Fructose recognition using new “Off-On” fluorescent chemical probes based on boronate-tagged 1,8-naphthalimide

Article abstract of DOI:10.1039/c8nj05092a

It has recently been found that hereditary fructose intolerance (HFI) alongside the blood glucose level can be an important indicator in diabetes or some metabolic disorders. Over time, fluorescent organic molecules have grown in popularity for selective recognitions of a wide range of analytes. This work introduces new synthesized molecules based on boronate-tagged 1,8-naphthalimide with a planar structure and high fusion loops for rapid optical detection of fructose. Boronic acids which form five-membered boronate esters with diols have become increasingly popular in the synthesis of small chemical receptors. Two new water-soluble fluorescent selectophores have been successfully synthesized and their selectivity has been studied in sensing saccharides. The synthesized selectophores indicated selectivity toward fructose compared to other tested saccharides within the range of 0.16 to 50 mM at the physiological pH 7.4. The designed fluorescent probes were used to measure the percentage of fructose in fructose corn syrup. The results revealed that the (2-(((2-((7-oxo-7H-benzo[d,e]benzo[4,5]imidazo[2,1-a]isoquinolin-3-yl)amino)ethyl)amino)methyl)phenyl)boronic acid (NONB) probe was able to determine the concentration of fructose satisfactorily, suggesting that this probe can potentially be applied for quantitative detection of fructose in real samples.

Full text of DOI:10.1039/c8nj05092a

NJC
View Article Online
View Journal | View Issue
PAPER
Fructose recognition using new ‘‘Off–On’’
fluorescent chemical probes based on
boronate-tagged 1,8-naphthalimide
Cite this: New J. Chem., 2018,
42, 19872
ab
c
Sanaz Seraj,a Shohre Rouhani
*
and Farnoush Faridbod
It has recently been found that hereditary fructose intolerance (HFI) alongside the blood glucose level can
be an important indicator in diabetes or some metabolic disorders. Over time, fluorescent organic
molecules have grown in popularity for selective recognitions of a wide range of analytes. This work
introduces new synthesized molecules based on boronate-tagged 1,8-naphthalimide with a planar structure
and high fusion loops for rapid optical detection of fructose. Boronic acids which form five-membered
boronate esters with diols have become increasingly popular in the synthesis of small chemical receptors.
Two new water-soluble fluorescent selectophores have been successfully synthesized and their selectivity
has been studied in sensing saccharides. The synthesized selectophores indicated selectivity toward fructose
compared to other tested saccharides within the range of 0.16 to 50 mM at the physiological pH 7.4.
The designed fluorescent probes were used to measure the percentage of fructose in fructose corn
syrup. The results revealed that the (2-(((2-((7-oxo-7H-benzo[d,e]benzo[4,5]imidazo[2,1-a]isoquinolin-3-
yl)amino)ethyl)amino)methyl)phenyl)boronic acid (NONB) probe was able to determine the concentration
of fructose satisfactorily, suggesting that this probe can potentially be applied for quantitative detection of
fructose in real samples.
Received 7th October 2018,
Accepted 9th November 2018
DOI: 10.1039/c8nj05092a
rsc.li/njc
interact selectively with carbohydrates requires special recogni-
tion moieties capable of strong binding under physiological
Introduction
Carbohydrates are one of the three main categories of bio- conditions. Many molecules and biomolecules have been reported
molecules that are essential in biological systems. Glucose and for saccharide detection including antibodies,5 lectins,6 aptamers
fructose levels play a significant role in some diseases, such based on nucleic acids7 and small molecule lectin mimics,8 and
as hypoglycemia, diabetes mellitus, and hereditary fructose boronic acids and derivatives.9–11 Among the above-mentioned
intolerance (HFI). Therefore, quantitative monitoring of materials, boronic acid compounds have attracted more interest
saccharides such as fructose is of great importance in diagnoses thanks to their strong interactions with sugars. Boronic acids can
and therapies.1,2 HFI is due to a deficiency of fructose-1- bind tightly and reversibly to cis-1,2 or 1,3 diol moieties to form
phosphate aldose activity which can cause accumulation of five- or six-membered rings, respectively.12–14 The fluorescence of
fructose in the liver, small intestine, and kidneys. This disease boronic acid-based compounds changes when bound with
is currently diagnosed with an H2-exhalation test. This test is sugars.15–21 According to these properties, they have been used
dangerous to the patient and causes complications such as the for a wide variety of applications including solution carbohydrate
use of fructose by the patient during the test. On the other hand, sensing,22,23 cell labeling based on cell surface carbohydrate
only limited laboratories have the ability to detect the disease biomarkers,24 and carbohydrate separation.25 The first fluores-
at the blood level by molecular biological diagnostic methods. cence sensor for saccharide detection was designed by Czarnik in
For that reason, monitoring blood fructose levels is of great 1992.26 In this design, the boronic acid group was directly
relevance.3,4 Development of small molecules which are able to attached to the anthracene ring. After that, various fluorescence
probes were made by Shinkai and coworkers.27,28 The designed
probes had high fluorescence emission, which was quenched
in the presence of fructose. After the investigation by Shinkai,
a Department of Organic Colorants, Institute for Color Science and Technology,
Tehran, Iran. E-mail: rouhani@icrc.ac.ir
b Center of excellence for Color Science and Technology (CECST), Institute for Color
different fluorescence probes were developed to detect
saccharides.29,30 One of the major challenges in designing probe
Science and Technology, Tehran, Iran
c Center of Excellence in Electrochemistry, School of Chemistry, College of Science,
studies was the lack of fluorescence emission changes. To
improve this problem, the photoinduced electron transfer (PET)
University of Tehran, Tehran, Iran
19872 | New J. Chem., 2018, 42, 19872--19880 This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online
Paper
NJC
mechanism was introduced by James and coworkers in 1994.31 Equipment
Based on this mechanism, the probe is designed in three parts
All fluorescence spectral measurements were performed using a
Perkin-Elmer LS55 fluorescence spectrophotometer. UV-Vis absorp-
tion of the synthesized dyes in PBS solution was collected on a
including donor, spacer, and acceptor. They used the amine
group as a Lewis base and boronic acids as a Lewis acid for the
first time.32 1,8-Naphthalimide is well known for its typical
photoinduced electron transfer (PET) and intramolecular
charge transfer (ICT), strong absorption and emission in the
visible region and large Stokes shift. Design, synthesis, and
application of some 1,8-naphthalimide derivatives for sensor
applications have been undertaken by our team.33–35 Until now,
various fluorescent probes including a boronic acid group
based on naphthalimide dyes have been designed and devel-
oped to detect monosaccharides.15,17,36–39 In recent years, the
use of different structures to improve the PET mechanism, the
type of interactions between the amine and boron groups
1
CECIL-CE9200 spectrophotometer. H NMR and 13C NMR spectra
were recorded using a Brucker DRX AVANCE NMR spectrophoto-
meter at 500 MHz and 125 MHz in DMSO. FTIR spectra were
recorded on a SPECTRUM ONE spectrometer using KBr pellets.
