Influence of cytosine methylation on ultraviolet-induced cyclobutane pyrimidine dimer formation in genomic DNA

Influence of cytosine methylation on ultraviolet-induced cyclobutane pyrimidine dimer formation in genomic DNA

Mutation Research 665 (2009) 7–13 Contents lists available at ScienceDirect Mutation Research/Fundamental and Molecular Mechanisms of Mutagenesis jo...

528KB Sizes 0 Downloads 10 Views

Mutation Research 665 (2009) 7–13

Contents lists available at ScienceDirect

Mutation Research/Fundamental and Molecular Mechanisms of Mutagenesis journal homepage: www.elsevier.com/locate/molmut Community address: www.elsevier.com/locate/mutres

Influence of cytosine methylation on ultraviolet-induced cyclobutane pyrimidine dimer formation in genomic DNA Patrick J. Rochette a,b,c , Sandrine Lacoste a,b,d , Jean-Philippe Therrien a , Nathalie Bastien a,b , Douglas E. Brash c , Régen Drouin a,b,∗ a

Division of Pathology, Department of Medical Biology, Université Laval, Québec, QC, Canada Division of Genetics, Department of Pediatrics, Faculty of Medicine and Health Sciences, Université de Sherbrooke, Sherbrooke, QC, Canada c Department of Therapeutic Radiology, Yale School of Medicine, New Haven, CT, USA d Manitoba Institute of Cell Biology, University of Manitoba, Winnipeg, MB, Canada b

a r t i c l e

i n f o

Article history: Received 20 October 2008 Received in revised form 25 January 2009 Accepted 17 February 2009 Available online 28 February 2009 Keywords: Cyclobutane pyrimidine dimers Cytosine methylation Ultraviolet light Photocarcinogenesis Skin cancer

a b s t r a c t The ultraviolet (UV) component of sunlight is the main cause of skin cancer. More than 50% of all non-melanoma skin cancers and >90% of squamous cell carcinomas in the US carry a sunlight-induced mutation in the p53 tumor suppressor gene. These mutations have a strong tendency to occur at methylated cytosines. Ligation-mediated PCR (LMPCR) was used to compare at nucleotide resolution DNA photoproduct formation at dipyrimidine sites either containing or lacking a methylated cytosine. For this purpose, we exploited the fact that the X chromosome is methylated in females only on the inactive X chromosome, and that the FMR1 (fragile-X mental retardation 1) gene is methylated only in fragile-X syndrome male patients. Purified genomic DNA was irradiated with UVC (254 nm), UVB (290–320 nm) or monochromatic UVB (302 and 313 nm) to determine the effect of different wavelengths on cyclobutane pyrimidine dimer (CPD) formation along the X-linked PGK1 (phosphoglycerate kinase 1) and FMR1 genes. We show that constitutive methylation of cytosine increases the frequency of UVB-induced CPD formation by 1.7-fold, confirming that methylation per se is influencing the probability of damage formation. This was true for both UVB sources used, either broadband or monochromatic, but not for UVC. Our data prove unequivocally that following UVB exposure methylated cytosines are significantly more susceptible to CPD formation compared with unmethylated cytosines. © 2009 Elsevier B.V. All rights reserved.

1. Introduction Exposure to the ultraviolet (UV) component of sunlight is the preeminent risk factor in skin cancer development [1,2]. Cyclobutane pyrimidine dimers (CPDs) are the primary promutagenic DNA adducts generated by all UV wavelengths, i.e., UVC (200–280 nm), UVB (280–320 nm) and UVA (320–400 nm) [3–5]. UV also induces (6-4) pyrimidine-pyrimidone photoproducts (6-4PPs) and oxidative DNA damage to a lesser extent. Although 6-4PPs were found to be mutagenic in E. coli [6], CPDs are considered the most important for skin carcinogenesis based on their relative abundance (approximately 85% of DNA adducts after UV), relatively slow removal by nucleotide excision repair, and known high mutagenic potential in human cells [7,8].

∗ Corresponding author at: Dept. of Pediatrics, Centre Hospitalier Universitaire de Sherbrooke (CHUS), Hospital Fleurimont, 3001, 12th Avenue North, Sherbrooke, Quebec, Canada J1H 5N4. Tel.: +1 819 820 6827; fax: +1 819 564 5217. E-mail address: [email protected] (R. Drouin). 0027-5107/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.mrfmmm.2009.02.008

The p53 tumor suppressor gene is mutated in more than 50% of all cancers and in approximately 90% of squamous cell carcinomas of the skin [9,10]. The vast majority of p53 mutations in non-melanoma skin cancers are C → T and CC → TT tandem double transition mutations, considered the “mutational signature” of UV exposure [11]. Such signature mutations preferentially occur at 5 YCG sequences (i.e. 5 TCG or 5 CCG). In fact, even though 5 YCG represents only 3% of all trinucleotides in p53, more than 30% of all tumor-derived mutations occur at these sites [12]. Of the 8 most common hotspots for p53 mutations in skin cancer, 6 occur at the cytosine of a 5 YCG sequence [13,14]. In vertebrates, the cytosine in a 5 CG sequence is a target for DNA methylation at position 5 of the pyrimidine ring (5m CG) [15]. In the coding region of exons 5–8 of the human p53 gene, 42 5 CG sites are fully methylated [12]. In the case of internal cancers, it is believed that a high frequency of spontaneous deamination of 5-methylcytosine (5m C) to thymine at 5 CG sites in p53 engenders the observed preponderance of C to T transitions at these sites [9,16]. A 5m CG preference for UV-induced mutations in the p53 gene is not observed after UVC exposure [17,18], but is clearly observed after UVB [17,18]. Moreover, the 5m CG sensitivity to UVB seems

