9
Applied Animal Behaviour Science, 22 (1989) 105-113 105 Elsevier Science Publishers B.V., Amsterdam -- Printed in The Netherlands Genetic Engineering and Animal Welfare* MICHAEL W. FOX Bioethics and Farm Animals, The Humane Society of the United States, 2100 L Street, N. W. Washington, DC, 20037 (U.S.A.) ABSTRACT Fox, M.W., 1989. Genetic engineering and animal welfare. Appl. Anim. Behav. Sei., 22: 105-113. Transgenic engineering of animals has already caused animals to suffer and this was not antic- ipated by the researchers involved. As more animals are subjected to transgenic intervention, the probability of animal suffering increases. Types and sources of suffering include the following. (1) Developmental abnormalities. Once the anticipated genetic changes have been accom- plished and the new animal prototypes developed as foundation breeding stock, problems are to be anticipated and have already been shown to occur in transgenic animals. (2) Deleterious pleiotropic effects. The term pleiotropism refers to multiple effects by one or more genes on the animal's phenotype. Pervasive suffering can arise from such genetic engineer- ing. This is so in the case of the U.S. Department of Agriculture's (U.S.D.A.'s) sickly transgenic pigs. Predictions and assurances as to the safety and humaneness of genetic engineering cannot be generalized from one animal species to other species. (3) New health problems. Veterinary medical knowledge will be inadequate to deal with the special requirements of animals subjected to genetic reprogramming. Research will be needed to correct health problems and associated suffering. (4) Disease resistance. Simply endowing an animal with resistance to a particular disease will not protect it from other pathogens or the stress factors and contingent suffering that make it susceptible to disease in the first place, such as transportation stress, overcrowding, etc. (5) Genetic disorders. Following the successful insertion of the gene for human insulin into diabetic mice, their diabetes was cured. However, after several weeks the mice died of excess insulin. Simply stopping the practice of inbreeding and breeding defective animals is the best prevention. (6) Productivity and suffering. Genetic engineering biotechnology to increase the productivity of farm animals (growth rates, milk or egg yield, etc.) will increase the severity and incidence of animal suffering. The term "production-related diseases" acknowledges that animal sickness and suffering are an unavoidable and integral aspect of modern livestock and poultry farming. (7) Selective Breeding. Veterinary medical research has only recently begun to recognize that so many of the health problems of companion animals are genetic in origin. Few of the disorders that affect farm animals have been looked at from this perspective. The genetic engineering of farm animals for these same reasons will, therefore, have similar consequences. (8) Suffering in the absence of disease. The stress and primary suffering that arise from the *This article is taken in part from the author's forthcoming book entitled'Silent World: Genetic Engineering -- Nature's End?

Genetic engineering and animal welfare

Embed Size (px)

Citation preview

Page 1: Genetic engineering and animal welfare

Applied Animal Behaviour Science, 22 (1989) 105-113 105 Elsevier Science Publishers B.V., Amsterdam - - Printed in The Netherlands

Genetic Engineering and Animal Welfare*

MICHAEL W. FOX

Bioethics and Farm Animals, The Humane Society of the United States, 2100 L Street, N. W. Washington, DC, 20037 (U.S.A.)

ABSTRACT

Fox, M.W., 1989. Genetic engineering and animal welfare. Appl. Anim. Behav. Sei., 22: 105-113.

Transgenic engineering of animals has already caused animals to suffer and this was not antic- ipated by the researchers involved. As more animals are subjected to transgenic intervention, the probability of animal suffering increases. Types and sources of suffering include the following.

(1) Developmental abnormalities. Once the anticipated genetic changes have been accom- plished and the new animal prototypes developed as foundation breeding stock, problems are to be anticipated and have already been shown to occur in transgenic animals.

(2) Deleterious pleiotropic effects. The term pleiotropism refers to multiple effects by one or more genes on the animal's phenotype. Pervasive suffering can arise from such genetic engineer- ing. This is so in the case of the U.S. Department of Agriculture's (U.S.D.A.'s) sickly transgenic pigs. Predictions and assurances as to the safety and humaneness of genetic engineering cannot be generalized from one animal species to other species.

(3) New health problems. Veterinary medical knowledge will be inadequate to deal with the special requirements of animals subjected to genetic reprogramming. Research will be needed to correct health problems and associated suffering.

