Resistin Promotes Nasopharyngeal Carcinoma Metastasis through TLR4-Mediated Activation of p38 MAPK/NF-κB Signaling Pathway

Simple Summary Chronic inflammation is associated with the development of nasopharyngeal carcinoma (NPC). Mounting evidence has indicated that resistin is an inflammatory cytokine that is associated with the risk of tumorigenesis. However, the correlation between serum resistin levels and the risk of NPC remains unclear. Here, we found that high serum resistin levels in NPC patients were positively correlated with lymph node metastasis and that resistin promoted the metastasis of NPC cells both in vitro and in vivo. Furthermore, we elucidated the underlying molecular mechanisms through which resistin promotes metastasis in NPC cells by inducing the epithelial-mesenchymal transition (EMT). Abstract NPC is a type of malignant tumor with a high risk of local invasion and early distant metastasis. Resistin is an inflammatory cytokine that is predominantly produced from the immunocytes in humans. Accumulating evidence has suggested a clinical association of circulating resistin with the risk of tumorigenesis and a relationship between blood resistin levels and the risk of cancer metastasis. In this study, we explored the blood levels and the role of resistin in NPC. High resistin levels in NPC patients were positively associated with lymph node metastasis, and resistin promoted the migration and invasion of NPC cells in vitro. These findings were also replicated in a mouse model of NPC tumor metastasis. We identified TLR4 as a functional receptor in mediating the pro-migratory effects of resistin in NPC cells. Furthermore, p38 MAPK and NF-κB were intracellular effectors that mediated resistin-induced EMT. Taken together, our results suggest that resistin promotes NPC metastasis by activating the TLR4/p38 MAPK/NF-κB signaling pathways.


Introduction
Nasopharyngeal carcinoma (NPC) is a malignant tumor that originates from the nasopharyngeal epithelium [1]. The incidence of NPC is familial, with regional clustering existing in Southeast Asia and South China [2,3]. Established risk factors for NPC include Epstein-Barr virus (EBV) infection, a family history of NPC, and environmental factors [4][5][6]. In addition, the development of NPC is usually accompanied by chronic inflammation and metabolic dysregulation. Emerging evidence indicates that the immune system and cytokines may play an important role in the diagnosis and prognosis of NPC [7][8][9][10]. Indeed, our own previous study found that decreased levels of macrophage inflammatory protein (MIP)-1α and MIP-1β could increase the tumorigenic risk of NPC [11].

Cell Viability and Proliferation Assays
The NPC cells were cultured in 96-well plates and were treated with different concentrations of resistin for 48 h. After the incubation of the CCK-8 solution was carried out according to the provided instructions (Sangon Biotech, Shanghai, China), the absorbance was measured at 450 nm with a microplate reader (Infinite F50, Tecan Group Ltd., Mannedorf, Switzerland). The relative cell viability was calculated as the percentage of untreated cells. Cell proliferation was measured using plate clone formation and was carried out as previously described [32].

Wound-Healing Assay
The wound-healing assay was performed as previously described [31]. The NPC cells were plated in 12-well culture plates, and cell confluence was 100% after adherence. After the monolayer cells were scraped with a plastic 200 µL pipette tip, the cells were washed with phosphate-buffered saline (PBS) and incubated with fresh medium. The cell-scratch images were photographed using a microscope at 0 h and 24 h. The relative migration rates were calculated using the following calculation: cell-covered area (0 h)/cell-covered area (24 h).

Migration and Invasion Assays
The migration and invasion assays were performed using 24-well Transwell inserts (BD Biosciences, San Jose, CA, USA) coated with or without growth factor-reduced Matrigel (Corning Incorporated, Corning, NY, USA), as previously described [31]. The cells were resuspended in a 200 ul serum-free medium that had been treated with or without resistin, which was added to the Transwell's upper chamber, and a medium with 20% FBS was added to the Transwell's lower chamber. For the signaling blockade, cells were pre-incubated with an inhibitor for 2 h. After 24 h of incubation, membrane-trapped cells were fixed with 4% paraformaldehyde, stained with 10% crystal violet solution and counted using a light microscope.

Transient Transfection with Small Interfering RNA (siRNA)
TLR4, p38 MAPK and scrambled control siRNAs were synthesized by RiboBio (Guangzhou, Guangdong, China) and transfected using a transfection reagent kit (RiboBio) according to the manufacturer's protocols. The siRNA sequences used for this study are listed in Table S1.

RNA Extraction and qRT-PCR
RNA samples were extracted with Trizol reagent (Sigma-Aldrich, St. Louis, MO, USA). A cDNA synthesis was performed with the HiScript II Q RT kit (Vazyme Biotech, Nanjing, China). The quantitative real-time PCR (qRT-PCR) analysis was performed as previously described [32]. Primers were synthesized by Sangon Biotech (Shanghai, China). The primer sequences are listed in Table S2.

Western Blot Analysis
A Western blot analysis was performed as previously described [31]. Briefly, protein samples were extracted with radio-immunoprecipitation (RIPA) lysis buffer containing a protease inhibitor (Beyotime Biotechnology, Shanghai, China) and quantified with the bicinchoninic acid (BCA) protein assay kit (Beyotime Biotechnology). Cytosolic and nuclear proteins were extracted using the Nuclear and Cytoplasmic Protein Extraction Kit according to the manufacturer's instructions (Beyotime Biotechnology). Protein samples were separated using SDS-PAGE and were transferred to a polyvinylidene fluoride (PVDF) membrane (Millipore, Billerica, MA, USA). The membranes were blocked with 5% nonfat dried milk and incubated overnight at 4 • C with the primary antibody and then incubated with the secondary antibody for 1 h at room temperature. The blots were tested with the ECL detection system (Thermo, Waltham, MA, USA) using the ChemiDoc XRS+ system (Bio-Rad, Hercules, CA, USA). The antibodies used are listed in Table S3.

Immunofluorescence Staining
Immunofluorescence staining was carried out as previously reported [31]. The NPC cells were cultured onto glass-bottom cell culture dishes (Wuxi NEST Biotechnology Co., Ltd., Jiangsu, China). After incubation with resistin, the cells were fixed with 4% paraformaldehyde, permeabilized in 0.2% Triton X-100 PBS buffer, and then blocked with Immunol Staining Blocking Buffer (Beyotime), incubated with primary antibody rabbit anti-p65 (CST; 1:400) overnight at 4 • C, and then incubated with goat anti-rabbit Alexa 555 fluorescent secondary antibody (CST; 1:1000) for 2 h at room temperature. After washing with PBS, the cells were mounted with an anti-fade solution with DAPI (Beyotime). The images were obtained using a fluorescent microscope.

Dual-Luciferase Reporter Assay
The pNFκB-luc, pRL-TK plasmids, and dual-luciferase reporter assay kit were purchased from Beyotime. The dual-luciferase reporter assay was determined as previously reported [31] using a Varioskan LUX multimode microplate reader (Thermo). Relative luminescence units = Firefly luciferase activity/Renilla luciferase activity.

Immunohistochemistry Staining
Immunohistochemistry staining was carried out as described elsewhere [31]. After dewaxing and rehydrating, microwave heating for antigen retrieval was performed on the sections in a citrate antigen retrieval solution. After blocking with 3% H 2 O 2 for 15 min, the sections were incubated with 5% goat serum buffer for 1 h at room temperature, followed by overnight incubation at 4 • C with the primary antibodies mouse anti-p-p65 (CST; 1:50), rabbit anti-Vimentin (CST; 1:50) and rabbit anti-E-cadherin (CST; 1:100). Then, the sections were incubated with the secondary antibody for 1 h at room temperature. The sections were incubated with a developing solution (diaminobenzidine, DAB) and counterstained with hematoxylin (Wuhan Servicebio Technology, Hubei, China).

