is it true that selenium is an antidote to mercury and fish that have a greater quantity of selenium than mercury are safe to eat in large quantities?
There is a real scientific basis for the idea, but the stronger claim — “if selenium exceeds mercury, the fish is safe to eat in unlimited or very large quantities” — is not established and is not the FDA/EPA position.
Selenium does appear to reduce mercury toxicity. Mercury binds very strongly to selenium, and methylmercury can interfere with selenium-dependent enzymes that protect cells, particularly in the nervous system. Adequate selenium may therefore mitigate some of mercury's toxic effects. This mercury–selenium interaction is well documented experimentally.
There is even a proposed way of evaluating seafood called the selenium-to-mercury molar ratio (Se:Hg ratio) or, in a more developed form, the Selenium Health Benefit Value (HBV-Se). Under this theory, fish containing substantially more selenium molecules than mercury molecules should present less mercury risk than you would predict from mercury concentration alone.
The important catch is that a Se:Hg ratio greater than 1 is not a proven safety threshold. A detailed review specifically examined this question and concluded that treating 1:1 as a bright dividing line between "safe" and "unsafe" is misleading. The authors concluded that selenium probably provides some protection, especially when present in substantial molar excess, but that people who eat fish frequently should still pay attention to the actual mercury concentration. Another review by U.S. Geological Survey-affiliated researchers found significant uncertainties and concluded there wasn't enough evidence to revise fish-consumption advisories based on Se:Hg ratios.
So I would describe selenium as a protective factor, rather than an "antidote." And FDA/EPA do not currently say that high-selenium fish can be eaten without regard to their mercury content. Their advice continues to classify fish primarily according to mercury exposure and recommends choosing lower-mercury species.
This becomes quite interesting for your earlier question about StarKist chunk light tuna. Chunk light tuna is generally made primarily from skipjack, and skipjack tends to have considerably more selenium than mercury on a molar basis. One published analysis actually found positive selenium health-benefit values for skipjack tuna.
If you'd like, I can take your 3 cans of StarKist chunk light tuna per day example and calculate both the estimated mercury dose and selenium dose, including the Se:Hg molar ratio, and compare that with the conventional EPA mercury limit. That would give us a much better picture of how much difference the selenium argument actually makes.
Yes. Using the 5-ounce StarKist Chunk Light Tuna in Water cans, the selenium-to-mercury ratio looks quite favorable — but the absolute mercury intake from 3 cans every day would still be high by EPA standards.
StarKist says a 5-oz can contains 113 g drained tuna and 140% of the Daily Value for selenium. Since the U.S. Daily Value for selenium is 55 µg, that works out to about 77 µg selenium per can, or 231 µg/day from 3 cans.
For mercury, FDA's large dataset for canned light tuna found an average concentration of 0.126 ppm, with a median of 0.077 ppm. Using the average:
| Per can | 3 cans/day | |
|---|---|---|
| Drained tuna | 113 g | 339 g |
| Mercury at 0.126 ppm | 14.2 µg | 42.7 µg |
| Selenium | ~77 µg | ~231 µg |
The interesting part is the molar ratio. You can't simply compare 231 µg selenium with 42.7 µg mercury because selenium and mercury atoms have different atomic weights. Converting them to moles gives approximately:
Selenium: 2.93 micromoles/day
Mercury: 0.213 micromoles/day
So the estimated selenium-to-mercury molar ratio is about 13.8 : 1.
That means there are roughly 14 selenium atoms available for every mercury atom, at least based on the total amounts in the food. From the selenium-mercury hypothesis, that is a very favorable ratio. It is nowhere near the borderline 1:1 situation.
However, here is why I would not conclude that three cans per day are therefore mercury-safe. Your estimated mercury intake would be 42.7 µg/day. EPA's methylmercury reference dose remains 0.1 µg/kg of body weight per day.
That means a person would have to weigh about:
42.7 ÷ 0.1 = 427 kg, or about 941 lb
for 42.7 µg/day to equal the EPA reference dose. For an ordinary-sized adult, three cans daily therefore exceeds the EPA RfD by several-fold.
There is also an interesting selenium issue going the opposite direction. Three cans give about 231 µg selenium/day. That's below the U.S. adult tolerable upper intake level of 400 µg/day, but it is already more than four times the 55-µg Daily Value. Other foods would add additional selenium. NIH notes that chronic selenium excess can eventually cause selenosis, including hair/nail changes and neurological or gastrointestinal symptoms.
What I think the selenium calculation actually tells us
The result does make three cans of light tuna look less concerning than the mercury number alone suggests. A roughly 14:1 Se:Hg molar ratio is substantial, and there is plausible biology behind selenium mitigating methylmercury toxicity.
But we currently do not have enough human evidence to say that a 14:1 ratio cancels out 42.7 µg/day of mercury. EPA has not incorporated selenium into its mercury RfD, so scientifically we can't turn that 14:1 ratio into something like "your effective mercury exposure is only 3 µg."
One other important point: 0.126 ppm is an average, not what's necessarily in your particular StarKist cans. FDA's 545 canned-light-tuna samples ranged from below detection to 0.889 ppm, so there is considerable variability.
