Replication of N2-Ethyldeoxyguanosine DNA Adducts in the Human

N2-Ethyldeoxyguanosine (N2-ethyldGuo) is a DNA adduct formed by reaction ... The 293T cell line is a variant of 293 that is postreplication mismatch r...
0 downloads 0 Views 200KB Size
Download PDF
960

Chem. Res. Toxicol. 2006, 19, 960-967

Replication of N2-Ethyldeoxyguanosine DNA Adducts in the Human Embryonic Kidney Cell Line 293 Dana C. Upton,† Xueying Wang,† Patrick Blans,‡ Fred W. Perrino,† James C. Fishbein,*,‡ and Steven A. Akman*,† Wake Forest UniVersity Health Sciences, 1 Medical Center BouleVard, Winston-Salem, North Carolina 27157, and UniVersity of Maryland, Baltimore County, Baltimore, Maryland 21250 ReceiVed April 21, 2006

N2-Ethyldeoxyguanosine (N2-ethyldGuo) is a DNA adduct formed by reaction of the exocyclic amine of dGuo with the ethanol metabolite acetaldehyde. Because ethanol is a human carcinogen, we assessed the biological consequences of replication of template N2-ethyldGuo, in comparison to the well-studied adduct O6-ethyldeoxyguanosine (O6-ethyldGuo). Single chemically synthesized N2-ethyldGuo or O6ethyldGuo adducts were placed site specifically in the suppressor tRNA gene of the mutation reporting shuttle plasmid pLSX. N2-EthyldGuo and O6-ethyldGuo were both minimally mutagenic in double-stranded pLSX replicated in human 293 cells; however, the placement of deoxyuridines on the complementary strand at 5′- and 3′-positions flanking the adduct resulted in 5- and 22-fold enhancements of the N2ethyldGuo- and O6-ethyldGuo-induced mutant fractions, respectively. The fold increase in the N2ethyldGuo-induced mutant fraction in deoxyuridine-containing plasmids was similar after replication in 293T cells, a mismatch repair deficient variant of 293 cells, indicating that postreplication mismatch repair has little role in modulating N2-ethyldGuo-mediated mutagenesis. The mutation spectrum generated by N2-ethyldGuo consisted primarily of single base deletions and adduct site-targeted transversions, in contrast to the exclusive production of adduct site-targeted transitions by O6-ethyldGuo. The yield of progeny plasmids after replication in 293 cells was reduced by the presence of N2-ethyldGuo in parental plasmids with or without deoxyuridine to 39 or 19%, respectively. Taken together, these data indicate that N2-ethyldGuo in DNA exerts its principal biological activity by blocking translesion DNA synthesis in human cells, resulting in either failure of replication or frameshift deletion mutations. Introduction The association between alcoholic beverage consumption and the risk of aerodigestive cancer development is well-documented (1-4). In this regard, ethanol has been classified as a known human carcinogen (5). Ethanol is predominantly converted in vivo to acetaldehyde (6), which may form a Schiff base with exocyclic amino groups of DNA bases that, upon reduction, causes the formation of stable exocyclic alkylamino adducts. Recent work by Wang et al. indicates that the primary adduct caused by acetaldehyde is N2-ethylidenedeoxyguanosine (7), which may be converted by in vivo reducing agents such as glutathione or ascorbate to N2-ethyldeoxyguanosine (N2ethyldGuo)1 (8). N2-EthyldGuo has been found in the DNA of ethanol-treated mice (8), the DNA of human alcoholics (9), and the urine of human volunteers who have abstained from alcohol for at least a week (10). While the biological consequences of N2-ethyldGuo adducts in human DNA are unknown, in vitro data indicate that N2ethyldGuo blocks translesion DNA synthesis catalyzed by a variety of DNA polymerases. Choi and Guengerich observed that N2-ethyldGuo blocks the viral polymerases bacteriophage T7 exonuclease- and HIV-1 reverse transcriptase (11). Similarly, * To whom correspondence should be addressed. (J.C.F.) Tel: 410-4552190. E-mail: [email protected]. (S.A.A.) Tel: 336-716-0231. Fax: 336716-0255. E-mail: [email protected]. † Wake Forest University Health Sciences. ‡ University of Maryland. 1 Abbreviations: N2-ethyldGuo, N2-ethyldeoxyguanosine; O6-ethyldGuo, O6-ethyldeoxyguanosine; MMR, mismatch repair.