Statistical Analysis
In order to describe the cohort characteristics, the Chi-square (χ2) test and Wilcoxon rank-sum test were used to compare the differences between the case and control groups. The median levels of resistin among the cases and controls were compared using the Wilcoxon rank-sum test to compare the differences between groups. In the multivariable models, we adjusted for the established or suspected risk factors of NPC, including age, gender, EBV VCA-IgA and EBNA1 IgA. The odds ratios (ORs) and 95% confidence intervals (95% CIs) for the risk of NPC were computed using an unconditional logistic regression model. The differences in the serum resistin levels among NPC patients with different clinical characteristics were analyzed using the Wilcoxon rank-sum test. The correlation of the serum resistin levels with LN metastasis was analyzed by unconditional logistic regression. Statistical analyses were performed using the SAS statistical software, version 9.4 (SAS Institute, Cary, NC, USA). All hypothesis tests underwent two-sided testing, and a p < 0.05 was considered statistically significant.
Data are presented as mean ± SD. All the data (except clinical) were analyzed using a Student's t-test or one-way ANOVA with Sidak's multiple comparisons test using Graph-Pad Prism 8.0 (GraphPad Software, La Jolla, CA, USA). A value of p < 0.05 was considered statistically significant.

Clinical Correlation of Serum Resistin Levels with the Risk of NPC
The descriptive characteristics of the study subjects with NPC and the controls who provided serum are presented in Table 1. The serum resistin levels were significantly higher in the NPC patients compared with the controls (4.12 vs. 3.59 ng/mL; p < 0.001) ( Table 1). The sex-corrected resistin levels were higher among the cases compared to the controls; these differences were statistically significant among men (4.18 vs. 3.58 ng/mL; p < 0.001) and women (4.38 vs. 3.46 ng/mL; p = 0.003) ( Table 1). In multivariable logistic regression models, we observed that high serum resistin levels were associated with increased NPC risk after adjusting for established or suspected risk factors of NPC, including age, gender, EBV VCA-IgA and EBNA1 IgA ( Table 2). Through analyzing the differences in serum resistin levels with clinical characteristics, we observed different levels of serum resistin in NPC patients with different types of lymph node metastasis (Table 3). Moreover, the serum resistin level was a significant independent predictor for lymph node metastasis in NPC patients, according to multivariate logistic regression analysis after adjusting for established or suspected risk factors of NPC, including age, gender, EBV VCA-IgA and EBNA1 IgA (Table 4).

