Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

A Meta-Analysis of MicroRNA Expression in Liver Cancer

  • Jingcheng Yang,

    Affiliation First Affiliated Hospital of Guangxi Medical University, Nanning, Guangxi Zhuang Autonomous Region, China

  • Shuai Han,

    Affiliation First Affiliated Hospital of Guangxi Medical University, Nanning, Guangxi Zhuang Autonomous Region, China

  • Wenwen Huang,

    Affiliation First Affiliated Hospital of Guangxi Medical University, Nanning, Guangxi Zhuang Autonomous Region, China

  • Ting Chen,

    Affiliation Department of Management Information System, College of Computer and Information Engineering, Guangxi Teachers Education University, Nanning, Guangxi Zhuang Autonomous Region, China

  • Yang Liu,

    Affiliation First Affiliated Hospital of Guangxi Medical University, Nanning, Guangxi Zhuang Autonomous Region, China

  • Shangling Pan,

    Affiliation Department of Pathophysiology, Guangxi Medical University, Nanning, Guangxi Zhuang Autonomous Region, China

  • Shikang Li

    shikangli@hotmail.com

    Affiliation First Affiliated Hospital of Guangxi Medical University, Nanning, Guangxi Zhuang Autonomous Region, China

A Meta-Analysis of MicroRNA Expression in Liver Cancer

  • Jingcheng Yang, 
  • Shuai Han, 
  • Wenwen Huang, 
  • Ting Chen, 
  • Yang Liu, 
  • Shangling Pan, 
  • Shikang Li
PLOS
x

Abstract

MicroRNA (miRNA) played an important role in the progression of liver cancer and its diagnostic and prognostic values have been frequently studied. However, different microarray techniques and small sample size led to inconsistent findings in previous studies. We performed a comprehensive meta-analysis of a total of 357 tumor and 283 noncancerous samples from 12 published miRNA expression studies using robust rank aggregation method. As a result, we identified a statistically significant meta-signature of five upregulated (miR-221, miR-222, miR-93, miR-21 and miR-224) and four downregulated (miR-130a, miR-195, miR-199a and miR-375) miRNAs. We then conducted miRNA target prediction and pathway enrichment analysis to find what biological process these miRNAs might affect. We found that most of the pathways were frequently associated with cell signaling and cancer pathogenesis. Thus these miRNAs may involve in the onset and progression of liver cancer and serve as potential diagnostic and therapeutic targets of this malignancy.

Introduction

Liver cancer in men is the second most frequent cause of cancer death and is the sixth leading cause of cancer death in women, which causes approximately 700 thousand deaths per year with about 750 thousand new cases diagnosed worldwide. [1] Low survival is attributed to late diagnosis, resistance to chemotherapy, tumor recurrence, and metastasis, hence stressing the need for novel diagnostics and therapeutics. Numerous gene expression studies have shown that a general aberrant activation of signaling pathways was attributed to the oncogenicity, however, one signature or a single prominent characteristic pathway could not be defined in liver cancer. [2]

MicroRNAs (miRNAs), small non-coding RNAs of 18–25 nucleotides in length, which were seen to control gene expression in virtually all cancer cells, were abundantly investigated relating to the occurrence, progress, classification, diagnosis and treatment of tumors recently. The involvement of miRNAs in cancer pathogenesis is well established, as they can behave as oncogenes or tumor suppressor genes depending on the cellular function of their targets. [3] Understanding the biology of miRNA and its contribution to cancer development may promise early diagnosis and effective control of malignant tumors.

Recent findings from integrative and mechanism-based profiling studies have provided important information about the roles of miRNAs in normal cells and disease condition. These studies could improve our understanding of the molecular mechanisms of chronic liver diseases and liver cancer. Numerous studies linking to the deregulation of miRNA expression to liver cancer have been reported with multifarious methods. [4] However, due to the application of different technological platforms and small sample size, the miRNA expression profiling efforts have led to inconsistent results between the studies.

To overcome the limitations in current researches, we performed a meta-analysis applying the robust rank aggregation method, [5] followed by pathway analysis, to identify miRNA deregulation in liver cancer and the pathways that key miRNAs may impact. The leave-one-out cross-validation method was used to validate the results. Identification of miRNA meta-signature and invovled pathways would provide potential targets for further experimental studies of liver cancer development.

