Story
Manganese
Last updated
In one pass Manganese is among the nutrients you least need to manage on purpose.
Educational content, not medical advice — consult a clinician.
Story path
Chapter 1
Mostly plant foods
Manganese is among the nutrients you least need to manage on purpose. Whole grains, nuts, legumes, tea, and dark leafy greens all contain it, and true deficiency on an ordinary diet is extremely rare; in US dietary surveys, manganese intakes met or exceeded the Adequate Intake at every age. In the body it is a key part of a few enzymes: the antioxidant enzyme inside mitochondria, and enzymes of the urea cycle and of making new glucose, all need it.
What is more worth knowing is where the risk lies. Manganese you swallow has to pass two checkpoints, the gut wall and the liver; only a small share is absorbed, and the surplus is dumped into bile, so eating too much is almost impossible. The real problems come from routes that bypass both checkpoints: welding fumes or ore dust breathed into the lungs, years of drinking high-manganese well water, long-term intravenous nutrition, and liver disease that stops bile from draining.
What is more worth knowing is where the risk lies. Manganese you swallow has to pass two checkpoints, the gut wall and the liver; only a small share is absorbed, and the surplus is dumped into bile, so eating too much is almost impossible. The real problems come from routes that bypass both checkpoints: welding fumes or ore dust breathed into the lungs, years of drinking high-manganese well water, long-term intravenous nutrition, and liver disease that stops bile from draining.
Numbers · Why manganese is easy to get enough of
The content figures below are mainly here to show why manganese is so easy to get enough of.Content (mg per 100 g):
Dry tea leaves ~35–100 mg; one brewed cup of tea 0.2–0.5 mg, so three cups supply a good share of a day's needRice bran ~14, pine nuts ~8.8, chili powder ~5, oats (dry) ~4.9Dark chocolate ~2, spinach (cooked) ~0.9, pineapple ~0.9
Adequate Intake (, the reference value set when data are not enough for a recommended amount): men 2.3 / women 1.8 mg/day; Tolerable Upper Intake Level () 11 mg/day.
Compared with a normal diet: two cups of tea, a serving of whole grains, and a salad usually exceed the AI. People who drink a lot of tea naturally take in more manganese, and that in itself is not a problem.
In practice: a normal diet is almost never short of manganese, and there is no need to supplement. The point is not to take more, but to know which enzymes it works in and where the risk of excess lies: industrial inhalation, water sources, and long-term intravenous nutrition (TPN) are the real concerns.
Background · Is the manganese in tea a worry
People who drink tea take in more manganese; should they worry? On the evidence available, no. The US Office of Dietary Supplements (ODS) states that there is no evidence of manganese toxicity from high dietary intakes; the poisoning cases come from long-term occupational inhalation of manganese dust and from drinking high-manganese well water.The reason is the absorption checkpoint: people absorb only about 1–5% of the manganese in food, and the more they eat, the smaller the fraction absorbed (the gut-wall and liver checkpoints are explained in the chapter on the few enzymes that need manganese). Tea also contains polyphenols, caffeine, and other compounds, and manganese is only one of them; there is no reason to drink less tea because of manganese, and no reason to drink more for it either.
So manganese needs essentially no active management: an ordinary diet covers it, and extra supplements are not warranted.
Chapter 2
An antioxidant enzyme in mitochondria
Manganese's most important post is inside mitochondria: manganese superoxide dismutase (MnSOD, also called SOD2) is the mitochondrion's first line of defense against the superoxide anion. The electron transport chain leaks a little superoxide while it makes energy; this enzyme turns it into hydrogen peroxide, which catalase and the selenium-containing enzyme glutathione peroxidase then clear. Mice born entirely without this enzyme die within 10 days of birth.
But flooding the body with antioxidant pills cannot stand in for this enzyme system. Manganese is essential to it, yet the manganese in an ordinary diet already covers its needs, and eating more manganese does not make more of the enzyme. According to mechanistic studies, exercise training raises MnSOD expression in muscle, which comes closer to the real meaning of boosting antioxidant defense than taking manganese does.
