Muramyl dipeptide potentiates a Bacillus anthracis poly-γ-D-glutamic acid capsule surrogate that induces maturation and activation of mouse dendritic cells

Abstract

Poly-γ-D-glutamic acid (PGA), a component of the anthrax-causing bacterium, acts as a key factor in its ability to cause disease by preventing immune cells from engulfing it. Furthermore, PGA can stimulate mouse immune cells called macrophages to release signaling molecules known as cytokines, doing so through a receptor called Toll-like receptor 2. Peptidoglycan, a major part of bacterial cell walls, triggers inflammatory responses in the host. This study investigated whether PGA can induce maturation and cytokine production in immature mouse dendritic cells in the presence of muramyl dipeptide, which is the smallest active part of peptidoglycan known to stimulate the immune system. Stimulation of immature dendritic cells with either PGA or muramyl dipeptide alone increased the expression of costimulatory molecules and MHC class II proteins, all of which are markers on the cell surface indicating maturation. These effects were even stronger when PGA and muramyl dipeptide were used together. PGA alone was sufficient to trigger the production of TNF-α, IL-6, MCP-1, and MIP1-α, whereas muramyl dipeptide alone did not do so under the same conditions. Treatment with muramyl dipeptide enhanced the PGA-induced production of the inflammatory signaling molecules tested. However, the enhanced effect seen when PGA and muramyl dipeptide were combined was not observed in dendritic cells that lacked either Toll-like receptor 2 or nucleotide-binding oligomerization domain 2. Additionally, muramyl dipeptide increased the PGA-induced activation of MAP kinases and NF-κB, which are critical for the production of cytokines. Blocking MAP kinases and NF-κB reduced the muramyl dipeptide enhancement of PGA-induced cytokine production. Moreover, when immune cells from the spleen were grown together with dendritic cells matured by PGA and muramyl dipeptide, they showed higher production of IL-2 and IFN-γ compared to those grown with dendritic cells matured by PGA alone. Taken together, these results suggest that PGA and muramyl dipeptide work together to induce inflammatory responses in mouse dendritic cells through Toll-like receptor 2 and nucleotide-binding oligomerization domain 2 via the MAP kinase and NF-κB pathways, subsequently leading to the activation of lymphocytes.

Introduction

Dendritic cells are specialized immune cells that can capture antigens and present them to lymphocytes. They are found in almost all tissues of the body, acting as professional sentinel cells. Immature dendritic cells are highly efficient at engulfing substances but have a limited capacity to present antigens. When immature dendritic cells encounter molecules associated with pathogens, such as lipopolysaccharides or lipoteichoic acid, released by invading microbes, they are stimulated to mature. This maturation process equips them with the ability to present antigens effectively and increases their production of cytokines and maturation markers like CD80 and CD86.

Bacillus anthracis is a highly dangerous bacterium responsible for anthrax. It is classified as a Tier 1 Biological Select Agent due to its potential use as a biological weapon. Once Bacillus anthracis spores enter a host, they are taken up by phagocytic cells, such as macrophages and dendritic cells. In the regional lymph nodes, the spores develop into encapsulated bacteria that produce lethal and edema toxins, reaching high concentrations in the blood.

Toxins and poly-γ-D-glutamic acid are essential for the pathogenesis of anthrax. The toxins produced by the multiplying bacteria lead to swelling or death of the host. The PGA capsule enables the bacterium to resist antibodies, helping it to evade and prevent phagocytosis. Macrophage stimulation by the PGA capsule enhances cell death induced by the lethal toxin. Additionally, the PGA capsule triggers host immune responses in mouse macrophages that depend on Toll-like receptor 2.

Peptidoglycan, a major component of bacterial cell walls, is a glycan polymer linked by amino acids. It was suggested that Toll-like receptor 2 recognizes peptidoglycan. However, the peptidoglycan preparation used in that study was not pure and contained lipoteichoic acid or lipoproteins, which likely influenced the results. Subsequent research showed that highly purified peptidoglycan does not stimulate Toll-like receptor 2. Instead, peptidoglycan activates nucleotide-binding oligomerization domain 1 and 2, which sense D-diaminopimelic acid and muramyl dipeptide, respectively. Peptidoglycan from Bacillus anthracis can stimulate human monocytes to produce TNF-α and activate human platelets.

