DNA Repair 1 (2002) 559–569
Activities of human DNA polymerase ␬ in response to the major
benzo[a]pyrene DNA adduct: error-free lesion bypass and
extension synthesis from opposite the lesion
Yanbin Zhang a , Xiaohua Wu a , Dongyu Guo a , Olga Rechkoblit b , Zhigang Wang a,∗
a
Graduate Center for Toxicology, University of Kentucky, Lexington, KY 40536, USA
b Chemistry Department, New York University, New York, NY 10003, USA
Received 6 March 2002; accepted 26 April 2002
Abstract
In cells, the major benzo[a]pyrene DNA adduct is the highly mutagenic (+)-trans-anti-BPDE-N2 -dG. In eukaryotes, little
is known about lesion bypass of this DNA adduct during replication. Here, we show that purified human Pol␬ can effectively
bypass a template (+)-trans-anti-BPDE-N2 -dG adduct in an error-free manner. Kinetic parameters indicate that Pol␬ bypass
of the (−)-trans-anti-BPDE-N2 -dG adduct was ∼41-fold more efficient compared to the (+)-trans-anti-BPDE-N2 -dG adduct.
Furthermore, we have found another activity of human Pol␬ in response to the (+)- and (−)-trans-anti-BPDE-N2 -dG adducts:
extension synthesis from mispaired primer 3 ends opposite the lesion. In contrast, the two adducts strongly blocked DNA
synthesis by the purified human Pol␤ and the purified catalytic subunits of yeast Pol␣, Pol␦, and Polε right before the lesion.
Extension by human Pol␬ from the primer 3 G opposite the (+)- and (−)-trans-anti-BPDE-N2 -dG adducts was mediated by
a −1 deletion mechanism, probably resulting from re-aligning the primer G to pair with the next template C by Pol␬ prior
to DNA synthesis. Thus, sequence contexts 5 to the lesion strongly affect the fidelity and mechanism of the Pol␬-catalyzed
extension synthesis. These results support a dual-function model of human Pol␬ in bypass of BPDE DNA adducts: it may
function both as an error-free bypass polymerase alone and an extension synthesis polymerase in combination with another
polymerase. © 2002 Elsevier Science B.V. All rights reserved.
Keywords: Lesion bypass; Translesion synthesis; Polymerase ␬; Mutagenesis; Benzo[a]pyrene; DNA adducts
1. Introduction
Benzo[a]pyrene is a common environmental carcinogen produced by incomplete combustion of organic
materials. Racemic anti-benzo[a]pyrene-7,8-dihydrodiol-9,10-epoxide (anti-BPDE) are two metabolites
of benzo[a]pyrene, which react with DNA mainly at
the N2 position of guanine, forming stereoisomeric
∗ Corresponding author. Tel.: +1-859-323-5784;
fax: +1-859-323-1059.
E-mail address: [email protected] (Z. Wang).
bulky adducts (+)-trans-anti-BPDE-N2 -dG, (+)-cisanti-BPDE-N2 -dG, (−)-trans-anti-BPDE-N2 -dG, and
(−)-cis-anti-BPDE-N2 -dG [1,2]. In cells, the major
benzo[a]pyrene adduct is (+)-trans-anti-BPDE-N2 -dG
[2]. This DNA lesion is highly mutagenic in simian
COS cells, producing mostly G → T transversions
[3,4] in a defined sequence context. The mutagenic
potential of (+)-trans-anti-BPDE-N2 -dG in different
sequence contexts has been investigated [3–8].
Lesion bypass is an important cellular mechanism
in response to benzo[a]pyrene DNA adducts, which
directly copies the damaged DNA template during
1568-7864/02/$ – see front matter © 2002 Elsevier Science B.V. All rights reserved.
PII: S 1 5 6 8 - 7 8 6 4 ( 0 2 ) 0 0 0 5 5 - 1
560
Y. Zhang et al. / DNA Repair 1 (2002) 559–569
replication. Conceptually, actions of DNA polymerase
during lesion bypass can be separated into two distinctive steps: (a) nucleotide insertion (incorporation)
opposite the lesion; and (b) extension DNA synthesis
from opposite the lesion. Lesion bypass can be errorfree or error-prone. Error-free bypass predominantly
incorporates the correct nucleotide opposite the lesion;
whereas error-prone bypass frequently incorporates
an incorrect nucleotide opposite the lesion. Errorfree bypass of benzo[a]pyrene DNA adducts would
suppress the mutagenic and carcinogenic effects of
benzo[a]pyrene. In contrast, error-prone bypass of
the adducts would constitute a major mechanism
of induced mutagenesis, which likely plays a key role
in carcinogenesis of benzo[a]pyrene.
Recent studies indicate that the Y family DNA
polymerases are involved in lesion bypass [9]. Human
Pol␬ and Pol␩ are two members of the Y family polymerases [10]. In vitro, purified human Pol␬ is able to
perform error-prone translesion synthesis opposite a
template AP site and AAF-adducted guanine [11–14].
Remarkably, purified human Pol␬ efficiently bypasses
a template (−)-trans-anti-BPDE-N2 -dG adduct in
vitro in an error-free manner [11]. In contrast, purified
human Pol␩ performs error-prone nucleotide insertion opposite a template (+)-trans-anti-BPDE-N2 -dG
adduct [15]. Moreover, following nucleotide insertion
opposite the (+)-trans-anti-BPDE-N2 -dG adduct, further extension synthesis by human Pol␩ is greatly
inhibited by the lesion [15]. Thus, if Pol␩ indeed contributes to bypass of the (+)-trans-anti-BPDE-N2 -dG
adduct in cells, another DNA polymerase would be
needed to perform extension synthesis.
