Antimicrobial cathelicidin peptide LL-37 inhibits the pyroptosis of macrophages and improves the survival of polybacterial septic mice

LL-37 affects macrophages and inhibits sepsis

Abstract

LL-37 is the only known member of the cathelicidin family of antimicrobial peptides in humans. In addition to its broad spectrum of antimicrobial activities, LL-37 can modulate various inflammatory reactions. We previously revealed that LL-37 suppresses the LPS/ATP-induced pyroptosis of macrophages in vitro by both neutralizing the action of LPS and inhibiting the response of P2X₇ (a nucleotide receptor) to ATP. Thus, in this study, we further evaluated the effect of LL-37 on pyroptosis in vivo using a cecal ligation and puncture (CLP) sepsis model. As a result, the intravenous administration of LL-37 improved the survival of the CLP septic mice. Interestingly, LL-37 inhibited the CLP-induced caspase-1 activation and pyroptosis of peritoneal macrophages. Moreover, LL-37 modulated the levels of inflammatory cytokines (IL-1β, IL-6 and TNF-α) in both peritoneal fluids and sera, and suppressed the activation of peritoneal macrophages (as evidenced by the increase in the intracellular levels of IL-1β, IL-6 and TNF-α). Finally, LL-37 reduced the bacterial burdens in both peritoneal fluids and blood samples. Together, these observations suggest that LL-37 improves the survival of CLP septic mice by possibly suppressing the pyroptosis of macrophages, and inflammatory cytokine production by activated macrophages and bacterial growth. Thus, the present findings imply that LL-37 can be a promising candidate for sepsis because of its many functions, such as the inhibition of pyroptosis, modulation of inflammatory cytokine production and antimicrobial activity.

Introduction

Sepsis is the systemic inflammatory response syndrome caused by infection (1). Despite two decades of research, it is still the most common cause of death in the non-coronary intensive care unit (2). A dysregulated inflammatory reaction is considered the major mechanism for the susceptibility to sepsis. However, disappointingly, anti-inflammatory cytokine therapies are inefficient in clinical trials (2–4). In recent years, much attention has been focused on the mechanism of host cell death, which develops during sepsis and contributes to the dysregulated inflammatory reaction, immunosuppression and organ failure in sepsis (5, 6).

Pyroptosis is a caspase-1-dependent cell death and mainly occurs in macrophages and dendritic cells, accompanied with the release of pro-inflammatory cytokines (such as IL-1β) (7, 8). Moreover, pyroptosis results in the cellular lysis and release of cytosolic contents, which induce the augmentation of inflammatory reactions (7, 8). Induction of pyroptosis requires two distinct stimuli, microbial-pathogen-associated molecular patterns (PAMPs) (such as lipoproteins or LPS) and endogenous-damage-associated molecular patterns (DAMPs) (such as ATP and uric acid) (9, 10). In response to these stimuli, a multi-protein complex of inflammasome [typically including caspases-1, Nod-like receptors (NLRs)and apoptosis-associated speck-like protein containing a CARD (ASC)] is formed, where pro-caspase-1 is converted to active caspase-1 (11). Thereafter, the activated caspases-1 cleaves pro-IL-1β to release IL-1β and induces cell death (pyroptosis) (9, 11). Of note, caspase-1 knockout (KO) causes a protective effect on several murine sepsis models and improves the survival of mice, where the IL-1β level was completely depressed, and the organs (such as kidney and spleen) were protected (12–16). Thus, caspase-1 activation and possibly pyroptosis play a major role in the pathogenesis (organ failure) and mortality of sepsis (7, 8).

