Connection Between EIF3S3 Polymorphisms and Hepatocellular Carcinoma Prognosis

Background: EIF3S3 that encodes a p40 subunit of eukaryotic initiation factor 3 (eIF3), has been reported overexpression in several human malignancies. The expression of EIF3S3 in hepatocellular carcinoma (HCC) and its clinical role in the disease progression is still unclear. Here we aimed to study the effect of EIF3S3 on HCC prognosis. Methods: Quantitative real-time polymerase chain-reaction (qRT-PCR) was performed to assess the mRNA levels of EIF3S3 in 120 HCC tissues samples and 60 adjacent noncancerous specimens. Kaplan-Meier and Cox regression analysis was used to study the prognostic value of EIF3S3. Results: The expression of EIF3S3 was signicantly increased in HCC compared to the expression found in adjacent tissues (P<0.001). Furthermore, high EIF3S3 expression was positively correlated with vascular invasion, tumor size, cirrhosis, AFP levels and TNM stage (all P<0.05). Kaplan-Meier survival analysis showed that patients with high EIF3S3 expression had worse overall survival (OS) (P<0.001) and disease-free survival (DFS) (P<0.001) than those with low EIF3S3 expression. In addition, multivariable analysis revealed that EIF3S3 could be an independent prognostic factor of OS (P<0.001) and DFS (P<0.001) for patients with HCC. Conclusion: The results suggested that EIF3S3 might serve as a promising prognostic factor for OS and DFS of HCC patients.


Background
Hepatocellular carcinoma (HCC) is the fth most common human cancer and the third most common cause of cancer-related death worldwide [1,2], due to its high invasive and metastatic features. As the results of the prevalence of hepatitis B and C viral infections, the morbidity and mortality of HCC is more likely discovered in China [3]. Actually, most HCC patients have lost the best treatment opportunity when they were diagnosed for the rst time, because of the absence of the speci c symptoms in the early stage. Despite improvements in the treatment of HCC such as surgical techniques, radiotherapy, chemotherapy, drug intervention and molecular targeted therapy, the prognosis of HCC still remains poor with a 5-year survival rate of 11% [4,5]. Moreover, the recurrence and metastasis are the main factors for the frustrating outcome of HCC [6]. Therefore, identi cation of new potential biomarkers improving the outcome of patients with HCC is urgent.
EIF3 is the largest of eukaryotic initiation factors (eIF) with a molecular mass of 800 kDa, which is composed of 13 subunits (EIF3a to eIF3m) and plays an essential role in cellular and viral initiation of translation [7][8][9]. EIF3S3 encodes for a p40 subunit of eIF3, which is the largest initiation factor of the protein synthesis. EIF3S3 was identi ed by Kittler et al., who screened over 5000 genes for the necessity on Hela cell division and found 37 genes that were required for cell division, and two of them were eIF3 subunits, EIF3S3 and EIF3S10 [10][11][12]. Ampli cation and over expression of. EIF3S3 have been observed in prostate and breast cancers [13,14], and its ampli cation has been associated with advanced stages of prostate cancer [15]. These ndings suggested that EIF3S3 has been involved in cancer development. However, to date, there have been few reports on the role of EIF3S3 in patients with HCC, and the clinical signi cance of EIF3S3 in the progression of HCC is also unclear.
Therefore, the the present study was designed to investigate whether the expression level of EIF3S3 could be used as prognostic biomarker for HCC patients.

