Research-Grade Peptides | 99%+ Purity Verified | US Delivery 2–5 Days
Save 5% with Crypto
SaiyanMed

Your Cart

No products in the cart.

LL-37 Sequence and Structure: Why Its Alpha Helix Matters

Illustration of a flexible peptide chain in solution on the left and the same chain folded into two bent helical segments lying along the surface of a curved lipid bilayer on the right.

LL-37 is the 37-residue peptide cleaved from the C-terminal end of the human cathelicidin precursor hCAP-18, and its sequence — LLGDFFRKSKEKIGKEFKRIVQRIKDFLRNLVPRTES — combines a net positive charge that pulls it toward bacterial surfaces with a pattern of hydrophobic and cationic side chains that only fits an amphipathic helix. In water the peptide is largely disordered, and added anions such as 15 mM bicarbonate, sulfate or trifluoroacetate also induce helix formation, with the transition cooperative in peptide concentration[1]. When an anionic lipid interface is present, residues 2–31 fold into a curved helix-bend-helix and the last six residues stay flexible[2]. That conditional folding is why the α-helix matters here: the helix is the membrane-facing unit, and in the in vitro systems measured so far the amount of helix tracks the strength of antibacterial activity[1].

The 37-Residue Sequence: Charge, Amphipathicity and No Cysteine

hCAP-18 is a 170-residue precursor with a 30-residue signal peptide, a cathelin-like prodomain at residues 31–131 and the mature peptide at 134–170[3]. Proteinase 3 releases LL-37 by cleaving the precursor; cathepsin G and neutrophil elastase also cut it[3]. Processing continues after release: on the skin surface, serine proteases produce shorter peptides from the same region, including KR-20 (residues 151–170), LL-23 (134–156), LL-29 (134–162), KS-30 (141–170), RK-31 (140–170) and FF-33 (138–170)[3]. LL-37 is one product of a processing ladder, and the shorter forms are trimmed versions of the same sequence rather than unrelated peptides.

Counting ionisable side chains in the 37-mer gives a side-chain charge total of +6 at neutral pH: six lysines (positions 8, 10, 12, 15, 18, 25) and five arginines (7, 19, 23, 29, 34) against two aspartates (4, 26) and three glutamates (11, 16, 36). This total excludes the terminal groups; UniProt records that the N-terminus is blocked[3] but does not give the C-terminal chemistry, so the peptide’s overall net charge depends on the terminal groups: the ionisable side chains total +6; if both the alpha-amino and alpha-carboxyl groups were free, their +1 and −1 charges would cancel and the whole peptide would still be +6; with the N-terminus blocked as recorded and a free C-terminal carboxylic acid the value is +5; with an amidated C-terminus it returns to +6. UniProt does not specify the C-terminal form, so at approximately pH 7 the whole-peptide charge is +5 or +6 and cannot be reduced to one value without that missing terminal chemistry. No histidine is present, so no side chain titrates near pH 7 to soften that charge. The apolar residues are patterned as well: Phe-5, Phe-6, Phe-17 and Phe-27, together with isoleucines and leucines at 13, 20, 24, 28 and 31, form the face that turns away from water once the helix closes. NMR nuclear Overhauser effects place all four phenylalanine rings in contact with anionic micelles[2], which is what an amphipathic arrangement predicts.

There is no cysteine in LL-37. The consequence is structural: the peptide cannot pin itself into a fixed conformation with an intramolecular disulfide, so helix content, oligomer state and orientation stay reversible and depend on the surrounding solution. The precursor is different — the cathelin domain is disulfide-bonded at Cys-86–Cys-97 and Cys-108–Cys-125[3], so hCAP-18 is covalently constrained while the released 37-mer is not. The mature peptide’s mass is 4492.9 Da by electrospray[3], a value that pairs with the sequence when a sample’s identity needs checking.

The acidic residues are spread rather than clustered: Glu-11 and Glu-16 in the N-terminal half and Glu-36 two residues from the end. The hydroxyl-bearing side chains are Ser-9, Ser-37 and Thr-35. Each glutamate breaks a run of positive charge, and Glu-16 sits inside the bend that NMR assigns between Gly-14 and Glu-16[2] — the hinge between the two helical segments.

How LL-37 Folds: Helix-Bend-Helix and a Disordered Tail

Diagram of a peptide folded into two helical segments joined by a kink, with one face shaded dark blue for polar and charged side chains, the opposite face shaded amber for hydrophobic side chains, and an unstructured strand trailing from one end.

