Melatonin suppresses TLR9-triggered proinflammatory cytokine production in macrophages by inhibiting ERK1/2 and AKT activation

Toll-like receptor (TLR) signaling plays major roles in innate immune response in macrophages. Melatonin regulates TLR3- and TLR4-mediated innate immune responses in macrophages. However, it remains unknown whether melatonin regulates TLR9-mediated innate immune responses in macrophages. Here we demonstrated that melatonin suppressed TLR9 ligand-induced proinflammatory cytokines mRNA and protein production in peritoneal macrophages without interrupting the viability of peritoneal macrophages. Using a melatonin membrane receptors MT1/MT2 antagonist luzindole, we found that MT1 and MT2 were dispensable for melatonin’s inhibitory effects on TLR9-mediated proinflammatory cytokines production, even though melatonin upregulated mRNA expression of MT1 and MT2 in macrophages. Furthermore, melatonin did not affect mRNA expressions of TLR9 and MyD88 but attenuated TLR9 ligand-induced ERK1/2 and AKT phosphorylation without affecting p38 and NF-κB p65 phosphorylation. Also, melatonin inhibited TLR9-mediated proinflammatory cytokines production in vivo. Taken together, our results demonstrate that melatonin suppresses TLR9-triggered proinflammatory cytokines production in macrophages via melatonin membrane receptor-independent manners and probably through inhibiting ERK1/2 and AKT activation, which further elucidates the roles of melatonin in regulating TLR-mediated innate immune responses in macrophages.

signaling molecules or substances that fine-tune TLR pathways have been receiving increased attention in recent years. Many molecules have been found to negatively regulate TLR signaling by different mechanisms. These molecules include cell membrane or intracellular molecules, and even intranuclear epigenetic modifiers 2,[4][5][6][7] . In a previous study, Stk38 protein kinase was found to preferentially inhibit TLR9-activated inflammatory responses by promoting MEKK2 ubiquitination in macrophages 4 . Furthermore, various substances have been found to regulate TLR signaling, such as micheliolide, vasoactive intestinal peptide, and melatonin 8-12 . Melatonin is mainly produced by the pineal gland, but extrapineal tissues, such as gastrointestinal tract, retina, spleen, liver and kidney, and endothelial cells or immune cells can also produce melatonin 13,14 . Melatonin has a wide range of functions, which are divided into chronobiotic and nonchronobiotic ones 13 . The chronobiotic effects regulate the circadian rhythms and seasonal changes, and are mediated by melatonin in the circulation due to nocturnal pineal synsthesis, whereas melatonin produced by other cells, such as gastrointestinal and immune cells, is independent of the light/dark cycle and exerts nonchronobiotic effects, including antioxidant, oncostatic, antiaging, and immunomodulatory effects 13 .
The effects of melatonin in the body are mediated mainly by two pathways: melatonin receptor-dependent pathway and melatonin receptor-independent pathway 14,15 . Melatonin receptors include membrane receptors MT1/MT2, and nuclear receptors RORα/RZR 14,15 . Increasing evidences demonstrate that melatonin plays a fundamental role in regulating immune response, as well as infection, sepsis, inflammation, and autoimmunity 13,[15][16][17][18] . Recently, more and more studies show that melatonin modulates the functions of innate immune cells such as macrophages, monocytes, natural killer cells and neutrophils 13,17 . In particular, melatonin regulates TLR3 and TLR4 signalings in macrophages [8][9][10]13,17 . However, it remains to be elucidated whether melatonin regulates TLR9-mediated innate immune responses in macrophages.
In this study, we observed that melatonin suppressed TLR9 ligand-induced proinflammatory cytokines production in primary peritoneal macrophages. Moreover, its suppressive effects on TLR9 signaling did not depend on melatonin membrane receptors MT1 and MT2, but probably through inhibiting ERK1/2 and AKT activation without affecting mRNA expressions of TLR9 and MyD88. Furthermore, melatonin also decreased TLR9-mediated TNF-α and IL-6 production in mouse serum. Overall, the results of our study clarify the inhibitory effects of melatonin on TLR9-mediated innate immune responses in macrophages, probably through suppressing ERK1/2 and AKT activation.
