Selenium: The Nutrient We Discovered Backwards

Selenium entered nutrition as a poison. Ranchers watched horses and cattle lose hair, develop damaged hooves, waste away, and sometimes die after grazing in particular parts of the American Great Plains. By the 1930s, researchers had traced the problem through plants and soil to selenium.

For the next two decades, selenium was mainly something agricultural scientists tried to keep out of animals. Then researchers discovered that animals also became sick when selenium was missing. Later still, severe selenium deficiency would be implicated in a deadly human heart disease in parts of China.

The same element that poisoned livestock in North America turned out to be something animals and humans require in tiny amounts. Science met the toxicity long before it understood the requirement.

Named for the moon

Jöns Jacob Berzelius discovered selenium in 1817 while investigating material from a sulfuric-acid works in Sweden. He initially suspected tellurium. When the substance proved to be a new element, he named it after Selene, the Greek goddess of the moon, as a companion to tellurium, whose name came from the Earth.

Two centuries later, one of the first clinical signs of excessive selenium intake is a garlic odor on the breath, alongside a metallic taste in the mouth. Excess selenium is metabolized into volatile selenium compounds that leave through the lungs.

The animals got there first

Livestock disease on selenium-rich land was described long before anyone knew selenium was responsible. Nineteenth-century reports from the upper Missouri region described horses with hair loss, lameness, damaged hooves, and severe wasting after grazing particular areas. The chronic syndrome became known as alkali disease.

By the early twentieth century, investigators knew the problem followed the vegetation. Grain or forage grown on certain soils could reproduce the disease when fed elsewhere. In the 1930s, researchers traced the toxicity to selenium accumulated by plants growing on selenium-rich soils and geological formations.

Some plants are unusually effective accumulators. Certain species of Astragalus can concentrate selenium to levels dangerous to grazing animals, and ordinary crops can become high in selenium under the right soil conditions.

Factor 3

The reversal came in 1957. Klaus Schwarz and Calvin Foltz were studying severe liver damage in rats fed particular purified diets. Some ordinary foods contained an unknown protective substance that prevented the damage. The researchers called it Factor 3.

It turned out to be selenium.

Selenium was no longer merely a toxic contaminant. At a sufficiently low intake, it was an essential nutrient. Within a few years, deficiency syndromes were recognized in poultry, sheep, cattle, and other animals, including forms of muscle degeneration farmers had already been fighting. The mineral agricultural agencies had spent years warning producers about became something livestock feed sometimes needed to contain.

Then selenium got a job description

In 1973, researchers showed that selenium is a component of glutathione peroxidase, an enzyme involved in controlling oxidative damage. That changed the argument again. Selenium was not merely associated with better health in deficient animals. It occupied a defined position in an enzyme.

Glutathione peroxidase turned out to be one example among many. Humans have 25 known selenoproteins, including thioredoxin reductases, glutathione peroxidases, and selenoprotein P. They participate in antioxidant systems, thyroid-hormone metabolism, DNA synthesis, reproduction, and redox control.

While that was being worked out, people were dying in China

Keshan disease is an endemic cardiomyopathy first identified in 1935, in parts of China where the soil is low in selenium. The heart muscle enlarged and weakened, and acute cases could progress rapidly to heart failure and death. It fell most heavily on preschool children and women of childbearing age.

Its cause was not obvious. Researchers investigated infection, toxins, and general malnutrition.

Geography became the strongest clue. The disease occurred in a broad belt overlapping regions with extremely low selenium in soils, crops, and human tissues. Adults in affected areas had average selenium intakes of no more than 10 micrograms per day; roughly 20 micrograms per day appears to be protective.

Chinese researchers ran population-scale intervention trials with sodium selenite tablets. A 2018 systematic review pooled 41 studies covering nearly two million people and found a risk ratio of 0.14 for selenium supplementation, an approximately 86% reduction in Keshan disease incidence, with protection rates above 80% in most individual studies.

That does not make selenium deficiency the whole story. Reviews of this work conclude that selenium deficiency is the major cause while noting that other contributing factors cannot be ruled out, and the full etiology is still described as unresolved. Severe deficiency appears to create susceptibility rather than to guarantee disease.

