Selective fluorescence-based detection of dihydrouridine with boronic acids

Article abstract of DOI:10.1016/j.tetlet.2006.10.150

The first fluorescent sensing system for dihydrouridine detection is presented. Dihydrouridine is the single most frequently occurring post-transcriptional modification in tRNA from bacteria and eukaryotes. A series of 10 boronic acid derivatives was prepared and their fluorogenic behaviours towards dihydrouridine and uridine were investigated. Whereas uridine always quenches fluorescence via π-π stacking interactions, several boronic acid sensors have been found to show substantial fluorescence enhancement upon binding with dihydrouridine.

Full text of DOI:10.1016/j.tetlet.2006.10.150

Tetrahedron Letters 47 (2006) 9253–9256
Selective fluorescence-based detection of dihydrouridine
with boronic acids
Delphine Luvino, Michael Smietana* and Jean-Jacques Vasseur*
LCOBS UMR-5625 CNRS-UM II, Universite´ de Montpellier II, Place Bataillon, 34095 Montpellier, France
Received 26 September 2006; revised 23 October 2006; accepted 25 October 2006
Abstract—The first fluorescent sensing system for dihydrouridine detection is presented. Dihydrouridine is the single most fre-
quently occurring post-transcriptional modification in tRNA from bacteria and eukaryotes. A series of 10 boronic acid derivatives
was prepared and their fluorogenic behaviours towards dihydrouridine and uridine were investigated. Whereas uridine always
quenches fluorescence via p–p stacking interactions, several boronic acid sensors have been found to show substantial fluorescence
enhancement upon binding with dihydrouridine.
ꢀ 2006 Elsevier Ltd. All rights reserved.
The identification and characterization of ribonucleic
acids (RNAs) and their post-transcriptional modifica-
tions are essential for fully understanding their struc-
tural and functional roles. These modifications are
primarily found in stable RNAs such as transfer RNA
(tRNA) and ribosomal RNA (rRNA). Roughly 100
different modified nucleosides have been isolated and
characterized from tRNA.1 After having performed their
task, the tRNA molecules are cleaved to nucleosides.
Whereas, unmodified nucleosides are recycled for bio-
synthesis of new RNA, modified nucleosides cannot be
reutilized or further degraded but are either metabolized
or excreted intact in urine. While healthy adults have
stable baseline levels of modified nucleoside excretion,
numerous works have shown that an elevated amount
of modified nucleosides is related to carcinoma.2
is less rigid than the C-30-endo conformer, allowing con-
formational flexibility of RNA where tertiary interac-
tions and loop formation must be simultaneously
accommodated.3
The use of modified nucleosides excreted in urine as
potential tumor markers has prompted the development
of several analysis techniques. Traditionally urinary
nucleosides are analyzed and quantified using HPLC
analysis techniques. Most recently MALDI-TOF MS
methods have been reported.4 Because dihydrouridine,
unlike all other natural or modified nucleosides, pos-
sesses no significant chromophore, HPLC analysis using
UV detection is not practical due to poor sensitivity.5 On
the other hand, analysis using fluorescent sensors offers
the advantages of being convenient and highly sensitive.
Herein, we wish to report our efforts in developing the
first fluorescent detection method of dihydrouridine.
The single most frequently occurring post-transcrip-
tional modification in tRNA from bacteria and eukar-
yotes is dihydrouridine D in which the C5–C6 double
bond of the uracil nucleobase is hydrogenated. Despite
the widespread occurrence of dihydrouridine, little is
known about its functional roles. It seems that dihydro-
uridine favors the C-20-endo sugar conformation, which
An ideal fluorescent monitoring molecular system for
specific recognition of nucleobases would display an
enhanced signal upon binding. Unfortunately, fluo-
rescence quenching by purine and pyrimidine nucleo-
bases is a well known phenomenon which has been
studied by various groups.6 Usually, the nucleobases
exhibit a variable degree of quenching, with guanosine
being the most efficient, followed by adenosine, cytidine
and thymidine–uridine. This has notably been used to
monitor PCR,7 to design probes for protein–ligand
interaction8 or for the design of a modified molecular
beacon quenched by consecutive guanosines.9
Keywords: Fluorescence detection; Modified nucleosides; Tumor
markers.
