Increased mRNA levels of xeroderma pigmentosum complementation group B (XPB) and Cockayne's syndrome complementation group B (CSB) without increased mRNA levels of multidrug-resistance gene (MDR1) or metallothionein-II (MT-II) in platinum-resistant human ovarian cancer tissues.

Tumor tissue specimens from human ovarian cancer patients were assessed for relative mRNA abundance levels of several genes thought to be involved in the development of in vitro drug resistance in this disease. Higher mRNA levels of Xeroderma pigmentosum group B (XPB), which links DNA repair with DNA transcription, and of Cockayne's syndrome group B (CSB), which is essential for gene-specific repair, were observed in tumor tissues that were clinically resistant to platinum-based chemotherapy, as compared with tissues from patients responding to therapy. In a cohort of 27 patients, mRNA levels of XPB averaged 5-fold higher in platinum-resistant tumors (P = 0.001); and for CSB, mRNA levels averaged 6-fold higher but with greater variability (P = 0.033). Concurrently, these platinum-resistant tumor tissues did not exhibit significantly higher mRNA levels for the MDR1 (multidrug-resistance) gene (P = 0.134) or of the metallothionein-II (MT-II) gene (P = 0.598). Since these platinum-resistant tumors also show higher mRNA levels of ERCC1 and XPA, platinum resistance appears to be associated with concurrent up-regulation of four genes (XPA, ERCC1, XPB, and CSB). These four genes participate in DNA damage excision activity, gene-specific repair, and linkage of DNA repair with DNA transcription. These data suggest that concurrent up-regulation of genes involved in nucleotide excision repair may be important in clinical resistance to platinum-based chemotherapy in this disease.

Evidence for in the appearance of mRNAs of nucleotide excision repair genes, in human ovarian cancer tissues.

Nucleotide excision repair (NER) is the DNA repair pathway through which cisplatin-DNA intrastrand adduct is repaired. Clinical studies have shown that increased mRNA expression of selected genes involved in the rate-limiting step of NER, appear to be closely associated with clinical resistance to platinum agents. These specific studies have led to the possibility of an assessment of the order in which, the appearance of mRNAs of selected NER genes may occur. Included in this assessment are ERCC1, XPB, CSB, and XPA, studied in 28 ovarian cancer tumor tissue specimens. The study of these four genes, in pairs, from 28 ovarian cancer specimens, results in 168 separate observations. Based on the mRNA expression patterns of these genes in these tissues, it is suggested that ERCC1 mRNA may appear before the mRNA of any of the other genes, in an obligate fashion. This is followed by XPB mRNA; which appears before the mRNA of XPA; which in turn, may appear before CSB. This pattern is consistent with what we have reported previously, in non-malignant human bone marrow specimens from a cohort of 52 patients.

Alternative splicing of ERCC1 and cisplatin-DNA adduct repair in human tumor cell lines.

Alternative splicing is a common natural tool for the inhibition of function of full length gene products. We explored whether there was evidence that alternative splicing of ERCC1 may serve such a function for nucleotide excision repair. The ratio of alternatively spliced species to full length species was assessed for the protein and/or for the mRNA, for a series of human cell lines and tissues. This ratio was plotted against the amount of cisplatin-DNA adduct repair in each cell line (n=9), as measured by atomic absorbance spectrometry. As the percentage of alternatively spliced protein and/or mRNA increased, the amount of cisplatin-DNA adduct that was repaired was reduced. This inverse relationship was associated with a substantial amount of scatter (r=0.635), particularly at low levels of repair. These data demonstrate an association between alternative splicing of ERCC1, and reduction in cellular capability to repair cisplatin-DNA adduct.

Comparative analyses of relative ERCC3 and ERCC6 mRNA levels in gliomas and adjacent non-neoplastic brain.

Nucleotide excision repair (NER) is an ordered process in nonmalignant cells, in both human and nonhuman systems. We previously reported that in human brain there is discordant mRNA expression of excision repair cross-complementing (ERCC) 1 and ERCC2 in malignant tissues, concurrent with excellent concordance of these genes in nonmalignant tissues from the same patients. Here we have extended these studies to compare low-grade tumors to high-grade tumors and to include ERCC3 (which links DNA repair with DNA transcription) and ERCC6 (which is essential for gene-specific repair). Glial tumor and adjacent normal brain specimens from 19 individuals were studied. Paired malignant and nonmalignant tissues were obtained from 12 of these patients. For ERCC3, there was excellent concordance of mRNA expression between malignant and nonmalignant tissues from the same individuals (P = 0.003). For ERCC6, no concordance was observed (P = 0.314). Tumor tissue from patients with high-grade gliomas exhibited marked discordance of mRNA expression patterns in situations in which good concordance was observed in tumor tissue from low-grade gliomas. We previously established that malignant brain tumors show increased disorder of genes in the NER process, as compared with nonmalignant tissues. These data suggest that increasing disorder in the NER process may occur as cells move from low-grade to high-grade malignancy.

