Development of fluorescent double-strand probes labeled with 8-(p-CF 3-cinnamyl)-adenosine for the detection of cyclin D1 breast cancer marker

Article abstract of DOI:10.1016/j.ejmech.2014.03.081

Fluorescent nucleoside analogs replacing natural DNA bases in an oligonucleotide have been widely used for the detection of genetic material. Previously, we have described 2-((4-(trifluoromethyl) phenyl)-trans-vinyl)- 2′-deoxy-adenosine, 6, a nucleoside a

Full text of DOI:10.1016/j.ejmech.2014.03.081

European Journal of Medicinal Chemistry 79 (2014) 77e88
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Development of fluorescent double-strand probes labeled with
8-(p-CF3-cinnamyl)-adenosine for the detection of cyclin D1 breast
cancer marker
Lital Zilbershtein-Shklanovsky a, Pinhas Kafri b, Yaron Shav-Tal b, Eylon Yavin c,
,
Bilha Fischer a
*
a Department of Chemistry, Gonda-Goldschmied Medical Research Center, Bar-Ilan University, Ramat-Gan 52900, Israel
b Faculty of Life Sciences and Institute of Nanotechnology, Bar-Ilan University, Ramat-Gan 52900, Israel
c School of Pharmacy, Institute for Drug Research, The Hebrew University of Jerusalem, Ein-Kerem, Jerusalem 91120, Israel
a r t i c l e i n f o
a b s t r a c t
Article history:
Fluorescent nucleoside analogs replacing natural DNA bases in an oligonucleotide have been widely used
for the detection of genetic material. Previously, we have described 2-((4-(trifluoromethyl) phenyl)-
trans-vinyl)-20-deoxy-adenosine, 6, a nucleoside analog with intrinsic fluorescence (NIF). Analog 6 ex-
hibits a quantum yield 3115-fold higher than that of adenosine (4 0.81) and maximum emission which is
120 nm red shifted (lem 439 nm). Here, we incorporated this analog in one or several positions of cyclin
D1-targeting 15-mer oligonucleotides (ONs). The fluorescence of 6 was quenched upon incorporation
into an oligonucleotide (ca. 1.5e22 fold), and was further reduced upon duplex formation. Specifically,
ON7 exhibited a fluorescence decrease of ca. 2- or 3-fold upon duplex formation with complementary
DNA or RNA strand, respectively. We determined the kinetics of dehybridization/rehybridization process
in the presence of ssDNA or ssRNA targets to optimize our probes length and established the probes’
selectivity towards a specific target. Furthermore, we proved specificity of our probe to the target vs.
singly mismatched targets. Our most promising ds-NIF-probe, ON7:RNA, was used for the detection of
cyclin D1 mRNA marker in cancerous cells total RNA extracts. The ds-probe specifically recognized the
target as observed by a 2-fold fluorescence increase within 30 s at RT. These findings illustrate the po-
tential of ds-NIF-probes for the diagnosis of breast cancer.
Received 5 February 2014
Received in revised form
24 March 2014
Accepted 29 March 2014
Available online 30 March 2014
Keywords:
Oligonucleotide
Fluorescence
Hybridization
Cyclin D1 mRNA
Diagnostics
Ó 2014 Elsevier Masson SAS. All rights reserved.
1. Introduction
Fluorescent nucleoside analogs with intrinsic fluorescence,
based on the elongation of the natural purine/pyrimidine p-system
During the past decade, there has been a remarkable growth in
the use of synthetic oligonucleotide probes in nucleic acid research
for the detection of specific DNA or RNA sequences [1e3]. The
sensitivity of fluorescence detection and its simplicity makes it an
important tool in many areas and is being used for diagnostics of
genetic and infectious diseases, discovery of gene-targeted drugs,
molecular biology, and other biomedicinal studies [4,5].
replacing natural DNA bases in an oligonucleotide, have been
developed by us and others as probes for detection of genetic
material [6e12].
New fluorescent nucleoside probes have to meet the following
requirements: they should be structurally similar to native nucle-
obases and capable of WatsoneCrick pairing, their absorption and
emission spectra should be red-shifted relative to those of native
nucleosides, and they should exhibit 1000s-times higher quantum
yields vs. natural nucleosides. Moreover, to become an effective tool
for hybridization analysis, the labeled oligonucleotide has to
display a considerable change of fluorescence upon hybridization
with the complementary strand.
Purine nucleosides with intrinsic fluorescence include, for
instance, 2-aminopurine (2AP), 1, Fig. 1, (labs 303 nm, lem 370 nm, 4
0.68 in neutral aqueous solution) [13,14]. The fluorescence of 2AP is
strongly quenched within DNA, and this quenching is sensitive to
Abbreviations: A, adenosine; G, guanosine; C, cytosine; T, thymidine; ON,
oligonucleotide; NIF, nucleoside with intrinsic fluorescence; TPPTS, tris(3-
sulfonatophenyl) phosphine trisodium salt; MALDI, matrix-assisted laser desorp-
tion/ionization; BLAST, basic local alignment search tool; OD, optical density; PBS,
phosphate buffer saline.
* Corresponding author.
E-mail address: bilha.fischer@biu.ac.il (B. Fischer).
http://dx.doi.org/10.1016/j.ejmech.2014.03.081
0223-5234/Ó 2014 Elsevier Masson SAS. All rights reserved.

78
L. Zilbershtein-Shklanovsky et al. / European Journal of Medicinal Chemistry 79 (2014) 77e88
Fig. 1. Previously reported fluorescent purine and pyrimidine nucleoside analogs.
local and global changes in DNA conformation. Therefore this flu-
orophore has been employed to probe the structural and dynamic
changes that characterize folding of ribozymes [15], DNA contain-
ing damage or mismatches [16,17], etc. Another known purine
manner as adenosine by testing the probe vs. singly mismatched
targets. We analyzed the dependence of photophysical properties
and thermal stability of labeled ds-ONs on the number of labels,
and on the neighboring nucleobases. Furthermore, we determined
the kinetics of dehybridization/rehybridization process of ds-NIF-
probes of different lengths with ssDNA or ssRNA targets, and also
established the probes’ selectivity towards a specific target. Finally,
we demonstrated the use of the optimal ds-NIF-probe, ON7:RNA, for
the detection of cyclin D1, a breast cancer mRNA marker, in total
RNA extracts from cancerous cells.
nucleoside probe is 8-vinyl-deoxyadenosine (8VdA), 2,
(labs
290 nm, lem 388 nm, 4 0.66 in neutral aqueous solution) [18e20].
8VdA exhibits similar base-pairing and base-stacking properties to
those of adenine when introduced into double-stranded DNA
(dsDNA). Moreover, the fluorescence quantum yield of 8VdA is
sensitive to base stacking, making it a useful real-time probe of DNA
structure [18]. However, both 2AP and 8VdA emit at relatively short
wavelengths.
Pyrimidine nucleosides with intrinsic fluorescence include, for
instance, 6-phenyl-pyrrolocytidine (PhpC), 3, (labs 360, lem 465 nm,
4 0.31 in phosphate buffer) [21,22]; pyrimidindole cytosine (dCPPI),
4, (lmax 374, lem 513 nm, 4 0.006 in phosphate buffer) [23,24] and 5-
2. Results
2.1. Design of NIF-labeled-oligonucleotide probes
((4-methoxy-phenyl)-trans-vinyl)-20-deoxy-uridine,
320 nm, lem 478 nm, 4 0.12 in H2O) [8].
5
(
labs
We prepared 15-mer long ONs targeting cyclin D1 mRNA. Cyclin
D1 protein is a major player in the control of the cell cycle [25]. The
cyclin D1 gene has been convincingly implicated in oncogenesis,
and it is highly expressed in many cancer types and has been
specifically implicated in breast cancer [26]. From the original
mRNA sequence of cyclin D1 consisting of 4304 nucleic acids, we
selected a 15-mer sequence containing several T and A residues. We
chose a 15-mer long oligonucleotide since this is the minimal
length that would ensure specific binding to target cyclin D1. The
specificity of the sequence was assessed by BLAST (basic local
alignment search tool) [27].
Monomer 6 was used for the preparation of cyclin D1-targeting
15-mer ONs containing a single label or multi-labels replacing A at
different positions of the oligonucleotide sequence. Since spectral
properties of fluorescent nucleosides within ONs are sensitive to
neighboring nucleobases [28e30], we incorporated monomer 6 at
various positions of the studied 15-mer oligonucleotides. In addi-
tion, differently multi-labeled ONs were prepared to examine the
influence of more than one label on fluorescence of the single-
strand and of the resulting duplex.
Recently, we reported on the promising properties of a novel
adenosine analog with intrinsic fluorescence (nucleoside with
intrinsic fluorescence, NIF), 2-((4-(trifluoromethyl) phenyl)-trans-
vinyl)-20-deoxy-adenosine, 6, (labs 340 nm, lem 439 nm, 4 0.81 in
MeOH) [12]. This fluorescent adenosine analog bears a minimal
chemical modification, at a position not involved in base-pairing,
resulting in relatively long absorption and emission wavelengths
and high quantum yield. Analog 6 exhibits a quantum yield that is
3115-fold higher than that of adenosine and maximum emission
which is 120 nm red shifted. In addition, since analog 6 adopts an
anti conformation and S sugar puckering, favored by B-DNA, it
makes a promising nucleoside analog to be incorporated in an
oligonucleotide probe for the detection of genetic material by a
hybridization assay.
Here, we report on labeling cyclin D1-targeting 15-mer oligo-
nucleotides by one or several monomer(s) of 6, to be used as ss-NIF-
probes or as ds-NIF-probes. Next, we determined the competency of
monomer 6 to create WatsoneCrick H-bond pairing in the same

