A few weeks ago, the Union of Concerned Scientists released a policy paper entitled "The Healthy Farm: A Vision for U.S. Agriculture," which is exactly what it sounds like.

A healthy farm practices sustainable agriculture, which means it must do three things well:

Productivity. A healthy farm produces food in abundance.

Economic viability. A healthy farm is a thriving business that provides a good living and fair working conditions to those who work on it, and contributes to a robust local and regional economy.

Environmental stewardship. A healthy farm maintains the fertility of the soil and the health of the surrounding landscape for future generations.

Current industrial farming practices in the US accomplish the first and second goals quite well, but these practices tend to be unsustainable and fail the "Environmental Stewardship" plank pretty miserably. The UCS's concern about the dire state of our food system is well-founded, and I applaud their efforts to get out in front of the policy debate. There's just one problem: they oppose using all of our technology to help combat this problem. Specifically, I'm talking about genetic engineering (GE) and genetically modified organisms (GMO).

Conversations of this sort inevitably devolve into ad hominem attacks on the GMO supporter's credibility, so before going further, let me state clearly and for the record that I do not now, nor have I ever, nor do I ever plan to work for any company that produces GMOs. Neither have I ever received any form of compensation from any such company.

Nevertheless, I think that using genetic engineering to improve our crops can help move us towards more productive, healthier, and yes, more sustainable farming practices. Unfortunately, there's a lot of misinformation standing in the way of public acceptance of this technology. Since I'm an immunologist, today I'm just going to address a single piece of that misinformation. From UCS:

[GE crops] may produce new allergens and toxins[...]

This statement is at best wildly misleading and at worse an all-out fabrication. For an organization dedicated to informing citizens about science, I'm a bit appalled that they got this one so wrong. But in order to explain why, I first need to explain a bit about genes, proteins and how these things interact with the immune system.

From Genes to Proteins

If you're already well acquainted with the Central Dogma of molecular biology, feel free to skip ahead. For the rest of you, your memories of genetics may be a foggy recollection of a monk and his peas. But don't worry, I'm not going to ask you to draw any punnett squares. The key thing to know is that your genetic information, encoded in your DNA, is a blueprint for the production of proteins*.

Proteins are the things that do work in the cell. They can do everything from providing structure and support, to communicating information between cells, to sensing the outside world, to catalyzing chemical reactions. Basically, if there's a job to be done in a cell, it's a protein that's doing it. Proteins are fundamentally a linear sequence of small units called "amino acids." In the same way that you can take a finite set of lego blocks and build almost any shape, evolution has selected for a finite set of about 20 amino acids, but these 20 blocks can be fit together in many different ways to make many different shapes of protein. Those different shapes determine the multitude of different functions that proteins have in a cell.

Because of the molecular biology revolution, we now have a pretty firm understanding of how a cell reads a particular sequence of nucleic acid (that's the "NA" in DNA), and translates the code into a sequence of amino acids that becomes a protein of a certain shape and function. And one of the most amazing features of this process is that the language is the same regardless of the sort of cell you're talking about, be it plant, bacteria, virus or mammal**. This is all very neat in theory, but it has profound consequences in practice.

For example, the insulin that diabetics need to stay alive is just a protein. Before genetic engineering, the vast majority of insulin was isolated from the blood of cows or pigs - these sources were not particularly reliable, and insulin from animals is not exactly the same as human insulin, leading to potential adverse reactions. In the 1980's, scientists realized that they could use genetic engineering to make actual human insulin in bacteria. They isolated the DNA sequence code for the human version of the protein and inserted it into the genome of E. coli bacteria. The bacteria don't know the difference between a human gene and a bacterial gene - it's all just DNA! The bacteria read the code, and turned it into protein - the exact same protein that your own β-islet cells make in your own pancreas; it's identical.

This is the same process used in genetic engineering of crops - moving a gene code for a protein or group of proteins from one organism into another. More on that later.

Proteins and Allergies

An allergy is essentially an immune response to something that's not normally dangerous. Those pollen grains that are the source of so much misery don't actually pose a threat, but your immune system may react as if it is. Your immune system makes particular antibodies called IgE that are able to bind some protein from the pollen. Those IgE antibodies coat the surface of mast cells, which are filled with a bunch of reactive molecules like histamines that make your immune system freak out. Mast cells evolved to combat parasitic worms and other infections, and when the immune response is directed appropriately, it's a good defense - a little bit of inflammation is better than an infection.

When it's directed against something abundant and harmless though, that's when suffering ensues. Immune responses to all sorts of things have been reported, from the relatively common seasonal allergies to different types of pollen, to dust mites, to semen. Though these allergies can be quite unpleasant for the afflicted, but are usually not life threatening. Allergies to food, on the other hand, can be significantly more severe.

Because food allergies can lead to anaphylaxis and death, it's perhaps understandable that people are worried about manipulation of food. But remember - allergies are a response to a particular protein. Our immune systems can distinguish between different proteins quite well, but is completely unaware of the source of that protein.

