Exposed: the great GM-food scare made up by the Independent

The Independent recently published an article with the startling headline:

Exposed: the great GM crops myth
Major new study shows that modified soya produces 10 per cent less food than its conventional equivalent

Wow. That's a pretty bold statement. Let's look further into it, shall we?

Genetic modification actually cuts the productivity of crops, an authoritative new study shows, undermining repeated claims that a switch to the controversial technology is needed to solve the growing world food crisis.

The study – carried out over the past three years at the University of Kansas in the US grain belt – has found that GM soya produces about 10 per cent less food than its conventional equivalent, contradicting assertions by advocates of the technology that it increases yields.

Hmmm.

Professor Barney Gordon, of the university's department of agronomy, said he started the research – reported in the journal Better Crops – because many farmers who had changed over to the GM crop had "noticed that yields are not as high as expected even under optimal conditions".

I've noticed in the past that news organizations don't always do the best job with science stories, so let's look at the original citation, shall we?

But first, some background (yes, yes, I know that the Independent wouldn't require any background, but they also wrote a factually wrong, irresponsible, scaremongering article, so we'll do some background). Soybean plants have been genetically modified in recent years to be resistant to a broad-spectrum herbicide called Roundup (aka glyphosate). This provides a neat trick for farmers, because they can grow their Roundup resistant soybeans, and spray them with Roundup to kill any weeds (without killing the soybeans). This is a neat trick, because Roundup isn't very toxic to people, and breaks down pretty readily in the soil, providing a neat way for farmers to decrease weeds in their fields, without having to use more expensive and toxic compounds.

Ok, so what was the question that Dr. Gordon asked? From the paper:

There is evidence to suggest that glyphosate may interfere with Mn metabolism and also adversely affect populations of soil micro-organisms responsible for reduction of Mn to a plant-available form. Manganese availablity is also strongly influenced by soil pH. As soil pH increases, plant-available Mn decreases. It is unlikely that Mn deficiencies will occur on acid soils. It stands to reason that the addition of supplemental Mn at the proper time may correct deficiencies and result in greater GR soybean yields.

Ok, so he wanted to ask if there are manganese deficiencies in plants that are treated with glyphosate. And what did he see?


He sees that he can increase yield of soybeans by applying manganese to the soil. Cool, huh? Now, if you're a journalist at the Independent, you may look at that graph and say, "Hey, wait, the conventional soybeans have higher levels of production than the Roundup resistant soybeans!". Well, you could say that, but you'd be wrong. You can't ask that question from these data. Firstly, these strains aren't isogenic. Meaning that their genetic content isn't just differentiated by the fact that the GM strain is Roundup resistant, so increased growth could be due to different genetic content. Secondly, the Roundup resistant strain was sprayed with Roundup. And the other one wasn't. That's a rather large difference. Indeed, the conventional strain of soybean is really only a control to show that increasing manganese levels doesn't increase yield in the absence of Roundup. Thirdly, both these crops were worked over by hand to remove weeds, as they were interested in only studying the effect of manganese on the experiment, not on the effect of weeds in the crop. So this experiment simply can't be used to ask "Do Roundup resistant crops produce more or less than conventional crops?" The experiment wasn't designed to ask that question, and you can't use these data even incidentally to ask that question. And what does the Independent say?

The GM crop – engineered to resist Monsanto's own weedkiller, Roundup – recovered only when he added extra manganese, leading to suggestions that the modification hindered the crop's take-up of the essential element from the soil.

Hmmm.... but they applied Roundup to the Roundup resistant plants. That's another pretty huge difference.

Indeed, the author of the study, Dr. Barney Gordon, concurs. From an e-mail he wrote me:

The article you saw in the Independant was a total distortion of the research. My research concerned manganese application on soils that are known to respond to Mn application. We used one conventional variety and a glyphosate-tolerant near isoline (not genetically identical). The objective of the research was to improve soybean yields under optimum management conditions, not to make any statement about GM crops.

Now, I can almost hear the anti-GM crowd crowing now, "Just because he didn't ask that question, doesn't mean you can't extrapolate from the data he gathered", but as I've outlined above, you can't even extrapolate. The experiment isn't properly controlled to ask that particular question, just to ask the question about manganese suplementation.

