Serum microRNA expression profiling identifies serum biomarkers for HCV-related hepatocellular carcinoma

Abstract

BACKGROUND:

The identification of high-sensitivity biomarkers for detection of hepatocellular carcinoma (HCC) from high-risk individuals is essential.

OBJECTIVE:

The present study was undertaken to identify and validate serum microRNAs (miRNAs) as potential biomarkers for hepatitis C virus (HCV)-related HCC.

METHODS:

Illumina sequencing was employed to screen the expression profiles of miRNAs in serum samples of HCV-related HCC patients and liver cirrhosis (LC) patients. RT-qPCR was used to confirm the altered miRNAs between the two groups. Moreover, candidate miRNAs were examined in serum samples of 40 HCC patients, 54 LC patients, 55 patients with chronic HCV hepatitis and 45 healthy controls. Receiver operating characteristic (ROC) curve analysis was used to evaluate the diagnostic performance of the miRNAs for the detection of HCC.

RESULTS:

Four miRNAs (miR-122-5p, miR-331-3p, miR-494-3p, miR-224-5p) were significantly increased and two miRNAs (miR-185-5p, miR-23b-3p) were significantly decreased in HCC patients compared to LC patients. ROC curve analysis demonstrated that the six miRNAs could be used as potential biomarkers for HCC detection. Combination of the six miRNAs could efficiently detect HCC in LC patients with the area under the ROC curve (AUC) of 0.995 and combination of the six miRNAs also provided high diagnostic accuracy (AUC $=$ 0.961) for detection of HCC in non-HCC subjects.

CONCLUSIONS:

The six serum miRNAs can be utilized as a surrogate and non-invasive biomarker for HCV-related HCC diagnosis.

Keywords

Hepatocellular carcinoma MiRNAs serum hepatitis c virus liver cirrhosis

1. Introduction

Hepatocellular carcinoma (HCC), the most prevalent type of liver cancer, has a very poor prognosis [1]. Hepatitis C virus (HCV) is a major etiologic agent of chronic hepatitis, cirrhosis and HCC in the world [2, 3, 4]. In China, the number of individuals infected with HCV is increasing and the incidence of HCV infection has increased from 0.7 to 15.0 cases per 100 000 persons in the past 10 years [3].

While great progress has been made in the treatment of HCC, the HCC prognosis remains poor and adverse effects of chemotherapy are common [1]. Early diagnosis of HCC and timely treatment can greatly improve life expectancy and reduce mortality. However, only 30% to 40% of patients with HCC are qualify for potential curative intervention at diagnosis because of the lack of clinical presentation and the abscence of potential early identification biomarkers [5]. Thus, new tumor markers with high sensitivity and specificity for HCC are urgently needed.

MicroRNAs (miRNAs) are small, noncoding regulatory RNAs. They inversely regulate the expression of their target genes at the posttranscriptional level by inhibiting translation or causing the degradation of the target messenger RNA (mRNA) [6]. MiRNAs are involved in various biological processes including proliferation, differentiation and apoptosis but also play an important role in the occurrence and development of a variety of diseases including HCC [7, 8, 9, 10]. Evidence has shown differential expression of miRNAs between tumor and normal tissue and miRNAs may act as tumour suppressors, oncogenes, or even possess a dual nature playing both roles depending on the cellular requirements [8, 10]. MiRNAs are tissue-specific and are stable in plasma or serum even after being subjected to severe conditions such as extreme temperature and low pH, which establish the role of miRNAs as non-invasive biomarkers in disease diagnosis [11, 12, 13, 14, 15].

In this study, we performed profiling of miRNAs by high throughput small RNA sequencing in order to identify a set of miRNAs served as efficient biomarkers for detection of HCV-related HCC from high-risk individuals. Candidate miRNAs were selected and validated by using a stem-loop RT-qPCR assay. A combined panel of six miRNAs (miR-122-5p, miR-331-3p, miR-494-3p, miR-224-5p, miR-185-5p, miR-23b-3p) was demonstrated could discriminate HCC efficiently from high-risk individuals.

2. Material and methods

2.1 Study design and patients

In the initial stage, pooled serum samples from 5 HCV-related HCC patients and 5 liver cirrhosis (LC) patients served as controls underwent Illumina sequencing to identify miRNAs that showed significant differences between the HCC patients and LC controls. Subsequently, we performed a biomarker confirmation analysis with an RT-qPCR assay. This analysis was carried out in 2 phases: (a) 15 HCV-related HCC patients and 15 LC patients formed the training set. (b) The selected miRNAs were examined by RT-qPCR in the serum samples from 40 HCC patients, 54 LC patients, 60 HCV carriers and 45 healthy controls.

HCV-related HCC patients, HCV-related LC patients and patients with chronic hepatitis C (CH) were recruited from our institution during the study period. All HCC patients were on top of HCV cirrhosis and HCC diagnosis was based on American Association for the Study of Liver Diseases (AASLD) Practice Guidelines [16] and staged by the Barcelona Clinic Liver Cancer (BCLC) system [17]. Patients with HCV-related LC were diagnosed by abdominal ultrasonography and confirmed by histopathology examination. Patients with CH were characterized by the persistent increase of the alanine aminotransferase (ALT) values (more than three times normal) for at least 6 months [18]. Patients with other neoplasm, virus infection or live disease were excluded from the study. 45 healthy volunteers without liver diseases and any tumors were served as healthy controls. The study is approved by the ethic committee of You’an Hospital Affiliated to Capital Medical University and patients signed the informed consent.

