Trimethylglycine (TMG)

Overview

Trimethylglycine (TMG) is the same molecule as betaine. The body makes it from choline, and small amounts come from beets, spinach, cereals, and seafood. Its job is to donate methyl groups to homocysteine. The catch: the human trials in this review used grams a day, while the NMN + NR serving from Mt. Angel Vitamins lists 50 mg per serving.

What is Trimethylglycine (TMG)?

Quick facts

Key takeaways

• TMG and betaine are one molecule. The Cystadane prescribing information states that the chemical name of betaine anhydrous is trimethylglycine.
• In a 12-week trial of 100 adults with high homocysteine, 1 gram of betaine a day plus low-dose B vitamins lowered homocysteine 10.1% more than placebo (Lu 2023).
• Betaine alone lowered fasting homocysteine 12% to 20% over six weeks at 1.5 to 6 grams a day (Olthof 2003).
• At 6 grams a day, betaine raised LDL cholesterol by 0.36 mmol/L against placebo in a pooled analysis of four trials (Olthof 2005).
• The NMN + NR serving lists 50 mg of TMG, one-twentieth of the 1 gram in the lowest human dose reviewed here.
• In mice, the enzyme NNMT uses SAM to clear nicotinamide, and knocking it down raised adipose SAM and NAD+ (Kraus 2014).
• No human trial we found measured NAD+ after betaine or TMG (PubMed search, 2026-10-20).

What is trimethylglycine?

Trimethylglycine is glycine with three methyl groups attached to its nitrogen. Chemists call it N,N,N-trimethylglycine. Nutrition papers call it betaine, and supplement labels call it TMG. The Cystadane prescribing information makes the link explicit: the chemical name of betaine anhydrous is trimethylglycine. This page uses TMG for the supplement and betaine for the nutrient and the trials, because most of the trials were run on betaine.

Here is the fact most readers miss. The enzyme that clears extra nicotinamide, NNMT, spends a methyl group from S-adenosylmethionine (SAM) each time it works. In a 2014 Nature study, knocking NNMT down in mice raised SAM and NAD+ in fat tissue (Kraus 2014). Methyl groups are not free, so every nicotinamide cleared carries a methyl cost. That cost is the argument for adding a methyl donor to an NAD+ formula. It is an argument, not a tested result.

The body makes betaine from choline, and the Cystadane label lists beets, spinach, cereals, and seafood as food sources. Typical dietary intake is about 0.5 to 2 grams a day (Olthof 2003). The 50 mg in the MAV serving is between 2.5% and 10% of that range, depending on whether you use the high or low end.

In the body, betaine is an alternative methyl donor. The main donor comes from the folate pathway that depends on MTHFR. Betaine gives homocysteine a second route back to methionine, a point a mouse study measured directly (Schwahn 2003). Methionine is the starting material for SAM, the methyl donor that many enzymes use.

Pharmacokinetic data are thin. The Cystadane label says plasma betaine concentrations after dosing have not been measured in patients, and no pharmacokinetic study is available. In the 2023 trial, blood betaine rose over 12 weeks, and the authors read that rise as a sign that participants were taking the capsules (Lu 2023).

Two names and two salts appear on labels. TMG and betaine anhydrous are the same compound. Betaine hydrochloride is a different salt. Where a trial paper names its form, the 2026 trials used anhydrous betaine. We found no head-to-head absorption comparison of the forms.

The NMN + NR two-capsule serving from Mt. Angel Vitamins lists Trimethylglycine (TMG) (Betaine anhydrous) at 50 mg per serving. The label gives 30 servings per 60-capsule bottle (live product page, checked 2026-10-20). The next sections compare that dose with the trials.

How it works

How it works

Betaine's most documented job is donating a methyl group. The Cystadane prescribing information describes it as a methyl group donor in the remethylation of homocysteine to methionine (section 12.1). The same reaction runs in healthy people. A major share of the methyl groups comes from the folate-dependent MTHFR pathway, and betaine, which the body derives from choline, offers a second route (Schwahn 2003).

Methionine then becomes S-adenosylmethionine (SAM), the methyl donor that many enzymes draw on. Kraus and colleagues note that SAM supplies methyl groups for histone methylation, along with other jobs (Kraus 2014). A methyl-donor argument for NAD+ formulas therefore starts from SAM, not from betaine itself.

