Current and new regulations, including requirements to implement the 3Rs (replacement, refinement, and reduction), along with public desire to reduce the number of animals used, could potentially impact the speed and quality of biomedical research, noted Roberto Caminiti, professor of physiology at Sapienza University of Rome and session chair. Panelists discussed the role of animals in neuroscience research, benefits and costs (administrative, economic, social, animal welfare), mechanisms to maintain public confidence, and the impact of the laws, regulations, and policies on animal-based research in neuroscience (key points are summarized at the end of the chapter in Box 4-2).
RODENTS IN NEUROSCIENCE RESEARCH
Rodents are the dominant mammalian animal species used in neuroscience research, said Bill Yates, professor of otolaryngology and neuroscience at the University of Pittsburgh, but the Animal Welfare Act excludes mice and rats, so the exact number used in the United States is not available. The number of higher animals used is known because the U.S. Department of Agriculture (USDA) requires research institutions to submit an annual report of the number of animals used. The use of most animal species tracked by the USDA has declined over the past decades (Figure 4-1). Only the use of non-human primates has increased slightly. Yates noted that during this time, National Institutes of Health (NIH) grant funding has increased tremendously, suggesting that if more animal research is being done, it must be in species such as rodents, which are not regulated by the Animal
FIGURE 4-1 U.S. Department of Agriculture (USDA)-tracked animal use data for the United States, 1979 and 2009. (Data for rats, mice, birds, and cold-blooded vertebrates are not tracked.) NHP = non-human primate.
SOURCE: Yates presentation citing USDA Annual Reports.
Welfare Act. The UK Home Office tracks the number of procedures (not the number of animals used) and does include rodents. Over the past 20 years, the use of all animal species except mice has decreased (Figure 4-2).
Increased Use of Rodents
Prior to the mid-1980s, cats were popular research animals for classical neurophysiological procedures because they could withstand the extensive surgeries required, were large enough to accommodate bulky instrumentation, and were inexpensive models. However, in the mid-1980s, new regulations substantially increased the economic cost and administrative burden of feline models. In addition, public opinion shifted against the use of companion animals in research.
Miniaturization of instrumentation has allowed rodents to serve as replacements for felines in some studies. Refinement of techniques such
FIGURE 4-2 UK Home Office–tracked animal procedure data, 1988 and 2009. NHP = non-human primate.
SOURCE: Yates presentation citing UK Home Office Web site.
as chronic recording techniques enables the study of a single animal over a prolonged period of time, leading to greater data collection from single animals. This results in fewer animals needed per study. Thus, the use of non-human primates, which can be trained for more elaborate tasks than cats, has become more economically feasible. Refinement, Yates pointed out, does not always lead to use of a lower species.
Transgenic Mouse Models
The most significant contributor to the increased use of rodents in biomedical research has been the development of transgenic mouse models. In the late 1980s, Capecchi, Evans, and Smithies developed principles for introducing specific gene modifications in mice by the use of embryonic stem cells leading to the development of the first knockout mouse. Today, human genes can be inserted into a mouse or overexpress a particular gene.
Through breeding, it is possible to obtain a line of animals that expresses a new phenotype. Most procedures now are done using transgenic animals. Data suggest that transgenic mice likely account for two-thirds or more of the mice, and more than half of the mammals used in biomedical research.
Use of transgenic animals has allowed neuroscientists to decipher the function of particular genes and to create disease models, Yates said. Knockout models have been used in the study of Alzheimer’s disease, for example, and have been critical in understanding the neural basis of learning and memory.
Use of transgenic mouse models does have limitations, Yates noted. Genetic diseases involving multiple genes can be difficult to model in transgenic animals. In addition, some genetic diseases have different phenotypes in mice and humans. For example, transgenic models of Parkinson’s disease often do not exhibit the same neural degeneration observed in humans. In addition, compensation for the gene manipulation during development can lead to false conclusions about the role of particular genes.
Rodents Versus Higher Mammals
Even considering the limitations, transgenic animals and rodents in general have provided a significant boost to biomedical research. But are they the ideal research model? Yates highlighted some of the advantages and disadvantages of using rodents (Box 4-1).
Expanding Transgenic Technology to Other Species
The technology is now available to create other transgenic species. Zinc-finger nuclease technology has allowed the creation of knockout rats, and theoretically, the technique could work for inactivating genes in any species, including humans. As more types of transgenic animals become available, the balance of species used in biomedical research may shift, Yates noted.
