ABSTRACT
Host defense peptides are immediate responders of the innate immunity that express antimicrobial, immunoregulatory, and wound-healing activities. Neutrophils are a major source for oral host defense peptides, and phagocytosis by neutrophils is a major mechanism for bacterial clearance in the gingival tissue. Dysfunction of or reduction in the numbers of neutrophils or deficiency in the LL-37 host defense peptide was each previously linked with proliferation of oral Aggregatibacter actinomycetemcomitans which resulted in an aggressive periodontal disease. Surprisingly, A. actinomycetemcomitans shows resistance to high concentrations of LL-37. In this study, we demonstrated that submicrocidal concentrations of LL-37 inhibit biofilm formation by A. actinomycetemcomitans and act as opsonins and agglutinins that greatly enhance its clearance by neutrophils and macrophages. Improved uptake of A. actinomycetemcomitans by neutrophils was mediated by their opsonization with LL-37. Enhanced phagocytosis and killing of A. actinomycetemcomitans by murine macrophage-like RAW 264.7 cells were dependent on their preagglutination by LL-37. Although A. actinomycetemcomitans is resistant to the bactericidal effect of LL-37, our results offer a rationale for the epidemiological association between LL-37 deficiency and the expansion of oral A. actinomycetemcomitans and indicate a possible therapeutic use of cationic peptides for host defense.
INTRODUCTION
Antimicrobial peptides are positively charged amphipathic components of the innate immunity in insects, vertebrates, and humans that mediate a broad range of antimicrobial activity (1). Their production is induced by injury or microbial burden, and their microbial targets include the outer and inner membranes and cytoplasmic components. In mammals, apart from their direct microbicidal activity, they act as multifunctional effectors that elicit cellular processes to promote anti-infective and tissue repair responses (2). Since the recognition of their immunoregulatory functions, antimicrobial peptides have been referred to as alarmins (3) or host defense peptides (HDPs), and their protective immunomodulatory activities are being tested as a novel therapeutic approach (4).
Apart from protection against systemic and skin pathogens (5–7) and against lung infections (8), HDPs also maintain a balance in the oral microflora (1, 9, 10). The oral HDPs include α- and β-defensins, histatins, and the cathelicidin LL-37 (11–13).
Periodontitis, the primary cause of tooth loss after the age of 35 (14), is a common disease (15) that involves damage to the tooth-supporting tissue. Periodontal disease results largely from an inappropriate immune response to dysbiotic communities in bacterial biofilms at subgingival sites (16–19).
Dysfunction of neutrophils or reduction in neutrophil numbers was previously associated with the outgrowth of the periodontopathogenic Aggregatibacter actinomycetemcomitans and with the appearance of an aggressive, rapidly progressing periodontal disease (20, 21). Neutrophils are a major source for LL-37 (22). The importance of oral LL-37 has been shown in patients with Kostmann syndrome treated with granulocyte colony-stimulating factor (GCSF) and in patients with Papillon-Lefevre syndrome. Patients with morbus Kostmann suffer from severe congenital neutropenia. Treatment with recombinant GCSF restores their levels of neutrophils. However, despite treatment with GCSF, these patients remain deficient in LL-37 and α-defensin HNP-1 and develop severe periodontal disease (9). Patients suffering from Papillon-Lefevre syndrome lack LL-37 because of an inherited deficiency in serine proteinases that activate LL-37 by cleaving it from its hCAP-18 precursor, and similarly to individuals with Kostmann syndrome, they suffer from severe periodontal disease (23). The aggressive periodontal disease that develops in LL-37-deficient individuals with morbus Kostmann or Papillon-Lefevre syndrome is believed to be promoted by an overgrowth of A. actinomycetemcomitans (9, 23).
Surprisingly, although a lack of LL-37 was correlated with disease caused by A. actinomycetemcomitans overgrowth, this bacterium does not appear to be sensitive to LL-37 when tested in vitro (24). We therefore searched for additional mechanisms by which LL-37 might control the growth of A. actinomycetemcomitans in the oral cavity.
