Debating Genetically Modified Salmon Biotech company AquaBounty has not yet received FDA approval for its fast-growing, genetically-modified salmon. Biotechnologist Alison Van Eenennaam and environmental scientist Anne Kapuscinski discuss the food safety and environmental concerns associated with transgenic fish.

Debating Genetically Modified Salmon

Debating Genetically Modified Salmon

  • Download
  • <iframe src="" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript

Biotech company AquaBounty has not yet received FDA approval for its fast-growing, genetically-modified salmon. Biotechnologist Alison Van Eenennaam and environmental scientist Anne Kapuscinski discuss the food safety and environmental concerns associated with transgenic fish.


This is SCIENCE FRIDAY. I'm Joe Palca. Ira Flatow is away this week. The biotech company AquaBounty Technologies of Waltham, Massachusetts, has developed a genetically modified Atlantic salmon that grows twice as fast as regular salmon. How has it done this? By tinkering with the salmon's genome, adding a growth hormone gene from one fish plus an antifreeze gene from another.

The result: fish that grow to market size rapidly. AquaBounty's application to market these bioengineered fish has been under FDA review for 15 years. Last fall, the Food and Drug Administration held a public hearing and convened a panel of experts to review the food safety and environmental risks posed by these salmon. Their conclusion: AquaBounty salmon was safe, as safe as food from conventional Atlantic salmon.

But some scientists and environmental groups have said there are questions about the safety of genetically modified fish remain unanswered. This hour: the science and safety of bioengineered fish. And joining us for this discussion is Dr. Alison Van Eenennaam. She's an extension specialist in animal genomics and biotechnology at University of California, Davis. She was on the FDA's panel of experts that evaluated AquaBounty's proposal last year. She co-authored a piece in Nature Biotechnology about the company's regulatory battle. She joins us from Eugene, Oregon. Welcome, Dr. Van Eenennaam.

ALISON VAN EENENNAAM: Good afternoon, Joe.

PALCA: Good afternoon. And also with us is Dr. Anne Kapuscinski. She's a professor of sustainability science and chair of the Environmental Studies Program at Dartmouth College. She has been a scientific advisor to the federal government on several issues relating to genetically modified organisms, and she joins us from Hannover, New Hampshire. Welcome to the program, Dr. Kapuscinski.

ANNE KAPUSCINSKI: Good afternoon, Joe.

PALCA: Good afternoon. And if you want to weigh in on the science and the safety of genetically modified salmon, give us a call. Our number is 1-800-989-8255. That's 1-800-989-TALK. If you're on Twitter, you can tweet us your question by writing the @ sign followed by scifri, and there's also more information on the website,, where you'll find links to our topic.

So Dr. Van Eenennaam, let's start with you. You were on this committee that was evaluating. How long did you have to evaluate the company's data saying this food is - this fish is safe to eat and safe to grow and release into the marketplace?

EENENNAAM: So the veterinary medicine advisory committee that looked at all that data had about two weeks prior to the meeting, which was held last September. It was actually released to the committee the same day it was made publicly available on the FDA's website.

PALCA: And what sorts of material were available to you?

EENENNAAM: It was about 172-page briefing package, which included information on the safety of the fish from a food consumption standpoint and also information about how the construct was created, whether or not the fish grew faster or not and information also included there on the environmental assessment that the company had done to look at the environmental concerns associated with the fish.

I guess one thing, just I know environmental concerns are always something that people are concerned about, this fish actually getting out into wild populations, and the way that the product was being regulated through the FDA, it was proposed, and it is proposed, that this particular fish is going to be raised in land-based tanks, so on land in tanks, and it would also be triploid female product, which means that all of the fish would be female so they couldn't interbreed with each other, and also triploid, and triploidy would result in those animals being sterile.

And so the company tried to have some risk mitigation in place to prevent any possibility of those fish interbreeding with wild populations.

PALCA: Right, and just finally on this meeting, and this meeting took place last September, 2010, the conclusion of the meeting was that this group said to the FDA we think that this is a product that you could approve for marketing.

EENENNAAM: Basically the idea of the veterinary medicine advisory committee was to look at all of the data that the FDA had looked at. The FDA's conclusion was there was a reasonable certainty of no harm from a food safety perspective, and also they looked at the environmental assessment. And basically, the committee was there to look and see if there was anything that they had overlooked.

And I think the conclusion of the committee was that they agreed with the FDA that there was a reasonable certainty of no harm from food safety perspective and also from the environmental assessment perspective.

PALCA: And that reasonable certainty of no harm is one of those terms of art that means, we don't think it's a problem.

EENENNAAM: Well, yeah.

