| Hepatocellular carcinoma (HCC), one of the most common cancers worldwide,represents the third cause of cancer-related death. The rate of incidence is the highest inAfrica and South East Asia. It is estimated that about600,000new cases of primary livercancer occur each year, more than half of these people are in China. The development ofHCC is a multi-step process, hepatitis B and hepatitis C infection, alcohol abuse and livercirrhosis have been implicated in liver cancer development. Although the treatment ofHCC has made a great progress, the incidence of tumor metastasis, and tumor recurrenceafter surgical resection is still high. The occurrence and metastasis of HCC was caused bymalignant proliferation and migration of tumor cells. In order to find moreeffective means of prevention and treatment for HCC, genes and proteins, which may leadto malignant proliferation and migration, have been extensively studied and found thataseries of genes and signaling pathways including Wnt/β-catenin, p53, Rb and Ras/MAPKwere involved in the development of HCC. To explore the mechanism of HCC invasionand metastasis, and looking for new diagnostic markers and therapeutic targets is thekey to improve the therapeutic effect for HCC.microRNAs (miRNAs) are a class of non-coding single-stranded molecules, whichhave been found in recent years, is considered a major breakthrough in the field ofregulation of gene expression. miRNAs were conserved among different species. The firstmiRNA has been found in C. elegans since1993, a large number of miRNAs have beenfound in tobacco, fruit flies, zebrafish, mice and humans. miRNAs are transcribed ashairpin pri-miRNAs which, inturn, are processed by the Drosha-DGCR8and Dicer,resulting in22bp long mature miRNAs Duplexes. Mature miRNAs are then loaded intomiRISC (miRNA-induced silencing complex, RISC) and upon recognition of sequencecomplementarity between the miRNA "guide strand" and its target sequence generallylocated at the3' unstranslated region (3' UTR) of mRNAs, are responsible for messengerRNAs degradation and decay or repression of their translation. The silencing of targetgenes by miRNAs is a post-transcriptional regulatory mechanism, which is a fine-tuning ofprotein expression of widespread presence in cells. More importantly, most miRNAsdisplay an imperfect base-pairing with their complementary sequences at the3' UTR of target mRNAs, a miRNA can act on multiple target genes. Computational predictionsindentified more than30%of protein-coding mRNAs as modulated by miRNAs. It hasbeen confirmed that miRNAs was widely involved in many major human diseases,including cancer, neurological diseases, immune system and cardiovascular systemdiseases as well as apoptosis, proliferation, differentiation, and migration and otherimportant cellular processes. A large number of investigations have shown that miRNAs isrelated to tumor cell proliferation and migration, as well as the occurence of HCC. Thoughthe analysis of HCC animal models and clinical HCC samples, it has been found that,relative to normal liver tissue,some miRNAs expression is downregulated in HCC, whilesome is upregulated. miR-224, miR-21, miR-222and miR-18were found upregulated inHCC, while miR-112a, miR-375and miR-195were downregulated.In this study, by using a miRNA array-based analysis, we explore the function andmechanism of miR-210and its target gene SIN3A in tumor invasion and metastasis ofHCC in order to find a new target point of HCC.1. Screening and identification of aberrant expression ofmiRNAs in HCCTo identify the miRNAs which may be involved in the development of HCC, wecollected six different human clinical specimens and then used Agilent humanmiRNA-array analysis to screen differentially expressed miRNAs between HCC tissuesand respective side normal tissues.26miRNAs are upregulated in HCC, such as miR-210,miR-10b, miR-1181, miR-1246, miR-1290, miR-130b and miR-134(Foldchange>2,P<0.05), whereas39miRNAs are down-regulated, such as miR-144,miR-145,miR-146a,miR-150and miR-194(Foldchange <0.5, P<0.05). To verify the microarray results,real-time quantitative PCR assays of stem-loop method were used to validate thedifferenence of miRNAs expression between HCC and adjacent-nontumorous.Theverification results indicated that miR-21and miR-222were upregulated in HCC, whereasmiR-99awas downregulated in HCC, of which was consistent with the results of the chipscreening, showing that the chip screening results was accurate and reliable. It wasconfirmed that the expression of miRNAs between HCC tissue and adjacent tissue wasdifferent, indicating that aberrant expressed miRNAs may play an important role in thedevelopment and progression of HCC. Screened by preliminary microarray data andretrieved relevant literature, we found that the expression of miR-210was significantlyupregulated in the breast cancer, pancreatic cancer, ovarian cancer, urinary system tumor and head and neck cancer and was associated with the prognosis of patients. However, theexpression and function of miR-210in HCC have not been reported. Additionally, theexpression of miR-210was induced by hypoxia-inducible factor-1α (HIF-1α). Theexpression of HIF-1α was significantly upregulated in HCC, indicating that there is acertain correlation between HIF-1α signaling pathway and miR-210, so, miR-210waschosen for the further study.2. The expression and function of microRNA-210in HCCThe expression level of miR-210in HCC, adjacent noncancerous tissues and normalliver tissues were detected by quantitative PCR and were found that miR-210wassignificantly up-regulated in HCC compared with adjacent tissues and normal liver tissues.The expression of miR-210in HCC is closely related to the progress of HCC, theexpression level of miR-210in chronic hepatitis liver tissues and cirrhotic liver tissues wassignificantly higher than the normal liver tissues, and expression of miR-210in HCC wassignificantly higher than the chronic hepatitis liver tissues and cirrhotic liver tissues. Theexpression of miR-210in the patients of Barcelona Clinic Liver Cancer (BCLC) stage Band C was significantly