(4) Disease resistance. Simply endowing an animal with resistance to a particular disease will not protect it from other pathogens or the stress factors and contingent suffering that make it susceptible to disease in the first place, such as transportation stress, overcrowding, etc.

(5) Genetic disorders. Following the successful insertion of the gene for human insulin into diabetic mice, their diabetes was cured. However, after several weeks the mice died of excess insulin. Simply stopping the practice of inbreeding and breeding defective animals is the best prevention.

(6) Productivity and suffering. Genetic engineering biotechnology to increase the productivity of farm animals (growth rates, milk or egg yield, etc.) will increase the severity and incidence of animal suffering. The term "production-related diseases" acknowledges that animal sickness and suffering are an unavoidable and integral aspect of modern livestock and poultry farming.

(7) Selective Breeding. Veterinary medical research has only recently begun to recognize that so many of the health problems of companion animals are genetic in origin. Few of the disorders that affect farm animals have been looked at from this perspective. The genetic engineering of farm animals for these same reasons will, therefore, have similar consequences.

(8) Suffering in the absence of disease. The stress and primary suffering that arise from the

*This article is taken in part from the author's forthcoming book entitled'Silent World: Genetic Engineering - - Nature's End?

Page 2: Genetic engineering and animal welfare

106

consequences of how they are selectively bred, raised and handled have led to dependence upon vaccines and other biologics to protect their weakened immune systems. With such artificial sup- ports, primary suffering will increase as producers are able to adopt even more intensive methods of animal production. The absence of actual disease does not mean an end to animal suffering under current farm animal husbandry conditions.

We should all ask why we need to genetically engineer farm animals, especially in these times of agricultural surpluses and chronic overproduction. Are the risks and costs of potential and actual animal suffering worth the benefits, and who will be the primary beneficiaries?

INTRODUCTION

Will genetic engineering of animals cause suffering? The answer is tha t it already has.

A major concern of all humani tar ians is whether or not genetic engineering will cause animals to suffer. The answer is being evaded by proponents of bio- technology who claim tha t "unnecessary" suffering will be avoided and exist- ing Federal animal care guidelines and regulations will take care of the prob- lem. As will be shown, transgenic engineering of animals has already caused animals to suffer and this was not anticipated by the researchers. Federal an- imal welfare regulations contain no reference to genetically engineered ani- mals and have to do with the care of animals; they have nothing to do with preventing or alleviating animal suffering following genetic reprogramming.

Acknowledging the fact tha t as more animals are subjected to transgenic intervention (or reprogramming), the probability of animal suffering in- creases, we should specify what types and sources of suffering we are dealing with. These types are as follows.

DEVELOPMENTAL ABNORMALITIES

Following gene insertion into embryos, the embryos often fail to develop and are aborted. Some may develop abnormally and die in utero, being aborted or resorbed, or are born with a variety of developmental defects, some resulting from so-called insertional mutations. These may not be manifested until later in life; hence, there can be no accurate prediction as to whether or not engi- neered animals are going to suffer and because of the nature of genetic repro- gramming, there can be no safeguards to prevent animal suffering. These prob- lems are to be expected in the initial phase of creating transgenic animals and in other genetic manipulations.

Anderson (1985) has emphasized tha t the microinjection of eggs with for- eign genes "can produce deleterious results because there is no control over where the injected DNA will re-integrate in the genome". This can mean tha t a gene, say, for a certain hormone or other protein, expresses itself in an in- appropriate tissue. He notes, "There have been several cases reported where

Page 3: Genetic engineering and animal welfare

107

integration of microinjected DNA has resulted in a pathological condition". This is one of the major reasons why he is opposed to transgenic germ line therapy in humans and it is a valid reason for concern over the welfare of animals subjected to this kind of t reatment in early embryonic life.

Once the anticipated genetic changes have been accomplished and the new animal prototypes developed as foundation breeding stock, additional prob- lems are to be anticipated. These have already been shown to occur in trans- genic animals (Palmiter et al., 1982; Hammer et al., 1985; Simon et al., 1987).

DELETERIOUS PLEIOTROPIC EFFECTS

These problems are now well recognized by biotechnologists (Hammer et al., 1985). The term pleiotropism refers to multiple effects by one or more genes on the animal's phenotype where the phenotype is the entire physical, biochemical and physiological make-up of an individual.

