NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Brown TA. Genomes. 2nd edition. Oxford: Wiley-Liss; 2002.

Cover of Genomes

Genomes. 2nd edition.

Show details

Chapter 4Studying DNA

Learning outcomes

When you have read Chapter 4, you should be able to:

  • Give outline descriptions of the events involved in DNA cloning and the polymerase chain reaction (PCR), and state the applications and limitations of these techniques
  • Describe the activities and main applications of the different types of enzyme used in recombinant DNA research
  • Identify the important features of DNA polymerases and distinguish between the various DNA polymerases used in genomics research
  • Describe, with examples, the way that restriction endonucleases cut DNA and explain how the results of a restriction digest are examined
  • Distinguish between blunt- and sticky-end ligation and explain how the efficiency of blunt-end ligation can be increased
  • Give details of the key features of plasmid cloning vectors and describe how these vectors are used in cloning experiments, using pBR322 and pUC8 as examples
  • Describe how bacteriophage λ vectors are used to clone DNA
  • Give examples of vectors used to clone long pieces of DNA, and evaluate the strengths and weaknesses of each type
  • Summarize how DNA is cloned in yeast, animals and plants
  • Describe how a PCR is performed, paying particular attention to the importance of the primers and the temperatures used during the thermal cycling

The toolkit of techniques used by molecular biologists to study DNA molecules was assembled during the 1970s and 1980s. Before then, the only way in which individual genes could be studied was by classical genetics, using the procedures that we will examine in Chapter 5. Classical genetics is a powerful approach to gene analysis and many of the fundamental discoveries in molecular biology were made in this way. The operon theory proposed by Jacob and Monod in 1961 (Section 9.3.1), which describes how the expression of some bacterial genes is regulated, was perhaps the most heroic achievement of this era of genetics. But the classical approach is limited because it does not involve the direct examination of genes, information on gene structure and activity being inferred from the biological characteristics of the organism being studied. By the late 1960s these indirect methods had become insufficient for answering the more detailed questions that molecular biologists had begun to ask about the expression pathways of individual genes. These questions could only be addressed by examining directly the segments of DNA containing the genes of interest. This was not possible using the current technology, so a new set of techniques had to be invented.

The development of these new techniques was stimulated by breakthroughs in biochemical research which, in the early 1970s, provided molecular biologists with enzymes that could be used to manipulate DNA molecules in the test tube. These enzymes occur naturally in living cells and are involved in processes such as DNA replication, repair and recombination (see Chapters 13 and 14). In order to determine the functions of these enzymes, many of them were purified and the reactions that they catalyze studied in the test tube. Molecular biologists then adopted the pure enzymes as tools for manipulating DNA molecules in pre-determined ways, using them to make copies of DNA molecules, to cut DNA molecules into shorter fragments, and to join them together again in combinations that do not exist in nature (Figure 4.1). These manipulations, which are described in Section 4.1, form the basis of recombinant DNA technology, in which new or ‘recombinant’ DNA molecules are constructed in the test tube from pieces of naturally occurring chromosomes and plasmids. Recombinant DNA methodology led to the development of DNA or gene cloning, in which short DNA fragments, possibly containing a single gene, are inserted into a plasmid or virus chromosome and then replicated in a bacterial or eukaryotic host (Figure 4.2). We will examine exactly how gene cloning is performed, and the reasons why this technique resulted in a revolution in molecular biology, in Section 4.2.

Figure 4.1. Examples of the manipulations that can be carried out with DNA molecules.

Figure 4.1

Examples of the manipulations that can be carried out with DNA molecules.

Figure 4.2. DNA cloning.

Figure 4.2

DNA cloning. In this example, the fragment of DNA to be cloned is inserted into a plasmid vector which is subsequently replicated inside a bacterial host.

Gene cloning was well established by the end of the 1970s. The next major technical breakthrough came some 5 years later when the polymerase chain reaction (PCR) was invented (Mullis, 1990). PCR is not a complicated technique - all that it achieves is the repeated copying of a short segment of a DNA molecule (Figure 4.3) - but it has become immensely important in many areas of biological research, not least the study of genomes. PCR is covered in detail in Section 4.3.

Figure 4.3. The polymerase chain reaction (PCR) is used to make copies of a selected segment of a DNA molecule.

Figure 4.3

The polymerase chain reaction (PCR) is used to make copies of a selected segment of a DNA molecule. In this example, a single gene is copied.

4.1. Enzymes for DNA Manipulation

Recombinant DNA technology was one of the main factors that contributed to the rapid advance in knowledge concerning gene expression that occurred during the 1970s and 1980s. The basis of recombinant DNA technology is the ability to manipulate DNA molecules in the test tube. This, in turn, depends on the availability of purified enzymes whose activities are known and can be controlled, and which can therefore be used to make specified changes to the DNA molecules that are being manipulated. The enzymes available to the molecular biologist fall into four broad categories:

Figure 4.4. The activities of (A) DNA polymerases, (B) nucleases, and (C) ligases.

Figure 4.4

The activities of (A) DNA polymerases, (B) nucleases, and (C) ligases. In (A), the activity of a DNA-dependent DNA polymerase is shown on the left and that of an RNA-dependent DNA polymerase on the right. In (B), the activities of endonucleases and exonucleases (more...)

Box Icon

Box 4.1

DNA labeling. Attachment of radioactive, fluorescent or other types of marker to DNA molecules. DNA labeling is a central part of many molecular biology procedures, including Southern hybridization (Section 4.1.2), fluorescent in situ hybridization (FISH; (more...)

4.1.1. DNA polymerases

Many of the techniques used to study DNA depend on the synthesis of copies of all or part of existing DNA or RNA molecules. This is an essential requirement for PCR (Section 4.3), DNA sequencing (Section 6.1), DNA labeling (Technical Note 4.1) and many other procedures that are central to molecular biology research. An enzyme that synthesizes DNA is called a DNA polymerase and one that copies an existing DNA or RNA molecule is called a template-dependent DNA polymerase.

The mode of action of a template-dependent DNA polymerase

A template-dependent DNA polymerase makes a new DNA polynucleotide whose sequence is dictated, via the base-pairing rules, by the sequence of nucleotides in the DNA molecule that is being copied (Figure 4.5). The mode of action is very similar to that of a template-dependent RNA polymerase (Section 3.2.2), the new polynucleotide being synthesized in the 5′→3′ direction: DNA polymerases that make DNA in the other direction are unknown in nature.

Figure 4.5. The activity of a DNA-dependent DNA polymerase.

Figure 4.5

The activity of a DNA-dependent DNA polymerase. New nucleotides are added on to the 3′ end of the growing polynucleotide, the sequence of this new polynucleotide being determined by the sequence of the template DNA.

One important difference between template-dependent DNA synthesis and the equivalent process for synthesis of RNA is that a DNA polymerase is unable to use an entirely single-stranded molecule as the template. In order to initiate DNA synthesis there must be a short double-stranded region to provide a 3′ end onto which the enzyme will add new nucleotides (Figure 4.6A). The way in which this requirement is met in living cells when the genome is replicated is described in Chapter 13. In the test tube, a DNA copying reaction is initiated by attaching to the template a short synthetic oligonucleotide, usually about 20 nucleotides in length, which acts as a primer for DNA synthesis. At first glance, the need for a primer might appear to be an undesired complication in the use of DNA polymerases in recombinant DNA technology, but nothing could be further from the truth. Because annealing of the primer to the template depends on complementary base-pairing, the position within the template molecule at which DNA copying is initiated can be specified by synthesizing a primer with the appropriate nucleotide sequence (Figure 4.6B). A short specific segment of a much longer template molecule can therefore be copied, which is much more valuable than the random copying that would occur if DNA synthesis did not need to be primed. You will fully appreciate the importance of priming when we deal with PCR in Section 4.3.

