1 INTRODUCTION

Esophageal cancer is one of the most common malignancies with the highest morbidity and mortality in the world, and it consists of two histological types, esophageal squamous cell carcinoma (ESCC) and esophageal adenocarcinoma.^1-4 Among them, ESCC is the main type of esophageal cancer, ranking sixth and fourth in the morbidity and mortality of all malignant tumors in China.^{5, 6} Although current chemotherapy, radiotherapy and surgery have improved the overall survival of ESCC patients, the prognosis of ESCC patients is poor owing to the high recurrence probability of ESCC and limited treatment methods. In China, the 5-year survival rate of ESCC patients is only approximately 20%.^{7, 8} Therefore, it is necessary to explore new biomarkers for ESCC diagnosis, prognosis and treatment.

Immunotherapy has achieved remarkable results in a variety of tumors, but the interaction between ESCC and the immune system is unclear.^{9, 10} Tumor-infiltrating lymphocytes (TILs) have been shown to have prognostic and predictive roles in ESCC, and TILs can be effective prognostic markers for ESCC.^{11, 12} TILs can affect the tumor microenvironment by triggering an inflammatory response to form an immunostimulatory, antitumor or protumor microenvironment.¹³ Studies have shown that the level of T helper cells is related to the prognosis of ESCC.¹⁴ Therefore, it is of great clinical significance to study the infiltration of immune cells in ESCC.

The SYNGR2 gene is located at 17q25.3 and is a member of the synaptogyrin family. SYNGR2 has a transcript length of 1.6 kb and is highly expressed in all organs and tissues except the brain.¹⁵ Current studies have indicated that SYNGR2 plays an important role in cellular exocytosis, the storage and transport of GLUT4 at the cytoplasmic membrane, and the formation and maturation of microvesicles in neuronal cells.¹⁶ A recent study showed that SYNGR2 was significantly up-regulated during infection with severe fever with thrombocytopenia syndrome bunyavirus. Overexpression of SYNGR2 could induce the formation of synaptic vesicle-like microbubbles in neuroendocrine cells. SYNGR2 triggers the formation of vesicles or the reconstruction of existing vesicles through interaction and aggregation with virtual non-structural proteins, including membrane rearrangement or the fusion of lipid droplets on non-structural proteins, so as to reshape lipid droplets into unique formations of inclusion bodies or virus-like structures, which is conducive to the spread of bunyavirus.^{17, 18} A study on the cell cycle and apoptosis of lymphocytes induced by the Aggregatibacter actinomycetemcomitans cytolethal distending toxin (cdt) showed that SYNGR2 was highly expressed in human macrophages, promoted the combination of cholesterol in the lipid-rich membrane microdomain and cdtb (the active subunit B of cdt), and internalized and translocated cdtb to intracellular compartments, which played a key role in the cytotoxicity of cdt.¹⁹ At present, the achievements regarding synapsin family members in the field of tumor research are limited. Chromophobe renal cell carcinoma and renal oncocytoma are two different types of renal tumors. They have strong morphological and genetic similarities, but the current techniques for detecting the differences between them are very limited. Tan et al. found that SYNGR3 was expressed in the cytoplasm of chromophobe renal cell carcinoma, but not in the cytoplasm of renal oncocytoma. Therefore, SYNGR3 may play an important role in the differential diagnosis of chromophobe renal cell carcinoma and renal oncocytoma.²⁰ However, the function of SYNGR2 in other diseases has been poorly studied, and there are no reports on the role of SYNGR2 in ESCC development and progression.

To better understand the role of the SYNGR2 gene in ESCC, we comprehensively assessed the relationship between the expression level of SYNGR2 and prognosis in the The Cancer Genome Atlas (TCGA), Gene Expression Omnibus (GEO) and Kaplan–Meier Plotter databases. Additionally, we validated the correlation between SYNGR2 and TILs in the TIMER and TISIDB databases. Our findings suggest that SYNGR2 can be a biomarker for predicting prognosis and immune cell infiltration in ESCC patients.

2 MATERIAL AND METHODS

3 RESULTS

3.2 SYNGR2 is highly expressed in ESCC

To explore the expression level of SYNGR2 in normal and tumor tissues, we downloaded and analyzed the expression levels of SYNGR2 mRNA in different tumors and normal tissues from the TCGA using the R package. The results showed that SYNGR2 expression was significantly higher in tumor tissues such as esophageal carcinoma, bladder urothelial carcinoma, breast cancer, colon adenocarcinoma, head and neck squamous cell carcinoma, glioblastoma multiforme, kidney renal clear cell carcinoma, kidney renal papillary cell carcinoma, liver hepatocellular carcinoma, lung adenocarcinoma, lung squamous cell carcinoma, pheochromocytoma and paraganglioma, prostate adenocarcinoma, rectum adenocarcinoma, stomach adenocarcinoma, thyroid carcinoma and uterine corpus endometrial carcinoma than in normal tissues. In addition, lower expression was observed in kidney chromophobe (Figure 1A). We used the TCGA database to evaluate the mRNA expression levels of SYNGR2 in ESCC patients and compared these with the expression levels in adjacent tissues. The results showed that the expression level of SYNGR2 in ESCC was significantly higher than that in paracarcinoma tissues (p < 0.001; Figure 1B). The results were verified in ESCC and paired paracarcinoma tissues (Figure 1C). Next, we used logistic analysis to analyze the relationship between different clinical parameters and the prognosis of patients, and we found that the T stage was associated with OS (Table 2). To analyze the correlation between the SYNGR2 expression and clinical characteristics in ESCC patients, we analyzed the mRNA expression levels of SYNGR2 in different clinical categories in the TCGA database. The association identified between SYNGR2 expression and clinical features in patients with ESCC is summarized in Table 3. The results showed that high expression of SYNGR2 was significantly related to T stage (p < 0.05) and OS (p < 0.05) (Figure 1D and Table 3). These data suggest that SYNGR2 is significantly upregulated in ESCC.

TABLE 2. Logistic analysis of the association between SYNGR2 expression and clinical characteristics

Characteristics	Total (N)	Odds ratio	p-Value
T stage (T3 and T4 vs. T1 and T2)	79	0.372 (0.146–0.917)	0.034
N stage (N1, N2 and N3 vs. N0)	78	1.900 (0.767–4.826)	0.169
M stage (M1 vs. M0)	73	2.118 (0.194–46.801)	0.548
Pathological stage (stages III and IV vs. stages I and II)	79	1.167 (0.450–3.043)	0.750
Histological grade (G2 and G3 vs. G1)	72	0.787 (0.245–2.477)	0.681
Age (>60 vs. ≤60)	82	1.113 (0.449–2.773)	0.817
Gender (male vs. female)	82	1.482 (0.432–5.432)	0.534
BMI (>25 vs. ≤25)	78	1.250 (0.434–3.693)	0.679
Smoker (yes vs. no)	79	0.929 (0.363–2.362)	0.876
Alcohol history (yes vs. no)	80	2.020 (0.713–6.090)	0.193
Radiation therapy (yes vs. no)	76	1.218 (0.481–3.101)	0.677
Residual tumor (R1 and R2 vs. R0)	71	0.456 (0.060–2.505)	0.383
Race (Black or African American and White vs. Asian)	80	1.008 (0.417–2.437)	0.987
Weight (>70 vs. ≤70)	81	0.727 (0.247–2.082)	0.553
Height (≥170 vs. <170)	78	0.875 (0.348–2.184)	0.775
Reflux history (yes vs. no)	67	1.940 (0.558–7.868)	0.315
Tumor central location (mid and proximal vs. distal)	81	1.421 (0.593–3.445)	0.432
Columnar mucosa dysplasia (high-grade dysplasia vs. negative/no dysplasia)	23	0.583 (0.058–5.759)	0.626

TABLE 3. Univariate and multivariate cox regression analyses of clinical characteristics associated with overall survival

