In this unit, we look at the basics of choosing an animal model, including species-specific biological features and considerations for biomethodology (handling and restraint, sample collection, anesthesia and surgery, etc.)
“Selection of Animal Models,” by Michael S. Rand, DVM, is a useful starting point for understanding animal models. It’s part of a course on Research Animal Methods at the University of Arizona.
There are a variety of considerations for using animals in research, and the previous Units in this course have examined some of them. This Unit seeks to provide an introduction to some of the scientific and practical issues in choosing a particular animal to use in a particular research study, understanding that, for many beginning researchers, that decision has already been made by the laboratory director. Choosing an animal model is obviously, and primarily, a scientific matter, but there is also an ethical component. Indeed, proper experimental design in general is essential to avoid unnecessary (wasteful) use of animals. Designing an experiment requires scientific expertise, which translates into knowledge of the animal used, but also knowledge of the disease or system being studied and the methodology (laboratory techniques, but also statistics) employed. Recall also from Unit Two that regulations require, as part of protocol review, that the PI provide a justification for using animals, and for the specific species and numbers of animals to be used.
A related component, again with both scientific and ethical aspects, has been referred to as biomethodology. This is taken here to mean the various techniques involved with actual animal manipulations, such as handling, restraint, substance administration, specimen collection, anesthesia and surgery. Competency in these methods is key to obtaining meaningful data, but is also is a regulatory requirement and an important refinement, as discussed in Units Two and Three.
Wessler provides a classic definition (see text-block) of an animal model of human disease, i.e., for biomedical research. However, animals are used in other types of research, and additional components are in the definition given by Calabrese: "…a living organism in which normative biology or behavior can be studied, or in which the phenomenon in one or more respects resembles the same phenomenon in humans or other species of animals." (E. J. Calabrese. IN: Principles of Animal Extrapolation. John Wiley & Sons, NY, 1983.)
These definitions seem to miss another important use of animals referred to by Hau (see citation below) as predictive models, or those studies that use animals to evaluate the effects of some treatment or administered compound. This includes extensive use of animals in the validation of new drugs and treatments, for both humans and animals, as well as the testing of household products and cosmetics for potentially toxic effects. This category would also include the many studies that use animals for the development of new surgical techniques, instruments and equipment, as well as the use of domestic animal species in clinical trials intended to test new drugs or treatments in the target species.
Hau also defines exploratory and explanatory models. Although his discussion is oriented towards animal models for humans, a great deal of research is conducted primarily for understanding the basic biology or behavior of a particular species. Basic biological research, as compared with applied research, seeks to understand how an organism works. Far-reaching biological principles applicable to many species may be uncovered, but not necessarily, and the relevance of the work lies in increasing our understanding of the natural world. Animals are also studied in their natural setting, and this type of research also may be basic or applied; e.g., applied field research may be focused on conservation of endangered species. Finally, livestock species are used in studies that aim to improve the efficiency of food or fiber production.
Animal models of disease can be viewed in the simplest terms as spontaneous (naturally-occurring) vs. induced (experimental). Spontaneous animal models are those that involve some type of genetic variant or mutation. These models may closely resemble a human disease condition or may simply involve a genetic defect that aids in the study of normal function. One example is the muscular dystrophy that occurs in golden retrievers, which is similar in many respects to Duchenne Muscular Dystrophy of humans. This model has been used, among other things, to study gene therapy to correct the genetic defect, and there are a number of other animal models that have yielded useful information about this disease. The utility of spontaneous models rests upon recognition of the abnormality (mutation) by animal care or research staff, and many (especially mouse) models over the years have been so recognized and developed. See, for example, the listing of animal models offered by The Jackson Laboratory.
Induced animal models are those in which a normal animal is manipulated to create the condition of interest. Examples of induced models involve such things as the surgical removal of endocrine organs to study deficiency diseases or the administration of infectious disease agents. Today, manipulation of the genome is at the heart of many research areas, and these manipulations can be considered a subset of the induced animal model category (even though the genetic modification may become a permanent part of the genome) . Genetically modified organisms may be created by gene transfer (transgenics), homologous recombination (gene targeting), or chemical mutagenesis. (See, for example, the Induced Mutant Resource from The Jackson Laboratory, or the Transgenic Animal Model Core from the University of Michigan. A more general introduction to transgenics is here.