8

P.J. Rochette et al. / Mutation Research 665 (2009) 7–13

1) and FMR1 (fragile-X mental retardation 1) are subject to strong transcriptional silencing by methylation-related X inactivation [40]. The PGK1 5 CG-rich island is unmethylated in the active X chromosome [41] and methylated at 119 of 121 5 CG dinucleotides in the inactivated X [42,43]. Female X-inactivation is not the only condition in which FMR1 is methylated at 5 CG. In fragile-X syndrome male patients, the FMR1 gene is mutated resulting in the hypermethylation of the gene, as in the inactivated X chromosome in female [44]. In the present study, we used ligation-mediated PCR (LMPCR) to analyze the efficiency of different UV wavelengths, i.e., UVC (254 nm) and UVB (290–320 nm, 303 and 313 nm) to generate CPD at 5 YCG sites that were methylated or not in the endogenous PGK1 and FMR1 genes. As such, we were able to unequivocally determine the effect of DNA methylation on CPD formation. Fig. 1. Emission spectrum of the UVB lamp (FS20/T12/UVB/BP) after filtering through a Kodacel TA-407 clear 0.015 in. filter. Measurements were made using an International Light double monochromator spectroradiometer (IL1700/760D/790).

to increase with the wavelength. In a mouse model, UV-specific tumor mutations in p53 occurred at 5m CG sites at a frequency of 33% following irradiation with UVB (290–320 nm), 43% with broad spectrum simulated sunlight (SSL, >300 nm), and 66% with UVA (320–400 nm) [19–21]. What might explain this wavelength-dependent specificity for C → T mutations at 5m CG sites? CPDs are efficiently bypassed in a relatively error-free manner by DNA polymerase ␩ (encoded by XPV gene). When polymerase ␩ bypasses a 5 TT, 5 TU or 5 UU dimer (T or U generated by deamination of 5-methylcytosine or cytosine, respectively), adenine is incorporated with high preference across from T or U [22–24], thereby generating a C → T mutation. The halflife for cytosine deamination in dsDNA is ∼25,000 years [25–27] and ∼2000 years when methylated [28]. However, when cytosine is covalently bound to another pyrimidine to form a CPD, the deamination half-life is reduced to 2–100 h [25,27,29–31]. When the same cytosine is methylated in a CPD, the deamination is further accelerated [32]. As a result, when implicated in a CPD, cytosine methylation increases the probability of generating a mutation. Moreover, (i) CPD repair in Chinese hamster ovary cells is slower when the DNA is methylated [33,34] and (ii) CPD formation is actually enhanced by the methylation of cytosines. Indeed, some p53 mutational hotspots found in non-melanoma skin cancers at 5 YCG sites are as much as 15-fold more susceptible to CPD formation after exposure to sunlight (>295 nm) [13] or UVB [17]. On synthetic residues, Mitchell observed a 2-fold increase in UVB-induction of CPD at a methylated cytosine when compared with the cytosine at the same position but unmethylated [35]. On the other hand cytosine methylation did not enhance CPD formation by UVC [36]. In conclusion, the methylated cytosines in dipyrimidine sites are the nucleotide residues with the highest mutation rate upon UVB and SSL irradiation. One hypothesis frequently proposed to explain the results is the different UV absorbance spectra for C and 5m C, which manifest a maximum at 267 and 273.5 nm, respectively [37,38]. In the studies cited above, the efficiency of different UV wavelengths for inducing CPD at a given 5 Ym CG site was compared. However, comparisons of the same 5 YCG site when methylated versus unmethylated using endogenous genomic sequences has not yet been performed. We therefore analyzed the influence of cytosine methylation on UV-induced CPD formation by comparing the same 5 YCG methylated or not in genomic DNA. To do so, we exploited the natural methylation-related inactivation of the X chromosome. In female mammalian cells, one of the two X chromosomes becomes genetically silent during early embryogenesis. The inactivation of one of the X chromosomes in females is a methylation-related gene-silencing mechanism that affects almost all genes of the X chromosome (reviewed in [39]). PGK1 (phosphoglycerate kinase

