Field of Science

Why did The Atlantic publish this piece trying to link miRNAs and GMOs?

[Note: Updates on this piece are at the bottom. Writer Ari Levaux explains (1) that he was not responsible for headline or dek, which I'd suspected, and (2) that he's going to rewrite the piece, updating the scientific portions of it.]

A study from a Chinese group led by Chen-Yu Zhang of Nanking University and published in Cell Research, has uncovered the fascinating result that when people eat rice, they can absorb microRNAs (miRNAs)--tiny sequences of RNA--from the rice into the blood. These rice-originating miRNAs turn up in blood and tissues of people who eat rice's the type of rice miRNA interacts with human proteins that are responsible for removing LDL ("bad" cholesterol) from the blood (!). It's the first report of plant miRNAs ending up in people by way of diet and the finding that at least one of them alters an important process in the body.

The implications could extend in many a direction, but not as far as writer Ari Levaux would like to take them in this remarkably confusing article published on the Atlantic Website. Before taking on the errors and the overstretch that are that piece, let's look at something far more interesting: miRNAs themselves.

These little bits of RNA, consisting of 22 building blocks linked in a single strand (a human DNA molecule has billions) get around with surprising facility, and their purpose is to regulate genes. They don't regulate by latching directly onto a DNA sequence but instead lurk in the cell and interfere with processes that come after the gene's role is complete. If you consider the gene sequence as the directions for building a protein, one job of RNA is to serve as a copy of those directions. It takes on the risky business of toting that copy out of the safety of the nuclear vault in our cells and into the big, bad scary cytoplasm outside. In the cytoplasm, the fluid-ish environment of the cell, RNA has many, many roles, but all of them center on executing the directions encoded in the gene for building proteins, the molecules that help make up our tissues and perform the tasks required to keep us alive.

In some cases, though, RNA occurs in the form of miRNAs, and their job may well be to bollix up the protein-building works. These little molecules--which researchers have identified in the hundreds in humans--can, for example, latch onto an RNA that is a copy of the protein code and cause it to break down or keep the cell from using it. These tiny RNA sequences help fine-tune the process of protein building well beyond the starting point of directions copied from a gene sequence. Thanks to miRNAs and many other steps that can promote or interfere with protein building, the cell--and the organism--has several chances to modulate how much of a specific protein it makes, allowing agile, real-time responses to changing conditions. 

Researchers have discovered myriad ways that miRNA influences human development and disease, and these discoveries open the way to using that information to cure disease. But all of the miRNAs investigated thus far in people have come from people themselves, either present for normal functions or overabundant and linked to disease. The flashy take-home from this latest rice study is, We can pick up these tiny regulators from what we eat...and they can interfere with the functions of proteins we make

This take-home could have huge implications for how diet influences our health and development if other non-human miRNAs turn up that fit the same profile: absorbable after we eat them and modifying how our bodies function. The effects could be good, bad, ugly, or neutral. This paper is simply an open door. Now, for years and years, investigators will walk through it to find a number of research paths to explore, from seeking more non-human miRNAs and identifying their effects to evaluating how modifying diet might influence disease or human development via miRNAs. 

In spite of how much lies ahead and how relatively little lies in the present about this discovery--one rice miRNA, one human effect--the piece that appeared today in the Atlantic argues that the implications are immediate and dire and related to genetically modified organisms. I initially read the piece trying to identify how someone could make that leap but instead found myself distracted by how poorly the article presents the science itself.

First, the headline: The Very Real Danger of Genetically Modified Foods. I read the Cell Research paper. I can't find mention of GMOs in it. I don't find mention in the paper the the rice miRNA in question derives from a genetically modified rice strain. So, I don't see that this headline appropriately represents the science here.

Then there's the dek: "New research shows that when we eat we're consuming more than just vitamins and proteins. Our bodies are absorbing information, or DNA." That's not what this research shows. It shows that the body takes up a specific rice miRNA when people consume it. Not DNA or "information."

The lede leaves out a crucial modifier: the word "rice": "Chinese researchers have found small pieces of ribonucleic acid (RNA) in the blood and organs of humans who eat rice." Actually, miRNAs are present in the blood and organs of...all humans, whether they eat rice or not. I think the writer here means "small pieces of rice ribonucleic acid." 

