*kbp = kilo base pairs, i.e. thousands of nucleotide monomers.

Amazingly, genes only seem to comprise about 2% of the DNA in a cell. The majority of the DNA does not form genes and doesn’t seem to do anything. The purpose of this junk DNA remains a mystery!

RNA is a nucleic acid like DNA, but with 4 differences:

• RNA has the sugar ribose instead of deoxyribose
• RNA has the base uracil instead of thymine
• RNA is usually single stranded
• RNA is usually shorter than DNA

mRNA carries the "message" that codes for a particular protein from the nucleus (where the DNA master copy is) to the cytoplasm (where proteins are synthesised). It is single stranded and just long enough to contain one gene only. It has a short lifetime and is degraded soon after it is used.

rRNA, together with proteins, form ribosomes, which are the site of mRNA translation and protein synthesis. Ribosomes have two subunits, small and large, and are assembled in the nucleolus of the nucleus and exported into the cytoplasm.

tRNA is an "adapter" that matches amino acids to their codon. tRNA is only about 80 nucleotides long, and it folds up by complementary base pairing to form a looped clover-leaf structure. At one end of the molecule there is always the base sequence ACC, where the amino acid binds. On the middle loop there is a triplet nucleotide sequence called the anticodon. There are 64 different tRNA molecules, each with a different anticodon sequence complementary to the 64 different codons. The amino acids are attached to their tRNA molecule by specific enzymes. These are highly specific, so that each amino acid is attached to a tRNA adapter with the appropriate anticodon.

The sequence of bases on DNA codes for the sequence of amino acids in proteins. But there are 20 different amino acids and only 4 different bases, so the bases are read in groups of 3. This gives 4³ or 64 combinations, more than enough to code for 20 amino acids. A group of three bases coding for an amino acid is called a codon, and the meaning of each of the 64 codons is called the genetic code.

There are several interesting points from this code:

• The code is degenerate, i.e. there is often more than one codon for an amino acid. The degeneracy is on the third base of the codon, which is therefore less important than the others.
• One codon means "start" i.e. the start of the gene sequence. It is AUG.
• Three codons mean "stop" i.e. the end of the gene sequence. They do not code for amino acids.
• The code is only read in one direction along the mRNA molecule.

DNA is copied, or replicated, before every cell division, so that one identical copy can go to each daughter cell. The method of DNA replication is obvious from its structure: the double helix unzips and two new strands are built up by complementary base-pairing onto the two old strands.

1. Replication starts at a specific sequence on the DNA molecule called the replication origin.
2. An enzyme unwinds and unzips DNA, breaking the hydrogen bonds that join the base pairs, and forming two separate strands.
3. The new DNA is built up from the four nucleotides (A, C, G and T) that are abundant in the nucleoplasm.
4. These nucleotides attach themselves to the bases on the old strands by complementary base pairing. Where there is a T base, only an A nucleotide will bind, and so on.
5. The enzyme DNA polymerase joins the new nucleotides to each other by strong covalent bonds, forming the sugar-phosphate backbone.
6. A winding enzyme winds the new strands up to form double helices.
7. The two new molecules are identical to the old molecule.

DNA replication can takes a few hours, and in fact this limits the speed of cell division. One reason bacteria can reproduce so fast is that they have a relatively small amount of DNA.

This replication mechanism is sometimes called semi-conservative replication, because each new DNA molecule contains one new strand and one old strand. This need not be the case, and alternative theories suggested that a "photocopy" of the original DNA could be made, leaving the original DNA conserved (conservative replication. The evidence for the semi-conservative method came from an elegant experiment performed in 1958 by Meselson and Stahl. They used the bacterium E. coli together with the technique of density gradient centrifugation, which separates molecules on the basis of their density.

DNA never leaves the nucleus, but proteins are synthesised in the cytoplasm, so a copy of each gene is made to carry the "message" from the nucleus to the cytoplasm. This copy is mRNA, and the process of copying is called transcription.

1. The start of each gene on DNA is marked by a special sequence of bases.
2. The RNA molecule is built up from the four ribose nucleotides (A, C, G and U) in the nucleoplasm. The nucleotides attach themselves to the bases on the DNA by complementary base pairing, just as in DNA replication. However, only one strand of RNA is made. The DNA stand that is copied is called the template or sense strand because it contains the sequence of bases that codes for a protein. The other strand is just a complementary copy, and is called the non-template or antisense strand.
3. The new nucleotides are joined to each other by strong covalent bonds by the enzyme RNA polymerase.
4. Only about 8 base pairs remain attached at a time, since the mRNA molecule peels off from the DNA as it is made. A winding enzyme rewinds the DNA.
5. The initial mRNA, or primary transcript, contains many regions that are not needed as part of the protein code. These are called introns (for interruption sequences), while the parts that are needed are called exons (for expressed sequences). All eukaryotic genes have introns, and they are usually longer than the exons.
6. The introns are cut out and the exons are spliced together by enzymes
7. The result is a shorter mature RNA containing only exons. The introns are broken down.
8. The mRNA diffuses out of the nucleus through a nuclear pore into the cytoplasm.

A single piece of mRNA can be translated by many ribosomes simultaneously, so many protein molecules can be made from one mRNA molecule. A group of ribosomes all attached to one piece of mRNA is called a polysome.

