Chemistry Reference
In-Depth Information
few years, but most will also be in the inactive form. This lack of structures is
contributing to the shortage of safe and efficacious drugs that target GPCRs.
Availability of GPCR structures, experimental or predicted, can lay the foundation
for rational structure-based drug design. So, a fast but accurate computational
approach is needed that can generate structures for all important conformations of
a target receptor and any other receptors implicated for off-target therapeutic side-
effects and determine their ligand binding efficacies for developing highly selective
drug candidates with potentially minimal side-effects. We have developed one such
FF based method.
The structural topology of GPCRs consists of seven transmembrane (TM)
a
-
helices that span the membrane and are connected by both intracellular and
extracellular loops. To characterize this topology quantitatively with respect to a
common reference frame, the middle of the membrane is assumed to correspond to
the
z
0 plane or the hydrophobic plane that cuts the 7-helix bundle into two
halves. Each GPCR structure can then be characterized by the six orientation
parameters of the seven helices shown in Fig.
4
, which shows how the helix position
and tilt are defined. Helix position on the hydrophobic plane is then given by
x
and
y
. Value
h
corresponds to the hydrophobic center residue from the helix that will be
positioned on the hydrophobic plane. Two angles,
ΒΌ
y
and
f
, specify the tilt angles of
the helix and the angle
Z
corresponds to the helix rotation angle about its axis. The
two tilt angles (
) require a definition of the helical
axis which needs to account for the reality of bent helices as prolines are commonly
found in the TM helices. We use a helical axis that corresponds to the least moment
of inertia vector for the helix obtained by eigensolution of the moment of inertia
matrix for the helix using only heavy backbone atoms.
The structural analysis of available experimental structures shows large varia-
tions in helix tilts and rotations. Considering a
y
,
f
) and the rotation angle (
Z
10
sampling of the
y
tilt angle,
30
sampling of the
and
f
and
Z
angles, for each of the seven helices, leads to
5)
7
, ~10 trillion possible conformations, for each of which the amino acid
side chains must be optimized. To make such a huge sampling computationally
feasible, we developed the SuperBiHelix sampling (BiHelix sampling only sampled
the
(3
5
Z
rotation angle) procedure. As indicated by 12 double arrows in Fig.
5
(left)
a typical class A GPCR template has 12 important pair-wise interactions. For
each such pair of helices, we will sample all combinations of
over some
grid. During this sampling, the other five helices are ignored, as indicated in Fig.
5
y
,
f
,
Z
Fig. 4 Definitions of
orientation parameters of a
transmembrane helix
Search WWH ::
Custom Search