Unique Characteristics of Biotechnology Products

Susan K. Harlander

Director, Dairy Foods Research & Development
Land O’Lakes, Inc.
PO Box 116
Minneapolis, MN 55440

[published February 1994]

Are biotechnology-derived foods any different than foods derived from traditional agricultural breeding and selection methods? Do biotechnology-derived foods have unique characteristics that may require special consideration? Under what conditions should biotechnology-derived foods be labeled? What kind of information would be useful to consumers and how should this information be conveyed on the label? To answer these questions requires an understanding of biotechnology, an understanding of how biotechnology-derived foods fit within the context of traditional agricultural practices and an understanding of the regulatory framework which assures the safety of the food supply.

Understanding biotechnology

What is biotechnology? The Office of Technology Assessment (OTA) defines biotechnology as “any technique that uses living organisms or parts of organisms to make or modify products, to improve plants or animals or to develop microorganisms for specific uses.” This is a very broad definition which could encompass many of the techniques (e.g., fermentation, breeding and selection, induced mutagenesis) used for centuries to improve the food supply. What separates traditional biotechnology from the “new” biotechnology is genetic engineering. Genetic engineering allows the transfer of deoxyribonucleic acid (DNA, the genetic blueprint) from one organism to another. Because the DNA in every living organism is structurally and functionally identical, genetic engineering techniques developed in the early 1970’s can be used to isolate, characterize and transfer DNA between viruses, microorganisms, plants, animals and even humans. The same basic techniques can be used to improve any living organism. Genetic engineering allows the transfer of single, well-characterized genes across species barriers in predictable, precise and controllable ways. The process is much faster than traditional methods currently used to genetically improve the food supply.

Understanding traditional agricultural practices

Prior to the discovery of genetic engineering, how were traditional techniques used to genetically improve the food supply? Traditional fermentation processes utilize living bacteria and yeast (starter cultures) to produce valuable products such as cheese, yogurt, sausage, sauerkraut, bread and wine. Starter cultures do not always possess the right combination of desirable properties, and are frequently exposed to various mutagenic agents, such as ultraviolet light or EMS (ethylmethylsulfonate) in order to improve their metabolic characteristics. Following mutagenesis, improved strains are selected based on specific properties such as improved flavor-producing ability, resistance to bacterial viruses (bacteriophage), or ability to grow at reduced or elevated temperatures. In addition to the property of interest, mutagenic agents create numerous and random changes in the DNA. Since the function of most microbial genes has not been identified, there is no way to monitor unanticipated changes in nontarget genes. The fact that strains have been “genetically improved” is not revealed on the label. In fact, in most cases, the label does not even include the scientific names of the organisms used in the fermentation. Traditional breeding and selection techniques in plant and animal agriculture take advantage of the tremendous genetic diversity available in different species. In corn, for example, genetic improvement can be achieved using cross hybridization. In traditional breeding, the 60,000 to 100,000 genes from one parental species are mixed and randomly sorted with an equivalent number of genes from another parent of the same species. Improved plants or animals are then selected from the offspring of the cross and back-crossed with parental strains to eliminate undesirable traits. An example is hybrid seed corn, the preferred variety in the U.S., which took over 20 years to develop using traditional breeding and selection techniques. Mutagenesis has also been used to genetically improve crops. An interesting example of how mutagenesis has been used in the plant area is the single-serving lettuce developed by the U.S. Department of Agriculture (USDA). Lettuce seeds were exposed to the mutagenic agent EMS, and plants were selected that produced a much smaller head size, probably due to mutation in gene(s) controlling leaf size. Once these lettuce heads reach the marketplace, they will probably not be identified with a label indicating that mutagenesis was used in the process. The traditional methods used to improve microorganisms, plants and animals have some limitations. The random process of mixing and sorting 60,000 to 100,000 genes during the breeding event is imprecise and uncontrollable. Although one might be screening for improved yield, there are lots of other mutations that go undetected. In addition to being a time consuming process, it is not possible to transfer a valuable trait from wheat into corn due to natural species barriers. Therefore, the gene pool cannot be expanded using traditional methods. The U.S. Food and Drug Administration (FDA) has the authority to evaluate new plant varieties developed by traditional breeding methods before they reach the marketplace, but the agency has not exercised this authority. The FDA does not routinely assess new plant varieties for safety because established agronomic practices used by plant breeders and farmers (e.g., performance, yield, disease resistance), and food processors (e.g., chemical composition, nutritional quality, presence of natural toxicants, processing characteristics) have been considered sufficient to ensure human safety.