Then we assessed the effects of melatonin on protein expression of inflammatory cytokines in macrophage induced by CpG-ODN. As shown in Fig. 2A-D, 2 μg/ml CpG-ODN significantly induced the production of TNF-α, IL-6, IL-12 p70 and IL-10 in macrophages, whereas, as expected, it induced no production of INF-β (Data not shown). Melatonin effectively suppressed the production of TNF-α, IL-6 and IL-12 p70 induced by CpG-ODN in a dose-dependent manner ( Fig. 2A-C), whereas it exerted no notable suppression on anti-inflammatory IL-10 production (Fig. 2D). Of note, melatonin alone could inhibit the constitutive IL-10 production in macrophages (Fig. 2D). The results above indicate that melatonin could markedly suppress TLR9-mediated inflammatory responses in macrophages.
Notably, melatonin inhibited IL-10 mRNA expression but not IL-10 protein expression in CpG-ODN-stimulated macrophages (Figs 1E and 2D), indicating the possible posttranscriptional regulation by melatonin. The result also showed that melatonin suppressed TLR9-mediated proinflammatory cytokines production in macrophages not through upregulation of anti-inflammatory IL-10.
To exclude the possibility that melatonin may affect cell viability of macrophages, we used Annexin V/7AAD staining to examine the apoptosis of primary peritoneal macrophages after melatonin and/or CpG-ODN treatment. As shown in Fig. 3A,B, 1 mM melatonin did not increase the apoptotic cell percentage of macrophages irrespective of CpG-ODN stimulation, indicating that melatonin had no cytotoxic effects on macrophages and the suppression of TLR9-mediated proinflammatory cytokines production could not be attributable to a direct cytotoxic effect by melatonin.
Melatonin membrane receptors MT1 and MT2 are dispensable for the suppressive effects of melatonin. Melatonin exerts a variety of physiological and pharmacological effects on the immune system through activation of two melatonin membrane receptors MT1 and MT2, which belong to G protein-coupled receptors (GPCRs) 19 . We found that peritoneal macrophages expressed melatonin membrane receptors MT1 and MT2, and melatonin treatment upregulated mRNA expressions of MT1 and MT2 in macrophages, whereas CpG-ODN stimulation did not affect mRNA expressions of MT1 and MT2 (Fig. 4A,B). To investigate whether MT1 and MT2 were required for the suppressive effects of melatonin on TLR9-mediated inflammatory cytokines production, luzindole, a nonselective antagonist of both MT1 and MT2, was used. As shown in Fig TLR9-mediated TNF-α and IL-6 production. These results demonstrated that melatonin membrane receptors MT1 and MT2 were dispensable for the suppressive effects of melatonin on TLR9-mediated inflammatory cytokines production in macrophages.

Melatonin inhibits TLR9-mediated proinflammatory cytokines production in vivo.
To explore the role of melatonin in regulating TLR9 signaling in macrophages in vivo, C57BL/6J mice were intraperitoneally injected with CpG-ODN in the presence or absence of melatonin pretreatment after mobilizing peritoneal macrophages through thioglycollate (TG) treatment, and then the levels of TNF-α and IL-6 in mouse serum were detected. As shown in Fig. 8A,B, CpG-ODN induced high levels of TNF-α and IL-6 in serum and melatonin pretreatment significantly inhibited CpG-ODN-induced TNF-α and IL-6 production in serum, indicating that melatonin inhibited TLR9-mediated proinflammatory cytokines production in vivo.

Discussion
Increasing evidences show that melatonin not only exerts an immunosuppressive effect on macrophages but also exerts an immunostimulatory effect on macrophages, possibly depending on the manners of melatonin administration or the dose of melatonin 13 . For example, it has been reported that melatonin stimulated a significant increase of IL-12 production in human synovial macrophages and THP-1 cells, however, a significant reduction of IL-12 production was observed when macrophages were previously activated with LPS and then treated with melatonin 23 . Regarding TLRs signaling, it has been demonstrated that melatonin mainly exerts a suppressive effect on TLRs signaling, including TLR3 and TLR4 signalings [8][9][10]13 . In the present study, we demonstrated for the first   time that melatonin suppressed TLR9-mediated inflammatory responses in macrophages, indicating that melatonin was a negative regulator of TLR9-mediated innate immunity in macrophages. Although melatonin upregulated melatonin membrane receptors MT1 and MT2 mRNA expression in macrophages, MT1 and MT2 were dispensable for the suppressive effects of melatonin on TLR9-mediated inflammatory responses. Furthermore,  melatonin did not affect TLR9 and MyD88 mRNA expression, but possibly attenuated ERK1/2 and AKT activation to suppress TLR9 signaling.
