The United States Department of Agriculture (USDA) plays an important role in the nutritional well-being of the country and throughout the world as part of its leadership role related to the production of food. Part of its mission is to conduct research that helps define an optimal and safe diet, to conduct research to enhance agricultural production, and to provide dietary guidance based on the latest research. These activities are carried out largely through the Agricultural Research Service, which is the intramural research arm of the department. In addition, the USDA plays an important role in administering a number of food assistance programs, which in order to be effective must be based on the best nutritional information available. The promise of nutrigenomics holds exceptional opportunities for all of the areas of importance to the department. In this report, I will examine the promise and challenges regarding the use of nutritional genomics in agriculture.
Genomic Prediction in Dairy Cows
The USDA has for over 100 years been active in programs to improve dairy cattle, which today is manifested in the National Cooperative Dairy Herd Improvement Program [1]. This program is aimed at improving the breeding of dairy cows. Through the widespread use of artificial insemination it has been extremely successful in keeping up with the demand for food, improving productivity and breeding, improving the quality of milk, and improving profitability for producers. This program has been of great importance to the dairy industry.
Milk production in dairy cows is a trait that is actually transferred through the male sire. Traditionally, through what is known as prodigy testing, predictions of the value of a bull are based on pedigree and milk production information that was maintained by the Animal Improvement Laboratory at the Beltsville Agricultural Research Center in Beltsville, Md. With the emerging understanding of the bovine genome [2, 3], it has opened up the door for genetic prediction. Working as part of a Cooperative Research and Development Agreement with Illumina Inc., Beltsville scientists and their collaborators began an ambitious examination of single nucleotide polymorphisms (SNPs) in cows. They have developed a commercially available SNP chip that can be used for screening a bull to produce a dairy cow that has the desired characteristics with regard to milk production. This newer method is called genomeenhanced improvement evaluation [4].
A comparison between the 2 methods of evaluating bulls is presented in table 1. The prodigy testing method of prediction is accurate to about 35%, the results are not known until the cow is about 5 years old, and the cost is about USD 50,000. Using an Illumna chip of over 50,000 SNPs, the accuracy of the prediction increases to greater than 70% and the determination can be done at birth of the calf. Significantly, the cost of the analysis is only about $250. This newer approach has changed the dairy industry in a very short time. It will lead to greater increases in productivity and milk production, which as a source of food is of significance for the human population. Beyond the dairy industry, this new approach represents a proof-of-concept of genetic prediction in agriculture. The use of this genomics approach can be used for any traits that one might be interested in. As indicated in the table, it has been used to predict a number of qualities important to the dairy industry. Furthermore, this SNP analysis approach can be used for genetic prediction in other commodities. Scientists at the Beltsville Agricultural Research Center are taking a similar approach to soybeans, by looking at qualities that are traditionally bred into them, such as protein content, disease resistance, yield, drought resistance and their storage characteristics.
Dietary Guidance
In the United States, the Department of Health and Human Services and the Department of Agriculture share the responsibility for development of the Dietary Guidelines for Americans. These guidelines form the basis for the government's dietary advice and are revised every 5 years to include the latest nutritional research. In light of the role that nutrition plays in maintaining health and the importance of a proper diet, it is attractive to consider that nutritionally related chronic diseases can be prevented by an improved diet. With the escalation of health care costs, an approach based on prevention is particularly attractive. These guidelines, like much of our nutritional advice, are a population-based recommendation. That is to say, the advice has been ‘one size fits all’ despite the fact that we know that tremendous variation can exist within the population. Despite this, however, the guidelines do serve a purpose for enlightening people about sound nutritional information. The promise of nutrigenetics is to develop individualized dietary advice that more accurately represents the risk to an individual with regard to nutritionally related diseases. Today one can see the evolution toward this goal with the availability of information contained on the MyPyramid.gov website, where an individual's lifestyle and basic background can produce an individually tailored dietary pyramid.
Comparison of traditional prodigy testing in dairy cows and genomic-enhanced improvement evaluation
a Predictions determined for net merit, milk produced, fat, protein, productive life, pregnancy rate and calving ease.
As we become more aware of the genes that are involved in health and the polymorphisms associated with those genes, we will be confronted with the need to be able to make meaningful dietary recommendations. The overall approach and pitfalls associated with it are depicted in figure 1. Diets are a complex interaction of a variety of foods, nutrients and non-nutritive components. Changing one or a number of components of the diet will no doubt affect the overall impact of one's diet in ways that we do not fully understand. It is important to recognize that metabolism and interactions between nutrients and genes take place within the metabolic context of a cell or an individual. Many of the diseases that are associated with diet are known to be the results of multiple gene interactions [5, 6]. These interactions are poorly understood and a simple attempt to alter the interaction or to perturb the system might have no effect or even negative effects.
