CCL18-NIR1/GPR3 promotes oral cancer cell growth and metastasis by activating the JAK2/STAT3 signalling pathway CURRENT

Background Chemokine (C-C motif) ligand 18 (CCL18) affects the malignant progression of varying cancers by activating chemokine receptors. Our previous study have shown that CCL18 promotes hyperplasia and invasiveness of oral cancer cells; however, the cognate receptors of CCL18 involved in the pathogenesis of oral squamous cell carcinoma (OSCC) has not yet been identified. This study aimed to investigate the underlying molecular mechanisms through which CCL18 promotes OSCC progression by binding to specific functional receptors.Methods The properties of CCL18 receptors (i.e., NIR1, CCR6, CCR8, and GPR3) in OSCC were detected by conducting western blotting, immunofluorescence, and immunocytochemistry assays. The binding between CCL18 and its receptors was verified by performing coimmunoprecipitation (CO-IP) assays. The χ2 test was applied to analyze the relationship between CCL18 receptor expression patterns and clinicopathological factors. Recombinant CCL18 (rCCL18) and receptor siRNA were used to confirm the effects of CCL18 receptor axis on the morphology of cancer cells (i.e., proliferation, and metastasis), epithelial-mesenchymal transition (EMT) and the expression of the JAK2/STAT3 signalling pathway.Results NIR1 and GPR3 as specific receptors of CCL18 in OSCC were found to be significantly increased and positively related to the TNM stage of OSCC patients. rCCL18 induced phenotype alterations of oral cancer cells including cell growth, metastasis, and EMT. Knockdown of NIR1 and/or GPR3 expression could block the effects of rCCL18-induced OSCC. Furthermore, JAK2/STAT3 signalling was confirmed to be a downstream pathway of the CCL18-NIR1/GPR3 axis.Conclusion CCL18 can promote the progression of OSCC by binding specific receptors (NIR1 and GPR3), and the CCL18-receptor signalling can activate


4
JAK2/STAT3 pathway. The identification of the mechanisms of CCL18 promoting OSCC progression could implicate potential therapeutic targets for treating oral cancer.

Background
Oral squamous cell carcinoma (OSCC) as the most common type of oral cancer is well-known to be characterized by a high risk of local invasion and cervical lymph node metastasis [1]. Chemokines have been shown to promote the metastasis of OSCC by facilitating the proliferation, survival, and migration of cancer cells, as well as shaping the tumour microenvironment [2]. Chemokine receptors can selectively bind to chemokines and further transduce intercellular signals that can impact the oncogenic processes of cancer [3]. Chemokines-receptors signalling axis has been therefore considered as the hallmark of cancer and potential therapeutic strategy lack of understanding about the involvement of CCL18-receptor signalling axis in the cancer biology of OSCC. It remains a question that if the dysregulation of the 5 receptors listed above (e.g., NIR1, CCR6, CCR8 and GPR3) are also involved in the progression of OSCC by binding to CCL18. Consequently, the identification of specific receptors of CCL18 appears to be essential for researchers to obtain a better understanding of the implication of CCL18-receptors signalling pathway in OSCC.
The signalling pathways are well-known to interact with each other by activating or inhibiting their downstream pathways, thereby forming a complicated pathway interaction network to influence the immune regulation of cancer [10]. Based on this theory, we believe that it is necessary to identify the downstream pathway of CCL18-receptors signalling pathway. JAK2/STAT3 (Janus kinase 2/signal transducers and activators of transcription 3) signalling as an oncogenic pathway involved in many solid cancers including OSCC [11]has been shown to be activated by several chemokine-signalling axes, for instance, CXCL12-CXCR4 axis [12], CXCL8-CXCR1/CXCR2 axis [13], and CXCL9-CXCR3 axis [14]. The interaction between JAK2/STAT3 and these chemokine-receptor axes arouse our interest and made us wonder: if the CCL18-receptors signalling is coupled with the JAK2/STAT3 pathway and if the interaction between both pathways plays a contributing role in the metastasis of OSCC. To our knowledge, this is the first report that identified the specific receptors of CCL18 that are involved in the regulation of OSCC, and further investigated the underlying mechanism of CCL18-receptor axis in promoting metastasis of OSCC by determining its coupling downstream pathway.

