IRF6 Regulates the Expression of IL-36γ by Human Oral Epithelial Cells in Response to Porphyromonas gingivalis

Toll-like receptors are critical mediators of host immunity (1). The specificity of TLR signaling is determined in part by the differential use of adapter proteins, for example, MyD88 and Toll/IL-1R domain–containing adapter inducing IFN-β (TRIF). TLR signaling can therefore be broadly divided into the MyD88-dependent and TRIF-dependent pathways (2). The MyD88-dependent pathway uses the protein kinase IL-1R–associated kinase 1 (IRAK1) to promote the activation of NF-κB and IFN regulatory factors (IRFs), which then transactivate inflammatory genes (e.g., TNF, IL-8, CCL5) (2, 3). The transactivation of inflammatory genes by the TRIF-dependent pathway (e.g., IFN-β) occurs via the TAK1-mediated activation of NF-κB and IRF3 activation by TBK1 (2, 3). Importantly, the differential transactivation of inflammatory genes by IRFs allows them to shape the immune response to pathogens by imparting signaling specificity to individual TLR pathways (4–7).

Porphyromonas gingivalis is a keystone pathogen in chronic periodontitis, an inflammatory disease that causes tooth loss due to the destruction of supporting tissues, including the resorption of alveolar bone (8–10). P. gingivalis grows in the superficial layers of a polymicrobial biofilm that accretes to the tooth root (11). Crucially, the epithelial cells adjacent to the biofilm express TLRs, thereby enabling them to detect and respond to P. gingivalis by eliciting the activation and recruitment of immune cells via the production of inflammatory cytokines (e.g., IL-8). TLR2 is a key mediator of the innate immune response to P. gingivalis (12, 13). However, P. gingivalis expresses various virulence factors, including the extracellular proteinases Kgp and RgpA/B, which allows it to dysregulate local host immunity, for example by manipulating TLR2 signaling (14, 15). The resulting disruption of the homeostasis that normally exists between the immune system and the tooth-accreted biofilm causes dysbiosis and proliferation of the biofilm. Importantly, the increased pathogenicity of the biofilm leads to the chronic activation of inflammatory cells (e.g., neutrophils) and collateral inflammatory tissue damage (8–10). Therefore, the stimulation of TLR2-elicited inflammation by P. gingivalis is likely to be central to the pathogenesis of chronic periodontitis.

The constant transit of neutrophils through periodontal tissues is critical for protection against infection (16). However, excessive neutrophil recruitment causes tissue damage due to their inflammatory properties (e.g., reactive oxygen species and matrix metalloproteinase production) (17). Macrophages are major sources of inflammatory cytokines (e.g., TNF and IL-1β) in periodontal tissues, and although the cytokines exert protective effects they can also cause connective tissue breakdown and alveolar bone resorption (18, 19). Indeed, we recently demonstrated that macrophage depletion reduces alveolar bone loss in experimental periodontitis (20). The production of IL-17 by Th17 cells, the dominant T cell subset in human periodontal lesions (21), can cause alveolar bone resorption by promoting the development and activation of osteoclasts in periodontal tissues (17).

IRF6 plays a key role in differentially regulating TLR2-elicited inflammatory cytokine responses by oral epithelial cells (22). These cells are important initiators of inflammation to P. gingivalis, as they are the first host cells to encounter the pathogen. Therefore, in the present study we investigated a role for IRF6 in regulating the inflammatory cytokine responses of oral epithelial cells to P. gingivalis. We demonstrate that IRF6 promoted the expression of the IL-1 family cytokine IL-36γ when human oral epithelial cells were challenged with P. gingivalis. IL-36γ was shown to not only stimulate the expression of neutrophil- and Th17-attracting chemokines by human monocyte-derived dendritic cells and macrophages, but IL-36γ also stimulated their expression by oral epithelial cells. Significantly, IL-17 stimulated IL-36γ expression by human oral epithelial cells, thus raising the possibility that in vivo IL-36γ and IL-17 may form an inflammatory axis. Therefore, our data potentially position IRF6 as an important mediator of inflammation to P. gingivalis through its regulation of IL-36γ.

