Man or Mouse
Making Genetically Modified Animals
What exactly are GM, mutant and transgenic animals?
Many methods exist that can alter the genetic material of laboratory animals with differing outcomes. A brief summary of terms and techniques of production is included here, before a discussion of their uses.
Genetically modified animals are animals whose DNA has been artificially manipulated in some way. This umbrella term includes cloned animals and some animals in which a gene has been mutated (mutants) to alter its function. In addition, it includes transgenic animals, which have been altered to carry a 'foreign' gene ('transgene') within their natural genome.
The main focus of this report is transgenic animals, which constitute the majority of GM work undertaken in biomedical research. Most transgenesis (the term for creating transgenic animals via the insertion of a transgene into their DNA) involves one of two basic approaches:
Knock-outs are used to study a gene's function by halting its expression, and observing the effects of its absence. This usually involves preparing a copy of the gene in the laboratory with a segment deleted and into which a foreign segment of DNA (usually a 'marker' gene to track its presence) is inserted. This new piece of DNA is then used to replace the normal, fully functioning gene in a target animal. Because the coding sequence of the gene has been disrupted, its function will be 'knocked out' because it is unable to be expressed properly, and cannot give rise to the protein for which it codes.
Knock-ins are used to study the function of a 'foreign' gene by introducing it into a target animal to observe the effects of its expression.
So how are they made?
"The Canadian Council on Animal Care classifies transgenic experiments in the second-most severe 'category of invasiveness,' with the potential to cause 'moderate to severe distress or discomfort."
There are many different combinations of techniques using different transgene constructs that are currently used to generate transgenic animals. Broadly speaking, however, all methods fit into three categories. A concise overview of these methods is given here, along with a brief consideration of proposed improvements and other uses of GM technology.
This is the oldest of the three main technologies, which involves engineering a virus to deliver the transgene into target sperm and eggs. It had until recent developments been overtaken by the other methods due to its inherent shortcomings. These include multiple insertions of the transgene into the host genome - making interpretation of results difficult if not impossible: a very low rate of production of viable transgenic offspring: and the creation of so-called 'mosaic' animals, where the transgene is carried and expressed only in some cells of the body. Furthermore, genetic material from the viruses used has been known to combine with DNA sequences from the host's genome to regenerate virulent viruses, and a phenomenon known as 'insertional oncogenesis' exists which can cause the host cells to become cancerous.[7-9]
Lately the use of a different type of virus has increased the efficiency of this technique - a development that has had some impact on animal welfare in terms of the numbers of animals surviving the procedure, and the proportion of these who express the transgene and are therefore 'useful'. There are, however, concerns surrounding size limits for the genes that can be transferred that could rule out this method for many human genes, and the other caveats described in this report still apply.
2. Pronuclear microinjection
This commonly used method became the norm due to its comparative efficiency relative to viral transgenesis, though the typical range of 'success' of between 1% and 10% still represents 90-99% wastage of the animals involved.
In this technique, female animals (mice, for example) are injected with a hormone that causes them to produce many eggs, which are fertilised by a male. The female is then killed so that her embryos can be harvested, at which point the transgenes are introduced to the embryos via injection through a small glass needle. 20 or 30 embryos are then transplanted into a number of 'pseudopregnant' female mice (these are mice that have been previously mated with castrated males to prepare their uteruses for implantation: they are clearly not pregnant following this mating, but their bodies act as though they are). About three weeks following the transplantation of embryos, comes the birth of any pups that have survived the process.
Typically 20-30% of the embryos develop, to term. 20% of these, in other words 4-6% of the initial embryos, contain the transgene. Successful uptake and function of the transgene is then confirmed via DNA analysis of these offspring, usually by cutting off the end of their tails or ears.
Successful confirmation of transgenesis at this stage is not enough. To be of any use, a suitable population of transgenic animals from which breeding can take place must be generated. This is usually done by the mating of these individuals with 'normal' mice through two generations.
There are some major caveats with this methodology. Up to 200 copies of the transgene can be randomly inserted into the host genome - an outcome with potentially disastrous consequences:
- A gene's function is highly dependent upon its environment: its site of insertion will determine if it is expressed at all and, if so, to what level.