There is then a series of claims about what the research implies, including, mysteriously, that it will help us learn how some "herbal medicines function." The original paper makes no mention of herbal medicines, although some research indicates that "natural agents" can alter expression of human miRNA. Also among the potential implications described in the piece is, "And it reveals a pathway by which genetically modified (GM) foods might influence human health." That's an enormous leap to make from "one rice miRNA in blood and tissues influences activity of one human protein." A number of steps would be required for a GM food to exert a similar effect, none of which have been investigated yet. These steps include identifying that the modified sequence in the target food either also encodes a miRNA sequence or interacts with its expression or, later in the gene-to-protein process, somehow evades normal miRNA regulation thanks to this change. 

Then suddenly, there's Monsanto and a strange effort to explain the central dogma of molecular biology (DNA-->RNA-->protein) using a pizza/pizza restaurant analogy that involves the "DNA" knowing what kind of pizza "it wants," although in truth, the cell is the entity in charge of which parts of the DNA it uses. The central dogma, a linear representation of how a cell copies DNA into RNA and then uses the RNA copy instructions to build proteins, is too simple for what we know today about how cells regulate protein expression. But the core dogma remains intact, including that DNA serves as the template for making RNA.

The article makes a number of other scientific errors, including in a bold pull quote claiming, "The Chinese RNA study threatens to blast a major hole in Monsanto's claim. It means that DNA can code for microRNA (italics mine), which can, in fact, be hazardous." No. That's not what the Chinese study "means." It's not news that DNA encodes RNA of all kinds. It encodes the messenger form that carries the copy of the code. It encodes the ribosomal form that is a component of ribosomes, the cell factory workers that take the code copy and use it as an instruction book for building proteins. It encodes the RNAs that bring those factory workers the molecular blocks the cell uses for building proteins. And it encodes miRNAs. This latest paper does not carry the meaning that DNA encodes miRNAs--that's a longstanding part of the Central Dogma, ironically, and not news. Nor does it threaten in any discernible way to "blast a hole" in much of anything. As I noted, the study opens a door. 

In closing, Levaux writes,
The news that we're ingesting information as well as physical material should force the biotech industry to confront the possibility that new DNA can have dangerous implications far beyond the products it codes for. Can we count on the biotech industry to accept the notion that more testing is necessary? Not if such action is perceived as a threat to the bottom line.
"Ingesting information"? The miRNAs are not "information" (they are noncoding molecules), and like all other things of this world that we've identified, they're not somehow distinctive from "physical material." There is naught in this study that implies that "new DNA" can have "dangerous implications" far beyond the products it "codes for." The miRNAs in this paper are not "new." They are from rice, the most-prevalent grain crop in Asia, and presumably something humans have been taking in for hundreds of years. It's unclear from this study even what the implications of the findings are for consumers of regular rice, much less what they'd be for modified organisms. Furthermore, we are not the only entities that modify organisms. Nature does so, often by way of viruses. I wonder why the fact that miRNAs are also present in viruses and could "potentially regulate host genes" didn't set off the anti-GMO alarms, too. 

The article goes on for several grafs about Monsanto and substantial equivalence--indeed, the writer devotes a mere 180 words or so of 908 to the study itself--and observes that the lead author on the Cell Research paper (wisely) declined to comment on any implications about these findings for GM foods. If only the Atlantic and Ari Levaux had done the same, the real implications of this remarkable work could simply stand on their own. 
For an article that focuses more on the research findings from the study, including design and other dietary miRNAs identified, see this piece by Anne-Marie C. Hodge at Scientific American.

ETA: As for the study itself, the effects the authors found weren't earthshattering, and it seems that there was an issue with images provided that required a rapid erratum after the paper was published. 

Follow-up: The author of the piece, Ari Levaux, has responded here, and I have replied just below that. 

Follow-up follow-up: Ari Levaux has tweeted that he is going to rewrite the piece, taking the scientific critiques into account. I'm looking forward to seeing the update. 


  1. Thank you for this post and setting the fact straight. The Atlantic piece is one of the worst articles I have seen in a while...

  2. Just the other day I was looking at my breakfast of sausage gravy and biscuits with suspicion, thinking, "This is going to be converted into mass, somehow" , a profound understatement.

