Designed boronate ligands for glucose-selective holographic sensors

Article abstract of DOI:10.1002/chem.200600442

In this study, 2-acrylamido-phenylboronate (2-APB) was synthesised and its ability to bind with glucose was investigated both in solution and when integrated into a holographic sensor. Multiple forms of 2-APB, resulting from the neighbouring effect of the

Full text of DOI:10.1002/chem.200600442

FULL PAPER
DOI: 10.1002/chem.200600442
Designed Boronate Ligands for Glucose-Selective Holographic Sensors
Xiaoping Yang,[a] Mei-Ching Lee,[b] Felicity Sartain,[a] Xiaohan Pan,[a] and
Christopher R. Lowe*[a]
Abstract: In this study, 2-acrylamido-
phenylboronate (2-APB) was synthes-
ised and its ability to bind with glucose
was investigated both in solution and
spectrometry. It was found that 2-APB
predominantly adopts zwitterionic
and generates a five-membered-ring
a
complex. Furthermore, a 2-APB-based
holographic sensor displayed a signifi-
cant response to glucose with little in-
terference from lactate, and with no
dependence on pH in the physiological
pH range. These features suggest that
the new ligand 2-APB is a potential
candidate for the development of glu-
cose-selective sensors.
tetrahedral form at physiological pH
values. The complex formation of 2-
APB with glucose and lactate was in-
vestigated in DMSO; 2-APB favours
binding with glucose rather than lactate
when integrated into
a holographic
sensor. Multiple forms of 2-APB, re-
sulting from the neighbouring effect of
the amido group with the boronic acid
À
through an intramolecular B O-coordi-
Keywords: boronic acid · glucose ·
holography · pH · sensors
nated interaction, were shown to exist
in solution by using multinuclear NMR
Introduction
to boronate-based glucose sensors has rarely been men-
tioned.[4]
It is well known that boronates rapidly and reversibly form
cyclic esters in alkaline media with diols such as glucose. A
number of novel boronate analogues containing functional
groups have been designed to improve their affinity to glu-
cose at physiological pH values.[1–3] Such analogues have
been included in glucose-responsive cyclodextrin, fluores-
cence, coloured, electrochemical, polymeric and imprinted
receptors, although most of the novel boronate receptors re-
ported still display a pH dependency in the physiological
range, both in solution as well as in the sensors. Further-
more, interference from the metabolic end product, lactate,
Holographic glucose sensors based on 3-acrylamidophe-
nylboronate (3-APB) were recently developed in our
group.[4] The holographic sensors are based on planar, small-
volume polymer hydrogels containing silver halide, which
acts as a holographic recording material. A reflection holo-
gram is recorded within the polymer hydrogel matrix, pro-
ducing an optical sensor that only reflects a narrow band of
wavelengths when illuminated with white light. Holographic
glucose sensors have been fabricated from 3-APB incorpo-
rated into acrylamide-based “smart” hydrogels that respond
to glucose by changing their swelling state or degree of hy-
dration upon glucose binding with 3-APB. The change in
the volume of the polymer causes a change in the wave-
length of the light diffracted by the holographic grating
within the hydrogel. The narrow band of wavelengths dif-
fracted is determined by the spacing between silver (Ag0)
nanoparticle fringes. Thus, the hologram acts as a mono-
chromatic mirror, in that light with a wavelength of approxi-
mately double the distance between the fringes will emerge
in phase at a specific angle and constructive interference
occurs. The diffraction wavelength of the hologram is gov-
erned by Braggꢀs law. The results to date show that 1) The
sensor displays reversible changes in diffraction wavelength
as a function of monosaccharide concentration at physiologi-
cal pH values; 2) maximum sensitivity to glucose is observed
at a functional monomer concentration of 20–25 mol%;
[a]Dr. X. Yang, Dr. F. Sartain, X. Pan, Prof. C. R. Lowe
Institute of Biotechnology
University of Cambridge
Tennis Court Road, Cambridge, CB2 1QT (UK)
Fax : (+44)1223-334-162
E-mail: crl1@biotech.cam.ac.uk
+
[b]Dr. M.-C. Lee
PA Consulting Group
123 Buckingham Palace Road, London SW1W 9SR (UK)
[+]A former member of Professor Loweꢀs group.