There is a natural tendency to equate the value of an animal as an experimental model with its fidelity, or how closely it resembles the biological structure of the target organism (e.g., human). In some areas of research, close homology (structural similarity) may be important. For example, if one were developing a new surgical procedure or instrument for humans, similarity of anatomical structure in the animal chosen might well be essential. However, the idea that overall anatomic or genetic (evolutionary) similarity is, in general, necessary, or even desirable, may be erroneous. Much more important is the concept of discrimination, or the ability of an animal model to reproduce a particular property desired, or to provide good predictive ability for a condition (or normal biology) of the target species. Discrimination is related to the concept of biological reductionism, which is the view that complex biological systems can be studied by teasing out components of the process in less complex systems.
Russell and Burch, who were introduced in Unit Two in the context of the "3 Rs" (replacement, reduction and refinement alternatives), also discussed the concept of the "hi-fi (high-fidelity) fallacy." This view emphasizes the importance of not placing inordinate weight on the fidelity of an animal model. The fallacy arises when one assumes, for purposes of human disease research, that mammalian models are inherently better and, by extrapolation, that the use of nonhuman primates is inherently best.
(There are those who take this argument even further, arguing that no animal provides a suitable experimental model for humans, and only human clinical trials, or other human-based research, are reliable. We’ve left this controversial issue for a Discussion Question, below.)
A key question in deciding upon an animal model should be how best to answer the specific scientific question posed. The answer may be that nonhuman primates are necessary, e.g., in many types of infectious disease research where other animals simply do not become infected; one classic example is Hepatitis B virus infection. In other cases, the greater simplicity and ability to manipulate the system in lower species may be scientifically more powerful. A good example is the tremendous amount of basic genetic information gleaned over the years from studies of the fruit fly. Another example has been the study of HIV and AIDS. Although the chimpanzee is the only animal species that becomes infected with the human immunodeficiency virus, the virus does not behave in the same way as it does in humans, and clinical signs do not develop. Nonhuman primates have their own immunodeficiency viruses, and study of these has provided valuable clues in our understanding of HIV. However, a great deal of information also has been derived from studies of feline immunodeficiency virus, a natural disease of cats that resembles HIV/AIDS in many respects.
The list of considerations above rightfully begins with scientific issues that should be at the heart of a decision on an animal model. However, those considerations are not sufficient, and practical matters also must be weighed. Nonhuman primates again provide a useful example of this. Many researchers believe that use of nonhuman primates carries a greater ethical cost than other "lower" species, and similar views held by the public means that use of these animals carries a greater risk of media attention and public criticism. Nonhuman primates, especially larger species such as macaques or baboons, are by no means easy to work with, and are dangerous not only because of their strength and sharp teeth, but because they can carry diseases transmissible to humans. Finally, nonhuman primates are expensive to purchase, and maintenance costs are high because of the need for large, sturdy cages, along with programs to provide for their "psychological well-being" (e.g., group housing and appropriate environmental enrichments).
Considerations for choosing an animal model include two other areas that deserve additional comment: microbiological and genetic characterization. Each of these has scientific relevance, but also relates to control of unwanted variation, a concept discussed in Unit Four in the context of animal husbandry and housing.
Keeping animals free of disease is relevant to both animal welfare and the scientific endeavor. Most research animals used today can be referred to as "specific pathogen free," or SPF, a term that means an animal is free of specified disease agent(s); the key to understanding this term is that the pathogens must be specified. Rodents used commonly in research are typically free of almost all known pathogens, and suppliers of these animals (e.g., Charles River Laboratories) go to great lengths to maintain this health status. For other species, such as dogs, cats and livestock species, suppliers help ensure healthy animals by vaccination of animals and parasite control programs, but certain disease agents may be endemic in the colony. It is important to investigate the specific health status of any research animal used, and reputable vendors of research animals should have such information readily available.
