Chemistry Reference
In-Depth Information
Unfortunately, it was very difficult to produce larger amounts of the target protein and the
fragment-based screening project was stopped at this point.
It is also important to test the temperature stability of the target protein before the fragment
screening. Partly unfolded proteins will transiently expose hydrophobic patches to the
solvent (and the fragments) to a much higher degree than fully folded proteins and will
therefore pick up more nonspecific binders. Partly unfolded proteins may also form large,
but soluble, aggregates.
Target proteins with large solvent-exposed binding pockets.
Large amounts of bound water
in the targeted binding pocket will pose a special problem for fragment binding. The binding
affinity of fragments may not be sufficient to displace the bound water in a solution where
the water concentration is
55 M. This will result in a very low hit rate. Our experiences
include a phosphatase and a protease, both with large solvent-exposed active sites, and hit
rates below 1%. The low hit rate can be counteracted to some extent by using higher protein
and fragment concentrations and also by performing the NMR screen at low temperature
in order to increase the tumbling time of the protein.
∼
4.7 Future Directions
4.7.1 Multitarget Drugs
Due to the redundancy in biological networks, modulating the function of multiple pro-
tein targets simultaneously can be beneficial for treating complex diseases.
[
128, 129
]
Many
currently marketed drugs act via multiple targets, but the discovery of their multiple mech-
anism of action was usually serendipitous and retrospective. Compounds that are designed
to modulate functions of several targets are in general larger and more lipophilic than com-
pounds in the clinic or in the market.
[
130, 131
]
The reason for this is the current 'framework
combination strategy', where two selective ligands are combined into one dual ligand,
which most likely will contain features that are important for one of the target proteins
only. The result of this strategy is contradictory to the notion that larger compounds are
more selective than smaller compounds,
[
130, 132
]
i.e. multitarget compounds should typic-
ally be smaller than target-selective compounds. Fragment-based screening is the ideal
approach to find a core scaffold capable of binding to two or several target proteins. The
core scaffold could then be optimized to a compound with appropriately balanced affinities
between the target proteins. Naturally, the greatest likelihood of success would be when
the target proteins share a conserved binding site, e.g. kinases. The RAMPED-UP NMR
method,
[
133
]
where several differently labeled target proteins can be screened for binders
simultaneously, will probably prove to be useful in finding multitarget core scaffolds.
4.7.2 Membrane Protein Targets
Currently published examples of fragment-based screening by NMR are only applied to
soluble target proteins. There are, however, NMR techniques that have been used to detect
the binding of small molecules to integral membrane proteins reconstituted in liposomes.
STD has been used to detect the binding of peptides to integrin embedded in liposomes
[
100
]
and the bound conformation of a GPCR-bound peptide was determined by transferred
NOE experiments.
[
134
]
The problem is rather the difficulty in producing and successfully