After stimulation with TLR ligands such as LPS, poly I:C or CpG-ODN, activated macrophages produce various cytokines including TNF-α, IL-6, IL-12, IL-1β, IL-10, and type 1 interferon including IFN-α and IFN-β. TLR9 ligand CpG-ODN induces macrophages and dendritic cells to secret TNF-α, IL-6, IL-12, IL-10 and IFN-α/β (mainly produced by plasmacytoid dendritic cells) 24,25 . The present study demonstrated that melatonin suppressed the mRNA and protein expression of TNF-α, IL-6 and IL-12 p70 induced by CpG-ODN, indicating that melatonin could suppress TLR9-mediated proinflammatory cytokines production in macrophages. Similarly, previous studies showed that melatonin suppressed LPS-induced TNF-α, IL-6, IL-1β, and IL-8 expression in RAW264.7 cells 9,10 . IL-10 induced by CpG-ODN, an anti-inflammatory cytokine, could suppress IL-12 p70 production in macrophages 26 . Our present study demonstrated that melatonin did not suppress or promote IL-10 protein production induced by CpG-ODN, which is different from another study showing that melatonin suppressed LPS-induced IL-10 expression 10 . This indicates that IL-10 is not involved in the inhibitory effects of melatonin on TLR9-mediated inflammatory responses in macrophages.
The molecular mechanisms of action of melatonin mainly include the following four models: (i) via its interaction with membrane receptors such as MT1 and MT2 (at low concentration), (ii) by binding to its nuclear receptors such as RORα and RZR (at low concentration), (iii) via an interaction with cytoplasmic proteins such as calmodulin and hydroquinone (at high concentration), and (iv) via directly, receptor-independent actions such as scavenging ROS/RNS 13,15,27 . It has been shown that the effects of melatonin on immune system are mediated through melatonin receptor-dependent or -independent mechanisms. In multiple sclerosis (MS), melatonin blocks Th17 differentiation and boosts the generation of IL-10-secreting Treg cells by an MT1-dependent mechanism 28,29 . Melatonin inhibits the infectivity of Leishmania by murine peritoneal macrophages and enhances the phagocytosis of zymosan by human colostrum mononuclear cells, both of which are attenuated by luzindole (an antagonist of MT1 and MT2), indicating the involvement of melatonin membrane receptors in these effects 30,31 . The present study showed that luzindole did not reverse the inhibitory effects of melatonin on TLR9-mediated inflammatory responses in macrophages, so melatonin exerts its inhibitory effects on TLR9 signaling through melatonin membrane receptor-independent way.
Given that melatonin at high concentration also exerts its effects through receptor-independent manners in addition to its interaction with either melatonin membrane or nuclear receptors 13,15,27 , in the present study, high-concentration melatonin (reaching the level of millimole) probably directly regulated the expression or activation of TLR9 signal molecules, similar to its mechanisms in regulating TLR3 or TLR4 signaling in macrophages [8][9][10] . Melatonin decreased TLR-3-mediated TNF-α and iNOS mRNA expression in respiratory syncytial virus-infected RAW264.7 cells via inhibition of NF-κB activation without influencing TLR3 and MyD88 mRNA expression 8 . In LPS-stimulated RAW264.7 cells, melatonin attenuated the expression of MyD88 and the activation of NF-κB, and also inhibited TLR4-mediated AKT phosphorylation and attenuated the elevation of IRF3, which was involved in TLR4-mediated TRIF-dependent signaling pathway 10 . In another study, melatonin inhibited LPS induced production of nitric oxide and IL-6 in RAW264.7 cells by suppressing NF-κB and STAT1 activation 9 . In the present study, melatonin inhibited CpG-ODN-induced ERK1/2 and AKT activation without affecting NF-κB and p38 activation in macrophages. Furthermore, melatonin did not affect TLR9 and MyD88 mRNA expression. Hence, melatonin might suppress TLR9-mediated innate immune responses through inhibiting ERK1/2 and AKT activation. This study, together with other previous studies, indicate that melatonin regulates TLR signaling in macrophages through different mechanisms, dependent on different TLR subtypes and their ligands [8][9][10] . It remains to be further investigated how melatonin regulates TLR9-mediated ERK1/2 and AKT activation in macrophages. In our previous study, MEKK2 knockdown (specifically inhibiting ERK1/2 activation) significantly suppressed CpG-ODN-induced TNF-α and IL-6 production in macrophages 4 . In addition, other studies also showed that wortmannin, a PI-3K inhibitor, inhibited the production of IL-12 induced by CpG DNA in bone marrow derived dendritic cells 32 , and another PI-3K inhibitor, LY294002, also reduced TNF-α production induced by CpG DNA in macrophages 33 . These studies further support that ERK1/2 and AKT activation is important for TLR9-mediated inflammatory cytokines production in macrophages.