2. Materials and methods 2.1. Cell culture, DNA extraction and UV irradiation DNA was harvested from human diploid fibroblasts of a male, a female, or a male with the fragile-X syndrome (mutated on the FMR1 gene). For monochromatic irradiations, DNA was extracted from female human diploid lymphoblasts. DNA extraction from fibroblasts and lymphoblasts was performed as previously published [45]. Prior to irradiation, purified DNA was dissolved at a concentration of 60 ␮g/mL in irradiation buffer (150 mM KCl, 10 mM NaCl, 1 mM EDTA, and 10 mM Tris–HCl pH 7.4). The UVB (290–320 nm) source consisted of two fluorescent tubes (FS20/T12/UVB/BP, Philips, city) delivering a dose rate of 7.45 J/(m2 s), which was filtered through a sheet of cellulose acetate to eliminate wavelengths below 290 nm (Kodacel TA-407 clear 0.015 in.; Eastman-Kodak Co., Rochester, NY). The emission spectrum of the UVB lamp is depicted in Fig. 1. A Philips G15T8 TUV 15W germicidal lamp was used to irradiate cells with 254 nm UV at a fluency of 6.25 J/(m2 s). Doses for UVB and UVC irradiation (10,000 and 200 J/m2 , respectively) were chosen to give approximately 1 CPD per kb of genomic DNA [46]. For the monochromatic irradiations (GM 252, Spectral Energy Co. Washingtonville, NY), 302 and 313 nm were used at doses of 12,000 and 75,000 J/m2 , respectively. Doses were chosen to give similar CPD frequency as of polychromatic UVB and UVC. 2.2. Hpa II digestion of female DNA In females, the active X chromosome is undermethylated whereas the inactive one is hypermethylated. In males, there is only one X chromosome and it is undermethylated. In this study, the goal was to use the hypermethylated inactive X chromosome from the female to compare it with the undermethylated X chromosome from the male. To do so, we had to eliminate the active X chromosome in the female DNA. Hpa II, a restriction enzyme sensitive to methylation, cleaves specifically at an unmethylated 5 CCGG sites allowing us to digest the active X chromosome in females leaving the inactive one for LMPCR analysis. This approach has been successfully used previously [42,47–49]. Ten units of Hpa II restriction enzyme (New England BioLabs) were used to cleave 10 ␮g of genomic DNA incubated 1 h at 37 ◦ C. 2.3. Ligation-mediated PCR (LMPCR) The LMPCR protocol has been described previously in detail [45]. Briefly, irradiated genomic DNA was extracted and digested with T4 endonuclease V to incise the DNA at CPD sites (including 5 YU dimers [50]). The resulting 5 -pyrimidine overhangs were then removed by photoreactivation using E. coli CPD photolyase, in order to generate ligatable ends. A gene-specific oligonucleotide was annealed downstream of the break site, and the set of genomic cleavage products was extended using cloned Pfu exo- DNA polymerase (Strategene, LaJolla, CA). The Hpa II digestion of the female DNA insured that only the molecules from the inactivated X chromosome could undergo this primer extension. An asymmetric double-stranded linker was then ligated to the phosphate groups at the fragment 5 termini, providing a common sequence on the 5 -end of all fragments. A linker-specific primer, in conjunction with another gene-specific primer, was used in a PCR reaction to amplify the cleavage products of interest. These products were subjected to electrophoresis on 8% polyacrylamide sequencing gels alongside a Maxam and Gilbert sequencing ladder, transferred to nylon membranes, hybridized to a 32 P-labeled gene-specific probe, and visualized by autoradiography. Each experimental condition was assayed in duplicate. A screening sequencing gel was initially run using a portion of the DNA to ensure that there was no significant variation between samples. The two samples were then pooled on a combined gel, and the resulting autoradiogram analyzed using a Fuji BAS 1000 phosphorimager (Fuji Medical Systems, Stanford, CT, USA). Each band represents a nucleotide position where a break was induced by CPD cleavage, and the intensity of the band reflects the number of DNA molecules with ligatable ends at that position. Therefore, it is possible to make a quantitative correlation between the band intensity and the CPD formation at each dipyrimidine site.

P.J. Rochette et al. / Mutation Research 665 (2009) 7–13

9

Fig. 2. Graphical representation of the phosphorimager quantification of the LMPCR data shown in Figs. 3 and 4. Signal from methylated FMR1 mutated patient (Mt) and methylated inactive female X-chromosome (F) is compared with the signal at the same dipyrimidine site from the unmethylated male (M). Details of the quantification technique are explained in Section 3. Briefly, we attribute a value of 1 to each dipyrimidine site of the unmethylated male DNA (M). Then, the ratio between a methylated dipyrimidine (5 Ym CG) site from the female inactive X-chromosome (F) or the FMR1 mutated patient (Mt) and the same site but unmethylated from male DNA (M) is calculated. A ratio over 1 means methylation has a positive effect on CPD formation. Statistical significance using Student’s t-test is shown (*p value < 0.02, **p value <0.002, ***p value < 0.0002). Abbreviations: CPDs (cyclobutane pyrimidine dimers), M (male DNA, unmethylated), Mt (FMR1-mutated patient DNA, methylated), F (female inactive X-chromosome DNA, methylated). 2.4. Primers for LMPCR Primers specific for the PGK1 [42] and FMR1 gene were used for the LMPCR analysis. For FMR1 gene: primer X1: AAGTGAAACCGAAACGGAGC (position +39 to +20), primer X2: GAGCTGAGCGCCTGACTGAGGCCG (position +23 to −1); primer Y1: TCCACCAAGCCCGCGCA (position −203 to −187), primer Y2: TCTGTCTTTCGACCCGGCACCCCG (position −171 to −148) [17]. For PGK1 gene: primer A1: AAGTCGGGAAGGTTCCTT, primer A2: AAGGTTCCTTGCGGTTCGCGGCG (position −230 to −208); primer G1: TCACGTCCGTTCGCAGC, primer G2: CCGTTCGCAGCGTCACCCGGATC (position −311 to −289).

