Allosteric indicator displacement enzyme assay for a cyanogenic glycoside

Article abstract of DOI:10.1002/chem.201302801

Indicator displacement assays (IDAs) represent an elegant approach in supramolecular analytical chemistry. Herein, we report a chemical biosensor for the selective detection of the cyanogenic glycoside amygdalin in aqueous solution. The hybrid sensor consists of the enzyme β-glucosidase and a boronic acid appended viologen together with a fluorescent reporter dye. β-Glucosidase degrades the cyanogenic glycoside amygdalin into hydrogen cyanide, glucose, and benzaldehyde. Only the released cyanide binds at the allosteric site of the receptor (boronic acid) thereby inducing changes in the affinity of a formerly bound fluorescent indicator dye at the other side of the receptor. Thus, the sensing probe performs as allosteric indicator displacement assay (AIDA) for cyanide in water. Interference studies with inorganic anions and glucose revealed that cyanide is solely responsible for the change in the fluorescent signal. DFT calculations on a model compound revealed a 1:1 binding ratio of the boronic acid and cyanide ion. The fluorescent enzyme assay for β-glucosidase uses amygdalin as natural substrate and allows measuring Michaelis-Menten kinetics in microtiter plates. The allosteric indicator displacement assay (AIDA) probe can also be used to detect cyanide traces in commercial amygdalin samples.

Full text of DOI:10.1002/chem.201302801

FULL PAPER
DOI: 10.1002/chem.201302801
Allosteric Indicator Displacement Enzyme Assay for a Cyanogenic Glycoside
D. Amilan Jose,[a, b] Martin Elstner,[a] and Alexander Schiller*[a]
Abstract:
Indicator
displacement
at the allosteric site of the receptor
(boronic acid) thereby inducing
changes in the affinity of a formerly
bound fluorescent indicator dye at the
other side of the receptor. Thus, the
sensing probe performs as allosteric in-
dicator displacement assay (AIDA) for
cyanide in water. Interference studies
with inorganic anions and glucose re-
vealed that cyanide is solely responsi-
ble for the change in the fluorescent
signal. DFT calculations on a model
compound revealed a 1:1 binding ratio
of the boronic acid and cyanide ion.
The fluorescent enzyme assay for b-
glucosidase uses amygdalin as natural
substrate and allows measuring Mi-
chaelis–Menten kinetics in microtiter
plates. The allosteric indicator displace-
ment assay (AIDA) probe can also be
used to detect cyanide traces in com-
mercial amygdalin samples.
assays (IDAs) represent an elegant ap-
proach in supramolecular analytical
chemistry. Herein, we report a chemical
biosensor for the selective detection of
the cyanogenic glycoside amygdalin in
aqueous solution. The hybrid sensor
consists of the enzyme b-glucosidase
and a boronic acid appended viologen
together with a fluorescent reporter
dye. b-Glucosidase degrades the cyano-
genic glycoside amygdalin into hydro-
gen cyanide, glucose, and benzalde-
hyde. Only the released cyanide binds
Keywords: amygdalin
·
boronic
acid · chemical biosensors · density
functional calculations
cence
· fluores-
Introduction
IDA) has been presented by the groups of Anslyn and
Ariga.[7] All mentioned types of displacement assays are
based on the competition between an analyte and an indica-
tor for binding to the receptor (host) at the same binding
site. In contrast, a different system was reported by Singar-
am and co-workers,[8] in which the indicator is displaced by
means of an allosteric interaction of an analyte with a recep-
tor. Herein, the analyte binds at another site (allosteric site)
of the host thereby inducing changes in the affinity of the in-
dicator to the receptor. This can be called an allosteric indi-
cator displacement assay (AIDA, Scheme 1).[9]
Indicator displacement assays have found prominent ap-
plications in supramolecular tandem enzyme assays in the
group of Nau.[10] These assays rely on different binding affin-
ities of indicator dye, substrate, and the corresponding prod-
uct with the supramolecular receptor. They have been clev-
erly applied for monitoring enzymatic transformations in-
volving amino acids, biogenic amines, amino aldehydes, and
nucleotides.[11] Recently, supramolecular real-time fluores-
cent assays for the monitoring of the activity of carbohy-
drate active enzymes have been reported by Singaram
group and our group. These were the first AIDA enzyme
assays.[12] As a continuation of our ongoing endeavor to ex-
plore the AIDA system in real-time fluorescent enzyme
assays, we identified cyanogenic glycosides as challenging
substrates.[13]
Selective and sensitive detection of biological and environ-
mentally important agents in water is of significant interest
in molecular biology, environmental monitoring, and food
safety.[1] Hence, designing new sensors is one of the most im-
portant topics especially in analytical chemistry.[2] The recep-
tor/spacer/reporter paradigm requires a receptor, which is
covalently tethered to a reporter, albeit a chromophore or
fluorophore. Often, analyte binding alters optical properties
by a photoinduced electron transfer (PET) mechanism.[1d,3]
However, supramolecular analytical chemists were fascinat-
ed of Anslynꢀs revitalized indicator displacement assays
(IDAs).[4] They have become popular as supramolecular sen-
sors because of their advantages over traditional receptor/
spacer/reporter systems.[5] To date, a variety of IDAs have
been introduced, such as colorimetric IDAs (C-IDA), fluo-
rometric IDAs (F-IDA), metal complex IDAs (M-IDA), and
enantiomeric IDAs (E-IDA).[6] Very recently, even a me-
chanically controlled indicator displacement assay (MC-
[a] Dr. D. A. Jose, Dipl. Chem. M. Elstner, Prof. Dr. A. Schiller
Institute for Inorganic and Analytical Chemistry
Abbe Center of Photonics, Friedrich Schiller University Jena
Humboldtstrasse 8, 07743 Jena (Germany)
E-mail: alexander.schiller@uni-jena.de
Cyanogenic glycosides are of potential danger for mam-
mals, because hydrogen cyanide (HCN) is produced by hy-
drolysis (spontaneous or enzymatically regulated reac-
tions).[14] The amount of cyanogenic glycosides in plants is
usually reported as the level of releasable HCN. The most
familiar cyanogenic glycoside is amygdalin ([O-b-d-gluco-
[b] Dr. D. A. Jose
Current address: Department of Chemistry
National Institute of Technology
Kurukshetra, Haryana 136119 (India)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201302801.
