Synthesis, DNA polymerase incorporation, and enzymatic phosphate hydrolysis of formamidopyrimidine nucleoside triphosphates

Article abstract of DOI:10.1021/ja065525r

The nucleoside triphosphates of N6-(2-deoxy-α,β-D-erythro- pentofuranosyl)-2,6-diamino-4-hydroxy-5-formamidopyrimidine (Fapy·dGTP) and its C-nucleoside analogue (β-C-Fapy·dGTP) were synthesized. The lability of the formamide group required that nucleoside

Full text of DOI:10.1021/ja065525r

Published on Web 10/25/2006
Synthesis, DNA Polymerase Incorporation, and Enzymatic
Phosphate Hydrolysis of Formamidopyrimidine Nucleoside
Triphosphates
Shuhei Imoto, Jennifer N. Patro, Yu Lin Jiang, Natsuhisa Oka, and
Marc M. Greenberg*
Contribution from the Department of Chemistry, Johns Hopkins UniVersity, 3400 North Charles
Street, Baltimore, Maryland 21218
Received July 31, 2006; E-mail: mgreenberg@jhu.edu
Abstract: The nucleoside triphosphates of N6-(2-deoxy-R,â-D-erythro-pentofuranosyl)-2,6-diamino-4-hy-
droxy-5-formamidopyrimidine (Fapy‚dGTP) and its C-nucleoside analogue (â-C-Fapy‚dGTP) were syn-
thesized. The lability of the formamide group required that nucleoside triphosphate formation be carried
out using an umpolung strategy in which pyrophosphate was activated toward nucleophilic attack. The
Klenow fragment of DNA polymerase I from Escherichia coli accepted Fapy‚dGTP and â-C-Fapy‚dGTP as
substrates much less efficiently than it did dGTP. Subsequent extension of a primer containing either modified
nucleotide was less affected compared to when the native nucleotide is present at the 3′-terminus. The
specificity constants are sufficiently large that nucleoside triphosphate incorporation could account for the
level of Fapy‚dG observed in cells if 1% of the dGTP pool is converted to Fapy‚dGTP. Similarly, polymerase-
mediated introduction of â-C-Fapy‚dG could be useful for incorporating useful amounts of this nonhydro-
lyzable analogue for use as an inhibitor of base excision repair. The kinetic viability of these processes is
enhanced by inefficient hydrolysis of Fapy‚dGTP and â-C-Fapy‚dGTP by MutT, the E. coli enzyme that
releases pyrophosphate and the corresponding nucleoside monophosphate upon reaction with structurally
related nucleoside triphosphates.
DNA lesions are produced when nucleic acids are exposed
to oxidative stress and are associated with aging and a variety
of diseases.1-3 When present in DNA, damaged nucleotides can
be mutagenic and/or cytotoxic.4-7 A lesion can also be
incorporated in DNA following damage to the nucleoside
triphosphate pool provided the respective triphosphate is a
substrate for DNA polymerase(s). Organisms have developed
elaborate, redundant repair systems that excise DNA lesions
into duplex DNA. Hydrolysis of the nucleoside triphosphates
before they affect replication or transcription.8,9 Enzymes that
by the hydrolase from E. coli (MutT) that acts on the related
guard against mutations by cleansing the nucleoside triphosphate
8-oxopurine lesion OxodGTP was also examined. The results
pool also exist.10 Described below are the results of investiga-
of these experiments raise interesting possibilities regarding
tions on two nucleoside triphosphates that are related to the
DNA mutagenesis and repair. The efficiency with which Klenow
formamidopyrimidine derived from 2′-deoxyguanosine (N6-(2-
incorporates the modified nucleosides in DNA and the reluctance
deoxy-R,â-D-erythro-pentofuranosyl)-2,6-diamino-4-hydroxy-5-
of MutT to hydrolyze the triphosphates indicate that Fapy‚dGTP
formamidopyrimidine, Fapy‚dG). We examined the ability of
could be a source of Fapy‚dG in DNA and that â-C-Fapy‚dGTP
a model Escherichia coli polymerase (Klenow) to incorporate
could be useful as a DNA repair inhibitor.
