Getting to the Bottom of Subway’s Tuna Sandwich

Back in January, you may have noticed articles floating around the web with headlines like “Subway’s Tuna Is Not Tuna” or something else along those lines. These all stemmed from the initial filing of a lawsuit against Subway in federal court with a shocking claim: Subway tuna sandwiches allegedly “lack tuna and are completely bereft of tuna as an ingredient.” The story died down for some time as the plaintiffs waited for their day in court, then blew up again in June after The New York Times sent samples of Subway tuna to a lab for genetic testing and found that they contained no identifiable tuna DNA. The article itself was cautious in drawing conclusions, noting that such a result could have simply been due to the high level of processing inherent in canned tuna (nevertheless, it caused another round of clickbait to ricochet around the internet). There’s a lot more to the story, however. As I discovered while writing this, the rabbit hole goes pretty deep. Before we get to that, though, let’s first dive into those lab results and how they work.

The Times tested their tuna samples with a polymerase chain reaction test, better known as a PCR test. If that sounds familiar, it’s because it’s the same sort of test that’s used to test people for COVID-19. (There are other ways to detect COVID, but that’s for another article.) The process was almost identical to Swarthmore’s COVID testing program — they took a sample of tuna, dropped it in a tube, and shipped it to a lab where it was tested for DNA matching one of five different tuna species.

The “polymerase” in “polymerase chain reaction” refers to DNA polymerase. You might remember from high school biology that DNA polymerase is the enzyme that creates copies of DNA during cell division. In a PCR test, a DNA polymerase enzyme is mixed with the collected sample (tuna, in this case) in a soup of building blocks and primers. The primers are short sequences of DNA that find the start and end of a specific section of DNA that is being tested for. For this reason, they’re specific to the test being performed (a COVID test would have different primers than a test for tuna DNA), and poorly chosen primers can result in unreliable results. 

The sample is then heated to a specific temperature to separate the strands of DNA from each other, cooled down to allow the primers to attach, and then heated once more to activate the DNA polymerase. When activated, the polymerase takes the original DNA segment and creates a copy of it between the attachment points of the two primers (if no DNA matches the primers, then the polymerase can’t attach, and no replication occurs). This cycle is repeated 30 to 40 times, each time doubling the amount of DNA in the solution. Eventually, if even a tiny amount of DNA was present in the original sample, it will have been amplified exponentially (a single DNA strand duplicated 30 times results in about one billion copies) and is easily detected and analyzed using other methods. 

PCR tests are extremely good at their job, when used properly. The PCR tests commonly used for COVID can pick up incredibly minute traces of virus and multiply them billions of times. That assumes, however, that there’s viable genetic material in the first place, and that primers have been properly designed to attach to it. This isn’t particularly hard for a COVID swab that’s taken from your nose, dropped into a test tube, and tested the next day. Testing for something like tuna is a different matter — especially when the tuna in question has been caught, frozen, thawed, chopped up, cooked, sterilized during canning, and left to sit in brine for a couple of months, before being mixed with mayonnaise and spread on a bread roll. It stands to reason that this sort of treatment would cause significant degradation of DNA. In fact, a study released shortly before this whole controversy showed just that: standard processing steps for canned tuna caused significant errors to show up in DNA test results, particularly in the case of brine canning like that used by Subway. By the time the tuna had been canned in brine, one of the two DNA fragments the researchers were looking for was impossible to detect, and the other one was so degraded that the correct tuna species was identified only 50% of the time.

In such difficult conditions, it’s not particularly surprising that the PCR test commissioned by the Times failed to pick anything up. When DNA degrades it breaks up into small pieces, and since primers must attach to both ends of a given DNA sequence, a break anywhere in the sequence makes the DNA unrecognizable. The ability of a PCR test to amplify what it’s supposed to amplify is very dependent on the choice of primers and targets, which in cases like these can be highly difficult — shorter sequences pick up spurious signals, longer ones break up before you can test them. That said, it’s still possible — just maybe not with the lab The New York Times used. In a much-less-covered story, the TV show Inside Edition ran similar tests on 50 pounds of Subway tuna using a lab that specializes in highly-processed tuna. Tuna, and specifically skipjack or yellowfin tuna, was confirmed to be present in all the samples.

