Animal Aid

BRED TO SUFFER - Transgenic animal disease models

In this second main section of the Bred to Suffer report, we look at transgenic animal disease models and expose their massive failure rate, as well as outlining the ethical, moral and religious concerns.

Animals have been genetically manipulated to model all of the diseases mentioned above and many more. In fact many scientists think that animals can now be engineered to represent practically any human complaint simply by adding or disabling bits of DNA.

Numbers rocketing

Many species of animals are used in this research. But mice are the favourite (at present) and their use is rocketing - up by 960% over the past 10 years, with this rate of increase predicted to continue for the foreseeable future. (40) There are already over 650 different transgenic mouse models sold commercially through catalogues, as though they were just another piece of laboratory equipment. (41) The RSPCA has expressed concern that 'GM animals may be produced simply because it is possible, and not because it is necessary'. (42)

Sheep, cattle, pigs and chickens have all been genetically modified to increase their production of milk and meat, which is already beyond the limits their bodies can bear without damage. They have also been engineered to secrete therapeutic protein products (which could be obtained more safely and cheaply from transgenic micro-organisms or plants) for human medicine; a process called 'gene pharming'. These abuses are outside of the scope of this report, but see The Gene and the Stable Door - a Compassion in World Farming Trust report, available at

Suffering at every step: creation of transgenic animals

In order to create a new strain of transgenic mice, young females are injected with powerful hormones to make them superovulate. After mating, they are killed to extract the embryos, which are microinjected with the foreign DNA. These altered embryos are then surgically implanted into many surrogate mothers, who have also been hormone-injected to assist implantation and who will later be killed before or after giving birth. Many of the resulting baby mice are malformed and die before or shortly after birth. The surviving babies have to be tested to see if they have the new gene: this can be done by saliva or faecal sampling but is more often conducted by cutting off the tips of their tails or a notch from their ears.

Massive failure rate: millions of animals killed as 'rubbish'

Only 1-10% of the baby mice will have successfully incorporated the new gene. The other 90-99% will be destroyed as 'failures'. This translates into so much killing that many of the animal technicians responsible for killing all the 'waste' animals find it traumatic and are left feeling 'physically and emotionally exhausted'. (43) While hundreds of animals are sacrificed to produce a new transgenic 'model', life for the survivors can be even worse than for the failures.

Multiple misery

A gene is not a unit, but part of an integrated system. When introduced into a foreign environment it may take effect in the wrong tissue, switch on at the wrong time, or be uncontrolled in its effects and inflict damage on non-target organs or tissues. As a consequence, there is always a likelihood that the animals will suffer unpredicted side effects in addition to the intended suffering resulting from their designer disease. For example, 'giant' mice were given a human growth hormone gene to make them bigger than normal. But they also suffered unplanned-for liver and kidney damage, grossly deformed hearts, spleens and genitalia, together with high infant mortality and a shortened life-span. (44)

Often, scientists create a 'model' by removing or disabling a gene. The resulting animals are called 'knockouts'. The effects cannot be predicted in advance. Researchers can guess, for example, that knocking out a receptor gene for thrombin (a blood-clotting enzyme) in mice will affect their control of blood coagulation. But only by creating the animals can they discover that such a deletion causes half of the altered embryos to bleed from multiple sites so that they die in the womb. (45) Other mice have been accidentally produced with no legs or with only one eye. (46)

Models of dubious value

Just as physically damaging animals results in poor 'models' of human disease, human conditions cannot be replicated in mice simply by giving them a human gene or two.

For example, none of the current 'cystic fibrosis' mouse strains accurately models the human condition, in which the major symptoms are excess mucus in the lungs, leading to lung infections. The mice, in contrast, suffer principally from bowel disorders and are clearly not a very helpful model of the disease. (47)

As already discussed, many human cancers have been 'replicated' in animals by inserting some of the genes involved. 'One might expect that these animals would mimic human symptoms, not just the genetic mutations. In fact, that is usually the exception, not the rule.' (48)

Even the industry's own Lab Animal magazine stated, 'Mice are actually poor models of the majority of human cancers.' (49) Yet the media constantly announces 'breakthrough' cancer treatments (developed in mice), raising false hopes in patients and their families. Dr Richard Klausner, director of America's National Cancer Institute commented, 'The history of cancer research has been a history of curing cancer in the mouse. We have cured mice of cancer for decades, and it simply didn't work in humans.' (50)

Fundamental flaw

The whole concept of modelling diseases on the basis of their genetic component alone is fundamentally flawed. There is indeed a genetic element to our susceptibility to many diseases, but our genes are not an automatic ticket to illness or health. In all the fanfare about the sequencing of the human genome, their contribution has been massively exaggerated. Other factors such as diet, lifestyle and environmental pollution are far more important in determining whether or not we will succumb to a particular disease at a particular time. Most of us are carrying the genes for a variety of serious diseases but are not suffering from them. This is because these 'disease genes' are not switched on unless triggered through, for instance, exposure to cigarette smoke, a high-fat diet or some other environmental risk factor. Even if one identical twin suffers from a particular disease, the other twin usually does not, (51) showing that genes alone are not enough to cause disease. (Except, of course, inherited disorders like cystic fibrosis.)