3. Results 3.1. Experimental design and quantification technique For this study, two different systems were used to compare the same chromosomal sequence, either methylated or not. The first takes advantage of the inactive methylated X chromosome in females. In males, there is only one X chromosome which is unmethylated (active). In females, one of the two X chromosomes is active as in males whereas the other is inactivated by cytosine methylation at 5 CG. After DNA extraction, the female active X chromosome was specifically degraded by a methyl-sensitive restriction enzyme (Hpa II). This procedure allows a comparison between male unmethylated and female methylated X chromosomes. Using this system, we studied methyl-specificity of CPD induction after different wavelengths in the X-linked PGK1 and FMR1 genes. The second system exploits one characteristic of the fragile-X syndrome. In afflicted male patients, FMR1 is mutated resulting in methylation of the gene. We thus compared CPD formation in the FMR1 gene from a male fragile-X patient versus a wild-type male. To prevent cellular context to affect CPD formation, genomic DNA was purified from cells, then UV irradiated. Five and seven 5 YCG sites were analyzed on PGK1 and FMR1 genes, respectively. CPD formation as measured following LMPCR

by the intensity of bands at each 5 YCG site. The relative intensity (in percentage) of each dipyrimidine site in a particular lane was calculated as the fraction of counts at this site in comparison with the total number of counts at all bands corresponding to all other dipyrimidine sites in that lane, after correction for background by subtracting the corresponding value in the unirradiated control lane (NoUV). In other words, the corrected number of counts for a specific band was divided by the total number of counts of all bands in that lane. The signal percentage at each dipyrimidine site from the methylated DNA (female or FMR1-mutated male) was then compared with the corresponding site in the male (unmethylated) and a ratio 5 Ym C/5 YC is derived. The point of this mode of quantification is to ensure that any given observed difference, even subtle, cannot be caused by differences between samples in DNA quantity, reaction efficiency or loading quantity during the experiment. At a particular site, a ratio <1 means that methylation inhibits CPD formation whereas a ratio >1 reflects that methylation enhances formation. The average and standard deviation of these ratios were calculated for the 12 sites. 3.2. Cytosine methylation at a given site in genomic DNA enhances CPD induction by UVB but not by UVC For CPD induction using UVC irradiation (254 nm), no significant difference was seen between methylated (5 Ym CG) and unmethylated (5 YCG) dipyrimidine sites (Fig. 2). The ratio 5 Ym CG/5 YCG was 1.06 ± 0.08 (Table 1). Within the UVB range (including 302 and 313 nm conditions), the CPD induction was significantly higher in methylated DNA compared to non-methylated DNA (Fig. 2). More precisely, the 5 Ym CG/5 YCG ratio was >1 in the two genes studied (FMR1 and PGK1) using both systems (female inactivated-X and male FMR1-mutated). Whether each system is considered separately or the results combined, the increases are

10

P.J. Rochette et al. / Mutation Research 665 (2009) 7–13

Table 1 Ratio of CPD frequency at methylated DNA/unmethylated DNA at each 5 YCG site analyzed on PGK1 and FMR1 genes (Figs. 3 and 4), quantified using a Fuji BAS 300 phosphorimager. The female inactive X-chromosome (F) and a fragile-X syndrome male patient (Mt) are used as methylated DNA whereas the normal male X-chromosome (M) is used as unmethylated DNA. 5 CCA sites are used as unmethylated control. Gene (Figure)

Site #

UVB (290–320 nm)

302 nm

M

Mt

F

M

Mt

F

M

Mt

F

M

Mt

F

1 2 3

5 CCA 5 CCG 5 TCG 5 TCG

1 1 1 1

X X X X

0.61 1.43 1.02 1.10

1 1 1 1

X X X X

0.84 1.40 1.80 1.82

1 1 1 1

X X X X

0.95 1.15 2.67 1.86

1 1 1 1

X X X X

0.88 1.12 2.70 2.28

4 5

5 CCA 5 TCG 5 CCG

1 1 1

X X X

0.87 0.95 1.07

1 1 1

X X X

0.68 1.88 1.66

1 1 1

X X X

0.74 1.44 2.09

1 1 1

X X X

1.00 2.69 1.66

6 7

5 CCA 5 CCG 5 CCG

1 1 1

1.20 0.93 1.14

1.01 0.95 0.83

1 1 1

0.81 2.30 1.94

0.98 1.57 1.29

1 1 1

1.04 2.16 1.75

1.01 1.46 1.08

1 1 1

0.82 1.58 1.31

0.96 1.22 0.82

8 9 10 11 12

5 CCA 5 CCG 5 CCG 5 CCG 5 TCG 5 CCG

1 1 1 1 1 1

0.89 1.10 1.02 1.05 0.98 1.29

0.91 1.22 1.30 0.88 0.87 0.97

1 1 1 1 1 1

0.70 1.52 2.27 2.02 1.42 2.69

0.95 1.53 1.96 1.28 1.04 1.46

1 1 1 1 1 1

0.86 1.68 1.74 1.68 1.30 2.43

0.80 2.64 1.70 1.28 1.31 1.39

1 1 1 1 1 1

0.83 2.50 1.67 1.79 2.44 2.37

0.76 2.80 1.82 1.34 2.11 1.12

PGK1 (Fig. 3A)

PGK1 (Fig. 3B)

FMR1 (Fig. 4A)

FMR1 (Fig. 4B)