*
Corresponding authors. Tel.: +33 4 67143837; fax: +33 4 67042029
(M.S.); e-mail addresses: msmietana@univ-montp2.fr; vasseur@
univ-montp2.fr
0040-4039/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2006.10.150

9254
D. Luvino et al. / Tetrahedron Letters 47 (2006) 9253–9256
Table 1. Synthesis of fluorescent phenylboronic acid derivatives
NMR and X-ray crystallography studies have shown
that the nonplanar dihydrouridine resists stacking
because of the absence of the C5–C6 double bond.10
Therefore, we thought of taking advantage of the loss
of aromaticity and thus of its quenching ability, upon
saturation of the 5,6-double bond in dihydrouridine,
for its specific fluorescence detection. This was done
using boronic acids via formation of a boronate ester
with the ribose 2,3-cis-diol group as depicted in Scheme 1.
HO
OH
B
HO
OH
B
H
N
NaBH3CN
+
R1
R2
CHO
N R1
R2
1a-c
2a-j
Entry
Starting
materiala
Product
Compound
(Yield %)b
R1
R2
The design and synthesis of boronic acid-based fluores-
cent sensors for saccharide detection have been the sub-
ject of intense research in recent years.11 There is general
agreement that boronic acids covalently react with
1,2 or 1,3 diols to form reversible five- or six-membered
cyclic esters. It has been demonstrated that these sensors
show an increase in fluorescence intensity upon binding
with a diol through different and still studied signalling
mechanisms.12 To date, and presumably, because of
their quenching abilities, attempts to detect nucleosides
or nucleotides with boronic acids have been made by
using dye receptors,13 polycations,14 or molecular
imprinting,15 but never using fluorescent reporters. In
this project, we chose to use a series of 10 fluorescent
boronic acids which were synthesized according to
reported procedures by reductive amination of readily
available 2-, 3- or 4-formylbenzeneboronic acid (com-
pounds 1a–c, respectively), with the corresponding
amine (Table 1). In this study, the sensors have been
classified depending on their respective signalling mech-
anism, as ICT (internal charge transfer) fluorophore
(Table 1, compounds 2a–f), and PET (photoinduced
electron transfer) fluorophores (Table 1, compounds
2g–2j). In this latter case, it should be noted, that recent
reports dispute the PET mechanism over a competing
solvolysis process.12
1
2
3
4
5
6
1a
1b
1c
1a
1b
1c
Phenyl
Phenyl
Phenyl
2-Fluorenyl
2-Fluorenyl
2-Fluorenyl
H
H
H
H
H
H
2a (70)
2b (62)
2c (78)
2d (80)
2e (55)
2f (53)
CH2
CH2
7
8
1a
1a
H
2g (31)
2h (87)
Me
CH2
CH2
9
1a
1a
H
2i (25)
10
Me
2j (80)
a Compounds 1a–c refer to 2-, 3- or 4-formylbenzeneboronic acid,
respectively.
b Isolated yields.
The fluorescence titrations of sensors 2a–j (2 · 10ꢀ5 M)
were carried out in a pH 8.21 buffer (52.1% MeOH/
H2O phosphate buffer solution) in the presence of
uridine or dihydrouridine (0–50 mM).16 To our satisfac-
tion, all candidates stood out showing modest to good
fluorescence intensity enhancements upon addition of
dihydrouridine. Fluorescence intensity enhancements
induced by D and quenching efficiencies of U calculated
from these titrations for sensors 2a–j are given in Table
2. In the ICT series (Table 2, entries 1–6), there does not
HO
R
O
HO
HO
OH
R
B
O
Table 2. Association constants (Ka) and fluorescence intensity changes
of boronic acids with D and U
O
O
B
Ka (Mꢀ1 a
)
DID (fold)
QE (%)
b
c
OH
OH
Entry
Sensor
F
1
2
3
4
5
6
7
8
9
2a
2b
2c
2d
2e
2f
64
30
1.5
2
86
92
82
71
54
11
21
F
Weak Fluorescence
162
125
146
162
383
981
224
479
5.5
2.9
2.8
5.1
1.39
2.65
1.6
5.2
O
NH
Strong Fluorescence
R =
N
H
O
2g
2h
2i
NDd
83
O
10
2j
29
NH
R =
No Fluorescence
a Average of at least two measurements with r2 P 0.99.