The response of human Pol␬ to the major benzo[a]
pyrene DNA adduct, (+)-trans-anti-BPDE-N2 -dG, is
unknown. Little is known about the lesion bypass enzymology of this important DNA adduct in eukaryotes.
To help identify the polymerases involved in bypass of
the major benzo[a]pyrene DNA adduct, we have analyzed biochemical activities of purified human Pol␬ toward a template (+)-trans-anti-BPDE-N2 -dG adduct.
In this report, we show two activities of human Pol␬
in response to a template (+)-trans-anti-BPDE-N2 -dG
adduct: error-free lesion bypass and extension DNA
synthesis from mispaired primer 3 ends opposite
the lesion. Our results suggest that Pol␬ plays an
important role in error-free bypass of the (+)- and
(−)-trans-anti-BPDE-N2 -dG DNA adducts, and that
Pol␬ may also function as an extension DNA synthesis polymerase following nucleotide insertion by
another polymerase opposite these lesions.
2. Materials and methods
2.1. Materials
A mouse monoclonal antibody against the His6 tag
was obtained from Qiagen (Valencia, CA). Alkaline
phosphatase conjugated anti-mouse IgG was obtained
from Sigma Chemical Co. (St. Louis, MO). Human
Pol␬ was purified to near homogeneity as previously
described [11]. A 33-mer DNA template, 5 -CTCGATCGCTAACGCTACCATCCGAATTCGCCC-3,containing a site-specific (+)-trans-anti-BPDE-N2 -dG or
(−)-trans-anti-BPDE-N2 -dG adduct at the N2 position of the underlined G was prepared as previously
described [16–18].
2.2. Purification of DNA polymerases
The catalytic subunit of yeast Pol␣ Pol␦, and
Polε were expressed in yeast cells from plasmids
pEGUh6-POL1, pECUh6-POL3, and pEGUh6-POL2,
respectively. Expression of yeast Pol␦ was induced by
adding CuSO4 to 0.3 mM in the growth media as previously described [11]. Expression of yeast Pol␣ and
Polε was induced by adding galactose to 2% in the
growth media as previously described [19]. Yeast cells
(∼100 g) were homogenized by zirconium beads in a
bead-beater in an extraction buffer containing 50 mM
Tris–HCl, pH 7.5, 1 M KCl, 5 mM ␤-mercaptoethanol,
10% sucrose, and protease inhibitors. The clarified
extract (∼120 ml) was loaded onto two connected HiTrap chelating columns charged with NiSO4 (Amersham Pharmacia Biotech, 2 ml × 5 ml), followed by
washing the column sequentially with 100 ml of Ni
Buffer A (20 mM KH2 PO4 , pH 7.4, 1 M NaCl, 10%
glycerol, 5 mM ␤-mercaptoethanol, and protease inhibitors) containing 10 mM imidazole and 100 ml of
Ni buffer A containing 35 mM imidazole. Bound proteins were eluted with a linear gradient of 35–108 mM
imidazole. The His6 -tagged protein was identified
by western blot analyses using a mouse monoclonal
antibody specific to the His6 tag. The pooled sample
was concentrated by polyethylene glycol 10,000 and
Y. Zhang et al. / DNA Repair 1 (2002) 559–569
then desalted through five connected Sephadex G-25
columns (Amersham Pharmacia Biotech, 5 ml × 5 ml)
in FPLC buffer A (50 mM Tris–HCl, pH 7.5, 1 mM
EDTA, 10% glycerol, and 5 mM ␤-mercaptoethanol)
containing 50 mM KCl. The resulting sample was
loaded onto an FPLC Mono S HR5/5 column (Amersham Pharmacia Biotech) and eluted with a 30 ml
linear gradient of 50–500 mM KCl in FPLC buffer
A. For Polε, the Ni-column sample was desalted to
30 mM KCl and an FPLC Resource S column (Amersham Pharmacia Biotech) was used in place of Mono
S. Pol␣ and Pol␦ were further purified as the following. The Mono S fractions were concentrated by PEG
10,000 and resolved on an FPLC Superdex 200 gel
filtration column that had been equilibrated in FPLC
buffer A containing 300 mM KCl.
561
or a (−)-trans-anti-BPDE-N2 -dG adduct, we performed in vitro DNA syntheses with purified polymerases. A 17-mer primer was labeled with 32 P at its 5
end and annealed with the damaged DNA template two
nucleotides before the BPDE adduct (Fig. 1). DNA
polymerase assays were then performed. As shown in
Fig. 1 (lanes 2, 3, 5, 6, 8 and 9), both (+)-trans-antiBPDE-N2 -dG and (−)-trans-anti-BPDE-N2 -dG adducts completely blocked the catalytic subunits of yeast
DNA polymerases ␣, ε, and ␦ right before the lesion.
Purified human Pol␤ was also completely blocked
by the (+)- and (−)-trans-anti-BPDE-N2 -dG DNA
2.3. DNA synthesis assays
Standard DNA synthesis assays were performed
at 30 ◦ C for 10 min as described before [11], using
50 fmol of an indicated DNA substrate containing a
32 P-labeled primer and a purified DNA polymerase.
The reaction products were resolved on a 20% polyacrylamide gel containing 8 M urea and visualized by
autoradiography.
