Bemethyl

Bemitil — 2-ethylthiobenzimidazole

Bemethyl (Bemitil — 2-ethylthiobenzimidazole) is a performance & endurance research compound. A Russian 'actinoprotector' — the drug class built around improving tolerance to physical work under hypoxia. Its described action is on mitochondrial protein synthesis: rather than stimulating output acutely like a beta-agonist, it is reported to raise the enzymatic capacity for gluconeogenesis and aerobic metabolism over a course of days, which is why it is dosed as a course rather than pre-workout.

How Bemethyl works

A Russian 'actinoprotector' — the drug class built around improving tolerance to physical work under hypoxia. Its described action is on mitochondrial protein synthesis: rather than stimulating output acutely like a beta-agonist, it is reported to raise the enzymatic capacity for gluconeogenesis and aerobic metabolism over a course of days, which is why it is dosed as a course rather than pre-workout.

Where to get Bemethyl

The evidence for Bemethyl

Graded by what exists behind each claim.

✅ Clinically validated

• Russian and Soviet clinical literature, where it is classed as an 'actinoprotector' — a drug category that essentially does not exist outside that research tradition. Studied for physical work capacity, hypoxia tolerance and recovery, typically at 250-500mg.
• Almost none of it has been replicated independently. Single-group designs, one research school, one language. See [[evidence-framing-rule]] — that is an absence of evidence, not evidence of absence, and the two get confused constantly with this class.

📊 Correlative data

• It sits in the same evidence position as the Khavinson bioregulators already in the Vault: decades of use inside one system, and a near-total gap in the Western literature. Read it the same way.

🧪 Theoretical / extrapolated

• Described as acting on mitochondrial protein synthesis — raising the enzymatic capacity for gluconeogenesis and aerobic metabolism rather than stimulating output acutely.
• That is why it is dosed as a course, not pre-workout. If the mechanism is adaptive rather than stimulant, a single dose before training is the wrong way to use it and the wrong way to judge whether it worked.

How to read these tiers: they say how much human evidence exists, not how well something works — and ✗ flags harm, never a disappointing trial. How the evidence tiers work →

What Bemethyl actually does

Bemethyl is 2-ethylthiobenzimidazole, usually supplied as the hydrobromide salt, and the sulfur atom is the whole molecule. A benzimidazole ring with an ethylthio group at position 2: a small, lipophilic, weakly basic heterocycle of about 178 Da as the free base. That thioether is simultaneously the reason the compound is absorbed orally, the reason it reaches the brain, and — as the metabolism section below shows — the reason it costs the body something to clear.

It belongs to a drug class that has no equivalent outside the Russian-language literature. An actoprotector is defined by what it is supposed to do: maintain physical work capacity under hypoxia without stimulating the central nervous system. Oliynyk 2012 is the English-language review of the class and it is the right place to start, because the framing matters: this is not a stimulant, and a page that treats it as one has already made the central error. Stimulants raise output and raise oxygen cost. The claim here is the opposite — that output is maintained because the cost falls.

The proposed mechanism is protein synthesis, and it explains the unusual time course. The description in the class literature is that bemethyl increases synthesis of mitochondrial and gluconeogenic enzymes — the machinery, not the substrate Oliynyk 2012. Two things follow. First, an effect that requires new enzyme to be transcribed and translated cannot appear in an hour; the class literature describes courses of days to weeks, and that is a mechanistic prediction rather than a dosing convention. Second, increased gluconeogenesis is a specific and testable claim — it means lactate is cleared back to glucose faster, which is exactly the phenotype "tolerance to work under hypoxia" describes.

What is not established is a receptor. No binding site, no cloned target, no dose-response curve at a purified protein. The mechanism above is a description of downstream effects, and calling it a mechanism is a courtesy the compound has not yet earned. That is a different situation from having no mechanism at all — the metabolism is known in detail, and that is where this page gets specific.

Cell, rodent, human — and where it stops

Step one, metabolism in rats, and it is the best-characterized thing about this molecule. Belinskaia 2021 gave rats a single oral dose of 330 mg/kg and analyzed 24-hour urine by LC-MS/HRMS. Nine metabolites across six molecular formulas were identified, and the most abundant was a benzimidazole-acetylcysteine conjugate. Molecular docking in the same paper demonstrated the conjugation mechanism with glutathione.

Read that result properly, because it is the most useful sentence on this page. An acetylcysteine conjugate in urine is the end product of the mercapturic acid pathway: glutathione S-transferase attaches glutathione, the glutamate and glycine are trimmed off, and the remaining cysteine is N-acetylated and excreted. So the dominant route of elimination for bemethyl spends glutathione, one molecule at a time. Nothing in the marketing for this compound mentions that, and it is the basis of both the prediction and the caution below.

