Linker-free fluorophore-labeled oligonucleotides: Synthesis of 3′-fluoresceinylthymidine building blocks and their coupling reactions

Article abstract of DOI:10.1055/s-2005-870024

Reported here are the syntheses of two thymidine derivatives with an amide-linked 6-carboxyfluorescein residue at their 3′-position that are suitable for the synthesis of fluorophore-labeled oligonucleotides. The first is a 5′-phosphoramidite with a pivaloyl-protected carboxyfluorescein residue in the lactone form. It was prepared from 3′-azido-3′- deoxythymidine (AZT) in three steps and 73% overall yield and coupled on solid support. The second is an imidazolide of thymidine 5′-monophosphate that was obtained from AZT in five steps and 48% overall yield. The imidazolide can be coupled to amino-terminated nucleic acids in aqueous solution. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-2005-870024

PAPER
2327
Linker-Free Fluorophore-Labeled Oligonucleotides: Synthesis of
3¢-Fluoresceinylthymidine Building Blocks and their Coupling Reactions
F
luoresceinylthym
i
idine
B
e
uilding Blo
l
cks s Griesang, Eric Kervio, Clemens Richert*
Institut für Organische Chemie, Universität Karlsruhe (TH), 76131 Karlsruhe, Germany
Fax +49(721)6084825; E-mail: cr@rrg.uka.de
Received 2 February 2005; revised 5 April 2005
cise location and a possible engagement in tightly folded
complexes. Further, such rigidly linked constructs may fa-
cilitate the in situ detection of binding through fluores-
cence anisotropy measurements.
Abstract: Reported here are the syntheses of two thymidine deriv-
atives with an amide-linked 6-carboxyfluorescein residue at their
3¢-position that are suitable for the synthesis of fluorophore-labeled
oligonucleotides. The first is a 5¢-phosphoramidite with a pivaloyl-
protected carboxyfluorescein residue in the lactone form. It was
prepared from 3¢-azido-3¢-deoxythymidine (AZT) in three steps and
73% overall yield and coupled on solid support. The second is an
imidazolide of thymidine 5¢-monophosphate that was obtained from
AZT in five steps and 48% overall yield. The imidazolide can be
coupled to amino-terminated nucleic acids in aqueous solution.
A report on linker-free deoxyadenosine-fluorescein hy-
brids incorporated in synthetic oligonucleotides has re-
cently appeared.7 In these hybrids, the covalent link
between a carboxyfluorescein and the nucleotide involved
the nucleobase. In order to allay concerns that this may af-
fect the base pairing properties, we decided to pursue an
alternative approach, where the direct link is between the
deoxyribose and carboxyfluorescein. For azo dyes it is
known that they can be directly attached to the 3¢- or 5¢-
position of nucleic acids via ester bonds.8 We were inter-
ested in producing building blocks that can be employed
both in solid phase syntheses of oligonucleotides and in
non-enzymatic primer extension reactions involving
deprotected nucleic acids in aqueous solution. The former
synthetic approach calls for phosphoramidites of protect-
ed nucleosides,9 whereas the latter synthetic approach is
commonly based on imidazolides of nucleotides.10 We
chose carboxyfluorescein as fluorophore, as this chro-
mophore is among the best established in biomolecular
applications. Carboxyfluorescein is commercially avail-
able and shows favorable absorption and emission proper-
ties, even when it is linked to DNA.11 An amidic link was
considered most desirable, as it is small, uncharged, rigid,
and stable toward the deprotection conditions typically
employed at the end of a solid phase synthesis of oligonu-
Key words: DNA, nucleosides, oligomerization, oligonucleotides,
solid-phase synthesis
Introduction
Fluorophore-labeled nucleic acids play important roles in
biomedical applications. For example, DNA sequencing
by the chain terminator method1 is predominantly per-
formed with fluorophore-labeled dideoxynucleotides.2
Massively parallel detection of gene expression can be
achieved with DNA chips that detect fluorophore-labeled
DNA or RNA targets.3 Direct, in situ detection of specific
sequences, including those formed during polymerase
chain reactions (PCR) can be achieved with molecular
beacons,4 which carry a combination of a fluorophore and
a quencher.5 Förster resonance energy transfer (FRET)
can be employed to measure spatial closeness of two chro-
mophores in a structural motif.
Common to the fluorophore labeling approaches currently cleotides. Of the four deoxynucleotides that make up
favored is the use of a flexible linker between the nucleic DNA, thymidine was chosen for the current study, as it
acid and the luminescent chromophore. This results in does not complicate syntheses with a need for protection
limited precision in the location of the labeled compound, of the nucleobase. This led to compounds 1 and 2 as our
a loss of correlation between fluorophore and DNA in primary synthetic targets (Figure 1).
fluorescence anisotropy measurements, an increase in the
O
O
molecular weight, and often an increased likelihood of ad-
sorption on surfaces. We have recently shown that small
appendages at the termini of oligonucleotides can be suc-
cessfully engaged in complex formation with the termini
of DNA duplexes.6 A more precise location in folded
structures of nucleic acids is also desirable for fluoro-
phores. Developing a construct where the precise location
of the fluorophore is achieved through a direct amidic link
to a nucleotide is one first step towards achieving the pre-
(i-Pr)2N
NH
O
P
NH
N
N
P
O
O
N
O
N
O
NCC2H4O
O
O
O
Na
NH
NH
O
O
O2C
1
2
O
O
HO
PivO
O
O
OPiv
O
SYNTHESIS 2005, No. 14, pp 2327–2334
Advanced online publication: 14.07.2005
DOI: 10.1055/s-2005-870024; Art ID: T01405SS
© Georg Thieme Verlag Stuttgart · New York
x
x.
x
x
.
2
0
0
5
Figure 1 Building blocks for the synthesis of linker-free fluo-
rescein-labeled oligonucleotides with 3¢-terminal thymidine residues.