Sequence

UVC (254 nm)

statistically significant (Fig. 2). The ratio 5 Ym CG/5 YCG for CPD formation were 1.73 ± 0.21, 1.73 ± 0.24 and 1.81 ± 0.31 when using UVB (290–320 nm), 302 or 313 nm, respectively. This methylationrelated increase of CPD formation by UVB wavelengths was higher in the FMR1-mutated X (Mt) than in the female inactive X (F) (Fig. 2). This held for all UVB sources (UVB (290–320 nm), 302 and 313 nm) but was only significant after polychromatic UVB. Four 5 CCA sites were used as unmethylated control for all conditions (male, female inactivated-X and male FMR1-mutated). The (female inactivatedX and male FMR1-mutated)/male ratios at these sites were 0.92, 0.86, 0.93 and 0.85 when using UVC, polychromatic UVB, 302, and 313 nm, respectively. 3.3. Increasing the wavelength within the UVB range does not affect CPD formation at 5 Ym CpG To investigated whether CPD formation at 5 Ym CG sites was influenced by the wavelength used within UVB range, irradiation with a monochromator at either 302 and 313 nm was used in addition to broadband UVB (290–320 nm). A small increase in CPD formation was observed at 5 Ym CG sites using 313 nm when compared with UVB or 302 nm. The 5 Ym CG/5 YCG ratio was 1.86, 1.73 and 1.73 for 313 nm, 302 nm and broadband UVB, respectively (Table 1, Fig. 2). However, this difference was not statistically significant. We also observed that the first pyrimidine in the 5 Ym CG sequence did not influence CPD formation: 5 Tm CG and 5 Cm CG were statistically indistinguishable (data not shown). 4. Discussion 4.1. UVB induces a higher frequency of CPDs at dipyrimidine sites containing a methylated cytosine Reports showing a higher CPD formation at dipyrimidine sites containing methylcytosine have used synthetic residues [35,51] or compared different sites for methylated versus non-methylated states [13,17]. We designed this study to unequivocally evaluate the effect of cytosine methylation on CPD formation following irradiation with different UV wavelengths and exclude potential effects of DNA sequence and/or chromatin on the frequency of CPD formation. Here we compare the same dipyrimidine sites methylated or not using human genomic DNA. We clearly show that methylation of cytosine has a positive effect on CPD induction by UVB.

313 nm

A dipyrimidine containing a cytosine is 1.7 times more sensitive to UVB when the cytosine is methylated compared with the same cytosine unmethylated (Table 1, Fig. 2). This is in accord with the 2-fold effect that has been published previously with synthetic residues [35]. This increased susceptibility can, at least in part, be explained by the fact that the absorption peak of cytosine is 267 nm whereas the absorption peak of methylated cytosine is 273.5 at pH 7.2. As a result, the molar absorption coefficient of methylated cytosine at 290 nm is about 5-fold higher than when unmethylated [37,38] (Fig. 5). The molar absorption coefficient of methylated cytosine is 1.3 times lower compared with an unmethylated cytosine at 254 nm UVC [37,38] (Fig. 5). Since UVC (254 nm) is below the absorption peak of cytosine, one might also expect that methylation of cytosine has an effect by decreasing UVC-induction of CPDs. This was not observed in our system, nor in previous studies. This difference in absorption may be insufficient to induce a sufficient change in CPD formation to be detected by the LMPCR technique. However, methylation of cytosine does have an effect by decreasing the probability of formation of the other UVC-induced dipyrimidine lesions, i.e., the 6-4PPs [52]. 4.2. The effect of methylation on CPD formation is greater on the FMR1-mutated male X chromosome than on the inactivated female X In this study, we used two systems to study the effect of methylation. The first was the FMR1-mutated male X-chromosome and the other was the inactivated female X-chromosome. In both systems, all 5 YCG are methylated within the regions analyzed. Unexpectedly, the methylation-related increase in CPD formation by UVB wavelengths was higher in the FMR1-mutated X (Mt) than in the female inactive X (F) (Fig. 2). This effect could be due, at least in part, to possible incomplete digestion of the unmethylated active female X-chromosome, slightly diluting the intensity of the bands from methylated DNA. Primers for the LMPCR reaction were designed to have a Hpa II site nearby, so the restriction cut in the unmethylated Hpa II-sensitive X chromosome would prevent amplification. However, some Hpa II sites can be seen in the autoradiograms (Figs. 3 and 4). This means that Hpa II digestion of the active X chromosome was not complete; as a result, the signal from 5 Ym CG sites was diluted with residual unmethylated 5 YCG from the active X containing lower amounts of CPD.

P.J. Rochette et al. / Mutation Research 665 (2009) 7–13

Fig. 3. LMPCR autoradiogram depicting UV-induced CPD formation at nucleotide resolution along a portion of the PGK1 promoter from the unmethylated male X chromosome (M) or the inactive (methylated) female X chromosome (F). (A) primer set A, (B) primer set G (see Section 2). The first four lanes from the left are Maxam–Gilbert cleavage reactions. “NoUV” lanes show LMPCR of unirradiated DNA treated with T4 endonuclease V plus photolyase. “UVB” lanes are LMPCR of DNA irradiated with 10,000 J/m2 UVB light (290–320 nm) (see Fig. 1 for the output spectrum). The “UVC” lanes are LMPCR of DNA irradiated with 200 J/m2 UVC light (254 nm). “302” and “313” are LMPCR of DNA irradiated with 12,000 or 75,000 of 302 or 313 nm light, respectively. Arrows indicate dipyrimidine sites that were quantified using a Fuji BAS 1000 phosphorimager using the Image Gauge V3.0 program. The sequences of those sites are written 5 → 3 . The site of Hpa II digestion is indicated. The open arrow is a background site containing no CPD.