Fluorescence quantum yield was evaluated based on the absorp-
tion and fluorescence spectra, using fluorescein as a reference
(jst = 0.95).40 The melting points (m.p.) were determined using a
Perkin-Elmer (USA) Pyris 6 differential scanning calorimeter.
Synthesis
and the effect of an amine group on the detection rate of The intermediate compound 1 (as shown in Scheme 1) was
saccharides has become a very attractive field. On the other synthesized from acenaphthene according to the literature.41–43
hand, the design of fluorescent chemical sensors for neutral The synthesis of the fluorescent probes is also illustrated in
organic species presents a more serious challenge due to Scheme 1.
the lack of sufficient electronic changes upon inclusion.
Synthesis of 3-(and 4-) bromo-7H-benzimidazo[2,1-a] benz[d,e]-
Saccharides are generally soluble in polar protic solvents but isoquinolin-7-ones (NO). 4-Bromo-1,8-naphthalic anhydride
such solvents are competitive with the guest in a hydrogen (36 mmol) and o-phenylene diamine (55 mmol) were refluxed
bonding receptor. However, the design of a fluorescence in glacial acetic acid (120 ml) for 7 h and washed off with water
chemical sensor based on the boronic acid–saccharide inter- to obtain the desired product. The crude product was recrys-
action has been difficult due to the lack of sufficient electronic tallized with acetic acid to yield the compound NO (73%).
variations found in either the boronic acid moiety or the
Synthesis of 3-(and 4-)(2-aminoathylamine)-7H-benzimid-
saccharide moiety. These above-mentioned disadvantages of azo(2,1-a)benz(d,e) iso-quinolin-7-ones (NON). NO (0.17 mmol),
boronic acid–saccharide interactions can be overcome by mod- ethylenediamine (5 ml) and CuSO4Á5H2O (0.2 g) were refluxed
ifying the boronic acid binding site to create an electron-rich in 2-methoxyethanol (100 ml) for 10 h. The liquor was poured
center around the boronic acid moiety. Given the stronger into water to yield NON dye recrystallized from toluene as
acidity of boronic acid upon polyol-binding as well as the tight orange needles (57%).
and reversible binding between sugars and boronic acid, we
Synthesis of (2-(((2-((7-oxo-7H-benzo[d,e]benzo[4,5]imidazo
synthesized two PET probes for the measuring of saccharides. A [2,1-a]isoquinolin-3yl)amino)ethyl)amino)methyl)phenyl) boronic
new water-soluble naphthalimide dye was designed with fused acid (NONB). The NONB probe was synthesized by refluxing
loops. Increasing the number of rings in the naphthalimide
fluorophore can lead to greater electron absorption from the
electron-rich center at the lower part of the molecule to create a
probe with low fluorescence intensity based on the PET mechanism.
In addition, a nitrogen around the boronic acid moiety in the
(2-(((2-((7-oxo-7H-benzo[d,e]benzo[4,5]imidazo[2,1-a]isoquinolin-
3yl)amino)ethyl)amino)methyl)phenyl)boronic acid (NONB)
probe and sulphur around the boronic acid moiety in the
4-thiophene-N-propyl (triethoxysilane)1,8-naphthalimide (NST)
probe were designed. The effects of the designed groups for
detection of monosaccharides are discussed further.
Experimental
Materials
Acenaphthene, ortho-phenylenediamine, and bromomethyl-
phenyl boronic acid pinacol ester were purchased from
Merck Chemical Co. and used as received. All solvents were
of analytical grade. 2-5-Thiophenediyl-bisboronic acid was
Scheme 1 Synthesis procedure for the NONB and NST probe:
(a) o-phenylenediamine, glacial acetic acid, reflux 7 h, 73%; (b) ethylene
diamine, 2-methoxyethanol, reflux 10 h, 57%; (c) (2-bromomethylphenyl)
purchased from Aldrich Co. The following experiments were
conducted in PBS/DMF (99 : 1).
