Genetic engineering: animal welfare and

ethics.
1 INTRODUCTION
1.1 What does genetic modification of animals involve?
Genetic modification, or genetic 'engineering'a of animals involves the addition or
deletion of part of the genetic code (DNA) of an animal in order to change the animal's
characteristics (its phenotype). Change in phenotype can be brought about either
through expression of introduced DNA, or through addition, deletion or substitution of
some part of the animal's own genetic material. The aim is usually that the genetic
alteration should also be present in the germ line cells, so that the changes can be
passed on from generation to generation.
A range of methods is available for altering the genetic material. Techniques include
pro-nuclear micro-injection (available since 1980, and used in a range of species);
embryonic stem cell manipulation (in mice, and very recently, primates); and (also
recently) the ability to modify farm animals by nuclear transfer (1,2).
This paper is concerned with non-human vertebrate animals, which are genetically
modified and used
• in research and testing (e.g. in studying gene function and regulation, as models
of human disease and in toxicity testing);
• to synthesise medically important proteins;
• in developing animals which might, in future, be used as sources of organs and
tissues for xenotransplantation;
• as farm animals modified to have increased 'productivity' or disease resistance.
Latest Home Office statistics record that in Britain in 1997, 352 752 scientific
procedures performed under the auspices of the Animals (Scientific Procedures) Act
1986 involved the use of genetically modified animals, amounting to 13% of all recorded
scientific procedures (3). The majority (98%) of the animals involved were mice, and the
remainder mainly rats, pigs and sheep (see Table). Forty percent of the animals were
used for fundamental and applied biomedical research and testing, whilst the remainder
were 'breeding stock', used to maintain the genetically modified lines. The latter animals
were bred from genetically modified parents but were not used in research or testing
procedures - the regulations apply to such animals because their welfare may be
affected by the genetic modification itself.
Scientific procedures peformed on genetically modified animals in 1997
Source: Home Office(3)
Type of genetically No. procedures No. procedures Total number
modified animal for research & testing for breeding procedures
Mouse 137426 207556 344982
Rat 2115 2343 4458
Pig 266 1219 1485
Sheep 1472 0 1472
Fish 200 0 200
Domestic fowl 119 0 119
Rabbit 28 8 36
TOTAL PROCEDURES: 141626 211126 352752

1.2 Why genetically modify animals?

A range of benefits is sought from genetic modification of animals. Most work is basic or
applied medical or biological research, aimed at understanding gene function and
regulation, or studying human or animal disease. The ability to replace or alter individual
genes, or delete them, can assist in investigating the natural functions of a gene in
health and disease, the factors within the body that control it or interact with it, and the
interplay between genes and external factors, such as diet or environment. Much
experimental genetic modification is done on cells, plants or simple organisms, such as
the nematode worm, C. elegans - and such alternatives are used in preference to
vertebrate (or complex invertebrate) animals wherever possible.
The following are some examples of the benefits that can come from using genetically
modified mice and other vertebrate animals:
(i) Use of laboratory animals - mainly mice The ability to insert or delete genes in the
genomes of mice has assisted in unravelling complex developmental processes in
which genes are switched on and off, and tissues become differentiated.
Many different mouse models mimicking human disease, and/or possessing relevant
human receptor sites, are used in studying the mechanisms by which disorders (such
as immunological, neurological, inflammatory, metabolic and developmental
abnormalities and cancers) are caused, and in working towards developing more
effective treatments, such as pharmaceutical or gene therapy (4).
(ii) Use of farm animals
In Britain, a minority of scientific procedures involves genetically modified farm animals.
In 1997, around 1 860 such animals (sheep, pigs and domestic fowl) were used in
scientific research procedures, the majority for 'applied' purposes. A range of potential
and actual benefits can come from such work, including:
• better understanding of disease processes and mechanisms of resistance in
livestock such as chickens;
• production of pharmaceuticals in milk, for example production in sheep's milk of
human blood clotting Factor IX, and human α -1-anti-trypsin which can be used
to treat hereditary emphysema and cystic fibrosis (5);
 • genetic modification of cows which, in future, might allow production of milk with
enhanced nutritional quality more suitable for premature infants (an example of a
so-called 'nutriceutical', or 'functional food');
• improving the chances of successful xenotransplants in future, by genetically
modifying pigs so that, for example, they carry human complement regulating
proteins on the surface of their cells which will help to inhibit hyperacute (i.e.
immediate) rejection when pig organs are transplanted into humans.
