Down-regulation of long non-coding RNA GAS5-AS1 and its prognostic and diagnostic significance in hepatocellular carcinoma

Abstract

BACKGROUND:

Hepatocellular carcinoma (HCC) is the most common solid tumor in global range, with high degree of malignancy and poor prognosis. But the relationship between the expression of GAS5-AS1 and HCC is not documented. This study aimed to profile GAS5-AS1 expression signature and then to explore its clinical significance in HCC.

METHODS:

Quantitative real-time PCR (RT-qPCR) was performed to detect the expression of GAS5-AS1 in 83 pairs of HCC surgical tissues and adjacent normal liver tissues. We also performed RT-qPCR on plasma samples of 156 patients and 58 healthy controls.

RESULTS:

We found that GAS5-AS1 was down-regulated in HCC tissues ( $P<$ 0.01). Correlation analysis showed that the expression of GAS5-AS1 was notably associated with differentiation (High/Moderate vs Low, $P=$ 0.031), tumor-node-metastasis (TNM) stage (I $\sim$ II vs III $\sim$ IV, $P=$ 0.020) and glucose levels ( $<$ 6.2 vs $\geqq$ 6.2, $P=$ 0.047) in HCC patients. The overall survival analysis showed that patients with lower GAS5-AS1 expression had a relatively poor prognosis. Univariate and multivariate analysis elaborated that GAS5-AS1 was an independent prognostic factor for HCC patients. The area under the ROC (AUC ${}_{\text{ROC}}$ ) demonstrated that GAS5-AS1 presented a high accuracy (AUC $=$ 0.824, 95% CI: 0.741–0.906) for distinguishing HCC from the cirrhosis. When differentiating HCC cases with AFP $<$ 200 ng/ml from the cirrhosis and hepatitis B whose AFP levels were also below 200 ng/ml, GAS5-AS1 had the high sensitivity (89.5%, 89.5%, respectively).

CONCLUSIONS:

GAS5-AS1 could be considered as a potential prognostic and diagnostic marker in HCC. However, the potential clinical application value of GAS5-AS1 still needs to be further illustrated.

Keywords

Hepatocellular carcinoma long non-coding RNA GAS5-AS1 prognosis diagnosis

1. Introduction

Hepatocellular carcinoma (HCC) is one of the most common and aggressive malignancies, and ranks as the third leading cause of cancer-related death worldwide [1, 2]. As one of the most malignant cancers, HCC occupies 85%–90% of primary liver cancer [3]. It causes nearly 695,900 deaths per year, half of which occur in China, as a result of the high chronic hepatitis B virus (HBV) infection incidence [4]. Surgical resection and liver transplantation are the first-line treatment for HCC [5]. Even after surgical resection, the 5-year survival rate of HCC patients remains poor [6]. Invasion, metastasis and recurrence are the primary factors that affect clinical treatment and prognosis [7, 8]. The serum AFP level has long been used as a prognostic marker for HCC, but even in the advanced stage, 15%–30% of their AFP levels remain normal [9]. Therefore, a good prognostic and diagnostic marker for patients with HCC is urgently needed.

Recently, advanced techniques discovered that more than 90% of the mammalian genome can be transcribed into non-coding RNAs [10]. According to the transcript size, non-coding RNAs are divided into small non-coding RNAs and long non-coding RNAs (lncRNAs) [11]. In 1992, Brockdorff et al. [12] indicated that lncRNAs are molecules with a length of longer than 200 bps that cannot code protein products. Emerging evidences have indicated that lncRNAs play critical roles in biological processes, such as cell proliferation, differentiation, apoptosis, and cell cycle progression [13]. Meanwhile, researchers find that lncRNAs can participate in the process of the occurrence and development of cancers [14]. For example, HOX transcript antisense RNA (HOTAIR) played an oncogenic role in HCC development, and could be a candidate biomarker for predicting tumor recurrence in HCC patients who have undergone liver transplant therapy [15]. In 2016, Jia et al. [16] elaborated that lincRNA-p21, down-regulated in HCC, contributed to the inhibition of tumor metastasis through Notch signaling when overexpressed in vivo. In HCC, prostate cancer-associated transcript 1 (PCAT-1) played an oncogenic effect and silencing PCAT-1 might be a potential novel therapeutic strategy for HCC [17]. Accumulating studies have been explored potential diagnostic and prognostic roles of lncRNAs for HCC.

The antisense RNA of growth arrest-specific transcript 5 (GAS5-AS1) is a novel lncRNA transcript which maps on chromosome 1q25.1. In 2016, Wu et al. [18] suggested GAS5-AS1 could be as a tumor suppressor in non-small cell lung cancer (NSCLC), correlated with tumor size, tumor-node-metastasis (TNM) stage, and lymph node metastasis. Overexpression of GAS5-AS1 in NSCLC cells could reduce those molecules that played critical roles for epithelial-mesenchymal transition (EMT). GAS5, sense RNA of GAS5-AS1, was found significantly down-regulated in HCC tissues, and overexpression of GAS5 could decrease the hepatoma cell proliferation and invasion [19]. However, the prognostic or diagnostic feature of GAS5-AS1 in HCC need to be elucidated.

