5-Fluoro-4-thiouridine phosphoramidite: New synthon for introducing photoaffinity label into oligodeoxynucleotides

Article abstract of DOI:10.1016/j.bmc.2011.08.035

The synthesis of phosphoramidite of 5-fluoro-4-thio-2′-O- methyluridine is described. An appropriate set of protecting groups was optimized including the 4-thio function introduced via 4-triazolyl as the 4-(2-cyanoethyl)thio derivative, and the t-butyldimethyl silyl for 2′ and 3′ hydroxyl protection, enabling efficient synthesis of the phosphoramidite. These protecting groups prevented unwanted side reactions during oligonucleotide synthesis. The utility of the proposed synthetic route was proven by the preparation of several oligonucleotides via automated synthesis. Photochemical experiments confirmed the utility of the synthon.

Full text of DOI:10.1016/j.bmc.2011.08.035

Bioorganic & Medicinal Chemistry 19 (2011) 6098–6106
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
5-Fluoro-4-thiouridine phosphoramidite: New synthon for introducing
photoaffinity label into oligodeoxynucleotides
Jan Milecki a, , Joanna Nowak b, Bohdan Skalski a, , Stefan Franzen c
⇑
a Adam Mickiewicz University, Faculty of Chemistry, Grunwaldzka 6, 60-780 Poznan, Poland
b Adam Mickiewicz University, Centre of Advanced Technologies, Grunwaldzka 6, 60-780 Poznan, Poland
c Department of Chemistry, North Carolina State University, Raleigh, NC 27695, United States
a r t i c l e i n f o
a b s t r a c t
The synthesis of phosphoramidite of 5-fluoro-4-thio-20-O-methyluridine is described. An appropriate set of
protecting groups was optimized including the 4-thio function introduced via 4-triazolyl as the 4-(2-cya-
noethyl)thio derivative, and the t-butyldimethyl silyl for 20 and 30 hydroxyl protection, enabling efficient
synthesis of the phosphoramidite. These protecting groups prevented unwanted side reactions during oli-
gonucleotide synthesis. The utility of the proposed synthetic route was proven by the preparation of several
oligonucleotides via automated synthesis. Photochemical experiments confirmed the utility of the synthon.
Ó 2011 Elsevier Ltd. All rights reserved.
Article history:
Received 27 April 2011
Revised 9 August 2011
Accepted 14 August 2011
Available online 22 August 2011
Keywords:
Photocrosslink
5-Fluoro-4-thiouridine
Oligonucleotides
Phosphoramidite
1. Introduction
properties.5 The relatively small steric modification of the pyrimi-
dine base shown in Figure 1 should not interfere with base pairing.
Photocrosslinking has been shown to be an extremely useful and
versatile technique for structural studies in relation to nucleic acids
and protein interactions,1 thus becoming a routine laboratory
tool.1f,1g Macromolecules can form associates due to different inter-
actions such as hydrogen bonds, charge transfer, and electrostatic
attraction. Such weakly and reversibly bound associates, when irra-
diated with light of a proper wavelength, can form covalent bonds
between reactive sites of the macromolecules. Occurrence of this
phenomenon is based on the ability of the reactive sites to absorb
the energy of the applied light and use it to form the covalent bond.
The applied wavelengthhas tobe specificfor thereacting parts of the
macromolecule and different from the absorption region of the bulk
of chromophores present in the molecule. Among the different can-
didates for the photoprobe moiety, 4-thio-5-halogenopyrimidines
appear to be especially suitable.2 Sulfur substitution causes a bath-
ochromic shift of the main absorption band3 and increases yield of
the first excited triplet state. Halogen substituents enhance the
reactivity because of their lability in the excited state.4 We have
observed2c that these beneficial effects culminate in the case of
4-thio-5-fluoro-uracil (FSU) nucleoside 1a (Fig. 1).
We hypothesize, that if this modified nucleoside (FSU) were
introduced into an oligonucleotide, and the duplex of this oligonu-
cleotide with its complementary chain were irradiated with UV
light of k >300 nm, efficient cross-links would occur that will
engage FSU and thymine residue located in close proximity to
the complementary strand. Preliminary experiments suggest that
a thymidine moiety next to a complementary adenosine unit
would be involved.5
2. Results and discussion
For more detailed studies of the photocrosslinking process on
the oligomer level, we needed simple and efficient synthetic access
to FSU phosphoramidite. Our initial studies with a ribo-FSU 1a unit
S
N
O
N
F
F
NH
NH
HO
HO
O
O
O
FSU efficiently produces a photodimer with a thymidine upon
irradiation, and the product exhibits pronounced fluorescent
O
a X = OH
b X = H
OH
X
1 a-c
OH
X
2 a-c
c X = OCH3
⇑
Corresponding author. Tel.: +48 61 829 1312; fax: +48 61 829 1505.
E-mail address: janmil@amu.edu.pl (J. Milecki).
Author of the photochemical part.
Figure 1. Structures of the modified pyrimidine nucleosides.
0968-0896/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2011.08.035

J. Milecki et al. / Bioorg. Med. Chem. 19 (2011) 6098–6106
6099
indicated that it has considerable drawbacks due to the necessity
of the 20-hydroxyl protection and low reactivity. From two other
candidates, 20-deoxy-5-fluoro-4-thio-uridine 1b and 20-O-methyl-
5-fluoro-4-thio-uridine 1c, the 20-deoxy analog was excluded due
to instability in the conditions applied during synthesis.
Fortunately the ease of methylation and selectivity was pre-
served in the case of 5-fluoro-4-(2-cyanoethylthio)uridine 7b.
The 50-O-DMTr derivative gave the desired 20-O-methylated prod-
uct 8b with excellent 81% isolated yield together with an 8% yield
of the 30-O-derivative 9b. Both isomers could be efficiently sepa-
rated by flash chromatography.
We turned our attention to 20-O-methyl-FSU nucleoside 1c. It
appeared to be the compound of choice for several reasons. Meth-
ylation of the 20-hydroxyl while masking the hydroxyl reactivity6
does not affect ability of the oligonucleotide to form duplexes with
DNA or RNA7 nor does it change considerably the conformation of
the ribose ring.8 Oligonucleotides methylated at 20 position are
resistant to base or nuclease hydrolysis9 and are much less prone
to glycosidic bond cleavage under acidic conditions.
As the starting material, we have chosen 5-fluorouridine 2a. It is
easily accessible commercially or synthetically.10 In the view of lit-
erature reports on FSU reactivity,11 the sulfur atom in the position
4 had to be protected in order to avoid undesired modification dur-
ing the oligonucleotide synthesis.
From the available protecting groups, we have tested pivaloyl-
oxymethyl12 and 2-cyanoethyl,13 which were successfully applied
in 4-thiopyrimidines and 6-thiopurines syntheses (Fig. 2). The
pivaloyloxymethyl group turned out to be difficult to introduce
into the FSU system using the reported procedure.12 Only a very
modest yield of the desired product 4a was achieved, and consid-
erable decomposition occurred.
Initial attempts to introduce 2-cyanoethyl group via the reac-
tion of 3 with 2-cyanoethyl bromide13b did not lead to the desired
product 4b.