Terashima et al. noted that the adduct blocks Escherichia coli DNA polymerase I at, or one base prior, to the lesion, suggesting that this adduct might block translesion DNA synthesis in E. coli in vivo (12). Template N2-ethyldGuo strongly blocks synthesis catalyzed by the human replicative DNA polymerase R (13), while both the human translesional polymerases η and ι bypass the N2-ethyldGuo adduct (13-15). Although N2-ethyldGuo strongly blocked translesion synthesis catalyzed by E. coli polymerase I (12), insertion of dGuo was favored in the case of DNA polymerase I-catalyzed insertion across from and extension beyond N2-ethyldGuo. These data suggest that N2-ethyldGuo might induce G:C f C:G transversions in vivo (12). However, the mutation spectrum of a single chemically synthesized N2-ethyldGuo adduct serving as a template in E. coli in vivo consisted of G:C f T:A transversions and single base deletions, at either the adduct site or 3-5 nucleotides downstream of the adduct (16). The in vitro data demonstrating that N2-ethyldGuo blocks translesion DNA synthesis catalyzed by certain B family human DNA polymerases while permitting efficient translesion synthesis by Y family “bypass” polymerases (13) suggest that N2ethyldGuo adducts may affect faithful DNA replication in vivo. Considering the possibility that N2-ethyldGuo might mediate some of the procarcinogenic effects of ethanol consumption, we sought to determine if N2-ethyldGuo perturbs DNA replication in vivo. To do so, we inserted this adduct site specifically in a mutation reporting shuttle vector and observed the consequences of adduct-containing vector replication in human embryonic kidney 293 and 293T cell lines. The 293T cell line

10.1021/tx060084a CCC: $33.50 © 2006 American Chemical Society Published on Web 07/01/2006

N2-Ethyldeoxyguanosine DNA Adducts in Human Cells

Chem. Res. Toxicol., Vol. 19, No. 7, 2006 961

is a variant of 293 that is postreplication mismatch repair (MMR) deficient due to hypermethylation of the hMLH1 promoter region, which results in epigenetic silencing (17). Our data indicate that the N2-ethyldGuo adduct blocks translesion DNA synthesis in 293 cells in vivo. Consistent with its replication blocking activity, N2-ethyldGuo caused a decrease in the yield of progeny plasmids and a high prevalence of frameshift deletions. N2-EthyldGuo differed from the well-known mutagenic DNA adduct O6-ethyldeoxyguanosine (O6-ethyldGuo) in both the magnitude of the adduct-induced mutant fraction and the mutation spectrum. N2-EthyldGuo induced a smaller mutant fraction than did O6-ethyldGuo. Also, whereas O6-ethyldGuo caused G:C f A:T transitions exclusively, N2-ethyldGuo caused a majority of frameshift deletions and a minority of G:C f T:A transversions in both 293 and 293T cells. Finally, there was no significant difference in the mutation fraction of N2-ethyldGuo in 293 vs 293T cells, suggesting that MMR plays little or no role in minimizing N2-ethyldGuoinduced mutagenesis in vivo.

