Technology News - February 12, 2004

Improving fish food

The aquaculture industry is reeling in the wake of a widely publicized report showing that farmed salmon contain significantly more contaminants than wild salmon. The report, published last month (Science 2003, 303, 226–229), has focused worldwide attention on fish nutrition by implicating the food fed to farmed salmon as the source of the contamination.

Experts in the field say that many technologies have the potential to reduce the concentrations of PCBs, dioxins, and pesticides in fish food. Until now, however, much of the research on reducing the amount of contaminants in fish food formulations has foundered because of a lack of funding, they say.

The Science study, which was funded by the Pew Charitable Trusts, represents the most comprehensive analysis of contamination in farmed fish conducted to date. A group led by Indiana University researcher and ES&T Associate Editor Ron Hites analyzed more than 2 metric tons of farmed and wild salmon and found that aquacultured salmon from Europe and North America contained significantly higher levels of 14 organochlorine contaminants than the wild fish.

Because farm-raised salmon that comprises the bulk of the salmon eaten today has been shown to harbor contaminants like PCBs, scientists are researching ways to include more plant-based proteins and oils in their diets.

According to the U.S. EPA’s risk-based guidelines for cancer, the combined concentrations of PCBs, toxaphene, and dieldrin in all the tested samples of farmed salmon were so high that consumers could only safely eat no more than one serving of the fish per month. Salmon from farms in the European Union (EU) had the highest levels of contaminants. The EPA guidelines are much more stringent than the levels set by the U.S. Food and Drug Administration or Health Canada, which would give both wild and farmed salmon a clean bill of health.

Hites and his colleagues stressed that a number of the contaminants they documented in the farmed fish are also associated with noncancer heath issues, including endocrine disruption, which were not considered in their analysis.

The contaminant levels found in European fish would be lower if the researchers had tested salmon on the market after July 1, 2002, when EU legislation limiting dioxin levels in fish went into effect, says Stuart Barlow, director general of the International Fishmeal and Fish Oil Organization, an industry group. Five European fish-processing plants are now using carbon filtration systems that reduce PCBs and dioxins, he says. However, Barlow acknowledges that the filters don’t target PCBs as efficiently as dioxins, and they are unlikely to reduce pesticide levels.

Although filtering can reduce the amount of contaminants in the fish oils fed to the farmed fish, which serve as the source of most of the fat-soluble contaminants in their diets, the approach is generally too expensive to be practical, says Michael Ikonomou, a research scientist for Canada’s Department of Fisheries and Oceans (DFO). In collaboration with Canada’s University of British Columbia (UBC), DFO is in the process of conducting what they believe is an even more comprehensive study that compares wild and farmed salmon. Ikonomou says it mirrors the size and scope of the Pew study but also examines heavy metal and pesticide contamination as well as hormone and nutrient profiles. A major aim is to determine the sources and magnitude of the contaminants of concern by examining their uptake patterns observed in the tissues of various stocks of wild and farmed salmon. Unfortunately, no results are yet available.

The more economical way to reduce the amount of contaminants in farmed fish is to use plant-based materials as a significant source of some of the protein and oils in their diet, Ikonomou says. Plant-based oils such as canola are not expected to contain dioxins or other global environmental contaminants, he explains.

“We could make dramatic reductions in the amount of dioxins and PCBs in farmed fish,” says David Higgs, the research scientist who heads DFO’s fish nutrition program. He has been researching ways to use plant-based materials in fish food for more than 25 years. According to research conducted on trout by North American and European investigators, oilseed and protein concentrates based on canola and soy offer great promise to extensively replace fish meal in diets for salmon, he says. Higgs says that he and Ikonomou have tried unsuccessfully to get funding for a study that would investigate just how much contaminants could be reduced in salmon that were fed diets with higher percentages of plant-based materials.

Barlow agrees that additional sources of fish meal and oil need to be developed. “There’s a limit to how much fish meal and oil can be produced [from fish]. Our worldwide production has been more or less static for the last eight years…. That’s not going to change unless something very surprising happens,” Barlow explains. Because the percentage of fish raised by aquaculture is projected to grow steadily (Environ. Sci. Technol. 2003, 37, 429A-430A), fish food formulators must find ways to use even more alternative oil sources, Barlow says.

Plant oil sources can also help fish farmers during El Niño years when drops in the supply of pelagic fish caught off the coast of South America, which are the main source of world’s fish oil and fish meal, cause prices for these commodities to skyrocket, Higgs says. In fact, he says that the program he has been working for since 1977 to formulate fish diets that are based on more protein and lipid from plant sources originated in response to a particularly difficult El Niño year.