L. Zilbershtein-Shklanovsky et al. / European Journal of Medicinal Chemistry 79 (2014) 77e88
79
The synthesized oligonucleotides were based on a 20-OMe-RNA
scaffold. 20-OMe-RNA oligonucleotides are not substrates for RNa-
seH and show resistance to degradation by RNA- or DNA-specific
nucleases [31]. In addition to being stable to standard handling
and nuclease resistance, 20-OMe-RNA oligomers form more stable
hybrids with complementary RNA strands than equivalent DNA and
RNA sequences [32,33].
Since fluorescence enhancement rather than quenching should
be associated with a positive identification of a target, we propose
two prototypes of NIF-labeled-oligonucleotide probes: a ss-NIF-
probe and a ds-NIF-probe. Specifically, ss-NIF-probe is relatively
dark, but, if it becomes fluorescent upon hybridization with a
ssDNA or a ssRNA target, then ss-NIFs will be used as probes
(Scheme 1).
activated nucleoside phosphoramidite, 10, starting from 8-Br-20-
deoxy-adenosine. [34] The latter was protected with 4, 40-dime-
thoxytrityl group at C50-OH, to obtain 7 in 73% yield, which was
then employed as a substrate in a Suzuki coupling reaction with
trans-2-[4-(trifluoromethyl)phenyl]vinylboronic acid to give 8 in
80% yield. Compound 8 was protected at the exocyclic N6 with N,N-
dimethylformamide dimethyl acetal to give 9 in 99% yield. Finally,
the protected and activated nucleoside phosphoramidite, 10, was
obtained in 82% yield upon treatment of 10 with phosphoramidite
chloride (Scheme 3).
The modified ONs, containing analog 6, were synthesized by
standard solid-phase oligonucleotide synthesis and purified by
reverse-phase HPLC. MALDI-TOF mass spectroscopy confirmed
their identity.
However, if there is a decrease in the fluorescence intensity of
the duplex as compared to ss-NIF-DNA fluorescence, a ds-NIF-probe
will be used (Scheme 2). The ds-NIF-probe, III, (Scheme 2A) is
relatively dark due to nucleobase interactions which result in
quenching of fluorophore. Next, this ds-probe, III, is allowed to
hybridize with a target sequence, IV (Scheme 2B). Here, we expect
the formation of a new duplex, V, (Scheme 2B) and the release of
fluorescent ss-NIF-DNA, I, thus indicating the presence of the
target.
The requirements for application of this methodology for the
detection of target DNA/RNA are that the free ss-NIF-DNA should be
relatively fluorescent and NIF-DNA:DNA duplex should be rela-
tively dark. Moreover, the ss-NIF-DNA should be shorter than its
complementary oligonucleotide to faciliate dehybridization of III
and re- hybridization with IV. If the target is mRNA the more stable
DNA: mRNA duplex V, will be preferred over DNAeDNA duplex III.
To apply the NIF methodology a considerable difference of fluo-
rescence intensity (at least two fold) between the fluorescence of
the free ss-NIF-DNA and that of the duplex NIF-DNA:DNA/RNA is
required.
2.3. Photophysical characterization of cyclin D1-targeting sseNIFe
probes
To identify an optimal ss-NIF-probe, five variously labeled single
strand oligonucleotides were synthesized (ONs 1e5). Four of these
oligonucleotides were mono-labeled with monomer 6 (ONs 1e4),
at position 3 or 8 or 13 or 14, and one oligonucleotide was di-
labeled at positions 3 and 8 (ON5), Scheme 4. For each oligomer
the corresponding duplex was prepared. We studied both
DNA:DNA duplexes and DNA:RNA duplexes due to possible differ-
ences in the fluorescent properties upon hybridization of ONs 1e5
with complementary DNA or RNA.
Excitation and emission spectra of both single strands of labeled
ONs and the corresponding duplexes were measured in PBS buffer
(pH 7.4) in duplicates on two days, and the determined quantum
yields are the average values of those measurements.
All studied ONs, either single or double strands, were excited at
338e344 nm and emit at 436e443 nm (Table 1). The quantum
yields of all mono-labeled oligonucleotides decreased by ca. 1.5e
2.5-fold upon DNA:DNA duplex formation. This trend was also
observed for the corresponding DNA:RNA duplexes, however in
these cases the decrease was more significant ca. 4.5e10-fold
(Table 1). The di-labeled oligonucleotide did not exhibit a signifi-
cant change in quantum yield upon DNA:DNA duplex formation
(0.1 vs. 0.098), however, the quantum yield was decreased by ca.
2.5-fold upon DNA:RNA duplex formation. Since the ss-NIF probe
method requires an increase of the quantum yield of the duplex
compared to the single strand, none of these oligonucleotides can
be used as a probe for the detection of mRNA.
Two series of oligonucleotides were prepared to form the two
prototypes of NIF-labeled-oligonucleotide probes. Specifically, the
oligonucleotide was either complementary to the target mRNA (for
the ss-NIF-probe) or identical to the target mRNA (for the ds-NIF-
probe). These series of oligomers were used to explore whether our
monomer of choice, 6, when incorporated into a ss-oligonucleotide,
would trigger a significant quantum yield change upon hybridiza-
tion to a complementary sequence.
2.2. Incorporation of 8-(p-CF3-cinnamyl)-20-deoxy-adenosine (6)
into oligonucleotides
2.4. Photophysical characterization of cyclin D1-targeting ds-NIF-
probes
To develop fluorescent oligonucleotide probes for the detection
of DNA or RNA, we incorporated NIF-analog 8-(p-CF3-cinnamyl)-20-
deoxy-adenosine, 6, replacing 20-deoxy-adenosine, into a single-
strand oligonucleotide. We first prepared the protected and
In this series, six labeled single strands oligonucleotides were
synthesized. For each oligomer the corresponding duplex was
prepared. Since this method requires the dehybridization of the
duplex probe followed by the rehybridization of the target with its
complementary strand and the release of the fluorescent single-
strand (Scheme 2), the complementary strand was longer than
the labeled strand (35-mer vs. 15-mer, respectively), to allow
eventually formation of a longer and more stable duplex with the
target RNA (see below optimization of the length of the comple-
mentary strand).
To identify an optimal ds-NIF-probe, six variously labeled single
strand oligonucleotides were synthesized (ONs 6e11). Three of
these oligonucleotides were mono-labeled at position 4, 9 or 14
(ONs 6e8), two were di-labeled at positions 4 and 9 (ON9), and 4
and 14 (ON10), and one was tri-labeled at positions 4, 9 and 14
(ON11), Scheme 5.
Scheme 1. A schematic illustration of the ss-NIF-probe methodology.

80
L. Zilbershtein-Shklanovsky et al. / European Journal of Medicinal Chemistry 79 (2014) 77e88
Scheme 2. A schematic illustration of the mechanism of action of a ds-NIF-probe. methodology.
Scheme 3. Synthesis of phosphoramidite 10. Reaction conditions: a) 4, 40-dimethoxytrityl chloride (1.6 eq), TEA (2 eq), DMAP (0.25 eq), pyridine, 73% yield; b) trans-2-[4-
(trifluoromethyl)phenyl]vinylboronic acid (1.25 eq), Na2CO3 (3 eq), TPPTS (0.25 eq), Pd(OAC)2 (0.05 eq), CH3CN:H2O (1:2), 80% yield; c) N,N-Dimethylformamide dimethyl acetal
(6.5 eq), DMF, 99 % yield; d) N,N-diisopropylcyanoethyl-phosphoramidite chloride (1.2 eq), DIPEA (2.5 eq), DCM, 82% yield.
All studied ONs, either single or double strands, were excited at
336e343 nm and emit at 437e442 nm (Table 2). The measure-
ments were performed in duplicates on two days, and the quantum
yield values are averaged. Here, as for the ss-NIF-probe, we exam-
ined the change in fluorescence of the single strand upon DNA:DNA
Unmodified ON: 5'- CUA CGC UAC UGU AAC -3'
* Complementary ON: 5'- GUU ACA GUA GCG UAG -3'
ON1: 5'- CUA CGC UAC UGU AAC -3'
ON2: 5'- CUA CGC UAC UGU AAC -3'
ON3: 5'- CUA CGC UAC UGU AAC -3'
ON4: 5'- CUA CGC UAC UGU AAC -3'
ON5: 5'- CUA CGC UAC UGU AAC -3'
and DNA:RNA duplex formation.
Oligonucleotides mono-labeled at position 4 or 14 (ONs 6 and 8)
did not exhibit a significant decrease of fluorescence upon duplex
formation of the corresponding NIF-DNA:DNA or NIF-DNA:RNA
(1.5, 1.2-fold or 1.8, 1.3-fold, respectively), Table 2. However, oligo-
nucleotide mono-labeled at position 9 (ON7) displayed quantum
yield value of 0.039, and this value was decreased by ca. 2-fold upon
NIF-DNA:DNA duplex formation and by ca. 3-fold upon NIF-
DNA:RNA duplex formation. The di-labeled oligonucleotides (ONs
9 and 10) exhibited a slight increase in quantum yields upon duplex
formation, either NIF-DNA:DNA or NIF-DNA:RNA, while the quan-
tum yield of the tri-labeled oligonucleotide (ON11) remained
A denotes fluorescent adenosine label. All other bases are 2’-OMe RNA.
* The complementary oligonucleotide is either a DNA or a RNA strand.
Scheme 4. Labeled-ONs for the preparation of ss-NIF-probes for the detection of cyclin
D1 mRNA.

L. Zilbershtein-Shklanovsky et al. / European Journal of Medicinal Chemistry 79 (2014) 77e88
81
Table 1
Photophysical properties of labeled cyclin D1-targeting ONs (1e5) designed as ss-NIF probes and their corresponding duplexes.a
labs max (nm)
lem max (nm)
4
Single
Duplex
Duplex
Single
Duplex
Duplex
Single
Duplex
Duplex
strand (ss)
with DNA
with RNA
strand (ss)
with DNA
with RNA
strand (ss)
with DNA
with RNA
ON1
ON2
ON3
ON4
ON5
343
340
341
340
342
339
340
342
344
338
340
340
340
340
343
437
436
440
443
439
440
436
436
443
441
438
440
440
439
443
0.14
0.25
0.37
0.55
0.1
0.056
0.176
0.29
0.37
0.098
0.033
0.044
0.084
0.055
0.04
a
All measurements were carried out at 2 mM in a PBS buffer (pH 7.4) at room temperature.
unchanged upon NIF-DNA:DNA duplex formation, but exhibited a
ca. 1.5-fold increase in quantum yield upon NIF-DNA:RNA duplex
formation (Table 2).
concluded that the dehybridization of the ds-probe (Fig. 4CeG)
occurred due to specific target recognition and not due to sponta-
neous opening of the probe in the presence of any ON.
2.7. Reaction kinetics with double-stranded probes and a ssRNA
target
2.5. Evaluating the ability of monomer 6 to retain the hybridization
preference of adenosine
The ds-probes mentioned above were made up of two com-
plementary ONs at different length, when only at 5 nt difference a
detectable hybridization/dehybridization process occurred. How-
ever, for RNA detection the two complementary ONs can be at the
same length, or at least with a smaller nt difference between the
two strands, since RNA forms a more stable duplex than DNA with a
complementary strand. Since our ultimate goal was using these
probes for the detection of mRNA, we tested the ds-NIF-probes for
their hybridization kinetics with a target 30-mer ssRNA at room
temperature. Fig. 7 shows the kinetic curves of the ds-probes hy-
bridized with the target.
As shown in Fig. 5, when the two strands of the probe were of
equal length (Fig. 5A), or, when the complementary ON was 17-mer
(Fig. 5B), a small increase in fluorescence was observed, implying
that dehybridization/rehybridization process has occurred (unlike
for a DNA target). However, the reaction was very slow and after
35 min the reaction reached only 1/5 of the maximum fluorescence
intensity. At 5 nt difference (Fig. 5C), after 35 min the reaction still
did not reach its maximum. However, when the difference was
10 nt (Fig. 5D), maximum reaction occurred after 5 min. When the
length difference exceeded 12 nt, maximum fluorescence was ob-
tained after only 15 s (Fig. 5EeF).
Analog 6 was designed to be incorporated into an oligonucleo-
tide that upon hybridization will exhibit different fluorescence
characteristics than the single-strand. Furthermore, monomer 6
should be capable of WatsoneCrick base-pairing similarly to
adenosine. For this purpose, ON7 was hybridized to its perfectly
matched complementary ON, and to singly mismatched ONs
(Fig. 2).
Upon hybridization of ON7 with its perfectly matched comple-
mentary ON, the fluorescence was quenched by ca. 3.5-fold. How-
ever, when the labeled ON7 is hybridized with mismatched A/C/G
ONs, the fluorescence was quenched only by ca. 2-fold, Fig. 3. Since
similar results were obtained for all three mismatched ONs, we
concluded that monomer 6 specifically recognizes T as an adeno-
sine base does.
2.6. Reaction kinetics with double-stranded probes and a ssDNA
target
Subsequent to proving the selectivity of 15-mer NIF probe ON7,
we set to optimize the length of the complementary strand aiming
at a facile dehybridization of the ds-NIF-probe followed by rehy-
bridization with target mRNA.
In addition, we tested ds-NIF-probe containing equally long
strands (15-mer each) with a 15-mer RNA target in which only 5 nt
were complementary to the tested probe (Fig. 5G). However, when
the ds-NIF-probe was added to this target RNA no reaction was
observed.
For this purpose, we prepared a series of double-stranded ONs
composed of ON7 (15-mer) and complementary strands of different
lengths (15- to 35-mer). The ds-probes were tested for their hy-
bridization kinetics with a target single strand at room tempera-
ture. Specifically, instead of using the full mRNA of cyclin-D1, we
used a 35-mer complementary DNA sequence. Fig. 4 shows the
kinetic curves of the ds-NIF-probes hybridized to the 35-mer DNA
target.
2.8. Using ON7 for the application of ds-NIF-probe methodology
To prove the applicability of analog 6 labeled ONs for detection
of target ssDNA or ssRNA, we selected an ON exhibiting relatively
significant change of fluorescence upon hybridization/dehybrid-
ization. Specifically, we selected a cyclin D1-targeting ON labeled at
position 9 (ON7), the fluorescence of which is reduced ca. 2-fold
upon NIF-DNA:DNA duplex formation and ca. 3-fold upon NIF-
DNA:RNA duplex formation (Fig. 6).
When the two strands were of equal length (Fig. 4A), or, when
the complementary ON was a 17-mer oligonucleotide (Fig. 4B), no
increase of fluorescence was observed over time, meaning no
dehybridization/rehybridization process occurred. However, as the
length difference of the complementary ONs increased, the reac-
tion rate became faster (Fig. 4CeF). Specifically, when the overall
difference between the strands was 5 nucleotides (nt) (Fig. 4C), the
reaction still was not completed after 35 min. However, when the
difference was 10 or 12 nt (Fig. 4DeE), maximum reaction was
reached after 12 and 3 min, respectively. When the length differ-
ence exceeded 15 nt, maximum fluorescence was obtained after
only 15 s (Fig. 4FeG).
2.9. Application of ds-NIF-probe for detection of cyclin D1 mRNA in
total RNA extracts of cancerous cells
Next, to prove the applicability of the ds-NIF methodology for
identification of a specific mRNA, we monitored the fluorescence
intensity of the probe in total RNA extracted from the human U2OS
osteosarcoma cell line. The selected ds-NIF-probe consisted of ON7
Moreover, we have added a random 35-mer DNA to the ds-NIF-
probe composed of (15-mer) ON7:DNA (35-mer) (Fig. 4H). Since no
change in fluorescence intensity was observed in this case, we

