Dissecting the differences between the α and β anomers of the oxidative DNA lesion FaPydG

Article abstract of DOI:10.1002/chem.200701373

The oxidative DNA lesion, FaPydG rapidly anomerizes to form a mixture of the α and β anomer. To investigate the mutagenic potential of both forms, we prepared stabilized bioisosteric analogues of both configurational isomers and incorporated them into oligonucleotides. These were subsequently used for thermodynamic melting-point studies and for primer-extension experiments. While the β compound, in agreement with earlier data, prefers cytidine as the pairing partner, the α compound is not able form a stable base pair with any natural base. In primer-extension studies with the high-fidelity polymerase Bst Pol I, the polymerase was able to read through the lesion. The β compound showed no strong mutagenic potential. The α compound, in contrast, strongly destabilized DNA duplexes and also blocked all of the tested DNA polymerases, including two low-fidelity polymerases of the Y-family.

Full text of DOI:10.1002/chem.200701373

FULL PAPER
DOI: 10.1002/chem.200701373
Dissecting the Differences between the a and b Anomers of the Oxidative
DNA Lesion FaPydG
Florian Büsch,[a] J. Carsten Pieck,[a] Matthias Ober,[a] Johannes Gierlich,[a]
Gerald W. Hsu,[b] Lorena S. Beese,[b] and Thomas Carell*[a]
Abstract: The oxidative DNA lesion,
FaPydG rapidly anomerizes to form a
mixture of the a and b anomer. To in-
vestigate the mutagenic potential of
both forms, we prepared stabilized bio-
isosteric analogues of both configura-
tional isomers and incorporated them
into oligonucleotides. These were sub-
sequently used for thermodynamic
melting-point studies and for primer-
extension experiments. While the b
compound, in agreement with earlier
data, prefers cytidine as the pairing
partner, the a compound is not able
form a stable base pair with any natu-
ral base. In primer-extension studies
with the high-fidelity polymerase Bst
Pol I, the polymerase was able to read
through the lesion. The b compound
showed no strong mutagenic potential.
The a compound, in contrast, strongly
destabilized DNA duplexes and also
blocked all of the tested DNA poly-
merases, including two low-fidelity
polymerases of the Y-family.
Keywords:
DNA repair
DNA
lesions
DNA
·
·
·
DNA ·
mutagenesis · oxidative stress
Introduction
Scheme 1, to give a mixture of the a and b anomers 1 and
3.[4] Additionally, hydrolysis of the glycosidic bond can occur
at elevated temperatures. So far, it is neither known if such
an anomerization occurs in DNA, nor has the reactivity of
One of the most common DNA lesions that is found in cells
after oxidative stress or g radiation[1,2] is the 2,6-diamino-4-
hydroxy-5-formamidopyrimi-
dine (FaPydG).[3] This lesion 1
is a hydrolysis product of gua-
nine that features an opened
imidazole ring, which strongly
increases the reactivity and
flexibility of the compound.
This structural element also
leads to rapid anomerization of
Scheme 1. Anomerization of the b-FaPydG lesion 1 via the open intermediate 2 to its configurational a-isomer
3.
1
via the ring-opened inter-
mediate 2, as depicted in
the lesion been fully explored. Because repair of a-config-
[a] Dipl.-Chem. F. Büsch, Dipl.-Ing. J. C. Pieck, Dipl.-Chem. M. Ober,
Dr. J. Gierlich, Prof. Dr. T. Carell
ured nucleobases is a well-known fact, it is essential to study
the mutagenic effect of the maybe rare but important a ver-
sions of lesions.
Center for Integrative Protein Science (CiPSM)
at the Department of Chemistry and Biochemistry
Ludwig-Maximilians University Munich
To clarify the mutagenic potential of the FaPydG lesion,
several attempts to prepare the compound have been re-
ported. Cadet et al. obtained the formamidopyrimidine
lesion by direct irradiation of an oxygen-free aqueous solu-
tion that contained the nucleotide dG with gamma rays.[5]
They obtained an a/b mixture of the pyranosyl-FaPydG in
low yields, which was the first direct proof of the instability
of the compound. Greenberg et al. reported an elegant syn-
Butenandtstrasse 5–13, Haus F, 81377 Munich (Germany)
Fax : (+49)89-218077756
E-mail: Thomas.Carell@cup.uni-muenchen.de
[b] Dr. G. W. Hsu, Prof. Dr. L. S. Beese
Department of Biochemistry, Duke University Medical Center
Durham, North Carolina 27710 (USA)
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
Chem. Eur. J. 2008, 14, 2125 – 2132
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2125

thesis of the lesion, and this groupwas also able to incorpo-
rate it into DNA.[6–8] To address the question of which
anomer is finally present in DNA, a biochemical repair
study was performed, which suggested that the lesion exists
in DNA mainly as the b anomer.[7,8] Experiments by our
groupto introduce the lesion into DNA by solid-phase syn-
thesis were thwarted by the rapid anomerization of the com-
pound during its synthesis or handling. In all cases, we ob-
tained a mixture of both anomers, particularly in DNA
single strands.[9] To circumvent
tential of the a-configured analogue by using high and low-
fidelity polymerases.
Results and Discussion
Synthesis: Synthesis of a-cFaPydG was achieved in about 15
steps as depicted in Scheme 2. The cyclopentane core 4 was
available in only four steps in good yields.[14,15] This inter-
this
problem,
we[10]
and
others[11,12] developed different
non-hydrolyzable derivatives of
the FaPy-lesion. To minimize
the influence of the modifica-
tion on the base-pairing proper-
ties of the lesion, we designed a
stabilized isosteric analogue of
the b anomer (b-cFapydG), in
which the oxygen atom in the
ribose ring is replaced by a
methylene unit. This eliminates
any anomerization reaction.
The strategy also allows the
synthesis of both “anomers” in
a pure form for a direct com-
parison, as reported here. In
recent studies, b-cFapydG was
incorporated into DNA and its
influence on DNA structure
was evaluated.[10] In addition, a
crystal structure of the stabi-
lized lesion in DNA in complex
with the FPG repair enzyme
could be obtained; this shows
that the preparation of bioiso-
steric lesion analogues, such as
b-cFaPydG and a-cFaPydG,
provides detailed information
about the biological and bio-
physical properties of fragile
DNA lesions.[13]
Scheme 2. Synthesis of the a-cFaPydG phosphoramidite building block 15. a) p-TsCl, pyridine, RT, 18 h, 79%;
b) NaN3, DMF, 608C, 16 h, 94%; c) BCl3, CH2Cl2, ꢀ788C!RT, 99%; d) TBDMSCl, imidazole, DMF, RT,
16 h, 86%; e) 10% Pd/C, H2, EtOH, 15 h, quant.; f) DIEA, 10, 30 min RT, then 45 min at 608C, 71%; g) 10%
Pd/C, H2, EtOH/CH2Cl2 6:1, EDC·HCl, formic acid, THF/pyridine 5:3, RT, 10 h, 79%; h) HF–pyridine com-
plex, pyridine, THF, RT, 15 h, quant.; j) DMTCl, pyridine, RT, 3.5 h, 74%; k) 2-cyanoethyl-tetraisopropylphos-
phordiamidite, tetrazolate, CH3CN, RT, 6 h, 84%. DIEA=diisopropylethylamine.
mediate was subsequently transformed into azide 6 in two
steps, which included the tosylation of the hydroxyl group to
give 5, followed by a SN2 reaction with sodium azide. Depro-
tection of the benzyl ethers with boron trichloride in di-
chloromethane furnished compound 7.
