Fluoride-cleavable, fluorescently labelled reversible terminators: Synthesis and use in primer extension

Article abstract of DOI:10.1002/chem.201001952

Fluorescent 2′-deoxynucleotides containing a protecting group at the 3′-O-position are reversible terminators that enable array-based DNA sequencing-by-synthesis (SBS) approaches. Herein, we describe the synthesis and full characterisation of four reversi

Full text of DOI:10.1002/chem.201001952

FULL PAPER
DOI: 10.1002/chem.201001952
Fluoride-Cleavable, Fluorescently
ACHTUNGTRNELNUNG abelled Reversible Terminators:
Synthesis and Use in Primer Extension
Diana C. Knapp,[a] Saulius Serva,[b] Jennifer D’Onofrio,[a] Angelika Keller,[a]
Arvydas Lubys,[b] Ants Kurg,[c, d] Maido Remm,[d] and Joachim W. Engels*[a]
Abstract: Fluorescent 2’-deoxynucleoti-
des containing a protecting group at
the 3’-O-position are reversible termi-
nators that enable array-based DNA
sequencing-by-synthesis (SBS) ap-
proaches. Herein, we describe the syn-
thesis and full characterisation of four
nucleotide analogue has a different flu-
orophore attached to the base through
a fluoride-cleavable linker and a 2-cya-
noethyl moiety as the 3’-blocking
group, which can be removed by using
a fluoride treatment as well. Further-
more, we identified a DNA poly-
merase, namely, RevertAid M-MuLV
reverse transcriptase, which can incor-
porate the four modified reversible ter-
minators. The synthesised nucleotides
and the optimised DNA polymerase
were used on CodeLink slides spotted
with hairpin oligonucleotides to dem-
onstrate their potential in a cyclic re-
versible terminating approach.
reversible terminators bearing
a 3’-
Keywords: fluorescence · fluoride
ions · nucleotides · reverse tran-
scriptase · reversible terminators
blocking moiety and linker-dye
a
system that is removable under the
same fluoride-based treatment. Each
Introduction
terventions, and drug response. For this reason, there has
been a rapid development in genotyping technology. For ex-
ample, many systems for detecting mutations or single nu-
cleotide polymorphisms (SNPs) on a large-scale are current-
ly commercially available.
The completion of the two Human Genome Projects in 2001
implemented the ultimate goal of sequencing the whole
human genome.[1,2] This milestone not only provided a refer-
ence genome, but also unexpected opportunities, such as
new questions, goals, and hopes, increased demands for im-
provements in the cost efficiency and throughput of DNA
sequencing to an astonishing extent. These requirements not
only address sequencing of whole novel genomes, but also
the search for individual variation within the human
genome. In fact, the latter is a very important issue of the
postgenome era, as it promises to elucidate how genetic var-
iation interacts with the environment to confer individual
resistance or susceptibility to disease, success of medical in-
One genotyping technology, arrayed primer extension
(APEX), is a minisequencing microarray assay[3] capable of
detecting different types of genetic variations, while combin-
ing the efficiency of microarrays (an alternative to gel-based
methods) and Sanger sequencing technology.[4,5] In general,
this method can be viewed as DNA sequencing by termina-
tion with the use of labelled dideoxynucleotides in a DNA
polymerase reaction. However, there is one important dif-
ference between APEX and conventional Sanger sequenc-
ing. Instead of using one primer and analysing hundreds of
extension products with polyacrylamide gel electrophoresis
(PAGE), hundreds to thousands of primers are spatially sep-
arated beforehand as a two-dimensional array of oligonucle-
otides. These primers are immobilised by attaching the 5’-
end to a glass surface, and each oligonucleotide is extended
at the 3’-end by only one dye-labelled dideoxynucleotide
complementary to the nucleotide at the variable site. As a
result, each primer identifies one base in the target se-
quence. The advantages of APEX include parallel analysis
of hundreds to thousands of genetic variations in a single re-
action, high allelic discrimination by the use of a DNA poly-
merase and four labelled terminators, and the possibility of
locus- or disease-specific array design. Therefore, all four
possible sequence variants can be detected simultaneously
in one reaction. One disadvantage is that one primer is nec-
essary for each position to be identified. This approach can
lead to a very high number of oligonucleotides on the array,
which can cause problems in fluorescence detection. De-
[a] D. C. Knapp, Dr. J. D’Onofrio, Dr. A. Keller, Prof. Dr. J. W. Engels
Institut fꢀr Organische Chemie und Chemische Biologie
J.W. Goethe Universitꢁt Frankfurt am Main
Max-von-Laue Strasse 7, 60438 Frankfurt am Main (Germany)
Fax : (+49)6979829148
E-mail: joachim.engels@chemie.uni-frankfurt.de
[b] Dr. S. Serva, Dr. A. Lubys
Thermo Fisher Scientific (formerly Fermentas)
V. Graiciuno 8, 02241 Vilnius (Lithuania)
[c] Prof. Dr. A. Kurg
Asper Biotech Ltd, Vaksali 17A, 50410 Tartu (Estonia)
[d] Prof. Dr. A. Kurg, Prof. Dr. M. Remm
Institute of Molecular and Cell Biology
University of Tartu, Riia 23, 51010 Tartu (Estonia)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem201001952.
Chem. Eur. J. 2011, 17, 2903 – 2915
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2903

pending on the region of DNA that has to be sequenced, it
could even exceed the scope of the array.
firmed by a crystal-structure study of a rat DNA/primer/nu-
cleotide complex.[9] Thus, the linker has to be cleavable
under conditions that match the cleavage of the 3’-blocking
group to allow both the regeneration of the 3’-OH group
and the removal of the linker-dye system in a single depro-
tection step. Thirdly, the cleavage of the reversibly terminat-
ing group and the linker should be quantitative without af-
fecting the DNA-template stability. Fourthly, a polymerase
is needed that accepts the 3’-modification and nucleotide
modifications and still discriminates strictly between the
four bases during the incorporation reaction. Whereas
Sanger sequencing has proven that modifications at the 7-
position of 7-deazapurines and the 5-position of pyrimidines
are well tolerated, the choice of a suitable 3’-modification
seems more difficult because the editing properties of poly-
merases must also be considered.[10] During the last several
years, academic and industrial research groups have focused
on the design of such reversible terminators. Within the
scope of this study, several 3’-blocking groups were investi-
gated, including bulky esters[7] and ethers[8] with the label at-
tached to the blocking group and small groups. Some exam-
ples are the 3’-O-(2-nitrobenzyl) group investigated by
Metzker and co-workers[6] and Welch and Burgess,[8,11] the
3’-O-allyl group reported by Metzker,[6] Ju,[12] and Kim,[13] or
the 3’-O-azidomethyl group, which was used by Ju and co-
workers[14,15] and was also realised in a commercially avail-
able device, the Genome Analyzer developed by Illumina/
Solexa.[16,17] Other interesting groups are the 3’-O-NH2
group from Benner and co-workers,[18] the 3’-O-(2-cyanoe-
thoxy)methyl group reported by us,[19] or some 3’-blocking
groups removable under mild reducing or mild acidic condi-
tions reported by Kwiatkowski.[20] The terminators with
bulky 3’-modifications exhibited problems with polymerase
acceptance. Ester and carbonate linkages are easily cleaved
by polymerases, thus leading to multiple incorporation
events. Some cases in which the reporter is attached to the
base by a cleavable linker, the cleavage conditions of the
linker differ from those of the 3’-blocking group. Two differ-
ent kinds of chemical treatments make these strategies more
time consuming, thus showing that this research is very chal-
lenging and encourages further investigation of new candi-
dates for reversibly terminating groups.
We have designed and synthesised a new generation of
fluorescently labelled, reversibly terminating nucleotides,
identified a DNA polymerase that accepts these nucleotides,
and optimised the reaction conditions under which they are
incorporated into the DNA fragments. The use of these re-
versible terminators has the potential to fulfil the needs of
repeated primer extension reactions on APEX DNA arrays.
In this approach, one cycle consists of three steps: 1) DNA
polymerase-mediated incorporation of the complementary
reversible terminator onto the immobilised oligonucleotide
primer sequence, 2) detection of the fluorescence signals
specific for each of the four bases, and 3) cleavage of the ter-
minating moiety and the reporter molecule to restore the
free 3’-OH group and remove fluorescence signals of al-
ready incorporated nucleotides. Repetition of this cycle
leads to the template sequence. This approach, which uses
our reversible terminators together with the DNA poly-
merase, provides an opportunity to revolutionise the future
of APEX technology because the number of oligonucleotide
features on the array could be decreased according to the
number of consecutive primer extension cycles on the chip
surface.
This idea of using 3’-reversibly blocked nucleotides for se-
quencing was proposed in the beginning of the 1990s. The
first examples of potential reversible terminators were re-
ported in 1994 by Metzker and co-workers[6] and Canard
and Sarfati.[7] However, the demands these molecules have
to meet are challenging, and therefore implementation is
difficult. The general structure of a reversible terminator 1
is shown in Scheme 1.
Scheme 1. General structure 1 of a fluorescently labelled reversible ter-
minator.
Herein, we present the synthesis of a complete set of four
reversible terminators that bear the fluoride-cleavable 3’-O-
(2-cyanoethyl) group as a 3’-OH blocking moiety[21] and a
suitable fluoride-cleavable linker[22] to connect the nucleo-
side to a fluorescent dye. The polymerase incorporation ex-
periments and first applications of the reversible terminators
in an APEX system are also presented.
The structural requirements for these nucleotides include
a reversibly terminating moiety at the 3’-position and a re-
porter molecule, such as a dye, attached to the base by a
cleavable linker. In the design of a suitable reversible termi-
nator, several important issues have to be considered: First-
ly, the 3’-blocking group has to be stable during the poly-
merase-mediated extension step to ensure effective abortion
of elongation after incorporating a single nucleotide. Sec-
ondly, a cleavable linker has to be designed to attach the re-
porter moiety to the base. It is disadvantageous to combine
the reporter and blocking group at the 3’-position. Welch
and Burgess reported the lack of acceptance of bulky 3’-
modifications by DNA polymerases,[8] which was also con-
Results and Discussion
Synthesis of the four dye-labelled reversible terminators:
Our target molecules are modified at the 3’-position with
the fluoride-cleavable 2-cyanoethyl group[21] and at the base
with our recently reported fluoride-cleavable linker
2904
www.chemeurj.org
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 2903 – 2915

