G protein-coupled receptors not currently in the spotlight: free fatty acid receptor 2 and GPR35

Abstract

It is widely appreciated that G protein-coupled receptors have been the most successfully exploited class of targets for the development of small molecule medicines. Despite this, to date, less than 15% of the non-olfactory G protein-coupled receptors in the human genome are the targets of a clinically used medicine. In many cases, this is likely to reflect a lack of understanding of the basic underpinning biology of many G protein-coupled receptors that are not currently in the spotlight, as well as a paucity of pharmacological tool compounds and appropriate animal models to test in vivo function of such G protein-coupled receptors in both normal physiology and in the context of disease. 'Open Innovation' arrangements, in which pharmaceutical companies and public-private partnerships provide wider access to tool compounds identified from ligand screening programmes, alongside enhanced medicinal chemistry support to convert such screening 'hits' into useful 'tool' compounds will provide important routes to improved understanding. However, in parallel, novel approaches to define and fully appreciate the selectivity and mode of action of such tool compounds, as well as better understanding of potential species orthologue variability in the pharmacology and/or signalling profile of a wide range of currently poorly understood and understudied G protein-coupled receptors, will be vital to fully exploit the therapeutic potential of this large target class. I consider these themes using as exemplars two G protein-coupled receptors, free fatty acid receptor 2 and GPR35.

Figure 1

Many GPR35 agonists display markedly different potency at human and rodent orthologues. Representative concentration–response curves for four ligands with agonist potency at GPR35 are shown. All experiments reflect ligand‐induced interactions between transiently co‐expressed forms of β‐arrestin‐2 and the indicated species orthologue of GPR35. For human GPR35, the short isoform (GPR35a) was used, which corresponds to the single form of GPR35 expressed by rat and mouse. Zaprinast displays modest but similar potency at each orthologue, whilst each of lodoxamide, pamoic acid and ‘compound 1’ (4‐{(Z)‐[(2Z)‐2‐(2‐fluorobenzylidene)‐4‐oxo‐1,3‐thiazolidin‐5‐ylidene]methyl}benzoic acid) are markedly more potent at human GPR35 than at mouse GPR35 and, as such, are inappropriate to explore functions of the mouse orthologue. Data are adapted from Jenkins et al., 2010, Neetoo‐Isseljee et al., 2013 and Mackenzie et al., 2014.

Figure 2

Mouse and human orthologues of FFA2 receptors display extreme differences in antagonist ligand pharmacology. For both mouse and human FFA2 receptors, short‐chain fatty acids induce activation of a panoply of signals via different members of the family of heterotrimeric G proteins (see Figure 3). However, antagonists from two reported chemical series, exemplified by GLPG0974 and CATPB, display no significant affinity at mouse FFA2 receptors, although they are high‐affinity, orthosteric antagonists at the human FFA2 receptor.

Figure 3

Species orthologues of FFA2 receptors display G protein selectivity. SCFAs promote phosphorylation and activation of ERK1/2 MAP kinases in HEK293 cells transfected to express either human or mouse FFA2. However, combinations of studies with the G protein inhibitors Pertussis toxin (G_i) and FR900359 (G_q/G₁₁) and the G_i‐biased FFA2 agonist AZ1729 demonstrate that while activation of the human orthologue transmits this signal largely via G_q/G₁₁, for the mouse orthologue the signal is transmitted largely via G_i. The indicated size of the noted G protein subtypes illustrates the relative contribution of each to SCFA‐mediated activation of ERK1/2 MAP kinases. See text and Bolognini et al. (2016b) for further details.

Figure 4

Mouse and human orthologues of GPR35 display marked differences in antagonist ligand pharmacology. Although the tryptophan metabolite kynurenic acid is able to activate GPR35, it is more potent at the mouse orthologue that at either splice variant of the human orthologue. This has raised questions of its effectiveness as an agonist at GPR35 in human. Exemplars from two chemical series, illustrated as ML‐145 and CID‐27455687, have been shown to be high‐potency antagonists of human GPR35. However, at least in transfected cell systems, neither appears able to block agonist actions at mouse GPR35 (Jenkins et al., 2012). However, in certain reports, these ligands have been reported to block effects at GPR35 in rodent cells and tissue (see text for details). These dichotomies remain to be understood, but improved pharmacological study is likely to do so. G protein‐mediated signalling resulting from activation GPR35 in different cells and tissues can involve both Gα₁₃ and Pertussis toxin‐sensitive members of the G_i‐subfamily.