Farm animals have also been genetically modified in attempts to increase aspects of
their 'productivity', such as growth rate or wool production, though no such work is
currently carried out in Britain. There are major difficulties in using genetic engineering
to modify such complex traits, which are controlled by a number of genes, and
conventional selective breeding currently proves more economically viable as a means
of modifying such characteristics (6).
1.3 What are the ethical concerns?
Alongside the potential benefits, genetic modification of animals raises a variety of
ethical concerns. In thinking about these concerns it may be helpful to distinguish
between:
a. fundamental moral objections either
i. to the human use of animals generally, or
ii. specifically to their genetic modification; and
b. concerns about the consequences of genetic modification (about the various
harms which might be caused, and about the goals and potential benefits of the
work).
(a) Fundamental moral objections
(i) Objections to the use of animals in general
Most people, nowadays, would agree that animals can have 'interests' (interests not to
be caused pain and suffering, for example), but there is considerable debate about
whether,
and to what extent, these interests may be forfeited for human interests. Many
arguments (about consciousness, self-consciousness, cognitive ability, language
capacity, moral sense, quality of life, and evolutionary status, for example) have been
used in attempts to find morally relevant differences (or, conversely, similarities)
between humans and animals which could justify (or preclude) treating animals as
means to human ends. None of these arguments so far has succeeded in attracting
general philosophical agreement.
There is a spectrum of opinion regarding the relative weightings that should be
accorded to human and animal interests. At the ends of the spectrum are the absolutist
positions - that human interests are always sufficiently important to outweigh animal
interests, or that they are never sufficiently important. The latter view, at its simplest and
strongest, is that if it is wrong to conduct certain experiments on humans, it is also
wrong to conduct them on animals. This moral position is not negotiable, although the
animal interests focused on may differ. For example, some animal welfarists object to
any experiment which causes animals pain and suffering, whereas some advocates of
animal rights object to all human uses of animals, whether or not pain and suffering is
involved.
(ii) Objections to the genetic modification of animals in particular
A different kind of moral objection is specifically concerned about the nature of genetic
modification. The concern may be expressed, for example, by objecting that genetic
engineering is 'unnatural', that it amounts to 'playing God', and that it 'debases animals'
by treating them as 'commodities'. A related view is that there are special moral
objections to the creation of animal strains which suffer throughout their lives because
of their genetic make-up.
(b) Concerns about the consequences of genetic modification of animals
The argument that it is acceptable to use animals as means to at least some human
ends usually appeals to the benefits of that use - that, in at least some cases, the
benefits of using animals can outweigh the harms that are caused. Here, therefore, the
main ethical concerns are about the consequences. In the case of genetic modification,
there may be concern about consequences for the welfare of modified animals, and
about the harms caused during their production. There may also be concern about the
hazards which modified organisms might pose to human and animal health and to the
environment. Or, again, there may be concern about the balance of harms and benefits
arising from genetic modification.
The remainder of this paper concentrates on a (ii) and (b) above - fundamental
objections and consequential concerns specific to genetic modification.

2 FUNDAMENTAL MORAL OBJECTIONS TO GENETIC MODIFICATION

Fundamental moral objections to genetic modification may be expressed in the
argument that genetic engineering 'fails to respect the genetic integrity' of animals,
because it involves 'mixing' of genetic material between different species and even
between different Kingdoms (between animals and plants for example)b. Anxiety,
distaste, or even revulsion, may be expressed about the 'unnatural' mixing of kinds -
about creating chimeras, about altering the 'telos' of species (so as to interfere with a
pig's 'pig-ness', for example), about crossing the species barrier, and about the mixing
of genes between humans and other animals. These moral objections may arise, for
example, from widely held philosophical or religious world-views, or from deep-seated
emotions or aesthetic values. (Associated with these fundamental objections may be
consequentialist fears that limited experiments in such areas can lead down 'slippery
slopes', perhaps culminating ultimately in ethically indefensible human eugenic
practices, or creating bizarre animals and/or treating animals as mere commodities.)
In response to these objections it can be argued that talk of 'mixing' genomes does not
reflect the nature of genetic engineering as currently practised. Although there is a
random element, present practice usually involves the relatively precise transfer of only
one or two genes - a small fraction of the genome of most recipient organisms (which
may contain upwards of 100 000 genes). Each gene codes for a specific protein, and it
is only the combined effects of expression of a multitude of genes within the living
organism that confer, say, its 'pig-ness' or its 'human-ness'. Furthermore, many genes
are conserved (are similar) between different species.