Phagocytosis by neutrophils is a major host defense mechanism for bacterial clearance in the space between the tooth and the surrounding gingival tissue (called gingival sulcus) (25–27). A. actinomycetemcomitans, a Gram-negative capnophilic microorganism (28, 29), can evade neutrophils by utilizing its 116-kDa pore-forming leukotoxin, which lyses human neutrophils and monocytes that express the β2-integrin lymphocyte function-associated molecule 1 (LFA-1) on their surface (30–33). Leukotoxin production varies among A. actinomycetemcomitans strains. The JP2 genotype produces large amounts of leukotoxin due to a 530-bp mutational deletion in the promoter region of the lkt/ltx gene, which encodes leukotoxin (34). Strains of this genotype were associated with aggressive periodontitis in subjects of African origin (35, 36).
Several reports regarding the susceptibility of A. actinomycetemcomitans to neutrophils have been contradictory. Some reported efficient phagocytosis and killing (37), while others found complement-mediated phagocytosis of A. actinomycetemcomitans to be generally inefficient and uptake of antibody-opsonized bacteria to result in the rapid cell death of neutrophils (38). Oral A. actinomycetemcomitans strains were divided into seven serotypes, a, b, c, d, e, f, and g (39, 40). Increased resistance to phagocytic killing has been shown for serotype b strains. This increased resistance was reduced by mutations preventing the formation of the serotype b-specific polysaccharide antigen (41).
Cationic proteins, including the full-length hCAP-18 precursor of LL-37, were previously shown to be capable of acting as opsonins that enhance phagocytosis of systemic pathogens by macrophages and monocytes (42–45). LL-37 at low concentrations was shown to inhibit biofilm formation by Pseudomonas aeruginosa (46). Here we show that at subbactericidal concentrations, LL-37 eliminates A. actinomycetemcomitans by inhibition of biofilm formation and by opsonization.
MATERIALS AND METHODS
Peptides.LL-37 (LLGDFFRKSKEKIGKEFKRIVQRIKDFLRNLVPRTES), tetramethylrhodamine-labeled LL-37 (TMR-LL-37), scrambled LL-37 (sLL-37) (GLKLRFEFSKIKGEFLKTPEVRFRDIKLKDNRISVQR), and 6-carboxyfluorescein (6-FAM)-labeled scrambled LL-37 were purchased from Genemed Synthesis Inc. (San Antonio, TX). The peptides were purified by high-pressure liquid chromatography (HPLC), and purity (greater than 90%) was determined by mass spectrometry.
Bacterial strains, cell lines, and growth conditions.Aggregatibacter actinomycetemcomitans ATCC 29523, JP2, and Y4 were cultured in 0.5% yeast extract, 1.5% Bacto tryptone, 0.75% d-glucose, 0.25% NaCl, 0.075% l-cysteine, 0.05% sodium thioglycolate, and 4% NaHCO3. Strains VT726S, VT726S Ltx− (a kind gift of D. Galli), and ATCC 33384 (a kind gift of U. K. Gursoy) (38) were cultured in tryptic soy broth supplemented with 0.6% yeast extract (TSBYE medium). Kanamycin (100 μg/ml) was added for growth of the leukotoxin-deficient strain VT726S Ltx−. A. actinomycetemcomitans was grown at 37°C in 5% CO2 or under anaerobic conditions, as specified. Escherichia coli ATCC 25922 was grown in LB (Difco, MD, USA) under aerobic conditions. Bacterial purity was determined by phase-contrast microscopy and by Gram staining. The RAW 264.7 murine macrophage cell line (kindly supplied by G. Nussbaum) was grown as described before (47) in Dulbecco modified Eagle medium (DMEM) supplemented with 10% fetal calf serum, 2 mM L-glutamine, and penicillin and streptomycin (both at 100 μg/ml). All cell growth reagents were purchased from Biological Industries (Kibbutz Beit Haemek, Israel).