Material and Method

Study selection and data extraction

A systematic literature search was performed for the identification of liver cancer miRNA expression profiling studies using a two-level search strategy. First, we undertook a web-based search in Gene Expression Omnibus (GEO, www.ncbi.nlm.-nih.gov/geo/) using search term (("Neoplasms"[Mesh] AND "Liver"[Mesh]) AND "MicroRNAs"[Mesh]) AND "Humans"[Mesh]. To perform a comprehensive retrieval, searching in ArrayExpress (www.ebi.ac.uk/arrayexpress), the Pubmed and the Embase database were also performed. Second, the reference lists of all relevant and existing studies were reviewed through a manual search for further identification of potential relevant studies.

Abstracts were screened carefully and full texts of relevant potential abstracts were evaluated. Studies with original experimental design that analyzed the miRNA expression profiling in human between liver cancer tissues and non-tumorous liver tissues were included. Meanwhile, studies were not eligible for meta-analysis if they met the following selection criteria: 1) using only cell lines, 2) preselected candidate genes research, 3) profiling different clinical or histologic subtypes without including non-tumorous tissues.

Lists of statistically significant expressed miRNAs were extracted from publications. Authors were contacted when the lists could not be obtained. All miRNA names were standardized through miRBase version 20. MiRNAs that cannot be related to either -3p or -5p in miRBase were designated with hsa-miR-*, such as hsa-miR-210. Those identified as dead entry in miRBase were remained their names mentioned in the literatures.

Statistical analysis

The lists of miRNAs were extracted based on statistical test p-values (<0.05 was considered significant). Then, miRNAs in every list were prioritized by fold changes or other values that could indicate the degree of deregulation. To ensure that extracted miRNAs can be ranked in a more reliable way, we used robust rank aggregation method. [5] The method is based on the comparison of actual data with a null model that assumes random order of input lists. A P-value assigned to each element in the aggregated list described how much better it was ranked than expected. In case of false positive results, Bonferroni correction was performed. Meanwhile, to assess the stability of acquired p-values, leave one out cross-validation was applied on the robust rank aggregation algorithm. An averaged p-value was obtained from random gene lists after repeating the analyses 10,000 times.

Clustering analysis

To investigate the correlations among the miRNA expression profiles of individual studies, we performed hierarchical clustering using the deregulated miRNAs. Two-dimensional average-linkage hierarchical clustering of a Spearman rank correlation similarity matrix constructed from separate analyses for upregulated and downregulated gene lists was performed.

Integrative identification of miRNA targets

The meta-signature miRNAs were selected for target prediction by using TargetScan database version 6.2 [6], PicTar (predictions from miRWalk database) [7] and DIANA-microT-CDS v5.0 (using miTG score threshold 0.7) [8]. TarBase v6.0database [9] and CLIP-Seq database starBase [10] were used to acquire validated targets. To improve the accuracy of the target prediction, consensus targets were extracted for the overlapping targets predicted by at least two algorithms plus validated targets from TarBase and starBase.

Enrichment analysis

To identify the pathways of predicted miRNA targets, Kyoto Encyclopedia of Genes and Genomes (KEGG), Panther pathways and Gene Ontology terms were carried out with GeneCodis web tool (http://genecodis.dacya.ucm.es/). [11]

Results

Study selection and data extraction

Database searches initially yielded a total of 251 publications and 16 studies met the inclusion criteria. (Fig. 1) Four researches [12][15] were excluded in the final analysis because the lists of ranked miRNA were not neither publicly nor personally available from the corresponding authors. Most of the studies were published between 2009 and 2013. Two-thirds of the researches came from Asian region and the remainings were come from America and Europe. Moreover, included studies used various microarray platforms and the average number of miRNA probes was 626 (ranging from 121 to 1205). A total of 357 tumor and 283 noncancerous samples were included. The majority of the studies focused on hepatocellular carcinoma (HCC) compared with adjacent nontumorous tissues or normal liver tissues except Cario 2010 and Selaru 2009, which respectively focused on hepatoblastoma (HB) and cholangiocarcinoma (CCA). The main characteristics of these studies were listed in Table 1.

In total, 136 miRNAs were reported as significantly upregulated and 138 as significantly downregulated in included studies. The number of significantly deregulated miRNAs varied greatly across studies (ranging from 14 to 91).