Cell lines and reagents
The HSC6 cell line was purchased from CinoAsia Co., Ltd. (Shanghai, China). CAL27, SCC9 and HOK cell lines were purchased from TongPai Biotechnology Co., Ltd.

Immunohistochemistry
The OSCC tissues and adjacent normal tissues were analysed using immunohistochemistry (IHC). Tissues were dewaxed in xylene and rehydrated in a graded alcohol series. After antigen retrieval with Tris-EDTA, the slides were The sections were covered with secondary antibody and incubated at room temperature for 30 min. Next, the tissue sections were visualized with DAB (Gene, Shanghai, China). The expression of NIR1 and GPR3 was quantified using a visual grading system based on the average optical density (+: positive, ++: strong positive).

Immunofluorescence
Cells were seeded in glass bottom cell culture dishes for 24 h. Then, the cells were

Coimmumoprecipitation assay
Protein coimmunoprecipitation (CO-IP) assays were performed using a Pierce TM Crosslink Magnetic IP/CO-IP Kit (Thermo Fisher Scientific, Waltham, MA, USA). Cells were lysed in IP lysis/wash buffer for 5 min and then centrifuged at 4°C for 15 min.
Magnetic beads were coated with anti-CCL18 (Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA), anti-GPR3 (Santa Cruz Biotechnology Inc., USA) or anti-IgG (CST, USA). After crosslinking the antibodies with DSS, the cancer cell lysates and prepared magnetic beads were mixed at 4°C overnight. After washing with IP lysis/wash buffer and ultrapure water, the CCL18-associated and GPR3-associated proteins were eluted with elution buffer and detected by Western blotting.

CCK-8 assay
Cells were treated with siRNA-NIR1 and/or siRNA-GPR3 for 48 h, and 5000 cells were then added to 96-well plates and treated with 20 ng/ml rCCL18. At 24 h, 48 h and 72 h, CCK-8 reagent (Sigma-Aldrich, Louis, MO, USA) was added, and the absorbance values of each well were read with a microplate reader (Thermo Fisher Scientific. Waltham, USA) at 450 nm.

Clone formation assay
Forty-eight hours after siRNA-NIR1 and/or siRNA-GPR3 transfection, cells were plated in 6-well plates at 1000 cells per well and exogenously stimulated with 20 ng/ml rCCL18. The number of cell clones was counted by crystal violet staining 14 days later.

Transwell assays
Cell migration and invasion were detected by transwell assays (Corning, New York, NY, USA). The upper chamber was precoated with 50 μl 20% Matrigel (Gibco, USA) for the invasion assay. Cells were transfected with siRNA for 48 h and then suspended in serum-free medium with or without 20 ng/ml rCCL18. The prepared cells were seeded in the upper insert, and the lower chamber was filled with DMEM containing 15% FBS. Then, the transwell plates were incubated at 37°C with 5% CO 2 for 24 h. Cells that did not invade through the pores were gently removed with cotton tips. The upper chamber was fixed with 4% formaldehyde for 15 min and

