How do cells shape and interpret PIP3 signals?

Multi-cellular organisms rely on a large array of different transmitter substances to allow certain cells to control the behavior of others. The more sophisticated the organism the more complex the cell to cell communication. In mammals this language probably involves hundreds of fundamentally different types of transmitter. Clearly such systems need a large collection of specialized receptor molecules that can detect the individual presence of any particular transmitter. Further, these receptors, typically found on the outer surface of the cell's limiting membrane, have to signal their specific stimulation by passing a molecular message into the cells interior, effectively informing the cell that the receptor has been activated. Clearly, if a cell has many different types of receptors on its surface the molecular signal generated inside the cell by each different receptor (often called an intracellular message) must identify and distinguish which specific receptor has been stimulated. Otherwise the cell could not discriminate between the transmitters present on the outside of the cell and could not respond correctly. Hence, mammalian cells have vastly complex intracellular signalling mechanisms continuously informing the cell of what is happening in other parts of the organism or its environment. One such intracellular signalling molecule or 'message' is PIP3. It is a phospholipid molecule found on the inside surface of the cell's limiting membrane. Levels of PIP3 rise rapidly on activation of a large number of receptors. This is surprising given the problems the cell faces in knowing precisely which receptor has been activated when it detects an intracellular signal. This grant application is to understand how it is possible that rises in PIP3 can encode specific messages from so many different receptors. We have performed some experiments that have, in fact, shown that PIP3 in cells is not a single type of molecule. At least four tiny variants of PIP3 can be detected, called molecular species of PIP3. Interestingly, we find that these different molecular species of PIP3 do not respond equivalently to different ways of activating the cells we work with. We and others have also found that the different receptors can make the levels of PIP3 rise for different times and to different maximum levels. We propose that these small differences are very important inside the cell for discriminating whether a certain receptor has been stimulated. This is a 'clever' economy or efficiency on the part of the cell and allows it to use similar mechanisms to perform many different jobs. Although on the surface these might appear trivial details in the business of understanding biology, it has recently been discovered that many different cancers are caused by mutations in genes that regulate PIP3 levels in cells. Mutations that by chance cause the production of PIP3 to be increased without any need for receptor stimulation make cancers much more likely to occur. Mutations that by chance stop the enzymes that normally break down PIP3 from working also make cancer more likely to occur. As a result it is clear that understanding how PIP3 is made and then interpreted by cells is crucial for us to better understand how cancer occurs and how to treat it. Many companies are already trying to design drugs that will reduce PIP3 levels to fight cancer. This work will help us understand how to make better drugs of that type.

This proposal is a collaboration between biochemistry groups at BI and mathematical biologists at the EBI to achieve a detailed and quantitative understanding of a major mammalian signal transduction pathway, the PI3K network. Several PI3K isoforms exist in cells that can be selectively engaged by a variety of cell surface receptors to generate the membrane phospholipid PIP3. PIP3 is the initial signal, which is then transduced by 10-50 effector proteins into the regulation of complex cell responses, such as cell growth and movement. Our strategy is to focus on collecting robust, high quality data sets in a panel of isogenic, non-transformed breast cell lines (MCF10a) in which key endogenous components of the pathway can be manipulated and to embed iteration between experiment and modelling to arrive at a more satisfactory explanation of: 1) the key factors which shape the magnitude and spatiotemporal properties of PIP3 signals in response to hormonal stimulation (EGF, insulin, LPA) and oncogenic mutation; 2) The way in which different PIP3 effectors interpret these PIP3 signals and 3) the relative importance of individual PIP3 effectors in delivering regulation of chemokinesis, growth and global transcription. We plan to use homologous gene targetting, siRNA suppression and pharmacological inhibition of pathway components and measure the impact of these perturbations, in several relevant cellular contexts, on i) the levels of PIP3 and other phosphoinositides measured by a novel, quantitative mass spectrometry assay that allows systematic analysis of fatty acid composition; ii) the activity and spatial distribution of several PIP3 effectors (in some cases via knock-in of endogenous GFP-fusion proteins); iii) chemokinesis, markers of growth and global transcription (using next generation sequencing). Models will be built at several levels in the pathway and integrated to allow a deeper understanding of this network and guide more effective therapeutic intervention

1) Identify the beneficiaries of this work. See the section 'the beneficiaries'. To restate, ignoring our proximal research community, they would include, within the life-time of the grant; (a) an international and broad group of commercial and academic researchers, (b) the BBSRC, our host Institutions (Babraham and EBI), (c) the post-doc researchers on the grant through the training they receive and (d) our IPA partner Astra Zeneca. In addition, in the longer term, (e) the health care sector, patients and the UK's economic competitiveness. 2) How would they benefit? a) From the technologies and approaches we propose to apply in this application. Most signficantly the lipidomics strategies we have developed to enable sensitive, medium through-put analyses of the different molecular species of PIP3. This advance enables the development of potentially direct read-outs of the effectiveness of PI3K inhibitors in a clinical setting, through the opportunity to take frozen cell or tissue samples and sensitively and quantitatively analyse their PIP3 content. Until now companies have relied on surrogate read-outs of PI3K activity that have a variety of technical and intellectual weaknesses. This approach will also change the phosphoinositide research community, historically there has ben a huge pausity of data in the field through the technical or financial challenges in capturing this type of data. This has severly limited attempts to model this pathway. b) See beneficiaries section. The BBSRC would gain from delivery of their objectives in their recently formulated strategic plan, specifically in the Bioscience for Health priority. As an IPA application, this project has evidence of its commercial relevance as 'basic research underpinning the pharmaceutical sector'. Its use of modelling is in line with the BBSRCs drive to change the culture of modern biology towards being more mathematically based. c) Both the EBI and BI have internationally competitive research environments where the post-docs would learn within a project that has both direct commercial relevance, substantial contact time with a major pharmaceutical company and a multi-disciplinary approach. d) Will benefit from accelerated access to a internationally competitive grouping working in a field in which AZ have direct interest in their inflammation and oncology programmes and from having first option on any IP that might emerge from the project. More specifically they will gain direct leverage from results with the cell lines and inhibitors we have chosen to focus upon; both are used within their own, internal, research programmes. They will also gain early insights into the lipidomics approach we have developed and how they might use it to solve long-standing problems with bio-markers of activity in the PI3K pathway in whole animal or clinical studies. e) These could be longer terms outcomes resulting from improved focus on the most appropriate PI3K targets and hence the best PI3K-selectivity profile of potential drugs in specific therapeutic setting and the development of more relevant and useful bio-markers. As our IPA partners oncology programme is based in the UK (Alderly Edge, Macclesfield) the above should give competitive advantage to UK business. 3) How would we ensure the above potential benefits are realised a) Presentations at international science meetings, publications, good data sharing practice, seminars at companies b) Through the projects funding and execution c) Regular meetings between EBI and BI labs and with AZ researchers (see work plan) d) Through (c) and see Case for support e) The IPA status of this application and the success of past research collaborations (£300K) with AZ are based on our approach to, and conduct within, collaborations with industry. AZ have confidence we will put effort into transferring knowledge and skill and that it will give them competitive advantage in getting good therapeutics into the market-place.