Local Induction of B Cell Interleukin-10 Competency Alleviates Inflammation and Bone Loss in Ligature-Induced Experimental Periodontitis in Mice

ABSTRACT

Interleukin-10 (IL-10)-producing B cells (B10 cells) play a critical role in the immune system balance by negatively regulating inflammatory responses. This study was conducted to determine the effect of local B10 cell induction on periodontal inflammation and bone loss in ligature-induced experimental periodontitis in vivo. Purified spleen B cells from C57BL/6J mice (8 to 10 weeks old) were cultured with CD40 ligand (CD40L) and the Toll-like receptor 9 (TLR9) agonist cytidine-phosphate-guanosine oligodeoxynucleotide (CpG) to determine effective IL-10 induction in vitro. Silk ligatures (size 7-0) were tied around the mouse maxillary second molars on day 0, followed by the injection of CD40L and CpG into the palatal gingiva on days 3, 6, and 9. All the mice were sacrificed, and samples were collected on day 14. CD40L and CpG significantly increased the level of IL-10 production by B cells in vitro, although the frequencies of CD1d^hi CD5⁺ and IL-10-producing (IL-10⁺) CD45⁺ cells were decreased. IL-10 was predominantly produced by the CD1d^hi CD5⁺ subpopulation of B cells. In vivo, both IL-10 mRNA expression and the number of IL-10⁺ CD45⁺ cells were significantly increased after gingival injection of CD40L and CpG. Periodontal bone loss was significantly decreased and the gingival expression of IL-1β, tumor necrosis factor alpha, and RANKL was significantly reduced. The number of multinucleated tartrate-resistant acid phosphatase-positive cells along the alveolar bone surface was significantly decreased after gingival injection of CD40L and CpG. This study indicates for the first time that the local induction of B10 cell activity could inhibit periodontal inflammation and bone loss.

INTRODUCTION

B cells are traditionally thought to positively activate immune responses by serving as antigen-presenting cells (APCs) and producing antibodies. However, a distinct subset of B cells that play a suppressive role in immune responses, known as regulatory B cells (Bregs), has recently been identified in both humans and mice (1–4). One of the most widely studied Breg subsets is termed interleukin-10 (IL-10)-producing B cells (B10 cells) (5) due to their pivotal capability to produce the anti-inflammatory cytokine IL-10. Currently, due to the lack of unique phenotypic markers, these IL-10-competent B cells are functionally identified by their production of IL-10 (6). Spleen B10 cells are predominantly enriched within the CD1d^hi CD5⁺ subpopulation, where among them more than 50% are competent to express IL-10 (6, 7). Although the frequency is low (1 to 3% of splenic B cells in mice and less than 1% of circulating B cells in humans), there is emerging evidence demonstrating that a deficiency of B10 cells exacerbates the symptoms of many inflammatory and autoimmune diseases (4, 8–11), suggesting the critical role of B10 cells in maintaining immune system balance and their therapeutic potential for the treatment of many immune-related disorders (12). For example, the transfer of agonist anti-CD40 IL-10-producing B cells into mice with arthritis induced benefits against collagen-induced arthritis (CIA) through inhibition of the Th1 response and suppression of the inflammation in the synovia (1, 2). The inflammatory responses in ulcerative colitis are negatively regulated by B10 cells, and the adoptive transfer of B10 cells reduced the level of inflammation in CD19^−/− mice (5).

Recent studies have revealed that various signals are essential for B10 cell function, including CD40 ligation and Toll-like receptor (TLR) signaling (13, 14). The CD40 ligand (CD40L), a member of the tumor necrosis factor (TNF) receptor superfamily protein, is primarily expressed on activated CD4⁺ T cells and binds to CD40 on APCs, leading to many effects, depending on the target cell type. The cross-linking of CD40 with CD40L is required for the differentiation and activation of B cells to produce IL-10 in murine autoimmune disease models (7, 15, 16). TLRs are pattern recognition receptors (PRRs) expressed by B cells and other immune cells to recognize pathogen-associated molecular patterns (PAMPs) and play a key role in the innate and adaptive immune responses (17). Previous investigations indicated that spleen marginal zone (MZ) B cells and B1 B cells can act as a significant source of IL-10 through TLR9 stimulation via the TLR9 agonist cytidine-phosphate-guanosine oligodeoxynucleotide (CpG) (18). Other studies have also suggested that B cells respond to CpG and can induce IL-10-competent B10 cells to produce IL-10 in vitro (19, 20). While the activation of B10 cells by CD40 ligation and TLR9 activation appears to be evident in vitro, it remains to be determined how they affect B10 cell activity and the overall inflammatory response in vivo.

