The short version

Plain English first. Everything below this section is the same material with the sources, the numbers and the caveats attached.

What Cartalax is

A synthetic tripeptide: alanine, glutamate, aspartate. Three residues, 333 daltons. “Cartalax” is a trade name and it is essentially absent from the scientific literature — which is the first thing anyone trying to check the claims runs into.

On the CAS number: the figure 85806-95-7 circulates widely, and we traced it to a single chemical supplier's product page where it appears as a bare identifier with no registry reference and no third-party confirmation. Against it, the compound's Wikipedia entry carries IUPAC name, formula, SMILES and InChI — and no CAS field at all, which is unusual and suggests no editor found a citable registry entry. We would not print that number on a label or a certificate until it is confirmed against the registry itself.

It is Epitalon minus one glycine, exactly

Epitalon is Ala-Glu-Asp-Gly. Cartalax is Ala-Glu-Asp. The arithmetic is clean:

Nor is the missing glycine inert. In a molecule three residues long, removing one is a third of the chain; it moves the terminal carboxylate onto aspartate and changes the charge distribution at that end. The developing group's own modelling assigns the two peptides different behaviour. Treating them as interchangeable would be wrong in both directions.

The counterion problem is nine times worse here than on a normal peptide

No vendor we checked discloses the salt form. On a 333-dalton molecule that is not a rounding error. Cartalax has exactly one basic site — the N-terminal amine — so it can carry at most one acid counterion, but which one matters enormously:

Our arithmetic from published formulae, not a finding from any source. No source we found performs it.

The organ-extract problem, which is sharper here than for Epitalon

Cartalax exists in the literature alongside a bovine cartilage polypeptide complex — a crude extract containing short peptides from 75 to 846 daltons, among them AED. Those are two different substances: a mixture of unknown composition, and one synthetic tripeptide.

The Russian manufacturer's own catalogue describes them both ways, under the same trade name. The capsules list the active ingredient by its three constituent residues — implying the synthetic tripeptide. The adjacent sublingual product lists it as “extract of cartilage tissue peptides.” Same name, same manufacturer, two different substances described on two adjacent pages.

How it is said to work

The proposal is that short peptides cross the cell and nuclear membranes, bind DNA in a sequence-specific way, and switch genes on. It is the mechanistic spine of the whole bioregulator programme, and vendors compress it to “it turns genes on.” For this compound specifically, the evidence for it is one computer simulation, in which Cartalax scores weakly.

The flagship paper is a 2016 molecular docking study of nineteen short peptides against four-base-pair DNA duplexes. Peptides flexible, DNA bases held rigid, solvent treated as a continuum. For Cartalax:

Be precise about what that is. No DNA was mixed with peptide. No binding constant was measured. No cell was treated. It is a set of computed poses and interaction energies. Docking is a hypothesis-generating device, not a measurement.

Three gaps between the model and the claim

• Nuclear delivery is assumed, not shown. The paper's own conclusion concedes the point, verbatim: “It remains unclear how this regulation is realized: whether short peptides penetrate through the cytoplasmic and nuclear membrane and then bind to DNA or they bind to receptors.” Nothing in a docking study addresses whether the molecule crosses a membrane, survives the cytosol, or reaches a nucleus.
• Four base pairs is not enough information to regulate anything. A four-base-pair site occurs roughly every 256 base pairs; the human genome contains millions of copies of that sequence. A preference at that scale cannot select a promoter.
• Cartalax is the weak case, not the strong one. The group's own model ranks it weak, three-fold below its best performer. If the peptide-DNA model is right, this is one of the peptides for which it works least well.

There is also a chemical objection. Cartalax carries a net negative charge of about −3 at physiological pH. DNA is the archetypal polyanion. Strong, specific association between two negatively charged molecules is electrostatically unfavourable, not favourable.

What is better evidenced, and what it does and does not show

Separately from the DNA claim, there is real measured data showing Cartalax changes protein and transcript levels in cultured cells. That is genuine and should not be thrown out with the mechanism. But it shows an effect, not a mechanism — nothing in it demonstrates the effect comes from direct DNA binding rather than a receptor, a transporter, amino acid supply, or an artefact of serum-free culture.

Pharmacokinetics

There is none. Not a half-life, not a distribution study, not an absorption figure, in any species, in any form.

