Bemethyl (Bemitil)

Identity, and nine names for one salt

PubChem CID 9816609 resolves bemethyl as 2-ethylsulfanyl-1H-benzimidazole hydrobromide, molecular formula C9H11BrN2S, molecular weight 259.17, InChIKey BKUKXOMYGPYFJJ-UHFFFAOYSA-N, UNII TG4SAK8A5M. The structure is a benzimidazole carrying an ethyl thioether at the 2-position, isolated as a one-to-one salt with hydrogen bromide. It is a small molecule, not a peptide. The free base is indexed separately as CID 720878 and ChEMBL1559140, C9H10N2S, 178.26 g/mol; no ChEMBL record was found for the salt itself. Two CAS numbers, 63513-71-3 and 109628-14-0, sit on the same FDA Global Substance Registration System entry. The record as fetched on 18 August 2026 types 109628-14-0 as PRIMARY and 63513-71-3 as NO STRUCTURE GIVEN, while the registry number the National Library of Medicine attaches to its own indexing concept is 63513-71-3. Both are carried here, because one summary quoting the first and another quoting the second are not actually in disagreement.

The naming is a practical obstacle, though a narrower one than it is usually described as. Bemethyl, bemitil, bemithyl, bemetil and bemactor, the last also spelled bemaktor, are transliterations of one Cyrillic name, and Metaprot, Bemiton, Betamyl and Antihot are trade names layered on top: five spellings and four trade names. PubMed maps all of these strings onto a single supplementary concept, so single-spelling queries run on 18 August 2026 returned 93 records for bemethyl, 96 for bemitil, 96 for bemithyl, 93 for bemetil and 96 for metaprot, against a union of 105. One spelling therefore recovers 88 to 91 per cent of the indexed set. The naming bites at the retrieval of full texts, and in secondary summaries that follow a single spelling, rather than in PubMed indexing.

Development is attributed to the pharmacology department of the Military Medical Academy in Leningrad during the 1970s, under Vladimir Vinogradov. Russian drug-reference sources describe a registration under the trade name Metaprot, and Kwiatkowska and colleagues describe the substance as the active ingredient of a dietary supplement marketed in Ukraine (PMID 30346653). Queries against the FDA drug approval database on four of its names returned no record, and no European Medicines Agency authorisation was found. It is not an approved medicine in the United States or the European Union, and nothing on this page establishes availability, legality or suitability for any person.

A hundred records, almost none of them in English

The corpus has an unusual shape. Of the 105 records, the great majority are Russian-language papers in Eksperimental'naia i klinicheskaia farmakologiia, Farmakologiia i toksikologiia, Biulleten' eksperimental'noi biologii i meditsiny and military or occupational-medicine journals, published with English abstracts only. A second tier is the English translation series of the Bulletin of Experimental Biology and Medicine, which carries roughly a dozen rodent studies in full. Genuinely international work amounts to five papers: a cirrhotic-liver study in Experimental and Toxicologic Pathology, two in Mutation Research, a Polish excretion study in Drug Testing and Analysis, and a Russian metabolite study in the International Journal of Molecular Sciences.

The figure of 105 is a raw hit count rather than a count of papers about the compound. The set includes at least one record that never administers it: a 1999 Japanese study of rubber antioxidants in rats (PMID 10496380), which carries the supplementary concept only because the indexer attached registry number 63513-71-3 to it. Several further records mention the compound in a single list, among them a 2021 review of plant adaptogens (PMID 34445021). The number of papers that actually study this substance is therefore lower than 105, and any summary quoting the raw count as a paper count overstates the literature.

That distribution has a methodological cost. An English abstract states a conclusion and sometimes a dose. It rarely states a sample size, almost never states randomisation or blinding, and does not permit anyone outside the language to check whether the analysis supports the sentence. Most of the ledger rows below therefore record sample size as absent from the indexed abstract, because that is the honest description of what is retrievable rather than a failure to look.

