© JOEL HOLLAND
In the mid-1990s, microbiologist Cecil Forsberg of the University of Guelph in Ontario and his colleagues thought they’d achieved a pig production breakthrough: they had genetically engineered swine that could digest the phosphorous compounds in their feed. Phytase, an enzyme that breaks down phosphorus-containing phytate in plants, is produced by the gut bacteria of cows and other ruminants, but it is not made by pigs. Forsberg’s team borrowed a phytase gene from E. coli and a fragment of mouse DNA that mediated the enzyme’s production in the salivary glands, injected the genetic construct into pig zygotes, then inserted those zygotes into fertile sows. “In the end, we had approximately 30 different lines of pigs,” Forsberg recalls. The researchers screened the animals for levels of phytase production in the salivary glands, narrowing the field to the four most promising. “Then we came down to one line”—the Cassie line, named for its founding animal—“which performed satisfactorily and contained three copies of the transgene,” he says.
The Enviropig, as it came to be known, produced manure with lower levels of phosphorus, notorious for leaching into groundwater beneath pig farms and fueling algal growth in local streams and lakes. The transgene would also eliminate the high cost of adding mineral phosphorus or commercially produced phytase to feed to ensure that the pigs get the nutritional phosphorus they need. Years of research demonstrated that the trait was stable for more than nine generations and that phytase expression was primarily limited to the salivary glands, with only trace amounts of the enzyme found in any of the animals’ other tissues; the protein composition of the meat appeared normal. Forsberg and his colleagues submitted the Enviropig for regulatory review by the US government in 2007, and to Canadian agencies two years later, and the assessments, while slow and still preliminary, had been positive, Forsberg says.
But like most plants or animals whose genomes have been engineered, commonly called genetically modified (GM) organisms, the pigs were not so well received by the public. Anti-GM activists argued that, because of their low-phosphorus manure, if the Enviropigs were approved for human consumption, “big, bad agriculture [would have] an excuse to put them in even more concentrated facilities,” says Alison Van Eenennaam of the University of California, Davis, who studies how DNA technologies can be applied in the beef-production industry. “They really targeted it and made it a bad thing.”
The lack of public acceptance trickled down to the industry, and in 2011, the Ontario Pork Producers Marketing Board, a long-time funder of the project, withdrew its support. Unable to find another industry backer, the researchers could not afford the costs of raising the pigs and navigating the long and extensive regulatory process, and more economical ways of giving pigs phytase—such as developing a phytase-producing GM corn, which can be sold as an enriched powder to be added to the animals’ diets—had begun to crop up. The following May, the researchers euthanized their herd and withdrew their applications from regulatory review. “These pigs were healthy pigs that did perform as they were designed to perform. They just didn’t meet the social requirement,” says Forsberg. The Enviropig was dead.
In the past 30 years, dozens of GM animals have been developed, none managing to get a foothold in a market that is wary of genetically engineered food products in general, and GM meat in particular. “Who would have thought when we started [manipulating animal genomes] in the early 1980s that at this point we would have no animals approved?” says James Murray of the University of California, Davis, who develops goats whose milk contains an antibacterial protein found in human breast milk that could help treat childhood diarrhea. “It’s been over 30 years. I made my first transgenic sheep in 1985. We were all making [GM] mice before that, with an eye toward agriculture.”
Some speculate that the unreceptive social environment has led the US Food and Drug Administration (FDA) to drag its feet on green-lighting a faster-growing GM salmon, which has seemingly been on the brink of market approval for the past three years. Of course, even if the GM salmon, developed by Massachusetts-based AquaBounty Technologies, is approved for human consumption, a GM-resistant public does not make for a fertile market. Already, dozens of grocery chains have refused to carry the GM salmon, even if or when it’s legal. Such resistance has stunted funding, limiting the amount and scale of research into developing GM food animals. Some projects have moved to other countries in search of a more welcoming climate; others, like Enviropig, have discontinued work altogether. “It’s a very harsh environment,” says Murray.
The field may be at a turning point, however. Within the past five years, genome-editing tools—namely, TALENs (transcription activator-like effector nucleases), the CRISPR/Cas9 system, and improved zinc finger nucleases (ZFNs)—have come online, allowing researchers to make much more precise changes to DNA, inducing mutations in a specific gene, even substituting a single base pair. Groups are now using these tools to introduce “natural” alleles that exist in closely related populations, thereby improving livestock without the introduction of transgenes from distantly related taxa. “The possibilities that these new editing tools present us with will be the game changer that GM technology’s been waiting for,” says molecular biologist Tim Doran of the Commonwealth Scientific and Industrial Research Organisation (CSIRO) Australian Animal Health Laboratory.
“There’s probably more excitement now in the animal industries over genetic engineering than there has been ever, since this whole process started,” agrees Charles Long of the Texas A&M University College of Veterinary Medicine and Biomedical Sciences. “We really believe that the science part, the hardest part, is now behind us.”
