184 Pingoud, Alves, and Geiger- Preparative Applications
For a variety of different purposes, in particular for the cloning of
DNA and sequencing using the Maxam-Gilbert method (182), it is
necessary to produce fragments in preparative amounts. This is done
by a scale-up of analytical digests and/or an extension of the incuba-
tion time. The derived restriction fragment is usually isolated from the
product mixture either by preparative electrophoresis or when larger
amounts are needed by HPLC (for detailed protocols see refs. 21,185).
An application for which restriction digests have to be carried out
on a preparative scale that is becoming increasingly more important is
the preparation of DNA for site-directed mutagenesis. The gapped-
duplex protocol (198) requires, in addition to the single-stranded cir-
cular template DNA, a complementary piece of DNA that covers most
of the template and leaves uncovered the region in which the mutation
is to be introduced. The gap is prepared by a suitable restriction digest
of the double-stranded mutagenesis vector, e.g., M 13RF, pEMBL, and
so on, isolation of the appropriate fragment, and annealing to the single-
stranded mutagenesis vector.
Preparative restriction digests can also be used to increase the molar
yield ofa mutagenesis experiment. Suppose the circular template strand
of a gapped-duplex carries a particular singular restriction site within
the selection marker gene (e.g., the 13-1actamase gene), whereas the
gap-forming complementary strand owing to a silent point mutation
does not contain this site. Annealing of the mismatch oligodeoxy-
nucleotide, polymerization, and ligation will yield a double-stranded
circular DNA that carries in its template strand the wild-type sequence
of the gene as well as the recognition sequence for the restriction
enzyme and, in its newly formed strand, the mutated sequence of the
gene, but no recognition sequence for the restriction enzyme. Trans-
formation into repair-deficient E. coli cells will yield a more or less
equal distribution of two plasmids in the clones, only one of which
carries the mutated gene. The yield of this one can be improved by
isolating plasmid DNA from the E. coli culture and cleavage of the
DNA by the restriction enzyme, which attacks only the wild-type
DNA. Retransformation of the DNA resulting from this reaction will
lead to preferential transformation of the supercoiled DNA, i.e., the
mutant DNA. This procedure leads to much increased yields of the