We searched for serum and plasma half-life, peptidase degradation, biodistribution, labelled tracer studies, oral bioavailability and absorption — under all three of the compound's names, plus at family level. Nothing returned a pharmacokinetic measurement.

One figure does circulate. A vendor page states the half-life is “minutes to hours” — an unsourced restatement of the general prior above — and then adds that “biological effects persist for weeks to months through epigenetic gene regulation.” That second clause has no support in anything we found, and it is the consequential half of the sentence, because it is what licenses infrequent dosing.

Why this gap is the central problem with the whole compound

Every result in the evidence section was obtained by holding Cartalax in continuous contact with cells at a fixed concentration for days — medium replaced every three days in one study, five days in culture in another.

What the evidence shows

Eight studies, as far as we could establish. One in a living animal, and it was a kidney study. One in cartilage, and it was in a dish. None in a human being.

Note the pattern in the last two columns. Six of the eight do not state how much compound was used, and seven of the eight do not state how many samples. That is itself a finding about the quality of this literature, not an omission on our part.

The cartilage evidence, in full, because there is only one study

Primary chondrocytes from young and old rats, grown in culture, assessed at passage 4 over five days. Cartalax was tested at 20, 200 and 2000 ng/mL against the crude bovine cartilage extract.

The result is positive and we are not dismissing it. At 200 ng/mL, cell numbers rose 1.4 to 1.8-fold in cultures from young animals and 1.6 to 2.1-fold from old ones. That is a real, on-target in vitro finding.

Everything else is about other tissues, and this is what reframes the product

• Human stem cells, 2019 — the best-designed study in the set, and Cartalax underperformed in it. Ten donors, defined concentration, ten days, four peptides tested side by side for neuronal differentiation. The effect was carried by the mixture and by a different peptide. Cartalax was not the standout.
• Human stem cell line, 2020 — the headline finding runs the wrong way for a joint product. The measured Cartalax-specific result was upregulation of NF-κB, the master pro-inflammatory transcription factor. In osteoarthritis, NF-κB signalling drives the catabolic cascade that destroys cartilage. No vendor page we read mentions it.
• Human skin fibroblasts, 2020 — the collagen result, and it is the wrong collagen. The finding is that Cartalax activates synthesis of sirtuin-1, sirtuin-6 and collagen I, in skin fibroblasts. Articular cartilage is built from collagen II. Collagen I in a joint is fibrocartilaginous scar tissue. This is the result that vendor copy relabels as cartilage repair.
• Rat kidney explants, 2015 — a second loss to an organ extract. Verbatim: Cartalax and a sibling peptide “stimulate proliferation and reduce apoptosis of the kidney cells, but to a lesser degree than the polypeptide complex.”
• Old rats, 2019 — the only living-animal study, and it is about kidneys. Diuresis up, urine protein down, no nephrotoxicity observed. But dose, route and duration are not stated, and neither are strain, age or group sizes. In the only in vivo study that exists, we cannot tell you how much was given to a rat, by what route, or for how long.

The direct answers

On how these negatives were reached: by positive searching across an accessible literature, including the originating group's own publication list of 200+ articles, not by failure to search. But the main biomedical database served challenge pages throughout this session, and much of this corpus is Russian-language and non-indexed. So every negative here is “searched and not found,” not “does not exist.” Russian national trial registries were not reachable at all.

Reconstitution

No measured solubility figure for this compound exists in any source we could find.

We searched for solubility, aqueous solubility and stability under all three of its names. The entirety of the record is one chemical supplier's spec sheet saying “soluble in water or aqueous buffers” — a qualitative claim with no figure, no method and no temperature.

The 28-day rule does not transfer here

No stability data of any kind was found — not in solution, not lyophilised, not at any temperature, over any time course. No forced-degradation study, no shelf-life study, no reconstituted-stability study.

Our storage guide gives 28 days for a reconstituted peptide in bacteriostatic water, and we have a whole explainer on it. That figure is half a sterility rule and half a stability convention. The sterility half transfers here unchanged — benzyl alcohol works the same way regardless of what is dissolved in it. The stability half has never been measured for this molecule.

Anyone presenting “stable for 28 days once reconstituted” as a Cartalax-specific finding is importing generic peptide guidance wholesale. It is a house convention, not data on this compound, and we would rather say so.