The well publicized health problems of the U.S. Department of Agriculture's (U.S.D.A.'s) transgenic pigs that carry the human growth gene were unex- pected, since mice and rabbits reprogrammed with this same gene did not man- ifest deleterious pleiotropic effects to anywhere near the same degree (Palmi- ter et al., 1982 ). These pigs were arthritic, lethargic, had defective vision, arising from abnormal skull growth, and did not grow twice as big or twice as fast, as was anticipated on the basis of the effects of the human growth gene in mice. These pigs had high mortality rates and were especially prone to pneumonia, the conclusion being that the genetic change had seriously impaired their im- mune systems.

This illustrates that pervasive suffering can arise from genetic engineering. It also demonstrates another principle: that a genetic change in one species may cause little apparent sickness and suffering, as in the case of transgenic mice (Palmiter et al., 1982), but this does not mean tha t the same genetic change in another species will have comparable consequences (as in the case of the U.S.D.A's sickly transgenic pigs). In other words, predictions and as- surances as to the safety and humaneness of genetic engineering cannot be generalized from one animal species to other species.

NEW HEALTH PROBLEMS

Pigs treated with somatotropin to stimulate growth require additional di- etary lysine and possibly other essential amino acids according to Kansas State University researchers (Anon., 1987 ). What this research indicates is that spe- cial diets and other health-corrective treatments will be needed following some forms of genetic reprogramming. Veterinary medical knowledge will be inad- equate to deal with the special requirements of animals subjected to genetic

Page 4: Genetic engineering and animal welfare

108

reprogramming, additional research being needed to correct health problems and associated suffering.

DISEASE RESISTANCE

Biotechnologists contend that through genetic engineering animals can be made disease resistant and this will help reduce animal suffering; shipping fever in cattle, which is a cause of considerable economic loss to the livestock industry, is a commonly used example.

However, the notion that genetically engineered disease resistance will re- duce animal suffering is scientifically naive. It is naive because it reflects a single cause (bacteria/virus) approach to disease. Simply endowing an animal with resistance to a particular disease will not protect it from other pathogens or the stress factors and contingent suffering that make it susceptible to dis- ease in the first place, such as transportation stress, overcrowding, etc.

It has been claimed that genetically engineering livestock to be resistant to various tropical diseases (such as sleeping sickness) and to extremes in climate would benefit them as well as the industry, but this would not benefit other animals, i.e., threatened and endangered wildlife species, since they would be displaced and exterminated as their habitats are taken over by the livestock industry.

GENETIC DISORDERS

There are other erroneous claims of the potential benefits of biotechnology to the animals themselves. It has been claimed that genetic engineering could be used to help cure animals of genetic disorders. This claim is made despite the almost 200 diseases of genetic origin that have been identified in highly inbred "purebred" dogs, and there are dozens that afflict other domesticated species (Patterson, 1974). Aside from the fact that it would be a poor invest- ment to correct germ line defects since disease termination in this way is not profitable (and this point is also relevant to corporate investment and involve- ment in human medicine), simply stopping the practice of inbreeding and breeding defective animals is the best prevention.

One transgenic study designed to cure an inherited disease resulted in ani- mal suffering and death. Following the successful insertion of the gene for human insulin into diabetic mice, their diabetes was cured. However, after several weeks the mice died of an excess of insulin (Seldon et al., 1987 ).

PRODUCTIVITY AND SUFFERING

Using genetic engineering biotechnology to increase the productivity and

Page 5: Genetic engineering and animal welfare

109

"efficiency" of farm animals (growth rates, milk or egg yield, etc. ) will increase the severity and incidence of animal suffering and sickness. It is already ex- tensively documented that farm animals raised under intensive confinement husbandry systems, in order to maximize production and efficiency, suffer from a variety of so-called production-related diseases (Ekesbo, 1973; Dawkins, 1980; Craig, 1981 ). The argument that if animals are suffering they will not be pro- ductive and farmers will not profit is demolished by the fact that animal sci- entists, in using the term "production-related diseases", acknowledge that an- imal sickness and suffering are an unavoidable and integral aspect of modern livestock and poultry farming (Fox, 1984). Using biotechnology to make ani- mals even more productive and efficient under these conditions will place their overall welfare in greater jeopardy than ever because the severity and incidence of production-related diseases will be increased.