In the present study, we investigated the levels of GAS5-AS1 in HCC patients, analyzed the association between expression level of GAS5-AS1 and clinical characteristics, and then evaluated the prognostic and diagnostic value of GAS5-AS1. In conclusion, our aim was to explore potential biomarker for prognosis and diagnosis in HCC patients.

2. Materials and methods

2.1 Specimens

Surgical specimens were obtained from 83 HCC patients (76 men and 7 women, mean age 55 $\pm$ 11) who had hepatectomy from 2012 to 2016 at Zhongnan Hospital of Wuhan University. None of the patients had received preoperative chemotherapy or radiotherapy treatment before hepatectomy. The samples were stored at $-$ 80 ${}^{\circ}$ C until use. All of the patients were selected based on pathology reports. Follow-up data ranged from 2 to 48 months. All participants were from China.

Whole blood samples of 156 patients were obtained from Zhongnan Hospital of Wuhan University during 2016, and separated them into three groups: 63 patients with HCC (54 males and 9 females, mean age 56 $\pm$ 9), 46 patients with hepatitis B (34 males and 12 females, mean age 54 $\pm$ 12), and 47 cirrhosis patients (39 males and 8 females, mean age 55 $\pm$ 10), which were collected in EDTA tubes. Plasma was separated from whole blood sample by two-step centrifugation (2000 g for 5 min at 4 ${}^{\circ}$ C and then 12000 g for 5 min at 4 ${}^{\circ}$ C) to spin down the blood cells. The plasma was then stored at $-$ 80 ${}^{\circ}$ C until use. Patients who underwent preoperative chemotherapy or radiotherapy before were excluded. Meantime, from the Physical Examination Center, we collected 58 healthy blood samples (49 males and 9 females, mean age 55 $\pm$ 10), which all were without hepatitis, hepatic diseases or abnormal liver biochemical outcome.

2.2 Ethical approval

All subjects gave their informed consent for inclusion before they participated in the study. The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Ethics Committee of Zhongnan Hospital of Wuhan University (Ethical Approval No. 2013059). All the experiments were conducted according to the institutional ethical guidelines.

Figure 1.

GAS5-AS1 expression in HCC tissues. A. GAS5-AS1 levels in tumor tissues were significantly lower than in non-tumor tissues. The expression levels were calculated using the comparative cycle threshold (Ct) method ( $\Delta$ Ct) B. Waterfall plot shows the fold change of GAS5-AS1 expression. $\Delta\Delta$ Ct $=$ (Ct GAS5-AS1–Ct GAPDH) of HCC – (Ct GAS5-AS1–Ct GAPDH) of NT]. (NT: paired non-tumor tissues). ** $P<$ 0.01.

2.3 RNA extraction and reverse transcription

Total RNA from tissues were extracted using Trizol reagent (Invitrogen, CA, USA), and Total RNA Seperate Extraction Kit (Bioteke, Beijing, China) was used to extract total RNA from plasma according to manufacturer’s instruction. The concentration and purity of RNA were quantified by NanoDrop ND2000 (Thermo Fisher Scientific, CA, USA). Then RNA was reverse transcribed to cDNA by using PrimeScriptTM RT reagent Kit with gDNA Eraser (Takara, Japan), following: 42 ${}^{\circ}$ C for 2 min, and then 37 ${}^{\circ}$ C for 15 min, 85 ${}^{\circ}$ C for 5 sec.

2.4 Real-time PCR analysis

According to manufacturer’s instructions, the expressions of GAS5-AS1 were performed on the Bio-Rad CFX96 (Bio-Rad Laboratories, Inc., Hercules, CA, USA) using SYBR Premix Ex Taq Kit (TaKaRa, Japan). The reactions started at 95 ${}^{\circ}$ C for 5 min, followed by 40 cycles of 95 ${}^{\circ}$ C for 30 s, 64 ${}^{\circ}$ C for 30 s and 72 ${}^{\circ}$ C for 30 s. Ct values of the target GAS5-AS1 were normalized in relation to GAPDH. The primers were designed based on (NR_037605.1) GAS5-AS1 gene sequence obtained from the gene database of NCBI. The primer sequences were as follows: GAS5-AS1 (forward: 5’-TCCCAGCCTCAGACTCAACA-3’ and reverse: 5’-GTTTCATAGGCCCCTGTGCT-3’); GAPDH (forward: 5’-AGAAGGCTGGGGCTCATT TG-3’ and reverse: 5’-GCAGGAGGCATTGCTGAT GAT-3’). The comparative Ct method formula 2 ${}^{-\Delta\text{Ct}}$ was used to calculate the relative gene expression. All experiments were carried out in duplicate for each data point.