Nucleotide–Fluorophores in Primer Extension
FULL PAPER
Scheme 2. Structures of the fluorescently labelled reversible terminators 2–5.
(Scheme 2).[22] The four fluorescent dyes chosen for labelling
the four reversible terminators are 5- and 6-carboxyfluores-
cein for thymidine, cyanine 3.0 (Cy 3.0) for 2’-deoxycytidine,
5- and 6-carboxy-X-rhodamine for 2’-deoxyadenosine, and
cyanine 5.0 (Cy 5.0) for 2’-deoxyguanosine. These dyes were
chosen as they are spectrally well separated from each
other, well known, and commercially available. In addition,
the same fluorescent dyes have been already used as di-
deoxy terminator conjugates for a couple of years in classi-
cal APEX reactions on a Genorama platform for SNP geno-
typing,[23] mutation detection,[24] and APEX-based rese-
quencing.[25]
To synthesise the four reversible terminators, there are
two crucial steps: 1) introduction of the 3’-OH modification
and 2) introduction of the linker-dye system. The linker was
incorporated by the well-known Sonogashira cross-coupling
reaction.[26,27] As a prerequisite for this method of attach-
ment, the four iodonucleosides (i.e., 5-iodo-2’-deoxyuridine
(6), 5-iodo-2’-deoxycytidine (7), 7-deaza-7-iodo-2’-deoxyade-
nosine (8), and 7-daza-7-iodo-2’-deoxyguanosine (9; shown
in Scheme 3) are required starting compounds. Compounds
6 and 9 were purchased from commercial sources, whereas 7
and 8 were synthesised in our laboratory by combining steps
from reported procedures that are not described in detail
herein.[28–32]
For selective introduction of the 2-cyanoethyl group at
the 3’-position, it is necessary to protect all the other func-
tional groups. Saneyoshi et al. described the protection of
the 2’-OH group with the 2-cyanoethyl group of all four
RNA nucleosides with acrylonitrile in a Michael addition re-
action with Cs2CO3 as a heterogeneous base in tBuOH.[33]
For DNA, the modification of the 3’-OH group with the 2-
cyanoethyl group has been reported by us for thymidine and
6.[21,22] For the other three nucleosides 7–9, we developed
Scheme 3. The four starting compounds 6–9.
protecting-group strategies to enable the selective introduc-
tion of the 3’-(2-cyanoethyl) group (Schemes 4 and 5).
We used base-labile protecting groups for the amino func-
tions and 5’-OH group. The synthesis of the 3’-O-(2-cya-
noethyl)-modified 2’-deoxyuridine analogue 10 was accom-
plished, as recently reported by us.[22] The same strategy was
used for the protection of the functional groups of 7 and 8.
The exocyclic amino groups of both nucleosides were pro-
tected with formamidine. No acidic proton should be left to
prevent the reaction of its conjugated base in the Michael
addition. Both reactions were carried out following a proce-
dure described for 5-iodo-2’-deoxycytidine.[34] The formami-
dino-protected 2’-deoxycytidine analogue 11 was obtained in
a moderate yield of 62%. Whereas the previously reported
yield[35] for the 2’-deoxyadenosine analogue 14 could be en-
hanced from 85 to 95%. By carrying out the 5’-selective
Chem. Eur. J. 2011, 17, 2903 – 2915
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2905

J. W. Engels et al.
and 16, could be obtained by using very similar protecting-
group strategies, which effectively enabled the selective
modification of the 3’-OH group. Another protecting group
strategy had to be applied for the fourth nucleoside 9 be-
cause guanosine analogues have an additional functional
group that has to be protected. The developed synthetic pro-
cedure was optimised first by using natural 2’-deoxyguano-
sine (not shown) and then transferred to the 7-deaza-7-iodo
derivative 9. The synthesis of the 3’-O-(2-cyanoethyl) modi-
fied derivative 21 is shown in Scheme 5.
Scheme 5. Protecting-group strategy and introduction of the 3’-modifica-
tion into 9. Reagents and conditions: a) 1,1,3,3-tetraisopropyldichlorodisi-
loxane, dry pyridine, 08C!rt, 1 h; b) N,N-dimethylformamide dimethyl-
acetal, dry DMF, rt, 24 h; c) BzCl, dry pyridine, dry CH2Cl2, 08C!rt,
2 h; d) Et3N·3HF, THF, rt, 1 h; e) MMTrCl, DMAP, dry pyridine, rt,
18 h; f) acrylonitrile, Cs2CO3, tBuOH, rt, 2 h; g) PTSA, CH2Cl2/EtOH
1:1, rt, 1 h; h) 32% aqueous NH3, MeOH, rt, 18 h. DMAP=N,N-dime-
thylaminopyridine, MMTr=monomethoxytrityl, PTSA=para-toluenesul-
fonic acid.
Scheme 4. Protecting-group strategy and introduction of the 3’-modifica-
tion into 6,[22] 7, and 8. Reagents and conditions: a) N,N-dimethylforma-
mide dimethylacetal, dry DMF, 558C, 2.5 h; b) BzCl, dry pyridine/dry
DMF=4:1, 08C!rt, 2 h; c) acrylonitrile, Cs2CO3, tBuOH/dry DMF 2:1,
rt, 3 h; d) saturated methanolic ammonia, rt, 2.5 h; e) N,N-dimethylfor-
mamide dimethyl acetal, dry DMF, 508C, 2 h; f) BzCl, dry CH2Cl2, dry
pyridine, ꢀ158C, 1 h; g) acrylonitrile, Cs2CO3, tBuOH, rt, 2 h; h) satu-
rated methanolic ammonia, 508C, 2 h. Bz=benzoyl, BzCl=benzoyl chlo-
ride.
For the protection of the N1 atom, we chose benzoyl as a
blocking group during the synthesis of the 2’-deoxyguano-
sine analogue 21. For the intermediate protection of the 3’-
and 5’-OH groups, we used the Markiewicz procedure.[36]
For the protection of the exocyclic amino group, the forma-
midino group was used. These two steps were carried out
without an intermediate purification step, thus furnishing 17
in 95%. The subsequent benzoylation of the N1 atom was
carried out between 08C and room temperature and pro-
ceeded in 69% yield. When followed by deprotection of the
5’- and 3’-OH groups with Et3N·3HF in THF, the amino-
protected compound 18 could be isolated in 97% yield.
During the synthesis of the 2’-deoxyguanosine analogue 21,
we deviated from the exclusively base-labile protecting-
group strategy. To improve the solubility properties for the
Michael addition reaction, we used the MMTr group to pro-
tect the 5’-OH group of 18 to obtain fully-protected 19 in
71% yield. The 3’-modification was accomplished by follow-
ing the same procedure employed for the 2’-deoxyuridine
benzoylation of 11 at 08C at room temperature, 12 was pro-
duced in a moderate yield of 60%. The 2’-deoxyadenosine
analogue 15 could be obtained in excellent yield by carrying
out the reaction at a low temperature (ꢀ158C). The Michael
addition was accomplished by following procedures reported
by Saneyoshi et al.[33] and by our group.[21,22] Freshly distilled
acrylonitrile was used in tBuOH as the solvent and Cs2CO3
as the base. One modification of the procedure was necessa-
ry for 12; in this case, DMF was used as a cosolvent to en-
hance the solubility of the starting material. After purifica-
tion, the fully protected intermediate was isolated in 88%
yield. The fully protected 3’-modified 7-deaza-7-iodo-2’-de-
oxyadenosine analogue was obtained in a quantitative yield.
The deprotection reactions that lead to compounds 13 and
16 were carried out in saturated methanolic ammonia to
result in yields of 86 and 96% of 13 and 16, respectively.
The first three 3’-modified key compounds, namely, 10, 13
2906
www.chemeurj.org
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 2903 – 2915

