Ethics statement

All animal experiments were approved by the Institutional Review Board and Animal Care and Use Committee of the Jiangsu Institute of Parasitic Diseases (Ethical Approval No. JIPD-2025-015). All procedures were performed in accordance with the National Guidelines for the Care and Use of Laboratory Animals in China.

Animals

C57BL/6 wild-type mice were purchased from Vital River Laboratory Animal Technology Co. Ltd. Cramp knockout mice (Cramp^−/−), generated on a C57BL/6 background, were also obtained from Vital River. These mice carry a targeted deletion of exons 3 and 4 of the Cramp gene, which encode the entire mature CRAMP peptide, resulting in a complete functional knockout [29]. All mice were housed under standard conditions with a 12-hour light/dark cycle with free access to food and water. All mice were acclimatized to the housing for 1 week before the study. All animals were anaesthetized by isoflurane inhalation (2%) and then euthanized by cervical dislocation. All efforts were made to minimize animal suffering.

Peptides

All peptides were synthesized by GL Biochem Ltd. (Shanghai, China). The synthesized peptides were analyzed by reversed-phase high-performance liquid chromatography and mass spectrometry to confirm purity higher than 98%. All peptides are water-soluble and were dissolved in PBS to make a 20 mM stock solution. The sequences of peptides involved are as follows:

1. LL-37: LLGDFFRKSKEKIGKEFKRIVQRIKDFLRNLVPRTES;
2. Mouse CRAMP: GLLRKGGEKIGEKLKKIGQKIKNFFQKLVPQP;
3. LE-16: LLGDFFRKSKEKIGKE;
4. FR-13: FKRIVQRIKDFLR.

Parasite strains and cells

All the parasite strains used in the study, including P. falciparum 3D7 (drug-sensitive, MR4 Cat. No. MRA-102), P. falciparum 803 (artemisinin-resistant) [30], P. falciparum Dd2 (chloroquine-resistant, MR4 Cat. No. MRA-156G) [31], and P. berghei ANKA (MR4 Cat. No. MRA-311), were stored in our laboratory. Human erythrocytes were obtained from Wuxi Blood Center.

Plasmodium culture

Blood stages of the laboratory clones P. falciparum 3D7 (sensitive strain), P. falciparum 803 (artemisinin-resistant strain), and P. falciparum Dd2 (chloroquine-resistant strain) were cultured in vitro according to our previous method [32]. Briefly, cultures were maintained in O-positive human RBCs as host cells at 2% hematocrit and maintained in complete medium (CM) [RPMI-1640 culture medium supplemented with NaHCO₃ (2 mg/mL), hypoxanthine (50 μg/mL), HEPES (5.96 mg/mL), Albumax II (1%), and gentamicin (40 μg/mL)]. The cultures were diluted to 1% parasitemia and 2% hematocrit in CM and incubated in a humidified atmosphere with 5% O₂, 5% CO₂, and 90% N₂ at 37°C. Prior to experiments, the cultures were highly synchronized for ring stages using 5% sorbitol twice at a 40-h interval. Briefly, ring stage cultures were centrifuged and the pellet was washed once in incomplete medium (ICM, RPMI-1640 without addition albumax). Then, the pellet was resuspended in 5% sorbitol and incubated at 37°C for 10 min. After incubation, the suspension was centrifuged and the pellet was washed once and resuspended in medium.

For schizont enrichment, parasitized erythrocytes from asynchronous Plasmodium falciparum cultures (>5% parasitemia) were concentrated by centrifugation (800 × g, 5 min) and resuspended in 1 mL RPMI 1640. A discontinuous Percoll gradient was prepared in a 15-mL tube by layering 3 mL of 40% iso-osmotic Percoll (IOP; density 1.075 g/mL) over 3 mL of 70% IOP (density 1.118 g/mL). The cell suspension was gently layered atop the gradient and centrifuged at 1,000 × g for 20 min at room temperature (brake disengaged). Schizonts were collected from the 40%/70% interface, washed twice in complete medium, and resuspended for downstream applications.

The development of the parasite was observed using light microscopy, and parasitemia was determined by counting at least 10,000 RBCs on Giemsa’s solution-stained thin blood smears. To ensure systematic and non-overlapping counts, a standardized protocol was followed using a customized gridded eyepiece reticule. The slide was moved methodically in a unidirectional pattern (left-to-right, top-to-bottom), with all RBCs within each sequential grid field counted until the target of 10,000 RBCs per sample was reached. This approach eliminated field duplication and ensured accurate parasitemia calculation.

Assessment of In vitro antimalarial activity of peptides

A 3-day inhibition assay was used to test the antimalarial activity of LL-37 and its two truncated peptides FR-13 and LE-16. Briefly, after highly synchronization of ring stages, an aliquot of parasite inoculum (200 μL) with 1% parasitemia and 2% hematocrit was added into each well of a 96-well plate followed by a series of concentrations of LL-37, FR-13 or LE-16 peptide (dissolved in PBS and diluted with RPMI 1640). The concentrations of LL-37, FR-13 or LE-16 peptide were 1, 3, 4, 5, 7, 9, 12 and 18 μM. After 72 hours of incubation, the antimalarial effects of peptides were estimated by IC₅₀. Thin blood films stained with Giemsa’s solution were counted under a microscope. IC₅₀ was calculated by GraphPad Prism. Three independent assays with three technical replicates each time were conducted.

