Contents

This edition first published 2012
© 2012 by John Wiley & Sons, Ltd.

Wiley-Blackwell is an imprint of John Wiley & Sons, formed by the merger of Wiley's global Scientific, Technical and Medical business with Blackwell Publishing.

Registered office
John Wiley & Sons, Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK

Editorial offices
9600 Garsington Road, Oxford, OX4 2DQ, UK
The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK
111 River Street, Hoboken, NJ 07030-5774, USA

For details of our global editorial offices, for customer services and for information about how to apply for permission to reuse the copyright material in this book please see our website at .

The right of the author to be identified as the author of this work has been asserted in accordance with the UK Copyright, Designs and Patents Act 1988.

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by the UK Copyright, Designs and Patents Act 1988, without the prior permission of the publisher.

Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The publisher is not associated with any product or vendor mentioned in this book. This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold on the understanding that the publisher is not engaged in rendering professional services. If professional advice or other expert assistance is required, the services of a competent professional should be sought.

Library of Congress Cataloging-in-Publication Data

Dale, Jeremy, Professor.
From genes to genomes : concepts and applications of DNA technology / Jeremy W. Dale, Malcolm von Schantz, and Nick Plant. – 3rd ed.
p. ; cm.
Includes bibliographical references and index.
ISBN 978-0-470-68386-6 (cloth) – ISBN 978-0-470-68385-9 (pbk.)
I. Schantz, Malcolm von. II. Plant, Nick. III. Title.
[DNLM: 1. Genetic Engineering. 2. Cloning, Molecular. 3. DNA, Recombinant. QU 450]
LC classification not assigned
660.6′5–dc23

2011030219

A catalogue record for this book is available from the British Library.

This book is published in the following electronic formats: ePDF 9781119953159; ePub 9781119954279; Mobi 9781119954286

Preface

The first edition of this book was published in 2002. By the time of the second edition (2007) the emphasis had moved away from just cloning genes, to embrace a wider range of technologies, especially genome sequencing, the polymerase chain reaction and microarray technology. The revolution has continued unabated, indeed even accelerating, not least with the advent of high-throughput genome sequencing. In this edition we have tried to introduce readers to the excitement engendered by the latest developments – but this poses a considerable challenge. Our aim has been to keep the book to an accessible size, so including newer technologies inevitably means discarding some of the older ones. Some might maintain that we could have gone further in that direction. Some methods that have been kept are no longer as important as they once were, and maybe there is an element of sentimentality in keeping them – but there is some virtue in retaining a balance so that we can maintain a degree of historical perspective. There is a need to understand, to some extent, how we got to the position we are now in, as well as trying to see where we are going.

The main title of the book, From Genes to Genomes, is derived from the progress of this revolution. It also indicates a recurrent theme within the book, in that the earlier chapters deal with analysis and investigation at the level of individual genes, and then later on we move towards genome-wide studies – ending up with a chapter directed at the whole organism.

Dealing only with the techniques, without the applications, would be rather dry. Some of the applications are obvious – recombinant product formation, genetic diagnosis, transgenic plants and animals, and so on – and we have attempted to introduce these to give you a flavour of the advances that continue to be made, but at the same time without burdening you with excessive detail. Equally important, possibly more so, are the contributions made to the advance of fundamental knowledge in areas such as developmental studies and molecular phylogeny.

The purpose of this book is to provide an introduction to the concepts and applications of this rapidly moving and fascinating field. In writing it, we had in mind its usefulness for undergraduate students in the biological and biomedical sciences (who we assume will have a basic grounding in molecular biology). However, it will also be relevant for many others, ranging from research workers and teachers who want to update their knowledge of related areas to anyone who would like to understand rather more of the background to current controversies about the applications of some of these techniques.

Jeremy W. Dale
Malcolm von Schantz
Nick Plant

1

From Genes to Genomes

1.1 Introduction

The classical approach to genetics starts with the identification of variants that have a specific phenotype, i.e., they differ from the wildtype in some way that can be seen (or detected in other ways) and defined. For Gregor Mendel, the father of modern genetics, this was the appearance of his peas (e.g., green versus yellow, or round versus wrinkled). One of the postulates he arrived at was that these characteristics assorted independently of one another. For example, when crossing one type of pea that produces yellow, wrinkled peas with another that produces green, round peas, the first generation (F1) are all round and yellow (because round is dominant over wrinkled, and yellow is dominant over green). In the second (F2) generation, there is a 3 : 1 mixture of round versus wrinkled peas, and independently a 3 : 1 mixture of yellow to green peas.

