Analisi integrata dell'espressione di geni/microrna in un modello di epatocancerogenesi sperimentale

Hepatocellular carcinoma (HCC) is the fifth most common cancer in the world, accounting for an estimated half million deaths annually, and represents one of the major health problems. Although much is known about the cellular changes that lead to HCC, the molecular pathogenesis of HCC is not yet well understood. Gene expression studies conducted with microarray techniques and real-time PCR, suggest that tumors are characterized by an aberrant activation of signal transduction pathways involved in proliferation, survival, cell differentiation and angiogenesis. However, for HCC, these studies don’t allow the identification of a "signature" or a single specific pathway that is predominantly involved in the development and prognosis of the malignancy. Recently it has become clear that the classification and stratification of cancer can be performed not only through the analysis of gene expression, but also by analyzing the expression of microRNAs, small non-coding RNA molecules that negatively control gene expression and protein synthesis. In the present study we performed an integrated analysis of genome-wide mRNA and microRNA (miR) expression profiles to characterize the molecular events involved in the step-by-step progression (preneoplastic nodules-adenoma-early HCC-advanced HCC) of hepatocellular carcinoma (HCC) in the rat Resistant-Hepatocyte (R-H) model. Interestingly, while analysis of the transcriptome clustered together preneoplastic lesions and advanced HCC, suggesting that the majority of the genes dysregulated in HCC are already aberrantly expressed in early lesions, miRNome analysis did not co-cluster the two populations but, very interestingly, stratified the lesions according to their stage of progression to HCC. The results also unveiled specific genes/miRs, altered in the very early steps of the carcinogenic process, in the transition from adenoma to early HCC or in the progression to advanced HCC. By assessing the correlation between the expression of each miRNA and its targets, we determined that distinct pathways are aberrantly activated in different stages of the carcinogenic process. This integrated approach was also able to identify molecular events discriminating the preneoplastic lesions that will progress to HCC from those that spontaneously regress. Finally, 110 orthologous genes were almost super imposable between rat and human HCC signatures, supporting the value of the R-H model in recapitulating human liver cancer. Conclusions: This systematic analysis deciphered the molecular phenotypes of the several steps involved in the onset and progression of HCC and investigated their variations at mRNA and miR levels. In view of the striking similarity between mRNA and miRs commonly dysregulated in rat and human HCC, our results provide a valuable source for future studies and highlight promising genes, miRNAs, pathways and processes which may be useful for diagnostic or therapeutic applications.