I’ve been reading about copper for years, and the longer I read the less the toxicity story holds together.

That’s an awkward position to hold in the mineral world. Copper toxicity arrived a long time ago and has been repeated so often that questioning it reads as eccentricity rather than as ordinary skepticism. But the evidence underneath it is thinner than its reputation, and the way the idea got assembled is not hard to trace.

Here’s my short version. Copper isn’t a toxin looking for an opening. It’s a regulator. Most of what gets blamed on copper looks to me like what happens when copper stops working.

How the idea got built

The early researchers writing about unbound copper were doing the best they could with what they had. They could measure how much copper sat in a tissue. They could not tell you what state it was in or what it was doing there.

That gap got filled with an assumption. The assumption was that copper found in tissue is copper running loose. It hardened into received wisdom in conventional and alternative medicine alike, and it’s still load-bearing today.

Nobody has demonstrated anything in copper that parallels the labile iron pool. Iron has a documented loosely bound fraction that is genuinely reactive and genuinely destructive. Copper doesn’t. Copper travels bound, to enzymes, to ceruloplasmin, to albumin and other carriers. That looks like a system built specifically to keep copper from doing free chemistry.

Which leaves the measurement problem sitting in plain view. A biopsy reports total copper. So does a hair analysis. Neither can tell you whether that copper is inside a working enzyme, held by a transport protein, or actually adrift. A high reading is equally consistent with excellent enzyme function and with metabolic collapse. The literature has generally read it as the second, and I don’t know why.

What Wilson’s disease might actually show

Samuel Alexander Kinnier Wilson described his disease in 1912 as a neurological disorder with liver involvement. The copper connection didn’t turn up until 1948, when John Cumings found elevated copper in brain and liver tissue. After that, copper accumulation was the whole story.

Nobody was looking at iron. Iron overload has since been documented in Wilson’s patients, including in people without HFE mutations, which complicates the standard account more than it usually gets credit for.

The disease involves ATP7B, a copper transport gene. But read the same facts differently. When copper transport fails, copper-dependent enzymes stop working. Ceruloplasmin is one of them, and ceruloplasmin is the ferroxidase that oxidizes iron so it can be moved safely. Lose that and iron regulation falls apart. The damage that follows would be iron damage, showing up in a disease we named after copper.

Aceruloplasminemia says the same thing more plainly. A ceruloplasmin defect produces an iron overload disorder, with iron piling up in the brain. Copper function fails and iron pathology follows.

The experiments underneath all this

Most of the animal evidence for copper toxicity comes from sheep, and a lot of it from Western Australia, where flocks developed liver disease that researchers attributed to copper.

Those sheep were eating pasture whose mineral content shifted with soil, season, and the interactions between minerals. Molybdenum matters enormously here, since molybdenum and sulfur together determine how much copper an animal can actually use. The studies found a correlation between copper exposure and liver disease. What they mostly didn’t measure was copper-dependent enzyme activity or iron status. The mechanism was inferred, not observed.

The cell culture work has a worse problem. These experiments drop copper ions directly onto cells, which is a situation that doesn’t occur in a living animal. Inside a body, copper arrives bound and chaperoned. Studying free copper ions and concluding something about copper in people is a mistake about what’s being studied, not a small methodological quibble.

The iron we added on purpose

American flour enrichment started in 1941, when the FDA set a standard of identity for enriched flour that included iron. The recommendation had come the year before from what is now the Food and Nutrition Board. Within about two years most of the white bread in the country was enriched.

The part I find more interesting is what came after. In 1974 the National Academy of Sciences recommended raising iron levels, and the FDA revised the standards upward. Iron intakes among young men went up. Concerns about excess iron followed quickly. In 1977 the FDA lowered the standards again.

The regulators ran into the ceiling themselves, less than three years in.

This matters because of how iron leaves the body, which is to say it mostly doesn’t. There’s no regulated excretion route. Iron exits through blood loss. We’re built to hold onto iron and we’re not built to get rid of it, which was sensible for most of human history and is a problem in a fortified food supply. Meanwhile the enzymes that manage iron all run on copper, and dietary copper has gone the other direction.

