Search

Increasing evidence has indicated that perturbation of cyclins is one of the major factors leading to cancer. The aim of this study was not only to investigate various cell cycle-related kinase activities in hepatocellular carcinoma (HCC), but also to analyze the difference of cell cycle-related kinase activity levels between hepatitis C virus (HCV)-induced HCC and HCV-induced cirrhosis. The protein levels of cyclins D1, E, A, and H, and of cyclin dependent kinase 1 (Cdk1), Cdk2, Cdk4, Cdk6, and Cdk7 in HCC and in surrounding nontumorous cirrhosis were determined by Western blot. The enzymatic activities of cyclins D1, E, A, Cdk1, Cdk4, Cdk6, Cdk7, and Wee1 were measured using in vitro kinase assays. Protein levels and kinase activities of cyclin D1, Cdk4, cyclin E, cyclin A, and Wee1 were significantly elevated in HCC compared with surrounding cirrhotic tissues. The enhanced cyclin D1-related kinase activity in HCC was accompanied by the up-regulation of Cdk4 activity, but not Cdk6 activity. The kinase activities of Cdk6, Cdk7, and Cdk1 did not differ between HCC and surrounding cirrhotic tissues. In addition, the protein levels and kinase activities of cyclin D1, Cdk4, and cyclin E were higher in poorly differentiated HCC and advanced HCC. In conclusion, the increases of cyclin D1, Cdk4, cyclin E, cyclin A, and Wee1 play an important role in the development of HCC from cirrhosis. Cyclin D1, Cdk4, and cyclin E activation may be closely related to the histopathologic grade and progression of HCC.

Cyclin D1 is considered to play a critical role in the progression from G1 to S phase of the cell cycle, and its overexpression is seen in many human tumors. However, previous studies in cell lines have shown that cyclin D1 is not sufficient to trigger cell replication. To directly test the role of cyclin D1 in the progression of the cell cycle, we have examined the proliferative response of hepatocytes to the hepatomitogen 1,4-bis[2-(3,5-dichloropyridyloxy)]benzene (TCPOBOP) in mice with homozygous disruption of the cyclin D1 gene. We found that 24 hours after administration of TCPOBOP, the number of bromodeoxyuridine (BrdU)-positive hepatocytes was significantly reduced in cyclin D1(-/-) (labeling index was 1.9% in knockout mice vs. 9.7% of heterozygous mice); however, no difference in the number of proliferating hepatocytes was found 36 or 72 hours after treatment (labeling index was 16% and 43% in cyclin D1(-/-) mice vs. 20% and 41% of heterozygous mice), indicating that lack of cyclin D1 may transiently delay entry into S phase but is not sufficient to inhibit the response of hepatocytes to mitogenic stimuli. The results also show that although there was no difference in hepatic protein levels of cyclin D2 and D3 between untreated cyclin D1(-/-) and cyclin D1(+/-) mice, messenger RNA (mRNA) and protein levels of cyclin E were much higher in the former. In conclusion, our results show that cyclin D1 is not essential for liver development and hepatocyte proliferation induced by mitogenic stimuli and suggest that overexpression of cyclin E may compensate for the lack of cyclin D1.

Amplification found in a number of cyclin genes, especially in cyclin D and E, is an important event process that takes place in cancers, including hepatocellular carcinoma (HCC). The activities of a wide range of cell cycle-related kinases remain obscure in HCC. The purpose of the present study is to determine the cyclins and kinase activities of HCC in Long-Evans Cinnamon (LEC) rats. Cyclin D1, E, A, H, Cdk1(cyclin-dependent kinase; Cdc2), Cdk4, and Cdk6 protein levels were determined by Western blot analysis at different pathologic stages of liver tissues exhibiting HCC. Enzymatic activities of cyclin D1, E, A, Cdk4, Cdk6, Cdc2, Cdk7, and Wee1 kinase were measured by in-gel kinase assay. Protein levels and kinase activities of cyclin D1, E, Cdk4, cyclin A, and Wee1 increased proportionally with the development of HCC, especially in the transition process from chronic hepatitis to HCC. Although Cdc2 kinase activity was found to increase slightly from normal liver to chronic hepatitis, its activity remained unchanged in the process from chronic hepatitis to HCC. Cdk6 and Cdk7 activities remained unchanged in the process from normal liver to HCC. These data suggest that the increase in Cdc2 kinase may play a role in the process from normal liver to chronic hepatitis, whereas the predominant increase in cyclin D1, Cdk4, cyclin E, cyclin A, and Wee1 suggests involvement not only in the process from normal liver to chronic hepatitis, but also during transition into HCC.

