Enzymatic synthesis of perfluoroalkylated DNA

Article abstract of DOI:10.1016/j.bmc.2009.03.063

Thymidine analogues 5-trifluoromethyl-, 5-pentafluoroethyl- and 5-(heptafluoro-n-propyl)-2′-deoxyuridines were synthesised and converted into the corresponding 5′-triphosphates 1a-c. Performing DNA polymerase-catalyzed primer extension reactions these mod

Full text of DOI:10.1016/j.bmc.2009.03.063

Bioorganic & Medicinal Chemistry 17 (2009) 3653–3658
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Enzymatic synthesis of perfluoroalkylated DNA
Bastian Holzberger, Andreas Marx *
Department of Chemistry and Konstanz Research School Chemical Biology, University of Konstanz, Universitätsstrasse 10, P.O. Box 726, 78457 Konstanz, Germany
a r t i c l e i n f o
a b s t r a c t
0
Article history:
Thymidine analogues 5-trifluoromethyl-, 5-pentafluoroethyl- and 5-(heptafluoro-n-propyl)-2 -deoxyuri-
Received 27 February 2009
Revised 23 March 2009
Accepted 29 March 2009
Available online 7 April 2009
0
dines were synthesised and converted into the corresponding 5 -triphosphates 1a–c. Performing DNA
polymerase-catalyzed primer extension reactions these modified nucleotides were incorporated into
DNA to create perfluoroalkylated nucleic acids. Although single modified nucleotides were enzymatically
incorporated and further elongated quite similar to the natural TTP, the enzymatic synthesis of multi-
modified nucleic acids was initial only feasible with modifications at every fourth base. Nevertheless,
as the effects of the modified dUTPs on DNA polymerases varied significantly with the used enzyme,
Keywords:
DNA
Modified nucleotide
DNA polymerase
Fluor
0
Therminator DNA polymerase was proficient in incorporating 11 adjacent 5-trifluoromethyl-2 -deoxyur-
idine moieties.
Ó 2009 Elsevier Ltd. All rights reserved.
SELEX
1
. Introduction
Aside from being the fundamental genetic material for the stor-
ious approaches in bioorganic and medicinal chemistry.10 Fluorine
is highly electronegative and shows unique nuclear magnetic res-
onance (NMR) spectroscopic parameters. The substitution of
hydrogen by fluorine leads to significant electronic effects and
age and transfer of information, DNA has recently gained signifi-
cant attention as building block in nanotechnology. The
1
can modify properties of a molecule like pK
a
-values, dipole mo-
introduction of chemically modified DNA building blocks may offer
the possibility to modify the physical and geometrical properties of
DNA. The generation of new DNA-based materials through enzy-
matic synthesis is particularly promising, since this approach has
the potential to introduce non-nucleic acid like functionalities
through usage of modified building blocks and their incorporation
ments or chemical reactivities and stabilities of adjacent functional
1
1
groups. Although fluorine is used as an isosteric replacement for
hydrogen, the introduction of fluorine into a molecule leads to ste-
ric changes as well. This is caused by different van der Waals radii
of fluorine (1.47 Å) and hydrogen (1.2 Å) and different bond
1
1
lengths of C–H (1.09 Å) and C–F (1.35 Å). However, fluorinated
substrate analogues are often used as mechanistic probes to study
2
into long DNA stretches by DNA polymerases. Although this ap-
proach has only sparsely been exploited so far,2 several different
modifications were successfully incorporated to create highly
enzyme catalysis. Furthermore, fluorine is applied in monitoring
10c
19
functional important transitions in biological systems via F NMR
3
12
functionalized DNA double strands. Aside from amino acid groups,
spectroscopic methods.
0
aromatic, basic, acidic and lipophilic modifications, functionalized
dNTPs were also used to introduce Ru(II) and Os(II) complexes into
The fluorinated thymidine analogue 5-trifluoromethyl-2 -deoxy-
uridine (CF
adenosine. By replacing an internal thymidine residue in DNA oli-
gonucleotides with CF dU DNA double helices are formed that still
adopt the B-form conformation. Thereby, the substitution of a
thymidine residue by CF dU causes a slight decrease in DNA du-
3
dU) has been shown to form regular base pairs with
4
DNA double strands. Furthermore, alkyne- and azido-labelled oli-
gonucleotides offer the possibility to post-synthetically modify
3
5
13
DNA oligonucleotides by ‘Click’ chemistry or Staudinger ligation.
Recently, we used DNA polymerases for the construction of DNA-
3
6
7
14
based networks and site-specific multi-spinlabelled DNA. In
addition, modified DNA has been employed in the generation of
aptamers and catalytically active DNA.8 These entities were ob-
tained by SELEX that requires that the respective modified nucleo-
tides are substrates for DNA polymerases.
plex stability. Apart from chemical methods to introduce modi-
fied nucleosides into DNA oligonucleotides by using suitable
building blocks in automated DNA synthesis, enzymatic incorpora-
,9
0
tion of modified nucleoside 5 -triphosphates by DNA polymerases
2
allows the synthesis of long multi-modified DNA oligonucleotides.
0
Due to its small van der Waals radius and the stability of the
highly polarized bond formed with carbon, fluorine is used in var-
Albeit it was claimed that CF
3
dU 5 -triphosphate is a substrate for
DNA polymerases in vitro, experimental results have not been
documented. In addition, the effect of extended C5-perfluoro mod-
ifications in 2 -dU 5 -triphosphates has not been explored so far.
Here we present the introduction of 5-trifluoromethyl-, 5-penta-
1
5
0
0
*
Corresponding author. Tel.: +49 7531 88 5139; fax: +49 7531 88 5140.
E-mail address: Andreas.Marx@uni-konstanz.de (A. Marx).
0
968-0896/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2009.03.063

3
654
B. Holzberger, A. Marx / Bioorg. Med. Chem. 17 (2009) 3653–3658
fluoroethyl- and 5-(heptafluoro-n-propyl)-dU-residues into DNA
by exploiting DNA polymerase-catalyzed template directed reac-
tions and modified triphosphates 1a–c.
(a)
Primer (23 nt):
Template (35 nt):
5’---TC
3’---AGG GCA CGG TCG CG
(
b)
35
(c)
35
2
. Results and discussion
First, we synthesized the 5-perfluoroalkyl-2 -deoxyuridine-5 -
0
0
triphosphates 1a–c (Scheme 1).
The synthesis started with 2a–c that were synthesized follow-
ing published procedures.16 2a was deacetylated according to the
reported procedure by using a 7 M ammonia solution of metha-
2
7
1
6
27
nol. However, this method gave unsatisfactory results for 2b
and c due to the formation of side-products. Due to the known
reactivity and instability of the trifluoromethyl group of 5-trifluo-
1
7
romethyluracil towards nucleophiles and under basic conditions,
23
23
we applied a different method for deacetylation of 2b and 2c. By
use of toluene-4-sulfonic acid monohydrate in dichloromethane
and methanol, we were able to synthesize 3b and 3c. Compounds
0
3
a–c were subsequently transformed to the nucleoside 5 -triphos-
1
8
phates 1a–c.
Next, we explored the action of 1a–c on DNA polymerases and
investigated the ability of the Klenow fragment of Escherichia coli
DNA polymerase I (3 ?5 exonuclease-deficient variant, KF(exo-))
and the Klenow fragment of Thermus aquaticus DNA polymerase
Figure 1. Primer extension studies: acceptance of 1a–c by KF(exo-) (b) and KlenTaq
c). (a) Partial primer template sequences employed. (b) and (c) Primer: primer
only; ꢀTTP: primer extension in the presence of dATP, dGTP and dCTP; +TTP: as
TTP but in the presence of TTP; CF
dUTP: as ꢀTTP but in the presence of 1a;
dUTP: as ꢀTTP but in the presence of 1b; C dUTP: as ꢀTTP but in the presence
of 1c.
(
0
0
ꢀ
3
C
2
F
5
3 7
F
(
KlenTaq) to accept 1a–c in primer extension experiments. There-
3
2
0
fore, we used a P-5 -end labelled primer (23 nt)/template (35 nt)
complex with an adenosine residue at position 27 of the template
to direct the usage of TTP and the thymidine analogues 1a–c,
respectively, after incorporation of three natural nucleotides
(25 nt) in most cases. In general, incorporation of the modified
thymidines was comparable to that of the natural counterpart
(Fig. 2). Especially 1a led to similar product yields opposite the four
natural nucleotide residues (quantified by phosphor imaging).
(Fig. 1a).
Subsequently all reactions were analyzed by denaturing poly-
acrylamide gel electrophoresis (PAGE) and phosphor imaging. In
the presence of dGTP, dCTP and dATP and the absence of any TTP
analogue, primer elongation was paused at position 27 when using
KF(exo-) or KlenTaq (Fig. 1b and c, ꢀTTP). When all four natural
dNTPs were present, formation of full length products was moni-
tored with both enzymes (Fig. 1b and c, +TTP). Substitution of
TTP by 1a–c led with both enzymes to full length products as well
(a)
T*TP
Primer:
5’---ACA
Template: 3’---TGT NCT GTC TGC GTG
(b) N = A
T*TP: TTP
T*TP: 1a
T*TP: 1b
T*TP: 1c
(
Fig. 1b and c, CF
3
dUTP, C
2
F
5
dUTP, C
3
F
7
dUTP). The incorporation of
25
an additional nucleotide in a non-templated manner led to the
3
exonuclease-deficient DNA polymerases.
To get first insights into the efficiency and selectivity of the
enzymatic incorporation step in particular, we examined the
acceptance of 1a–c by KF(exo-) catalysis performing single incor-
poration experiments in the presence of different concentrations
of TTP and 1a–c, respectively. Thus, we used four different P-5 -
end labelled primer (24 nt)/template (36 nt) complexes with all
24
0
0
6 nt long products. This has been observed before using 3 ?5
0
1 2 3
(c) N = G
T*TP: TTP
0 1 2
3
0 1 2
3
0 1 2 3
19
T*TP: 1a
T*TP: 1b
T*TP: 1c
2
2
5
4
3
2
0
0
1 2 3
0 1 2 3
0 1 2
3
0 1 2 3
0
four natural nucleotide residues at the first position after the 3 -
end of the primer (Fig. 2a). Doing so, we observed two bands after
PAGE analysis, the given 24 nt long primer and the elongated one
(d) N = C
T*TP: TTP
T*TP: 1a
T*TP: 1b
T*TP: 1c
2
2
5
4
0
1
2 3
0 1 2 3
0 1 2 3
0 1 2 3
(e) N = T
T*TP: TTP
T*TP: 1a
T*TP: 1b
T*TP: 1c
2
2
5
4
0
1 2 3
0 1 2 3
0 1 2 3
0 1 2 3
Figure 2. Single incorporation employing TTP and 1a–c by KF(exo-) with differing
nucleotide concentrations. (a) Partial primer template sequences employed.
(b) Incorporation opposite A (matched case, t = 5 min); lanes 0: primer only, lanes
Scheme 1. Synthesis of 5-perfluoroalkylated dUTP analogues 1a–c. Reagents and
*
*
*
yields: (a) NH
6% (3c); (c) 1,8-bis-(dimethylamino)-naphthalene, POCl
Bu NH) , Et N, 8% (1a), 10% (1b), 16% (1c).
