Bioorganic Chemistry
23
development of very efFcient coupling re-
actions. If the coupling efFciency is 95%, a
high yield under normal circumstances, in
the process of making a 20-mer, 0.95
20
or
36%oftheresu
ltantpeptidewil
lbeofthe
sequence speciFed. The remainder will be
a mixture of similar but incorrect peptides.
Modern solid-phase peptide synthesizers
are able to attain efFciencies on the order
of 99%, which allows good purities to be
attained.
5
Combinatorial Chemistry
A central area of bioorganic chemistry is
the identiFcation of small-molecule lig-
ands for macromolecules. Such molecules
can be used as biochemical probes as well
as potential lead therapeutics. There are
several paths that can lead to high-afFnity
ligands for macromolecules. Molecules
that
mimic
the
structures
of
known
substrates can act as lead compounds.
Three-dimensional structures of macro-
molecules determined by X-ray or NMR
can be used as a guide to ‘‘rationally’’
design molecules that bind to identiFed
binding sites. If efFcient screening meth-
ods exist, large collections of compounds
can be individually tested until a ‘‘hit’’
is found, which can then be optimized
through the synthesis of derivatives. An
additional method, that of combinatorial
chemistry, has proven itself to be useful
for identifying ligands for a range of sys-
tems. It has been made possible by the
revolution of solid-phase organic synthe-
sis, which allows the rapid synthesis of
collections of molecules containing arbi-
trarily large numbers of species.
An example of this is shown in ±ig. 18. It
depicts the synthesis of a simple tripeptide
library using 10 standard L amino acids.
The peptide is constructed on an insoluble
resin in a stepwise fashion. If each position
during the synthesis is allowed to vary and
be any one of 10 amino acids (we will
discuss how this is done shortly), there
will be a mixture of 10
×
10
×
10
=
1000
possible tripeptides present at the end of
the synthesis. The challenge then is to
determine which, if any, of the members
of this library bind to the target of interest.
There are many approaches to this process,
known as
deconvolution
of a library. Two
of the most widely used approaches have
been positional scanning and iterative
deconvolution. In both these approaches,
mixtures of compounds (sublibraries of
the larger library) are tested and the results
are used to identify an individual hit
or hits.
In the process of positional scanning,
sublibraries are constructed by using a
speciFc amino acid at one given position
and
mixtures
of
amino
acids
at
the
remaining positions (±ig. 18). This is done
multiple times, moving the speciFc amino
acid along the length of the peptide.
Therefore, at the end of the process for
the tripeptide library described above, one
will have three collections, each of which
contain 10 sublibraries. As shown in the
Fgure, each of the 10 sublibraries contains
a mixture of products. This is a mixture of
peptides that have a random amino acid at
two positions and a speciFc amino acid at
the remaining position. When all of these
sublibraries are assayed, if there is a single
‘‘winner’’ to be found in the total library
(for the purpose of argument Gly-Pro-Asp),
then the subelements that contain that
sequence in high concentration will show
activity. The sequence of the ‘‘hit’’ can
therefore be read directly from the activity
results. It can then be resynthesized as
an individual species and conFrmed as
an active molecule. In practice, there can
previous page 697 Encyclopedia of Molecular Cell Biology and Molecular Medicine read online next page 699 Encyclopedia of Molecular Cell Biology and Molecular Medicine read online Home Toggle text on/off