After protecting the free hydroxyl groups as tert-butyldi-
methylsilanes (TBDMS) to afford compound 8, the azide
function was reduced to the amine 9. Compound 9 was sub-
sequently coupled with heterocycle 10[9,16] to give the lesion
precursor 11. The absolute configuration of this compound
Here, we report the synthesis of the a anomer of the
FaPydG lesion as its stabilized carbacyclic analogue, a-cFa-
PydG. We report the pairing properties of the a analogue
and primer-extension studies to decipher the mutagenic po-
was verified by using NOE NMR spectroscopy. Reduction
of the nitro groupof 11 and subsequent formylation of the
amine intermediate to 12 required stringently anhydrous
and anaerobic conditions. The best yields were obtained
2126
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2125 – 2132

FaPydG Lesions
FULL PAPER
with
N-(3-dimethylaminopropyl)-N’-ethylcarbodiimide
(EDC) and formic acid. After completing the synthesis of
the silyl-protected lesion nucleotide analogue, compound 12
had to be transformed into the required solid-phase synthe-
sis building block. This involved deprotection of the hydrox-
yl groups in 12 with HF–pyridine complex in THF to yield
diol 13. Additional pyridine was needed for the transforma-
tion to protect the acid-labile formyl group from cleavage.
Compound 13 was finally converted into the dimethoxytrityl
(DMT)-protected phosphoramidite 15 by using standard
procedures. With this building block in hand, the stage was
set to perform solid-phase synthesis. The lability of the
formyl group, however, required us to change the standard
procedures. The coupling time of building block 15 was in-
creased, and the typical capping reagents, phenoxyacetyl or
acetic anhydride, were changed to isobutyryl anhydride to
avoid any transamidation of the formamidopyrimidine.
Cleavage of the DNA from the solid support was performed
with concentrated ammonia in water/ethanol 3:1 at 158C
overnight. All oligonucleotides were purified by reverse-
phase HPLC and characterized by ESI-FTICR mass spec-
trometry. The sequences of the oligonucleotides that were
prepared for this study are listed in Table 1. DNA that con-
tained b-cFaPydG was prepared by following previously
published procedures.[9,10]
Figure 1. HPLC chromatogram and mass spectrum of an enzymatic
digest of ODN 1. The peak at 10.1 min was assigned to a-cFaPydG by co-
injection and by mass spectrometry.
Thermodynamic measurements: After successful incorpora-
tion into oligonucleotides, the influence of the pseudo-a
anomer on duplex stability and its base-pairing properties
were addressed. Two strands (ODN 2 and ODN 3) were hy-
bridized with four different counter strands. The counter
strands contained all four of the canonical nucleobases op-
posite the lesion to form all possible base pairs. UV melting
points of these duplexes were measured with DNA concen-
trations ranging from 0.3 to 27 mm. The thermodynamic pa-
rameters DH8, DS8 and DGꢁ298 K of the duplex dissociation
process were determined by using van’t Hoff plots (see the
Table 1. DNA sequences that were prepared for melting-point studies
and primer-extension investigations. X=a-cFaPydG; Y=b-cFaPydG.
Name
Sequence
ODN 1
ODN 2
ODN 3
ODN 4
ODN 5
5’-CTC TTT X TTT CTC G
5’-GCG ATX TAG CG
5’-TGC AGT XAC AGC
5’-TAC XCC TGG TCA TT
5’-TAG YCC TGG TCA TT
Supporting Information).[17] For
a better comparison,
Table 2 also contains data for b-cFaPydG, which were taken
from previous work.[10]
Enzymatic digest: To clarify that the correct incorporation
of a-cFaPydG into oligonucleotides has taken place, an en-
zymatic digest of the DNA strand followed by HPLC-MS/
MS characterization is needed. For this experiment, DNA
(ODN 1) was treated with four different enzymes, which are
able to cleave the DNA into the corresponding nucleosides.
The obtained mixture was subsequently separated by HPLC
on a reversed-phase column that was coupled to a FTICR
mass spectrometer. In the HPLC chromatogram that is de-
picted in Figure 1, three sharp signals are seen, which could
be assigned, based on the molecular weights, to the three
normal nucleosides (dC, dG and dT) that are present in the
strand. The signal at 10.1 min possessed the same molecular
weight and retention time as the deacetylated derivative of
compound 13, which is the compound that is expected to be
the product from the enzymatic digest of an a-cFaPydG-
containing oligonucleotide. This result proves that a-cFa-
PydG was not modified during DNA synthesis or purifica-
tion. The cis/trans-isomerization of the formamide on the
column is responsible for the rather broad appearance of
the signal.
From the data listed in Table 2, it is evident that the pres-
ence of either of the two cFaPydG lesion anomers in DNA
induces large duplex destabilizations, but while b-cFaPydG
still favours base pairing with dC (DGꢁ298 K =ꢀ10.2 kcal
molꢀ1), no preferred pairing partner was observed in the
case of a-cFaPydG. The absolute value of DGꢁ298 K for the
melting process of duplexes that contain a a-cFaPydG:dN
base pair is strongly reduced compared even to a dG:dA
mismatch;[9] a-cFaPydG induces a large destabilization of
the DNA duplex regardless of the opposing base, and is not
able to form any stabilizing interaction with any of the can-
onical bases.
Primer-extension experiments: To measure the thermody-
namic properties of the lesion in double-stranded DNA
alone is not sufficient to evaluate how it influences DNA
replication, because the basis for the replication fidelity is
the base-pairing situation inside the polymerase. This, how-
ever, is influenced by the steric constraints that are estab-
lished by the active site of the DNA polymerase. To investi-
gate the mutagenic potential, we prepared a DNA template/
Chem. Eur. J. 2008, 14, 2125 – 2132
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2127

T. Carell et al.
Table 2. Thermodynamic data for two different double strands, containing either a- or b-cFaPydG with all natural bases as pairing partners.[a] For b-cFa-
PydG, see also reference [10].
5’-GCGATXTAGCG
3’-CGCTAYATCGC
5’-TGCAGTXACGC
3’-ACGTCAYTGCG
base pair[a] X:Y
DGꢁ298 K
DH8
DS8
DGꢁ298 K
DH8
DS8
[kcalmolꢀ1
]
[kcalmolꢀ1
]
[calmolꢀ1 Kꢀ1
]
[kcalmolꢀ1
]
[kcalmolꢀ1
]
[calmolꢀ1 Kꢀ1
]
[b]
[b]
[b]
[b]
a-cFaPydG:dA
a-cFaPydG:dC
a-cFaPydG:dG
a-cFaPydG:dT
b-cFaPydG:dA
b-cFaPydG:dC
b-cFaPydG:dG
b-cFaPydG:dT
ꢀ7.0
ꢀ7.7
ꢀ7.7
ꢀ7.8
ꢀ7.0
ꢀ10.2
ꢀ6.9
ꢀ8.3
ꢀ44ꢂ1.1
ꢀ41ꢂ1.7
ꢀ40ꢂ2.0
ꢀ38ꢂ1.6
ꢀ42ꢂ2.0
ꢀ51ꢂ1.0
ꢀ43ꢂ2.0
ꢀ47ꢂ1.0
ꢀ125ꢂ7.8
ꢀ111ꢂ4.7
ꢀ109ꢂ3.2
ꢀ100ꢂ6.1
ꢀ118ꢂ7.0
ꢀ137 ꢂ3.2
ꢀ121ꢂ6.8
ꢀ129ꢂ3.4
ꢀ10.4
ꢀ10.7
ꢀ11.1
ꢀ11.4
ꢀ10.4
ꢀ13.4
ꢀ9.9
ꢀ70ꢂ1.3
ꢀ64ꢂ1.9
ꢀ74ꢂ1.0
ꢀ77ꢂ1.1
ꢀ64ꢂ1.6
ꢀ80ꢂ1.8
ꢀ61ꢂ1.1
ꢀ68ꢂ1.1
ꢀ201ꢂ8.0
ꢀ180ꢂ6.5
ꢀ211ꢂ7.2
ꢀ219ꢂ3.7
ꢀ181ꢂ5.0
ꢀ222ꢂ5.7
ꢀ173ꢂ3.7
ꢀ193ꢂ3.7
ꢀ10.8
[a] Conditions: 150 mm NaCl, 10 mm Tris-HCl (pH 7.4, coligo =0.3–27 mm). [b] The error was calculated by propagation of the standard error of the linear
regression.
