A genetically engineered salmon (top) grows twice as fast as its wild counterpart (bottom). Credit: AquaBounty

In the remote highlands of Panama, in tanks protected by netting, barbed wire and guard dogs, swim the world’s most expensive and scrutinized fish. These swift-growing salmon have been at the centre of a 18-year, US$60-million battle to bring the first genetically modified (GM) animal to US dinner tables — a struggle that may be nearing its end.

Last week marked the end of the public’s opportunity to weigh in on a US Food and Drug Administration (FDA) draft assessment of the salmon. Genetically engineered to grow twice as fast as their unaltered brethren, the fish pose no significant environmental threat to the United States when grown in landlocked tanks, says the FDA. The agency needs only to finalize that assessment before deciding whether to approve the fish for human consumption. The number of opportunities for a surprise delay — a recurring theme in the history of these salmon — is dwindling (see ‘Against the current’).

boxed-text

Environmental groups are preparing to take the battle to consumers by fighting the sale of the fish in grocery stores across the country. Others point out that it will be years before the salmon are anything more than a curiosity. At full capacity, the Panama facility can produce only about 100 tonnes of salmon a year, says Gregory Jaffe, director of biotechnology at the Center for Science in the Public Interest, a consumer group in Washington DC that monitors the regulation of GM foods. That amount is a trifle compared to the roughly 230,000 tonnes of farmed Atlantic salmon that the United States imported in 2012. “You’d have to try hard to eat it,” says Jaffe. “It won’t be as hard as winning the lottery, but it will be close.”

nature.com/gmcrops

For the firm that developed the fish, AquaBounty Technologies of Maynard, Massachusetts, those 100 tonnes are a hard-won prize. In 1989, the salmon were engineered to overexpress a growth-hormone gene. The result: ‘AquAdvantage’ fish that grew to full size in around 18 months rather than the usual 3 years. The company applied for FDA approval in 1995 and has been stuck in regulatory limbo ever since. AquaBounty has had to demonstrate the food’s safety, and gauge the environmental risk of the sterile fish escaping its tanks and successfully mating with wild salmon. By contrast, the FDA approved the first GM crop for human consumption — the Flavr Savr tomato — after just three years of regulatory consideration.

Cash crisis

The uncertainty has taken its toll. To save money, AquaBounty has reduced its staff by more than half. Last year, the company sold off its research and development arm and lost one of its biggest investors. In March, AquaBounty came within a week of running out of cash, says chief executive Ronald Stotish. The firm was saved by last-minute refinancing and fresh investment from Intrexon, a synthetic-biology company based in Blacksburg, Virginia.

At first glance, the Panama facility hardly seems to be the key to financial prosperity. With salmon selling for around $6.50 per kilogram, AquaBounty would make less than $1 million each year from the salmon. It would take decades for the company to make back its $60-million investment if it relied solely on the Panama farm.

Stotish says that the company must expand. Following FDA approval, AquaBounty hopes to sell its salmon eggs to farmers and expand to markets in Argentina, Canada, Chile and China.

To sell AquAdvantage fish in the United States, each farm would require separate FDA approval, but because the food safety of the fish has already been vetted, the approval process would require only an environmental evaluation, says Jaffe.

Yet even with regulatory approval, the battle over AquaBounty’s salmon will be far from over. In March, several speciality grocery stores, including Whole Foods, an inter­national chain based in Austin, Texas, said that they would not sell AquAdvantage fish. Lawmakers in Alaska and Oregon, which both export wild salmon, have repeatedly tried to block the GM fish because they fear contamination of the wild stock and worry that it could drive down the price of farmed salmon.

AquaBounty’s long struggle has discouraged other US companies from producing GM animals for food. Mark Walton, chief marketing officer at Recombinetics, an animal-biotechnology company in St Paul, Minnesota, says that his company will focus initially on medical applications — using modified farm animals as disease models, for example — rather than on livestock for food. Medical applications of GM technology do not stir consumer passions in the same way as GM foods, and there is a regulatory precedent: in 2009, the FDA approved a goat that makes an anti-clotting drug in its milk. If Recombinetics invests in agricultural products, Walton adds, the items will probably be marketed outside the United States first. “The AquaBounty example has [made] the company very sceptical about how much investment to pour into the US regulatory process,” he says.

Yet Stotish says that GM animal products will inevitably find their way to grocery stores. He points to heavy investment in the technology in China, where dozens of GM farm animals are in development. “I think we will end up eating genetically modified animals of a variety of species,” says Stotish. “But they’ll come from other countries.”