For some agonists these stable GPR responses were resistant
For some agonists, these stable GPR119 responses were resistant to washing. Thus, sustained activation could continue, for at least a number of hours, even after removal of free agonist. These data are consistent with reports for other GPCR systems that do not desensitize (Calebiro et al., 2009, Ferrandon et al., 2009, Mullershausen et al., 2009). Much of the emphasis in the literature is focused on the role of receptor binding kinetics in antagonist function. However, whilst an understanding and role for GPCR agonist binding kinetics is largely overlooked, there is some precedent for this effect in the literature. Sustained receptor responses linked to agonists with slow dissociation rates have been reported for the calcitonin (Andreassen et al., 2014), parathyroid hormone (Dean et al., 2008, Ferrandon et al., 2009, Okazaki et al., 2008), mealocortin (Haskell-Luevano et al., 1996) and muscarinic M1 (Jakubik et al., 2002) receptors. An inability of washing/dilution to successfully remove the agonist from the vicinity of the receptor, rather than persistent occupation of the receptor binding site per se is an alternative mechanism to explain sustained wash-resistant signalling. Rebinding, where upon dissociation from the receptor an agonist is more likely to bind again than diffuse away and discontinue its influence (Vauquelin and Charlton, 2010); or membrane deposition, where cellular membranes act as a reservoir of the agonist that can later access the receptor binding site (Anderson et al., 1994) have both been postulated to play a role in sustained wash-resistant agonist responses. A key distinction between the roles of binding kinetics and rebinding/membrane deposition is that the former involves pre-established binding interactions formed during the initial agonist incubation that remain after removal of unbound ligand, whilst the latter relies on new binding interactions being made after removal of excess ligand as unbound ligand is not effectively removed by dilution of drug in the surrounding fluid. These post-washing binding events can be identified by responses that are readily reversed by antagonist. As the responses to MBX-2982 and AZ1 were reversed by antagonist, this suggests that agonist dissociation events occur over the time-course of the experiments as receptor gp120 are left unoccupied for antagonist binding. Our observations that rapid washing (infinite dilution protocol) was less effective than many washes over time (“chronic incubation/washout” experiments) were also consistent with rebinding/membrane deposition being involved in the sustained effects of the agonists MBX-2982 and AZ1 as they highlighted that free concentration of the ligand was not the only factor driving the effect. Moreover, these findings indicate that receptor-independent mechanisms are involved, and cannot be explained by binding kinetics alone. Evidence that membrane deposition of MBX-2982 and AZ1 may play a prominent role over rebinding come from findings that the sustained effects are sensitive to a reduction in buffer volume during chronic pre-treatment, suggesting that a large number of agonist molecules are needed to evoke the response via their accumulation in cell membranes (Teschemacher and Lemoine, 1999). High lipophilicity is often cited as a key determinant in ligand membrane deposition as it favours interactions with lipid membranes. Indeed we find that the lipophilicity (logD values) and protein binding of MBX-2982 and AZ1 are high (Sup. Table S1) and could accommodate this role. However, other agonists that do not exert a sustained wash-resistant effect also have high lipophilicity (e.g. AZ2), so this is unlikely to be the sole mechanism. Membrane deposition has been proposed as a mechanism of the prolonged cellular and clinical effects of long-acting β2-adrenergic receptor agonists such as salmeterol (Anderson et al., 1994, Szczuka et al., 2009, Teschemacher and Lemoine, 1999). Whilst a number of GPR119 agonists, including GSK-1292263 and MBX-2982, have advanced to phase II clinical trials for type-2 diabetes, it appears that a common problem with these ligands may be a lack of clinical efficacy (Kang, 2013). Indeed, clinical development of GSK-1292263 has been discontinued. It is tempting to speculate that efficacy could be improved by targeting ligands with enhanced sustained effects, as this could provide more prolonged receptor signalling. There is a growing appreciation that targeting ligands with a long residence time could be useful strategy in drug discovery for increased target coverage and decreased toxicity (Copeland et al., 2006, Guo et al., 2014). We find that the sustained effects of GSK-1292263 are weak and not statistically significant, and may therefore limit its ability to activate GPR119 in a physiological setting. Although MBX-2982 did cause robust sustained wash-resistant receptor activation, the effects with AR-231,453 were higher in magnitude (Fig. 4A), as well as being more stable, more potent and less susceptible to exposure conditions (Fig. 5, Fig. 6). Encouragingly, for AR-231,453 we also observe sustained responses downstream of cAMP in an endogenous expression system, which supports the potential (patho)physiological relevance of this phenomena.