We previously showed that Lactobacillus strains having a rigid cell wall resistant to intracellular digestion can stimulate macrophages to induce large a quantity of interleukin-12 (IL-12). In this study, we examined the influence of lactobacilli and bacterial cell wall components on IL-12 production by macrophages that was induced by Lactobacillus casei, which has a rigid cell wall. Easily digestible lactobacilli such as Lactobacillus johnsonii and Lactobacillusplantarum or their intact cell walls (ICWs) weakly or very weakly induced IL-12 production by macrophages, and inhibitedL. casei-induced IL-12 production. While the ICW of L. casei was resistant to intracellular digestion and did not inhibit L. casei-induced IL-12 production, its polysaccharide-depleted ICW, i.e. intact peptidoglycan, was sensitive to intracellular digestion and inhibited L. casei-induced IL-12 production. Furthermore, the peptidoglycans of L. johnsonii, L. plantarum and Staphylococcus aureus also inhibited L. casei-induced IL-12 production. Peptidoglycans from lactobacilli suppressed L. casei-induced expression of IL-12p40 but not IL-12p35 mRNA. Inhibition of IL-12 production by peptidoglycan was mitigated in Toll-like receptor 2 (TLR2)-deficient macrophages compared with the inhibition in wild-type macrophages. A derivative of the minimal structural unit of peptidoglycan (6-O-stearoyl-muramyl dipeptide) recognized by nucleotide-binding oligomerization domain 2 (NOD2) could also suppress L. casei-induced IL-12 production. These findings demonstrate that easily digestible bacteria and peptidoglycan suppress IL-12 production through pattern recognition receptors such as TLR2 and NOD2. IL-12 production in the gut may be negatively regulated by the simultaneous inhibitory actions of various resident bacteria that are susceptible to intracellular digestion.
Lactobacilli are members of the indigenous intestinal microflora in humans, and they are currently recognized as the most popular probiotics, i.e. live micro-organisms that, when administered in adequate amounts, confer a health benefit on the host.1 Of the several health-promoting effects of probiotic lactobacilli, their immunoregulatory functions have been studied with the greatest interest.2,3 Some strains of lactobacilli stimulate macrophages and dendritic cells to secrete higher amounts of interleukin-12 (IL-12), which elicits the innate immune response, and thereby the incidence of infectious diseases and cancers decreases.4,5 The IL-12-inducing ability of Lactobacillus strains is now considered as a critical index in the selection of immunostimulatory probiotic strains. However, dysregulated overproduction of IL-12 and the related cytokine IL-23, which shares the p40 subunit with IL-12, causes some types of inflammation.6,7 Therefore, the production of IL-12 must be tightly regulated to maintain homeostasis in the host.
IL-12 is a heterodimeric protein composed of two disulphide-linked subunits, namely, p35 and p40. The regulation of IL-12 expression is a complex process, because the genes encoding these subunits are located on different chromosomes, and their expression is differentially regulated by various stimuli. Recent reviews5,8 provide good, detailed summaries of the regulation of IL-12 expression. (i) The expression of p35 and p40 involves the activation of nuclear factor (NF)-κB, and in particular c-Rel, which is dependant on a myeloid differentiation primary response gene 88 (MyD88)-mediated signalling pathway, after Toll-like receptors (TLRs) are stimulated by microbial components. (ii) The Toll/IL-1 receptor-domain-containing adaptor protein inducing interferon (IFN)-β (TRIF)-mediated pathway activates IFN regulatory factor 3 (IRF3) and enhances p35 expression both directly and indirectly by inducing the secretion of IFN-β. Nucleotide-binding oligomerization domain 2 (NOD2) is a cytoplasmic receptor for the by-products of peptidoglycan digestion; the effect of NOD2 stimulation on IL-12 production is a subject of controversy. Tada et al.9 demonstrated the synergistic effect of muramyl dipeptide (MDP), a NOD2 ligand, with that of the TLR3, TLR4 or TLR9 ligand on the induction of IL-12 production by human dendritic cells, while Watanabe et al.10 revealed that MDP inhibited IL-12 production by murine macrophages. Moreover, a recent study showed that certain strains of lactobacilli that have weak IL-12-inducing ability inhibit the IL-12 production that is potently induced by some other strains in human and murine dendritic cells.11,12 However, the precise mechanisms of this inhibitory effect have not yet been elucidated. Taken together, these findings suggest that IL-12 production is tightly regulated through various mechanisms, probably because of the pivotal biological functions of this cytokine.