SELECTIVE BREEDING

Biotechnologists argue that genetic engineering is Simply an extension of selective breeding and that since mutations (spontaneous genetic changes) occur naturally, then there is nothing morally wrong or unethical about alter- ing animals through genetic engineering. In so doing they totally ignore the scientific and medical evidence of the harmful consequences to animals of de- liberate genetic manipulation, as witness the wide variety of disorders in pure- bred dogs.

The domesticated dog, man's closest and oldest animal companion, is af- flicted by almost 200 diseases of hereditary origin. These have been produced (like the genetic disorders of farm animals) through traditional selective breeding and inbreeding procedures in order to "fix" various traits for reasons of utility and aesthetics.

Without the careful attention of their owners and often needed veterinary expertise, many dogs afflicted with these diseases would never survive to sexual maturity or, at least, be able to breed or raise their offspring successfully.

So-called "semi-lethal" inherited traits are those that can cause dogs and other domesticated animals considerable sickness and suffering or chronic dis- comfort, as the case may be. Under natural (i.e., wild) conditions, animals so afflicted would soon cease to exist due to the rigors of natural selection.

Some of these traits in dogs and other domesticated animals are actually mutations. Examples in the dog include the deformed face of the peke and bulldog, the pendulous ears of the Cocker Spaniel, the skinfolds and wrinkles of the Shar-pei, giantism in the Irish wolfhound and Great Dane, and achon- droplastic dwarfism in the Basset hound and dachshund. Other lethal and semi- lethal inherited diseases common in many breeds are expressed structurally or physiologically, such as hip dysplasia, glaucoma, epilepsy, hemophilia and im- mune system dysfunction.

Some diseases are linked with selectively bred-for mutations like the merle

Page 6: Genetic engineering and animal welfare

110

coat color of Shetland sheepdogs and albinism in the bull terrier, which are associated, respectively, with blindness and deafness.

Veterinary medical research has only recently begun to recognize that so many of the health problems of companion animals are genetic in origin. Few of the disorders that affect farm animals have been looked at from this per- spective, but the research that has been done reveals the same trend: an in- creasing incidence of health problems, many of them of genetic origin, arising as a consequence of selectively breeding for reasons of utility. The genetic en- gineering of farm animals for these same reasons.will, therefore, have similar consequences - an increasing incidence of structural and functional disorders of genetic origin that will be the cause of even more animal suffering. For those animals used in biomedical research that are subjected to genetic engineering to serve as "models" of various human disease conditions and as "tools" to test new diagnostic and treatment procedures, suffering will also be their burden (Hinrichs et al., 1987 ) and to what final end or purpose? Surely the betterment of humanity and social progress should not become ever more dependent upon the exploitation and suffering of other sentient, non-human beings.

The geneticist F.B. Hutt (1978) concluded that for every disease of farm animals that has been adequately investigated for evidence of a genetic basis, such evidence has been found and while selective breeding in poultry has been found to be successful in reducing the incidence of some diseases, "breeding efforts ended when the problem could be resolved by vaccination or medica- tion" according to poultry scientist W. Hartmann (1985). The latter empha- sized that the "need for maximum short-term improvement of economical ef- ficiency has determined selection principles in commercial (farm animal breeding) programmes which rarely left room for selection of viability".

As a consequence, the genetic basis of disease problems in farm animals has been long neglected. According to the American Veterinary Medical Associa- tion (1985); "Genetic defects hit the livestock owner where it hurts - - in the pocketbook. Kansas State University pathologist Horst Leipold is leading the research effort into the causes of disease which are estimated to cost cattle breeders between $5-10 million annually. So far, scientists have identified 88 genetic defects in cattle. Leipold discovered 12 of those himself. He has re- ceived referrals (and defective animals ) from most of the 50 states...The work is especially important because artificial insemination is so widely used in cat- tle. One bull with a genetic defect could, through artificial insemination, pass the defect on to as many as 70,000 calves".

According to geneticists, birth defects occur in at least 1% of all newborn pigs. While the causes for ~ 75% of these congenital abnormalities are un- known, a combination of environmental factors and genetic influences, espe- cially related to inbreeding and selective breeding for high performance, are most likely involved and need to be identified.