Microbiological characterization of the animal may go beyond control of disease agents, and can include complete control of an animal’s microflora. Gnotobiotic animals are animals that are free of disease, but also have a specifically defined flora in their intestinal tract, skin, etc. A sub-group of gnotobiotes is the axenic or germ-free animal, which is literally free of all microbial agents. Once produced, these animals must be maintained in isolators under strict aseptic conditions in order to maintain their original status. Axenic or gnotobiotic animals have been used in a variety of research areas; one example has been in the study of intestinal diseases such as inflammatory bowel disease.
Genetic characterization of an animal is another key consideration in experimental planning. Like microbiological status, inappropriate genetics of an animal or an animal colony may lead to unwanted variation, or seriously hinder the interpretation of experimental data.
Our companion animals and livestock species, as well as many animals used in research, are considered outbred; for rodents, it is standard terminology to refer to stocks of outbred animals. This means that individuals within a species are genetically distinct from one another. There may be similarities of type, but each individual has its own unique genetic make-up. Outbred stocks of rodents are similar to breeds of dogs and cats, although there is generally no attempt to exaggerate characteristics of interest. Examples of common stocks are the Swiss and Swiss-Webster mouse, and the Sprague-Dawley and Long-Evans rat.
In contrast, there are many inbred strains of rats and, especially, mice. Inbred animals are technically the result of 20+ generations of brother X sister matings, at which point each animal is almost genetically identical to each other. Examples of common mouse strains are the BALB/c and C57BL/6. Festing suggests that, "If mice or rats are being used, the use of isogenic [inbred] strains should be considered because they are usually more uniform phenotypically than commonly used outbred stocks. Experiments using such animals either should be more powerful and able to detect smaller treatment responses or could use fewer animals." (MFW Festing and DG Altman. Guidelines for the design and statistical analysis of experiments using laboratory animals. ILAR Journal 43(4), 2002.) A somewhat different perspective is offered in the article by Hartl.
Mouse genetics is an increasingly complex field. In addition to outbred and inbred animals, there are hybrids, congenics, and a myriad of transgenic and knockout strains. Two helpful sources of information are the on-line text by Silver on Mouse Genetics, and a recent issue of the ILAR Journal devoted to "Mouse Models of Human Disease."
Like microbiological status, the ability of others to reproduce one’s work requires that the genetics of the colony status be described accurately. Scientific journals are increasingly requiring that these elements be included in their publications. There are now accepted rules for genetic nomenclature of rats and mice.
It is useful to think in terms of a team approach in the choosing of an animal model and the planning of an experiment in general. The principal investigator is certainly the key responsible individual, as the scientific expert and also the supervisor of the laboratory and the research staff and students. Other key members of the team include the animal care staff, who are experienced in meeting the logistical needs of an animal-based study and can advise on potential limitations in available support. The laboratory animal veterinarian is trained in health and welfare issues associated with the use of research animals (see Unit Three), but also is knowledgeable about animal models and comparative biology.
Finally, a statistician may be an invaluable part of the team. Although many PIs have a good appreciation for statistical design and analysis, it is often useful, if not essential, to consult with a statistician in the planning of an experiment. Such an approach helps ensure the most productive science as well as the most efficient use of animals. The quotes by Festing (boxes) are from an issue of the ILAR Journal dedicated to "Experimental Design and Statistics in Biomedical Research," and the last paper in this issue ("Guidelines for the design and statistical analysis of experiments using laboratory animals") is an excellent resource for all researchers using animals in their work.
A good, basic introduction to animal models is a chapter by Hau. (Hau, J. Animal models, pp. 1-9 IN (J Hau and G Van Hoosier, eds.): Handbook of Laboratory Animal Science, 2nd ed., Volume II. CRC Press, 2003.) That chapter is actually an introduction to the volume which discusses a variety of animal models. An excellent, comprehensive review chapter on animal models is: FM Quimby. Animal models of biomedical research, Chap. 30 IN: JG Fox et al., eds. Laboratory Animal Medicine, 2nd ed. Academic Press, 2002.