Patients and samples
A total of 120 fresh tumor tissue samples and 60 randomly selected adjacent noncancerous tissue samples were obtained from HCC patients who underwent curative resection at Chinese PLA General Hospital-Sixth Medical Center. HCC was con rmed by pathological examination and none of these patients had received any radiotherapy or chemotherapy prior to surgical treatment. The stages of tumor were classi ed according to the TNM staging system of UICC. All clinicopathological characteristics of 120 patients were collected and summarized in Table 1. Total RNA from snap-frozen samples were extracted using TRIzol (Invitrogen, USA) reagent according to the manufacturer's protocol. Then, the concentration of isolated RNA was measured by the NanoDrop ND-1000 spectrophometer (NanoDrop Technologies, USA). Complementary DNA (cDNA) was synthesized from total RNA using Superscript Reverse Transcriptase (Invitrogen) following the manufacturer's directions. Afterwards, the cDNA was subjected to real-time PCR that was carried out using an Applied Biosystems 7500 real-time PCR system. For standardization of RNA quality control, the GAPDH was used as an internal control. The primer sequences were as follows: EIF3S3 sense 5′-AAGGAGTTCACTGCCCAAAA-3′ and antisense 5′-AAGAGTTGCCCTGGTGTGAC-3′; GAPDH sense 5′-TGACTTCAACAGCGACACCCA-3′ and 5′-CACCCTGTTGCTGTAGCCAAA-3′. Each sample was examined in triplicate.

Statistical analysis
The statistical analyses were performed using SPSS statistics software (Version 18.0, Chicago) and Prism 5.0 software (GraphPad, La Jolla, CA, USA). The correlation of EIF3S3 expression with clinicopathological features was assessed by Chi-square test, and quantitative data were analyzed by Student's t-test. Survival rate was constructed by the Kaplan Meier with log rank test, and the Cox regression model was used for univariate and multivariate survival analysis. All data in this study are presented as mean±standard deviation (SD). P<0.05 was considered statistically signi cant.

Results
The expression of EIF3S3 is upregulated in HCC To investigate the role of EIF3S3 in HCC, we explored the EIF3S3 mRNA levels in 120 HCC and 60 adjacent non-cancerous tissues samples using qRT-PCR. As shown in Figure 1, the mRNA expression of EIF3S3 was signi cantly increased in HCC tissues in comparison to adjacent nontumor tissues (mean±SD: 0.61 ± 0.56 vs 0.22 ± 0.37, P<0.05). The results suggested that EIF3S3 might serve as an oncogene in HCC.

Correlation of EIF3S3 expression with clinicopathological features of HCC patients
To further verify this observation that EIF3S3 might play a role in the progression of HCC, the patients were divided into two groups according to EIF3S3 expression (high and low) as de ned by the mean. Associations between EIF3S3 expression and the clinicopathological parameters in HCC were shown in Table 1. The expression of EIF3S3 was found to correlate closely with vascular invasion (P=0.007), tumor size (P=0.033), cirrhosis (P=0.007), AFP levels (P=0.012) and TNM stage (P=0.001). However, no statistical correlations were discovered between EIF3S3 expression and age (P=0.841), sex (P=0.869), and differentiation (P=0.570).

Relationship between EIF3S3 expression and survival
Next, Kaplan-Meier curve analysis was applied to evaluate the relationship between EIF3S3 expression and survival of patients with HCC. The log-rank test revealed that patients with high EIF3S3 expression had a signi cantly shower overall survival (OS) when compared with those of low EIF3S3 expression (P<0.001, Figure 2A). Likewise, the disease free survival (DFS) time of HCC patients with high EIF3S3 expression was markedly shorter than those with low expression of EIF3S3 (P<0.001; Figure 2B).
Furthermore, univariate analysis of prognostic factors revealed that the following variables including vascular invasion, cirrhosis, AFP level, TNM stage and EIF3S3 expression correlated signi cantly with OS (all P<0.05; Table 2). Multivariate analysis indicated that EIF3S3 expression was an independent prognostic factor for OS (HR=4.810, 95% CI 2.233-10.361; P<0.001), along with tumor size, cirrhosis, AFP level and differentiation (all P<0.05). In addition, univariate analysis of prognostic factors showed that EIF3S3 expression, cirrhosis, AFP level and TNM stage had signi cant prognostic in uences on DFS (all P<0.05; Table 3). Moreover, multivariate survival analysis revealed that EIF3S3 expression and cirrhosis were independent factors that affected the DFS (both P<0.05).

