B7-33 Research

B7-33 research is entirely preclinical, and every experimental B7-33 result below comes from in vitro work or from rats and mice ^{[1][2][3][4][5][6][7][8][9][11][12]}. No published study has given B7-33 to a human ^[16]. Serelaxin is a different substance, the parent hormone made as a drug, and the large human trial record belongs to it rather than to B7-33 ^[13][14].

What B7-33 is

B7-33 is a synthetic 27-residue single-chain peptide derived from human gene-2 relaxin ^[1]. It covers residues 7 to 29 of the relaxin B chain, with both cysteines replaced by serine and four residues, KRSL, added at the C-terminus ^[1]. It has no A chain and no disulfide bond, so it is neither relaxin nor serelaxin ^[1]. The design steps described in the founding paper are physical rather than pharmacological ^[1]. Solution NMR, an in vitro method, then showed that the peptide is largely unstructured ^[1].

The declared CAS number 1818415-56-3 resolves to PubChem CID 162662592, which returns the formula C131H229N41O36S and a mass of 2,986.5 g/mol ^[17]. That formula holds one sulfur atom, which matches the one methionine in the B7 to B29 sequence and no cysteine ^[1][17]. The absence of cysteine is consistent with the two serine substitutions, although a sulfur count alone cannot identify serine ^[1][17]. One caveat attaches to the identifier. Its only PubChem substance record was deposited by a peptide supplier rather than by a registry ^[17].

The B7 immune co-stimulatory proteins, including CD80, CD86 and B7-H3, are unrelated to this peptide ^[16]. A substance-registry search for the name B7-33 returns antibody-drug conjugates directed at those targets instead ^[16].

B7-33 mechanism of action

B7-33 binds relaxin family peptide receptor 1, or RXFP1, and preferentially drives ERK1/2 phosphorylation over cAMP in vitro, in cells that natively express the receptor ^[1]. That preference was measured after the peptide was made, rather than designed into it ^[1]. B7-33 is a full cAMP agonist of weak potency, not a peptide that avoids the cAMP arm ^[1]. Deliberate design for low cAMP potency begins with the same group's later analog work ^[11].

In HEK-293T cells expressing human RXFP1, B7-33 gave a pKi of 5.54 against 8.96 for relaxin, roughly 2,600-fold weaker binding in vitro ^[1]. cAMP potency in those cells was pEC50 5.12 against 10.49, roughly 230,000-fold weaker ^[1]. Weak cAMP potency also appeared in THP1 cells, which express RXFP1 natively ^[1]. The claim that B7-33 matches relaxin is an ERK1/2 result in fibroblasts, and it does not extend to binding or to cAMP ^[1]. The originating group's later paper repeats the weakness, reporting that low binding affinity made comparison of B7-33 analogs difficult ^[11]. That group had to build an engineered receptor ectodomain construct to obtain a measurable B7-33 binding curve in vitro ^[11].

Free B7-33 has an in vitro serum half-life of about six minutes ^[12]. Fatty-acid conjugation raised that to about sixty minutes, and the longer half-life belongs to the lipidated analog ^[12].

Preclinical research on B7-33

In a mouse ischemia-reperfusion model, B7-33 reduced infarct size to 21.99% against 45.32% for vehicle and preserved fractional shortening at 29% against 23%, both at P=0.02 ^[3]. That study was co-authored with the originating Melbourne group ^[3]. In mouse isoprenaline-induced cardiomyopathy, B7-33 and relaxin equivalently reduced left ventricular fibrosis, inflammation and cardiomyocyte hypertrophy, and B7-33 reduced fibrosis where perindopril did not ^[4].

In rats, an equimolar intravenous injection of B7-33 or serelaxin enhanced bradykinin-mediated endothelium-dependent relaxation in the mesenteric artery ^[2]. Neither had any overall effect in the small renal artery or the abdominal aorta ^[2]. That vascular result is vessel-bed-selective ^[2].

The same paper also ran an ex vivo arm using mouse mesenteric arteries ^[2]. Those arteries were pre-incubated in placental trophoblast conditioned media, which induces the endothelial dysfunction characteristic of preeclampsia ^[2]. Co-incubation with B7-33 or serelaxin prevented that dysfunction from developing ^[2]. That result is isolated artery tissue, not a preeclampsia model in a living animal ^[2].

