Search

Glycinamide ribonucleotide transformylase (GART) is a folate-dependent enzyme in the de novo purine pathway that has been the target of antineoplastic intervention for almost 2 decades. Until now, its expression and functional significance in hepatocellular carcinoma (HCC) have been unclear. We demonstrated by Western blotting that the expression of GART was markedly up-regulated in HCC patients. Immunohistochemistry staining was used to determine the expression of GART in HCC and adjacent nontumor tissues from 96 patients. Increased expression of GART correlated positively with the histologic grade (P = .001), tumor size (P = .043), number of tumorous nodes (P = .020), and intrahepatic metastases (P = .031), suggesting a role for GART in the progression of HCC. Patients with higher GART expression had a much worse overall survival rate than those with low expression (P = .002). Furthermore, multivariate analysis showed that GART expression was an independent predictor of overall survival (hazard ratio, 2.265; 95% confidence interval, 1.335-3.842; P = .002). Depletion of GART by small interfering RNA inhibited cell proliferation and blocked S-phase and mitotic entry in cultured HepG2 and BEL-7404 cells. Western blot analyses showed that GART depletion decreased the proliferating cell nuclear antigen concentration. Collectively, our clinical and in vitro data indicate that GART expression may be one of the causative factors for a poor prognosis in HCC.

p53-Induced ring-H2 protein (Pirh2), a recently identified ubiquitin-protein ligase, interacts with p27(Kip1) to promote ubiquitination of p27(Kip1) independently of p53. High Pirh2 and low p27(Kip1) immunoreactivity are associated with a poor prognosis in several cancers, including resistant phenotypes. In the present study, we investigated the role of Pirh2 and p27(Kip1) in human hepatocellular carcinoma (HCC) progression. Immunohistochemical analysis was performed on formalin-fixed paraffin sections of 87 specimens. Statistical analysis showed that expression of Pirh2 was negatively related to p27(Kip1) expression (r = 0.787; P < .05), and Pirh2 expression correlated significantly with histologic grade (P < .001), venous invasion (P = .004), tumor size (P = .024), and the presence of multiple tumor-bearing lymph nodes (P = .017), whereas p27(Kip1) expression correlated significantly with histologic grade (P < .001), venous invasion (P = .048), and cirrhosis (P = .028). By Kaplan-Meier analysis, the survival curves of low versus high expressers of Pirh2 and p27(Kip1) showed significant separation (P < .01). Molecular interaction could be demonstrated between Pirh2 and p27(Kip1) in three HCC cell lines. In vitro, following release of two HCC cell lines from serum starvation, the expression of Pirh2 was upregulated, whereas p27(Kip1) was downregulated. Our results suggest that Pirh2 mediates the degradation of p27(Kip1) and participates in cell proliferation in human HCC. These findings provide a rational framework for further development of Pirh2 inhibitors as a novel class of anti-tumor agents.

Early mitotic inhibitor-1 (Emi1) is a key cell-cycle regulator that promotes S-phase and M-phase entry by inhibiting anaphase-promoting complex/cyclosome (APC/C) activity. Immunohistochemical analysis was performed in 114 human hepatocellular carcinoma (HCC) samples, and the data were correlated with clinicopathologic features. Univariate and multivariate survival analyses were performed to determine the prognostic significance of the proteins. Expression of Emi1 correlated directly with the stage of HCC. More importantly, high expression of Emi1 was associated with a poor outcome. Western blot analysis showed that Emi1 was highly expressed in HCC compared with the adjacent noncancerous tissue. In vitro, after the release of HCC cell lines from serum starvation, the expression of Emi1 APC/C substrates (cyclins A, B) and Skp2 was up-regulated, whereas p27(Kip1) was down-regulated. In addition, we used small interfering RNA to knock out Emi1 expression and observed its effects on HCC growth in vitro to determine whether loss of Emi1 could inhibit cell proliferation by blocking S-phase and mitotic entry. Western blot analyses indicated that deletion of Emi1 was positively correlated with APC/C substrates (cyclins A, B) and Skp2 but was negatively correlated with p27(Kip1). Emi1 inhibits APC/C activity, whereas Skp2 degradation is mediated by APC/C, and degradation of Skp2 can stabilize p27(kip1). These results suggested that Emi1 participates in HCC cell proliferation and that progression is controlled by APC/C inhibition, which stabilized Skp2 and enabled p27(kip1) degradation. These findings provide a potential therapeutic strategy for HCC.