Platinum-sensitive and platinum-resistant ovarian cancer tissues show differences in the relationships between mRNA levels of p53, ERCC1 and XPA.

We investigated the association between p53 mRNA expression and clinically relevant surrogates of nucleotide excision repair (ERCC1 and XPA) in 28 ovarian cancer specimens. We observed that platinum-resistant tumors showed higher mRNA levels of p53, ERCC1, and XPA than platinum-sensitive tumors; mRNA expression patterns in responders differed substantially from nonresponders; and p53 expression showed a strong correlation with the expression of ERCC1, and of XPA in platinum-sensitive tumor tissues, but not with platinum-resistant tumors. 47% of the mutations from p53 sequence analysis were not related to clinical response to chemotherapy. We conclude that the p53 influence on DNA repair in human malignancy may vary substantially from tumor to tumor, and that such differences are not necessarily related to the mutational status of p53.

Effects of interleukin-1 alpha on DNA repair in human ovarian carcinoma (NIH:OVCAR-3) cells: implications in the mechanism of sensitization of cis-diamminedichloroplatinum(II).

The cytokine interleukin-1 alpha (IL-1 alpha) showed a cytostatic effect on human ovarian carcinoma cells and significantly enhanced the antiproliferative activity of cis-diamminedichloroplatinum(II) (cisplatin) toward the NIH:OVCAR-3 tumor cell line in culture. The factor of sensitization was 15-20-fold. The maximum levels of sensitization were observed both with simultaneous exposure to cisplatin and IL-1 alpha and with 24-hr pretreatment with IL-1 alpha. Synergy between these agents was diminished when cells were pretreated with an IL-1 alpha-specific receptor antagonist, indicating that synergistic interaction was receptor mediated. Using atomic absorption spectroscopy, we evaluated the cellular accumulation of cisplatin and the DNA platination; the results showed that IL-1 alpha increased cellular accumulation of cisplatin and DNA platination. Cisplatin did not affect IL-1 alpha accumulation in NIH:OVCAR-3 cells. Further studies showed that IL-1 alpha reduced the removal of platinum from DNA. These results strongly suggest that IL-1 alpha inhibits DNA repair, and this decrease in DNA repair may explain, in part, the strong synergistic interaction between IL-1 alpha and cisplatin in NIH:OVCAR-3 cells.

Paclitaxel, cisplatin, and cyclophosphamide in human ovarian cancer: molecular rationale and early clinical results.

The three most active types of agents in the treatment of cancer of the ovary are platinum compounds (cisplatin or carboplatin), bifunctional alkylating agents (cyclophosphamide, melphalan, etc), and the recently developed natural product paclitaxel (Taxol; Bristol-Myers Squibb Company, Princeton, NJ). In an effort to improve long-term disease-free survival in patients with advanced disease, we have developed a three-drug regimen consisting of cisplatin, paclitaxel, and cyclophosphamide. Granulocyte-colony stimulating factor is given as bone marrow support using a flexible dosing approach. The molecular basis for this approach is founded in the positive molecular interactions between cisplatin and bifunctional alkylating agents and between paclitaxel and DNA-damaging agents. Cisplatin and cyclophosphamide damage DNA by two very different mechanisms; the respective importance of DNA strand "kinking" versus DNA strand "cross-linking" may explain the positive cell kill interaction between these two drugs. Paclitaxel appears to markedly slow the repair of DNA lesions caused by DNA-damaging agents, which may include radiation-induced lesions, cisplatin adducts, and cyclophosphamide cross-links. Clinical data to date from several different groups strongly suggest that these molecular interactions translate into positive clinical benefit in human ovarian cancer. Preliminary data from our clinical trial show that these three agents are well tolerated in the doses administered and that this combination shows exceptional promise as a possible therapeutic advance in this disease.

Malignant and nonmalignant brain tissues differ in their messenger RNA expression patterns for ERCC1 and ERCC2.