Antimicrobial peptides (AMPs) represent the first line of defense against invading pathogens (17). The cathelicidin family of AMPs has been identified in various mammalian species, and LL-37 is the only known human cathelicidin, primarily produced by neutrophils and epithelial cells (17). In addition to its antimicrobial activity, LL-37 exhibits diverse biological activities, including regulation of inflammatory responses and wound healing (17, 18). Importantly, we previously found that LL-37 reduces the LPS-induced apoptosis of endothelial cells by neutralizing the action of LPS and inhibits the spontaneous apoptosis of neutrophils via the purinergic receptor X₇ (P2X₇) and formyl-peptide receptor-like 1 (FPRL1) (19, 20), suggesting that LL-37 is involved in the modulation of cell death. Furthermore, we have recently revealed that LL-37 suppresses the LPS/ATP-induced pyroptosis of macrophages by both neutralizing the action of LPS on CD14/TLR4 (toll-like receptor 4) and inhibiting the P2X₇ response to ATP in vitro (21), Thus, we hypothesize that LL-37 may inhibit the pyroptosis of macrophages in vivo, thereby exhibiting the protective action on septic model. In the present study, we intravenously administered LL-37 into a cecal ligation and puncture (CLP) septic model of mice, and evaluated the effect of LL-37 on pyroptosis by analyzing caspase-1 activation, cell death and IL-1β release. We revealed that LL-37 improves the survival of CLP septic mice by possibly suppressing the pyroptosis of macrophages, production of inflammatory cytokines (IL-6 and TNF-α, as well as IL-1β) and bacterial growth.

Methods

Reagents

A 37-mer peptide of hCAP18 (LL-37; L¹LGDFFRKSKEKIGKEF KRIVQRIKDFLRNLVPRTES³⁷) was synthesized by the solid-phase method on a peptide synthesizer (model PSSM-8; Shimadzu Scientific Instruments, Kyoto, Japan) by fluorenylmethoxycarbonyl chemistry, as described earlier (19). 7-Amino-actinomycin D (7AAD), a fluorescent chemical compound with a strong affinity for DNA, was purchased from Beckman Coulter (Brea, CA, USA). Fixable Viability Dye eFluor 660 (a viability dye that binds free amine groups on both surface and intracellular proteins and can be used to irreversibly label dead cells prior to fixation and/or permeabilization procedures), brefeldin A, anti-mouse TNF-α FITC antibody and anti-mouse IL-6 FITC antibody were purchased from eBioscience (San Diego, CA, USA). Anti-mouse IL-1β fluorescein antibody was purchased from R&D Systems (Minneapolis, MN, USA). Anti-F4/80 RPE antibody was purchased from AbD Serotec (Oxford, UK). Saline was purchased from Otsuka Pharmaceutical (Tokushima, Japan). Trypto-Soya agar (peptone 15.0g, soy peptone 5.0g, NaCl 5.0g, agar 15.0g, pH = 7.3) was purchased from Nissui Pharmaceutical (Tokyo, Japan), dissolved in 1 liter of distilled water and autoclaved.

Mice

Male BALB/c mice (7–10 weeks old; Sankyo Labo Service, Tokyo, Japan) were used in the experiments. Mice were bred under specific-pathogen-free conditions and housed in temperature-controlled, air-conditioned facilities with 12/12-h light/dark cycles and food and water ad libitum. All experiments were approved by the Ethics Committee for the Use of Laboratory Animals of Juntendo University, Graduate School of Medicine (Permit Number: 260213).

CLP procedure

The CLP procedure was performed according to the general guidelines (22). Mice were anesthetized using 2% isoflurane (Wako Pure Chemical Industries, Osaka, Japan) in oxygen. After disinfection of the abdomen, a 1-cm midline laparotomy was performed and the cecum was exposed. Cecal contents were massaged to the tip, and the distal 0.5cm of the tip was ligated with 3-0 silk suture. An 18-G needle was used to perforate the ligated portion of the cecum once in a through-and-through manner, and a small amount of stool (1mm in length) was extruded. The cecum was returned to the abdominal cavity, and the abdomen and skin were respectively closed using 3-0 nylon suture. Following the surgery, 1ml of saline was subcutaneously administered in the neck.

In the Sham group, mice underwent the same procedure but without the cecal ligation and the puncture. In the LL-37 group, 2 μg per mouse of LL-37 were administered intravenously, immediately after the surgery. The survival rates of the mice were monitored every day for 7 days. In some experiments, 1 μg per mouse of LL-37 was administered intravenously, immediately after the surgery.

Preparation of the peritoneal fluids and blood samples

At the indicated period (5 or 15h) after the surgery, 3ml of cold PBS were intraperitoneally injected, and the peritoneal fluids were collected after gentle massage for 1min. In addition, blood was obtained by cardiac puncture, and sera were prepared by centrifugation of blood at 3000rpm for 20min.

