The relationship between RECK Gene Polymorphisms and the prognosis of cholangiocarcinoma

The study was designed to examine the reversion inducing cysteine rich protein with Kazal motifs (RECK) levels in patients with cholangiocarcinoma (CCA) and assess its role in CCA prognosis. Quantitative real-time PCR (qRT-PCR) was used to determine the expression of RECK mRNA in 127 pairs of CCA samples and controls. Chi-square test was conducted to analyze the effects of clinical features on RECK expression. Kaplan-Meier curves were plotted to determine the overall survival rate of CCA patients with different RECK expression. The prognostic biomarkers for CCA patients were identied using the Cox regression analysis.


Abstract Background
The study was designed to examine the reversion inducing cysteine rich protein with Kazal motifs (RECK) levels in patients with cholangiocarcinoma (CCA) and assess its role in CCA prognosis.

Methods
Quantitative real-time PCR (qRT-PCR) was used to determine the expression of RECK mRNA in 127 pairs of CCA samples and controls. Chi-square test was conducted to analyze the effects of clinical features on RECK expression. Kaplan-Meier curves were plotted to determine the overall survival rate of CCA patients with different RECK expression. The prognostic biomarkers for CCA patients were identi ed using the Cox regression analysis.

Results
Signi cantly down-regulated expression of RECK mRNA was determined in CCA tissues compared to noncancerous controls (P < 0.05). Chi-square test suggested reduced RECK expression was related with invasion depth (P = 0.026), differentiation (P = 0.025), lymphatic metastasis (P = 0.010) and TNM stage (P = 0.015). However, age, sex, tumor size and family history had no signi cant links with RECK expression (all, P > 0.05). The survival curves showed that patients with low RECK expression had a shorter overall survival rate than those with high RECK expression. Both the univariate analysis (P = 0.000, HR = 5.290, 95%CI = 3.195-8.758) and multivariate analysis (P = 0.000, HR = 5.376, 95%CI = 2.231-8.946) demonstrated that RECK was an independent biomarker for predicting the outcomes of CCA patients.

Conclusions
Taken together, the expression of RECK was down-regulated in CCA and it might be an e cient biomarker for CCA patients.

Background
Cholangiocarcinoma (CCA) is one of the most frequent cancers in the biliary tract with high malignancy.
It constitutes approximately 2% of all reported cancers anfd accounts for about 3% of all gastrointestinal malignancies [1,2]. According to the tumor location, CCA can be anatomically divided into two large subtypes: extrahepatic cholangiocarcinoma (EHCC) and intrahepatic cholangiocarcinoma (IHCC), and the former is further classi ed as perihilar CCA and distal CCA [3,4] The mortality and morbidity of CCA have steadily increased in recent years all over the world [5]. At present, there are several treatment approaches for CCA patients, such as endoscopic stent placement, radiation therapy, surgical resection, chemotherapy and photodynamic therapy [6]. Surgery is the only curative method for patients with early stage, however, due to late diagnosis and metastasis, the resection rate of CCA patients is always low and the recurrent rate is sometimes high [7,8]. Therefore, the prognosis of CCA patients is extremely poor, which is often less than 40% in general [9,10]. Thus, it is urgent to nd novel biomarkers to predict the outcomes of patients with this disease and improve the treatment results.
Tumor invasion and metastasis of malignancies, which are signaled by the degradation of extracellular matrix (ECM), are the important factors for cancer-related deaths [11,12]. Matrix metalloproteinases (MMPs) and their inhibitors have been reported to play important roles in the ECM degradation [13,14]. RECK (reversion inducing cysteine rich protein with Kazal motifs) is a newly found tumor suppressor gene, which could inhibit the expression and activity of MMPs as an inhibitor [15,16]. The RECK protein is a 110-kDa glycosyphosphatidylinositol (GPI)-anchored protein in mammal cells, which is involved in a variety of biological processes, such as tissue remodeling, remodeling enzyme control and vascular grow during the development [17,18]. Reduced expression of RECK has been found in various cancers, including breast cancer, bladder carcinoma and prostate cancer [19][20][21].
In the present study, we attempted to measure the expression of RECK in CCA patients and then to identify its prognostic role in CCA.

Patients and specimens
A total of 127 CCA patients who received surgical resection in the PLA Rocket Force Characteristic Medical Center were enrolled in this retrospective study, including 67 male cases and 60 female cases. None patients were subjected to radio-or chemo-therapy before surgery. The resected tissue samples, including 127 CCA specimens and 127 adjacent noncancerous controls were collected immediately after surgery and stored at -80℃ until use. A 5-year follow-up was taken and the clinical information of patients was updated every 3 months through telephone calls. This study was approved by the Ethics Committee of the PLA Rocket Force Characteristic Medical Center. The informed consents were provided by the patients before surgery.

Quantitative real-time PCR (qRT-PCR)
Total RNA was isolated from CCA and control tissues with the RNeasy Mini Kit (Qiagen, Germantown) according to the manufacture's instructions. The SuperScript ® III Reverse Transcriptase Kit (Life Technologies, Carlsbad, CA, USA) was adopted for cDNA synthesis. Quantitative real time PCR was performed with QuantiTect SYBR Green RT-PCR Kit (Qiagen). The gene GAPDH was used as an endogenous control. The relative expression levels of RECK were calculated by the 2 -ΔΔCt method. All experiments were performed in triple.

Statistical analysis
All data were statistically analyzed on SPSS 18.0 and Sigmaplot 12.5 softwares. The relationships of RECK expression with various clinical characteristics were detected by Chi-square test. The overall survival rate was evaluated by the Kaplan-Meier analysis and log-rank test . Cox regression analysis was applied to identify the prognostic factors for CCA prognosis. All data were presented as mean±SD. The results were signi cant with P values of less than 0.05.