Of course Mendel did not know why this happened. We now know that if two genes are located on different chromosomes, which will segregate independently during meiosis, the genes will be distributed independently amongst the progeny. Independent assortment can also happen if the two genes are on the same chromosome, but only if they are so far apart that any recombination between the homologous chromosomes will be sufficient to reassort them independently. However, if they are quite close together, recombination is less likely, and they will therefore tend to remain associated during meiosis. They will therefore be inherited together. We refer to genes that do not segregate independently as linked; the closer they are, the greater the degree of linkage, i.e., the more likely they are to stay together during meiosis. Measuring the degree of linkage (linkage analysis) is a central tool in classical genetics, in that it provides a way of mapping genes, i.e., determining their relative position on the chromosome.

Bacteria and yeasts provide much more convenient systems for genetic analysis, because they grow quickly, as unicellular organisms, on defined media. You can therefore use chemical or physical mutagens (such as ultraviolet irradiation) to produce a wide range of mutations, and can select specific mutations from very large pools of organisms – remembering that an overnight culture of Escherichia coli will contain some 109 bacteria per millilitre. So we can use genetic techniques to investigate detailed aspects of the physiology of such cells, including identifying the relevant genes by mapping the position of the mutations.

For multicellular organisms, the range of phenotypes is even greater, as there are then questions concerning the development of different parts of the organism, and how each individual part influences the development of others. However, animals have much longer generation times than bacteria, and using millions of animals (especially mammals) to identify the mutations you are interested in is logistically impossible, and ethically indefensible. Human genetics is even more difficult as you cannot use selected breeding to map genes; you have to rely on the analysis of real families, who have chosen to breed with no consideration for the needs of science. Nevertheless, classical genetics has contributed extensively to the study of developmental processes, notably in the fruit fly Drosophila melanogaster, where it is possible to study quite large numbers of animals, due to their relative ease of housing and short generation times, and to use mutagenic agents to enhance the rate of variation.

However, these methods suffered from a number of limitations. In particular, they could only be applied, in general, to mutations that gave rise to a phenotype that could be defined in some way, including shape, physiology, biochemical properties or behaviour. Furthermore, there was no easy way of characterizing the nature of the mutation. The situation changed radically in the 1970s with the development of techniques that enabled DNA to be cut precisely into specific fragments, and to be joined together, enzymatically –techniques that became known variously as genetic manipulation, genetic modification, genetic engineering or recombinant DNA technology. The term ‘gene cloning’ is also used, since joining a fragment of DNA with a vector such as a plasmid that can replicate in bacterial cells enabled the production of a bacterial strain (a clone) in which all the cells contained a copy of this specific piece of DNA. For the first time, it was possible to isolate and study specific genes. Since such techniques could be applied equally to human genes, the impact on human genetics was particularly marked.

The revolution also depended on the development of a variety of other molecular techniques. The earliest of these (actually predating gene cloning) was hybridization, which enabled the identification of specific DNA sequences on the basis of their sequence similarity. Later on came methods for determining the sequence of these DNA fragments, and the polymerase chain reaction (PCR), which provided a powerful way of amplifying specific DNA sequences. Combining those advances with automation, plus the concurrent advance in computer power, led to the determination of the full genome sequence of many organisms, including the human genome, and thence to enormous advances in understanding the roles of genes and their products. In recent years, sequencing technology has advanced to a stage where it is now a routine matter to sequence the full genome of many individuals, and thus attempt to pinpoint the causes of the differences between them, including some genetic diseases.

Furthermore, since these techniques enabled the cloning and expression of genes from any one organism (including humans) into a more amenable host, such as a bacterium, they allowed the use of genetically modified bacteria (or other hosts) for the production of human gene products, such as hormones, for therapeutic use. This principle was subsequently extended to the genetic modification of plants and animals – both by inserting foreign genes and by knocking out existing ones – to produce plants and animals with novel properties.

As is well known, the construction and use of genetically modified organisms (GMOs) is not without controversy. In the early days, there was a lot of concern that the introduction of foreign DNA into E. coli would generate bacteria with dangerous properties. Fortunately, this is one fear that has been shown to be unfounded. Due to careful design, genetically modified bacteria are, generally, not well able to cope with life outside the laboratory, and hence any GM bacterium released into the environment (deliberately or accidentally) is unlikely to survive for long. In addition, one must recognize that nature is quite capable of producing pathogenic organisms without our assistance – which history, unfortunately, has repeatedly demonstrated through disease outbreaks.

The debate on GMOs has now largely moved on to issues relating to genetically modified plants and animals. It is important to distinguish the genetic modification of plants and animals from cloning of plants and animals. The latter simply involves the production of genetically identical individuals; it does not involve any genetic modification whatsoever. (The two technologies can be used in tandem, but that is another matter.) There are ethical issues to be considered, but cloning plants and animals is not the subject of this book.

