Magnesium L-Threonate vs Glycinate: The Numbers Behind the Brain Claim
TL;DR: Magnesium L-threonate carries about half the magnesium per milligram that bisglycinate does. US labels declare 7.2% elemental magnesium, against 14.1% for bisglycinate, so a product advertising “2,000 mg” delivers 144 mg of actual magnesium. Five human trials have tested the form. Four were paid for by companies selling it, and the one funded by a public research agency found no benefit on any outcome it measured.
The other two magnesium comparisons turn on absorption. In glycinate versus oxide nobody has tested it properly; in glycinate versus citrate, citrate has the better evidence and the worse marketing.
Magnesium L-threonate is sold on a different promise: that it reaches the brain. That promise has a specific research chain behind it, short enough to trace end to end. So that is what this page does — the arithmetic on the label, then the studies, then who paid for them.
Quick comparison
| Magnesium L-threonate | Magnesium bisglycinate | |
|---|---|---|
| Chemical class | Organic salt (magnesium + threonic acid, a sugar acid) | Organic salt (magnesium + two glycine molecules) |
| Elemental magnesium, theoretical | 8.3% (anhydrous, MW 294.50) | 14.1% (MW 172.42) |
| Elemental magnesium, as declared on labels | 7.0–7.65%, clustering at 7.2% | 14.1% where labels show the conversion |
| Median declared serving, US products | 144 mg elemental (n = 30 labels) | 350 mg elemental (n = 45 labels) |
| Products over the 350 mg supplemental limit | 0 of 30 | 48.9% |
| Branded, patented raw material | Yes — 28 of 32 labels name Magtein | No — commodity ingredient |
| Human trials of the form | 5, four industry-funded | 1 head-to-head, in 12 bowel-resection patients |
| Trial comparing the two with each other | None exists | None exists |
| Longest published human trial | 12 weeks | Single-dose crossover only |
2,000 mg on the front of the bottle, 144 mg of magnesium inside
The gap between the number on the label and the magnesium in the capsule is wider for L-threonate than for any other common form. Threonate is a bulky carrier: two threonate ions weigh more than eleven times as much as the magnesium ion they hold. The front label counts the whole molecule.
Magnesium L-threonate, anhydrous, has a molecular weight of 294.50 g/mol, of which magnesium is 24.305 — 8.3%, the figure in our table of ten magnesium forms. Real labels declare less. Of the 20 US labels stating both the compound weight and the elemental magnesium, the median conversion is 7.2%, with a range of 7.0% to 7.34%. The trials agree: Ni et al. 2023 describe 1.2 g as “91.8 mg elemental magnesium” (7.65%), and Lopresti and Smith 2026 describe 2 g as 145 mg (7.25%).
In plain terms: a serving marked “2,000 mg Magtein” gives you about 144 mg of real magnesium. The other 1,856 mg is the carrier. The shortfall against the 8.3% theoretical figure is what you would expect if the commercial ingredient carries water of crystallisation — roughly two water molecules’ worth — but no label we read says so, and we are inferring.
Two capsules of a typical bisglycinate product and two capsules of a typical L-threonate product are not the same purchase: the first carries nearly twice the mineral for the same weight of powder.
What we found on 32 US magnesium L-threonate labels
We queried the NIH Dietary Supplement Label Database for on-market magnesium L-threonate products, found 36 labels and retrieved 32 of them, across 24 brands. The picture is unusually uniform, and uniform in a direction that surprised us: these are the most transparent magnesium labels we have surveyed.
The dose is effectively a single number. The median declared serving is 144 mg of elemental magnesium, and 20 of the 30 labels that declare it state exactly 144 mg, with a full range of 48 to 280 mg. Almost everyone is buying the same raw material at the same inclusion rate: 28 of the 32 labels name Magtein.
Not one product exceeds the supplemental upper limit. Zero of 30 declare more than 350 mg per serving, against 48.9% of bisglycinate, 44.4% of oxide and 31.1% of citrate products in our earlier surveys. This is not restraint — it is the carrier. A 350 mg serving would need 4.9 g of compound.
