B7-33: the single-chain relaxin analogue, and what has and has not been tested

Relaxin-2, sometimes called human gene-2 relaxin or H2 relaxin, is a hormone best known from pregnancy, where it relaxes blood vessels and softens connective tissue. It also has powerful anti-fibrotic properties, meaning it works against the scarring that stiffens organs after injury, and a recombinant version called serelaxin was developed and evaluated as a treatment for acute heart failure. The problem is the molecule. Relaxin has two chains, A and B, held together by disulfide bridges in an insulin-like arrangement, which makes it awkward and expensive to manufacture. Its strong activation of cAMP signalling has also been linked to tumour-promoting effects, which is an unwelcome property in a drug meant for long-term use.

In 2016 a team led by Mohammed Akhter Hossain, Ross Bathgate and Chrishan Samuel at the Florey Institute of Neuroscience and Mental Health and Monash University reported a solution: take the B-chain alone. Their analogue, described in later papers as the single-chain relaxin mimetic, bound RXFP1 and preferentially activated the pERK pathway over cAMP in cells that naturally express the receptor, making it the first functionally selective agonist of that receptor. It prevented or reversed fibrosis and organ dysfunction in three preclinical rodent models of heart and lung disease with potency similar to relaxin, and, in contrast to relaxin, it did not promote prostate tumour growth in vivo ^[1].

That combination, a single linear chain instead of a two-chain disulfide-bonded hormone, is why the paper described the work as the first known example of minimising a two-chain cyclic insulin-like peptide to a single-chain linear peptide that keeps the useful effects ^[1]. Our guide to what research peptides are explains what the research-use label on a vial means in practice. B7-33 is often shelved alongside repair-oriented compounds such as BPC-157 and TB-500, but the relaxin peptide analogue B7-33 is a receptor-targeted analogue of a human hormone and works by a completely different route. Searches for the B7-33 relaxin connection are asking about exactly that lineage.

The B7-33 RXFP1 interaction is where the design does its work. RXFP1 is a G protein-coupled receptor that can drive several downstream pathways. Full-length relaxin activates both cyclic AMP signalling and ERK1/2 phosphorylation strongly. B7-33 was reported to shift that balance, favouring pERK over cAMP in cells that express RXFP1 naturally ^[1]. Pharmacologists call this functional selectivity, or biased agonism: same receptor, different mix of downstream signals, potentially different effects and side effects.

The anti-fibrotic mechanism the authors identified runs through receptor pairing. B7-33 activated RXFP1-angiotensin II type 2 receptor heterodimers, which in turn drove pERK1/2 signalling and matrix metalloproteinase-2, a collagen-degrading enzyme ^[1]. In plainer terms, the peptide switches on the machinery that breaks down excess scar tissue rather than simply stopping new scar being laid down.

A 2019 review of single-chain peptide agonists of relaxin receptors sets out how the design was arrived at and where the chemistry has gone since ^[2], and later work has continued to refine it: a 2023 study reported further development towards a minimal potent derivative of human relaxin-2 ^[3], and a lipidated single-B-chain derivative improved serum stability in vitro without altering activity ^[4]. Short linear peptides are cleared quickly, and most of the follow-up chemistry has been about making the molecule last longer in the body.

All of the B7-33 benefits reported so far come from rodents, isolated blood vessels and cell culture. Within those limits the findings are reasonably consistent and they cluster around two themes: fibrosis and the blood vessel wall. Most B7-33 fibrosis work has been done in the heart and the lung.

Heart, after a heart attack: a 2020 study in the Journal of the American Heart Association gave B7-33 to mice subjected to 30 minutes of coronary artery ligation followed by reperfusion. Infarct size fell from 45.32 per cent to 21.99 per cent compared with vehicle, and fractional shortening, a measure of heart pumping, was better preserved at 29 per cent against 23 per cent at 24 hours, widening to 29 against 20 per cent at seven days. In isolated adult cardiomyocytes, B7-33 at 50 and 100 nmol/L improved cell survival and reduced a marker of endoplasmic reticulum stress in an ERK1/2-dependent way ^[5].

Heart, in established cardiomyopathy: a 2023 study compared B7-33 with perindopril, a standard ACE inhibitor, in an experimental model of cardiomyopathy. B7-33 maintained the cardioprotective effects of relaxin and reduced left ventricular fibrosis more rapidly than the drug it was compared against ^[6].

Blood vessels: a 2017 study found that a single tail-vein injection of B7-33 at 13.3 µg/kg, or serelaxin at 26.6 µg/kg, selectively enhanced bradykinin-mediated relaxation in rat mesenteric arteries three hours later, by increasing endothelium-derived hyperpolarisation, with no overall effect in the small renal artery or the aorta. In mouse mesenteric arteries incubated with placental trophoblast conditioned media, a laboratory model of the endothelial dysfunction seen in pre-eclampsia, both B7-33 and serelaxin at 15 and 30 nmol/L prevented that dysfunction developing ^[7].

Kidney and implants: B7-33 has been used as the anti-fibrotic comparator in work developing a stain-free imaging platform for renal fibrosis ^[8], and coatings that release B7-33 reduced fibrotic encapsulation of implanted material ^[9]. A 2025 study attached anti-fibrotic peptides including this one to glycinated nanoparticles taken up by immune cells, reporting prolonged activity and, notably, oral administration in animals ^[10]. Pain: a 2021 study reported an analgesic effect from activating central relaxin receptors in a mouse model of persistent inflammatory pain ^[11].

There is no published clinical trial of B7-33 and no registered trial on ClinicalTrials.gov. Nothing in the human record describes what it does in a person, at any dose, by any route.