Currently, the debate on genetic modification can be envisaged as largely revolving around two factors: food safety and environmental impact. The first thing to be clear about is that there is no imaginable reason why genetic modification, per se, should make a foodstuff hazardous in any way. There is no reason to suppose that cheese made with rennet from a genetically modified bacterium is any more dangerous than similar cheese made with ‘natural’ rennet. It is possible to imagine a risk associated with some genetically modified foodstuffs, due to unintended stimulation of the production of natural toxins – remembering, for example, that potatoes are related to deadly nightshade. But this can happen equally well (or perhaps is even more likely) with conventional cross-breeding procedures for developing new strains, which are not always subject to the same degree of rigorous safety testing as GM plants.

The potential environmental impact is more difficult to assess. The main issue here is the use of genetic modification to make plants resistant to herbicides or to insect attack. When such plants are grown on a large scale, it is difficult to be certain that the gene in question will not spread to related wild plants in the vicinity (although measures can be taken to reduce this possibility), or the knock-on effect that such resistance may have on the ecosystem – if all the insects are killed, what will small birds and animals eat? But these concerns may be exaggerated. As with the bacterial example above, these genes will not spread significantly unless there is an evolutionary pressure favouring them. So we would not expect widespread resistance to weedkillers unless the plants are being sprayed with those weedkillers. There might be an advantage in becoming resistant to insect attack, but the insects concerned have been around for a long time, so the wild plants have had plenty of time to develop natural resistance anyway. In addition, targeted resistance in a group of plants may arguably have less environmental impact than the less targeted spraying of insecticides. We have to balance the use of genetically modified plants against the use of chemicals. If genetic modification of the plants means a reduction in the use of environmentally damaging chemicals, then that is a tangible benefit that could outweigh any theoretical risk.

The purpose of this book is to provide an introduction to the exciting developments that have resulted in an explosion of our knowledge of the genetics and molecular biology of all forms of life, from viruses and bacteria to plants and mammals, including of course ourselves – developments that continue as we write. We hope that it will convey some of the wonder and intellectual stimulation that this science brings to its practitioners.

1.2 Basic molecular biology

In this book, we assume you already have a working knowledge of the basic concepts of molecular and cellular biology. This section serves as a reminder of the key aspects that are especially relevant to this book.

1.2.1 The DNA backbone

Manipulation of nucleic acids in the laboratory is based on their physical and chemical properties, which in turn are reflected in their biological function. Intrinsically, DNA is a remarkably stable molecule. Indeed, DNA of sufficiently high quality to be analysed has been recovered from frozen mammoths thousands of years old. This stability is provided by the robust phosphate–sugar backbone in each DNA strand, in which the phosphate links the 5′ position of one sugar to the 3′ position of the next (). The bonds between these phosphorus, oxygen and carbon atoms are all covalent bonds, meaning they are strong interactions that require significant energy to break. Hence, the controlled degradation of DNA requires enzymes (nucleases) that catalyse the breaking of these covalent bonds. These enzymes are divided into endonucleases, which attack internal sites in a DNA strand, and exonucleases, which nibble away at the ends. (We can for the moment ignore other enzymes that attack, for example, the bonds linking the bases to the sugar residues.) Some of these enzymes are non-specific, and lead to a generalized destruction of DNA. It was the discovery of restriction endonucleases (or restriction enzymes), which cut DNA strands at specific positions, coupled with DNA ligases, which can join two double-stranded DNA molecules together, that opened up the possibility of recombinant DNA technology (‘genetic engineering’).

RNA molecules, which contain the sugar ribose (), rather than the deoxyribose found in DNA, are less stable than DNA, often surviving only minutes within the cell. They show greater susceptibility to attack by nucleases (ribonucleases), and are also more susceptible to chemical degradation, especially by alkaline conditions.

1.2.2 The base pairs

In addition to the sugar (2′-deoxyribose) and phosphate, DNA molecules contain four nitrogen-containing bases (): two pyrimidines, thymine (T) and cytosine (C), and two purines, guanine (G) and adenine (A). It should be noted that other bases can be incorporated into synthetic DNA in the laboratory, and sometimes others occur naturally, but T, C, G and A are the major DNA bases. Because the purines are bigger than the pyrimidines, a regular double helix requires a purine in one strand to be matched by a pyrimidine in the other. Furthermore, the regularity of the double helix requires specific hydrogen bonding between the bases so that they fit together, with an A opposite a T, and a G opposite a C (). We refer to these pairs of bases as complementary, and hence to each strand as the complement of the other.

