Pancreatic cancer poses a grave prognosis due to the advanced and inoperable stage at which it is usually diagnosed and lack of effective therapy. Frondanol®-A5, produced from an edible sea cucumber inhibits growth of pancreatic cancer cells. In the present study, we examined the effects of a sub-fraction, Frondanol®-A5P on cell cycle and apoptosis. Cells were cultured in serum-free conditions 24 hours prior to treatment. Cell cycle and apoptosis-related proteins were evaluated by western blotting. Cells were stained with propidium iodide for cell cycle analysis. Messenger RNA was quantified by real-time RT-PCR. Frondanol®-A5P inhibited proliferation of pancreatic cancer cells by arrest in the G2/M phase of the cell cycle and induction of apoptosis. Thymidine incorporation was concentration-dependently decreased by Frondanol®-A5P and was markedly inhibited in AsPC-1 (P<0.005) and S2-013 (P<0.02), at 15.6 and 31.2 μg/ml, respectively. Marked cleavage of caspase 3 and PARP at 24 hours and increasing by 48 hours of treatment confirmed the induction of apoptosis. Western blots explained the G2/M phase arrest, with a decrease in cyclin A, cyclin B and CDC25c in treated cells versus controls at 24 hours. Frondanol®-A5 increased the apoptotic p21waf1 mRNA by 55-fold in AsPC-1 and 8-fold S2013 cells at three hours. The increase in p21waf1 protein was 6-fold in AsPC-1 cells and 11-fold in S2-013 cells at three hours, persisting beyond 24 hours. Frondanol®-A5P induced marked phosphorylation of p38 kinase within five minutes. Treating the cells simultaneously with the p38 kinase inhibitor, SB203580 markedly reduced the Frondanol®-A5P-induced expression of p21 waf1. Apoptosis induced by Frondanol®-A5P is associated with a marked increase in expression of the p-21 waf1 triggered via p38 kinase. It is likely that p21waf1 mediates the Frondanol®-A5P-induced cell cycle arrest. Since Frondanol®-A5P comes from as edible, nontoxic source, it may be valuable for therapy or prevention of pancreatic cancer.

Retinoids are promising agents for the prevention and treatment of several human malignancies, including pancreatic cancer. The effects of retinoic acid (RA) in pancreatic adenocarcinoma are controversial and the mechanisms of growth inhibition not clearly defined. Using optimized treatment conditions, we tested the effects of retinoic acid in pancreatic adenocarcinoma cells with varying degrees of differentiation. RA induced time- and concentration dependent growth arrest in the G1 phase of the cell cycle in CD-18 and CAPAN-2 cells and the G2/M phase in Hs766T cells. In CD-18 and CAPAN-2, RA reduced cyclin E and cyclin D1 proteins in a time- and concentration-dependent manner, respectively. Expression of the cyclin subunit, CDK2 associated with cyclin E in CD-18 cells was reduced, but CDK4 was unchanged. In Hs766T cells, cyclin B did not change, but CDK2 expression decreased. The changes in expression of cell cycle proteins were concomitant with the cell cycle arrest mediated by retinoic acid. Expression of mRNA of the CDK inhibitors, p21waf1 and p27kip1, were increased 60- and 40-fold, repectively, in CD-18 cells at 18 hours. However, p27 protein expression did not change but p21 protein was decreased. In Hs766T cells, there was a smaller increase in p21 mRNA followed by a similar decrease in p21 protein. In CAPAN-2 cells, expression of p21 and p27 depended on the concentration of RA. At low RA concentrations p21 and p27 proteins decreased, while at higher concentrations their expression was unchanged. These results indicate that the growth inhibitory effects of retinoic acid are mediated through the modulation of the activities of cyclins and their respective CDKs. This may be independent of changes in the CDK inhibitors, p21 and p27.

Retinoic acid (RA) induces growth inhibition by modulation of cell cycle associated genes, including induction of p21 in various types of cancer cells. The effects of RA on growth of pancreatic cancer cells are controversial. Using optimized treatment conditions under which all pancreatic cancer cells tested responded to RA, we investigated whether p21 could be involved in RA-induced growth inhibition. In CD-18 cells, RA caused a massive but transient 60-fold increase in p21 mRNA levels at 18 hours, accompanied by a three-fold increase in p21 protein levels. Levels of p21mRNA had normalized by 48 hours, while p21 protein declined to levels 80% lower than control. In HS766T cells the transient increase in p21 mRNA came later (24–48 hours) and was smaller (8-fold). No early increase in p21 protein was seen in HS766T, but levels declined at later times as in CD-18 cells. The decrease of p21 protein levels following an increase in the mRNA suggested that degradation of the protein was induced. The RA-induced decrease in p21 was blocked by the proteosome inhibitors, MG132 and lactacystin, but not by the caspase-3 inhibitor III. These findings suggest that the RA-induced p21 degradation is mediated by the proteosome but not by executioner caspases. However, we also detected ~32 fold increase in the mRNA expression of the recently discovered p53 inducible ring finger protein (p53RFP), which is an E3 ligase. This suggests that some form of p21 is targeted for ubiquitination. These results demonstrate RA-induced p21 proteosomal degradation which may be ubiquitin-independent. This decrease in p21 is unexpected in the face of RA-induced growth arrest and may represent a cell-survival mechanism.