Dysregulation of ECRG4 is associated with malignant properties and of prognostic importance in human gastric cancer

Abstract

BACKGROUND:

We have previously characterized esophageal carcinoma-related gene 4 (ECRG4) as a novel tumor suppressor gene, which is frequently inactivated in nasopharyngeal carcinoma and breast cancer. Nevertheless, the expression status and prognostic significance of ECRG4 maintain elusive in human gastric cancer. Herein, we examined ECRG4 expression profile in gastric cancer and assessed its association with clinicopathological characteristics and patient survival.

METHODS:

Online data mining, real-time RT-PCR and immunohistochemistry were employed to determined ECRG4 expression at transcriptional and protein levels in tumors vs. noncancerous tissues. Statistical analyses including the Kaplan-Meier survival analysis and the Cox hazard model were utilized to detect the impact on clinical outcome. Moreover, ECRG4 expression was silenced in gastric cancer SGC7901 cells, and cell proliferation, colony formation and invasion assays were carried out.

RESULTS:

ECRG4 mRNA and protein levels were obviously downregulated in cancer tissues than noncancerous tissues. Statistical analyses demonstrated that low ECRG4 expression was found in 34.5% (58/168) of primary gastric cancer tissues, which was associated with higher histological grade ( $P=$ 0.018), lymph node metastasis ( $P=$ 0.011), invasive depth ( $P=$ 0.020), advanced tumor stage ( $P=$ 0.002) and poor overall survival ( $P<$ 0.001). Multivariate analysis showed ECRG4 expression is an independent prognostic predictor ( $P<$ 0.001). Silencing ECRG4 expression promoted gastric cancer cell growth and invasion. Western blot analysis revealed the anti-metastatic functions of ECRG4 by downregulating of E-cadherin and $\alpha$ -Catenin, as well as upregulating N-cadherin and Vimentin.

CONCLUSIONS:

Our observations reveal that ECRG4 expression is involved in gastric cancer pathogenesis and progression, and may serve as a candidate prognostic biomarker for this disease.

1. Introduction

Gastric cancer is the fourth most common type of cancer worldwide in terms of incidence and second in terms of mortality [1, 2]. The prognosis is heavily influenced by recurrence and metastasis. Surgery has so far been the only curative treatment choice for gastric cancer, and it must be conducted promptly. Currently, the 5-year overall survival rate of gastric cancer remains no more than 25% [2]. The molecular mechanisms underlying gastric tumorigenesis are mostly unknown. Consequently, developing novel molecular biomarkers during carcinogenesis and progression is urgently required to more accurately predict patient outcomes and individualize treatments for gastric cancer.

Esophageal carcinoma-related gene 4 (ECRG4), also called augurin or C2ORF40, was originally characterized via comparing gene expression differentiation between human normal esophageal epithelium and cancer tissues [3]. ECRG4 has been identified as a candidate tumor suppressor gene which is frequently inactivated in various cancer types, including human esophageal cancer, thyroid carcinoma, prostate cancer, breast cancer, gastric cancer, hepatocellular carcinoma, renal cell cancer, colorectal carcinomas and glioma, mainly due to its gene promoter methylation [4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15]. Consistently, we previously stated that promoter methylation-mediated silencing of ECRG4 in nasopharyngeal carcinoma and its overexpression inhibits the malignant phenotypes of cancer cells [16]. In addition, we observed that decreased ECRG4 expression is significantly correlated with unfavorable patient survival and is an independent prognostic indicator in breast cancer [17]. Our results suggest that ECRG4 might not only offer novel tailor-made therapeutic targets for cancers, but may also contribute to prediction of patient outcomes. Therefore, in the current study we first examined ECRG4 expression profiles in primary gastric cancer tissues. Following that, we analyzed these results in combination with clinicopathologic parameters and survival data. Finally, we elucidated the biological functions of silencing ECRG4 regarding cancer cell proliferation and invasion.

