Affinity proteomics of cell signalling and cancer

In order to derive full benefit from the unique collection of binders generated in this consortium, ideally suited to target specific functions, they will be applied in a functional proteomics approach focused on signal transduction and cancer. Data obtained using binders will complement those from other approaches to extract the biological meaning of normal and aberrant signalling behaviour, providing new opportunities for molecular diagnosis and therapies. Characterisation of signal transduction proteins is of enormous biomedical importance, both for their roles in normal cell activities and because dysregulation is linked to major diseases, including cancers, diabetes and inflammatory disorders. They are major targets for drug development, as their selective inhibition or activation modulate pathogenic events. In a recent study investigating somatic mutations in all protein kinase genes in a range of cancers, 119 had mutations considered causal (‘drivers’). Because of the similarity among the 503 protein kinases, specificity of binding reagents is a particularly important consideration, yet existing (commercial) antibodies often do not discriminate effectively and in many cases no quality antibodies are available. Moreover, because of the high similarity of binding sites, there are few specific inhibitors available, and the use of RNAi is limited because of the partial nature of its inhibition and its limitation to newly synthesized protein. There is a parallel requirement for specific binders to SH2-containing proteins, which recognise the phosphotyrosine products of protein kinase action and couple them to effectors. They regulate the effects (on/off, kinetics, location, pathways) of kinase action and are central in spatial organisation of dynamic signalling complexes; mutations are associated with leukaemias, solid tumours, diabetes, neurodegenerations and immunodeficiency. Phosphotyrosine pathways are completed by tyrosine phosphatases, where pathological outcomes of mutations resemble those for kinases; the focus of the pharmaceutical industry on the phosphatases was strong during the 1990's, but failures in optimising chemical inhibitors for this target class has led more recently to a diminished interest from the private sector. Thus, the mission for the public sector should clearly be to discover novel approaches to target this important family of enzymes, and thereby open up new avenues for treatment. Other cancer related proteins which will be potential targets in AFFINOMICS have also been catalogued.

In order to target specific functions of proteins in interaction networks, such as the kinome, the analysis will make use of different types of binder specificity which can be tailored to the task. Reagents which are highly specific for a given kinase, phosphatase or SH2 protein would be of great utility in the total quantitation of those targets, e.g. by binder arrays (ULUND, BBT, DKFZ), as well as in spatio-temporal studies (UU, UNIKASSEL). The interaction of a kinase with a scaffolding protein, with its activating upstream kinase and with its downstream substrate may be visualised, giving an additional dimension of observing the action of drugs. Ideally, a particular molecular inhibitor might interrupt only one of these interactions, which might be made visible directly inside a cell. A second type of binder discriminating between forms of the proteins with and without typical PTMs, such as protein phosphorylation, could be used to observe the effect of upstream activators or inhibitors (e.g. from extracellular ligands) and identification of disease-specific variations. As noted , binders capable of a biological effect when expressed inside the cell will be particularly useful, both in tracking the target and producing functional intracellular inhibition in cell lines; the development of intrabodies will allow high throughput functional screening, complementing RNAi to provide insights into aberrant pathways and suggesting keys to the targeted use of small chemical inhibitors. Intrabodies that recognise docking domains in protein interactions promise to be powerful tools for disrupting the pathways and complexes and probing the biological role of the more poorly defined kinases. DARPins with this property have already been isolated (UZH). Intrabody binders would clarify the flux of information in kinase signalling and enormously improve our toolbox for quantifying these parameters, which in turn would be the input for systems biology. Binders which interfere with the functions of PTMs will most likely influence distribution of protein complexes by displacing components from their cognate intracellular location, thereby identifying potential sites for targeted disruption. An exciting study would also be to test the heterogeneity of tumour samples.

AFFINOMICS binders will be used to analyse the composition of multiprotein signal transduction complexes, down to the single cell level. In recent years, advances in binder technologies have created the potential for an unprecedented view on protein expression and distribution patterns in intact cells and tissues, and now increasingly on protein function. They include methods to study protein interaction from cells and tissues after pull-down and co-sedimentation coupled to mass spectrometry (MS), immobilised binders on arrays and actions within intact cells. Such methods taken together allow an assessment of composition and wiring of protein networks, with quantitative and topological information. In an integrated approach in WP6, using pre-existing and novel experimental strategies and several layers of information gathering, investigations will include analysis of composition of complexes, linking affinity purification with MS, while for cellular complexes in situ, detection of proximity will be achieved by DNA amplification in multiplexed proximity ligation (UU) and bioluminescence (Bioluminescence Resonance Energy Transfer, BRET) assays (UNIKASSEL). Recombinant binders expressed in the cytoplasm will be used for in cell analyses, such as target localisation and functional disruption of pathways through the effect of binders on protein interactions (UZH, VIB). Recombinant binders will also improve target crystallisation for structure determination (SGC); for computer-aided drug design, all signalling related structures are of interest and value increases with completeness. With these methodology tools in place, a means to analyse signalling pathways in a systems biology approach is provided.