Publication of the International Union Against Cancer Publication de l’Union Internationale Contre le Cancer Int. J. Cancer: 70, 551–555 (1997) r 1997 Wiley-Liss, Inc. DNA REPAIR CAPACITY AND CISPLATIN SENSITIVITY OF HUMAN TESTIS TUMOUR CELLS Beate KÖBERLE1, Keith A. GRIMALDI2, Andrew SUNTERS2, John A. HARTLEY2, Lloyd R. KELLAND3 and John R.W. MASTERS1* 1Institute of Urology and Nephrology, London, UK 2CRC Drug-DNA Interactions Research Group, Department of Oncology, University College London, London, UK 3CRC Centre for Cancer Therapeutics, Institute of Cancer Research, Sutton, Surrey, UK Metastatic testicular germ cell tumours can be cured using cisplatin-based combination chemotherapy. To investigate the role of DNA repair in cisplatin sensitivity, we measured the formation and removal of cisplatin adducts in the whole genome and in specific genomic regions in 3 testis and 3 bladder cancer cell lines. Following a 1 hr exposure to 50 µM cisplatin, the mean level of DNA platination was lower in the testis tumour cell lines. During a 72 hr post-treatment incubation period, the 3 bladder cell lines removed platinum from the overall genome, whereas 2 of the testis tumour cell lines showed relatively little reduction of DNA platination. The third testis tumour cell line, SuSa, showed an intermediate capacity to remove cisplatin. Cisplatin-induced damage and repair in selected regions of the actively transcribed N-ras gene and the inactive CD3d gene were measured using quantitative PCR. The data were in agreement with those obtained with atomic absorption spectroscopy for the whole genome, showing that the bladder lines were repair-proficient: 2 of the testis tumour cell lines showed no repair, and the third testis line, SuSa, showed an intermediate level of repair in these 2 genes. Our findings confirm that reduced capacity to repair cisplatin-damaged DNA may contribute to the hypersensitivity of testis tumour cells to DNA-damaging agents. Int. J. Cancer 70:551–555, 1997. r 1997 Wiley-Liss, Inc. Metastatic testicular germ cell tumours are cured in over 80% of patients using cisplatin-based combination chemotherapy (Einhorn, 1990). In contrast, most other types of metastatic cancer rarely are cured using chemotherapy. Therefore, the sensitivity of testicular germ cell tumours to chemotherapy may have more general relevance as it may provide clues to improving therapy for other types of cancer. Cell lines derived from testis tumours are also more sensitive to chemotherapeutic drugs than those derived from other types of cancer, indicating that these cells retain their relative sensitivity to DNA-damaging agents in vitro, reflecting the clinical response. We have compared the sensitivity of cell lines derived from testicular germ cell tumours with those derived from bladder cancers. Cisplatin is also the most effective single agent for bladder cancer, but only 40–50% of metastatic bladder cancers respond to combination chemotherapy and long-term survival is rare (Seidman and Scher, 1991). Testis tumour cell lines are more sensitive than bladder cancer cells to the cytotoxic effects of irradiation (Parris et al., 1988) and many chemotherapeutic drugs (Walker et al., 1987; Masters et al., 1993; Pera et al., 1995). Cells derived from certain DNA repair disorders are also more sensitive to chemotherapeutic drugs and ionising radiation. Testis tumour cells are similar to cells derived from patients with xeroderma pigmentosum (XP) and ataxia telangiectasia in their sensitivities to UV and gamma radiation (Pera et al., 1987) and to excision repair-deficient rodent cell mutants in their sensitivity to cisplatin (Lee et al., 1993). In keeping with this analogy, studies on repair in the whole genome indicate that some testis tumour cell lines are repair-deficient (Bedford et al., 1988; Hill et al., 1994; Sark et al., 1995; Kelland et al., 1992). This deficiency may contribute to their sensitivity to