Understanding the regulatory framework

As outlined by Ed Korwek in another paper in this volume, the FDA has broad authority for regulating the food supply. Although the FDA has always had the authority to regulate new plant varieties, they have not exercised this authority in the past. On May 29, 1992 (Federal Register, Vol. 57, No. 104), the FDA published a statement of policy regarding foods derived from new plant varieties. These guidelines relate specifically to plants, including genetically engineered plants, as well as those derived from traditional breeding processes, but do not include applications involving transgenic animals or starter cultures. The policy indicates that no new regulations are required for oversight of biotechnology-derived food crops and that regulation will be identical in principle to that applied to foods developed by traditional plant breeding. Regulatory status is dependent upon objective characteristics of the food and its intended use rather than the process by which it was produced.

As in the past, producers have an obligation to ensure that foods are safe and in compliance with applicable legal requirements. Producers are encouraged to informally consult with the FDA prior to marketing new foods and a “decision tree” approach provides guidance as to when contact with the FDA would be required. For example, insertion of a gene which codes for a component not currently consumed in the human diet could be a trigger for FDA consultation. A crop with such a gene might be considered to have a new food additive and be subject to oversight under the Federal Food, Drug and Cosmetic Act (FFDCA).

The guidelines contain no requirement that foods derived from new plant varieties be labeled. On April 28, 1993, the FDA published a request for additional comments on the labeling issue (Federal Register, Vol. 58, No. 80). A step-by-step analysis of the questions proposed in the Federal Register announcement provides an introduction to possible consumer concerns about biotechnology-derived foods, as well as potential logistical challenges facing food producers and processors if mandatory labeling were proscribed. This discussion will include some of the applications of biotechnology currently on the horizon and how some of the questions raised by the FDA might apply to these foods. In some cases, particularly in the case of potentially allergenic proteins in an unexpected food source, labeling might be a reasonable alternative. In other cases, particularly those for which no safety issue is involved, labeling may be impractical and costly, and provide no additional consumer benefit.

A small number of biotechnology-derived processing aids (e.g., enzymes) have been approved by the FDA and are currently being used commercially. An example is the enzyme rennet, the first biotechnology-derived product which was approved for use in 1990 for the production of cheese. The traditional source of rennet for cheesemaking is the stomach of veal calves. Biotechnology has been used to duplicate the calf gene which codes for the enzyme and genetically engineer a microorganism to produce the enzyme via fermentation. The enzyme is identical to calf rennet, an ingredient which is “generally recognized as safe” (GRAS) by the FDA. The label on the final cheese product does not indicate the source of the rennet used in manufacture. Biotechnology can be used to enhance production of many other enzymes and microbially-derived food ingredients where the final product will be identical to a counterpart that is already a GRAS food ingredient. Since the products are identical to their natural counterpart, labeling is probably not a significant issue.

For other biotechnology-derived foods, labeling issues will be more complicated. The FDA has requested additional information on specific characteristics of foods derived from genetically engineered plants that distinguish them from other foods.

Should labeling be required for foods that contain proteins not previously found in foods?