The GM gauntlet
© JOEL HOLLANDThere are currently no GM animals marketed for human consumption anywhere in the world. The product closest to approval is the AquAdvantage salmon, created by Garth Fletcher and colleagues at Memorial University of Newfoundland in 1989 with the insertion of a growth hormone gene from Chinook salmon and a promoter from the eel-like ocean pout into fertilized Atlantic salmon eggs. “By the end of 1990, we could say, ‘Yep, those rapidly growing fish are the ones that are carrying the growth-hormone gene,’” says Fletcher, a now semiretired fish physiologist who oversaw the development of the salmon, which grows to market size in half the time it takes conventionally farmed fish.
Who would have thought when we started manipulating animal genomes in the early 1980s that at this point we would have no animals approved?—James Murray,
University of California, Davis
AquaBounty, then known as A/F Protein, initiated discussions with the FDA in 1993, officially filing two years later for permission to develop the genetic modification of the fish as an Investigational New Animal Drug (INAD), a first step on the road to approval. In the course of the GM salmon’s long regulatory review, AquaBounty researchers have demonstrated to the FDA’s satisfaction that its flesh is compositionally the same as meat from wild-type fish. The only difference is that the GM fish grow nearly twice as fast while consuming 25 percent less food. To prevent escape and interbreeding with wild fish, the AquAdvantage salmon are all female and triploid (like many farm-raised fish), making them sterile, and they are kept in inland, indoor enclosures with multiple physical barriers between them and the outside world. Thanks to these measures, a 2010 environmental impact assessment found that breeding these salmon will likely not harm wild fish. That September, the Veterinary Medicine Advisory Committee (VMAC) appointed by the FDA unanimously agreed that the fish “is as safe as food from conventional Atlantic salmon, and that there is a reasonable certainty of no harm from the consumption of food from this animal.” Approval was expected from the agency soon thereafter.
Still, as this article goes to press, the FDA had released no word of the AquAdvantage salmon’s approval. “Theoretically there’s meant to be a decision then, and not four years later,” says Van Eenennaam, who coauthored a 2011 commentary on the GM product’s regulatory journey in Nature Biotechnology.1 “It’s just stuck. It’s like nothing’s happening.”
Rumors are circulating that the holdup is political, driven by a GM-resistant public and the wild-caught salmon industry. “The FDA has said it’s safe for human consumption, and they said it’s safe for the environment—so it’s in the hands of the administration,” says Scott Fahrenkrug of Recombinetics, a biotech company that focuses on the use of genetic technologies for agriculture and biomedicine.
The FDA declined The Scientist’s request for comment on the AquAdvantage salmon review, though a spokesperson noted that much of the process is publicly available on the agency’s website. “Although the existence of an application is generally considered to be confidential business information, we have been able to provide information on the AquAdvantage Salmon application because the applicant chose to publicly disclose their submission,” Theresa Eisenman of the FDA Office of Media Affairs said in an e-mail.
But the Web page hasn’t been updated in over a year, and the field is growing antsy. “It sure would help if we knew that [the FDA] was a science-based regulatory organization—that once they’ve determined it’s safe, the product shows up on the market; it doesn’t get put in the pocket of politicians,” says Fahrenkrug. Van Eenennaam agrees, “I thought we had a scientific evaluation system in place here, but clearly I was misguided.”
As if the regulatory uncertainties weren’t stymieing enough, funding to develop new GM food animals is increasingly hard to come by. Fewer than 0.1 percent of US Department of Agriculture (USDA) research grants have supported work on GM food animals since 1999, according to Nature News (October 17, 2012). In fact, for a brief period about 10 years ago, the agency even included language in its calls for proposals essentially stating that researchers hoping to develop GM animals for agricultural purposes need not apply. And much of the research on genetically altered livestock rests in the hands of academics who depend on such public funds, as the field lacks big industrial backers. “Fifteen or 20 years in, $70 million down the drain, and no decision,” Murray says of AquaBounty’s salmon. “Who wants to invest in the next transgenic animal product?”
For researchers who continue to develop GM food animals despite the challenges, the future of their work now lies in the hands of the FDA. Approval of AquaBounty’s GM salmon could set the stage for a new industry, instill in GM researchers confidence that their goals are realistic, and ignite a flurry of excitement, exposure, and controversy. But how—and when—will the government rule? “The rumors are out there that it’s going to be soon,” says Murray, “but we’ve heard that before.”