The NAD+ connection runs through one enzyme. NNMT methylates nicotinamide using SAM as its donor (Kraus 2014). NMN and NR both feed nicotinamide into the NAD+ cycle. A supplement that raises nicotinamide flow could, in principle, draw more SAM toward NNMT, and a methyl donor is the proposed counterweight. That chain is a hypothesis. The mouse data show the enzyme and its methyl cost. They do not show that people lose methyl capacity on NAD+ precursors.

The closest human check comes from NR, not betaine. In an eight-week trial of NIAGEN (nicotinamide riboside chloride) at 100, 300, and 1,000 mg a day in overweight adults, NR raised whole-blood NAD+ dose-dependently and did not dysregulate one-carbon metabolism (Conze 2019). Two of the three authors list an affiliation with Chromadex Spherix Consulting, and NIAGEN is the NR product the trial tested.

Blends are the last piece. A 2021 mouse study tested a four-ingredient cofactor mix of betaine, N-acetylcysteine, L-carnitine, and nicotinamide riboside in mice with diet-induced fatty liver. The mix reduced liver fat and liver injury markers. The study did not separate betaine's share from the rest of the mix (Quesada-Vázquez 2021). The NAD+ longevity guide covers the precursor side of this chain.

Evidence at a glance

What the evidence doesn't support yet

No human trial in this review measured NAD+ after betaine or TMG. The NAD+ link rests on mouse work and on NR data, and the NR data concern NR, not betaine.

The homocysteine effect at 50 mg has not been tested. The lowest human dose in the trials reviewed here was 1 gram, and the one positive homocysteine trial in this review gave betaine together with B vitamins.

No trial in this review measured heart events. The two signals these trials did measure are LDL cholesterol, which rose at 6 grams a day, and plasminogen activator inhibitor-1 (PAI-1), a clotting-related marker, which rose in the betaine group of the six-month trial at 4 grams a day (Schwab 2010).

Nothing in this review tests aging outcomes, lifespan, or biological age. Studies of NAD+ precursors in people measure NAD+ levels and short-term safety, which is a different question.