Chem. Eur. J. 2013, 19, 14451 – 14457
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
14451

and cyanide in water by using fluorescence spectroscopy.
The sensing system works with the AIDA approach; cyanide
ions, produced during an enzymatic reaction of amygdalin,
can be detected in situ.
Results and Discussion
Selection of the receptor and experimental design: The Sin-
garam group has generated a two-component sensing system
for continuous monitoring of the blood-glucose level.[8b,22]
The system is originally composed of 8-hydroxypyrene-1,3,6-
trisulfonic acid trisodium salt (HPTS) as the reporter unit
and derivatives of bis-boronic acid-appended mono violo-
gens (4,4’-o-BBV and 3,3’-o-BBV) as receptors for saccha-
rides.[23] In collaboration with the Singaram group, we have
implemented 4,4’-o-BBV/HPTS and 3,3’-o-BBV/HPTS in se-
lective enzymatic assays for sucrose phosphorylase (SPO)
and phosphoglucomutase (PGM).[12] The sensing system is
perfectly suited for the screening of glycosylation reactions
with SPO and sucrose as donor, because fructose is always
released as product.[9c,13]
Bis-viologens are superior quenchers compared to mono-
viologen derivatives (Scheme 2).[24] In general, boronic acid
based receptors have been utilized for the detection of
Scheme 1. Recent developments in the field of indicator displacement
assays (IDA), such as mechanically controlled IDA[7] and allosteric IDA
(AIDA).[9a]
pyranosyl-(1-6)-b-d-glucopyranosyloxy]benzeneacetoni-
trile).[15] Amygdalin can be found in kernels of food plants,
such as apples, almonds, peaches, cherries, and apricots.[16]
Although amygdalin is not toxic itself, HCN, liberated as
a result of hydrolysis by acids or enzymes, causes acute tox-
icity.[17] Isolated amygdalin is widely sold as an antitumor
natural product.[18] However, recent studies showed cyano-
genic compounds to be dangerously toxic, as well as amyg-
dalin itself as clinically ineffective in the treatment of
cancer.[19] Taken orally, amygdalin is potentially lethal, be-
cause internal b-glucosidase releases HCN upon metabo-
lism. b-Glucosidase is a glycoside hydrolase that acts upon
Scheme 2. Chemical structures of isomers of bis-boronic acid receptors
(o-Bis-BBV, m-Bis-BBV and p-Bis-BBV) and indicator dye (HPTS) used
for the two-component sensor system.
À
b-1 4 bonds linking two glucose or glucose-substituted mol-
ecules. It is an exocellulase with specificity for a variety of
b-d-glycoside substrates, for example, cellobiose. Recently,
b-glucosidases have been used for enzymatic glycosylation
of terpenoids.[13,20] Because of the potential health hazard as-
sociated with the ingestion of cyanide through consumption
of cyanogenic plants and food, several analytical methods,
such as colorimetric methods, HPLC, GC-MS, potentiomet-
ric, and electrochemical methods have been developed for
detection of cyanogenic glycosides and their metabolites.
Most of these labor-intensive methods need sophisticated
analytical instruments.[21] Hence, the development of rapid
and robust detection methods is a very important task.
Herein, we report a boronic acid appended bis-viologen-
based hybrid sensor system for the detection of amygdalin
anionic species including fluoride and cyanide ions.[25] Thus,
we used the second-generation bis-viologen isomers for our
study (Scheme 2). All bis-viologens were synthesized by
a previously reported procedure and characterized by stan-
dard analytical methods.[24] Modification in the synthesis
protocol can be found in the Supporting Information. The
quenching efficiency and preliminary binding studies of
monosaccharides with the Bis-BBV isomers has been report-
ed by Singaram and co-workers,[9a,24] but the application of
o-, m-, p-Bis-BBV receptors in enzyme assays with cyano-
genic glycosides and for the detection of cyanide anions at
physiological conditions is not known.
