Nothing To Pout About: The Kosher Status of Genetically Modified Salmon

AquAdvantage® Salmon has DNA from a non-kosher fish. Is it fighting an upstream battle in the kosher market?

This special post is an iylui neshama (merit for the soul) of my dear mother, Mrs. Joan Maybruch, Yocheved Pesha bas Hillel a”h, on her yahrzeit (date of passing).

What’s Orange and Flavorful and Eaten All Over

The unique, tasty and familiar orange-hued meat of salmon has been an international staple for centuries. In contemporary Jewish cuisine, it is what puts lox on a bagel, puts pizzazz in a salmon roll, and takes the cake as a salmon steak.

Against this backdrop, a company headquartered outside of Boston, named AquaBounty Technologies, came up with a bold invention. It’s a small step for them, but a giant leap for fish production. They discovered a way to combine parts of the DNA of two other fish with a standard Atlantic salmon to make the Atlantic salmon grow at approximately twice the normal rate. Since one of the greatest expenditures salmon producers face is feeding the fish, the rapid growth provides significant savings on feed. Approximately two decades ago, AquaBounty began the process of applying to the Federal Drug Administration for approval of their genetically modified salmon, AquAdvantage. It was finally approved last year.

AquAdvantage salmon (AAS) raises potential kashrus concerns. The genetic material inserted into the Atlantic salmon comes from two fish, the related Pacific Chinook salmon – which is kosher, and the ocean pout – an eel like fish that is not kosher. Does using material extracted from a non-kosher fish render the modified AquAdvantage salmon as non-kosher? A closer examination of scientific and halachic background provides bountiful evidence to suggest that the new salmon has no kashrus concerns at all.

Grafting Two Animals Together

The process of making the new salmon is complex, and will be discussed later. But let’s imagine that the way that they created this innovative fish was by combining some actual material from the ocean pout into the Atlantic salmon. For example, suppose that they actually took some of the ocean pout’s meat and grafted it onto a salmon. Would that provide kashrus concerns on the new fish? Could the original meat of the salmon be consumed? How about the actual ocean pout meat that takes to the salmon and grows as part of it? If both of those would be permissible, there would be strong reason to maintain that the scientific transplanting of a genetic part of the pout into salmon should certainly be OK. Based on similar discussions in contemporary authorities regarding plant and human transplants, one could cogently argue that theoretical animal splicing would yield permissible food.

In 1958, Rabbi Moshe Feinstein (Igros Moshe Y.D. vol. 1, 230/ 8) was asked by Rabbi Isaac Liebes about a conundrum faced by Kohanim. Kohanim are prohibited from contact with a corpse or organ from one because those are ritually impure. Yet, skin grafts or transplants sometimes come from cadavers. May a skin graft or transplant be performed on a Kohen? In a lengthy responsum, Rabbi Feinstein permits the implant. One of the points Rabbi Feinstein raises is based on the Talmud’s ruling removing tumah (ritual impurity) from the skin of a cadaver that one designates for another purpose. Although it is considered inappropriate to use skin as leather, there are tribes who do so. In a case where one begins to tan the skin of a cadaver, it is no longer considered “skin” and is not off limits to be handled by a Kohen. Rabbi Feinstein extends that to a transplant. When the cadaverous skin is readied or attached to the Kohen, it ceases to be considered part of a corpse. Rather, it takes on a new identity once it is designated to be part of the recipient, and is no longer ritually impure.

Likewise, if part of a pig is grafted onto a cow, it should become part of the cow as soon as it is grafted. Logic may dictate the necessity to wait until the graft takes hold and becomes an integral part of the cow for the grafted pig meat to be considered kosher, but it is kosher. Similarly, if one would graft part of an ocean pout onto a salmon, the ocean pout section and, certainly the rest of the salmon, should be kosher. There is a caveat. Rabbi Feinstein is discussing ritual impurity and our case is regarding the permissibility of eating, which is a different area of halacha. It might be difficult to compare the two.

