- Email: [email protected]
UV-induced base substitution mutations in a shuttle vector plasmid propagated in group C xeroderma pigmentosum cells
Mutation Research, DNA Repair, 273 (1992) 213-220
213
© 1992 Elsevier Science Publishers B.V. All rights reserved 0921-8777/92/$05.00
MUTDNA 00192
UV-induced base substitution mutations in a shuttle vector plasmid propagated in group C xeroderma pigmentosum cells Takashi Yagi, Mayumi Sato, Junko Tatsumi-Miyajima and Hiraku Takebe Department of Experimental Radiology, Faculty of Medicine, Kyoto University, Yoshida-konoe.cho, Sakyo-ku, Kyoto 606 (Japan) (Accepted 9 May 1991)
Keywords: Point mutation; Ultraviolet; Xeroderma pigmentosum group C
Summary To assess the contribution to mutagenesis of human DNA repair defects, the UV-irradiated shuttle vector plasmid pZ189 was propagated in fibroblasts derived from a xeroderma pigmentosum (XP) patient in DNA repair complementation group C. In comparison to results with DNA repair-proficient human cells (WI-38 VAI3), UV-irradiated pZ189 propagated in the XP-C (XP4PA(SV)) cells showed fewer surviving plasmids and a higher frequency of mutated plasmids. Base sequence analysis of 67 mutated plasmids recovered from the XP-C cells revealed similar classes of point mutations and mutation spectrum, and a higher frequency of G : C to A : T transitions along with a lower frequency of transversions among plasmids with single or tandem mutations compared to plasmids recovered from the normal line. Most single-base substitution mutations (83%) occurred at G : C base pairs in which the 5'.adjacent base of the cytosine was thymine or cytosine, These results indicate that the DNA repair defects in XP-C, in comparison to data previously reported for XP-A, XP-D and XP-F, result in different UV survival and mutation frequency but in similar types of base substitution mutations.
This study was focused on investigating the contribution to UV-induced mutagenesis of the DNA repair defect in a xeroderma pigmentosum (XP) patient in complementation group C. XP is an autosomal recessive disease that is characterized by a high frequency of neoplasms in sun-ex-
Correspondence: Dr. T. Yagi, Department of Experimental
Radiology, Faculty of Medicine,Kyoto University,Yoshidakonoe-cho, Sakyo-ku, Kyoto 606 (Japan).
XP, xeroderma pigmentosum;XP-A, XP.C, XP-F, xeroderma pigmento~um complement~:[3n groups A,
Abbre~'iations:
C, F; UV, ultraviolet; UDS, unscheduled DNA synthesis,
posed skin (reviewed in Kraemer et al., 1987; Takebe et al., 1987). Cultured cells derived from XP patients are hypersensitive to UV radiation, have a reduced capacity for DNA repair and produce a high frequency of UV-induced mutations. There are at least 7 DNA excision-repair complementation groups (A-G) in XP cells (Kraemer et al., 1987; Takebe et al., 1987). XP DNA repair complementation group C is the most common worldwide. A shuttle vector plasmid, pZ189 (Seidman et al., 1985), was utilized to measure UV-induced point mutations following replication in skin fibroblast cells derived from an Algerian XP pa-
214
tient in complementation group C compared with normal human cells, pZ189 carries a 190-bp bacterial suppressor tRNA-g-cn¢, ~,-s'ap"~;~-s-a~,-~rker gene for mutations. It also contains the origin of replication and T antigen gene from SV40 to permit replication in human cells, and the origin of replication and the fl-lactamase gene from pBR327 for growth in E. coil and selection for ampicillin resistance. The plasmid is irradiated with UV and then transfected into the human cells where DNA repair, mutation and replication occur in a process dependent on host cell enzymes. Replicated plasmid is harvested and used to transform an indicator strain of bacteria. The number of ampicillin-resistant bacterial colonies reflects plasmid survival. The supF gene suppresses an amber mutation in the//-galactosidase gene (lacZ) in the indicator strain of E. coll. Bacteria with ti~e wild-type supF gene make dark blue colonies while bacteria containing the mutant supF gene make white or light blue colonies on agar plates containing 5-bromo-4chloro-3-indoyl/3-D-galactoside (X-gal). Thus, the mutation frequency can be determined by counting the proportion of white or light blue colonies compared to the total number of colonies. Mutared plasmids are purified from the bacterial colonies and the nucleotide sequence of the supF gen¢ is determined, The frequency and type of UV-induced base substitution mutations occurring in pZI89 propagated in cells from XP-A (Brcdberg et al., 1986; Yagi et aL, 1991), XP-D (Seethram ¢t al., 1 9 8 7 ) and XP.F (Yagi et al., 1991) patients was previously reported to be different from those in normal human cells. Hence, we examined whether the frequency and type of those mutations occurring in XP-C cells are different from those in XP cells in other complementation groups and from those in normal cells, Materials and methods
Cells Established fibroblast cell lines were used in this study. XP4PA(SV) is a cell line immortalized by transfection of a plasmid containing replication origin defective promoter and early region of SV40 virus (Daya-Grosjean et al., 1987). The
original XP4PA patient is an Algerian boy who was prenatally diagnosed as XP (complementation group C) (Halley et al., 1979), and XP4PA fibroblast cells were established after his birth. The XP4PA fibroblasts had 12% of the normal UV-induced unscheduled DNA synthesis (UDS) ability (a measure of DNA excision repair) (HalIcy et al., 1979) and were moderately sensitive to killing by UV (D37 = 1.2 j/m2). To compare the XP-C cells with normal human cells, an SV40-transformed normal fibroblast cell line, WI-38 VA13 (Girardi et al., 1965), was used. The parental cell line WI-38 was established by L. Hayflick from normal embryonic lung tissue of a Caucasian female, and showed normal sensitivity to killing by UV (D37 = 6.6 j/m2). The UV-induced mutation data of pZ189 passed through WI-38 VA13 cells have been published by us (Yagi et al., 1991) and are used in this study for comparison. All cells were cultured in Dulbecco's modified minimum esential medium (Nissui, Tokyo) supplemented with 10% fetal bovine serum (Hyclone, Logan, UT).
Methods pZ189 was extracted from overnight cultures of E, coli DHI having pZ189 by alkoline lysis and purified by cesium chloride ultracentrifugation (Maniatis et al., 1982). The closed circular pg189 band was carefully removed from the centrifuge tube with a needle and the plasmid solution was dialyzed against TE buffer and used for experiments. pZI89 dissolved in TE buffer at a eoncentration of 45/zg/ml was irradiated with a bank of two 10.W Toshiba germicidal lamps (predominately 254-nm UV light) at a dose rate of 3.4 J/m2/s. The human cells were trypsinized, washed and suspended in Duibecco's phosphatebuffered saline (pH 7.4-7.6). 2 X 107 cells plus 9 /zg UV-treated pZI89 in 0,2 ml buffer were placed in an electroporation chamber (electrodes 0.3 cm apart) (ZA-1200, PDS lnc., Madison, WI) and transfected with plasmid by electric pulses (500 V, 4 times) (Tatsuka et al., 1988). The cells were plated in 10-cm dishes and incubated in a CO 2 incubator at 37°C, After 72 h incubation, the plasmid was extracted from the cells by the
215
method of Hirt (1967), and digested by restriction endonuclease Dpnl (Boehringer-Yamanouchi, Tokyo) to eliminate,.~on~replicated input plasmid with the bacterial methyrlation pattern (Seidman et ai., 1985). E. coli MBM7070 which has an amber mutation in its lacZ gene was transformed by the plasmid and spread onto LB agar plates which contained ampicillin, the colorless dye, 5-bromo4-chloro-3-indoyl/3-D-galactoside (X-gaD and isopropyl fl-o-thiogalactoside (IPTG) (Nacalai Tesque, Kyoto, Japan). After 24 h incubation a t 37°C, blue colonies with normal supF and white or light blue colonies with mutated supF appeared on the plate. The total number of colonies w a s counted as a n indication of plasmid survival and the proportion of white or light blue colonies determined as a measure of mutation frequency, Mutated plasmids were purified from overnight cultures and the nucl¢otide sequence of the supF gene of the plasmid was determined with K / R T universal dideoxy sequencing kit (Promega, Madison, WI). To obtain independent mutant clones, fewer than 5 mutants were isolated from each independent transfection and only 1 mutant was scored if several plasmids had identical mutations.
