Fidelity of Uracil-initiated Base Excision DNA Repair in DNA
Polymerase 
-Proficient and -Deficient Mouse Embryonic Fibroblast
Cell Extracts*
Samuel E.
Bennett
,
Jung-Suk
Sung, and
Dale W.
Mosbaugh§¶
From the Departments of Environmental and Molecular
Toxicology and § Biochemistry and Biophysics and the
¶ Environmental Health Sciences Center, Oregon State University,
Corvallis, Oregon 97331-7301
Received for publication, July 3, 2001, and in revised form, September 5, 2001
 |
ABSTRACT |
Uracil-initiated base excision DNA repair was
conducted using homozygous mouse embryonic fibroblast DNA polymerase
(+/+) and (
/) cells to determine the error frequency and
mutational specificity associated with the completed repair process.
Form I DNA substrates were constructed with site-specific uracil
residues at U·A, U·G, and U·T targets contained within the
lacZ
gene of M13mp2 DNA. Efficient repair was observed
in both DNA polymerase (+/+) and (/) cell-free extracts. Repair
was largely dependent on uracil-DNA glycosylase activity because
addition of the PBS-2 uracil-DNA glycosylase inhibitor (Ugi) protein
reduced (~88%) the initial rate of repair in both types of cell-free
extracts. In each case, the DNA repair patch size was primarily
distributed between 1 and 8 nucleotides in length with 1 nucleotide
repair patch constituting ~20% of the repair events. Addition of p21 peptide or protein to DNA polymerase (+/+) cell-free extracts increased the frequency of short-patch (1 nucleotide) repair by ~2-fold. The base substitution reversion frequency associated with
uracil-DNA repair of M13mp2op14 (U·T) DNA was determined to be
5.7-7.2 × 10
4 when using DNA polymerase (+/+)
and (/) cell-free extracts. In these two cases, the error frequency
was very similar, but the mutational spectrum was noticeably different.
The presence or absence of Ugi did not dramatically influence either
the error rate or mutational specificity. In contrast, the combination
of Ugi and p21 protein promoted an increase in the mutation frequency associated with repair of M13mp2 (U·G) DNA. Examination of the mutational spectra generated by a forward mutation assay revealed that
errors in DNA repair synthesis occurred predominantly at the position
of the U·G target and frequently involved a 1-base deletion or
incorporation of dTMP.
| |
INTRODUCTION |
Uracil residues accumulate in DNA following incorporation of dUMP
in place of dTMP during DNA synthesis as well as by spontaneous deamination of existing cytosine residues that form premutagenic U·G
mispairs (1-3). In mammalian cells, the presence of uracil in DNA may
have cytotoxic, mutagenic, and lethal consequences (1, 4-6). To avoid
these deleterious effects, the uracil-initiated base excision repair
pathway has evolved to remove uracil residues from the genome (2, 3,
7). The fundamental chemo-mechanical steps of uracil-mediated
BER1 have been described
previously in considerable molecular detail (8, 9).
Several enzymes capable of uracil-excision (UNG, TDG, MED1, and SMUG)
have been identified in mammalian cells (10-13); however, the extent
to which each participates in BER in vivo remains to be
determined. Although these uracil-DNA glycosylases share a common mode
of action for cleavage of the N-glycosylic bond linking the
uracil base to the deoxyribose phosphate DNA backbone, they differ in
substrate specificity. The major uracil-DNA glycosylase activity
detected in mammalian cell extracts is UNG (UNG1 and UNG2) (14-16).
UNG preferentially recognizes uracil residues in single-stranded DNA
and is also active on duplex DNA containing U·G mispairs and U·A
base pairs (17). In contrast, human thymine-DNA glycosylase does not
recognize uracil in single-stranded DNA but does excise uracil residues
in double-stranded DNA with the following specificity: U·G > U·C > U·T 
U·A (18, 19). The human MED1 (also termed
MBD4) enzyme acts on U·G mispairs in the context of methylated or
unmethylated CpG sites, but activity against uracil-containing
single-stranded DNA and U·A, U·C, or U·T double-stranded DNA
substrates was not detected (20). Finally, the single strand-selective monofunctional uracil-DNA glycosylase (SMUG1) removes uracil residues preferentially from single-stranded DNA relative to U·G-containing duplex DNA; however, the catalytic efficiency of xSMUG1 compared with
UNG for both single- and double-stranded DNA was reduced 10- and
70-fold, respectively (13).
The uracil-DNA glycosylases described above can also be differentiated
based on their sensitivity to inhibition by the bacteriophage PBS-1 and
-2 uracil-DNA glycosylase inhibitor (Ugi) protein (1, 13, 21). Ugi acts
as a DNA mimic and binds to UNG, forming a stable 1:1 complex that is
essentially irreversible under physiological conditions (22-25). High
resolution x-ray structures of Ugi in complex with the human UNG
84
(22) or Escherichia coli Ung (23) have revealed an intricate
shape, electrostatic and hydrophopic complementarity between the
interacting proteins. The observation that Ugi inactivates these
biologically distinct enzymes was not surprising because these proteins
share 55.7% identical amino acid residues (23, 26). In contrast, TDG,
MED1, and SMUG, which do not share similar amino acid homology with
UNG, are insensitive to inhibition by Ugi (13, 18, 27, 28). Thus,
addition of Ugi to cell-free extracts has been effectively used to
investigate Ugi-sensitive and Ugi-insensitive modes of uracil-initiated
BER (21, 29).
Following removal of the uracil base and incision of the ensuant
apyrimidinic site, a DNA repair tract is synthesized by DNA polymerase(s) that results in either a short (1 nucleotide) or long (2 or more nucleotides) repair patch (30-33). Short-patch BER of a
uracil-containing duplex oligonucleotide (51-mer) has been
reconstituted in vitro using the following four purified human enzymes: uracil-DNA glycosylase (UNG84), apurinic/apyrimidinic endonuclease (HAP1), DNA polymerase (POL ), and DNA ligase I
(LIG I) (34). Alternatively, DNA ligase III (LIG III) has been
successfully substituted for LIG I in short-patch BER reactions (35).
In the latter case, addition of XRCC1 protein was found to suppress
strand displacement DNA synthesis by POL and promote efficient
ligation after the single nucleotide gap-filling reaction that is
necessary for short-patch BER (35, 36). In BER reactions containing
plasmid DNA bearing a single AP site and extracts of xrcc1 mutant hamster cells, a partial defect in
ligation was detected that disrupted only the short-patch BER pathway
(37). It is important to indicate that short-patch BER has also been
observed in extracts of POL -deficient mouse embryonic fibroblast
cells (31, 38, 39) indicating that one or more other DNA
polymerases, such as POL 
or POL 
, can act in the single
nucleotide gap-filling reaction. Recently, DNA polymerase 
(POL )
was shown to possess an intrinsic 5'-deoxyribose phosphodiesterase
activity similar to that of POL (40). Furthermore, in
vitro reconstitution reactions were conducted that showed POL can substitute for POL and can perform complete BER with UNG, HAP1,
and LIG I using an oligonucleotide (34-mer) substrate containing either
a U·A or U·G target (40). A reconstituted enzyme system has also
been developed for long-patch BER of a reduced AP site located in a defined oligonucleotide (60-mer) utilizing purified HAP1, POL ,
DNase IV (FEN 1), and LIG I (41). In this system, addition of
proliferating cell nuclear antigen (PCNA) was observed to strongly stimulate the repair reaction and promote strand displacement gap-filling DNA synthesis involving two to six nucleotides (41). Furthermore, purified POL was found to substitute for POL , providing PCNA was present in the reaction (41). In extracts of mouse
embryonic fibroblast POL -deficient cells, both short- and
long-patch BER were found to be strictly dependent on PCNA (31).