PALCA: That's basically. I just wanted to make sure people understood that. OK, so you said things looked OK, but Dr. Kapuscinski, I mean obviously you and other scientists think either - well, tell me what you think. Did they not look at the data correctly, or did they not realize that more data was needed or were needed?

KAPUSCINSKI: Well, I think my main concern was that the kind of data presented had gaps, and the quality of the analysis of the data, especially the statistical analysis, was really quite a low bar. So my main concern is that this application is really setting a precedent, and it's actually an application for a fairly small grow-out facility to raise some of these salmon in an undisclosed location in Panama.

So this one facility doesn't really represent what's coming down the road. It's really more, you can think of it as sort of putting the camel's nose under the tent. But because it's going to set a precedent, it's really important that the quality of the science be as high as scientific standards would normally expect and that the risk assessment is complete.

So, I was concerned that there were some problems with small sample sizes, some problems with statistical analysis, and I was even more concerned that there were key parts missing from the risk assessment. It seemed like the approach taken, the risk assessment, wasn't really up to speed with the state of the art risk assessment.

So there were really three things that I thought were missing. One was the lack of what we would call a failure mode analysis, basically lack of a quantitative analysis of what would - what could go wrong in the multiple confinement system that the company proposed.

As Alison just explained, they have some biological confinement methods, which I think is a very good thing that they have that, and then they had also a number of physical confinement methods combined with that. The problem is, with confinement systems, there are always possibilities of things going wrong, and their interactions can be complex.

And it's now a pretty standard practice, when you're assessing a technology, that you do a failure analysis. So that was missing. The other part that was missing is, if some of the fish did escape, they didn't really go the next step of answering the question, well, what could happen if they did escape. And one might argue that there would be a very low number of fish escaping, but it doesn't obviate the need to still do that second step of the risk assessment.

And then finally, and in a way the most important, is the risk assessment lacked a formal uncertainty analysis. And this is really important, especially in environmental risk assessment, because there are always going to be scientific uncertainties.

Living organisms and ecosystems where these fish might end up, are very complex. There's a lot of variability. There are always things that the scientists don't fully understand. That doesn't need to paralyze, though, the risk-assessment process. Instead, the state of the art in risk assessment nowadays is that you carry out a formal uncertainty analysis throughout the risk assessment. And you gather the results of that at the end, and you make that part of the conclusions that you had to the decision-makers, so that the decision-makers are much better informed about, really, what are they accepting, and what assumptions are they making if they give an approval or if they give a denial.

So I think really in a sense I was worried that if this application is approved with these low standards of science and these missing parts of a risk assessment, and it sets the message, the precedent that this is what the U.S. government will expect.

And also, many other countries are actually watching to see what the U.S. government will do because there are other groups around the world developing transgenic fish. That would just be a really unfortunately way too low bar.

PALCA: I'm just curious, though, Alison Van Eenennaam said that the expectation was that these fish would all be - the ones that were released or the ones that would leave the breeding facility - were all female and were triploid, meaning they were sterile. Doesn't that mean that they - I mean, if they got out, nothing would happen, they'd live their lives, and then they'd die, and that would be the end of it?

KAPUSCINSKI: Well, it's - unfortunately, it's not quite that simple. There are two things that we still have to think about. The first one is that there is some low percentage of fish in which the triploid induction which makes them sterile doesn't always succeed. And again because of this one application being fairly small, that low number, you know, it might be anything as low as .1 percent to one percent of the fish not being totally sterile, that low percentage might not be a problem for this particular case.

But there should have been an actual failure analysis of that, if this is again going to set the model for what you do in the future. So that's the first problem because in future applications you might have much larger numbers of fish, total numbers escaping, especially if this approval ends up triggering a proliferation of genetically engineered salmon being taken up by the salmon farming industry, which is a global industry, a global commodity, then small percentages can add up when you're starting to have operations that raise anything from 500,000 to a million fish in a particular fish farm.

But then the second issue is if fish escape, and they're sterile, they don't die out immediately. They still live in the ecosystem and interact with other organisms. So again the question of scale becomes important. For this particular application, we were given really no information about the organisms living in the undisclosed - in the river in the undisclosed place in Panama. So it's really hard to know are there other fish in that ecosystem that any escapes could interact with?

If there are, are they fish that are endangered? We know that freshwater fish species throughout the world are in decline or in a lot of trouble because of other human impacts. So we need to have those kinds of questions at least addressed.

PALCA: I'm sorry, I don't mean to interrupt you, but we have to take a quick break, and I want to make sure that we let Dr. Van Eenennaam get some thoughts in, as well. But stay with us. We're going to go through all this, and hopefully we'll come out of it with some clearer idea, although it's a very complicated issue. So stick around. we'll be right back after a short break.




