Synthesis and enzymatic incorporation of modified deoxyuridine triphosphates

Article abstract of DOI:10.1039/b906956a

We describe the synthesis of 2′-deoxyuridine-5′-triphosphate derivatives bearing linkers of varying length, bulk and flexibility, at position 5 of the pyrimidine base. Nucleotide analogues with terminal functional groups are of interest due to their appli

Full text of DOI:10.1039/b906956a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Synthesis and enzymatic incorporation of modified deoxyuridine
triphosphates†
Vinciane Borsenberger,‡ Mikiembo Kukwikila‡ and Stefan Howorka*
Received 6th April 2009, Accepted 24th June 2009
First published as an Advance Article on the web 21st July 2009
DOI: 10.1039/b906956a
We describe the synthesis of 2¢-deoxyuridine-5¢-triphosphate derivatives bearing linkers of varying
length, bulk and flexibility, at position 5 of the pyrimidine base. Nucleotide analogues with terminal
functional groups are of interest due to their application potential for the functional labelling of DNA
strands. In the course of the synthesis of the nucleotide analogues, the methodology for the Yoshikawa
phosphorylation procedure was optimised, resulting in an approach which reduces the amount of
side-products and is compatible with labile functional groups attached to the base. The effect of linker
composition on the enzymatic incorporation into DNA was systematically investigated using two
different DNA polymerases. Deep VentR exo- from the B-polymerase family accepted most nucleotide
analogues as substrates, while Taq from the A-family was slightly less proficient. Both polymerases had
difficulties incorporating 5-(3-amino-prop-1-ynyl)-2¢-deoxyuridine triphosphate. A molecular model of
the active site of the polymerase was used to rationalise why this nucleotide was not accepted as a
substrate.
catalysed cross-coupling reactions. For the Sonogashira and
Introduction
Suzuki–Miyaura reactions,14,15 halogenated nucleosides are used in
combination with linkers carrying terminal alkyne or aryl/vinyl
boronic acids groups, respectively. The cross-coupling reactions
are attractive due to their mild conditions which enable the
tagging of thermally sensitive substrates. In addition, the reactions
can be performed in aqueous conditions to simplify work-up.
The attractive cross-coupling reactions have been exploited12,16–24
in the pre-modification approach to introduce tags such as
fluorophores,9,25 peptides,26 and transition metal complexes.27,28
The synthesis and use of nucleotides for the post-modification
approach has, however, attracted less attention. Nucleotides have
been prepared with terminal alkyne groups for click-couplings,3 a
diene for Diels–Alder cycloadditions,7,29 and an amino group12,22
for amide coupling. In general, linkers for this strategy must meet
criteria with respect to linker length and terminal functional group.
For example, the linker must be long enough to ensure that the
functional group remains accessible for covalent coupling to the
functional tags. At the same time, the linkers have to be short
enough to avoid back-folding onto the DNA strand which would
reduce the rate of reaction. Furthermore, the terminal functional
group of the linker must be tolerated by DNA polymerase
enzymes.
Herein we present a systematic study on the effect of sub-
stituent length, charge, and bulkiness on the enzyme-catalysed
incorporation of deoxyuridine triphosphate analogues. Firstly, we
synthesised nine nucleotides which bear functional substitutions
at position 5 of deoxyuridine (Fig. 1). The modifications include
amine and carboxyl functionalities, alkyne and diene groups,
as well as bulky Diels–Alder-coupled peptide tags to probe
the influence of steric factors. The compounds were synthe-
sised via Sonogashira cross-coupling reactions from 5-iodo-2¢-
deoxyuridine using the corresponding acetylene-based linkers.
The resulting nucleosides were converted into deoxyribonucleotide
triphosphates using the Yoshikawa procedure.
The generation of chemically modified deoxyribonucleotide
triphosphates is of importance in light of their application
in chemical biology, structural biology, DNA sensing, and
nanobiotechnology.1 In many cases, deoxyribonucleotide triphos-
phates carrying functional tags such as fluorophores are enzy-
matically incorporated into DNA strands. This pre-modification
approach is complemented by the two-step post-modification
strategy in which nucleotide precursors carrying a linker are first
enzymatically introduced into DNA and then covalently deriva-
tised with functional tags. The coupling to the tags can exploit
e.g. click chemistry,2–4 Diels–Alder cycloaddition,5–7 or maleimide,
disulfide,8 or amide couplings. The post-modification approach
is particularly useful for large tags which are not accepted by
polymerases such as oligosaccharides3 or metal nanoparticles.
In deoxyuridine and deoxycytidine, position 5 of the pyrimidine
ring is widely used to introduce new functionalities.9 Within the
pre-modification approach, nucleotides carrying e.g. a fluorophore
attached via a long linker at position 5 are usually well accepted
by DNA polymerases,3,9–12 as this site is not involved in Watson–
Crick base-pairing thereby minimising any distortion in the shape
of the base pairs. With regard to the post-modification approach,
substituents at position 5 protrude from the major groove of the
DNA duplex thereby ensuring good steric accessibility to facilitate
coupling to reagents.13
Nucleotide analogues carrying a non-biogenic substitution
at position 5 can be chemically synthesised using palladium-
University College London, Department of Chemistry, 20 Gordon Street,
London, UK WC1H OAJ. E-mail: s.howorka@ucl.ac.uk; Fax: +44 (0) 20
7679 7463; Tel: +44 (0) 20 7679 4702
† Electronic supplementary information (ESI) available: Details of
the optimisation of the Yoshikawa phosphorylation procedure. See
DOI: 10.1039/b906956a
‡ V. B. and M. K. contributed equally to this work.
3826 | Org. Biomol. Chem., 2009, 7, 3826–3835
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
product (Scheme 1, step ii) or reacted with tributylammonium
pyrophosphate affording the triphosphate deoxyribonucleotide
(Scheme 1, step iii).3,33–35 The reaction with POCl3 is often per-
formed with unprotected deoxyribonucleosides to predominantly
achieve 5¢-selective phosphorylation. However, the regioselectiv-
ity is not perfect, and, consequently, 3¢-phosphate and 3¢,5¢-
diphosphate by-products are encountered.36 While 3¢-protection
strategies can be enlisted to address this issue, it is advantageous
to use unprotected starting material. Indeed, several publications
have reported that the extent of side reactions can be suppressed
by replacing the traditionally used trialkyl phosphate solvent
with other solvent systems or by employing a reaction time of
2–5 hours.35,37 However, this approach cannot be universally ap-
plied to all deoxyribonucleosides due to solubility issues. Our study
therefore investigated, secondly, the Yoshikawa phosphorylation
procedure in an attempt to optimise yield and regioselectivity for
non-protected nucleosides.
Finally, the third aim of our study was to test the incorporation
of the nucleotide analogues using DNA polymerases. We chose
DNA polymerases Taq and Deep Vent (exo-) which represent
the A and B family, respectively. Both enzymes are widely used
in the thermal amplification reaction and functional tagging of
DNA.3,7,10
Fig. 1 2¢-Deoxyuridine-5¢-triphosphate bearing different substituents at
position 5 of the pyrimidine. R¢ are peptide tags H6, GVKRKKKP, or
GKDDDDYD.