Keratinocyte serum-free medium, RPMI 1640, DMEM, FCS and supplements (human EGF, bovine pituitary extract, penicillin/streptomycin, and GlutaMAX-1), TaqMan universal master mix II, and PCR primers were from Life Technologies. DermaLife keratinocyte medium and supplements (TGF-α, insulin, epinephrine, apo-transferrin, hydrocortisone, bovine pituitary extract, and glutamine) were from Lifeline Cell Technology. The ON-TARGETplus IRF6, IRAK1, and TLR2 small interfering RNA (siRNA) were from GE Healthcare. Human GM-CSF, IL-4, IL-17A, and IL-36γ, rat anti–IL-36γ mAb, and human CCL20 ELISA kit were from R&D Systems. The human IL-8 ELISA kit was from Life Research; human M-CSF was a gift from Chiron. FSL-1 was from InvivoGen, and Complete protease inhibitors were from Roche. The rabbit anti-IRF6 Ab was from Cell Signaling Technology; the mouse anti–heat shock protein (HSP) 90 mAb was from BD Biosciences. Restriction enzymes and T4 DNA ligase were from New England Biolabs.

Human OKF6/TERT-2 oral epithelial cells (23) (hereafter referred to as OKF6 cells) were cultured in keratinocyte serum-free medium supplemented with 0.2 ng/ml human EGF, 25 μg/ml bovine pituitary extract, 0.4 mM CaCl₂, 100 U/ml penicillin, 100 μg/ml streptomycin, and 2 mM GlutaMAX-1. The human gingival epithelial cell line, TIGK (24, 25), was cultured in DermaLife keratinocyte medium supplemented with 0.5 ng/ml TGF-α, 5 μg/ml insulin, 1 μM epinephrine, 5 μg/ml apo-transferrin, 100 ng/ml hydrocortisone, 0.4% bovine pituitary extract, and 6 mM glutamine. Human monocytes were purified from buffy coats using a RosetteSep Ab mixture (StemCell Technologies), followed by Ficoll-Paque (GE Healthcare) density gradient centrifugation, and then cultured overnight in RPMI 1640 supplemented with 10% FCS, 100 U/ml penicillin, 100 μg/ml streptomycin, and 2 mM GlutaMAX-1. Monocyte-derived dendritic cells were generated by culturing monocytes in the presence of GM-CSF (20 ng/ml) and IL-4 (10 ng/ml) for 7 d (both cytokines were replenished after 4 d). Monocyte-derived macrophages were generated by culturing monocytes in the presence of M-CSF (5000 U/ml) for 7 d (M-CSF was replenished after 4 d). HEK293T cells were cultured in DMEM supplemented with 10% FCS, 100 U/ml penicillin, 100 μg/ml streptomycin, and 2 mM GlutaMAX-1. All cells were cultured at 37°C in a humidified atmosphere of 5% CO₂.

Freeze-dried cultures of P. gingivalis (ATCC 33277) were obtained from the culture collection of the Melbourne Dental School, University of Melbourne. The bacterium was maintained on horse-blood agar plates at 37°C in an anaerobic atmosphere of 5% H₂, 80% N₂, and 15% CO₂. Bacterial colonies were used to inoculate brain heart infusion broth supplemented with 5 mg/ml cysteine, 5 μg/ml hemin, and 5 μg/ml menadione as previously described (26).

P. gingivalis cell concentrations were determined spectrophotometrically and confirmed retrospectively by counting viable cell colonies on horse-blood agar plates (27). The bacteria were harvested by centrifugation at 7000 × g for 20 min at 4°C and resuspended in antibiotic-free keratinocyte culture medium. OKF6 cells and TIGK, which had been allowed to adhere overnight in antibiotic-free culture medium, were challenged with P. gingivalis at a bacterium-to-cell ratio of 100:1.

A reverse-transfection protocol was used for siRNA transfections. Briefly, the siRNAs were diluted to 120 nM with 100 μl Opti-MEM I reduced serum medium, mixed with 100 μl Opti-MEM I reduced serum medium containing 1 μl Lipofectamine RNAiMAX (Life Technologies), and incubated at room temperature for 15–20 min. OKF6 cells and TIGK (2 × 10⁵ cells in 1 ml antibiotic-free culture medium) were seeded into 12-well plates and the transfection mixture was then added. The medium was replaced 16 h later, and the cells either challenged with P. gingivalis or stimulated (FSL-1 or IL-36γ) 48 h posttransfection.

