Discussion
In the present study, we demonstrated that OVA-induced allergic airway inflammation occurred in parallel with increased-expression of TLR2, activation of NLRP3 inflammasome and decreased biosynthesis of melatonin. Deletion of TLR2 effectively alleviated allergic airway inflammation and concomitantly inhibited the activation of NLRP3 inflammasome as well as restored the level of melatonin. Furthermore, exogenous addition of melatonin to OVA-challenged WT mice pronouncedly ameliorated airway inflammation, decreased OVA-induced TLR2 expression and NLRP3 activity, and increased melatonin biosynthesis to similar level as that in OVA-challenged TLR2−/− mice. However, although luzindole significantly reduced the expression of AANAT and ASMT and subsequent level of melatonin in OVA-challenged TLR2−/− mice, it exhibited null effect on OVA-induced airway inflammation and activation of NLRP3, it only aggravated allergen-induced mucus hyper-secretion. These results are the first to show the existence of TLR2-melatonin feedback loop in allergic airway diseases, which regulates NLRP3 inflammasome activity and may represent a mechanism underlying the initiation and persistence of airway inflammation.
An allergic airway disease is usually dominated by a strong Th2 response which can be redirected to Th1 response by regulation of TLRs signaling (4, 31, 32), making TLRs attractive therapeutic targets. During the last decade, accumulating evidence has been focused on the role of TLRs in the pathogenesis of airway inflammatory diseases such as asthma and the possibility of using TLRs-based therapies for asthma (33, 34). To date, TLR2 has been considered as the most relevant to the onset of asthma. Asthmatic patients who ultimately die have increased expression of TLR2 (2), activation of TLR2 promotes Th2-biased immune responses, which may be correlated with the imbalance of Th1/Th2 in asthma (3, 4). In line with these data, our present study demonstrated a significant increase of TLR2 expression in WT mice post OVA challenge, and this increase was accounted from its induced-expression on various cell types. Previous studies also indicate its wide spectral expression (35–37), and reports from our group and others have indicated that TLR2 on macrophages (38), group 2 innate lymphoid cells (39) and epithelial cells (40) may contribute to allergic airway inflammation. Furthermore, the increased expression of TLR2 was accompanied with lung inflammation exacerbation in WT mice post OVA challenge in this study. However, such OVA-induced airway inflammation, including leukocyte recruitment to bronchial, mucus metaplasia in the airways, total number or composition of the BALF cellularity, the level of OVA-specific IgE, as well as IL-4 and IL-13, was notably alleviated in OVA-challenged TLR2−/− mice. Based on these data, TLR2 was considered to mediate allergic airway inflammation, and targeting TLR2 may have therapeutic benefit in allergic airway diseases.
However, the mechanism underlying TLR2 mediation of allergic airway inflammation is still unknown. NLRP3 inflammasome is one of the most extensively characterized NLRs due to its relevance in human inflammatory disorders such as asthma. Invading pathogens including viral or bacterial which commonly associated with asthma exacerbation, have been shown to trigger NLRP3 activation (5, 41). IL-1β is increased in the serum and BALF of human asthmatics, and administration of IL-1β induces airway hyper-reactivity (12, 13). Increased expressions of NLRP3, caspase-1 and IL-1β are found in macrophages as well as neutrophils in sputum of neutrophilic asthma (14). Consistent with these studies, our present study showed that in WT mice, OVA challenge substantially increased protein expression of NLRP3, mature IL-1β and caspase 1(p20) in lung tissues, markedly elevated the levels of NLRP3-associated IL-1β and IL-18 in BALF. However, this effect was only seen in WT mice, but not in TLR2−/− mice. The interaction between TLR2 and NLRP3 inflammasome has been also reported in various cell types. In monocytes (42), macrophages (43, 44), bone marrow-derived DCs (45) and several different cell lines (46, 47), lacking TLR2 failed to upregulate NLRP3 inflammasome as well as its substrate IL-1β. Therefore, our results have suggested that activation of NLRP3 inflammasome induced by OVA requires TLR2 signaling, thus TLR2 may mediate allergic airway inflammation through regulating NLRP3 inflammasome activity.
A question arises now is, by which TLR2 crosstalks with NLRP3 inflammasome and consequently mediates allergic airway inflammation. Melatonin has been reported to exert important immune-modulating effects in allergic airway diseases (19, 20, 48, 49), and the level of melatonin in saliva or serum of asthma patients were significantly lower than those in healthy controls (21, 50, 51). Particularly, it has been shown that TLR9 signaling regulates endogenous melatonin synthesis in allergic airway inflammation (22). Considering TLR2 and TLR9 are structurally similar, we asked whether TLR2 regulates endogenous melatonin synthesis and thereby modulated NLRP3 inflammasome activity. Indeed, the current study demonstrated that TLR2 protein expression was remarkably increased accompanying with decreased expression of melatonin synthetase ASMT and lower level of melatonin in lung homogenate in WT mice post OVA challenge, while deletion of TLR2 significantly rescued melatonin biosynthesis. Since it have been reported that C57BL/6 mice were pineal melatonin-deficient mice (52, 53) or produce much lower melatonin in their pineal gland (17, 54), suggested an extra-pineal melatonin synthesis might compensate for the pineal deficiency (17). Actually, there are results reporting that several extra-pineal tissues, such as thymus, spleen and skin, in C57BL/6 mice can synthesize melatonin (17, 55), our current study further found that another extra-pineal tissue, i.e., lung synthesized melatonin, and this process maybe modulated by TLR2 signal.