However, transfer of, for example, even a single human gene into a pig can result in
expression within that pig of something typically human - a human protein, such as
human growth hormone. The human protein may be only very slightly different from the
pig protein but nevertheless it is found naturally only in humans. Furthermore, whilst
currently it is feasible to transfer only a few genes between species, in future it may be
possible to transfer many more genes - and we therefore need to be alert to the
biological implications and related ethical concerns that might arise.
A further response to fundamental objections to genetic modification is the argument
that talk of 'transgressing the species barrier' is inappropriate, because species
boundaries are not necessarily hard and fast - species change naturally through
evolution, for example. Similarly, the characteristics of many animal and plant species,
traditionally, have been altered artificially through selective breeding, so it can be
argued that direct genetic modification is merely an extension of these traditional
breeding techniques, and thus poses no new fundamental ethical concerns. Thus, if
genetic modification of animals falls prey to charges of 'playing God', 'unnaturalness'
and 'treating animals as commodities', these same charges should be levelled at
selective breeding.
However, although species change through natural events, it is extremely difficult to
challenge species boundaries in selective breeding. Direct genetic modification is
different from both these processes in that, potentially, it offers limitless possibilities for
transferring specific genes between widely different species (although not every
combination will be viable). Genes can also be transferred from a variety of different
species into the same animal, and such genetic change(s) can be achieved within a
single generation. (Further points of comparison with selective breeding are explored
below.)
Members of the Boyd Group differ in their evaluations of these arguments. This is not
surprising. Value judgements are inevitable in ethical discussions: different people
respond to the same situations in a variety of ways, and arrive at different conclusions.
With regard to genetic engineering of animals, moreover, perception of the fundamental
issues can be complicated by concerns about what will be possible in the future, and
whether scientists can be trusted not to stray into ethically controversial or objectionable
territory. The latter concern, in particular, is fuelled by worries about the pressures
brought to bear by increasing commercialisation of such research, or from questions
about the ethics of research raised in the media.
Fundamental moral objections to the genetic engineering of animals, and concerns
about what might be possible in future, are sufficient to persuade some members of the
Boyd Group that such work should be severely limited, or abandoned altogether. The
consequences for animal welfare, moreover, may be judged so detrimental that it is
argued that the benefits of using such animals can never (or only in the most
exceptional cases) outweigh the harms.
For other members of the Boyd Group, fundamental objections do not, in themselves,
provide sufficient reason for avoiding research involving the genetic manipulation of
animals in order to achieve the kinds of benefits described above. They agree, however,
that such concerns give pause, and should be addressed, since perceptions of what is
morally acceptable are contestable and can change. At a time of rapid scientific change,
it is particularly important to listen to all reasoned arguments that highlight areas of
ethical concern.

3 CONSEQUENCES OF GENETIC MODIFICATION

Whatever their position on fundamental moral objections to genetic modification,
members of the Boyd Group are agreed that whenever animals are genetically modified
and used in science, there should be careful, detailed and critical scrutiny of the
consequences of that use. This should include consideration of the special potential of
genetic modification to cause welfare problems in animals.
3.1 Consequences for animal welfare
Here, two aspects need to be considered: the harms that can be caused during the
production of genetically modified animals, and the welfare of the resulting modified
animals. In the UK, all scientific procedures likely to cause pain and distress to animals
are regulated under the Animals (Scientific Procedures) Act 1986 and must be licensed
by the Home Office. The harms caused to the animals must be minimised, and the
procedures can only be carried out if the likely benefits are judged acceptable in relation
to the harms.
3.1.1 Production of genetically modified animals
The techniques used in genetic manipulation of animals include administration of drugs
to donor female animals, in order to induce superovulation, followed by timed matings
and collection of fertilised eggs (by killing the donor animals if mice are used, or by
laparotomy in larger animals). After they have been genetically manipulated in vitro, the
modified embryos are then implanted into surrogate mothers, by laparotomy.
Both induction of superovulation and laparotomy are established techniques which,
increasingly, are employed in selective breeding of farm and laboratory animals.
