Cellular response and molecular mechanism of antitumor activity by leinamycin in MiaPaCa human pancreatic cancer cells.

Previous in vitro biochemical studies have revealed that the antitumor drug leinamycin causes oxidative DNA damage and DNA alkylation. However, it is still not clear whether the same mechanism(s) of action operate in cultured human tumor cells. Here, we evaluated the effects of leinamycin in the human pancreatic carcinoma cell line MiaPaCa. Leinamycin was highly toxic to MiaPaCa cells in vitro, with an IC50 value of 50 nM, and extensive DNA fragmentation was observed in leinamycin-treated MiaPaCa cells. Flow cytometric experiments showed that leinamycin was able to disrupt normal cell cycle progression, resulting in an initial arrest of the cells in S phase. With increased time or at higher concentrations of leinamycin, the population of cells in the sub-G1 phase gradually increased, indicative of apoptotic cell death due to DNA damage. Mammalian Chk2, but not Chk1 kinase, was found to be activated in MiaPaCa cells treated with leinamycin, indicating that cellular responses to leinamycin could be attributed to DNA strand break formation rather than DNA adduct formation. Like other DNA-damaging anticancer drugs, the downregulation of telomerase activity was also observed in MiaPaCa cells at cytotoxic concentrations. However, leinamycin failed to induce DNA ligase I expression in MiaPaCa cells, unlike other DNA-damaging agents, which are known to inhibit DNA replication by arresting DNA replication forks. Taken together, the results from our study indicate that the DNA strand breakage caused by the oxidative DNA-damaging property of leinamycin is directly related to the cellular responses of this drug in MiaPaCa cells over the DNA alkylation property in a dose-responsive manner.

Development of non-electrophoretic assay method for DNA ligases and its application to screening of chemical inhibitors of DNA ligase I.

A new rapid assay method for DNA ligases has been developed, which allows direct quantification of enzyme activity without using the traditional polyacrylamide gel electrophoretic technique. In this method, the 5'-biotinylated nicked duplex was used as a substrate for the ligase reaction, in which the 5'-end of the first oligonucleotide (19-mer) on the nicked strand is biotinylated and the second oligonucleotide (20-mer) on the same strand is labeled with radioactive 32P at the 5'-end. After ligation of the biotinylated 19-mer oligonucleotide into the second oligonucleotide with the reaction of DNA ligases, the biotinylated 19-mer oligonucleotide is converted into the radioactive 39-mer oligonucleotide. The ligase reaction products were heat-denatured to release both ligated and unligated biotinylated oligonucleotides. The biotinylated oligonucleotides were then captured on a streptavidin-coated microtiter plate and counted. The results obtained using this method correlated very well with those from the standard assay method using electrophoresis. Using this assay method, we were able to screen a chemical library and identify new DNA ligase inhibitors structurally related to resorcinol, which has growth inhibitory effects on the human breast cancer cell, MCF-7. The method described here is anticipated to be very useful for screening DNA ligase inhibitors from chemical libraries.

Effects of cisplatin on expression of DNA ligases in MiaPaCa human pancreatic cancer cells.

The effect of the broad-spectrum anticancer agent, cisplatin, on the expression of DNA ligase I in human pancreatic carcinoma MiaPaCa cells was examined in this study, since DNA ligase I is known to be involved in various DNA repair pathways. Upon exposure of MiaPaCa cells to cisplatin at near IC(50) value (2.5-5 microM), about 2-3-fold increase of DNA ligase I levels was observed within 24h, while levels of other DNA ligases (III and IV) remained unchanged or slightly decreased. The same fold-increase in DNA ligase I levels was also observed in MiaPaCa cells exposed to cytostatic concentrations, but not cytotoxic concentrations of cisplatin, which significantly reduced the number of cells. Flow cytometric analysis revealed that normal cell cycle progression was disrupted in the cells treated with cisplatin, resulting in an initial arrest of the cells in the S-phase, concomitant with a decrease of cells in G0/G1-phase. With time elapsing, the transition from S- to G2 + M-phase was observed, but further progression into G0/G1-phase was blocked. Overall, the increase of DNA ligase I expression seems to correlate well with the arrest of the cell cycle between the S- and G2-phases in response to cisplatin treatment. Interestingly, the cisplatin-induced DNA ligase I increase was abrogated by caffeine treatment in MiaPaCa cells, suggesting that caffeine sensitive kinases might be important mediators in the pathway, leading to the increase of DNA ligase I levels in response to cisplatin. We propose that the increase of DNA ligase I expression after exposure to cisplatin might be required for aiding the cells to recover from the damage by facilitating the repair process.