No registered trial exists. ClinicalTrials.gov was searched on bemethyl, bemitil, bemithyl, metaprot, ethylthiobenzimidazole, ethylsulfanyl benzimidazole and actoprotector; every one of the seven queries returned zero studies. The human papers described below predate registry practice, fall outside its jurisdiction, or both, and none has been replicated by a group outside the former Soviet Union. That combination, no registration and no independent replication and no accessible methods, is the governing constraint on everything in the next section, and it does not improve with the number of papers, because the papers share authors, institutions and journals.

What the mechanistic work actually measured

The clearest mechanistic experiment is Zarubina and Mironova's 2002 study in rats (PMID 12227091). Animals received 25 mg/kg intraperitoneally thirty minutes before acute hypoxic hypoxia, modelled by raising them in a pressure chamber to a simulated altitude of 8,000 to 11,000 metres for thirty minutes. The treated group did not show the fall in reduced glutathione, sulfhydryl groups, glutathione reductase and glutathione peroxidase activity seen otherwise, and an inhibition analysis was used to argue that the protection depended on new enzyme synthesis. The abstract describes the inhibitor, actinomycin D, as a protein synthesis inhibitor. Actinomycin D blocks transcription rather than translation, so what the experiment constrains is an RNA-dependent step.

Plotnikov and colleagues had reported a complementary negative result in 1989 (PMID 2736293). In rats given 50 mg/kg intraperitoneally, the fall in oxygen tension and the rise in diene conjugates and Schiff bases were limited under circulatory hypoxia; the abstract confines that particular result to the circulatory model, although both circulatory and hypoxic hypoxia were studied. In separate preparations the compound restricted malondialdehyde accumulation in rat brain homogenate under free-radical activation and limited organelle degradation in mitochondrial suspension, while showing no antiradical activity of its own. An antioxidant effect that operates without radical scavenging has to be enzymic or transcriptional, which is at least directionally consistent with the glutathione work and with the transcriptional pattern reported for the related actoprotector bromantane.

A finding absent from every secondary summary reviewed here concerns drug metabolism. Sorokina and colleagues reported in 2002 that single and five-fold administration at 50 mg/kg orally in rats produced mixed-type induction of hepatic cytochrome P-450, raising total P-450 content and the activity of both aryl-hydrocarbon-receptor-dependent and independent isoforms, with the explicit exception of aniline-p-hydroxylase activity, which was not induced (PMID 12227093). Induction of a major drug-metabolising system is the kind of property that governs interactions, and it appears in none of the popular descriptions.

Distribution has been mapped only coarsely. Sergeeva and Gulyaeva recovered the compound from liver, brain, kidney, spleen, heart, skeletal muscle, lung, adipose tissue and testis after oral administration to rats, with liver concentrations highest after a single dose and lower after repeated dosing, which they attributed to induced biotransformation. Their stated conclusion is that accumulation occurs in blood rather than in tissue. Boiko and colleagues had reported in 1987 that concentrations in rat brain tissue were significantly higher than in plasma, while cerebrospinal fluid concentrations in patients with craniocerebral trauma were lower than plasma; the authors read the pair as consistent with penetration of the blood-brain barrier (PMID 3609285).

The human record, and what kind of evidence it is

The largest human study located is Aleksandrovskii and colleagues' 1988 report in 130 patients with asthenic conditions of borderline severity, comparing the compound against piracetam and pyritinol. The authors described a faster onset than the comparators and a mild psychostimulant character. Neither randomisation nor blinding is described in the indexed abstract, and the abstract states that the treatment scheme was elaborated in the course of the study. No placebo arm and no blinded comparison are described, and the abstract states no route of administration.

Bobkov and Epishkin's 1988 paper is the source of the operator-performance figures that circulate. Operators performing 56 hours of continuous work under simulated spaceflight conditions were compared with placebo controls: compensatory tracking quality averaged 10 per cent higher, pursuit tracking errors were 1.8 times lower, and visual signal detection time was 2.4 times shorter. The paper is indexed as a controlled clinical trial. Its abstract reports neither a sample size nor a route of administration, and it appears in a Soviet space-medicine journal that publishes no English full text.