Laparotomy is carried out under general anaesthetic. Nevertheless, laparotomy can
cause post-operative pain, and superovulation can cause discomfort. In both cases,
appropriate analgesia should be administered. Preparation of surrogate mothers
involves mating them with sterile males to produce a 'pseudopregnancy', and the males
must therefore undergo vasectomy under general anaesthetic. Sometimes, the donor
female animals are mated when very young, and this can be stressful (7).
Aside from the direct effects of the techniques involved, fetal death can occur during
development in utero, and some additional deaths can occur post-natally. One study (8)
found that, in experiments involving pro-nuclear micro-injection of seven different gene
constructs into mouse embryos, 1360 out of a total of 1585 embryos survived micro-
injection (and in some cases overnight culture) and were implanted into
pseudopregnant females. 29% of the implanted embryos survived to weaning (with a
range of 21% to 42% between the seven experiments). Just under a quarter of these
pups proved to be successfully genetically manipulated (that is, 7% of the implanted
embryos, range 3% to 11% across the experiments).
Such proportions are likely to vary considerably from case to case. It is uncertain at
what stage in development fetuses can experience pain and distress, or how far the
welfare of the mother is compromised by fetal death. However, in the larger farm
animals it is known that miscarriages cause distress to the mother. Losses during
production mean that relatively large numbers of donor and recipient animals must
usually be used in order to produce a relatively low yield of genetically modified animals.
In general, there is a lack of published data on mortality rates and ages at which death
occurs during production of genetically modified animals, and there is a need for more
detailed analysis of the full range of welfare problems caused during such production.
3.1.2 Welfare of resulting animals
In some cases, genetic modification appears to have no impact on the welfare of
resulting animals; in theory, it might in some cases benefit animal welfare; and in other
cases there are certainly adverse welfare effects, which encompass a range of
severities (9,10).
Welfare can be compromised in two main ways.
(a) For research purposes, gene deletions ('knock-outs'), mutations, or defective genes
may be introduced in order deliberately to cause or simulate a wide range of genetic
diseases or developmental or gene function abnormalities.
(b) In any case of genetic manipulation, unintended deleterious or harmful side effects
can occur. Such side effects may be caused when the new genetic material is
expressed, and unpredicted physiological changes occur; or they may be caused when
the introduced DNA disrupts the function of one or more of the animal's own genes. The
latter is a result of randomness of integration of the new genetic material into the
recipient animal's genome, in particular when the pro-nuclear micro-injection technique
is used. Many such disruptions prove fatal to the developing embryo. When the effect is
not lethal, the welfare of the resulting animal can be seriously compromised (mice have
been born with deformed limbs or kidney malfunction, for example - see reviews in 2,9).
In mice, the embryonic stem cell manipulation technique now offers the possibility of
better gene targeting.
Whilst most genetic modifications tend not to benefit the animals concerned, genetic
modification might also aim to benefit animal welfare, for example by producing better
disease resistance. To take a hypothetical example, if all pigs (everywhere) were
resistant to foot and mouth disease, there would be enormous benefits for pig welfare.
In other cases, genetic modification might be welfare 'neutral'. This might be the case if
there is 'no change from the average for unmodified animals' (as for example in most,
but not all, animals modified to produce medically important proteins in their milk - 11);
or welfare is no different from that of animals produced by selective breeding (1). This
last point again raises the question of whether there are significant differences in the
effects of direct genetic modification and of conventional selective breeding.
3.1.3 Comparison with selective breeding
Both direct genetic modification and selective breeding have the potential to improve
animal welfare and to produce welfare problems. Direct genetic modification can
produce particular changes in the genetic material more rapidly than selective breeding
- yet, along the way to the desired goal (the desired phenotype), the outcomes of
genetic manipulations can be less predictable than the outcomes of selective breeding.
Selective breeding tends to build on previous results in step-wise fashion, whereas
direct modification of the genome can produce more novel, surprising and wide-ranging
phenotypic effects, with greater potential to compromise welfare, in one step. This is
particularly the case in knock-out experiments, when it can be difficult to predict the
effects of genetic deletions; and, more generally, when pro-nuclear micro-injection is
used, because, as already noted, the introduced DNA is randomly integrated into the
recipient animal's genome.
It can also be argued, however, that because the welfare changes brought about in
selective breeding may be more gradual, they can also be more insidious, and difficult
to spot at an early stage. The gradual nature of these changes can also lead society to
accept features - in some breeds of pet, for instance - which would generally be
considered unacceptable if introduced by genetic modification.