Another discussion that might shed light on the status of non-kosher meat grafted onto a kosher animal is posed by several authorities in the past century, and referred to by Rabbi Eliezer Yehudah Waldenberg (Tzitz Eliezer vol. 10, 25/26). If one transplants ovaries from one woman to another, who is considered the mother of children then born? Furthermore, if one transplants a heart, does the recipient now acquire a new identity? Rabbi Waldenberg posits definitively that a transplant is considered part of the recipient in all respects. A similar assertion was made by Rabbi Yekusiel Kamelhar in his Hatalmud Umadaei Hatevel, discussed by Rabbi J. David Bleich in Volume 1 of Contemporary Halakhic Problems.

Similarly, one could further their points and maintain that transplanting from one animal to another is no different. Once an organ or flesh from one animal is transplanted to another, it may be considered part of the recipient, and completely kosher. As with the responsum of Rabbi Feinstein, one could also differentiate between people and animals, and kashrus and other areas of Jewish law. Yet, it is an interesting and significant claim to consider.

A related point might be gleaned from a Talmudic ruling regarding orlah (the prohibition of eating or deriving benefit from a new tree for its first three years). The Talmud (Sotah 43b) rules that if one grafts a branch from a new tree, which is currently non-kosher, onto an established tree that is older than three years old, the sapling joins the older tree and no longer has the status of orlah. Similarly, it would stand to reason that if one would add some non-kosher animal to a kosher animal, the result should be a completely kosher animal.

Too Small, That’s All

The previous point deals with grafting and splicing on a macro level. Genetic engineering is done on the microscopic level, which creates several more reasons to be lenient. The most obvious one would seem to be that since Halacha only prohibits things that one can see with the naked eye, the microscopic amount of fish extracted is halachically negligible. However, that leniency doesn’t work. Once something is not kosher in large quantities, one can’t take a minute amount and then ingest it. For example, a person cannot take a really small amount of pork and eat it intentionally. Since it comes from a larger amount that was prohibited, reducing its size doesn’t remove the prohibition. Once it’s forbidden, it’s forbidden, regardless of the size or amount. In contrast, an entity that is entirely microscopic, such as bugs or bacteria that one cannot see without magnification were never prohibited in the first place. Therefore, one need not examine food because of a concern about eating them.

A further point as to why the ocean pout extract is not considered too small is made by Rabbi Shlomo Zalman Auerbach (Minchas Shlomo Tinyana 100, 7). There is a prohibition from grafting two plant species together. Rabbi Auerbach was asked if genetic recombination of the DNA of two species is also restricted, since it is microscopic. Rabbi Auerbach matter-of-factly prohibits it. He explains that genetic engineers purposely maneuver and recombine DNA, so the DNA fragments are considered halachically significant and “visible,” even though they are microscopic.

Nevertheless, the small nature of DNA might still provide a way to permit ocean pout in a salmon, similar to the splicing of two animals that we discussed above. The Talmud discusses (Menachos 69a) that a digested item sometimes loses its independent status and becomes part of the animal that digested it. Tosafos (s.v. dibala) there quotes several opinions regarding how much digestion is necessary and which area or areas of halacha the concept of digestion is relevant to. Yet, it is possible that all opinions in Tosafos would agree that if food is completely digested by the animal and assimilated into its body, the food is permissible. Similarly, since the DNA of the ocean pout is completely absorbed into the host salmon, and might be considered as if it completely digested, posing no halachic problems altogether.

The Culture of Cloning

As we examine the process of creating AAS using genetic engineering, it becomes more evident that potential kashrus problems disappear. To create the part of the DNA that is given to the salmon, DNA is removed from each of the other two fish. Then that DNA has to be copied, or cloned, by being mixed with a medium, in this case bacteria.  These bacteria clone multiple copies of the DNA and create a pool from which to select a DNA sample that best expresses the desired gene or gene fragment. The fact that the DNA is extracted from the ocean pout and then put into bacteria provides us with another potential reason to be lenient. The tiny bit of ocean pout derivative is batel (rendered legally insignificant) in the large volume of the bacterial cloning medium to which it is inserted. The halachically required ratio necessary to permit a non-kosher ingredient that fell into a kosher one is 1:60. Here, presumably, the ratio is much larger, resulting in copious bitul (nullification).