2.2 Serum collection and RNA extraction

Before curative treatment, blood samples were obtained. After centrifugation at 3000 revolutions/min for 15 min at room temperature, the supernatant was transferred into a new tube. The supernatant was then centrifuged at 15,000 revolutions/min using a high-speed refrigerated centrifuge for 30 min. Then, the supernatant sera were stored at $-$ 80 ${}^{\circ}$ C. The total RNA of each pool was extracted by using TRIzol reagent (Invitrogen) according to the manufacturer’s instructions.

Table 1
Demographic and clinical characteristics of the patients for Illumina sequencing in the profiling study

	Male	Age (years)	AST (IU/l)	ALT (IU/l)	Total bilirubin (mg/dl)	AFP (ng/ml)	Albumin (g/dl)	Hemoglobin (g/dl)	BCLC score (stage)
HCC patients
1	No	57	103.62	39.82	19.51	27.47	32.2	128	B
2	No	60	28.62	21.77	12	2365	33.8	138	B
3	Yes	57	137.9	204.3	17.1	41.7	32.9	141	B
4	Yes	65	50.51	33.7	51.76	50.8	36.4	149	A
5	Yes	57	39.5	43.19	18.14	15.8	33.8	138	A
LC patients
1	No	57	51.6	55.5	16.9	33	39.9	145
2	No	59	36.8	59.7	15.1	43	39.1	126
3	Yes	62	77.7	31.7	50.6	7.6	40.7	136
4	Yes	61	39.4	18.8	15.8	26	42.2	140
5	Yes	59	95.4	74.3	12.2	3.6	39.8	143

Figure 1.

Length of distribution of total clean reads.

2.3 Small RNA library construction and Illumina sequencing

Through PAGE purification, total small RNA mole-cules under 30 bp were isolated. After ligating a pair of adaptors to their 5’ and 3’ ends, the small RNA molecules were amplified for 17 cycles and then fragments about 90 bp were isolated after gel purification. Illumina HiSeq 2500 was used for sequencing analysis according to the manufacturer’s instructions. Then we processed the data by computational analysis.

2.4 Quantification of miRNAs by RT-qPCR analysis

To validate the miRNAs discovered by sequencing, RT-qPCR was used. The RNA concentration was quantified with a DU 800 spectrophotometer (Beckman Coulter), Reverse transcription was carried out using a TaqMan MicroRNA Reverse Transcription Kit (Applied Biosystems) according to the manufacturer’s protocol: 16 ${{}^{\circ}}$ C for 30 min, 42 ${{}^{\circ}}$ C for 30 min and 85 ${{}^{\circ}}$ C for 5 min to generate a library of miRNA cDNAs. PCR was subsequently carried out using the Roche 480 real-time PCR system using a TaqMan MicroRNA assay (Applied Biosystems) for the following reaction: 95 ${{}^{\circ}}$ C for 10 min, 40 cycles of 95 ${{}^{\circ}}$ C for 15 s, and 60 ${{}^{\circ}}$ C for 1 min. Cel-miR-39 was used as an external control gene and the 2(- $\Delta$ Ct) method was employed to assess miRNA expression. A cel-miR-39 Ct value higher than 30 was considered exclusion criteria [19, 20].

Table 2
Demographic and clinical characteristics of the patients in the first confirmation analysis set

Parameter	LC ( $n=$ 15)		HCC ( $n=$ 15)		$P$ value
Age (years)	58.73 $\pm$	7.69	61.15 $\pm$	8.97	NS
Male, $n$ (%)	10 (66.67)		10 (66.67)		NS
ALT (IU/l)	77.89 $\pm$	66.93	79.63 $\pm$	42.39	NS
AST (IU/I)	68.89 $\pm$	65.63	65.89 $\pm$	41.23	NS
Total bilirubin (mg/dl)	27.56 $\pm$	18.56	24.98 $\pm$	14.96	NS
Albumin (g/dl)	37.59 $\pm$	4.47	36.45 $\pm$	4.62	NS
Hemoglobin (g/dl)	130.24 $\pm$	29.58	130.69 $\pm$	31.22	NS
AFP (ng/ml)	27.56 $\pm$	41.23	165.49 $\pm$	214.56	$<$ 0.001
BCLC score $N$ (%)
Stage 0			0 (0)
Stage A			6 (40)
Stage B			8 (53.33)
Stage C			1 (6.66)

3. Statistical analysis

MiRNAs values were normalized to cel-miR-39 and are expressed as 2- $\Delta$ CT (CT sample-CT cel-miR-39) [19, 20]. The quantitative data are presented as the mean $\pm$ standard deviation. All continuous variables were checked with the Kolmogorov-Smirnov normality test to show their distributions. Continuous variables with normal distributions were compared using the unpaired Student’s t-test and analysis of variance (ANOVA). Continuous variables with abnormal distributions were compared using the Mann-Whitney U test and ANOVA. For categorical variables, the chi-square test was used. Correlation between the miRNAs was analyzed using Pearson’s correlation coefficient. Receiver operating characteristic (ROC) curves were constructed, AUC values were estimated and analyzed to evaluate the accuracy of individual and combined markers to discriminate between HCC and controls as described [21, 22]. $P$ values $<$ 0.05 were considered statistically significant. Statistical calculations were performed using SPSS 16.0.