2.5 Statistical analysis

All statistical analysis was carried out using the SPSS version 17.0 (SPSS, Inc. Chicago, IL, USA) and GraphPad Prism 5.0 (GraphPad Software, La Jolla, CA, USA). All $P$ -values shown were two-sided, $P<$ 0.05 was considered to be statistically significant. Statistical significance was assigned at $P<$ 0.05 (*) or $P<$ 0.01 (**). The normally distributed numeric variables were evaluated by Student’s t-test. Correlations were analyzed using the Spearman correlation. Survival curves were plotted using the Kaplan-Meier method. Cox proportional hazard regression was used for univariate and multivariate analysis to explore potential prognosis factors. To validate the GAS5-AS1 expression among subgroups in plasma, Oneway ANOVA was used. The categorical variables were analyzed using Chi-square test. Finally, the area under the corresponding ROC curve analysis was performed to estimate the diagnostic values.

3. Results

3.1 GAS5-AS1 level was significantly lower in HCC tissues

We detected the expression levels of GAS5-AS1 in 83 pairs of HCC surgical tissues and adjacent normal liver tissues by RT-qPCR. Expression of GAS5-AS1 relative to GAPDH in tumor tissues was obviously lower than that in non-tumor tissues ( $P<$ 0.01; Fig. 1A). Compared with adjacent normal liver tissues, 67.47% (56/83) cases showed down-expression of GAS5-AS1 in HCC tissues, 42.17% (35/83) of cases were reduced at least by two-fold (Fig. 1B). The results suggested that GAS5-AS1 may play tumor suppressor’s role in HCC.

Table 1
Association of GAS5-AS1 expression with clinical parameters in HCC

Characteristics	GAS5-AS1			$P$
	All cases	Low expression $n=$ 41	High expression $n=$ 42
Gender				0.223
Male	76	36	40
Female	7	5	2
Age				0.746
$<$ 55	44	21	23
$\geqq$ 55	39	20	19
Smoking				0.435
Negative	38	17	21
Positive	45	24	21
Alcoholism				0.43
Negative	47	25	22
Positive	36	16	20
Differentiation*				0.031
High/Moderate	70	31	39
Low	13	10	3
Tumor size (cm)				0.573
$<$ 5	22	12	10
$\geqq$ 5	61	29	32
Tumor nodes				0.272
Single	74	35	39
Multi	9	6	3
TNM stage*				0.02
I $\sim$ II	49	19	30
III $\sim$ IV	34	22	12
HBV DNA (IU/mL)				0.45
$<$ 500	12	5	7
$\geqq$ 500	35	19	16
Cirrhosis				0.099
Negative	43	25	18
Positive	40	16	24
AFP (ng/mL)				0.996
$<$ 200	41	20	21
$\geqq$ 200	39	19	20
CEA ( $\mu$ g/L)				0.822
$<$ 7.2	50	29	21
$\geqq$ 7.2	2	1	1
ALT (u/L)				0.734
$<$ 46	45	23	22
$\geqq$ 46	38	18	20
AST (u/L)				0.582
$<$ 46	42	22	20
$\geqq$ 46	41	19	22
GGT (u/L)				0.591
$<$ 55	34	18	16
$\geqq$ 55	49	23	26
5’-NT (u/L)				0.208
$<$ 10	58	26	32
$\geqq$ 10	2	0	2
GLU (mmol/L)*				0.047
$<$ 6.2	69	32	37
$\geqq$ 6.2	10	8	2

Abbreviation: * $p<$ 0.05; HCC, hepatocellular carcinoma; TNM, tumor-node-metastasis; HBV, hepatitis B virus; AFP, $\alpha$ -fetoprotein; CEA, carcinoembryonic antigen; ALT, alanine aminotransferase; AST, aspartate aminotransferase; GGT, $\gamma$ -glutamyl transferase; 5’-NT, 5’-nucleotidase; GLU, glucose.

Table 2

Univariate and multivariate analysis of overall survival in HCC patients

Factors	Univariate analysis		Multivariate analysis
	HR (95% CI)	$P$	HR (95% CI)	$P$
Gender	1.216 (0.358–4.134)	0.754
Age	1.003 (0.961–1.047)	0.897
Smoking	2.028 (0.816–5.040)	0.128
Alcoholism	1.287 (0.499–3.318)	0.602
Cirrhosis	1.623 (0.687–3.843)	0.269
Tumor number	0.968 (0.325–2.886)	0.954
ALT	0.506 (0.185–1.386)	0.185
AST	0.543 (0.210–1.405)	0.208
AFP	0.912 (0.386–2.154)	0.834
TNM	1.618 (0.652–4.016)	0.3
Differentiation (low vs high/moderate)	2.762 (1.126–6.776)	0.027	2.593 (1.049–6.410)	0.039
GAS5-AS1 expression (high vs low)	0.354 (0.147–0.851)	0.02	0.370 (0.153–0.898)	0.028

Abbreviation: $P<$ 0.05 was considered statistically significant; ALT, alanine aminotransferase; AST, aspartate aminotransferase; AFP, $\alpha$ -fetoprotein; TNM, tumor-node-metastasis.

