Manipulating Genomes

By the early 1970s, the structure of DNA was known, as were the sequences of base triplets that coded for the various amino acids. However, at this time it was difficult to work out the sequence of the nucleotide base triplets in genes.

In 1969, a gene was isolated from a bacterial chromosome. In 1972, a Belgian molecular biologist sequenced a gene that codes for the protein coat of a virus, MS2. Both scientists worked from the mRNA transcribed from the gene, and not the raw DNA. RNA is unstable and this whole process was extremely slow and only suitable for very short genes. In 1975, the British biochemist Fred Sanger developed a method that ultimately allowed scientists to sequence whole genomes.

Sanger's approach was to use a single strand of DNA as a template for four experiments in separate dishes. Each dish contained a solution with the four bases A, T C and G plus an enzyme, DNA polymerase. 

To each dish, a modified version of one of the DNA bases was added. The base was modified in such a way that, once incorporated into the synthesised complementary strand of DNA no more bases could be added. Each modified base was also labelled with a radioactive isotope. 

As the reaction progressed, thousands of DNA fragments of varying lengths were generated. The DNA fragments were passed through a gel by electrophoresis. Smaller fragments travelled further, so the fragments became sorted by length. 

The nucleotide base at the end of each fragment was red according to its radioactive label. 

• If the first one-base fragments had thymine at the end, then the first base in the sequence is T. 
• If the two-base fragments have cytosine at the end, then the sequence is TC. 
• If the three-base fragment ends with guanine, then the base sequence is TCG. 

This method was efficient and safe. Sanger used it to sequence the genome of a phage virus (a virus that infects bacteria) called Phi-X174, the first DNA-based organism to have its genome sequenced. He had to count off the bases, one by one, from the bands in a piece of gel -a very time-consuming and therefore expensive. 

The gene to be sequenced was isolated, using restriction enzymes, from a bacterium. 

The DNA was then inserted into a bacterial plasmid (the vector) and then into an Escherichia coli bacterium host that, when cultured, divided many times, enabling the plasmid with the DNA insert to be copied many times.

Each new bacterium contained a copy of the candidate gene. These lengths of DNA were isolated using plasmid preparation techniques for and were then sequenced. 

In 1986, the first automated DNA sequencing machine was developed at the California Institute of Technology, based on Fred Sanger's method. Fluorescent dyes instead of radioactivity were used to label the terminal bases. These dyes glowed when scanned with a laser beam, and the light signature was identified by computer. This method dispensed with the need for technicians to read autoradiograms.

In the first decade of the twenty-first century, a variety of approaches was used to develop fast, cheap methods to sequence genomes. One of them was pyrosequencing.

This method was developed in 1966 and uses sequencing by synthesis, not by chain termination as in the Sanger method. It involves synthesising a single strand of DNA, complementary to the strand to be sequenced, one base at a time, whilst detecting, by light emission, which base was added at each step. 

1. A long length of DNA to be sequenced is mechanically cut into fragments of 300 800 base pairs, using a nebuliser. 
2. These lengths are then degraded into single-stranded DNA (ssDNA). These are the template DNAs and they are immobilised. 
3. A sequencing primer is added and the DNA is then incubated with the enzymes DNA polymerase, ATP sulfurylase, luciferase, apyrase and the substrates adenosine 5' phosphosulfate (APS) and luciferin. Only one of the four possible activated nucleotides, ATP, TTP, CTP and GTP is added at any one time and any light generated is detected.

4. One activated nucleotide (a nucleotide with two extra phosphoryl groups), such as TTP (thymine triphosphate), is incorporated into a complementary strand of DNA using the strand to be sequenced as a template. As this happens, the two extra phosphoryls are released as pyrophosphate (PP). In the presence of APS, the enzyme ATP sulfurylase converts the pyrophosphate to ATP. In the presence of this ATP the enzyme luciferase converts luciferin to oxyluciferin. This conversion generates visible light which can be detected by a camera. The amount of light generated is proportional to the amount of ATP available, and therefore, indicates how many of the same type of activated nucleotide were incorporated adjacently into the complementary DNA strand. 