In eukaryotes, proteins often need to be altered before they become fully functional. Modifications are carried out by other enzymes and include: chain cutting, adding methyl or phosphate groups to amino acids, or adding sugars (to make glycoproteins) or lipids (to make lipoporteins).

Mutations are changes in genes, which are passed on to daughter cells. DNA is a very stable molecule, and it doesn't suddenly change without reason, but bases can change when DNA is being replicated. Normally replication is extremely accurate but very occasionally mistakes do occur (such as a T-C base pair). Changes in DNA can lead to changes in cell function like this:

There are basically three kinds of gene mutation, shown in this diagram:

The actual effect of a single mutation depends on many factors:

• A substitution on the third base of a codon may have no effect because the third base is less important (e.g. all codons beginning with CC code for proline).
• If a single amino acid is changed to a similar one (e.g. both small and uncharged), then the protein structure and function may be unchanged, but if an amino acid is changed to a very different one (e.g. an acidic R group to a basic R group), then the structure and function of the protein will be very different.
• If the changed amino acid is at the active site of the enzyme then it is more likely to affect enzyme function than if it is part of the supporting structure.
• Additions and Deletions are Frame shift mutations and are far more serious than substitutions because more of the protein is altered.
• If a frame-shift mutation is near the end of a gene it will have less effect than if it is near the start of the gene.
• If the mutation is in a gene that is not expressed in this cell (e.g. the insulin gene in a red blood cell) then it won't matter.
• If the mutation is in an non coding section of DNA then it probably won't matter.
• Some proteins are simply more important than others. For instance non-functioning receptor proteins in the tongue may lead to a lack of taste but is not life-threatening, whereas non-functioning haemoglobin is fatal.
• Some cells are more important than others. Mutations in somatic cells (i.e. non-reproductive body cells) will only affect cells that derive from that cell, so will probably have a small local effect like a birthmark (although they can cause widespread effects like diabetes or cancer). Mutations in germ cells (i.e. reproductive cells) will affect every single cell of the resulting organism as well as its offspring. These mutations are one source of genetic variation.

As a result of a mutation there are three possible phenotypic effects:

• Most mutations have no phenotypic effect. These are called silent mutations, and we all have a few of these.
• Of the mutations that have a phenotypic effect, most will have a negative effect. Most of the proteins in cells are enzymes, and most changes in enzymes will stop them working (because there are far more ways of making an inactive enzyme than there are of making a working one). When an enzyme stops working, a metabolic block can occur, when a reaction in cell doesn't happen, so the cell's function is changed. An example of this is the genetic disease phenylketonuria (PKU), caused by a mutation in the gene for the enzyme phenylalanine hydroxylase. This causes a metabolic block in the pathway involving the amino acid phenylalanine, which builds up, causing mental retardation.
• Very rarely a mutation can have a beneficial phenotypic effect, such as making an enzyme work faster, or a structural protein stronger, or a receptor protein more sensitive. Although rare beneficial mutations are important as they drive evolution.

The kinds of mutations discussed so far are called point or gene mutations because they affect specific points within a gene. There are other kinds of mutation that can affect many genes at once or even whole chromosomes. These chromosome mutations can arise due to mistakes in cell division. A well-known example is Down syndrome (trisonomy 21) where there are three copies of chromosome 21 instead of the normal two.

Mutations are normally very rare, which is why members of a species all look alike and can interbreed. However the rate of mutations is increased by chemicals or by radiation. These are called mutagenic agents or mutagens, and include:

• High energy ionising radiation such as x-rays, ultraviolet rays, a, b, or g rays from radioactive sources. These ionise the bases so that they don't form the correct base pairs.
• Intercalating chemicals such as mustard gas (used in World War 1), which bind to DNA separating the two strands.
• Chemicals that react with the DNA bases such as benzene, nitrous acid, and tar in cigarette smoke.
• Viruses. Some viruses can change the base sequence in DNA causing genetic disease and cancer.

During the Earth's early history there were far more of these mutagens than there are now, so the mutation rate would have been much higher than now, leading to a greater diversity of life. Some of these mutagens are used today in research, to kill microbes or in warfare. They are often carcinogens since a common result of a mutation is cancer.

The DNA molecule in a single human cell is 99 cm long, so is 10 000 times longer than the cell in which it resides (< 100mm). (Since an adult human has about 10¹⁴ cells, all the DNA is one human would stretch about 10¹⁴ m, which is a thousand times the distance between the Earth and the Sun.) In order to fit into the cell the DNA is cut into shorter lengths and each length is tightly wrapped up with histone proteins to form a complex called chromatin. During most of the life of a cell the chromatin is dispersed throughout the nucleus and cannot be seen with a light microscope. At various times parts of the chromatin will unwind so that genes on the DNA can be transcribed. This allows the proteins that the cell needs to be made.

Just before cell division the DNA is replicated, and more histone proteins are synthesised, so there is temporarily twice the normal amount of chromatin. Following replication the chromatin then coils up even tighter to form short fat bundles called chromosomes. These are about 100 000 times shorter than fully stretched DNA, and therefore 100 000 times thicker, so are thick enough to be seen under the microscope. Each chromosome is roughly X-shaped because it contains two replicated copies of the DNA. The two arms of the X are therefore identical. They are called chromatids, and are joined at the centromere. (Do not confuse the two chromatids with the two strands of DNA.) The complex folding of DNA into chromosomes is shown below.

Since the DNA molecule extends from one end of a chromosome to the other, and the genes are distributed along the DNA, then each gene has a defined position on a chromosome. This position is called the locus of the gene, and the loci of thousands of human genes are now known. There are on average about 3 000 genes per chromosome, although of course the larger chromosomes have more than this, and the smaller ones have fewer.

If a dividing cell is stained with a special fluorescent dye and examined under a microscope during cell division, the individual chromosomes can be distinguished. They can then be photographed and studied. This is a difficult and skilled procedure, and it often helps if the chromosomes are cut out and arranged in order of size.

This display is called a karyotype, and it shows several features:

• Different species have different number of chromosomes, but all members of the same species have the same number. Humans have 46 (this was not known until 1956), chickens have 78, goldfish have 94, fruit flies have 8, potatoes have 48, onions have 16, and so on. The number of chromosomes does not appear to be related to the number of genes or amount of DNA.
• Each chromosome has a characteristic size, shape and banding pattern, which allows it to be identified and numbered. This is always the same within a species. The chromosomes are numbered from largest to smallest.

• Chromosomes come in pairs, with the same size, shape and banding pattern, called homologous pairs ("same shaped"). So there are two chromosome number 1s, two chromosome number 2s, etc, and humans really have 23 pairs of chromosomes. Homologous chromosomes are a result of sexual reproduction, and the homologous pairs are the maternal and paternal versions of the same chromosome, so they have the same sequence of genes

• One pair of chromosomes is different in males and females. These are called the sex chromosomes, and are non-homologous in one of the sexes. In humans the sex chromosomes are homologous in females (XX) and non-homologous in males (XY). (In birds it is the other way round!) The non-sex chromosomes are sometimes called autosomes, so humans have 22 pairs of autosomes, and 1 pair of sex chromosomes.

Cells are not static structures, but are created and die. The life of a cell is called the cell cycle and has three phases:

In different cell types the cell cycle can last from hours to years. For example bacterial cells can divide every 30 minutes under suitable conditions, skin cells divide about every 12 hours on average, liver cells every 2 years.

The mitotic phase can be sub-divided into four phases (prophase, metaphase, anaphase and telophase). Mitosis is strictly nuclear division, and is followed by cytoplasmic division, or cytokinesis, to complete cell division. The growth and synthesis phases are collectively called interphase (i.e. in between cell division). Mitosis results in two "daughter cells", which are genetically identical to each other, and is used for growth and asexual reproduction. The details of each of these phases follows.

Asexual reproduction is the production of offspring from a single parent using mitosis. The offspring are therefore genetically identical to each other and to their "parent"- in other words they are clones. Asexual reproduction is very common in nature, and in addition we humans have developed some new, artificial methods. The Latin terms in vivo ("in life", i.e. in a living organism) and in vitro ("in glass", i.e. in a test tube) are often used to describe natural and artificial techniques. These different methods are summarised in the table.

Spores. These are simply specialised cells that are released from the parent (usually in very large numbers) to be dispersed. Under suitable conditions each cell can grow into a new individual. This method is used by most fungi and by the lower plants (mosses and ferns).

Fragmentation. This is when an organism spontaneously breaks into two or more fragments, each of which then develops into a new individual. It is used by some simple animals such as sponges, flatworms and starfish. Do not confuse this with regeneration, where some animals simply regenerate a part of their body if it is cut off.

Vegetative Reproduction. This term describes all the natural methods of asexual reproduction used by plants. A bud grows from a vegetative (i.e. not reproductive) part of the plant (usually the stem) and develops into a complete new plant, which eventually becomes detached from the parent plant. There are numerous forms of vegetative reproduction, including bulbs (e.g. onion, daffodil), corms (e.g. crocus, gladiolus), rhizomes (e.g. iris, couch grass), stolons (e.g. blackberry), runners (e.g. strawberry, creeping buttercup), tubers (e.g. potato, dahlia), tap roots (e.g. carrot, turnip), and tillers (e.g. grasses). Many of these methods are also perenating organs, which means they contain a food store and are used for survival over winter as well as for asexual reproduction. Since vegetative reproduction relies entirely on mitosis, all offspring are clones of the parent.

Parthenogenesis. This is used by some plants (e.g. the citrus fruits) and some invertebrate animals (e.g. honeybees, aphids, some crustaceans) as an alternative to sexual reproduction. Egg cells simply develop into adult clones without being fertilised. These clones may be haploid, or the chromosomes may replicate to form diploid cells.

Cloning is of great commercial importance, as brewers, pharmaceutical companies, farmers and plant growers all want to be able to reproduce "good" organisms exactly. Natural methods of asexual reproduction are often quite suitable for some organisms (such as yeast, potatoes and strawberries), but many important plants and animals do not reproduce asexually, so artificial methods have to be used.

Tissue Culture (or micropropagation). This is a more modern, and very efficient, way of cloning plants. Small samples of plant tissue, called an explant, can now be grown on agar plates in the laboratory in much the same way that bacteria can be grown. Initially the explant had to be meristem tissue (i.e. undifferentiated buds), but the technique has improved so that any tissue can now be used (e.g. from a leaf). The plant tissue can be separated into individual cells, each which can grow into a mass of cells called a callus, and if the correct plant hormones are added these cells can develop into whole plantlets, which can eventually be planted outside, where they will grow into normal-sized plants. Conditions must be kept sterile to prevent infection by microbes.