THE ROLE OF NON-HUMAN PRIMATE MODELS IN NEUROSCIENCE
Roger Lemon, Sobell Chair of Neurophysiology at the University College London Institute of Neurology, showed data from the UK Home Office spanning from 1995 through 2010 that indicates a gradual increase in the use of old-world monkeys (primarily macaques) and a gradual decrease in the use of new-world monkeys (mainly marmosets). Overall, nonhuman primates were used in a very small percentage, less than 0.1 percent, of the total number of procedures involving animals in the United Kingdom.
Are Rodents Ideal Research Models?
• Rodents typically live < 2 years, which facilitates aging studies.
• The small size of rodents allows many animals to be maintained in a limited space.
• In some cases, rodents are better models of human diseases than other animal models.
• Rapid aging can confound repeated measures over time in the same subject.
• The small size of rodents causes constraints on manipulations and measurements.
• Some human disease conditions cannot be mimicked in rodents.
• Some findings in rodents might be unique to rodents.
• Rodents lack the ability to vomit, so carnivores remain a better model animal to study emesis and conditions that elicit emesis (e.g., motion sickness).
• Rodents lack respiratory responses such as coughing and sneezing; carnivores remain the most appropriate model for studying cough-suppressing drugs.
SOURCE: Yates presentation.
The majorities, about 81 percent, were involved in applied research (e.g., toxicological tests), often due to a statutory requirement for testing of new drugs in a non-human primate model before entering human clinical trials.
The Case for Non-Human Primate Models
Recital 17 of European Union (EU) Directive 2010/63 states that “the use of non-human primates in scientific procedures is still necessary in biomedical research,” and that “the use of non-human primates should be permitted only in those biomedical areas essential for the benefit of human beings, for which no other alternative replacement methods are yet available.” Recital 13 states that the methods selected should “require the use of species with the lowest capacity to experience pain, suffering, distress or lasting harm, that are optimal for extrapolation into the target species.” In essence, then, Lemon said, both recitals urge that non-human primates be used only in those areas that are likely to be of ultimate benefit for humans.
Articles 5 and 8 of the directive state that non-human primates can be used for basic research. Much of the basic research work in the United Kingdom using non-human primates involves understanding the role of the prefrontal cortex, which may inform progress toward treatment of human neurological and psychiatric disorders of the frontal lobe. The rodent, Lemon noted, is not a particularly good model for studies of higher-level cognitive processes as it lacks the cortical complexity of the human brain.
Independent Policy Reports
Several reports have outlined the scientific case for continued use of non-human primates in biomedical research, including the Weatherall Report in 2006 (MRC, 2006) and the EU Scientific Committee on Health and Environmental Risks (SCHER) report in 2009, both of which identified neuroscience in particular as an area where evidence supports the use of primates (MRC, 2006; SCHER, 2009). A 2004 report from the Academy of Medical Sciences highlighted the need to promote translation of basic science into clinical practice to improve neurorehabilitation, including better therapies for rehabilitation of hand function. This is very clinically relevant, Lemon noted, as in the United Kingdom there are 100,000 new cases of stroke every year and about half of these patients will have some form of serious hand disability. Loss of hand function is also associated with spinal injury (800 new cases per year in the United Kingdom) and cerebral palsy (1,800 new cases each year in the United Kingdom). Lemon also alerted participants to the Bateson report, a retrospective survey of research in the United Kingdom using non-human primates, which was expected to be released the same week as the Institute of Medicine workshop (MRC, 2011).1
Lemon noted that a review by Courtine and colleagues (2007) concluded that there are “crucial differences in the organization of the motor system and behaviors among rodents, non-human primates, and humans” and that “studies in non-human primates are critical for the translation of some potential interventions to treat spinal cord injury in humans.”
There are major differences in the organization of the corticospinal system across species, Lemon said. Examples include the size and numbers of fibers involved; the trajectory that neurons follow within the spinal cord; the extent to which they reach within the spinal cord; and how
1 The report of the independent panel chaired by Patrick Bateson was released on July 27, 2011. The findings were not discussed at the workshop because the report was not publicly available until the second day of the workshop.
they terminate within the spinal gray matter. In primates, the extent of cortico-motoneuronal connections correlates with dexterity, and all dexterous primates that use tools in the wild have highly developed cortico-motoneuronal connections (Lemon, 2008).