On the other hand, melatonin has been evidenced to significantly inhibit airway inflammation (49, 56), and suppress TLR3/4-mediated inflammation in liver injury (23). Most notably, NLRP3 is a novel molecular target for melatonin in murine model of septic response, liver injury and acute lung injury (24–26). Consistent with these findings in animals, melatonin also has been shown to exert inhibitory effect on TLRs including TLR3/4/9 signaling in macrophage (57–59) and NLRP3 inflammasome in epithelial cells (60). Therefore, we asked whether endogenous melatonin, which was regulated by TLR2 signal, alleviated allergic airway inflammation through directly suppressing NLRP3 inflammasome activity or feedback controlling TLR2-NLRP3 signal. Our current study showed that treatment of melatonin notably alleviated OVA-induced airway inflammation in WT mice, which was consistent with previous findings (19, 56). However, we extended previous observations by showing that melatonin treatment strongly inhibited OVA-induced protein expressions of TLR2, NLRP3, mature IL-1β and caspase1(p20), as well as lowered the levels of NLRP3-associated IL-1β and IL-18 in BALF in WT mice, suggesting that melatonin mitigates allergic airway inflammation by inhibiting TLR2 and NLRP3 inflammasome activity. Considering the data shown above that TLR2 signaling regulated melatonin synthesis, we speculated that a TLR2-melatonin feedback loop may exist in allergic airway disease, and melatonin elicits its effect on activation of NLRP3 inflammasome through TLR2 signal. Additionally, we found that exogenous addition of melatonin further increased protein expression of ASMT, as well as elevated the level of 5-HT in BALF and melatonin in lung homogenate in OVA-challenged WT mice. These interesting data suggested that the proven effect of exogenous melatonin in the resolution of inflammation was paralleled by the effect of endogenous synthesized melatonin.
Among the three melatonin receptors, the anti-inflammatory effect of melatonin is reported to be mainly mediated by MT2 melatonin receptor (61). Luzindole is considered as a non-selective MT1/MT2 receptor antagonist, showing 15–26 times higher affinity for MT2 receptor, which pharmacologically blocks the effect of melatonin (62). Therefore, luzindole was used as indicated (28) to further confirm that melatonin elicited its effect on NLRP3 inflammasome activity via TLR2 signal. Our results showed that although administration of luzindole to OVA-challenged TLR2−/− mice significantly reduced the expression of melatonin synthetase AANAT and ASMT, and subsequent level of melatonin, it exhibited null effects on the level of 5-HT in BALF, OVA-induced leukocyte infiltration, the level of IgE, Th2 cytokines production, and NLRP3 inflammasome activity, it only aggravated allergen-induced mucus hyper-secretion. To our knowledge, this is the first study showing that luzindole inhibited lung melatonin synthesis, which may be due to the blockage of 5-HT conversion to melatonin. These combined data suggested that the effect of melatonin on airway inflammation and NLRP3 inflammasome activity requires TLR2 signaling, and melatonin which is modulated by TLR2 signal feedback regulates TLR2 signaling, and subsequently regulates NLRP3 inflammasome activity and airway inflammation. However, the inhibitory effect of melatonin on airway remodeling characterized by mucus hyper-secretion may not require TLR2 signal, because luzindole still deteriorates mucus production even in the absence of TLR2 signal.
In conclusion, the present study demonstrates that a TLR2-melatonin feedback loop regulates the activation of NLRP3 inflammasome in allergic airway inflammation, and suppression of melatonin synthesis by TLR2 activation in turn results in the loss of its inhibitory effect on the TLR2 signaling (Figure 7). This important endogenous regulatory feedback loop may drive the onset of allergic airway inflammation, and melatonin may be a promising therapeutic medicine for airway inflammatory disease such as asthma. The present study used OVA model, which has been the most widely used pre-clinical allergic asthma model and recapitulates many of the hallmarks of allergic asthma in humans, however, bio-models offer other allergen-induced allergic model, such as house dust mite (HDM), Alternaria, papain and IL-33, with each mimicking the major features of human asthma. When OVA, HDM, or Alternaria was given via inhalation or intranasally, each of them can be recognized and sensed by TLRs reportedly, especially TLR2 (40). Similarly, intranasally given IL-33 is reported can induce direct stimulation of TLR2 on ILC2s (39). However, papain, which is homologous of HDM-derived Der p1, activates innate immune responses in a manner distinct from that of PAPMs and can induce production of an alarmin IL-33 (63). We hypothesize our findings may be also applicable to other allergens–induced allergic airway inflammation models in which TLR2 is activated by allergen. However, the true fact needs future study to discover.
Figure 7  Proposed schematic of the mechanism that TLR2-melatonin feedback loop regulates the activation of NLRP3 inflammasome in allergic airway inflammation. After TLR2 is activated by allergen, it subsequently induces NLRP3 inflammasome activation. Meanwhile, TLR2 inhibits the expression of melatonin synthetase ASMT, accordingly blocks 5-HT converting to melatonin, ultimately leading to the decreased level of melatonin. This inhibition of melatonin synthesis results in the loss of its inhibitory effect on the TLR2 signaling. The result of the cycle is the persistent of allergic airway inflammation.