Top row: In a mouse model of Alzheimer’s disease, lithium deficiency (right) dramatically increased amyloid beta deposits in the brain compared with mice that had normal physiological levels of lithium (left). Bottom row: The same was true for the Alzheimer’s neurofibrillary tangle protein tau. Images: Yankner Lab
By Jo Nova
A bit of a blockbuster…
Wow. A major new study this month suggests for the first time, finally, what might be a causal link between a deficiency in lithium and Alzheimer’s. The team at Harvard asks: Could Lithium Explain — and Treat — Alzheimer’s Disease?
A few weeks ago, they released a big paper in Nature. They had analyzed brain tissue from people who had died, and found that lithium levels declined dramatically in people with mild cognitive decline, in other words, in the earliest stages, before Alzheimer’s was diagnosed. When they deprived mice of lithium, the mice showed accelerated brain pathology and their memory declined. But when they fed deprived mice lithium, they were able to restore their memory. It’s quite remarkable. There is hope.
We’ve known for years lithium might be essential
For a century or more there have been hints that lithium is an essential element for our health, and especially for our brains. In 1949 an Australian, John Cade, discovered he could treat people with bipolar disorder with lithium. By 1990 researchers knew that low levels of lithium in drinking water were associated with higher levels of suicide suggesting that lithium might be essential for a healthy brain. Studies in the 1970s and 1980s showed rats and goats had higher mortality rates if they were deprived of lithium.
In 2011, researchers showed that in area of Japan with higher levels of lithium in the water, people actually lived longer. In 2017 a group in Texas found that in counties with higher lithium levels, people died less often of Alzheimer’s. In 2018 lithium was shown to enhance the clearance of amyloid-β (Aβ) which is commonly found in people with Alzheimer’s. And yet, there is no RDA for lithium. Like so many things that can’t be patented and sold for profit, our government agencies are slow to get interested in cheap treatments. Imagine if the cure was cents a day instead of $56,000 a year? (Big Pharma will hate it).
But long ago, in the 1860s there was a health craze to bathe and drink bottled water from lithium enriched springs, which maybe wasn’t so crazy.
Lithium deficiency looks like a cause of the build-up of the infamous amyloid plaques?
Their research fills a big gap — we know amyloid plaques are associated with Alzheimer’s, but they are not definitive. There are people with high levels of amyloid who don’t suffer memory loss, and people who do have cognitive decline who don’t seem to have much in the way of plaques. And we’ve spent twenty years, and billions of dollars, trying to treat dementia by reducing the plaques, yet been unable to show any improvement in cognitive abilities.
This study is the first to show that amyloid plaques seem to sequester lithium. In effect, the lithium concentrates in the plaques, but this starves the rest of the brain, and we now know lithium protects brains in several ways. So the real issue appears to be the lack of lithium, not so much the plaques.
The researchers looked at brain tissue from people who were cognitively healthy when they died and compared it to people with mild cognitive impairment (MCI) and people with Alzheimer’s disease (AD). They compared levels of 20 different micronutrients and minerals across the three types of tissue and the one that stood out was lithium. Levels fell in the cortex of people with mild cognitive impairment before they developed Alzheimer’s,
See the graph below, by crikey. It’s rare the medical world to see such a stark difference. The graphs show that lithium, and only lithium is declining in brain tissue compared to blood in people with some form of cognitive impairment. (NCI means no cognitive impairment).
Figure 1: a,b, Volcano plots showing changes in metal cortex-to-serum ratios in the PFC of MCI versus NCI (a) and AD versus NCI (b) cases, along with their statistical significance, determined by one-way analysis of variance (ANOVA) with Tukey’s post-hoc test, followed by the Benjamini–Hochberg correction for the number of metals assessed. (Aron et al 2025: )
Lithium stands alone. No other mineral showed such a dramatic decline in the cortex in association with cognitive impairment. They looked at half the periodic table (nearly) including sodium, potassium, calcium, magnesium, phosphorus, iron, copper, zinc, manganese, cobalt, nickel, chromium, molybdenum, lead, aluminum, cadmium, arsenic, mercury, lithium, strontium, barium, rubidium, boron, and selenium.