But flooding the body with antioxidant pills cannot stand in for this enzyme system. Manganese is essential to it, yet the manganese in an ordinary diet already covers its needs, and eating more manganese does not make more of the enzyme. According to mechanistic studies, exercise training raises MnSOD expression in muscle, which comes closer to the real meaning of boosting antioxidant defense than taking manganese does.
Evidence · What mice and fruit flies show
MnSOD (SOD2) is manganese's most important working site and one of the most studied enzymes in the biology of aging.It clears the superoxide anion (O₂•⁻) in the mitochondrial matrix. Superoxide is a by-product of the electron transport chain that damages mitochondrial DNA, proteins, and lipids; MnSOD turns it into hydrogen peroxide (H₂O₂), which catalase and the selenium-containing glutathione peroxidase (GPx) then clear.
What animal experiments show:
Mice with MnSOD completely knocked out (SOD2 -/-): die within 10 days of birth from dilated cardiomyopathy, with fat building up in the liver and skeletal muscle and metabolic acidosis (Li 1995)Fruit flies induced to overexpress MnSOD: mean lifespan about 16% longer on average, and 30–33% longer in some lines (Sun 2002)
Know the limits: these are experiments that switch a gene off or turn it up in animals. They show the enzyme is a matter of life and death for mitochondria; they do not show that eating more manganese or taking antioxidants extends human life, and fruit-fly results cannot be carried straight over to mammals or people.
This is one of the roots of the antioxidant-supplement theory, and also where it fails. The body's own antioxidant enzyme network (SOD, catalase, GPx) regulates itself, and swallowed vitamins C and E cannot replace it. Summarizing randomized trials in JAMA in 2013, Bjelakovic and colleagues concluded that antioxidant supplements are not associated with lower all-cause mortality, and that beta-carotene, vitamin E, and higher doses of vitamin A may be associated with higher mortality. Mechanistic studies indicate that exercise training raises MnSOD expression through the PGC-1α signaling line; a similar effect of calorie restriction comes mainly from animal studies.
The lesson: antioxidant defense is not a supplement word but an enzyme network with a clear division of labor. Manganese, copper and zinc, and selenium each have their place, and loading up on any one of them cannot carry the whole system. Manganese is essential for MnSOD, but an ordinary diet already provides enough.
bjelakovic-2013-jama-antioxidants
Evidence · MnSOD in human studies
In people, research has mostly looked at genetic differences. SOD2 has a common variant called Ala16Val (whether position 16 carries the amino acid alanine or valine). It has been reported in association with some cancers, cardiovascular disease, and neurodegeneration, but the effect sizes are small, and these are observed associations that cannot show the variant causes anything.Another common question: does low manganese make MnSOD fail? The mechanism suggests the enzyme binds manganese tightly, and an ordinary diet supplies enough for it; the US ODS also notes that manganese deficiency is very rare in people and that its symptoms have never been firmly established.
So the practical aim is not to supplement MnSOD but to let the MnSOD you already have express more: regular exercise and a varied diet get closer to that than manganese pills do.
Chapter 3
The few enzymes that need manganese
Manganese is often described as the cofactor of more than two hundred enzymes, but in most of them magnesium is the first choice and manganese only a backup: both are divalent metal ions and close chemical relatives, so in many enzymes one can stand in for the other. Only a few truly require manganese: arginase, the last step of the urea cycle; pyruvate carboxylase, in making new glucose; glutamine synthetase, which recycles ammonia; and MnSOD in the mitochondria.
The body keeps a tight grip on manganese, which is another reason clinical deficiency is rare: how much the gut wall lets in adjusts to how much manganese the body holds, and the liver then dumps the surplus into bile. With one checkpoint rising as the other falls, the swings of an everyday diet are smoothed out.
The body keeps a tight grip on manganese, which is another reason clinical deficiency is rare: how much the gut wall lets in adjusts to how much manganese the body holds, and the liver then dumps the surplus into bile. With one checkpoint rising as the other falls, the swings of an everyday diet are smoothed out.
Mechanism · Mostly magnesium, a few need manganese
Manganese is the cofactor of over 200 enzymes is a frequently quoted line, but in reality most of those enzymes prefer magnesium, and manganese is an auxiliary or a backup.Manganese and magnesium are close chemical relatives: both are divalent metal ions (Mn²⁺ / Mg²⁺) with similar properties, both often serve as enzyme cofactors (DNA and RNA polymerases, kinases, phosphatases, and others), and in many enzymes they can substitute for each other with slightly different activity.