Total RNA was purified using the ReliaPrep RNA cell miniprep system (Promega), which included an on-column DNase-treatment step. RNA was reverse transcribed into cDNA using random primers and SuperScript III reverse transcriptase (Life Technologies) according to the manufacturer’s instructions.

Two micrograms total RNA was reverse transcribed as described above. The cDNA (100 ng/μl) was mixed with TaqMan OpenArray real-time master mix (Life Technologies), and then loaded onto an OpenArray human inflammation plate (Life Technologies) using the OpenArray AccuFill System. PCR was performed on the QuantStudio 12K flex real-time PCR system, and the data were normalized against hypoxanthine phosphoribosyltransferase (HPRT) gene expression using ExpressionSuite software (version 1.0.1).

Quantitative real-time PCR (qPCR) was performed in triplicate (5–10 ng cDNA per reaction) using TaqMan universal master mix II and predeveloped TaqMan assays (Life Technologies) for the following genes: CCL5 (Hs00174575_m1), CCL20 (Hs01011368_m1), CXCL1 (Hs00236937_m1), G-CSF (Hs00738432_g1), GM-CSF (Hs00929873_m1), human β-defensin-2 (hBD2; Hs00175474_m1), IL-1R accessory protein (IL-1RAcP; Hs00895050_m1), IL-6 (Hs00985639_m1), IL-8 (Hs00174103_m1), IL-36α (Hs00205367_m1), IL-36β (Hs00758166_m1), IL-36γ (Hs00219742_m1), IL-36R (Hs00909276_m1), IL-36Ra (Hs01104220_g1), IRAK1 (Hs01018347_m1), IRF1 (Hs00971960_m1), IRF2 (Hs00180006_m1), IRF3 (Hs00155574_m1), IRF4 (Hs01856533_m1), IRF5 (Hs00158114_m1), IRF6 (Hs00196213_m1), IRF7 (Hs00185375_m1), IRF8 (Hs01128710_m1), IRF9 (Hs00196051_m1), and TLR2 (Hs00152932_m1). PCR was performed on an ABI Prism 7900HT sequence detection system (Applied Biosystems), and data were normalized against HPRT gene expression using the ∆Ct (cycle threshold) method (28).

ELISA assays were performed by incubating diluted culture supernatants and standards in 96-well microplates for 2 h. The wells were washed prior to the addition of biotinylated anti–IL-8 and anti-CCL20 Abs according to the manufacturers’ instructions. After incubation at room temperature for 2 h, the plates were washed and then incubated with a streptavidin-HRP conjugate for 2 h. The plates were again washed, 3,3′,5,5″-tetramethylbenzidine substrate was added, and color development was measured at 450 nm using a microplate reader (model 680, Bio-Rad).

Cells were washed twice with ice-cold PBS and then lysed (20 mM Tris-HCl [pH 7. 4], 150 mM NaCl, 1 mM EDTA, 1% Nonidet P-40, 10% glycerol, 10 mM β-glycerol phosphate, 10 mM NaF, and protease inhibitors) on ice for 60 min. The lysates were clarified by centrifugation (13,000 × g for 10 min at 4°C) and the protein concentrations were measured using a protein assay kit (Bio-Rad). Cell lysates were subjected to electrophoresis on 10% NuPAGE gels using MOPS buffer (Life Technologies), followed by Western blotting according to standard protocols. Immunoreactive bands were visualized using ECL reagents (Millipore) and an LAS-3000 imager (Fujifilm) or by exposure to x-ray film (Fujifilm), which were subsequently scanned on a GS-800 calibrated imaging densitometer (Bio-Rad).