Unincorporated activated nucleotides are degraded by apyrase and the reaction starts again with another nucleotide. 1 million reads occur simultaneously, so a 10-hour run generates 400 million bases of sequencing information. Software packages assembles these sequences into longer sequences. 

A branch of biology called bioinformatics has grown out of this research, to store the huge amounts of data generated. It would have been impossible to store and analyse these data prior to computers and microchips. Software packages are specially designed for this purpose.

Scientists predicted that the human genome would contain about 100 000 genes. In 1990, the Human Genome Project was launched, and the genome was sequenced by 2003. Scientists were surprised to learn that the human genome contained only about 24 000 genes, not many more than in the mouse genome. 

Whole genome sequencing determines the complete DNA sequence of an organism's genome - in the case of eukaryotic cells, that is the genetic material of the chromosomes, mitochondria and, if plants and algae, also of chloroplasts. Sequences genomes are stored in gene banks.

When the human genome was compared with those of other species, it became clear that few human genes are unique to us. Most of our genes have counterparts in other organisms. We share over 99% of our genes with chimpanzees. This verifies that genes that work well tend to be conserved by evolution. For example, pigs and humans have similar genes for insulin, which is why, prior to genetically-modifying bacteria to make insulin, pig insulin was used to treat patients with diabetes. 

Sometimes, as evolution progresses, some genes are co-opted to perform new tasks. Tiny changes to a gene in humans called FOXP2, which is found in other mammals including mice and chimpanzees, mean that in humans this gene allows speech. 

Many of the differences organisms are not because the between organisms have totally different genes, but some of their because shared genes have been altered and now work in subtly different ways. Some changes to the regulatory regions of DNA that do not code directly for proteins have also altered the expression of the genomes - regulatory and coding genes interact in such ways that, without increasing the number of genes, the numbers of proteins made may be increased.

Comparing genomes of organisms thought to be closely related species has helped confirm their evolutionary relationships or has led to new knowledge about the relationships and, in some cases, to certain organisms being reclassified. 

The DNA from bones and teeth of some extinct animals can be amplified and sequenced, so that the animals' evolutionary history can be verified. 

Recently, samples of extinct cave bear (Ursus spelaeus) genomes were sequenced using high throughput techniques, and the sequence data obtained was compared to those of dogs. Dogs and bears diverged about 50 million years ago and share 92% of their genome.

All humans are genetically similar. Except for rare cases, where a gene has been lost by deletion of part of a chromosome, we all have the same genes, but we have different alleles. About 0.1% of our DNA is not shared with others. This sounds very small but given that our genome contains three billion DNA base pairs this means that there are three million places on the DNA lengths where our DNA sequences can differ due to random mutations such as substitution.

The places on the DNA where these substitutions occur are called single nucleotide polymorphisms, or SNPs (pronounced 'snips"). Some have no effect on the protein, some can alter a protein or alter the way a piece of RNA regulates the expression of another gene.

Methylation of certain chemical groups in DNA plays a major role in regulating gene expression in eukaryotic cells. Methods to map this methylation of whole human genomes can help researchers to understand the development of certain diseases, for example certain types of cancer and why they may or may not develop in genetically similar individuals. The study of this aspect of genetics is called epigenetics.

Determining the sequence of amino acids within a protein is laborious and time consuming. However, if researchers have the organism's genome sequenced and know which gene codes for a specific protein, by using knowledge of which base triplets code for which amino acids, they can determine the primary structure of proteins. The researchers need to know which part of the gene codes for exons and which codes for introns. 

Synthetic biology is an interdisciplinary science concerned with designing and building useful biological devices and systems. It encompasses biotechnology, evolutionary biology, molecular biology, systems biology and biophysics. Its ultimate goals may be to build engineered biological systems that store and process information, provide food, maintain human health and enhance the environment.