3
/MeOH (7 M), 76% (3a); (b) p-TsOHꢁH
2
O, DCM/MeOH (9:1), 74% (3b),
1: 1
lM T TP, lanes 2: 10 lM T TP, lanes 3: 100 lM T TP. (c–e) Incorporation
6
3
,
(CH O) PO then (n-
3
3
opposite G/C/T (mismatched cases, t = 30 min); lanes 0: primer only, lanes 1: 10 lM
*
*
*
3
H
2 2
P
2
O
7
3
T TP, lanes 2: 50 lM T TP, lanes 3: 200 lM T TP.

B. Holzberger, A. Marx / Bioorg. Med. Chem. 17 (2009) 3653–3658
3655
Higher concentrations of 1b and 1c inhibited the incorporation
opposite the guanosine residue (Fig. 2c, right two panels, lanes 3).
Even more unexpected was that TTP, 1a and 1b were not, but 1c
was used as substrate for incorporation opposite the non-canonical
cytosine residue in the template (Fig. 2d).
Next, we investigated the multiple incorporations of the modi-
fied nucleotides into DNA by using different templates that call for
usage of TTP and 1a–c, respectively, at every fourth, every second
or every singular nucleotide position. Figure 3 shows the results
obtained in multiple incorporation experiments by using KF(exo-).
Using the template with an adenosine residue at every fourth posi-
tion, we obtained full length products when natural TTP was
incorporates only about 4–6 adjacent nucleotides under the same
conditions (Fig. 4b, middle and right panel). This inhibition of
DNA synthesis is in part surprising since it has been shown that
bulkier nucleotides are accepted by the same DNA polymerase.2
However, the nucleotide perfluoroalkyl modification hampers
catalysis to a greater extent. Thereby, the pentafluoroethyl- and
heptafluoro-n-propyl-modified residues exhibit enhanced inhibi-
,3
3
tion of DNA synthesis in comparison to the CF dU residue. Pertur-
bation of DNA polymerase/DNA contacts, for example, H-bonds
between amino acid side chains and the nucleobases which are re-
quired for binding and activity of the enzyme might be the origin of
this effect.
substituted by 1a–c (Fig. 3a, CF
acceptance of 1a–c at every second position led only in the case
of 1a to the full length product (Fig. 3b, CF dUTP). The template
with an adenosine residue at every position could not be extended
to full length with any of the TTP analogues (Fig. 3c, CF dUTP,
dUTP, C dUTP). Although the modified nucleotides were
3 2 5 3 7
dUTP, C F dUTP, C F dUTP). The
3
. Conclusions
3
In summary, we report the synthesis of 5-trifluoromethyl-, 5-
0
3
pentafluoroethyl- and 5-(heptafluoro-n-propyl)-2 -deoxyuridine-
2
C F
5
3
F
7
0
5
-triphosphates 1a–c and their usage for site-specific introduction
incorporated opposite the adenosine residue in comparable yields
to the natural counterpart (Fig. 2b), the multiple incorporation of
additional nucleotide analogues was inhibited. With employed
conditions we obtained reaction products that were only elongated
by one or two (1b, 1c) or four incorporated thymidine analogues in
of perfluorinated residues into DNA in DNA polymerase-catalyzed
template directed reactions. Since fluorine has properties that are
distinct from other polar or non-polar modifications the herein de-
picted approach presents new routes for the generation of highly
modified nucleic acids for future DNA biotechnological applica-
tions like in SELEX. Directed evolution of DNA polymerases might
be suited for the improvement of enzymes’ proficiencies to process
3 2 5 3 7
the case of 1a (Fig. 3c, CF dUTP, C F dUTP, C F dUTP).
However, we found that the effects of 1a–c on DNA polymer-
ases significantly vary with the used enzyme. It was found that
Therminator DNA polymerase (A485L mutant of Thermococcus spe-
cies 9°N DNA polymerase) is most proficient in processing multiple
incorporations of these modified nucleotides. In order to form one
2
0
non-natural substrates.
4. Experimental
4.1. General
0
entire DNA helix turn of 5-perfluoroalkyl-2 -deoxyuridines, we
used a template with eleven adjacent adenosine residues (Fig. 4a).
Due to the known decrease in stability of DNA helices with
internal CF
3
dU residues,14 we examined the multiple incorporation
NMR spectra were recorded on a Bruker AC-250 Cryospec, a
Bruker DRX-600 and a Jeol ECX-400 spectrometer. ESI-IT mass
spectra were recorded on a Bruker Daltonics esquire 3000+. ESI-
TOF mass spectra were recorded on a Bruker micrOTOFII. All re-
agents are commercially available and were used without further
purification. Solvents were stored over molecular sieves (Fluka)
and used directly without further purification.
experiments at temperatures between 37 and 70 °C. Interestingly,
Therminator DNA polymerase is proficient in incorporating eleven
adjacent CF
3
dU nucleotides at about 40 to 60 °C (Fig. 4b, left panel,
lanes 3–8). In case of 1b and 1c Therminator DNA polymerase
(
a)
(b)
(c)
5’---TC
3’---AG(A)
0
4
.1.1. 5-Trifluoromethyl-2 -deoxyuridine 3a
5
’---TC
’---AG(ANNN) AT
5’---TC
3’---AG(AN)
To cleave the acetyl protecting groups, compound 2a (445 mg,
1.17 mmol) was treated with a 7 M ammonia solution of methanol
4.5 mL) for 19 h at room temperature (TLC analysis EtOAc/petro-
leum ether 5/1). The solvent was removed in vacuo and the crude
product was purified by silica gel chromatography (eluent: EtOAc/
3
11
23
46
(
6
9
69
69
petroleum ether 5/1) to yield 3a (264 mg, 76%). TLC (EtOAc/petro-
1
leum ether 5/1) R
f
0.34. H NMR (600 MHz; MeOH-d
4
) d (ppm)
0
0
2
.25–2.29 (m, 1H, H2 a), 2.35–2.39 (m, 1H, H2 b), 3.75 (dd, 1H,
0
2
3
0
2
H5 a,
J
H5
0
b
= 12 Hz,
J
H4 H5 a
= 3 Hz), 3.84 (dd, 1H, H5 b, J = 12 Hz,
0
= 3 Hz), 3.97 (m, 1H, H4 ), 4.41 (m, 1H, H3 ), 6.24 (dd, 1H,
0
J = 6 Hz), 8.79 (s, 1H, H6). C NMR (151 MHz; MeOH-d )
H2 4
0
3
0
0
J
H4
0
0
3
13
H1 ,
d (ppm) 42.15, 62.17, 71.73, 87.58, 89.34, 105.35 (q, 1C, C5,
2
1
J
F
= 33 Hz), 123.98 (q, 1C, CF
= 6 Hz), 151.37, 161.27.
