Dynamic mechanisms of FGFR activation in cancer by kinase mutations

The way in which cells divide, proliferate and, in turn, die and become 'recycled' must be very carefully regulated in a highly programmed manner. Both development and maturation, as well as normal functioning of the adult organism, need to follow well-defined paths, and responses to environmental influences such as temperature, availability of food etc. must occur in a predictable manner. These responses require very fine control of complex cellular processes at the level of individual molecules. When this fine control breaks down, diseases such as cancer, degenerative disorders (e.g. Alzheimer's disease) and inflammatory conditions can result. Understanding these cellular and molecular processes in detail is important both to understand normal growth and development, and to provide us with insights into how serious diseases can be treated.

Fibroblast growth factors (FGFs) are protein 'hormones' produced by certain cells to stimulate the growth of other cells involved in important processes such as the development of an embryo, the growth of new blood vessels and the repair and healing of wounds. FGF molecules bind to the outer parts of FGF receptors (FGFRs), which are proteins that span across the cell's protective outer membrane, and cause FGFR molecules to pair up. The parts of the receptor proteins that are inside the cell, known as kinase domains, are then close enough to activate one another through addition of phosphate 'chemical labels' that induce a change in the shape of the kinase domains from an inactive to an active conformation, causing the kinase domains to activate other proteins in the cell in a 'signalling cascade' that tells the cell to start dividing and proliferating. In turn, this process results in the formation of new tissues. The role of FGFs and FGFRs in formation of new blood vessels is also significant in cancer, where tumour cells often artificially elevate FGFR signalling within and between themselves as a way of securing a supply of nutrients and oxygen for further growth. Starving cancers of their new blood supply by inhibiting FGFR signalling is a promising avenue for treatment, and drug companies are currently developing new medicines that inhibit the activity of FGFRs.

Although we understand some of the mechanisms by which the kinase domain of FGFR is activated from static 'snapshots' of the protein by X-ray crystallography, we still lack knowledge of how the flexibility of the kinase protein contributes to this role. Most proteins are not rigid, but need to flex to change their shape, or parts of their shape, in subtle ways to allow them to perform their functions in the cell. We will use an innovative combination of experimental methods including nuclear magnetic resonance spectroscopy (NMR), surface plasmon resonance (SPR) and isothermal titration calorimetry (ITC), together with advanced computational methods, to understand the role of flexibility of the protein in the transition between inactive and active conformations. NMR is a particularly powerful method for investigating flexibility in protein function at the level of individual atoms or groups of atoms, and here we will combine experimental information from NMR with cutting-edge computational modelling of kinase motion to describe these movements in much more detail than has been previously achieved.

By understanding the protein motions that govern FGFR kinase activity, we can understand better how FGFRs function in normal tissues and how they can malfunction in certain diseases such as cancers and developmental disorders. For example, mutated forms of FGFRs are found in many cancers. These contain amino acid changes that short-circuit the normal activation process and result in a kinase that is permanently switched 'on'. Our work will lead to enhanced understanding of how to design drugs that specifically inhibit these mutant forms of FGFR, leading ultimately to better treatments for cancers and developmental disorders.

Fibroblast growth factor receptors (FGFRs) play essential roles in a wide range of biological processes including embryonic development, angiogenesis and wound healing. As they are also key actors in a number of developmental disorders and cancers, FGFRs are important drug targets.

The core of the proposed programme is the implementation of an innovative, powerful combination of experiment and simulation to study protein kinase regulation in normal and disease biology. Using this approach, we will develop an understanding of the free energy landscapes that describe the conformational transitions accompanying rearrangements of the activation loop of wild-type and mutant FGFR1 kinases. Our combined NMR and computational approach will allow us to understand in unparalleled detail the molecular mechanism of a novel autoinhibitory mechanism for FGFR kinase activity, and of its subversion by disease-causing mutations. This understanding will improve our knowledge of how FGFRs regulate their catalytic activity in response to external stimuli under both normal and disease settings, as well as informing future strategies for development of conformationally-selective inhibitors as potential mutation-targeted therapeutics. Building on our previous published work in this area, we will address the following three main objectives:
1. To establish how the pathogenic N546K and/or N546S gain-of-function mutants of FGFR1 alter the conformational dynamics of the kinase.
2. To establish how another pathogenic FGFR1 gain-of-function mutation, R675G, alters FGFR conformational dynamics and compare with the N546 mutation.
3. To develop a comprehensive understanding of the mechanism of important structural transitions in activation and inhibition of FGFR1 in the presence and absence of disease-associated mutations, by using in silico metadynamics simulations guided by NMR-derived experimental restraints.

Scientific beneficiaries:
By elucidating novel mechanisms of FGFR kinase regulation, our work will benefit the kinase and signalling communities, and those working in areas of disease biology encompassing oncology (both oncogenesis and progression/metastasis) and developmental disorders. Enhanced understanding of the conformational energetics of allosteric kinase inhibition and its subversion by activating mutations will also benefit scientists working in the fields of kinase inhibitor design and biomolecular recognition more broadly. The methods we will employ (specifically the combination of experimental restraints from NMR and advanced computational methods) will be publicly available and will serve as exemplars for similar studies of rare conformational excursions in other biological systems.

Industrial beneficiaries:
Our project will transform our understanding of structural and dynamic features of FGFR kinases that govern activation via the 'DFG-in/out' status of the activation loop. This will, in turn, benefit ongoing efforts within the pharmaceutical and biotech sectors to develop more effective type II inhibitors of medically important kinases such as FGFRs that have intrinsically low DFG-out propensities. By exploiting regions of the kinase that are less highly conserved than the ATP binding site targeted by archetypal type I kinase inhibitors, type II inhibitors that bind to the DFG-out state afford enhanced selectivity and reduced off-target effects. In addition, our insights will inform strategies for conformationally-selective inhibition of FGFRs, potentially leading to mutation-specific therapeutics.

Delivering highly skilled people:
Our project forms part of a programme of work in the ALB lab and more widely within ACSMB on RTK-mediated signalling in normal and disease biology which entails training of UG placement students, Masters and PhD students and PDRAs in a wide range of cutting-edge scientific approaches across NMR, structural biology and biophysics. This project will train a suitably qualified and motivated PDRA in state-of-the-art NMR, biophysics, and computational methodology, and facilitate dissemination of the new computational techniques acquired through the collaboration with the MV lab into the ALB group and the wider Leeds/ACSMB community. In addition, collaboration with the ALB group and Leeds/ACSMB will bring exposure to new biological questions and experimental approaches into the MV group through the PDRA's time spent in Cambridge working on the computational aspects of the project. At Leeds, we actively promote research-led teaching for both undergraduate and postgraduate students. Hence, progress made in the field of FGFR mechanistic structural biology would enhance many aspects of the teaching delivered as well as contributing to the students' learning experience. In terms of our own research, we intend to exploit the powerful combination of experimental and computational tools that we will develop for this programme in investigations of other systems of biological and medical interest. Thus, we anticipate multiple opportunities to apply for further grant funding from research councils and charities.

Wider impacts for general public:
FGFRs are important drug targets for a range of common cancers that affect the lives of millions of patients. The prospect of improved medicines emerging as a result of the enhanced fundamental understanding of kinase regulation we will develop in this project could in the longer term benefit both length and quality of life for many such patients and their families.