LL-37
Cathelicidin LL-37, hCAP18(134-170), human cathelicidin antimicrobial peptide, CAMP gene product, LL37
LL-37 is the only human cathelicidin: a 37-residue cationic peptide released from the hCAP18 precursor by neutrophils and epithelial cells. It kills bacteria by disrupting membranes and separately drives host-defence signalling. It also has a well-documented role in driving autoimmune inflammation in psoriasis, which makes framing it as a benign immune booster inaccurate. Human efficacy evidence amounts to two topical wound trials, and the larger one failed.
Mechanism
LL-37 is the mature, secreted form of the only cathelicidin encoded in the human genome. The CAMP gene produces an 18 kDa precursor, hCAP18, stored in neutrophil secondary granules and produced by keratinocytes and other epithelia. Extracellular proteolysis (proteinase 3 in neutrophils, kallikreins 5 and 7 in skin) releases the C-terminal 37 residues, which begin with two leucines, hence the name. The peptide carries a net charge of about +6 and folds into an amphipathic alpha-helix on contact with anionic membranes.
It works through two largely separable mechanisms. The first is direct antimicrobial action: the cationic face binds negatively charged bacterial lipopolysaccharide, lipoteichoic acid and membrane phospholipid, and the peptide inserts and permeabilises the membrane. Being physical rather than target-specific, this is harder for bacteria to evolve around than conventional antibiotic mechanisms, but it is also strongly inhibited by physiological salt concentrations, divalent cations and serum proteins, which is a large part of why LL-37 has never become a systemic antibiotic despite thirty years of interest. The second mechanism is host-defence signalling largely independent of killing: LL-37 is an agonist at formyl peptide receptor 2 (FPR2, formerly FPRL1), transactivates the epidermal growth factor receptor, engages P2X7, and neutralises LPS to blunt TLR4-driven inflammation. Through these it recruits neutrophils, monocytes and T cells, promotes keratinocyte migration and angiogenesis, and reshapes cytokine output, the basis for the wound-healing hypothesis that was actually tested in humans.
The same chemistry that makes LL-37 useful also makes it dangerous, and this is not a hypothetical concern. Lande and colleagues showed in Nature in 2007 that LL-37 binds extracellular self-DNA released from dying cells, forming aggregated and condensed structures retained in early endocytic compartments of plasmacytoid dendritic cells, thereby converting inert self-DNA into a potent TLR9 trigger and driving interferon-alpha production, now understood as an initiating step in psoriasis. In 2014 the same group showed that LL-37 is itself a T-cell autoantigen in psoriasis: roughly two-thirds of patients with moderate-to-severe plaque disease harbour LL-37-specific CD4+ and/or CD8+ T cells producing IFN-γ and Th17 cytokines, and their frequency correlates with disease activity. Related work implicates LL-37 in complex with neutrophil extracellular trap material in lupus and in rosacea. LL-37 is therefore best understood as a double-edged effector of innate immunity with a documented causal role in autoimmune inflammation, not as an immune tonic.
What the research shows
Human efficacy evidence for exogenous LL-37 amounts to two topical wound-healing trials in venous leg ulcers, and the honest reading is that the second did not confirm the first.
Grönberg and colleagues (2014) ran the first-in-human phase I/IIa study in 34 patients with hard-to-heal venous leg ulcers: a three-week placebo run-in, then four weeks of twice-weekly topical LL-37 at 0.5, 1.6 or 3.2 mg/mL versus placebo, then four weeks of follow-up. The two lower concentrations produced substantially higher healing rate constants than placebo, with mean ulcer area reductions of 68% and 50% respectively. The highest concentration, 3.2 mg/mL, showed no benefit at all. An inverted dose-response in a 34-patient study spread across four arms is a result that should prompt caution rather than confidence, though it is biologically arguable given that cationic peptides become cytotoxic to host cells above a concentration threshold. The specific p-values and fold-changes quoted in some secondary sources could not be confirmed from the published abstract and are not reproduced here.
The follow-up was a proper confirmatory test: a phase IIb multicentre, double-blind, randomised, placebo-controlled, parallel-group trial across sites in Poland and Sweden, 148 patients treated (mean age 67.6, median ulcer duration 20.3 months, mean wound area 11.6 cm²), receiving 0.5 or 1.6 mg/mL twice weekly alongside compression therapy. It missed its primary endpoint: the analysis did not identify any significant improvement in healing with LL-37 compared with placebo. A post-hoc analysis of patients with large target wounds of 10 cm² or larger showed statistically significant improvement in healing parameters. Post-hoc subgroups within a failed trial generate hypotheses; they do not establish efficacy, and reporting them as if they did is one of the more common ways failed trials are laundered into marketing claims.
Everything else attributed to LL-37 (broad antibacterial and antibiofilm activity, antiviral effects, anti-cancer activity, systemic immune modulation) rests on cell culture and animal models. There are no human trials of injected or systemic LL-37 for any indication anywhere in the registry record. Its salt sensitivity, protease susceptibility and cytotoxicity at higher concentrations are precisely why that development path was never taken. Anyone encountering LL-37 sold for injection should understand they are attempting a route of administration that has never been tested in a human trial.
Evidence assessment
Mixed evidence
Two randomised placebo-controlled human trials exist for a single topical indication, but they conflict: the 34-patient phase I/IIa was positive on healing rate, and the larger 148-patient phase IIb missed its primary endpoint, with only a post-hoc subgroup favouring treatment. Everything else claimed for LL-37 is cell and animal work.