higher than patients of BCLC stage0and A, which shows that theexpression level of miR-210is closely associated with HCC progress. In order to verify therelationship between miR-210expression levels and the prognosis of HCC patients,120HCC patients underwent surgery were randomly selected, and expression levels ofmiR-210in HCC were detected. All120patients were divided into two groups accordingto the expression level of miR-210in120cases, the cumulative disease-free survival (DFS)rates and overall survival (OS) rates were compared between the two groups. It was foundthat the DFS and OS rates were significantly higher in patients with lower miR-210expression levels. Cox proportional hazard model was used to estimate the independentimpact of each variable on DFS and OS, and found that miR-210expression levels, theAFP expression levels, vascular invasion, spreading nodules and tumor size werepredictive of shorter DFS. Vascular invasion, spreading nodules and tumor size werepredictive of shorter OS. Although, miR-210expression levels can not be considered as anindependent risk factor of postoperative survival, it reflected the impact at least in part onthe survival of patients (P=0.076). In addition, by comparing clinical and pathologicalfeatures of HCC patients of two groups, it was shown that the proportion of combinedvascular invasion was significantly higher in patients with higher expression levels ofmiR-210compared to the patients with lower expression of miR-210. These results indicated that expression levels of miR-210was associated with patient prognosis, and itshigh expression indicated poorer prognosis of patients and more malignant andinvasiveness of tumor. To test the influence of miR-210on cell proliferation and invasiveability, MTT assay was used to investigate whether the overexpression of miR-210aggravate cell proliferation. miR-210transfected hepatocytes HL-7702revealed that cellproliferation was significantly enhanced, suggesting that miR-210can improve the abilityof cell proliferation. Next, we established miR-210stable transfected cell lines onSMMC-7721cells by lentivirus systems. Over-expression of miR-210significantlyenhanced cell ability of invasion and metastasis. These results were consistent with thecondition that the proportion of combined vascular invasion and intrahepatic metastasiswas significantly higher in patients with high expression levels of miR-210.Tumourigenicity assay in nude mice indicated that overexpression of miR-210in HL-7702and SMMC-7721promoted tumor formation compared with that of control cells. Theseresults, including the data from HCC patients, showed that expression of miR-210reflectedthe biological characteristics of the tumor, indicating the malignant degree of the tumor. Itshigh expression showed more malignant and invasiveness of tumor.miRNAs exert biological function mainly through inhibiting their target gene.Identification the target genes of miRNAs can facilitate to disclose its mechanism. In orderto reveal the mechanism of miR-210in the promotion of HCC development, invasion andmetastasis, we searched the putative target genes of miR-210though bioinformatics, andfound SIN3A as a possible downstream target gene of miR-210.SIN3A is a corecomponent of a large multi-protein co-repressor complex with associated histonedeacetylase (HDAC) activity. Physical interactions of SIN3A with many sequence-specifictranscription factors has linked the SIN3A co-repressor complex to the regulation ofdiverse signaling pathways and associated biological processes. It has recently beenreported that tumor suppressor proteins, including p53, pRb, and Menin, repress theirtarget genes through interaction with the SIN3A/HDAC complex. Moreover, the aberrantrecruitment of this complex by altered transcription factors has been demonstrated to bepathogenic in several human malignancies, such as acute promyelocytic leukemia andacute myeloid leukemia. A conditional SIN3A deletion analysis revealed the deregulationof genes involved in cell cycle regulation, DNA replication, DNA repair, apoptosis,chromatin modifications, and mitochondrial metabolism. Thus, we speculated that SIN3Ainteracted with a number of tumor-associated genes and play an important role in tumor development through different signal transduction pathways.To verify whether SIN3A is a direct target of miR-210, a dual-luciferase reportersystem was used by co-transfection of miR-210and a luciferase reporter plasmidcontaining the3' UTR of human SIN3A. The luciferase activity was significantly inhibitedby miR-210co-transfection, and further miR-210failed to inhibit the expression ofluciferase construct on target site deletion, suggesting that miR-210can directly target the3' UTR of SIN3A. Moreover, in BEL-7402and SMMC-7721cells, endogenous expressionof SIN3A protein was suppressed by miR-210transfection, further proving that Sin3A isdirectly targeted and regulated by miR-210expression.Then the expression level of SIN3A was detected in the clinical samples of HCCpatients by real-time quantitative PCR assay and was found that expression of miR-210was negatively correlated with expression of SIN3A. The expression of SIN3A in HCCparaffin samples was analyzed by immunohistochemical method, and association betweenthe SIN3A expression and patient's prognosis was investigated. We found that theexpression of SIN3A was negatively correlated with that of miR-210, and patients withhigh expression of SIN3A exhibit higher cumulative DFS rates and OS rates than patientswith low expression of SIN3A. This was opposite to the relationship between expressionlevels of miR-210and patients prognosis. It also indirectly validated that SIN3A wasindeed a target gene of miR-210, and miR-210exerted its biological function by targetingSIN3A.Taken together, our results demonstrated that miR-210, upregulated in HCC, is anoncomiRNA. miR-210promotes proliferative and invasion ability of HCC cells both invitro and in vivo by targeting SIN3A. Furthermore, miR-210have been identified tocorrelate with prognosis, which maybe developed to be potential therapeutic targets ofHCC. |