3
,
F
F
J = 269 Hz), 143.82 (q, 1C, C6,
3J
19
NMR (376 MHz; MeOH-d
d
F
4
)
ꢀ
(
2
ppm) ꢀ64.4 (s, 3F, CF
3
). ESI-IT-MS calcd for (MꢀH) 295.1, found
23
ꢀ
23
23
94.8. ESI-TOF-MS calcd for (MꢀH) 295.0542, found 295.0536.
0
4
.1.2. 5-Pentafluoroethyl-2 -deoxyuridine 3b
The deacetylation of compound 2b (140 mg, 0.33 mmol) was
performed in dichloromethane and methanol (9:1; 10 mL) with
toluene-4-sulfonic acid monohydrate (309 mg, 1.63 mmol) for
Figure 3. Multiple incorporation of modified nucleotides. Primer template
sequences are indicated on the top of every panel. (a–c) Incorporation by KF(exo-)
catalysis; Primer: primer only; ꢀTTP: primer extension in the presence of dATP,
3
days at room temperature to get full turnover of the reactant
(
TLC analysis EtOAc/petroleum ether 5/1). The solvent was re-
dGTP and dCTP; +TTP: as - TTP but in the presence of TTP; CF
3
dUTP: as ꢀTTP but in
dUTP: as ꢀTTP
moved in vacuo and the crude product was purified by RP-MPLC
(linear grade from water with 5% acetonitrile to 100% acetonitrile)
the presence of 1a; C
2
F
5
dUTP: as ꢀTTP but in the presence of 1b; C
3 7
F
but in the presence of 1c.

3
656
B. Holzberger, A. Marx / Bioorg. Med. Chem. 17 (2009) 3653–3658
(
a)
Primer (23 nt):
Template (49 nt):
5’---TC
3’---AGA AAA AAA AAA AGG TTG CGT GGT CCG T
(
b)
49
34
23
0
1
12 0 1
12 0 1
12
CF dUTP
C F dUTP
C F dUTP
3 7
3
2
5
Figure 4. Temperature dependent polymerization of 1a–c by Therminator DNA polymerase; lanes 0: primer only, lanes 1–12: primer extension in the presence of dATP,
dGTP, dCTP and 1a, 1b, or 1c, respectively; incubation temperatures: 37.0, 37.8, 40.1, 43.7, 47.6, 51.5, 55.4, 59.4, 63.3, 66.9, 69.2 and 70.0 °C.
to yield 3b (84 mg, 74%). TLC (dichloromethane/methanol 9:1) R
f
added dropwise at 0 °C. After full turnover of the reactant (TLC
analysis 2-propanol/NH /H O 3/1/1) a 0.5 M bis-tri-n-butylammo-
nium pyrophosphate solution of DMF (2.25 mL, 1.13 mmol) and n-
tributylamine (540 L, 2.27 mmol) were added simultaneously.
After stirring for additional 20 min 1 M TEAB-buffer (Et NH(HCO ),
pH 7.5) was added (10 mL). The aqueous layer was washed with
EtOAc (2 ꢂ 10 mL) and the solvent was removed in vacuo. The res-
idue was resolved in 0.1 M TEAB-buffer (3 mL) and purified by
sephadex ion exchange chromatography (eluent: linear grade from
0.1 M to 1 M TEAB-buffer). The salts were removed by RP-MPLC
1
0
.30. H NMR (600 MHz; MeOH-d
4
) d (ppm) 2.23–2.29 (m, 1H,
3
2
0
0
0
2
H2 a), 2.37–2.41 (m, 1H, H2 b), 3.74 (dd, 1H, H5 a,
J
H5
0
b
= 12 Hz,
3
0
2
= 12 Hz, 3
J
H4
0
= 3 Hz), 3.83 (dd, 1H, H5 b,
J
H5
0
a
J
H4
0
= 3 Hz), 3.98
l
0
0
0
3
(
m, 1H, H4 ), 4.41 (m, 1H, H3 ), 6.24 (dd, 1H, H1 ,
J
H2
0
= 6 Hz),
3
3
1
3
8
6
.80 (s, 1H, H6). C NMR (151 MHz; MeOH-d
4
) d (ppm) 42.30,
2
2.16, 71.80, 87.79, 89.45, 103.05 (t, 1C, C5,
J
F
= 24 Hz), 113.71
1
2
(
tq, 1C, CF
J
2
,
J
F
= 254 Hz,
= 40 Hz), 145.85 (t, 1C, C6,
60.97. F NMR (376 MHz; MeOH-d
J = 40 Hz), 120.49 (tq, 1C, CF3,
F
1
2
3
F
= 286 Hz,
J
F
F
J = 10 Hz), 151.29,
1
9
1
CF
4
) d (ppm) ꢀ114.8 (s, 1F,
2
a), ꢀ114.8 (s, 1F, CF
2
b), ꢀ86.0 (s, 3F, CF
3
). ESI-IT-MS calcd for
3
(eluent: grade from 50 mM TEAA-buffer (Et NH(OAc), pH 7.0) with
ꢀ
ꢀ
(
3
MꢀH) 345.1, found: 344.8. ESI-TOF-MS calcd for (MꢀH)
5% acetonitrile to 100% acetonitrile) to yield 1a (18 mg, 8%, trieth-
1
45.0510, found 345.0609.