14452
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 14451 – 14457

Enzyme Assay for a Cyanogenic Glycoside
FULL PAPER
Scheme 3. Working principle of cyanide detection from amygdalin by an allosteric indicator displacement enzyme assay.
Enzyme assay with the cyanogenic glycoside amygdalin:
Enzyme assays are necessary to evaluate the performance of
enzyme variants in high-throughput screening.[12b] Almond
b-glucosidase has been used for the enzymatic glycosylation
of terpenoids. However, reaction conditions and enzyme
variants have to be still screened and improved for higher
product yields.[13,20] Thus, real-time enzyme assays for b-glu-
cosidases are of high interest. Natural substrates, such as cy-
anogenic glycosides, liberate cyanide through enzymatic hy-
drolysis by b-glucosidase.[26] Upon reaction of the cyanogen-
ic glycoside amygdalin, the products glucose, benzaldehyde,
and HCN are formed (Scheme 3).
tained by performing the enzyme assay in a classical fluores-
cence spectrometer with a better time resolution. We show
also the change in emission intensity at different enzyme
concentrations (0 to 1.19 UNmLÀ1). A hyperbolic saturation
curve was obtained by plotting the initial velocities Vi
against the concentration of the enzyme (Figure 1, inset).
For the enzyme assay, we used the o-Bis-BBV isomer due to
the very low glucose response compared to the m-and p-Bis-
BBV isomers.[24]
We presume that the only enzymatic product, which binds
to the boronic acid receptor, is cyanide. It displaces the indi-
cator HPTS from the two-component sensing ensemble and
the fluorescence turns on. This allows real-time monitoring
of the b-glucosidase-mediated reaction (Figure 1). In con-
trast, existing enzyme assays for b-glucosidase use labeled
substrates with fluorogenic or chromogenic groups, such as
resorufin b-d-glucopyranoside, fluorescein di-(b-d-glucopyr-
anoside, and p-nitrophenyl-b-d-glucopyranoside.[27] Upon
enzymatic hydrolysis, the substrates split into glucose and
reporter agent. The signaling mechanism in our system in-
volves ground-state complex formation between dye and
a cationic boronic acid receptor that facilitates electron
transfer from the dye to the bipyridinium salt, resulting in
a fluorescence quenching of the dye (Scheme 3). When the
enzymatic product cyanide strongly interacts with the boron-
ic acid moieties of the ground-state complex at physiological
pH, the cationic viologen is partially neutralized. This weak-
ens the complex and its quenching efficiency. As a conse-
quence, an increase in fluorescence intensity is observed.
In principle, the hydrolysis products of amygdalin glucose
and cyanide, or both of them, could be responsible for the
increase of fluorescence intensity. Thus, we have carried out
several experiments to elucidate unambiguously the “work-
ing” analyte. Among receptors for the recognition of fluo-
ride and cyanide ions developed to date, boron compounds
are shown to be one of the most effective molecular plat-
forms due to their high affinity toward nucleophilic
anions.[25,28] Many chemical receptors are available for the
detection of cyanide ions.[1d,29] However, most of them failed
to bind in pure water, and cyanide binding often interferes
with other anionic species. Therefore, the search for a highly
selective cyanide sensor in pure water is still a challenging
task.[25b,28,30]
The reaction of amygdalin with b-glucosidase in the pres-
ence of o-Bis-BBV/HPTS was studied by time-dependent
emission intensity measurements (F/F0; Figure 1). Fluores-
Figure 1. Time-dependent fluorescence measurements of o-Bis-BBV/
HPTS (4ꢂ10À5/4ꢂ10À6 m) in the presence of amygdalin (60.5 mm) with
varying concentrations of b-glucosidase (from down to top 0, 0.048,
0.238, 0.476, 0.714, and 1.19 UNmLÀ1
) in water (100 mm HEPES,
pH 7.20) at 378C. Inset: plot of initial rate Vi versus b-glucosidase con-
centration.