The IRF6 (pEF-HA-IRF6) and IRAK1 (pEF-V5-IRAK1) expression plasmids have been described (22, 29, 30). The pGL4-IL36γ gene promoter reporter plasmid was created by amplifying the 5′-flanking region of the IL-36γ gene from OKF6 cell genomic DNA using Pfu DNA polymerase (Promega) and the PCR primers, 5′-CTAGATCTTAGGGTGAAAAGTAAAGACG-3′ (forward) and 5′-TGAAGCTTAGTGTGGTTGTCTCAGCACCT-3′ (reverse). The PCR product was digested with BglII and HindIII, gel purified, ligated into pGL4.11 (Promega), and verified by DNA sequencing. The promoter fragment (−997 nt/+633 nt) ends 5′ adjacent to the adenine nucleotide of the IL-36γ translational (AUG) start site (31). The pGL2-CCL20 gene promoter reporter plasmid was provided by Dr. Toshifumi Matsuyama (Nagasaki University) (32). The pRL-TK Renilla luciferase reporter plasmid was from Promega.

HEK293T cells were seeded in 12-well tissue culture plates at a density of 3 × 10⁵ cells per well and transfected (in duplicate) the next day using FuGENE 6 (Promega). The total amount of plasmid in each transfection was kept constant using empty plasmid. The cells were lysed 24 h posttransfection with passive lysis buffer (Promega), and firefly and Renilla luciferase activities were assayed using the Dual-Glo luciferase assay system (Promega). Renilla luciferase activity was used to normalize transfection efficiencies.

Data combined from three or more independent experiments are presented as the mean ± SEM. Statistical analyses were performed using GraphPad Prism software version 6.01 (GraphPad Software). Differences between two groups were evaluated using the Student t test. For multiple comparisons, statistical analysis was performed using a one-way ANOVA. A p value <0.05 was considered to be statistically significant.

We recently demonstrated that IRF6 functions in the TLR2 signaling pathway to induce inflammatory cytokine expression in oral epithelial cells (22). Given the role of TLR2 in eliciting host inflammation to P. gingivalis (12–15), we investigated whether IRF6 expression was modulated in response to P. gingivalis. As shown in Fig. 1A, IRF6 gene expression was strongly upregulated in human oral epithelial cells (e.g., OKF6 cells) in response to P. gingivalis. In contrast, the expression levels of IRF1, IRF2, IRF3, IRF5, and IRF9 were only weakly upregulated, and the expression levels of IRF4 and IRF7 were unchanged. IRF8 expression was not detected (Fig. 1A). Thus, IRF6 mRNA levels were on the order of 30-fold higher than those of other IRFs following challenge with P. gingivalis. We next investigated the kinetics of the upregulation of IRF6 expression. Increased IRF6 mRNA levels were detected as early as 4 h postchallenge, and by 24 h had increased ∼10-fold (Fig. 1B). Lysates of the cells were subjected to Western blotting with an anti-IRF6 Ab to establish whether IRF6 protein levels were also increased. This revealed that IRF6 protein levels had increased ∼4-fold by 24 h postchallenge (Fig. 1C, 1D). To determine whether the increase in IRF6 mRNA levels was due to increased IRF6 transcription and/or mRNA stabilization, cells were treated with actinomycin D to inhibit gene transcription, and the effects on the upregulation of IRF6 mRNA levels by P. gingivalis were measured. The increase in IRF6 mRNA levels were inhibited by treatment with actinomycin D (Fig. 1E), indicating that the upregulation of IRF6 mRNA expression in response to P. gingivalis was due to increased IRF6 transcription. To determine whether the increased IRF6 transcription was mediated by TLR2 signaling, the effects of gene silencing of TLR2 on IRF6 expression were determined. The data presented in Fig. 1F indicate that the upregulation of IRF6 expression by P. gingivalis was not mediated by TLR2. Consistent with this conclusion was the finding that the stimulation of oral epithelial cells with the TLR2 agonist FSL-1 resulted in only a small (<2-fold) and transient increase in IRF6 mRNA levels (Fig. 1G). Collectively, these data indicate that IRF6 expression in human oral epithelial cells is preferentially upregulated in response to P. gingivalis, consistent with IRF6 playing an important role in mediating the inflammatory response of oral epithelial cells to P. gingivalis.