DNA-damaging agents. The rate of DNA repair is not uniform throughout the genome. There are 2 repair pathways: one deals with the rapid and efficient removal of lesions from transcribed genes, while the other accom- plishes the slower and less efficient repair of bulk DNA (Bohr, 1991; Link et al., 1991). It has been suggested that the repair of certain essential genes may be more critical for survival than repair of the whole genome (Bohr et al., 1986). In repair-deficiency disorders the defect can affect either one or both of the repair pathways. In XP group C the repair defect is limited to the overall genome repair (Venema et al., 1990a), whereas in Cockayne syndrome cells the repair defect is limited to preferential genespecific repair (Venema et al., 1990b). The published descriptions of DNA repair capacity in testis tumour cells (Bedford et al., 1988; Hill et al., 1994; Sark et al., 1995; Kelland et al., 1992) are restricted to repair in the whole genome. The aim of our study was to extend these studies to investigate the potential contribution of repair in active and inactive genes. To confirm the results on DNA repair in the whole genome, we compared DNA repair after treatment with cisplatin in 3 testis tumour cell lines with that in 3 bladder tumour cell lines in the whole genome using atomic absorption spectroscopy. To investigate repair in selected genes, we measured damage induction and repair in defined regions of the actively transcribed N-ras gene and the inactive CD3d gene using quantitative (Q-PCR) (Grimaldi et al., 1994). MATERIAL AND METHODS Cell culture and drug treatment Details of the origins of the cell lines have been described previously (Masters et al., 1993). All cell lines were grown routinely as monolayers in 25 cm2 tissue culture flasks in RPMI 1640 medium supplemented with 10% heat-inactivated FCS and 2 mM 1-glutamine at 36.5°C in a humidified atmosphere of 5% CO2 in air. Each cell line was used over a maximum of 15 passages to minimize changes that might occur during long-term culture. Cisplatin (Sigma, Poole, UK) was prepared immediately before use by dissolving in water to a concentration of 1 mg/ml and sterilization through a 0.22 µM filter. Cisplatin sensitivity measurements Cytotoxicity was determined by plating 2,500 cells (testis tumour cell lines) or 750 cells (bladder tumour cell lines) in 60 3 15 mm Petri dishes containing 5 ml medium. Four dishes were set up for untreated controls and each concentration of cisplatin. Following incubation overnight at 36.5°C, cells were treated with a range of concentrations of cisplatin for 1 hr, then washed twice with PBS and incubated in fresh medium for 10–14 days. Colonies were fixed with methanol, stained with 10% Giemsa and those consisting of 50 or more cells counted. Inhibition of colony formation in Contract grant sponsor: the Cancer Research Campaign; contract grant number: SP 1997. *Correspondence to: Institute of Urology and Nephrology, University College London, 67 Riding House Street, 3rd Floor, London W1P 7PN, UK. Fax: 004401716377076. E-mail: Regnjrm@Ucl.ac.uk Received 19 August 1996. 552 KÖBERLE ET AL. treated dishes was expressed as a percentage of colony formation in untreated controls. Measurement of platinum binding to DNA Between 5 3 107 and 108 cells were grown to near confluence in 175 cm2 culture flasks, exposed to cisplatin (50 µM) for 1 hr and harvested either immediately after drug exposure or after recovery periods of 24, 48 and 72 hr. DNA was isolated by treatment with proteinase K, 0.5% SDS and lysis buffer (150 mM NaCl, 10 mM Tris [pH 8.0], 10 mM EDTA [pH 8.0]) overnight at 37°C. After phenol/chloroform extraction (23) and chloroform extraction (13), the DNA solution was treated with RNase (10 mg/ml) at 37°C for 1 hr followed by phenol/chloroform (13) and chloroform extraction (13). DNA was precipitated with a 1/10 vol of sodium acetate (3 M) and 2 vol of cold ethanol (absolute), washed twice with ethanol (70%), dried and dissolved in 0.1 M nitric acid. The amount of platinum bound to DNA was measured by atomic absorption spectroscopy and the amount of DNA determined by measuring the absorbance at 260 nm. To correct for dilution due to DNA synthesis during the recovery period, cells were labelled with 50 nCi/ml [ 14C]thymidine for 24 hr before drug treatment. Dilution factors (specific activity of DNA at time point/specific activity of DNA at 0 hr) were then determined for the 24, 48 and 72 hr repair periods. The reported values are the mean of up to 6 independent experiments corrected for dilution due to DNA synthesis. Damage and repair of cisplatin-induced lesions in specific genes Cell culture and drug treatment were performed in 6-well plates (Nunc, Roskilde, Denmark): 106 cells were treated with a range of concentrations of cisplatin for 5 hr, after which cells were washed with PBS and either trypsinised for DNA isolation or incubated for another 24 hr in fresh medium for repair studies. DNA isolation and Q-PCR were performed according to Grimaldi et al. (1994). Briefly, cells were lysed with a solution composed of 340 µl lysis buffer (400 mM Tris-HCl [pH 8.0], 60 mM EDTA, 150 mM NaCl, 1% w/vol SDS) and 100 µl 5 M sodium perchlorate. After vortexing, the cell lysate was incubated on a shaking platform at 37°C for 20 min and then at 65°C for 20 min; 580 µl of chloroform were added to the solution and the mixture was rotated for 20 min at room temperature and centrifuged at 11,500g in a microfuge; 330 µl of the upper layer were transferred to a separate Eppendorf tube and the DNA precipitated with 660 µl ethanol (absolute), washed twice with ethanol (70%), dried and resuspended in 400 µl H2O. Then, PCR was performed with 25 µl DNA suspension (equivalent to 5 3 104 cells) in a 100 µl reaction mixture containing 50 pmoles of each primer (Pharmacia, Uppsala, Sweden); 2 units Red Hot DNA polymerase (Advanced Biotechnologies, Columbia, MD); 120 µM each dATP, dCTP, dGTP and dTTP (Pharmacia); 0.75 mM MgCl2, buffer IV (Advanced Biotechnologies); and 1 µCi (a32P)dCTP (Amersham, Aylesbury, UK). The CD3d gene fragment was amplified using the forward primer 58-TGA GGA CAG AGT GTT TGT GAA-38 and the reverse primer 58-AGA GTA ACT CCC AGC TGA GAC-38. The amplification product was 934 bp in length and contained part of exon 2, intron 2 and exon 3. Amplification of the N-ras gene, which gave a 1,053 bp fragment covering the first intron, was performed with the forward primer 58-GCC TGG TTA CTG TGT CCT GT-38 and the reverse primer 58-GCC AGC CAC ATC TAC AGT AC-38. DMSO (5%) was added to the reaction mixture for amplification of the fragment of the N-ras gene. The mixture was overlaid with 100 µl mineral oil. PCR was carried out in a Perkin-Elmer 480 Thermal Cycler (Norwalk, CT) under the following conditions: the CD3d gene fragment was amplified after an initial denaturation step of 5 min at 96°C by 25 cycles of 96°C for 1 min, 60°C for 1 min and 70°C for 1 min, followed by a final incubation of 4 min at 70°C at the end of the cycling; the N-ras fragment was amplified after an initial denaturation step of 2 min at 94°C, followed by 25 cycles of 94°C for 1 min, 59°C for 1 min and 72°C for 1 min, with a final incubation of 4 min at 72°C at the end of the cycling. Preliminary experiments ensured that under these conditions the PCR was still in exponential phase when the PCR reaction stopped. The PCR product was measured by precipitating 40 µl of PCR mixture with 1 ml TCA (5% w/v TCA, 20 mM tetrasodium pyrophosphate). The precipitate was captured on Whatman (Maidstone, UK) filter discs held in a 12 position vacuum manifold (Millipore, Bedford, MA). Filters were washed free of unincorporated a-32P-dCTP with 10 ml TCA (5%) and 10 ml absolute ethanol, placed in vials containing 5 ml scintillation fluid (Ecoscint; National Diagnostics, Manville, NJ) and counted on a Beckman (Palo Alto, CA) LS1800 scintillation counter. The lesions per region were determined according to the formula: lesions per region 5 2ln (Ad/A), where A is the PCR product from the undamaged template and Ad is the product from the damaged template. To demonstrate that the PCR reaction was specific, resulting in only 1 product, 10 µl of PCR product were subjected to electrophoresis on a 1.5% agarose gel overnight at 42V and the gel was dried and exposed to autoradiographic film (Kodak, Rochester, NY), to visualize the band. RESULTS Survival curves of the 3 testis (SuSa, 833K, GCT27) and 3 bladder (MGH-U1, RT112, HT1376) tumour cell lines following a 1 hr exposure to a range of concentrations of cisplatin are shown in Figure 1. Comparing the mean IC50 values, the testis tumour cell lines were on average 4.1-fold more sensitive to cisplatin than the bladder tumour cell lines (Table I). The amount of platinum bound to DNA directly after exposure to cisplatin (50 µM) for 1 hr is shown in Table I. There was no significant difference in the initial platination in the 3 testis tumour cell lines and one of the bladder cancer cell lines (MGH-U1). However, the other 2 bladder cancer cell lines contained approximately twice the levels of DNA platination of the testis tumour cell lines. Removal of platinum from the DNA of the 3 testis and the 3 bladder tumour cell lines is shown in Figure 2. The amount of DNA platination of the 3 bladder tumour cell lines was reduced by 50–60% within 24 hr. In contrast, 2 of the 3 testis tumour cell lines FIGURE 1 – Clonogenic cell survival curves of testis tumour cell lines (open symbols) and bladder cancer cell lines (closed symbols) after exposure to cisplatin for 1 hr. DNA REPAIR IN TESTIS TUMOUR CELLS 553 TABLE I – CISPLATIN SENSITIVITY AND INITIAL PLATINATION OF TESTIS AND BLADDER CANCER CELL LINES FOLLOWING A 1 HR EXPOSURE TO CISPLATIN Cell line Testis SuSa 833K GCT27 Bladder MGH-U1 RT112 HT1376 IC50 (µM) DNA platination (nmol Pt/gDNA) 5.6 6 0.3 4.3 6 0.4 6.3 6 0.9 41.8 6 14.9 55.1 6 17.7 67.0 6 19.7 14.2 6 1.7 16.1 6 2.9 20.1 6 3.9 48.3 6 10.4 120.4 6 17.5 108.1 6 33.2 (833K and GCT27) showed less removal of platinum (approximately 30%) during the post-incubation time (Fig. 2). The third testis line, SuSa, removed platinum at an intermediate rate at 24 hr and to the same extent as the bladder cancer cells at 48 and 72 hr. To investigate damage induction and repair in single-gene regions, cells were treated with cisplatin at concentrations ranging from 25 to 200 µM for the bladder tumour cells and from 25 to 200 µM for the testis tumour cells for 5 hr. Under these conditions, testis tumour cell lines showed minimal new DNA synthesis (,1%) at cisplatin concentrations of 25 µM (833K, SuSa) or 50 µM (GCT27) as measured by incorporation of 14C-thymidine. With the bladder tumour cell lines there was less than 5% new DNA synthesis at 50 µM (HT1376, RT112) or 100 µM (MGH-U1) cisplatin. Cells were harvested either directly after treatment or after a 24 hr recovery period. Using Q-PCR, the cisplatin-induced lesions were measured in a 934 bp fragment of the CD3d gene and a 1,053 bp fragment of the N-ras gene. Some of these data were also used in a study of the repair of cisplatin-induced DNA damage in cisplatin-sensitive parental lines compared with their cisplatin-resistant sublines (Köberle et al., 1996). Figure 3 shows an example of an autoradiograph of PCR products of the CD3d gene region of HT1376 cells directly after treatment or after a 24 hr recovery period. There is only one band, reflecting the specificity of the PCR reaction. With increasing cisplatin concentration the intensity of the band decreases, reflecting an increase in damage. This is reversed after 24 hr, showing repair of the lesion. The data for the bladder and testis tumour cell lines are summarized in Tables II and III, respectively. Comparing the 2 gene fragments, there was no significant