Detection of caspase-1 activation in tissue sections

Organs (livers, kidneys and spleens) were collected at 5h after the surgery, embedded in Tissue-Tek OCT compound (Sakura, Tokyo, Japan) and frozen in liquid nitrogen. Sections (5 μm) were cut in a Leica CM305S cryostat (Leica Microsystems, Wetzlar, Germany), thawed onto silane-coated slides (Muto Pure Chemicals, Tokyo, Japan) and air dried. Thereafter, the activation of caspase-1 was evaluated using these sections. Briefly, the sections were fixed with acetone, and then incubated in 3 × FLICA-YVAD-FMK (FLICA™ Caspase-1 Assay Kit, Immunochemistry Technologies, Bloomington, MN, USA) for 1h at 37°C, anti-F4/80 RPE antibody (5 μg ml^–1) for 20min at 4°C and Hoechst33342 (0.8 μg ml^–1) for 5min at 4°C in the dark, sequentially. Then, the sections were mounted by coverslips (Deckglaser, Freiburg, Germany) using an aqueous medium Vectashield (Vector Labs, Burlingame, CA, USA) and were photographed with a fluorescence microscope system Axioplan 2 (Carl Zeiss, Jena, Germany). To quantify the percentage of macrophages with activated caspase-1, 3–5 high power fields (×400) derived from each animal were examined and the numbers of F4/80-positive cells, and FLICA- and F4/80-double-positive cells were counted.

The percentage of macrophages with activated caspase-1 = (number of FLICA- and F4/80-double-positive cells)/(number of F4/80 positive cells) × 100%.

Detection of caspase-1 activation and pyroptosis among peritoneal macrophages

Peritoneal cells were collected from peritoneal fluid at 5h after the surgery by centrifugation at 400 × g for 5min. Thereafter, peritoneal cells were evaluated used for pyroptosis assay. Briefly, the cells (about 2×10⁶ cells ml^–1) were incubated in 1 × FLICA-YVAD-FMK for 1h at 37°C, anti-F4/80 RPE antibody (5 μg ml^–1) for 20min at 4°C,and 7AAD (20 μl per test) for 5min at 4°C in the dark, sequentially. Then, the cells were assayed for cell death and caspase-1 activation by flowcytometry (FACSCalibur, BD Biosciences, Rutherford, NJ, USA).

Moreover, peritoneal cells were washed and incubated in 7AAD (20 μl per test) for 5min at 4°C in the dark. Then, viable cells were gated using 7AAD, and the percentage of granulocytes (neutrophil) and macrophages was determined by forward scatter/side scatter (FSC/SSC). The numbers of granulocytes (neutrophil) and macrophages were calculated based on total cell counts and percentages.

Detection of the activated form of caspase-1 (p10 subunit) in the peritoneal fluids

Peritoneal fluids (40 µl) recovered 5h after CLP were subjected to sodium dodecyl sulfate–polyacrylamide gel electrophoresis on a 15% polyacrylamide gel under reducing condition, and proteins were transferred to polyvinylidene fluoride membranes (Immobilon-P; Millipore, Billerica, MA, USA). The membranes were probed with rabbit anti-mouse caspase-1 p10 antibody (Santa Cruz Biotechnology, Santa Cruz, CA, USA), and HRP-conjugated goat anti-rabbit IgG (Millipore). The signals were detected with SuperSignal West Dura Chemiluminescent Substrate (Pierce Biotechnology, Rockford, IL, USA) and quantified using a LAS-3000 luminescent image analyzer (Fujifilm, Tokyo, Japan) and MultiGauge software (Fujifilm).

Quantification of cytokine levels in peritoneal fluids and sera

Peritoneal fluids and sera were assayed for cytokines. Cytokines (IL-1β, IL-6 and TNF-α) were determined using commercially available mouse cytokine detection ELISA kits (eBioscience), according to the manufacturer’s instructions. Moreover, MCP-1 was determined in the peritoneal fluids using a commercially available Cytometric Bead Array Mouse Inflammation Kit (BD), according to the manufacturer’s instructions.