Statistics Statistical comparisons were performed with Fisher's exact test for difference in proportions (Armitage, 1971). P values for a 1-tailed test are presented,
Results
Plasmid surt,it,al and mutagenesis The shuttle vector plasmid pZ189 was irradiated with UV, transfected into the XP-C and normal cells, incubated for 3 days to permit repair, mutation and growth, extracted and used to transform bacteria to ampicillin resistance. UV irradiation of pZ189 showed a dose-dependent reduction in the number of ampicillin-resistant bacterial colonies (Fig. 1). Survival of the plasmid recovered from the XP-C line was lower than that from the normal cell line. At a dose of 1 0 0 0 J / m 2 to the plasmid the relative number of bac-
1
~,~..,.., ~'x... ~
,
,
\"
~ ~ ~ m.1 ~ ~c, a-
,
,
",
".\ \\ \
• \
\ "\ \\\ "
10-2
I o
I
I
I
I
soo looo lsoo 20o0 UV Dose (j/m2)
Fig. 1. Survival of UV-irradiated pZ189 replicated in XP-C (XP4PA(SV)) (e) and normal (WI-38 VAI3) (broken line) cell lines. The relative number of ampiciilin-resistant bacterial colonies obtained after repair and replication of UV-irradiated pZ189 in the cells followed by transformation of the indicator E. coil is shown. The data show mean values from 2-4 independent transfection experiments for each cell line. The standard error is smaller than the symbols for most points. The data of WI-38 VAI3 are from Yagi et al. (1991).
terial colonies recovered from the XP-C line was about 3.4-fold less than from the normal line. UV irradiation to pZ189 increased the proportion of white and light blue colonies with a mutated supF gene in a dose-dependent manner (Fig. 2). The background plasmid mutation frequeneies were 2.1 × 10 -'~ and 0.9 x 10 -'~ with the XP-C and normal lines, respectively, and increased 10-190-fold following treatment of the plasmid with 500-2500 J / r e ' UV. The UV-induced l~mtation frequency was greater with the XP-C th~n with the normal line. At 1000 J / m : UV to the plasmid the mutation frequencies with the XP-C and the normal lines were 65 × 10- "~ and 17 × 10 -3, respectively (Fig. 2). These results reflect the UV hypermutability of XP cells.
Base sequence analysis Analysis of 97 supF mutant plasmids with the XP-C cell line by agarose gel electrophoresis and nucleotide sequence analysis revealed that 31% of the plasmids contained large deletions (average size 84 base pairs). The mutant plasmids without deletions contained point mutations. Table 1 shows the classes of point mutations
216 1
,
,
I
,
,
, $_
e~ ~
~ 10-1 -
//'/e/"/" t~ m-2 =1
_ ........
~ //
tE l o _ a _ t,,
I 104
0
_
I I I I I 500 1000 1500 2000 2fi00 UV Dose (J/m 2)
Fig. 2. Mutation of UV-irradiateo pZ189 replicated in XP-C (XP4PA(SV)) (e) and normal (WI-38 VAI3) (broken line) cell lines. The frequency of white and light blue bacterial colonies obtained after repair and replication of UV-irradiated p Z 1 8 9 in the cells followed by transformation of the indicator E. c o l i is shown. The data show mean values from 2-4 independent transfection experiments for each cell line. The standard error is smaller than the symbols for most points. The data of W1-38 VAI3 are from Yagi et al. (1991).
identified following base sequence analysis of 67 independent mutant plasmids of wild-type size recovered from the XP-C cell line. Plasmids are classified as to whether they have a single-base substitution, tandem base substitutions (2 base substitutions 0-2 bases apart, or 3 adjacent base
substitutions), multiple base substitutions (2 base substitutions m o r e than 3 bases a p a r t or 3 or m o r e base substitutions), and single-base insertion or deletion. Single-base substitutions were slightly m o r e f r e q u e n t with X P - C (72%, P = 0.058) t h a n with the normal cells (58%). T h e p r o p o r t i o n o f mutated plasmids with multiple base substitutions was slightly lower with XP-C (12%, P = 0.076) than with the n o r m a l cells (22%) (Table 1). T h e maximum n u m b e r o f base substitutions f o u n d in the supF gene in one plasmid was 3 with the XP-C cells and 4 with the normal line. Single-base frameshift mutations were not seen with the XP-C cell line. A m o n g the m u t a n t plasmids with single and t a n d e m base substitution mutations, transition mutations were p r e d o m i n a n t with both X P - C and
normal cell lines (Table 2). The transitions were slightly more frequent with the XP-C (75%, P = 0.057) than with the normal cells (62%). It is quite striking that the most frequent type of base substitution mutation was the G : C to A : T transition, accounting for more than half of the single or tandem base substitution mutations with both cell lines, The frequency of G : C to A : T transitions was significantly higher with the XP-C (71%, P = 0.040) cells than with the normal cells (56%). The frequency of G: C to C:G transversions was significantly lower with the XP-C (3%, P - 0.050)
TABLE I CLASSES OF POINT MUTATIONS INDUCED IN UV-IRRADIATED pZ189 a Number of plasmids with base changes (%)
Independentplasmidssequenced Point mutations Single-base substitution Tandem-base substitution Multiple-base substitution Frameshifl Single-base insertion
Single.basedeletion " 250-2500 J/m "~UV irradiation to pZ189. h The data are from Yagi el el, (1991), c These mutants also have a single-base substitution. * 0.05 < P < 0.1 vs. normal.
XP-C 67 (100%)
Normal h 91 (100%)
48 (72%)* II (17%) 8 (12%)*
53 (58%) 15 (16%) 20 (22%)
(I (0%) (1 ((1%)
i
(1%)
2 c (2%)
217
cells than with the normal cells (II%). While 89% of the single-base substitution mutations involved G : C base pairs with the XP-C cells, only 11% involved A : T base pairs.
TABLE 2 TYPES O F SINGLE O R T A N D E M BASE SUBSTITUTION M U T A T I O N S I N D U C E D IN U V - I R R A D I A T E D
pzI89 ~ Number of base changes (%) XP-C
Normal b
Transitions G : C to A : T A:TtoG:C
54 ( 7 5 % ) * 51 (71%) * * 3 (4%)
52 (62%) 47 (56%)
Transversions G:CtoT:A G : C to C : G A : T to T : A
18 ( 2 5 % ) * I! (15%)
32 (38%) 18 (21%)
A:T to C:G
I (1%)
5 (6%)
2
(3%) * *
9 (11%)
4
(6%)
3
(4%)
2 (2%)
72 (100%)
Total
Mutation spectrum The distribution of single and tandem base substitutions in the supF gene with the XP-C a n d normal cells is shown in Fig. 3. All mutations were found between base pairs 99 and 177. The mutational patterns show some similarities and s o m e differences between the cell lines. The point mutations were not distributed randomly in the supF gene but appeared preferentially at certain site s. There was a G : C to A: T transition hotspot at base pair 156 with both lines accounting for 17% and 15% of all single and tandem mutations with
84 (100%)
~' 250-1000 ,l/re" U V irradiation to pZ189. The data are from Yagi et al. (1991). • 0.05 < P < 0.1 vs. normal: * * P < 0.05 vs. normal,
4"
4'
4.
4.
4"
4.
4"
4"
4"
4"
1"/0
180
ATTA•CTGTGGTGGGGTT•••GAG•GG••AAAGG•AG•AGA•TCTAAATCTGC•GTCAT•GA•TT•GAAGGTT•GAAT••TT••C••A••AC•A XP-C 90
100
110
180
130
140
150
160
4" . . . . . . . . . .4. . . . . . . . . . .4-. . . . . . . . . . 4". . . . . . . . . . 4". . . . . . . . . . 4,. . . . . . . . . . +. . . . . . . . . . 4,. . . . . . . . . . 4". . . .
A.A AA A~A
90
100
AT
110
G AA.~ TA TA TA T_A
120
4" . . . . . . . . .4". . . . . . . . . 4". . . . . . . . . 4..