difference in the initial frequency of lesions in the CD3d gene fragment when compared with the N-ras gene fragment in all 6 tumour cell lines examined. The repair studies showed that all 3 bladder tumour cell lines removed cisplatin-induced damage from both gene fragments. Comparing the repair rates after treatment with cisplatin at a concentration of 100 µM for 5 hr, almost complete repair of the CD3d and N-ras fragments of MGH-U1 cells after 24 hr was observed (Table II). RT112 and HT1376 showed proportionately less repair compared with MGH-U1, particularly at the higher concentrations of 150 and 200 µM cisplatin. Moreover, in both cell lines, cisplatin-induced lesions were repaired less efficiently from the transcriptionally active N-ras gene compared with the inactive CD3d gene (Table II). However, RT112 and HT1376 contained more lesions per fragment following exposure to 100 µM than MGH-U1. Therefore, when the number of lesions repaired was calculated, it became apparent that the amount of repair was almost identical for both genes in all 3 cell lines. In contrast to the bladder cancer cell lines, neither 833K nor GCT27 showed repair in the N-ras or the CD3d gene but showed an increase in damage after 24 hr (Table III). However, the other testis tumour cell line, SuSa, removed cisplatin-induced damage from the CD3d gene and, to a lesser extent, from the N-ras gene, but its repair capacity was lower than that of the 3 bladder tumour cell lines (Table III). FIGURE 2 – Removal of platinum from total genomic DNA of testis tumour cell lines (open symbols) and bladder cancer cell lines (closed symbols) measured by atomic absorption spectroscopy, expressed as fraction of that present immediately following a 1 hr exposure to cisplatin (50 µM). FIGURE 3 – Autoradiograph of PCR products of HT1376 cells treated with a range of cisplatin concentrations. C, untreated control; Oh, directly after treatment; 24 h, after a 24 hr recovery period. DISCUSSION Our results have shown that testis tumour cells have relatively little capacity to repair cisplatin-damaged DNA and that this deficiency can be detected at the level of specific genes. Following cisplatin exposure, the levels of DNA platination were the reverse of what might be expected. DNA platination was, on average, lower in the testis tumour cells, despite their greater sensitivity to cisplatin. This lack of correlation between cisplatin sensitivity and the level of DNA platination has previously been observed in ovarian cancer cell lines (Johnson et al., 1994; Shellard et al., 1991), lung carcinoma cell lines (Shellard et al., 1993), colon carcinoma cell lines (Sark et al., 1995) and testis tumour cell lines (Pera et al., 1987; Hill et al., 1994; Bedford et al., 1988). DNA damage does not appear to be associated with inherent sensitivity to cisplatin. However, cisplatin-resistant sublines of ovarian cancer cells (Zhen et al., 1992) and testis tumour cells (Timmer-Bosscha et al., 1993) showed less initial DNA platination than their parental lines. One of the testis tumour cell lines (SuSa) showed overall genome repair but at a decreased efficiency compared to the bladder tumour cell lines. This suggests that, although DNA repair deficiency might contribute to the hypersensitivity of testis tumour cell lines, this might not be the sole mechanism. Our findings, however, are in direct contrast to previous results, where SuSa was KÖBERLE ET AL. 554 TABLE II – NUMBER OF LESIONS IN SELECTED REGIONS OF THE CD3 GENE AND N-ras GENE IN BLADDER TUMOUR CELL LINES AFTER TREATMENT WITH CISPLATIN CD3 N-ras MGH-U1 lesions/kg RT112 lesions/kb HT1376 lesions/kb Cisplatin (µM) 0 hr 24 hr 0 hr 24 hr 0 hr 24 hr 25 50 100 150 200 25 50 100 150 200 0.58 6 0.37 0.80 6 0.52 1.28 6 0.38 1.81 6 0.14 2.35 6 0.14 0.35 6 0.42 0.65 6 0.52 0.90 6 0.28 1.36 6 0.32 1.69 6 0.26 0.0 6 0.0 0.0 6 0.0 0.0 6 0.0 0.24 6 0.24 0.64 6 0.29 0.0 6 0.0 0.0 6 0.0 0.02 6 0.02 0.23 6 0.32 