Novel fluorescence sensor for 'small' saccharides

Article abstract of DOI:10.1039/a606552j

The small binding pocket of compound 2 can exclude 'large' saccharides such as D-glucose, whereas 'small' saccharides such as D-sorbitol are strongly bound as 1:1 complexes; the binding events can be sensitively monitored by changes in the fluorescence in

Full text of DOI:10.1039/a606552j

View Article Online / Journal Homepage / Table of Contents for this issue
Novel fluorescence sensor for ‘small’ saccharides
Tony D. James,*a Hideyuki Shinmorib and Seiji Shinkaib
a The University of Birmingham, School of Chemistry, Edgbaston, Birmingham, UK B15 2TT
b Department of Chemical Science and Technology, Faculty of Engineering, Kyushu University, Fukuoka 812, Japan
The small binding pocket of compound 2 can exclude ‘large’
saccharides such as D-glucose, whereas ‘small’ saccharides
such as D-sorbitol are strongly bound as 1:1 complexes; the
binding events can be sensitively monitored by changes in
the fluorescence intensity.
fluorescence intensity (I/I0; emission maximum 347 nm) is
illustrated in Fig. 1. In the absence of saccharides, the I/I0 is
nearly constant from pH 1.5 to 8.0. Judging from the pH
dependence of 1 a change from 2H+ to 2 (cf. Scheme 2) should
cause a large decrease in the fluorescence intensity. This means
that deprotonation of 2H+ occurs in a more acidic region. On the
other hand, the I/I0 gradually decreases above pH 8.0. This
reflects the dissociation of boronic acid groups to 22 and
The development of boronic acid receptors for saccharides has
recently gained much attention.1 One problem with many of
these systems is that they only function in basic aqueous media.
Wulff was the first to report that a neighbouring nitrogen
enhances the formation of boronate esters even under neutral
pH conditions.2 In a series of recent papers we employed the
interaction between boronic acid and amine3–7 to create
photoinduced electron transfer (PET)8,9 sensory systems for
saccharides. When saccharides form cyclic boronate esters with
boronic acids, the acidity of the boronic acid is enhanced10 and
therefore the Lewis acid–base interaction between the boronic
acid and the tertiary amine is strengthened. The strength of this
acid-base interaction modulates the PET from the amine (acting
as a quencher) to anthracene (acting as a fluorophore). These
compounds show increased fluorescence at pH 7.77 through
suppression of the photoinduced electron transfer from nitrogen
to anthracene on saccharide binding; a direct result of the
stronger boron–nitrogen interaction.3–7
222
.
In the presence of saccharides, the I/I0 increases from pH 4.0
and reaches a maximum at around pH 8.0. This change is
ascribed to the intensified B–N interaction caused by the
saccharide binding to the boronic acid group. Since the largest
I/I0 increase was observed at around pH 8.0, we determined the
association constants (K) at pH 8.0 from plots of I/I0 vs.
NH2
Br
B
Br
O
B
O
B
+ 2.2
O
With diboronic acid 1 glucose selectivity was predicted from
CPK models and then observed experimentally.4 With other
diboronic acids chiral discrimination5 and metal ion alloster-
ism6 have been achieved. In these systems the formation of a
1:1 intramolecular complex which is classified as a saccharide-
containing macrocycle was shown to be the important fluores-
cent species and the origin of the circular dichroism (CD)
activity. With this work we have reduced the size of the cleft in
order to modulate the selectivity of the boronic acid unit
towards ‘small’ saccharides. From CPK models glucose cannot
easily bind as a cyclic 1:1 complex with diboronic acid 2 and
should be excluded from the binding pocket.
Br
i
B(OH)
2 B(OH)2
N
2
Scheme 1 Reagents: i, K2CO3, MeCN, heat, 70%
Compound 2 was synthesised from 1-aminomethylnaph-
thalene according to Scheme 1 and characterised using IR and
1H NMR spectral evidence and elemental analysis.
1.6
The absorption spectrum (25 °C, H2O–MeOH, 300:1 v/v)
and lmax were not affected over a large pH range (pH 1.7–11.8).
The pKa values were determined using fluorescence spectro-
scopy (excitation 296 nm). The pH dependence of the relative
1.4
1.2
1.0
I
I
Me
0.8
0.6
N
B
B(OH)2
HO OH
OH
B(OH)2
HO
O
N
0
2
4
6
8
10
12
14
HO OH
B
pH
N
Fig. 1 pH Dependence of the relative fluorescence intensity (I/I0 at 347 nm)
Me
in the absence and presence of saccharides: [2] = 1.00 3 1025 mol dm23
,
[saccharide] = 0.10 mol dm23, 25 °C, H2O–MeOH, 300:1 v/v buffered
with HCl and NaOH, excitation 296 nm: (5) d-fructose, («) pentaery-
thritol, (2) d-threitol, ( ~ ) d-glucose, (+)-glycerine, (-) no saccharide
OH
OH
1
2
3
Chem. Commun., 1997
71

View Article Online
–
–
–
B(OH)2
B(OH)2
B(OH)
B(OH)
B(OH)2
2 B(OH)2
2 B(OH)2
B(OH)2
H
N
N
N
N
+
Naph =
Naph
Naph
Naph
Naph
2H+
2
2–
22–
Scheme 2
the two diols, Dn = 0 in Table 1), do a large fluorescence
increase and a large K value result. Glycerine and propane-
1,3-diol, which have only one binding site, cannot enhance the
fluorescence intensity. d-Glucose (Dn = 1) and mannitol
derivative 3 (Dn = 2), in which two binding sites are separated
by one or two carbons, cannot enhance the fluorescence either.
Other compounds which can enhance the fluorescence intensity
have two (or more than two) binding sites and Dn = 0 without
exception. Even in those compounds with a choice of binding
site, Dn = 0–2, the most stable complex should be the one with
Dn = 0.
1.6
1.4
1.2
I
I
The formation of the cyclic structures is further supported by
the fact that the complexes become CD active. For example, 2
(1 3 1025 mol dm23) in the presence of sorbitol (0.10
mol dm23) gave [q]max 22.16 3 104 deg cm2 dmol21 for d-
sorbitol and 2.16 3 104 deg cm2 dmol21 for l -sorbitol at 228
nm (25 °C, pH 8.0 with 50 mmol dm23 phosphate buffer).
In conclusion we have clearly demonstrated that the selectiv-
ity of a diboronic acid cleft towards saccharides can be modified
in a controlled manner by the correct spacing of two boronic
acid units. This work illustrates the power of the synthetic
saccharide sensor. Unlike enzymatic systems, selectivity can be
designed for one particular saccharide. This offers the possibil-
ity of monitoring the concentrations of biologically important
saccharides which are less abundant than d-glucose, in a variety
of industrial and medicinal applications.
1.0
0.8
0.00
0.02
0.04
0.06
[Sugar] / M
0.08
0.10
0.12
Fig. 2 Plots of I/I0 vs. [saccharide] for selected saccharides: (5) d-fructose,
(«) pentaerythritol, (2) d-threitol, ( ~ ) d-glucose, (+)-glycerine, (-) no
saccharide
Table 1 Association constants (K), number of binding sites and number of
carbon separating two binding sites (Dn) at 25 °C and pH 8.0 in H2O–
MeOH, 300:1 v/v
Sugar
log K
Binding site
Dn
T. D. J. wishes to acknowledge the Royal Society for support
through the award of a University Fellowship.
d-Sorbitol
d-Fructose
Dulcitol
2.54
2.52
2.4
2
2
2
2
2
2
2
2
2
2
2
2
2
1
1
2
0–2
0–1
0–2
0–1
0–2
0
0–1
0
0–1
0
References
Xylitol
2.23
2.15
2.11
1.97
2.03
2.13
1.91
1.50
1.45
a
1 G. Deng, T. D. James and S. Shinkai, J. Am. Chem. Soc., 1994, 116,
4567 and references cited therein; J. Yoon and A. W. Czarnik, J. Am.
Chem. Soc., 1992, 114, 5874; G. T. Morin, M. P. Hughes, M.-F. Paugam
and B. D. Smith, J. Am. Chem. Soc., 1994, 116, 8895 and references
cited therein.
2 G. Wulff, Pure Appl. Chem., 1982, 54, 2093.
3 T. D. James, K. R. A. S. Sandanayake and S. Shinkai, J. Chem. Soc.,
Chem. Commun., 1994, 477.
4 T. D. James, K. R. A. S. Sandanayake and S. Shinkai, Angew. Chem.,
Int. Ed. Engl., 1994, 33, 2207.
5 T. D. James, K. R. A. S. Sandanayake and S. Shinkai, Nature, 1995, 374,
345.
6 T. D. James and S. Shinkai, J. Chem. Soc., Chem. Commun., 1995,
1483.
7 T. D. James, P. Linnane and S. Shinkai, Chem. Commun., 1996, 281; (b)
T. D. James, K. R. A. S. Sandanayake and S. Shinkai, Angew. Chem.,
Int. Ed. Engl., 1996, 108, 2038.
d-Mannitol
Pentaerythritol
Ribitol
d-Threitol
d-Arabitol
d-Ribose
d-Fucose
Erythritol
d-Glucose
Glycerine
Propane-1,3-diol
3b
0
0
1
—
—
2
a
a
a
a The K could not be determined because of the small fluorescence change.
b H2O–MeOH, 2:1 v/v (because of low solubility in water).
8 R. A. Bissel, A. P. de Silva, H. Q. N. Gunaratna, P. L. M. Lynch,
G. E. M. Maguire, C. P. McCoy and K. R. A. S. Sandanayake,
Top. Curr. Chem., 1993, 168, 223.
9 A. W. Czarnik, Fluorescent Chemosensors for Ion and Molecular
Recognition, ACS Books, Washington, 1993.
saccharide concentration (Fig. 2). The results are summarised in
Table 1.
Examination of Table 1 reveals two important features of
compound 2 as a PET sensor: that is, (i) 2 shows selectivity
towards ‘small’ saccharides and (ii) only when they possess two
binding sites (i.e. two diol units associative with a boronic acid)
and they are not separated (i.e. the number of carbons separating
10 J. P. Lorand and J. D. Edwards, J. Org. Chem., 1959, 24, 769.
Received, 24th September 1996; Com. 6/06552J
72
Chem. Commun., 1997