Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
4 Alternative Methods in Biomedical and Behavioral Research In recent years a great deal of attention has been focused on the use of alternative methods in animal exper~rnentation (National Institutes of Health, 1981, National Research Council, 1977; Office of Technology Assessment, 1986~. This interest has arisen in part because of a concern for the animals' welfare and the increasing costs of animal purchase ~d care. However, the term "alternative" has caused a great deal of confusion, because it implies that there are replacements for animals ~ many exper~rnental situations. In reality, there are few situations in which computer simulations, in vitro techniques, or other methods are suitable replacements for animals. By expanding what is considered to be an alternative to include reductions in the use of animals and refinements in experimental protocols that lessen the pain of the animals involved, the possibility of using alternatives increases. In addition, the replacement of one animal species with another, particularly if the substituted species is nonmammalian, can be considered another alternative method. In the following chapter we will apply this broader definition of aiterna- tives, one that arises directly from the concepts of Russell and Burch (1959) and that was used by the Office of Technology Assessment (1986) in its report Alternatives to Animal Use in Research, Testing, and Education. Scientists searching for alternative methods have Eked: How 38
ALTERNATIYE RESEARCH METHODS 39 can nonmammalian organisms, in vitro techniques, and nonbiolog- ical approaches be used? To answer that question, one must first determine how alternative approaches can provide results that are relevant to humans, axed how knowledge of more complex forms (or- gan~srns, organs, and tissues) can be inferred from research on cells and molecules. Many sirn~larities in structure and function among mammals make them obvious candidates for research applicable to humans. Rodents rats, mice, guinea pigs, and hamsters have been used because their small size makes them suitable for laboratory experi- ments and because they can be bred readily in captivity. Less well known has been the ongoing and successful use of lower vertebrates, invertebrates, and microorgan~srns in biomedical research. A variety of organisms have been used in achieving the progress that has been made in biomedical research during this century (Na- tional Research Council, 1985b). For instance, of the 135 recipients of the Nobel Prize in physiology or medicine frown 1901 to 1984, the majority of organisms used in their prize-winning work were mam- mals. On~third of the recipients were cited though for work that involved no warm-blooded vertebrates. An additional 17 were cited for work involving only humans. Twenty-five of the Nobel Prize winners based their work on a combination of different experimental subjects, including vertebrates, invertebrates, and cultures. Even higher plants have been used as sources of mode! systems. A survey of this series of awards is but one indication of the contributions made by a variety of organisms to biomedical research. RELATIONSHIPS AMONG LIFE FOBMS The principle on which the search for alternatives to mammals in research depends is that of Sanity in diversity (National Research Council, 1985b). Diversity is seen in the trillions of species that have existed or now exist, each of which has characteristics sufficiently different to enable them to be distinguished from one another. Unity is seen in common anatomical features and ~ the universality of the cell theory. For example, the development of all vertebrate embryos follows a program of blastula formation and development of ecto- derm, mesoderm, and endoderm a program that is characteristic of most invertebrates as well. Unity is also seen in the universal scheme of intermediary met am olism, which can be displayed on a chart and involves the relatively
40 USE OF LABORATORY ANIMALS small number of approximately 1,000 intermediates. The intermedi- ary metabolism of any species ~ a subset of the universal scheme, which therefore has as much generality for biochemistry as the peri- odic table does for chemistry (Sallach, 1972~. Furthermore, discov- eries in molecular biology have demonstrated the universality of the genetic code, which applies from the simplest virus Al the way to humans. The feature that makes it possible to substitute different species and other systems such as cell and tissue cultures, single cells, and nonliving systems is the presence in biology of generalizations that apply quite broadly. ANIMAL MODELS In the last decade, knowledge and use of modem and the capabil- ity of computer systems have expanded. For instance, a committee of the National Research Council (1985b) has recommended in its report Modem for Biomedical Research: A New Perspective that NTH support those proposals aimed at the development of mode} systems for specific fields of research. The committee also recom- mended that NTH regard proposals for the study of