Characteristics	Total (N)	Univariate analysis		Multivariate analysis
		Hazard ratio (95% CI)	p-Value	Hazard ratio (95% CI)	p-Value
T stage	79
T1 and T2	35	Reference
T3 and T4	44	0.942 (0.414–2.142)	0.887
N stage	78
N0	46	Reference
N1, N2 and N3	32	2.337 (1.007–5.425)	0.048	0.983 (0.287–3.367)	0.979
M stage	73
M0	70	Reference
M1	3	3.197 (0.909–11.238)	0.070	1.028 (0.247–4.275)	0.970
Pathological stage	79
Stages I and II	54	Reference
Stages III and IV	25	2.178 (0.958–4.952)	0.063	1.748 (0.563–5.430)	0.334
Histological grade	72
G1	15	Reference
G2 and G3	57	1.293 (0.434–3.851)	0.645
Smoker	79
No	27	Reference
Yes	52	1.661 (0.620–4.447)	0.313
Age	82
≤60	53	Reference
>60	29	1.592 (0.676–3.749)	0.288
Gender	82
Male	70	Reference
Female	12	0.100 (0.013–0.756)	0.026	0.000 (0.000-Inf)	0.998
BMI	78
≤25	60	Reference
>25	18	1.066 (0.434–2.619)	0.889
Residual tumor	71
R0	65	Reference
R1 and R2	6	2.454 (0.804–7.489)	0.115
Reflux history	67
No	54	Reference
Yes	13	1.348 (0.524–3.469)	0.536
Alcohol history	80
No	19	Reference
Yes	61	3.087 (0.722–13.204)	0.128
SYNGR2	82
Low	41	Reference
High	41	3.756 (1.546–9.128)	0.003	3.309 (1.116–9.812)	0.031

• Note: Values with p-value < 0.05 are in bold.

3.3 Validation using independent external databases and clinical specimens

To further verify the expression level of SYNGR2 in ESCC, we selected another three independent external GEO datasets (validation cohort) to analyze the SYNGR2 transcription levels of cancer tissues and adjacent tissues in SYNGR2, including GSE26886, GSE45670 and GSE161533. The results showed that the transcription level of SYNGR2 in ESCC was significantly higher than that in non-cancerous adjacent tissues from ESCC datasets (GSE26886, GSE45670 and GSE161533, p < 0.001; Figure 2A–C). The results of this comparison were also verified in ESCC tissues and accurately matched non-cancerous tissues (GSE161533, p < 0.001; Figure 2D). These data verify that SYNGR2 is highly expressed in ESCC.

3.4 High expression of SYNGR2 is an independent prognostic factor for the overall survival of ESCC

To identify whether SYNGR2 expression affects patient survival, we classified ESCC patients in the TCGA database into a high SYNGR2 expression group (the top 50% of samples with the highest expression) and a low SYNGR2 expression group (the remaining 50% of the samples) to perform survival analysis according to the mean expression value of SYNGR2. The Kaplan–Meier survival analysis showed that the high expression of SYNGR2 was related to the overall survival (HR = 4.34, p = 0.002) and disease-specific survival (HR = 3.64, p = 0.015) of ESCC patients (Figure 3A and B). Subgroup analysis showed that high SYNGR2 expression was significantly correlated with poor prognosis in ESCC in the following cases: T1, T2 and T3 stages, HR = 4.23, p = 0.006; T2 and T3 stages, HR = 3.67, p = 0.015; T3 and T4 stages, HR = 4.58, p = 0.025; N0 stage, HR = 5.20, p = 0.044; pathological stages I and II, HR = 4.91, p = 0.023; pathological stages II and III, HR = 5.10, p = 0.005; histological grades 1 and 2, HR = 4.31, p = 0.023; histological grades 2 and 3, HR = 3.53, p = 0.029; male sex, HR = 3.37, p = 0.008; and age ≤60 years old, HR = 3.66, p = 0.024. These data are shown in Figure 3C–L. Univariate Cox analysis demonstrated that high SYNGR2 expression was significantly correlated with poor OS (HR = 3.309, 95% CI = 1.126–9.812, p = 0.031; Figure 4, Table 3). These data suggest that high expression of SYNGR2 is an independent prognostic factor for the OS of ESCC patients.

