INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Lipopolysaccharides (LPS), a major component of the outer membrane of gram-negative bacteria, is a highly potent effector of immune responses for many immunocompetent cells (15). Monocytes/macrophages, known to be primary targets for LPS, produce proinflammatory cytokines such as tumor necrosis factor alpha (TNF-), interleukin-1 (IL-1), IL-6, and IL-8 upon activation (25). Anti-inflammatory cytokines such as IL-10 (14) or IL-1 receptor antagonist (IL-1ra) (1) are also up-regulated with delayed kinetics in monocytes after LPS stimulation and are presumably important for down-regulating the inflammation.

Repeated administration of LPS in vivo induces a refractoriness to its pathophysiological effects, known as LPS tolerance or desensitization (12, 35). This phenomenon is controlled at the monocyte/macrophage level and is associated with down-regulation of cytokine production upon secondary LPS stimulation (12). In this respect, in vitro incubation of monocytic cells with low-dose LPS also renders the cells refractory to subsequent optimal LPS stimulation in terms of production of proinflammatory cytokines such as TNF-, IL-1, and IL-6 (34); however, the mechanism of this process is not clear. It has been suggested that LPS tolerance does not always lead to a down-regulation of all cell functions, and it is not necessarily a passive process that occurs in exhausted cells (35). In this respect, LPS-tolerant cells in vitro produce IL-1ra (22) and granulocyte colony-stimulating factor (12) in response to LPS secondary stimulation in amounts similar to or even higher than those produced by naive cells. Recently, it was reported that IL-10 was also up-regulated in LPS-tolerant Mono Mac 6 cells (8). Furthermore, Morrison's group reported a biphasic enhancement/suppression of TNF- responsiveness compared with a reciprocal suppression/enhancement of NO secretion in the various threshold doses of LPS-pretreated mouse macrophages, a phenomenon they termed LPS reprogramming (9, 33).

Porphyromonas gingivalis, a gram-negative black-pigmented anaerobic rod, is suspected to be one of the major periodontopathic organisms in chronic inflammatory periodontal disease (30). The chemical and biological properties of P. gingivalis LPS differ from those of the classical enterobacterial LPS, and its endotoxicity is much less than that of enterobacterial preparations (17, 19). Furthermore, P. gingivalis LPS and its lipid A, the endotoxic and bioactive center of LPS, induce very weak production of proinflammatory cytokines such as IL-1 and TNF- in human peripheral blood monocytes compared to Escherichia coli LPS and its synthetic lipid A (compound 506) (21). On the contrary, P. gingivalis LPS and its lipid A induce greater or almost comparable production of IL-1ra, IL-6, and IL-8 (20, 21). However, the effect of low-dose P. gingivalis LPS on cytokine production by monocytes/macrophages has not been defined.

This study was undertaken to assess the production of IL-6, IL-8 and IL-10 by human peripheral blood monocytes pretreated with a low dose of P. gingivalis LPS in comparison with those from naive cells and from E. coli LPS-pretreated cells. We further investigated the association of IL-10 in the establishment of P. gingivalis LPS-induced tolerance of human monocytes.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

LPS preparations and reagents. P. gingivalis 381 was anaerobically grown in GAM broth (Nissui, Tokyo, Japan) supplemented with hemin (5 mg/ml; Wako Pure Chemicals, Osaka, Japan) and menadione (10 µg/ml; Wako) for 26 h at 37°C. Bacterial cells were collected by centrifugation, washed three times with pyrogen-free water, and lyophilized. LPS was extracted from lyophilized cells by the hot phenol-water method (31), and the crude extract was purified by repeated ultracentrifugation (100,000 × g, 3 h) followed by treatment with nuclease P₁ (Yamasa Shoyu Co., Choshi, Japan) and finally lyophilized. E. coli O55:B5 LPS was purchased from List Biological Laboratories (Campbell, Calif.). Human recombinant IL-10 (rIL-10) was obtained from PeproTech Inc. (Rocky Hill, N.J.), and rabbit anti-human IL-10 polyclonal antibody was purchased from Chemicon International Inc. (Temecula, Calif.).