And it's rather disingenuous of the Independent to say otherwise.

Horizontal transfer makes worms photosynthesize?

Horizontal transfer is cool. This is the process whereby cells pick up DNA from other cells that are totally unrelated to them. Bacteria do this all the time. They pick up DNA from viruses and other bacteria and integrate this DNA into their genomes. If it turns out to be useful, it remains in the genome. If not, it's slowly lost over time.

People have always assumed that horizontal transfer doesn't happen in larger organisms. When was the last time you met a person who had picked up a little cabbage DNA. This has historically been one of the complaints about anti-genetically modified food campaigners. Putting genes into a plant that come from a bacterium is somehow "unnatural". Well, turns out horizontal transfer is very natural (in the sense that it’s starting to look like it might be quite common in nature).

In the last few months, there has been more and more data suggesting that horizontal transfer *does* occur in larger organisms. A group at the J. Craig Venter Institute recently published a paper in Science showing large chunks of the Wolbachia genome has been transferred into some fruitfly genomes. Wolbachia is a bacterial obligate intracellular parasite in insects. It grows inside the germ cells of insects, and transfers itself to the following generation inside the egg or sperm of the infected parent (usually through the egg). Well turns out that in at least one detectable case, the Wolbachia genome has actually been incorporated into the fruitfly genome. Very cool! This shows at least 2 things:

1) Horizontal transmission in fruitflies (and possibly other metazoans) is probably not astronomically improbable. If it were, this discovery would represent an incredibly unlikely find. (Of course, until someone finds more examples of this, we could wave this away as just a bizarre anomaly, but I doubt it - more in a bit).

2) This demonstrates an early step in the endosymbiont hypothesis. The endosymbiont hypothesis suggests that organelles that contain their own DNA come from previously endosymbiotic bacteria (like mitochondria and choloroplasts). These organelles contain their own DNA, but not all the proteins in them are come from their own genomes. Some of the proteins come from the host genome. This new Wolbachia finding is a demonstration of an early step in moving gene products from an endosymbion into the host genome. Very cool!

Well, today I heard a talk by a gentleman named Debashish Bhattacharya who had recently sequenced parts of several protist genomes. These are organisms like Paramecium, that are single-celled organisms that feed on bacteria (they’re quite a bit larger than bacteria, so they swim around and gobble them up). They don’t photosynthesize. However! He found the remains of several hundred genes from photosynthetic bacteria in their genomes. This shows yet another example of horizontal transfer in eukaryotes.

The coolest part of his story comes last. Dr. Bhattacharya hypothesized (and is looking for supporting evidence) that he would find genes from photosynthetic algae within the genome of a particular worm Placida cf. dendritica. This worm does something pretty cool. It eats photosynthetic algae, but it preserves the plastids within its own body, and uses them to generate energy from light. Holy photosynthesizing worms, Batman! But how do they maintain these complex organelles in a viable state that can provide energy? He hypothesizes that he will find hundreds of genes from photosynthetic algae that are involved in keeping the plastids functioning. Stay tuned!

Mark this day in 2007, the day that gene networks were discovered

There are at least two major ways that science articles in newspapers can fail.

The first is the "No, duh." coverage. That is, the article is presenting things that scientists have known for years if not decades, but presenting it like it's a major new discovery that will require major rethinking of how that area is studied. The second is the "the goblins will get you!" coverage, that is that if it weren't for the genius working at PETA, we wouldn't know that we're all going to die, when in fact the data say nothing of the sort.

Denise Caruso managed to hit a double this weekend in the New York Times entitled (stupidly): A Challenge to Gene Theory, a Tougher Look at Biotech. Let's start with her grandiose opening statement:

The $73.5 billion global biotech business may soon have to grapple with a discovery that calls into question the scientific principles on which it was founded.

Gee whiz! That sounds exciting! If only it were true...

Last month, a consortium of scientists published findings that challenge the traditional view of how genes function... (snip) ...To their surprise, researchers found that the human genome might not be a “tidy collection of independent genes” after all... (snip) ...Instead, genes appear to operate in a complex network, and interact and overlap with one another and with other components in ways not yet fully understood.