We next sought to establish whether IRF6 regulates the stimulation of specific inflammatory cytokines in oral epithelial cells by P. gingivalis. Briefly, IRF6 gene expression in OKF6 cells was silenced and the cells then challenged with P. gingivalis. Changes in inflammatory gene expression were identified with the OpenArray human inflammation panel, and the effects of IRF6 gene silencing on the stimulation of specific genes were confirmed by real-time PCR. This analysis revealed that the novel IL-1 family cytokine IL-36γ was not only one of the most strongly induced inflammatory genes, but its induction was IRF6-dependent (Fig. 2A). The IL-36 subfamily of IL-1 cytokines consists of IL-36α, IL-36β, and IL-36γ (33). Neither IL-36α nor IL-36β expression was significantly induced in response to P. gingivalis (Fig. 2B), thus indicating selectivity in the induction of IL-36γ expression. To establish whether the stimulation of IL-36γ mRNA expression resulted in a corresponding increase in IL-36γ protein levels, cell culture supernatants, as well as cell lysates, were subjected to Western blotting with an anti–IL-36γ Ab. Although IL-36γ was not detected in the culture supernatants (data not shown), a marked increase in IL-36γ protein in the cell lysates was demonstrated (Fig. 2C). Because IRF6 functions downstream of IRAK1 in the TLR2 pathway in oral epithelial cells (22), we again used a gene silencing approach to establish whether IRAK1 is also important for the stimulation of IL-36γ gene expression by P. gingivalis. As shown in Fig. 2D, the silencing of IRAK1 inhibited the stimulation of IL-36γ expression. IL-36γ expression was also shown to be strongly stimulated in an IRF6-dependent and IRAK1-dependent manner in human gingival epithelial cells (e.g., TIGK) in response to P. gingivalis (Fig. 2E, 2F). Collectively, these data indicate that IRF6 and IRAK1 function as critical regulators of the stimulation of IL-36γ expression in human oral epithelial cells by P. gingivalis.

In view of the findings above, we sought to establish whether the expression of IL-36γ is regulated by TLR2 signaling. The stimulation of OKF6 cells with FSL-1 resulted in the induction of IL-36γ mRNA expression (Fig. 3A). In contrast, IL-36α and IL-36β expression was not significantly induced (Fig. 3A). As for the stimulation of IL-36γ expression by P. gingivalis (Fig. 2A), the stimulation of IL-36γ expression by FSL-1 was demonstrated to be IRF6-dependent (Fig. 3B). Likewise, the gene silencing of IRAK1 also inhibited the stimulation of IL-36γ expression by FSL-1 (Fig. 3C). The coordinated regulation of IL-36γ gene expression by IRF6 and IRAK1 was investigated further by performing gene promoter reporter assays. The ectopic expression of IRF6 was not sufficient to transactivate the IL-36γ promoter (Fig. 3D). However, the coexpression of IRAK1 resulted in the robust transactivation of the IL-36γ promoter by IRF6 (Fig. 3D). These data strongly suggest that TLR2 stimulates IL-36γ expression by promoting the IRAK1-mediated activation of IRF6.

IL-36γ was recently shown to be a potent stimulator of inflammatory cytokine expression in mouse bone marrow–derived dendritic cells (34). Similar results were obtained in this study where IL-36γ stimulated IL-8, CXCL1, CCL20, IL-6, and CCL5 expression in human monocyte-derived dendritic cells (Fig. 4A–E), although the effects were not statistically significant. Qualitatively similar responses were obtained when monocyte-derived macrophages were stimulated with IL-36γ (Fig. 4F–J). The upregulation in cytokine expression was validated by ELISA for IL-8 (Fig. 4K). Interestingly, the overall magnitudes of the responses elicited in dendritic cells were typically higher than those in macrophages. The IL-36 cytokines exert their effects on cells via IL-36R and IL-1RAcP (35). Therefore, the significantly higher levels of IL-36R and IL-1RAcP expression by dendritic cells could potentially explain their greater responsiveness to IL-36γ stimulation (Fig. 4L, 4M). However, the stronger responses of dendritic cells to IL-36γ also appear to be cytokine specific. For example, the IL-36γ–induced expression levels of CXCL1 and CCL20 in dendritic cells were ∼40-fold and 4-fold higher, respectively, than those in macrophages (Fig. 4B, 4C versus Fig. 4G, 4H), whereas the magnitudes of the stimulation of IL-8 expression in dendritic cells and macrophages by IL-36γ were more comparable (Fig. 4A versus Fig. 4F). Therefore, the magnitudes of the inflammatory cytokine responses elicited by IL-36γ may also be influenced by factors in addition to IL-36R and/or IL-1RAcP expression levels.