3.2 Correlation between GAS5-AS1 and clinical variables

According to the median expression levels (median $=$ 0.011) of GAS5-AS1, HCC patients were divided into low expression group ( $n=$ 41) and high expression group ( $n=$ 42). Then compared with different clinical features, including gender, age, smoking, alcoholism, tumor size, cirrhosis, AFP, HBV-DNA, differentiation, TNM stage and other biochemistry indexes. Correlation analysis showed that the expression of GAS5-AS1 in HCC was significantly associated with tumor differentiation (High/Moderate vs Low, $P=$ 0.031), TNM stage (I $\sim$ II vs III $\sim$ IV, $P=$ 0.020) and glucose level ( $<$ 6.2 vs $\geqslant$ 6.2, $P=$ 0.047) (Table 1).

Figure 2.

Association between down-regulation of GAS5-AS1 and prognosis in HCC.

3.3 Association between GAS5-AS1 levels and patients’ survival

In order to explore the prognostic value of GAS5-AS1 in HCC, we assessed the relationship between the GAS5-AS1 expression and overall survival (OS) by using the Kaplan-Meier method. The survival signatures were analyzed in 51 patients, and 28 of them were in the high-GAS5-AS1 group, 23 patients were in the low group. Comparing with high-GAS5-AS1 group, low GAS5-AS1 expression levels were significantly associated with a shorter survival time (median OS: 21 month) ( $P=$ 0.020, Fig. 2), indicating that GAS5-AS1 expression level could be consider as a novel indicator for prognosis in HCC.

Then univariate and multivariate analysis were conducted to explore the prognostic factors in HCC patients. Univariate analysis showed that differentiation ( $P=$ 0.027) and GAS5-AS1 expression ( $P=$ 0.020) were significantly associated with OS in HCC patients. Multivariate analysis was performed using the Cox hazards model, which confirmed that both GAS5-AS1 expression ( $P=$ 0.039) and differentiation ( $P=$ 0.028) were prognostic factors for HCC patients (Table 2).

Figure 3.

GAS5-AS1 in plasma among subgroups. A. GAS5-AS1 in HCC was lower than that in hepatitis B, cirrhosis and the controls. The expression levels were calculated using the comparative cycle threshold (Ct) method ( $\Delta$ Ct) B. Correlation between plasma GAS5-AS1 and the tissues. The relative expression level was calculated using 2 ${}^{-\Delta\text{CT}}$ method. * $P<$ 0.05, ** $P<$ 0.01.

Figure 4.

Receiver operating characteristic (ROC) curves. The ROC curves of the expression of GAS5-AS1 for A. HCC vs Hepatitis B; B. HCC (AFP $<$ 200 ng/ml) vs Hepatitis B (AFP $<$ 200 ng/ml); C. HCC vs Cirrhosis; D. HCC (AFP $<$ 200 ng/ml) vs Cirrhosis (AFP $<$ 200 ng/ml); E. HCC vs Controls; F. Cirrhosis vs Controls.

3.4 GAS5-AS1 expression in plasma among subgroups

In addition to evaluate GAS5-AS1 expression in HCC tissues, we also tested its expression in plasma in 63 HCC patients, 46 hepatitis B patients, 47 cirrhosis patients, and 58 control cases. The main demographic and clinical characteristics of studied subjects were illustrated in Table 3. No difference was observed in important risk factors including gender, age, smoking, and alcoholism ( $P>$ 0.05) in the four groups.

The levels of GAS5-AS1 expression in subgroups were measured by RT-qPCR. The result elaborated that the GAS5-AS1 expression level in HCC was lower than that in hepatitis B, cirrhosis and control groups (HCC vs hepatitis B: $P<$ 0.01; HCC vs cirrhosis: $P<$ 0.01; HCC vs the controls: $P<$ 0.01). Beyond that, comparing with the controls, the expression of GAS5-AS1 in cirrhosis groups were higher (cirrhosis vs controls: $P<$ 0.05). However, no difference was observed in another two groups (controls vs hepatitis B: $P=$ 0.23; hepatitis B vs cirrhosis: $P=$ 0.22) (Fig. 3A).

Among 63 HCC plasma, 23 samples were from the patients whose tissues were also collected. Then we analyzed the association between tissues and plasma by detecting GAS5-AS1 expression in the 23 paired samples. There was a positive relationship between tissues and plasma in the expression of GAS5-AS1. Spearman correlation analysis showed that the correlation coefficient was 0.480 ( $P=$ 0.02) (Fig. 3B). Therefore, we demonstrated that the expression of GAS5-AS1 in human circulation was closely correlated with the original HCC tissues.