2. Materials and methods

2.1 Patients and tissue samples

20 gastric cancer tissues and paired adjacent noncancerous mucosa were collected from an independent cohort of gastric cancer patients undergoing endoscopy in 2020. These specimens were obtained and immediately processed using an RNALater reagent (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA) and kept at $-$ 80 ${}^{\circ }$ C for RNA extraction, reverse transcription and Real-time quantitative PCR analysis. In addition, two tissue microarray chips were purchased from Outdo Biotech Co., Ltd. (Shanghai, China) for immunohistochemistry analysis, providing primary gastric cancer tissues along with their paired adjacent noncancerous tissues from 180 patients undergoing curative surgery between February 2008 and June 2010. The median follow-up time was 41 months (range, 1–88 months) from the date of surgery. During follow-up, 65.6% (118/180) of patients died due to recurrence or metastasis. Tumor grade and stage were identified in accordance with the Union of International Cancer Control (UICC)/American Joint Committee on Cancer (AJCC) pathologic tumornodemetastasis (TNM) classification, 7th edition (2010). The clinicopathological parameters are summarized in Table 1. None of patients had received neoadjuvant therapy before surgery. Signed informed consents were received from patients according to the principles stated in the Declaration of Helsinki. The current study was approved by the Institution Review Board of the People’s Hospital of Ningxia Hui Autonomous Region.

Table 1

Correlation between ECRG4 expression and clinicopathological variables

Parameters	No. of patients	ECRG4 expression		${\chi }^2$	$P$ value
		Low, $n$ (%)	High, $n$ (%)
Age (years)
$<$ 65	66	23	43	0.005	0.005
$⩾$ 65	102	35	67
Sex
Female	48	22	26	3.663	0.056
Male	119	36	83
Null	1
Histological grade
I/II	67	16	51	5.585	0.018
III/IV	101	42	59
T classification
T1/T2	24	3	21	5.456	0.020
T3/T4	130	48	82
Null	14
N classification
N0	47	9	38	6.531	0.011
N1–N3	120	48	72
Null	1
M classification
M0	165	58	107	1.611	0.204
M1	3	0	3
Clinical stage (pTNM)
I/II	65	13	52	9.257	0.002
III/IV	95	41	54
Null	8

2.2 Online data mining analyses

ECRG4 mRNA expression was examined in various cancer types and cancer cell lines using the ONCOMINE database (http://www.oncomine.org) [18, 19] and the Cancer Cell Line Encyclopedia (CCLE) database (http://portals.broadinstitute.org/ccle) [20], respectively. The transcriptional expression difference between gastric cancer tissues and normal tissues was characterized using the UALCAN database (http:// ualcan.path.uab.edu) [21] and the Gene Expression Profile Interactive Analysis (GEPIA) database (http:// gepia.cancer-pku.cn) [22]. The genomic alteration profile of ECRG4 containing gene mutations and copy number variance (CNV) in gastric cancer as well as their impacts on overall survival, disease-free survival, progression-free survival and disease-specific survival were examined using the cBioPortal database (www. cbioportal.org) [23, 24]. The distribution and prognostic implications of ECRG4 CNV in gastric cancer were determined using the Gene Set Cancer Analysis (GSCA) database (http://portals.broadinstitute.org/ccle) [25].

2.3 RNA extraction and reverse transcription- quantitative polymerase chain reaction (RT-qPCR)

Total RNA was extracted from cells transfected with ECRG4 siRNAs or scrambled siRNA using a TRIzol reagent (Invitrogen) and was reversely transcribed using an M-MLV First Strand kit (Invitrogen) according to the manufacturer’s protocols. The cDNA was subjected to qPCR with the following gene-specific primers: ECRG4 forward 5 ${}^{\prime }$ -CCGGTTCTCCCTCGCAGCAC-3 ${}^{\prime }$ and reverse 5 ${}^{\prime }$ -CGCTTCTGGCGCTTCAGGCT-3 ${}^{\prime }$ ; GADPH forward 5 ${}^{\prime }$ -GACCCCTTCATTGACCTCAAC-3 ${}^{\prime }$ and reverse 5 ${}^{\prime }$ -CTTCTCCATGGTGGTGAAGAC-3 ${}^{\prime }$ .

2.4 Western blot analyses

This procedure has been described before [16, 17, 20, 26, 27, 28, 29, 30]. In brief, equivalent amounts of proteins extracted from cell lysates were separated using 10% SDS-PAGE and then transferred to PVDF membranes (Millipore, Billerica, MA, USA). After blocked in 5% skim milk in TBS containing 0.1% Tween-20 (TBST) buffer, the membranes were were incubated with the indicated primary antibodies overnight at 4 ${}^{\circ }$ C, i.e., rabbit polyclonal antibodies against ECRG4 (Thermo Fisher Scientific, Inc.), E-cadherin and N-cadherin (Santa Cruz Biotechnology, Inc., Dallas, TX, USA) and mouse monoclonal antibodies against Vimentin (Abcam, Cambridge, MA, USA). The horseradish peroxidase-con- jugated secondary antibodies against rabbit IgG (Abcam) or mouse IgG (Abcam) were added for 2 h at room temperature. Signals were captured using an enhanced chemiluminescence system following the manufacturer’s instructions (GE Healthcare Life Sciences, Chalfont, UK). The blots were reprobed with a rabbit polyclonal antibody against GAPDH (Abcam) to confirm equal loading of different samples.

