[Ed. note: In December 1996, AWIC staff were invited by Dr. Doug Erbeck to present a workshop on the Animal Welfare Act to his animal technology class at Murray State University. Paul Jaco, a senior, was the winner of a poster competition between members of the class. He is currently in graduate school at Murray State University. We invited him to write this article for the AWIC Newsletter.]
In the past century, scientists have moved toward using fewer animals in research by choosing alternatives models. The concept of "alternatives" was developed in 1959, with the publication of The Principles of Human Experimental Technique by W.M.S. Russell and R.L. Burch. They proposed the three R's of research: replacement, reduction, and refinement.
Replacement is more than substituting animals with nonanimal models. In fact, replacement can be as simple as replacing a warm-blooded animal with a cold-blooded animal. An example of this is Teleost fish used in research of diabetes mellitus. Many times companion animals are replaced with less sentient animals such as rodents. Other avenues of replacement are computer simulations, tissue cultures, and many other models.
Reduction may be accomplished by simply reducing the numbers of animals used in the research. One example is proper application of statistical information. Reduction can also involve organs and tissues taken at the animal's time of death to be used for research that does not involve whole animals.
Refinement is the best quality care that can be afforded the animal. This may involve training the staff in the area of husbandry, or training the animals with behavior enhancement. It has been found that social interaction with animals leads to less distress on both staff members and the animals being studied in research.
In addition to the 3 R's of Russell and Burch, a fourth R has recently been introduced to the scientific community. The fourth R overlaps some of the refinement techniques. Ronald Bank, D.V.M., (1995) proposed the 4th R of research as being Responsibility. He states, "Responsibility toward research animals focuses new facility design and facility renovation toward accommodation of social interaction and behavioral interplay performing approved experimentation in a manner as distress free as possible, with analgesics or anesthetics used when necessary, of sufficient efficacy and dosage to ameliorate pain and distress." We also share a responsibility to educate the public and show them that we do care about the welfare of the animals.
Using animal models in diabetes research began in 1890, when Von Merhring and Minkowski began studying the digestive functions of the pancreas (Von Mehring and Minkowski 1890, Herbery 1988). They noticed that the animals on which they had performed a pancreatectomy developed acute diabetes mellitus. Things have changed considerably since the 1890's when Von Merhring and Minkowski did their research. Today, very little research is done involving animals without considering possible alternatives. Before involving animals in research, we must consider the 3 R's of Russell and Burch in order to satisfy Institutional Animal Care and Use Committees and funding agencies such as the National Institutes of Health (Hart 1995, Erbeck 1996).
Perhaps the most sought-after alternative is replacement. Scientists doing research on diabetes are finding several replacements for higher animals. One such replacement involves the Goby fish (Gillichthys mirabilis) (Bern et al. 1992, Kelly 1995). This advanced Teleost has a single mesentery-bound pancreatic endocrine organ that is separate from the exocrine pancreatic acini. The organ is located next to the portal vein, and its removal (isletectomy) is very simply done through a small incision on the ventral surface of the fish. Many fish have these separate pancreatic endocrine organs that may be referred to as principal islets or Brockmann bodies (Kelly 1992).
After the isletectomy, these cold-blooded vertebrates show clinical signs of a typical Insulin Dependent Diabetes Mellitus (IDDM) mammal. The fish lose body weight and show slow skeletal growth and skeletal retardation which is measured by body length. They also show 50 percent reduction in cartilage. After these animals are given insulin, they respond like mammals. Kevin Kelly (1992) states, "The establishment of this unique model of IDDM in an ectothermic vertebrate should prove valuable for future comparative studies on the role of insulin and other pancreatic factors in the regulation of metabolic and growth processes."
Another way replacement is shown in diabetes research is with laboratory rodents. Rodents are beneficial in several ways. Their low maintenance and their short generation time allow careful and controlled experiments to be conducted. Since laboratory rodents usually breed very easily, diabetes can be genetically established in a colony. This makes an excellent model for the study of genetic factors. Environmental factors such as pharmaceutical actions on diabetes mellitus are also easily observed.
