The endogenous anti-microbial peptide LL-37/hCAP-18 is an effector molecule of the innate host defense system at surfaces of the body. Besides its direct anti-microbial activity, the peptide interacts with different cell types. Dendritic cells (DCs) play a central role in mucosal host defense. It was the aim of the study to determine whether LL-37 modulates the response of DCs to pathogen-associated molecular patterns. Monocyte-derived DCs were stimulated with the Toll-like receptors (TLRs) ligands LPS, lipoteichoic acid and flagellin. We measured classical markers of DC maturation and assayed the ability of the DCs to activate T cell responses. Co-incubation with LL-37 resulted in suppressed activation of DCs. Levels of released IL-6, IL-12p70 and TNF-α and surface expression of HLA-DR, CD80, CD83, CD86 and the chemokine receptor CCR7 were decreased. Exposure of DCs to LL-37 during LPS exposure induced co-cultured naive T cells to produce less IL-2 and IFN-γ and decreased their proliferation. The response of memory T cells to a recall antigen was also decreased. In conclusion, we demonstrate that the anti-microbial peptide LL-37 inhibits the activation of DCs by TLR ligands. We propose that LL-37 is a regulator of host defense responses at the intersection of innate and adaptive immune systems.
Dendritic cells (DCs) are the most potent antigen-presenting cells and play a central role in the mammalian host defense and immune system (1). They are located at the intersection of the innate and adaptive parts of the immune system, integrate information about the initial innate host defense reaction and orchestrate the adaptive immunity (2). Immature dendritic cells (iDCs) are derived from circulating hematopoietic precursors and pre-DC populations under the instruction of specific cytokines and other factors (3). Following activation by contact with micro-organisms, iDCs are transformed into antigen-processing and -presenting mature dendritic cells (mDCs). mDCs migrate to secondary lymphoid organs and interact with naive T lymphocytes. The turnover of mucosal DCs during physiological conditions or during microbial challenge is a dynamic process. The local environment at mucosal surfaces has a significant impact on this process. Factors that are released from epithelial cells influence DC biology. As one example, human epithelial cells trigger DC-mediated allergic inflammation by producing human thymic stromal lymphopoietin (TSLP) (4).
Anti-microbial peptides (AMPs) are effector molecules of the innate immune system and are secreted by epithelial and other cell types (5, 6). AMPs have a broad anti-microbial spectrum and inactivate micro-organisms by direct interaction with biomembranes or other organelles. Besides their direct anti-microbial function, it has been suggested that they play multiple roles as mediators of inflammation with impact on epithelial and inflammatory cells influencing diverse processes such as cytokine release, cell proliferation, angiogenesis, wound healing, chemotaxis, immune induction, and protease–antiprotease balance (7, 8). Peptide antibiotics of the cathelicidin family contain a highly conserved signal sequence and pro-region (termed ‘cathelin’ = cathepsin L inhibitor) but show substantial heterogeneity in the C-terminal domain that encodes the mature peptide (9). The only human cathelicidin, LL-37/hCAP-18, was first isolated from human bone marrow (10, 11). The peptide is expressed in myeloid and epithelial cells (10, 12–15). The peptide has broad anti-microbial activity and in addition induces angiogenesis (16), attracts neutrophils, monocytes and CD4+ T cells and activates mast cells (17). In human primary monocytes, LL-37 activates the extracellular signal-regulated kinase and p38 kinase via an unknown G protein-coupled receptor independent mechanism (18). iDCs generated from blood monocytes in the presence of LL-37 have significantly up-regulated endocytic capacity, modified expression of phagocytic receptors, enhanced co-stimulatory molecule expression and increased secretion of Th1-inducing cytokines (19). In addition to these activities, LL-37 is known to bind to LPS and to inhibit its effects on monocytes (20). In contrast, no data are available that test a potential influence of LL-37 on the activation of DCs by LPS and other ligands of Toll-like receptors (TLRs).
Here we aimed to test the hypothesis that LL-37 inhibits the maturation and activation of monocyte-derived DCs by LPS and other pathogen-associated molecular patterns. We measured classical markers of DC maturation and assayed the ability of the DCs to induce T cell responses.