Periodontitis is one of the most prevalent chronic inflammatory diseases in the oral cavity, with the incidence of periodontitis among U.S. adults being 46% in 2012 (21). The hallmarks of periodontitis are inflammation and alveolar bone resorption caused by the host immune response to bacterial infection (22, 23). The conventional principle of therapies for periodontitis focused on the reduction of infection through the use of antibiotics and the mechanical removal of pathogenic agents. These treatments are not always sufficient, since the unbalanced host immune response, which is the direct cause of the tissue destruction during the pathogenesis of periodontitis, is not effectively addressed. Accumulating evidence has demonstrated the potential benefits of promoting B10 cell function in the amelioration of many inflammatory and autoimmune diseases. For instance, in a murine lupus model, transfer of splenic CD1d^hi CD5⁺ B cells from wild-type NZB/W mice into CD19^−/− NZB/W mouse recipients significantly prolonged the survival of the CD19^−/− NZB/W mice (10). However, no investigation of the effect of B10 cell activity on the inflammation and bone loss that occur during periodontitis has been reported.

The purpose of the present study was to determine the effect of the local induction of B10 cell activity on periodontal inflammation and bone loss in ligature-induced experimental periodontitis in vivo.

RESULTS

CD40L- and CpG-induced IL-10 production by B cells in vitro.

The levels of IL-10 mRNA expression and protein secretion by cultured B cells were determined after stimulation with CD40L and CpG for 24 h, 48 h, and 72 h (Fig. 1a and b). The levels of both IL-10 mRNA expression and protein secretion were significantly increased at all time points after stimulation with CD40L and CpG (P < 0.01). A combination of low doses of CD40L and CpG (0.1 μg/ml of CD40L and 1 μM CpG) appeared to be more effective for the induction of IL-10 than a combination of high doses of CD40L and CpG (1 μg/ml of CD40L and 10 μM CpG) at all time points (P < 0.05). Moreover, after B cells were cultured with CD40L and CpG, the IL-10 mRNA level was increased significantly from 24 h to 48 h (P < 0.01) and at 72 h remained at a level similar to that at 48 h (P > 0.05) (Fig. 1a), whereas the secreted IL-10 protein level was significantly elevated continuously from 24 h to 72 h (P < 0.01) (Fig. 1b).

FIG 1.

Levels of IL-10 production by mouse spleen B cells in vitro. Purified mouse spleen B cells were cultured with CD40L and CpG at the indicated dose for 24 h, 48 h, and 72 h. (a and b) After each treatment, the level of IL-10 mRNA expression by cultured B cells (a) and the amount of IL-10 protein secreted from cultured B cells (b) were detected by RT-qPCR and ELISA, respectively. (c and d) Subsets of B cells (CD1d^lo CD5⁻, CD1d^hi CD5⁻, CD1d^hi CD5⁺, and CD1d^lo CD5⁺ cells) cultured for 48 h with no treatment (c) and CD40L+CpG treatment (d) were sorted by flow cytometry, and the level of IL-10 mRNA expression in each subset was detected by RT-qPCR. Data are presented as means and SDs (n = 4 animals/group). *, P < 0.05; **, P < 0.01.

IL-10 production was mainly derived from CD1d^hi CD5⁺ B cells.

In order to verify the source of IL-10, B cells stimulated with CD40L and CpG for 48 h were sorted into 4 subpopulations: CD1d^lo CD5⁻, CD1d^hi CD5⁻, CD1d^hi CD5⁺, and CD1d^lo CD5⁺ cells. The level of IL-10 mRNA expression of each subset was determined by reverse transcription (RT)-quantitative PCR (qPCR) (Fig. 1c and d). In untreated B cells, the level of IL-10 mRNA expression by the CD1d^hi CD5⁺ subset was significantly higher than that by any other subset (Fig. 1c) (P < 0.01), whereas no difference in the level of IL-10 mRNA expression was observed among CD1d^lo CD5⁻, CD1d^hi CD5⁻, and CD1d^hi CD5⁺ cells. After treatment with CD40L and CpG, while the CD1d^hi CD5⁺ B cells remained the most predominant subset of IL-10-expressing cells (P < 0.01), the level of IL-10 mRNA expression by the CD1d^lo CD5⁺ and CD1d^hi CD5⁻ subsets was higher than that by the CD1d^lo CD5⁻ subset (Fig. 1d).

The frequency of CD1d^hi CD5⁺ B cells was decreased after CD40L and CpG stimulation in vitro.

Despite the significant IL-10 production (Fig. 1a and b), flow cytometry results demonstrated that the frequencies of CD1d^hi CD5⁺ cells were significantly lower after stimulation with either low doses or high doses of CD40L and CpG than after stimulation with the control and CD40L alone (Fig. 2a and b) (P < 0.05). The result of immunocytochemistry (ICC) was consistent with the flow cytometry data, indicating a decrease in the amount of IL-10-expressing CD45⁺ B cells after stimulation with CD40L and CpG than after stimulation with CD40L only (Fig. 2c to f).

FIG 2.

Evaluation of B10 cell frequencies by flow cytometry and ICC. Purified mouse spleen B cells were cultured with CD40L and CpG at the indicated dose and for the indicated times. (a and b) The frequencies of CD1d^hi CD5⁺ B cells were detected by flow cytometry and are presented as means ± SDs (n = 4). Q1 to Q4, quadrants 1 to 4, respectively. (c to e) Alternatively, the frequencies of IL-10⁺ CD45⁺ B cells were determined by ICC staining for IL-10 (red) and CD45 (green). Representative images of 48-h cultures are shown. Arrows, cells that stained double positive. Magnifications, ×400. The results for isotype control staining are presented in Fig. S1. (f) Frequencies of IL-10⁺ CD45⁺ B cells presented as means ± SDs (n = 4 animals/group). *, P < 0.05; **, P < 0.01.

