Recently, I discovered in my Facebook feed an interview with Dr. Neil deGrasse Tyson in which a question was asked to Dr. Tyson about his opinion of genetically modified organisms. Here is a link to the original video:
Dr. Tyson brought up some good points, but predicated his argument on something that I’ve heard a lot with regards to recombinant DNA modification of organisms: that this manipulation is just like the manipulation that we have done with selective breeding over many thousands of years.
I have debated this internally for quite some time when forming my own opinions, and while there is more truth than falsehood to this, it misses a subtle difference between the two: that laboratory manipulation of genes is not bound by evolutionary fitness, whereas selective breeding is. Selective breeding happens in the field, where the organisms are still required to sustain fundamental fitness; transgenic organisms are not subjected to the feedback loops which determine that an organism is unfit.
We are able to make rapid, discontinuous alterations in the genotype of these organisms. The systems in which we have selectively bred our food have played as important a role in their current characteristics as we ourselves have; the behavior of a change to this system becomes increasingly uncertain as the change becomes increasingly divergent from the status quo. In other words, it’s very, very hard to predict the outcome of any change we make to an ecosystem, even if it’s to a single plant in that system: how will it respond to our alterations? With selective breeding, we are able to alter the circumstances in a gradual, controlled way, and tweak our attempt as time goes on. The single leap of direct gene manipulation opens up vulnerabilities for unintended feedback from the system. Therefore, extra caution is a practical step to take with transgenic organisms; we cannot treat them on par with selectively bred varieties.
A single very fast iteration requires a different perspective than many minor slow iterations.
I decided to construct an open letter to Dr. Tyson which hopefully captures this difference, and how the distinction is being undervalued specifically in the pre-release and approval studies done to transgenic organisms.
I decided to open the letter to general reading because I feel, as Dr. Tyson seems to, that there are a lot of extraneous issues being dragged into the discussion of the technology of genetically modified organisms. I wished to offer up a counter-argument that focuses specifically on the science, and not on food justice or corporate practices. Hopefully any reader will find the letter an inspiration to learn more about this fascinating technology and the good and bad it can bring us.
I know you’re getting a lot of criticism, not all of it constructive, about the issue of genetically modified organisms. I know you can’t possibly read all the comments, but I would respectfully request that you consider the one that follows. I hold your insight across various topics in the highest regard, and believe you do as good a job as any man can do in being a figurehead for science. To this end, I hope that my experience as an entrepreneur in the ethical food space will help provide clarity specifically on the politics and consequences of recombinant DNA manipulation which you can integrate into your ever-practical scientific viewpoint.
I apologize if the statements which I have viewed of your perspective have been taken out of context (as most controversy-causing sound bites are), but I was struck by an underlying assumption that you expressed in your comments: that modification by recombinant DNA is equivalent to selective breeding, and thus we have been eating genetically modified food for millennia. I have, I believe, a concrete argument for why these techniques are not qualitatively the same. I am skeptical that it is safe to assume all methods of modifying a genome are equivalent, and perhaps I can convince you to share in my skepticism.
You are quite correct in that we humans have been eating domesticated variants of natural organisms for thousands of years, and it’s not an inherently valuable thing to only eat wild food. Corn is far removed from teosinte, and potatoes are tamed versions of nightshade. The wild variants are certainly unpalatable; you can even die by eating wild variants of food we eat on a daily basis. Then again, you can die from eating foods that we do eat on a daily basis: apples and stone fruit contain cyanide-like compounds in their seeds. Eat enough of them, and you can be poisoned.
The critical thing to understand is that we have learned the proper consumption of domesticated food variants by induction, having experienced generations of practical testing: though we can cause an organism’s genome to drift, the differences between parent and child are minor enough on each successive generation to imply that the characteristics of the child will be substantially equivalent to the characteristics of the parents. The pace of development allows us to generate observational rules regarding that organism, and, crucially, the accuracy of our knowledge comes from the degree with which that organism has consistent characteristics. In simpler terms, our ability to understand a food is based on the food’s consistency across time, which generally changes slowly. We can see the trends as they happen, compare yesterday’s food with today’s, and adjust as necessary.