Stage-specific and time-dependent parasite inhibition assay

To analyze both stage-specific and time-dependent antimalarial activity, highly synchronized ring-stage P. falciparum 3D7 parasites were cultured in 48-well plates at a starting 0.5-1% parasitemia and 2% hematocrit. LL-37 (IC₅₀ concentration) was added to parasites at distinct developmental stages: rings (5 hours post-invasion, hpi), trophozoites (24 hpi), and schizonts (36 hpi). At each developmental stage, parasites were exposed to the peptide for specific durations: 3, 6, 9, or 12 hours. Following peptide exposure, parasites were washed three times with complete medium (CM) and transferred into new wells to remove the peptide. Re-invaded parasites were subsequently analyzed at 50 hpi. Parasitemia at 50 hpi was diluted 1:40 with fresh human red blood cells (RBCs) and cultured for an additional 4 days, with CM as the blank control. Parasitemia levels were determined by microscopic examination of Giemsa-stained smears. Viability was calculated as (parasitemia of LL-37-treated group/ parasitemia of untreated control group) × 100%. Lysed iRBCs were excluded from counting by excluding ghost cells (translucent, shrunken morphology) in Giemsa-stained smears; only intact RBCs (infected or uninfected) were counted to ensure accuracy. All assays were performed in three independent biological replicates, each consisting of three technical replicates.

Hemolytic activity

In vitro, hemolysis assays were performed using both uninfected human RBCs [32] and P. falciparum-infected RBCs. For uninfected RBCs, different peptide dilutions were incubated in triplicate with 2% hematocrit RBCs in PBS at 37°C for 15, 30, and 60 min. To evaluate potential hemolytic toxicity of LL-37, C57BL/6 mice infected or not were administered LL-37 intravenously at the maximum dose used in efficacy studies (16 mg/kg/day) for 4 consecutive days. Control mice received equivalent volumes of saline. Peripheral blood was collected 24 h after the final dose. After centrifugation (1,000 × g, 10 min), supernatant absorbance at 540 nm was measured. 1% Triton X-100 and PBS served as maximum hemolysis and negative controls, respectively, with hemolysis percentage calculated as [(A540 (peptide)- A540 (PBS))/ (A540 (Triton X-100)- A540 (PBS))]×100. For infected RBCs, schizont-enriched cultures (>70% purity) were tested. To assess stage-specific activity, enriched schizonts were diluted 1:10 with fresh RBCs and incubated to allow reinvasion, yielding ~30% ring-stage parasitemia, followed by tight synchronization with 5% sorbitol. LL-37 was then applied to synchronized cultures at ring, trophozoite, and schizont stages followed by identical hemolysis quantification.

Mechanistic Analysis of LL-37 Targeting to Infected Erythrocytes via Membrane Modulation

To investigate the mechanism by which LL-37 targets Plasmodium falciparum-infected erythrocytes, the membrane properties of uninfected erythrocytes were modulated to mimic the infected phenotype through PS externalization and cholesterol depletion. PS externalization was induced by incubating erythrocytes with calcium ionophore A23187 (MCE, HY-N6687) (0.125, 0.25, 0.5, 0.75 and 1 μM) in PBS containing 1 μM CaCl₂ for 3 h at 37°C [33]. Cholesterol depletion was achieved using methyl-β-cyclodextrin (MβCD; MCE, HY-101461) (1.5, 3, 4.5, 6, and 7.5 mM) in PBS for 2 h at 37°C [18,33]. For combined treatments, cells were either pre-treated with fixed MβCD (1.5 mM) followed by varying A23187, or pre-treated with varying MβCD followed by fixed A23187 (0.75 μM), with washing between steps. All treated erythrocytes were adjusted to 2% hematocrit and incubated with LL-37 (concentration of IC₅₀) for 15 min at 37°C, and hemolysis was quantified.

Cholesterol repletion and DOTAP competition assays to validate LL-37 targeting mechanism