Periodontal bone loss was inhibited by local administration of CD40L and CpG.

The ligature-induced experimental periodontitis model was used to determine the effect of the local induction of B10 cell activity on periodontal bone resorption in vivo. The areas of bone loss around maxillary second molars were measured after each treatment (Fig. 3). As expected, the resorption area on the ligation side was significantly larger than that on the control side (P < 0.05), indicating that the ligature severely induced periodontal bone loss. The bone resorption area was significantly decreased after the injection of low-dose CD40L and CpG (P < 0.05) but not that of high-dose CD40L and CpG.

FIG 3.

Measurement of alveolar bone resorption. Silk ligatures were tied around the maxillary secondary molars on day 0, and injection of CD40L and CpG at two different doses (0.1 μg/ml of CD40L + 1 μM CpG [CD40L+CpG] or 1 μg/ml of CD40L and 10 μM CpG [CD40L_H+CpG_H]) was performed on days 3, 6, and 9. (a) Maxillae were collected on day 14, and the area of palatal alveolar bone resorption around the maxillary secondary molars was measured. (b) Area of bone resorption. Data are presented as the bone resorption area per square millimeter determined at a magnification of ×30. Bar charts show the mean area of palatal alveolar bone resorption ± SD (n = 8 animals/group). *, P < 0.05; **, P < 0.01.

The number of B10 cells in periodontal tissue was elevated by local administration of CD40L and CpG.

In order to localize IL-10-producing B cells in periodontal tissues in vivo, CD45⁺ IL-10-producing (IL-10⁺) cells were detected by immunofluorescence staining (Fig. 4). The results showed that CD45⁺ IL-10⁺ cells were barely noticeable in healthy gingiva, whereas they were readily detectable in gingival tissues after ligation (P < 0.05). In the mice in which both sides were ligatured, the number of CD45⁺ IL-10⁺ cells was significantly increased on the side injected with low doses of CD40L and CpG compared to that on the side injected with phosphate-buffered saline (PBS) (P < 0.01). However, there was no significant difference in the number of CD45⁺ IL-10⁺ cells between the side injected with high doses of CD40L and CpG and the side injected with PBS.

FIG 4.

Detection of IL-10- and CD45-positive B cells in gingiva by immunofluorescence staining. Silk ligatures were tied around the maxillary secondary molars on day 0, and injection of CD40L and CpG at two different doses (0.1 μg/ml of CD40L plus 1 μM CpG or 1 μg/ml of CD40L plus 10 μM CpG) was performed on days 3, 6, and 9. (a to c) Tissues were collected on day 14, stained for IL-10 and CD45, and evaluated by confocal microscopy. Images of gingival tissues were analyzed at a magnification of ×630. The results for isotype control staining are presented in Fig. S2. Arrows, IL-10⁺ CD45⁺ cells. (d) Bar chart showing the mean number of IL-10⁺ CD45⁺ cells per square millimeter ± SD (n = 6 animals/group). *, P < 0.05; **, P < 0.01.

Gingival IL-10 mRNA expression was increased but RANKL and IFN-γ mRNA expression was decreased after local injection of CD40L and CpG.

The levels of expression of genes for inflammatory cytokines were detected using RT-qPCR (Fig. 5). Our results showed that the levels of gingival expression of inflammatory cytokines IL-1β, RANKL, IL-6, IL-17, gamma interferon (IFN-γ), and monocyte chemoattractant protein 1 (MCP-1) were significantly higher on the ligation side than the control side, while no significant difference in the levels of expressions of IL-10, TNF alpha (TNF-α), osteoprotegerin (OPG), or intercellular adhesion molecule 1 (ICAM-1) was observed. Notably, the level of anti-inflammatory cytokine IL-10 expression was significantly increased but the levels of gingival RANKL and IFN-γ expression were decreased after gingival injection of low doses of CD40L and CpG (P < 0.05). There were no differences in the levels of gingival expression of IL-1β, OPG, TNF-α, IL-6, IL-17, MCP-1, or ICAM-1 between the side treated with low-dose CD40L and CpG and the side treated with the PBS control. Gingival injection of high-dose CD40L and CpG did not produce any change in the level of cytokine expression compared to that on the control side, except for an increase in the level of IL-10 expression.

FIG 5.

Levels of expression of mRNA for inflammatory cytokines determined by RT-qPCR. Silk ligatures were tied around the maxillary secondary molars on day 0, and injection of CD40L and CpG at two different doses (0.1 μg/ml of CD40L plus 1 μM CpG or 1 μg/ml of CD40L plus 10 μM CpG) was performed on days 3, 6, and 9. The levels of expression of mRNA for 10 genes (IL-10, IL-1β, RANKL, OPG, TNF-α, IL-6, IL-17, IFN-γ, MCP-1, and ICAM-1) in gingival tissues on day 14 were detected by RT-qPCR, as indicated, and their relative levels of expression were normalized to the level of GAPDH expression. Data are presented as means ± SDs (n = 8 animals/group). *, P < 0.05; **, P < 0.01.