Few enzymes truly require manganese; these are the ones that count as must be manganese:
Arginase: the last step of the urea cycle, splitting arginine into urea and ornithinePyruvate carboxylase: a key step in making new glucose, turning pyruvate into oxaloacetateGlutamine synthetase: at the core of nitrogen metabolism, fixing ammonia into glutamineMnSOD: the antioxidant enzyme in mitochondria (the chapter on the mitochondrial antioxidant enzyme)
Clinical meaning: when manganese runs low, magnesium can partly take over in most metabolic enzymes, so clinical deficiency is uncommon; for the few enzymes it cannot replace, the body tightly controls their manganese supply. Substitution plus tight control, a double safeguard, is part of why manganese deficiency is so rare in people.
In practice: combination supplements of manganese plus magnesium plus zinc are mostly marketing combinations, not physiological partnerships. How trace minerals really interact is far more complicated than a label, involving competition between ions, transport, and local differences in concentration. Manganese is one of the trace elements that least needs active management.
Myth · Needing X does not mean taking more X
Nutrition has many enzymes that can be described as needing element X, but very few cases where needing X leads to should take more X. Manganese is a clear counterexample: it is indispensable to life, yet the body controls it tightly and an ordinary diet is enough; the US ODS also notes that no group of people is known to be likely to run short.So the manganese plus magnesium plus zinc blends you can buy are mostly marketing combinations rather than physiological ones. Real synergy between trace minerals involves ion competition, transport channels, and local concentrations, which a one-size-fits-all blend cannot deliver. Looked at the other way, single-ingredient manganese supplements often contain 5–20 mg per serving, close to or above the 11 mg/day Tolerable Upper Intake Level, which makes them even less necessary.
Mechanism · How the body keeps manganese in check
The body keeps strict control over manganese, but how does that control actually work? The answer is not inside the cell. It sits on two gates: one in the gut wall, one in the liver.First gate: the gut wall lets only a small share through
The manganese in the whole grains, nuts, and cup of strong tea you chew has to cross the lining cells of the small intestine before it can enter the blood. That layer is not wide open: people absorb only about 1–5% of the manganese in food, and most of it leaves unchanged with the food residue. More important, this gate regulates in the opposite direction: when the body holds more manganese it lets less through, and when it holds less it lets more through. So if you go from one cup of tea to three, the manganese entering your blood does not double with it. It is flattened at the door.
Second gate: the liver dumps surplus manganese into bile
The small share that crosses the gut wall does not join the general circulation directly. It first follows the portal vein into the liver. Liver cells pull manganese out of the blood and pack it into bile to be excreted; bile flows into the small intestine and leaves the body in the feces. More than 90% of absorbed manganese leaves by this bile route, and very little goes out in urine. So manganese's main exit is the biliary tract, not the kidney, the opposite of sodium, potassium, and phosphorus, which leave in urine.
Why these two gates decide who gets poisoned
Read the chain backwards and the high-risk list in the chapter on inhaled manganese harming the brain is no longer a list to memorize. You can derive it yourself:
Cholestasis: bile cannot flow out, the exit is blocked in the pipe, manganese gets in but cannot leave, and it builds up in the bodyCirrhosis or liver failure: the pump that pulls manganese out and dumps it into bile is itself broken; the exit machinery failsLong-term total parenteral nutrition: the nutrient solution goes straight into a vein, so the first gate is bypassed entirely and manganese enters the blood without the gut wall checking it; these patients often have liver problems as well, so the entrance is uncontrolled and the exit is sluggish
The rule is simple: whenever bile is blocked, the liver is damaged, or the gut wall is bypassed, manganese accumulates. This also explains why manganese poisoning almost never comes from eating: the eating route has two gates, while inhaled manganese (welding fumes, ore dust) goes from the lung straight into the blood without passing either.
Looked at the other way, the same two gates explain why deficiency is rare: when intake runs low, the gut wall raises the share it lets through and the liver lowers how much it dumps into bile; one up, one down, and the day-to-day swings disappear. So what the body keeps a tight grip on manganese really means is this: your blood manganese is set mainly by these two gates, not by how much you ate today.
