Vitamin A: The Nutrient That Kept Turning Out to Be Something Else

Ancient Egyptian and Greek medical texts describe liver as a remedy for people who could not see well at dusk. The Ebers Papyrus prescribes roasted liver for the condition, and a Hippocratic treatise on the eye recommends raw ox liver dipped in honey. The reason would not be understood for another two thousand years: liver stores unusually large amounts of preformed vitamin A, and night blindness is one of the earliest signs of deficiency.

Modern vitamin A research began somewhere else entirely. In 1913, scientists studying growth found that animals could receive what appeared to be adequate protein, carbohydrate, fat, and minerals and still fail to thrive. The missing factor eventually became known as vitamin A, but growth was only the first of several roles researchers would uncover.

1913: something in fat that animals needed to grow

Elmer McCollum and Marguerite Davis at Wisconsin and Thomas Osborne and Lafayette Mendel at Yale were independently feeding rats diets assembled from purified ingredients. The diets appeared complete by the nutritional standards of the time, yet the animals stopped growing normally.

Growth resumed when butter or egg yolk was added. Lard and olive oil did not have the same effect, which meant the missing factor was not fat itself but something carried in certain fats. Both groups published in 1913, in the same volume of the Journal of Biological Chemistry.

The naming came later. The growth factor was called fat-soluble A in 1918, distinguishing it from the water-soluble material that would eventually be separated into the B vitamins, and became vitamin A in 1920. At that stage it had no known chemical identity and no unified biological explanation.

Vitamin A becomes part of the visual cycle

Deficiency was soon connected with the eye. Animals adapted poorly to darkness, and severe human deficiency caused night blindness, xerophthalmia, corneal damage, and eventually irreversible blindness.

George Wald supplied the molecular explanation beginning in the 1930s. He showed that the light-sensitive pigment rhodopsin contains retinal, a derivative of vitamin A, bound inside the protein opsin. When rhodopsin absorbs a photon, its 11-cis retinal isomerizes to the all-trans form. That change in the molecule’s shape forces a conformational change in the surrounding opsin, which activates the G protein transducin and begins the phototransduction cascade that lets a photoreceptor signal light.

Wald shared the 1967 Nobel Prize in Physiology or Medicine with Ragnar Granit and Haldan Keffer Hartline, for discoveries concerning the primary physiological and chemical visual processes in the eye. Vitamin A deficiency was therefore doing something more specific than making an eye generally unhealthy. It was depriving the visual system of a molecule built directly into the machinery that detects light.

Carrots revealed a second route to vitamin A

Plant foods created an apparent contradiction. Carrots, leafy greens, and other plants showed strong vitamin A activity even though they did not contain the preformed vitamin A found in liver, eggs, dairy foods, and fish.

In 1930, Thomas Moore demonstrated that animals fed carotene accumulated vitamin A in their livers. Plants were supplying provitamin A carotenoids that the body could convert, rather than supplying retinol itself. That distinction remains part of modern nutrition.

The conversion is also feedback-regulated rather than fixed. Retinoic acid signaling induces an intestinal transcription factor called ISX, which in turn reduces expression of the transporter that takes carotenoids up and the enzyme that cleaves beta-carotene into retinal. As vitamin A status rises, that loop turns the conversion down. It does not switch it off, and it does not apply to preformed vitamin A, which bypasses the conversion step entirely. The difference between what a label lists and what the body absorbs shows up throughout vitamin A chemistry, and the RAE system exists because of it.

The liver stores enough vitamin A to create both a reserve and a risk

Most of the body’s vitamin A is stored in the liver as retinyl esters, held in lipid droplets inside hepatic stellate cells. Those cells account for roughly 80 percent of the body’s vitamin A, and the reserve is large enough to keep status adequate through periods when intake falls.

The same storage system is why animal liver can be dangerous to eat in quantity. Rodahl and Moore measured polar bear liver in 1943 and found on the order of 13,000 to 18,000 IU of vitamin A per gram. Stellate cells in Arctic top predators such as polar bears and Arctic foxes store 20 to 100 times the vitamin A found in humans or rats. Acute hypervitaminosis A can cause severe headache, blurred vision, nausea, dizziness, aching muscles, and problems with coordination.

Chronic excess is a different problem, involving dry skin, painful muscles and joints, fatigue, depression, and abnormal liver tests. The tolerable upper intake level for adults is 3,000 mcg RAE per day of preformed vitamin A. Excess preformed vitamin A during pregnancy can also disrupt fetal development, and malformations of the eye, skull, lungs, and heart have been reported, which is why pregnant women are advised not to take supplements above that amount.

Prescription systemic retinoids such as isotretinoin act directly on the same developmental signaling system and carry strict pregnancy precautions of their own. That is a pharmacologic exposure at drug doses, not a dietary one, and the two should not be treated as the same risk.

Provitamin A carotenoids behave differently because the conversion step is regulated. Very high carotenoid intake can produce carotenodermia, a harmless yellow-orange coloration of the skin that reverses when intake drops, without producing vitamin A toxicity. Beta-carotene is not known to be teratogenic. The visible color change can look dramatic while remaining physiologically very different from excess preformed vitamin A.

Deficiency also changed barriers and immunity

Vitamin A was linked to infection long before the underlying biology was understood. Harry Green and Edward Mellanby described it as an anti-infective vitamin in the British Medical Journal in 1928, on the basis that deficient animals showed increased susceptibility to infection.