Figure 4.6. The role of the primer in template-dependent DNA synthesis.

Figure 4.6

The role of the primer in template-dependent DNA synthesis. (A) A DNA polymerase requires a primer in order to initiate the synthesis of a new polynucleotide. (B) The sequence of this oligonucleotide determines the position at which it attaches to the (more...)

A second general feature of template-dependent DNA polymerases is that many of these enzymes are multifunctional, being able to degrade DNA molecules as well as synthesize them. This is a reflection of the way in which DNA polymerases act in the cell during genome replication (Section 13.2.2). As well as their 5′→3′ DNA synthesis capability, DNA polymerases can also have one or both of the following exonuclease activities (Figure 4.7):

Figure 4.7. The DNA synthesis and exonuclease activities of DNA polymerases.

Figure 4.7

The DNA synthesis and exonuclease activities of DNA polymerases. All DNA polymerases can make DNA and many also have one or both of the exonuclease activities.

  • A 3′→5exonuclease activity enables the enzyme to remove nucleotides from the 3′ end of the strand that it has just synthesized. This is called the proofreading activity because it allows the polymerase to correct errors by removing a nucleotide that has been inserted incorrectly.
  • A 5′→3exonuclease activity is less common, but is possessed by some DNA polymerases whose natural function in genome replication requires that they must be able to remove at least part of a polynucleotide that is already attached to the template strand that the polymerase is copying.

The types of DNA polymerases used in research

Several of the template-dependent DNA polymerases that are used in molecular biology research (Table 4.1) are versions of the Escherichia coli DNA polymerase I enzyme, which plays a central role in replication of this bacterium's genome (Section 13.2.2). This enzyme, sometimes called the Kornberg polymerase, after its discoverer Arthur Kornberg (Kornberg, 1960), has both the 3′→5′ and 5′→3′ exonuclease activities, which limits it usefulness in DNA manipulation. Its main application is in DNA labeling, as described in Technical Note 4.1.

Table 4.1. Features of template-dependent DNA polymerases used in molecular biology research.

Table 4.1

Features of template-dependent DNA polymerases used in molecular biology research.

Of the two exonuclease activities, it is the 5′→3′ version that causes most problems when a DNA polymerase is used to manipulate molecules in the test tube. This is because an enzyme that possesses this activity is able to remove nucleotides from the 5′ ends of polynucleotides that have just been synthesized (Figure 4.8). It is unlikely that the polynucleotides will be completely degraded, because the polymerase function is usually much more active than the exonuclease, but some techniques will not work if the 5′ ends of the new polynucleotides are shortened in any way. In particular, DNA sequencing is based on synthesis of new polynucleotides, all of which share exactly the same 5′ end, marked by the primer used to initiate the sequencing reactions. If any nibbling of the 5′ ends occurs, then it is impossible to determine the correct DNA sequence. When DNA sequencing was first introduced in the late 1970s, it made use of a modified version of the Kornberg enzyme called the Klenow polymerase. The Klenow polymerase was initially prepared by cutting the natural E. coli DNA polymerase I enzyme into two segments with a protease. One of these segments retained the polymerase and 3′→5′ exonuclease activities, but lacked the 5′→3′ exonuclease of the untreated enzyme. The enzyme is still often called the Klenow fragment in memory of this old method of preparation, but nowadays it is almost always prepared from E. coli cells whose polymerase gene has been engineered so that the resulting enzyme has the desired properties. But in fact the Klenow polymerase is now rarely used in sequencing and has its major application in DNA labeling (see Technical Note 4.1). This is because an enzyme called Sequenase (see Table 4.1), which has superior properties as far a sequencing is concerned, was developed in the 1980s. We will return to the features of Sequenase, and why they make the enzyme ideal for sequencing, in Box 6.1

Figure 4.8. The 5′→3′ exonuclease activity of a DNA polymerase can degrade the 5′ end of a polynucleotide that has just been synthesized.

Figure 4.8

The 5′→3′ exonuclease activity of a DNA polymerase can degrade the 5′ end of a polynucleotide that has just been synthesized.

The E. coli DNA polymerase I enzyme has an optimum reaction temperature of 37 °C, this being the usual temperature of the natural environment of the bacterium, inside the intestines of mammals such as humans. Test-tube reactions with either the Kornberg or Klenow polymerases, and with Sequenase, are therefore incubated at 37 °C, and terminated by raising the temperature to 75 °C or above, which causes the protein to unfold or denature, destroying its activity. This regimen is perfectly adequate for most molecular biology techniques but, for reasons that will become clear in Section 4.3, PCR requires a thermostable DNA polymerase - one that is able to function at temperatures much higher than 37 °C. Suitable enzymes can be obtained from bacteria such as Thermus aquaticus, which live in hot springs at temperatures up to 95 °C, and whose DNA polymerase I enzyme has an optimum working temperature of 72 °C. The biochemical basis of protein thermostability is not fully understood, but probably centers on structural features that reduce the amount of protein unfolding that occurs at elevated temperatures.

One additional type of DNA polymerase is important in molecular biology research. This is reverse transcriptase, which is an RNA-dependent DNA polymerase and so makes DNA copies of RNA rather than DNA templates. Reverse transcriptases are involved in the replication cycles of retroviruses (Section 2.4.2), including the human immunodeficiency viruses that cause AIDS, these having RNA genomes that are copied into DNA after infection of the host. In the test tube, a reverse transcriptase can be used to make DNA copies of mRNA molecules. These copies are called complementary DNAs (cDNAs). Their synthesis is important in some types of gene cloning and in techniques used to map the regions of a genome that specify particular mRNAs (Section 7.1.2).

4.1.2. Nucleases

A range of nucleases have found applications in recombinant DNA technology (Table 4.2). Some nucleases have a broad range of activities but most are either exonucleases, removing nucleotides from the ends of DNA and/or RNA molecules, or endonucleases, making cuts at internal phosphodiester bonds. Some nucleases are specific for DNA and some for RNA; some work only on double-stranded DNA and others only on single-stranded DNA, and some are not fussy what they work on. We will encounter various examples of nucleases in later chapters when we deal with the techniques in which they used. Only one type of nuclease will be considered in detail here: the restriction endonucleases, which play a central role in all aspects of recombinant DNA technology.

Table 4.2. Features of important nucleases used in molecular biology research.

Table 4.2

Features of important nucleases used in molecular biology research.

Restriction endonucleases enable DNA molecules to be cut at defined positions

A restriction endonuclease is an enzyme that binds to a DNA molecule at a specific sequence and makes a double-stranded cut at or near that sequence. Because of the sequence specificity, the positions of cuts within a DNA molecule can be predicted, assuming that the DNA sequence is known, enabling defined segments to be excised from a larger molecule. This ability underlies gene cloning and all other aspects of recombinant DNA technology in which DNA fragments of known sequence are required.

There are three types of restriction endonuclease. With Types I and III there is no strict control over the position of the cut relative to the specific sequence in the DNA molecule that is recognized by the enzyme. These enzymes are therefore less useful because the sequences of the resulting fragments are not precisely known. Type II enzymes do not suffer from this disadvantage because the cut is always at the same place, either within the recognition sequence or very close to it (Figure 4.9). For example, the Type II enzyme called EcoRI (isolated from E. coli) cuts DNA only at the hexanucleotide 5′-GAATTC-3′. Digestion of DNA with a Type II enzyme therefore gives a reproducible set of fragments whose sequences are predictable if the sequence of the target DNA molecule is known. Over 2500 Type II enzymes have been isolated and more than 300 are available for use in the laboratory (Brown, 1998). Many enzymes have hexanucleotide target sites, but others recognize shorter or longer sequences (Table 4.3). There are also examples of enzymes with degenerate recognition sequences, meaning that they cut DNA at any of a family of related sites. HinfI (from Haemophilus influenzae), for example, recognizes 5′-GANTC-3′, where ‘N’ is any nucleotide, and so cuts at 5′-GAATC-3′, 5′-GATTC-3′, 5′-GAGTC-3′ and 5′-GACTC-3′. Most enzymes cut within the recognition sequence, but a few, such a BsrBI, cut at a specified position outside of this sequence.

