DNA Cloning
Michael Andrew Quail, The Wellcome Trust Sanger Institute, Cambridge, UK
Deoxyribonucleic acid (DNA) cloning is the art of creating recombinant DNA
molecules that can be introduced into living cells, replicated and stably inherited, such
that multiple ‘clonal’ copies of that DNA are produced. Techniques for cloning and
the various vectors that can be used are described.
Introduction
Deoxyribonucleic acid (DNA) cloning is a cornerstone
of molecular biology. Its power is such that without it
a large proportion of modern biological science would
simply not be possible.
A clone can be defined as an identical copy. DNA
cloning involves joining, or ‘ligating’, DNA with a
‘vector’ that enables the resulting construct to be
introduced into a cell, replicated and passed on to
daughter cells as that cell divides (Figure 1). The result
is a population of cells all containing clones of the
original DNA. Thereafter that DNA is immortalized,
since cells can be grown to order and DNA extracted
for study or further manipulation whenever it is
required.
DNA cloning has been made possible by the
discovery of two types of protein, those that break or
modify DNA such that they have suitable termini for
ligation and those that are capable of ligating those
molecules of DNA.
How to Clone
DNA exists as a double helix of antiparallel polymer
strands with the nucleotide units in each joined by 50
–
30
phosphodiester bonds. The ability to selectively
break and join these bonds is the basis of DNA
cloning. For the detailed experimental protocols
described in this article the reader is referred to
Sambrook and Russell (2001), an excellent threevolume
laboratory manual of molecular biology.
Ligation
DNA strands can be joined through the action of the
enzyme DNA ligase, the normal biological role of
which is to join the series of Okazaki fragments that
are generated on the lagging strand during replication.
Typically the ligase from the Escherichia coli
bacteriophage T4, known as ‘T4 DNA ligase’, is used
in cloning experiments as it is both easily purified and
highly active. Ligase can be purchased commercially
and is normally supplied with a 106 concentrate of
reaction buffer, containing Mg2þ
and adenosine 50
triphosphate (ATP) that are both required for the
activity of this enzyme. Ligase utilizes the Mg2þ
cation
to contact the free 30
hydroxyl group at the end of a
DNA strand and uses the energy stored in the
phosphodiester bonds of ATP to covalently attach
this position to the 50
phosphate at the terminus of
another DNA strand, so joining the two strands.
Ligating double-stranded DNA thus consumes two
molecules of ATP and requires the two DNA
molecules to have compatible termini. However, only
one of the two 50
termini needs to have a phosphate
group. This is useful because dephosphorylation of the
vector DNA prevents its self-ligation. DNA can be
ligated with either blunt ends or, more efficiently, as
compatible staggered (‘sticky’), cohesive ends
(Figure 2), the conditions used being slightly different
in each case.
For sticky-end ligation, insert DNA is incubated in
16 ligase buffer in the presence of approximately
equimolar amounts (typically 10–50 ng) of vector that
has compatible termini, with 0.1–1 units of T4 DNA
ligase in a total volume of 10–50 ml for 2–8 h at 16
C,
or overnight at 4
C. DNA molecules possessing sticky
ends transiently base pair while in solution so bringing
their ligatable termini into close proximity. It is for this
reason that sticky-end ligation is more efficient.
Blunt-end ligation can be driven by increasing the
concentration of termini such that ligation is favoured.
Typically vector is mixed with a threefold molar excess
of insert DNA in 16 ligase buffer, with 1 unit of T4
DNA ligase, in a total volume of 4–10 ml and
incubated overnight at 16
C.
Preparation of DNA for ligation
Prior to ligation, compatible insert and vector DNA
must be prepared. Both need compatible termini and
must be free of contaminating chemicals such as
phenol, ethylenediaminetetraacetic acid (EDTA) or
ethanol that may inhibit ligation. Vector DNA is
typically prepared by digestion with the relevant
Advanced article
Article contents
 Introduction
 How to Clone
 Vectors
 Introduction of Deoxyribonucleic Acid into the Cell
 Epilog
doi: 10.1038/npg.els.0005344
DNA Cloning
ENCYCLOPEDIA OF LIFE SCIENCES & 2005, John Wiley & Sons, Ltd. www.els.net 1
restriction enzyme, then dephosphorylated by
incubation with alkaline phosphatase to remove the
terminal 50
phosphate groups and so prevent vector
self-ligation. Insert DNA is typically prepared using
one of the methods described below, then if it is part of
a mixture of DNA fragments, it is purified by
separating that mixture on a preparative (normally
low-melting-point) agarose gel, cutting out the slice of
agarose containing the fragment one wishes to clone
and extracting the DNA from the agarose matrix.
