Cloning strategies - Principles of Gene Manipulation - page 117

Box 6.6 A landmark publication. Subtraction cloning of the human
Duchenne muscular dystrophy ( DMD ) gene

While most subtractive cloning experiments involve
cDNAs, this publication reports one of the few
successful attempts to isolate a gene using a
subtracted genomic library. The study began with
the identification of a young boy, known as 'BB',
who suffered from four X-linked disorders, including
DMD. Cytogenetic analysis showed that the boy
had a chromosome deletion in the region Xp21,
which was known to be the DMD locus. Kunkel's
group then devised a subtraction-cloning procedure
to isolate the DNA sequences that were deleted
in BB. Genomic DNA was isolated from BB and
randomly sheared, generating fragments with
blunt ends and non-specific overhangs. DNA was
also isolated from an aneuploid cell line with four
(normal) X chromosomes. This DNA was digested
with the restriction enzyme Mbo I, generating sticky
ends suitable for cloning. The Mbo I fragments were
mixed with a large excess of the randomly sheared
DNA from BB, and the mixture was denatured and
then persuaded to reanneal extensively, using phenol
enhancement. The principle behind the strategy was

that, since the randomly fragmented DNA was
present in a vast excess, most of the DNA from the
cell line would be sequestered into hybrid DNA
molecules that would be unclonable. However,
those sequences present among the Mbo I fragments
but absent from BB's DNA due to the deletion would
only be able to reanneal to complementary strands
from the cell line. Such strands would have intact
Mbo I sticky ends and could therefore be ligated into
an appropriate cloning vector. Using this strategy,
Kunkel and colleagues generated a genomic library
that was highly enriched for fragments corresponding
to the deletion in BB. Subclones from the library
were tested by hybridization against normal DNA
and DNA from BB to confirm that they mapped
to the deletion. To confirm that the genuine DMD
gene had been isolated, the positive subclones
were then tested against DNA from many other
patients with DMD, revealing similar deletions
in 6.5% of cases.

From Kunkel (1986) Nature 322: 73-77.

primer used for first-strand cDNA synthesis. In the
differential-display PCR technique (Liang & Pardee
1992), the antisense primer is an oligo-dT primer
with a specific two-base extension, which thus
binds at the 3

Liang and Pardee (1992), the technique was used
to study differences between tumour cells and
normal cells, resulting in the identification of a
number of genes associated with the onset of cancer
(Liang et al . 1992). Further cancer-related gene
products have been discovered by other groups
using differential display (Sager et al . 1993, Okamato
& Beach 1994). The technique has also been used
successfully to identify developmentally regulated
genes (e.g. Adati et al . 1995) and genes that are
induced by hormone treatment (Nitsche et al . 1996).
An advantage of display techniques over sub-
tracted libraries is that changes can be detected in
related mRNAs representing the same gene family.
In subtractive-cloning procedures, such differences
are often overlooked because the excess of driver
DNA can eliminate such sequences (see review by
McClelland et al . 1995).

end of the mRNA. Conversely, in the
arbitrarily primed PCR method (Welsh et al . 1992),
the antisense primer is arbitrary and can in prin-
ciple anneal anywhere in the message. In each case,
an arbitrary sense primer is used, allowing the
amplification of partial cDNAs from pools of several
hundred mRNA molecules. Following electropho-
resis, differentially expressed cDNAs can be excised
from the gel and characterized further, usually to
confirm its differential expression.
Despite the fact that these display techniques
are problematical and appear to generate a large
number of false-positive results, there have been
remarkable successes. In the original report by

′