Recently, we showed that, for Lactobacillus strains to induce IL-12 production, it is critical for their cell walls to be resistant to intracellular digestion.13Lactobacillus strains that have a rigid cell wall that is resistant to intracellular digestion, such as Lactobacillus casei and Lactobacillus rhamnosus, potently induced IL-12 production by mouse peritoneal macrophages, whereas those strains that have an easily digestible cell wall, such as Lactobacillus johnsonii, hardly induced IL-12 production. Our results also revealed that L. casei induces IL-12 production through TLR2-independent mechanisms and that the insoluble intact cell wall (ICW) and not the solubilized cell wall of this strain is responsible for inducing IL-12 production. These findings strongly suggest that the ICW structure rather than soluble cell wall components of lactobacilli is essential for the effective induction of IL-12 production, at least in the case of mouse peritoneal macrophages. Sekine et al.14 examined the antitumour effects of three kinds of morphologically distinct cell wall preparations isolated from Bifidobacterium infantis and showed that the antitumour activity of the cell wall preparation increases in proportion to the degree of integrity of its physical form. The potent antitumour activity of the cell wall preparation having physical integrity of the cell wall structure, i.e. ICW, might be related to its potent ability to induce IL-12 production.
In the present study, we further investigated the IL-12 production by macrophages stimulated with a combination of L. casei and other strains possessing easily digestible cell walls. The data obtained showed that the latter strains inhibited L. casei-induced IL-12 production and that the peptidoglycan in the cell walls of these strains was responsible for this inhibitory effect. This study reveals a novel mechanism by which peptidoglycan from Lactobacillus cell walls mediates the negative regulation of IL-12 production.
Materials and methods
Reagents and culture medium
Trypsin and cytochalasin D were purchased from Sigma (St Louis, MO). N-acetylmuramidase SG isolated from Streptomyces globisporus was obtained from Seikagaku Corp. (Tokyo, Japan). Benzon nuclease and pronase were purchased from Merk (Darmstadt, Germany) and Roche Diagnostics (Indianapolis, IN), respectively. Purified insoluble peptidoglycan from Staphylococcus aureus, MDP, 6-O-stearoyl-MDP (L18-MDP), and lauroyl-γ-d-glutamyl-meso-diaminopimelic acid (C12-DAP) were purchased from InvivoGen (San Diego, CA). RPMI-1640 medium (Sigma) supplemented with 10% heat-inactivated fetal calf serum, 100 U/ml penicillin, 100 μg/ml streptomycin, and 0·05 mm 2-mercaptoethanol was used to culture macrophages.
Female BALB/c mice and male C57BL/6 mice were purchased from Japan SLC (Shizuoka, Japan). Male TLR2-deficient mice with a C57BL/6 genetic background were obtained from Oriental Bioservice (Kyoto, Japan). Animals aged between 8 and 12 weeks were used. All the experiments were performed in accordance with the guidelines for the care and use of laboratory animals set by Yakult Central Institute.
Heat-killed lactobacilli were used in this study, unless otherwise specified. On the basis of its resistance to gastric juices and bile acid and its ability to survive in the gastrointestinal tract, the probiotic strain L. casei strain Shirota (YIT 9029) was originally isolated at Yakult Central Institute. Lactobacillus johnsonii JCM 2012T and Lactobacillus plantarum ATCC 14917T were obtained from the Japan Collection of Microorganisms (Wako, Japan) and American Type Culture Collection (Rockville, MD), respectively. The bacteria were cultured at 37° for 20 hr in lactobacilli-de Man, Rogosa and Sharpe (MRS) broth (Difco, Detroit, MI), washed with sterile distilled water, heated at 100° for 30 min, and then lyophilized.
Fluorescein isothiocyanate (FITC)-labelled lactobacilli and cell wall components were prepared using FITC isomer-I (Dojindo Lab., Kumamoto, Japan), as described previously.13
Preparation of cell wall components
Cell wall components were obtained from heat-killed lactobacilli, and the preparation steps are summarized in Fig. 1. ICW was obtained using the method of Sekine et al.14 with minor modifications.13 Briefly, heat-killed cells were boiled in 0·3% sodium dodecyl sulphate (SDS) solution for 15 min and washed with acetone. The cells were treated with pronase, and delipidated by successive refluxing with methanol, methanol–chloroform–water (1 : 1 : 1), and methanol–chloroform (1 : 1). The delipidated preparation was treated with benzon nuclease and pronase. The insoluble material was washed with distilled water and lyophilized; the resultant product was used as the ICW preparation. To obtain intact peptidoglycan and cell wall polysaccharides (PS), ICW was treated with 47% hydrogen fluoride at 4° for 20 hr, whereby the polysaccharide moiety was removed from the linkage region of ICW. The phosphodiester bonds were selectively cleaved under these conditions.15 After centrifugation, the precipitate was washed with distilled water and used as the intact peptidoglycan preparation. It was likely that the peptidoglycan preparation was free of contaminants such as lipoteichoic acid (LTA), wall teichoic acid and nucleic acid, as phosphorus was below the level of detection (< 10 nmol/mg) in the preparation by the method of Lowry et al.16 The supernatant was neutralized with 10 m NaOH, dialysed against distilled water, lyophilized, and used as the PS preparation. Soluble peptidoglycan (sPGN) was prepared by exhaustive digestion of intact peptidoglycan by N-acetylmuramidase treatment for 16 hr. The digested material solution was then treated with trypsin, dialysed against distilled water, lyophilized, and used as the sPGN preparation.