3.1.4 Improving understanding of welfare effects
Consideration of the consequences of producing and using genetically modified animals
is complicated by the difficulties involved in predicting both the welfare 'costs' to
animals, and the benefits likely to be afforded by the modified phenotypes. As in other
research areas, it is possible that potential costs and benefits can be assessed from
scientific understanding and previous experience (including the results of similar
experiments), but in this area much new ground is being covered rapidly and the
potential effects of the procedures are often uncertain. It is therefore especially
important that the justification for such work is reassessed as the work progresses: that
is, as it becomes more possible to predict likely costs and benefits from experience of
previous related work. On the welfare side, it is important that the effects of genetic
manipulations are documented in as much detail as possible - and, equally, on the
question of benefits, to record whether the desired benefits actually are achieved.
Databases on the characteristics of genetically modified animals tend not to indicate
welfare problems. Some effects - the more 'cryptic' abnormal effects, such as changes
in behaviour - may be difficult to spot, and 'tolerance' of adverse effects can depend on
the scale of animal use, and the size of animals involved. For example, some people
may have relatively little interest in the apparently minor side-effects of genetic
manipulations of a few mice used in laboratory research, whereas in larger-scale
production of farm animals there will usually be an attempt to assess all possible
effects.
There is a need for greater awareness of the welfare problems posed by abnormal
effects, improvement in surveillance and data gathering on such effects, and
improvement in data sharing. In particular, there is a need for:
• greater commitment to monitoring, collecting and reporting data on adverse side
effects of genetic manipulations. Good practice should be followed, in that
adverse effects should be looked for actively and data gathering should involve
people with responsibility for the husbandry of the animals (the animal
technicians and Named Persons). Welfare problems should be recorded in
databases on the characteristics of genetically modified animals, and journals
should require scientists reporting novel genetic manipulations to document fully
the effects on the animals of the procedures. Reporting should include aspects
such as deaths in utero occurring during production of genetically modified
animals, as well as adverse effects experienced by the resulting animals. The
latter should include any morbidity or mortality, changes in health status,
changes in weight/growth of the animals, behavioural changes, changes in
breeding success, and results of post mortem examinations of gross morphology;
 • assessment of actual outcomes of experiments - for example, analysis of the
results of experiments in order to find the proportion in which the desired
phenotype actually is achieved.
3.2 Concerns about the numbers of animals used
There has been a rapid rise in the production and use of genetically modified animals.
Comparative data are available from 1995 to 1997, and show that number of scientific
procedures performed on genetically modified animals has risen by 64% over the period
(3)c. There is currently a phase of relatively large-scale 'grasping of the opportunity' to
do new studies using the new technology. Many more projects are employing
genetically modified animals - though it might be the case that, within each individual
project, the validity of the animal models is enhanced, so that 'results may be obtained
more quickly and, ultimately, after the use of fewer animals' (1).
Arguably, the increase in use of genetically modified animals can be regarded as a
'necessity' of the moment, in that the justification is based on circumstances that did not
exist 15 years ago and which might change over time, as benefits are realised and the
need to use animals diminishes. On the other hand, with upwards of 100 000 genes in
the mammalian genome and rapid progress in sequencing the human genome, the
potential for new studies is vast, and the number of animals used could well continue to
increase in the foreseeable future.
It would be helpful if the Home Office statistics included more information on the scale
and nature of the different uses of genetically modified animals, and the extent to which
welfare is compromised, in terms both of losses in production and of effects on the
resulting animals. It is understood that the Animal Procedures Committee is considering
whether more such information might be included in the statistics (12). (See also the
recommendation of Langley and D'Silva [13] with respect to pigs produced in
xenotransplantation research.)
3.3 Concerns about safety
A utilitarian justification for producing and using genetically modified animals must take
into account potential risks to humans and other animals, as well as to the wider
environment. While this is a major concern of regulatory bodies, these vary in scope
and efficacy between countries, and much more research on safety aspects is needed
to inform their decisions.
Here, we simply note that several related categories of concern about risks to safety
need to be considered:
• concern that modified animals might 'escape' and breed with other domestic or
wild animals, so transferring the new gene(s) to these other populations;
• concern about risks from the use of retroviruses as DNA vectors during
production of genetically modified animals: e.g. risks that genes might
inadvertently be transferred to other individuals or species, or that retroviruses
might infect other organisms;
• concern about possible risks to human and animal health from consumption of
genetically modified animals and their products;
• concern about risks that drug resistance gene markers used in some genetic
engineering procedures might inadvertently be transferred and expressed;
 • 'ecological' concerns, e.g. about the wider effects of producing disease-resistant
animals;
• in xenotransplantation, concern about risks that human recipients of animal
organs might become infected with animal viral diseases, which might then infect
the wider population.
