Design, synthesis and applications of fluoride probe based on aromatization of isoquinolinium salts

Article abstract of DOI:10.1016/j.dyepig.2020.108547

A novel ratiometric fluorescent probe based on aromatization drive of isoquinolinium salts was designed, synthesized to recognize fluoride. The applications have shown that the probe has high selectivity and sensitivity to fluoride. Fluoride induces a cascade reaction of hydroxyl deprotection and 1,6-oxidative elimination, and causes a remarkable fluorescence enhancement. The relative fluorescent intensities (I₃₉₈/I₅₂₂) increased linearly with F^? concentration in the range of 0–9.0 μmol L^?1. The limit of detection was very low (3.7 nM) due to the intrinsic aromatization drive of probe induced by F^?. The recognition mechanism was demonstrated by mass spectrum (MS) and ¹H NMR. The in vitro imaging results showed that the new probe BIQS (2-[4-(tert-butyl-diphenyl-silanyloxy)-benzyl]-3-(4-methoxy- phenyl) -isoquinolinium nitrate) was membrane-permeable and could be applied into the determination of fluoride ions in living cells.

Full text of DOI:10.1016/j.dyepig.2020.108547

Dyes and Pigments 181 (2020) 108547
Contents lists available at ScienceDirect
Dyes and Pigments
journal homepage: http://www.elsevier.com/locate/dyepig
Design, synthesis and applications of fluoride probe based on aromatization
of isoquinolinium salts
a,
*
a
b
,
**
c
a
d,***
Yun-Hui Zhao , Yajun Yu , Tao Guo
, Zhihua Zhou , Zilong Tang , Li Tian
a
School of Chemistry and Chemical Engineering, Hunan University of Science and Technology, Xiangtan, Hunan, 411201, PR China
School of Chemistry and Chemical Engineering, Henan University of Technology, Zhengzhou, Henan, 450001, PR China
b
c
Hunan Provincial Key Laboratory of Controllable Preparation and Functional Application of Fine Polymers, Hunan University of Science and Technology, Xiangtan, PR
China
d
School of Materials Science and Engineering, Hunan University of Science and Technology, Xiangtan, PR China
A R T I C L E I N F O
A B S T R A C T
Keywords:
A novel ratiometric fluorescent probe based on aromatization drive of isoquinolinium salts was designed, syn-
thesized to recognize fluoride. The applications have shown that the probe has high selectivity and sensitivity to
fluoride. Fluoride induces a cascade reaction of hydroxyl deprotection and 1,6-oxidative elimination, and causes
Isoquinolinium salts
Fluoride determination
Ratiometric chemosensor
Aromatization
À
a remarkable fluorescence enhancement. The relative fluorescent intensities (I398/I522) increased linearly with F
À 1
À
concentration in the range of 0–9.0
μ
mol L . The limit of detection was very low (3.7 nM) due to the intrinsic
aromatization drive of probe induced by F . The recognition mechanism was demonstrated by mass spectrum
(
MS) and 1H NMR. The in vitro imaging results showed that the new probe BIQS (2-[4-(tert-butyl-diphenyl-
silanyloxy)-benzyl]-3-(4-methoxy- phenyl) -isoquinolinium nitrate) was membrane-permeable and could be
applied into the determination of fluoride ions in living cells.
1
. Introduction
With the rapid development of fluorescent analytical sensing and
ratiometric fluorescent probes that harvest the target-induced fluores-
cent intensity changes from two different emission bands were explored.
Most importantly, the ratiometric fluorescent sensing characteristic
improves signal-to-noise ratios and enhances accuracy for much reliable
quantification of specific analytes.
optical imaging, fluorescent probes based on organic dyes have become
important tools in the research of biological and medical fields because
of their high sensitivity and technical simplicity. The fluorescent signal
of chemosensors will be inhibited or enhanced during the interaction
with the target species by quenching off or removing quenching effects.