2328
N. Griesang et al.
PAPER
ther, the duplex showed intense fluorescence (lmax = 518
Results and Discussion
nm upon excitation at 495 nm). This was also true for the
The synthesis of 1 started from 3¢-azido-3¢-deoxythymi- second oligonucleotide prepared (Scheme 2). Besides the
dine (3) (Scheme 1). This antiviral compound is commer- 3¢-fluoresceinyl residue, dodecamer 15 features a lysine
cially available, and larger quantities could be prepared as residue at its 5¢-terminus, linked to the DNA via a 2,2-
described.12 Hydrogenation gave amine 4,13 ready for cou- dimethyl-3-hydroxypropionic acid residue.20 This allows
pling to the fluorophore. The N-hydroxysuccinimide for immobilization of oligonucleotides on aldehyde-dis-
(NHS) ester of 6-carboxyfluorescein dipivalate 5 was playing glass slides in the form of microarrays.21 Chain
chosen as reaction partner. To generate it, the 6-isomer of assembly started from functionalized support 1221
carboxyfluorescein dipivalate was prepared from the (Scheme 2) and proceeded to doubly modified 15, which
commercial mixture of 5/6-carboxyfluorescein isomers was obtained in mass spectrometrically pure form after
following a literature protocol.14 The NHS ester 5 was RP-HPLC. The fluorophore-labeled oligonucleotide
used, rather than an in situ activated acid, to avoid side re- elutes four minutes later than the unlabeled control, and
actions involving the lactone functionality. Such side re- fractions containing this compound are easy to spot by
actions have been described for related xanthene dyes15 their color and fluorescence. Again, the fluoresceinyl sub-
and the 5-isomer of carboxyfluorescein16 when coupling stituent had a minor effect on the melting point of the du-
to a primary amine. For the conversion of carboxyfluores- plex with the target strand. When compared to the
cein to the NHS ester, a protocol similar to that described unmodified control duplex 16:17, a DTm of –2.0 °C was
by Laurent and collaborators for the mixture of fluores- measured (Table 1). Lysine-terminated 15 is currently be-
cein isomers17 was employed. The use of EDC as coupling ing used to optimize the immobilization procedure for oli-
agent, in combination with an acidic aqueous work-up, gonucleotides on aldehyde slides.
gave sufficiently pure 5, avoiding low-yielding chromato-
graphic steps. The acylation occurred under conditions
similar to those reported for other acylations of aminothy-
midine.18 In this instance, amine 4 and NHS ester 5 were
reacted in DMF without addition of a base to avoid the
loss of pivaloyl groups. Thymidine-fluorescein hybrid 6
was thus obtained in 85% yield after chromatographic pu-
rification.
Compound 6 was 5¢-phosphitylated to 1 (Scheme 1),
which was purified by column chromatography after a ba-
sic aqueous work-up. Compound 1 was used in the last
coupling step of DNA syntheses involving commercial 5¢-
phosphoramidites.19 First, self-complementary hexamer
10 was prepared, starting from controlled pore glass (cpg)
loaded with the 5¢-terminal deoxyadenosine residue 7.
Automated synthesis gave pentamer 8, followed by man-
ual coupling of 1 to give the protected hexamer 9. UV-
melting curves for the duplex of deprotected 10 (Figure 2)
showed a melting point (Tm) decrease of no more than
2.6 °C per modification when compared to that of un-
modified control hexamer ACGCGT (11) (Table 1). Fur-
Figure 2 Representative UV-melting curves of the modified oli-
gonucleotide 15 (filled squares) or unmodified 5¢- TCATTC-
TGTTCT-3¢ (16) (open circles), each with their fully complementary
strand 17 at 2.25 mM strand concentration, 1 M NaCl, 10 mM phos-
phate buffer, pH 7; monitored at 260 nm.
O
O
N
b,
PivO
O
O
OPiv
NH
HO
NH
O
O
N
O
O
O
HO
O
N
O
O
O
NH
5
c
O
R
1
O
O
3, R = N3
4, R = NH2
a
PivO
O
OPiv
6
Scheme 1 Reagents and conditions: a) H2, Pd/C, MeOH, 92%; b) 5, DMF, 85%; (c) DIEA, chloro-N-(diisopropylamino)cyanoethoxy-
phosphoramidite, MeCN, 86%.
Synthesis 2005, No. 14, 2327–2334 © Thieme Stuttgart · New York

PAPER
Fluoresceinylthymidine Building Blocks
2329
Since the linker-free labeling of the 3¢-terminus of oligo- in non-enzymatic replication.28 Two possible routes to the
nucleotides seemed to be tolerated, we next turned to the imidazolide were considered: the oxidative generation of
synthesis of 5¢-imidazolide 2, which is envisioned to func- an imidazolide from a 5¢-H-phosphonate under Atherton
tion in non-enzymatic, template-directed extension reac- conditions,29 and the coupling of imidazole to a 5¢-phos-
tions.23–26 Primer extension products from reactions phate. The phosphate could also have been generated di-
involving 2, when performed in microarray format, can be rectly from the 5¢-alcohol,30 or through oxidation of an H-
expected to be detectable by fluorescence scanning of phosphonate, which in turn should have been accessible
chip surfaces. To evaluate routes without the complication from the alcohol using more reactive phosphorus(III) re-
of the hydrolytic lability of pivaloyl-protected fluores- agents.31 Since the H-phosphonate promised the greatest
ceinyl residues, a model synthesis was performed with 3¢- flexibility, compound 19 was chosen as intermediate.
TBDMS-protected thymidine 18 (Scheme 3).