The National Institutes of Health (NIH) has an excellent website (Model Organisms for Biomedical Research) that provides useful information on mice and rats, but also several important non-mammalian animal models. The ILAR Journal has published a number of excellent review articles/issues over the years on animal models, and many are available on-line. These sites, and the readings, focus mostly on animal models of biomedical research (research focused on the diseases of humans), but there is of course a great deal of research using animals in other areas.
Perhaps implicit in the discussion on choosing an animal model is the importance of being familiar with that species’ normal biology. Species differences are abundant, and may in fact be the basis for choosing a particular animal model. Providing details on the comparative biology of even the common laboratory animal species is beyond the scope of this course, but there are helpful resources. Harkness and Wagner’s text (JE Harkness and JE Wagner. Biology and Medicine of Rabbits and Rodents, 4th ed. Williams & Wilkins, 1995) is a reliable, concise source of information that includes unique anatomical and physiological features as well as normal values for physiological variables, vital signs (e.g, body temperature, heart rate), growth and reproduction, hematology, serum chemistries, etc. One of the most comprehensive sources of information is the multi-authored text published by the American College of Laboratory Animal Medicine, which includes heavily-referenced, species-specific chapters on the common laboratory animals (JG Fox et al., eds. Laboratory Animal Medicine, 2nd ed. Academic Press, 2002). Websites from the University of Iowa and the University of Arizona include some useful species-specific biological data.
A discussion of specific techniques for working with laboratory animal species is ideally left for hands-on training by the PI and/or workshops that often are offered by the institution. The American Association for Laboratory Animal Science has annual meetings that include a variety of hands-on workshops, typically including those on biomethodology. The key message for individuals who are using animals is that there is both an ethical and legal obligation to become knowledgeable about the animal species they work with, and to develop competence in the manipulations being performed. Recall that some institutions may have a training requirement for these techniques that includes completing a course and/or demonstrating competency.
Following is a list of techniques that may be important to master in preparation for conducting a research project using animals.
• Basic handling of animal, with minimal stress, for purposes of examination
• Ability to determine sex, vital signs, health status, and pain and distress
• Safe and effective restraint, with minimal stress, for purposes of substance administration or sample collection
• Administration of substances by oral, intraperitoneal, intramuscular, and intravenous routes, with attention to appropriate site, gauge of needle, and safe volumes
• Blood collection, with attention to appropriate sites, gauge of needle, safe withdrawal volumes, and need for anesthesia
• Anesthesia, including appropriate preparation, monitoring, and post-procedural care
• Surgery, including aseptic technique and postoperative analgesia and supportive care
• Euthanasia, consistent with guidelines and experimental needs
• Other specific procedures that will be used in the research
There are some on-line resources available to help with species-specific techniques, such as those through the University of Iowa, the University of Arizona, and the Australian and New Zealand Council for the Care of Animals in Research and Teaching.
The IACUC.org site contains links to a number of training sites, and training materials are listed on the Office of Laboratory Animal Welfare site. The "Electronic Zoo" from NetVet has species-specific lists of internet links on a variety of subjects. There also is a Comparative Medicine home page that has links to many sites relevant to the use of animals in research. A good introduction to experimental surgery is available through the University of Missouri.
1. What regulations are involved with the use of animals in the testing of new drugs?
2. Under what conditions can toxicity testing be done in non-animal systems? Are there examples of approved non-animal replacement alternatives?
3. Using the web site from an established rodent vendor, find the standardized nomenclature for one inbred strain of mouse. What does each of the symbols stand for?
4. What is meant by genotype and phenotype? What is the special significance of phenotypic analysis when new transgenic lines are created?
1. One of the arguments against using animals in research for human disease or testing is that animal models are inherently poor because they don’t sufficiently resemble the target species (human). What are some ethical and practical limitations in limiting biomedical research to human clinical trials?
2. Is it appropriate or necessary to use animals in the toxicity testing of household and personal-care products?
3. "Random-source" dogs are those obtained from pounds, shelters, or licensed dealers. In each case, there is minimal background information available on the animals’ health history, age, breeding, etc. What, if any, are the ethical and practical limitations to the use of random-source dogs? Under what conditions is their use justifiable?
Go to Home page <--