    After reading this, I've had fruit and cottage cheese for breakfast.

    Edited to add, you really ARE what you eat!At least for a while...

  3. It's funny, but when I heard about this Cell Research paper originally I rather wondered whether the anti-GMO movement wouldn't try to capitalize on it. They're so predictable in that way...but it's highly unfortunate the Atlantic has decided to provide them with a platform. Moreover, I think the press also should have been more careful about how they covered the Cell Research paper itself. If I were a reporter writing about this subject, I would want to wait until these results were replicated -- for obvious reasons.

  4. Excellent discussion. Thank you! Something puzzles me. Rice miRNA shows up in the bloodstream and can affect a cellular process. Presumably, over time, we'll discover that hundreds or even thousands of miRNAs present in the diet end up in the bloodstream and interact with a variety of cellular processes. It's not, presumably, a GMO-specific process, yet it's singled out as a risk of GMO foods. I don't get it.

  5. I wondered the same thing. Also...I'm not sure how the miRNA would have survived digestion. When we worked with RNA in the lab, we took extreme measures to avoid degradation because of RNases, etc. From what I'd understood, it's a very degradable molecule because it's not intended to persist indefinitely but serves transient purposes. So...I've been wondering but didn't include in this piece how RNA would survive the mouth and stomach in the first place. Think I'll tweet that question.

  6. Well, that's precisely the problem. RNA is much more labile than DNA owing to that 2' hydroxyl group. It's not clear how you could get these small RNAs to survive digestion, get them into the bloodstream and get them into cells. That's why I would like to see these results replicated before I buy into them. For now, consider me skeptical.

  7. Right. So I looked around, and I can't find that there's oral administration of miRNA. I find siRINA, but that's double stranded and stable. So...the REAL implication of this work, if it bears out, seems to me to be that miRNAs--which are of great interest therapeutically but have proved difficult to deliver--might survive the gut and be absorbed. That would be a VERY BIG DEAL for therapeutics. But yes...skepticism is obviously in order here.

  8. Why would the route be necessarily oral? We inhale food particulates during preparation and ingestion. i.e. Things like BSE can spread from contaminated feed because the animals inhale it when they stick their nose in the trough, rather than swallowing it.

  9. Whether or not Nature alters cell DNA by means of viruses, Nature also destroys stuff all the time by means of earthquakes, hurricanes, and volcanoes. Do we have to emulate everything Nature does, especially for silly reasons? (Exhibit A: claiming that GMOs are about increasing food supply when there is in fact food rotting in storehouses all over the world. Sounds like world hunger is a *distribution* problem to me! Cure *that* with Frankengenes.)

    I hadn't read the Atlantic article and only saw this one referenced via a link on Facebook and haven't got a dog in this particular fight either way. But I can't help wondering what percentage of the pro-GMO crowd stands to benefit financially from GMO technology in some way. Because even if you could argue GMOs are completely safe both in the human body and in the environment, the *economics* of growing the stuff are set up to benefit Monsanto and their ilk, and put small farmers out of business.

    Growing food is not about running a food factory that has to be made ever more efficient. People have to eat or they will not live. It follows that there should be as much redundancy built into the food-producing system as possible--as many individuals as possible must have at least some control over their food supply. This is the same principle ruling over the fact that we have two kidneys instead of one. If one unit fails the other can take over. Having all our food production centered on and run by Big Agribusiness means more people will starve.

    Tell me you already worked all that out and have figured out a solution. Otherwise I am not the one being irrational here.

  10. I made no arguments about GMOs in this piece one way or another. I made an argument that the science as presented in the original article was awful and that the research in question forms no link to GM foods. The effort to make the link was irrational. The question about the viruses was a serious one.

    You start your comment with a straw man argument. I don't recall anyone's having argued that we need to "emulate everything nature does." So, yes, your premise is irrational. Of interest, however, you seem to think it's fine to use nature as a model (two kidneys) for what you think needs to be done if it suits your purposes. We don't have two kidneys because nature somehow anticipated the need for a backup in case of kidney failure, which generally happens after reproduction would have occurred. Likely, we have two kidneys because the origins of the kidney develop bilaterally during embryogenesis.

    I do not benefit financially in any way from GMO technology.


Comments for this blog are closed.

Note: Only a member of this blog may post a comment.