Related reading

• Folate and folic acid: from 1930s Bombay to fortified bread

Clinical evidence

• Effects of low-dose B vitamins plus betaine supplementation on lowering homocysteine concentrations among Chinese adults with hyperhomocysteinemia: a randomized, double-blind, controlled preliminary clinical trial — Who: 100 adults aged 18 to 65 with high homocysteine in South China. How long: 12 weeks. Dose and form: 1 g betaine daily plus low-dose folic acid and B vitamins, against placebo. Measured: fasting homocysteine, plus blood folate, B vitamins, and betaine. Result: homocysteine fell by a mean group difference of 3.87 against placebo, a 10.1% reduction (covariate-adjusted P = 0.012 for the difference; P below 0.001 for the reduction rate). Limitation: betaine was given with folate and B vitamins, so its own share is not isolated, and the authors call it a preliminary trial.
• Low dose betaine supplementation leads to immediate and long term lowering of plasma homocysteine in healthy men and women — Who: four groups of 19 healthy men and women. How long: six weeks. Dose and form: betaine at 1.5, 3, or 6 g a day, or placebo. Measured: fasting and post-methionine-load homocysteine. Result: after six weeks, fasting homocysteine was 12%, 15%, and 20% lower than placebo at the three doses (P from below 0.01 to below 0.0001). Limitation: small groups. The abstract does not state how allocation was randomized, so the study type is flagged for review.
• Effect of homocysteine-lowering nutrients on blood lipids: results from four randomised, placebo-controlled studies in healthy humans — Who: healthy adults pooled from four placebo-controlled studies: three betaine studies (n=151), two folic acid studies, and one phosphatidylcholine study. How long: six weeks. Dose and form: betaine 6 g a day in the main comparison, with lower doses also assessed. Measured: serum lipids. Result: LDL cholesterol rose 0.36 mmol/L against placebo (95% CI 0.25 to 0.46), triacylglycerol rose 0.14 mmol/L, and the total-to-HDL ratio rose 0.23. HDL was unchanged. Lower doses also raised LDL, not significantly, and the rise was evident by two weeks. Limitation: pooled from small trials, and the analysis measured lipids, not heart outcomes.
• Long-term effect of betaine on risk factors associated with the metabolic syndrome in healthy subjects — Who: 63 healthy young adults (mean age 27), double-blind and placebo-controlled. How long: six months. Dose and form: 4 g betaine a day in mineral water, against water alone. Measured: serum lipids, plasma homocysteine, hemostatic factors, liver and kidney function. Result: plasma homocysteine did not change in either group, and the authors call the lowering effect weak. Serum lipids showed no long-term effect. Plasma PAI-1 rose in the betaine group (P = 0.028). Limitation: one dose, a young cohort, and six months of follow-up.
• The effects of chronic betaine supplementation on body composition and performance in collegiate females: a double-blind, randomized, placebo controlled trial — Who: 23 untrained young women in a collegiate setting. How long: eight weeks of resistance training. Dose and form: 2.5 g betaine a day (n=11) against placebo (n=12). Measured: body composition, muscle thickness, vertical jump, and squat and bench one-rep maxes. Result: no significant group difference in lean mass, muscle thickness, jump, or strength. Body fat percentage fell 3.3% with betaine against 1.7% with placebo, and fat mass fell 2.0 kg against 0.8 kg (significant interaction). Limitation: 23 women, a single sex and population, and one author lists a DuPont Nutrition and Health affiliation.
• Creatine but not betaine supplementation increases muscle phosphorylcreatine content and strength performance — Who: untrained adults, double-blind and placebo-controlled (n not given in the abstract). How long: 10 days. Dose and form: betaine 2 g a day, creatine 20 g a day, both, or placebo. Measured: muscle phosphocreatine by magnetic resonance spectroscopy, strength, and power. Result: betaine did not raise muscle phosphocreatine against placebo (P = 0.78) and did not change strength or power. Creatine did raise both. Limitation: a 10-day window and an untrained population.
• Betaine supplementation modulates mitochondrial and redox-related adaptations in premenopausal women with overweight: a randomized controlled trial — Who: premenopausal women with overweight or obesity (n not given in the abstract). How long: eight weeks. Dose and form: anhydrous betaine 3 g a day against placebo. Measured: citrate synthase and cytochrome c oxidase in blood mononuclear cells, plus plasma antioxidant enzymes. Result: citrate synthase rose more with betaine (between-group difference 102.3 µg/mg protein, 95% CI 3.19 to 201.4), and plasma catalase rose only in the betaine group. Other markers did not change. Limitation: exploratory cell and blood markers, not clinical outcomes, and the within-group citrate synthase changes were not significant.
• Betaine supplementation has no effect on sex hormones in premenopausal women with overweight or obesity: A randomized controlled trial — Who: 36 premenopausal women with overweight or obesity. How long: eight weeks. Dose and form: betaine 3 g a day against placebo. Measured: plasma and salivary testosterone, DHEA-S, SHBG, estradiol, and cortisol. Result: no significant time-by-treatment interaction for any hormone (P values from 0.378 to 0.882). Limitation: short duration; the authors call for larger, longer studies.
• Safety and Metabolism of Long-term Administration of NIAGEN (Nicotinamide Riboside Chloride) in a Randomized, Double-Blind, Placebo-controlled Clinical Trial of Healthy Overweight Adults — Who: healthy overweight men and women (n not given in the abstract). How long: eight weeks. Dose and form: nicotinamide riboside chloride (NIAGEN) at 100, 300, or 1,000 mg a day against placebo. This is NR, not betaine. Measured: whole-blood NAD+ and metabolites, LDL cholesterol, one-carbon metabolism, and adverse events. Result: whole-blood NAD+ rose 22%, 51%, and 142% at the three doses within two weeks and held. No flushing, no adverse-event difference, no LDL rise, and no disruption of one-carbon metabolism. Limitation: two of the three authors list an affiliation with Chromadex Spherix Consulting, and NIAGEN is a branded ingredient.
• Nicotinamide N-methyltransferase knockdown protects against diet-induced obesity — Who: mice in diet-induced obesity models (n not given in the abstract). How long: not stated in the abstract. Dose and form: genetic knockdown of Nnmt in white fat and liver. Measured: adiposity, cellular energy expenditure, adipose SAM and NAD+ levels. Result: knockdown protected against diet-induced obesity, and NNMT inhibition raised adipose SAM and NAD+. The abstract states that NNMT methylates nicotinamide using SAM as the methyl donor. Limitation: mouse data only, with no human arm.
• Homocysteine-betaine interactions in a murine model of 5,10-methylenetetrahydrofolate reductase deficiency — Who: wild-type mice and mice with mild or severe MTHFR deficiency (n not given). How long: not stated in the abstract. Dose and form: betaine administered to the animals, which the abstract describes as the alternate choline-derived methyl donor. Measured: plasma homocysteine and liver choline metabolites. Result: betaine lowered homocysteine in all three genotypes, restored liver betaine and phosphocholine pools, and prevented severe steatosis in MTHFR-deficient mice. The abstract also reports a negative correlation between plasma betaine and homocysteine in people with cardiovascular disease. Limitation: the main results are from mice; the human correlation is observational.
• Supplementation with a Specific Combination of Metabolic Cofactors Ameliorates Non-Alcoholic Fatty Liver Disease, Hepatic Fibrosis, and Insulin Resistance in Mice — Who: male mice fed a high-fat, high-fructose and sucrose diet (n not given in the abstract). How long: 16 weeks of diet, then 4 weeks of treatment. Dose and form: a four-ingredient mix of betaine, N-acetylcysteine, L-carnitine, and nicotinamide riboside, against vehicle. Measured: liver steatosis, liver injury markers, fibrosis, and insulin sensitivity. Result: the mix reduced liver fat and liver injury markers, reduced fibrosis, and improved insulin sensitivity. Limitation: mice only. The mix cannot show betaine's share, and the authors' human suggestion is untested.

