Mutagenic activity of ambient oxygen and mitomycin

Mutagenesis vol.12 no.2 pp.69-77, 1997

Mutagenic activity of ambient oxygen and mitomycin C in Fanconi's anaemia cells

Wolfgang Liebetrau1'4, Thomas M.Riinger2, Beatrix E.Mehling1, Martin Poot3 and Holger Hoehn1 'Department of Human Genetics, University of WQrzburg, Biozentmm, Am Hubland, D-97074 WUrzburg, 2Department of Dermatology, University of Gottingen, Germany, and 'Molecular Probes, Eugene, Oregon, USA ^To whom correspondence should be addressed

Introduction Fanconi's anaemia (FA) is an autosomal recessive disease that has recently received major attention as a cancer-prone syndrome and as a candidate for gene therapy (Walsh et al, 1994). Since cells from FA patients are chromosomally unstable and hypersensitive to DNA cross-linking agents such as mitomycin C (MMC) and diepoxybutane (DEB) (Auerbach, 1993) it has been suggested that they may be defective in © UK Environmental Mutagen Society/Oxford University Press 1997

69

Downloaded from mutage.oxfordjournals.org by guest on July 12, 2011

Cellular evidence suggests that Fanconi's anaemia (FA) might be a condition of increased oxygen sensitivity. In order to test this hypothesis, a common shuttle vector assay with the plasmid pZ189 was utilized. We transfected intact, circular plasmid into FA and control lymphoblast and fibroblast host cells maintained at 5 and 20% O2 (v/v). In parallel experiments, host cells were exposed to different concentrations of mitomycin C (MMC), a cross-Unking agent towards which FA cells are known to be hypersensitive. Baseline mutation frequencies at 20% oxygen were significantly higher in plasmids passaged through FA lymphoblasts or FA fibroblasts in comparison with passage through the corresponding control cells. Lowering the oxygen concentration during the 48 h transfection period to 5% resulted in a significant decrease of mutation frequencies in plasmids passaged through FA cells. Sequence analysis of plasmids recovered from FA lymphoblasts revealed a mutation hot spot (22% of point mutations with G:C to A:T base substitutions) at base 117 of the supF tRNA gene. This hot spot was present only at 20% oxygen. 59% of the base changes at the hot spot and 39% of the changes elsewhere in the supF gene were C to T transitions (the corresponding figures are 0 and 27% at 5% oxygen), the most common type of base change induced by oxygen. The mutation spectrum observed suggests a role for 8hydroxydeoxyguanosine in G:C to A:T base substitutions: at 20% oxygen, FA cells displayed 4 times as many G:C to T:A transversions than FA cells kept at 5% O2. In MMC treated cells the decrease in plasmid survival is dose dependent and more pronounced in FA than control cells. Mutation analysis shows similar rates of deletions for both control and FA cells . However, FA cells generate a specific type of deletion whose breakpoint involves an indirect repeat that corresponds to a heptamer signal sequence commonly seen at recombination sites. Together our data provide compelling evidence that the genetic defect in FA causes oxygen sensitivity and recombinational types of DNA lesions following exposure to MMC.

some component of the DNA repair cascade. The genes for the complementation groups C and A have been cloned (Strathdee et al, 1992; Fanconi Anaemia/Breast Cancer Consortium, 1996; Lo Ten Foe et al, 1996). The FAC gene product has been localized to the cytoplasm (Youssoufian, 1994; Yamashita et al, 1994), a finding that contradicts its direct function in a nuclear repair pathway. Apart from the fact that the FAC protein appears to form a multimeric complex with other proteins (Youssoufian et al, 1995), its function remains enigmatic, as is true for the function of the FAA gene. The main part of the present paper deals with a phenotype of FA cells that is even less understood than their sensitivity to cross-linking agents. For this particular phenotype, sensitivity to oxygen, may be fundamentally important for the understanding of the clinical manifestations of the disease: (i) FA patients show a highly variable pattern of congenital malformations. This pattern is reminiscent of defective protection against developmental oxidative stress observed in transgenicp55 deficient mice (Nicol et al., 1995); (ii) 50% of FA patients develop bone marrow failure (pancytopaenia) by age 8, and the bone marrow is known to be one of the most oxygen-sensitive tissues in the human body; (iii) FA patients have a 10—20% lifetime cumulative risk of neoplasia the most frequent type of which is myelomonocytic leukaemia. Monocytes and their precursors are cells that actively generate reactive oxygen species such that the genome of these cells might be especially vulnerable in a situation of increased oxygen sensitivity (Schindler et al, 1987; Szatrowski and Nathan, 1991). Hypersensitivity of FA cells to oxygen was first reported by Joenje et al. (1981) who noted a dramatic increase in chromosome breakage when primary blood lymphocytes from FA patients were exposed to higher than ambient oxygen (an elevated rate of chromosome breakage at ambient oxygen culture conditions was noted first by Schroeder et al. (1964). Unusual sensitivity towards oxygen was likewise established for FA fibroblast cultures whose exceedingly poor growth and cloning performance at ambient oxygen could be restored to near normal by lowering the oxygen supply in the cell culture environment to 5% (v/v) (Schindler and Hoehn, 1988). Most recently, Poot et al. (1996) showed that lymphoblastoid cell lines from FA patients are also more sensitive towards reactive oxygen species than control cell lines. While confirming oxygen hyper-sensitivity as an uniform phenotype in primary FA fibroblast cultures, Saito et al. (1993) have questioned whether this phenomenon should be considered a primary expression of the underlying gene defect. Their reservations were based on the observation that SV40 large T-transformed FA fibroblasts lose their oxygen sensitivity. Accordingly, Saito et al. considered oxygen sensitivity not as a primary, but rather as only a secondary manifestation of the FA genotype. However, in a subsequent paper (Saito et al, 1995) the authors themselves showed that loss of oxygen-mediated control of cell proliferation is a general phenomenon following transformation by SV40, and that this is in no way limited to FA cells.

W.Liebetrau et al

Since there is evidence that reactive oxygen species can induce at least some types of mutations in mammalian cells (Hsie et al, 1986; Oiler and Thilly, 1992; Gille et al., 1994) we decided to test the oxygen-sensitivity hypothesis in FA by analysing mutation frequencies and mutational spectra caused during passage of indicator plasmids through FA cells in the presence of either 5 or 20% (v/v) oxygen. If the underlying genetic defect(s) in fact increases the vulnerability of FA cells towards oxygen, elevated mutation frequencies and mutational spectra suggestive of oxygen action should be observed. We used the plasmid pZ189 (Seidman et al, 1985) that carries the supF marker gene and that is known to replicate in human cells. In contrast to conventional shuttle vector assays, we did not expose the plasmids to mutagenic agents prior to transfection, but varied the conditions of our host cells during the 48 h transfection period. Since FA cells are known to be highly sensitive towards MMC, experiments in which plasmids were passaged through MMC treated cells served as positive controls for our oxygen studies and gave additional interesting information about the MMC hypersensitivity of FA cells at the molecular level.