To achieve good overall yield of the synthesis, proper arrange-
ment of protecting groups was required. After testing three sets
of protections for the starting 5-fluorouridine: 20,30,50-tri-O-acetyl,
50-O-dimethoxytrityl-20,30-di-O-phenoxyacetyl19 and 50-O-dime-
thoxytrityl-20,30-di-O-t-butyldimethylsilyl, we found the latter as
the most efficient and straightforward, giving good yield of the de-
sired product without excessive by-products formation. Removal
of the silyl protection required careful adjustment of the fluoride
applied for deprotection. A commonly used solution of TBAF in
THF20 caused almost complete removal of the cyanoethyl protec-
tion from sulfur. Milder fluorides such as TEBA–HF hydrate or
TEA–HF hydrate21 were not efficient in removing silyl groups. A
popular reagent used for deprotection of silylated oligoribonucleo-
tides, TEA–3HF complex,22 appeared as the reagent of choice,
removing both tBDMS groups within several hours without any
noticeable side reactions. The synthetic pathway to the 50-O-dime-
thoxytrityl-4-cyanoethylthio-5-fluoro-20-O-methyluridine 8b and
the phosporamidate 14 is shown in Scheme 1.
Phosphitylation of the protected nucleoside was achieved with
a known procedure23 and gave good yield of product in the form of
an approximately 1:1 mixture of diastereoisomers. The phospho-
ramidite synthon was applied for the automated synthesis of oligo-
nucleotides intended for photocrosslinking studies (for the
sequences of the oligonucleotides see Table 1).
Another way was to use 2-cyanoethane thiol14 to displace the
good leaving group15 in position 4 of the 5-fluoropyrimidine-2-
one ring. Approach through displacement of 4-methoxy group15
in several reaction conditions tested (K2CO3, DIPEA or TEA as a
base, dioxane, DMF or neat thiol as solvents, elevated temperature,
microwave irradiation). Contrary to this, 1,2,4-triazolyl leaving
group,16 in the derivative 6, was smoothly substituted by thiol in
the presence of excess DIPEA, leading to the desired 5-fluoro-4-
(2-cyanoethylthio)uridine 4b (Fig. 3).
The reactivity of the 5-fluoro-4-thiopyrimidine system, had to
be taken int account during de-blocking of the synthesized oligo-
mer and ‘soft’ amino-protecting groups were applied as a precau-
tion. The thiol-protecting 2-cyanoethyl group was removed first
by treating the oligomer with DBU base (0.4 M in anhydr acetoni-
trile) for 15 min while still attached to the solid support. Removing
the oligomer from the solid support and de-blocking of base-labile
groups followed a routine protocol for ‘soft’ blocking groups. The
optimal procedure consisted of thoroughly washing the support
from DBU with anhydrous acetonitrile and drying before deprotec-
tion with methanolic ammonia as the next step. With phenoxyace-
tyl (for adenosine) and 4-isopropylphenoxyacetyl (for guanosine)
protection, a 3 h treatment with methanolic ammonia at room
temperature was optimal (sufficient for deprotection and removal
from the solid support without undue sulfur loss). After evapora-
tion of the ammonia solution, the residue was dissolved in 0.1 M
ammonium acetate and freed from low molecular weight contam-
inants by gel filtration. Final purification was achieved on HPLC
(see Fig. 5). As a convenient test for identification of the appropri-
ate HPLC peak and approximate verification of the content of the
Introduction of the 20-O-methyl group posed another problem.
Fluorouridine 2c methylated at 20 position is commercially avail-
able, but is also prohibitively expensive. Direct methylation of
5-fluorouridine can lead to formation of undesired 3-methyl
derivative17 in considerable amounts. Having possession of the
4-O-methyl 5-fluorouridine, we tried the methylation of its 50-O-
dimethoxytrityl derivative 7a with the known diazomethane–
SnCl2 procedure18 (Fig. 4). The methylation proceeded easily and
with rather strong selectivity towards the desired 20-O-methyl
nucleoside product 8a (8a:9a isomer ratio equal to 87:13 by 1H
NMR). Unfortunately, as was explained earlier, this good result
could not be exploited further due to the resistance by the 4-meth-
oxy group to substitution by a sulfur nucleophile.
Figure 2. Reagents and conditions: (i) Pivalyloxy methyl chloride, K2CO3; (ii) 2-cyanobromoethane, K2CO3; (iii) 2-cyanoethanethiol, K2CO3 or DIPEA, or TEA, solvents: DMF,
THF, neat thiol, w. or w/o MW.

6100
J. Milecki et al. / Bioorg. Med. Chem. 19 (2011) 6098–6106
N
NC
NC
N
S
N
N
N
S
F
F
F
N
N
N
HO
AcO
AcO
O
O
N
O
ii
i
O
O
O
4c
4b
Figure 3. Reagents and conditions: (i) 2-cyanoethanethiol, DIPEA, CH3CN; (ii) 1.5% aq ammonia.
6
OH OH
OAc OAc
OAc OAc
R
N
R
N
R
N
F
F
F
N
N
N
DMTrO
DMTrO
DMTrO
O
O
O
i
O
O
O
+
a R = OCH3
b R = SCH2CH2CN
OH OCH3
8a,b
OCH3 OH
9a,b
OH OH
7a,b
Figure 4. Reagents and conditions: (i) diazomethane (Et2O solution), SnCl2 hydrate, CH2Cl2.
N
N
N
O
N
O
N
F
F
F
N
NH
NH
DMTrO
DMTrO
DMTrO
N
O
O
O
O
iii
i
ii
O
O
TBDMSO
OTBDMS
OH OH
10
TBDMSO
OTBDMS
11
12
NC
NC
NC
S
S
S
N
F
F
F
N
N
N
DMTrO
DMTrO
DMTrO
N
O
iv
O
O
N
O
O
v
O
TBDMSO
OTBDMS
OH OH
7b
NC
OH OCH3
8b
13
S
F
N
DMTrO
N
vi
N
O
O
O
OCH3
14
P
O
CN
Scheme 1. Reagents and conditions: (i) TBDMSCl, imidazole, DMF; (ii) POCl3, TEA, 1,2,4-triazole; (iii) CH3CN, NC(CH2)2SH, DIPEA; (iv) TEA–3HF; (v) CH2N2, SnCl2ꢀ2H2O,
CH2Cl2; separation of isomers; (vi) DIPEA, NC(CH2)2OP(Cl) N(iPr)2, THF.

J. Milecki et al. / Bioorg. Med. Chem. 19 (2011) 6098–6106
6101
Table 1
Oligonucleotide
Sequence (X = 5-fluoro-4-thio-20-O-methyl-U), MALDI peaks
OD after gel OD after purification
filtration
and removal of TEAA
BS-3
BS-4
BS-5
BS-6
50-ATA GCA XAG C-30; 3085 (M+H+, calcd for C98H123FN40O56P9S+ 3085.5339), 3107 (M+Na+), 3123 (M+K+)
50-CGA TAC GAX A-30; 3085 (M+H+, calcd for C98H123FN40O56P9S+ 3085.5339), 3107 (M+Na+), 3123 (M+K+)
50-AXA TGC ATA T-30; 3075 (M+H+, calcd for C99H125FN36O58P9S+ 3075.5271), 3097 (M+Na+), 3113 (M+K+)
89
84
119
62
53
41
56
50-AXA GCA TAG C-30; 3085 (M+H+, calcd. for C98H123FN40O56P9S+ 3085.5339), 3107 (M+Na+), 3123 (M+K+) 111
4-S modification, absorbance ratio A260/A340 was applied. A mean
value of 5.11:1 was observed (Fig. 6).