Fapy‚dGTP and its C-nucleotide analogue (â-C-Fapy‚dGTP)
Fapy‚dG is believed to result from a radical intermediate
(Scheme 1) that serves as a common precursor to 7,8-dihydro-
(1) Lombard, D. B.; Chua, K. F.; Mostoslavsky, R.; Franco, S.; Gostissa, M.;
8-hydroxy-2′-deoxyguanosine (OxodG).11 The lesions are de-
Alt, F. W. Cell 2005, 120, 435-567.
(2) Evans, M. D.; Dizdaroglu, M.; Cooke, M. S. Mutat. Res. 2004, 567, 1-61.
tected in comparable quantities, particularly when DNA is
subjected to oxidative stress under O2-deficient conditions.12,13
Fapy‚dG and OxodG affect polymerases and repair enzymes
(3) DePinho, R. A. Nature (London) 2000, 408, 248-254.
(4) Neeley, W. L.; Essigmann, J. M. Chem. Res. Toxicol. 2006, 19, 491-505.
(5) Kamiya, H. Nucleic Acids Res. 2003, 31, 517-531.
(6) Cooke, M. S.; Evans, M. D.; Dizdaroglu, M.; Lunec, J. FASEB 2003, 17,
1195-1214.
(7) Bjelland, S.; Seeberg, E. Mutat. Res. 2003, 531, 37-80.
(8) Scha¨rer, O. D. Angew. Chem., Int. Ed. 2003, 42, 2946-2974.
(9) Stivers, J. T.; Jiang, Y. L. Chem. ReV. 2003, 103, 2729-2759.
(10) Bessman, M. J.; Frick, D. N.; O’Handley, S. F. J. Biol. Chem. 1996, 271,
25059-25062.
(11) Candeias, L. P.; Steenken, S. Chem.sEur. J. 2000, 6, 475-484.
(12) Cadet, J.; Douki, T.; Gasparutto, D.; Ravanat, J. L. Mutat. Res. 2003, 531,
5-23.
(13) Dizdaroglu, M.; Jaruga, P.; Birincioglu, M.; Rodriguez, H. Free Radical
Biol. Med. 2002, 32, 1102-1115.
9
14606
J. AM. CHEM. SOC. 2006, 128, 14606-14611
10.1021/ja065525r CCC: $33.50 © 2006 American Chemical Society

Formamidopyrimidine Nucleoside Triphosphate Effects
A R T I C L E S
Scheme 1. Formation of Fapy‚dG and OxodG via a Common Intermediate
in similar ways when they are present in DNA. Both lesions
induce DNA polymerases to misinsert dA when they are present
in duplex DNA.14-16 G f T transversions result when this
misreplication event occurs in cells.17-20 Base excision repair
enzymes, such as formamidopyrimidine-DNA glycosylase (Fpg)
also excise Fapy‚dG and OxodG, but discriminate with respect
to the opposing nucleotide to prevent transforming a promu-
tagenic base pair (e.g., Fapy‚dG:dA) into a mutation.21-23
Introduction of OxodG into DNA via its nucleoside triphosphate
(OxodGTP) has also been investigated. OxodGTP is a substrate
for several DNA polymerases.24,25 The efficiency with which
the modified nucleotide is incorporated into DNA relative to
dG via its respective triphosphate varies. For instance, the
Klenow fragment of DNA polymerase I from E. coli that lacks
proofreading capability (Klenow exo-) accepts OxodGTP more
than 150000-fold less efficiently than dGTP. However, incor-
poration of OxodG by the R-subunit of the major replicative
polymerase in E. coli occurs at almost 3% the rate of that of
dG, suggesting that OxodGTP could be a significant source of
the lesion in DNA.25 This hypothesis is corroborated by the
determination that OxodGTP is hydrolyzed by the E. coli