Experimental Procedures Synthesis of Oligonucleotides Containing Exocyclic Alkylamino DNA Adducts. The methods for synthesis, purification, and chemical analysis of the phosphoramidites of O6-ethyldGuo and N2-ethyldGuo, as well as those for the synthesis and purification of oligodeoxynucleotides containing these adducts, have been previously published (13). Purine base analysis of the oligodeoxynucleotides was performed by acid hydrolysis and quantitation of the natural and adducted bases by HPLC/MS by methods previously reported (13). In all cases, the ratio of adducted to natural base was within less than 4% of theoretical. MALDI-TOF MS analysis of each oligodeoxynucleotide gave a molecular ion with a mass within 0.005% of theoretical. Construction of Adduct-Containing Plasmids. Oligodeoxynucleotides were 5′-phosphorylated using 200 pmol of oligodeoxynucleotide and 20 units of T4 DNA polynucleotide kinase in 60 mM Tris-HCl, pH 7.6, 15 mM MgCl2, 7.5 mM dithiothreitol, 2.5% PEG-8000, and 5 mM ATP. After incubation for 1 h at 37 °C, the reaction was terminated at 70 °C for 10 min. Phosphorylated oligodeoxynucleotides were then annealed by incubation of 200 pmol of each oligodeoxynucleotide in buffer (1 M NaCl, 1 M Hepes) at 68 °C for 15 min followed by a decrease of -1 °C/min to 25 °C. All restriction enzymes used to make adduct-containing constructs were obtained from New England Biolabs (Ipswich, MA) and were reacted in the supplied buffers. Mutation studies were carried out using plasmid pLSX (Figure 1), a derivative of pZ189 (18). pLSX contains β-lactamase for ampicillin resistance, the T antigen coding region and SV40 origin of replication for replication in human cells, the pBR322 origin of replication for replication in E. coli, and the supF tRNA gene. Double-stranded pLSX DNA was prepared in E. coli strain JM109 cultured in Luria-Bertani broth supplemented with 75 µg/mL ampicillin. Plasmid DNA was isolated using the GenElute HP Plasmid Maxiprep kit (Sigma-Aldrich, St. Louis, MO) according to the manufacturer’s instructions. Plasmid DNA was then digested with five units/µg of XhoI and BspEI, and the resultant doubly digested plasmids were separated from the 82 base pair stuffer fragment and purified by agarose gel electrophoresis (QIAquick gel extraction kit, Qiagen, Valencia, CA). Ten picomoles of doublestranded adduct-containing oligodeoxynucleotides and 1.5 pmol of gel-purified XhoI/BspEI-cleaved pLSX were ligated with 1600 units of T4 DNA ligase at 16 °C overnight. The adduct-containing constructs were recovered by precipitation with 70% ethanol. The constructs were then transfected into human embryonic kidney 293 cells or 293T cells. PCR-Based Quantification of Adduct-Containing Plasmids. The quantity of plasmid generated by ligation was estimated by

Figure 1. (A) Scheme of the reporting plasmid pLSX containing a single exocyclic amino purine adduct. pLSX has an 82 nucleotide “stuffer fragment” containing an XbaI site placed in the supF tRNA gene that is removed by cleavage with XhoI (located at nucleotides 101-106 of the final construct) and BspEI (located at nucleotides 124129 of the final construct) (18). Double-stranded oligodeoxynucleotides containing a single O6-ethyldGuo or N2-ethyldGuo adduct are ligated into the XhoI-BspEI site in the plasmid, recreating the wild-type supF gene with the adduct in place in the acceptor stem of the tRNA. (B) The sequences of the double-stranded oligodeoxynucleotides ligated into the XhoI-BspEI site of the supF tRNA gene are listed. X indicates the adducted deoxyguanosine.

PCR. Primers were designed to amplify a 200 base region immediately downstream of the site of the adduct with a forward primer that annealed over the site of the adduct (forward primer, 5′-d(pTGCTCGAGACTTCXAAGGGT)-3′ where X indicates the site complementary to the adducted deoxyguanosine; reverse primer, 5′-d(pTTGAGCGTCGATTTTTGTGA)-3′; primers synthesized by Operon Biotechnologies, Huntsville, AL). Control experiments showed that the adduct did not interfere with replication catalyzed by the PCR polymerase. Also, no product was generated in the absence of ligated insert (data not shown), such that the presence of a PCR product confirmed incorporation of a double-stranded oligomer into the construct. Appropriately diluted aliquots of linear adduct-containing plasmid DNA were amplified in a 25 µL reaction containing 2× PCR Master Mix (Promega, Madison, WI), 5 pmol of forward primer, 5 pmol of reverse primer, and 1 mM additional MgCl2. PCR was carried out as follows: an initial cycle of 94 °C for 5 min, followed by 20 cycles of 94 °C × 30 s, 60 °C × 30 s, 72 °C × 1 min, and a final extension at 72 °C for 5 min. Amplified material was run on a 1% agarose gel at 100 V for 20 min and then imaged using a Typhoon 8600 imager (Molecular Dynamics, Amersham Biosciences, Piscataway, NJ) and quantified with ImageQuant 5.2 software (Molecular Dynamics, Amersham Biosciences).