Materials and methods Cells Epstein-Barr virus-transformed lymphoblastoid cells from a patient with FA (GM8010; complementation group unknown) and from a clinically normal individual (GM3715) were obtained from the Human Genetic Cell Repository (Camden, NJ, USA). They were routinely grown as suspension cultures in Roswell Park Memorial Institute (RPMT) 1640 medium (Gibco BRL, Eggenstein, Germany) supplemented with 10% fetal calf serum (Gibco BRL). High humidity incubators (Haereus; Hanau, Germany) were equipped with automatic CO 2 and O 2 sensors. Hypoxic culture conditions were generated (within 5-10 min) by purging air with nitrogen. 6% CO2 was used with lymphoblasts, 5% CO 2 (v/v) was used for fibroblasts. Primary fibroblastoid cells were obtained from a 18 week male fetus homozygous for Fanconi's anaemia of unknown complementation group (H9438) and a 20 week male fetus affected by the fragile X syndrome (H94-17) as a control for oxygen experiments. Fibroblast cultures were routinely grown in MEM medium (Gibco BRL) supplemented with 10% fetal calf serum (Gibco BRL) and incubated at 5% CO 2 and 5 or 20% O 2 . Plasmid and bacteria The shuttle vector plasmid pZ189 (Seidman et al, 1985) can replicate in both mammalian (including human) and bacterial cells. The plasmid contains the origin of replication and large tumor antigen from simian virus 40 plus the origin of replication from pBR327 and a mutagenesis marker gene, the bacterial supF suppressor tRNA (Ranger and Kraemer, 1989). A functional 150-bp supF gene in pZ189 suppresses an amber mutation in the lac Z gene in the Escherichia coli indicator strain (MBM 7070), resulting in blue colonies on indicator plates. Mutations inactivating supF result in white or light blue colonies. The E.coli MBM7070 has the genotype [F-lac Z(Am) CA 7020 lacYl hsdR hsdM A(araABC-leu) 7679 gal U gal K rpsL thi].

70

Transformation of bacteria and mutation analysis The replicated purified plasmid was used to transform competent MBM7070 E.coli indicator bacteria by electroporation. Competent bacteria (40 ill) were placed in a Bio-Rad cuvette (0.2 cm gap) on ice along with 1.8 ill of the replicated plasmid in distilled water The bacteria were pulsed once with 2500 V with settings of 200 £i and 25 up. After a 1 h incubation at 37°C in a dry shaking incubator, the bacteria were spread onto LB agar dishes containing ampicillin, iPTG (isopropyl ^-D-thiogalactopyranoside), and Xgal (5-bromo-4-chloro-3-indolyl fi-D-galactopyranoside). The proportion of mutant white compared to the total number of colonies was evaluated as mutation frequency. In order to exclude borderline phenotypes only white (with completely inactivated supF) and not light blue colonies (with incompletely inactivated supF ) were isolated and the plasmid was further purified from 5 ml overnight culture with Qiagen tip-20 columns (Qiagen, Hilden, Germany). The total number of colonies reflect plasmid survival. For the further characterization of the recovered mutants, their size was compared to the size of the wild-type plasmid by agarose gel electrophoresis and the base sequence of the supF gene was determined by the dideoxy-sequencing procedure using the PRISM™ Ready Reaction DyeDeoxy™ Terminator Cycle Sequencing Kit and an Applied Biosystem Sequencer (Applied Biosystems, Weiterstadt, Germany). For the sequencing reaction the following ohgonucleotides were used: HUG10: 3 ' TATCACGAGGCCCTTTC y HUG37: 5 ' CCTAACTACGGCTACAC 3 ' To ensure that the mutants were of independent origin, we considered identical mutations only if they were derived from different transfections. Statistical analysis was performed by calculating mutation frequencies for each independent experiment (n = 3-6), averaging the values, and using Student's /-test to test for differences.

Results Oxygen experiments Mutation frequencies at 5 versus 20% (v/v) oxygen: Table IA summarizes mutant colony counts of plasmids passaged through FA and control cells maintained at 5 and 20% oxygen during the 48 h transfection period. As in previous studies using the plasmid pZ189 (Protic-Sabljic et al, 1986; Seetharam et al, 1987; GozUkara et al, 1994), our baseline mutation frequencies in plasmids recovered from lymphoblastoid cell lines are in order of 0.01-0.04%. Plasmids passaged through the control lymphoblastoid cell line displayed no difference in mutant counts as a function of oxygen concentration. In contrast, plasmids replicated through FA cells yielded twice as many mutants at 5%, and nearly four times as many mutants at 20% oxygen. The oxygen dependence of mutation frequencies was even more striking when plasmids were passaged through fetal lung fibroblast cultures (strain H94-17 and H94-38). With this cell type, both control and FA strains yielded significantly higher mutation frequencies at ambient compared to 5% oxygen. Baseline values at 5% oxygen were considerably higher in fibroblasts than in lymphoblasts, and there was no difference between the control and the FA strain


Following reductive activation MMC is converted into an alkylating agent (Tomasz and Lipman, 1981) that can cause cross-links and strandbreaks, such that deletions may result (Tomasz et al, 1987; Basu et al, 1993). Since in theory at least part of the sensitivity of FA cells towards MMC could be due to the generation of oxygen radicals (Pritsos and Sartorelli, 1986; Druse et al, 1990), FA and control cells were exposed to this drug using the shuttle vector assay. In contrast to previous studies (Skikant et al, 1994; Bredberg et al, 1995), we have not treated the plasmid but have exposed the transfected cells to different concentrations of the cross-linking agent. In this report we describe, in addition to the oxygen experiments, the detection of a hot spot for a specific deletion that is nearly exclusive for FA cells exposed to MMC. Finally, we define a specific breakpoint in the sequence of the mutants with the MMC-induced deletion pattern.