Preparation and culture of peripheral blood monocytes. Peripheral blood mononuclear cells (PBMC) were prepared by Histopaque (Sigma Chemical Co., St. Louis, Mo.) centrifugation of heparinized venous blood from healthy adult donors (4). PBMC (2.5 × 10⁵/well) in RPMI 1640 medium (Nikken Biomedical Laboratory, Suita, Japan) supplemented with 10% fetal bovine serum (HyClone Laboratories Inc., Logan, Utah) was incubated for 2 h at 37°C in humidified air containing 5% CO₂ in a 96-well flat-bottom microtiter plate (Corning Glassware, Corning, N.Y.), followed by the removal of nonadherent cells. Under these conditions, adherent cells contained >90% monocytes as assessed by morphological analysis using phase-contrast microscopy and flow cytometric analysis. The adherent monocytes were pretreated with or without various concentrations of P. gingivalis or E. coli LPS for various periods, and a secondary stimulation with 1 µg of LPS per ml was performed after the cells were washed. Culture supernatants were collected and stored at 80°C until used for the assays.

Cytokine assays. Cytokine levels were measured by an enzyme-linked immunosorbent assay (ELISA) for secreted IL-6 (DuoSet; Genzyme Diagnostics, Cambridge, Mass.), IL-8 (DuoSet; Genzyme Diagnostics), and IL-10 (Biotrak; Amersham Life Science, Buckinghamshire, England). The assay was performed according to the manufacturer's instructions, and the IL-6 and IL-8 ELISA systems were developed by a one-step tetramethylbenzidine-hydrogen peroxide method (Sigma). The data were determined by using a standard curve prepared for each assay.

RT-PCR analysis. A semiquantitative reverse transcription-PCR (RT-PCR) amplification protocol (24) was performed to obtain IL-6 and IL-8 mRNAs. Adherent monocytes were cultured in 10-cm-diameter petri dishes as described above, and the stimuli were added to the culture in a volume of 10 ml. Supernatants were removed at specified time points, and total RNA was extracted from the cells by using a single-step extraction procedure with RNAzol B (Cinna/Biotecx Laboratories, Houston, Tex.) (6). Reverse transcription of RNA was carried out in 40 µl of 50 mM Tris-HCl buffer (pH 8.3) containing 3 mM MgCl₂, 75 mM KCl, 10 mM dithiothreitol, deoxynucleoside triphosphate mixture (dATP, dCTP, and dGTP, 0.5 mM each; Takara Shuzo, Ohtsu, Japan), RNase inhibitor (1.4 U/µl; Life Technologies, Gaithersburg, Md.), Moloney murine leukemia virus reverse transcriptase (5 U/µl), 2 ml of random hexamers [pd(N)6; 10³ U/µl; Pharmacia Biotech, Milwaukee, Wis.], and 1 µg of RNA sample. Each sample tube was placed in a thermal cycler (GeneAmp PCR System 9600; Perkin-Elmer Co., Norwalk, Conn.), incubated for 60 min at 37°C followed by 5 min at 99°C, and then kept at 5°C. Primer sequences were 5'-TTCTCCACAAGCGCCTTCGGT-3' (sense) and 5'-TAGATTCTTTGCCTTTTTCTG-3'), (antisense) for IL-6 and 5'-ATGACTTCCAAGCTGGCCCTGGCT-3' (sense) and 5'-TCTCAGCCCTCTTCTTCAAAAACTTCTC-3') (antisense) for IL-8. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) amplification was performed in parallel as a positive control of the efficiency of the reaction, using primers 5'-TGAAGGTCGGAGTCAACGGATTTGGT-3' (sense) and 5'-CATGTGGGCCATGAGGTCCACCAC-3') (antisense). IL-6 and IL-8 primers were selected based on published sequences (16, 32) and were prepared by Iwaki Glasswares (Funahashi, Japan). GAPDH primers were purchased from Clontech Laboratories Inc. (Palo Alto, Calif.). Each pair of primers was designed to anneal with sequences in distinct exons, thereby ensuring amplification of only the appropriate mRNA, not the genomic DNA, sequences. For each PCR amplification, 10 µl was electrophoresed on a 1.5% agarose gel in Tris-acetate-EDTA buffer and visualized by ethidium bromide staining.