boronic acid pinacol ester, dry THF/MeOH, reflux 6 h, 25%; (d) APTS, ethanol,
85% (e) thiophene-2,5-diyldiboronicacid, Pd(PPh3), toluene, reflux, 28 h.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018 New J. Chem., 2018, 42, 19872--19880 | 19873

View Article Online
NJC
Paper
compound NON (0.56 mmol) with (2-bromomethylphenyl) Anal. calcd (%) for: C20H16N4O: C, 73.17; H, 4.88; N, 17.07.
boronic acid pinacol ester (0.94 mmol) in a 40 ml mixed Found (%): C, 73.15; H, 5.11; N, 16.98.
solution of THF : CH3OH (v : v = 1 : 1). Then, 2 ml of triethyl-
Synthesis of (2-(((2-((7-oxo-7H-benzo[d,e]benzo[4,5]imidazo
amine was added drop-wise to the above solution for 6 h. The [2,1-a]isoquinolin-3-yl)amino)ethyl)amino)methyl)phenyl) boronic
solvent was removed under reduced pressure and the product acid (NONB). Yield (25%), orange powders, m.p. 140–141 1C, lmax
was purified on silica gel using [5 : 1 CH2Cl2 : CH3OH] to yield (log e) = 467 nm, FTIR (KBr) cmÀ1: 3200–3400 (–NH st and BO-H
orange powders (25%).
st), 3056 (aromatic C–H st), 2919 (aliphatic C–H st.), 1712 (CQO),
Synthesis of 4-bromo-N-propyl (triethoxysilane)1,8-naphthal- 1695 (CQN), 1227 (C–N), 3430 (OH), 1100 (B–O), 659 (B–C).
imide (NS). The intermediate compound NS was prepared 1H NMR (DMSO, 500 MHz) d ppm: 9.46 (1H,s, naphthalimide–
by the reaction of 0.275 mmol of compound 1 with 2.2 mmol NH–), 7.21–8.27 (13H, m, Ar-H), 4.21 (2H, s, –OH), 4.16 (1H, s,
of 3-aminopropyl-triethoxysilane in ethanol. The solvent was –NH–), 3.77 (2H, s, –CH2–), 3.36 (2H, t, –CH2–), 2.64 (2H, t, –CH2–).
evaporated via rotary evaporation and the crude product was Elemental analysis: Anal. calcd (%) for: C27H24N4O: C, 80.20;
recrystallized with ethanol to generate compound NS (85%).44
Synthesis of 4-thiophene-N-propyl (triethoxysilane)1,8-
naphthalimide (NST). The NST compound was synthesized
H, 5.94; N, 13.86. Found (%): C, 80.24; H, 5.92; N, 13.90.
Photophysical properties
via the palladium-catalyst Suzuki reaction.45 The intermediate NONB and NST dyes, as probes, are designed based on the
(NS) (0.5 mmol), Pd(PPh3)4 (0.01 mmol), K3PO4 (0.75 mmol) photoinduced electron transfer (PET) mechanism inhibition.
and 2,5-thiophenediyl-bisboronic acid (0.75 mmol) in DMF Sensors based on the PET mechanism are in the fluorophore–
were mixed, stirred, and refluxed for 28 h. The crude products spacer–receptor format.48–50 If PET occurs between the
were then concentrated under vacuum and further purified by fluorophore and the receptor, the fluorescence emission of
chromatography on silica gel with a solution of CH2Cl2 and the dye turns off. If it can occur efficiently, the fluorescence
CH3OH as a diluent.
of the fluorophore will be quenched. When the receptor binds
to the guest molecule, the PET in the system is inhibited and
the fluorescence intensity grows. Nowadays, PET is one of the
most extensively adopted mechanisms for the development of
chemical bulk sensors. When an electron donating group is
substituted on the naphthalimide ring, it usually enhances the
Results and discussion
General procedure for synthesis of the fluorescent probes
The general procedure for the synthesis of NONB and NST is PET mechanism at the C-4 position. Amine and thiophene
displayed in Scheme 1. 4-Bromo-1,8-naphthalic anhydride (1) was groups in the 4-position can enhance the fluorescence proper-
prepared according to the procedure reported previously.44–47 ties and acts as a donor group while the carbonyl group can act
Compound NO was obtained by the reaction of compound 1 as the acceptor group. N-Substituted 1,8-naphthalimide is
with ortho-phenylenediamine in glacial acetic acid. Compound an excellent D–p–A chromophore with good photophysical
NONB was synthesized by the condensation of compound and photochemical properties. 1,8-Naphthalimide is a polar
NO with (2-bromomethylphenyl) boronic acid pinacol ester in molecule in both the ground and excited states. Substitution
THF : CH3OH (v : v = 1 : 1). Next, compound NS was synthesized of electron donating or electron withdrawing groups at the C-4
via the reaction of compound 1 with APTS in ethanol. Finally, position of the naphthalimide core strongly affects their emis-
compound NST was prepared via the condensation of com- sion spectra. Fig. 1 demonstrates the absorption and emission
pound NS with thiophene-2,5-diyilboronic acid, catalyzed by spectra of NONB and NST probes. The Stokes shifts for the
Pd(PPh3)4 in DMF.