ylammonium salt of the triphosphate). H NMR (400 MHz; MeOH-
0
d
4
) d (ppm) 1.30 (t, Et
3
0
N, Me), 2.32 (m, 2H, H2 ), 3.17 (q, Et
3
N, CH
2
),
0
0
0
4
.1.3. 5-(Heptafluoro-n-propyl)-2 -deoxyuridine 3c
4.09–4.22 (m, 2H, H5 ), 4.31 (m, 1H, H4 ), 4.59 (m, 1H, H3 ), 6.19
0
3
31
The deacetylation of compound 2c (161 mg, 0.33 mmol) was
(dd, 1H, H1 ,
MeOH-d
P
J
H2
0
= 7 Hz), 8.29 (s, 1H, H6). P NMR (162 MHz;
2
performed in dichloromethane and methanol (9:1; 10 mL) with tol-
uene-4-sulfonic acid monohydrate (319 mg, 1.68 mmol) for 3 days
at room temperature (TLC analysis EtOAc/petroleum ether 5/1). The
solvent was removed in vacuo and the crude product was purified
by RP-MPLC (linear grade from water with 5% acetonitrile to 100%
4
) d (ppm) ꢀ22.94 (dd, 1P, P
b
,
J
P
= 21 Hz), ꢀ10.66 (d, 1P,
2
2 19
a
,
J
Pb = 21 Hz), ꢀ9.56 (d, 1P,
P
c
,
J
Pb = 21 Hz).
F NMR
(376 MHz; MeOH-d
4
) d (ppm) ꢀ64.7 (s, 3F, CF
3
). ESI-IT-MS calcd
ꢀ
ꢀ
for (MꢀH) 534.9, found 534.6. ESI-TOF-MS calcd for (MꢀH)
534.9537, found 534.9651.
acetonitrile) to yield 3c (87 mg, 66%). TLC (dichloromethane/meth-
1
0
0
anol 9:1) R
f
0.36. H NMR (600 MHz; MeOH-d
4
) d (ppm) 2.22–2.28
4.1.5. 5-Pentafluoroethyl-2 -deoxyuridine-5 -triphosphate 1b
The nucleoside 3b (32 mg, 0.09 mmol) and 1,8-bis(dimethyl-
amino)naphthalene (30 mg, 0.14 mmol) were dried in the dark in
vacuo overnight. Then the mixture was dissolved under argon in
0
0
0
(
m, 1H, H2 a), 2.35–2.43 (m, 1H, H2 b), 3.74 (dd, 1H, H5 a,
2
3
3
0
2
J
J
H5
0
b
= 12 Hz,
J
H4
0
= 2 Hz), 3.82 (dd, 1H, H5 b,
H5 a
J = 12 Hz,
0
0
0
0
= 3 Hz), 3.98 (m, 1H, H4 ), 4.41 (m, 1H, H3 ), 6.25 (dd, 1H,
H4
0
3
13
H1 , JH2
0
= 6 Hz), 8.81 (s, 1H, H6). C NMR (151 MHz; MeOH-d
4
) d
trimethyl phosphate (1 mL) and POCl
added dropwise at 0 °C. After full turnover of the reactant (TLC
analysis 2-propanol/NH /H O 3/1/1) a 0.5 M bis-tri-n-butylammo-
nium pyrophosphate solution of DMF (0.92 mL, 0.46 mmol) and n-
tributylamine (220 L, 0.92 mmol) were added simultaneously.
After stirring for additional 20 min 1 M TEAB-buffer (Et NH(HCO ),
pH 7.5) was added (10 mL). The aqueous layer was washed with
EtOAc (2 ꢂ 10 mL) and the solvent was removed in vacuo. The res-
idue was resolved in 0.1 M TEAB-buffer (3 mL) and purified by
sephadex ion exchange chromatography (linear gradient from
0.1 M to 1 M TEAB-buffer). The salts were removed by RP-MPLC
3
(20 lL, 0.22 mmol) was
(
ppm) 42.38, 62.17, 71.85, 87.79, 89.50, 103.21 (t, 1C, C5,
2
2
1
J
J
F
= 24 Hz), 110.36 (m, 1C, CF
2
b), 115.69 (tq, 1C, CF
2
a
, J
F
= 256 Hz,
3
2
1
2
F
= 33 Hz), 119.49 (tq, 1C, CF3,
J
F
= 287 Hz, J
F
= 34 Hz), 146.19 (t,
3
19
1
C, C6, J
F
= 10 Hz), 151.26, 160.95. F NMR (376 MHz; MeOH-d
4
)
l
d (ppm) ꢀ127.5 (s, 2F, CF
2
b), ꢀ111.5 (m, 1F, CF
2
a
), ꢀ82.4 (t, 3F,
3
3
4
ꢀ
CF
3
,
J
Fa
= 10 Hz). ESI-IT-MS calcd for (MꢀH) 395.1, found 394.8.
ꢀ
ESI-TOF-MS calcd for (MꢀH) 395.0478, found 395.0595.