And a mouse with a gene for a human disease is still a mouse, whose 30,000 or so other genes will affect the expression and behaviour of the gene in question. The gene will perform in a completely different way in the mouse from the way it is expressed in its natural human environment. As Philip Abelson, editor of the prestigious journal Science commented, 'Are humans to be regarded as behaving biochemically like huge, obese, inbred, cancer-prone rodents?' (52)

Even when scientists think they have a 'good model' it is difficult to determine how much its attributes are due to its genes or to environmental factors. Wildly differing results have been found to occur in different laboratories using the same strains of animal in the same procedures. (53) Part of the explanation is that the stress of handling, confinement and isolation alter an animal's physiology in various ways - increasing susceptibility to certain diseases and tumours and altering levels of hormones and antibodies. (54) But new research has also shown that the brains of animals housed in standard barren laboratory cages are severely abnormal. (55) The sheer boredom of cage life literally drives them insane, causing brain damage, which must surely render much accepted research invalid. (56)

Transgenic animals also used to test poisons and carcinogens

Transgenic rats and mice are used in toxicity tests, for example, to measure the carcinogenic (cancer-causing) potential of various chemicals. The animals are designed to be genetically susceptible to cancer and it is claimed that this is beneficial to animal welfare because the tests should be less prolonged and use fewer animals than the traditional 'chronic rodent bioassay', which consumes 400-500 animals per compound. However, human hazard would be better predicted by using human cells. 'Toxicogenomics' (or pharmacogenetics) is a new technique using DNA arrays: tiny glass plates or 'chips' covered with a matrix of DNA fragments are washed over with fluorescent 'probes' that can detect which fragments have been affected by the substance in question. Thousands of chips can be processed in a matter of hours. The results are more accurate and sensitive than animal tests and (when human DNA is used) are directly relevant to humans. (57)

Legal protection inadequate

Having read this far, it will be apparent that no laboratory animals are properly protected under the 1986 Animals (Scientific Procedures) Act. But there are certain problems unique to GM animals, which require changes in the law to afford them due consideration. This is, not least, because their use, certainly on its current scale, was not foreseen when that legislation was introduced. Even the Home Office recognised this inadequacy and, in 1999, published guidance notes for project licence applicants who were intending to create or use GM animals. These notes stipulate, for example, that mice should be at least five weeks old before they can be superovulated by repeat hormone injections - a week after which they will be killed for egg/embryo harvesting. The notes also specify a maximum of 0.5cm tail-tip removal, or a maximum 15% of total blood volume removal by tail-bleeding for DNA-typing. However, DNA can be typed by faecal or saliva-sampling: clearly these more humane methods should be mandatory. The massive wastage of animals as 'failures' should be prohibited. There are methods that achieve much greater levels of success and these should be mandatory. Equally significant is that the Home Office notes still classify the production and maintenance of GM animals as 'mild' severity procedures. Yet, as we have seen, the consequences of transgenesis cannot be predicted and often seriously compromise the welfare of the resulting animals.

Ethical, moral and religious concerns

Altering the genetic material of animals raises a whole host of ethical, moral and religious questions.

In the final section of the Bred to Suffer report, we draw our conclusions about the use of animals as models for human disease and outline the other methods that are available.


40. C Blakemore, as quoted in The Guardian, 4th July 2000
41. e.g.
42. The Guardian, 21st August 2000
43. A Coghlan, New Scientist, 8th May 1999
44. T Poole in Animals, Alternatives and Ethics (eds Zutphen & Balls) Elsevier, 1997 p277-82
45. C Griffin et al, Science Vol 293 (5535) p1666, 31st August 2001
46. BUAV Campaign Report, Summer 2000
47. JR Dorin et al, Nature, 1992 Vol 359 p211-215
48. T Jacks, Science, 7th November 1997, Vol 278 p1041
49. Lab Animal, June 2001, Vol 30 no.6 p13
50. Los Angeles Times, 6th May 1998
51. New Scientist, No. 2279, 24th Feb 2001, p8
52. Science, 1992 Vol 255 p141
53. JC Crabbe et al, Science, 4th June 1999 Vol 284 (5420) p1670-73
54. New Scientist, 25th April 1992
55. J Garner, Nature, 16th August 2001 Vol 412 p669
56. J Meek, The Guardian, 28th August 2001
57. RA Lovett, Science, 28th July 2000 Vol 289 (5479) p536-37
58. Animal Experimentation, A Harvest of Shame, Hidden Springs Press, 1996

Send this page to a friend

Read about how we treat your data: privacy policy.

© Copyright Animal Aid 2014