3.5 Diagnostic value of SYNGR2 expression in ESCC

To analyze the diagnostic value of SYNGR2 expression in ESCC, we drew the ROC curve and performed nomogram analyses on the SYNGR2 gene expression data from the TCGA database to evaluate the diagnostic value of the gene. The area under the ROC curve (AUC) was 0.881, suggesting a higher diagnostic value, as shown in Figure 5A. A time-dependent survival ROC curve of SYNGR2 was created to predict the 1, 3 and 5 year survival rates. All of these AUC values were >0.55, which is considered suitable for prediction (Figure 5B). Then, we combined the expression level of SYNGR2 with the clinical variables to construct a nomogram to predict the survival probability of patients at 1, 3 and 5 years. The nomogram indicated that the prognostic prediction of the expression level of SYNGR2 was better than that of the traditional clinical features of age, T stage, M stage and histological grade (Figure 5C).

3.6 Functional enrichment and analyses of SYNGR2-related genes in ESCC

To understand the biological function of SYNGR2 in ESCC, we used the LinkFinder module of the LinkedOmics website to detect the coexpression pattern of SYNGR2 in ESCC in the TCGA database. The red dot indicates the top 25 genes that were positively correlated with SYNGR2, and the green dot represents the bottom 25 genes that were negatively correlated with SYNGR2 (Figure 6A). We used DAVID Functional Annotation Bioinformatics Microarray Analysis to identify the enriched GO functional enrichment and KEGG pathways among the SYNGR2-related genes (top 600) and found that the steroid metabolic process, small molecule catabolic process and steroid biosynthetic process were enriched among these genes (Figure 6B and Table 4). To identify genes with the same regulatory direction in high SYNGR2 and non-survival patients, we crossed the 471 genes (the absolute value of correlation coefficient ≥0.3) with the highest SYNGR2 correlation with the 913 survival-related upregulated genes (p Cox < 0.05) in ESCC and detected 38 genes at the intersection that were related to SYNGR2 and ESCC survival (Figure 6C). These 38 protein-coding genes may be potential genetic biomarkers for ESCC patients. GO functional enrichment and KEGG pathway analysis were performed for these 38 genes, and the results showed that differentially expressed genes were significantly enriched in retinal cone cell differentiation, retinal cone cell development and protein polyubiquitination (Figure 6D and Table 5). After discovering significantly different pathways, protein–protein interactions and correlated analysis were used to identify the interactions between these 38 proteins. We found that there was a stronger enrichment network among these proteins than in random proteins (Figure 6E). Gene coexpression correlation analysis showed that most of the proteins in the network had a strong positive correlation with each other (Figure 6F). Therefore, these SYNGR2-associated established genes have a strong intertwined interaction and can be used as multigene biomarkers to predict the survival of ESCC patients.

TABLE 4. Gene sets enriched in the SYNGR2-related genes (top 600)