2.5 Immunohistochemistry and evaluation

Immunohistochemistry (IHC) was performed using a standard EnVision complex method described previously [17, 20, 23, 29, 30]. Briefly, after undergoing deparaffinization and rehydration, endogenous peroxidase blocking and antigen retrieval, 4- $\mu$ m sections were cut from specimens were incubated with a rabbit polyclonal antibody against ECRG4 (Thermo Fisher Scientific, Inc.) at a dilution of 1:400. IHC was perform using an EnVision antibody complex (anti-mouse/rabbit) method with an Envision™ Detection kit (ZSGB-BIO, Beijing, China) and 3,3 ${}^{\prime }$ -diaminobenzidine as the chromogen substrate. Nuclei were counterstained with hematoxylin. Sections immunostained with a normal rabbit IgG as the primary antibody were employed as negative controls.

The staining evaluation was conducted using the following standard: ten random 400 $\times$ fields per slide were assessed by two independent pathologists blinded to the clinical information. ECRG4 immunostaining was determined by semi-quantitative method synthesizing the intensity and percentage of positively stained cells. The mean percentage of positively stained cells was scored as follows: 0% (0); 1–25% (1); 26–50% (2); 51–75% (3); and 76–100% (4). The staining intensity was classified as follows: absent (0); weak staining exhibited as light-yellow (1); moderate staining exhibited as yellow brown (2); and strong staining exhibited as brown (3). The multiplication of percentage and intensity scores was utilized as the final staining score. For statistical assessment, specimens with a final staining score of $⩽$ 3 were designated as low expression and those with a final score $>$ 3 as high expression.

2.6 Cell culture and transfection

One established gastric cancer cell line SGC-7901 were cultured as described before [30]. A pre-designed validated siRNA against 5 ${}^{\prime }$ -AACGAGAAGCACCTGTTC-3 ${}^{\prime }$ sequence of human ECRG4 was synthesized by Genetimes ExCell Tech Company (Shanghai, China). Cell transfection was performed using a Lipofectamine 2000 reagent (Invitrogen) in line with the manufacturer’s protocol. Cells were harvested 48 h post-transfection for the following experiments mentioned below.

2.7 Cell proliferation and colony formation assays

Cell proliferative and clonogenic capacities were determined using a 3-(4,5-dimethylthiazol-2-yl)-2 5-diphenyl-2H-tetrazolium bromide (MTT) colorimetric assay and a colony formation (soft agar culture) assay, respectively, as described previously [16, 26, 27, 28].

2.8 Cell invasion assay

Cell invasiveness was evaluated using a Matrigel invasion chamber assay as described before [6, 16, 26, 27, 28]. Briefly, cells were plated into 24-well Matrigel-coated Transwell inserts (EMD Millipore, Billerica, MA, USA) at the density of 1 $\times$ 10 ${}^5$ cells per 500 $\mu$ L of Dulbecco’s modified Eagle’s medium (Gibco; Thermo Fisher Scientific, Inc.). The lower chamber of the insert was filled with 750 $\mu$ l of DMEM with 10% fetal bovine serum (Gibco; Thermo Fisher Scientific, Inc.). After 24 h incubation, cells on the underside surfaces were fixed with 4% paraformaldehyde, stained with 2.5% crystal violet and counted in five random fields under the microscope.

2.9 Statistical analyses

Statistical analyses were conducted using the SPSS 26.0 statistical software package (SPSS Inc., Chicago, IL, USA). The difference of ECRG4 expression between tumors and nancancerous tissues was calculated by Student’s $t$ -test. Correlations between ECRG4 expression and clinicopathologic features were assessed using a ${\chi }^2$ test. Survival curves were evaluated using the Kaplan-Meier method with a log-rank test. The impacts of ECRG4 expression and clinicopathological features on patient survival were identified by univariate analysis and then assessed by multivariate regression analysis with a Cox regression model. A $P$ -value $<$ 0.05 (two-tailed) was defined as statistically significantly different.

