A D-glucose selective fluorescent assay

Article abstract of DOI:10.1016/S0040-4039(01)02102-5

A D-glucose selective competition assay has been devised using Alizarin Red S and diboronic acid 2 in buffered aqueous solution.

Full text of DOI:10.1016/S0040-4039(01)02102-5

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 303–305
A
D-glucose selective fluorescent assay
Susumu Arimori, Christopher J. Ward and Tony D. James*
Department of Chemistry, University of Bath, Bath BA2 7AY, UK
Received 28 September 2001; accepted 5 November 2001
Abstract—A D-glucose selective competition assay has been devised using Alizarin Red S and diboronic acid 2 in buffered aqueous
solution. © 2002 Elsevier Science Ltd. All rights reserved.
Over the last few years we have been interested in
developing molecular sensors using boronic acids.1–3
The systems we are developing contain a receptor and
reporter (fluorophore or chromophore) as part of a
discrete molecular unit. This is, however, not the only
approach towards boronic acid based sensors. Anslyn
has recently demonstrated that boronic acid receptors
and a separate reporter unit can be used in competitive
assays.4–6 A competitive assay requires that the receptor
and reporter (typically a commercial dye) associate
under the measurement conditions. The receptor–
reporter complex is then selectively dissociated by the
addition of appropriate guests. When the reporter dis-
sociates from the receptor a measurable response is
produced.
plexity of the receptor. Recently, a competitive assay
for -fructose has been developed using two commer-
cial reagents, Alizarin Red S (ARS) and phenylboronic
acid (PBA).8 The system displays
-fructose selectivity,
D
D
which is the inherent selectivity of all monoboronic
acids.3,9
However, a
D
-glucose selective system is much more
-glucose is
desirable. Detection and monitoring of
D
particularly important for the rapidly increasing num-
bers of diabetics. Recent research provides clear evi-
dence that a tight control of the blood glucose levels in
diabetics reduces the risk of long terms complications.
From our previous work with fluorescent PET sensors,
D
-glucose selectivity can be achieved through the cor-
rect spacing of two boronic acids.3
In this communication, we report our work towards the
development of a
D
-glucose fluorescent sensor using
-glucose
ARS in a competitive assay. We based our
D
selective receptor on the successful fluorescent PET
sensor 1.2 The receptor components of 2 are identical to
those of 1, the difference is the lack of a fluorescent
signalling unit (pyrene). Synthesis of 2 was achieved
from readily available starting materials10 (Scheme 1).
We have been exploring the use of coloured and
fluorescent dyes to signal binding with boronic acids.7
Our goal has been to develop effective dye molecules
for competitive assays. We are interested in such com-
petitive systems because they reduce the synthetic com-
Scheme 1. Reagent (yield): (i) benzaldehyde, THF/MeOH, (ii)
NaBH4 (83%, two steps), (iii) 2-(2-bromobenzyl)-1,3,2-dioxa-
borinane, K2CO3, MeCN (35%).
* Corresponding author. Tel.: +44 1225 323810; fax: +44 1225
826231; e-mail: t.d.james@bath.ac.uk
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)02102-5

304
S. Arimori et al. / Tetrahedron Letters 43 (2002) 303–305
UV–vis absorbance titrations of Alizarin Red S (9.6×
10−5 mol dm−3) with PBA and 2 were performed in a
pH 8.21 buffer (52.1 wt% methanol in water with KCl,
0.01000 mol dm−3; KH2PO4, 0.002752 mol dm−3;
Na2HPO4, 0.002757 mol dm−3).11 As the concentration
of boronic acid is increased, a visible colour change
from burgundy to yellow was observed. Fig. 1 shows
the absorption changes with added boronic acid 2.
The absorbance of the free dye at 530 nm decreases as
boronic acid 2 is added, and a new absorbance at 464
nm appears. The stability constants (K) of ARS with
the boronic acids were calculated by fitting the
absorbance at 530 nm versus boronic acid concentra-
tion.12 The calculated K values are 1719 22 for PBA
and 8394 790 for 2. The higher binding constant of 2
with ARS means that a lower concentration of recep-
tor can be used to create the same colour or fluores-
cence response. This is particularly important if a
competitive system is to be developed into a working
sensor.
Figure 2. Absorption spectral changes of ARS and 2 with
increasing concentration of -glucose.
D
We then performed UV–vis absorbance titrations and
fluorescence titrations with
D-glucose and D-fructose in
the presence of PBA (5.0×10−4 mol dm−3) and 2 (5.0×
10−5 mol dm−3) with ARS (9.6×10−5 mol dm−3 for the
UV–vis titrations and 1.0×10−4 mol dm−3 for the
fluorescence titrations) in a pH 8.21 buffer (52.1 wt%
methanol in water with KCl, 0.01000 mol dm−3;
KH2PO4, 0.002752 mol dm−3; Na2HPO4, 0.002757 mol
dm−3).11 As the concentration of
D-glucose and D-fruc-
Figure 3. Fluorescence spectral changes of ARS and 2 with
tose increased the colour changed from yellow (464
nm) to burgundy (530 nm) and the fluorescence emis-
sion intensity at 570 nm (excitation 495 nm) decreased.
The absorption and fluorescence changes with added
increasing concentration of
D-glucose.
D
-glucose are shown in Figs. 2 and 3. The observed
change in absorbance and fluorescence represents
release of free ARS as the saccharide competes for the
boronic acid in solution (Scheme 2). The apparent
stability constants (K) of
D-glucose and D-fructose
with the boronic acids were calculated by fitting the
increase in absorbance at 530 nm and decrease in
fluorescence intensity at 570 nm versus saccharide con-
centration.12 The calculated K values are shown in
Table 1.
Scheme 2.
Table 1. Apparent stability constant K (coefficient of
determination; r2) for saccharide complexes of PBA and
2, in pH 8.21 buffer
Saccharide
PBA
2
2/PBA
Fluorescence (uex 495 nm, uem 570 nm)
D
D
-Glucose
-Fructose
2696 (0.96)
7894 (1.00)
99911 (0.98)
9598 (0.99)
4
1
UV–vis (530 nm)
D
D
-Glucose
-Fructose
1191 (0.99)
7394 (1.00)
6698 (0.99)
14198 (1.00)
6
2
Figure 1. Absorption spectral changes of ARS with increas-
ing concentration of 2.

S. Arimori et al. / Tetrahedron Letters 43 (2002) 303–305
305
Cooperative binding of the two boronic acid groups is
clearly observed as illustrated by the stability constant
differences between PBA and 2. The stability constants
2. Arimori, S.; Bell, M. L.; Oh, C. S.; Frimat, K. A.; James,
T. D. Chem. Commun. 2001, 1836.
3. Hartley, J. H.; James, T. D.; Ward, C. J. J. Chem. Soc.,
K for diboronic acid sensor 2 with
D-glucose are four
Perkin Trans. 1 2000, 3155.
(from fluorescence) and six (from UV–vis) times greater
than with PBA. Whereas the stability constants K of 2
4. Metzger, A.; Anslyn, E. V. Angew. Chem., Int. Ed. Engl.
1998, 37, 649.
5. Lavigne, J. J.; Anslyn, E. V. Angew. Chem., Int. Ed. 1999,
38, 3666.
with
D
-fructose are only one (from fluorescence) and
two (from UV–vis) times stronger than PBA. These
results are not surprising since it is well known that
-glucose easily forms 1:1 cyclic complexes with
6. Cabell, L. A.; Monahan, M. K.; Anslyn, E. V. Tetra-
hedron Lett. 1999, 40, 7753.
7. Arimori, S.; Ward, C. J.; James, T. D. Chem. Commun.
D
diboronic acids, whereas
D
-fructose tends to form 2:1
acyclic complexes with diboronic acids.3
2001, 2019.
8. Springsteen, G.; Wang, B. Chem. Commun. 2001, 1608.
9. Lorand, J. P.; Edwards, J. O. J. Org. Chem. 1959, 24,
769.
In conclusion sensor 2 and ARS produces a very
efficient
D
-glucose assay. Sensor 2 and ARS show an
enhanced response to
D
-glucose when compared to
10. Selected data for 2: mp 110–113°C; m/z (FAB) 1106
simple PBA (six-fold enhancement). Sensor 2 can can
also be used at much lower concentrations (ten times)
than PBA.
([M+H+4(3-HOCH2C6H4NO2)−4(H2O)]+,
100%);
HRSMS found: 1105.47177. C62H62B2N6O12 requires:
1105.464538; lH (300 MHz, CDCl3+CD3OD (a few
drops), Me4Si): 1.04 (4H, bs, NCCCH2), 1.30 (4H, bs,
NCCH2), 2.62 (4H, t, NCH2), 4.09 (4H, s, PhCH2N),
4.21 (4H, s, PhB(OH)CH2N), 7.18 (4H, d, J=3.96 Hz,
Ar-H), 7.25–7.34 (2H, m, Ar-H), 7.43 (10H, m, Ar-H),
7.68 (2H, d, J=7.14 Hz, Ar-H); lC (75 MHz, CDCl3+
CD3OD (a few drops), Me4Si): 22.7, 26.1, 31.8, 55.3, 64.2,
126.5, 126.9, 127.2, 127.8, 128.3, 129.1, 129.3, 130.5,
130.7, 135.0; lB (96 MHz, CDCl3+CD3OD (a few
drops)): 9.64.
Acknowledgements
We wish to acknowledge the Royal Society, the
EPSRC, Beckman–Coulter and Avecia Ltd for support.
We would also like to acknowledge the support of the
University of Bath.
11. Perrin, D. D.; Dempsey, B. Buffers for pH and Metal Ion
Control; Chapman and Hall: London, 1974.
12. The K values were analysed in KaleidaGraph using non-
linear (Levenberg–Marquardt algorithm) curve fitting.
The errors reported are the standard errors obtained
from the best fit.
References
1. Arimori, S.; Bosch, L. I.; Ward, C. J.; James, T. D.
Tetrahedron Lett. 2001, 42, 4553.