PCNA-dependent long-patch BER has been also demonstrated in
reconstituted systems, where it was shown to be heavily dependent on
the circularity of the in vitro BER DNA substrate (30,
42-44).
Specific interactions have been detected between PCNA and a number of
proteins involved in DNA repair, such as LIG I (45), FEN1 (46, 47),
UNG2 (48), and p21(Cip1/WAF1) (49). These proteins share a
consensus PCNA-binding motif composed of eight amino acids (50) and
bind to the same region of PCNA (51, 52). p21 is a p53-inducible
bi-modal inhibitor of DNA replication that binds PCNA with 1:1
stoichiometry at the C-terminal domain while maintaining the PCNA
homotrimeric ring structure (49, 52). Thus, the inhibitory effect of
p21 appears to lie in masking the interface required for protein
interactions (52). Intriguingly, a p21 peptide (20 amino acids)
spanning the C-terminal PCNA-binding motif was found to be sufficient
for inhibition of PCNA function in DNA replication and repair (51).
Both the p21 peptide and protein have proved to be useful tools with
which to probe the role of PCNA in DNA repair (53, 54).
Although the biochemical steps of both uracil-initiated short- and
long-patch BER pathways have been studied, the fidelity and mutational
specificity associated with mammalian BER have not been directly
investigated. Whereas in vitro studies to determine the
fidelity of the gap-filling DNA synthesis for various purified mammalian DNA polymerases provide useful information (55-57), an assessment of the mutation frequency associated with completed uracil-initiated BER remains to be fully elucidated. In the present study, we have developed M13mp2 lacZ DNA-based assays to
assess the accuracy of uracil-initiated BER using mouse embryonic
fibroblast cells. We have examined the efficiency, fidelity, mutational
specificity, and patch size of uracil-initiated BER in extracts of
isogenic POL -proficient and POL -deficient cells in the presence
and absence of the uracil-DNA glycosylase inhibitor protein. In
addition, the influence of p21 peptide and protein on the kinetics,
patch size, and fidelity of uracil-initiated BER was investigated.
| |
EXPERIMENTAL PROCEDURES |
Materials--
PCR primers FP-18-mer, RP-18-mer, and DNA
sequencing primer S-21-mer have been described previously (58).
Oligodeoxynucleotides U-23-mer and A-23-mer, used in the construction
of M13mp2op14 (U·T)2
heteroduplex and M13mp2op14 (A·T) homoduplex, have been described (58). Oligodeoxynucleotides CCCAGTCACGUCATTGTAAAACG
(U80-23-mer),3
GGTTTTCCCAGTCUCGTCATTGAAAACG (U83-29-mer), CCAGGGTTTTCCCAGTCACGACG (C89-23-mer), and CCAGGGTTTTCUCAGTCACGACG (U89-23-mer) were obtained from Oligos Etc. and were used to prepare M13mp2op14 (U·A) and (5'-U·T)4 DNA and M13mp2
(C·G) and (U·G) DNA, respectively. Oligodeoxynucleotide preparations were purified by denaturing polyacrylamide gel
electrophoresis and 5'-end phosphorylated as described (29).
E. coli strains NR9162 and CSH50 were obtained from T. A. Kunkel (NIEHS, National Institutes of Health), as was M13mp2 phage. DNA polymerase -proficient (MB16tsA, clone 1B5) or -deficient (MB19tsA, clone 2B2) SV40-immortalized mouse embryonic fibroblast (MEF)
cell lines were obtained from American Type Culture Collection. Purified human DNA polymerase (Mono S fraction) (59) and rabbit anti-DNA polymerase polyclonal antibodies were provided by S. H. Wilson (NIEHS, National Institutes of Health). Sources of purified enzymes (E. coli Ung, Nfo, and Xth; PBS-2 Ugi; T4 DNA
polymerase, T4 polynucleotide kinase, and T4 DNA ligase; restriction
endonucleases EcoRI and SmaI) have been reported
previously (58). The p21 peptide GRKRRQTSMTDFYHSKRRLIFS that
binds PCNA was synthesized by Research Genetics and resuspended in p21
peptide solubilization buffer (10 mM Tris-HCl (pH 8.0), 1 mM EDTA, 100 mM NaCl) as described by Pan
et al. (53). The human C-terminal hexahistidine-tagged p21
expression plasmid pET-Cip1 (49) was provided by J. Massague (Sloan-Kettering Cancer Institute), and p21 protein was purified as
described below.
Purification of p21 Protein--
His-tagged p21 was overproduced
in E. coli BL21-DE3/pET-Cip1 cells (1.5 liters) according to
established protocols (pET Manual, Novagen). Bacterial cells were
pelleted by centrifugation at 4000 × g for 15 min,
resuspended in 40 ml of sonification buffer (50 mM Tris-HCl
(pH 7.5), 1 mM EDTA, 1 mM dithiothreitol)
supplemented with 1 mg/ml chicken egg white lysozyme (Sigma), and
stored at 80 °C. After thawing on ice, cells were subjected to
sonification at 0 °C (Branson 450 Sonifier, 1/2-inch horn, 50% duty
cycle, 20 pulses), and inclusion bodies containing p21 protein were
collected by centrifugation at 12,000 × g for 15 min
(fraction I). The supernatant fraction was discarded, and the pellet
was washed briefly with ice-cold distilled H2O and
centrifuged as before. After discarding the supernatant, the pellet was
resuspended in 10 ml of denaturing buffer (100 mM sodium
phosphate, 10 mM Tris-HCl, 8 M urea (pH 8.0))
and again subjected to centrifugation. The resultant supernatant (fraction II) was applied directly to a
Ni2+-nitrilotriacetic acid-agarose (Qiagen) column (0.79 cm2 × 3 cm) equilibrated in denaturing buffer and washed
with 5 column volumes of denaturing buffer (pH 8.0) followed by 15 column volumes of denaturing buffer (pH 6.3). Finally, the p21 protein
was eluted from the column by the addition of 20 ml of denaturing
buffer (pH 4.5). Fractions (1 ml) were collected, and those enriched for p21 protein as determined by 12% SDS-polyacrylamide gel
electrophoresis (25) were pooled and extensively dialyzed at 4 °C
against AE buffer (2 M urea, 50 mM Tris-HCl (pH
7.5), 1 mM EDTA, 1 mM dithiothreitol, 10%
(w/v) glycerol). The dialysate was applied to a DEAE-Sephadex A-50
(Amersham Pharmacia Biotech) column (1.79 cm2 × 5 cm)
equilibrated in AE buffer, washed with 12 column volumes of
equilibration buffer, and eluted with a linear gradient (40 ml) from 0 to 1 M NaCl in AE buffer. Fractions (2 ml) were collected, and those containing p21 were pooled, dialyzed against p21 storage buffer (50 mM Tris-HCl (pH 7.5), 1 mM EDTA, 1 mM dithiothreitol, 500 mM NaCl, 10% (w/v)
glycerol), concentrated using Centricon 30 concentrators, and stored at
80 °C. Protein concentrations were determined by the Bradford
reaction (60) using the Bio-Rad Protein Assay reagent with BSA as the standard.