Reported protocols

There is no protocol with a defensible source. What exists is a set of mutually contradictory conventions, and the contradictions are large enough to be the story.

The injection protocols disagree by 25 to 50-fold

Look at the first two rows: the same vendor gives the same 200 mcg dose as oral on one page and subcutaneous on another. A dose that is indifferent to route is a dose that was never derived from anything. And across sources the spread is 200 micrograms against 5 to 10 milligrams — a twenty-five to fifty-fold difference for the same compound and the same indication.

Why the one real concentration cannot be turned into a dose

The only figure with an actual experiment behind it is 200 ng/mL, the effective concentration in rat chondrocyte culture. A reader might try to convert it. Here is why that fails.

A 70 kg adult holds roughly 15 litres of extracellular fluid. Reaching 200 ng/mL throughout it would take about 3 mg — which lands, coincidentally, between the two circulating protocols. That calculation is worthless, and the reason it is worthless is the point. It assumes instant, complete, uniform distribution and, fatally, zero clearance. A free tripeptide is expected to be degraded in minutes. The dish was held at a fixed concentration continuously for five days; a syringe cannot reproduce that, and no pharmacokinetic study exists to say how far short it falls.

Where the vendor claims come from

Several are traceable distortions of real studies rather than inventions, which is more useful to show than to dismiss.

Published side-effect lists, contraindications and stacking guidance for this compound are all uncited, on a molecule with zero recorded human exposure and zero toxicology. A side-effect list for something never given to a person in a documented study is not a safety profile; it is a genre convention.

Storage

No stability or storage data specific to this compound was found in any primary source. What exists is commercial guidance, and it is generic peptide handling rather than a Cartalax finding — and the sources contradict each other:

The first two directly contradict each other because they are describing different products — a research powder and a food supplement capsule. That is worth knowing before taking either as guidance for a vial.

Our general conventions are in the storage guide. Freeze the solid, keep it dry and dark, and see the reconstitution section on why the 28-day figure carries no compound-specific evidence here.

Reported effects

There is no toxicology for this compound. No lethal dose study, no acute toxicity study, no repeat-dose study, no genotoxicity, no reproductive toxicity, no immunogenicity assessment. Searched under all three of its names; the results were vendor pages.

No adverse event reports were found either. No case reports, no pharmacovigilance signal, no regulatory communication naming it.

The nearest thing to a safety observation in the entire literature is the 2019 rat kidney study, which reported no nephrotoxicity — at an unstated dose, by an unstated route, for an unstated duration, with unstated group sizes. That is the whole in vivo safety record.

Two things worth flagging

• The NF-κB finding. The one well-documented gene-expression effect of this compound in human cells is upregulation of NF-κB — the central driver of the inflammatory and catabolic programme that degrades cartilage in osteoarthritis. A compound sold for joints, whose best-documented effect is on the pathway that destroys joints, deserves a sentence. We are not claiming this makes it harmful; a gene-expression change in a cultured cell line is a long way from a joint. But it is the opposite of what the marketing implies, and nobody selling it mentions it.
• An immunogenicity concern about the near-identical parent. FDA's stated concern about Epitalon is that compounded products containing it “may pose risk for immunogenicity for certain routes of administration.” Cartalax is Epitalon minus one glycine. No such assessment has been performed on Cartalax — but the regulator's concern about the molecule one residue away is on the record.

Nobody has opened a vial

We found no published analysis of the contents of Cartalax vials sold online — no mass spectrometry, no chromatography, by any regulator, academic group or independent laboratory. Searches returned marketing for testing services, not a study that bought product and analysed it.

One point on identity risk, stated fairly. Cartalax and Epitalon differ by 57 daltons in a 333-dalton molecule — a 17% mass difference, trivially resolved by any mass spectrometer. The risk is not analytical; a competent certificate distinguishes them easily. The risk is that the two sit adjacent in catalogues, adjacent in name and adjacent in the same inventories. That is a labelling and picking problem, not a chemistry one — and it is an argument for lot-specific certificates rather than against them. Ours are on the certificate register.

Regulatory status

Cartalax is absent from every peptide enforcement action and every compounding list we checked — while its siblings are named in them. That is not a safety signal. It is the signature of a compound too obscure to have been picked up.