Figure 4.9. Cuts produced by restriction endonucleases.

Figure 4.9

Cuts produced by restriction endonucleases. In the top part of the diagram, the DNA is cut by a Type I or Type III restriction endonuclease. The cuts are made in slightly different positions relative to the recognition sequence, so the resulting fragments (more...)

Table 4.3. Some examples of restriction endonucleases.

Table 4.3

Some examples of restriction endonucleases.

Restriction enzymes cut DNA in two different ways. Many make a simple double-stranded cut, giving a blunt or flush end; others cut the two DNA strands at different positions, usually two or four nucleotides apart, so that the resulting DNA fragments have short single-stranded overhangs at each end. These are called sticky or cohesive ends because base-pairing between them can stick the DNA molecule back together again (Figure 4.10A). Some sticky-end cutters give 5′ overhangs (e.g. Sau3AI, HinfI) whereas others leave 3′ overhangs (e.g. PstI) (Figure 4.10B). One feature that is particularly important in recombinant DNA technology is that some pairs of restriction enzymes have different recognition sequences but give the same sticky ends, examples being Sau3AI and BamHI, which both give a 5′-GATC-3′ sticky end even though Sau3AI has a 4-bp recognition sequence and BamHI recognizes a 6-bp sequence (Figure 4.10C).

Figure 4.10. The results of digestion of DNA with different restriction endonucleases.

Figure 4.10

The results of digestion of DNA with different restriction endonucleases. (A) Blunt ends and sticky ends. (B) Different types of sticky end: the 5′ overhangs produced by BamHI and the 3′ overhangs produced by PstI. (C) The same sticky (more...)

Examining the results of a restriction digest

After treatment with a restriction endonuclease, the resulting DNA fragments can be examined by agarose gel electrophoresis (see Technical Note 2.1) to determine their sizes. Depending on the concentration of agarose in the gel, fragments between 100 bp and 50 kb can be separated into sharp bands after electrophoresis (Figure 4.11). Fragments less than 150 bp can be separated in a 4% or 5% agarose gel, making it possible to distinguish bands representing molecules that differ in size by just a single nucleotide. With larger fragments, however, it is not always possible to separate molecules of similar size, even in gels of lower agarose concentration. If the starting DNA is long, and so gives rise to many fragments after digestion with a restriction enzyme, then the gel may simply show a smear of DNA because there are fragments of every possible length that all merge together. This is the usual result when genomic DNA is restricted.

Figure 4.11. Separation of DNA molecules by agarose gel electrophoresis.

Figure 4.11

Separation of DNA molecules by agarose gel electrophoresis. The range of fragment sizes that can be resolved depends on the concentration of agarose in the gel. Electrophoresis has been performed with three different concentrations of agarose. The labels (more...)

If the sequence of the starting DNA is known then the sequences, and hence the sizes, of the fragments resulting from treatment with a particular restriction enzyme can be predicted. The band for a desired fragment (for example, one containing a gene) can then be identified, cut out of the gel, and the DNA purified. Even if its size is unknown, a fragment containing a gene or another segment of DNA of interest can be identified by the technique called Southern hybridization, providing that some of the sequence within the fragment is known or can be predicted. The first step is to transfer the restriction fragments from the agarose gel to a nitrocellulose or nylon membrane. This is done by placing the membrane on the gel and allowing buffer to soak through, taking the DNA from the gel to the membrane, where it becomes bound (Figure 4.12A). This process results in the DNA bands becoming immobilized in the same relative positions on the surface of the membrane.

Figure 4.12. Southern hybridization.

Figure 4.12

Southern hybridization. (A) Transfer of DNA from the gel to the membrane. (B) The membrane is probed with a radioactively labeled DNA molecule. On the resulting autoradiograph, one hybridizing band is seen in lane 2, and two in lane 3.

The next step is to prepare a hybridization probe, which is a labeled DNA molecule whose sequence is complementary to the target DNA that we wish to detect. The probe could, for example, be a synthetic oligonucleotide whose sequence matches part of an interesting gene. Because the probe and target DNAs are complementary, they can base-pair or hybridize, the position of the hybridized probe on the membrane being identified by detecting the signal given out by a label attached to the probe. To carry out the hybridization, the membrane is placed in a glass bottle with the labeled probe and some buffer, and the bottle gently rotated for several hours so that the probe has plenty of opportunity to hybridize to its target DNA. The membrane is then washed to remove any probe that has not become hybridized, and the signal from the label is detected (see Technical Note 4.1). In the example shown in Figure 4.12B the probe is radioactively labeled and the signal is detected by autoradiography. The band that is seen on the autoradiograph is the one that corresponds to the restriction fragment that hybridizes to the probe and which therefore contains the gene that we are searching for.

4.1.3. DNA ligases

DNA fragments that have been generated by treatment with a restriction endonuclease can be joined back together again, or attached to a new partner, by a DNA ligase. The reaction requires energy, which is provided by adding either ATP or NAD to the reaction mixture, depending on the type of ligase that is being used.

The most widely used DNA ligase is obtained from E. coli cells infected with T4 bacteriophage. It is involved in replication of the phage DNA and is encoded by the T4 genome. Its natural role is to synthesize phosphodiester bonds between unlinked nucleotides present in one polynucleotide of a double-stranded molecule (Figure 4.13A). In order to join together two restriction fragments, the ligase has to synthesize two phosphodiester bonds, one in each strand (Figure 4.13B). This is by no means beyond the capabilities of the enzyme, but the reaction can occur only if the ends to be joined come close enough to one another by chance - the ligase is not able to catch hold of them and bring them together. If the two molecules have complementary sticky ends, and the ends come together by random diffusion events in the ligation mixture, then transient base pairs might form between the two overhangs. These base pairs are not particularly stable but they may persist for sufficient time for a ligase enzyme to attach to the junction and synthesize phosphodiester bonds to fuse the ends together (Figure 4.13C). If the molecules are blunt ended, then they cannot base-pair to one another, not even temporarily, and ligation is a much less efficient process, even when the DNA concentration is high and pairs of ends are in relatively close proximity.

Figure 4.13. Ligation of DNA molecules with DNA ligase.

Figure 4.13

Ligation of DNA molecules with DNA ligase. (A) In living cells, DNA ligase synthesizes a missing phosphodiester bond in one strand of a double-stranded DNA molecule. (B) To link two DNA molecules in vitro, DNA ligase must make two phosphodiester bonds, (more...)

The greater efficiency of sticky-end ligation has stimulated the development of methods for converting blunt ends into sticky ends. In one method, short double-stranded molecules called linkers or adaptors are attached to the blunt ends. Linkers and adaptors work in slightly different ways but both contain a recognition sequence for a restriction endonuclease and so produce a sticky end after treatment with the appropriate enzyme (Figure 4.14). Another way to create a sticky end is by homopolymer tailing, in which nucleotides are added one after the other to the 3′ terminus at a blunt end (Figure 4.15). The enzyme involved is called terminal deoxynucleotidyl transferase, which we will meet in the next section. If the reaction contains the DNA, enzyme, and only one of the four nucleotides, then the new stretch of single-stranded DNA that is made consists entirely of just that single nucleotide. It could, for example, be a poly(G) tail, which would enable the molecule to base-pair to other molecules that carry poly(C) tails, created in the same way but with dCTP rather than dGTP in the reaction mixture.