Resistin Does Not Affect the Proliferation but Promotes the Migration and Invasion in NPC Cells
To determine the manifestations of the clinical correlation of resistin, we further explored whether resistin affected the activity of NPC cells. A co-culture of different concentrations of resistin for 48 h with NPC cells did not affect cell viability ( Figure 1A). Furthermore, resistin treatment did not affect the proliferation of the NPC cells in a colony formation assay after long-term incubation ( Figure 1B).
To determine the manifestations of the clinical correlation of resistin, we further e plored whether resistin affected the activity of NPC cells. A co-culture of different con centrations of resistin for 48 h with NPC cells did not affect cell viability ( Figure 1A). Fu thermore, resistin treatment did not affect the proliferation of the NPC cells in a colon formation assay after long-term incubation ( Figure 1B). Interestingly, resistin treatment of the NPC cells enhanced wound healing and th migration and invasion activities in a dose-dependent manner (Figure 2A-C). The epith lium-mesenchymal transition (EMT) plays an important role in tumor cell invasion an cancer metastasis [33]. We found that resistin also induced the expression of EMT-pro moting transcription factors, such as ZEB1 and Snail and Slug, via the Western blot assay however, the level of β-catenin was not altered by the resistin treatment ( Figure 2D). Th loss of E-cadherin expression is a hallmark of an EMT [34]. The resistin treatment signif cantly suppressed the expression of E-cadherin in NPC cells as well as other epitheli markers, such as claudin-1 and ZO-1 ( Figure 2D). Conversely, the levels of N-cadher and vimentin, which are the hallmarks of mesenchymal cells, were significantly increase after the resistin treatment ( Figure 2D). Importantly, resistin also elevated the expressio of matrix metalloproteinase 2 (MMP-2) and matrix metalloproteinase 9 (MMP-9) (Figu 2D), both of which are essential for cell motility and invasion. Thus, resistin can promo migration and invasion by inducing an EMT in NPC cells. Interestingly, resistin treatment of the NPC cells enhanced wound healing and the migration and invasion activities in a dose-dependent manner (Figure 2A-C). The epitheliummesenchymal transition (EMT) plays an important role in tumor cell invasion and cancer metastasis [33]. We found that resistin also induced the expression of EMT-promoting transcription factors, such as ZEB1 and Snail and Slug, via the Western blot assays; however, the level of β-catenin was not altered by the resistin treatment ( Figure 2D). The loss of E-cadherin expression is a hallmark of an EMT [34]. The resistin treatment significantly suppressed the expression of E-cadherin in NPC cells as well as other epithelial markers, such as claudin-1 and ZO-1 ( Figure 2D). Conversely, the levels of N-cadherin and vimentin, which are the hallmarks of mesenchymal cells, were significantly increased after the resistin treatment ( Figure 2D). Importantly, resistin also elevated the expression of matrix metalloproteinase 2 (MMP-2) and matrix metalloproteinase 9 (MMP-9) ( Figure 2D

TLR4 Is Necessary for Resistin-Induced NPC Cell Migration
Resistin is a type of cysteine-rich polypeptide hormone that functions through its purported receptor, Toll-like receptor 4 (TLR4), which plays a critical role in regulating inflammation and is also involved in tumor cell proliferation, invasion and metastasis [25,28,29]. The expression of TLR4 is widely observed in head and neck squamous cell carcinoma (HNSC) and NPC tissues ( Figure S1A,B), as well as in several NPC cell lines ( Figure S1C). Although there is no discernable difference in TLR4 expression in NPC, as compared with the normal tissues ( Figure S1D), high levels of TLR4 expression were correlated with increasing tumor grade and nodal metastasis ( Figure S1E,F). The blockade of TLR4 with the pharmaceutical inhibitor LPS-RS Ultrapure, a specific TLR4 antagonist, suppressed cell migration and invasion induced by the resistin treatment ( Figure 3A-D). Vimentin, Claudin-1, ZEB-1, ZO-1 and β-catenin in cultured S18 and 5-8F cells after treatment with or without 25 ng/mL resistin for 24 h. Results are presented as mean ± SD of three independent experiments performed in triplicate. ** p < 0.01, *** p < 0.001, # p < 0.0001.

TLR4 Is Necessary for Resistin-Induced NPC Cell Migration
Resistin is a type of cysteine-rich polypeptide hormone that functions through its purported receptor, Toll-like receptor 4 (TLR4), which plays a critical role in regulating inflammation and is also involved in tumor cell proliferation, invasion and metastasis [25,28,29]. The expression of TLR4 is widely observed in head and neck squamous cell carcinoma (HNSC) and NPC tissues ( Figure S1A,B), as well as in several NPC cell lines ( Figure S1C). Although there is no discernable difference in TLR4 expression in NPC, as compared with the normal tissues ( Figure S1D), high levels of TLR4 expression were correlated with increasing tumor grade and nodal metastasis ( Figure S1E,F). The blockade of TLR4 with the pharmaceutical inhibitor LPS-RS Ultrapure, a specific TLR4 antagonist, suppressed cell migration and invasion induced by the resistin treatment ( Figure 3A-D). Similarly, the knockdown of TLR4 expression significantly nullified the resistin-induced elevation of N-cadherin, MMP-2 and MMP-9 expression, as well as a reduction of E-cadherin in the NPC cells (Figures 3E and S2). These results unequivocally demonstrate that resistin induces the migration and invasion of NPC cells through TLR4. Similarly, the knockdown of TLR4 expression significantly nullified the resistin-induced elevation of N-cadherin, MMP-2 and MMP-9 expression, as well as a reduction of E-cadherin in the NPC cells ( Figures 3E and S2). These results unequivocally demonstrate that resistin induces the migration and invasion of NPC cells through TLR4.

