Determining muscarinic receptor agonist subtype-selectivity by rational data analysis

Greg Osborne, Richard Mould, Edward Hurrell, Kirstie Bennett, Christopher Langmead. Heptares Therapuetics, Pharmacology, United Kingdom.

Defining the selectivity of a ligand to interact with a target receptor in comparison with other receptors is an important challenge in drug discovery. Agonist potency as measured in a biochemical or cellular assay is comprised of both agonist affinity and efficacy. Furthermore, efficacy depends not only on the ability of an agonist to activate a receptor, but on the level of receptor expression and the degree of signal amplification between the receptor and the endpoint being monitored. Hence, a simple comparison of agonist potencies across receptor, cellular and assay systems can be highly misleading. Using muscarinic acetylcholine receptor agonists as a prototype, a data analysis method has been developed which can determine a more robust measure of agonist selectivity.

Agonist affinities for the M1-M4 receptor subtypes were determined using [3H]NMS inhibition binding assays in CHO-hM1 – hM4 cell membranes (7 μg/well) in Krebs buffer containing 200 μM GTP and 0.5 nM radioligand. Non-specific binding was defined by 10 μM atropine. After 1h incubation at 25°C, assays were terminated by filtration through 96-well GF/B filter plates and washed with 5 × 0.5 mL water. Plates were dried and bound ligand measured using a Microbeta counter. Inhibition curves were fitted to a four-parameter logistic equation to determine IC50 values which were converted to KI values using KD value of a 0.1 nM as determined by saturation binding (n = 3). For functional ERK1/2 phosphorylation assays, cells were plated at 25K/well in 96 well half-area tissue culture plates, incubated for 4h at 37°C then serum starved overnight. Cells were lysed following a 5 minute incubation period (37°C) with compound added using a matrix channel pipette and lysate transferred to a 384 well microplate before addition of p-ERK detection kit in low light (PerkinElmer SureFire). Plates were foil sealed and incubated at 25°C for two hours prior to measurement on a PHERAStar plate-reader. Data was fitted to a four-parameter logistic equation to determine EC50 and % Emax with respect to acetylcholine. Slope factors were typically in the range x.x to x.x.

Agonist efficacy was estimated using the simple index (Emax × KA)/EC50. Measurement of affinity and efficacy values relative to the endogenous agonist, acetylcholine, accounts for any system-dependent variation and yields an estimate of the Relative Activity (RA) of the test agonist compared to acetylcholine (KACh/KTest × EfficacyTest/EfficacyACh). Affinity estimates cancel to reduce the expression to (EmaxTest/EC50Test × EC50ACh/EmaxACh); however determination of a pKI provides a guide as to whether any observed selectivity is affinity-based. RA values are analogous to RAi values as described by Tran et al. (2009). RAi values, however, estimate relative affinity for a specific receptor-G protein complex, whereas RA values take no account of this. However, the RA values generated for a range of agonists across M1-M4 subtypes allow for a much closer approximation of subtype-selectivity than a comparison of potency values (Table 1).

Table 1: Relative Activity and affinity estimates calculated for six muscarinic receptor agonists across the M1-M4 subtypes. All RA values are generated from affinity and potency estimates where n ≥ 3 at each target.