LL-37 vs. Defensins vs. Magainin: Antimicrobial Peptide Comparison

If you are trying to understand the antimicrobial peptide field, three names come up constantly: LL-37, defensins, and magainin. They are the most studied AMPs in existence, and they represent three fundamentally different approaches to the same problem -- killing bacteria.

But they are not interchangeable. They differ in structure, mechanism, spectrum of activity, clinical development status, and practical limitations. This comparison breaks down each peptide class across the dimensions that matter most for understanding their therapeutic potential.

A 2025 computational and biophysical analysis published in PMC directly compared the structural and mechanistic differences between LL-37, HNP-1 (a defensin), and magainin-2, confirming that these three peptides represent "mechanistically distinct AMPs" with "three major functional classes of membrane-interacting peptides."

Structure at a Glance

Feature	LL-37	Defensins	Magainin-2
Source	Human (cathelicidin precursor hCAP-18)	Human (neutrophils, epithelial cells, Paneth cells)	African clawed frog (Xenopus laevis)
Length	37 amino acids	18-45 amino acids (varies by subtype)	23 amino acids
Structure	Long amphipathic alpha-helix	Rigid beta-sheet (3 disulfide bonds)	Short, stable alpha-helix
Disulfide bonds	None	Three (defining structural feature)	None
Net charge	+6 at physiological pH	+1 to +5 (varies)	+4
Flexibility	High -- disordered in solution, folds on membrane contact	Low -- locked by disulfide bridges	Moderate -- well-ordered helix

LL-37: The Shape-Shifter

LL-37 is the only cathelicidin-derived AMP in humans. It is processed from its precursor protein hCAP-18 during inflammation and has 37 amino acid residues (hence the name). In solution, LL-37 is largely disordered. When it contacts a bacterial membrane, it folds into a long amphipathic alpha-helix -- a structural shift that activates its antimicrobial function.

This "structural plasticity" is one of LL-37's defining features. The 2025 computational analysis found that LL-37 has moderate pLDDT scores and significant terminal disorder, indicating a peptide that adapts its shape to different membrane environments. That adaptability makes LL-37 effective against diverse pathogens but also means its behavior is harder to predict and control pharmacologically.

Defensins: The Armored Tanks

Defensins take the opposite structural approach. Their three disulfide bonds create a rigid, compact beta-sheet fold that is extremely resistant to heat, pH changes, and proteolytic degradation. This stability is both their greatest strength and, paradoxically, a challenge for drug development -- the complex disulfide pattern is expensive to reproduce synthetically.

Human alpha-defensins (HNP-1 through HNP-4, HD5, HD6) and beta-defensins (hBD-1 through hBD-4) differ in their disulfide connectivity, tissue expression patterns, and antimicrobial spectra. For clinical purposes, hBD-3 is the standout: it has the broadest activity, including potent killing of MRSA even in high-salt environments.

Magainin-2: The Frog's Legacy

Magainin was discovered in 1987 by Michael Zasloff, who noticed that surgical incisions on African clawed frogs healed without infection in non-sterile aquarium water. The peptides secreted from frog skin turned out to have broad antimicrobial activity.

Magainin-2 forms a short, stable alpha-helix -- simpler than LL-37's long helix and lacking the defensins' disulfide architecture. This structural simplicity made magainin-derived peptides (especially pexiganan/MSI-78) attractive for drug development early on, as they were relatively easy to synthesize.

Mechanisms of Action

Mechanism	LL-37	Defensins	Magainin-2
Primary model	Carpet model / mixed	Membrane perturbation	Toroidal-pore model
Membrane interaction	Accumulates on surface, causes dissolution	Electrostatic binding, pore formation	Inserts to form lipid-peptide pores
Intracellular targets	DNA binding, LPS neutralization	Varies by subtype	Minimal (primarily membrane-active)
Immunomodulation	Strong (chemokine-like activity, immune cell recruitment)	Moderate (hBD-2/3 recruit dendritic cells, T cells)	Minimal
Anti-biofilm activity	Strong (active below MIC)	hBD-3 effective against staphylococcal biofilms	Limited data
Quorum sensing interference	Yes (downregulates Las/Rhl systems)	Limited evidence	Not documented

How LL-37 Kills

LL-37 is a generalist. It uses the carpet model of membrane disruption -- accumulating on the bacterial surface until it reaches a critical concentration, then dissolving the membrane like detergent. But that is only part of the story.