Cluster analysis

To assess the degree of concordance between miRNA lists and possible correlation according to the subgroups of tumor histology, region, and sample size, hierarchical clustering analysis was performed (S1 Figure). The clustering of these lists showed that the results of Shih 2012, Sato 2011 and Li 2008 using the HCC tissues were more similar to each other than any other studies. The most similar results were Li 2008-1 and Li 2008-2, conducted by the same workgroup using the same platform though the sample size differs greatly. It is not clear whether there was some overlap of the samples between the two studies, but if this is the case, it may be one explanation for the high similarity of the results. No more obvious similarities were seen between other subgroups.

MiRNA meta-signature

We identified a statistically significant meta-signature of five upregulated miRNAs and four downregulated miRNAs in liver cancer samples compared to noncancerous liver tissue according to the permutation p-value. (Table 2) Only two upregulated but not downregulated miRNAs reached statistical significance after Bonferroni correction. The number of meta-signature miRNAs studies reported varied greatly but at least three deregulated miRNAs were reported by each study, with an exception of Cairo 2010 and Selaru 2009, which just separately reported two upregulated miRNAs. The most significantly deregulated miRNAs, miR-221, miR-222, are respectively reported by nine and ten datasets. Furthermore, the permutation p-values of another three upregulated miRNAs, miR-93, miR-21 and miR-224, and four downregulated miRNAs, miR-130a, miR-195, miR-199a and miR 375 are <0.05, but do not reach the corrected significance.

Meta-signature miRNA genes are scattered on different chromosomal locations with an exception of miR-221 and miR-222 genes, which are both located on Xp11.3. MiR-224 genes are located in the same chromosomal region, Xq28, which is a cytogenetic region that contains a large portion of the cancer/testis antigen gene family. [16]

Target prediction and enrichment analysis

Consensus targets for miRNAs reaching robust rank aggregation significance were extracted for the overlapping targets predicted by at least two algorithms plus experimentally validated targets from two databases. The summery of target counts is presented in Fig. 2. MiR-130a and miR-195 have more targets than other miRNAs, whereas miR-199a has no targets because it was predicted by only one algorithm.

Enrichment analyses were performed using predicted target genes with GeneCodis web tool. Several pathways enriched by KEGG and Panther pathways were relatively significant and most of them were frequently associated with cell signaling (e.g. neurotrophin, Wnt, FGF, and p53 signaling pathway) and cancer. (Table 3, Fig. 3, S2 Figure)

thumbnail
Figure 3. Pathway enrichment of meta-signature miRNA targets.

The intensity of color represents the FDR-corrected p-value. Only those pathways, which were significant for more than four miRNAs are shown (full data are available as S2 Figure).

https://doi.org/10.1371/journal.pone.0114533.g003

thumbnail
Table 3. GO processes and pathways most strongly enriched by meta-signature miRNA targets.

https://doi.org/10.1371/journal.pone.0114533.t003

Discussion

Using robust rank aggregation method, 14 prioritized miRNA lists from the 12 published studies were analyzed and finally identified five upregulated and four downregulated miRNAs. But after Bonferroni correction only two upregulated miRNAs reached the statistical difference.

The deregulated miRNAs identified by different research centers did not allow a consistent conclusion. Differences among microarray techniques, stage of the tumor, histologic appearance and etiological factors attributed to the heterogeneity. [17] There have been attempts to divide the datasets into subgroups on the basis of platform and tumor-type subtype, but none of two studies have used same platform and the majority of tumor samples were from HCC tissues. However, the cluster analysis demonstrated that two datasets conducted by one work group used an identical platform made high similarity and anther meta-analysis used robust rank aggregation method also led to the same conclusion as us. [18]

In this study, we used the Bonferroni method to control the false positive rate and made the results more reliable. Ultimately, miR-221 and miR-222 were only two statistically significant meta-signature miRNAs. Furthermore, the permutation p-values of another three upregulated (miR-93, miR-21 and miR-224) and four downregulated miRNAs (miR-130a, miR-195, miR-199a and miR-375) were <0.05, but their corrected p-values were not significant. The following limitations may explain these finding: 1) there were not sufficient datasets for integration, 2) the sample sizes of the datasets were relatively small, 3) different methodology researchers used made more discrepant.