Nucleotide–Fluorophores in Primer Extension
FULL PAPER
analogue and the 2’-deoxyadenosine analogue 15, thus pro-
viding fully protected 20 in an excellent yield of 88%. After
deprotection, first with PTSA for removal of the MMTr
group followed by aqueous ammonia for cleavage of the
amino protecting groups, the fourth key compound 21 could
be obtained in 62% yield over the last two steps.
Recently, we described the synthesis of the 5- and 3’-
modified 2’-deoxyuridine analogue 24.[22] We coupled 5-io-
donucleoside 10 to 1,1,1-trifluoroacetyl-protected propargyl-
amine and, after subsequent deprotection, attached the
linker to the free amino group through the N-hydroxysucci-
nimidyl carbonate group. Therein, we describe an improve-
ment of this procedure by synthesising the linker derivative
23 in a one-pot procedure (Scheme 6).
The Sonogashira coupling was applied to all four nucleo-
sides 10, 13, 16 and 21 under the same conditions (i.e., a
Pd0/CuI catalyst system, DMF, Et3N, rt). The desired deriva-
tives were obtained in good-to-very-good yields and with
very high purities. The triphosphate synthesis was accom-
plished by following the procedure of Ludwig and Eckstein
reported in 1989.[37] In this step, 2-chloro-4H-1,2,3-benzo-
dioxaphosphorin-4-one is used as the phosphorylating re-
agent that is subsequently opened with pyrophosphate and
finally oxidised with a 2% iodine solution in pyridine/water.
In general, the reaction proceeds well, as demonstrated by
analysing the crude reaction product by 31P NMR spectro-
scopic analysis (not shown). These crude products were
treated directly with TFA/water to remove the Boc protect-
ing group from the amino function of the linker moiety. The
compounds were obtained in yields of around 20% for 28,
30 and 31 and only 8% for 29 due to the multistep purifica-
tion of these compounds, which is crucial after this step.
Thus, two purification steps had to be performed to achieve
the necessary purity. Anion-exchange chromatography at
48C on a fast-protein liquid chromatography (FPLC) system
was the first step followed by reverse phase (RP)-HPLC to
yield 28–31 with a high degree of purity in quantities be-
tween 22 and 120 mg (31 and 28, respectively). The obtained
triphosphate derivatives produced positive results in the fol-
lowing labelling reactions with the selected fluorescent dyes.
All the dyes (i.e., 5- and 6-carboxyfluorescein, 5- and 6-car-
boxy-X-rhodamine, Cy 3.0, and Cy 5.0) were used in the ac-
tivated N-hydroxysuccinimidyl ester form. The coupling re-
actions were carried out in dry DMF with KHCO3 as a het-
erogeneous base in the case of 28 and N,N-diisopropylethyl-
amine (DIPEA) as a base for
Scheme 6. Synthesis of the linker derivative 23. Reagents and conditions:
a) 1. DSC, anhydrous K2CO3, dry CH3CN, 08C, 20 h; 2. propargylamine,
KHCO3, 08C !rt, 5 h. Boc=tert-butoxycarbonyl, DSC=N,N’-hydroxy-
succinimidyl carbonate.
Compound 23 enables a direct Sonogashira coupling to
the 3’-modified iodonucleosides 10, 13, 16 and 21
(Scheme 7). This route saves two synthetic steps, and the
yield of 24 could be improved from 49% over three steps to
80% in one step.
the other nucleotides 29–31.
The purification of the final
compound was also very labori-
ous; prepurification on the RP-
FPLC system with self-packed
columns was necessary to
remove the starting-material
dye before final purification
with RP-HPLC was possible.
After this procedure, the four
reversible
terminators
2–5
(Scheme 2) were obtained with
a high degree of purity in quan-
tities between 1.5 and 6 mg (for
5 and 2, respectively) and con-
firmed by 1H and 31P NMR
spectroscopic and mass-spectro-
metric analysis. The 31P NMR
spectra of all four compounds
prove the structure of the ob-
tained compounds (Figure 1).
The four labelled reversible
terminators 2–5 were used in
Scheme 7. Linker attachment and triphosphate synthesis with all four 3’-O-(2-cyanoethyl) modified nucleosides
10, 13, 16, and 21. Reagents and conditions: a) dry DMF, triethylamine, CuI (0.2 equiv), [PdACTHNUTRGNE(UNG PPh3)4]
(0.1 equiv), alkyne 23 (1.2–2.0 equiv), rt, 3.5–5 h; b) 1. dry pyridine/dry dioxane (1:30, 2-chloro-4H-1,2,3-benzo-
dioxaphosphorin-4-one, rt, 15 min; 2. dry DMF, bis(tributylammonium)pyrophosphate, tributylamine, rt,
20 min; 2. iodine in pyridine/water (2%, 98:2), rt, 20 min; 4. 5% Na2SO3 solution; c) TFA (40 equiv) in water,
rt, 5–6 h. TFA=trifluoroacetic acid.
polymerase
incorporation
assays and in the demonstration
Chem. Eur. J. 2011, 17, 2903 – 2915
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2907

J. W. Engels et al.
of a proof-of-principle for reversible primer extension on
immobilised template oligonucleotides.
Polymerase identification for primer extension: Several as-
pects of the system appear to be critical to incorporate re-
versible terminators enzymatically into the nucleic acid sub-
strate. The polymerase must have the proper affinity and se-
lectivity towards modified nucleotides along with adequate
turnover rate and propensity to form terminated DNA mol-
ecules that remain stable until the 3’-modifications are re-
moved by chemical means. In turn, the reversible terminator
itself must display an appropriate balance between chemical
stability during enzymatic incorporation and the ability to
be unblocked under conditions mild enough for the DNA
structure to be retained for further enzymatic treatment. A
solution-based DNA primer extension system that is suita-
ble for simultaneous evaluation of substrate extension by
either single or multiple T residue(s) was used for initial
screening of a wide range of polymerases that represent
major families of DNA polymerases (A, B, X, Y, and re-
verse transcriptases) and some of their mutants. This system
enables the gel-based detection of primer extension prod-
ucts with single nucleotide resolution. By following the
primer-extension step, natural thymine triphosphate (dTTP)
was added to check the stability of the termination of the re-
sulting +1 product. Polymerases that exhibited even mini-
mal proofreading activity were found to be incompatible
with the assay due to primer truncation (data not shown).
We found 3’-O-(2-cyanoethyl)-dTTP[21] to be incorporated
into DNA (Figure 2A) by several reverse transcriptases
only (Figure 2B). All enzymes possess the ability to incorpo-
rate the compound (see even-numbered lanes for bands that
appear just above the primer), although the efficiency of in-
corporation varies. RevertAid M-MuLV reverse transcrip-
tase yielded a single primer-extension product (Figure 2B,
lane 4) that was resistant towards extension by supplement-
ed dTTP (Figure 2B, lane 5) due to its 3’-terminal modifica-
tion. Optimisation of the reaction conditions was necessary
due to the low efficiency of the +1 extension (reaction time
was 60 min) and was accomplished by including 1 mm MnCl2
into the reaction mixture. The reaction rate increased signif-
icantly, thus driving the primer-extension reaction to com-
pletion in just 15 min (Figure 2C). Incorporation of the dye-
conjugated 3’-modified compounds 2–5 under identical con-
ditions is of comparable speed. Figure 2D presents the pro-
files of the primer extension reactions that involve com-
pounds 2–5 after 5 min.
The specificity of primer extension was addressed by per-
forming the reaction under multiplex conditions, in which
four DNA substrates of different length, each competent for
extension by a particular nucleotide only (Figure 3A), were
simultaneously present in the reaction mixture. Incorpora-
tion of the individual 3’-modified nucleotides (Figure 3B,
Figure 1. 31P NMR spectra of the four dye-labelled reversible terminators
2–5.
2908
www.chemeurj.org
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 2903 – 2915