1.0
0.5
0.0
BS3
BS4
BS5
BS6
3. Photocrosslinking
To examine the feasibility of photocrosslinking, the photoreac-
tion of BS-4 with its complementary strand ODN 1 was performed.
According to the earlier results,5 the label X can form interstrand
crosslink either with thymidine at 50 side:
BS-4
5’ C G A T A C G A X A 3’
ODN 1
or at 3’ side
BS-4
3’ T G C T A T G C T A T T 5’
220
260
300
[nm]
340
380
5’ C G A T A C G A X A 3’
λ
ODN 1
3’ T G C T A T G C T A T T 5’
Figure 6. Normalized UV spectra of oligodeoxynucleotides BS3–BS6. Mean
A260/A340 = 5.11:1. (Theoretical A260/A340, calculated from the sum of molar coeffi-
cients of individual nucleosides, without adjusting for hipochromicity was 5.32:1.)
or both crosslinks can be formed, each of them in two possible dia-
stereomeric forms.
Duplex BS-4—ODN 1 (A260 ꢁ 2.3 in 0.1 M phosphate buffer (pH
7)) was irradiated using argon–ion laser emitting the light at
351 nm. To be sure that all molecules are in duplex form small ex-
cess of complementary strand ODN 1 was added (1.2 equiv) and
reaction was carried out at 15 °C (the estimated melting tempera-
ture of duplex was above 30 °C). The course of the reaction was fol-
lowed with UV spectrophotometry. During the reaction, along with
the formation of photocrosslink, the pronounced rise of fluores-
cence was observed (Fig. 7). HPLC analysis (Fig. 8) showed that
BS-4 and ODN 1 were consumed after 60 s of irradiation. The reac-
tion gave two fluorescent photoproducts (ODN 2 and ODN 3), with
one of them (ODN 3) dominating (it was formed with 90% yield).
MALDI-TOF MS indicated that both isolated photoproducts have
the same molecular mass which corresponds to calculated mass
of single intermolecular crosslink [calcd 6693 for (M+H)+, found
6694.75] and is equal to the sum of mass of BS-4 and ODN 1 minus
HF (eliminated during crosslink formation). In the absorption and
fluorescence spectra (Fig. 9) of photoproducts the shifts of maxi-
mum bands of fluorescence and absorption are noticeable (ODN
2: A kmax = 374 nm, F kmax = 462 nm; ODN 3: A kmax = 370 nm, F
kmax = 457 nm). The same properties had been observed on nucle-
otide level.5 It prompted us to the suggestion that photoproducts
ODN 2 and ODN 3 are diastereoisomers.
Enzymatic digestion of the isolated crosslink ODN 3 revealed the
formation of appropriate amounts dC, dG, T, dA and dI as a result of
deamination of dA (Fig. 10). Fluorescent remaining fragment
(dAMP-crosslink) was assumed to be a crosslink core bound to aden-
osine monophosphate (Fig. 11) on the basis of MALDI-TOF MS [calcd
828.71 for (M+H)+, found 828.96]. The same mixture was obtained in
case of enzymatic digestion of ODN 2 with formation of dAMP-cross-
link fragment in second diastereomeric form, respectively.
4. Conclusions
Different approaches for the synthesis of potential useful photo-
affinity probe, phosphoramidite of 5-fluoro-4-thio-20-O-methyluri-
dine showed that the system possesses unique reactivity of the
pyrimidine ring, easily leading to decomposition or poor synthetic
yield. Adjustment of synthetic strategy required proper choice of
protecting groups and the optimal arrangement was found to be
50-O-DMTr, 20,30-di-O-TBDMS and 4-S-(2-cyanoethyl). When apply-
ing this set of protections, phosphoramidite synthon was synthe-
sized with good yield and purity and was applied in the
oligonucleotide synthesis. Optimizing the deprotection and purifi-
cation procedure of the product oligonucleotide showed that when
using ‘soft’ exoamine function protection and deprotection with
methanolic ammonia at room temperature, nucleophilic substitu-
tion of sulphur atom does not occur and the 5-fluoro-4-thiouracil
system is preserved.
200
100
0
1.0
0.5
0.0
2.5 min
220 260 300 340 380
λ [nm]
260 nm
340 nm
0.0
1.5
3.0
4.5
time [min]
6.0
7.5
When FSU-containing oligonucleotides were irradiated as a du-
plex with their complementary strand, efficient formation of cross-
link with the thymidine next to the opposite adenosine occurred.
Properties of the photoproduct followed the properties of
analogous product obtained on the nucleoside level and prelimin-
Figure 5. Example HPLC trace for the BS3 oligomer. Separation conditions: Sample
injected: 0.5 l of oligodeoxynucleotide in 1 mL H2O/Acetonitrile, 95/5, v/v; Waters
XBridge OST C18 Column, 2.5
B = 0.1 M TEAA/Acetonitrile, 50/50, v/v, Flow rate: 1 mL/min, column temp = 40 °C,
gradient: 15–25%B in 5 min, then to 50%B in 5 min.
l
l
m, 4.6 ꢂ 50 mm, mobile phases: A = 0.1 M TEAA,

6102
J. Milecki et al. / Bioorg. Med. Chem. 19 (2011) 6098–6106
20
200
100
0
ODN 3
ODN 3
ODN 1
10
0
BS-4
ODN 2
ODN 2
0
2
4
6
8
10
12
0
2
4
6
8
10
12
retention time [min]
retention time [min]
Figure 7. Upper part—HPLC absorption elution profile for irradiation of duplex BS4-ODN 1. Dotted line refers to mixture before irradiation and solid line to mixture after 60 s
of irradiation to ca. 98% conversion of BS-4. Lower part—HPLC fluorescence elution profile after irradiation of duplex recorded at 460 nm (kEx = 370 nm). Separation
conditions: Waters XBridge OST C18 Column, 2.5
temp = 45 °C, gradient: 13–20%B in 10 min.
lm, 4.6 ꢂ 50 mm, mobile phases: A = 0.1 M TEAA, B = 0.1 M TEAA/Acetonitrile, 50/50, v/v, flow rate: 1 ml/min, column
6
1.00
0.75
0.50
0.25
0.00
120
irradiation time [s]:
dG
4
2
0
0
60
T
80
40
0
dC dI
250
300
350
λ [nm]
400
450
dAMP-crosslink
dA
400
450
λ [nm]
500
550
0.0
2.5
5.0
7.5 10.0 12.5 15.0 17.5 20.0
retention time [min]
Figure 8. Changes in fluorescence spectra of duplex BS-4-ODN 1 during irradiation.
Figure 10. HPLC absorption elution profile for ODN 3 after enzymatic digestion.