Although no strong significance was seen in all meta-signature miRNAs, experimental studies in recent years demonstrated that they were highly associated with liver cancer. MiR-221 was the most frequently studied among the meta-signature miRNAs. Fornari et al. [19] have proved that miR-221 functioned as an oncogene in hepatocarcinogenesis by targeting CDKN1B/p27 and CDKN1C/p57 in 2008. Meanwhile, results of Gramantieri et al. [20] have indicated that miR-221 inhibited apoptosis by targeting Bmf and its overexpression was associated with a more aggressive phenotype in 2009. Together, these findings implicated that miR-221 might be a potential target for nonconventional treatment against HCC. In addition, Park et al. [21] found that miR-221 silencing blocked hepatocellular carcinoma and promoted survival in a valid orthotopic mouse model of HCC and Callegari et al. [22] and Pineau et al. [23] have respectively confirmed the oncogenic role of miR-221 in the mice model. Moreover, Li et al. [24] demonstrated that serum miR-221 might provide predictive significance for prognosis of HCC patients and He et al. [25] have proved that miR-221 silencing inhibited liver cancer malignant properties in vitro and in vivo. Together, miR-221 has been deeply studied on its impact on liver cancer, and it might be a potential candidate target for liver cancer diagnosis and therapy.

Besides miR-221, some other miRNAs have been experimentally proved to associate with the occurrence, development, and metastasis. Three upregulated meta-signature miRNAs, miR-222, miR-21 and miR-224, might exacerbate HCC through AKT signaling pathways. [26]-[28] MiR-222 overexpression is common in HCC and could confer metastatic potentials in HCC cells possibly by enhancing AKT signaling. [26] MiR-21 suppresses PTEN and hSulf-1 expression and promotes HCC progression through AKT/ERK pathways, and as for miR-224, it possibly actives the AKT signaling pathways by targeting PPP2R1B. [27], [28] A recent study showed that miR-21 and miR-222 expressions were differentially modulated by Hepatitis B Virus X protein in malignant hepatocytes. [29] In addition, miR-224 might play significant role in migration and invasion of HCC cell. [30]-[32] Interestingly, two downregulated meta-signature miRNAs, miR-195 and miR-375, might plays important inhibitory roles in HCC progression. [33][38] For example, miR-195 might inhibit HCC progression by targeting LATS2 and the NF-κB signaling pathway [33], [34] and suppress angiogenesis and metastasis of hepatocellular carcinoma by inhibiting the expression of VEGF, VAV2, and CDC42. [35] Xu et al. have proved that miR-195 played an important role in cell cycle control and in the molecular etiology of HCC. [36] Moreover, miR-195 may target PCMT1 in hepatocellular carcinoma that increases tumor life span [39] and negatively regulate protein levels of steroid receptor coactivator-3 through targeting its 3'-untranslated region in HCC cells [40]. Another downregulated miRNA, miR-375, also has been proved to suppress liver cancer cell growth in vitro or in vivo. [37], [38]

Although we have demonstrated that miRNAs may involve in promoting or inhibiting liver cancer progression by targeting some genes in the key pathways of cancer regulation, there is still a long way to go in interpreting the impact of miRNA on liver cancer. Future work should keep continuous focus on the mechanism through which miRNAs regulate occurrence, progression and metastasis of liver cancer. In addition, studies with large sample size and the same platform are needed.

In conclusion, we suggest that the meta-signature miRNAs are key regulatory drivers of the oncogenic process, which may be good potential targets for the diagnosis and therapy of liver cancer.

Supporting Information

S1 Figure.

Cluster analysis of miRNA list. Possible correlation was shown according to the subgroups of tumor histology, region, and sample size. Clustering was performed using Pearson correlation and average linkage method.

https://doi.org/10.1371/journal.pone.0114533.s001

(TIF)

S2 Figure.

Pathway enrichment of meta-signature miRNA targets. The intensity of color represents the FDR-corrected p-value. Clustering was performed using Pearson correlation and average linkage method.

https://doi.org/10.1371/journal.pone.0114533.s002

(TIF)

Acknowledgments

We thank Mr. Hui Chen and Ms. Shanshan Zeng for their support on workroom and retrieval network.