Fluorescent probes are commonly used in many areas such as analytical
chemistry, biochemistry, medicine and environmental monitoring
because of the real-time and on-line detection or monitoring of target
objects with the help of changes in fluorescent signals [1–5]. The fluo-
rescent probes are roughly divided into three types: fluorescence
quenching, fluorescence enhancement and ratiometric probes. The first
two types provide only a single emission feature, and it may be difficult
to quantify the target analyte due to the interference from various
analyte-independent factors, such as instrument parameters, the
microenvironment surrounding the probe molecule, and the local con-
centration of the probe [6–10]. To conquer this limitation, the
Fluoride is one of the important trace elements which play crucial
roles in human body and is an indispensable component of teeth and
bones [11,12]. The proper amount of fluoride in body not only makes
the bone hard, but also benefits the uptake of calcium and phosphorus to
consolidate the formation of bone [13,14]. Therefore, fluoride ion is
often observed in toothpaste and in pharmaceutical agents as an
essential ingredient. On the other hand, over deposition of fluoride in
human body would lead to dental and skeletal fluorosis, which can cause
severe toxicity to the biological tissue and even lead to many serious
neurodegenerative diseases [15–18]. For example, in cell biology sys-
tem, an excessive absorption of sodium fluoride can affect a number of
essential cellular signal transduction processes and obstruct normal
cellular metabolism [19]. In view of the importance of its homeostasis in
physiology, tremendous effort has been devoted to the development of
*
Corresponding author.
*
*
* Corresponding author.
** Corresponding author.
E-mail address: zhao_yunhui@163.com (Y.-H. Zhao).
https://doi.org/10.1016/j.dyepig.2020.108547
Received 16 April 2020; Received in revised form 10 May 2020; Accepted 12 May 2020
Available online 18 May 2020
0
143-7208/© 2020 Elsevier Ltd. All rights reserved.

Y.-H. Zhao et al.
Dyes and Pigments 181 (2020) 108547
Scheme 1. Possible mechanism of isoquinolinium salt for fluoride detection.
highly selective and sensitive chemosensors for the detection of fluoride,
in biological systems, even in living animals. Among the reported
methods, such as ion selective electrodes (ISE) [20–22], high perfor-
mance liquid chromatography (HPLC) [23,24], spectrophotometry [25],
ion chromatography [26,27], fluorimetry and fluorescent probes
2.2. Synthesis of benzyl isoquinolinium salts
General procedure: AgNO (0.5 mmol, 85 mg) was added to the so-
3
lution of 2-(arylethynyl)benzaldehyde (0.5 mmol) and benzylamine
(0.50 mmol, 1.0 euqiv) in 1,2-dichloroethane (DCE, 3 ml). Then, the
ꢀ
[
28–32], fluorescence sensing is a powerful method for recognition of
reaction mixture was stirred at 80 C for 12 h. After been cooled to room
fluoride in biological systems, even in living animals due to its superi-
orities of high sensitivity and selectivity, simple operation, low cost.
In recent years, several types of fluorescent probes for recognition of
fluoride were performed by different reaction mechanisms, e.g. the
formation of hydrogen bonds between NH and fluoride ion, the
complexation of boron-fluoride, and the desilylation reaction by fluoride
temperature, H
extracted with DCM (3 � 10 ml). The combined extracts were washed
with brine and dried over anhydrous Na SO4, filtered, and the solvent
2
O (5 ml) was added to the mixture. The mixture was
2
evaporated to dryness. The crude residue was subjected to silica gel
(200–300 mesh) chromatography eluted with DCM/MeOH (20:1, v/v)
to afford the desired products. All synthesized dyes were fully charac-
1
13
[
33–37]. However, there are some drawbacks associated with existing
terized and confirmed by H NMR, C NMR, IR and HRMS spectroscopy
(see supporting information Figs. S1–S27).
sensing methods. For example, the “turn-on” fluorescent anion chemo-
sensors for selective fluoride determination often require complicated
preparation procedures [38–40]. In addition, the addition of cetyl tri-
methyl ammonium bromide (CTAB) was necessary to increase the rate
of desilylation in aqueous solution for some probes [41]. In order to
address the solubility in aqueous solution, several cationic probes were
designed and exhibited excellent selectivity and sensitivity for fluoride
2-Benzyl-3-(4-methoxy-phenyl)-isoquinolinium nitrate (3a), pale
yellow solid, M.p 94–97 ꢀC. IR (KBr) cm 3431, 2924, 2853, 2421,
À 1
2049, 2025, 1639, 1608, 1504, 1458, 1380, 1251, 1180, 1093, 1021,
1
905, 840, 765, 716, 695, 575, 540, 512, 473. H NMR (500 MHz, CDCl
3
)
δ 10.769 (s, 1H), 8.615 (d, J ¼ 8.0 Hz, 1H), 8.025 (t, J ¼ 7.5 Hz, 1H),
7.959–7.936 (m, 2H), 7.828 (t, J ¼ 7.5 Hz, 1H), 7.227–7.186 (m, 2H),
[
42,43]. These studies further indicated that the N-alkylated cationic
7.159–7.101 (m, 3H), 6.945 (d, J ¼ 8.0 Hz, 2H), 6.839 (d, J ¼ 7.5 Hz,
þ
probes have the potential ability to target mitochondria of living cells.