The deficiency may even change the pathogen

An animal model later suggested one way nutrition and infection might interact. An amyocarditic strain of coxsackievirus B3, a strain that does not normally damage heart muscle, converted to virulence when inoculated into selenium-deficient mice. The conversion came with changes in the genetic structure of the virus, so that its genome came to resemble those of known virulent strains of the same virus.

The deficiency had not simply weakened the host. In this model it created conditions under which the pathogen itself changed. This is an animal-model finding. It is a compelling mechanism for how a nutritional deficiency could interact with infection, not a demonstration that it explains human Keshan disease.

China also produced the opposite extreme

An endemic disease was discovered in 1961 in parts of Enshi County, Hubei Province. During the years of highest prevalence, from 1961 to 1964, morbidity reached almost 50% among the 248 inhabitants of the five most heavily affected villages.

Hair fell out. Nails became damaged or were lost. Skin and nervous-system involvement appeared. The cause was selenium.

The source began underground. The ultimate environmental source was a stony coal of very high selenium content, averaging more than 300 micrograms per gram, with one sample exceeding 80,000. Weathering carried selenium from the coal into local soils, where it was available to crops, helped along by the traditional use of lime as fertilizer.

A drought then caused the rice crop to fail, forcing villagers to eat more of the high-selenium local crops. Daily selenium intake, estimated after the peak had subsided, averaged 4,990 micrograms, with a range of roughly 3,200 to 6,690. The current U.S. adult RDA is 55 micrograms.

Comparing the two Chinese regions, selenium concentrations in vegetables, cereals, hair, blood, and urine differed by up to a thousandfold between the selenosis areas and the Keshan disease areas. Both extremes began in geology and reached people through food.

Why sulfur keeps appearing in the selenium story

Selenium sits directly below sulfur on the periodic table, and the chemical resemblance matters. Plants take up selenium compounds through transport systems used for sulfur compounds and route selenium through parts of sulfur metabolism. Depending on the form of selenium and the plant, it can end up in compounds such as selenomethionine and selenocysteine.

Why hair, nails, and hooves are hit so visibly

Hair, nails, and hooves are rich in keratin, which gets much of its mechanical strength from sulfur-containing cysteine residues and the disulfide bonds they form. One proposed explanation for the characteristic hair, nail, and hoof damage of selenosis is that excessive selenium interferes with normal sulfur-containing keratin chemistry and alters the proteins being built. It is a plausible and well-discussed mechanism rather than a settled single explanation, and selenium toxicity almost certainly involves several mechanisms at once. What is clear is that the same chemical resemblance that lets selenium travel through sulfur-related pathways becomes a liability when exposure is excessive.

The controlled version is the twenty-first amino acid

At normal physiological levels, selenium is not an accidental sulfur substitute. The body deliberately places it into specific proteins as selenocysteine, generally described as the twenty-first proteinogenic amino acid. Selenocysteine resembles cysteine, except that selenium occupies the position sulfur normally holds.

Getting it into a protein requires unusual machinery. In the standard genetic code, the RNA codon UGA means stop. In a selenoprotein message, the cell reads that UGA differently and inserts selenocysteine instead, guided by specialized RNA structures and translation factors that tell the ribosome this particular stop codon is not the end of the protein.

The body therefore maintains dedicated machinery for placing selenium atoms at exact positions in particular proteins. Selenium becomes part of the amino-acid sequence itself rather than binding to a finished protein as a mineral cofactor.

Brazil nuts are not a dose

Brazil nuts are one of the strangest entries in a nutrient database. USDA data list an average of 544 micrograms of selenium per ounce, about six to eight nuts. That is roughly the entire adult daily requirement in a single average nut.

The average is the least interesting part. Brazil nut trees take up selenium from soil, and soil selenium varies enormously across the Amazon. One study of nuts and soils from the Brazilian Amazon found median nut concentrations ranging from 2.07 mg/kg in Mato Grosso to 68.15 mg/kg in Amazonas. Depending on origin, the authors calculated that a single nut could supply anywhere from 11% to 288% of an adult man’s daily requirement, using 70 micrograms as their reference.

NIH’s own fact sheet notes that values from analyses other than the USDA average vary widely. Brazil nuts are not a dangerous food. The mistake is treating an unusually variable food as though it were a standardized selenium tablet.