Dosage & forms

• Anhydrous betaine powder (TMG, betaine anhydrous):
• Single-ingredient TMG tablet or capsule:
• Betaine hydrochloride (HCl):
• Minor component in a multi-ingredient NAD+ capsule:
• Prescription betaine anhydrous oral solution (Cystadane):

Safety & interactions

How much to take

The trials used grams, not milligrams. The lowest human dose reviewed here was 1 gram a day, given with folate and B vitamins for 12 weeks (Lu 2023). Healthy adults in a six-week trial took 1.5, 3, or 6 grams a day, and the homocysteine drop grew at each step (Olthof 2003). Recent trials in women used 2.5 to 3 grams a day for eight weeks (Cholewa 2018; Zawieja 2026b). Diet supplies roughly 0.5 to 2 grams a day (Olthof 2003).

Single-ingredient TMG labels mostly sit at 1 gram. One NIH Dietary Supplement Label Database record, a 1,000 mg tablet, says to take one tablet up to two times daily, preferably with meals (record 36315, checked 2026-10-20). Two tablets reach 2 grams a day, which sits inside the trial range.

Our two-capsule NMN + NR serving lists 50 mg of TMG, taken once daily. That is one-twentieth of the 1 gram in the lowest trial and one-fortieth of the 2 grams used in the 2011 strength trial (del Favero 2011). Nothing in this review shows that 50 mg changes any outcome. The NMN and NR dose comparison covers the precursor doses.

No trial in this review compared timing. The prescription label says to mix the powder into liquid or food (Cystadane label, section 2.2).

The 2003 trial saw homocysteine change within six weeks, and the 2023 trial measured at 12 weeks. The prescription label reports homocysteine responses within days at 6 grams a day, with steady state within a month (section 2.1).

The 2003 trial showed more homocysteine lowering at each step up to 6 grams. No trial in this review tested healthy adults above 6 grams a day.

Safety profile

The strongest safety data come from healthy adults given 4 grams a day for six months. Liver and kidney function markers did not change in that trial (Schwab 2010). Two signals need watching. LDL cholesterol rose 0.36 mmol/L at 6 grams a day in the pooled analysis (Olthof 2005), and plasma PAI-1, a clotting-related marker, rose in the betaine group of the six-month trial (Schwab 2010). The LDL rise comes from the pooled 6-gram data. The 4-gram six-month trial did not show one.