Step two, pharmacokinetics, which exist — the card is wrong to say they do not. Rat pharmacokinetics were published in 1987 Boiko 1987, and human pharmacokinetics in healthy volunteers were published in 2011 under the trade name Metaprot Kibal'chich 2011. Both are in Russian-language journals and neither is open access, which is a different problem from the data not existing. The card at the top of this page says the half-life is "not well characterized in English-language sources", and that is precisely accurate and easy to misread as "unknown".

Step three, in humans, clinically. The clinical literature is Soviet and post-Soviet, small, and mostly not randomized by modern standards: bemitil in progressive muscular dystrophies Lobzin 1992, and an effect on bronchial epithelial regeneration and respiratory protective mechanisms in chronic inflammation Kobylianskii 1999. These are real published human reports. They are not randomized controlled trials, and reading them as if they were is the standard error made with this entire pharmacopoeia.

The obstacles, named one at a time. (1) No trial meeting modern standards has ever been run: no placebo control with a pre-registered endpoint in a performance population. (2) The 330 mg/kg rat metabolism dose is enormous relative to the 250–500 mg human doses described in practice, so the metabolite profile at ordinary exposure may differ. (3) Almost the entire record is in one language, which means it has never been through the adversarial reading that makes a literature trustworthy. (4) The claimed endpoint — work capacity under hypoxia — is the hardest kind to measure and the easiest to produce by expectation.

Bemethyl pharmacokinetics — how much of it actually gets in

The card says the half-life is not well characterized in English-language sources, which is true and is not the same as unknown. Here is what the chemistry and the two pharmacokinetic papers bound it to.

What degrades it. Not a peptidase — this is a small heterocycle, and its clearance is conjugation. The dominant route is glutathione conjugation at the thioether sulfur followed by the mercapturic acid pathway, producing the acetylcysteine conjugate that is the most abundant urinary metabolite Belinskaia 2021. Nine metabolites in six formulas means oxidative steps run alongside it, consistent with cytochrome-mediated S-oxidation of a thioether to sulfoxide and sulfone — the standard fate of that functional group. Excretion is renal, as urinary conjugates.

The oral barrier, which is the easy half. A lipophilic 178 Da benzimidazole is a textbook orally absorbed small molecule: well under the 500 Da mass ceiling, few hydrogen bond donors, unionized at intestinal pH. This is the opposite situation from the peptides elsewhere in this Vault — there is no absorption problem to argue about, and the hydrobromide salt exists to make it dissolve, not to make it cross.

The arithmetic, and why the schedule looks the way it does. Russian clinical practice as described in Oliynyk 2012 uses 250–500 mg once or twice daily in courses of days to weeks. Twice-daily dosing implies an elimination half-life on the order of hours rather than minutes — a half-life of 30 minutes would make a twice-daily schedule pointless, and one of 24 hours would make it accumulate. The dosing interval is itself evidence, and it points to something in the 4–12 hour range, which is what the volunteer study Kibal'chich 2011 was measuring. Set against that, the effect is described as taking days — and a compound whose clearance is hours and whose effect is days is telling you the effect is not the drug being present. It is what the drug caused to be built.

The injectable comparator does not exist and should not. There is no parenteral bemethyl in the described practice; a compound with unproblematic oral absorption has nothing to gain from injection and would gain a first-pass-free exposure nobody has characterized.

What would have to be true, and how you would know it was not

Three predictions, and the first comes straight out of the metabolism paper and is not in any account of this compound anywhere.

1. Chronic dosing should show up as a shift in glutathione turnover, and GGT is the enzyme that reports it. The dominant metabolite is a glutathione-derived acetylcysteine conjugate Belinskaia 2021, so every molecule cleared costs a glutathione. Gamma-glutamyl transferase is the membrane enzyme that recycles extracellular glutathione and it rises when glutathione turnover is pushed. Draw GGT as part of a CMP at baseline and at 4 weeks of continuous dosing. A rise with no other liver abnormality would be the first direct human evidence that this pathway is loaded, and a flat GGT would say the load is trivial at ordinary doses. Either answer is new.

2. The claim is that oxygen cost falls, so measure oxygen cost. An actoprotector is defined by maintained work under hypoxia Oliynyk 2012, which is a statement about efficiency, not about maximum. The test is a fixed submaximal workload with VO2 measured at that load, before and after a two-week course — not a maximal test, because a maximal test measures ceiling and the claim is about cost. If oxygen consumption at a fixed watt output does not fall, the central claim of the entire drug class has failed its cheapest test.