Growth inhibition by LL-37.A microdilution assay was used. Portions (150 μl) containing approximately 1 × 105 (for strains JP2, ATCC 29523, and Y4) or 1 × 106 (for strains VT726S, VT726S Ltx−, and ATCC 33384) bacteria from overnight cultures were distributed into wells of 96-well plates (Nunc, Denmark) containing 50 μl of LL-37 or scrambled LL-37 diluted in phosphate-buffered saline (PBS) (as specified). Plates were incubated for 16 h for E. coli and 24 h for A. actinomycetemcomitans. Bacterial growth was determined by measuring the absorbance at 595 nm using a GENios microplate reader (Tecan, Austria). Percent growth inhibition was calculated relative to growth of bacteria not treated with LL-37. The MIC was defined as the concentration that inhibited bacterial growth by more than 95% compared to that of controls. Data presented are means and standard deviations from three independent experiments performed in triplicate.
LL-37 binding to A. actinomycetemcomitans.Overnight-cultured A. actinomycetemcomitans (∼109 CFU/ml, verified by plating on blood agar plates) was labeled with fluorescein isothiocyanate (FITC) (0.1 mg/ml in PBS; Sigma-Aldrich) for 30 min at room temperature and washed three times in PBS. Labeled bacteria (approximately 6 × 106 bacteria) were incubated for 10 min at room temperature with tetramethylrhodamine-labeled LL-37 (4.4 μM and 11.1 μM final concentrations). Bacteria were washed twice with PBS and resuspended in 50 μl PBS. From each sample, 5 μl was analyzed using fluorescence microscopy.
Binding of labeled LL-37 and scrambled LL-37 to A. actinomycetemcomitans strains was also quantified by flow cytometry. Overnight-cultured bacteria (20 μl) were incubated with tetramethylrhodamine-labeled LL-37 (0, 4.4, 6.6, 8.8, and 11.1 μM) or 6-FAM-labeled scrambled LL-37 (6.6, 8.8, and 11.1 μM) for 30 min at room temperature. The percentage of peptide-bound bacteria was measured by flow cytometry using an Eclipse fluorescence-activated cell sorter (FACS) analyzer (Icyte, USA). Analysis of raw data was performed using the FlowJo 7.6.5 software (TreeStar, USA).
Inhibition of biofilm formation by LL-37.Overnight cultures of A. actinomycetemcomitans (20 μl, approximately 2 × 107 cells) were inoculated in 180 μl fresh appropriate broth in 96-well plates (Nunc, Denmark) containing LL-37 or sLL-37 diluted in PBS (0,4.4, 6.6, 8.8, 11.1, and 22.2 μM final dilutions). Plates were incubated for 24 h. Biofilm quantification using crystal violet staining was performed as described previously (48). Wells were washed twice with PBS, stained with 0.1% crystal violet for 30 min, and rinsed twice. Two hundred microliters of 30% acetic acid was added to each well, and the plates were incubated for 30 min at room temperature with shaking. The crystal violet solution was collected, and the optical density (OD) at 595 nm was measured using the GENios microplate reader (TECAN, Austria). Percent inhibition of biofilm formation was calculated relative to biofilm formation by bacteria not treated with LL-37.
Preparation of neutrophils.Human peripheral blood was collected from healthy donors in accordance with a protocol that was approved by the institutional ethics committee. Neutrophils were isolated using Histopaque Percoll (Sigma) as described before (49) or by negative selection (50) and brought to a concentration of 1 × 107 cells/ml in Hanks buffer (Biological Industries, Kibbutz Beit Haemek, Israel). Neutrophil purity was determined by flow cytometry using antibodies to CD66b and CD16 (BioLegend, USA).
A. actinomycetemcomitans opsonization by LL-37.A. actinomycetemcomitans (∼109 CFU/ml) was labeled with FITC as described above and added at a multiplicity of infection of 125 to neutrophils and a multiplicity of infection of 100 to RAW cells in the absence or presence of increasing concentrations of LL-37 (6.6 μM, 8.8 μM, and 11.1 μM), or scrambled LL-37 (11.1 μM). Samples were incubated at 37°C in 5% CO2 for 1 h, and binding of labeled bacteria to the cells was measured by flow cytometry using the Accuri C6 FACS analyzer (BD, USA). Analysis of raw data was performed using the FlowJo 7.6.5 software (Treestar, USA). In addition, 5-μl samples were fixed with 1% paraformaldehyde and analyzed by fluorescence microscopy.