Lemon highlighted the work of Schwab and others as an example of how studies in non-human primates can lead to clinical trials. In the late 1980s, Schwab discovered that axons on the spinal cord contained a protein that inhibits the growth of neurons, which he subsequently named Nogo, for “NO GrOwth” (Schnell and Schwab, 1990). In vitro studies showed that neuron growth in culture was strongly suppressed by the myelin inhibitory factor Nogo. Over the next 15 years, Schwab conducted studies in mice and rats to characterize the properties and mechanism of action of Nogo, and developed a means of neutralizing it with antibody (anti-Nogo). Only after this extensive fundamental research, Lemon said, did Schwab decide it was necessary to move to a primate model. The first primate study assessed anti-Nogo as a potential treatment for spinal cord injury (Freund et al., 2006). Non-human primates with untreated spinal lesions permanently lost the ability to make hand movements smoothly, efficiently, and accurately; non-human primates treated with the Nogo-specific antibody largely recovered their ability to make dexterous movements. Lemon stressed that it would be very difficult to assess the impact of spinal lesions on hand function in a rodent. As a result of this successful study in macaques, a Phase I clinical trial in humans has been completed and a Phase II trial began in 2010 (Zörner and Schwab, 2010).
The Future of Non-Human Primate Research
Non-human primate research will continue to be needed, especially research directed at lifelong conditions such as neurodegenerative diseases and psychiatric disorders, Lemon opined. Studies using non-human primates complement other data collection approaches, such as in vitro studies, in silico modeling, human brain imaging, and parallel investigations in rodents. The number of animals used will be relatively low; however, long-term study of a single primate can involve a significant number of independent assessments, resulting in reliable statistical answers from relatively small numbers of animals.
There is a very positive culture of non-human primate care in the United Kingdom, Lemon said. The UK National Center for the 3Rs has played an important role in training and raising the standards of care and knowledge among those working with primates, including technicians, animal care staff, postdoctoral fellows, and principal investigators.
Lemon noted cost, regulatory burden, and training as issues impacting the use of non-human primates in neuroscience research in the United
Kingdom. Cost is a significant obstacle for UK researchers. The purchase cost for a single purpose-bred macaque (excluding taxes) is about £20,000 (more than $30,000) and per diem costs for housing and care range from £50 to £70 per macaque per day (about $80 to $110 per day). Some of the cost stems from increasing standards of welfare that are required and additional security needs. On the upside, the high cost effectively ensures that no trivial or unnecessary work is done in non-human primates. The downside, however, is that high economic costs threaten serious non-human primate research in the United Kingdom. A participant commented that similar financial challenges face non-human primate researchers in other countries. Without additional investment in infrastructure, Lemon observed, centers that are using non-human primates may find it difficult to compete with other types of research in the long term.
The possible reclassification of “moderate procedures” involved in long-term neuroscience studies as “severe” is another problem facing EU researchers. Lemon suggested that reclassification may lead to large restrictions in the types of neuroscience research that can be conducted on nonhuman primates. Finally, training is important for the long-term future of non-human primate research. Lemon suggested that the perceived difficulty of conducting research with non-human primates may negatively affect the ability to attract the best young scientists to the field.
ETHICAL AND PRACTICAL DILEMMAS OF RESEARCH WITH NON-HUMAN PRIMATES
Basic Versus Applied Research
Stuart Zola, director, Yerkes National Primate Research Center at Emory University, noted that the definitions of basic (or fundamental) research and applied (or translational) research are not necessarily clear. In the early 1600s, Sir Francis Bacon divided research into experimenta lucifera, experiments shedding light, and experimenta fructifera, experiments yielding fruit. The distinction between basic and applied research is relevant to the use and regulation of animals in research. Biomedical ethics committees and Institutional Animal Care and Use Committees (IACUCs), for example, must consider the potential benefits of the proposed research for humans and animals. In addition, animal rights groups are often concerned that basic or fundamental research using animals has no immediate application to humans.