cence experiments were carried out by using a microtiter
plate reader in 384-well plates with 100 mL as a working
volume (time resolution 30 s). The receptor isomer o-Bis-
BBV (4ꢂ10À5 m) containing HPTS (4ꢂ10À6 m) dye was fresh-
ly prepared in 4-(2-hydroxyethyl)-1-piperazineethanesulfon-
ic acid (HEPES) buffer (100 mm, pH 7.20). In the absence
of b-glucosidase, the fluorescence of o-Bis-BBV/HPTS with
amygdalin (60.5 mm) was in a quenched state and virtually
stable. In presence of b-glucosidase, the fluorescence intensi-
ty increased significantly, and the reaction was followed in
regular time intervals (Figure 1). Similar results were ob-
Anion binding in water: We have investigated the binding
properties of the isomers o-, m-, p-Bis-BBV, and HPTS (4ꢂ
Chem. Eur. J. 2013, 19, 14451 – 14457
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14453

A. Schiller et al.
moiety. For computational simplicity, bis-BBV receptors
have been modeled as phenyl boronic acid (pB(OH)2;
Figure 3). Previously, Larkin and co-workers showed the us-
ability of different DFT methods for energy calculations
with boronic acids.[33] Because of the known problems to es-
timate dative bond energies with the B3LYP functional, we
have also used the M06-2X functional.[34] It has been shown
that the M06-2X functional predicts accurately interaction
energies in boron–nitrogen adducts and Raman spectra of
protected boronic acids.[35] The calculated results show that
À
the formation of an anionic pB(OH)3 through the addition
of a hydroxyl ion is an exergonic process with a Gibbs
energy of À187.4 kJmolÀ1. In contrast, this reaction is with
À51.8 kJmolÀ1 less exergonic in the PCM model calculations
(in the case of the M06-2X functional, it is À91.7 kJmolÀ1,
see Table S1 in the Supporting Information).
Addition of a single cyanide ion to pB(OH)2 forms a tetra-
hedral species pB(OH)2CNÀ in an exergonic reaction, with
DG of À48.6 kJmolÀ1. It changes to slightly endergonic
values in aqueous-phase calculations (DG=33.7 kJmolÀ1
(B3LYP) and DG=13.7 kJmolÀ1 (M06-2X)). However, the
substitution of three hydroxide groups by three cyanide ions
is a strong endergonic process in all calculated cases
(Table S1 in the Supporting Information). These theoretical
calculations indicate for the first time that threefold cya-
À
nide-coordinated boron complex pB(CN)3 is not favorable
as shown in earlier reports,[25a,b] and formation of one cya-
nide-coordinated boron complex pB(OH)2CNÀ is the most
favorable.[36]
The generation of an anionic boronate displaces the
HPTS molecule, which is responsible for the fluorescent en-
hancement (Scheme 3). Other anions show very small to
negligible changes in the emission intensity (Figure 2). They
could not form anionic boronates in water. Fluorescence ti-
tration experiments were carried out to determine the sensi-
tivity of the cyanide sensing system o-Bis-BBV/HPTS. From
these experiments, we found that the probe is able to detect
selectively cyanide in the millimolar range with a detection
limit of around 4 mm in water (Figure S4 in the Supporting
Information). In a similar way, fluorescence titrations of m-
and p-Bis-BBV with anionic substrates were conducted (Fig-
ures S5 and S6 in the Supporting Information). Interestingly,
from the three isomers of the boronic acid receptor, o-Bis-
BBV isomer showed the strongest fluorescence intensity
modulation (F/F0). Fitting of fluorescent spectral data with
a nonlinear binding isotherm[8b] gave similar apparent bind-
ing constants of 207Æ20, 135Æ20, and 137Æ20mÀ1 for o-
Bis-BBV, m-Bis-BBV, and p-Bis-BBV isomers, respectively
(Figures S4–S6 in the Supporting Information). We were not
able to calculate the binding affinities with other anions, be-
cause they showed very small changes in fluorescence emis-
sion (Figure 2 and S2 and S3 in the Supporting Informa-
tion).
Figure 2. a) Fluorescence spectra of o-Bis-BBV/HPTS=4ꢂ10À5/4ꢂ10À6
m
,
À
with inorganic anions, such as FÀ, CNÀ, BrÀ, ClÀ, SO42À, H2PO4À, NO3
SCNÀ, and AcOÀ (5ꢂ10À3 m); b) fluorescence emission intensity modula-
tion (F/F0) of o-, m-, and p-Bis-BBV/HPTS with anions. Reaction condi-
tions: water with 100 mm HEPES, pH 7.20 at room temperature.
10À5/4ꢂ10À6 m) with a collection of inorganic anions, such as
À
À
FÀ, CNÀ, BrÀ, ClÀ, SO42À, H2PO4 , NO3 , SCNÀ, and AcOÀ
(5ꢂ10À3 m) in water at pH 7.20 by using fluorescence spec-
troscopy measurements.
Without addition of anions, the fluorescence intensity of
the o-Bis-BBV/HPTS system was quenched (Figure 2). A
strong fluorescence signal modulation (F/F0) was only ob-
served in the presence of the cyanide anion. Inorganic
anions obtain high free energies of hydration DGHydr. There-
fore, anion hosts have to compete more effectively with the
surrounding medium. That makes it especially difficult to
bind fluoride, sulfate, and phosphate anions in aqueous solu-
tion.[31] It is also known from literature that cationic ammo-
nium boranes can be used for cyanide complexation because
of favorable Coulombic receptor-cyanide attractions.[32]
A stepwise OHÀ/CNÀ exchange process of a boronic acid
with the cyanide ion in aqueous solution is possible.[25a]
Thus, we performed DFT energy calculations on the reac-
tion of cyanide and hydroxyl ions with a boronic acid
Interference study: A disadvantage of many optical chemo-
sensors for cyanide is that they are easily disturbed by other
anions. Most of the systems that are based on boron deriva-
14454
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 14451 – 14457

Enzyme Assay for a Cyanogenic Glycoside
FULL PAPER
glycosides in aqueous solution.