Some cytokines can negatively regulate the ability of cells to respond to further stimulation by downregulating the expression levels of their cognate receptors or by inducing the expression of receptor antagonists. In the case of the IL-36 cytokines, their binding to IL-36R is inhibited by the receptor antagonist IL-36Ra (36). IL-36γ did not significantly affect the expression levels of IL-36R, IL-1RAcP, and IL-36Ra in human monocyte-derived dendritic cells or macrophages (data not shown), thus suggesting that IL-36γ does not modulate, at least at the receptor level, the capacity for these cells to respond to further IL-36γ stimulation.

To establish whether IL-36γ can also stimulate inflammatory cytokine expression in human oral epithelial cells, OKF6 cells were treated with IL-36γ and the induction of specific inflammatory cytokines was measured. IL-36γ stimulated the expression of IL-8, CXCL1, CCL20, and IL-6 (Fig. 5A–D). The ability of IL-36γ to elicit inflammatory cytokine responses in human gingival epithelial cells was also examined. IL-36γ induced strong IL-8, CXCL1, CCL20, and IL-6 expression in TIGK (Fig. 5E–H). The ability of IL-36γ to stimulate IL-8 and CCL20 expression was also demonstrated by ELISA (Fig. 5I, 5J). Interestingly, the magnitudes of the cytokine responses elicited in TIGK were comparable to those in monocyte-derived dendritic cells (Fig. 4A–D). Taken together, this suggests that epithelium-derived IL-36γ may also be capable of functioning in an autocrine manner to amplify inflammatory responses.

The intracellular domains of IL-1R and IL-36R are similar, and hence IRAK1 is assumed to mediate signaling by IL-36R (37). To formally establish whether IRAK1 regulates the inflammatory responses elicited by IL-36γ in oral epithelial cells, gene silencing experiments were performed. The silencing of IRAK1 significantly inhibited the stimulation of IL-8, CXCL1, and CCL20 expression by IL-36γ in OKF6 cells (Fig. 6A–C), consistent with IRAK1 functioning downstream of IL-36R in oral epithelial cells. Given that IRAK1 can activate the transactivator function of IRF6 (22), we next sought to establish whether IL-36γ exerts its effects on oral epithelial cells by promoting the IRAK1-mediated activation of IRF6. In contrast to IRAK1, the silencing of IRF6 did not inhibit the stimulation of IL-8, CXCL1, or CCL20 expression by IL-36γ in OKF6 cells (Fig. 6D–F). Comparable findings were obtained with TIGK; specifically, the stimulation of IL-8, CXCL1, and CCL20 expression by IL-36γ was shown to be IRAK1-dependent and IRF6-independent (Fig. 6G–I). We recently demonstrated that IRAK1 mediates the TLR2-inducible expression of IL-8 in human oral epithelial cells independently of IRF6 (22). Consequently, we sought to establish whether IRAK1 can likewise activate the CCL20 promoter in an IRF6-independent manner. Gene promoter reporter assays revealed robust activation of the CCL20 promoter by IRAK1 (Fig. 6J). Consistent with the results from the silencing of IRF6 (Fig. 6I), the coexpression of IRF6 did not increase further the activation of the CCL20 promoter by IRAK1 (Fig. 6J). Thus, these data establish IRAK1 as a key regulator of the inflammatory effects of IL-36γ on human oral epithelial cells.