NIR1 and GPR3 expression in OSCC and its clinical significance
Differential expression levels of NIR1, GPR3, CCR6 and CCR8, which are receptorbinding proteins of CCL18, have been reported in different tumours. We used immunohistochemistry (IHC) to detect NIR1, GPR3, CCR6 and CCR8 expression in 10 pairs of OSCC tissues and adjacent normal tissues. Positive NIR1 and GPR3 expression was found to be localized primarily in the cellular membrane and cytoplasm of oral cancer cells. CCR6 and CCR8 expression was not found in oral cancer cells or normal oral mucosa cells (Fig. 1a). Western blotting results showed that the expression of NIR1 and GPR3 was significantly higher in cancer tissues than in adjacent normal tissues. CCR6 and CCR8 protein expression in cancer tissues and adjacent tissues was low (Fig. 1b). To further evaluate the clinical significance of NIR1 and GPR3, 43 OSCC tissues were analysed with NIR1 and GPR3 antibodies for IHC. All OSCC tissues displayed positive NIR1 and GPR3 expression. Highly NIR1-or GPR3-positive tissues were closely related to the clinical TNM stage (P=0.042, Table   1; P=0.040, Table 2; Fig. 2a and b). However, the expression of NIR1 and GPR3 was not significantly related to other clinical features of the patients, such as age, sex or histological grade.
In addition, we examined the protein levels of NIR1, GPR3, CCR6 and CCR8 in 3 OSCC cell lines (HSC6, CAL27 and SCC9) and in normal human oral epithelial keratinocytes (HOK). Western blotting and immunofluorescence (IF) verified that NIR1 and GPR3 were highly expressed in all OSCC cells compared with HOK cells and that CCR6 and CCR8 were rarely expressed in OSCC cells and HOK cells (Fig. 3a and   c). qRT-PCR also showed increased mRNA expression of NIR1 and GPR3 in OSCC cells (Fig. 3b).
NIR1 is the most common receptor of CCL18, and their potent combination has been verified in breast cancer [15]. GPR3, as a receptor of CCL18, has been implicated in pancreatic cancer. However, there has been no experimental evidence that CCL18 and GPR3 bind effectively. We immunoprecipitated total protein extracts from CAL27 cells with anti-CCL18 or anti-GPR3 antibodies. As shown in Fig. 3d, Western blotting revealed specific bands for CCL18 or GPR3 in the precipitated proteins.
Neither of these two proteins appeared when an irrelevant antibody (IgG) was used for immunoprecipitation. These data suggest that NIR1 and GPR3 may be receptors of CCL18 and thus participate in the progression of OSCC.

Effective interference sequences of NIR1 and GPR3 in OSCC cells
We next sought to confirm that CCL18 regulates the progression of OSCC through NIR1 and GPR3. Three different siNIR1 segments and siGPR3 segments were designed to screen effective interference sequences. Our results showed that the mRNA levels of NIR1 and GPR3 were significantly decreased by siNIR1-1, siNIR1-2, and siNIR1-3 and siGPR3-2 (Fig. 4a). Western blotting results showed that siNIR1-1 and siNIR1-3 could decrease the protein level of NIR1 and that siGPR3-1, siGPR3-2, and siGPR3-3 could decrease the protein expression of GPR3 (Fig. 4b). However, there were no significant changes in mRNA or protein levels in untreated cells or cells transfected with negative siRNA sequences. Overall, the data suggested that siNIR1-3 and siGPR3-2 were effective sequences for subsequent experiments.

NIR1 and GPR3 are required for OSCC cell proliferation via CCL18
To investigate the function of the CCL18-NIR1/GPR3 axis in OSCC, HSC6 and CAL27 cells were transfected with siNIR1 and/or siGPR3 and then stimulated with 20 ng/ml rCCL18. The proliferation of each group of cells was detected by CCK8 and clone formation assays. CCK8 analysis results showed that the proliferation of OSCC cells increased upon rCCL18 stimulation for 48 h and 72 h. After knocking down NIR1 and/or GPR3, the proliferation effect of rCCL18 on OSCC cells was reduced (Fig. 5a).
As shown in Fig. 5b, OSCC cells cultured with rCCL18 had strong clone formation ability for 14 days. Compared to the control conditions, transfection with siNIR1 and/or siGPR3 in cells cultured with rCCL18 resulted in a significant decrease in colony numbers.

