NOTE: Monsanto is introducing a new GM soy engineered to contain higher levels of omega-3 fatty acids, which it's promoting as offering health benefits to consumers:
http://dmedia.ucsc.edu/~bsharris/film%20170a/proj3/monsanto/monsanto/layout/featured/omega.html
http://www.indyinasia.com/2011/04/monsanto%E2%80%99s-new-gm-soy-to-grow-%E2%80%9Cfish-oil%E2%80%9D/

Natural fatty acids do have health benefits - some research suggests that they reduce the risk of heart disease and cancer, and improve the behaviour of violent and antisocial children and teenagers. They appear to help reduce inflammation in the body. They are present in foods such as oily fish, flax seeds, chia seed, and walnuts.

But scientists have warned us that genetically engineered omega-3 fatty acid soy is a completely different matter - and is dangerous. Fatty acids are directly incorporated into brain lipids and they play major roles in many diseases such as Alzheimer's and diabetes. It's recently become known that there are specific receptors for individual fatty acids on the surface of cells that can control inflammation and glucose metabolism. If the GM soy plants make low levels of some biologically active molecule related to the intended product, this substance could drastically alter the physiology of those who eat it - much like the infamous l-tryptophan impurities, which killed and sickened many Americans.

A 2008 paper (excerpts below) discusses the risks of GM nutritionally enhanced plants, including the new omega-3 fatty acid soy.

More information: http://www.gmwatch.org/latest-listing/1-news-items/11657
http://gmo-journal.com/index.php/2009/12/11/gm-soy-now-to-contain-omega-3s-a-recipe-for-indigestion/
---
---
Perspective: The Problem with Nutritionally Enhanced Plants
David R. Schubert
Cellular Neurobiology Laboratory, The Salk Institute for Biological Studies, La Jolla, California
JOURNAL OF MEDICINAL FOOD
J Med Food 11 (4) 2008, 000 000
http://www.liebertonline.com/doi/abs/10.1089/jmf.2008.0094

ABSTRACT Among the next generation of genetically modified (GM) plants are those that are engineered to produce elevated levels of nutritional molecules such as vitamins, omega-3 fatty acids, and amino acids. Based upon the U.S. current regulatory scheme, the plants and their products may enter our food supply without any required safety testing. The potential risks of this type of GM plant are discussed in the context of human health, and it is argued that there should be very careful safety testing of plants designed to produce biologically active molecules before they are commercially grown and consumed. This will require a mandatory, scientifically rigorous review process.

CONCLUSIONS
The above paragraphs summarize published data that clearly show the following: (1) Compounds structurally related to a common small molecule can have a lethal effect when present as even a minor contaminant in a food supplement. (2) The GM enhancement of a metabolic pathway by the overexpression of genes for that pathway can have unpredictable consequences in the form of synthesizing a toxin. (3) Finally, in the case of golden rice, it is argued that biologically active compounds derived from aberrant plant carotenoid synthesis could have profound effects on human development. Similar arguments can be made for NEP-derived fatty acids that are directly incorporated into brain lipids and about NEPs overproducing vitamin E. Aberrant fatty acid composition of brain lipids is implicated in Alzheimer's disease, and vitamin E has a role similar to RA in mammalian development. The excess consumption of a nutrient can also have negative effects. For example, a clinical trial with vitamin E supplementation showed that a relatively small dose increased the risk of heart failure, and smokers who supplemented their diet with -carotene had an increased risk of lung cancer. Therefore, there is a potential for nutrient toxicity in NEPs because upper tolerable levels of many nutrients are not well established and are likely to vary between individuals and lifestyles.

The information presented here shows that not only the potential harm of the product should be considered for risk assessment, but the GM process itself. The data clearly invalidate the argument that "the regulatory trigger for risk assessment should be based upon the physical features of the product rather than the process by which the product was generated." While it is true that traditional breeding methods can give rise to potentially hazardous products, the most recent assessment of GM food safety by the National Research Council stated that GM "has a higher probability of producing unanticipated changes than some genetic modification methods", but it curiously concludes by stating that the risk of GM technology is no greater than conventional breeding methods. There are, in fact, no data comparing the food safety profiles of GM versus conventional breeding, and the ubiquitous argument that since there is no evidence that GM products make people sick, they are safe (see, for example, McHughen and Smyth, Bradford et al., and Miller et al.) is both illogical and false. There are, again, simply no data or even valid assays to support this contention. Without proper epidemiological studies, most types of harm will not be detected, and no such studies have been conducted. The necessity of labeling all GM products and particularly NEPs is therefore critical if there is any hope of monitoring adverse health consequences due to their consumption. For example, it would have been impossible to identify the source of the toxic tryptophan supplement if the product were not traceable through labeling.

It follows that before NEPs producing biologically active molecules such as beta-carotene, omega-3 fatty acids, or vitamin E are introduced into the food chain, great care must be taken to do rigorous, multigenerational animal safety assessments with the hope of identifying risks to health (for methods, see, for example, the 2007 publication by the National Toxicology Program and Pusztai and Bardocz). In addition, the products must be labeled and traceable, and the unpredictable and unintended metabolic changes that may occur in NEPs require the thorough testing of the entire edible portion of the plant, not just the designated product as is almost always done by biotech companies. To date there is essentially no multigenerational animal safety testing published for GM plants and no required labeling in the United States for any GM product. In an excellent review of our current GM regulatory process, Mandel concluded that for second-generation GM products, like NEPs, "it is necessary to establish a comprehensive, efficient and scientifically rigorous regulatory system." As discussed herein there are very valid scientific concerns to support this conclusion.