Transfection and recovery of the plasmids For each experiment, three to six independent transfections of pZ189 into FA and control cell lines were achieved by electroporation with a Bio-Rad GenePulser (Bio-Rad, Munich, Germany): 15X106 cells suspended in serumfree RPMI 1640 medium respectively. The MEM medium was placed in a Bio-Rad electroporation cuvette (0.4 cm gap) together with 8 |ig of pZ189 DNA in a total volume of 400 JJJ and then pulsed once with 250V with a capacity setting of 960 \sF. After electroporation the cells were transferred to 25 ml of prewarmed (37°C) complete medium and allowed to grow in a humidified incubator at 6% CO2/20% Oj or 5% O 2 for lymphoblasts and 5% CC>2/20% O2 or 5% O2 for fibroblasts. During experiments with MMC complete medium was supplemented with different MMC concentrations. After a 48 h period allowing for DNA damage formation and DNA repair and replication of transfected plasmids, plasmid DNA was isolated from the cell cultures with Qiagen plasmid kit (Qiagen, Hilden, Germany). Digestion with the endonuclease Dpn I ensured that only plasmids replicated within the host cells were recovered.

Mutagenicity in Fanconl's anaemia cells

Table IA. Mutation frequencies in plasmids pZ189 after passage through control (CIK) or FA cells, incubated at different concentrations of oxygen Cell lines

Conditions during plasmid replication (% oxygen)

No. of iupF"/total colonies

Percentage mutants

SD

Student's t-test

Lymphoblastoid cell lines GM3715 (CTR) GM3715 (CTR) GM8010 (FA) GM8010 (FA)

5 20 5 20

24/141170 34/179692 38/114850 80/104795

0.017 0.019 0.033 0.076

0.001 0.002 0006 0.021

P = 0.556

Fetal lung fibroblasts H94-17 (CTR) H94-17 (CTR) H94-38 (FA) H94-38 (FA)

5 20 5 20

102/147680 79TO904 26/35075 214/118004

0.069 0.108 0.074 0.181

0.026 0.012 0.023 0.055

P < 0.001

P < 0.001

P < 0.001

Additional (-tests: 5% oxygen: GM8010 versus GM3715, P = 0.003; 20% oxygen: GM8010 versus GM3715, P < 0.001; 5% oxygen: H94-38 versus H9417, P = 0.495; 20% oxygen: H94-38 versus H94-17, P < 0.001

Table IB. Mutation frequencies following passage of plasmids in control (CTR) and FA host cells treated with different concentrations of MMC Cell lines

No. of supF~/total colonies

Percentage mutants

SD

Student's r-test

1.5 3 9 1.5 3 9

122/265738 32/43233 124/99919 164/158994 124/66437 107/57712

0.046 0.074 0.124 0.103 0.187 0.185

0.012 0.024 0.023 0.049 0.039 0.084

P < 0.001 P < 0.001 Downloaded from mutage.oxfordjournals.org by guest on July 12, 2011

Lymphoblastoid cell lines GM3715 (CTR) GM3715 (CTR) GM3715 (CTR) GM8010 (FA) GM8010 (FA) GM8010 (FA)

Conditions during plasmid replication (uM MMC)

P < 0.001 P = 0.927

Additional /-tests: 1.5 uM MMC: GM8010 versus GM3715, P < 0.001; 3 uM MMC: GM8010 versus GM3715, P < 0.001; 9 uM MMC: GM8010 versus GM3715, P < 0.001.

Table HI. Types of base substitution mutations in plasmids pZ189 after passage through FA lymphoblasts, incubated at different oxygen concentrations. Figures in parentheses are percentages

Table n. Characteristics of supF mutants: Types of mutations observed following passage of plasmid pZ189 through FA and control (CTR) lymphoblastoid cell lines at different oxygen conditions. Figures in parentheses are percentages

Base substitutions FA GM8010 20% O 2 Point mutations Deletions Total number of plasmids analysed

38 (70)° 16 (30) 54 (100)

FA GM8010 5% O2 b

11 (79) 3(21) 14(100)

CTR GM3715 20% O 2 C

23 (82) 5(18) 28 (100)

GM8010 (20% oxygen)

GM8010 (5% oxygen)

Transitions G:C->A:T A:T->G:C

35 (47) 5(7)

13 (29.5) 3 (6.8)

Transversion G:C->T:A G:C->C:G A:T-C'G A:T->T:A Total

14(19) 4(5) 1 (1) 16(21) 75 (100)

2 (4.6) 7(15 9) 12 (27.3) 7 (15.9) 44(100)

CTR GM3715 c

19 (83) 4 (17) 23 (100)

'Including five tandem and one triple base substitution. ^Including four tandem and one triple base substitution. Including two tandem base substitutions.

maintained at 5% oxygen. The higher sensitivity of fibroblasts as compared to lymphoblasts is not unexpected (Schroeder et al., 1964; Schindler and Hoehn, 1988), but irrespective of cell type mutation frequencies of plasmids were uniformly oxygen dependent when passaged through FA cells. Table IB depicts the corresponding results of transfections carried out while host lymphoblasts were exposed to increasing concentrations of MMC. As expected, there was a concentration-dependent increase of mutant yields, with FA cells displaying higher overall levels and earlier saturation of mutation frequencies than controls. These results show that the wellknown sensitity of FA cells towards MMC translates directly into higher mutation frequencies under the conditions of our shuttle vector assay. The increased sensitivity of FA cells towards oxygen was also reflected by a reduction of plasmid survival. Relative to

5% oxygen conditions, only 68% of the plasmids passaged through FA lymphoblasts at 20% oxygen survived. The corresponding figure in control cells is 89%. Mutation analysis Plasmids with mutations inactivating the supF suppressor gene were purified and characterized by plasmid size, presence of restriction sites and DNA-sequencing. The classes of mutations found in 119 independent plasmids recovered from FA cells are shown in Table n. As in earlier studies (Seetharam et al., 1991) four major classes of mutations were observed: those with single base substitutions, those with tandem mutations, those with multiple mutations and those with deletion mutations. Some of the deletions were diagnosed by their loss of the Eco RI restriction site or their increased mobility in agarose gel electrophoresis. All detected deletions were larger 71

W.Liebetrau et aL

T T T AC T A T T TTGATATGAT GCGCCCCGCT TCCCGATAAG GGAGCAGGCC AGTAAAAGCA 50 60 70 80 90 AACTATACTA CGCGGGGCGA AGGGCTATTC CCTCGTCCOQ TCATTTTCGT > pr«-tRHA

T T T

T T T T T T AT

(56-98)

T T T T T T T TT

T

> pre-tRNA

TCT

(58-98)

C T C A A A T C A A A A CAA C C AT T T A C T TTACCTGTGG TGGGGTTCCC GAGCGGCCAA AGGGAGCAGA CTCTAAATCT 100 110 120 130 140 AATGGACACC ACCCCAAGGG CTCGCCGGTT TCCCTCGTCT GAGATTTAGA

A T A T A T G C A G A C T T TT A G TTACCTGTGG TGGGGTTCCC GAQCGGCCAA AGGGAGCAGA CTCTAAATCT 100 110 120 130 140 AATGGACACC ACCCCAAOOG CTCOCCGGTT TCCCTCGTCT GAGATTTAOA