Finland changed the soil on purpose

Selenium geography is not only an old story from remote valleys. Finnish soils are selenium-poor, and by the 1970s average selenium intake in Finland had fallen to about 25 micrograms per day.

In 1984, an official decision was made to add selenium, as sodium selenate, to multinutrient fertilizers. Almost all fertilizer used in Finland since 1985 has contained it.

The effect was measurable in the food supply. Selenium concentrations in spring cereals rose roughly fifteenfold compared with levels before the program. Selenium in beef, pork, and milk rose about six-, two-, and threefold respectively, and the selenium status of the population rose with them.

Selenium’s margin is not uniquely narrow

Selenium has a reputation for having almost no room between enough and too much. The visible history explains the reputation. Poisoned livestock lose hooves and hair. Human selenosis damages nails and hair. The toxicity is memorable.

But comparing the U.S. recommended intake with the U.S. Tolerable Upper Intake Level does not show selenium to have a uniquely narrow margin.

By this crude measure, selenium has more room between the recommended intake and the upper limit than zinc, manganese, or iron for premenopausal women. That does not make selenium safer than those minerals; toxicity cannot be ranked from a ratio. It means the familiar description of selenium as having almost no usable window is not what these numbers show.

What selenium does have is unusually recognizable toxicity. Hair and nail changes are hard to forget, and the U.S. upper limit is itself based on them.

The current U.S. adult UL is 400 micrograms per day. The European Food Safety Authority set a lower adult UL of 255 micrograms per day in 2023, derived from a large human trial and an uncertainty factor. Both are population safety boundaries, not cliffs where one more microgram becomes toxic.

Once selenium is sufficient, more is a different intervention

Selenium-sensitive markers such as glutathione peroxidase activity and selenoprotein P respond to selenium when status is low. Once status is adequate, those systems approach saturation and additional selenium no longer produces the same response.

That matters for interpreting who might benefit. Average U.S. intake from foods and beverages is about 108 micrograms per day, nearly twice the RDA of 55, and modern food distribution means few people eat almost entirely from one local geological zone. The extremes of Keshan, Enshi, and historical Finland do not describe the average American diet.

A result obtained by correcting severe deficiency therefore cannot be assumed to transfer to an already-sufficient population. Those are different biological situations. Selenium nonetheless attracted the hope attached to many essential nutrients: if deficiency is harmful, perhaps extra is protective.

The cancer-prevention story did not hold up

Observational work and an early prevention trial created interest in selenium as a possible cancer-protective nutrient. Larger and later work did not justify it. SELECT, the Selenium and Vitamin E Cancer Prevention Trial, randomized 35,533 men aged 50 and older and was discontinued after 5.5 years when analyses showed no reduction in prostate cancer risk from 200 micrograms per day of selenium as selenomethionine.

A separate finding concerns diabetes, and it came from the earlier Nutritional Prevention of Cancer trial, not from SELECT. In a secondary analysis of that trial, participants assigned 200 micrograms of selenium per day developed more type 2 diabetes, with the strongest signal among those who entered with the highest selenium status.

The lesson is narrower than “selenium causes diabetes.” A nutrient that prevents problems when deficient does not necessarily help when added to a diet that already supplies enough.

What the selenium story teaches

Selenium was recognized as a livestock poison before anyone knew animals required it. Twenty years later it became an essential nutrient, and sixteen years after that it acquired a specific job inside an enzyme.

China then demonstrated both environmental extremes: selenium-poor soils where a lethal cardiomyopathy was endemic, and selenium-rich coal a few hundred miles away that poisoned villagers through their crops.

Finland showed that a country could change its own selenium status by changing what went into fertilizer. Brazil nuts show the same geology operating on the scale of a single food. And at the molecular level, the body overrides what is normally a stop signal so that selenium can be placed into particular proteins at particular positions.

“Toxic” and “essential” were never opposites. They were observations made at different doses, in different places, before anyone understood how the pieces fit together.

Related: Selenium · Iodine · Vitamin E

This article is educational and does not constitute medical advice. It describes the history, geography, and biology of selenium rather than individual intake recommendations. See the Selenium reference page for current intake values and nutrient-specific guidance.