invertebrates, lower Vertebrates, microorganisms, cell and tissue culture systems, and mathematical models as having the same potential relevance to biomedical research as proposals for work on mammalian models. In addition, it recom~nended that NTH strive to make information on mode} systems readily available to the research community. The unportance of both mammalian and nonmarnmalian animal models to basic research and to the understanding of human disease is illustrated ~ the story of how researchers came to understand myasthenia gravy (Morowitz, 1986~. Myasthenia gravis is a disorder characterized by muscular weakness that can proceed to complete paralysis of some muscle groups. Perhaps the first link forged in the chain of knowledge concerning the cause of myasthenia gravis was Bernard's research with frogs on the mode of action of curare, which causes paralysis of muscles. I,ater, it was demonstrated that muscles of myasthenic patients, when st~rnulated through their nerves, fail to respond, as though they have been poisoned with curare. Fifty years ago, Loewi, Dale, Feldberg, and Vogt established in frogs and other laboratory animals that transmission of a signal from nerve to muscle was effected by the release of a chemical, acety~chol~ne, from the nerve ending (McGrew, 1985~. The concept
ALTERNATIVE RESI£ARCH METHODS 41 soon evolved that acety~choline interacted with receptor molecules on muscle where the nerve terminated. Curare blocked the action of acety~cholme and so decreased its effectiveness. Subsequently, two chemists in Taiwan isolated a powerful toxin from snake venom that paralyzed animals by binding to and blocking (inactivating) the receptor for acety~choline on the surface of muscle cells. Having a chemical that could tightly bind to the acety~choline receptor, other investigators used the toxin to obtain large quanti- ties of acety~choline receptor molecules from the electric eel Torpedo, whose electricity-produc~ng organ contains large quantities of acetyI- choline and acety~choline receptors. The acety~choline receptor, a protein, is now under intense investigation to determine its amino acid structure and its mode of response to acety~chol~ne (Kandel and Schwartz, 1985~. Receptor proteins can also produce antibodies. Indeed, in an at- tempt to make such antibodies, scientists injected the acety~choline receptor protein fir to rabbits. Unexpectedly, the rabbits developed a complete clinical picture of myasthenia gravis, which led to the recog- nition that my~thenia gravis is an autoimmune disease. In fact, it is now one of the most completely understood autoimmune diseases. For some reason, the body produces antibodies that specifically bud to and decrease the functional activity of acety~choline receptors. The search for an understanding of myasthenia gravis in hu- mans has invol`tec} frog muscles, rodent neuromuscular synapses, snake toxin, electric eel receptors, and rabbit antibodies. Additional research wall be needed before a fud cure to the disease is found research that will probably continue to use nonmarnmalian models. The preceding example illustrates how both mammalian and nonmammalian models can be used in the discovery of causer and treatments of human diseases. It also demonstrates that biomedical research requires the use of anneals, whether they be frogs, rabbits, snakes, or electric eels. ALTERNATIVES TO MAMMALS As discussed in Chapter 2, rodents (rats, mice, guinea pigs, and hamsters) and lagomorphs (rabbits) are the mammals used most of- ten in research. The use of some kinds of marrunals is limited by their size, cost, and availability en cl by the emotional attachment of humans to them. Depending on the type of research in question,
42 USE OF LABORATORY ANIMALS marnrnals sometimes can be replaced with nonmammalian verte- brates, invertebrates, microorganisms, cell and tissue cultures, and nonbiological systems, as discussed below. The necessary verifica- tion of experimental results still requires the use of some mammals in establishing a mode] system. Nonn~ian Vertebrates Nonmammalian vertebrates fish, amphibians, reptiles, and birds are rather closely related to mammals. Most of the basic properties of chern~cal transmission In nerve cells were learned by studying the frog neuromuscular junction (the synapse between nerve and skeletal muscle). Many similarities in embryonic development are present throughout the vertebrate class. Vertebrate Among the invertebrates, the largest number of species are in- sects. A great deal of research h" been conducted on insects, and much of it has provided fundamental insights into the processes of aD living things. For instance, research on the eye pigmentation of Drosophila led to the hypothesis that each gene controls a single enzyme a concept that has proved fundamental to modern molec- ular biology (Ephrussi, 1942~. Other invertebrates have also been studied; for example, research on the squid giant axon provided the basis for the concept of the ionic nature of the electrical action potential in nerve transmission (Hodgkin and Huxley, 1952~. Microorgan~Q~rm Microorganisms me acceptable as models in metabolism, genet- ics, and biochemistry, and they can sometimes serve as models of more complex systems. For instance, insights into the fundamental mechanisms of gene expression are applicable to the study of the nor- mal and pathological development of human embryos. Investigators have also shown that yeast has receptors for estrogen that appear identical in affinity with those of the rat uterus (National Research Council, 1985b). Cell and Tisme Cultures Cell and tissue culture systems are used in basic research, in
