Sequencing and mutagenesis - Principles of Gene Manipulation - page 132

Eventually, with the accumulation of new
sequences, sequence conservation will become the
definitive criterion for gene identification and the
contribution of statistical methods will decrease.
Nevertheless, it is still likely that some genes will not
have identifiable homologues and statistical and ex-
perimental approaches will remain necessary for their
detection. Furthermore, even for genes that have
homologues, statistical methods of coding-potential
analysis will remain useful for localizing frame shifts
and choosing among the possible initiation codons.

for purification and assay must also be considered
before embarking on protein mutagenesis.

Cassette mutagenesis

In cassette mutagenesis, a synthetic DNA fragment
containing the desired mutant sequence is used to
replace the corresponding sequence in the wild-type
gene. This method was originally used to generate
improved variants of the enzyme subtilisin (Wells
et al. 1985). It is a simple method for which the
efficiency of mutagenesis is close to 100%. The
disadvantages are the requirement for unique
restriction sites flanking the region of interest and
the limitation on the realistic number of different
oligonucleotide replacements that can be synthes-
ized. The latter problem can be minimized by the use
of doped oligonucleotides (Fig. 7.8; Reidhaar-Olson
and Sauer, 1988), as described on pp. 107-109.

Changing genes: site-directed
mutagenesis

Three different methods of site-directed mutagenesis
have been devised: cassette mutagenesis, primer
extension and procedures based on the PCR. All
three are described below but the reader wishing
more detail should consult the review of Ling and
Robinson (1998).
In some cases, the goal of protein engineering is to
generate a molecule with an improvement in some
operating parameter, but it is not known what
amino acid changes to make. In this situation,
a random mutagenesis strategy provides a route to
the desired protein. However, methods based on
gene manipulation differ from traditional mutagen-
esis in that the mutations are restricted to the gene
of interest or a small portion of it. Genetic engineer-
ing also provides a number of simple methods of
generating chimeric proteins where each domain
is derived from a different protein.
It should not be forgotten that constructing the
mutant DNA is only part of the task. The vector for
expression, the expression system and strategies

Primer extension:
the single-primer method

The simplest method of site-directed mutagenesis is
the single-primer method (Gillam et al. 1980, Zoller
& Smith 1983). The method involves priming in
vitro DNA synthesis with a chemically synthesized
oligonucleotide (7-20 nucleotides long) that carries
a base mismatch with the complementary sequence.
As shown in Fig. 7.9, the method requires that the
DNA to be mutated is available in single-stranded
form, and cloning the gene in M13-based vectors
makes this easy. However, DNA cloned in a plasmid
and obtained in duplex form can also be converted to
a partially single-stranded molecule that is suitable
(Dalbadie-McFarland et al. 1982).

Fig. 7.8 Mutagenesis by means of
doped oligonucleotides. During
synthesis of the upper strand of the
oligonucleotide, a mixture of all four
nucleotides is used at the positions
indicated by the letter N. When the
lower strand is synthesized, inosine
(I) is inserted at the positions shown.
The double-stranded oligonucleotide
is inserted into the relevant position
of the vector.

Arg Glu Ile Glu Met Glu Ala Val Ser Met

** * ** *

N G N G

II

T CG GA A AC T N GA

CC

AT N GA A GC GT T AG A T

C C

CT T TA II I T TA I CT T CG CA A TC

A
Xho I

T GTAC

Sph I