Isolation and culture of peritoneal macrophages
Peritoneal macrophages were isolated from BALB/c mice, unless otherwise specified. Mice were intraperitoneally injected with 2 ml of 4% thioglycollate broth (Difco). After 4 days, peritoneal macrophages were isolated as described previously.13 For the analysis of cytokine secretion, peritoneal macrophages (1 × 105 cells) were cultured in a 96-well culture plate (Nunc, Roskilde, Denmark) with heat-killed lactobacilli or their cell wall components (1–30 μg/ml) in the presence or absence of heat-killed L. casei (10 μg/ml) in 200 μl of RPMI-1640 medium. The supernatants of these cultures were collected at 24 hr to determine the levels of IL-12p70, tumour necrosis factor-α (TNF-α), and IL-10.
Analysis of IL-12 mRNA expression
Peritoneal macrophages (1·2 × 106 cells) were cultured in a 12-well culture plate (Nunc) with peptidoglycan (10 μg/ml) in the presence or absence of heat-killed L. casei (10 μg/ml) in 2·4 ml of RPMI-1640 medium. Total RNA was extracted from the cells at the indicated time-points (0–24 hr) using the RNAqueous kit (Ambion, Austin, TX), and reverse transcription was performed using the High-Capacity cDNA Reverse Transcription kit (Applied Biosystems, Foster City, CA). Quantitative real-time polymerase chain reaction (PCR) was performed using TaqMan Universal PCR master mix and TaqMan gene expression assays for IL-12p35 and IL-12p40 expression (Applied Biosystems) in an ABI 7500 Real Time PCR System (Applied Biosystems). The glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene was used as an endogenous control to normalize the expression of the IL-12 genes.
Light microscopic analysis of macrophages to observe intracellular digestion
Peritoneal macrophages (2 × 105 cells) were plated onto round, 12-mm collagen type I-coated cover glasses (Asahi Techno Glass, Tokyo, Japan) and cultured in a 24-well culture plate (Nunc) in 1·2 ml of RPMI-1640 medium. After preincubation of the culture for 16 hr, heat-killed lactobacilli were introduced at a concentration of 10 μg/ml. The macrophages were harvested after 24 hr, washed with phosphate-buffered saline (PBS), fixed with methanol for 10 min, and then stained with Giemsa (Sigma). Intracellular digestion of bacteria was observed under a BH-2 light microscope (Olympus, Tokyo, Japan).
Confocal laser scanning microscope analysis of macrophages to observe intracellular digestion
Peritoneal macrophages (2 × 105 cells) on round, 12-mm collagen type I-coated cover glasses were cultured with FITC-labelled lactobacilli or cell wall components (10 μg/ml) in a 24-well culture plate in 1·2 ml of RPMI-1640 medium. The macrophages were harvested after 24 hr, washed with PBS, and fixed with methanol. After the plasma membrane and the nucleus had been stained with phycoerythrin (PE)-conjugated rat anti-mouse CD11b monoclonal antibody (BD PharMingen, San Diego, MO) and TO-PRO-3 dye (Molecular Probes, Eugene, OR), respectively, the macrophages were observed and photographed with a LSM 510 confocal laser scanning microscope (Carl Zeiss, Oberkochen, Germany).
Flow cytometric analysis of phagocytosis
Peritoneal macrophages (6 × 105 cells) were cultured in a 24-well culture plate with FITC-labelled lactobacilli (10 μg/ml) in 1·2 ml of RPMI-1640 medium for 24 hr. The macrophages were dislodged by treating them with 10 mm ethylenediaminetetraacetic acid (EDTA)/PBS for 10 min, and they were then washed with 3 mm EDTA/PBS. Flow cytometric analysis was performed using an EPICS Altra flow cytometer equipped with expo32 software (Beckman Coulter, Miami, FL). In order to investigate whether peptidoglycan inhibits phagocytosis of L. casei by macrophages, macrophages were cultured with FITC-labelled L. casei (10 μg/ml) in the presence or absence of peptidoglycan (10 μg/ml) and cytochalasin D (5 μg/ml) for 16 hr, and analysed by flow cytometry.
The sensitivity of lactobacilli and their cell wall components to N-acetylmuramidase was evaluated on the basis of the decrease in their optical density at 600 nm (OD600), as described previously.13 In brief, heat-killed lactobacilli or their cell wall components suspended at a concentration of 2 mg/ml in 50 mm Tris–maleate buffer (pH 7·0) containing 4 mm MgCl2 were treated with N-acetylmuramidase (10 μg/ml) for 10, 30, 60 and 120 min. After heat-denaturation of the enzyme, the test materials were further treated with 2% SDS to dissolve the protoplasts formed as a result of cell wall digestion. The OD600 values were determined and expressed as a percentage of the OD600 values before enzyme treatment.
