1: Basic Molecular Biology Techniques

The ability to study processes at the molecular level has had and is still having a profound effect on many areas of biosciences. New and powerful ways of isolating, analysing, manipulating and exploiting nucleic acids have all been developed and refined over many years. The completion of numerous genome projects has allowed the continued development of new, exciting areas of biological sciences such as biotechnology, genome mapping, molecular medicine and gene therapy. Perhaps one of the most startling applications of molecular biology to date is gene editing and the development of gene modification methods such as the CRISPR/cas9 system that are likely to have a profound impact on many areas of biosciences, from crop improvement to potentially treating genetic-based diseases.¹  This and other methods are having a profound impact in many areas of molecular biology and biotechnology and will be a highly important resource. In considering the potential utility of molecular biology techniques, it is important to understand the basic structure of nucleic acids and gain an appreciation of how this dictates the function in vivo and in vitro. Indeed, many techniques used in molecular biology mimic in some way the natural functions of nucleic acids such as replication and transcription. In recent years, there has been much focus on ‘omics’ technology in a number of fields such as ‘genomics’, ‘proteomics’ and ‘transcriptomics’, among many others.²  This developing area attempts to address critical biological problems as a whole and the interactions within the area. This type of integrative approach using the techniques outlined in the following chapters will no doubt provide many insights into biological processes. This chapter is intended to provide an overview of the general features of nucleic acid structure and function and to describe some of the basic methods used in their isolation and analysis.

DNA and RNA are macromolecular structures composed of regular repeating polymers formed from nucleotides.³  These are the basic building blocks of nucleic acids and are derived from nucleosides which are composed of two elements: a five-membered pentose carbon sugar (2-deoxyribose in DNA and ribose in RNA) and a nitrogenous base (Figure 1.1). The carbon atoms of the sugar are labelled with ‘primes’ (l′, 2′, 3′, etc.) to distinguish them from the carbons of nitrogenous bases, of which there are two types: either a purine or a pyrimidine. A nucleotide or nucleoside phosphate is formed by the attachment of a phosphate to the 5′-position of a nucleoside by an ester linkage. Such nucleotides can be joined together by the formation of a second ester bond by reaction between the phosphate of one nucleotide and the 3′-hydroxyl of another, thus generating a 5′ to 3′ phosphodiester bond between adjacent sugars; this process can be repeated indefinitely to give long polynucleotide molecules. DNA has two such polynucleotide strands; however, since each strand has both a free 5′-hydroxyl group at one end and a free 3′-hydroxyl at the other end, each strand has a polarity or directionality. The polarities of the two strands of the molecule are in opposite directions and DNA is therefore described as having an ‘anti-parallel’ structure.

The purine bases (composed of fused five- and six-membered rings) adenine (A) and guanine (G) are found in both RNA and DNA, as is the pyrimidine (a single six-membered ring) cytosine (C). The other pyrimidines are each restricted to one type of nucleic acid: uracil (U) occurs exclusively in RNA, whilst thymine (T) is limited to DNA. Hence it is possible to distinguish between RNA and DNA on the basis of the presence of ribose and uracil in RNA and deoxyribose and thymine in DNA. However, it is the sequence of bases along a molecule that distinguishes one DNA (or RNA) from another.

The two polynucleotide chains in DNA are usually found in the shape of a right-handed double helix, in which the bases of the two strands lie in the centre of the molecule, with the sugar–phosphate backbones on the outside. A crucial feature of this double-stranded structure is that it depends on the sequence of bases in one strand being complementary to that in the other. A purine base attached to a sugar residue on one strand is always hydrogen bonded to a pyrimidine base attached to a sugar residue on the other strand. Moreover, adenine (A) always pairs with thymine (T) or uracil (U) in RNA, via two hydrogen bonds, and guanine (G) always pairs with cytosine (C), via three hydrogen bonds (Figure 1.2). When these conditions are met, a stable double-helical structure results in which the backbones of the two strands are, on average, a constant distance apart. Therefore, if the sequence of one strand is known, that of the other strand can be deduced. The strands are designated as plus (+) and minus (−) and an RNA molecule complementary to the minus (−) strand is synthesised during transcription. The base sequence may cause significant local variations in the shape of the DNA molecule and these variations are vital for specific interactions between the DNA and various proteins to take place. Although the three-dimensional structure of DNA may vary, it generally adopts a double-helical structure termed the B form or B-DNA in vivo.

DNA as a structure is known to undergo a number of chemical modifications without the underlying DNA sequence being altered. One of the most important alterations is the addition of a methyl (CH₃) group to cytosine, termed DNA methylation, which is catalysed by DNA methyltransferases. This results in what some describe as the fifth base in DNA. Approximately 1.5% of human DNA is methylated (termed the epigenome) and the methylation status seems to have a profound effect on gene expression, where hypomethylation appears to promote gene expression. This feature of gene expression control is termed epigenetics and is a complex process that is also extended to the modification of histone proteins and some small RNA molecules involved in gene expression control. Importantly, epigenetics appears to play a role in a number of disease states such as cancers and certain neurological diseases. This has opened up the potential for new means of treatments of those diseases.

The two anti-parallel strands of DNA are held together only by the weak forces of hydrogen bonding between complementary bases and partly by hydrophobic interactions between adjacent, stacked base pairs, termed base stacking. Little energy is needed to separate a few base pairs, hence at any instant a few short stretches of DNA will be opened up to the single-stranded conformation. However, such stretches immediately pair up again at room temperature, so the molecule as a whole remains predominantly double stranded.

If, however, a DNA solution is heated to approximately 90 °C or above, there will be sufficient kinetic energy to denature the DNA completely, causing it to separate into single strands. The temperature at which 50% of the DNA is melted is termed the melting temperature (T_m), and it depends on the nature of the DNA. If several different samples of DNA are melted, the T_m is highest for those DNAs molecules which contain the highest proportion of cytosine and guanine. The T_m can actually be used to estimate the percentage of (C + G) in a DNA sample. This relationship between T_m and (C + G) content arises because cytosine and guanine form three hydrogen bonds when base paired, whereas thymine and adenine form only two. Because of the different numbers of hydrogen bonds between A–T and C–G pairs, those sequences with a predominance of C–G pairs will require greater energy to separate or denature them. The conditions required to separate a particular nucleotide sequence is also dependent on environmental conditions such as salt concentration. If melted DNA is cooled, it is possible for the separated strands to reassociate, a process known as renaturation.

Strands of RNA and DNA will also associate with each other, if their sequences are complementary, to give double-stranded, hybrid molecules. Similarly, strands of labelled RNA or DNA, when added to a denatured DNA preparation, can act as probes for DNA molecules to which they are complementary. This hybridisation of complementary strands of nucleic acids is a cornerstone of many molecular biology techniques and is very useful for isolating a specific fragment of DNA from a complex mixture. It is also possible for small single-stranded fragments of DNA (up to 40 bases in length), termed oligonucleotides, to hybridise to a denatured sample of DNA. This type of hybridisation is termed annealing and again is dependent on the base sequence of the oligonucleotide and the salt concentration of the sample.

The use of DNA for analysis or manipulation usually requires that it is isolated and purified to a certain extent.²  DNA is recovered from cells by the gentlest possible method of cell rupture to prevent the DNA from fragmenting by mechanical shearing. This is usually in the presence of EDTA, which chelates the Mg²⁺ ions needed for enzymes that degrade DNA, termed DNase. Ideally, cell walls, if present, should be digested enzymatically (e.g. lysozyme treatment of bacterial cell walls) and the cell membrane should be solubilised using a detergent. If physical disruption is necessary, it is usually kept to a minimum and should involve cutting or squashing of cells, rather than the use of shear forces. Cell disruption (and most subsequent steps) should be performed at 4 °C, using glassware and solutions that have been autoclaved to destroy DNase activity (Figure 1.3).

