GM mosquito study under fire by industry experts
Last week, New Atlas reported that a pest control experiment in Brazil involving genetically-modified mosquitoes had gone wrong, with new genes mixing into the native population of the insects. This, it was said, “very likely” resulted in the gene pool of the wild bugs getting stronger. But this conclusion may be unfounded, according to the company behind the mosquitoes, Oxitec, as well as other experts in the field.
More than a nuisance, mosquitoes can be carriers of deadly diseases. Trying to control them has been an ongoing problem for centuries, and in recent years a new, apparently effective method has emerged.
Genetically engineered mosquitoes are designed to be unwitting traitors to their species, bringing their numbers down from the inside with little collateral damage. Oxitec’s bugs, for instance, have two new genes inserted into their genomes: one is a dominant lethal gene that males pass down to their offspring, ensuring that less than five percent survive. The second gene makes the bugs fluorescent under certain light, so they’re easily distinguishable from wild mosquitoes.
By releasing these lab-grown mozzies into the wild, the idea is that local populations should be cut down drastically. There’s no need to spray poisons, or introduce new species, or anything else that could harm the environment. And because the engineered mosquitoes have the self-limiting gene, their influence should fade away from the local gene pool fairly quickly.
For the better part of a decade, Oxitec has been conducting scientific studies that largely show the technique to be safe and effective – but last week a paper was published in the journal Scientific Reports with a different conclusion. That study focused on how genes from the engineered mosquitoes had entered the native Brazilian population, and went on to speculate that this transfer had “very likely” served to make the wild mosquitoes more genetically robust. This declaration led to some hyperbolic headlines (including, admittedly, our own).
To get the full story, New Atlas has spoken to Oxitec and Jason Rasgon, a Professor of Entomology at Pennsylvania State University with no ties to the company or the study.
The controversy regarding the paper centers on a phenomenon called “hybrid vigor.” According to this idea, an organism that’s a first-generation hybrid between two closely related species will be stronger and fitter, genetically speaking, than its parents. Hybrid vigor can usually be seen in action under carefully-controlled circumstances, such as selectively breeding plants or animals like dogs to specific desired traits.
According to the researchers on the study, thanks to hybrid vigor the GM mosquitoes may have inadvertently strengthened the wild population. Oxitec’s bugs were developed by first crossing two existing strains of Aedes Aegypti mosquitoes from Cuba and Mexico. Genes from these two strains were found in the local population a few years after millions of these bugs were released into the area around the city of Jacobina, Brazil.
One sentence in particular, from the paper’s Discussion section, seemed to be the most inflammatory.
“The three populations forming the tri-hybrid population now in Jacobina (Cuba/Mexico/Brazil) are genetically quite distinct, very likely resulting in a more robust population than the pre-release population due to hybrid vigor.”
Oxitec, as well as independent experts like Rasgon, disagree.
“Strictly speaking, it’s the ‘very likely’ in that sentence that I disagree with,” Rasgon tells New Atlas. “It is not very likely, there are no data in the paper supporting that statement. It is a hypothesis, and thus a good argument for doing the experiments to see if it is true.”
“If the paper had said something like ‘…it is possible that this will result in a more robust population etc etc…, future experiments should be conducted to confirm or refute this hypothesis,’ there would be no issues here at all. And we probably wouldn’t be having this conversation."
The strong wording of that phrase, which was echoed in a Yale University press release, was latched onto by the media. In turn, Oxitec had to go into damage control mode. The company isn’t denying that genes from its own mosquitoes made it into the wild population – this has long been known to occur. Instead, Oxitec finds fault with the idea that this gene transfer is dangerous in any way.
“The data published in this paper and in the entire body of peer-reviewed literature do not support this (hybrid vigor) hypothesis,” an Oxitec representative told us. “The natural, background genetics passed into the local population declined over time, after releases of Oxitec mosquitoes stopped.”
Two traits in particular were discussed in the paper as potentially troublesome: whether the new hybrid population would be more resistant to insecticides, and whether they could transmit disease more readily. Of course, these would be concerning to find in the new hybrid population – but despite the alarming tone of the paper, no trace of these traits was found in this or any earlier study.
“The study itself describes that the Oxitec mosquito shows no greater capacity for disease transmission than do wild ones,” the Oxitec rep tells us. “And a previous published study (which shares some of the same authors as this paper) found that the Oxitec mosquito is not resistant to commonly used insecticides. The paper describes data that shows the Oxitec mosquito to be safe.”
Also of note is that the two inserted genes – the self-limiting gene and the fluorescent gene – weren’t found in the wild population, indicating that the engineered mosquito was working as intended.
The saga is far from over. Nature Research, the publisher of the paper, has acknowledged the criticisms and said that an editorial response will be released when the issue is resolved. And it’s not like the paper doesn’t have useful things to say – it might just have been saying them the wrong way.
“This paper highlights the importance of monitoring changes in the genetics of host populations during a genetic control trial, and as far as I am aware, it is one of the first to do so,” Rasgon tells us. “It is very important work. It also suggests ways that these issues can be avoided during future releases.
“I have no problems with speculating in a paper discussion (I mean, where else would one speculate?) but it needs to be very clearly identified that it is speculation and not data-driven.”