Enzyme-linked immunosorbent assay (ELISA) for detection of cytokines
A sandwich ELISA was used to determine the levels of IL-12p70, IL-10 and TNF-α in the supernatants of the cultures. Rat anti-mouse IL-12 (clone 9A5) and IL-10 (clone JES5-SXC1) monoclonal antibodies were used as the capture antibodies, and the corresponding detection antibodies for these were biotinylated rat anti-mouse IL-12 (clone C17.8) and IL-10 (clone JES5-2A5) monoclonal antibodies. These antibodies and recombinant mouse IL-12 and IL-10 were purchased from BD PharMingen. Mouse TNF-α Duoset (Genzyme, Cambridge, MA) was used to determine the TNF-α levels in the cultures.
Sensitivity of lactobacilli to digestion
The results of our previous study showed that the phagocytosis and subsequent digestion of lactobacilli by macrophages are closely related to IL-12 production and that the sensitivity of Lactobacillus strains to digestion with N-acetylmuramidase is negatively correlated to their ability to induce IL-12 production.13 Therefore, we initially tested the in vitro sensitivity of heat-killed L. casei, L. johnsonii and L. plantarum to N-acetylmuramidase. While L. casei was resistant to enzyme treatment, L. johnsonii and L. plantarum were sensitive (Fig. 2a). Digestion of these three strains by mouse peritoneal macrophages was also examined. Light microscopic analysis revealed that, while L. casei was hardly digested at all after 24 hr of macrophage culture, L. johnsonii was easily digested (Fig. 2b,c). Some of the L. plantarum cells that were phagocytosed by macrophages seemed to retain their cell morphology during the culture period (Fig. 2d). Confocal laser scanning microscope analysis using FITC-labelled lactobacilli clearly showed that L. casei was hardly digested at all by the macrophages, L. johnsonii was easily digested, and L. plantarum showed an intermediate pattern between the former two strains (Fig. 2e–g). Flow cytometric analysis confirmed that all these strains were phagocytosed well by macrophages (Fig. 2h–j).
IL-12 production by macrophages stimulated with a single Lactobacillus strain or a combination of two strains
Macrophages were cultured with heat-killed L. casei, L. johnsonii or L. plantarum for 24 hr, and the concentrations of IL-12p70, TNF-α and IL-10 in the supernatants were measured by ELISA. As shown in Fig. 3, L. casei potently induced IL-12 production in a dose-dependent manner at the tested doses (1–30 μg/ml), whereas L. plantarum weakly induced IL-12 production at an optimal dose of 3 μg/ml. Lactobacillus johnsonii hardly induced any IL-12 production. The induction pattern of TNF-α production was similar to that of IL-12, except that L.johnsonii induced similar levels of TNF-α to L. plantarum. In contrast, IL-10 production was strongly induced by L. plantarum and weakly induced by L. casei, especially at low doses. As was the case for IL-12 production, L. johnsonii hardly induced any IL-10 production.
Next, macrophages were cultured with these three strains along with 10 μg/ml heat-killed L. casei, and the effects of the strains on L. casei-induced cytokine production were examined. Lactobacillus casei-induced IL-12 production was suppressed in a dose-dependent manner by the addition of L. johnsonii and L. plantarum (Fig. 3). Lactobacillus casei-induced TNF-α production was inhibited only in the presence of L. plantarum, and the induction of IL-10 production was not inhibited by either of the strains.
When live lactobacilli were used in these experiments, L. casei was resistant but L. johnsonii and L. plantarum were sensitive to N-acetylmuramidase digestion, and only L. casei showed strong induction of IL-12 production; live L. johnsonii and L. plantarum, but not L. casei, inhibited L. casei-induced IL-12 production, as was observed in the case of the corresponding heat-killed cells (data not shown).
ICW is responsible for the ability of lactobacilli to induce and inhibit IL-12 production
We previously showed that the insoluble ICW of L. casei is responsible for IL-12 production.13 Therefore, we prepared ICWs from the three Lactobacillus strains and tested their ability to induce cytokine production; we also tested the ability of the ICWs to inhibit L. casei-induced cytokine production. Similar to the corresponding intact heat-killed cells, the ICWs of L. casei, L. johnsonii and L. plantarum induced the production of IL-12, TNF-α and IL-10 (Fig. 4). Furthermore, the ICWs of L. johnsonii and L. plantarum inhibited the L. casei-induced production of IL-12, as was observed in the case of the corresponding intact cells. The ICW of L. plantarum also inhibited L. casei-induced TNF-α production, as did intact L. plantarum cells. These data suggest that the ICW is responsible for the ability of these strains to induce and inhibit IL-12 production.