A~_ AT A A AA TT C _C.._T T_T A
e A
GT GT
130
A A
140
C G
AAA A A A A A A A A A A A T
150
AC A A G G G G
G
G
TT TT TT T T T T T T T T
160
- . . . . . . . 4.. . . . . . . . . 4. . . . . . . . . .4- . . . . . .
T AA AA A(~
A A A A A A T
G_~ TA~_.ATC T,A~ A TA A AA T AA T CA A
A A A A A A T
170
- . 4" . . . . . . . .
T
T C
4.
TT T T AT
- 4. . . . . . . . .
IBO - 4- . m - -
T_T~C T ~ A TT TT TT T T
A A
Fig. 3. Location of single and tandem base substitution mutations found in the s'upF gene of UV-irradiated pZ189 replicated in XP-C (XP4PA(SV)) and normal (WI-38 VAI3) cells. The supF suppressor tRNA sequences start at base pair 99 and end at base pair 183. Each lelter represents a single-base-pair substitution mutation found in an independent plasmid. Tandem base substitution mutations are underlined. The data of WI-38 VAI3 are from Yagi et al. (1991).
218
the XP-C and normal lines, respectively. At base pair 159 the (3: C to A: T transition and G: C to T : A transversion were frequent with the XP-C (10%) and less frequent with the normal cell line (7%). At base pair 168 the G : C to A : T transition in single point mutations was slightly more frequent with the XP-C line than with the normal line (23% vs. 9 % , / ' = 0.057). Fig. 3 shows that most single-base substitution mutations occurred at G : C base pairs where the 5' neighboring base of the mutated cytosine is a cytosine or thymine (XP-C 83%, normal 91%). In contrast, about a half of the single-base substitutions occurred at G : C base pairs where the 3' base adjacent to the mutated cytosine was a pyrimidine (XP-C 52%, normal 51%). All tandem mutations occurred at 5'-CC sequences and only 8% of the single base substitution mutations were found at 5'-TT sequences with XP-C cells, Discussion Cells derived from XP patients show hypersensitivity to killing by UV and UV hypermutability (Kraemer et al., 1987; Takebe et al., 1987). These cellular abnormalities are reflected in the increased plasmid killing and mutations observed in UV-irradiated pZ189 passed through the XP-C cells (Figs. I and 2). These results are in agreement with similar studies performed with UV. irradiated pZ189 passed through cells from XP-A (Bredberg et al., 1986; Yagi et al., 1 9 9 1 ; Seetharam et al., 1990, 1991), XP.D (Seetharam et al., 1987) and XP-F (Yagi et al., 1991) patients, The survival and mutation frequency of the UVirradiated pZ189 passed through XP-C cells are between those through XP-A and normal (Bredberg et al., 1986; Yagi et al., 1991). Thus the present research provides further support for the use of pZ189 to measure DNA repair defects in human cells, Analysis of the classes of UV-associated mutations recovered in pZI89 revealed a slight increase in the proportion of plasmids with singlebase substitution mutations and a slight decrease in plasmids with multiple.base substitutions with the XP-C line compared to the normal line (0,05 < P < 0.1 vs. normal) (Table 1), The residual level of excision repair ability (UDS 12% of nor-
mal) in the XP-C cells may n o t have caused the significantly large difference between the XP-C and normal cells. A significantly abnormal proportion of these base substitutions was clearly observed with XP-A (Bredberg et al., 1986; Yagi et al., 1991) and XP-D fibroblast lines (Seetharam et al., 1987) in comparison to the normal lines. These studies strongly suggest that defective DNA repair is linked to a low frequency of UV-induced multiple-base substitutions and to a high frequency of single-base substitutions. The multiplebase substitutions have been shown to be dramatically increased by creating a single-strand nick in the plasmid (Seidman et al., 1987). Normal cells, unlike XP cells, can make a nick at the UV-induced DNA damage, which leads to induction of the multiple-base substitutions. DNA repair deficiency in the XP cells is also associated with an increased frequency of G: C to A : T transitions (with consequent decreased frequency of transversions) in UV-treated pZ189. In the present study, among the single and tandem base substitution mutations, the percentage of the G : C to A : T transition was significantly higher in the XP-C than in the normal cells (P = 0.04) (Table 2). This was also observed with XP-A (Bredberg et al., 1986; Yagi et al., 1991), XP.F (Yagi et al., 1991) and XP-D (Seetharam et al., 1987) fibroblast lines. Excision repair-proficient monkey cells also had a significantly lower frequency of the G : C to A : T transition with UV-treated pZI89 than XP cells (Keyse et al., 1988; Hauser et al., 1986). Most base substitution mutations (89%) were found at G: C base pairs (Table 2) and 83% of 5' neighboring bases of the mutated cytosines in the single-base substitutions are thymine or cytosine with the XP-C line (Fig. 3). These data are consistent with earlier studies with the XP-A, XP-D and XP-F cells (Bredberg et al., 1986; Seetharam et al., 1987; Yagi et al., 1991). These results suggest that the Y cytosine of 5°-TC and 5'-CC sequences is the major mutational target of UV radiation. All tandem mutations occurred at 5'CC sequences with the XP-C cells (Fig. 3). These dipyrimidine sites can be involved in either cyclobutane dimer or pyrimidine (6-4) pyrimidone photoproducts (Brash et al., 1987; Mitchell and Nairns, 1989) which have been shown to be muta-
219
genic in bacterial cells (Brash and Haseltine, 1982; Conlondre and Miller, 1977; Wood et al., 1984). No mutation occurred at 5'-Tr sequences that form the major UV photoproduct, cyclobutane thymine dimer (about 60% of the total photoproducts in pZ189) (Brash et al., 1987; Kraemer et al., 1988), As in earlier studies with I V-irradiated pZ189, the major UV photoproduct is not the major mutagenic lesion. In pZ189, (6-4) photoproducts have been measured to occur at a frequency 70% of that of cyclobutane dimers at 5'-TC sites and 30% at 5'-CC sites (Brash et al., 1987; Kraemer et al., 1988). Studies of photoreactivation of UV-treated pZ189 indicate that both cyclobutane dimers and (6-4) photoproducts are mutagenic lesions (Brash et al., 1987; Kraemer et al., 1988). This suggests that 3' cytosine in unrepaired 5'-TC and 5'-CC sites causes G : C to A : T transition during DNA replication, and a high frequency of this mutation in XP cells is ascribed to the faulty repairing of these lesions, The mutation spectrum with UV-treated pZ189 passed through XP-C cells showed a tandem mutation hotspot at base pairs 122-123 and singlebase substitution hotspots at base pairs 156, 159 and 168 (Fig. 3)..The proportion of the number of mutants included in each hotspot with XP-C is not significantly different from that with normal cells. The normal cells have a transversion hotspot at base pair 133 but the XP-C cells have no mutation at the base pair ( P - 0.02). Base pair 168 showed a high frequency of G : C to A : T transition mutations with all XP cell lines tested including XP-A (Bredberg et al., 1986; Yagi et al., 1991), XP-D (Seetharam et al., 1987) and XP-F (Yagi et al., 1991) fibroblasts, XP-A lyrephoblasts (Seetharam et al., 1990, 1991) and the repair-proficient lymphoblasts (Seetharam et al., 1990, 1991) but not with the repair-proficient WI-38 VA13 (Yagi et al., 1991; Fig. 3) or GM0637 (SV) cells (Bredberg et al., 1986). This is in keeping with the observation of mutational hotspot variability that has recently been discussed (Seetharam et al., 1990) which suggested that the major determinants of mutational hotspots were not genetic polymorphisms or repair capacity of the cells but cellular factors that can influence the probability of mutagenesis at particular sites. The XP-C cells have very low but significantly
measurable amounts of DNA excision repair capacity (12% of the normal level of UDS) (Halley et al., 1979). This residual DNA repair specificaily works on transcribed genes but not on nontranscribed genes in the genome of UV-irradiated XP-C cells (Venema et al., 1990). The significantly higher frequenc~ of UV-induced G : C to A : T transition mutations in XP-C cells than in normal cells indicates that the residual DNA repair does not work efficiently on UV-damaged pZ189, despite the classes of point mutations (Table 1) and the mutation spectrum (Fig. 3) of the XP-C cells, in contrast to XP-A, XP-D and XP-F cells, were comparable with those of the normal cells. XP-C patients have markedly different clinical features from those in other complementation groups that were studied earlier. The general clinical features of XP-C are severe sun sensitivity of the skin and a high incidence of multiple skin cancers at an early age but no neurological abnormality (Kraemer et al., 1987; Takebe et al., 1987). The cultured fibroblast cells originating f r o m the XP-C patient, XP4PA, are hypersensitive to UV irradiation, and the extent of hypersensitivity is similar to that of XP-F patients (Kraemer et al., 1987; Takebe et al., 1987). XP-A patients generally have severe clinical symptoms including severe sun sensitivity of the skin, onset of multiple skin cancers at an early age and neurological abnormalities (Kraemer et al., 1987; Takebe et ai., 1987). in contrast, XP-F patients generally have mild clinical symptoms with roodcrate sun sensitivity, late onset of skin cancer and no neurological abnormality (Kraemer et al., 1987; Takebe et al., 1987; Yamamura et al., 1989). With these complementation groups of XP, we could not see an apparent correlation between the abnormalities detected with UV-irradiated pZ189 passed through these cells and the disease severity, age of onset or number of skin cancers, or presence of neurological abnormalities in the XP patients. The high incidence of multiple skin cancers at an early age in XP-C patients cannot be entirely explained by our results.