nucleoside-triphosphate pyrophosphohydrolase, MutT, as much
as 200-fold more efficiently than dGTP.10,25,26 These data have
given rise to a widely accepted proposal that a biological role
for MutT is to sanitize the nucleoside triphosphate pool by
removing OxodGTP. Consequently, MutT is often referred to
as a member of the GO (guanine oxidation) family of repair
enzymes. Fpg and the mismatch repair enzyme MutY are the
other members of this repair enzyme family. However, this
model was recently questioned when OxodGTP was not detected
in cell extracts of E. coli mutT mutants that were subjected to
oxidative stress.27 An unknown compound, presumably a
different nucleoside triphosphate, was detected. It is conceivable
that this molecule is Fapy‚dGTP or one of several oxidation
productsofOxodGTPthathaverecentlybeencharacterized.4,20,28-30
We sought to determine whether Fapy‚dGTP could be a
source of Fapy‚dG in DNA by synthesizing this molecule and
characterizing its interactions with MutT and Klenow. In
addition, we carried out the analogous synthesis and study of
â-C-Fapy‚dGTP. We were motivated to carry out this exercise
due to the considerable interest in analogues of DNA lesions
asmechanisticprobesandasbaseexcisionrepairinhibitors.21,22,31-33
Duplex DNA containing hydrolytically stable â-C-Fapy‚dG was
previously shown to strongly bind the enzyme that cleaves the
glycosidic bond in Fapy‚dG.21 Synthesizing â-C-Fapy‚dGTP
enabled us to determine whether this molecule was a viable
base excision repair inhibitor.
Scheme 2. Umpolung Preparation of Fapy‚dGTPa
(14) Wiederholt, C. J.; Greenberg, M. M. J. Am. Chem. Soc. 2002, 124, 7278-
7279.
(15) Shibutani, S.; Takeshita, M.; Grollman, A. P. Nature (London) 1991, 349,
431-434.
(16) Lowe, L. G.; Guengerich, F. P. Biochemistry 1996, 35, 9840-9849.
(17) Kalam, M. A.; Haraguchi, K.; Chandani, S.; Loechler, E. L.; Moriya, M.;
Greenberg, M. M.; Basu, A. K. Nucleic Acids Res. 2006, 34, 2305-2315.
(18) Tan, X.; Grollman, A. P.; Shibutani, S. Carcinogenesis 1999, 20, 2287-
2292.
(19) Wood, M. L.; Esteve, A.; Morningstar, M. L.; Kuziemko, G. M.; Essigmann,
J. M. Nucleic Acids Res. 1992, 20, 6023-6032.
(20) Henderson, P. T.; Delaney, J. C.; Muller, J. G.; Neeley, W. L.; Tannenbaum,
S. R.; Burrows, C. J.; Essigmann, J. M. Biochemistry 2003, 42, 9257-
9262.
(21) Wiederholt, C. J.; Delaney, M. O.; Pope, M. A.; David, S. S.; Greenberg,
M. M. Biochemistry 2003, 42, 9755-9760.
a Conditions and reagents: (a) pyrophosphate; (b) 1,1-carbonyldiimida-
zole; (c) Fap‚dGMP; (d) hν (350 nm), MeOH.
9
J. AM. CHEM. SOC. VOL. 128, NO. 45, 2006 14607

A R T I C L E S
Imoto et al.
Scheme 3. Synthesis of Fapy‚dGMPa
a Conditions and reagents: (a) (i) bis(2-cyanoethyl)-N,N-diisopropylaminophosphine, tetrazole, CH3CN, 25 °C; (ii) t-BuOOH, 25 °C; (b) (i) H2, Pd/C,
DlPEA, THF; (ii) pyridine, formic acetic anhydride, 0 °C; (c) (i) DBU, BSA, pyridine, 25 °C; (ii) AcOH, TBAF‚3H2O, MeOH; (iii) K2CO3, MeOH, 25 °C.