Figure 4.14. Linkers are used to place sticky ends on to a blunt-ended molecule.

Figure 4.14

Linkers are used to place sticky ends on to a blunt-ended molecule. In this example, each linker contains the recognition sequence for the BamHI restriction endonuclease. DNA ligase attaches the linkers to the ends of the blunt-ended molecule in a reaction (more...)

Figure 4.15. Homopolymer tailing.

Figure 4.15

Homopolymer tailing. In this example, a poly(G) tail is synthesized at each end of a blunt-ended DNA molecule. Tails comprising other nucleotides are synthesized by including the appropriate dNTP in the reaction mixture.

4.1.4. End-modification enzymes

Terminal deoxynucleotidyl transferase (see Figure 4.15), obtained from calf thymus tissue, is one example of an end-modification enzyme. It is, in fact, a template- independent DNA polymerase, because it is able to synthesize a new DNA polynucleotide without base-pairing of the incoming nucleotides to an existing strand of DNA or RNA. Its main role in recombinant DNA technology is in homopolymer tailing, as described above.

Two other end-modification enzymes are also frequently used. These are alkaline phosphatase and T4 polynucleotide kinase, which act in complementary ways. Alkaline phosphatase, which is obtained from various sources, including E. coli and calf intestinal tissue, removes phosphate groups from the 5′ ends of DNA molecules, which prevents these molecules from being ligated to one another. Two ends carrying 5′ phosphates can be ligated to one another, and a phosphatased end can ligate to a non-phosphatased end, but a link cannot be formed between a pair of ends if neither carries a 5′ phosphate. Judicious use of alkaline phosphatase can therefore direct the action of a DNA ligase in a pre-determined way so that only desired ligation products are obtained. T4 polynucleotide kinase, obtained from E. coli cells infected with T4 phage, performs the reverse reaction to alkaline phosphatase, adding phosphates to 5′ ends. Like alkaline phosphatase, the enzyme is used during complicated ligation experiments, but its main application is in the end-labeling of DNA molecules (see Technical Note 4.1).

4.2. DNA Cloning

DNA cloning is a logical extension of the ability to manipulate DNA molecules with restriction endonucleases and ligases. Imagine that an animal gene has been obtained as a single restriction fragment after digestion of a larger molecule with the restriction enzyme BamHI, which leaves 5′-GATC-3′ sticky ends (Figure 4.16). Imagine also that a small E. coli plasmid has been purified and treated with BamHI, which cuts the plasmid in a single position. The circular plasmid has therefore been converted into a linear molecule, again with 5′-GATC-3′ sticky ends. Mix the two DNA molecules together and add DNA ligase. Various recombinant ligation products will be obtained, one of which comprises the circularized plasmid with the animal gene inserted into the position originally taken by the BamHI restriction site. If the recombinant plasmid is now re-introduced into E. coli, and the inserted gene has not disrupted its replicative ability, then the plasmid plus inserted gene will be replicated and copies passed to the daughter bacteria after cell division. More rounds of plasmid replication and cell division will result in a colony of recombinant E. coli bacteria, each bacterium containing multiple copies of the animal gene. This series of events, as illustrated in Figure 4.16, constitutes the process called DNA or gene cloning.

Figure 4.16. An outline of gene cloning.

Figure 4.16

An outline of gene cloning. See the text for details.

4.2.1. Cloning vectors and the way they are used

In the experiment shown in Figure 4.16, the plasmid acts as a cloning vector, providing the replicative ability that enables the cloned gene to be propagated inside the host cell. Plasmids replicate efficiently in bacterial hosts because each plasmid possesses an origin of replication which is recognized by the DNA polymerases and other proteins that normally replicate the bacterium's chromosomes (Section 13.2.1). The host cell's replicative machinery therefore propagates the plasmid, plus any new genes that have been inserted into it. Bacteriophage genomes can also be used as cloning vectors because they too possess origins of replication that enable them to be propagated inside bacteria, either by the host enzymes or by DNA polymerases and other proteins specified by phage genes. The next two sections describe how plasmid and phage vectors are used to clone DNA in E. coli.

Plasmids are uncommon in eukaryotes, although Saccharomyces cerevisiae possesses one that is sometimes used for cloning purposes; most eukaryotic vectors are therefore based on virus genomes. Alternatively, with a eukaryotic host the replication requirement can be bypassed by performing the experiment in such a way that the DNA to be cloned becomes inserted into one of the host chromosomes. These approaches to cloning in eukaryotic cells are described later in the chapter.

Vectors based on E. coli plasmids

The easiest way to understand how a cloning vector is used is to start with the simplest E. coli plasmid vectors, which illustrate all of the basic principles of DNA cloning. We will then be able to turn our attention to the special features of phage vectors and vectors used with eukaryotes.

One of the first plasmid vectors to be developed was pBR322 (Bolivar et al., 1977), which was constructed by ligating restriction fragments from three naturally occurring E. coli plasmids: R1, R6.5 and pMB1. The pBR322 plasmid is small (just 4363 bp) and, as well as the origin of replication, it carries genes coding for enzymes that enable the host bacterium to withstand the growth-inhibitory effects of two antibiotics: ampicillin and tetracycline (Figure 4.17). This means that cells containing a pBR322 plasmid can be distinguished from those that do not by plating the bacteria onto agar medium containing ampicillin and/or tetracycline. Normal E. coli cells are sensitive to these antibiotics and cannot grow when either of the two antibiotics is present. Ampicillin and tetracycline resistance are therefore selectable markers for pBR322.

Figure 4.17. pBR322.

Figure 4.17

pBR322. The map shows the positions of the ampicillin-resistance gene (amp R), the tetracycline-resistance gene (tet R), the origin of replication (ori) and the recognition sequences for seven restriction endonucleases.

The manipulations shown in Figure 4.16, resulting in construction of a recombinant plasmid, are carried out in the test tube with purified DNA. Pure pBR322 DNA can be obtained quite easily from extracts of bacterial cells (Technical Note 4.2), but how can the manipulated plasmids be re-introduced into the bacteria? The answer is to make use of the natural processes for transformation of bacteria, which result in the uptake of ‘naked’ DNA by a bacterial cell. This is the process studied by Avery and his colleagues in the experiments which showed that bacterial genes are made of DNA (Section 1.1.1). Transformation is not a particularly efficient process in many bacteria, including E. coli, but the rate of uptake can be enhanced significantly by suspending the cells in calcium chloride before adding the DNA, and then briefly incubating the mixture at 42 °C. Even after this enhancement, only a very small proportion of the cells take up a plasmid. This is why the antibiotic-resistance markers are so important - they allow the small number of transformants to be selected from the large background of non-transformed cells.

Box Icon

Box 4.2

DNA purification. Techniques for the preparation of pure samples of DNA from living cells play a central role in molecular biology research. The first step in DNA purification is to break open the cells from which the DNA will be obtained. With some types (more...)