Isoquinoline derivatives are nitrogen-containing compounds with good
2H), 5.976 (s, 2H), 3.820 (3.820, 3H). HRMS calcd for C23
326.1539, found 326.1539.
H20NO [M]
fluorescence emission [44,45]. Due to the “push-pull”
π
-electron system,
2-Benzyl-3-p-tolyl-isoquinolinium nitrate (3b), yellow oily liquid, IR
À 1
isoquinolinium salts have better fluorescence emission than isoquino-
lines. If a isoquinoline is combined with a reactive site to form a fluo-
rescent probe isoquinolinium salt for ratio sensing and imaging of
anions, the probe will release the leaving group and restore to the parent
structure isoquinoline in the presence of the target detection analyte.
Methylene benzoquinone is a very good leaving molecule via oxidative
elimination because of the stable electronic structure [46–49]. In this
work, we developed a novel fluorescent probe which takes advantage of
a fluoride-induced hydroxyl deprotection and aromatization of iso-
quinolinium salt for the detection of fluoride (Scheme 1). The probe has
three advantages: detecting fluoride in aqueous solution without the
addition of CTAB; high detection sensitivity due to the intramolecular
aromatization; offering visual detection of fluoride on paper strip.
(KBr) cm 3432, 2923, 2853, 2048, 2025, 1635, 1500, 1383, 573, 515.
1
H NMR (500 MHz, CDCl
3
) δ 10.786 (s, 1H), 8.633 (d, J ¼ 8.0 Hz, 1H),
8.043 (t, J ¼ 7.5 Hz, 1H), 7.990–7.963 (m, 2H), 7.840 (t, J ¼ 7.5 Hz, 1H),
7.259 (d, J ¼ 8 Hz, 2H), 7.198–7.152 (m, 3H), 7.117 (t, J ¼ 7.5 Hz, 2H),
1
3
6.847 (d, J ¼ 7.5 Hz, 2H), 5.980 (s, 2H), 2.399 (s, 3H). C NMR (125
MHz, CDCl ) δ 153.2, 146.2, 141.4, 137.8, 137.7, 133.5, 131.8, 131.2,
3
129.9, 129.7, 129.1, 129.1, 128.9, 128.2, 127,5, 127.3, 126.6, 62.1,
þ
21.5. HRMS calcd for C23
H
20N [M] 310.1590, found 310.1590.
2-Benzyl-3-phenyl-isoquinolinium nitrate (3c), yellow oily liquid. IR
À 1
(KBr) cm 3432, 2921, 2845, 2048, 2025, 1715, 1635, 1381, 1227,
572. 1H NMR (500 MHz, CDCl
) δ 10.822 (s, 1H), 8.648 (d, J ¼ 7 Hz,
3
1H), 8.059 (t, J ¼ 7.5 Hz, 1H), 8.005–7.983 (m, 2H), 7.854 (t, J ¼ 7.5 Hz,
1H), 7.521 (t, J ¼ 7.5 Hz, 1H), 7.441 (t, J ¼ 7.5 Hz, 2H), 7.301 (d, J ¼
7
.5 Hz, 2H), 7.157 (t, J ¼ 7.5 Hz, 1H), 7.100 (t, J ¼ 7.5 Hz, 2H), 6.797 (d,
13
2
. Experimental
J ¼ 7.5 Hz, 2H), 5.978 (s, 2H). C NMR (125 MHz, CDCl
3
) δ 153.3,
1
45.9, 137.8, 137.7, 133.3, 131.8, 131.8, 131.3, 130.9, 129.8, 129.2,
2
.1. Chemicals and instrumentation
129.1, 129.1, 128.2, 127.5, 127.4, 126.7, 62.3. HRMS calcd for C22
H
18
N
þ
[
M] 296.1434, found 296.1433.