*gasp!* Really? They operate in networks? Sort of like this paper by Jacob and Monod suggested in 1961? The statement that "genes appear to operate in a complex network" is like saying that "the liver operates in a complex environment". Of course the liver works with other organ systems. This is not news. We've known about networks for a really long time, and with each year we're understanding more and more of these networks, and finding more and more spokes to them. In the last ten years, the study of networks has even become a discipline of its own, called systems biology. This is not new.

Biologists have recorded these network effects for many years in other organisms. But in the world of science, discoveries often do not become part of mainstream thought until they are linked to humans.

Or at least, in the world of newspapers. This has been mainstream for years.

Unfortunately, Francis Collins (director of the National Human Genome Research Institute) gave a misleading quote on their website, which is quoted in the New York Times article:

Because of the hard work and keen insights of the ENCODE consortium, the scientific community will need to rethink some long-held views about what genes are and what they do, as well as how the genome's functional elements have evolved.

While Ms. Caruso seems to think this means that he's discovered that gene networks exist, it seems more likely that he means that they have discovered particular networks that will make it easier to understand particular diseases. I admit, he's not being clear, and she bases the first half of her article off of this quote. (And I find it hard to believe that Francis Collins is stupid enough to think this study marks the discovery of gene networks).

So where does she take this credulous party off to next? Why not fearmongering?! (I've covered her scaremongering here in the past). Enter fearmonger Jack Heinemann, (director of the Center for Integrated Research in Biosafety):

“Because gene patents and the genetic engineering process itself are both defined in terms of genes acting independently,” he said, “regulators may be unaware of the potential impacts arising from these network effects.”

They're also unaware of the possible interactions between cheese and tomato sauce, and yet we don't test these in combination (and cheese is a living food product, after all!). This is yet another case of someone demanding that testing be done of every possible interaction between different genes and the food that they are placed in. Why? Knowing that your food has a complex network of genes that may be changed by the introduction of a foreign gene doesn't change the fact that the constituents of that food don't change.

Here's an analogy. If you remove someone's kidneys, you will find that they now require dialysis because they can no longer process out the urea from their blood. Without dialysis, they will eventually poison themselves and have multiple organ failure. However, you can say for sure that they will not grow wings and breathe fire, just because you messed up their organ homeostasis.

In the same way, adding foreign genes to plants, you can only get what you put in. You may increase the amount of protein, or decrease the amount of carbohydrate, or make the plant grow slower, but you won't get a corn plant with wings that is poisonous to humans merely by inserting the BT-gene. You can only get what you put in.

From Denise Caruso:

Now that the consortium’s findings (that there are gene networks) have cast the validity of (the gene) theory into question, it may be time for the biotech industry to re-examine the more subtle effects of its products, and to share what it knows about them with regulators and other scientists.

Oy. This just makes me cringe. Please, Ms. Caruso. Please. The stupid. It burns...

Genetic engineering meets vaccinology: or how to get the attention of all kinds of protesters

ERV wrote yesterday about a cool, new genetically-modified food that has the potential to act as a vaccine against cholera. The basic idea is that they are expressing a gene from cholera in rice. When you eat the rice, you are exposed to the fragment of a protein from cholera and will make antibodies to it. That way when you are exposed to cholera in the future, you will be immune to it. What are the advantages of this approach? From ERV:

So really this rice is going to be ground up and put into pill form, which is fantastic on so many levels!

1. Dont need a needle. Ive said over and over and OVER (are you listening, HIV-Circumcision 'researchers'??) its really damn hard to make sure your needles/equipment is appropriately sterilized in the third world. I dont care how many seminars youve had on how to sterilize needles. It doesnt always get done. Plus, I dont think youll hear any kids complaining about a lack of needles ;)

2. Can be kept at room temperature. No refrigeration. Aint gonna always have that in the third world either. Sure you might have a fridge, but it might not have power. And they tested batches that were left at room temp for 1.5 years-- Still worked. AWESOME!

3. Introducing the antigen through the intestinal mucosa. These folks found that introducing the cholera antigen orally works great because cholera normally infects through your digestive tract! Convinces lots of sentinel immune cells to hang out in your intestine and wait for another cholera bug to float by!

