Introduction
Chronic rhinosinusitis (CRS), affecting 14% of adults in the United States and 8% in China [
1,
2], is a chronic mucosal inflammation mediated by innate and adaptive immune cells and mediators in nasal sinus. CRS is typically classified into two types of clinical phenotypes: chronic rhinosinusitis without nasal polyps (CRSsNP) and chronic rhinosinusitis with nasal polyps (CRSwNP) [
1]. Although emerging evidences have indicated that different regions of the world might have different endotypes of CRS [
3‐
6], a T helper (Th) 2-predominant eosinophilic endotype in NP has been well documented in the western world [
3,
7‐
9], and notably has increased over past 10 years in oriental countries, such as Thailand [
10] and Korea [
11]. However, the cellular and molecular mechanisms driving the Th2-predominant immune response in NP remain unclear.
With potent antigen presenting capacity, dendritic cells (DCs) are a heterogeneous population of cells, consisting of multiple subtypes. It is clear now that different DC subsets perform different tasks. For instance, some subsets of DCs are better at cross-presentation of antigen to CD8 T cells on major histocompatibility complex (MHC)- I molecules and others better at presenting endocytosed antigen to CD4 T cells on MHC-II molecules [
12,
13]. On the other hand, DCs can be divided into functional subsets according to their polarizing function on naïve T cells, such as a Th1-skewing subset and a Th2- skewing subset, which usually are both differentiated from myeloid DCs (mDCs) [
14]. In NP, it has been reported that mDCs are increased in NP tissues [
15,
16], and two distinct DC subsets, OX40L/PDL1
+ DCs with Th2-cell-priming ability and low OX40L/PDL1-expressing DCs with Th1/17-cell-inducing ability, are associated with forming eosinophilic and non-eosinophilic endotype of NP respectively [
17], indicating an important role DCs play in the modulation of T cell response in NP. However, the molecular factors in regulating functional DC subsets to induce Th2 inflammation in NP have not been fully understood.
Thymic stromal lymphopoietin (TSLP), IL-25, and IL-33 are three cytokines predominantly produced by epithelial cells at mucosal surfaces in response to a wide range of environmental stimuli, and their expression during type 2 diseases, including NP, in humans has been widely documented [
18‐
26]. It is now evident that these cytokines play an important role in initiating type 2 immunity in mammals by activating resident mucosal group 2 innate lymphoid cells (ILC2s) to produce Th2-type cytokines (IL-5 and IL-13) and skewing CD4+ T cells toward Th2 differentiation [
27]. Furthermore, accumulating evidence show that these cytokines can also activate DCs to induce Th2-type immune responses. TSLP induces DC activation in nasal mucosa and enhances their capacity to initiate Th2 responses [
28]. In a mouse model of house dust mite-induced airway inflammation, IL-25 was shown to promote Th2 and Th9 inflammation in lungs by targeting DCs [
29]. Upon IL-33 exposure, DCs exhibited increased expression of CD40 and OX40 ligand (OX40L) and became very potent at inducing Th2 responses [
30]. In addition, we recently reported that mDCs in peripheral blood mononuclear cells (PBMCs) in atopic subjects expressed higher levels of IL-17RB (IL-25 receptor), ST2 (IL-33 receptor) and TSLPR (TSLP receptor) than those of non-atopic subjects [
31]. Of note, DCs generated from peripheral blood monocytes of atopic subjects with GM-CSF and IL-4 in vitro also expressed higher level of IL-17RB and could enhance a Th2-type response, suggesting that DCs expressing IL-17RB might be a Th2-skewing subset [
31]. However, whether mDCs in NP have similar receptor expression patterns as those seen in atopic subjects remains unknown.
The aim of this study was to explore the phenotypic characteristics of mDCs, especially the expression of IL-17RB, ST2 and TSLPR, and their potential contribution to the Th2 inflammation and disease severity in NP.
Discussion
The data presented in our study demonstrate that mDCs (CD1c
+ DCs) accumulated in NP tissues expressed increased IL-17RB and ST2, which were positively correlated with their counterpart ligands IL-25 and IL-33 mRNA levels, as well as IL-5 mRNA level and eosinophil numbers in NP tissues. Given both IL-25 and IL-33 are predominantly produced by epithelial cells and have been shown to play an important role in the initiation and development of type 2 immune response [
27], our data provide evidence for a link between mDCs and IL-25- and IL-33-induced type 2 responses and eosinophilic inflammation in NP.
NP has been known to be a Th2-skewed eosinophilic inflammation in the nasal sinus mucosa. T helper type 2-associated cytokines IL-4, IL-5 and IL-13 are involved in the pathogenesis of the eosinophilic airway diseases such as allergic asthma, allergic rhinitis and NP [
37,
38]. IL-25, IL-33 and TSLP have recently been considered as potential therapeutic targets because of their important roles in initiating the type 2 inflammation [
39]. In the present study, we found significant elevation in transcript levels of IL-25, IL-33 and TSLP, as well as IL-4, IL-5 and IL-13, but not IFN-γ and IL-17A, in patients with NP when compared with control subjects, which are largely consistent with previous reports [
23,
25,
40], confirming the Th2-skewed innate and adaptive immune responses in the pathogenesis of NP.