Author Contributions

Conceived and designed the experiments: JCY SH WWH. Analyzed the data: JCY SH WWH. Contributed reagents/materials/analysis tools: TC YL. Wrote the paper: SH SLP SKL.

References

  1. 1. Jemal A, Bray F, Center MM, Ferlay J, Ward E, et al. (2011) Global cancer statistics. CA Cancer J Clin 61:69–90.
  2. 2. Giordano S, Columbano A (2013) MicroRNAs: new tools for diagnosis, prognosis, and therapy in hepatocellular carcinoma? Hepatology 57:840–847.
  3. 3. Lujambio A, Lowe SW (2012) The microcosmos of cancer. Nature 482:347–355.
  4. 4. Borel F, Konstantinova P, Jansen PL (2012) Diagnostic and therapeutic potential of miRNA signatures in patients with hepatocellular carcinoma. J Hepatol 56:1371–1383.
  5. 5. Kolde R, Laur S, Adler P, Vilo J (2012) Robust rank aggregation for gene list integration and meta-analysis. Bioinformatics 28:573–580.
  6. 6. Grimson A, Farh KK, Johnston WK, Garrett-Engele P, Lim LP, et al. (2007) MicroRNA targeting specificity in mammals: determinants beyond seed pairing. Mol Cell 27:91–105.
  7. 7. Dweep H, Sticht C, Pandey P, Gretz N (2011) miRWalk—database: prediction of possible miRNA binding sites by "walking" the genes of three genomes. J Biomed Inform 44:839–847.
  8. 8. Maragkakis M, Alexiou P, Papadopoulos GL, Reczko M, Dalamagas T, et al. (2009) Accurate microRNA target prediction correlates with protein repression levels. BMC Bioinformatics 10:295.
  9. 9. Vergoulis T, Vlachos IS, Alexiou P, Georgakilas G, Maragkakis M, et al. (2012) TarBase 6.0: capturing the exponential growth of miRNA targets with experimental support. Nucleic Acids Res 40:D222–229.
  10. 10. Yang JH, Li JH, Shao P, Zhou H, Chen YQ, et al. (2011) starBase: a database for exploring microRNA-mRNA interaction maps from Argonaute CLIP-Seq and Degradome-Seq data. Nucleic Acids Res 39:D202–209.
  11. 11. Nogales-Cadenas R, Carmona-Saez P, Vazquez M, Vicente C, Yang X, et al. (2009) GeneCodis: interpreting gene lists through enrichment analysis and integration of diverse biological information. Nucleic Acids Res 37:W317–322.
  12. 12. Noh JH, Chang YG, Kim MG, Jung KH, Kim JK, et al. (2013) MiR-145 functions as a tumor suppressor by directly targeting histone deacetylase 2 in liver cancer. Cancer Lett 335:455–462.
  13. 13. Navon R, Wang H, Steinfeld I, Tsalenko A, Ben-Dor A, et al. (2009) Novel rank-based statistical methods reveal microRNAs with differential expression in multiple cancer types. PLoS One 4:e8003.
  14. 14. Burchard J, Zhang C, Liu AM, Poon RT, Lee NP, et al. (2010) microRNA-122 as a regulator of mitochondrial metabolic gene network in hepatocellular carcinoma. Mol Syst Biol 6:402.
  15. 15. Oishi N, Kumar MR, Roessler S, Ji J, Forgues M, et al. (2012) Transcriptomic profiling reveals hepatic stem-like gene signatures and interplay of miR-200c and epithelial-mesenchymal transition in intrahepatic cholangiocarcinoma. Hepatology 56:1792–1803.
  16. 16. Wallden B, Emond M, Swift ME, Disis ML, Swisshelm K (2005) Antimetastatic gene expression profiles mediated by retinoic acid receptor beta 2 in MDA-MB-435 breast cancer cells. BMC Cancer 5:140.
  17. 17. Diaz G, Melis M, Tice A, Kleiner DE, Mishra L, et al. (2013) Identification of microRNAs specifically expressed in hepatitis C virus-associated hepatocellular carcinoma. Int J Cancer 133:816–824.
