Biomedical Engineering Reference
In-Depth Information
Notably, the central Merck compound was already known for several
other biological activities [165,166], but these details were left out of the
original Merck paper.
In 2002, Novo Nordisk and DGI Biotech reported on insulin mimetic
peptides identified from screening of phage display peptide libraries
(peptides expressed on the surface of viral phages) [167]. Two different
motifs for each of the alleged receptor-binding epitopes were identified
and subsequently linked, to provide fully-active insulin agonists (and
antagonists), although of quite long sequences [168]. Testing
in vivo
showed
a
correlation
between
length
of
peptide
sequence
and
bioavailability.
7.10
PUSHING THE LIMITS OF INSULIN
ENGINEERING
The currently marketed insulins are produced by recombinant methods,
in some cases supplemented by chemical transformations. Recombinant
methods are traditionally limited to the 20 genetically-encoded amino
acids. Recently, novel recombinant methods have been developed in
which stop codons are used as codes for the introduction of unnatural
amino acids in recombinant sequences [169]. However, these methods
have apparently not yet been applied to insulin or matured to industrial
scale.
Various chemical methods are known for the addition or exchange of
amino acids with unnatural residues in the terminals of native insulin [7],
but these methods can only be used in a very limited number of positions.
Full chemical synthesis of insulin has been described by a couple of
methods, but the chemistry is difficult and low-yielding [170]. The rela-
tively long sequences of insulin near the limits of modern peptide synth-
esis, but the worst problem is the required pairing of three disulfide
bridges.
Recent efforts to improve synthetic insulin disulfide pairing include
orthogonal protections and directed pairings of each disulfide bridge
[171,172], as well as synthetic use of a solubilizing extension accom-
panied by artificial C-peptide to direct the correct folding and dis-
ulfide formation [173], followed by enzymatic processing to two-
chain insulin (Figure 7.14) [174]. However, the yields of these meth-
ods
are
still
low
and
neither
seems
ready
for
industrial-scale
production.
Search WWH ::
Custom Search