Determination of ligand binding conformations at α1-adrenergic receptor subtypes based on NMR and MD studies

Abstract

Adrenergic receptor (AR) subtypes (α1A, α1B, α1D, α2A, α2B, α2C, β1, β2, β3) are G-protein coupled receptors (GPCRs) activated by the same endogenous catecholamines, adrenaline and nor-adrenaline. The two subtypes α1A- and α1B-AR maintain a complex balance in modulating the functions of the sympathetic and central nervous systems, whereby chronic activity can be either detrimental or protective for both heart and brain function. Regulation is believed to be mediated through the distinct activation of individual α1-AR subtypes and thus, subtype selective activation or blocking may have major clinical implications.

Despite having tremendous clinical importance, there are no approved α1A- or α1B-AR selective marketed drugs. Firstly, the conservation of the orthosteric binding site within a GPCR family makes it challenging to achieve receptor subtype selectivity for competitive compounds. In such a case, allosteric modulators, which interact with binding sites that are topographically distinct from the binding site of the endogenous ligand, offer an alternative. Secondly, lack of structural information of a majority of GPCR members makes it difficult to predict ligand binding. Intrinsic instability of these receptors make crystallisation challenging and as yet, no crystal structure has been reported for α1-AR subtypes while crystallising weakly binding ligands to GPCRs is difficult. The broad aim of the thesis was to gain molecular insight into the structures of weakly binding GPCR ligands by applying atomic resolution NMR methods.

In this study, we have used variants of α1A- and α1B-AR which were thermostabilised using the directed evolution method, Cellular high throughput encapsulation, solubilisation and screening (CHESS). The long-term stability of these receptors in detergents has enabled us to investigate binding of a variety of ligands including native agonists, adrenaline and noradrenaline, an α1A-AR selective agonist A-61603, as well as allosteric modulators, benzodiazepines, by employing solution-based ligand-observed Nuclear Magnetic Resonance methods including STD-NMR (saturation transfer difference NMR), Water-LOGSY (water-ligand observed via gradient spectroscopy), Tr-NOESY (transferred nuclear overhauser effect spectroscopy) and INPHARMA (interligand noes for pharmacophore mapping) experiments. These are nuclear overhauser effect based methods, which rely on the transfer of magnetisation from the target protein or other molecules (such as bulk water and ligand) to ligands through dipole–dipole interactions. STD-NMR and Water-LOGSY experiments are most commonly used to detect weak ligand binding and in fragment screening projects, while Tr-NOESY and INPHARMA are used to detect and map ligand binding conformations. The INPHARMA experiment takes advantage of known or expected orientations of a bound ligand, for example adrenaline bound to its GPCRs, to determine the respective orientation of novel ligands for which there are no known structures. We obtained experimental NMR data for the ligand A-61603 (α1A-AR selective agonist) with respect to adrenaline for both α1A- and α1B-AR. These data were compared to the back-calculated spectra obtained from molecular dynamics simulations. The results helped mechanistically explain the selectivity of A-61603 towards α1A-AR. Overall, we have shown that this solution-based methodology provides valuable information on ligand-binding poses inside the highly conserved orthosteric binding site of ARs, dissecting out subtle structural variations across the subtypes and thereby may aid in future subtype selective drug development.