Research Lab Review

LL-37 and rosacea: how a skin defense peptide may drive inflammation

News · LL-37

New reviews, mouse work and a 204-patient trial link the skin peptide LL-37 to rosacea inflammation via mast cells, but direct tests in patients are missing.

LL-37 is an antimicrobial peptide the skin makes to help fight microbes. Two 2026 narrative reviews pull together human tissue, cell and animal evidence suggesting it may also help drive the redness and inflammation of rosacea, possibly through mast cells and a receptor called MRGPRX2 [1, 2]. A new 204-patient randomized trial adds a data point by measuring LL-37 in blood [3], but no study has yet directly tested this pathway in people with rosacea [1].

Key takeaways

  • A 2026 review reports that human rosacea skin shows changes in mast cells, and that MRGPRX2 is found on mast cells in lesions [1].
  • In human mast cells studied in the lab, LL-37 can activate MRGPRX2 and trigger degranulation, the release of inflammatory contents [1].
  • Much of the support comes from mouse models, and the mouse receptor behaves differently from the human one [1].
  • A randomized trial of 204 patients found that serum LL-37 fell more in its active group than in its sham group. This shows an association, not proof that LL-37 or MRGPRX2 causes symptoms [3].

3-D fluorescence image of bioprinted skin tissue with dermis, capillaries and a top epidermis layer

What LL-37 and MRGPRX2 are

LL-37 is the active form of cathelicidin, the human antimicrobial peptide. Researchers often use it to produce rosacea-like inflammation in mice [1, 4]. Mast cells are immune cells found in the skin. They hold granules full of signaling molecules, and they release these when activated.

MRGPRX2 is a receptor on human mast cells that responds to certain peptides, including LL-37 [1]. Mice have a related receptor, MrgprB2. The 2026 review stresses that the human and mouse receptors are pharmacologically distinct, so a result with one does not automatically carry over to the other [1].

The 2026 mechanism review

The main source here is a critical narrative review published in Clinical, Cosmetic and Investigational Dermatology [1]. The authors searched PubMed and Web of Science and also checked the reference lists of key papers [1]. A narrative review summarizes and interprets studies. It does not pool data the way a systematic review or meta-analysis does, so how studies were chosen and weighed depends partly on the authors’ judgment.

The review groups the evidence into several layers [1]:

  • Human rosacea tissue. Studies report changes in how many mast cells are present, or in staining linked to degranulation. They also place MRGPRX2 on mast cells within lesions [1].
  • Human mast cells and receptor studies in the lab. LL-37 triggers calcium signaling through MRGPRX2 and leads to degranulation. Blocking MRGPRX2 with drugs reduces these responses [1].
  • Mouse models. In LL-37-induced models, mast cells and the mouse receptor MrgprB2 appear to contribute to inflammation [1].

The authors are clear about the gaps. Other LL-37-responsive pathways and downstream regulatory steps may produce similar mast-cell effects, so MRGPRX2 may not be the only route [1]. Direct, selective testing of MRGPRX2 function in patients with rosacea is still lacking [1]. The receptor-level findings and the patient-level observations also come mostly from different study settings [1]. In other words, the pieces fit together, but nobody has yet shown all the steps in the same people.

The neurogenic rosacea review

A second 2026 narrative review, in Medicina, looks at “neurogenic rosacea.” The term describes a presentation dominated by burning, stinging and other facial sensations, rather than mainly redness or bumps [2]. The authors searched several databases through mid-2026 and sorted the evidence by strength: phenotype-specific human evidence, broader human rosacea evidence, case reports, and animal or cell studies [2].

They found that direct evidence for neurogenic rosacea as its own presentation comes mainly from small observational series and case reports [2]. Human rosacea studies support a role for nerves and blood vessels. However, many proposed links, including the LL-37-MRGPRX2-mast cell pathway, remain indirect or experimental [2]. No validated diagnostic criteria exist for this presentation, and the authors say current evidence does not establish it as an independent subtype [2]. This review places the LL-37 story in a wider picture of nerve, vessel and immune interactions, and it agrees with the first review that human confirmation is missing [1, 2].

