The highlight of the summer was definitely Jurassic World. (What a movie!) Genetically engineered extinct animals are super cool and awe-inspiring, but unfortunately not real. In this post, Deepti Mathur and Amy Jobe tackle some genetically modified organisms that do exist: GMO crops.
GMOs are controversial, and our goal here is to provide the most comprehensive yet understandable discussion from a scientific standpoint. The GMO debate is very polarized, and although it would be convenient to have neat yes-or-no answers about benefits and drawbacks, we will demonstrate the nuances in the debate.
We’re scientists, so we’re only going to focus on the scientific aspects of GMOs rather than try to incompletely address policy or economics, which is a whole other side to the story. Here, we’ll present unadulterated facts of the technology, distinctly annotated with our comments and conclusions on various divisive issues. To understand both the benefits and downsides to GMO crops, let’s start with some background info.
What is a GMO?
GMO: genetically modified organism
GM: genetically modified, or genetic modification
Genome: all of the long DNA sequences that make up the genetic “instructions” in an organism.
Transgene: a gene that is transferred to an organism; in the context of this article, a transgene refers to such a gene transferred in a lab as an intended genetic modification.
A genetically modified organism is a plant, animal, etc. which has had its DNA altered by humans. In the case of GMO crops, this is achieved with one of several methods that allow a gene of interest to penetrate the cell, either by sheer force or by creating holes in the cell. The gene enters the plant cell and inserts itself into the cell’s native DNA, thus incorporating itself permanently – as this plant reproduces, the new gene is now treated like one of its own and is passed down to future generations. The inserted genetic material may be something that enhances or suppresses a gene that already exists in the target plant, or it might be a gene from a different species entirely.
The two most popularly used GMO crops are those engineered for herbicide resistance, and those engineered to produce pesticides. The former have genes inserted that protect the plant from herbicides, like Roundup, that are sprayed on fields, thus allowing the crop to live while surrounding weeds do not. The latter is a bit different: Bt cotton has a transgene – a gene transferred to the organism -- which causes the cotton to produce a toxin from bacteria. Insects that eat the transgenic cotton plant die from the toxin, and the plant is protected from pests.
How is genetic modification different from selective breeding?
The practice of selective breeding began thousands of years ago, as soon as humans began to plant more of the seeds from the plants that they liked to eat the most. We see selective breeding today in plants, like ever bigger, juicier fruits, and in animals – for instance, in the differences between Upper East Side purse dogs and the Great Danes that could accidentally inhale them.
A common argument in favor of GMOs is that farmers have been selecting crops for centuries. So then how can a new form of artificial modification be that different?
The selective breeding process would go roughly as follows: in a given season, a farmer finds a mix of small tomatoes and medium-sized tomatoes naturally occurring in his harvest. Plants, like humans, have inherent variation among individuals. The farmer decides that larger tomatoes are better, so for his next round he only plants seeds from the medium-sized tomato plants. The following season, his average tomato size is indeed a little larger, although there is still some variation. He again plants only seeds from the largest of the plants, and continues the same process for many years. In time, his average tomato size will be much larger than when he started out.
The genetic engineering process for the same trait would go differently: in the first year itself, a gene for large tomatoes would be isolated and/or created, and within one or two generations the farmer’s entire crop would produce extra-big tomatoes.
So what’s the big deal? The key difference between these two methods is time. Even though selective breeding directs evolution, it is much slower than the abruptness of genetic engineering. This is important because no form of life lives in a bubble – a tomato plant in the field is continuously interacting with numerous other species, from birds to bacteria. Selective breeding allows interacting species to co-evolve and adapt, while genetic engineering does not. Upsetting the genes and, consequently, the traits of one organism has unintended snowball effects on others. The size of a tomato might not matter in the big picture, but what if it was the sweetness of the fruit that was altered, which changed other animals’ diets? Or what if the larger tomatoes had a lower nutritional content?
Genes from other species
Even that change is somewhat conservative, if the tomato plant is getting modified tomato genes. But what usually occurs in GMO plants is the insertion of genes from a foreign species, like the Bt cotton described in the section above. This exposes another key difference: genetic engineering introduces a new gene that has never before existed in an organism, as opposed to selecting over generations for a pre-existing trait.
Deepti: Mixing genes from different species can hold surprises, even if we are totally sure about each part on its own. At my own lab meeting a couple weeks ago, my boss balked at the idea of putting a human gene in a mouse for a certain experiment, because he acknowledged that the well-studied human gene could behave completely differently in the mouse in ways we wouldn’t be able to easily detect. E.g., we could know for sure that a given gene affects insulin signaling in its host animal. But, it’s entirely possible that in a new setting it can have sneaky side consequences in the immune system, for example.
Amy: The origin of a transgene – whether it comes from the same species or a different species – may be less important than other aspects of its makeup. Genes of all species have regulatory sequences like bookends around them and might even have some in the middle, too. If a gene is used without regulatory sequences, which we can almost always identify, then the risk of unintended effects is reduced dramatically. Additionally, we can check the transgene for protein binding sites: the regulation of genes and their products happens when proteins bind to them. So we can check to see whether certain binding events might happen that would have unintended consequences. There are a few other variables here, but I think that those two measures would eliminate the majority of risk of trans-species genes.
In a GMO crop setting, even if studies find that the inserted gene doesn’t seem to interfere with the plant’s biology in other ways, it still affects organisms that feed on or pollinate it. Messing with an ecosystem in a rapid way that prevents co-evolution and adaptation of surrounding species can have long-lasting consequences, and we’re already beginning to see it.
Horizontal gene transfer
One such consequence that has been demonstrated is called horizontal gene transfer. The primary route of inheriting genes is from parents to offspring – “vertical” gene transfer – but genes can also pass “horizontally” through crop populations via cross-pollination and infections from viruses and bacteria. In the case of infections, bacteria or viruses take up genes from other bacteria or even from the cells of other species that they infect, and transfer them to plant or animal hosts during later infection events. Horizontal gene transfer is known to play a role in the spread of antibiotic resistance, and has now been demonstrated in the context of GMO gene uptake into the bacteria in soil.
This is a problem because it means that the transgene that was created for a certain crop is not being contained – it’s being released into the surrounding environment. Even though bacteria are small, they affect the ecosystem and food chain from the very beginning, thus affecting every higher organism. E. coli with GMO transgenes for herbicide or antibiotic resistance have been found, indicating gene transfer from plant to bacteria.
This speeds up the rate at which an ecosystem needs to adapt to these changes, and makes co-evolution harder and more unpredictable. And, of course, if an antibiotic resistance gene gets out, it has more obvious and immediate consequences. A few studies on horizontal gene transfer say that although it has been demonstrated both in the lab and in the field, it’s not a hazard because it takes many months for a transgenic bacterial population to outgrow the naturally occurring population. Frankly, that’s a ridiculous argument because the GMO plants are there for years.
Creation of monocultures
If a particular gene from a GMO were to sweep through, say, the corn population in the US, then most corn in the country would have very little variation in that gene. This is a great situation, and a fairly probable one, as long as the gene confers some big advantage, like heat resistance, or long shelf life. But this lack of genetic variation, even at one locus, could be a problem if a disease or a drastic change in climate afflicts these crops. Populations with little genetic variation may not harbor very many, or any, individuals that can resist a change in environment, so genetically homogenous populations stand at high risk for extinction, because you can always count on threatening environmental changes to occur at some point. On the commercial front, unchecked spread of GM transgenes could cause enormous problems for farmers attempting to grow non-GM crops, only to find their product infiltrated by a patented gene that they didn’t want to begin with.
Amy: Genetically homogenous populations are already a problem on farms that grow very large amounts of a single crop. It also requires homogeneity in many genes to become vulnerable to environmental changes. Spread of GMO genes to non-GMO farms hurts the chances that we can confidently label different foods as “GMO” or “non-GMO” in grocery stores, but my opinion is that such labeling is nonessential anyway. Herbicide resistance that is spreading into weeds is a more pressing issue.
Single proteins in food, pollen, and so on are the culprits of allergic reactions. It can be difficult to predict whether a given protein will cause an allergic reaction in humans, including the proteins produced by the newly cut-and-pasted genes in GMOs. GMOs are tested for production of proteins that are similar in composition to known allergens, but there remains a small risk that a completely novel allergen will arise along with a gene insertion.