NONB probe and NST probe are large enough to avoid self-
Synthesis of 3-(and 4-) bromo-7H-benzimidazo[2,1-a] benz[d,e] quenching. On the other hand, the fluorescence quantum
isoquinolin-7-ones (NO). Yield (73%), yellow needles, m.p. 215– yields of NONB and NST (Table 1) in PBS were relatively low,
216 1C, FTIR (KBr) cmÀ1: 3055 (Ar C–H st), 1712 (CQO), 1695 suggesting the potential of both probes in the detection of
(CQN), 1227 (C–N), 706 (C–Br). 1H NMR (DMSO, 500 MHz) saccharides under physiological conditions.
d ppm: 8.79–8.74 (1H, m, naphthyl ring), 8.61–8.63 (1H, m,
The water solubility of both designed probes in PBS 50 mM
naphthyl ring), 8.47–8.50 (1H, m, naphthyl ring), 8.39 (1H, m, (pH = 7.4) was examined by measuring the absorbance of the
naphthyl ring), 8.21–8.28 (1H, m, naphthyl ring), 8.99–8.05 probes at various concentrations (0–3 Â 10À2 mM) according to
(1H, m, Ar-H), 7.86 (1H, m, Ar-H), 7.50 (2H, m, Ar-H). Elemental the Beer Lambert law. Fig. 1 indicates that both probes were
analysis: Anal. calcd (%) for: C18H9BrN2O: C, 61.91; H, 2.59; completely soluble in PBS within the concentration range of
N, 8.02. Found (%): C, 61.88; H, 2.55; N, 8.11.
(0–3 Â 10À2 mM) (5 Â 10À2 mM was used in this work). Therefore,
Synthesis of 3-(and 4-)(2-aminoathylamine)-7H-benzimid- the subsequent tests were conducted at pH = 7.4 PBS/DMF (99 : 1).
azo(2,1-a) benz (d,e) iso-quinolin-7-ones (NON). Yield (57%), The structure of the NONB and NST probes, called selectophores,
orange needles, m.p. 207–208 1C, FTIR (KBr) cmÀ1: 3500 (–NH2–), are displayed in Scheme 2.
2940 (CH2), 1700 (CQO). 1H NMR (DMSO, 500 MHz) d ppm:
9.41 (2H, s, naphthalimide–NH–) 8.27–7.31 (5H, m, naphthyl
Sensing properties
ring), 7.55–7.21 (4H, m, imidazole ring), 3.31 (2H, t, ethylene), 2,73
Spectral response of the fluorescent probes toward sugars.
(2H, t, ethylene), 1.49 (2H, s, amine –NH2). Elemental analysis: In order to study the sensing properties of the synthesized
19874 | New J. Chem., 2018, 42, 19872--19880 This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online
Paper
NJC
Fig. 2 (a) UV-Vis spectral changes of NONB in the presence of different
concentrations of D-fructose (0–2 M), [NONB] = 5 Â 10À2 mM, 50 mM
PBS, pH = 7.4, lex = 391 nm, and (b) UV-Vis spectral changes of NST in
the presence of different concentrations of D-fructose (0–2 M), [NST] =
5 Â 10À2 mM, 50 mM PBS, pH = 7.4, lex = 379 nm.
Fig. 1 (a) Absorption and emission spectra of the NONB probe in PBS,
[NONB] = 5 Â 10À2 mM, lexNONB = 391 nm, 50 mM PBS, pH = 7.4, (b) the
linear correlation between the absorbance and the NONB probe concen-
tration, 50 mM PBS, pH = 7.4, (c) absorption and emission spectra of
the NST probe in PBS, [NST] = 5 Â 10À2 mM, lex NST = 379 nm, 50 mM PBS,
pH = 7.4, and (d) the linear correlation between the absorbance and the
NST probe concentration, 50 mM PBS, pH = 7.4.
Table 1 The association constants (Kas) between NONB and NST with
fructose and their detection limits (LOD) and quantum yield in the
presence and absence of fructose
Kas (MÀ1
)
LOD (mM)
jfa
FEb
a
Probe
NONB-fructose
NONB blank
NST-fructose
NST blank
80.64
—
35.84
—
0.049
—
0.1
—
0.2
2.22
—
2
Fig. 3 (a) Fluorescence change of NONB (5 Â 10À2 mM) in the presence
of D-fructose (0–2 M) in PBS (50 mM, pH 7.4), (b) linear range of the
fluorescence change of NONB at 514 nm in the presence of D-fructose
(lex = 391 nm), (c) fluorescence change of NST (5 Â 10À2 mM) in the
presence of D-fructose (0–2 M) in PBS (50 mM, pH 7.4), and (d) linear
range of the fluorescence change of NST at 515 nm in the presence of D-
fructose (lex = 379 nm).
0.09
0.14
0.07
—
a
Fluorescein (j = 0.95 in ethanol) was used as the reference, [NONB] =
5 Â 10À2 mM, [NST] = 5 Â 10À2 mM, [fructose] = 100 mM, 50 mM PBS,
b
pH 7.4. Fluorescent enhancement (FE) is the ratio of quantum yield
with sugar and that without sugar.
The fluorescence emission of the probes revealed enhance-
ments upon the addition of 0–2 M of D-fructose. The fluores-
cence titration spectra of NONB and NST with various amounts
of D-fructose are displayed in Fig. 3. Fluorescence enhance-
ments of both probes are observed upon the addition of
D-fructose. The introduction of D-fructose induced about
6.4-fold and 4-fold fluorescence enhancements for NONB and
NST, respectively. The relative signal changes in transitioning
from zero to 100 mM of D-fructose are highly dependent on
buffer pH (Fig. 4). By varying the pH from 2 to 11, the fluores-
cence of the probes without D-fructose is higher at acidic pH
than at neutral and basic pHs. In probes based on the PET
fluorophores (NONB and NST), and introduce them as selecto-
phores, the probes (10À2 mM in PBS solution) were titrated by
different concentrations of D-fructose. As shown in Fig. 2, by
increasing the fructose concentration, the absorption of both
probes grew.
Fig. 4 Fluorescence change of NONB and NST probes in the presence of
Scheme 2 Structure of the NONB and NST selectophores.