Quantification of intracellular cytokine levels

Peritoneal cells were washed and stained with 1 μl ml^–1 Fixable Viability Dye eFluor 660 (containing 3 μg ml^–1 brefeldin A) for 30min at 4°C in the dark. Then, the cells were fixed with 4% paraformaldehyde for 30min at room temperature (RT) in the dark and permeated with 0.2 % Triton X-100/PBS. Subsequently, the cells were incubated in anti-IL-1β fluorescein antibody (2.5 μg ml^–1), anti-IL-6 FITC antibody (2.5 μg ml^–1) or anti-TNF-α FITC antibody (0.6 μg ml^–1) for 30min at RT in the dark. Thereafter, macrophages were gated by FSC/SSC, and the intracellular levels of IL-1β, IL-6 and TNF-α were measured by flowcytometry (FACSCalibur).

Quantification of bacterial burden

The Peritoneal fluids and blood samples were serially diluted in PBS. Diluted samples (100 μl) were plated on Trypto-Soya agar plates, and the plates were incubated for 20h at 37°C. The colony forming units (CFU) were counted and corrected for the dilution factor.

Statistical analyses

Data are shown as the mean ± SEM. Statistical testing was performed by one-way ANOVA, followed by Bonferroni’s multiple comparison test (GraphPad Prism; GraphPad Software, San Diego, CA, USA). Survival data were analyzed using the Kaplan–Meier method and survival curves were compared using the log-rank test and Gehan–Breslow–Wilcoxon test in univariate analysis (GraphPad Prism). A P value of < 0.05 was considered to be significant.

Results

LL-37 improves the survival of CLP septic mice

First, we evaluated the effect of LL-37 administration on the survival of CLP-operated mice. Mice were divided into Sham, CLP and LL-37 groups, and the survival rates were monitored for 7 days. As shown in Fig. 1, the survival rate in the CLP group (6.7%) was significantly lower than that in the Sham group (100%, P < 0.001). Interestingly, the administration of LL-37 significantly improved the survival rate to 36.4% (P < 0.05).

Fig. 1.

Effect of LL-37 on the survival of CLP septic mice. Mice were divided into the Sham (□), CLP (●) and LL-37 (○) groups. In the LL-37 group, mice were intravenously administered with 2 μg per mouse LL-37, and the survival rates of the mice were monitored for 7 days. Survival data were analyzed using the Kaplan–Meier method and survival curves were compared using the log-rank test and Gehan–Breslow–Wilcoxon test in univariate analysis. n = 8–15 per group *P < 0.05.

LL-37 inhibits the pyroptosis of peritoneal macrophages

Next, to evaluate the effect of LL-37 on pyroptosis of macrophages, peritoneal cells were harvested at 5h after the operation, and peritoneal macrophages (gated as F4/80 positive) were evaluated for pyroptosis by detecting caspase-1 activation (FLICA positive) and cell death (7AAD positive). As shown in Fig. 2, caspase-1 activation (FLICA positive) was significantly higher in the CLP group (13.3 %) than that in the Sham group (5.1%, P < 0.05). Similarly, pyroptosis (caspase-1-activated dead/dying cells; FLICA/7AAD-double positive) was obviously induced in the CLP group (CLP group, 6.8%, versus Sham group, 1.9%, P < 0.01). Of note, the administration of LL-37 significantly suppressed the CLP-induced caspase-1 activation and pyroptosis of peritoneal macrophages to 4.4% (P < 0.05) and 2.9% (P < 0.01), respectively (Fig. 2B and C). Moreover, the activated form of caspase-1 (p10 subunit) was increased in the peritoneal fluids of the CLP group, and the increase was suppressed in the LL-37 group (Supplementary Figure 1, available at International Immunology Online).

Fig. 2.

Effect of LL-37 on the pyroptosis of peritoneal macrophages. Peritoneal cells were collected from mice of Sham, CLP and LL-37 groups at 5h after the surgery. Thereafter, peritoneal cells were evaluated for pyroptosis by detecting caspase-1 activation (FLICA positive) and cell death (7AAD positive) of the peritoneal macrophages (F4/80 positive). In panel A, upper halves, right halves and upper-right quadrants show cell death, caspase-1 activation and pyroptosis (FLICA/7AAD-double positive), respectively, among peritoneal macrophages. Images are representative of 6–7 separate experiments. Panels B and C show the percentage of caspase-1 activation and pyroptosis, respectively. Data are the mean ± SEM of 6–7 separate experiments, and values are compared between the CLP and LL-37 groups. *P < 0.05, **P < 0.01.