Preparation of Base Excision Repair DNA
Substrates--
M13mp2op14 DNA containing an opal codon located at
nucleotide positions 78-80 of lacZ gene was constructed
and isolated as described previously by Sung et al. (58);
M13mp2 DNA was isolated using the same procedure. Briefly, purified
single-stranded M13 DNA was annealed to the appropriate 5'-end
phosphorylated oligodeoxynucleotide, and the primed template DNA was
subjected to a primer extension reaction as described previously (29).
Covalently closed circular duplex DNA reaction products were subjected
to cesium chloride/ethidium bromide gradient equilibrium
centrifugation, and form I DNA was isolated, extracted, concentrated,
and buffer exchanged into TE buffer (10 mM Tris-HCl (pH
8.0), 1 mM EDTA) as described (58). The isolated DNA was
>95% form I as determined by 0.8% agarose gel electrophoresis
(29).
Preparation of Mouse Embryonic Fibroblast Cell-free
Extracts--
MEF cells (POL (+/+)) were grown in DMEM/Ham's F-12
medium (1:1) (Life Technologies, Inc.) supplemented with 10% fetal
bovine serum, 4.8 mg/ml NaHCO3, and 50 units/ml each of
penicillin and streptomycin. Cell cultures were incubated at 37 °C,
5% CO2, and 90% humidity and grown to ~90% confluency
in 175-cm2 tissue culture flasks. Cells were harvested
using trypsin/EDTA (0.25% trypsin, 1 mM EDTA) and
resuspended in phosphate-buffered saline (8.1 mM
Na2HPO4, 1.5 mM
KH2PO4, 2.7 mM KCl, 137 mM NaCl, pH 7.4). Cell-free extracts were prepared as
described by Sanderson and Mosbaugh (29), dialyzed against 25 mM Hepes-KOH (pH 7.9), 100 mM KCl, 12 mM MgCl2, 1 mM EDTA, 2 mM dithiothreitol, and 17% (w/v) glycerol, and
flash-frozen in liquid N2 as small droplets. The protein
concentration of the cell-free extracts (5-10 mg/ml) was determined
using Bio-Rad Protein Assay reagent.
Western Blot Analysis of MEF POL Cell-free
Extracts--
Proteins were resolved by 12.5% SDS-polyacrylamide gel
electrophoresis (25) with a Mini-Protean II apparatus (Bio-Rad), equipped with 0.8-mm spacers, and transferred to a polyvinylidene difluoride membrane (Immobilon-P, Millipore) using a semi-dry transfer
method (Milliblot-Graphite Electroblotter I, Millipore). After transfer
at 2.5 mA/cm2 for 25 min, the membrane was blocked
overnight in BSA buffer (10 mM Hepes-KOH (pH 7.6), 150 mM NaCl, 0.2% BSA, 0.02% Tween 80). Anti-POL serum
was diluted 1:5,000 in BSA buffer and incubated with the membrane
1 h at room temperature. After washing three times with BSA
buffer, alkaline phosphatase secondary antibody (goat anti-rabbit IgG,
Roche Molecular Biochemicals) was diluted 1:2,000 in BSA buffer and
incubated with the membrane for 30 min. The membrane was again washed
and then developed with nitro blue tetrazolium and
5-bromo-4-chloro-3-indoyl phosphate according the manufacturer's
recommendations (Bio-Rad).
Uracil-DNA Glycosylase Inhibitor Assays--
Two assays were
employed to measure Ugi-mediated inhibition of uracil-DNA glycosylase
activity in MEF POL (+/+) cell-free extracts. In the first assay,
reaction mixtures (100 µl) contained M13mp2op14 (U·T)
[32P]DNA (1 µg), 100 mM Tris-HCl (pH 7.5),
5 mM MgCl2, 0.1 mM EDTA, 1 mM dithiothreitol, and Ugi (0-1000 units) as indicated.
Following incubation at 30 °C for 5 min, the reactions were
terminated by heating at 70 °C for 3 min, treated with RNase A and
proteinase K, and the DNA recovered by electroelution as described
below. The DNA from each reaction was resuspended in 20 µl of TE
buffer, and a sample (~100 ng) was subjected to digestion with excess EcoRI and SmaI for 1 h at 25 °C;
reactions were terminated by heating at 70 °C for 10 min. After each
sample was treated with Nfo (1 unit) for 30 min at 37 °C, the
reaction products were resolved by 12% polyacrylamide, 8.3 M urea gel electrophoresis (29). After drying the gel under
vacuum, autoradiography was performed, and the amount of
32P radioactivity associated with the substrate (40-mer)
and product (19-mer) band for each reaction was quantified using a
PhosphorImager. The uracil-DNA glycosylase activity in each reaction
was determined as the quotient of the 32P radioactivity
associated with the product band divided by the sum of the
32P radioactivity associated with both product and
substrate bands. In turn, these values were expressed as a percentage
of the amount of uracil-DNA glycosylase activity observed in the
absence of Ugi. To calculate the degree of inhibition caused by a Ugi
addition, the amount (%) of uracil-DNA glycosylase activity was
subtracted from 100.
In the second uracil-DNA glycosylase inhibitor assay, uracil-DNA
glycosylase activity was quantified by liquid scintillation counting of
free [3H]uracil as excised from
[uracil-3H]calf thymus DNA. Uracil-DNA
glycosylase reaction mixtures (100 µl) contained 70 mM
Hepes-KOH (pH 7.9), 1 mM EDTA, 1 mM
dithiothreitol, 10 µg [uracil-3H]calf thymus
DNA (specific activity 150 cpm/pmol), 50 µg of MEF POL (+/+)
cell-free extract, and Ugi as indicated. Following incubation at
30 °C for 30 min, reactions were terminated on ice by the addition
of 250 µl of 10 mM ammonium formate (pH 4.2). Free
[3H]uracil was resolved from non-hydrolyzed
[uracil-3H]DNA using a Bio-Rad 1-X8 (formate
form) column equilibrated in 10 mM ammonium formate (pH
4.2) as described previously by Bennett and Mosbaugh (24).
Base Excision DNA Repair Reactions--
Standard BER reaction
mixtures (100 µl) contained 100 mM Tris-HCl (pH 7.5), 5 mM MgCl2, 1 mM dithiothreitol, 0.1 mM EDTA, 2 mM ATP, 0.5 mM -NAD,
20 µM each of dATP, dTTP, dGTP, and dCTP, 5 mM phosphocreatine, 200 units/ml phosphocreatine kinase, 10 µg/ml of the appropriate M13mp2 (form I) DNA substrate, and 0.5 mg/ml
cell-free extract. Following incubation at 30 °C, the reactions were
terminated after various times by adjustment to 20 mM EDTA and heated at 70 °C for 3 min. RNase A was then added to 80 µg/ml, and the reaction mixtures were incubated at 37 °C for 10 min. Each
reaction was then adjusted to 0.5% SDS and 190 µg/ml proteinase K,
incubated for 30 min at 37 °C, and the duplex M13mp2 DNA was isolated and resuspended in 20 µl of TE buffer as described by Sanderson and Mosbaugh (29).