3.5 Diagnostic value of GAS5-AS1 in circulation

To test whether GAS5-AS1 could be used as a marker for HCC patients, we performed ROC, which has been established as a standard to determine the value of biomarker. ROC was conducted in 4 groups: HCC vs the controls, HCC vs hepatitis B, HCC vs cirrhosis, and cirrhosis vs controls (Table 4). The AUC ${}_{\text{ROC}}$ indicated that GAS5-AS1 had a high sensitivity of 85.7% (AUC $=$ 0.744, 95% CI: 0.648–0.839) (Fig. 4A) in discriminating HCC patients from patients with hepatitis B. When we defined the cut-off value of AFP at 200 ng/ml both in HCC and hepatitis B cases, the sensitivity was up to 89.5% (AUC $=$ 0.743, 95% CI: 0.647–0.841) (Fig. 4B). Meanwhile, AUC ${}_{\text{ROC}}$ showed that GAS5-AS1 could be considered as a better diagnostic indicator for differentiating HCC patients from the cirrhosis (AUC $=$ 0.824, 95% CI: 0.741–0.906) (Fig. 4C). Moreover, ROC result suggested that plasma GAS5-AS1 conferred a higher sensitivity for detecting HCC patients ( $n=$ 57) with AFP levels below 200 ng/ml from cirrhosis cases ( $n=$ 45) (AUC $=$ 0.819, 95% CI: 0.733–0.906) (Fig. 4D) whose AFP levels were also below 200 ng/ml. However, the diagnostic value of GAS5-AS1 in another two groups was not noticeable (Fig. 4E and F).

Table 3
Characteristics of the studied subjects

Characteristics	HCC	Hepatitis B	Cirrhosis	Control	$P$
	$N=$ 63	$N=$ 46	$N=$ 47	$N=$ 58
Gender					0.404
Male	54	34	39	49
Female	9	12		9
Age					0.095
$<$ 55	22	21	15	13
$\geqq$ 55	41	25	32	45
Smoking					0.161
Negative	40	30	38	36
Positive	23	16	9	22
Alcoholism					0.464
Negative	47	33	29	38
Positive	16	13	18	20

Table 4

Comparisons of the AUC of the expression of GAS5-AS1 for subgroups

Group	AUC	95% CI	$P$	Se (%)	Sp (%)
HCC vs Hepatitis B	0.744	0.648–0.839	$<$ 0.001	85.7	56.5
HCC vs Hepatitis B#1	0.743	0.647–0.841	$<$ 0.001	89.5	53.3
HCC vs Cirrhosis	0.824	0.741–0.906	$<$ 0.001	87.3	70.2
HCC vs Cirrhosis#2	0.819	0.733–0.906	$<$ 0.001	89.5	68.9
HCC vs Control	0.683	0.584–0.781	0.001	93.7	46.6
Cirrhosis vs Control	0.622	0.515–0.729	0.032	48.3	74.5

Abbreviation: Se: Sensitivity; Sp: Specificity; #1 HCC with AFP levels below 200 ng/ml and Hepatitis B whose AFP levels also below 200 ng/ml; #2 HCC with AFP levels below 200 ng/ml and Cirrhosis whose AFP levels also below 200 ng/ml.

4. Discussion

HCC is a prevalent liver malignancy with poor prognosis, despite the therapeutic advances in HCC treatment recently [20, 21]. The pathogenesis of HCC is concealed, its progress is rapid, and the mortality rate is high [22]. Thus, novel biomarkers for HCC are intensively desired in clinic. Researches showed that HCC cells could express some fetal liver specific transcript factors and signal molecules [23]. Recently, mounting evidence has pointed to a relationship between lncRNAs and cancers, including metastasis, migration, apoptosis and clinical outcome [24]. Several lncRNAs have been implicated in human liver diseases [25]. In 2013, Xie et al. [26] indicated that highly up-regulated in liver cancer (HULC) is a specific gene which was markedly up-regulated both in HCC tissue and plasma. Metastasis Associated Lung Adenocarcinoma Transcript 1 (MALAT1) was up-regulated in HCC and acted as a proto-oncogene through Wnt pathway activation and induction of the oncogenic splicing factor serine/arginine-Rich Splicing Factor 1 (SRSF1) [27]. In 2016, Lv et al. [28] demonstrated that Unigene56159 significantly promoted cell migration and invasion through EMT in HCC. It is believed that lncRNAs play important regulatory roles in cancer progression [29]. Hence, it is particularly prospective to explore diagnostic and prognostic of lncRNAs for HCC.

For the first time, we investigated the expression pattern and clinical value of GAS5-AS1 in HCC patients. We found that the GAS5-AS1 expression level was lower in HCC tissues than in the paired adjacent normal tissues and its level was significantly associated with differentiation, TNM stage and glucose level. Comparing to the low-GAS5-AS1 group, the high group had a better prognosis and a longer survival time. The Cox hazards model elaborated that poorer differentiation and lower expression level of GAS5-AS1 were risk factors for HCC patients. Meanwhile, we detected the expression of GAS5-AS1 in plasma. The results suggested that GAS5-AS1 expressions in HCC were significantly lower than those in the controls, the hepatitis B and the cirrhosis. Remarkably, the levels of GAS5-AS1 expression in cirrhosis were higher than in controls. The possible reason might be that GAS5-AS1 was released when the normal tissues developed premalignant lesion. Unfortunately, the exact mechanisms was still undiscovered, needing further studies to explain. Moreover, we detected the relationship between the circulating RNA and the RNA in tissues. We found that there was a positive correction between them. In 2014, Kumarswamy et al. [30] indicated that circulating lncRNAs were from the tissues, which were confirmed by our result. Based on these results, it is believed that GAS5-AS1 could be used as a candidate prognostic and non-invasive biomarker for HCC.