The affected tissues offered part of the explanation. Vitamin A helps maintain epithelial differentiation in the respiratory tract, gastrointestinal tract, eye, and other surfaces. With severe deficiency, those tissues can lose their normal specialized character and become dry and keratinized, weakening barriers that separate the body from the outside environment. Vitamin A also influences immune-cell development and function directly.

The public-health consequences are large. Vitamin A deficiency affects an estimated 190 million children under five, and the current Cochrane review of 19 trials covering more than 1.2 million children found that supplementation reduced all-cause mortality by about 12 percent. It also reduced deaths from diarrhea by about 12 percent and lowered the incidence of measles, night blindness, and Bitot’s spots. The same review found no clear effect on measles mortality specifically, so the benefit is best described as a broad reduction in childhood death and eye disease rather than protection against one illness.

Retinoic acid connected vitamin A to gene regulation

A major piece of the biology arrived in 1987, when two groups independently identified nuclear receptors for retinoic acid, a metabolite of vitamin A. Both reports appeared in Nature that year, one from Pierre Chambon’s laboratory and one from Ronald Evans’s. These receptors bind DNA and regulate transcription in response to retinoic acid.

Retinal and retinoic acid therefore perform very different jobs. Retinal participates directly in the chemistry of vision. Retinoic acid acts as a signaling molecule that changes which genes a cell expresses.

This helped explain observations that had previously looked unrelated. Embryonic development requires cells to receive positional and developmental signals, epithelial tissues need instructions that maintain their differentiated state, and immune cells alter their behavior as they mature. Retinoic-acid signaling participates in all of these. The growth failure seen in the purified-diet experiments was no longer just evidence that an unknown nutrient was missing; vitamin A metabolites were involved in controlling developmental programs throughout the body.

Retinoid biology became medicine

Retinoids entered medicine through several routes. Topical tretinoin and systemic isotretinoin changed the treatment of acne and other skin disorders by altering epithelial-cell growth and differentiation.

The more unusual application appeared in acute promyelocytic leukemia, or APL. Most APL involves a translocation between chromosomes 15 and 17 that creates the fusion protein PML-RARA, which disrupts normal retinoic-acid signaling and blocks granulocytic differentiation. Immature promyelocytes become stuck at an early stage and accumulate rather than completing normal maturation.

Researchers in Shanghai treated patients in the 1980s with pharmacologic doses of all-trans retinoic acid, or ATRA. Their 1988 report in Blood described 24 patients, 16 previously untreated and 8 who had not responded to chemotherapy, and all 24 attained complete remission. One patient who proved resistant to ATRA reached remission only after low-dose cytarabine was added. In marrow cultures from the patients studied, the leukemic cells did not simply die; they differentiated into granulocytes, the mature cells the promyelocytes had been prevented from becoming.

This differed conceptually from conventional cytotoxic chemotherapy. Instead of primarily killing rapidly dividing cells, ATRA restored a developmental program the leukemia had interrupted, and the patients reached remission without the bone marrow hypoplasia that cytotoxic treatment produces. Modern treatment combining ATRA with arsenic trioxide has made APL one of the most curable forms of acute leukemia, with long-term survival above 90 percent, although death within the first 30 days of diagnosis remains the main barrier and is considerably more common outside clinical trials.

Beta-carotene supplements challenged a simple cancer-prevention idea

Observational studies in the late twentieth century repeatedly found lower lung-cancer rates among people who ate more carotenoid-rich fruits and vegetables. That raised a reasonable experimental question: would giving beta-carotene itself reduce cancer risk in people at high risk?

Two large randomized trials answered it. The Alpha-Tocopherol, Beta-Carotene Cancer Prevention Study gave 20 mg of beta-carotene daily to male smokers aged 50 to 69 in southwestern Finland, enrolling 29,133 men. The Beta-Carotene and Retinol Efficacy Trial, or CARET, gave 30 mg of beta-carotene combined with preformed vitamin A to 18,314 smokers, former smokers, and asbestos-exposed workers in the United States.

Neither found protection. ATBC reported an 18 percent higher incidence of lung cancer among men receiving beta-carotene, and CARET found a relative risk of 1.28 for lung cancer in the active group and stopped 21 months early. The mechanisms remain a subject of interpretation, but the clinical result is not ambiguous: associations seen with carotenoid-rich foods did not transfer to high-dose beta-carotene supplements in these high-risk populations.

Beta-carotene and ATRA are both part of vitamin A chemistry, but they are not the same molecule and were not doing the same thing. One is a dietary precursor given at supplement doses to healthy people at risk of cancer; the other is a metabolite given at pharmacologic doses to correct a specific molecular defect in an existing leukemia. Chemical form, dose, physiology, and population all mattered.

What vitamin A turned out to be

Vitamin A now refers to a related group of compounds with distinct jobs.

The researchers who identified fat-soluble A as a growth factor were right. So were the researchers who connected it to vision, to epithelial integrity and immunity, to gene regulation and development, and later to medicine. Each result described a real part of vitamin A biology, and each also showed that the previous definition had been incomplete.

The purified-diet experiments had identified a genuine nutritional requirement long before anyone could see how many different biological processes depended on it. That pattern, where a nutrient’s classification lags behind its biology, is common in nutrition history. Vitamin A is the case where it happened most often to the same nutrient.

Related: Vitamin A · Vitamin D · Vitamin E

This article is educational and does not constitute medical advice. It describes the history and biology of vitamin A rather than individual intake recommendations. Preformed vitamin A has a tolerable upper intake level, and excessive intake carries real risks, including during pregnancy. See the Vitamin A reference page for current intake guidance, and follow individualized medical guidance for prescription retinoids.