> suppressor-tRNA

(99-183)

(99-183)

160

170

180

C C

c

c

c

c

C TT. AA C A T A CA C T GCCGTCATCG ACTTCGAAGG TTCGAATCCT TCCCCCACCA CCATCACTTT 150 160 170 180 190 CGGCAGTAGC T3AAGCTTCC AAGCTTAGGA AGGGGGTCGT GGTAGT3AAA

190

CGGCAGTAGC TGAAGCTTCC AAGCTTAOGA AGGGGGTGOT OGTAGTGAAA T T T TTC CAAAAGTCCG 200 GTTTTCAGGC

C TT

CAAAAGTCCG 200 GTTCTCAGGC

3 ' f l a n k i n g r e g i o n (184-200)

3'flanking region (184-200)

B £C£

T T C A A G

Q

T T AX G T T C

T T G T A A A TTGATATGAT GCGCCCCGCT TCCCGATAAG GGAGCAOGCC AGTAAAAGCA 50 60 70 80 90 AACTATACTA CGCGGGGCGA AGGGCTATTC CCTCGTCCGG TCATTTTCGT

TTGATATOAT GCGCCCCGCT TCCCGATAAG GGAGCAGGCC AGTAAAAGCA

> pre-WNA (58-98)

50 60 70 80 90 AACTATACTA CGCGGGGCGA AGGGCTATTC CCTCGTCCGG TCATTTTCGT > pre-tHNA (58-98) Q

G C C CAT G A G C TTACCTOTGG TGGGGTTCCC GAGCGGCCAA AGGGAGCAGA CTCTAAATCT 100 110 120 130 140 AATGGACACC ACCCCAAGGG CTCGCCGGTT TCCCTCGTCT GAGATTTAGA

G G G G C A G G A G G A C C A A T G G G A T GAGC TTACCTGTGG TGGGGTTCCC GAGCGGCCAA AGGGAGCAGA CTCTAAATCT 100 110 120 130 140 AATGGACACC ACCCCAAGGG CTCGCCGGTT TCCCTCGTCT GAGATTTAGA > suppressor-tRNA

(99-183)

> suppressor-tRNA (99-183)

T T A T . A T T G C G

T T G

C C C A A

GCCGTCATCG ACTTCGAAGG TTCGAATCCT TCCCCCACCA CCATCACTTT 150 160 170 180 190 CGGCAGTAGC TGAAGCTTCC AAGCTTAGGA AGGGGGTGGT GGTAGTGAAA

72

A C T A CA T A CA A C£ G C T ACT A A T A GCCOTCATCG ACTTCGAAGG TTCGAATCCT TCCCCCACCA CCATCACTTT 150 160 170 180 190 CGGCAGTAGC TGAAGCTTCC AAGCTTAGGA AGOGGGTGGT GGTAOT3AAA

C C C C T CAAAAOTCCG 200 GTTTTCAGGC

T C C A CAAAAGTCCG 200 GTTTTCAGGC




C A T A T A T A A T TTT T T A T AT T CA GCCGTCATCG ACTTCGAAGG TTCOAATCCT TCCCCCACCA CCATCACTTT 150

G A A A A C T

TTGATATGAT GCGCCCCGCT TCCCGATAAG GGAGCAGGCC AGTAAAAGCA 50 60 70 80 90 AACTATACTA CGCGGGGCGA AGGGCTATTC CCTCGTCCGG TCATTTTCGT

A A A A A A A T T T T T T T T T

> «ijppresBor-tRNA

G G

Mutageniclty in Fanconi's anaemia cells

than 100 bp and the plasmids often lacked the complete supF gene. The proportion of deletions was highest in plasmids passaged through FA cells at 20% oxygen, whereas those passaged through FA cells at 5% oxygen approached the level of deletions observed in controls at 5% and 20% oxygen. The types of base substitutions found in the plasmids propagated through FA cells grown at 5% and 20% oxygen are listed in Table in. All six possible types of base substitutions were observed. Plasmids passaged through FA cells at 20% oxygen exhibited an about equal distribution of transitions and transversions, whereas those passaged at 5% exhibited more transversions (64 versus 46%). Among the latter, however, the percentage of the G:C to T:A type decreased from 19% at 20% oxygen to 4.6% at 5% oxygen, whereas the A:T to C:G type increased from only 1% to 27.3%. Thus there are differences in the frequency of the different types of single base substitutions depending on whether plasmids were transfected into FA cells maintained at ambient versus hypoxic oxygen conditions.

MMC experiments Following exposure to MMC a total of 116 plasmids were analysed; 67% of 56 plasmids passaged through the control cells exhibited deletions. The corresponding figure was 68% of 60 plasmids recovered after passage through FA cells. Figure 2 shows the survival of the plasmid pZ189 after passage through FA and normal lymphoblasts exposed to increasing concentrations of MMC. As expected, there is a concentration-dependent decrease of plasmid survival, with FA cells displaying lower overall levels and earlier saturation than controls. These results show that the well-known sensitivity of FA cells towards MMC translates directly into reduced plasmid survival and higher mutation frequencies under the conditions of our shuttle vector assay. Deletion-breakpoint analysis A detailed analysis of deletion breakpoints was carried out in plasmids recovered from MMC-treated FA and control cells, because twice as many deletions resulted from MMC treatment in comparison with 20% oxygen cultures. Using agarose gel electrophoresis a total of 79 deletion mutants were detected. 3.6% of deleted plasmids derived from control cells but 25% derived from FA cells had a size of 4.3 kb. Given an initial plasmid size of 5.5 kb, the recovered plasmids therefore carry a 1.2 kb deletion (Table IV). The other deletions were distributed randomly starting from within the supF gene. An attempt was made to sequence all plasmids with the primer HUG 10 that binds upstream to the supF gene. However, none of the deletion mutants of the 1.2 kb type could be sequenced with this primer, indicating that the primer bindingsite must have been mutated or deleted. These mutants could only be sequenced using the downsteam primer HUG 37. Sequence analysis showed that the complete supF gene was deleted. A remarkable finding was that the 1.2 kb deletion occurred predominantly in plasmids recovered after passage through FA cells (7-fold excess). Moreover, these particular types of deletions had a common downsteam breakpoint localized at the same base position, i.e. position 211 which is 11 bp downstream from the supF gene (Figure 3). The sequence homology to the pZ189 vector resumed upstream around 100 bp adjacent to the coding region of the ampicillin gene. The breakpoint is flanked by indirect repeats of the signal heptamer structure described as a breakpoint in the HPRT gene (Laquerbe et ai, 1995). These kinds of deletions were found exclusively in lymphoblastoid cell lines incubated with MMC. Discussion There are at least five complementation groups in FA, three of which have been assigned to different chromosomal loci (Whitney et al, 1995; Pronk et ai, 1995; Joenje et ai, 1995). In the context of the present study it is important to realize that parameters such as chromosomal breakage, cross-link

Fig. 1. Map of sites of point mutations in the double-stranded supF gene of the plasmid pZ189 after passage through (A) the FA lymphoblast line GM8010 at 20% O2; (B) GM8010 at 5% O 2 ; (C) GiM3715 at 20% O 2 and (D) GM3715 at 5% O2. Each base substitution is indicated above the mutated base in the upper strand. Tandem and triple base substitutions are indicated by a line. The mutations were of independent origin from three to six independent transfection experiments.