As Anderson (1978) has shown, many of the health problems that afflict

Page 7: Genetic engineering and animal welfare

111

livestock and poultry have been related to selectively breeding for certain util- ity traits such as rapid rate of growth, which has been shown to be linked with calving difficulties and higher mortality rates. Genetic engineering aimed at increasing the utility of farm animals is thus likely to intensify these already existing problems in livestock and poultry.

SUFFERING IN THE ABSENCE OF DISEASE

That biotechnologists promise that genetic engineering will also be used to improve the health/disease resistance as well as the utility of farm animals, and thus actually reduce their overall suffering, is a falsehood of considerable magnitude. The new and profitable vaccines and other biologics that are being developed by the biotech industry will help reduce the incidence of certain diseases and associated secondary suffering in farm animals. The same can be said of genetically engineering the animals themselves to be resistant to spe- cific diseases. However, the stress and primary suffering that arise from the consequences of how they are selectively bred, raised and handled, and which have led to dependence upon vaccines and other biologics to protect their weakened immune systems, will not be eliminated.

In fact, with such artificial supports (vaccines and drugs), primary suffering will increase as producers are able to adopt even more intensive methods of animal production. In other words, the absence of actual disease does not mean an end to animal suffering under current farm animal husbandry conditions. Using biotechnology to control infectious and contagious diseases, even though the animals are chronically stressed under overcrowded confinement condi- tions (which are standard treatments on large factory-scale commercial live- stock and poultry operations) will do little, therefore, to improve the welfare of farm animals and in the final analysis, may actually jeopardize it even more.

FUTURECONCERNS

The genetic engineering of animals {along with other biotechnologies, such as superovulation, in vitro fertilization, embryo transplantation, cloning and the creation of chimeras (such as the sheep with goats' heads and the half- human early stage embryos of hamster and chimpanzee ova artificially fertil- ized with human sperm) is a relatively recent development. This means that, at present, there is a total lack of evidence that the welfare of animals, sub- jected to this technology, can be guaranteed and that coincidental and contin- gent suffering to these animals will be avoided. It is only presumed and prom- ised by the biotechnology industry that animal welfare will not be placed in jeopardy and that "unnecessary" suffering will be avoided.

Can these promises and presumptions be believed, since there is clear evi- dence of suffering in neogenomic animals (animals with new genetic construc-

Page 8: Genetic engineering and animal welfare

112

tions) whose welfare has already been jeopardized by this new biotechnology? It should be emphasized that as much as we may not like it as a culture, the suffering of billions of farm and laboratory animals is justified today on the grounds of unavoidable necessity. The suffering of millions more animals is to be anticipated as new uses for animal life are discovered by the genetic engi- neers and marketed. This is likely to occur, since whatever forms of suffering arise, once society becomes economically, medically, and in other ways depen- dent upon the creation and exploitation of neogenomic animals, they will be justified on the grounds of unavoidable necessity, as is the situation today with other forms of animal utilization for the purported benefit of society.

ALTERNATIVES

There are many alternatives that are already available or could be developed so that animals do not need to be subjected to the risks of genetic engineering.

(1) Improved handling, transportation and housing/husbandry practices are alternatives to developing genetically engineered disease-resistant animals.

(2) Bacteria can be engineered to produce insulin and other pharmaceuti- cals. The genetic engineering of farm animals to perform the same task [as by reprogramming cows and sheep to produce such biologics in their milk (Simon et al., 1987) ] is to be questioned. Animals are sentient beings and can suffer, while bacteria are clearly not capable of suffering.

(3) Alternatives to genetically engineering animals as "models" for various human diseases should also be sought, if the application of research findings to human patients are not primarily preventive in nature. Prevention of ge- netic and developmental disorders in humans includes such non-animal alter- natives as genetic screening and counselling, and decontamination of environ- mental chemical pollutants that are teratogenic and mutagenic (i.e., cause developmental and genetic abnormalities).

In conclusion, we should all ask why we need to genetically engineer farm animals, especially in these times of agricultural surpluses and chronic over- production. Most of the health problems of farm animals are best addressed by making much needed improvements in overall handling, transportation, housing and husbandry. Are the risks and costs of potential and actual animal suffering worth the benefits, and who will be the primary beneficiaries? The animals will certainly not be the latter. Additionally, the genetic engineering of laboratory animals is more the domain of profitable interventive human medicine. This is surely of lesser importance than public health, environmen- tal and preventive medicine, which no amount of genetic engineering of ani- mals can advance.