Chapter 4
Manganese in joint supplements
Joint-cartilage supplements often include manganese, on the grounds that a class of enzymes that build the proteoglycans of cartilage needs it. But the enzyme needs X does not lead to taking X treats disease; there is no logical bridge between those two sentences. The manganese in joint formulas has never been tested on its own for clinical outcomes, and even the formula's headline ingredient, glucosamine, did not do significantly better than placebo overall in a large randomized trial.
For knee osteoarthritis, the evidence supports a different list: weight loss for people who are overweight, targeted strength training, short courses of pain relievers, and, when needed, joint injections or joint replacement considered by a doctor.
For knee osteoarthritis, the evidence supports a different list: weight loss for people who are overweight, targeted strength training, short courses of pain relievers, and, when needed, joint injections or joint replacement considered by a doctor.
Evidence · Does the manganese in joint formulas help
Joint supplements often contain a set of ingredients: glucosamine, chondroitin, MSM, manganese, vitamin C, fish oil, and assorted plant extracts.Why manganese? Because a class of enzymes that build the proteoglycans of cartilage needs it. That is a reasonable mechanistic inference, and it belongs to the the enzyme needs X kind of logic.
And the actual evidence: no has shown that the manganese in joint formulas improves clinical outcomes such as pain, function, or imaging, and whether adding it helps has never been tested separately. Even the headline ingredient did not pass: the GAIT trial (NEJM 2006) randomized 1583 people with painful knee osteoarthritis for 24 weeks. Counting a pain reduction of at least 20% as a response, 60.1% responded on placebo and 64.0% on glucosamine (P = 0.30), no significant difference; glucosamine plus chondroitin was 6.5 percentage points higher (P = 0.09), also not significant; the pain reliever celecoxib, included as a comparison, was 10.0 points higher (P = 0.008). In the moderate-to-severe pain stratum the combination responded 79.2% versus 54.3%, but the authors themselves called this exploratory and in need of confirmation.
What has evidence for knee osteoarthritis:
Weight loss for people who are overweight: in overweight and obese older adults with knee osteoarthritis, Messier 2005 measured that each 1 kg of weight lost reduced the force on the knee during walking by about 4 kgTargeted training, especially strengthening the quadriceps at the front of the thighShort courses of pain relievers (such as nonsteroidal anti-inflammatory drugs), used as your doctor advises or as the label directsWhen pain is significant, joint injections assessed by a doctor; joint replacement when the disease is severe
The core lesson: the enzyme needs X and taking X treats disease are completely different claims. Chromium, boron, silicon, and manganese supplements have all made the same leap. A nutrient appearing in a metabolic pathway means it is essential (an adequate diet covers it); it does not mean that taking extra treats disease.
Myth · In the pathway does not mean take more
Manganese supplements have no evidence-based place in osteoarthritis: adding manganese to joint formulas is only a mechanistic inference; in the GAIT trial even the headline ingredient, glucosamine, did not differ significantly from placebo; and what has evidence is weight loss, training, short-term pain relief, and, when needed, injections or joint replacement.The bone side is the same: in animals, manganese deficiency impairs bone formation, but the US ODS notes that no clinical trial has tested manganese supplements alone on human bone, and studies linking blood manganese to in people have given inconsistent results.
The broader lesson is a judgment shortcut worth keeping: appearing in a mechanistic pathway means essential; it does not mean extra supplements treat disease. The same rule keeps coming up with chromium, boron, silicon, and other trace-element supplements.
Chapter 5
Inhaled excess harms the brain
Manganese poisoning almost never comes from eating. Swallowed manganese has to pass two checkpoints, the gut wall and the liver; only a small share is absorbed, and the surplus is dumped into bile. Inhaled manganese (welding fumes, ore dust) goes from the lungs straight into the blood, passes neither checkpoint, and then settles preferentially in the basal ganglia, the deep brain structures that control movement, especially the globus pallidus.