D-fructose in 50 mM PBS with different pH values.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018 New J. Chem., 2018, 42, 19872--19880 | 19875

View Article Online
NJC
Paper
mechanism, with reduction of pH, the donor moiety at the end
of the molecule would be protonated; PET is interrupted and the
fluorescence emission grows. As a result, enhanced fluorescence
emission confirms the performance of the probes according to
the PET mechanism by decreasing the pH. In the absence of any
sugar, the fluorescence intensity of NONB and NST diminished
upon altering the pH from 2 to 11. Also, an apparent pKa of 7.8
was observed for NST, which is assigned to the boronic acid
moiety. An apparent pKa of 6 was observed for NONB. In the
presence of fructose, the fluorescence intensity was tested within
the pH 2–11 region. The fluorescence intensity increased with
elevation of pH. An apparent pKa of 5.5 for NST and 4 for NONB
were observed. As has been demonstrated by past studies,
boronic acids with lower pKa values tend to have higher binding
constants. The apparent pKa of 6 for NONB was lower than that
of NST (7.8) determined under similar conditions. Due to the
presence of a nitrogen group near the boron atom, the NONB
Fig. 6 Plots of (Imax À I0)/(Ic À I0) against cÀ1 for NONB and NST probes
with fructose sugar. [NONB] = [NST] = 5 Â 10À2 mM, [fructose] = 50 mM,
50 mM PBS, pH 7.4.
probe has been able to reduce pKa to 4 when binding to fructose. As displayed in Fig. 6, the line slope for the NONB probe is less
The nitrogen atom has formed a dative bond with the boron than that of the NST probe. On the other hand, the association
atom, resulting in a reduction in the pKa. However, the NST probe constant value of the NONB probe is greater than that of the
only dropped to 5.5, due to the lack of binding between the sulfur NST probe. According to the results obtained, the line slope and
atom and the boron atom when linked to the fructose. As a result, the association constant values, the sensitivity of the NONB
the greater reduction in the pKa of the NONB probe because of the probe to fructose is greater than that of the NST probe.
NÁÁÁB bonding can probably lead to a higher binding constant as
The linear response ranges for the NONB probe were
compared to the NST probe. According to the results, the neutral obtained within [0.16–50 mM] with a regression R2 = 0.991,
pH (7.4) was selected for the following experiments.
and the linear response ranges for the NST probe were found
within [0.34–50 mM] with a regression R2 = 0.988. The associa-
tion constant, the detection limit (LOD) and quantum yield (Qy)
Selectivity
To investigate the selectivity of both probes, the fluorescence in the presence of fructose were measured and outlined
emission changes were evaluated in the presence of different in Table 1.
sugars. As illustrated in Fig. 5, in the presence of 50 mM of
LOD was measured using the equation 3Sb/K where Sb is the
fructose, sorbitol, mannose, glucose, galactose, and glucosamine, standard deviation of the blank. A total of 10 measurements
the designed probes only showed fluorescent changes toward were carried out, where K is the slope of the fitted area.
fructose. However, as can be observed, no significant changes Fluorescence measurements as well as the association constant
occurred in the case of other saccharides.
and LOD suggest that the NONB probe has a lower detection
The behavior of NONB and NST probes against saccharides limit than NST for fructose detection. The results of Fig. 5
follows a trend that is typical for aromatic boronic acids, where indicate a selectivity to fructose for both probes. Furthermore,
its affinity is as follows: D-fructose 4 sorbitol 4 D-galactose 4 10 replicate measurements of 100 mM fructose yielded a
D-glucose 4 mannose 4 glucoseamine. The selectivity patterns reproductive intensity with the relative standard deviation
represent the reorganization of the diol (cis or trans configu- (R.S.D) of 1.31% and 1.2% for NONB and NST, respectively.
ration), and the sterical hindrance via e.g. the hydroxymethyl Furthermore, five independent optical sensors were prepared
group of the saccharides.51
using the same procedure and used for the determination of
The binding interaction between the probes and fructose was fructose (100 mM) with the R.S.D of 1.7 and 1.8% for NONB and
further studied via association constant measurement (Fig. 6). NST, respectively, suggesting good repeatability and reproduci-
bility of this current optical sensor.
Sensing mechanism
Boronic acids are capable of binding to compounds containing
diol groups through the formation of reversible ester compounds.
The interaction between boronic acid and diol is highly
pH-dependent. The boronic acid group is an electron-deficient
Lewis acid with a sp2-hybridized boron atom and a trigonal
structure. With the increase of pH, the anionic form of boronic
acid develops with a more electron rich sp3-hybridized boron
atom with a tetrahedral structure. The changes in the electronic
and structural properties in the boron atom influence the
Fig. 5 Intensity changes of NONB (5 Â 10À2 mM) and NST (5 Â 10À2 mM)
upon addition of different saccharides (50 mM) in PBS (50 mM, pH7.4) in
the absence and presence of (50 mM) fructose.
