With an estimated 573,000 new cases and 212,000 deaths in 2020, urinary bladder cancer (BCa) has become one of the most common malignancies worldwide (1). In the United States, BCa accounted for 7% of all malignancy cases and ranked fourth among all types of malignancies in 2020 (2). Chemical components from cigarettes and industrial products are well-known risk factors; however, the etiology of BCa is poorly understood.

Previous evidence has shown that the immune system might play important roles in the development and progression of BCa, involving immune surveillance and immune tolerance. In immunoregulatory pathways, chemokines serve as key factors (3). Tumor immunotherapy represented by PD1/PD-L1 has attracted great attention in recent years. The effect of chemokines on tumors has been proven to produce a two-way regulation pattern: first, chemokines may enhance antitumor immunity by promoting immune cell recruitment to the tumor site (4); second, autonomous secretion of chemokines and receptors from tumor cells may promote tumor growth by inhibiting antitumor immunity (5).

A unique class of chemokines-the CX3C family has been reported playing a critical role in tumor immune regulation, in which CX3CL1 is the only member (6). It contains a unique 29nm long pedicle structure (mucus domain) in N prime/C prime. This structure allows chemokines to extend from epithelial cells to contact flowing white blood cells, and is independent of selectins and integrins (7). In addition, it can also induce intracellular calcium flow and chemotaxis without activating integrins (7). The receptor of CX3CL1, CX3CR1, is a transmembrane domain receptor, which mediates cell adhesion to CX3CL1 without G protein activation (8). Therefore, the unique structure and function of CX3CL1 has attracted the attention of many researchers.

As an important molecule in the chemokine family, C-X3-C Motif Chemokine Ligand 1 (CX3CL1) was found to be significantly associated with the development and progression of multiple cancers (9–13). However, the roles of CX3CL1 and its receptors are still controversial in terms of the biological regulation of tumors. Studies have suggested that CX3CL1 exerts its tumor-suppressive effect by activating immune cells, including natural killer T cells, and inhibiting tumor cell aggregation and adhesion (14, 15). For example, in a study of breast cancer, the high expression of CX3CL1 might recruit immune cells, such as CD8+ T cells and NK cells, to exert anti-tumor activity (16). Another study showed that patients with high expression of CX3CL1 and CX3CL1 receptors would have a significantly better prognosis in liver cancer (15). However, CX3CL1 may also produce a cancer-promotive effect as an adhesion molecule (9, 13). It was reported that high expression of CX3CL1 can induce tumor cell proliferation and promote the transformation of the cell cycle to S phase in gastric cancer (17). Shulby et al. indicated that CX3CL1 involved in the invasion process of prostate cancer metastasis to bone marrow which could be significantly reduced by anti-CX3CL1 antibody (13). This two-way regulation pattern of CX3CL1 is still unclear in BCa.

The objective of the current study was to evaluate the association between CX3CL1 and BCa and to investigate the underlying biological mechanisms involving CX3CL1 and its receptors in BCa.

The expression level of CX3CL1 mRNA in bladder cancer tissues was significantly higher than that in normal tissues ( Figure S2 , P=0.0268). A total of 277 patients were enrolled in the present study, with an average follow-up time of 49 months (1 to 72 months). Among them, 50 (18.1%) were negative for CX3CL1 expression, and 227 (81.9%) were considered positive for CX3CL1 expression. The demographic and clinical characteristics of the patients are summarized in Table 1 . The patients were divided into 2 groups according to their expression of CX3CL1. By univariable analysis, the expression status of CX3CL1 was significantly associated with tumor stage [odds ratio (OR)=2.55, 95% confidence interval (95%CI): 1.36-4.76, P=0.003], tumor size (OR=2.06, 95%CI: 1.10-3.86, P=0.025), tumor grade (OR=1.91, 95%CI: 1.02-3.55, P=0.043), disease recurrence (OR=2.05, 95%CI: 1.08-3.88, P=0.029), metastatic disease (OR=3.98, 95%CI: 1.95-8.13, P<0.001), Ki67 expression (OR=1.02, 95%CI: 1.01-1.04, P=0.002) and mortality (OR=4.90, 95%CI: 2.53-9.47, P<0.001, Table S1 ). Multivariable analysis showed that CX3CL1 expression remained significantly associated with tumor size, tumor stage, Ki67 expression, disease recurrence, and mortality (all P<0.05).

CX3CL1 was an independent predictor of BCa recurrence in both a univariable analysis (OR=2.44, 95%CI: 1.24-2.83, P=0.001) and a multivariable analysis (OR=2.37, 95%CI: 1.18-2.75, P=0.006) after adjusting for multiple variables, as shown in Table 2 . As shown in the cumulative hazard curve of disease recurrence ( Figure 1A ), the 6-year cumulative recurrence rate of the BCa CX3CL1-positive group was 62.9%, which was significantly higher than that of the BCa CX3CL1-negative group (31.8%, P=0.006). Similarly, the BCa-specific survival rate of CX3CL1-positive group was 23.2%, which was significantly lower than that of the negative expression group (41.1%, P<0.0001, Figure 1B ). Patients with high expression of CX3CL1 had an ~2-fold increased risk of cancer-specific death based on univariable (HR=2.16, 95%CI: 1.59-2.93, P<0.001) and multivariable (HR=2.17, 95%CI: 1.08-2.39, P=0.006) analyses ( Table 3 ). This indicated that CX3CL1 was also a significant and independent risk factor for disease-specific death. A total of 406 BCa samples were obtained from the TCGA database. Survival analysis suggested that the patients with high expression of CX3CL1 (FPKM≥1.0) were significantly associated with poor survival compared with the patients with low CX3CL1 expression (log-rank P value=0.039, Figure 1C ). These results remained significant after adjusting for age at diagnosis and tumor stage via Cox regression (HR=1.76, 95%CI: 1.14-2.71, P=0.011)

To evaluate whether CX3CL1 expression is associated with disease aggressiveness with functional experiments, we evaluated the expression levels in 3 different types of cell lines, TCCSUP (grade IV cancer), 5637 (grade II cancer), and T24 cells (high-grade and invasive TCC). The mRNA expression of CX3CL1 in TCCSUP cells was significantly higher than that in 5637 and T24 cells (P<0.001, Figure 2A ). No significant difference in CX3CL1 expression was observed between the 5637 and T24 cell lines (P=0.766). Thus, TCCSUP cells were used for further evaluation. CX3CL1 was then knocked down by lentiviral infection. The knockdown efficiency was then verified by qPCR (efficiency reached 56.5%, P=0.0032, Figure 2B ). We then evaluated whether CX3CL1 expression is associated with cancer cell proliferation and invasiveness. MTT assay results showed that TCCSUP cell proliferation was significantly decreased in the KD group compared to the NC group ( Figure 2C , P<0.001). Transwell assays suggested a significant decrease in cell invasion in the KD group compared with the NC group ( Figures 2D, E , P<0.001).

To further evaluate whether CX3CL1 expression is associated with BCa aggressiveness in vivo, we established xenografts in the shoulder of immunocompromised nude mice. Subcutaneous tumor formation in the nude mice was verified by pathologists, and the success rate was 100% ( Figure 3 ). From the 10th day to the 35th day, the tumor volume in the KD group was significantly smaller than that in the NC group ( Figure 3 , days 10-35, P<0.05). The tumor weight in the KD group was significantly lower than that in the NC group ( Figure 3 , day 35, P=0.017).

To investigate potential biological functions, we performed gene expression profiling using an mRNA microarray. Totals of 182 and 367 genes were found to be significantly upregulated and downregulated, respectively, in CX3CL1 KD TCCSUP cells. IPA-based disease and function analysis suggested that after CX3CL1 was inhibited, tumor cell proliferation (P=1.62×10^-11, z-score=-2.32), cell malignant transformation (P=6.47×10^-13, z-score=-2.289), tumor cell interphase (P=1.57×10^-12, z-score=-2.078) and differentiation of stem cells (P=4.02×10^-5, z-score=-2.465) were significantly inhibited, and tumor cell apoptosis (P=2.35×10^-7, z-score=2.236) was significantly increased ( Figure 4A ). CDH1, ETS1, RAF1, and EIF4E were genes with major effects in the enrichment analysis. In terms of classic pathway enrichment analysis based on IPA, the ERK/MAPK signaling pathway was significantly inhibited (P=2.19×10^-4, z-score=-1.604). ETS1, RAF1, and EIF4E were closely related to the ERK/MAPK pathway and significantly downregulated ( Figure 4A and Table S2 ). A volcano plot and hierarchical clustering for significant difference analysis are shown in Figure 4B . Briefly, the tumor cell apoptosis (P=2.35×10^-7, z-score=2.236) pathway was significantly activated, while tumor cell proliferation (P=1.62×10^-11, z-score=-2.32), tumor cell interphase (P=1.57×10^-12, z-score=-2.078), cell viability of tumor cell lines (P=7.12×10^-14, z-score=-2.092), differentiation of stem cells (P=4.02×10^-5, z-score=-2.465), cell malignant transformation (P=6.47×10^-13, z-score=-2.289), etc. were significantly inhibited.