LL-37 also downregulates quorum sensing systems (the Las and Rhl pathways in P. aeruginosa), neutralizes bacterial lipopolysaccharide to prevent septic shock, and activates the NLRP3 inflammasome in macrophages. A 2025 review confirmed LL-37 combats over 38 bacteria, 16 fungi, and 16 viruses through various mechanisms including membrane rupture, biofilm suppression, and disruption of viral envelopes.

This multifunctionality -- direct killing plus immunomodulation plus biofilm disruption -- is what makes LL-37 unique. No other single AMP does this much. But it also makes LL-37 harder to develop as a drug, because its effects are complex and context-dependent.

How Defensins Kill

Defensins kill primarily through membrane perturbation. Their rigid, cationic structure binds electrostatically to negatively charged bacterial membranes, disrupting integrity and causing cell death. The specific mechanism varies by defensin subtype.

HD5 (alpha-defensin 5) from Paneth cells has a well-characterized membrane perturbation mechanism that is effective against both gram-positive and gram-negative bacteria. HD6 does not directly kill bacteria at all -- instead, it self-assembles into nanonets that physically trap microorganisms.

Beta-defensin 3 (hBD-3) has the broadest spectrum and shows particular strength against staphylococcal biofilms. In titanium implant models, hBD-3 reduced intact bacterial colonies in a dose-dependent manner even against bacteria already producing biofilm. Read more about defensin drug development in our defensins clinical development article.

How Magainin Kills

Magainin-2 is primarily a membrane-disrupting peptide that works through the toroidal-pore model. The peptides insert into the membrane and cause the lipid bilayer to curve inward, creating pores lined by both peptide and lipid. Cellular contents leak out and the bacterium dies.

The mechanism is straightforward -- and that simplicity is both an advantage (predictable behavior, easier to optimize) and a limitation (narrower therapeutic applications compared to multifunctional peptides like LL-37).

Antimicrobial Spectrum

Pathogen Type	LL-37	Defensins	Magainin-2/Pexiganan
Gram-positive bacteria	Active (moderate potency)	Active (hBD-3 strongest; effective vs. MRSA)	Active (broad spectrum)
Gram-negative bacteria	Active	Active (varies by subtype)	Active (broad spectrum)
MRSA	Active but moderate	hBD-3 highly active even in high salt	Pexiganan active (no resistance detected)
Pseudomonas aeruginosa	Active (biofilm prevention noted)	hBD-2/3 active	Active
Fungi	Active against 16+ species	Variable	Some activity
Viruses	Active against 16+ viruses	Active (especially against enveloped viruses)	Limited data
Anaerobic bacteria	Active	Limited data	Pexiganan tested against 3,109 isolates including anaerobes
Biofilms	Strong anti-biofilm below MIC	hBD-3 anti-biofilm on titanium	Limited

LL-37 Spectrum

LL-37 has genuinely broad-spectrum activity, but its direct antimicrobial potency under physiological conditions is often described as "weak to moderate." Where LL-37 excels is in anti-biofilm activity (effective at 1/16 to 1/32 of its MIC against planktonic bacteria) and immunomodulation.

Clinically relevant finding: LL-37 showed greater anti-biofilm activity against S. aureus on titanium surfaces than either silver nanoparticles or conventional antibiotics. For applications like implant infection prevention, this matters more than raw bactericidal potency.

Defensin Spectrum

Among defensins, hBD-3 stands out for clinical relevance. It was tested against 44 clinical S. aureus isolates (22 MRSA), and showed the greatest activity of any human beta-defensin. It maintained activity in high-salt conditions -- a property that matters for wound environments and body fluids where other AMPs lose effectiveness.

Alpha-defensin HD5 is potent against gut pathogens, consistent with its role in intestinal Paneth cell defense. HD6's nanonet mechanism is active against Salmonella but through physical entrapment rather than direct killing.