Analysis of Base Excision DNA Repair Reaction
Products--
Samples (5 µl) of the M13mp2 DNA isolated from base
excision DNA repair reactions were treated with Ung (100 units) for 30 min at 37 °C to remove uracil residues that had escaped repair during the BER reaction. After quenching the uracil-DNA glycosylase reaction by adding Ugi (1000 units), the samples were further incubated
with Nfo (1 unit) for 30 min at 37 °C to incise AP sites. Nfo was
inactivated by heating at 70 °C for 3 min, and form I DNA that was
resistant to the combined Ung/Nfo treatment (i.e. repaired) was resolved from form II DNA by 0.8% agarose gel
electrophoresis. The amount of form I and form II DNA was determined by
comparing the fluorescent intensity at 300 nm of the ethidium
bromide-stained DNA reaction products to that of co-electrophoresed
form I and form II DNA standards (58, 61). The fluorescent intensity data was captured with a gel documentation system (Ultraviolet Products) and quantified using the ImageQuant computer program (Molecular Dynamics).
Determination of Repair Patch Size--
Standard BER reaction
mixtures were prepared as described except that a
2'-deoxyribonucleoside -thiotriphosphate was used in place of each
of the four 2-deoxyribonucleoside triphosphates, and
32P-labeled M13mp2op14 (U·T) DNA (form I) was used as the
BER substrate. The [32P]DNA substrate was constructed as
described previously (61, 62) and contained a 32P
radiolabel located 13 nucleotides upstream of the target uracil and 7 nucleotides downstream of the SmaI restriction site on the transcribed strand of the lacZ gene sequence. Following
the BER reactions, DNA products were isolated as described above and
resuspended in 20 µl of TE buffer. Samples (4 µl, ~200 ng) were
removed for digestion with 10 units of EcoRI for 1 h at
25 °C. The reaction was terminated at 70 °C for 10 min, and
samples were incubated in the absence or the presence of various
amounts of E. coli exonuclease III (Xth) for 1 h at
37 °C. After incubation, each sample was heated at 70 °C for 10 min, and the [32P]DNA was then cleaved with 10 units of
SmaI for 1 h at 25 °C. An equal volume of denaturing
formamide dye buffer was added, and [32P]DNA products
were resolved by 12% polyacrylamide, 8.3 M urea gel
electrophoresis. After drying the gel under vacuum, autoradiography was
performed, and the amount of 32P radioactivity associated
with each band was quantified using a PhosphorImager and the ImageQuant
computer program.
Isolation of Repaired Form I DNA--
Repaired form I DNA was
isolated from standard BER reaction mixtures following Ung/Nfo
treatment and 0.8% agarose gel electrophoresis, as described above.
Form I DNA was recovered from agarose gel slices by electroelution
using an Elutrap apparatus (Schleicher & Schuell); electroelution was
conducted for 3 h at 150 V in TAE buffer (40 mM Tris
acetate, 1 mM EDTA, pH 8.0). The recovered form I DNA was
concentrated using a Centricon 30 concentrator (Amicon) and
buffer-exchanged into distilled water for further analysis.
Determination of Mutation Frequencies and Mutational
Spectra--
E. coli NR9162 (mutS) cells were
transfected with repaired M13 DNA (form I), combined with E. coli CSH50 (indicator strain) and top agar containing 0.4 mM -D-thiogalactopyranoside and 1 mg/ml
5-bromo-4-chloro-3-indolyl--D-galactopyranoside, and
plated on M9 plates as described (29). After scoring the phage plaques as either colorless or blue, the mutation frequency was calculated as
the ratio of the number of mutant plaques to total plaques detected.
Mutant plaques were blue in reversion assays (M13mp2op14 (A·T),
(U·T), (5'-U·T), and (U·A) DNAs) and white in forward mutation assays (M13mp2 (C·G) and (U·G) DNAs). Secondary screening of each mutant plaque was then conducted in preparation for PCR amplification of the lacZ gene (58) and subsequent nucleotide sequence
analysis that was conducted by the Center for Gene Research and
Biotechnology (Oregon State University).
| |
RESULTS |
Uracil-initiated Base Excision DNA Repair Assay--
Sanderson and
Mosbaugh (29) previously developed an M13mp2op14 lacZ
(U·T) DNA-based reversion assay to detect uracil-initiated BER in
cell-free extracts and determine the base substitution error frequency
associated with the completed repair reaction (Fig.
1, A and B). By
replacing the arginine amino acid codon 14 (CGT) with the opal codon
(TGA), lacZ was rendered incapable of -complementation
(29). Heteroduplex DNA (form I) was synthesized that contained a uracil
residue in the complementary ()-strand at the first position of the
opal codon to create a U·T mispair at position +78; excision of the
target uracil residue served to initiate the BER reaction (Fig.
1B). If BER DNA synthesis was unfaithful and incorporated
either dGMP, dCMP, or dTMP at position 78, reversion of the opal codon
occurred, resulting in restored lacZ complementation and
a blue plaque phenotype. As control, homoduplex DNA containing an A·T
base pair at position +78 was also constructed to evaluate the
background level of non-uracil-mediated reversion of the opal codon in
the cell-free extract (Fig. 1A). By using this assay with
M13mp2op14 lacZ (U·T) DNA and other DNA substrates
(Fig. 1, C-F), the fidelity and mutational specificity of
the uracil-initiated BER pathway was investigated using isogenic DNA
polymerase -proficient (+/+) and DNA polymerase -deficient (/) mouse embryonic fibroblast cells.
View larger version (24K):
[in this window]
[in a new window]
|
Fig. 1.
Substrates for measurement of
uracil-initiated base excision DNA repair fidelity in mammalian
cell-free extracts. M13mp2 lacZ DNA substrates
(A-F) were constructed as described under "Experimental
Procedures." The EcoRI (+59 cleavage site) and
SmaI (+98 cleavage site) restriction endonuclease
recognition sites within the lacZ gene of M13mp2 DNA are
indicated, as is the direction of DNA synthesis (arrow)
during BER. Nucleotide positions 59-95 of the E. coli
lacZ gene are depicted where position +1 corresponds to the
first lacZ transcribed base. The transcribed and template
DNA strands are labeled as () and (+), respectively. Substrates
A-D are based on M13mp2op14 DNA (29), in which positions
78-80 have been changed to an opal codon, and position 98 has been
altered to create a unique SmaI site; substrates
E and F are based on M13mp2 DNA. Substrates
A-F are referred to in the text as M13mp2op14 (A·T),
(U·T), (U·A), (5'-U·T) DNA, and M13mp2 (C·G) and (U·G) DNA
substrates, respectively. The uracil target for initiation of BER is
underlined, and a 20 nt-long horizontal arrow
indicates the direction of uracil mediated DNA repair synthesis; a
line denotes the complementary ()-strand nucleotide
sequence.
|
|
Detection of Uracil-initiated BER in DNA Polymerase -Proficient
Cell Extracts--
To detect uracil-initiated BER, M13mp2op14 (U·T)
DNA (form I) was combined with MEF POL (+/+) cell-free extract, and
the reaction products produced after various times (0-60 min) were resolved by agarose gel electrophoresis (Fig.