Scheme 4. Synthesis of â-C-Fapy‚dGMPa
Results and Discussion
Synthesis of Fapy‚dGTP and â-C-Fapy‚dGTP. Nucleoside
triphosphate synthesis is typically achieved from the respective
nucleoside via the monophosphate, which is activated for
nucleophilic attack by pyrophosphate.34 Isolated yields of the
triphosphates are often low, but pure products are obtained
following anion exchange or reversed-phase chromatography
in sufficient quantities for enzymology experiments. These issues
notwithstanding, it was not possible to synthesize Fapy‚dGTP
or â-C-Fapy‚dGTP by this method. For instance, attempted
activation of the corresponding monophosphates with 1,1-
carbonyldiimidazole resulted in deformylation. Consequently,
we developed an umpolung-type approach for synthesizing
Fapy‚dGTP and â-C-Fapy‚dGTP from the respective nucleoside
monophosphates (Scheme 2). The o-nitroveratrole ether of
pyrophosphate (1) was activated by 1,1-carbonyldiimidazole.35
After excess electrophile was quenched with methanol, the
photolabile pyrophosphate was coupled to Fapy‚dGMP and â-C-
Fapy‚dGMP. The photolabile, protected nucleoside triphosphates
were partially purified by anion exchange chromatography and
photolyzed (350 nm) to yield Fapy‚dGTP and â-C-Fapy‚dGTP,
which were purified using a mono-Q column on an FPLC
instrument.
a Conditions and reagents: (a) (i) bis(2-cyanoethoxy)-N,N-diisopropy-
laminophosphine, pyridine hydrochloride, pyridine, -20 °C; (ii) t-BuOOH,
25 °C; (b) (i) DBU, BSA, pyridine, 25 °C; (ii) K2CI3, MeOH, 25 °C.
Figure 1. Determination of stereochemistry in â-C-Fapy‚dGMP via NOE.
Fapy‚dGMP was synthesized using an approach (and inter-
mediate) employed during the preparation of the corresponding
phosphoramidite (3, Scheme 3).36-38 These previous studies
determined that it was necessary to introduce the phosphate
group prior to transformation of the nitro group into a formamide
to prevent rearrangement to the pyranose isomer. Consequently,
Fapy‚dGMP was prepared as a mixture of R- and â-anomers
from previously reported 3 (Scheme 3).38 The anomers of 5
were separated by chromatography, but it was unnecessary to
do so for practical preparative purposes because the diastere-
omers readily equilibrated. After treatment of 5 with excess N,O-
bis(trimethylsilyl)acetamide and DBU, the amine and alcohol
groups of the crude monophosphate were sequentially depro-
tected.39 The anomeric mixture of Fapy‚dGMP, which was
purified by anion exchange chromatography, was coupled as
described above (Scheme 2) to form the desired nucleoside
triphosphate as a 1:1 mixture of epimers. The coupling yield
and subsequent photochemical deprotection proceeded in 59%
yield.
(22) Tchou, J.; Bodepudi, V.; Shibutani, S.; Antoshechkin, I.; Miller, J.;
Grollman, A. P.; Johnson, F. J. Biol. Chem. 1994, 269, 15318-15324.
(23) Karakaya, A.; Jaruga, P.; Bohr, V. A.; Grollman, A. P.; Dizdaroglu, M.
Nucleic Acids Res. 1997, 25, 474-479.
(24) Einholf, H. J.; Schnetz-Boutaud, N. C.; Guengerich, F. P. Biochemistry
1998, 37, 13300-13312.
(25) Maki, H.; Sekiguchi, M. Nature (London) 1992, 355, 273-275.
(26) Xia, Z.; Azurmendi, H. F.; Mildvan, A. S. Biochemistry 2005, 44, 15334-
15344.
(27) Tassotto, M. L.; Mathews, C. K. J. Biol. Chem. 2002, 277, 15807-15812.
(28) Luo, W.; Muller, J. G.; Rachlin, E. M.; Burrows, C. J. Chem. Res. Toxicol.
2001, 14, 927-938.
(29) Kornyushyna, O.; Berges, A. M.; Muller, J. G.; Burrows, C. J. Biochemistry
2002, 41, 15304-15314.
(30) Kino, K.; Sugiyama, H. Chem. Biol. 2001, 8, 369-378.
(31) Deng, L.; Scha¨rer, O. D.; Verdine, G. L. J. Am. Chem. Soc. 1997, 119,
7865-7866.