Intact peptidoglycan is sensitive to digestion
Our previous finding that the removal of the polysaccharide moiety from the ICW of L. casei resulted in the loss of its IL-12-inducing ability suggested that the polysaccharide-depleted ICW of L. casei might not be resistant to intracellular digestion.13 We prepared polysaccharide-depleted ICWs, i.e. intact peptidoglycans, from the ICWs of the three strains by hydrogen fluoride treatment, and we examined their sensitivity to N-acetylmuramidase treatment and macrophage intracellular digestion. The intact peptidoglycan as well as ICW retained the rod-like morphology before being phagocytosed by macrophages (insets in Fig. 5c,d). The intact peptidoglycan, but not the ICW, of L. casei was susceptible to N-acetylmuramidase treatment (Fig. 5a,b). The ICWs of L. johnsonii and L. plantarum were sensitive to N-acetylmuramidase treatment, and removal of the polysaccharide moiety seemed to increase their sensitivity to N-acetylmuramidase treatment. Moreover, the intact peptidoglycan, but not the ICW, of L. casei that was phagocytosed by the macrophages rapidly lost its morphology, suggesting that it might be susceptible to intracellular digestion (Fig. 5c,d).
Intact peptidoglycan inhibits Lactobacillus casei-induced IL-12 production
We evaluated the inhibitory effects of cell wall components obtained from L. casei on L. casei-induced IL-12 production. Only the intact peptidoglycan, not the ICW or the cell wall polysaccharides, had an inhibitory effect (Fig. 6a). The intact peptidoglycans of L. johnsonii, L. plantarum and S. aureus inhibited L. casei-induced IL-12 production but did not inhibit L. casei-induced TNF-α production; this result was different from that obtained for the intact cells and ICW of L. plantarum (Fig. 6b). The intact peptidoglycans of these strains inhibited the IL-12 production that was potently induced by other Lactobacillus strains having a rigid cell wall, such as L. rhamnosus (data not shown). We could not examine whether intact peptidoglycan inhibits IL-12 production induced by pathogenic stimuli, because any such stimuli tested (lipopolysaccharide, flagellin, heat-killed S. aureus and Escherichia coli) could not induce IL-12 production effectively in the peritoneal macrophage cultures used in this study. The solubilized form of the L. casei peptidoglycan did not effectively inhibit L. casei-induced IL-12 production, suggesting the importance of the recognition of intact peptidoglycan by the receptors on macrophages or of the steps involved in the phagocytosis of peptidoglycan (Fig. 6a).
Peptidoglycan inhibits IL-12 production by reducing the IL-12p40 mRNA expression
We investigated whether peptidoglycan inhibits the expression of IL-12p35 and IL-12p40 mRNAs. The expression of IL-12 mRNAs in macrophages cultured with L. casei cells or the peptidoglycans of L. casei and L. plantarum was analysed. Lactobacillus casei cells induced the expression of both IL-12p35 and IL-12p40 mRNAs in the macrophages, with peak expression at 16 hr of culture; however, the peptidoglycans of L. casei and L. plantarum induced IL-12p35 mRNA expression more effectively than did L. casei cells, with peak expression observed at 8 hr of culture, but the peptidoglycans hardly induced any IL-12p40 mRNA expression (Fig. 7a,b). When macrophages were cultured for 8 and 16 hr with a combination of L. casei and the peptidoglycans of L. casei and L. plantarum, the peptidoglycans markedly inhibited the L. casei-induced expression of IL-12p40 but not IL-12p35 mRNA (Fig. 7c,d). These data suggest that the inhibition of IL-12 production by the peptidoglycans results from a reduction in IL-12p40 mRNA expression.
Peptidoglycan inhibits IL-12 production by neither suppressing phagocytosis nor inducing IL-10 production
We determined whether peptidoglycan inhibits the phagocytosis of L.casei by macrophages because we had previously observed that the phagocytosis of L.casei is essential for it to induce IL-12 production.13 Macrophages were cultured with FITC-labelled L. casei for 16 hr with or without unlabelled peptidoglycans or cytochalasin D, an inhibitor of actin polymerization; phagocytosis was then analysed by flow cytometry. We observed that the peptidoglycans of L. casei and L. plantarum did not interfere with the phagocytosis of L. casei, whereas cytochalasin D markedly suppressed it (Fig. 8a).
It is well known that IL-10 inhibits the secretion of IL-12 by macrophages.5 Therefore, we used a neutralization antibody against IL-10 in order to determine whether IL-10 mediated the inhibition of IL-12 production by peptidoglycan. Macrophages were cultured with a combination of L. casei and cell wall components of L. casei and L. plantarum in the presence of the anti-IL-10 or control antibody. As shown in Fig. 8b, the ICW of L. plantarum induced high amounts of IL-10 and inhibited L. casei-induced IL-12 production in the absence of the anti-IL-10 antibody. Addition of the anti-IL-10 antibody abrogated the inhibitory effect of the L. plantarum ICW on IL-12 production, suggesting that IL-10 mediated the inhibition of IL-12 production by L. plantarum ICW. Moreover, the ICW of L. plantarum could effectively induce IL-12 production in the presence of the anti-IL-10 antibody. In contrast, the inhibition of IL-12 production by the peptidoglycans of L. casei and L. plantarum was not affected by the addition of the anti-IL-10 antibody, suggesting that peptidoglycan inhibits L. casei-induced IL-12 production through IL-10-independent mechanisms.