The synthesis of 18 followed the protocol of Lan and col-
Imidazolides of nucleotides have been known for over 40 leagues.32 A more recent synthetic procedure33 gave lower
years,27 and have been established as activated monomers yields. Conversion to H-phosphonate 19 and the oxidative
O
5'
HO
FluoPiv2
3'
HN
3'
A
C
G
C
G
T
O DMT
OH
T
GiBu
CBz
GiBu
CBz
GiBu
CBz
GiBu
CBz
O
N
a
O
b
BzA
c
NH
O
O
O
O
O
ABz
ABz
NH
Fluo
3'
NH
O
5'
5'
O2C
O
O
NH
O
HO
cpg
O
7
8
9
10
O
FluoPiv2
HN
O2C
3'
3'
T
OH
NH2
HN
CBz
CBz
O
T
T
Lys
T
GiBu
T
GiBu
O
O
N
DP
O
PO2
O
5'
NH
O
T
T
T
C
A
T
T
C
T
G
T
T
C
T
CBz
CBz
O
T
T
O
OH
NH
T
T
ABz
CBz
ABz
CBz
O
H
N
T
T
TFA
b
c
a
5'
5'
O
O
O
NH
DP
DP
O
(CH2)11
NH
N
O
Lys
Lys
O
O
NH
3'
Fluo
NH
O
O2C
O
O
HO
12
14
15
13
O
Scheme 2 Reagents and conditions: a) DNA synthesis with 5¢-phosphoramidites; b) coupling cycle with 1; c) NH4OH. Fluo = fluorescein
residue.
Synthesis 2005, No. 14, 2327–2334 © Thieme Stuttgart · New York

2330
N. Griesang et al.
PAPER
The free H-phosphonate of protected 21 was too insoluble
in acetonitrile, however, for conversion to 2 under the
conditions established for 19. Both DMF and DMSO as
solvents failed to give detectable conversion. Only DMF
and Et3N (1:1) gave conversion. In situ addition of water
then led to the removal of the pivaloyl groups. Imida-
zolide 2 could thus be isolated as sodium salt after precip-
itation in the presence of sodium perchlorate.
Table 1 UV Melting Points and Hyperchromicities of DNA Du-
plexes
Duplex
Tm (°C)a
DTm (°C)b Hyperchromicity (%)b
(11)2 (control)
(10)2
31.1 0.322
25.8 2.3
6.6 1.422
–2.6
–2.0
6.2 2.3
23.1 0.2
19.7 1.0
16:17 (control) 49.0 0.3
15:17 47.0 0.6
An exploratory experiment with nucleosidic amine 4 as
reaction partner in the absence of a template performed in
aqueous buffer (Scheme 5) showed that 2 couples to ami-
no-terminal nucleic acids, producing an (elongated)
chain. The reaction was analyzed by MALDI MS under
conditions that allow for quantitation.34 After 40 hours,
the imidazolide had disappeared from the spectrum and
peaks for three products were detected at a ratio of ap-
proximately 9:2:3. These were for free phosphate 25, the
elongation product 24, and a compound whose mass fits
that of a dimerization product of 2 with hydrolyzed imida-
zolide group, respectively. Since small peaks compatible
with condensation of 24 with an additional residue of 2 as
well as a trimerization product of 2 were detected, it is fair
to assume that the self-condensation of 2 observed in the
absence of a template involves the phenolic hydroxyl
groups of fluoresceinyl moieties and not the phosphate,
which would have led to a pyrophosphate-linked dimer
only. Oligomers of fluoresceinylnucleosides, such as
those formed accidentally here, may be of interest as
FRET fluorophore arrays with large photon capture cross
sections.35 Further, the detectable rate of coupling be-
tween 4 and 2, even in the absence of a template, suggests
a Average of four melting points one standard deviation; conditions:
1 M NaCl, 10 mM phosphate buffer, pH 7, strand concentration: 3.5
mM (entries 1, 2), 2.25 mM (entries 3, 4).
b Melting point difference to control strand (per fluoresceinyl group).
O
N
O
N
NH
O
H P
O
NH
a
O
HO
O
O
O
O
HNEt3
OTBDMS
OTBDMS
18
19
O
O
P
NH
b
N
N
O
N
O
O
O
Na
OTBDMS
20
O
Scheme 3 Reagents and conditions: a) 1. diphenyl phosphite, pyri-
dine, 2. H2O, Et3N, 50%; b) TMS-imidazole, CCl4, Et3N, MeCN,
36%.
O
H P
O
NH
a
c
O
6
2
N
O
O
coupling to 20 were achieved in moderate yield, using an
NH
adaptation of the method of Ushioda and colleagues.29
R
O
The same route was employed for fluoresceinylthymidine
6 as starting material (Scheme 4). Phosphitylation with
diphenyl phosphite generated a phenyl phosphite whose
in situ hydrolysis with water–triethylamine gave H-phos-
phonate 21. Initial attempts to purify 21 led to a mixture
containing 22 and 23, whose fluoresceinyl moieties lack
pivaloyl groups. Hydrolysis was probably induced by re-
sidual triethylamine introduced during chromatography,
though precipitation with NaClO4 also gave partial loss of
pivaloyl groups. Subsequently, the mixture was deliber-
ately converted to 23 with sodium methoxide. Attempts to
convert 23 to 2 analogously to the activation of 19 to 20
were unsuccessful, however, due to the low solubility of
the deprotected intermediate. Ion exchange to the pyridin-
ium form did not help. In a mixture of 21 and 23, only the
former was reactive toward the Atherton reagents. Sepa-
ration of 2 and 23 by chromatography or precipitation
failed.
O
21 R =
O
PivO
O
OPiv
OPiv
CO2
O
b
22 R =
O
23 R =
CO2
O
OH
O
Scheme 4 Reagents and conditions: a) 1. diphenyl phosphite, pyri-
dine, 2. H2O, Et3N, 85%; b) NaOMe, MeOH, 83%; c) 1. TMS-imida-
zole, CCl4, Et3N, DMF, 2. H2O, 73%.