(32) Wiederholt, C. J.; Delaney, M. O.; Greenberg, M. M. Biochemistry 2002,
41, 15838-15844.
(33) Ober, M.; Muller, H.; Pieck, C.; Gierlich, J.; Carell, T. J. Am. Chem. Soc.
2005, 127, 18143-18149.
(34) Burgess, K.; Cook, D. Chem. ReV. 2000, 100, 2047-2059.
(35) Dinkel, C.; Schultz, C. Tetrahedron Lett. 2003, 44, 1157-1159.
(36) Haraguchi, K.; Greenberg, M. M. J. Am. Chem. Soc. 2001, 123, 8636-
8637.
â-C-Fapy‚dGMP was prepared from 6 (Scheme 4) using the
same strategy employed above for the synthesis of Fapy‚
(37) Haraguchi, K.; Delaney, M. O.; Wiederholt, C. J.; Sambandam, A.; Hantosi,
Z.; Greenberg, M. M. J. Am. Chem. Soc. 2002, 124, 3263-3269.
(38) Jiang, Y. L.; Wiederholt, C. J.; Patro, J. N.; Haraguchi, K.; Greenberg, M.
M. J. Org. Chem. 2005, 70, 141-147.
(39) Sekine, M.; Tsuruoka, H.; Iimura, S.; Kusuoku, H.; Wada, T. J. Org. Chem.
1996, 61, 4087-4100.
9
14608 J. AM. CHEM. SOC. VOL. 128, NO. 45, 2006

Formamidopyrimidine Nucleoside Triphosphate Effects
A R T I C L E S
Table 1. Specificity Constants for Hydrolysis of Nucleoside
Triphosphates by MutT
1
1
dNTP
kcat/Kma (M-1 s-
)
dNTP
kcat/Kma (M-1 s-
)
dGTP
(4.3 ( 0.9) × 104 â-C-Fapy‚dGTP (2.2 ( 0.5) × 102
Fapy‚dGTP (3.8 ( 1.3) × 102 8-OxodGTPb
(5.4 ( 0.7) × 106
a Results are the average ( standard deviation of at least three
experiments. b Taken from ref 26.
dGMP.40 The anomers of â-C-Fapy‚dGMP were separated by
anion exchange chromatography. The stereochemistry of the
â-anomer was determined via an NOE experiment in which a
2.8% enhancement of the pseudoanomeric proton was observed
when the respective 4′-proton was irradiated (Figure 1). Fol-
lowing synthesis from the monophosphate using the above-
described umpolung approach, â-C-Fapy‚dGTP was purified in
22% yield and characterized.
Fapy‚dGTP and â-C-Fapy‚dGTP Are Poor Substrates for
MutT. If Fapy‚dGTP and/or â-C-Fapy‚dGTP are to serve as
sources of the respective nucleotides in DNA, they must escape
hydrolysis by nucleotide hydrolases. MutT was chosen as a
potential hydrolase because of the biochemical similarity
between Fapy‚dG and OxodG. The latter is an excellent
substrate for this enzyme.14,21,41 Preliminary studies indicated
that the Km for Fapy‚dGTP and â-C-Fapy‚dGTP hydrolysis by
MutT was very high. Saturation was not achieved at triphosphate
concentrations >400 µM. To spare valuable triphosphate
substrates, we utilized a steady-state kinetic method that provides
the specificity constant by measuring the amount of product as
a function of time at a single concentration of substrate. This
simplification of Michaelis-Menten kinetics assumes that the
amount of enzyme-bound substrate is much smaller than the
total amount of enzyme in solution.42 This approximation is valid
when the substrate concentration is much lower than the Km.