2.4. Kinetic analysis
Kinetic analysis of nucleotide incorporation opposite benzo[a]pyrene DNA adducts by human Pol␬
was performed as previously described [20,21], using
50 fmol of a primed DNA template, 0.7 ng (7 fmol) of
purified Pol␬, and increasing concentrations of dATP,
dCTP, dTTP, or dGTP. Kinetic analysis yielded Vmax
and Km values for the incorporation of the correct and
the incorrect nucleotides. The misincorporation error
rate was calculated from the equation: f inc = (V max /
K m )incorrect /(Vmax /Km )correct .
3. Results
3.1. The (+)- and (−)-trans-anti-BPDE-N2 -dG
adducts in DNA strongly block many eukaryotic
DNA polymerases
To examine the response of various eukaryotic DNA
polymerases to a template (+)-trans-anti-BPDE-N2 -dG
Fig. 1. Strong blockage of various eukaryotic DNA polymerases
by the (+)- and (−)-trans-anti-BPDE-N2 -dG DNA adducts. A
17-mer primer was labeled with 32 P (asterisk) at its 5 end and
annealed two nucleotides before the lesion as shown below the
gel. DNA synthesis assays were performed with purified DNA
polymerases as indicated, using 50 fmol (5 nM) DNA substrate
and 100 fmol of DNA polymerases (10 nM). The yPol␣, yPol␦,
and yPolε indicate the purified catalytic subunits of yeast Pol␣,
Pol␦, and Polε, respectively. DNA size markers in nucleotides are
indicated on the right.
562
Y. Zhang et al. / DNA Repair 1 (2002) 559–569
adducts right before the lesion (Fig. 1, lanes 11 and
12). Thus, the (+)- and (−)-trans-anti-BPDE-N2 -dG
bulky DNA adducts are strong blockers to many
eukaryotic DNA polymerases.
3.2. Error-free bypass of the (+)-trans-anti-BPDEN2 -dG DNA adduct by human Polκ
To examine if the major benzo[a]pyrene DNA
adduct, (+)-trans-anti-BPDE-N2 -dG, can be bypassed
by purified human Pol␬, we annealed a 32 P-labeled
19-mer primer to the damaged template right before the lesion (Fig. 2) and subsequently, performed
DNA synthesis assays. As shown in Fig. 2 (lane 1),
human Pol␬ was able to bypass the (+)-trans-antiBPDE-N2 -dG DNA adduct. To identify the nucleotide
incorporated opposite the lesion, we performed lesion
bypass assays with only one deoxyribonucleoside
triphosphate. As shown in Fig. 2 (lanes 2–5), human Pol␬ predominantly incorporated the correct C
opposite the template (+)-trans-anti-BPDE-N2 -dG.
Fig. 2. Bypass of a template (+)-trans-anti-BPDE-N2 -dG adduct
by human Pol␬. A 19-mer primer was labeled with 32 P (asterisk)
at its 5 end and annealed right before a template (+)-transanti-BPDE-N2 -dG as shown on the right. Polymerase reactions
were performed with 2 ng (20 fmol) of human Pol␬ in the presence
of a single dATP (A), dCTP (C), dTTP (T), or dGTP (G), or all
four dNTPs (N4 ). DNA size markers in nucleotides are indicated
on the left.
These results demonstrate that human Pol␬ is capable of error-free lesion bypass opposite a template
(+)-trans-anti-BPDE-N2 -dG.
3.3. Effect of stereochemistry of BPDE adducts on
lesion bypass efficiency by human Polκ
To determine if the stereochemistry of BPDE adducts affect lesion bypass by human Pol␬, we compared
the bypass efficiency of the (+)-trans-anti-BPDE-N2 dG adduct versus the (−)-trans-anti-BPDE-N2 -dG
adduct. A 19-mer primer was labeled with 32 P at its
5 end and annealed right before the template (+)or (−)-trans-anti-BPDE-N2 -dG lesion. Lesion bypass
assays were then performed with increasing concentrations of purified human Pol␬. As shown in Fig. 3A,
bypass of the (−)-trans-anti-BPDE-N2 -dG adduct
by human Pol␬ was significantly more efficient than
that of the (+)-trans-anti-BPDE-N2 -dG adduct. More
efficient bypass of the (−)-trans-anti-BPDE-N2 -dG
adduct compared to the (+)-trans-anti-BPDE-N2 -dG
adduct was also evident when the primer was annealed two nucleotides before the BPDE adduct
(Fig. 3B). Significant accumulation of the 19-mer
DNA band was observed with the damaged DNA
template, but not with the undamaged template
(Fig. 3B, compare lanes 3–5 with lanes 8–10 and
13–15), indicating an inhibitory effect of the (+)and (−)-trans-anti-BPDE-N2 -dG adducts on human
Pol␬. Furthermore, some DNA synthesis stopped
opposite the (+)-trans-anti-BPDE-N2 -dG adduct,
as evidenced by the accumulation of the 20-mer
DNA band (Fig. 3B, lanes 8–10). This 20-mer DNA
band, however, was not accumulated opposite the
(–)-trans-anti-BPDE-N2 -dG adduct (Fig. 3B, lanes
13–15). These results show that the stereochemistry
of the BPDE DNA adducts significantly affects lesion
bypass efficiency by human Pol␬.
