Biology Reference
In-Depth Information
Fig. 8.
The principle of fragment-based drug design (DD): chemical building
blocks are docked separately into an active site. New putative ligands are constructed
from these maps of favorable poses by linking several building blocks together or
eventually to a known lead compound according to geometric constraints.
We are using a fragment-based approach to design small organic
IDO ligands. This approach is based on the work of Bemis and
Murcko,
71,72
who analyzed commercially available drugs and came up
with a limited set of a few tens of molecular frameworks and side chains
that are able to describe the chemical space of many drugs. In an
in silico
drug design approach, this concept can be used by separately
docking these fragments into the protein active site (Fig. 8). For each
fragment, a map with the 50 most favorable poses in the IDO active site
is produced using EADock.
73
Geometric constraints, which are derived
from molecular bond lengths, bond angles, and dihedral angles, are then
used to determine whether several fragments can be linked together to
create a new virtual compound. If so, the putative ligand is subjected to
the docking procedure to check if it adapts the intended binding mode
and makes favorable interactions. This approach can be adopted for both
virtual lead design and optimization.
A strong hint for the validity of this approach is given by the redis-
covery of the IDO ligand PIM from the X-ray structure
70
(Fig. 9). When
the most favorable pose of the imidazole and the second most favorable
pose of the phenyl rings are linked, the structure of the PIM-bound
X-ray structure is reproduced with a root mean square deviation (RMSD)
lower than 1 Å.
So far, we have identified 21 promising candidate ligands that we have
tested experimentally. Thirteen out of these 21 molecules have shown some
IDO inhibitory activity in an
in vitro
enzymatic assay. This high success