Increasing evidence suggests that bacterial cell-wall components, such as lipopolysaccharides, lipoteichoic acid, and lipoproteins, can work together with peptidoglycan breakdown products to activate various cell types and induce cytokine production. Agonists of nucleotide-binding oligomerization domain 1 and muramyl dipeptide enhance the lipopolysaccharide-induced expression of activation markers and pro-inflammatory signaling molecules in dendritic cells through their respective activities. In addition, muramyl dipeptide enhances the expression of lipoteichoic acid-induced inflammatory signaling molecules, including TNF-α, IL-12, nitric oxide, and cyclooxygenase-2, in macrophages or dendritic cells. Muramyl dipeptide also stimulates lipoprotein-induced cytokine production in dendritic cells and macrophages. However, it was not known whether PGA can interact cooperatively with other bacterial components or peptidoglycan breakdown products to induce immune responses. Therefore, this study investigated the cooperative activity of PGA and a synthetic muramyl dipeptide, free of lipoteichoic acid and lipoproteins, on the maturation and activation of dendritic cells.

Materials and methods

Reagents
All fluorochrome-labeled antibodies were obtained from Biolegend (San Diego, CA, USA). Muramyl dipeptide (MDP) was purchased from InvivoGen (Cayla SAS, Toulouse, France). MAP kinase inhibitors were sourced from Calbiochem (Darmstadt, Germany). Parthenolide was purchased from Sigma-Aldrich Chemical Inc. (St. Louis, MO, USA). Antibodies against IκB-α, β-actin, and all MAP kinases used for Western blotting were purchased from Cell Signaling Technology (Beverly, MA, USA). Cell culture reagents were obtained from Invitrogen (Carlsbad, CA, USA). Mouse granulocyte-macrophage colony-stimulating factor (GM-CSF) was obtained from Peprotech (Rocky Hill, NJ, USA).

Purification of PGA
The poly-γ-D-glutamic acid (PGA) capsule was isolated from the culture supernatants of Bacillus licheniformis. Bacillus licheniformis was cultured in E medium (prepared in-house) to maximize the production of the D-enantiomer of PGA. Under these conditions, the produced PGA is approximately 80–90% D-enantiomeric. The purified capsule was fragmented to a size of less than 50 kilodaltons using 0.6 M hydrochloric acid, as previously described. After dialysis to remove salts, the fragmented PGA preparation was used for experiments. The molecular identity of the PGA preparation was confirmed using nuclear magnetic resonance spectroscopy. The preparation was found to be free of contaminating endotoxin, lipoprotein, lipoteichoic acid, or peptidoglycan.

Animals
The animal studies were approved by the Korea National Institute of Health. C57BL/6 mice were obtained from Orient Bio (Seoul, South Korea). Toll-like receptor 2 (TLR2)- and nucleotide-binding oligomerization domain 2 (NOD2)-knockout mice were purchased from the Jackson Laboratory (Bar Harbor, ME, USA).

Differentiation of DCs
Bone marrow cells were prepared from TLR2- and NOD2-knockout mice and their wild-type control mice as previously described. For dendritic cell (DC) differentiation, samples of the bone marrow cell preparation, from which red blood cells had been removed, were plated onto 100-mm Petri dishes containing 20 nanograms per milliliter of mouse GM-CSF and 50 micromolar β-mercaptoethanol. The cells were then cultured for 8 days with medium replacement every 3 days. On day 8, the differentiated immature DCs (iDCs) were collected by pipetting and used for experiments.

Treatment of iDCs with PGA and/or MDP
Immature DCs (1 × 10⁶ cells/ml) were plated into the wells of 12-well culture plates to which PGA (at concentrations of 10 or 30 micrograms per milliliter) had been added with or without MDP (at a concentration of 1 microgram per milliliter). The cells were incubated for various time periods (24 hours for flow cytometry and 0.25, 0.5, 1, 2, or 4 hours for Western blotting).

DCs-splenocytes co-culture
Immature DCs (5 × 10⁴ cells/ml) were incubated with 30 micrograms per milliliter of PGA and/or 1 microgram per milliliter of MDP for 24 hours in 96-well plates. The cells were then co-cultured with mouse splenocytes (5 × 10⁵ cells/ml) for 24 hours.

ELISA
The expression levels of mouse TNF-α, IL-6, MCP-1, MIP-1α, IL-2, and IFN-γ were evaluated using ELISA kits (Biolegend) according to the manufacturer’s protocol.

Flow cytometry
Cell surface marker expression was evaluated using flow cytometry as previously described. Briefly, DCs were stained with peridinin chlorophyll protein-labeled anti-CD11c, allophycocyanin-labeled anti-CD80, fluorescein isothiocyanate-conjugated anti-CD86, or anti-MHC class II I-Ab antibodies. Positive cells for CD11c and each of the aforementioned maturation markers were gated and examined using a flow cytometer (Becton Dickinson, Mountain View, CA, USA) with FlowJo v.6 software (Tree Star Inc., Ashland, OR, USA).