Fluorescence microscopy of phagocytosis.RAW 264.7 cells were grown to confluence at 37°C in 5% CO2 in Lab-Tek 8-well glass-bottom chamber slides (Nunc, USA). The wells were washed three times in ice-cold PBS. After the third wash, 200 μl of FITC-labeled A. actinomycetemcomitans cells (∼2 × 105 CFU) preincubated with or without increasing concentrations of tetramethylrhodamine-labeled LL-37 (6.6 μM, 8.8 μM, and 11.1 μM) were added to the adherent cells in each chamber. Slides were incubated for 2 h at 37°C in 5% CO2 and washed three times in ice-cold PBS. The slides were then fixed with 3.7% paraformaldehyde for 5 min and washed twice with ice-cold PBS, and 200 μl of 100% glycerol was added. Opsonization of A. actinomycetemcomitans by tetramethylrhodamine-labeled LL-37 and phagocytosis by murine RAW cells were analyzed by fluorescence microscopy.
Killing of A. actinomycetemcomitans by murine macrophages.Approximately 2 × 105 RAW 264.7 cells were added to 24-well plates (Nunc, Denmark), allowed to adhere for 2 to 4 h at 37°C in 5% CO2, and washed three times in PBS. A. actinomycetemcomitans JP2 preincubated with or without increasing concentrations (6.6 μM, 8.8 μM, and 11.1 μM in DMEM) of LL-37 or sLL-37 was added at a 1:10 ratio of infection. Plates were incubated for 2 h at 37°C in 5% CO2. The medium from each well, which contained extracellular bacteria, was collected, and the remaining adherent cells were lysed with double-distilled water (DDW) for 20 min at room temperature. The cell lysate containing intracellular and cell-bound bacteria was combined with the medium containing the extracellular bacteria, diluted, plated on blood agar, and incubated for 48 h at 37°C in a Bactron anaerobic (85% N2, 10% H2, and 5% CO2) environmental chamber (Sheldon Manufacturing Inc., Cornelius, OR). The resulting colonies were counted, and the percentage of viable bacteria in each sample was then determined by comparing the number of CFU from a control tube without added macrophages (100% viability) to the number of CFU obtained for each bacterial strain incubated with macrophages.
Fluorescence microscopy and Image analysis.An Olympus Fluoview 300 (FV300) confocal microscope (Olympus, Japan) was used. Fluorescence intensity thresholds were set manually for red and green pixels. All analyses were done using FV-10 Fluoview software (Olympus, Japan) and the Image-Pro analyzer (Media Cybernetics, USA). Fold intensity of green and red pixels was measured relative to no LL-37 treatment. Colocalization was measured for green and red pixels, and value shown are Pearson's correlation calculated by Image-Pro software.
Statistical methods.Unless otherwise specified, all data presented are means and standard deviations from three independent experiments performed in triplicate. For statistical significance, Student t test analysis was used. A statistical test was considered significant (*) when the P value was <0.05.
RESULTS AND DISCUSSION
A. actinomycetemcomitans is resistant to LL-37.As previously observed (24), A. actinomycetemcomitans treated with LL-37 at physiological ionic strength maintained growth at all tested LL-37 concentrations (up to 100 μg/ml) (Fig. 1A). The antimicrobial potency of the LL-37 used was confirmed using E. coli, and as expected (51), it resulted in complete growth inhibition at concentrations equal to or greater than 4.4 μM (Fig. 1A and B). Of the six tested A. actinomycetemcomitans strains, growth of four was not affected by LL-37. Growth of strains ATCC 29523 and Y4 was partially (67% and 56%, respectively) inhibited at LL-37 concentrations of ≥8.8 μM (40 μg/ml) (Fig. 1B). A. actinomycetemcomitans Y4 is a serotype b strain. It therefore appears that in contrast to resistance to complement-mediated phagocytosis, where access of complement to the lipopolysaccharide is blocked (41), the serotype b-specific polysaccharide antigen does not provide increased resistance to LL-37.
