It’s like Seralini with caterpillars. While the estimable Professor Giles-Eric had his infamous rats with tumours, this time we get deformed butterflies. The only surprise is that the media has not so far picked up the story, despite the catchy photographs helpfully included by the authors (see below). This is probably a good thing, because a read of the paper in question suggests two things: one, it is irrelevant, and two, it is likely wrong anyway.
Some background: Scientists at the government-funded Rothamsted Research institute in the UK have been developing omega-3 fatty acids in the oilseed crop camelina, using genetic engineering to transfer the relevant genes into the target plant. The object is to develop a sustainable source of feed for fish farming: currently farmed salmon are dependent for these essential oils on fish harvested from the marine environment. To reduce the burden of overfishing we need a land-based supply of feed, hence the project.
But how safe would this new GMO camelina be? Would the omega-3 fatty acids EPA and DHA which are produced by the camelina affect insects consuming the crop? A new study published in PLOS One aimed to examine this question, using lab-reared cabbage white butterflies, a pest of cultivated brassicas such as camelina. The study, authored by Hixson et al, fed their lab-reared caterpillars artificial feed, some with the EPA and DHA fatty acids, and a control group without.
The results? Supposedly the groups fed the novel fatty acids got higher levels of wing deformities. And hence the predictable howls of outrage from the anti-GMO lobby, which is against all genetic engineering in plants and animals always, whatever the purpose or result, and saw an easy propaganda win: ‘Nutritionally-enhanced GM crops? Too bad about the deformed butterflies’. Did you hear that? DEFORMED BUTTERFLIES! FRANKEN-CABBAGE WHITES!! BAN GMOS NOW!!! Etc.
Something fishy here methinks. So let’s look at the study methodology. What is rather reminiscient of Seralini is the low number of controls (25, with odd numbers of 4 different treatment groups totalling around 90 insects) and the high number of wing deformities in the controls: fully a third of the controls also had deformed wings, an inconvenient fact that the anti-GMO people predictably failed to note.
So why the deformed wings? The study authors admit that they got the humidity wrong when raising the insects and that high humidity atmosphere is associated with wing deformities – but they go on to insist that there is a statistically significant correlation between dose and deformed wings anyway. Looks like another typically inappropriate use of p-values to me, but anyway you decide (see graph).
In any case, the entire exercise is unfortunately rather irrelevant, as Rothamsted has gently pointed out in a response statement. Most importantly, Hixson et al fed their caterpillars on artificial feed that is not their natural food. Why not just feed them GMO brassica leaves containing EPA and DHA? Because they claim they couldn’t get them, but when I queried Rothamsted about this, the scientists there told me Hixson et al never asked for leaf material. Instead the paper gives a long justification for why their artificial food was a good approximation:
In order to put this into context, we estimated the amount of EPA and DHA found in a genetically engineered camelina seed with what might be found in leaf tissue based on the following approximate calculation. We point out that concentrations of EPA and DHA in vegetative tissues of genetically engineered oilseed crops have not yet been reported… Therefore, the highest amounts of EPA and DHA included in the artificial diets in this study are close to, but slight underestimates of, those in leaves that would be consumed by leaf specialists (such as P. rapae) in genetically engineered oilseed crop fields.
Whoops. Wrong! Rothamsted scientists have confirmed that there is no EPA and DHA expressed in the camelina leaves at all, because its expression is controlled by a seed-specific promoter. Cabbage whites don’t eat brassica seeds, so the whole experiment is rather pointless.
As Rothamsted says in its statement:
Rothamsted Research scientists have analysed leaf tissue of GM Camelina and non-GM Camelina plants to examine whether there is any production of EPA and DHA in the leaves of the plants, even if they have not been designed to do so. Rothamsted Research has unpublished data that demonstrate that the leaves of GM Camelina plants have exactly the same fatty acid profile as non-GM camelina plants and are devoid of EPA and DHA. These data will be submitted for peer review as part of a broader study in the near future.
The authors Hixson et al reference their feeding methodologies to work published by Cornell University entomologist Professor Tony Shelton. I contacted Professor Shelton (note: I have worked with Shelton at Cornell in other projects, such as Bt brinjal in Bangladesh) and received the following response:
The authors stress caution in interpreting their results, and this is fully justified. In their study they used an artificial diet and such simple diets do not mimic the complex physiological interactions that occur when insects feed on their natural host plants. Furthermore, the authors acknowledge the insects they used were “highly adapted” to the artificial diet and may not mimic those found in nature. Therefore, the authors are correct in stating that field trials, with proper oversight, should be conducted to advance our knowledge about the risks and benefits of plant production of EPA and DHA.
So we had best ignore the howls of protest from the antis and move towards proper field trials in order to get some more meaningful results. Jolly good. Thankfully only last week the UK government approved Rothamsted’s request for consent to proceed to further field trials, so this can now happen. These won’t at this stage be looking at ecosystem effects as the scientists are still refining the balance of the oils produced in the camelina crop. However, field-level effects on invertebrates would certainly be a subject of future study before any commercial release.
Anti-GMO campaigners have tried to stop the trial and generate public alarm on the basis that EPA and DHA are “novel” in terrestrial ecosystems and therefore potentially to be feared. Swiss anti-GM scientist Angelica Hilbeck is quoted saying:
The fact that these compounds [long-chain omega-3 fatty acids] are novel in terrestrial systems has been entirely overlooked until this study. I congratulate the authors for having raised the issue of this important ecological risk before these crops are planted on a significant scale.
Sounds scary. And Hixson et al also justify their study with the same novelty argument:
Because EPA and DHA are largely novel FA [fatty acids] at the level of terrestrial primary producers and terrestrial insects, and have not yet entered the agroecosystem, the effects of their consumption on terrestrial insect growth and development are unknown, and have not been the subject of any scientific study to date.
But actually it turns out EPA and DHA aren’t so novel at all. As Professor Johnathan Napier, lead Rothamsted scientist on the project told me:
The study by Stefanie Hixson and colleagues is potentially interesting, but as the authors themselves acknowledge, their experimental system is testing something quite distinct from our GM camelina plants which accumulate omega-3 fish oils only in their seeds. And their overall rationale, the suggestion that such fatty acids as DHA, and especially EPA, are not produced in terrestrial ecosystems is not always the case. In fact, one of the genes we used to produce EPA in our camelina plants is from the moss Physcomitrella patens (commonly known as spreading earth-moss), which is found in many temperate parts of the world. [See reference here.]
Yes, yes. I know cabbage whites don’t eat moss. But the argument that these fatty acids are so novel that they’ve never been out of the sea before, and could therefore run riot across terrestrial ecosystems, is clearly bogus. Both EPA and DHA are also found in algae that colonise rivers and ponds on land, Professor Napier told me.
To be clear in closing, I am not alleging that Hixson et al were motivated by rabid ideological bias in the same way that Seralini so obviously was. Their motivation for the study may well have been just the spirit of pure scientific inquiry. I have no problem with that. Nor do I think it’s so bad as to merit being retracted, as Seralini’s infamous work eventually was. As Professor Napier said, it’s “potentially interesting”. It’s a shame though that Hixson et al didn’t design a study that might have yielded seriously relevant or interesting results, rather than one which begs more questions than it answers.
A final thought: cabbage whites, as I mentioned earlier, are a pest. I hold a personal grudge against them for trashing my allotment broccoli crop every summer. So my humble suggestion is that Rothamsted scientists consider putting a Bt gene in there to just kill the damn things. That would protect the camelina from Lepidopteran insect damage, and no doubt reduce the need for insecticide sprays and thereby protect the wider ecosystem. Just a thought…