Statistical analysis. Data are presented as mean ± standard error (SE). Student's t test was used for paired comparison, and P values less than 0.05 were considered significant.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Cytokine production from monocytes pretreated with a low dose of P. gingivalis LPS. Monocytes were preincubated for 12 h in RPMI 1640 medium alone or with various doses of P. gingivalis or E. coli LPS (0.001 to 100 ng/ml), washed, and restimulated with 1 µg of the same LPS per ml. Supernatants were collected 36 h after restimulation and assayed for IL-6 and IL-8 production. As shown in Fig. 1, relatively low doses (~0.1 ng/ml) of P. gingivalis LPS caused a significant reduction in IL-6 production, up to 70% (at 0.1 ng/ml), compared with that of non-LPS-pretreated control, whereas the higher pretreatment dose (100 ng/ml) of LPS reversed this decrease. IL-8 production was not affected by the LPS dose used for pretreatment (0.001 to 100 ng/ml). On the other hand, low doses (0.01 to 0.1 ng/ml) of E. coli LPS pretreatment showed a priming effect on both IL-6 and IL-8 production. These results showed that equivalent conditions of pretreatment with P. gingivalis and E. coli LPS led to reciprocal enhancement/suppression patterns in terms of IL-6 production.

View larger version (15K): [in this window] [in a new window]   FIG. 1.   Modulation of IL-6 (A) and IL-8 (B) responses in P. gingivalis or E. coli LPS-pretreated human monocytes. Adherent monocytes were obtained by culturing PBMC (5 × 105/well) for 2 h in 96-well flat-bottom plates. Adherent cells were pretreated in the presence or absence of various doses of P. gingivalis or E. coli LPS for 12 h. After extensive washing, cells were restimulated with 1 µg of P. gingivalis () or E. coli () LPS per well for a further 36 h. Supernatants were harvested, and cytokine levels were determined by ELISA. The data represent the mean (±SE) of triplicate measurements from one representative experiment of three performed. **, P < 0.01 compared with cells pretreated in the absence of P. gingivalis LPS.

View larger version (15K): [in this window] [in a new window]   FIG. 2.   Time course of P. gingivalis LPS pretreatment effect on IL-6 (A) and IL-8 (B) production from monocytes after restimulation with LPS. Adherent monocytes were pretreated with 0.1 ng of P. gingivalis LPS per ml for 0 to 24 h. After extensive washing at the times indicated, cells were restimulated with 1 µg of P. gingivalis () or E. coli () LPS for a further 36 h. Supernatants were harvested, and cytokine levels were determined by ELISA. Control cultures were pretreated with medium alone for the indicated times and restimulated with LPS for a further 36 h. All data are expressed as percent production of cytokines in supernatants from control cultures. * P < 0.05; **, P < 0.01 compared with cells pretreated in the absence of LPS.

Cytokine mRNA expression in P. gingivalis LPS-pretreated monocytes. Monocytes preincubated for 12 h with 0.1 ng of P. gingivalis LPS per ml were subsequently stimulated for 2 h (Fig. 3A) or 10 h (Fig. 3B) with 1 µg of LPS per ml. At 2 h after restimulation, IL-6 and IL-8 mRNA was not lower in LPS-treated cells than in non-LPS-pretreated and stimulated cells. Furthermore, IL-8 mRNA expression was detected in non-LPS-pretreated cells without secondary stimulation.

View larger version (47K): [in this window] [in a new window]   FIG. 3.   Cytokine mRNA expression in P. gingivalis LPS-pretreated monocytes. Adherent monocytes were obtained by culturing PBMC (5 × 107/dish) for 2 h in petri dishes. Adherent cells were pretreated with 0.1 ng of P. gingivalis LPS or medium alone per ml for 12 h. After extensive washing, cells were restimulated with or without 1 µg of LPS per ml. RNA was extracted at 2 h (A) and 10 h (B) after the secondary stimulation and subjected to RT-PCR analysis. The data shown have been normalized to GAPDH gene expression and are derived from a single experiment that is representative of three independent experiments performed.