73


Mutation spectrum analysis The location of base substitution mutations within the coding sequence of the supF gene is shown in Figure 1A for the FA cell line GM8010 at 20% oxygen, in Figure IB for GM8010 at 5% oxygen, in Figure 1C for GM3715 at 20% oxygen and in Figure ID for GM3715 at 5% oxygen. Single base substitutions are distributed throughout the supF gene, but most point mutations are localized in the hypothetical cloverleaf-structure of this gene (Kraemer and Seidman, 1989). There is an uniformly high mutation frequency involving G:C basepairs at ambient oxygen (71% for GM8010). In plasmids recovered from FA cells kept at 20% oxygen a mutational hot spot (G:C to T:A base substitution) was located at base 117 within the supF gene of pZ189 (Figure 1A). This hot spot accounted for 22% of the point mutations in plasmids recovered from the FA cell line. In other words, every third re-isolated and mutated plasmid from FA lymphoblasts at ambient oxygen was mutated at base 117 (18 of 54 plasmids had a point mutation at base 117). This hot spot was detected in each of six independent transfections. In contrast, in plasmids recovered from FA cell lines grown at 5% O2, just 6.8% of the point mutations involved base 117 and consisted of G:C to C:G transversions rather than G:C to T:A events. Regardless of whether control cell lines were grown at 20% or 5% oxygen, 4.7% to 5% of the point mutations involved base 117. As in FA cells grown at 5% oxygen, only G:C to C:G transversions were noted in 5% oxygen control experiments. In mutant plasmids with complete inactivation of supF due to point mutations (white bacterial colonies) we found a mean number of base substitution mutations of one to three per mutant. Tandem mutations of the CC to TT and CCC to TTT types were observed only in plasmids recovered from FA cells exposed to 20% oxygen. The quality and distribution of mutations throughout the supF gene therefore is quite distinctive for plasmids passaged through FA cells at ambient oxygen, whereas the distribution of mutations in plasmids passaged

through FA cells at 5% oxygen resembles that of plasmids passaged through control cells at 5% and 20% oxygen.

W.Liebetrau et aL Plasmid * -

survival

100 -Control call line GM3715

80-

- FA cell line QM8010 0 (jM

15 (IM

3 (lM

9 (lM

MMC concentrations

Fig. 2. Genotoxicity of MMC shown as decreasing survival of the plasmid pZ189 after passage through normal and FA lymphoblastoid lines GM3715 and GM8010.

Table IV. Characterization of deletions observed following passage of plasmid pZ189 through FA and control lymphoblasts in the presence of MMC GM8010 (FA) 41/60 (68%)

GM3715 (CTR) 38/56 (67%)

Deletion type 4.3 kb Other Total

Percentage 25 75 100

Percentage 3.6 96.4 100

sensitivity, cell cycle blockage, and oxygen sensitivity are common to cells derived from any FA patient, irrespective of complementation group (Auerbach, 1993; Saito et al, 1993; Seyschab et al, 1995). Since our FA cell cultures were repeatedly tested and retained these cellular features throughout the course of our study, we can be quite confident that they are representative of the FA cellular phenotype even though their complementation group is as yet unknown. In other words, the classification of FA cells according to complementation group appears to be irrelevant for studies that focus on the behaviour of cells rather than on the function of a single gene. The centra] finding of our study is a clear-cut oxygen dependence of the yield of supF mutants in plasmids that underwent replication in FA cells. This oxygen dependence was observed with two different FA cell types, and, because of the large number of bacterial colonies screened, has a solid statistical basis. We conclude from this observation that oxygen has a strong influence on the replication and repair fidelity in FA cells. Considerable evidence has accrued supporting the hypothesis that reactive oxygen species may be an important cause of DNA damage leading to mutation (Moody and Hassan, 1982; Hsie et al, 1986; Loeb et al, 1988; Moraes et al, 1989; Moraes et al, 1990). In the mutagenesis process active oxygen species, such as superoxide and hydrogen peroxide, are thought to induce mutations through the formation of DNA adducts that result in misincorporation during replication. Analysis of in vitro damage products (Aruma et al, 1989), in vitro misincorporation studies (Shibutani et al, 1991), and in vivo genetic studies of E.coli strains defective for the DNA adduct repair enzyme 8-hydroxyguanosine endonuclease (Cabrena et al, 1988; Bessho et al., 1992) suggest that 8-hydroxydeoxyguanosine (8-OHdG) represents the principal lesion involved in mutagenesis by oxygen species. 8-OHdG mispairs with deoxyadenosine, leading to G:C to T: A transversions (Shibutani 74

A second important finding of our study is the presence of a single mutational hot spot in plasmids isolated after passage through FA lymphoblasts maintained at 20% oxygen (Figure 1A). Under these conditions, every third mutated plasmid showed this G:C to T:A change at position 117 in the putative clover-leaf structure of the supF gene (Laquerbe et al, 1995). The preponderance of C to T transitions at this particular hot spot is indicative for a causative role of oxygen because this type of transition is the most common type of base change induced by oxygen (Tkeshelashvili et al, 1991). This type of base change is caused by incorporation of 5-hydroxy-2'deoxycytidine into DNA leading to C to T transitions. 5hydroxy-2'-deoxycytidine is more mutagenic than any previously identified oxidative DNA lesion (Feig et al, 1994). Plasmids recovered after passage through FA cells at 5% oxygen showed a much less pronounced series of mutations that include a different spectrum of base substitutions at this site. The position of the hot spot within a G/C rich area argues against deaminations as its likely cause. However, since polymerase alpha arrest sites are known to be G/C rich, it is conceivable that the observed type of mutation arises as a consequence of an impaired polymerase function (Weaver and DePamphilis, 1982). Supporting this notion are experiments by Feig and Loeb (1994) showing that the mutational spectra induced by oxygen radicals reflect the type of DNA polymerase that is concerned with the repair of the respective DNA damage. Likewise, studies by Oiler and Thilly (1992) show that oxygen-induced mutational spectra in the HPRT gene of human lymphoblastoid cells differ substantially from the spontaneous pattern. We therefore conclude that our observation of a mutational hot spot that occurs in plasmids passaged through FA cells at ambient oxygen supports the hypothesis of an oxygen-related impairment of the replication apparatus in FA cells. A final point concerning the spectrum of mutations present in plasmids that were passaged through FA cells at 20% oxygen relates to the observation of a CC to TT and a CCC to TTT change at nucleotides 173 to 176 of the supF gene. According to Reid and Loeb (1993) such tandem