Results and discussion
Synthesis of modified nucleosides
The Yoshikawa procedure30,31 is a simple and popular phospho-
rylation approach1,32 that involves the reaction of a deoxyribonu-
cleoside with phosphorus oxychloride (POCl3) (Scheme 1, step i)
to yield a highly reactive phosphorodichloridate intermediate.30,31
The reactive intermediate can subsequently either be hydrolysed
by triethylammonium bicarbonate (TEAB) to the monophosphate
The synthesis of the modified nucleosides and their corresponding
nucleotides is outlined in Scheme 2. The generation of each nucle-
oside began with a Sonogashira cross-coupling reaction between
5-iodo-2¢-uridine and acetylene-based moieties in the presence
of Et3N, CuI, Pd(PPh3)4 and DMF. In each reaction, a ratio
of approximately 2:1 palladium:copper was employed to avoid
oxidation of the catalyst. Furthermore, air was excluded from
reaction vessels by repetitive purging and evacuation with argon.
Using this procedure, the modified nucleosides were obtained in
yields ranging from 33–73% (see Experimental).
The first nucleoside, 5-ethynyl-2¢-deoxyuridine, 2a was obtained
via a cross-coupling reaction with trimethylsilylacetylene, followed
by deprotection to remove the silyl group (Scheme 2, step ii).3
Deprotection with TBAF proved to be more successful than with
sodium carbonate in methanol/water.
The amino-propargyl-substituted derivative, 3c, was synthesised
through a cross-coupling reaction with trifluoroacetate-protected
propargylamine yielding 3a (Scheme 2, step iii). The protected nu-
cleoside was then phosphorylated using an optimised Yoshikawa
procedure (see below). The trifluoroacetate protecting group was
removed by treatment with concentrated ammonia affording the
free amino-terminated nucleotide 3c (Scheme 2, step iv).
3a was also used to form diene-terminated 3d (Scheme 2) by,
firstly, deprotecting the amine (Scheme 2, step iv) whereby repeated
lyophilisation steps were key to eliminate the volatile TFA and
ammonium salts. Thereafter, the free amine 3b was reacted in a
one-pot EDC·HCl-mediated amide coupling to potassium sorbate
(Scheme 2, step v). HPLC monitoring of the reaction of 3b showed
near quantitative conversion of the amine to 3d (data not shown).
It is noted that this one-pot reaction with EDC did not result
in any side-products stemming from the possible reaction of the
Scheme 1 The Yoshikawa phosphorylation via the phosphorodichlo-
ridate intermediate yields mono- and triphosphorylated nucleotides.
(i) proton sponge, POCl3; (ii) TEAB; (iii) nBu3NH+·H3P2O7-; TEAB.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 3826–3835 | 3827
©

View Article Online
Scheme 2 Synthesis of modified nucleotides. Reagents: (i) POCl3, proton sponge, TEAB, nBu3NH+·H3P2O7-; TEAB; (ii) CuI, Et3N, DMF, Pd(PPh3)4,
trimethylsilylacetylene; TBAF; (iii) CuI, Et3N, DMF, Pd(PPh3)4, N-propynyltrifluoroacetamide; (iv) TEAB; conc. NH3; (v) DMF, potassium sorbate,
EDC·HCl; (vi) CuI, Et3N, DMF, Pd(PPh3)4, pent-4-ynoic methyl ester; (vii) conc. NH3; (viii) CuI, Et3N, DMF, Pd(PPh3)4, pent-4-ynoic acid
hexa-2,4-dienyl ester; (ix) maleimide dienophile R¢1 = H6, R¢2 = GVKRKKKP, R¢3 = GKDDDDYD.
3828 | Org. Biomol. Chem., 2009, 7, 3826–3835
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
activated ester with the hydroxyl groups of the nucleoside. By
contrast, coupling routes using N-hydroxysuccimide or other more
reactive activated esters exhibit less selectivity for the amine and
show more side-reactions with hydroxyl groups.38
unsatisfactory results prompted a detailed investigation into the
Yoshikawa reaction with respect to yield and regioselectivity as
a function of reaction duration, temperature and molar excess of
POCl3.
For the synthesis of the carboxyl-terminated derivative 4b, pent-
4-ynoic methyl ester was first coupled to position 5 (Scheme 2,
step vi). The protected nucleoside 4a was then phosphorylated
to the triphosphate 4b (see below). Treatment with concentrated
ammonia removed the protecting methyl group and afforded the
free carboxylic acid.
To examine the regioselectivity and yield, the model nucleoside
5-iodo-2¢-deoxyuridine was reacted with 1.5 equivalents of POCl3
◦
at 0 C, similar to published conditions.9,33 At intervals, aliquots
of the reaction mixture were withdrawn, treated with TEAB to
hydrolyse any phosphorodichloridate intermediate (Scheme 1, step
iii), and analysed by RP-HPLC. The chromatogram of the reaction
mixture after 3.5 min indicates that the starting material (Fig. 2A,
0 min, symbol S) was partially converted into peaks A, B, and C
(Fig. 2A 3.5 min). These peaks constitute the 5¢,3¢-diphosphate,
5¢-monophosphate, and 3¢-phosphate derivatives, respectively, as
demonstrated by ESI-MS and 1H analysis. Continued incubation
of up to 27 min led to a further decrease in starting material, an
increase of 5¢-monophosphate product peak B and diphosphate
by-product peak A, but no major changes for 3¢-phosphate
peak C (Fig. 2A). A summary of the observed reaction kinetics
(Fig. 2B) highlighted that the level of the 5¢-monophosphate
rose fast within the first 5 min to around 70% and levelled off
thereafter (Fig. 2B, filled circles). By comparison, diphosphate
increased steadily and constituted 15% at 5 min and close to
30% at 27 min (Fig. 2B, empty squares). This suggests an
optimal reaction window between 3 to 7 min that achieves high
levels of target and a low level of side-product. The short time
needed to complete the phosphorylation reaction is an important
finding as most other reports use a reaction time of at least
2 h.9,33
After investigating yield and regioselectivity as a function of
time, we examined the influence of two other parameters: the
reaction temperature and the molar ratio of POCl3 to deoxyri-
bonucleoside. As detailed in the ESI†, the phosphorylation of 5-
iodo-2¢-deoxyuridine 1 was conducted at either 0 ◦C or -13 ◦C, and
at molar ratios of 1.0, 1.5, or 3.0 equivalents of POCl3. The kinetic
analysis via HPLC found, firstly, that a reduction in temperature
to -13 ◦C at a ratio 1.5 had only a minor effect on the generation
of 5¢-phosphate, but a beneficial consequence by slowing down the
formation of the diphosphate by-product. Secondly, increasing the
ratio to 3.0 equivalents at -13 ◦C neither affected product nor by-
product formation. Thirdly, a ratio of 1.0 avoided the generation
of diphosphate by-product but at the cost of decreased yield of the
product.
A second diene-modified nucleoside, 5a, was prepared via
a cross-coupling reaction with sorbyl-4-pentyonate (Scheme 2,
step viii). Subsequent phosphorylation (see below) yielded target
nucleotide 5b.7 The diene-modified nucleotide was reacted with
a series of maleimide-terminated peptides to form three Diels–
Alder adducts 5c-R1, 5c-R2 and 5c-R3 (Scheme 2, step ix). The
peptides were obtained by solid-phase peptide synthesis, and
their sequences were chosen to produce functionally relevant tags.