Perturbation of the DNA repair process appears to be responsible for the occurrence of a number of human diseases, which are usually associated with a propensity to develop internal malignancies and/or disorders of the central nervous system. We have been interested in the possibility that a subtle abnormality in DNA repair competency might be associated with the transformation of nonmalignant cells to the malignant state. To study this question, we assayed malignant and nonmalignant brain tissues from 19 individuals for mRNA expression levels of the human DNA repair genes ERCC1, ERCC2, and XPAC and for differential splicing of the ERCC1 transcript. We separately compared expression levels of these genes in the following situations: concordance of expression within malignant tissues; concordance of expression within nonmalignant tissues; concordance between malignant and nonmalignant tissues within individuals of the cohort; and concordance of gene expression between two nonmalignant tissue sites within a single individual. Linear regression analyses of mRNA values obtained suggested orderly concordance of these three DNA repair genes in nonmalignant tissues within the patient cohort and an excellent concordance of these genes between two separate biopsy sites from the same individual. In contrast, malignant tissues showed disruption of concordance between the full-length ERCC1 transcript and ERCC2, which have excision and helicase functions, respectively. Furthermore, within the same individuals, malignant tissues were discordant with nonmalignant tissues for ERCC1 and ERCC2, although concordance for XPAC was preserved. These data suggest that one molecular characteristic of human malignancy may be the disruption of the normal relationship between the excision and the helicase functions of the nucleotide excision repair pathway.

Expression of an alternatively spliced ercc1 messenger-RNA species, is related to reduced DNA-repair efficiency in human T-lymphocytes.

We have previously shown that in non-drug-selected human T lymphocytes, DNA repair is the primary determinant of cellular resistance to cisplatin (1). In this system, we have assessed mRNA levels of expression of the nucleotide excision repair (NER) genes ERCC1 and XPA, as well as the alternatively spliced species of ERCC1 which lacks exon VIII. The focus of these studies, was to try to identify the possible relative roles of normal XPA, full-length ERCC1, and alternatively spliced ERCC1, in a system where DNA repair is a clear determinant of cisplatin resistance. ERCC1 expression was directly related to cisplatin-DNA adduct repair capability, as well as directly related to cisplatin resistance, suggesting a primary role for ERCC1 in effecting DNA repair. XPA expression was approximately equivalent in each cell line, regardless of the level of DNA repair activity, suggesting a helper role for the product of this gene. The mRNA levels of the alternatively spliced species of ERCC1 were strongly inversely related to DNA repair activity, suggesting a possible inhibitory influence on the DNA repair process. This interpretation is consistent with alternative splicing of several known oncogenes, where the alternatively spliced species has an inhibitory effect on the full-length gene product. The NER pathway appears to be vitally important in effecting cisplatin resistance in non-drug-selected T lymphocytes. Further, it appears that NER may have at least one inhibitory regulatory component.

Messenger RNA levels of XPAC and ERCC1 in ovarian cancer tissue correlate with response to platinum-based chemotherapy.

Nucleotide excision repair is a DNA repair pathway that is highly conserved in nature, with analogous repair systems described in Escherichia coli, yeast, and mammalian cells. The rate-limiting step, DNA damage recognition and excision, is effected by the protein products of the genes ERCC1 and XPAC. We therefore assessed mRNA levels of ERCC1 and XPAC in malignant ovarian cancer tissues from 28 patients that were harvested before the administration of platinum-based chemotherapy. Cancer tissues from patients whose tumors were clinically resistant to therapy (n = 13) showed greater levels of total ERCC1 mRNA (P = 0.059), full length transcript of ERCC1 mRNA (P = 0.026), and XPAC mRNA (P = 0.011), as compared with tumor tissues from those individuals clinically sensitive to therapy (n = 15). In 19 of these tissues, the percentage of alternative splicing of ERCC1 mRNA was assessed. ERCC1 splicing was highly variable, with no difference observed between responders and nonresponders. The alternatively spliced species constituted 2-58% of the total ERCC1 mRNA in responders (median = 18%) and 4-71% in nonresponders (median = 13%). These data suggest greater activity of the DNA excision repair genes ERCC1 and XPAC in ovarian cancer tissues of patients clinically resistant to platinum compounds. These data also indicate highly variable splicing of ERCC1 mRNA in ovarian cancer tissues in vivo, whether or not such tissues are sensitive to platinum-based therapy.

Platinum-DNA adduct in leukocyte DNA of a cohort of 49 patients with 24 different types of malignancies.

Using atomic absorbance spectrometry with Zeeman background correction, we measured platinum-DNA adduct levels in leukocyte DNA of 49 patients receiving therapy consisting of only carboplatin and cisplatin. Twenty-four histological types of malignancy were included in the cohort. Peripheral blood leukocytes were collected at defined times during the first two cycles of treatment. The relationship between adduct level and disease response was highly statistically significant during cycle 1 of therapy (two-sided P = 0.007 at day 2), but statistical significance was lost during cycle 2. On all days studied, median and mean adduct levels were consistently higher in responders as compared to nonresponders (summary two-sided P = 0.0004). These data suggest that the processes which protect cellular DNA may be common to malignant and nonmalignant rapidly dividing tissues of the same individual, regardless of the type of tumor that individual may harbor.

