In proteomics, we will develop a platform for protein forward arrays in which a collection of antibodies spotted on microarrays are used to monitor expression of particular protein isoforms (e.g., phosphoproteins, ubiquitinated proteins etc.). This technology has already been successfully applied for profiling activation of both receptor tyrosine kinases and downstream signaling molecules. Similarly, we will develop a platform for reverse protein arrays, which are based on the principle that complex protein mixtures such as cell lysates are spotted in an array format and probed with selected antibodies in a multiplexed manner. These reverse protein arrays will be used to monitor dynamic network responses. Both technologies require the availability of highly specific antibodies that can be obtained from commercial sources. These antibodies will be screened for specificity and validated in several experimental settings (e.g., cell lines, tissue extracts, cross-species reactivity). Future applications will depend on the generation of broad collections of high quality antibodies, for instance in projects such as the ‘Human Protein Atlas Initiative’ (URL: www.proteinatlas.org, part of the Human Antibody Initiative of the Human Proteome Organization, HUPO). To promote access to such data it is our goal to further develop our own collection of phosphoprotein affinity tags based on such technologies as DARPINs. We will collaborate with Molecular Partners, a Swiss biotech company, to develop a large collection of ankyrin repeat-based high affinity target binding proteins. Such proteins will be selected from complex phage expression libraries and can be used for analyses such as immunoprecipitation, western blotting or immunofluorescence microscopy. DARPINs specific for a series of phosphopeptide sequences will be selected and produced in bacteria. These reagents will complement the antibody collections used in our protein arrays and can be used, for instance, to monitor protein phosphorylation in signaling networks.
Access to MS services for proteomics will be essential for our project and is guaranteed through facilities at the FMI and Biozentrum. In both places, new techniques (e.g., for phosphoproteomics or lipidomics) will be implemented for the benefit of the Kompetenzzentrum and then the NCCR. The Department of Chemistry of the University of Basel will be available for joint projects in which low molecular weight molecules or peptides will be used for system analysis.
We will use flow cytometry-based single-cell proteomics and fluorescence microscopy based screening technology for analyzing signaling molecules in live and fixed cells. These technologies, which can be automated for high throughput analysis, allow measuring a limited number of signaling events simultaneously in single cells. This technology will complement our protein array platforms.
Signaling research also relies on the systematic analysis of gene function in signaling pathways and cellular processes, made possible by the development of cell based genome-scale approaches with over 20’000 individual genes interrogated in highly multiplexed experiments. Such technologies aim to quantify the effects of either over expression of individual proteins using full-length cDNAs or the inhibition of gene expression by RNAi reagents. They require the creation of genome-scale collections of reagents and their optimization through computational approaches. These reagents can be applied in, for example, reporter-gene assays and in phenotypic screens, where the read out is based on image analysis. To leverage these approaches we will develop an experimental platform in collaboration with SystemsX.ch and selected commercial partners for the synthesis of RNA- and DNA-based reagents.
The implementation of the genomic, proteomic and metabolomic methods in the network requires evolution of existing methods and development of new algorithms to analyze the experimental data. Integrating data from different experiments is a challenge that will be addressed, for example, using Relevance Networks or Bayesian inference to uncover new pathways. Methods to overlay gene expression data with genome-wide transcription factor location data obtained by ChIP-on-Chip experiments will be used to identify previously unknown regulatory networks. Such methods rely on the growing number of databases – of which KEGG (URL:http://www.genome.jp/keg/kegg2.html) is probably the best known – that collect such information from scientific publications using literature mining and manual curation. This approach – called Gene Set Enrichment Analysis (GSEA) – removes the undue bias of selecting individual up-regulated genes by focusing on entire sets of genes. In addition, more recent clustering algorithms based on unsupervised methods are able to identify network modules without the prior manual (and therefore biased) selection of gene sets. Taken together, these approaches enable the discovery of co-regulated network entities and thereby allow unraveling new signaling pathways. Furthermore, signaling networks will be described as mathematical models using, for example, ordinary differential equations and/or agent based technologies. These models will be informed with the dynamic data obtained from our proteomics approaches and simulations will be run to verify the validity of the models and to generate new hypotheses. Indeed, systems biology is an essential aspect of any initiative in modern signaling biology and the inclusion of systems biology as one leg of the proposed center is intentional and important.