Two controlled studies exist outside the performance setting. Ratnikova allocated 66 patients with recurrent erysipelas to conventional treatment plus either the compound or placebo, reporting that the treated arm was free of intoxication symptoms and local manifestations sooner and was discharged earlier (PMID 1942990). Lisitsyna and colleagues gave 15 patients with systemic lupus erythematosus the compound alone and 16 the compound in combination with cyclophosphamide over eight weeks, reporting lower anti-DNA autoantibody titres and lower SLEDAI and ECLAM scores against prednisolone comparator arms of the same size (PMID 10572751).

Human pharmacokinetics does not rest on a single study. A PubMed query pairing the compound's spellings with pharmacokinetics, re-run on 18 August 2026, returns eight records, of which three are human: Boiko and colleagues in 1986 (PMID 3770166), the same group in 1991 (PMID 1679400), and Kibal'chich and colleagues in 2011 (PMID 21870773). A fourth human distribution paper from the same group falls outside that query (PMID 3609285). The two earlier Boiko abstracts report comparative rather than absolute parameters: the 1986 paper describes kinetic curves differing by clinical response in patients with asthenic states, with a larger area under the curve and a shorter half-excretion period where the psychoactivating component predominated, and the 1991 paper reports correlation between test-dose parameters and overall therapeutic effect, under monotherapy and in combination with phenazepam.

Absolute figures come from one study. Kibal'chich and colleagues gave healthy adult volunteers a single 250 mg oral dose and measured ethylthiobenzimidazole by high-performance liquid chromatography, reporting a peak concentration of 0.91 plus or minus 1.05 micrograms per millilitre at 1.06 plus or minus 0.16 hours. The abstract is internally inconsistent about the matrix, stating that the assay was run on blood serum and then reporting the peak concentration in plasma. The standard deviation on the peak concentration exceeds the mean, and the authors describe the distribution of parameters across the group as polymodal. No half-life, clearance, bioavailability figure or sample size appears in the indexed abstract.

Monitored for four years, then dropped

The anti-doping history was checked against the source documents rather than against summaries of them. The 2018 Monitoring Program lists, as its fourth entry, 2-ethylsulfanyl-1H-benzimidazole (bemitil), In- and Out-of-Competition. The same entry appears in the 2019, 2020 and 2021 programmes. The Monitoring Program covers substances that are not on the Prohibited List; the World Anti-Doping Code provision under which it operates describes substances whose patterns of use the agency wishes to observe.

Two analytical papers were published while the monitoring ran. Kwiatkowska and colleagues gave six healthy volunteers, three men and three women aged 26 to 49, six doses of a bemitil-containing supplement across three consecutive days, and collected urine before, during and for 30 days afterwards. Parent compound and glucuronide conjugate were both detectable, the conjugate more abundant and over a longer window. Belinskaia and colleagues then characterised the rodent side: five male Wistar rats per group, a single 330 mg/kg intragastric dose, 24-hour urine, nine metabolites across six molecular formulas, the most abundant a benzimidazole-acetylcysteine conjugate.

The 2022 Monitoring Program does not list it, and neither do the 2023, 2024 or 2026 programmes. Text extracted from the 2026 Prohibited List returns no match on bemitil, benzimidazole or ethylsulfanyl, while bromantan and meldonium both appear on that list by name. Hypoxen, another Russian antihypoxant, was added to monitoring in 2023 as the benzimidazole came off. The extractions were run on the published PDFs on 18 August 2026.

No public explanation for the removal was located. A substance can leave the programme because prevalence data proved uninteresting, because analytical follow-up ended, or for reasons the agency does not publish. The compound was monitored for four consecutive years, was never prohibited, and is absent from both the 2026 Prohibited List and the 2026 Monitoring Program as extracted here. The Prohibited List and the Monitoring Program are reissued annually. Secondary pages written between 2018 and 2021 and never revised describe a status that ended four years ago.