Acknowledgements A part of the nucleotide sequence analysis was done by Ms Chisako Hashimoto at the Biomedi-
220 cal E x p e r i m e n t T r a i n i n g C o u r s e for M e d i c a l Stu-
dents, Faculty of Medicine, Kyoto University. X P 4 P A ( S V ) a n d W I - 3 8 V A 1 3 cells w e r e kindly
supplied by Alan Sarasin and by the American Type Culture Collection, respectively. This research was supported by Grants-in-aid from the Ministry o f E d u c a t i o n , S c i e n c e a n d C u l t u r e o f
Japan. References Armitage, P. (1971)Statistical Methods in Medical Research. Wiley. New York. pp. 135-138, Brash. D.E.. and W.A. Haseltine (1982) UV-induced mutation hotspots occur at DNA damage hotspots, Nature (London). 298. 189-192. Brash, D.E., S. Seetharam. K,H. Kraemer, M.M. Seidman and A. Bredberg 11987) Photoproduct frequency is not the major determinant of UV base substitution hot spots or cold spots in human cells. Prec. Natl. Acad. Sci. (U.S.A.), 84, 3782-3786. Bredhcrg. A.. K.H. Kraemer and N.M. Seidman 11986) Restricted ultraviolet mutational spectrum in a shuttle vector propagated in xeroderma pigmentosum cells, Prec. Natl. Acad. Sci. (U.S.A.). 83, 8273-8277. Conlondre. C., ,'rod J.H. Miller (1977)Genetic studies of the hw repress,r. IV. Mutagenic specificity in the loci gene of Escherichia coll. J. Mol. Biol.. 117. 577-606. Daya-Grosjean, L., M.R. James, C. Drougard and A. Sarasin 11987) An immortalized xeroderma pigmentosum, group C. cell line which replicates SV40 shuttle vectors, Mutalion Rcs., 183. 185=196. Girardi, AJ., F.C, Jensen and H. Koprowski (191~5) SV411.indated transformation of human diploid cells: crisis and recovery, J. C¢11. Ctnnp. Ph~'si(~l.. 65.69~84, Halley. D.J.J,, W. Keijzer, N.G,J. Jaspers, M.F, Niermeijer, W.J, Kleijer, A. Boue, A, Bouc and D, Bootsma (1979) Prenatal diagnosis of xeroderma pigmentosum (group C) using assays of unscheduled DNA synthesis and postreplication repair, Clin. Genet,, 16, 137-146. Hauser, J,, M.M. Seidman, K, Sidur and K. Dixon (1986) Sequence specificity of point mutations induced during passage of a UV-irradiated shuttle vector plnsmid in menkey cells, Mol. C¢11, Biol,, 6, 277~285. Hirt, B. (19671 Selective extraction of polyoma DNA from infected mouse cell cultures, J. Mol. Biol., 20, 365-369. Keyse. S.M., F. Amaudruz and R.M, Tyrell (1988) Determinalion of the spectrum of mutations induced by defined wavelength solar UVB (313 nm) radiation in mammalian cells by use of a shuttle vector, Mol, Cell. Biol,, 8, 54255431. Kraemer, K.H., M.M. Lee and L Scotto (1987) Xeroderma pigmentosum: cutaneous, ocular, and neurologic abnerrealities in 830 published cases, Arch. Dermatol., 1 2 3 , 241-250. Kraemer, K.H., S. Seetharam, M, Proti~ Sablji~, D,E. Brash, A. Bredberg and M,M. Seidman (1988) Defective DNA repair and mutagenesis by dimer and non-dimer photo-
products in xeroderma pigmentosum measured with plasmid vectors, in: E. Friedberg and P. Hanawalt (Eds.), Mechanisms and Consequences of DNA Damage Processing, UCLA Symposia on Molecular and Cellular Biology, New Series, Vol. 83, Liss, New York, pp. 325-335. Maniatis, T., E.F. Fritsch and J. Sambrook 11982) Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Mitchell, D.L., and R.S. Nairns 11989) The biology of the (6-4) photoproduct, Photochem. Photobiol., 49, 805-819. Seetharam, S., M. Protic-Sabljic, M. Seidman and K.H. Kraemer 11987) Abnormal ultraviolet mutagenic spectrum in plasmid DNA replicated in cultured fibroblasts from a patient with skin cancer-prone disease, xeroderma pigmentosum. J. Clin. Invest., 80, 1613-1617. Seetharam, S.. K.H. Kraemer, H.L. Waters and M.M. Seidman (1990) Mutational hotspur variability in an ultraviolet treated shuttle vector plasmid propagated in xeroderma pigmentosum and normal human lymphoblasts and fibreblasts, J. Mol. Biol., 212, 433-436. Seetharam, S, K.Fi. Kraemer, H.L. Waters and M.M. Seidman (1991) Ultraviolet mutational spectrum in a shuttle vector propagated in xerodcrma pigmentosum lymphoblastold cells and fibroblasts, Mutation Res., 254, 97-105. Seidman, M.M.. K. Dixon, A. Razzaque and M,L. Berman 11985) A shuttle vector plasmid for studying carcinogen-induced point mutations in mammalian cells, Gene, 38, 233-237. Seidman, M.M., A, Bredberg. S. Seetharam and K.H. Kra¢mer (1987) Multiple point mutations in a shuttle vector propagated in hum:m cells: evidence for an error-prone DNA palymerase activity, Proc, Null. Acad. Sci. (U.S.A.), 84, 4944=4948. Takebe, H., C. Nishigori and Y, Sarah (1987) Genetics and skin cancer of seroderma pigmenlosum in Japan, Jpn. J. Cancer Res, (Gann), 78, 1135-1143. Tatsuka, M., S. Orita, T. Yogi and T. Kakunaga (1988) An improved method of electmporation for introducing hiplogically active foreign genes into cultured mammalian cells, Exp, Cell Res., 178, 154-162. Venema, J., A. van Hoffen, A,T. Natarajan, A.A. van Zeeland and L.H.F. Mullenders (I990)The residual repair capacity of xerodcrma pigmentosum complemenlation group C fibroblasts is highly specific for transcriptionally active DNA, Nucleic Acids Res., 18, 443-448. Wood, R.D., T.R. Scopek and F. Hutchinson (1984)Changes in DNA base sequence induced by targeted mutagenesis of Lambda phage by ultraviolet light, J. Mol. Biol., 173, 273-291. Yogi, T,, J, Talsumi-Miyajlma, M, Sate, K,H, Kraemer and H. Takebe (1991) Analysis of Ix)int mutations in a UV-irradiated shuttle vector plasmid propagated in cells from Japanese xeroderma pigmentosum patients in complemenration groups A and F, Cancer Res,, 51, 3177-3182. Yamamura, K,, M. ichihashi, T, Hiramoto, M. Ogoshi, K. Nishioka and Y, Fujiwara (1989)Clinical and photobiological characteristics of xeroderma pigmentosum complementation group F: a review of cases from Japan, Br. J. Dermatol,, 121,471-480.