In addition, the numbers of granulocytes (neutrophils) and macrophages were determined in the peritoneal cavity of CLP mice. The numbers of granulocytes (neutrophils) and macrophages markedly increased in the CLP group (Sham group versus CLP group, P < 0.001) and significantly suppressed in the LL-37 group (P < 0.05), suggesting that LL-37 inhibits the infiltration of these cells into the peritoneal cavity after CLP (Supplementary Figure 2A and B, available at International Immunology Online).

We further evaluated the effect of LL-37 on the CLP-induced pyroptosis of macrophages in the organs. Thus, we tried to evaluate the caspase-1 activation in macrophages of the spleen. As shown in Supplementary Figure 3A and B (available at International Immunology Online), caspase-1 was activated in macrophages (F4/80 positive) in the white pulp of spleen in the CLP group. Moreover, the percentage of caspase-1-activated macrophages was significantly increased in the CLP group (Supplementary Figure 3C, available at International Immunology Online); however, the activation was suppressed in the LL-37 group. In contrast, we could not detect caspase-1 activation in macrophages of the liver and or kidney (data not shown).

LL-37 suppresses the increase of IL-1β in the peritoneal fluids and sera

Furthermore, we measured the IL-1β levels in the peritoneal fluids and sera at 15h after the operation. Comparing with the levels of IL-1β in peritoneal fluids and sera in the Sham group (48 pg ml^–1 and 19 pg ml^–1, respectively), those levels significantly increased in the CLP group (1269 pg ml^–1, P < 0.001 and 230 pg ml^–1, P < 0.01, respectively; Fig. 3A and B). Interestingly, LL-37 administration significantly suppressed the IL-1β levels both in peritoneal fluids (768 pg ml^–1, P < 0.05) and sera (49 pg ml^–1, P < 0.001; Fig. 3A and B).

Fig. 3.

Effect of LL-37 on the levels of IL-1β in peritoneal fluids and sera. Peritoneal fluids and sera were collected from mice of Sham, CLP and LL-37 groups at 15h after the surgery, and assayed for IL-1β. Panels A and B show the IL-1β levels in the peritoneal fluids and sera, respectively. Data are the mean ± SEM of 5–9 separate experiments, and values are compared between the CLP and LL-37 groups. *P < 0.05, **P < 0.01.

LL-37 regulates the levels of IL-6 and TNF-α in the peritoneal fluids and sera

In addition, we measured the levels of other inflammatory cytokines (IL-6 and TNF-α) in the peritoneal fluids and sera at 15h after the operation. Comparing with the level of IL-6 in the peritoneal fluids and sera in the Sham group (0.6ng ml^–1 and 0.7ng ml^–1, respectively), those levels significantly increased in the CLP group (34ng ml^–1, P < 0.001, and 92ng ml^–1, P < 0.001, respectively; Fig 4A and B). Interestingly, LL-37 inhibited the IL-6 levels in both the peritoneal fluids (23ng ml^–1, P < 0.05) and sera (77ng ml^–1, P < 0.001). In addition, comparing with the level of TNF-α in the peritoneal fluids and sera in the Sham group (49 pg ml^–1 and 46 pg ml^–1, respectively), those levels significantly increased in the CLP group (376 pg ml^–1, P < 0.05 and 521 pg ml^–1, P < 0.001, respectively). Importantly, LL-37 significantly suppressed the TNF-α level (308 pg ml^–1, P < 0.05) in the sera, but not in the peritoneal fluids (444 pg ml^–1).

Fig. 4.

Effect of LL-37 on the levels of IL-6 and TNF-α in peritoneal fluids and sera. Peritoneal fluids and sera were collected from mice of Sham, CLP and LL-37 groups at 15h after the surgery, and assayed for IL-6 and TNF-α. Panels A and B show the IL-6 levels in the peritoneal fluids and sera, respectively, and panels C and D show the TNF-α levels in the peritoneal fluids and sera, respectively. Data are the mean ± SEM of 5–9 separate experiments, and values are compared between the CLP and LL-37 groups. *P < 0.05, ***P < 0.001.