NIR1 and GPR3 are required for cell mobility and EMT in OSCC cells via CCL18
NIR1 and GPR3 were associated with the tumour stage in OSCC patients. We thus characterized the effects of the CCL18-NIR1/GPR3 axis on the migration and invasion of OSCC cells using a transwell assay. HSC6 and CAL27 cells were treated with rCCL18, siNIR1+rCCL18 and/or siGPR3+rCCL18; untreated cells and siNC+rCCL18 treatment served as controls. The results in Fig. 6a show that rCCL18 14 could promote OSCC cell migration through the transwell membrane. In the presence of siNIR1 and/or siGPR3, the number of cells on the submembrane surface decreased despite rCCL18 stimulation. Similar results were observed in the cell invasion assay (Fig. 6b).
EMT plays a critical role in oral cancer metastasis by enhancing migration and invasion. Therefore, we examined the expression of E-cadherin, N-cadherin and ZEB2 in HSC6 and CAL27 cells subjected to the different treatments stated above.
Compared with that in the untreated group, the expression of E-cadherin decreased, and the expression of N-cadherin and ZEB2 increased in the rCCL18 group. Knocking down NIR1 and/or GPR3 reversed the decreases in E-cadherin levels and the increases in N-cadherin and ZEB2 levels caused by rCCL18 (Fig. 6c). All data suggested that CCL18 enhanced the EMT property of OSCC by binding to the receptors NIR1 and GPR3 and then promoting OSCC cell invasion and migration.

CCL18-NIR1/GPR3 activates the JAK2/STAT3 signalling pathway in OSCC cells
The JAK2/STAT3 pathway is known to be involved in the growth and EMT process in OSCC. As shown in Fig. 7, P-JAK2 (Tyr1007/1008) and P-STAT3 (Tyr705) levels increased, but there was no change in JAK2 and STAT3 protein expression in rCCL18treated OSCC cells. Silencing NIR1 and/or GPR3 in HSC6 and CAL27 cells could reverse the activation effect of rCCL18 on JAK2 and STAT3. In addition, decreased STAT3 expression was found in HSC6 and CAL27 cells treated with siNIR1+siGPR3+rCCL18. These findings indicated that CCL18-NIR1/GPR3 could promote the malignant progression of OSCC by activating the JAK/STAT3 signalling pathway.

