Friday, August 23, 2013

Are GMOs safe? Define 'safe'

GMO: thumbs up or thumbs down?
People tend to feel strongly about genetically modified organisms, all good or all bad, but of course the story is too complex for a simple up or down. Organisms are genetically modifed for many reasons including to protect against disease, or to protect against pesticides or herbicides, but also to alter some characteristic such as ability to withstand freezing temperatures (tomatoes) or to speed their growth (salmon).  Each case needs to be evaluated on its own merits.

This undated 2010 handout photo provided by AquaBounty Technologies shows two 18 month old salmon, a genetically modified salmon, rear, and a non-genetically modified salmon, foreground.  (AP Photo/AquaBounty Technologies, Huffington Post)

GMO oranges
Amy Harmon had a piece in the New York Times a few weeks ago reporting on sour orange disease affecting oranges in Florida and around the world, and efforts to genetically modify trees to be resistant to the bacterium that's killing them.  No citrus that is naturally resistant has been found, so no 'natural' solution to the problem was possible.  Instead, after a number of transgenic trials, a gene from the spinach plant that attacks the bacterium has been introduced into orange trees for testing.  It's too early to know whether or not it will save oranges, but it seems more likely than any other transgene tried to date.

Harmon notes in her piece that, "Leading scientific organizations have concluded that shuttling DNA between species carries no intrinsic risk to human health or the environment, and that such alterations can be reliably tested."  The link there is to a 2012 statement by the American Association for the Advancement of Science on GMOs, which begins:
There are several current efforts to require labeling of foods containing products derived from genetically modified crop plants, commonly known as GM crops or GMOs. These efforts are not driven by evidence that GM foods are actually dangerous. Indeed, the science is quite clear: crop improvement by the modern molecular techniques of biotechnology is safe. Rather, these initiatives are driven by a variety of factors, ranging from the persistent perception that such foods are somehow “unnatural” and potentially dangerous to the desire to gain competitive advantage by legislating attachment of a label meant to alarm. Another misconception used as a rationale for labeling is that GM crops are untested.
So, broadly speaking, science says GMOs are safe, it's lay people who say it's not.  But not so fast.  Leading food reporter Michael Pollan took issue with Harmon's treatment of the GMO question.  He said on Twitter in a now famous 140 characters, that the piece included too many industry talking points, and he then went on vacation without following up.  He has now explained himself to a reporter for the Columbia Journalism Review, and Harmon responds. Their differences seem primarily to be differences in worldview, rather than arguments over facts -- Pollan critiques Harmon for not challenging the monoculture approach to orange growing, among other things, for example and Harmon responds that she was talking about a single problem, not the general issue.  And so on. 

Harmon's piece does pick up on the widespread misconceptions about GMOs.  She quotes someone wondering if their orange juice will now be green, with the spinach gene introduced into the trees.  It's certainly true that much of the mistrust of transgenics is based on faulty knowledge about genetics and the technology itself, but there is also some well-grounded mistrust, fueled by science.

In the US most of our corn, soy, canola oil and others is transgenic.  We've been eating it for years, to apparently no ill effect, or at least none that has been clearly documented. But it's not only human health that is of concern.  There are also environmental issues.  Glyphosate resistant weeds, for example.  Monsanto's Roundup Ready transgenic crops, bred to be resistant to glyphosate-containing herbicides, have led to "superweeds", weeds that are also resistant to glyphosate, and thus require more toxic herbicides to kill. We've written about this subject before (here, e.g.), but it's also all over the web.

Similarly, transgenic crops, bred to be resistant to pesticides, have naturally lead to resistant pests, and the increased use of more toxic pesticides, and other sequelae (we wrote about that here, but this issue, too, is all over the web). Transgenic orange trees bred to kill bacteria could potentially lead to a similar kind of problem, with intense selection for bacteria that are able to resist.  That's evolution.  Only time will tell if this becomes a problem. 