XR
XR
MCS
+
Recombinant DNA
+
Cells
(a)
+
(d)
(e) (f)
(c) (b)
Figure 1 A schematic representation of a typical cloning experiment. (a) The vector is cut (;) within its multicloning site (MCS).
(b) The target DNA is cut (;) so as to produce termini compatible with the vector. (c) The vector and insert are ligated to produce
recombinant DNA. (d, e) Recombinant DNA is introduced into appropriate host cells. In this illustration the vector encodes resistance
to an antibiotic, X. (f ) If the cells are plated out onto medium containing X, only cells that have been transformed will grow and divide
to form colonies (groups of around a million cells or ‘clones’ that have arisen from the same original transformed cell).
(b)(a)
pCpCpCOH
GpGpGp
+ +3Ј
5Ј
5Ј
3Ј
5Ј
3Ј
3Ј
5Ј
pGpGpG
HOCpCpCp CpTpTpApAp
pGOH pApApTpTpC
HOGp
GpGpGpCpCpCp
pCpCpCpGpGpG
CpTpTpApApGp
pGpApApTpTpC
Figure 2 Schematic diagram depicting (a) blunt end ligation and (b) ligation of cohesive (‘sticky’) ends. (a) Two double-stranded
molecules with blunt ends are ligated to produce a single molecule. (b) 50
protruding sticky ends, which were created by digestion with
the enzyme EcoRI, are ligated to produce a single double-stranded DNA molecule with the EcoRI recognition site being recreated.
DNA Cloning
2
Extraction of DNA from agarose can be performed by
one of a number of methods including melting
followed by phenol extraction, agarase digestion of
the agarose backbone or absorption and subsequent
elution of the DNA from a silica matrix.
Generation of sticky ends
DNA molecules with sticky ends are usually generated
through the use of type II restriction endonucleases.
These enzymes, whose biological function is to digest
or ‘restrict’ foreign DNA that enters the cell, recognize
specific sequences, ‘recognition sites’ and catalyze the
double-stranded cleavage of DNA within that site to
generate either molecules with protruding 50
, protruding
30
or blunt termini (Table 1).
Some restriction enzymes generate protruding
termini that are compatible, for example BamHI,
BclI, BglII, DpnII, MboI and Sau3AI all cleave to
give GATC 50
overhangs, and so DNA cleaved by any
of these enzymes will ligate to that cleaved by any of
the others. This is of particular use when one wishes to
generate genomic libraries. A DNA library is a ligation
that contains a mixture of DNA inserts. For wholegenome
studies, libraries are often made so as to
contain inserts representing all sequences in that
genome. The patterns of cleavage of a restriction
enzyme with a six-base (and longer) recognition
sequence are often nonrandom across a whole genome.
Therefore a common technique is to digest the DNA
with an enzyme with a four-base recognition site such
as Sau3AI (GATC). Normally such an enzyme would
cut every 44
(that is, 256), bases, but if the digestion
time is kept short or the amount of enzyme limited,
then a partial digestion will occur and so by
controlling the conditions a spectrum of fragments in
the desired size range can be produced. Since it is
important to cleave the vector DNA only once,
preferably within the desired cloning site region, and
Sau3AI would digest a typical vector molecule into
many pieces, the vector can be prepared using an
enzyme such as BamH I that cuts less frequently but
leaves compatible sticky ends.
Generation of blunt ends
DNA with blunt ends can be prepared using
restriction enzymes that leave a blunt end, through
cleavage with DNaseI or after end-repair of DNA
molecules that have staggered termini. Although less
efficient, this method has more versatility as there is
no absolute requirement for the presence of a
particular recognition site. DNA can be broken
randomly by mechanical breakage or digested with
any enzyme, then repaired. For ends with protruding
50
termini, repair is commonly achieved by
incubation with nucleotides and the Klenow fragment
of E. coli DNA polymerase I. Protruding 30
ends are removed by taking advantage of the high
30
–50
exonuclease activity of T4 DNA polymerase
which digests the single-stranded bases in the overhang.
If the nature of the ends is unknown, as
would be the case for random breakage, then ends
are normally repaired by digesting protruding singlestranded
termini by incubation with mung bean or
Bal3I nuclease or by incubating with a mixture of
both T4 and Klenow polymerase. (See DNA:
Mechanical Breakage.)