To further validate the roles of phosphatidylserine and cholesterol in the antiplasmodial activity of LL-37, cholesterol reconstitution experiments were performed on normally cultured parasites, along with the addition of the cationic lipid DOTAP (1,2-dioleoyl-3-trimethylammonium propane (HY-112754A, MCE)) to competitively bind externalized phosphatidylserine and interfere with LL-37 binding [34]. For cholesterol replenishment, a 20 mg/mL cholesterol solution in ethanol was mixed with a 5% (wt/vol) aqueous methyl-β-cyclodextrin (MβCD) solution: 10 µL of cholesterol solution was added to 500 µL MβCD solution, heated at 80 °C for 10 min with repeated mixing until clear, repeated four times for a total of 50 µL cholesterol, then lyophilized in a speed vacuum concentrator until a fluffy powder remained, which was stored at −20 °C [35]. Before use, the MβCD-cholesterol complex was reconstituted in 400 µL culture medium, vortexed, and filter-sterilized through a 0.22 μm syringe filter. Ring-stage parasites (0–3 h post-invasion) were treated with the MβCD-cholesterol complex at dilution ratios of 1:20, 1:40, and 1:80 in a 96-well plate for 38 h at 37 °C, then washed with incomplete culture medium (ICM) to remove the complex, resuspended in complete medium (CM), transferred to a new 96-well plate, and incubated with LL-37 at its IC₅₀ for 6 h followed by smear preparation and parasitemia quantification. Membrane cholesterol levels in iRBCs post-cholesterol repletion were quantified using the Amplex Red Cholesterol Assay Kit (Beyotime, China; Catalog No.S0211S) strictly following the manufacturers’ instructions, with all samples run in triplicate. For DOTAP supplementation assays, late-stage parasites (36–40 h, ∼4% parasitemia) were treated with DOTAP at 5, 10, or 20 μM, alongside blank control, and vehicle control controls. After incubation with or without LL-37 for 6 h, parasitemia was determined from smears. DOTAP was dissolved in dimethyl sulfoxide (DMSO) to prepare a high-concentration stock solution. Working concentrations of DOTAP (including 20 μM and other test concentrations) were generated by serial dilution of the stock solution with RPMI 1640 culture medium. The vehicle control group was prepared by adding an equal volume of pure DMSO to the same culture medium, ensuring the final DMSO percentage was identical across all DOTAP treatment groups and the vehicle control (≤0.1%). Flow cytometry was used to analyze the competitive binding of LL-37 and DOTAP to PS on late-stage iRBCs. Enriched late-stage parasites were washed with pre-warmed PBS, aliquoted into tubes (400 μL per tube), and divided into five groups: Hoechst single-staining, Annexin V single-staining, negative control (N group), LL-37 (5 μM), and DOTAP (20 μM). The two single-staining groups were stained separately to establish gating parameters for flow cytometry analysis. All groups were supplemented with 4 μL Hoechst staining solution, incubated at 37°C for 5 min, washed twice with 1× Binding Buffer, and resuspended in 400 μL 1× Binding Buffer. Annexin V staining solution, DOTAP, and LL-37 were added sequentially according to grouping, incubated at 37°C for 20 min in the dark, and samples were analyzed by flow cytometry. Unbound LL-37 in the supernatant was quantified by ELISA to verify DOTAP-mediated PS binding competition. Enriched late-stage parasites were washed with pre-warmed PBS, aliquoted into tubes (400 μL per tube), and divided into four groups: LL-37 (5 μM) alone, LL-37+DOTAP (5 μM), LL-37+DOTAP (10 μM), and LL-37+DOTAP (20 μM). Pre-diluted DOTAP was added to achieve the indicated working concentrations; the LL-37 alone group received equal volume of DMSO at the same dilution ratio. All groups were supplemented with LL-37 to a final concentration of 5 μM, mixed thoroughly, incubated at 37°C for 5 min, and centrifuged at 3,000 rpm for 50 s to collect the supernatant. The concentration of LL-37 in supernatant was examined by Human LL-37 ELISA Kit (Sangon Biotech, Shanghai, China; Catalog No. D711299) following the manufacturer’s instructions. For the combined cholesterol and DOTAP treatment, ring-stage parasites (0–3 h) were pretreated with MβCD-cholesterol complex at a 1:40 dilution for 38 h, washed with ICM, resuspended in CM, then co-treated with 20 μM DOTAP (or DMSO) and LL-37 for 6 h before parasitemia assessment.

Parasite infection and in vivo antimalarial assay

For in vivo infection, Plasmodium berghei ANKA (P.b ANKA) parasites were thawed rapidly at 37°C and injected intraperitoneally (i.p.) into donor mice. Parasites were passaged in C57BL/6J mice. Donor mice exhibiting approximately 30% parasitemia were euthanized, and blood was collected for infection of mice in the antimalarial assay.

The antimalarial activity of test peptides was evaluated using two therapeutic regimens in female C57BL/6J mice. For the prophylactic/early treatment model, the classical 4-day suppressive test was used as previously described [36]. On day 0, mice were inoculated with 1.0 × 10⁶ P.berghei-infected erythrocytes harvested from donor mice with the infection route differing by assessment batch: mice in the Endpoint-Assay Batch received an intravenous (i.v.) injection to achieve synchronous parasitemia for acute analysis, while mice in the Survival-Assay Batch received a standard intraperitoneal (i.p.) injection for long-term monitoring. One hour post-infection, treatment was initiated intravenously once daily for four days, administering the test peptides at 1, 4, or 16 mg/kg, chloroquine diphosphate (16 mg/kg) as a positive control, or saline as a negative control (n = 6 per group).

For the therapeutic intervention model, designed to mimic therapy after infection establishment, all mice were infected via i.p. injection. Treatment with a high dose of the test peptides (16 mg/kg), chloroquine (16 mg/kg), or saline (n = 6 per group) was administered intravenously for four consecutive days only after peripheral parasitemia reached 8–15%, with this regimen applied to separate Endpoint- and Survival-Assay Batches that were both infected via the i.p. route.

In the Endpoint-Assay Batches for both models, mice were euthanized on day 4, 24 hours after the final dose for terminal sampling. Parasitemia was evaluated from thin tail blood smears from Day 0 to Day 4. These smears were fixed with methanol, stained with 10% Giemsa for 15 min, and examined under a light microscopy (100× objective). Parasitemia was quantified by counting parasitized erythrocytes among ≥3,000 red blood cells per smear. For histopathology, liver, spleen and brain were collected immediately after euthanasia, and rinsed three times in ice-cold phosphate-buffered saline (PBS) to remove residual blood, and fixed in 4% paraformaldehyde for 24 h at 4°C. Fixed tissues were paraffin-embedded, sectioned, and stained with hematoxylin and eosin (H&E) for histological evaluation. Plasma was isolated by centrifugation at 3,500 × g for 30 min at room temperature. Cytokine concentrations (IL-1β, IL-6, IFN-γ, TNF-α) were measured in plasma using commercial ELISA kits (Dakewe, China; Catalog Nos.: 1210122, 1210602, 1210002, 1217202) according to the manufacturer’s protocols.