TLR2 is involved in the inhibition of IL-12 production by peptidoglycan
Using macrophages isolated from TLR2-deficient mice, we examined whether the recognition of peptidoglycan by TLR2 is involved in the inhibition of IL-12 production. IL-12 production by both TLR2-deficient and wild-type macrophages was induced by stimulation with L. casei, suggesting that the recognition of L. casei cell components by TLR2 is not essential for IL-12 production (Fig. 9), as has previously been reported by us.13 However, the peptidoglycans of L. casei and L. plantarum inhibited L. casei-induced IL-12 production by the TLR2-deficient macrophages only by 12 and 48%, respectively, whereas these values were 61 and 97%, respectively, in the case of the wild-type macrophages. Lipopolysaccharide inhibited IL-12 production by the wild-type and TLR2-deficient macrophages. These results suggest that the TLR2-mediated recognition of peptidoglycan plays an important role in the inhibition of IL-12 production, and we assume that some other mechanisms may also be involved.
Digestion products of peptidoglycan inhibit IL-12 production in a TLR2-independent manner
Watanabe et al.10 reported that MDP, a product of peptidoglycan digestion, was recognized by the NOD2 receptor, and subsequently acted as a negative regulator of IL-12 production. Therefore, the effects of MDP and L18-MDP on IL-12 production were examined. L18-MDP is a stearoyl fatty acid derivative of MDP and is more effectively internalized into cells. We observed that L18-MDP but not MDP inhibited L. casei-induced IL-12 production, although the inhibitory effect was slightly weaker than that of L. casei peptidoglycan (Fig. 10a). C12-DAP, a derivative of γ-d-glutamyl-meso-diaminopimelic acid, is a ligand for NOD1, and it did not inhibit IL-12 production. The inhibitory effect of L18-MDP on IL-12 production was observed in TLR2-deficient macrophages as well as wild-type macrophages (Fig. 10b), suggesting that the digestion products of peptidoglycan can inhibit IL-12 production in a TLR2-independent manner, probably through NOD2 receptor recognition.
We previously reported that the ICW structure of certain Lactobacillus strains is essential for inducing IL-12 production by macrophages, and Lactobacillus strains that have a rigid cell wall are resistant to intracellular digestion by macrophages and can effectively stimulate them to secrete IL-12.13 The results of the present study revealed that, when two types of lactobacilli (sensitive to intracellular digestion by macrophages and resistant to intracellular digestion) are used to simultaneously stimulate macrophages, the sensitive Lactobacillus strains inhibit the IL-12 production induced potently by the resistant strains. The findings demonstrate that the sensitive strains not only rapidly lose the cell wall structure responsible for their ability to induce IL-12 production but also have the ability to actively inhibit IL-12 production. Further, our results revealed that the peptidoglycan from lactobacilli was responsible for the inhibition of IL-12 production.
Peptidoglycan is a characteristic cell wall component of Gram-positive bacteria and can stimulate macrophages and dendritic cells to secrete proinflammatory cytokines.17,18 In our study, we found that peptidoglycan potently induced TNF-α production in macrophages at similar levels to those induced by L. casei. In addition, it induced IL-12 production, albeit weakly, and also induced the expression of IL-12p35 mRNA more potently than did L. casei (Figs 6b and 7a). However, peptidoglycan could inhibit the IL-12p40 mRNA expression potently induced by the resistant Lactobacillus strains, thereby inhibiting bioactive IL-12p70 secretion. This novel observation suggests that peptidoglycan can act in both proinflammatory and anti-inflammatory ways. Dysregulated overproduction of IL-12 and the subsequent overactivation of T helper (Th)1 cells are considered to be one of the causes of autoimmune diseases and inflammatory bowel diseases.6,19 Recently, it has been reported that IL-17-producing Th17 cells also play critical roles in the development of inflammation in such diseases.7,20 IL-12p40 is a subunit of IL-23, which is an important cytokine for the proliferation and maintenance of Th17 cells. Therefore, peptidoglycan may aid in the elimination of these inflammatory diseases by suppressing the production of both IL-12 and IL-23. Further studies are required to assess the possible anti-inflammatory activity of peptidoglycan.
