LL-37 is the only cathelicidin made in humans, and the published record describes it less as a wound antiseptic than as a repair signal the body switches on within hours of injury. The mechanistic and animal evidence for that role is strong and reads as causal; the human clinical evidence is one small positive trial that a larger controlled trial did not reproduce. That gap between well-mapped biology and an unconfirmed treatment is the whole story for anyone weighing what LL-37 can currently do in a wound.
LL-37 is a well-characterized endogenous repair signal whose early therapeutic signal in a 34-patient trial was not confirmed by a 148-patient Phase IIb trial, and no Phase III trial has been reported.
The peptide never touches the receptor that does the work. LL-37 drives keratinocyte migration through the epidermal growth factor receptor indirectly, by way of a shedding step at the cell surface, and the blocking experiments are what established that indirect route. This is the arm with the most direct causal evidence behind it, and it is also the arm with a hard ceiling, since concentrations that move keratinocytes in the low micromolar range become membrane-disruptive to those same cells at higher exposure.
Blocking EGFR, inhibiting the metalloproteinase step, or neutralizing heparin-binding EGF each abolishes the migratory response, and that pattern is what established the transactivation model rather than a direct receptor interaction.
Angiogenesis is the arm where LL-37 behaves most cleanly like a classical signaling molecule: one receptor, one G protein-coupled pathway, and effects that vanish when the receptor is blocked. For a stalled ulcer that arm may carry as much weight as the epidermal one, since a wound bed without adequate perfusion cannot supply the oxygen and nutrient load that granulation tissue and a migrating epidermis demand.
Mice lacking the cathelicidin ortholog CRAMP form wounds with reduced vascularization, which ties the peptide to vessel growth in repairing tissue specifically rather than in a generic assay.
Two seemingly opposite jobs sit in the same molecule, and context decides which one dominates. The peptide amplifies the early inflammatory response and then helps shut it down, and the record treats that duality as its defining feature in this phase rather than as a contradiction in the data. Dose and timing are what separate orderly repair from the sustained recruitment seen in inflammatory skin disease.
| Function | Amplifying arm | Restraining arm |
|---|---|---|
| Primary action | FPR2-mediated chemotaxis of neutrophils, monocytes, mast cells, T cells | Direct binding of lipopolysaccharide and lipoteichoic acid |
| Cytokine effect | Raises IL-8, IL-6, MCP-1 output | Blunts TLR4 and TLR2 engagement |
| Cell fate | Delays neutrophil apoptosis | Promotes macrophage uptake of dying neutrophils |
| Effect on the phase | Sustains inflammation when levels stay high | Allows the pivot into the proliferative phase |
Early, moderate LL-37 clears the field and then helps close the response down, while sustained high levels keep recruiting cells and feeding cytokine output, a picture closer to inflammatory skin disease than to orderly repair.
This is the thinnest of the repair arms, and it deserves to be described that way. Dermal fibroblasts clearly respond to LL-37, but the direction of the collagen response flips depending on the experiment, and no single answer survives across them.
Animal wound studies generally show faster closure when cathelicidin is present without a clear signal that the resulting scars are better or worse, so any claim that LL-37 improves cosmetic outcome runs ahead of the data.
The obvious reading, that chronic ulcers are simply short of cathelicidin, does not survive contact with the measurements. Immunostaining at chronic venous ulcer edges finds hCAP18 and LL-37 reduced or absent, while chronic wound fluid often carries as much peptide as an acute wound or more. Both readings can hold at once, because total peptide present is not the same as functional peptide able to reach a receptor.
Gene transfer of LL-37 into wounds in diabetic mice improved closure and vascularization, which supports reading the peptide as a plausible target in stalled wounds rather than only a marker of them.
Both arms are real, but they operate at different concentrations and under different conditions, and the host-directed arm is the stronger candidate for the healing effect. Direct killing needs concentrations that are hard to sustain in tissue, and physiological salt, serum, and the anionic material filling a wound bed all weaken it. Receptor signaling runs an order of magnitude lower and survives the conditions that abolish bactericidal activity.
| Criteria | Antimicrobial arm | Host-directed arm |
|---|---|---|
| Working concentration | High, hard to sustain in tissue | Low micromolar |
| Mechanism | Electrostatic membrane disruption | FPR2 signaling and EGFR transactivation |
| Effect of salt and serum | Potency drops sharply | Largely unaffected |
| Evidence in sterile wounds | Not applicable | Closure still accelerates |
LL-37 still accelerates closure in wound models with no bacterial burden, and cathelicidin-deficient mice show impaired vascularization and slower closure after sterile injury, which cannot be attributed to lost microbial control.
No single number marks the line, but the shape of the curve is consistent: benefit at low concentrations, harm at high ones, with the crossover moving by cell type and assay conditions. Rosacea and psoriasis are what excess cathelicidin activity looks like in people, through aberrant kallikrein-5 processing in the first, and through LL-37 complexing with self-DNA and self-RNA at TLR9 and TLR7 in the second. Reported thresholds vary considerably, since serum, salt, and anionic matrix all bind free peptide and effectively lower the delivered dose.
Receptor-mediated benefit appears in the low micromolar range while cytotoxicity toward keratinocytes and fibroblasts generally emerges around an order of magnitude higher, and the width of that window in human wound tissue has not been mapped precisely.
Human evidence is genuinely limited and should not be described as more than it is. Two controlled trials exist, they point in different directions, and the larger of the two is the one that failed.
Topical LL-37 cleared an early human safety hurdle and showed a signal of efficacy in one 34-patient trial that a larger 148-patient controlled trial did not confirm, and it has not been established as an effective wound treatment.
The biology is not what has stalled this molecule. Four practical walls sit between a well-mapped repair signal and a usable product, and they compound: the first two destroy or immobilize an applied dose, which pushes toward higher dosing, which collides with the third. Formulation and sequence engineering are the responses under investigation, and none has produced a marketed product.
Each truncated or D-amino-acid analog designed to resist cleavage restarts the regulatory clock as a new molecular entity rather than inheriting the parent peptide's safety record.
Human cathelicidin regulation carries an oddity with real consequences: the CAMP gene holds a functional vitamin D response element in its promoter, inserted by an Alu retrotransposon in the primate lineage. Cathelicidin output in human skin is tied to vitamin D signaling in a way that has no equivalent in mice or rats. That single difference means the large preclinical literature built on rodent wound models cannot test this pathway at all.
Cell and tissue studies show reduced cathelicidin induction when 25-hydroxyvitamin D substrate is low, but whether correcting deficiency measurably improves wound healing in patients is unsettled, since vitamin D acts through many genes beyond CAMP.
Educational use only. This article describes what the published scientific and clinical literature reports about LL-37. It is not medical advice, and it does not recommend, prescribe, or tell anyone to use anything described here. The regulatory status shown at the top of this page reflects what the record showed on the date given there and can change. mdpep.com does not sell any substance described here, does not endorse human use of it, and does not direct anyone to obtain it.
This is not guidance for your situation. Nothing here accounts for your medical history, your current medications, or anything else specific to you, and none of it should be used to make a decision about your own health.
Affiliate disclosure. Some links on this site are affiliate links, and mdpep.com may earn a commission when they are used. That never affects what is written here, it is not an endorsement of any vendor, and it is not a statement that anything described on this page is available to buy.
Every claim here ties to a named primary source and a date, written as plain documentation with nothing for sale. That is how MD PEP covers the whole peptide market.
