B7-33

The parent hormone and its problem. Relaxin-2 rises in pregnancy and remodels connective tissue, softening it and dilating vessels. That made it an obvious candidate for fibrotic disease and acute heart failure, and serelaxin - recombinant human relaxin-2 - was developed for the latter. It did not succeed in phase 3, which is the context every B7-33 paper is written against.

Relaxin is also structurally awkward: two chains joined by three disulfide bonds, like insulin, which makes it costly to synthesise.

What B7-33 is. A single-chain derivative of the relaxin B-chain. The 2016 paper that introduced it showed something more interesting than simplification: B7-33 is a functionally selective agonist at RXFP1. Full relaxin activates both cAMP and pERK signalling through that receptor; B7-33 activates pERK - the arm that carries the anti-fibrotic effect - without cAMP [1]. Biased agonism of this kind is the respectable version of "all the benefit without the side effects", because here it was demonstrated at the receptor rather than asserted.

Animal results. B7-33 replicates the vasoprotective functions of serelaxin in vascular work [2]. In an experimental cardiomyopathy model it maintained relaxin's cardioprotective effects and reduced left ventricular fibrosis more rapidly than perindopril, an ACE inhibitor and standard comparator [3]. Subsequent chemistry has pursued further minimal derivatives [4], a lipidated version with better serum stability [5], and nanoparticle conjugates allowing prolonged activity and oral administration [6].

Human evidence. None. No trial, no pharmacokinetic study, no case series.

How to weigh it. The biology is good and the functional selectivity is a genuine finding. The cautionary note is not about B7-33's data but about the pathway: serelaxin produced encouraging early results in acute heart failure and then failed when properly tested. A more selective agonist at the same receptor inherits that uncertainty until someone runs the trial.

RXFP1 is a leucine-rich repeat G-protein-coupled receptor. Full-length relaxin engages it to raise cAMP and to activate ERK phosphorylation, and the anti-fibrotic actions - suppression of TGF-beta-driven collagen deposition, increased matrix metalloproteinase activity - track the pERK arm.

B7-33 activates that arm selectively [1]. The practical implication is a molecule that reduces fibrosis with less of the systemic haemodynamic effect that cAMP signalling contributes - relevant because serelaxin's development was partly about vasodilation in acute heart failure, and partly limited by it.

The structural achievement is separate: relaxin's two-chain, three-disulfide architecture is replaced by a single chain that folds well enough to retain receptor engagement [1][4]. The cost is stability - B7-33 is cleared quickly and degrades in serum, which is what the lipidated derivative addresses [5].

A synthetic single-chain peptide, injected in the published animal work. Follow-up chemistry has produced a lipidated derivative with improved in vitro serum stability and unchanged activity [5], and glycinated nanoparticle conjugates taken up by immune cells that enable prolonged activity and oral administration in animals [6]. There is no human formulation.

No human safety data exist, and the animal studies were designed to measure efficacy rather than toxicity.

Two mechanism-derived considerations. Relaxin remodels connective tissue throughout the body, not only in the organ of interest - that is what it does physiologically in pregnancy, including at ligaments and the pelvic girdle. Sustained pharmacological activation of that pathway has not been characterised in any species over a long period. And functional selectivity established in a receptor assay does not guarantee selectivity of downstream effect in a whole animal over months.

The most relevant safety context is that the closely related serelaxin was given to large numbers of patients in trials without a safety catastrophe but also without demonstrated benefit - so the pathway is not obviously dangerous, and is also not obviously useful.

No interaction studies exist. The one comparator relationship in the literature is with perindopril, an ACE inhibitor, against which B7-33 reduced left ventricular fibrosis more rapidly in an animal model - a comparison, not a combination [3].