19876 | New J. Chem., 2018, 42, 19872--19880 This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online
Paper
NJC
(in the presence of fructose). For this reason, the NONB probe with
the nitrogen moiety near to the boron atom has a higher sugar
binding constant than that of the NST probe with a sulfur
moiety. Consequently, the NONB probe tends more positively
to the trigonal form due to the dative bond formation (by
lowering the pKa) and a more stable complex is established in
the presence of a nitrogen atom with the sugar molecule. In
addition to the presence of the nitrogen group in reducing pKa,
in the structure of the NONB probe, with the closure of the
naphthalimide rings and the increase in rings, the probe
becomes more rigid. However, there is a flexible chain on the
top of the naphthalimide loop in the NST probe. Increasing the
rigidity of the NONB structure helps to strengthen the PET
Scheme 3 Reaction mechanism of boronic acid with diol with high pKa in
alkaline solution.45
fluorescence emission changes in the probes.52 The interaction mechanism and the electrons are stretched with more power
mechanism of the boronic acid with diol is illustrated in from the receptor. The NONB probe examined in this study had
Scheme 3. As revealed in Scheme 3, in the reaction systems a lower detection limit in detecting saccharides compared to
consisting of phenylboronic acid (pKa = 8.72) with fructose other fluorescent probes in other studies. Thus, this probe can
(pKa = 12.3) at pH = 7.4, the problem of proton ambiguity potentially be a good candidate for measuring fructose in real
is resolved, as the ligand concentrations are [H2L c HLÀ] at samples at physiological pH. The binding of the designed probes
pH = 7.4 and the kinetics reactivity order is H2L 4 HLÀ. With with fructose is shown in Scheme 4.
reference to Scheme 3, the structure changes from the state of
The studies performed on naphthalimide fluorescence
trigonal (RB(OH)2) to the tetrahedral state (intermediate I). probes containing boronic acids for the detection of fructose
With the structure change, the coordination number varies are shown in Table 2.
from three to four, where these changes occur very fast. It is
As reported in Table 2, the performance of the fluorescence
followed by rate-determining chelate-ring closure in the tetra- probe designed in the present work was in most cases better
hedral intermediate I to form intermediate HBL, which may be than the previous literature. A study conducted in 2002 focused
in fast trigonal-tetrahedral equilibrium with the chelated trigo- on the position of the group of boronic acids placed above the
nal species (BL).53 The stability of the boronate ester is pH naphthalimide ring in the ortho and para positions. Since 2006,
dependent. The optimal pH for the binding of boronic acid various naphthalimide dyes have been designed based on the
compounds to diols is above the pKa of the boronic acid PET mechanism for the detection of fructose. In this research
species. Therefore, reduction of the pKa of a boronic acid is work, more fused loops were used for the first time in the
known to increase the binding constant of boronic acids.45 detection of saccharides in comparison with other studies
Conversion of the boronic acid to the ester results in dimin- on improving the performance of naphthalimide dyes based
ished pKa.53 When using boronic acid with a high pKa value on the PET mechanism. The presence of fused loops causes
(pKa = 7–9), RB(OH)2 is always reactive irrespective of the pH of stronger PET in the system due to higher rigidity of the
the solution; it is reactive in a strong alkaline solution since the
reactivity of RB(OH)2 is far higher than that of RB(OH)3À. In
this study, we used the neutral pH and boronic acid with a high
pKa value (pKa = 7–9) and we expect the RB(OH)2 (trigonal form)
À
formation to be greater than the RB(OH)3 formation (tetra-
hedral form). As has been concluded by past studies, boronic
acids with lower pKa values tend to have higher binding
constants. The apparent pKa of NONB (6) is lower than that
of NST (7.8) in the absence of sugars determined under similar
conditions. This might be a major contributing factor to the
higher binding constant for NONB toward fructose. A further
decrease of pKa yields further increase in the sugar binding
constant. For this reason, the NONB probe including a nitrogen
moiety reduces the pKa to 4 and the NST probe including a
sulfur moiety decrease the pKa to 5.5. As a result, the NONB
probe has a higher sugar binding constant than the NST probe.
The reason is that the sulfur atom can be weakly bonded to the
boron atom.54 However, with the reaction between boronic acid
and fructose, the Lewis acidity of boron increases and the nitrogen
heteroatom can create a dative bond with the free orbit of boronic
Scheme 4 Binding of fluorescent probes with fructose: (1) NONB
and (2) NST.
acid, where this bond formation leads to a further reduction in pKa
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018 New J. Chem., 2018, 42, 19872--19880 | 19877

View Article Online
NJC
Paper
Table 2 Naphthalimide fluorescent probes for fructose detection
Stock
shift (nm)
Mechanism of
detection
Association
Linear
range (mM)
LOD
(mM)
Structure
pH
7.4
constant (Ka (MÀ1))
DIf
Year
Ref.
50
55
Flourescence enhancement
Quenching
80.64
0.16–50
NR
3.5
0.049
—
2018
2002
This work
1.6,
1.3
7.5
7.5
—
32
15
120
Flourescence enhancement
170
1–20
2
0.1 mM
2006
77
77
Flourescence enhancement
Flourescence enhancement
7.4
7.4
57
28
NR
NR
2.5
2.5
2–2.3 M
2007
2008
30
31
—
85
Flourescence enhancement
Flourescence enhancement
9
0–50 mM
0–2000
2.8
50
2.7
2014
2015
29
17
NR
130
7.4
75.8
0.045
NR = no report.
backbone of the dye and greater reduction in the initial Different chromatographic methods were developed for
fluorescence of the dye, resulting in a further increase in saccharide determination. Some of the analytical methods for
fluorescence emission and a lower detection limit by binding fructose determination are described in Table 3, and as can be
to the sugar molecule. In addition to their remarkable fluores- seen the present work can detect low amounts of fructose and is
cence properties, they are known as an important class of drugs a suitable candidate for developing a new sensing method for
in antitumor therapy and thus characterized by high cytotoxi- fructose detection.
city activity against a variety of tumor cells55 due to the
presence of a coplanar fluorophore, p-deficient aromatic sys-
tem and one or two basic side chains, some of which have
shown promising antitumor activities. On the other hand, a
Table 3 Comparing the LODs of different analytical techniques for
fructose
new generation of probes for saccharide detection with drug
activity and a lower detection limit has the potential to be
created and studied. This new structure can also be used in the HPLC-RI
fluorescence imaging of carbohydrates in future studies.