The levels of IL-1β, TNF-α, and RANKL protein production in gingival tissues were reduced after local injection of CD40L and CpG.

To determine the level of inflammatory cytokine response in situ, immunohistochemistry (IHC) was performed to localize and evaluate the levels of cytokine production in gingival tissues (Fig. 6). The numbers of IL-1β-, RANKL-, and TNF-α-positive cells were increased after ligation compared to the numbers after the PBS control treatment (P < 0.01). In contrast, the numbers of cells positive for IL-1β, RANKL, or TNF-α were greatly reduced after gingival injection of low-dose CD40L and CpG compared to the numbers after the PBS control treatment (P < 0.05). Gingival injection of high-dose CD40L and CpG did not change the number of positive cells compared to the numbers on the side receiving the PBS control treatment.

FIG 6.

Detection of IL-1β, TNF-α, and RANKL levels in gingiva by IHC staining. Silk ligatures were tied around the maxillary secondary molars on day 0, and injection of CD40L and CpG at two different doses (0.1 μg/ml of CD40L plus 1 μM CpG or 1 μg/ml of CD40L plus 10 μM CpG) was performed on days 3, 6, and 9. After tissue collection on day 14, IHC staining was performed on tissue sections, and images of periodontal tissues were analyzed for IL-1β (a to d), TNF-α (e to h), and RANKL (i to l) at a magnification of ×400. The results for isotype control staining are presented in Fig. S3. All the bar graphs show the mean number of positive cells in gingiva per square millimeter ± SD (n = 3 animals/group). *, P < 0.05; **, P < 0.01.

The number of multinucleated TRAP-positive cells was decreased after local administration of CD40L and CpG.

Tartrate-resistant acid phosphatase (TRAP) staining was performed to evaluate the osteoclastogenic activities within periodontal tissues (Fig. 7). The results showed that the number of multinucleated TRAP-positive cells along the alveolar bone surface was dramatically increased after ligation (P < 0.01), whereas the number of such cells was significantly decreased after injection of low-dose CD40L and CpG (P < 0.05). The number of multinucleated TRAP-positive cells on the side injected with high-dose CD40L and CpG was not changed from that on the side injected with the PBS control.

FIG 7.

Evaluation of osteoclastogenesis in periodontal tissues by TRAP staining. Silk ligatures were tied around the maxillary secondary molars on day 0, and injection of CD40L and CpG at two different doses (0.1 μg/ml of CD40L plus 1 μM CpG or 1 μg/ml of CD40L plus 10 μM CpG) was performed on days 3, 6 and 9. (a to c) TRAP staining was performed on tissue sections, and images of periodontal tissues were analyzed at a magnification of ×200. (d) The bar chart shows the mean number of multinucleated TRAP-positive cells along the alveolar bone surface per square millimeter ± SD (n = 3 animals/group). *, P < 0.05; **, P < 0.01.

DISCUSSION

The current means of resolution of periodontitis is focused more on management of the infection, while they neglect the homeostatic imbalance of immune responses and bone metabolism. Thus, immunological intervention can be effective for the direct curtailment of immune-mediated pathogenesis during periodontitis. The current work tested the efficacy of the local induction of B10 cells on the amelioration of periodontal inflammation and bone loss.

We first investigated the effect of CD40L and CpG signaling on the induction of B10 cells in vitro. Even though it has been established that both CD40L and CpG play an important role in the activation of B10 cells, the synergy, doses, and time effects of the two stimuli on the function of B10 cells were unknown. We demonstrated that CpG in combination with CD40L dramatically increased the levels of IL-10 mRNA expression and secretion by B cells (Fig. 1a and b). According to recent studies, B10 cells are believed to develop from a B10 progenitor (B10pro) cell population. CD40 signals appeared to predominantly induce B10pro and B10 cell maturation so that they become competent for cytoplasmic IL-10 expression during B10 cell development instead of inducing the direct secretion of IL-10 (8, 19). CpG, which is a strong inducer of IL-10 expression in murine B cells, is able to further induce B10 cells to secret IL-10 (4, 18). More importantly, our results indicated that the effect of CD40L and CpG on B10 cell activation was associated with the dose and the duration of dosing. Low doses of CD40L and CpG appeared to be more effective than high doses in stimulating IL-10 production by B cells (Fig. 1a and b). It has long been demonstrated that the outcome of CD40 engagement on B cells might depend on the stage of B cell maturation and on the duration or the strength of the mutual signals between T cells and B cells that dictate their further differentiation into Bregs, memory cells, or plasma cells (24–26). Our data indicate that CD40L and CpG signals at various intensities may differentially regulate the activation and function of B10 cells.