0.56 6 0.13 0.69 6 0.22 1.28 6 0.59 1.87 6 0.07 1.81 6 0.29 2.03 6 0.14 0.42 6 0.06 0.98 6 0.37 1.56 6 0.04 1.73 6 0.19 1.92 6 0.32 0.06 6 0.01 0.32 6 0.14 0.59 6 0.16 1.81 6 0.14 2.20 6 0.14 0.09 6 0 0.42 6 0.06 0.84 6 0.18 1.64 6 0.06 2.06 6 0.39 1.17 6 0.53 1.87 6 0.97 2.84 6 0.11 3.31 6 0.0 3.10 6 0.67 0.51 6 0.46 1.15 6 0.44 1.90 6 0.44 2.21 6 0.46 2.58 6 0.59 0.33 6 0.32 0.80 6 0.22 1.60 6 0.48 2.46 6 0.52 2.99 6 0.44 0.0 6 0.0 0.40 6 0.28 1.14 6 0.37 1.69 6 0.56 2.49 6 0.59 TABLE III – NUMBER OF LESIONS IN SELECTED REGIONS OF THE CD3 GENE AND N-ras GENE IN TESTIS TUMOUR CELL LINES AFTER TREATMENT WITH CISPLATIN CD3 N-ras GCT27 lesions/kg 833K lesions/kb SuSa lesions/kb Cisplatin (µM) 0 hr 24 hr 0 hr 24 hr 0 hr 24 hr 25 50 100 150 200 25 50 100 150 200 0.16 6 0.21 0.37 6 0.07 0.38 6 0.21 1.07 6 0.14 1.49 6 0.0 0.23 6 0.32 0.51 6 0.46 1.09 6 0.28 1.31 6 0.13 1.83 6 0.19 0.67 6 0.84 1.60 6 1.28 1.07 6 1.07 2.08 6 1.58 1.71 6 0.44 0.0 6 0.0 1.17 6 0.32 1.88 6 0.18 2.06 6 0.39 2.02 6 0.06 0.56 6 0.05 0.80 6 0.33 0.98 6 0.25 1.51 6 0.19 2.14 6 0.44 0.65 6 0.52 1.26 6 0.19 1.64 6 0.37 2.03 6 0.37 2.30 6 0.19 1.42 6 0.26 1.41 6 0.19 1.73 6 0.37 2.16 6 0.32 1.97 6 0.07 1.92 6 0.46 2.39 6 0.19 2.39 6 0.23 2.53 6 0.32 2.49 6 0.06 0.18 6 0.02 0.69 6 0.22 1.19 6 0.27 1.45 6 0.34 1.81 6 0.29 0.08 6 0.01 0.47 6 0.13 1.03 6 0.09 1.31 6 0.39 1.5 6 0.39 0.32 6 0.14 0.37 6 0.07 0.74 6 0.53 1.20 6 0.53 0.58 6 0.37 0.46 6 0.38 0.56 6 0.13 0.75 6 0.09 1.12 6 0.39 0.89 6 0.19 unable to repair platinated DNA while 833K proved to be repair-proficient (Bedford et al., 1988). The identity of the lines was checked using locus-specific probes, and the SuSa cell line was shown to be different from 833K but identical to a cisplatinresistant subline of SuSa we had developed previously. We therefore have no explanation for these earlier data, which were obtained in a different laboratory. An increase in cisplatin-induced damage during a postincubation period, as observed for 833K and GCT27 at the level of the gene, has also been reported at the level of the whole genome in testis tumour cell lines (Hill et al., 1994; Sark et al., 1995). Cisplatin binds to DNA, forming mainly intrastrand cross-links and, in a slower reaction, interstrand cross-links (Eastman, 1983). The increase in damage, therefore, may be due to the continuous reaction leading to interstrand cross-links. However, as interstrand cross-links account for only 1% of cisplatin-induced damage, this does not account for all of the observed increase. The increase in damage observed at the level of the gene in 833K and GCT27 might in addition be due to degradation of DNA. This would result in less template for the PCR reaction, which would be reflected as damage. In contrast to the other testis tumour cell lines, SuSa repaired cisplatin-induced damage in both gene regions, in agreement with the data for the whole genome. The CD3d gene was repaired more efficiently compared with the N-ras gene, as was seen with the bladder tumour cell lines. Interestingly, the removal of damage was more efficient from the inactive CD3d gene than from the transcriptionally active N-ras gene. In a number of studies it has been found that actively transcribed genes are repaired faster and more efficiently than inactive genes and bulk DNA after treatment with a number of DNA-damaging agents, including cisplatin (Islas et al., 1991; Jones et al., 1991; Zhen et al., 1993). The removal of damage from actively transcribed genes has been associated with transcription coupled repair, which seems to be completed after a few hours, whereas repair in the whole genome takes up to 24 hr (Venema et al., 1992). Therefore, it is likely that a preference in repair in the actively transcribed gene would have been shown after a shorter time period. In addition, the discrepancy with published results may be explained by the fact that we investigated repair in only a fragment of the