Model Fish:

Fish in research and toxicity testing and the need for alternatives

By Nina Mak, MS, AAVS Research Analyst

Since the early 1990s, fish have been increasingly used in biomedical research as models of human development and disease, in toxicity testing as subjects for measuring harmful effects of chemicals, and in aquaculture research to expand the farming of fish for food.^[1] It is not possible, however, to determine the exact number of fish used in the U.S. because fish are not covered by the Animal Welfare Act, the law that sets reporting requirements and minimum standards of humane care and treatment for certain animals used in laboratories.

The National Institutes of Health (NIH), though, do report that fish make up approximately seven percent of the animals used there, second only to rats and mice (91 percent).^[2] According to the USDA, 1,012,713 covered animals (i.e., not including rats, mice, and fish) were used in research in 2006.^[3] If the NIH percentages hold for the entire U.S., then the USDA figure reflects just two percent of animal use, meaning that in actuality more than 50 million animals are used every year for research, including some 3.5 million fish.^[4]

The increased use of fish in research can be traced back to the growth of interest in aquaculture.^[5] Researchers began investigating conditions that could maximize growth of fish in farms, focusing on areas such as nutrition, water quality, and treatment of diseases that flourish in farmed settings. Beginning in 1985, scientists began creating transgenic fish (fish with genes added from other species) to develop strains of fish with traits of commercial interest.^[6] Fish farming, however, carries with it many of the same animal welfare problems as other intensive farming systems, and also has the potential to harm wild fish and the environment.

As the exploitation of fish for aquaculture increased, researchers began directing their attention to the possibilities of using fish in biomedical research and toxicity testing. From the researcher’s point of view, fish provide simpler systems for the study of complex processes. Because fish are small, inexpensive, and relatively easy to house, they have become a ‘convenient’ test subject for many scientists. In addition, scientists are increasingly looking towards fish as an alternative to studies that more traditionally use mice and rats, thus, in theory, reducing animal costs by using a ‘phylogenetically lower species.’^[7]

The zebrafish, in particular, has become ‘the fish of choice’ for a variety of investigations. Since zebrafish embryos are transparent, develop outside of the mother, and grow rapidly (hatching in just three days), they are frequently used to study vertebrate development and physiology. For example, zebrafish have been used to study the development of sensory systems,^[8] and were also among a variety of fish used to determine how the kidney develops and functions.^[9]

Disturbingly, scientists are increasingly creating mutant zebrafish to identify genes that are essential for normal development. In these investigations, large numbers of zebrafish are exposed to particular mutagens that damage their DNA and then are observed for any developmental defects. Scientists try to uncover the gene(s) affected by the mutations in order to discover what genes control those processes. These approaches belie a lack of appreciation for the animal lives that are harmed and ultimately taken in the process.

There have been several large zebrafish mutagenesis projects undertaken in labs around the world,^[10] and advances in the sequencing of the zebrafish genome have fueled interest in this line of research. The Wellcome Trust Sanger Institute in the UK, for example, not only leads the zebrafish genome sequencing project but also identifies, characterizes, and distributes large numbers of mutants to researchers around the world.^[11] The Institute further offers to create mutants on demand.

In 2006, the zebrafish research community submitted a request to NIH for increased funding for efforts to expand the zebrafish model. Focusing on models of human disease, they proposed to generate 2000 independent mutant lines of zebrafish; several thousand lines of transgenic zebrafish with genes expressed in inappropriate tissues or with proteins, cells, or organs labeled with green fluorescent protein (causing them to glow and allowing visualization of their function and development); and models for large-scale chemical screening to aid in drug development.^[12] A similar large-scale project⎯the Zebrafish Models for Human Development and Disease (ZF-MODELS) ⎯is under way in Europe, running over a period of five years and funded with 12,000,000 Euros from the European Commission.^[13]

The sheer number of fish that are used in these efforts is mind boggling, and the cumulative impact that creating all these mutants has on animal welfare can only be imagined. Yet zebrafish research continues to grow. NIH set up the Trans-NIH Zebrafish Initiative (ZFIN) in 1997 to coordinate and encourage zebrafish research. In 2000, 256 labs were registered on ZFIN,^[14] compared to 527 labs and 4,336 researchers in 2007.^[15]

Zebrafish, however, are not the only fish species being exploited in biomedical research. Platyfish and swordtails, another kind of small fish, have been used as cancer models for more than 70 years.^[16] When platyfish are crossed with swordtails, creating a hybrid, they develop melanomas spontaneously. This has led researchers to study how tumors form in these fish, how chemicals and UV radiation affect the formation and growth of tumors, and the genetic factors associated with carcinogenesis. However, while human and fish tumors share certain characteristics, humans do not possess the kind of cell that becomes cancerous in these fish,^[17] thus questioning the usefulness of these investigations.

Just as fish have been used to study how cancer develops, they have also been used to determine if a chemical will cause cancer. In addition to the zebrafish, platyfish, and swordtails, the Japanese medaka is one of the most commonly used fish models for carcinogenicity testing.^[18] Fish are also used in environmental toxicity testing to measure the effects of exposure to chemicals or pollution in the environment. For example, fish were used to determine how polychlorinated biphenyls (PCBs) concentrate in the body and affect development.^[19] Similarly, fish are also used as sentinels of environmental contamination.^[20]

Small fish are used in carcinogenicity testing because they can be bred in large numbers, are cheap to maintain, have low background incidence of tumors, are sensitive to a wide variety of carcinogens, and develop tumors soon after exposure to a test substance. As such, it is faster and cheaper to conduct the tests on fish rather than on mammals.^[21] However, just as the standard carcinogenesis assays cause a significant amount of pain and distress for rodents, it is likely that fish suffer as well.

The number of fish used and subjected to painful or distressful procedures is further increased by recent interest in developing transgenic fish for carcinogenicity and ecotoxicology testing. Scientists have attempted to create transgenic medaka, mummichog, and zebrafish to reveal when exposure to a contaminant occurs or how that contaminant damages DNA and affects development.^[22] Like biomedical research, these studies also involve mutation assays.

While the use of fish is considered a refinement alternative to the use of mammals, the danger is, because fish are so small and cheap, and because we do not have a good intuitive sense of how they feel or suffer, we will consider them expendable, not value their lives, and use them in tremendous quantities. Indeed, as evidenced by the research project describe above, this is already happening.

It is important to consider that experiments are stressful for fish and can cause pain and distress. For some time, there had been debate in the scientific community about whether or not fish can feel pain. However, while some scientists continue to dispute whether fish have the capacity to experience pain in a human sense, all generally agree that fish “are capable of behavioral, physiological and hormonal responses to stressors (including noxious stimuli) which can be detrimental to their well-being.”^[23] Even routine handling can cause morbidity and mortality in fish, and it can take a long time for fish to recover from stress.^[24] Thus, we have a responsibility to be concerned about their welfare.

Despite this, researchers are subjecting countless fish to procedures that are considered extremely painful for mammals. In addition to the studies described above, the lethal dose 50 test (LD₅₀), for example, is still being conducted in fish even though it is generally frowned upon in mammals.^[25] Little attention is being paid to identifying humane endpoints in fish models to reduce pain and distress, and in the U.S., researchers are not required to consider alternatives or minimize pain and distress because fish are not protected under the Animal Welfare Act. Furthermore, lack of standardization and harmonization of fish test protocols and welfare monitoring lead to unreliable results and even greater numbers of fish being used.

The scientific community should not rest comfortably just because it is using fish instead of mammals. Fish, just as with any other animal, should be afforded the principles of the 3Rs ( reduce, refine, replace). It is necessary to develop alternatives that reduce the number of fish used and the levels of pain and distress they experience, as well as alternatives that replace the use of fish entirely.

Fortunately, researchers in Canada and Europe are beginning to do just that. The Canadian Council on Animal Care recently published “Guidelines on the Care and Use of Fish in Research, Teaching and Testing,” which provides recommendations on issues such as nutrition, housing, and handling of fish.^[26] In addition, ECVAM validated an approach for ecotoxicological testing developed by the European Commission’s Joint Research Centre that substantially reduces the number of fish used.^[27] Other researchers are working to expand the use of fish cell lines and mammalian cell lines to completely replace fish in carcinogenicity assays and toxicity testing.^[28]{28]

It is important to remember that the goal is to end the use of all animals in research, testing, and education. The use of fish should not provide a justification for continuing to rely on animal models. Efforts to find alternatives to the use of fish are beginning and must be strongly encouraged.