After release of nucleic acids from the cells, RNA can be removed by treatment with ribonuclease (RNase) that has been heat treated to inactivate any DNase contaminants; RNase is relatively stable to heat as a result of its disulfide bonds, which ensure rapid renaturation of the molecule on cooling. The other major contaminant, protein, is removed by shaking the solution gently with water-saturated phenol or with a phenol–chloroform mixture, either of which will denature proteins but not nucleic acids. Centrifugation of the emulsion formed by this mixing produces a lower, organic phase, separated from the upper, aqueous phase by an interface of denatured protein. The aqueous solution is recovered and deproteinised repeatedly, until no more material is seen at the interface. Finally, the deproteinised DNA preparation is mixed with two volumes of absolute ethanol and the DNA is allowed to precipitate out of solution in a freezer. After centrifugation, the DNA pellet is redissolved in a buffer containing EDTA to inactivate any DNases present and stored at 4 °C. DNA solutions can be stored frozen although repeated freezing and thawing tend to damage long DNA molecules by shearing.

The procedure described is suitable for total cellular DNA. If the DNA from a specific organelle or viral particle is needed, it is best to isolate the organelle or virus before extracting its DNA, since the recovery of a particular type of DNA from a mixture is usually rather difficult. When a high degree of purity is required, DNA may be subjected to density gradient ultracentrifugation through caesium chloride, which is particularly useful for the preparation of plasmid DNA. It is possible to check the integrity of the DNA by agarose gel electrophoresis and determine the concentration of the DNA by using the fact that 1 absorbance unit equates to 50 µg mL⁻¹ of DNA, hence

Contaminants may also be identified in the sample by employing scanning UV spectrophotometry from 200 to 300 nm. A ratio of 260 nm : 280 nm of approximately 1.8 indicates that the sample is free of protein contamination, as protein absorbs strongly at 280 nm.

The methods used for RNA isolation are very similar to those described above for DNA. However, RNA molecules are relatively short and therefore less easily damaged by shearing, so cell disruption can be more vigorous. RNA is, however, vulnerable to digestion by RNases, which are present endogenously at various concentrations in certain cell types and exogenously on fingers. Gloves should therefore be worn and a strong detergent should be included in the isolation medium to denature any RNases immediately. Subsequent deproteinisation should be particularly rigorous, since RNA is often tightly associated with proteins. DNase treatment can be used to remove DNA and RNA can be precipitated by ethanol. A reagent that is commonly used in RNA extraction is guanidinium thiocyanate (GTC), which is both a strong inhibitor of RNase and a protein denaturant. It is possible to check the integrity of an RNA extract by analysing it by agarose gel electrophoresis. The most abundant RNA species are the rRNA molecules. For prokaryotes these are 16S and 23S and for eukaryotes the molecules are slightly heavier at 18S and 28S. These appear as discrete bands following agarose gel electrophoresis and importantly indicate that the other RNA components, such as mRNA, are likely to be intact. This is usually carried out under denaturing conditions to prevent secondary structure formation in the RNA. The concentration of the RNA may be estimated by using UV spectrophotometry. At 260 nm, 1 absorbance unit equates to 40 µg mL⁻¹ of RNA, hence

Contaminants may also be identified in the same way as for DNA by scanning UV spectrophotometry; however, in the case of RNA a 260 nm : 280 nm ratio of approximately 2 would be expected for a sample containing no protein.

In many cases, it is desirable to isolate eukaryotic mRNA, which constitutes only 2–5% of cellular RNA, from a mixture of total RNA molecules. This may be carried out by affinity chromatography on oligo(dT)–cellulose columns. At high salt concentrations, the mRNA containing poly(A) tails binds to the complementary oligo(dT) molecules of the affinity column, so mRNA will be retained; all other RNA molecules can be washed through the column with further high-salt solution. Finally, the bound mRNA can be eluted using a low concentration of salt. Nucleic acid species may also be sub-fractionated by more physical means such as electrophoretic or chromatographic separations based on differences in nucleic acid fragment sizes or physicochemical characteristics. Analysis of nucleic acid from single cells is also proving to be an important method in molecular analysis. This relies on the preparation of single cells from a mixed population of cells and tissues and can be achieved using a number of methods, one of the most common being laser capture microdissection (LCM). Here a few cells or single cells are typically prepared using a UV pulsed laser under automated control or by manual microscopy of the tissue.

Many current molecular biology methods and their reagents can now be found in the form of a kit from manufacturers such as QIAGEN, Sigma and ThermoFisher. In addition, there is an increasing move towards automation of manual processes of standard molecular biology methods. The extraction of nucleic acids by automated means is no exception. The advantage of their use lies in the fact that the reagents are standardised and quality control tested, providing a high degree of reliability. For example, glass bead preparations for DNA purification have been used increasingly and with reliable results. Small, compact, column-type preparations such as QIAGEN spin columns are also used extensively in research and in routine DNA extraction. Essentially the same reagents for nucleic acid extraction may be used in a format that allows reliable and automated extraction. The process can also be automated with a low-throughput QIAcube system. Further methods are also available using kit-based extraction methods for RNA; these in particular have overcome some of the problems of RNA extraction such as RNase contamination. A number of fully automated nucleic acid extraction machines, such as the QIAsymphony system, are now employed in areas where high throughput is required, e.g. in clinical diagnostic laboratories. Here the raw samples, such as blood specimens, are placed in 96- or 384-well microtitre plates and these follow a set computer-controlled processing pattern carried out robotically. In this way, the samples are rapidly manipulated and extracted in approximately 45 min without any manual operations being undertaken. Nucleic acids can be extracted from a variety of samples, including blood, serum, frozen tissue sections, formalin-fixed paraffin-embedded (FFPE) tissue sections and biopsies, among others, and all have their own unique challenges in the extraction process.

The discovery and characterisation of a number of key enzymes have allowed the development of various techniques for the analysis and manipulation of DNA. In particular, the enzymes termed type II restriction endonucleases have come to play a key role in all aspects of molecular biology.⁴  These enzymes recognise certain DNA sequences, usually 4–6 base pairs (bp) in length, and cleave them in a defined manner. The sequences recognised are palindromic or of an inverted repeat nature (Figure 1.4), that is, they read the same in both directions on each strand. When cleaved they leave a flush-ended or staggered (also termed a cohesive-ended) fragment, depending on the particular enzyme used. An important property of staggered ends is that those produced from different molecules by the same enzyme are complementary (or ‘sticky’) and so will anneal to each other. The annealed strands are held together only by hydrogen bonding between complementary bases on opposite strands. Covalent joining of the ends of each of the two strands may be carried out using the enzyme DNA ligase. This is widely exploited in molecular biology to enable the construction of recombinant DNA, i.e. the joining of DNA fragments from different sources. Approximately 500 restriction enzymes have been characterised that recognise over 100 different target sequences. A number of these, termed isoschizomers, recognise different target sequences but produce the same staggered ends or overhangs. A number of other enzymes have proved to be of value in the manipulation of DNA, as summarised in Table 1.1 and indicated at appropriate points in the text (Figure 1.5).