The detection limit of amygda-
line is 9 mm after 30 min of
enzyme reaction.
We investigated the Michae-
lis–Menten (MM) behavior of
the b-glucosidase assay in the
presence of o-Bis-BVV/HPTS
by measuring the kinetics with
varying amygdalin concentra-
tions (Figure S8 in the Support-
ing Information). Initial rate
constants
were
calculated
within the first five minutes of
the b-glucosidase assay. The
rate constants were plotted
against the concentration of
amygdalin (Figure S9 in the
Supporting Information). The
plot followed a typical MM sat-
uration curve and gave ap-
Figure 3. Possible OHÀ/CNÀ exchange process in phenyl boronic acid (pB(OH)2) used for the DFT calcula-
tions and comparison of relative Gibbs energy (DG) for different species of phenyl boronic acid obtained
upon OHÀ/CNÀ exchange process, the values were calculated by using B3LYP/aug-cc-pVTZ.
AHCTUNGTRENpGUNN arent Km and Vmax values
of 23.8Æ1 mm and 0.05Æ
tives show high competition between cyanide and fluo-
ride.[32,37] To investigate the anion interference in cyanide
binding, we have measured the fluorescence response of cy-
anide (8.0ꢂ10À3 m) together with common anions, such as
FÀ, BrÀ, ClÀ, SO42À, H2PO4 , NO3 , SCNÀ, and AcOÀ. These
anions were added in excess (4.0ꢂ10À2 m). Very small or
negligible fluorescence changes can be observed in the pres-
ence of all mentioned anions (Figure S7 in the Supporting
Information). These results indicate that the sensing system
is highly selective towards cyanide over other anions in
water. We have also investigated the interference of mono-
saccharides with cyanide. The monosaccharides glucose and
galactose did not show any fluorescence changes at concen-
trations of 4.0ꢂ10À2 m, fructose showed very little interfer-
ence with cyanide (Figure S7 in the Supporting Informa-
tion). These findings further validate that cyanide is solely
responsible for the change in emission intensity of the b-glu-
cosidase assay (Figure 3).
In addition, we performed control experiments with the
substrates cellobiose, lactose, and maltose (Figure S11 in the
Supporting Information). b-Glucosidase has also the ability
to hydrolyze these substrates and produce glucose as one of
the products.[38] As shown in Figure 4, the sensing system o-
Bis-BBV/HPTS did not show any detectable changes in the
fluorescence spectra even when measuring the fluorescence
after 6 h. These results clearly demonstrate that glucose,
produced during hydrolysis of cellobiose, lactose, or maltose,
has no or a very weak effect on the fluorescence signal
(Figure 4). As a control, amygdalin did not show any fluo-
rescent change in absence of b-glucosidase (Figure S10 in
the Supporting Information). These experiments demon-
strate how amygdalin can be selectively determined of other
À
À
Figure 4. Fluorescence spectra of o-Bis-BBV/HPTS (4ꢂ10À5/4ꢂ10À6 m,
blank) with 60.5 mm of substrates amygdalin, maltose, cellobiose, and lac-
tose in water (HEPES, 100 mm, pH 7.20) at 378C in presence of enzyme
b-glucosidase (0.356 UNmLÀ1). The emission spectra were recorded after
6 h.
0.001 mmminÀ1, respectively. These measured kinetic values
for b-glucosidase activity are different from literature values
(Km =1.45 mm at 508C in 50 mm 2-(N-morpholino)ethane-
sulfonic acid (MES) buffer, pH 6.2).[38b,39] Changes in assay
conditions and the use of a different monitoring method
could be the reason.
Cyanide contamination in commercial amygdalin sources:
Amygdalin has been used as an anticancer treatment in
humans worldwide. Although many anecdotal reports are
available, no controlled clinical trials of amygdalin have
ever been conducted.[23,24] However, commercial prepara-
tions of amygdalin are still sold through internet. These
problematic products may contain contaminations of CNÀ
and/or glucosidases. To quantify the amount of CNÀ, fluores-
Chem. Eur. J. 2013, 19, 14451 – 14457
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14455

A. Schiller et al.
cent measurements with potassium cyanide (KCN)-spiked
amygdalin samples were carried out. The fluorescence inten-
sity at l=518 nm increased in a linear fashion with a higher
percentage of KCN in the amygdalin samples (0–40%).