3.4. Fidelity of human Polκ for nucleotide
incorporation opposite (+)- and
(−)-trans-anti-BPDE-N2 -dG DNA adducts
To quantitatively determine the fidelity of nucleotide incorporation by human Pol␬ opposite the (+)and (−)-trans-anti-BPDE-N2 -dG DNA adducts, we
performed kinetic measurements. The Vmax and Km
Y. Zhang et al. / DNA Repair 1 (2002) 559–569
563
values were determined using a previously described
method [20]. The efficiency of nucleotide incorporation is indicated by Vmax /Km , and the misincorporation
frequency is determined by finc , which is calculated
as (Vmax /Km )incorrect /(Vmax /Km )correct . As shown in
Table 1, misincorporation error rates of human Pol␬
opposite the (+)- and (−)-trans-anti-BPDE-N2 -dG
DNA adducts were very similar. Furthermore, these
misincorporation rates opposite the damaged template
G were in the order of 10−2 to 10−3 , which were not
drastically different from those opposite the undamaged template G (in the order of 10−3 ) (Table 1). In
contrast, the (+)- and (−)-trans-anti-BPDE-N2 -dG
DNA adducts inhibited the catalysis efficiency of
human Pol␬ by ∼170- and ∼70-fold, respectively
(Table 1). Consistent with the results of Figs. 3 and 4,
nucleotide incorporation opposite the (−)-trans-antiBPDE-N2 -dG adduct was 2.3-fold more efficient
than that opposite the (+)-trans-anti-BPDE-N2 -dG
adduct. These results show that although the (+)and (−)-trans-anti-BPDE-N2 -dG DNA adducts significantly inhibit human Pol␬, these lesions do not
drastically alter the fidelity of this polymerase.
3.5. Extension DNA synthesis by human Polκ from
opposite the (+)- and (−)-trans-anti-BPDE-N2 -dG
adducts
Fig. 3. Comparison of (−)- and (+)-trans-anti-BPDE-N2 -dG lesion bypass by human Pol␬. (A) A 19-mer primer was labeled
with 32 P at its 5 end and annealed right before a template (+)- or
(−)-trans-anti-BPDE-N2 -dG adduct for DNA synthesis assays, using increasing concentrations of purified human Pol␬ as indicated.
Quantitation of the bypassed 31-mer and 32-mer DNA products
is shown. (B) A 17-mer primer was labeled with 32 P (asterisk) at
its 5 end and annealed two nucleotides before the template (+)or (−)-trans-anti-BPDE-N2 -dG adduct as shown at the top. DNA
synthesis assays were performed with increasing concentrations
of purified human Pol␬ as indicated. The arrowhead indicates the
lesion site. DNA size markers in nucleotides are indicated on the
left.
To examine extension DNA synthesis by human
Pol␬ from opposite a template (+)- or (−)-trans-antiBPDE-N2 -dG adduct, we performed DNA synthesis
assays using four DNA substrates containing a C, A,
T, or G, respectively, at the primer 3 end opposite
the lesion (Fig. 4). As shown in Fig. 4B and C (lanes
1 and 6), human Pol␬ effectively catalyzed extension
DNA synthesis from opposite the BPDE lesion when
the primer 3 end was either a C or a T. The extension
DNA synthesis was less efficient when the primer 3
end was either an A or a G (Fig. 4A and D, lanes 1
and 6).
To identify the nucleotide incorporated opposite
the undamaged template C 5 to the BPDE lesion,
we performed extension synthesis assays again with
only one deoxyribonucleoside triphosphate. When
the primer 3 end was either an A or a C opposite the
BPDE lesion, the correct G was incorporated opposite
the undamaged template C during extension DNA
synthesis (Fig. 4A and B, lanes 5 and 10). When the
564
Y. Zhang et al. / DNA Repair 1 (2002) 559–569
Table 1
Kinetic measurement of nucleotide incorporation opposite (+)- and (−) trans-anti-BPDE-N2 -dG by human Pol␬
dNTP
Vmax (fmol/min; mean ± S.D.)
Undamaged G
dATP
dCTP
dTTP
dGTP
2.21
2.61
3.05
2.52
Km (␮M; mean ± S.D.)
Vmax /Km
finc a
0.10
11.9
0.020
0.043
8.4 × 10−3
1
1.7 × 10−3
3.6 × 10−3
±
±
±
±
0.12
0.12
0.12
0.13
21.2
0.22
50.5
58.6
±
±
±
±
5.60
0.05
8.80
11.9
(+)-trans-anti-BPDE-N2 -dG
dATP
1.11 ±
dCTP
2.11 ±
dTTP
0.58 ±
dGTP
0.36 ±
0.02
0.16
0.03
0.02
532
30.6
634
315
±
±
±
±
24.6
9.80
111
35.0
0.0021
0.069
0.00091
0.0011
3.0 × 10−2
1
1.3 × 10−2
1.6 × 10−2
(−)-trans-anti-BPDE-N2 -dG
dATP
0.58 ±
dCTP
2.31 ±
dTTP
0.86 ±
dGTP
0.41 ±
0.03
0.25
0.04
0.02
178
14.8
510
376
±
±
±
±
35.0
6.29
74.3
72.4
0.0033
0.16
0.0017
0.0011
2.1 × 10−2
1
1.1 × 10−2
6.9 × 10−3
a
f inc = (V max /K m )incorrect /(Vmax /Km )correct .
primer 3 end was a T opposite the lesion, T was significantly incorporated in addition to G incorporation
(Fig. 4C, lanes 4, 5, 9 and 10). Surprisingly, when
the primer 3 end was a G opposite the lesion, T was
predominantly incorporated (Fig. 4D, lanes 4, and
9), whereas the correct G incorporation was barely
detected (Fig. 4D, lanes 5 and 10).
Extension DNA synthesis from mispaired primer
3 ends by human Pol␬ was significantly more efficient from opposite the (−)-trans-anti-BPDE-N2 -dG
adduct than that from the (+)-trans-anti-BPDE-N2 -dG
adduct (compare lanes 1 and 6 of Fig. 4A–D). To
quantitatively determine if the stereochemistry of
the BPDE DNA adducts also affects the efficiency
of extension DNA synthesis from a paired C-GBPDE
(primer-template), we measured the kinetic Vmax and
Km values of this reaction. As shown in Table 2, extension DNA synthesis from opposite the (−)-trans-antiBPDE-N2 -dG adduct was indeed ∼18-fold more
efficient than that from the (+)-trans-anti-BPDE-N2 dG adduct. While extension DNA synthesis from the
paired C-GBPDE primer end was error-free (Fig. 4B,
lanes 5 and 10), the (−)- and (+)-trans-anti-BPDEN2 -dG adducts inhibited such extension synthesis by
5- and 83-fold (Table 2), respectively. Together, these
results show that the efficiency and fidelity of extension DNA synthesis by human Pol␬ from opposite
the (−)- and (+)-trans-anti-BPDE-N2 -dG adducts
are strongly influenced by the nucleotide residing
opposite the template lesion.
3.6. Mechanisms of extension DNA synthesis from
mispaired primer 3 ends opposite the (−)- and
(+)-trans-anti-BPDE-N2 -dG adducts by human Polκ
The major product of extension synthesis from
opposite the G-GBPDE (primer-template) mispair was
one nucleotide shorter (Fig. 4D) than that from the
Table 2
Kinetic measurement of extension DNA synthesis by human Pol␬ from a C-GBPDE (primer-template) pair
dGTP
Vmax (fmol/min; mean ± S.D.)
Km (␮M; mean ± S.D.)
Vmax /Km
fext a
C-G
C-G(+)-trans -anti -BPDE
C-G(−)-trans -anti -BPDE
2.75 ± 0.09
1.98 ± 0.19
2.16 ± 0.08
0.10 ± 0.02
6.07 ± 0.29
0.37 ± 0.08
27.5
0.33
5.84
1
1.2 × 10−2
2.1 × 10−1
a
f ext = (V max /K m )damaged
G /(Vmax /Km )undamaged G .
Y. Zhang et al. / DNA Repair 1 (2002) 559–569
565
Fig. 4. Extension DNA synthesis of human Pol␬ from opposite the (+)- or (−)-trans-anti-BPDE-N2 -dG DNA adducts. Four
20-mer 32 P-labeled primers were separately annealed to a DNA template with the primer 3 end opposite a template (+)- or
(−)-trans-anti-BPDE-N2 -dG adduct as shown at the top. Each primer differed only at the 3 end. DNA synthesis assays were then performed with 5 ng (50 fmol) of purified human Pol␬ in the presence of a single dATP, dCTP, dTTP, or dGTP, or all four dNTPs. (A) A
at the primer 3 end; (B) C at the primer 3 end; (C) T at the primer 3 end; and (D) G at the primer 3 end. DNA size markers in
nucleotides are indicated on the sides. Asterisk, 32 P-label; underline, DpnII recognition sequence; arrows, DpnII cleavage sites.
566
Y. Zhang et al. / DNA Repair 1 (2002) 559–569
C-GBPDE pair (Fig. 4B). Increasing human Pol␬ concentration in the reaction did not alter this product
pattern (Fig. 5C, lanes 9 and 11). Thus, extension
synthesis from opposite the G-GBPDE mispair might
be mediated by a –1 deletion mechanism. To directly
test this possibility, we digested the extension synthesis products with the DpnII restriction endonuclease. Cleavage of the products would yield a 26-mer
DNA band, if normal extension DNA synthesis had
occurred. This was indeed observed when the undamaged DNA template was copied by human Pol␬
(Fig. 5A, lane 2). The presence of a template (−)- or
(+)-trans-anti-BPDE-N2 -dG adduct did not change
the DpnII cleavage pattern (Fig. 5A, lanes 4 and 6).
Thus, extension synthesis from the C-GBPDE pair by
human Pol␬ proceeded by a normal DNA synthesis
mechanism as expected. Similar results were obtained
with extension synthesis from the A-G mispair, regardless of whether or not the template G contained
the (−)- and (+)-trans-anti-BPDE-N2 -dG adducts
(Fig. 5B). In contrast, following extension synthesis from the G-GBPDE mispair, the DpnII cleavage
yielded predominantly a one-nucleotide shorter DNA
band (25-mer) (Fig. 5C, lanes 10 and 12), indicating
a −1 deletion mechanism. Even without the lesion,
extension synthesis from the G-G mispair yielded
∼40% −1 deletion products (Fig. 5C, lane 8).