Magainin/Pexiganan Spectrum

Pexiganan (MSI-78), the synthetic magainin analog, has the most extensively documented antimicrobial spectrum of the three. In one comprehensive study, it was tested against 3,109 clinical isolates -- gram-positive and gram-negative, aerobic and anaerobic. Susceptibility was equivalent regardless of whether bacteria were antibiotic-sensitive or resistant, and no resistance to pexiganan was detected.

This broad-spectrum, resistance-proof profile explains why pexiganan attracted significant pharmaceutical investment despite ultimately failing to gain FDA approval.

Clinical Development Status

Clinical Metric	LL-37	Defensins	Magainin/Pexiganan
Most advanced trial	Phase 1/2 (LL-37-derived, melanoma)	Phase 2 completed (brilacidin, multiple indications)	Phase 3 completed (pexiganan, diabetic foot ulcers)
FDA approval	None	Alpha-defensin ELISA approved (diagnostic use only)	Denied (1999); additional trials conducted
QIDP designation	No	Yes (brilacidin)	No
Total human subjects tested	Limited	500+ (brilacidin)	Hundreds (across Phase 3 trials)
Active clinical programs	LL-37-derived antitumor peptide	Brilacidin (oral mucositis Phase 3 pathway); hBD-2 (immune disorders)	Limited current activity

LL-37 Clinical Status

LL-37 clinical development has focused on oncology rather than antimicrobial applications. A Phase 1/2 trial completed in 2024 evaluated an LL-37-derived peptide injected directly into melanoma tumors. The study demonstrated safety, tolerability, and modulation of the tumor microenvironment.

For antimicrobial applications, LL-37's cytotoxicity to human cells at bactericidal concentrations limits systemic use. Topical and derivative formulations are under investigation but have not reached advanced trials.

Defensin Clinical Status

Brilacidin, the defensin-mimetic, is the clear leader here with completed Phase 2 trials in skin infections, oral mucositis, inflammatory bowel disease, and COVID-19. Its QIDP designation and agreed Phase 3 pathway for oral mucositis make it the AMP drug candidate closest to potential approval.

Defensin Therapeutics ApS continues developing recombinant hBD-2 for immune-mediated diseases. HD5-inspired peptidomimetics are in preclinical development.

Magainin/Pexiganan Clinical Status

Pexiganan reached Phase 3 twice for diabetic foot ulcers but was denied FDA approval in 1999 because it failed to demonstrate superiority over standard care. Additional randomized trials (NCT01594762, NCT01590758) were conducted, but pexiganan has not been approved.

The pexiganan story is often cited as a cautionary tale for AMP drug development: strong in vitro data and a clean safety profile are necessary but not sufficient. The clinical bar is efficacy superior to existing treatments, and that bar has proved difficult for topical AMPs to clear.

Pros and Cons Comparison

LL-37

Advantages:

Only human cathelicidin -- endogenous, well-tolerated at physiological levels
Strongest immunomodulatory profile of any AMP
Potent anti-biofilm activity at sub-MIC concentrations
Quorum sensing interference documented
Broad antiviral activity

Disadvantages:

Cytotoxic to human cells at therapeutic concentrations
Susceptible to protease degradation in serum
High production cost
Complex, context-dependent behavior complicates drug development
No approved antimicrobial applications

Defensins

Advantages:

Naturally stable structure (disulfide bonds resist proteases and heat)
hBD-3 active against MRSA in high-salt conditions
Defensin-mimetic (brilacidin) furthest along in clinical development
Immunomodulatory and wound-healing properties (hBD-3 promotes keratinocyte migration)
Diagnostic applications already FDA-approved (alpha-defensin)

Disadvantages:

Complex disulfide folding makes manufacturing expensive
Activity inhibited by serum and physiological salt (except hBD-3)
Natural defensins have narrow therapeutic windows
Limited clinical trial data for defensin peptides themselves (most data is on mimetics)

Magainin/Pexiganan

Advantages:

Simplest structure -- easiest to synthesize and modify
Largest documented antimicrobial spectrum (3,109 clinical isolates tested)
No resistance development detected
Activity independent of antibiotic resistance status
Good safety profile in humans

Disadvantages:

Failed to outperform standard care in Phase 3 trials
Minimal immunomodulatory activity
Limited anti-biofilm data
Derived from frog (not human) -- potential immunogenicity concerns for systemic use
FDA rejection has dampened pharmaceutical investment

Practical Applications: Which Peptide for Which Use?