Table 3
Demographic and clinical characteristics of the patients in the second confirmation analysis set

Parameter	Health ( $n=$ 45)		CH ( $n=$ 60)		LC ( $n=$ 54)		HCC ( $n=$ 40)		$P$ value
Age (years)	46.52 $\pm$	10.24	39.22 $\pm$	11.18	57.89 $\pm$	8.96	62.37 $\pm$	9.68	$<$ 0.001
Male, $n$ (%)	26 (57.78)		32 (58.18)		34 (62.96)		27 (65.96)		0.055
ALT (IU/l)	22.54 $\pm$	9.67	70.35 $\pm$	48.22	84.73 $\pm$	87.94	78.91 $\pm$	42.39	$<$ 0.001
AST (IU/I)	27.05 $\pm$	8.87	56.47 $\pm$	24.22	72.5 $\pm$	45.63	68.41 $\pm$	36.35	$<$ 0.001
Total bilirubin (mg/dl)	8.56 $\pm$	4.21	14.23 $\pm$	9.23	28.41 $\pm$	24.66	24.53 $\pm$	13.32	$<$ 0.001
Albumin (g/dl)	43.82 $\pm$	5.23	43.66 $\pm$	4.69	37.99 $\pm$	5.29	36.58 $\pm$	5.2	$<$ 0.001
Hemoglobin (g/dl)	145.63 $\pm$	17.12	143.9 $\pm$	18.69	123.27 $\pm$	35.45	128.65 $\pm$	23.6	$<$ 0.001
AFP (ng/ml)	3.72 $\pm$	1.15	9.33 $\pm$	1.82	35.56 $\pm$	41.23	171.75 $\pm$	332.23	$<$ 0.001
BCLC score $N$ (%)
Stage 0							0 (0)
Stage A							14 (35)
Stage B							20 (50)
Stage C							4 (10)
Stage D							2 (5)

4. Results

4.1 Patients and sample characteristics

In the profile study, we included 5 HCV-related HCC patients and 5 LC patients as controls. The characteristics of the patients are shown in Table 1. In the first confirmation analysis set, 15 HCV-related HCC patients and 15 LC patients were included. The characteristics of the patients are shown in Table 2. There were no significant differences in the age, gender and transaminase between the two groups. The level of AFP was significantly higher in HCC patients than in LC patients. In the second confirmation analysis set, 40 HCC patients, 54 LC patients, 60 CH patients and 45 healthy controls (HC) were included. The characteristics of the patients are shown in Table 3. As showed by the table, patients differed across groups with regard to age ( $P<$ 0.001). The differences in ALT, AST, total bilirubin and AFP were statistically significant among groups ( $P<$ 0.001). During liver disease progression, levels of albumin, hemoglobin, and platelets tended to decrease among studied groups ( $P<$ 0.001, for each).

4.2 Sequencing of serum miRNAs

We obtained 8,892,160 clean reads (60.84%) from 14,616,725 total reads in the LC group and 10,825,202 clean reads (72.49%) from 14,933,961 total reads in the HCC group (Table 4). Length distribution analysis showed that most reads were distributed around 22 nt (Fig. 1).

Table 4
Sequencing data

Sample	LC	HCC
Total raw reads	14,616,725	14,933,961
Total raw bases	730,836,250	746,698,050
Total clean reads	8,892,160	10,825,202
Total clean bases	216,309,318	249,012,479
Clean reads rate (%)	60.84	72.49
Low quality reads	5427	747
Low quality rate (%)	0.04	0.01
Without 3p adaptor reads	1,313,383	546,859
Without 3p adaptor rate (%)	8.99	3.66
Without insert reads	492,168	7,429
Without insert rate (%)	3.37	0.05
Poly A/T reads	667,670	3,464,732
Poly A/T rate (%)	4.57	23.2
Ex-length reads	3,245,917	88,992
Ex-length rate (%)	22.21	0.6
Raw reads Q20 (%)	97.26	96.68
Raw reads Q30 (%)	94.51	95.49
Clean reads Q20 (%)	97.49	97.4
Clean reads Q30 (%)	94.59	96.44

Table 5

Ten miRNAs were selected from sequencing data

miRNA name	HCC count	LC count	Fold change	$\log_{2}$ fold change
hsa-miR-185-5p	246	2471	0.15	$-$ 2.69
hsa-miR-494-3p	886	163	8.45	3.08
hsa-miR-331-3p	866	191	7.05	2.82
hsa-miR-150-5p	3183	1018	4.86	2.28
hsa-miR-4433b-3	1278	422	4.71	2.23
hsa-miR-224-5p	1538	550	4.35	2.12
hsa-miR-122-5p	11200	4611	3.77	1.92
hsa-miR-877-5p	1332	629	3.29	1.72
hsa-miR-127-3p	563	2531	0.35	$-$ 1.53
hsa-miR-23b-3p	508	2171	0.36	$-$ 1.46

Table 6

The expression quantities of the miRNAs (2- $\Delta$ CT) in HCC group versus LC group in first confirmation phase

miRNA name	HCC ( $n=$ 15)	LC ( $n=$ 15)	Fold change	$\log_{2}$ fold change	$P$ value
hsa-miR-185-5p	0.22	0.89	0.25	$-$ 2.02	0.040
hsa-miR-494-3p	0.99	0.18	5.5	2.46	0.025
hsa-miR-331-3p	0.67	0.19	3.53	1.82	0.020
hsa-miR-150-5p	2.07	0.83	2.49	1.32	0.073
hsa-miR-4433b-3	1.46	0.57	2.56	1.36	0.096
hsa-miR-224-5p	1.71	0.62	2.73	1.46	0.018
hsa-miR-122-5p	2.69	0.69	3.89	1.96	0.011
hsa-miR-877-5p	1.47	0.79	1.86	0.89	0.204
hsa-miR-127-3p	0.37	0.83	0.44	$-$ 1.66	0.081
hsa-miR-23b-3p	0.27	0.77	0.35	$-$ 1.51	0.038

Figure 2.

Relative expression of miR-122-5p, miR-331-3p, miR-494-3p, miR-224-5p, miR-185-5p and miR-23b-3p in different groups (2- $\Delta$ CT).

Figure 3.

ROC curves for miR-122-5p, miR-331-3p, miR-494-3p, miR-224-5p, miR-185-5p and miR-23b-3p for discrimination of HCC from LC.

Figure 4.

ROC curves for miR-122-5p, miR-331-3p, miR-494-3p, miR-224-5p, miR-185-5p and miR-23b-3p for discrimination of HCC from non-malignant controls (HC $+$ CH $+$ LC).

Table 7

Significant relations among enrolled serum miRNAs in hepatocellular carcinoma patients

miRNA name	miR-331-3p	miR-122-5p	miR-494-3p	miR-224-5p	miR-185-5p	miR-23b-3p
miR-331-3p		$r=$ 0.338	NS	$r=$ 0.347	$r=$ 0.365	$r=$ 0.415
miR-331-3p		$P=$ 0.033		$P=$ 0.028	$P=$ 0.021	$P=$ 0.008
miR-122-5p	$r=$ 0.338		NS	NS	$r=$ 0.429	$r=$ 0.350
miR-122-5p	$P=$ 0.033				$P=$ 0.006	$P=$ 0.027
miR-494-3p	NS	NS		NS	NS	NS
miR-494-3p	$r=$ 0.347	NS	NS		$r=$ 0.377	$r=$ 0.519
$P=$ 0.028				$P=$ 0.018	$P=$ 0.001
miR-185-5p	$r=$ 0.365	$r=$ 0.429	NS	$r=$ 0.377		$r=$ 0.302
miR-185-5p	$P=$ 0.021	$P=$ 0.006		$P=$ 0.018		$P=$ 0.058
miR-23b-3p	$r=$ 0.415	$r=$ 0.350	NS	$r=$ 0.519	NS
miR-23b-3p	$P=$ 0.008	$P=$ 0.027		$P=$ 0.001

Figure 5.

ROC curves for combination of miR-122-5p, miR-331-3p, miR-494-3p, miR-224-5p, miR-185-5p and miR-23b-3p for discriminate of HCC. A diagnosis of HCC against LC B diagnosis of HCC against non-malignant controls (HC $+$ CH $+$ LC).

MiRBase database (release 21) was used to identify known miRNAs. 502 miRNAs were found in LC patients and 839 miRNAs were found in HCC patients. A miRNA was considered altered if sequencing detected 500 copies in any patient group and the miRNA showed at least a 1.5 log2 fold change or less than $-$ 1.5 log2 fold change difference between the HCC patients and LC patients. Based on these criteria, ten miRNAs were selected as candidate miRNAs (Table 5).

4.3 Evaluation of miRNA expression by RT-qPCR

We used RT-qPCR assay to confirm the expression of candidate miRNAs. In the first confirming set, 15 HCV-related HCC patients and 15 LC patients were used (Table 2).

The expression quantities of the miRNAs (2- $\Delta$ CT) in the HCC group versus LC group were shown in Table 6. MiRNAs significantly different between the two groups ( $P<$ 0.05) were selected for further analysis. Thus, Six miRNAs were selected. Comparing to LC group, miR-122-5p, miR-331-3p, miR-494-3p and miR-224-5p were increased significantly in HCC group and the P values were 0.011, 0.020, 0.025 and 0.018 respectively. MiR-185-5p and miR-23b-3p were decreased significantly in the HCC group compared to the LC group, and the $P$ values were 0.040 and 0.038 respectively.

In the second conforming phase, the six selected miRNAs of miR-122-5p, miR-331-3p, miR-494-3p, miR-224-5p, miR-185-5p and miR-23b-3p were measured in all groups of HCC, CH, LC and HC. MiR-122-5p, miR-494-3p and miR-224-5p were all upregulated significantly early in CH patients versus health controls. The level of miR-494-3p was then significantly downregulated in LC patients compared to CH patients. The levels of the four miRNAs of miR-122-5p, miR-494-3p, miR-224-5p and miR-331-3p were all significantly higher in HCC patients versus LC patients or CH patients. Except miR-331-3p, levels of the other three miRNAs were also significantly higher in HCC patients versus health controls. The level of miR-185-5p was found significantly downregulated in CH patients and then upregulated in LC patients. In HCC patients the level of miR-185-5p was significantly downregulated compared to LC patients. No significant differences were found in levels of miR-23b-3p between CH, LC and health groups. MiR-23b-3p was significantly downregulated in HCC patients compared to LC patients, CH patients or health controls (Fig. 2).

4.4 ROC curve analysis

ROC analysis revealed that studied miRNAs could discriminate between HCC and LC with AUC $=$ 0.808 for miR-122-5p (95% CI 0.721-0.895, $P<$ 0.001), 0.726 for miR-331-3p (95% CI 0.616–0.836, $P<$ 0.001), 0.709 for miR-494-3p (95% CI 0.605–0.812, $P<$ 0.001), 0.762 for miR-224-5p (95% CI 0.672–0.851, $P<$ 0.001), 0.721 for miR-185-5p (95% CI 0.618-0.824, $P<$ 0.001), and 0.664 for miR-23b-3p (95% CI 0.551–0.777, $P=$ 0.007), respectively (Fig. 3).

For discriminating HCC from non-malignant controls (HC $+$ CH $+$ LC), the AUC of miR-122-5p was 0.747 (95% CI 0.664-0.829, $P<$ 0.001), miR-331-3p was 0.658 (95% CI 0.543–0.773, $P=$ 0.002), miR-494-3p was 0.607 (95% CI 0.499–0.714, $P=$ 0.033), miR-224 was 0.759 (95% CI 0.688–0.830, $P<$ 0.001), miR-185-5p was 0.662 (95% CI 0.564–0.761, $P=$ 0.002), and miR-23b-3p was 0.758 (95% CI 0.688–0.849, $P<$ 0.001), respectively (Fig. 4).

AUC of combined the six miRNAs for discriminating HCC from LC patients was 0.995 (95% CI 0.987–1.003, $P<$ 0.001) and AUC of combined the six miRNAs for discriminating HCC from non-malignant controls (HC $+$ CH $+$ LC) was 0.961 (95% CI 0.930–0.992, $P<$ 0.001) (Fig. 5).

4.5 Serum miRNAs’ correlation in the HCC study group

MiR-331-3p was correlated significantly with miR-122-5p ( $r=$ 0.338, $P=$ 0.033), miR-224-5p ( $r=$ 0.347, $P=$ 0.028), miR-185-5p ( $r=$ 0.365, $P=$ 0.021) and miR-23b-3p ( $r=$ 0.415, $P=$ 0.008), respectively. MiR-122-5p was correlated with miR-185-5p ( $r=$ 0.429, $P=$ 0.006) and miR-23b-3p ( $r=$ 0.350, $P=$ 0.027). miR-224-5p was correlated with miR-185-5p ( $r=$ 0.377, $P=$ 0.018) and miR-23b ( $r=$ 0.519, $P=$ 0.001) (Table 7).

5. Discussion

Patients with LC are known to be at high risk of developing HCC. Early diagnosis of HCC among these patients is important to reduce the mortality of the disease. Imaging examinations and biopsy are the major means for diagnosis of HCC at present. But imaging examinations can not distinguish the benign cancer and malignant cancer well. [23] and biopsy has the risk of tumor cells seeding along the needle track [24]. So biomarkers in peripheral blood can efficiently distinguish HCC are urgently needed.

MiRNAs, an integral component of gene regulation networks controlling tumorigenesis are stably maintained in a variety of biological fluids, including human serum/plasma [14, 15]. The quantities of circulating miRNAs reflect the physiological and pathological conditions of the human body and circulating miRNAs have been suggested candidates for a blood-based biomarker for diagnosis and prognosis in many cancers [11].

In the present study, we used small RNA sequencing technology as a discovery step, followed by RT-qPCR-based validation steps to discovery HCV-related HCC specific miRNAs. Our miRNA profiling analysis in sera from LC patients and HCC patients found tent miRNAs with different expressions. The further validation analysis confirmed that six miRNAs of miR-122-5p, miR-331-3p, miR-494-3p, miR-224-5p, miR-185-5p, miR-23b-3p were differentially expressed in serum of HCV-related HCC patients versus LC patients, implicating these miRNAs as surrogate biomarkers of HCC.

MiR-122, regulating lipid and cholesterol metabo-lism, comprises 70% of the total miRNAs in the liver. MiR-122 is important for HCV stability in liver cells and is known to regulate HCV stability by binding to an unstructured 5’UTR of HCV-RNA [25]. It has been reported that miR-122 serves as a tumor suppressor in HCC development by targeting multiple target genes such as metalloprotease 17 (ADAM17) [26], Bcl-w (anti-apoptotic gene) [27] and cyclin G1 (CCGN1) [28] and miR-122 expression was downregulated in HCC tumor tissues [28] while its serum levels were demonstrated significantly upregulated in HCC patients in comparison to LC patients or healthy controls [18, 29, 30, 31]. These contrary results may be due to its increased release [31, 32]. In the present study, miR-122-5p was demonstrated to be significantly upregulated in CH patients compared to healthy controls and the serum levels of miR-122-5p were upregulated significantly in HCV related HCC patients compared to LC patients, CH patients or healthy controls. Our findings also confirmed the role of miR-122 as a diagnostic marker for detection HCC in the LC group and control group.

In our study, levels of serum miR-494-3p were found upregulated significantly in HCC patients compared to all controls including LC patients, CH patients and healthy controls. MiR-494-3p was identified as a tumor driver in lung cancer [33], glioma [34], and HCC [35]. In HCC, miR-494-3p promotes cell proliferation, migration and invasion by targeting phosphatase and tensin homolog (PTEN) [35]. MiR-224-5p was also supposed an oncogenic miRNA during hepatocarcinogenesis and was demonstrated higher expressed in HCC tissue than paired adjacent noncancerous liver tissue [36]. Consistent with its tissue expression, in the current study, serum MiR-224-5p was demonstrated to be significantly unregulated in HCC patients compared to all controls. MiR-331-3p, another tumorigenic miRNA, promotes proliferation and metastasis of HCC through suppression of leucine-rich repeat protein phosphatase (PHLPP) mediated dephosphorylation of protein kinase B (AKT) [37]. In the present study, we found serum levels of miR-331-3p were upregulated significantly in HCC patients compared to CH patients and LC patients. Similar results were previously reported by Chen L, et al. [38] who demonstrated higher levels of serum miR-331-3p in HCC patients than patients with benign liver diseases. In the present study, we demonstrated higher levels of serum miR-494-3p, miR-224-5p and miR-331-3p in HCC patients, suggesting the roles of the miRNAs as potential biomarkers for HCC detection.

MiR-185-5p and miR-23b-3p were two downregulated serum miRNAs in HCC patients in our study. It has been reported that miR-185 acts as a tumor suppressor in HCC by targeting multiple genes [39, 40, 41]. Besides, recently Li et al. [42] found miR-185-5p was downregulated by HCV core protein, participating in the occurrence of liver steatosis which is a frequent histologic finding in CH. In the present study, serum miRNA-185-5p was downregulated significantly in HCC patients compared to LC patients, suggesting the diagnostic role of miRNA-185-5p in discriminating HCC from LC patients. MiR-23b-3p is a tumor suppressor which may regulate HCC migration and invasion by targeting Pyk2 via regulation of epithelial-mesenchymal transition (EMT) and may be a therapeutic target for HCC treatment [43]. In tissue, miR-23b-3p was reported aberrantly downregulated in HCC in contrast with the counterpart adjacent non-cancerous HCC [44]. Consistent with its reported tissue expression, levels of serum miRNA were significantly downregulated in HCC patients compared to all controls including LC patients, CH patients and healthy controls in our study. This suggests that serum miRNA may be a novel, non-invasive biomarker for the diagnosis of HCC.

To further analyze the diagnostic capacities of miR-122-5p, miR-331-3p, miR-494-3p, miR-224-5p, miR-185-5p, miR-23b-3p, receiver operating characteristic curves were done. Areas under the curves showed all of the six miRNAs could significantly discriminate HCV-related HCC from LC or from non-malignant controls. For discriminating HCC from LC, the AUC of miR-122-5p was largest with 0.808, second by miR-224-5p with AUC of 0.762. For discriminating HCC from non-malignant controls, the AUC of miR-224-5p was largest with 0.758, second by miR-23b-3p with AUC of 0.762. Combined, the six miRNAs could efficiently discriminate HCC from LC with AUC of 0.995 and AUC of 0.961 to discriminate HCC from non-malignant controls.

In conclusion, six miRNAs (miR-122, miR-331, miR-494, miR-224, miR-185 and miR-23b) were identified that may be useful as biomarkers of HCV-related HCC, especially for high-risk individuals. When used in combination the six miRNAs had a high degree of accuracy in detecting HCC in high risk individuals. Further studies may be indicated to validate these findings.

Footnotes

Acknowledgments

This study was funded by the National Grand Program on Key Infectious Diseases (No. 2015ZX100048 01).

Conflict of interest

The authors declare no conflict of interest.

References

Zhu

R.X.

Seto

W.K.

Lai

C.L.

and Yuen

M.F.

, Epidemiology of hepatocellular carcinoma in the Asia-Pacific region, Gut Liver 10 (2016), 332–339.

Lee

M.H.

Yang

H.I.

Yuan

L’Italien

and Chen

C.J.

, Epidemiology and natural history of hepatitis C virus infection, World J Gastroenterol 20 (2014), 9270–9280.

Qin

Smith

M.K.

Wang

Guo

Wang

Cui

and Wang

, Hepatitis C virus infection in China: An emerging public health issue, J Viral Hepat 22 (2015), 238–244.

Zhang

Ouyang

Jiang

Lin

Kong

S.K.

and Xie

, Dysregulated serum MicroRNA expression profile and potential biomarkers in hepatitis C virus-infected patients, Int J Med Sci 12 (2015), 590–598.

Tan

Pan

Wen

Chen

Zhou

and Gan

, A serum MicroRNA panel as potential biomarkers for hepatocellular carcinoma related with hepatitis B virus, PLoS One 9 (2014), e107986.

Bartel

D.P.

, MicroRNAs: Genomics, biogenesis, mechanism, and function, Cell 116 (2004), 281–297.

Berindan-Neagoe

Monroig Pdel

Pasculli

and Calin

G.A.

, MicroRNAome genome: A treasure for cancer diagnosis and therapy, CA Cancer J Clin 64 (2014), 311–336.

Gao

Ning

Liu

Wei

Tang

Feng

and Zhang

, Integrated analysis of differentially expressed mRNAs and miRNAs between hepatocellular carcinoma and their matched adjacent normal liver tissues, Oncol Rep 34 (2015), 325–333.

Kloosterman

W.P.

and Plasterk

R.H.

, The diverse functions of MicroRNAs in animal development and disease, Dev Cell 11 (2006), 441–450.

10.

Esquela-Kerscher

and Slack

F.J.

, Oncomirs–MicroRNAs with a role in cancer, Nat Rev Cancer 6 (2006), 259–269.

11.

Chen

Cai

Yin

Wang

Guo

Zhang

Chen

Guo

Wang

Zhang

Wang

Jiang

Xiang

Zheng

Zhang

Shang

Gong

Ning

Wang

Zen

Zhang

and Zhang

C.Y.

, Characterization of MicroRNAs in serum: A novel class of biomarkers for diagnosis of cancer and other diseases, Cell Res 18 (2008), 997–1006.

12.

Kelly

B.D.

Miller

Sweeney

K.J.

Durkan

G.C.

Rogers

Walsh

and Kerin

M.J.

, A circulating MicroRNA signature as a biomarker for prostate cancer in a high risk group, J Clin Med 4 (2015), 1369–1379.

13.

Zhu

Ren

Han

Ding

Dai

Shen

and Jin

, A five-MicroRNA panel in plasma was identified as potential biomarker for early detection of gastric cancer, Br J Cancer 110 (2014), 2291–2299.

14.

Mitchell

P.S.

Parkin

R.K.

Kroh

E.M.

Fritz

B.R.

Wyman

S.K.

Pogosova-Agadjanyan

E.L.

Peterson

Noteboom

O’Briant

K.C.

Allen

Lin

D.W.

Urban

Drescher

C.W.

Knudsen

B.S.

Stirewalt

D.L.

Gentleman

Vessella

R.L.

Nelson

P.S.

Martin

D.B.

and Tewari

, Circulating microRNAs as stable blood-based markers for cancer detection, Proc Natl Acad Sci U S A 105 (2008), 10513–10518.

15.

Jiang

Yao

Guo

and Ding

, Stability analysis of liver cancer-related MicroRNAs, Acta Biochim Biophys Sin (Shanghai) 43 (2011), 69–78.

16.

Bruix

and Sherman

, Management of hepatocellular carcinoma, Hepatology 42 (2005), 1208–1236.

17.

Forner

Reig

M.E.

de Lope

C.R.

and Bruix

, Current strategy for staging and treatment: The BCLC update and future prospects, Semin Liver Dis 30 (2010), 61–74.

18.

Zekri

A.N.

Youssef

A.S.

El-Desouky

E.D.

Ahmed

O.S.

Lotfy

M.M.

Nassar

A.A.

and Bahnassey

A.A.

, Serum MicroRNA panels as potential biomarkers for early detection of hepatocellular carcinoma on top of HCV infection, Tumour Biol 37 (2016), 12273–12286.

19.

Wen

Han

Chen

Dong

Xia

Liu

Jiang

Dai

Jin

Han

Wei

Shen

Sun

and Hu

, Plasma miRNAs as early biomarkers for detecting hepatocellular carcinoma, Int J Cancer 137 (2015), 1679–1690.

20.

Livak

K.J.

and Schmittgen

T.D.

, Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Δ⁢Δ C(T)) method, Methods 25 (2001), 402–408.

21.

Jiang

Cheng

Zhang

B.H.

and Zhang

M.Z.

, Circulating MicroRNAs as biomarkers in hepatocellular carcinoma screening: A validation set from China, Medicine (Baltimore) 94 (2015), e603.

22.

Ali

H.E.A.

Abdel Hameed

Effat

Ahmed

E.K.

Atef

A.A.

Sharawi

S.K.

Ali

Abd Elmageed

Z.Y.

and Abdel Wahab

A.H.

, Circulating MicroRNAs panel as a diagnostic tool for discrimination of HCV-associated hepatocellular carcinoma, Clin Res Hepatol Gastroenterol 41 (2017), e51–e62.

23.

Block

Mehta

A.S.

and London

W.T.

, Hepatocellular carcinoma of the liver, Cancer Biomark 9 (2010), 375–383.

24.

Stigliano

Marelli

Davies

Patch

and Burroughs

A.K.

, Seeding following percutaneous diagnostic and therapeutic approaches for hepatocellular carcinoma. What is the risk and the outcome? Seeding risk for percutaneous approach of HCC, Cancer Treat Rev 33 (2007), 437–447.

25.

Janssen

H.L.

Reesink

H.W.

Lawitz

E.J.

Zeuzem

Rodriguez-Torres

Patel

van der Meer

A.J.

Patick

A.K.

Chen

Zhou

Persson

King

B.D.

Kauppinen

Levin

A.A.

and Hodges

M.R.

, Treatment of HCV infection by targeting MicroRNA, N Engl J Med 368 (2013), 1685–1694.

26.

Tsai

W.C.

Hsu

P.W.

Lai

T.C.

Chau

G.Y.

Lin

C.W.

Chen

C.M.

Lin

C.D.

Liao

Y.L.

Wang

J.L.

Chau

Y.P.

Hsu

M.T.

Hsiao

Huang

H.D.

and Tsou

A.P.

, MicroRNA-122, a tumor suppressor MicroRNA that regulates intrahepatic metastasis of hepatocellular carcinoma, Hepatology 49 (2009), 1571–1582.

27.

Lin

C.J.

Gong

H.Y.

Tseng

H.C.

Wang

W.L.

and Wu

J.L.

, miR-122 targets an anti-apoptotic gene, Bcl-w, in human hepatocellular carcinoma cell lines, Biochem Biophys Res Commun 375 (2008), 315–320.

28.

Zhu

Yang

Wang

and Wu

, MicroRNA-122 suppresses cell proliferation and induces cell apoptosis in hepatocellular carcinoma by directly targeting Wnt/beta-catenin pathway, Liver Int 32 (2012), 752–760.

29.

Che

Wang

Zhang

Huang

Tan

Wang

and Lin

, Circulating MicroRNAs, miR-21, miR-122, and miR-223, in patients with hepatocellular carcinoma or chronic hepatitis, Mol Carcinog 50 (2011), 136–142.

30.

Cheng

S.Q.

Wang

Chen

Y.F.

and Gao

C.F.

, Serum MicroRNAs as biomarkers for hepatocellular carcinoma in Chinese patients with chronic hepatitis B virus infection, PLoS One 6 (2011), e28486.

31.

Hung

C.H.

T.H.

S.N.

Kuo

F.Y.

Chen

C.H.

Wang

J.H.

Huang

C.M.

Lee

C.M.

Lin

C.Y.

Yen

Y.H.

and Chiu

Y.C.

, Circulating MicroRNAs as biomarkers for diagnosis of early hepatocellular carcinoma associated with hepatitis B virus, Int J Cancer 138 (2016), 714–720.

32.

Giordano

and Columbano

, MicroRNAs: New tools for diagnosis, prognosis, and therapy in hepatocellular carcinoma? Hepatology 57 (2013), 840–847.

33.

Faversani

Amatori

Augello

Colombo

Porretti

Fanelli

Ferrero

Palleschi

Pelicci

P.G.

Belloni

Ercoli

Degrassi

Baccarin

Altieri

D.C.

Vaira

and Bosari

, miR-494-3p is a novel tumor driver of lung carcinogenesis, Oncotarget 8 (2017), 7231–7247.

34.

X.T.

Wang

H.Z.

Z.W.

Yang

T.Q.

Zhao

Z.H.

Chen

G.L.

Xie

X.S.

Wei

Y.X.

Huang

Y.L.

Zhou

Y.X.

and Du

Z.W.

, miR-494-3p regulates cellular proliferation, invasion, migration, and apoptosis by PTEN/AKT signaling in human glioblastoma cells, Cell Mol Neurobiol 35 (2015), 679–687.

35.

Liu

Zhang

Jia

Tan

Jin

and Liu

, miR-494 promotes cell proliferation, migration and invasion, and increased sorafenib resistance in hepatocellular carcinoma by targeting PTEN, Oncol Rep 34 (2015), 1003–1010.

36.

Shi

K.Q.

Lin

Chen

X.J.

Song

Wang

Y.Q.

Cai

Y.J.

Yang

N.B.

Zheng

M.H.

Dong

J.Z.

Zhang

and Chen

Y.P.

, Hepatocellular carcinoma associated MicroRNA expression signature: Integrated bioinformatics analysis, experimental validation and clinical significance, Oncotarget 6 (2015), 25093–25108.

37.

Chang

R.M.

Yang

Fang

J.F.

and Yang

L.Y.

, MicroRNA-331-3p promotes proliferation and metastasis of hepatocellular carcinoma by targeting PH domain and leucine-rich repeat protein phosphatase, Hepatology 60 (2014), 1251–1263.

38.

Chen

Chu

Cao

Shao

and Wang

, Serum miR-182 and miR-331-3p as diagnostic and prognostic markers in patients with hepatocellular carcinoma, Tumour Biol 36 (2015), 7439–7447.

39.

Qadir

X.V.

Han

Zhang

and Wu

, miR-185 inhibits hepatocellular carcinoma growth by targeting the DNMT1/PTEN/Akt pathway, Am J Pathol 184 (2014), 2355–2364.

40.

Zhu

S.M.

Chen

C.M.

Jiang

Z.Y.

Yuan

F.H.

and Jin

, MicroRNA-185 inhibits cell proliferation and epithelial-mesenchymal transition in hepatocellular carcinoma by targeting Six2, Eur Rev Med Pharmacol Sci 20 (2016), 1712–1719.

41.

Zhi

Zhu

Guo

Xue

Zhou

Chen

Zhao

and Kuang

, Metastasis-related miR-185 is a potential prognostic biomarker for hepatocellular carcinoma in early stage, Biomed Pharmacother 67 (2013), 393–398.

42.

Wang

Liu

S.A.

Zhang

J.Q.

Quan

Feng

S.H.

Dong

J.L.

Gao

and Cheng

, MicroRNA-185-5p mediates regulation of SREBP2 expression by hepatitis C virus core protein, World J Gastroenterol 21 (2015), 4517–4525.

43.

Cao

Liu

Long

Huang

Liu

Zhang

and Yan

, MicroRNA-23b suppresses epithelial-mesenchymal transition (EMT) and metastasis in hepatocellular carcinoma via targeting Pyk2, Biomed Pharmacother 89 (2017), 642–650.

44.

R.Q.

P.R.

Xiang

X.L.

Yang

Liang

H.W.

Qiu

X.H.

Yang

L.H.

Peng

Z.G.

and Chen

, Downregulated miR-23b-3p expression acts as a predictor of hepatocellular carcinoma progression: A study based on public data and RT-qPCR verification, Int J Mol Med 41 (2018), 2813–2831.

Serum microRNA expression profiling identifies serum biomarkers for HCV-related hepatocellular carcinoma

Abstract

BACKGROUND:

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSIONS:

Keywords

1. Introduction

2. Material and methods

2.1 Study design and patients

2.2 Serum collection and RNA extraction

Table 1 Demographic and clinical characteristics of the patients for Illumina sequencing in the profiling study

2.4 Quantification of miRNAs by RT-qPCR analysis

Table 2 Demographic and clinical characteristics of the patients in the first confirmation analysis set

Table 3 Demographic and clinical characteristics of the patients in the second confirmation analysis set

4.1 Patients and sample characteristics

4.2 Sequencing of serum miRNAs

Table 4 Sequencing data

4.4 ROC curve analysis

4.5 Serum miRNAs’ correlation in the HCC study group

5. Discussion

Footnotes

Acknowledgments

Conflict of interest

References

Table 1
Demographic and clinical characteristics of the patients for Illumina sequencing in the profiling study

Table 2
Demographic and clinical characteristics of the patients in the first confirmation analysis set

Table 3
Demographic and clinical characteristics of the patients in the second confirmation analysis set

Table 4
Sequencing data