Electrophoresis in agarose or polyacrylamide gels is the standard method for separating DNA molecules according to size. The technique can be used analytically or preoperatively and can be qualitative or quantitative. Large fragments of DNA such as chromosomes may also be separated by a modification of electrophoresis termed pulsed field gel electrophoresis (PFGE). The easiest and most widely applicable method is electrophoresis in horizontal agarose gels, followed by staining with ethidium bromide. This dye binds to DNA by insertion between stacked base pairs (intercalation) and it exhibits a strong orange–red fluorescence when illuminated with ultraviolet light. Very often electrophoresis is used to check the purity and intactness of a DNA preparation or to assess the extent of a enzymatic reaction during for example the steps involved in the cloning of DNA (Figure 1.6). For such checks, ‘mini-gels’ are particularly convenient, since they need little preparation, use small samples and provide results quickly. Agarose gels can be used to separate molecules larger than about 100 bp. For higher resolution or for the effective separation of shorter DNA molecules, polyacrylamide gels are the preferred method.

When electrophoresis is used preparatively, the piece of gel containing the desired DNA fragment can be physically removed with a scalpel. The DNA may be recovered from the gel fragment in various ways. This may include crushing with a glass rod in a small volume of buffer, using agarase to digest the agarose, leaving the DNA, or by the process of electroelution. In this method, the piece of gel is sealed in a length of dialysis tubing containing buffer and is then placed between two electrodes in a tank containing more buffer. Passage of an electric current between the electrodes causes DNA to migrate out of the gel piece, but it remains trapped within the dialysis tubing and can therefore be recovered easily. There are now many kits that allow this process to be undertaken efficiently, such as the QIAquick gel extraction kit from QIAGEN.

Gel electrophoresis remains the established method for the separation and analysis of nucleic acids. However, a number of automated systems using pre-cast gels and standardised reagents are available that are now very popular. This approach is especially useful in situations where there are a large number of samples or high-throughput analysis is required. In addition, systems such as the Agilent Bioanalyzer have been developed that obviate the need to prepare electrophoretic gels. These employ microfluidic circuits constructed on small cassette units that contain interconnected microreservoirs. The sample is applied in one area and driven through microchannels under computer-controlled electrophoresis. The channels lead to reservoirs allowing, for example, incubation with other reagents such as dyes for a specified time. Electrophoretic separation is thus carried out in a microscale format. The small sample size minimises sample and reagent consumption and the units, being computer controlled, allow data to be captured within a very short timescale. In addition, dedicated spectrophotometers such as the ThermoFisher Scientific NanoDrop system can provide nucleic acid concentrations quickly with a limited sample volume. Alternative methods of analysis, including high-performance liquid chromatography-based approaches, have gained in popularity, especially for DNA mutation analysis. Mass spectrometric methods traditionally used in protein analysis, such as matrix-assisted laser desorption/ionisation (MALDI-TOF), are also becoming increasingly used for nucleic acid analysis owing to their rapidity and increasing reliability.

Single-cell DNA analysis has also been a focus of much research in the recent past. Techniques such as fluorescent-activated cell sorting (FACS), laser capture microdissection (LCM) and microfluidic analysis have greatly improved the capability for analysis of these cells. Many commercial methods have been employed recently, an example being a process developed by 10x Genomics. This has allowed an interesting and novel automated process termed 10x Chromium whereby nucleic acid from thousands of cells can be prepared, barcoded and analysed for single-cell gene expression profiling and characterisation by sequencing using Illumina HiSeq systems (see Section 1.8.6). New processes such as these allow effective methods for efficient further analysis of cells and tissues in both healthy and diseased states.

Electrophoresis of DNA restriction fragments allows separations based on size to be carried out, but it provides no indication regarding the presence of a specific, desired fragment among the complex sample. This can be achieved by transferring the DNA from the intact gel onto a piece of nitrocellulose or nylon membrane placed in contact with it. This provides a more permanent record of the sample since DNA begins to diffuse out of a gel that is left for a few hours. First the gel is soaked in alkali to render the DNA single stranded. It is then transferred to the membrane so that the DNA becomes bound to it in exactly the same pattern as that originally on the gel. This transfer, named a Southern blot after its inventor Edwin Southern, can be performed electrophoretically or by drawing large volumes of buffer through both gel and membrane, thus transferring DNA from one to the other by capillary action.⁵  The point of this operation is that the membrane can now be treated with a labelled DNA molecule, for example a gene probe. This single-stranded DNA probe will hybridise under the right conditions to complementary fragments immobilised on the membrane (Figure 1.7). The conditions of hybridisation, including the temperature and salt concentration, are critical for this process to take place effectively. This is usually referred to as the stringency of the hybridisation and it is particular for each individual gene probe and for each sample of DNA. A series of washing steps with buffer are then carried out to remove any unbound probe and the membrane is developed, after which the precise location of the probe and its target may be visualised. It is also possible to analyse DNA from different species or organisms by blotting the DNA and then using a gene probe representing a protein or enzyme from one of the organisms. In this way, it is possible to search for related genes in different species. This technique is generally termed zoo blotting.

The same basic process of nucleic acid blotting can be used to transfer RNA from gels onto similar membranes.⁶  This allows the identification of specific mRNA sequences of a defined length by hybridisation to a labelled gene probe and is known as northern blotting. With this technique, it is not only possible to detect specific mRNA molecules but also it may be used to quantify the relative amounts of the specific mRNA. It is usual to separate the mRNA transcripts by gel electrophoresis under denaturing conditions since this improves resolution and allows a more accurate estimation of the sizes of the transcripts. The format of the blotting may be altered from transfer from a gel to direct application to slots on a specific blotting apparatus containing the nylon membrane. This is termed slot or dot blotting and provides a convenient means of measuring the abundance of specific mRNA transcripts without the need for gel electrophoresis; it does not, however, provide information regarding the size of the fragments.

A further method of RNA analysis that overcomes the problems of RNA blotting is termed the ribonuclease protection assay (RPA). This is a solution-based method in which a probe that is complementary to the mRNA of interest is bound to form a hybrid that is resistant to digestion with RNase. Hence while other single-stranded RNA molecules are digested the intact hybrids can be further analysed by gel electrophoresis.

The availability of a gene probe is essential in many molecular biology techniques but in many cases it is one of the most difficult steps.⁷  In recent years, the emphasis has shifted from gene probe production to primers for the amplification of DNA by the polymerase chain reaction (PCR). Indeed, probes and primers are very similar, being single stranded with some primers being labelled with a fluorescent dye as gene probes are. The information needed to produce a gene probe or primer may come from many sources but with the development and sophistication of genetic databases this is usually one of the first stages. There are a number of databases throughout the world and it is possible to search these over the Internet and identify particular sequences relating to a specific gene or protein. In some cases, it is possible to use related proteins from the same gene family to acquire information on the most useful DNA sequence. Similar proteins or DNA sequences but from different species may also provide a starting point with which to produce a so-called heterologous gene probe. Although in some cases probes are already produced and cloned, it is possible, armed with a DNA sequence from a DNA database, to synthesise chemically a single-stranded oligonucleotide probe. Commercial providers with computer-controlled gene synthesisers are normally used. The process employs phosphoramidite synthesis, which links dNTPs together based on a desired sequence. It is essential to carry out certain checks before probe production to determine that the probe or primer is correct and unique, is not able to self-anneal or that it is self-complementary, all of which may compromise its use (Figure 1.8).

When little DNA information is available to prepare a gene probe, it is possible in some cases to use the knowledge gained from analysis of the corresponding protein. Thus it is possible to isolate and purify proteins and sequence part of the N-terminal end of the protein. From our knowledge of the genetic code, it is possible to predict the various DNA sequences that could code for the protein and then synthesise appropriate oligonucleotide sequences chemically. Owing to the degeneracy of the genetic code, most amino acids are coded for by more than one codon, hence there will be more than one possible nucleotide sequence that could code for a given polypeptide. The longer the polypeptide, the greater is the number of possible oligonucleotides that must be synthesised. Fortunately, there is no need to synthesise a sequence longer than about 20 bases, since this should hybridise efficiently with any complementary sequences and should be specific for one gene. Ideally, a section of the protein should be chosen that contains as many tryptophan and methionine residues as possible, since these have unique codons and there will therefore be fewer possible base sequences that could code for that part of the protein. The synthetic oligonucleotides can then be used as probes in a number of molecular biology methods. Indeed, the chemical solid-phase synthesis of DNA for probe or gene synthesis has improved over the years since it was first introduced and limits on length, construction and error correction or fidelity have also been developed. Recent advances have investigated enzymatic synthesis using terminal transferase (TdT) and new formats termed DNA printing. Refinement of these exciting methods may eventually bring DNA synthesis to benchtop computer-controlled machines.