Media and reagents
iDCs were cultured and stimulated in RPMI 1640 (Invitrogen, Karlsruhe, Germany) containing 10% fetal bovine serum (FBS) (Invitrogen), 50 ng ml−1 granulocyte macrophage colony-stimulating factor (GM-CSF) (Strathmann Biotec, Hamburg, Germany) and 20 ng ml−1 IL-4 (Strathmann Biotec). LL-37 (LLGDFFRKSKEKIGKEFKRIVQRIKDFLRNLVPRTES-COOH) and a scrambled form of LL-37, sLL-37 (RSLEGTDRFPFVRLKNSRKLEFKDIKGIKREQFVKIL-COOH), were chemically synthesized (Charité, Humboldt-Universität, Berlin, Germany). LPS from Salmonella abortus equi and lipoteichoic acid (LTA) were purchased from Sigma–Aldrich (Munich, Germany). Flagellin was purchased from InvivoGen (San Diego, CA, USA) and IL-1ß, IL-6 and tumor necrosis factor-α (TNF-α) from R&D Systems (Wiesbaden-Nordenstadt, Germany).
Cell purification and culture
DCs were prepared from PBMCs by standard procedures (18). Briefly, PBMCs were isolated from a buffy coat made from 500 ml blood of a healthy volunteer donor by magnetic cell sorting with anti-CD14 MicroBeads according to the manufacturer's protocol (Miltenyi Biotec, Bergisch Gladbach, Germany). PBMCs were seeded in petri dishes and cultured in RPMI 1640 (Invitrogen) containing 10% FBS (Invitrogen), 50 ng ml−1 GM-CSF (Strathmann Biotec) and 20 ng ml−1 IL-4 (Strathmann Biotec) to generate iDC. On day 3, fresh medium containing GM-CSF and IL-4 was added to the cells. After 7 days of culture, cells were harvested, washed once with PBS and used for the experiments.
Induction of DC maturation
iDCs were seeded in six-well (5 × 105 cells per well) or 24-well (1 × 105 cells per well) plates in complete RPMI containing GM-CSF and IL-4. DC maturation was induced by addition of LPS (10 or 100 ng ml−1), LTA (10 μg ml−1), flagellin (100 ng ml−1) or a cytokine cocktail consisting of IL-1ß (10 ng ml−1), IL-6 (150 ng ml−1) and TNF-α (10 ng ml−1). To test the effect of LL-37 on DC maturation, the cells were incubated with LL-37 alone or together with LPS, LTA, flagellin or cytokine cocktail. As a control, a scrambled form of LL-37, sLL-37, was also used alone and together with LPS. The following concentrations of LL-37 and sLL-37 were used: 2, 5 and 20 μg ml−1. To test the involvement of a G protein-coupled receptor, iDCs were pre-incubated with 50 ng ml−1 pertussis toxin (Calbiochem, Bad Soden, Germany) for 1 h before adding LPS with or without LL-37. After 48 h of incubation, the cell culture supernatants were removed and analyzed for cytokines. The cells were harvested, washed once with PBS and used for FACS analysis. In some experiments, DCs were pre-incubated with LL-37 for 2 h. Then the cells were washed in medium without LL-37 and then stimulated with LPS.
Determination of cytokine concentrations
The concentrations of IL-6, TNF-α, IL-12p70, IL-2 and IFN-γ in the cell culture supernatants were determined by commercially available sandwich-type ELISA, according to the manufacturer's instructions (R&D Systems).
DCs were incubated for 30 min at room temperature with the following mAb: FITC-conjugated anti-CD14, FITC-conjugated anti-CCR7, PE-conjugated anti-CD83, PE–Cy5-conjugated anti-CD80 and anti-CD86, and allophycocyanin-conjugated anti-HLA-DR (all BD PharMingen, Heidelberg, Germany) or with their corresponding isotype controls (BD Pharmingen) in PBS supplemented with 1% FBS. To block unspecific binding of the antibodies, the samples were pre-incubated with 20 μg ml−1 purified mouse IgG (Dako, Hamburg, Germany). Then the cells were washed twice with PBS containing 1% FBS, re-suspended in CellFix (BD PharMingen) and analyzed using a FACSort flow cytometer (Becton Dickinson, Heidelberg, Germany).
T cell stimulation
iDCs were stimulated with or without LPS (10 ng ml−1) and with or without LL-37 (20 μg ml−1) for 48 h. Cells were harvested, washed once with PBS and used for T cell stimulation. CD4+ T cells were isolated from blood of a healthy volunteer donor by MACS technique using magnetic beads conjugated with anti-CD4. For allogeneic stimulation, a total of 1.5 × 105 CD4+ cells per well were seeded in triplicate cultures in a 96-well plate and 5 × 104 of the differently stimulated DCs per well were added. Similar conditions were chosen for antigen-specific stimulation of the T cells; however, for this purpose, DCs and T cells from the same volunteer were co-cultured in the presence or absence of 10 U tetanus toxoid (Chiron Vaccines, Marburg, Germany). After 72 h, the cell culture supernatants was removed and analyzed for IFN-γ and IL-2 by ELISA.