Note that the two DNA strands run in opposite directions. In a conventional representation of a double-stranded sequence the ‘top’ strand has a 5′ hydroxyl group at the left-hand end (and is said to be written in the 5′ to 3′ direction), while the ‘bottom’ strand has its 5′ end at the right-hand end. Because the two strands are complementary, there is no information in one strand that cannot be deduced from the other one. Therefore, to save space, the convention is to represent a double-stranded DNA sequence by showing the sequence of only one strand, in the 5′ to 3′ direction. The sequence of the second strand is inferred from that, and you must remember that the second strand runs in the opposite direction. Thus a single strand sequence written as AGGCTG (or more fully 5′AGGCTG3′) would have as its complement CAGCCT (5′CAGCCT3′) (see Box 1.1).

Thanks to this base-pairing arrangement, the two strands can be separated intact – both in the cell and in the test tube – under conditions that, although disrupting the relatively weak hydrogen bonds that exist between the bases on complementary strands, are much too mild to pose any threat to the covalent bonds that join nucleotides within a single strand. Such a separation is referred to as denaturation of DNA, and unlike the denaturation of many proteins it is reversible. Because of the complementarity of the base pairs, the strands will easily join together again and renature by reforming their hydrogen bonds. In the test tube, DNA is readily denatured by heating, and the denaturation process is therefore often referred to as melting, even when it is accomplished by means other than heat (e.g. by NaOH). Denaturation of a double-stranded DNA molecule occurs over a short, specific temperature range, and the midpoint of that range is defined as the melting temperature (Tm). This is influenced by the base composition of the DNA. Since guanine:cytosine (GC) base pairs have three hydrogen bonds, they are stronger (i.e., melt less easily) than adenine:thymine (AT) pairs, which have only two. It is therefore possible to estimate the melting temperature of a DNA fragment if you know the sequence. These considerations are important in understanding the technique known as hybridization, in which gene probes are used to detect specific nucleic acid sequences. We will look at hybridization in more detail in Chapter 3.

In addition to the hydrogen bonds, the double-stranded DNA structure is maintained by hydrophobic interactions between the bases. The hydrophobic nature of the bases means that a single-stranded structure, in which the bases are exposed to the aqueous environment, is unstable. Pairing of the bases enables them to be removed from interaction with the surrounding water, introducing significant stability to the DNA helix. In contrast to hydrogen bonding, hydrophobic interactions are relatively non-specific. Thus, nucleic acid strands will tend to stick together even in the absence of specific base-pairing, although the specific interactions make the association stronger. The specificity of the interaction can therefore be increased by the use of chemicals (such as formamide) that reduce the hydrophobic interactions.

What happens if there is only a single nucleic acid strand? This is normally the case with RNA, but single-stranded forms of DNA (ssDNA) also exist, in some viruses, for example. A single-stranded nucleic acid molecule will tend to fold up on itself to form localized double-stranded regions, producing structures referred to as hairpins or stem-loop structures (). This has the effect of removing the bases from interaction with the surrounding water, again increasing stability. At room temperature, in the absence of denaturing agents, a single-stranded nucleic acid will normally consist of a complex set of such localized secondary structure elements; this is especially evident with RNA molecules.

A further factor is the negative charge on the phosphate groups in the nucleic acid backbone. This works in the opposite direction to the hydrogen bonds and hydrophobic interactions; the strong negative charge on the DNA strands causes electrostatic repulsion that tends to make the two strands repel each other. In the presence of salt, this effect is counteracted by a cloud of counterions surrounding the molecule, neutralizing the negative charge on the phosphate groups. However, if you reduce the salt concentration, any weak interactions between the strands will be disrupted by electrostatic repulsion. Hence, at low ionic strength, the strands will only remain together if the hydrogen bonding is strong enough, and therefore we can use low salt conditions to increase the specificity of hybridization (see Chapter 3). Of course, within the cell the salt concentration is such that the double-stranded DNA is quite stable.

1.2.3 RNA structure

Chemically, RNA is very similar to DNA. The fundamental chemical difference is that whereas DNA contains 2′-deoxyribose (i.e., ribose without the hydroxyl group at the 2′ position) in its backbone, the RNA backbone contains ribose (). This slight difference has a powerful effect on some properties of the RNA molecule, especially on its stability. For example, RNA is destroyed under alkaline conditions while DNA is stable. Although the DNA strands will separate, they will remain intact and capable of renaturation when the pH is lowered again. However, under such conditions, RNA will quickly be destroyed. A further difference between RNA and DNA is that the former contains uracil rather than thymine ().