Consequently, we measured the hydrolysis of Fapy‚dGTP and
â-C-Fapy‚dGTP at 20 µM (Table 1).43 Furthermore, to validate
the method, we determined the specificity constant for MutT
hydrolysis of dGTP (10 µM). The value of kcat/Km determined
for dGTP hydrolysis (Table 1) is within error of that reported
in the literature.26 MutT hydrolyzes Fapy‚dGTP and â-C-Fapy‚
dGTP 100-200 times less efficiently than even dGTP. These
data indicate that MutT will not protect the nucleoside tri-
phosphate pool against Fapy‚dGTP. Inefficient hydrolysis of
â-C-Fapy‚dGTP is also potentially significant. DNA containing
â-C-Fapy‚dG strongly binds the glycosylase that excises Fapy‚
dG.21 MutT’s reluctance to act on the C-nucleotide suggests
that â-C-Fapy‚dGTP will be useful as a base excision repair
enzyme inhibitor if the triphosphate is accepted as a substrate
by DNA polymerase(s) and an alternative nucleotide hydrolase
does not act on the molecule.10
Figure 2. Qualitative incorporation of â-C-Fapy‚dG (A) and Fapy‚dG (B)
at multiple positions in 8. Pause sites were identified using an independently
synthesized oligonucleotide ladder.
the kinetics.44,45 The kinetic behavior can be explained in one
of two ways. Either epimerization of the anomers is accelerated
in the presence of MutT or the two anomers are comparable
substrates for the enzyme. In the absence of evidence supporting
the former, mechanistic economy favors the latter.
Effect of Fapy‚dGTP and â-C-Fapy‚dGTP on Replication
by the Klenow Fragment of DNA Polymerase I. The ability
of Klenow to incorporate the formamidopyrimidine triphos-
phates and extend the primer following modified nucleotide
incorporation was qualitatively examined. Extension of the
radiolabeled primer in 8 in the presence of Fapy‚dGTP (200
µM) or â-C-Fapy‚dGTP (200 µM) and dATP, dCTP, and dTTP
(10 µM) was followed as a function of time (Figure 2). Greater
than 85% of the primer was extended to full-length product in
the presence of either nonnative nucleoside triphosphate. No
full-length product was observed in the absence of Fapy‚dGTP
or â-C-Fapy‚dGTP (data not shown). The template contained
3 2′-deoxycytidines in the single-stranded region, indicating that
Klenow was able to incorporate 3 of the modified nucleotides
in a short sequence (19 nucleotides). In addition, pause sites
were only observed at the position prior to incorporation of â-C-
Fapy‚dG. None were detected for Fapy‚dG incorporation under
these conditions. If extension of the primer following modified
nucleotide incorporation was strongly inhibited, additional pause
sites opposite dC would have been observed. Although full-
length extension of 8 required a higher concentration of Fapy‚
dGTP or â-C-Fapy‚dGTP than the native nucleoside triphos-
phate (dGTP, 10 µM, data not shown), the qualitative experiments
indicated that neither modified nucleotide blocked replication
by Klenow. The ability to tolerate multiple incorporations of
Fapy‚dG and â-C-Fapy‚dG in a growing primer distinguishes
these molecules from other nonnative nucleoside triphosphates,
in which extension of primers containing them at their 3′-termini
An additional observation worth noting is that although the
starting material is a 1:1 mixture of R- and â-anomers, the
hydrolysis of Fapy‚dGTP, which proceeds to ∼75% conversion
in 45 min, is fit to a single exponential. The reactions are
completed in too short a period of time for equilibration of the
anomers and preferential hydrolysis of one isomer to explain
(40) Delaney, M. O.; Greenberg, M. M. Chem. Res. Toxicol. 2002, 15, 1460-
1465.
(41) Greenberg, M. M. Biochem. Soc. Trans. 2004, 32, 46-50.
(42) Meyer-Almes, F.-J.; Auer, M. Biochemistry 2000, 39, 13261-13268.
(43) See the Supporting Information.
(44) Greenberg, M. M.; Hantosi, Z.; Wiederholt, C. J.; Rithner, C. D.
Biochemistry 2001, 40, 15856-15861.
(45) Burgdorf, L. T.; Carell, T. Chem.sEur. J. 2002, 8, 293-301.
9
J. AM. CHEM. SOC. VOL. 128, NO. 45, 2006 14609

A R T I C L E S
Imoto et al.