Russia registered it — as a food supplement

This is the claim most worth getting right, because “registered in Russia” is true and misleading at the same time.

Cartalax holds a Russian state registration dated 10 April 2014. But the instrument is a state registration certificate — the mechanism for biologically active food supplements — and not a registration certificate from the Ministry of Health, which is the instrument for medicines. The manufacturer's own page carries the mandatory Russian disclaimer verbatim: “a dietary supplement is not a medicine.”

Sentient Biogenix supplies Cartalax for laboratory research use only. It is not for human or veterinary use.

Questions we actually get

What is it?

A synthetic tripeptide — alanine, glutamate, aspartate. In the literature it is written AED, and in some papers coded T-31. “Cartalax” is a trade name that barely appears in the science. It is one of the Khavinson bioregulators, a family of short peptides developed from the 1970s at a St Petersburg institute.

Is it the same as Epitalon?

No, but it is one glycine away — 57.05 daltons, exactly one residue. They are separate compounds with separate evidence bases, and the near-identity transfers nothing in either direction. Worth knowing that Epitalon's own headline human data used a crude pineal extract rather than the synthetic peptide.

Does it repair cartilage?

There is one study in cartilage cells, in a dish, unreplicated. It counted cells. It did not measure type II collagen, aggrecan, COMP or SOX9 — any of the things that would show cartilage matrix being made. And the crude extract it was compared against produced a larger effect. There is no animal model of osteoarthritis, no joint, and no human study.

Why is a review of cartilage peptides not mentioning it?

Good question, and it is the sharpest thing on this page. A 2024 review in the journal Cartilage, on peptides for cartilage regeneration, covers BPC-157, thymosin beta-4 and AOD-9604 — and does not mention Cartalax at all. That is a fair measure of where it sits in the field it is sold into.

Isn't it registered in Russia?

Yes — as an oral food supplement, in 2014, in the same regulatory class as a vitamin tablet. The manufacturer's own page carries the required line: a dietary supplement is not a medicine. Russia approved capsules. Nobody anywhere has approved an injection.

What dose should I use?

We cannot tell you, and neither can anyone else honestly. The circulating figures differ by twenty-five to fifty-fold — 200 micrograms a day against 5 to 10 milligrams — and none of them carries a citation. One vendor publishes the same number as an oral dose on one page and a subcutaneous dose on another. At the low figure a vial lasts a hundred days; at the high one it lasts two.

What's the half-life?

Unknown. No pharmacokinetic study of this compound exists in any species. A free tripeptide would be expected to survive minutes in serum, but that has not been measured here and we will not print a number for it. The claim that effects “persist for weeks to months” has no source we could find.

Does it bind DNA and switch genes on?

That is the proposal. The evidence for it, for this compound, is a single computer docking simulation in which Cartalax scored weak — three times below the model's best performer — published by the group that proposes the model. No DNA was mixed with peptide. The authors' own conclusion concedes it is unclear whether these peptides reach a nucleus at all.

Are there side effects?

Nobody knows. There is no toxicology, no adverse-event literature, and no recorded human exposure. The side-effect lists you will find published for it are uncited. Absence of reports is not a safety record when nobody has been looking.

So why do you sell it?

Because it is a legitimate research compound and researchers have their own reasons for wanting it. What we will not do is tell you it repairs cartilage. One study, in a dish, that counted cells and lost to a crude extract is not a joint-repair claim, and you should know that before you buy anything from us.