0
0
4
.1.4. 5-Trifluoromethyl-2 -deoxyuridine-5 -triphosphate 1a
The nucleoside 3a (67 mg, 0.23 mmol) and 1,8-bis-(dimethyl-
amino)naphthalene (72 mg, 0.34 mmol) were dried in the dark in
vacuo overnight. Then the mixture was dissolved under argon in
3
(gradient from 50 mM TEAA-buffer (Et NH(OAc), pH 7.0) with 5%
acetonitrile to 100% acetonitrile) to yield 1b (9 mg, 10%, triethyl-
1
trimethyl phosphate (1.8 mL) and POCl
3
(55
l
L, 0.60 mmol) was
ammonium salt of the triphosphate). H NMR (400 MHz; MeOH-

B. Holzberger, A. Marx / Bioorg. Med. Chem. 17 (2009) 3653–3658
3657
0
d
4
) d (ppm) 1.24 (t, Et
3
0
N, Me), 2.44 (m, 2H, H2 ), 3.26 (q, Et
3
N, CH
2
),
contained DNA polymerase, 200 nM template, 150 nM radioactive
0
0
0
0
0
4
.11–4.32 (m, 3H, H4 , H5 ), 4.66 (m, 1H, H3 ), 6.24 (dd, 1H, H1 ,
labelled primer (5 -d(GAC CCA CTC CAT CGA GAT TTC TC)-3 ) and
200
3
31
J
H2
0
= 7 Hz), 8.28 (s, 1H, H6). P NMR (162 MHz; MeOH-d
4
) d
l
M dNTPs in the according 1ꢂ reaction buffer. Used DNA poly-
2
(
(
ppm) ꢀ22.43 (m, 1P, P
b
19
), ꢀ11.04 (d, 1P, P
a
,
J
Pb = 19 Hz), ꢀ9.56
merase concentrations were (i) for reactions depicted in Figure 1b:
30 nM DNA polymerase KF(exo-), (ii) for reactions depicted in Fig-
ure 1c: 30 nM DNA polymerase KlenTaq, (iii) for reactions depicted
in Figure 3: 10 nM DNA polymerase KF(exo-) and (iv) for reactions
2
d, 1P, P
c
,
J
Pb = 17 Hz). F NMR (376 MHz; MeOH-d
4
) d (ppm)
ꢀ
ꢀ
114.6 (s, 2F, CF
2
), ꢀ85.5 (s, 3F, CF
3
). ESI-IT-MS calcd for (MꢀH)
ꢀ
5
84.9, found 584.5. ESI-TOF-MS calcd for (MꢀH) 584.9505, found
5
84.9688.
depicted in Figure 4: 0.05 U/lL Therminator DNA polymerase. The
reactions were incubated (i) for reactions depicted in Figure 1b and
c: for 20 min at 37 °C (KF(exo-)) and 72 °C (KlenTaq), (ii) for reac-
tions depicted in Figure 3: for 60 min at 37 °C and (iii) for reactions
depicted in Figure 4: for 60 min at 37.0, 37.8, 40.1, 43.7, 47.6, 51.5,
55.4, 59.4, 63.3, 66.9, 69.2 and 70.0 °C, respectively (lanes 1–12) in
0
0
4
.1.6. 5-(Heptafluoro-n-propyl)-2 -deoxyuridine-5 -
triphosphate 1c
The nucleoside 3c (31 mg, 0.08 mmol) and 1,8-bis-(dimethyl-
amino)naphthalene (25 mg, 0.12 mmol) were dried in the dark in
vacuo overnight. Then the mixture was dissolved under argon in
a thermocycler and were stopped by addition of 40 lL stop solu-
trimethyl phosphate (1 mL) and POCl
added dropwise at 0 °C. After full turnover of the reactant (TLC
analysis 2-propanol/NH /H O 3/1/1) a 0.5 M bis-tri-n-butylammo-
nium pyrophosphate solution of DMF (0.78 mL, 0.39 mmol) and n-
tributylamine (190 L, 0.80 mmol) were added simultaneously.
After stirring for additional 20 min 1 M TEAB-buffer (Et NH(HCO ),
pH 7.5) was added (10 mL). The aqueous layer was washed with
EtOAc (2 ꢂ 10 mL) and the solvent was removed in vacuo. The res-
idue was resolved in 0.1 M TEAB-buffer (3 mL) and purified by
sephadex ion exchange chromatography (eluent: linear grade from
3
(29
l
L, 0.32 mmol) was
tion (80% [v/v] formamide, 20 mM EDTA, 0.025% [w/v] bromophe-
nol blue, 0.025% [w/v] xylene cyanol). After denaturing at 95 °C for
5 min the reaction mixtures were separated using a 12% denatur-
ing PAGE gel. Visualization was performed using phosphorimaging.
The used templates were (i) for reactions depicted in Figure 1b
3
2
l
0
3
3
and c: 5 -d(GCG CTG GCA CGG GAG AAA TCT CGA TGG AGT GGG
0
TC) (35 nt), (ii) for reactions depicted in Figure 3a: 5 -d(TAG GGA
TCT ATT TAC TTA CCG ATT CAT TTA CGC AGT CAG GCA TTC AGA
GAA ATC TCG ATG GAG TGG GTC) (69 nt), (iii) for reactions de-
0
picted in Figure 3b: 5 -d(TAG ACA TAG ACA TAG ACA TAG ACA
0
(
5
.1 M to 1 M TEAB-buffer). The salts were removed by RP-MPLC
eluent: grade from 50 mM TEAA-buffer (Et NH(OAc), pH 7.0) with
% acetonitrile to 100% acetonitrile) to yield 1c (13 mg, 16%, trieth-
TAG ACA TAC AGA CAG ACA TAC AGA GAA ATC TCG ATG GAG
TGG GTC) (69 nt), (iv) for reactions depicted in Figure 3c: 5 -
0
3
d(AAA AAA AAA AAA AAA AAA AAA AAA AAA AAA AAA AAA AAA
AAA AAA AGA GAA ATC TCG ATG GAG TGG GTC) (69 nt) and (v)
1
ylammonium salt of the triphosphate). H NMR (400 MHz; d
4
-
0
0
MeOH) d (ppm) 1.30 (m, Et
m, Et N, CH
.15 (dd, 1H, H1 ,
162 MHz; MeOH-d
3
N, Me), 2.25–2.43 (m, 2H, H2 ), 3.17
), 4.09–4.31 (m, 3H, H4 , H5 ), 4.58 (m, 1H, H3 ),
for reactions depicted in Figure 4: 5 -d(TGC CTG GTG CGT TGG
0
0
0
(
6
(
ꢀ
3
2
AAA AAA AAA AAG AGA AAT CTC GAT GGA GTG GGT C) (49 nt).
