Chemistry Reference
In-Depth Information
Ideally, an entirely in silico approach to screening a large compound library
would significantly improve efficiency and reduce costs [ 140 , 141 ]. However,
several assessments of virtual screens have concluded that, without prior in-depth
analysis of the protein's ligand binding site, only a marginal improvement in
finding successful leads is observed relative to standard HTS [ 32 ]. NMR can
complement a virtual screen by providing rapid experimental validation of lead
compounds. NMR allows for a ligand-binding event to be directly observed instead
of relying on false-positive prone activity assays. Also, NMR provides detailed
structural information about the ligand binding site and the orientation of the bound
ligand. An NMR ligand affinity screen can be used to validate upwards of thousands
of predicted hits from a virtual screen [ 142 ]. Thus, combining NMR with virtual
screens may provide a more efficient approach to lead identification and drug
discovery.
4.1
Identification of New Therapeutic Targets
The functional assignment of unannotated proteins is essential to the drug discovery
process. Greater than 40% of protein sequences encoded in eukaryotic genomes
consist of proteins of unknown function and represent an important opportunity to
identify new therapeutic targets [ 143 ]. Assigning a function to an uncharacterized
protein is an arduous and time-consuming task. The process often requires detailed
biochemical studies that may include analyzing cell phenotypes through knockout
libraries, monitoring of gene expression levels, or utilizing pull-down assays
[ 144 - 147 ].
Since the interactions of proteins with other biomolecules or small molecules is
the basis of a functional definition or classification, identifying the functional
ligand, the functional epitope or ligand binding site, and the 3D structure of the
protein-ligand complex are invaluable for a functional annotation. A functional
epitope or ligand binding site is evolutionarily conserved relative to the rest of the
protein structure in order for the protein to maintain its biological function. There-
fore, proteins that share similar binding site structures are expected to be functional
homologs and bind a similar set of ligands [ 28 , 29 ]. Correspondingly, numerous
in silico approaches attempt to infer a function for an uncharacterized protein by
predicting ligand binding sites using geometry-based, information-based, and
energy-based algorithms [ 148 - 150 ]. Unfortunately, unambiguously identifying
the ligand binding site on a protein can be challenging without experimental
evidence, especially for proteins with no known function.
Functional Annotation Screening Technology using NMR (FAST-NMR) [ 28 , 29 ]
is one approach that combines HTS by NMR with molecular docking and bio-
informatics analysis in order to assign a function to a protein (Fig. 7 ). In this
process, a compound library that contains approximately 430 biologically relevant
compounds [ 151 ] is screened by NMR using a multistep approach [ 152 ]. First,
a ligand-based screen using 1D NMR 1 H line-broadening experiments identifies
Search WWH ::




Custom Search