respective genes, while in the other studies repair in the whole gene (10–20 kb) was investigated using Southern blotting techniques. Alternatively, a more open structure of the CD3 gene could allow better accessibility to repair enzymes and, therefore, better repair. The fact that N-ras seems to be expressed at lower levels (Futscher and Erickson, 1990) than genes often used to study gene-specific damage and repair, such as DHFR and c-myc, may also be a contributory factor. These results indicate that the hypersensitivity of testis tumour cells to cisplatin can be related to a deficiency in DNA repair. Differential sensitivity of testis tumour cells and other types of cell to DNA-damaging agents has been associated with an increased propensity to undergo apoptosis (Lowe et al., 1993; Chresta et al., 1996). Further studies investigating which steps in the repair process are defective in testis tumour cells will help to clarify whether testis tumours die as a result of a defect in DNA repair or because of their propensity to undergo apoptosis. REFERENCES BEDFORD, P., FICHTINGER-SCHEPMAN, A.M.J., SHELLARD, S.A., WALKER, M.C., MASTERS, J.R.W. and HILL, B.T., Differential repair of platinum-DNA adducts in human bladder and testicular tumour continuous cell lines. Cancer Res., 48, 3019–3024 (1988). mammalian cells correlates with efficient DNA repair in essential genes. Proc. nat. Acad. Sci. (Wash.), 83, 3830–3833 (1986). BOHR, V., Gene specific DNA repair. Carcinogenesis, 12, 1983–1992 (1991). CHRESTA, C.M., MASTERS, J.R.W. and HICKMAN, J.A., Hypersensitivity of human testicular tumors to etoposide-induced apoptosis is associated with functional p53 and high Bax:Bcl-2 ratio. Cancer Res., 56, 1834–1841 (1996). BOHR, V., OKUMOTO, D.S. and HANAWALT, P.C., Survival of UV-irradiated EASTMAN, A., Characterization of the adducts produced in DNA by DNA REPAIR IN TESTIS TUMOUR CELLS cis-diamminedichloroplatinum(II) and cis-dichloro(ethylenediamine)platinum(II). Biochemistry, 22, 3927–3933 (1983). EINHORN, L.H., Treatment of testicular cancer: a new and improved model. J. clin. Oncol., 8, 1777–1781 (1990). FUTSCHER, B.W. and ERICKSON, L.C., Changes in c-myc and c-fos expression in a human tumour cell line following exposure to bifunctional alkylating agents. Cancer Res., 50, 62–66 (1990). GRIMALDI, K.A., BINGHAM, J.P., SOUHAMI, R.L. and HARTLEY, J.A., DNA damage by anticancer agents: mapping in cells at the subgene level with quantitative PCR. Analyt. Biochem., 222, 236–242 (1994). HILL, B.T., SCANLON, K.J., HANSSON, J., HARSTRICK, A., PERA, M., FICHTINGER-SCHEPMAN, A.M.J. and SHELLARD, S.A., Deficient repair of cisplatin-DNA adducts identified in human testicular teratoma cell lines established from tumours from untreated patients. Europ. J. Cancer, 30A, 832–837 (1994). ISLAS, A.L., VOS, J.M.H. and HANAWALT, P.C., Differential induction and repair of psoralen photoadducts to DNA in specific human genes. Cancer Res., 51, 2867–2873 (1991). JOHNSON, S.W., PEREZ, R.P., GODWIN, A.K., YEUNG, A.T., HANDEL, L.M., OZOLS, R.F. and HAMILTON, T.C., Role of platinum-DNA adduct formation and removal in cisplatin resistance in human ovarian cancer cell lines. Biochem. Pharmacol., 47, 689–697 (1994). JONES, J.C., ZHEN, W., REED, E., PARKER, R.J., SANCAR, A. and BOHR, V.A., Gene specific formation and repair of cisplatin intrastrand adducts and interstrand crosslinks in Chinese hamster ovary cells. J. biol. Chem., 266, 7101–7107 (1991). KELLAND, L.R., MISTRY, P., ABEL, G., FREIDLOS, F., LOH, S.Y., ROBERTS, J.J. and HARRAP, K.R., Establishment and characterization of an in vitro model of acquired resistance to cisplatin in a human testicular nonseminomatous cell line. Cancer Res., 52, 1710–1716 (1992). KÖBERLE, B., PAYNE, J., GRIMALDI, K.A., HARTLEY, J.A. AND MASTERS, J.R.W. DNA repair in cisplatin-sensitive and resistant human cell lines measured in specific genes by quantitative polymerase chain reaction. Biochem. Pharmacol., 