In a mouse model of renal fibrosis, relaxin, B7-33 and perindopril all reduced collagen fiber thickness and cross-reticulation, while increasing collagen fiber counts ^[5]. That paper validates an imaging platform, and B7-33 is a tool compound within it ^[5].

In mice with inflammatory paw pain, B7-33 injected into the cerebral ventricles reduced mechanical and thermal sensitivity at 30 minutes and not thereafter ^[6]. An RXFP1 antagonist blocked the mechanical component but not the thermal one ^[6]. That route carries no information about B7-33 reaching the brain from the bloodstream.

Three further animal results belong to formulations rather than to the peptide alone. A coating releasing B7-33 locally cut fibrotic capsule thickness around implants in mice by 49.2% over six weeks ^[7]. Iron-oxide nanoparticle conjugates stayed anti-fibrotic when given orally to mice, and the oral route belongs to the nanoparticle ^[8]. B7-33-carrying nanovesicles inhibited tumor growth by 67.7% in a mouse cholangiocarcinoma xenograft (p<0.001), as one part of a combination construct ^[9].

Human research on B7-33

There is no published or registered human research on B7-33 ^[16]. ClinicalTrials.gov returns zero studies for the term, and the EU Clinical Trials Register returns no matching protocol ^[16]. PubMed returns 15 records for the term B7-33, of which 12 concern this peptide and none involve human subjects ^[16]. openFDA returns no label and no approval record for B7-33 ^[16]. No US or EU regulatory record for the peptide appears in the registries searched for this article ^[16].

Serelaxin is a different molecule

Serelaxin is recombinant human relaxin-2, the parent hormone manufactured as a drug ^[13]. In RELAX-AHF-2, 6,545 patients hospitalized for acute heart failure received serelaxin or placebo ^[13]. Cardiovascular death at 180 days occurred in 8.7% against 8.9% (hazard ratio 0.98, 95% CI 0.83 to 1.15, P=0.77) ^[13]. Worsening heart failure at day 5 occurred in 6.9% against 7.7% (hazard ratio 0.89, 95% CI 0.75 to 1.07, P=0.19) ^[13]. Both co-primary endpoints were missed ^[13]. Death from any cause at 180 days also showed no significant difference between the groups ^[13]. Nor did the composite of cardiovascular death or heart failure or renal rehospitalization at 180 days ^[13]. The length of the index hospital stay did not differ either ^[13]. Novartis Pharma funded the trial ^[13].

A later fixed-effect meta-analysis of six serelaxin trials reported less 5-day worsening heart failure, 6.0% against 8.1% (hazard ratio 0.77, 95% CI 0.67 to 0.89, P=0.0002) ^[14]. It also reported that serelaxin was associated with lower all-cause mortality (hazard ratio 0.87, 95% CI 0.77 to 0.98, P=0.0261) ^[14]. That mortality estimate is borderline, because its confidence interval reaches 0.98 ^[14]. It is a pooled secondary endpoint ^[14]. The analysis found no significant effect on length of stay, cardiovascular death or rehospitalization ^[14]. Four further qualifiers attach to the meta-analysis. It appeared after the confirmatory trial had already missed, and the serelaxin program's own investigators led it ^[14]. The pooled trials were coordinated by a single sponsor, Corthera, which is a wholly owned Novartis subsidiary ^[14]. Several of its authors disclose Novartis grants and personal fees during the conduct of the study ^[14].

Marketing authorization for serelaxin under the name Reasanz was refused in the European Union ^[18]. The European Medicines Agency's CHMP adopted a negative opinion on 23 January 2014 and confirmed the refusal on re-examination on 22 May 2014 ^[18]. The European Commission issued the refusal decision, which closes the file, on 5 August 2014 ^[18]. That was a refusal by the regulator, not a withdrawal by the applicant ^[18]. openFDA holds no serelaxin approval or label record ^[16].

Neither half of that record transfers to B7-33. Serelaxin's earlier positive signals are not evidence for B7-33, because no human has been given B7-33 in a published study ^[16]. Serelaxin's failure is not evidence against B7-33, because the two molecules differ in structure and in receptor pharmacology ^[1][13]. The founding B7-33 paper's abstract separately states that relaxin gained provisional FDA approval for acute heart failure ^[1]. Neither that approval nor that regulatory category exists ^[16].