In SDS micelles, 13C/15N-labelled LL-37 adopts a curved amphipathic helix-bend-helix spanning residues 2–31 followed by a disordered C-terminal tail, with the bend between Gly-14 and Glu-16. Chemical shifts and 15N NOE patterns in dioctanoylphosphatidylglycerol (D8PG) micelles indicate the same structure, and intermolecular NOEs between D8PG and the rings of Phe-5, Phe-6, Phe-17 and Phe-27, as well as arginine side chains, show the whole amphipathic helix touching anionic lipid rather than a short inserted stub[2].

Two independently determined structures cover the same construct. PDB 2K6O is an NMR entry and 5NMN a 0.95 Å X-ray structure, both spanning residues 134–170 of the precursor; the helix annotation in the crystal structure runs 135–163, which is LL-37 residues 2–30[3]. The helix-to-tail boundary therefore survives a change of method.

Truncation tests the same question from the sequence side. KR-12, residues 18–29, was the smallest fragment that retained antibacterial activity, folded as a short three-turn amphipathic helix rich in positively charged side chains, and showed selective toxicity toward bacteria rather than human cells in that study[2]; UniProt separately annotates residues 17–29 as the active core[3]. Removing the N-terminal helix costs more than length alone: FF-33 (four residues shorter) and SK-29 (eight shorter) need higher anion concentrations to become helical and are less active than LL-37 against E. coli D21[1]. The N-terminus also contributes to proteolytic resistance and to haemolytic activity, though not to antimicrobial activity in the liposome assays that tested it[4]. Those findings describe different readouts rather than a conflict: the N-terminal helix governs how readily the peptide folds and how it disrupts host-cell membranes, while the cationic core carries much of the antibacterial action in the systems measured.

Reading the Evidence: What CD, NMR and FTIR Each Measure

Each technique answers a different question, and each answer is only as general as the sample it came from.

Method What it reports Sample used What it cannot establish
Circular dichroism Average helical content across the peptide population Water or defined buffers, micromolar peptide, varied salt and pH Which residues are helical; where the peptide sits on a membrane
NMR Atomic-resolution structure and direct lipid contacts SDS and D8PG micelles Behaviour in a bilayer or on a bacterium
Oriented ATR-FTIR Secondary structure plus helix orientation relative to the membrane plane Supported zwitterionic lipid bilayers Atomic detail, or the bend-helix model itself

CD gives the folding conditions. At micromolar concentration in water, LL-37 produces a disordered spectrum. Adding 15 mM bicarbonate, sulfate or trifluoroacetate converts it to a helix with roughly equal efficiency, whereas 160 mM chloride is less effective, and the transition becomes cooperative as peptide concentration rises, consistent with oligomer formation. Helical content falls below pH 5, the peptide is fully disordered at pH 2, and the helix survives above pH 13[1].

FTIR gives orientation, and it changed the mechanistic picture. On zwitterionic bilayers, the lipid composition used to mimic mammalian membranes, the peptide is predominantly α-helical and lies nearly parallel to the membrane surface; that result argues against a channel spanning the bilayer and for detergent-like, carpet-style disruption[4]. UniProt nevertheless annotates membrane permeabilisation through transmembrane pores as an in vitro function[3]. Both descriptions exist because the preparations differ. Treat "pore-forming" as one model supported by some systems, not as an established mechanism.

Helical Content and Activity: What the In Vitro Data Show

Where helix and killing were measured in the same assays, they moved together: the extent of α-helicity correlated with antibacterial activity against both Gram-positive and Gram-negative bacteria. In those experiments the minimal inhibitory concentration against E. coli was 5 µM, the peptide was cytotoxic to several eukaryotic cell lines at 13–25 µM, and human serum inhibited both the antibacterial and the cytotoxic activity[1]. Those figures belong to the buffers, strain and cells used; they are assay constants, not dose information of any kind.

Quaternary structure is a second variable. Engineered LL-37 dimers forced into parallel or antiparallel orientations by disulfide bonds gained a higher propensity to form stacked helices in solution and on anionic or neutral model membranes, and both their antibacterial and their cytotoxic effects depended strongly on which dimer geometry was used[5]. Because native LL-37 has no cysteine, those dimers are laboratory tools for fixing an oligomer state, not a version of the natural peptide; what they establish is that self-association changes the readout, consistent with the concentration-dependent folding seen by CD[1].