Table 1 A selection of restriction endonucleases commonly used in DNA cloning
Enzymea
Source Recognition siteb
Ends produced
BamHI Bacillus amyloliquefaciens H 50
G^GATCC 30
50
overhang
EcoRI Escherichia coli 50
G^AATTC 30
50
overhang
HindIII Haemophilus influenzae 50
A^AGCTT 30
50
overhang
NcoI Nocardia corallina 50
C^CATGG 30
50
overhang
NdeI Neisseria denitrificans 50
CA^TATG 30
50
overhang
NotI Nocardia otitidis-caviarum 50
GC^GGCCGC 30
50
overhang
PstI Providencia stuartii 50
CTGCA^G 30
30
overhang
Sau3AI Staphylococcus aureus 3A 50
^GATC 30
50
overhang
SmaI Serratia marcescens 50
CCC^GGG 30
Blunt
a
Enzymes commonly get their names from abbreviations of the organism from which they were first isolated. Roman
numerals after the name depict the order in which enzymes were discovered from an organism, for example, HindIII
was the third enzyme to be isolated from Haemophilus influenzae.
b
The recognition sequence for one strand in the 50
–30
direction is shown. Typically restriction sites have twofold
dyad symmetry, so the same site is seen when reading in the 50
–30
direction on the other strand. Enzymes cleave
the DNA strand within each recognition site to give a 30
hydroxyl group and a 50
phosphate group, at the
position denoted by ^.
DNA Cloning
3
Conversion of blunt to sticky ends
There are several methods by which blunt ends can be
given sticky tails:
 Through the action of the enzyme terminal transferase,
tracts of a single nucleotide base can be added
to the 30
hydroxyl group at each terminus. If different
but complementary bases are added to both
vector and insert, then the two will be compatible
and can be ligated.
 Single base overhangs can be added to the 30
position
of blunt termini by the action of Taq DNA
polymerase. Taq DNA polymerase, commonly used
in the polymerase chain reaction (PCR), frequently
adds an extra adenosine at the 30
position of the
amplified DNA. Thus PCR products can readily be
cloned by ligating to a thymine tailed vector, socalled
TA cloning. By incubating DNA with blunt
termini in the presence of Taq DNA polymerase and
a single deoxynucleotide triphosphate at 72
C, DNA
with single base 30
overhangs can easily be prepared.
 Short double-stranded adaptor or linker DNAs
containing restriction enzyme recognition sequences
can be ligated to blunt termini. Adaptors have one
blunt end and one sticky end; thus when ligated they
convert the blunt ends to sticky ones. Linkers have
two blunt ends but contain internal restriction sites.
After linker addition, the resulting molecules can be
digested with the appropriate restriction enzyme
and sticky ends created.
Generation of deoxyribonucleic acid for cloning by
polymerase chain reaction
PCR is an extremely powerful technique as it allows
for minute quantities of DNA (or messenger ribonucleic
acid (mRNA)) to be amplified up to cloneable
quantities. PCR products synthesized using enzymes
such as Taq DNA polymerase (that lack 30
–50
exonuclease activity) have a single adenosine 30
overhang
and those synthesized with ‘proofreading’ polymerases
which possess 30
–50
exonuclease activity are
generated with blunt ends. Furthermore, since the
primers used for PCR are incorporated into the
products (and mismatches between primer and target
sequence are often tolerated, especially at the 50
end of
the primer) PCR can be used to introduce restriction
sites for cloning at the termini of the amplified
fragments. Of particular use in this respect are the
recognition sites for the enzymes NdeI and NcoI, which
contain the initiation codon sequence ATG. Here PCR
can be used to introduce an NdeI or NcoI site at the
initiation codon of a particular gene so that it can be
cloned in-frame into an expression vector.
Vectors
Vectors are DNA molecules that are capable of
accepting selected fragments of DNA and replicating
the resulting hybrid when it is introduced into living
cells. Typically vectors are modified versions of
naturally occurring independent replicons or viruses.
Due to its extremely well-characterized genetic systems,
the majority of vectors propagate in laboratory
strains of the enteric bacterium E. coli, although
‘shuttle vectors’ that can also replicate in other
organisms and vectors that exclusively replicate in
other organisms are widely available. Vectors can be
obtained commercially from molecular biology
reagent suppliers, from authors of published papers
or from collections such as ATCC (see Web Links).
Features of vectors
Required features
Cloning site
The vector must possess a recognition site for at least
one restriction endonuclease so that the vector can
accept DNA. Most common vectors have been
constructed so as to have a multicloning site (MCS).
Here a short piece of DNA containing the recognition
sequences for several enzymes has been added into the
region of the vector where foreign DNA is to be
inserted.