Side effects

In a retrospective physician survey of 111 people taking the prescription powder, the most often reported reactions were nausea and gastrointestinal distress, each in two patients. A bad taste, diarrhea, and an odor each appeared once. The survey was open-label and uncontrolled, so it cannot give rates (Cystadane label, section 6.1). Cerebral edema and rising methionine have been reported in people with CBS deficiency who took the prescription powder (sections 5.1 and 6.2).

Drug and supplement interactions

We did not find an interaction study of betaine with common drug classes in the sources reviewed. The table lists what the sources do show.

Who should ask first

• Pregnant or breastfeeding. The prescription label reports only limited case-report data in pregnancy, no animal reproduction studies, and no data on milk (Cystadane label, sections 8.1 and 8.2). We found no safety trial of betaine supplements in pregnancy. The six records we checked were animal or prenatal exposure studies.
• Homocystinuria or another inherited metabolic condition. Betaine is prescribed for homocystinuria and dosed by specialists (Cystadane label, section 2.1).
• High LDL cholesterol or cholesterol medication. The pooled 6-gram result applies here (Olthof 2005).
• Kidney disease. The 2005 analysis notes earlier studies in kidney patients that found cholesterol rises (Olthof 2005). Ask first.
• Children. The prescription label covers pediatric use under specialist care. The sources reviewed include no pediatric trial at supplement doses.

Upper limits and signs of too much

We did not find an established upper limit for betaine in the sources reviewed. The highest trial dose here was 6 grams a day. Prescription dosing reached up to 20 grams a day in some patients under specialist care (Cystadane label, section 2.1). The label reports no human overdose information.

The reactions reported most often are gastrointestinal, so nausea, stomach discomfort, or diarrhea are the early signs to watch for. Stop and ask a clinician if they persist.

If you are pregnant or nursing, take a cholesterol medicine, have kidney disease, or are under care for a metabolic condition, talk to your clinician before you start betaine or TMG.

Buying guide

Match the dose to the research

The trials reviewed here used 1 to 6 grams a day. Read any TMG label against that range. The Mt. Angel Vitamins two-capsule NMN + NR serving lists 50 mg of TMG per serving, which is below every trial dose in this review. No MAV product meets the dose check on this page, so the rest of this section gives the checklist for any brand.

Check the form

TMG and betaine anhydrous are the same compound. Betaine hydrochloride is a different salt, and the trials that name a form used anhydrous betaine. We found no comparison of absorption between the forms. A label should state the amount of TMG itself, in milligrams, the way the NIH label record does ("Trimethylglycine (TMG) 1000 mg, as betaine anhydrous").

Look for proof of testing

This is the one place the manufacturer proof stack appears on this page. The NMN + NR serving is made and tested in Mt. Angel, Oregon, in a facility the product page describes as UL-certified and cGMP. The product page links a lot-specific certificate of analysis for the batch, and the lab-testing page describes each lot's certificate as third-party lab-verified. Ask any brand, including us, for four things: the lot number, the lab that ran the test, the tests run, and the result against the label claim. The MAV certificates are on the lab-testing page.

Read the other ingredients line

The MAV serving lists a hypromellose capsule shell, organic brown rice, and magnesium stearate as other ingredients (live product page, checked 2026-10-20). Those are routine capsule and processing materials. They tell you nothing about the dose, so check the active amount first.

Work out the price per day

The MAV serving costs $59.99 for 30 servings of two capsules, which is $2.00 per day for all six ingredients (live product page, checked 2026-10-20). That is the price of the whole serving, not a TMG price. For a single-ingredient product, divide the price per day by the grams per day on the label. That gives the price per gram of TMG, which is the number to compare.

Red flags

Watch for four things: a proprietary blend that hides the TMG amount; "clinically proven" with no study named; a dose far below the trial range, such as 50 mg presented as if it matched a studied dose; and claims that betaine lowers homocysteine or raises NAD+ with no named trial behind them.

For the label-reading steps in the NAD+ category, see the NMN and NR label-check guide. For the category buying guide, see the NMN and NR buying guide.

