The highlands of western Panama are an odd place for a fish farm. But then the Atlantic salmon reared there is unusual. It’s genetically engineered to grow twice as fast as typical farmed salmon, containing a growth-hormone gene from Chinook salmon and DNA from an eel-like species. It could boost production and reduce the environmental burden of salmon farming. If it ever gets into the shops, that is.

In 2012, the AquAdvantage salmon, reared by the US-based AquaBounty Technologies, looked set to become the first GM animal approved for human consumption. A panel appointed by the US Food and Drug Administration (FDA) said the fish is safe to eat and poses no threat to the environment. Approval seemed imminent, despite opposition from environmental groups alarmed by the prospect of a GM salmon escaping. But the FDA stalled. Today, AquaBounty’s fast-growing salmon is stuck in the pipeline.

Researchers have developed dozens of GM animals over the years, from pigs full of healthy omega-3 fatty acids to chickens resistant to bird flu. They insist their work can help solve one of our greatest problems: how to feed a swelling global population with dwindling natural resources. “We need to produce more food with less land and water while not degrading the environment for future generations,” says James Murray of the University of California, Davis, who has developed GM goats that produce milk containing antibacterial proteins that can prevent diarrhoea. “We need to use all the tools we have available, and genetic engineering is one of them.”

So can we expect to see GM meat and milk on our shelves any time soon?

Genetic modification can enrich animal products and boost production efficiency, says Murray, whilst simultaneously improving animal welfare. But so far not one transgenic animal - an animal with genes transferred from another species - has made it to market. In the developed world at least, there is strong resistance from environmental groups, sections of the public are wary, and regulatory authorities appear to be dragging their heels. Such a hostile landscape has made investors cautious and strangled government funding, stifling research.

“It’s extremely frustrating,” says Murray, who has transferred his GM goat project to Brazil, where children still die from diarrhea and the government is more receptive.

Accelerated selective breeding

The technology, however, has continued to advance in recent years. Powerful new genome-editing tools mean researchers can make very precise changes to DNA, altering specific genes without changing other parts of an animal’s genome. Now, instead of inserting genes from distantly related species, researchers can “improve” livestock by replicating small genetic variations found naturally in different breeds of the same species. “We’re not slapping in genes from other species, we’re making changes in the exact places we want them to create mutations that exist in animals we already eat,” says Bruce Whitelaw of the Roslin Institute at the University of Edinburgh.

In other words, it’s an accelerated form of selective breeding – and some in the field hope it will be a game changer. They argue that gene-edited animals with mutations swapped around within their own species are less risky than traditional transgenic animals and therefore should not have to clear the same regulatory hurdles. Some even argue that they should not be regulated at all.

One thing is clear: such debates remain theoretical until researchers come up with a more compelling case than fast-growing salmon. “We need to produce animals with real and distinct benefits that can’t be produced any other way,” says Whitelaw, “then we can ask society, ‘Do you want it?’”

Whitelaw is using genome-editing tools to make European domestic pigs resistant to a deadly viral disease called African swine fever, for which there is no vaccine. Wild pigs in Africa are resistant to the disease but they cannot be crossbred with European domestic pigs. So Whitelaw and his colleagues pinpointed a tiny genetic variation that they think bestows resistance – a mutation in a single base pair among three billion – and precisely replicated it in the fertilised eggs of a European domestic pig. In October 2013, Whitelaw’s team announced the birth of five pigs carrying the mutation. It remains to be seen how these GM pigs cope when infected with the virus. But if they are resistant, the benefits are clear. “It would be good news for the animals and good news for producers,” says Whitelaw.

Tweaking the genomes

Across the Atlantic, Scott Fahrenkrug is betting that hornless cattle can serve up a similar win-win situation that will finally push a GM animal into mainstream farming. In 2010, Fahrenkrug, a molecular geneticist then at the University of Minnesota, saw a TV news story about how farmhands use hot irons to burn off horns from dairy cows. The practice is necessary to stop the animals from injuring their handlers and each other, but it’s also painful for cows and expensive for farmers. “It’s a pretty nasty practice and everyone wants to get rid of it,” says Fahrenkrug. But it would take decades to produce hornless dairy cattle with conventional selective breeding methods.

Instead, Fahrenkrug’s St. Paul, Minnesota-based startup Recombinetics is tweaking a few letters in the genomes of dairy cattle to render them hornless, while maintaining cherished milk-production qualities. They precisely edit the DNA of cells from Holstein bulls – prized for producing huge amounts of milk – to replicate the letters that make Red Angus cattle – a breed prized for beef – hornless. The researchers then use cloning to transform edited cattle cells into embryos, direct embryo editing, to produce hornless calves.

For Whitelaw, these hornless cattle and his disease-resistant pigs should be treated differently than traditional transgenic animals. Think of it as a spectrum, he says. At one end are transgenic animals carrying genes from different species. Pretty much everyone agrees that such animals require in-depth review. At the other end are gene-edited animals in which a few DNA letters have been tweaked to replicate a gene found in the same species. Here the risk is much smaller, says Whitelaw, and the regulatory pathway should be less onerous: “I think [gene-edited animals] should be regulated. But if we’re talking about changes to one or a few base pairs, it should be fast-tracked in some way.”

Ethical minefield

Fahrenkrug goes further, arguing that his hornless cattle should not be regulated. We’ve been swapping around traits within species ever since we began selecting and breeding animals, he says, and “regardless of the breeding technique you use, that’s never been the purview of regulatory bodies”.

The FDA is still figuring out how to deal with gene-edited animal products, as is the European Food Standards Agency. But even if some gene-edited animal products do eventually get the nod, will the public want to eat them? “We don’t know the extent to which people distinguish between gene editing and GM, so it’s hard to predict how people will react,” says Ann Bruce, a social scientist at the University of Edinburgh’s ESRC Institute for Innovation Generation in the Life Sciences.

What we do know is that many people’s objections to GM animals seem to be on ethical grounds. For people opposed to the very idea of genetic tinkering, the method by which that is achieved are irrelevant. Still others insist that any genetic modification to animals, regardless of how small and precise, can have unintended and potentially dangerous consequences.

Ultimately, then, the onus remains on the scientists to come up with a must-have application: a gene-edited animal that is demonstrably too good to turn down. “To persuade the public you need a product people really want, something transformative,” says Bruce.