Figure 1.

ECRG4 is down-regulated in gastric cancer tissues. (A) The mRNA levels of ECRG4 in different cancer types. A graphic obtained from Oncomine indicates the numbers of datasets with significant over-expression (Red) or under-expression (Blue) at the transcriptional level in cancer tissues relative to normal tissues. Cell color was determined by the best gene rank percentile for the analyses within the cell, and the gene rank was analyzed by percentile of target genes in the top of all genes measured in each analysis. (B) ECRG4 was distinctively high expressed in gastric cancer cell lines according to CCLE analysis. (C) ECRG4 mRNA levels in gastric cancer were analyzed using the UALCAN database. NG, normal gastric ( $n=$ 34); GC, gastric cancer ( $n=$ 415). (D) ECRG4 mRNA levels were analyzed using the GEPIA database. NG ( $n=$ 211); GC ( $n=$ 408). (E) ECRG4 mRNA levels were determined by RT-qPCR using gastric cancer tissues and paired noncancerous tissues. NG ( $n=$ 20); GC ( $n=$ 20). (F) The immunohistochemistry score of ECRG4 was plotted. NG ( $n=$ 168); GC ( $n=$ 168). * $P<$ 0.05, ** $P<$ 0.01, *** $P<$ 0.001 by Student’s $t$ -test.

Figure 2.

Genetic alterations of the ECRG4 gene and its association with patient survival in gastric cancer via the cBioPortal database analysis. (A) Oncoprint in cBioPortal represented the proportion and distribution of samples with ECRG4 gene alterations. The figure was cropped on the right to exclude samples without alterations. (B) Alterations frequency of the ECRG4 gene (mutation and amplification) in subtypes of gastric cancer. (C-F) The impacts of genetic alterations of ECRG4 on overall survival (C), disease-free survival (D), progression-free survival (E) and disease-specific survival (F) in patients with gastric cancer. (G) Analysis of the top 10 genes with concomitant alteration of ECRG4.

Figure 3.

Copy number variance (CNV) analysis of the ECRG4 gene in gastric cancer as well as their impacts on overall survival. (A) Gene copy number of ECRG4 in GC was obtained from the Oncomine database. (B) CNV distribution in stomach adenocarcinoma (STAD) specimens. (C) Spearman correlation analysis between CNV and mRNA expression of ECRG4 in STAD. (D) Relationship between ECRG4 CNV and overall survival in STAD patients.

Figure 4.

Immunohistochemical staining of ECRG4 in human gastric cancer tissues and adjacent noncancerous tissue. (A) Strong staining of ECRG4 in gastric cancer tissue. (B) Moderate ECRG4 staining in gastric cancer tissue. (C) Negative ECRG4 staining in gastric cancer tissue. (D) Positive ECRG4 staining in adjacent noncancerous tissue. Original magnification, 400 $\times$ .

3. Results

3.1 Differential expression analysis of ECRG4 in gastric cancer

We first performed differential expression analysis of ECRG4 in primary gastric cancer and adjacent noncanerous tissues using several online databases. As displayed in Fig. 1A, the Oncomine database revealed lower ECRG4 mRNA level in 24.6% of (64/260) unique analyses. Regarding gastric cancer, all four unique analyses indicated its down-regulation. The CCLE database analysis revealed that ECRG4 mRNA expression level in gastric cancer cell lines are listed in the 29 ${}^{\text{th}}$ highest position among multiple cancer types (Fig. 1B). In addition, the UALCAN and GEPIA databases demonstrated lower ECRG4 transcripts in cancer tissues than in normal tissues ( $P<$ 0.01 and $P<$ 0.05, respectively; Fig. 1C and D). We subsequently performed RT-qPCR and IHC analyses on gastric cancer tissues and paired adjacent noncancerous tissues to corroborate the findings above. Accordingly, noncancerous tissues exhibited higher ECRG4 expression than cancer tissues at mRNA and protein levels ( $P<$ 0.001 and $P<$ 0.01, respectively; Fig. 1E and F). Collectively, these observations clearly suggest that ECRG4 is downregulated in cancer tissues compared to normal controls, consistent with our hypothesis that ECRG4 might function as a tumor suppressor in gastric cancer.