Bottom line

Read any TMG label against the 1 to 6 gram trial range, and treat a 50 mg amount in a blend as a small, untested dose. If you have high LDL cholesterol, take a cholesterol medicine, are pregnant or nursing, or are under specialist care for a metabolic condition, ask a clinician before you start.

Related reading

• the trimethylglycine ingredient profile

Frequently Asked Questions

What is trimethylglycine and is it the same as betaine?

Yes. Trimethylglycine (TMG) and betaine are the same molecule, N,N,N-trimethylglycine. Supplement brands tend to say TMG, and the nutrition literature says betaine. The Cystadane prescribing information lists trimethylglycine as the chemical name of betaine anhydrous. The body makes betaine from choline, and small amounts come from beets, spinach, cereals, and seafood. A label that says betaine and a label that says TMG describe the same ingredient, so compare the milligrams, not the names.

Why would an NAD+ formula include a methyl donor?

Because NAD+ precursors may create a methyl cost that a donor could offset. The enzyme NNMT uses S-adenosylmethionine (SAM) to methylate nicotinamide, the fragment that NMN and NR feed into. In mice, knocking NNMT down raised SAM and NAD+ in fat tissue (Kraus 2014). No human trial we found shows that NAD+ precursors deplete methyl donors, and none tested TMG inside an NAD+ formula. It is a reasonable hypothesis, not a tested result.

What does the nicotinamide methylation pathway use up?

It uses S-adenosylmethionine (SAM). NNMT takes a methyl group from SAM and attaches it to nicotinamide. The Nature study states that NNMT methylates nicotinamide using SAM as the methyl donor (Kraus 2014). The methyl group used in that reaction is no longer available to other methylation reactions, which is the reason the pathway matters for a methyl-donor argument. In the mouse knockdown experiments, lowering NNMT raised SAM and NAD+ in fat tissue, which fits a pathway that draws on the same SAM pool (Kraus 2014).

What did the human trials of betaine measure?

Mostly homocysteine, plus lipids, body composition, strength, hormones, and cell markers. Fasting homocysteine fell 12% to 20% against placebo at 1.5 to 6 grams a day over six weeks (Olthof 2003), and 10.1% with 1 gram plus B vitamins over 12 weeks (Lu 2023). LDL cholesterol rose at 6 grams a day (Olthof 2005). Strength did not change in two small trials, and fat mass fell more with betaine in one trial of 23 women (Cholewa 2018). Hormones did not change in a 2026 trial of 36 women.

How does 50 mg of TMG compare with the doses in human studies?

The trials used 1 to 6 grams a day. The Mt. Angel Vitamins two-capsule NMN + NR serving lists 50 mg of TMG, which is one-twentieth of the lowest human dose reviewed here (1 gram, Lu 2023) and one-fortieth of the 2 grams used in the 2011 strength trial. None of the trials reviewed here tested a dose that small, so their results do not transfer to 50 mg.

Who should ask a clinician before taking TMG?

Ask first if you are pregnant or breastfeeding, if you have homocystinuria or another inherited metabolic condition, if you take a cholesterol-lowering medicine or have high LDL cholesterol, or if you have kidney disease. Betaine is prescribed for homocystinuria under specialist care, at doses in grams a day. The 2005 pooled analysis found LDL rises at 6 grams a day. Stop and ask if nausea, stomach discomfort, or diarrhea persists.

Do I need TMG at all?

Only if you have a reason to. Betaine's most tested effects are on homocysteine and LDL cholesterol. The human data for NAD+ and for aging outcomes are missing. Food supplies some betaine, including beets, spinach, cereals, and seafood (Cystadane label). The NR trials show NAD+ rising with NR alone (Conze 2019), so the TMG question is a separate question from the precursor question. The [NMN vs NR comparison](/compare/nmn-vs-nicotinamide-riboside) covers the precursor side.

Is TMG the same as choline?

No. Choline is the precursor, and betaine is a metabolite of choline (Cystadane label, section 12.1). Choline is a separate nutrient. [Citicoline](/ingredients/citicoline) is a different choline compound, not betaine. The Schwahn mouse study describes betaine as the alternate, choline-derived methyl donor (Schwahn 2003). The trials in this review studied betaine itself, not choline. A result for one does not automatically carry over to the other.