IL-10 production from P. gingivalis LPS-pretreated human monocytes. Anti-inflammatory cytokine IL-10 production from low-dose P. gingivalis LPS-pretreated monocytes was investigated. As shown in Fig. 4, the presence of a low dose (0.1 ng/ml) of P. gingivalis LPS during primary culture induced an early up-regulation of IL-10 production from monocytes after secondary stimulation (10 h) compared to non-LPS-pretreated cells. In contrast, E. coli LPS pretreatment at the same dose resulted in the down-regulation of IL-10 production upon restimulation with the same LPS. Also, we detected no enhanced IL-10 production without an optimal LPS restimulation.

View larger version (16K): [in this window] [in a new window]   FIG. 4.   Effect of low-dose P. gingivalis or E. coli LPS pretreatment on LPS-stimulated IL-10 production by human monocytes. Adherent monocytes were pretreated in the presence or absence of a low dose of P. gingivalis or E. coli LPS (0.1 ng/ml) for 12 h and then restimulated with 1 µg of P. gingivalis LPS (), E. coli LPS (), or medium alone () per ml for a further 10 h. Supernatants were harvested, and cytokine levels were determined by ELISA. The data represent the mean (±SE) of triplicate measurements from one representative experiment of three performed. *, P < 0.05; **, P < 0.01 compared with cells pretreated in the absence of LPS.

Prevention of IL-6 down-regulation in P. gingivalis LPS-pretreated cells by neutralization of endogenous IL-10. Since anti-inflammatory IL-10 production was up-regulated in LPS-pretreated monocytes 10 h after the secondary stimulation, we wondered whether IL-10 produced during the secondary culture may be involved in a down-regulation of IL-6 production from LPS-pretreated monocytes. Blocking of IL-10 by a neutralizing polyclonal IL-10 antibody during the secondary culture led to a reversal of IL-6 down-regulation observed in P. gingivalis LPS-pretreated monocytes (Fig. 5A). However, the same IL-10 neutralization did not affect IL-8 production by LPS-pretreated cells (Fig. 5B).

View larger version (31K): [in this window] [in a new window]   FIG. 5.   Neutralizing effect of anti-IL-10 antibody on IL-6 (A) and IL-8 (B) production from P. gingivalis LPS-pretreated monocytes. Adherent monocytes were pretreated with or without 0.1 ng of P. gingivalis LPS per ml for 12 h and then restimulated with 1 µg of P. gingivalis LPS (), E. coli LPS (), or medium alone () per ml for a further 36 h in the presence or absence of anti-IL-10 antibody (Ab; 10 µg/ml). Cytokine levels were determined by ELISA, and the data represent the mean (±SE) of triplicate measurements from one representative experiment of three performed.

Effect of exogenous IL-10 on production of IL-6 and IL-8 by monocytes. We tested whether addition of exogenous IL-10 to non-LPS-pretreated cell culture could lead to the suppression of IL-6 production after secondary stimulation. As shown in Fig. 6A, human rIL-10 dose dependently inhibited IL-6 synthesis from non-LPS-pretreated monocytes after restimulation with LPS, and the coexisting polyclonal anti-IL-10 antibody partially blocked the IL-10-induced down-regulation of IL-6. Blocking of IL-8 synthesis was observed only when 100 ng of IL-10 per ml was added to the naive cell culture (Fig. 6B).

View larger version (35K): [in this window] [in a new window]   FIG. 6.   Effect of exogenous IL-10 on IL-6 (A) and IL-8 (B) production from nontreated monocytes. Adherent monocytes were prepared from PBMC and precultured in medium alone for 12 h. After extensive washing, cells were stimulated with 1 µg of P. gingivalis or E. coli LPS per ml for 36 h in the presence of various concentrations of human rIL-10. Anti-IL-10 antibody (Ab; 10 µg/ml) was added to the culture simultaneously with IL-10. Supernatants were harvested, and cytokine levels were determined by ELISA. *, P < 0.05; **, P < 0.01 compared with cells in the absence of human rIL-10.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

