Determination of orphan GPCR signalling pathways in a ligand-independent fashion: towards better de-orphanization assays

J Hanson, N Dupuis, J Gilissen, A Derj, C Laschet, A Soni, JC Twizere, B Pirotte. University of Liège, Liège, Belgium

G protein-coupled receptors (GPCRs), characterized by seven transmembrane domains, are known to be involved in many physiological and pathophysiological processes and constitute the target of around 30% of all marketed therapies. Nonetheless, ~100 human non-odorant GPCRs have no known ligand and are called orphan [1]. Identification of a ligand is critical to understand their function and validate them as potential drug targets. This project focuses on defining signalling pathways of orphan receptors, in order to select adequate assays for ligand identification through chemical library screening. GPR27 [2], a rhodopsin-like orphan, has been prioritized in our study due to its involvement in the regulation of insulin promoter activity and insulin secretion [3], and its high expression in gastric ghrelin cells [4], suggesting an important role in metabolism.

GPCR signalling investigations are usually performed after addition of an activating ligand. In order to study signalling pathways in a ligand-independent fashion, we developed a luciferase (FLuc) complementation assay to assess the binding capacity of GPCRs to β-arrestin. We used a constitutively active mutant of G protein-coupled receptor kinase 2 (GRK2mod [5]) able to induce a constitutive phosphorylation of the receptor, a phenomenon known to induce β-arrestin recruitment. In addition, we generated a truncated mutant of β-arrestin1 (ß-arr1T) able to generate more stable complexes and thus allow us to detect weaker interactions [6].

We validated our approach using two known receptors diverging in their interaction with β-arrestin. First, the archetypal rhodopsin-like GPCR, the β2 adrenoceptor (β2AR), closely related to GPR27 and characterized by a transient interaction with arrestins (type A interaction [7]). Secondly, the chemokine receptor CXCR7, know to strongly interact with ß-arrestins (type B interaction [8]). We successfully confirmed with our assay the persistent (type B) interaction of β-arrestin binding with CXCR7 when co-transfected with GRK2mod (GRK2mod = 673±36 % compared to control non GRK2mod transfected cells = 100%. Results expressed in Relative Luminescent Units or RLU). Moreover, we could detect weak interactions (type A) between the β2AR and β-arr1T (200±12% RLU). When applied to GPR27, we identified an inability of this receptor to interact with β-arr1T in a GRK2-dependent manner (88±12% RLU). This is similar to other rhodopsin-like GPCRs, including the β3 adrenoceptor (β3AR, 76±4% RLU) and we classified it as atypical. These results have been confirmed by confocal microscopy.

This project aims to identify the signalling pathways of orphan receptors, with the aim of designing adequate pharmacological assays to facilitate deorphanization. Luciferase complementation between GPR27 and β-arrestin could not be enhanced by a constitutively active GRK2, suggesting that either other GRKs are necessary to phosphorylate GPR27 or β-arrestin binding is absent for this receptor. Precise identification of signalling pathways is an absolute prerequisite to de-orphanization campaigns. Our original assay will be used to characterize all the rhodopsin-like orphan receptors with regard to β-arrestin interaction (type A, type B or atypical). Furthermore, other GRK subtypes and similar assays to identify signalling partners will be used to extend our investigation of GPR27 signalling pathways.

[1] Davenport, et al. (2013). Pharmacol Rev, 65, 967. [2] O\'Dowd, et al. (1998). Genomics, 47, 310. [3] Ku, et al. (2012). PLoS Genet, 8, e1002449. [4] Engelstoft, et al. (2013). Mol Metab, 2, 376. [5] Oakley, et al. (2006). Methods Enzymol, 414, 50. [6] Oakley, et al. (2001). J Biol Chem, 276, 19452. [7] Oakley, et al. (2000). J Biol Chem, 275, 17201. [8] Rajagopal, et al. (2010). Proc Natl Acad Sci U S A, 107, 628.