Independence, replication and conflicts of interest

Eleven of the 12 B7-33 papers indexed in PubMed carry an author from the Florey Institute, the University of Melbourne or Monash University ^[16]. The twelfth is a Chinese nanomedicine study in which B7-33 is one part of a construct ^[9]. No clean independent replication of B7-33 on its own has been published ^[16]. That is a replication gap rather than a failed replication, because few groups outside the network have tried.

US patent 10,081,662 B2, covering modified relaxin B chain peptides and naming B7-33, was granted in September 2018 to the Florey Institute of Neuroscience and Mental Health ^[19]. Its named inventors include authors of the founding paper ^[19]. That paper carries no conflict-of-interest declaration and does not mention the patent ^[1]. Australian public grants funded it rather than industry, which counts in its favor ^[1].

Sanofi's published relaxin-mimetic work developed a lipidated single-chain analog rather than B7-33 itself, and that successor's subcutaneous bioavailability and extended half-life belong to it ^[15]. The originating group records that those Sanofi analogs regained cAMP potency similar to relaxin's, so the successor gave up the signaling preference that distinguishes B7-33 ^[11]. Short single-chain peptides from the H1 relaxin isoform showed no meaningful RXFP1 affinity and no relaxin-like activity in vitro, while serelaxin worked in the same assays ^[10]. Those H1 peptides are not B7-33, and that paper does not test B7-33 ^[10]. Their failure is a caution that short relaxin B-chain peptides are not a reliably active class ^[10]. The cytotrophoblast line of B7-33 preeclampsia work has no peer-reviewed paper behind it, and exists only as conference abstracts ^[16]. The only indexed peer-reviewed preeclampsia-related B7-33 result is an ex vivo mouse artery arm in the 2017 vascular study ^[2][16]. No B7-33 preeclampsia study in a living animal or in a human has been published ^[16].

Conclusion

B7-33 remains a preclinical tool compound roughly ten years after it was first described ^[1]. Its receptor binding in vitro is about 2,600-fold weaker than relaxin's, and its free serum half-life in vitro is about six minutes ^[1][12]. Its evidence base is animal and in vitro, and the group holding the patent produced most of it ^[16][19]. No human has received B7-33 in a published study ^[16].

References

1. 1

   Hossain MA, Kocan M, Yao ST, et al. A single-chain derivative of the relaxin hormone is a functionally selective agonist of the G protein-coupled receptor, RXFP1. Chem Sci. 2016 Jun 1;7(6):3805-3819. PMID 30155023. DOI: 10.1039/c5sc04754d (full text read at PMC6013806; the pKi, pEC50, design, NMR and THP1 statements are quoted from its Results and Discussion; no conflict-of-interest declaration and no mention of a patent appear anywhere in that text)
2. 2

   Marshall SA, Leo CH, Girling JE, et al. B7-33 replicates the vasoprotective functions of human relaxin-2 (serelaxin). Eur J Pharmacol. 2017 Jul 15;807:190-197. PMID 28478069. DOI: 10.1016/j.ejphar.2017.05.005 (two arms: a rat tail-vein bolus followed by wire myography in three vessel beds, and a separate ex vivo arm in virgin female mouse mesenteric arteries pre-incubated in placental trophoblast conditioned media; abstract re-read live by efetch on 2026-08-28)
3. 3

   Devarakonda T, Mauro AG, Cain C, et al. B7-33, a functionally selective relaxin receptor 1 agonist, attenuates myocardial infarction-related adverse cardiac remodeling in mice. J Am Heart Assoc. 2020 Apr 21;9(8):e015748. PMID 32295457. DOI: 10.1161/JAHA.119.015748 (publication types include "Research Support, N.I.H., Extramural"; affiliations include both the VCU Pauley Heart Center and the Florey Institute)
4. 4

   Alam F, Gaspari TA, Kemp-Harper BK, et al. The single-chain relaxin mimetic, B7-33, maintains the cardioprotective effects of relaxin and more rapidly reduces left ventricular fibrosis compared to perindopril in an experimental model of cardiomyopathy. Biomed Pharmacother. 2023 Apr;160:114370. PMID 36753958. DOI: 10.1016/j.biopha.2023.114370
5. 5