The DpnII cleavage analysis of DNA products
from extending the T-GBPDE mispair indicated DNA
synthesis without deletion as the major mechanism
(Fig. 5C, lanes 4 and 6). However, some extension
synthesis products were shorter by two nucleotides
(Fig. 5C, lanes 4 and 6), indicating a −2 deletion
mechanism during DNA synthesis by human Pol␬,
presumably as a result of re-aligning the primer
end T to pair with the template A two nucleotides
downstream. This explains the significant T incorporation in Fig. 4C (lanes 4 and 9). These results
show that the mechanism of extension DNA synthesis by human Pol␬ from opposite the (−)- and
Fig. 5. Analysis of BPDE lesion bypass products by DpnII restriction digestion. Extension synthesis was performed from opposite a template
(+)- and (−)-trans-anti-BPDE-N2 -dG DNA adduct, using 20 ng (200 fmol) of purified human Pol␬. After the polymerase reaction, DNA
was purified by phenol/chloroform extractions. The DNA was either untreated (DpnII, −) or treated (DpnII, +) with the DpnII restriction
endonuclease (20 U) at 37 ◦ C for 4 h. DNA was separated by electrophoresis and visualized by autoradiography. Sequences of the primer
and the template are shown in Fig. 4. The primer 3 end and its opposing template base are indicated below the gels as primer-template.
DNA size markers in nucleotides are indicated on the right.
Y. Zhang et al. / DNA Repair 1 (2002) 559–569
(+)-trans-anti-BPDE-N2 -dG adducts is strongly influenced by the nucleotide residing opposite the template
lesion. These results also suggest that −1 deletion
occurs efficiently during extension DNA synthesis
by human Pol␬ when the primer 3 end opposite the
BPDE adducts can pair with the next undamaged
template base.
4. Discussion
We have shown that purified human Pol␬ can effectively bypass the template (+)- and (−)-trans-antiBPDE-N2 -dG adducts in an error-free manner, and
determined the kinetic parameters for the bypass.
Thus, Pol␬ may play an important role in suppressing benzo[a]pyrene-induced mutagenesis in
humans. Furthermore, we have found another activity of human Pol␬ in response to the (+)- and
(−)-trans-anti-BPDE-N2 -dG adducts: extension synthesis from mispaired primer 3 ends opposite these
lesions. Pol␬ extension from a G opposite the 3 T of
a template TT dimer has also been observed recently
[22].
Human Pol␬ inserts the correct C 2.3-fold more
efficiently opposite the (−)-trans-anti-BPDE-N2 -dG
adduct and performs subsequent extension DNA
synthesis ∼18-fold more efficiently. Thus, bypass of the (−)-trans-anti-BPDE-N2 -dG adduct by
human Pol␬ is (2.3 × 18) ∼41-fold more efficient than that of the (+)-trans-anti-BPDE-N2 -dG
adduct. In contrast, purified human Pol␩ performs
error-prone translesion synthesis opposite the (+)and (−)-trans-anti-BPDE-N2 -dG DNA adducts,
with higher efficiency opposite the (+)-stereoisomer
(manuscript in preparation). If these observations are
valid for most sequence contexts, the (+)-trans-antiBPDE-N2 -dG DNA adduct would be predicted to be
more mutagenic than the (−)-trans-anti-BPDE-N2 -dG
adduct. Supporting this prediction, Moriya et al.
[3] reported that the (+)-trans-anti-BPDE-N2 -dG
DNA adduct is indeed more mutagenic than the
(−)-trans-anti-BPDE-N2 -dG DNA adduct in a defined sequence context in simian kidney cells. Furthermore, it was also reported that the (+)-anti-benzo
[a]pyrene dioepoxide is more mutagenic than the
(−)-anti-benzo[a]pyrene dioepoxide in the hamster
V79 cells [23].
567
During error-free bypass of human Pol␬, extension
is (5.83/0.16) ∼36-fold and (0.33/0.069) ∼5-fold
more efficient than insertion for the (−)- and the
(+)-trans-anti-BPDE-N2 -dG DNA adducts, respectively. Thus, a rate limiting step during bypass of
these adducts by human Pol␬ is at the nucleotide
insertion step opposite the lesion. This explains
our observation that a DNA band was accumulated
right before the lesion, but such accumulation was
absent opposite the (−)-trans-anti-BPDE-N2 -dG
adduct, or was only slightly accumulated opposite the
(+)-trans-anti-BPDE-N2 -dG DNA adduct (Fig. 3B).
During extension synthesis of human Pol␬ from
opposite the (−)- and (+)-trans-anti-BPDE-N2 -dG
DNA adducts, −1 deletion occurs as the predominant mechanism from the mispair G-GBPDE at the
sequence context 5 -CGBPDE C. This is probably mediated by re-aligning the primer 3 G with the next
template C prior to DNA synthesis. Such −1 deletion
mechanism is unlikely specific to the combination
of the primer 3 G and the next template C, since a
similar −1 deletion mechanism also occurs during
extension synthesis by human Pol␬ from the primer
3 A opposite an AP site that is flanked by a 5 T in
the template [11,12]. Thus, the −1 deletion mechanism is probably determined by both the nucleotide
residing opposite the BPDE adduct and the next undamaged template base. When these two bases can
form a Watson–Crick base pairing, −1 deletion is
predicted to occur efficiently during extension DNA
synthesis by human Pol␬ (Fig. 6). Therefore, the efficiency and fidelity of the Pol␬-catalyzed extension
synthesis from opposite the BPDE adduct, hence the
overall lesion bypass, could be strongly affected by
the sequence context 5 to the lesion site. A similar
conclusion was also reported by Zhuang et al. [24]
based on studies with the Klenow fragment.