Example

Description of application 

Information storage

Scientists can encode vast amounts of digital information onto a single strand of synthetic DNA. One project has encoded the complete works of William Shakespeare onto a strand of synthetic DNA.

Production of medicines

Escherichia coli and yeast have both been genetically engineered to produce the precursor of a good antimalarial drug, artemisinin, previously only available by extracting it from certain parts of the Artemisia plants at particular times in the plant's life cycle.

Novel proteins

Designed proteins have been produced, for example one that is similar to haemoglobin and binds to oxygen, but not to carbon monoxide.

Biosensors 

Modified bioluminescent bacteria, placed on a coating of a microchip, glow if air polluted with petroleum pollutants.

Nanotechnology

Material can be produced for nanotechnology e.g. amyloid fibres for making biofilms for functions such as adhesion.

Synthetic biology raises issues of ethics and biosecurity. Extensive regulations are already in place, due to 30-40 years of using genetically-modified organisms. There are many advisory panels and many scientific papers have been written on how to manage the risks. Synthetic biology is not about making synthetic life forms from scratch, but is about a potential for new systems with rewards and associated risks to be managed.

There are slightly different procedures, but the principles underlying them all are: 

1. DNA is obtained from the individual - either by a mouth swab, from saliva on a toothbrush, from blood or hair or in the case of ancient remains, from bone. 
2. The DNA is then digested with restriction enzymes. These enzymes cut the DNA at specific recognition sites. They will cut it into fragments, which will vary in size from individual to individual.
3. The fragments are separated by gel electrophoresis and stained. Larger fragments travel the shortest distance in the gel. 
4. A banding pattern can be seen. 
5. The DNA to which the individual's is being compared is treated with the same restriction enzymes and also subjected to electrophoresis. 
6. The banding patterns of the DNA samples can then be compared. 

The first method involved restriction fragment length polymorphism analysis. This method is laborious and is no longer used. Today, short tandem repeat (STR) sequences of DNA are used. These are highly variable short repeating lengths of DNA. The exact number of STRs varies from person to person.

STR sequences are separated by electrophoresis. Each STR is polymorphic, but the number of alleles in the gene pool for each one is small. Thirteen STRs are analysed simultaneously, so although each STR is present in between 5% and 20% of individuals, the chances of two people sharing STR sequences at all the loci is 1 x 1018. This is a greater number than the number of people on Earth. However, there are an estimated 12 million identical twins on Earth. 

The technique is very sensitive, and even a trace of DNA left when someone touches an object can produce a result. Samples must be tested carefully to avoid contamination. DNA can be stored for many years if a crime case is unsolved. It can then later be used to assess new evidence. 

DNA profiling has transformed forensic science. Not only has it brought about convictions, and established the innocence of many suspects and of people previously wrongly convicted, it has been used to: 

• identify Nazi war criminals hiding in South America 
• identify the remains of the Romanov family and to refute a person's claim to be the survivor, Anastasia 
• identify remains found in Leicester as those of Richard III 
• identify victims' body parts after air crashes, terrorist attacks other disasters 
• match profiles from descendants of those lost during World War I with the unidentified remains of the soldiers who fell on battlefields in Northern France.

Half of every child's genetic information comes from the mother and half from the father; hence half the short tandem repeat (STR) fragments come from the mother and half from the father. Comparing the DNA profiles of mother, father and child can therefore establish maternity and/or paternity.

Protein electrophoresis can detect the type of haemoglobin present and aid diagnosis of sickle cell anaemia. A varying number of repeat sequences for a condition such as Huntington disease can be detected by electrophoresis.

Kary Mullis developed a technique, the polymerase chain reaction (PCR), to amplify (increase the amount of) DNA, enabling it to be analysed. The PCR soon became incorporated into forensic DNA analysis and into the protocols for analysis of DNA for genetic diseases. 