Tiers are applied consistently across the library and re-checked when new trials read out. Read the grading method.
Key studies
Treatment with LL-37 is safe and effective in enhancing healing of hard-to-heal venous leg ulcers: a randomized, placebo-controlled clinical trial Mixed evidence
Healing rate constants at 0.5 and 1.6 mg/mL were substantially higher than placebo, with mean ulcer area reductions of 68% and 50% respectively; no benefit at 3.2 mg/mL; no adverse events documented.
Evaluation of LL-37 in healing of hard-to-heal venous leg ulcers: A multicentric prospective randomized placebo-controlled clinical trial High-quality evidence
Primary endpoint not met. The analysis did not identify any significant improvement in healing with LL-37 versus placebo. A post-hoc analysis of patients with large target wounds (≥10 cm²) showed statistically significant improvement in healing parameters. Well tolerated and safe at both dose strengths.
Plasmacytoid dendritic cells sense self-DNA coupled with antimicrobial peptide Preclinical only
LL-37 binds extracellular self-DNA released from dying cells, forming aggregated and condensed structures retained within early endocytic compartments of plasmacytoid dendritic cells, converting inert self-DNA into a potent TLR9 trigger and driving interferon-alpha production, identifying an initiating mechanism in psoriasis.
The antimicrobial peptide LL37 is a T-cell autoantigen in psoriasis Preclinical only
Approximately two-thirds of patients harboured LL-37-specific CD4+ and/or CD8+ T cells producing IFN-γ and Th17 cytokines; these cells infiltrated lesional skin and their frequency correlated significantly with disease activity, establishing LL-37 as a driver of adaptive as well as innate autoimmune inflammation.
Safety
In the two topical wound-healing trials LL-37 was well tolerated: the phase I/IIa documented no adverse events, and the phase IIb reported the peptide as safe and well tolerated at both dose strengths. That record covers topical application to open ulcers at concentrations up to 3.2 mg/mL, twice weekly, for up to 13 weeks. It says nothing about any other route.
Several risks follow directly from what LL-37 is. Cationic amphipathic peptides are cytotoxic to mammalian cells above a threshold concentration, and LL-37 is haemolytic in vitro in the tens-of-micromolar range, which is a plausible explanation for the loss of benefit at the highest concentration in the phase I/IIa trial. Systemic administration has never been tested in humans, so there is no dose-toxicity relationship, no data on cytokine release, and no data on effects on circulating cells. Most importantly, elevated LL-37 is causally implicated in psoriasis through the self-DNA/TLR9 pathway and functions as a T-cell autoantigen in that disease; it is also associated with rosacea and with autoantibody production in systemic lupus erythematosus. Introducing exogenous LL-37 in a person with, or predisposed to, any of these conditions is a plausible route to worsening them, and no study has examined this. Material sold online additionally carries the usual unverified identity, purity, sterility and endotoxin problems.
Regulatory status
| Jurisdiction | Status |
|---|---|
| United Kingdom | No MHRA marketing authorisation. Not a licensed medicine. The phase IIb wound-healing trial was conducted at sites in Poland and Sweden under EU clinical trial authorisation and did not lead to a licence application. (The specific EudraCT numbers circulated for these trials could not be independently confirmed in this audit and are therefore not stated.) |
| United States | Not approved by the FDA for any indication. No LL-37 product has completed phase 3. It is sold online under 'research use only' labelling, a legal formula that permits sale of an unapproved compound rather than any statement about safety or evidence. |
| WADA (sport) | Not named on the WADA Prohibited List. Athletes should nonetheless consider section S0 (unapproved substances), which prohibits pharmacological substances not currently approved by any governmental regulatory health authority for human therapeutic use, a description that fits LL-37, which is approved nowhere. A formal ruling should be sought rather than permitted status assumed. |
Questions
Not in the body, on current evidence. LL-37 kills bacteria efficiently in low-salt laboratory conditions by disrupting their membranes, but its activity is strongly inhibited by physiological salt concentrations, divalent cations and serum proteins, and it is rapidly degraded by proteases. This is the central reason that thirty years of interest has not produced a systemic antimicrobial drug. No human trial has ever tested injected or systemic LL-37 for infection.
The first did and the second did not. A 34-patient phase I/IIa in venous leg ulcers found substantially higher healing rates than placebo at the two lower concentrations, and none at the highest. The confirmatory phase IIb (148 patients across sites in Poland and Sweden) missed its primary endpoint, with no significant improvement over placebo. A post-hoc subgroup with wounds of 10 cm² or more favoured treatment, but a post-hoc subgroup in a failed trial is a hypothesis, not a result.
Yes, causally, and this is well established rather than speculative. LL-37 binds self-DNA released from dying cells and converts it into a potent TLR9 trigger for plasmacytoid dendritic cells, driving interferon-alpha production, an initiating step in psoriasis, shown in Nature in 2007. In 2014 the same group demonstrated that LL-37 is itself a T-cell autoantigen, found in roughly two-thirds of patients with moderate-to-severe plaque psoriasis. It is also implicated in rosacea and in lupus autoantibody production. That makes the framing of LL-37 as a general immune booster misleading.
There is no evidence either way, because it has never been done in a human trial. Both human studies used a topical gel on open ulcers. LL-37 is haemolytic to red cells in vitro at concentrations in the tens of micromolar, has no established systemic pharmacokinetics, and has a documented role in driving autoimmune skin and systemic inflammation. Anyone injecting it is operating entirely outside the tested evidence base, with material of unverified identity, sterility and endotoxin content.