To further confirm whether the genes in the ERK/MAPK pathway are influenced by the expression of CX3CL1, the protein expression of the enriched genes including RAF1, EIF4E, ETS1, and CDH1 was evaluated by Western blotting. The results suggested that RAF1, EIF4E, and ETS1 protein expression was significantly downregulated by 67.2%, 52.0%, and 43.6%, respectively (all P<0.05), while CDH1 protein expression was significantly upregulated by 74.6% (P<0.05, Figure 4C ).

Tumor immunotherapy is considered as an important and hopeful curative therapy for cancers, such as PD-1/PD-L1. Immune surveillance and immune tolerance usually serve with a two-side effect throughout cancer development and progression. Scientists have found that chemokines family was the key factor in this process by affecting the immune system (3). Among different chemokines, CX3CL1 is considered as one of the most important one, and may enhance local anti-tumor immunity by polarizing immunocompetent cells around the tumor microenvironment, and therefore inhibit tumor growth (4); on the other hand, tumors spontaneously secrete chemokines and receptors, which can promote the growth of cancer cells (5). We found that high expression of CX3CL1 was significantly associated with higher risk of disease recurrence and cancer-specific death; through in vivo and in vitro functional study, we further confirmed that high expression of CX3CL1 was associated with tumor cell proliferation and invasion. Enrichment analysis and functional study based on expression array data suggested that the regulation of tumor growth by CX3CL1 might due to the enhance of the ERK/MAPK signaling pathway.

CX3CL1 is a large cytokine (373 amino acids) and the only known member of the CX3C chemokine family. It is also known as fractalkine (in humans) and neurotactin (in mice) (6, 20). It has been proved correlating with cell migration and interaction in multiple diseases. Shulby et al. presented the first evidence of the expression of CX3CR1 (the specific receptor of fractalkine) in human prostate cancer cells and the expression of fractalkine in human bone marrow endothelial cells. They also found that a neutralizing antibody against fractalkine might significantly reduce the adhesion of prostate cancer cells to human bone marrow endothelial cells, suggesting the important role of fractalkine in this process (13). By silencing CX3CR1, Jamieson et al. further defined the roles of this chemokine receptor in the adhesion of prostate cancer cells to bone marrow endothelial cells and prostate cancer cell migration towards media conditioned by different types of bone cells (21). Tardaguila et al. found that CX3CL1 expression was downregulated in HER2/neu tumors; however, paradoxically, adenovirus-mediated CX3CL1 expression in the tumor milieu enhanced mammary tumor numbers in a dose-dependent manner. Additionally, CX3CL1 triggered cell proliferation by inducing ErbB receptors through the proteolytic shedding of an ErbB ligand, and these findings support the conclusion that CX3CL1 acts as a positive modifier of breast cancer in concert with ErbB receptors (22). In another study on breast cancer, Tsang et al. reported that high CX3CL1 expression was detected in 33.3% of primary invasive cancers and that CX3CL1 expression correlated positively with increased tumor-infiltrating lymphocytes (TILs; P=0.005). Furthermore, adverse features in breast cancers, including lymph node involvement, α-β crystalline expression and high Ki67, were positively correlated with the expression of CX3CL1. Notably, in the same study, the authors showed that patients with high levels of CX3CL1 had poorer overall survival. The findings indicated an oncogenic role for CX3CL1, despite its previously suggested role in enhancing antitumor immunity, and highlighted the complicated functions of CX3CL1 in breast carcinogenesis (23). However, a study on the prognosis of patients with liver cancer showed that patients with high expression of CX3CL1 and its receptor had a better prognosis than those with low expression, with a significantly lower local recurrence rate and distant metastasis rate and longer disease-free survival and overall survival (15). These results indicate that CX3CL1 has complex functions in tumors. At present, there are few in-depth studies on the role of CX3CL1 in BCa, and the mechanism of action of the chemokine CX3CL1 is still unclear. We conducted a series of experimental studies to investigate the role of CX3CL1 in BCa.

