The chaperone cycle of fibroblast growth factor receptor kinases in molecular detail

As cells in our body age or encounter external stresses, such as toxins derived from diet or the environment, proteins within the cell can accumulate damage (for example mutations) or alterations in expression levels that can cause them to form aggregates that are damaging to cell function and survival. Diseases of aging such as Alzheimer's and Parkinson's are examples of conditions associated with the formation of large, aggregated protein structures within certain groups of neuronal cells, leading to manifestations including dementia, tremor or loss of motor control. In response to stresses, cells marshal a set of proteins called chaperones whose function is to rescue such aberrantly folded or aggregated proteins. Often, however, the chaperone response itself can become compromised through accumulation of cellular damage, which can blunt the organism's ability to remediate protein aggregation and lead to disease progression with advancing age.

One such arm of the chaperone response involves the protein HSP90, which is assisted in its recognition of 'client' proteins by various co-chaperones that include Cdc37. Cdc37 is the specific co-chaperone for an important family of proteins involved in cellular signalling known as kinases. Receptor tyrosine kinases (RTKs) are a type of 'signalling' kinase that transduce signals from outside to inside cells and include fibroblast growth factor receptor kinases (FGFRs), the subject of this study. FGFRs are important in embryonic development, wound-healing, generation of new blood vessels, and in altered forms are responsible for driving several types of cancer. We are interested in establishing the molecular details of how FGFRs interact with the Cdc37-HSP90 chaperone system, and in particular in how Cdc37 is able to distinguish between forms of FGFRs that require chaperone intervention (for example, altered forms of FGFRs responsible for diseases such as cancer) and those that do not (for example, wild-type variants of most FGFRs).

Our project seeks to fill in some of the important gaps in our knowledge, by using a range of cutting-edge technologies and approaches. In particular, cryo-electron microscopy (cryo-EM) has recently undergone a dramatic improvement in the level of structural detail it can provide, thanks to new technologies and methods, and in Leeds and UCL we have access to the latest-generation cryo-EM equipment and expertise. Other structural techniques such as NMR and mass spectrometry have also seen significant advances in capability, and again, we have expertise and state-of-the-art equipment in Leeds and UCL to undertake this work. Moreover, over the last 5 years we have built up a strong body of pilot data, generating the components required, in-particular the protein complexes, which are often the limiting step in these challenging projects.

Understanding the molecular-level 'rules of engagement' of Cdc37 and the chaperone system with FGFR kinases and their altered forms will help us to understand not only the normal functioning of the chaperone system in respect of this important class of protein clients (RTKs), but also how it might become overwhelmed or subverted in acute or cumulative cellular stress. Such understanding can also assist in efforts to target the chaperone system selectively to treat cancers driven by oncogenic kinases that are 'addicted' to chaperone intervention to maintain their activity.

In this 3-year project grant we will combine and integrate insights from an array of advanced structural techniques, including cryo-EM, H-D exchange- and ion mobility-MS, and methyl-resolved and paramagnetic NMR, to effect a step change in our understanding of how FGFRs (hence RTKs more broadly) interact with the Cdc37-HSP90 chaperone system.

Over a number of years we have established in our labs the means to generate heterotrimeric/tetrameric complexes that represent defined points on the kinase-co-chaperone-chaperone interaction cycle through in vitro reconstitution of individually expressed and purified components carrying functional mutations, e.g. E47A and D93N ATP-binding/hydrolysis HSP90 mutants, I538F DFG-latch mutant FGFR3, thus overcoming what can often be a major limiting factor in structurally-based projects.

Our methyl-resolved NMR platform will enable us to define the rules for discrimination by Cdc37 between kinase variants on the strong-weak client continuum, and how these features are selected for presentation to HSP90. Furthermore, we will be able to probe the effect of HSP90 inhibitors such as PU-H71 on the full complex structure and dynamics, and answer the question of whether there is obligate order and homogeneity in the assembly of the kinase-co-chaperone-chaperone complexes.