Yet, there is a factor that might undermine halachic nullification here, the rule of davar hama’amid – something that creates substance or volume. An ingredient that is responsible for a significant material change in a mixture is not nullified since its effects are still tangible. An example is rennet, the natural product (containing the enzyme renin) that is used to turn milk into cheese. Rennet is produced in the stomach of ruminating animals, so it used to be common to insert a bit of stomach meat into a large volume of milk to create cheese. The ratio of meat to the milk was less than 1:60, nevertheless the halacha is that if the meat comes from an animal that was not properly shechted (ritually slaughtered), the cheese is prohibited (Shulchan Aruch Y.D. 87, 11). The meat is the davar hama’amid – the ingredient that catalyzes the cheese to be in its current state. Similarly, when a bit of ocean pout DNA is inserted into a bacterial cloning medium, the nature of the entire mixture is to clone ocean pout DNA. It is arguable that, in accordance with Rabbi Shlomo Zalman Auerbach’s approach that even small amounts of DNA are scientifically and halachically significant, the ocean pout derivative it is not batel (nullified) in the mixture, even though the amount inserted is negligible.

A related assertion has come up in several volumes (34 and 36) of the recent Journal Techumin (published by Machon Tzomet in Alon Shvut, Gush Etzion) regarding laboratory grown meat. Lab grown meat is created from real meat stem cells that are then grown and multiplied in a culture. Some authors have suggested that the meat is nullified in the large amount of culture. Others have retorted that since the goal is to replicate meat, the few stem cells are the davar hama’amid, and would not be nullified.

Even if one would side with the stricter view in the meat dispute, there might still be room to say that the ocean pout DNA is nullified in its medium. In the meat culture, the result is meat, which comes from the stem cells. In contrast, in the bacterial medium, the result is still millions of bacteria. It is just enlightened scientists that know that those bacteria contain replicated DNA.


The next stage of the process seems to leave room for nullification without a doubt. The new DNA sequence is created from the combination of the kosher Chinook salmon and the ocean pout. This is termed recombinant DNA, since it combines different genes for a new DNA that does not exist in nature. The ocean pout is a small amount of the entire rDNA or transgene, and an even smaller amount of the entire DNA of the salmon, and an insignificant amount once it is in the fish itself. The ocean pout genetic material is batel (nullified) in the rest of the rDNA, and certainly in the entire strand of DNA, and even more certainly in the entire fish.

Here, the idea of davar hama’amid would not seem to pose a problem. The ocean pout genetic material is not a davar hama’amid. You might wonder why they need to take genetic material from two fish to make the Atlantic salmon grow quicker. The two materials serve different purposes. The Chinook gene codes for the growth hormone which helps salmon grow. The ocean pout genetic material is a promoter sequence – a fragment of DNA that signals to the gene when to be active and when not to be, like a switch that indicates to the gene when to be on and when to be off. Salmon usually stop and start growing during periods of their life, which is why it takes them approximately three years to reach maturity. On the other hand, the promoter sequence from the ocean pout constantly signals the fish to grow, without stopping. The resulting fish reach their mature size much quicker. The ocean pout genetic material merely catalyzes the growth hormone to stay on; it does not cause the growth. Therefore, it does not directly cause any volume or character, so it is not a davar hama’amid, and would be batel.

A Fishy Situation

As we further explore the process of creating AquAdvantage salmon, we will see several more reasons to assuage any kashrus concerns. AquaBounty reports that their current salmon has no ocean pout derivatives. Rather, ocean pout DNA is only used to engineer a salmon founder fish to create a lineage. They began with salmon eggs that had their recombinant DNA transgene injected into it. The fish that hatched from those eggs now had the transgene. The first fish to mature was then bred and crossbred to produce multiple generations of fish that carry the transgene. These offspring, several generations removed from the founder fish, are the ones that are used to breed AAS. Stopping at this point in the story already seems to give clear thumbs up for their kashrus. No commercially marketed AAS has a trace of non-kosher ocean pout in it. All they have are salmon, salmon, salmon. While part of their transgene is from a non-kosher pout,  in the next generations, it is only copied genetic material and there is no ocean pout bodily fluid present.