The first peptide in 5c-R1 is the hexahistidine tag7 which can be
useful for metal affinity chromatography purification. The nuclear
localisation sequence of the second tag (GVKRKKKP in 5c-R2)
is used in cell-permeating peptide–oligonucleotide conjugates.39
Finally, the FLAG epitope tag is employed in cell biological
fluorescence labelling (GKDDDDYD in 5c-R3). The Diels–Alder
couplings between the peptide–maleimide dienophile and the
nucleotide diene were conducted at pH 6.1 to avoid the base-
catalysed hydrolysis of the maleimide.7
All nucleosides and their corresponding nucleotides were puri-
fied by RP-HPLC or flash chromatography and their chemical
identity was confirmed by ESI-MS, MALDI-TOF 1H-NMR,
13C-NMR and 31P-NMR. The modified dNTPs were stored as
lyophilisates at -20 ◦C and all are stable for several months except
5c-R3 (see Experimental).
Phosphorylation of nucleosides
The phosphorylation of the nucleosides was conducted using
the Yoshikawa procedure. Following a published protocol,9 we
first reacted 5-iodo-2¢-deoxyuridine 1 with POCl3 followed by
incubation with tributylammonium phosphate for 2 h to obtain
the triphosphate nucleotide (Scheme 1, steps i and iii). The
methodology yielded the target compound (not shown), but also
produced a high content of the diphosphate by-product. These
Fig. 2 (A) HPLC traces for the reaction of 1 with 1.5 equivalents of POCl3 at 0 ◦C. Y-scale: Absorbance at 260 nm. (B) Corresponding plot of peak
areas vs. time. (C) IEX chromatogram of the product mixture for the phosphorylation of 3e with POCl3 and pyrophosphate.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 3826–3835 | 3829
©

View Article Online
The optimal reaction conditions inferred from the mono-
phosphorylation of 1 (1.5 equivalents of POCl3, temperature
of at least -13 ◦C) were applied to generate the triphosphate
derivative of our nucleosides. Following Scheme 1 (steps i to iii), the
functionalised deoxyuridines were triphosphorylated by treatment
with pyrophosphate of the dichloridate species, and subsequently
hydrolysed. Fig. 2C illustrates a typical example of an anion
exchange chromatogram (IEX) of the crude reaction mixture,
with one major component (Fig. 2C, peak C) corresponding
to the expected triphosphate, and minor species being mono-
, di- and tetra-phosphate nucleotides (Fig. 2C, peaks A, B,
and D, respectively). The improved Yoshikawa methodology was
extended to the other modified nucleosides giving rise to the
corresponding triphosphate derivatives 2b, 3c, 3e, 4b, 5b, 5c-R1, 5c-
R2, 5c-R3 (Scheme 2) in quantitative yields. The successful outcome
of the phosphorylation reactions illustrates that the improved
Yoshikawa methodology is applicable to a wider range of nucleo-
side analogues. Furthermore, it is noted that the phosphorylation
of 5b carrying the labile allylic ester was well tolerated and did not
lead to hydrolysis.
into the primer. Consequently, no extension is expected to occur
for dNTP mix 0. Conversely, dNTP mix F contains a full set of
natural bases, and the primer should be extended by 12 nt to
the same length as the template (Fig. 3B). By comparison, dNTP
mix H lacks dGTP, and the primer is to be extended halfway
by 6 nt (Fig. 3B). Similarly, the assay to test for the enzymatic
incorporation of the nucleotide analogues was conducted via
primer extension by six or 12 bases. The corresponding nucleotide
mixes dNTP 6 and dNTP 12 contain the chemically tagged uridine
derivates in place of dTTP (Fig. 3B). Based on the sequence of the
template, the primer extension should lead to the incorporation of
two and three dU*TP bases, respectively (Fig. 3B).
The primer extension assays were analysed by determining the
length of the extension products via gel electrophoresis, followed
by staining of the DNA bands. The gel analysis is summarised
in Fig. 4. Panel A shows the control reactions which confirm the
viability of the primer extension assay. In Fig. 4A, lane 1 and 2
correspond to the 36 nt-long template and the 24 nt-long primer,
respectively. By comparison, the use of complete dNTP mix F led
to the disappearance of the non-extended primer due to the full
extension of the primer which co-migrated with the template in a
single band (Fig. 4A, lane 4). By contrast, dNTP mix H which is
designed to direct the extension by 6 nt resulted in an up-shift of
the primer band (Fig. 4A, lane 5) but not to the co-migration as
seen for dNTP mix F (Fig. 4A, lane 4). Furthermore, no extension
occurred for dNTP 0 (Fig. 4A, lane 3), in line with expectations.
The results of the 6-nt extension reaction using modified
deoxyuridine derivatives and Deep Vent is shown in Fig. 4B. The
enzymatic reaction with 1a, 2b, 4b, 3e, 5b, and peptide-tagged
analogues 5c-R1, 5c-R2, 5c-R3 proceeded successfully, as indicated
by a major up-shifted band stemming from the extended primer
(Fig. 4B, lanes 1, 2, 4 to 9). Enzymatic incorporation of 5-ethynyl-
2¢-deoxyuridine, 2b and diene-modified nucleotide 5b was also
observed in other studies.3,7 These two nucleotides hence serve
as positive control and reference for our extension assay. The
extended primers carrying the modified nucleotides migrated at
about the same height as the strand from natural nucleotide mix
H (compare Fig. 4B, lanes 1, 2, 4 to 6 with Fig. 4A, lane 5). The
absence of any tag-induced gel shift is most likely due to the small
running distance in the gel and the low mass of the chemical
tags. By contrast, the bands obtained from peptide-modified
nucleotides 5c-R1, 5c-R2, and 5c-R3 showed a considerable gel
shift (Fig. 4B, lanes 7 to 9) reflecting the approximately ten-fold
higher masses of the peptide tags compared to the other chemical
tags. These results suggest that in each case the enzyme is able to
extend the primer by 6 nucleotides and thereby incorporate two
modified nucleotides (Fig. 3B, dNTP mix 6). Interestingly, almost
no incorporation occurred for the amino-propargyl derivative 3c
(Fig. 4B, lane 3). The inability of 3c to serve as substrate is most
likely related to the terminal amino group of the substituent at
position 5 (see further below).
Enzymatic incorporation of modified nucleotides
All deoxyuridine triphosphate analogues were utilised in primer
extension assays to evaluate their ability to act as substrates for
commercially available DNA polymerases Deep Vent exo- and
Taq. The principle of the assays is schematically summarised in
Fig. 3A. A short primer is hybridised to a complementary section
of a longer DNA oligonucleotide. Its non-hybridised section serves
as a template to direct the extension of the primer via the enzyme-
catalysed addition of nucleotides. As the template contains three
adenine bases (underlined in Fig. 3A), the primer can be extended
by incorporating up to three modified deoxyuridine residues
dU*TP. The sequence of the template is designed so that the
extension can be controlled by the inclusion or omission of
particular bases (Fig. 3B). For instance, dNTP mix 0 lacks a
uridine or thymine which is the first nucleotide to be incorporated
Similar results were found for the extension by 12 nt using Deep
Vent (Fig. 4C). Fully extended primers were obtained for 1a, 2b,
4b, 3e and 5b (Fig. 4C, lanes 1, 2, 4 to 6), while amino-propargyl
3c was not accepted as substrate (Fig. 4C, lane 3). A lower
incorporation extent was found for peptide-modified nucleotides
5c-R1, 5c-R2, and 5c-R3 (Fig. 4C, lanes 7 to 9). In particular, the
addition of a third nucleotide into the partly extended primer
was likely blocked due to steric reasons. Complete incorporation
Fig. 3 Template-directed primer extension. The primer and template
form a duplex, and the non-hybridised section of the template directs
the polymerase-catalysed addition of nucleotides to the 3¢ of the primer.