In addition, we tried to detect a chemokine (MCP-1) in the peritoneal fluids. The level of MCP-1 in the peritoneal fluids was significantly increased in the CLP group (Sham group versus CLP group, P < 0.001; Supplementary Figure 2C, available at International Immunology Online). Similar to the actions on the cytokine production, LL-37 significantly reduced the level of MCP-1 in the CLP group (P < 0.05).

LL-37 inhibits the intracellular levels of cytokines (IL-1β, IL-6 and TNF-α) in peritoneal macrophages

To further evaluate the effect of LL-37 on the activation of macrophages in the local milieu of inflammation, we measured the intracellular levels of cytokines in peritoneal macrophages at 15h after the operation. As shown in Fig. 5, the intracellular levels of IL-1β, IL-6 and TNF-α in the macrophages significantly increased in the CLP group, comparing with those in the Sham group (P < 0.05; Fig. 5). Interestingly, LL-37 significantly suppressed the intracellular levels of IL-1β, TNF-α and IL-6 (P < 0.05).

Fig. 5.

Effect of LL-37 on the intracellular levels of cytokines in peritoneal macrophages. Peritoneal cells were collected from mice of Sham, CLP and LL-37 groups at 15h after the surgery, and assayed for the intracellular cytokine. Macrophages were gated by FSC/SSC. Panels A–C show the levels of intracellular cytokines (IL-1β, IL-6 and TNF-α), respectively, in viable macrophages (Fixable Viability Dye negative). Images are representative of 3–6 separate experiments. Panels D–F show the percentage of cells expressing IL-1β, IL-6 and TNF-α, respectively. Data are the mean ± SEM of 3–6 separate experiments, and values are compared between the CLP and LL-37 groups. *P < 0.05, ***P < 0.001.

LL-37 inhibits the bacterial burdens in the peritoneal fluids and blood samples

Finally, we evaluated the effect of LL-37 on the bacterial burdens at 15h after the operation. Comparing with the CFU of the peritoneal fluids and blood samples in the Sham group (144 per ml and 0 per ml, respectively; Fig. 6), those significantly increased in the CLP group (2.5×10⁹ per ml, P < 0.001 and 4.2×10⁵ per ml, P < 0.001, respectively). Interestingly, LL-37 significantly reduced the bacterial burdens in both the peritoneal fluids (0.7×10⁹ ml^–1, P < 0.001) and blood samples (1.8×10⁵ ml^–1, P < 0.05).

Fig. 6.

Effect of LL-37 on the bacterial burdens in the peritoneal fluids and blood samples. Peritoneal fluids and blood samples were collected from mice of Sham, CLP and LL-37 groups at 15h after the surgery, and serially diluted in PBS. Then, diluted samples were plated on Trypto-Soya agar plates, and the plates were incubated for 20h at 37°C. CFU were counted and corrected for the dilution factor. Panels A and B show the CFU in the peritoneal fluids and blood samples, respectively. Data are the mean ± SEM of 5–7 separate experiments, and values are compared between the CLP and LL-37 groups. *P < 0.05, ***P < 0.001.

Discussion

LL-37 is the only known member of the cathelicidin family of AMPs in humans, cleaved from a human cationic antimicrobial polypeptide of 18-kDa (hCAP18) (17, 18). In addition to its broad spectrum of antimicrobial activities, LL-37 can modulate various inflammatory reactions (17, 18). We previously reported that LL-37 suppresses TNF-α production by macrophages and improves the survival of LPS-induced endotoxin mice via the neutralization of LPS (23, 24). Moreover, we also found that LL-37 represses the LPS-induced apoptosis of endothelial cells by neutralizing LPS and inhibits the spontaneous apoptosis of neutrophils via P2X₇ (a nucleotide receptor) and FPRL1 (a formyl peptide receptor) (19, 20), suggesting that LL-37 may have a potential to modulate the host cell death. On the basis of this hypothesis, we have recently revealed that LL-37 suppresses the LPS/ATP-induced pyroptosis of macrophages in vitro by both neutralizing the action of LPS and inhibiting the response of P2X₇ to ATP (21). Thus, in this study, we further evaluated the effect of LL-37 on pyroptosis in vivo using a CLP sepsis model. As a result, the intravenous administration of LL-37 improved the survival of the CLP septic mice (Fig. 1). Interestingly, LL-37 inhibited the CLP-induced caspase-1 activation of peritoneal and splenic macrophages, and pyroptosis of peritoneal macrophages (Fig. 2). Moreover, LL-37 modulated the levels of inflammatory cytokines (IL-1β, IL-6 and TNF-α) in both the peritoneal fluids and sera (Figs 3 and 4) and suppressed the activation of peritoneal macrophages (as evaluated by the increase in the intracellular levels of IL-1β, IL-6 and TNF-α; Fig. 5). Finally, LL-37 reduced the bacterial burdens in both the peritoneal fluids and blood samples (Fig. 6). Thus, our present study suggests that LL-37 improves the survival of CLP septic mice by possibly suppressing the pyroptosis of macrophages and inflammatory cytokine production by activated macrophages, as well as bacterial growth.