Application	Best Candidate	Why
Chronic wound infection	LL-37 derivatives or hBD-3	Anti-biofilm activity + wound healing promotion
Implant infection prevention	hBD-3 or LL-37	Both active against biofilms on titanium; hBD-3 stable in high salt
Topical skin infection (MRSA)	hBD-3 or brilacidin	Potent anti-MRSA activity; brilacidin has Phase 2 data
Broad-spectrum topical	Pexiganan analogs	Widest documented spectrum; good safety
Immunomodulatory therapy	LL-37	Strongest immune-activating profile
Anti-biofilm coating	LL-37 (sub-MIC)	Effective biofilm prevention at very low concentrations
Oral/systemic delivery	Brilacidin (defensin mimetic)	Small molecule; resists proteases; oral bioavailability possible
Acne / skin inflammation	LL-37/defensin derivatives	Both upregulated in acne; AMP-based acne treatments under development

Cost and Manufacturing Comparison

Factor	LL-37	Defensins	Magainin/Pexiganan
Synthesis complexity	Moderate (37 residues, no disulfides, but long chain)	High (3 disulfide bonds must fold correctly)	Low (23 residues, no disulfides)
Recombinant production	Difficult (toxic to host cells; requires fusion protein)	Difficult (disulfide folding in bacteria)	Moderate (simpler structure)
Estimated cost per gram	High ($10,000-40,000+)	Very high (correct folding adds cost)	Moderate to high (simplest of the three)
Mimetic available?	LL-37 fragments (FK-16, KR12, GF-17)	Yes (brilacidin -- small molecule)	Yes (pexiganan IS the optimized analog)
Patent situation	Limited (natural human peptide)	Broad (mimetics and analogs patented)	Mixed (pexiganan patents; original magainin natural)

The manufacturing comparison strongly favors the mimetic approach. Brilacidin, as a non-peptide small molecule, avoids the synthesis challenges of all three natural peptide classes. Pexiganan, while still a peptide, benefits from its short length and lack of disulfide bonds.

For researchers weighing which AMP class to develop, the manufacturing question often determines the answer: the peptide with the best activity profile may not be the peptide that can be produced at commercial scale. This practical reality has shaped the clinical pipeline, where defensin mimetics and magainin analogs lead while natural LL-37 and defensins remain largely in preclinical development.

The Convergence: Hybrid and Engineered Peptides

The future of AMP therapeutics may not require choosing between these three classes. Researchers are now creating hybrid peptides that combine the best features of each.

An LL-37/Renalexin hybrid peptide published in Applied Microbiology and Biotechnology (2024) demonstrated antimicrobial activity at lower MIC values than either parent peptide alone. Multi-domain peptides like 8DSS-C8-P113 combine antimicrobial, signaling, and tissue-repair functions in a single molecule.

AI-driven design is accelerating this approach. Machine learning models trained on the structural and functional data of all three peptide classes can generate novel sequences that combine LL-37's immunomodulatory properties with defensin stability and magainin-like broad-spectrum killing. These computationally designed peptides are expected to enter preclinical testing in 2026-2027.

The Bottom Line

There is no single "best" antimicrobial peptide. LL-37 is unmatched for immunomodulation and biofilm disruption. Defensins (and their mimetic brilacidin) are closest to clinical approval. Pexiganan has the broadest documented spectrum and the most human safety data.

For the antimicrobial peptide field to reach its potential as a real alternative to failing antibiotics, the most likely path forward involves engineered peptides and mimetics that draw on the strengths of all three classes -- combined with the advanced delivery systems (wound dressings, nanoparticles, hydrogels) needed to get these molecules to infection sites safely and effectively.

Related Reading on PeptideJournal.org: