Genetically Engineered Foods: Fears and Facts
An Interview with FDA's Jim Maryanski

    Genetic engineering of fruits and vegetables and FDA's policy concerning   
these foods have been the subject of many consumer questions recently. To      
help answer the questions, FDA Consumer writer Mary Alice Sudduth talked to    
James Maryanski, biotechnology coordinator in FDA's Center for Food Safety     
and Applied Nutrition.

Q: What is "new biotechnology" in reference to food plants, and how does it    
differ from old biotechnology?
A: All plant breeding involves genetic manipulation of plants. There are       
hundreds of new plant varieties introduced every year in the United States,    
and all have been genetically modified through traditional plant breeding      
techniques--such as cross-fertilization of selected plants--to produce         
desired traits. This is "old biotechnology."
    The new biotechnology--known variously as gene splicing, recombinant DNA,  
or genetic engineering--is actually an extension of traditional plant          
breeding. It involves direct modification of DNA, a living thing's genetic     
material. This new technique is more precise, making it possible to direct     
and predict changes without introducing extraneous, undesirable traits. The    
new technique also will allow scientists to introduce genes from essentially   
any organism into a plant.

Q: Why do we need these plants and the foods they produce?
A: Plant breeders have a limited pool of genes--and, therefore,                
traits--available for use in improving plants. By looking at bacteria and      
animals, scientists can find other traits that may expand the number of        
potentially useful traits. These may include size, solids content, or          
resistance to certain viruses or fungi.

Q: Under what circumstances will FDA require labeling of genetically           
engineered foods?
A: One important area is that of potential allergens. If a gene from a food    
that commonly causes allergic reactions, like fish or peanuts, is inserted     
into tomatoes or corn, where people would not expect to find allergens, then   
the vegetables would have to be labeled to alert sensitive consumers. If       
companies can demonstrate scientifically that the allergenic component was     
not transferred to the vegetable, no special label will be required. FDA's     
policy states that proteins taken from commonly allergenic foods are presumed  
to be allergens unless demonstrated otherwise.
    Labeling also could be required if the nutritional content of the food is  
changed. Tomatoes are a major source of vitamin C, and if someone develops a   
tomato that no longer contains vitamin C, then that will have to be            
disclosed. So we envision a number of circumstances where labeling will be     
necessary, and we'll use the same labeling regulations we've always used       
under the FD&C Act. We've invited public comment on this issue, because we     
anticipate consumers will have diverse opinions about genetic engineering and  
about what information should appear on labels.

Q: FDA has emphasized the importance of proper labeling of foods and has       
initiated legal action against certain products--Citrus Hill "fresh" orange    
juice, for example--because of misleading labeling. How does this differ from  
labeling biotechnology-derived foods? Isn't the fundamental issue the          
same--full disclosure?
A: The law says labeling for foods must disclose information that's material,  
as well as avoid false or misleading statements.
    It's our view that the method by which a plant is developed by a plant     
breeder is not material information in the sense of the law. For example, we   
do not require sweet corn to be labeled "hybrid sweet corn" because it was     
developed through cross-hybridization. And plant breeders have many other      
traditional techniques through which they coax nature to change genes that     
would not occur otherwise. A process called somoclonal variation allows        
breeders to take advantage of natural mutations in plant cells that produce    
desired traits. Through embryo rescue, breeders nurture embryos produced by    
crossing two plant varieties that would not breed naturally, producing         
potentially useful plants that would not have survived on their own.           
Historically, we have not required this information to be on labels. It would  
not be practical.
    If genetic engineering or any other technique changes the composition of   
a tomato in a way that it's really not the same tomato anymore, then it would  
have to be called something different. It would have to have a different       
varietal name, or, if it was a significant difference, it may not even be      
called a tomato.
    We are reviewing the comments from the public on this issue to see if      
there is a basis for modifying our policy. However, if we were to require      
labeling of all genetically engineered fruits and vegetables, it would not be  
merely a matter of putting a sticker on a tomato or a banana. Producers would  
have to segregate the genetically engineered foods from other varieties,       
whereas normally those are all just lumped into a barrel. And then what        
happens when you make tomato paste, and that tomato paste is used to make      
pizzas? Does the label have to follow along through the food processing        
chain? It would increase the cost of these foods to consumers and would        
disrupt our complex food distribution system.

Q: Are there environmental risks involved in producing genetically altered     
foods?
A: Potential environmental risks from these crops are the same as those that   
occur in plants developed by traditional methods. There are many complex       
issues--such as potential transfer of traits to other plants and potential     
adverse effects on other organisms, particularly endangered species--that      
need to be taken into account. For example, if a plant has many wild           
relatives, it could pass a gene to one of those, resulting in an outcross      
plant species with some undesirable traits. The wild plants could develop      
into a fast-growing weed species, for instance. For some crops, like           
tomatoes, this really isn't a risk in the United States because there are      
few, if any, wild relatives of tomatoes that could be bred accidentally. But   
there are other plants, such as soybeans and squash, where there will be       
potential for an outcross species. Whether outcrossing is a problem depends    
on the trait that's introduced, as well.
    Environmental risks are looked at during the research and development      
phase. Right now, crops developed using recombinant DNA methods are reviewed   
by the U.S. Department of Agriculture's Animal and Plant Health Inspection     
Service. If FDA acts on a food additive petition or GRAS [generally            
recognized as safe] petition for a new ingredient, we will do an               
environmental assessment. We will look at what the other agencies have done,   
and, to the extent possible, we will rely on their information. If there are   
other issues, they will have to be considered.

Q: How can consumers be sure companies producing these foods will test them    
adequately and take all necessary measures to ensure they are safe for         
consumption and will not harm the environment?
A: All the companies that we've talked to are doing the kinds of tests that    
we would think need to be done. In May, FDA published a notice in the Federal  
Register providing a guide for companies that establishes a standard of care.  
What's happening now is that companies are coming to us and telling us the     
kinds of tests they're proposing based on the guidance in our Federal          
Register notice. And they're asking for our advice.

Q: How long will it be before most of the foods now being developed through    
new biotechnology are available to consumers?
A: That's a good question. People have this perception that all these things   
are just coming out of the sky and landing in the grocery stores next week,    
and that's just not true. There are products, like the Flavr Savr tomato,      
that are nearing commercialization. Assuming all the safety questions are      
answered, it probably will be introduced within the next year. USDA has        
received a request to rule on whether a variety of virus-resistant squash may  
be grown without USDA oversight, so one would assume it's close to             
commercialization. But most of the products--such as insect-resistant          
produce, vegetables with increased amino acids, and low-caffeine coffee--are   
two to five years away. We're going to see a gradual introduction of these     
products over the next several years, not an avalanche.

Q: What about the possibility of plants containing animal genes?
A: Several experimental plants have been developed that have copies of genes   
found in animals, such as the "antifreeze protein" gene from the Arctic        
flounder that may make tomato paste freeze and thaw better.
    However, there really aren't any plants with animal genes, that we know    
of, that are going to be marketed foods in the near future. So we have a good  
deal of time to think about the issue of animal genes. We believe the safety   
of the proteins produced by these genes should be evaluated based on their     
characteristics. If, for example, the flounder antifreeze protein is a         
component of fish fillet, it likely would be safe to eat as a component of     
tomato paste. But proteins derived from animals that have not been consumed    
safely will be treated as new food additives.

Q: Has FDA considered the ethical or religious implications of injecting       
animal genes into plants? How will this affect vegetarians?
A: FDA is considering these issues. There are thousands of genes in a plant.   
When a scientist adds new genes from an animal, it gives that plant several    
new proteins. But these proteins would not seem to give animal                 
characteristics to the vegetable. 
    A breeder does not actually take something from the animal and introduce   
it into the plant. For example, a scientist copies the fish "antifreeze        
protein" and modifies the gene. We know the characteristic of the fish gene,   
and we can tinker with it to make a different version. The copy is what is     
introduced into the plant, and the new gene works just like any other plant    
gene. So you don't really have pieces of animals in vegetables. You have       
pieces of plant DNA that are the same as, or nearly the same as, pieces of     
animal DNA.
    One of the things people probably don't realize is that there are genes    
in humans and animals that are in plants. There is a gene that occurs in rice  
that also occurs in the human brain. Vegetarians would not avoid rice because  
of that.
    Our current view is that these modifications will not result in foods      
that violate any ethical or religious considerations. However, we recognize    
people will have different views, and we specifically invited comment on this  
issue. We're also trying to get comments from various religious and other      
authoritative leaders so that we can get some sort of official opinion.

Q: What values will these genetically engineered plants have--more nutrients,  
better taste? 
A: Both of those, and many agronomic values: better processing, freeze         
resistance. That's what using a flounder gene is all about--making a tomato    
freeze and thaw better. That protein also is being used as a model for         
developing a food additive to use in ice cream so ice crystals don't form. 
    In addition, scientists actually will be able to make food safer. They'll  
be able to reduce the natural toxins. We're already looking at ways to         
identify the allergenic proteins in foods like milk. But those things are      
many years in the future.
    Right now, it's more a matter of giving fruits and vegetables better       
shelf life and shipping properties. Most of the traits will have economic      
values for farmers and processors.

Q: Surveys of consumer attitudes have shown that most people will eat          
genetically engineered foods, but that a considerable portion will not. Is it  
FDA's role to ensure public acceptance of these foods?
A: No, it is not. Our role is to tell the public how we ensure the safety of   
foods under the FD&C Act. We tell people about the important scientific        
questions that need to be answered and about the kinds of tests that should    
be done. But we can't be proponents of the products. We can't say genetically  
engineered food is something consumers should buy. However, we can explain to  
the public how these foods are the same and how they are different from other  
varieties. 
    The U.S. government has a policy to foster biotechnology, and FDA          
recognizes that there are immense potential benefits to be derived from this   
science. It's not our responsibility to promote individual products, but we    
see that this technology does have beneficial applications. We are             
encouraging industry by working with them to ensure that safety questions are  
resolved. We're expecting rigorous testing and will not accept unsafe          
products. n
Splicing to the Nitty Gritty
    FDA's policy for genetically engineered foods covers all foods produced    
from new plant varieties developed by any method of plant breeding. The        
policy is based on the Food, Drug, and Cosmetic Act's requirements of          
post-market surveillance of foods and pre-market approval of new substances.   
This system has ensured the safety of foods and food additives for many years.
    Under the FD&C Act, GRAS substances (those "generally recognized as        
safe") are excluded from the requirement for pre-market approval. But a new    
substance introduced via breeding for which safety has not been established    
must be approved as a food additive before marketing. Genetically engineered   
food crops that do not contain substances significantly different from         
substances already in the diet will not require approval as food additives. A  
substance that is significantly different from those already in consumers'     
diets will have to be approved by FDA. 
    All foods are subject to FDA's post-market authority under the             
"adulteration" provisions of the act, and producers have a legal duty to       
ensure that the foods they place on the market meet the safety standards of    
these provisions. Adulterated foods are subject to seizure; producers and      
distributors who fail to meet their statutory duties are subject to            
injunction or criminal prosecution. These provisions have been FDA's primary   
tools for ensuring the safety of new varieties of fruits, vegetables and       
grains.
    New plant varieties routinely go through many years of testing and         
evaluation before marketing, and foods from these plants are tasted and        
tested. 
    Plants developed by genetic engineering are being subjected to the usual   
tests for quality (Is the fruit firm? Does it look good?) and agronomic        
traits, such as improved processing and pest resistance. In addition,          
companies are using new tools of molecular biology and genetics to look at     
the very nature of these genetic changes. Biotechnology not only allows        
scientists to make new products, it also provides better tools to assess       
safety.
    In addition, companies are testing for known plant toxins, comparing       
levels in new varieties with levels in parent varieties. For example,          
tomatine is a natural toxin in tomatoes, but when using traditional breeding   
techniques, scientists don't usually look for it. However, the developer of a  
tomato that is genetically engineered to stay fresh longer is evaluating       
tomatine levels in that tomato.
    FDA has told companies that they will need to do more tests on the first   
foods developed using new biotechnology, to ensure that these foods are safe.  
But as developers gain more experience with these techniques, they will not    
need to do as many tests.

--M.A.S.