The cell walls of L. casei as well as L. johnsonii are composed of peptidoglycan and associated uncharged polysaccharides, whereas the cell wall of L. plantarum is composed of peptidoglycan and associated anionic polysaccharides, i.e. wall teichoic acids.21 Although the ICW of L. casei was resistant to N-acetylmuramidase treatment, polysaccharide-depleted ICW, i.e. intact peptidoglycan, was sensitive to the enzyme treatment. The ICWs of L. johnsonii and L. plantarum were sensitive to the enzyme treatment, and their peptidoglycans were more sensitive than the corresponding ICWs (Fig. 4). These results suggest that the peptidoglycan-associated polysaccharide moiety plays a role in protecting peptidoglycan from enzyme digestion by sterically hindering access of the enzyme to the corresponding digestion sites in peptidoglycan.22,23 Furthermore, the ICWs of these three strains showed different abilities to induce cytokines in macrophages and to inhibit L. casei-induced IL-12 production (Fig. 4), while the peptidoglycans of all these strains similarly induced very low levels of IL-12 and inhibited L. casei-induced IL-12 production (Fig. 6). The findings suggest that differences in the characteristics of the peptidoglycan-associated polysaccharide moiety, such as chemical composition, physical length and density, and electrostatic properties, are important factors determining the abilities of Lactobacillus strains to regulate cytokine production.
While the peptidoglycans of these three strains and the ICWs of L. johnsonii and L. plantarum were sensitive to intracellular digestion and inhibited IL-12 production, the ICW of L. casei was resistant to intracellular digestion and did not inhibit IL-12 production. On the basis of these results, we proposed the following hypothesis: the products of peptidoglycan digestion, such as MDP, are responsible for the inhibitory activity of peptidoglycan. Unexpectedly, the inhibitory activity of peptidoglycan was very low in TLR2-deficient macrophages, although TLR2 cannot recognize digests of peptidoglycan;24 this suggests that the recognition of undigested peptidoglycan by TLR2 plays an important role in the inhibition of IL-12 production. This apparent contradiction may be explained by the possibility that the accessibility of TLR2 to peptidoglycan was merely correlated to the accessibility of cell wall-digesting enzymes to peptidoglycan, and that the important component for the inhibition of IL-12 production is actually undigested peptidoglycan and not the products of peptidoglycan digestion. In other words, although the recognition of peptidoglycan by TLR2 might be sterically hindered by the polysaccharide moiety in the case of the ICW of L. casei, the absence of the polysaccharide moiety might allow TLR2 to recognize peptidoglycan and simultaneously change the sensitivity of the L. casei ICW to enzymatic digestion.
Recently, Travassos et al.25 reported that highly purified peptidoglycan dose not seem to signal via TLR2. They suggested that the TLR2 stimulatory activity of crude peptidoglycan preparations obtained from Gram-positive bacteria is likely to be mediated by contaminating LTA. In contrast, Dziarski and Gupta24 re-evaluated activation of TLR2 by peptidoglycan using highly purified peptidoglycan, and showed that peptidoglycan actually stimulates TLR2. In the present study, we obtained highly purified intact peptidoglycan with no detectable contaminating LTA by treatment of heat-killed lactobacilli with detergent, organic sorbent, nuclease, pronase, and hydrogen fluoride, which is similar to the peptidoglycan purification method of Travassos et al.25 The TLR2-mediated ability of the intact peptidoglycan to inhibit L. casei-induced IL-12 production was abolished by treatment with N-acetylmuramidase, which digests peptidoglycan but not possible traces of LTA and other contaminants in the peptidoglycan preparation (Fig. 6a). The results support the idea that the intact peptidoglycan of lactobacilli actually signals via TLR2 to inhibit IL-12 production, although they do not completely exclude the possibility that contaminants in the peptidoglycan preparation may be responsible for the inhibitory effect on IL-12 production.
Although the recognition of peptidoglycan by TLR2 was essential for the inhibition of IL-12 production, 12–48% of the IL-12 production in TLR2-deficient macrophages was inhibited by peptidoglycan, suggesting that other TLR2-independent mechanisms might also be involved. Watanabe et al.10 reported that MDP inhibited IL-12 production after it was recognized by the intracellular receptor NOD2. We also observed that L18-MDP, a derivative of MDP, could inhibit IL-12 production. MDP might not be inhibitory because of the low efficacy of its internalization into peritoneal macrophages, which we used in our study. Collectively, these results suggest that peptidoglycan can inhibit IL-12 production by the recognition of both its undigested form via TLR2 and its digested form (MDP) via NOD2. We obtained the preliminary finding that peptidoglycan enhanced the mRNA expression of TLR2 and NOD2 in macrophages (data not shown). The regulation of these receptors by their ligand will be an interesting issue to be examined.
The inhibition of IL-12 production by peptidoglycan was mediated by neither the inhibition of phagocytosis nor the induction of the suppressive cytokine IL-10. Although we showed that IL-12 production was inhibited as a result of the reduction in IL-12p40 mRNA expression, the details of the underlying mechanisms remain to be clarified. Recently, Kuwata et al.26 demonstrated that the nuclear transcription factor IκBNS inhibits IL-12p40 mRNA expression, and it was reported to be highly expressed in the macrophages of the colonic lamina propria, which hardly produced any IL-12p40 in response to bacterial components.27 We observed strong expression of IκBNS mRNA in peritoneal macrophages after stimulation with the peptidoglycan of L. casei; however, this was not observed after stimulation with its ICW (K. Shida, J. Kiyoshima-Shibata and M. Nanno, unpublished data). Peptidoglycan might suppress IL-12 production by inducing such inhibitory transcription factors.