  18. 18. Vosa U, Vooder T, Kolde R, Vilo J, Metspalu A, et al. (2013) Meta-analysis of microRNA expression in lung cancer. Int J Cancer 132:2884–2893.
  19. 19. Fornari F, Gramantieri L, Ferracin M, Veronese A, Sabbioni S, et al. (2008) MiR-221 controls CDKN1C/p57 and CDKN1B/p27 expression in human hepatocellular carcinoma. Oncogene 27:5651–5661.
  20. 20. Gramantieri L, Fornari F, Ferracin M, Veronese A, Sabbioni S, et al. (2009) MicroRNA-221 targets Bmf in hepatocellular carcinoma and correlates with tumor multifocality. Clin Cancer Res 15:5073–5081.
  21. 21. Park JK, Kogure T, Nuovo GJ, Jiang J, He L, et al. (2011) miR-221 silencing blocks hepatocellular carcinoma and promotes survival. Cancer Res 71:7608–7616.
  22. 22. Callegari E, Elamin BK, Giannone F, Milazzo M, Altavilla G, et al. (2012) Liver tumorigenicity promoted by microRNA-221 in a mouse transgenic model. Hepatology 56:1025–1033.
  23. 23. Pineau P, Volinia S, McJunkin K, Marchio A, Battiston C, et al. (2010) miR-221 overexpression contributes to liver tumorigenesis. Proc Natl Acad Sci U S A 107:264–269.
  24. 24. Li J, Wang Y, Yu W, Chen J, Luo J (2011) Expression of serum miR-221 in human hepatocellular carcinoma and its prognostic significance. Biochem Biophys Res Commun 406:70–73.
  25. 25. He XX, Guo AY, Xu CR, Chang Y, Xiang GY, et al. (2014) Bioinformatics analysis identifies miR-221 as a core regulator in hepatocellular carcinoma and its silencing suppresses tumor properties. Oncol Rep 32:1200–1210.
  26. 26. Wong QW, Ching AK, Chan AW, Choy KW, To KF, et al. (2010) MiR-222 overexpression confers cell migratory advantages in hepatocellular carcinoma through enhancing AKT signaling. Clin Cancer Res 16:867–875.
  27. 27. Bao L, Yan Y, Xu C, Ji W, Shen S, et al. (2013) MicroRNA-21 suppresses PTEN and hSulf-1 expression and promotes hepatocellular carcinoma progression through AKT/ERK pathways. Cancer Lett 337:226–236.
  28. 28. Ma D, Tao X, Gao F, Fan C, Wu D (2012) miR-224 functions as an onco-miRNA in hepatocellular carcinoma cells by activating AKT signaling. Oncol Lett 4:483–488.
  29. 29. Bandopadhyay M, Banerjee A, Sarkar N, Panigrahi R, Datta S, et al. (2014) Tumor suppressor micro RNA miR-145 and onco micro RNAs miR-21 and miR-222 expressions are differentially modulated by hepatitis B virus X protein in malignant hepatocytes. BMC Cancer 14:721.
  30. 30. Li Q, Ding C, Chen C, Zhang Z, Xiao H, et al. (2014) miR-224 promotion of cell migration and invasion by targeting Homeobox D 10 gene in human hepatocellular carcinoma. J Gastroenterol Hepatol 29:835–842.
  31. 31. Zhang Y, Takahashi S, Tasaka A, Yoshima T, Ochi H, et al. (2013) Involvement of microRNA-224 in cell proliferation, migration, invasion, and anti-apoptosis in hepatocellular carcinoma. J Gastroenterol Hepatol 28:565–575.
  32. 32. Yu L, Zhang J, Guo X, Li Z, Zhang P (2014) MicroRNA-224 upregulation and AKT activation synergistically predict poor prognosis in patients with hepatocellular carcinoma. Cancer Epidemiol 38:408–413.
  33. 33. Yang X, Yu J, Yin J, Xiang Q, Tang H, et al. (2012) MiR-195 regulates cell apoptosis of human hepatocellular carcinoma cells by targeting LATS2. Pharmazie 67:645–651.
  34. 34. Ding J, Huang S, Wang Y, Tian Q, Zha R, et al. (2013) Genome-wide screening reveals that miR-195 targets the TNF-alpha/NF-kappaB pathway by down-regulating IkappaB kinase alpha and TAB3 in hepatocellular carcinoma. Hepatology 58:654–666.