When I started writing this article, I thought it would end here. I thought the lawsuit was just a misguided but well-intentioned effort; media outlets misrepresented the significance of the test results, but that’s par for the course when it comes to science journalism. In fact, Subway has some history with this sort of thing: a court case in Ireland last year found that Subway bread isn’t legally bread. (It’s more boring than it actually sounds: the issue was that the high amount of sugar in the bread disqualified it from tax-exempt status as a staple food.) Then I decided to dig into the court filings of the lawsuit in question, and that’s where things got a bit weirder.

While the lawsuit’s initial complaint (the document where you explain who you’re suing and why) contained plenty of claims that “the [sandwich] filling … has no scintilla of tuna at all” and that it “entirely lack[s] any trace of tuna,” the plaintiffs didn’t actually provide any concrete evidence other than vaguely mentioning that “independent testing has repeatedly affirmed” said lack of tuna. Remember, the Times analysis wouldn’t be published until six months after this lawsuit was filed. Then it appears that lawyers on both sides went through a couple months of lawyerly meetings before filing a joint case management statement late last May. In the statement, Subway’s lawyers claim to have provided “information” (without further detail) to the plaintiffs that proves that their sandwiches contain tuna. Meanwhile, the plaintiffs switch slightly from the whole “completely bereft of tuna as an ingredient” thing to a milder accusation that Subway “mislabeled the products as ‘100% wild caught’ ‘skipjack and yellowfin tuna,’ even though this is not the case,” and said that they would file an amended complaint to “clarify their allegations.”

A month later, the plaintiffs filed a new, amended complaint with a significant change of tune. All allegations that the tuna sandwiches did not contain tuna had been removed, replaced by a considerably tamer claim: the sandwiches “do not contain 100% skipjack and yellowtail tuna, and/or do not consist of 100% tuna with respect to the fish portion of the product Defendants represent as tuna.” Still, though, no evidence was provided for the claim, and certainly nothing in the way of test results. This may be compared with this other lawsuit against Subway claiming that some of their Footlong sandwiches were, in fact, slightly less than a foot long. The complaint provides several specific examples with names of customers, dates, and locations, and includes photographs of the undersized sandwiches that were received. None of this is present in either tuna complaint.

So … do things seem a bit fishy here to anyone else? The plaintiffs file a lawsuit with a shocking, yet believable, claim with no evidence provided to the court. Then they walk it back just as quickly, well before the claim ever reaches a courtroom and indeed before Subway even filed a response! It certainly doesn’t seem like the kind of thing the plaintiffs would do if they had evidence that “independent testing has repeatedly affirmed” that Subway tuna sandwiches don’t contain tuna. As Subway’s lawyers pointed out in their motion to dismiss the case, it does seem consistent with an attempt to get a quick settlement out of Subway to make the accusations and media coverage go away.

Now, I’m not a lawyer, and I’m not going to make any statements with complete confidence until the case is concluded (a hearing is set for October 7). Maybe the plaintiffs have some bombshell evidence that they’ll pull out of nowhere and have been making all their court filings as suspicious as possible for unrelated reasons. The lawsuit as it stands, however, certainly isn’t grounds to believe that Subway tuna isn’t what Subway says it is. I have no real stake in this — I don’t even like tuna sandwiches! But if you do, and you find Subway to be a good way of satisfying your tuna needs, then I encourage you to go forth and consume them without elevated concern for their contents. As a general rule: if you see another story about a shocking lawsuit filed against some company for a similar claim, you should treat it with a healthy dose of skepticism.

Special thanks to Professor Dawn Carone for reviewing the biology portions of this article.

If you have any further questions, would like to see a column on a specific topic, or think that I got something wrong, feel free to email me at zrobins2@swarthmore.edu.