Micropropagation is used on a large scale for fruit trees, ornamental plants and plantation crops such as oils palm, data palm, sugar cane and banana. The advantages are:

• thousands of clones of a particularly good plant can be made quickly and in a small space
• disease-free plants can be grown from a few disease-free cells. In the field, almost all crop plants are infected with viruses.
• the technique works for plants species that cannot be asexually propagated by other means, such as palms and bananas.
• a single cell can be genetically modified and turned into many identical plants

Although some animal cells can be grown in culture, they cannot be grown into complete animals, so tissue culture cannot be used for cloning animals.

Embryo Cloning (or Embryo Splitting). The most effective technique for cloning animals is to duplicate embryo cells before they have irreversibly differentiated into tissues. It is difficult and quite expensive, so is only worth it for commercially-important farm animals, such as prize cows, or genetically engineered animals. A female animal is fed a fertility drug (FSH) so that she produces many mature eggs (superovulation). The eggs are then surgically removed from the female’s ovaries. The eggs are fertilised in vitro (IVF) using selected sperm from a prize male. The fertilised eggs (zygotes) are allowed to develop in vitro for a few days until the embryo is at the 16-cell stage. This young embryo can be split into 16 individual cells, which will each develop again into an embryo. (This is similar to the natural process when a young embryo splits to form identical twins.) The identical embryos can then be transplanted into the uterus of surrogate mothers, where they will develop and be born normally.

Could humans be cloned this way? Almost certainly yes. A human embryo was split and cloned to the stage of a few cells in the USA in 1993, just to show that it is possible. However experiments with human embryos are now banned in most countries including the UK for ethical reasons.

Somatic Cell Cloning (or Nuclear Transfer). The problem with embryo cloning is that you don’t know the characteristics of the animal you are cloning. By selecting good parents you hope it will have good characteristics, but you will not know until the animal has grown. It would be far better to clone a mature animal, whose characteristics you know. Until recently it was thought impossible to grow a new animal from the somatic cells of an existing animal (in contrast to plants). However, techniques have gradually been developed to do this, first with frogs in the 1970s, and most recently with sheep (the famous "Dolly") in 1996.

The technique used to create Dolly is similar to embryo cloning, but has one crucial difference. The cell used for Dolly was from the skin of the udder, so was a fully differentiated somatic cell. This cell was fused with a unfertilised egg cell which had had its nucleus removed. This combination of a diploid nucleus in an unfertilised egg cell was a bit like a zygote, and sure enough it developed into an embryo. The embryo was implanted into the uterus of a surrogate mother, and developed into an apparently normal sheep, Dolly. It took hundreds of attempts to achieve success with Dolly, but once the technique is improved it will be possible to combine this technique with embryo cloning to make many clones of an adult animal. Dolly’s "mother" was just an ordinary sheep, but in the future prize animals (or genetically engineered ones) could be cloned in this way.

Sexual reproduction is the production of offspring from two parent using gametes. The cells of the offspring have two sets of chromosomes (one from each parent), so are diploid. Sexual reproduction involves two stages:

• Meiosis- the special cell division that makes haploid gametes
• Fertilisation- the fusion of two gametes to form a diploid zygote

These two stages of sexual reproduction can be illustrated by a sexual life cycle:

Meiosis is a form of cell division. It starts with DNA replication, like mitosis, but then proceeds with two divisions one immediately after the other. Meiosis therefore results in four daughter cells rather than the two cells formed by mitosis. It differs from mitosis in two important aspects:

• The chromosome number is halved from the diploid number (2n) to the haploid number (n). This is necessary so that the chromosome number remains constant from generation to generation. Haploid cells have one copy of each chromosome, while diploid cells have homologous pairs of each chromosome.
• The chromosomes are re-arranged during meiosis to form new combinations of genes. This genetic recombination is vitally important and is a major source of genetic variation. It means for example that of all the millions of sperm produced by a single human male, the probability is that no two will be identical.

You don’t need to know the details of meiosis at this stage (It's covered in module 4).

The usual purpose of meiosis is to form gametes- the sex cells that will fuse together to form a new diploid individual.

In some algae and fungi the gametes are roughly the same size. This is called isogamy. There are no male and female sexes, but there can be + and - strains, who reproduce together.

In all plants and animals the gametes are different sizes. This is called heterogamy.

• The larger gametes tend to be stationary and contain food reserves (lipids, proteins, carbohydrates) to nourish the embryo after fertilisation. These are the female gametes (ova or eggs in animals, ovules in plants), and they are produced in fairly small numbers. Human females for example release about 500 ova in a lifetime.
• The smaller gametes can move.

• If they can propel themselves they are called motile (e.g. animal sperm)
• If they can easily be carried by the wind or animals they are called mobile (e.g. plant pollen).

It is this difference in gametes that actually defines the sex of an individual. Those individuals that produce small mobile gametes are the males, and those that produce the larger gametes are the females. In some species (such as most flowering plants) the same individual organisms can produce both male and female gametes, so they do not have distinct sexes and are called hermaphrodites. In other species (such as mammals) there are two distinct sexes, each producing their own gametes. These are called unisexual.