Ortho-alkynylarylaldehydes were synthesized according to
2-Benzyl-3-(4-chloro-phenyl)-isoquinolinium (3d), yellow oily
À 1
Ref. [50–54] Benzylamine (98%) and silver nitrate (99%) were pur-
chased from Guoyao Chemical Reagent Co. (Shanghai). All other normal
reagents and solvents were of analytical-reagent grade and used directly
without further purification. Deionized water was employed throughout
liquid. IR (KBr) cm 3432, 2924, 2026, 1711, 1638, 1605, 1491, 1359,
1
1088, 839, 763. H NMR (400 MHz, CDCl
3
) δ 10.63 (s, 1H), 8.58 (d, J ¼
8.4 Hz, 1H), 8.05–8.04 (m, 3H), 7.85–7.81 (m, 1H), 7.41 (d, J ¼ 8.4 Hz,
2H), 7.31 (d, J ¼ 8.4 Hz, 2H), 7.20–7.11 (m, 3H), 6.79 (d, J ¼ 7.2 Hz,
À 3
À 1
13
the whole detection procedure. The stock solution of 1.0 � 10 mol L
3
2H), 5.94 (s, 2H). C NMR (100 MHz, CDCl ) δ 153.1, 144.7, 137.8,
fluoride ion was prepared by dissolving 26.1 mg of Bu
4
NF in 100 ml
137.7, 137.2, 133.1, 131.6, 131.4, 131.3, 130.2, 129.4, 129.1, 128.8,
þ
deionized water. Stock solutions of other control anions were obtained
by dissolving appropriate amounts of their soluble sodium salts in
deionized water.
128.1, 127.7, 127.4, 126.9, 62.6. HRMS calcd for C22
H17ClN [M]
330.1044, found 330.1043.
2-Benzyl-3-(4-chloro-phenyl)-7-methyl-isoquinolinium nitrate (3e),
ꢀ
À 1
yellow solid. M.p 173–175 C. IR (KBr) cm 3434, 3033, 2960, 2920,
2

Y.-H. Zhao et al.
Dyes and Pigments 181 (2020) 108547
2
8
851, 2049, 2025, 1636, 1612, 1497, 1454, 1376, 1198, 1088, 1015,
1
3
53, 831, 765, 736, 703, 560, 558, 511, 474. H NMR (500 MHz, CDCl )
δ 10.574 (s, 1H), 8.335 (s, 1H), 7.918–7.879 (m, 3H), 7.387 (d, J ¼ 8 Hz,
H), 7.264 (d, J ¼ 8 Hz, 2H), 7.168 (t, J ¼ 7.5 Hz, 1H), 7.115 (t, J ¼ 7.5
2
1
3
Hz, 2H), 6.786 (d, J ¼ 7.5 Hz, 2H), 5.938 (s, 2H), 2.543 (s, 3H). C NMR
(
125 MHz, CDCl
3
) δ 152.5, 144.0, 142.6, 140.2, 137.2, 136.0, 133.3,
1
2
31.3, 130.3, 130.1, 129.4, 129.2, 128.1, 127.8, 127.2, 126.5, 62.1,
þ
1.9. HRMS calcd for C23H19ClN [M] 344.1201, found 344.1200.
2
-Benzyl-7-chloro-3-(4-chloro-phenyl)-isoquinolinium nitrate (3f),
ꢀ
À 1
yellow solid. M.p 161–164 C. IR (KBr) cm 3417, 3061, 3030, 2924,
854, 2054, 2026, 1643, 1602, 1516, 1490, 1451, 1379, 1189, 1082,
2
1
1
022, 1000, 929, 833, 724, 697, 672, 602, 578, 511, 474. H NMR (500
MHz, CDCl
3
) δ 10.744 (s, 1H), 8.520 (s, 1H), 8.016–7.969 (m, 2H), 7.882
d, J ¼ 8.5, 1H), 7.368 (d, J ¼ 8.5 Hz), 7.306 (d, J ¼ 8.5 Hz, 2H), 7.158
t, J ¼ 7.5 Hz, 1H), 7.097 (t, J ¼ 7.5 Hz, 2H), 6.773 (d, J ¼ 7.5 Hz, 2H),
(
(
5
1
1
.949 (s, 2H). 13C NMR (125 MHz, CDCl
37.4, 136.1, 132.9, 131.3, 130.0, 129.7, 129.4, 129.3, 129.2, 128.6,
28.2, 128.2, 127.5, 62.9. LCMS calcd for C22 l2N [M]＋ 364.0,
16Cl N [M] 364.0654, found
3
) δ152.5, 145.2, 138.4, 137.6,
16
H C
þ
found 363.9. HRMS calcd for C22
64.0653.
