Chemistry Reference
In-Depth Information
the density of an inhibitor is of great interest. For irreversible inhibitors which
deactivate the enzyme by forming a chemical bond to the active site, the influence is
expected to be also strong since a chemical reaction takes place. The other type of
inhibitors, the so-called noncovalent reversible inhibitors, is attached to the active
site by weaker intermolecular interactions which may be too weak to cause a
noticeable influence on the ED. In the present text, we review recent theoretical
works which investigate whether the ED can be used as an indicator for the
inhibition potency of a substance. As discussed above, theoretical data should be
sufficiently accurate to reveal correct trends. Using theoretical and experimental
data simultaneously is unlikely to be of any advantage as the uncertainties on both
approaches are expected to be in the order of or larger than the effects of interest.
Additionally, experimental EDs are only available for a single enzyme-inhibitor
complex [
65
-
67
]. In the following, we will focus on the question whether the
electron densities of inhibitors in crystals of the pure compound are comparable
to those of the same compound in the active sites of enzymes [
83
]. The studies
required theoretical methods that simulate the influence of different surroundings
on EDs on an equal basis, i.e., the inhibitors within the enzyme or the crystal
environment. The method of choice to describe such complicated assemblies are
combined quantum mechanics/molecular mechanics (QM/MM) approaches [
87
]
since pure quantum chemical approaches are too expensive and MM approaches do
not provide an ED. QM/MM methods [
88
-
92
] divide the total system (enzyme,
solvent, and inhibitor) into the active center and the rest. The active site is described
by QM approaches, while the influence of the surrounding protein environment and
the solvent is captured at the MM level. The QM and MM regions interact with each
other through electrostatic and dispersive terms. In the work reviewed in the
following, the electrostatic QM/MM interactions are represented by an electronic
embedding scheme [
93
] incorporating the MM charges into the one-electron QM
Hamiltonian and thereby allowing the ED of the QM system to adapt to the field
exerted by the environment. Dangling bonds at the QM/MM boundary are capped
with hydrogen link atoms [
94
-
98
] in the framework of the charge shift method. For
the present applications, the inhibitor is described by quantum chemical methods
(QM part), while the environment is represented by a force field obtained from
molecular mechanical simulations (MM part). These potentials possess atomic
resolution, that is, they also contain finer details arising from the molecular nature
of the surrounding, for example, due to hydrogen bonds or salt bridges. The
geometrical arrangements of the inhibitors in crystals of the pure compound and/
or in enzyme-inhibitor complexes were derived from available crystal structures. In
addition to crystal and enzyme environments, the influence of polar solvents was
also studied using the conductor-like screening model (COSMO) [
99
,
100
]. It
should be noted that COSMO is well suited to include the overall polarization by
a solvent but has problems to describe stronger interactions, e.g., hydrogen bond-
ing. Completely unpolarized EDs were determined by single molecule computa-
tions. This situation corresponds to the inhibitor molecule in vacuum and will be
designated as such in the following. The ED of a molecule is a strong function of its
geometrical structure. Thus, already different conformers may show strong