Generally, while most of the DNA we encounter is double-stranded, most of the RNA we meet consists of a single polynucleotide strand. However, DNA can also exist as a single-stranded molecule, and RNA is able to form double-stranded molecules. Thus, this distinction between RNA and DNA is not an inherent property of the nucleic acids themselves, but is a reflection of the natural roles of RNA and DNA in the cell, and of the method of production. In all cellular organisms (i.e., excluding viruses), DNA is the inherited material responsible for the genetic composition of the cell, and the replication process that has evolved is based on a double-stranded molecule. By contrast, the roles of RNA in the cell do not require a second strand, and indeed the presence of a second, complementary, strand would preclude its role in protein synthesis. However, there are some viruses that have double-stranded RNA (dsRNA) as their genetic material, as well as some viruses with single-stranded RNA. In addition, some viruses (as well as some plasmids) replicate via single-stranded DNA forms. Double-stranded RNA is also important in the phenomenon known as RNA interference, which we will come to in later chapters.

1.2.4 Nucleic acid synthesis

We do not need to consider here all the details of how nucleic acids are synthesized. The fundamental features that we need to remember are summarized in , which shows the addition of a nucleotide to the growing end (3′-OH) of a DNA strand. The substrate for this reaction is the relevant deoxynucleotide triphosphate (dNTP), i.e., the one that makes the correct base-pair with the corresponding residue on the template strand. The DNA strand is always extended at the 3′-OH end, thus the nucleotide strand grows in the 5′ to 3′ direction. For this reaction to occur it is essential that the existing residue at the 3′-OH end, to which the new nucleotide is to be added, is accurately base-paired with its partner on the other strand.

RNA synthesis occurs in much the same way, as far as this simplistic description goes, except that of course the substrates are nucleotide triphosphates (NTPs) rather than the deoxynucleotide triphosphates (dNTPs). There is one very important difference though. DNA synthesis only occurs by extension of an existing strand – it always needs a primer to get it started. In contrast, RNA polymerases are capable of starting a new RNA strand, complementary to its template, from scratch, given the appropriate signals.

1.2.5 Coiling and supercoiling

DNA can be denatured and renatured, deformed and reformed, and still retain unaltered function. This is a necessary feature, because such a large molecule as DNA will need to be packaged if it is to fit within the cell that it controls. The DNA of a human chromosome, if it were stretched into an unpackaged double helix, would be several centimetres long. Thus, cells are dependent on the packaging of DNA into modified configurations for their very existence.

Double-stranded DNA, in its relaxed state, normally exists as a right-handed double helix with one complete turn per 10 base pairs; this is known as the B form of DNA. Hydrophobic interactions between consecutive bases on the same strand contribute to this winding of the helix, as the bases are brought closer together enabling a more effective exclusion of water from interaction with the hydrophobic bases.

The DNA double helix can exist in other forms, notably the A form (also right-handed but more compact, with 11 bases per turn) and Z-DNA, which is a left-handed double helix with a more irregular appearance (a zigzag structure, hence its designation). However, that is not the complete story. Higher orders of conformation are known to exist. The double helix is in turn coiled on itself – an effect known as supercoiling. There is an interaction between the coiling of the helix and the degree of supercoiling. As long as the ends are fixed, changing the degree of coiling will alter the amount of supercoiling, and vice versa. DNA in vivo is constrained; the ends are not free to rotate. This is most obviously true of circular DNA structures such as (most) bacterial plasmids, but the rotation of linear molecules (other than very short oligonucleotides) is also constrained within the cell. The net effect of coiling and supercoiling (a property known as the linking number) is therefore fixed, and cannot be changed without breaking one of the DNA strands. In nature, there are enzymes known as topoisomerases (including DNA gyrase in bacteria) that do just that: they break the DNA strands, and then in effect rotate the ends and reseal them. This alters the degree of winding of the helix, and thus affects the supercoiling of the DNA. Topoisomerases also have an ingenious use in the laboratory, which we will consider in Chapter 2.

The plasmids that we will be referring to frequently in later pages are naturally supercoiled when they are isolated from the cell. However, if one of the strands is broken at any point, the DNA is then free to rotate at that point and can therefore relax into a non-supercoiled form. This is known as an open circular form (in contrast to the covalently closed circular form of the native plasmid).