In combination with another DNA polymerase, the
extension DNA synthesis activity of human Pol␬ may
contribute to the bypass of some BPDE adducts by the
two-polymerase two-step mechanism. That is, following nucleotide insertion opposite a BPDE adduct by
another polymerase such as Pol␩, subsequent extension synthesis step could be catalyzed by Pol␬. In this
regard, Pol␬ would play a dual-function role during
bypass of BPDE lesions, similar to the dual-function
role of Pol␨ in lesion bypass as we recently postulated [25]. One function of Pol␬ would be to catalyze
568
Y. Zhang et al. / DNA Repair 1 (2002) 559–569
Fig. 6. A dual-function model of human Pol␬ in lesion bypass of DPBE DNA adducts. This model postulates that Pol␬ can catalyze
error-free lesion bypass of BPDE DNA adducts (one-polymerase two step-mechanism). Additionally, Pol␬ can also catalyze the extension
DNA synthesis following nucleotide insertion opposite BPDE adducts by other polymerases (two-polymerase two-step mechanism). In
the one-polymerase two-step mechanism, if the 5 undamaged template base (X) is a G, −1 deletion may also occur to some extent as
indicated by the thin arrow.
error-free lesion bypass of some BPDE adducts
(one-polymerase two-step mechanism) (Fig. 6).
The second function of Pol␬ would be to catalyze
the extension DNA synthesis following nucleotide
insertion opposite some BPDE adducts by other
polymerases (two-polymerase two-step mechanism)
(Fig. 6). Indeed, lesion bypass by the two-polymerase
two-step mechanism can be demonstrated in vitro
by the sequential actions of REV1 insertion and
Pol␬ extension in response to a template (+)- and
(−)-trans-anti-BPDE-N2 -dG adduct or a template
1,N6 -ethenoadenine adduct [26]. Consistent with an in
vivo function of Pol␬ in lesion bypass of BPDE DNA
adducts, DNA polymerase IV (Pol␬ homologue) is
required for both error-free and −1 deletion bypass
of a template (+)-trans-anti-BPDE-N2 -dG adduct in
E. coli cells [27], and the mouse Polκ gene can be
up-regulated by the AhR receptor [28].
Acknowledgements
We thank Dr. Nicholas E. Geacintov for supporting
the studies of Olga Rechkoblit and for his critical
review of this manuscript. This work was supported
by a New Investigator Award in Toxicology from
Burroughs Wellcome Fund (Z.W.), and NIH grants
CA92768 (Z.W.) and CA20851 (Nicholas E. Geacintov).
References
[1] S.C. Cheng, B.D. Hilton, J.M. Roman, A. Dipple, DNA
adducts from carcinogenic and noncarcinogenic enantiomers
of benzo[a]pyrene dihydrodiol epoxide, Chem. Res. Toxicol.
2 (1989) 334–340.
[2] K. Peltonen, A. Dipple, Polycyclic aromatic hydrocarbons:
chemistry of DNA adduct formation, J. Occup. Environ. Med.
37 (1995) 52–58.
[3] M. Moriya, S. Spiegel, A. Fernandes, S. Amin, T. Liu, N.
Geacintov, A.P. Grollman, Fidelity of translesional synthesis
past benzo[a]pyrene diol epoxide-2 -deoxyguanosine DNA
adducts: marked effects of host cell, sequence context, and
chirality, Biochemistry 35 (1996) 16646–16651.
[4] A. Fernandes, T. Liu, S. Amin, N.E. Geacintov, A.P.
Grollman, M. Moriya, Mutagenic potential of stereoisomeric
bay region (+)- and (−)-cis-anti-benzo[a]pyrene diol epoxide-N2-2 -deoxyguanosine adducts in Escherichia coli and
simian kidney cells, Biochemistry 37 (1998) 10164–10172.
[5] W. Mackay, M. Benasutti, E. Drouin, E.L. Loechler,
Mutagenesis by (+)-anti-B[a]P-N2-Gua, the major adduct of
activated benzo[a]pyrene, when studied in an Escherichia
coli plasmid using site-directed methods, Carcinogenesis 13
(1992) 1415–1425.
Y. Zhang et al. / DNA Repair 1 (2002) 559–569
[6] S.A. Jelinsky, T. Liu, N.E. Geacintov, E.L. Loechler, The
major, N2-Gua adduct of the (+)-anti-benzo[a]pyrene diol
epoxide is capable of inducing G → A and G → C, in addition
to G → T, mutations, Biochemistry 34 (1995) 13545–13553.
[7] R. Shukla, T. Liu, N.E. Geacintov, E.L. Loechler, The major,
N2-dG adduct of (+)-anti-B[a]PDE shows a dramatically
different mutagenic specificity (predominantly, G → A) in a
5 -CGT-3 sequence context, Biochemistry 36 (1997) 10256–
10261.
[8] J.E. Page, B. Zajc, T. Oh-hara, M.K. Lakshman,
J.M. Sayer, D.M. Jerina, A. Dipple, Sequence context
profoundly influences the mutagenic potency of trans-opened
benzo[a]pyrene 7,8-diol 9,10-epoxide-purine nucleoside
adducts in site-specific mutation studies, Biochemistry 37
(1998) 9127–9137.
[9] Z. Wang, Translesion synthesis by the UmuC family of DNA
polymerases, Mutat. Res. 486 (2001) 59–70.
[10] H. Ohmori, E.C. Friedberg, R.P.P. Fuchs, M.F. Goodman, F.