The LOD of this sensor is 0.049 mM (0.45 ppb) for fructose.
Chromatography is the general method for all sugar analysis. Fluorescence sensor
Techniques
LOD
Ref.
57.86 mg LÀ1
56
57
58
59
LC
HPLC-RID
HPLC -SPE
3.0 mg LÀ1
0.03–0.56 mg LÀ1
(3.30–16.48) Â 103 mg LÀ1
0.049 mM (0.00045 mg LÀ1
)
This work
19878 | New J. Chem., 2018, 42, 19872--19880 This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online
Paper
NJC
8 C. Ke, H. Destecroix, M. Crump and A. Davis, Nat. Chem.,
2012, 4, 718–723.
Analytical performance of the optical probes
To check the analytical performance of the probes in real
matrixes, the fructose content of a standard solution of fructose,
named high-fructose corn syrup 55 (HFCS-55) was determined
using the standard addition method. As expected, the NONB
probe was able to determine the fructose concentration of the
standard fructose sample successfully (the relative error was less
than 4.7% for three replicate measurements). This suggests that
this probe can potentially be used for quantitatively detecting
fructose levels in real samples like blood and urea.
9 Z. Xu and P. Deng, Mater. Sci. Eng., 2014, 40, 228–235.
10 C. H. Wang and H. Zhu, Anal., 2013, 138, 7146–7151.
11 H. Chen and L. Li, RSC Adv., 2015, 5, 13085–13092.
12 W. Zhai, X. Sun, T. D. James and J. S. Fossey, Chem. – Asian
J., 2015, 10, 1836–1848.
13 J. P. Lorand and J. Q. Edwards, J. Org. Chem., 1959, 24,
769–774.
14 S. D. Bull, M. G. Davidson, J. M. H. Van Denelson, J. S. Fossey,
A. T. A. Jenkins, Y. Jiang, Y. Kubo, F. Marken, K. Sakurai,
J. Zhao and T. D. James, Acc. Chem. Res., 2013, 46, 312–326.
15 S. Trupp, A. Schweitzer and J. Mohr, Org. Biomol. Chem.,
2006, 4, 2965–2968.
Conclusions
In summary, we developed selective ‘‘turn-on’’ fluorescence
sensors for fructose detection by novel naphthalimide probes
based on photoinduced electron transfer. The photophysical
characteristics of the dyes were evaluated in the presence of
fructose. The probe including the amine moiety near the boron
atom revealed greater selectivity and sensitivity to fructose than
the probe including a sulfur moiety. The remarkable enhance-
ment in the fluorescence intensity of the NONB probe upon the
addition of fructose makes it potentially a good candidate for
fabricating a highly sensitive and selective probe for quantita-
tively detecting fructose levels in real samples such as blood
and urea.
16 X. Gao, Y. Zhang and B. Wang, New J. Chem., 2005, 29,
579–586.
17 H. Chen, L. Li, H. Guo, X. Wang and W. Qin, RSC Adv., 2015,
5, 13805–13811.
18 Y. J. Huang, W. J. Quyang, X. Wu, Z. Li, J. S. Fossey,
T. D. James and Y. B. Jiang, J. Am. Chem. Soc., 2013, 135,
1700–1721.
19 L. N. Neupane, S. Y. Han and K. H. Lee, Chem. Commun.,
2014, 50, 5854–5862.
20 J. S. Hansen, M. Ficker, J. F. Peterson, J. B. Christensen and
T. Hoeg-Jensen, Tetrahedron Lett., 2013, 54, 1849–1862.
21 B. Xu, J. Hou, K. Li, Z. Lu and X. Yu, Chin. J. Chem., 2015, 33,
101–111.
22 G. Kaur, H. Fang, X. Gao, H. Li and B. Wang, Tetrahedron,
2006, 62, 2583–2589.
Conflicts of interest
23 Z. T. Xing, H. C. Wang, Y. X. Cheng, T. D. James and
C. J. Zhu, Chem. – Asian J., 2011, 6, 3054–3058.
24 X. Sun, W. Zhai, J. S. Fossey and T. D. James, Chem.
Commun., 2016, 52, 3456–3469.
25 M. Faraco, D. Fico, A. Pennetta and G. E. D. Benedetto,
Talanta, 2016, 159, 40–46.
26 J. Yoon and A. W. Czarnik, J. Am. Chem. Soc., 1992, 114,
5874–5875.
27 H. Suenaga, M. Mikami, K. R. A. S. Sandanayake and
S. Shinkai, Tetrahedron Lett., 1995, 36, 4825–4828.
28 H. Suenaga, H. Yamamoto and S. Shinkai, Pure Appl. Chem.,
1996, 68, 2179–2186.
There are no conflicts to declare.
Acknowledgements
We would like to acknowledge the Center of Excellence for
Color Science and Technology (CECST), Institute for Color
Science and Technology, Tehran-Iran and University of Tehran
for providing us with a good environment and facilities to
complete this project.
Notes and references
29 N. DiCesare and J. R. Lakowicz, J. Phys. Chem. A, 2001, 105,
6834–6840.
30 N. DiCesare and J. R. Lakowicz, Tetrahedron Lett., 2001, 42,
9105–9108.
1 C. Tran, Nutrients, 2017, 9, 356–366.
2 T. Mogos and A. Evelin Iacobini, Rom J. Diabetes Nutr.
Metab. Dis., 2016, 23, 081–085.