Expression of excision repair genes in non-malignant bone marrow from cancer patients.

The patterns of expression of 3 human DNA-repair genes (ERCC1, ERCC2, ERCC6) were assessed in 52 bone-marrow specimens obtained from cancer patients prepared for autologous bone-marrow transplantation. Marrow was collected prior to the initiation of treatment in patients with sarcoma or testicular cancer; marrow was collected after initial cytoreductive therapy for patients with non-Hodgkin's lymphoma, Hodgkin's disease, and other tumors. Slot-blot analysis of marrow RNA showed a bimodal pattern of ERCC1, ERCC2 and ERCC6 gene expression with relative expression values ranging more than 200-fold. This pattern was seen in all patient groups and appeared to be independent of whether or not patients had received prior chemotherapy. In all patient groups, when expression was low for ERCC1, expression was also low for ERCC2 and ERCC6, suggesting that expression of these genes may be coordinated within an individual although they are located on two different chromosomes. Southern blot analyses of Pst I digests of DNA from 6 bone-marrow samples indicate no differences in ERCC1 gene copy number between high expressors and low expressors. There is absence of restriction fragment length polymorphism for ERCC1 suggesting that the different levels of expression in high and low expressors were not due to major deletions or rearrangements of the ERCC1 gene. We conclude that expression of these ERCC genes may vary widely between individuals, and that within an individual, their expression may be linked and coordinated by a common regulatory mechanism.

Taxol effect on cisplatin sensitivity and cisplatin cellular accumulation in human ovarian cancer cells.

Taxol and cisplatin are the two most effective agents discovered to date for treating advanced-stage cancer of the ovary. Learning how best to combine these agents is the focus of preclinical and clinical studies conducted at a number of institutions. Taxol's effect on cellular sensitivity to cisplatin was studied in paired cisplatin-sensitive A2780 and cisplatin-resistant A2780/CP70 human ovarian cancer cell lines. Cisplatin growth curves were generated under conditions of specific sequencing with Taxol, and IC50s (concentrations at which growth is inhibited to 50% of control) for cisplatin were obtained and compared. Taxol was used at an IC10 dose in all experiments. Taxol treatments were for 24 hours and cisplatin treatments were for 1 hour in all experiments. Dimethyl sulfoxide (DMSO) was the diluent for all Taxol stock solutions. Separately, the effects of Taxol and DMSO on cisplatin cellular accumulation were measured. End points reported include measures of cytotoxicity and Taxol effects on cisplatin cellular accumulation. Using a microculture tetrazolium assay, cisplatin growth curves were obtained under the influence of Taxol, at a Taxol dose of 3 nM for both cell lines. DMSO alone had no effect on tumor cell growth. In A2780 cells, the influence of Taxol on cisplatin cytotoxicity was modest, whereas cisplatin-induced cell kill was augmented 1.5-fold when cisplatin was given immediately after Taxol. In A2780/CP70 cells, Taxol augmented cisplatin-induced cell kill by 30-fold when cisplatin was given immediately after Taxol; 75-fold when cisplatin was given 24 hours after completion of Taxol; and 19-fold when cisplatin was given 48 hours after completion of Taxol.(ABSTRACT TRUNCATED AT 250 WORDS)

Platinum-DNA adduct in head and neck-cancer patients receiving Cisplatin and Carboplatin chemotherapy.

Sixteen patients with squamous cell carcinoma of the head and neck region were studied for the formation of platinum-DNA adduct in leukocyte DNA. Every patient received carboplatin and cisplatin, each given intravenously on day 1 of 21-day treatment cycles. Blood was obtained for DNA isolation 24 hours after treatment on cycles 1 and 2; and on the morning before cycle 2 of therapy (21 days after cycle 1 drug delivery). Adduct was measured by atomic absorbance spectrometry with Zeeman background correction. As has been shown in previous studies, adduct persistence and accumulation could be demonstrated in a portion of the cohort. Adduct removal from blood cell DNA during the 21 days of cycle number 1, was determined for 12 patients. In these patients, adduct removal was directly i-elated to clinical resistance to the cisplatin/carboplatin treatment regimen; and adduct removal was inversely related to leukocyte toxicity. These data demonstrate that changes in adduct levels with time may provide valuable information, in the study of the human in vivo biology of platinum-DNA damage and repair.