Strain Plasmids del/total

et al, 1991; Cheng et al, 1992). Likewise, Retel et al (1993) report that irradiation of an oxygenated solution of pUC18 plasmid DNA produces mainly G:C to T:A transversions at one of the mutational hot spots found in their system. Our results tend to confirm the putative role of 8-OHdG in the process leading to enhanced mutagenesis in FA cells. As shown in Table HI, G:C to T:A transversions were increased four-fold in FA cells at 20% oxygen compared with 5% oxygen. This could be the result of G-A mispairing caused by the presence of 8-OHdG. In this context it is interesting that two reports claim an increased production of 8-OHdG by FA cells (Takeuchi and Morimoto, 1993; Degan et al, 1995), but additional studies are required to settle this issue (Lindahl, 1993). As early as 1987 Decuyper-Debergh et al. showed that G:C base pairs are the preferential target for mutations caused by singlet oxygen. The high specificity of singlet oxygen was established because an analysis of mutations obtained after treatment with other agents such as psoralen (Piette et al, 1985) or depurination by heat and acid (Kunkel, 1984) yielded different patterns. The obvious correlation between sequences with guanine residues and singlet oxygen mutagenesis support a direct role of some guanine oxidation products in the mutagenesis process. However, the available data cannot discriminate between cyanuric acid or 8-oxoguanine as the likely major premutational lesions.

Mutagenicity in Fanconi's anaemia cells

H 211

d.lot.d i t q i u n breakpoint *t 211

plasmida with 1.2 Kb d«l«tion»

Breakpoint-sequence:

201 pZ189-sequence. signal heptamer: signal heptamer: deletion-sequence:

216

AAAGAATTGAGCGTCAG 5

C^dSTG 3 3 GTGIC&C 5

gttccacJEGAGC£IC&3

T breakpoint

(indirect repeat sequence is underlined)

Fig. 3. Observed breakpoint in reisolated plasmids from MMC induced FA cells with a 1.2 kb deletion (drawing not to scale) and the observed breakpoint sequence.

(Pritsos and Sartorelli, 1986; Druse et al, 1990). It should be kept in mind, however, that in addition to oxidative damage MMC induces alkylation damage as well; it is therefore difficult to decide whether the deletions observed in these cells were due to alkylation or oxidative damage. Srikanth et al. (1994) have reported that reductively activated MMC causes predominantly base substitutions in the vector pSP189 (a pZ189 derivate). Because our studies yielded a preponderance of deletions, we suggest that the effect of MMC in our studies may be mediated by reactive oxygen species. Acknowledgements This work was supported by a grant from the Bundesministerium ftir Bildung, Wissenschaft, Forschung und Technologie (grant: FKZ 0310721). We thank Petra Busch and Wolfgang HSdelt for technical assistence.


transitions represent a kind of 'signature' for the mutagenic action of oxygen. Concerning complex mutations, Table II shows that plasmids passaged through FA cells at ambient oxygen exhibit more deletions than plasmids passaged through FA cells at hypoxic culture conditions. Such mutations are observed after exposure of mammalian cells to oxidative stress (Oiler and Thilly, 1992) and might result from crosslinks and double strand breaks. The repair of such lesions is initiated by specific DNA glycosylases and completed by polymerases and ligases (Kornberg and Baker, 1992; Lindahl, 1993). In this context it is perhaps relevant that Riinger et al. (1993) have previously shown that the ligation process in FA cells, while not deficient, results in increased levels of complex mutations (tested in a ligation assay employing plasmid pZ189). The highest proportion of deletion types of mutations was found in plasmids passaged through host cells in the presence of MMC. Following reductive activation MMC is converted into an alkylating agent (Tomasz and Lipman, 1981) that can cause cross-links and strandbreaks, such that deletions may result (Tomasz et al, 1987; Basu et al, 1993). An important observation of our study was that MMC induces a very specific type of deletion in plasmids passaged through FA cells (Table TV). The base at the breakpoint is a guanine that is known as the target base for MMC (Li and Kohn, 1991). Moreover, the breakpoint was found to be flanked by indirect repeats of the signal heptamer type described by Laquerbe et al (1995) as a common breakpoint in the HPRT gene of FA lymphoblasts. These authors speculate that on the basis of similarity of this signal sequence with the heptamer sequence that directs cleavage and rejoining in the assembly of immunoglobulin and T-cell receptor genes, steps in common with the V(D)J recombinational process may be defective in FA cells. This interpretation is in line with the statement by Gille et al (1994) that 'mammalian cells under oxidative stress exhibit a hyper-recombinational phenotype* and with the result of our experiments involving replication of plasmids through FA cells in the presence of MMC if one assumes that at least part of the MMC effects are mediated by reactive oxygen species

References Auerbach,A.D. (1993) Fanconi anemia diagnosis and the diepoxybutane (DEB) test Exp. Hematol. (New York), 21, 731-733. Aruma,0.I., Halliwell.B., Gajewski,E. and Dizdaroglu.M. (1989) Damage to the basis in DNA induced by hydrogen peroxide and ferric ion chelates. J. Biol. Chem., 264, 20509-20512. Basu,A.K., Hanrahan.C.J., Malia,S.A., Kumar,S., Bizanek,R. and Tomasz,M. (1993) Effect of site-specifically located mitomycin C DNA monoadducts on in vitro DNA synthesis by DNA polymerase. Biochemistry, 32,4708-4718. Bessho,T., Tano.K., Kasai.H. and Nishimura,S. (1992) Deficiency of 8-hydroxyguanosine endonuclease activity and accumulation of 8hydroxyguanosine in mutator strain (mutm) of Escherichia coli. Biochem. Biophys. Res. Commun., 188, 372—378. Bredberg.A., Sandor,Z. and Brant^M. (1995) Mutational response of Fanconi anemia cells to shuttle vector site-specific psoralen cross-links. Carcinogenesis, 16, 555—561. Cabrenajvl, Nghiem.Y. and MillerJ.H. (1988) mutM, a second mutator locus in E.coli that generates G:C to T:A transversions. J. Bacteriol., 170, 5404-5407. Cheng,K.C, Cahill.D.S., Kasai.H., Nishimura,S. and Loeb.L.A. (1992) 8hydroxyguanine, an abundant form of oxidative DNA damage, causes G to T and A to C substitutions. J. Biol. Chem., 267, 166-172. Decuyper-DeberghJD., PietteJ. andd Van de Vorst,A. (1987) Singlet oxygeninduced mutations in M13 lacZ phage DNA. EMBO J., 6, 3155-3161. Degan,P., Bonassi.S., De Caterina>l., KokinaX.G., Pinto.L., ScopacasaJ., Zatterale.A., Calzone,R. and Pagano.G. (1995) In vivo accumulation of 8hydroxy-2'-deoxyguanosine in DNA correlates with release of reactive oxygen species in Fanconi's anemia families. Carcinogenesis, 16, 735—741.

