Creative Touch Permanent Makeup and Aesthetics

LL-37 is the only cathelicidin expressed in humans – a distinction that makes its biology simultaneously elegant and precarious. Every function the cathelicidin family provides in human skin rests on a single molecule, and that molecule’s behaviour depends not on whether it is present, but on how much of it reaches the skin surface and in what form. At physiological concentrations, LL-37 kills bacteria through membrane disruption, neutralises bacterial endotoxins before they can trigger TLR-mediated inflammation, promotes wound healing through EGFR transactivation and angiogenesis, disrupts biofilm at concentrations 100-fold below those required for bactericidal activity, and recruits neutrophils and monocytes to sites of challenge. At excess concentrations – generated when KLK5 serine protease activity escapes its normal regulatory controls – LL-37 degranulates mast cells, drives vascular permeability and angiogenesis, recruits Th17 cells, and activates NLRP3 inflammasome signalling. That is rosacea.

In atopic dermatitis, the opposite problem obtains: Th2 cytokines suppress CAMP gene expression, alkaline skin pH reduces KLK5 processing efficiency, and reduced vitamin D pathway activity removes the upstream induction signal, leaving the skin surface with substantially insufficient LL-37 – directly enabling Staphylococcus aureus colonisation through a mechanism that no topical product addresses from the outside. Understanding LL-37 is understanding how the same molecule can be both the solution and the problem, separated only by the regulatory environment controlling its production. ^[7]

Of the dozens of antimicrobial peptides expressed in human skin, LL-37 is in a category of its own. Every other mammalian species expresses multiple cathelicidins – mice have CRAMP, pigs have protegrins and PR-39, cows express multiple bovine cathelicidins. Humans have one. Whatever the cathelicidin family contributes to skin immunity in every other mammal is, in humans, carried by a single 37 amino acid peptide named for the two leucines at its N-terminus. That evolutionary narrowing is one reason LL-37’s dysregulation has such outsized clinical consequences: there is no backup.

Structure, synthesis, and activation

LL-37 is not synthesised directly. Keratinocytes, neutrophils, mast cells, sweat gland epithelial cells, and sebocytes produce an 18 kDa inactive precursor protein – hCAP18 – encoded by the CAMP gene on chromosome 3. hCAP18 consists of a conserved N-terminal cathelin domain (which inhibits the active peptide’s function while providing protease resistance to the precursor) and a C-terminal cationic domain that, once cleaved, becomes LL-37. ^[5]

Activation requires extracellular protease cleavage. At the skin surface, the primary enzyme is KLK5 – a serine protease secreted by keratinocytes into the stratum corneum. KLK5 cleaves hCAP18 at a specific arginine- glycine bond, releasing the 37 amino acid C-terminal fragment as active LL-37. A second cleavage pathway involves neutrophil-derived proteinase 3 in wound fluid and inflammatory exudate – generating the same LL-37 peptide through an independent route that ensures AMP availability even when surface KLK5 activity is compromised. ^[8]

The two-step synthesis and activation process is not architectural redundancy, but a regulatory mechanism. By storing the cathelicidin as an inactive precursor and requiring extracellular protease activation, the skin maintains a reservoir of potential LL-37 activity that can be rapidly deployed at sites of injury or infection – where local protease activity rises sharply – without constitutively producing the active, membrane-disrupting peptide in every keratinocyte at all times. The regulation is built into the processing requirement.

KLK5 activity itself is regulated by two overlapping controls: LEKTI (lympho-epithelial Kazal-type inhibitor, encoded by SPINK5), which directly inhibits KLK5 in the stratum corneum; and pH, with KLK5 activity optimised at approximately pH 5.5 and measurably reduced below pH 5. Both controls fail in specific disease states – and when they do, the consequences follow directly from what KLK5 does to hCAP18.

What LL-37 does at physiological concentrations

Direct antimicrobial activity. LL-37 disrupts bacterial cell membranes through the cationic mechanism described in the parent Anitmicrobial Peptides page – electrostatic attraction to negatively charged bacterial membranes, insertion through the amphipathic helix, and membrane destabilisation through pore formation or carpet-model solubilisation. It is active against gram-positive bacteria (including S. aureus and MRSA), gram-negative bacteria, fungi, and enveloped viruses. It kills S. aureus at concentrations achievable in healthy skin, directly contributing to the competitive advantage commensals hold over the pathogen. ^[7]

