Two recent conceptual advances have led to an exciting renaissance in genetic regulation of biological processes involved in cell growth, development, and function. The first was the recognition that many portions of the human genome that do not encode conventional protein-coding genes can encode biologically active RNA species. The second was the identification of mechanisms by which small RNA molecules could regulate gene and protein expression. Consequently, there has been an increasing recognition of the role of genetically encoded regulatory RNA molecules in biological processes. These advances have resulted in several new hypotheses on the mechanistic role of noncoding RNA in physiological processes, and their deregulation in disease. In turn, these hypotheses are beginning to provide new insights into disease classifications and pathogenesis.

Several types of regulatory RNA have been identified in the vast portion of the human genome that is “noncoding”. These include microRNAs (miRNAs), a group which is recognized based on their structural characteristic of a hairpin loop, and which have been shown to modulate messenger RNA (mRNA) expression in plants and animals. Mature miRNAs result from the sequential processing of primary transcripts by two ribonuclease III enzymes that act sequentially, Drosha in the nucleus and Dicer in the cytoplasm. Mature miRNAs can negatively regulate protein expression either by translational inhibition or by mRNA degradation of specific mRNA. One mechanism involves the incorporation of one strand of the mature miRNAs into an effector complex called RNA-induced silencing complex (RISC), while the other strand is eliminated. Perfect or nearly perfect complementarity between a miRNA and its target 3′ untranslated region induces RISC to cleave the target mRNA, whereas imperfect matching induces mainly translational silencing of the target, but can also reduce the amount of the mRNA target.

Alterations in miRNA expression can modulate key cellular processes involved in tumor formation such as cell differentiation, proliferation, and apoptosis. As a result, systematic evaluation of changes in global and individual miRNA expression in phenotypically defined tumors could provide insight into basic mechanisms of tumorigenesis. Such studies have been performed in many human cancers. They have led to the identification of aberrantly expressed miRNA which can function as either tumor suppressors or oncogenes and thus can be directly implicated in tumor formation. The changes observed in miRNA expression show considerable specificity for the tissue type, and this offers the possibility that they may be useful for diagnostic purposes. Indeed, the informative nature of miRNA expression for tissue type of tumor origin has been shown to be even more useful than mRNA-based profiling.1

Several studies have reported miRNA expression profiling in human hepatocellular cancer (HCC) cells and tissues and in experimental models of HCC.2–11 Consistently, these studies have all shown that specific miRNAs are aberrantly expressed in malignant HCC cells or tissues compared to nonmalignant hepatocytes or tissue. Thus, these miRNAs may provide insights into cellular processes involved in malignant transformation or behavior, or they may represent markers of malignancy. Among the miRNAs that are aberrantly increased in HCC are miR-21, miR-34a, miR-221/222, and miR-224. Likewise, certain miRNAs have been noted to be decreased in HCC compared to nontumoral tissue. These include miRNAs such as miR-122, miR-145, and mir-199a. Murakami et al. showed a correlation between miR-222, miR-106a, miR-92, miR-17-5p, miR-20, and miR-18 and the degree of differentiation suggested an involvement of specific miRNAs in the progression of the disease.5 Similarly, Budhu et al. highlighted a 20-miRNA signature that is associated with metastatic HCC.7 The miRNA expression profiles in malignant hepatocytes differ from those of malignant cholangiocytes12 (Table 1). These studies indicate that miRNA profiling studies could be used for defining clinical phenotypes, as well as potentially useful molecular diagnostic markers.

Table 1. Selected miRNAs that are Aberrantly Expressed in Human Liver Tumors
Hepatocellular AdenomaHepatocellular cancerCholangiocarcinoma
  1. Data is summarized from reported studies of miRNA profiling from hepatocellular adenoma or carcinoma relative to nontumoral controls, or from cholangiocarcinoma cell lines relative to nonmalignant cholangiocytes.2, 4, 5, 9–13


In this issue, Ladeiro et al. provide the first report of miRNA expression in benign liver cancers and add to the literature on miRNA expression profiling in hepatocellular neoplasms.13 Several miRNAs that are associated with either benign or malignant hepatocellular tumors are identified. These carefully validated studies show that selected miRNAs such as miR-21, miR-224, miR-106a, and miR-203 are up-regulated in HCC compared to benign hepatocellular tumors such as adenomas or focal nodular hyperplasia. Increased expression of miR-10b and miR-222 with decreased expression of miR-122a and miR-422b was also identified in HCC relative to nontumoral liver. Interestingly, the altered expression of some miRNAs was associated with distinctive risk factors, such as miR-96 with hepatitis B virus infection and miR-126* with alcohol use. Further studies will be needed to evaluate whether or not these findings are generalizable. Clearly these findings need to be further explored and integrated with studies to validate specific targets of these miRNAs and to define their potential contribution to etiologically distinct cellular pathways of tumorigenesis.

An additional observation is that deregulation of expression of specific miRNAs was associated with specific gene mutations. One of these, miR-375, is decreased in expression in benign and malignant tumors associated with β-catenin mutations in hepatocellular adenoma. Not surprisingly a correlation was also noted between miR-375 expression in hepatocellular adenoma and HCC, and three β-catenin–activated genes. However, these studies fall short of either demonstrating a direct link or defining the mechanism by which β-catenin mutations result in repression of miR-375, and no hypotheses to explain this link are provided. Interestingly, similar changes in miR-375 expression were also associated with β-catenin mutations in HCC. Because previous work has shown that the presence of β-catenin mutations would indicate a higher tendency toward malignant transformation, it may be expected that miR-375 may also have prognostic potential but this was not evaluated in this study.

Another miRNA, miR-107, is reported as repressed in HNF-1α–mutated adenomas. The relationship between miR-107 and hepatocyte nuclear factor 1α (HNF-1α) is supported by reciprocal changes in miR-107 expression in response to RNA interference–mediated inhibition of HNF-1α. Although HNF-1α binding sites upstream of the miR-107 preinitiation site occur, no functional studies to establish a direct link were performed. Details of the expression of other miRNAs such as miR-194 that can be modulated by HNF-1α were also not provided, although their expression would also have been expected to be altered in HNF-1α–inactivated adenomas. Functional studies that experimentally verify the role of HNF-1α in miR-107 are needed to explore the implications of these observations. The association between miR-107 and regulatory pathways involved in metabolism suggests an intriguing link between miR-107 and steatosis, a histologic feature common to these lesions, that also warrants further evaluation.

Regulation of miRNA expression by transcription factors other than HNF-1α has also been recognized. For example, the myocyte enhancer factor-2 transcription factor can increase the expression of miR-1-2 and miR-133a-1 during cardiac and skeletal muscle development, and the transcription factor Myc can either up-regulate the expression of miR17-92 cluster or down-regulate several other miRNAs in tumorigenesis.14–16 Similarly, the functional expression of transcription factors can also be regulated by miRNAs, as we have recently shown for signal transducer and activator of transcription 3 (STAT-3).17 The direct modulation of transcription factors by miRNAs adds another layer of complexity to the role played by these small RNA molecules in regulating gene expression.

Hepatocellular tumors comprise diverse benign and malignant neoplasms. Although the phenotypes can be broadly distinguished histologically or immunologically, these tumors can vary widely in their clinical behavior and prognosis. The use of miRNA-based classifications that correlate with etiology, pathogenetic changes, or malignant tendency will enhance molecular diagnosis and enable further definition of these phenotypes. In turn, this may yield clinically useful predictive markers of tumor behavior, as well as identify individual genetic and molecular contributors to tumorigenesis.


  1. Top of page
  2. References