Intermediate 21 could be isolated in satisfactory yields
with 5% acetic acid in the eluent during chromatography.
Synthesis 2005, No. 14, 2327–2334 © Thieme Stuttgart · New York

PAPER
Fluoresceinylthymidine Building Blocks
2331
matrixes were a mixture of 2,4,6-trihydroxyacetophenone (0.3 M in
EtOH) and diammonium citrate (0.1 M in H2O) (2:1) or diammoni-
um tartrate (0.1 M in H2O), abbreviated below as TC21. For small
molecules, 2,5-dihydroxybenzoic acid (0.1 M in MeCN) or 6-aza-
thio-2-thiothymidine (sat. solution in MeCN) was used as MALDI
matrix. UV/Vis and Fluorescence spectra were recorded using a
Perkin-Elmer spectrophotometer Lambda 10 and a Perkin-Elmer lu-
minescence spectrometer LS50B respectively in 600 mL quartz cu-
vettes.
that a moderate template effect should suffice for elonga-
tion of a primer to dominate over hydrolysis of 2 and self-
coupling reactions of the activated nucleotide. Imida-
zolide 2 may therefore become useful for screening cata-
lysts for non-enzymatic primer extension reactions,
including assays performed on microarrays analyzed by
fluorescence microscopy. This use could complement the
application of directly linked fluorescein-nucleoside hy-
brids as building blocks for oligomers with precisely po-
sitioned fluorophores, such as 10 and 15.
6-Carboxyfluorescein-3¢,6¢-dipivalate NHS-Ester (5)
The compound was synthesized as reported for the mixture of iso-
mers,17 starting from isomerically pure 6-carboxyfluorescein dipiv-
alate; yield 92%; Rf 0.35 (CH2Cl2–MeOH, 98:2).
2
4
+
1H NMR (400 MHz, CDCl3): d = 8.40 (d, J = 7.9 Hz, 1 H), 8.18 (d,
J = 7.9 Hz, 1 H), 7.95 (s, 1 H), 7.10 (d, J = 2.1 Hz, 2 H), 6.77–6.84
(m, 4 H), 2.90–2.89 (m, 4 H), 1.36 (br s, 18 H).
a
O
N
N-(3¢,6¢-Dipivaloylfluorescein-6-ylcarbonyl)-3¢-amino-3¢-de-
oxythymidine (6)
NH
HO
To a stirred solution of amine 4 (230 mg, 0.954 mmol) in anhyd
DMF (10 mL) was added the NHS ester 5 (916 mg, 1.19 mmol) un-
der stirring. After 16 h at r.t., the mixture was diluted with CH2Cl2
and washed with 5% aq NaHCO3 solution (2 ×) and brine (1 ×). The
combined organic phases were dried (Na2SO4), and the solvent was
removed by rotary evaporation. The residue was purified by column
chromatography (silica gel) with a stepwise gradient of CH2Cl2–
MeOH from 95:5 to 9:1 to give 620 mg of 6 (85%) as a yellow solid;
Rf 0.35 (CH2Cl2–MeOH, 9:1).
1H NMR (500 MHz, CDCl3): d = 9.85 (br s, 1 H), 8.16 (d, J = 7.8
Hz, 1 H), 8.03 (d, J = 7.8 Hz, 1 H), 7.65 (2 s, each 1 H), 7.03–6.99
(m, 2 H), 6.77–6.67 (m, 4 H), 6.03 (t, J = 5.6 Hz, 1 H), 4.55 (m, 1
H), 3.83–3.70 (m, 3 H), 2.27–2.24 (m, 2 H), 1.79 (s, 3 H), 1.36 (s, 9
H), 1.35 (s, 9 H).
O
O
O
N
O
N
HN
O
NH
O
P
NH
P
O
HO
O
+
O
O
O
O
O
O
NH
NH
O2C
O
O2C
O
O
O
HO
HO
24
25
O
O
condensation
products of 2
13C NMR (126 MHz, CDCl3). d = 176.9, 176.8, 168.1, 166.2, 164.1,
153.2, 152.6, 152.5, 151.4, 151.3, 150.8, 139.8, 136.1, 129.8, 128.8,
128.7, 128.4, 125.5, 122.7, 117.9, 117.8, 115.5, 115.3, 110.9, 110.5,
110.4, 85.7, 84.5, 81.7, 62.0, 50.2, 37.0, 12.5.
+
Scheme 5 Reagents and conditions: a) 0.2 M MgCl2, 1 M NaCl, 0.5
M aq 4-(2-hydroxyethyl)piperazin-1-ethansulfonic acid (HEPES)
buffer, pH 7.9.
MALDI-TOF MS (linear, positive, TC21): m/z calcd for
C41H41N3O12 [M + H]+: 768.8; found: 769.1.