75

W.Liebetrau et al

76

Protic-Sabljic.M., Seetharam.S., Seidman,M.M. and Kraemer.ICH. (1986) An S V40-transformed xeroderma pigmentosum group D cell line: establishment, ultraviolet sensitivity, transfection effiency and plasmid mutation induction. Mutat. Res., 166, 287-294. Reid.T.M. and Loeb.L.A. (1993) Tandem double CC to TT mutations are produced by reactive oxygen species. Proc. Natl. Acad. Sci. USA, 90, 3904-3907. RetelJ., Hoebee.B., BraunJ.E., LutgerinkJ.T, Akker van der,E., WanamartaA-H., Joenje.H. and Lafleur.M.V. (1993) Mutational specificity of oxidative DNA damage. Mutat. Res., 299, 165-182. RUnger.T.M., SobottaJ1., Dekant,B., Moller.K., Bauer.C. and Kraemer.K.H. (1993) In-vivo assessment of DNA ligation efficiency and fidelity in cells from patients with Fanconi's anemia and other cancer-prone disorders. Toxicol. Lett., 67, 309-324. RQnger.T.M. and Kraemer.K H. (1989) Joining of linear plasmid DNA is reduced and error-prone in Bloom's syndrome cells. EMBOJ., 8,1419-1425. Saito.H., Hammond.A.T. and Moses.R.E. (1993) Hypersensitivity to oxygen is a uniform and secondary defect in Fanconi anemia cells. Mutat. Res., 294, 255-262. Saito.H., HammondAT. and Moses.R.E. (1995) The effect of low oxygen tension on the in vitro- replicative life span of human diploid fibroblast cells and their transformed derivates. Exp. Cell Res., Ill, 272-279. Seidman.M.M., Dixon.K., Razzaque.A., Zagursky.RJ. and Bergman,M.L. (1985) A shuttle vector plasmid for studying carcinogen-induced point mutations in mammalian cells. Gene, 38, 233-237. Schindler.D , Kubbiesjvi., Hoehn.H., Schinzel.A. and Rabinovitch.P.S. (1987) Confirmation of Fanconi's anemia and detection of a chromosomal aberration (1Q12-32 triplication) via BrdU/Hoechst flow cytometry. Am. J. Pediatr Hematol. Oncol, 9, 172-177. Schmdler, D., Hoehn, H. (1988) Fanconi anemia mutation causes cellular susceptibility to ambient oxygen. Am. J. Hum. Genet., 43, 429-435. Schroeder.T M., Anschuetz,F. and KnoppA (1964) Spontane Chromosomenaberrationen bei familiarer Panmyelopathie. Humangenetik, 1, 194-196. Seetharam.S., Protic-Sabljic,M., Seidman.M.M. and Kraemer.K.H. (1987) Abnormal ultraviolet mutagenic spectrum in plasmid DNA replicated in cultured fibroblasts from a patient with me skin cancer-prone disease, xeroderma pigmentosum. /. Clin. Invest., 80, 1613—1617. Seetharam.S., Kraemer.K.H., Waters,H.L. and Seidman,M.M. (1991) Ultraviolet mutational spectrum in a shuttle vector propagated in xeroderma pigmentosum lymphoblastoid cells and fibroblasts. Mutat. Res., 254,97-105. Seyschab.H., Friedl.R., Sun,Y., Schindler.D., Hoehn.H., Hentze.S. and Schroeder-Kurth.T. (1995) Comparative evaluation of diepoxybutane sensitivity and cell cycle blockage in the diagnosis of Fanconi anemia. Blood, 85, 2233-2237. Shibutani,S., Takeshita>I. and GrollmanAP (1991) Insertion of specific bases during DNA synthesis past the oxidation-damaged base 8-oxodG. Nature, 349, 431^34. Skrikant,S., MudipalliA, MaccubbinAE. and Gurtoo.H.L. (1994) Mutations in a shuttle vector exposed to activated mitomycin C. Mol. Carcinogen., 10, 23-29. Strathdee.CA, Gavish.H., Shannon.W. and Buchwald.M. (1992) Cloning of cDNAs for Fanconi's anemia by functional complementation. Nature, 356, 763-767. Szatrowski.T.P. and Nathan.C.F. (1991) Production of large amounts of hydrogen peroxide by human tumor cells Cancer Res., 51, 794-798. Takeuchi.T. and Morimoto.K. (1993) Increased formation of 8-hydroxydeoxyguanosine, an oxidative DNA damage, in lymphoblasts from Fanconi anemia's patients due to possible catalase deficiency. Carcinogenesis, 14, 1115-1120. Tkeshelashvili.L.K., McBride.T, Spence.K. and Loeb.L.A. (1991) Mutation spectrum of copper-induced DNA damage. J. BioL Chem., 266, 6401-6406. Tomasz,M. and Lipman.R. (1981) Reductive metabolism and alkylating activity of mitomycin C induced by rat liver microsomes. Biochemistry, 20, 5056-5061. Tomasz^M., Lipman.R., Chowdary.D., PawlakJ., Nerdine.G.L. and Nakanishi.K. (1987) Isolation and structure of a covalent cross-link adduct between mitomycin C and DNA. Science, 235, 1204-1208. Walsh,C.E., Grompe,M., Vanin,E., BuchwakLM., YoungJM.S., Nienhuis.A.W. and Liu,J.M. (1994) A functionally active retrovmis vector for gene therapy in Fanconi anemia group C. Blood, 84, 453—459. Weaver D.T. and DePamphilis.M.L. (1982) Specific sequences in native DNA that arrest synthesis by DNA polymerase alpha. J. BioL Chem., 257, 2075-2086. Whitneyjvl, Thayerjvl., Reifsteck,C, Olson.S., SmithJL, Jakobs,P., Leach.R. et aL (1995) Microcell mediated chromosom transfer maps the Fanconi anemia group D gene to chromosome 3p. Nature Genet., 11, 341-343.