primer complex (Figure 2) that contains one of the two com-
pounds, b-cFaPydG (ODN 5) or a-cFaPydG (ODN 4), at
position X, respectively.
that contains b-cFaPydG with dATP and dCTP. The primer
strand was detected with a retention time of 25.2 min, while
the elongation products appear at a slightly longer retention
time. An efficient elongation is
clearly taking place in the pres-
ence of dCTP, which is nicely
inserted opposite b-cFaPydG.
This is in full agreement with
our earlier data.[10] The increas-
ing n+2 peak in the right part
of Figure 2 is caused by the se-
quence of the template strand
ODN 5. At the position after
the lesion analogue, a dG base
is instructing the incorporation
of a second dCTP, so that in
our case, the primer is elongat-
ed stepwise by two bases: First
the incorporation opposite the
b-cFaPydG lesion is observed,
then a second dC is incorporat-
Figure 2. Capillary electrophoresis coupled to laser-induced fluorescence detection of the elongation reaction
of a fluorescent-labelled primer with a b-cFaPydG that contains a template by using Bst Pol I as the poly-
merase (conditions: 0.2 mm primer, template 0.4 mm ODN 5, 3 nm Bst Pol I, 0.2 mm dATP or dCTP).
ed in a regular extension step.
To avoid the n+2 peak, we
changed the base that follows
the lesion to dC, and indeed
The template was hybridized to the fluorescently (Fl) la-
belled primer so that the first base that is read by the poly-
merase for replication is the DNA lesion analogue (standing
start conditions).[18,19] For the analysis of the fluorescence-la-
belled primer and its extension products, capillary electro-
phoresis that was coupled to a laser-induced fluorescence
detector was employed. This allows exact quantification of
the extension products with single-base resolution.
In the first experiments, we studied how the lesion ana-
logues influence the fidelity of the polymerase Bst Pol I
from Geobacillus stearothermophilus. The experiment was
conducted by using 0.2 mm primer, 0.4 mm template, 3 nm Bst
Pol I and 0.2 mm dNTP at 258C. Every 5 min, aliquots from
the assay were quenched and subsequently analyzed by
using capillary electrophoresis. Figure 2 depicts the chroma-
tograms of the primer extension reaction with the template
only one incorporation step opposite the lesion analogue is
observed (data not shown). In experiments in which dATP
was used instead of dCTP, one observes only very inefficient
incorporation of dATP opposite the b-cFaPydG lesion.
These results are in agreement with a recent primer-exten-
sion study with a FaPydG triphosphate, which showed pre-
ferred incorporation of the FaPydG triphosphate opposite
dC.[20] The observation is also in agreement with our own
primer-extension data,[10] and with recent in vivo studies.[21]
In summary, b-cFaPydG has very low mutagenic potential in
the replication process.
To our surprise, when the primer extension was conducted
under the same conditions by using the DNA template that
contains a-cFaPydG (ODN 4), no primer extension could be
observed. Even extensive screening of different enzyme con-
centrations (1, 5, 10, 100 nm) at 50 mm dNTP revealed no
2128
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2125 – 2132

FaPydG Lesions
FULL PAPER
elongation within 3.0 min of the assay time regardless of the
dNTP that was used. We also studied the low-fidelity poly-
merases Pol h and DinB to investigate if the blocking char-
acter of a-cFaPydG is limited to high-fidelity polymerase.
These polymerases are able read through certain DNA le-
sions, and belong to the recently discovered Y-family of
polymerases.[22] To our surprise, the same blocking effect
was also observed with S. cerevisiae Pol h and with DinB
from Geobacillus stearothermophilus. Even after increasing
the enzyme concentration by a factor of 100, no primer ex-
tension was obtained; this shows that the a anomer is a very
efficient replication blocker for high- and low-fidelity poly-
merases.
merase, and it fully retains the coding potential of dG. The
b version of the lesion is, therefore, not strongly mutagenic.
In the case of a-cFaPydG, we observed different properties:
The thermodynamic data show that this anomer strongly de-
stabilizes the duplex irrespective of the counter base. One
possible explanation for this behaviour is that the a-config-
ured base might be outside the helix, and, therefore, unable
to form a base pair inside the base stack. In primer exten-
sion studies, a-cFaPydG turned out to be a strong blocking
unit for high- (Bst Pol I) and low-fidelity polymerases (DinB
and Pol h), but if the enzyme, triphosphate concentrations
and the reaction time are strongly increased, dCTP is inco-
porated by Bst Pol I.
We finally decided to drastically increase the reaction
time and the triphosphate concentration. Indeed at a final
concentration of 6.5 mm Bst Pol I (about 2100 times higher
than in case of the b-cFaPydG), 2 mm dNTPs and 1.5 h reac-
tion time we could measure an elongation past the lesion
(see Figure 3). After 90 min reaction time, 55% of the
The presented data allow a direct comparison between
the a and b version of a lesion that is based on the studies
of appropriate model compounds for the first time. From
these analogue studies, it is clear that the b version of the
lesion is neither blocking nor mutagenic, but the a version is
mainly blocking. Because the role of the anomerization of
the natural FaPydG lesion is still unknown in biological sys-
tems, the results presented here indicate that the rate of
anomerization is an important parameter that needs to be
investigated in more detail in duplex DNA. This is in ac-
cordance with recent in vivo experiments in which a mixture
of both anomers was employed.
Experimental Section
Chemicals were purchased from Sigma–Aldrich, Acros or Lancaster and
were used without further purification. Solvents were of reagent grade
and were purified by standard methods. The reactions were monitored
on Merck Silica 60 F254 TLC plates. Detection was carried out by irradia-
tion with UV light (254 nm) and staining with 0.5% KMnO4 in aq.