3. It is not a stimulant, and the resting numbers should prove it. This is the prediction that distinguishes bemethyl from what people expect of it. If the class description is right, resting heart rate and blood pressure should be unchanged, and a CBC should be unremarkable. If a user reports the jittery, appetite-suppressed, sleep-disrupting profile of a stimulant, either the class description is wrong or the vial is not what the label says — and given the supply chain, the second is at least as likely.

What nobody has tested yet

Four experiments that have never been done and that a well-equipped exercise physiology lab could do in a term.

Nobody has measured whole-blood glutathione on bemethyl. The metabolism is known Belinskaia 2021, glutathione assays are routine, and no one has connected the two in any species at steady-state dosing. This is the single most obvious missing experiment for the compound and it would take one cohort and two blood draws.

Nobody has run it against N-acetylcysteine. If the mercapturic pathway is rate-limiting at chronic doses, co-dosing NAC — a cysteine donor — should either increase tolerability or increase clearance, and those two outcomes have opposite practical meanings. A crossover with urinary conjugate measurement would separate them.

Nobody has replicated any of the human work outside the Russian-language literature. Not one study. For a compound in clinical use in one country for decades, the absence of a single independent replication anywhere else is the most informative fact about its evidence base, and it is a fact about publishing rather than about pharmacology.

Nobody has tested the hypoxia claim in the setting it was written for. The class was developed for work at altitude and under heat and hypoxic stress Oliynyk 2012. A modern altitude or normobaric-hypoxia protocol with time-to-exhaustion and blood lactate kinetics as endpoints is a standard exercise-science design, and it has never been pointed at this drug.

Bemethyl — its own safety story, not its class's

The class block above cannot be about this compound, because there is no Western drug class it belongs to. Here is what is specific.

The glutathione question is the real one and it is mechanistic. Clearance runs through glutathione conjugation Belinskaia 2021. Glutathione is also the body's primary defense against reactive metabolites — it is what acetaminophen overdose depletes and what makes that overdose lethal. Nothing suggests bemethyl at ordinary doses approaches that, and the interaction logic is nonetheless worth stating: anything else loading the same pathway at the same time competes for the same pool. Alcohol, high-dose acetaminophen, and a heavy training load all qualify. This is extrapolation from the metabolic route, not an observed interaction, and it is the sort of extrapolation that is cheap to respect and expensive to ignore.

The second specific risk is supply, not pharmacology. Bemethyl is not approved in the United States or the European Union, so every unit in circulation comes from a chain with no regulatory oversight. A benzimidazole is trivially cheap to synthesize and trivially easy to substitute with a different benzimidazole. There is no home test.

What the human literature does report. The clinical papers describe use in muscular dystrophy Lobzin 1992 and chronic respiratory inflammation Kobylianskii 1999 without prominent adverse-event reporting, which in a 1990s Soviet clinical paper is weak evidence of safety rather than strong evidence — adverse event capture was not what those studies were built for.

The honest summary. This is a drug with a fully mapped metabolism, a real human pharmacokinetic study, decades of clinical use in one country, and no controlled efficacy trial anyone outside that country can read. It is neither the mysterious nootropic it is sold as nor a validated ergogenic aid.

Sources read for this page

• Belinskaia DA, et al. Investigation of Bemethyl Biotransformation Pathways by Combination of LC-MS/HRMS and In Silico Methods. International Journal of Molecular Sciences 2021 · PMID 34445727
• Oliynyk S, Oh S. The pharmacology of actoprotectors: practical application for improvement of mental and physical performance. Biomolecules and Therapeutics 2012 · PMID 24009833
• Kibal'chich DA, et al. [Pharmacokinetics of domestic actoprotector drug Metaprot in healthy volunteers]. Eksperimentalnaia i Klinicheskaia Farmakologiia 2011 · PMID 21870773
• Boiko SS, et al. [Bemetil pharmacokinetics in an experiment on rats]. Farmakologiia i Toksikologiia 1987 · PMID 3691781
• Kobylianskii VI, et al. [Effect of bemithyl (bemactor) on the bronchial epithelium regeneration and the main protector mechanisms of the respiratory system in chronic inflammation]. Tsitologiia 1999 · PMID 10420471
• Lobzin VS, et al. [The use of bemitil in patients with progressive muscular dystrophies]. Likars'ka Sprava 1992 · PMID 1441389

Bemethyl — safety, predicted from mechanism

Bemethyl — interference & stacking

Bloodwork to run alongside Bemethyl

Baseline first, then again at 8–12 weeks.

The Basics — Start Here panel covers these in one order — 4 markers, $32.40 with the discount applied.

Check results you already have → · All 103 markers A–Z

Bemethyl — frequently asked questions

What Bemethyl is used for

Bemethyl appears under 1 goal in the goal router.

Related Performance & Endurance compounds

← Explore the full Protocol Vault

↑ Back to on this page