Insert: UV spectrum of dAMP-crosslink. Separation conditions: Agilent Poroshell
120 EC-C18 Column, 2.7
lm, 4.6 ꢂ 150 mm, mobile phases: A = 0.1 M CH3COONH4/
Acetonitrile, 95/5, B = 0.1 M CH3COONH4/Acetonitrile, 50/50, v/v, flow rate: 0.7 ml/
min, column temp = 40 °C, gradient: 0–3%B in 5 min, then to 15%B in 5 min and to
100%B in 5 min.
1.00
1.0
50.0
ODN 2
ODN 3
0.5
37.5
25.0
12.5
0.0
0.75
0.50
0.25
0.00
0.0
300 325 350 375 400 425 450
chemical shifts are reported in parts per million relative to TMS
(1H and 13C spectra), CF3COOH (19F) or 85% H3PO4 31P).
(
High-performance liquid chromatography (HPLC) was per-
formed with an Agilent 1200 system with a binary gradient-form-
ing module and diode-array UV–Vis detector. The preparative
separation was carried out using the same conditions (specified
at Fig. 5).
The MS analyses were performed using the MALDI-TOF MS
instrument model Autoflex II equipped with a reflectron (resolution
about 5000 at m/z 1000), on a MALDI metal target plate (Bruker, Bre-
men, Germany). The instrumentwas equipped with a SmartBeam la-
ser and operated under FlexControl. Spectra were calibrated in
FlexAnalysis using the Protein Calibration Standard I from Bruker.
Steady-state irradiation and UV–vis and fluorescence spectra:
UV–vis spectra were measured at room temperature using a Cary
300 Bio Varian spectrophotometer. Fluorescence emission spectra
were measured at room temperature using a Luminescence Spec-
trometer LS 50B Perkin Elmer.
225 275 325 375 425 475 525 575
λ [nm]
Figure 9. Normalized absorption and fluorescence emission spectra of photoprod-
ucts ODN 2 and ODN 3.
ary analytical data confirm formation of two diastereoisomes, with
one of them dominating.
Steady-state photochemical irradiation experiments were car-
ried out in 1 cm ꢂ 1 cm rectangular fluorescence cell with stirring
bar on a standard optical bench system equipped with an argon ion
laser (Coherent INNOVA 400, equipped with a special UV grade
tube and UV resonator optics). A double Pellin–Broca prismline
separator was used to select narrow band irradiation wavelengths
5. Experimental
5.1. Instruments
1H, 13C, 19F and 31P NMR spectra were recorded on Varian
300 MHz Mercury system in deuterated solvents as indicated. All

J. Milecki et al. / Bioorg. Med. Chem. 19 (2011) 6098–6106
6103
O
OCH3
O
N
CH3
N
OH
OH
O
N
HN
O
N
N
NH2
O
S
O
N
O
O
N
HO
P
OH
O
OH
Figure 11. Tentative structure of fragment (dAMP-crosslink) containing photocrosslink produced on enzymatic digestion of ODN 2 and ODN 3.
(351.1 and 351.4 nm) with approximately 100 mW optical power,
as measured by a Newport 1918-C laser-power meter with an
818P-010-12 thermopile head. The beam profile was expanded
(5ꢂ) in a home-built beam expander. The approximately circular
cross-sectional beam with 1 cm diameter was directed to the tem-
perature-controlled cell holder (Quantum Northwest, model TC
125), placed on a micrometer one-axis translation stage.
A small portion (<1%) of the laser beam was reflected by a beam
splitter to the monitoring CCD spectrometer (Edmund Optics, mod-
el BRC111 A-USB) to monitor the wavelength.
5.2.2. 20,30,50-Tri-O-acetyl-4-(2-cyanoetylthio-)-5-fluorouridine
4b
Compound 6 (1.9 g crude), was dissolved in anhyd. acetonitrile
(10 mL), 2-cyanoethanothiol (2 mL, 2 g, 23 mmol) and N,N-diiso-
propylethylamine (DIPEA) (4 mL, 2.97 g, 22.96 mmol) were added.
After 3 h TLC (solvent B) showed no starting material. Solvents
were evaporated, residue dissolved in methylene chloride, succes-
sively washed with 0.1 M KH2PO4, water and evaporated. Chroma-
tography (gradient of methanol in dichloromethane from 0% to 2%)
gave compound 4b (2.3 g) as yellow foam. NMR showed that it
contained free mercaptane, and after repeated chromatography
1.7 g (3.71 mmol) of pure 4b was obtained. TLC: Rf = 0.43 (B).
1H NMR (CDCl3): 7.78 (d, J = 4.5 Hz, 1H); 6.05 (dd, J = 1.3 Hz,
J = 3.6 Hz, 1H); 5.39 (dd, J = 3.6 Hz, J = 5.2 Hz, 1H); 5.29 (t,
J = 5.6 Hz, 1H); 4.46–4.39 (m, 3H); 3.47 (t, J = 6.6 Hz, 2H); 2.80 (t,
J = 6.6 Hz, 2H); 2.17; 2.14; 2.10 (3 ꢂ s, 3 ꢂ 3H).
5.2. Materials
Kieselgel 60H (MERCK) was used for flash chromatography.
Acetonitrile for HPLC (J. T. Baker) was used as received. Pyridine
was dried by successive distillation from KOH and CaH2, methylene
chloride was distilled from K2CO3. THF was distilled from CaH2 di-
rectly before use. Triethylamine and DIPEA were distilled from
CaH2 and stored over 4 Å molecular sieves. Water used in the prep-
aration of aqueous solutions for HPLC solutions was distilled and
purified using a Millipore Simpak1 system.
Eluting solvents used for TLC were: chloroform–ethyl acetate
7:3 (A), methylene chloride–methanol 19:1 (B), methylene chlo-
ride–methanol 9:1 (C), hexane–ethyl acetate–triethylamine
50:40:3 (D). TLC was performed on aluminum-backed SiO2 plates
(HF254 Merck 5554) cut to a size 3 ꢂ 10 cm.
19F NMR (CDCl3): ꢃ155.91 (d, JH-F = 4.5 Hz). 13C NMR (CDCl3) see
Table 2.
Elemental Anal. Calcd for C18H20FN3O8S: C, 47.26; H, 4.41; N,
9.19; S, 7.01. Found: C, 47.11; H, 4.63; N, 9.23; S, 7.20.
5.2.3. (2-Cyanoetylthio-)-5-fluorouridine 4c
Compound 4b (1.5 g, 3.28 mmol) was dissolved in dimethoxy-
ethane–water 2:1 (20 mL), containing 1.5% ammonia. After 1.5 h at
room temp the solvents were evaporated, the residue was dissolved
in 1 mL of methanol, 7 mL of dichloromethane was added and the
resulting emulsion was applied onto a chromatography column.
Elution with gradient of methanol in dichloromethane from 0% to
6% gave 400 mg of compound 4c. TLC Rf = 0.27 (solvent C).
1H NMR (DMSO-d6): 8.70 (d, J = 5 Hz, 1H); 5.65 (m, 2H); 5.39
(m, 1H); 5.05 (m, 1H); 3.95 (m, 3H); 3.79 (dt, J = 2.3 Hz, J = 10 Hz
1H); 3.64 (dt, J = 2.2 Hz, J = 10 Hz, 1H); 3.44 (t, J = 6.6 Hz); 2.98 (t,
J = 6.6 Hz, 2H).
