Detection of low levels of DNA fragmentation in ovarian tumor cell lines.

The techniques presented in this chapter describe the experimental procedure for the identification of the nonrandom DNA fragmentation associated with apoptosis. The major benefits of this method are its ability to detect a low level of DNA fragmentation and its ability to detect large DNA fragments (50 kb and 300-1000 kb) as opposed to the internucleosomal DNA ladder. It is a qualitative technique although it can be of limited quantitative value. It can be used to detect apoptosis in cell lines following exposure to a wide variety of agents.

Cisplatin- and paclitaxel-induced apoptosis of ovarian carcinoma cells and the relationship between bax and bak up-regulation and the functional status of p53.

We investigated the roles of p53 and Bcl-2 homologues in the induction of apoptosis by cisplatin and paclitaxel in wild-type p53-expressing human ovarian carcinoma cells and cisplatin-resistant derivatives that have lost p53 function. Cisplatin induced apoptosis in parental A2780 but not in cisplatin-resistant A2780/cp70 cells, whereas paclitaxel induced apoptosis in both cell lines. Immunoprecipitation of p53 using antibodies specific for p53 conformation (pAb 1620 and pAb 240) showed that there were no relative changes in p53 conformation before and after cisplatin treatment in either cell line. A2780/cp70 cells have lost p53 function, yet they have wild-type p53 gene sequence. However, A2780/cp70 cells constitutively express more p53 in a form detected by pAb 240, an antibody that also detects mutant conformations of p53 that are transcriptionally inactive. There were no changes in levels of Bcl-2, Bcl-XL, or 24-kDa Bax over 72 hr after exposure to cisplatin or paclitaxel, but each agent led to up-regulation of Bak and 21-kDa Bax in A2780 cells. Paclitaxel, but not cisplatin, increased Bak and 21-kDa Bax levels in A2780/cp70 cells. These data suggest that apoptosis in A2780 and A2780/cp70 is associated with an increased level of Bak and 21 kDa Bax after drug-induced damage and that functional p53 may be required for this effect after cisplatin but not after paclitaxel.

hMLH1 expression and cellular responses of ovarian tumour cells to treatment with cytotoxic anticancer agents.

Loss of expression of the hMLH1 and hPMS2 subunits of the MutL alpha-mismatch repair complex is a frequent event (9/10) in independent cisplatin resistant derivatives of a human ovarian carcinoma cell line. However, only hMLH1 mRNA is decreased in these MutL alpha-deficient lines. No alterations in the levels of the hMSH2 and hMSH6 (GTBP) subunits of the MutS alpha-complex are observed. An increase in the proportion of ovarian tumours negative for the hMLH1 subunit is observed in samples taken at second look laparotomy after chemotherapy (36%: 4/11), compared to untreated tumours (10%: 4/39). No significant difference is observed for hMSH2, hMSH6 or hPMS2. Furthermore, cisplatin and doxorubicin-resistant ovarian lines deficient in hMLH1 expression are cross-resistant to 6-thioguanine and the methylating agent N-methyl-N-nitrosourea (MNU). Depletion of O6-alkylguanine-DNA-alkyltransferase (ATase) activity confers only limited increased sensitivity to MNU. Thus the mismatch repair deficient lines retain DNA damage tolerance even after ATase depletion. The hMLH1 deficient lines also lose ability to engage G1 and G2 cell cycle arrest after cisplatin damage. Together these data suggest that loss of hMLH1 expression may be a high frequency event following exposure of ovarian tumour cells to cisplatin and may be critically involved in the development of drug resistance. Thus, the hMLH1 status of these cells appears to be highly correlated with the ability to engage cell death and cell cycle arrest after DNA damage induced by cisplatin.

Microsatellite instability, apoptosis, and loss of p53 function in drug-resistant tumor cells.

We have examined microsatellite instability and loss of p53 function in human tumor cell line models of acquired anticancer drug resistance. We observe acquisition of an RER(+) phenotype in cell lines selected for resistance to cisplatin or doxorubicin. The majority of independent cisplatin-resistant sublines are RER(+), whereas the parental line shows no evidence of microsatellite instability. Microsatellite mutations in random, nonselected subclones of a cislatin-resistant line are observed in the absence of further drug exposure, suggesting that the RER(+) phenotype is a stable phenotype rather than being transiently induced by DNA damage. Furthermore, a cisplatin-resistant derivative shows reduction in a G:T mismatch recognition activity compared to the parental line. Independent lines selected by multiple exposure to cisplatin show resistance factors of up to a 5-fold by clonogenic assay and have reduced cisplatin-induced apoptosis. The resistant lines that are RER(+) show evidence of loss of p53-dependent functions, as measured by a loss of radiation-induced G(1) arrest and reduced CIP1 mRNA. Induced loss of p53 function by transfection of mutant TP53 does not cause a detectable RER(+) phenotype. We speculate that tolerance of DNA damage and expansion of cells with an RER(+) phenotype may select for reduced ability to engage apoptosis and loss of p53 function.

Cell-cycle arrest and p53 accumulation induced by geldanamycin in human ovarian tumour cells.

We have analysed the cell-cycle arrests and cytotoxicity of the A2780 human ovarian cell line in response to geldanamycin, a benzoquinoid ansamycin that can inhibit tyrosine kinases. Geldanamycin causes a dose-dependent G2 arrest and reversible inhibition of entry into the S phase in A2780 cells. After a 3-h exposure to 0.1 microM geldanamycin, the cells show an increase in accumulation of p53 protein that is maximal at 24 h after drug exposure. Increased p53 levels can be induced in cells by DNA-damaging agents; however, using alkaline elution and sister chromatid exchange assays we detect no DNA damage induced by geldanamycin. Using dominant negative mutant TP53 transfectants of A2780 we have analysed the possible dependence of geldanamycin-induced cell-cycle arrests on the presence of functional p53. We observe no difference in cell-cycle arrests in mutant p53 transfectants known to have the p53-DNA damage-response pathway inactivated as compared with vector-alone controls. Similarly, we observe no difference in clonogenic resistance to the cytotoxicity of geldanamycin in these cells. These results suggest that geldanamycin can induce increased p53 protein by a mechanism not involving DNA damage. Furthermore, the cell-cycle arrests and cytotoxic effects of geldanamycin in these cells are not mediated by p53-dependent pathways.

Cell cycle arrests and radiosensitivity of human tumor cell lines: dependence on wild-type p53 for radiosensitivity.

Loss of p53 function has been shown to cause increased resistance to ionizing radiation in normal murine cells; however, the role of p53 in radioresistance of human tumor cells is less clear. Since wild-type p53 function is required for radiation-induced G1 arrest, we measured G1 arrest in 12 human tumor cell lines that have a wide range of radiosensitivities (surviving fraction at 2 Gy, 0.11-0.8). We observed a significant correlation between the level of ionizing radiation-induced G1 arrest and radiosensitivity. Cell lines having G1 arrest are more radiosensitive. There is no correlation between maximal G2 arrest and radiosensitivity. Expression of a dominant-negative mutant of p53 (codon 143, Val to Ala) in transfectants of the radiosensitive human ovarian cell line A2780 abrogates the radiation-induced G1 arrest. Such mutant p53 transfectants are more resistant to ionizing radiation than the parental line and vector-alone transfectants, as measured by clonogenic assays. These results support the concept that wild-type p53 function is required for sensitivity of tumor cells to DNA-damaging agents, such as ionizing radiation, and that the loss of p53 function in certain human tumor cells can lead to resistance to ionizing radiation.