962 Chem. Res. Toxicol., Vol. 19, No. 7, 2006 Best-fit regression lines were calculated using the signal intensities of the bands corresponding to 0.1-0.75 ng of control parental pLSX plasmid and then used to back-calculate the amount of DNA present for each construct. R2 ) 0.94 ( 0.08 for the regression lines. Transfection and Recovery of Adduct-Containing Plasmids. Human embryonic kidney 293 cells were maintained in Dulbecco’s modified Eagle medium supplemented with 1% L-glutamine, 1% penicillin-streptomycin, and 10% fetal bovine serum. 293T cells were maintained in the above medium with the addition of 1.5 g/L sodium bicarbonate. Cells were seeded into six well tissue culture dishes and grown to 90-95% confluency. The cells were transfected with 0.5 µg of each construct using 6 µL of Lipofectamine cationic lipid reagent (Invitrogen, Carlsbad, CA). After transfection, the cells were allowed to grow at 37 °C in 5% CO2 for 48 h, after which plasmid DNA was recovered and purified using the Hirt procedure. Recovered plasmid DNA was digested with 0.1 units of DpnI at 37 °C for 1 h to eliminate unreplicated DNA. After DpnI cutting, plasmid DNA was recovered by ethanol precipitation and resuspended in 20 µL of 0.5× TE. Transformation of Reporter Bacteria. Thirty-five microliters of electrocompetent E. coli strain MBM7070 cells was transformed with 1 µL of recovered plasmid DNA in a 2 mm gapped electroporation cuvette pulsed at 2.5 kV, 2.5 µFd, and 200 Ω. Transformed E. coli were recovered in 500 µL of SOC medium (20 g/L tryptone, 5 g/L yeast extract, 0.5 g/L NaCl, 20 mM glucose, 10 mM MgCl2, and 2.5 mM KCl) and incubated at 37 °C for 20 min. One hundred microliters of cells was then plated onto S-Gal/ LB agar plates [12 g/L agar, 10 g/L tryptone, 10 g/L NaCl, 5 g/L yeast extract, 0.5 g/L ferric ammonium citrate, 0.3 g/L 3,4cyclohexenoesculetin-β-D-galactopyranoside (S-Gal), 0.03 g/L isopropyl-1-thio-β-D-galactoside (IPTG)] supplemented with 75 µg/ mL ampicillin. Colonies were scored as black (wild-type supF) or white (mutant supF). Mutant colonies were confirmed by secondary streaking. The mutant fraction was determined by the number of white colonies divided by the total number of colonies. Plasmid DNA was harvested from mutant colony minipreparations (QIAprep spin miniprep kit, Qiagen), after which the supF gene was sequenced by automated DNA sequencing (Wake Forest University School of Medicine DNA Sequencing Laboratory, Winston-Salem, NC, and Macrogen, Seoul, South Korea). Western Blot Analysis. hMLH1 protein expression was analyzed in 293 and 293T cells by Western blotting. Cell lysates (10 µg protein per sample) were separated on a NuPAGE Novex Bis-Tris 4-12% gradient gel (Invitrogen) and transferred to PVDF membranes, which were blocked in 5% milk/TBST buffer (150 mM NaCl, 10 mM Tris-HCL, and 0.1% Tween) for 1 h. hMLH1 protein was detected by blotting with 1:300 rabbit anti-human MLH1 (N20) polyclonal antibody (Santa Cruz Biotechnology Inc., Santa Cruz, CA) in 5% milk/TBST for 1 h. The primary antibody was detected by blotting with 1:10000 secondary goat anti-rabbit-HRP antibody (Bio-Rad, Hercules, CA) in 5% milk/TBST for 1 h. The membrane was reacted with ECL Plus (Amersham Biosciences) according to the manufacturer’s instructions and imaged on film.