GMO rice 
So, even if Monsanto's Roundup Ready crops are benign for human consumption, there are environmental consequences.  And the mistrust of GM is likely to spread, given new evidence.  A paper published in New Phytologist (Wang et al.) and described in Nature last week, addresses the concern of gene flow from transgenic crops to their wild relatives.  The paper describes the results of introduction of the Monsanto RoundUp Ready transgene into weedy rice.  Wang et al. introduced the gene experimentally, but it also does happen in the field naturally.

Weedy rice can pick up transgenes from genetically modified crop rice through cross-pollination.
XIAO YANG; Nature Aug 16 2013
They. found that the transgenic weedy rice, a noxious weed that infests rice fields everywhere, "produced 48-125% more seeds per plant than nontransgenic controls in monoculture- and mixed-planting designs without glyphosate application."  That is, the weedy rice with the transgene seems to have substantial fitness advantages, even when not exposed to the herbicide.  (This still needs to be replicated -- see Andrew Kniss's blog post for discussion.  But if it's not true in this particular case, everything we know about evolution suggests it could be, and it's only a matter of time.)  And, hybridization, "pollen-mediated gene flow from rice to weedy rice has been reported."  If this is so, Wang et al. can't say whether it's also true of other plants or whether weedy rice is unique, but if increasing herbicide resistance in a crop increases the hardiness of an already noxious weed, this is potentially a huge problem.

And this is aside from the problem of the spontaneous evolution of resistance to glyphosate that has been documented in at least 24 weed species.  The superweeds.  Wang et al. write, "Our findings have broad implications for plant biology, crop breeding, weed management, biotech risk assessment, and the ongoing evolution of herbicideresistant weeds."

So, are genetically modified organisms safe?  It depends on what you mean by 'safe'. 

23 comments:

  1. There's a nice piece about diversifying weed management in Australia in last week's Science issue on pesticides. Monsanto may reap benefits from transgenic weed management in the short term, but they can never be one step ahead of evolution for very long.

    ReplyDelete
  2. Nice, nuanced piece. But while you ably handle the question of safe/not safe, you didn't address the "what for?" question. I would be much more inclined to support GMOs to save the American elm tree or to boost the biosynthesis of lysine in traditional landraces of maize than I would the merchandizing of seeds carrying "suicide genes" so that poor farmers are required to buy their seedstock anew each year. The former "whatfors" are noble, the latter is vicious.

    ReplyDelete
    Replies
    1. You're right, of course. Monsanto is in business, and so, ok, must attend to the bottom line. But, damn it, so are farmers, and so must they.

      Delete
    2. Yes, Monsanto is a business and should be expected to act like one. But the suicide gene thing crosses the line. It not only makes smallholder farmers in the developing world permanently dependent on commercial seed sources, it also completely removes the control of genetic resources, including artificial selection, from their hands. I am extremely reluctant to call anything in the world "evil" (since I'm not sure what that means), but this is evil.

      Delete
    3. Yes, that and many more examples that it's hard to imagine thinking up. It's a testament to the power of the current agribusiness, monoculture, chemical-dependent model that Monsanto can so treat the people they depend on to buy their products, and still be so successful.

      Delete
    4. It's a bit more complicated, I think. Often AgriBiz gets its natural stock, that it later modifies, from the indigenes to whom the suicided seed is then sold.

      On the other hand, if the recipient farmers could buy one bag of MonsantoSeed and then plant it indefinitely, how would Monsanto make its money? There must be a precedent or else there wouldn't be so much outrage, but I don't know what it is.

      Delete
    5. Jim, there is no "suicide gene" anywhere near production.

      Ken, people have been buying a bag of seed they can't replant for decades. It is called "hybrids" and no next-generation from the hybrid will perform well. That's how seed companies always could stay in business an fund the next innovation.

      Delete
    6. I stand corrected if no 'suicide genes' are yet implemented, because I thought that several had been.

      The point is no different from that of hybrid seed, though the suicide gene issue comes up repeatedly in attempts to paint agribusiness with a bad brush (which is why I thought some GMOs like BT or Roundup Ready were of that sort.