11

Fig. 4. LMPCR autoradiogram depicting UV-induced CPD formation at nucleotide resolution along a portion of FMR1 in DNA from the unmethylated male X chromosome (M), a male fragile-X syndrome patient (methylated) (Mt), or the inactive (methylated) female X chromosome (F). (A) primer set X, (B) primer set Y. The first four lanes from the left are Maxam–Gilbert cleavage reactions. Other abbreviations as in Fig. 3.

4.3. Increasing the wavelength within the UVB range does not affect CPD formation at 5 Ym CG Little is known about the absorption of cytosine or of methylated cytosine at wavelengths >300 nm. In fact, the effects of light scattering (i.e. turbidity) make the measurement of the absorption spectrum of DNA at wavelengths above 300 nm very complex [53]. To measure the effect of increasing the wavelength within the UVB range on CPD formation, we used monochromatic UV at 302 and 313 nm. In addition, we used polychromatic UVB (290–320 nm with

12

P.J. Rochette et al. / Mutation Research 665 (2009) 7–13

Fig. 5. Molar extinction coefficiency of cytosine and 5-methylcytosine at pH 7.2 using different wavelengths. Results depicted in this graph are derived from Shugar and Fox [37].

a peak at 310 nm). According to the measured output of this broadband UVB lamp, more than 50% of all UVB energy of this lamp is between 302 and 313 nm (Fig. 1). Our results show that increasing the wavelength in the UVB range (313 nm) leads to an increase in CPD formation at 5 Ym CG (i.e. 1.83 vs 1.73 and 1.73 at 313 nm vs UVB and 302 nm, respectively) (Table 1, Fig. 2). However, this difference is not significant (p value = 0.27 and 0.60 for 313 nm vs UVB and 302 nm, respectively). These experimental results are consistent with the literature indicating an increase of the influence of methylation with the irradiation wavelength. In summary, methylation of cytosine increases UVB-induction of CPD formation by a factor of 1.7. We think this increased in CPD formation in methylated cytosine dipyrimidine sites have a major effect on the probability of a mutation to occur. First, there is the direct obvious effect of having more DNA damage that leads to more mutation by polymerase bypass or sliding. In addition to that, the deamination plays an important role. A cytosine in a CPD deaminates hundred times faster [25,27,29–31] compared to cytosines not covalently bound to another pyrimidine. If the deamination occurs in an unmethylated cytosine, it converts it into an uracile whereas if the cytosine is methylated it deaminates into a thymidine. Then, a methylated cytosine in a dipyrimidine site has 1.7 times more chances to form a CPD after a UVB exposure, which leads to an increase in deamination and this deamination will likely cause a C → T transition. Conflict of interest None declared. Acknowledgements The authors are grateful to Dr. R. Stephen Lloyd and Tim O’Connor for supplying T4 endonuclease V and photolyase, respectively. The research project was supported by the National Cancer Institute of Canada (NCIC) (with funds from the Canadian Cancer Society and the Terry Fox Run) and the Canada Research Chairs Program to R.D.R. Drouin holds the Canada Research Chair in “Genetics, Mutagenesis and Cancer”. P.J. Rochette is the recipient of a postdoctoral fellowship from the Fonds de la Recherche en Santé du Québec (FRSQ). References [1] H.S. Black, F.R. deGruijl, P.D. Forbes, J.E. Cleaver, H.N. Ananthaswamy, E.C. deFabo, S.E. Ullrich, R.M. Tyrrell, Photocarcinogenesis: an overview, J. Photochem. Photobiol. B 40 (1997) 29–47.