Results N2-EthyldGuo

Presence of in Double-Stranded Plasmids Reduces the Progeny Yield: Comparison to O6-EthyldGuo. The biological consequences of the exocyclic alkylamino purine adduct N2-ethyldGuo as template for DNA synthesis in vivo were studied by placing single chemically synthesized adducts into the supF tRNA gene of plasmid pLSX (Figure 1A). Adducts were placed site specifically in the double-stranded plasmid that corresponds to a position located in the acceptor stem of the transcribed tRNA, such that a base substitution or deletion at that position would cause a loss of tRNA function (18, 19). Replication of the adduct-containing plasmids in human cells was initiated from the SV40 origin. Progeny plasmids were

Upton et al.

recovered, and the yields and mutant fractions were assessed after transformation into the mutation reporting E. coli strain MBM7070. Considering that in vitro data indicated that N2-ethyldGuo blocks the human replicative polymerase R (13), we assessed the yield of progeny plasmids after replication of adductcontaining plasmids in human 293 cells in order to determine whether N2-ethyldGuo blocks translesion synthesis in vivo. The amount of input plasmid was estimated by quantitative PCR, and similar amounts of control and adduct-containing plasmids were used. The presence of N2-ethyldGuo or O6-ethyldGuo markedly reduced the average number of progeny colonies recovered after replication in 293 cells (Figure 2). Comparing the ratio of progeny colonies to the amount of input plasmid DNA, there was ca. 23% percent decrease in the number of progeny recovered from O6-ethyldGuo-containing parental plasmid and ca. 80% decrease in the number of progeny recovered from N2-ethyldGuo-containing parental plasmid. For deoxyuridine-containing constructs, there was ca. 45% decrease in the number of progeny recovered from O6-ethyldGuo-containing parental plasmid and ca. 60% decrease in the number of progeny recovered from N2-ethyldGuo-containing parental plasmid. Mutagenicity of N2-EthyldGuo in Double-Stranded Plasmids with or without Deoxyuridine in the Strand Complementary to the Adduct: Comparison to O6-EthyldGuo. The mutant fractions of double-stranded pLSX containing either a single N2-ethyldGuo or a O6-ethyldGuo adduct were determined after replication in human embryonic kidney 293 cells. N2EthyldGuo and O6-ethyldGuo only caused a minimal enhancement of the mutant fraction as compared to the mutant fraction of control dGuo-containing plasmids (Table 1 and Figure 3). The mutagenicity of these adducts was also studied in doublestranded plasmids in which deoxyuridines were placed on the complementary strand at 5′- and 3′-positions flanking the adduct (Figure 1B). The presence of deoxyuridines did not result in a statistically significant change in the mutant fraction of control dGuo-containing constructs replicated in human 293 cells (Table 1 and Figure 3). The presence of the O6-ethyldGuo adduct enhanced the mutant fraction of constructs containing deoxyuridines by 22-fold in 293 cells (p ) 0.00006) (Table 1 and Figure 3). The presence of the N2-ethyldGuo adduct caused a smaller but still statistically significant 5-fold increase in the mutant fraction above that of the deoxyuridine-containing control constructs in 293 cells (p ) 0.02) (Table 1 and Figure 3). Mutagenicity of N2-EthyldGuo and O6-EthyldGuo in Double-Stranded Plasmids Containing Deoxyuridine in the Strand Complementary to the Adduct: Comparison of MMR Competent 293 and MMR Deficient 293T Cells. The deoxyuridine-containing constructs were transfected into both MMR competent 293 cells and MMR deficient 293T cells. The presence of hMLH1 protein in 293 cells and its absence in 293T cells was confirmed by Western blotting (Figure 4). There was no statistically significant difference in the mutant fraction of control dGuo-containing constructs between the two cell lines (1.4 ( 0.7 vs 0.57 ( 0.1, p ) 0.2). O6-EthyldGuo caused a 38-fold increase in the mutant fraction above that of deoxyuridine-containing control construct in 293T cells, while N2ethyldGuo caused a 4-fold increase in the mutant fraction (p ) 0.00008 and p ) 0.01, respectively) (Table 2 and Figure 3). The fold increase in the O6-ethyldGuo- or N2-ethyldGuo-induced mutant fraction in deoxyuridine-containing plasmids was similar after replication in either 293 or 293T cells (p ) 0.2 and 0.8, respectively).