213
© 1992 Elsevier Science Publishers B.V. All rights reserved 0921-8777/92/$05.00
MUTDNA 00192
UV-induced base substitution mutations in a shuttle vector plasmid propagated in group C xeroderma pigmentosum cells Takashi Yagi, Mayumi Sato, Junko Tatsumi-Miyajima and Hiraku Takebe Department of Experimental Radiology, Faculty of Medicine, Kyoto University, Yoshida-konoe.cho, Sakyo-ku, Kyoto 606 (Japan) (Accepted 9 May 1991)
Keywords: Point mutation; Ultraviolet; Xeroderma pigmentosum group C
Summary To assess the contribution to mutagenesis of human DNA repair defects, the UV-irradiated shuttle vector plasmid pZ189 was propagated in fibroblasts derived from a xeroderma pigmentosum (XP) patient in DNA repair complementation group C. In comparison to results with DNA repair-proficient human cells (WI-38 VAI3), UV-irradiated pZ189 propagated in the XP-C (XP4PA(SV)) cells showed fewer surviving plasmids and a higher frequency of mutated plasmids. Base sequence analysis of 67 mutated plasmids recovered from the XP-C cells revealed similar classes of point mutations and mutation spectrum, and a higher frequency of G : C to A : T transitions along with a lower frequency of transversions among plasmids with single or tandem mutations compared to plasmids recovered from the normal line. Most single-base substitution mutations (83%) occurred at G : C base pairs in which the 5'.adjacent base of the cytosine was thymine or cytosine, These results indicate that the DNA repair defects in XP-C, in comparison to data previously reported for XP-A, XP-D and XP-F, result in different UV survival and mutation frequency but in similar types of base substitution mutations.
This study was focused on investigating the contribution to UV-induced mutagenesis of the DNA repair defect in a xeroderma pigmentosum (XP) patient in complementation group C. XP is an autosomal recessive disease that is characterized by a high frequency of neoplasms in sun-ex-
Correspondence: Dr. T. Yagi, Department of Experimental
Radiology, Faculty of Medicine,Kyoto University,Yoshidakonoe-cho, Sakyo-ku, Kyoto 606 (Japan).
XP, xeroderma pigmentosum;XP-A, XP.C, XP-F, xeroderma pigmento~um complement~:[3n groups A,
Abbre~'iations:
C, F; UV, ultraviolet; UDS, unscheduled DNA synthesis,
posed skin (reviewed in Kraemer et al., 1987; Takebe et al., 1987). Cultured cells derived from XP patients are hypersensitive to UV radiation, have a reduced capacity for DNA repair and produce a high frequency of UV-induced mutations. There are at least 7 DNA excision-repair complementation groups (A-G) in XP cells (Kraemer et al., 1987; Takebe et al., 1987). XP DNA repair complementation group C is the most common worldwide. A shuttle vector plasmid, pZ189 (Seidman et al., 1985), was utilized to measure UV-induced point mutations following replication in skin fibroblast cells derived from an Algerian XP pa-
214
tient in complementation group C compared with normal human cells, pZ189 carries a 190-bp bacterial suppressor tRNA-g-cn¢, ~,-s'ap"~;~-s-a~,-~rker gene for mutations. It also contains the origin of replication and T antigen gene from SV40 to permit replication in human cells, and the origin of replication and the fl-lactamase gene from pBR327 for growth in E. coil and selection for ampicillin resistance. The plasmid is irradiated with UV and then transfected into the human cells where DNA repair, mutation and replication occur in a process dependent on host cell enzymes. Replicated plasmid is harvested and used to transform an indicator strain of bacteria. The number of ampicillin-resistant bacterial colonies reflects plasmid survival. The supF gene suppresses an amber mutation in the//-galactosidase gene (lacZ) in the indicator strain of E. coll. Bacteria with ti~e wild-type supF gene make dark blue colonies while bacteria containing the mutant supF gene make white or light blue colonies on agar plates containing 5-bromo-4chloro-3-indoyl/3-D-galactoside (X-gal). Thus, the mutation frequency can be determined by counting the proportion of white or light blue colonies compared to the total number of colonies. Mutared plasmids are purified from the bacterial colonies and the nucleotide sequence of the supF gen¢ is determined, The frequency and type of UV-induced base substitution mutations occurring in pZI89 propagated in cells from XP-A (Brcdberg et al., 1986; Yagi et aL, 1991), XP-D (Seethram ¢t al., 1 9 8 7 ) and XP.F (Yagi et al., 1991) patients was previously reported to be different from those in normal human cells. Hence, we examined whether the frequency and type of those mutations occurring in XP-C cells are different from those in XP cells in other complementation groups and from those in normal cells, Materials and methods
Cells Established fibroblast cell lines were used in this study. XP4PA(SV) is a cell line immortalized by transfection of a plasmid containing replication origin defective promoter and early region of SV40 virus (Daya-Grosjean et al., 1987). The
original XP4PA patient is an Algerian boy who was prenatally diagnosed as XP (complementation group C) (Halley et al., 1979), and XP4PA fibroblast cells were established after his birth. The XP4PA fibroblasts had 12% of the normal UV-induced unscheduled DNA synthesis (UDS) ability (a measure of DNA excision repair) (HalIcy et al., 1979) and were moderately sensitive to killing by UV (D37 = 1.2 j/m2). To compare the XP-C cells with normal human cells, an SV40-transformed normal fibroblast cell line, WI-38 VA13 (Girardi et al., 1965), was used. The parental cell line WI-38 was established by L. Hayflick from normal embryonic lung tissue of a Caucasian female, and showed normal sensitivity to killing by UV (D37 = 6.6 j/m2). The UV-induced mutation data of pZ189 passed through WI-38 VA13 cells have been published by us (Yagi et al., 1991) and are used in this study for comparison. All cells were cultured in Dulbecco's modified minimum esential medium (Nissui, Tokyo) supplemented with 10% fetal bovine serum (Hyclone, Logan, UT).
Methods pZ189 was extracted from overnight cultures of E, coli DHI having pZ189 by alkoline lysis and purified by cesium chloride ultracentrifugation (Maniatis et al., 1982). The closed circular pg189 band was carefully removed from the centrifuge tube with a needle and the plasmid solution was dialyzed against TE buffer and used for experiments. pZI89 dissolved in TE buffer at a eoncentration of 45/zg/ml was irradiated with a bank of two 10.W Toshiba germicidal lamps (predominately 254-nm UV light) at a dose rate of 3.4 J/m2/s. The human cells were trypsinized, washed and suspended in Duibecco's phosphatebuffered saline (pH 7.4-7.6). 2 X 107 cells plus 9 /zg UV-treated pZI89 in 0,2 ml buffer were placed in an electroporation chamber (electrodes 0.3 cm apart) (ZA-1200, PDS lnc., Madison, WI) and transfected with plasmid by electric pulses (500 V, 4 times) (Tatsuka et al., 1988). The cells were plated in 10-cm dishes and incubated in a CO 2 incubator at 37°C, After 72 h incubation, the plasmid was extracted from the cells by the
215
method of Hirt (1967), and digested by restriction endonuclease Dpnl (Boehringer-Yamanouchi, Tokyo) to eliminate,.~on~replicated input plasmid with the bacterial methyrlation pattern (Seidman et ai., 1985). E. coli MBM7070 which has an amber mutation in its lacZ gene was transformed by the plasmid and spread onto LB agar plates which contained ampicillin, the colorless dye, 5-bromo4-chloro-3-indoyl/3-D-galactoside (X-gaD and isopropyl fl-o-thiogalactoside (IPTG) (Nacalai Tesque, Kyoto, Japan). After 24 h incubation a t 37°C, blue colonies with normal supF and white or light blue colonies with mutated supF appeared on the plate. The total number of colonies w a s counted as a n indication of plasmid survival and the proportion of white or light blue colonies determined as a measure of mutation frequency, Mutated plasmids were purified from overnight cultures and the nucl¢otide sequence of the supF gene of the plasmid was determined with K / R T universal dideoxy sequencing kit (Promega, Madison, WI). To obtain independent mutant clones, fewer than 5 mutants were isolated from each independent transfection and only 1 mutant was scored if several plasmids had identical mutations.