Label transparency is the best of the four forms, by a wide margin. 22 of 32 labels declare both the compound weight and the elemental magnesium, so a shopper can check the conversion. Only 3 of 45 glycinate labels did, and none of the oxide or citrate labels. The reason is commercial rather than ethical: “2,000 mg” is the number the marketing wants, and the panel must state the mineral anyway.
Two labels break the pattern, and three are blends. Barlowe’s Herbal Elixirs declares 800 mg of magnesium L-threonate per capsule with no elemental figure, and Absorb Health declares 500 mg the same way. Three other products mix L-threonate with different magnesium salts, and two of those three also contain magnesium oxide — the same quiet substitution we found in one citrate product in six.
Does it actually reach the brain?
The claim rests on rat studies, and no human study has measured brain magnesium after taking it. The foundational paper is Slutsky et al., Neuron 2010, which reported that magnesium L-threonate improved learning and memory in rats and increased hippocampal synapse density. It was funded by a US National Institute of Neurological Disorders and Stroke grant, NS37342.
The figure everyone quotes from it is a 7% to 15% rise in rat cerebrospinal fluid magnesium over 24 days. We could not open that paper’s full text to check it; the wording we could trace appears in the introduction of Zhang et al. 2022, citing Slutsky 2010. We say so rather than present the number as if we had verified it — see our evidence grading method.
Two things about that chain matter more than the percentage.
The animal doses are an order of magnitude above the human ones. A 2025 mouse study used magnesium L-threonate at 455 and 910 mg/kg/day in drinking water, citing earlier work for 910 mg/kg/day as the dose that “effectively increases brain magnesium levels” (Xiong et al., Exp Neurobiol 2025). A 2 g human serving is about 29 mg/kg/day for a 70 kg adult — 16 to 32 times lower.
The one human trial that tried to measure the mechanism could not. The 2016 trial used red blood cell magnesium as a stand-in and said so plainly: “current technology does not permit safe quantification of intracellular magnesium of neurons in human,” adding that the red cell measure “may not be a true indicator of brain magnesium” (Liu et al. 2016).
Every statement that this form crosses the blood-brain barrier better than others traces back to animals; nobody has measured brain magnesium in a person taking it.
The five human trials, and who paid for them
We found five randomized controlled trials in people, totalling 431 analysed participants, and four were funded by companies selling the ingredient. They are small, short and inconsistent with each other. The fifth, the only one paid for by a public research agency, is also the only clearly negative one.
| Trial | Size and dose | Funder | What it reported |
|---|---|---|---|
| Liu 2016, J Alzheimers Dis | n = 44, 12 weeks, 1.5–2 g/day | Neurocentria Inc. | Composite cognitive score improved vs placebo, p = 0.003 |
| Zhang 2022, Nutrients | n = 102, 30 days, 2 g/day plus phosphatidylserine and vitamins | Wu-xi Magtein Bioscience | All five memory subtests improved, p < 0.001 |
| Ni 2023, Breast Cancer | n = 109, 12 weeks, 1.2 g/day | National Natural Science Foundation of China | No difference in persistent pain (p = 0.67) or cognition |
| Hausenblas 2024, Sleep Med X | n = 76, 21 days, 1 g/day | AIDP Inc. | Insomnia Severity Index not significant (p = 0.390); some subscales were |
| Lopresti 2026, Front Nutr | n = 100, 6 weeks, 2 g/day | Threotech Inc. | Cognition composite p = 0.043; objective sleep no different |
Three details sit in those papers and almost never in the articles that cite them. The 2016 trial’s text says it was “funded by Neurocentria Inc.” and that Neurocentria “wrote the paper” and ran the statistical analysis, with three authors at the company. The 2024 sleep trial’s corrected disclosure says AIDP “commissioned and funded the study and owns patents”, and names three authors it pays. The 2026 trial says its funder “also provided the IP in this study”.
None of that makes a result wrong. It does mean the positive findings and the commercial interest come from the same place, while the one trial without that interest found nothing.
Is L-threonate better absorbed than glycinate?
