated in several forms of human cancer, like colorectal cancer. This pathway is highly conserved during evolution and controls diverse biological processes, including cellular differentiation, stem cell behaviour and cell proliferation. Wnt proteins are secreted factors that activate target gene expression in the receiving cell. We have developed cell-based assays using Drosophila and human cells suitable for high-throughput pathway analysis with a focus on the Wnt signalling pathway. We performed genome-wide RNA interference (RNAi) screens using our in-house developed dsRNA library for Drosophila screens and a commercial synthetic siRNA library for human cells to systematically knockdown all predicted genes. For computational analysis of large-scale RNAi screens we have developed and continuously improved data analysis software and databases that store and present phenotype data. The open-source software packages cellHTS automates the analysis of high-throughput cell-based screens and is also broadly used in the screening community http://web-cellHTS2.dkfz.de. The RNAi database GenomeRNAi developed in our group stores and publicly presents available RNAi phenotypes from cell-based RNAi screens in Drosophila and Homo sapiens http://www.genomernai.org. RNAi screening data analysis resulted in a list of candidate genes that were validated by retest and rescue experiments. For in vivo analyses we focused on highly conserved candidates that passed retest studies such as knockdown specificity and efficiency measured on mRNA and protein level. A few candidates from Drosophila dsRNA and human siRNA screens were characterized in detail by applying biochemical and in vivo studies to reveal their function in the respective signalling pathways. Indeed, experiments on tumour samples provided a link of some candidate genes to cancer development. In particular we have identified the potential transmembrane protein CG6210 as a novel positive regulator of the Wnt signalling pathway and named it Evenness interrupted (Evi) due to its in vivo phenotype (Bartscherer et al., Cell 2006). It is required for Wnt signalling in flies and in humans, and we confirmed Evi as a new core component of this pathway. Evi mutant flies have patterning defects that phenocopy Wnt loss-of-function alleles, and fail to express Wnt target genes. Evi function seems to be specific for Wnt signalling. We narrowed down the site of Evi action to the secreting cell, and showed that producing cells retain Wnt in the absence of Evi. Understanding the function of Evi will be the key to an understanding of how Wnts are secreted. Experiments have been initiated to generate Evi knockdown and overexpression mouse models to functionally study this Wnt signalling regulator in vertebrate models. Taken together we have successfully developed approaches to systematically dissect signalling cascades using genome-wide perturbation studies by RNA interference in Drosophila and human cells. We have identified novel and core components that have been confirmed in vivo and found novel connections between pathways that have been previously not described. These results are being used to study Wnt signalling in vertebrate models and to develop novel disease models.