In the Survival-Assay Batches for both models, a minimal confirmation of parasitemia was performed on Day 4, after which mice were monitored daily for 21 days. To assess disease progression, mice were evaluated according to the rapid murine coma and behavior scale (RMCBS) for quantitative assessment of murine cerebral malaria (CM), as described [37] with minor modifications. Briefly, six parameters were used to evaluate the animals, including ruffled fur, hunched posture, wobbly gait, limb paralysis, convulsions, and coma. Each parameter was scored from 0 to 2. For example, fur condition was scored as 0 for ruffled/piloerected hair, 1 for dusty hair or hair out of place, and 2 for normal, clean, and shiny hair. In this scale, the minimum RMCBS score was 0 and the maximum score was 12. Following infection, the assessment was performed every 24 hours until 7 days post-infection (p.i.) and every 12 hours thereafter. Based on Evans blue staining confirming blood-brain barrier disruption, an RMCBS score of ≤ 6 was defined as indicating the onset of experimental cerebral malaria (ECM). Mice were euthanized when they reached a humane endpoint criterion of RMCBS scores of 6 or less, and this timepoint was defined as “end stage”. All tissue/blood samples were collected 24 hours after the final dose. Humane endpoints were strictly enforced: moribund mice exhibiting irreversible symptoms were euthanized by cervical dislocation under anesthesia (isoflurane 2%) and the time of death was recorded. All surviving mice were humanely euthanized at 21 d.p.i. for endpoint analysis.

Detection and statistical analysis of serum LL-37 expression in human and mouse models

The study involving human subjects was conducted in accordance with relevant ethical guidelines and regulations. This study utilized existing, anonymized plasma samples obtained from the Jiangsu Malaria Biobank. The samples consisted of specimens from 63 patients with confirmed P.falciparum-infection and 63 age- and sex-matched healthy controls. The secondary use of these de-identified samples for research is permitted under national ethical regulations, which grant an exception to the requirement for project-specific informed consent. The operational protocols of the biobank and the specific research protocol for this study were reviewed and granted an exemption by the Ethics Committee of the Jiangsu Institute of Parasitic Diseases (Approval No. JIPD-2022–005).

Meanwhile, 12 female C57BL/6 mice (6–8 weeks old) were infected with 1.0 × 10⁶ P.berghei ANKA-infected erythrocytes via intraperitoneal injection alongside 12 uninfected control mice in compliance with animal care protocols. Mice were culled at day 4 post-infection (when parasitemia reached ~8–15%) for plasma collection. Following infection, mouse blood was collected via cardiac puncture under anesthesia. Human and mouse plasma samples were separated by centrifugation at 3,000 × g for 10 minutes and stored at −80°C until further analysis. For quantification of LL-37, sample concentrations were measured using species-specific commercial ELISA kits, human LL-37 ELISA Kit (Sangon Biotech (Shanghai) Co., Ltd.) and murine CRAMP ELISA Kit (Cat# CSB-E15061m, CUSABIO, USA), strictly following the manufacturer’s instructions, with all samples run in triplicate. Finally, statistical analysis was performed using GraphPad Prism software, where data were expressed as mean ± SD and median values, and group comparisons were carried out with the non-parametric Mann-Whitney U test, considering a p-value of less than 0.05 as statistically significant.

Statistical analysis

Continuous variables were summarized as means ± standard error of the mean (SEM). Two-group comparisons used two-tailed unpaired Student’s t-test. Multi-group analyses employed one-way analysis of variance (ANOVA) with Dunnett’s multiple comparisons test (comparisons versus control group). Survival curves were compared by log-rank test. Statistical significance was defined as p < 0.05 (two-tailed). All analyses were performed using GraphPad Prism v10.0.

1. In Vitro antimalarial activity of LL-37 and analysis of its truncated peptides

To evaluate the antimalarial activity of LL-37 and the functional region responsible for the antimalarial activity of LL-37, two truncated peptides, the highly hydrophobic LE-16 fragment (residues 1–16) and the amphipathic α-helical FR-13 fragment (residues 17–29) (Fig 1a, 1b), were designed and evaluated alongside the full-length peptide. The efficacy against the Plasmodium falciparum 3D7 strain of LL-37 and its truncated peptides was assessed in vitro. LL-37 exhibited antimalarial activity, with a half-maximal inhibitory concentration (IC₅₀) of about 5.3 ± 0.3 μM against P. falciparum 3D7 strain (Fig 1c). Interestingly, at lower concentrations (below ~5 μM), all peptides, including both fragments, reached a similar partial inhibition plateau of about 40%. However, only the full-length LL-37 demonstrated a capacity to surpass this plateau at higher concentrations, achieving complete growth inhibition. Although both fragments retained partial antiplasmodial activity, their overall efficacy was markedly reduced compared to intact LL-37 (Fig 1c), suggesting that neither the C-terminal helical domain nor the N-terminal hydrophobic segment alone is sufficient to recapitulate the full activity of LL-37. The markedly enhanced potency of the full-length peptide implies that its structural integrity, likely involving cooperative interactions between domains, is critical for optimal antimalarial action. Furthermore, LL-37 displayed consistent efficacy against drug-resistant strains, with IC₅₀ values of 5.5 ± 0.3 μM for artemisinin-resistant 803 and 5.4 ± 0.3 μM for chloroquine-resistant Dd2 (Fig 1d), indicating that its mechanism remains effective across genetically distinct parasites.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. In vitro antimalarial activity of LL-37 and its truncated peptides.

(a) Helical wheel analysis of antimicrobial peptide LL-37. The helical wheel diagram was generated using HeliQuest to visualize the spatial distribution of residues. Residues are color-coded by physicochemical properties: yellow (hydrophobic: I, L,V,F), blue (positively charged: K,R), purple (uncharged polar: S,T), red (negatively charged: D,E), pink (uncharged polar amide: Q), green (non-polar: P), gray (aliphatic: G,A). The projection angle is rotated to maximize amphipathicity with hydrophobic residues (yellow) clustered on one face and polar/charged residues (blue/red/purple) on the opposing face. (b) Hydrophobicity profile of antimicrobial peptide LL-37. Residue numbers are indicated on the x-axis. Hydrophobicity index was calculated using the Kyte-Doolittle scale (window size = 5). Positive values indicate hydrophobic regions, negative values indicate hydrophilic regions. (c) IC₅₀s of LL-37 and its truncated peptide against P. falciparum 3D7. (d) IC₅₀s of LL-37 against sensitive and resistant strains of Plasmodium falciparum. Data in c and d are presented as means ± SEM from three biological replicates.