[2] R.B. Setlow, The wavelengths in sunlight effective in producing skin cancer: a theoretical analysis, Proc. Natl. Acad. Sci. U.S.A. 71 (1974) 3363–3366. [3] E.C. Friedberg, G.C. Walker, W. Siede, DNA Repair and Mutagenesis, ASM Press, Washington, DC, 1995. [4] D. Perdiz, P. Grof, M. Mezzina, O. Nikaido, E. Moustacchi, E. Sage, Distribution and repair of bipyrimidine photoproducts in solar UV-irradiated mammalian cells. Possible role of Dewar photoproducts in solar mutagenesis, J. Biol. Chem. 275 (2000) 26732–26742. [5] P.J. Rochette, J.P. Therrien, R. Drouin, D. Perdiz, N. Bastien, E.A. Drobetsky, E. Sage, UVA-induced cyclobutane pyrimidine dimers form predominantly at thymine–thymine dipyrimidines and correlate with the mutation spectrum in rodent cells, Nucleic Acids Res. 31 (2003) 2786–2794. [6] B.W. Glickman, R.M. Schaaper, W.A. Haseltine, R.L. Dunn, D.E. Brash, The C–C (6-4) UV photoproduct is mutagenic in Escherichia coli, Proc. Natl. Acad. Sci. U.S.A. 83 (1986) 6945–6949. [7] D.E. Brash, S. Seetharam, K.H. Kraemer, M.M. Seidman, A. Bredberg, Photoproduct frequency is not the major determinant of UV base substitution hot spots or cold spots in human cells, Proc. Natl. Acad. Sci. U.S.A. 84 (1987) 3782– 3786. [8] G.P. Pfeifer, Formation and processing of UV photoproducts: effects of DNA sequence and chromatin environment, Photochem. Photobiol. 65 (1997) 270–283. [9] D.E. Brash, J.A. Rudolph, J.A. Simon, A. Lin, G.J. McKenna, H.P. Baden, A.J. Halperin, J. Ponten, A role for sunlight in skin cancer: UV-induced p53 mutations in squamous cell carcinoma, Proc. Natl. Acad. Sci. U.S.A. 88 (1991) 10124–10128. [10] A.J. Nataraj, J.C. Trent 2nd, H.N. Ananthaswamy, p53 gene mutations and photocarcinogenesis, Photochem. Photobiol. 62 (1995) 218–230. [11] F. Hutchinson, Induction of tandem-base change mutations, Mutat. Res. 309 (1994) 11–15. [12] S. Tornaletti, G.P. Pfeifer, Complete and tissue-independent methylation of CpG sites in the p53 gene: implications for mutations in human cancers, Oncogene 10 (1995) 1493–1499. [13] S. Tommasi, M.F. Denissenko, G.P. Pfeifer, Sunlight induces pyrimidine dimers preferentially at 5-methylcytosine bases, Cancer Res. 57 (1997) 4727–4730. [14] A. Ziegler, D.J. Leffell, S. Kunala, H.W. Sharma, M. Gailani, J.A. Simon, A.J. Halperin, H.P. Baden, P.E. Shapiro, A.E. Bale, et al., Mutation hotspots due to sunlight in the p53 gene of nonmelanoma skin cancers, Proc. Natl. Acad. Sci. U.S.A. 90 (1993) 4216–4220. [15] G.P. Pfeifer, S.D. Steigerwald, S. Grunwald, The DNA methylation system in proliferating and differentiated cells, Cell. Biophys. 15 (1989) 79–86. [16] W.M. Rideout 3rd, G.A. Coetzee, A.F. Olumi, P.A. Jones, 5-Methylcytosine as an endogenous mutagen in the human LDL receptor and p53 genes, Science 249 (1990) 1288–1290. [17] R. Drouin, J.P. Therrien, UVB-induced cyclobutane pyrimidine dimer frequency correlates with skin cancer mutational hotspots in p53, Photochem. Photobiol. 66 (1997) 719–726. [18] S. Tornaletti, D. Rozek, G.P. Pfeifer, The distribution of UV photoproducts along the human p53 gene and its relation to mutations in skin cancer, Oncogene 8 (1993) 2051–2057. [19] H. Ikehata, H. Kudo, T. Masuda, T. Ono, UVA induces C → T transitions at methylCpG-associated dipyrimidine sites in mouse skin epidermis more frequently than UVB, Mutagenesis 18 (2003) 511–519. [20] H. Ikehata, T. Masuda, H. Sakata, T. Ono, Analysis of mutation spectra in UVBexposed mouse skin epidermis and dermis: frequent occurrence of C → T transition at methylated CpG-associated dipyrimidine sites, Environ. Mol. Mutagen. 41 (2003) 280–292. [21] H. Ikehata, S. Nakamura, T. Asamura, T. Ono, Mutation spectrum in sunlightexposed mouse skin epidermis: small but appreciable contribution of oxidative stress-mediated mutagenesis, Mutat. Res. 556 (2004) 11–24. [22] P.E. Gibbs, C.W. Lawrence, U–U and T–T cyclobutane dimers have different mutational properties, Nucleic Acids Res. 21 (1993) 4059–4065. [23] N. Jiang, J.S. Taylor, In vivo evidence that UV-induced C → T mutations at dipyrimidine sites could result from the replicative bypass of cis-syn cyclobutane dimers or their deamination products, Biochemistry 32 (1993) 472– 481. [24] C. Masutani, R. Kusumoto, S. Iwai, F. Hanaoka, Mechanisms of accurate translesion synthesis by human DNA polymerase eta, EMBO J. 19 (2000) 3100–3109. [25] Y. Barak, O. Cohen-Fix, Z. Livneh, Deamination of cytosine-containing pyrimidine photodimers in UV-irradiated DNA. Significance for UV light mutagenesis, J. Biol. Chem. 270 (1995) 24174–24179. [26] L.A. Frederico, T.A. Kunkel, B.R. Shaw, A sensitive genetic assay for the detection of cytosine deamination: determination of rate constants and the activation energy, Biochemistry 29 (1990) 2532–2537. [27] W. Peng, B.R. Shaw, Accelerated deamination of cytosine residues in UV-induced cyclobutane pyrimidine dimers leads to CC → TT transitions, Biochemistry 35 (1996) 10172–10181. [28] X. Zhang, C.K. Mathews, Effect of DNA cytosine methylation upon deaminationinduced mutagenesis in a natural target sequence in duplex DNA, J. Biol. Chem. 269 (1994) 7066–7069. [29] J. Burger, Assessing perceptions about ecosystem health and restoration options in three east coast estuaries, Environ. Monit. Assess 83 (2003) 145–162. [30] R.B. Setlow, W.L. Carrier, Pyrimidine dimers in ultraviolet-irradiated DNA’s, J. Mol. Biol. 17 (1966) 237–254. [31] Y. Tu, R. Dammann, G.P. Pfeifer, Sequence and time-dependent deamination of cytosine bases in UVB-induced cyclobutane pyrimidine dimers in vivo, J. Mol. Biol. 284 (1998) 297–311.