Bacillus thuringiensis (Bt), a soil microorganism, produces a protein which is toxic to certain insects. The organism has been used as a natural biopesticide for over 20 years. The gene which codes for the toxic protein has been genetically engineered into tomatoes, corn, cotton and many other crops, and the engineered crops are resistant to insect damage. Humans may have consumed very small quantities of the Bt toxic protein if fruit and vegetables harvested from fields where Bt was used as a natural pesticidal agent were not washed thoroughly prior to consumption. However, the quantities ingested would be extremely low. Therefore, Bt might be considered a new protein in the diets of humans. Some important questions to ask might be: How much of the protein is going to end up in the food? Are there any conditions under which the protein might be toxic to humans? The same questions need to be asked whenever genetic material from exotic plants or microorganisms not previously a part of the traditional food supply is engineered into the food crops. There are numerous examples of plants which have been engineered to resist microbial, fungal and viral diseases that fall into this category.

Should foods containing higher or lower concentrations of proteins already present in the food be labeled?

An emerging technique which has found application in plant genetic engineering is antisense technology that has been used to selectively inactivate single genes. In effect, antisense can be used to decrease or totally eliminate production of a specific protein. This has been used to inactivate an enzyme involved in softening of tomato fruit, thus extending the shelf-life and improving the quality. One might imagine genetically engineering plants to contain elevated levels of amino acids, vitamins such as beta-carotene, or antioxidants. It is possible to increase the level of structural components such as starch in potatoes; a relatively small increase in solids content dramatically decreases the absorption of oil into french fries or potato chips during deep fat frying. Supplementation of pigs with recombinant DNA-derived porcine somatotropin decreases fat content of pork.

Certain microbial starter cultures are capable of producing natural inhibitory agents (e.g., bacteriocins) that function as antibiotics. For example, many of the lactic acid bacteria used to make cheese, yogurt, and fermented meat and vegetable products naturally produce these antibiotic-like compounds. Bacteriocins are not very effective as natural preservatives in foods because they are present at fairly low levels. Strains could be genetically engineered to overproduce bacteriocins to help ensure the safety and extend the shelf-life of fermented products.

The first genetically engineered organism to receive regulatory approval is a strain of Saccharomyces cerevisiae (baker’s yeast) that produces elevated levels of carbon dioxide, the compound responsible for leavening of bread. This is accomplished by insertion of additional copies of genes coding for two enzymes that are naturally present in yeast and are responsible for starch metabolism. The strain was approved for use in England in March 1990.

Should products containing unexpressed genetic material be labeled?

There are numerous examples of crops where biotechnology has been used to improve the nutritional quality or processing characteristics of a food. Most cereal grains are deficient in certain amino acids. Genetic engineering can be used to construct varieties of corn, rice and wheat that overproduce these amino acids; the result is a higher quality protein which provides more nutritious cereal-based products. Changing the ratio of amino acids in cereal grains could have a dramatic impact on the nutritional quality of the diet, particularly in Third World and developing countries. As we understand more about the role of saturated and unsaturated fats in health and chronic disease, it will be possible to use biotechnology to alter the degree of saturation or chain length of fatty acids in oil seeds such as canola and soybean. It may be possible to increase the level of vitamins, trace elements, minerals or antioxidants using biotechnology. It may even be possible to improve the digestibility and bioavailability of nutrients using genetic engineering.

Improved food processing of foods is also a goal of biotechnology. Many starches are chemically or enzymatically modified to confer specific processing characteristics. Genetic engineering could be used to allow the plant to modify the starch as part of the plant’s normal metabolism.

Should ingredients that have been derived from genetically engineered plants but not affected by the genetic modification be labeled?

Many crops which serve as source material for food ingredients may be genetically engineered to enhance unrelated properties such as disease-resistance, insect-tolerance or stress-resistance. For example, the gene(s) introduced to confer disease- or virus-resistance in corn may not affect starch biosynthesis in any way. Ingredients might also be derived from stress-tolerant plants engineered to thrive in colder or warmer climates, or in saline or alkaline soils where the biosynthesis of the ingredients of interest is not influenced by inserted genetic information. Milk and meat from bovine somatotropin (bST)-supplemented cows might also fit in this category, as the protein hormone bST affects metabolism but does not affect the composition of the final products.