Origin of replication
To be able to replicate and propagate the DNA
construct, vectors must contain origin of replication
(ori) sequences that are recognized by the host cell
DNA replication systems. It is the ori that largely
determines vector copy number.
Selectable marker
A vector must contain sequences that confer a positive
advantage on cells that contain it. This is because
maintenance and propagation of cloned DNA places a
burden upon that cell, and without an advantage cells
that did not contain the clone would grow preferentially
resulting eventually in loss of the clone from the
population. Furthermore, when DNA is introduced
into a cell population only a small fraction of those
cells take up the DNA and one needs a mechanism
to select for those cells that contain the vector
construct. Therefore vectors are designed with selectable
markers. These typically fall into two main classes:
 genes that encode resistance to a certain antibiotic
(e.g. ampicillin, G418, kanamycin, chloramphenicol,
tetracycline, etc.) and
DNA Cloning
4
 auxotrophic markers which complement a host cell
mutation in an essential biosynthetic pathway. For
example, the yeast URA3 gene is commonly used in
yeast vectors in combination with a host cell that has
a mutation resulting in a requirement for uracil.
When cells are grown in a medium that does not
contain uracil only those containing the vector with
the URA3 marker can grow.
Optional features
There are a great many vectors now available to the
molecular biologist. Each has slightly different features
allowing it to be more suitable for one application
than another, and so a careful look at the
features that a vector contains is essential. Common
features are noted below.
Sequences that enable recombinant selection
Many vectors have their cloning sites within the coding
region of a gene that confers a selectable phenotype,
such that when DNA is inserted the function of that
gene is disrupted. Since the vector can self-ligate to
give nonrecombinant molecules, this is a valuable
means of selecting for cells that contain recombinant
clones. The most widely available selection is
a-complementation, where the cloning site is located
within coding sequences for the a fragment of the
enzyme b-galactosidase (encoded by the lacZ gene).
The a fragment is able to complement a host cell
mutation (lacZDM15) in this gene. Thereby on
selective agar plates containing the colorimetric
substrate for this enzyme (X-gal) and the inducer
isopropyl-b-D-thiogalactopyranoside (IPTG), nonrecombinants
have b-galactosidase activity, break down
X-gal into a blue compound and appear blue whereas
recombinants where this gene has been disrupted
remain white/colorless. Another recombinant selection
method utilizes ccdB (or the cell-death gene) the
product of which strongly inhibits bacterial DNA
gyrase and thereby DNA synthesis, resulting in cell
death. If the cloning site is positioned within this gene,
only cells containing vector with an insert will grow.
Likewise the sacB gene (present in P1 and some
bacterial artificial chromosome (BAC) vectors)
encodes the enzyme levansucrase, which can convert
the common sugar sucrose into a cytotoxic compound.
When cells are grown on medium containing sucrose,
only recombinants where DNA has been inserted into
the cloning site to disrupt this gene will grow.
Primer binding sites
Sequences that bind well-characterized (often commercially
available) primers are often found flanking
the multicloning site. These are useful for sequencing,
or amplification by PCR, of the inserted DNA.
Reporter genes
Reporter genes encode a product with an easily
detectable activity such as lacZ (b-galactosidase),
chloramphenicol acetyltransferase (CAT) or green
fluorescent protein (GFP). The promoter activity of
a given piece of DNA can be quantified by cloning that
sequence upstream of a promoterless reporter gene.
Likewise if a gene being studied is cloned in-frame with
a reporter gene, the activity of the reporter gene will
give a measure of the expression level of the
corresponding gene product. In the case of GFP,
information regarding the subcellular localization of
the product can also be obtained.
Strong promoter sequences
1. Riboprobe generation. Strong RNA polymerase
promoters (e.g. SP6, T3 or T7) often flank cloning
sites; this enables single-stranded probes from the
termini of the inserted DNA to be generated. This
can be carried out in vitro when the construct is
incubated with the corresponding polymerase and
ribonucleotides.
2. Overexpression. For overexpression of a gene, for
example, as a prelude to protein purification, that gene
can be cloned into a vector that has a known strong
promoter sequence, such as T7 polymerase promoter,
Ptac, ParaBAD or PCMV, just upstream of the cloning site.
These promoters are often inducible such that expression
can be induced at the right time. In some instances
the overexpressed protein can accumulate to over 50%
of the total cell protein.