Resistance of A. actinomycetemcomitans to LL-37. (A) MICs of A. actinomycetemcomitans strains tested in this study. (B and C) Growth inhibition (see Materials and Methods) of A. actinomycetemcomitans strains and E. coli ATCC 25922 grown in the presence of increasing concentrations of LL-37 (B) or scrambled LL-37 (C) (22.2 μM). *, P < 0.05 compared to untreated control (no LL-37) using the Student t test.
Leukotoxin production was not found to affect the ability of A. actinomycetemcomitans to maintain growth in the presence of LL-37. As can be seen in Fig. 1B, the high-leukotoxin-producing JP2 strain, the intermediate-leukotoxin-producing VT726S strain, and the leukotoxin-deficient VT726S mutant VT726S Ltx− strain were all similarly resistant to LL-37.
The ability of LL-37 to partially inhibit growth of A. actinomycetemcomitans ATCC 29523 and Y4 (as well as its ability to completely inhibit growth of E. coli) was sequence specific and was lost by sequence scrambling (Fig. 1C).
A previous report described growth inhibition of A. actinomycetemcomitans with a lower LL-37 concentration of 8.5 μM (38 μg/ml) (52). However, this inhibition was obtained in a nonphysiological, low-ionic-strength environment (10 mM sodium phosphate buffer) (52).
Oral concentrations of LL-37 were shown to reach 0.1 μM (0.5 μg/ml) in saliva (53, 54) and 2.2 μM (10 μg/ml) in gingival crevicular fluid (GCF) of periodontitis patients (55), seemingly lower than that necessary to effectively inhibit A. actinomycetemcomitans growth. Saliva and GCF are constantly secreted, and therefore it is possible that newly secreted LL-37 might continuously accumulate on the pathogen's membrane to reach bactericidal concentrations (51). However, subinhibitory LL-37 concentrations might restrain the growth of A. actinomycetemcomitans by indirect mechanisms.
Binding of A. actinomycetemcomitans by subinhibitory concentrations of LL-37 is mediated by hydrophilic and hydrophobic interactions.Though growth of 4 of the tested A. actinomycetemcomitans strains was not inhibited at LL-37 concentrations as high as 22.2 μM (100 μg/ml) (Fig. 1), strong binding of fluorescently labeled LL-37 to all of the tested strains of A. actinomycetemcomitans could be detected at concentrations as low as 4.4 μM (20 μg/ml) (Fig. 2). Similar to the partial growth inhibition of A. actinomycetemcomitans ATCC 29523 and Y4 by LL-37 (Fig. 1C), binding of LL-37 to A. actinomycetemcomitans JP2 was also sequence specific and was 90% diminished by scrambling (Fig. 3A, 6.6 μM). Binding of LL-37 to A. actinomycetemcomitans JP2 was not reduced by the addition of unlabeled sLL-37 (Fig. 3B), demonstrating that sLL-37 cannot compete with LL-37 for binding to A. actinomycetemcomitans. Scrambled LL-37 has the same charge as LL-37 but lacks the amphipathic structure and is disrupted in its hydrophobic domain. As both LL-37 and sLL-37 share the same charge, it appears that hydrophobic interactions, which are disrupted by scrambling the LL-37 sequence, are important for the ability of LL-37 to bind A. actinomycetemcomitans. However, binding of LL-37 to A. actinomycetemcomitans is also charge sensitive and was reduced by 85% and 50% in the presence of 50 mM MgCl2 and 200 mM NaCl, respectively (Fig. 3C). These results demonstrate that both hydrophobic and hydrostatic interactions are involved in the binding of LL-37 to A. actinomycetemcomitans.
LL-37 binds A. actinomycetemcomitans at submicrocidal concentrations. (A) Fluorescence microscopy analysis of binding of 4.4 μM and 11.1 μM tetramethylrhodamine-labeled LL-37 (red) to FITC-stained A. actinomycetemcomitans JP2 (green). (B) Flow cytometry analysis of A. actinomycetemcomitans strains incubated with increasing concentrations of tetramethylrhodamine-labeled LL-37. Graphs indicate mean cell percentages ± standard deviations from three independent experiments.