In this study, we found that preincubation of human peripheral blood monocytes with 0.1 ng of P. gingivalis LPS per ml renders the cells refractory to a subsequent optimal LPS challenge in terms of IL-6 but not IL-8 production. The equivalent pretreatment with E. coli LPS reciprocally primed the cells for production of both IL-6 and IL-8. Thus, IL-6 down-regulation seems to be a specific phenomenon in P. gingivalis LPS-pretreated cells. Previous studies reported that exposure of human monocytic cells to classical enterobacterial LPS in vitro, even at a very low dose, suppressed the responses of proinflammatory cytokines such as TNF-, IL-1, and IL-6 to a second high-dose LPS challenge (1, 12). However, low-dose LPS exposure does not always lead to suppression of all cytokine production upon subsequent stimulation with an optimal dose. In this respect, Mengozzi et al. (13) reported that LPS-pretreated monocytes still produced IL-8, IL-1, and IL-6, but not TNF-, after restimulation with LPS. Furthermore, Hirohashi and Morrison (9) reported that substimulatory-dose LPS pretreatment could effectively reprogram macrophages to induce biphasic and reciprocal dose-dependent enhancement and inhibition of TNF-, IL-6, and NO. Our data are consistent with these results and suggest that different regulatory mechanisms are involved in IL-6 and IL-8 production in P. gingivalis LPS-pretreated monocytes.

Pretreatment with low-dose E. coli LPS induced a pattern of IL-6 production reciprocal to that of P. gingivalis LPS-pretreated cells and also augmented IL-8 production. P. gingivalis lipid A differs from the E. coli-type lipid A in chemical structure, and it induced very weak production of IL-1 and TNF- but higher or almost comparable production of both IL-6 and IL-8 (18). Partial structures of LPS with reduced toxicity, including diphosphoryl lipid A from Rhodobacter sphaeroides (5), monophosphoryl lipid A (2), and the monophosphoryl 3-acyl compound SDZMRL 953 (10), were previously tested for the capacity to induce tolerance. These studies revealed that these LPS derivatives with low endotoxicity were potent inducers of LPS tolerance, although a substantially higher dose was required for an effect equivalent to that of LPS. It was also reported that tolerance was induced only by compounds of diphosphoryl lipid A analogues that were capable of stimulating cells (11). Most likely, our results suggest that the differences in biological activities and chemical structure are important to determine whether LPS preexposure preferentially induces a refractoriness or a hypersensitivity to itself. Therefore, these results suggest that cellular activation and tolerance induction are closely associated events, and LPS or its derivatives with a variety of structure and biological activities lead cells to different degrees of activation which are critical for the cellular response upon subsequent optimal LPS stimulation.

The down-regulation of IL-6 production in P. gingivalis LPS-tolerant human monocytes is dependent on the time of LPS pretreatment (Fig. 2). However, the ability of monocytes to produce IL-8 was not significantly decreased even by prolonged (24-h) preexposure. These results again indicated that preexposure to P. gingivalis LPS did not cause a state of exhaustion in the monocytes and suggest that independent regulatory mechanisms are involved in controlling proinflammatory IL-6 and IL-8 production in P. gingivalis LPS-pretreated monocytes. To speculate how IL-6 down-regulation occurred in our experimental model, we investigated the kinetics of the IL-6 message (Fig. 3). RT-PCR analysis revealed that IL-6 mRNA expression was down-regulated in P. gingivalis LPS-pretreated cells 10 h after a secondary challenge with the same LPS, whereas no difference from nontreated cells was observed at 2 h. These results suggested that down-regulation of IL-6 was established during the secondary stimulation in P. gingivalis LPS-tolerant cells.