There is only one published research reporting on the relationship between melatonin and TLR9 signaling. This study showed that JNK-enhanced TLR9 signal pathway mediated allergic airway inflammation through suppressing melatonin biosynthesis 34 . However, there is no research addressing the regulatory function of melatonin on TLR9 signaling. Our present study showed for the first time that melatonin regulated TLR9-mediated innate immune responses and TLR9 signaling pathways.
TLR9-mediated innate immunity and inflammation plays important roles in infectious diseases, autoimmune diseases and cancer 25,35 . Previous studies showed that polymicrobial sepsis in mouse experimental peritonitis model induced by caecal ligation and puncture (CLP) was TLR9-dependent 36,37 . It has been demonstrated that melatonin protects mice against septic shock or septic organs injury in CLP sepsis models through different mechanisms: these include antioxidative effects, anti-nitrosative effects, protective effects against mitochondrial dysfunction and anti-inflammatory effects 18,38-43 . Hence, it is possible that melatonin also provides protections against sepsis through suppressing TLR9-mediated inflammation, which requires to be further investigated. Furthermore, our present study demonstrated that in vivo administration of melatonin exerted notable suppressive effects on TLR9-mediated inflammatory cytokines production in mice, further raising this possibility.
Taken together, our study demonstrated that melatonin suppressed TLR9-triggered proinflammatory TNF-α, IL-6 and IL-12 p70 production in macrophages not through melatonin membrane receptors-dependent manners but probably through inhibiting TLR9-mediated ERK1/2 and AKT activation. These results proved that melatonin was a negative regulator of TLR9-mediated innate immunity in macrophages.
Cell culture and treatments. Thioglycollate-elicited mouse peritoneal macrophages were isolated and cultured as previously described 4 . C57BL/6J mice (6-8 weeks old) were intraperitoneally injected with 1 ml of 3% sodium thioglycollate (Merck) solution per mouse. After 3 days, cells in the abdominal cavity were collected by lavaging with 10-15 ml FBS-free DMEM and centrifugated. Cells were cultured and maintained in DMEM with 10% (vol/vol) FBS for 24 hours. After removing the floating cells, adhered peritoneal macrophages were subjected for further experiments. The cells were pre-treated with various concentrations of melatonin (stock concentration 200 mM, dissolved in ethanol) for 1 hour and then stimulated with 2 μg/ml CpG-ODN for indicated times (Supplementary Fig. 1). In some experiments, the cells were incubated with 10 μM luzindole for 1 hour before pretreatment with melatonin ( Supplementary Fig. 1C).
Detection of cytokine production. The concentrations of TNF-α, IL-6, IL-12p70 and IL-10 in culture supernatants or serum were measured with ELISA kits from R&D systems (Minneapolis, MN) according to the manufacturer's instructions.
Detection of cell apoptosis. After treatment with melatonin and/or CpG-ODN ( Supplementary Fig. 1B), peritoneal macrophages were harvested and stained with Annexin V-APC and 7AAD according to the manufacturer's instructions. Then the apoptosis of macrophages was assayed by flow cytometry (LSR II, BD biosciences) and the data were analyzed with FlowJo Version 5.7.2 software (Tree Star).
Quantitative reverse transcription-PCR. Total RNA was extracted with RNA extraction kit (RNAfast200, Fastagen Biotech Co., China) and reversed-transcripted with reverse transcriptase M-MLX (Takara) following the manufacturer's instructions. Real-time PCR was performed with a SYBR RT-PCR kit (Takara) on LightCycler system Figure 8. Melatonin inhibits TLR9-mediated proinflammatory cytokines production in vivo. C57BL6/J mice were intraperitoneally injected with thioglycollate (TG) to elicite macrophages on day 0 and then received 5 mg/ kg melatonin (MT) or ethanol (E) on day 3 and day 4. On day 4, mice were further stimulated by intraperitoneal injection of 50 μg CpG-ODN two hours after melatonin pretreatment. TNF-α (A) and IL-6 (B) in serum of mice in various groups (n = 5) were assayed by ELISA 2 hours after CpG-ODN injection. Error bars represent s.d. **P < 0.01 (ANOVA).