Cisplatin-DNA damage and repair in peripheral blood leukocytes in vivo and in vitro.

We have extended our studies on the relationship between cisplatin/carboplatin-induced DNA damage in readily accessible tissue(s) and clinical response to therapy. Such an approach may assist in the study of cancer drug resistance and in establishing parameters for assessing human populations for sensitivity to DNA damaging agents in the environment. Platinum-DNA adduct levels were measured by atomic absorbance spectrometry. DNA repair capacity was assessed in human T-lymphocytes by the ability to repair cisplatin lesions in cellular DNA or in transfected plasmid DNA. In a "blinded" study of 21 patients receiving combination cisplatin/carboplatin drug therapy, there was a direct relationship between DNA damage in leukocytes and disease response (summary two-sided p = 0.00011). The cohort of patients had 15 different tumor types, suggesting that blood tissue and tumor tissue of an individual may process platinum-DNA damage similarly regardless of the tissue of origin of the tumor. In leukocytes in vivo, persistence and accumulation were prominent features of the cisplatin-DNA adduct profile. Functional DNA repair capacity has been studied in eight human leukocyte cell lines in vitro (three, T-cells; three, B-cells; one, monocytic; one, promyelocytic), using a host cell reactivation assay with cisplatin-damaged pRSVcat. In the three T cell lines studied, host cell reactivation efficiency was directly related to the cells' abilities to repair cisplatin-damaged cellular DNA (correlation coefficient = 0.993).(ABSTRACT TRUNCATED AT 250 WORDS)

ERCC1 and ERCC2 expression in malignant tissues from ovarian cancer patients.

BACKGROUND: ERCC1 and ERCC2 are human DNA repair genes that are associated with in vitro resistance to selected DNA-damaging agents. PURPOSE: Fresh tumor tissues from 26 patients with ovarian cancer were analyzed for the RNA levels of expression of these genes to determine possible clinical relevance. METHODS: Tumor tissues were harvested from patients immediately before they entered a cisplatin- or carboplatin-based treatment protocol. Clinical response was assessed by standard criteria. Gene expression level was assessed by slot blot analysis, using beta-actin as a control. Relative expression levels were determined by comparing each tumor sample with a Chinese hamster ovary cell line that had a stable transfection of the human ERCC1 gene. RESULTS: Patients who were clinically resistant to platinum-based therapy had a 2.6-fold higher expression level of ERCC1 in their tumor tissue than did patients who responded to that therapy (P = .015). Results obtained by slot blot analysis were qualitatively confirmed by polymerase chain reaction analysis. Relative levels of expression of ERCC2 did not differ significantly between responders and nonresponders. CONCLUSION: We conclude that ERCC1 expression levels in human tumor tissue may have a role in clinical resistance to platinum compounds. These data appear to be consistent with the assertion that ERCC1 serves as an excision nuclease, whereas ERCC2 serves as a helicase.

Determinants of cisplatin sensitivity in non-malignant non-drug-selected human T cell lines.

We have studied molecular mechanisms of cisplatin sensitivity and resistance in 3 non-malignant, non-drug-selected human T lymphocyte cell lines. HuT 78, H9, and MOLT-4 cells were assessed for sensitivity to cisplatin, DNA damage levels following defined drug exposures, drug accumulation, and DNA repair efficiency as measured by adduct removal from cellular DNA and by host-cell reactivation of cisplatin-modified plasmid DNA. Based on 3-day continuous drug exposures, the IC50 values for the cell lines were: HuT 78, 0.83 microM; H9, 0.45 microM; and MOLT-4, 0.33 microM. These cells retained this order with respect to DNA repair capability, whether measured by platinum-DNA adduct removal from cellular DNA or by host-cell reactivation assays. DNA repair values measured by these two assays were directly related to one another with a linear correlation coefficient of 0.993. At sublethal cisplatin doses the more resistant cells showed the highest levels of drug uptake. When drug uptake levels were 'corrected' for drug-induced cell kill, there were equal levels of DNA repair efficiency for a given level of drug uptake. Absolute levels of cisplatin-DNA adduct repair increased with increasing drug dose. However, at supralethal doses of drug, efficient DNA repair could be overcome in all 3 cell lines with percentage-adduct-removal dropping from a 60-80% range to a less than 30% range. We conclude that in non-malignant non-drug-selected human T cells, DNA repair appears to be the primary determinant of cisplatin sensitivity/resistance and that enhanced DNA repair may be a biologic compensatory mechanism for cells that cannot prevent cellular uptake of DNA-damaging agents.