NaOH (1n) or acidic anisaldehyde solution. Flash chromatography was
performed on Silica 60 (Merck, 230–400 mesh). NMR spectra were re-
corded on the following spectrometers: Varian Oxford 200, Bruker
AC 300, Varian XL 400 and Bruker AMX 600. Chemical shifts (d) are
given in ppm, the coupling constants (J) in Hz. Mass spectra were record-
ed on the following machines: Finnigan MAT 95 (EI), Bruker Autoflex II
(MALDI-TOF) and Thermo Finnigan LTQ-FT (ESI-ICR). IR spectra
were measured on a Nicolet 510 FTIR spectrometer in a KBr matrix or
with a diamond-ATR (attenuated total reflection) setup. DNA synthesis
was performed on a PerSeptive Biosystems Expedite 8900 Synthesizer
and an ¾kta Oligopilot 10 (Amersham Biosciences). Analysis and purifi-
cation of the oligonucleotides was performed on Merck LaChrome
HPLC systems with UV and diode array detectors by using 5 m silica-C18
RP columns and 0.1m NEt3/AcOH in H2O/MeCN as eluent. UV spectra
and melting points were measured on a Cary 100 UV/Vis spectrometer
by using 1 mL quartz cuvettes with 1 cm pathlength. The samples con-
tained NaCl (150 mm), Tris-HCl (10 mm, pH 7.4) and 0.3–27 mm of each
oligonucleotide. For every strand, five temperature cycles from 85 to 08C
were recorded. The melting point was calculated computationally by
using Microcal Origin. ESI spectra of DNA strands were measured in
flow injection analysis mode or coupled to chromatographic separation
(eluent: 2 mm NEt3/AcOH in H2O/MeCN). In flow injection mode, a
2 mL sample (30 mM DNA, 100 mm NH4OAc) was injected in a steady
flow of H2O/MeCN (8:2; 200 mLminꢀ1). The capillary temperature was
3008C, spray voltage 4–5 kV (negative mode).
Figure 3. Capillary electrophoresis with laser-induced fluorescence detec-
tion of the elongation reaction of a fluorescent-labelled primer with a-
cFaPydG that contained template by using Bst Pol I (conditions: 0.2 mm
primer, 0.4 mm template ODN 4, 6.5 mm Bst Pol I, 2 mm dNTP).
primer starting material was elongated under these enforced
conditions with dCTP, but only 28% was elongated with
dATP. We observed 10% elongation if we used dGTP, and
only 4% elongation was observed with dTTP. To our sur-
prise, dCTP remained the “best” triphosphate, followed by
dATP. Under these extremely harsh conditions, the high-fi-
delity polymerase is able to read through the a-cFaPydG
compound. dCTP, and to a lesser extent dATP, are preferen-
tially inserted.
Outlook and Conclusions
In this publication, we report the synthesis of the a-cFa-
PydG cyclopentane analogue of the FaPydG lesion, its in-
corporation into DNA strands, base-pairing properties and
finally primer-extension data. As shown by previous studies
from our groupand other laboratories, the favoured base
pair of the b-cFaPydG lesion is formed with dC. The b isom-
er destabilizes the DNA, but it does not stopthe oply-
AHCTREUNG
4
Chem. Eur. J. 2008, 14, 2125 – 2132
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2129

T. Carell et al.
Toluenesulfonyl chloride (6.4 g, 33.6 mmol) was added in portions over a
period of 30 min. After the addition, the reaction was stirred at room
temperature for 18 h. The suspension was diluted with ethyl acetate
(300 mL) and H2O (200 mL). The organic phase was separated, washed
with sat. NH4Cl solution (3”200 mL), brine (100 mL) and dried over
MgSO4. The solvent was evaporated and the residue was purified by
flash chromatography (silica gel (500 mL), hexane/ethyl acetate 5:1) to
give 5 (8.2 g, 17.6 mmol, 79%) as a colourless oil. Rf =0.26 (20% ethyl
acetate in hexane); 1H NMR (400 MHz, CDCl3): d=7.79 (m, 2H), 7.38–
7.27 (m, 12H), 5.05–4.99 (m, 1H), 4.50 (s, 2H), 4.45 (s, 2H), 3.96–3.92
(m, 1H), 3.48–3.40 (td, 2H, J=6.3 Hz), 2.45 (s, 3H), 2.33–2.20 (m, 2H),
2.14–2.01 (m, 2H), 1.69–1.62 ppm (m, 1H); 13C NMR (100 MHz, CDCl3):
d=144.5, 138.2 (2C), 129.7, 128.3 (2C), 127.7 (4C), 127.5 (4C), 127.4
(4C), 82.1, 80.1, 73.0, 71.6, 71.2, 44.0, 38.8, 34.0, 21.5 ppm; IR (ATR): n˜ =
3436 (br), 3063 (s), 3031 (m), 2927 (m), 2862 (m), 1598 (m), 1454 (m),
1360 (s), 1189 (s), 1175 (vs), 1096 (s), 988 (m), 894 (s), 668 cmꢀ1; HRMS
(ESI+): m/z: calcd for C27H30O5S: 466.1814; found: 466.1799 [M]+.
4.4 Hz, 1H), 1.90–1.74 (m, 2H), 1.72–1.63 (m, 1H), 0.88 (s, 18H), 0.05 (s,
6H), 0.03 ppm (s, 6H); 13C NMR (100 MHz, CDCl3): d=73.1, 62.6, 59.5,
49.1, 41.2, 32.7, 25.9 (3C), 25.8 (3C), 18.3, 18.0, ꢀ4.6, ꢀ4.8, ꢀ5.4,
ꢀ5.5 ppm; IR (ATR): n˜ =2956 (s), 2930 (s), 2896 (m), 2858 (s), 2098 (s),
1472 (m), 1464 (m), 1389 (w), 1361 (w), 1257 (s), 1108 (s), 1006 (m), 939
(w), 874 (m), 836 (vs), 813 (m), 776 (s), 669 cmꢀ1 (w); HRMS (ESI+):
m/z: calcd for C18H39N3O2Si2: 385.2581; found: 385.2579 [M]+.
AHCTRE(UGN 1S,3S,4R)-3-(tert-Butyldimethylsilanyloxy)-4-[(tert-butyldimethylsilanyl-
oxy)methyl]cyclopentane-1-amine (9): 10% Pd/C (300 mg) was added to
a suspension of compound 8 (2.94 g, 7.6 mmol) in dry EtOH (20 mL).
The reaction was evacuated twice to exchange the inert gas atmosphere,
and then connected to two balloons that were filled with H2. The suspen-
sion was vigorously stirred at room temperature for 15 h. The palladium
catalyst was removed by using a PTFE-Filter (Whatman Puradisc) and
the solvent was removed under reduced pressure. The amine 9 (2.73 g,
6.0 mmol, quant.) was obtained as a colourless liquid, and was used with-
out further purification. Rf =0.31 (10% MeOH in CHCl3); 1H NMR
(400 MHz, CDCl3): d=4.04 (dd, J=10.6, 5.1 Hz, 1H), 3.47 (ddd, J=28.0,
10.0, 5.4 Hz, 2H), 3.29 (ddd, J=11.7, 6.8, 4.9 Hz, 1H), 2.21–2.12 (m, 1H),
2.03 (td, J=12.8, 6.3, 6.3 Hz, 1H), 1.75–1.67 (m, 3H), 1.61 (ddd, J=13.5,
9.1, 4.4 Hz, 1H), 1.44 (td, J=13.0, 5.2, 5.2 Hz, 1H), 0.87 (s, 9H), 0.86 (s,
9H), 0.04 (s, 3H), 0.03 (s, 3H), 0.02 ppm (s, 6H); 13C NMR (100 MHz,
CDCl3): d=74.9, 64.0, 51.2, 49.6, 45.1, 37.7, 25.9 (3C), 25.8 (3C), 18.3,
18.0, ꢀ4.6, ꢀ4.8, ꢀ5.4, ꢀ5.5 ppm; IR (ATR): n˜ =2956 (s), 2930 (s), 2887
(m), 2858 (s), 1472 (m), 1464 (m), 1389 (w), 1361 (w), 1257 (s), 1103 (s),
1006 (m), 939 (w), 874 (m), 836 (vs), 813 (m), 775 (s), 668 cmꢀ1 (w);
HRMS (ESI+): m/z: calcd for C18H41NO2Si2: 359.2676; found: 359.2679
[M]+.