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author. It contains select-
1
ed H, 13C and 11B NMR spectra and FTIR spectra of 2-APB and 3-
APB. It also contains more details of the synthesis of the copolymer
films of 2-APB and 3-APB and the hologram construction.
Chem. Eur. J. 2006, 12, 8491 – 8497
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
8491

3) the potential for using this holographic sensor to detect
real-time changes in bacterial cell metabolism has been
demonstrated by monitoring the vegetative metabolism of
Bacillus subtilis. However, such boronate-based glucose sen-
sors display strong interference from lactate. This feature
makes them especially unsuitable for application in glucose-
selective holographic sensors for monitoring tear fluid and
bacterial culture media, because lactate is a metabolic by-
product that is present at high concentrations in both these
fluids. Thus, there is a need for upgrading the current 3-
APB-based holographic glucose sensor into one that dis-
plays little or no response to lactate as well as a limited pH
sensitivity to binding affinity within the physiological pH
range. To achieve this, one solution is to design new ligands
that can selectively bind with glucose and obviate lactate in-
terference. It is known that the tetrahedral form of the bor-
onic acid favours binding with glucose, whereas the trigonal
form preferentially binds with lactate.[5] Thus, if it is possible
to generate a tetrahedral boronate analogue as the domi-
nant species at physiological pH values, then the interfer-
ence from lactate is likely to be diminished. Such a tetrahe-
dral form can be induced by designing boronic acids with
ference from lactate and a glucose-binding profile that was
almost independent of pH over the physiological pH range,
both of which are desired features for development of glu-
cose-selective sensors.
Experimental Section
Materials: All chemicals used were of analytical grade unless otherwise
stated. 2-Aminophenylboronic acid and 3-aminophenylboronic acid were
purchased from Avocado Research Chemicals. Acryloyl chloride was pur-
chased from Sigma Chemical Co. All other chemicals related to holo-
gram preparation and holographic measurement are the same as descri-
bed in previous work.[4] All deuterated solvents, acids and bases were
purchased from Cambridge Isotope Laboratories, Inc.
Equipment: A 400 MHz JEOL NMR instrument was used for NMR
analysis. A glass pH microelectrode (combined with the NMR spectrom-
eter) was supplied by ThermoOrion. A PerkinElmer FTIR spectrometer
was used for collecting IR data. A UV exposure unit (lꢀ350 nm, model
no. 555-279) was purchased from RS Components. A frequency-doubled
Nd:YAG laser (E=350 mJ, l=532 nm, Brilliant B, Quantel) was used in
hologram construction. Holographic measurements were performed by
using an Avantes AVS-MC2000-2 reflectance spectrometer with AvaSoft
5 processing software (Knight Scientific).
À
lower pKa values or with intramolecular B X bonds that
confer a tetrahedral conformation at the boron centre
through the neighbouring effect of an ortho group. The
latter approach is more appealing, because it is believed
À
that if a strong intramolecular B X bond can be generated,
then such boronic acids would bind with glucose even under
mildly acidic conditions. Although a number of boronic
Synthesis of 2-acrylamidophenylboronic acid (2-APB): In an ice–water
bath, a well-stirred solution of 2-aminophenylboronic acid·HCl (0.75 g,
4.4 mmol) in aqueous sodium hydroxide (5n, 5 mL) was treated with
drop-wise addition of acryloyl chloride (0.5 mL) over a period of about
10 min. After 30 min of stirring on ice, the reaction mixture was allowed
to reach room temperature and was stirred for a further 2 h. The mixture
was adjusted to pH 8 by using dilute HCl (0.1m). The resulting beige-col-
oured precipitate was filtered off and washed with water and acetone. A
fine, white powder was obtained in a yield of ꢀ50% after drying.
1H NMR (400 MHz, [D6]DMSO, 258C, TMS): d=12.08 (s, 1H; NH), 7.63
À
acids with B N bonds have been reported and there are
concerns over the strength of the B N bond,
[1a,2b,6–7]
À
very
few of them have been further employed in the develop-
ment of glucose sensors.[8] Acknowledging that an intramo-
[9]
À
lecular B O bond is much stronger, it is believed that the
B O bond should maintain the boron centre in the favoura-
ble tetrahedral form across a wide range of pH, although
little of the literature has discussed pH stability of the tetra-
hedral species and its binding affinity to glucose.