IL-17 is an important mediator of the adaptive immune response to P. gingivalis (38, 39). CCL20 can direct the recruitment of IL-17–producing Th17 cells to sites of infection (40), and thus the stimulation of CCL20 expression by IL-36γ may be important in vivo for the adaptive immune response to P. gingivalis. In addition to immune cells, IL-17 can also activate nonimmune cells (e.g., fibroblasts) (41). Therefore, the effects of IL-17 on the inflammatory properties of human oral epithelial cells were investigated. The stimulation of OKF6 cells with IL-17 resulted in the expression of the neutrophil chemokines IL-8 and CXCL1 (Fig. 7A, 7B), as well as the growth factor involved in neutrophil development and function, G-CSF (Fig. 7C). IL-17 also stimulated the expression of GM-CSF (Fig. 7D), a key regulator of the maturation and inflammatory properties of dendritic cells and macrophages (42). Likewise, IL-17 stimulated CCL20 expression (Fig. 7E). Oral epithelial cells also contribute to host defense by producing antimicrobial peptides (43), and IL-17 was demonstrated to stimulate hBD2 expression in these cells (Fig. 7F). Notably, IL-17 also stimulated the strong expression of IL-36γ (Fig. 7G). This effect was specific to IL-36γ, as the expression levels of IL-36α and IL-36β were not significantly increased (Fig. 7H, 7I). Thus, IL-17 can stimulate human oral epithelial cells to express a wide range of factors, including IL-36γ, that mediate host defense (Fig. 8).

IRFs are key regulators of host defense, as they help to shape appropriately the immune response to pathogens by providing signaling specificity to individual TLR pathways (3–7). In this study, IRF6 was shown to be important for the expression of IL-36γ by human oral epithelial cells in response to P. gingivalis. IL-36γ not only stimulated the expression of neutrophil- and Th17-attracting chemokines (IL-8/CXCL1 and CCL20, respectively) by human dendritic cells and macrophages, but it also stimulated their expression by oral epithelial cells. Moreover, the ability of IL-17 to stimulate oral epithelial cells to express IL-36γ raises the possibility that these two cytokines might form an inflammatory axis in the oral mucosa. Our findings therefore potentially position IRF6 as an important mediator of inflammation to P. gingivalis through its regulation of IL-36γ.

We recently demonstrated that IRF6 differentially regulates TLR2-elicited inflammatory cytokine expression by human oral epithelial cells (22). This prompted us to investigate whether IRF6 also regulates their cytokine responses to P. gingivalis. IRF6 was shown to be important for the expression of IL-36γ by human oral epithelial cells. Notably, IL-36γ was one of the most strongly stimulated inflammatory cytokines, consistent with it playing an important role in the host response to P. gingivalis. Interestingly, IL-36γ expression was selectively stimulated over other IL-36 family cytokines, namely IL-36α and IL-36β. The reason for this is unclear given that all three signal via IL-36R (35), although one possibility is that in vivo they might differ in bioavailability, which may allow individual IL-36 cytokines to elicit inflammatory responses that differ in strength and/or duration.

Our data suggest that IRF6 functions downstream of TLR2 to regulate IL-36γ expression in human oral epithelial cells. IRAK1, which promotes the activation of IRF6 by the TLR2 pathway in oral epithelial cells (22), was required for the stimulation of IL-36γ expression by P. gingivalis. The stimulation of IL-36γ expression by the TLR2 agonist FSL-1 was likewise inhibited by the gene silencing of IRAK1 and IRF6, and gene promoter assays demonstrated that IRAK1 and IRF6 can act in concert to transactivate the IL-36γ promoter. This suggests that TLR2 induces the expression of IL-36γ in oral epithelial cells by promoting the IRAK1-mediated activation of IRF6. IRAK1 also stimulated the activation of the IL-36γ promoter to a lesser degree independently of IRF6. NF-κB has previously been shown to functionally cooperate with IRFs to regulate gene expression (44, 45). Therefore, optimal stimulation of IL-36γ gene expression may require cooperation between IRF6 and NF-κB.

Little is known about the specific roles of IL-36 cytokines in eliciting inflammation. In vitro studies have demonstrated that IL-36 cytokines had potent inflammatory effects on mouse bone marrow–derived dendritic cells (34). Similar findings were obtained in the present study with human monocyte-derived dendritic cells. IL-36γ also exerted, albeit weaker, inflammatory effects on monocyte-derived macrophages. In particular, IL-36γ stimulated the expression of IL-8 and CXCL1, whose primary functions are to direct the recruitment of neutrophils to sites of infection and/or injury. Thus, its ability to stimulate IL-8 and CXCL1 expression potentially positions IL-36γ as a key regulator of neutrophil recruitment in response to infection. Indeed, the transgenic expression of IL-36α in the epidermis of mice caused cutaneous inflammation that was characterized by the infiltration of neutrophils, along with mononuclear cells (46).