52, 1729–1734 (1996). LEE, K.B., PARKER, R.J., BOHR, V., CORNELISON, T. and REED, E., Cisplatin sensitivity/resistance in UV repair deficient Chinese hamster ovary cells of complementation groups 1 and 3. Carcinogenesis, 14, 2177–2180 (1993). LINK, C.J., BURT, R.K. and BOHR, V.A., Gene specific repair of DNA damage induced by UV irradiation and cancer chemotherapeutics. Cancer Cells, 3, 427–436 (1991). LOWE, S.W., RULEY, H.E., JACKS, T. and HOUSMAN, D.E., p53-dependent apoptosis modulates the cytotoxicity of anticancer agents. Cell, 74, 957–967 (1993). MASTERS, J.R.W., OSBORNE, E.J., WALKER, M.C. and PARRIS, C.N., Hypersensitivity of human testis-tumour cell lines to chemotherapeutic drugs. Int. J. Cancer, 53, 340–346 (1993). PARRIS, C.N., ARLETT, C.F., LEHMANN, A.R., GREEN, M.H.L. and MASTERS, 555 J.R.W., Differential sensitivities to gamma radiation of human bladder and testicular tumour cell lines. Int. J. Rad. Biol., 53, 599–608 (1988). PERA, M.F., FRIEDLOS, F., MILLS, J. and ROBERTS, J.J., Inherent sensitivity of cultured human embryonal carcinoma cells to adducts of cis-diamminedichloroplatinum(II) on DNA. Cancer Res., 47, 6810–6813 (1987). PERA, M.F., KÖBERLE, B. and MASTERS, J.R.W., Exceptional sensitivity of testicular germ cell tumour cell lines to the new anti-cancer agent, temozolamide. Brit. J. Cancer, 71, 904–906 (1995). SARK, M.W.J., TIMMER-BOSSCHA, H., MEIJER, C., UGES, D.R.A., SLUITER, W.J., PETERS, W.H.M., MULDER, N.H. and DEVRIES, E.G.E., Cellular basis for differential sensitivity to cisplatin in human germ cell tumour and colon carcinoma lines. Brit. J. Cancer, 71, 684–690 (1995). SEIDMAN, A.D. and SCHER, H.I., The evolving role of chemotherapy for muscle infiltrating bladder cancer. Semin. Oncol., 18, 585–595 (1991). SHELLARD, S.A., FICHTINGER-SCHEPPMAN, A.M.J., LAZO, J.S. and HILL, B.T., Evidence of differential cisplatin-DNA adduct formation, removal and tolerance of DNA damage in three human lung carcinoma cell lines. Anticancer Drugs, 4, 491–500 (1993). SHELLARD, S.A., HOSKING, L.K. and HILL, B.T., An anomalous relationship between cisplatin sensitivity and the formation and removal of platinumDNA adducts in two human ovarian carcinoma cell lines in vitro. Cancer Res., 51,4557–4564 (1991). TIMMER-BOSSCHA, H., TIMMER, A., MEIJER, C., DE VRIES, E.G.E., DE JONG, B., OOSTERHUIS, J.W. and MULDER, N.H., cis-Diamminedichloroplatinum(II) resistance in vitro and in vivo in human embryonal carcinoma cell lines. Cancer Res., 53, 5705–5713 (1993). VENEMA, J., BARTOSOVA, Z., NATARAJAN, A.T., VAN ZEELAND, A.A. and MULLENDERS, L.H.F., Transcription affects the rate but not the extent of repair of cyclobutane pyrimidine dimers in the human adenosine deaminase gene. J. biol. Chem., 267, 8852–8856 (1992). VENEMA, J., MULLENDERS, L.H.F., NATARAJAN, A.T., VAN ZEELAND, A.A. and MAYNE, L.V., The genetic defect in Cockayne syndrome is associated with a defect in repair of UV-induced damage in transcriptionally active DNA. Proc. nat. Acad. Sci. (Wash.), 87, 4707–4711 (1990b). VENEMA, J., VAN HOFFEN, A., NATARAJAN, A.T., VAN ZEELAND, A.A. and MULLENDERS, L.H.F., The residual repair capacity of xeroderma pigmentosum complementation group C fibroblasts is highly specific for transcriptionally active DNA. Nucleic Acids Res., 18, 443–448 (1990a). WALKER, M.C., PARRIS, C.N. and MASTERS, J.R.W., Differential sensitivities to chemotherapeutic drugs between testicular and bladder cancer cells. J. nat. Cancer Inst., 79, 213–216 (1987). ZHEN, W., EVANS, M.E., HAGGERTY, C.M. and BOHR, V.A., Deficient gene specific repair of cisplatin-induced lesions in xeroderma pigmentosum and Fanconi’s anemia cells. Carcinogenesis, 14, 919–924 (1993). ZHEN, W., LINK, C.J., O’CONNOR, P.M., REED, E., PARKER, R., HOWELL, S. and BOHR, V.A., Increased gene specific repair of cisplatin interstrand crosslinks in cisplatin resistant human ovarian cancer cell lines. Mol. cell. Biol., 12, 3689–3698 (1992).