Isolation of nucleic acids from plasma might be an ideal method for screening patients at an early stage and prefiguring their prognosis [31]. Meantime, circulating RNA has been an emerging field for non-invasive diagnostic applications in HCC [32]. Therefore, we evaluated the diagnostic value of GAS5-AS1 in HCC plasma. The results demonstrated that GAS5-AS1 could be helpful for distinguishing HCC patients from the healthy, with AUC of 0.683 and a good sensitivity of 93.7%. Additionally, the AUC ${}_{\text{ROC}}$ elaborated that GAS5-AS1 could be considered as a good diagnostic marker for making a distinction between HCC patients and the cirrhosis. What was more, GAS5-AS1 yielded a high sensitivity of 85.7% for differentiating HCC from patients with hepatitis B. When differentiating HCC cases with AFP $<$ 200 ng/ml from the cirrhosis and hepatitis B whose AFP levels were also below 200 ng/ml, GAS5-AS1 had a higher sensitivity (89.5%, 89.5%), respectively. Unfortunately, statistical significance in another two groups was not obvious. Although these data indicated that GAS5-AS1 could not be used to monitor the process which hepatitis B developed into cirrhosis, it could be valuable to diagnosis the premalignant disease from HCC.

Epithelial-mesenchymal transition (EMT) is not only a mechanism that forms fibroblast-like cells, it can result in cancer cell migration, invasion, metastasis and therapeutic resistance, which plays a key role in step toward cancer metastasis [33, 34, 35]. In addition, cells lose their adhesive properties and undergo alterations in polarity and reorganization of the cytoskeleton during the process of EMT [36]. Accumulating studies have revealed that the involvement of several transcription factors, known as key EMT regulators, including the Snai family and the $\delta$ EF1 family [37]. In recent years, the role of EMT in the advancement of HCC has gained increasing attention [38]. Li et al. [39] demonstrated that EMT enhanced the carcinogenesis of HCC, and therapies targeting EMT appeared to be a promising treatment. In 2016, Zhang et al. [40] suggested that EMT played a pivotal role in the dissemination of malignant hepatocytes during HCC progression. Meantime, GAS5-AS1 in NSCLC cells could regulate a cohort of molecules including ZEB1, N-cadherin, Vimentin, and/or Snail1, which are critical molecules for EMT [18]. On account of these studies, we supposed that GAS5-AS1 might play an important role in the development of HCC, being a suppressor as well as in NSCLC. However, the exact mechanism of GAS5-AS1 is still unclear in HCC.

In summary, our study investigated the expression levels of GAS5-AS1 in HCC patients for the first time. We found that GAS5-AS1 expression levels were down-regulated in HCC tissues and correlated with differentiation, TNM stage and glucose levels. Furthermore, the survival analysis indicated that GAS5-AS1 was an independent prognostic factor for HCC patients. Comparing with the high GAS5-AS1 group, the low group had a poor prognosis and a shorter survival time. Although our data supported that plasma GAS5-AS1 could be a potential diagnostic biomarker for HCC patients, it still needed further vitro and vivo trial to validate the precise role of GAS5-AS1 in HCC.

Footnotes

Acknowledgments

This study was supported by the National Basic Research Program of China (973 Program) (2012CB720 605).

Conflict of interest

The authors have declared that no conflict of interest.

References

Tian

Gao

Zhang

Chen

Zhang

and Yuan

, Identification a novel tumor-suppressive hsa-miR-599 regulates cells proliferation, migration and invasion by targeting oncogenic MYC in hepatocellular carcinoma, Am J Transl Res 8 (2016), 2575–84.

Zhou

Song

Zhao

Cheng

Zhao

and Jia

, CHST11/13 regulate the metastasis and chemosensitivity of human hepatocellular carcinoma cells via mitogen-activated protein kinase pathway, Dig Dis Sci 61 (2016), 1972–85.

Fan

J.L.

Yang

Y.F.

Yuan

C.H.

Chen

and Wang

F.B.

, Circulating tumor cells for predicting the prognostic of patients with hepatocellular carcinoma: A meta analysis, Cell Physiol Biochem 37 (2015), 629–40.

Liu

Y.R.

Tang

R.X.

Huang

W.T.

Ren

F.H.

R.Q.

Yang

L.H.

Luo

D.Z.

Dang

Y.W.

and Chen

, Long noncoding RNAs in hepatocellular carcinoma: Novel insights into their mechanism, World J Hepatol 7 (2015), 2781–91.

Zhu

Wang

Huang

Liu

Gao

Tian

and Fan

, LncBRM initiates YAP1 signalling activation to drive self-renewal of liver cancer stem cells, Nat Commun 7 (2016), 13608.

El-Serag

H.B.

, Hepatocellular carcinoma, N Engl J Med 365 (2011), 1118–27.

Lan

Chang

and Yuan

, Downregulation of ZEB2-AS1 decreased tumor growth and metastasis in hepatocellular carcinoma, Mol Med Rep 14 (2016), 4606–4612.