Ontology	ID	Description	Gene ratio	Bg ratio	p-Value	p-Adjust	q-Value
BP	GO:0008202	Steroid metabolic process	26/528	331/18670	2.91 × 10−6	0.012	0.011
BP	GO:0044282	Small molecule catabolic process	30/528	445/18670	1.11 × 10−5	0.017	0.016
BP	GO:0006694	Steroid biosynthetic process	18/528	196/18670	1.22 × 10−5	0.017	0.016
BP	GO:0010498	Proteasomal protein catabolic process	30/528	477/18670	4.12 × 10−5	0.042	0.041
BP	GO:0031146	SCF-dependent proteasomal Ubiquitin-dependent protein catabolic process	11/528	95/18670	7.62 × 10−5	0.057	0.055
CC	GO:0005774	Vacuolar membrane	27/558	412/19717	5.09 × 10−5	0.021	0.019
CC	GO:0098852	Lytic vacuole membrane	24/558	355/19717	8.15 × 10−5	0.021	0.019
CC	GO:0098798	Mitochondrial protein complex	19/558	262/19717	1.81 × 10−4	0.021	0.019
CC	GO:0000152	Nuclear ubiquitin ligase complex	7/558	43/19717	1.86 × 10−4	0.021	0.019
CC	GO:0005765	Lysosomal membrane	23/558	354/19717	2.04 × 10−4	0.021	0.019
MF	GO:0016655	oxidoreductase activity, acting on NAD(P)H, quinone or similar compound as acceptor	10/530	60/17697	1.05 × 10−5	0.005	0.004
MF	GO:0016616	Oxidoreductase activity, acting on the CH-OH group of donors, NAD or NADP as acceptor	14/530	119/17697	1.30 × 10−5	0.005	0.004
MF	GO:0004032	Alditol: NADP+ 1-oxidoreductase activity	5/530	13/17697	2.49 × 10−5	0.005	0.005
MF	GO:0016614	Oxidoreductase activity, acting on CH-OH group of donors	14/530	128/17697	3.00 × 10−5	0.005	0.005
MF	GO:0016627	Oxidoreductase activity, acting on the CH–CH group of donors	9/530	58/17697	5.22 × 10−5	0.007	0.007
KEGG	hsa05012	Parkinson disease	23/245	249/8076	1.72 × 10−6	4.73 × 10−4	4.35 × 10−4
KEGG	hsa05010	Alzheimer disease	28/245	369/8076	6.23 × 10−6	8.60 × 10−4	7.90 × 10−4
KEGG	hsa05020	Prion disease	21/245	273/8076	7.86 × 10−5	0.007	0.006
KEGG	hsa05022	Pathways of neurodegeneration—multiple diseases	30/245	475/8076	9.78 × 10−5	0.007	0.006
KEGG	hsa00190	Oxidative phosphorylation	13/245	133/8076	1.88 × 10−4	0.010	0.010

• Abbreviations: BP, biological process; Bg, background; CC, cellular component; ES, enrichment score; FDR, false discovery rate; KEGG, Kyoto Encyclopedia of Genes and Genomes; MF, molecular function; NOM, nominal. Gene sets with NOM p-value < 0.05 and FDR q-value < 0.25 were considered as significantly enriched.

TABLE 5. Gene sets enriched in the 38 genes at the intersection that were related to SYNGR2 and esophageal squamous cell carcinoma (ESCC) survival

Ontology	ID	Description	Gene ratio	Bg ratio	p-Value	p-Adjust	q-Value
BP	GO:0042670	Retinal cone cell differentiation	2/34	11/18670	1.75 × 10−4	0.061	0.050
BP	GO:0046549	Retinal cone cell development	2/34	11/18670	1.75 × 10−4	0.061	0.050
BP	GO:0000209	Protein polyubiquitination	5/34	310/18670	2.29 × 10−4	0.061	0.050
BP	GO:1902036	Regulation of hematopoietic stem cell differentiation	3/34	72/18670	3.02 × 10−4	0.061	0.050
BP	GO:0060218	Hematopoietic stem cell differentiation	3/34	83/18670	4.59 × 10−4	0.063	0.051
MF	GO:0015078	Proton transmembrane transporter activity	3/32	133/17697	0.002	0.093	0.066
MF	GO:0044769	ATPase activity, coupled to transmembrane movement of ions, rotational mechanism	2/32	36/17697	0.002	0.093	0.066
MF	GO:0019842	Vitamin binding	3/32	138/17697	0.002	0.093	0.066

• Abbreviations: BP, biological process; CC, cellular component; ES, enrichment score. Gene sets with NOM p-value < 0.05 and FDR q value < 0.25 were considered as significantly enriched.