In a similar situation, Unilever was approved by the FDA to use a derivative of ocean pout blood for ice cream. Ocean pouts have a unique ability to descend to the depths of the ocean into cold water without their blood freezing. Unilever plans to use that antifreeze protein to reduce ice crystals in ice cream. They extracted it from an ocean pout, but grow it through yeast that are genetically modified to produce it. The FDA told Unilever that they do not need to specially label products using that protein as a fish allergen, since it is far removed from fish. Similarly, the genetic material in AAS (a fragment of the same gene that Unilever is using) is just as distant from a real ocean pout.

The ideas of nullification and founder fish also seem to solve an issue raised by several contemporary authorities. The foremost Ashkenazic codifier, Rav Moshe Isserles, (Rema Y. D. 81, 7) quotes a tradition that even though it is permissible for a Jewish child to nurse from a woman who eats non-kosher, the custom is not to do so. He explains that ingesting nourishment indirectly produced from non-kosher substances can have a spiritually negative influence, termed timtum halev (dulling of the heart). It is evident that a spiritually negative influence is possible by ingesting certain foods derived from a non-kosher substance even if eating them is permissible. Therefore, some contemporary authors, such as Rabbi Shlomo Revach (Tenuvos Hasadeh vol. 102, pp. 20 – 25) and Rabbi J. David Bleich, (Genetic Engineering, 2003, pp. 79 – 80) discuss the possibility that negative influence could come from AAS since it does have ocean pout derivative.

It is difficult to make absolute statements when dealing with the esoteric concept of timtum halev. At the same time, it is hard to imagine that a mixture that is permissible based on the rule of nullification would spiritually dull one’s heart. If so, one should refrain from any mixture that a small amount of prohibited substance erroneously entered, and this is not the normative approach in halacha. Furthermore, in the specific case of AAS, there is no actual ocean pout derivative in the fish one eats, so negative spiritual influence is even more unlikely.

Sunny Side Up

Furthermore, AquaBounty explains that the fish they use are not even results of the crossbreeding. Rather, they take eggs from standard, nontransgenic (with no genetic engineering) female Atlantic broodstock (fish used especially for producing other fish) and then add milt (reproductive excretion) from male salmon that have the transgene. They mix the two in their factory to produce salmon that have the transgene.  The Mishna in Bechoros (5b), codified in Shulchan Aruch (Y. D. 79, 2), states that the kashrus of the female parent of an animal is the determining factor in the kashrus of its offspring. Rabbi Nasan Gestetner (Lehoros Nasan 6, 57) explains that the consensus is that the rule applies to mothers who lay eggs, in addition to those that give birth to live offspring. In this case, AAS salmon have perfectly kosher mothers. Like mother, like daughter, the resulting AAS offspring should be kosher, too.

A Finny Finish

Another reason to be lenient is highlighted by many, including Rabbi Menachem Genack of the Orthodox Union, when he was interviewed by the New York Times (not to be confused with last month’s New York Times article about another kosher fish topic, the exponential growth and popularity of sushi in Boro Park and in kosher restaurants!). Rabbi Genack explained that if AAS has fins and scales, it should be kosher. This can be buttressed by an intriguing oral ruling issued by Rabbi Shlomo Zalman Auerbach, published posthumously (Halichos Shlomo, Moadim, Arbaas Haminim, p. 189). On Sukkos, the most famous of the four species that is required is the esrog, a specific subspecies of a citrus fruit called citron. A grafted esrog is not suitable. Rabbi Auerbach was asked about the suitability of a genetically engineered esrog. He responded that if the esrog looks different than the standard subspecies looks, it is unacceptable for the mitzvah. (The editor of that book elaborates that the ruling of Rabbi Auerbach is in accordance with his aforementioned written ruling that genetic manipulation is halachically significant, even though it is microscopic.) Here, the opposite is true. The genetically engineered salmon looks just like all other Atlantic salmon, so it should be kosher.