An essential feature of a gene probe is that it can be visualised by some means. In this way, a gene probe that hybridises to a complementary sequence may be detected and identify that desired sequence from a complex mixture.⁷  There are two main ways of labelling gene probes. Fluorescent labelling is now a popular method for tagging nucleic acids and includes dyes such as fluorescein amidite (FAM) and digoxigenin-labelled nucleotides. Additionally, fluorescent dyes such as 4′,6-diamidino-2-phenylindole (DAPI), PicoGreen and RiboGreen are commonly used. Radioactive labelling with phosphorus-32 (³²P) or for certain techniques sulfur-35 (³⁵S) and tritium (³H) can also be used. These may be detected by the process of autoradiography in which the labelled probe molecule, bound to sample DNA, located for example on a nylon membrane, is placed in contact with an X-ray-sensitive film. Following exposure, the film is developed and fixed just as a black-and-white negative and reveals the precise location of the labelled probe and therefore the DNA to which it has hybridised.⁸ 

Non-radioactive fluorescent labels are increasingly being used to label DNA gene probes and now many have similar sensitivities, which, combined with their improved safety profile, has led to their greater acceptance.

The labelling systems are termed either direct or indirect. Direct labelling allows an enzyme reporter such as alkaline phosphatase to be coupled directly to the DNA. Although this may alter the characteristics of the DNA gene probe, it offers the advantage of rapid analysis since no intermediate steps are needed. However, indirect labelling is at present more popular. This relies on the incorporation of a nucleotide that has a label attached. Three of the main labels currently in use are biotin, fluorescein and digoxigenin. These molecules are covalently linked to nucleotides using a carbon spacer arm of 7, 14 or 21 atoms. Specific binding proteins may then be used as a bridge between the nucleotide and a reporter protein such as an enzyme. For example, biotin incorporated into a DNA fragment is recognised with very high affinity by the protein streptavidin. This may either be coupled or conjugated to a reporter enzyme molecule such as alkaline phosphatase. This is able to convert a colourless substrate, p-nitrophenol phosphate (PNPP), into a yellow compound, p-nitrophenol (PNP), and also offers a means of signal amplification. Alternatively, labels such as digoxigenin incorporated into DNA sequences may be detected by monoclonal antibodies, again conjugated to reporter molecules including alkaline phosphatase. Thus, rather than the detection system relying on autoradiography, which is necessary for radiolabels, a series of reactions resulting in either a colour, light or chemiluminescence reaction take place. This has important practical implications since autoradiography may take 1–3 days whereas colour and chemiluminescent reactions take minutes.

The simplest form of labelling DNA is by 5′ or 3′ end labelling. The 5′ end labelling involves a phosphate transfer or exchange reaction in which the 5′-phosphate of the DNA to be used as the probe is removed and in its place a labelled phosphate is added. This is usually carried out by using two enzymes. The first, alkaline phosphatase, is used to remove the existing phosphate group from the DNA. Following removal of the released phosphate from the DNA, a second enzyme, polynucleotide kinase, is added, which catalyses the transfer of a labelled phosphate group to the 5′ end of the DNA. The newly labelled probe is then purified, usually by chromatography through a Sephadex column, and may be used directly (Figure 1.9).

Using the other end of the DNA molecule, the 3′ end, is slightly less complex. Here a new dNTP which is labelled (e.g. [³²P]dATP or biotin-labelled dNTP) is added to the 3′ end of the DNA by the enzyme terminal transferase. Although this is a simpler reaction, a potential problem exists because a new nucleotide is added to the existing sequence and so the complete sequence of the DNA is altered, which may affect its hybridisation to its target sequence. End labelling methods also suffer from the fact that only one label is added to the DNA so they are of lower activity in comparison with methods that incorporate labels along the length of the DNA. Alternatively, fluorescent labels such as FAM may be used as an alternative to a radiolabel.

The DNA to be labelled is first denatured and then placed under renaturing conditions in the presence of a mixture of many different random sequences of hexamers or hexanucleotides. These hexamers will, by chance, bind to the DNA sample wherever they encounter a complementary sequence and so the DNA will rapidly acquire an approximately random sprinkling of hexanucleotides annealed to it. Each of the hexamers can act as a primer for the synthesis of a fresh strand of DNA catalysed by DNA polymerase since it has an exposed 3′-hydroxyl group. The Klenow fragment of DNA polymerase is used for random primer labelling because it lacks 5′ to 3′ exonuclease activity. This is prepared by cleavage of DNA polymerase with subtilisin, giving a large enzyme fragment that has no 5′ to 3′ exonuclease activity, but still acts as a 5′ to 3′ polymerase. Thus, when the Klenow enzyme is mixed with the annealed DNA sample in the presence of dNTPs, including at least one that is labelled, many short stretches of labelled DNA will be generated. In a similar way to random primer labelling, the PCR may also be used to incorporate radioactive or non-radioactive labels (Figure 1.10).

A traditional method of labelling DNA is by the process of nick translation to produce high-sensitivity fluorescent probes used in applications such as in situ hybridisation. Low concentrations of DNase I are used to make occasional single-strand nicks in the double-stranded DNA that is to be used as the gene probe. DNA polymerase then fills in the nicks, using an appropriate deoxyribonucleoside triphosphate (dNTP), at the same time making a new nick to the 3′ side of the previous one. In this way, the nick is translated along the DNA. If fluorescently labelled dNTPs such as fluorescein-12-dUTP and tetramethylrhodamine-5-dUTP are added to the reaction mixture, they will be used to fill in the nicks and so the DNA can be labelled to a very high specific activity.

There have been a number of key developments in molecular biology techniques but the one that has had the greatest impact in recent years has been the polymerase chain reaction (PCR). One of the reasons for the adoption of the PCR is the elegant simplicity of the reaction and the relative ease of the practical manipulation steps. Frequently this is one of the first techniques used when analysing DNA; it has opened up the analysis of cellular and molecular processes to those outside the field of molecular biology and biotechnology. In addition, many alternative methods to amplify nucleic acids have also been developed.⁹ 

The PCR is used to amplify a precise fragment of DNA from a complex mixture of starting material, usually termed the template, which may be DNA from microbes, mouth swabs, blood, urine, tissue biopsy, etc. It requires the knowledge of some DNA sequence information flanking the fragment of DNA to be amplified (target DNA). From this sequence information, two oligonucleotide primers are chemically synthesised, each complementary to a stretch of DNA to the 3′ side of the target DNA. Two oligonucleotides need to be synthesised, one for each of the two DNA strands. The result is an amplification of a specific DNA fragment that obviates the need for more time-consuming cloning procedures. In some respects, the PCR can be regarded as analogous to molecular cloning since it results in the generation of new DNA molecules based exactly upon the sequence of existing ones. The use of relevant bioinformatics resources for the design of oligonucleotide primers and for the determination of the required experimental conditions allows a rapid means for DNA amplification, analysis and identification.¹⁰ 

One problem with early PCR reactions was that the temperature needed to denature the DNA also denatured the DNA polymerase. However, the availability of a thermostable DNA polymerase enzyme isolated from the thermophilic bacterium Thermus aquaticus found in hot springs provided the means to automate the reaction. Taq DNA polymerase has a temperature optimum of 72 °C and survives prolonged exposure to temperatures as high as 96 °C and so is still active after each of the denaturation steps.

The PCR is often used to amplify a fragment of DNA from a complex mixture of starting material usually termed the template DNA. In many cases a relatively impure DNA source can be used successfully. However, in contrast to conventional cell-based cloning, PCR requires knowledge of the DNA sequences that flank the target DNA (Figure 1.11). For many applications PCR has replaced the traditional DNA cloning methods as it fulfils the same function, the production of large amounts of DNA from limited starting material; however, this is achieved in a fraction of the time needed to clone a DNA fragment. Although not without its drawbacks, the PCR is a remarkable technique that has changed the approach of many scientists to the analysis of nucleic acids and continues to have a profound impact on core genomic and genetic analysis.¹¹ 

The PCR consists of three well-defined times and temperatures termed steps: (i) denaturation at high temperature, (ii) annealing of primer and target DNA and (iii) extension in the presence of a thermostable DNA polymerase. A single round of denaturation, annealing and extension is termed a ‘cycle’. A typical PCR experiment consists of 30–40 cycles. In the first cycle, the double-stranded ‘high molecular weight’ template DNA is denatured by heating the reaction mixture to above 90 °C. Within the complex mass of DNA strands, the region to be specifically amplified (target) is thus made accessible to the primers. The temperature is then decreased to between 40 and 60 °C to allow the hybridisation of the two oligonucleotide primers, which are present in excess, to bind to their complementary sites that flank the target DNA. The annealed oligonucleotides act as primers for DNA synthesis, since they provide a free 3′-hydroxyl group for DNA polymerase. The DNA synthesis step is termed ‘extension’ and is carried out at 72 °C by a thermostable DNA polymerase, most commonly Taq DNA polymerase (Figure 1.12).

DNA synthesis proceeds from both of the primers until the new strands have been extended along and beyond the target DNA to be amplified. It is important to note that, since the new strands extend beyond the target DNA, they will contain a region near their 3′ ends that is complementary to the other primer (Figure 1.13). Therefore, if another round of DNA synthesis is allowed to take place, not only the original strands will be used as templates but also the new strands. The products obtained from the new strands will have a precise length, delimited exactly by the two regions complementary to the primers. As the system is taken through successive cycles of denaturation, annealing and extension, all of the new strands will act as templates and so there will be an exponential increase in the amount of DNA produced. The net effect is to amplify selectively the target DNA and the primer regions flanking it, leading to the production of millions of effectively identical copies. The original template strands are still present and a background of extension of these also takes place.

For efficient annealing of the primers, the precise temperature at which the annealing occurs is critical and each PCR system has to be defined and optimised by the user for each primer set and described in the primer design section. One useful practical technique for optimisation of the annealing temperature is called touchdown PCR, in which a specialised programmable thermocycler is used to decrease the temperature incrementally until the optimum annealing is reached. Reactions that are not optimised may give rise to other mis-primed DNA products in addition to the specific target or may not produce any amplified products at all.

A further approach to reducing spurious non-specific amplification during the early stages of the reaction is termed ‘hot start’ PCR. In this method, the reaction components are heated to the melting temperature before adding the polymerase. At one time, hot start was achieved by introducing a physical wax barrier between the Taq polymerase and the remainder of the reaction components. The wax is melted at the denaturation temperature, thus allowing the Taq polymerase access to the reaction mixture. Modified polymerase enzyme systems have also been developed that inhibit polymerisation at ambient temperature, either by the binding ligands such as antibodies or by the presence of bound inhibitors that dissociate only after a high-temperature activation step is performed. Taq polymerase has a number of limitations such as the lack of proof reading, resulting in misincorporation errors (some as high as 1 × 10⁻⁵ errors per base), and sensitivity to inhibitors such as those found in blood, e.g. haem, immunoglobulin M, or in extraction buffers (e.g. proteinase K, phenol). Unusually, it also leaves an A residue at the end of the amplicon, which can be useful for dA:dT cloning. Engineered versions of the enzyme that overcome some of these problems have been produced, in addition to the use of other enzymes such as Pfu (Pyrococcus furiosus) DNA polymerase that have proof reading ability, and combinations of enzymes can be especially useful when amplifying fragments of 5 kb or longer. Further cocktails of DNA polymerases can be used in the reaction to amplify PCR products of 20–30 kb. There have been a number of enhancements of the PCR, and typically they try to improve the accessibility of the polymerase to the template, using chemicals such as betaine, dimethyl sulfoxide (DMSO), T4 gene 32 protein and single-stranded binding protein (SSBP). Another interesting approach is nanoparticle PCR (nanoPCR), where adding gold nanoparticles at nanomolar concentrations can increase the sensitivity of the reaction by as much as a 1000-fold.¹²  This may be due to a number of factors, including the enhanced thermal transfer properties of the particles and/or assisting with primer–template binding. Indeed, the potential use of nanoparticles is changing the way in which PCR may be undertaken in the future, and with increasing use of carbon nanotubes and quantum dots the specificity and sensitivity may be improved.

The specificity of the PCR lies in the design of the two oligonucleotide primers. These not only have to be complementary to sequences flanking the target DNA but must not be self-complementary or bind each other to form dimers, since both prevent authentic DNA amplification. They also have to be matched in their GC content and to have similar annealing temperatures and must be incapable of amplifying unwanted genomic sequences. Manual design of primers can be time consuming and often haphazard, although equations such as the following are still used to derive the annealing temperature (T_a) for each primer:

where T_m is the melting temperature of the primer–target duplex and G, C, A and T are the numbers of the respective bases G, C, A and T in the primer. In general, the T_a is set 3–5 °C lower than the T_m. On occasions, secondary or primer dimer bands may be observed on the electrophoresis gel in addition to the authentic PCR product. In these situations, touchdown or hot start regimes may help. Alternatively, raising the T_a closer to the T_m can enhance the specificity of the reaction. The increasing use of bioinformatics resources, such as Bioinformatics Primer3Plus, NCBI Primer-Blast, NEB Tm Calculator and numerous biotechnology company-based programs, makes the design and selection of primers and reaction conditions much more straightforward. These computer-based resources allow the input of the target sequence, primer length, product size, GC content, etc., and following analysis provide a choice of matched primer sequences. Indeed, the initial selection and design of primers without the aid of bioinformatics would now be unnecessarily time consuming. Finally, before ordering or synthesising the primers, some groups submit proposed sequences to a nucleotide sequence search program such as BLAST, which can be used to interrogate GenBank or other comprehensive public DNA sequence databases to increase confidence that the reaction will be specific for the intended target sequence only.

It is possible to amplify more than one target sequence in a single reaction by including multiple sets of primer pairs in a process termed multiplex PCR. The main advantage of the technique is that it is not only cost-effective but also allows information to be obtained from limited amounts of template. In general, there are two types of multiplex PCR, one using a single template with a number of primer pairs and the other employing multiple templates with primer pairs. Multiplex PCR has been used effectively in many areas of nucleic acid analysis such as microbial identification, mutation analysis and single nucleotide polymorphism (SNP) genotyping. One main consideration of the technique is the design of the primers and, although the same basic rules are followed as for singleplex PCR, the primers need to be specific and not interfere or form dimers. The annealing temperature is also a critical feature with multiple primer sets. Bioinformatic assistance in the primer design process using software such as PrimerPlex is essential and is able to design and optimise primer sets while minimising potential mismatches to ensure high specificity and specific amplification take place.