Nucleotide–Fluorophores in Primer Extension
FULL PAPER
have an internal self-complementary region that allows
folding into a hairpin structure, thus providing a double-
stranded 3’-terminus for extension by the polymerase. All
the oligonucleotides were functionalised with a 5’-C6
amino modification to enable their immobilisation onto
the amino-reactive array surface. We chose CodeLink-acti-
vated slides because their glass surface is coated by a syn-
thetic material that is compatible with our deprotection
conditions (fluoride ions) in contrast to glass slides that
would be harmed by a prolonged fluoride treatment (data
not shown). The spotting was accomplished by using a
10 mm solution of each oligonucleotide in 100 mm carbon-
ate buffer (pH 9.0) with the Versarray Chipwriter Pro Ar-
rayer. The spotting layout of the CodeLink slides is shown
in Figure 4B and is arranged so that the first primer-exten-
sion event is conveniently followed by signals that appear
vertically (Figure 4B, red letters) and the second event
represented by signals that appear horizontally (blue let-
ters). The primer layout shown in Figure 4 was spotted
twice on each slide (array). The composition of the reac-
tion mixture for the extension of immobilised primer was
Figure 2. The use of 3’-O-(2-cyanoethyl)-dNTP derivatives for primer ex-
tension by reverse transcriptases. A) DNA substrate used (* indicates the
radiolabelled 5’-terminus). The position in the template strand that di-
rects the nucleotide to be incorporated is in bold type. B) Incorporation
of 3’-O-(2-cyanoethyl)-dTTP. Lanes 1: starting-material primer, poly-
merases; 2 and 3: M-MuLV; 4 and 5: RevertAid M-MuLV; 6 and 7: Re-
vertAid H minus M-MuLV; 8 and 9: AMV (cloned); 10 and 11: Thermo-
Script; 12 and 13: SuperScript III; 14 and 15: Transcriptor. dTTP was
added into the reaction mix after the primer extension step, as indicated
for lanes 3, 5, 7, 9, 11, 13, and 15. C) Time course of incorporation of 3’-
O-(2-cyanoethyl)-dTTP by RevertAid M-MuLV reverse transctiptase in
the presence of Mn2+ ions. D) Extension of the corresponding substrates
by the dye-labelled 3’-O-(2-cyanoethyl)-modified nucleotides 2–5 during
a reaction time of 5 min.
lanes 2, 4, 6 and 8) or a mixture of all four (Figure 3B, lane
10) demonstrated the performance of the system under con-
ditions typical for sequencing reactions. DNA substrates are
extended by relevant 3’-modified nucleotides only and
remain unaffected in presence of natural 2’-deoxynucleotide
triphosphates (Figure 3B, lanes 3, 5, 7, 9 and 11). At this
stage, a more detailed assessment of polymerase specificity
was not performed due to differences between the condi-
tions for the liquid-state reactions and the ones involving
the microchip surface, typical for APEX.
Extension of immobilised DNA templates: The complete
system was tested in a reversibly terminating approach by
using the set of our four fluorescently labelled reversible ter-
minators 2–5, the RevertAid M-MuLV reverse transcriptase,
and the reaction buffer optimised for primer extension.
To test the labelled reversible terminators 2–5 for their in-
corporation into immobilised DNA, a set of 16 oligonucleo-
tides that differ at positions Q and Z was designed to pro-
vide a readout of all possible variants of two consecutive nu-
cleotides at X and Y positions (Figure 4A). These primers
Figure 3. Specific primer extension with 3’-O-(2-cyanoethyl)-dNTP deriv-
atives by RevertAid M-MuLV reverse transctiptase. A) Set of DNA sub-
strates used for specificity studies (* indicates radiolabelled 5’-termini).
The position in the template strand that directs the nucleotide to be in-
corporated is in bold type. B) The initial primer-extension step (lanes 2,
4, 6, 8, and 10) was followed by the addition of a dNTP mixture to the so-
lution containing all four natural dNTPs dissolved in water and incuba-
tion for 5 min (lanes 3, 5, 7, 9, and 11). Lanes 1 and 12: the starting-mate-
rial primer set; 2 and 3: 3’-O-(2-cyanoethyl)-dGTP; 4 and 5: 3’-O-(2-cya-
noethyl)-dATP; 6 and 7: 3’-O-(2-cyanoethyl)-dTTP; 8 and 9: 3’-O-(2-cya-
noethyl)-dCTP; 10 and 11: full set of 3’-O-(2-cyanoethyl)-dNTP nucleo-
tides.
Chem. Eur. J. 2011, 17, 2903 – 2915
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2909

J. W. Engels et al.
arrays spotted on the slide. The specificity of a particular
fluorescent-nucleotide incorporation into the first position X
of the primers was calculated by combining the signals from
four primers located within the same row and that bear the
same nucleotide at the Q position (Figure 4A) and by evalu-
ating the percentile of detected fluorescence within each of
the four primers. This approach enables the evaluation of
the polymerase specificity within the same sequence context.
The imaging results obtained after the first step showed that
all four reversible terminators were incorporated with high
selectivity (as summarised in Figure 5). We calculated that
the specificity of the incorporation of 5 (channel read G)
into primer GG reached 94.2% and the highest nonspecific
incorporation of 5 is 5.1% of the total signal into primer
AG. Incorporation of 2 (channel read T) was the most spe-
cific and varied from 96.8 to 99.6%; lower specificity was
observed for 3 and 4 (channel read A and C; 74.2–89.3 and
80.5–89.5%, respectively).
Figure 4. Schematic illustration of the hairpin primers and the spotting
layout of the slides. A) The variable positions in the template part are
marked as Z for the first template nucleotide and as Q for the second.
The variable positions immediately after the 3’-terminus are marked as X
and Y for the first and second nucleotide to be incorporated, respectively.
B) Each of the 16 oligonucleotides is applied on the slide in four spots ar-
ranged on the grid in squares. In a column (vertically), all the four
squares encode the same base for the first incorporation event (X). Each
square within a row (horizontally) encodes one of the four nucleotides
for the second incorporation event (Y).
Preliminary experiments revealed that a fraction of the
spotted oligonucleotides remains unextended by reverse
transcriptase during the first elongation step (data not
shown), thus resulting in the appearance of a high back-
ground during the second elongation step. The terminal de-
oxynucleotide transferase (TdT) catalyses a template-inde-
pendent addition of 2’-deoxyribonucleotides to the 3’-OH
further improved by the addi-
tion of bovine serum albumin
(BSA), a reported compound
to counteract protein adsorp-
tion and inactivation on wafer
surfaces.[38] Furthermore, nu-
merous
osmolytes
were
screened and 2-hydroxyme-
thyl[18]crown-6 was determined
to be the most efficient enhanc-
er of primer extension. Evalua-
tion of spotted CodeLink slides
for the acceptance of all four
fluorescently labelled reversible
terminators 2–5 confirmed the
time-dependent character of
the primer extension. The maxi-
mum signal was obtained after
a reaction time of 30 min (not
shown). This time-point was
used in the reversible primer-
extension
(Figure 5).
experiment
Figure 5 depicts the results of
the fluorescence detection after
the first and second elongation
steps in which each picture rep-
resents a scan of one channel.
The tables in the middle of
Figure 5 show the specificity
values of the first elongation
Figure 5. Imaging for illustration and analysis of the primer-extension events. Signals of the first elongation
appear vertically (in column), signals of the second elongation appear horizontally (in rows). Evaluation of the
specificity of the primer extension is given in the tables, and the target nucleotide is in bold type. Fluorescence
readout channels are shaded and present the data in columns (vertically) for the first extension step. Signals in
step estimated from the two circles are artefacts.
2910
www.chemeurj.org
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 2903 – 2915

Nucleotide–Fluorophores in Primer Extension
FULL PAPER
group of various DNA and RNA substrates and efficiently
incorporates the polymerisation terminator 2’,3’-didesoxy-
ACHTUNGTRENNUNGadenosine-5’-triphosphate (ddATP). On the other hand, we
TdT treatment (Figure 5, compare the images on the left
and right). However, this finding is true for the incorpora-
tion of all the modified nucleotides, except for compound 3
(Figure 5; see the channel read A) in which no increased
fluorescence was observed after the second elongation step
in column A. On the basis of this observation, it might be
speculated that ddATP incorporation by TdT was the most
complete when the complementary T nucleobase was pres-
ent in the template strand (all four primers of such structure
are located within the A column). On the other hand, unex-
tended primers might become available not only if treat-
ment by TdT is incomplete, but also as a result of pyrophos-
phorolytic dismutation.[39] One also should consider that all
four nucleotides employ a linker between the nucleobase
and fluorescent dye, which results in a propargylamine resi-
due after the cleavage of the dye. The modification of the
nucleobase is an additional stimulus for triphosphate-main-
tained phosphorolysis,[40] thus resulting in truncated primers
available for further extension.
found that TdT does not extend DNA substrates that bear
the 3’-O-(2-cyanoethyl) protecting group (results not
shown). Therefore, the first elongation step was followed by
incubation of the spotted slides with TdT in the presence of
ddATP. During this treatment, it was expected that unla-
belled (and uncleavable) 3’-dideoxy terminators would be
incorporated into the structure of previously unextended
primers and prevent them from being extended during the
second elongation step. Afterwards, the slide was imaged
and treated with a 1:1 solution of tetrabutylammonium fluo-
ride (TBAF; 1m)/THF and DMF at 458C for 15 min to
cleave the fluorescent dye and 3’-O-(2-cyanoethyl) group.
To ensure that the linker-dye moiety was completely re-
moved, this process was controlled by a separate imaging
step in which only the background-level fluorescence signal
could be detected (not shown).
On these deprotected slides, the second round of primer
extension was executed under conditions identical to the
first round. The pictures of fluorescence detection after the
second elongation are displayed in Figure 5 as well. During
this step, the same nucleotide is expected to be incorporated
into primers positioned in the same row (horizontally).
These pictures demonstrate the feasibility of the second
primer extension because the spots are detected at expected
positions. However, the efficiency of the incorporation of
the individual fluorescently labelled reversible terminators
2–5 was 15–30 times lower relative to that observed during
the first elongation step. The selectivity suffered mainly
from incorporation into those primers that were substrates
during the first elongation step. The latter phenomenon is
manifested by the appearance of additional signals in col-
umns (images on the right part of Figure 5) the positions of
which coincide with those labelled during the first elonga-
tion step (images on the left part of Figure 5). Several fac-
tors might account for this discrepancy during the primer ex-
tension on immobilised templates. During the experiments
on the slides, the efficiency of the unblocking process was
monitored by measuring the remnant fluorescence and ex-
pecting the 3’-O-(2-cyanoethyl) group to be removed as effi-
ciently as the linker-dye system. This outcome was true for
chemical unblocking in solution or on a controlled-pore
glass (CPG)-immobilised oligonucleotide,[22] but not necessa-
rily for hairpin-forming oligonucleotides immobilised on a
slide. Therefore, a probable reason for the decreased incor-
poration efficiency during the second elongation step could
be the nonquantitative chemical deprotection of the 3’-OH
group. On the other hand, we observed that the enzyme we
used extends a fraction rather than all spotted oligonucleo-
tides, and this feature should also be responsible for the par-
tial decrease in the signal intensity after the second elonga-
tion step. The apparently less-selective incorporation of
modified nucleotides 2–5 during the second elongation step
appears to be related to a fraction of substrates that remain
unblocked during the first elongation step and subsequent
Altogether, we have demonstrated the feasibility of the
template-directed incorporation of the 3’-terminated dye-la-
belled nucleotides 2–5 that can be cleaved in one step to
enable the next addition, thus providing the proof-of-princi-
ple of a functional system also targeted to run in cyclic
mode.
Conclusion
Herein, we have described the synthesis and full characteri-
sation of four fluorescently labelled reversible terminators
2–5. In these molecules, the 3’-blocking moiety and the
linker-dye system are removable during the same fluoride-
based deprotection treatment. The synthesised molecules
are recognised and accepted by an identified DNA poly-
merase. The reversible terminators and identified poly-
merase, which was able to incorporate them, were also used
in a cyclic reversibly terminating approach on CodeLink
slides spotted with hairpin oligonucleotide probes. These re-
sults demonstrate that the entire system is applicable in a
cyclic reversibly terminating approach. Further studies that
address the specificity and efficiency of the complete system
are ongoing.
Experimental Section
Materials and methods: 1H NMR spectra were recorded on Bruker AM,
DPX, and AV instruments at 250, 300, or 400 MHz and 300 K. 13C spec-
tra were recorded on Bruker AM, DPX, and AV instruments at 62.5, 75,
100, or 150 MHz. The chemical shifts d in 1H and 13C NMR spectra are
reported in ppm relative to the solvent signal. The fine structure of
proton signals was specified by s (singlet), d (doublet), t (triplet), q (quar-
tet), m (multiplet), dd (doublet of doublets), or br (broad) and in quota-
tions for pseudofine structures. Assignments in the 1H and 13C NMR
spectra were made by DEPT, COSY, HSQC, and HMBC experiments.
TLC analysis was carried out on polygram Sil G/UV254 by Macherey–
Nagel & Co. (Dueren; thickness=0.2 mm), 60 F254 (Merck KGaA, Darm-
stadt; thickness=0.2 mm), or RP-18W (Sigma–Aldrich Chemie GmbH,
Chem. Eur. J. 2011, 17, 2903 – 2915
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2911

