LL-37: the body's own natural germ-killer
A germ-fighting peptide the human body makes naturally; it kills a broad range of bacteria and is used as a reference standard in lab research, not as an approved drug.
A researcher, an agent, or an algorithm wrote down the sequence and picked a target to hit.
An AI model like OpenFold3 or AlphaFold built a 3D structure and scored how well it fits the binding site.
A second contributor repeated the computation on their own hardware and the scores matched.
Reference scaffold compound — documented synthesis
Fork this card to add platform evidence →
Reference scaffold — used as positive control in bioassays; activity well established
Fork this card to add platform evidence →
What this is
LL-37 is a 37-amino-acid antimicrobial peptide and the only cathelicidin found in humans. It is cut from a larger precursor protein called hCAP-18 (human cathelicidin antimicrobial peptide-18), which sits stored in the granules of neutrophils and is also made by epithelial cells lining the skin, gut, lungs, and reproductive tract. When infection or injury occurs, proteases — primarily proteinase 3 in neutrophils, and additional proteases at epithelial surfaces (Murakami and colleagues 2004) — cleave hCAP-18 and release the active LL-37 fragment into local tissues and secretions. LL-37 is also known as Human Cathelicidin Antimicrobial Peptide and cathelicidin.
LL-37 biology has been one of the most intensively studied chapters of innate immunity for three decades. Despite this depth of characterization, LL-37 has not reached regulatory approval as a therapeutic.
History
LL-37 was identified in the mid-1990s as the C-terminal antimicrobial fragment of hCAP-18. The gene encoding hCAP-18 — called CAMP — was characterized in 1996 by Gudmundsson and colleagues (Eur J Biochem 1996), establishing the structural framework for the precursor-to-active-peptide processing story. The name LL-37 derives from its two N-terminal leucine residues and its length of 37 amino acids.
Over the following decades, research documented LL-37's roles beyond direct microbial killing: as an immunomodulator that shapes dendritic cell differentiation and TLR signaling, as a promoter of wound healing and angiogenesis, and — notably — as a peptide whose expression is directly controlled by vitamin D via a vitamin D response element (VDR element) in the CAMP gene. The vitamin D–cathelicidin axis became one of the canonical examples linking micronutrient status to innate immune defense.
A distinct thread in the literature documents contexts where LL-37 is harmful rather than protective: in rosacea, psoriasis subsets, and lupus, aberrant LL-37 processing contributes to inflammatory pathology rather than resolving it (Durr and colleagues 2006).
LL-37 entered clinical investigation primarily in wound care. Multiple randomized trials have tested topical LL-37 in chronic venous leg ulcers and diabetic foot ulcers, and a small trial tested an oral formulation for SARS-CoV-2 infection. No formulation has reached regulatory approval.
What it does
LL-37 acts as both a direct antimicrobial and a signaling molecule for the immune system. In its antimicrobial role, it disrupts the membranes of bacteria, viruses, and fungi — Turner and colleagues (Antimicrob Agents Chemother 1998) characterized its activity against gram-negative and gram-positive organisms including Pseudomonas aeruginosa and multiply antibiotic-resistant pathogens. In its immune-signaling role, it draws immune cells to sites of infection, modulates how innate immune receptors respond to bacterial toxins, and can neutralize circulating bacterial lipopolysaccharide (LPS), a property characterized by Ciornei and colleagues (Antimicrob Agents Chemother 2005).
Beyond fighting infection, LL-37 promotes tissue repair by driving keratinocyte migration and proliferation, which is the mechanistic basis for the wound-healing trial program.
A critical nuance is that these effects are concentration- and context-dependent. At higher concentrations, LL-37 is toxic to mammalian cells as well — selectivity for pathogen membranes over host cell membranes is not absolute (Ciornei and colleagues 2005). In rosacea, psoriasis subsets, and lupus, elevated or aberrantly processed LL-37 drives inflammation rather than resolving it.
Evidence
- Human: Completed randomized controlled trials for topical LL-37 in hard-to-heal venous leg ulcers (including a multicentric placebo-controlled trial, PMID 34687253) and in diabetic foot ulcers (randomized double-blind trial, PMID 37480520), with modest but positive effects on wound healing. A small randomized trial of oral LL-37 for SARS-CoV-2 omicron BA.5.1.3 infection reported efficacy and safety outcomes (PMID 37605995). Multiple vitamin D supplementation RCTs establish that correcting vitamin D deficiency reliably raises circulating LL-37 levels, consistent with the known VDR element in the CAMP gene. Systemic exogenous LL-37 administered by injection has not been tested in a controlled human trial.
- Animal: Moderate preclinical evidence in infection and wound models, including a murine sepsis model (PMID 32825174) and antimicrobial efficacy studies against ESKAPE pathogens and cystic fibrosis isolates (Saiman and colleagues 2001; Narayana and colleagues 2019).