-Benzyl-7-chloro-3-phenyl-isoquinolinium nitrate (3g), yellow
H
2
3
2
ꢀ
À 1
solid, M.p 89–91 C. IR (KBr) cm 3430, 3050, 3016, 2920, 2851, 2049,
025, 1638, 1598, 1515, 1492, 1451, 1383, 1188, 1102, 1078, 930, 907,
2
8
1
1
33, 765, 729, 701, 601, 571, 510, 479. H NMR (500 MHz, CDCl
3
) δ
0.897 (s, 1H), 8.588 (s, 1H), 8.022–7.989 (m, 2H), 7.921 (dd, J ¼ 1.5
Hz, 8.5 Hz, 1H), 7.514 (t, J ¼ 7.5 Hz, 1H), 7.432 (t, J ¼ 7.5 Hz, 2H),
7
2
.325 (d, J ¼ 7.5 Hz, 2H), 7.152 (t, J ¼ 7.5 Hz, 1H), 7.089 (t, J ¼ 7.5 Hz,
13
H), 6.800 (d, J ¼ 7.5 Hz, 2H), 5.997 (s, 2H). C NMR (125 MHz,
) δ 152.5, 146.4, 138.5, 137.5, 136.2, 133.0, 131.6, 131.0, 129.8,
CDCl
3
1
29.3, 129.2, 129.1, 128.5, 128.3, 128.2, 127.4, 62.6. HRMS calcd for
þ
C
22
H
17ClN [M] 330.1044, found 330.1043.
2
-Benzyl-7-methoxy-3-phenyl-isoquinolinium nitrate (3h), pale yel-
ꢀ
À 1
low solid, M.p 98–101 C. IR (KBr) cm 3430, 3055, 3008, 2923, 2853,
051, 2026, 1614, 1498, 1452, 1382, 1216, 1180, 1106, 1021, 937, 838,
2
7
1
63, 613, 566, 514. H NMR (500 MHz, CDCl ) δ 10.717 (s, 1H), 8.014
3
(
s, 1H), 7.878–7.852 (m, 2H), 7.650 (dd, J ¼ 2 Hz, 9 Hz, 1H), 7.512 (t, J
7.5 Hz, 1H), 7.432 (t, J ¼ 7.5 Hz, 2H), 7.267 (d, J ¼ 7.5 Hz, 2H), 7.155
t, J ¼ 7.5 Hz, 1H), 7.099 (t, J ¼ 7.5 Hz, 2H), 6.797 (d, J ¼ 7.5 Hz, 2H),
¼
Fig. 1. (a) Fluorescent emission of compounds 3a and BIQS in MeCN. (b)
Fluorescent emission of compounds 3a-3h in MeCN.
(
5
1
1
3
.926 (s, 2H), 3.966 (s, 3H). 13C NMR (125 MHz, CDCl
43.8, 133.4, 131.6, 130.8, 129.9, 129.2, 129.1, 129.1, 128.1, 127.9,
3
) δ 161.4, 150.9,
ꢀ
incubate at 37 C for half an hour in 5% CO . Then the cells were washed
2
26.9, 107.9, 62.2, 56.6. LCMS calcd for C23
H20NO [M]＋ 326.1, found
with phosphate buffer saline (PBS) and new prepared fluoride solution
þ
ꢀ
26.0. HRMS calcd for C23
H
20NO [M] 326.1539, found 326.1538.
was added. After incubation for 30 min at 37 C and washing twice with
2
-[4-(Tert-butyl-diphenyl-silanyloxy)-benzyl]-3-(4-methoxy-
PBS, the fluorescence images were acquired with an Olympus IX81
inverted microscope with an Olympus FV1000 confocal scanning
system.