A similar change in emission intensity was also observed
with enzymatic hydrolysis of amygdalin. The content of the
total cyanide content present upon decomposition of amyg-
dalin can be estimated with a calibration curve generated
from the cyanide-spiked amygdalin (Figure 5, inset). The de-
under allosteric indicator displacement assay. The binding
behavior of cyanide with boronic acid receptors was investi-
gated by comparative DFT methods to confirm a single
OHÀ/CNÀ exchange process in a model compound. We
showed that threefold cyanide-coordinated boronate com-
plex is not favorable. The sensing mechanism has the poten-
tial of being used in amygdalin detection and in the monitor-
ing of different levels of cyanide contamination in the com-
mercial amygdalin samples by fluorescent spectroscopy. We
anticipate that allosteric indicator displacement enzyme
assays will be used in future as powerful tools for the detec-
tion of complex bioanalytes and for the evaluation of new
enzyme variants.
Acknowledgements
All authors thank the EC for financial support through the FP7-project
“Novosides” (grant agreement nr. KBBE-4-265854; www.novosides.eu).
A.S. is grateful to the Carl Zeiss Foundation for a Junior Professor fel-
lowship. We also acknowledge the support of the Center of Medical
Optics and Photonics (CeMOP) and the Jena Center of Soft Matter
(JCSM) at the Friedrich Schiller University Jena.
Figure 5. Fluorescence intensity change of o-Bis-BBV/HPTS (4ꢂ10À5/4ꢂ
10À6 m) with amygdalin (75 mm) spiked with KCN in water (100 mm
HEPES, pH 7.20). Inset: plot of spiked percentage of KCN in amygdalin
versus change in emission intensity at l=518 nm.
[1] a) V. M. Mirsky, A. K. Yatsimirsky, Artificial Receptors for Chemical
Sensors, Wiley-VCH, Weinheim, 2011, p.469; b) O. S. Wenger,
Chem. Rev. 2013, 113, 3686–3733; c) J. Rebek, Proc. Natl. Acad. Sci.
USA 2009, 106, 10423–10424; d) R. Martꢃnez-MaÇez, F. Sancenon,
Chem. Rev. 2003, 103, 4419–4476; e) A. W. Czarnik, Nature 1998,
394, 417–418; f) S. W. Thomas, G. D. Joly, T. M. Swager, Chem. Rev.
2007, 107, 1339–1386.
[2] a) A. P. Umali, E. V. Anslyn, Curr. Opin. Chem. Biol. 2010, 14, 685–
692; b) X. Lou, D. Ou, Q. Li, Z. Li, Chem. Commun. 2012, 48,
8462–8477; c) J. Chan, S. C. Dodani, C. J. Chang, Nat. Chem. 2012,
4, 973–984.
[3] C. R. Wade, A. E. J. Broomsgrove, S. Aldridge, F. P. Gabbai, Chem.
Rev. 2010, 110, 3958–3984.
[4] a) S. L. Wiskur, H. Ait-Haddou, J. J. Lavigne, E. V. Anslyn, Acc.
Chem. Res. 2001, 34, 963–972; b) D. Leung, S. O. Kang, E. V.
Anslyn, Chem. Soc. Rev. 2012, 41, 448–479.
[5] a) E. V. Anslyn, J. Org. Chem. 2007, 72, 687–699; b) D.-J. Oh, K.-M.
Kim, K.-H. Ahn, Chem. Asian J. 2011, 6, 2034–2039; c) M. Hu, G.
Feng, Chem. Commun. 2012, 48, 6951–6953; d) S.-K. Sun, K.-X. Tu,
X.-P. Yan, Analyst 2012, 137, 2124–2128; e) X. Zhang, E. V. Anslyn,
X. Qian, Supramol. Chem. 2012, 24, 520–525.
[6] B. T. Nguyen, E. V. Anslyn, Coord. Chem. Rev. 2006, 250, 3118–
3127.
[7] K. Sakakibara, L. A. Joyce, T. Mori, T. Fujisawa, S. H. Shabbir, J. P.
Hill, E. V. Anslyn, K. Ariga, Angew. Chem. Int. Ed. 2012, 51, 9643–
9646.
[8] a) S. Gamsey, N. A. Baxter, Z. Sharrett, D. B. Cordes, M. M. Olm-
stead, R. A. Wessling, B. Singaram, Tetrahedron 2006, 62, 6321–
6331; b) S. Gamsey, A. Miller, M. M. Olmstead, C. M. Beavers, L. C.
Hirayama, S. Pradhan, R. A. Wessling, B. Singaram, J. Am. Chem.
Soc. 2007, 129, 1278–1286; c) A. Schiller, R. A. Wessling, B. Singa-
rum, Angew. Chem. 2007, 119, 6577–6579; Angew. Chem. Int. Ed.
2007, 46, 6457–6459.
[9] a) A. Schiller, B. Vilozny, R. A. Wessling, B. Singaram, Rev. Fluo-
resc. 2011, 8, 155–191; b) M. Elstner, K. Weisshart, K. Muellen, A.