https://doi.org/10.1371/journal.ppat.1014062.g001

2. LL-37 exhibits stage-specific and time-dependent anti-malarial activity

We further evaluated the activity of LL-37 against different blood stages using highly synchronized P. falciparum parasites. Parasites were exposed to LL-37 at IC₅₀ for different treatment times (1, 3, 6, 9, 12h) at the ring (5 hours post invasion (hpi)), trophozoite (24 hpi), and schizont (36 hpi) stages (Fig 2a). Comprehensive stage-specific and time-dependent assays revealed that LL-37 possesses significant inhibitory activity against P. falciparum 3D7 parasites across all major intraerythrocytic developmental stages of rings (R), trophozoites (T), and schizonts (S). Quantitative analysis of parasitemia viability at 50 hpi, following LL-37 exposure, showed a clear sensitivity level that ring stages were the least sensitive (highest viability), followed by trophozoites, with schizonts exhibiting the highest sensitivity and lowest viability (Fig 2b). Furthermore, the antimalarial activity of LL-37 displayed a pronounced time-dependent relationship within each developmental stage (Fig 2b). This time-dependent effect indicates the importance of exposure duration for achieving maximal inhibitory activity against the parasites, irrespective of its developmental phase. Also, the same results were shown in further 4 days growth analysis (Fig 2c). Collectively, LL-37’s efficacy is maximized by extended exposure and targets schizonts most potently.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Stage-specific and time-dependent antiplasmodial activity of LL-37 in the blood stage.

(a) Schematic of the experimental design. Highly synchronized P. falciparum 3D7 parasites were treated with LL-37 at its IC₅₀ for 1, 3, 6, 9 and 12 h across three consecutive periods of the intraerythrocytic life cycle. Subsequently, the parasitemia of re-invasion was analyzed at 50 hpi, and another analysis was conducted after a 1:40 dilution and 4 days of culture. (b) Viability of parasites re-invasion at 50 hpi was measured. (c) Parasites were washed, followed by 1:40 dilution, and cultured for 4 days while parasitemia was measured. Data in b and c are presented as means ± SEM of 10,000 RBCs from three biological replicates.

https://doi.org/10.1371/journal.ppat.1014062.g002

3. LL-37 selectively targets infected erythrocytes by exploiting synergistic phosphatidylserine externalization and cholesterol depletion

Given the heightened susceptibility of schizonts, we hypothesized that LL-37 selectively targets pathological alterations in infected erythrocytes (iRBCs). LL-37 exhibited stage-selective hemolytic activity against P. falciparum-infected erythrocytes while demonstrating minimal toxicity to uninfected red blood cells (uRBCs) (Fig 3a). At its IC₅₀, LL-37 induced minimal hemolysis (<10%) in uRBCs, even after 60 minutes, a resistance maintained at 10 × IC₅₀ (Fig 3a). In contrast, schizont-enriched infected RBCs (iRBCs) showed significant lysis of 18.3% ± 2.0% at IC₅₀, escalating to 34.9% ± 1.5% at 10 × IC₅₀ (Fig 3a). Schizonts were 2.3-fold and 1.8-fold more sensitive to LL-37 than ring or trophozoite stages, respectively (Fig 3b).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. LL-37 selectively targets infected erythrocytes by exploiting synergistic phosphatidylserine externalization and cholesterol depletion.

(a) Comparative hemolysis of uninfected erythrocytes (uRBC) and schizont-infected erythrocytes (iRBC) exposed to LL-37 at its IC₅₀ and 10 × IC₅₀ concentrations. (b) Stage-specific hemolysis of P. falciparum-infected erythrocytes treated with LL-37 (IC₅₀). R: Ring stage; T: Trophozoite stage; S: Schizont stage. (c) Hemolysis of LL-37 under increasing A23187 concentrations with or without 1.5 mM MβCD. (d) Hemolysis of LL-37 under increasing concentrations of MβCD with or without 0.75 μM calcium ionophore A23187. (e) Parasite parasitemia under treatment with different proportions of MβCD-cholesterol complexes with or without LL-37 present. (f) Parasite parasitemia under treatment with different concentrations of DOTAP with or without LL-37 present. (g) Parasite parasitemia of LL-37 treatment with MβCD-cholesterol complexes and/or DOTAP. All data are presented as means ± SEM from three biological replicates. ns: not significant; *p < 0.05, *p < 0.01, ***p < 0.001 indicate statistically significant differences.

https://doi.org/10.1371/journal.ppat.1014062.g003

To dissect this selectivity, we reconstituted iRBC-like membrane pathology in normal RBCs. Individual PS externalization (A23187) or cholesterol depletion (MβCD) dose-dependently sensitized RBCs to LL-37. For instance, 1 μM A23187 or 7.5 mM MβCD combined with IC₅₀ of LL-37 induced hemolysis of 54.8% and >90%, respectively, compared to <10% by either modulator alone (Fig 3c, 3d). Crucially, combined subthreshold doses acted synergistically, adding 0.75 μM A23187 to 1.5 mM MβCD boosted LL-37-mediated hemolysis from 8.4% to 52.4%, a 43.9% increase (Fig 3d).