ACHTREUNG(1S,3S,4R)-3-(Benzyloxy)-4-[(benzyloxy)methyl]cyclopentane-1-azide (6):
Compound 5 (8.0 g, 17.1 mmol) was dissolved in dry DMF (30 mL) and
NaN3 (12.2 g, 18.8 mmol) was added. The mixture was stirred at 608C for
14 h. After the addition of ethyl acetate (300 mL), the organic layer was
washed with sat. NaHCO3 solution (2”100 mL) and brine (100 mL) and
dried over MgSO4. The solvent was removed under reduced pressure.
The residue was purified by flash chromatography (silica gel (100 mL),
hexane/ethyl acetate/CHCl3 10:1:0.5) to give 6 (5.4 g, 16 mmol, 94%) as
a
colourless oil. Rf =0.54 (20% ethyl acetate in hexane); 1H NMR
(400 MHz, CDCl3): d=7.39–7.25 (m, 10H), 4.52 (dd, J=26.0, 18.6 Hz,
2H), 4.49 (s, 2H), 3.97–3.89 (m, 1H), 3.86 (td, J=7.0, 5.1 Hz, 1H), 3.44
(d, J=5.5 Hz, 2H), 2.54–2.43 (m, 1H), 2.24 (td, J=13.8, 6.8 Hz, 1H),
1.99 (dddd, J=13.4, 8.7, 4.7, 1.3 Hz, 1H), 1.84 ppm (dddd, J=28.2, 20.8,
9.5, 4.2 Hz, 2H); 13C NMR (100 MHz, CDCl3): d=138.2 (2C), 129.7
(2C), 128.3 (2C), 127.7 (2C), 127.5 (3C), 127.4, 80.1, 73.0, 71.6, 71.2,
44.0, 38.8, 34.0, 21.5 ppm; IR (ATR): n˜ =3064 (w), 3030 (w), 2931 (m),
2858 (m), 2097 (vs), 1730 (w), 1491 (w), 1454 (m), 1360 (m), 1263 (m),
1206 (w), 1097 (s), 1028 (m), 736 (s), 697 cmꢀ1; HRMS (ESI+): m/z: calcd
for C20H23N3O2: 337.1790; found: 337.1591 [M]+.
N-{4-[3-(tert-Butyldimethylsilanyloxy)-4-[(tert-butyldimethylsilanyloxy)-
methyl]cyclopentylamino]-5-nitro-6-oxo-1,6-dihydropyrimidin-2-yl}acet-
amide (11): The amine 9 (2.73 g, 7.6 mmol) was dissolved in dry DMF
(50 mL) and treated with diisopropylethylamine (1.5 mL, 8.4 mmol) and
the heterocycle chloride 10 (1.77 g, 7.6 mmol, 1 equiv). This reaction was
first stirred at room temperature for 30 min then at 608C for 45 min.
TLC analysis with ninhydrin staining indicated the completion of the re-
action. The mixture was diluted with ethyl acetate (300 mL) and the or-
ganic phase was washed with brine (3”). After drying over MgSO4, the
solvent was evaporated and the residue was dried overnight under high
vacuum. Purification by flash chromatography (silica gel (500 mL), 5%
MeOH in CHCl3) gave 11 (3.0 g, 5.4 mmol, 71%) as a yellow powder.
Rf =0.38 (10% MeOH in CHCl3); m.p. 220–2258C; 1H NMR (400 MHz,
CDCl3): d=11.43 (brs, 1H), 10.36 (brs, 1H), 10.00 (d, J=7.8 Hz, 1H),
4.62–4.52 (m, 1H), 4.18 (td, J=5.5, 2.8, 2.8 Hz, 1H), 3.51 (dd, J=10.1,
5.8 Hz, 1H), 3.38 (dd, J=10.1, 6.1 Hz, 1H), 2.29 (s, 3H), 2.22 (m, 1H),
2.00 (ddd, J=13.7, 6.7, 5.6 Hz, 1H), 1.90–1.82 (m, 1H), 1.71 (m, 2H),
0.87 (s, 18H), 0.07 (s, 3H), 0.06 (s, 3H), 0.03 (s, 3H), 0.02 ppm (s, 3H);
13C NMR (100 MHz, CDCl3): d=172.6, 157.4, 155.2, 149.8, 112.9, 75.6,
64.0, 52.4, 49.9, 41.7, 35.0, 25.9 (6C), 18.3, 18.2, ꢀ4.6 (2C), ꢀ5.4,
ꢀ5.5 ppm; IR (ATR): n˜ =3436 (m), 3214 (m), 2955 (s), 2931 (s), 2887
(m), 2858 (s), 1669 (vs), 1623 (vs), 1589 (vs), 1537 (s), 1472 (m), 1427
(m), 1373 (m), 1330 (m), 1232 (s), 1151 (w), 1108 (m), 1034 (w), 1005
(w), 939 (vw), 837 (s), 776 (m), 699 cmꢀ1 (w); HRMS (ESI+): m/z: calcd
for C24H45N5O6Si2: 556.2987; found: 556.2993 [M+H]+.
A
(7):
Compound 6 (5.2 g, 15.4 mmol) was dissolved in dry CH2Cl2 (100 mL)
and cooled to ꢀ788C. A solution of 1m BCl3 (200 mL) in CH2Cl2 was
added by means of a dropping funnel over a period of 45 min and the
mixture was stirred at ꢀ788C for 3 h, then warmed upto room tempera-
ture. The reaction was quenched with dry MeOH (80 mL) at ꢀ788C and
was allowed to warm upto room temperature overnight. The solvents
were evaporated and the residue was purified on silica gel (10 cm pad of
350 mL silica gel, 10% MeOH in CHCl3). Product 7 (2.4 g, 15.3 mmol,
99%) was obtained as a yellow oil. Rf =0.27 (10% MeOH in CHCl3);
1H NMR (400 MHz, CDCl3): d=4.07 (dd, J=13.3, 6.0 Hz, 1H), 4.02–3.95
(m, 1H), 3.79 (ddd, J=10.4, 5.2, 3.0 Hz, 1H), 3.56 (dd, J=10.4, 8.0 Hz,
1H), 2.36–2.20 (m, 4H), 1.99–1.92 (m, 1H), 1.77 (dddd, J=14.0, 6.0, 4.5,
1.6 Hz, 1H), 1.58 ppm (ddd, J=13.9, 9.8, 6.8 Hz, 1H); 13C NMR
(100 MHz, CDCl3): d=75.8, 65.0, 60.2, 48.4, 40.7, 33.3 ppm; IR (ATR):
n˜ =3351 (vs), 2935 (s), 2499 (w), 2103 (vs), 1651 (w), 1442 (m), 1338 (m),
1261 (s), 1165 (m), 1076 (s), 1048 (s), 941 (m), 878 (w), 687 (m),
559 cmꢀ1; HRMS (ESIꢀ): m/z: calcd for C6H11N3O2Cl: 192.0545; found:
192.0544 [M+Cl]ꢀ.