82
L. Zilbershtein-Shklanovsky et al. / European Journal of Medicinal Chemistry 79 (2014) 77e88
Unmodified ON: 5'- GUU ACA GUA GCG UAG -3'
used for the detection of cyclin D1 mRNA, a breast cancer marker.
We selected a 15-mer sequence from the cyclin D1 mRNA sequence,
which is specific to this mRNA only and does not appear in any
other endogenous mRNA, as was determined by BLAST.
*Complementary ON: 5'- CACGGGCACGCUACGCUACUGUAACCAAGAGGGUA -3'
ON6: 5'- GUU ACA GUA GCG UAG -3'
ON7: 5'- GUU ACA GUA GCG UAG -3'
We examined the influence of the incorporation of monomer 6
into an ON on the fluorescence of the former. ONs 1e11 displayed
reduced quantum yields relative to that of the free monomer 6
(0.033e0.55 compared to 0.74). We observed that the fluorescence
intensities (which correlate with quantum yield) of the ONs are
not dependent on the location of the label in the ON, but, on the
neighboring nucleobases. In particular, ONs in which at least one
of the neighboring nucleobases of the labeled monomer was
guanine exhibited the lowest quantum yields (e.g., ONs 7e11). On
the contrary, ONs in which at least one of the neighboring
nucleobases of the labeled residue was adenine, exhibited the
highest quantum yields, as long as the other neighbor was not
guanine (e.g., ONs 3 and 4). The highest quantum yield was ob-
tained for ON4 with adenine and cytidine neighbors. In general,
ONs in which the label had thymidine and cytidine as neighbors
exhibited high quantum yields, 4 ꢁ 0.1 (e.g., ONs 1, 2, 5 and 6). No
additive effect was observed for multi-labeled ON. Moreover, di/
tri-labeled ONs exhibited an average quantum yield typical of
mono-labeled ONs (e.g., ON10 vs. ONs 6 and 8, 4 0.072, 0.296 and
0.037, respectively).
Development of a hybridization-based fluorescent probe re-
quires not only different photophysical properties in labeled single
strand vs. labeled duplex, but also duplex stability. If the labeled
duplex probe is unstable, namely, Tm value of the labeled duplex is
too low, it is of no use. Thus, we studied the dependence of the
thermo-stability of the labeled-duplexes on the site of the label and
number of labels. Surprisingly, we found that incorporation of
monomer 6 in ONs increases the stability of the resulting duplexes,
and in most labeled duplexes, Tm increased by 4e10 ꢀC as compared
to the unmodified duplex.
Moreover, no significant change was observed for mono-, di-, or
tri-labeled duplexes in terms of stability. For instance, ONs 1, 2, and
5 (mono-labeled at positions 3 or 8, or di-labeled at positions 3 and
8, respectively) exhibited almost the same Tm value (Tm ¼ 59e
60 ꢀC) for the DNA and for the RNA duplexes. The increased stability
of these duplexes may indicate favorable stacking interactions of
monomer 6, as compared to those of A itself with its neighboring
nucleobases.
We synthesized two prototypes of probes: ss-NIF-probes and ds-
NIF-probes. However, since a decrease of fluorescence was obtained
upon duplex formation in both cases, only the ds-NIF-probe
methodology was further pursued. For testing our methodology we
used our most promising probes-ON7:DNA and ON7:RNA, which
exhibited the most significant change in fluorescence upon duplex
formation (2- and 3-fold, respectively).
ON8: 5'- GUU ACA GUA GCG UAG -3'
ON9: 5'- GUU ACA GUA GCG UAG -3'
ON10: 5'- GUU ACA GUA GCG UAG -3'
ON11: 5'- GUU ACA GUA GCG UAG -3'
A denotes fluorescent adenosine label. All other bases are 2’-OMe RNA.
* The complementary oligonucleotide is either a DNA or a RNA strand.
Scheme 5. Labeled-ONs for the preparation of ds-NIF-probes for the detection of cyclin
D1 mRNA.
(15-mer) hybridized with a complementary 30-mer RNA, since this
probe exhibited the most significant change in fluorescence upon
hybridization, exhibiting 3-fold decrease of fluorescence. The ds-
NIF-probe (ON7:RNA) was added to total cell RNA extracts at a final
concentration of 1 mM. The first extract was obtained from a cell
line stably overexpressing the cyclin D1 gene, therefore containing
high levels of cyclin D1 mRNA (up to 3-fold, as measured by
quantitative RT-PCR, see Experimental section), while the second
cell extract contained basal levels of cyclin D1 mRNA found in U2OS
cells. Fluorescent ON7 was used as a positive control, while ds-
ON7:RNA probe was the negative control, and the fluorescence of
RNA extract alone was tested as well.
Upon addition of ds-ON7:RNA probe to RNA extract containing
high levels of cyclin D1, we observed an immediate increase in the
fluorescent intensity up to 2-fold increase after only 30 s, relative to
the negative control, indicating the release of the fluorescent 15-
mer ON7 and the formation of a RNA:cyclin D1 RNA duplex (Fig. 7).
Yet, addition of ds-ON7:RNA probe to total RNA extract of cells
expressing basal levels of cyclin D1 did not result in any increase of
fluorescence intensity (Fig. 8).
2.10. Thermostability of duplexes
The application of the above ONs, or corresponding duplexes, for
the detection of genetic material depends on the thermostability of
the related duplexes. The Tm values of the ds-NIF-probe should not
be not too high to allow dehybridization before rehybridization
with the target DNA/RNA (Scheme 2).
The Tm values of all hybrids fluorescent ONs 1e11 with the
complementary DNA or RNA strands are 48e66 ꢀC, indicating that
they are thermally stable (Table 3).
In most cyclin D1-targeting labeled duplexes Tm increased by 4e
10 ꢀC as compared to the unmodified duplex, for the DNA and for
the RNA duplexes (Table 3). Exception to this increase was ON3
(mono-labeled at position 13) that upon hybridization with DNA
formed a less stable duplex by 7 ꢀC and upon hybridization with
RNA formed a duplex with the same stability as the unmodified
duplex (Tm ¼ 56 ꢀC). Another exception was ON4 (mono-labeled at
position 14) that upon hybridization with DNA formed a less stable
duplex by 5 ꢀC (Tm ¼ 50 ꢀC).
We first proved that monomer 6 exhibits the same base-pairing
pattern as adenosine, and therefore shows the same hybridization
preference. Monomer 6 labeled ON7 was hybridized with four ONs:
a perfectly matched complementary ON, and three singly mis-
matched ONs. ON7 exhibited larger fluorescence quenching upon
hybridization to its perfectly matched complementary ON, as
compared to hybridization to the singly mismatched ONs (3.5-fold
vs. 2-fold), indicating that monomer 6 and adenosine recognize T in
the same manner.
Next, we optimized the length of our ds-NIF-probe (ON7:DNA)
by studying the reaction kinetic of duplexes having complementary
strands of different lengths upon addition of cyclin D1-based ssDNA
or ssRNA targets, at room temperature. Moreover, the targets were
either perfectly matched to the studied probes or random DNA
targets to evaluate the probes’ selectivity towards a specific target
sequence.
In general, there was no significant Tm difference (0e3 ꢀC) be-
tween duplexes in which the complementary strand was DNA or
RNA, neither in the modified nor in the unmodified duplexes.
3. Discussion
We described the design and synthesis of NIF-labeled-
oligonucleotide probes based on the incorporation of the fluores-
cent adenosine analog, 6. The oligonucleotides were designed to be