Should foods containing genes derived from foods known to be commonly allergenic be labeled?

Potential allergenicity of proteins derived from introduced gene(s) is an issue of concern to many people and the FDA will be requesting further comments in this area. A key question will be what foods should be considered allergenic? Much is known about commonly allergenic products such as peanut, fish and shellfish, and milk. Clearly, if corn has been genetically engineered to contain a commonly known allergenic protein (e.g., peanut protein) the FDA will require labeling of the product. This could be handled in a relatively straightforward manner by indicating the presence of peanut protein on the ingredient label. There are approximately 125 known allergens although a comprehensive data base is not readily available. Other issues include the technical feasibility of testing for the presence of allergenic potential. Is it possible to predict allergenic potential of proteins? There is particular concern about the allergenicity of proteins from sources not previously a part of the food supply.

Should foods be labeled relative to the source of DNA?

Biotechnology provides a means for introducing DNA from unrelated species into plants. Even though proteins derived from animal genes are made up of amino acids and would be digested just like any other protein, additional concerns arise when these transfers occur. As discussed earlier, antifreeze proteins from Arctic fish have been inserted into vegetables. It may be possible in the future to transfer animal genes into plant food sources. Strict vegetarians and members of some religious groups may not want to consume a vegetable product with animal-derived proteins in it. Labeling for religious or dietary preferences is not currently considered within the perview of the FDA; certification mechanisms are in place for religious groups with dietary laws.

Interestingly, due to conservation of certain genes throughout evolution, many living organisms from diverse species already possess identical genes; therefore, it could be difficult to determine whether or not a plant has been genetically engineered or to determine the source of the DNA.

What are the practical difficulties and economic impact of labeling genetically engineered foods?

Labeling of biotechnology-derived foods and ingredients raises several challenges for the food processing industry:

  1. In many cases, it will be impossible to distinguish between constituents that are introduced via genetic engineering versus traditional plant breeding techniques, particularly in those cases where the cloned gene products are already present in that food. In addition, for most plant varieties the normal range of concentrations of specific constituents and the variability of those constituents among and between species are not currently available as a basis for comparison of native and engineered varieties.
  2. Rapid, reliable and inexpensive methods would need to be developed to identify engineered varieties. These do not currently exist for many of the genes or gene products which are of interest to biotechnologists.
  3. In the absence of analytical methods for determining whether or not a plant is engineered, tracking systems for plants and plant-derived ingredients at every stage of the food chain (from the seed to the processor to the grocery store to the consumer) would need to be instituted. For example, a genetically engineered variety of corn may serve as the source of several different ingredients used in processed foods (e.g., starch, corn oil, etc.). If a label were to be required on the final product, farmers would need to segregate seed and maintain separate harvesting systems for native vs. engineered varieties. Food processors routinely receive the same ingredient from several suppliers and they would need to maintain systems for tracking and segregating supplies in storage facilities and processing plants. Logistically, it might be easier to label fresh vs. processed products; however, this would also require segregation and some kind of certification system. The impact and total cost for instituting tracking and separate handling systems is not known; however, the cost of the systems would probably be passed on to consumers through higher food prices.
  4. Labeling would encourage more vertical integration in the food industry, as food processors would want to control the source of their raw materials and document the history of their products from the seed to the processing plant. This would influence the structure of agriculture.
  5. Labels or symbols on labels are commonly used to alert consumers when a safety issue might exist for certain individuals (e.g., labeling of products containing aspartame due to the negative health effects on individuals with phenylketonuria). A label indicating that a product was derived using biotechnology might be perceived by consumers as a warning signal even when no safety issue exists. Mandatory labeling of biotechnology-derived foods is obviously a complex issue. How it will be resolved will have a dramatic impact on every sector of the food system from the farmer, to the processor, to the retailer, to the end consumer. Readers are encouraged to participate in the ensuing dialogue and public debate about these issues. All of us who eat food have a vested interest in the outcome of the debate.