Fishing for answers with science

The Vancouver Sun: Fishing for answers with science

Bob Devlin, a leading world authority on GE fish, has been involved with project for two decades
By Sarah Schmidt, Postmedia News February 24, 2011 Bob Devlin has been working toward this moment for more than two decades.

The world-renowned scientist at the Department of Fisheries and Oceans started his research on genetically engineered coho salmon back in 1989 at a government laboratory in West Vancouver. That's when the pioneering researcher set out to generate strains of GE fish to help government regulators handle what is now before them: the first application to commercialize GE fish to be mass produced at a fish farm destined for the dinner plate.

AquaBounty Technologies is already close to getting a decision from the Food and Drug Administration in the United States about its GE salmon, grown at the company's research facility in Prince Edward Island using technology developed at Memorial University. The company cleared an important hurdle in the U.S. last August, when the FDA's preliminary analysis concluded that the salmon -engineered to grow twice as fast as regular salmon -are safe to eat and not expected to have a significant impact on the environment.

The company is now consulting with Canadian regulators in preparation for a formal application to seek Canada's approval to transform the company's hatchery in P.E.I. into a commercial operation. The eggs would then be shipped to an inland fish farm in Panama to raise GE salmon for American consumers.

But Devlin's environmental risk assessment is front and centre of a raging international debate about the merits of AquaBounty's quest.

With the decline in wild salmon stocks, the company says its AquAdvantage salmon is an economically viable and environmentally sustainable alternative for the farmed Atlantic salmon industry. The fish contains a growth hormone gene from the Chinook salmon that is kept active all year round by a genetic on-switch from an eel-like fish called the ocean pout, so it reaches market size twice as fast.

"It's kind of embarrassing to admit you're on a 20-year project," but progress is slow "when you're dealing with long-cycle organisms," Devlin told Postmedia News in a rare interview.

Until now, Devlin, considered one of the world's leading authorities on risk assessment of GE fish, has been elusive during this very public debate.

The fisheries minister's office has usually turned down requests for interviews with him, internal DFO records released under access to information indicate. Political sensitivity over the work of Devlin and his team is hardly surprising, given he has published or co-written dozens of academic journal articles focusing on scientific uncertainty regarding the environmental risks of GE salmon.

Devlin's work has found, among other things, that his growth-enhanced GE coho salmon had ravenous appetites that out-competed and even ate native salmon in a laboratory environment. The genetic engineering process also alters their behaviour, so they are likely to explore novel prey and new areas. This may expose them to predators.

But their lower fitness when they escape does not necessarily translate into permanently lower environmental risk, the research shows. And despite efforts to achieve 100-per-cent sterility of all-female populations to prevent them from passing on their genes, Devlin's team has never achieved it, with sterility between 97 to 99.8 per cent effective.

This "is really high, but if there are large numbers of escapes, then that's not quite high enough for biological containment yet, so we're still working on trying to improve methods," Devlin said in an interview.

Devlin's work is so highly regarded that when two other leading scholars in the field spoke during special FDA hearings last September about AquaBounty's application, Anne Kapuscinski of Dartmouth College in New Hampshire and Fredrik Sundstrom of Uppsala University in Sweden relied heavily on Devlin's published work to flag "major concerns" with the FDA's preliminary analysis and to ask the FDA to conduct a full Environmental Impact Statement that assesses the potential genetic and ecological impacts that AquAdvantage salmon could have on wild fish and other aspects of the environment.

Internal DFO records released to Postmedia News under access to information also show Kapuscinski asked Devlin to sign on to the presentation, but he declined. "I think I should not participate in this statement as it could be interpreted to be a policy statement from Canada," Devlin wrote colleagues at the Department of Fisheries.

In an interview, Devlin says he leaves the regulatory work to regulators, and continues to focus on generating scientific data unencumbered by any political constraints or commercial interests.

Devlin spent the 1990s generating strains of GE coho salmon, also called transgenic salmon, so his team could conduct risk assessment research. Devlin's lab then developed semi-natural environments over the next decade to test possible environmental risks.

This shift to semi-natural environments dramatically slowed down growth rate of his GE salmon compared to wild fish -raising a new set of questions to be answered.

"Under simple laboratory environments, those fish have a huge advantage and will out-compete wild type fish. . . . But nature isn't so simple. And so when we set up these naturalized environments where there are predators in the environment and there is limited food supply, what we find is the transgenic fish are out hustling for food more. And in nature, you have to trade off whether you're going to go get food but you're taking a risk that you're going to be food, too," said Devlin.

The dilemma for scientists is to make sense of "a plus and a minus from very simple experiments, and try and combine those in a meaningful way to what the net effect as would occur in the full complexity of nature. It's extremely difficult. Well, I can't do that yet. I do not have enough data," said Devlin.

The other glitch is laboratory rearing conditions themselves change how wild-type fish grow, so Devlin's team is trying to chalk up differences in his GE salmon to the transgene or the lab environment. He's edging closer to an answer, thanks to more recent experiments conducted in "mesocosms" -large circular containers that hold one million litres of salt water to mimic nature.

"Wild type fish that are being grown in the mesocosms are starting to look like wild type fish from nature, so now we can start to believe, hopefully, the data for transgenic fish that are being raised in the same environment. So the key is to be able to know that your controls in the laboratory equate to wild type fish in nature. If you can make that relationship, then you can start to believe the transgenic fish data," said Devlin.

"My role is to keep focusing on the science and provide advice, and you know they can choose to take some of that advice or not. But if I can't provide objective scientific information, I'm not a happy person."