Binding of LL-37 to A. actinomycetemcomitans is sequence specific and sensitive to high ionic strength. (A) Flow cytometry analysis of A. actinomycetemcomitans JP2 incubated with increasing concentrations of tetramethylrhodamine-labeled LL-37 and 6-FAM-labeled scrambled LL-37. (B) A. actinomycetemcomitans incubated with tetramethylrhodamine-labeled LL-37 in the presence (+sLL-37) or absence (no sLL-37) of unlabeled 11.1 μM scrambled LL-37. (C) Flow cytometry histogram of binding of 11.1 μM tetramethylrhodamine-labeled LL-37 (TMR-LL-37) to A. actinomycetemcomitans JP2 at physiological ionic strength (PBS) (red) or high ionic strength (200 mM NaCl [black line] or 50 mM MgCl2 [blue line]). The gray filled line represents JP2 without LL-37. Graphs and numbers indicate cell percentages ± standard deviations from two independent experiments.
Based on our results so far, we concluded that at subinhibitory LL-37 concentrations, LL-37 binds to A. actinomycetemcomitans. Next, we tested whether LL-37 at low, sub-growth-inhibitory concentrations can exert nonbactericidal protective actions against the invading A. actinomycetemcomitans.
LL-37 inhibits A. actinomycetemcomitans biofilm formation at low concentrations.In order to withstand and avoid removal by the flow of the saliva or the gingival crevicular fluid, dental pathogens must attach to oral surfaces by constructing biofilms (16). Biofilms provide a protected mode of growth and are difficult to treat due to their high inherent resistance to antimicrobial agents (56). Biofilm formation or integration in an existing oral biofilm (also known as the dental plaque) is therefore essential for dental pathogens. LL-37 was previously shown to inhibit the formation of Pseudomonas aeruginosa biofilms at concentrations far below those which are required to kill or inhibit growth (46). As low concentrations of LL-37 were found to bind A. actinomycetemcomitans, we tested its ability to inhibit biofilm formation by A. actinomycetemcomitans. Biofilm formation of all tested A. actinomycetemcomitans strains was inhibited by LL-37 (though not in a dose-dependent manner in the tested range). LL-37 treatment of strains Y4 and ATCC 29523 (both of which are the most sensitive to growth inhibition by LL-37 [Fig. 1B]) led to a decrease of approximately 50% and 70% in biofilm mass (Fig. 4A). Biofilm formation of both these strains was also most sensitive to sLL-37 (Fig. 4B), though to a lesser extent than to LL-37. Leukotoxin deletion in strain VT726S did not hamper biofilm formation of the mutant progeny VT726S Ltx− (data not shown). For a reason yet unclear, leukotoxin deletion decreased the ability of the VT726S Ltx− mutant to form biofilm in the presence of LL-37 (Fig. 4A) and sLL-37 (Fig. 4B).
LL-37 inhibits biofilm formation of A. actinomycetemcomitans. Biofilm formation of A. actinomycetemcomitans strains on 96-well microtiter plates grown in the presence of increasing concentrations of LL-37 (A) or 22.2 μM scrambled LL-37 (B) for 24 h (see Materials and Methods) is shown.
LL-37 at low concentrations opsonizes A. actinomycetemcomitans, enhances its uptake by neutrophils, and increases its killing by macrophages.Phagocytosis by neutrophils is a major host defense mechanism for bacterial clearance in the gingival sulcus (25, 26). However, previous in vitro studies found A. actinomycetemcomitans to be relatively resistant to phagocytosis (38, 41, 52). The MCP-1 and MCP-2 host defense peptides (currently known as β-defensin A and B) were shown to act as opsonins and to enhance the ability of rabbit alveolar macrophages to ingest Staphylococcus aureus, Klebsiella pneumoniae, Bordetella bronchiseptica, and Candida albicans (45). The LL-37 precursor CAP18 was previously shown to opsonize S. aureus and to increase its phagocytosis by monocytes (but not granulocytes) (44). We therefore tested the ability of subbactericidal concentrations of LL-37 to opsonize A. actinomycetemcomitans and increase its uptake by human neutrophils and murine macrophages.