À
À
À
À
(m, 1H; Ph H), 7.54 (m, 1H; Ph H), 7.28 (m, 1H; Ph H), 7.13 (m, 1H;
À
Ph H), 6.33 (dd, J1 =16.8, J2 =1 Hz, 1H; =CH), 6.23 (quad, J1 =J2 =
10 Hz, 1H; =CH), 5.88 ppm (dd, J1 =10, J2 =1 Hz, 1H; =CH); 13C NMR
(100.6 MHz, [D6]DMSO, 258C, TMS): d=163.2, 139.5, 133.7, 129.7,
128.7, 125.5, 117.1 ppm; 11B NMR (128 MHz, [D6]DMSO, 258C, boric
acid): À5.63 ppm; MS (ESI): m/z (%): 191 (100) [M+].
In this paper, two boronates, 2-APB and 3-APB, have
been designed, synthesised and characterised, and their
boron geometries and binding affinity towards glucose and
lactate have been investigated by using 1H, 11B and
13C NMR spectrometry, and IR and mass spectrometry. The
pH titration results show that the tetrahedral forms are the
dominant species present in 2-APB, whereas the trigonal
and tetrahedral forms coexist in 3-APB, with the trigonal
form being the dominant species at neutral pH values. A
holographic sensor based on an acrylamide hydrogel, con-
taining 2-APB as the glucose-responsive functionality, re-
sponded to the addition of glucose at physiological pH
values. The observed contraction of the 2-APB-based holo-
gram in the presence of glucose suggests that glucose cross-
links proximal phenylboronate moieties due to the favoura-
ble tetrahedral form being the predominant species in the
matrix, in contrast to the observed swelling of the 3-APB-
based hologram in the presence of glucose. Furthermore,
the 2-APB-based hologram displayed significantly less inter-
Synthesis of 3-acrylamidophenylboronic acid (3-APB): 3-APB was syn-
thesised according to a method previously reported.[4]
Preparation of copolymer films and hologram construction: The 3-APB-
based copolymer films were made according to a method previously de-
scribed.[4]
Preparation of 2-APB-based copolymer film: The monomer 2-APB
(20 mol%) mixed with glucose (5.0 equiv) was dissolved in DMSO con-
taining the photoinitiator 2,2-dimethoxy-2-phenylacetophenone (2%
w/v). The solid-to-solvent ratio of monomer/DMSO was kept constant
throughout at 0.452:1 (w/v). A droplet (100 mL) of monomer solution was
pipetted onto the polyester surface of an aluminised polyester sheet sit-
ting on a glass plate. A glass microscope slide, pretreated with 3-(trime-
thoxysilyl)propyl methacrylate, was then gently lowered, silane-treated-
side down, onto the monomer mixture. Films were polymerised by means
of a UV-initiated free-radical reaction at room temperature for 60 min.
Polymerised films were peeled off the polyester sheet and then exhaus-
tively washed with water. Consequently, the binding complex was con-
verted to the free 2-APB, thereby releasing the favourable tetrahedral
species present in the copolymer matrix. The process is termed “imprint-
ing photopolymerisation”.
8492
www.chemeurj.org
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8491 – 8497

Holographic Sensors
FULL PAPER
The 11B NMR chemical shift is sensitive to the type of
substituents attached to, as well as to the amount of nega-
tive charge residing on the boron centre. Within a series of
compounds bearing a similar substitution pattern at the
boron atom, the 11B chemical shift values may be used to
deduce their hybridisation states, and thus result in structur-
al identification. Structure determination using 11B NMR
spectrometry has been applied to many different types of
boron-containing compounds.[12] Hence, in order to gauge
Hologram construction: The holograms were constructed according to a
method previously described.[4]
11B NMR spectroscopic pH titration: A boronate (2-APB or 3-APB;
ꢀ3 mg) was dissolved in D2O (0.5 mL), and the pH was adjusted to a de-
sired value by using DCl or NaOD (ꢀ0.1m) and measured by using a mi-
croelectrode pH probe. 11B chemical shifts of samples with different pH
values at room temperature were collected and compared with boric acid
as a standard reference.