- In vitro: Extensive and mechanistically well-characterized — broad-spectrum antimicrobial activity, LPS neutralization, keratinocyte migration, TLR modulation, and concentration-dependent host cell cytotoxicity are all documented.
Myths and misconceptions
- "LL-37 is a natural antibiotic so it's safe and broadly beneficial to inject" — LL-37 is endogenously regulated in tightly controlled local and systemic ways. Exogenous administration bypasses that regulation. The peptide can be cytotoxic to mammalian cells at higher concentrations, and in rosacea, psoriasis subsets, and lupus, LL-37 dysregulation drives pathology rather than preventing it. "Natural antimicrobial" is not the same as "safe to inject chronically."
- "Injecting LL-37 reproduces the immune benefits seen with vitamin D supplementation" — Vitamin D upregulates endogenous cathelicidin expression in tissues where it is biologically appropriate and corrects the deficit at the regulatory level. Injecting synthetic LL-37 systemically delivers peptide to compartments and concentrations that endogenous regulation does not produce, and this has not been shown in clinical studies to reproduce vitamin D's effects.
- "LL-37 is FDA-approved for wound healing" — Topical LL-37 has been studied in clinical trials for chronic wounds with positive but modest results. It is not FDA-approved for wound healing or any other indication.
- "Research-chemical LL-37 is the same as the LL-37 used in published trials" — Clinical trial preparations are produced under quality systems with identity, purity, sterility, and endotoxin controls. Research-chemical preparations offer none of those guarantees — particularly important for an immune-active peptide where endotoxin contamination could independently trigger inflammatory responses.
Known effects
- Antimicrobial activity (bacteria, viruses, fungi) — Extensively characterized in vitro; activity against gram-positive and gram-negative organisms including P. aeruginosa, multiply resistant cystic fibrosis isolates, and ESKAPE pathogens (Turner and colleagues 1998; Saiman and colleagues 2001; Narayana and colleagues 2019)
- Chronic venous leg ulcer healing (topical) — Phase III / Completed RCTs; modest effect sizes
- Diabetic foot ulcer healing (topical) — Completed RCT; limited replication to date
- Oral antiviral (SARS-CoV-2) — One small completed RCT; no replication
- Endogenous level modulation via vitamin D — Well-supported through multiple RCTs; CAMP VDR element mechanism well characterized
- Wound healing and angiogenesis (preclinical) — Preclinical; mechanistic data in tissue models
- LPS neutralization — In vitro and preclinical; anti-endotoxin activity characterized (Ciornei and colleagues 2005)
Safety signals
Concentration-dependent cytotoxicity: LL-37 can be toxic to mammalian cells at higher concentrations. Selectivity for pathogen membranes over host membranes is not absolute (Ciornei and colleagues 2005). The safety margin for systemic exogenous use in humans has not been characterized.
Context-dependent pro-inflammatory effects: In rosacea, psoriasis subsets, and lupus, dysregulated LL-37 processing contributes to inflammatory pathology. Exogenous LL-37 in individuals with these conditions may worsen rather than improve symptoms.
No pharmacokinetic data for systemic injection: Absorption, distribution, half-life, and tissue uptake of synthetic LL-37 injected systemically have not been characterized in humans.
Research-chemical quality: Synthetic LL-37 sold outside regulated supply chains has no guaranteed identity, purity, sterility, or endotoxin control. For an immune-active peptide, endotoxin contamination is a particularly relevant concern.
Cancer context: LL-37 has pro-angiogenic activity; complex context-dependent effects on tumor biology are described in available sources.
Regulatory status
- US (FDA): Not approved for any indication. Topical and oral investigational formulations have been tested in trials but have not reached approval. Synthetic LL-37 sold through research-chemical channels is not authorized for human therapeutic use; the FDA's 2023 review of compounded peptides narrowed the legitimate compounding pathway for peptides in this category.
- EU (EMA): No marketing authorization.
- UK (MHRA): No license.
- Canada: Investigational only.
- Australia (TGA): No approval; general enforcement action taken against unapproved peptide sales.
- WADA: LL-37 is not specifically named on the prohibited list but falls under the S0 catch-all for substances not approved by any governmental health authority for human therapeutic use. Current list status not independently refreshed in this card.
Mechanism
LL-37's antimicrobial activity operates primarily through direct physical disruption of pathogen membranes. The peptide adopts an amphipathic alpha-helical structure in lipid environments (Wang and colleagues 2008, J Biol Chem), with positively charged and hydrophobic faces that allow it to insert into and disrupt bacterial membranes via carpet-like or toroidal pore mechanisms. The N-terminal region does not form an amphipathic helix, while the core region (approximately residues 17–32) contains the primary antimicrobial pharmacophore (Wang and colleagues 2018). The same mechanism underlies both antimicrobial activity and host-cell cytotoxicity at higher concentrations.