Toxicology, and a reproductive finding that reverses with duration

Acute toxicity in rats was reported by Bugaeva and colleagues in 2000 as an LD50 of 581.48 mg/kg with confidence limits of 350.17 to 965.57 (PMID 11202514). The same paper describes a qualitative inversion with dose: ergotropic effects predominate across the 20 to 80 mg/kg range and trophotropic effects at doses above 100 mg/kg. The sign of the dominant autonomic effect therefore changes somewhere between 80 and 100 mg/kg, and findings recorded on one side of that boundary cannot be extrapolated across it in either direction.

The reproductive work is the part that secondary summaries omit, and it is the part where duration changes the sign of the result. Bugaeva and colleagues gave male rats 20 or 200 mg/kg orally and reported that three days of administration raised pairing motivation and spermatogenesis at both doses, while 60 days of administration lowered male sexual activity and the spermatogenesis index (PMID 16579058). The authors attributed the reversal to effects on hypothalamic and pituitary structures. Short-course rodent data and long-course rodent data point in opposite directions.

Developmental studies exist and are small. Daily administration at 20 mg/kg to pregnant rats from gestational day 6 to 16 was reported to reduce post-implantation fetal death and raise fetal body weight, with faster development of physical condition and sensorimotor reflexes in the offspring. Administration through 16 days of lactation at 20 and 100 mg/kg transferred the compound in milk and produced uneven development in the newborns, with effects on muscle strength, sensorimotor reflex development and later pubescence.

Genotoxicity was examined early and returned negative. Seredenin and colleagues found no recessive lethal mutations in Drosophila, no dominant lethal mutations in mammalian germ cells and no chromosomal damage in murine bone marrow or human peripheral blood cultures, and reported a twofold reduction in aberrant cells induced by the alkylating agents fotrin and fopurin (PMID 3089347). Arutyunyan and colleagues reproduced an anticlastogenic effect in lymphocyte cultures from 12 healthy donors and 12 patients with urticaria, with the protection absent in the patient cultures against one of the two clastogens tested (PMID 7508559).

A second Mutation Research paper is rarely cited alongside these. Korkina and colleagues in 1992 tested the compound as one of several oxygen-radical scavengers, with rutin and ascorbic acid, against the mutagenic effect of chrysotile asbestos fibres on human lymphocytes in whole blood, and reported that the mutagenic effect was inhibited by the scavengers, rutin being the most effective (PMID 1370723). That is the only indexed human-cell result in which this compound is treated as a radical scavenger, and it sits in tension with Plotnikov's report that the compound showed no antiradical activity. Neither result has been reconciled against the other in the retrievable literature.

What is not known

The human pharmacokinetic record is thin rather than absent: three indexed human papers plus one human distribution paper, three of the four from one group, and only the 2011 study reports absolute parameters. No half-life, clearance or bioavailability figure for this compound appears in any of them; the 2011 study reports a peak concentration whose standard deviation exceeds its mean, calls the distribution across the group polymodal without explaining it, and is internally inconsistent about whether the matrix was serum or plasma. No registered trial exists in any form: seven distinct search terms returned zero ClinicalTrials.gov records. The human studies that do exist are Russian-language publications with English abstracts only, mostly without stated sample sizes and several without stated routes, none replicated by a group outside the former Soviet Union, and several published in journals with no English full text at all. The upstream molecular target has not been identified, so what is described as a mechanism is a sequence of downstream measurements: preserved glutathione enzymes under hypoxia, absent antiradical activity in one rat study against radical-scavenging behaviour in one human-lymphocyte study, and induced hepatic cytochrome P-450. The interaction consequences of that P-450 induction have not been studied in humans. Repeated-administration toxicology is limited to a small rodent reproductive series in which the sign of the effect reversed between three days and sixty. No carcinogenicity study was located. The reason the World Anti-Doping Agency ended four years of monitoring in 2022 is not published, and this record does not know it.

Questions