2A). In a second BER reaction,
Ugi was added to the reaction mixture to inactivate uracil-DNA
glycosylase activity (Fig. 2B). Following the BER reaction, each sample was treated with excess Ung and Nfo to remove uracil residues that had escaped excision during the BER reaction and to nick
the resultant AP sites, ensuring the conversion of the uracil-DNA (form
I) substrate to form II DNA molecules. In contrast, DNA molecules that
had undergone complete uracil-DNA repair (i.e. uracil excision, AP site incision, BER DNA synthesis, and DNA ligation)
were resistant to Ung/Nfo treatment and migrated as form I DNA. This
step was taken to eliminate unrepaired or incompletely repaired DNA
molecules from further analysis. Inspection of the ethidium
bromide-stained agarose gels showed that repaired form I DNA
accumulated in a time-dependent manner in both BER
reactions (Fig. 2, A and B, lanes
1-6). After 10 min, the percentage of Ung/Nfo-resistant form I
DNA had increased substantially (~35%) in the POL (+/+) reaction
and, by 30 min, had essentially reached a plateau, whereas the
appearance of repaired form I DNA in the BER reaction containing Ugi
was considerably less (~8%) and its accumulation more gradual (Fig.
2C).
View larger version (35K):
[in this window]
[in a new window]
|
Fig. 2.
Detection of uracil-initiated BER in POL
-proficient mouse embryonic fibroblast cell-free
extracts. Standard BER reaction mixtures (100 µl) containing 1 µg of M13mp2op14 (U·T) DNA and 50 µg of MEF POL (+/+)
cell-free extract were incubated at 30 °C for 0, 10, 20, 30, 40, and
60 min (lanes 1-6, respectively) in the absence
(A) or presence (B) of Ugi (1000 units).
Following each reaction, the M13mp2op14 DNA was recovered, subjected to
Ung/Nfo treatment, and analyzed by 0.8% agarose gel electrophoresis as
described under "Experimental Procedures." Untreated and
mock-reacted M13mp2op14 (U·T) DNA (80 ng each) were used as reference
standards (lanes S and C, respectively); only the
latter was Ung/Nfo-treated. C, the fluorescent intensity of
ethidium bromide-stained DNA bands was quantified using the method
described by Sung et al. (58). After adjusting for the
reduced staining intensity of form I DNA relative to form II
(~0.7-fold), the percentage of form I DNA was calculated by dividing
the amount of form I DNA by the sum of form I and form II DNA and
multiplying by 100. The percentage of repaired (form I) DNA in each
sample was plotted as a function of incubation time:  , POL (+/+)
minus Ugi;  , POL (+/+) plus Ugi. D, the effect of
various amounts of Ugi on the uracil-DNA glycosylase activity of MEF
POL (+/+) cell-free extract (50 µg) was measured using calf
thymus [uracil-3H]DNA (24). In the absence of
Ugi, 0.2 units (1 unit of uracil-DNA glycosylase is defined as
the amount that released 1 nmol of uracil/h under standard conditions)
of uracil-DNA glycosylase were detected, which represented 100%
activity. E, Ugi-mediated inhibition of uracil-DNA
glycosylase activity in MEF POL (+/+) cell-free extracts was
measured using M13mp2op14 (U·T) DNA (form I) as described under
"Experimental Procedures." In the absence of Ugi, ~95% of the
uracil residues in M13mp2op14 (U·T) DNA (form I) were converted to AP
sites during a 5-min incubation; this level of uracil-excision
represented 100% activity.
|
|
To establish that the amount of Ugi added to the reaction mixture
was sufficient to inhibit the Ugi-sensitive uracil-DNA glycosylase activity, two different assays were conducted. In the first assay, the
uracil-DNA glycosylase activity in the POL (+/+) extract was
measured in the presence of various amounts of Ugi using a calf thymus
[uracil-3H]DNA substrate, which contained
U·A base pairs (Fig. 2D). The results showed that 20 units
of Ugi was sufficient to inhibit all but ~12% of uracil-DNA
glycosylase activity and that additional Ugi did not produce further
inhibition. In the second assay, uracil-DNA glycosylase activity was
measured using an M13mp2op14 (U·T) DNA substrate. The results
indicated that 20 units of Ugi was sufficient to produce maximal
inhibition of uracil-DNA glycosylase activity (Fig. 2E).
Taken together, these observations confirmed that 1000 units of Ugi was
sufficient to inactivate the Ugi-sensitive uracil-DNA glycosylase
activity present in the MEF POL cell-free extracts. The small
amount of Ugi-resistant uracil-DNA glycosylase activity detected was
attributed to uracil-excision enzymes other than UNG.
Kinetics of Uracil-initiated BER in POL (+/+) and POL (/) Cell Extracts--
Because initial experiments suggested that
uracil-initiated BER in POL (+/+) extracts was in large part
complete by 20 min, the kinetics of repair at shorter times in the BER
reaction was examined in both POL (+/+) and POL (/)
cell-free extracts (Fig. 3). In addition,
experiments were also conducted to determine the effect of Ugi on the
BER repair kinetics. The results indicated that in both cell extracts
the kinetics of uracil-initiated BER were essentially biphasic in the
absence of Ugi. Because the rate and extent of uracil-initiated BER
were remarkably similar in both POL (+/+) and POL (/) cell
extracts, Western blot analysis of both cell extracts was conducted to
ascertain the POLB genotype of the two cell lines (Fig.
3B, inset). Inspection of the Western blot showed that DNA
polymerase protein was detected in the POL (+/+) but not in the
POL (/) cell extract. When Ugi was added to the POL (+/+)
and POL (/) BER reactions, the initial rate of repaired was
reduced by 9.1- and 7.5-fold, respectively (Fig. 3, A and
B).
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 3.
