Chemistry Reference
In-Depth Information
described above is the 1,3-dipolar cycloaddition of organic azides and alkynes to form
1,2,3-triazoles. Sharpless and co-workers demonstrated the use of this completely bio-
orthogonal reaction (termed 'click chemistry') as a useful tool for lead generation by a
fragment linking approach.
[
41
]
In the database of 3.8 million commercially available compounds that Evotec main-
tains, the number of compounds that qualify as a 'fragment' by rule of three criteria is
only 65 000 (see above). Further incorporation of the criterion of synthetic expandabil-
ity (to have multiple synthetic handles) in fragment library design inevitably reduces the
maximum size of the fragment library considerably. This may be acceptable if the screen-
ing method chosen is NMR or X-ray crystallography due to the relatively low throughput
of the methodology. However, if the library is to be screened by high-concentration bio-
chemical assays, potentially all the available 65 000 fragments could be screened within
a reasonable cost and time irrespective of synthetic tractability. Although there is clearly
merit in the synthetic handle approach, ultimately it is how a fragment specifically interacts
with its biological target that will determine whether the potential for synthetic elabora-
tion can be exploited for a specific fragment. A key issue with this approach is that for
relatively low molecular weight species it is likely that any masked chemical functional-
ity may be intimately involved in fragment binding to the target protein and so may not
provide a suitable vector for fragment decoration. Furthermore, scientists at Novartis also
pointed out that the concept of a potential chemical linker not involved in the protein bind-
ing of the fragment, but available for chemical modification, conflicts with the aim of a
maximum ligand efficiency requiring that all parts of the fragment contribute to protein
binding.
[
39
]
3.3 Evotec's NMR Fragment Library
From the previous section, it is apparent that there are many considerations to be taken
into account when assembling a fragment library and a variety of different approaches and
philosophies. Evotec has two separate fragment libraries which were constructed independ-
ently and on the basis of different criteria. One has been built up for fragment screening by
high-concentration biochemical assays whereas the other was built up for fragment screen-
ing by NMR. Each library comprises approximately 20 000 compounds but there are fewer
than 1000 compounds in common between the two libraries. In this section we describe
the process that was followed for the construction of the library for NMR-based fragment
screening (Figure 3.2) and in the following section we describe the assembly of a fragment
library for use in high-concentration biochemical screening (Figure 3.3).
Due to its high sensitivity for detecting weak binders and its unique binding site res-
olution, protein-observed NMR screening is ideal for progressing to lead molecules from
fragment hits, as demonstrated by the pioneering work of Fesik's group at Abbott Labor-
atories over a decade ago.
[
42
]
Being significantly lower in molecular weight than leads,
fragment hits occupy sub-pockets within the active or allosteric site of interest rather than
the entire pocket. As nicely exemplified by a 168 Da fragment bound to PDE4d,
[
43
]
the most
binding efficient fragment hits will preferentially sit in the most druggable sub-pocket(s)
of the target, mediate attractive target interactions
via
the majority of its atoms and possess
a low entropic barrier to binding. Following this concept of fragment-target interaction,