N-{4-[3-(tert-Butyldimethylsilanyloxy)-4-[(tert-butyldimethylsilanyloxy)-
methyl]cyclopentylamino]-5-formylamino-6-oxo-1,6-dihydropyrimidin-2-
yl}acetamide (12): Compound 11 (1.9 g, 3.4 mmol) was dissolved in dry
EtOH (30 mL) and CH2Cl2 (5 mL). The reaction was evacuated twice to
exchange the inert gas atmosphere and then connected to two balloons
that were filled with H2. The suspension was vigorously stirred at room
temperature for 15 h. The palladium catalyst was removed by using a
PTFE-Filter (Whatman Puradisc) and the solvent was removed under re-
duced pressure. The amine was obtained as a colourless powder, which
was directly converted into the formamide; the residue was dissolved in
dry THF (50 mL) and dry pyridine (30 mL). EDC·HCl (1.7 g, 8.8 mmol)
was added in one portion and after the majority of the suspension was
dissolved, formic acid (162 mL, 4.3 mmol) was introduced. The reaction
was stirred for 10 h, and then the solvent was removed under reduced
pressure. The obtained residue was dissolved in ethyl acetate (200 mL)
ACHTREUNG
7
was dissolved in dry DMF (20 mL), and imidazole (2.61 g, 38.3 mmol)
was added at room temperature. After the portionwise addition of
TBDMSCl (5.77 g, 38.3 mmol) at 08C, the mixture was stirred at room
temperature for 16 h. The reaction was diluted with CH2Cl2 (200 mL),
washed once with sat. NH4Cl (150 mL) and with brine (100 mL) and
dried over MgSO4. The solvent was removed in vacuo and the crude
product was purified by flash chromatography (300 mL silica gel, 5%
ethyl acetate in hexane) to give compound 8 (5.1 g, 13.2 mmol, 86%) as
colourless oil. Rf =0.59 (ethyl acetate/hexane 20:1); 1H NMR (400 MHz,
CDCl3): d=4.04 (q, J=6.5, 6.5, 6.5 Hz, 1H), 3.87–3.77 (m, 1H), 3.60–3.52
(m, 2H), 2.23 (td, J=13.7, 6.9, 6.9 Hz, 1H), 2.10 (tdt, J=8.6, 8.6, 6.2, 4.4,
2130
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2125 – 2132

FaPydG Lesions
FULL PAPER
and washed with H2O (2”100 mL), sat. NaHCO3 solution (2”100 mL)
and brine (100 mL). After drying over MgSO4, the solvent was evaporat-
ed and the residue was purified by flash chromatography (silica gel
(350 mL), 5% MeOH in CHCl3). Compound 12 (1.5 g, 2.7 mmol, 79%)
was isolated as yellow powder after recrystallization from heptane. Rf =
0.22 (10% MeOH in CHCl3); m.p. 115–1198C; 1H NMR (400 MHz,
DMSO): d=11.27 (brs, 2H), 8.73 (d, J=1.6 Hz, 0.61H), 8.04 (d, J=
1.6 Hz, 0.39H), 8.04 (d, J=11.3 Hz, 0.61H), 7.80 (d, J=11.5 Hz, 0.39H),
6.55 (d, J=8.2 Hz, 1H), 6.18 (d, J=8.3 Hz, 1H), 4.37 (qd, J=14.3, 7.17,
7.17, 6.86 Hz, 1H), 3.98–3.90 (m, 1H), 2.15 (s, 3H), 2.18–2.02 (m, 2H),
1.77–1.61 (m, 2H), 1.53 (ddd, J=25.82, 12.87, 6.47 Hz, 1H), 0.86 (s, 9H),
0.85 (s, 9H), 0.03 (s, 3H), 0.03 (s, 3H), 0.03 (s, 3H), 0.02 ppm (s, 3H);
13C NMR (100 MHz, DMSO): d=174.1, 174.0, 166.8, 161.3, 158.5, 157.5,
157.4, 148.9, 148.8, 74.4, 63.8, 49.4, 49.1, 42.2, 34.6, 26.2 (3C), 24.3, 18.4,
18.2, ꢀ4.2, ꢀ4.4, ꢀ5.0 ppm (2C); IR (ATR): n˜ =3436 (s), 2956 (m), 2950
(m), 2858 (m), 1652 (s), 1472 (w), 1254 (m), 1088 (w), 837 (m), 777 cmꢀ1
(m); HRMS (ESI+): m/z: calcd for C25H47N5O5Si2: 554.3194; found:
554.3190 [M+H+]+. The cis/trans isomers of the formyl groupare re-
sponsible for the signals at d=8.67/8.04 and 8.04/7.8 ppm.
0.12 mmol) and 2-cyanoethyl-tetraisopropylphosphordiamidite (636 mL,
2.0 mmol) were suspended in acetonitrile (15 mL) and stirred for 6 h
under an argon atmosphere at room temperature. The acetonitrile was
evaporated under reduced pressure. The crude product was purified by
flash chromatography (50 mL of deactivated silica gel (the silica gel was
suspended in 1% pyridine in dichloromethane, stirred for 1 h, evaporated
and dried in an oven overnight) 8 cm pad, 5% MeOH in CHCl3 + 5%
pyridine). The phosphoramidite 15 (468 mg, 0.57 mmol, 84%) was ob-
tained as a colourless film. Rf =0.27 (10% MeOH in CHCl3); 1H NMR
(600 MHz, CDCl3): d=8.64–8.60 (m, 1H), 8.22 (s, 1H), 7.71 (m, 1H),
7.67 (tt, J=7.6, 7.6, 1.8, 1.8 Hz, 1H), 7.49–7.40 (m, 2H), 7.36–7.25 (m,
6H), 7.21 (t, J=7.4, 7.4 Hz, 1H), 6.86–6.76 (m, 3H), 6.33 (t, J=7.1,
7.1 Hz, 1H), 4.40–4.08 (m, 2H), 3.83–3.71 (m, 5H), 3.70–3.46 (m, 2H),
3.20–3.01 (m, 2H), 2.63 (m, 2H), 2.52 (t, J=6.0, 6.0 Hz, 1H), 2.39–2.30
(m, 2H), 2.21–2.05 (m, 3H), 2.05–1.80 (m, 2H), 1.75–1.60 (m, 1H), 1.53–
1.28 (m, 2H), 1.28–1.23 (m, 2H), 1.24–1.18 (m, 2H), 1.18–1.06 (m, 6H),
1.04 (d, J=6.8 Hz, 2H), 0.90 ppm (m, 1H); 13C NMR (150 MHz, CDCl3):
d=172.1, 171.8, 159.9, 168.6 (2C), 154.8, 154.7, 150.0, 145.5, 136.6, 136.2,
131.1, 130.4 (2C), 130.3, 128.5, 128.4 (2C), 126.9, 123.9, 118.2, 118.0,
113.4, 113.2, 86.0 (2C), 75.6, 75.5, 64.3, 64.1, 58.0, 57.9, 57.4, 57.3, 55.4,
50.7, 50.4, 50.3, 45.8, 45.5, 43.3 (2C), 43.2 (2C), 41.1, 41.0, 35.0, 34.7, 24.9
(2C), 24.8 (2C), 24.7 (2C), 24.6 (2C), 20.8, 20.6 ppm; 31P NMR (81 MHz,
CDCl3): d=147.5, 148,4 ppm; IR (ATR): n˜ =3423 (s), 3242 (s), 2962 (m),
1655 (s), 1584 (s), 1510 (s), 1444 (m), 1415 (m), 1365 (w), 1250 (s), 977
(w), 728 cmꢀ1 (w); HRMS (ESIꢀ): m/z: calcd for C43H53N7O8P: 828.3693;
found: 828.3696 [MꢀH]ꢀ.