RT-PCR is an extremely useful variation of the standard PCR that essentially allows the amplification of RNA molecules, such as mRNA transcripts via a cDNA reaction, from very limited sample amounts.¹⁴  This may be undertaken using a two-step process following RNA extraction or, in some cases, a single-step process in which both reactions are performed in a single tube. One advantage of the two-step process is the ability to archive the original sample for later analysis and the removal of aliquots when required. In some cases, the need for the rigorous extraction procedures associated with mRNA purification for conventional cloning purposes is obviated.¹³  For single-step PCR the dNTPs, buffer, Taq polymerase, oligonucleotide primers, reverse transcriptase (RT) and RNA template are added together to the reaction tube. The reaction mixture is heated to 37 °C, thus allowing the RT to work, and permits the production of a cDNA copy of the RNA strands that anneal to one of the primers in the mixture. Following ‘first strand synthesis’, a normal PCR is carried out to amplify the cDNA product, resulting in ‘second strand synthesis’, and subsequently a dsDNA product is amplified as usual. The choice of primer for the first strand synthesis depends on the experiment. The mRNAs in the cell extract may be converted to cDNA using an oligo(dT) primer that would anneal to all the poly(A) tails or a random hexamer providing the same cDNA. If a specific cDNA is sought, then a coding region specific primer can be used with success. The method is fast, accurate and simple to perform. It has many applications, such as the assessment of transcript levels in different cells and tissues when combined with quantitative PCR (qPCR). When combined with allele-specific primers, it also allows the amplification of cDNA from single chromosomes (Figure 1.14).

RT-PCR is widely used as a diagnostic tool in microbiology and virology. In particular, the diagnostic test for the SARS-CoV-2 coronavirus uses three primer and probe sets to detect nucleocapsid and RNase P, which can be multiplexed into one reaction. A further useful development in this area to try to quantitate rare alleles or viral nucleic acid copies is termed digital PCR (dPCR). Here the template is diluted or partitioned out to millions of single strands distributed in miniature chambers, emulsions and other surfaces. A PCR reaction is then performed such that a positive (1) or negative (0) will result. A determination of how many copies were in the original sample can then be assessed. Here, technological improvements have allowed the partitioning and amplification in titrated emulsions of oil, which lowers the cost in comparison with other quantitative methods such as qPCR. This approach of solid-phase PCR is also used in the template preparation for some of the next-generation sequencing methods.¹⁵ 

An important evolution of the PCR method is the development of quantitative PCR (qPCR). This method has been gaining popularity for many applications because of the rapidity of the method compared with conventional PCR amplification while simultaneously providing a lower limit of detection and greater dynamic range.¹⁶  It is possible to track the reaction in real time and the method allows assessments of any problems with the amplification process without the need for end-point analysis. Early qPCR methods involved the comparison of a standard or control DNA template amplified with separate primers at the same time as the specific target DNA. These types of quantification rely on the reaction being exponential and so any factors that affect the reaction may also affect the result. Other methods have involved the incorporation of a radiolabel through the primers or nucleotides and their subsequent detection following purification of the amplicon. In its simplest form, a DNA binding dye such as SYBR Green is included in the reaction. As amplicons accumulate, SYBR Green binds the dsDNA proportionally. Fluorescence emission of the dye is detected following excitation. The binding of SYBR Green is non-specific but most qPCR thermal cycler machines will produce a melt curve, which allows a degree of identity of the amplicon through its melting temperature profile. Indeed, a refinement of this method, termed precision melt analysis, is able to differentiate between amplicons with one base difference and has value in mutation analysis. Subsequent DNA sequencing can provide a definitive confirmation. In order to detect specific amplicons during the PCR, an oligonucleotide probe labelled with a fluorescent reporter and quencher molecule at either end can be included in the reaction in place of SYBR Green. This method is termed the 5′ fluorogenic exonuclease detection system or more commonly the TaqMan approach. When the oligonucleotide probe (TaqMan probe) binds to the target sequence, the 5′ exonuclease activity of Taq polymerase degrades and releases the reporter from the quencher during extension. A signal is thus generated, which increases in direct proportion to the number of starting molecules.¹⁷  Hence the detection system is able to induce and detect fluorescence in real time as the PCR proceeds. Importantly, it provides confirmation that the correct DNA sequence has been amplified (Figure 1.15).

Refinements of the PCR process are also under development and promise rapid amplification and reporting. This is mainly due to advances in miniaturising the amplification system and using microfluidics on a small chip. Efficiency, multiplex throughput and low reagent costs are advantages of these systems and similar methods are regularly employed in analysing nucleic acid and protein samples using technology such as the Agilent Bioanalyzer. Currently there is a move to provide a set of guidelines or information on the development and reporting of qPCR termed MIQE Guidelines or Minimum Information for Publication of Quantitative Real-time PCR Experiments.¹⁷  There are numerous applications for the PCR, as indicated in Table 1.2.

The PCR has remained with the same basic format for a number of years, but there have been many incremental advances on the original method. This has largely been in the standardisation and quality control of reagents and many are available as pre-mixed aliquots in what is termed a ‘master mix’; this minimises pipetting errors and is important since small volumes are usually used. The PCR also uses automated thermal cyclers that heat and cool the reagents located in plastic tubes with thin walls for efficient thermal transfer or in 96-well microtitre plates. There are numerous makes and models of different configurations on the market that employ various methods for heating and cooling; the most common use Peltier effect heating blocks and water or fan cooling. High-end thermal cyclers may also include a robotic sample preparation system and many can be programmed directly on the cycler or through a connected PC.

Many traditional methods of analysis in molecular biology have been superseded by the PCR and the applications of the technique appear to be almost unlimited.⁹  The success of the PCR process has given impetus to the development of other amplification techniques where the PCR has limitations. New amplification methods are based on either thermal cycling or non-thermal cycling (isothermal) methods. Indeed, the development of isothermal systems, such as the LAMP (loop mediated isothermal amplification) DNA amplification system obviates the need for a thermal cycler as it operates at a constant temperature of 60–65 °C. This is a highly efficient method for amplifying DNA in which 4–6 primers bind to a target of 6–8 regions in the DNA. A strand displacing DNA polymerase is employed and dumbbell-like products are produced that can be amplified exponentially. The inclusion of a fluorescent dye such as SYBR Green allows a relatively straightforward method of detection. Reverse transcriptase LAMP (RT-LAMP) also combines a prior cDNA synthesis step allowing mRNA detection and analysis. This has been a useful method in the detection of certain viruses such as Ebola, Zika and SARS-CoV-2. A potential disadvantage of the method is the two or three sets of primers that need to be designed and synthesised in contrast to the two primers used in the PCR. However, the simplicity of the setup and the fact that expensive thermal cyclers or electrophoresis are not required make this a useful technique for diagnostics and analysis in non-laboratory remote settings.

The determination of the linear order or sequence of nucleotide bases along a length of DNA is one of the central techniques in molecular biology and has played a key role in numerous genome mapping and sequencing projects. Two basic techniques were developed in the 1970s for efficient DNA sequencing, one based on an enzymatic method, frequently termed Sanger sequencing after its developer, and a chemical method termed Maxam and Gilbert sequencing, named for the same reason.^18,19  For large-scale DNA analysis Sanger sequencing and its variants are by far the most effective methods and many commercial kits are available. However, there are certain applications, such as the sequencing of short oligonucleotides, where the Maxam and Gilbert method might be more appropriate.