      There is a legitimate question of how a for-profit agribusiness can make their money. The idea that (especially in developing countries) poor farmers have to buy new seed each year is what has motivated much anger (and, especially, if as is alleged, agribiz gets some of its starter material from the same countries, as has been alleged in regard to India and perhaps Mexico).

      Our point in this regard is not to take sides, but to point out the issues that are so polarizing. I think that this polarizing stands in the way of attaining some reasonably fair and objective solutions.

      Of course, agribiz might not see that there is any need for a 'solution', when they argue that with whatever imperfections they're feeding the world with their current practices.

      Delete
    7. We've tried to be measured, and to point out that each instance has to be considered on its own merits. Even so, if we got some things wrong, it shows just how much misinformation is out there, and how easy it is to absorb.

      Delete
    8. Kevin, your points are well taken. But suicide genes (an admittedly melodramatic label) are in development, no?

      You are, of course, quite right about hydrid seeds. They interbreed (including through selfing) in the next generation and thereby lose 50 percent of the hybrids. But, as someone who works in the developing world, I have always believed that the dependence of the farmer on commercial seeds should have been discussed as a major counterweight to all the hype about hybrids before hybrids were widely marketed to poor smallholder farmers. Yes, hybrids have definite virtues, but they also have this singularly unfortunate vice.

      Delete
  3. But we have to feed the world... 's steaks.

    ReplyDelete
    Replies
    1. And it's not just the cattle that lose their hides...

      Delete
  4. What do you think of this article -

    http://www.examiner.com/article/fear-of-a-gm-planet-ii-bad-science-and-good-science

    ReplyDelete
    Replies
    1. Anti GM, pro GM; that illustrates our point perfectly - it makes no sense, to me, to be pro or anti a technology that can be used for such a wide diversity of purposes. Each instance needs to be evaluated on its own merits. It can make sense to be against the uses to which the products of the technology are put, but they aren't *all* evil, as Jim puts it in his comments above.

      And, as we tried to note in the post, even if a GMO turns out to be completely benign for human consumption, that does not mean it has no environmental consequences. Those are two separate issues. (Just as the argument that people shouldn't spend money on organic foods because they aren't any better for you avoids the issue of whether they are environmentally better.) Evolution can't be outsmarted.

      Delete
    2. The arguments are mostly advocacy pieces, so it is difficult to judge who's argument is accurate.

      Clearly the facts are not unambiguous. It would only take one huge mistake for GM to bear the burden of having forced itself on wary eaters. But the occasional minor or rare event, say allergic reaction, will trigger as much panicky reaction, in today's arena, no?

      As we will try to point out (in our next post, on Monday), people's motives are often mixed and complex.

      In that sense, I personally think that the issues are as much political and social as they are 'scientific'.

      Delete
  5. Thank you. You guys help me maintain my sanity, because it is difficult to find intelligent discussion on various topics in this polarized society. Everyone starts with putting a person into a 'Republican' or 'Democrat' bin and then assumes the next statement before responding.

    I also enjoyed the previous commentary on creationism. In that topic too, extreme polarization rather than social needs fuels the debate. I asked a college kid the other day, whether their college teaches about evolution. She told me that their's is a liberal college and so evolution, gay rights, etc. are all allowed. Then I asked her a couple of questions on evolution, but did not find that she had much clue. (She is a science student).

    So, although colleges talk a lot about teaching evolution, I do not think they are making much progress on transferring real knowledge. I wonder how the schools can be any better.

    ReplyDelete
    Replies
    1. Thanks for your comments.

      Truth to tell, lots of professional biologists and geneticists haven't a very sophisticated understanding of evolution and treat it as a rather closed, fully understood package, which it isn't.

      So (and given the teacher training situation in the country) how can you expected teachers to know the subject that well?

      This is to a considerable extent about cultural competition for resources and respect.

      Trying to see other points of view should be what citizenship is all about, and the same for science. But it's very difficult, especially when one is strongly convinced s/he understands something and the opponents don't.