P.J. Rochette et al. / Mutation Research 665 (2009) 7–13 [32] D.H. Lee, G.P. Pfeifer, Deamination of 5-methylcytosines within cyclobutane pyrimidine dimers is an important component of UVB mutagenesis, J. Biol. Chem. 278 (2003) 10314–10321. [33] V.A. Bohr, D.S. Okumoto, L. Ho, P.C. Hanawalt, Characterization of a DNA repair domain containing the dihydrofolate reductase gene in Chinese hamster ovary cells, J. Biol. Chem. 261 (1986) 16666–16672. [34] L. Ho, V.A. Bohr, P.C. Hanawalt, Demethylation enhances removal of pyrimidine dimers from the overall genome and from specific DNA sequences in Chinese hamster ovary cells, Mol. Cell. Biol. 9 (1989) 1594–1603. [35] D.L. Mitchell, Effects of cytosine methylation on pyrimidine dimer formation in DNA, Photochem. Photobiol. 71 (2000) 162–165. [36] P. Monti, A. Inga, G. Scott, A. Aprile, P. Campomenosi, P. Menichini, L. Ottaggio, S. Viaggi, A. Abbondandolo, P.A. Burns, G. Fronza, 5-Methylcytosine at HpaII sites in p53 is not hypermutable after UVC irradiation, Mutat. Res. 431 (1999) 93–103. [37] D. Shugar, J.J. Fox, Spectrophotometric studies of nucleic acid derivatives and related compounds as a function of pH. I. Pyrimidines, Biochim. Biophys. Acta 9 (1952) 199–218. [38] Y.H. You, C. Li, G.P. Pfeifer, Involvement of 5-methylcytosine in sunlight-induced mutagenesis, J. Mol. Biol. 293 (1999) 493–503. [39] S.G. Grant, V.M. Chapman, Mechanisms of X-chromosome regulation, Annu. Rev. Genet. 22 (1988) 199–233. [40] C.J. Brown, R.G. Lafreniere, V.E. Powers, G. Sebastio, A. Ballabio, A.L. Pettigrew, D.H. Ledbetter, E. Levy, I.W. Craig, H.F. Willard, Localization of the X inactivation centre on the human X chromosome in Xq13, Nature 349 (1991) 82–84. [41] A.P. Bird, CpG-rich islands and the function of DNA methylation, Nature 321 (1986) 209–213. [42] G.P. Pfeifer, S.D. Steigerwald, R.S. Hansen, S.M. Gartler, A.D. Riggs, Polymerase chain reaction-aided genomic sequencing of an X chromosome-linked CpG island: methylation patterns suggest clonal inheritance, CpG site autonomy, and an explanation of activity state stability, Proc. Natl. Acad. Sci. U.S.A. 87 (1990) 8252–8256.

13

[43] G.P. Pfeifer, R.L. Tanguay, S.D. Steigerwald, A.D. Riggs, In vivo footprint and methylation analysis by PCR-aided genomic sequencing: comparison of active and inactive X chromosomal DNA at the CpG island and promoter of human PGK-1, Genes Dev. 4 (1990) 1277–1287. [44] I. Oberle, F. Rousseau, D. Heitz, C. Kretz, D. Devys, A. Hanauer, J. Boue, M. Bertheas, J. Mandel, Instability of a 550-base pair DNA segment and abnormal methylation in fragile X syndrome, Science 252 (1991) 1097–1102. [45] R. Drouin, J.P. Therrien, M. Angers, S. Ouellet, In vivo DNA analysis, Methods Mol. Biol. 148 (2001) 175–219. [46] R. Drouin, S. Gao, G.P. Holmquist, in: G.P. Pfeifer (Ed.), Technologies for Detection of DNA Damage and Mutations, Plenum Press, New York, 1996, pp. 37–43. [47] D.H. Keith, J. Singer-Sam, A.D. Riggs, Active X chromosome DNA is unmethylated at eight CCGG sites clustered in a guanine-plus-cytosine-rich island at the 5 end of the gene for phosphoglycerate kinase, Mol. Cell. Biol. 6 (1986) 4122–4125. [48] J. Singer-Sam, J.M. LeBon, R.L. Tanguay, A.D. Riggs, A quantitative HpaII-PCR assay to measure methylation of DNA from a small number of cells, Nucleic Acids Res. 18 (1990) 687. [49] S. Tommasi, J.M. LeBon, A.D. Riggs, J. Singer-Sam, Methylation analysis by genomic sequencing of 5 region of mouse Pgk-1 gene and a cautionary note concerning the method, Somat. Cell. Mol. Genet. 19 (1993) 529–541. [50] K. Takasawa, C. Masutani, F. Hanaoka, S. Iwai, Chemical synthesis and translesion replication of a cis-syn cyclobutane thymine-uracil dimer, Nucleic Acids Res. 32 (2004) 1738–1745. [51] T. Douki, J. Cadet, Formation of cyclobutane dimers and (6-4) photoproducts upon far-UV photolysis of 5-methylcytosine-containing dinucleotide monophosphates, Biochemistry 33 (1994) 11942–11950. [52] G.P. Pfeifer, R. Drouin, A.D. Riggs, G.P. Holmquist, In vivo mapping of a DNA adduct at nucleotide resolution: detection of pyrimidine (6-4) pyrimidone photoproducts by ligation-mediated polymerase chain reaction, Proc. Natl. Acad. Sci. U.S.A. 88 (1991) 1374–1378. [53] J.C. Sutherland, K.P. Griffin, Absorption spectrum of DNA for wavelengths greater than 300 nm, Radiat. Res. 86 (1981) 399–409.