Western blot analysis
Detailed procedures for Western blotting have been previously reported. In brief, 30 micrograms of cell lysates were separated using an SDS-PAGE gel (10% weight/volume acrylamide) and then probed with antibodies against phosphorylated-ERK (p-ERK), ERK, phosphorylated-p38 kinase (p-p38), p38 kinase, phosphorylated-JNK (p-JNK), and JNK. Western blot signals were analyzed using a chemiluminescent reagent (Pierce, Rockford, IL, USA) and a chemiluminescence detection system (Bio-Rad, Hercules, CA, USA). Relative band intensities were quantified by densitometry using AlphaView SA software (Cell Biosciences, Santa Clara, CA, USA).

Statistical analysis
Experimental values are expressed as mean values ± standard deviation (SD). All experiments were conducted at least three times. Statistical significance was assessed using the Student’s t-test or a multiple-comparison test after one-way ANOVA.

Results

PGA and MDP cooperatively augment the expression of DC maturation markers

Immature dendritic cells increase the levels of co-stimulatory molecules during their maturation, which is crucial for initiating adaptive immunity. To investigate whether PGA and/or MDP influence the levels of CD80 and CD86 or the expression of MHC class II proteins, immature dendritic cells were treated with PGA and MDP for 24 hours. Treatment with PGA or MDP alone increased these phenotypic markers, and the enhancing effects were greater when cells were treated with both PGA and MDP together. Figure 1 shows the mean fluorescence intensity for CD80, CD86, and MHC class II proteins under different treatment conditions, with statistically significant increases observed in the combined treatment group compared to untreated cells and cells treated with PGA or MDP alone.

MDP enhances PGA-induced cytokine expression

The study examined whether MDP enhances the production of inflammatory cytokines mediated by PGA in dendritic cells. Immature dendritic cells were treated with PGA and MDP for 24 hours. PGA alone increased the secretion of TNF-α, IL-6, MCP-1, and MIP-1α, whereas MDP alone did not induce their production. However, the combination of PGA and MDP significantly increased the levels of all four inflammatory mediators compared to the levels induced by PGA alone. Furthermore, the combination increased IL-6 and MIP-1α levels in a dose-dependent manner, while such dose-dependent enhancements were not observed for TNF-α and MCP-1. This indicates that MDP enhances the capability of PGA to induce cytokine expression in dendritic cells.

TLR2 and NOD2 are required for MDP enhancement of PGA-mediated cytokine expression

PGA is recognized by TLR2, whereas MDP is recognized by NOD2. To examine whether TLR2 and/or NOD2 are critically involved in the PGA/MDP-mediated expression of TNF-α and IL-6, TLR2- and NOD2-knockout immature dendritic cells and their wild-type counterparts were treated with PGA and MDP for 24 hours. MDP synergistically enhanced PGA-mediated TNF-α and IL-6 expression in wild-type dendritic cells, but this effect was not observed in TLR2- or NOD2-knockout dendritic cells.

PGA/MDP induces cytokine expression via MAP kinase pathways

Since MAP kinase pathways are necessary for cytokine expression and PGA-induced signaling pathways, the study assessed whether MDP can modulate PGA-induced activation of p38 kinase, ERK, and JNK by Western blotting. MDP enhanced PGA-induced phosphorylation of p38 kinase up to 4 hours, particularly at 1 hour. Phosphorylation of JNK and ERK by PGA increased in the presence of MDP at 0.5 hours. Additionally, the levels of TNF-α and IL-6 induced by PGA/MDP were significantly decreased by inhibitors against ERK, p38 kinase, or JNK. These data indicate that the MAP kinase pathways contribute to the MDP enhancement of PGA-induced TNF-α and IL-6 secretion in dendritic cells.

MDP increases PGA-mediated cytokine production through NF-κB activation

NF-κB is crucially involved in the expression of inflammatory cytokines. Accordingly, Western blotting was used to examine whether MDP could enhance PGA-induced activation of NF-κB. When immature dendritic cells were treated with PGA and/or MDP, PGA-induced phosphorylation of IκB-α, an inhibitor of NF-κB, was enhanced by MDP up to 4 hours, with maximal induction observed at 1 hour. Concurrently, the degradation of IκB-α was markedly increased by PGA/MDP at both 1 and 2 hours after treatment compared to that by PGA alone. Furthermore, the NF-κB inhibitor parthenolide suppressed PGA/MDP-mediated TNF-α and IL-6 production. These results suggest that the NF-κB pathway is necessary for the MDP enhancement of PGA-mediated cytokine production.