Dusre,L., Rajagopalan.S., Eliot,H.M., CoveyJ.M. and SinhaJJ.K. (1990) DNA interstrand cross-link and free radical formation in a human multidrugresistent cell line from mitomycin C and its analogues. Cancer Res., 50, 648-652. Fanconi Anaemia/Breast Cancer Consortium (1996) Positional cloning of the Fanconi anaemia group A gene. Nature Genet., 14, 324-328. FeigJD.I. and Loeb.L.A. (1994) Oxygen radical induced mutagenesis is DNA polymerase specific. J. Mol. BioL, 235, 33-41. Feig.D.L, Sowers.L.C. and Loeb.L.A. (1994) Reverse chemical mutagenesis: identification of the mutagenic lesions resulting from reactive oxygen species-mediated damage to DNA. Proc. Natl. Acad Sci. USA, 91, 66096613. GilleJ.J., Berkel van.C.G. and Joenje.H. (1994) Mutagenicity of metabolic oxygen radicals in mammalian cell cultures. Carcinogenesis, IS, 2695—2699. Gozukara,E.M, Parris.C.N., Weber.C.A., Salazar.E.P., Seidman.M.M., WatkinsJ.F., Prakash, L. and Kraemer,K.H. (1994) The human DNA repair gene, ERCC2 (XPD), corrects ultraviolet hypersensitivity and ultraviolet hypermutability of a shuttle vector replicated in Xeroderma pigmentosum group D cells. Cancer Res., 54, 3837-3844. Hsie,A.W., Recio.L., Katz,D.S., Lee,C.Q., Wagner,M. and Schenley.R.L. (1986) Evidence for reactive oxygen species inducing mutations in mammalian cells. Proc. Natl. Acad. Sci. USA, 83, 9616-9620. Joenje.H., ArwertF., ErikssonA-W., De Koning.H. and OostraA-B- (1981) Oxygen-dependence of chromosomal alterations in Fanconi's Anemia. Nature, 290, 142-143 Joenje.H., Lo Ten FoeJ.R., OostraA.B., Berkel van.C.G.M., Roolmans,M.A., Schroeder-KurthJ., Wegner,R.D. et al. (1995) Classification of Fanconi anemia patients by complementation analysis: Evidence for a fifth genetic subtype. Blood, 86, 2156-2160. Kraemer.H.K. and Seidman.M.M. (1989) Use of supF, an E.coli tyrosine suppressor tRNA gene, as a mutagenic target in shuttle vector plasmids. Mutat. Res., 220, 61-72. Kornberg.A. and Baker.T.A. (1992) Repair, Recombination, Transformation, Restriction and Modification in DNA Replication. W.H.Freeman and Co, New York, p. 771. Kunkel.T.A. (1984) Mutational specificity of depurination. Proc. Natl. Acad. Sci. USA, 81, 1494-1498. LaquerbeA-, Moustacchi.E., FuscoeJ.C. and Papadopoulo,D. (1995) The molecular mechanism underlying formation of deletions in Fanconi anemia cells may involved a site-specific recombination. Proc. Natl. Acad Sci. USA, 92, 831-835. Li.V.S. and Kohn.H. (1991) Studies on the binding specificity for mitomycin c-DNA monoalkylation processes. J. Am. Chem. Soc., 113, 275-283. Lindahl,T. (1993) Instability and decay of the primary structure of DNA. Nature, 362, 709-715. Lo Ten FoeJ.R., Rooimans,M.A., Bosnoyan-Collins.L., Alon.N., Wijker^M., Parker,L., Lightfoot,F. et al. (1996) Expression cloning of a cDNA for the major Fanconi aneamia gene, FAA. Nature Genet., 14, 320-323. Loeb,L.A., James,E.A., WaltersdorphA-M. and Klebanoff.S.J. (1988) Mutagenesis by autoxidation of iron with isolated DNA. Proc. Natl. Acad Sci. USA, 85,3918-3922. Moody,C.S. and Hassan.H.M. (1982) Mutagenicity of oxygen free radicals. Proc. Natl. Acad. Sci. USA, 79, 2855-2859. Moraes,E.C, Keyse.S.M., Pidoux.M. and Tyrrell.R.M. (1989) The spectrum of mutations generated by passage of a hydrogen peroxide-damaged shuttle vector plasmid through mammalian host Nucleic Acids Res., 17,8301-8312. Moraes,EC., Keyse.S.M. and Tyrrell.R.M. (1990) Mutagenesis by hydrogen peroxide treatment of mammalian cells: a molecular analysis. Carcinogenesis, 11, 283-293. Nicol.CJ., Harrison.M.L., Laposa,R.R., Gimelstein.I.L. and WeUs.P.G. (1995) A teratologic suppressor role for p53 in benzo(a)pyrene-treated transgenic p53-deficient mice. Nature Genet., 10, 181-187. OllerA-R- and Thilly.W.G. (1992) Mutational spectra in human B-cells. J. Mol. BioL. 228, 813-826. PietteJ., Decuyper-DeberghJ}. and Gamperji. (1985) Mutagenesis of the lac promoter region in M13 mplO phage DNA by 4'-hydroxymethyl-4,5',8trimethylpsoralen. Proc. Natl. Acad. Sci. USA, 82, 7355-7359. Poo^M., Gross.O., Epe.B., Pflaum,M. and Hoehn.H. (1996) Cell cycle defect in connection with oxygen and iron sensitivity in Fanconi anemia lymphoblastoid cells. Exp. Cell Res., Ill, 262-268. Pritsos.C.A. and SartorelliAC. (1986) Generation of reactive oxygen radicals through bioactivation of mitomycin antibiotics. Cancer Res., 46, 35283532. PronkJ.C, Gibson.R.A., SavoiaA. Wijker.M., Morgan,N.V., Melchionda.S., FonLD. et al. (1995) Localisation of the Fanconi aneamia complementation group A gene to chromosome 16q24.3. Nature Genet., 11, 338-340.

Mutagenicity in Fanconi's anaemia cells Youssoufian,H. (1994) Localization of Fanconi anemia C protein to the cytoplasm of mammalian cells. Proc. Natl. Acad. Sci. USA, 91, 7975-7979. YamashitaX, Barter.D.L., Zhu.Y, Wu,N. and D'Andrca,A.D. (1994) The Fanconi anemia polypeptide FACC is localized to the cytoplasm. Proc. Natl. Acad. Sci. USA, 91, 6712-6716. Youssoufian.H., Auerbach.A.D., Verlander.P.C, Steimle,V. and Mach.B. (1995) Identification of cytosolic proteins that bind to the Fanconi anemia complementation group C polypeptide in vitro. Evidence for a multimeric complex. /. Biol. Chem., 270, 9876-9882. Received on July 19, 1996; accepted on December 16, 1996


77