Previous studies have shown that CX3CL1 may promote or suppress tumors in different tissues. Our research confirms the value and function of CX3CL1 in BCa. Compared with previous studies, our study identified possible signaling pathways and downstream genes of CX3CL1 in BCa through GeneChip PrimeView Human technology and explored the possible mechanisms in more depth. Through bioinformatic data analysis and statistics, the significant differences were further verified. We found that after CX3CL1 was inhibited, tumor cell proliferation, cell malignant transformation, tumor cell interphase, and differentiation of stem cells were significantly inhibited, and tumor cell apoptosis was significantly increased. These results suggest that CX3CL1 may play a role in promoting the proliferation of BCa cells. In this experiment, the ERK/MAPK pathway was significantly inhibited, and the downstream genes ETS1 (↓), RAF1 (↓), and EIF4E (↓), which are closely related to this pathway, were significantly downregulated. In addition, the downstream gene CDH1 (↑) was significantly upregulated. Further detection of the expression of the proteins encoded by the above genes was also confirmed. The RNA-Seq results, and protein expression results are consistent with a previous clinical prognostic analysis and in vivo and in vitro experimental results, which together indicate the tumor-promotive effect of CX3CL1 in BCa.

This study suggests that CX3CL1 might promote the cancer progression through the ERK/MAPK signaling pathway. Mitogen-activated protein kinase (MAPK) is a threonine/serine protein kinase that is widely present in cells. It is closely related to cell proliferation, apoptosis, tumor proliferation, invasion and metastasis. The MAPK pathway is one of the most studied signaling pathways in recent years (24, 25). Among MAPK family members, extracellular signal-regulated kinase (ERK) was the first discovered and is the most widely studied member. An abnormal ERK/MAPK pathway can lead to abnormal cell proliferation, which is related to tumorigenesis and can further affect the biological behavior of tumors (26). Current studies indicate that the ERK/MAPK signaling pathway plays an important role in tumor occurrence and development by promoting cancer cell proliferation, inhibiting tumor apoptosis, and inducing tumor neovascularization, invasion, and metastasis (27–30). Studies have also pointed out that the ERK/MAPK signaling pathway is also related to tumor resistance (31, 32). To date, many studies have found that the ERK/MAPK signaling pathway is related to many tumors, such as colon cancer, gastric cancer, liver cancer, melanoma, and BCa (33–38). According to previous research reports, among the four significantly regulated downstream genes found in this experiment, CDH1 was considered to be an anti-oncogene, and ETS1, RAF1 and EIF4E play roles in promoting cancer (39–48). Our research results are consistent with those in previous research reports.

To the best of our knowledge, this was the first study investigating the association between CX3CL1 expression and BCa. The results showed that CX3CL1 overexpression was independently and significantly associated with poor prognosis (increased risk of recurrence and cancer-specific death rate). In addition, we were able to confirm our results in TCGA dataset. This finding may be applied to clinical practice as a valuable prognostic biomarker via additional translational research approaches.

Our study has carried out a series of elaboration on the role of CX3CL1 in BCa. However, several limitations should be noted in the present study. First, the sample size was relatively small. However, we were able to validate our results in TCGA database. The results should be investigated in a large-scale study in the future. Second, the subcutaneous xenograft animal model could not fully simulate the tumor microenvironment. Although orthotopic bladder cancer model would be a better option, the current method was the most common used model in BCa research. Last but not least, only one cell line was chosen in the present study. However, the main objective of this study was to evaluate the association of CX3CL1 expression and BCa prognosis.

CX3CL1 was a significant and independent predictor for bladder cancer prognosis. It could promote the proliferation and invasion of bladder cancer cells via inhibiting the ERK/MAPK signaling pathway.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by The Institutional Review Board at Huashan Hospital. The written informed consent was obtained from all patients for institutional biobank. The animal study was reviewed and approved by The Institutional Review Board at Huashan Hospital.

KX, RN, and NZ conceived and designed the study. GJ, HW, YW, WD, and QZ acquired the data. GJ, HW, and DH analyzed and interpreted the data. GJ, HW, and DH drafted the manuscript. KX, RN, NZ, and QD contributed to the critical revision of the manuscript. KX and RN supervised the study. All authors contributed to the article and approved the submitted version.

This research was supported by the National Natural Science Foundation of China (Grant No. 81472379 and 81402339).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

The reviewer HW declared a shared affiliation, with no collaboration, with several of the authors HW, YW, WD, QZ, QD, KX to the handling editor at the time of review.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.