Antimicrobial cyclic peptide hits acne from four directions in laboratory tests

A ring shaped antimicrobial cyclic peptide cut oil production in cultured skin cells, killed the two bacteria most associated with acne and lowered several inflammatory signals, according to a paper in Scientific Reports. The work is entirely laboratory based. The authors tested the antimicrobial cyclic peptide on human cell lines and on bacteria grown in broth, and measured how much of it crossed a barrier when mixed with two penetration helpers. Nobody with acne was treated. The paper appeared in October 2026 from a group at Jiangnan University in Wuxi working with two companies.

What the peptide did to sebum producing cells

Acne starts with sebum. The study used SZ95 sebocytes, an immortalised human cell line that behaves like the oil producing cells of a hair follicle and is the standard tool for this question. The team pushed those cells into producing more lipid in two ways, with a potent androgen and with a fatty acid, then added the peptide. Lipid synthesis fell by about forty percent in the androgen driven condition and by roughly twenty nine percent in the fatty acid driven one, at the single concentration the abstract reports. Two different triggers responding to the same treatment is a better result than one, because it suggests the peptide is not simply blocking one hormone receptor.

Alongside the oil measurements the group looked at inflammation in sebocytes and in HaCaT keratinocytes, a cell line derived from skin surface cells. Interleukin 6, interleukin 8, interleukin 1 beta, tumour necrosis factor alpha and prostaglandin E2 all fell, and the effect grew as the amount of peptide rose, which is the dose response pattern reviewers look for. The same cells were also used to check migration and the expression of differentiation genes, standard readouts for whether a compound damages the skin barrier or helps it close. None of that establishes safety on living skin, but it is the right order in which to ask.

How it fared against Cutibacterium acnes and Staphylococcus aureus

The antibacterial part is the most conventional. Cutibacterium acnes, the slow growing resident of blocked follicles, and Staphylococcus aureus, a frequent secondary coloniser, were both killed, with the concentration needed to stop growth matching the concentration needed to kill. That equivalence is a point in the peptide's favour, since a compound that only slows bacteria gives resistance more room to develop. Cutibacterium acnes was the more sensitive of the two, needing half the concentration that Staphylococcus aureus required. Standard broth tests of this kind say nothing about behaviour inside a follicle packed with keratin and lipid, where drug access is the real constraint.

Standard broth tests of this kind say nothing about behaviour inside a follicle packed with keratin and lipid, where drug access is the real constraint.

Put together, the paper claims four separate actions for one molecule, which is where the word multitarget in its title comes from. Current acne treatments divide that work between a retinoid, an antibiotic and sometimes a hormonal agent, so a single compound covering several of those jobs would be useful if it held up. The usual caution applies in reverse as well. Broad activity in a dish is common, because many peptides disrupt membranes and many membrane active compounds look busy in cell assays, so the burden shifts to showing selectivity for microbial and sebaceous targets over everything else. The available summary does not report what the peptide does to ordinary skin cells at the concentrations that kill bacteria.

The delivery problem and the surfactant fix

Peptides are poor at crossing skin. They are large by the standards of topical chemistry, they carry charge, and the outer layer of the skin is built to keep exactly that kind of molecule out. The group addressed this directly, measuring how much peptide crossed a barrier over eight hours on its own and with two common helpers, a small amount of propylene glycol and a trace of a bacterial surfactant. The combination moved about five times as much peptide across in that window. This is the most practical result in the paper, because an antimicrobial cyclic peptide that cannot reach a follicle is a laboratory curiosity rather than a candidate treatment.

It is also the result that needs the most follow up. Permeation experiments of this sort are usually run on excised skin or on synthetic membranes, and both are generous compared with intact living skin that is being washed, rubbed and exposed to sunlight. Neither the amount reaching the sebaceous gland nor the fraction still intact on arrival is reported in the material available. Skin permeation enhancers also raise their own questions about irritation, and acne skin is frequently already inflamed and treated with drying agents. A fivefold improvement on a very low starting point can still be a very low number.

Why a dish is not a face

The gap between these experiments and an acne treatment is wide and the authors say so. There were no animals, no reconstructed skin models and no people in this work, so nothing here shows that the peptide clears a single spot. Acne is driven by follicular plugging, androgen signalling, bacterial communities and inflammation acting together over months, and a compound that touches each of those in isolation may do little when they interact. Cell lines are also convenient rather than faithful. SZ95 sebocytes are immortalised and HaCaT keratinocytes are transformed, which is why both grow reliably in a dish and why neither behaves quite like skin. The resident bacterial community of a real follicle is also absent from a broth culture.

Commercial context belongs in the reading as well. Two of the affiliations on the paper are companies, one in Shanghai and one in Singapore, and the work was funded by one of them, which the authors disclose. That does not make the measurements wrong, and the paper has been through peer review at a journal that publishes technically sound work without judging importance. It does mean independent replication matters, and that the natural next steps, a reconstructed skin model and then a controlled trial on human skin, have not been taken. Until they are, this is a promising set of assays and nothing more.

Conclusion

The paper describes an antimicrobial cyclic peptide that reduces oil production, kills two acne associated bacteria, lowers inflammatory signals in skin cells and crosses a barrier better when formulated with two helpers. All of it happened in dishes, with cell lines, broth cultures and a permeation test. No animal or person was treated, selectivity and irritation were not addressed in the available summary, and company funding sits behind the work. The result is a reasonable starting point for the experiments that would actually test an acne treatment.

Source Yu N and colleagues. An antimicrobial cyclic peptide with multitarget efficacy for acne. Scientific Reports, October 2026. DOI 10.1038/s41598-026-75285-7

Amino Weekly reports research. It is not medical advice and never gives doses.