Role of membrane composition and the extracellular loops for ligand binding to the neuronal human adenosine receptor type 2A

G protein-coupled receptors (GPCRs) are integral membrane proteins encoded by more than 800 genes in the human genome. They play essential role in cellular physiology and are targeted by approximately 40% of all modern medicinal drugs. Biophysical studies of these receptors are particularly challenging because they are transmembrane proteins heavily dependent on the membrane environment for proper function. In addition, the extracellular loops of GPCRs play a key role in early stage of ligand recognition. However, a quantitative description of the process of ligand binding to GPCR in native membrane environment is still largely missing. Molecular Dynamics (MD) simulations are being instrumental in assessing the effect of native membrane environment on conformational properties of GPCRs and accurately quantifying the structural properties and the energetics of ligand/GPCR binding process. Here, building on the existing state-of-the-art molecular dynamics, I address the important issue of the role of neuronal membrane composition and loop structural dynamics in ligand binding and recognition for a neuronal GPCR. This is the human adenosine receptor type 2A (hA2AR). I first addressed the importance of membrane lipid composition in receptor conformation and ligand binding stability by performing unbiased μs-scale molecular dynamics simulations of the receptor in complex with one of its naturally-occurred antagonists, caffeine. My results suggest cholesterol molecule in neuronal-like membrane allosterically stabilizes a specific caffeine-binding pose against others visited during the simulation. This indicates that cholesterol significantly present in neuronal cell membranes but neglected in X-ray determination of membrane proteins, affects the population of the ligand binding poses. I then studied the binding process of one potent antagonist of hA2AR, ZM241385, in neuronal cell-like membrane. Free energy calculations and structural analysis indicate that the first extracellular loop ECL1 and the second extracellular loop ECL2, along with a lipid molecule, form a previously overlooked and significantly populated “vestibular binding site,” which stabilizes the ligand prior entering the orthosteric binding site. Bioinformatics analysis indicates that this site is evolutionarily correlated with the orthosteric binding site. Evolutionary correlated sites suggest the existence of intramolecular allosteric communication between these two binding regions in synergistically modulating ligand potency and efficacy. To summarize, this thesis suggests that including a realistic description of neuronal membranes may be very important for computer-aided design of ligands targeting hA2AR and possibly other neuronal GPCRs. Our findings further uncover a vestibular site stabilized by membrane lipids, which might lead to novel GPCR-extracellular surface-targeting drug design