N
H
O
b Fluorescence intensity changes (I/I0) upon addition of 50 mM of
dihydrouridine D.
Scheme 1. Fluorescent response of boronic acids upon binding with
uridine and dihydrouridine with different fluorophores F.
c Quenching efficiencies upon addition of 10 mM of uridine U.
d No quenching detected.

D. Luvino et al. / Tetrahedron Letters 47 (2006) 9253–9256
9255
missing and the boronic acid groups in 2h and 2j could
theoretically complex the residual 30,50-dihydroxy
group. On the other hand, in DMP, the 50-OH position
is phosphorylated and the 2,3-diol remains the sole com-
plexation site. We found that the affinity of 2h and 2j
with dD dropped severely compared to dihydrouridine.
Whereas a low Ka of 11.6 Mꢀ1 was observed with 2j,
no spectral change was induced with 2h, thus indicating
the crucial role of the 20-OH group in the complexation.
In contrast, association constants of 2h and 2j obtained
with DMP are, respectively, one half (297 Mꢀ1) and
one third (214 Mꢀ1) of that for D presumably due to
the enhanced hydrophilicity caused by the phosphate
group.13 These results clearly confirm the key role
displayed by the 2,3-diol unit in the complexation.
5000
4000
3000
2000
1000
0
50 mM
0 mM
410
360
370
380
390
400
420
430
440
Wavelength (nm)
Figure 1. Fluorescence spectra change of 2j (2 · 10ꢀ5 M) with different
concentrations of dihydrouridine D (0–50 mM) in 52.1% MeOH pH
8.21 phosphate buffer. kex = 343 nm.
Finally, the binding of D in the presence of U was
assessed by a competitive fluorescence experiment. Sen-
sor 2j was choosen for this titration as it shows a 30%
fluorescence quenching with 10 mM of U, a high affinity
for D and one of the largest fluorescence intensity
enhancement. The results show less sensitivity to D in
the presence of U in agreement with the observed
(quenched) complex formation with uridine. However,
addition of a molar D/U ratio of 0.15 induced a gradual
fluorescence increase indicating a high selectivity for D
(Fig. 2). Moreover, with an equimolar amount of D
and U, the fluorescence intensity increased up to almost
2 times the background fluorescence. The 5-fold fluores-
cence intensity increase observed with D is, however,
never reached in the presence of U.
seem to be any correlation to explain the differences ob-
served in fluorescence intensity changes with the nature
of the fluorophore. However, the para position appears
to induce a superior emission intensity increase as shown
by sensors 2c and 2f (5.5 and 5.1, respectively). In gen-
eral, the fluorescence intensity increases are comparable
with those obtained in the PET series. In this case, the
largest intensity increase is observed with pyrene deriva-
tive 2j (Table 2, entry 10). Addition of dihydrouridine at
pH 8.21, induced a 5.2-fold intensity increase at 377 nm
in a concentration-dependent manner (Fig. 1).
On the other hand, all sensors but one induced a gradual
fluorescence intensity decrease upon addition of uridine
(11–92% with 10 mM of U, see Supplementary data). In
fact, the anthracene-based sensor 2h (Table 2, entry 8),
initially developed by the Shinkai group, binds strongly
with D, but does not exhibit any important fluorescence
change upon addition of 50 mM of D.17 Most impor-
tantly, upon addition of U, the fluorescence intensity
of sensor 2h first augments and then diminishes slowly
as the concentration of U increases. A high concentra-
tion of U (85 mM) is needed to recover the background
fluorescence making this sensor inappropriate for
dihydrouridine detection in the presence of uridine.
The present study has demonstrated the usefulness of
boronic acid fluorescent sensors to detect the highly
important dihydrouridine modified nucleoside. This
was done by taking advantage of the reduction of the
5,6-double bond in uridine. Among the 10 sensors
tested, five have been found to show significant fluores-
cence enhancement (Df > 5). The transformation of the
boronic acid function into a boronate ester selectively
resulted in a pronounced OFF–ON-type fluorescent sig-
naling behaviour. To the best of our knowledge, this is
the first time that synthetic receptors have been used
to fluorescently label a modified nucleobase. Moreover,
these results revealed the sensitivity and selectivity of
To examine the binding of sensors 2a–j with D in a more
quantitative fashion, the stability constants (Ka) were
evaluated using the Benesi–Hildebrand equation for
1:1 stoichiometric binding (Table 2). In all cases a linear
relationship is obtained when 1/(I ꢀ I0) versus 1/([D]0) is
plotted (see Supplementary data).18 As it can be seen,
the largest affinities of these various boronic sensors
with dihydrouridine are obtained with PET sensors, 2h
6
D
5
D + U
4
3
2
1
0
and 2j being the stronger ones (Ka = 981 and 479 Mꢀ1
,
respectively). The solvation (size of the aromatic p-sur-
face) and the steric crowding (number of peri-hydrogens
of each fluorophore) need to be considered to interpret
these values. Indeed, these binding constants are higher
than those observed with phenylboronic acids and
ribose.19
-3.5
-3
-2.5
-2
-1.5
-1
log [D]
In order to specify the binding requirements, we com-
pared the binding constants of compounds 2h and 2j
with 20-deoxydihydrouridine dD, and dihydrouridine
50-monophosphate DMP. In dD, the 20-OH group is
Figure 2. Fluorescence titration of boronic acid 2j (2 · 10ꢀ5 M) with
dihydrouridine in the presence of uridine (10 mM) in 52.1% MeOH pH
8.21 phosphate buffer.

9256
D. Luvino et al. / Tetrahedron Letters 47 (2006) 9253–9256
3911; (d) Kawai, K.; Wata, Y.; Ichinose, N.; Majima, T.
Angew. Chem., Int. Ed. 2000, 39, 4327; (e) Kawai, K.;
Yokoohji, A.; Tojo, S.; Majima, T. Chem. Commun. 2003,
2840; (f) Amann, N.; Pandurski, E.; Fiebig, T.; Wagen-
knecht, H.-A. Angew. Chem., Int. Ed. 2002, 41, 2978; (g)
Wagenknecht, H.-A. Angew. Chem., Int. Ed. 2003, 41,
2454.
these types of sensors allowing 2j to probe small amount
of D in a U/D mixture. These fluorescent reporter com-
pounds might be very useful for subsequent application
in the systematic study of the dihydrouridine urinary
level in cancer patients.
7. Crockett, A. O.; Wittwer, C. T. Anal. Biochem. 2001, 290,
89.
Acknowledgments
8. Dhar, G.; Bhaduri, A. J. Biol. Chem. 1999, 274, 14568.
9. Kurata, S.; Kanagawa, T.; Yamada, K.; Torimura, M.;
Yokomaku, T.; Kamagata, Y.; Kurane, R. Nucleic Acids
Res. 2001, 29, e34.
10. (a) Sundaralingam, M.; Rao, S. T.; Abola, J. Science 1971,
172, 725; (b) Suck, D.; Saenger, W.; Zechmeister, K. Acta
Crystallogr., Sect. B 1972, 28, 596; (c) Deslauriers, R.;
Lapper, R. D.; Smith, I. C. P. Can. J. Biochem. 1971, 49,
1279.
The authors wish to thank Dr. S. Arseniyadis for helpful
´
discussions. The CNRS and the Region Languedoc-
Roussillon are gratefully acknowledged for a doctoral
fellowship to D.L.
Supplementary data
11. (a) James, T. D. In Boronic Acids; Hall, D. G., Ed.; Wiley-
VCH: Weinheim, 2005, Chapter 12; (b) Yang, W.; Gao,
X.; Wang, B. In Boronic Acids; Hall, D. G., Ed.; Wiley-
VCH: Weinheim, 2005, Chapter 13; (c) Hartley, J. H.;
James, T. D.; Ward, C. J. J. Chem. Soc., Perkin Trans. 1
2000, 3155; (d) James, T. D.; Shinkai, S. Top. Curr. Chem.
2002, 218, 159.
1
Synthesis procedures and H and 13C NMR of D, dD,
DMP and compounds 2a–j. Experimental fluorescence
measurements. Supplementary data associated with this
article can be found, in the online version, at doi:10.1016/
j.tetlet.2006.10.150.
12. (a) James, T. D.; Sandanayake, K. R. A. S.; Inguchi, R.;
Shinkai, S. J. Am. Chem. Soc. 1995, 117, 8982; (b) Ni, W.;
Kaur, G.; Springsteen, G.; Wang, B.; Franzen, S. Bioorg.
Chem. 2004, 32, 571; (c) Zhu, L.; Shabbir, S. H.; Gray, M.;
Lynch, V. M.; Sorey, S.; Anslyn, E. V. J. Am. Chem. Soc.
2006, 128, 1222.
13. Takeuchi, M.; Taguchi, M.; Shinmori, H.; Shinkai, S. Bull.
Chem. Soc. Jpn. 1996, 69, 2613.
14. (a) Patterson, S.; Smith, B. D.; Taylor, R. E. Tetrahedron
Lett. 1997, 38, 6323; (b) Kanekiyo, Y.; Naganawa, R.;
Tao, H. Chem. Commun. 2004, 1006.
15. (a) Kanekiyo, Y.; Sano, M.; Iguchi, R.; Shinkai, S. J.
Polym. Sci. A 2000, 38, 1302; (b) Kanekiyo, Y.; Ono, Y.;
Inoue, K.; Sano, R.; Shinkai, S. J. Chem. Soc., Perkin
Trans. 2 1999, 557; (c) Zayats, M.; Lahav, M.; Kharito-
nov, A. B.; Willner, I. Tetrahedron 2002, 58, 815.
16. (a) Bosch, L.; Mahon, M. F.; James, T. D. Tetrahedron
Lett. 2004, 45, 2859; (b) Arimori, S.; Consiglio, G. A.;
Phillips, M. D.; James, T. D. Tetrahedron Lett. 2003, 44,
4789.
17. James, T. D.; Sandanayake, K. R. A. S.; Shinkai, S. J.
Chem. Soc., Chem. Commun. 1994, 477.
18. Benesi, H.; Hildebrand, J. H. J. Am. Chem. Soc. 1949, 71,
2703.
References and notes
1. Helm, M. Nucleic Acids Res. 2006, 34, 721.
2. Schram, K. H. Mass Spectrom. Rev. 1998, 17, 131.
3. Dalluge, J. J.; Hashizume, T.; Sopchik, A. E.; McCloskey,
J. A.; Davis, D. R. Nucleic Acids Res. 1996, 24, 1073.
4. Kammerer, B.; Frickenschmidt, A.; Gleiter, C. H.; Laufer,
S.; Liebich, H. J. Am. Soc. Mass Spectrom. 2005, 16, 940.
5. (a) Cerutti, P.; Ikeda, K.; Wiktop, B. J. Am. Chem. Soc.
1965, 87, 2505; (b) Gehrke, C. W.; Kuo, K. C. J.
Chromatogr. 1989, 477, 3; (c) Magrath, D. I.; Shaw, D.
C. Biochem. Biophys. Res. Commun. 1967, 26, 32; (d)
Molinaro, M.; Sheiner, L. B.; Neelson, F. A.; Cantoni, G.
L. J. Biol. Chem. 1968, 243, 1277; (e) Jacobson, M.;
Hedgcoth, C. Anal. Biochem. 1970, 34, 459; (f) Randerath,
K.; Gupta, R. C.; Randerath, E. Meth. Enzymol. 1980, 65,
638; (g) Johnson, J. D.; Horowitz, J. Biochim. Biophys.
Acta 1971, 247, 262; (h) Dalluge, J. J.; Hashizume, T.;
McCloskey, J. A. Nucleic Acids Res. 1996, 24, 3243.
6. (a) Marras, S. A. E.; Russel Kramer, F.; Tyagi, S. Nucleic
Acids Res. 2002, 30, e122; (b) Seidel, C. A. M.; Schulz, A.;
Sauer, M. H. M. J. Phys. Chem. 1996, 100, 5541; (c)
Marquez, C.; Pishel, U.; Nau, W. M. Org. Lett. 2003, 5,
19. Springsteen, G.; Wang, B. Tetrahedron 2002, 58, 5291.