Cardiogen (AEDR)

Three different substances answer to the name

The string Cardiogen resolves to three unrelated things across the primary databases, and only one of them is a peptide. A search of the FDA Global Substance Registration System returns exactly one record: rubidium chloride Rb-82, UNII F0Z746KRKQ, classified under diagnostic radiopharmaceuticals. That is the isotope eluted from the CardioGen-82 generator used for myocardial perfusion imaging, a device with its own recall history and its own literature. A fourth database lands in the same place: the NCI chemical structure resolver, queried on the bare string cardiogen, returns a structure, and the structure it returns is rubidium chloride Rb-82, CAS 132486-03-4.

PubChem returns something else again. A name query there returns CID 10917, L-carnitine, which carries Cardiogen among its several dozen trade synonyms. The tetrapeptide is in PubChem, but under no name. It sits at CID 11583989 with a single synonym attached: SCHEMBL3194515, an identifier generated automatically by SureChEMBL when a structure is extracted from patent text. UniChem returns two sources for its InChIKey, SureChEMBL and PubChem, and nothing else. ChEMBL has no entry. The molecule reached the chemical databases as a drawing lifted out of a patent specification, and nothing has been added to that record since.

Collisions of this kind stop being trivia when material changes hands. CAS 406-76-8, which belongs to DL-carnitine (PubChem CID 288, C7H15NO3, molecular weight 161.20), is printed alongside the word Cardiogen on raw-material marketplace listings, so a 161 g/mol quaternary ammonium compound and a 489 g/mol tetrapeptide are offered under one heading, and the identifier printed there points to the former. Nothing in that arrangement distinguishes the two for anyone reading the listing, and the number is the only machine-checkable thing on it.

What the chemical record supports, and what it does not

PubChem gives C18H31N7O9 and 489.5 g/mol for CID 11583989, with the IUPAC name describing an all-S chain: alanine, then glutamic acid linked through its alpha carboxyl, then aspartic acid, then arginine. The composition is not in dispute. The patent that first disclosed the sequence states 489.48 for the same formula. Its analytical section reports peptide content of 98.01% by reversed-phase HPLC, amino acid ratios of Glu 1.08, Asp 1.08, Ala 1.00 and Arg 1.07 after 24-hour hydrolysis, a TLC Rf of 0.59, moisture of 8%, a pH of 4.58 and a specific rotation of minus 24 degrees. It reports no melting temperature; the finished tetrapeptide is described twice as a white amorphous powder, and the single figure of 154 degrees Celsius in the specification belongs to a benzyl-protected tripeptide intermediate three steps earlier in the synthesis.

PubChem carries no CAS registry number on the entry, and neither does GSRS or UniChem. No UNII has been assigned. The compound is in neither the European nor the Japanese pharmacopoeia. A CAS number, 857267-11-9, circulates on vendor product pages and aggregator listings; it returns nothing in PubChem, produces no structure from the NCI chemical resolver, which answers both that number and the sequence Ala-Glu-Asp-Arg with a server error, and appears nowhere outside that commercial layer. The absences are the more useful part of this record. The figure is untraced, and it is recorded below in those terms and not repeated in the identity block.

The sequence itself places the molecule in a family. Several peptides from the same programme share an AED- opening: AEDG, also called epitalon, AEDL, AEDP and the tripeptide AED. Cardiogen differs from those by its C-terminal arginine, the residue that gives it a net positive charge at that end. That structural neighbourhood is the source of a good deal of borrowed evidence, since findings reported for one member of the series are routinely summarised as though they described the group.

The cardiac evidence base is a patent

Every cardiac number attached to this peptide originates in one document. Russian application RU2255756C1 carries a priority date of 22 June 2004. The PCT filing, WO2006001728A1, was lodged on 7 December 2004; the European grant is EP1758923 and the United States grant is US 7,662,789 B2, issued 16 February 2010 to Khavinson, Ryzhak, Grigoriev and Ryadnova, assigned first to Access Bioscience CJSC. Google Patents lists the legal status of the US grant as expired for non-payment of fees. Nine worked examples fill the specification, and they are the whole preclinical record.