(B) dNTP mixes of varying composition determine the number of
incorporated nucleotides comprising the nucleotide analogue dU*TP.
3830 | Org. Biomol. Chem., 2009, 7, 3826–3835
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
Fig. 4 Denaturing PAGE analysis of extension reactions with modified nucleotides using Deep Vent exo- (B,C) and Taq (D, E) for dNTP mixes 6 and
12 as defined in Fig. 3B, conducted at 57 ◦C. (A) are control reactions with dNTP mixes 0, F, H, as defined in Fig. 3B. Lane 1 and 2 contain template, T,
and primer, P, respectively. (F) shows the extension conducted at 70 ◦C using mix H, and mixes 6 or 12 containing 1a or 3c.
was probably hindered because the third and first tagged base were
separated by an almost complete helical turn (10 bases) potentially
leading to steric crowding between the two peptides positioned at
the same side of the duplex.
The assay with Taq polymerase yielded results as shown in
Fig. 4D and Fig. 4E. Incorporation by 6 nt was mostly complete
for 1a, 2b, 4b, 3e, 5b, 5c-R1, and 5c-R2 (Fig. 4D, lanes 1, 2, 4 to 8). In
comparison to Deep Vent, the extent of incorporation by Taq was
slightly lower, as exemplary illustrated by a range of minor bands
stemming from prematurely terminated extension product of 3e
(Fig. 4D, lane 5). As in the case of Deep Vent, amino-propargyl
nucleotide 3c was a poor substrate for enzymatic polymerisation by
Taq (Fig. 4D, lane 3), even though another study reported good
incorporation, albeit with the different enzyme Tth.12 Our low
extent of incorporation for 5c-R3 carrying a FLAG peptide epitope
(Fig. 4D, lane 9) reflected the partial hydrolysis of the triphosphate
(IEX analysis not shown) which was most likely caused by the low
pH in the solution of the acidic FLAG peptide tag. Re-purification
of 5c-R3 resulted in full incorporation (not shown). The assays for
the extension by 12 nt (Fig. 4E) supported the previous findings
on the incorporation of most nucleotide analogues as well as for
the poor incorporation of 3c (Fig. 4E, lane 3) and the sterically
blocked extension of peptide-tags 5c-R1 and 5c-R2 (Fig. 4E, lanes 7
and 8). Hydrolysis accounted for the poor incorporation of 5c-R3
(Fig. 4E, lane 9).
The likely molecular reason for the poor incorporation of
amino-propargyl nucleotide 3c was investigated with a molecular
model of a polymerase. A model was constructed using the
X-ray structure of the T7 DNA polymerase,40 which—like Taq—
belongs to the type A family and shares a conserved active site. The
X-ray structure of T7 DNA polymerase is a ternary complex with
the double stranded DNA template containing a single stranded
overhang, and an incoming dTTP nucleotide complementary to
base A in the overhang.40 Our model of the active site (Fig. 5)
shows the protein (grey), the DNA template (magenta), and the
nucleotide. The nucleotide 5-(3-amino-propynyl)-2¢-deoxyuridine
5¢-triphosphate 3c was placed in the same position as dTTP in
the X-ray file. Visual inspection of the active site (Fig. 5) revealed
that the primary amino group of the base substituent is close to
Fig. 5 Molecular model of the active site of the T7 DNA polymerase from
the A-type family. The model shows the electrostatic interactions between
the amino group of 5-(3-amino-propynyl)-2¢-deoxyuridine 5¢-triphosphate
3c and the side chains of aspartic acid and lysine. The protein is in grey,
the DNA strand of the template is in magenta, and the two Mg2+ ions are
represented as green spheres. The metal ions electrostatically interact with
the triphosphate moiety of the nucleotide. The nucleotide 3c was placed
into the position of the incoming dTTP nucleotide in the crystal structure
of the ternary protein·DNA·dTTP complex (PDB ID 1T7P).
two charged amino acid moieties. As measured from its nitrogen
˚
atom, the amino-propargyl group is 4.5 A apart from the nitrogen
˚
of lysine at position 522, and 4.7 A away from the closest oxygen
of aspartic acid at position 519 (Fig. 5). Based on the importance
of substrate recognition by key residues in DNA polymerases,41,42
we propose that the electrostatic interaction between the amino
group and the charged residues is the key to the inhibition of the
polymerase activity. In particular, the electrostatic attraction of
amino-propargyl to Asp, and the electrostatic repulsion to Lys,
is most likely altering the position of the pyrimidine base within
the active site (not shown) thereby affecting the ability of the
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 3826–3835 | 3831
©

View Article Online
polymerase to accept the nucleotide as a substrate. Several
observations are consistent with the notion that electrostatic
interactions cause the poor acceptance of amino-propargyl nu-
cleotide 3c. Firstly, abolishing the ionic interaction by neutralising
the amino group of 3c via acylation with hexa-2,4-dienoic acid
(thereby yielding nucleotide 3e, see Scheme 2) resulted in a high
polymerisation extent for Taq (lanes 5 in Fig. 4B and C). In addi-
tion, weakening the interaction by increasing the temperature of
polymerisation from 57 to 70 ◦C improved the acceptance of 3c by
the enzyme, as seen for dNTP mix 6 (Fig. 4F, lane 4) and dNTP mix
12 (Fig. 4F, lane 5). The extent of incorporation of 3c was, however,
not complete. While the ionic interaction between nucleotide and
enzyme is a plausible explanation, we cannot, however, rule out
other reasons. For example, a modified nucleotide added to a DNA
strand could affect the interaction between primer and template
strands and thereby lead to DNA duplex stability problems.12
Further work using point-mutated enzymes would be required
to unequivocally clarify whether the charged residues in the
active site are the key to the poor acceptance of amino-propargyl
derivative 3c.
Preparative ion exchange chromatography was performed on a
Sephadex A-25 DEAE (120 mL) column, a Resource Q (1 mL)
or a Resource S (1 mL) column on an Akta Purifier system.
¨
Analytical HPLC was performed on a Varian Pro Star system
equipped with a dual wavelength 325 detector and a Gemini C18
column (250 ¥ 4.6 mm). Preparative HPLC was performed on
the same system using a Polaris C18 column (100 ¥ 21.2 mm,
5 mm beads). Mass spectrometric analysis was performed on a
Waters Acquity Ultra Performance LC/MS system equipped with
an Acquity UPLC BEH C18 column (50 ¥ 2.1 mm, 1.7 mm beads).
NMR spectra were recorded at 293K on a Bruker AVANCE500
spectrometer (for 1H- and 13C-NMR) and on a Bruker AMX300
spectrometer (for 31P-NMR). Coupling constants are reported in
Hz. Polyacrylamide gels were prepared with 18% acrylamide in
tris-borate buffer containing 7 M urea and 18% formamide, and
stained using ethidium bromide.
Synthesis of nucleosides
5-Ethynyl-2¢-deoxyuridine (2a). 5-iodo-2¢-deoxyuridine, 1,
(370 mg, 1.04 mmol), palladium tetrakis(triphenylphosphine)
(113 mg, 98 mmol), and copper iodide (47 mg, 250 mmol)
were placed in a dried flask flushed with argon. Anhydrous
DMF (12.0 mL) was added via syringe, followed by triethy-
lamine (0.4 mL, 2.9 mmol) and trimethylsilylacetylene (0.71 mL,
5.0 mmol). The mixture was stirred at room temperature for 3.5 h,
then solvents were removed in vacuo. The resulting residue was
submitted to flash column chromatography (10% MeOH in DCM)
affording 381 mg of yellow foam. The foam was dissolved in
THF (4 mL) and TBAF in THF (1.2 mL, 1M, 1.2 mmol). After
stirring the reaction mixture at room temperature for 3 h, the
resulting white precipitate was removed by filtration and rinsed
first with THF and then with 10% MeOH in DCM (2 mL). The
combined filtrates were concentrated and purified by flash column
chromatography (10% MeOH in DCM) to afford 191 mg (73%)
of white powder. dH(300 MHz; DMSO-d6) 2.12 (2H, dd, J = 5.1,
5.1), 3.60 (2H, q, J = 11.7), 3.78 (1H, q, J = 3.3), 4.09 (1H, s),
4.20–4.25 (1H, m), 5.13 (1H, t, J = 4.6), 5.24 (1H, d, J = 4.1),
6.09 (1H, t, J = 6.6), 8.29 (1H, s), 11.62 (1H, s, H-3). dC(300 MHz;
DMSO-d6) 40.1, 60.7, 69.9, 76.4, 84.7, 87.5, 97.5, 144.5, 149.4,
161.6. ESI-MS(pos): 275.3 m/z [M + Na]+, calculated mass for
C11H12N2O5: 252.2.
Conclusions
In this study, we have described the synthesis of deoxyuridine
triphosphate derivates using the Sonogashira cross-coupling.
As the triphosphate analogues were being used for the post-
modification labelling of DNA strands, we explored pyrimidine
derivatives that carried at position 5 linkers of varying length,
flexibility and terminal functional groups. In the course of the
synthesis of the nucleotide derivatives, the Yoshikawa procedure
was optimised. Compared to the long incubation time in the
literature, we found an optimal shorter reaction window between
3 to 7 min that achieved high levels of target and a low level of side-
product. The procedure is also compatible with labile functional
groups attached to the base. As shown by primer extension assays,
almost all nucleotides were well accepted by two widely used DNA
polymerases. Blockade of nucleotide incorporation was observed
for sterically demanding peptide-tagged nucleotides as well as
for the amino-propargyl derivative of deoxyuridine. The poor
extent of incorporation was most likely related to electrostatic
interactions with amino acid side chains, as demonstrated using a
molecular model of the active site of an enzyme. In conclusion, by
covering the chemistry and biology of functional deoxyuridine
triphosphates our study provides important insights into their
5-[3-(Trifluoroacetamido)propynyl]-2¢-deoxyuridine
(3a). 5-
iodo-2¢-deoxyuridine, 1, (740 mg, 2.1 mmol), palladium
tetrakis(triphenylphosphine) (61 mg, 53 mmol), and copper
iodide (29 mg, 152 mmol) were placed in a dried flask
flushed with argon. Anhydrous DMF (7.0 mL) was added
via syringe, followed by triethylamine (0.6 mL, 4.0 mmol) and
N-propynyltrifluoroacetamide (762 mg, 5.0 mmol) in anhydrous
DMF (7 mL). After stirring the mixture at room temperature for
4.5 h, the solvents were removed in vacuo, and the residue was
purified by flash column chromatography (8% MeOH in DCM)
affording 790 mg of yellow foam. dH(300 MHz; DMSO-d6) 2.11
(3H, t, J = 5.4), 3.52–3.62 (2H, m), 3.78 (1H, q, J = 2.7, 3.2),
4.21 (3H, d, J = 5.35), 5.08 (1H, t, J = 4.3), 5.23 (1H, d, J = 3.7),
6.09 (1H, t, J = 6.4), 8.19 (1H, s), 10.05 (1H, t, J = 5.3), 11.62
(1H, s). dC(300 MHz; DMSO-d6) 29.4, 40.0, 60.9, 70.1, 75.3, 84.7,
87.4, 87.6, 144.1, 149.3, 161.5. dF(282 MHz; DMSO-d6) -74.71.
synthesis and biochemical use, thereby complementing existing
43
reports on purine-containing nucleotide triphosphates.3,9–12
,
Experimental
General
˚
Trimethylphosphate was dried under a 3A-molecular sieve prior
to use. All nucleosides and proton sponge were stored in a
desiccator over P2O5 several days before use. Anhydrous tributyl
ammonium pyrophosphate solution was prepared following a
published procedure.44 All other chemicals and solvents were of
-
R
commercial quality and used as received. Deep VentRꢀ (exo )
polymerase was purchased from New England Biolabs (NEB)
and DNA oligonucleotides from Integrated DNA Technologies.
3832 | Org. Biomol. Chem., 2009, 7, 3826–3835
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
5-(Hexa-2,4-dienoic acid prop-2-ynylamide)-2¢-deoxyuridine
(3d). 3a was taken up in a 1:1 solution of conc. ammonia and
water (10 mL). After 2 h, solvents were evaporated in vacuo. The
residue was taken in water and lyophilised. The resulting foam
(3b) was dissolved in anhydrous DMF, and mixed with potassium
sorbate (153 mg, 1.0 mmol) and EDC·HCl (304 mg, 2.0 mmol).
After stirring under argon for 48 h, the reaction mixture was
filtered, and DMF was removed using co-evaporation with water.
The brown residue was taken up in 5% MeOH in DCM and
triturated, which resulted in the precipitation of a yellow solid. The
solid was filtered off and dried in air, then washed with water, and
a small amount of methanol, affording 125 mg of white crystals
(33%). dH(300 MHz; DMSO-d6) 1.79 (3H, d, J = 5.9, H-16), 2.10
(2H, t, J = 5.4, H-2¢), 3.50–3.64 (2H, m, H-5¢), 3.78 (1H, q, J =
3.2, H-4¢), 4.14 (2H, d, J = 5.4, H-9), 4.18–4.24 (1H, m, H-3¢), 5.09
(1H, t, J = 4.8, 5¢-OH), 5.22 (1H, d, J = 4.3, 3¢-OH), 5.87 (1H, d,
J = 14.9, H-12), 6.07–6.25 (3H, m, H-1¢, H-13, H-14), 7.03 (1H,
dd, J = 10.2, 10.2, H-15), 8.16 (1H, s, H-6), 8.47 (1H, t, J = 5.4),
11.62 (1H, s).
Triphosphorylation of nucleotides
Time-course on the monophosphorylation of 5-iodo-2¢-
deoxyuridine. 5-Iodo-2¢-deoxyuridine (78 mg, 0.22 mmol) and
1-8-bis (dimethylamino) naphthalene (proton sponge) (71 mg,
0.33 mmol) were placed in a dried 2-neck flask, which was
subsequently flushed with argon. Trimethyl phosphate (2.0 mL)
was added via syringe, and the mixture was stirred at room
temperature for at least 10 min until the compound was completely
dissolved. A sample (1.5 mL) was withdrawn while maintaining
a positive pressure of argon, and diluted with TEAB (0.1 M,
150 mL). The reaction mixture was then brought to the desired
temperature using an ice bath (0 ◦C experiment) or an ice-salt
bath (-13 ◦C experiment) and POCl3 (20–60 mL, 0.22–0.66 mmol)
was added via syringe. Samples were taken at regular intervals
and TEAB was added within 10 sec as described to hydrolyse the
phosphorodichloridate intermediate. The samples were analysed
by HPLC. Peak areas of starting material and phosphorylated
derivatives were normalised to the peak area of the proton sponge.
dP(121.4 MHz; D2O) -2.26. ESI-MS(neg): 433 m/z [M – H]-,
calculated mass for C9H12IN2O8P: 434.1.
5-(Pent-4-ynoic acid methyl ester)-2¢-deoxyuridine (4a). 5-
iodo-2¢-deoxyuridine, 1, (354 mg, 1.0 mmol), palladium
tetrakis(triphenylphosphine) (110 mg, 95 mmol), and copper
iodide (46 mg, 158 mmol) were placed in a dried flask filled with
argon. Anhydrous DMF (6.0 mL) was added via syringe, followed
by triethylamine (0.4 mL, 2.9 mmol) and a solution of pent-4-ynoic
acid methyl ester (5.7 ml, 5.1 mmol) in anhydrous DMF (5 mL).
After 3 h stirring at room temperature, the solvent was removed
in vacuo. The resulting brown oil was purified by flash column
chromatography (10% MeOH in DCM), yielding 410 mg of orange
oil (125%) (contamination by triethylammonium). dH(300 MHz;
DMSO-d6) 2.09 (2H, t, J = 7.2), 2.57 (4H, m), 3.61 (3H, s), 3.57
(2H, d, J = 4.8), 3.77 (1H, d, J = 3.2), 4.21 (1H, m), 5.09 (1H,
t, J = 5.1), 5.25 (1H, d, J = 4.3), 6.09 (1H, t, J = 6.7), 8.10 (1H,
s). dC(126 MHz, DMSO-d6) 14.7, 32.6, 39.9, 51.5, 60.9, 70.1, 84.6,
87.5, 143.0. ESI-MS(pos): 449.5 m/z [M + Na + Et3N]+, calculated
mass for C15H18N207: 338.3.
General method for the triphosphorylation of nucleosides. The
nucleoside (0.22 mmol) and proton sponge (71 mg, 0.33 mmol)
were placed in a dried 2-neck flask, and dissolved in trimethyl
phosphate (2.0 mL) The reaction mixture was cooled to between
-13 ◦C and -20 ◦C, and POCl3 (30 mL, 0.33 mmol) was added
via syringe. After 20 min stirring, a suspension of tributylam-
monium pyrophosphate in anhydrous DMF (0.5 M, 2 mL) and
tributylamine (0.16 mL) was added. After 45 min, the mixture
was combined with TEAB (0.1M, 20 mL), and warmed up to
room temperature. Volatile solvents were removed in vacuo, and
the resultant residue purified by ion exchange chromatography
(Sephadex A-25 DEAE) followed by HPLC.
5-Iodo-2¢-deoxyuridine 5¢-triphosphate triethylammonium salt
(1a). Made from commercially available 5-iodo-2¢-deoxyuridine.
dH(400 MHz; D2O) 1.22 (38H, t, J = 7.4), 2.34–2.39 (2H, m), 3.09
(24H, q, J = 7.4), 4.15–4.27 (3H, m), 4.62–4.67 (1H, m), 6.26 (1H,
t, J = 6.6), 8.26 (1H, s) dC(126 MHz; D2O) 39.3, 65.9, 71.0, 85.8,
145.9. dP(121.4 MHz; D2O) -21.30 (1P, t, J = 21.4), -10.35 (1P,
d, J = 21.4), -5.26 (1P, d, J = 21.4). ESI-MS(neg): 593 m/z [M –
H]-, calculated mass for C9H14IN2O14P3: 594.0.
5-(Pent-4-ynoic acid hexa-2,4-dienyl ester)-2¢-deoxyuridine (5a).
5-iodo-2¢-deoxyuridine, 1, (395 mg, 1.05 mmol), palladium
tetrakis(triphenylphosphine) (55 mg, 48 mmol), and copper iodide
(28 mg, 121 mmol) were placed in a dried flask flushed with argon.
Anhydrous DMF (5.0 mL) was added via syringe, followed by
triethylamine (0.4 mL, 2.9 mmol) and a solution of pent-4-ynoic
acid hexa-2,4-dienyl ester (1.0 g, 6.02 mmol) in anhydrous DMF
(5 mL). After 22 h stirring at room temperature, the solvent was
removed in vacuo. The resulting brown oil was purified by flash
column chromatography using a gradient from 3–5% MeOH in
DCM, yielding 158 mg of white powder (39%). dH(500 MHz,
DMSO-d6) 1.71 (3H, d, J = 6.6,), 2.09 (2H, dd, J = 6.6, 4.7,),
2.53–2.63 (4H, m), 3.54 (1H, ddd, J = 11.8, 5.0, 3.8,), 3.59 (1H,
ddd, J = 11.8, 5.0, 3.6,), 3.78 (1H dd, J = 6.6, 3.5,), 4.19–4.23
(1H, m), 4.56 (1H, d, J = 6.3,), 5.07 (1H, t, J = 4.9,), 5.22 (1H,
d, J = 4.1), 5.62 (1H, dt, J = 15.1, 6.6,), 5.72 (1H, dq, J = 15.1,
6.6,), 6.05 (1H, ddd, J = 15.1, 10.4, 1.6), 6.09 (1H, t, J = 6.6), 6.25
(1H, dd, J = 10.4, 15.1), 8.10 (1H, s), 11.55 (1H, s). dC(126 MHz,
DMSO-d6) 171.1 161.6, 149.4, 143.0, 134.0, 130.6, 130.5, 124.2,
98.6, 91.5, 87.5, 84.6, 73.1, 70.1, 64.3, 60.9, 39.7, 32.8, 17.8, 14.8.
HRMS 403.15, calculated mass for C20H23N2O7: 403.15.
5-Ethynyl-2¢-deoxyuridine 5¢-triphosphate triethylammonium
salt (2b). Made from 2a (55 mg, 0.22 mmol). dH(400 MHz; D2O)
1.28 (27 H, t, J = 7.4), 2.33–2.46 (2H, m), 3.20 (18H, q, J = 7.4),
3.60 (1H, s),4.19–4.23 (3H, m), 4.61–4.65 (1H, m), 6.28 (1H, t,
J = 6.7), 8.22 (1H, s). ESI-MS(neg): 491 m/z [M – H]-, calculated
mass for C11H15N2O14P3: 492.2.