Table 2. Specificity Constants for Nucleoside Triphosphate
Incorporation by Klenow
Scheme 5. Determination of Relative Velocities for Incorporation
of a Modified Nucleotide and Extension
1
1
dNTP
X
kcat/Kma (M-1 min-
)
dNTP
X
kcat/Kma (M-1 min-
)
dGTP
dC (2.0 ( 0.3) × 107 Fapy‚dGTP
dA (4.8 ( 1.6) × 102
dA (1.2 ( 0.1) × 107
Fapy‚dGTP dC (1.2 ( 0.5) × 104 â-C-Fapy‚dGTP dA 26b
â-C-Fapy‚ dC (1.5 ( 0.4) × 103 dTTP
dGTP
dGTP
dA (4.3 ( 0.7) × 104
a Results are the average ( standard deviation of at least two experiments
containing three replicates. b The result is from a single experiment (three
replicates).
Table 3. Relative Kinetic Parameters for Nucleoside Triphosphate
Extension by Klenowa
(rel Vmax)/Km
(µ )
M-
by Klenow can be inefficient, and rarely does one observe
multiple incorporation sites in such a short region.46-50
Klenow incorporation of Fapy‚dG and â-C-Fapy‚dG and
subsequent extension were quantitatively evaluated under steady-
state conditions. Incorporation of the modified nucleotides in 9
was examined using a standing start process (Table 2).51
Preliminary experiments again established that the Km values
for the triphosphate substrates were very high. Consequently,
specificity constants were extracted from plots of product
formation as a function of time using a single concentration of
triphosphate.42,43 Experiments with Fapy‚dGTP were carried out
using 5 µM opposite dC and 100 µM when dA was the opposing
template nucleotide. The specificity constant for incorporation
of â-C-Fapy‚dG opposite dC was determined using three
different concentrations of the triphosphate (50-200 µM). The
small variation of the specificity constant indicates that the
experiments were carried out considerably below the Km and
justifies using the kinetic method.42 Insertion product grew in
linearly as a function of time, because even at the low
concentrations of triphosphates used the substrate was in large
excess relative to DNA. Hence, the substrate concentration did
not vary. Fapy‚dG and â-C-Fapy‚dG were incorporated ∼1000
and 10000 times less efficiently, respectively, than dG opposite
dC in the template. Klenow also incorporated the modified
nucleotides opposite dA considerably less efficiently than it did
dG. Moreover, Fapy‚dG and â-C-Fapy‚dG were incorporated
more than 40000-fold less efficiently opposite dA than was
dTTP. Incorporation of Fapy‚dG opposite dT or dG was not
observed under these conditions (data not shown). These data
indicate that if Fapy‚dG or â-C-Fapy‚dG are incorporated into
DNA via their respective nucleoside triphosphates, it will very
likely be opposite dC.
1
dNTP
rel Vmax
Km
(
µ
M)
Fapy‚dGTP
â-C-Fapy‚dGTP
6.4 ( 0.1
10.6 ( 3.2
1.7 ( 0.5
1.9 ( 0.8
4.6 ( 0.3
5.7 ( 0.6
a Results are the average ( standard deviation of at least two experiments
(three replicates).
ment. The ratio (Vrel) of the velocity for extending the primer
after incorporation of the modified nucleotide relative to the
velocity for its incorporation was obtained by measuring the
amounts of products (It-1, It, It+1) as a function of dATP
concentration (Scheme 5, Table 3).43,52 The radiolabeled primer
in 10 was extended in the presence of dCTP (10 µM) to prevent
removal of product by proofreading (“rescue nucleotide”) and
either Fapy‚dGTP (70 µM) or â-C-Fapy‚dGTP (200 µM). The
amount of product extended past Fapy‚dGTP or â-C-Fapy‚dGTP
depended on the concentration of dATP (1-10 µM). Primer
(It-2) extension was not observed in the absence of Fapy‚dGTP
or â-C-Fapy‚dGTP. The relative velocities (Vrel) were plotted
versus [dATP] and fitted via nonlinear regression analysis.51
These data indicate that primer extension is faster following
modified nucleotide incorporation than for incorporation of the
modification (rel Vmax > 1).52 Moreover, this implies that
Klenow does not treat the presence of Fapy‚dG or â-C-Fapy‚
dG like it does a mismatch (rel Vmax < 1), suggesting that these
modified nucleotides do not significantly perturb the structure
of the nascent duplex.
