Biomedical Engineering Reference
In-Depth Information
genome, in which case the target sequence could be
diluted over a millionfold by unwanted genomic
DNA. We need to find some way of rapidly sifting
through, or
screening
, large numbers of unwanted
sequences to identify our particular target.
There are two major strategies for isolating
sequences from complex sources such as genomic
DNA. The first, a cell-based cloning strategy, is to
divide the source DNA into manageable fragments
and
clone everything
. Such a collection of clones,
representative of the entire starting population, is
known as a
library
. We must then
screen the library
to
identify our clone of interest, using a procedure that
discriminates between the desired clone and all the
others. A number of such procedures are discussed
later in the chapter. The second strategy is to select-
ively amplify the target sequence directly from the
source DNA using the polymerase chain reaction
(PCR), and then clone this individual fragment.
Each strategy has its advantages and disadvantages.
Note that, in the library approach, screening is
carried out after the entire source DNA population
has been cloned indiscriminately. Conversely, in the
PCR approach, the screening step is built into the
first stage of the procedure, when the fragments
are generated, so that only selected fragments are
actually cloned. In this chapter we consider prin-
ciples for the construction and screening of genomic
and complementary DNA (cDNA) libraries, and
compare the library-based route of gene isolation to
equivalent PCR-based techniques.
Cloning genomic DNA
Genomic DNA libraries
Producing representative genomic
libraries in
λ
cloning vectors
Following the example above, let us suppose that we
wish to clone a single-copy gene from the human
genome. How might this be achieved? We could
simply digest total human DNA with a restriction
endonuclease, such as
Eco
RI, insert the fragments
into a suitable phage-
vector and then attempt to
isolate the desired clone. How many recombinants
would we have to screen in order to isolate the right
one? Assuming
Eco
RI gives, on average, fragments
of about 4 kb, and given that the size of the human
haploid genome is 2.8
λ
10
6
×
kb, we can see that
10
5
independent recombinants must be
prepared and screened in order to have a reasonable
chance of including the desired sequence. In other
words we have to obtain a very large number of
recombinants, which together contain a complete
collection of all of the DNA sequences in the entire
human genome, a human
genomic library
. The sizes
of some other genomes are listed in Table 6.1.
There are two problems with the above approach.
First, the gene may be cut internally one or more
times by
Eco
RI so that it is not obtained as a single
fragment. This is likely if the gene is large. Also, it
may be desirable to obtain extensive regions flanking
over 7
×
Table 6.1
Genome sizes of selected
organisms.
Genome size (kb)
Organism
(haploid where appropriate)
Escherichia coli
4.0
×
10
3
Yeast (
Saccharomyces cerevisiae
)
1.35
×
10
4
Arabidopsis thaliana
(higher plant)
1.25
×
10
5
Tobacco
1.6
×
10
6
Wheat
5.9
×
10
6
Zea mays
1.5
×
10
7
Drosophila melanogaster
1.8
×
10
5
Mouse
2.3
×
10
6
Human
2.8
×
10
6
Xenopus laevis
3.0
×
10
6