J. W. Engels et al.
Steinheim; fluorescence
indicator=254 nm, thickness=0.15 mm).
Following GP2, the reaction of 5-iodo-3’-O-(2-cyanoethyl)-2’-deoxyuri-
dine (10; 250mg, 0.61mmol, 1.0 equiv) with alkyne 23 (305mg,
0.74mmol, 1.2 equiv) gave, after flash-column chromatography (CH2Cl2/
MeOH 99:1!90:10), 24 as a slightly yellow oil (340mg, 80%). RF =0.34
Column chromatography was carried out on silica gel 60 (Merck KGaA,
Darmstadt; 40–63 mm) at normal pressure or on silica gel 60 (Merck
KGaA, Darmstadt; 15–40 mm) at a pressure of 2–3 bar (flash chromatog-
raphy). FPLC was performed at 48C on a Pharmacia FPLC system
equipped with a single-path monitor UV-1 UV detector (l=254 nm) and
self-packed columns of different sizes with diethylamnoethyl (DEAE) se-
pharose material for ion-exchange FPLC (Sigma–Aldrich; 0.05m triethy-
lammonium hydrogencarbonate (TEAB) buffer pH 8.0 (A)/0.8m TEAB
buffer pH 8.0 (B) as the eluent) or octadecyl-functionalised silica gel
(Sigma–Aldrich; water (A)/CH3CN (B) as the eluent) for RP-FPLC. RP-
HPLC was performed on a Jasco LC-2000Plus HPLC system equipped
with a Jasco UV 2075Plus detector (detection at l=254 nm) and a Shi-
madzu RF-353 fluorescence detector (excitation and emission at the spe-
cific wavelength of the dye used). For the reversed-phase separation,
Phenomenex Jupiter 4m Proteo 90 A 4 mm columns (250ꢃ15 mm for
preparative and 250ꢃ4.6 mm for analytical separations) were used with
1m triethylammonium acetate (TEAA) buffer pH 6.5 (A)/water (B)/
CH3CN (C) as the eluents. UV detection was accomplished at l=
254 nm. Ion-exchange HPLC was performed on a Jasco LC-900 HPLC
system equipped with a Jasco UV-970 detector (detection at l=254 nm)
and a Dionex BioLC DNAPac PA-100 column (250ꢃ9 mm) with water/
0.25m tris(hydroxymethyl)aminomethane hydrochloride (Tris·HCl) buffer
(pH 8)/1m sodium chloride solution as the eluents. ESI mass-spectrome-
try was performed on a Fisons instrument equipped with a VG platform
II with quadrupol analyser. UV spectroscopy was performed on a Jasco
V-650 spectrometer with 0.1 cm cuvettes. Fluorescence spectroscopy was
performed on a Hitachi F4500 fluorescence spectrometer with 0.3 cm
cuvettes. Elemental analyses were recorded on a Foss-Heraeus CHN-O
Rapid instrument. The numbering of the atoms to assign the NMR signal
are not related to the IUPAC numbering or the numbers used in the
1
(CH2Cl2/MeOH 90:10); H NMR (300 MHz, [D6]DMSO, 300 K): d=1.37
(s, 23-H; 9H), 2.16–2.31 (m, 2’-H; 2H), 2.77 (t, 25-H; 2H), 2.88 (m, 12-
H; 2H), 3.06 (q, 20-H; 2H), 3.37 (t, 19-H; 2H), 3.48–3.65 (m, 5’-H, 24-H,
14-H, 15-H, 16-H, 17-H, 18-H; 7ꢃ2H), 3.93–3.97 (m, 4’-H; 1H), 4.02 (d,
9-H; 2H), 4.14 (quintet, 3’-H; 1H), 4.95 (quintet, 11-H; 1H), 5.13–5.19
(m, 5’-OH; 1H), 6.07 (t, 1’-H; 1H), 6.68–6.77 (m, 20-NH; 1H), 7.93 (t, 9-
NH; 1H), 8.14 (s, 6-H; 1H), 11.64 ppm (s, N3H, J9,9ꢀNH =5.7, J19,20
=
6.0Hz; 1H); 13C NMR (75 MHz, [D6]DMSO, 300 K): d=18.23 (25-C),
19.72 (12-C), 28.21 (23-C), 30.68 (9-C), 36.90 (2’-C), 39.69 (20-C), 61.14
(5’-C), 63.56 (24-C), 67.91 (11-C), 69.17 (19-C), 69.49, 69.60, 69.73, 70.21,
70.30 (14-C, 15-C, 16-C, 17-C, 18-C), 74.49 (7-C), 77.58 (22-C), 79.08 (3’-
C), 84.64 (1’-C), 84.89 (4’-C), 89.57 (8-C), 98.23 (5-C), 117.64 (13-C),
119.26 (26-C), 143.40 (2-C), 143.56 (6-C), 154.84 (10-C), 155.54 (21-C),
161.49 ppm (4-C); ESI+-MS (m/z): calcd: 692.71 [C31H44N6O12]; found:
710.5 [M+H2O]+, 715.4 [M+Na]+; elemental analysis (%) calcd for
C31H44N6O12: C 53.75, H 6.40, N 12.12; found: C 53.53, H 6.51, N 11.88.
General procedure for the synthesis of N-deprotected triphosphates
(GP3): The N-Boc protected nucleoside was vacuum dried in a Schlenk
flask for 2 days, the flask was flushed with argon, and the nucleoside
(0.1 mmol) dissolved in dry pyridine (100 mL) and dry dioxane (300 mL).
A freshly prepared solution of 2-chloro-4H-1,2,3-benzodioxaphosphorin-
4-one (1m, 1.15 equiv) in dry dioxane was added with a syringe and the
reaction mixture stirred for 15 min at room temperature. Dry DMF
(150 mL), bis(tributylammonium)pyrophosphate (1m, 1.5 equiv), and trib-
utylamine (1m, 4.35 equiv) were added simultaneously with syringes. The
reaction mixture was stirred for 20 min before a solution of iodine in pyr-
idine/water (2%, 98:2) was added (1 mL). The reaction mixture was left
for a further 20 min. Afterwards, the excess of iodine was quenched with
5% Na2SO3 solution, and the mixture was evaporated to dryness. The
crude triphosphate derivatives (0.1 mmol) were dissolved in water (5 mL)
containing TFA (40 equiv). The solution was stirred for 5–6 h at room
temperature and subsequently concentrated in vacuum. For the ion-ex-
change FPLC purification, the deprotected triphosphate derivatives were
dissolved in Millipore water (8–12 mL) and filtered through a 0.45 mm sy-
ringe filter. The following gradient was used for the separation at a flow
names of the compounds. Some signals in the H and 13C NMR spectra of
1
the triphosphate compounds are doubled, although this did not occur in
the spectra of nucleoside derivatives 24–27. We assume that this outcome
is due to the diastereoisomers that are obtained by introduction of the
linker as a racemic mixture. These nucleoside derivatives are given in the
13C NMR data as X.XX/Y.YY.
General procedure for the Sonogashira coupling (GP2): In a heat-gun-
dried Schlenk flask in an Ar atmosphere, the nucleoside was dissolved in
dry DMF (0.1 mmolmLꢀ1) and triethylamine (5 equiv). The mixture was
degassed by using the freeze/thaw technique for three times. After the re-
action mixture had been warmed to room temperature, CuI (0.2 equiv)
rate of 4 mLminꢀ1
:
0%
B
(0 mL)!50%
B
(500 mL)!100%
B
(650 mL). The combined fractions were lyophilised, dissolved in 1–2 mL
of Millipore water, and filtered through a 0.45 mm syringe filter before
the second purification step by preparative RP-HPLC. The following gra-
dient was used at a flow rate of 6 mLminꢀ1: 0 min: 10:90:0 A/B/C,
15 min: 10:60:30 A/B/C; 16 min: 0:0:100 A/B/C; 22 min: 0:0:100 A/B/C;
23 min: 10:90 A/B/C, 30 min: 10:90:0 A/B/C. The final yields were calcu-
and [PdACHTUNGTRENNUNG(PPh3)4] (0.1 equiv) were added. In a separate heat-gun-dried
Schlenk flask, alkyne 23 (1.2–2.0 equiv) was dissolved in dry DMF
(250 mL) and added in two portions to the reaction mixture with a sy-
ringe. One half portion was added immediately and the other half por-
tion after 1 h. The reaction mixture was stirred for 3.5–5h at room tem-
perature in the dark. The reaction mixture was evaporated to dryness,
the crude product redissolved in CH2Cl2 (10mL) and 5% ethylenediami-
netetraacetate (EDTA; 5mL) solution, and stirred vigorously for 10min.
The layers were separated and the organic layer was washed with 5%
EDTA solution (5mL), dried over sodium sulphate, and concentrated in
vacuum. The residue was purified by flash-column chromatography. For
example, the synthesis of 5-(prop-2-ynyl)carbamic acid-1-{2-[2-(2-tert-bu-
toxycarbonylaminoethoxy)ethoxy]ethoxymethyl}-2-cyanoethyl ester-3’-O-
(2-cyanoethyl)-2’-deoxyuridine (24).
1
lated from the H NMR spectrum of the HPLC purified triphosphate de-
rivatives because the products contained traces of triethylammonium ace-
tate after RP-HPLC purification (for example, the synthesis of 28).
Following GP3, the reaction of the N-Boc-protected nucleoside 24
(500 mg, 0.72 mmol, 1.0 equiv) yielded triphosphate 28 as a colourless oil
(140 mg, 23%). RF =0.39 (iPr/H2O/EtMe2N 6:3.8:0.2); IE-FPLC: elution
concentration=0.29m TEAB/buffer (32% B); RP-HPLC: retention
time=14.92 min; 1H NMR (300 MHz, D2O, 300 K): d=2.32 (ddd, 2’-H;
1H), 2.55 (ddd, 2’-H’; 1H), 2.81 (t, 22-H; 2H), 2.89–2.99 (m, 12-H; 2H),
3.25 (m, 20-H; 2H), 3.67–3.79 (m, 14-H, 15-H, 16-H, 17-H, 18-H, 19-H;
2912
www.chemeurj.org
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 2903 – 2915

