Biomedical Engineering Reference
In-Depth Information
ing mutagenesis with phage display, up to 10
9
differ-
ent mutants can
theoretically
be screened in a single
experiment. The constraint with this approach is
that phage display is really best suited to selecting
proteins with altered binding characteristics, rather
than the other properties one might wish to engin-
eer. Despite this, a number of groups (Atwell &
Wells 1999, Demartis
et al.
1999, Olsen
et al.
2000)
were able to select variant proteases with novel
substrate specificities.
mutagenesis, error-prone PCR and spiked synthetic
oligonucleotides. The key element in the process is
the ability to screen large numbers of mutants. An
example is the isolation of a more thermostable
subtilisin. Up to 1000 mutant clones are gridded
out on replica plates and, once they are grown, one
plate is incubated at an elevated temperature long
enough to inactivate the wild-type enzyme. When
an assay for hydrolytic activity is subsequently
performed, only mutants with stability greater than
that of the wild type will display measurable activity.
Once stable mutants have been identified, the replic-
ate colony can be grown to identify the mutation.
Once stabilizing single amino acid changes have
been identified, building a highly stable subtilisin
can be accomplished by combining individual amino
acids into the same molecule (Pantoliano
et al.
1988,
1989, Zhao & Arnold 1999).
One of the disadvantages of the screening method
described above is that it is labour-intensive and the
maximum feasible number of mutants that can be
examined in a single screen is 10
4
-10
5
. By combin-
Gene families as aids to
protein engineering
An alternative approach to directed evolution is
'DNA shuffling', which is also known as 'molecular
breeding' (Minshull & Stemmer 1999, Ness
et al.
2000). This method can only be adopted if the target
protein belongs to a known protein family. If it does,
the genes for the different family members are
isolated and artificial hybrids created (Fig. 14.14).
As an example of this approach, Ness
et al.
(1999)
Ancestral species
Evolution
Species 1
Species 2
Species 3
Species 4
DNA shuffling
in vitro
Hybrid genes
Fig. 14.14
Schematic representation of
gene shuffling.
