Chemistry Reference
In-Depth Information
Ligand efficiency provides a means, as discussed by Hadjuk,
[
26
]
of extrapolating from
a screening hit to determine if a potency objective can be achieved through additions to
the molecule while maintaining a desired physicochemical property profile. One of the
advantages of using fragments is that they leave more scope for improvement based on
a typical medicinal chemistry approach (Figure 1.2), where studies
[
34
]
have shown that
lead development increases both the size and lipophilicity of the original hits. The fact
that existing data suggest fragments exhibit higher ligand efficiencies may also be a con-
sequence of traditional development strategies where suboptimal hits were taken as starting
points and/or suboptimal modifications were made in the development process. A clearer
picture may well develop as current practices of using ligand efficiency in both the selec-
tion and development of compounds feeds into our knowledge base. It is also the case that
the different features around a receptor binding site lead to an uneven distribution in the
maximum binding affinity across the site. The same can be said for screening hits where
different components of a ligand will contribute differently to the total binding affinity.
Without additional SAR information, it can be very difficult to establish, even with know-
ledge of the binding pose, the contribution of each component to the total binding affinity,
thereby necessitating the deconstruction of that hit and subsequent testing of the individual
components, i.e. fragment-based screening.
Figure 1.2
Fragments provide greater scope for development into drug-like molecules as
compared with HTS molecules that exhibit suboptimal binding to the target protein.
The dynamic, complex nature of receptor binding sites also highlights another very
important benefit of the fragment-based approach, that of the probability of identifying
an optimal match to the receptor. Hann
et al
.
[
22
]
calculated the probability of a binding
event with varying complexity of ligand-receptor interactions and noted that the chance
of observing a useful interaction falls rapidly with increasing complexity. The process of
binding was reduced to a simple functional model of molecular recognition between the
features of the receptor and a ligand. All the standard pharmacophoric features, such as