No trial has compared them, and the claim that it is traces to studies that did not test glycinate. The widely repeated line is that magnesium L-threonate shows “higher absorption and higher retention” than chloride, citrate, glycinate and gluconate. We followed it to its source.
That sentence appears in the introduction of Zhang et al. 2022, which cites two papers for it. The first is Sadir et al. 2019, a rat study comparing L-threonate with magnesium chloride and sulphate — citrate, glycinate and gluconate do not appear in it. The second is Sun et al. 2016, a cultured-neuron and rat study reporting that other magnesium anions “failed to have the same results” without naming them or measuring absorption. The claim that L-threonate out-absorbs glycinate rests on two studies, neither of which tested glycinate.
Bisglycinate’s own absorption evidence is just as thin: one trial in 12 ileal-resection patients found no significant overall difference against oxide, and in a 2024 industry-funded crossover its bisglycinate arm showed no significant plasma rise at all.
Worth knowing: when neither side of a comparison has published absorption data in healthy adults, the form that looks better evidenced is usually just the one with more marketing behind it.
Cost: you buy twice the compound for the same magnesium
Because the carrier is heavier, reaching any given magnesium dose takes 1.96 times as much L-threonate compound as bisglycinate — and the L-threonate is a patented branded material rather than a commodity. Both factors push the same way, which is why this is the widest cost gap among the magnesium forms.
We do not quote retailer prices, for the reason set out in our methodology: shop listings are not a source we can stand behind, and they change weekly. What we can cite is ConsumerLab’s November 2026 magnesium review, which found that across the products it bought and tested, obtaining 200 mg of magnesium cost “from less than 10 cents to more than $1” (ConsumerLab). In plain terms: divide the price of the bottle by the elemental magnesium per serving, never by the number on the front.
Dose, tolerance and safety
The safety rules are the same for both forms, because the risk comes from magnesium rather than from what it is attached to. The adult Tolerable Upper Intake Level is 350 mg of supplemental magnesium a day — supplements and magnesium-containing medicines, not food — set at the level likely to pose no risk of diarrhea (NIH Office of Dietary Supplements).
Two differences follow from the label arithmetic rather than from biology. A 144 mg L-threonate serving sits well under that limit and under the 310–420 mg RDA, so this form is a poor way to close an intake gap. And because none of the 30 products we surveyed exceeded the limit, it is hard to overshoot from this form alone — though a magnesium-containing antacid or laxative alongside it still counts.
Reduced kidney function remains the risk that matters most, identically for both forms. The quantified risk in chronic kidney disease, and the two opposite directions of drug interaction, are set out in the oxide comparison. Do not supplement magnesium with impaired kidney function without medical supervision.
On the carrier itself: at 2 g a day you take roughly 1.86 g of non-magnesium material. The longest published human trial ran 12 weeks, so nobody has reported what daily use beyond three months looks like. Adverse events in those trials were mild and no more frequent than on placebo — in the 2024 sleep trial, four against thirteen.
Seek medical attention for severe or persistent diarrhea, muscle weakness, difficulty breathing or an irregular heartbeat while taking any magnesium supplement.
Who should choose what
If you are trying to raise a low magnesium intake: bisglycinate, or any form with a higher elemental share. A 144 mg L-threonate serving covers under half the adult RDA, and the median bisglycinate serving is 350 mg.
If you are buying it for the cognitive positioning: read the trial table above and decide what you make of it. Four of the five trials were paid for by sellers, the publicly funded one found nothing, none ran past 12 weeks, and none measured brain magnesium.
If you want to verify what you are buying: this is the one magnesium category where the label usually lets you. Look for both the compound weight and the elemental magnesium, and check that the second is about 7.2% of the first.
If you specifically want to avoid magnesium oxide: check the form list, not the product name. Two of the three L-threonate blends we found contained oxide.
If you have reduced kidney function, are pregnant, or take prescription medication: the form is the least important variable here. Talk to a pharmacist before starting either.
Frequently asked questions
Is magnesium L-threonate better than magnesium glycinate?