Nucleotide–Fluorophores in Primer Extension
FULL PAPER
6ꢃ2H), 3.84 (t, 21-H; 2H), 4.12–4.26 (m, 5’-H, 9-H; 2ꢃ2H), 4.37 (brs,
4’-H; 1H), 4.45–4.49 (m, 3’-H; 1H), 5.12–5.23 (m, 11-H; 1H), 6.26 (dd,
for additional 5min at the same temperature. The reactions were stopped
by adding an equal volume of the STOP solution: 95% formamide and
100mm EDTA. The products were resolved by using a denaturing poly-
acrylamide (PAA) gel (7m urea; 15% 29:1). The gel was dried on What-
man paper and autoradiographed by using a Fuji phosphorimager screen.
For the evaluation of the polymerase specificity, the four substrate sys-
tems were used (Figure 3B). Concentrations of 5 nm of each DNA
duplex and 20 UmLꢀ1 of the RevertAid M-MuLV reverse transcriptase
were used. The reaction buffer was the same as described above with the
addition of MnCl2 (1mm) and dithiothreitol (DTT; 5mm). The reactions
were performed for 5min at 378C, then an aliquot of the reaction mix-
ture was supplemented with all four natural dNTP derivatives (up to
100 mm each). The reaction was allowed to proceed for additional 5min
at the same temperature. Samples were analysed as described above.
1’-H; 1H), 8.17 ppm (s, 6-H,
J2’,2’’ =14.5, J1’,2’ =8.4, J1’,2’’ =5.7, J21,22 =
6.0 Hz; 1H); 13C NMR (75 MHz, D2O, 300 K): d=18.91 (22-C), 20.21
(12-C), 31.49 (9-C), 37.28/37.30 (2’-C), 39.67 (21-C), 64.43 (21-C), 66.31
(d, 5’-C), 66.95, 70.09, 70.12, 70.19, 70.84, 71.18 (14-C, 15-C, 16-C, 17-C,
18-C, 19-C), 69.30 (11-C), 73.98/74.01 (7-C), 80.45/80.47 (3’-C), 84.31 (d,
4’-C), 86.50 (1’-C), 90.93 (8-C), 99.77 (5-C), 118.78 (13-C), 120.45 (23-C),
145.34/145.35 (6-C), 151.03 (2-C), 157.27/157.29 (10-C), 164.99 ppm (4-C,
J
4’,aꢀP =8.82, J
5’,aꢀP =5.60 Hz); 31P NMR (121 MHz, D2O, 300 K): d=
ꢀ10.97 (d, g-P), ꢀ11.69 (d, a-P), ꢀ23.43 ppm (t, b-P); ESIꢀ-MS (m/z):
calcd for [C26H39N6O19P3]: 832.54; found: 831.4 [MꢀH]ꢀ, 849.6 [MꢀH+
H2O]ꢀ.
Synthesis of 2: KHCO3 (3mg, 30 mmol, 2.0 equiv) was added to a solu-
tion of triphosphate 28 (12mg, 14 mmol, 1.0 equiv) in dry DMF (500 mL).
The reaction mixture was cooled to 08C, and 5- and 6-carboxyfluores-
Spotting of DNA arrays and APEX reactions on CodeLink arrays with
fluorescently labelled 3’-O-(2-cyanoethyl)-dNTP derivatives. All oligonu-
cleotide primers were diluted to a con-
centration of 10 mm in carbonate
buffer (100mm, pH 9.0) and spotted
onto CodeLink-activated slides from
SurModics (Eden Prairie, MN, USA)
with a Versarray Chipwriter Pro Ar-
rayer (BioRad Laboratories, Hercules,
CA, USA) as 8ꢃ8-probe squares ac-
cording to the scheme shown in
Figure 4. The oligonucleotide primers
were immobilised onto the slide sur-
face and the slides blocked before the
APEX reactions, as suggested by the
cein-(N-hydroxysuccinimidyl)ester (10mg, 20 mmol, 1.5 equiv) was dis-
solved in dry DMF (250 mL) and added in two portions to the reaction
mixture, one half immediately and the other after 1h at 08C. The reac-
tion mixture was stirred for 20 h at 08C. After evaporation of the solvent,
the crude product was purified by RP-FPLC at 48C and preparative RP-
HPLC (see GP4 in the Supporting Information). The triethylammonium
salt of the labelled triphosphate 2 was obtained as a yellow solid (6mg,
34%) as a mixture of the two regioisomers, each consisting of two diaste-
reomers. RF =0.65 (H2O/CH3CN 80:20); RP-FPLC: elution concentra-
tion=2.5!6% B; RP-HPLC: retention time=16.49 min isomer A,
vendor. APEX reaction cycles on Co-
deLink arrays consisted of the following steps: 1) Blocking of DNA
arrays with blocking solution (50 mL, Tris-acetate (33mm, pH 7.9 at
378C); magnesium acetate (10mm), potassium acetate (66mm), BSA
(1mgmLꢀ1), DTT (10mm), Tween-20 (1%), NP-40 (1%)) for 10min at
378C. 2) The first primer-extension reaction with 2–5 was carried out by
treating the slide with the set of the fluorescently labelled reversible ter-
minators 2–5 and RevertaAid M-MuLV reverse transcriptase at 378C for
30min. An aliquot of the reaction mixture (35 mL) was used for one slide
containing Tris–acetate (33mm, pH 7.9 at 378C), magnesium acetate
(10mm), potassium acetate (66mm), BSA (0.1mgmLꢀ1), MnCl2 (1mm),
DTT (10mm), Tween-20 (0.5%), Nonidet P-40 (0.5%), 2-hydroxyme-
thyl[18]crown-6 (1%), 2–5 (100 mm each), and RevertAid M-MuLV re-
verse transcriptase (10 UmLꢀ1). 3) Blocking of the starting-material oligo-
nucleotide primers on the array was accomplished with TdT (MBI Fer-
mentas, Vilnius, Lithuania) and ddATP by incubating the arrays for
20min at 378C. The TdT reaction mixture consisted of the reaction mix-
ture with an aliquot (40 mL) of potassium cacodylate (200mm), Tris
(25mm), Triton X-100 (0.01%, v/v), CoCl2 (1mm, pH 7.2 at 258C),
ddATP (1mm), and TdT enzyme (80 U). 4) The reactions were stopped
by washing with deionized water at 958C. The slides were dried, covered
with a droplet of Slowfade light antifade reagent (Molecular Probes, OR,
USA), and imaged with Genorama QuattroImager detector (Asper Bio-
tech Ltd.). The fluorescence signal intensities were extracted and ana-
lysed with BaseCaller module of the Genorama genotyping software
package (Asper Biotech Ltd.). 5) Cleavage of the fluorescence dye and
the 3’-O-(2-cyanoethyl) group from the incorporated fluorescence la-
belled 3’-O-(2-cyanoethyl)-blocked nucleotides was accomplished with
TBAF (1m) in THF. Before the unblocking reaction, the DNA arrays
were washed three times with THF. The unblocking reaction was carried
out by immersing the DNA arrays for 15min at 458C into a 1:1 solution
of TBAF/THF (Fluka, Buchs, Switzerland) and dry DMF. 6) The DNA
arrays were washed with deionized water, dried, and read with the Gen-
orama QuattroImager detector to be sure that the unblocking reaction
was successful. 7) Before the second primer-extension reaction, the slides
were blocked again with blocking solution (see above, step 1) for 10min
at 378C. 8) The second primer-extension reaction using the fluorescently
labelled 3’-O-(2-cyanoethyl)-dNTP derivatives 2–5 and the reverse tran-
scriptase was carried out under the reaction conditions described for the
first extension step (step 2). 9) The reactions were stopped by washing
16.90 min isomer
B (UV detection), 16.53 min isomer A, 16.94 min
isomer B (fluorescence detection); UV absorption: l=494, 277, 235 nm;
fluorescence: lex =492, lem =512 nm; 1H NMR (250 MHz, D2O, 300 K):
d=2.15–2.37 (m, 2’-H; 1H), 2.44–2.59 (m, 2’-H’; 1H), 2.81–2.93 (m, 12-H,
23-H; 2ꢃ2H), 3.57–3.91 (m, 14-H, 15-H, 16-H, 17-H, 18-H, 19-H, 20-H,
22-H; 8ꢃ2H), 4.00 (brs, 9-H; 2H), 4.10–4.32 (m, 4’-H, 5’-H; 3H), 4.10–
4.32 (m, 3’-H; 1H), 4.99–5.14 (m, 11-H; 1H), 6.04–6.25 (m, 1’-H; 1H),
6.69–7.04 (m, fluorescein-H; 4H), 7.09–7.32 (m, fluorescein-H; 2H),
7.56–8.46 ppm (m, 6-H, fluorescein-H; 5H); 31P NMR (121 MHz, D2O,
300 K): d=ꢀ10.63 (d, g-P), ꢀ11.56 (dd, a-P), ꢀ23.09 ppm (t, b-P, Ja,b
=
19.6, Jg,b =19.6Hz); ESIꢀ-MS (m/z): calcd for [C47H49N6O25P3]: 1190.85;
found: 594.6 [Mꢀ2H]2ꢀ
,
1110.8 [MꢀPO3-H]ꢀ, 554.8 [MꢀPO3-2H]2ꢀ
,
1190.4 [MꢀH]ꢀ.
Solution-based primer-extension experiments: The oligodeoxynucleotides
(shown in Figure 2A) were used to form the DNA substrate for the ex-
tension and termination experiments using 3’-O-(2-cyanoethyl)-modified
nucleotides and an extension experiment using fluorescence-labelled re-
versible terminators 2–5. The primer strand was 5’-radiolabelled and an-
nealed to the complementary (template) strand by heating for 5min to
958C and gradually cooling to room temperature over 2 h. For the
primer extension, 10 nm DNA duplex, polymerases in at least 10-fold
molar excess, and 50 mm of the 3’-modified nucleotides was used. The 3’-
modified nucleotides were additionally purified by enzymatic depletion
of the natural nucleotide counterpart before use.[41] The reactions were
performed for 60 min at 378C in a total volume of the reaction mixture
of 20 mL, containing 33mm Tris–acetate (pH 7.9, at 378C), 10mm magne-
sium acetate, 66mm potassium acetate, and BSA (0.1mgmLꢀ1). After
completion of the reaction, an aliquot of the reaction mixture was supple-
mented by dTTP (up to 50 mm), and the reaction was allowed to proceed
Chem. Eur. J. 2011, 17, 2903 – 2915
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2913