1.3 What is a gene?

The definition of a ‘gene’ is rather imprecise. Its origins go back to the early days of genetics, when it was used to describe the unit of inheritance of a phenotype. This meaning persists in non-scientific usage, rather loosely, as the ‘gene for blue eyes’, or ‘the gene for red hair’. As it became realized that many characteristics were determined by the presence or properties of individual proteins, the definition became refined to relate to the chromosomal region that carried the information for that protein, leading to the concept of ‘one gene, one protein’. As the study of genetics and biochemistry progressed further, it was realized that many proteins consist of several distinct polypeptides, and that the chromosomal regions coding for the different polypeptides could be distinguished genetically. So the definition was refined further to mean a piece of DNA containing the information for a single specific polypeptide (‘one gene, one polypeptide’). With the advent of DNA sequencing, it became possible to consider a gene in molecular terms. So we now often use the term ‘gene’ as being synonymous with ‘open reading frame’ (ORF), i.e., the region between the start and stop codons. In bacteria, this is (usually) a simple uninterrupted sequence, but in eukaryotes, the presence of introns (see below) makes this definition more difficult, since the region of the chromosome that contains the information for a specific polypeptide may be many times longer than the actual coding sequence. We also have to be careful as we may want to refer to the whole transcribed region, which will be longer than the translated open reading frame, or indeed we may want to include the control regions that are necessary for the start of transcription.

Furthermore, this definition, by focusing solely on the regions that code for proteins (or polypeptides), is too limited in its scope. It ignores many regions of DNA that, although not coding for proteins, are nevertheless important for the viability of the cell, or influence the phenotype in other ways. The most obvious of these are DNA sequences that are templates for so-called non-coding RNA molecules. The most well known of these are ribosomal and transfer RNA, although we will encounter other RNA molecules that play significant roles in gene regulation and other activities. Other DNA regions are important in gene regulation because they act as binding sites for regulatory proteins.

In organisms with small genomes, such as bacteria, a high proportion of the genome is accounted for by these coding and regulatory regions (together with elements that may be described as ‘parasitic’ such as integrated viruses and insertion sequences). There is relatively little DNA to which we can ascribe no likely function, compared to eukaryotic cells, and especially animal and plant cells with much larger genomes, where there is a much higher proportion of non-coding DNA. But we should not be too hasty in writing these sequences off as ‘junk’. Increasingly, much of it is recognized as having important functions within the cell. These include enabling the DNA to be folded correctly and in ensuring that the coding regions are available for expression under the appropriate conditions, as well as coding for small (non-translated) RNA molecules that play a major role in modulating gene expression.

Thus, we have to accept that it is not possible to produce an entirely satisfactory definition of the word ‘gene’. However, this is rarely a serious problem. We just have to be careful how we use it depending on whether we are discussing only the coding region (ORF), or the length of sequence that is transcribed into mRNA (including untranslated regions), or whether we wish to include DNA regions with regulatory functions as well as coding sequences. In this context, we will also encounter the words allele and locus. A locus is used in the same way as ‘gene’ in the broad meaning – i.e., it could be a coding sequence, a regulatory region, or any other region we wish to consider. An allele is one version of that locus. So variation of a genetic characteristic between individuals would be due to different alleles at one locus (or several loci).

1.4 Information flow: gene expression

The way in which genes are expressed is so central to the subsequent material in this book that it is worth reviewing briefly the salient features. The basic dogma () is that while DNA is the fundamental genetic material (ignoring RNA viruses) that carries information from one generation to the next, its effect on the characteristics of the cell requires firstly its copying into RNA (transcription), and then the translation of the mRNA into a polypeptide by ribosomes. Further processes are required before its proper activity can be manifested: these include the folding of the polypeptide, possibly in association with other subunits to form a multisubunit protein, and in some cases modification, for example by glycosylation or phosphorylation. It should be remembered that in some cases, RNA rather than protein is the final product of a gene (e.g., ribosomal and transfer RNA molecules).

1.4.1 Transcription

Transcription is carried out by RNA polymerase. RNA polymerase recognizes and binds to a specific sequence (the promoter), and initiates the synthesis of mRNA from an adjacent position.

A typical bacterial promoter carries two consensus sequences (i.e., sequences that are closely related in all genes): TTGACA centred at position −35 (i.e., 35 bases before the transcription start site), and TATAAT at −10 (). It is important to understand the nature of a consensus: few bacterial promoters have exactly the sequences shown, but if you line up a large number of promoters you will see that at any one position a large number of them have the same base (Box 1.2). The RNA polymerase has higher affinity for some promoters than others – depending not only on the exact nature of the two consensus sequences but also, to a lesser extent, on the sequence of a longer region of DNA. The nature and regulation of bacterial promoters, including the existence of alternative types of promoters, is considered further in Chapter 5.

In eukaryotes, by contrast, the promoter is a considerably larger area around the transcription start site, where a number of trans-acting transcription factors (i.e., DNA-binding proteins encoded by genes in other parts of the genome) bind to various cis-acting elements (i.e., elements that affect the expression of the gene next to them) in a considerably more complex scenario. The need for this added complexity can easily be imagined; if cells carrying the same genome are differentiated into a multitude of cell types fulfilling very different functions, a very sophisticated control system is needed to provide each cell type with its specific repertoire of proteins, and to fine-tune the degree of expression for each one of them. Nonetheless, the promoter region, however simple or complex, gives rise to different levels of transcription of various genes. A further complexity in eukaryotes is that there are commonly additional regulatory elements known as enhancers, which may further alter the level of transcription of one, or several, genes. By definition, enhancers are position and orientation independent, and are often remote from the actual start site of transcription by several thousand base pairs.