   Bhuiyan S, Kaluarachchi C, Alam F, et al. Assessment of renal fibrosis and anti-fibrotic agents using a novel diagnostic and stain-free second-harmonic generation platform. FASEB J. 2021 May;35(5):e21595. PMID 33908676. DOI: 10.1096/fj.202002053RRR (the increased collagen fiber count appears in the same abstract sentence as the reduced fiber thickness)
6. 6

   Abboud C, Brochoire L, Drouet A, et al. Analgesic effect of central relaxin receptor activation on persistent inflammatory pain in mice: behavioral and neurochemical data. Pain Rep. 2021;6(2):e937. PMID 34159282. DOI: 10.1097/PR9.0000000000000937 (route was injection into the lateral cerebral ventricle; sensitivity assessed at 30 to 120 minutes)
7. 7

   Welch NG, Mukherjee S, Hossain MA, et al. Coatings releasing the relaxin peptide analogue B7-33 reduce fibrotic encapsulation. ACS Appl Mater Interfaces. 2019 Dec 11;11(49):45511-45519. PMID 31713411. DOI: 10.1021/acsami.9b17859
8. 8

   Somanader-Livera DVN, Wang Y, Widdop RE, et al. Immune cell uptake of glycinated nanoparticles conjugated to anti-fibrotic peptides enables their prolonged activity and oral administration. J Biomed Sci. 2025 Dec 12;32(1):104. PMID 41382190. DOI: 10.1186/s12929-025-01198-8
9. 9

   Zhang L, Ji T, Li J, et al. Dual-functional nanovesicles simultaneously inhibit stromal fibrosis and angiogenesis to suppress cholangiocarcinoma progression. J Nanobiotechnology. 2025 Dec 22;23(1):781. PMID 41430305. DOI: 10.1186/s12951-025-03833-w (tumor growth inhibition reported as 67.7 +/- 17.6%; affiliations are Zhengzhou University, the National Center for Nanoscience and Technology and Tsinghua University, with no Australian author)
10. 10

    D'Ercole A, Nistri S, Pacini L, et al. Synthetic short-chain peptide analogues of H1 relaxin lack affinity for the RXFP1 receptor and relaxin-like bioactivity. Clues to a better understanding of relaxin agonist design. Front Pharmacol. 2022;13:942178. PMID 36034864. DOI: 10.3389/fphar.2022.942178 (this paper tests H1 relaxin peptides and does not test B7-33)
11. 11

    Handley TNG, Praveen P, Bathgate RAD, Hossain MA. Further developments towards a minimal potent derivative of human relaxin-2. Int J Mol Sci. 2023 Aug 11;24(16):12670. PMID 37628851. DOI: 10.3390/ijms241612670 (full text read at PMC10454739; its Introduction states the goal of low cAMP potency for the new analogs, and describes the Sanofi analogs as having regained cAMP potency)
12. 12

    Praveen P, Kocan M, Valkovic A, et al. A lipidated single-B-chain derivative of relaxin exhibits improved in vitro serum stability without altering activity. Int J Mol Sci. 2023 Apr 1;24(7):6616. PMID 37047588. DOI: 10.3390/ijms24076616 (abstract states t1/2 approximately 6 min for B7-33, raised to 60 min by fatty-acid conjugation)
13. 13

    Metra M, Teerlink JR, Cotter G, et al. Effects of serelaxin in patients with acute heart failure. N Engl J Med. 2019 Aug 22;381(8):716-726. PMID 31433919. DOI: 10.1056/NEJMoa1801291 (RELAX-AHF-2, NCT01870778; the abstract's own funding line reads "Funded by Novartis Pharma")
14. 14