5-(3-Amino-propynyl)-2¢-deoxyuridine 5¢-triphosphate triethy-
lammonium salt (3c). Made from 3a. The synthesis was per-
formed as described above. After neutralisation by TEAB, 20 mL
of concentrated ammonia was added to the reaction mixture to
deprotect the amine. After 18 h stirring at room temperature, the
product was purified in order to remove the acetamide protecting
group. dH(500 MHz; D2O) 1.26 (18H, t, J = 7.4), 2.29–2.43 (2H,
m), 3.18 (12H, q, J = 7.4), 3.14 (2H, s), 4.14–4.26 (3H, m), 4.58–
4.64 (1H, m), 6.28 (1H, t, J = 6.2), 8.31 (s, 1H). dP(121.4 MHz;
D2O) -21.38 (1P, t, J = 21.4), -10.40 (1P, d, J = 21.4), -0.05
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 3826–3835 | 3833
©

View Article Online
(1P, d, J = 21.4). ESI-MS(neg): 520 m/z [M – H]-, calculated
Diels–Alder reaction of nucleotide triphosphate (5c). A solution
of 100 mM of the maleimide-terminated peptide was prepared
by dissolving lyophilised peptide in 500 mM Na2HPO4 buffer
(final pH 6.1).7 This solution (15 mL, 1.5 mmol) was added
to the nucleotide solution (30 mM, 25 mL), and the reaction
mixture was incubated at 25 ◦C for 2 h. LC-MS and ion exchange
chromatography analysis indicated full conversion of the starting
material for all reactions. The product was purified by ion exchange
chromatography using a Resource Q column for the negatively
charged products (His6 tag or FLAG nucleotides) or a Resource
S column for the positively charged product (NLS). The fractions
containing products were lyophilised.
mass for C12H18N3O14P3: 521.2.
5-(Hexa-2,4-dienoic acid prop-2-ynylamide)-2¢-deoxyuridine 5¢-
triphosphate triethylammonium salt (3e). Made from 3d.
dH(500 MHz; D2O) 1.27 (27H, t, J = 7.4), 1.82 (3H, d, J = 6.0),
2.36 (1H, dd, J = 14.0, 6.3,), 2.41 (1H, ddd, J = 14.0, 6.3, 4.4),
3.18 (18H, q, J = 7.4,), 4.17–4.26 (5H, m), 4.61–4.66 (1H, m), 5.99
(1H, d, J = 15.1), 6.19–6.33 (3H, m), 7.14 (1H, dd, J = 15.1, 10.1),
8.17 (1H, s). ESI-MS(pos): 516.4 m/z [M + H]+, calculated mass
for C18H24N3O15P3: 615.3.
5-(Pent-4-ynoic acid)-2¢-deoxyuridine 5¢-triphosphate triethy-
lammonium salt (4b). Made from 4a. The synthesis was done
as described above. After neutralisation by TEAB, 20 mL of
concentrated ammonia was added to the reaction mixture to
deprotect the carboxylic acid. After 24 h stirring at 4 ◦C, the
product was purified in order to remove the protecting group.
dH(300 MHz; D2O) 2.35 (2H, t, J = 6.9), 2.53 (2H, t, J = 6.9),
2.68 (2H, t, J = 6.9,), 4.16 (2H, d, J = 4.2), 4.60 (1H, m), 6.25
(1H, t, J = 6.6), 8.03 (1H, s). dC(126 MHz; D2O) 15.8, 34.2, 39.0,
70.9, 85.9, 144.3. dP(121.4 MHz; D2O) -21.6 (1P, m), -10.41 (1P,
d, J = 18.8), -1.75 (1P, d, J = 18.8). ESI-MS(neg): 767.4 m/z [M +
2Et3N + H]+, calculated mass for C14H19N2O16P3: 564.2.
Purification gradient for His6 nucleotide (R1). Resource Q
column at a flow rate of 2.5 mL/min using 0.1 M TEAB pH 7.5
for 5 mL followed by a gradient of 0.1 M to 0.45 M TEAB
over 10 mL. HRMS: 1693.48, calculated mass for C65H80N22O27P3:
1693.47.
Purification gradient for NLS nucleotide (R2). Resource S
column at a flow rate of 2.5 mL/min using 5 mM TEAB pH 7.5
for 5 mL followed by a gradient of 5 mM to 0.4 M TEAB over
10 mL. ESI-MS(neg): 1734.1 m/z [M - H]-, calculated mass for
C69H114N18O28P3: 1735.7.
Purification gradient for FLAG nucleotide (R3). Resource Q
column at a flow rate of 2.5 mL/min using 0.45 M TEAB pH 7.5
for 5 mL followed by a gradient of 0.45 M to 0.8 M TEAB over
10 mL, followed by a step of 10 mL at 0.8 M.
5-(Pent-4-ynoic acid hexa-2,4-dienyl ester)-2¢-deoxyuridine 5¢-
triphosphate triethylammonium salt (5b). Made from 5a.
dH(500 MHz; D2O) 1.19 (27H, t, J = 7.3), 1.61 (3H, d, J = 6.6,),
2.26 (2H, dd, J = 14.0, 6.6,), 2.31 (1H, ddd, J = 14.0, 6.3, 3.9),
2.58–2.62 (2H, m); 2.63–2.67 (2H, m), 3.10 (18H, q, J = 7.3,),
4.07–4.15 (3H, m), 4.51–4.55 (1H, m), 4.58 (2H, d, J = 6.3,), 4.58
(1H, dt, J = 15.1, 6.6), 5.68 (1H, dq, J = 15.1, 6.6), 5.95 (1H, ddd,
J = 15.1, 10.4, 1.6), 6.17 (1H, t, J = 6.6,), 6.20 (1H, dd, J = 15.1,
10.4), 7.92 (1H, s). dC(126 MHz; D2O) 175.0, 164.8, 150.8, 144.2,
135.3, 132.5, 130.5, 123.7, 100.3, 94.1, 85.9, 85.8, 72.2, 70.9, 66.0,
46.9, 38.8, 33.4, 17.7, 15.5, 8.5. dP(121.4 MHz; D2O) -21.69 (1P, t,
J = 21.1), -10.17 (1P, d, J = 21.1), -8.75 (1P, d, J = 21.1). HRMS:
643.05, calculated mass for C20H26N2O16P3: 643.05.
Primer extension. Samples containing 150 pmol of DNA
oligonucleotide primer of sequence 5¢-ATG GGA CTA ACT
AAT CTT TGC TTA-3¢ and 170 pmol of template 5¢-TAT TCC
ATG TGA TAA GCA AAG ATT AGT TAG TCC CAT-3¢ were
dissolved in a polymerase extension mix. The mix contained 0.5
-
TM
R
U Deep VentRꢀ exo (NEB), 1X Thermopol buffer (NEB),
and 40 mM of natural dNTP nucleotides or dUTP derivatives in
a total volume of 25 mL. The reaction mixtures were heated to
85 ◦◦C for 1 min to denature DNA duplexes, and incubated at
57 C for 10 min to anneal and extend the primers, followed by
immediate chilling on ice. The extension products were visualised
by denaturing PAGE (18%) using 4–5 mL aliquots of the mixtures
denatured by addition of formamide and subsequent heating at
95 ◦C.
Peptide synthesis. The maleimide–peptide conjugates were
synthesised using standard Fmoc solid-phase peptide synthesis
(SPPS) on a Syro automated system using chloro-trityl resin
pre-loaded with His, or Wang resin pre-loaded with Gly, and
HBTU/DIPEA coupling chemistry. Four equivalents of coupling
agent and amino acid were used in the coupling step of 40 min
duration. Fmoc deprotection was carried out by shaking the resin
in 40% piperidine for 3 min, followed by a second step in 20%
piperidine for 10 min. Maleimide-b-alanine was added as the last
amino acid using the same coupling chemistry. After completion
of the synthesis, peptides were deprotected and cleaved from the
resin by incubation in TFA/TES/H2O (95:2.5:2.5 v/v/v) for 3 h.