Eukaryotes have three different RNA polymerases. Only one of these, RNA polymerase II, is involved in the transcription of protein-coding transcripts, plus the transcription of a group of small non-coding RNAs called micro-RNAs, which we will encounter in Chapter 11. RNA polymerase I is responsible for the synthesis of large ribosomal RNAs, whilst RNA polymerase III makes small RNAs such as transfer RNA (tRNA) and 5S ribosomal RNA.

In eukaryotes, the primary transcript from a protein-coding gene, produced by RNA polymerase II, is called a heterogeneous or heteronuclear RNA (hnRNA). It is very short-lived as such, being rapidly processed in a number of steps called maturation. A specialized nucleotide cap is added to the 5′ end; this is the site recognized by the ribosomes in protein synthesis (see below). The precursor mRNA is cleaved at a specific site towards the 3′ end and a poly-A tail, consisting of a long sequence of adenosine residues, is added to the cut end. This is a specific process, governed by polyadenylation recognition sequences in the 3′ untranslated region. Nature's ‘tagging’ of mRNA molecules comes in very useful in the laboratory for the isolation of eukaryotic mRNA (see Chapter 3). This transcript contains intervening sequences (introns) between the exons that carry the coding information (see below). The final step is the process of splicing, by which the introns are removed and the exons are joined together. This process is quite complex; some introns are removed by specific proteins known as splicing factors, whereas other introns are removed independently through autocatalysis. To complicate things further, in some cases the transcript is edited, leading, for example, to the introduction of a tissue-specific earlier stop codon that is not encoded in the corresponding DNA.

In bacteria, the processes of transcription and translation take place in the same compartment and simultaneously. In eukaryotes, by contrast, the mature mRNA molecule is transported out of the nucleus to the cytoplasm, where translation takes place.

The resulting level of protein production is dependent on the amount of the specific mRNA available, rather than just the rate of production. The level of an mRNA species will be affected by its rate of degradation as well as by its rate of synthesis. In bacteria, most mRNA molecules are degraded quite quickly (with a half-life of only a few minutes), although some are much more stable. The instability of the majority of bacterial mRNA molecules means that bacteria can rapidly alter their profile of gene expression by changing the transcription of specific genes. By contrast, the lifespans of most eukaryotic mRNA molecules are measured in hours rather than minutes, although some short-lived mRNA molecules do exist. This greater persistence of eukaryote mRNA molecules is, again, a reflection of the fact that an organism (or a cell) that is able to control its own environment to a substantial extent is subjected to less radical environmental changes. Consequently, mRNA molecules tend to be more stable in multicellular organisms than in, for example, yeast. Nonetheless, the principle remains – the level of an mRNA in a cell is a function of its production and degradation rates. We will discuss how to study and disentangle these parameters in Chapter 10.

1.4.2 Translation

In bacteria, translation starts when ribosomes bind to a specific site (the ribosome binding site, RBS), which is adjacent to the start codon. The sequence of the ribosome binding site (also known as the Shine–Dalgarno sequence) has been recognized as being complementary to the 3′ end of the 16S rRNA (). The precise sequence of this site and its distance from the start codon affect the efficiency of translation, although in nature this is less important than transcriptional efficiency in determining the level of gene expression. Translation efficiency will also depend on the codon usage, i.e., the match between synonymous codons and the availability of tRNA that will recognize each codon. This concept is explored more fully in Chapter 7.

In bacterial systems, where transcription and translation occur in the same compartment of the cell, ribosomes will bind to the mRNA, a process known as initiation, as soon as the RBS has been synthesized. Thus, there will be a procession of ribosomes following close behind the RNA polymerase, translating the mRNA in the process of elongation as and when it is being produced. So, although the mRNA may be very short-lived, the bacteria are capable of producing substantial amounts of the corresponding polypeptide in a short time. Translation stops when a termination codon is reached.

In eukaryotes, the mechanism is much more complicated. Instead of binding just upstream of the initiation codon, the ribosome binds to the cap at the 5′ end of the mRNA, and reads along the 5′ untranslated region (UTR) until it reaches an initiation codon. The sequence AUG may be encountered on the way without initiation, because the surrounding sequence is also important to define the start of protein synthesis. The fact that the 5′ UTR is scanned in its full length by the ribosome makes it an important region for specifying translation efficiency, and different secondary structures can have either a positive or a negative effect on the amount of protein that is produced. Sometimes, translation may be initiated at an internal ribosome entry site (IRES); the best studied of these occur in some viruses, but they also occur in some transcripts encoded naturally by the cell.