Commonly recognized phenotypic markers, CD1d and CD5, were used to characterize the competency of B10 cells for IL-10 production and the effect of CD40L and CpG stimulation on B10 cell expansion in vitro. Our data showed that IL-10-expressing B cells were enriched in the CD1d^hi CD5⁺ cell subset (Fig. 1c and d). This finding was in agreement with the findings of previous studies showing that the IL-10 production of splenic B cells was predominantly restricted to the CD1d^hi CD5⁺ cell subset (7). Furthermore, our results demonstrated that IL-10 expression by the CD1d^hi CD5⁻ and CD1d^lo CD5⁺ subpopulations also increased after stimulation with CD40L and CpG, indicating that the CD1d^hi CD5⁺ subset should not be considered the only population representative of the source of B10 cells. Indeed, it has been suggested that B10 cells are functionally defined by their competence for IL-10 expression without unique markers (6). Interestingly, treatment of cultured B cells with CD40L and CpG was not able to induce the expansion of the CD1d^hi CD5⁺ B cells in vitro, as evidenced by both flow cytometry and ICC (Fig. 2). This is in contrast to the results observed in vivo showing that the number of IL-10⁺ CD45⁺ B cells in periodontal tissue was elevated by the local administration of CD40L and CpG (Fig. 4). It is noted that the difference observed in vitro versus in vivo could be partially derived from the different B cell markers used. These data underscore the lack of unique phenotypic markers for B10 cells, and so far the most reliable method for the identification of B10 cells still relies on their production of IL-10 (6). It has been indicated that CD40 signals do not induce B cell clonal expansion but stimulate CD1d^hi CD5⁺ B cells to become more functionally competent to produce IL-10 (8), while CpG oligonucleotides induce IL-10-competent B cells to produce IL-10 (19). More importantly, studies have shown that additional T cell help and B cell receptor (BCR) signaling are required for optimal B10 expansion (27–29). Xiao et al. found that Tim-1 is indispensable for the induction of Bregs and that Tim-1-deficient B cells had reduced levels of IL-10 production as well as increased levels of proinflammatory cytokine production and increased Th1/Th17 responses (27). The maturation of B10 cells into B10 effector cells also required an IL-21-dependent cognate interaction with T cells (28). Moreover, recent findings demonstrated that B10 cells are positively selected by self-reactivity and that a higher BCR signaling strength promotes B10 cell development (29). Our results may have well substantiated the notion that B10 cells represent a transient functional program rather than a cell lineage or stage of differentiation, and their development requires cognate interaction, innate signals, as well as BCR signaling (6).

This study is the first attempt to reveal the effect of local B10 cell induction on periodontal inflammation and bone loss in vivo. It has been suggested that IL-10-producing Bregs appeared under intestinal inflammatory conditions but could not be detected in the normal state (30, 31). Likewise, in this study, we found that B10 cells rarely existed in healthy periodontal tissues, whereas they could be detected in the periodontal tissue of mice with experimental periodontitis (Fig. 4), suggesting that a chronic inflammatory microenvironment is crucial for the differentiation and activation of B10 cells. It is presumed that these rare B10 cells naturally occurring under inflammatory conditions are not sufficient to antagonize disease progression due to the overwhelming proinflammatory cytokine responses, including the IL-1β, TNF-α, IL-6, IL-17, IFN-γ, and MCP-1 responses, and upregulation of RANKL leads to the activation of osteoclasts (Fig. 5 to 7). These cytokines have been demonstrated to be involved in the pathogenesis of periodontitis. For example, TNF-α plays an important role at an early stage in the inflammatory cascade (32). IL-1β and TNF-α stimulate bone loss by increasing the number of osteoclasts formed (33). IL-1β also induces RANKL expression and the osteoclastogenic effect (34). Our results suggest that the observed reduction of bone resorption probably occurs via suppression of the proinflammatory cytokines IL-1β and TNF-α and inhibition of RANKL production.

After the local administration of low-dose CD40L and CpG, the number of B10 cells in periodontal tissue was increased (Fig. 4). In addition, the level of bone resorption was decreased (Fig. 3) and was accompanied by decreased levels of proinflammatory cytokines, including IL-1β, TNF-α, and RANKL, in situ (Fig. 6) and reduced numbers of osteoclasts along the alveolar bone surface (Fig. 7). Importantly, the level of IL-10 expression in gingival tissues was highly increased (Fig. 5). Taken together, these results suggest that the amelioration of bone resorption is associated with the suppressive role of locally induced B10 cells in inflammation and the host immune response via IL-10 production.

We noted that at the mRNA level there was no significant difference in the levels of TNF-α expression between healthy and ligatured gingiva; however, at the protein level, the difference, detected by IHC staining, was significant (Fig. 5 versus 6). Additionally, when the levels of expression of IL-1β and TNF-α on the side treated with low-dose CD40L and CpG and the side treated with the PBS control were compared, the levels of both IL-1β (Fig. 6) and TNF-α (Fig. 6) were significantly reduced at the protein level, as detected by IHC staining, but no such difference was observed at the mRNA level (Fig. 5). Two reasons might explain this discrepancy. First, the level of mRNA for a gene does not always predict its protein level. For example, Chen et al. (35) reported that the protein levels appeared to be strongly correlated with the mRNA level for only a small subset of proteins. The other reason might be related to the divergence of the targeted region between the two analysis approaches. IHC staining was focused on the gingiva mesial and distal to the second molar, which was more specific to the injection site. However, the area of gingival tissue collected for mRNA analysis was larger than the area of the injection site, which may contain surrounding healthy gingiva. Thus, the mRNA expression of some cytokines could be diluted, which could hinder the detection of subtle changes.