phenyl)-isoquinolinium nitrate (BIQS), pale yellow solid, M.p 99–102
C, IR (KBr) cm 3441, 2927, 2853, 2048, 2025, 1704, 1635, 1506,
ꢀ
À 1
1
1
383, 1256, 1179, 1110, 1028, 916, 826, 703, 572. H NMR (500 MHz,
CDCl
3
) δ 10.675 (s, 1H), 8.580 (d, J ¼ 7.5 Hz, 1H), 7.999 (t, J ¼ 7.5 Hz,
3. Results and discussion
1
7
H), 7.936 (d, J ¼ 8.0 Hz, 1H), 7.880 (s, 1H), 7.804 (t, J ¼ 7.5 Hz, 1H),
.572 (d, J ¼ 7.5 Hz, BIQS), 7.342 (t, J ¼ 7.5 Hz, 2H), 7.272 (t, J ¼ 7.5
The traditional method for preparing isoquinoline salt is to utilize the
benzylation of isoquinoline. However, this method has the disadvan-
tages of low yield and difficult purification. We developed a new method
for preparation of benzyl isoquinolinium salts from 2-ethynylbenzalde-
hyde 1 and benzylamine 2. The optimal reaction conditions were ob-
tained by screening the reaction conditions using 2-((4-methoxyphenyl)
ethynyl)benzaldehyde 1a and benzylamine 2 as model substrates. The
Hz, BIQS), 7.123 (d, J ¼ 8.0 Hz, 2H), 6.842 (d, J ¼ 8.0 Hz, 2H), 6.563 (d,
J ¼ 8.5 Hz, 2H), 6.510 (d, J ¼ 8.0 Hz, 2H), 5.809 (s, 2H), 3.766 (s, 3H),
0
1
1
.975 (s, 9H). 13C NMR (125 MHz, CDCl
3
) δ161.3, 156.3, 152.9, 145.9,
37.7, 137.4, 135.4, 132.4, 131.7, 131.3, 131.1, 130.1, 129.7, 127.8,
27.4, 126.6, 125.8, 123.9, 120.3, 114.6, 62.0, 55.6, 26.4, 19.4. HRMS
þ
calcd for C39
.3. Cell imaging
The bioimaging application of BIQS for fluoride sensing in living
H
38NO
2
Si [M] 580.2666, found 580.2668.
desired product 3a could be obtained in 66% yield under the presence of
ꢀ
2
100% mol AgNO
3
in DCE at 80 C. The target compound was confirmed
by comparison with the known isoquinolinium salts [55] with a typical
3
chemical shifts at 10.7 ppm. The substituents at the benzyl group R
cells was researched. CellTiter 96® AQueous One Solution Cell Prolif-
eration Assay was purchased from Promega (Madison, WI, USA). Cell
culture media was purchased from Thermo Scientific HyClone (MA,
USA). Hela cells were cultured in Dulbecco modified eagle medium
would have little effect on the fluorescent property of isoquinolinium
salts because the conjugated structure of the dye molecular skeleton is
not affected. In order to verify this assumption, the probe BIQS was
synthesized under the same reaction condition. As seen from Fig. 1a, the
fluorescent emission spectra of compounds 3a and BIQS were almost the
same. The result confirmed the previous deduction. Consequently, a
(
DMEM) supplemented with 10% (v/v) heat-inactivated fetal calf
ꢀ
serum, 100
μ
g/ml penicillin, under an atmosphere of 5% CO
2
at 37 C in
a humidified air. The cells were treated with BIQS and were allowed to
3

Y.-H. Zhao et al.
Dyes and Pigments 181 (2020) 108547
3
Scheme 2. Different substituent isoquinolinium salts were prepared under the presence of AgNO in DCE.
series of isoquinolinium salts with no substituents at benzyl R3 was
prepared to study their spectral property without the need to synthesize
relatively complex probe molecules. We would then synthesize the
target probe based on the spectral properties of these compounds.
Therefore, other benzylisoquinolinium salts having different sub-
1
2
stituents at R and R were obtained under optimal reaction conditions
in moderate to good yields (Scheme 2).