Anhyd solvents were purchased over molecular sieves and used
without further purification. Reagents were the best available grade
from Acros (Geel, Belgium) or Aldrich/Fluka/Sigma (Deisenhofen,
Germany). Samples of 3¢-azido-3¢-deoxythymidine (AZT) were
purchased from Toronto Research Chemicals Inc. (Ontario, Cana-
da) or were prepared by the authors;12 the isomeric mixture of 5(6)-
carboxyfluorescein was from Fluka (Deisenhofen, Germany). Re-
versed or 5¢-phosphoramidites were from Chemgenes (Ashland,
MA, USA). All other reagents for DNA synthesis were from Proli-
go (Hamburg, Germany). Oligonucleotides were purified by re-
verse-phase HPLC, with a gradient of MeCN in 0.1 M Et3NOAc
(pH 7.0) and detection at 260 nm, using Nucleosil C4 columns for
modified oligonucleotides and C18 columns for unmodified oligo-
nucleotides (both 250 × 4.6 mm; Macherey-Nagel, Düren, Germa-
ny). Yields of oligonucleotides are based on the intensity of the
product peaks in the integration of the HPLC traces of crude prod-
ucts. For modified oligonucleotides, the extinction coefficients
were calculated as the sum of the extinction coefficient of the un-
modified oligonucleotide and the extinction coefficient of any non-
nucleosidic residue. MALDI-TOF spectra were acquired on a Bru-
ker REFLEX IV spectrometer in negative, linear mode, using the
software XACQ 4.0.4 and XTof 5.1. MALDI-TOF spectra were re-
corded in positive, linear mode in the case of small molecules, and
in negative, linear mode in the case of oligonucleotides. MALDI
N-(3¢,6¢-Dipivaloylfluorescein-6-ylcarbonyl)-3¢-amino-3¢-de-
oxythymidine-5¢-O-yl-cyanoethyl-N,N-diisopropylphosphor-
amidite (1)
To a stirred solution of 6 (75 mg, 100 mmol) and diisopropylethyl-
amine (DIEA) (20 mL, 200 mmol) in anhyd MeCN (600 mL) was
added 2-cyanoethyl-N,N-diisopropylchlorophosphoramidite (24
mL, 110 mol). After stirring for 45 min, the mixture was poured into
aq sat. NaHCO3 solution (20 mL) and extracted with CH2Cl2
(3 × 20 mL). The organic phase was dried (Na2SO4), and the solvent
was removed by rotary evaporation to 0.3 mL. The residue was pu-
rified by column chromatography (silica gel previously neutralized
with 0.5% Et3N in elution mixture) using a mixture of cyclohexene–
acetone (1:1). The title compound was obtained as pale yellow foam
(83 mg, 86%); Rf 0.7 (hexane–acetone–Et3N, 66:33:0.5).
31P NMR (202 MHz, CD3CN): d = 150.1, 149.6.
MALDI-TOF MS (linear, positive, ATT): m/z calcd for
C50H58N5O13P ([M + H]+): 968.9; found: 969.3.
Nucleoside 5¢-O-Phosphonates 19 and 21; General Procedure
To a stirred solution of diphenyl phosphite (~10 equiv, 10 mmol,
1.93 mL) in anhyd pyridine (10 mL), was added dropwise a solution
of the alcohol 18 or 6 (~1 mmol) in anhyd pyridine (2 mL). The mix-
ture was stirred at r.t. under argon and the reaction progress was
Synthesis 2005, No. 14, 2327–2334 © Thieme Stuttgart · New York

2332
N. Griesang et al.
PAPER
monitored by TLC or MALDI-TOF MS until the phosphitylation
was complete. After stirring for 1 h, the mixture was treated with
H2O–Et3N (~500 mL). After stirring for 1 h, the solvents were evap-
orated in vacuo. The crude product was taken up in CH2Cl2 and
washed with 1 M Et3NHCO3 buffer (2 ×) and brine (1 ×). The com-
bined organic phases were dried (Na2SO4), and the solvent was re-
moved by rotary evaporation. After purification by column
chromatography (silica gel) the 5¢-O-phosphonates 19 and 21 were
obtained as yellow oils.
31P NMR (202 MHz, CD3OD): d = 4.71 (J = 620 Hz).
MALDI-TOF MS (linear, positive, TT21): m/z calcd for
C41H42N3O14P ([M + H]+): 832.7; found: 833.2.
Sodium 3¢-Amino-3¢-deoxy-N-(fluorescein-4-ylcarbonyl)thy-
midine-5¢-O-phosphonate (23)
Crude 21 (166 mg, 0.2 mmol), prepared as described above, was
dissolved in MeOH (3 mL) and treated with a solution of NaOMe
in MeOH (1 M, 600 mL, 0.6 mmol, 3 equiv). After 2 h, the red solu-
tion was added to CH2Cl2 (40 mL). After stirring at 0 °C for 30 min,
the red precipitate was isolated by filtration. The solid was washed
twice with anhyd acetone (20 mL total) to give 113 mg (83%) of 23
as a red solid; Rf 0.8 (CH2Cl2–MeOH–Et3N, 47:47:6).
1H NMR (500 MHz, CD3OD): d = 8.08 (m, 2 H), 7.88 (m, 1 H), 7.74
(m, 1 H), 7.06–7.04 (m, 2 H), 6.56–6.52 (m, 4 H), 6.85 (d, J = 620
Hz, 1 H, HP), 6.37 (t, J = 6.5 Hz, 1 H), 4.76–4.74 (m, 1 H), 4.22–
4.10 (m, 3 H), 2.48–2.44 (m, 2 H), 1.96 (s, 3 H).
Triethylammonium 3¢-O-(tert-Butyldimethylsilyl)thymidine-5¢-
O-phosphonate (19)
Prepared from 18 in 50% yield (261 mg) according to the general
protocol. Column (silica gel): with a stepwise gradient of CH2Cl2–
MeOH–Et3N from 90:5:5 to 85:10:5; Rf 0.5 (CH2Cl2–MeOH–Et3N,
90:5:5).
1H NMR (CDCl3, 250 MHz): d = 7.69 (s, 1 H), 6.85 (d, J = 620 Hz,
1 H, HP), 6.30 (t, J = 6.4 Hz, 1 H), 4.41–4.39 (m, 1 H), 4.01–3.94
(m, 3 H), 3.04 [q, J = 7.3 Hz, 6 H, HN(CH2CH3)3], 2.12–2.07 (m, 2
H), 1.87 (s, 3 H), 1.25 [t, J = 7.3 Hz, 9 H, HN(CH2CH3)3], 0.86 (s,
9 H), 0.01 (s, 6 H).
13C NMR (126 MHz, D2O, NaOD): d = 180.2, 179.9, 173.8, 169.2,
166.4, 158.7, 158.6, 157.6, 156.3, 151.6, 151.5, 142.6, 137.1, 133.9,
131.8, 131.2, 131.1, 128.7, 128.5, 128.4, 112.5, 111.6, 103.5, 84.7,
82.8, 63.2, 50.4, 36.2, 11.6.