Discussion
Our previous studies showed that CCL18 which was predominantly secreted by oral cancer cells promoted the malignancy of OSCC. A unique feature of chemokines has been commonly demonstrated to be the nature of receptors to which they bind.
Chemokines have been shown to elicit their effects mainly by activating specific transmembrane receptors which belong to the large family of G protein-coupled receptors (GPCRs) [16]. The receptor of CCL18 involved in mediating tumour pathogenesis has been investigated in many cancers; however, the specific receptor of CCL18 in oral cancer remains uncertain. In this study, we examined the expression profile of several putative receptors of CCL18 (i.e., NIR1, GPR3, CCR6, and CCR8) in OSCC tissues and cells through multiple immunoassays. Our results revealed that NIR1 and GPR3 were significantly upregulated in OSCC samples, whereas CCR6 and CCR8 were downregulated in both OSCC and control specimens.
Additionally, clinical analysis of 43 OSCC patients showed that the expression of NIR1 and GPR3 was higher in TNM stage III/IV samples compared to TNM stage Ⅰ/Ⅱ samples, suggesting close association between NIR1 and GPR3 and malignancy of oral cancer. This section will discuss the potential roles of NIR1 and GPR3 respectively in oral cancers by interacting with CCL18. Taking the example of NIR1, this receptor as a member of putative six transmembrane GPCRs has been firstly confirmed to be dysregulated in breast cancer [15]. Apart from its aberrant expression in breast cancer, the overexpression of NIR1 was also found to be associated with the clinical stage and histological grade of hepatocellular carcinoma (HCC) [17]. In addition, CCL18-NIR1 axis has been suggested to activate the intracellular calcium signaling and further promote the growth and invasion of breast and lung cancer cells [8,18].
Looking at the case of GPR3, this receptor belongs to the family of constitutively active Gs-coupled receptors and is predominantly expressed in both brain and oocytes [19]. Previous investigations about GPR3 focused mainly on its involvement in nervous system and follicle development in vertebrates [20]. Most studies showed that GPR3 belongs to orphan receptor and thus is lack of endogenous ligand [21]. Surprisingly, however, another study implied that abnormal expression of GPR3 acts as a receptor of CCL18 in pancreatic ductal adenocarcinoma (PDAC) [7].
This study for the first time reports the crosslink between CCL18 and GPR3 in OSCC.
Our findings unveiled critical potential of interaction between CCL18 and GPR3 in the chemokine-receptor network.
To further explore the role of CCL18-NIR1/GPR3 in oral cancer, the OSCC models of NIR1 and/or GPR3 silencing were established. It was observed that depletion of NIR1 and/or GPR3 in HSC6 and CAL27 cells impaired CCL18-induced proliferation, migration and invasion. Furthermore, expression of EMT markers was also altered.
While rCCL18 treatment led to down-regulation of E-cadherin and elevation of Ncadherin and ZEB2 in OSCC cells, silencing NIR1 and/or GPR3 caused the opposite, indicating that NIR1 and GPR3 may play essential roles in the CCL18-induced malignant progression of OSCC. JAK2/STAT3 signalling is an evolutionarily conserved signalling pathway and known to interfere with tumour growth and metastasis of OSCC [11,22]. Activation of JAK2 protein kinase can catalyze STAT3 protein phosphorylation which regulates the expression of oncogenic genes [23]. E-cadherin, N-cadherin and ZEB2 are wellknown downstream molecules of JAK2/STAT3 signalling pathway [24][25][26]. Numerous studies have authenticated that JAK2/STAT3 can be activated by certain chemokines, including CXCL3, IL-6 ,CXCL9 and CCL20 [27][28][29][30]. However, the relationship between CCL18 and JAK2/STAT3 signalling pathway remains elusive. In our study, we found that both P-JAK2 and P-STAT3 were upregulated in rCCL18stimulated oral cancer cells. But such CCL18-induced JAK2/STAT3 activation can be considerably diminished by siNIR1 and/or siGPR3 treatment. Intriguingly, when NIR1 and GPR3 were silenced simultaneously, the total protein levels of STAT3 in HSC6 and CAL27 cells were also significantly decreased. Overall, abovementioned results indicated that JAK2/STAT3 signalling contributes to the CCL18-NIR1/GPR3 mediated proliferation and migration in OSCC.
It is worthwhile to clearly state the limitation of this study. First, the relationship between NIR1 and GPR3 was not investigated in this study, in other words, if they play synergistically or antagonistically in OSCC needs to be elucidated in the future study. Another limitation that needs to be acknowledged is the 5-year survival rate of OSCC patients was not reported in this study since several patients fail to participate in the follow-up appointments. However, this study also provides implications for future research. First, the biomarkers identified in this study could be used as therapeutic targets in the gene delivery therapy of OSCC. In addition, the identification of these critical biomarkers will also advance the development of multi-target drugs; however, if these target drugs could synergistically work with conventional chemotherapeutic agents remains a question. Furthermore, these biomarkers related to metastasis could be regarded as novel evaluation indexes for prognosis of OSCC. The testing kits based on these biomarkers could be also developed for guiding clinal practitioners to promptly make subsequent treatment plans individually for patients with poor prognosis. 18

Conclusion
In summary, two receptors (i.e., NIR1 and GPR3) of CCL18 were identified to be upregulated in OSCC and associated with advanced tumour stage of OSCC. The CCL18-NIR1/GPR3 axis was found to regulate the proliferation, metastasis and EMT of OSCC cells by activating JAK2/STAT3 signalling pathway. Our results could facilitate a better understanding of the underlying mechanisms of CCL18-induced OSCC progression, and also indicate that NIR1 and GPR3 might be potential therapeutic targets for treating oral cancer.   Positive NIR1 and GPR3 expression in OSCC tissues is associated with cancer cells metastasis Figure 3 As potential receptors of CCL18, NIR1 and GPR3 are highly expressed in OSCC cell lines. A W Figure 4 Effective siRNA segments of NIR1 and GPR3 were screened in HSC6 cells. Western blotting an Table1.tif Table2.tif