      You might want to go to my web page (or the journal Evolutionary Anthropology web page) and fine my installment in my series 'Crotchets and Quiddities' entitled 'The Scopes trial' to see how even that icon of the battle between evolution and religious fundamentalism is widely misunderstood.

      Delete
  6. I think it is useful to draw a distinction between the safety of GM crops and whether the FDA approval framework is adequate for assessing safety.

    Under substantial equivalence it is not necessary to test crops on humans under the assumption that the modifications are well understood. This assumption is incorrect because it is impossible to know if there are unknown repercussions from a particular modification. Many crops, for instance, were put into the food chain with incorrect understandings of the molecular mechanisms (e.g. many antisense knockouts were used before biologists realized they were turning on the siRNA pathways).

    So assuming that human testing is not needed seems seems to be standing on sketchy logic, and it is in stark contrast to what is required for other products on the market, like food additives or drugs.

    And then the rat toxicology tests they do require use only a dozen or so rats. If you do a power analysis on that it should be clear that they can really only detect really extremely harmful effects that kill like 1/3 of the rats. Smaller effects cannot be detected.

    Then, thirdly, there is no labeling. Epidemiological survey studies are hard enough, but if biologists really did screw up a modification (i.e. turning on a pathway or mechanism they didn't know existed that produces an active molecule) the lack of labeling destroys the last safety net that could catch it.

    The structure of the US safety testing isn't an intrinsic characteristic of genetic modification. It is an economic structure. But the lack of substantial and systematic testing surely adds to the fear of GM food.

    So without systematic and adequate testing as a part of the approval process you are left with a hodgepodge of studies on certain specific modifications which themselves are often even smaller than the FDA tests, and very few individual modifications with large human tests, like the tests that actually were done on Golden Rice.

    The GM food debate seems at times like an argument between camps about what the data would say if it existed. I understand that scientist think they know more about modifications than non-scientists, but if the data does not exist they should know enough not to speculate.

    ReplyDelete
    Replies
    1. These are good points and we often post about scientists' various reasons, from bad science to vested interests, for being blind to or ignoring weaknesses in their data or methods.

      Monday's post will follow up on Friday's to make some points that are relevant here. That's because, beside the empirical issues such as you raise, there are other reasons why people (including scientists) take the positions on GM food that they do.

      There are somewhat similar issues in terms of drug safety, among other things. If there is a 'cure' for the problem, it would be to remove any vested interests in the generation and analysis of data--as if that were possible.

      But also, each of us in or out of science will have his/her own calculus for what benefits outweigh what risks.

      Delete
    2. Gawd, these are important points!

      Delete
  7. Hello,

    Please explain how "everything we know about evolution suggests" that the evolution of herbicide resistance leads to an increase of fitness in an environment where herbicide is not present. As an evolutionary student, I would say that to the opposite, everything we know about evolution suggests that herbicide resistance leads to a decrease of fitness in an environment where herbicide is not present because ressources spent for building the mechanisms preserving the plant from the herbicide are spent elsewhere in plants lacking the resistance (this is obviously not the case for RR crops since the resistance is confered by a mutated EPSPS gene, but then again nothing in evolution suggests that this mutation would confer other type of increase in fitness, otherwise you probably need to explain why it did not appear before we started using Round Up).

    Finally I always have problem with this RR crops = superweeds appearing shortcut. There is a blog post about this also at weedcontrolfreaks, bottom line is herbicide resistance appearance is an herbicide problem, not a RR crop problem.

    ReplyDelete
    Replies
    1. You are describing what one would expect if organisms always followed the (human-made) rules of evolution. But one of the points we, and of course many others, have tried to make about GMOs is that there can always be unanticipated consequences. If it's true that transgenic weedy rice is more fit than the non-transgenic, as Wang et al. suggest, even without applications of the herbicide, that's an unanticipated, unintended consequence. Life is complex. Every genome can be expected to incorporate a transgene in a different way.

      As for what superweeds are a consequence of, it's the RR/herbicide cycle, of course, not just RR -- without RR, we wouldn't have so much glyphosate being applied to crops, and weeds being exposed to glyphosate, and so forth.

      Delete