One absolute requirement for Sanger sequencing is that the DNA to be sequenced is in a single-stranded form. Traditionally, this demanded that the DNA fragment of interest be inserted and cloned into a specialised bacteriophage vector such as M13, which is naturally single stranded. Although M13 is still a viable option, the advent of the PCR has provided a means not only to amplify a region of any genome or cDNA for which primer sequences are available, but also very quickly to generate the corresponding nucleotide sequence. This has led to an explosion in not only the accumulation of DNA sequences but also the development of many other methods based on the technology.

The Sanger method is simple and elegant and in many ways mimics the natural ability of DNA polymerase to extend a growing nucleotide chain based on an existing template.¹⁸  Initially, the DNA to be sequenced is allowed to hybridise with an oligonucleotide primer, which is complementary to a sequence adjacent to the 3′ side of DNA within a vector such as M13 (or within an amplicon in the case of the PCR). The oligonucleotide will then act as a primer for synthesis of a second strand of DNA, catalysed by DNA polymerase. Since the new strand is synthesised from its 5′ end, virtually the first DNA to be made will be complementary to the DNA to be sequenced. One of the dNTPs required for DNA synthesis is labelled (or all four in some methods) with a fluorescent molecule or less commonly with ³⁵S and so the newly synthesised strand will be labelled. Alternatively, the primer may be labelled to provide the means of detection following the completion of the method.

The reaction mixture is then divided into four aliquots, representing the four dNTPs: A, C, G and T. Using the adenine (A) tube as an example, in addition to all of the dNTPs being present in the mixture, an analogue of dATP is added [2′,3′-dideoxyadenosine triphosphate (ddATP)], which is similar to A except that it has no 3′-hydroxyl group. Since a 5′ to 3′ phosphodiester linkage cannot be formed without a 3′-hydroxyl group, the presence of the ddATP will terminate the growing chain. The situation for tube C is identical except that ddCTP is added; similarly, the G and T tubes contain ddGTP and ddTTP, respectively (Figure 1.16).

Since the incorporation of a ddNTP rather than a dNTP is a random event, the reaction will produce new molecules varying widely in length, but all terminating at the same type of base. Thus four sets of DNA sequence are generated, each terminating at a different type of base, but all having a common 5′ end (the primer). The four labelled and chain-terminated samples are then denatured by heating and loaded next to each other on a polyacrylamide gel for electrophoresis. Electrophoresis is performed at approximately 70 °C in the presence of urea, to prevent renaturation of the DNA, since even partial renaturation alters the rates of migration of DNA fragments. In the original format, very thin, long electrophoresis gels were used for maximum resolution over a wide range of fragment lengths. After electrophoresis, the positions of labelled DNA bands on the gel can be determined by autoradiography. Since every band in the lane from the ddATP sample must contain molecules that terminate at adenine and those in the ddCTP terminate at cytosine, etc., it is possible to read the sequence of the newly synthesised strand from the autoradiogram, provided that the gel can resolve differences in length equal to a single nucleotide. Under ideal conditions, sequences up to about 300 bases in length can be read from one gel (Figure 1.17).

It is also possible to undertake nucleotide sequencing from double-stranded molecules such as plasmid cloning vectors and PCR amplicons directly. The double-stranded DNA must be denatured prior to annealing with primer. In the case of plasmids, an alkaline denaturation step is sufficient. However, for PCR amplicons this is more problematic. Unlike plasmids, amplicons are short and reanneal rapidly. Denaturants such as formamide or dimethyl sulfoxide (DMSO) have been used to prevent the reannealing of PCR strands following their separation. Another strategy is to bias the amplification towards one strand by using one primer in excess, which also overcomes this problem to a certain extent.

It is possible to separate and retain one PCR product strand physically by incorporating a molecule such as biotin in one of the primers. Following PCR, the strand that contains the biotinylated primer may be removed by affinity chromatography with streptavidin-coated magnetic beads, leaving the complementary PCR strand. This magnetic affinity purification provides single-stranded DNA derived from the PCR amplicon and, although somewhat time consuming, it does provide high-quality single-stranded DNA for sequencing.

Advances in fluorescent labelling chemistry led to the development of high-throughput automated sequencing techniques.^20,21  Most systems involve the use of dNTPs labelled with different fluorochromes (often referred to as dye terminators). The advantage of this modification is that since a different label is incorporated with each ddNTP, it is unnecessary to perform four separate reactions. Therefore, the four chain-terminated products are run on the same track of a denaturing electrophoresis gel. Each product with a base specific dye, such as fluorescein or tetramethylrhodamine, is excited by a laser and the dye then emits light at its characteristic wavelength. A diffraction grating separates the emissions and these are detected by a charge-coupled device (CCD) and the sequence is interpreted by a program on a dedicated PC. The advantages of the techniques include real-time detection of the sequence and read lengths in excess of 500 bp.

Capillary electrophoresis is increasingly being used for the detection of sequencing products. Here liquid polymers in thin capillary tubes are used, obviating the need to pour sequencing gels and requiring little manual operation. This substantially reduces the electrophoresis run times and allows higher throughputs to be achieved. A number of large-scale sequence facilities employing this type of sequencing facility are now fully automated, allowing the rapid acquisition of sequence data. Automated sequencing for genome projects is usually based on cycle sequencing using instruments such as the ABI PRISM 3700 DNA Analyzer. This can be formatted to produce simultaneous reads in 384-well cycle sequencing reaction plates. The derived nucleotide sequences are downloaded automatically to databases and manipulated using a variety of bioinformatics resources (Figure 1.18).

Further advances in sequencing methods and developments in scalability and automation led to what is termed next-generation sequencing (NGS).²²  This collection of methods, based on different platforms, offers rapid, high-throughput analysis with the use of a scaled-up approach that can determine the sequence of large numbers of different DNA strands at one time. Currently, the main NGS systems involve template preparation with some form of enrichment by PCR, usually followed by a ‘sequencing by synthesis’ approach. Various systems or platforms have been developed to undertake these processes. They include Illumina sequencing, Roche 454 sequencing and ABI SOLiD sequencing, Ion Torrent: Proton/PGM and Pacific Biosciences' real-time sequencing system, among others.²³ 

NGS has numerous applications in many fields, including whole-genome sequencing (WGS), RNA sequencing, chromatin immunoprecipitation sequencing (ChIP-seq), exome sequencing and epigenomics.²⁴  One general feature of these platforms is that they perform the methods in parallel, termed ‘massively parallel sequencing’, so that vast amounts of sequence data are produced simultaneously. Many of the methods also allow each fragment being sequenced to be uniquely tagged with a short identifying barcode in the initial sample preparation stage. The management, processing and analysis of enormous amounts of data in the reconstruction of large fragments of DNA derived from shorter reads also require powerful computer hardware processing. At present, the setup and capital costs for NGS systems are high in comparison with automated Sanger sequencing, although this will decrease as the uptake of NGS increases. With the huge increase in data that NGS and massively parallel sequencing bring, there is also a requirement for efficient bioinformatics resources and large amounts of data storage.

The Illumina method is similar to Sanger sequencing and is an example of a sequencing by synthesis method. Initially, genomic DNA is fragmented into shorter fragments and adapters are ligated to each end of the DNA. The DNA fragments are then immobilised on a slide or flow cell containing a lawn of primers²⁵ . The DNA fragments hybridise at one end of the immobilised primers and then bend over and bind to a complementary primer, also bound on the slide surface. The bridge formed is amplified by PCR (bridge PCR) and produces DNA colonies or clusters in each channel of the flow cell. The sequencing part of the technology involves the use of reversible dye terminators (Figure 1.19). Following denaturation, one strand is removed and sequencing primers, polymerase and nucleotides are added. Each base is tagged with a different fluorophore and a 3′ block and hence, as a complementary base is flowed across the cell, it is incorporated and a laser is used to excite the fluorescence, thus identifying the base. This is recorded in an image and occurs in parallel at each cluster. Following this, the fluorophore is cleaved, the block is removed and the second cycle begins. All the information is processed by a sophisticated computer program to build up the sequence in each of the clusters.