PALCA: This is SCIENCE FRIDAY. I'm Joe Palca. We're talking this hour about how a proposal to market genetically modified salmon is raising concerns. My guests are Dr. Alison Van Eenennaam, an extension specialist in animal genomics and biotechnology at the University of California, Davis; and Dr. Anne Kapuscinski, professor of sustainability science and chair of the Environmental Studies Program at Dartmouth College.

And I think I'd like to ask Dr. Van Eenennaam: You heard some of the concerns about missing data or inadequate date or safety data that she would like to see. How does all that strike you?

EENENNAAM: Well, a couple of points, I think. The FDA made it very clear that this particular approval was for the particular location in Panama that the company was proposing to grow these fish in, which again is an inland tank location.

And that's an FDA-inspected site, and the FDA has been there and seen that there's multiple levels of physical containment, things like nets and cages and things to stop any fish from escaping there, and also the physical location of where the actual site is has a number of geographical attributes that would prevent those fish from if fish did actually escape from that location from ever reaching the sea.

There's a number of thermal lethal rivers that the fish would have to get through if they ever were to escape. And so I guess we have to think about risk assessment in terms of there is a potential hazard, that is that the fish escaped, but there's a number of risk mitigation approaches that the company's put into place. And these are numerous and multiple and redundant.

So let's just talk about the triploid, which I agree is not 100 percent effective. So let's just say we've got .1 percent of the fish that are actually fertile, but then you've got all of these multiple physical and biological containment measures in place to try to reduce the risk of any fish ever escaping down to zero.

And I think you'll never say zero, but there's a number of multiple factors in place to try to minimize that risk as a result of all of these risk mitigation measures.

PALCA: All right, fair enough, but I want - now I want to include our listeners to this conversation because they, I'm sure, will have questions. So let's go first to Luke(ph) in Kansas City, Missouri. Luke, you're on SCIENCE FRIDAY, welcome.

LUKE: Hi, my question is that: Why is there so much attention surrounding the genetically modified salmon when we've been consuming, like, modified vegetables and other foods injected with who knows what for years now?

PALCA: Interesting question. Maybe Dr. Kapuscinski, you'd like to try that.

KAPUSCINSKI: Sure. I think that there's that much attention for two reasons. One is this will be the first genetically modified animal approved for widespread commercial production and human food. But secondly, this kind of animal, a fish especially, is not that removed from its wild relatives, and are - have much closer interactions with ecosystem than some of the vegetables that we grow.

Also, we're dealing with a species - Atlantic salmon - that many of the places where it's farmed are also the native range of wild Atlantic salmon, and those are - unfortunately, those wild Atlantic salmon populations are in deep trouble around the world.

So that's not directly the case for this particular application, as Alison pointed out, but I want to come back to my earlier point that we have to keep thinking about the broader context here. This application is setting the precedent for what would be expected of an applicant to show environmental safety to a reasonable degree in the future.

And if this application is approved, and if the salmon farming industry decides that this is a good product for their business, then it's going to be adopted and farmed in places where there may not be as good confinement and where if the fish escape in some of those places - like eastern Canada, the state of Maine, parts of Europe - where they can escape, interact with wild Atlantic salmon.

PALCA: Right.

KAPUSCINSKI: So, you know, vegetables don't move around as easily.

PALCA: Right, but there is also the case, at least as Dr. Van Eenennaam said, about they'll need their own applications, and someone will have to decide if their control measures are adequate. But let's go now to Roger(ph) in Commerce, Michigan. Roger, welcome to SCIENCE FRIDAY.

ROGER: Hey there.


ROGER: Yeah, my main concern was, when I hear about genetically modified foods, in general, I hear from a lot of people the complaint about not being safe to eat, which always drive me crazy. It's perfectly safe to eat. My only concern is like what people were talking about earlier, from the genetic diversity and getting involved with wild species.

But couldn't it be perfectly safe if it was just in a controlled environment, unlike the certain Asian carp where it wasn't in a proper place? If they were raise some place, you know, flood plains, or totally landlocked, away from the ocean. Give an extreme example like Nevada. We have no problem growing it over there and even encouraging it. And I'll take my comments off the air.

PALCA: Okay, thanks. So the question is, you know, absent the ecological questions, is this fish safe to eat? And are you satisfied, Dr. Kapuscinski, that that question has been answered?

KAPUSCINSKI: Well, the food safety area is not as much my area of expertise, but I did - I did attend the VMAC...

PALCA: That's that meeting, the veterinary...