3.2 Impact of the ECRG4 gene alterations on patient survival

We next examined the gene alteration status of ECRG4 and its impact on patient survival in gastric cancer using the GSCA online database. The genetic alteration rate for the ECRG4 gene was found to be 2.6% in seven studies containing 1512 gastric cancer patient (Fig. 2A). Interestingly, ECRG4 gene amplification occurred most frequently in single ring cell carcinoma of the stomach, while ECRG4 gene mutations appeared most frequently in stomach adenocarcinoma, and especially in intestinal and tubular stomach adenocarcinoma (Fig. 2B). Following analyses by Kaplan-Meier plot with a log-rank test, no significant association was found between ECRG4 gene alteration and patient survival regarding overall survival, progression-free survival and disease-specific survival (Fig. 2C, E and F). Nevertheless, ECRG4 gene alteration was observed to be significantly associated with disease-free survival for gastric cancer ( $P=$ 0.0341; Fig. 2D). Further correlation analysis revealed that the top 10 most frequent co-altered genes whose genetic alterations were correlated with the ECRG4 gene, including UXS1, TTN, TP53, NCK2, GPR45, RGPD3, ABCA13, ST6GAL2, DNAH11 and PLGLA (Fig. 2G).

Figure 5.

Kaplan-Meier survival curves with univariate analyses (log-rank) according to the expression statues of ECRG4 in patients with gastric cancer. (A) The survival curve of ECRG4 expression in all gastric cancer patients. (B and C) The survival curves of ECRG4 expression in N0 and N1–N3 patients. (D and E) The survival analysis of ECRG4 expression in M0 patients. (F) The survival curve of ECRG4 in stage I/II and stage III/IV patients.

To corroborate the above findings, we characterized the CNV status of the ECRG4 gene in gastric cancer. The amplification of the ECRG4 gene in gastric cancer was observed by analyzing DNA copy numbers using the Oncomine database (Fig. 3A). In addition, the GSCA database analysis revealed that the CNV distribution rates for homo-amplification, hetero-amplification and hetero-deletion were 2.27%, 17.23% and 6.12%, respectively (Fig. 3B). Spearman correlation analysis revealed a negligible correlation between mRNA expression and CNV status of ECRG4 in gastric cancer ( $P=$ 0.0341; Fig. 3C) with an unobvious false discovery rate (FDR) ( $P=$ 0.58; Fig. 3C). Additionally, no significant association was observed between ECRG4 CNV status and overall survival of gastric cancer patients (Fig. 3D)

3.3 Correlation of ECRG4 expression with clinico- pathological characteristics and patient survival

We further assessed the correlation between ECRG4 expression and various clinicopathological characteristics in gastric cancer, using two tissue microarray chips containing total 180 primary gastric cancer specimens. However, 12 gastric cancer tissue samples in tissue microarray chips were lost during IHC staining. Positive ECRG4 immunostaining was found primarily in the cytoplasm of cancer cells (Fig. 4). In a cohort of 168 informative gastric cancer specimens, high ECRG4 expression was observed in 65.5% (110/168) of primary tumors, whereas low expression was found in 34.5% (58/168) of primary tumors. Low ECRG4 expression was significantly associated with older patient age ( $⩾$ 65 vs. $<$ 65; $P=$ 0.005), higher tumor grade (grade III/IV vs. grade I/II; $P=$ 0.018), more invasive depth (T3/T4 vs.T1/T2; $P=$ 0.020), lymph node metastasis (N1–N3 vs. N0, $P=$ 0.011) and advanced clinical stage (stage III/IV vs. stage I/II, $P=$ 0.002). However, no significant correlation was observed between ECRG4 expression and age or distant metastasis (M classification). Accordingly, these results indicate that decreased ECRG4 expression is associated with malignant properties involving invasion/metastasis and tumor advancement, supporting that dysregulated ECRG4 expression may contribute to gastric cancer pathogenesis and progression.

To evaluate the prognostic impacts of ECRG4 expression on patient outcomes, Kaplan-Meier survival analyses was conducted. As demonstrated in Fig. 5A, high ECRG4 expression was observed to be correlated with a better overall survival rate in patients with gastric cancer ( $P<$ 0.001; Fig. 5A). In particular, we evaluated the correlation between ECRG4 expression and OS individually in patients with and without lymph node metastasis (N1–N3 and N0). This subgroup analysis showed that high ECRG4 expression predicted longer overall survival rates in both groups ( $P=$ 0.002 for N0 and $P<$ 0.001 for N1–N3; Fig. 5B and C). High ECRG4 expression was also associated with a favorable overall survival rate in patients without distant metastasis ( $P<$ 0.001; Fig. 5D). Moreover, subgroup analysis illustrated that high ECRG4 expression represented a favorable overall survival in patients with stage I/II and stage III/IV tumors, separately ( $P=$ 0.013 for stage I/II and $P<$ 0.001 for stage III/IV; Fig. 5E and F). These above findings indicate that dysregulation of ECRG4 is strongly correlates with patient survival of gastric cancer.