So far, the majority of strategies for the application of B10 cell-mediated treatment in disease models that have been reported use adoptive cellular transfer. However, the high cost and time-consuming procedures limit implementation in the clinic. The systemic administration of a B10 cell agonist to humans could be also troublesome because it may cause severe adverse events, such as cytokine release syndrome (36). Therefore, the local induction of B10 cell function would be a more viable regimen to curtail periodontal disease inflammation and bone loss. Indeed, our results demonstrated that on day 14 there was no difference in the total serum IgG level among the three groups of mice tested (see Fig. S4 in the supplemental material). A more extensive panel of serum markers of systemic inflammation is required for the definitive determination of whether local gingival administration of CD40L and CpG causes systemic immunological changes. Furthermore, we examined the bacterial load in oral swab specimens from each animal via quantification of total bacterial DNA (Fig. S5). The results showed no difference in the bacterial load in the presence or absence of gingival injection of CD40L and CpG, suggesting that the local administration of CD40L and CpG does not affect oral bacterial colonization.

Our results showed that high doses of a combination of CD40L and CpG could not suppress inflammation and bone resorption (Fig. 3, 5, and 6). It has been reported that the activation of CD40 leads to the expression of CD80/CD86 and the production of IL-12, thus intensifying T cell costimulation and skewing Th1/Th2 responses (37). Hassan et al. (38) suggested that the level of CD40L/CD40 signaling was the determinant of a tilt to a pathogenic role or a protective role. It has also been shown that TLR9-mediated inflammation triggers alveolar bone loss in experimental murine periodontitis (39). Thus, the anti-inflammatory effect of B10 cells may be neutralized by the inflammation upregulated due to TLR9 activation in response to high-dose CpG. Further studies are warranted to determine the mechanisms by which more specific and targeted strategies can be achieved to induce the local IL-10 competency of B10 cells without the activation of proinflammatory immune responses in vivo.

In summary, the present study demonstrates that the local induction of the IL-10 competency of B10 cells is associated with the inhibition of inflammation and bone loss in ligature-induced experimental periodontitis in vivo, which probably occurs via the suppression of proinflammatory cytokine and RANKL production. The findings of the current study may promote the development of a novel strategy for immunoregulatory therapy for periodontitis.

MATERIALS AND METHODS

Mice.

Wild-type (WT) C57BL/6 mice were purchased from The Jackson Laboratory (Bar Harbor, ME). All the mice used for the experiments were 8 to 10 weeks of age and were housed in sterile, specific-pathogen-free units. The experiments involving animals were approved by the Animal Care and Use Committee of the Forsyth Institute.

B cell purification and culture conditions.

Spleens were harvested from sacrificed mice and gently grinded on a metal mesh in Iscove's modified Dulbecco's medium (IMDM; Gibco). A suspension of single splenocytes was obtained by passage of the spleens through a 40-μm-mesh-size cell strainer after lysis of red blood cells with ACK lysis buffer (Life Technologies). A pan-B cell isolation kit (Miltenyi Biotec) was used for the isolation of B cells through magnetic columns (Miltenyi Biotec). Purified B cells (>95%, 5 × 10⁶ cells/well) were cultured in a 96-well tissue culture plate in 0.2 ml of complete medium (IMDM containing 10% fetal bovine serum, 2 mM l-glutamine, 100 U/ml penicillin, 100 U/ml streptomycin, 2.5 μg/ml amphotericin B [Fungizone], 0.1% 2-mercaptoethanol) for 24 h, 48 h, and 72 h in the presence or absence of CD40 ligand (CD40L; eBioscience) and CpG (ODN 2006 oligodeoxynucleotides; 5′-TCGTCGTTTTGTCGTTTTGTCGTT-3′; Hycult Biotech). The isolated B cells were divided into four groups (n = 4 animals/group): a control group (which received no stimulation), the CD40L group (which was stimulated by 0.1 μg/ml CD40L), the CD40L+CpG group (which was stimulated by 0.1 μg/ml of CD40L and 1 μM CpG), and the CD40L_H+CpG_H group (which was stimulated by 1 μg/ml of CD40L and 10 μM CpG).

Animal model and local administration of CD40L and CpG.