LL-37 added together with A. actinomycetemcomitans increased bacterial binding by neutrophils (as determined after 1 h of incubation). Opsonization of A. actinomycetemcomitans with 11.1 μM (50 μg/ml) LL-37 increased the binding of A. actinomycetemcomitans JP2 by neutrophils by approximately 10-fold (Fig. 5A) and increased the uptake of the four other tested strains (excluding strain ATCC 29523 where the increase in uptake was not statistically significant) by 2.8 to 9.5-fold (Fig. 5B). Leukotoxin inactivation in VT726S Ltx− increased its LL-37-mediated uptake by neutrophils by 4-fold compared to that of its leukotoxin-producing VT726S parent strain. This is in agreement with the increased biofilm formation sensitivity to LL-37 generated by leukotoxin inactivation in the VT726S Ltx− mutant.
LL-37 acts as an opsonin of A. actinomycetemcomitans. (A) FACS analysis (fold increase of mean fluorescence intensity [MFI]) of human neutrophils incubated with FITC-labeled A. actinomycetemcomitans JP2 in the absence (black lines) or presence (red lines) of 6.6 μM, 8.8 μM, and 11.1 μM LL-37 and 11.1 μM scrambled LL-37. Neutrophils without labeled bacteria and without LL-37 were used as baseline, which is represented by a filled blue histogram. (B) Fold increase of mean fluorescence intensity of A. actinomycetemcomitans (A. a) opsonization in the presence of increasing concentrations of LL-37. Neutrophils with A. actinomycetemcomitans only were used as a baseline. (C) Fold increase of mean fluorescence intensity of A. actinomycetemcomitans VT726S Ltx− opsonization after incubation with LL-37 or scrambled LL-37. Neutrophils with labeled bacteria and no LL-37 were used as a baseline. (D) Fluorescence microscopy of human neutrophils (arrows) incubated with FITC-labeled A. actinomycetemcomitans JP2 (A. a) in the absence or presence of LL-37. *, P < 0.05 compared to baseline using the Student t test.
Scrambled LL-37, which shares the same charge as LL-37 but lacks the amphipathic structure, produced a poor, nonstatistically significant improvement in binding of A. actinomycetemcomitans JP2 (Fig. 5A) and VT726S Ltx− (Fig. 5C) by neutrophils. This is in agreement with the inability of sLL-37 to inhibit growth (Fig. 1) and to bind A. actinomycetemcomitans (Fig. 3). The weaker opsonization and growth inhibition activity by sLL-37 (compared with LL-37) is likely to result from its weaker affinity to A. actinomycetemcomitans, as demonstrated in Fig. 3.
In contrast to the case for neutrophils, adding LL-37 together with A. actinomycetemcomitans to nonactivated murine macrophage-like RAW 264.7 cells barely improved bacterial uptake as determined after 1 h (Fig. 6). However, incubating LL-37 with A. actinomycetemcomitans prior to its addition to the macrophage-like cells induced agglutination of the bacteria, and the bacterial aggregates were readily phagocytized by the RAW cells in an LL-37 dose-dependent manner (Fig. 7). A 3-fold increase in bacterial uptake was observed with an LL-37 concentration of 11.1 μM (Fig. 7, bottom panel). Three-dimensional reconstruction of the confocal imaging confirmed that the bacteria were indeed phagocytized and internalized by the RAW cells (see Movie S1 in the supplemental material). LL-37 agglutination improved killing of A. actinomycetemcomitans JP2 by the RAW cells in a statistically significant, dose-dependent manner, rising from 46% without addition of LL-37 to 97% when cells were preopsonized with 11.1 μM LL-37 (Fig. 8). Adding sLL-37 slightly increased the killing of A. actinomycetemcomitans JP2 by RAW cells but not in a statistically significant manner, in agreement with the sequence dependence of the antimicrobial action of LL-37 described above (Fig. 1, 3, and 5).