I thought I would write about another super cool approach to vaccines of the 21st century, and this one is a little closer to home as my wife and I both work on things that are indirectly related to this. (Though I confess, I'm a little worried that this post will attract the anti-vaccination mob along with the anti-genetic-engineering crowd).

Live-vector vaccines are what a lot of folks are hoping will be the vaccines of the future. How do they work? Over the last 30 years, we've learned an awful lot about how several different bacterial pathogens work. Pathogens like Salmonella and Listeria actually enter your cells and express their genes inside you (these are not viruses, mind you, but bacteria that go intracellular). We also know an awful lot about how to stop pathogenesis at certain steps. There are all kinds of mutants that are able to go intracellular, but are blocked at later stages of pathogenesis, so are unable to spread, or unable to express certain toxins, or otherwise unable to complete the disease process.

How does this help us build vaccines? These mutants that are blocked at later stages of pathogenesis actually elicit an immune response from the host. That is, when your body sees foreign bacteria in their cells, you are able to mount a potent immune response to it, so the second time you see Salmonella or Listeria, you are able to fight it off (though given that Salmonella has so many different varieties, it is quite difficult to vaccinate against all of them). It also allows you to mount both an antibody response and a cell-mediated response (these are two different immune responses that I will write about later). The long and the short of it is that most vaccines only elicit an antibody response, which doesn't protect well against pathogens that climb inside your cells to evade the antibodies.

Well, these days it is rather trivial to genetically engineer Salmonella. And Listeria is also able to be genetically engineered. So it's easy to put genes from another pathogen into them. (And before you cry, "What?! You're building a superbug?!! - it takes hundreds of genes working in concert to create a pathogen. One or two genes does not a pathogen make). One example of a live vector vaccine is here. Folks at the Center for Vaccine Development put part of the tetanus toxin into Salmonella. They then gave the Salmonella to mice, and found that the mice made antibodies to tetanus that have been previously shown to be able to protect them from infection.

So, big fat hairy deal. They've built a vaccine for an infection that there is already a potent vaccine available. Well, this is a potent demonstration that this technology is viable. There are a bunch of labs working to build vaccines against disease in a similar manner (including for diseases that don't currently have available vaccines).

What are the advantages of this approach? It's similar to the rice vaccine above:

1. You can get both cell-mediated and antibody immune responses, whereas most subunit vaccines that are available only give an antibody response. In non-science language: your immune system responds in a way that's more likely to prevent disease.

2. No needles! (You can just drink this).

3. Easier shipment and administering of the vaccine. And shipping to remote locations (think parts of Africa & Asia) is much simpler. That, and you don't need to have anyone who is highly trained giving the vaccine. Merely an eyedropper and a bottle to drop it into a glass of water will do.

4. Cheap. Growing Salmonella is easy. And building genetically modified Salmonella is infinitely easier than genetically modified plants or animals.

5. Generalizable. It is easy to imagine building multiple different vaccines in this manner, and then administering them simultaneously (either as a single organism, or a cocktail of several different bugs).

But what is the downside? The upfront costs will be high. New vaccines are expensive as the manpower to make them is highly skilled, and the number of tests that they have to undergo to demonstrate safety is high.

The other major downside is the PR side. The public at large is so woefully uneducated about basic science issues, and even conventional vaccines have been controversial in the public arena. I can only imagine the anti-vaccination folks now:

Genetically modified pathogens! *gasp* You're playing God!

Yep, every time you save a kid from a disease you're playing God. Some people think that's okay. I think that the time is right to start telling people about these vaccines so that when they do start to become commercially available that there isn't a visceral, knee-jerk reaction against them. When I tell people about what I do, they say, "You genetically engineer bacteria? Why would you want to do that?". If they're calm enough to sit down for a few minutes and talk to me about it, they often ultimately agree that it is a good idea.

GM crops and bees, an addendum

At the moment, Colony Collapse Disorder (and this Spiegel article particularly) are causing a minor storm on the blogosphere. Many of the blogs are being fairly uncritical in their condemnation of GM-corn, with knee-jerk reactions predominating over reasonable analysis.