3.7 Correlation of SYNGR2 expression with immune characteristics

To explore the correlation between the expression level of SYNGR2 and the tumor immune response, we used the TIMER database to investigate immune infiltration in ESCC with different SYNGR2 expression levels. The expression of SYNGR2 was negatively correlated with T helper cells (Figure 7A). We further analyzed the correlation between the expression level of SYNGR2 and immune infiltration in ESCC, and the results showed that the expression level of SYNGR2 was negatively correlated with the infiltrating levels of T helper cells (r = −0.264, p = 0.017; Figure 7B). Therefore, we used TISIDB to analyze the correlation between the methylation level of SYNGR2 and immune cell chemokines and chemokine receptors in ESCC. The heatmap results showed that several chemokines and chemokine receptors were significantly correlated with the methylation level of SYNGR2 in ESCC (Figure 8A, B). These results demonstrate that SYNGR2 methylation may play an important role in tumor immunity. To further clarify the relationship between SYNGR2 methylation and immune cell migration, we comprehensively analyzed the correlation between SYNGR2 methylation and chemokine/chemokine receptors. The results showed that SYNGR2 methylation was positively correlated with CCL2 (r = 0.404, p < 1.6 × 10⁻⁸), CXCL12 (r = 0.393, p < 4.04 × 10⁻⁸), CCR1 (r = 0.321, p < 9.65 × 10⁻⁶) and CCR2 (r = 0.337, p < 3.21 × 10⁻¹⁴; Figure 8C–F). These data indicated that SYNGR2 methylation is positively correlated with the expression of chemokines/chemokine receptors in ESCC. Immune checkpoint inhibitors are a significantly novel strategy for tumor immunotherapy that has gradually improved the prognosis of patients with many types of cancers. Subsequently, we analyzed the correlation between SYNGR2 methylation and the expression of immunoinhibitors and immunostimulators in different types of human cancers using the TISIDB database (Figure 9A, B). Interestingly, SYNGR2 methylation was positively correlated with CCL12 (r = 0.393, p < 4.04 × 10⁻⁸), TNFRSF4 (r = 0.338, p < 3 × 10⁻⁶), CSF1R (r = 0.395, p < 3.5 × 10⁻⁸) and PDCD1LG2 (r = 0.348, p < 1.46 × 10⁻⁶) (Figures 9C–F). Therefore, these results suggest that SYNGR2 may play a role in regulating tumor immunity.

4 DISCUSSION

SYNGR2 plays an important role in biological processes such as cell exocytosis, the storage and transport of GLUT4 in the cytoplasmic membrane, and the formation and maturation of microvesicles in neuronal cells.¹⁶ A recent study indicated that SYNGR2 can be used as a promoter to enhance the infectivity of novel tick-borne bunyaviruses in humans.¹⁷ However, there are no reports of SYNGR2 in tumors. We integrated multiple bioinformatic analysis methods to determine the biological functions and potential regulatory pathways in ESCC. In this study, we first determined the expression and prognostic value of the SYNGR2 gene in cancer and found that the expression level of SYNGR2 was upregulated in ESCC. High SYNGR2 expression was associated with poorer OS in ESCC. Moreover, a high SYNGR2 expression level was associated with poorer T stage and poorer overall survival and disease-specific survival. In addition, we also found that SYNGR2 has an important reference value for the diagnosis of ESCC. These findings strongly suggest that SYNGR2 can be used as a biomarker for ESCC diagnosis and prognosis.

To explain the underlying molecular mechanism by which SYNGR2 affects ESCC prognosis, we defined 38 genes in the SYNGR2 gene network as our hub genes, which included 913 genes closely associated with ESCC prognosis and 471 genes with the highest correlation with SYNGR2 intersection. Then, we further analyzed the biological processes and signaling pathways of the hub genes using GO functional enrichment and KEGG pathway analyses. The enrichment results suggested that most of the hub genes were related to biological processes such as protein polyubiquitination. Protein polyubiquitination is a common biological process in cells, and a large number of studies have suggested that it plays an important role in the progression of ESCC.^{30, 31} A study indicated that the protein FLOT1 in ESCC cells promotes the recruitment of tumor necrosis factor-alpha receptors to lipid rafts, promotes polyubiquitination of tumor necrosis factor receptor-associated factor 2 and receptor-interacting proteins, and persists in activating NF-κB.³² Zhao et al. pointed out that the protein FAM175B inhibits ATF4 ubiquitination and promotes ESCC cell apoptosis in a p53-independent manner.³³ Therefore, SYNGR2 may be involved in regulating ESCC progression by regulating protein polyubiquitination.