L. Zilbershtein-Shklanovsky et al. / European Journal of Medicinal Chemistry 79 (2014) 77e88
83
Table 2
Photophysical properties of labeled cyclin D1-targeting ONs (6-11) designed as ds-NIF probes and their corresponding duplexes.a
labs max (nm)
lem max (nm)
4
Single
Duplex
Duplex
Single
Duplex
Duplex
Single
Duplex
Duplex
strand (ss)
with DNA
with RNA
strand (ss)
with DNA
with RNA
strand (ss)
with DNA
with RNA
ON6
ON7
ON8
ON9
ON10
ON11
340
337
336
339
342
340
343
343
336
343
343
342
342
343
339
341
343
342
441
439
437
441
439
440
440
440
437
439
441
442
441
440
439
439
441
439
0.296
0.039
0.037
0.042
0.072
0.033
0.197
0.021
0.032
0.062
0.079
0.031
0.164
0.014
0.029
0.067
0.086
0.046
a
All measurements were carried out at 2 mM in a PBS buffer (pH 7.4) at room temperature.
experimental procedures for the attachment of a fluorophore prior
to detection.
Our findings are a stepping stone to a further design of new and
more efficient probes based on NIF-method.
ON7: 5'- GTT ACA GTA GCG TAG -3'
match: 3'- CAA TGT CAT CGC ATC -5'
mismatch-A: 3'- CAA TGT CAA CGC ATC -5'
mismatch-C: 3'- CAA TGT CAC CGC ATC -5'
mismatch-G: 3'- CAA TGT CAG CGC ATC -5'
Fig. 2. The sequences of matched and mismatched ONs for hybridization with ON7.
4. Experimental
4.1. General
Compounds were characterized by NMR using a Bruker AC-200,
DPX-300 or DMX-600 spectrometers. Chemical shifts are expressed
in ppm downfield from Me4Si (TMS), used as internal standard. The
values are given in
d scale. All air and moisture sensitive reactions
For the DNA targets, even though a detectable reaction between
the ds-NIF-probe and the target was observed at 5 nt difference
(Fig. 6C), only when the difference was ꢁ10 nt, the reaction reached
its maximum within 12 min (Fig. 6D). The optimal probes reacting
immediately, in less than 15 s, were obtained for ds-NIF-DNA:DNA
in which the difference was ꢁ15 nt (Fig. 6F).
were carried out in flame-dried, nitrogen flushed flasks sealed with
rubber septa, and the reagents were introduced with a syringe. All
reactants in moisture sensitive reactions were dried overnight in a
vacuum oven. Progress of the reactions was monitored by TLC on
precoated Merck silica gel plates (60F-254). Visualization was
accomplished by UV light. Flash chromatography was carried out
on silica gel (Davisil Art. 1000101501). Medium pressure chroma-
tography was carried out using automated flash purification system
(Biotage SP1 separation system, Uppsala, Sweden). New com-
pounds were analyzed under ESI (electron spray ionization) con-
ditions on a Q-TOF micro-instrument (Waters, UK) and high
resolution MS-MALDI-TOF spectra were recorded with an autoflex
TOF/TOF instrument (Bruker, Germany). Absorption spectra were
measured on a UV instrument (UVe2401PC UVeVIS recording
spectrophotometer, Shimadzu, Kyoto, Japan) using a 10 mm quartz
cell with 1 cm path length. Emission spectra were measured using
However, when using the latter probes for the detection of a
RNA target (Fig. 7), even though a detectable reaction was observed
when the two strands of the probe were of equal length, only when
the difference was ꢁ10 nt, the reaction reached its maximum
within 5 min (Fig. 7D). An immediate reaction (less than 15 s) was
achieved when the difference between the lengths of the strands in
the ds-NIF-probe was ꢁ12 nt (Fig. 7EeF).
These probes were designed to be used for the detection of
mRNA. Therefore, after proving their ability to detect selectively
ssDNA and ssRNA sequences representing part of target mRNA, we
demonstrated the ability of the optimal ds-NIF-probe ON7:RNA to
detect selectively cyclin D1 mRNA in total RNA extracts from
cancerous U2OS cells. An immediate increase in fluorescence up to
2-fold was observed following the addition of the probe to the
extract containing high levels of cyclin D1 vs. extract containing
basal levels cyclin D1, indicating the recognition of the target mRNA
by the probe.
To conclude, we propose a novel probe for the efficient and
selective detection of cyclin D1 mRNA. This probe is based on the
incorporation of the fluorescent monomer, 6, into an oligonucleo-
tide to create a fluorescent single-strand. When the latter labeled
strand forms a duplex with a complementary unmodified strand,
the fluorescence is quenched, resulting in a relatively dark duplex
probe. Upon addition of a target single-strand to the dark probe a
new and more stable duplex is formed with the release of the
fluorescent oligonucleotide.
Aminco-Bowman series
2 (AB2) Luminescence Spectrometer
(Thermo electron corporation, Markham, Ontario, Canada). Reac-
tion kinetics of displacement hybridization experiments were
measured using Cary Eclipse Fluorescence Spectrophotometer.
Thermal denaturation curves measurements were performed on a
Cary 300 UVevisible spectrophotometer (Varian Inc.). Absorption
and fluorescence spectra were recorded in PBS buffer containing
NaC1 8.0 g, KC1 0.2 g, Na2HPO4 1.15 g, KH2PO4 0.2 g, and water
100 mL (pH 7.4). The unmodified oligonucleotides were purchased
from Integrated DNA Technologies (Coralville, Iowa, USA). Modified
oligonucleotides were synthesized by standard automated solid-
phase method on an AKTA oligopilot plus oligonucleotide synthe-
sizer. All commercial reagents were used without further
purification.
We validated our methodology by successfully detecting DNA
and RNA variants of cyclin D1 as well as cyclin D1 mRNA from cell
extracts using the relatively dark ON7:RNA duplex probe. These
results support the potential usefulness of ON7:RNA probe for the
diagnosis of breast cancer.
4.2. Extinction coefficients calculations
The following 260 nm extinction coefficients were used to
determine the concentration of the modified oligonucleotides:
εG ¼ 11,500 Mꢂ1 cmꢂ1, εA ¼ 15,400 Mꢂ1 cmꢂ1, εU ¼ 9900 Mꢂ1 cmꢂ1
,
A significant advantage of our new methodology is that the ON
itself is intrinsically fluorescent and there is no need for additional
εC ¼ 7200 Mꢂ1 cmꢂ1 and ε (modified)A ¼ 17,640 Mꢂ1 cmꢂ1. The
extinction coefficients of the duplexes (εD) are less than the sum of

84
L. Zilbershtein-Shklanovsky et al. / European Journal of Medicinal Chemistry 79 (2014) 77e88
where fAT and fGC are the fractions of the AT and GC base pairs,
respectively.
4.3. Oligonucleotide synthesis and purification
The modified 20-OMe oligonucleotides were synthesized at
3
mmol scale using a conventional automated solid-phase protocol
using natural -cyanoethylphophoramidite nucleosides. The
b
coupling time of the modified base was increased from 2 to 10 min.
Cleavage from the solid support and oligonucleotide deprotection
was carried out in a 3.5 M solution of NH3 in MeOH (300
mL) at RT
for 2 h, followed by treatment with 33% NH4OH (250
m
L) at 55 ꢀC for
overnight. The residue was washed with EtOH:H2O 1:1 (0.4 mL),
and following spin down, the resulting aqueous solution was
removed by speed-vac. Final purification of the oligonucleotides
was achieved on a HPLC (MerckeHitachi) system using a semi-
preparative reverse phase column (Gemini 5u C-18 110A,
Fig. 3. Emission spectra of ON7 alone and hybridized with the four targets (match;
mismatch-A; mismatch-C; mismatch-G) after excitation at 340 nm. All duplexes and
single strand were measured at 2 mM in a PBS buffer (pH 7.4) at room temperature.
250 mm ꢃ 10 mm, 5
mm, Phenomenex, Torrance, CA, USA), in TEAA
Fig. 4. Reaction kinetics of displacement hybridization between ds-ONs and a 35-mer DNA target. The final concentrations of the ds-ON probe and the target strand in the solution
were 2 mM and 4 mM, respectively. The sequences of the ds-ONs tested are listed at the top of each graph, where underlining identifies the modified nucleoside substituent. In
graphs A-G the listed ds-ONs were treated with a perfectly complementary target 50-TACCCTCTTGGTTACAGTAGCGTAGCGTGCCCGTG-30, and in graph H the listed ds-ON was treated
with a random 35-mer DNA target 50-CTATTCTCGCTACCGGAAGTATCGTGTCCGTGCGC-30.
the extinction coefficients of their complementary strands (εS1, εS2),
due to hypochromic effect that should be taken into account [35].
Therefore, the extinction coefficients were calculated by the
following equation:
buffer (0.1 M, pH 7.0): ACN 87:13 to 75:25 in 30 min at a flow rate of
5 mL/min. Purity was determined by analytical HPLC and confirmed
>95% pure. All modified ONs were confirmed by MALDI-TOF mass
spectrometry.
ON1 calcd for C164H210F3N53O102P14: 5045, found 5048.
ON2 calcd for C164H210F3N53O102P14: 5045, found 5047.
ON3 calcd for C164H210F3N53O102P14: 5045, found 5047.
ON4 calcd for C164H210F3N53O102P14: 5045, found 5045.
εD ¼ ð1 ꢂ h260 nmÞðεS1 þ εS2
Þ
h260 nm ¼ 0:287fAT þ 0:059fGC