Kinetics of uracil-initiated BER in extracts
of mouse embryonic fibroblast POL -proficient
and POL -deficient cells. Standard BER
reaction mixtures (100 µl) containing 1 µg of M13mp2op14 (U·T)
DNA and 50 µg of (A) POL (+/+) or (B) POL
(/) cell-free extract were incubated at 30 °C for 0, 2, 5, 10, 20, and 30 min in the absence (open symbols) or presence
(closed symbols) of Ugi (1000 units) as described under
"Experimental Procedures." Analysis of the recovered M13mp2op14 DNA
was carried out as described in Fig. 2, and the percentage of repaired
(form I) DNA in each sample was plotted as a function of incubation
time. Western blot analysis of the MEF POL cell-free extracts was
conducted as described under "Experimental Procedures," and the
results are shown in the inset of B. Samples
contained 20 ng of purified human POL (lane 1) or 50 µg of MEF POL (+/+) (lane 2) or MEF POL (/)
(lane 3) cell-free extracts. An arrow
indicates the position of the band corresponding to purified POL
.
|
|
Patch Size Associated with Uracil-initiated BER--
The patch
size of DNA repair synthesis associated with uracil-initiated BER of
M13mp2op14 (U·T) [32P]DNA in MEF POL (+/+) and POL
(/) cell-free extracts was determined using the method
developed by Huang et al. (63) and modified by Sung et
al. (58). Briefly, this approach relies on the incorporation of
2'-deoxyoligonucleoside -thiol monophosphates during the BER
reaction to render the repaired DNA strand resistant to in
vitro digestion by E. coli exonuclease III (Xth). As
illustrated in Fig. 4, the uracil target
residue was located on the ()-strand 20 nucleotides 5' and 19 nucleotides 3' to the unique EcoRI and SmaI
restriction endonuclease recognition sites, respectively. A
site-specific [32P]dCMP residue was introduced 5' to the
uracil target at position 90 as described previously (58). Two
reference standards were generated to assist in determining the BER
patch size. First, a 32P-labeled 40-mer standard was
produced by treating the M13mp2op14 (U·T) [32P]DNA
substrate with EcoRI and SmaI and resolving the
reaction products by denaturing gel electrophoresis (Fig.
4B, lane 1). Second, a
32P-labeled 19-mer was generated after the uracil target,
and the 32P-labeled 40-mer was excised by Ung, and the
ensuing AP site was cleaved by Nfo (Fig. 4B, lane
2). The 19-mer standard corresponded to the BER intermediate prior
to DNA synthesis and defined the 5'-boundary of the repair patch. To
determine the 3'-boundary of a repair patch, M13mp2op14 (U·T)
[32P]DNA recovered from BER reactions was first
linearized with EcoRI and then digested in the 3'- to
5'-direction with excess Xth. Because Xth-mediated DNA degradation of
the ()-strand was expected to terminate upon encountering the first
phosphorothioate linkage (64), this 3'-terminal 2'-deoxyribonucleoside
-thiol monophosphate residue defined the 3'-boundary of the repair
patch. Subsequent cleavage with SmaI produced a
[32P]DNA fragment, the length of which indicated the BER
patch size (Fig. 4A).
View larger version (29K):
[in this window]
[in a new window]
|
Fig. 4.
Analysis of the DNA repair patch size
associated with uracil-mediated BER. A, scheme for
detecting uracil-initiated BER DNA synthesis patch size. A segment of
the M13mp2op14 (U·T) [32P]DNA is shown that contains
the EcoRI and SmaI restriction endonuclease
sites. The location of the [32P]dCMP residue at
nucleotide (nt) position 90 on the ()-strand is indicated
by an asterisk. Uracil-mediated BER DNA synthesis resulting
in 2'-deoxyribonucleoside -thiol monophosphate incorporation is
represented by  for patch sizes of 1-4 nucleotides. Following
EcoRI, Xth, and SmaI treatments, the BER patch
sizes of 1-4 nucleotides in length correspond to
[32P]DNA fragments of 20-23 nucleotides, respectively.
B, standard BER reaction mixtures (100 µl) containing 1 µg of M13mp2op14 (U·T) [32P]DNA, 20 µM
each of dATP[S], dTTP[S], dGTP[S], and
dCTP[S] and 50 µg of either MEF POL (+/+) or POL (/)
cell-free extract were incubated for 60 min at 30 °C in presence (+)
or absence () of Ugi. Following the reaction, DNA products were
isolated, and samples (~100 ng) were digested with EcoRI
and subsequently incubated with 0 (lanes 3, 6, 9, 12, and 15), 2 (lanes 4, 7, 10, 13, and 16), and 20 (lanes 5, 8, 11, 14, and 17) units of E. coli exonuclease
III. After exonuclease III digestion, the DNA was cleaved with excess
SmaI, and the 32P-labeled DNA fragments were
resolved by 12% polyacrylamide, 8.3 M urea gel
electrophoresis as described under "Experimental Procedures." The
[32P]DNA size markers, 40-nucleotide (lane 1),
generated by digesting 200 ng of M13mp2op14 (U·T)
[32P]DNA with EcoRI and SmaI, and
19-nucleotide (lane 2), produced by treatment of the 40-mer
with Ung and Nfo, are indicated by arrows. C, the amount of
32P radioactivity detected in each band in B was
quantitatively measured using a Phos- phorImager. The relative amount of 32P label in each
band was determined by dividing the amount of 32P
radioactivity detected per band by the total 32P signal
detected for all bands in the particular lane and multiplying by 100. The frequency with which a specific BER patch size (i.e. one
nucleotide) occurred was expressed as the percent distribution of that
patch size. The results for reactions digested with 20 units of
E. coli exonuclease III are shown as follows: POL (+/+)
without Ugi (black bars), POL (+/+) with Ugi
(white bars), POL (/) without Ugi (horizontal
stripes), POL (/) with Ugi (diagonal stripes).
Mean values and standard deviations of the patch size distributions
from four experiments are indicated.
|
|
Patch size analysis was conducted following uracil-initiated BER in POL
(+/+) and POL (/) cell-free extracts both in the presence
and absence of Ugi (Fig. 4B). As a control, mock-treated M13mp2op14 (U·T) [32P]DNA was examined. For each of the
five reaction sets, the EcoRI-SmaI 32P-labeled 40-mer restriction fragment, not treated with
Xth, is shown first (Fig. 4B, lanes 3, 6, 9, 12, and 15), followed by digestion of the 40-mer fragment with 2 (Fig. 4B, lanes 4, 7, 10, 13, and 16)
or 20 units (Fig. 4B, lanes 5, 8, 11, 14, and 17) of Xth. As expected, a repair patch was not detected for
the mock-treated [32P]DNA since 2'-deoxyribonucleoside
-thiol monophosphates were not incorporated into the
[32P]DNA substrate in the absence of cell-free extract
(Fig. 4B, lanes 3-5). When treated similarly,
each of the other four reactions yielded discrete
[32P]DNA fragments, following digestion with Xth, which
were used to define the 5'-boundary of BER DNA synthesis. Quantitative
determination of the BER repair patch size was made using a
PhosphorImager, and the patch size distribution for each reaction was
expressed as a percentage of the total BER detected (Fig.
4C). The results showed that one-nucleotide repair patches
were the single most prevalent patch size in both POL (+/+) and POL
(/) BER reactions, constituting ~23 and ~18% of total
repair patches, respectively. Greater than 50% of the repair patches
produced in both POL (+/+) and POL (/) BER reactions were 2 to 8 nucleotides in length. The results also illustrated that Ugi
addition did not appreciably alter the distribution of patch size in
BER reactions containing either cell-free extract.
