Gene Transfer Is Neither Unnatural Nor Dangerous

The transformation of wild plants and animals into the foods we eat today is – by far – the single most dramatic experiment in genetic engineering the human species has undertaken

By Michael Eisen

Last week I wrote about the anti-science campaign being waged by opponents of the use of genetically modified organisms in agriculture. In that post, I promised to address a series of questions/fears about GMOs that seem to underly peoples’ objections to the technology. I’m not going to try to make this a comprehensive reference site about GMOs and the literature on their use and safety (I’m compiling some good general resources here.)

I want to say a few things about myself too. I am a molecular biologist with a background in infectious diseases, cancer genomics, developmental biology, classical genetics, evolution and ecology. I am not a plant biologist, but I understand the underlying technology and relevant areas of biology. I would put myself firmly in the “pro GMO” camp, but I have absolutely nothing material to gain from this position. My lab is supported by the Howard Hughes Medical Institute, the National Institutes of Health and the National Science Foundation. I am not currently, have never been in the past, and do not plan in the future, to receive any personal or laboratory support from any company that makes or otherwise has a vested interest in GMOs. My vested interest here is science, and what I write here, I write to defend it.

So, without further ado:

Question 1: Isn’t transferring genes from one species to another unnatural and intrinsically dangerous?

The most striking thing about the GMO debate is the extent to which it contrasts “unnatural” GMOs against “natural” traditional agriculture, and the way that anti-GMO campaigners equate “natural” with “safe and good”.  I’ll deal with these in turn.

The problem with the unnatural/natural contrast is not that it’s a mischaracterization of GMOs – they are unnatural in the strict sense of not occurring in Nature – rather that it is a frighteningly naïve view of traditional agriculture.

Far from being natural, the transformation of wild plants and animals into the foods we eat today is – by far – the single most dramatic experiment in genetic engineering the human species has undertaken. Few of the species we eat today look anything like their wild counterparts, the result of thousands of years of largely willful selective breeding to optimize these organisms for agriculture and human consumption. And, in the past few years, as we have begun to characterize the genetic makeup of crops and farm animals, we are getting a clear picture of the extent to which traditional agricultural practices have transformed their DNA.

Let’s take a few examples. This is a Mexican grass known as teosinte and its seed.

Thousands of years of selection transformed this relatively nondescript plant into one of the mainstays of modern agriculture – corn. The picture below – which shows the seeds of teosinte on the left, and an ear of modern corn on the right – gives a pretty good sense of the scope of change involved in the domestication and improvement for agriculture of teosinte.

Thanks to the pioneering work of geneticist John Doebley, and more recently an international consortium who have sequenced the genome of maize and characterized genetic variation in teosinte and maize, we now have a good picture of just what happened to the DNA of teosinte to accomplish the changes in the structure of the plant and its seed: a recent paper that characterized the DNA of 75 teosinte and maize lines identified hundreds of variants that appear to have been selected during the process of domestication. And maize is not weird in this regard – virtually all agriculturally important plants have a similar story of transformation from wild ancestors as generations of farmers adapted them to be easier to grow, safer to eat, more nutritious, resistant to pests and other stresses, and tastier.

For most of history this crop domestication and improvement has been a largely blind process, with breeders selecting crossing individuals with desired traits and selecting the offspring who have inherited them until they breed true – unaware of the molecular changes underlying these traits and other changes to the plants that may have accompanied them.

Modern genetics has fundamentally altered this reality. It has increased the power breeders have to select for desirable traits using traditional methods, and makes it far easier ensure that undesirable have not come along for the ride. And it also gives us the ability to engineer these changes directly by transferring just the DNA that confers a trait from one individual in a species to another. There are many ways to accomplish this – the most common involves extracting the DNA you want to transfer from the donor, placing it into a bacterium whose natural life-cycle involves inserting its DNA into that of its host, and then infecting the target individual with this bacterium. But recently developed technologies make it possible to effectively edit the genome in a computer and then make the desired changes in the living organism.

When applied to transfer genetic information from one individual in a species to another, this is an intrinsically conservative form of  crop improvement around since is all but eliminates the random genetic events that accompany even the most controlled breeding experiment.

The only difference between this and the generation of GMOs is that the transfered DNA comes not from a member of the same species, but from somewhere else on the tree of life. I understand why some people see this is a big difference, but modern molecular biology has shown us that all living things share a remarkably similar molecular toolkit, with the distinct properties of each species coming more from how these pieces are wired together than which ones are where.

Transferring a gene from a fish into a plant does not make the plant swim any more than stealing the radio from someone’s Maserati and putting it into my Honda Civic would turn it into a high-performance sports car. Indeed, scientists routinely use genes from mice, fungi, plants and even bacteria to substitute for their human counterparts, and vice-versa – which they often do perfectly.

And this doesn’t just happen in the lab. There are countless examples of genes moving naturally between species. Microorganisms swap DNA all the time – this is how antibiotic resistance spreads so quickly between species. Our own genome contains genes that got their start in bacteria and were subsequently taken up by one of our ancestors.

The relatively low rate of such “horizontal gene transfer” in multicellular organisms like plants and animals compared to bacteria is more a reflection of reproductive barriers and the defenses they have evolved to prevent viruses from hitchhiking in their DNA, than from a fundamental molecular incompatibility between species.

This is why I do not find the process of making GMOs unnatural or dangerous – certainly no more so than traditional breeding. And why I find the obsession with, and fearmongering about, GMOs to be so bizarre and irrational.

Of course the fact that making GMOs is not inherently dangerous does not mean that every GMO is automatically safe. I can think of dozens of ways that inserting a single gene into, say, soybeans could make them lethal to eat. But it would be because of what was inserted into them, not how it was done.

For what its worth, it would also be relatively easy to make crops plant dangerous to eat by strictly non-GM techniques. Essentially all plants make molecules that help them fight off insects and other pests. In the foods we eat regularly, these molecules are present at sufficiently low levels that they no longer constitute a threat to humans eating them. But it is likely that the production of these molecules could be ramped up when crossing crop varieties with wild stocks, or by introducing new mutations, and selecting for toxicity, much as one would do for any other trait. Indeed, there have been reports of potatoes that produce toxic levels of solanines and celery that produce unhealthy amounts of psoralens, both chemicals present at low levels in the crops.