Example 2 is the one the secondary literature repeats. Forty white mongrel rats weighing 180 to 200 grams were randomised into two groups of twenty. Infarction was induced by ligation of the left coronary artery, and the tetrapeptide was given intraperitoneally at 0.05 micrograms per animal at one, three and five hours after occlusion, with saline as control. Mortality in the first 24 hours was 45% in control and 15% in treated animals. The specification also reports blood glucose lowered by roughly 1.5-fold, glycogen preserved where controls fell threefold, and a mitochondrial energy-efficiency coefficient of 93.2 plus or minus 2.3 against 41.5 plus or minus 2.4.

Other examples run the same design across other injuries. Isolated hearts from 60 male guinea pigs were perfused by Langendorff in groups of six, with 30 minutes of total ischaemia; at 0.005 micrograms per millilitre the contraction amplitude recovered to 75 plus or minus 12 percent of baseline against 43 plus or minus 7 percent in control. Three groups of eight isolated rat hearts were used for the strained-contraction series. Adrenalin dystrophy was studied in 30 rats, calcium chloride arrhythmia in 60 rats across six groups, and thiuram-disulfide myocardiopathy in 40 rats across four.

Example 9 covers toxicity: 66 mice given a single intramuscular dose, 60 rats dosed daily for 90 days, and 96 male guinea pigs dosed daily for six months. The specification states that no toxic reaction followed a single dose 5000 times the amount it describes as therapeutic, and that long-term administration at 100 to 1000 times that amount produced no side effects it records. It also reports a rise in guinea pig leukocyte counts at three and six months, which it does not account for. A patent specification is written to support a claim of invention. It is not peer reviewed, its tables are not deposited, and no independent group has published any of this work.

What PubMed indexes

A title-and-abstract search for cardiogen returns 19 records. Fifteen are false matches: seven concern the CardioGen-82 rubidium generator or rubidium chloride Rb-82 itself, one is the Szeged CardioGen cardiomyopathy and ion-channel registry, one is a French paper on cardiogenetics, and the rest are Hungarian, Austrian and Polish papers in which the string is matched through cardiogenic. Four describe the peptide, and all four come from the originating programme in St Petersburg. None reports a sample size in its abstract.

Zakutskii and colleagues in 2006 applied cardiogen, bronchogen, prostamax and pancragen to organotypic explant cultures from the heart, lung, prostate and pancreas of three-week-old and 18-month-old rats, and reported a stimulating effect at 0.05 nanograms per millilitre in the matching tissue. Chalisova and colleagues in 2009 compared the tetrapeptide against all 20 individual amino acids at 10 to the minus 12 molar in myocardial explants from three-month and 24-month rats. Seven of the amino acids stimulated proliferation in young tissue and two in aged tissue; the peptide was reported active in both, with reduced p53 expression on immunohistochemistry.

Two results sit awkwardly with the tissue-specificity framing. Levdik and Knyazkin in 2009 transplanted M-1 sarcoma into senescent rats and reported dose-dependent inhibition of tumour growth, with raised apoptosis and haemorrhagic necrosis, and specifically noted that the effect was not a direct cytostatic action on the tumour but appeared to run through its vascular network. Kheifets and colleagues in 2010 applied cardiogen alongside two other short peptides to ageing human prostate fibroblast cultures and reported increased expression of CXCL12, WEDC1 and ghrelin in all of them.

A fifth paper names the compound only by its sequence. Khavinson and colleagues in 2012, in Bulletin of Experimental Biology and Medicine, cultured mouse embryonic fibroblasts treated with the tetrapeptide and reported actin, tubulin and vimentin expression raised two- to fivefold and lamins A and C raised two- to threefold. The abstract states that this explains the previously reported cardioprotective activity of the peptide. The previous report it refers to is the patent.

How a patent became a citation

In 2022 a review in Cells summarised the compound in four sentences. It stated that AEDR was detected in the polypeptide complex of the heart by highly sensitive chromato-mass spectrometry, that administration in experimental infarction produced a threefold decrease in mortality, that necrotic zones shrank and glycogen was preserved, and that mitochondria were protected. One citation carries all of it, and that citation is US 7,662,789. The mortality figures line up: 45% falling to 15% is threefold. The mass-spectrometry sentence does not. The patent describes the tetrapeptide as obtained using the classical method of peptide synthesis in solution, and reports no detection of it in cardiac tissue by any method.