3.4 Goals and potential benefits of genetic modification of animals
The main uses of genetically modified animals include: uses in biomedical research and
testing generally; production of medically important proteins; development of
xenotransplantation; and as farm animals with increased 'productivity' or disease
resistance.
3.4.1 Biomedical research
By far the most common use of genetically modified animals is in biomedical research
(fundamental and applied research into human and animal development, gene
regulation, brain receptor chemistry, genetic disorders and development of human gene
therapy, for example) and the related development and testing of new pharmaceutical
products. These uses tend to involve small laboratory mammals - in the vast majority,
mice.
Case-by-case judgements of the necessity to use genetically modified animals are
difficult. There is a large 'middle ground' in which judgements about the relative weights
of potential benefits and harms (both to animals and humans, and in terms of safety
risks and harms which, potentially, might be caused through inappropriate application of
the results of the research) are uncertain, and especially so in the case of genetic
engineering, because of the difficulties in predicting the phenotypic effects of
manipulations. It is important to examine actual as well as potential benefits and to
reassess the benefits as work progresses. Judgements can also depend on the
perspective from which the question is asked. For example, people's judgements about
the necessity to use animals in medical research can depend on whether or not they, or
someone close to them, actually suffers from the disease or condition in question. In
this context, there is value in widening the process of ethical review, to bring a range of
perspectives to bear on such issues. In the UK, the possibility of lay representation on
local ethical review processes for animal research (see 12) is a step in this direction -
providing a wider perspective, which will act as an adjunct to the expertise and legal
authority of the Home Office inspectorate.
- Possible refinements?
Aside from the potential benefits of the research itself, it is possible that the use of
genetically modified mice might result in refinements in animal use - although, overall,
any such improvements will need to be balanced against the potential animal welfare
problems already outlined.
For example:
• although harms might be caused to animals during production of, say, a
genetically modified mouse 'model' for Parkinson's disease, further modified
animals could be obtained by breeding, and this might result in less harm than
giving a toxin or performing a surgical procedure on each animal in order to
produce a lesion in the substantia nigra of the brain and so mimic the disease -
 although, against this, the modified animals would suffer the effects of
Parkinson's disease from birth, whereas the animals with lesions caused by
toxins or surgery would suffer the disease only from the time of the intervention;
• improvements in pharmaceutical testing can arise because genetic modification
can be used to develop mice that express human genes, enabling, for example,
human receptor sites and clinical disease to be studied in mice. These improved
'models' of human disease can help to reduce the overall burden of harm caused
to animals, by allowing short-term studies of toxicity or pharmacology in
genetically modified mice as an alternative to long-term studies in conventionally
bred mice.
- Increasing use of non-human primates?
An area of special concern within biomedical research generally is the use of non-
human primates. Some recent developments related to the use of genetic modification
raise the possibility that work on non-human primates might increase.
• In cystic fibrosis (CF) research, the limitations of mouse models have led
researchers in the USA to search for primate models of the disease. Although
genetically modified mice have assisted in understanding CF and working on
therapies, their usefulness is limited because the mice do not develop the same
kind of lung disease as humans. Researchers are currently screening large
populations of primates in US National Institute of Health primate centres to find
animals carrying the recessive gene mutation which causes cystic fibrosis, and
have begun a programme to breed primates with CF (14).
• Recent reports of studies on primate embryonic stem (ES) cells raise the
possibility of future work involving genetically modified non-human primates.
Primate ES cells have now been cultured successfully and are being used in
studies aimed at understanding early human development and disease (15).
Such studies may in future involve genetic manipulation of the cells, and
production of genetically modified animals.
Although there is room for debate about the relative capacities for suffering of the
different species of non-human primate, and their capacities when compared with other
species (such as mice, or pigs, for example), it is widely acknowledged that it is difficult
to satisfy the welfare requirements of non-human primates in the laboratory and thus,
that the potential for animal suffering is increased. In addition, some species are only
available in the wild and their capture and transport can impose severe stress, as well
as a threat to the natural population. All of these considerations need to be taken into
account when evaluating the need to use non-human primates (the Boyd Group is
currently examining this area).