In-frame genes encoding easily purifiable proteins
To facilitate purification of an overexpressed protein,
the corresponding gene can be cloned in-frame with
sequences that code for an easily purifiable protein or
peptide such as glutathione-S-transferase (GST), maltose
binding protein (MalE) or polyhistidine peptide
(His-TAG). This results in the overexpression of a
fusion protein comprising the protein under study and
the easy to purify protein, which can be at either the Nor
C-terminus depending on vector and cloning site
used. Subsequently this fusion can be readily purified
by virtue of the characteristics of the particular fusion
partner. His-TAG, which is probably the most
commonly used fusion, readily binds nickel ions, and
so His-TAGged fusions can be purified by affinity
chromatography on immobilized nickel, washed, then
subsequently eluted with imidazole. The fusion partner
often has no effect on the activity of the protein and so
is not normally removed, though some vectors
incorporate protease cleavage sites such as that for
factor Xa, at the fusion junction.
DNA Cloning
5
loxP sites
loxP is the recognition sequence for the bacteriophage
P1 recombinase enzyme, Cre, which catalyzes recombination
between sequences that are flanked by such
sites. By cloning DNA into vectors that possess loxP
sites that DNA can, for example, be recombined into
another vector containing loxP sites or into the host
chromosome, when in the presence of either the Cre
recombinase in vitro or a plasmid carrying the cre gene
in vivo.
cos sites
cos sites are the recognition sequences for phage
packaging systems. The DNA present between two
such sites, provided it is within the packaging limits for
that system, will be packaged into phage particles.
Types of vectors
An outline of the features of each vector type is given
in Table 2.
Plasmids
Plasmids are the most widely used vectors. This is
probably because of the ease with which plasmid DNA
can be isolated from the host cell (see Sambrook and
Russell, 2001) and the variety of plasmids that are
available. They are double-stranded independent
replicons that range in size from approximately 2 to
10 kb and which are capable of accepting DNA inserts
up to about 12 kb, though up to 5 kb is perhaps more
common. Figure 3 shows the structure of pUC18, a
typical plasmid that replicates in E. coli.
M13 bacteriophage
M13 is a filamentous bacteriophage, with a 6.4 kb
single-stranded genome, that infects E. coli. An
infected cell produces 100–200 double-stranded (or
replicative form (RF)) copies of the phage genome.
These encode genes that coordinate packaging of one
of the phage genome DNA strands into a proteinaceous
viral coat and their extrusion from the cell where
they are then free to infect further cells. A series of
Table 2 A summary of the properties of the various classes of vectors used in DNA cloning
Vector
Cloning
capacity Use
Common
examples
Method of
introduction
into cell
Type of
growth on
agar plate
Copy
number
Original
reference
Plasmids Up to 12 kb Multiple
(e.g. mutagenesis,
overexpression,
sequencing,
screening,
manipulation)
pUC,
pBR322,
pGEM
Transformation or
electroporation
into bacteria.
Various methods
for eukaryotes
(see text)
Colonies 1–700 Cohen et al.
(1973) PNAS 70:
3240–3244
M13 Normally
up to 6 kb
Probe generation,
sequencing, in vitro
mutagenesis
M13mp18 Transfection Plaques 100–200 Messing et al.
(1977) PNAS 74:
3642–3646
Bacteriophage
lambda (l)
9–23 kb Genomic and
cDNA libaries,
expression
screening
lZAP,
lFIX
Packaging of
construct followed
by infection
Plaques N/A Murray and Murray
(1974) Nature 251:
476–481
Cosmids Up to 47 kb Genomic libraries,
genome mapping
Supercos,
lawrist,
tropist
Packaging of
construct followed
by infection
Colonies Up to 50 Collins and Hohn
(1978) PNAS 75:
4242–4246
Fosmids Up to 43 kb Genomic libraries,
genome mapping
pFOS1 Packaging of
construct followed
by infection
Colonies 1–2 Kim et al. (1992)
NAR 20: 1083–1085
P1 bacteriophage 30–100 kb Genomic libraries,
genome mapping
pAd10sacBII Packaging of
construct followed
by infection
Colonies 1–2 Sternberg (1990)
PNAS 87: 103–107
BACs Up to 300 kb Genomic libraries,
genome mapping
pBACe3.6,
pBeloBACII
Electroporation Colonies 1–2 Shizuya et al. (1992)
PNAS 89: 2629–2633
PACs Up to 300 kb Genomic libraries,
genome mapping
pCYPAC2 Electroporation Colonies 1–2 Ioannou et al. (1994)
Nature Genetics 6:
84–89
YACs 20–2000 kb Genomic libraries,
genome mapping
pYAC4 Electroporation,
spheroplast absorption,
or lithium-mediated
transformation
Colonies 1 Burke et al. (1987)
Science 236: 806–812
DNA Cloning
6
M13-based cloning vectors have been developed
(Messing, 1983), and found capable of accepting
inserts several times larger than the M13 genome.
