Uncovering the pharmacology of the G protein-coupled receptor GPR40

G protein-coupled receptors (GPCRs) are both the largest family of transmembrane signalling proteins in man and the most successfully targetted as sites of action by therapeutically active small molecule medicines. Despite this, the ligands that activate many GPCRs remain either unknown or poorly characterised. Recently, the group of GPCRs named GPR40-GPR43 were shown to respond to medium- and long-chain fatty acids that circulate in the bloodstream. GPR40 is highly expressed by pancreatic beta-cells that produce and secrete insulin. It is known that, in the short term, elevation of blood fatty acid levels increase insulin secretion. However, in the longer term elevated blood fatty acids are detrimental to beta-cell function and it is known that the concentration of fatty acids is elevated in the blood of obese diabetics. At the moment, no GPCR targeted medicines are used to treat diabetes. However, we have recently shown that a group of clinically used medicines, called glitazones, activate GPR40 although their accepted mode of action is via activation of a completely different group of non-GPCR, nuclear receptors called Peroxisome Proliferator-Activated Receptors of the gamma subtype (PPARgamma). Although glitazones are used clinically they have side effects that limit their use and because they require to be used for significant amounts of time to produce benefits it is possible that at least some of their side effects reflect their ability to activate GPR40. Recently, models of how molecules related to medium-chain length fatty acids bind to their target GPCRs have been published. Comparing these models with the sequence of GPR40 has provided hypotheses on the molecular basis of how fatty acids interact with GPR40 and I will use combinations of mutagenesis and novel assays we have recently developed to measure activation of GPR40 to test these. I will also test if these mutations interfere with the activation of GPR40 by glitazones and by a group of small molecule chemicals also recently reported to activate GPR40. The major difference between fatty acids that activate GPR40 and those that active GPR41 and GPR43 is the length of the molecule. Comparisons of the sequences of GPR40 with GPR41 and GPR43 suggest key amino acids that may control this selectivity. This will also be tested by combinations of mutagenesis and functional assay. It can be hypothesised that a molecule that activates PPARgamma but inhibits GPR40 would be useful in the treatment of diabetes. I will thus also assess the activity of a range of other ligands that are related to the active glitazones but do not activate PPARgamma, for their ability to either activate or block activation of GPR40. Such mapping of the requirements for ligands to bind and to active GPR40 will provide new insights into how this GPCR is regulated and may, in the longer term help with further development of small molecules able to either activate or inhibit the function of this receptor.

GPR40, GPR41 and GPR43 are recently de-orphanised GPCRs. GPR40 responds to medium- and long-chain free fatty acids (FFAs) whilst both GPR41 and GPR43 respond to short-chain FFAs. Combinations of molecular modelling and mutagenesis have provided insights into the mechanism of binding of leukotriene B4 to its cognate GPCR BLT1. A key residue that acts to orientate leukotriene B4 between transmembrane helices 3 and 5 is an arginine near the top of helix 5. An equivalent arginine is present in each of GPR40, GPR41 and GPR43. Alteration of this arginine eliminates agonist function at GPR41 and GPR43 but has more modest effect in GPR40. A hypothesis to explain this is that the medium and long-chain FFAs extend substantially further down the gap between helices 3 and 5. Models of the binding of a synthetic GPR40 agonist to GPR40 suggest obvious residues that may contribute to the binding of FFAs. I hypothesise that mutation of a key residue in helix 5 will have greater effect on longer chain FFAs than those with shorter chains whilst mutation of the residue in helix 3 will have equal effects no matter FFA chain length because the helix 5 residue in deeper in the pocket. Comparisons between the sequences of GPR40 and both GPR41 and GPR43 suggest a patch of residues in helix 3 that may define chain length selectivity. This will be assessed. Particularly for GPR40 we have recently discovered that endogenous agonist ligand(s) become associated with the receptor during membrane preparation and removal of this ligand is required to 'uncover' the pharmacology of the receptor. Without this, the receptor appears, incorrectly, to be highly constitutively active. Key assays in this study will be based on this unique set of observations. Glitazones are anti-diabetic agents believed to target PPARgamma isoforms. However, they are also GPR40 agonists. Sustained activation of GPR40 is detrimental to beta cell function and I will thus also explore the SAR of GPR40/PPARgamma pharmacology.