The relevant human history belongs to the parent hormone rather than the analogue. Serelaxin, the recombinant full-length relaxin-2, reached large phase 3 trials in acute heart failure and had received provisional regulatory recognition at the time the B7-33 design paper was written ^[1]. Its clinical development in heart failure did not ultimately produce an approved treatment for that indication. That history matters in both directions: it shows that the pathway B7-33 targets has been taken seriously enough to fund large trials, and it shows that promising anti-fibrotic and vasodilatory pharmacology in animals is not a reliable guide to what happens in patients.

There is one more human-relevant point in the founding paper. The reason for wanting a cAMP-sparing analogue was a concern that strong cAMP signalling through RXFP1 might promote tumours, and B7-33 did not promote prostate tumour growth in vivo where relaxin did ^[1]. That is a single animal experiment addressing a theoretical risk, not a clean bill of health.

In the vascular study, a single injection produced measurable changes in artery function three hours later ^[7]. In the heart attack model, the difference in infarct size was measured at 24 hours and the difference in pumping function widened over seven days ^[5]. The cardiomyopathy comparison ran over weeks and reported that fibrosis fell faster with B7-33 than with the comparator drug ^[6].

Those are animal timelines. No pharmacokinetic study of B7-33 in humans has been published, so nobody can say how long it circulates in a person. The fact that much of the follow-up chemistry has gone into lipidation and nanoparticle carriers to extend its activity ^{[4] [10]} suggests the unmodified peptide is cleared quickly.

These figures record what animal and cell experiments used. They are not recommendations, and no B7-33 dosage has been tested in a human being. The vascular study injected 13.3 µg/kg into the tail vein of male Wistar rats as a single bolus, chosen as the equimolar match to 26.6 µg/kg of serelaxin, which is a useful reminder that the smaller peptide needs roughly half the mass for the same number of molecules ^[7]. Isolated arteries were exposed to 15 and 30 nmol/L ^[7], and isolated cardiomyocytes to 50 and 100 nmol/L ^[5]. The cardiac and cardiomyopathy studies used repeated systemic dosing over days to weeks ^{[5] [6]}.

Note how small those doses are compared with the milligram quantities in which research peptides are typically sold. If you need to work out what a stated vial content means as a concentration, our peptide calculator and the B7-33 calculator page handle that, but a correctly calculated dose of something never tested in people is still untested.

Published work used injection: intravenous in the vascular study ^[7] and systemic dosing in the cardiac studies ^{[5] [6]}. Suppliers sell B7-33 as a lyophilised powder for reconstitution. There is no published human study of any route. The only oral data we found is the 2025 nanoparticle study, in which the peptide was conjugated to glycinated nanoparticles taken up by immune cells, which enabled oral administration in animals ^[10]. That is a result about a delivery system, and it does not mean that swallowing the plain peptide does anything.

One genuinely interesting property, from a manufacturing point of view, is that B7-33 is a single linear chain. Full-length relaxin needs two chains correctly joined by disulfide bridges, which is difficult to get right and easy to get wrong; a single chain removes that particular failure mode ^{[1] [2]}.

No human safety data exist. Nothing has been measured in a person, so any description of B7-33 side effects in humans is invention rather than reporting.

From the animal work, the closest thing to a safety signal is what the peptide does to blood vessels. It enhances endothelium-dependent relaxation in mesenteric arteries ^[7], and relaxin-pathway activation is vasodilatory by nature, so effects on blood pressure are the mechanistically plausible concern in anything given systemically. None of the published studies measured blood pressure in a conscious animal over a prolonged course.

The tumour question is unresolved rather than answered. The founding paper reported that B7-33 did not promote prostate tumour growth in vivo where relaxin did ^[1], which is encouraging, but a single model of a single tumour type is thin ground on which to say anything general. Nothing is known about immunogenicity, effects in pregnancy, effects on fertility, or interactions with medicines, and no study has examined long-term or repeated use in any species. We found no study of B7-33 and muscle growth, body composition or athletic performance; claims in that direction have nothing behind them.

B7-33 is not an approved medicine anywhere we are aware of, has no marketing authorisation and no registered clinical trial. It is sold as a research chemical. The parent hormone’s recombinant form, serelaxin, went through a clinical programme in acute heart failure without becoming an established treatment, and that programme does not confer any regulatory status on a derived analogue.

What buyers are allowed to do with research peptides varies by country: see our legal status overview and the pages for the UK, the US and Australia, the country where B7-33 was designed. Our report on the FDA advisory committee meeting on compounded peptides describes how the regulatory position on research-use peptides has been changing. We have not confirmed B7-33’s status on the WADA Prohibited List.

B7-33 is supplied as a freeze-dried powder. Peptides are generally kept cold, dry and out of the light, and reconstituted solutions refrigerated and used within a limited period. See how to store peptides and our guide to bacteriostatic water. The published chemistry work on improving the serum stability of single-B-chain relaxin derivatives ^[4] is a reminder that this class is not especially robust once it is in solution.

There is a specific problem with buying B7-33: PubChem has no record under that name, so there is no independent public reference mass to check a certificate of analysis against, and different suppliers may not be selling exactly the same sequence. The name refers to a defined region of the relaxin B-chain in the original paper ^[1], but a listing that does not state the sequence is asking you to take the label on trust. Ask for the sequence, and check that the measured mass on the certificate matches that sequence rather than matching a number the store has printed on its own page.

Compare B7-33 prices across the suppliers we track, including listings in the UK, the US and Europe. Our guides to how to read a peptide COA, third-party peptide testing and how to spot a fake peptide supplier cover what the paperwork should show, and the stores that publish independent results appear on our third-party tested suppliers list. Why near-identical vials differ so much in price is covered in peptide prices explained.