Salmon farmers can never get away entirely from fish oil, Barlow claims. Fish oil is used to supplement the diet of fatty fish like salmon to ensure that they contain a sufficiently high percentage of the two long-chain, highly unsaturated, omega 3 fatty acids found in fishæ eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), he explains. These fatty acids are in high demand by consumers because they have been linked to various human health benefits, including reduced risk of heart disease, he says.

That may not always be true. In the last 10 years, researchers have shown that humans can metabolize and produce long-chain fatty acids from shorter-chain vegetable sources, says Miriam Jacobs, a nutritionist and toxicologist who has also studied the contaminants in salmon (Environ. Sci. Technol. 2002, 36, 267A–268A). But she acknowledges that the process “is not terribly efficient—you need to have 8 or 10 times as much quantity-wise.”
Scientists like Higgs are focusing their efforts on using oils from oilseeds that are rich in linolenic acid, an omega 3 fatty acid that is a precursor to EPA and DHA. The oils include flaxseed, hempseed, soy, and canola. They’re also looking at wheat, barley, corn, oats, canola, sunflower, and soybeans as sources of dietary proteins for salmon and other fish.

“We have reduced the [amount of fish in] fish meal level [from the levels] of a few years ago by 60%; we can get it down to 17 or 20% of the meal fed to the fish without breaking a sweat,” says Ron Hardy, director of the Aquaculture Research Institute at the University of Idaho. “Getting the fish oil level down presents a bigger challenge,” he says. If he received as much money for addressing the problem as the Pew Charitable Trust donated to defining it, he could solve it easily, he claims.

Reducing the amount of fish material used in the oils is challenging because the price of fish oil tends to be lower than that of vegetable-based oils in the non-El Niño years, Higgs says. Getting away from the oils is also difficult because “the feed companies want to make sure that the omega 3 acids, especially EPA and DHA, are there at the right levels in the salmon going to market,” he adds.

Higgs says that UBC and DFO researchers have been able to replace up to 75% of the supplemental fish oils fed to fish with cold-pressed flaxseed oil. But he is particularly enthusiastic about an approach that he and co-workers from CSH Innovations Ltd., of Vancouver, B.C., hope to patent, which uses animal proteins and lipids from fish or poultry together with pretreated oilseeds. This technology, which involves co-processing the oilseeds with the animal-based materials, is unique because animal and fish feeds generally include protein and lipid from oilseeds, fish, or poultry that are processed separately, not together.

The process capitalizes on what Higgs refers to as “under-utilized or poorly utilized animal sources—processing wastes, or poultry offal, cod heads, and associated viscera.” The animal wastes are cooked with the oilseeds to denature proteins and free the cellular water bound in the animal cell proteins. This cellular water is then used to wash the water-soluble oilseed meats. The heating and washing processes can markedly reduce the anti-nutritional factors, such as protease inhibitors, phytates, oligosaccharides, and glucosinolates, that can be found in oilseeds, he says.

If some fish or fish offal is used in the process, the oil rendered out of the process will contain the beneficial EPA and DHA fatty acids. “You can even use different combinations of oilseeds to create all kinds of neat blends of lipids with different fatty acid compositions with different purposes in mind,” he says. The process may be cost-effective because fish meal is generally two times as expensive, on a per-kilogram-protein basis, as poultry lipids or plant-based meals made from grains like soybeans. Higgs says that he and his colleagues can use the technology to produce a protein concentrate that derives 90% of its protein from oilseeds and is also enriched in minerals from the animal protein.

“We’ve also done in vivo digestibility work using Atlantic salmon as the test species and found that the availability of protein in these concentrates was among the best ever seen, relative to fish meal protein—up around 94–96% available protein—but we haven’t taken it past that due to lack of available funds,” he says.

Higgs does not know how consumers will react to salmon raised on feeds containing these new protein and lipid sources. Fish fed diets too high in vegetable matter can taste different. Jacobs says that fish producers with whom she has spoken say that such fish have been rejected by Japanese consumers because they don’t taste fishy enough.

However, Higgs says that researchers throughout the world are trying to determine the best nutritional strategies so that fish like salmon can be raised using diets that are largely plant-based and then “finished” during the last months of their lives with diets higher in the EPA and DHA omega 3 highly unsaturated fatty acids.

Using plant-based material or products that would otherwise not be directed into the diets of fish increases the eco-efficiency of fish farming, according to environmentalists. At present, it takes 1.3 kilograms of wild fish to produce each kilogram of aquacultured fish, says Rebecca Goldburg, a senior scientist with Environmental Defense, an environmental group. The small pelagic fish used to produce fish meal and oil are important parts of the marine food web, she says.

The future of farmed fish is bright, Higgs claims. In the long term, farmed salmon have the potential to contain lower levels of contaminants than their wild counterparts, he says. —KELLYN S. BETTS