Biofilm disruption. This is one of the least-discussed and most clinically significant properties of LL-37. It disrupts bacterial biofilm – the structured, matrix-protected community that makes S. aureus and C. acnes IA1 resistant to both antibiotic treatment and host immune clearance – at concentrations approximately 100-fold lower than those required for direct bactericidal activity. That concentration threshold matters enormously: it means that even in skin where LL-37 levels are insufficient for direct killing, biofilm disruption may still be occurring, and that relatively modest increases in LL-37 expression can meaningfully affect biofilm-resident pathogen communities. ^[7]

LPS and LTA neutralisation. LL-37 binds lipopolysaccharide from gram-negative bacterial cell walls and lipoteichoic acid from gram-positive cell walls with high affinity, neutralising these TLR4 and TLR2 agonists before they can trigger the full inflammatory cascade. This function is particularly important at the impaired barrier, where microbial products penetrate into the epidermis: LL-37 acts as a molecular buffer between the bacterial products entering through a compromised barrier and the TLR-expressing keratinocytes and immune cells that would otherwise generate sustained NF-κB activation in response. ^[7]

Wound healing. LL-37 promotes keratinocyte migration and proliferation through EGFR transactivation – it binds the EGFR co-receptor P2X7 and triggers downstream ERK1/2 and PI3K/Akt signalling that drives re-epithelialisation. It promotes angiogenesis through VEGF upregulation, supporting the vascular ingrowth that supplies the rebuilding epidermis with oxygen and nutrients. These are not incidental properties. They are active contributions to the wound repair cascade that place LL-37 at the interface of infection defence and tissue regeneration – operating simultaneously as pathogen killer and repair signal at every wound site. ^[1]

Immune cell chemotaxis. LL-37 acts as a chemoattractant for neutrophils, monocytes, and mast cells through formyl peptide receptor 2 (FPR2). At wound sites and sites of infection, this recruitment function bridges the initial AMP-mediated killing with the cellular immune response – LL-37 is not waiting for neutrophils to arrive and then acting alongside them; it is actively summoning them while simultaneously managing the pathogen load. ^[1]

The concentration switch: when LL-37 becomes inflammatory

Everything described above occurs within the concentration range that healthy, regulated skin maintains. When that range is exceeded – when hCAP18 production is upregulated or KLK5 activity is disinhibited beyond normal parameters – LL-37 transitions from protective peptide to inflammatory driver. The transition is not abrupt; it is a gradient. But the clinical consequences of sustained excess are specific and severe.

Mast cell degranulation. Excess LL-37 activates mast cells through a TLR2/JAK2/STAT3 pathway, as characterised by Fan et al. (2025). Mast cell degranulation releases histamine (producing immediate vasodilation and flushing), prostaglandins and leukotrienes (sustaining the vascular response and recruiting further inflammatory cells), TSLP (driving Th2 skewing in adjacent tissue), and tryptase (a serine protease that activates PAR2 on sensory neurons, generating neurogenic inflammation). In skin with excess LL-37, every mast cell encounter produces a disproportionate inflammatory output – and rosacea skin is densely populated with mast cells relative to healthy skin, so the amplification is structural as well as biochemical. ^[4]

Vascular remodelling. LL-37’s VEGF-upregulating activity – protective in wound healing – becomes pathological when sustained by chronic excess. Persistent VEGF elevation drives angiogenesis and vascular permeability that cannot be turned off between triggering events. The persistent erythema and developing telangiectasia of rosacea are direct structural consequences of chronic LL-37-driven vascular remodelling. The blood vessels are not simply dilated; they are anatomically changed. ^[8]

Th1/Th17 recruitment and adaptive immune education. Excess LL-37 recruits Th17 cells and activated dendritic cells, establishing an adaptive immune response primed to the triggers that generated the original LL-37 excess. Each acute episode of excess LL-37 production deepens the adaptive immune sensitisation – which is why rosacea triggers, over time, tend to produce increasingly severe responses to smaller stimuli. The immune system is being educated by every flare. ^[2]

NLRP3 inflammasome activation. At elevated concentrations, LL-37 drives NLRP3 inflammasome activation through mitochondrial reactive oxygen species generation, producing IL-1β release that further amplifies the inflammatory environment. This is the same inflammasome that pathogenic C. acnes CAMP factors activate in acne – a convergence point that is relevant when atopic, acne, and rosacea presentations overlap. ^[1]

The rosacea cascade – named in full

Rosacea is, at its mechanistic core, a disease of LL-37 dysregulation. Understanding the cascade precisely is the difference between treating symptoms and understanding what you are managing:

1. Trigger exposure (UV radiation, heat, certain foods, alcohol, emotional stress, Demodex mite activity) activates KLK5 expression and activity in the epidermis – either directly or through the protease activation cascade in which KLK14 and KLK8 upstream activate KLK5