N2-Ethyldeoxyguanosine DNA Adducts in Human Cells

Chem. Res. Toxicol., Vol. 19, No. 7, 2006 963

Figure 2. Yield of progeny plasmids after replication of adduct-containing plasmids in 293 cells either without (white bars) or with (gray bars) deoxyuridines. The total number of colonies per transfection was normalized to the amount of parental DNA (estimated by quantitative PCR), and then, the colonies/ng DNA were normalized to a percentage of control (dGuo) plasmid. Significant p values are indicated by an asterisk (p ) 0.01 for O6-ethyldGuo and p ) 0.0004 for N2-ethyldGuo in plasmids without deoxyuridines and p ) 0.005 for N2-ethyldGuo in plasmids with deoxyuridines). Table 1. Mutant Fractions Associated with Adducts Replicated in Human 293 Cells construct dGuo O6-ethyldGuo N2-ethyldGuo dGuo/dU in complementary strand O6-ethyldGuo/dU in complementary strand N2-ethyldGuo/dU in complementary strand

mutant colonies

wild-type colonies

mutant fraction (%)

11 6 7 17 21 14 14 19 21 14 48 17 503 350 383 70 44 124

1306 1016 841 1058 1022 962 954 905 919 1818 2141 1406 1413 911 1265 1020 627 3215

0.84 0.59 0.83 1.58 2.01 1.43 1.45 2.06 2.23 0.76 2.19 1.19 26.25 27.76 23.24 6.42 6.56 3.71

mean mutant fraction ( SD (%)

fold increase ( SD

p value

0.75 ( 0.1 1.7 ( 0.3

2.4 ( 0.9a

0.008a

1.9 ( 0.3

2.6 ( 0.9a

0.01a

1.4 ( 0.7

0.2a

25.8 ( 2.3

22.2 ( 11.1b

5.6 ( 1.6

4.8 ( 3.1b

0.00006b 0.02b

a As compared to control (dGuo) plasmid. b As compared to control (dGuo) plasmid containing dU on the complementary strand. c “Fold increase” was calculated as the mean and standard deviation of the adduct-mediated mutant fraction divided by the control (dGuo) mutant fraction determined for each individual experiment. d Table 1 shows the total number of colonies while Figure 2 shows number of colonies normalized to the amount of DNA (ng) used per experiment. Each experiment was conducted with newly synthesized constructs.

Mutation Spectrum of Deoxyuridine Containing Control dGuo Constructs. The mutation spectrum observed in deoxyuridine-containing control dGuo constructs was consistent with mutations arising during base excision repair. Nine of 10 mutations determined by sequencing of progeny of deoxyuridine-containing dGuo control constructs in human 293 cells were deletions spanning 5-34 nucleotides (Figure 5). Six of the nine deletions spanned the sequence d(pUUCGAAGU) and an additional two specifically deleted d(pCGAAGU). Additionally, one 12-nucleotide duplication of the sequence d(pAAGGGTTCCGGA) immediately downstream of the BspEI restriction site was observed.

Mutation Spectra Induced by O6-EthyldGuo and N2EthyldGuo Adducts. The mutation spectrum induced by N2ethyldGuo contrasted sharply with that of the O6-ethyldGuo adduct. One hundred percent (9/9) of the sequenced O6ethyldGuo-induced mutants were adduct site-targeted G:C f A:T transitions, which is consistent with what has been reported for this adduct (20-25). The N2-ethyldGuo adduct caused adduct site-targeted single base deletions and transversions, as well as a substantial portion of a single base deletion, -G at d(pGGG) located 3-5 nucleotides downstream of the adduct (Table 3). Ten of 17 N2-ethyldGuo adduct site-targeted mutations were single base deletions, and 7/17 were transversions.