Induction of DNA ligase I by 1-beta-D-arabinosylcytosine and aphidicolin in MiaPaCa human pancreatic cancer cells.

Exposure of MiaPaCa cells to 1-beta-D-arabinosylcytosine (ara-C) resulted in an increase in DNA ligase levels up to threefold compared to that in the untreated control cells, despite significant growth inhibition. Increased levels of DNA ligase I protein appear to correlate with the appearance of increased mRNA levels. The [(3)H]thymidine incorporation experiment and the biochemical assay of total polymerase activity revealed that an increase in DNA ligase I levels after treatment with ara-C was not accompanied by an increase of DNA synthesis or an increased presence of DNA polymerase activity inside cells. When cells resumed DNA synthesis after drug treatment, DNA ligase I levels began to drop, indicating that increased DNA ligase I is not required for DNA synthesis. An increase in DNA ligase I was also observed in cells treated with aphidicolin, another inhibitor of DNA synthesis that inhibits DNA polymerases without incorporating itself into DNA, indicating that an increase in DNA ligase I levels could be caused by the arrest of DNA replication by these agents. Interestingly, caffeine, which is a well-known inhibitor of DNA damage checkpoint kinases, abrogated the increase in DNA ligase I in MiaPaCa cells treated with ara-C and aphidicolin, suggesting that caffeine-sensitive kinases might be important mediators in the pathway leading to the increase in DNA ligase I levels in response to anticancer drugs, including ara-C and aphidicolin. We propose that ara-C and aphidicolin induce damage to the DNA strand by arresting DNA replication forks and subsequently increase DNA ligase I levels to facilitate repair of DNA damage.

Enhancement of DNA ligase I level by gemcitabine in human cancer cells.

PURPOSE: DNA ligase I is an essential enzyme for completing DNA replication and DNA repair by ligating Okazaki fragments and by joining single-strand breaks formed either directly by DNA-damaging agents or indirectly by DNA repair enzymes, respectively. In this study, we examined whether the DNA ligase I level could be modulated in human tumor cell lines by treatment with gemcitabine (2', 2'-difluoro-2'-deoxycytidine), which is a nucleoside analogue of cytidine with proven antitumor activity against a broad spectrum of human cancers in clinical studies. EXPERIMENTAL DESIGN: To determine the effect of gemcitabine on DNA ligase I expression, Western blot analysis was used to measure the DNA ligase I levels in MiaPaCa, NGP, and SK-N-BE cells treated with different concentrations of gemcitabine and harvested at different time intervals. Cell cycle analysis was also performed to determine the underlying mechanism of DNA ligase I level enhancement in response to gemcitabine. In addition, other agents that share the same mechanism of action with gemcitabine were used to elucidate further details. RESULTS: When different types of tumor cell lines, including MiaPaCa, NGP, and SK-N-BE, were treated with gemcitabine, the level of DNA ligase I increased severalfold despite significant cell growth inhibition. In contrast, other DNA ligases (III and IV) either remained unchanged or decreased with treatment. Cell cycle analysis showed that arrest in S-phase corresponded to an increase of DNA ligase I levels in gemcitabine treated cells. Other agents, such as 1-beta-D-arabinofuranosylcytosine and hydroxyurea, which partly share mechanisms of action with gemcitabine by targeting DNA polymerases and ribonucleotide reductase, respectively, also caused an increase of DNA ligase I levels. However, 5-fluorouracil, which predominantly targets thymidylate synthase, did not cause an increase of DNA ligase I level. CONCLUSIONS: Our results suggest that an arrest of DNA replication caused by gemcitabine treatment through incorporation of gemcitabine triphosphate into replicating DNA and inhibition of ribonucleotide reductase would trigger an increase in DNA ligase I levels in cancer cells. The elevated presence of DNA ligase I in S-phase-arrested cells leads us to speculate that DNA ligase I might have an important role in repairing DNA damage caused by stalled replication forks.