Schiller, J. Am. Chem. Soc. 2012, 134, 8098–8100; c) B. Pignataro,
A. Schiller in Molecules at Work: Going Beyond Glucose Sensing
with Boronic Acid Receptors, Wiley-VCH, Weinheim, 2012, pp. 315–
338.
tection limit of CNÀ in amygdalin is around 2 wt%. In addi-
tion, we were able to discriminate different amygdalin sam-
ples by their purity. We used samples from the following
commercial sources: Aldrich (purity 96%), Alfa–Aesar
(purity 98%), and TCI-Europe (purity 99%). We measured
the fluorescence intensities at different amygdalin concen-
trations (22, 55, 82 mm) in independent triplicates. It is im-
portant to note that we are analyzing traces of CNÀ in these
“pure” commercial samples. The dataset was analyzed by
linear discriminant analysis.[8c,23a] A clear separation of all
amygdalin samples was obtained with a small variance (Fig-
ure S13 in the Supporting Information). This shows that the
current AIDA probe not only works well in a fluorescent b-
glucosidase assay and for the detection of amygdalin among
other oligosaccharides. They also offer a high potential of
being used in the detection of cyanide in amygdalin samples
and other cyanogenic glycosides.
Conclusion
We have described the ability of boronic acid appended bis-
viologen quenchers (o-, m-, p-Bis-BBV) to be used as
probes for monitoring the enzymatic reaction of amygdalin
with b-glucosidase. We confirmed that among different en-
zymatic products, cyanide is exclusively responsible for the
change in emission intensity of the sensing system o-Bis-
BBV/HPTS at physiological conditions. It is the first fluores-
cent displacement assay to detect the cyanide ion released
from amygdalin. The cyanide sensing mechanism works
14456
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 14451 – 14457

Enzyme Assay for a Cyanogenic Glycoside
FULL PAPER
[10] R. N. Dsouza, A. Hennig, W. M. Nau, Chem. Eur. J. 2012, 18, 3444–
3459.
Dyes Pigm. 2005, 64, 49–55; c) J. V. Ros-Lis, R. Martinez-Manez, J.
Soto, Chem. Commun. 2005, 5260–5262; d) Y. Kim, H.-S. Huh,
M. H. Lee, I. L. Lenov, H. Zhao, F. P. Gabbai, Chem. Eur. J. 2011,
17, 2057–2062.
[11] a) G. Ghale, N. Kuhnert, W. M. Nau, Nat. Prod. Commun. 2012, 7,
343–348; b) M. Florea, S. Kudithipudi, A. Rei, M. J. Gonzalez-Al-
varez, A. Jeltsch, W. M. Nau, Chem. Eur. J. 2012, 18, 3521–3528;
c) G. Ghale, V. Ramalingam, A. R. Urbach, W. M. Nau, J. Am.
Chem. Soc. 2011, 133, 7528–7535; d) M. Florea, W. M. Nau, Org.
Biomol. Chem. 2010, 8, 1033–1039; e) W. M. Nau, G. Ghale, A.
Hennig, H. Bakirci, D. M. Bailey, J. Am. Chem. Soc. 2009, 131,
11558–11570.
[12] a) B. Vilozny, A. Schiller, R. A. Wessling, B. Singaram, Anal. Chim.
Acta 2009, 649, 246–251; b) J.-L. Reymond, V. S. Fluxa, N. Maillard,
Chem. Commun. 2009, 34–46.
[13] T. Desmet, W. Soetaert, P. Bojarova, V. Kren, L. Dijkhuizen, V.
Eastwick-Field, A. Schiller, Chem. Eur. J. 2012, 18, 10786–10801.
[14] J. Vetter, Toxicon 2000, 38, 11–36.
[26] T. Yamashita, T. Sano, T. Hashimoto, K. Kanazawa, Int. J. Food Sci.
Technol. 2007, 42, 70–75.
[27] a) G. M. King, Appl. Environ. Microbiol. 1986, 51, 373–380;
b) A. K. Grover, D. D. Macmurchie, R. J. Cushley, Biochim. Bio-
phys. Acta Enzymol. 1977, 482, 98–108.
[28] K. C. Song, K. M. Lee, N. V. Nghia, W. Y. Sung, Y. Do, M. H. Lee,
Organometallics 2013, 32, 817–823.
[29] a) Z. Xu, X. Chen, H. N. Kim, J. Yoon, Chem. Soc. Rev. 2010, 39,
127–137; b) U. Reddy, G. P. Das, S. Saha, M. Baidya, S. K. Ghosh,
A. Das, Chem. Commun. 2013, 49, 255–257; c) S. Saha, A. Ghosh, P.
Mahato, S. Mishra, S. K. Mishra, E. Suresh, S. Das, A. Das, Org.
Lett. 2010, 12, 3406–3409.
[15] A. V. Morant, N. Bjarnholt, M. E. Kragh, C. H. Kjaergaard, K. Joer-
gensen, S. M. Paquette, M. Piotrowski, A. Imberty, C. E. Olsen,
B. L. Moeller, S. Bak, Plant Physiol. 2008, 147, 1072–1091.
[16] D. A. Jones, Phytochemistry 1998, 47, 155–162.