Visvader

J.E.

, Cells of origin in cancer, Nature 469 (2011), 314–22.

Jing

Gao

Zhu

Luo

Jing

Chai

and Tu

, Potential diagnostic value of lncRNA SPRY4-IT1 in hepatocellular carcinoma, Oncol Rep 36 (2016), 1085–92.

10.

Ponting

C.P.

Oliver

P.L.

and Reik

, Evolution and functions of long noncoding RNAs, Cell 136 (2009), 629–41.

11.

M.Z.

Chu

B.F.

Zhang

Weng

M.Z.

Qin

Y.Y.

Gong

and Quan

Z.W.

, Long non-coding RNA CCAT1 promotes gallbladder cancer development via negative modulation of miRNA-218-5p, Cell Death Dis 6 (2015), e1583.

12.

Brockdorff

Ashworth

Kay

G.F.

McCabe

V.M.

Norris

D.P.

Cooper

P.J.

Swift

and Rastan

, The product of the mouse Xist gene is a 15 kb inactive X-specific transcript containing no conserved ORF and located in the nucleus, Cell 71 (1992), 515–26.

13.

Venkatraman

X.C.

Thorvaldsen

J.L.

Sugimura

Perry

J.M.

Tao

Zhao

Christenson

M.K.

Sanchez

J.Y.

Peng

Haug

J.S.

Paulson

Zhong

X.B.

Clemens

T.L.

Bartolomei

M.S.

and Li

, Maternal imprinting at the H19-Igf2 locus maintains adult haematopoietic stem cell quiescence, Nature 500 (2013), 345–9.

14.

Maruyama

and Suzuki

, Long noncoding RNA involvement in cancer, BMB Rep 45 (2012), 604–11.

15.

Yang

Zhou

L.M.

Lai

M.C.

Xie

H.Y.

Zhang

and Zheng

S.S.

, Overexpression of long non-coding RNA HOTAIR predicts tumor recurrence in hepatocellular carcinoma patients following liver transplantation, Ann Surg Oncol 18 (2011), 1243–50.

16.

Jia

Jiang

Wang

Y.D.

Huang

J.Z.

and Xue

H.Z.

, lincRNA-p21 inhibits invasion and metastasis of hepatocellular carcinoma through Notch signaling-induced epithelial-mesenchymal transition, Hepatol Res 46 (2016), 1137–1144.

17.

Wen

Sun

Xing

and Yin

, Upregulation of long non coding RNA PCAT-1 contributes to cell proliferation, migration and apoptosis in hepatocellular carcinoma, Mol Med Rep 13 (2016), 4481–6.

18.

Lyu

Liu

Shi

Song

and Liu

, Downregulation of the long noncoding RNA GAS5-AS1 contributes to tumor metastasis in non-small cell lung cancer, Sci Rep 6 (2016), 31093.

19.

Chang

Lan

Yuan

Liu

and Liu

, Decreased expression of long non-coding RNA GAS5 indicates a poor prognosis and promotes cell proliferation and invasion in hepatocellular carcinoma by regulating vimentin, Mol Med Rep 13 (2016), 1541–50.

20.

Yang

Zhong

Xie

Lai

Nie

Liu

and Wan

Y.J.

, Induction of the liver cancer-down-regulated long noncoding RNA uc002mbe.2 mediates trichostatin-induced apoptosis of liver cancer cells, Biochem Pharmacol 85 (2013), 1761–9.

21.

Huang

M.D.

Chen

W.M.

F.Z.

Xia

Sun

T.P.

Yin

Zhang

E.B.

and Shu

Y.Q.

, Long non-coding RNA ANRIL is upregulated in hepatocellular carcinoma and regulates cell apoptosis by epigenetic silencing of KLF2, J Hematol Oncol 8 (2015), 50.

22.

Tang

Jiang

Deng

Zhang

Wang

and Sun

, Circulation long non-coding RNAs act as biomarkers for predicting tumorigenesis and metastasis in hepatocellular carcinoma, Oncotarget 6 (2015), 4505–15.

23.

Wang

Yuan

J.H.

Wang

S.B.

Yang

Yuan

S.X.

Yang

Zhou

W.P.

W.L.

and Sun

S.H.

, Oncofetal long noncoding RNA PVT1 promotes proliferation and stem cell-like property of hepatocellular carcinoma cells by stabilizing NOP2, Hepatology 60 (2014), 1278–90.

24.

Wang

T.H.

Lin

Y.S.

Chen

Yeh

C.T.

Huang

Y.L.

Hsieh

T.H.

Shieh

T.M.

Hsueh

and Chen

T.C.

, Long non-coding RNA AOC4P suppresses hepatocellular carcinoma metastasis by enhancing vimentin degradation and inhibiting epithelial-mesenchymal transition, Oncotarget 6 (2015), 23342–57.

25.

Takahashi

Yan

I.K.

Kogure

Haga

and Patel

, Extracellular vesicle-mediated transfer of long non-coding RNA ROR modulates chemosensitivity in human hepatocellular cancer, FEBS Open Bio 4 (2014), 458–67.