However, DNA in M13 can be unstable and so insert
sizes are normally kept below that of the vector and
this system is not used for the longer-term maintenance
of DNA. M13 has no selectable marker;
instead infected cells are identified by their ability to
form zones of clearing, or ‘plaques’. If bacterial cells
are mixed with a soft agar and allowed to grow on top
of a normal agar plate, the cells in the soft agar will
grow and give the agar layer an opaque appearance.
Infected cells grow a lot more slowly than noninfected
ones and so can be visualized as a plaque in the
bacterial lawn.
Since the extruded phage contain copies of one
particular DNA strand, this system has proved very
useful for techniques that utilize a single-stranded
template such as sequencing, site-directed mutagenesis
and producing radiolabeled probe DNA for
hybridization.
Bacteriophage lambda
Lambda (l) is a naturally occurring phage virus of
E. coli. It has a 48 kb linear double-stranded genome,
with 12-base cohesive ends (cos sites) at each 50
end,
inside an icosahedral proteinaceous head to which is
attached a long noncontractile tail. Infection occurs as
a result of the tail binding to the outer membrane
maltose receptor and injection of the phage genome
through the tail and into the cell. Once in the cell the
DNA can follow a lytic or lysogenic pathway
depending on environmental or intracellular conditions
and the genotype of the individual phage. In the
lysogenic pathway, the phage DNA is integrated into
the E. coli genome providing the geneticist with a tool
to integrate DNA into genomic DNA. In the lytic
pathway, the DNA is replicated under a rolling circle
mechanism to produce a long concatemer of around
100 lambda genome units. Lambda genes encode head
and tail proteins and an enzyme that recognizes the cos
sites and cleaves the concatemer into individual
genomes. Subsequently each 48 kb unit is packaged
into lambda heads, the tails attach and the infected cell
lyses to release approximately 100 infective phage. In
order to be packaged the cos sites need to be around 38–
52 kb apart and so different lambda vectors have been
produced where nonessential regions have been deleted
to make space for DNA to beinserted.A typical lambda
vector lZAPII is shown in Figure 4. There are two main
advantages to using lambda vectors:
 the packaging and infection process is very efficient
(perhaps 100 times more efficient than transformation/transfection);
and
 since the cells die during lysis, lambda can maintain
some inserts that carry cytotoxic or unstable/poorly
tolerated sequences.
Cosmids/fosmids
Cosmids were created by engineering cos sites into
plasmid vectors to create hybrid vectors with the high-
efficiencypackaging/infectionadvantagesofthelambda
system and the ease of isolation advantage of plasmids.
In recent years, cosmids have been superseded as the
vector of choice for creating genomic DNA libraries
by artificial chromosome vectors that can accommodate
larger inserts. Since they are maintained at quite a
high copy number in the cell, cosmid clones can
become unstable and are often prone to deletion. To
overcome this, fosmids, which are based on the singlecopy
bacterial F factor, have been developed.
P1 bacteriophage vectors
P1 is another naturally occurring E. coli phage. It has a
linear 110 kb genome that upon infection is circularized
through the action of the P1 encoded Cre
recombinase which catalyzes recombination at loxP
recognition sites. Vectors based upon P1 that can
ori
ampR
lacZ
lacI
pUC18
(2.69 kb)
HindIII
SphI
PstI
SalI/AccI/HindII
Xba I
BamHI
SmaI/Xma I
KpnI
SacI
EcoRI
M13F
pUCR
Figure 3 Diagram of the plasmid cloning vector, pUC18.
This 2.69 kb double-stranded, closed-circular DNA molecule has
an origin of replication (ori) allowing multicopy propagation in
E. coli, a b-lactamase gene (ampR
) giving resistance to the antibiotic
ampicillin and a lacI gene, the product of which induces the
expression of the lacZ gene, which in turn encodes b-galactosidase.
The multicloning site (dark bar, shown expanded to the right) is close
to the 50
end of the lacZ gene, such that insertion at any of the points
within the multicloning site (MCS) should disrupt b-galactosidase
expression, thereby allowing blue/white recombinant selection on
media containing the colorimetric substrate X-gal and inducer
isopropyl-b-D-thiogalactopyranoside (IPTG). The MCS contains
closely nested restriction sites for a number of enzymes, in the order
shown. It is flanked by sequences to which the M13 forward and
pUC reverse primers bind, facilitating polymerase chain reaction
amplification or sequencing of the inserted DNA.