To analyze T cell proliferation, the carboxyfluorescein diacetate succinimidyl ester (CFSE) assay was performed according to the manufacturer's instructions (Molecular Probes, Karlsruhe, Germany) with some minor modifications. Briefly, allogeneic naive CD4+ T cells isolated by negative magnetic sorting (Miltenyi Biotec) from peripheral blood of a healthy donor were incubated with 5 μM of CFSE for 10 min at 37°C. The washed T cells (1 × 106) were co-cultured with differently treated DCs (1 × 105) during 5 days in the presence of 1 μg ml−1 anti-CD3 mAb (clone UCHT1; R&D Systems). After co-incubation period, the cells were harvested and assessed for their CFSE content dilution using a FACScalibur (BD Biosciences) sorter. Given that each division cycle is associated with a 2-fold decrease in CFSE fluorescence intensity, the percentages of the daughter cells under each CFSE peak, corresponding to the number of divisions, were determined. As a negative control, CFSE-labeled T cells cultured in the presence of anti-CD3 but without DCs have been analyzed.
Cell viability and apoptosis detection
To exclude an effect of LL-37 on cell viability, 2 × 105 iDCs were seeded in a 24-well plate and incubated with 2, 5 and 20 μg ml−1 LL-37 or 5 μM paclitaxel (Sigma–Aldrich) as positive control for 48 h. Cells were harvested, washed once with PBS and incubated with Annexin V–Biotin (Molecular Probes) in binding buffer (10 mM HEPES, 140 mM NaCl, 2.5 mM CaCl2, pH 7.4) for 15 min. Cells were washed once with binding buffer and incubated with streptavidin-conjugated Alexa Fluor 488 (Molecular Probes) for 30 min at 4°C. After another washing step, cells were re-suspended in binding buffer and 1 μg ml−1 propidium iodide (Molecular Probes) was added immediately before fluorescence detection using a FACSort flow cytometer (Becton Dickinson). In addition, cell viability was also measured using a WST-1 assay according to the manufacturer's protocol (Roche Applied Science, Mannheim, Germany).
Values are displayed as mean ± SD. Comparisons between groups were analyzed by t-test (two-sided), or analysis of variance for experiments with more than two subgroups. Post hoc range tests were performed with the t-test (two-sided) with Bonferroni adjustment. Results were considered statistically significant for P < 0.05.
LL-37 inhibits DC responses to LPS
LL-37 modulates the phenotype of monocytes that are differentiating into DCs (19) and also alters the reaction of monocytes to LPS (20). Because LPS is one of the strongest activators of DC maturation, we wanted to test whether LL-37 might also modulate the response of DCs to LPS. We incubated monocyte-derived iDCs with 10 ng ml−1 LPS with or without LL-37. Stimulation of iDCs with LPS resulted in a release of IL-6 and TNF-α. The release of these cytokines was suppressed by the addition of LL-37 in a concentration-dependent manner (Fig. 1A). Addition of 20 μg ml−1 LL-37 resulted in a total inhibition of IL-6 and TNF-α secretion (Fig. 1A). Incubation of iDCs with LL-37 alone had no effect on cytokine release (Fig. 1A). As the G protein-coupled receptor formyl peptide receptor-like 1 (FPRL1) is one of the proposed receptors for LL-37, we tested the effect of pertussis toxin on the LL-37-dependent inhibition of DC maturation. We found that pertussis toxin had no effect on the inhibition of IL-6 and TNF-α release by LL-37 (data not shown). In a next step, we analyzed the effect of LL-37 on the expression of DC surface activation markers. LPS stimulation of DCs resulted in an up-regulation of HLA-DR, CD80, CD83, CD86 and the chemokine receptor CCR7. This up-regulation was significantly inhibited by LL-37 and reached the level of the unstimulated control cells at a concentration of 20 μg ml−1 LL-37 (Fig. 1B). Incubation of iDCs with LL-37 alone had no effect on the expression of the surface markers (data not shown). In some experiments, a scrambled peptide with the same amino acids as LL-37, called sLL-37, was also used as control to exclude an unspecific effect. sLL-37 had no effect on LPS-induced cytokine release or HLA-DR, CD86, and CD83 up-regulation (Fig. 1A and B). When DCs were pre-incubated with 20 μg ml−1 LL-37 and then washed to remove LL-37, we could not detect a significant impact on the response to LPS as compared with cells that were not pre-incubated (data not shown).
