SS-31 and Elamipretide: One Tetrapeptide, Several Names

SS-31, MTP-131, Bendavia and elamipretide are one peptide. Its structure, where each name comes from, its trial record and 2025 FDA decision.

Published 28 October 2026

SS-31, MTP-131, Bendavia and elamipretide are four names for one molecule, and DrugBank adds a fifth, RX-31. They are not different grades or versions. They are the names the same four-residue peptide has carried as it moved from an academic laboratory, through a company's development programme, to an international non-proprietary name. This page sets out what the molecule is, where each name comes from, what research has been done with it, and what happened to it at the US regulator in 2025, since that is the part most often misreported.

The molecule#

The free-base formula, weight (639.802), monoisotopic mass and CAS number are from DrugBank entry DB11981. The sequence and the trihydrochloride figures (C32H49N9O5·3HCl, 749.2) are from the Description section of the US prescribing information, which names the compound D-arginyl-2,6-dimethyl-L-tyrosyl-L-lysyl-L-phenylalaninamide. We checked the free-base arithmetic from the sequence ourselves.

Two features make it unlike most peptides in a catalogue. The first residue is D-arginine, the mirror-image form of the amino acid, not the one proteins are built from. The second is Dmt, a tyrosine carrying two extra methyl groups, which is not one of the twenty standard amino acids at all. So writing SS-31 in one-letter code as "RYKF" throws away exactly the information that defines it. Arg-Tyr-Lys-Phe-NH2 with ordinary residues is a different compound, about 28 daltons lighter (611.8).

The arrangement alternates aromatic and basic residues, and the peptide carries a net positive charge of three at neutral pH. That motif was deliberate: it is what the designers built the series around.

Where the names come from#

SS-31. "SS" stands for Szeto–Schiller, after Hazel Szeto of Weill Cornell Medicine in New York and Peter Schiller of the Montreal Clinical Research Institute (IRCM). Schiller's laboratory worked on opioid peptide chemistry, including the Dmt-containing peptide [Dmt1]DALDA. The IRCM's own account says the SS peptides were co-designed with Szeto and first synthesised in Schiller's laboratory (IRCM). Szeto has described noticing that one compound from her opioid receptor work "did not behave like I thought it would" and was acting on mitochondria (Cornell Chronicle, 2011). SS-31 is one number in that series; others, such as SS-02 and SS-20, are different molecules.

MTP-131 and Bendavia. Szeto founded a company, Stealth Peptides, in 2006 to develop the work (later Stealth BioTherapeutics). MTP-131 was its development code, and Bendavia the name used in its early cardiac trials. The 2011 Cornell article above calls it Bendavia.

Elamipretide. The international non-proprietary name, used in all later trials and regulatory documents. When you see "elamipretide hydrochloride", that is the salt form used in the US-approved product, which is why its label weight (749.2) differs from the free-base figure (639.8).

Research material is often supplied as an acetate or trifluoroacetate salt instead, which changes the weight again. None of these differences changes the peptide, but each changes the number on paper. A certificate should say which form it describes.

The research literature#

The SS-31 literature runs to several hundred papers, the large majority preclinical.

Origins and mechanism. Szeto's 2006 review in The AAPS Journal summarises the early work: small peptides with alternating aromatic and basic residues that the authors reported to concentrate more than a thousandfold in the inner mitochondrial membrane, with the tyrosine or Dmt residue described as the source of their antioxidant activity (Szeto, 2006). The explanation shifted over the following decade. A 2013 study in rats subjected to kidney ischaemia reported that SS-31 binds cardiolipin, a lipid found almost only in the inner mitochondrial membrane, and inhibits the peroxidase activity of cytochrome c bound to it (Birk et al., 2013, J Am Soc Nephrol). Most later papers describe it as a cardiolipin-targeted compound rather than a simple antioxidant.

Animal and cell models. The preclinical work covers ischaemia–reperfusion models in kidney and heart, neurodegeneration models, and many cell-culture studies of mitochondrial function. One early example tested SS-31 in neuronal cells and in the G93A mouse model of amyotrophic lateral sclerosis (Petri et al., 2006, J Neurochem).