Monitoring holographic responses: The holographic sensors were interro-
gated by using a procedure described in previous work.[4]
Holographic pH titration: Holographic pH titration was performed by
equilibrating the 2-APB- or 3-APB-based hologram samples at each pH
value and then recording the diffracted peak wavelength. For each pH
value, 1 mL of the appropriate buffer was added and removed from the
cuvette three times to ensure the buffering capacity of the hologram was
cancelled out. Thereafter, the buffer (1 mL) was left in the cuvette and
the hologram was allowed to equilibrate. The wavelength was recorded
before the buffer was removed and the sample was adjusted to the next
pH value. The following pH buffer systems were used: phosphate (pH 2),
formate (pH 3), acetate (pH 4 and 5), 2-morpholinoethanesulfonic acid
(MES; pH 6 and 7), tris(hydroxymethyl)methylamine (Tris; pH 7.4 and
8), N,N-bis(2-hydroxyethyl)glycine (Bicine; pH 9), 2-(N-cyclohexylami-
no)ethanesulfonic acid (CHES; pH 10) and piperidine (pH 11 and 12);
all of the buffers prepared had a constant concentration (10 mm) and a
fixed ion strength (154 mm).
À
the strength of the B O bond of the intermediate, the
boron geometry of 2-APB relative to 3-APB was explored
by using 11B NMR spectroscopic pH titration (Figure 1). The
Results and Discussion
Equilibria of 2-APB and 3-APB in solution: Boronates 2-
APB and 3-APB were synthesised from aminophenylboron-
ic acids with good yields, then fully characterised by using
multinuclear NMR spectrometry, IR spectrometry and high-
resolution mass spectrometry (HRMS).[4,10] Our previous
work demonstrated that a 3-APB-based hologram responds
to both glucose and lactate; thus, utilising this sensor to
monitor glucose levels in biofluids is problematic, because it
is likely to suffer interference from lactate, which is a fre-
quent metabolic byproduct. Therefore, the boronate 2-APB
was designed to improve the utility of the 3-APB-based
hologram. By moving the acrylamide group from the meta
to the ortho position, the neighbouring group participation
Figure 1. 11B NMR spectroscopic pH titration curves of 3-APB ( ) and 2-
~
APB ( ) in D2O (2–5 mgmLÀ1) at 238C. DCl (0.1m) and NaOD (0.1m)
*
were used for the titration.
titration curve of 3-APB shows a pH-dependent interconver-
sion between the trigonal and tetrahedral forms. Below
pH 6, the major species of 3-APB is the trigonal form (d=
11.0 ppm); above pH 10, the tetrahedral form is dominant
(d=À15.3 ppm); between pH 6 and 10, a broad peak is ob-
served, which shifts from d=11.0 to À15.5 ppm and suggests
the presence of a mixture of the two species due to a fast ex-
change on the 11B NMR timescale. A pKa value of 8.87 for
3-APB was determined from the curve. It was no surprise
that 2-APB behaved differently, with two species present at
d=À13.50 and À15.56 ppm, over the whole pH range. The
latter form exists at pH>11.0 and corresponds to species C,
which possesses the normal tetrahedral structure (Scheme1)
that has been observed in 3-APB. An interesting fact is that
over the wide pH range between 2.0 and 10.0, only one
11B NMR peak at d=À13.50 ppm was observed, which is
most likely due to a contribution from the boron di-hetero-
cyclic tetrahedral zwitterionic adducts A and B. It is be-
lieved that the acrylic amido group is active enough to add
readily and reversibly to the Lewis acid site at the boron
centre in a process that generates zwitterionic adduct A.
Similar zwitterionic boronic adducts have been reported
previously.[9d] With increasing pH, the deprotonation of spe-
cies A on N4 occurs, to generate species B, although
11B NMR analysis is unable to differentiate them because
À
of the amido group through an intramolecular B O-coordi-
nated interaction occurs, resulting in the tetrahedral form of
2-APB being the dominant species. A number of boronic
compounds containing ortho substituents, most of which
À
contain B N interactions, have been reported, and research
into their binding affinity to analytes has been mainly based
on a change in fluorescence or colour,[11] rather than on the
geometry at the boron centre. The latter is a crucial factor
that influences both their affinity for and selectivity towards
glucose, because the boron geometry is closely related to
the electron density on the boron atom as well as to that on
the substituents attached to the boron centre. Consideration
of such effects helps to explain why a variety of structures
exist in a particular boronic acid, and hence their variable
affinity towards different analytes. Therefore, investigating
the structures of 2-APB in solution should enable a greater
understanding of the behaviour of the 2-APB-based holo-
graphic sensors.