There is talk of lithium being used as an early indicator now, used as a way to diagnose (and hopefully prevent) dementia. Though blood levels of lithium are not that useful — it’s the amount in the brain that matters — which is hard to test. (Who wants a brain biopsy?)
If lithium really is the key to Alzheimer’s, or even just one of the main causes, it could explain why some people with no plaques could still suffer memory loss if they were deprived of lithium some other way. Likewise, people with plaques who had a high intake of lithium, could perhaps keep their abilities intact. More studies need to be done to find out.
The paradox is that lithium is concentrated in amyloid plaques, and yet getting more lithium into the brain might be the solution, both to overcome the deficiency, and prevent more plaques forming. The irony is that a lithium deficiency may increase the rate of the plaque formation, but the plaques then hoard the lithium. It’s a nasty spiral.
In Alzheimers, lithium was concentrated in the amyloid plaques.
Lithium acts in several ways that protect the brain
What impresses me is that there are now so many mechanisms. This is not just another study of some loose correlation or association. Lithium inhibits an enzyme called GSK3β — which is normally elevated in Alzheimer’s. GSK3β is involved in building both plaques and tau tangles, so it’s a good thing that lithium puts a brake on this process.
When lithium is low, genes involved in synaptic signaling and structure were broadly down-regulated, and lithium-deficient mice had thinner myelin sheaths around neuronal axons, which can’t be good. (Myelin is like insulation for our nerves). Many of the genes that were up-regulated were risky genes that are also up-regulated in Alzheimer’s.
How much lithium do we need?
Dr Bruce Yanker (the lead researcher at Harvard) obviously tells people we need more studies before he could recommend anyone take lithium.
For those who don’t want to wait another ten years, or who are at risk now, it might be prudent to look up the lithium levels in your water and food (which is easy in the US where most wells have been assayed, but harder in Australia, though I found one map). People who are concerned can take a small over-the-counter supplement to make sure they are not deprived. It seems safe to take doses of lithium that are at the same low level that some people naturally get from food and water.
Lithium is found in leafy greens, legumes, eggs, potatoes and milk but at very low levels, especially in areas where the soil doesn’t have much lithium. Some forms of mineral water have higher levels of lithium too. Or you could move to Portugal or Argentina.
The natural intake of lithium is between 0.5mg to 1mg per day, but in some parts of the world it reaches up to 5mg a day. Doses used to treat bipolar disorder are vastly higher — as much as 100 to 200mg a day of elemental lithium for years on end. This is via a prescription for something like 1,000mg of lithium carbonate. There are sometimes side effects at these high doses that need monitoring (talk to your doctor, etc).
The form of lithium that seemed most useful in this study was Lithium Orotate which is available over the counter in many countries, or at least by mail order from Australia.
Lithium is a favourite of mine, and I’ll be writing more soon. I might start a campaign.
h/t to Michael Rae (20 years ago) and hello to Maurice and Charles who share an interest.
Related posts: One in six dementia cases might be avoided with Vitamin D
ABBREVIATIONS
- NCI: no cognitive impairment,
- MCI: Mild Cognitive impairment
- AD: Alzheimer’s Disease
- GSK3β — Glycogen Synthase Kinase 3 beta
REFERENCES
Aron, L., Ngian, Z.K., Qiu, C. et al. Lithium deficiency and the onset of Alzheimer’s disease. Nature (2025). https://doi.org/10.1038/s41586-025-09335-x
Ohgami, H, Terao, T, Shiotsuki, I, Ishii, N, Iwata, N. Lithium levels in drinking water and risk of suicide. Br J Psychiatry 2009; 194: 464–5.
Schrauzer, GN, Shrestha, KP. Lithium in drinking water and the incidences of crimes, suicides, and arrests related to drug addictions. Biol Trace El Res 1990; 25: 105–13.