It is well known that peptidoglycan, LTA, lipoproteins and genomic DNA from Gram-positive bacteria, including lactobacilli, stimulate TLR2 and TLR9, leading to the secretion of proinflammatory cytokines.28 However, other published reports have revealed that TLR2- or TLR9-deficient as well as wild-type macrophages respond to certain Lactobacillus strains that belong to the L. casei group and secrete IL-12.13,29 In this study, we reconfirmed that L. casei induced a high level of IL-12 production in both wild-type and TLR2-deficient macrophages, while peptidoglycan induced a low level of IL-12 production in wild-type macrophages and even lower levels in TLR2-deficient macrophages (Fig. 9). TLR2 may be responsible for the lower level of IL-12 production induced by peptidoglycan or Lactobacillus strains sensitive to intracellular digestion, while some unidentified receptors other than TLRs may play an important role in recognizing the three-dimensional structure of the cell wall of Lactobacillus strains resistant to intracellular digestion, leading to the production of a higher amount of IL-12. Receptors that are strongly associated with the phagocytosis of bacteria, such as C-type lectin receptors, might be candidate receptors of the second type.30
Zeuthen et al.12 reported that Lactobacillus reuteri and Bifidobacterium bifidum, which are weak inducers of IL-12, inhibited the Lactobacillus acidophilus-induced potent production of IL-12 by human dendritic cells in an IL-10-independent manner. They further revealed that the B. bifidum-induced inhibition of IL-12 production was mediated by the recognition of lipoproteins and partially ruptured insoluble cell walls by TLR2 on murine dendritic cells, and they suggested that TLR2-mediated signals could act as negative regulators of proinflammatory responses.31 Our findings showed that L. johnsonii and L. plantarum, which were weak inducers of IL-12, inhibited IL-12 production by involving the TLR2-mediated signalling pathway and they agree well with the results of Zeuthen et al. However, our results showed that peptidoglycan was responsible for the inhibitory effect of L. johnsonii and L. plantarum, mainly through its recognition by TLR2; in contrast, Zeuthen et al. proposed the possibility that peptidoglycan promotes IL-12 production via the recognition of its digestion products by NOD2. These contradictory findings may be explained by the different responses elicited by dendritic cells and macrophages, as other researchers have also revealed a difference in responses of these cells: NOD2 mediated synergistic induction of IL-12 production in dendritic cells, while it was found to mediate the inhibition of IL-12 production in macrophages.9,32 The cross-regulatory effect of certain probiotics has also been observed in an in vivo study, wherein it was found that the anti-allergic effect of a mixture of L. rhamnosus GG and three other probiotic strains was weaker than that elicited by the L. rhamnosus GG strain alone in the treatment of infant atopic dermatitis.33 These findings suggest that these cross-regulatory effects should be considered when developing novel probiotic mixtures.
Probiotics having a potent IL-12-inducing ability are expected to exhibit suppressive effects on infections and cancers, as IL-12 induces a Th1-dominant immune response and augments the innate immune response. However, some researchers are afraid that certain probiotics, which induce excessive production of IL-12, may increase the risk of inflammatory diseases.34 Our present finding that easily digestible lactobacilli and peptidoglycan inhibit the IL-12 production potently induced by a probiotic L. casei strain supports the idea that, when orally administered, such probiotics may stimulate mucosal macrophages simultaneously with intestinal commensal lactobacilli or their peptidoglycan to produce suitable levels of IL-12, and thereby have beneficial rather than adverse effects on the host.
Intestinal macrophages and dendritic cells have been reported to secrete only low levels of proinflammatory cytokines such as IL-12 in response to bacterial stimuli, and this may contribute to the maintenance of intestinal homeostasis. The suppressed responses of intestinal macrophages and dendritic cells are thought to be mediated by various mechanisms, such as the constitutive expression of inhibitory nuclear transcription factors in these cells and the supply of suppressive mediators by intestinal epithelial cells.27,35,36 Our study showed that peptidoglycans from both S. aureus and lactobacilli had an inhibitory effect on IL-12 production, suggesting that peptidoglycans from many bacteria in addition to lactobacilli might inhibit the IL-12 production potently induced by certain other bacteria. The intestine hosts various commensal bacteria, and some of them may be able to inhibit IL-12 production through recognition of peptidoglycan and its digestion products via TLR2 and NOD2. This inhibitory mechanism may also contribute to the maintenance of intestinal homeostasis.
We are grateful to Dr Satoshi Matsumoto (Yakult Central Institute) for valuable discussions. We also gratefully acknowledge the staff of the animal facility of the Yakult Central Institute for their expertise in breeding mice.