Nevertheless, dozens of governments are showing interest in GM livestock, with international workshops scheduled for this August in Brazil and October in Australia to discuss strategies for regulation of food products from such animals. A similar international meeting in Argentina in 2011 drew researchers and regulators from more than 30 countries. Along with updates on the latest GM food animals in development, representatives from the FDA and the European Food Safety Authority (EFSA) spoke about their regulatory approaches. And at informal side sessions, groups of attendees gathered to discuss the future of GM agriculture, says Murray, who spoke at the 2011 meeting and is helping to organize this year’s conference in Brazil. “We’re trying to get regulators from like-minded countries to get together and try to move the regulatory environment forward,” he says.
Why fight the GM battle?
Despite the regulatory, financial, and social uncertainties that accompany the development of GM livestock, some researchers are forging ahead to design animals that could serve as founders for future human food products. Advocates of the approach argue that GM technologies stand to improve animal welfare, reduce disease (both animal and human), and expedite production, as well as create enriched animal products that could help feed a growing populace and prevent nutritional deficiencies.
“If you look at the increasing world population, we essentially have to double [meat] production by 2050,” says Murray. “And we have to do that with less land, less water, and in an environmentally sustainable manner. I don’t think we can actually do that without using genetic engineering.”
An early and still-popular target is milk, with researchers around the world using genetic engineering tools to enrich the milk of cows, goats, and sheep. Many have focused on the antibacterial proteins lysozyme and lactoferrin, which are found in extremely high abundance in human breast milk, but are absent from most animal milks. In 2011, Chinese scientists reported that cow’s milk enriched with lysozyme offered nutritional benefits similar to those of human breast milk.2 Last year, Murray, UC Davis colleague Elizabeth Maga, and others published findings indicating that lysozyme-containing milk from the GM goat they created could help cure young pigs of diarrhea.3 The researchers are now preparing for human trials using milk from new goat lines currently under development in Brazil, where diarrheal diseases still kill a significant number of young children. Scientists are also trying to create cows that produce low-lactose or allergen-free milk. “Milk [is] a good system to study because it [is] external to the animal,” says Murray.
For all intents and purposes, the resulting animal is just the same as a wild-type animal. I just cannot see any risk attached to it whatsoever.—Bruce Whitelaw, The Roslin Institute
Other researchers are hoping to develop animals that grow faster or produce more-nutritious meat. In 2006, nutrition and metabolism researcher Jing Kang of Harvard Medical School and Massachusetts General Hospital reported the creation of pigs carrying a C. elegans gene called fat-1, which encodes a fatty acid desaturase that converts omega-6 fatty acids to the healthy omega-3s that livestock typically lack.4 His group has done the same with fish, hoping to boost their omega-3 levels, and they are now working to add the fat-2 gene, which codes for an enzyme that converts a monounsaturated fatty acid to omega-6. “[We are using these] technologies to increase the nutritional value to balance the modern food,” says Kang. Meanwhile, Long and his colleagues at Texas A&M University are targeting the gene for myostatin, a negative regulator of muscle growth, to increase the muscle mass of cattle and pigs. The faster-growing animals could allow farmers to “have fewer animals on the same amount of land, producing the same amount of meat,” says Long.
Genetic engineers have also sought to protect food animals from lethal pathogens. In 2011, molecular virologist Laurence Tiley of the University of Cambridge and colleagues published their work on GM chickens that express an RNA decoy that distracts the avian influenza virus from replication.5 “The decoys sequester the [viral] polymerase and divert it,” Tiley explains. The birds were partially resistant to the disease, and the researchers are now working to improve the efficacy of the molecular deception. “If we could make genetic modifications that made chickens fully resistant to bird flu, then there’s the potential to scale that up,” says collaborator Helen Sang of The Roslin Institute, an animal research center at the University of Edinburgh.
Another approach to protect animals against their viral foes—one that has already proven itself in the plant science community—involves the transfection of viral genetic fragments into an animal’s genome to trigger the inhibitory phenomenon known as RNA interference (RNAi). The expression of bits of viral RNA blocks the activity of the transcripts involved in viral replication. Researchers around the world are using this approach to develop transgenic fish that are resistant to viral hemorrhagic septicemia (VHS), a deadly fish disease that affects more than 50 types of fish, including many commercially relevant species. Others are also looking at the use of RNAi to protect prawns against viral disease, and CSIRO’s Doran is in the early stages of testing RNAi techniques for improving chickens’ resilience to avian influenza. “There is a lot of work going on at the proof-of-principle stage,” Doran says.
The important step now is to demonstrate how these animals can have a positive impact on agriculture, researchers agree. “Until we can really demonstrate that value [of GM organisms], I don’t think we’ll overcome public qualms,” says Sang.
© JOEL HOLLANDSomething that might go a long way toward winning over a GM-resistant public is the advent of new, precision genome editors. These new tools enable researchers to rewrite genes, making targeted changes to an organism’s DNA with unprecedented specificity. “With the advent of TALENs and CRISPRs, the [GM] landscape is going to change,” says Murray.