ALTERNATIVE RESEARCH AfE17IODS 43 applied research on such subjects as cancer chemotherapy, and in testing of potentially toxic substances. They are relatively easy to manipulate, and living cells can be observed with a light microscope while various components of the system are changed. For instance, one can observe the beating of cultured heart cells and note the effects of adding various chern~cals to the culture medium. Buman Tissues The use of human tissues removed at surgery or at autopsy is another alternative to the use of live anunals In research. Such material is as table at most research centers and is similar to tissues that are the targets of the research. One example of a human tissue used in research is the pituitary gland. Hormones from pituitary glands have been characterized, and in the past growth hormone was extracted and used to treat children with growth hormone deficiency. More recently, the latter prac- tice has been discontinued because a few recipients have contracted Creutzfeld-Jakob disease, apparently as a result of infection with a slow virus contained in pituitaries of infected persons (Gibbs et al., 1985~. A bioeng~neered growth hormone produced by the bacterium Escherschia cold that has recently become available elirn~nates the possibility of contracting the disease. Human placental tissue is also used. The endothelial cells har- vested from umbilical cords are used for tissue culture; the mem- branes are studied to further the understanding of human labor processes and have displaced, to a degree, experiments in sheep; and the placenta proper is used to study lamunin and other basement membrane proteins (Charpin et al., 1985~. Various other tissues are collected at autopsy for an array of research uses. For instance, Brent tissue is used to investigate the pathogenesis of cancer, and other organs are used in cardiovascular arid pulmonary research. In Vitro Systems and Mathematical Modem In vitro approaches are appropriate for some research in biology. For instance, much of the study of intermediary metabolism uses synthesized biochemicals in a manner similar to that of any chemistry experiment. Studies of reaction rates and the role of catalysts are typical examples.
44 USE OF LABORATORY ANIMALS Mathematical models can supplement experimental work or oc- casionally replace it. Such models can increase the effectiveness of experiments by defining variables and checking theories, thus making experiments on biological systems more effective and economical. Refinements The preceding discussion has focused primarily on means of re- ducing the number of animals used or replacing mammals with other organisms or with in vitro or mathematical models. The third alter- native is to refine experunental techniques to lessen the discomfort of the animals involved. Such refinements of protocol constantly occur, as researchers expand the range and uses of anesthetics, improve or eliminate restraining devices, rn~nimize invasive techniques, or use noninvasive methods to obtain the required results. Each success in this area minimizes discomfort to experimental animals. ALTERCATE METHODS IN TESTING In recent years more attention has been focused on finding aiter- natives to the use of animals in testing. Several centers are looking for alternatives to such tests as the Draize test and the LDso test. Many of the tests are described in Chapter 2. This section examines efforts to reduce the numbers of animals used, replace a mammal or vertebrate with a lower organism, or refine procedures to reduce the pain and suffering experienced by animals in these tests. A listing of alternative methods follows. · Acute toxicity tests Alternatives to the I`D50 test have been developed that use far fewer animals, with more attention being paid to morbidity and symptoms than to a statistical estimate of the median lethal close (Rowan and Goldberg, 1985~. ~ Eye and skin irritation tests Modifications to the Draize test have been developed that use smaller or more dilute doses of irritant substances and result in less trauma and distress to the test animal (GIoxhuber, 1985~. Corneal and other cell cultures might also prove to be replacements for the Draize test. In one test system being developed, fertilized chicken eggs are used to evaluate both skin and eye irritancy; the irritant is applied to the chorioallantoic membrane, which surrounds the developing embryo (L`uepke, 1985~. Research and development is in progress with the hope that one or more of these methods can be validated.