A randomized trial that measured serum LL-37

The newest human data come from a single-center, randomized, sham-controlled, assessor-masked trial published in JAMA Dermatology [3]. It enrolled 204 adults with rosacea who were already receiving standard first-line care, with 102 in each group [3]. Participants were mostly women (84.3% in each arm), with an average age in the early 30s [3].

The trial tested auricular acupressure, meaning pressure on points of the outer ear, as an add-on to usual care [3]. The active group used a site on the ear supplied by a branch of the vagus nerve, and the sham group used a site with only ordinary skin sensation [3]. The intervention lasted 8 weeks [3].

At week 8, 56.9% of the active group reached the main goal, a near-clear or clear rating on an investigator scale, compared with 35.3% of the sham group [3]. Serum LL-37, along with the nerve-related peptide CGRP and the inflammatory signals TNF-α and IL-6, fell more in the active group [3]. The authors reported no serious adverse events [3].

This trial was not designed to test the LL-37-MRGPRX2-mast cell pathway. It measured LL-37 in blood, not in skin, and it did not measure MRGPRX2 or mast-cell activity [3]. A drop in serum LL-37 alongside clinical change is an association. It does not show that LL-37 drives symptoms or that lowering it causes the change. Other limits include a single site, a mostly young female group, and the difficulty of masking participants to a hands-on intervention [3]. The results need independent replication.

Mouse evidence: paeonol in an LL-37 model

A 2026 study in Frontiers in Immunology shows how LL-37 is used in lab research [4]. Researchers injected LL-37 into mice to create rosacea-like skin inflammation, then gave paeonol, a compound extracted from the root bark of tree peony [4]. They first used bioinformatics to predict overlap between rosacea-related genes and the compound’s targets [4].

In these mice, paeonol was linked to less severe rosacea-like lesions [4]. It was also linked to lower levels of S100A9, an inflammatory protein, and changes in LL-37 and NF-κB signaling [4]. The abstract does not report how many animals were used.

This is an animal study, and its results cannot be applied to people. LL-37 injection mimics some features of rosacea but is not the human disease [1]. The study also did not focus on MRGPRX2, so it supports LL-37’s role as a trigger in mice more than it supports the specific mast-cell receptor pathway [4].

What we still don’t know

  • Whether MRGPRX2 matters in patients. No study has selectively blocked or measured MRGPRX2 function in people with rosacea and tracked the effect on mast cells and symptoms [1].
  • How much the mouse data apply. Human MRGPRX2 and mouse MrgprB2 differ pharmacologically, so mouse findings may not carry over [1].
  • Whether other pathways explain the same effects. LL-37 acts on more than one pathway, and some of these could produce similar mast-cell responses [1].
  • What serum LL-37 means. Blood levels may not reflect what happens in skin lesions, and the trial measured only two time points [3].
  • Which patients this applies to. Rosacea varies, and the role of LL-37 may differ between presentations such as sensation-dominant forms [2].

The first review suggests that future studies should bring the steps together in the same patients: check MRGPRX2 status in lesions, selectively perturb the receptor, and measure mast-cell responses alongside tissue and clinical changes [1]. Matching results would strengthen the case. Partial or mixed results would help define how far the pathway’s role extends [1].

References

  1. LL-37–MRGPRX2–Mast Cell Signaling in Rosacea: Current Evidence and Directions for Human Studies. Clinical, Cosmetic and Investigational Dermatology (2026). doi:10.2147/CCID.S648503
  2. Neurogenic Rosacea and Facial Dysesthesia: Neurovascular–Immune Mechanisms and Translational Therapeutic Perspectives. Medicina (2026). doi:10.3390/medicina62091689
  3. Device-Free Auricular Acupressure for Rosacea. JAMA Dermatology (2026). doi:10.1001/jamadermatol.2026.3999
  4. Ameliorative effects and mechanisms of paeonol against rosacea through network pharmacology and experimental validation. Frontiers in Immunology (2026). doi:10.3389/fimmu.2026.1949398

Image credits

Free to use (CC0 / public domain); credited as a courtesy.

  • Mastocyte.jpg, by Makysm (CC0, via wikimedia-commons): source
  • 3-D Image of Bioprinted Skin Tissue (28746740398).jpg, by National Center for Advancing Translational Sciences from Bethesda, MD (Public domain, via wikimedia-commons): source