5.2.1. 20,30,50-Tri-O-acetyl-5-fluoro-4-(1,2,4-triazol-1-yl)
uridine 6
1,2,4-Triazole (3.7 g, 53.62 mmol) was placed in 50 mL flask,
dried in vacuo for 3 h, dissolved in anhyd acetonitrile (17 mL)
and triethylamine (7.2 mL, 5.22 g, 51.6 mmol) was added. The flask
was closed with a rubber septum, solution was cooled to 0 °C and
phosphorus oxychloride (1.2 mL, 1.97 g, 12.87 mmol) was added
with syringe. After 10 min stirring at 0 °C, solution of 2,3,5-tri-O-
acetyl-5-fluorouridine (1.7 g, 4.38 mmol) in pyridine–acetonitrile
(2:1, 35 mL) was added during 5 min. The mixture was allowed
to reach room temperature and was left stirring for 20 h. Water
(0.34 mL) was added to the dark solution and after 30 min the sol-
vents were evaporated, residue dissolved in dichloromethane
(40 mL), washed with satd NaHCO3, water, dried with MgSO4 and
evaporated. The crude product (1.9 g) showed one spot on TLC
(SiO2, solvent A, Rf = 0.12, blue fluorescence) and was essentially
pure by NMR spectrum. It was used for the next step without
purification.
19F NMR (DMSO-d6): ꢃ158.24 (d, JH-F = 5.3 Hz). 13C NMR (CDCl3)
see Table 2.
Elemental Anal. Calcd for C12H14FN3O5S: C, 43.50; H, 4.26; N,
12.68; S, 9.62. Found: C, 43.21; H, 4.53; N, 12.87; S, 9.50.
5.2.4. 50-O-Dimethoxytrityl-4-(2-cyanoetylthio-)-5-
fluorouridine 7b
From 4c. Compound 4c (400 mg, 1.03 mmol) was coevaporated
with pyridine three times and dissolved in pyridine (12 mL). Dime-
thoxytrityl chloride (375 mg, 1.13 mmol) was added and the mix-
ture left for 5 h. Pyridine was evaporated to thick oil, which was
dissolved in dichloromethane (25 mL), washed with KH2PO4, dried
over MgSO4 and solvent was evaporated, leaving yellow oil
(770 mg). Separation on silicagel (gradient of methanol in dichlo-
romethane 0–2%) gave product 7b as a foam (720 mg,). TLC
Rf = 0.67 (solvent B).
1H NMR (CDCl3): 9.27 (s, 1H); 8.43 (d, J = 6.5 Hz, 1H), 8.23 (s,
1H), 6.14 (d, J = 2.4 Hz, 1H), 5.47 (dd J = 2.4 Hz, J = 5.5 Hz, 1H),
5.31 (d, J = 5.5 Hz, 1H), 4.52 (m, 1H), 4.45 (m, 2H), 2.20, 2.16,
2.11 (3 ꢂ s, 3 ꢂ 3H).
From 12. Compound 12 (420 mg, 0.49 mmol) was dissolved in
neat triethylamine–3HF complex (0.6 mL). After 8 h methylene
19F NMR (CDCl3): ꢃ157.19 (d, JH-F = 6.5 Hz).

6104
J. Milecki et al. / Bioorg. Med. Chem. 19 (2011) 6098–6106
Table 2
13C NMR shifts (ppm from TMS) and 13C–19F coupling constants (Hz) for new compounds
Carbon atom
4b*
4c
7b
8b
9b
11
13
C-2
C-4
152.32
167.60
152.88
167.48
153.38
167.64
151.55
167.68
152.32
167.50
148,52
156.57
151.6742
167.27
(d, JC-F = 18 Hz)
143.38
(d, JC-F = 241 Hz)
135.02
(d, JC-F = 18 Hz)
143.51
(d, JC-F = 242 Hz)
135.12
(d, JC-F = 18 Hz)
143.60
(d, JC-F = 243 Hz)
134.94
(d, JC-F = 18 Hz)
143.30
(d, JC-F = 243 Hz)
135.22
(d, JC-F = 18 Hz)
143.34
(d, JC-F = 243 Hz)
135.10
(d, JC-F = 27 Hz)
140.40
(d, JC-F = 238 Hz)
124.22
(d, JC-F = 18 Hz)
143.16
(d, JC-F = 248 Hz)
125.89
C-5
C-6
(d, JC-F = 25 Hz)
—
(d, JC-F = 26 Hz)
—
(d, JC-F=27 Hz)
158.64
(d, JC-F = 26 Hz)
158.60
(d, JC-F = 25 Hz)
158.68
(d, JC-F = 31 Hz)
158.74
(d, JC-F = 33 Hz)
158.66
Arom.DMTr
(i-OCH3)
Arom. DMTr
Arom. DMTr
Arom. DMTr
Arom. DMTr
Arom. DMTr
Arom. DMTr
Arom. DMTr
(o- OCH3)
Ar3C-OCH250
SCH2CH2CN
SCH2CH2CN
SCH2CH2CN
C10
—
—
—
—
—
—
—
—
—
—
—
—
—
—
143.97
129.80
127.80
127.13
126.38
125.93
113.59
144.21
129.98
128.02
127.93
127.04
126.06
113.32
144.07
129.90
129.10
128.05
127.82
127.74
113.32
143.97
134.98
130.08
129.11
128.02
127.18
113,30
143.78
134.90
130.12
128.18
127.80
127.06
113.31
—
—
93.85
117.69
24.63
18.09
87.20
77.28
72.50
86.68
62.98
—
88.42
117.73
24.62
18.07
84.41
67.76
83.36
83.57
60.64
58.93
55.24
—
92.05
117.77
24.57
18.10
87.14
79.08
74.97
82.28
62.09
58.54
55.26
—
88.96
—
—
—
87.39
75.94
71.75
84.17
62.40
—
55.24
25.72; 25.79
17.91; 17.95
ꢃ4.25; ꢃ4.62;
ꢃ4.77; ꢃ4.89
87.02
117.72
24.43
17.98
91.76
75.77
69.60
81.95
60.95
—
55.12
25.63; 25.77
17.75; 17.93
ꢃ3.94; ꢃ4.18;
ꢃ5.19; ꢃ5.32
117.70
24.61
18.09
88.19
73.63
69.62
79.84
62.63
—
—
—
—
—
117.70
24.60
18.08
90.00
74.40
59.36
84.31
59.36
—
—
—
—
—
C20
C30
C40
C50
20(30)-OCH3
DMTr OCH3
Si-C-(CH3)3
Si-C-(CH3)3
Si-(CH3)2
55.22
—
—
—
—
—
—
—
*
Acetyl C@O: 170.02; 169.53; 169.45; acetyl CH3: 20.72; 20.48; 20.49.
chloride was added (10 mL) and the solution was washed with satd
NaHCO3, dried with MgSO4 and evaporated, giving oily residue
(500 mg). Separation on silicagel (gradient of methanol in dichloro-
methane 0–2%) gave product 7b as a foam (270 mg,). TLC Rf = 0.67
(solvent B).