Statistics Statistical comparisons were performed with Fisher's exact test for difference in proportions (Armitage, 1971). P values for a 1-tailed test are presented,
Results
Plasmid surt,it,al and mutagenesis The shuttle vector plasmid pZ189 was irradiated with UV, transfected into the XP-C and normal cells, incubated for 3 days to permit repair, mutation and growth, extracted and used to transform bacteria to ampicillin resistance. UV irradiation of pZ189 showed a dose-dependent reduction in the number of ampicillin-resistant bacterial colonies (Fig. 1). Survival of the plasmid recovered from the XP-C line was lower than that from the normal cell line. At a dose of 1 0 0 0 J / m 2 to the plasmid the relative number of bac-
1
~,~..,.., ~'x... ~
,
,
\"
~ ~ ~ m.1 ~ ~c, a-
,
,
",
".\ \\ \
• \
\ "\ \\\ "
10-2
I o
I
I
I
I
soo looo lsoo 20o0 UV Dose (j/m2)
Fig. 1. Survival of UV-irradiated pZ189 replicated in XP-C (XP4PA(SV)) (e) and normal (WI-38 VAI3) (broken line) cell lines. The relative number of ampiciilin-resistant bacterial colonies obtained after repair and replication of UV-irradiated pZ189 in the cells followed by transformation of the indicator E. coil is shown. The data show mean values from 2-4 independent transfection experiments for each cell line. The standard error is smaller than the symbols for most points. The data of WI-38 VAI3 are from Yagi et al. (1991).
terial colonies recovered from the XP-C line was about 3.4-fold less than from the normal line. UV irradiation to pZ189 increased the proportion of white and light blue colonies with a mutated supF gene in a dose-dependent manner (Fig. 2). The background plasmid mutation frequeneies were 2.1 × 10 -'~ and 0.9 x 10 -'~ with the XP-C and normal lines, respectively, and increased 10-190-fold following treatment of the plasmid with 500-2500 J / r e ' UV. The UV-induced l~mtation frequency was greater with the XP-C th~n with the normal line. At 1000 J / m : UV to the plasmid the mutation frequencies with the XP-C and the normal lines were 65 × 10- "~ and 17 × 10 -3, respectively (Fig. 2). These results reflect the UV hypermutability of XP cells.
Base sequence analysis Analysis of 97 supF mutant plasmids with the XP-C cell line by agarose gel electrophoresis and nucleotide sequence analysis revealed that 31% of the plasmids contained large deletions (average size 84 base pairs). The mutant plasmids without deletions contained point mutations. Table 1 shows the classes of point mutations
216 1
,
,
I
,
,
, $_
e~ ~
~ 10-1 -
//'/e/"/" t~ m-2 =1
_ ........
~ //
tE l o _ a _ t,,
I 104
0
_
I I I I I 500 1000 1500 2000 2fi00 UV Dose (J/m 2)
Fig. 2. Mutation of UV-irradiateo pZ189 replicated in XP-C (XP4PA(SV)) (e) and normal (WI-38 VAI3) (broken line) cell lines. The frequency of white and light blue bacterial colonies obtained after repair and replication of UV-irradiated p Z 1 8 9 in the cells followed by transformation of the indicator E. c o l i is shown. The data show mean values from 2-4 independent transfection experiments for each cell line. The standard error is smaller than the symbols for most points. The data of W1-38 VAI3 are from Yagi et al. (1991).
identified following base sequence analysis of 67 independent mutant plasmids of wild-type size recovered from the XP-C cell line. Plasmids are classified as to whether they have a single-base substitution, tandem base substitutions (2 base substitutions 0-2 bases apart, or 3 adjacent base
substitutions), multiple base substitutions (2 base substitutions m o r e than 3 bases a p a r t or 3 or m o r e base substitutions), and single-base insertion or deletion. Single-base substitutions were slightly m o r e f r e q u e n t with X P - C (72%, P = 0.058) t h a n with the normal cells (58%). T h e p r o p o r t i o n o f mutated plasmids with multiple base substitutions was slightly lower with XP-C (12%, P = 0.076) than with the n o r m a l cells (22%) (Table 1). T h e maximum n u m b e r o f base substitutions f o u n d in the supF gene in one plasmid was 3 with the XP-C cells and 4 with the normal line. Single-base frameshift mutations were not seen with the XP-C cell line. A m o n g the m u t a n t plasmids with single and t a n d e m base substitution mutations, transition mutations were p r e d o m i n a n t with both X P - C and
normal cell lines (Table 2). The transitions were slightly more frequent with the XP-C (75%, P = 0.057) than with the normal cells (62%). It is quite striking that the most frequent type of base substitution mutation was the G : C to A : T transition, accounting for more than half of the single or tandem base substitution mutations with both cell lines, The frequency of G : C to A : T transitions was significantly higher with the XP-C (71%, P = 0.040) cells than with the normal cells (56%). The frequency of G: C to C:G transversions was significantly lower with the XP-C (3%, P - 0.050)
TABLE I CLASSES OF POINT MUTATIONS INDUCED IN UV-IRRADIATED pZ189 a Number of plasmids with base changes (%)
Independentplasmidssequenced Point mutations Single-base substitution Tandem-base substitution Multiple-base substitution Frameshifl Single-base insertion
Single.basedeletion " 250-2500 J/m "~UV irradiation to pZ189. h The data are from Yagi el el, (1991), c These mutants also have a single-base substitution. * 0.05 < P < 0.1 vs. normal.
XP-C 67 (100%)
Normal h 91 (100%)
48 (72%)* II (17%) 8 (12%)*
53 (58%) 15 (16%) 20 (22%)
(I (0%) (1 ((1%)
i
(1%)
2 c (2%)
217
cells than with the normal cells (II%). While 89% of the single-base substitution mutations involved G : C base pairs with the XP-C cells, only 11% involved A : T base pairs.
TABLE 2 TYPES O F SINGLE O R T A N D E M BASE SUBSTITUTION M U T A T I O N S I N D U C E D IN U V - I R R A D I A T E D
pzI89 ~ Number of base changes (%) XP-C
Normal b
Transitions G : C to A : T A:TtoG:C
54 ( 7 5 % ) * 51 (71%) * * 3 (4%)
52 (62%) 47 (56%)
Transversions G:CtoT:A G : C to C : G A : T to T : A
18 ( 2 5 % ) * I! (15%)
32 (38%) 18 (21%)
A:T to C:G
I (1%)
5 (6%)
2
(3%) * *
9 (11%)
4
(6%)
3
(4%)
2 (2%)
72 (100%)
Total
Mutation spectrum The distribution of single and tandem base substitutions in the supF gene with the XP-C a n d normal cells is shown in Fig. 3. All mutations were found between base pairs 99 and 177. The mutational patterns show some similarities and s o m e differences between the cell lines. The point mutations were not distributed randomly in the supF gene but appeared preferentially at certain site s. There was a G : C to A: T transition hotspot at base pair 156 with both lines accounting for 17% and 15% of all single and tandem mutations with
84 (100%)
~' 250-1000 ,l/re" U V irradiation to pZ189. The data are from Yagi et al. (1991). • 0.05 < P < 0.1 vs. normal: * * P < 0.05 vs. normal,
4"
4'
4.
4.
4"
4.
4"
4"
4"
4"
1"/0
180
ATTA•CTGTGGTGGGGTT•••GAG•GG••AAAGG•AG•AGA•TCTAAATCTGC•GTCAT•GA•TT•GAAGGTT•GAAT••TT••C••A••AC•A XP-C 90
100
110
180
130
140
150
160
4" . . . . . . . . . .4. . . . . . . . . . .4-. . . . . . . . . . 4". . . . . . . . . . 4". . . . . . . . . . 4,. . . . . . . . . . +. . . . . . . . . . 4,. . . . . . . . . . 4". . . .
A.A AA A~A
90
100
AT
110
G AA.~ TA TA TA T_A
120
4" . . . . . . . . .4". . . . . . . . . 4". . . . . . . . . 4..