No trial has compared them, so this cannot be answered from evidence. On the one certain measure, bisglycinate wins: 14.1% elemental magnesium against 7.2% on L-threonate labels. L-threonate has five human trials behind its cognitive positioning, four funded by sellers; bisglycinate has essentially none for anything.
How much magnesium is in 2,000 mg of magnesium L-threonate?
About 144 mg of elemental magnesium, a conversion of 7.2%. That is the figure 20 of the 30 US labels we surveyed declare. The theoretical figure from the anhydrous salt’s molecular weight is 8.3%, and real labels sit consistently below it.
Does magnesium L-threonate cross the blood-brain barrier?
Animal studies report that it raises magnesium in rat cerebrospinal fluid, at doses 16 to 32 times higher per kilogram than a human serving. No human study has measured brain or cerebrospinal fluid magnesium after taking it. The one human trial that tried used red blood cell magnesium instead and said that measure “may not be a true indicator of brain magnesium.”
Can I take magnesium L-threonate and glycinate together?
The relevant limit is the total, not the number of products. Supplemental magnesium from all sources, including magnesium-containing antacids and laxatives, has an adult upper limit of 350 mg per day. A 144 mg L-threonate serving plus a 350 mg bisglycinate serving is 494 mg, well past it. Add up both Supplement Facts panels.
Why is magnesium L-threonate so expensive?
Two reasons compound: you need 1.96 times as much compound for the same magnesium, and 28 of the 32 labels we surveyed use Magtein, a patented branded material rather than a commodity. ConsumerLab’s 2026 testing found the cost of 200 mg of magnesium ranged from under 10 cents to more than $1 across all forms.
Is Magtein the same as magnesium L-threonate?
Magtein is a branded, patented form of magnesium L-threonate, and it dominates the category: 28 of the 32 on-market US labels we retrieved name it. Four of the five human trials used it or a formula built on it, and companies with a commercial interest in it funded those same four.
Verdict
| Winner | |
|---|---|
| Elemental magnesium per mg | Bisglycinate — 14.1% vs 7.2% declared |
| Usefulness for closing an intake gap | Bisglycinate — median 350 mg vs 144 mg per serving |
| Number of human trials of the form | L-threonate — five, against one |
| Independence of that evidence | Neither — four of five L-threonate trials were seller-funded; the public one was negative |
| Head-to-head evidence against each other | Neither — no trial has been designed to compare them |
| Label transparency | L-threonate — 22 of 32 labels let you check the conversion, vs 3 of 45 |
| Staying under the 350 mg supplemental limit | L-threonate — 0 of 30 products exceed it, vs 48.9% |
| Cost per mg of elemental magnesium | Bisglycinate — half the compound, and not a patented material |
| Demonstrated effect on brain magnesium in humans | Neither — never measured |
Overall: this is the rare comparison where the more expensive, more heavily marketed form has the better labels and the worse value. The 7.2% elemental share is the whole story on the shelf: L-threonate is a low-dose product by construction, which makes it a poor way to raise magnesium intake and an unusually frank one about how little it contains. The case for paying the premium rests on five small trials, four of them paid for by sellers of the ingredient, and on a mechanism shown in rats at doses no human takes. That is not nothing. It is also not what the front of the bottle implies.
Sources
- Slutsky I, Abumaria N, Wu LJ, et al. Enhancement of learning and memory by elevating brain magnesium. Neuron. 2010;65(2):165-77. PMID 20152124 — rat study; funded by NIH NINDS grant NS37342.
- Liu G, Weinger JG, Lu ZL, Xue F, Sadeghpour S. Efficacy and safety of MMFS-01, a synapse density enhancer, for treating cognitive impairment in older adults. J Alzheimers Dis. 2016;49(4):971-90. PMC4927823 — funded by Neurocentria Inc., which the paper states designed the study jointly, performed the statistical analysis and wrote the paper; three authors affiliated with the company.
- Zhang C, Hu Q, Li S, et al. A Magtein, magnesium L-threonate, -based formula improves brain cognitive functions in healthy Chinese adults. Nutrients. 2022;14(24):5235. PMC9786204 — funded by Wu-xi Magtein Bioscience; product supplied by Magceutics; multi-ingredient formula.