The map of pBR322 shown in Figure 4.17 indicates the positions of the recognition sequences for seven restriction enzymes, each of which cuts the plasmid at just one location. Note that six of these sites lie within one or other of the genes for antibiotic resistance. This means that if a new fragment of DNA is ligated into one of these six sites, then the antibiotic-resistance properties of the plasmid become altered - the plasmid loses the ability to confer either ampicillin or tetracycline resistance on the host cells. This is called insertional inactivation of the selectable marker and is the key to distinguishing a recombinant plasmid - one that contains an inserted piece of DNA - from a non-recombinant plasmid that has no new DNA. Identifying recombinants is important because the manipulations illustrated in Figure 4.16 result in a variety of ligation products, including plasmids that have recircularized without insertion of new DNA. To identify recombinants, the resistance properties of colonies are tested by transferring cells from agar containing one antibiotic onto agar containing the second antibiotic. For example, if the BamHI site has been used then recombinants will be ampicillin resistant but tetracycline sensitive, because the BamHI site lies within the region that specifies resistance to tetracycline. After transformation, cells are plated onto ampicillin agar (Figure 4.18). All cells that contain a pBR322 plasmid, whether recombinant or not, divide and produce a colony. The colonies are then transferred onto tetracycline agar by replica plating, which results in the colonies on the second plate retaining the relative positions that they had on the first plate. Some colonies do not grow on the tetracycline agar because their cells contain recombinant pBR322 molecules with a disrupted tetracycline-resistance gene. These are the colonies we are looking for because they contain the cloned gene, so we return to the ampicillin plate, from which samples of the cells can be recovered.

Figure 4.18. Recombinant selection with pBR322.

Figure 4.18

Recombinant selection with pBR322. See text for details. The inset shows how replica plating is performed.

Replica plating is not a difficult technique but it takes time. It would be much better if recombinants could be distinguished from non-recombinants simply by plating onto a single agar medium. This is possible with most of the modern plasmid cloning vectors, including pUC8 (Figure 4.19; Vieira and Messing, 1982). This vector carries the ampicillin-resistance gene from pBR322, along with a second gene, called lacZ′, which is part of the E. coli gene for the enzyme β-galactosidase. The remainder of the lacZ gene is located in the chromosome of the special E. coli strain that is used when cloning genes with pUC8. The proteins specified by the gene segments on the plasmid and on the chromosome are able to combine to produce a functional β-galactosidase enzyme. The presence of functional β-galactosidase molecules in the cells can be checked by a histochemical test with a compound called X-gal (5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside), which the enzyme converts into a blue product. The lacZ′ gene contains a cluster of unique restriction sites; insertion of new DNA into any one of these sites results in insertional inactivation of the gene and hence loss of β-galactosidase activity. Recombinants and non-recombinants can therefore be distinguished simply by plating the transformed cells onto agar containing ampicillin and X-gal (Figure 4.19). All colonies that grow on this medium are made up of transformed cells because only transformants are ampicillin resistant. Some colonies are blue and some are white. Those that are blue contain cells with functional β-galactosidase enzymes and hence with undisrupted lacZ′ genes; these colonies are therefore non-recombinants. The white colonies comprise cells without β-galactosidase activity and hence with disrupted lacZ′ genes; these are recombinants.

Figure 4.19. Recombinant selection with pUC8.

Figure 4.19

Recombinant selection with pUC8. See the text for details.

Cloning vectors based on E. coli bacteriophage genomes

E. coli bacteriophages were developed as cloning vectors back in the earliest days of the recombinant DNA revolution. The main reason for seeking a different type of vector was the inability of plasmids such as pBR322 and pUC8 to handle DNA fragments greater than about 10 kb in size, larger inserts undergoing rearrangements or interfering with the plasmid replication system in such a way that the recombinant DNA molecules become lost from the host cells. The first attempts to develop vectors able to handle larger fragments of DNA centered on bacteriophage λ. The infection cycle of λ is similar to that of the T2 phages studied by Hershey and Chase in the experiments that alerted molecular biologists to the fact that genes are made of DNA (Section 1.1.1), but there is one important difference. As well as following the lytic infection cycle (see Figure 1.4B), the λ genome is able to integrate into the bacterial chromosome, where it can remain quiescent for many generations, being replicated along with the host chromosome every time the cell divides. This is called the lysogenic infection cycle (Figure 4.20).

Figure 4.20. The lysogenic infection cycle of bacteriophage λ.

Figure 4.20

The lysogenic infection cycle of bacteriophage λ. Compare with the lytic infection cycle of T2 bacteriophage, shown in Figure 1.4B. The special feature of the lysogenic cycle is the insertion of the phage genome into the bacterium's chromosomal (more...)

The λ genome is 48.5 kb, of which some 15 kb or so is ‘optional’ in that it contains genes that are only needed for integration of the phage DNA into the E. coli chromosome (Figure 4.21A). These segments can therefore be deleted without impairing the ability of the phage to infect bacteria and direct synthesis of new λ particles by the lytic cycle. Two types of vector have been developed (Figure 4.21B):

Figure 4.21. Cloning vectors based on bacteriophage λ.

Figure 4.21

Cloning vectors based on bacteriophage λ. (A) In the λ genome, the genes are arranged into functional groups. For example, the region marked as ‘protein coat’ comprises 21 genes coding for proteins that are either components (more...)

  • Insertion vectors, in which part or all of the optional DNA has been removed and a unique restriction site introduced at some position within the trimmed down genome;
  • Replacement vectors, in which the optional DNA is contained within a stuffer fragment, flanked by a pair of restriction sites, that is replaced when the DNA to be cloned is ligated into the vector.

The λ genome is linear, but the two natural ends of the molecule have 12-nucleotide single-stranded overhangs, called cos sites, which have complementary sequences and so can base-pair to one another. A λ cloning vector can therefore be obtained as a circular molecule which can be manipulated in the test tube in the same way as a plasmid, and re-introduced into E. coli by transfection, the term used for uptake of naked phage DNA. Alternatively, a more efficient uptake system called in vitro packaging can be utilized (Hohn and Murray, 1977). This procedure starts with the linear version of the cloning vector, the initial restriction cutting the molecule into two segments, the left and right arms, each with a cos site at one end. The ligation is carried out with carefully measured quantities of each arm and the DNA to be cloned, the aim being to produce concatamers in which the different fragments are linked together in the order left arm-new DNA-right arm, as shown in Figure 4.22. The concatamers are then added to an in vitro packaging mix, which contains all the proteins needed to make a λ phage particle. These proteins form phage particles spontaneously, and will place inside the particles any DNA fragment that is between 37 and 52 kb in length and is flanked by cos sites. The packaging mix therefore cuts left arm-new DNA-right arm combinations of 37–52 kb out of the concatamers and constructs λ phages around them. The phages are then mixed with E. coli cells, and the natural infection process transports the vector plus new DNA into the bacteria.

Figure 4.22. Cloning with a λ insertion vector.

Figure 4.22

Cloning with a λ insertion vector. The linear form of the vector is shown at the top of the diagram. Treatment with the appropriate restriction endonuclease produces the left and right arms, both of which have one blunt end and one end with the (more...)

After infection, the cells are spread onto an agar plate. The objective is not to obtain individual colonies but to produce an even layer of bacteria across the entire surface of the agar. Bacteria that were infected with the packaged cloning vector die within about 20 minutes because the λ genes contained in the arms of the vector direct replication of the DNA and synthesis of new phages by the lytic cycle, each of these new phages containing its own copy of the vector plus cloned DNA. Death and lysis of the bacterium releases these phages into the surrounding medium, where they infect new cells and begin another round of phage replication and lysis. The end result is a zone of clearing, called a plaque, which is visible on the lawn of bacteria that grows on the agar plate (Figure 4.23). With some λ vectors, all plaques are made up of recombinant phages because ligation of the two arms without insertion of new DNA results in a molecule that is too short to be packaged. With other vectors it is necessary to distinguish recombinant plaques from non-recombinant ones. Various methods are used, including the β-galactosidase system described above for the plasmid vector pUC8 (see Figure 4.19), which is also applicable to those λ vectors that carry a fragment of the lacZ gene into which the DNA to be cloned is inserted.