Previous studies suggested an involvement of regulatory cytokines in the establishment of LPS tolerance. Among these regulatory cytokines, IL-10 is an essential endogenous mediator that induces LPS tolerance (22, 23, 27). We speculate that enhanced IL-10 production may occur to down-regulate IL-6 secretion in our experimental model. A significantly elevated level of IL-10 from LPS-pretreated cells was seen at 10 h after restimulation with an optimal dose of P. gingivalis LPS (Fig. 4). Contrasting results were reported in terms of IL-10 production from LPS-tolerant cells. Frankenberger et al. reported the up-regulation of IL-10 production from LPS-toleralized Mono Mac 6 cells (8). However, Randow et al. reported the down-regulation of IL-10 production in LPS-toleralized cells, although they indicated an importance of IL-10 in primary culture for induction of LPS tolerance (22). Shnyra et al. also failed to detect measurable levels of IL-10 secretion in culture supernatants from LPS-primed cells without subsequent optimal stimulation (27). Our results, in contrast, indicate that an early up-regulation of IL-10 occurs after P. gingivalis LPS pretreatment and secondary stimulation, and secondary optimal stimulation is required for IL-10 production. Furthermore, we found that pretreatment of monocytes with a low dose of E. coli LPS induces a refractoriness in IL-10 production upon secondary stimulation (Fig. 4). A reciprocal regulation of IL-6 and IL-10 in P. gingivalis and E. coli LPS-pretreated cells is strongly suggested.

Blocking of endogenous IL-10 in the secondary culture by a neutralizing antibody could prevent IL-6 down-regulation and significantly increase IL-6 production upon restimulation of P. gingivalis LPS-pretreated cells (Fig. 5A). However, IL-8 production from LPS-pretreated and nontreated cells was not affected by addition of an anti-IL-10 antibody (Fig. 5B). Also, an exogenous IL-10 could substitute for P. gingivalis LPS pretreatment and blocked IL-6 production more effectively than IL-8 production (Fig. 6). Taken together, these results indicate that IL-10 in secondary culture is essential for the selective suppression of IL-6 production in our experimental model. In contrast, a previous study showed that neutralization of IL-10 in primary but not secondary culture prevent the down-regulation of TNF- (22). Addition of exogenous IL-10 and transforming growth factor  in primary culture was also shown to induce LPS hyporesponsiveness and substitute for LPS preexposure (23). However, it was noted that the synthesis of IL-10 was down-regulated as was that of TNF- in these studies.

Although IL-10 has been shown to inhibit LPS-induced production of IL-1, IL-6, IL-8, and TNF- by murine macrophages similarly by blocking gene transcription (7), our results suggested that the different regulatory mechanisms were attributable to IL-10-mediated IL-6 down-regulation in P. gingivalis LPS-pretreated cells. Recently, Takeshita et al. (28) reported that IL-10 enhanced the degradation of IL-6 mRNA and regulated IL-6 levels posttranscriptionally in human monocytic cell lines. Bogdan et al. (3) also reported that IL-10 mediated degradation of cytokine mRNAs of TNF-, IL-1, and IL-1 in murine macrophages. The 3' AU-rich sequences have been reported to regulate terminal deadenylation of the mRNA poly(A) tract and caused mRNA degradation by RNase (26). Further, IL-6 mRNA also contains six of these sequences (29) and is presumably susceptible to mRNA degradation. In contrast to IL-6 mRNA, the IL-8 gene lacks an AU-rich sequence in the 3' untranslated region (16). These studies may suggest that IL-10 can inhibit IL-6 synthesis by enhancing its mRNA degradation and/or affecting gene transcription in P. gingivalis LPS-pretreated cells. Our results showed that a higher amount of exogenous IL-10 was required to suppress IL-8 production (Fig. 6), suggesting that IL-6 mRNA is susceptible to IL-10-mediated degradation. Thus, IL-6, but not IL-8, was down-regulated in LPS-pretreated human monocytes by an autocrine IL-10 in our experimental conditions.

We have demonstrated that a low-dose preexposure to P. gingivalis LPS selectively induced IL-6 down-regulation in human peripheral blood monocytes, in contrast to up-regulation of IL-6 and IL-8 production in E. coli LPS-pretreated cells. Our data also emphasized the pivotal role of autocrine IL-10 in the regulation of these proinflammatory cytokines. The structural difference of the LPS molecule is suggested to be important for determining the cellular responses after primary culture. The possible clinical significance of LPS tolerance in the development of periodontal disease is unknown; however, it is attractive to hypothesize that chronic exposure to LPS from periodontopathic bacteria such as P. gingivalis may mobilize the host cells for selective cytokine production and create an imbalance of inflammatory responses in periodontal lesions. The up- and down-regulation of different cytokines induced by P. gingivalis LPS may play a role in the development and progression of chronic inflammation in periodontal tissues.