Effect of p21 on Uracil-initiated BER Patch Size--
Repair patch
size experiments were conducted to examine the influence of the
PCNA-binding p21 peptide and protein on uracil-initiated BER of
M13mp2op14 (U·T) [32P]DNA in POL (+/+) cell-free
extracts (Fig. 5). As before, the two
reference standards, 32P-labeled 40-mer (Fig. 5,
lanes 1 and 7) and 32P-labeled 19-mer
(Fig. 5, lanes 2 and 8), were produced as markers and resolved by electrophoresis. An examination of the BER reaction products revealed that the p21 buffer had no apparent influence on
patch size (Fig. 5A, lanes 3 versus
4 and 9 versus 10). In contrast, addition of p21 peptide (8 µM) or p21 protein
(0.7 µM) provoked a readily discernible increase in the
one-nucleotide repair patch and a corresponding decrease in longer
repair patches (Fig. 5A, lanes 5 and
11, respectively). Supplementation of the BER reactions with
the higher concentration of p21 peptide (32 µM) or
protein (2.8 µM) resulted in severe curtailment of longer BER repair patches (Fig. 5A, lanes 6 and
12, respectively). The results were quantitatively analyzed
and plotted for the p21 peptide and protein BER reactions as shown in
Fig. 5, B and C, respectively. Addition of p21
peptide (8 µM) increased the percentage of the one-nucleotide repair patch ~2-fold relative to the buffer control (52 versus 27%), and p21 protein (0.7 µM)
promoted a similar ~2.1-fold change (62 versus 29%).
Longer patch sizes (2-8 nucleotides) decreased 36% upon addition of
p21 peptide and 46% with p21 protein supplementation. Overall, the
amount of BER DNA synthesis was reduced 69 and 72% relative to
controls in the presence of p21 peptide (Fig. 5A, lane
6) and p21 protein (Fig. 5A, lane 12),
respectively. These results suggested that PCNA plays a significant
role in determining the patch size associated with uracil-initiated
BER.
View larger version (35K):
[in this window]
[in a new window]
|
Fig. 5.
Effect of p21 peptide and protein on the DNA
repair patch size associated with uracil-mediated BER. Standard
BER reaction mixtures (100 µl) containing 1 µg of M13mp2op14
(U·T) [32P]DNA, 20 µM each of
dATP[S], dTTP[S], dGTP[S], and dCTP[S], and
50 µg of MEF POL (+/+) cell-free extract were incubated for 60 min at 30 °C. The following additions (2.5 µl) were included in
the reaction mixtures: standard BER reaction buffer (lanes 3 and 9); p21 peptide solubilization buffer (lane
4); p21 storage buffer (lane 10); p21 peptide
(lanes 5 and 6, 8 and 32 µM,
respectively) or p21 protein (lanes 11 and 12, 0.7 and 2.8 µM, respectively). After terminating the BER
reactions, the [32P]DNA products were isolated, digested
with EcoRI as described in Fig. 4, and subsequently
incubated with 20 units of E. coli exonuclease III.
Following exonuclease III digestion, the [32P]DNA was
cleaved with SmaI, and the reaction products were resolved
by 12% polyacrylamide, 8.3 M urea gel electrophoresis as
described under "Experimental Procedures." The DNA size markers,
40-mer (lanes 1 and 7) and 19-mer (lanes
2 and 8), are indicated by arrows. The
amount of 32P radioactivity detected in BER reactions
containing p21 peptide (B) or p21 protein (C) was
quantitatively measured using a PhosphorImager, and the relative amount
of 32P label in each band was determined as described in
Fig. 4. The frequency with which a specific patch size occurred is
expressed as the percent distribution of that patch size, and the
height of the bars in the histogram correspond to the observed
frequencies. The four BER reaction conditions are represented by bars
as follows: standard BER reaction buffer, white; p21 storage
buffer, dotted; lower p21/peptide concentration,
diagonal stripes; higher p21/peptide concentration,
black.
|
|
Error Frequency and Mutational Specificity Associated with BER of
M13mp2op14 (U·T) DNA--
The frequency of mutations produced during
uracil-initiated BER of M13mp2op14 (U·T) DNA was determined using the
M13mp2op14 lacZ-based reversion assay. First, the
background reversion frequency of M13mp2op14 (A·T) DNA was measured
in extracts of MEF POL (+/+) and POL (/) cells (Table
I). A similar reversion frequency (~0.1 × 104) was observed for both cell extracts.
Second, the reversion frequency associated with BER of M13mp2op14
(U·T) DNA was determined to be 7.18 × 104 and
5.71 × 104 for reactions conducted with POL (+/+) and POL (/) cell-free extracts, respectively. Third, Ugi
was added to reaction mixtures, and the reversion frequencies
associated with the BER reactions were found to be 6.56 × 104 for POL (+/+) and 7.19 × 104
for POL (/) cell-free extracts (Table I). Thus, the error frequency of uracil-initiated BER was essentially unaffected by the
addition of Ugi and did not appear to correlate with the
POLB gene status.
View this table:
[in this window]
[in a new window]
|
Table I
Frequency of mutations produced by BER of U · T-containing
M13mp2op14 DNA in mouse embryonic fibroblast DNA polymerase (+/+)
and (/) cell-free extracts
Standard BER reaction mixtures (100 µl) were prepared that contained
50 µg of MEF pol (+/+) or pol (/) cell-free extract, 1 µg of M13mp2op14 (A · T) or (U · T) DNA, and Ugi (1000 units) as indicated. After incubation at 30 °C for 60 min, the
reactions were terminated and DNA products recovered, and form I DNA
resistant to Ung/Nfo treatment was isolated by 0.8% agarose gel
electrophoresis as described under "Experimental Procedures."
E. coli NR9162 cells were then transfected with the form I
DNA, and the M13mp2 lacZ DNA-based reversion assay was
performed as described under "Experimental Procedures."
|
|
The mutational spectrum of M13mp2op14 (A·T) DNA was determined to
define the nature and distribution of background base substitutions induced by incubation with POL (+/+) and POL (/) cell-free extracts (Fig. 6 A and
D, respectively). Examination of the mutational spectrum
produced in POL (+/+) reactions showed that ~60% of the
reversion mutations occurred at the third position (A) of the opal
(TGA) codon (Fig. 6A), whereas only ~35% of the total reversions occurred at the first position (T) of the opal codon. The
spectrum observed in POL (/) cell-free extracts revealed that
~80% of all reversions occurred at the third position of the opal
codon and ~15% resided at the first position (Fig. 6D). Taken together, these results indicate a bias toward third position mutations in the background error spectrum.
View larger version (32K):
[in this window]
[in a new window]
|
Fig. 6.
Specificity of mutations produced during
uracil-initiated BER of M13mp2op14 (U·T) DNA. Six standard BER
reactions were conducted as described in Table I using 50 µg of MEF
POL -proficient (A-C) or POL -deficient
(D-F) cell-free extracts and M13mp2op14 DNA as shown in
each panel with the target uracil residue underlined in the
opal codon 14. Two reactions contained M13mp2op14 (A·T) DNA
(A and D), whereas the other four reactions were
prepared with M13mp2op14 (U·T) DNA (B, C,
E, and F); Ugi (1000 units) was added to two
reactions (C and F). Following the BER reactions,
the M13mp2op14 DNA was recovered, treated with excess Ung and Nfo, and
used to conduct the M13mp2 lacZ DNA-based reversion assay
as described under "Experimental Procedures." Revertant (blue and
light blue) plaques were isolated and subjected to PCR-mediated
lacZ DNA amplification, and DNA sequence analysis was
conducted as described under "Experimental Procedures." The
nucleotide sequence of the opal codon (TGA) used as the template for
uracil-initiated BER DNA synthesis is shown, as are the four
deoxyribonucleoside triphosphates available for incorporation into the
primer strand during BER; the corresponding amino acid encoded by each
single deoxyribonucleoside triphosphate incorporation event is
indicated within parentheses. For each BER reaction, the
number of base substitutions detected by DNA sequence analysis at each
nucleotide position was divided by the total number of mutants
sequenced and plotted as a percentage. In BER reactions containing POL
(+/+) cell-free extract (A-C), the number of mutants
sequenced was 20, 30, and 30, respectively. In BER reactions containing
POL (/) cell-free extract (D-F), the number of
mutants sequenced was 20, 31, and 30, respectively.