To induce experimental periodontitis in mice, a ligature-induced experimental periodontitis model was used as previously described with a slight modification (40). Briefly, silk ligatures (size 7-0; Fisher Scientific) were tied around the maxillary second molars on day 0 and remained in place for 2 weeks. WT mice were randomly divided into three groups. In group 1 (n = 11 animals/group), the right side was untreated and the left maxillary second molar was ligated. In groups 2 and 3 (n = 11 animals/group), both maxillary second molars were ligated, the left side was injected with a CD40L and CpG mixture (0.1 μg/ml of CD40L plus 1 μM CpG for group 2 or 1 μg/ml of CD40L plus 10 μM CpG for group 3), and the right side was injected with PBS. Two microliters of the CD40L and CpG mixture or PBS was injected into the palatal gingival papilla between the first and second molars and the second and third molars of each mouse on days 3, 6, and 9. All the mice were euthanized by CO₂ inhalation on day 14.

ELISA.

The secreted IL-10 protein levels in the supernatants of cultured B cells were measured by use of a mouse IL-10 enzyme-linked immunosorbent assay (ELISA) Max Standard kit (BioLegend, San Diego, CA) following the manufacturer's protocol. For each sample, the assay was performed in duplicate and a standard curve was run with each assay. The absorbance at 450 nm was measured in a microplate reader (BioTek), and the IL-10 concentration (in picograms per milliliter) was calculated on the basis of the standard curve.

Flow cytometry analysis and cell sorting.

After the times indicated above, cultured B cells were washed with cell staining buffer and incubated with blocking buffer containing anti-mouse CD16/32 antibody (Ab). Cells were then stained with both phycoerythrin-labeled anti-mouse CD1d and allophycocyanin-labeled anti-mouse CD5 Ab. All the reagents and Abs were from BioLegend (San Diego, CA). Data were collected on a FACSAria flow cytometer (BD Biosciences) and analyzed using FlowJo software (TreeStar, Inc., San Carlos, CA). In addition, the cultured B cells stimulated with CD40L and CpG for 48 h were sorted into CD1d^lo CD5⁻, CD1d^hi CD5⁻, CD1d^hi CD5⁺, and CD1d^lo CD5⁺ subsets by use of the FACSAria flow cytometer. The expression of IL-10 mRNA of each B cell subset was further detected by real-time PCR (n = 4).

Real-time PCR.

Gingival tissues around the second molars in the maxilla were collected at the termination of the experiment (day 14) under a surgical microscope and homogenized in lysis buffer using a tissue homogenizer (Omni). Total RNA was extracted from cultured B cells or individual homogenized gingival tissue samples using a PureLink RNA minikit (Ambion). cDNA was generated using a SuperScript II reversed transcriptase kit (Invitrogen) according to the manufacturer's protocol. The expression of mRNA for IL-10, IL-1β, RANKL, OPG, TNF-α, IL-6, IL-17, IFN-γ, MCP-1, and ICAM-1 by the samples in vivo was detected by real-time qPCR using LightCycler SYBR green I master mix and a LightCycler 480 instrument system (Roche). The sequences of the primers are shown in Table 1. The GAPDH (glyceraldehyde-3-phosphate dehydrogenase) gene was used as an internal control. The cycle threshold (C_T) value of each gene was recorded by LightCycler software. The levels of expression of the target gene relative to that of the GAPDH gene were calculated by the equation 2^−ΔCT · −ΔC_T = C_T for the target gene − C_T for the GAPDH gene (n = 8 animals/group).

TABLE 1.

Primers and sequences used for PCR

Target	Orientation	Sequence
GAPDH	Forward	5′-CCCCAGCAAGGACACTGAGCAA-3′
	Reverse	5′-GTGGGTGCAGCGAACTTTATTGATG-3′
IL-1β	Forward	5′-CCAGCTTCAAATCTCACAGCAG-3′
	Reverse	5′-CTTCTTTGGGTATTGCTTGGGATC-3′
IL-6	Forward	5′-TCCAGTTGCCTTCTTGGGAC-3′
	Reverse	5′-GTACTCCAGAAGACCAGAGG-3′
IL-10	Forward	5′-TGGCCCAGAAATCAAGGAGC-3′
	Reverse	5′-CAGCAGACTCAATACACACT-3′
IL-17	Forward	5′-AACACTGAGGCCAAGGACTT-3′
	Reverse	5′-ACCCACCAGCATCTTCTCG-3′
TNF-α	Forward	5′-CACAGAAAGCATGATCCGCGACGT-3′
	Reverse	5′-CGGCAGAGAGGAGGTTGACTTTCT-3′
OPG	Forward	5′-AGCAGGAGTGCAACCGCACC-3′
	Reverse	5′-TTCCAGCTTGCACCACGCCG-3′
RANKL	Forward	5′-GGGTGTGTACAAGACCC-3′
	Reverse	5′-CATGTGCCACTGAGAACCTTGAA-3′
MCP-1	Forward	5′-GAAGGAATGGGTCCAGACAT-3′
	Reverse	5′-ACGGGTCAACTTCACATTCA-3′
ICAM-1	Forward	5′-TCGGGAAGGGAGCCAAGTAACT-3′
	Reverse	5′-GATCCTCCGAGCTGGCATT-3′
MCP-1	Forward	5′-AGAGCCAGATTATCTCTTTCTACCTCAG-3′
	Reverse	5′-CCTTTTTCGCCTTGCTGTTG-3′

Immunocytochemistry (ICC).