ALTERNATIVE RESEARCH AfE=ODS 45 ~ Repeated-dose chronic toxicity tests Cell cultures may be use- ful adjuncts to animals for specific target organs add tissues and thus may prove useful in routine screening tests (Office of Technology AS sessment, 19863. ~ Carcinogenicity tests A battery of short-term tests using mostly bacterial, yeast, cell-culture, ~d in vitro assays has been proposed as a predictor of the carc~nogenicity of new and existing chemicals (Leave and Omenn, 1986~. A recent study, however, has shown that four such tests have a concordance of only 60 percent with rodent carcinogenicity tests (Tennant et al., 1987~. ~ Developmental and reproductive toxicity The chick embryo has been investigated as a possible screen for teratogens, and fish and amphibian embryos, as well as other systems, might also prove useful. No single in vitro system can yet replace animal testing. Neurotoxicity tests Invertebrates can be used for some screen- ing purposes because their nervous systems are sufficiently complex and biochern~cally related to the human nervous system. The devel- oping chick embryo is being used to measure the effects of certain drugs, because the activity of the embryo can easily be observed and recorded (Norton, 1981~. ~ Mutagenicity tests The most commonly used test for mll- tagenicity is the Salmonelia/rnicrosome, or Ames test, which uses microorganisms and animal tissues (Ames et al., 1975~. However, whole-animal use is also needed in certain instances for example, to test hereditability of mutations (Office of Technology Assessment, 1986). ~ Biological screening team Cell and fixe~l-enzyme systems are used for ~creen~g whenever possible. If suitable alternative methods can be found for these tests, reductions in the number of anneals used are possible. Testing for pregnancy once relied exclusively on live annnab mostly rabbits but also truce and frogs. It is now conducted with such procedures as agglutination, radio~rnmunoassay, and enzyme ~rnmunoa~ay. The search for alternatives to the use of animals in testing is growing rapidly. Tox-Tips, a journal published since 1976 by the National Library of Medicine in Bethesda, Maryland, ~ designed specificaDy to prevent cluplication of toxicity-testing programs and to provide citations of tests that minimize the use of live Mornay. For example, the June 1986 volume included references to Mends Egg Chorioaliantoic Membrane Test for Irritation Potentials (T~uepke, 1985), ~Biopharmaceutical Test of Ocular Irritation in the Mouser
46 USE OF LABORATORY ANVILS (Ester and Wildhaber, 1985), Testing for the Toxicity of Chemicals with Tetrabymena pyriformis" (Yoshioka et al., 1985), and "An Approach to the Detection of Environmental Tumor Promoters by a Short-Term Cultured-Cell Assay" (Motile, 1984~. Despite these and other efforts, success in eliminating the use of animate from tests has been minimal. This lack of success is due both to the paucity of suitable alternatives and to regulations that require the use of specific animal tests. Although better models may become available that eluninate the use of animals, for the immediate future more realistic goad are reductions in the number of animals used, replacements of mammals with nonmamrnalian systems, and experimental refinements that lead to a reduction in the pain and discomfort of the animals being tested. ~ any discussion on alternatives, it should be noted that most research using cells, tissue cultures, or nonmammaliaD systems is conducted not as an alternative to the use of mammab but because the system best answers the question under study. Thus, a physi- ologmt may conduct experiments on an insect not as an alternative to mammab, but because there are questions to be answered about insects. For the same reason, when the molecular or cell biologist uses in vitro systems it is because they are the best ones available to answer hm or her questions. SUMMARY Although the search for alternatives to the use of animate, par- ticularly mammals, remains a valid goal of researchers, there is no chance of replacing all animals In research and testing in the foresee able future. Nevertheless, some successes have occurred in developing nonmamrnalian models, in reducing the numbers of animals used, and in refining experimental protocols to reduce the animals' pain. Such research should continue, but any hope for sudden success must be tempered by the realization that progress in this area has been slow.