Hanaoka, D. Hinkle, T.A. Kunkel, C.W. Lawrence, Z. Livneh,
T. Nohmi, L. Prakash, S. Prakash, T. Todo, G.C. Walker,
Z. Wang, R. Woodgate, The Y-family of DNA polymerases,
Mol. Cell 8 (2001) 7–8.
[11] Y. Zhang, F. Yuan, X. Wu, M. Wang, O. Rechkoblit, J.-S.
Taylor, N.E. Geacintov, Z. Wang, Error-free and error-prone
lesion bypass by human DNA polymerase ␬ in vitro, Nucleic
Acids Res. 28 (2000) 4138–4146.
[12] E. Ohashi, T. Ogi, R. Kusumoto, S. Iwai, C. Masutani, F.
Hanaoka, H. Ohmori, Error-prone bypass of certain DNA
lesions by the human DNA polymerase ␬, Genes Dev. 14
(2000) 1589–1594.
[13] V.L. Gerlach, W.J. Feaver, P.L. Fischhaber, E.C. Friedberg,
Purification and characterization of Pol␬, a DNA polymerase
encoded by the human DINB1 gene, J. Biol. Chem. 276
(2001) 92–98.
[14] N. Suzuki, E. Ohashi, K. Hayashi, H. Ohmori, A.P. Grollman,
S. Shibutani, Translesional synthesis past acetylaminofluorene-derived DNA adducts catalyzed by human DNA
polymerase ␬ and Escherichia coli DNA polymerase IV,
Biochemistry 40 (2001) 15176–15183.
[15] Y. Zhang, F. Yuan, X. Wu, O. Rechkoblit, J.-S. Taylor,
N.E. Geacintov, Z. Wang, Error-prone lesion bypass by
human DNA polymerase ␩, Nucleic Acids Res. 28 (2000)
4717–4724.
[16] M. Cosman, V. Ibanez, N.E. Geacintov, R.G. Harvey,
Preparation and isolation of adducts in high yield derived from
the binding of two benzo[a]pyrene-7,8-dihydroxy-9,10-oxide
stereoisomers to the oligonucleotide d(ATATGTATA), Carcinogenesis 11 (1990) 1667–1672.
569
[17] N.E. Geacintov, M. Cosman, B. Mao, A. Alfano, V. Ibanez,
R.G. Harvey, Spectroscopic characteristics and site I/site II
classification of cis and trans benzo[a]pyrene diolepoxide
enantiomer-guanosine adducts in oligonucleotides and
polynucleotides, Carcinogenesis 12 (1991) 2099–2108.
[18] O. Rechkoblit, S. Amin, N.E. Geacintov, Primer length
dependence of binding of DNA polymerase I Klenow fragment to template-primer complexes containing site-specific
bulky lesions, Biochemistry 38 (1999) 11834–11843.
[19] Y. Zhang, X. Wu, F. Yuan, Z. Xie, Z. Wang, Highly frequent
frameshift DNA synthesis by human DNA polymerase ␮,
Mol. Cell. Biol. 21 (2001) 7995–8006.
[20] S. Creighton, L.B. Bloom, M.F. Goodman, Gel fidelity
assay measuring nucleotide misinsertion, exonucleolytic
proofreading, and lesion bypass efficiencies, Methods
Enzymol. 262 (1995) 232–256.
[21] Y. Zhang, F. Yuan, X. Wu, Z. Wang, Preferential incorporation
of G opposite template T by the low fidelity human DNA
polymerase ␫, Mol. Cell. Biol. 20 (2000) 7099–7108.
[22] M.T. Washington, R.E. Johnson, L. Prakash, S. Prakash,
Human DINB1-encoded DNA polymerase ␬ is a promiscuous
extender of mispaired primer termini, Proc. Natl. Acad. Sci.
USA 99 (2002) 1910–1914.
[23] P. Brookes, M.R. Osborne, Mutation in mammalian cells by
stereoisomers of anti-benzo[a] pyrene-diolepoxide in relation
to the extent and nature of the DNA reaction products,
Carcinogenesis 3 (1982) 1223–1226.
[24] P. Zhuang, A. Kolbanovskiy, S. Amin, N.E. Geacintov,
Base sequence dependence of in vitro translesional DNA
replication past a bulky lesion catalyzed by the exo- Klenow
fragment of Pol I, Biochemistry 40 (2001) 6660–6669.
[25] D. Guo, X. Wu, D.K. Rajpal, J.-S. Taylor, Z. Wang,
Translesion synthesis by yeast DNA polymerase ␨ from templates containing lesions of ultraviolet radiation and acetylaminofluorene, Nucleic Acids Res. 29 (2001) 2875–2883.
[26] Y. Zhang, X. Wu, O. Rechkoblit, N.E. Geacintov, J.-S.
Taylor, Z. Wang, Response of human REV1 to different
DNA damage: preferential dCMP insertion opposite the
lesion, Nucleic Acids Res. 30 (2002) 1630–1638.
[27] R. Napolitano, R. Janel-Bintz, J. Wagner, R.P. Fuchs, All
three SOS-inducible DNA polymerases (Pol II, Pol IV and
Pol V) are involved in induced mutagenesis, EMBO J. 19
(2000) 6259–6265.
[28] T. Ogi, J. Mimura, M. Hikida, H. Fujimoto, Y. FujiiKuriyama, H. Ohmori, Expressions of human and mouse
genes encoding Pol␬: testis-specific developmental regulation
and AhR-dependent inducible transcription, Genes to Cells
6 (2001) 943–954.