2. Elevated KLK5 cleaves hCAP18 to generate LL-37 at concentrations substantially above the physiological range – and simultaneously generates aberrant processing fragments (including a specific N-terminal fragment) that are not produced in healthy skin and are independently pro-inflammatory, activating TLR2 and driving cytokine production through pathways distinct from intact LL-37 ^[8]

3. Excess LL-37 degranulates mast cells via TLR2/JAK2/STAT3 → histamine, prostaglandins, TSLP; drives vascular permeability and VEGF-driven angiogenesis; recruits neutrophils and Th17 cells via FPR2; activates NLRP3 in macrophages ^[4]

4. Mast cell and Th17 activation further stimulates KLK5 expression – creating the self-sustaining loop where KLK5-driven LL-37 excess produces an inflammatory environment that re-stimulates KLK5, maintaining the cycle independently of the original trigger ^[2]

5. Repeated cycles produce anatomical changes: permanent vascular remodelling (telangiectasia), sebaceous gland hyperplasia (rhinophyma), and progressive sensitisation of both the innate and adaptive immune response to triggering stimuli

The acid mantle connection here is direct and often overlooked. KLK5 activity is pH-sensitive, with its optimum around pH 5.5 and measurable inhibition below pH 5. An alkaline skin surface – from alkaline cleansers, hard water mineral deposition, or the progressive skin pH elevation that comes with barrier dysfunction – removes a primary regulatory brake on KLK5 activity. Maintaining the acid mantle in rosacea-prone skin is not just microbiome management. It is active KLK5 suppression.

LL-37 in atopic dermatitis – the opposite problem

If rosacea is the disease of LL-37 excess, atopic dermatitis is the disease of LL-37 insufficiency. And it is produced by a convergence of three independent suppression mechanisms operating simultaneously:

Th2 cytokine suppression. IL-4 and IL-13 – the dominant cytokines of the atopic inflammatory environment – suppress CAMP gene transcription in keratinocytes through STAT6-mediated inhibition. The approximately 50% reduction in LL-37 expression in lesional AD skin versus healthy skin reflects this direct cytokine-driven transcriptional suppression. Critically, this is the same Th2 environment that suppresses filaggrin and PPARα – all three deficits share a common upstream cause, which is why restoring the inflammatory environment (rather than individually supplementing each downstream deficit) is such a potent therapeutic strategy. ^[3]

Alkaline pH reducing KLK5 processing. Atopic skin has a consistently elevated surface pH compared to healthy skin – driven by barrier dysfunction, reduced lactic acid secretion, and impaired proton pump activity. Elevated pH reduces KLK5 efficiency in cleaving hCAP18 to active LL-37, meaning even the hCAP18 that is produced is less efficiently converted. This is a second, independent suppression route operating at the activation step rather than the synthesis step. ^[8]

Vitamin D pathway reduction. Vitamin D deficiency – prevalent in atopic populations and further compounded by sun avoidance behaviour in photosensitive or photosensitised AD skin – reduces VDR-driven CAMP gene transcription. Where the Th2 environment is suppressing LL-37 production at one regulatory level, vitamin D insufficiency suppresses it at another. ^[3]

The consequence of this triple suppression is a skin surface with substantially reduced cathelicidin activity – insufficient to suppress S. aureus Agr quorum sensing (which LL-37 achieves at concentrations present in healthy skin), insufficient to kill established S. aureus at lesional sites, and insufficient to neutralise the LPS and LTA that S. aureus generates in the now-colonised epidermis. The microbiological and immunological vulnerabilities of AD skin are not separable problems. They are the same problem – LL-37 deficit – producing consequences in two domains simultaneously. ^[9]

Vitamin D → CAMP → LL-37: the upstream regulatory route

The CAMP gene promoter contains a directly confirmed vitamin D response element (VDRE) to which the vitamin D receptor (VDR) binds in response to 1,25-dihydroxyvitamin D3 (the active metabolite). At concentrations achievable through vitamin D3 supplementation to sufficiency – serum 25(OH)D in the range of 75–100 nmol/L – hCAP18 protein expression in keratinocyte models increases approximately 13-fold compared to VDR-unstimulated controls. ^[6]

This is not a pharmacological dose effect. It is a physiological restoration – the VDR system driving CAMP expression as it is designed to do when vitamin D status is adequate, and failing to do so when vitamin D status is low. The implication is direct: for clients in whom LL-37 insufficiency is a factor (AD presentations, recurrent skin infection, aged skin with impaired AMP defence), correcting vitamin D deficiency addresses the deficit from the transcriptional level upward – not by adding more of the peptide topically, which is not a viable route, but by restoring the endogenous production signal that should have been generating it.