L. Zilbershtein-Shklanovsky et al. / European Journal of Medicinal Chemistry 79 (2014) 77e88
85
Fig. 5. Reaction kinetics of displacement hybridization between ds-ONs and a 30-mer RNA target. The final concentrations of the ds-ON probe and the target strand in the solution
were 2 mM and 4 mM, respectively. The sequences of the ds-ONs tested are listed at the top of each graph, where underlining identifies the modified nucleoside substituent. In
graphs A-F the listed ds-ONs were treated with a perfectly complementary target 50-UACCCUCUUGGUUACAGUAGCGUAGCGUGC-30, and in graph G the listed ds-ON was treated
with RNA target contains only 5 complementary nt 50-UACCG AGUAGAUCGU-30.
ON5 calcd for C172H215F6N53O101P14: 5187, found 5187.
ON6 calcd for C167H210F3N59O102P14: 5165, found 5168.
ON7 calcd for C167H210F3N59O102P14: 5165, found 5163.
ON8 calcd for C167H210F3N59O102P14Na: 5188, found 5184.
ON9 calcd for C175H213F6N59O101P14: 5307, found 5311.
ON10 calcd for C175H213F6N59O101P14: 5307, found 5308.
ON11 calcd for C183H216F9N59O100P14K: 5484, found 5489.
4.6. Quantum yield determination
The quantum yields (4) of the single-stranded and double-
stranded oligonucleotides were calculated from the observed
absorbance and the area of the fluorescence emission band. The
fluorescence quantum yields of all oligonucleotides were deter-
mined relative to quinine sulfate in 0.1 M H2SO4 (lex 350 nm, lem
446 nm, 4 0.54) [36]. The quantum yield was calculated according
to the following equation:
I ODR h2
4 ¼ 4R
4.4. UV measurements
hR2
IR OD
The concentrations of all single-stranded and double-stranded
Here, 4 and 4 are the fluorescence quantum yield of the sample
R
oligonucleotides were 2
mM in PBS buffer (pH 7.4), and were
and the reference, respectively, I and IR are areas under the fluo-
rescence spectra of the sample and of the reference, respectively,
OD and ODR are the absorption values of the sample and the
determined by UV absorption measurement at 260 nm. Samples
were measured at room temperature in a 10 mm quartz cell with a
1 cm path length.
reference at the excitation wavelength, and
h and hR are the
refractive index for the respective solvents used for the sample and
the reference.
4.5. Fluorescence measurements
4.7. Hybridization of oligonucleotides
Emission spectra of all the single-stranded and double-stranded
oligonucleotides were measured in PBS buffer (pH 7.4) at the con-
Solutions of labeled single-strands were mixed at room tem-
perature with an equimolar amount of the complementary single
strand oligonucleotides in PBS buffer (pH 7.4). Samples were hy-
bridized by heating to 90 ꢀC for 5 min and subsequently allowed to
cool to room temperature over 2 h prior to measurements.
centration of 2
mM. Measurements conditions included: excitation
at 340 nm, 650 V sensitivity, and a 5 nm slit. Samples were
measured at room temperature in a 70
length of 1 cm.
mL quartz cell with path

86
L. Zilbershtein-Shklanovsky et al. / European Journal of Medicinal Chemistry 79 (2014) 77e88
4.11. Fluorescence measurements of double-stranded probe in total
cell RNA extracts
Fluorescence measurements of double-stranded probe,
ON7:RNA, in RNA extracts were performed at room temperature in
PBS buffer (pH 7.4). The final concentration of the probe was 1
while the concentration of the total RNA extracts ranged from 6000
to 7500 ng/ L. The fluorescence was measured upon excitation at
mM,
m
340 nm. Samples were measured in a 10 mm quartz cell with a 1 cm
path length. Duplicate samples were measured and the experi-
ments were performed on two days using two different total RNA
extract batches.
4.12. Synthesis
Fig. 6. Emission spectra of Cyclin D1-derived ON7 vs. its duplexes with complemen-
tary DNA and RNA strands.
4.12.1. 5-(6-Amino-8-bromo-9H-purin-9-yl)-2-((bis(4-
methoxyphenyl)(phenyl) methoxy)methyl)tetrahydrofuran-3-ol (7)
[34]
8-Bromo-20-deoxyadenosine (1 g, 3.03 mmol) was co-
evaporated with dry pyridine and then dissolved in dry pyridine
(15 mL). Et3N (0.845 mL, 6.06 mmol), DMAP (93 mg, 0.7575 mmol),
and DMT-Cl (1.64 g, 4.848 mmol) were added under N2, and the
mixture was stirred at RT. for 5 h. Then, MeOH (2.5 mL) was added
and the mixture was extracted with DCM. The organic phase was
washed with sat. NaHCO3 soution, the combined organic layer was
dried (Na2SO4) and concentrated, and the residue was purified on a
silica gel column using DCM:MeOH, 98:2, as the eluant. Analog 50
was obtained as a light yellow foam (1.4 g, 73%). Mp 122 ꢀC; 1H NMR
4.8. Thermal denaturation measurements (Tm)
The Tm values of duplexes were measured in PBS buffer (pH 7.4)
at 1
mM concentrations. The absorbance of the samples was
monitored at 260 nm and the temperature ranged from 20 to 85 ꢀC
with heating rate of 1 ꢀC/min.
4.9. Reaction kinetics of displacement hybridization
(600 MHz, DMSO-d6):
d
7.99 (s, 1H), 7.39 (s, 2H), 7.26 (d, J ¼ 7.5 Hz,
2H), 7.20e7.12 (m, 7H), 6.78 (d, J ¼ 9 Hz, 2H), 6.75 (d, J ¼ 9 Hz, 2H),
6.30 (m, 1H), 5.35 (d, J ¼ 4.5 Hz, 1H), 4.71 (m, 1H), 3.95 (m, 1H), 3.72
(s, 3H), 3.71 (s, 3H), 3.48 (m, 1H), 3.17 (m, 1H), 3.07 (m, 1H) 2.27 (m,
The reaction kinetics experiments of double-stranded probes
were carried out at room temperature in a 70 mL quartz fluores-
cence cell with path length of 1 cm. Fluorescence was measured
upon excitation at 340 nm. The final concentration of the double-
1H) ppm; 13C NMR (150 MHz, DMSO-d6):
d 157.9, 157.9, 154.9, 152.4,
150.0, 144.9, 135.7, 135.6, 129.6, 129.5, 127.6, 127.2, 126.4, 119.5,
113.0, 112.9, 85.6, 85.5, 85.1, 70.4, 63.3, 55.0, 54.9, 48.6, 36.2 ppm.
MS ESþ m/z: 655 (MNaþ); HR MALDI calcd for C31H30BrN5NaO5:
654.1323, found 654.1270.
stranded probe and the target strand in the solution was 2
and 4 M, respectively (to ensure full hybridization).
mM
m
4.10. Cell culture and RNA cell extract
4.12.2. 5-(6-Amino-8-(4-(trifluoromethyl)styryl)-9H-purin-9-yl)-
2-((bis(4-methoxy-phenyl)(phenyl)methoxy)methyl)
tetrahydrofuran-3-ol (8) [37,38]
Human U2OS osteosarcoma cells were maintained in low
glucose Dulbecco’s Modified Eagle’s Medium (DMEM, Biological
Industries, Israel) containing 10% fetal bovine serum (FBS, HyClone,
Logan, UT). The clone overexpressing cyclin D1 contained a stable
integration of a GFP-cyclin D1 plasmid under the control of the
CMV promoter. RNA was extracted from wild-type U2OS cells and
cyclin D1 overexpressing cells using TRI Reagent SigmaeAldrich
protocol.
Watereacetonitrile (2: 1, 15 mL) mixture was added through a
septum to
a nitrogen purged vial containing 7 (500 mg,
0.791 mmol, 1 eq), trans-2-(4-fluorophenyl)vinylboronic acid
(214 mg, 0.989 mmol, 1.25 eq), Pd(OAc)2 (11 mg, 0.04 mmol,
0.05 eq), Na2CO3 (252 mg, 2.373 mmol, 3 eq), and TPPTS (113 mg,
For RT-PCR, DNA-free Kit (Ambion) was used to remove genomic
DNA contamination. cDNA (1 mg RNA) was synthesized using the
ReverseAid First Strand cDNA Synthesis Kit (Fermentas) with oligo-
dT as a primer. Semiquantitative RT-PCR was performed using
Eppendorf Thermocycler amplification for 19ꢂ38 cycles (depend-
ing on the saturation level of the genes amplified) using 1 min
denaturation at 94 ꢀC, 1 min annealing at 55 ꢀC for GAPDH and
62 ꢀC for cyclin D1,1 min extension at 72 ꢀC, and 72 ꢀC for 10 min for
final extension. Primers for GAPDH: sense, ACC ACA GTC CAT GCC
ATC AC; antisense, TCC ACC ACC CTG TTG CTG TA. Primers for cyclin
D1: sense, ATA CTC GAG CCA TGG AAC ACC AGC TCC TGT GC;
antisense, GCA ACG AAG GTC TGC GCG TGT TTG C.
Levels of cyclin D1 mRNA expression were three times higher in
U2OS cells overexpressing GFP-cyclin D1 than in wild-type U2OS
cells (WT), as measured by quantitative RT-PCR. Glyceraldehyde-3-
phosphate dehydrogenase (GAPDH) mRNA levels were used as a
control.
Fig. 7. Fluorescence spectra of ds-ON7:RNA probe in the presence of total RNA cell
extract containing high levels of cyclin D1 mRNA.

