Josh P. Roberts has an M.A. in the history and philosophy of science, and he also went through the Ph.D. program in molecular, cellular, developmental biology, and genetics at the University of Minnesota, with dissertation research in ocular immunology.
Some questions are best answered in large-animal models. But for researchers used to working with rodents or fish, the idea of experimenting on a companion animal, farm animal or non-human primate may be a bit daunting.
Questions like, ‘What is the most promising, informative model?’ must be weighed against such mundane considerations as regulatory requirements, budget, space, availability of subjects and ease of manipulation. By the time these various variables are taken into account, you're going to need a compelling reason to leave the mouse room behind. Here, we walk you through some of the arguments for and against.
Practical considerations
One of the first things to think about is availability of the animals. Mice and rats are plentiful in the research world. Nonhuman primates—which are perhaps the best scientific model for many translational applications—are fairly hard to come by, and they pose some unique ethical issues, notes Randy Prather, director of the National Swine Resource and Research Center. Most other commonly used large-animal models, on the other hand “are in-between.”
Size is another concern. “The biggest thing is cost and space—if you’re at a hospital, and you have a domestic pig, it may be kind of hard to get [it] in your animal facility,” Prather says. “We do work with miniature pigs, and those fit in a lot easier, but a pig is still going to be a lot more expensive than a mouse.”
The diminutive size of rodents means their housing requirements are low: Cages are small—with a recommended minimum of six to 15 square inches per group-housed mouse (depending on the size of the animals) [1], for example—and can be stacked, meaning that a small room is often sufficient to meet the needs of multiple projects. Contrast this to the minimum requirements for cats (three to four square feet), dogs (eight to 24 square feet) and swine (eight to 52 square feet), for example. Bedding, food and other size-related costs, such as anesthesia, are commensurate. Instrumentation, too, is easier with mice: Smaller animals can be imaged on smaller MRIs and PET/CT scanners, for example, rather than needing more expensive built-for-humans instrumentation.
Yet the benefits of larger-sized animal models can’t be discounted. “Say you have a cardiac device. You’re not going to make it small enough to fit into a rodent, and get information that’s going to be helpful for people. Size of the animal is oftentimes critical to the device that you’re testing,” says veterinary surgeon Theresa Fossum, vice president of research and strategic initiatives at Midwestern University. In addition, formulating drugs for small animals may not always translate well to human-sized animals.
The same can be said for surgical and other procedures. “If you’re dealing with eyes, for example, and you want to do cell-based therapy? You want to measure blood flow related to cardiovascular disease? Or do xenotransplantation, taking organs and putting them into humans? … Mice are pretty small. Pigs are probably a good size,” Prather says.
In fact, the U.S. Food and Drug Administration (FDA) is moving toward requiring testing on a large, nonrodent mammal—they tend to recommend pigs or dogs—prior to going into humans, because so many drugs are successfully tested in mice only to fail in the clinic, Prather says. “If you have an animal that’s similar in size, you can get dosage, safety studies, method of delivery.”
Anatomy, physiology, genotype and phenotype
But it’s not just size that makes one model more commensurate, or easier to work with, than another. Each discipline has its own favorite models, often for very sound scientific reasons. For example, the fact that armadillos were susceptible to infection by Mycobacterium leprae led to their being used to discover a treatment for leprosy [2]. And a recent paper concluded that ferrets could serve as a "useful expermental animal model" for studies of the eye, "filling the gap between rabbit and monkey animal research models for researchers in the field of ophthalmology” [3].
Other species have been studied more extensively. Pigs, for example, produce the correct post-translational modifications on proteins such as human coagulation factors XIII and IX (used to treat hemophilia), which can then be secreted in the milk, making them ideal bioreactors. The dermatology of pigs is similar enough to humans that swine are often used for wound-repair studies. Even more impressive is the phenotype of animals with a mutated CFTR gene—which in humans causes cystic fibrosis (CF). Pigs with a mutated CFTR gene get full-blown CF, including blockages in the gall bladder, pancreatic duct and vas deferens; liver lesions; and (if they live long enough) lung disease. “When you make that same mutation in mice, you have a defective chloride ion transporter, and that’s it,” says Prather.
Nonhuman primates are the closest to humans, making them in many respects the nearest we can come to an ideal animal model. For example, macaques are used extensively for AIDS research, because they are susceptible to infection with simian immunodeficiency virus (SIV). Other examples of research with nonhuman primates, undertaken at the eight U.S. national primate-research centers, include: reproductive and developmental biology, including fertility treatments; infectious disease research (including on Ebola virus); vaccine development and biodefense; chronic diseases, such as diabetes and obesity; imaging and pathology; and programs such as conservation and studies of social societies. “We also have a fairly large neuroscience program,” says Joyce Cohen, assistant director of the Division of Animal Resources at the main center of the Yerkes National Primate Research Center. “You can do a lot of cognitive testing when you’re looking at neurological disorders, [and get at questions] that you can’t necessarily ask in other species.”
But because primates are so close to humans, they also present the highest hurdles as research subjects, whether at the national centers or other facilities. “It’s exponentially more expensive than working with rodents,” Cohen says. “You’re going to have expenses related to not only daily husbandry and veterinary care. There are also components of colony management and behavior management that play into the costs associated with working with primates that you might not have with other species.” There are requirements like extra enrichment, socialization and opportunities for species-like behavior that need to be closely monitored and managed, and there is much more detailed regulatory oversight.
Natural models
Dogs, cats and to a lesser extent other larger animals may suffer from the same diseases as humans, and for the same reasons: They have the same or similar genetic mutations, they share the same environment and they live a relatively long time. “They live long enough that they get cancer. They have a lot of cardiac problems, both congenital and acquired cardiac disease,” and so forth, says Fossum. She thinks that much research—especially efficacy testing—can be done on such “naturally occurring animal models” and is spearheading an effort to establish the infrastructure necessary to enable that on a large scale. For starters, she says, “we’re looking for funding for this consortium between Midwestern University, Tgen [Translational Genomics Research Institute] and Mayo Clinic, to establish a dog-cancer registry.”
If you’re looking to establish your own large-animal program, the resources are there to move your research forward. Just be sure to match your model to the scientific questions at hand. And when in doubt, consult the experts.
References
[1] Guide for the Care and Use of Laboratory Animals: Eighth Edition. Washington, D.C.: The National Academies Press, 2011. [PDF]
[2] NIH Office of Science Education, “Armadillos and their role in treating leprosy,” available at http://science.education.nih.gov/animalresearch.nsf/Story1/Armadillos+and+Their+Role+in+Treating+Leprosy
[3] Sakamoto, Y, et al., “Ferrets as a new experimental animal model for corneal endothelial research,” Anim Eye Res, 31:3–11, 2012. [Article Link]