DNA Cloning
7
maintain inserts of 70–100 kb have been developed
(Sternberg, 1990). Advantages of P1 vectors include:
 stability of inserts, because the vector is single copy;
 DNA is present as a closed circle within the cell and
so can be isolated using a standard alkaline lysis
miniprep; and
 constructs are introduced into the cell by the highly
efficient process of packaging and infection.
Bacterial artificial chromosomes
BACs are vectors based on the E. coli F factor, which is
normally present at one to two copies per cell. They
can accept inserts of up to 300 kb and can be directly
introduced into deoR strains of E. coli by electroporation,
thus avoiding the use of packaging extracts. Their
stability and the ease by which clone DNA can be
miniprepped for analysis, or end sequencing, make
BACs and P1 artificial chromosomes (PACs) (see
below) the current vectors of choice for most wholegenome
libraries. For a detailed experimental description
of BAC library construction see Osoegawa et al.
(1998). A diagram of a typical BAC vector, pBACe3.6,
is shown in Figure 5.
P1 artificial chromosomes
PACs are a hybrid of P1 bacteriophage and BAC
vectors. They are low copy number, have good
recombinant selection and can be introduced into the
cell by electroporation.
Yeast artificial chromosomes
Yeast artificial chromosome (YAC) vectors are shuttle
vectors containing yeast telomeres, centromere, an
origin of replication and selectable markers allowing
replication in yeast and E. coli. They can accommodate
inserts of up to 2 megabases (Mb). The resulting
constructs are transformed into yeast by spheroplast
absorption and therein behave like an extra chromosome.
Saccharomyces cerevisiae is quite tolerant of
extreme base composition and repeats, and so YACs
are often useful for cloning sequences that are
unclonable in E. coli.
oror
lacZ
λZAPII
(39.9kb)
EcoRI
XhoI
SpeI
XbaI
NotI
SacI
T3 T7
A–J cIts857
IT
coscos
colE1
F1 ori
MCSMCS
pBluescript SK-
(2.9kb)
MCS
TI
ampR
Figure 4 Diagrammatic illustration of the bacteriophage lambda cloning vector lZAPII (Stratagene). This 39.9 kb linear DNA
molecule can accept inserts up to 10 kb. It has a multicloning site (MCS) (expanded above) flanked by T7 and T3 polymerase promoters
to which T7 and T3 primers bind. The MCS is within the lacZ gene allowing blue/white selection of inserts. The region of the vector
containing the insert can be excised to the plasmid pBluescript (SK-) as shown, by infection of cells harboring lZAPII with a helper
phage. lZAPII also contains terminal cos sites, lambda genes A–J and a temperature-sensitive mutation within the cI repressor that
represses lysis.
DNA Cloning
8
Human artificial chromosomes
With the goal of creating stable transgenic mammalian
cell lines, researchers have been developing mammalian
artificial chromosomes (MACs) and human
artificial chromosomes (HACs). This was achieved in
1997 when Harrington et al. (1997) reported the
construction and maintenance, in a human cell line,
of human artificial chromosomes containing 0.3–2 Mb
of inserted a-satellite DNA.
Introduction of Deoxyribonucleic
Acid into the Cell
Although some bacteria can spontaneously take up
DNA, this is not a widespread occurrence and so a
number of methods have been developed for the
directed transfer of cloned DNA into the cell.
Transfer into bacteria
There are four common methods for DNA transfer
into bacteria (see Sambrook et al., 1989; Hanahan
et al., 1991).
Transformation
Transformation is the process of plasmid uptake by
bacteria. This process is typically chemically induced
with the most common method involving treatment
with calcium chloride. Here log phase cells are washed
and resuspended in 0.1 M calcium chloride. These cells
are incubated on ice with the DNA whereupon a
calcium ion–DNA complex is formed on the outside of
the cell. If this mixture is subsequently heat-shocked
by incubating at 42
C for 60–90 s, the cell then takes
up this complex. Typically 0.01–0.1% of the DNA is
introduced generating up to 108
transformants per
microgram of pUC18 DNA.
Electroporation
Application of a brief high-voltage electric shock to a
suspension of cells transiently induces the presence of
pores in the cell membrane through which DNA can
enter the cell. Electroporation can be highly efficient
giving more than 1010
transformants per microgram of
pUC18 DNA.
Transfection
Transfection is defined as the transformation by viral
or phage DNA, and can be achieved by either of the
previous methods.
Infection
Infection of bacteria can reach efficiencies close to
100%. Typically cells are first grown under conditions
that favor induction of the phage receptor; for example
E. coli to be infected with phage lambda are cultured in
media containing magnesium ions and maltose.
Prepared cells can simply be incubated with phage
particles, typically at 37
C for 20–30 min, during
which time the phage are absorbed.
Transfer into eukaryotic cells
The introduction of DNA into eukaryotic cells is
technically more demanding and often less efficient
than transfer into bacteria. Nevertheless, a number of
methods have been developed (Ausubel et al., 1999;
James and Grosveld, 1987) involving calcium
phosphate-mediated transfection, DEAE dextran
transfection, electroporation, microinjection, infection
(viral vectors), electroporesis, protoplast/spheroplast
fusion and lipofection. The method used is dependent
on the type of genetic material being handled, the
ori
R
mc
R
mc
sacBII
pBACe3.6
(11.49kb)
SP6
pUC
fr
agm
ent
cm
R
T7
NotII
NotI
BamHI
BamHI
SacII
SacII
EcoRI
EcoRI
SacI
SacI
MluI
MluI
loxP
loxP
Figure 5 Diagrammatic illustration of the bacterial artificial
chromosome (BAC) cloning vector pBACe3.6. This 11.49 kb doublestranded,
closed-circular DNA molecule has an origin of replication
(ori) allowing single-copy propagation in E. coli, a gene encoding
resistance to the antibiotic chloramphenicol (cmR
) and loxP
recombination sites that flank the region into which DNA is inserted.
Dual multicloning sites (MCS) flank a pUC stuffer fragment. The
parent vector possesses this region to allow multicopy growth in
E. coli allowing easy preparation of the vector. Restriction of the
vector with any of the MCS restriction enzymes excises this
fragment. The MCS is flanked by T7 and T3 polymerase
promoters, to which T7 and T3 primers bind, and is located
between the sacBII gene and its promoter so that inserted DNA
disrupts expression of its cytotoxic product, ensuring that only
recombinants grow on media containing sucrose.
DNA Cloning
9
reason that DNA is being introduced and more
critically the type of cell being transformed.
Epilog
Cloning is at the heart of molecular biology. It can be
immensely rewarding, or just as easily extremely frustrating.
The key to success is usually to be found in
gaining a full understanding of the system to be used
and in careful experimental planning. As a starting
point the beginner is directed to the references and sug-
gestionsforfurtherreadingcitedbelow.Happycloning!
See also
Mammalian Artificial Chromosomes (MACs)
Recombinant DNA
Yeast Artifical Chromosome (YAC) Clones
References
Ausubel FM, Brent R, Kingston RE, et al. (eds.) (1999) Introduction
of DNA into mammalian cells. Current Protocols in Molecular
Biology, chap. 9. New York, NY: John Wiley.
Hanahan D, Jessee J and Bloom FR (1991) Plasmid transformation
of Escherichia coli and other bacteria. Methods in Enzymology
204: 63–113.
Harrington JJ, Van Bokkelen G, Mays RW, Gustashaw K and
Willard HF (1997) Formation of de novo centromeres and
construction of first-generation human artificial minichromosomes.
Nature Genetics 15: 345–355.
James RFL and Grosveld FG (1987) DNA-mediated gene transfer
into mammalian cells. In: Walker JM and Gaastraxs W (eds.)
Techniques in Molecular Biology, pp. 187–202. London, UK:
Croom Helm.
Messing J (1983) New M13 vectors for cloning. Methods in
Enzymology 101: 20–78.
Osoegawa K, Yeong Woon P, Zhao B, et al. (1998) An improved
approach for construction of bacterial artificial chromosome
libraries. Genomics 52: 1–8.
Sambrook J and Russell DW (2001) Molecular Cloning: A Laboratory
Manual, 3rd edn. Cold Spring Harbor, NY: Cold Spring Harbor
Laboratory Press.
Sternberg N (1990) Bacteriophage P1 cloning system for the
isolation, amplification, and recovery of DNA fragments as
large as 100 kilobase pairs. Proceedings of the National Academy of
Sciences of the United States of America 87: 103–107.
Further Reading
Birren B, Green ED, Klapholz S, Myers RM and Roskams J (eds.)
(1997) Genome Analysis: A Laboratory Manual. Volume 1, Analysing
DNA. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory
Press.
Brown TA (2000) Essential Molecular Biology: A Practical Approach.
Oxford, UK: Oxford University Press.
Markie D (ed.) (1996) YAC protocols. Methods in Molecular Biology,
vol. 54. Totowa, NJ: Humana Press Inc.
Web Links
ATCC (American Type Culture Collection). A global bioresource
center
http://www.atcc.org
DNA Cloning
10