- Multiple copies of a gene can give rise to large amounts of its protein product, which can have promiscuous effects and even be highly toxic.
- Transgene insertion can disrupt crucial host genes, rendering them useless, as well as critical 'control regions' of DNA that switch genes on and off. Again,disruption of these areas can have far reaching and catastrophic effects.
- Tissue-specific and species-specific effects can confound results. DNA molecules can be chemically modified in certain host cells of particular animals, and this has been known to result in severe developmental abnormalities, causing extreme suffering of offspring animals. Gene expression has been known to stray from its usual site, causing a gene that is usually expressed in one specific organ or tissue to be expressed in another.[10-12]
- The obvious welfare implications resulting from the above issues are compounded by breeding programmes subsequent to the experimental transgenesis, which can involve large numbers of animals and therefore high levels of suffering and mortality.[13,14]
3. Embryonic stem cell
This method utilises stem cells (cells that have the potential to develop into any kind of specialised cell later in life, such as brain, muscle, nerve etc.) from early stage embryos known as blastocysts. Blastocysts are embryos that have developed for only a few days post-fertilisation, consisting of a hollow sphere of around 100 cells and that hasn't yet implanted into the wall of the womb.
Using the ubiquitous GM mouse as an example again, embryos a few days old are removed from a freshly killed pregnant animal and the stem cells are isolated and incubated in vitro. During this time the transgene is introduced via an engineered virus or by using an electric current. The transgene is introduced in tandem with another 'marker' gene, such as a gene that confers resistance to an antibiotic. This allows those stem cells containing the transgene to be identified and selected using the antibiotic. Only those cells containing the transgene will be resistant to it, while all cells not containing the transgene will be sensitive to the drug and will not be viable. These stem cells are combined with a new 'host' blastocyst, which is then introduced into a pseudopregnant female.[16,17]
This process gives rise to progeny animals known as 'chimaeras' that are composed of some cells derived from the transgene-containing stem cells, and others from the blastocyst into which the original stem cells were introduced. Often, the mice used as a source of these cells are different colours, so that the proportion of each of the progeny mice that is actually transgenic can be estimated simply by observing their colouring. Subsequent breeding will give rise to offspring of different colours, and the potentially transgenic individuals can be selected and subjected to further testing.
One advantage of this technique is that insertion of the transgene into the host genome is not random, but directed, via specific pieces of DNA flanking the transgene that recombine with matching areas of the host's DNA. This lowers the chance of inducing mutations in host genes by random insertion. Unfortunately, the technique is still no more effective overall than microinjection in terms of final outcome, with unavoidably high numbers of wasted embryos and adult animals.
In addition, only a small number of strains of mice are suitable. It is doubtful whether these strains, with their limited and relatively invariable gene pools, can reflect the genetic complexity of human beings or serve as appropriate models for human disease. This is especially so when one takes into account the genetic changes that are known to occur in the stem cells themselves when they are removed from their natural environment of the blastocyst to be grown in vitro.
Recent developments and other uses of GM
In addition to technical improvements with the viral method mentioned previously, other attempts have been made to overcome the problems associated with transgenesis, including techniques such as sperm-mediated gene transfer (SMGT) and transposon-based gene delivery. The results have been variable. The former method involves simply incubating sperm cells in a solution of DNA containing the transgene, during which it is hoped that the sperm 'absorb' the transgene, thereby enabling them to transfer it during an in vitro fertilisation process. This promises to be very quick and easy if it can be made to be successful: unfortunately, the jury is out. Although there are reports of successful uptake of transgenes by sperm cells,[21,22] there are many examples of results showing the opposite, and some 15 years after it was first reported it has not yet been established as a reliable form of genetic modification.
As described earlier, genetic modification doesn't just comprise the transgenic techniques outlined above. Changes in DNA known as mutations occur naturally, and form the basis of evolution and natural selection as organisms change and, sometimes, by chance become more suited to their environment. Mutations have been used in genetic research for decades, both by detecting naturally-occurring gene mutations and linking them to observable traits in the organisms affected to elucidate the functions of that gene, or by deliberately inducing mutations to see what happens. In August 2005, £9 million was earmarked by the European Commission for the production of a 'library' of 20,000 mouse embryonic stem cells, each containing a specific mutant or 'knocked out' gene. Known as the EUCOMM Project, this endeavour is ostensibly to further our understanding of uniquely human diseases.