The vast majority of large genetic shifts across a single generation within selective breeding create an unsuitable organism: it dies young, is sterile, is environmentally uncompetitive, or is otherwise buffered against a rapid shift into being an entirely different thing. This is apparent on some of the foods you mentioned; our seedless varieties of grapes and watermelons, for example, have been pushed into being non-viable, reproducing through cloning instead of sexual reproduction, and they are no longer strains available for selective breeding. It’s as far as we can push those fruits; new varieties must be pursued with the reproductively viable upstream varieties.
This low pass filter enables genetic drift, but dissuades genetic discontinuities. If you bend the genome too hard it will break. This is one of the inherent safety factors that enables us to modify a genome with consistent understanding of the organism. The differences between generations are limited to some predictable subset of all possible outcomes: we don’t have apple trees, for instance, that will suddenly produce belladonna berries or their associated poisonous alkaloids. We know the hotspots to look for with apples through our experience with them and thus know how to interact with them safely.
Bypassing this inherent consistency is the point of recombinant DNA techniques: direct genetic modification is arbitrarily variable and can successfully modify the genome at a rate and for a purpose that cannot be achieved through natural selection. This forms the basis of my counterargument: recombinant DNA modification of a genome is not just different in degree from selective breeding, it is also different in kind.
Genetic manipulation’s aim is to progress rapidly to an improved version of the organism (for whichever definition of improvement we wish to create), out-of-band of the constraints of evolution, including configurations which would never occur even if selectively bred for. We’re modifying organisms orthogonally to the vectors which have in the past provided a framework which introduce some level of consistency of characteristics over time. We’re modifying organisms without being constrained by the generations upon generations of fitness that evolution has baked into the genome. We can no longer use the parent organism to conceptually prototype the new child variety, for better and worse. We are not limited by the parent’s restrictions, nor are we protected by the parent’s safeguards.
We have become the Creator, with His or Her role in intelligent design; we must take responsibility to carefully craft our children. Should we make a mistake, the fault, dear Dr. Tyson, is not in our stars, but in ourselves, that we are underlings. A requisite level of caution should be applied before we feel satisfied that our efforts have resulted in the exact set of characteristics we intended to create.
The crux of the debate over the application of recombinant DNA technology revolves around our ability to execute precisely our intent, no more, no less, including avoiding regressions to a less-fit organism than the progenitor. This involves biology, chemistry, agriculture, and anthropology, from microbiological manipulation of DNA to proteins passed in vivo through our feedstock to ourselves. This is extremely complex, and is not deserving of scientific scorn, but of scientific scrutiny.
It’s not that we understand nothing about our new varieties; in fact, genetically modified organisms have been studied in significantly greater detail than their conventional counterparts. However, just that there are studies does not imply that they were useful studies for the purpose of completely characterizing our creations..
It is important, of course, when evaluating the state of science to understand the sociopolitical framework around the issues. It is incomplete to study Galileo without Pope Urban VIII, nuclear fission without World War II.
The FDA leaves the responsibility for vetting a new variety of organism to the producer of the organism. Specifically, the rule states:
“Producers of new foods have an obligation under the act to ensure that the foods they offer consumers are safe and in compliance with applicable legal requirements.”
The companies producing the new food are the same ones studying their safety; not the FDA or an independent party. This creates a systematic bias towards what’s good for the producer, not what’s good for knowledge: it’s not necessary for regulators to characterize a company’s product, it’s necessary that a company, focusing on its product’s success, sounds to regulators as if they have characterized the product. When they’ve studied their own product enough to sound convincing they may offer their product for sale. This is not the same as fully understanding their product within the ecosystem or our food system.
Of course the product can be studied independently after release, but this is the prime regulatory hurdle between a new product and the marketplace. Once the product is out, it cannot be recalled; genetic material has a way of dispersing throughout the environment. It would behoove us to be supremely confident in a product before it is released.
In practice, the first step in evaluating a new food product is to compare the new variety to its progenitor, a policy known as “substantial equivalence”. This is outlined by the Food and Agriculture Organization of the United Nations and the World Health organization in this way:
“Substantial equivalence embodies the concept that if a new food or food component is found to be substantially equivalent to an existing food or food component, it can be treated in the same manner with respect to safety (i.e. the food or food component can be concluded to be as safe as the conventional food or food component).”
Just below this statement, the FAO/WHO follows with:
“Establishment of substantial equivalence is not a safety assessment in itself, but a dynamic, analytical exercise in the assessment of the safety of a new food relative to an existing food.”