Chem. Eur. J. 2006, 12, 8491 – 8497
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8493

C. R. Lowe et al.
the tetrahedral boron geometry
of 2-APB remains even in the
solid state.[10]
Binding affinity of 3-APB and
2-APB in solution: The binding
of 3-APB with glucose at pH>
9
was clearly confirmed by
Scheme 1. Equilibrium forms of 2-APB in D2O.
using 11B NMR spectrometry
because peaks of the free 3-
their boron geometries are the same and the NH peak is ex-
changed by D2O. However, pH titration of the 2-APB-based
hologram provides significant evidence to strongly support
the equilibrium between species A and B, as presented in
Scheme 1. On increasing the pH to above 10, species B is
converted quantitatively into species C, because a strong nu-
cleophilic reagent (HOÀ) is able to open the boron di-het-
APB and the complex were easily recognised at around d=
À13.5 and À10 ppm, respectively. However, the binding of
3-APB with lactate occurs in acidic and neutral media, for
example, its complex was observed at d=À11.8 ppm at
pH 6.5. It is believed that the five-membered-ring ester com-
plexes are formed in both additions; the significant response
to both glucose and lactate has been previously observed in
the 3-APB-based holographic sensor.[4]
À
erocycle by breaking the intramolecular B O bond.
It is known that the zwitterionic adducts are the preferred
form in polar and proton-donor solvents.[13] Interestingly, the
11B NMR spectrum of 2-APB in [D6]DMSO exhibits a
broad peak at d=À5.30 ppm, corresponding to a quasitetra-
hedral form instead of the zwitterionic tetrahedral species A
or B in water (Scheme 1). In the quasitetrahedral form, the
boron centre is not fully negatively charged like species A
Binding of 2-APB with glucose and lactate was carried
out in [D6]DMSO owing to the poor solubility of 2-APB in
1
aqueous solutions except at high pH (>10). The H NMR
spectrum shows significant changes in the chemical shifts of
2-APB observed after addition of glucose; for example, two
NH peaks appear in the mixture instead of one NH peak in
free 2-APB.[10] Furthermore, a fragment with m/z=334.6298
was observed in the mass spectrum of the mixture, corre-
sponding to the complex. These observations led us to the
conclusion that 2-APB binds with glucose in [D6]DMSO.
However, no change was observed in the 11B NMR spec-
trum, suggesting that the complex containing the five-mem-
bered ring with glucose remains in the same quasitetrahe-
dral form as free 2-APB (Figure 2).
À
or B, because a weakly coordinated intramolecular B O in-
teraction occurs between the amide carbonyl and the boron.
Furthermore, by increasing the temperature, a second spe-
cies at d=9.30 ppm appears, suggesting that the trigonal
form is generated due to a cleavage of the weak intramolec-
À
À
ular B O bond (Figure 2). The presence of such a B O
Titration of 2-APB with sodium lactate shifts the 11B
signal upfield from d=À5.6 (broad) to À10.8 ppm
(sharp),[10] suggesting that the boron configuration is con-
verted from the quasitetrahedral to the tetrahedral form. To
consider the structural features of lactate, the effect of the
carboxylic anion on the conversion was first investigated by
titration with sodium acetate. The 11B NMR spectrum shows
that the broad peak at d=À5.6 ppm displays no significant
shifts after adding sodium acetate to free 2-APB, suggesting
that the carboxylic anion does not affect the boron stereo-
À
centre because the intramolecular B O interaction in the
quasitetrahedral form blocks
the anion from attacking the
boron stereocentre. On the
other hand, it was found that in
methanol, 2-APB exists in a
zwitterionic tetrahedral form
Figure 2. 11B NMR spectrum of 2-APB in [D6]DMSO at 808C.