964 Chem. Res. Toxicol., Vol. 19, No. 7, 2006

Upton et al.

Figure 3. Mutant fractions of plasmids containing single dGuo (white bars), O6-ethyldGuo (black bars), or N2-ethyldGuo (gray bars) after replication in 293 and 293T cells. “Double-stranded” and “dU-containing” refer to the type of plasmids used for the transfections.

Figure 4. Western blot analysis of hMLH1 in 293 and 293T cells. The notation at the left indicates the position of the 98 and 62 kDa molecular mass markers and the actin loading controls run on the same gel. The molecular mass of hMLH1 is 85 kDa.

There was no difference in mutation spectra between the 293 and the 293T cells.

Discussion N2-EthyldGuo

Is a Replication Blocking and Mutagenic DNA Adduct in Human Cells. The minimal mutagenicity of N2-ethyldGuo and O6-ethyldGuo observed when these adducts were placed in the supF tRNA mutation reporting gene of the double-stranded shuttle plasmid pLSX suggests that these adducts serve predominantly as replication blocking lesions. This suggestion is supported by previous observations with O6ethyldGuo and other replication blocking adducts. The lack of observed mutagenicity of well-known mutagenic adducts O6ethyldGuo (20-25) and N-acetylaminofluorene (AAF) adducted DNA (26, 27) in assay systems using double-stranded DNA plasmids is well-documented (25, 28) and has been interpreted as reflecting the adduct-induced arrest of replication of the adducted strand. We assert that a similar explanation for the minimal adduct-mediated mutagenicity by N2-ethyldGuo or O6ethyldGuo in the double-stranded plasmid construct replicated in 293 cells is likely. This explanation is supported by the strong blocking effects observed during in vitro DNA synthesis of templates containing these two adducts during DNA synthesis reactions catalyzed by a variety of DNA polymerases (11-13)

and by the markedly diminished yield of progeny plasmids from adduct-containing parent plasmids observed in these experiments. If the yield of progeny plasmid is used as a semiquantitative indicator of a replication block in vivo, these data suggest that template N2-ethyldGuo poses a stronger block to replication than does O6-ethyldGuo. Replacement of thymidines by deoxyuridines that flank the adducted position on the complementary strand enhanced the mutagenic potential of N2-ethyldGuo. The use of deoxyuridinecontaining constructs was based on previous work of Pauly et al. (23), who demonstrated enhanced mutant fractions in constructs containing certain O6-alkyldGuo adducts when such constructs contained lesion-flanking deoxyuridines in the complementary strand. We observed a similar increase in the N2ethyldGuo- and O6-ethyldGuo-induced mutant fraction in deoxyuridine containing plasmids replicated in E. coli (16). Pauly et al. and Prakash et al. inferred that repair of the deoxyuridines by uracil DNA glycosylase and AP endonucleases creates a gap in the strand complementary to the adduct (24, 29). Gap filling forces DNA synthesis across the template adduct. The mutation spectrum observed in the deoxyuridinecontaining control pLSX constructs replicated in 293 cells is consistent with this inference. Six of 10 mutations observed in these constructs were 6-13 nucleotide long deletions that included the deoxyuridine-containing sequence, suggesting the transient presence of a single-stranded intermediate that arose during repair of the deoxyuridines. The exocyclic ethyl adducts N2-ethyldGuo and O6-ethyldGuo differed markedly with regard to both the adduct-induced mutant fractions and the mutation spectra. The principal differences were that O6-ethyldGuo caused a higher mutant fraction and produced transitions, while N2-ethyldGuo induced transversions and a high prevalence of deletions. This difference may in large part be attributed to differences in how these replication blocking adducts are bypassed. While the minor groove N2-ethyldGuo adduct is readily bypassed in an “error-free” manner by the human Y family polymerase η in vitro, the major groove O6ethyldGuo adduct is not (13, 14). This in vitro observation offers a plausible explanation for why the mutant fraction induced by

