Comparative Study of Receptor-, Receptor State-, and Membrane- Dependent Cholesterol Binding Sites in A2A and A1 Adenosine Receptors Using Coarse-Grained Molecular Dynamics Simulations

We used coarse-grained molecular dynamics (CG MD) simulations to study protein-cholesterol interactions for different activation states of the A2A adenosine receptor (A2AR) and the A1 adenosine receptor (A1R) and predict new cholesterol binding sites indicating amino acid residues with a high residence time in three biologically relevant membranes. Compared to 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC)-cholesterol and POPC-phosphatidylinositol-bisphos-phate (PIP2)-cholesterol, the plasma mimetic membrane best described the cholesterol binding sites previously detected for the inactive state of A2AR and revealed the binding sites with long-lasting amino acid residues. We observed that using the plasma mimetic membrane and plotting residues with cholesterol residence time >= 2 mu s, our CG MD simulations captured most obviously the cholesterol- protein interactions. For the inactive A2AR, we identified one more binding site in which cholesterol is bound to residues with a long residence time compared to the previously detected, for the active A1R, three binding sites, and for the inactive A1R, two binding sites. We calculated that for the active states, cholesterol binds to residues with a much longer residence time compared to the inactive state for both A2AR and A1R. The stability of the identified binding sites to A1R or A2AR with CG MD simulations was additionally investigated with potential of mean force calculations using umbrella sampling. We observed that the binding sites with residues to which cholesterol has a long residence time in A2AR have shallow binding free energy minima compared to the related binding sites in A1R, suggesting a stronger binding for cholesterol to A1R. The differences in binding sites in which cholesterol is stabilized and interacts with residues with a long residence time between active and inactive states of A1R and A2AR can be important for differences in functional activity and orthosteric agonist or antagonist affinity and can be used for the design of allosteric modulators, which can bind through lipid pathways. We observed a stronger binding for cholesterol to A1R (i.e., generally higher association rates) compared to A2AR, which remains to be demonstrated. For the active states, cholesterol binds to residues with much longer residence times compared to the inactive state for both A2AR and A1R. Taken together, binding sites of active A1R may be considered as promising allosteric targets.