The PCR is artificial replication of DNA. It relies on the facts that: 

• DNA is made of two antiparallel backbone strands 
• each strand of DNA has a 5' end and a 3' end
• DNA grows only from the 3' end 
• base pairs pair up according to complementary base pairing rules, A with T and G with C 

PCR differs from DNA replication in that:

• only short sequences, of up to 10 000 base pairs, of DNA can be replicated, not entire chromosomes
• It requires the addition of primer molecules to make the process start
• A cycle of heating and cooling is needed to separate the DNA strands, bind primers to the strands and for the DNA strands to be replicated. 

The PCR is a cyclic reaction. The steps are:

1. The sample of DNA is mixed with DNA nucleotides, primers, magnesium ions and the enzyme Taq DNA polymerase. 

2. The mixture is heated to around 94-96°C to break the hydrogen bonds between complementary nucleotide base pairs and thus denature the double-stranded DNA into two single strands of DNA.

3. The mixture is cooled to around 68°C, so that the primers can anneal (bind by hydrogen bonding) to one end of each single strand of DNA. This gives a small section of double-stranded DNA at the end of each single-stranded molecule.

4. The Taq DNA polymerase enzyme molecules can now bind to the end where there is double-stranded DNA. Taq polymerase is obtained from a bacterium that lives at high temperatures, 72°C is the optimum temperature for this enzyme. The temperature is raised to 72°C, which keeps the DNA as single strands. 

5. The Taq DNA polymerase catalyses the addition of DNA nucleotides to the single-stranded DNA molecules, starting at the end with the primer and proceeding in the 5' to 3' direction. 

6. When the Taq DNA polymerase reaches the other end of the DNA molecule, then a new double strand of DNA has been generated. 

7. The whole process begins again and is repeated for many cycles. The amount of DNA increases exponentially: 1, 2, 4, 8, 16, 32, 64, 128 and so on.

Since its inception, the PCR has been improved and elaborated on in many ways. It is used to amplify DNA samples for sequencing. It is commonly used for a wide variety of applications including:

• Tissue typing: donor and recipient tissues can be typed prior to transplantation to reduce the risk of rejection of the transplant. 
• Detection of oncogenes: if the type of mutation involved in a specific patient's cancer is found, then the medication may be better tailored to that patient. 
• Detecting mutations: a sample of DNA is analysed for the presence of a mutation that leads to a genetic disease. Parents can be tested to see if they carry a recessive allele for a particular gene; fetal cells may be obtained from the mother's blood stream for prenatal genetic screening; during IVF treatment, one cell from an eight-cell embryo can be used to analyse the fetal DNA before implantation 
• Identifying viral infections: sensitive PCR tests can detect small quantities of viral genome amongst the host cells' DNA. This can be used to verify, for example, HIV or hepatitis C infections.

• Monitoring the spread of infectious disease: the spread of pathogens through a population of wild or domestic animals, or from animals to human populations, can be monitored, and the emergence of new more virulent sub-types can be detected. 
• Forensic science: small quantities of DNA can be amplified for DNA profiling, to identify criminals or to ascertain parentage. 
• Research: amplifying DNA from extinct ancient sources such as Neanderthal or woolly mammoth bones, for analysis and sequencing. In extant organisms, tissues or cells can be analysed to find out which genes are switched on or off. 

Electrophoresis is used to separate different sized fragments of DNA. It can separate fragments that differ by only one base pair, and is widely used in gene technology to separate DNA fragments for identification and analysis. 

The technique uses an agarose gel plate covered by a buffer solution. Electrodes are placed in each end of the tank so that, when it is connected to a power supply, an electric current can pass through the gel. DNA has an overall negative charge, due to its many phosphate groups, and the fragments migrate towards the anode (positive electrode). Fragments of DNA all have a similar surface charge, regardless of their size.

1. The DNA samples are first digested with restriction enzymes to cut them, at specific recognition sites, into fragments. This is carried out at 35-40°C and may take up to an hour. 

2. While the restriction enzymes are cutting the DNA, the tank is set up. The agarose gel is made up and poured into the central region of the tank, whilst combs are in place at one end. Once the gel is set, buffer solution is added so that the gel is covered and the end sections of the tank contain buffer solution. Now the comb can be carefully removed, leaving wells at one end of the gel. 