Results
Down-regulation of RECK mRNA in CCA tissues The expression of RECK mRNA was measured using qRT-PCR. As shown in Figure 1, the relative expression level of RECK mRNA in CCA tissues was 1.705±0.476 (mean±SD), while that in adjacent normal controls was 3.188±0.431. Thus the expression of RECK in CCA was signi cantly lower than that in controls (P<0.05).

Association between RECK expression and clinical characteristics of CCA patients
To explore the relationship between RECK expression and clinical characteristics of CCA patients, Chisquare test was conducted. Signi cant correlations were observed between RECK expression and invasion depth (P=0.026, differentiation (P=0.025), lymphatic metastasis (P=0.010) and TNM stage (P=0.015). However, no signi cant relation was observed between RECK levels and age, sex, tumor size or family history (all, P>0.05) ( Table 1).

Low-expression of RECK was associated with poor prognosis in CCA patients
Among all the 127 CCA patients, 75 (59.05%) cases died of this disease during the 5-year follow-up, with the 5-year overall survival rate of 40.95%. As shown in Figure 2, the survival rate was 18.75% in low-RECK group, and 63.49% in high-RECK group (P<0.001). Univariate analysis demonstrated that invasion depth (P=0.021), differentiation (P=0.042), TNM stage (P=0.004) and RECK level (P=0.000) were related with prognosis of CCA patients ( Discussion CCA refers to those malignancies that originate from the epithelial cells of the bile duct system, and is only inferior to liver cancer. CCA is characterized by jaundice, stomachache, eallbladder enlargement, fever, and ascites. There are no speci c clinical manifestations for CCA at early stages, resulting in only a small part of patients could receive the curative resection. The 5-year survival rate of CCA patients is low and the the prognosis is dismal. Thus, new predictor biomarkers are in urgent need to improve the prognosis of CCA patients. So far, some prognostic markers have been identi ed for CCA patients. For instance, Wu et al. claimed that S100P was a candidate for prognosis, diagnosis and treatment of CCA [22]. In the study of Gu et al., CA9 was identi ed as a promising prognostic biomarker for CCA [23]. In the present study, we were engaged in nding novel and effective molecule to better improve the prognosis of CCA patients. RECK gene is discovered by Takahashi et al. in v-ki-ras-transfected NIH3T3 cells in 1998, and locates on human chromosome 9p12-p13 [24]. RECK is about 87 kb in length, including 21 exons and 20 introns. RECK protein consists of 971 amino acids with two hydrophobic regions in the N-terminal and C-terminal, respectively. RECK is a novel metastasis suppressor gene, which plays an important role in maintaining the balance of cell proliferation, differentiation and apoptosis. Moreover, RECK might be implicated in regulating cell growth and tumor invasion, migration and angiogenesis [25]. In addition, the prognostic role of RECK has been identi ed in several cancers, such as colorectal cancer [26]. So here we attempted to explore the prognostic role of RECK in CCA.
In this retrospective study, the expression levels of RECK mRNA in CCA tissues and adjacent normal controls were measured using qRT-PCR method. The results showed signi cantly reduced expression of RECK mRNA in CCA samples compared with noncancerous controls, which was consistent with the ndings in previous studies. The Chi-square test suggested that RECK down-regulation was affected by invasion depth, lymphatic metastasis, differentiation and TNM stage. With these ndings, we hypothesized that RECK might be related with the CCA development and prognosis. Therefore, the Kaplan-Meier and Cox regression analyses were carried out. It could be concluded from the survival curve that patients with low RECK expression were more likely to die than those with high RECK expression. The nal Cox regression analysis revealed that RECK was an independent biomarker for the prognosis of CCA patients with both univariate and multivariate analyses.
Though we have identi ed the predictive role of RECK in CCA, its mechanisms on CCA development and progression have not been well understood. MMPs are composed of over 20 zinc dependent enzymes, which could interact with ECM in tumor cells to destroy connective tissues and induce tumor invasion and migration [27]. Evidence has demonstrated that RECK is an inhibitor of MT1-MMP, MMP-2 and MMP-9 and therefore suppresses the development and progression of tumors [19]. In addition, in the study of Huang et al., RECK was reported to be a binding site of the miR-21, which was associated with cancer cell processes as a tumor suppressor gene [28]. Moreover, there were several investigations indicating that miR-21 was related with CCA occurrence and progression [29,30]. Therefore, we conjectured that RECK might function on CCA through interacting with miR-21. In a word, the precise mechanism of RECK on CCA development is complex and more further studies are needed in future.

Conclusions
In summary, RECK expression was signi cantly decreased in CCA tissues compared to adjacent normal controls. Down-regulation of RECK was in uenced by invasion depth, differentiation, lymphatic metastasis and TNM stage. RECK could be used as a candidate biomarker for CCA prognosis. The subjects had been informed the objective. Certainly, written consents were signed by every subject in this study.

Consent for publication
We obtaining permission from participants to publish their data.

Availability of data and materials
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.  Figure 1 The expression of RECK mRNA was measured by qRT-PCR and normalized to GAPDH. The result showed that RECK mRNA was negatively expressed in CCA tissues compared with adjacent normal controls (P<0.05).

Figure 2
Kaplan-Meier survival curves to describe the overall outcome of CCA patients. The results showed that patients with high RECK expression had a relatively higher overall survival rate than those with low RECK expression levels (P<0.001).