The p38 MAPK Signaling Pathway Is Involved in Resistin-Induced Migration in NPC Cells
We further examined the downstream signaling events of TLR4. While resistin did not affect the phosphorylation of AKT, it stimulated the phosphorylation of p38 mitogen-

The p38 MAPK Signaling Pathway Is Involved in Resistin-Induced Migration in NPC Cells
We further examined the downstream signaling events of TLR4. While resistin did not affect the phosphorylation of AKT, it stimulated the phosphorylation of p38 mitogenactivated protein kinase (MAPK) ( Figure 4A) and suppressed the level of ERK1/2 phosphorylation ( Figure 4A). Incubation of the NPC cells with a specific p38 inhibitor, SB203580, largely reversed resistin-induced migration ( Figure 4B), whereas the inhibitors of ERK1/2, JNK and AKT showed no effect on the migration ( Figure 4B). Consistent with this observation, siRNA's knockdown of p38 MAPK expression prevented cell migration and invasion in the resistin-treated NPC cells ( Figures 4C,D and S3). Due to the reduction in p38 MAPK expression, the resistin-induced changes in the EMT-related proteins were inhibited by transfection with p38 MAPK siRNA ( Figure 4E). In analyzing the whole cell lysates from resistin-treated cells, we also found that blocking TLR4 activity via LPS-RS Ultrapure or siRNA transfection abolished the resistin-induced activation of p38 MAPK ( Figure 4F,G), further proving that TLR4 was required for the resistin-induced activation of p38 MAPK in NPC cells.

Resistin Regulates Expression of EMT-Related Protein via NF-κB
The involvement of Nuclear factor-κB (NF-κB) in regulating the expression of EMTrelated protein has been well documented [35,36]. Co-cultures with resistin increased the phosphorylation of the p65 protein in a dose-dependent manner in the NPC cells ( Figure 5A). The resistin treatment promoted the phosphorylation of IκBα ( Figure 5B), which, in turn, led to the degradation of IκBα and to the activation of NF-κB. Consistent with these results, the proportion of nuclear translocation of the p65 and p50 proteins markedly increased following the resistin treatment ( Figure 5C). Moreover, pretreatment with the NF-κB inhibitors, BAY-117083 and PDTC, completely suppressed a resistin-induced EMT as well as the migration of NPC cells ( Figure 5D,E). These results demonstrate that resistin promotes the EMT of NPC cells, largely through the activation of the NF-κB pathway.
We further delineated the molecular mechanisms underlying the resistin-induced EMT alterations in the NPC cells, particularly with respect to NF-κB signaling. The transcriptional activation of NF-κB, induced by resistin, was suppressed by co-culturing with LPS-RS Ultrapure, a specific inhibitor of TLR4 ( Figure 6A). The immunofluorescence staining revealed that the resistin-induced nuclear translocation of p65 was abolished by the LPS-RS Ultrapure treatment in the NPC cells ( Figure 6B). Importantly, by impeding the activation of p38 MAPK via its pharmacological inhibitor, it also suppressed the resistin-mediated activation of NF-κB in the NPC cells ( Figure 6A,C) and reversed the resistin-induced phosphorylation of IκBα ( Figure 6D,E). These results, taken together, demonstrate that the induction of cellular migration by resistin depends on the TLR4/p38 MAPK/NF-κB signaling pathways.