3.4.2 Xenotransplantation
In the UK, research aimed at developing xenotransplantation involves genetically
modified pigs as sources of organs, and some use of non-human primates as
recipients. There is much on-going debate about the issues posed by such uses of
animals, and this broad area is outside the scope of this report. The issues are explored
in recent reports from the Nuffield Council on Bioethics (16), the UK Government
Advisory Group on the Ethics of Xenotransplantation (17), the United Kingdom Interim
Regulatory Authority on Xenotransplantation (UKXIRA - 18) and the British Union for the
Abolition of Vivisection with Compassion in World Farming (13).
3.4.3 Production of pharmaceutically important proteins
Sheep, cattle and goats have all been modified to produce pharmaceutically important
proteins for human use. Although it is sometimes possible to use alternative means of
producing such proteins, this is not always the case. Most pharmaceutical proteins are
needed in large quantities that would be prohibitively expensive to produce by large-
scale culture of human cells. Transgenic bacteria, plants and animals provide
alternative approaches to overcoming the problem of scale, though often only the use of
animals can ensure appropriate biological activity. Alpha-1-antitrypsin (AAT), for
example, is required for the treatment of lung diseases such as emphysema and cystic
fibrosis. This protein is normally produced in the liver. AAT can be produced in
transgenic plants but the product does not have carbohydrate groups which would
normally be added in the liver, and it is removed from the bloodstream about 50 times
as rapidly as the natural protein. Genetically modified animals that secrete the human
protein into their milk are therefore often the only way of producing sufficient quantities
of biologically active proteins at an affordable price.
3.4.4 Increasing productivity in farm animals
As noted, at the time of writing there is currently no work in the UK aimed at increasing
productivity in farm animals by genetic modification, though selective breeding
programmes are directed at achieving the same aim.
The production of genetically modified animals with the sole objective of increasing
productivity by enhancing growth rate, or related factors such as muscling, is an area
which deserves critical attention. This is due to the severe welfare problems which have
been encountered in the past, and also because the benefits arising from such work are
generally perceived as low. Ethical review of proposals to undertake such work should
take into account the scale of potential welfare problems to the animals concerned,
based on previous experiences; the likelihood that such adverse welfare effects will
arise in the new work; and how they are to be avoided. It should be remembered that
the welfare of some farm animals may already be compromised because they have
been bred to 'over produce', and genetically engineering such animals to boost their
productivity still further is likely to lead to more animal suffering. In large-scale
production and long-term use of genetically modified animals in agriculture, welfare-
negative effects caused by genetic modification should not be tolerated, and every effort
should be made to minimise such effects. The onus should be on the creators of such
animals to prove, under practical circumstances, that welfare is not compromised at any
stage during the animal's life.

4 CONCLUSIONS
4.1 Fundamental moral objections to genetic modification of animals
Members of the Boyd Group differ in their fundamental moral perspectives on genetic
modification of animals. For some the fundamental moral objections are sufficient to
persuade them that such work should be severely limited, or abandoned altogether. The
consequences for animal welfare, moreover, may be judged so detrimental that the
benefits of using such animals can never (or only in the most exceptional cases)
outweigh the harms.
For others, the fundamental objections do not, in themselves, provide sufficient reason
for avoiding research involving the genetic manipulation of animals, although the
majority believe that such concerns give pause, and should be addressed, since
perceptions of what is morally acceptable are contestable and can change. At a time of
rapid scientific change, it is particularly important to listen to all reasoned arguments
that highlight areas of ethical concern.
4.2 Concerns about the consequences of genetic modification of animals
Whatever their fundamental moral position, members of the Boyd Group are agreed that
whenever animals are genetically modified and used in science, there should be careful,
detailed and critical scrutiny of the consequences of that use, and serious, honest,
reflection on the need to use animals at all. This effort should involve everyone
associated with the use of genetically modified animals - researchers, funding bodies,
the institutional ethical review process, journal editors and those who care for the
animals. In addition, it is important to encourage wider public discussion leading to
greater understanding of the uses of genetically modified animals and of genetic
engineering generally.
The Boyd Group's discussions have focussed on the consequences for animal welfare
and on the necessity to produce and use genetically modified animals. It is noted that
such work can raise further controversial issues - such as those posed by cloning and
patenting genetically modified animals - but these are outside the scope of this paper.