J. W. Engels et al.
with deionized water at 958C. The slides were dried, read, and analysed
as described above for first primer-extension reaction.
[11] M. B. Welch, C. I. Martinez, A. J. Zhang, S. Jin, R. Gibbs, K. Bur-
gess, Chem. Eur. J. 1999, 5, 951–960.
[12] J. Y. Ju, D. H. Kim, L. R. Bi, Q. L. Meng, X. P. Bai, Z. M. Li, X. X.
Li, M. S. Marma, S. Shi, J. Wu, J. R. Edwards, A. Romu, N. J. Turro,
Proc. Natl. Acad. Sci. USA 2006, 103, 19635–19640.
[13] T. S. Kim, D. R. Kim, H. C. Ahn, D. Shin, D. R. Ahn, Chembiochem
2009, 11, 75–78.
Acknowledgements
[14] J. Guo, N. Xu, Z. M. Li, S. L. Zhang, J. Wu, D. H. Kim, M. S.
Marma, Q. L. Meng, H. Y. Cao, X. X. Li, S. D. Shi, L. Yu, S. Kala-
chikov, J. J. Russo, N. J. Turro, J. Y. Ju, Proc. Natl. Acad. Sci. USA
2008, 105, 9145–9150.
[15] J. Guo, L. Yu, N. J. Turro, J. Y. Ju, Acc. Chem. Res. 2010, 43, 551–
563.
We gratefully acknowledge financial support from the EU project
COOP-CT-2005–01/623-Array-SBS. We thank Stefan Bernhardt, Priit
Tomson, Hanno Roomere, and Triinu Kꢄressaar for technical assistance;
Ilona Priess for measuring the mass spectra; Yangzhou Li, Marek Kwiat-
kowski, Andras Fçldesi, Sarunas Zigmantas, and Alina Stura for the ini-
tial experiments and Chelsea Ann Stern for critical reading of the manu-
script.
[16] D. R. Bentley, S. Balasubramanian, H. P. Swerdlow, G. P. Smith, J.
Milton, C. G. Brown, K. P. Hall, D. J. Evers, C. L. Barnes, H. R.
Bignell, J. M. Boutell, J. Bryant, R. J. Carter, R. K. Cheetham, A. J.
Cox, D. J. Ellis, M. R. Flatbush, N. A. Gormley, S. J. Humphray, L. J.
Irving, M. S. Karbelashvili, S. M. Kirk, H. Li, X. Liu, K. S. Maising-
er, L. J. Murray, B. Obradovic, T. Ost, M. L. Parkinson, M. R. Pratt,
I. M. J. Rasolonjatovo, M. T. Reed, R. Rigatti, C. Rodighiero, M. T.
Ross, A. Sabot, S. V. Sankar, A. Scally, G. P. Schroth, M. E. Smith,
V. P. Smith, A. Spiridou, P. E. Torrance, S. S. Tzonev, E. H. Vermaas,
K. Walter, X. Wu, L. Zhang, M. D. Alam, C. Anastasi, I. C. Aniebo,
D. M. D. Bailey, I. R. Bancarz, S. Banerjee, S. G. Barbour, P. A. Bay-
bayan, V. A. Benoit, K. F. Benson, C. Bevis, P. J. Black, A. Boodhun,
J. S. Brennan, J. A. Bridgham, R. C. Brown, A. A. Brown, D. H.
Buermann, A. A. Bundu, J. C. Burrows, N. P. Carter, N. Castillo,
M. C. E. Catenazzi, S. Chang, R. N. Cooley, N. R. Crake, O. O.
Dada, K. D. Diakoumakos, B. Dominguez-Fernandez, D. J. Earn-
shaw, U. C. Egbujor, D. W. Elmore, S. S. Etchin, M. R. Ewan, M. Fe-
durco, L. J. Fraser, K. V. Fuentes Fajardo, W. S. Furey, D. George,
K. J. Gietzen, C. P. Goddard, G. S. Golda, P. A. Granieri, D. E.
Green, D. L. Gustafson, N. F. Hansen, K. Harnish, C. D. Haudens-
child, N. I. Heyer, M. M. Hims, J. T. Ho, A. M. Horgan, Nature 2008,
456, 53–59.
[1] E. S. Lander, L. M. Linton, B. Birren, C. Nusbaum, M. C. Zody, J.
Baldwin, K. Devon, K. Dewar, M. Doyle, W. FitzHugh, R. Funke,
D. Gage, K. Harris, A. Heaford, J. Howland, L. Kann, J. Lehoczky,
R. LeVine, P. McEwan, K. McKernan, J. Meldrim, J. P. Mesirov, C.
Miranda, W. Morris, J. Naylor, C. Raymond, M. Rosetti, R. Santos,
A. Sheridan, C. Sougnez, N. Stange-Thomann, N. Stojanovic, A.
Subramanian, D. Wyman, J. Rogers, J. Sulston, R. Ainscough, S.
Beck, D. Bentley, J. Burton, C. Clee, N. Carter, A. Coulson, R.
Deadman, P. Deloukas, A. Dunham, I. Dunham, R. Durbin, L.
French, D. Grafham, S. Gregory, T. Hubbard, S. Humphray, A.
Hunt, M. Jones, C. Lloyd, A. McMurray, L. Matthews, S. Mercer, S.
Milne, J. C. Mullikin, A. Mungall, R. Plumb, M. Ross, R. Show-
nkeen, S. Sims, R. H. Waterston, R. K. Wilson, L. W. Hillier, J. D.
McPherson, M. A. Marra, E. R. Mardis, L. A. Fulton, A. T. Chinwal-
la, K. H. Pepin, W. R. Gish, S. L. Chissoe, M. C. Wendl, K. D. Dele-
haunty, T. L. Miner, A. Delehaunty, J. B. Kramer, L. L. Cook, R. S.
Fulton, D. L. Johnson, P. J. Minx, S. W. Clifton, T. Hawkins, E. Bran-
scomb, P. Predki, P. Richardson, S. Wenning, T. Slezak, N. Doggett,
J. F. Cheng, A. Olsen, S. Lucas, C. Elkin, E. Uberbacher, M. Frazier,
Nature 2001, 409, 860–921.
[17] J. W. Milton, X. Smith, M. Brennan, J. S. Barnes, C. L. Liu, S. Rue-
diger, 2004, WO 2004/018497.
[2] J. C. Venter, M. D. Adams, E. W. Myers, P. W. Li, R. J. Mural, G. G.
Sutton, H. O. Smith, M. Yandell, C. A. Evans, R. A. Holt, J. D. Go-
cayne, P. Amanatides, R. M. Ballew, D. H. Huson, J. R. Wortman, Q.
Zhang, C. D. Kodira, X. H. Zheng, L. Chen, M. Skupski, G. Subra-
manian, P. D. Thomas, J. Zhang, G. L. Gabor Miklos, C. Nelson, S.
Broder, A. G. Clark, J. Nadeau, V. A. McKusick, N. Zinder, A. J.
Levine, R. J. Roberts, M. Simon, C. Slayman, M. Hunkapiller, R.
Bolanos, A. Delcher, I. Dew, D. Fasulo, M. Flanigan, L. Florea, A.
Halpern, S. Hannenhalli, S. Kravitz, S. Levy, C. Mobarry, K. Reinert,
K. Remington, J. Abu-Threideh, E. Beasley, K. Biddick, V. Bonazzi,
R. Brandon, M. Cargill, I. Chandramouliswaran, R. Charlab, K.
Chaturvedi, Z. Deng, V. Di Francesco, P. Dunn, K. Eilbeck, C.
Evangelista, A. E. Gabrielian, W. Gan, W. Ge, F. Gong, Z. Gu, P.
Guan, T. J. Heiman, M. E. Higgins, R. R. Ji, Z. Ke, K. A. Ketchum,
Z. Lai, Y. Lei, Z. Li, J. Li, Y. Liang, X. Lin, F. Lu, G. V. Merkulov,
N. Milshina, H. M. Moore, A. K. Naik, V. A. Narayan, B. Neelam,
D. Nusskern, D. B. Rusch, S. Salzberg, W. Shao, B. Shue, J. Sun, Z.
Wang, A. Wang, X. Wang, J. Wang, M. Wei, R. Wides, C. Xiao, C.
Yan, Science 2001, 291, 1304–1351.
[18] F. Chen, E. A. Gaucher, N. A. Leal, D. Hutter, S. A. Havemann, S.
Govindarajan, E. A. Ortlund, S. A. Benner, Proc. Natl. Acad. Sci.
USA 2010, 107, 1948–1953.
[19] A. Fçldesi, A. Keller, A. Stura, S. Zigmantas, M. Kwiatkowski, D.
Knapp, J. W. Engels, Nucleosides, Nucleotides Nucleic Acids 2007,
26, 271–275.
[20] M. Kwiatkowski, 1996, WO 9623087.
[21] A. C. Keller, S. Serva, D. C. Knapp, M. Kwiatkowski, J. W. Engels,
Collect. Czech. Chem. Commun. 2009, 74, 515–534.
[22] D. C. Knapp, J. D’Onofrio, J. W. Engels, Bioconjugate Chem. 2010,
21, 1043–1055.
[23] E. Dawson, G. R. Abecasis, S. Bumpstead, Y. Chen, S. Hunt, D. M.
Beare, J. Pabial, T. Dibling, E. Tinsley, S. Kirby, D. Carter, M. Pa-
paspyridonos, S. Livingstone, R. Ganske, E. Lohmmussaar, J. Zern-
ant, N. Tonisson, M. Remm, R. Magi, T. Puurand, J. Vilo, A. Kurg,
K. Rice, P. Deloukas, R. Mott, A. Metspalu, D. R. Bentley, L. R.
Cardon, I. Dunham, Nature 2002, 418, 544–548.
[24] T. Jaijo, E. Aller, G. Garcia-Garcia, M. J. Aparisi, S. Bernal, A.
Avila-Fernandez, I. Barragan, M. Baiget, C. Ayuso, G. Antinolo, M.
Diaz-Llopis, M. Kulm, M. Beneyto, C. Najera, J. M. Millan, Invest.
Ophthalmol. Visual Sci. 2010, 51, 1311–1317.
[3] A. C. Syvꢁnen, Hum. Mutat. 1999, 13, 1–10.
[4] T. Pastinen, A. Kurg, A. Metspalu, L. Peltonen, A. C. Syvanen,
Genome Res. 1997, 7, 606–614.
[25] N. Tonisson, J. Zernant, A. Kurg, H. Pavel, G. Slavin, H. Roomere,
A. Meiel, P. Hainaut, A. Metspalu, Proc. Natl. Acad. Sci. USA 2002,
99, 5503–5508.
[26] M. J. Robins, P. J. Barr, J. Org. Chem. 1983, 48, 1854–1862.
[27] K. Sonogashira, Y. Tohda, N. Hagihara, Tetrahedron Lett. 1975, 16,
4467–4470.
[5] A. Kurg, N. Tonisson, I. Georgiou, J. Shumaker, J. Tollett, A. Met-
spalu, Genet. Test. 2000, 4, 1–7.
[6] M. L. Metzker, R. Raghavachari, S. Richards, S. E. Jacutin, A. Civi-
tello, K. Burgess, R. A. Gibbs, Nucleic Acids Res. 1994, 22, 4259–
4267.
[7] B. Canard, R. S. Sarfati, Gene 1994, 148, 1–6.
[28] M. Bobek, I. Kavai, R. A. Sharma, S. Grill, G. Dutschman, Y. C.
Cheng, J. Med. Chem. 1987, 30, 2154–2157.
[29] R. G. Breiner, W. C. Rose, J. A. Dunn, J. E. Macdiarmid, T. J.
Bardos, J. Med. Chem. 1990, 33, 2596–2602.
[8] M. B. Welch, K. Burgess, Nucleosides Nucleotides 1999, 18, 197–201.
[9] H. Pelletier, M. R. Sawaya, A. Kumar, S. H. Wilson, J. Kraut, Sci-
ence 1994, 264, 1891–1903.
[10] B. Canard, B. Cardona, R. S. Sarfati, Proc. Natl. Acad. Sci. USA
1995, 92, 10859–10863.
[30] J. Davoll, J. Chem. Soc. 1960, 131–138.
2914
www.chemeurj.org
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 2903 – 2915