It looks like Parkinson's disease (slowed movement, poor balance, an odd gait), but it damages a different place, and levodopa, the Parkinson's drug, barely helps. Manganese from food almost never reaches excess; the real risks are occupational inhalation, high-manganese well water, long-term intravenous nutrition, and liver disease that stops bile from draining. If you work in one of these settings and develop tremor, unsteady walking, or mood changes, see a doctor early and tell them about your exposure.
It looks like Parkinson's disease (slowed movement, poor balance, an odd gait), but it damages a different place, and levodopa, the Parkinson's drug, barely helps. Manganese from food almost never reaches excess; the real risks are occupational inhalation, high-manganese well water, long-term intravenous nutrition, and liver disease that stops bile from draining. If you work in one of these settings and develop tremor, unsteady walking, or mood changes, see a doctor early and tell them about your exposure.
Safety · How inhaled manganese harms the brain
Manganese poisoning (manganism) is a classic occupational disease and one of the earliest described forms of metal neurotoxicity. In 1837 the Scottish physician John Couper described 5 men who ground manganese ore; they walked clumsily, had expressionless faces, drooled, and trembled, much like the shaking palsy that had recently been described (Parkinson 1817).High-risk occupations: manganese mining, welding with manganese, battery factories, glass and ceramic glazing, and fertilizer production.
Mechanism: inhaled manganese crosses the blood-brain barrier and deposits preferentially in the basal ganglia (especially the globus pallidus), affecting dopamine and glutamate neurons. But it does not follow the same path as Parkinson's disease: Parkinson's mainly damages the substantia nigra, while manganism mainly damages the globus pallidus. Manganese is paramagnetic, so the globus pallidus lights up on T1-weighted MRI, the imaging signature of the condition.
Telling it apart from Parkinson's: manganism brings slowed movement, balance problems, a tiptoe cock-walk gait, and involuntary grimacing or a crying face; Parkinson's more often brings a resting tremor, one-sided onset, and a reduced sense of smell. A key point of difference is the response to levodopa: most people with Parkinson's respond well, while manganism responds poorly. After early removal from exposure, symptoms may improve; late damage to the globus pallidus is irreversible. Chelating drugs (such as EDTA) are sometimes used, but the evidence that they work is limited.
Modern concerns:
Areas with high-manganese drinking water (India, Bangladesh, some rural well water in China) may affect children's cognitive development. Bouchard 2011 studied 362 children aged 6–13 in Quebec, Canada: each 10-fold rise in tap-water manganese was associated with an IQ about 2.4 points lower, and children in the highest fifth of water manganese scored 6.2 points lower than those in the lowest fifth (group medians 216 versus 1 µg/L). This is an observational study measuring a continuous association, not a threshold of the harmful only above 100 µg/L kindInfant formula: soy-based about 200–300 µg/L, cow's-milk-based 30–100 µg/L (so soy formula carries about 2–10 times as much); breast milk 3–10 µg/LPatients on long-term total parenteral nutrition (TPN): the solution bypasses the gut-wall checkpoint, so manganese can build up and cause nerve symptoms; newborns whose livers are still immature, especially premature babies, need extra careLiver failure: manganese leaves mainly through bile, and people with cirrhosis have higher blood manganese, which is partly linked to hepatic encephalopathy
In practice: manganese from food almost never reaches excess (only about 1–5% is absorbed, and the gut and liver regulate it well); the real risks are inhalation, water sources, and long-term TPN. Ordinary people do not need manganese supplements; common multivitamins usually already contain 1.0–4.5 mg, and single-ingredient manganese supplements have no clear medical indication.
Clinical · When parkinsonism may be manganese
Two practical points worth taking away:If a history shows metal exposure (welding, battery factories, mining, long-term TPN, cirrhosis) together with extrapyramidal symptoms (slowed movement, stiffness, problems with gait and balance), do not default to treating it as Parkinson's disease: get an MRI to look at the globus pallidus, and use the response to levodopa as the pivot for telling them apartManganese eaten at home almost cannot reach excess, but well water and water near industrial sites are genuinely worth checking, especially when there are young children in the household
The overall conclusion of the manganese story: deficiency is rare, but excess is real; the single most worthwhile thing an ordinary person can do is not to buy a standalone manganese supplement.