Figure 4.23. Bacteriophage infection is visualized as a plaque on a lawn of bacteria.

Figure 4.23

Bacteriophage infection is visualized as a plaque on a lawn of bacteria.

Vectors for longer pieces of DNA

The λ phage particle can accommodate up to 52 kb of DNA, so if the genome has 15 kb removed then up to 18 kb of new DNA can be cloned. This limit is higher than that for plasmid vectors, but is still very small compared with the sizes of intact genomes. The comparison is important because a clone library - a collection of clones whose inserts cover an entire genome - is the starting point for a project aimed at determining the sequence of that genome (Chapter 6). If a λ vector is used with human DNA, then over half a million clones are needed for there to be a 95% chance of any particular part of the genome being present in the library (Table 4.4). It is possible to prepare a library comprising half a million clones, especially if automated techniques are used, but such a large collection is far from ideal. It would be much better to reduce the number of clones by using a vector that is able to handle fragments of DNA longer than 18 kb. Many of the developments in cloning technology over the last 20 years have been aimed at finding ways of doing this.

Table 4.4. Sizes of human genomic libraries prepared in different types of cloning vector.

Table 4.4

Sizes of human genomic libraries prepared in different types of cloning vector.

One possibility is to use a cosmid - a plasmid that carries a λ cos site (Figure 4.24). Concatamers of cosmid molecules, linked at their cos sites, act as substrates for in vitro packaging because the cos site is the only sequence that a DNA molecule needs in order to be recognized as a ‘λ genome’ by the proteins that package DNA into λ phage particles. Particles containing cosmid DNA are as infective as real λ phages, but once inside the cell the cosmid cannot direct synthesis of new phage particles and instead replicates as a plasmid. Recombinant DNA is therefore obtained from colonies rather than plaques. As with other types of λ vector, the upper limit for the length of the cloned DNA is set by the space available within the λ phage particle. A cosmid can be 8 kb or less in size, so up to 44 kb of new DNA can be inserted before the packaging limit of the λ phage particle is reached. This reduces the size of the human genomic library to about a quarter of a million clones, which is an improvement compared with a λ library, but is still a massive number of clones to have to work with.

Figure 4.24. A typical cosmid.

Figure 4.24

A typical cosmid. pJB8 is 5.4 kb in size and carries the ampicillin-resistance gene (amp R), a segment of λ DNA containing the cos site, and an Escherichia coli origin of replication (ori).

The first major breakthrough in attempts to clone DNA fragments much longer than 50 kb came with the invention of yeast artificial chromosomes or YACs (Burke et al., 1987). These vectors are propagated in S. cerevisiae rather than in E. coli and are based on chromosomes, rather than on plasmids or viruses. The first YACs were constructed after studies of natural chromosomes had shown that, in addition to the genes that it carries, each chromosome has three important components:

  • The centromere, which plays a critical role during nuclear division (see Figure 2.7);
  • The telomeres, the special sequences which mark the ends of chromosomal DNA molecules (see Figure 2.10);
  • One or more origins of replication, which initiate synthesis of new DNA when the chromosome divides (Section 13.2.1).

In a YAC, the DNA sequences that underlie these chromosomal components are linked together with one or more selectable markers and at least one restriction site into which new DNA can be inserted (Figure 4.25). All of these components can be contained in a DNA molecule of 10–15 kb. Natural yeast chromosomes range from 230 kb to over 1700 kb, so YACs have the potential to clone Mb-sized DNA fragments. This potential has been realized, standard YACs being able to clone 600 kb fragments, with special types able to handle DNA up to 1400 kb in length. Currently this is the highest capacity of any type of cloning vector, and several genome projects have made extensive use of YACs. Unfortunately, with some types of YAC there have been problems with insert stability, the cloned DNA becoming rearranged into new sequence combinations (Anderson, 1993). For this reason there is also great interest in other types of vectors, ones that cannot clone such large pieces of DNA but which suffer less from instability problems. These vectors include the following:

Figure 4.25. Working with a YAC.

Figure 4.25

Working with a YAC. (A) The cloning vector pYAC3. (B) To clone with pYAC3, the circular vector is digested with BamHI and SnaBI. BamHI restriction removes the stuffer fragment held between the two telomeres in the circular molecule. SnaBI cuts within (more...)

  • Bacteriophage P1 vectors (Sternberg, 1990) are very similar to λ vectors, being based on a deleted version of a natural phage genome, the capacity of the cloning vector being determined by the size of the deletion and the space within the phage particle. The P1 genome is larger than the λ genome, and the phage particle is bigger, so a P1 vector can clone larger fragments of DNA than a λ vector, up to 125 kb using current technology.
  • Bacterial artificial chromosomes or BACs (Shizuya et al., 1992) are based on the naturally occurring F plasmid of E. coli. Unlike the plasmids used to construct the early cloning vectors, the F plasmid is relatively large and vectors based on it have a higher capacity for accepting inserted DNA. BACs can be used to clone fragments of 300 kb and longer.
  • P1-derived artificial chromosomes or PACs (Ioannou et al., 1994) combine features of P1 vectors and BACs and have a capacity of up to 300 kb.
  • Fosmids (Kim et al., 1992) contain the F plasmid origin of replication and a λ cos site. They are similar to cosmids but have a lower copy number in E. coli, which means that they are less prone to instability problems.

The sizes of human genome libraries prepared in these various types of vector are given in Table 4.4.

Cloning in organisms other than E. coli

Cloning is not merely an aid to DNA sequencing: it also provides a means of studying the mode of expression of a gene and the way in which expression is regulated, of carrying out genetic engineering experiments aimed at modifying the biological characteristics of the host organism, and of synthesizing important animal proteins, such as pharmaceuticals, in a new host cell from which the proteins can be obtained in larger quantities than is possible by conventional purification from animal tissue. These multifarious applications demand that genes must frequently be cloned in organisms other than E. coli.

Cloning vectors based on plasmids or phages have been developed for most of the well studied species of bacteria such as Bacillus, Streptomyces and Pseudomonas, these vectors being used in exactly the same way as the E. coli analogs. Plasmid vectors are also available for yeasts and fungi. Some of these carry the origin of replication from the 2 μm circle, a plasmid present in many strains of S. cerevisiae, but other plasmid vectors only have an E. coli origin. An example is YIp5, an S. cerevisiae vector that is simply a pBR322 plasmid that contains a copy of the yeast gene called URA3 (Figure 4.26A). What is the logic behind the construction of YIp5? When used in a cloning experiment, the vector is initially used with E. coli as the host, up to the stage where the desired recombinant molecule has been constructed by restriction and ligation. The recombinant vector is then purified from E. coli and transferred into S. cerevisiae, usually by mixing the DNA with protoplasts - yeast cells whose walls have been removed by enzyme treatment. Without an origin of replication the vector is unable to propagate independently inside yeast cells, but it can survive if it becomes integrated into one of the yeast chromosomes, which can occur by homologous recombination (Section 7.2.2) between the URA3 gene carried by the vector and the chromosomal copy of this gene (Figure 4.26B). ‘YIp’ in fact stands for ‘yeast integrative plasmid’. Once integrated the YIp, plus any DNA that has been inserted into it, replicates along with the host chromosomes.

Figure 4.26. Cloning with a YIp.

Figure 4.26

Cloning with a YIp. (A) YIp5, a typical yeast integrative plasmid. The plasmid contains the ampicillin-resistance gene (amp R), the tetracycline-resistance gene (tet R), the yeast gene URA3, and an Escherichia coli origin of replication (ori). The presence (more...)

Integration into chromosomal DNA is also a feature of many of the cloning systems used with animals and plants, and forms the basis of the construction of knockout mice, which are used to determine the functions of previously unknown genes that are discovered in the human genome (Section 7.2.2). The vectors are animal equivalents of YIps. Adenoviruses and retroviruses are used to clone genes in animals when the objective is to treat a genetic disease or a cancer by gene therapy (Lemoine and Cooper, 1998). A similar range of vectors has been developed for cloning genes in plants. Plasmids can be introduced into plant embryos by bombardment with DNA-coated microprojectiles, a process called biolistics (Klein et al., 1987), integration of the plasmid DNA into the plant chromosomes, followed by growth of the embryo, resulting in a plant that contains the cloned DNA in most or all of its cells. Some success has also been achieved with plant vectors based on the genomes of caulimoviruses and geminiviruses (Timmermans et al., 1994; Viaplana et al., 2001), but the most interesting types of plant cloning vector are those derived from the Ti plasmid (Hansen and Wright, 1999). This is a large bacterial plasmid found in the soil microorganism Agrobacterium tumefaciens, part of which, the T-DNA, becomes integrated into a plant chromosome when the bacterium infects a plant stem and causes crown root disease. The T-DNA carries a number of genes that are expressed inside the plant cells and induce the various physiological changes that characterize the disease. Vectors such as pBIN19 (Figure 4.27) have been designed to make use of this natural genetic engineering system (Bevan, 1984). The recombinant vector is introduced into A. tumefaciens cells, which are allowed to infect a cell suspension or plant callus culture, from which mature transformed plants can be regenerated.

Figure 4.27. The plant cloning vector pBIN19.

Figure 4.27

The plant cloning vector pBIN19. pBIN19 carries the lacZ′ gene (see Figure 4.19), the kanamycin-resistance gene (kan R), an Escherichia coli origin of replication (ori), and the two boundary sequences from the T-DNA region of the Ti plasmid. These (more...)

4.3. The Polymerase Chain Reaction (PCR)

In essence, DNA cloning results in the purification of a single fragment of DNA from a complex mixture of DNA molecules. Cloning is a powerful technique and its impact on our understanding of genes and genomes has been immeasurable. Cloning does, however, have one major disadvantage: it is a time-consuming and, in parts, difficult procedure. It takes several days to perform the manipulations needed to insert DNA fragments into a cloning vector and then introduce the ligated molecules into the host cells and select recombinants. If the experimental strategy involves generation of a large clone library, followed by screening of the library to identify a clone that contains a gene of interest (see Technical Note 4.3), then several more weeks or even months might be needed to complete the project.

Box Icon

Box 4.3

Working with a clone library. Clone collections used as a source of genes and other DNA segments. Clone libraries have been used in molecular biology research for many years and their importance extends well beyond their role as the starting point for (more...)

PCR complements DNA cloning in that it enables the same result to be achieved - purification of a specified DNA fragment - but in a much shorter time, perhaps just a few hours (Saiki et al., 1988). PCR is complementary to, not a replacement for, cloning because it has its own limitations, the most important of which is the need to know the sequence of at least part of the fragment that is to be purified. Despite this constraint, PCR has acquired central importance in many areas of molecular biology research. We will examine the technique first and then survey its applications.

4.3.1. Carrying out a PCR

PCR results in the repeated copying of a selected region of a DNA molecule (see Figure 4.3). Unlike cloning, PCR is a test-tube reaction and does not involve the use of living cells: the copying is carried out not by cellular enzymes but by the purified, thermostable DNA polymerase of T. aquaticus (Section 4.1.1). The reason why a thermostable enzyme is needed will become clear when we look in more detail at the events that occur during PCR.

To carry out a PCR experiment, the target DNA is mixed with Taq DNA polymerase, a pair of oligonucleotide primers, and a supply of nucleotides. The amount of target DNA can be very small because PCR is extremely sensitive and will work with just a single starting molecule. The primers are needed to initiate the DNA synthesis reactions that will be carried out by the Taq polymerase (see Figure 4.6). They must attach to the target DNA at either side of the segment that is to be copied; the sequences of these attachment sites must therefore be known so that primers of the appropriate sequences can be synthesized.

The reaction is started by heating the mixture to 94 °C. At this temperature the hydrogen bonds that hold together the two polynucleotides of the double helix are broken, so the target DNA becomes denatured into single- stranded molecules (Figure 4.28). The temperature is then reduced to 50–60 °C, which results in some rejoining of the single strands of the target DNA, but also allows the primers to attach to their annealing positions. DNA synthesis can now begin, so the temperature is raised to 72 °C, the optimum for Taq polymerase. In this first stage of the PCR, a set of ‘long products’ is synthesized from each strand of the target DNA. These polynucleotides have identical 5′ ends but random 3′ ends, the latter representing positions where DNA synthesis terminates by chance. When the cycle of denaturation-annealing- synthesis is repeated, the long products act as templates for new DNA synthesis, giving rise to ‘short products’ whose 5′ and 3′ ends are both set by the primer annealing positions (Figure 4.29). In subsequent cycles, the number of short products accumulates in an exponential fashion (doubling during each cycle) until one of the components of the reaction becomes depleted. This means that after 30 cycles, there will be over 250 million short products derived from each starting molecule. In real terms, this equates to several micrograms of PCR product from a few nanograms or less of target DNA.

Figure 4.28. The first stage of a PCR.

Figure 4.28

The first stage of a PCR. See the text for details.

Figure 4.29. The synthesis of ‘short’ products in a PCR.

Figure 4.29

The synthesis of ‘short’ products in a PCR. The first cycle products from Figure 4.28 are shown at the top. The next cycle of denaturation-annealing-synthesis leads to four products, two of which are identical to the first cycle products (more...)

The results of a PCR can be determined in various ways. Usually, the products are analyzed by agarose gel electrophoresis, which will reveal a single band if the PCR has worked as expected and has amplified a single segment of the target DNA (Figure 4.30). Alternatively, the sequence of the product can be determined, using techniques described in Section 6.1.1.

Figure 4.30. Analysing the results of a PCR by agarose gel electrophoresis.

Figure 4.30

Analysing the results of a PCR by agarose gel electrophoresis. The PCR has been carried out in a microfuge tube. A sample is loaded into lane 2 of an agarose gel. Lane 1 contains DNA size markers, and lane 3 contains a sample of a PCR done by a colleague. (more...)

4.3.2. The applications of PCR

PCR is such a straightforward procedure that it is sometimes difficult to understand how it can have become so important in modern research. First we will deal with its limitations. In order to synthesize primers that will anneal at the correct positions, the sequences of the boundary regions of the DNA to be amplified must be known. This means that PCR cannot be used to purify fragments of genes or other parts of a genome that have never been studied before. A second constraint is the length of DNA that can be copied. Regions of up to 5 kb can be amplified without too much difficulty, and longer amplifications - up to 40 kb - are possible using modifications of the standard technique. However, the >100 kb fragments that are needed for genome sequencing projects are unattainable by PCR.

What are the strengths of PCR? Primary among these is the ease with which products representing a single segment of the genome can be obtained from a number of different DNA samples. We will encounter one important example of this in the next chapter when we look at how DNA markers are typed in genetic mapping projects (Section 5.2.2). PCR is used in a similar way to screen human DNA samples for mutations associated with genetic diseases such as thalassemia and cystic fibrosis. It also forms the basis of genetic profiling, in which variations in microsatellite length are typed (see Figure 2.25).

A second important feature of PCR is its ability to work with minuscule amounts of starting DNA. This means that PCR can be used to obtain sequences from the trace amounts of DNA that are present in hairs, bloodstains and other forensic specimens, and from bones and other remains preserved at archaeological sites. In clinical diagnosis, PCR is able to detect the presence of viral DNA well before the virus has reached the levels needed to initiate a disease response. This is particularly important in the early identification of viral-induced cancers because it means that treatment programs can be initiated before the cancer becomes established.

The above are just a few of the applications of PCR. The technique is now a major component of the molecular biologist's toolkit and we will discover many more examples of its use in the study of genomes as we progress through the remaining chapters of this book.

Study Aids For Chapter 4

Self study questions


Draw diagrams that outline the events that occur during (a) DNA cloning, and (b) PCR. What are the limitations of each of these two techniques?


List the types of enzyme used in recombinant DNA research.


Distinguish between the two types of exonuclease activity that can be possessed by a DNA polymerase, and explain how these activities influence the potential applications of individual DNA polymerases in recombinant DNA research.


Using examples, describe the various types of end produced after digestion of DNA with a restriction endonuclease.


How are agarose gel electrophoresis and Southern hybridization used to examine the results of a restriction digest?


Explain why the efficiency of blunt-end ligation is less than that of sticky-end ligation. What steps can be taken to improve the efficiency of blunt-end ligation?


Draw diagrams of (a) pBR322, and (b) pUC8. Explain how the differences between these two vectors influence the ways in which they are used to clone DNA fragments.


Distinguish between the lytic and lysogenic infection cycles for a bacteriophage.


Write a short description of the way in which a bacteriophage λ vector is used to clone DNA. How does a cosmid differ from a standard λ vector?


Draw a diagram showing a typical YAC. Indicate the key features and explain how a YAC is used to clone DNA.


What problems might arise when a YAC is used to clone a large fragment of DNA? To what extent can these problems be solved by the use of other types of high-capacity cloning vector?


How is DNA cloned in organisms other than Escherichia coli?


Describe how a PCR is carried out, paying particular attention to the role of the primers and the temperatures used during the thermal cycling.

Problem-based learning


Soon after the first gene cloning experiments were carried out in the early 1970s, a number of scientists argued that there should be a temporary moratorium on this type of research. What was the basis of these scientists' fears and to what extent were these fears justified?


What would be the features of an ideal cloning vector? To what extent are these requirements met by any of the existing cloning vectors?


The specificity of the primers is a critical feature of a successful PCR. If the primers anneal at more than one position in the target DNA then products additional to the one being sought will be synthesized. Explore the factors that determine primer specificity and evaluate the influence of the annealing temperature on the outcome of a PCR.


  1. Anderson C. Genome shortcut leads to problems. Science. (1993);259:1684–1687. [PubMed: 8456291]
  2. Bevan M. Binary Agrobacterium vectors for plant transformation. Nucleic Acids Res. (1984);12:8711–8721. [PMC free article: PMC320409] [PubMed: 6095209]
  3. Bolivar F, Rodriguez RL, Greene PJ. et al. Construction and characterisation of new cloning vectors II. A multi-purpose cloning system. Gene. (1977);2:95–113. [PubMed: 344137]
  4. Brown TA (1998) Molecular Biology Labfax, 2nd edition, Vol. 1. Academic Press, London.
  5. Brown TA (2001) Gene Cloning and DNA Analysis: An Introduction, 4th edition. Blackwell Scientific Publishers, Oxford.
  6. Burke DT, Carle GF, Olson MV. Cloning of large segments of exogenous DNA into yeast by means of artificial chromosome vectors. Science. (1987);236:806–812. [PubMed: 3033825]
  7. Hansen G, Wright MS. Recent advances in the transformation of plants. Trends Plant Sci. (1999);4:226–231. [PubMed: 10366879]
  8. Hohn B, Murray K. Packaging recombinant DNA molecules into bacteriophage particles in vitro. Proc. Natl Acad. Sci. USA. (1977);74:3259–3263. [PMC free article: PMC431522] [PubMed: 333431]
  9. Ioannou PA, Amemiya CT, Garnes J. et al. P1-derived vector for the propagation of large human DNA fragments. Nature Genet. (1994);6:84–89. [PubMed: 8136839]
  10. Kim U-J, Shizuya H, de Jong PJ, Birren B, Simon MI. Stable propagation of cosmid and human DNA inserts in an F factor based vector. Nucleic Acids Res. (1992);20:1083–1085. [PMC free article: PMC312094] [PubMed: 1549470]
  11. Klein RM, Wolf ED, Wu R, Sanford JC. High velocity microprojectiles for delivering nucleic acids into living cells. Biotechnology. (1992);24:384–7386. [PubMed: 1422046]
  12. Kornberg A. Biologic synthesis of deoxyribonucleic acid. Science. (1960);131:1503–1508. [PubMed: 14411056]
  13. Lemoine N and Cooper D (1998) Gene Therapy. BIOS Scientific Publishers, Oxford.
  14. Mullis KB. The unusual origins of the polymerase chain reaction. Sci. Am. (1990);262(4):56–65. [PubMed: 2315679]
  15. Saiki RK, Gelfand DH, Stoffel S. et al. Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science. (1988);239:487–491. [PubMed: 2448875]
  16. Shizuya H, Birren B, Kim UJ. et al. Cloning and stable maintenance of 300-kilobase-pair fragments of human DNA in Escherichia coli using an F-factor-based vector. Proc. Natl Acad. Sci. USA. (1992);89:8794–8797. [PMC free article: PMC50007] [PubMed: 1528894]
  17. Sternberg N. Bacteriophage P1 cloning system for the isolation, amplification, and recovery of DNA fragments as large as 100 kilobase pairs. Proc. Natl Acad. Sci. USA. (1990);87:103–107. [PMC free article: PMC53208] [PubMed: 2404272]
  18. Timmermans MCP, Das OP, Messing J. Geminiviruses and their uses as extrachromosomal replicons. Ann. Rev. Plant Physiol. Plant Mol. Biol. (1994);45:79–112.
  19. Viaplana R, Turner DS, Covey SN. Transient expression of a GUS reporter gene from cauliflower mosaic virus replacement vectors in the presence and absence of helper virus. J. Gen. Virol. (2001);82:59–65. [PubMed: 11125159]
  20. Vieira J, Messing J. The pUC plasmids, an M13mp7-derived system for insertion mutagenesis and sequencing with synthetic universal primers. Gene. (1982);19:259–268. [PubMed: 6295879]

Further Reading

  1. Blackman K. The advent of genetic engineering. Trends Biochem. Sci. (2001);26:268–270.A personal account of the early days of gene cloning. [PubMed: 11295561]
  2. Dale JW (1998) Molecular Genetics of Bacteria, 3rd edition. Wiley, Chichester. —Provides a detailed description of plasmids and bacteriophages.
  3. Monaco AP, Larin Z. YACs, BACs, PACs and MACs - artificial chromosomes as research tools. Trends Biotechnol. (1994);12:280–286.A good review of high-capacity cloning vectors. [PubMed: 7765076]
  4. Southern EM. Blotting at 25. Trends Biochem. Sci. (2000);25:585–588.The origins of Southern hybridization. [PubMed: 11116181]
  5. Watson JD, Gilman M, Witkowski J and Zoller M (1992) Recombinant DNA, 2nd edition. W.H. Freeman, New York. —Detailed descriptions of basic recombinant DNA methodology.
Image ch1f6
Image ch13f12
Image ch13f17
Image ch1f4a
Image ch2f7
Image ch2f10
Image ch6f14
Image ch11f11
Image ch7f4
Image ch5f32
Image ch2f25
Copyright © 2002, Garland Science.
Bookshelf ID: NBK21129


Related information

  • PMC
    PubMed Central citations
  • PubMed
    Links to PubMed

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...