A~_ AT A A AA TT C _C.._T T_T A
e A
GT GT
130
A A
140
C G
AAA A A A A A A A A A A A T
150
AC A A G G G G
G
G
TT TT TT T T T T T T T T
160
- . . . . . . . 4.. . . . . . . . . 4. . . . . . . . . .4- . . . . . .
T AA AA A(~
A A A A A A T
G_~ TA~_.ATC T,A~ A TA A AA T AA T CA A
A A A A A A T
170
- . 4" . . . . . . . .
T
T C
4.
TT T T AT
- 4. . . . . . . . .
IBO - 4- . m - -
T_T~C T ~ A TT TT TT T T
A A
Fig. 3. Location of single and tandem base substitution mutations found in the s'upF gene of UV-irradiated pZ189 replicated in XP-C (XP4PA(SV)) and normal (WI-38 VAI3) cells. The supF suppressor tRNA sequences start at base pair 99 and end at base pair 183. Each lelter represents a single-base-pair substitution mutation found in an independent plasmid. Tandem base substitution mutations are underlined. The data of WI-38 VAI3 are from Yagi et al. (1991).
218
the XP-C and normal lines, respectively. At base pair 159 the (3: C to A: T transition and G: C to T : A transversion were frequent with the XP-C (10%) and less frequent with the normal cell line (7%). At base pair 168 the G : C to A : T transition in single point mutations was slightly more frequent with the XP-C line than with the normal line (23% vs. 9 % , / ' = 0.057). Fig. 3 shows that most single-base substitution mutations occurred at G : C base pairs where the 5' neighboring base of the mutated cytosine is a cytosine or thymine (XP-C 83%, normal 91%). In contrast, about a half of the single-base substitutions occurred at G : C base pairs where the 3' base adjacent to the mutated cytosine was a pyrimidine (XP-C 52%, normal 51%). All tandem mutations occurred at 5'-CC sequences and only 8% of the single base substitution mutations were found at 5'-TT sequences with XP-C cells, Discussion Cells derived from XP patients show hypersensitivity to killing by UV and UV hypermutability (Kraemer et al., 1987; Takebe et al., 1987). These cellular abnormalities are reflected in the increased plasmid killing and mutations observed in UV-irradiated pZ189 passed through the XP-C cells (Figs. I and 2). These results are in agreement with similar studies performed with UV. irradiated pZ189 passed through cells from XP-A (Bredberg et al., 1986; Yagi et al., 1 9 9 1 ; Seetharam et al., 1990, 1991), XP.D (Seetharam et al., 1987) and XP-F (Yagi et al., 1991) patients, The survival and mutation frequency of the UVirradiated pZ189 passed through XP-C cells are between those through XP-A and normal (Bredberg et al., 1986; Yagi et al., 1991). Thus the present research provides further support for the use of pZ189 to measure DNA repair defects in human cells, Analysis of the classes of UV-associated mutations recovered in pZI89 revealed a slight increase in the proportion of plasmids with singlebase substitution mutations and a slight decrease in plasmids with multiple.base substitutions with the XP-C line compared to the normal line (0,05 < P < 0.1 vs. normal) (Table 1), The residual level of excision repair ability (UDS 12% of nor-
mal) in the XP-C cells may n o t have caused the significantly large difference between the XP-C and normal cells. A significantly abnormal proportion of these base substitutions was clearly observed with XP-A (Bredberg et al., 1986; Yagi et al., 1991) and XP-D fibroblast lines (Seetharam et al., 1987) in comparison to the normal lines. These studies strongly suggest that defective DNA repair is linked to a low frequency of UV-induced multiple-base substitutions and to a high frequency of single-base substitutions. The multiplebase substitutions have been shown to be dramatically increased by creating a single-strand nick in the plasmid (Seidman et al., 1987). Normal cells, unlike XP cells, can make a nick at the UV-induced DNA damage, which leads to induction of the multiple-base substitutions. DNA repair deficiency in the XP cells is also associated with an increased frequency of G: C to A : T transitions (with consequent decreased frequency of transversions) in UV-treated pZ189. In the present study, among the single and tandem base substitution mutations, the percentage of the G : C to A : T transition was significantly higher in the XP-C than in the normal cells (P = 0.04) (Table 2). This was also observed with XP-A (Bredberg et al., 1986; Yagi et al., 1991), XP.F (Yagi et al., 1991) and XP-D (Seetharam et al., 1987) fibroblast lines. Excision repair-proficient monkey cells also had a significantly lower frequency of the G : C to A : T transition with UV-treated pZI89 than XP cells (Keyse et al., 1988; Hauser et al., 1986). Most base substitution mutations (89%) were found at G: C base pairs (Table 2) and 83% of 5' neighboring bases of the mutated cytosines in the single-base substitutions are thymine or cytosine with the XP-C line (Fig. 3). These data are consistent with earlier studies with the XP-A, XP-D and XP-F cells (Bredberg et al., 1986; Seetharam et al., 1987; Yagi et al., 1991). These results suggest that the Y cytosine of 5°-TC and 5'-CC sequences is the major mutational target of UV radiation. All tandem mutations occurred at 5'CC sequences with the XP-C cells (Fig. 3). These dipyrimidine sites can be involved in either cyclobutane dimer or pyrimidine (6-4) pyrimidone photoproducts (Brash et al., 1987; Mitchell and Nairns, 1989) which have been shown to be muta-
219
genic in bacterial cells (Brash and Haseltine, 1982; Conlondre and Miller, 1977; Wood et al., 1984). No mutation occurred at 5'-Tr sequences that form the major UV photoproduct, cyclobutane thymine dimer (about 60% of the total photoproducts in pZ189) (Brash et al., 1987; Kraemer et al., 1988), As in earlier studies with I V-irradiated pZ189, the major UV photoproduct is not the major mutagenic lesion. In pZ189, (6-4) photoproducts have been measured to occur at a frequency 70% of that of cyclobutane dimers at 5'-TC sites and 30% at 5'-CC sites (Brash et al., 1987; Kraemer et al., 1988). Studies of photoreactivation of UV-treated pZ189 indicate that both cyclobutane dimers and (6-4) photoproducts are mutagenic lesions (Brash et al., 1987; Kraemer et al., 1988). This suggests that 3' cytosine in unrepaired 5'-TC and 5'-CC sites causes G : C to A : T transition during DNA replication, and a high frequency of this mutation in XP cells is ascribed to the faulty repairing of these lesions, The mutation spectrum with UV-treated pZ189 passed through XP-C cells showed a tandem mutation hotspot at base pairs 122-123 and singlebase substitution hotspots at base pairs 156, 159 and 168 (Fig. 3)..The proportion of the number of mutants included in each hotspot with XP-C is not significantly different from that with normal cells. The normal cells have a transversion hotspot at base pair 133 but the XP-C cells have no mutation at the base pair ( P - 0.02). Base pair 168 showed a high frequency of G : C to A : T transition mutations with all XP cell lines tested including XP-A (Bredberg et al., 1986; Yagi et al., 1991), XP-D (Seetharam et al., 1987) and XP-F (Yagi et al., 1991) fibroblasts, XP-A lyrephoblasts (Seetharam et al., 1990, 1991) and the repair-proficient lymphoblasts (Seetharam et al., 1990, 1991) but not with the repair-proficient WI-38 VA13 (Yagi et al., 1991; Fig. 3) or GM0637 (SV) cells (Bredberg et al., 1986). This is in keeping with the observation of mutational hotspot variability that has recently been discussed (Seetharam et al., 1990) which suggested that the major determinants of mutational hotspots were not genetic polymorphisms or repair capacity of the cells but cellular factors that can influence the probability of mutagenesis at particular sites. The XP-C cells have very low but significantly
measurable amounts of DNA excision repair capacity (12% of the normal level of UDS) (Halley et al., 1979). This residual DNA repair specificaily works on transcribed genes but not on nontranscribed genes in the genome of UV-irradiated XP-C cells (Venema et al., 1990). The significantly higher frequenc~ of UV-induced G : C to A : T transition mutations in XP-C cells than in normal cells indicates that the residual DNA repair does not work efficiently on UV-damaged pZ189, despite the classes of point mutations (Table 1) and the mutation spectrum (Fig. 3) of the XP-C cells, in contrast to XP-A, XP-D and XP-F cells, were comparable with those of the normal cells. XP-C patients have markedly different clinical features from those in other complementation groups that were studied earlier. The general clinical features of XP-C are severe sun sensitivity of the skin and a high incidence of multiple skin cancers at an early age but no neurological abnormality (Kraemer et al., 1987; Takebe et al., 1987). The cultured fibroblast cells originating f r o m the XP-C patient, XP4PA, are hypersensitive to UV irradiation, and the extent of hypersensitivity is similar to that of XP-F patients (Kraemer et al., 1987; Takebe et al., 1987). XP-A patients generally have severe clinical symptoms including severe sun sensitivity of the skin, onset of multiple skin cancers at an early age and neurological abnormalities (Kraemer et al., 1987; Takebe et ai., 1987). in contrast, XP-F patients generally have mild clinical symptoms with roodcrate sun sensitivity, late onset of skin cancer and no neurological abnormality (Kraemer et al., 1987; Takebe et al., 1987; Yamamura et al., 1989). With these complementation groups of XP, we could not see an apparent correlation between the abnormalities detected with UV-irradiated pZ189 passed through these cells and the disease severity, age of onset or number of skin cancers, or presence of neurological abnormalities in the XP patients. The high incidence of multiple skin cancers at an early age in XP-C patients cannot be entirely explained by our results.