Interestingly, there are others that might argue on Rabbi Auerbach’s ruling regarding the unsuitability of the esrog. Rabbi Moshe Shternbuch (quoted by Rabbi J. David Bleich, ibid., p. 73 ) quotes the Talmudic ruling discussed earlier regarding Orlah. Just like a young sapling grafted onto an older branch becomes part of the older tree, a spliced gene becomes part of the host DNA. If so, it would stand to reason that genetic modification could result in a different looking fruit that is still an esrog, since an esrog was the host.

Fishing It Out

In conclusion, we have suggested several reasons that AquAdvantage salmon does not have any barriers from being kosher.

  • Even grafting actual parts from a non-kosher animal onto a kosher animal would create a fully kosher animal
  • Ocean pout substance is absorbed into the salmon, like digestion
  • Ocean pout substance is a negligible ratio in the bacterial medium
  • Even if it is not negligible in the cloning culture, it is negligible once it becomes part of the new (recombinant) DNA
  • It is negligible once it is in the entire salmon
  • AAS come from kosher salmon eggs
  • AAS’s fathers were already the product of crossbreeding, so there is no ocean pout in their bodies
  • AAS have fins and scales

Even if some of the reasons might be discounted, the nature of the process of creating AAS makes it so removed from any actual ocean pout that it seems evident that it is kosher. This happens to be good news. Aside from permitting us to taste AquAdvantage salmon, it also solves another problem. AquAdvantage salmon is considered so genetically harmless that the FDA ruled that it does not need to be labelled as genetically modified. Which means that your next bagel and lox might just have an AquAdvantage over any other one you’ve had. Enjoy it!

4 thoughts on “Nothing To Pout About: The Kosher Status of Genetically Modified Salmon”

  1. It seems to me that one might also consider that what’s being taken is not even a cell, but rather a sequence of instructions.
    I therefore wonder whether one need to even invoke bitul.
    It’s not that DNA is so small, but that there is nothing inherently forbidden in it that would even require bitul.

    I would also point out that in terms of its substance, all DNA is the same, in that all strings contain the same letters, the only difference being the sequence.


    1. Your point is very well taken. It would seem to be that even though we can whittle down DNA to only being AGCT or a double helix, when the DNA comes from a non-kosher source it is going to have the non-kosher status of its origin. A fitting parable is that the Torah mandates wearing techeiles – wool dyed with a bluish purple dye – on tzitizs (the fringes of a four-cornered garment). Techeiles must be extracted from a sea creature called the chilazon even though the same dye could be made from plants called k’la ilan. On a molecular and chemical level, the two might be the same. Yet, the Torah demands techeiles from a specific source anyway. It is not exactly comparable (although that was also a sea creature, like salmon!), but the parallel still illustrates the point. All the best! SM


      1. Hi. Just saw this again, I do not believe that k’la ilan and murex truncullus are “the same” on a molecular level — only certain aspects of the COLOR aspect of the dye are. For instance, the murex includes bromide, which is only found in the ocean. The murex dye is more colorfast. If the two were actually the same, on a molecular level, then royalty would never have bothered with murex in the first place. And perhaps the Torah’s specification that it comes from a sea creature was for the same reason — that plant indigo molecule does not bind as well. (Writing this as a non-chemist and from long-ago memory.)


  2. Dear Rabbi Maybruch,

    Thank you for the excellent article. Mr. Hodja makes an important point, but I think it needs to be better formulated. No cellular extract from an eel was ever actually injected into the salmon egg that gave rise to the founder salmon. Rather, a COPY of the regulatory DNA from the eel pout was used in the rDNA construct that was injected into the salmon eggs. It really is only the sequence of instructions, as Mr. Hodja puts it, that is being evaluated.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: