Association Between Two Genetic Variants in miRNA and Primary Liver Cancer Risk in the Chinese Population

Abstract

MicroRNAs (miRNAs) play an important role in the growth and development of human beings. Single nucleotide polymorphisms (SNPs) within miRNA could change their production or affinity with target genes, thus leading to malignant diseases. This case–control study conducted in Western China aimed to explore the relationship between polymorphisms in miR-146a (rs2910164 G>C) and miR-499 (rs3746444 T>C) and primary liver cancers in the Chinese population. 186 primary liver cancer cases and 483 healthy controls were genotyped using polymerase chain reaction–restriction fragment length polymorphism. No significant differences were observed between distributions of the two SNPs and susceptibility of primary liver cancer or diverse clinicopathologic features. However, we found that patients with genotype CG of the SNP in miR-146a tended to have earlier onset and better liver function than patients with genotype CC (average age: 49.9 vs. 54.9, p=0.038; average Child-Pugh grade: 5.55 vs. 6.15, p=0.021), and further analysis showed that patients who had at least one G allele were diagnosed at an earlier age (average age: 49.6 vs. 54.9, p=0.022) and had better liver function (average Child-Pugh grade:5.60 vs. 6.15, p=0.026). Our data suggested lack of association between the two SNPs and primary liver cancer risk, though, interestingly, the miR-146a SNP may influence the age of onset and Child-Pugh grade.

Introduction

P rimary liver cancer is the sixth most common cancer in the world (748,300 new cases per year), and the third most common cause of cancer deaths (695,900 deaths per year) (Jemal et al., 2011). It is prevalent in Middle and Western Africa and in East and Southeast Asia, especially in China, where almost half of the cases and deaths occurred, owing to high prevalence of chronic hepatitis B virus (HBV) and hepatitis C virus (HCV) infections (Yuen et al., 2009). Epidemiological evidence suggested that the most dominant risk factor for liver cancer is chronic infection of HBV and HCV. Moreover, alcohol consumption and flavacin contamination of foodstuffs also could contribute to primary liver cancer (El-Serag et al., 2007). Further, genetic alterations were also proven to be involved in the etiology and prognosis of primary liver cancer in recent years (Thorgeirsson and Grisham, 2002; Hagymási and Tulassay, 2008).

MicroRNAs (miRNAs) are small noncoding functional RNAs of ∼22 nucleotides (Bartels and Tsongalis, 2009). They function as negative gene regulators (Ambros, 2004; Bartel, 2004), complementarily binding to hundreds of mRNA targets implicated in the regulation of almost every biological process (Ambros, 2003; Zamore and Haley, 2005). Alteration of miRNA expression was observed in initiation and progression of human cancers: miR-15a and miR-16 in chronic lymphocytic leukemia (Calin et al., 2008); miR-145 in colon carcinoma, lung, breast, and prostate (Iorio et al., 2005; Akao et al., 2006, 2007; Bandrés et al., 2006; Yanaihara et al., 2006); and miR-21 in glioblastoma (Chan et al., 2005). In the pathogenesis of primary liver cancer, many miRNAs were found to be involved, for example, let-7, miR-1, miR-101, miR-122 and miR-124 were downregulated while miR-17-5p, miR-106b-25, miR-143, and miR151 were upregulated, and they targeted genes such as c-Myc, HDAC4, p21, and Cyclin G1 (Huang and He, 2011). Single nucleotide polymorphisms (SNPs), the most abundant form of DNA variation in the human genome, may have significant effects on miRNA generation. Sequence variations within miRNA can alter the secondary structure of the miRNA precursor, thus disturbing miRNA maturation (Xu et al., 2008) and, consequently, changing the production output or the binding to targets, and finally altering protein expression, contributing to cancer susceptibility (Bartels and Tsongalis, 2009). To date, many studies have indicated that SNPs in pre-miRNA or miRNA could increase the risk of human diseases and lead to phenotypic differences (Bartels and Tsongalis, 2009; Hu et al., 2009; Peng et al., 2009), and thus were expected to be used for cancer diagnosis, staging, and prognosis (Ludwig and Weinstein, 2005).

We hypothesized that the SNPs in miR-146a (rs2910164 G>C) and miR-499 (rs3746444 T>C) are associated with primary liver cancer. To test this hypothesis, we genotyped the polymorphisms and assessed their relationship in the present case–control study (186 primary liver cancer cases and 483 healthy controls) in the Chinese population.

Materials and Methods

Study population

669 unrelated subjects who visited West China Hospital of Sichuan University, including 186 patients with primary liver cancer and 483 healthy individuals undergoing routine medical examination without any medical illness, were enrolled in this study between December 2008 and April 2010. All the patients were diagnosed either by histopathologic or imaging evidence. Informed consent for sample collection and subsequent analysis was obtained from each subject at recruitment. This study was approved by West China Hospital.

Genotyping assay of SNPs within miRNA

Genomic DNA was extracted from peripheral blood using Chelex-100, then genotyped using polymerase chain reaction (PCR)–restriction fragment length polymorphism, and confirmed by DNA sequencing analysis. The PCR mixture consisted of 5 pmol of each primer described previously (TAKARA) (Hu et al., 2008), 1.5 mM MgCl₂, 200 μM dNTP Mixture (Promega), 0.7 U Taq DNA Polymerase (Promega), 1× PCR buffer, and 2.0 μL exacted DNA; finally, the total volume of 25 μL was reached by adding deionized water. Then, it was initially heated at 94°C for 8 min to activate the polymerase, and DNA amplification was achieved by 40 cycles of (denaturing in 94°C for 30 s, annealing for 30 s, and prolonging in 72°C for 30 s) in a PCR Amplifier (ABI 9700 cycler), the annealing temperature was 60°C for rs2910164 and 62°C for rs3746444. The final elongation step was maintaining 72°C for 8 min. The next step was digestion by restriction enzyme; it was performed with 2 μL PCR product. For rs2910164, PCR products were digested overnight with 5 U SacI (MBI Fermentas) in a final volume of 7.5 μL at 37°C, whereas for rs3746444, PCR products were digested with 5 U BcII (MBI Fermentas) in a final volume of 8 μL at 65°C. Genotype calling was made by 3% agarose gel electrophoresis at 150 V for 1 h and photograghing under ultraviolet light with Gel Doc (Bio-Rad). 600 ng PCR products were randomly selected for DNA sequencing (ABI 3130).

Liver function and tumor marker tests

Electrochemiluminescence was used for the quantitative determination of human serum alpha fetoprotein (AFP) by ROCHE-E170 (Roche). Alanine aminotransferase (ALT), glutamic-oxalacetic transaminease (AST), glutamyl transpeptidase (GGT), total bilirubin (TB) and serum albumin (ALB) were all tested by automatic biochemical analyzer ROCHE-P800 (Roche). Quantitation of HBV-DNA was performed using real-time fluorescence quota PCR by ROCHE Light Cycler 480 (Roche). The quality control (Bio-Rad) and result audit were done by an experienced medical technologist. All assays were performed in the department of laboratory medicine in west China hospital of Sichuan University, which is a College of American Pathologists–certified lab.

Statistical analysis

Genotype frequencies were assessed for Hardy–Weinberg equilibrium in the case and control groups. Dualvariate unconditional logistic regression analysis was performed to investigate the relationship between the risk of primary liver cancer and the two SNPs by adjusting for gender and age. Genotype and allele frequencies were compared for the two SNPs in patients versus controls using Chi-square test. Odds ratios (ORs) and 95% confidence intervals for the risk of primary liver cancer were also calculated simultaneously. Further, in a subsequent analysis of the association between the polymorphisms and clinical characteristics, Chi-square test was used when the data were qualitative, such as the presence of cancer embolus and satellite, single or multifocal, or differentiation degree. If the data were quantitative, analysis of variance was carried out when the data had homogeneity of variance, whereas nonparametric test was performed when the data had heterogeneity of variance or non-normal distributions. All statistical analyses were performed using SPSS version 17.0 software. A p<0.05 was considered significantly different, and all statistical tests were two sided.

Results

Baseline characteristics of patients with primary liver cancer

This study involved 186 primary liver cancer patients and 483 healthy controls matched with patients by age and gender from Western China. Demographic and other baseline characteristics of patients with primary liver cancer are summarized in Table 1.

Table 1.

Baseline Data of Patients with Primary Liver Cancer (n=186)

	Patients, n (%) or mean±SD
Age of onset	52.10±15.20
Sex
Male	154 (82.8)
Female	32 (17.2)
Serum level of tumor markers
ALT (IU/L)	59.59±56.39
AST (IU/L)	74.16±59.72
GGT (IU/L)	199.33±233.46
AFP (ng/mL)	727.9 (10.98–1210)
TB (μM)	24.38±30.48
Child-Pugh grade	5.83±1.20
HBV-DNA (copies/mL)	9.810E03 (2.418E03-2.305E05)
Clinicalpathological charateratics (n=49)
Maximum diameter of tumor (n=41)	6.60±3.20
Presense of satellite (n=34)	4 (8.2)
Presense of cancer embolus (n=37)	22 (44.9)
Number of lesions (n=40)
Singlefocal	37 (75.5)
Multifocal	3 (6.1)
Histological differentiation (n=49)
Well	5 (10.2)
Moderately	20 (40.8)
Poorly	24 (49.0)

For non-normal distributions, median (P₂₅–P₇₅) was used.

SD, standard deviation; TB, total bilirubin; HBV, hepatitis B virus; AFP, alpha fetoprotein.

Distribution of genotype and allele frequencies and their association with primary liver cancer susceptibility

669 samples were genotyped from 186 patients with primary liver cancer and 483 healthy individuals. Genotype distributions of the two SNPs in both case and control groups were all in accordance with the Hardy–Weinberg equilibrium (all p>0.05), suggesting the genetic equilibrium of population enrolled (Table 2).

Table 2.

Genotype and Allele Distribution of Single-Nucleotide Polymorphisms Within miR-146 a and miR-499 in Primary Liver Cancer Patients and Controls

Polymorphism	Patient, n (%)	Control, n (%)	χ²	p	OR	95% CI
rs2910164 G>C
Genotype
CC	67 (36.0)	158 (32.7)	2.303	0.316
CG	86 (46.2)	254 (52.6)
GG	33 (17.7)	71 (14.7)
CG versus GG			1.682	0.195	0.728	0.451–1.177
CC versus GG			0.128	0.720	0.912	0.552–1.507
Allele
G	152 (40.9)	396 (41.0)	0.002	0.964	0.994	0.780–1.268
C	220 (59.1)	570 (59.0)
rs3746444 T>C
Genotype
CC	4 (2.2)	12 (2.48)	0.195	0.907
CT	41 (22.0)	100 (20.7)
TT	141 (75.8)	371 (76.8)
CT versus TT			0.130	0.718	1.079	0.715–1.629
CC versus TT			0.000	1.000	0.877	0.278–2.765
Allele
C	49 (13.2)	124 (12.8)	0.027	0.870	1.030	0.722–1.469
T	323 (86.8)	842 (87.2)

OR, odds ratio; CI, confidence interval.

As shown in Table 2, the frequencies of the genotype CC, CG, and GG of rs2910164 were 36%, 46.2%, and 17.7% in the patient group, and 32.7%, 52.6%, and 14.7% in the control group, and the allele frequencies were nearly the same in patient (C=59.1%, G=40.9%) and control (C=59.0%, G=41.0%) groups. As to the rs3746444, the frequencies of the genotype CC, CT, and TT were 2.2%, 22.0%, and 75.8% and 2.48%, 20.7%, and 76.8% in the case and control groups, respectively. The allele frequencies were also similar in patient (C=13.2%, T=86.8%) and control groups (C=12.8%, T=87.2%). The two SNPs within miR-146a and miR-499 showed no significant difference between primary liver cancer patients and healthy controls in either genotype (p=0.316 and p=0.907, respectively) and allelic analysis (p=0.964 and p=0.870, respectively). With regard to the SNP rs2910164 G>C, OR of G versus C, CG versus GG, CC versus GG, and (CC+GG) versus GG were all extremely close to 1. As shown in Table 2, the conclusions for SNP rs3746444 T>C were the same. Our study had a power of 83.4% (rs2910164) and 96.33% (rs3746444) to detect a difference in the distribution of the SNPs within miR-146a and miR-499 between case and controls.

To evaluate the association of primary liver cancer and the two SNPs without the interference of age and gender, dualvariate unconditional logistic regression analysis of the relationship between the two SNPs and the risk of primary liver cancer was performed adjusting for age and gender. In accordance with the Chi-square test, the distribution of the genotypes of the two SNPs showed no significant differences between cases and controls (p=0.6∼0.8).

Association between genotypes and phenotypes of primary liver cancer

AFP, HBsAg status, HBV-DNA, ALT, AST, GGT, and TB are all correlated with primary liver cancer and can reflect its progression and prognosis. Child-Pugh classification, developed in 1973, is a comprehensive assessment of liver function comprising quantitation of bilirubin, albumin, and prothrombin time, degree of ascites, and presence of hepatic encephalopathy, and it is widely used in prognosis of liver cancer. We investigated whether the mutations affected clinical features in 102 patients; the indicators above were obtained before treatment when the patients' disease was first diagnosed as primary liver cancer. In addition, we compared the age of onset among the genotypes of the two SNPs in all of the patients enrolled. No marker showed any significant difference in the genotypes of the two SNPs (Table 3). Interestingly, we found that patients with genotype CG tended to had earlier onset and lower Child-Pugh grade, compared with CC (average age: 49.9 vs. 54.9, p=0.038; average Child-Pugh grade: 5.55 vs. 6.15, p=0.021). Further, when we compared genotype GG+CG with CC by analysis of variance, we found that patients who had at least one G allele were diagnosed at an earlier age (average age 49.6 vs. 54.9, p=0.022) and lower Child-Pugh grade (average Child-Pugh grade: 5.60 vs. 6.15, p=0.026; Table 4).

Table 3.

Indexes Correlated with Liver Function (n=102) and Age at Primary Liver Cancer Onset (n=186) with Different Genotypes of the Two Single-Nucleotide Polymorphisms Within miR- 146 a and miR- 499

	rs2910164 G>C				rs3746444 T>C
	CC (mean±SD)	CG (mean±SD)	GG (mean±SD)	p-value	CC (mean±SD)	CT (mean±SD)	TT (mean±SD)	p-value
ALT (IU/L)	51.15±27.55	70.85±77.40	48.13±27.38	0.815^a	101.7±104.9	44.00±24.81	61.49±59.02	0.535^a
AST (IU/L)	76.34±55.72	76.57±68.81	60.80±37.58	0.648	86.33±90.01	50.11±26.78	79.05±63.15	0.320^a
GGT (IU/L)	180.4±138.3	226.6±290.5	167.4±172.6	0.530	230.0±202.1	144.1±127.6	210.5±240.2	0.511
TB (μM)	20.07±10.34	24.42±31.21	36.03±55.38	0.500^a	14.73±9.616	17.73±8.420	26.21±33.74	0.489
Child-Pugh grade	6.14±1.35	5.54±1.02	5.79±1.12	0.069	5.500±0.707	5.722±1.018	5.861±1.258	0.844
AFP^b (ng/mL)	441.2 (11.63–1210)	1048 (11.28–1210)	413.3 (3.590–1210)	0.577^a	389.7 (3.020–1210)	244.1 (13.11–1210)	977.7 (10.42–1210)	0.987^a
HBV-DNA^b (copies/mL)	9.810E03 (1.897E03–23628E05)	1.330E04 (2.435E03–2.405E05)	5.225E03 (1.470E03–6.995E05)	0.889^a	7.950E03 (0.000–0.590E04)	9.860E03 (5.990E03–5.380E05)	9.760E03 (2.380E03–2.390E05)	0.685^a
Age of onset	54.89±14.78	49.85±14.38	49.00±13.90	0.069	49.00±19.90	51.03±14.36	51.80±14.62	0.901

Owing to heterogeneity of variance or non-normal distributions, nonparametric test was used.

Owing to non-normal distributions, median (Q _U−Q _L) was used to describe the distribution of AFP and HBV-DNA.

Table 4.

Age of Onset and Improved Child-Pugh Grade with Genotypes of miR- 146 a Single-Nucleotide Polymorphism

	Age of onset (n=186)		Improved Child-Pugh grade (n=102)
Genotype	Mean±SD	p-value	Mean±SD	p-value
CG	49.85±14.38	0.038^a	5.54±1.02	0.021^a
CC	54.89±14.78	0.022^b	6.14±1.35	0.026^b
CG+GG	49.64±14.20		5.60±1.04

CG versus CC.

CG+GG versus CC.

Association between genotypes and pathological features of primary liver cancer

A subset of 49 patients with hepatocellular cancer who underwent surgical resection also underwent a pathological examination at diagnosis. We analyzed the pathological features to reveal whether there was any relationship with the distribution of genotypes. The pathological features for comparison included cancer embolus, single or multifocal, differentiation, satellite (Chi-square test for all the previous indexes), and maximum diameter (nonparametric test: Kruskal–Wallis), which greatly affected prognosis. However, as shown in Table 5, no significant difference was found among the genotypes, suggesting lack of relationship of the two SNPs and the pathological characteristics.

Table 5.

Clinicalpathological Characteristics and the Genotypes of Two Single-Nucleotide Polymorphisms Within miR- 146 a and miR- 499 in 49 Primary Liver Cancer Patients

	rs2910164G>C (n (%))				rs3746444T>C (n (%))
	CC	CG	GG	p-value	CC	CT	TT	p-value
Cancer embolus
Presence	6 (27.3)	12 (54.5)	4 (18.2)	0.990	0 (0.00)	6 (27.3)	16 (72.7)	0.297
Absence	4 (26.7)	8 (53.3)	3 (20.0)		1 (6.67)	6 (40.0)	8 (53.3)
Single or multifocal
Single	10 (27.2)	20 (54.1)	7 (18.9)	0.758	1 (2.70)	10 (27.0)	26 (70.3)	0.939
Multifocal	1 (33.3)	1 (33.3)	1 (33.3)		0 (0.00)	1 (33.3)	2 (66.7)
Differentiation
Poorly	7 (29.2)	15 (62.5)	2 (0.02)	0.401	0 (0.00)	7 (29.2)	17 (70.83)	0.788
Oderately	6 (30.0)	8 (40.0)	6 (30.0)		1 (5.00)	6 (30.0)	13 (65.0)
Well	1 (20.0)	3 (60.0)	1 (20.0)		0 (0.00)	2 (40.0)	3 (60.0)
Satellite
Presence	2 (50.0)	2 (50.0)	0 (0.00)	0.482	0 (0.00)	2 (50.0)	2 (50.0)	0.508
Absence	8 (26.7)	16 (53.3)	6 (20.0)		1 (3.33)	7 (23.3)	22 (73.3)
Maximum diameter of tumor (mean±SD)
	3.31±1.00	2.65±0.55	4.11±1.55	0.121	2.50	7.68±3.73	6.28±2.87	0.194

Discussion

Primary liver cancer is a deadly cancer whose incidence is increasing worldwide; there is a growing understanding of the molecular mechanism of hepatocarcinogenesis, mainly the role of oncogenes (Zender et al., 2010). In recent years, noncoding small RNAs, miRNAs, have been shown to participate in tumorigenesis, such as inflammation, cell cycle regulation, differentiation, apoptosis, and invasion (Farazi et al., 2011). For instance, down-regulated let-7 was found in nonsmall cell lung cancer patients (Johnson et al., 2007) and was associated with poor prognosis (Yanaihara et al., 2006), whereas highly expressed miR-21, which was frequently in solid and hematological malignancies, resulted in a pre-B malignant lymphoid-like phenotype, and miR-21 inactivation led to apoptosis and tumor regression (Medina et al., 2010). Previous studies have demonstrated that alteration of miR-146a expression was found in various malignances, such as PTC, breast cancer, and pancreatic cancer (He et al., 2005; Pallante et al., 2006; Volinia et al., 2006; Li et al., 2010b). SNPs within miRNA could significantly change its binding and post-transcriptional regulation (Cai et al., 2009) and contribute to carcinogenesis. A SNP within miR-196a2 was reported to be associated with susceptibility to hepatocellular carcinoma with cirrhosis (Li et al., 2010a) and chronic HBV infection (Qi et al., 2010) in Chinese population. Xu et al. (2011) found that a SNP in the promoter region of miR-34b/c contributed to primary hepatocellular carcinoma risk. We hypothesized that primary liver cancer progression is also affected by miRNA SNPs. Reports have demonstrated that SNPs within miR-146a and miR-499 played an important role in carcinogenesis. Zeng et al. (2010) reported that subjects with the variant genotypes (GC+GG) of miR-146a SNP had a 58% increased risk of gastric cancer (OR=1.58), compared with CC genotype carriers. Pastrello et al. (2010) found that BRCA1/BRCA2-negative familial breast and ovarian cancer patients with C allele of miR-146a SNP had earlier onset. miR-499 may play important roles in cardiac differentiation of human embryonic stem cells (Wilson et al., 2010) and regulating mitochondrial fission machinery and apoptosis (Wang et al., 2011). Variation of miR-499 is associated with increased risk of dilated cardiomyopathy and breast cancer (Hu et al., 2009; Zhou et al., 2010). Variants of miR-146a and miR-499 may be associated with primary liver cancer, which were assessed in this study.

In the present study, we genotyped miR-146a SNP (rs2910164 G>C) and miR-499 SNP (rs3746444 T>C) in 186 primary liver cancer patients and 483 healthy controls in a Chinese population. Our results suggested that genetic polymorphism of miR-146a is associated with age of onset and Child-Pugh grade in primary liver cancer patients. The genotype GC may contribute to early onset of primary liver cancer and relatively good liver function. Further analysis indicated that patients with at least one G allele had earlier onset and better liver function. However, this variant seemed to lack association with risk of primary liver cancer and the pathological characteristics. Similarly, no association was found between the SNP miR-499 and the risk of primary liver cancer and the clinicopathologic outcome, and, moreover, the age of onset and liver function.

As alteration of miRNAs could lead to changes of protein expression by complementarily binding to protein-coding RNAs, polymorphisms of miRNAs, which can alter the expression and the sequence of miRNAs, are likely to make a significant contribution to phenotypic variation (Georges et al., 2007). Reports showed that G to C variation in the miR-146a precursor could reduce production of mature miRNA (Jazdzewski et al., 2008; Xu et al., 2008). The variant was located in the middle of the stem hairpin, leading to a change from C: U pair (WT) to G: U (MT) mismatch (www.patrocles.org), thus making the secondary structure unstable and thereby altering the expression of mature miR-146a, as reported in previous studies (Zeng and Cullen, 2003; Duan et al., 2007). The G allele of the precursor of miR-146a produced more mature miR-146a in hepatocellular carcinoma and PTC (Jazdzewski et al., 2008; Xu et al., 2008), but less mature miR-146a in familial breast/ovarian cancer (Shen et al., 2008). Jazdzewski et al. (2009) reported that the genotype GC produced two additional mature miRNA from the passenger strands (miR-146a*G and miR-146a*C) of miRNA precursor, which could target different sets of genes. Further, upregulation of miR-146a could result in dramatic loss of Kit transcript and protein, a key tumor suppressor in the thyroid carcinogenesis pathway (He et al., 2005). miR-146a can target TRAF6 and IRAK1, which are key adapter molecules downstream of the Toll-like and cytokine receptors in the signaling pathway that plays a crucial role in cell growth and immune recognition (Jazdzewski et al., 2008). Xu et al. (2008) demonstrated that miR-146a can promote cell proliferation as well as colony formation. Therefore, we hypothesize that the variation of miR-146a affected the age at primary liver cancer onset and liver function via target selection and/or alteration of the expression of miRNA. However, the association between miR-146a SNP and the age of onset and liver function in primary liver cancer may also be coincidental in our study. The number of subjects analyzed was relatively small; replicate studies and further research are required to verify our conclusion.

In addition, our results indicated no significant difference between the distribution of genotypes and alleles of miR-146a SNP and the risk of primary liver cancer, in contrast to Xu et al.'s (2008) report, which showed that the genotype of miR-146a SNP in HCC (hepatocellular carcinoma) cases was significantly different from that in control subjects (p=0.026). More than 90% primary liver cancer is hepatocellular carcinoma; the lack of consistency may be due to the cases we enrolled and differences in samples size and population-specific factors. They enrolled 479 HCC patients and 504 cancer-free controls who lived in Southern China, whereas the population we enrolled lived in Western China, which is a multi-nationality area. We suspect that different allele frequencies of different areas and ethnic groups may contribute to the discordance. Sample size of the present study is smaller, especially when analyzed in subgroups. Therefore, further large multi-center studies with different ethnic groups are warranted to assess the role of the miR-146a SNP in primary liver cancer risk.

Footnotes

Acknowledgments

We thank Dr. Haiyan Chen (Rush University Medical Center) for critical review and editorial assistance during article preparation. The research has been supported by National Natural Science Foundation of China (Grant No. 30900658 and 30901720).

Disclosure Statement

The authors declare that they have no competing interests that are directly relevant to the content of this study.

References

Akao

, Nakagawa

, Naoe

2006. MicroRNAs 143 and 145 are possible common onco-microRNAs in human cancers. Oncol Rep, 16:845–850.

Akao

, Nakagawa

, Naoe

2007. MicroRNA-143 and −145 in colon cancer. DNA Cell Biol, 26:311–320.

Ambros

2003. MicroRNA pathways in flies and worms: growth, death, fat, stress, and timing. Cell, 113:673–676.

Ambros

2004. The functions of animal microRNAs. Nature, 431:350–355.

Bandrés

, Cubedo

, Agirre

et al. 2006. Identification by real-time PCR of 13 mature microRNAs differentially expressed in colorectal cancer and nontumoral tissues. Mol Cancer, 5:29.

Bartel

D.P.

2004. MicroRNAs: genomics, biogenesis, mechanism, and function. Cell, 116:281–297.

Bartels

C.L.

, Tsongalis

G.J.

2009. MicroRNAs: novel biomarkers for human cancer. Clin Chem, 55:623–631.

Cai

, Yu

, Hu

et al. 2009. A brief review on the mechanisms of miRNA regulation. Genomics Proteomics Bioinform, 7:147–154.

Calin

G.A.

, Cimmino

, Fabbri

et al. 2008. MiR-15a and miR-16-1 cluster functions in human leukemia. Proc Natl Acad Sci U S A, 105:5166–5171.

10.

Chan

J.A.

, Krichevsky

A.M.

, Kosik

K.S.

2005. MicroRNA-21 is an antiapoptotic factor in human glioblastoma cells. Cancer Res, 65:6029–6033.

11.

Duan

, Pak

, Jin

et al. 2007. Single nucleotide polymorphism associated with mature miR-125a alters the processing of pri-miRNA. Hum Mol Genet, 16:1124–1131.

12.

El-Serag

H.B.

, Rudolph

K.L.

et al. 2007. Hepatocellular carcinoma: epidemiology and molecular carcinogenesis. Gastroenterology, 132:2557–2576.

13.

Farazi

T.A.

, Spitzer

J.I.

, Morozov

et al. 2011. MiRNA in human cancer. J Pathol, 223:102–115.

14.

Georges

, Coppieters

, Charlie

2007. Polymorphic miRNA-mediated gene regulation: contribution to phenotypic variation and disease. Curr Opin Genet Dev, 17:166–176.

15.

Hagymási

, Tulassay

2008. Epidemiology, risk factors and molecular pathogenesis of primary liver cancer. Orv Hetil, 149:541–548.

16.

, Jazdzewski

, Li

et al. 2005. The role of microRNA genes in papillary thyroid carcinoma. Proc Natl Acad Sci U S A, 102:19075–19080.

17.

, Chen

, Tian

et al. 2008. Genetic variants of miRNA sequences and non-small cell lung cancer survival. J Clin Invest, 118:2600–2608.

18.

Z.B.

, Liang

, Wang

Z.W.

et al. 2009. Common genetic variants in pre-microRNAs were associated with increased risk of breast cancer in Chinese women. Hum Mutat, 30:79–84.

19.

Huang

, He

2011. The role of microRNAs in liver cancer progression. Br J Cancer, 104:235–240.

20.

Iorio

M.V.

, Ferracin

, Liu

C.G.

et al. 2005. MicroRNA gene expression deregulation in human breast cancer. Cancer Res, 65:7065–7070.

21.

Jazdzewski

, Murray

E.L.

, Franssila

et al. 2008. Common SNP in pre-miR-146a decreases mature miR expression and predisposes to papillary thyroid carcinoma. Proc Natl Acad Sci U S A, 105:7269–7274.

22.

Jazdzewski

, Liyanarachchia

, Swierniakb

et al. 2009. Polymorphic mature microRNAs from passenger strand of pre-miR-146a contribute to thyroid cancer. Proc Natl Acad Sci U S A, 106:1502–1505.

23.

Jemal

, Bray

, Center

M.M.

et al. 2011. Global cancer statistics. CA Cancer J Clin, 61:69–90.

24.

Johnson

C.D.

, Esquela-Kerscher

, Stefani

et al. 2007. The let-7 microRNA represses cell proliferation pathways in human cells. Cancer Res, 67:7713–7722.

25.

X.D.

, Li

Z.G.

, Song

X.X.

et al. 2010a. A variant in microRNA-196a2 is associated with susceptibility to hepatocellular carcinoma in Chinese patients with cirrhosis. Pathology, 42:669–673.

26.

, Vandenboom

T.G.

II , Wang

et al. 2010b. miR-146a suppresses invasion of pancreatic cancer cells. Cancer Res, 70:1486–1495.

27.

Ludwig

J.A.

, Weinstein

J.N.

2005. Biomarkers in cancer staging, prognosis and treatment selection. Nat Rev Cancer, 5:845–856.

28.

Medina

P.P.

, Nolde

, Slack

F.J.

2010. OncomiR addiction in an in vivo model of microRNA-21-induced pre-B-cell lymphoma. Nature, 467:86–90.

29.

Pallante

, Visone

, Ferracin

et al. 2006. MicroRNA deregulation in human thyroid papillary carcinomas. Endocr Relat Cancer, 13:497–508.

30.

Pastrello

, Polesel

, Della

P.L.

et al. 2010. Association between hsa-mir-146a genotype and tumor age-of-onset in BRCA1/BRCA2-negative familial breast and ovarian cancer patients. Carcinogenesis, 31:2124–2126.

31.

Peng

, Kuang

Z.S.

, Sheng

C.Y.

et al. 2009. Association of microRNA-196a-2 gene polymorphism with gastric cancer risk in a Chinese population. Dig Dis Sci, 55:2288–2293.

32.

, Dou

T.H.

, Geng

et al. 2010. Association of a variant in MIR 196A2 with susceptibility to hepatocellular carcinoma in male Chinese patients with chronic hepatitis B virus infection. Hum Immunol, 71:621–626.

33.

Shen

, Ambrosone

C.B.

, DiCioccio1

R.A.

et al. 2008. A functional polymorphism in the miR-146a gene and age of familial breast/ovarian cancer diagnosis. Carcinogenesis, 29:1963–1966.

34.

Thorgeirsson

S.S.

, Grisham

J.W.

2002. Molecular pathogenesis of human hepatocellular carcinoma. Nat Genet, 31:339–346.

35.

Volinia

, Calin

G.A.

, Liu

C.G.

et al. 2006. A microRNA expression signature of human solid tumors defines cancer gene targets. Proc Natl Acad Sci U S A, 103:2257–2261.

36.

Wang

J.X.

, Jiao

J.Q.

, Li

et al. 2011. miR-499 regulates mitochondrial dynamics by targeting calcineurin and dynamin-related protein-1. Nat Med, 17:71–78.

37.

Wilson

K.D.

, Hu

, Venkatasubrahmanyam

et al. 2010. Dynamic microRNA expression programs during cardiac differentiation of human embryonic stem cells: role for miR-499. Circ Cardiovasc Genet, 3:426–435.

38.

, Zhu

, Wei

Q.K.

et al. 2008. A functional polymorphism in the miR-146a gene is associated with the risk for hepatocellular carcinoma. Carcinogenesis, 29:2126–2131.

39.

, Liu

et al. 2011. A potentially functional polymorphism in the promoter region of miR-34b/c is associated with an increased risk for primary hepatocellular carcinoma. Int J Cancer, 128:412–417.

40.

Yanaihara

, Caplen

, Bowman

et al. 2006. Unique microRNA molecular profiles in lung cancer diagnosis and prognosis. Cancer Cell, 9:189–198.

41.

Yuen

M.F.

, Hou

J.L.

, Chutaputti

2009. Hepatocellular carcinoma in the Asia pacific region. J Gastroenterol Hepatol, 24:346–353.

42.

Zamore

P.D.

, Haley

2005. Ribo-gnome: the big world of small RNAs. Science, 309:1519–1524.

43.

Zender

, Villanueva

, Tovar

et al. 2010. Cancer gene discovery in hepatocellular carcinoma. J Hepatol, 52:921–929.

44.

Zeng

, Cullen

B.R.

2003. Sequence requirements for microRNA processing and function in human cells. RNA, 9:112–123.

45.

Zeng

, Sun

Q.M.

, Liu

N.N.

et al. 2010. Correlation between pre-miR-146a C/G polymorphism and gastric cancer risk in Chinese population. World J Gastroenterol, 16:3578–83.

46.

Zhou

, Rao

et al. 2010. Common genetic polymorphisms in pre-microRNAs were associated with increased risk of dilated cardiomyopathy. Clin Chim Acta, 411:1287–1290.