Nucleotide–Fluorophores in Primer Extension
FULL PAPER
[31] J. S. Pudlo, M. R. Nassiri, E. R. Kern, L. L. Wotring, J. C. Drach,
L. B. Townsend, J. Med. Chem. 1990, 33, 1984–1992.
[32] F. Seela, M. Zulauf, Synthesis 1996, 726–730.
[33] H. Saneyoshi, K. Seio, M. Sekine, J. Org. Chem. 2005, 70, 10453–
10460.
[34] A. Okamoto, K. Kanatani, I. Saito, J. Am. Chem. Soc. 2004, 126,
4820–4827.
[35] F. Seela, M. Zulauf, Chem. Eur. J. 1998, 4, 1781–1790.
[36] W. T. Markiewicz, J. Chem. Res. Synop. 1979, 24–25.
[37] J. Ludwig, F. Eckstein, J. Org. Chem. 1989, 54, 631–635.
[38] I. Erill, S. Campoy, N. Erill, J. Barbe, J. Aguilo, Sens. Actuators, B
2003, 96, 685–692.
[39] R. S. Anderson, F. J. Bollum, K. L. Beattie, Nucleic Acids Res. 1999,
27, 3190–3196.
[40] P. R. Meyer, S. E. Matsuura, R. G. So, W. A. Scott, Proc. Natl. Acad.
Sci. USA 1998, 95, 13471–13476.
[41] M. L. Metzker, R. Raghavachari, K. Burgess, R. A. Gibbs, BioTech-
niques 1998, 25, 814–817.
Received: July 9, 2010
Revised: October 18, 2010
Published online: February 3, 2011
Chem. Eur. J. 2011, 17, 2903 – 2915
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2915