A2aR Oligomeric assemblies identified from MD simulations using in-vivo mimetic biomembranes

GPCR oligomerisation is known to play an important role in the receptor signalling. However, due to the technical challenges, the structural information of GPCR oligomerisation is still very limited, which hinders our understanding of GPCR signalling in a fuller picture. In this deposit, we provide the structural coordinates of various oligomeric assemblies of Adenosine A2a receptor that were sampled from unbiased MD simulations.

For more information regarding the MD simulation setup and definitions of the various calculated values, please check out our paper on BioRxiv: https://www.biorxiv.org/content/10.1101/2020.06.24.168260v1

** Simulation setup **
9 copies of A2aR were randomly inserted into an in-vivo mimetic biomembrane (of size of 45nm x 45nm) to build the initial configuration of the simulations. 10 such systems were set up for A2aR in the inactive state, 10 for the active state and 10 for the active in complex with the mini Gs state. These systems were represented by MARTINI 2 coarse-grained models and were simulated for 50 micro-seconds. The use of MARTINI coarse-grained force field would freeze the receptor conformation in the initial configuration, thus decoupled the oligomerization from such process as ligand-induced conformational change. The more efficient sampling of coarse-grained force field therefore allowed us to explore fully the protein-protein associations in the oligomerisation process. Protein-protein  associations were identified when any atoms from two protomers were getting closer than 0.75 nm. The oligomerisation process was monitored and the sampled various oligomeric assemblies were identified for calculation of oligomer residence time. 

** Coordinate file explained **
These pdb files contain the A2aR oligomer structures in atomistic models. The coarse-grained oligomeric structures were converted back to atomistic models using CHARMM 36 force field. The identified oligomeric structures from each oligomeric order were clustered. 10 structures were randomly taken from each cluster and stored as individual models in the pdb files with a naming format {Conf. State}_OS{Oligomeric Order}_cl{Cluster id}.pdb. The pdb files can be viewed by such visualization tools as PyMol, Chimera or JMol etc.

** Spreadsheet file explained **
The calculated properties, including residence time and geometry, of each identified oligomer were stored in the Excel shreadsheet (Oligomeric_Assembly_Distribution.xlsx). Each oligomeric order, i.e. oligomer order = 2,3,4,5, opens an individual spreadsheet page where the calculated data were grouped by oligomers' conformational states (i.e. Inactive, Active and Act + mini Gs) and then ranked by oligomers' residence time. The measurements for describing the oligomer geometry were shown in columns after "Cluster ID" and before "Count". For definitions of these measurements, please refer to our paper. Pictures of the oligomers viewed from the extracellular side and intracellular side were also provided in the the spreadsheet to assist visualisation.