The Cells review and the International Journal of Molecular Sciences review, published by the same group in the same year, give the patent two different issue dates. Cells cites it as issued 16 October 2010; the IJMS paper cites 16 February 2010. The USPTO date is 16 February 2010, which was a Tuesday, the day United States patents issue; 16 October 2010 was a Saturday. The two reviews disagree about the same document, and the shorter of the two summaries is the one that circulates.

Two other strands attach to the sequence rather than the trade name. Fedoreyeva and colleagues in 2013 measured fluorescence quenching of FITC-labelled histones and reported that Ala-Glu-Asp-Arg, along with five sibling peptides, binds histones H1, H2B, H3 and H4. The histones were from wheat. Khavinson and colleagues in 2023 docked 26 ultrashort peptides against transporter structures and returned ICM scores for AEDR of minus 35.35 at LAT1, minus 27.54 at LAT2 and minus 27.50 at PEPT1, with a positive and therefore unfavourable 13.15 at PEPT2. That work is computational throughout, and the authors describe it as a feasibility assessment.

Registered trials and regulatory footprint

ClinicalTrials.gov holds nothing. A query on cardiogen returns four studies, none of them of the peptide; the closest is a 2011 evaluation of strontium exposure in patients scanned with the rubidium generator. An intervention-field query on cardiogen returns that same study. AEDR returns zero. A control query on Ala-Glu-Asp-Arg returns 25 unrelated amino-acid and nutrition studies, which confirms the endpoint was answering. No dose-ranging, pharmacokinetic, safety or efficacy study in people has been registered.

No GSRS entry exists for the peptide, which means the substance has never been through the identification process that assigns a UNII. It was not among the seven peptides considered by the FDA Pharmacy Compounding Advisory Committee on 23 and 24 July 2026, which covered BPC-157, KPV, TB-500, MOTS-c, emideltide, semax and epitalon, nor is it among the five scheduled for review before the end of February 2027. It is not an approved medicine in any jurisdiction, and the patent that describes it has lapsed.

Material described by this name circulates outside any registered trial. Russian-language distributor copy presents it as a food supplement; elsewhere it appears in research-chemical catalogue listings that print a molecular weight, a purity figure and a CAS number no registry recognises. None of that material has been subject to identity, purity or sterility controls, and the compound is not approved for human use in any jurisdiction. Claims of institute studies in elderly patients showing improved ejection fraction over twelve months appear on aggregator pages with no citation attached, and no such publication exists in PubMed. The distance between what the patent examples report and what is asserted downstream is the whole subject of the ledger below.

What is not known

No human data exists in any form. ClinicalTrials.gov searches on cardiogen and AEDR, run as general terms and against the intervention field on 18 August 2026, returned no study of the peptide, so nothing has been registered about dose-ranging, pharmacokinetics, safety or efficacy in people, let alone reported. Absorption, distribution, metabolism and clearance are uncharacterised in every species: no pharmacokinetic study was located in PubMed or Europe PMC, no half-life has been measured, and the relationship between an administered amount and any tissue concentration is unestablished. Immunogenicity has not been assessed. The toxicity work that exists is inside the patent, was conducted by the applicant, and reports an unexplained rise in guinea pig leukocyte counts at three and six months that no follow-up addresses. Reproductive and developmental effects, carcinogenicity and chronic administration beyond six months are unstudied. No sample size is stated in any of the retrievable abstracts of the PubMed-indexed papers used here; the full texts of the Advances in Gerontology and Bulletin of Experimental Biology and Medicine articles could not be obtained, so whether the underlying papers report an n is unestablished rather than known to be absent. The proposed mechanism rests on histone binding measured in wheat and on transporter affinities that were computed and never measured; no experiment has shown the peptide entering a mammalian cardiomyocyte. Nothing in the record has been reproduced by a group unconnected to the originating programme, and the cardiac findings have never been through peer review at all. Cardiogen is not an approved medicine in any jurisdiction, appears in no major pharmacopoeia, has no FDA substance registry entry, and material sold outside a registered trial has not been subject to identity, purity or sterility controls.

Questions