References

1. Khavinson VKh, Popovich IG, Lin'kova NS, et al. Peptide regulation of gene expression: a systematic review. Molecules 2021;26(22):7053. DOI 10.3390/molecules26227053. PMID 34834147. pmc.ncbi.nlm.nih.gov — read in full. Its Table 5 is the group's own mapping of “AED, Cartalax” to function, and is what makes this literature findable at all. Note that this document calls itself a systematic review and is written by the programme's own founder; it is not an independent appraisal.
2. Myakisheva S, Lin'kova N, Polyakova V, Ryzhak G. Peptides of cartilage tissue: regulation of chondrocyte proliferation, geroprotection and prospects for use in osteoarthrosis. Vrach 2023;(10):46–49. DOI 10.29296/25877305-2023-10-08. vrachjournal.ru — the only cartilage study that exists. Rat chondrocytes, five days, passage 4. Cell counts only. The crude extract produced the larger effect.
3. Khavinson VKh, Lin'kova NS, Tarnovskaya SI. Short peptides regulate gene expression. Bulletin of Experimental Biology and Medicine 2016;162(2):288–292. DOI 10.1007/s10517-016-3596-7. PMID 27909961. — the docking study. Read in full as a PDF from the group's own site. Cartalax graded “weak” at −3 kcal/mol against a four-base-pair duplex.
4. Caputi S, Trubiani O, Sinjari B, et al. Effect of short peptides on neuronal differentiation of stem cells. International Journal of Immunopathology and Pharmacology 2019;33:2058738419828613. DOI 10.1177/2058738419828613. journals.sagepub.com — the best-designed study in the set: 10 human donors, defined concentration, 10 days. Cartalax was not the standout. The programme's principals are co-authors — a collaboration, not an independent replication.
5. Ashapkin V, Khavinson V, Shilovsky G, Lin'kova N, Vanyushin B. Gene expression in human mesenchymal stem cell aging cultures: modulation by short peptides. Molecular Biology Reports 2020;47(6):4323–4329. DOI 10.1007/s11033-020-05506-3. PMID 32399807. — the source of the NF-κB finding, and of the IGF1 figure that vendors attribute to Cartalax alone when the paper states it for three peptides collectively. Concentrations given only as “nanomolar,” with no value.
6. Comparison of the effects of KE and AED peptides on functional activity of human skin fibroblasts during their replicative aging. Bulletin of Experimental Biology and Medicine 2020;170:154–157. DOI 10.1007/s10517-020-05022-1. — the collagen I result, in skin fibroblasts. Not collagen II, and not cartilage.
7. Chalisova N, et al. Peptide regulation of cell renewal processes in kidney tissue cultures from young and old animals. Bulletin of Experimental Biology and Medicine 2015;159(1):124–127. DOI 10.1007/s10517-015-2906-9. PMID 26033601. — the source of the T-31 code, and the first of two head-to-head losses to a crude organ extract.
8. Zamorskii I, et al. The influence of peptides on the morphofunctional state of kidneys in old rats. Advances in Gerontology 2019;9:75–80. DOI 10.1134/S207905701901017X. — the only study in a living animal, and it is about kidneys. Dose, route, duration, strain and group sizes all unstated.
9. Liao H-J, Chen H-T, Chang C-H. Peptides for targeting chondrogenic induction and cartilage regeneration in osteoarthritis. CARTILAGE 2026;17(3):660–673, published online 18 September 2024. DOI 10.1177/19476035241276406. — read in full. Covers BPC-157, thymosin beta-4 and AOD-9604. Does not mention Cartalax anywhere.
10. Health Canada. Unauthorized injectable peptide drugs seized and sold by Canada Peptide. Advisory, 1 August 2025. recalls-rappels.canada.ca — names eight Khavinson bioregulators. Cartalax is not among them.
11. Health Canada. Think twice before injecting peptides bought online. Public advisory, 9 April 2026. recalls-rappels.canada.ca — source of the “examples… many other unauthorized peptide drugs exist” qualifier quoted above.
12. US Food and Drug Administration. Certain bulk drug substances for use in compounding that may present significant safety risks. Updated 22 April 2026. fda.gov — Cartalax absent entirely; Epitalon present as nominated-but-withdrawn, with the immunogenicity concern quoted above.
13. World Anti-Doping Agency. Prohibited List 2026, section S0. — wording read verbatim from a published official copy. Cartalax is not named; the application of S0 is ours.

Listed rather than hidden, what we could not check. The main biomedical literature database served challenge pages on every attempt this session, so no query was run against it — the identifiers above come from journal landing pages and reference lists. The chemical database entry renders only under JavaScript; we read its formula from page metadata and nothing else. We could not verify the CAS number against any registry. We could not open the paper identifying this peptide inside the cartilage extract, because we could not find one. We could not reach Australian regulatory sources in either of two sessions. Russian national trial registries were not reachable at all, and much of this literature is Russian-language and non-indexed — so every negative on this page is stated as searched and not found, never as does not exist. That distinction matters more here than on any other page in this library.