These diagrams of human gametes illustrate the differences between male and female.

Fertilisation is the fusion of two gametes to form a zygote.

In humans this takes place near the top of the oviduct. Hundreds of sperm reach the egg and use their tails to swim through the follicle cells (shown in this photo). When they reach the jelly coat surrounding the ovum they bind to receptors and this stimulates the rupture of the acrosome membrane in the sperms, releasing digestive enzymes, which make a path through the jelly coat. When a sperm reaches the ovum cell the two membranes fuse and the sperm nucleus enters the cytoplasm of the ovum. This triggers a series of reactions in the ovum that cause the jelly coat to thicken and harden, preventing any other sperm from entering the ovum. The sperm and egg nuclei then fuse, forming a diploid zygote.

In plants fertilisation takes place in the ovary at the base of the carpel. The haploid male nuclei travel down the pollen tube from the pollen grain on the stigma to the ovules in the ovary. In the ovule two fusions between male and female nuclei take place: one forms the zygote (which will become the embryo) while the other forms the endosperm (which will become the food store in the seed). This double fertilisation is unique to flowering plants.

For most of the history of life on Earth, organisms have reproduced only by asexual reproduction. Each individual was a genetic copy (or clone) of its "parent", and the only variation was due to random genetic mutation. The development of sexual reproduction in the eukaryotes around one billion years ago led to much greater variation and diversity of life. Sexual reproduction is slower and more complex than asexual, but it has the great advantage of introducing genetic variation (due to genetic recombination in meiosis and random fertilisation). This variation allows species to adapt to their environment and so to evolve. This variation is clearly such an advantage that practically all species can reproduce sexually. Some organisms can do both, using sexual reproduction for genetic variety and asexual reproduction to survive harsh times.

Genetic engineering, also known as recombinant DNA technology, means altering the genes in a living organism to produce a Genetically Modified Organism (GMO) with a new genotype. Various kinds of genetic modification are possible: inserting a foreign gene from one species into another, forming a transgenic organism; altering an existing gene so that its product is changed; or changing gene expression so that it is translated more often or not at all.

Genetic engineering is a very young discipline, and is only possible due to the development of techniques from the 1960s onwards. These techniques have been made possible from our greater understanding of DNA and how it functions following the discovery of its structure by Watson and Crick in 1953. Although the final goal of genetic engineering is usually the expression of a gene in a host, in fact most of the techniques and time in genetic engineering are spent isolating a gene and then cloning it. This table lists the techniques that we shall look at in detail.

These are enzymes that cut DNA at specific sites. They are properly called restriction endonucleases because they cut the bonds in the middle of the polynucleotide chain. Some restriction enzymes cut straight across both chains, forming blunt ends, but most enzymes make a staggered cut in the two strands, forming sticky ends.

The cut ends are "sticky" because they have short stretches of single-stranded DNA with complementary sequences. These sticky ends will stick (or anneal) to another piece of DNA by complementary base pairing, but only if they have both been cut with the same restriction enzyme. Restriction enzymes are highly specific, and will only cut DNA at specific base sequences, 4-8 base pairs long, called recognition sequences.

Restriction enzymes are produced naturally by bacteria as a defence against viruses (they "restrict" viral growth), but they are enormously useful in genetic engineering for cutting DNA at precise places ("molecular scissors"). Short lengths of DNA cut out by restriction enzymes are called restriction fragments. There are thousands of different restriction enzymes known, with over a hundred different recognition sequences. Restriction enzymes are named after the bacteria species they came from, so EcoR1 is from E. coli strain R, and HindIII is from Haemophilis influenzae.

In biology a vector is something that carries things between species. For example the mosquito is a disease vector because it carries the malaria parasite into humans. In genetic engineering a vector is a length of DNA that carries the gene we want into a host cell. A vector is needed because a length of DNA containing a gene on its own won’t actually do anything inside a host cell. Since it is not part of the cell’s normal genome it won’t be replicated when the cell divides, it won’t be expressed, and in fact it will probably be broken down pretty quickly. A vector gets round these problems by having these properties:

• It is big enough to hold the gene we want (plus a few others), but not too big.
• It is circular (or more accurately a closed loop), so that it is less likely to be broken down (particularly in prokaryotic cells where DNA is always circular).
• It contains control sequences, such as a replication origin and a transcription promoter, so that the gene will be replicated, expressed, or incorporated into the cell’s normal genome.
• It contain marker genes, so that cells containing the vector can be identified.

Many different vectors have been made for different purposes in genetic engineering by modifying naturally-occurring DNA molecules, and these are now available off the shelf. For example a cloning vector contains sequences that cause the gene to be copied (perhaps many times) inside a cell, but not expressed. An expression vector contains sequences causing the gene to be expressed inside a cell, preferably in response to an external stimulus, such as a particular chemical in the medium. Different kinds of vector are also available for different lengths of DNA insert:

Plasmids are by far the most common kind of vector, so we shall look at how they are used in some detail. Plasmids are short circular bits of DNA found naturally in bacterial cells. A typical plasmid contains 3-5 genes and there are usually around 10 copies of a plasmid in a bacterial cell. Plasmids are copied separately from the main bacterial DNA when the cell divides, so the plasmid genes are passed on to all daughter cells. They are also used naturally for exchange of genes between bacterial cells (the nearest they get to sex), so bacterial cells will readily take up a plasmid. Because they are so small, they are easy to handle in a test tube, and foreign genes can quite easily be incorporated into them using restriction enzymes and DNA ligase.