3. A loading dye is added to the tubes containing the digested DNA. 

4. The digested DNA plus loading dye is added to wells in the electrophoresis gel. To do this, a pipette is used and this is held, in the buffer solution, just above one of the wells. The loading dye is dense and carries the DNA down into the well. The pipette should not be placed right into the well, otherwise you might pierce the bottom of the well. 

5. Once all the wells have been loaded with the different DNA samples, the electrodes are put into place and connected to an 18V battery. This is then left to run for up to 6-8hours. Alternatively, a higher voltage power pack can be used and the gel run for a shorter time (less than 2 hours); do not use a higher voltage unless the current is limited to 5mA or less, otherwise there is a risk of severe electric shock from the electrodes or gel. 

6. The DNA fragments move through the gel at different speeds. Smaller fragments travel faster, so in a fixed period they travel further. 

7. At the end of the period, the buffer solution is poured away and a dye is added to the gel. This dye adheres to the DNA and stains the fragments. 

The principle for separating proteins is the same as for separating DNA fragments, but is often carried out in the presence of a charged detergent such as sodium dodecyl sulfate (SDS), which equalises the surface charge on the molecules and allows the proteins to separate as they move through the gel, according to their molecular mass. In some cases the proteins can be separated according to mass and then, without SDS, according to their surface charge. 

This technique can be used to analyse the types of haemoglobin proteins for diagnosis of conditions such as:

• Sickle cell anaemia, where the patient has haemoglobin S and not the normal haemoglobin A. 
• Aplastic anemia, thalassaemia and leukaemia, where the patients have higher than normal amounts of foetal haemoglobin (haemoglobin F), and lower then normal amounts of haemoglobin A. 

A DNA probe is a short (50-80 nucleotides) single-stranded length of DNA that is complementary to a section of the DNA being investigated. The prober may be labelled using:

• A radioactive marker, usually with 32P in one of the phosphate groups in the probe strand. Once the probe has annealed by complementary base pairing, to the piece of DNA, it can be revealed by exposure to photographic film. 
• A fluorescent marker that emits a colour on exposure to UV light. Fluorescent markers may also be used in automated DNA sequencing. 

Probes are useful in locating specific DNA sequences, for example:

• To locate a specific gene needed for use in genetic engineering 
• To identify the same gene in variety of different genomes from different species when conducting genome comparison studies
• To identify the presence or absence of a specific allele for a particular genetic disease or that given susceptibility to a particular condition. 

Scientists can place a number of different probes on a fixed surface, known as a DNA microarray. Applying the DNA under investigation to the surface can reveal the presence of mutated alleles that match the fixed probes, because the sample DNA will anneal to any complementary fixed probes. 

The sample DNA must first be broken into smaller fragments, and it may also be amplified using the polymerase chain reaction (PCR). A DNA microarray can be made with fixed probes, specific for certain sequences found in mutated alleles that cause genetic diseases, in the well. 

Reference and test DNA samples are labelled with fluorescent markers. Where a test subject and a reference marker both bind to a particular probe, the scan reveals fluorescence of both colours, indicating the presence of the particular sequence in the DNA. 

Genetic engineering is also known as recombinant DNA technology, because it involves combining DNA from different organisms. It is also called genetic modification. Genes are isolated from one organisms and inserted into another organism, using suitable vectors. 

The following stages are necessary:

1. The required gene is obtained. 

2. A copy of the gene is placed inside a vector. 

3. The vector carries the gene into a recipient cell. 

4. The recipient expresses the novel gene. 

1. Obtaining the required gene

• mRNA can be obtained from cells where the gene is being expressed. An enzyme, reverse transcriptase, can then catalyse the formation of a single strand of complementary DNA (cDNA) using the mRNA as a template. The addition of primers and DNA polymerase can make this cDNA into a double-stranded length of DNA, whose base sequence codes for the original protein.
• If scientists know the nucleotide sequence of the gene, then the gene can be synthesised using an automated polynucleotide synthesiser. 
• If scientists know the sequence of the gene, they can design PCR primers to amplify the gene from the genomic DNA. 
• A DNA probe can be used to locate a gene within the genome and the gene can then be cut out using restricting enzymes. 

