Biomedical Engineering Reference
In-Depth Information
20 kb, which are suitable for insertion into a
re-
placement vector. Packaging
in vitro
(p. 58) ensures
that an appropriately large number of independent
recombinants can be recovered, which will give an
almost completely representative library.
λ
endonuclease that cuts frequently, only short frag-
ments of insert DNA are left attached to these
promoters. This allows RNA probes to be generated
corresponding to the
ends
of any genomic insert.
These are ideal for probing the library to identify
overlapping clones as part of a chromosome walk
(p. 107) and have the great advantage that they can
be made conveniently, directly from the vector, with-
out recourse to subcloning. Vector maps of
Development of
replacement vectors for
genomic library construction
λ
λ
DASH
In the Maniatis strategy, the use of two different
restriction endonucleases with completely unrelated
recognition sites,
Hae
III and
Alu
I, assists in obtaining
fragmentation that is nearly random. These enzymes
both produce blunt ends, and the cloning strategy
requires linkers (see Fig. 6.2). Therefore, in the early
days of vector development, a large number of
different vectors became available with alternative
restriction sites and genetic markers suitable for
varied cloning strategies. A good example of this
diversity is the Charon series, which included both
insertion- and replacement-type vectors (Blatttner
et al
. 1977, Williams & Blattner 1979).
A convenient simplification can be achieved by
using a
single
restriction endonuclease that cuts fre-
quently, such as
Sau
3AI. This will create a partial
digest that is slightly less random than that achieved
with a pair of enzymes. However, it has the great
advantage that the
Sau
3AI fragments can be readily
inserted into
and
λ
FIX are shown in Fig. 6.4.
λ
FIX is similar to
λ
DASH, except that it incorporates additional
Xho
I
sites flanking the stuffer fragment. Digestion of the
vector with
Xho
I followed by partial filling of the
sticky ends prevents vector religation. However,
partially filled
Sau
3AI sticky ends are compatible
with the partially filled
Xho
I ends, although not with
each other. This strategy prevents the ligation of
vector arms without genomic DNA, and also prevents
the insertion of multiple fragments.
Genomic libraries in high-capacity vectors
In place of phage-
derivatives, a number of higher-
capacity cloning vectors such as cosmids, bacterial
artificial chromosomes (BACs), P1-derived artificial
chromosomes (PACs) and yeast artificial chromo-
somes (YACs) are available for the construction of
genomic libraries. The advantage of such vectors
is that the average insert size is much larger than
for
λ
EMBL3
(Frischauf
et al
. 1983), which have been digested
with
Bam
HI (Fig. 6.3). This is because
Sau
3AI and
Bam
HI create the same cohesive ends (see p. 32). Due
to the convenience and efficiency of this strategy,
the
λ
replacement vectors, such as
λ
replacement vectors. Thus, the number of
recombinants that need to be screened to identify a
particular gene of interest is correspondingly lower,
large genes are more likely to be contained within
a single clone and fewer steps are needed for a chro-
mosome walk (p. 107). Generally, strategies similar
to the Maniatis method discussed above are used
to construct such libraries, except that the partial
restriction digest conditions are optimized for larger
fragment sizes, and size fractionation must be pre-
formed by specialized electrophoresis methods that
can separate fragments over 30 kb in length. Pulsed-
field gel electrophoresis (PFGE) and field-inversion
gel electrophoresis (FIGE) have been devised for this
purpose (p. 10). High-molecular-weight donor DNA
fragments can also be prepared using restriction
enzymes that cut very rarely.
Cosmids may be favoured over
λ
EMBL series of vectors have been very widely
used for genomic library construction (p. 58). Note
that
λ
EMBL vectors also carry the
red
and
gam
genes
on the stuffer fragment and a
chi
site on one of the
vector arms, allowing convenient positive selec-
tion on the basis of the Spi phenotype (see p. 58).
Most
λ
vectors currently used for genomic library
construction are positively selected on this basis,
including
λ
λ
2001 (Karn
et al
. 1984),
λ
DASH and
λ
FIX (Sorge 1988).
λ
DASH and
λ
FIX (and recently
improved versions,
FIXII) are par-
ticularly versatile because the multiple cloning sites
flanking the stuffer fragment contain opposed
promoters for the T3 and T7 RNA polymerases. If
the recombinant vector is digested with a restriction
λ
DASHII and
λ
vectors because
they accept inserts of up to 45 kb. However, since
λ