LL-37 also oligomerises under physiological solution conditions and binds molecular surfaces as an oligomer, and its functional range extends well past direct killing: alarmin activity, local immunomodulation, angiogenesis, wound healing and possibly elimination of abnormal cells have all been described[6]. Several of those effects run through host-cell receptors rather than through membrane lysis, so the helix explains membrane engagement without explaining every biological response attributed to the peptide.

Where the Evidence Stops: Micelles, Model Membranes and Living Systems

Three side-by-side panels showing a helical peptide wrapped on a spherical detergent micelle, the same helix lying parallel to a flat supported lipid bilayer, and a curved bacterial envelope with an outer polysaccharide layer and embedded membrane proteins.

Everything cited above comes from in vitro systems: water and defined buffers, SDS and D8PG micelles, zwitterionic and anionic liposomes, supported bilayers for FTIR, bacterial suspensions including E. coli D21, and cultured eukaryotic cells. No animal model and no human study appears in this evidence set, so nothing here supports a statement about efficacy, safety, pharmacokinetics or clinical use in people or animals.

The gaps between those systems matter in specific ways. A micelle is a curved detergent aggregate without a bilayer interior, so helix-bend-helix solved in SDS describes a lipid-facing conformation, not a pore. A liposome has a bilayer but no lipopolysaccharide, capsule, membrane proteins or cell wall. A supported planar bilayer reports orientation relative to a flat surface that no bacterial membrane resembles. A minimal inhibitory concentration is defined by inoculum, medium and strain, and a peptide that inhibits growth at 5 µM in one buffer is inhibited by serum[1]. Cytotoxicity toward cultured eukaryotic cells is not a prediction of human toxicity. Questions about a specific manufactured peptide — identity, purity, counterion, aggregation state of a given lot — are separate from the structural literature and are not settled by it.

Why the Alpha Helix Matters

Four points carry the answer. Helix formation depends on solution conditions — anions, peptide concentration and pH — and lipid interfaces, so the same sequence can be disordered or helical, and membrane activity is linked to local conditions rather than determined by a permanently exposed hydrophobic face. The helix is amphipathic, so binding and disruption come from the same surface: cationic residues reach for anionic lipid while the hydrophobic face stays in contact with it. The helix is interrupted, so the Gly-14–Glu-16 bend and the disordered tail mean the peptide is not one rigid rod, which fits the whole helix lying along a membrane surface rather than crossing it[2][4]. And the helix is unconstrained by disulfides; salt, pH and peptide concentration affect its helical content. Separately, human serum inhibits antibacterial and cytotoxic activity; this observation does not establish a change in helical content. These condition-dependent observations are why structures solved in micelles should not be read as a description of LL-37 on a living cell.

Related research: What Is LL-37? Antimicrobial Activity and Immune Signaling; Peptide Identity: Sequences, Fragments, Modifications, Salt Forms; LL-37.

Research Use Only: LL-37 and the related peptides discussed here are research materials. This article does not provide human or veterinary dosing, administration, compounding or clinical guidance, and nothing in it should be used to treat any condition.

References

  1. Johansson J, Gudmundsson GH, Rottenberg ME, Berndt KD, Agerberth B. Conformation-dependent antibacterial activity of the naturally occurring human peptide LL-37. J Biol Chem 1998 (PMID 9452503)
  2. Wang G. Structures of human host defense cathelicidin LL-37 and its smallest antimicrobial peptide KR-12 in lipid micelles. J Biol Chem 2008 (PMID 18818205)
  3. UniProtKB P49913 — Cathelicidin antimicrobial peptide (CAMP_HUMAN)
  4. Oren Z, Lerman JC, Gudmundsson GH, Agerberth B, Shai Y. Structure and organization of the human antimicrobial peptide LL-37 in phospholipid membranes. Biochem J 1999 (PMID 10417311)
  5. Xhindoli D, Pacor S, Guida F, Antcheva N, Tossi A. Native oligomerization determines the mode of action and biological activities of human cathelicidin LL-37. Biochem J 2014 (PMID 24117320)
  6. Xhindoli D, Pacor S, Benincasa M, Scocchi M, Gennaro R, Tossi A. The human cathelicidin LL-37 — a pore-forming antibacterial peptide and host-cell modulator. Biochim Biophys Acta 2016 (PMID 26556394)