DCs, particularly in the airway, play a critical role in the induction of peripheral tolerance and maintenance of immune homeostasis [
41]. Therefore, DCs in the airway require some degree of activation to exert their function. This might explain why no difference in the percentage of activated mDCs in TMCs between patients with NP and control subjects was observed in the present study. In addition, we found PDL1 expression on CD86
+CD1c
+ DCs was lower in NP tissues than control tissues and negatively correlated with IL-5 mRNA level, this is in line with a recent study by Kortekaas et, al [
42], showing that the PDL1 mRNA level was lower, but PD1 was higher and positively correlated with IL-5 mRNA level, in NP tissue. Furthermore, we found increased number of activated mDCs in PBMCs, but not in polyp tissues, in patients with NP, suggesting that the mDC pathogenicity might not be restricted to the local inflammatory responses in NP. However, these results are inconsistent with a recent study by Shi et, al, showing no significant difference in the percentage of mDCs and activated DC subsets between NP and control subjects [
17]. The discrepancy may arise from differences in technical approaches. For example, we used CD1c as the marker of mDCs, whereas Shi et al. chose CD11c, which could also be expressed on some macrophage population [
43].
From previous reports, IL-25, IL-33 and TSLP can affect the properties and functions of DCs. For example, IL-25 instructs DCs to promote Th2 and Th9 inflammation in mouse models of allergic airway inflammation [
29,
44]. IL-33 activates DCs to express CCL17 and CCL22 through ST2 signaling [
45]. TSLP-licensed DCs are responsible for the initiation of allergic airway inflammation [
13,
28,
46]. However, whether DCs in NP respond to these innate type 2 cytokines remains unknown. In the present study, we demonstrated that mDCs accumulated in NP tissues exhibit elevated surface expression of IL-17RB, ST2 and TSLPR, suggesting that mDCs in NP have the potential to respond to IL-25, IL-33 and TSLP. These findings are parallel to our recent study in patients with allergic rhinitis [
31].
Ample evidences have indicated that the signals provided by the surface of DC subsets dictate Th1-Th2 differentiation. For example, it has well been shown that surface expression of OX40L is critical for the induction and maintenance of type 2 immune response elicited by TSLP-activated DCs [
13,
46]. PD-1/PDL1 interactions play an important role in maintaining peripheral tolerance [
47,
48]. ICOSL has been reported to be involved in DC-driven Th2 response to allergens [
49]. In our study, we found that activated mDCs expressed higher level of OX40L and ICOSL, but lower level of PDL1, in NP than those of control subjects. Furthermore, increased OX40L expression on activated mDCs was positively correlated with IL-4 and IL-25 mRNA levels, as well as tissue eosinophil numbers. In contrast, increased ICOSL and decreased PDL1 were negatively correlated with IFN-γ and IL-5 respectively. These results further suggest that mDCs might play a potential role in the IL-25- and IL-33-induced Th2 inflammation via expression of functional surface molecules, such as OX40L, ICOSL and PDL1, in NP.
Previous study showed that TSLP-induced OX40L expression on DCs is required for initiation of Th2 cell polarization, proposing an important role OX40L on DCs plays in determining T cell differentiation [
46]. However, although increased expression of TSLP mRNA was noted in NP, we did not find significant association between TSLP mRNA level and OX40L expression on activated mDCs in NP, implying that TSLP might induce Th2 immune response not through OX40L expression on mDCs in NP. Further functional studies are needed to address this hypothesis.
Group 2 innate lymphoid cells (ILC2s) are a recently identified innate cell subset that produces large amounts of IL-5 and IL-13 and therefore serve an important role in orchestrating the type 2 inflammation [
50‐
52]. IL-25, IL-33 and TSLP have been shown to be the key for production of type 2 cytokines by ILC2s [
50‐
52]. Several studies have demonstrated that ILC2s are increased in NP tissues [
24,
53,
54]. However, our recent study [
26] found that the IL-17RB expression on ILC2 in NP was relatively low and unable to be upregulated by IL-25 in vitro, implying that ILC2 might not be critical in mediating IL-25-induced Th2 inflammation in NP. Whether this is the case for IL-33 remains to be investigated.
Although we have provided new information that elevated expression of IL-17RB and ST2 on mDCs might underlie the pathogenesis of IL-25- and IL-33-induced Th2 inflammation in NP, several limitations still need to be addressed before more definitive conclusions can be drawn. First, we were unable to perform functional experiments to analyze the direct effect of IL-25 and IL-33 on mDCs due to the technical limitations of the isolation of pure DCs from NP tissue. Second, the present study was an ex vivo study. It did not provide direct evidence of the role of IL-17RB and ST2 on mDCs in promoting Th2 inflammation in vivo. Third, the sample size is relatively small, since we did not observe several expected correlations such as between ST2 expression on mDCs and IL-4 and IL-13. Moreover, it should be noted that several correlations were weak, such as between OX40L expression on CD86+CD1c+ DCs and IL-25 mRNA level and tissue eosinophil number (r = 0.394 and 0.368, respectively), ST2 expression on CD1c+ DCs and tissue eosinophil number (r = 0.366), although the P values were less than 0.05. Further studies are needed to address these issues.
Authors’ contributions
RZ, DW and KW performed the flow cytometry, qPCR, data analysis, and prepared the manuscript; W-XG performed some flow cytometry; Q-TY participated in data analysis and manuscript preparation; L-JJ, MZ and Y-JC participated in sample collection; YS and JS contributed to the conception and design of the study. All authors reviewed and approved the final manuscript.