Amsacrine-promoted DNA cleavage site determinants for the two human DNA topoisomerase II isoforms alpha and beta.

Site-specific DNA cleavage by topoisomerase II (EC 5.99.1.3) is induced by many antitumour drugs. Although human cells express two genetically distinct topoisomerase II isoforms, thus far the role and determinants of drug-induced DNA cleavage have been examined only for alpha. Here we report the first high-resolution study of amsacrine (mAMSA) induced DNA breakage by human topoisomerase II beta (overexpressed and purified from yeast) and a direct comparison with the recombinant alpha isoform. DNA cleavage in plasmid pBR322 and SV40 DNA was induced by alpha or beta in the absence or presence of the antitumour agent mAMSA, and sites were mapped using sequencing gel methodology. Low-resolution studies indicated that recombinant human alpha promoted DNA breakage at sites akin to those of beta, although some sites were only cleaved by one enzyme and different intensities were observed at some sites. However, statistical analysis of 70 drug-induced sites for beta and 70 sites for alpha revealed that both isoforms share the same base preferences at 13 positions relative to the enzyme cleavage site, including a very strong preference for A at +1. The result for recombinant alpha isoform is in agreement with previous studies using alpha purified from human cell lines. Thus, alpha and beta proteins apparently form similar ternary complexes with mAMSA and DNA. Previous studies have emphasized the importance of DNA topoisomerase II alpha; the results presented here demonstrate that beta is an in vitro target with similar site determinants, strongly suggesting that beta should also be considered a target of mAMSA in vivo.

Drug sensitivity and sequence specificity of human recombinant DNA topoisomerases IIalpha (p170) and IIbeta (p180).

Effective anticancer agents, such as epipodophyllotoxins and anthracyclines, exert their antitumor activity through stabilization of cleavable topoisomerase II/DNA complexes, which may result in DNA breakage on detergent addition. Two isozymes (alpha and beta) of DNA topoisomerase II are present in human cells; however, their roles as drug targets have not been completely defined. We determined the in vitro isoenzyme sensitivities to VM-26 (teniposide) and 4-demethoxy-3'-deamino-3'-hydroxy-4'-epi-doxorubicin (an anthracycline analog) and established the sequence selectivity of isoenzyme-mediated DNA cleavage. Human topoisomerases IIalpha and IIbeta were purified from yeast cells overexpressing the corresponding plasmid-borne cDNA. Enzyme sensitivities to drugs were measured by a DNA cleavage assay using 32P-labeled simian virus 40 DNA fragments, and cleavage sites were mapped using agarose and sequencing gels. Both isozymes were sensitive to the studied poisons. They stimulated similar cleavage intensity patterns in agarose and sequencing gels; however, minor differences could be detected. The results showed that local base preferences for DNA cleavage without drugs were different at positions -2 and -1. On the other hand, sequence specificities of VM-26 and 4-demethoxy-3'-deamino-3'-hydroxy-4'-epi-doxorubicin were identical for both isozymes and corresponded to those of the native murine enzyme. The identical drug sequence specificities suggested that molecular interactions of the tested drugs in the ternary complex are likely similar between the two isozymes. The current findings indicate that both topoisomerase IIalpha and IIbeta may be in vivo targets of antitumor poisons.

Gene expression of DNA topoisomerases I, II alpha and II beta and response to cisplatin-based chemotherapy in advanced ovarian carcinoma.

DNA topoisomerases, nuclear enzymes that regulate DNA topology, are recognized as the primary targets of effective anti-tumor drugs. These enzymes may also have a role in the repair of DNA damage induced by alkylating agents and platinum compounds; therefore, their expression may be a determinant of tumor response to chemotherapy. Our study was undertaken in an attempt to establish a correlation between the enzyme expression and response of ovarian cancer to cisplatin-based chemotherapy. The expression of topoisomerase I, II alpha and II beta genes was assessed by RNase protection assay in tumor specimens obtained from 37 untreated patients with advanced epithelial ovarian cancer at initial surgery and from 13 pre-treated patients at subsequent laparotomy. The expression levels were compared with those found in 5 specimens from benign ovarian tissue and 5 specimens from normal ovarian tissue. The expression levels in untreated patients were used to establish a correlation with response to high-dose cisplatin therapy. A significant intertumor variability of mRNA expression was noted for all the genes examined. However, a comparison of median values indicated a remarkable increase of expression in malignant tumors over benign or normal tissues only for topoisomerase II alpha. This change is not related to alterations or amplification of topoisomerase II alpha gene. Interestingly, a correlation was found between tumor response to chemotherapy and the expression level of the isoform alpha (but not of topoisomerase II beta and topoisomerase I). The observed correlation suggests a contribution of the enzyme in determining tumor sensitivity. Alternatively, increased expression levels of the alpha isoenzyme gene in responsive tumors might reflect higher fractions of proliferating tumor cells that may be more drug-sensitive than resting cells.

A comparative study of p53 gene mutations, protein accumulation, and response to cisplatin-based chemotherapy in advanced ovarian carcinoma.

The p53 protein is a multifunctional transcriptional regulator involved in cellular response to DNA damage and has been implicated as a putative determinant of sensitivity of tumor cells to cytotoxic agents. Since the p53 gene becomes inactivated in over one-half of advanced ovarian carcinoma, in this study we have examined the relationships between p53 gene alterations, p53 immunoreactivity, and response to cisplatin-based chemotherapy in ovarian cancer patients. All patients had advanced (FIGO stage III or IV) ovarian carcinoma and, with one exception, were untreated at the time of collection of tumor specimens. After initial debulking surgery, patients received high-dose cisplatin therapy. Tumor samples were analyzed for p53 gene mutations and for p53 protein accumulation, and the findings were correlated with tumor responsiveness. Of the 33 tumors examined, p53 gene mutations were found in 20 cases, including 15 missense mutations, 2 deletions, 2 nonsense mutations, and a base substitution at splice site. Twenty tumors showed positive immunostaining for p53. Only missense mutations were associated with positive immunostaining. In addition, p53 overexpression was detected in five tumors in the absence of mutations. Most (12 of 14) of the missense mutations associated with p53 protein stabilization were found refractory to therapy, as well as tumors overexpressing wild-type p53 (4 of 5). A significant correlation has been found between p53 accumulation, type of mutation (i.e., missense mutations), and pathological response to cisplatin-based therapy. In conclusion, the present results are consistent with a role of p53 as a determinant of chemosensitivity of ovarian carcinoma.