All isoquinolinium salts were stable under ambient conditions, and
were soluble in common organic solvents, such as DCM, MeCN, EtOH
and THF. The maximum absorption wavelengths of these compounds
were between 350 and 380 nm (Fig. S28). To our delight, these iso-
quinolinium salts exhibited excellent fluorescence emission at 400–600
nm. As seen from Fig. 1b, the maximum emission wavelengths shifted
2
from 522 to 445, 421, and 428 nm, when the substituents at R changed
from methoxyl (3a) to methyl (3b), hydrogen (3c), and chlorine (3d),
respectively. Compound 3a showed the largest emission wavelength and
highest intensity among these isoquinolinium salts because of the
strongest electron-donating group methoxyl [56]. However, when the
2
substituent at R was diethylamino (not show), the maximum emission
Fig. 2. Fluorescent emissions of BIQS, 4 and the mixed solution of BIQS
peak shifted to a shorter wavelength than that of compound 3a and the
intensity was also very weak. This phenomenon may be explained by the
intramolecular charge transfer due to the presence of diethylamino. On
the other hand, the maximum emission wavelengths exhibited an
and fluoride.
medium in the following experiments. The color of probe system tuned
from yellow-green to colorless quickly after the addition of an equiva-
lent fluoride. Coincidentally, the fluorescence spectrum of the probe
after the addition of fluoride was very similar to that of isoquinoline 4
containing the corresponding parent structure (Fig. 2). The above results
indicated that after the diphenyl-tert-butylsilyl group (TBDPS) of the
dye BIQS was captured by fluoride, an oxidative elimination reaction
was initiated by aromatization to afford the isoquinoline 4.
1
increasing trend when the substituents at R changed from methyl (3e)
to hydrogen (3d) and chlorine (3f). Based on the above results, the dye
BIQS having the same dye moiety of 3a, which has the best fluorescence
performance among the obtained isoquinolinium salts, was chosen as
the parent structure for the determination of fluoride.
A detailed study of BIQS’s spectral properties was conducted before
the recognition of fluoride. The UV–vis spectra of BIQS (10
μM in MeCN)
Then the titration experiments were carried out via the addition of
displayed an absorption peak at 348 nm and the fluorescent spectra
showed an emission peak at 522 nm (Fig. S29). In order to apply this dye
to an aqueous environment, the effect of MeCN volume concentration on
fluorescence properties was investigated, and the results were summa-
rized in Fig. S30. It indicated that the fluorescence intensity enhanced
gradually with the increased concentration of MeCN in mixed solution.
The fluorescent emission quenching may be caused by the aggregation
of probe molecule in aqueous solution. To facilitate actual sample
À
increasing amounts of F to the BIQS in 90% MeCN–H
2
O mixture. It can
be seen from Fig. 3 that the emission wavelength of BIQS appeared at
around 520 nm and the Stokes shift was about 170 nm (λex ¼ 350 nm).
À
After the addition of F , the peak at around 520 nm weakened gradually,
and a new peak at around 400 nm began to emerge. The large shift
(
about 120 nm) between the two peaks would reduce the interference of
background noise and increase the detection accuracy, which is a better
advantage than the normal ratiometric or single signal sensors. With an
2
analysis, a 90% MeCN–H O solution was selected as the reaction
4

Y.-H. Zhao et al.
Dyes and Pigments 181 (2020) 108547
À 1
Fig. 4. Fluorescent emission changes of BIQS (10
various anions in 90% MeCN–H
μ
mol L ) upon addition of
2
O solution. Incubation reaction was performed
for 30 min. The fluorescence responses were obtained with 350 nm excitation
and collected at 398 and 522 nm.
À
À
2À
À
2À
2À
NO
2
, NO
3
, CO
3
, AcO , SO
3
, SO in CH
4
3
CN/H
2
O ¼ 9/1 (v/v). After
each of these anions was added for half an hour, the fluorescence spectra
of solution BIQS were recorded. It can be seen from Fig. 4, a small
À
amount of F (0.5 equiv.) gave sharp increase in fluorescence spectra at
3
98 nm and drastically decrease at emission 522 nm. No obvious
changes were observed upon addition of other anions even at high
concentration (3 equiv.). Due to the high affinity of fluoride to silicon,
the reaction-based probe BIQS only triggered by fluoride and undergo
an oxidative-elimination process to provide isoquinoline 4. Conse-
quently, these results indicated that our probe has a high selectivity
À 1
Fig. 3. (a) Fluorescent emission changes of BIQS (10
μ
mol L ) in the presence
À
of increasing amounts of F (0, 0.2, 0.4, 0.6, 0.8, 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0,
8
.0, 9.0, 10
μ
mol LÀ 1) in 90% MeCN–H
2
O (v/v) and the linear relationship
À
toward F ions.