Resources

[1] Borski, R.J., & Hodson, R.G. (2003). Fish research and the Institutional Animal Care and Use Committee. ILAR Journal, 44(4), 286-294.
[2] NIH Animal Research Facility Orientation Course. Retrieved Nov. 2007, from http://oacu.od.nih.gov/powerpoint/Facilityguide.pdf.
[3] USDA. (2006). Annual Report. Retrieved Nov. 2007, from http://www.aphis.usda.gov/animal_welfare/downloads/awreports/awreport2006.pdf.
[4] These numbers do not include fish embryos produced and used for research, as these are not covered by NIH guidelines. Office of Laboratory Animal Welfare. (2007). PHS Policy on Humane Care and Use of Laboratory Animals. Frequently Asked Questions. Retrieved Nov. 2007, from http://grants.nih.gov/grants/olaw/faqs.htm#App_4.
[5] Borski, R.J., & Hodson, R.G. (2003). Fish research and the Institutional Animal Care and Use Committee. ILAR Journal, 44(4), 286-294.
[6] Winn, R.N. (2001). Transgenic fish as models in environmental toxicology. ILAR Journal, 42(4), 322-329.
[7] Borski, R.J., & Hodson, R.G. (2003). Fish research and the Institutional Animal Care and Use Committee. ILAR Journal, 44(4), 286-294.
[8] Moorman, S.J. (2001). Development of sensory systems in zebrafish (Danio rerio). ILAR Journal, 42(4), 292-298.
[9] Reimschuessel, R. (2001). A fish model of renal regeneration and development. ILAR Journal, 42(4), 285-291.
[10] Epstein, F.H., & Epstein J.A. (2005). A perspective on the value of aquatic models in biomedical research. Experimental Biology and Medicine, 230, 1-7.
[11] Wellcome Trust Sanger Institute. (2007). Zebrafish Mutation Resource. Retrieved Nov. 2007, from http://www.sanger.ac.uk/Projects/D_rerio/mutres.
[12] Zon, L.I. (2006). The Zebrafish & Disease Project: Zebrafish as a model system to study and cure human diseases. White Paper. Retrieved Nov. 2007, from http://www.nih.gov/science/models/zebrafish/reports/ZonFundingZebrafish052206.pdf.
[13] ZF Models. Retrieved Nov. 2007, from http://www.zf-models.org/documents/ZF-MODELS_brochurepdf.
[14] Lardelli, M. (2000). Zebrafish – Do we need another vertebrate model?. ANZCCART News, 13(4), 1-3.
[15] Zebrafish Information Network. Retrieved Nov. 2007, from http://www.zfin.org.
[16] Walter, R.B., & Kazianis, S. (2001). Xiphophorus interspecies hybirds as genetic models of induced neoplasia. ILAR Journal, 42(4), 299-321.
[17] Walter, R.B., & Kazianis, S. (2001). Xiphophorus interspecies hybirds as genetic models of induced neoplasia. ILAR Journal, 42(4), 299-321.
[18] McHugh Law, J. (2001). Mechanistic considerations in small fish carcinogenicity testing. ILAR Journal, 42(4), 274-284.
[19] NIEHS News. (1998). Fishing for answers. Environmental Health Perspectives, 102(3). Retrieved Nov. 2007, from http://www.ehponline.org/docs/1994/102-3/niehsnews.html.
[20] McHugh Law, J. (2001). Mechanistic considerations in small fish carcinogenicity testing. ILAR Journal, 42(4), 274-284.
[21] McHugh Law, J. (2001). Mechanistic considerations in small fish carcinogenicity testing. ILAR Journal, 42(4), 274-284.
[22 Winn, R.N. (2001). Transgenic fish as models in environmental toxicology. ILAR Journal, 42(4), 322-329.
[23] Canadian Council on Animal Care. (2005). Guidelines on the Care and Use of Fish is Research, Teaching and Testing. Retrieved Nov. 2007, from http://www.ccac.ca/en/.
[24] Canadian Council on Animal Care. (2005). Guidelines on the Care and Use of Fish in Research, Teaching and Testing. Retrieved Nov. 2007, from http://www.ccac.ca/en/.
[25] Johansen, R., Needham, J.R., Colquhoun, D.J., et al. (2006). Guidelines for health and welfare monitoring of fish used in research. Laboratory Animals, 40, 323-340.
[26] Canadian Council on Animal Care. (2005). Guidelines on the Care and Use of Fish in Research, Teaching and Testing. Retrieved Nov. 2007, from http://www.ccac.ca/en/.
[27] European Commission. (2005). JRC-developed toxicology test set to reduce experimentation on fish. Retrieved Nov. 2007, from http://cordis.europa.eu
[28] Castano, A., Bols, N., Braunbeck, T., et al. (2003). The use of fish cells in ecotoxicology. The report and recommendations of ECVAM Workshop 47. Retrieved Nov. 2007, from http://altweb.jhsph.edu/publications/ECVAM/ecvam47.htm.
[29] Gulden, M., Morchel, S., & Seibert, H. (2004). Comparison of mammalian and fish cell line cytotoxicity: Impact of endpoint and exposure duration. Aquatic Toxicology, 71(3), 229-236.

Mak, Nina. (Winter 2008). AV Magazine. Pages 2-5.

Print View

Web ViewText Size: