262
LUIZ F. ONUCHIC AND GREGORY G. GERMINO
Vector and host selection
Recombinant DNA strategies require the ability to distinguish and select
specific fragments of interest. Finding a specific fragment buried in a large
population of similarly sized molecules and then obtaining enough of it to
allow its manipulation would be a daunting challenge if not for the availability of cloning techniques and libraries. Rodent-human somatic cell hybrids are one way of propagating very large human DNA fragments, but the
fraction of total DNA that is of human origin is relatively small, and the
human fragments are often too large tobe easily characterized or manipulated. Another drawback is that the human and rodent chromosomes cannot be separated except by flow-sorting or by preparing metaphase spreads.
Thus, a variety of vectors and hosts have been constructed and derived to
allow more efficient and manageable propagation ofDNA fragments: YACs
that are capable of propagating very large fragments spanning several hundred kilobases (kb) to megabases; and plasmids, lambda bacteriophage and
cosmids for smaller fragments of 1 -50 kb. Larger fragments contain more
information but are more difficult to characterize and separate from host
DNA. There are a number of factors that determine selection of a vector ( the
cloning vehicle) and host ( the cellline infected/transfected or transformed
with the recombinant molecule) for a given library, which are presented and
discussed below.
Desired size of cloned inserts
In chromosome "walking", larger inserts are desirable in order to reduce the
number of "walks" required. Each walk yields, on average, "new DNA"
equal to one half the insert size of the vector being used: e.g., lambda bacteriophage with 20 kb inserts yield 10 kb, cosmids with 40 kb yield 20 kb, P1s
with 100 kb yield 50 kb and YACs with 1000 kb yield 500 kb (Figure 1). It is
obvious why YACs and P1 clones have replaced cosmids and lambda bacteriophage in most physical mapping and chromosome walking projects.
Smaller insert sizes are preferable, however, for many other projects requiring genomic fragments. Examples include mapping intron-exon boundaries and transcriptional regulatory sequences or developing constructs
for gene knock-outs via homologaus recombination. Smaller inserts are
more easily characterized, isolated and subcloned. Lambda bacteriophage
and cosmids are the vectors most commonly used to generate such multipurpose libraries. They are highly efficient cloning vehicles and carry inserts
in the appropriate range of sizes.
LUIZ F. ONUCHIC AND GREGORY G. GERMINO
Vector and host selection
Recombinant DNA strategies require the ability to distinguish and select
specific fragments of interest. Finding a specific fragment buried in a large
population of similarly sized molecules and then obtaining enough of it to
allow its manipulation would be a daunting challenge if not for the availability of cloning techniques and libraries. Rodent-human somatic cell hybrids are one way of propagating very large human DNA fragments, but the
fraction of total DNA that is of human origin is relatively small, and the
human fragments are often too large tobe easily characterized or manipulated. Another drawback is that the human and rodent chromosomes cannot be separated except by flow-sorting or by preparing metaphase spreads.
Thus, a variety of vectors and hosts have been constructed and derived to
allow more efficient and manageable propagation ofDNA fragments: YACs
that are capable of propagating very large fragments spanning several hundred kilobases (kb) to megabases; and plasmids, lambda bacteriophage and
cosmids for smaller fragments of 1 -50 kb. Larger fragments contain more
information but are more difficult to characterize and separate from host
DNA. There are a number of factors that determine selection of a vector ( the
cloning vehicle) and host ( the cellline infected/transfected or transformed
with the recombinant molecule) for a given library, which are presented and
discussed below.
Desired size of cloned inserts
In chromosome "walking", larger inserts are desirable in order to reduce the
number of "walks" required. Each walk yields, on average, "new DNA"
equal to one half the insert size of the vector being used: e.g., lambda bacteriophage with 20 kb inserts yield 10 kb, cosmids with 40 kb yield 20 kb, P1s
with 100 kb yield 50 kb and YACs with 1000 kb yield 500 kb (Figure 1). It is
obvious why YACs and P1 clones have replaced cosmids and lambda bacteriophage in most physical mapping and chromosome walking projects.
Smaller insert sizes are preferable, however, for many other projects requiring genomic fragments. Examples include mapping intron-exon boundaries and transcriptional regulatory sequences or developing constructs
for gene knock-outs via homologaus recombination. Smaller inserts are
more easily characterized, isolated and subcloned. Lambda bacteriophage
and cosmids are the vectors most commonly used to generate such multipurpose libraries. They are highly efficient cloning vehicles and carry inserts
in the appropriate range of sizes.