To fix the cells, 10% formalin was added to the B cell culture for 15 min at room temperature. After the cells were washed with distilled water twice, they were resuspended in distilled water with the same volume as the volume of the culture medium. Eight microliters of the cell suspension was spotted on Superfrost Plus microscope slides (Fisher). The slides were then placed in an oven at 40°C until they were dry and stored at 4°C before they were stained. Briefly, cell spots were incubated with 0.25% Triton X-100 for 10 min. After the cells were blocked with 1% bovine serum albumin (BSA) for 30 min at room temperature, the cells were incubated with a mixture of rat anti-mouse CD45 Ab (Abcam) diluted 1:1,000 in PBS–Tween 20 (PBST) and goat anti-mouse IL-10 Ab (Santa Cruz) at a dilution of 1:100 for 1 h at room temperature and washed three times in PBST. A secondary Ab double-staining kit (catalog number DS206C; GBI) containing Emerald-conjugated anti-rat Ab and Permanent Red-conjugated anti-goat Ab was used to visualize cells double positive for CD45 and IL-10. Images of nine randomly chosen fields for each cell spot were obtained, and the percentage of double-positive cells among all counted cells was calculated (n = 4 animals/group).

Bone morphometric analysis.

The maxillae were removed, and the flesh was removed by a dermestid beetles colony. After the bone was bleached with 3% hydrogen peroxide, it was stained with 1% toluidine blue. Bone resorption measurements were assessed under a Nikon microscope (Nikon SMZ745T; Nikon Instruments Inc., Japan). Images of the palatal surfaces of the maxillae were captured, and the polygonal area of bone loss was measured using ImageJ software (NIH). The area was enclosed longitudinally from the cementoenamel junction (CEJ) to the alveolar bone crest (ABC) and transversely from the area distal to the first maxillary molar to the area mesial of the third maxillary molar as previously described (41). The results are presented in square millimeters (n = 8 animals/group).

Tissue histological analysis.

The collected maxillae were fixed in 4% formaldehyde overnight followed by decalcification in 10% EDTA for 2 weeks at 4°C with agitation. Paraffin-embedded specimens were cut into sections of 5 μm along the long axis of the molars. For immunohistochemistry staining, sections were deparaffinized, rehydrated, and incubated with 3% hydrogen peroxide for 10 min, followed by heat-induced antigen retrieval in 10 mM sodium citrate buffer at pH 6.0. After the samples were blocked with 1% BSA for 30 min at room temperature, they were incubated with primary Abs. For immunofluorescence staining, a mixture of rat anti-mouse CD45 Ab (also known as B220; catalog number ab25386; Abcam) diluted 1:200 in PBST and goat anti-mouse IL-10 Ab (catalog number sc-1783; Santa Cruz Biotechnology) at a dilution of 1:50 was applied and the samples were incubated overnight at 4°C. Incubation with donkey anti-goat IgG (1:100; Alexa Fluor 555; catalog number ab150130; Abcam) was followed by incubation with goat anti-rat IgG (1:100; Alexa Fluor 633; catalog number A-21094; Thermo Fisher) in the dark at room temperature for 1 h. Before application of a coverslip, the cells were counterstained with DAPI (4′,6-diamidino-2-phenylindole; 1:50; BioLegend) for 30 s. The stained slides were immediately observed using a confocal microscope system (Zeiss LSM780). Images of both mesial and distal periodontal tissues were captured at a magnification of ×630. The number of IL-10⁺ CD45⁺ cells in 4 to 5 serial sections (9 fields per section) of each mouse were counted, and the values were averaged (n = 6). For immunohistochemical staining, tissue sections were incubated with rabbit anti-mouse Abs against IL-1β (1:200; catalog number ab9722; Abcam), RANKL (1:500; catalog number ab62516; Abcam), or TNF-α (1:100, catalog number ab6671; Abcam) for 1 h at room temperature and were washed three times with Tris-buffered saline–Tween 20. A rabbit-specific horseradish peroxidase-diaminobenzidine staining kit (catalog number ab64261; Abcam) was applied according to the manufacturer's instruction to visualize the location of the antigen. For tartrate-resistant acid phosphatase (TRAP) staining, tissue sections were stained using an acid phosphatase kit (catalog number 378A; Sigma). After the sections were counterstained with hematoxylin, images of gingival areas mesial and distal to the second molars from each section were acquired under a light microscope at an objective magnification of ×40. The number of cells double positive for CD45 and IL-10, the number of cells positive for IL-1β, RANKL, and TNF-α, and the number of cells positive for multinucleated TRAP along the alveolar bone surface were counted (n = 6 animals/group).

Statistical analysis.

Data obtained by flow cytometry, ELISA, determination of IL-10 gene expression, and ICC of B cells were analyzed by one-way analysis of variance (ANOVA) and Tukey's multiple-comparison test with the Bonferroni correction for comparisons among groups. Unpaired Student's t test was used to evaluate the differences in the levels of bone resorption, the levels of cytokine gene expression, and the histology of tissues within each group. Data analysis was performed using SPSS (version 16.0) software. A P value of <0.05 was considered statistically significant.