bond in the quasitetrahedral form was also confirmed in the
1
À
H NMR spectrum, in which the N H resonance was ob-
(d=À11.85 ppm;
Figure 3);
such an effect of methanol on
served at d=12.2 ppm at room temperature. The NH signal
is shifted upfield slightly and becomes broader with increas-
ing temperature, suggesting that two species are present at
À
similar boronic acids with B O
Figure 3. Zwitterionic tetrahe-
dral structure of 2-APB in
[D4]methanol (R=CH3) and
bonds has been reported previ-
ously.[14] Hence, the conversion
of the boron stereocentre is due
to ester formation with the a-
À
higher temperature. Similar N H peaks have been reported
previously.[14] FTIR data revealed that there are very strong,
relative to 3-APB, intermolecular hydrogen bonds among
the OH groups present in the solid state of 2-APB and that
2-APB
with
lactate
in
[D6]DMSO
(R=
hydroxyl group of lactate, CH3CH(OH)COONa).
8494
www.chemeurj.org
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8491 – 8497

Holographic Sensors
FULL PAPER
which plays a similar role to methanol (Figure 3), rather
than due to the formation of a five-membered-ring lactate
complex as observed with 3-APB. Investigations in solution
suggest that boronic acid 2-APB is able to possess variable
structures under different conditions (pH, solvent or temper-
ature) because of the neighbouring effect of the substituents
on the boron stereocentre, and can consequently generate a
variety of affinities. In terms of the development of glucose-
selective sensors, the tetrahedral forms may be the preferred
species because they are stable and insensitive to pH; fur-
there is a relatively static response to pH, because species B
is the dominant form present in the hydrogel matrix. On
continuously decreasing the pH from 5 to 3, a significant
contraction (lꢀ50 nm) was observed, due to the fact that
species B with a negative charge on B1 is able to be proto-
nated as the pH value falls and is converted to the zwitter-
ionic species A, which possesses a negative charge on B1
and a positive charge on N4. The generation of species A
neutralises the charge within the matrix, thus a contraction
is observed. The holographic pH titration curve proves that
the 2-APB-based hologram is relatively independent of the
pH in the physiological pH range, and its reactivity in hy-
drogels is comparable to that in solution, both of which are
important features for further application of this sensor.
À
thermore, the coordinated B O interaction is believed to be
strong enough to eliminate lactate interference.
Holographic pH titration of 2-APB and 3-APB: With a
thorough understanding of the behaviour of 2-APB in solu-
tion established, an acrylamide-hydrogel-based hologram
containing 2-APB as the responsive functionality was fabri-
cated by means of photopolymerisation. The pH titration
curves of the 2-APB and 3-APB holograms were recorded
in order to evaluate the effect of pH on the holographic re-
sponse (Figure 4). As the pH was increased, the 3-APB-
Response of 2-APB-based holograms: The response of the
2-APB-based holograms to glucose and lactate was also in-
vestigated using the methods reported previously.[4] The 2-
APB-based holograms showed a significant contraction in
response to glucose at physiological pH (7.1), whereas no
significant response to lactate was observed (Figure 5). Fur-
Figure 5. Response of a 2-APB hologram (20 mol%) in phosphate buffer
*
solution (pH 7.1) to lactate ( ), glucose (”) and glucose mixed with lac-
~
tate (10 mm) ( ).
*
Figure 4. pH titration curves of 3-APB-based (12 mol%; ) and 2-APB-
*
based holograms (20 mol%; ).
thermore, no significant change was observed in the overall
response to an addition of glucose in the presence of lactate
(10 mm) compared to the response in the absence of lactate.
These results suggest that the interference from lactate is in-
consequential for the 2-APB-based hologram in aqueous
media. This may result from the predominance of the tetra-
hedral forms of 2-APB within the hydrogel matrix, because
the tetrahedral forms favour yielding a cyclic ester with glu-
cose via cis-diols, rather than complexing with lactate. Com-
pared with a similar situation in 3-APB-based holograms,[4]
it is apparent that employing the new ligand 2-APB signifi-
cantly diminishes lactate interference. It is also interesting
that the two holographic sensors behave differently towards
glucose. The blueshift observed in the 2-APB-based holo-
gram indicates that the hydrogel matrix contracts upon the
addition of glucose, which is due to the formation of a 2:1
based hologram was found to swell in the pH range from
ꢀ7.5 to 12, resulting in a diffraction-wavelength shift of l
ꢀ160 nm. The swelling can be explained by the generation
of negative charges within the matrix due to the conversion
from the trigonal to the tetrahedral form of 3-APB, in
agreement with the pH titration results in solution
(Figure 1). In the 2-APB-based hologram, however, substan-
tial matrix swelling with a wavelength shift of lꢀ330 nm
was observed in the pH range from 8 to 11, which is be-
lieved to result from the ring opening in a conversion from
species B to C, rather than a charge effect occurring in the
3-APB-based
hologram.