2.   Placing the gene into a vector 

• Plasmids can be obtained from organisms such as bacteria and mixed with restriction enzymes that will cut the plasmid at specific recognition sites. 
• The cut plasmid has exposed unpaired nucleotide bases, called sticky ends. 
• If free nucleotide bases, complementary to the sticky ends of the plasmid, are added to the ends of the gene to be inserted, then the gene and cut plasmid should anneal. DNA lipase enzymes catalysed the annealing. 
• A gene may be sealed into an attenuated (weakened) virus that could carry it into a host cell. 

3.   Getting the vector into the recipient cell

• DNA does not easily cross the recipient cell's plasma membrane. Various methods can be used to aid the process:
• Heat shock treatment - if bacteria are subjected to alternating periods of cold (0C) and heat (42C) in the presence of calcium chloride, their walls and membranes will become more porous and allow in the recombinant vector. This is because the positive calcium ions surround the negatively charged parts of both the DNA molecules and phospholipids in the cell membrane, thus reducing repulsion between the foreign DNA and the host cell membranes. 
• Electroporation - a high voltage pulse is applied to the cell to disrupt the membrane. 
• Electrofusion - electrical fields help to introduce DNA into cells
• Transfection - DNA can be packaged into a bacteriophage, which can then transfect the host cell. 
• T1 (recombinant) plasmids are inserted into a bacterium Agrobacterium tumefaciens, which infects some plants and naturally inserts its genome into the host cell genomes. 

4.   Direct method of introducing gene into recipient 

If plants are not susceptible to a A. tumefaciens, then direct methods can be used. Small pieces of gold or tungsten are coated with the DNA and shot into the plant cells. This is called a 'gene gun'. 

Retroviruses, such as HIV, which contain RNA that they inject into the host genome, have reverse transcriptase enzyme that catalyses the production of cDNA (complementary DNA) using their RNA as a template. This is the reverse of transcription. These enzymes are useful for genetic engineering, as outlined previously.

Bacteria and Archaea have restriction enzymes, called restriction endonucleases, to protect them from attack by phage viruses. These enzymes cut up the foreign viral DNA, by a process called restriction, preventing the viruses from making copies of themselves. The prokaryotic DNA is protected from the action of these endonucleases by being methylated at the recognition sites.

The restriction endonucleases are named according to the bacterium from which they have been obtained. The first one used was EcoR1 - it was obtained from E. coli and was restriction endonuclease number 1. Restriction enzymes are useful to molecular biology and biotechnology as molecular scissors, as they recognise specific sequences within a length of DNA and cleave the molecule there. Some make a staggered cut leaving sticky ends. Others make a cut that produces blunt ends. 

DNA ligase enzyme is used in molecular biology to join DNA fragments. It catalyses condensation reactions that join the sugar groups and phosphate groups of the DNA backbone. These enzymes catalyse such reactions during DNA replication in cells and are also used in the PCR.

Scientists can obtain mRNA from beta cells of islets of Langerhans in the human pancreas, where insulin is made. 

1. Adding reverse transcriptase enzyme makes a single strand of cDNA and treatment with DNA polymerase makes a double strand - the gene. 

2. Addition of free unpaired nucleotides at the ends of the DNA produces sticky ends. 

3. Now, with the help of ligase enzyme, the insulin gene can be inserted into plasmids extracted from E. coli bacteria. These are now called recombinant plasmids, as they contain inserted DNA. 

4. E. coli bacteria are mixed with recombinant plasmids and subjected to heat shock in the presence of calcium chloride ions, so that they will take up the plasmids. 

Genetically modified bacteria are cultured in large numbers to produce insulin.

Because transformed (transgenic) bacteria have resistance to some antibiotics, we do not want them to 'escape' from laboratories into the wild.