|
|
The mutational spectrum of uracil-initiated BER of M13mp2op14 (U·T)
DNA in reactions containing POL (+/+) cell extract is presented in
Fig. 6B. A large majority (87%) of reversions was identified at the first position of the opal codon where the uracil target was located. The most common mutation was T to G (37%), followed by T to C (30%) and T to A (20%). In reactions containing POL (/) cell-free extract, 90% of all base substitutions
occurred at the first position, where T to G transversions also
prevailed (Fig. 6E). Supplementation of BER reactions with
Ugi had a modest effect on the specificity on mutations in POL (+/+) reactions (Fig. 6C). However, in reactions containing
POL (/) cell-free extract, the propensity for forming T·C
mispairs increased to 83% of mutations sequenced (Fig.
6F).
Error Frequency and Mutational Specificity Associated with BER of
M13mp2op14 (U·A) DNA--
Mutational analysis was repeated using a
uracil target contained within a U·A base pair of the opal codon 14 (Fig. 1C). The frequency of mutations produced during
uracil-initiated BER of M13mp2op14 (U·A) DNA was found to be
6.75 ± 2.06 × 104 and 2.60 ± 0.91 × 104 for reactions conducted with POL (+/+) and POL
(/) cell-free extracts, respectively (Table
II). When Ugi was added to the BER reactions, the error frequencies associated with POL (+/+) and POL
(/) cell-free extracts were 2.64 ± 1.23 × 104 and 5.02 ± 2.02 × 104
(Table II). Comparison of the mutation frequencies obtained using the
U·A target, located at the third position of the opal codon 14, with
the U·T target situated at the first position (Table I), showed that
the frequency of reversion did not vary significantly for BER reactions
containing POL (+/+) cell extract. However, the same comparison for
BER reactions containing POL (/) cell extracts indicated that
U·A repair was somewhat less mutagenic than U·T repair.
View this table:
[in this window]
[in a new window]
|
Table II
Frequency of mutations produced by BER of U · A- or 5'-U
· T-containing M13mp2op14 DNA in mouse embryonic fibroblast DNA
polymerase (+/+) and (/) cell-free extracts
Standard BER reaction mixtures (100 µl) were prepared that contained
50 µg of MEF pol (+/+) or pol (/) cell-free extract, 1 µg of M13mp2op14 (U · A) or (5'-U · T) DNA, and Ugi
(1000 units) as indicated. After incubation at 30 °C for 60 min, the
reactions were terminated and DNA products recovered, and form I DNA
resistant to Ung/Nfo treatment was isolated by 0.8% agarose gel
electrophoresis as described under "Experimental Procedures."
E. coli NR9162 cells were then transfected with the form I
DNA, and the M13mp2 lacZ DNA-based reversion assay was
performed as described under "Experimental Procedures."
|
|
The mutational spectrum of uracil-initiated BER of M13mp2op14 (U·A)
DNA in reactions containing POL (+/+) cell extract is presented in
Fig. 7A. The dominant base
substitution mutation (77%) was insertion of dCMP opposite A, where
the uracil residue was located. Addition of Ugi to the BER reaction did
not appear to alter the mutational specificity (Fig.
7B).
View larger version (33K):
[in this window]
[in a new window]
|
Fig. 7.
Specificity of mutations produced during
uracil-initiated BER of M13mp2op14 (U·A) DNA. Four standard BER
reactions were conducted as described in Table II using 50 µg of MEF
POL -proficient (A and B) or POL -deficient
(C and D) cell-free extracts and M13mp2op14
(U·A) DNA as shown in each panel; Ugi (1000 units) was
added to two reactions (B and D). The M13mp2op14
(U·A) DNA was recovered from each BER reaction, processed, and
analyzed as described in Fig. 6. Revertant (blue and light blue)
plaques were isolated and subjected to PCR-mediated lacZ
DNA amplification, and DNA sequence analysis was conducted as described
under "Experimental Procedures." For each BER reaction, the number
of base substitutions detected by DNA sequence analysis at the
positions of the opal codon was divided by the total number of mutants
sequenced and plotted as a percentage. The number of mutant sequences
determined for A-D was 30, 29, 30, and 30, respectively.
|
|
An examination of the mutational spectrum of uracil-initiated BER on
the same DNA substrate but containing POL (/) cell extract is
shown in Fig. 7C. In this spectrum, insertion of dGMP opposite A competed equally with dCMP insertion (33%). The addition of
Ugi to the POL (/) BER reactions altered the specificity of
opal codon 14 reversion such that 93% (28 of 30) of reversions was the
result of dCMP insertion opposite A at the third position.
Error Frequency and Mutational Specificity Associated with BER of
M13mp2op14 (5'-U·T) DNA--
Determination of the error frequency of
uracil-initiated BER on the 3'-side of the uracil excision site was
made using M13mp2op14 (5'-U·T) DNA (Fig. 1D). In this
substrate, the uracil target (U·T) was situated three nucleotides on
the 5'-side of the last base of the TGA opal codon. Following excision
of the "upstream" uracil residue, DNA synthesis must extend 6 nucleotides "downstream" to traverse the opal codon where base
substitution errors could result in reversion mutations. The mean
reversion frequency of M13mp2op14 (5'-U·T) DNA recovered from BER
reactions containing extracts of POL (/) cells was 0.39 ± 0.04 × 104 and 0.32 ± 0.1 × 104 in the presence and absence of Ugi, respectively
(Table II). These values were only slightly elevated over the
background observed for M13mp2op14 (A·T) DNA (Table I), indicating
that the fidelity of downstream BER DNA synthesis was relatively
high. However, the mutational specificity associated with the repaired
(5'-U·T) DNA differed from the reversion spectrum of the (A·T) DNA.
Reversion at opal codon 14 in repaired (5'-U·T) DNA occurred more
frequently at the first position (57%) than at the third position
(40%) (Fig. 8A), whereas
reversion in (A·T) DNA occurred predominantly at the third position
(80%) relative to the first position (20%) of opal codon 14 (Fig.
6D). In addition, the specificity of BER DNA synthesis
errors contrast with that obtained using the (U·T) substrate (Fig. 6,
E and F), where errors introduced at the site of
uracil excision consisted predominantly of incorporation of dCMP
opposite T and not dGMP opposite T, as observed in Fig. 8, A
and B.
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 8.
Specificity of mutations produced during
uracil-initiated BER of M13mp2op14 (5'-U·T) DNA. Two standard
BER reactions were conducted as described in Table II using 50 µg of
MEF POL -deficient cel |
|
TOP
ABSTRACT
INTRODUCTION