Rodents, whether rat, mouse, hamster, or guinea pigs, and the rabbit make suitable models for spontaneous diabetes. They are used to study the natural development of the disease. In the early days of diabetes research, the best and quickest way to induce diabetes was to perform a pancreatectomy just as Von Mehring and Minkowski did with cats, rabbits, pigs, pigeons, ducks, dogs, and primates.
In recent years, scientists and technologists have worked diligently toward refining techniques that have led to the discovery of chemical agents that physiologically alter the function of the pancreas. The advantage to using such chemicals is that body changes during and after the induction of diabetes can be observed. The five major classes of diabetogenic agents are chemicals, biological agents, peptides, potentiators, and steroids. Perhaps the most commonly used chemicals agents are alloxan and streptozotocin (Mendez and Ramos 1994). These compounds target the Langerhans islets beta cells. Because of its pathology potential, alloxan is preferred for the rat.
Still another avenue of refinement is gene alteration and transgenic models (Miyazaki and Tashiro 1993). A transgenic organism is "an organism in which foreign DNA is integrated into the genome early enough in development to transform all cell lineage. The foreign gene would then be passed on to progeny" (Elseth and Baumgardner 1995). With transgenic models, the animal cannot be completely substituted for a nonanimal model such as cell culture. Instead, transgenic models are used mostly as a refinement technique in mice. Mice allow us to look at the molecular mechanisms of the autoimmunity against beta cells as well as the disease-influenced genes. There is no surgeryrequired for animals being studied, which makes this model a very humane source for obtaining research data.
In some cases, replacement can be attained by completely eliminating animals. One replacement is the use of a glucose kinetics simulation model that estimates the glucose utilization rate in IDDM and IIDM (Insulin Independent Diabetes Mellitus) individuals. In this complex model, there are at least 16 different variables to consider (Boroujerdi et al. 1995). The factors are plugged into a mathematical equation that allows interpretation of the results of the synthetic model.
The 3 R's, when demonstrated in actual research, may overlap. For example, the glucose kinetics simulator was proposed as a replacement tool in IDDM studies. This model is also an excellent source of reduction. To gather the original data for diabetic patients that the model simulates, diabetic patients were closely examined after three regimens of insulin therapy. The first regimen was 24 hours after insulin withdrawal. The data was recorded after conventional insulin therapy and after an overnight insulin infusion. Investigators compared the results with those that had already been published from glucose clamp experiments (Boroujerdi et al. 1995). They were able to collect a range of glycemic levels and insulin concentrations. The kinetics simulator not only reduced the number of animals being used, but also reduced the labor of the research staff. For a laboratory to work with large numbers of research animals, it must have the correct facility design as well as people properly trained in animal husbandry. These concerns were greatly reduced when the glucose kinetic simulator was used.
Nutrition is becoming a major concern in the prevention of diabetes (Herbery 1988). In much of the nutrition-related diabetic research, animals with the disease are closely monitored with a set sugar intake, restricted carbohydrate levels, and a strict regimen of exercise. This enables the investigator to view the longterm effects of diabetes while using the same subjects. Again, the number of animals used is minimized. A certain amount of refinement in the health concerns of these animals is also seen. The exercise is clearly a commendable part of husbandry and can be considered environmental enrichment. In this research the data does not suffer, rather it is helped.
Researchers understand that the word "alternative" has many sides. Each avenue should be studied to fit the proposed research. Depending on the type of research, there may be several models that employ the 3 R's. Then researchers can examine the affordability of the models and how well the results will simulate the desired outcome. Deciding which model to use in research is usually not a quick or easy decision. The examples given in this paper succeeded by giving reliable and consistent data for research in diabetes while considering the concept of alternatives.
Go to:
Contents, Animal Welfare Information Center
Newsletter
Top of Document
The Animal Welfare Information Center
U.S. Department of Agriculture
Agricultural Research Service
National Agricultural Library
10301 Baltimore Ave.
Beltsville, MD 20705-2351
Phone: (301) 504-6212
FAX: (301) 504-7125
Contact us: http://www.nal.usda.gov/awic/contact.php