References · 8
- National Institutes of Health, Office of Dietary Supplements. (2021). Manganese — Fact Sheet for Health Professionals. Breast milk 3–10 mcg/L; cow's-milk formula 30–100 mcg/L; soy-based formula 200–300 mcg/L (higher than milk-based, not a 50–75× ratio). Fact sheet (updated March 29, 2021; Wayback snapshot 15 September 2026): humans absorb only about 1% to 5% of dietary manganese; more than 90% of absorbed manganese is excreted via bile into the faeces; multivitamin/mineral supplements that contain manganese typically provide 1.0 to 4.5 mg, and manganese-only or few-nutrient supplements mostly contain 5 to 20 mg; people with chronic liver disease eliminate manganese poorly (fact sheet). ods.od.nih.gov/factsheets/Manganese-HealthProfessional
- Aschner, J. L., & Aschner, M. (2005). Nutritional aspects of manganese homeostasis. Molecular Aspects of Medicine, 26(4-5), 353–362. Oral/parenteral Mn, infant formula, TPN, manganism. Does not report SOD2-knockout phenotypes or Drosophila lifespan. 10.1016/j.mam.2005.07.003
- Li, Y., Huang, T. T., Carlson, E. J., Melov, S., Ursell, P. C., Olson, J. L., Noble, L. J., Yoshimura, M. P., Berger, C., Chan, P. H., Wallace, D. C., & Epstein, C. J. (1995). Dilated cardiomyopathy and neonatal lethality in mutant mice lacking manganese superoxide dismutase. Nature Genetics, 11(4), 376-381. Sod2-/- mice die within the first 10 days with dilated cardiomyopathy, lipid accumulation in liver and skeletal muscle, and metabolic acidosis. Abstract does not report neurodegeneration. 10.1038/ng1295-376
- Sun, J., Folk, D., Bradley, T. J., & Tower, J. (2002). Induced overexpression of mitochondrial Mn-superoxide dismutase extends the life span of adult Drosophila melanogaster. Genetics, 161(2), 661-672. Mean lifespan +16% on average, with some lines showing 30-33% increases. Maximum lifespan +15% average. 10.1093/genetics/161.2.661
- Clegg, D. O., Reda, D. J., Harris, C. L., Klein, M. A., O'Dell, J. R., Hooper, M. M., et al. (2006). Glucosamine, chondroitin sulfate, and the two in combination for painful knee osteoarthritis. New England Journal of Medicine, 354(8), 795-808. 1,583 patients, 24 weeks: glucosamine 1,500 mg, chondroitin 1,200 mg, both, celecoxib 200 mg or placebo. Response (>= 20% pain reduction) with placebo 60.1%; glucosamine +3.9 points (P = 0.30), chondroitin +5.3 (P = 0.17), combination +6.5 (P = 0.09), celecoxib +10.0 (P = 0.008). Assignment was stratified by pain severity (mild n = 1,229; moderate to severe n = 354); in the moderate-to-severe stratum the combination responded 79.2% vs 54.3% (P = 0.002), which the authors call exploratory (abstract, PMID 16495392). 10.1056/NEJMoa052771
- Messier, S. P., Gutekunst, D. J., Davis, C., & DeVita, P. (2005). Weight loss reduces knee-joint loads in overweight and obese older adults with knee osteoarthritis. Arthritis & Rheumatism, 52(7), 2026-2032. 10.1002/art.21139
- Erikson, K. M., Thompson, K., Aschner, J., & Aschner, M. (2007). Manganese neurotoxicity: a focus on the neonate. Pharmacology & Therapeutics, 113(2), 369–377. 10.1016/j.pharmthera.2006.09.002
- Bouchard, M. F., Sauvé, S., Barbeau, B., Legrand, M., Brodeur, M.-È., Bouffard, T., Limoges, E., Bellinger, D. C., & Mergler, D. (2011). Intellectual impairment in school-age children exposed to manganese from drinking water. Environmental Health Perspectives, 119(1), 138-143. Québec, n=362 children aged 6–13. Median tap-water Mn 34 µg/L (range 1–2700). A 10-fold increase in water Mn associated with −2.4 IQ points (95% CI −3.9 to −0.9). Highest vs lowest MnW quintile: −6.2 IQ points (quintile medians 216 vs 1 µg/L). 10.1289/ehp.1002321