They have a gene knocked out, which means they cannot synthesise a particular nutrient. They survive in the laboratory where they are given that nutrient in their growth medium but will not survive outside of the laboratory. 

Humans have been genetically modifying plants for thousands of years. All the crops grown around the world have genomes vastly different from those of their wild relatives, as a result of human intervention and selective breeding. Agriculture has changed the face of the landscape and produced domesticated breeds of animals and plants. However, selective breeding relies on a rather 'hit or miss' technique and may produce unexpected results.

In the 1970s, the techniques of recombinant DNA technology allowed scientists to splice new genes into plant genomes. This gave a more exact way of transforming crop plants, compared with inducing mutations or crossing different strains. Now specific genes conferring desirable traits can be excised from one organism and inserted into another, using a vector or a 'gene gun'.

Some people are concerned about potential hazards and risks associated with genetic modification (GM). However, the potential benefits have to be recognised and weighed against the potential hazards. Anti-GM campaigners want the technology abolished, which would also abolish any choice and the opportunity to exploit potential benefits. 

The basic principle of gene therapy is to insert a functional allele of a particular gene into cells that contain only mutated and non- functioning alleles of that gene. If the inserted allele is expressed, then the individual will produce a functioning protein and no longer have the symptoms associated with the genetic disorder. 

Knowledge gained from the Human Genome Project has led to further possibilities, such as using interference RNA to silence genes by blocking translation. Interference RNA has been used to treat cytomegalovirus infections in AIDS patients by blocking replication of the cytomegalovirus.

Some metabolic disorders such as cystic fibrosis occur when an individual inherits two faulty recessive alleles for a particular gene. As a result, the differentiated cells where this gene should normally be expressed lack the protein product of that gene. If functioning alleles for this gene can be put into specific cells so that these cells then make the protein, these cells will function normally. Various methods are available for delivering the functioning alleles to the patient's body cells. Somatic cell gene therapy affects only certain cell types. The alterations made to the patient's genome in those cells are not passed to the patient's offspring. 

Patients with cystic fibrosis lack a functioning CFTR gene. The alleles, which are lengths of DNA, can be packaged within small spheres of lipid bilayer to make liposomes. If these are placed into an aerosol inhaler and sprayed into the noses of patients, some will pass through the plasma membrane of cells lining the respiratory tract. If they also pass through the nuclear envelope and insert into the host genome, the host cell will express the CFTR protein - a transmembrane chloride ion channel. Epithelial cells lining the respiratory tract are replaced every 10-14 days, so this treatment has to be repeated at regular short intervals.

Viruses have been used as vectors. If a virus that usually infects humans is genetically modified so that it encases the functioning allele to be inserted into the patient, whilst at the same time being made unable to cause a disease, it can enter the recipient cells, taking the allele with it. In 1999, a patient taking part in a trial for such a technique died, and in 2002 some trials were interrupted after several patients developed leukaemia. 

There are potential problems with using viruses as gene delivery agents:

• Viruses, even though not virulent, may still provoke an immune or inflammatory response in the patient 
• The patient may become immune to the virus, making subsequent deliveries difficult or impossible. The virus may insert the allele into the patient genome in a location that disrupts a gene involved in regulating cell division, increasing the risk of cancer 
• The virus may insert the allele into the patient's genome in a location that disrupts the regulation of the expression of other genes. 

Research is being carried out into the possibility of inserting genes into an artificial chromosome that would co-exist with the other 46 chromosomes in the target cells. 

Germ line gene therapy involves altering the genome of gametes or zygotes. Not only will all the cells of that individual altered, their offspring may also inherit the foreign allele(s). Thus, this type of therapy has the potential to change the genetic makeup of many people, the descendants of the original patient, none of whom would have given consent. There are also concerns about how the genes may be inserted - they may find their way into a location that could disrupt the expression or regulation of other genes or increase the risk of cancer.

Strict guidelines drawn up by regulatory bodies and ethics committees consider germ line gene therapy for humans to be ethically impermissible.