0
3
31
J
H2
0
= 7 Hz), 8.15 (s, 1H, H6).
) d (ppm) ꢀ22.90 (dd, 1P, P
2J
Pb = 21 Hz).
P
NMR
= 21 Hz),
4
b
,
P
4.1.9. Single nucleotide incorporation assays (Fig. 2)
The 24 nt long primer was 5 - P-labelled using [
according to standard techniques. The reaction mixture (20
2
2
19
0
32
c
32P]-ATP
10.60 (d, 1P, P
a
,
J
Pb = 21 Hz), ꢀ9.56 (d, 1P, P
c
,
J
F
NMR (376 MHz; MeOH-d
4
) d (ppm) ꢀ127.1 (m, 2F, CF2b), ꢀ111.5
l
L)
4
4
0
(
q, 2F, CF
2
a
,
J
F
= 10 Hz), ꢀ82.2 (t, 3F, CF
3
,
J
F
= 10 Hz). ESI-IT-MS
contained 2 nM DNA polymerase KF(exo-), 60 nM template (5 -
d(GTG CGT CTG TCA/G/C/T TGT CTG TCA GAA ATT TCG CAC CAC),
ꢀ
ꢀ
calcd for (MꢀH) 634.9, found 634.5. ESI-TOF-MS calcd for (MꢀH)
0
6
34.9473, found 634.9729.
40 nM radioactive labelled primer (5 -d(GTG GTG CGA AAT TTC
0
TGA CAG ACA)-3 ) and 1, 10, 100
lM dNTPs (matched case) or
4
.1.7. Materials for DNA polymerase experiments
10, 50, 200 lM dNTPs (mismatched cases) in 1x reaction buffer
dNTPs were purchased from Roche.
KF(exo-). The reactions were incubated for 5 min (matched case)
or 30 min (mismatched cases) at 37 °C in a thermocycler and were
4
.1.7.1. Primers and templates.
Figures 1, 3 and 4: Tem-
stopped by addition of 40 lL stop solution (80% [v/v] formamide,
plates were obtained from Purimex (2ꢂ HPLC purified). The 23 nt
primer strand was synthesized on an Applied-Biosystems-392
DNA synthesizer and afterwards purified by preparative PAGE. Fig-
ure 2: Used primer and templates were synthesized on the DNA
synthesizer and afterwards purified by HPLC and preparative
PAGE.
20 mM EDTA, 0.025% [w/v] bromophenol blue, 0.025% [w/v] xylene
cyanol). After denaturing at 95 °C for 5 min the reaction mixtures
were separated using a 12% denaturing PAGE gel. Visualization
was performed using phosphorimaging.
Acknowledgements
4
.1.7.2. Enzymes.
merase I (KF(exo-)) was generated and purified as described.
The Klenow fragment of Thermus aquaticus DNA polymerase I
The Klenow fragment of E. coli DNA poly-
We thank M. Liebmann for her contributions to this work and
Konstanz Research School Chemical Biology for financial support.
2
1
KlenTaq) was generated and purified as described.2 The Thermi-
0f
(
References and notes
nator DNA polymerase (A485L mutant of Thermococcus species 9°N
DNA polymerase) was purchased from New England Biolabs.
1. (a) Seeman, N. C. Nature 2003, 421, 427; (b) Feldkamp, U.; Niemeyer, C. M.
Angew. Chem., Int. Ed. 2006, 45, 1856; (c) Lin, C. X.; Liu, Y.; Rinker, S.; Yan, H.
ChemPhysChem 2006, 7, 1641; (d) Deng, Z. X.; Lee, S. H.; Mao, C. D. Nanosci.
Nanotechnol. 2005, 5, 1954.
4
.1.7.3. Buffers.
10x reaction buffer KF(exo-): 500 mM Tris–
, 10 mM DTT, 0,5% Triton X-100, pH 7.3. 10ꢂ
HCl, 100 mM MgCl
reaction buffer KlenTaq: 500 mM Tris–HCl, 160 mM (NH ) SO ,
4 2 4
2
2.
(a) Weisbrod, S. H.; Marx, A. Chem. Commun. 2008, 5675; (b) Hocek, M.; Fojta,
M. Org. Biomol. Chem. 2008, 6, 2233.
3.
(a) Sakthivel, K.; Barbas, C. F., III Angew. Chem., Int. Ed. 1998, 37, 2872; (b)
Capek, P.; Cahova, H.; Pohl, R.; Hocek, M.; Gloeckner, C.; Marx, A. Chem. Eur. J.
2
5 mM MgCl
2
, 1% Tween 20, pH 9.2. 10ꢂ reaction buffer Thermina-
tor: 200 mM Tris–HCl, 20 mM MgSO
4 4 2 4
, 100 mM (NH ) SO , 100 mM
2007, 13, 6196; (c) Cahova, H.; Havran, L.; Brazdilova, P.; Pivonkova, H.; Pohl,
KCl, 1% Triton X-100, pH 8.8.
R.; Fojta, M.; Hocek, M. Angew. Chem., Int. Ed. 2008, 47, 2059; (d) Jaeger, S.;
Rasched, G.; Kornreich-Leshem, H.; Engeser, M.; Thum, O.; Famulok, M. J. Am.
Chem. Soc. 2005, 127, 15071; (e) Jaeger, S.; Famulok, M. Angew. Chem., Int. Ed.