Acknowledgements A part of the nucleotide sequence analysis was done by Ms Chisako Hashimoto at the Biomedi-
220 cal E x p e r i m e n t T r a i n i n g C o u r s e for M e d i c a l Stu-
dents, Faculty of Medicine, Kyoto University. X P 4 P A ( S V ) a n d W I - 3 8 V A 1 3 cells w e r e kindly
supplied by Alan Sarasin and by the American Type Culture Collection, respectively. This research was supported by Grants-in-aid from the Ministry o f E d u c a t i o n , S c i e n c e a n d C u l t u r e o f
Japan. References Armitage, P. (1971)Statistical Methods in Medical Research. Wiley. New York. pp. 135-138, Brash. D.E.. and W.A. Haseltine (1982) UV-induced mutation hotspots occur at DNA damage hotspots, Nature (London). 298. 189-192. Brash, D.E., S. Seetharam. K,H. Kraemer, M.M. Seidman and A. Bredberg 11987) Photoproduct frequency is not the major determinant of UV base substitution hot spots or cold spots in human cells. Prec. Natl. Acad. Sci. (U.S.A.), 84, 3782-3786. Bredhcrg. A.. K.H. Kraemer and N.M. Seidman 11986) Restricted ultraviolet mutational spectrum in a shuttle vector propagated in xeroderma pigmentosum cells, Prec. Natl. Acad. Sci. (U.S.A.). 83, 8273-8277. Conlondre. C., ,'rod J.H. Miller (1977)Genetic studies of the hw repress,r. IV. Mutagenic specificity in the loci gene of Escherichia coll. J. Mol. Biol.. 117. 577-606. Daya-Grosjean, L., M.R. James, C. Drougard and A. Sarasin 11987) An immortalized xeroderma pigmentosum, group C. cell line which replicates SV40 shuttle vectors, Mutalion Rcs., 183. 185=196. Girardi, AJ., F.C, Jensen and H. Koprowski (191~5) SV411.indated transformation of human diploid cells: crisis and recovery, J. C¢11. Ctnnp. Ph~'si(~l.. 65.69~84, Halley. D.J.J,, W. Keijzer, N.G,J. Jaspers, M.F, Niermeijer, W.J, Kleijer, A. Boue, A, Bouc and D, Bootsma (1979) Prenatal diagnosis of xeroderma pigmentosum (group C) using assays of unscheduled DNA synthesis and postreplication repair, Clin. Genet,, 16, 137-146. Hauser, J,, M.M. Seidman, K, Sidur and K. Dixon (1986) Sequence specificity of point mutations induced during passage of a UV-irradiated shuttle vector plnsmid in menkey cells, Mol. C¢11, Biol,, 6, 277~285. Hirt, B. (19671 Selective extraction of polyoma DNA from infected mouse cell cultures, J. Mol. Biol., 20, 365-369. Keyse. S.M., F. Amaudruz and R.M, Tyrell (1988) Determinalion of the spectrum of mutations induced by defined wavelength solar UVB (313 nm) radiation in mammalian cells by use of a shuttle vector, Mol, Cell. Biol,, 8, 54255431. Kraemer, K.H., M.M. Lee and L Scotto (1987) Xeroderma pigmentosum: cutaneous, ocular, and neurologic abnerrealities in 830 published cases, Arch. Dermatol., 1 2 3 , 241-250. Kraemer, K.H., S. Seetharam, M, Proti~ Sablji~, D,E. Brash, A. Bredberg and M,M. Seidman (1988) Defective DNA repair and mutagenesis by dimer and non-dimer photo-
products in xeroderma pigmentosum measured with plasmid vectors, in: E. Friedberg and P. Hanawalt (Eds.), Mechanisms and Consequences of DNA Damage Processing, UCLA Symposia on Molecular and Cellular Biology, New Series, Vol. 83, Liss, New York, pp. 325-335. Maniatis, T., E.F. Fritsch and J. Sambrook 11982) Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Mitchell, D.L., and R.S. Nairns 11989) The biology of the (6-4) photoproduct, Photochem. Photobiol., 49, 805-819. Seetharam, S., M. Protic-Sabljic, M. Seidman and K.H. Kraemer 11987) Abnormal ultraviolet mutagenic spectrum in plasmid DNA replicated in cultured fibroblasts from a patient with skin cancer-prone disease, xeroderma pigmentosum. J. Clin. Invest., 80, 1613-1617. Seetharam, S.. K.H. Kraemer, H.L. Waters and M.M. Seidman (1990) Mutational hotspur variability in an ultraviolet treated shuttle vector plasmid propagated in xeroderma pigmentosum and normal human lymphoblasts and fibreblasts, J. Mol. Biol., 212, 433-436. Seetharam, S, K.Fi. Kraemer, H.L. Waters and M.M. Seidman (1991) Ultraviolet mutational spectrum in a shuttle vector propagated in xerodcrma pigmentosum lymphoblastold cells and fibroblasts, Mutation Res., 254, 97-105. Seidman, M.M.. K. Dixon, A. Razzaque and M,L. Berman 11985) A shuttle vector plasmid for studying carcinogen-induced point mutations in mammalian cells, Gene, 38, 233-237. Seidman, M.M., A, Bredberg. S. Seetharam and K.H. Kra¢mer (1987) Multiple point mutations in a shuttle vector propagated in hum:m cells: evidence for an error-prone DNA palymerase activity, Proc, Null. Acad. Sci. (U.S.A.), 84, 4944=4948. Takebe, H., C. Nishigori and Y, Sarah (1987) Genetics and skin cancer of seroderma pigmenlosum in Japan, Jpn. J. Cancer Res, (Gann), 78, 1135-1143. Tatsuka, M., S. Orita, T. Yogi and T. Kakunaga (1988) An improved method of electmporation for introducing hiplogically active foreign genes into cultured mammalian cells, Exp, Cell Res., 178, 154-162. Venema, J., A. van Hoffen, A,T. Natarajan, A.A. van Zeeland and L.H.F. Mullenders (I990)The residual repair capacity of xerodcrma pigmentosum complemenlation group C fibroblasts is highly specific for transcriptionally active DNA, Nucleic Acids Res., 18, 443-448. Wood, R.D., T.R. Scopek and F. Hutchinson (1984)Changes in DNA base sequence induced by targeted mutagenesis of Lambda phage by ultraviolet light, J. Mol. Biol., 173, 273-291. Yogi, T,, J, Talsumi-Miyajlma, M, Sate, K,H, Kraemer and H. Takebe (1991) Analysis of Ix)int mutations in a UV-irradiated shuttle vector plasmid propagated in cells from Japanese xeroderma pigmentosum patients in complemenration groups A and F, Cancer Res,, 51, 3177-3182. Yamamura, K,, M. ichihashi, T, Hiramoto, M. Ogoshi, K. Nishioka and Y, Fujiwara (1989)Clinical and photobiological characteristics of xeroderma pigmentosum complementation group F: a review of cases from Japan, Br. J. Dermatol,, 121,471-480.