MDP augments lymphocyte activation induced by PGA-matured DCs

To examine whether MDP enhances the ability of dendritic cells to mediate lymphocyte activation, splenocytes were co-incubated for 24 hours with dendritic cells matured by PGA and/or MDP. The expression of cytokines IL-2 and IFN-γ was then measured by ELISA. The combination of PGA and MDP significantly increased IL-2 and IFN-γ expression compared to cells treated with PGA or MDP alone. These results imply that MDP enhances the PGA-induced activation of lymphocytes.

Discussion

PGA is a significant factor in how anthrax causes disease because it can resist being engulfed by immune cells and is not affected by antibodies. Notably, as this study has shown, purified PGA itself stimulates the immune system and triggers cytokine production in macrophages through TLR2 and TLR6. To further understand how PGA affects cytokine production in the context of Bacillus anthracis infection, the cooperative activity of PGA and MDP, a minimal structural component of peptidoglycan known to stimulate immunity, was investigated. The findings demonstrated that MDP not only enhances the PGA-induced expression of maturation markers and cytokines in dendritic cells but also boosts the activation of lymphocytes by dendritic cells matured with PGA. Moreover, the study showed that TLR2, NOD2, MAP kinase, and NF-κB pathways are necessary for MDP to enhance PGA-mediated cytokine production. These data suggest that PGA and MDP work together in the inflammatory responses against Bacillus anthracis infection.

TLR2 and NOD2 appear to be essential for the cytokine expression mediated by PGA and MDP, as the potentiation of PGA-induced cytokine expression by MDP was not observed in immature dendritic cells lacking either TLR2 or NOD2. The synergistic increase in cytokine expression by PGA and MDP might result from enhanced expression of TLR2 and/or NOD2 due to the combined stimulation, leading to increased activation of downstream signaling pathways. Indeed, it has been shown that costimulation with lipoteichoic acid and MDP enhances TLR2 and NOD2 expression in bovine mammary epithelial cells. Additionally, NOD2 expression increases in response to lipoteichoic acid in odontoblast-like cells, and both lipopolysaccharide and lipoteichoic acid increase NOD2 levels in human periodontal cells. However, the enhancement of TLR2 and/or NOD2 expression is likely to result at least in part from the synergistic action of PGA and MDP because MDP-enhanced phosphorylation of MAP kinases and degradation of IκB-α were observed within 4 hours after stimulation.

Conversely, the synergistic activation of the signaling pathways induced by PGA and MDP may be involved in this potentiation effect. Indeed, MDP enhanced PGA-induced phosphorylation of MAP kinases and degradation of IκB-α in dendritic cells. Additionally, inhibitors of all three MAP kinases and an NF-κB inhibitor significantly reduced the production of TNF-α and IL-6 induced by PGA and MDP. Among the MAP kinases, p38 kinase may be particularly important in the MDP enhancement of PGA-mediated cytokine expression because a p38 kinase inhibitor attenuated cytokine expression induced by PGA/MDP, but not by PGA alone. Consistent with these results, MDP enhances PAM3CSK4-induced activation of NF-κB in human monocytes and increases lipopolysaccharide-mediated MAP kinases and NF-κB activation in macrophages. Furthermore, MDP decreases the expression of suppressor of cytokine signaling 1, an inhibitor of TLR responses.

Septicemia, a severe condition characterized by excessive cytokine production in response to pathogen infection, contributes to blood clotting, unstable blood flow, and leakage from blood vessels. Peptidoglycan from Bacillus anthracis can induce a strong cytokine response in immune cells and activate human platelets through FcγRII and complement activity. It also causes changes in a rat model consistent with widespread blood clotting within vessels. Additionally, anthrax infection increases NOD2-dependent IL-1β secretion in mouse macrophages. PGA also induces macrophages to produce cytokines and nitric oxide, which are key factors in the development of septicemia. High levels of PGA have been detected in the blood of anthrax infection models. In the later stages of infection, Bacillus anthracis can reach concentrations of 10⁷ to 10⁸ bacteria per milliliter of blood. Thus, the release of peptidoglycan and PGA by Bacillus anthracis in the later phase of infection might contribute to Bacillus anthracis-mediated septicemia.

In conclusion, the findings demonstrate that PGA and MDP are recognized by TLR2 and NOD2, respectively, and that this recognition synergistically enhances the maturation and activation of dendritic cells via the MAP kinase and NF-κB pathways, resulting in enhanced lymphocyte activation by dendritic cells matured with PGA and MDP. This study is the first to report the synergistic induction of inflammatory responses mediated by PGA and MDP, and the findings suggest that these two factors may contribute to the pathogenesis of Bacillus anthracis infection.