DNA synthesis: Oligonucleotide synthesis was performed on an ¾KTA
Oligopilot 10 from Amersham Biosciences by using standard DNA syn-
thesis conditions (deblocking solution: 2.5% dichloroacetic acid in tolu-
ene; activator: 0.25m 5-benzylmercaptotetrazole in acetonitrile; oxida-
tion: 150 mm I2 in 2,6-lutidine/MeCN/H2O 1:11:5; capping solutions: 1m
Ac2O, 2,6-lutidin (11%) in acetonitrile and N-methylimidazole (16%) in
MeCN. Ultra-mild phosphoramidites for dA, dC, dG, dT and CPG carri-
ers were obtained from Amersham, Glen Research or PE Biosystems.
Because oligonucleotides that contained the cFaPydG building block
were not compatible with the regular capping procedure, the standard
capping solutions were replaced by 2,6-lutidine/isobutyrylanhydride/THF
1:1:8, and N-methylimidazole (16%) in acetonitrile, and were used for all
couplings after the modification. The coupling of a-cFaPydG itself was
carried out by using an eight-fold excess of phosphoramidite in a double-
coupling protocol with 15 min for each step. After the modification, all
the phosphoramidites were also coupled with a double-coupling protocol
by using 10 equiv for 3 min. The DNA was cleaved off of the solid sup-
port by using EtOH and sat. NH3 solution (1:3) at 158C overnight. The
DNA was purified by RP-HPLC by using a gradient of 0.1m triethylam-
moniumacetate in H2O/0.1m triethylammoniumacetate in MeCN/H2O
80:20. The purity of the strands was verified by analytical RP-HPLC and
MALDI-TOF or ESI-FTICR mass spectrometry.
N-{4-[3-(hydroxy)-4-[(hydroxy)methyl]cyclopentylamino]-5-formylamino-
6-oxo-1,6-dihydropyrimidin-2-yl}acetamide (13): Compound 12 (250 mg,
0.45 mmol) was dissolved in dry THF (4.5 mL) and pyridine (125 mL) in a
10 mL plastic tube. After addition of HF–pyridine complex (125 mL), the
reaction was stirred overnight. The resulting orange precipitate was cen-
trifuged, and the supernatant was decanted off. The solvent was discard-
ed. Traces of HF–pyridine in the precipitate were quenched by adding
methoxytrimethylsilane (0.6 mL). The crude product was washed twice
with dry THF, and dried completely for 2 d under high vacuum. Com-
pound 13 (145 mg, 0.9 mmol, quant.) was obtained as a yellow powder.
Rf =0.67 (C18-TLC, MeOH/H2O 1:1); m.p. 152–1558C; 1H NMR
(400 MHz, DMSO): d=11.34 (brs, 1H), 11.31 (brs, 1H), 8.75 (d, J=
1.4 Hz, 0.61H), 8.12 (d, J=11.5 Hz, 0.37H), 8.06 (d, J=1.5 Hz, 0.65H),
7.81 (d, J=11.5 Hz, 0.37H), 6.57 (d, J=8.2 Hz, 0.33H), 6.27 (d, J=
8.2 Hz, 0.62H), 4.76 (dd, J=11.5, 4.2 Hz, 1H), 4.49 (brs, 1H), 4.33 (sept.,
J=7.4, 7.4, 7.4, 7.4, 7.4 Hz, 1H), 3.83–3.74 (m, 1H), 3.43–3.23 (m, 2H),
2.15 (s, 3H), 2.07 (td, J=12.8, 6.4, 6.4 Hz, 1H), 2.01–1.91 (m, 1H), 1.73–
1.62 (m, 2H), 1.50 ppm (ddd, J=26.6, 13.1, 6.7 Hz, 1H); 13C NMR
(75 MHz, DMSO): d=202.7, 173.9, 161.1, 157.2, 113.6, 92.8, 67.4, 62.9,
49.6, 48.8, 35.0 (2C), 25.5, 24.2 ppm; IR (ATR): n˜ =3420 (s), 2929 (w),
1645 (vs), 1587 (s), 1505 (m), 1440 (w), 1422 (m), 1375 (w), 1241 (s), 1152
(w), 1044 (m), 1013 (w), 879 (w), 710 (w), 563 cmꢀ1 (w); HRMS (ESIꢀ):
m/z: calcd for C13H18N5O5: 324.1308; found: 324.1315 [MꢀH]ꢀ.
N-{4-[3-(hydroxy)-4-[(dimethoxytrityloxy)methyl]cyclopentylamino]-5-
formylamino-6-oxo-1,6-dihydropyrimidin-2-yl}acetamide (14): Compound
13 (352 mg, 1.08 mmol) was dissolved in pyridine (15 mL) at room tem-
perature. DMTCl (477 mg, 1.41 mmol) was added in 3 portions (2”
0.5 equiv + 1”0.3 equiv) every 30 min. After 3.5 h stirring, TLC analysis
indicated that the reaction was complete and the mixture was quenched
with MeOH (3 mL). The solvent was evaporated under reduced pressure,
and the crude product was purified by flash chromatography (silica gel,
10% MeOH and 5% pyridine in CHCl3. The DMT-protected carbocycle
14 (500 mg, 7.97 mmol, 74%) was obtained as a powder. Rf =0.16 (10%
MeOH in CHCl3); m.p. 145–1488C; 1H NMR (400 MHz, CDCl3): d=
11.34 (brs, 1H), 11.31 (brs, 1H), 8.75 (d, J=1.30 Hz, 0.61H), 8.12 (d, J=
11.5 Hz, 0.39H), 8.06 (d, J=1.5 Hz, 0.61H), 7.98 (d, J=11.5 Hz, 0.39H),
7.24–7.17 (m, 9H), 6.75–6.71 (m, 4H), 6.14 (d, J=8.2 Hz, 0.33H), 6.05 (d,
J=8.2 Hz, 0.62H), 4.46–4.30 (m, 1H), 3.99–3.89 (m, 1H), 3.70 (s, 6H),
3.00–2.89 (m, 2H), 2.10 (s, 3H), 2.35–2.25 (m, 1H), 2.13–1.98 (m, 2H),
1.83 (ddd, J=13.0, 8.9, 4.1 Hz, 1H), 1.63–1.52 ppm (m, 2H); 13C NMR
(100 MHz, CDCl3): d=172.6, 159.8, 157.8 (2C), 157.1, 155.2, 149.2, 144.6,
135.6, 135.4, 129.4 (4C), 127.5, 127.2, 126.1 (2C), 123.2, 12.5 (4C), 94.1,
85.2, 74.6, 64.9, 54.6, 49.8, 46.2, 41.4, 35.3, 23.5 ppm; IR (ATR): n˜ =3436
(s), 2920 (s), 1732 (m), 1650 (m), 1445 (w), 1384 (m), 1249 (m), 1156 (w),
1070 (m), 612 cmꢀ1 (w); HRMS (ESI+): m/z: calcd for C34H38N5O7:
628.2771; found: 628.2776 [M+H]+.