What copper does when it works

Through ceruloplasmin, copper governs iron’s oxidation state. Through superoxide dismutase, it handles oxidative stress. It supports cellular respiration and the building of connective tissue.

When those functions drop, the resulting picture looks a great deal like copper poisoning. Ceruloplasmin activity falls and iron metabolism stumbles. Superoxide dismutase activity falls and oxidative stress climbs. What presents is oxidative damage, and copper takes the blame for damage that followed from copper not working.

There’s also the question of what accumulation actually means. Copper sitting in tissue may be a consequence rather than a cause, the way a pile of undelivered mail during a postal strike tells you the system stopped rather than that the mail is dangerous.

None of this gets revisited quickly, and I don’t think that’s mysterious. Reconsidering copper toxicity means reconsidering iron fortification, which is a public health measure with real successes attached and a great deal of institutional weight behind it. Funding flows toward work that extends a frame rather than tests it. And iron and copper metabolism are so entangled that studying either alone will usually miss the thing that explains what you’re seeing. Most research studies them alone.

What would change my mind

I’ve just spent several pages criticizing a framework for being hard to test, so let me say what would falsify mine.

I’m claiming that copper dysfunction rather than copper excess drives most of what gets called copper toxicity, and that iron dysregulation does the actual damage. If people with the classic symptom picture consistently showed intact ceruloplasmin activity and normal iron handling, I’d be wrong. If lowering iron burden in that group reliably did nothing, I’d be wrong. Before concluding anything in either direction I’d want to see ceruloplasmin, ferritin, transferrin saturation, and serum iron.

That’s the standard I’d ask of anyone working in this space, including the people I disagree with.

The questions I care about now are different from the ones I started with. Not how much copper is present, but whether copper-dependent enzymes are working. Not whether copper is high, but whether iron regulation has broken. That’s harder to assess and it tells you more.

Medicine has been here before. Pellagra was read for years as a toxic reaction to something in corn, and the observation was real enough. The causal story was simply backwards, and it stayed backwards for a long time while people died of a vitamin deficiency.

I don’t think any of this throws out what came before. It rearranges it. And I notice that once you stop treating copper as the threat, a lot of things that didn’t fit start to.

References

Bull, P. C., Thomas, G. R., Rommens, J. M., Forbes, J. R., & Cox, D. W. (1993). The Wilson disease gene is a putative copper transporting P-type ATPase similar to the Menkes gene. Nature Genetics, 5(4), 327-337.

Collins, J. F., Prohaska, J. R., & Knutson, M. D. (2010). Metabolic crossroads of iron and copper. Nutrition Reviews,68(3), 133-147.

Cumings, J. N. (1948). The copper and iron content of brain and liver in the normal and in hepato-lenticular degeneration. Brain, 71(4), 410-415.

Gulec, S., & Collins, J. F. (2014). Molecular mediators governing iron-copper interactions. Annual Review of Nutrition,34, 95-116.

Holmberg, C. G., & Laurell, C. B. (1948). Investigations in serum copper II: Isolation of the copper containing protein, and a description of some of its properties. Acta Chemica Scandinavica, 2, 550-556.

Institute of Medicine. Overview of food fortification in the United States and Canada. In Dietary Reference Intakes: Guiding Principles for Nutrition Labeling and Fortification (2003).

Kell, D. B. (2009). Iron behaving badly: inappropriate iron chelation as a major contributor to the aetiology of vascular and other progressive inflammatory and degenerative diseases. BMC Medical Genomics, 2(1), 2.

Laurell, C. B. (1947). Studies on the transportation and metabolism of iron in the body. Acta Physiologica Scandinavica,14(Suppl 46), 1-129.

Scheinberg, I. H., & Gitlin, D. (1952). Deficiency of ceruloplasmin in patients with hepatolenticular degeneration (Wilson’s disease). Science, 116(3018), 484-485.

Sternlieb, I., & Scheinberg, I. H. (1963). The role of radiocopper in the diagnosis of Wilson’s disease. Gastroenterology,44, 550-553.

Wilson, S. A. K. (1912). Progressive lenticular degeneration: a familial nervous disease associated with cirrhosis of the liver. Brain, 34(4), 295-507.