J = 2.2 Hz, 2H); 3.44 (t, J = 6.5 Hz, 2H); 2.91 (t, J = 6.6 Hz, 2H);
2.53 (d, J = 9.7 Hz, exchangeable with D2O).
19F NMR (CDCl3): ꢃ155.63 (d, J = 4.2 Hz). 13C NMR (CDCl3) see
Table 2.
Elemental Anal. Calcd for C34H34FN3O7S: C, 63.05; H, 5.29; N,
6.49; S, 4.95. Found: C, 63.44; H, 5.35; N, 6.31; S, 4.82.
Compound 9b: 1H NMR (CDCl3): 8.06 (d, J = 4.2 Hz, 1H); 7.37–
7.24 (m, 9H); 6.83 (m, 4H); 5.85 (d, J = 2.5 Hz, 1H); 4.45 (m, 1H);
4.30 (m, 1H); 4.01 (t, J = 5.2 Hz, 1H); 3.80 (s, 3H);); 3.55 (d,
J = 3.5 Hz, 1H); 3.51–3.40 (m, 5H, overlapped with s, 3H at 3.47);
2.90 (t, m, 2H).
1H NMR (CDCl3): 7.88 (d, J = 4.6 Hz), 7.25 (m, 9H); 6.80 (m, 4H);
5.72 (d, J = 4.6 Hz, 1H); 5.35 (s, 1H); 4.57 (t, J = 4.6 Hz,1H); 4.43–
4.38 (m, 2 H); 3.80 (s, 3H); 3.50 (m, 2H); 3.31 (m, 3H); 2.92 (t,
J = 6.6 Hz, 2H).
19F NMR (CDCl3): 155.41 (d, JH-F = 4.6 Hz). 13C NMR (CDCl3) see
Table 2.
Elemental Anal. Calcd for C33H32FN3O7S: C, 62.55; H, 5.09; N,
6.63; S, 5.06. Found: C, 62.36; H, 5.29; N, 6.87; S, 5.01.
19F NMR (CDCl3): ꢃ155.63 (d, JH-F = 4.2 Hz). 13C NMR (CDCl3) see
Table 2.
Elemental Anal. Calcd for C34H34FN3O7S: C, 63.05; H, 5.29; N,
6.49; S, 4.95. Found: C, 63.43; H, 5.47; N, 6.30; S, 4.86.
5.2.5. 50-O-Dimethoxytrityl-20-O-methyl-4-(2-cyanoetylthio-)-5-
fluorouridine 8b and 50-O-dimethoxytrityl-30-O-methyl-4-(2-
cyanoetylthio-)-5-fluorouridine 9b
5.2.6. 50-O-Dimethoxytrityl-20,30-di-O-t-butyldimethylsilyl-5-
fluorouridine 11
Compound 7b (592 mg, 0.93 mmol) was dissolved in anhydrous
methylene chloride (100 mL). The solution was cooled in ice and
SnCl2 dihydrate (68 mg, 0.3 mmol) was added. To the stirred solu-
tion diazomethane in diethyl ether (approx 0.5 M) was added in
aliquots of 2 mL in 15 min intervals. The progress of the reaction
was monitored by TLC. When no starting material remained (6
additions) the solution was successively washed with satd NaH-
CO3, water, dried over MgSO4 and evaporated, giving 610 mg of
yellow oil. Separation on a SiO2 column (gradient of ethanol in
dichloromethane 0–2%) gave pure product 8b (490 mg) and 9b
(48 mg). Analytical sample was dissolved in benzene, frozen and
freeze-dried to give solvent free fluffy solid.
50-O-Dimethoxytrityl-5-fluorouridine (700 mg, 1.24 mmol) and
imidazole (700 mg, 10.28 mmol) were placed in flask and dried
in vacuo (30 °C, 1 torr) for 2 h. Anhydrous DMF (2 mL) and t-butyl-
dimethylsilyl chloride (500 mg, 3.32 mmol) were added and the
reaction mixture was left overnight at rt. TLC showed complete
disappearance of starting nucleoside. Satd NaHCO3 (30 mL) was
added and the mixture was extracted with ethyl acetate
(2 ꢂ 10 mL) and combined organic layers were washed with H2O,
dried with MgSO4 and evaporated. Chromatography on SiO2 (gradi-
ent of methanol in dichloromethane (0–3%) gave pure disililated
product (580 mg, 0.73 mmol, 58%) followed by mixture of 20,30-
O-monosilylated isomers (300 mg, 0.44 mmol, 36%).
Compound 8b 1H NMR (CDCl3): 8.26 (d, J = 4.2 Hz, 1H); 7.42–
7.24 (m, 9H); 6.80 (m, 4H); 5.84 (s, 1H); 4.46 (m, 1H); 4.03 (m,
1H); 3.86 (d, J = 5 Hz, 1H); 3.80 (s, 3H); 3.75 (s, 3H); 3.56 (d,
1H NMR (CDCl3): 8.05 (d, J = 6.2 Hz, 1H) 7.40–7.24 (m, 9H); 6.80
(m, 4H); 5.90 (d, J = 4.6 Hz, 1H); 4.24 (t, J = 4.1 Hz, 1H); 4.09 (m,

J. Milecki et al. / Bioorg. Med. Chem. 19 (2011) 6098–6106
6105
Table 3
Enzymatic digestion of oligonucleotides
Oligonucleotide
BS3
Nucleoside
Area (HPLC)
e
ꢂ 10ꢃ3
9.1
7.4
Area/
e
ꢂ 10ꢃ3
Experimental nucleoside count**
Theoretical nucleoside count
dC
dI
dA
dG
T
237.7
306.9
373.4
334.5
156.6
378.7
0.026
0.041
0.025
0.025
0.016
0.018
1.62
2.57
1.55
1.52
1.00
1.12
2
4
14.9
13.6
9.7
2
1
1
X
20.92*
BS4
BS5
BS6
dC
dI
dA
dG
T
277.3
201.8
554.5
361.2
152.8
375.0
9.1
7.4
0.030
0.027
0.037
0.027
0.016
0.018
1.93
1.73
2.36
1.69
1.00
1.14
2
4
14.9
13.6
9.7
2
1
1
X
20.92*
dC
dI
dA
dG
T
116.1
336.8
354.6
166.2
521.4
452.2
9.1
7.4
0.013
0.046
0.024
0.012
0.054
0.022
0.71
2.54
1.32
0.68
3.00
1.20
1
4
14.9
13.6
9.7
1
3
1
X
20.92*
dC
dI
dA
dG
T
354.3
252.0
764.2
483.0
203.5
467.3
9.1
7.4
0.039
0.034
0.051
0.036
0.021
0.022
1.86
1.62
2.44
1.69
1.00
1.06
2
4
14.9
13.6
9.7
2
1
1
X
20.92*
*
e
of 5-fluoro-4-thiouridine was assumed for 5-fluoro-4-thio-20-O-methyluridine (X).
was determined relative to signal of thymidine.