    Teerlink JR, Davison BA, Cotter G, et al. Effects of serelaxin in patients admitted for acute heart failure: a meta-analysis. Eur J Heart Fail. 2020 Feb;22(2):315-329. PMID 31886953. DOI: 10.1002/ejhf.1692 (Teerlink is first author and Metra is senior author; both led serelaxin trials. Indexed publication type "Meta-Analysis". The paper carries no funding statement and Europe PMC records no grants for it. The single-sponsor sentence is the paper's own wording, naming "a single sponsor (Corthera, a whole owned subsidiary of Novartis)". Its disclosure section, read via Wiley Online Library at 10.1002/ejhf.1692, records Novartis grants and personal fees "during the conduct of the study" for Teerlink, Cotter and others. Mortality is reported as HR 0.87, 95% CI 0.77-0.98, P=0.0261, and 5-day worsening heart failure as HR 0.77, 95% CI 0.67-0.89, P=0.0002)
15. 15

    Illiano S, Poirier B, Minoletti C, et al. Characterization of a new potent and long-lasting single chain peptide agonist of RXFP1 in cells and in vivo translational models. Sci Rep. 2022 Nov 28;12(1):20435. PMID 36443381. DOI: 10.1038/s41598-022-24716-2 (Sanofi R&D, with one Florey Institute co-author; describes SA10SC-RLX, not B7-33)
16. 16

    Database and registry sweep, re-run live for this draft on 2026-08-28. PubMed E-utilities esearch for "B7-33" returns count 15; 12 of those PMIDs are B7-33 papers (30155023, 28478069, 32295457, 36753958, 33908676, 34159282, 31713411, 37628851, 37047588, 30641102, 41430305, 41382190) and 3 are string artifacts from HLA-B7 and Child-Pugh B7 records (8406520, 1633650, 23582346). Author affiliations were pulled by efetch for all 12: 11 carry a Florey, University of Melbourne or Monash author, and 41430305 does not. ClinicalTrials.gov API v2 with countTotal=true returns totalCount 0 for both query.intr=B7-33 and a quoted term search. The EU Clinical Trials Register returns "Trials with a EudraCT protocol (0)" and "Query did not match any clinical trials". openFDA drug/label and other/substance return NOT_FOUND for "B7-33"; openFDA drug/label and drug/drugsfda return NOT_FOUND for "serelaxin" under both substance_name and generic_name. The FDA GSRS substance registry does return many records for the string "B7-33", but they are CD276/B7-H3 antibody-drug conjugates from the unrelated B7 co-stimulatory family. The Pantho cytotrophoblast preeclampsia work appears only as meeting abstracts, and a PubMed search for its subject matter (marinobufagenin AND relaxin) returns no indexed full paper. That scope does not cover the ex vivo trophoblast-conditioned-media arm of PMID 28478069., which is peer-reviewed and is reported in the body. A separate esearch for "B7-33" AND (preeclampsia OR pre-eclampsia), run live on 2026-08-28, returns count 1 and that single PMID, so Marshall 2017 is the only indexed B7-33 preeclampsia paper and its preeclampsia arm is ex vivo mouse artery tissue.
17. 17

    NCBI PubChem PUG-REST, retrieved 2026-08-28. A compound-name lookup on "1818415-56-3" returns CID 162662592, and that CID's property record returns MolecularFormula C131H229N41O36S and MolecularWeight 2986.5, so the CAS and the structure round-trip. The formula contains exactly one sulfur atom, consistent with one methionine and no cysteine. The only PubChem substance record carrying this CAS was deposited by a peptide supplier, and the linked ChEMBL entry has no preferred name, so the chain of custody is weaker than for a well-registered small molecule.
18. 18

    European Medicines Agency, medicine page for Reasanz (serelaxin), EMA product number EMEA/H/C/002817, retrieved 2026-08-28. Status label: "Refused - This medicine has been refused authorisation". Applicant Novartis Europharm Ltd. CHMP negative opinion 23 January 2014; refusal confirmed on re-examination 22 May 2014; application details field reads "Refusal of marketing authorisation 05/08/2014". Under the EU centralized procedure the CHMP adopts the opinion and the European Commission issues the binding decision, so the 5 August 2014 refusal is the Commission's act.
19. 19

    US Patent 10,081,662 B2, "Modified relaxin B chain peptides". Filed 17 April 2015, priority 17 April 2014, granted 25 September 2018. Assignee: Florey Institute of Neuroscience and Mental Health. Named inventors include Mohammed A. Hossain and John D. Wade, both authors of reference 1. Legal status recorded as active. Its figure legends name the B7-33 C11.23S peptide directly. Record retrieved 2026-08-28.