31 T. D. James, K. Sandanayake and S. Shinkai, J. Chem. Soc.,
Chem. Commun., 1994, 477–478.
3 H. Li, H. M. Byers and A. Diaz, Mol. Genet. Metab., 2018, 123,
428–432.
32 T. D. James, K. R. A. S. Sandanayake, R. Iguchi and
S. Shinkai, J. Am. Chem. Soc., 1995, 117, 8982–8987.
33 S. Seraj and S. Rouhani, J. Fluoresc., 2017, 27, 1877–1883.
34 S. Rouhani and S. Haghgoo, Sens. Actuators, B, 2015, 209,
957–965.
4 A. Quaglia, E. A. Roberts and M. Torbenson, Macsween’s
Pathology of the Liver, Developmental and Inherited Liver
Disease, 7th edn, 2018, ch. 3, pp. 111–274.
5 U. Boas, P. Lind and U. Riber, Anal. Biochem., 2014, 465,
73–80.
35 S. Rouhani, K. Gharanjig and M. Hosseinnezhad, Green
Chem. Lett. Rev., 2014, 7, 174–178.
6 Q. Hou, J. Wang, X. Liu and Z. Wang, Chem. Res. Chin. Univ.,
2012, 28, 947–952.
36 X. Sun, B. Zhu, D. K. J. Ji, Q. Chen, X. P. He, G. R. Chen and
T. D. James, ACS Appl. Mater. Interfaces, 2014, 6, 10078–10082.
7 W. Sun, L. Du and M. Li, Curr. Pharm. Des., 2010, 16,
2269–2278.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018 New J. Chem., 2018, 42, 19872--19880 | 19879

View Article Online
NJC
Paper
37 J. Wang, S. Jin, S. Akay and B. Wang, Eur. J. Org. Chem., 50 Y. Fu, X. Pang, Z. Wang and H. Qu, Molecules, 2018, 23,
2007, 2091–2099. 376–390.
38 S. Jin, J. Wang, M. Li and B. Wang, Chem. – Eur. J., 2008, 14, 51 J. Zhao and T. D. James, J. Mater. Chem., 2005, 15,
2795–2804. 2896–2901.
39 N. Dicesare, D. P. Adhikari, J. Heynekamp, M. Heagy and 52 E. Watanabe, C. Miyamoto, A. Tanaka, K. Lizuka, S. Iwatsuki,
J. R. Lakowicz, J. Fluoresc., 2002, 12, 147–154.
40 A. M. Brouwer, Pure Appl. Chem., 2011, 83, 2213–2228.
M. Inamo, H. D. Takagi and K. Ishihara, Dalton Trans., 2013,
42, 8446–8453.
´
´
´
´
´
41 R. Herraez-Hernandez, P. Campıns-Falko and J. Verdu´-Andres, 53 Y. Furikado, T. Nagahato, T. Okamoto, T. Sugaya, S. Iwatsuki,
J. Biochem. Biophys. Methods, 2002, 54, 147–167.
42 N. I. Georgiev, I. S. Yaneva, A. R. Surleva, A. M. Asiri and
V. B. Bojinov, Sens. Actuators, B, 2013, 184, 54–63.
43 V. B. Bojinov, I. P. Panova and J.-M. Chovelon, Sens.
Actuators, B, 2008, 135, 172–180.
M. Inamo, H. D. Takagi, A. Odani and K. Ishihara,
Chem. – Eur. J., 2014, 20, 1–10.
54 G. F. Whyte, R. Vilar and R. Woscholski, J. Chem. Biol., 2013,
6, 161–174.
55 U. H. Sk, A. S. P. Gowda, M. A. Crampsie, J. K. Yun,
T. E. Spratt, S. Amin and A. K. Sharma, Eur. J. Med. Chem.,
2011, 46, 3331–3338.
44 P. H. Grayshan, A. M. Kadhim and A. T. Peters, J. Heterocycl.
Chem., 1974, 11, 33–37.
45 L. Xiea, J. Cui, X. Qian, Y. Xu, J. Liu and R. Xu, Bioorg. Med. 56 A. A. F. Zielinski, C. M. Braga, I. M. Demiate, F. L. Beltrame,
Chem., 2011, 19, 961–967.
A. L. Nogueira and G. Wo, Food Sci. Technol., 2014, 34,
46 Q. Meng, X. Zhang, C. He, P. Zhou, W. Su and C. Duan,
Talanta, 2011, 84, 53–59.
38–43.
´
´
57 P. D. Duarte-Delgado, C. Narvaez-Cuenca, L. Restrepo Sanchez,
´
47 Y. Nagao, T. Yoshida, K. Arimitsu and K. Kozawa,
Heterocycles, 2010, 80, 1197–1213.
A. Kushalappa and T. Mosquera-Vasqueza, J. Chromatogr. B:
Anal. Technol. Biomed. Life Sci., 2015, 975, 18–23.
48 X. Qian, Y. Xiao, Y. Xu, X. Guo, J. Qian and W. Zhu, Chem. 58 R. Hadjikinova, N. Petkova, D. Hadjikinov, P. Denev and
Commun., 2010, 46, 6418–6436. D. Hrusavov, J. Pharm. Sci. Res., 2017, 9, 1263–1269.
49 P. Kucheryary, R. Khatmullin, E. M. Irzakulova, D. Zhou and 59 W. Xu, L. Liang and M. Zhu, Int. J. Food Prop., 2015, 18,
K. D. Glusac, J. Phys. Chem. A, 2011, 115, 11606–11614. 547–557.
19880 | New J. Chem., 2018, 42, 19872--19880 This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018