KAPUSCINSKI: Yes, and I heard some of the other comments and read some of them and also actually read the report from the committee. And, you know, some of the concerns that were made that sort of resonated for me based on my having skimmed the food safety section of it was again concerns about small sample sizes and some problems with the statistical analysis.

So for example, one of the legitimate things to ask about is whether the engineering of these salmon has increased their allergenicity, and the data that was used to conclude the conclusions about that involved only six fish. And even the statistician on the veterinary medicine advisory committee commented that, you know, there could be some ways to improve the statistical analysis.

And there was a paragraph about that in the final report from the chair of that committee. So my concern there is also really about the quality of the science.

PALCA: Okay, Dr. Van Eenennaam, what about you at this point? Are you more or less satisfied that this is safe to eat, or do you think there's still more to find out?

EENENNAAM: I'm comfortable that it's safe to eat. I guess I will touch on the allergenicity question because that's always an issue with genetically engineered foods. And the concern is that the protein that's being introduced through the genetic engineering would create allergens.

For example, if you brought, I don't know, a peanut protein into a fish, it might result in an allergen. And that is not the case in the case of the growth hormone that's in the salmon. And so the allergenicity question got down to the question of whether or not people who were allergic to fish would be more likely to be allergic to this particular fish.

KAPUSCINSKI: And there's really no consensus in the scientific or medical communities regarding the magnitude of an increase in kind of the endogenous allergens, the fish allergens, if you will, That would pose an additional risk to public health. And I think in the absence of knowing what level you're looking for, it's difficult to know what would be the appropriate work to do.

EENENNAAM: We don't even know, really, the levels of allergens that are in naturally occurring salmon, and so in the absence of that information, it's difficult to know what levels would trigger a concern. And we didn't see - in that absence of that information, it's difficult to make a determination.

PALCA: Okay, let's take another call now and this time go to - let's see, how about Jerry(ph) in - no, Brian(ph) in Portland, Oregon. Brian, welcome to the program.

BRIAN: Yeah, I had a question about the failure analysis. I read an article about either this fish or a very similar fish, and the article I read is about 10 years ago, and it stated that if some of these fish that grow extra fast were to escape into the wild, within 50 years or something, it would supplant the wild fish because it grows so much faster.

And then they would destroy themselves because they would eat so much of the food so quickly, there would be a population collapse. And this is a catastrophic failure, and this is a computer analysis I heard about, again about 10 years ago. I wanted to know if your panel has heard (unintelligible).

PALCA: Fair enough, thanks, Brian. What about that, Dr. Kapuscinski, maybe you've heard of this?

KAPUSCINSKI: I suspect that the caller is referring to a study that described a theoretical Trojan gene effect. And in this case, I anticipate that Allison and I probably agree.

PALCA: Wait, excuse me, can you explain what a Trojan gene effect is?

KAPUSCINSKI: Sure, the idea is that this engineered gene would give a mating advantage to the fish, for example because larger salmon are maybe more successful at competing for a mate. So it would give them a mating advantage that would drive the engineered into a wild population.

But then there's some other aspect of the gene that causes reduced viability in the offspring, and so over generations, that drives the population to extinction. Now, the problem was that the original model that was used to come up with the theoretical prediction was really very simplistic.

And some geneticists questioned it at that time. Now, we actually have some additional studies that have tried to add some of the things that were left out of that model - for example, the possibility of a evolutionary process; the fact that environmental factors can actually influence the actual traits of a fish, the way the gene gets expressed in the final - things like the final size. So we now have pretty good evidence that all points in the same direction, that the Trojan gene effect is not very likely. However, I don't think that that means then that there's absolutely nothing to worry about. There still are other important ecological effects questions.

PALCA: OK. Dr. Van Eenennaam, do you have anything you'd like to add to that?

EENENNAAM: You know, I mean, I think there's some data that has come out that in this particular fish's case shows that there actually would have reduced reproductive performance, at least the males would, relative to controlled. That's the study that came out this year. But I guess my question is the relevance of that given the proposed containment that's associated with this particular application, and that these fish are not going to be interacting with the environment. The proposal is to have them on - in land-based facilities.

PALCA: OK. Fair enough. Let's now go to Jerry in Ehrhardt, South Carolina. Jerry, welcome to SCIENCE FRIDAY. You're on the air.

JERRY: Hi, good afternoon. If you would ask your guest to comment on the origin of the extreme resistance of genetic modification because it does seems to me that for a millennia we've been practicing husbandry with animals. We've been breeding them. We've been grafting plants. And the only objection to me seems to be the speed involved rather than the quality of the product because, ironically, the, excuse me, the European resistance is on the quality and is it safe, where for the layman standpoint it seems like the speed involved is the main thing and not the quality of the result because we've been doing this for generations, for millennia.

PALCA: Well, thank you for that, Jerry. So the question, basically, is genetic modification is the same as, I mean, the engineering it in the lab is just the faster way of doing the same thing that people have always done in terms of breeding crops that they want or fish that they want or anything that they want. Dr. Van Eenennaam, what about that?

EENENNAAM: Yeah, it's a very interesting question, the level of resistance to this. And I guess one of the things that I think is really important when we're talking about risks is having a look at what the current methods of producing food, in this case salmon, are. And we're comparing, I mean, all of the Atlantic salmon that's farmed is effectively imported into the U.S. And it's raised in net pens in countries like Scotland. And net-pen aquaculture of salmon has its own ecological concerns. There's some pollution concerns. There's some disease concerns. There's actually escapes from most net pens are of those diploid, fertile growth - animals that have been selected for growth just sort of natural breeding ways that are getting out into the wild.

And there's certainly some risks associated with that particular approach to aquaculture. And so I think when we're looking at the risks associated with genetic engineering, it's always important to have a look at the risks associated with the current approaches to raising fish. And this particular application, I might argue, is actually a more sustainable approach to raising salmon for aquaculture because you're taking the whole product on land and removing any risks associated with net-pen aquaculture and producing a more efficient, sustainable product for human consumption.

PALCA: We're talking about the science and safety of bioengineered fish. I'm Joe Palca. And this is SCIENCE FRIDAY from NPR. So, Dr. Kapuscinski, what about that? I mean, is there something special about genetically modified organisms that needs a higher level of scrutiny?

KAPUSCINSKI: Well, the actual - the reason why people are interested in doing genetic engineering is because it has new powers, and it allows you to actually introduce genes that are either were never in that animal or that are expressed at times that have never been expressed. So it really does have the potential to fundamentally change the biology of the whole organism. It doesn't mean it'll do that all the time, but that potential is there. And I think that's why there's interest in greater scrutiny. So an example with salmon is there's one study with genetically engineered Coho salmon that also had a growth hormone gene added to them that showed that their tolerance of warmer temperatures change, so that they actually grew faster at a warmer temperature than they would have at a colder temperature that's usually the optimal temperature for salmon.

And with selective breeding, traditional breeding, we hadn't yet seen that with salmon. So that's an example of something new that you can do. In a way, the dilemma is that the very power of genetic engineering that makes it exciting and potentially a very useful tool for some applications also raises some new questions about unexpected effects, which is I think why we're gaining more scrutiny.

PALCA: So - well, I'm just curious, Dr. Kapuscinski, I mean, in the end of the day, can you imagine a time when you will have enough information to feel comfortable about saying, this particular salmon or any genetically modified animal is safe to bring in to the marketplace?

KAPUSCINSKI: I think if the things that I was saying were missing were addressed, if there was a quantitative failure analysis, if some of the fairly important questions about ecological effects - if fish did escape - were answered, and if you did a good uncertainty analysis so that the overall quality of the signs and the completeness of the risk assessment were there, I would be much more comfortable with it. And the reason I keep pushing on this is I don't think it actually really makes sense to focus only on this one application, because it is such a small scale application. And we know that if it gets approved, it's then going open the door to much larger-scale use of this fish and many other fish farms. So if we, for example, come back to Dr. Van Eenennaam's idea that if you were to shift the farming of salmon from cages to inland facilities, that that would be more sustainable, what we have to ask is, first of all, I don't know if the salmon farming industry would really be willing to do that, but let's say they did. When you shift it to inland, although you are now reducing, greatly reducing the escapes compared to cages, you are going to have much higher energy use and some other uses that...

PALCA: I'm sorry. I don't like to cut you off, but...


PALCA: ...I'm afraid we've run out of time for this segment.


PALCA: It's one of those topics that's got people very interested and excited, but thank you, Dr. Kapuscinski, for joining us today. She's a professor of sustainable - sustainability science and chair of the Environmental Studies Program at Dartmouth. And thank you also to Dr. Alison Van Eenennaam, extension specialist in animal genomics and biotechnology at the University of California, Davis.

After the break, we'll talk to molecular biologist Leslie Leinwand and what she learned about her unusual choice from her - unusual choice for a lab animal: a python. Stay with us. This is SCIENCE FRIDAY from NPR.

Copyright © 2011 NPR. All rights reserved. Visit our website terms of use and permissions pages at for further information.

NPR transcripts are created on a rush deadline by an NPR contractor. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.