Numerous studies have pointed out that the development of cancer is closely related to the tumor microenvironment.³⁴ The tumor microenvironment is composed of immune cells, extracellular matrix and inflammatory mediators, which interact with each other to promote the growth, survival and metastasis of tumor cells.^{35, 36} Although an effective immune response can produce antitumor effects, cancer cells can evade attack by immune cells through antigen presentation dysfunction and the recruitment of immunosuppressive cells.^{37, 38} Previous studies have reported that the degree of immune cell infiltration in tumors can affect the prognosis of patients, and the grade of tumor-infiltrating lymphocytes is an independent predictor of the prognosis of tumor patients.^{39, 40} This study explored the correlation between SYNGR2 expression and the level of immune infiltration in ESCC. Our results showed that the expression of SYNGR2 was negatively correlated with the infiltration level of T helper cells. Studies have shown that the frequency of Th cells in ESCC patients is significantly reduced, and the average expression of IFN-γ in Th cells is also greatly reduced, indicating that the dysregulation of Th-cell subsets contributes to the occurrence and development of ESCC.^{14, 41, 42} These findings suggest that SYNGR2 may play an important regulatory role in the tumor immune microenvironment and ESCC development.

Chemokines and their receptors play an important role in the directed migration of immune cells. In this study, the TISIDB database was used to analyze the correlation between the methylation level of SYNGR2 and the expression of immune cell chemokines and chemokine receptors in ESCC. The results showed that the methylation level of SYNGR2 was positively correlated with the expression of CCL2, CXCL12, CCR1 and CCR2, suggesting that the methylation of SYNGR2 may inhibit the migration of immune cells to the tumor microenvironment. The CCL2–CCR2 axis, one of the major chemokine signaling pathways, increases tumor cell proliferation and invasiveness during tumor progression and creates the tumor microenvironment by increasing angiogenesis and recruitment of immunosuppressive cells.^43-45 The interaction of CXCL12 and its receptor CXCR4 stimulates downstream signaling pathways, which affect tumor angiogenesis, tumor cell proliferation and chemotherapy resistance.^46-48 In addition, studies have pointed out that CCL3/CCR1 can promote the chemotaxis of monocytes in tumors and affect tumor angiogenesis and cell proliferation.^{49, 50} This may explain the mechanism by which SYNGR2 regulates immune infiltration in ESCC.

In conclusion, our study found that the expression of SYNGR2 was significantly upregulated and closely related to the poor prognosis of ESCC patients. SYNGR2 has a certain reference value for the diagnosis and prognosis of ESCC. SYNGR2 may affect ESCC progression through immune infiltration and it could serve as a novel premarker for ESCC patients.

However, even though we systematically analyzed SYNGR2 and performed cross-validation using different databases, this study has limitations. First, the role of SYNGR2 in ESCC may vary in the reproducibility of microarray data generated by different laboratories. Second, we need in vivo and in vitro experiments to demonstrate the effect of SYNGR2 on ESCC to increase the reliability of our results. Third, our results show that the AUC area of SYNGR2 is 0.881, indicating that SYNGR2 has a good diagnostic effect on esophageal cancer. At present, the relevant experiments for detecting SYNGR2 in peripheral blood and liver tissues are in progress, and the relevant research results will be published in the follow-up research reports. Finally, although we concluded that SYNGR2 is closely associated with immune infiltration and prognosis in ESCC, we lack direct evidence that SYNGR2 affects prognosis through its involvement in immune infiltration. We will further explore these questions in future experiments.

CONFLICT OF INTEREST

The authors declare that they have no competing interests.

AUTHOR CONTRIBUTIONS

B.L., MY.R., YZ.C. and YQ.M. wrote the main manuscript text and prepared Figures 1-9, B.L., TN.S., ZP.S. and B.Y. prepared Tables 1–5. All authors reviewed the manuscript.