- Ni Y, Deng F, Yu S, et al. A randomized, double-blind, placebo-controlled trial to evaluate the therapeutic effect of magnesium-L-threonate supplementation for persistent pain after breast cancer surgery. Breast Cancer (Dove Med Press). 2023;15:495-504. PMC10386839 — publicly funded (National Natural Science Foundation of China 81901146); authors report no conflicts.
- Hausenblas HA, Lynch T, Hooper S, Shrestha A, Rosendale D, Gu J. Magnesium-L-threonate improves sleep quality and daytime functioning in adults with self-reported sleep problems. Sleep Med X. 2024;8:100121. PMC11381753 — funded by AIDP Inc., which per the 2025 erratum commissioned the study and owns patents; three authors paid by AIDP. Erratum
- Lopresti AL, Smith SJ. The effects of magnesium L-threonate (Magtein) on cognitive performance and sleep quality in adults. Front Nutr. 2026;12:1729164. Full text — funded by Threotech Inc., which also provided the IP and was involved in study design; a correction has been applied to the article.
- Xiong Y, Yang Y, Ruan Y, et al. Magnesium-L-threonate ameliorates cognitive deficit by attenuating adult hippocampal neurogenesis impairment in a mouse model of Alzheimer’s disease. Exp Neurobiol. 2025;34(2):53-62. PMC12069926 — mouse study; doses of 455 and 910 mg/kg/day.
- Sadir S, et al. Neurobehavioral and biochemical effects of magnesium chloride, magnesium sulphate and magnesium-L-threonate supplementation in rats. Pak J Pharm Sci. 2019;32(1 Suppl):277-83. PMID 30829204 — rat study; did not test citrate, glycinate or gluconate.
- Sun Q, et al. Regulation of structural and functional synapse density by L-threonate through modulation of intraneuronal magnesium concentration. Neuropharmacology. 2016;108:426-39. PMID 27178134 — cultured neurons and rats.
- Schuette SA, Lashner BA, Janghorbani M. Bioavailability of magnesium diglycinate vs magnesium oxide in patients with ileal resection. JPEN J Parenter Enteral Nutr. 1994;18(5):430-5. PMID 7815675
- Pajuelo D, Meissner JM, Negra T, Connolly A, Mullor JL. Comparative clinical study on magnesium absorption and side effects after oral intake of microencapsulated magnesium versus other magnesium sources. Nutrients. 2024;16(24):4367. PMID 39770988 — industry-funded; conflict of interest declared.
- Varga P, Lehoczki A, Fekete M, et al. The role of magnesium in depression, migraine, Alzheimer’s disease, and cognitive health: a comprehensive review. Nutrients. 2025;17(13):2216. PMC12252419
- ConsumerLab. Magnesium Supplements Review, updated 2 November 2026.
- NIH Office of Dietary Supplements. Magnesium — Health Professional Fact Sheet.
- Linus Pauling Institute, Oregon State University. Magnesium.
- NIH Dietary Supplement Label Database (DSLD) — label data retrieved November 2026.
About our label survey. The DSLD search for on-market magnesium L-threonate products returned 36 labels. The database’s rate limiting blocked four of them (two duplicate listings of products already in the sample, plus one Bluebonnet and one Metagenics product), so all figures in this page are from the 32 labels we retrieved, across 24 brands. Thirty of those 32 declare elemental magnesium; the two that do not are counted separately where relevant. We did not re-survey bisglycinate, oxide or citrate for this page; those figures are carried over from our earlier surveys of 45 labels per form.
Elemental magnesium percentages calculated from molecular formulas (Mg 24.305; magnesium L-threonate anhydrous 294.50; magnesium bisglycinate 172.42 g/mol).
This page provides general information and does not replace advice from a qualified healthcare professional. Talk to your doctor or pharmacist before starting a magnesium supplement, particularly if you have kidney disease, are pregnant or breastfeeding, or take prescription medication.
Information only, not medical advice. Talk to a doctor or pharmacist before starting a supplement, particularly if you have kidney disease, are pregnant or breastfeeding, or take prescription medication.