Conclusions
The kinetic data raise the possibility that if Fapy‚dGTP is
produced in the nucleoside triphosphate pool as a result of
oxidative stress, it could be a source of Fapy‚dG in duplex DNA.
The triphosphate is a kinetically competent source of Fapy‚dG
in DNA because it is a poor substrate for MutT, but is
incorporated opposite dC by Klenow, a model DNA polymerase.
Fapy‚dGTP is incorporated opposite dC ∼1000 times less
efficiently than is dGTP. If Fapy‚dGTP is present at even 1%
of the concentration level of dGTP, our kinetic data lead us to
estimate that the nucleotide would be incorporated opposite 1
every 100000 dC’s (10-3 × 10-2 ) 10-5) as a substitute for
dG. This would yield several Fapy‚dG lesions per million base
pairs of DNA and is of sufficient efficiency to achieve the levels
The qualitative observation that primer extension following
modified nucleotide incorporation was not strongly inhibited
(Figure 2) was confirmed using a running start kinetic experi-
(46) Leconte, A. M.; Matsuda, S.; Hwang, G. T.; Romesberg, F. E. Angew.
Chem., Int. Ed. 2006, 45, 4326-4329.
(47) Leconte, A. M.; Matsuda, S.; Romesberg, F. E. J. Am. Chem. Soc. 2006,
128, 6780-6781.
(48) Henry, A. A.; Olsen, A. G.; Matsuda, S.; Yu, C.; Geierstanger, B. H.;
Romesberg, F. E. J. Am. Chem. Soc. 2004, 126, 6923-6931.
(49) Morales, J. C.; Kool, E. T. Nat. Struct. Biol. 1998, 5, 950-954.
(50) Moran, S.; Ren, R. X. F.; Kool, E. T. Proc. Natl. Acad. Sci. U.S.A. 1997,
94, 10506-10511.
(51) Creighton, S.; Bloom, L. B.; Goodman, M. F. Methods Enzymol. 1995,
262, 232-256.
(52) Creighton, S.; Goodman, M. F. J. Biol. Chem. 1995, 270, 4759-4774.
9
14610 J. AM. CHEM. SOC. VOL. 128, NO. 45, 2006

Formamidopyrimidine Nucleoside Triphosphate Effects
A R T I C L E S
of Fapy‚dG detected in cellular DNA.12,13 Admittedly, this
estimate is based upon data obtained using a bacterial enzyme.
However, the Klenow fragment of DNA polymerase I from E.
coli is often used to study DNA lesions. In view of the fact that
it is less certain that 8-OxodGTP is the source of dG f dT
transversions, these data suggest that one should consider Fapy‚
dGTP as the source of these mutations.27
have advantages over other base excision repair inhibitors
because it is a substrate for a DNA polymerase.31
Acknowledgment. We are grateful for support of this research
by the National Cancer Institute of the NIH (Grant CA-74954)
to M.M.G. We thank Professor Maurice Bessman for providing
MutT and Dr. Ranabir Sinha Roy for bringing ref 42 to our
attention.
The interaction of â-C-Fapy‚dGTP with MutT and the DNA
polymerase suggests that the triphosphate could be useful as
an inhibitor of base excision repair. It is known that when the
nonhydrolyzable analogue â-C-Fapy‚dG is in DNA, it strongly
binds the repair enzyme (Fpg) responsible for removing many
purine lesions.21 Provided â-C-Fapy‚dGTP could be delivered
or produced in cells (i.e., via a prodrug), this molecule would
Supporting Information Available: Experimental procedures
for the syntheses of Fapy‚dGTP and â-C-Fapy‚dGTP and
enzyme experiments, NMR spectra of previously unreported
compounds, and sample kinetic plots. This material is available
free of charge via the Internet at http://pubs.acs.org.
JA065525R
9
J. AM. CHEM. SOC. VOL. 128, NO. 45, 2006 14611