One of the most common plasmids used is the R-plasmid (or pBR322). This plasmid contains a replication origin, several recognition sequences for different restriction enzymes (with names like PstI and EcoRI), and two marker genes, which confer resistance to different antibiotics (ampicillin and tetracycline).

The diagram below shows how DNA fragments can be incorporated into a plasmid using restriction and ligase enzymes. The restriction enzyme used here (PstI) cuts the plasmid in the middle of one of the marker genes (we’ll see why this is useful later). The foreign DNA anneals with the plasmid and is joined covalently by DNA ligase to form a hybrid vector (in other words a mixture or hybrid of bacterial and foreign DNA). Several other products are also formed: some plasmids will simply re-anneal with themselves to re-form the original plasmid, and some DNA fragments will join together to form chains or circles. Theses different products cannot easily be separated, but it doesn’t matter, as the marker genes can be used later to identify the correct hybrid vector.

Vectors containing the genes we want must be incorporated into living cells so that they can be replicated or expressed. The cells receiving the vector are called host cells, and once they have successfully incorporated the vector they are said to be transformed. Vectors are large molecules which do not readily cross cell membranes, so the membranes must be made permeable in some way. There are different ways of doing this depending on the type of host cell.

• Heat Shock. Cells are incubated with the vector in a solution containing calcium ions at 0°C. The temperature is then suddenly raised to about 40°C. This heat shock causes some of the cells to take up the vector, though no one knows why. This works well for bacterial and animal cells.

• Electroporation. Cells are subjected to a high-voltage pulse, which temporarily disrupts the membrane and allows the vector to enter the cell. This is the most efficient method of delivering genes to bacterial cells.

• Viruses. The vector is first incorporated into a virus, which is then used to infect cells, carrying the foreign gene along with its own genetic material. Since viruses rely on getting their DNA into host cells for their survival they have evolved many successful methods, and so are an obvious choice for gene delivery. The virus must first be genetically engineered to make it safe, so that it can’t reproduce itself or make toxins. Three viruses are commonly used:

• Plant Tumours. This method has been used successfully to transform plant cells, which are perhaps the hardest to do. The gene is first inserted into the Ti plasmid of the soil bacterium Agrobacterium tumefaciens, and then plants are infected with the bacterium. The bacterium inserts the Ti plasmid into the plant cells' chromosomal DNA and causes a "crown gall" tumour. These tumour cells can be cultured in the laboratory and whole new plants grown from them by micropropagation. Every cell of these plants contains the foreign gene.
• Gene Gun. This extraordinary technique fires microscopic gold particles coated with the foreign DNA at the cells using a compressed air gun. It is designed to overcome the problem of the strong cell wall in plant tissue, since the particles can penetrate the cell wall and the cell and nuclear membranes, and deliver the DNA to the nucleus, where it is sometimes expressed.

• Micro-Injection. A cell is held on a pipette under a microscope and the foreign DNA is injected directly into the nucleus using an incredibly fine micro-pipette. This method is used where there are only a very few cells available, such as fertilised animal egg cells. In the rare successful cases the fertilised egg is implanted into the uterus of a surrogate mother and it will develop into a normal animal, with the DNA incorporated into the chromosomes of every cell.

• Liposomes. Vectors can be encased in liposomes, which are small membrane vesicles (see module 1). The liposomes fuse with the cell membrane (and sometimes the nuclear membrane too), delivering the DNA into the cell. This works for many types of cell, but is particularly useful for delivering genes to cell in vivo (such as in gene therapy).

These are needed to identify cells that have successfully taken up a vector and so become transformed. With most of the techniques above less than 1% of the cells actually take up the vector, so a marker is needed to distinguish these cells from all the others. We’ll look at how to do this with bacterial host cells, as that’s the most common technique.

A common marker, used in the R-plasmid, is a gene for resistance to an antibiotic such as tetracycline. Bacterial cells taking up this plasmid can make this gene product and so are resistant to this antibiotic. So if the cells are grown on a medium containing tetracycline all the normal untransformed cells, together with cells that have taken up DNA that’s not in a plasmid (99%) will die. Only the 1% transformed cells will survive, and these can then be grown and cloned on another plate.

Replica plating is a simple technique for making an exact copy of an agar plate. A pad of sterile cloth the same size as the plate is pressed on the surface of an agar plate with bacteria growing on it. Some cells from each colony will stick to the cloth. If the cloth is then pressed onto a new agar plate, some cells will be deposited and colonies will grow in exactly the same positions on the new plate. This technique has a number of uses, but the most common use in genetic engineering is to help solve another problem in identifying transformed cells.

This problem is to distinguish those cells that have taken up a hybrid plasmid vector (with a foreign gene in it) from those cells that have taken up the normal plasmid. This is where the second marker gene (for resistance to ampicillin) is used. If the foreign gene is inserted into the middle of this marker gene, the marker gene is disrupted and won't make its proper gene product. So cells with the hybrid plasmid will be killed by ampicillin, while cells with the normal plasmid will be immune to ampicillin. Since this method of identification involves killing the cells we want, we must first make a master agar plate and then make a replica plate of this to test for ampicillin resistance.

Once the colonies of cells containing the correct hybrid plasmid vector have been identified, the appropriate colonies on the master plate can be selected and grown on another plate.

The R-plasmid with its antibiotic-resistance genes dates from the early days of genetic engineering in the 1970s. In recent years better plasmids with different marker genes have been developed that do not kill the desired cells, and so do not need a replica plate. These new marker genes make an enzyme (actually lactase) that converts a colourless substrate in the agar medium into a blue-coloured product that can easily be seen. So cells with a normal plasmid turn blue on the correct medium, while those with the hybrid plasmid can't make the enzyme and stay white. These white colonies can easily be identified and transferred to another plate. Another marker gene, transferred from jellyfish, makes a green fluorescent protein (GFP).

Genes can be cloned by cloning the bacterial cells that contain them, but this requires quite a lot of DNA in the first place. PCR can clone (or amplify) DNA samples as small as a single molecule. It is a newer technique, having been developed in 1983 by Kary Mullis, for which discovery he won the Nobel prize in 1993. The polymerase chain reaction is simply DNA replication in a test tube. If a length of DNA is mixed with the four nucleotides (A, T, C and G) and the enzyme DNA polymerase in a test tube, then the DNA will be replicated many times. 

1. Start with a sample of the DNA to be amplified, and add the four nucleotides and the enzyme DNA polymerase.
2. Normally (in vivo) the DNA double helix would be separated by the enzyme helicase, but in PCR (in vitro) the strands are separated by heating to 95°C for two minutes. This breaks the hydrogen bonds.
3. DNA polymerisation always requires short lengths of DNA (about 20 bp long) called primers, to get it started. In vivo the primers are made during replication by DNA polymerase, but in vitro they must be synthesised separately and added at this stage. This means that a short length of the sequence of the DNA must already be known, but it does have the advantage that only the part between the primer sequences is replicated. The DNA must be cooled to 40°C to allow the primers to anneal to their complementary sequences on the separated DNA strands.
4. The DNA polymerase enzyme can now extend the primers and complete the replication of the rest of the DNA. The enzyme used in PCR is derived from the thermophilic bacterium Thermus aquaticus, which grows naturally in hot springs at a temperature of 90°C, so it is not denatured by the high temperatures in step 2. Its optimum temperature is about 72°C, so the mixture is heated to this temperature for a few minutes to allow replication to take place as quickly as possible.
5. Each original DNA molecule has now been replicated to form two molecules. The cycle is repeated from step 2 and each time the number of DNA molecules doubles. This is why it is called a chain reaction, since the number of molecules increases exponentially, like an explosive chain reaction. Typically PCR is run for 20-30 cycles.

PCR can be completely automated, so in a few hours a tiny sample of DNA can be amplified millions of times with little effort. The product can be used for further studies, such as cloning, electrophoresis, or gene probes. Because PCR can use such small samples it can be used in forensic medicine (with DNA taken from samples of blood, hair or semen), and can even be used to copy DNA from mummified human bodies, extinct woolly mammoths, or from an insect that's been encased in amber since the Jurassic period. One problem of PCR is having a pure enough sample of DNA to start with. Any contaminant DNA will also be amplified, and this can cause problems, for example in court cases.

Complementary DNA (cDNA) is DNA made from mRNA. This makes use of the enzyme reverse transcriptase, which does the reverse of transcription: it synthesises DNA from an RNA template. It is produced naturally by a group of viruses called the retroviruses (which include HIV), and it helps them to invade cells. In genetic engineering reverse transcriptase is used to make an artificial gene of cDNA as shown in this diagram.

Complementary DNA has helped to solve different problems in genetic engineering:

It makes genes much easier to find. There are some 70 000 genes in the human genome, and finding one gene out of this many is a very difficult (though not impossible) task. However a given cell only expresses a few genes, so only makes a few different kinds of mRNA molecule. For example the b cells of the pancreas make insulin, so make lots of mRNA molecules coding for insulin. This mRNA can be isolated from these cells and used to make cDNA of the insulin gene.

These are used to identify and label DNA fragments that contain a specific sequence. A probe is simply a short length of DNA (20-100 nucleotides long) with a label attached. There are two common types of label used:

• a radioactively-labelled probe (synthesised using the isotope ³²P) can be visualised using photographic film (an autoradiograph).
• a fluorescently-labelled probe will emit visible light when illuminated with invisible ultraviolet light. Probes can be made to fluoresce with different colours.

Probes are always single-stranded, and can be made of DNA or RNA. If a probe is added to a mixture of different pieces of DNA (e.g. restriction fragments) it will anneal (base pair) with any lengths of DNA containing the complementary sequence. These fragments will now be labelled and will stand out from the rest of the DNA. DNA probes have many uses in genetic engineering:

• To identify restriction fragments containing a particular gene out of the thousands of restriction fragments formed from a genomic library. This use is described in shotguning below.
• To identify the short DNA sequences used in DNA fingerprinting.
• To identify genes from one species that are similar to those of another species. Most genes are remarkably similar in sequence from one species to another, so for example a gene probe for a mouse gene will probably anneal with the same gene from a human. This has aided the identification of human genes.
• To identify genetic defects. DNA probes have been prepared that match the sequences of many human genetic disease genes such as muscular dystrophy, and cystic fibrosis. Hundreds of these probes can be stuck to a glass slide in a grid pattern, forming a DNA microarray (or DNA chip). A sample of human DNA is added to the array and any sequences that match any of the various probes will stick to the array and be labelled. This allows rapid testing for a large number of genetic defects at a time.

This is used to find one particular gene in a whole genome, a bit like finding the proverbial needle in a haystack. It is called the shotgun technique because it starts by indiscriminately breaking up the genome (like firing a shotgun at a soft target) and then sorting through the debris for the particular gene we want. For this to work a gene probe for the gene is needed, which means at least a short part of the gene’s sequence must be known.

These are used to turn off the expression of a gene in a cell. The principle is very simple: a copy of the gene to be switch off is inserted into the host genome the "wrong" way round, so that the complementary (or antisense) strand is transcribed. The antisense mRNA produced will anneal to the normal sense mRNA forming double-stranded RNA. Ribosomes can’t bind to this, so the mRNA is not translated, and the gene is effectively "switched off".

It is possible to chemically synthesise a gene in the lab by laboriously joining nucleotides together in the correct order. Automated machines can now make this much easier, but only up to a limit of about 30bp, so very few real genes could be made this way (anyway it’s usually much easier to make cDNA). The genes for the two insulin chains (xx bp) and for the hormone somatostatin (42 bp) have been synthesisied this way. It is very useful for making gene probes.

This is a form of chromatography used to separate different pieces of DNA on the basis of their length. It might typically be used to separate restriction fragments. The DNA samples are placed into wells at one end of a thin slab of gel made of agarose or polyacrylamide, and covered in a buffer solution. An electric current is passed through the gel. Each nucleotide in a molecule of DNA contains a negatively-charged phosphate group, so DNA is attracted to the anode (the positive electrode). The molecules have to diffuse through the gel, and smaller lengths of DNA move faster than larger lengths, which are retarded by the gel. So the smaller the length of the DNA molecule, the further down the gel it will move in a given time. At the end of the run the current is turned off.

Unfortunately the DNA on the gel cannot be seen, so it must be visualised. There are three common methods for doing this:

• The gel can be stained with a chemical that specifically stains DNA, such as ethidium bromide or azure A. The DNA shows up as blue bands. This method is simple but not very sensitive.
• The DNA samples at the beginning can be radiolabelled with a radioactive isotope such as ³²P. Photographic film is placed on top of the finished gel in the dark, and the DNA shows up as dark bands on the film. This method is extremely sensitive.
• The DNA fragments at the beginning can be labelled with a fluorescent molecule. The DNA fragments show up as coloured lights when the finished gel is illuminated with invisible ultraviolet light.

This means reading the base sequence of a length of DNA. Once this is known the amino acid sequence of the protein that the DNA codes for can also be determined, using the genetic code table. The sequence can also be compared with DNA sequences from other individuals and even other species to work out relationships.

DNA sequencing is based on a beautifully elegant technique developed by Fred Sanger, and now called the Sanger method.

There is now a modified version of the Sanger method called cycle sequencing, which can be completely automated. The primers are not radiolabelled, but instead the four dideoxy nucleotides are fluorescently labelled, each with a different colour (A* is green, T* is red, C* is blue and G* is yellow). The polymerisation reaction is done in a single tube, using PCR-like cycles to speed up the process. The resulting mixture is separated using capillary electrophoresis, which gives good separation in a single narrow gel. The gel is read by a laser beam and the sequence of colours is converted to a DNA sequence by computer program (like the screenshot below). This technique can sequence an amazing 12 000 bases per minute.

Thousands of genes have been sequenced using these methods and the entire genomes of several organisms have also been sequenced. A huge project is underway to sequence the human genome, and it delivered a draft sequence in June 2000. The complete 3 billion base sequence should be complete by 2003. This information will give us unprecedented knowledge about ourselves, and is likely to lead to dramatic medical and scientific advances.

We have now looked at some of the many techniques used by genetic engineers. What can be done with these techniques? By far the most numerous applications are still as research tools, and the techniques above are helping geneticists to understand complex genetic systems. Despite all the hype, genetic engineering still has very few successful commercial applications, although these are increasing each year. The applications so far can usefully be considered in three groups.

The biggest and most successful kind of genetic engineering is the production of gene products. These products are of medical, agricultural or commercial value. This table shows a few of the examples of genetically engineered products that are already available.

The products are mostly proteins, which are produced directly when a gene is expressed, but they can also be non-protein products produced by genetically-engineered enzymes. The basic idea is to transfer a gene (often human) to another host organism (usually a microbe) so that it will make the gene product quickly, cheaply and ethically. It is also possible to make "designer proteins" by altering gene sequences, but while this is a useful research tool, there are no commercial applications yet.

Since the end-product is just a chemical, in principle any kind of organism could be used to produce it. By far the most common group of host organisms used to make gene products are the bacteria, since they can be grown quickly and the product can be purified from their cells. Unfortunately bacteria cannot not always make human proteins, and recently animals and even plants have also been used to make gene products. In neither case is it appropriate to extract the product from their cells, so in animals the product must be secreted in milk or urine, while in plants the product must be secreted from the roots. This table shows some of the advantages and disadvantages of using different organisms for the production of genetically-engineered gene products.