Resistin Promotes NPC Tumor Metastasis in Animal Models
To clarify whether intravenously administered resistin would exhibit a pharmacokinetic profile suitable for an in vivo evaluation, we measured the serum concentrations of resistin in nude mice after the intravenous administration of 20 µg/kg of resistin. We found that the concentration of resistin was 20.79 ng/mL at 15 min ( Figure 7A), which was consistent with the concentration of resistin that promoted migration in vitro. and invasion in the resistin-treated NPC cells ( Figures 4C,D and S3). Due to the reduction in p38 MAPK expression, the resistin-induced changes in the EMT-related proteins were inhibited by transfection with p38 MAPK siRNA ( Figure 4E). In analyzing the whole cell lysates from resistin-treated cells, we also found that blocking TLR4 activity via LPS-RS Ultrapure or siRNA transfection abolished the resistin-induced activation of p38 MAPK ( Figure 4F,G), further proving that TLR4 was required for the resistin-induced activation of p38 MAPK in NPC cells.  turn, led to the degradation of IκBα and to the activation of NF-κB. Consistent results, the proportion of nuclear translocation of the p65 and p50 proteins ma creased following the resistin treatment ( Figure 5C). Moreover, pretreatment wi κB inhibitors, BAY-117083 and PDTC, completely suppressed a resistin-induce well as the migration of NPC cells ( Figure 5D,E). These results demonstrate th promotes the EMT of NPC cells, largely through the activation of the NF-κB pa To understand the effects of resistin in metastasis in vivo, we established luciferaseexpressing 5-8F-Luc cells. A lung metastasis model was established by intravenously injecting 5-8F-Luc cells into nude mice, and the tumor metastasis was monitored by bioluminescence imaging ( Figure 7B). The intravenous delivery of the exogenous recombinant resistin proteins significantly increased lung metastasis at 6 weeks post-injection ( Figure 7C-E), with the lungs showing more and larger metastatic nodules in the resistintreated group than in the control group ( Figure 7F,H), accompanied by the lungs exhibiting an increased wet weight ( Figure 7G). The immunohistochemical staining showed that treatment with exogenous resistin markedly elevated the levels of phospho-p65 and vimentin ( Figure 7I,J). These results, taken together, unequivocally validated the concept that elevation of blood resistin could enhance the metastasis ability of NPC cells in the metastatic animal model. the LPS-RS Ultrapure treatment in the NPC cells ( Figure 6B). Importantly, by impe the activation of p38 MAPK via its pharmacological inhibitor, it also suppressed the tin-mediated activation of NF-κB in the NPC cells ( Figure 6A,C) and reversed the res induced phosphorylation of IκBα ( Figure 6D,E). These results, taken together, de strate that the induction of cellular migration by resistin depends on the TLR4 MAPK/NF-κB signaling pathways.

Discussion
Using a case-control cohort of 100 patients and 100 controls, we revealed, time, that high serum resistin levels were associated with an increased risk of portantly, we showed that the serum resistin levels were positively correlated w node metastases in NPC patients. Consistent with these clinical findings, the res ment promoted the invasion and migration of NPC cells in cultured cells as w

