Genetic engineering has been performed for centuries in animals and plants by selective breeding. This enhances particular genetic traits based on outward appearance, by choosing, for example, which boars to mate which sows to develop, over many generations, leaner pig meat. From the early beginnings in the 1970's, however, it has now become possible to manipulate specific genes at a molecular level, using laboratory procedures on material taken from living organisms, which can be replaced in the organism, or put into a different one. In principle, this ought to be much more specific than selective breeding, but the uptake of the relevant modified gene is often quite low, particularly in animals. It also allows the creation of "transgenic" organisms, where a short section of genetic material from an unrelated species can be introduced into another. Genetic engineering is being developed in animals:

With some notable exceptions, it has not proved as straightforward to produce "transgenic" animals as originally thought, but various manipulations have been performed, many of which are still at a fairly early stage of development. Human growth hormone was introduced in mice and pigs in early experiments, but many problems were found and this work has mostly been discontinued. In general, attempts to genetically engineer farm animals to enhance production - more specifically and rapidly - have not been promising. At present the best prospects for this type of "production" genetic engineering seem to be mainly in fish. Most of the applications of genetic engineering in animals have been in finding novel uses for the animal.

By far the largest of these has been in producing transgenic mice to "model" human diseases. Sufficient similarity has been found that once a human gene has been identified, one of the easiest ways to find out its function is to disable the equivalent gene in a mouse and observe the effect. Alternatively mice have been produced which contain a genetic defect which is likely to produce the symptoms of a human disease, like cycstic fibrosis and many forms of cancer. The first and most famous (or infamous) of these was the Harvard "oncomouse", a mouse engineered to develop a cancer for use in testing potential cancer drugs. This caused immense controversy when the mouse became the subject of a patent application. Partly this was over the patenting of an animal as such, and partly because of the inevitable suffering which the animal would undergo. A long-awaited hearing at the European Patent Office in Munich in late November will decide whether a series of moral and ethical objections to the patent will be upheld.

Generally less controversial has been the novel idea of genetically engineering mammals so that in their milk they produce proteins of potential medical benefit as pharmaceutical products. The leading example of this is the production of alpha-1-antitrypsin in the milk of a sheep called Tracy and her progeny in Edinburgh. Sufferers from the lung disease emphysema have a deficiency of this protein, and this method is being developed as a convenient source of it in fairly large quantities, which appears to have no ill effects on the sheep and which has the prospect of being safe from the cross contamination which can arise if human blood is used as the source. The preliminary clinical trials are awaited. Other applications are being attempted using the same basic idea in Edinburgh and elsewhere.

A third novel area is to xenografting - the potential use of animal organs as transplants into humans, such as hearts and kidneys - where there is a significant shortfall between patients and realistic potential donors. By genetically engineering a pig's heart with a human gene, researchers at Cambridge hope to produce a "layer" of proteins around the heart which would send the signal "human". This might be able to convince the human body not to put into action the rapid rejection of tissue belonging to another species. No one knows whether this will be successful. There would still remain a number of other problems to be solved, including the need to supress the body's slower rejection that is familiar in human-human heart transplants.

Although the basic coding system is the same in all organisms, the fine details of gene control often differ. A gene from a bacterium, say, will often not work correctly if it is introduced unmodified into a plant or animal cell. The genetic engineer must first construct a transgene - the gene to be introduced; this is a segment of DNA containing the gene of interest and some extra material that correctly controls the gene's function in its new organism. The transgene must then be inserted into the second organism.

Making a transgene: All genes are controlled by a special segment of DNA found on the chromosome next to the gene and called a promoter sequence. When making a transgene, scientists generally substitute the organism's own promoter sequence with a specially designed one that ensures that the gene will function in the correct tissues of the animal or plant and also allows them to turn the gene on or off as needed. For example, a promoter sequence that requires a dietary "trigger" substance can be used to turn on genes for important hormones in animals; the animal would not produce the new hormone unless fed the appropriate trigger.

Inserting the transgene in animals: Copies of the transgene are usually injected directly into a fertilised egg which is then implanted in the female's reproductive tract. However, it is difficult to control where in the chromosome the transgene is inserted, and this sometimes causes variations in the way the gene is expressed. As well, the process is demanding and has a low success rate. Currently less than 5 per cent of injected embryos result in offspring with the gene integrated into their DNA and able to be passed on consistently to successive generations. Researchers are therefore investigating new methods of gene transfer.