Mutations and mutagenesis
DNA mutations occur spontaneously as a result of mistakes made by parts of a cell's machinery responsible for copying and repairing that DNA, or through external agents such as chemicals, radiation and ultra-violet light in a process known as mutagenesis. In normal circumstances such 'errors' are rare and repaired as a matter of course with no ill-effects, although the fidelity of this process is not 100%, especially when the DNA is damaged at multiple sites and/or when that damage is severe. Even the smallest error can have extreme effects: people with sickle-cell anaemia, which affects more than 6000 people in the UK alone, are the victims of a 'point' mutation that results in the substitution of just one amino-acid in the haemoglobin of their red blood cells when the gene responsible is expressed.
The detection of gene mutations and the linking of them to observable characteristics ('phenotypes') of individuals carrying them have, over time, contributed greatly to the understanding of gene function. In the past decade this led to the random mutagenesis and screening of mice in large-scale projects, in the hope that some light could be shed upon the function of many novel genes discovered in, for example, the human genome project.
The mouse mutagenesis
Certain chemicals exist that are especially potent DNA-damaging agents, which have formed the basis of some large-scale research programmes in recent years involving hundreds of thousands of mice. A favourite chemical of the mouse geneticist is ENU (N-ethyl-N-nitrosourea) due to its high potency. This is injected into male mice, whose mutated sperm via subsequent matings produces an array of progeny with a huge variety of genetic mutations. Specific individuals with 'interesting' phenotypes are then chosen for further study in an attempt to identify which of their tens of thousands of genes have been mutated, and which of these may be responsible for their abnormal appearance and/or behaviour.
This approach has been extensively used in an attempt to reveal gene function in the mouse, in the hope that any information will be similar to the human situation and therefore be applicable to human disease research. Applications in this collaborative 'Mouse Mutagenesis Project'[24-27] range from studies into complex behavioural traits including drug and alcohol responses, circadian rhythms, epilepsy and psychiatric diseases,[28-30] to studies into skull and eye abnormalities[31,32] and indeed a whole host of human diseases.[33,34] Many hundreds of types of mutant mice have been created via the screening of many tens of thousands of animals,[35,36] all of which exhibit some form of deformity, behavioural abnormality and/or physiological dysfunction.
The ethical cost of these projects is unavoidably high. This is brought into clearer focus when one considers that only 1-2% of the animals are retained - these having exhibited (by the researchers' criteria) an interesting phenotypic change: 98-99% are killed immediately as a result of having nothing 'novel' to offer. Of the 1-2% 'successes', on average only one quarter will possess a new mutation for further investigation. Therefore, three quarters of the initially intriguing mice are discarded when it is discovered that their mutations are duplicates or not of sufficient interest. Factoring in additional duplicates from other institutions participating in the project, plus mutations that kill progeny animals during prenatal or postnatal development, the extra 'non-randomly mutagenised' animals used in breeding, the males subjected to the ENU mutagenesis in the first place, and the subsequent analytical procedures the ultimate animals of interest must endure, we can see that this area of research begins to look ethically troubling by anyone's standards.
Most people, then, would surely agree that this type of endeavour ought to demonstrate promise of a huge pay-off for human health, in order to balance these troubling animal welfare aspects. In reality there is a significant level of feeling that this does not constitute good science, but poorly defined and crude research that is of limited value in furthering the understanding of human genetics and disease. Of much more relevance to comprehending the function of human genes in a human environment and discerning the genetic basis of human diseases, would be human-specific studies using tissue samples, cultured cells and DNA microarrays ('gene chips'). The latter are small pieces of glass, upon which are spotted tiny amounts of DNA representing thousands of human genes. These can be 'probed' using DNA samples from healthy and diseased people, and scanned using a computer to reveal huge amounts of comparative information about gene activity relevant to specific diseases.