N2-Ethyldeoxyguanosine DNA Adducts in Human Cells

Chem. Res. Toxicol., Vol. 19, No. 7, 2006 965

Table 2. Mutant Fractions Associated with Adducts Replicated in Human 293T Cells construct dGuo/dU in complementary strand O6-ethyldGuo/dU in complementary strand N2-ethyldGuo/dU in complementary strand a

mutant colonies

wild-type colonies

mutant fraction (%)

7 10 9 451 170 653 26 43 31

1091 1517 2209 1785 765 2271 1268 1260 1528

0.64 0.65 0.41 20.17 18.18 22.33 2.01 3.30 1.99

mean mutant fraction ( SD (%)

fold increase ( SD

p value

20.2 ( 2.1

38.1 ( 14.8a

0.00008a

2.4 ( 0.8

4.4 ( 1.1a

0.57 ( 0.1

0.01a

As compared to control (dGuo) plasmid containing dU on the complementary strand. Each experiment was conducted with newly synthesized constructs.

Figure 5. Mutations observed in control (dGuo), O6-ethyldGuo-, and N2-ethyldGuo-containing constructs containing deoxyuridine at specific sites on the nonadducted strand. The sequence shown includes 17 nucleotides of the supF pre-tRNA and the 33 5′-nucleotides of the supF tRNA. X indicates the adducted deoxyguanosine. The dA positions complementary to deoxyuridines on the nonadducted strand are shown in bold. The XhoI and BspEI restriction sites are underlined. Mutations in control (dGuo) constructs are shown above the pLSX sequence. Deletions are depicted as lines where the line closest to the sequence represents an 88 nucleotide deletion spanning the sequence shown and the line above that represents a 34 nucleotide deletion beginning at the XhoI site and continuing beyond the sequence shown. A 12 nucleotide duplication is shown immediately above the sequence. O6-ethyldGuo-mediated mutations are shown above the sequence. All nine O6-ethyldGuo-mediated mutations were transitions that occurred at the site of the adduct (indicated as A). N2-ethyldGuo-mediated mutations are shown below the sequence. The N2-ethyldGuomediated mutations were point deletions (indicated as -), transversions (indicated as T, C, or G), insertions (indicated as +T), or large deletions (a 10 nucleotide deletion and a 31 nucleotide deletion).

O6-ethyldGuo was higher than that of N2-ethyldGuo in the deoxyuridine-containing constructs. It is possible that when translesion synthesis catalyzed by a replicative DNA polymerase across from template N2-ethyldGuo is blocked, Y family polymerase(s) might accomplish error-free translesion synthesis. In this regard, polymerase η has been shown to catalyze translesion synthesis across N2-ethyldGuo in the presence of polymerase R (13). O6-ethyldGuo, however, may not be readily bypassed by Y family polymerase(s) in vivo and must be copied by more error-prone mechanisms (13). The high prevalence of frameshift deletions induced by N2ethyldGuo is consistent with an adduct-induced replication blockade. A large number of single base deletions has been

observed with other replication blocking DNA adducts, e.g., C8-N-acetylaminofluorene (30). The proposed mechanism for such deletions involves “skipping” over the blocking adduct and base pairing to the 5′-base (31). The single base deletions may also reflect the involvement of Y family polymerases in bypassing template N2-ethyldGuo, some members of which have a tendency to generate frameshift errors (29). Further studies in cell lines that are deficient in various Y family member polymerases are currently underway to characterize their involvement in bypass of N2-ethyldGuo. Interestingly, in deoxyuridine-containing constructs, a -G deletion at the sequence d(pGGG) located three nucleotides downstream of the N2-ethyldGuo adduct was frequently ob-

966 Chem. Res. Toxicol., Vol. 19, No. 7, 2006

Upton et al.

Table 3. Mutation Spectrum of N2-EthyldGuo mutation single base deletion downstream of adduct a -G at site of adduct transversions at the site of the adduct transversion downstream of adduct other total

no. no. total % of 95% confidence (293) (293T) no. total interval 14

2

16

40

24-56

7 6b

3 1c

10 7

25 18

9-41 2-34

1d

1e

2

5