  35. 35. Wang R, Zhao N, Li S, Fang JH, Chen MX, et al. (2013) MicroRNA-195 suppresses angiogenesis and metastasis of hepatocellular carcinoma by inhibiting the expression of VEGF, VAV2, and CDC42. Hepatology 58:642–653.
  36. 36. Xu T, Zhu Y, Xiong Y, Ge YY, Yun JP, et al. (2009) MicroRNA-195 suppresses tumorigenicity and regulates G1/S transition of human hepatocellular carcinoma cells. Hepatology 50:113–121.
  37. 37. Chang Y, Yan W, He X, Zhang L, Li C, et al. (2012) miR-375 inhibits autophagy and reduces viability of hepatocellular carcinoma cells under hypoxic conditions. Gastroenterology 143:177–187.e178.
  38. 38. He XX, Chang Y, Meng FY, Wang MY, Xie QH, et al. (2012) MicroRNA-375 targets AEG-1 in hepatocellular carcinoma and suppresses liver cancer cell growth in vitro and in vivo. Oncogene 31:3357–3369.
  39. 39. Amer M, Elhefnawi M, El-Ahwany E, Awad AF, Gawad NA, et al. (2014) Hsa-miR-195 targets PCMT1 in hepatocellular carcinoma that increases tumor life span. Tumour Biol: ahead of print.
  40. 40. Jiang HL, Yu H, Ma X, Xu D, Lin GF, et al. (2014) MicroRNA-195 regulates steroid receptor coactivator-3 protein expression in hepatocellular carcinoma cells. Tumour Biol 35:6955–6960.
  41. 41. Cairo S, Wang Y, de Reynies A, Duroure K, Dahan J, et al. (2010) Stem cell-like micro-RNA signature driven by Myc in aggressive liver cancer. Proc Natl Acad Sci U S A 107:20471–20476.
  42. 42. Chung GE, Yoon JH, Myung SJ, Lee JH, Lee SH, et al. (2010) High expression of microRNA-15b predicts a low risk of tumor recurrence following curative resection of hepatocellular carcinoma. Oncol Rep 23:113–119.
  43. 43. Elyakim E, Sitbon E, Faerman A, Tabak S, Montia E, et al. (2010) hsa-miR-191 is a candidate oncogene target for hepatocellular carcinoma therapy. Cancer Res 70:8077–8087.
  44. 44. Koh YS, Kim JH, Cai H, Li LH, Kim HS, et al. (2013) Dysregulated microRNAs in non-cirrhotic hepatocellular carcinoma. Genes & Genomics 35:759–765.
  45. 45. Li W, Xie L, He X, Li J, Tu K, et al. (2008) Diagnostic and prognostic implications of microRNAs in human hepatocellular carcinoma. Int J Cancer 123:1616–1622.
  46. 46. Meng F, Henson R, Wehbe-Janek H, Ghoshal K, Jacob ST, et al. (2007) MicroRNA-21 regulates expression of the PTEN tumor suppressor gene in human hepatocellular cancer. Gastroenterology 133:647–658.
  47. 47. Sato F, Hatano E, Kitamura K, Myomoto A, Fujiwara T, et al. (2011) MicroRNA profile predicts recurrence after resection in patients with hepatocellular carcinoma within the Milan Criteria. PLoS One 6:e16435.
  48. 48. Selaru FM, Olaru AV, Kan T, David S, Cheng Y, et al. (2009) MicroRNA-21 is overexpressed in human cholangiocarcinoma and regulates programmed cell death 4 and tissue inhibitor of metalloproteinase 3. Hepatology 49:1595–1601.
  49. 49. Shih TC, Tien YJ, Wen CJ, Yeh TS, Yu MC, et al. (2012) MicroRNA-214 downregulation contributes to tumor angiogenesis by inducing secretion of the hepatoma-derived growth factor in human hepatoma. J Hepatol 57:584–591.
  50. 50. Su H, Yang JR, Xu T, Huang J, Xu L, et al. (2009) MicroRNA-101, down-regulated in hepatocellular carcinoma, promotes apoptosis and suppresses tumorigenicity. Cancer Res 69:1135–1142.
  51. 51. Yang L, Ma Z, Wang D, Zhao W, Chen L, et al. (2010) MicroRNA-602 regulating tumor suppressive gene RASSF1A is overexpressed in hepatitis B virus-infected liver and hepatocellular carcinoma. Cancer Biol Ther 9:803–808.