Case Studies on GMOs and Allergies

The Premise

Before getting started, let's go back to the statement from UCS that I find so objectionable:

[GE crops] may produce new allergens and toxins [emphasis mine]

This is patently false - genetic engineering techniques allow us to precisely add genes of known structure and function to crops. It would in principle be possible to engineer corn that expresses anthrax toxin, or introduce peanut allergens into soybeans, but this would have to be by malicious intent of the scientists, not some accident. We know how genes work, and we know what kind of protein an individual gene will make.

Contrast this with a common tool of breeding in organic and non-GMO farming: Mutation Breeding. This is a technique whereby farmers expose seeds to large doses of radiation or chemical mutagens, and then selectively breed the seeds that have useful traits. This process may introduce hundreds or thousands of mutations into the genomes, and breeders cannot know where those mutations are. These mutations will change the shape and functions of proteins, and could, in principle produce new allergens. Despite the fact that this process is manipulating the genome, it's not considered genetic engineering, and is allowed to be called organic.

Now, some examples of the most common types of GE crops.

Bt Corn

Different strains of the bacterium Bacillus thuringiensis (Bt) can produce proteins that are toxic to various invertebrates. These proteins, called "Cry toxins," have been used in agriculture for almost 100 years - bacteria cultured in a certain way can be induced to create these proteins, and then sprayed onto crops. Certain types of insects are susceptible to eating these toxins and will die upon ingesting them. Bt Cry proteins are among the safest insecticides that can be used in agriculture, and there are many varieties that target different types of insect pests. Since Cry toxins are proteins, that means they are coded for by genes, and scientists realized that they could do away with the bacterium entirely.

In much the same way we can produce human insulin in bacteria, we can get corn (and other plants) to produce bacterial Cry proteins - and scientists did. The protein is produced predominantly in the leaves of the corn, and insects attempting to feed on the leaves ingest the Cry proteins at the same time and die. The protein isn't expressed much in the corn kernels themselves, which is actually a problem for farmers wanting to use these crops to stave off insects that attack the ear, but it also means that humans enjoying that corn-on-the-cob are not going to be ingesting much either.

So, Cry proteins are safe to consume, they're expressed in very low levels in the food we eat, and they're sprayed on organic crops in huge quantities (and have been for almost a hundred years). There's no reason to assume that Cry produced by corn is any different than Cry made by bacteria - it's the same gene, so it's the same protein.

Fishy Tomatoes

One of the horror stories often trotted out by GMO opponents is a tomato plant that was genetically engineered to resist frost. The winter flounder fish has "antifreeze" in its blood to allow it to survive in extremely cold waters. Scientists realized that antifreeze in plants would be incredibly useful - frost damage costs farmers hundreds of millions of dollars every year in lost crops or decreased productivity.

Now, I can understand why antifreeze in your food might sound scary, but this isn't the stuff you put in your car. The antifreeze in the fish is just a protein called AFA3, and as you've probably gathered by now, that means it's coded for by a gene. Unfortunately, when this gene was put into tomatoes, it didn't actually provide much frost resistance, and these tomatoes were never brought to market, but I think this is an instructive example - if you could eat flounder without an allergic reaction, you could eat these tomatoes.

Potential for Harm

There are many examples of new GMO varieties that are using genes for proteins that don't have a 100 year history like Bt, or aren't usually ingested the way that flounder is. But there's nothing magical about genetic engineering - it's just about proteins. Most proteins are readily destroyed in our stomach and small intestine, broken down into their constituent amino acids and absorbed into our bloodstream, regardless of whether that protein comes from a cow or a tomato or a bacterium. Our digestive systems and our immune systems are oblivious to their origin.

It's impossible to claim that there's zero risk from using GMO technology in our food, and it's worth testing the safety of anything new that we put into our mouths. Safety tests are done of course, but it would be impossible to eliminate all risk.

But a possibility of risk alone is not a valid reason to avoid a technology. As I mentioned above, mutation breeding is at least as likely to generate new allergens, if not more so. At least with GE, we know what genes are being changed, and we have better tools for testing the proteins that they code for. We’ve embraced many technologies that have risks, from microwave ovens to cell phones, and there’s more at stake here than quick meals or communication. In order to feed the billions of people on our planet without doing (more) irreparable harm to the environment, we need to be thinking about all of our options.

It's also worth noting as Pamela Ronald did in this space two years ago:

There is broad scientific consensus that genetically engineered crops currently on the market are safe to eat. After 14 years of cultivation and a cumulative total of 2 billion acres planted, no adverse health or environmental effects have resulted from commercialization of genetically engineered crops.

Please note: I will not address comments here related to the myriad other complaints about GMOs - this is a post about allergens, but there are a number of other resources to check out:

GMOs do not cause cancer.

GMO's have not driven Indian farmers to suicide

Herbicide resistant weeds are a problem, but not one unique to GMO

A Survey of Long Term GM Food Studies (collected papers)


* Not all genes code for protein. There are also gene products like microRNAs, but these largely have an effect by regulating the expression of proteins.

** Not exactly the same, it turns out. Some species have slight modifications to the code, but it's more like having different dialects rather than a different language.

Images: top: by author; Jessica Reuter; Dcastor; United States Department of Agriculture.

UPDATE (6/1/13): The UCS believes I have misrepresented their position here. They do not oppose GE technology per se, but (I'm paraphrasing here) think that it is under-regulated and that funding and research should be focused on other technologies. See comments below and also their website.