1.5 Gene structure and organisation

1.5.1 Operons

In bacteria, it is quite common for a group of genes to be transcribed from a single promoter into one long RNA molecule; this group of genes is known as an operon (). If we are considering protein-coding genes, the transcription product, messenger RNA (mRNA), is then translated into a number of separate polypeptides. This can occur by the ribosomes reaching the stop codon at the end of one polypeptide-coding sequence, terminating translation and releasing the product before reinitiating (without dissociation from the mRNA). Alternatively, the ribosomes may attach independently to internal ribosome binding sites within the mRNA sequence. Generally, the genes involved are responsible for different steps in the same pathway, and this arrangement facilitates the coordinate regulation of those genes, i.e., expression of each gene in the operon goes up (or down) together in response to changing conditions.

In eukaryotes, by contrast, the way in which ribosomes initiate translation is different, which means that they cannot usually produce separate proteins from a single mRNA in this way. Although there are some examples where polygenic transcripts analogous to bacterial operons are produced, these are very much the exception rather than the rule. Generally, when a single mRNA gives rise to different proteins, this is due to alternative processing of the mRNA (see below) or by producing one long polyprotein or precursor, which is then cleaved into different proteins (as occurs in some viruses). However, in eukaryotes, there is one type of RNA that is usually produced in a polycistronic fashion, namely microRNA (miRNA). We will consider this in more detail in Chapter 11.

1.5.2 Exons and introns

In bacteria there is generally a simple one-for-one relationship between the coding sequence of the DNA, the transcribed mRNA and the translated protein. This is usually not true for eukaryotic cells, where the initial transcript is many times longer than that needed for translation into the final protein. This pre-mRNA contains blocks of sequence (introns) that are removed by processing to generate the final mature mRNA for translation ().

Introns do occur in bacteria, but quite infrequently. This is partly due to the need for economy in a bacterial cell. The smaller genome and generally more rapid growth create an evolutionary pressure to remove unnecessary material from the genome. A further factor arises from the nature of transcription and translation in a bacterial cell. Since the ribosomes translate the mRNA while it is being made, there is less opportunity for sections of the RNA to be removed before translation.

1.6 Refinements of the model

The simple model described earlier in this chapter is that the inherited properties of a cell are determined by the sequence of bases in the DNA. These properties are manifested by transcription of the DNA into RNA, which (in eukaryotes) is processed by removal of introns to produce a messenger RNA that is then translated into protein. The levels of the protein are regulated by control of the frequency and rate of transcription and/or translation, as well as by the stability of the RNA and protein molecules.

We now know that several aspects of this model are inadequate. Firstly, there can be differences between cells that have identical DNA sequences, and these differences can be passed on from one cell to its progeny. This phenomenon is known as epigenetic inheritance. The clearest example of how this can happen is due to methylation of specific bases in the DNA. The extent and position of this methylation can be passed on from one cell to its progeny, hence it is inherited, but you will not see that difference by conventional DNA sequencing. Later on (in Chapter 10) we will consider ways in which this can be investigated, together with other differences such as variation in the number of copies of regions of the DNA (copy number variation).

Secondly, we now know that much of the DNA that does not code for proteins is not actually junk at all. It plays an important role in the regulation of the activity of the cell through the action of small RNA molecules. This includes the process of RNA interference, which is the action of miRNAs to alter mRNA stability and translational efficiency, with a net effect of reducing protein production. RNA interference probably developed to protect the cell from infection by RNA viruses, which replicate via a double-stranded RNA intermediate. It may also be exploited as a way of silencing gene expression in the laboratory, one that demands much less resources and is much more amenable to high-throughput screening than other methods (see Chapters 8 and 11).

Thirdly, sequencing of eukaryotic genomes (see Chapter 8), especially those of humans and other mammals, showed that the genome appeared to code for far fewer proteins than expected. Human DNA is thought to contain only about 21 000 distinct protein-coding genes, which can be compared to a bacterium such as Mycobacterium tuberculosis, with over 4000 such genes. However, the number of potential protein species in a human cell is much larger than 21 000. The main reason for this is that the splicing mechanism, by which introns are removed to produce the mRNA, is capable of variation, so that in some cells two exons are adjacent in the final product while in others they are separated by an intron that has not been removed. This alternative splicing, which can result in quite different proteins arising from a single protein-coding region, is a significant cause of the physiological variation between cells in different tissues, which have identical DNA structures.