What are you feeding your lab animals? It seems like a simple question, but in my experience, often the answer is “I don’t know”, “Whatever the animal facility has” or even worse, “It doesn’t matter, it’s just the food”.

Diet does matter, and the control over this environmental component is important, given efforts to reduce the number of lab animals used. So how can you take steps to reduce variability that may be coming from the diet? First, understand the diet choices.

Diets made from refined ingredients including corn starch, sucrose, soybean oil and casein (a protein from cow’s milk) are called purified ingredient diets. Each ingredient basically contains one nutrient. Grain-based (GB) diets are made with ingredients closer to their ‘raw’ state—soybean meal, ground oats, ground wheat and alfalfa. Each ingredient contains many nutrients, as well as a number of non-nutrients that are generally absent from purified diets.

While many researchers are familiar with GB diets (they are what most lab animals are being fed), it might surprise you to learn that purified diets have been used in lab animal research for 90 years. In those early days, they were an indispensable tool that allowed researchers to remove single nutrients from the diet and so discover those that were required for the animal’s survival. Purified diets helped to define the vitamin and mineral requirements of the rats and mice in use today.

The fact that purified diets are ‘clean’ enough to induce vitamin and mineral deficiencies tells us that they can be controlled and have little variability. In fact it wouldn’t have been possible to induce deficiencies using GB diets, given that each ingredient contains multiple nutrients.

Maybe this history of purified diets contributes to the misconception that they are only needed for very specific research goals, like studying nutrient deficiencies. However, purified diets have already proven themselves in other areas. For example, most of the diet-induced metabolic disease research done in the past 20 years used high-fat, sucrose, fructose or sodium containing purified diets, allowing researchers to understand mechanisms of disease and test potential therapies. This would not have been possible with GB diets, since adding fat, sucrose, etc. on top of the GB diet dilutes other nutrients, potentially making that high-fat diet a low-protein diet as well.

At the same time, purified diets were also being appreciated and used for what they lack—non-nutrient materials, many of which can unwittingly affect the animal’s phenotype. For example, heavy metals like arsenic are undetectable in purified diets while GB diets contain levels known to affect gene expression. And because purified diets don’t use plant materials like soybean meal, there are no phytoestrogens to interfere with the animal’s endocrine system and no chlorophyll to interfere with in vivo imaging.

Reducing data variability is an unwritten goal of every experiment. Tighter data means that fewer animals can be used, also saving time and money.

So, what are you feeding your animals? Maybe it’s time to find out.

Matthew R. Ricci, Ph.D. is Science Director and Vice-President at Research Diets, Inc., 20 Jules Lane, New Brunswick, NJ USA. +1 732-317-5241,