Overexpression of HMGA1 promotes anoikis resistance and constitutive Akt activation in pancreatic adenocarcinoma cells.

HMGA1 proteins are architectural transcription factors that are overexpressed by pancreatic adenocarcinomas. Roles of HMGA1 in mediating the malignant phenotype of this cancer are poorly understood. We tested the hypothesis that overexpression of HMGA1 promotes resistance to anoikis (apoptosis induced by anchorage deprivation) in pancreatic cancer cells. HMGA1 cDNA was stably transfected into MiaPaCa2 human pancreatic adenocarcinoma cells (which have low baseline expression levels of HMGA1). Cells were grown in suspension on PolyHEMA-coated plates and their susceptibility to anoikis was assayed using flow cytometry. Overexpression of HMGA1 was associated with marked reductions in susceptibility to anoikis in concert with increases in Akt phosphorylation (Ser473) and in Akt kinase activity and with reductions in caspase 3 activation. Inhibition of phosphoinositidyl-3 (PI3-K)/Akt pathway with either the small molecule inhibitor LY294002 or dominant-negative Akt resulted in reversal of anoikis resistance induced by HMGA1 overexpression. Further, RNA interference-mediated HMGA1 silencing in MiaPaCa2 and BxPC3 (a human pancreatic adenocarcinoma cell line with high baseline levels of HMGA1 expression) cells resulted in significant increases in susceptibility to anoikis. Our findings suggest HMGA1 promotes anoikis resistance through a PI3-K/Akt-dependent mechanism. Given the putative associations between anoikis resistance and metastatic potential, HMGA1 represents a potential therapeutic target in pancreatic adenocarcinoma.

Absence of PIK3CA hotspot mutations in hepatocellular carcinoma in Japanese patients.

A recent study revealed that the p110alpha (PIK3CA), catalytic subunit of phosphatidylinositol 3-kinase (PI3K), is somatically mutated in many types of cancer. For example, PIK3CA is mutated in an estimated 35.6% of hepatocellular carcinoma (HCC) cases. To measure the frequency of PIK3CA hotspot mutations in Japanese HCC patients, exons 9 and 20 of the PIK3CA gene were sequenced in 47 clinical HCC samples. Contrary to expectations, no hotspot mutations were found any of the HCC samples. In addition, we found abnormally migrating waves near the end of exon 9 in the PCR chromatograms from 13 of the 47 samples. PCR amplification and subsequent cloning and sequencing revealed that these chromatograms contained two distinct sequences, the wild-type p110alpha sequence and a different sequence found on human chromosome 22q11.2, the Cat Eye Syndrome region, which contains a putative pseudogene of PIK3CA. These abnormally migrating waves were also found in noncancerous liver tissue, indicating that this was not a result of HCC-associated mutations. Therefore, it is likely that the percentage of hotspot mutations in the PIK3CA gene of Japanese HCC patients is lower than was previously reported.

The hepatitis B virus X protein enhances AP-1 activation through interaction with Jab1.

Hepatitis B virus X protein (HBx) has many cellular functions and is a major factor in hepatitis and hepatocellular carcinoma caused by HBV infection. A proteomic approach was used to search for HBx-interacting proteins in order to elucidate the molecular mechanism of hepatocarcinogenesis. HBx was attached to myc and flag tags (MEF tags) and expressed in 293T cells; the protein complex formed within the cells was purified and characterized by mass spectrometry. COP9 signalosome (CSN) subunits 3 and 4 were subsequently identified as HBx-interacting proteins. In addition, CSN subunit 5, Jun activation domain-binding protein 1 (Jab1), was shown to be a novel cellular target of HBx. In vivo and in vitro interactions between HBx and Jab1 were confirmed by standard immunoprecipitation and GST pull-down assays. An analysis of HBx deletion constructs showed that amino acids 30-125 of HBx were responsible for binding to Jab1. Confocal laser microscopy demonstrated that HBx was mainly localized in the cytoplasm, while Jab1 was found mainly in the nucleus and partially in the cytoplasm, and that the two proteins colocalized in the cytoplasm. The cotransfection of HBx and Jab1 resulted in substantial activator protein 1 (AP-1) activation and knockdown of endogenous Jab1 attenuated AP-1 activation caused by HBx. In addition, the coexpression of HBx and Jab1 potentiated phosphorylation of JNK, leading to the subsequent phosphorylation of c-Jun, whereas the level of c-Jun and JNK phosphorylation induced by HBx was decreased in Jab1 knockdown cells. These results suggest that the interaction between HBx and Jab1 enhances HBx-mediated AP-1 activation.