Beyond membrane disruption, LL-37 acts as an immune modulator through the formyl peptide receptor 2 (FPR2/ALX), a G-protein-coupled receptor expressed on immune cells (Cattaneo and colleagues 2013). Activation of FPR2 by LL-37 elicits distinct signaling cascades depending on ligand context. LL-37 also binds and neutralizes LPS by competing with its interaction at TLR4. The CAMP gene carries a vitamin D response element (VDR element); in multiple tissue types, vitamin D directly upregulates cathelicidin transcription, making LL-37 expression a functional readout of vitamin D status (Gudmundsson and colleagues 1996).
Open questions
- Systemic injection efficacy — The most common community self-administration route has no controlled human evidence. Whether injectable LL-37 produces any of the wound-healing effects demonstrated with topical preparations is unknown.
- Safety margin for systemic use — Concentration-dependent cytotoxicity and the absence of PK/PD data in humans leave the therapeutic window for systemic exogenous LL-37 undefined.
- Long-term safety — Chronic immune modulation by exogenous LL-37 has not been studied at any dosing scale or duration.
- Disease-context selectivity — In rosacea, psoriasis subsets, and lupus, LL-37 can be pathogenic. Predictive markers for benefit vs. harm are lacking.
- Research-chemical equivalence — Whether research-chemical preparations produce equivalent peptide to clinical-trial preparations, and how endotoxin variation affects immunological outcomes, is not characterized.
Research directions for this peptide, selected from the current sources — hypotheses you can explore and model. None of it is proven yet; tap any one to see the full thinking.
Is there a precise concentration threshold below which LL-37 safely directs immune cells and above which it starts destroying both germs and your own cells indiscriminately?
If this dose-switch is real, drug designers could keep LL-37 below the danger level to get the immune benefits without the toxicity, which is the main reason this natural peptide has never become an approved medicine.
If LL-37 is chemically attached to a collagen-based wound dressing, will it slowly release at levels that kill bacteria but stay too low to hurt the patient's own cells?
For people with hard-to-heal wounds like diabetic foot ulcers, this could mean a single dressing that fights infection for two weeks without the tissue damage that has blocked LL-37 from clinical use, requiring no changes to the peptide itself.
At safe, natural concentrations, does LL-37 dampen a key inflammation trigger in immune cells, reducing the harmful flare-up signals behind gout attacks and related conditions?
If confirmed, LL-37 or a close relative could open a new treatment path for gout, rare inherited fever syndromes, and other diseases driven by this inflammation pathway, conditions where current options are limited or cause side effects.
Is the standard label for LL-37, "antimicrobial," hiding its more important role as a molecule that switches on specific immune receptors in the body?
Reclassifying LL-37 by its receptor targets could let researchers compare it to a whole family of related drugs, design better versions, and unlock immune-modulating uses that its current germ-killer label obscures.
Would a topical LL-37 treatment clear acne better than antibiotics alone by both killing Cutibacterium acnes and blocking the inflammation signal that turns an infection into an angry red pimple?
With antibiotic-resistant acne on the rise and roughly 650 million people affected worldwide, a single topical agent that addresses both the microbial cause and the inflammatory reaction could offer a genuinely new option, especially for people for whom current treatments have stopped working.
Is the loose, unstructured front end of LL-37 actually muffling its own activity, so that a shorter version without it would activate immune receptors more strongly and kill bacteria at lower doses?
If that first segment acts as a natural brake, trimmed versions of LL-37 could be more targeted, less toxic, and more useful as a starting point for wound-healing or anti-inflammatory drug candidates.
Instead of simply grabbing the bacterial toxin (LPS) in the blood, does LL-37 actually dock onto the immune receptor itself and lock it in a closed position so it cannot fire the dangerous sepsis cascade?
If LL-37 works at the receptor level rather than just soaking up toxin, it would point to a precise docking site that chemists could use to design new anti-sepsis drugs, a condition that kills millions of people each year and has very few targeted therapies.
▸full evidence table1 metrics
| metric | value | tool |
|---|---|---|
| ranking score | 0.8009225726127625 | boltz-2 |
▸3-letter notation
▸recipeboltz-2 2.2.1
| parameter | value |
|---|---|
| model | boltz-2 2.2.1 |
| weights | — |
| hardware | vast_v100_32gb |
| mlx version | — |
| python | — |
| random seed | 1 |
| msa strategy | none_monomer |
| runtime | — |
| predicted by | — |
| predicted at | 2026-05-23 |
▸citationbibtex
@peptide{pep00002,
sequence = {LLGDFFRKSKEKIGKEFKRIVQRIKDFLRNLVPRTES},
target = {antimicrobial},
author = {peptidemodel},
year = {2026},
status = {bioassayed}
}