Discussion
Because of the high incidence of recurrence and metastasis after hepatic resection, the long-term survival of HCC patients still remains poor. Currently, despite the BCLC staging system and CLIP score of patients with HCC is critical for the guidance of therapy selection, HCC patients with the same stage are diverse [16,17]. Early diagnosis of HCC is di cult due to its high-grade malignancy and tumor invasion [4]. Nowadays, AFP has been used as a diagnostic and prognostic marker for HCC [18], moreover, gene therapy and molecular targeted therapy are used in clinical research and various genes for targeted therapy of HCC are effective than the traditional forms of treatments [19]. Therefore, it is important to nd effective molecular biomarkers with high sensitivity and speci city for HCC treatment and prognosis.
EIF3 is the largest initiation factor of the protein synthesis. Originally, eIF3 was identi ed as a factor that binds to the 40S ribosomal subunit and thereby prevents the association of the 40 and 60S subunits with one another. eIF3 complex has been the essential for initiation of protein synthesis for both cells and virus [20]. EIF3S3 located at 8q23, encodes the p40 subunit of eIF3. Previous studies have indicated that EIF3S3 gene was abundantly expressed in several cancers. For example, Quti Saramaki et al. showed that high level ampli cation of EIF3S3 was found in prostate tumors and it was associated with poor cancerspeci c survival of the carcinoma [21]. Moreover, the effect of EIF3S3 on cell proliferation was also reported. In the study of Savinainen et al., they found that EIF3S3 overexpression regulated cell growth and viability and the overexpression of the gene may provide growth advantage to the cancer cells [22]. What's more, a study carried out by Okamoto et al. revealed that EIF3S3 was ampli ed in HCCs, and the expression of EIF3S3 was signi cantly associated with large tumor size,and hepatitis B virus infection. They concluded that EIF3S3 encodes the p40 subunit of the eIF3 and may be involved in the progression of HCC [23]. However, the clinical role of EIF3S3 in human HCC prognosis was unclear.
In the present study, the expression of EIF3S3 was signi cantly upregulated in HCC tissues compared with in adjacent noncancerous liver tissues. Moreover, correlation analysis between EIF3S3 expression and clinical parameters of HCC patients revealed that high expression of EIF3S3 were signi cantly related to vascular invasion, tumor size, cirrhosis, AFP levels and TNM stage, indicating that EIF3S3 expression may be responsible for tumor progression in HCC. These results were consistent with the previous study [23]. Moreover, the data suggested that EIF3S3 presented an unfavorable outcome in HCC, and might be a potential prognostic marker. Our data revealed that HCC patients with high EIF3S3 expression had a signi cantly shorter OS and DFS than those with low expression. In addition, multivariate analysis identi ed EIF3S3 as an independent prognostic factor for HCC patients. However, the molecular mechanisms of EIF3S3 in HCC that high expression of EIF3S3 results in poor outcome still remain unclear.
Until now, little is known about the actual role of EIF3S3 in carcinogenesis as well as the mechanisms EIF3S3 on the regulation of cancer. Dysregulated translation of the gene may be involved because several other initiation factors such as eIF4E, eIF4G, and eIF5A2 are also ampli ed and over expressed in various types of cancers [24][25][26]. Therefore, the speci c mechanism by which EIF3S3 overexpression could improve the growth of HCC cells still remains to be elucidated.

Conclusion
In summary, the present study suggest that increased expression of EIF3S3 could be involved in the progression of HCC and might be a novel biomarker of poor prognosis for patients with HCC. 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 materialsAll data generated or analysed during this study are included in this published article.
Competing interestsThe authors declare that they have no competing interests.
Funding Not applicable.
Authors' contributions C.Y. design of the work; P.D. the acquisition, analysis, W.L. interpretation of data; P.D. the creation of new software used in the work; C.Y. have drafted the work or substantively revised it. All authors read and approved the nal manuscript.  Figure 1 Relative EIF3S3 mRNA levels. The expression of EIF3S3 in HCC tissues was signi cantly higher than in adjacent normal specimens (P<0.001).