[17] T. Tatsuma, K. Komori, H. H. Yeoh, N. Oyama, Anal. Chim. Acta
2000, 408, 233–240.
[30] a) Y.-K. Yang, J. Tae, Org. Lett. 2006, 8, 5721–5723; b) C. Mꢄnnel-
Croisꢅ, F. Zelder, Inorg. Chem. 2009, 48, 1272–1274; c) C. Mꢄnnel-
Croisꢅ, F. Zelder, ACS Appl. Mater. Interfaces 2012, 4, 725–729.
[31] Z.-h. Lin, S.-j. Ou, C.-y. Duan, B.-g. Zhang, Z.-p. Bai, Chem.
Commun. 2006, 624–626.
[32] T. W. Hudnall, F. P. Gabbaie, J. Am. Chem. Soc. 2007, 129, 11978–
11986.
[18] C. Holden, Science 1976, 193, 982–985.
[19] a) J. D. Khandekar, H. Edelman, JAMA J. Am. Med. Assoc. 1979,
242, 169–171; b) E. S. Schmidt, G. W. Newton, S. M. Sanders, J. P.
Lewis, E. E. Conn, JAMA J. Am. Med. Assoc. 1978, 239, 943–947.
[20] F. Rivas, A. Parra, A. Martinez, A. Garcia-Granados, Phytochem.
Rev. 2013, 1–13.
[21] a) T. Tatsuma, K. Tani, N. Oyama, H.-H. Yeoh, J. Electroanal.
Chem. 1996, 407, 155–159; b) U. Backofen, F. M. Matysik, G.
Werner, Fresenius J. Anal. Chem. 1996, 356, 271–273; c) H. H.
Yeoh, V. D. Truong, Food Chem. 1993, 47, 295–298; d) H. H. Yeoh,
Biotechnol. Tech. 1993, 7, 761–764; e) J. L. Lambert, J. Ramasamy,
J. V. Paukstelis, Anal. Chem. 1975, 47, 916–918.
[33] a) J. D. Larkin, K. L. Bhat, G. D. Markham, B. R. Brooks, H. F.
Schaefer, C. W. Bock, J. Phys. Chem. A 2006, 110, 10633–10642;
b) J. D. Larkin, J. S. Fossey, T. D. James, B. R. Brooks, C. W. Bock, J.
Phys. Chem. A 2010, 114, 12531–12539.
[34] a) H. A. LeTourneau, R. E. Birsch, G. Korbeck, J. L. Radkiewicz-
Poutsma, J. Phys. Chem. A 2005, 109, 12014–12019; b) B. G. Jane-
sko, J. Chem. Theory Comput. 2010, 6, 1825–1833.
[35] D. N. Reinemann, A. M. Wright, J. D. Wolfe, G. S. Tschumper, N. I.
Hammer, J. Phys. Chem. A 2011, 115, 6426–6431.
[36] L. Feng, Y. Wang, F. Liang, W. Liu, X. Wang, H. Diao, Sens. Actua-
tors B 2012, 168, 365–369.
[22] a) D. B. Cordes, B. Singaram, Pure Appl. Chem. 2012, 84, 2183–
2202; b) B. Vilozny, A. Schiller, R. A. Wessling, B. Singaram, J.
Mater. Chem. 2011, 21, 7589–7595.
[23] a) A. Schiller, B. Vilozny, R. A. Wessling, B. Singaram, Anal. Chim.
Acta 2008, 627, 203–211; b) J. N. Camara, J. T. Suri, F. E. Cappuccio,
R. A. Wessling, B. Singaram, Tetrahedron Lett. 2002, 43, 1139–1141.
[24] Z. Sharrett, S. Gamsey, P. Levine, D. Cunningham-Bryant, B. Viloz-
ny, A. Schiller, R. A. Wessling, B. Singaram, Tetrahedron Lett. 2008,
49, 300–304.
[37] Z. Ekmekci, M. D. Yilmaz, E. U. Akkaya, Org. Lett. 2008, 10, 461–
464.
[38] a) Y. B. Tewari, R. N. Goldberg, J. Biol. Chem. 1989, 264, 3966–
3971; b) F. F. Zanoelo, M. d. L. T. d. M. Polizeli, H. F. Terenzi, J. A.
Jorge, FEMS Microbiol. Lett. 2004, 240, 137–143.
[39] S. Yang, Z. Jiang, Q. Yan, H. Zhu, J. Agric. Food Chem. 2008, 56,
602–608.
Received: July 17, 2013
Revised: August 30, 2013
Published online: October 7, 2013
[25] a) R. Badugu, J. R. Lakowicz, C. D. Geddes, J. Am. Chem. Soc.
2005, 127, 3635–3641; b) R. Badugu, J. R. Lakowicz, C. D. Geddes,
Chem. Eur. J. 2013, 19, 14451 – 14457
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14457