Expression of cell-cycle modulators at the G1-S boundary, retinoblastoma gene product (pRb), p21, p16, p27, p53, cyclin D1, and cyclin E was investigated with 104 hepatocellular carcinomas (HCC), as well as 90 of their adjacent noncancerous lesions and 9 normal liver control specimens. The labeling indices (LI) of pRb, p21, p16, and p27 were higher in HCC lesions than in the adjacent noncancerous lesions and normal controls. Especially, p27 LI in noncancerous lesions was significantly higher than that in normal livers (P =.011). Aberrant p53 expression and cyclin D1 and E overexpression were observed exclusively in HCC lesions. pRb was positive in 85.6% of the HCC cases and was not related to any clinicopathological parameters. The p21 LI was generally low (average, 5.5 +/- 9.8). Although a negative regulator, p21 LI was higher in cases with intrahepatic metastasis (P =.0359). The p16 LI was significantly decreased (P =.0121) in cases with advanced stage. p27 LI was significantly decreased in cases with portal invasion (P =.0409), poor differentiation (P <.0001), larger size (P =.0421), and intrahepatic metastasis (P =.0878, borderline significance). On the other hand, aberrant p53 expression showed positive relationships with poor differentiation (P =.0004) and Ki-67 LI (P =. 0047). Cyclin D1 overexpression was found in 32.6% of the cases and occurred more frequently in those with high Ki-67 LI (P =.0032), pRb expression (P =.0202), poor differentiation (P =.0612, borderline significance), and intrahepatic metastasis (P =.0675, borderline significance). Cyclin E was overexpressed in 35.5% and had positive relationships with Ki-67 LI (P =.0269) and stage (P =.0125). In univariate analysis, cases with p27 LI < 50 (P =.0004), cyclin D1 overexpression (P =.0041), and cyclin E overexpression (P =.0572, borderline significance) showed poorer outcomes for disease-free survival (DFS). In multivariate analysis, p27 expression could be recognized as an independent prognostic marker for DFS. These findings suggest that in HCC: 1) p27 is active against HCC progression in early phases and, possibly, hepatocarcinogenesis as a negative regulator and can be a novel prognostic marker for DFS; and 2) cyclin D1 predominantly works for cell-cycle progression at the G1-S boundary.

Partial hepatectomy (PH) triggers the entry of rat liver cells into the cell cycle. The signals leading to cell-cycle activation converge into a family of kinases named cyclin-dependent kinases (cdks). Specific cyclin-cdk complexes are sequentially activated during the cell cycle. Cyclin D-cdk4 and cyclin E-cdk2 are activated during the G1 phase, cyclin A-cdk2 is activated during the S phase, and cyclin B-cdk1 during mitosis. In the present study, we have examined the timing of the activation of cdk4 and cdk2, the intracellular location of G1/S cyclins and cdks, and the relationship between location and cdk4 and cdk2 activities during rat liver regeneration after a PH. Results showed that the activity of both kinases started at 13 hours and showed maximal levels at 24 hours after hepatectomy. In quiescent cells, cyclin D3 and cdk4 were cytoplasmatic, whereas cyclin D1 was nuclear. At 5 hours after hepatectomy, cyclin D3 and cdk4 began to move into the nucleus, and at 13 hours, they were mostly nuclear. During the first 13 hours after hepatectomy, significant amounts of cyclin D1-cdk4 and cyclin D3-cdk4 complexes were formed, but they were mostly inactive. At 24 hours, these complexes were maximally activated. This activation was associated with the accumulation of cyclin D1, cyclin D3, and cdk4 in a nuclear subfraction extractable with nucleases. At 28 hours, the activity of cdk4 in this nuclear subfraction decreased when cyclin D1 moved from this fraction to the nuclear matrix (NM) and the levels of cyclin D3 diminished. The maximal activation of cdk2 at 24 hours was also associated with the accumulation of cyclin E, cyclin A, and cdk2 in this nuclease-sensitive fraction. The inactivation of cdk2 at 28 hours was associated with a strong decrease in cdk2 in this nuclear subfraction. Thus, results reported here indicate that the activation of cdk4 and cdk2 observed in rat liver cells after a PH is associated with a specific intranuclear location of these cdks and their associated cyclins.