26.

Xie

and Zhou

, Plasma HULC as a promising novel biomarker for the detection of hepatocellular carcinoma, Biomed Res Int 2013 (2013), 136106.

27.

Malakar

Shilo

Mogilavsky

Stein

Pikarsky

Nevo

Benyamini

Elgavish

Zong

Prasanth

K.V.

and Karni

, Long noncoding RNA MALAT1 promotes hepatocellular carcinoma development by SRSF1 up-regulation and mTOR activation, Cancer Res (2016).

28.

Fan

H.X.

Zhao

X.P.

Fan

J.Y.

Zhang

Liu

and Tang

, Long non-coding RNA Unigene56159 promotes epithelial-mesenchymal transition by acting as a ceRNA of miR-140-5p in hepatocellular carcinoma cells, Cancer Lett 382 (2016), 166–175.

29.

Fang

T.T.

Sun

X.J.

Chen

Zhao

Sun

R.X.

Ren

and Liu

B.B.

, Long non-coding RNAs are differentially expressed in hepatocellular carcinoma cell lines with differing metastatic potential, Asian Pac J Cancer Prev 15 (2014), 10513–24.

30.

Kumarswamy

Bauters

Volkmann

Maury

Fetisch

Holzmann

Lemesle

de Groote

Pinet

and Thum

, Circulating long noncoding RNA, LIPCAR, predicts survival in patients with heart failure, Circ Res 114 (2014), 1569–75.

31.

Wang

Tang

Jiang

Zhang

and Sun

, HULC and Linc00152 Act as Novel Biomarkers in Predicting Diagnosis of Hepatocellular Carcinoma, Cell Physiol Biochem 37 (2015), 687–96.

32.

Bao

Wang

Zhang

Tao

Cui

and Jiang

, The plasma lncRNA acting as fingerprint in non-small-cell lung cancer, Tumour Biol (2015).

33.

Xie

Nie

F.Q.

Sun

Xia

Liu

Y.W.

Zhou

and Liu

X.H.

, Decreased long noncoding RNA SPRY4-IT1 contributing to gastric cancer cell metastasis partly via affecting epithelial-mesenchymal transition, J Transl Med 13 (2015), 250.

34.

Kalluri

and Weinberg

R.A.

, The basics of epithelial-mesenchymal transition, J Clin Invest 119 (2009), 1420–8.

35.

Zhang

Yang

Yuan

J.H.

Yuan

S.X.

Zhou

W.P.

Huo

X.S.

H.S.

Wang

and Sun

S.H.

, Epigenetic activation of the MiR-200 family contributes to H19-mediated metastasis suppression in hepatocellular carcinoma, Carcinogenesis 34 (2013), 577–86.

36.

X.L.

Liu

D.D.

Y.P.

Guo

L.H.

Sun

L.P.

Liu

L.N.

H.X.

and Zhang

X.P.

, Integrin beta4 promotes cell invasion and epithelial-mesenchymal transition through the modulation of Slug expression in hepatocellular carcinoma, Sci Rep 7 (2017), 40464.

37.

Saitoh

, Epithelial-mesenchymal transition is regulated at post-transcriptional levels by transforming growth factor-beta signaling during tumor progression, Cancer Sci 106 (2015), 481–8.

38.

Jayachandran

Dhungel

and Steel

J.C.

, Epithelial-to-mesenchymal plasticity of cancer stem cells: Therapeutic targets in hepatocellular carcinoma, J Hematol Oncol 9 (2016), 74.

39.

Song

Wang

Yao

Tian

Cai

and Wang

, Suppression of epithelial-mesenchymal transition in hepatocellular carcinoma cells by Kruppel-like factor 4, Oncotarget 7 (2016), 29749–60.

40.

Zhang

P.F.

K.S.

Shen

Y.H.

Gao

P.T.

Dong

Z.R.

Cai

J.B.

Zhang

Huang

X.Y.

Tian

M.X.

Z.Q.

Gao

D.M.

Fan

A.W.

and Shi

G.M.

, Galectin-1 induces hepatocellular carcinoma EMT and sorafenib resistance by activating FAK/PI3K/AKT signaling, Cell Death Dis 7 (2016), e2201.

Down-regulation of long non-coding RNA GAS5-AS1 and its prognostic and diagnostic significance in hepatocellular carcinoma

Abstract

BACKGROUND:

METHODS:

RESULTS:

CONCLUSIONS:

Keywords

1. Introduction

2. Materials and methods

2.1 Specimens

2.2 Ethical approval

2.4 Real-time PCR analysis

2.5 Statistical analysis

3. Results

3.1 GAS5-AS1 level was significantly lower in HCC tissues

Table 1 Association of GAS5-AS1 expression with clinical parameters in HCC

3.5 Diagnostic value of GAS5-AS1 in circulation

Table 3 Characteristics of the studied subjects

Footnotes

Acknowledgments

Conflict of interest

References

Table 1
Association of GAS5-AS1 expression with clinical parameters in HCC

Table 3
Characteristics of the studied subjects