Because the tools are more exact, researchers are able to mimic naturally occurring genomic alterations, by tweaking a gene to match an allele found in related populations. This same outcome could be achieved by conventional breeding methods, but it would take years of back-crossing to preserve the animals’ desired production traits. Using precision genome editing, on the other hand, “[you can] take the trait you want—and is natural—and get it in a quick way,” says Fahrenkrug of Recombinetics. And in the end, he adds, “it’s just replicated a natural mutation out there. . . . There’s no other change to the genome anywhere else.”
Fahrenkrug’s group has used TALENs and the CRISPR/Cas9 system to target a gene called POLLED in the cells of dairy cattle, changing it to a version that resembles that of the hornless beef cattle. So far, the researchers have successfully used the editors to develop cell lines with the desired genotype,6 and they hope to soon use these cells to create GM dairy cows that, hopefully, will lack horns. Dehorning dairy cows is a labor-intensive, dangerous, and costly process that dairy farmers would be happy to avoid. “Producers have been wanting to get rid of this trait for a long time, and it just takes too long” using traditional breeding methods, Fahrenkrug says.
Bruce Whitelaw of The Roslin Institute is also using precision genome editing—specifically TALENs and ZFNs—to make minor, arguably natural changes to animal genomes. Last fall he and colleagues announced the generation of five founder pigs7 carrying mutations in the RELA gene?8 of the NF-κB signaling cascade, a key regulator of the immune responses that drive symptoms of African swine fever. The mutations rendered the genes only partially functional, just like those of wild African pigs, which are remarkably resistant to the deadly virus. “We’re trying to convert the Eurasian pig [allele] to an African allele, on the assumption that if we do that, we’ll be able to subdue the [NF-κB–mediated immune] response that the pig would mount if infected,” says Whitelaw, adding that the real test will come when he and his colleagues challenge the animals with the virus later this year or next.
For cases such as Whitelaw’s pigs and Fahrenkrug’s cows, the researchers argue that the products should be viewed differently than traditional transgenics. “For all intents and purposes, the resulting animal is just the same as a wild-type animal,” says Whitelaw, who is hopeful that GM animals created in this way will face a lesser regulatory burden. “At one end you have a transgenic” that carries foreign genes; these products appear harmless so far, but do warrant in-depth regulatory review, he says. “At the other end, we’ve got the editing, which is inherently safe, and is just the same as another animal in that breed. I just cannot see any risk attached to it whatsoever.”
Fahrenkrug adds that such “natural” products should also make the idea of genetic engineering more palatable to the meat-eating public. “It’s likely to get consumer acceptance more quickly because it’s, again, replicating a natural phenotype.” However, with no formal assessments of the public’s reaction to precision genome editing, the jury remains out. “These modern [gene-editing] techniques have got scientists excited in terms of what can be done, . . . [but] we don’t necessarily know how people would react to the more recent developments,” says Ann Bruce, a social scientist at the University of Edinburgh’s ESRC Institute for Innovation Generation in the Life Sciences (INNOGEN) who has studied public perception of GM foods.
Perhaps something more than precision genome editing will be necessary to win over a wary public. “I would suspect it will be a very long time before people accept GM animals,” says Forsberg, who is now wrapping up the final publications on the Enviropig project. “My thinking when we started the project [in 1997] was it might take us 10 years to get [the science] there, but within 10 years, consumers would be accepting of transgenic projects, and indeed I was wrong. It’s difficult to predict a time line here.”
- A.L. Van Eenennaam, W.M. Muir, “Transgenic salmon: a final leap to the grocery shelf?” Nat Biotech, 29:706-10, 2011.
- B. Yang et al., “Characterization of bioactive recombinant human lysozyme expressed in milk of cloned transgenic cattle,” PLOS ONE, 6:e17593, 2011.
- C.A. Cooper et al., “Consuming transgenic goats’ milk containing the antimicrobial protein lysozyme helps resolve diarrhea in young pigs,” PLOS ONE, doi:10.1371/journal.pone.0058409, 2013.
- L. Lai et al., “Generation of cloned transgenic pigs rich in omega-3 fatty acids,” Nat Biotech, 24:435-36, 2006.
- J. Lyall et al., “Suppression of avian influenza transmission in genetically modified chickens,” Science, 331:223-26, 2011.
- W. Tan et al., “Efficient nonmeiotic allele introgression in livestock using custom endonucleases,” PNAS, doi: 10.1073/pnas.1310478110, 2013.
- S.G. Lillico et al., “Live pigs produced from genome edited zygotes,” Scientific Reports, 3:2847, 2013.
- C.J. Palgrave et al., “Species-specific variation in RELA underlies differences in NF-κB activity: a potential role in African swine fever pathogenesis,” J Virol, 85:6008-14, 2011.
Editor's note (June 4, 2014): The date of the conference on removing barriers to GM animals in Australia has changed from September to October.