Table 2

Univariate and multivariate Cox proportional hazards model for overall survival in gastric cancer patients

Variables	Univariate analysis			Multivariate analysis
	HR	95% CI	$P$ value	HR	95% CI	$P$ value
ECRG4 expression	0.360	0.246–0.527	$<$ 0.001	0.447	0.291–0.688	$<$ 0.001
Age	1.606	1.078–2.392	0.020	1.758	1.148–2.691	0.009
Histological grade	2.098	1.523–2.889	$<$ 0.001	1.613	1.106–2.354	0.013
T classification	1.832	1.352–2.483	$<$ 0.001	1.215	0.785–1.881	0.382
N classification	1.878	1.559–2.261	$<$ 0.001	1.499	1.051–2.138	0.025
M classification	6.481	1.969–21.333	0.002	6.880	1.233–38.409	0.028
Clinical stage	4.635	2.871–7.482	$<$ 0.001	1.341	0.575–3.130	0.497
Sex	0.983	0.648–1.491	0.936

HR, hazard ratio; CI, confidence interval.

To determine the feasibility that ECRG4 expression may act as a prognostic predictor for gastric cancer, we employed univariate and multivariate Cox-regression models for further analysis. On univariate analysis, we found that patient age ( $P=$ 0.020), histological grade ( $P<$ 0.001), invasion depth (T classification; $P<$ 0.001), lymph node metastasis (N classification; $P<$ 0.001), distant metastasis (M classification; $P=$ 0.002), clinical tumor stages ( $P<$ 0.001) and ECRG4 expression ( $P<$ 0.001) were responsible for patient survival (Table 2). However, the multivariate analysis represented that patient age ( $P=$ 0.009), histological grade ( $P=$ 0.013), lymph node metastasis ( $P=$ 0.025), distant metastasis ( $P=$ 0.028) and ECRG4 expression ( $P<$ 0.001) were independent prognostic factors. Thus, we concluded that ECRG4 expression may serve as an independent predictive indicator for the prognosis of gastric cancer patients.

3.4 Knockdown of ECRG4 expression promotes gastric cancer cell growth and invasion

We finally investigated the biological consequences of silencing ECRG4 expression in gastric cancer cells. The efficiency of ECRG4 knockdown using a pre-design validated siRNA was confirmed by western blot analysis (Fig. 6A). We found that SGC-7901 cells with silenced ECRG4 expression represented a significant increase in cell proliferation and colony formation than those transfected with non-specific control siRNAs (Fig. 6B and C). In addition, ECRG4 knockdown impact on cell invasiveness was assessed using a Matrigel assay, in which SGC-7901 cells with ECRG4 knockdown indicated a significant increase in invasiveness than their control cells (Fig. 6D). Western blot analysis revealed that ECRG4 knockdown led to markedl downregulation of representative junction proteins such as E-cadherin and $\alpha$ -Catenin, with concomitant upregulation of mesenchymal markers Vimentin and N-cadherin. Altogether, these data consistently demonstrate that knockdown of ECRG4 increases the proliferation and the migration-invasion of gastric cancer cells.

Figure 6.

Knockdown of ECRG4 promotes gastric cancer cell growth and invasion. (A) Knockdown of ECRG4 in SGC-7901 cells as indicated western blot. Knockdown of ECRG4 promotes SGC-7901 cell proliferation assessed by MTT method (B) and colony formation assay (C). (D) Knockdown of ECRG4 promotes the invasive ability of SGC-7901 cells using a Matrigel assay. All experiments were repeated at least three times. siNS, non-specific control siRNA. ** $P<$ 0.01, *** $P<$ 0.001, as compared to the control group.