To definitively validate this mechanism, we investigated whether reversing these alterations could protect iRBCs. We first identified an effective rescue strategy that used an MβCD-cholesterol complex to supply cholesterol. Direct quantification of membrane cholesterol (S1a Fig) showed that treatment with 1:80, 1:40, and 1:20 MβCD-cholesterol complexes all significantly increased iRBC membrane cholesterol relative to the infected control group (*p < 0.05; ***p < 0.001), with levels rising in a dose-dependent manner (1:80 < 1:40 < 1:20). We then evaluated the protective effect of these cholesterol-replete iRBCs against LL-37. While all MβCD-cholesterol ratios increased membrane cholesterol, only the 1:40 dilution yielded optimal protection, approximately 20% restoration of iRBC survival as shown in Fig 3e. This discrepancy could potentially be explained by parasite tolerance to membrane cholesterol levels: the 1:20 dilution (highest cholesterol) likely disrupted the delicate lipid balance required for parasite development (inducing non-specific anti-parasitic toxicity, even without LL-37), while the 1:80 dilution (lowest cholesterol) failed to sufficiently reverse the LL-37-sensitive “cholesterol-depleted” membrane state. The 1:40 dilution, by contrast, restored cholesterol to a level that counteracted LL-37’s targeting (by reducing membrane vulnerability) without impairing parasite viability, thus representing the optimal balance between cholesterol repletion and parasite fitness.

In parallel, we evaluated DOTAP (a PS-binding cationic lipid) and confirmed its competition with LL-37 for PS binding. Flow cytometry analysis (S1b Fig) showed that both LL-37 and DOTAP reduced Annexin V positivity in PS-exposed iRBCs (Q2 quadrant), reflecting PS binding by these molecules (rather than reduced PS exposure). Quantification of Annexin V positivity (S1c Fig) further verified that both LL-37 and DOTAP treatment decreased PS accessibility to Annexin V. To directly assess competition, we measured LL-37 levels in supernatants post-co-incubation with DOTAP (S1d Fig). DOTAP concentration-dependently increased free LL-37 (e.g., 20 μM DOTAP raised supernatant LL-37 by ~30%), confirming reduced LL-37 binding to iRBCs. The protective effect was concentration-dependent but plateaued at higher concentrations, with 20 μM considered a saturating concentration that restored approximately 30% of the lost survival (Fig 3f). Most conclusively, the combination of the optimized 1:40 MβCD-cholesterol and saturating 20 μM DOTAP yielded a strong synergistic effect, restoring nearly 65% of the lost survival rate (Fig 3g).

Together, these results demonstrate that the membrane-specific pathology induced by Plasmodium infection, characterized by cholesterol depletion and phosphatidylserine externalization, is essential for the selective hemolytic activity of LL-37. Reversing these alterations effectively protects infected cells, providing direct functional evidence for the mechanism underlying LL-37’s anti-malarial action.

4. LL-37 expression is significantly elevated in malaria-infected hosts

Having established LL-37’s antiplasmodial mechanism, we asked whether hosts naturally mobilize this defense during infection. Quantitative ELISA confirmed significant upregulation of serum LL-37. Quantitative ELISA analysis demonstrated significant upregulation of serum LL-37 in Plasmodium-infected hosts, with human malaria patients (n = 63) exhibiting 2.2-fold higher levels than healthy controls (33.7 ± 6.2 ng/mL vs 15.1 ± 9.1 ng/mL; median 35.4 vs 15.4 ng/mL, p < 0.0001, Mann-Whitney U test), where 92% (58/63) of patients exceeded the 30 ng/mL threshold compared to only 13% (8/63) of controls (Fig 4a). This threshold was statistically derived as the mean LL-37 level of the healthy controls plus two standard deviations, representing the upper limit of the normal range. The serum LL-37 levels observed in our healthy controls (15.1 ± 9.1 ng/mL) are consistent with previously reported ranges for normal individuals (e.g., 5–20 ng/mL), while the elevated levels in malaria patients resemble those documented in other inflammatory conditions [38]. Meanwhile, P. berghei-infected mice (n = 12) exhibited a 6-fold elevation in serum CRAMP relative to uninfected controls (n = 12; 262.4 ± 126.6 ng/mL vs 43.6 ± 16.3 ng/mL; median 240.4 vs 39.2 ng/mL, p < 0.0001, Mann-Whitney U test), with all infected animals exceeding 100 ng/mL while controls remained below 73 ng/mL, confirming species-conserved induction of this antimicrobial peptide (Fig 4b). This conserved induction indicates LL-37’s physiological role in malaria defense.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. Elevated plasma levels of LL-37 and CRAMP in malaria-infected individuals and Plasmodium- infected mice.

(a) Plasma levels of LL-37 in Plasmodium falciparum (P.f) - infected patients and healthy donors. (b) Plasma levels of CRAMP in Plasmodium berghei (P.b) - infected mice and non-infected control mice. Each dot represents an individual sample. Horizontal lines indicate the median values. ***p < 0.001 indicate statistically significant differences.

https://doi.org/10.1371/journal.ppat.1014062.g004

5. Prophylactic, but not therapeutic, administration of exogenous LL-37/CRAMP reduces malaria pathogenesis and improves host survival