Pyroptosis is a caspase-1-dependent cell death and mainly occurs in macrophages and dendritic cells (7). Caspase-1 is typically activated by the combination with PAMPs (such as LPS, flagellin) and DAMPs (such as ATP, uric acid) via the formation of inflammasomes (9, 25). Thereafter, activated caspase-1 induces the release of pro-inflammatory cytokines (e.g. IL-1β) and cell death (pyroptosis) (9, 26). Moreover, pyroptosis results in the cellular lysis by forming pores in the cell membrane, and the extracellular release of cytosolic contents, such as DAMPs (ATP, DNA and histones) (7). Thereafter, DAMPs elicit the inflammatory responses and tissue damage; for example, histones (H3 and H4) can induce apoptosis of endothelial cells (27). In addition, during pyroptosis, caspase-1-containing inflammasomes are extracellularly released to activate the pro-caspase-1 in the extracellular milieu and internalized (endocytosed) by the surrounding macrophages to amplify the caspase-1 activation and pyroptosis (28). In the present study, we demonstrated that the LL-37 administration inhibited the CLP-induced caspase-1 activation and pyroptosis of peritoneal macrophages (Fig. 2), and caspase-1 activation in splenic macrophages (Supplementary Figure 2A and B, available at International Immunology Online). Moreover, LL-37 suppressed the processing of caspase-1, as evidenced by the reduction of an active p10 subunit in peritoneal fluids (Supplementary Figure 1, available at International Immunology Online). Thus, these results likely suggest that LL-37 inhibits the caspase-1 activation and pyroptosis of macrophages in the CLP mice. Consistent with this, LL-37 lowered the IL-1β levels in both the peritoneal fluids and sera (Fig. 3). Importantly, it is reported that the inactivation of caspase-1 (such as, caspase-1 KO or administration of ac-YVAD-cmk, a caspase-1 inhibitor) improves the survival of Escherichia coli-induced or LPS-induced septic mice, where the plasma IL-1β level was completely depressed (12–15). Moreover, KO of an inflammasome component (NLRP3 or ASC) significantly improves the survival of CLP septic mice (29). Furthermore, we confirmed that LL-37 dose dependently improves the survival rate of CLP septic mice (Fig. 1 and Supplementary Figure 4, available at International Immunology Online) and inhibits the pyroptosis activation of peritoneal macrophages (Supplementary Figure 5A–C, available at International Immunology Online) and IL-1β levels in the peritoneal fluids (Supplementary Figure 5D, available at International Immunology Online), indicating a relationship between the LL-37-mediated inhibition of pyroptosis and improvement of survival in our CLP model. Therefore, the present study indicates that LL-37 improves the survival of the CLP mice possibly by inhibiting caspase-1 activation, IL-1β production and pyroptosis that amplifies the inflammatory reactions.

However, it is not clear how LL-37 inhibits caspase-1 activation and pyroptosis in the CLP model. Of note, LL-37 has an antimicrobial activity (Fig. 6) and thus may rescue the CLP septic mice by decreasing the number of circulating bacteria; that is, killing of circulating bacteria by LL-37 may result in the decreased number of macrophages undergoing pyroptosis.