Enzymatic digestion of the DNA: For the enzymatic digestion, the DNA
(ca. 10 mg in 100 mL to clarify the mutagenic potential of the FaPydG
lesion) was incubated in 10 mL of buffer (300 mm ammonium acetate,
100 mm CaCl2, 1 mm ZnSO4, pH 5.7), 22 units Nuclease P1 (Penicilinum
citrium) and 0.05 units of calf spleen phosphodiesterase II. The sample
was shaken at 378C for 3 h. The digest was completed by adding 12 mL
buffer (500 mm Tris-HCl, 1 mm EDTA), 10 units of alkaline phosphatase
(CIP) and 0.1 units of snake venom phosphodiesterase I (Crotalus ada-
manteus venom). The sample was shaken for another 3 h at 378C. For
workup, 6 mL of 0.1m HCl was added and the probes were centrifuged
(6000 rpm, 5 min.). The digest was analyzed by HPLC (Interchim Inter-
chrom Uptisphere 3 HDO column (150”2.1 mm), Buffer A: 2 mm NEt3/
HOAc in H2O; Buffer B: 2 mm NEt3/HOAc in H2O/MeCN 20:80; 0!
30 min; 0!30% B; 30!32 min; 30!100% B; 32!36 min; 100% B;
36!38 min; 100!0% B; 38!60 min; 0% B; flow: 0.2 mLminꢀ1). The
different peaks were assigned by co-injection, UV and FTICR-HPLC-
MS-MS, by using the same conditions. The incorporated cFaPydG could
be detected by LCMS/MS and FTICR-ESI; calcd mass for a-cFaPydG,
C13H20N5O6ꢀ: m/z: 342.1419; found: 342.1416 [M+AcOHꢀH+]ꢀ; further
signals of the lesion: FTICR-ESI+: m/z: calcd for C11H16N5O4ꢀ: 282.1208;
N-{4-[3-(hydroxy)-4-[(dimethoxytrityloxy)methyl]cyclopentylamino]-5-
formylamino-6-oxo-1,6-dihydropyrimidin-2-yl}acetamide
phosphorami-
dite (15): Compound 14 (420 mg, 0.67 mmol), tetrazolate (20 mg,
Chem. Eur. J. 2008, 14, 2125 – 2132
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2131

T. Carell et al.
ꢀ
found: 282.1225 [MꢀH+]ꢀ; calcd for C10H16N5O3
254.1221 [MꢀCOꢀH+]ꢀ.
:
254.1259; found:
[6] K. Haraguchi, M. O. Delaney, C. J. Wiederholt, A. Sambandam, Z.
Hantosi, M. M. Greenberg, J. Am. Chem. Soc. 2002, 124, 3263–
3269.
Primer extensions: For primer extensions, a solution of template (4 mm)
and primer (2 mm) in buffer (KCl 50 mm, 10 mmK3PO4/HCl pH 7.5) was
heated to 808C and allowed to cool to 58C over several hours, then
stored at that temperature for the experiments. For the experiments with
concentrations as described in the text, primer/template solution (3 mL),
enzyme solution (3 mL) and dNTP solution (3 mL) were added to 21 mL
reaction buffer (Pol h: 100 mm KCl, 5 mm MgCl2, 10 mm DTT, 2.5 vol%
glycerin, 40 mm Tris-HCl pH 7.4, 2.5 vol% BSA (10 mgmLꢀ1); DinB:
5 mm MgCl2, 10 mm DTT, 250 mgmLꢀ1 BSA, 2.5 vol% glycerin, 40 mm,
pH 8 Tris-HCl; Bst Pol I: 10 mm KCl, 10 mm (NH4)2SO4, 2 mm MgSO4,
0.1% Triton X-100, 20 mm Tris-HCl buffer pH 8.8). The reaction was
quenched by adding 100 mm EDTA solution (30 mL, pH 7.4) and incuba-
tion at 958C for 15 min. After centrifugation at 5000 rpm, the DNA was
extracted from the supernatant by using ZipTip (Millipore). The eluted
probes (ca. 10 mL) were diluted with ddH2O (40 mL) and analyzed by
using capillary electrophoresis (9 kV for 30 min).
[7] K. Haraguchi, M. M. Greenberg, J. Am. Chem. Soc. 2001, 123,
8636–8637.
[8] J. N. Patro, K. Haraguchi, M. O. Delaney, M. M. Greenberg, Bio-
chemistry 2004, 43, 13397–13403.
[9] L. T. Burgdorf, T. Carell, Chem. Eur. J. 2002, 8, 293–301.
[10] M. Ober, H. Muller, C. Pieck, J. Gierlich, T. Carell, J. Am. Chem.
Soc. 2005, 127, 18143–18149; for the primer-extension studies re-
ported in this publication see also G. Hsu, Ph.D thesis, Duke Uni-
versity.
[11] Y. L. Jiang, C. J. Wiederholt, J. N. Patro, K. Haraguchi, M. M.
Greenberg, J. Org. Chem. 2005, 70, 141–149.
[12] C. J. Wiederholt, M. O. Delaney, M. A. Pope, S. S. David, M. M.
Greenberg, Biochemistry 2003, 42, 9755–9760.
[13] F. Coste, M. Ober, T. Carell, S. Boiteux, C. Zelwer, B. Castaing, J.
Biol. Chem. 2004, 279, 44074–44083.
[14] S. Masamune, B. M. Kim, J. S. Petersen, T. Sato, S. J. Veenstra, T.
Imai, J. Am. Chem. Soc. 1985, 107, 4549–4551.
[15] O. R. Ludek, C. Meier, Synthesis 2003, 2101–2109.
[16] W. Pfleider, E. Buhler, Chem. Ber. 1966, 99, 3022.
[17] L. A. Marky, K. J. Breslauer, Biopolymers 1987, 26, 1601–1620.
[18] S. Creighton, M. F. Goodman, J. Biol. Chem. 1995, 270, 4759–4774.
[19] S. Creighton, L. B. Bloom, M. F. Goodman, Methods Enzymol. 1995,
262, 232–257.
Acknowledgements
This work was supported by the Volkswagenstiftung, the Fonds der
Chemischen Industrie (FCI) and Novartis. We thank the DFB (SFB 749)
and the Leibniz program for financial support.
[20] S. Imoto, J. N. Patro, Y. L. Jiang, N. Oka, M. M. Greenberg, J. Am.
Chem. Soc. 2006, 128, 14606–14611.
[21] J. N. Patro, C. J. Wiederholt, Y. L. Jiang, J. C. Delaney, J. M. Essig-
mann, M. M. Greenberg, Biochemistry 2007, 46, 10202–10212.
[22] S. Prakash, R. E. Johnson, L. Prakash, Annu. Rev. Biochem. 2005,
74, 317–353.
[1] A. R. Collin, J. Cadet, L. Mçller, H. E. Poulsen, J. Vina, Arch. Bio-
chem. Biophys. 2004, 423, 57–65.
[2] J.-P. Pouget, S. Frelon, J.-L. Ravanat, I. Testard, F. Odin, J. Cadet,
Radiat. Res. 2002, 157, 589–595.
[3] C. J. Burrows, J. G. Muller, Chem. Rev. 1998, 98, 1109–1151.
[4] R. Saladino, E. Mincone, C. Crestini, R. Negri, E. Di Mauro, G.
Costanzo, J. Am. Chem. Soc. 1996, 118, 5615–5619.
Received: August 31, 2007
Revised: October 24, 2007
[5] M. Berger, J. Cadet, Z. Naturforsch. 1985, 40, 1519–1531.
Published online: January 14, 2008
2132
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2125 – 2132