L. Zilbershtein-Shklanovsky et al. / European Journal of Medicinal Chemistry 79 (2014) 77e88
87
123.3, 117.5, 112.9, 112.9, 85.2, 85.0, 83.4, 70.2, 63.1, 54.8, 40.7, 37.2,
34.6 ppm. MS ESþ m/z: 779 (MHþ); HR MALDI calcd for
C43H42F3N6O5: 779.3163 found 779.3130.
4.12.4. 2-((Bis(4-methoxyphenyl)(phenyl)methoxy)methyl)-5-(6-
((Z)-(dimethylamino) methyleneamino)-8-(4-(trifluoromethyl)
styryl)-9H-purin-9-yl)tetrahydrofuran-3-yl 2-cyanoethyl
diisopropylphosphoramidite (10) [34]
A solution of 9 (800 mg, 1.03 mmol, 1 eq) in dry DCM (9 mL) was
treated with DIPEA (448
diisopropyl-phosphoramidochloridite (276
m
L, 2.57 mmol, 2.5 eq) and 2-cyanoethyl
L, 1.23 mmol, 1.2 eq).
m
After stirring for 22 h at RT, the product was purified on a silica gel
column using hexane/AcOEt 90:10 - 40:60 (þ3% Et3N), as the
eluant, and was obtained as a yellow powder (810 mg, 81%). Mp
78e79 ꢀC; 1H NMR (200 MHz, CDCl3):
d 8.89 (s, 1H), 8.36 (s, 1H),
Fig. 8. Fluorescence spectra of ds-ON7:RNA probe in the presence of total RNA cell
8.09 (d, J ¼ 15.6 Hz, 1H), 7.65e7.33 (m, 7H), 7.22e7.12 (m, 6H), 6.65
(m, 5H), 6.54 (m, 1H), 4.93 (m, 1H), 4.26 (m, 1H), 3.71 (s, 6H), 3.70e
3.46 (m, 7H), 3.30 (s, 3H), 3.22 (s, 3H), 2.76e2.44 (m, 3H), 1.30e1.09
extract containing basal levels of cyclin D1 mRNA.
0.198 mmol, 0.25 eq). The mixture was stirred at 90 ꢀC under N2
atmosphere for 3 h until a brown-yellow residue was obtained. The
reaction was cooled to room temperature and few drops of
concentrated HCl were added (pHw7). The product was purified on
a silica gel column using CHCl3:MeOH, 97:3, as the eluant, and was
obtained as a yellow foam (458 mg, 80%). Mp 134 ꢀC; 1H NMR
(m, 12H) ppm; 13C NMR (150 MHz, CDCl3):
d 158.9, 158.4, 157.7,
151.9, 144.7, 139.3, 136.8, 136.7, 135.9, 135.9, 130.1, 130.0, 128.3,
128.2, 127.7, 127.6, 126.7, 126.7, 126.4, 125.7, 117.5, 116.3, 116.2, 113.0,
86.1, 86.1, 85.4, 85.4, 85.2, 84.0, 84.0, 73.1, 63.4, 63.1, 58.7, 58.5, 58.5,
58.4, 55.1, 45.4, 45.3, 43.3, 43.2, 41.3, 37.3, 37.0, 35.4, 24.7, 24.7, 24.6,
24.6, 24.5, 23.0, 22.9, 20.4, 20.4, 20.3, 20.2 ppm. 31P NMR (81 MHz,
149.4 and 149.2 ppm. MS ESþ m/z: 979 (MHþ); HR MALDI
(300 MHz, DMSO-d6):
d
8.07 (s, 1H), 7.94 (d, J ¼ 8.4 Hz, 2H), 7.84 (d,
CDCl3):
d
J ¼ 15.9 Hz, 1H), 7.75 (d, J ¼ 8.4 Hz, 2H), 7.71 (d, J ¼ 15.9 Hz, 1H), 7.30
(s, 2H), 7.21e7.08 (m, 9H), 6.70 (m, 5H), 5.36 (d, J ¼ 5 Hz, 1H), 4.76
(m, 1H), 3.94 (m, 1H), 3.65 (s, 3H), 3.64 (s, 3H), 3.55 (m, 1H), 3.20 (m,
1H), 3.05 (m, 1H), 2.30 (m,1H) ppm; 13C NMR (150 MHz, DMSO-d6):
calcd for C52H59F3N8O6P: 979.4242, found 979.4150.
Appendix A. Supplementary data
d
157.8, 155.7, 152.3, 149.8, 147.7, 144.9, 139.7, 135.6, 135.5, 133.8,
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2014.03.081.
129.5, 129.5, 129.0, 128.7, 128.4, 127.9, 127.6, 127.5, 126.4, 125.7,
125.7, 125.1, 123.3, 119.4, 117.8, 112.9, 112.9, 85.2, 85.0, 83.3, 79.1,
70.2, 63.1, 54.8, 37.1 ppm. MS ESþ m/z: 746 (MNaþ); HR MALDI
calcd for C40H36F3N5NaO5: 746.2561, found 746.2550.
References
[1] K.K. Karlsen, A. Pasternak, T.B. Jensen, J. Wengel, Pyrene-modified unlocked
nucleic acids: synthesis, thermodynamic studies, and fluorescent properties,
ChemBioChem 13 (2012) 590e601.
[2] F. Hoevelmann, L. Bethge, O. Seitz, Single labeled DNA FIT probes for avoiding
false-positive signaling in the detection of DNA/RNA in qPCR or cell media,
ChemBioChem 13 (2012) 2072e2081.
4.12.3. N0-(9-(5-((bis(4-methoxyphenyl)(phenyl)methoxy)methyl)-
4-hydroxytetra-hydrofuran-2-yl)-8-(4-(trifluoromethyl)styryl)-9H-
purin-6-yl)-N,N-dimethylformimidamide (9) [39]
Dimethylformamidine acetal (0.63 mL, 4.72 mmol, 6.5 eq) was
added to a solution of 64 (525 mg, 0.7262 mmol, 1 eq) in dry DMF
(16 mL). The mixture was stirred for 24 h at 60 ꢀC and the solvent
was removed under vacuum to produce a light-yellow solid
[3] I.V. Astakhova, T.S. Kumar, J. Wengel, Fluorescent oligonucleotides containing
a novel perylene 20-amino-
a-L-LNA monomer: synthesis and analytical po-
tential, Collection of Czechoslovak Chemical Communications 76 (2011)
1347e1360.
[4] S. Latif, I. Bauer-Sardina, K. Ranade, K.J. Livak, P.-Y. Kwok, Fluorescence po-
larization in homogeneous nucleic acid analysis II: 50-nuclease assay, Genome
Research 11 (2001) 436e440.
[5] S.A.E. Marras, S. Tyagi, F.R. Kramer, Real-time assays with molecular beacons
and other fluorescent nucleic acid hybridization probes, Clinica Chimica Acta
363 (2006) 48e60.
[6] D.W. Dodd, R.H.E. Hudson, Intrinsically fluorescent base-discriminating
nucleoside analogs, Mini-Reviews in Organic Chemistry 6 (2009) 378e391.
[7] R.H.E. Hudson, A. Ghorbani-Choghamarani, Selective fluorometric detection of
guanosine-containing sequences by 6-phenylpyrrolocytidine in DNA, Synlett
6 (2007) 870e873.
[8] M. Segal, B. Fischer, Analogues of uracil nucleosides with intrinsic fluorescence
(NIF-analogues): synthesis and photophysical properties, Organic & Biomol-
ecular Chemistry 10 (2012) 1571e1580.
[9] R.W. Sinkeldam, Y. Tor, New fluorescent nucleosides, Collection Symposium
Series 12 (2011) 101e107.
[10] A.S. Wahba, M.J. Damha, R.H.E. Hudson, RNA containing pyrrolocytidine base
analogs: increased binding affinity and fluorescence that responds to hy-
bridization, Nucleic Acids Symposium Series 52 (2008) 399e400.
[11] L.M. Wilhelmsson, A. Holmen, P. Lincoln, P.E. Nielsen, B. Norden, A highly
fluorescent DNA base analogue that forms Watson-Crick base pairs with
guanine, Journal of the American Chemical Society 123 (2001) 2434e2435.
[12] L. Zilbershtein, A. Silberman, B. Fischer, 8-(p-CF3-cinnamyl)-modified purine
nucleosides as promising fluorescent probes, Organic & Biomolecular Chem-
istry 9 (2011) 7763e7773.
(565 mg, 99%). Mp 114 ꢀC; 1H NMR (300 MHz, DMSO-d6):
d 8.93 (s,
1H), 8.32 (s, 1H), 8.01 (d, J ¼ 8.4 Hz, 2H), 7.97 (d, J ¼ 16.5 Hz, 1H),
7.77 (d, J ¼ 8.4 Hz, 2H), 7.74 (d, J ¼ 16.5 Hz, 1H), 7.19e7.05 (m, 9H),
6.68 (m, 5H), 5.39 (d, J ¼ 5.1 Hz, 1H), 4.78 (m, 1H), 3.94 (m, 1H), 3.63
(s, 6H), 3.60e3.40 (m, 2H), 3.23 (s, 3H), 3.16 (s, 3H), 3.06 (m, 1H),
2.30 (m, 1H) ppm; 13C NMR (150 MHz, DMSO-d6):
d 158.6, 157.8,
157.8,157.6,151.9,151.4,149.4,144.8,139.7,135.6,134.8,129.5,129.4,
128.9, 128.6, 128.1, 127.6, 127.5, 126.4, 126.0, 125.6, 125.6, 125.1,
Table 3
Thermal Denaturation Temperatures of cyclin D1-targeting ONs (1e11) measured at
1
m
M in a PBS buffer (pH 7.4).
Tm (ꢀC)
Tm (ꢀC)
Duplex with Duplex with
Duplex with Duplex with
DNA
RNA
DNA
RNA
Unmodified
ON
55
56
ON6 64
65
ON1
ON2
ON3
ON4
59
60
48
50
60
60
60
56
65
60
ON7 61
ON8 64
ON9 64
ON10 65
ON11 65
60
61
62
64
66
[13] K. Evans, D. Xu, Y. Kim, T.M. Nordlund, 2-Aminopurine optical spectra: sol-
vent, pentose ring, and DNA helix melting dependence, Journal of Fluores-
cence 2 (1992) 209e216.
[14] J. Smagowicz, K. Wierzchowski, Lowest excited states of 2-aminopurine,
Journal of Luminescence 8 (1974) 210e232.
ON5

88
L. Zilbershtein-Shklanovsky et al. / European Journal of Medicinal Chemistry 79 (2014) 77e88
[15] M. Menger, T. Tuschl, F. Eckstein, D. Porschke, Mg2þ-dependent conforma-
tional changes in the hammerhead ribozyme, Biochemistry 35 (1996) 14710e
14716.
[16] J.T. Stivers, 2-Aminopurine fluorescence studies of base stacking interactions
at abasic sites in DNA: metal-ion and base sequence effects, Nucleic Acids
Research 26 (1998) 3837e3844.
[17] C.R. Guest, R.A. Hochstrasser, L.C. Sowers, D.P. Millar, Dynamics of mis-
matched base pairs in DNA, Biochemistry 30 (1991) 3271e3279.
[18] G. Kodali, K.A. Kistler, M. Narayanan, S. Matsika, R.J. Stanley, Change in elec-
tronic structure upon optical excitation of 8-vinyladenosine: an experimental
and theoretical study, The Journal of Physical Chemistry A 114 (2010) 256e
267.
[19] M. Daniels, W. Hauswirth, Fluorescence of the purine and pyrimidine bases of
the nucleic acids in neutral aqueous solution at 300ꢀK, Science 171 (1971)
675e677.
[27] W.J. Kent, BLATdthe BLAST-like alignment tool, Genome Research 12 (2002)
656e664.
[28] M. Mizuta, K. Miyata, K. Seio, T. Santa, M. Sekine, Synthesis of fluorescent
cyclic cytosine nucleosides and their fluorescent properties upon incorpora-
tion into oligonucleotides, Nucleic Acids Symposium Series 50 (2006) 19e20.
[29] M. Segal, E. Yavin, P. Kafri, Y. Shav-Tal, B. Fischer, Detection of mRNA of the
cyclin D1 breast cancer marker by a novel duplex-DNA probe, Journal of
Medicinal Chemistry 56 (2013) 4860e4869.
[30] P. Sandin, L.M. Wilhelmsson, P. Lincoln, V.E. Powers, T. Brown, B. Albinsson,
Fluorescent properties of DNA base analogue tC upon incorporation into
DNAdnegligible influence of neighbouring bases on fluorescence quantum
yield, Nucleic Acids Research 33 (2005) 5019e5025.
[31] C. Boiziau, B. Larrouy, B.S. Sproat, J.-J. Tolume, Antisense 20-O-alkyl oligor-
ibonucleotides are efficient inhibitors of reverse transcription, Nucleic Acids
Research 23 (1995) 64e71.
[20] N.B. Gaied, N. Glasser, N. Ramalanjaona, H. Beltz, P. Wolff, R. Marquet,
A. Burger, Y. Mely, 8-vinyl-deoxyadenosine, an alternative fluorescent
nucleoside analog to 20-deoxyribosyl-2-aminopurine with improved proper-
ties, Nucleic Acids Research 33 (2005) 1031e1039.
[32] H. Inoue, Y. Hayase, A. Imura, S. Iwai, K. Miura, E. Ohtsuka, Synthesis and
hybridization studies on two complementary nona(20-O-methyl)ribonucleo-
tides, Nucleic Acids Research 15 (1987) 6131e6148.
[33] S.M. Freier, K.-H. Altmann, The ups and downs of nucleic acid duplex stability:
structure-stability studies on chemically-modified DNA: RNA duplexes,
Nucleic Acids Research 25 (1997) 4429e4443.
[21] A.S. Wahba, A. Esmaeili, M.J. Damha, R.H.E. Hudson,
A single-label
phenylpyrrolocytidine provides molecular beacon-like response
a
reporting HIV-1 RT RNase H activity, Nucleic Acids Research 38 (2010)
1048e1056.
[34] L. Clima, W. Bannwarth, Building-block approach for the straightforward
incorporation of a new FRET (fluorescence-resonance-energy transfer) system
into synthetic DNA, Helvetica Chimica Acta 91 (2008) 165e175.
[35] A.V. Tataurov, Y. You, R. Owczarzy, Predicting ultraviolet spectrum of single
stranded and double stranded deoxyribonucleic acids, Biophysical Chemistry
133 (2008) 66e70.
[22] C. Wanninger-Weiss, H.A. Wagenknecht, Synthesis of 5-(2-pyrenyl)-20-deox-
yuridine as a DNA modification for electron-transfer studies: the critical role
of the position of the chromophore attachment, European Journal of Organic
Chemistry 1 (2008) 64e71.
[23] M. Mizuta, K. Seio, A. Ohkubo, M. Sekine, Fluorescence properties of pyr-
imidopyrimidoindole nucleoside dCPPI incorporated into oligodeoxynucleo-
tides, The Journal of Physical Chemistry B 113 (2009) 9562e9569.
[24] M. Mizuta, K. Seio, K. Miyata, M. Sekine, Fluorescent pyrimidopyrimidoindole
nucleosides: control of photophysical characterizations by substituent effects,
Journal of Organic Chemistry 72 (2007) 5046e5055.
[25] Z. Guo, X. Hao, F. Tan, X. Pei, L. Shang, X. Jiang, F. Yang, The elements of human
cyclin D1 promoter and regulation involved, Clinical Epigenetics 2 (2011) 63e
76.
[26] M. Wei, L. Zhu, Y. Li, W. Chen, B. Han, Z. Wang, J. He, H. Yao, Z. Yang, Q. Zhang,
B. Liu, Q. Gu, Z. Zhu, K. Shen, Knocking down cyclin D1b inhibits breast cancer
cell growth and suppresses tumor development in a breast cancer model,
Cancer Science 102 (2011) 1537e1544.
[36] W. Melhuish, Quantum efficiencies of fluorescence of organic substances:
effect of solvent and concentration of the fluorescent solute1, The Journal of
Physical Chemistry 65 (1961) 229e235.
[37] P. Capek, R. Pohl, M. Hocek, Cross-coupling reactions of unprotected hal-
opurine bases, nucleosides, nucleotides and nucleoside triphosphates with 4-
boronophenylalanine in water. Synthesis of (purin-8-yl)- and (purin-6-yl)
phenylalanines, Organic & Biomolecular Chemistry 4 (2006) 2278e2284.
[38] A. Collier, G.K. Wagner, Suzuki-Miyaura cross-coupling of unprotected hal-
opurine nucleosides in water-influence of catalyst and cosolvent, Synthetic
Communications 36 (2006) 3713e3721.
[39] D. Odadzic, J.B. Bramsen, R. Smicius, C. Bus, J. Kjems, J.W. Engels, Synthesis of
20-O-modified adenosine building blocks and application for RNA interfer-
ence, Bioorganic & Medicinal Chemistry 16 (2008) 518e529.