Furthermore, LPS, a major component of bacteria, potently induces the inflammatory responses via CD14/TLR4 (30) and is increased in the peritoneal fluids and sera of sepsis models (31, 32). Moreover, ATP is extracellularly released from dead/dying cells; for example, the plasma ATP level increased to 10 μM in the CLP mice (2 μM in the Sham mice) (33), and the peritoneal ATP level increases to 1 μM in the E. coli-induced septic mice (0.1 μM in the Sham mice) (34). Importantly, we previously reported that the combination of LPS and ATP induced the pyroptosis of macrophages, which was inhibited by LL-37 via both the neutralization of LPS action and inhibition of the P2X₇ response to ATP (21). Thus, we speculate that LPS and ATP are major inducers of caspase-1 activation and pyroptosis in the CLP septic mice, as reported in various sepsis models (35, 36) and that LL-37 suppresses the actions of LPS and ATP in vivo and in vitro, thereby inhibiting pyroptosis (caspase-1 activation) and increasing the survival of CLP septic mice.

The levels of serum IL-6 and TNF-α correlate with the severity of sepsis; thus, these cytokines are recognized as important markers for sepsis (4). Importantly, LL-37 significantly inhibited the increased levels of IL-6 and TNF-α in sera of CLP septic mice (Fig. 4). Of note, LL-37 significantly lowered the increased level of IL-6 but not that of TNF-α in peritoneal fluids. Interestingly, intraperitoneal TNF-α is important for promoting peritoneal adhesion and preventing the spread of bacteria (37, 38). Moreover, the intraperitoneal administration of TNF-α antibody increases the mortality of the CLP mice (37). Thus, the maintained level of peritoneal TNF-α after the administration of LL-37 is assumed to be important for survival for the CLP mice. Moreover, it is reported that LL-37 inhibits LPS-induced macrophage activation (cytokine production) (17, 23, 39). Consistent with these observations, the present study confirms that LL-37 suppresses the up-regulated intracellular cytokine levels in peritoneal macrophages. Thus, it is reasonable to speculate that LL-37 inhibits LPS-induced activation of macrophages in vivo at the local milieu of inflammation. Furthermore, LL-37 has an antimicrobial activity both in vitro [minimum inhibitory concentration (MIC), 0.1~10 μg ml^–1] and in vivo (17), and the concentration of plasma LL-37 is estimated to be about 1 μg ml^–1 (LL-37, 2 μg per mouse; body weight, about 22g) in this study. Thus, LL-37 is expected to suppress the bacterial growth in vivo in the CLP model (Fig. 6).

Finally, LL-37 was originally identified as an AMP, which protects the host from invasive microbial infections because of its antimicrobial activity, but is now regarded as a multifunctional molecule, which participates in the innate immune system by exerting diverse biological activities, including regulation of inflammatory responses and modulation of immune cell death (17, 18). The present study has demonstrated for the first time an in vivo function of LL-37 to suppress the caspase-1 activation and pyroptosis of macrophage, and confirmed the modulatory actions of LL-37 on inflammation and antimicrobial activities using a polybacterial CLP sepsis model. As a therapeutic strategy, the blocking of LPS action (e.g. TLR4 KO) protects mice from systemic inflammation and tissue damage in endotoxemia (40), and the blocking of ATP action (e.g. administration of a P2X₇ antagonist) improve the survival rate in E. coli-induced sepsis (36, 41). We previously revealed that LL-37 suppresses LPS/ATP-induced pyroptosis by neutralizing LPS and inhibiting the P2X₇ response to ATP. Thus, the present findings imply that LL-37 can be a promising candidate for sepsis because it has many functions, including the inhibition of pyroptosis, modulation of inflammatory cytokine production and antimicrobial activity.

Supplementary data

Supplementary data are available at International Immunology Online.

Funding

This work was supported in part by a Grant-in-Aid (Grant number 26460538; http://www.jsps.go.jp/j-grantsinaid/inde​x.html) for Scientific Research from Japan Society for the Promotion of Science, and a Grants-in-Aid (Grant number S1201013; http://www.mext.go.jp/a_menu/koutou/shin​kou/07021403/002/002/1218299.htm) from the Ministry of Education, Culture, Sports, Science and Technology, Japan. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Conflict of interest statement: The authors declared no conflict of interests.