The 454 platform utilises the pyrosequencing format and involves direct analysis of DNA fragments, and this system allows the rapid sequencing of entire genomes by the ‘shotgun’ approach. First, genomic DNA is randomly sheared and ligated to linker sequences that permit individual molecules captured on the surface of a bead to be amplified while isolated within an emulsion droplet by emulsion PCR (emPCR). Each bead is designed to contain one DNA fragment. A very large collection of such beads is arrayed in the 1.6 million wells of a fibre-optic slide.

Each PCR template is hybridised to an oligonucleotide and incubated with DNA polymerase, ATP sulfurylase, luciferase and apyrase. During the reaction, the first of the four dNTPs is added and, if incorporated, releases pyrophosphate (PPi), hence the name ‘pyrosequencing’. The ATP sulfurylase converts the PPi to ATP, which drives the luciferase-mediated conversion of luciferin to oxyluciferin to generate light. Apyrase degrades the resulting component dNTPs and ATP. This is followed by another round of dNTP addition. A resulting pyrogram provides an output of the sequence. The method provides short reads very quickly and is especially useful for the determination of mutations or SNPs. Current technology of 454 sequencing allows one gigabase or a third of the human genome to be derived in a single day in comparison with Sanger sequencing which would take over a year to complete. A comparison of NGS methods is provided in Table 1.3.

An alternative NGS platform developed by Applied Biosystems is the SOLiD (sequencing by oligonucleotide ligation and detection) system. This process is similar to 454 technology in that it uses emPCR to generate clonal amplification of single-stranded DNA fragments (ligated to adapters) attached to beads. The beads are then attached to a glass at a high density.

A universal sequencing primer is hybridised to the adapter sequence on the beads containing the template. Ligase, together with a set of four fluorescently labelled dibase probes, are added to the reaction and compete for ligation to the sequencing primer. Specificity of the dibase probe is achieved by interrogating every first and second base in each ligation reaction. Hybridisation, ligation, fluorescence detection and cleavage are performed for a predetermined number of cycles, after which the extension product is removed and the template is reset with a primer offset by one base for a second round of ligation cycles. A comparison of selected DNA sequencing methods is provided in Table 1.3.

A different approach that has recently been employed is the observation of DNA single-molecule sequencing in real time. This unique process is used by Pacific Biosciences with their single-molecule real-time (SMRT) platform.²⁵  The method utilises phospho-linked nucleotides where a fluorescent dye is attached through the phosphate group in the molecule, each of the four nucleotides having a different fluorescent dye. Thus, during incorporation of the nucleotide into a growing chain by a DNA polymerase, the phosphate is cleaved while a detector identifies which of the four nucleotides has been incorporated. The technology is based around a laser light-illuminated nanophotonic containment cylinder termed a zero-mode waveguide within which the reaction takes place. Remarkably, these cylinders are smaller than the wavelength of light and a sophisticated CCD detects flashes of light when a nucleotide is incorporated and records this as a movie. There have been a number of initiatives for new DNA sequencing methods and an interesting one uses tiny semiconductors or biological nanopores through which DNA can be passed and a sequence deduced.²⁶  The nanopore is a transmembrane protein such as α-haemolysin spanning a lipid bilayer over which a voltage is passed. As the DNA fragment being analysed moves through the pore, the different bases alter the current, each base having a characteristic alteration. Devices such as those produced by Oxford Nanopore are ultraportable and thus very useful for field studies where laboratory analysis is difficult.

One firmly established technology used in molecular biology is the application of microarrays or DNA microchips.²⁷  These provide a radically different approach to current laboratory molecular biology research strategies in that large-scale analysis and quantification of genes and gene expression are possible simultaneously. Specific applications of microarrays include human and plant genotyping, methylation analysis and genome-wide association studies (GWAS). A microarray consists of an ordered arrangement of potentially hundreds of thousands of DNA sequences such as oligonucleotides or cDNAs deposited on a solid surface. The solid support may be either glass or silicon and currently the arrays are synthesised on or off the chip. They require complex fabrication methods similar to that used in producing computer microchips. Most commercial productions employ robotic ultrafine microarray deposition instruments that dispense volumes in the picolitre range. Alternatively, on-chip fabrication as used by the biotechnology company Affymetrix builds up layers of nucleotides using a process borrowed from the computer industry termed photolithography. Here wafer-thin masks with holes allow photoactivation of specific dNTPs that are linked together at specific regions on the chip. The whole process allows layers of oligonucleotides to be built up with each nucleotide at each position being defined by computer.

The arrays themselves may represent a variety of nucleic acid material. This may be mRNA produced in a particular cell type, termed cDNA expression arrays, or may alternatively represent coding and regulatory regions of a particular gene or group of genes.²⁸  A number of arrays are now available that may determine mutations in DNA, detection of mRNA transcript levels or other polymorphisms such as SNPs. The sample DNA is placed on the array and any unhybridised DNA is washed off. The array is then analysed and scanned for patterns of hybridisation by detection of fluorescence signals. Any mutations or genetic polymorphisms in relevant genes may be rapidly analysed by computer interpretation of the resulting hybridisation pattern and mutation, transcript level or polymorphism defined. Indeed, the collation and manipulation of data from microarrays present as big a problem as fabricating the chips in the first place.

Microarrays have been developed for the detection of various genetic mutations including the cystic fibrosis transmembrane regulator (CFTR) gene and the breast cancer gene BRCA1 and in the study of the human immunodeficiency virus (HIV).

At present, microarrays require DNA to be highly purified, which can limit their applicability. However, as DNA purification becomes automated and microarray technology develops, it is not difficult to envisage numerous laboratory tests on a single DNA microchip. This could be used for analysing not only single genes but also large numbers of genes or DNA representing microorganisms, viruses, etc. Since the potential for quantitation of gene transcription exists, expression arrays could also be used in defining a particular disease status. This technique may be very significant since it will allow large amounts of sequence information to be gathered very rapidly and assist in many fields of molecular biology, especially in large genome sequencing projects or in so-called resequencing projects where gene regions such as those containing potentially important polymorphisms require analysis in a number of samples.

One current application of microarray technology is the generation of a catalogue of SNPs across various genomes, including the human genome.²⁹  Estimates indicate that there are approximately 85 million SNPs and importantly some may point to the development of certain diseases. For example, an SNP in the F5 gene is implicated in a thrombophilia, factor V Leiden, a genetic blood clotting disorder. SNP analysis is therefore clearly a candidate for microarray analysis and genome-wide SNP arrays permit the simultaneous analysis of nearly one million SNPs on one gene chip. In order to simplify the problem of the vast numbers of SNPs that need to be analysed, the HapMap Project currently analyses SNPs that are inherited as a block and generally as few as 500 000 SNPs are required to genotype an individual.³⁰ 

The techniques developed and employed in the manipulation of nucleic acids are the cornerstone of the analysis of cells and tissues and interactions at the molecular level. There has been an explosion in recent years towards the scale-up and automation of many techniques, highlighted by high-throughput DNA sequencing. It took approximately 13 years and $3 billion to sequence the first human genome.²⁵  The current cost for the sequencing of a genome is in the hundreds of dollars. Indeed, there has been a huge expansion in various genome projects in various species and in many different countries. Large-scale initiatives such as the UK Biobank, Genome Asia 100K and various specialist disease projects such as the 100K genome project are proving vital in research and discovery. This has been due in part not only to the technology to undertake this but also to the parallel improvements that have taken place in computing and processing power and importantly the developments in bioinformatics data manipulation and analysis.