Project description:To demonstrate CD133+CD44+ and CD133+CD44- subpopulations of hepatocellular carcinoma as distinct subgroups, we have employed whole genome microarray expression profiling as a discovery platform to reveal the gene profiles of different subgroups and identify genes responsible for the enhanced metastatic potentials of CD133+CD44+ tumor cells. CD133+CD44+ and CD133+CD44- tumor cells were isolated from three human metastatic hepatocellular carcinoma specimens. A 76-gene consensus signature was identified that distinguished between CD133+CD44+ and CD133+CD44- subgroups. CD133+CD44+ and CD133+CD44- subgroups from different patients were well clustered as two distinct classes according to this signature, and many genes in this signature were reported involved in tumor metastasis. Expression of four genes (CCL4, DKK3, CCR5 and MMP12) from this signature was confirmed in another three metastatic HCC specimens by real-time PCR. CD133+CD44+ and CD133+CD44- subpopulations of hepatocellular carcinoma were isolated from three metastatic hepatocellular carcinoma specimens by flow cytometry. A total of 30K to 50K cells for each subgroup was obtained for each microarray.
Project description:We devised a novel insertional mutagenesis approach based on lentiviral vectors to induce hepatocellular carcinoma in three mouse models and identified four novel cancer initiating genes. Two genes are the well characterized Braf and Sos1, while the other two are Fign, encoding an AAA ATPase whose functions are poorly understood, and the paternally expressed gene Rtl1 within the complex Dlk1-Dio3 imprinted region recently involved in stemness. Interestingly, Fign and Braf regulate the expression of the maternal genes of the Dlk1-Dio3 imprinted region, suggesting that both maternally and paternally expressed genes of this region play a role in hepatocarcinogenesis. Moreover, all the genes identified are upregulated and/or amplified/deleted in human hepatocellular carcinoma and play a relevant role in human hepatocarcinogenesis, as their expression levels and/or transcriptional signatures induced by their deregulation predict a different clinical outcome in hepatocellular carcinoma patients. Primary human hepatocytes were transduced with SINLV.ET.trBRAF, SINLV.PGK.GFP or mock treated. RNA was collected at 5 different timepoints post-transduction: 24h, 36h, 48h, 72h, 144h (6d). Each experimental point was done in triplicate (A,B,C)