LL-37 has only a moderate effect on binding of A. actinomycetemcomitans by murine macrophages. Macrophage-like RAW 264.7 cells were incubated with FITC-labeled A. actinomycetemcomitans in the absence (black lines) or presence (red lines) of increasing concentrations of LL-37 (6.6 μM, 8.8 μM, and 11.1 μM). The fold increase in mean fluorescence intensity of A. actinomycetemcomitans JP2 binding to RAW cells in the presence of LL-37 is shown. RAW cells without labeled bacteria and with no LL-37 are represented by a filled blue line. RAW cells incubated with labeled bacteria without LL-37 were used as a baseline for fold increase calculations.
Preincubation of A. actinomycetemcomitans JP2 with LL-37 enhances its phagocytosis by murine macrophages. FITC-labeled A. actinomycetemcomitans (A. a) was incubated in the absence of tetramethylrhodamine-labeled LL-37 or in the presence of increasing concentrations of tetramethylrhodamine-labeled LL-37 (6.6 μM, 8.8 μM, and 11.1 μM) for 10 min at room temperature, washed twice, incubated with macrophage-like RAW 264.7 cells for 2 h, washed twice, and analyzed using fluorescence microscopy. Colocalization analysis appears above the colocalization panels for each concentration. Mean fluorescence intensities from two independent experiments are shown in the bottom panel.
LL-37 enhances killing of A. actinomycetemcomitans by murine macrophages. A. actinomycetemcomitans JP2 was incubated with macrophage-like RAW 264.7 cells at a multiplicity of infection of 10 for 2 h in the presence of LL-37 or scrambled LL-37 at increasing concentrations. Murine macrophage killing of A. actinomycetemcomitans JP2 was determined after 2 h of incubation by plating the samples on blood agar. A. actinomycetemcomitans JP2 without macrophages was used as a reference of bacteria number and no killing (0%).
Conclusion.Dysfunction of neutrophils or reduction in their numbers (20, 21), as well as deficiency in LL-37, was previously linked with outgrowth of A. actinomycetemcomitans and the development of an aggressive periodontal disease (9, 23). Surprisingly, A. actinomycetemcomitans was found to be highly resistant to LL-37 (Fig. 1A and B). However, LL-37, even at low concentrations, bound A. actinomycetemcomitans (Fig. 2 and 3) and was found to inhibit its biofilm formation at low, submicrocidal concentrations (Fig. 4). Since cationic peptides (including LL-37) were shown to act as opsonins (42–45), we tested whether A. actinomycetemcomitans can be opsonized and cleared by neutrophils and macrophages at relatively low concentrations of LL-37. Our results demonstrate that at sub-growth-inhibitory concentrations, LL-37 efficiently opsonized A. actinomycetemcomitans and enhanced its removal by neutrophils (which are known to be abundantly present the gingival sulcus) (Fig. 5).
A. actinomycetemcomitans agglutination by LL-37 (rather than opsonization of planktonic bacteria) is suggested as a possible mechanism by which LL-37 improves phagocytosis and killing of A. actinomycetemcomitans by macrophages. The LL-37 concentrations found in our study as necessary to improve the removal of A. actinomycetemcomitans by neutrophils and macrophages are still higher than those measured in saliva. Newly secreted LL-37 is likely to continuously accumulate on membranes of A. actinomycetemcomitans, reaching the concentrations required for mediating their clearance by local neutrophils and macrophages.
ACKNOWLEDGMENTS
We are very grateful to Avi-Hai Hovav and Gabriel Nussbaum for useful discussions and to Ronit Naor and Luba Eli-Berchoer for expert technical assistance.
This work was supported by the Israel Science Foundation (grant no. 208/10).
FOOTNOTES
- Received 20 November 2012.
- Returned for modification 16 December 2012.
- Accepted 2 July 2013.
- Accepted manuscript posted online 8 July 2013.
Supplemental material for this article may be found at http://dx.doi.org/10.1128/IAI.01288-12.
- Copyright © 2013, American Society for Microbiology. All Rights Reserved.