I found some interesting stuff from the Colony Collapse Disorder Working Group. This is a group of entomologists based out of Penn State that are trying to figure out why the bees in North America and Europe are dying. They're not quite as hysterical about the die-off as the conspiracy theorists at Spiegel:

During 2006, an alarming number of honey bee colonies began to die across the continental United States. Subsequent investigations suggest these outbreaks of unexplained colony collapse were experienced by beekeepers for at least the last two years. Reports of similar die offs are documented in beekeeping literature, with outbreaks possibly occurring as long ago as 1896. The current phenomenon, without a recognizable underlying cause, has been tentatively termed “Colony Collapse Disorder” (CCD), and threatens the pollination industry and production of commercial honey in the United States.

They have a number of likely suspects:

What potential causes of CCD is the Working Group investigating? The current research priorities under investigation by various members of the CCD working group, as well as other cooperators include, but is not limited to:
• Chemical residue/contamination in the wax, food stores and bees
• Known and unknown pathogens in the bees and brood
• Parasite load in the bees and brood
• Nutritional fitness of the adult bees
• Level of stress in adult bees as indicated by stress induced proteins
• Lack of genetic diversity and lineage of bees

but consider GM-corn (BT-corn) to be an unlikely suspect:

What are examples of topics that the CCD working group is not currently investigating?

GMO crops: Some GMO crops, specifically Bt Corn have been suggested as a potential cause of CCD. While this possibility has not been ruled out, CCD symptoms do not fit what would be expected in Bt affected organisms. For this reason GMO crops are not a “top” priority at the moment.

Radiation transmitted by cell towers: The distribution of both affected and non-affected CCD apiaries does not make this a likely cause. Also cell phone service is not available in some areas where affected commercial apiaries are located in the west. For this reason, it is currently not a top priority.

Now conspiracy theorists may point out that the working group hasn't completely ruled out BT-corn, but it is rather compelling that they have examined the deaths of these bees, and decided their deaths don't fit the symptoms expected by BT-poisoning. (Rather akin to finding a bullet hole in a corpse, and concluding that the person likely wasn't strangled).

Genetically modified crops are responsible for bee deaths? I don't think so...

An article from last week in Spiegel is just so breathtakingly stupid it begs to be commented on. The article: Are GM Crops Killing Bees? suggests that there is a link between the declining bee population and the introduction of GM crops, shows data that makes the answer clear (the answer is no) and continues merrily along, ending the article as if there is some doubt about what the answer is.

A bit of background is in order. Over the last decade, bee populations in the U.S. and Europe have been dropping. The last year has seen particularly large drops in commercial bee-keepers. The Spiegel article cites rates as high as 70%, though these numbers are cited for individual beekeepers. It's unclear from the article what the overall drop rate has been, and though they cite differences between the U.S. and Europe, it's hard to tell if these are meaningful because they're only citing the losses of individual beekeepers.

The bee die off is particularly important for our agriculture, as many of the flowering plants require bees to pollinate their flowers to make fruit. Loss of bees could result in major decreases in the production of flowering crops.

So what does this article say? These drops coincide with the introduction of genetically modified-insecticide producing corn, so this corn must be to blame. This is not a bad hypothesis. One could imagine that bees that find their way onto the corn might ingest some of the compound. However, let's look at the data. Germany has some pretty strict laws about introduction of genetically modified (GM) crops. GM crops made up less than one tenth of 1% of all crops grown in Germany, whereas almost all corn grown in the U.S. is genetically modified to produce the BT-insecticidal protein. So, if GM crops are to blame, one would expect bees in Germany would largely be spared the die off seen in bee colonies in the U.S. Sadly, this is not the case. German bees are dieing off at appreciable rates as well (though granted, at somewhat lower the rates than seen in the U.S. if you trust that the Spiegel numbers are representative). Game over. The fact that both countries are losing bees at high rates clearly demonstrates that GM crops are not to blame.

The National Academies of Science of the U.S. have looked into the bee die off (and have proposed spending more effort looking into it - it has serious economic consequences). Their main conclusions? Non-native mites have been responsible for a large number of deaths, as well as antibiotic-resistant bacterial pathogens (who knew? people treat their bees with antibiotics?). Their recommendations are for increased investment in bee-breeding, to allow creation of pathogen and mite-resistant bees.