4
.1.8. Primer extension assays (Figs. 1, 3 and 4)
2004, 43, 3337; (f) Thum, O.; Jaeger, S.; Famulok, M. Angew. Chem., Int. Ed. 2001,
0
32
32P]-ATP
L)
The 23 nt long primer was 5 - P-labelled using [
according to standard techniques. The reaction mixture (20
c
40, 3990; (g) Rasched, G.; Ackermann, D.; Schmidt, T. L.; Broekmann, P.; Heckel,
A.; Famulok, M. Angew. Chem., Int. Ed. 2008, 47, 967.
l

3
658
B. Holzberger, A. Marx / Bioorg. Med. Chem. 17 (2009) 3653–3658
4
.
(a) Vrabel, M.; Horakova, P.; Pivonkova, H.; Kalachova, L.; Cernocka, H.; Cahova,
H.; Pohl, R.; Sebest, P.; Havran, L.; Hocek, M.; Fojta, M. Chem. Eur. J. 2009, 15,
144; (b) Weizman, H.; Tor, Y. J. Am. Chem. Soc. 2002, 124, 1568.
(a) Gramlich, P. M. E.; Wirges, C. T.; Gierlich, J.; Carell, T. Org. Lett. 2008, 10, 249;
b) Salic, A.; Mitchison, T. J. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 2415; (c)
Weisbrod, S. H.; Marx, A. Chem. Commun. 2007, 1828; (d) Baccaro, A.;
Weisbrod, S. H.; Marx, A. Synthesis 2007, 1949.
Keller, S.; Wang, J.; Chandra, M.; Berger, R.; Marx, A. J. Am. Chem. Soc. 2008, 130,
13. Gmeiner, W. H.; Pon, R. T.; Lown, J. W. J. Org. Chem. 1991, 56, 3602.
14. Markley, J. C.; Chirakul, P.; Sologub, D.; Sigurdsson, S. T. Bioorg. Med. Chem. Lett.
2001, 11, 2453.
15. (a) Satake, H.; Takeda, S.; Wataya, Y. Nucleic Acids Symp. Ser. 1991, 25, 37; (b)
Satake, H.; Takeda, S.; Matsumura, A.; Sasaki, M.; Sugimoto, N.; Wataya, Y.
Nucleic. Acids Symp. Ser. 1992, 27, 189.
16. Tanabe, Y.; Matsuo, N.; Ohno, N. J. Org. Chem. 1988, 53, 4582.
17. (a) Takatori, S.; Matsuda, A.; Yamashita, J.-I.; Hayatsu, H.; Wataya, Y.
Nucleosides, Nucleotides Nucleic Acids 1994, 13, 2105; (b) Wataya, Y.; Sonobe,
Y.; Maeda, M.; Yamaizumi, Z.; Aida, M.; Santi, D. V. J. Chem. Soc., Perkin Trans. 1
1987, 2141; (c) Santi, D. V.; Sakai, T. T. Biochemistry 1971, 10, 3598.
18. (a) Yoshikawa, M.; Kato, T.; Takenishi, T. Bull. Chem. Soc. Jpn. 1969, 42, 3505; (b)
Kovács, T.; Ötvös, L. Tetrahedron Lett. 1988, 29, 4525.
1
5
.
(
6
.
.
1
3188.
Obeid, S.; Yulikov, M.; Jeschke, G.; Marx, A. Angew. Chem., Int. Ed. 2008, 47,
782.
7
6
8
9
.
.
Famulok, M.; Hartig, J. S.; Mayer, G. Chem. Rev. 2007, 107, 3715.
Recent examples: (a) Hollenstein, M.; Hipolito, C.; Lam, C.; Dietrich, D.; Perrin,
D. M. Angew. Chem., Int. Ed. 2008, 47, 4346; (b) Hollenstein, M.; Hipolito, C. J.;
Lam, C. H.; Perrin, D. M. Nucleic Acids Res. 2009, 37, 1638.
19. Clark, J. M. Nucleic Acids Res. 1988, 16, 9677.
20. (a) Ghadessy, F. J.; Ramsay, N.; Boudsocq, F.; Loakes, D.; Brown, A.; Iwai, S.;
Vaisman, A.; Woodgate, R.; Holliger, P. Nat. Biotechnol. 2004, 22, 755; (b)
Leconte, A. M.; Chen, L.; Romesberg, F. E. J. Am. Chem. Soc. 2005, 127, 12470; (c)
Vichier-Guerre, S.; Ferris, S.; Auberger, N.; Mahiddine, K.; Jestin, J.-L. Angew.
Chem., Int. Ed. 2006, 45, 6133; (d) Xia, G.; Chen, L.; Sera, T.; Fa, M.; Schultz, P. G.;
Romesberg, F. E. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 6597; (e) Sauter, K. B. M.;
Marx, A. Angew. Chem., Int. Ed. 2006, 45, 7633; (f) Gloeckner, C.; Sauter, K. B. M.;
Marx, A. Angew. Chem., Int. Ed. 2007, 46, 3115.
1
0. (a) Purser, S.; Moore, P. R.; Swallow, S.; Gouverneur, V. Chem. Soc. Rev. 2008, 37,
20; (b) Müller, K.; Faeh, C.; Diederich, F. Science 2007, 317, 1881; (c) Pongdee,
3
R.; Liu, H. Bioorg. Chem. 2004, 32, 393.
1
1. O’Hagan, D. Chem. Soc. Rev. 2008, 37, 308.
1
2. (a) Kreutz, C.; Micura, R. In Modified Nucleosides; Herdewijn, P., Ed.; Springer:
Berlin, 2008; p 3. Chapter 1; (b) Graber, D.; Moroder, H.; Micura, R. J. Am. Chem.
Soc. 2008, 130, 17230; (c) Kreutz, C.; Kählig, H.; Konrat, R.; Micura, R. J. Am.
Chem. Soc. 2005, 127, 11558; (d) Barhate, N. B.; Barhate, R. N.; Cekan, P.;
Drobny, G.; Sigurdsson, S. T. Org. Lett. 2008, 10, 2745.
21. Summerer, D.; Marx, A. Angew. Chem., Int. Ed. 2002, 41, 3620.