In the final analysis, is the public interest and good of society really being served (biotech industry promises aside) by the genetic engineering and pat- enting of animals? Long-term social and environmental consequences also need

Page 9: Genetic engineering and animal welfare

113

to be cons idered , as well as t he e th ics of th is new technology , bu t in view of the suf fer ing t h a t will resu l t in a n i m a l s sub jec ted to ce r t a in fo rms of genet ic re- p r o g r a m m i n g pu re ly for r ea sons of ut i l i ty , t i gh t e r a n d m o r e a p p r o p r i a t e a n i m a l welfare regula t ions , or b e t t e r g u a r a n t e e s of co rpo ra t e respons ib i l i ty wi th re- gard to the a n i m a l s ' wel lbe ing are no t suff icient . T h e bes t way to p r e v e n t an- imal suf fe r ing a n d reduce i ts p robab i l i t y of occur rence before th is new indus t ry e x p a n d s fu r the r , is sure ly to pu t a m o r a t o r i u m on all a n i m a l p a t e n t i n g a n d to es tab l i sh an A n i m a l Bioe th ics Counci l c o m p o s e d of r e p r e s e n t a t i v e s f rom gov- e r n m e n t , indus t ry , a c a d e m i a a n d publ ic in t e re s t groups, pa r t i cu l a r ly those in- vo lved in a n i m a l welfare , r igh ts a n d conse rva t ion , t h a t could give the ques t ion of a n i m a l genet ic eng inee r ing a n d a n i m a l p a t e n t i n g the a t t e n t i o n and consid- e ra t ion t h a t have been so lack ing to date.

REFERENCES

American Veterinary Medical Association, 1985. Animal Health News and Feature Tips, Summer, Vol. 1,p. 1.

Anderson, L., 1978. The Chance to Survive: Rare Breeds in a Changing World. Cameron and Tayleur, London.

Anderson, W.F., 1985. Human gene therapy: scientific and ethical considerations. J. Med. Philos., 10: 275-291.

Anonymous, 1987. Feedstuffs, December 7, p. 1. Craig, J.V., 1981. Domestic Animal Behavior. Prentice Hall, Englewood Cliffs, NJ. Dawkins, 1980. Animal Suffering. The Science of Animal Welfare. Chapman Hall, London. Ekesbo, I., 1973. Animal health, behaviour and disease prevention in different environments in

modern Swedish animal husbandry. Vet. Rec., 93: 36-39. Fox, M.W., 1984. Farm Animals: Husbandry, Behavior and Veterinary Practice. University Park

Press, Baltimore, MD, 285 pp. Hammer, R.E., Pursel, V.G., Rexroad, C.E., Jr., Wall, R.J., Bolt, D.J., Ebert, K.M., Palmiter, R.D.

and Brinster, R.L., 1985. Production of transgenic rabbits, sheep and pigs by microinjection. Nature (London), 315: 680-683.

Hartmann, W., 1985. The effect of selection and genetic factors on resistance to disease in fowls - A review. World's Poult. Sci., 41: 20-35.

Hinrichs, S.H., Nerenberg, M., Reynolds, R.K., Khoury, G. and Jay, G., 1987. A transgenic mouse model for human neurofibromatosis. Science, 237: 1340-1343.

Hutt, F.B., 1978. Genetic Resistance to Disease in Domestic Animals. Constable, London. Palmiter, R.D., Brinster, R.L., Hammer, R.E., Trumbauer, M.E., Rosenfeld, M.G., Birnberg, N.C.

and Evans, R.M., 1982. Dramatic growth of mice that develop from eggs microinjected with metallothionein -growth hormone fusion genes. Nature ( London ), 300:611-615.

Patterson, D.F., 1974. A catalog of genetic disorders in the dog. In: R.W. Kirk (Editor), Current Veterinary Therapy. Saunders, Philadelphia, pp. 61-75.

Seldon, R.F., Skoskiewicz, M.J., Russell, P.S. and Goodman, M.M., 1987. Regulation of insulin- gene expression. N. Engl. J. Med., 317: 1067-1076.

Simon, J.P., McClenaghan, M. and Clark, J.A., 1987. Alteration of the quality of milk by expres- sion of sheep fl-lactoglobulin in transgenic mice. Nature (London), 328: 530-532.