31P NMR (202 MHz, CDCl3): d = 4.52 (J = 620 Hz).
31P NMR (202 MHz, CD3OD): d = 4.78 (J = 620 Hz).
MALDI-TOF MS (linear, positive, DHB): m/z calcd for
C16H28N2O7PSi ([M + Na]+): 443.4; found: 443.7.
MALDI-TOF MS (linear, positive, TC21): m/z calcd for
C31H25N3O12P ([M + H]+): 663.5; found: 664.3.
Sodium 3¢-O-(tert-Butyldimethylsilyl)thymidine-5¢-O-phosphor-
imidazolide (20)
Sodium 3¢-Amino-3¢-deoxy-N-(fluorescein-4-ylcarbonyl)thy-
midine-5¢-O-phosphorimidazolide (2)
To a stirred solution of H-phosphonate 19 (57.1 mg, 0.109 mmol)
in a mixture of MeCN–CCl4 (2:1, 3 mL total volume) was added
Et3N (77 mL, 0.547 mmol). The mixture was then treated with tri-
methylsilylimidazole (80.8 mL, 0.547 mmol, 5 equiv) at r.t. After 2
h, the solution was then added in small portions to a solution of
NaClO4 (27 mg, 0.218 mmol) in anhyd acetone (18 mL) and anhyd
Et2O (9 mL). After stirring for 30 min, the white precipitate was
isolated by filtration. The solid was washed three times with anhyd
acetone (30 mL total) and twice with anhyd Et2O (20 mL total) to
give 20 mg (36%) of 20 as a white solid; Rf 0.45 (CH2Cl2–MeOH–
Et3N, 90:5:5).
To a stirred solution of the H-phosphonate 21 (61 mg, 0.073 mmol)
in a mixture of DMF–CCl4–Et3N (1:1:1, 1.5 mL total) was added
dropwise trimethylsilylimidazole (55 mL, 0.375 mmol, 5 equiv) at
r.t., and the reaction was allowed to proceed for 5 h until MALDI-
TOF MS showed full conversion. After stirring the yellow solution
for 5 h, the mixture was treated with H2O (13.5 mL, 0.75 mmol). The
deprotection of the pivaloyl groups was monitored by MALDI-TOF
MS. After 2 h, the dark red solution was then added in small por-
tions to a solution of NaClO4 (27 mg, 0.218 mmol) in anhyd acetone
(27 mL). After stirring for 30 min, the red precipitate was isolated
by filtration. The solid was washed three times with anhyd acetone
(30 mL total) and twice with anhyd Et2O (20 mL total), to give 40
mg (73%) of 2 as a red solid; Rf 0.8 (CH2Cl2–MeOH–Et3N,
47:47:6).
1H NMR (D2O, 400 MHz, conc. and pH/pD dependent): d = 7.83 (br
s, 1 H), 7.78 (br s, 1 H), 7.75 (br s, 1 H), 7.42 (br s, 2 H), 7.10 (br s,
1 H), 7.05 (br s, 1 H), 6.92–6.78 (m, 2 H), 6.53–6.49 (m, 2 H), 6.45–
6.40 (m, 2 H), 6.08 (t, J = 6 Hz, 1 H), 4.42 (q, J = 6 Hz, 1 H), 4.01–
3.94 (m, 1 H), 3.93–3.91 (m, 1 H), 3.85–3.78 (m, 1 H), 2.33–2.25
(m, 1 H), 2.25–2.16 (m, 1 H), 1.79 (s, 3 H).
1H NMR (CD3OD, 250 MHz): d = 7.81 (s, 1 H), 7.63 (s, 1 H), 7.21
(s, 1 H), 6.96 (s, 1 H), 6.21 (t, J = 6.1 Hz, 1 H), 4.28–4.26 (m, 1 H),
3.94–3.80 (m, 3 H), 2.14–2.05 (m, 2 H), 1.86 (s, 3 H), 0.80 (s, 9 H),
–0.03 (s, 6 H).
31P NMR (202 MHz, CD3OD): d = –9.31.
MALDI-TOF MS (linear, positive, ATT): m/z calcd for
C19H30N4O7PSi ([M + Na]+): 509.5; found: 509.5.
3¢-Amino-3¢-deoxy-N-(3¢,6¢-dipivaloylfluorescein-6-ylcarbon-
yl)thymidine-5¢-O-phosphonate (21)
31P NMR (202 MHz, D2O): d = –8.82.
MALDI-TOF MS: m/z calcd for C34H27N5O12P ([M + H]+): 730.8;
Prepared on a 0.086 mmol scale from 6 in 85% yield (61 mg) ac-
cording to the general procedure. The reaction was monitored by
MALDI-TOF MS. Column (silica gel) with the solvent mixture:
CH2Cl2–MeOH–AcOH, 85:10:5); Rf 0.3 (CH2Cl2–MeOH–AcOH,
85:10:5).
found: 730.7.
Coupling of 4 with 2
Solutions of amine 4 (550 mM, 0.4 mL, 225 nmol) and imidazolide
2 (48.8 mM, 4.6 mL, 224 nmol) in HEPES buffer solution (500 mM;
containing NaCl, 1 M and MgCl2, 200 mM; pH 7.9) were combined
to give a final concentration of 45 mM of either nucleoside/nucle-
otide. The mixture was kept at r.t., and the conversion was followed
by MALDI-TOF mass spectrometry.
1H NMR (500 MHz, CD3OD): d = 8.22 (dd, J = 8.2, 0.9 Hz, 1 H),
8.15 (d, J = 8.2 Hz, 1 H), 7.82 (d, J = 0.9 Hz, 1 H), 7.76 (s, 1 H),
7.19–7.18 (m, 2 H), 6.93–6.87 (m, 4 H), 6.80 (d, J = 620 Hz, 1 H,
HP), 6.28 (t, J = 6.5 Hz, 1 H), 4.67–4.65 (m, 1 H), 4.13–4.06 (m, 3
H), 2.44–2.32 (m, 2 H), 1.94 (s, 3 H), 1.37 (s, 18 H).
13C NMR (126 MHz, CD3OD): d = 176.5, 168.5, 166.6, 165.0,
153.1, 153.0, 151.4, 150.9, 140.9, 136.5, 129.8, 128.9, 128.3, 124.9,
122.6, 117.9, 115.7, 110.6, 110.1, 84.6, 83.3, 83.2, 82.0, 63.3, 63.2,
50.8, 38.8, 36.8, 26.0, 11.1.
DNA Synthesis
Oligodeoxynucleotides were prepared on a 1 mmol scale, following
the protocol recommended by the manufacturer of the DNA synthe-
sizer (8909 Expedite, Perseptive Biosystems, system software
Synthesis 2005, No. 14, 2327–2334 © Thieme Stuttgart · New York

PAPER
Fluoresceinylthymidine Building Blocks
2333
2.01). The couplings of the modified phosphoramidite 1 to solid-
support-bound DNA 8 and 13 were carried out in a polypropylene
vessel using a 0.1 M solution of the phosphoramidite in DCI activa-
tor solution (100 mL). The oligodeoxynucleotides were cleaved
from the solid support and deprotected with 30% aq ammonia (400
mL) for 15 h at r.t. Excess ammonia was removed from the aqueous
solutions with a gentle stream of compressed air. The solution was
filtered (pore size 0.2 mm) and used directly for HPLC purification.
(6) (a) Bleczinski, C. F.; Richert, C. J. Am. Chem. Soc. 1999,
121, 10889. (b) Tuma, J.; Richert, C. Biochemistry 2003, 42,
8957. (c) Dogan, Z.; Paulini, R.; Rojas Stütz, J. A.;
Narayanan, S.; Richert, C. J. Am. Chem. Soc. 2004, 126,
4762. (d) Tuma, J.; Paulini, R.; Rojas Stütz, J. A.; Richert, C.
Biochemistry 2004, 43, 15680.
(7) Sigmund, H.; Pfleiderer, W. Helv. Chim. Acta 2003, 86,
2299.
(8) Gunnlaugsson, T.; Kellya, J. M.; Nieuwenhuyzenb, M.;
O’Briena, A. M. K. Tetrahedron Lett. 2003, 44, 8571.
(9) (a) Letsinger, R. L.; Lunsford, W. B. J. Am. Chem. Soc.
1976, 98, 3655. (b) McBride, L. J.; Caruthers, M. H.
Tetrahedron Lett. 1983, 24, 245. (c) Caruthers, M. H.;
Barone, A. D.; Beaucage, S. L.; Dodds, D. R.; Fisher, E. F.;
McBride, L. J.; Matteucci, M.; Stabinsky, Z.; Tang, J.-Y.
Methods Enzymol. 1987, 154, 287.
(10) Lohrmann, R.; Orgel, L. E. Tetrahedron Lett. 1978, 34, 853.
(11) Jiao, G.-S.; Kim, T. G.; Topp, M. R.; Burgess, K. Org. Lett.
2004, 6, 1701.
(12) Czernecki, S.; Veréry, J. M. Synthesis 1991, 239.
(13) Lin, T. S.; Prusoff, W. H. J. Med. Chem. 1979, 21, 109.
(14) Rossi, F. M.; Kao, J. P. Y. Bioconjugate Chem. 1997, 8, 495.
(15) Lyttle, M. H.; Carter, T. G.; Dick, D. J.; Cook, R. M. J. Org.
Chem. 2000, 65, 9033.
(16) Theisen, P.; McCollum, C.; Upadhya, K.; Jacobson, K.; Vu,
H.; Andrus, A. Tetrahedron Lett. 1992, 33, 5033.
(17) Laurent, A.; Debart, F.; Lamb, N.; Rayner, B. Bioconjugate
Chem. 1997, 8, 856.
(18) Morr, M.; Waßmann, A.; Wray, V. Tetrahedron 1999, 55,
2985.
(19) (a) Uhlmann, E.; Engels, J. Tetrahedron Lett. 1986, 27,
1023. (b) v.d. Sande, J. H.; Ramsing, N. B.; Germann, M.
W.; Elhorst, W.; Kalisch, B. W.; v. Kitzing, E.; Pon, R. T.;
Clegg, R. C.; Jovin, T. M. Science 1988, 241, 551.
(c) Koga, M.; Moore, M. F.; Beaucage, S. L. J. Org. Chem.
1991, 56, 3757. (d) Beier, M.; Stephan, A.; Hoheisel, J. D.
Helv. Chim. Acta 2001, 84, 2089.
5¢-ACGCGT-3¢NH-Fluo (10)
HPLC: MeCN gradient 0% (in 0.1 M TEAA buffer) for 5 min to
15% in 55 min; tR 36 min; yield: 61%.
MALDI-TOF MS: m/z calcd for C79H86O39N24P5 [M – H]–: 2147.5;
found: 2147.9.
UV-Vis: l (%) = 239.1 (100), 257.2 (94.7), 323.2 (15.2), 494.6 nm
(59.4); e260 (calcd): 77000 M–1 cm–1.
5¢-DPLysTCATTCTGTTCT-3¢NH-Fluo (15)
HPLC: MeCN gradient 0% (in 0.1 M TEAA buffer) for 5 min to
20% in 40 min; tR 34 min; yield: 98%.
MALDI-TOF MS: m/z calcd for C150H157O88N36P12 [M – H]–:
4240.0; found: 4246.7.
UV: l (%) = 261.5 (100), 331.4 (14.1), 494.2 nm (35.5); e260
(calcd) = 122000 M–1 cm–1.
UV-Melting Experiments
UV-melting data were acquired at 260 nm wavelength and 1 cm
path length at heating and cooling rates of 1 °C/min, using solutions
with equimolar concentrations of the DNA strands in 10 mM sodi-
um phosphate buffer (NaH2PO4, Na2HPO4), pH 7, in previously
deionized H2O and the salt concentration given in Table 1. The
melting temperatures and the thermodynamic data were determined
with the program MeltWin 3.0.36
UV/Vis and Fluorescence Data
UV/Vis wavelengths scans were acquired between 200 and 600 nm
at 20 °C of 1.97 mM solution of 10 and 2.43 mM of 15 in 10 mM so-
dium phosphate (NaH2PO4, Na2HPO4) buffer at pH 7.4. The fluo-
rescence spectra were recorded between 510 and 650 nm for the
same solutions at 20 °C using a 600 mL cuvette, an excitation wave-
length of 495 nm, and emission and excitation slits of 3 nm.
(20) Schwope, I.; Bleczinski, C. F.; Richert, C. J. Org. Chem.
1999, 64, 4749.
(21) Dombi, K. L.; Griesang, N.; Richert, C. Synthesis 2002, 816.
(22) Connors, W. H.; Narayanan, S.; Kryatova, O. P.; Richert, C.
Org. Lett. 2003, 5, 247.
(23) (a) Zielinski, W. S.; Orgel, L. E. Nucleic Acids Res. 1987,
15, 1699. (b) Orgel, L. E. Nature 1992, 358, 203.
(c) Kozlov, I. A.; Pitsch, S.; Orgel, L. E. Proc. Natl. Acad.
Sci. U.S.A. 1998, 95, 13448.
Acknowledgment
The authors wish to thank Stefanie Harnisch and Marc Westermann
for expert technical support, Siegfried Herzberger for sharing pro-
tocols for the preparation of 5, and Brian Davies for a review of part
of the manuscript. This work was supported by DFG (grant RI
1063/1-3) and Fonds der Chemischen Industrie (project 164431).
(24) (a) Kurz, M.; Göbel, K.; Hartel, C.; Göbel, M. W. Helv.
Chim. Acta 1998, 81, 1156. (b) Hey, M.; Hartel, C.; Göbel,
M. W. Helv. Chim. Acta 2003, 86, 844.
(25) Kanavarioti, A.; Stronach, M. W.; Ketner, R. J.; Hurley, T.
B. J. Org. Chem. 1995, 60, 632.
(26) Luther, A.; Brandsch, R.; von Kiedrowski, G. Nature 1998,
396, 245.
(27) (a) Cramer, F.; Schaller, H.; Staab, H. A. Chem. Ber. 1961,
94, 1612. (b) Cramer, F.; Schaller, H.; Staab, H. A. Chem.
Ber. 1961, 94, 1621.
(28) Lohrmann, R.; Orgel, L. E. J. Mol. Biol. 1980, 142, 555.
(29) (a) Atherton, F. R.; Openshaw, H. T.; Todd, A. R. J. Chem.
Soc. 1945, 660. (b) Ushioda, M.; Kadokura, M.; Moriguchi,
T.; Kobori, A.; Aoyagi, M.; Seio, K.; Sekine, M. Helv. Chim.
Acta 2002, 85, 2930.
References
(1) Sanger, F.; Nicklen, S.; Coulson, A. R. Proc. Natl. Acad. Sci.
U.S.A. 1977, 74, 5463.
(2) Lee, L. G.; Connell, C. R.; Woo, S. L.; Cheng, R. D.;
McArdle, B. F.; Fuller, C. W.; Halloran, N. D.; Wilson, R.
K. Nucleic Acids Res. 1992, 20, 2471.
(3) Pirrung, M. C. Angew. Chem. Int. Ed. 2002, 41, 1276;
Angew. Chem. 2002, 114, 1326.
(4) Tyagi, S.; Kramer, F. R. Nature Biotechnol. 1996, 14, 303.
(5) Wittwer, C. T.; Herrmann, M. G.; Moss, A. A.; Rasmussen,
R. P. Biotechniques 1997, 22, 130.
(30) Bannwarth, W.; Trzeciak, A. Helv. Chim. Acta 1987, 70,
175.
(31) Wada, T.; Mochizuki, A.; Sato, Y.; Sekine, M. Tetrahedron
Lett. 1998, 39, 7123.
Synthesis 2005, No. 14, 2327–2334 © Thieme Stuttgart · New York

2334
N. Griesang et al.
PAPER
(35) Lee, L. G.; Spurgeon, S. L.; Heiner, C. R.; Benson, S. C.;
(32) Lan, T.; McLaughlin, L. W. J. Am. Chem. Soc. 2000, 122,
6512.
Rosenblum, B. B.; Menchen, S. M.; Graham, R. J.;
Constantinescu, A.; Upadhya, K. G.; Cassel, J. M. Nucleic
Acids Res. 1997, 25, 2816.
(33) (a) Tallman, K. A.; Greenberg, M. M. J. Am. Chem. Soc.
2001, 123, 5181. (b) Tronchet, J. M. J.; Kovacs, I.; Dilda, P.;
Seman, M.; Andrei, G.; Snoeck, R.; De Clercq, E.; Balzarini,
J. Nucleosides Nucleotides 2001, 20, 1927.
(34) Sarracino, D.; Richert, C. Bioorg. Med. Chem. Lett. 1996, 6,
2543.
(36) McDowell, J. A.; Turner, D. H. Biochemistry 1996, 35,
14077.
Synthesis 2005, No. 14, 2327–2334 © Thieme Stuttgart · New York