The organisational structure of class A GPCRs: Implications for pharmacology, function and therapeutic regulation

Proteins located at the surface of cells act as receptors for information provided by the presence of a bewildering number of hormones, neurotransmitters and various other cell modulators. In certain cases, such as the single transmembrane domain receptor-tyrosine kinases, it is well established that for binding of the messenger to be transmitted into a response inside the cell, the receptor has to dimerise, i.e. two copies of the receptor must come together into a complex and interact. The family of G protein-coupled receptors (GPCRs) are the targets for key hormones and neurotransmitters that control everything from heart rate to emotional responses. Given their importance in physiology and as the targets of a very broad group of medicines that alleviate major diseases, these proteins have been studied extensively for many years. Indeed, in a drive to both understand the molecular basis of action of medicines that function by binding to GPCRs and also to improve the next generation of such medicines, in recent years enormous progress has been made via crystallising certain GPCRs so that amazing detail of their structures can be observed. In a number of recent cases the receptor has crystallised as either a dimer (two copies) or a tetramer (containing four copies of the receptor) with well defined contacts between the individual copies of the receptor. Despite this, and clear evidence that GPCRs can and do exist as dimers or even, indeed, as multimers in living cells, it is also well established that such interactions are not absolutely required for their primary function. As such, key outstanding questions that so far lack answers include, what range of sizes of complexes of GPCRs exist, does this vary between different members of the family, does this vary in different cells and tissues in which the receptor is expressed and, if so, what are the consequences of this for function in both health and disease? However, as the moment there is little agreement on these topics. The first substantial component of the studies proposed is to address and answer each of these questions. These studies are designed to answer both fundamental questions about these receptors but also to explore the significiance of the answers for the effectiveness of various medicines and how such variation might be used to develop improved medicines.
Integral to developing approaches to address many of these questions has been a series of recent efforts to better incorporate mathematics into analysis of biological processes and the appreciation that photographic images of cells expressing such receptors labelled with fluorescent markers often contain far more information that is usually abstracted from them. Thus, in preparation for this application we have developed ways in which detailed mathematical analysis can uncover hitherto unappreciated insights into the size and the shape of a receptor complex inside a single cell. In isolation, this would be interesting but insufficient. Therefore, the second major element of the work will be to use information from the crystal structure data to attempt to generate modified receptors that disassemble or that prevent the formation of such complexes and then use these modified forms of the receptor to assess the functional consequences. Such studies will take advantage of a wide array of approaches in pharmacology, biochemistry and cellular signalling assays that my team and I have built up and used over many years. The other major component of the proposal reflects that individual GPCRs do not only interact to form complexes with other copies of the same receptor. They can also interact with other members of the GPCR superfamily to generate heteromers. Such heteromers have been reported to display very distinct pharmacology and function than the corresponding homomers and we will address similar questions as above for such complexes incorporating GPCRs for the neurotransmitters dopamine and glutamate.

Basic information on the ability of G protein-coupled receptors (GPCRs) to form dimers or higher-order complexes has been available for some time. However, there is currently widespread variation in conclusions on the extent, stability, size, shape and functional consequences of such interactions. Integration of mathematics with improvements in optical microscopy have recently begun to offer means to address many of these issues. In preparation for this application we have employed multi-photon FRET with spectral resolution to interrogate single pixel images from plasma membranes of cells. Analysis of such images indicate that the M3 muscarinic acetylcholine receptor exists as a mixture of dimers and tetramers that are organised as rhombi. We will now address if this is true for other class A GPCRs and if the proportions of the species differ at different expression levels and in different cellular compartments. Within the proposed programe of work we will also integrate each of faster scan speeds, total internal reflectance and two quasi-simultaneous excitation wavelengths to determine the total concentrations of donors and acceptors and to employ single molecule tracking studies. As an alternate approach we have been developing the use of Spatial Intensity Distribution Analysis (SpIDA) to address similar questions. Given information on the interfaces that may generate quaternary structure that are becoming available from both atomic level structures of GPCRs and models generated from these, we will employ mutagenesis to disrupt quaternary complexes and then employ detailed biochemical and pharmacological studies, including the generation and use of intramolecular FRET sensor forms of GPCRs to assess the effects on allosteric communication within receptor complexes and how this influences the potency and efficacy of receptor ligands, including currently employed therapeutic medicines, as well as if this may influence or define the signalling pathways activated.

As G protein-coupled receptors are routinely described as the most tractable group of targets for the development of therapeutic small molecule drugs, then the 'drug discovery' industry is clearly the section of the commercial sector and end users of basic research that will benefit most substantially from the proposed research. Milligan has developed wide ranging contacts and links with a large number of both UK-based and multinational companies that are attempting to develop both novel ligands and different approaches to target GPCRs for the treatment and amelioration of a broad range of human diseases. New understanding of the structural organisation of GPCRs, above and beyond what is being uncovered by the definition of the ligand binding pockets of individual receptors via crystallography, is integral to efforts to gain a competitive advantage in this area. The pharmaceutical industry is therefore greatly interested in all novel and developing ideas and concepts in GPCR research including applications of 'allosterism', 'ligand bias' and the implications of quaternary organisation and how this might be controlled. Thus, for example, efforts to identify ligands that selectively target GPCR heterodimers have employed an approach patented by Milligan/University of Glasgow and then licenced to a company which will launch an Intial Public Offering (IPO) of stock in early 2014. Equally, the concept that GPCR heterodimers display pharmacology and function that can be quite distinct from that anticipated from either partner receptor in isolation has both encouraged pharmaceutical companies to explore if 'unexpected' behaviour of various ligands in animal models of disease and, indeed, in clinical trials might reflect interaction with either a GPCR heteromer or with a distinct quaternary structure of a GPCR homomer. Although challenging to define, recent examples of ligands that display markedly different characteristics in either potency or efficacy at GPCR heteromers have posed questions but also helped to confirm the physiological relevance of such complexes. Equally the concept of 'allosteric' effects of ligands within dimers and/or high order complexes is one that may have substantial application in helping to define why what may appear to be the same receptor generates different responses in different cells and tissues. The questions posed in the current application are challenging and the implications far from clear at this point, but it is likely that the outcomes will be assessed carefully within the pharmaceutical industry. Novel therapeutic medicines, particularly with reduced side effect profiles, may well be a longer term impact of the work proposed. However, as with translating many aspects of basic understanding of key drug target classes, this is unlikely be to realised in less than a 10 year horizon.
The staff employed on the project are already highly qualified in areas of molecular pharmacology and cellular imaging and indeed, are currently working at the absolute forefront of the research areas proposed. Their skills could clearly be redeployed in many areas of molecular cell research. As all of the staff proposed to work on the project are currently members of the Glasgow 'Molecular Pharmacology Group' they actively work closely together as a team, as detailed both in the Case for Support and in Justification of Resources', and as is evident from the substantial list of joint publications and other outputs they have produced.