MFG-E8 (Milk fat globule-epidermal growth factor VIII), also called lactadherin or BA46, SED1 is a glycoprotein found in milk and mammary epithelial cells, it is a major protein component associated with milk fat globule membrane. Previously, our study showed that expression of MFG-E8 is gradually increased with hepatic differentiation of human embryonic stem cells (hESCs). Therefore, we hypothesized that MFG-E8 would be an early cancer stem cell marker, which may predict cancer progression. Our results showed that MFG-E8 was expressed in various human cancer cell lines such as HepG2, Hep3B, and Huh7. Production and secretion of the MFG-E8 were also confirmed in the conditioned media of those three cell lines using enzyme-linked immunosorbent assay. Next, we analyzed the MFG-E8 expression in 11 clinical cases of cholangiocellular carcinoma (CC) and 33 cases of hepatocellular carcinoma (HCC) by immunohistochemistry and examined the potential correlation with β-catenin and AFP, which are known cancer markers. According to hitological criteria, the progression of HCC and CC was evaluated and classified into high, low, metastatic, and well-, moderate-, poor-differentiated, respectively. Statistical analysis indicated that incidence of both HCC and CC is significantly associated with male compared to female (P<0.05). Tumor size also has positive correlation with age (r2=08948). Our immunohistochemistry data showed that MFG-E8 was expressed both HCC and CC tissue. Interestingly, the MFG-E8 expression was significantly increased with cancer progression (P<0.05) in both cases. Additionally, b-cateninexpression was increased and its localization was changed from membrane to cytoplasm and nucleus with the degree of HCC. Likely b-catenin, AFP was also increased with the degree of HCC but it was not correlated with severalty of CC. Importantly, both AFP and b-catenin were highly co-localized with MFG-E8 in HCC. These results suggest that MFG-E8 may have important physiological roles and its expression in HCC and CC would be considered as an important prognostic factor.

Hepatocytes derived from human embryonic stem cells (hESCs) may be a useful source for the treatment of diseased or injured liver. However, a low survival rate of grafted hepatocytes and immune rejection are still major obstacles to be overcome. We previously showed that secreted proteins (secretome) from hESC-derived hepatocytes had a potential therapeutic power in the tissue repair of injured liver without cell transplantation. The purpose of the present study was to discover key protein(s) in the secretome of hESC-derived hepatocytes using proteomic analysis and to study the tissue repair mechanism which may be operated by the secretomes. Purified indocyanine green+ hepatocytes derived from hESCs displayed multiple hepatic features, including expression of hepatic genes, production of albumin, and glycogen accumulation. The nano-LC/ESI-QTOF-MS analysis identified 365 proteins in the secretome of hESC-derived hepatocytes and the protein functional network analysis was conducted using the MetaCore TM from GeneGO. In addition, 20 tissue regeneration-related transcription factors (TFs) were extrapolated through further proteomic analysis. After intraperitoneal injection, the secretome significantly promoted the liver regeneration in a mouse model of acute liver injury. Protein functional network analysis on the secretome-induced regenerating liver confirmed 20 transcription factors (TFs) which were identified in the ICGhigh cells. The upreguation of these tissue repair-related TFs were validated by qPCR and western blotting on the regenerating liver tissues. These results demonstrate that application of the secretome analysis in combination with the protein functional network mapping would provide a reliable tool to discover new tissue-regenerating proteins as well as to expand our knowledge of the mechanisms of tissue regeneration.

family in the Brassica genome sequences by computational approach. The MITE family showed a total of 264bp length including 36bp terminal inverted repeats and remained 2bp (TA) targets it eduplication by its insertion. By searching the genome database of Brassica species, 516, 227, and 15 members were identified from 470Mbp of Brassica oleraceae, 154Mbp of B.rapa and 15Mbp of B.napus, respectively, indicating that there are approximately 692, 760, 1235 copies in B.oleracea, B.rapa and B.napus genomes,respectively. A total of 225 relatively intact MITE members, 146,68, and 11 members, which showed >80% sequence similarity and sequence coverage were identified and retrieved for MITE analysis from B.oleracea, B.rapa and B.napus genomes, respectively. Out of 225 MITE family members 159 having full structure of MITE and 66 having the truncated end either in right TIR or left TIR. Insertion polymorphism due to insertion or non-insertion of MITEs showed high level of polymorphism among accessions intra and inter species of Brassica. The new MITE would provide abetter tool for study molecular breeding in Brassica species and also helpful to understand their contribution in evolution and diversification of the highly duplicated Brassica genome.