IL-36γ also stimulated human monocyte-derived dendritic cells and macrophages to express CCL20, a key regulator of the recruitment of IL-17–producing Th17 cells (40). Th17 cells are important mediators of host defense against bacteria, particularly bacteria that colonize mucosal surfaces (47, 48). IL-17 can stimulate innate immune cells and nonimmune cells (e.g., fibroblasts) to express inflammatory factors. In this study, IL-17 was shown to exert inflammatory effects on human oral epithelial cells (e.g., OKF6 cells). In particular, IL-17 stimulated the expression of factors that promote neutrophil recruitment (IL-8 and CXCL1) and survival (G-CSF). IL-17 also stimulated the expression of factors that can further polarize Th17 cell responses, including CCL20 and GM-CSF. GM-CSF promotes the differentiation of dendritic cells and macrophages (42), which are major sources of IL-23, an important regulator of the survival and proliferation of Th17 cells (49). Notably, IL-17 also stimulated the strong expression of IL-36γ. This raises the possibility that by stimulating CCL20 expression, IL-36γ may form an inflammatory axis with IL-17 in response to P. gingivalis infection (Fig. 8). However, in vivo studies will be required to demonstrate the formation and importance of such an axis in the oral mucosa.

Interestingly, IL-36γ also stimulated the expression of IL-8, CXCL1, and CCL20 by human oral epithelial cells, with particularly strong responses elicited in gingival epithelial cells (e.g., TIGK). IL-36γ might therefore act in both a paracrine and autocrine manner. The inflammatory responses elicited by IL-36γ in oral epithelial cells were mediated by IRAK1 but independent of IRF6. Consequently, the strong upregulation of IRF6 in response to P. gingivalis might provide a mechanism for augmenting IL-36γ expression while maintaining tight control over the inflammatory effects IL-36γ may have on the oral epithelium.

P. gingivalis causes periodontitis by dysregulating local immunity, resulting in dysbiosis and chronic inflammation (8–10). Neutrophils are critical for protection against P. gingivalis–induced tissue damage and alveolar bone resorption (16). However, excessive numbers of neutrophils can also cause tissue breakdown (17). Therefore, the factors controlling the activation and recruitment of neutrophils into periodontal tissues (e.g., IL-8) must be tightly regulated to maintain homeostasis. Given its ability to stimulate the expression of neutrophil chemokines (e.g., IL-8 and CXCL1) by innate immune and oral epithelial cells, IL-36γ might be an important factor in driving the neutrophil response in periodontitis.

T cell responses are also dysregulated in periodontitis (10). P. gingivalis has been shown to inhibit, including in human oral epithelial cells (e.g., TIGK), the expression of chemokines that promote the recruitment of Th1 cells (50). Indeed, Th17 cells are the dominant T cell subset in human periodontal lesions (21), and IL-17 promoted alveolar bone loss in experimental periodontitis, most likely by stimulating the expression of osteoclastogenic factors (e.g., RANK ligand) (17). The dysregulation of an inflammatory axis between IL-36 and IL-17 in the skin appears to be important in the pathogenesis of psoriasis (51–53). Consequently, and by analogy, IL-36γ could potentially promote periodontitis due to its ability to also stimulate the expression of CCL20 by different cell types. Interestingly, high levels of IL-36γ and CCL20 have recently been shown to inhibit oral epithelial cell migration and wound healing (54). Therefore, a dysregulated IL-36γ response in the context of P. gingivalis infection might also compromise the ability of the host to maintain the integrity of the oral epithelium.

In conclusion, our findings position IRF6 as an important mediator of the inflammatory response of oral epithelial cells to P. gingivalis through its regulation of IL-36γ. Therefore, future studies should aim to define the nature of the relationship between IL-36γ and IL-17 in the oral mucosa, as well as its role in the immune response to P. gingivalis during periodontitis.

We thank Richard Lamont (University of Louisville, Louisville, KY) for providing the TIGK cell line.

The authors have no financial conflicts of interest.