Heat shock protein (HSP) 20, one of the low-molecular weight HSPs, is known to have versatile functions, such as vasorelaxation. However, its precise role in cancer proliferation remains to be elucidated. While HSP20 is constitutively expressed in various tissues including the liver, we have previously reported that HSP20 protein levels in human hepatocellular carcinoma (HCC) cells inversely correlate with the progression of HCC. In this study, we investigated the role of HSP20 in HCC proliferation. The activities of extracellular signal-regulated kinase (ERK), c-jun N-terminal kinase (JNK), and AKT were negatively correlated with the HSP20 protein levels in human HCC tissues. Since HSP20 proteins were hardly detected in HCC-derived cell lines, the effects of HSP20 expression were evaluated using human HCC-derived HuH7 cells that were stably transfected with wild-type human HSP20 (HSP20 overexpressing cells). In HSP20 overexpressing cells, cell proliferation was retarded, and the activation of the mitogen-activated protein kinases (MAPKs) signaling pathways, including the ERK and JNK, and AKT pathways, as well as cyclin D1 accumulation induced by either transforming growth factor-α (TGFα) or hepatocyte growth factor, were significantly suppressed compared with the empty vector-transfected cells (control cells). Taken together, our findings strongly suggest that HSP20 suppresses the growth of HCC cells via the MAPKs and AKT signaling pathways, thus suggesting that the HSP20 could be a new therapeutic target for HCC.

We did not find serum GPC3 level, measured by a commercially available ELISA kit with GPC3 antibody, to be useful in the diagnosis of HCC. However, we did observe increased GPC3 staining in HCC tissue with moderate or poor differentiation, suggesting that GPC3 is produced by HCC tumors. This lack of utility could have been due to the measuring procedure used in the present study. Further evaluation of GPC3 in HCC with other measuring procedures is needed.

Small heat shock proteins (HSPs) regulate a variety of cell functions. Among them, HSP22 and HSP20 are recognized to be ubiquitously expressed in various tissues. With regard to hepatocellular carcinoma (HCC) cells, we previously reported that phosphorylated HSP20 plays a suppressive role in transforming growth factor (TGF)-α-induced cell migration and invasion. In the present study, we investigated whether or not HSP22 is implicated in HCC cell migration. We detected HSP22 protein expression both in human HCC tumor (189.9±68.4ng/mg protein) and the adjacent non-tumor liver tissues (167.9±94.6ng/mg protein). The cases of low-quantity HSP22 protein level group (88.3≧ng/mg protein, the optimum cut-off value of HSP22) were increased in tumor tissues compared with the adjacent non-tumor tissues. The migration of human HCC-derived HuH-7 cells stimulated by TGF-α or hepatocyte growth factor (HGF) was significantly enhanced by the knockdown of HSP22 expression. Down-regulation of HSP22 protein in the cells markedly strengthened the AKT phosphorylation induced by TGF-α or HGF. Inhibitors of the phosphoinositide 3-kinase (PI3K)/AKT pathway, which suppressed the TGF-α-induced migration, significantly reduced the amplification by HSP22 knockdown. PI3K but not AKT was coimmunoprecipitated with HSP22 in HuH-7 cells. In addition, in human HCC tissues, a significantly lower HSP22 protein level in tumor tissues than in adjacent non-tumor tissues was observed more frequently in cases of moderately or poorly differentiated HCC than well-differentiated HCC. Taken together, our results strongly suggest that HSP22 represses HCC progression, especially HCC cell migration, by the down-regulation of the PI3K/AKT signaling pathway.

Aim:To clarify the diagnostic impact of liver fibrosis except for cirrhosis identified using shear wave elastography (SWE) in chronic hepatitis C (CHC) patients, and to compare the performance in diagnosing liver fibrosis among SWE and liver fibrosis indices.Methods: A total of 55 CHC patients who underwent liver biopsy were analyzed. The diagnostic performance for identifying significant liver fibrosis (F2-F3) for SWE, FIB-4 index, aspartate aminotransferase-to-platelet ratio index (APRI) and Forns' index was assessed using receiver-operator curve (ROC) analysis. Results:The median SWE elasticity value, FIB-4 index, APRI and Forns' index in the F0-F1 and F2-F3 groups were 6.3 kPa and 13.1 kPa; 1.52 and 4.45; 0.41 and 1.43; and 7.69 and 8.85, respectively (P < 0.001 for all four methods). Multivariate analysis showed that SWE was independently associated with the presence of significant liver fibrosis (odds ratio, 2.52; 95% confidence interval, 1.49-4.28; P < 0.001). The area under the ROC curve for SWE in diagnosing significant liver fibrosis was 0.94, indicating high diagnostic value, compared with 0.86, 0.88 and 0.83, for the FIB-4 index, APRI and Forns' index, respectively, which corresponds to moderate diagnostic value. The accuracy of SWE, FIB-4 index, APRI and Forns' index for diagnosing significant liver fibrosis was 90.9%, 76.4%, 74.5% and 67.2%, respectively.Conclusion: SWE has excellent ability for diagnosing significant liver fibrosis in CHC even when patients with cirrhosis are excluded. The diagnostic performance of SWE is superior to that of three liver fibrosis indices.