In practice, it can be difficult to distinguish to which domain an activity clearly belongs. For example, an experiment that involves the development of a behavioral task in non-human primates to assess functions of the hippocampus would seem to be very basic research. However, there is clear
application of the knowledge in terms of diagnostics or interventions with respect to a wide range of clinical diseases and conditions. Zola noted that what may look like basic research may have very clear applications. Basic research was critical to the development of medical breakthroughs such as coronary bypass surgery and magnetic resonance imaging (MRI), for example. In the United States and the United Kingdom, Zola noted, the focus is now “translational research,” bringing together basic scientists and clinicians to develop the best and most effective treatments and interventions.
The “Justification Rule”
An issue of concern for scientists is the idea that some clear applied benefits should come from the research itself, whether it is a diagnosis, treatment, prevention, or some other benefit to humans or animals. This “justification rule” is espoused in EU Directive 2010/63 (Para 17) which states that non-human primate research “should be permitted only in those biomedical areas essential for the benefit of human beings, for which no other alternative replacement methods are yet available” or when basic research is carried out in relation to potentially life-threatening conditions in humans or in relation to cases having a substantial impact on a person’s day-to-day functioning (i.e., debilitating conditions). Zola opined, however, that this need for justification is based on two presumptions that are incorrect: first, that there are clear distinctions between basic research and applied research, and second, that it is possible to predict direct benefits to humans or animals that result from research using animals. Instead, Zola said, we can recognize that the discovery of fundamental knowledge has value in its own right. This is not an “anything goes” approach, Zola stressed, but an approach of basing choices on science and value, and not on semantics and arbitrary distinctions.
Challenges to Non-Human Primate Research
Advances in technologies related to genomics, behavior, imaging, and microbiology/immunology are offering new avenues for non-human primate researchers to develop therapies, interventions, and diagnoses (Figure 4-3). Zola offered several examples of challenges and welfare concerns facing researchers related to some of these new technologies. Positron emission tomography (PET) imaging can be used to conduct brain imaging while the animal is engaged in cognitive tasks, very much the same way it would be done with humans, Zola noted. However, a number of concerns are associated with PET imaging of non-human primates. First, the animal is awake, raising questions about stress, not unlike the stress many humans feel when inside a much longer MRI tube.
FIGURE 4-3 Examples of technologies used in translating research with nonhuman primates to human applications.
NOTE: MHC = major histocompatibility complex; MRI = magnetic resonance imaging; PET = positron emission tomography.
SOURCE: Zola presentation.
The use of PET imaging has recently been used to show the localization of simian immunodeficiency virus (SIV) in individual macaques. The ability to track the virus in the body is revolutionary, Zola noted, and will aid development of HIV vaccines and interventions. However, this requires a lot of animals, and the particular macaque species used in this study, the sooty mangabey, is an endangered species. This use is relevant as the sooty mangabey is one of the species from which the mutation from SIV jumped to humans. Research in this species could help answer many questions about immunity to SIV, but invasive research is prohibited because the sooty mangabey is endangered.
A rapidly advancing area is the development of transgenic non-human primate models of inherited neurodegenerative diseases. Recently, researchers produced the first transgenic non-human primates that express the Huntington’s disease gene; the animals exhibit many of the defining signs of Huntington’s disease (Yang et al., 2008). Animals also have been
developed that carry risk factor genes for Alzheimer’s disease. Longitudinal studies of the animals are ongoing, including gene expression studies, MRI, and cognitive behavioral evaluation. This is a remarkable new era in the study of disease, Zola said, but there are ethical concerns about inducing disease in non-human primates. In addition, such studies require a large number of animals.
Overall, the most significant challenge is infrastructure. The lack of resources, space, animals, and funding is outweighing the ability to do the research. The precarious balance between science and infrastructure is really an ethical concern, Zola concluded, as the inability to do the science will lead to lives lost in the end.
ADMINISTRATIVE AND ECONOMIC COSTS
Charles J. Heckman, professor at Northwestern University Feinberg School of Medicine, offered his perspective on the regulatory burden from the viewpoint of an IACUC chair. Across Northwestern’s two campuses, there are a total of about 16,000 cages of mice at any given time, approximately 500 cages of rats, and a modest number of larger animals. Approximately 200 principal investigators are involved in animal research. The university is AAALAC-accredited, and receives NIH funding totaling around $300 million per year, about half of which is probably associated with animal research, Heckman estimated.