The peptide solution was collected and precipitated in diethyl
ether at -20 ◦C and centrifuged to a pellet. The supernatant was
removed and fresh diethyl ether was added. The pellet was broken
up by sonication, the fragments supended in ether, and the slurry
centrifuged to form a pellet. This purification process was repeated
three times. Afterwards, the precipitate was dissolved in deionised
water, frozen in a dry ice/acetone bath and freeze-dried overnight.
The purity and chemical identity of the peptides was confirmed
by HPLC and mass spectrometry, respectively.
Acknowledgements
This work has been supported by UCL Chemistry, UCL Business
PLC, and the EPSRC (EP/D030005). We thank A. B. Tabor for
use of a peptide synthesizer, and Hugh Martin for helping prepare
illustrations.
References
1 G. M. Blackburn, M. J. Gait, D. Loakes and D. Williams, Nucleic acids
in chemistry and biology, Royal Society of Chemistry, London, 2006.
2 V. V. Rostovtsev, L. G. Green, V. V. Fokin and K. B. Sharpless, Angew.
Chem., Int. Ed., 2002, 41, 2596–2599.
3 G. A. Burley, J. Gierlich, M. R. Mofid, H. Nir, S. Tal, Y. Eichen and T.
Carell, J. Am. Chem. Soc., 2006, 128, 1398–1399.
4 J. Gierlich, G. A. Burley, P. M. E. Gramlich, D. M. Hammond and T.
Carell, Org. Lett., 2006, 8, 3639–3642.
5 O. Diels and K. Alder, Liebigs Annalen der Chemie, 1928, 460, 98–122.
3834 | Org. Biomol. Chem., 2009, 7, 3826–3835
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
6 D. C. Rideout and R. Breslow, J. Am. Chem. Soc., 2000, 102, 7816–
7817.
7 V. Borsenberger and S. Howorka, Nucleic Acids Res., 2009, 37, 1477–
1485.
25 L. H. Thoresen, G. S. Jiao, W. C. Haaland, M. L. Metzker and K.
Burgess, Chem,–Eur. J., 2003, 9, 4603–4610.
26 P. Capek, R. Pohl and M. Hocek, Org. Biomol. Chem., 2006, 4, 2278–
2284.
8 H. A. Held and S. A. Benner, Nucleic Acids Res., 2002, 30, 3857–3869.
9 G. Giller, T. Tasara, B. Angerer, K. Muhlegger, M. Amacker and H.
Winter, Nucleic Acids Res., 2003, 31, 2630–2635.
10 T. Tasara, B. Angerer, M. Damond, H. Winter, S. Dorhofer, U.
Hubscher and M. Amacker, Nucleic Acids Res., 2003, 31, 2636–2646.
11 Z. Fo¨ldes-Papp, B. Angerer, W. Ankenbauer and R. Rigler, J. Biotech-
nol., 2001, 86, 237–253.
27 M. Vrabel, M. Hocek, L. Havran, M. Fojta, I. Votruba, B. Klepetarova,
R. Pohl, L. Rulisek, L. Zendlova, P. Hobza, I. H. Shih, E. Mabery and
R. Mackman, Eur. J. Inorg. Chem., 2007, 1752–1769.
28 H. Weizman and Y. Tor, J. Am. Chem. Soc., 2002, 124, 1568–1569.
29 V. Borsenberger, N. Mitchell and S. Howorka, J. Am. Chem. Soc., 2009,
131, 7530–7531.
30 M. Yoshikawa, T. Kato and T. Takenishi, Tetrahedron Lett., 1967, 5065–
12 O. Thum, S. Ja¨ger and M. Famulok, Angew. Chem., Int. Ed., 2001, 40,
3990–3993.
5068.
31 M. Yoshikawa, T. Kato and T. Takenishi, Bull. Chem. Soc. Jpn., 1969,
42, 3505–3508.
32 K. Burgess and D. Cook, Chem. Rev., 2000, 100, 2047–2059.
33 J. L. Ruth and Y. C. Cheng, Mol. Pharmacol., 1981, 20, 415–422.
34 J. Ludwig, in Biophosphates and their analogues- Synthesis, structure,
metabolism and activity, eds. K. S. Bruzik and W. J. Stec, Elesevier,
Amsterdam, first edn., 1987, pp. 131–133.
35 T. Sowa and S. Ouchi, Bull. Chem. Soc. Jpn., 1975, 48, 2084–2090.
36 W. H. Dawson, R. L. Cargill and R. B. Dunlap, J. Carb. Nucleos. Nucl.,
1977, 4, 363–375.
13 T. Kottysch, C. Ahlborn, F. Brotzel and C. Richert, Chem.–Eur. J.,
2004, 10, 4017–4028.
14 F. W. Hobbs, J. Org. Chem., 1989, 54, 3420–2422.
15 M. Hocek and M. Fojta, Org. Biomol. Chem., 2008, 6, 2233–
2241.
16 P. Baumhof, N. Griesang, M. Bachle and C. Richert, J. Org. Chem.,
2006, 71, 1060–1067.
17 L.-A. Fendt, I. Bouamaied, S. Thoeni, N. Amiot and E. Stulz, J. Am.
Chem. Soc., 2007, 129, 15319–15329.
18 T. Nguyen, A. Brewer and E. Stulz, Angew. Chem., Int. Ed., 2009, 48,
1974–1977.
37 A. Collier and G. Wagner, Org. Biomol. Chem., 2006, 4, 4526–4532.
38 J. M. Campagne, J. Coste and P. Jouin, J. Org. Chem., 1995, 60, 5214–
5223.
19 S. I. Khan, A. E. Beilstein and M. W. Grinstaff, Inorg. Chem., 1999, 38,
418–419.
39 I. Diala, S. Murao and M. Fujii, Nucleic Acids Symposium Series, 2008,
20 N. Amann, E. Pandurski, T. Fiebig and H. A. Wagenknecht, Angew.
Chem., Int. Ed., 2002, 41, 2978–2980.
21 E. Mayer-Enthart and H. A. Wagenknecht, Angew. Chem., Int. Ed.,
2006, 45, 3372–3375.
22 S. Ja¨ger and M. Famulok, Angew. Chem., Int. Ed., 2004, 43, 3337–
3340.
52, 679–680.
40 S. Doublie, S. Tabor, A. M. Long, C. C. Richardson and T. Ellenberger,
Nature, 1998, 391, 251–258.
41 M. C. Franklin, J. M. Wang and T. A. Steitz, Cell, 2001, 105, 657–667.
42 U. Hubscher, G. Maga and S. Spadari, Annu. Rev. Biochem., 2002, 71,
133–163.
23 V. R. Sirivolu, P. Chittepu and F. Seela, ChemBioChem, 2008, 9, 2305–
43 K. Kincaid, J. Beckman, A. Zivkovic, R. L. Halcomb, J. W. Engels and
R. D. Kuchta, Nucleic Acids Res., 2005, 33, 2620–2628.
44 J. Ludwig and F. Eckstein, J. Org. Chem., 1988, 54, 631–635.
2316.
24 J. He and F. Seela, Nucleic Acids Res., 2002, 30, 5485–5496.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 3826–3835 | 3835
©