The 10 minor realizations that flipped my thinking about GMOs

Written by Marc Brazeau

A recent story about GMO testing kicked off a conversation with a friend. The researchers tested the biochemicals from crops to suss out variations in food quality and composition due to genetic engineering. The new process allowed researchers to extract 1,000 or so biochemicals from the fruit of tomatoes. *

When the scientists compared the biochemicals of the GM tomato and a wide assortment other non-GM tomatoes, including modern and heirloom varieties, they found no significant differences overall. Thus, although the GM tomato was distinct from its parent, its metabolic profile still fell within the “normal” range of biochemical diversity exhibited by the larger group of varieties.

My friend was unimpressed. He said:

Can you explain how this is any different from the problematic way the USDA and certain diet plans defines nutrition. As long as foods have a similar nutrient and vitamin profiles they are interchangeable. Food isn’t just a collection of interchangeable pieces there is something that this type of researches misses about the value of whole foods taken whole and not taken apart and reconstituted into food like substances. By this logic it has been claimed that Rice Crispies are heart healthy and margarine is healthier then butter. I understand your desire to find equivalences between GM and heirloom varieties but I have my doubts that anyone who is interested in whole food nutrition would say that this is the proof you are searching for.

Adapted from an image by Amanda via Flickr.
Adapted from an image by Amanda via Flickr.

I understood where he was coming from. It’s very similar to the way I thought about the issue a few years ago when I first became aware of GMOs. I was learning about nutrition at the time and coming to similar conclusions as Michael Pollan did in his book In Defense of Food: nutrition is a science very much in its infancy and we don’t understand how its constituent parts interact very well. While waiting for credible and durable advice on sugar, sodium, and saturated fat, etc;  successful traditional diets based on whole and minimally processed foods currently provide more credible and durable guidance for making choices about what to eat.
When I first started learning about GMOs, my model was trans fats and vitamin supplements. We thought that we could engineer a food that was healthier than saturated animal fats and it blew up in our face. We thought that if eating vitamin rich foods conferred health benefits, then supplementation would be even better. Except in cases of malnutrition, that hasn’t proven to be the case. I wasn’t ideologically opposed to genetic engineering, I just figured that given our current understanding of nutrition and ecology, the technology wasn’t really ready for prime time. I figured if we couldn’t figure out margarine, then we weren’t ready to start tinkering with plants at a genetic level. Common sense, right?
It took a while to realize that was an incorrect model for thinking about GE breeding. There are a number of realizations that I went through before leaving that behind. Here are ten of them:

  1. It’s a single gene (sometimes two or three) that is being transferred out of 10’s of thousands.
    • The cartoon of a tomato crossed with a fish is wildly misleading. It’s one gene from a fish into the DNA of a tomato that contains 31,760 genes. *
    • A geneticist would say, “It doesn’t matter where the gene comes from it matters what it does.”
  2. I share half my DNA with a banana. Half the genes in me are also in a banana.
    • That doesn’t mean that taking the same gene from me or a banana would produce the same results if you inserted it into a plantain, for example. But it’s worth keeping in mind in regards to our squeamishness about boundaries that nature doesn’t recognize. (See also: 1. “a tomato crossed with a fish”)
    • Once again, a geneticist would say, “It doesn’t matter where the gene comes from, it matters what it does.” Or, “We’re all wearing the same genes.”
  3. The genes/proteins/traits for GE crops are very well thought out and chosen carefully.
    • None of them seem particularly risky if you understand them.
    • Round Up works on deactivating a plant enzyme called EPSPS. In RR crops they express a different version of EPSPS that is not deactivated by glyphosate. It has allowed farmers to use a relatively non-toxic herbicide and practice no-till farming.
    • The Arctic Apple simply silences the gene that produce an enzyme that causes browning.
    • Rainbow Papaya has a bit of the ringspot virus encoded into the DNA as a built in vaccine. Humans are not susceptible to ringspot virus. In fact if you have eaten organic papaya with a some green spots, you’ve eaten ringspot virus.
    • Bt crops express the protein from the organic pesticide Bt that is toxic to corn borers and other pests but harmless to humans. Lots of edible plants produce their own pesticides.
  4. Lots of plants produce their own pesticides. You can read more about this here and here.
  5. Introducing a food from another part of the world introduces more risk of unintended consequences than introducing a single gene into a soybean.
    • Cultivating kiwis for human consumption began just 100 years ago and only recently imported to the US market. There was no testing for allergenicity. Some people turned out to be allergic. Yet, no one thinks things like this should require new regulations or complicated testing.
    • My Irish and French (Canadian) ancestors did not co-evolve in any meaningful way with kiwis or mangos or chocolate or any of a number for foods I eat to great benefit and pleasure.
  6. Radiation and chemical mutagenesis breeding have been safely practiced for half a century.
    • Does anyone look at a bag of Calrose rice or a Rio Star grapefruit and think those are unwholesome foods?
    • Those forms of breeding are more likely to cause unintended consequences than GE breeding. You blast seeds with radiation or chemicals, get random mutations, choose the best ones and selectively breed to finish. That’s also roughly how nature works.
    • This is not to suggest that radiation and chemical mutagenesis breeding is dangerous, just that it’s all a question of relative risks.
  7. Traditional selective breeding has had negative unintended consequences. In the real world. Not just theoretically.
    • The Lenape potato is the most well known, but in trying to make a breed of celery that was more pest resistant, the breeders dialed up the amount of psoralens a variety of celery expressed. (see also: 4. Lots of plants produce their own pesticides.) That resulted in farm and grocery workers getting serious rashes and the product had to be pulled. That celery was released without allergen testing or compositional analysis. If you want to talk about being a human guinea pig, talk to those farm and grocery workers. Since those cases, breeders are more careful about voluntarily testing for potential unintended effects.
    • This is not to suggest that traditional selective breeding is dangerous, just that it’s all a question of relative risks.
  8. Contemporary selective breeding is incredibly … selective.
  9. Why isn’t the novelty of breeding with wild relatives an issue?
    • “…if a plant breeder chose to cross breed a wild relative with a plant in order to confer a desired trait of hardiness; drought, heat, flood, pest resistance – take your pick; nobody raises an eyeball. Keep in mind that we have no experience eating the wild relative, no mandated testing of toxins (which of course would be the desired trait in breeding pest resistance) or allergens. We have no experience with large scale cultivation of the wild relative, so it’s hard to extrapolate the environmental impact. “
  10. Or a novel new mutation from nature? And what if that novel plant is then selectively breed using genetic analysis to isolate which genes the breeder wants to move and whether they have indeed been moved?
    • Scientific American recently told the story of a breeder who had been sent seeds of a habanero pepper that didn’t produce capsaicin, the compound that provides the heat. They then went on to detail the way the breeder had identified the genes associated with the traits he desired and ran the genome of each new cross instead of waiting for the plant to express the traits, saving massive amounts of time and guesswork.

It took awhile to sink in, but I’ve come to understand that a GMO tomato is a tomato. The reason why testing shows that it is biochemically substantially equivalent to a GM tomato is because it is a tomato.
Instead, my question has become,  “Why aren’t concerns about tinkering with our food any less applicable to the Dutch in the 17th century, breeding nearly all carrots to be orange, in homage to William of Orange?
Or a contemporary potato breeder like Walter DeJong?

De Jong has been working with farmers long enough to know that our food supply is never more than a step ahead of devastating insect infestations and disease. Selective breeders like De Jong work hard to develop resistant crops, but farmers still have to turn to chemical pesticides, some of which are toxic to human health and the environment. De Jong enjoys dabbing pollen from plant-to-plant the old-fashioned way, but he knows that selective breeding can only do so much.

Like I said, today’s breeders know what they want to achieve and they will achieve it, regardless of breeding technique. I’ve learned that I can live with that. In fact, it’s quite exciting.

Written by Guest Expert

Marc Brazeau is the Chief Organizer and Editor at Food and Farm Discussion Lab, a food system think tank for people within the agriculture industry to share evidence-based solutions. With experience as a chef and restaurant owner as well as organizing unions in the agriculture space, he brings a unique perspective to the food-to-table movement.

Guest Expert

Written by Guest Expert

The strength of the discussions on Biofortified depend on the diversity of expertise, perspectives, and backgrounds of our contributors and guest experts.


  1. That’s pretty well summed-up Marc. Though I think there is an error in point 2: AFAIK humans only share 15% of their genome with plants (like bananas) and not 50%. That changes nothing to the rationale of your argument, but you should double-check that for the sake of accuracy.

  2. marcbrazeau, can you please make it clear whether this GM tomato is something in commerce, or something not available to consumers. A great deal of anti-GMO propaganda talks about GMO tomatoes as if they are an unlabeled fruit, while many of us keep saying that there are no GMO tomato in commercial use and haven’t been since the Flavr-Savr was withdrawn years ago.

  3. Charles, I’ve added a note on the tomato issues.
    Samuel, the 50% number seemed high to me, but a number of seemingly credible sources confirm it. Whether it’s 100% accurate or not, it’s a factoid that shifted my thinking, and as you said, the whether it’s 15 or 50%, the underlying point stands.
    Here are three sources for the claim.
    Sadly, Gene Cuisine’s source for the number, a pdf from what looks like college presentation, is no longer available.

  4. I do Ag in the classroom talks to grade 6ers, using materials from the major Ag assn ie poultry, beef, canola etc.
    I found most of the kids blissfully ignorant the meat they were eating came from an animal that was killed.
    Not sure if I addressed the issue or just created more vegans.
    Does the consumer want to know how their food is produced or just need to hear facts like the food is certified safe, produced by family (incorporated) farms that eat their own produce every day?

  5. I do not share your optimism, unfortunately, when genetic engineering is of the “protect the plant against pests” variety. I think this will end badly. It’s just another example of privitizing gains and socializing losses. We’re the guinea pigs.
    Any blanket dependency on patentable technology (by its definition implies that it’s not natural) leaves us vulnerable. No, this is bio-warfare with poisons, and it’s anyone’s guess as to who will be left standing (or crawling about on 6 legs) when all is said and done.
    I think that there are three (if not more!) very problematic aspects to herbicide/pesticide GM – the rush to market ahead of the science has caused the global confusion. If there was a decent regulatory structure, or even rigorous science at work, we wouldn’t be in this mess.
    1. Although these latest GM traits in our food supply allows for lighter or no spray for pests, that’s because the pesticide is actually now a cellular “trait” IN the plant, now to be ingested by human and animal(without the opportunity to wash or peel it off). Although we humans don’t possess the natural metabolic pathways to uptake the toxins inbedded in the food material, our gut microbiota DO possess these same shikimate pathways and therefore, the herbicide severely affects our digestive systems, where these microbes live.
    2.There was the incorrect assurance that glyphosates would not bio-accumulate in the human body, which has been proven false.
    3.Very few people would be comfortable allowing collateral damage to untargeted, benign species:,

    1. 1. Muddled, and contradicts later pieces. The pesticide is only a trait in the plant in the case of insect resistance – for herbicide tolerance the trait in the plant is merely an enzyme which is not inhibited by the herbicide in question (at least for RR). I haven’t seen anything which suggests that gut bacteria are impacted by Bt toxins, and thus in planta triats don’t impact guy bacteria at all. The suggestion that glyphosate might impact gut bacteria also falls down as the levels in food are far below those which have any observable effect on bacterial growth – to the extent one could literally fill the GI tract from top to bottom with only maximally impacted soy (ie soy with maximum allowable levels of glyposate) and you still wouldn’t achieve concentrations of glyphosate capable of having an observable effect on bacterial growth (for species imapcted).
      Samsel Seneff is crazy bunk nonsense of the highest order. So can be instantly dismissed without further discussion.
      The second link is more bafflegab from the illustrious liar in chief Dr. Seralini, who, once again, shows that if you drop surfactants onto cells in vitro they don’t like it. Literally golf clap worthy science. Meaningless when it comes to food however.
      2. The linked piece doesn’t show anything at all, even if true (MAM is notoriously poor on science (and when I say poor what I mean is that they literally just make things up and lie in a manner that’d make Joseph Smith uncomfortable), so I assume off the bat that it is made up – at best what is measured is statistical noise, although as the sample size is so small who can even tell) it doesn’t demonstrate bioaccumulation at all as the bulk of the samples test negative for glyphosate (and I posit that the positives are false, which is a possibility for any test done ever…) and given that they are not a time series claiming they demonstrate bioaccumulation is rather… odd. (I just noticed that MAM are also trying to make the claim, based on a single datapoint, that switching to organic both eliminates detectable glyphosate from urine (which, hey, it potentially could) and… and this is where we go down the rabbit hole… reverses autism symptoms. Extraordinary claims apparently require no evidence whatsoever.
      3. All farming, and indeed human activity, allows for collateral damage to benign species. It is unsurprising in the extreme that effective weed control might reduce populations of insects dependant on weeds during their life cycle. Population levels are reduced not because bugs are dying, but because habitat size is reduced – this would only be disasterous if all habitat was removed, as all habitat was not removed you get a natural population decrease dictated by available land. As happens any time land use is altered.

  6. Donna, I’m not sure those studies say what you think they say. In fact, the only vaguely scientific link completely goes against the next link. It is shown, with references, that glyphosate adsorbs to soil and is broken down by microorganisms, which in turn were shown to not be harmed by degrading glyphosate. And the breast milk thing you linked is spurious. Show a real study instead of just concerned moms.

  7. Rainbow Papaya has a bit of the ringspot virus encoded into the DNA as a built in vaccine. Humans are not susceptible to ringspot virus. In fact if you have eaten organic papaya with a some green spots, you’ve eaten ringspot virus.

    I’ll be a pedant here and point out one thing because I see it all the time and it bugs me. The Rainbow Papaya does not have “a bit of the ringspot virus encoded into the DNA as a built in vaccine”.
    As I understand from what I have read elsewhere, in these plants there’s a viral surface protein that is being expressed and it competes for receptor sites with the virus –thus denying it a chance to enter plant cells. It does not stimulate B cells to produce targeted antibodies that neutralize the virus or T cells to specifically destroy infected plant cells. It can’t do this because, as far as I know, plants don’t have an acquired immune system (though plants apparently share at least some components of the innate immune system with us in the form of what amounts to toll like receptors –but that’s totally unrelated to vaccination).
    I know it’s only a metaphor and I know it could be considered pedagogically appropriate as a lie-to-children but it still kind of bugs me.
    Besides, if we repeat this often enough the antis will start worrying about Hawaiian papayas getting autism and we shall be responsible for much anxiety.

  8. That doesn’t mean that taking the same gene from me or a banana would produce the same results if you inserted it into a plantain, for example.

    Actually, with a lot of genes, it very well might produce the same results. There’s been a lot of work which has involved rescuing yeast mutations with, for example, human genes*. One recent paper has looked at doing it systematically and it worked almost as often as not.
    * I think the reasoning is that if one gene in one system has the ability to rescue a lethal mutant of a homologue in a different system it provides evidence that their function might be the same in both systems (so if you know what it does in one system this kind of experimental evidence gives you insight as to the function in the other system).

Comments are closed.