Validation and quantification of ligand-biased signalling and allosteric modulation at CB₁ cannabinoid receptors
thesisposted on 27.02.2017, 05:12 authored by Khajehali, Elham
The CB₁ cannabinoid receptor (CB₁R) is a potential target for the treatment of numerous central nervous system disorders. Although a large number of CB₁R ligands exist, their therapeutic applications are limited due to adverse on-target effects. Selective activation of the receptor signalling events that mediate desired therapeutic effects at the expense of those that mediate adverse effects may overcome adverse on-target effects. This could be achieved via a phenomenon referred to as ligand-biased signalling. There is growing evidence that CB₁R ligands may activate selective signalling pathways and engender biased signalling (Bosier et al., 2008b). More strikingly, CB₁R allosteric modulators, such as Org27569 may also display pathway selective modulation or biased allosterism (Ahn et al., 2012). Several endogenous allosteric modulators at CB1Rs have also been suggested, including pregnenolone (Vallee et al., 2014), lipoxin A4 (Pamplona et al., 2012) and CRIP1a (Niehaus et al., 2007). The current investigation aimed to detect and quantify ligand-biased signalling and allosterism at CB₁Rs using sophisticated analytical methods, in order to establish potential CB₁R biased “fingerprints” that may guide structure-activity and drug discovery studies. Our results showed that 2-AG and WIN55,212-2 had little preference for cAMP inhibition and pERK1/2 activation (bias factor not dissimilar from 1). However, anandamide, Δ9-THC, CP55940 and in particular HU-210 and methanandamide with bias factors of over 20 and 15, respectively were biased towards cAMP inhibition. We also demonstrated that Org27569 reduced the CB₁R inverse agonist [ᶟH]SR141716A binding, indicated by a binding cooperativity (α) value close to 0. However, it had little effect on the binding of cannabinoid agonists (α close to 1). Org27569 completely abolished inhibition of cAMP by all the cannabinoids tested, indicated by functional cooperativity (β) values approaching 0. Interestingly, however, in pERK1/2 assays, Org27569 abolished the response to HU-210 and CP55940, had no significant effect on pERK1/2 activation by anandamide, methanandamide and Δ9-THC, and only partially inhibited 2-AG and WIN55,212-2-induced pERK1/2 activation, as indicated by β values ranging from 0 to 1. This clearly indicates strong probe-dependence and biased allosterism by Org27569. Furthermore, our results showed no inhibitory effects on Δ9-THC-induced pERK1/2 activation by pregnenolone, and no enhancing effects on anandamide-mediated inhibition of cAMP by lipoxin A4, in contrast to previous findings (Pamplona et al., 2012; Vallee et al., 2014). Our results also demonstrated that CRIP1a knockdown in NG108-15 cells abolished WIN55,212-2-induced cAMP inhibition and reduced KCl-induced Ca²⁺ influx. However, it had no effects on cannabinoid-mediated Ca²⁺ mobilisation. In recombinant HEK-CB1-TREx CRIP1a cells, cannabinoid-mediated cAMP and pERK1/2 signalling was unchanged in the absence or presence of CRIP1a. Therefore, further research is required to verify the allosteric nature of these endogenous ligands. In conclusion, this study quantifies, for the first time, ligand-biased signalling from CB1Rs, provides quantitative insights into biased allosterism and probe-dependence by the small molecule Org27569 at CB1Rs, provides evidence against the reported allosteric effects of the endogenous ligands lipoxin A4 and pregnenolone, and demonstrates the cell line-dependent effects of CRIP1a. These novel insights may contribute to the development of selective CB1R-targeted therapies.