**
2H); 3.79 (s, 6H); 3.53 (d, J = 8.8 Hz, 1H); 3.33 (d, J = 8.8 Hz, 1H);
0.89 (s, 9H); 0.81 (s, 9H); 0.09 (s, 3H); 0.08 (s, 3H); 0.04 (s, 3H);
ꢃ0.04 (s, 3H).
(hood, stench), residue was dissolved in methylene chloride,
washed with 0.5 M KH2PO4 solution and evaporated. Chromatogra-
phy (SiO2, methylene chloride as eluent) gave pure 13 as cream-
white foam (425 mg).
19F NMR (CDCl3): ꢃ155.56 (d, JH-F = 4.7 Hz). 13C NMR (CDCl3) see
Table 2.
1H NMR (CDCl3): 8.43 (d, J = 4.7 Hz, 1H); 7.40–7.24 (m, 9H);
6.80 (m, 4H); 5.66 (s, 1H); 4.28–4.24 (m, 2H); 4.12 (dd, J = 3.8 Hz,
J = 8.8 Hz, 1H); 3.80 (s overlapped with m, 6H+1H); 3.50–3.30 (m,
3H); 2.96–2.83 (m, 2H); 0.91 (s, 9H); 0.73 (s, 9H); 0.28 (s, 3H);
0.15 (s, 3H); 0.01 (s, 3H); ꢃ0.07 (s, 3H).
5.2.7. 50-O-Dimethoxytrityl-20,30-di-O-t-butyldimethylsilyl-5-
fluoro-4-(1,2,4-triazol-1-yl) uridine 12
1,2,4-Triazole (1.3 g, 18.81 mmol) was dried in vacuo (30 °C,
1 torr) for 3 h. Dry acetonitrile (6 mL) was added, followed by tri-
ethylamine (2.7 mL, 1.96 g, 19.36 mmol), the flask was closed with
septum, cooled in ice and phosphoryl chloride (0.5 mL, 822 mg,
5.36 mmol) was added. White suspension was stirred at 0 °C for
10 min and solution of compound 11 (570 mg, 0.718 mmol) in
1:2 mixture of acetonitrile and pyridine (10 mL) was added. The
mixture was stirred at rt overnight. TLC (hexane–ethyl acetate
30:1) showed complete reaction. Water (0.5 mL) was added, after
30 min solvents were evaporated, residue treated with satd NaH-
CO3 (20 mL) and extracted with ethyl acetate (2 ꢂ 10 mL). Organic
phases were combined, dried with MgSO4, and evaporated, leaving
essentially pure 12 (TLC) as white foam (698 mg), which was used
without purification.
19F NMR (CDCl3): ꢃ155.88 (d, JH-F = 4.7 Hz). 13C NMR (CDCl3) see
Table 2.
Elemental Anal. Calcd for C45H60FN3O7SSi2: C, 62.69; H, 7.01; N,
4.87; S, 3.72. Found: C, 62.56; H, 7.29; N, 4.87; S, 3.60.
5.2.9. 50-O-Dimethoxytrityl-20-O-methyl-30-O-[(2-cyanoethoxy),
(N,N-diisopropylamino)phosphino]-4-(2-cyanoetylthio-)-5-
fluorouridine 14
Nucleoside 8b (480 mg, 0.74 mmol) was placed in round-bot-
tomed flask (50 mL) and dried in vacuo (0.5 mmHg) for 3 h at room
temp. Argon was let to the flask, which was closed with septum.
Anhydrous THF (5 mL, freshly distilled from CaH2) was added, fol-
1H NMR (CDCl3): 9.22 (s, 1H); 8.86 (d, J = 6.3 Hz, 1H), 8.20 (s,
1H), 7.40–7.24 (m, 9H); 6.80 (m, 4H); 5.79 (s, 1H), 4.33 (dd
J = 1.1 Hz, J = 3.8 Hz, 1H), 4.30 (t, J = 2.3 Hz, 1H); 4.14 (dd
J = 12.0 Hz, J = 3.8 Hz, 1H); 3.80 (s, 6H); 3.77 (dd, J = 11.1 Hz,
J = 2.6 Hz, 1H); 3.40 (dd, J = 11.1 Hz, J = 2.6 Hz 1H), 0.92 (s, 9H);
0.76 (s, 9H); 0.27 (s, 3H); 0.16 (s, 3H); 0.02 (s, 3H); ꢃ0.05 (s, 3H).
19F NMR (CDCl3): ꢃ157.21 (d, JH-F = 5.0 Hz).
lowed by DIPEA (210
lL, 155 mg, 1.20 mmol) and chlorophosphine
(230 L, 244 mg, 1.04 mmol) and the solution was left at room
l
temperature. After 3 h no starting nucleoside was visible by TLC
(solvent D). Reaction was quenched with methanol (0.5 mL), then
additional triethylamine (1 mL) and ethyl acetate (30 mL) were
added. Solution was washed with satd NaHCO3, brine, dried over
MgSO4 and solvent was evaporated, leaving 685 mg of yellow
foam. Chromatography (SiO2, cyclohexane containing 2% of trieth-
ylamine with gradient of dichloromethane (20–40%) gave product
14 (white solid after freeze drying of benzene solution). Yield
250 mg.
5.2.8. 50-O-Dimethoxytrityl-20,30-di-O-t-butyldimethylsilyl-4-(2-
cyanoetylthio-)-5-fluorouridine 13
Compound 12 (690 mg, obtained from 0.710 mmol of 11) was
dissolved in dry acetonitrile (2 mL), 2-cyanoethanethiol (0.5 mL,
500 mg, 5.73 mmol) and DIPEA (0.33 mL, 245 mg, 1.89 mmol) were
added and the mixture was left at rt. After 1.5 h TLC showed com-
plete disappearance of the substrate 12. Solvent was evaporated
31P NMR (CDCl3) 150.61, 150.02; 19F NMR (CDCl3) ꢃ156.03
(JH-F = 4.2 Hz).
Elemental Anal. Calcd for C43H51FN5O9PS: C, 59.78; H, 5.95; N,
8.19; S, 3.71. Found: C, 60.13; H, 6.06; N, 8.15; S, 3.62.

6106
J. Milecki et al. / Bioorg. Med. Chem. 19 (2011) 6098–6106
5.3. Oligonucleotide synthesis
of Bioorganic Chemistry, Polish Academy of Sciences, for his help
with MALDI spectra. This work was supported by the funds from
the Ministry of Science and Higher Education, Republic of Poland.
Syntheses were performed on ABI 391PCR Mate instrument
applying built-in 1 lmol protocol and commercial CPG supports.
Iodine solution of 0.05 M concentration was used in the oxidation
References and notes
step.
1. (a) Favre, A. In Bioorganic Photochemistry; Morrison, H., Ed.; Wiley: New York,
1990; Vol. 1, p 379; (b) Favre, A.; Dubreiul, Y. L. New J. Chem. 1991, 15, 593; (c)
Favre, A.; Fourrey, J.-L. Acc. Chem. Res. 1995, 28, 375; (d) Meisenheimer, K. M.;
Koch, T. H. Crit. Rev. Biochem. Mol. Biol. 1997, 32, 101; (e) Wang, Z.; Huq, I.; Rana,
T. M. Bioconjugate Chem. 1999, 10, 512; (f) Naryshkin, N.; Reviakin, A.; Ebright,
R. Cold Spring Harbor Protoc. 2006. doi:10.1101/pdb.prot 4588; (g) Mc Kinnon,
A. L.; Taunton, J. Curr. Protocols Chem. Biol. 2009, 55; online version
doi:10.1002/9780470559277.ch090167.