LL-37 inhibits DC maturation induced by other TLR ligands
We then tested whether LL-37 also modulates the activation of iDCs by other TLR ligands and applied LTA or flagellin, ligands for TLR2 and TLR5, respectively. The presence of each substance resulted in the release of IL-6, IL-12p70 and TNF-α (Fig. 2A). This release was completely inhibited by 20 μg ml−1 LL-37 (Fig. 2A). To test whether the LL-37-dependent response is restricted to TLR signaling, we induced maturation of iDCs with a stimulatory cytokine cocktail containing IL-1β, IL-6 and TNF-α and found that this TLR-independent activation of DCs is not modified by LL-37 (Fig. 2B and C). The application of LTA or flagellin also resulted in an up-regulation of HLA-DR and CD83 that was again significantly suppressed by LL-37 (Fig. 2C).
LL-37 does not affect DC viability
LL-37 is known to be potentially cytotoxic at high concentrations (21). We measured cell viability using a WST-1 assay and found no decrease in the number of viable DCs after 48 h of incubation with LL-37 at concentrations up to 20 μg ml−1 as compared with the medium control (data not shown). The number of apoptotic and necrotic cells in response to stimulation of iDC with LL-37 for 48 h was measured by detection of annexin V and propidium iodide staining by FACS (Fig. 3). We could not detect a toxic effect of LL-37 at the concentrations that were used in the experiments.
LL-37 inhibits T cell responses induced by DCs
As LL-37 inhibits DC maturation and down-regulates the expression of molecules that are important for the stimulation of CD4+ T cells, we analyzed the stimulatory potential of LL-37-pre-treated DCs for CD4+ T cells from peripheral blood either in an allogeneic mixed lymphocyte reaction (Fig. 4A) or in a syngeneic antigen-specific response to tetanus toxoid (Fig. 4B). The co-culture of DCs pre-treated with LPS for 48 h with allogeneic CD4+ T cells induced the release of IFN-γ and IL-2 from these cells (Fig. 4A). LL-37 added during the pre-culture of DCs with LPS resulted in a significantly reduced stimulatory capacity of DCs for CD4+ T cells, as tested by IFN-γ and IL-2 release (Fig. 4A). We further analyzed the effect of LL-37 on the ability of DCs to induce naive T cell proliferation in a mixed lymphocyte reaction using CSFE assay. Five-day co-culture of naive T cells with LPS-stimulated DCs resulted in a significant increase of cells undergoing more than two divisions (Fig. 4C). Treatment of DCs with LL-37 before stimulation with LPS blocked their ability to induce naive CD4+ T cell proliferation, returning the percentage of the cells undergoing more than one division to the control level. Upon antigenic stimulation with DCs pre-treated with LPS for 48 h and tetanus toxoid, CD4+ T cells also released IL-2. As in the allogeneic mixed lymphocyte reaction pre-culture of DCs with LL-37 and LPS led to a reduced release of IL-2 by the stimulated CD4+ T cells (Fig. 4B).
The main finding of the present study is that the human anti-microbial peptide LL-37 suppresses the response of iDCs to LPS and other TLR ligands. The presence of LL-37 during the process of maturation renders DCs significantly less capable to activate T lymphocytes. This suggests that the peptide influences the development of an adaptive immune response.
Our data extent the observations of Davidson et al. (19) who described that LL-37 modulates the differentiation of blood-derived monocytes to DCs. DCs derived from monocytes by incubation with IL-4, GM-CSF and LL-37 displayed up-regulated endocytotic capacity, modified phagocytotic receptor expression and function, up-regulated co-stimulatory molecule expression, and enhanced secretion of Th1 cytokines. The authors speculated that these changes might increase antigen capture and presentation. LL-37-derived DCs showed normal maturation in response to LPS. The present data are not in contradiction to those results. Our data show that LL-37 at concentrations that are likely to occur under in vivo conditions inhibits the response of differentiated iDCs to LPS. Whereas Davidson et al. (19) cultured monocytes in the presence of LL-37 and activated these LL-37-derived DCs with LPS but without the further addition of LL-37, we investigated the impact of LL-37 on LPS maturation of iDCs that had not been in contact with LL-37 before. Davidson et al. (19) determined a modulation of LL-37 on iDC generation from monocytes, whereas our data show the influence of LL-37 on the maturation of iDCs. Both mechanisms might be relevant in vivo.