Consequences of Regulations for a Large Research University
An animal program the size of Northwestern’s leads to a large amount of work for the IACUC and the administration. At any one time, there are between 900 and 1,000 animal protocols, Heckman said, with about 200 new protocols each year. In the U.S. system, protocols must undergo a full review and renewal every 3 years, meaning about 50 to 100 de novo reviews each year. There are also 250 to 300 personnel addendums each year, some covering multiple people.
Facility Inspections and Reports
As mandated by the Public Health Service (PHS) Policy and the Animal Welfare Act, all animal housing spaces must be inspected twice per year. The inspection teams include several IACUC members, an IACUC staff person, and sometimes a safety staff member. It takes a minimum of 5 teams a minimum of 2 hours each to inspect a facility. The laboratory spaces of
the 200 investigators conducting surgical procedures must also be inspected twice per year and the semi-annual reports consume a modest amount of time and effort as well.
IACUC Staff Personnel and Volunteer Members
The total IACUC staff handling this administrative load at Northwestern is 7 full-time positions: a director, 3 IACUC program assistants, an administrative assistant, and 2 people responsible for postapproval monitoring. There are 23 volunteer IACUC members: 3 veterinarians, 3 community members, 2 members from the Office of Research Safety, and 15 principal investigators from various departments with significant animal programs. Heckman estimated his own effort as chair at around 15 to 20 percent, and noted that it is difficult to find a sufficient percentage of effort to sustain his scientific work while serving as IACUC chair. Based on the review burden, Heckman estimated that a typical committee member needs at least a couple of hours per week to review protocols and attends a 2- to 3-hour committee meeting each month. A subcommittee of the IACUC is also devoted to reviewing medical records, which takes 2 to 3 hours per week. Each lab working with a USDA-covered species also must conduct a monthly self-audit of at least some of their medical records and report the audit findings to the IACUC.
Investigators and Laboratory Members
Protocol preparation takes a significant amount of time. Each individual protocol is approximately 30 to 40 pages, taking at least 2 hours per protocol to draft, followed by several rounds of revisions and review. Most large laboratories with four or more protocols will usually have a laboratory manager who dedicates at least a third of his or her time to managing the protocols. The principal investigator is ultimately responsible for ensuring that fellows, students, and staff understand the importance of the process, and adhere to the approved protocols. This is not just paperwork, Heckman stressed.
Summary of Session Points
Basic vs. Applied Research
• Arbitrarily separating research into “basic” and “applied” categories could be harmful if used to determine the types of research that can or cannot be conducted:
o The line between basic and applied research is often blurred.
o The discovery of fundamental knowledge has value in its own right.
o Some basic research may have clear application toward development of treatments for nervous system diseases and disorders.
• It is not always possible to predict specific benefits to humans or animals that result from research using animals.
Animals in Neuroscience
• Rodents are the dominant mammalian species used in neuroscience research:
o Miniaturization of instrumentation has allowed rodents to replace larger animals.
o Development of transgenic mouse models has significantly increased the use of rodents.
• Refinement does not always lead to use of a species with lower cognitive abilities, such as rodents.
o Refinement of techniques can result in a decreased number of animals required for study, thus making the use of species of higher cognitive abilities, such as non-human primates, more economically feasible.
• Non-human primates continue to be used in biomedical research, including neuroscience:
o Studies in non-human primates can lead to human clinical trials.
o Non-human primate studies can complement in vitro studies, in silico model ing, human brain imaging, and parallel investigations in rodents.
o Long-term study of a single primate may involve numerous independent assessments, resulting in reliable statistical answers from a relatively small number of animals.
o Difficulties in using non-human primates in research include costs, regulatory burden, and attracting talent to the field in the face of mounting challenges.
o The potential reclassification of “moderate procedures” in long-term neuro-science studies as “severe” may impact the use of non-human primates.
Challenges Associated with New Regulations
• Increased recordkeeping requirements.
• Increased cost of conducting biomedical research without direct evidence that increased regulations result in improved animal welfare.
• Conflicting regulations from multiple agencies with multiple sets of rules.
• Compliance can require significant time and effort from dedicated IACUC staff, IACUC volunteer members, principal investigators, and laboratory staff, for writing and review of protocols and addendums, animal facility and laboratory inspections, monitoring, and reports. Training and accreditation activities can consume resources.
SOURCE: Individual panelists and participants.
This page intentionally left blank.