utilized to explore various aspects of proteins, such as protein
structure plasticity, domain folding, identification of key residues
for protein folding and function, and residue fluctuation
[13–19]. Particularly, the network analysis of protein structures
and their wiring diagram can be extended further to identify key
residues responsible for allostery.
Several computational methods present the opportunity to
explore allosteric events at scales inaccessible for experimental studies.
Most of the in silico methods are focused on the identification of
allosteric binding sites and their signaling propagation pathway
[20]. In our previous work, we studied the link between allosteric
signal transduction and functional dynamics in S-adenosyl-homocys-
teine (SAH) hydrolase using multifaceted computational approaches
[21]. In this chapter, we describe a simple, but powerful approach to
glean the mechanism of receptor regulation and identify the allosteric
hotspots for signal transductions and their pathways. This method has
been implemented in our previous study using A2Aadenosine receptor
(A2AAR) and other class A GPCRs, such asβ1 adrenergic receptor,
β2 adrenergic receptor, chemokine CXCR4 receptor, dopamine D3
receptor, histamine H1 receptor, and rhodopsin as model systems
[22]. Based on the structural ensemble derived from MD simulations,
we revealed the allosteric hotspots and signaling pathway of A2AAR by
utilizing betweenness centrality (CB)-based network analysis, glycine
scanning analysis, and cross-correlation analysis. The intramolecular
network analysis was applied to identify the key residues responsible
for GPCR allostery. The residues with high betweenness centralities
constitute physically connected sparse network linking the ligand
binding site to the G-protein binding site. These residues also
included several highly conserved microswitch residues which are
deemed important for GPCR function. Compared to other conven-
tional methods that utilize the information ofsequence coevolution or
variants of normal mode analysis (NMA), ourCB-based network
analysis approach is proven to be quite powerful in elucidating the
allosteric hotspots, and the results are in strong correlation with the
biochemical studies.2 Methods
In the following section, we describe the primary steps involved in
the mapping of the allosteric signal flow by adopting A2AAR as a
model system (Fig.1).2.1 Molecular
Dynamics Simulations
of Membrane Proteins
MD simulation has become a popular tool in the investigation of
the dynamics of proteins, membrane proteins, and more complex
systems, providing insights into the biological processes at atomis-
tic level which are inaccessible via experiments. MD simulations
generate successive configurations of the system by integratingMolecular Dynamics Approach for Investigation of GPCR Allostery 457