Maturation of DCs is initiated in non-lymphoid tissues upon exposure to exogenous danger signals such as LPS. In the lung, LL-37 is produced during infections and secreted from airway epithelial cells (AECs), neutrophils and macrophages (14, 15). Several studies demonstrated that expression of LL-37 is increased during infection or inflammation (22, 23). Our data suggest that LL-37 acts as a negative feedback signal to modulate the activity of the innate and adaptive immune system. As an example, LL-37 suppresses the capacity of DCs to secrete IL-12, an important factor in the induction of a Th1 response. The inability of T cell activation by LL-37-treated DCs may be explained by the decreased production of IL-12 by DCs in response to TLR signals. LL-37 as part of a fusion protein coding for M-CSFR (J6-1) enhanced anti-tumor responses when applied in a murine cancer model (24). Non-specific or immunological effects have not been ruled out by a control peptide in this study. Injection of LL-37 into mice results in the induction of an adaptive immune response to the peptide. A similar inhibitory effect of LL-37 similar to our results has been recently described for human PBMCs (25). In both settings, the response of immune cells to TLR agonists is inhibited. The exact mechanisms remain to be identified. The effect of LL-37 on DC activity likely depends on the situation.
The local environment plays an important role in the control of DC maturation and activation; however, the mechanisms and mediators are largely unknown. In mucosal tissues, a number of cell types contribute to the local environment that educates immune cells. Epithelial cells secrete different substances that modulate the biology of DCs. The role of TSLP in modulating DC function has already been mentioned in the introduction (4). AECs also secrete growth factors (e.g. GM-CSF) and various chemokines that are known to influence the biology of mucosal DCs (26, 27). Beside LL-37, other anti-microbial peptides are secreted by AECs and have been implicated in the regulation of the function of DCs. Epithelial human β-defensins are supposed to interact with CCR6 and to chemoattract iDCs in vitro (28). The contribution of an individual molecule to the interaction between DCs and AECs is difficult to determine. The results of our experiments indicate that the peptide LL-37 that is produced by AECs, neutrophils, macrophages and other host defense cells inhibits the activation of DCs. The mucosal milieu in the gastrointestinal and the respiratory tracts appears to induce tolerance of DCs. LL-37/hCAP-18 regulates the activation of the adaptive immune system. The biological consequences of this finding will have to be clarified in experiments using appropriate animal models.
The mechanisms by which LL-37 interferes with the activation of DCs by TLR ligands are not entirely clear. LL-37 is known to associate with LPS and potentially prevents binding of endotoxin to pattern recognition receptors (29). In addition to this mechanism, a direct modulation of effector cells is likely because the peptide also inhibits the effects of LTA and flagellin. In contrast, the effects of a cytokine mixture were not suppressed. We therefore propose that LL-37 directly interferes with the TLR pathway. It is not clear at which point of the TLR cascade and by which mechanisms LL-37 exerts its inhibiting effect. Different receptors have been proposed that could mediate effects of the peptide: FPRL1 (30), P2X7 (31) and EGFR (32). In our experiments, pertussis toxin did not significantly inhibit the activities of LL-37. Therefore, the involvement of a G protein-coupled receptor is not likely. In contrast, a synthetic peptide (WKYMVm) has similar effects on DCs as LL-37 and is supposed to exert its effect through the G protein-coupled receptors FPR and/or FPRL2 (33). The effect of LL-37 was not detectable when the peptide was removed after an incubation period. This shows that the effects of LL-37 are not due to a general cytotoxic effect. These data also indicate that LL-37 does not act to modify cellular pathways but potentially interacts with surface structures that are involved in the recognition of microbial structural components.
In conclusion, we demonstrate that the anti-microbial peptide LL-37 modulates DC function and inhibits the activation of these cells by TLR ligands. These data show one more time that so-called ‘anti-microbial peptides’ actually might have other functions in addition to the direct microbicidal effect.
This study was supported by grants of the Deutsche Forschungsgemeinschaft to R.B. (Ba 1641/5-1, 6-1; TR 22/1 A8) and M.L. (TR 22/1 A16).
airway epithelial cell
carboxyfluorescein diacetate succinimidyl ester
fetal bovine serum
formyl peptide receptor-like 1
granulocyte macrophage colony-stimulating factor
immature dendritic cell
mature dendritic cell
tumor necrosis factor-α
thymic stromal lymphopoietin