4. Discussion

This study builds on our previous work to decipher the relationship between gene expression profiles and cancer outcomes, as well as to identify novel prognostic biomarkers and molecular targets. We were prompted to study ECRG4 because it frequently appeared on the top list of genes whose expression correlates with cancer prognosis. Based on our previous findings that exhibit evidenced functions of ECRG4 as a potential tumor suppressor [16, 17], we evaluated ECRG4 expression profile utilizing multiple approaches, e.g. in silico data-mining and IHC analysis. In this regard, this is the first report to examine ECRG4 as a molecular biomarker by means of combining these two avenues. The above results not only indicated downregulated ECRG4 expression in gastric cancer tissues vs. noncancerous tissues, but also demonstrated its strong correlation with metastasis and tumor advancement. Out data suggest that ECRG4 may serve as an independent prognostic indicator in gastric cancer patients, consistent with our recent report in breast cancer [17].

We utilized multiple avenues to reveal the first comprehensive analysis of ECRG4 in a cohort of 168 specimens from patients with primary resected gastric cancer and relevant online databases. We found that ECRG4 expression, at both mRNA and protein levels, was significantly down-regulated in tumor tissues than paired noncancerous tissues, which was supported by data mining, RT-qPCR and IHC analyses. The immunostaining patterns of ECRG4 observed in gastric cancer are similar to previous reports in esophageal cancer, hepatocellular carcinoma, renal cell cancer, colorectal cancer, and breast cancer [4, 12, 13, 14, 17, 31]. These findings suggest the notion that loss of ECRG4 expression may play an important role in pathogenesis and progression of gastric cancer. Our analyses discovered that decreased ECRG4 expression was associated with several clinicopathologic indicators, including higher histological grade, lymph node metastasis, invasive depth and advanced disease stage. This result was supported by similar reports in colorectal cancer and hepatocellular carcinoma [12, 14]. Yet, we did not observe a significant relationship between ECRG4 expression and distant metastasis, disagreeable with findings studied in renal cell cancer [13]. One possibility may be due to the actual composition of samples in which M1 samples were barely 1.8% (3/168). Further research with a larger sample scale of distant metastasis tumors may be more informative to clarify this issue.

Although a recent study has provided that decreased ECRG4 expression determined by IHC analysis was closely related with lymph node metastasis in gastric cancer, the prognostic implications of ECRG4 expression in gastric cancer remain missing [9]. In this study, decreased ECRG4 expression was observed to be correlated with unfavorable overall survival of patients with gastric cancer. The prognostic value of ECRG4 expression observed in gastric cancer is consistent with its essential role as an independent prognostic predictor for patient survival reported in esophageal cancer, breast cancer, renal cell cancer and nasopharyngeal carcinoma [4, 6, 13, 17, 32]. Through IHC analysis and in vitro cell experiments, we identified that ECRG4 expression may serve as a prognostic biomarker that also holds great therapeutic promise for gastric cancer.

Because gastric cancer patients usually present with locally advanced disease, biomarkers other than lymph node metastasis are required to forecast patient outcomes [33]. In this study, high ECRG4 expression was found to be associated with better overall survival rates in patients without lymph node metastasis or distant metastasis, and in stage I/II patients. These observations indicate that ECRG4 expression may be valuable in predicting the prognosis of patients with early-stage disease.

The alterations of DNA could contribute to the cancer initiation, progress, diagnosis, prognosis and therapy [20]. Through online database analysis, we found a negligible correlation between mRNA expression and CNV status of the ECRG4 gene in gastric cancer. As a recognized integrator and an indispensable node of major growth factor signaling pathways, ECRG4 is modulated by multiple layers of mechanisms such as transcriptional repression, gene promoter hypermethylation, genetic alteration and RNA splicing. The intrinsic differences in each type of tumors might be another possible explanation. Further, an unobvious FDR was not statistically supported the association and more strict setting may be needed to avoid excessive false positive.

In conclusion, our results reveal that ECRG4 expression is downregulated in gastric cancer and this alteration was strongly correlated with metastasis and tumor advancement. ECRG4 expression is an independent prognostic factor that holds therapeutic promise for gastric cancer. Clearly, further examination of ECRG4-modulated signaling pathways may promote our ability to assess the prognosis in patient with gastric cancer.

Availability of data and materials

The dataset used and/or analyzed in the present study is available from the corresponding authors on reasonable request.

Author contributions

Conception: Yanjie You.

Interpretation or analysis of data: Yanjie You, Shengjuan Hu.

Preparation of the manuscript: Yanjie You, Shengjuan Hu.

Revision for important intellectual content: Yanjie You, Shengjuan Hu.

Supervision: Yanjie You, Shengjuan Hu.