5.4. Deprotection and purification of oligonucleotides
Column containing solid support with the synthesized oligonu-
cleotide was attached to a syringe containing 2 mL of 0.4 M solu-
tion of DBU in anhydr CH3CN. Approx. 0.7 mL of the solution was
pushed into the column, replaced by a new portion after 5 min
and another one after next 5 min. After total time of 15 min the
support was washed with dry CH3CN (10 mL) and dried in vacuo.
The dried support was placed into a vial, 1 mL of approx. 22%
methanolic ammonia was added and the vial closed tightly. After
3 h at rt the solid was filtered off, washed with methanol and water
and the filtrates were evaporated. The residue was dissolved in
0.1 m ammonium acetate (1 mL) and passed through a NAP 25 col-
umn. The column was washed with the same solution and 6 frac-
tions of 1.4 mL were collected and checked by UV. The oligomer
was present mainly in the fractions 3 and 4, with some residue
in 5th fraction. Combined fractions were evaporated, dissolved in
H2O/acetonitrile, 95/5, v/v (1 mL) and purified by HPLC (two
2. (a) McGregor, A.; Rao, M. V.; Duckworth, G.; Stockley, P. G.; Connolly, B. A.
Nucleic Acids Res. 1996, 24, 3173; (b) Favre, A.; Saintome, C.; Fourrey, J.-L.;
Clivio, P.; Laguaa, P. J. Photochem. Photobiol. 1998, 42, 109; (c) Wenska, G.;
´
´
Taras-Goslinska, K.; Lamparska-Kupsik, K.; Skalski, B.; Gdaniec, M.; Gdaniec, Z.
J. Chem. Soc., Perkin Trans. I 2002, 53.
3. Lipsett, M. N. J. Biol. Chem. 1965, 240, 3975; Rubin, Y. V.; Morozov, Y.;
´
Venkatesvarlu, D.; Leszczynski, J. J. Phys. Chem. A 1998, 102, 2194.
4. Taras-Gos´lin´ ska, K.; Wenska, G.; Skalski, B.; Maciejewski, A.; Burdzin´ ski, G.;
Karolczak, K. J. Photochem. Photobiol. A: Chem. 2004, 168, 227.
5. Skalski, B.; Taras-Gos´lin´ ska, K.; Dembska, A.; Gdaniec, Z.; Franzen, S. J. Org.
Chem. 2010, 75, 621.
6. Uhlmann, E.; Peyman, A. Chem. Rev. 1990, 90, 543.
7. Inoue, H.; Imura, A.; Iwai, S.; Miura, K.; Ohtsuka, E. Nucleic Acids Res. 1987, 15,
6131.
8. Hruska, F. E.; Mak, A.; Singh, H.; Shugar, D. Can. J. Chem. 1973, 51, 1099.
9. (a) Paolella, G.; Sproat, B. S.; Lamond, A. I. EMBO J. 1992, 11, 1913; (b) Lesnik, E.
A.; Guinoso, C. J.; Kawasaki, A. M.; Sasmor, H.; Zounes, M.; Cummins, L. L.;
Ecker, D. J.; Cook, P. D.; Freier, S. M. Biochemistry 1993, 32, 7832.
10. Vorbrüggen, H.; Krolikiewicz, K.; Bennua, B. Chem. Ber. 1981, 114, 1234.
11. Wempen, I.; Miller, N.; Falco, E. A.; Fox, J. J. J. Med. Chem. 1968, 11, 144.
12. Clivio, P.; Fourrey, J.-L.; Gasche, J.; Audic, A.; Favre, A.; Perrin, C.; Woisard, A.
Tetrahedron Lett. 1992, 33, 65.
500 ll injections, Waters XBridge OST C18 Column, 2.5 lm,
10 ꢂ 50 mm, phase A = 0.1 M TEAA, B = 0.1 M TEAA/Acetonitrile,
50/50, v/v, flow rate 1 ml/min, T = 40 °C, gradient 17–20%B in
10 min, then to 50%B in 5 min. Combined fractions were concen-
trated to approx. 0.5 mL and desalted by passage through HPLC col-
umn. Yields of the oligomers BS3–BS6 see Table 1. The oligomers
were verified by enzymatic digestion, results see Table 3. Enzy-
13. (a) Coleman, R. S.; Siedlecki, J. M. Tetrahedron Lett. 1991, 32, 3033; (b)
Christopherson, M. S.; Broom, A. D. Nucleic Acids Res. 1991, 19, 5719; (c)
Kadokura, M.; Wada, T.; Seio, K.; Sekine, M. J. Org. Chem. 2000, 65, 5104.
14. Klose, J.; Reese, C. B.; Song, Q. Tetrahedron 1997, 53, 14411.
_
15. Fischer, A.; Gdaniec, Z.; Biała, E.; Łozyn´ ski, M.; Milecki, J.; Adamiak, R. W.
matic digestion conditions: 0.2 OD of oligonucleotide in 150 lL
Nucleosides Nucleotides 1996, 15, 477.
of buffer (10 mM KH2PO4, 10 mM MgCl2, pH 7) was digested with
alkaline phosphatase bovine intestinal mucosa (27 DEA units, Sig-
ma–Aldrich, BioUltra) and phosphodiesterase I from Crotalus ada-
manteus venom (0.0055 units, Sigma–Aldrich, Purified) for 18 h
at 37 °C. The enzyme caused partial deamination of dA to dI (con-
firmed in model experiment).
16. Reese, C. B.; Ubasawa, A. Nucl. Acids Res. Symp. Ser. 1980, 7, 5.
17. (a) Kamimura, T.; Masegi, T.; Hata, T. Chem. Lett. 1982, 965; (b) Pathak, T.;
Chattopadhyaya, J. Chem. Scr. 1986, 26, 135.
18. Robins, M. J.; Naik, S. R.; Lee, A. S. K. J. Org. Chem. 1974, 39, 1891.
19. Chaix, C.; Duplaa, A. M.; Molko, D.; Teoule, R. Nucleic Acids Res. 1990, 18, 5143.
20. Ogilvie, K. K.; Beaucage, S. L.; Entwistle, D. W. Tetrahedron Lett. 1977, 1255.
21. Markiewicz, W. T. J. Chem. Res. (S) 1979, 24. J. Chem. Res. (M) 1979, 181.
22. Westman, E.; Stromberg, R. Nucleic Acids Res. 1994, 22, 2430.
23. (a) Sinha, N. D.; Biernat, J.; McManus, J.; Köster, H. Nucleic Acids Res. 1984, 12,
4539; (b) Milecki, J.; Dembek, P.; Antkowiak, W. Z.; Gdaniec, Z.; Mielewczyk, S.;
Adamiak, R. W. Nucleosides Nucleotides 1989, 8, 463.
Acknowledgments
The excellent technical help of Ms. Krystyna Sternal is greatly
appreciated. The authors thank Dr. Łukasz Marczak of the Institute