To translate in vitro and observational findings, we evaluated the efficacy of host defense peptides LL-37 (human) and its murine ortholog CRAMP using a P. berghei infection model. We first assessed their activity in a prophylactic 4-day suppressive test. Intravenous administration of LL-37 and CRAMP (1–16 mg/kg/day for 4 days) induced a significant, dose-dependent suppression of day 4 parasitemia (Fig 5a, 5b). Dynamic monitoring of parasitemia from day 0 to day 4 revealed that peptide treatment effectively delayed the exponential rise of parasite load compared to the untreated control group, with the 16 mg/kg dose exhibiting the most inhibitory effect across all time points (Fig 5b). Consistent with the kinetic data, a single-time-point analysis on day 4 also confirmed dose-dependent reduction of parasitemia in peptide-treated mice (Fig 5a, 5c), which aligned with the time-course observations and validated the antiplasmodial efficacy of LL-37 and CRAMP. LL-37 and CRAMP also significantly prolonged survival days compared to saline controls, the median survival day of saline controls was 11.5, while that of 1 mg/kg/day LL-37 and CRAMP were 14 and 13.5, 4 mg/kg/day LL-37 and CRAMP were 16 and 15, 16 mg/kg/day LL-37 and CRAMP were 18 and 17 (Fig 5d). Consistent with the extended survival, progression to ECM was delayed in both LL-37- and CRAMP-treated groups (S2 Fig). To address the potential hemolytic effects of LL-37/CRAMP, we measured plasma hemoglobin levels (A540) in treated and control mice. As shown in S3 Fig, no significant differences in hemolysis were observed between LL-37/CRAMP-treated groups (16 mg/kg/day) and saline-treated control groups in both P. berghei-infected and uninfected mice (all p > 0.05). This result indicates that the therapeutic doses of LL-37/CRAMP used in our study do not cause in vivo hemolysis, despite a minimal 3–5% in vitro lysis of uninfected RBCs (Fig 3a). The discrepancy between in vitro and in vivo results suggests that the physiological microenvironment in vivo may protect uninfected RBCs from peptide-induced lysis.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. Exogenous LL-37 and CRAMP treatment confers antiplasmodial and anti-pathological effects.

(a) Representative images of blood smears (day 4 post-infection) following treatments with different concentrations of LL-37 and CRAMP (n = 6 mice per group). (b) Dynamic changes in parasitemia from day 0 to day 4 post-P. berghei ANKA infection in untreated control mice and mice treated with LL-37/CRAMP (1, 4, 16 mg/kg/day, intravenous administration for 4 consecutive days). Parasitemia was determined by microscopic examination of Giemsa-stained thin blood smears at 24 h intervals. Data are presented as mean ± SEM (n = 6 mice per group). (c) Quantification of parasitemia (day 4, n = 6 mice/group) in mice treated with different concentrations of LL-37 and CRAMP. Data are presented as mean ± SEM (n = 6 mice per group). (d) Survival curves of mice administered different concentrations of LL-37 and CRAMP (day 4, n = 6 mice per group). Statistical analysis: Log-rank (Mantel-Cox) test, ns: not significant; *p < 0.05, **p < 0.01 vs. NS group. (e–h) Quantification of various inflammatory cytokines in mice treated with different concentrations of LL-37 and CRAMP (day 4, n = 6 mice per group). (i) Representative H&E-stained sections of liver, spleen, and brain tissues (day 4, n = 3 mice per group) from mice under distinct conditions: normal control, infected control, CQ positive control, and treatment with different concentrations of LL-37 or CRAMP. Black arrows point to malarial pigment; red arrows indicate the sequestration of RBCs; yellow arrows denote inflammatory cells; yellow boxes show fat vacuoles; red boxes indicate RePu or WhPu. Scale bar = 10 μm. ns: not significant; **p < 0.01, ***p < 0.001 indicate statistically significant differences. In all panels, NS (normal saline) served as a negative control, treated with normal saline (0.9% NaCl); NC (normal control) denotes the normal uninfected control group, serving as a baseline for normal physiological status and CQ (chloroquine) as a positive control.

https://doi.org/10.1371/journal.ppat.1014062.g005

Beyond parasite control, prophylactic treatment profoundly attenuated the malarial inflammatory response. Plasma levels of key pro-inflammatory cytokines IL-1β (Fig 5e), IL-6 (Fig 5f), TNF-α (Fig 5g), and IFN-γ (Fig 5h) measured on day 4 were significantly reduced compared to controls. Notably, the reduction in pro-inflammatory cytokines (IL-1β, IL-6, TNF-α, IFN-γ) observed in LL-37/CRAMP-treated mice (Fig 5e-5h) coincided with significantly lower parasitemia in these animals (Fig 5a,b). Therefore, it is likely that the decreased systemic cytokine levels are primarily attributable to the reduced parasite burden, which diminishes the antigenic stimulus for immune activation. Furthermore, histopathological analysis revealed a reduction of organ pathology (Fig 5i).

Among all treatment groups, the groups treated with 16 mg/kg LL-37 and CRAMP exhibited the greatest protective capacity for the tissues, including liver, spleen and brain, the therapeutic effect was equivalent to that of the chloroquine control group (Fig 5i). For liver, the hepatic lobule structure in the untreated group exhibited severe damage, the hepatic plates were not closely arranged. The liver contained a large number of fat vacuoles and hemozoin deposition. The groups from LL-37 and CRAMP-treated mice exhibited markedly reduced hemozoin deposition, and restored the normal structure of the hepatic lobules with a tighter arrangement between hepatocytes. While spleen sections showed a significantly widened red pulp (RePu) region and atrophy in the white pulp (WhPu) area in the untreated group, indicating severe anemia and more extramedullary hematopoiesis. The groups treated by LL-37 and CRAMP restored the WhPu area indicating improvement of anemia. Another indicator of the improvement of the spleen tissue in the treatment groups was a significant reduction in the deposition of malaria pigment (Fig 5i). In brain sections, the adhesion and accumulation of leukocytes in cerebral vessels are correlated with brain inflammation, which is a key feature of ECM. Concomitant with leukocyte aggregation, untreated mice exhibited evidence of vascular congestion characterized by increased RBC density within cerebral capillaries-a morphological manifestation of impaired microcirculation in experimental malaria. The treated groups effectively reduced the aggregation of inflammatory cells and alleviated this vascular congestion and associated increase in RBC density in brain capillaries (Fig 5i). The tissue-protective effects on liver, spleen and brain exhibited dose-dependent manner in both LL-37 and CRAMP groups.