Genetic modification of an organism can result in unintended effects on the phenotype of that organism, such as changes in growth or reduced tolerance to environmental stress, which are readily apparent and typically eliminated by appropriate selection procedures. However, other unintended effects such as alterations in the concentration of key nutrients or increases in the level of natural toxicants cannot be readily detected without specific safety assessment.
The intent seems clear: “substantial equivalence” is an effort at defining our traditional methodology of evaluating varieties, putting on paper what I outlined above: that a variety of grape, say, can be assumed to be “grape-like” enough to form a foundation for understanding a new variety, both intrinsically and within an ecosystem. It should not itself, however, be used to determine the surety of a genetically modified food, only provide a starting point for determining surety, with diligence moving forward from there. Whereas substantial equivalence plus experience has been enough to evaluate selectively bred varieties, we need substantial equivalence plus evaluation of direct and indirect effects of insertion of the new gene to evaluate laboratory modifications of organisms.
Here is the problem: we cannot use the paradigm of substantial equivalence as the primary rubric for judging genetically modified organisms because the modification is intended to alter the variety in a radical way. It’s important to understand that the technology is not creating a “new kind of” soy, which is how we think of natural selection. Recombinant DNA can create something that that at some point might not even be definable as “soy” anymore.
Hypothetically, imagine a complete replacement of the genetic material of the soybean with that of, say, the black bean. Conceptually, this is similar to recombinant DNA techniques where every gene from the black bean has been substituted for the genetic material of a soybean. It’s clear to see here that substantial equivalence is a faulty doctrine; the offspring will be very unlike soy. While this sounds like extremism for the sake of an academic argument, we did this in 2003 with the Southeast Asian banteng (an endangered cattle variety) being cloned from domestic cattle stock. The offspring was substantially equivalent not to cattle, but to banteng, as we would expect. At what point is the organism the same as the progenitor, the gene donor, or somewhere in between? This is an entirely more complex issue than simple genetic drift.
Clearly, man-made transgenic organisms are not qualitatively the same as selectively bred ones. This does not imply superiority or inferiority, just a need for paradigmatic introspection. We must contemplate genetically modified organisms differently than selectively bred ones.
Sadly, the misunderstanding that recombinant DNA techniques is similar to selective breeding has created a de facto legal loophole: if a company can provide a plausible case that the new variety appears like the progenitor, and the gene to be inserted is well understood in its host organism, then the diligence of testing the modified organism itself is unnecessary. It sounds like you’ve proven the food safe: safe corn and a safe gene is strictly as safe as the corn plus the gene. This is why the regulatory science surrounding recombinant DNA techniques is faulty: it has been done with the assumption that a reductionist view of a complex biological system proves safety.
But biology and reductionist science rarely mix. A zebra is not merely a horse with stripes.
Let’s take Bt crops as a practical example. Bacillus thuringiensis is a soil dwelling bacteria that produces a pro-endotoxin that paralyzes the digestive tract of certain susceptible insects, which causes them to stop eating and starve to death and/or have a rupture of the digestive tract. The Bt organism itself is widely used as a topical pesticide even in organic farming, because the bacteria readily breaks down in sunlight and in our digestive tract, and the protein is susceptible to degradation in acidic environments (as in our gastric fluids) but not in alkaline solutions (like the digestive tract of target insects). Finally, there is some evidence that microbial action within the insect’s digestive flora is important to the reduction of the pro-endotoxin; the bacteria doesn’t produce the bare toxin, it produces a precursor crystal to the toxin, and the crystal is converted by microbes within the insect to the toxin itself.
It is with this mode of action that Bt has been approved and understood as safe as an insecticide; however, the expression of the gene in Bt crops is a completely different application. The gene to produce the endotoxin itself, and not the pro-endotoxin, is what has been used in the modified plants; the extra synthesis within the insect has been bypassed. The protein is expressed in nearly every cell of the plant, and not just topically. This means that it is still present within the corn when it is consumed or processed, unlike the topical application which is easily degraded and washed off.
The tests for Bt crop acceptance used the approval of Bt as a topical insecticide as a springboard for proving safety; however, the foundation of topical Bt as safe for humans was not reevaluated with respect to the new method of application. The safety of topical Bt relied on the breakdown, or lack of synthesis, of the toxin before it could enter our bloodstream, and it was assumed that the behavior in transgenic crops would be similar. There are some reasons to suspect this is not the case: for example, a study which has shown that the toxin can be found not only in our bloodstream, but in the cord blood of unborn babies; thus passing the placental blood barrier. If this were verified, then it would indicate that some assumption about the behavior of Bt toxin in vivo (from whichever source, topical or transgenic) is wrong.
