The publisher's final edited version of this article is available at Animal models are invaluable resources for biomedical research, including research on the effects of diet on metabolism and disease. Usually, great care is taken to ensure comparable genetic backgrounds and environmental conditions when performing studies using animal models, since this minimizes introduction of variability that can confound detection of treatment-related phenotypic differences. However, many papers using animal models draw conclusions about dietary effects from comparison of natural ingredient chow with defined diets, despite marked differences in micro- and macronutrient content. When comparing the effects of a chow diet with a high-fat defined diet, the effects of the dietary fat will be confounded with the effects of other components in the diets. This issue is highlighted by a limited literature survey that was conducted to identify common problems in the use and reporting of rodent diets.
All original research papers identified by the keywords "mouse high fat" published in 2007 in five high-impact journals were included in this evaluation. Of the 35 papers examined, only 14% (5 papers) compared diets using identical nutrients differing only in relative amounts of fat and carbohydrate (Figure 1
). Specific details regarding the dietary comparisons made are often lacking, and frequently conclusions are drawn from comparisons of defined high-fat diets to chow.
). Specific details regarding the dietary comparisons made are often lacking, and frequently conclusions are drawn from comparisons of defined high-fat diets to chow. | Figure 1 Pie chart showing the percentage of 35 original research papers evaluated that used appropriate diet comparisons (14%), that compared chow and defined high-fat diets (43%), or that presented insufficient information to evaluate diet comparisons (34%). (more ...) |
Regular chow is composed of agricultural byproducts, such as ground wheat, corn, or oats, alfalfa and soybean meals, a protein source such as fish, and vegetable oil and is supplemented with minerals and vitamins. Thus, chow is a high fiber diet containing complex carbohydrates, with fats from a variety of vegetable sources. Chow is inexpensive to manufacture and is palatable to rodents. In contrast, defined high-fat diets consist of amino acid supplemented casein, cornstarch, maltodextrose or sucrose, and soybean oil or lard, also supplemented with minerals and vitamins. Fiber is often provided by cellulose. Chow and defined diets may exert significant separate and independent unintended effects on the measured phenotypes in any research protocol.
Two important difference between chow and defined diets are the phytoestrogen content from soy that is high but variable in chow diets but is absent from defined diets (reviewed in [1]) and sucrose that is used as a carbohydrate source in defined diets but is absent from chow. Dietary phytoestrogens influence food and water intake, anxiety related behaviors, locomotor activity, learning and memory, fat deposition, blood insulin, leptin and thyroid levels, and lipogenesis and lipolysis in isolated rat adipocytes ([2, 3] and reviewed in [4]). Sucrose is 50% fructose and there is abundant evidence that fructose can exacerbate weight gain and contribute to insulin resistance and dyslipidemia (reviewed in [5]). Other effects that differ between chow and defined diets and that may be related to either phytoestrogen or fructose content include bone-related changes, plasma estradiol, urinary calcium, urinary corticosterone [6], and solution taste preferences [7].
When comparing the effects of chow with a high-fat defined diet, the effects of the dietary fat will be confounded with the effects of other components that differ between the diets. Just as it is essential that mouse strains be specified, constituents of experimental diets must be specified [1, 8, 9]. Specifics to be considered are the objectives of the study, whether the endpoints are affected by several diet components, the maintenance diet, and palatability of the diet [1]. Although this correspondence focuses on high-fat diet comparisons, these issues are applicable to any specific nutrient comparisons.