Given the critical importance of exposure for efficacy, we evaluated the in vivo pharmacokinetics of LL-37/CRAMP. Plasma concentrations were measured 30 minutes after a single injection and 24 hours after the final dose of the 4-day regimen. At 30 minutes post-injection, peptide levels were significantly elevated in both uninfected and infected mice versus controls (p < 0.001), confirming successful delivery (S4a, S4b Fig). Notably, infected mice exhibited higher plasma CRAMP levels than uninfected (p < 0.01), suggesting infection may synergistically enhance CRAMP availability (S4a, S4b Fig). However, by 24 hours after the last dose, plasma CRAMP in infected mice was no longer distinguishable from baseline, and LL-37 was undetectable (S4c Fig). This indicates rapid in vivo degradation of the native peptides with minimal accumulation upon repeated dosing.

Based on the promising prophylactic results but considering their rapid clearance, we next assessed the peptides’ efficacy when treatment was initiated after infection establishment. A separate cohort received LL-37 or CRAMP (16 mg/kg/day for 4 days) only once parasitemia reached 8–15%. While this therapeutic regimen significantly reduced parasitemia and downregulated systemic pro-inflammatory cytokines (IL-1β, IL-6, TNF-α, IFN-γ) compared to saline controls, it failed to improve histopathological organ pathology in the liver, spleen, or brain, and did not extend host survival (S5a–S5h Fig). The lack of therapeutic benefit correlates with the pharmacokinetic data, suggesting that rapid peptide degradation prevents the accumulation of sufficient bioactive concentrations to reverse established tissue damage.

Collectively, these results demonstrate a critical distinction: prophylactic administration of LL-37/CRAMP effectively controls parasitemia, modulates inflammation, protects organs, and prolongs survival. However, their rapid in vivo clearance limits efficacy, and therapeutic intervention initiated after patent parasitemia, while reducing parasite burden and systemic inflammation, is insufficient to rescue established pathology or improve survival outcomes.

6. Endogenous Cathelicidin (CRAMP) is essential for parasite control

To determine the physiological contribution of endogenous cathelicidin in malaria infection, we employed Cramp knockout (KO) mice in our experimental model. The absence of functional CRAMP resulted in a significantly exacerbated disease course. On day 4 post-infection, Cramp

KO mice exhibited substantially higher parasitemia compared to wild-type (WT) controls (Fig 6a, 6b), indicating a severely impaired capacity to limit early parasite replication. This defect in controlling the rise of parasite burden directly translated to a worsened clinical outcome, as evidenced by a significant reduction in the survival time of KO mice (Fig 6c). These data establish that endogenous CRAMP plays a non-redundant and critical role in the host’s innate defense system against blood-stage P. berghei infection by limiting initial parasite expansion.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 6. Endogenous cathelicidin (Cramp) is essential for parasite control.

(a) Representative images of blood smears from C57BL/6 mice (day 4 post-infection), Cramp^⁻/⁻ mice. Scale bar = 5 μm. (b) Quantification of parasitemia (day 4, n = 6 mice/group) in C57BL/6 mice, Cramp^⁻/⁻ mice. (c) Survival curves (day 21, n = 6 mice/group) of C57BL/6 mice, Cramp^⁻/⁻ mice after infection. (d–g) Quantification of plasma levels of IL-1β (d), IL-6 (e), TNF-α (f), and IFN-γ (g) (day 4, n = 6 mice/group) in C57BL/6 mice, Cramp^⁻/⁻ mice. (h) Representative H&E-stained sections of liver, spleen, and brain tissues (day 4, n = 3 mice/group) from C57BL/6 mice, Cramp^⁻/⁻ mice. Black arrows point to malarial pigment; red arrows indicate the sequestration of RBCs; yellow arrows denote inflammatory cells; red boxes indicate RePu or WhPu. Scale bar = 10 μm. ns: not significant; **p < 0.01, ***p < 0.001 indicate statistically significant differences.

https://doi.org/10.1371/journal.ppat.1014062.g006

A notable and unexpected finding emerged from the analysis of the systemic inflammatory response. Despite the higher parasite burden, the absence of CRAMP did not lead to a corresponding increase in the plasma levels of key pro-inflammatory cytokines, including IL-1β, IL-6, TNF-α, and IFN-γ, which were measured at the same time point (Fig 6d-6g). The cytokine profiles of infected KO mice were not significantly different from those of their infected WT counterparts. This dissociation between parasite burden and systemic cytokine levels suggests that the primary role of endogenous CRAMP at this early stage of infection is direct parasite restriction rather than broad modulation of systemic inflammation. The absence of CRAMP leads to uncontrolled parasite replication, but this does not result in a proportionally amplified systemic cytokine response, indicating that the peptide’s essential function in host defense is mediated through its direct antiplasmodial activity. This suggests that the protective mechanism of endogenous CRAMP is independent of the modulation of these specific systemic inflammatory mediators during the early phase of infection. Consistent with the cytokine data, histopathological assessment at the defined endpoints also revealed no marked aggravation of tissue injury in Cramp KO mice relative to WT controls (Fig 6h). The degree of hemozoin deposition in the liver and inflammatory cell accumulation within cerebral vessels (a feature of vascular inflammation) was comparable between the two groups. This indicates that the increased mortality in KO mice is more directly attributable to their failure to restrain parasitemia rather than to a heightened immunopathological damage at the tissue level at the time points examined.

Collectively, these in vivo findings that impaired early parasite control without heightened systemic inflammation or tissue pathology converge with our in vitro evidence of direct lytic activity to demonstrate that endogenous CRAMP functions as a crucial first-line defender in malaria primarily by directly targeting and disrupting the infected red blood cell. This mechanism effectively delays the exponential rise in parasitemia and fatal outcome.