Testing in-vivo essentially means clinical trials, which have not been required due to time and expense. Neither the effects of the Bt toxin nor of a diet rich in Bt transgenic organisms were done on humans in a controlled study as a factor in Bt product evaluation, and even now in vivo research is mired in debate and critique. Though the organisms are out in the ecosystem, there is plenty of science still to be done.
Let me be clear: the technology of recombinant DNA itself is not evil, any more than any other technology is evil. It’s the indiscriminate or irresponsible use of the technology that should be our focus, the application of the technology. It should be treated as creating something wholly new, not as a variant of something that we’ve understood for generations, and we should be conservative before giving it carte blanche, yet almost all of our soy, 97% of US crop production at a value of $36.5 billion as of 2011, is genetically modified. 88% of US corn ($67.3 billion) is transgenic. This, in my mind, is a risky investment in technology which may well be safe, but is improperly vetted with regards to the scale of utilization.
Is the anti-GMO camp guilty of scientific fear mongering with regards to the technology and those who wield it? Partially, yes. This is a marketing technique some use to improve visibility of the issues, directed in opposition of the marketing stance of “trust us, science is on our side” that organizations with a financial stake have taken. One side encourages fear, the other discourages skepticism. The net effect is that discussions on the execution of the technology tend to be lost in the arguments floating around the network of food issues, especially when people can’t see or have difficulty understanding the fundamental techniques. Microbiology is a complicated subject, and it’s often easier to use a figurehead to represent the network of issues which are influence by the technology.
Wrapping gene patent issues, abusive business tactics, corporate irresponsibility, the destruction of small farms, loss of crop diversity, and first world diseases into the banner of genetic modification is hardly fair, but all of those things are to some degree exacerbated and influenced by the technology. With this understanding, contra-GMO movements are not anti-scientific pitchfork rallies, they’re metonymy: representing the loss of democratic control over our food production system, and how we make food and what we eat, with the lynchpin which enables these issues to persist.
This is why it is important that I write this open letter addressed to you, Dr. Tyson. You hold a role of great responsibility in fighting against the tide of uninformed or misleading science in the United States. You broaden the minds of the public, to get them to be curious about science and technology, to wow their worldview by showing them how stupendous our mundane existence really is. People listen to you when questions of a scientific nature arise.
It is important to educate the public about issues that are and are not technologically driven. It’s important to clarify the strengths and risks inherent with recombinant DNA techniques, and not dismiss the technology as an incremental improvement. It’s important to separate the technological issues from the marketing messages. It’s important to address how technology issues are separate yet intertwined with social issues. The only way we can do that is to have well-informed men of science who are trusted to give a concise and accurate digests of complex issues of concern to laymen. I urge you to scrutinize my statements, research the technology and social issues, and provide support to the general knowledge of the topic. Let’s remove fear and add transparency to our modern food system.
Should you like to rebut or otherwise respond to this letter, I encourage it and will publish to every fora in which I see this letter published. You can contact me at email@example.com. Should you not be able to respond, I completely understand. I imagine your stardust finds itself pulled in a great many directions, and I do not wish to distract you from saving science for the American people.
I shall remain respectfully yours,
Brain in the Sky, Inc.
 Aris A, Leblanc S. Maternal and fetal exposure to pesticides associated to genetically modified foods in Eastern Townships of Quebec, Canada. Reprod Toxicol. 2011 May;31(4):528-33. (Note: this study has been rightfully criticized as to its methodology, but the intent was never to prove GMO crops harmful, but to show that the science regarding GMO crops is hardly iron-clad; one of the basic assumptions in the classification of Bt crops as safe is suspect, and that there should be more study regarding this application of genetic modification.)
It appears that Dr. Tyson has posted a rebuttal to the comments he has received here before I had a chance to post my letter, which is also good reading on this subject. I have to respect the restraint that he has shown while sticking to his statements, and he addresses some of the things I talk about in my letter (e.g. aliasing all food rights issues under GMO technology).
I still feel, however, that there is a chance to discuss the science behind the technology. It still appears to me that the fundamental difference between the consequences of transgenic modification vs. genetic drift can be debated.