This
ring-opening
effect
(Scheme 1), verified in solution, plays the opposite role to
the cross-linking effect in polymer matrices; hence, it is rea-
sonable to generate such a large shift. Between pH 6 and 7.5
Chem. Eur. J. 2006, 12, 8491 – 8497
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8495

C. R. Lowe et al.
complex between 2-APB and glucose generating a cross-
linking effect. A similar cross-linking effect through two
binding sites on glucose with boronic acids has been pro-
posed previously.[15] In contrast, hydrogel swelling is ob-
served in the 3-APB-based hologram on addition of glucose.
It is known that in the equilibrium of 3-APB there is a coex-
istence of trigonal and tetrahedral forms, which favour bind-
ing with lactate and glucose, respectively. In the physiologi-
cal pH region, the trigonal species is dominant, and the 2:1
complex is less likely to occur due to the only small percent-
age of tetrahedral species present. As a result, the cyclic 1:1
complexes with glucose or lactate are predominantly gener-
ated in 3-APB-based holograms. Consequently, the response
of 3-APB-based holograms to both glucose and lactate was
swelling, because a charge effect in the hydrogels plays a
key role through the generation of the negatively charged
1:1 complexes.
Figure 7. Glucose responses of a 3-APB-based hologram (12 mol%) at
various physiological pH values:
pH 6.5, y=0.87x (R2 =0.9886);
pH 7.4, y=3.40x (R2 =0.9805);
pH 5.8, y=0.44x (R2 =0.969);
pH 7.0, y=2.30x (R2 =0.9823);
pH 7.8, y=2.01x (R2 =0.9964).
Additionally, the binding affinity to glucose in the 2-APB-
based holograms is relatively independent of the buffer pH
in the physiological region pH 5.8–7.8, which is shown by
the fact that the gradients of the calibration curves are
almost identical (Figure 6).[10] This relatively static response
Conclusion
In conclusion, the acrylamide-hydrogel-based holographic
sensor containing 2-APB as the glucose-binding functionali-
ty displays a significant response to glucose at physiological
pH values with only minor interference from lactate. Fur-
thermore, the binding affinity of 2-APB in the hydrogels to
glucose is pH independent. The current results suggest that
the boronate ligand 2-APB shows great potential for the de-
velopment of glucose-selective holographic sensors for use
in physiological media.
Acknowledgement
The authors would like to thank DTI, BBSRC, GSK and the ORS/Gates
Scholarship Scheme for their financial support, and Dr. A. Hussain and
Mr. J. Blyth for useful discussions.
Figure 6. Glucose responses of a 2-APB-based hologram (20 mol%) at
various physiological pH values:
pH 5.8, y=À2.08x (R2 =0.9867);
pH 6.5, y=À2.21x (R2 =0.9991);
pH 7.0, y=À2.39x (R2 =
[1]a) T. James, S. Shinkai, Top. Curr. Chem. 2001, 218, 159–200; b) W.
Wang, X. Gao, B. Wang, Curr. Org. Chem. 2002, 6, 1285–1387; c) S.
Striegler, Curr. Org. Chem. 2003, 7, 81–102.
0.9929);
pH 7.4, y=À2.27x (R2 =0.9968);
pH 7.8, y=À2.24x
(R2 =0.9973).
[2]a) A. M. Pia¸tek, Y. J. Bomble, S. L. Wiskur, E. V. Anslyn, J. Am.
Chem. Soc. 2004, 126, 6072–6077; b) J. Zhao, M. G. Davidson, M. F.
Mahon, G. Kociok-Kçhn, T. D. James, J. Am. Chem. Soc. 2004, 126,
16179–16186.
[3]S. A. Asher, V. L. Alexeev, A. V. Goponenko, A. C. Sharma, I. K.
Lednev, C. S. Wilcox, D. N. Finegold, J. Am. Chem. Soc. 2003, 125,
3322–3329.
[4]M. C. Lee, S. Kabilan, A. Hussain, X. Yang, J. Blyth, C. R. Lowe,
Anal. Chem. 2004, 76(19), 5748–5755.
[5]R. Pizer, P. Ricatto, J. Inorg. Chem. 1994, 33(55), 2402–2406.
[6]L. Zhu, Z. Zhong, E. V. Anslyn, J. Am. Chem. Soc. 2005, 127, 4260–
4269.
to pH across the physiological pH in the holographic pH ti-
tration curve suggests that species B is present in the matrix.
In contrast, the response of the 3-APB-based holograms to
addition of glucose varies significantly across the same pH
range (Figure 7). The holographic pH titration curves indi-
cate that the equilibrium of 3-APB shifts dramatically over
this pH range; the percentage of the tetrahedral form in-
creases and hence the binding to glucose changes according-
ly.
[7]a) T. D. James, K. R. A. S. Sandannayake, S. Shinkai, Nature 1995,
374, 345–347; b) J. C. Norrild, I. Sotofte, J. Chem. Soc. Perkin Trans.
2 2002, 303–311; c) G. Wulff, M. Lauer, H. Bohnke, Angew. Chem.
1984, 96, 714–-716; Angew. Chem. Int. Ed. Engl. 1984, 23, 741–742.
8496
www.chemeurj.org
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8491 – 8497

Holographic Sensors
FULL PAPER
[8]a) M. Mikami, S. Shinkai, J. Chem. Soc. Chem. Commun. 1995, 153;
b) C. Geddes, patent, WO 2005/000109.
Lett. 2002, 4, 4249–4251; c) G. Springsteen, B. Wang, Tetrahedron
2002, 58, 5291–5300; d) C. J. Ward, P. Patel, T. D. James, J. Chem.
Soc. Perkin Trans. 1 2002, 462–470.
[9]a) M. Yamashita, Y. Yamamoto, K. Akiba, D. Hashizume, F. Iwasa-
ki, N. Takagi, S. Nagase, J. Am. Chem. Soc. 2005, 127, 4354–4371;
b) M. P. Hughes, B. D. Smith, J. Org. Chem. 1997, 62, 4492–4499;
c) J.-C. Zhuo, A. H. Soloway, J. C. Beeson, W. Ji, B. A. Barnum, F.-
G. Ong, W. Tjarks, G. T. Jordan, IV, J. Liu, S. G. Shore, J. Org.
Chem. 1999, 64, 9566–9574; d) M. P. Groziak, A. D. Ganguly, P. D.
Robinson, J. Am. Chem. Soc. 1994, 116, 7597–7605; e) P. D. Robin-
son, M. P. Groziak, L. Yi, Acta Crystallogr. Sect. C 1996, 52, 2826–
2830.
[12]a) D. Y. Lee, J. C. Martin, J. Am. Chem. Soc. 1984, 106, 5745–5746;
b) R. D. Pizer, Polyhedron 1996, 15, 3411; c) K. Ishihara, Y. Mouri,
S. Funahashi, M. Tanaka, Inorg. Chem. 1991, 30, 2356–2360.
[13]A. J. Hirby, I. V. Komarov, V. A. Bilenko, J. E. Davies, J. M.
Rawson, Chem. Commun. 2002, 2106–2107.
[14]M. P. Hughes, B. D. Smith, J. Org. Chem. 1997, 62, 4492–4499.
[15]V. L. Alexeev, S. Das, D. N. Finegold, S. A. Asher, Clinical Chemis-
try 2004, 50, 2353–2360.
[10]See the Supporting Information.
[11]a) R. Badugu, J. R. Lakowicz, C. D. Geddes, Bioorg. Med. Chem.
2005, 13, 113; b) S. Arimori, M. L. Bell, C. S. Oh, T. D. James, Org.
Received: March 30, 2006
Published online: August 14, 2006
Chem. Eur. J. 2006, 12, 8491 – 8497
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8497