Human clinical trials. Elamipretide has been through a substantial clinical programme. Several large trials did not meet their primary endpoints, which matters for anyone reading the literature:

• Heart attack (EMBRACE STEMI). A phase 2a randomised, placebo-controlled trial in patients undergoing primary angioplasty for anterior ST-elevation myocardial infarction, run under the name MTP-131. The primary endpoint, an enzyme-based measure of infarct size, did not differ significantly from placebo (Gibson et al., 2016, Eur Heart J).
• Heart failure. A small randomised, placebo-controlled study in heart failure with reduced ejection fraction, measuring safety and heart-chamber volumes after a single administration. Its authors called for longer studies (Daubert et al., 2017, Circ Heart Fail).
• Primary mitochondrial myopathy (MMPOWER-3). A phase 3 randomised, placebo-controlled trial in 218 adults over 24 weeks. Neither primary endpoint, a six-minute walk test and a fatigue score, was met (Karaa et al., 2023, Neurology).
• Dry age-related macular degeneration (ReCLAIM-2). A phase 2 randomised, placebo-controlled trial in 176 participants with geographic atrophy over 48 weeks. Neither primary endpoint reached statistical significance (Ehlers et al., 2025, Ophthalmology Science).
• Barth syndrome (TAZPOWER). Barth syndrome is a rare inherited disorder of cardiolipin metabolism. TAZPOWER randomised 12 participants in a placebo-controlled crossover design, and neither primary endpoint was met in that part (Thompson et al., 2021, Genet Med). An open-label extension, without a placebo comparison, followed 10 participants, 8 of whom reached 168 weeks (Thompson et al., 2024, Genet Med).

What happened at the FDA#

This is where accounts online most often go wrong in both directions, so here it is plainly.

On 15 May 2025 the US Food and Drug Administration issued a complete response letter, meaning it declined to approve the application as it stood. The company resubmitted in August, and on 19 September 2025 the FDA granted accelerated approval to an elamipretide product (NDA 215244) for patients with Barth syndrome who weigh at least 30 kg. Both dates are recorded in the FDA approval letter.

Accelerated approval is a conditional route. The FDA based it on a measure of knee-extensor muscle strength, an intermediate measure it considered reasonably likely to predict benefit to patients. It is not a finding that such benefit has been confirmed. The letter requires a randomised, double-blind, placebo-controlled trial to verify it, and continued approval may depend on the result. That confirmatory study, 4TAZPower, is a 72-week randomised, placebo-controlled trial in Barth syndrome. It received a favourable UK research ethics opinion in February 2026 and lists the UK among its intended locations (HRA research summary). Running a trial here is not the same as being licensed here.

In the UK, as far as we can establish, elamipretide is not a licensed medicine. NICE's draft scope for an appraisal in Barth syndrome, issued in March 2026, records that it "does not currently have a marketing authorisation in the UK" (NICE). The MHRA agreed a paediatric investigation plan in January 2026 (MHRA decision), which is a step in a development programme, not a licence. We have found no UK marketing authorisation since. In the EU it holds an orphan designation for a different mitochondrial condition, and the EMA notes that orphan designation "does not mean the medicine is available or authorised for use."

One more distinction matters. A US approval covers one manufacturer's product, made to its own specification and salt form. A vial of research-grade SS-31 is the same molecule on paper, but it is not that product.

Checking identity#

As with any peptide, the useful figure on a certificate is the mass:

• Near 639.8 (or 640.4 for the protonated ion in a mass spectrum): SS-31 free base.
• Near 611.8: ordinary tyrosine in place of Dmt, which is a different compound.
• No mass: you cannot tell.

One limit is worth knowing. Mass spectrometry cannot tell D-arginine from L-arginine, because mirror-image forms weigh the same. Confirming the D-residue needs a chiral method or the manufacturer's synthesis records. See how to read a certificate of analysis, what an HPLC purity figure means and how to verify a certificate is genuine. Ask about batch documentation before ordering.

Handling#

SS-31 is supplied lyophilised and is a small, water-soluble peptide. The general guidance applies; see how to reconstitute research peptides and how to store research peptides.

What this page does not claim#

Nothing here concerns what SS-31 does in a person or animal outside the studies cited. The trials are listed as history: which conditions were studied, how, and whether the trials met the endpoints their designers set. For a similar case of one name covering more than it should, see TB-500: two different substances, one name. For how UK law applies to material sold for research, see what "research use only" actually means.

In our catalogue#

• SS-31 (elamipretide): D-Arg-Dmt-Lys-Phe-NH2, 639.8 g/mol free base

Supplied for laboratory research use only. Not for human or veterinary use.

Pioneer Research Peptides — UK supplier of research peptides for laboratory use. Ask about batch documentation before you order.

For laboratory research use only — not for human consumption. 18+.