4.2.1 Animal welfare
There is a need for more detailed analysis of the welfare problems (including mortality
rates and ages at which death occurs) caused during the production of genetically
modified animals.
Directly modifying an animal's genetic material can produce unpredictable, wide-ranging
effects, and thus the potential harms and benefits of such procedures are often
uncertain. It is therefore especially important that the justification for the work is
reassessed as it progresses. To assist in this, the welfare effects of genetic
manipulations should be documented in as much detail as possible and efforts should
be made to assess success rates in achieving desired phenotypes.
In particular, there is a need for greater commitment to monitoring, collecting and
reporting data on adverse/side effects of genetic manipulations. Good practice should
be followed, in that adverse effects should be looked for actively and data gathering
should involve those responsible for the husbandry of the animals. Welfare problems
should be recorded in databases on the characteristics of genetically modified animals,
journals should require scientists reporting novel genetic manipulations to document
fully the effects on the animals of the procedures. Reporting should include aspects
such as deaths in utero occurring during production of genetically modified animals, as
well as adverse effects experienced by the resulting animals. The latter should include
any morbidity or mortality, changes in health status, changes in weight/growth of the
animals, behavioural changes, changes in breeding success, and results of post
mortem examinations of gross morphology.
A utilitarian justification for producing and using genetically modified animals must also
take into account potential risks to humans and other animals, as well as to the wider
environment. While this is a major concern of regulatory bodies, these vary in scope
and efficacy between countries, and much more research on safety aspects is needed
to inform their decisions.
In recent years, there has been a rapid rise in the production and use of genetically
modified animals. The British Home Office statistics should include more information on
the scale of the different uses of genetically modified animals, and the extent to which
welfare is compromised.
4.2.2 Necessity to use genetically modified animals
A range of potential benefits has been derived, or is sought, from genetic modification of
animals. Case-by-case judgements about the necessity to use genetically modified
animals must take into account all the various potential harms (both to animals and
humans, and in terms of safety risks and harms which potentially might be caused
through inappropriate application of the results of the research) and benefits of the
work. Such judgements, however, are difficult, and especially so in the case of genetic
modification because of the uncertainties involved. Better documentation of the effects
of manipulations, as outlined above, should help in informing such judgements. The
judgements also depend on the perspective from which the question is asked, and in
this context there is value in widening the process of ethical review, to bring a range of
perspectives to bear.
The use of genetic modification to increase productivity in farm animals by enhancing
growth rate, or related factors such as muscling, is particularly controversial. In large-
scale production and long-term use of genetically modified animals in agriculture,
welfare-negative effects caused by genetic modification should not be tolerated, and
every effort should be made to minimise such effects. The onus should be on the
creators of such animals to prove, under practical circumstances, that welfare is not
compromised at any stage during the animal's life.
a. has increased by 66%; whereas the number of animals used in fundamental and
applied studies has increased by 61% over the period 1995-7.

REFERENCES
1. Gordon, J.W. (1997). Transgenic technology and laboratory animal science. ILAR
Journal 38, 32-39
2. Mepham, T.B., Combes, R.D., Balls, M. et al. (1998). The use of transgenic animals
in the European Union. The report and recommendations of ECVAM workshop 28.
Alternatives to Laboratory Animals 26, 21-43.
3. Home Office (1998). Statistics of Scientific Procedures on Living Animals Great
Britain 1997. Command 4025. TSO: London.
4. Bedell, M.A., Largaespada, D.A., Jenkins, N.A., & Copeland, N.G. (1997). Mouse
models of human disease II. Recent progress and future directions. Genes and
Development 11,11-43
5. Wilmut, I. & Whitelaw, C.B.A (1994). Strategies for production of pharmaceutical
proteins in milk. Reproduction Fertility and Development 6, 625-630.
6. Clark, J. (1999). Genetic modification of livestock. I: Roslin Institute Annual Report
1997/8 pp 32-42. Edinburgh: Roslin Institute.
7. Hubrecht, R . (1994). Genetically modified animals, welfare and UK legislation.
Animal Welfare 4, 163-170.
8. Wight, D.C. & Wagner, T.E. (1994). Transgenic mice: a decade of progress in
technology and research. Mutation Research 307, 429-440.
9. Moore, C.J. & Mepham, T.B. (1995). Transgenesis and animal welfare. Alternatives
to