Discussion
Using a case-control cohort of 100 patients and 100 controls, we revealed, for the first time, that high serum resistin levels were associated with an increased risk of NPC. Importantly, we showed that the serum resistin levels were positively correlated with lymph node metastases in NPC patients. Consistent with these clinical findings, the resistin treatment promoted the invasion and migration of NPC cells in cultured cells as well as metastasis in a human NPC cell-derived animal model. Resistin promoted the invasion and migration of NPC cells by inducing an EMT, a molecular event that was initiated by the interaction of resistin with its purported receptor, TLR4, and further mediated by the activation of the p38 MAPK and NF-κB pathways.
Nasopharyngeal carcinoma is typically characterized by heavy lymphocytic infiltration, suggesting that inflammation might be a potential risk factor for the progression of this cancer [11]. Indeed, a series of cytokines, such as leptin, adiponectin and visfatin, have been found in tumor microenvironments and have been implicated in cancer cell growth, apoptosis, invasion, angiogenesis and metastasis [13,19]. Resistin is a cytokine that is predominantly produced and secreted by macrophages, dendritic cells and monocytes in humans [17,37]. The purported ortholog receptor of resistin, TLR4, is usually expressed and has recently been identified on multiple tumor cells, including gastric cancer, breast cancer and lung adenocarcinoma [25,28,29]. Recent studies have already suggested that the polymorphisms and high expression of TLR4 are linked to an increased risk of NPC [38][39][40].
Our results indicate that TLR4 is widely expressed in NPC as well as in head and neck tumors; that its expression level is positively correlated with high grades of tumor and lymph node metastasis in HNSC; that the inhibition of TLR4 signaling prevents resistininduced migration and invasion; and that TLR4 knockdown prevents the resistin-induced expression of multiple critical EMT proteins. These findings are consistent with published reports, indicating that the activation of TLR4 could promote cancer cell proliferation, adhesion, EMT, invasion and migration [25,28,29]. Thus, TLR4 is the functional receptor of resistin signaling and is responsible for mediating the pro-metastatic effect of resistin in NPC cells.
Intracellular signaling pathways, such as MAPK and PI3K/AKT, are involved in mediating TLR4 functions [41,42]. In NPC cells, we only found the activation of p38 MAPK signaling after resistin treatment and that pretreatment with specific inhibitors of p38 MAPK largely reversed resistin-induced migration or invasion. Importantly, the blockade of TLR4 signaling reduced the resistin-induced activation of p38 MAPK signaling, proving that the TLR4/p38 MAPK signaling pathways are critical for resistin's induction of migration and invasion. The NF-κB proteins belong to a family of transcription factors that are involved in cellular functions, such as inflammation, immune responses, cell proliferation and apoptosis [41]. Moreover, NF-κB is an important regulator of the EMT process of tumor cells [35,36]. The activation of NF-κB by the cytokines from the tumor microenvironment plays an important role in the invasion and migration of NPC cells [43]. Indeed, pharmacological inhibition of the NF-kB signaling pathways attenuates resistininduced EMT-related protein expression, a process that depends on the activation of the TLR4/p38/NF-κB pathway ( Figure 8). These data provide an underlying mechanism describing how high blood levels of resistin promote the metastasis of NPC.

Conclusions
In conclusion, the findings of this study demonstrate that serum resistin levels are positively correlated with the risk of NPC development and can potentially serve as an independent predictor of lymph node metastasis in NPC cases. We propose that resistin promotes NPC metastasis through the induction of an EMT by activating the TLR4/p38 MAPK /NF-κB signaling pathways. The circulating levels of resistin may be considered for predicting the prognosis of NPC patients.

Conclusions
In conclusion, the findings of this study demonstrate that serum resistin levels are positively correlated with the risk of NPC development and can potentially serve as an independent predictor of lymph node metastasis in NPC cases. We propose that resistin promotes NPC metastasis through the induction of an EMT by activating the TLR4/p38 MAPK /NF-κB signaling pathways. The circulating levels of resistin may be considered for predicting the prognosis of NPC patients.

Supplementary Materials:
The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/cancers14236003/s1. The following supporting information, including supplementary methods; Figure S1: The expression of TLR4 in human HNSC and NPC; Figure S2: Knockdown expression of TLR4 in NPC cells; Figure: S3 Knockdown expression of p38 MAPK in NPC cells; Table S1: Sequences of small interfering RNA used in transfection; Table S2: Sequences of primers used in quantitative RT-PCR; Table S3: Antibodies list [44][45][46][47][48][49][50][51].  Institutional Review Board Statement: This study was approved by the Institutional Review Board of Sun Yat-sen University Cancer Center (SYSUCC) (NO. YP2009051). The study was carried out in compliance with the ARRIVE guidelines. All experimental animal procedures were approved by the Experimental Animal Academic Ethics Committee of the South China University of Technology (AEC2021059).

Informed Consent Statement:
Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The data supporting the findings of this study are available on request from the corresponding author.