A significant but unknown factor in any discussion of nutrigenomics is the uncertainty of the role that epigenetic influences play in altering the response to diet. These epigenetic influences are likely to be significant in light of the fact that nutrition is the result of a lifetime of ingestion of nutrients and non-nutritive components. It is not clear if there is a threshold of interactions before the biological effects are fixed or are actually observed. Equally unknown is whether the effects of the interactions can be repaired or reversed. Additional research is needed to determine if important nutrient gene interactions occur early in life and if there is significant imprinting of an effect, and what if anything can be done to prevent or reverse the biological effect from occurring.
Overview of significant interactions between diet and the genome of individuals as related to the development in individualized dietary advice. The shaded boxes highlight some of the concerns and potential problems in the development of the advice.
Discussion
The connection between nutrition and agriculture is a relationship that, because it is so fundamental, it is often overlooked or underappreciated. The example of genetic prediction in dairy cows has in a very short period of time changed the dairy industry and demonstrates the potential use of this technology for humans and nutrigenomic approaches to human nutrition. By itself, it demonstrates the value of genomics in making more food available for an expanding population. The ability to expand this approach to other commodities will offer an opportunity to increase yield, improve production and quality.
While genetic prediction might work well in agriculture and certainly reinforces the promise of nutrigenomics in the human population, there are reasons to be cautious. Agriculture has an advantage in that many of the animal and plant commodities are rather homogeneous genetically, while the human population is not. In agriculture, if a product does not meet the desired characteristics it could simply not be used or could be used for other purposes. Significantly, there exists a tremendous amount of phenotypic data for agricultural commodities and ready access to a huge collection of germplasm. Efforts are well underway to develop similar types of information and resources from ongoing nutritional studies and these resources must be available in order to understand and validate nutrigenomic approaches to delivering dietary guidance.
The nutritional status of an individual represents both the recent and lifetime intakes of substances in the diet. It is not clear to what extent the impact of these exposures can be reversed or mitigated. While there are well documented DNA repair mechanisms, their relationship to nutrition is now beginning to be understood [7]. The role of tissue remodeling and the ability to reverse nutritionally related biological effects are not clear. When dealing with genetic transformations, it is possible that imprinting occurs whereby effects on genetic material or gene expression may have long-lasting effects later in life. The role of epigenetic factors in influencing the response of an individual to dietary components will no doubt prove to be significant and need to be explored. The expanding importance and impact of nutrigenomics has the potential to raise some challenging ethical concerns. These and other issues will clearly need to be addressed as research in this important area proceeds.
As seen in the case of dairy production reported herein, the best genetic prediction is not 100% accurate. The guidance of individualized nutrition might be improved, but may never be completely accurate. No doubt the public will hear the terms genomics or individualized nutrition and interpret that as meaning highly accurate or absolute. This could open the door to unethical marketing of products and promises to the consumer or raise expectations that ultimately result in a loss of confidence in dietary guidance.
A significant issue that arises when taking a nutrigenomic approach to the development of dietary guidance is that we may be able to predict genetic predisposition but not know the actual genes that are of importance. It will clearly present a challenge to make meaningful dietary recommendations under such conditions. Even if we can identify the genes, the important point is determining how we can alter the diet in a way that produces the desired beneficial effect. In the simplest case, if we know how a particular dietary component interacts with a gene, then people carrying that gene can be advised to simply avoid that component. However, it is not likely we will be able to do that in many cases, particularly in light of the multiple genes that may be involved in many complex biological processes. Lastly, simply making dietary recommendations and getting people to change their dietary habits based on advice alone is never easy. Effective education programs for individuals, dieticians and health care providers will very much be needed. An important role that agriculture might play would be to develop varieties of commodities that would have desired nutrient profiles that would make it easier to meet the dietary recommendations. A 1998 perspective by Fink [8] highlighted the importance of Arabidopsis as a model organism in genetics. He discussed the shortfalls of nutrition and how this model organism might be helpful for studies of nutrition and how the optimal diet would be determined through selective breeding of plants and animals that might better meet our nutritional needs. Clearly we now have the tools to do these types of studies without the use of model systems but we need to have the nutritional rationale for particular studies.
Nutrigenetics/nutrigenomics will no doubt continue to provide information on metabolic processes and nutritional requirements. While the possibility of the development of individualized dietary guidance is becoming a reality, there are concerns and challenges. As we develop the capability to identify nutritional requirements for individuals, it is going to be a challenge to be able to make meaningful changes in a person's diet.