Amy: I’m no more concerned about allergies stemming from a new GMO than I am about their stemming from newly imported foods, which happens occasionally – like with the kiwi, when it was imported in the mid-1900s from China to New Zealand, and then to the rest of the world. I find the risk here acceptably low.
Deepti: I agree; allergies to preexisting allergens are on the rise, but I don’t think brand new ones are an issue yet.
Haven’t GMO crops been tested for problems?
Yes, they definitely have. Each crop must be FDA approved, and a company with a new GMO seed must present data attesting to the safety of the plant. There are just a couple problems.
First, an independent academic research lab in a university is not allowed to test a GMO seed unless the company who patented it allows the lab to have access. Normally, access is granted to most labs that request it. But the company is not required to give unfavorable data to the FDA; in fact, it can repeat experiments as often as it wants to get a favorable result, and only give the positive data to the FDA.
Second, it was found that of the abundant scientific literature on GMOs, studies with financial or industry interest significantly correlated (p<.001) with a pro-GMO result. This conflict of interest makes it hard for scientists to read the literature to accurately examine the real versus false dangers of GMOs, because which papers can we even trust?
Amy: It’s upsetting that the current equilibrium leaves pro-GMO scientists and pro-GMO results in a position of power relative to those who advise caution. Federal testing is probably the most complicated GMO-related issue that’s rooted in biology. GM seeds should be freely available for strictly testing purposes, and testing results should be as openly accessible as possible. Additionally, GM crops should be tested annually, because they, like all life forms, accrue mutations over generations and may significantly depart from their patented makeup.
Apart from the ecological and environmental issues, are there any risks to humans?
For every article saying GMOs are safe, you can find one saying that they’re not. Perhaps the discrepancy is simply due to the conflict of interest discussed above, but maybe it’s also impossible to know right now. GMOs have only been around for a few decades, which is not long enough to get a reliable epidemiological answer. The only way to know for sure would be to give one set of people only GMO food from birth, and another set only non-GMO food, and follow them for decades under controlled settings. That study is never going to happen. The best we can say is that countries with popular usage of GMO crops haven’t all dropped dead yet; however, as with any health hazard, that doesn’t mean that there aren’t small but significant changes. We have to look at the basic biology and use our heads to extrapolate what it might mean.
Here’s what we do know:
As described above, a large number of GMOs are designed for herbicide resistance. This means that farmers can spray their fields with increased volumes of herbicide without the fear of killing their crops. The most common herbicide currently used, Roundup, has an active ingredient called glyphosate. As far as herbicides go, glyphosate is actually not as bad as other previously used substances, but no one would argue that it’s good for you, and the sheer volume used doesn’t help, either.
Glyphosate was previously deemed “safe” because it interferes with a molecular pathway that exists in plants but not in humans. However, more recent research conducted by independent studies across the globe have found that glyphosate harms the beneficial bacteria that reside in our gut, by inhibiting an enzyme involved in the synthesis of amino acids. There’s a plethora of literature about how the well-being of our good gut bacteria is essential for human physiological health, and this key point may be the reason why some people didn’t find the harm in Roundup – they weren’t looking in the right place. Glyphosate is not broken down by water or sunlight, so Roundup runoff into water is another potential source for consumption. Ingestion of glyphosate may cause ailments from gastric pain to the onset of kidney damage, and we may feasibly encounter enough glyphosate in our food within the coming decades to cause these problems.
The silver lining is that glyphosate is the lesser of two evils – the other evil being the more toxic previously used herbicides. Actually, though, it might be the non-active ingredients in Roundup that are worse than glyphosate. Certain types of surfactants in the herbicide are known to be more toxic than glyphosate itself. Another hole in the silver lining is that some weed species are becoming resistant to Roundup, leading to progressively increasing amounts used in the fields.
A big controversy: Roundup
Okay then, so what’s with all the articles saying that no correlation has been found between Roundup-treated crops and health hazards? Even aside from the conflict-of-interest problems, part of the issue is with the duration of a study – looking at normal doses for a few months is obviously not going to show anything; we’re talking about something that accumulates over a lifetime. Peripheral tissues also would not show anything in the short to medium term, if the issue is originating in our GI tract.
Deepti: But nobody can ignore that rates of obesity, other metabolic problems, autism, etc. are on the rise – particularly in countries where Roundup is used. While this rise in disease occurrence (even after you account for increased detection and lifespan) is clearly due to many factors, there is a huge focus on the role of diet and GI tract health in overall health. Trace amounts of any toxin – herbicide or otherwise – are a part of what we eat, and ignoring that is just bad science.
This is actually what I find most frustrating about the GMO controversy – somehow, a lot of people have been convinced that not supporting GMOs is being a bad scientist. I’m a scientist myself, so I’m allowed to say that scientists can be surprisingly stubborn at times, with “Genetic technology is good! Anyone against it is bad!” repeated like some weird mantra. But I urge my colleagues who have fallen victim to this kind of thinking to stop and think.
Dangers of herbicide use in general have been known since the 1960s – just because glyphosate is “better,” does that mean we should rush to its defense? Every single grad student in biology knows that genetic modification (be it though RNA interference, CRISPR, etc) has off-target effects. It’s how we explain away data we don’t understand, and it’s why reviewers demand multiple ways of showing that our target is likely correct. Even gene therapy, a much loftier goal than making seeds, had problems with proper targeting, and led to leukemia development in patients. With all of this real science coming from academic and medical research, don’t we all need to be more careful and responsible?
Another problem with the studies deeming GMOs “perfectly” safe is statistics. A lot of potential problems that we discussed above are fully acknowledged to be true, but dismissed as unimportant because the chances of each event occurring are low (less than .1%). If you assume the probability of a single horizontal gene transfer being .0001%, and that that event causes ecological disruption or health issues with a probability of .0001%, then the chance that a single event will do any harm is .00000001% - extremely low. But multiply that by the number of bacteria in even one square inch of soil (100 million), and we’re back to a 1% chance, without even considering the thousands of acres and multitude of years for events to occur. Each change we see in the environment will be small and incremental, but it’s flat out bad math to assume it’s not there.
Do GMOs have benefits?
Deepti: Yes, I think so. Genetic engineering has also been used to create “golden rice,” a nutrient-enriched rice in the effort to prevent malnutrition in starving areas of the world. I think it’s a very positive use of the technology, even if some policy-makers and economists argue that there are more sustainable ways to reduce starvation than developing GMOs.
Amy: Also, there are a lot of developing and potential uses, like making it easier for crops to grow in harsh climate conditions. Growing robust cassava in extreme heat, for instance, or rice plants that remain healthy while completely submerged under water during monsoon season would both be well-received.
Deepti: While both these positive benefits also have the possibility to create the issues we discussed above, at least these are situations where we can consider whether the benefits outweigh the costs.
Amy: Yeah, GM technology may be used for good or for evil! Growing crops that produce toxins that kill pests, so that farmers don’t have to spray pesticides that are bad for both bugs and people, could be better than the alternative. On the other hand, genetic modification may be –- and already is –- harnessed to benefit biotech industries at the expense of consumers, like with herbicide-resistant crops; they boost output but are questionable in terms of human health.
Deepti: I think that’s really what it comes down to in the end: the fact that risks with GMOs are a reality is simply a matter of reading the literature and applying appropriate statistics. Whether these risks are worth it in terms of increasing yield and so on is a matter of policy and economics. And whether it’s ethical.
Are GMOs the best or only solution to the problems they claim to solve?
In a word, sometimes. For one, the prominent claim that GMOs such as golden rice can end world hunger is hollow. It would certainly be feasible to exploit GM to concentrate nutrients into staple crops that are easily grown in developing countries. However, the source of hunger is not insufficient food supply, but rather food waste and inefficient global distribution. As mentioned above, GM technology would not be the most sustainable – or the simplest – means to this particular end. As for growing crops that outgrow weeds and evade pests, GM may succeed in the long term, but scientists will periodically need to reinvent strategies to counter the perpetual evolution of crops’ weed and insect enemies. Meanwhile, the herbicide resistance tactic is not an optimal choice for consumers as long as produce arrives in supermarkets containing glyphosate. GM may, however, be the method best suited to make targeted changes to crops in order to grow them in new regions of the world, thereby improving food distribution, and to keep crops robust in the face of climate change.
A series of articles detailing each most controversial issues surrounding GMOs, including regulatory and ethical ones.
A discussion of studies that suggests GM feed is safe for farm animals. (There are other studes that suggest otherwise.)