between the relative fluorescent intensities (I398/I522, F) and concentration of
fluoride. Incubation reactions were performed for 20 min. The fluorescence
responses were obtained with 350 nm excitation and collected at 398 and 522
nm. (b) Solid-state fluorescence response of the BIQS with different concen-
trations of fluoride under UV illumination at 365 nm.
In order to verify our proposed fluoride-induced deprotection group,
1
followed by the oxidation-elimination mechanism, H NMR spectra was
À
applied to monitor the dynamic process of F recognition with BIQS in
CDCl
3
before and after the addition of fluoride. It could be found from
Fig. 5 that BIQS showed two sharp single peaks around 10.7 and 5.8
–
CH (benzyl iso-
quinolinium salts) and the protons CH Ph respectively. The H NMR
ppm, corresponding to the signals of the proton N
–
increasing amount of fluoride, the peak at around 400 nm enhanced
gradually, and the peak at around 520 nm weakened accordingly. The
calibration curve showed that there was an excellent linear relationship
1
2
signals of BIQS didn’t change obviously under the UV illumination for 3
h in the absence of fluoride ion, which demonstrated the dye was stable
. After the addition of fluoride in 1min, the signal of the N–
between the relative intensities (I398/I522) over the fluoride concentra-
À 1
in CDCl
3
–
CH
tion range of 0–9.0
calculated to be F ¼ 0.269c [
limit of detection was calculated from three times the standard deviation
μ
mol L . The linear regression equation was
À 1
2
was weakened dramatically and showed 1.4 ppm upfield shift, while a
μmol L ]-0.0313 (n ¼ 11, R ¼ 0.991). The
new peak at around 9.3 ppm emerged immediately. The new emerged
–
peak coincided with the N CH of isoquinoline 4 completely [50]. The
À 9
À 1
–
of the background noise to be 3.7 � 10 mol L . This value was much
lower than the limit concentration of fluoride (211 M) in drinking
phenomenon further confirmed the proposed mechanism that the probe
was induced by fluoride and undergone an oxidative-elimination pro-
cess to form isoquinoline immediately after the protected group been
deprotected. Further evidence was confirmed by the gradually disap-
μ
water determined by the United States Environmental Protection
Agency.
The sensing of fluoride on filter paper was subsequently conducted
c
À 3
pearance of the peak at 8.6 ppm corresponding to proton H . The signals
after drop-casting 12
μ
L of 1.0 � 10 M BIQS solution in MeCN and
a
c
of H and H were almost disappeared after 20 min. The nuclear mag-
netic dynamic monitoring process implies that the concentration of
fluoride can be determined in a short time. The fast detection process
indicated that the intrinsic aromatization drive of isoquinolinium salts
was very strong.
drying. Then, tetrabutylammonium fluoride (TBAF) solutions from 1.0
À 3
À 6
�
10 M to 1.0 � 10 M were added onto the stained filter paper, and
pictures were recorded under ultraviolet illumination, which were
shown in Fig. 3B. It was obvious that the dye BIQS on paper strip pro-
vided concentration-dependent fluorescence intensity decrease response
À 3
MS was further used to confirm the desilylation process induced by
fluoride. No desired peaks were observed during the deprotection pro-
cess. Fortunately, the desired intermediate, hydroxybenzyl iso-
quinolinium salt, was found in the synthesis of isoquinoline-like
reactions. The reaction between 4-hydroxybenzylamine and o-alkyny-
to fluoride, and 1.0 � 10 M fluoride anion could be easily identified by
the naked eye because the probe molecular structure was destroyed in
the presence of fluoride. This represented the dye which could provide a
simple and visual method for fluoride recognition on paper strip.
À
The higher selectivity towards F compare to other potentially
3
larylalde catalyzed by AgNO produced isoquinoline 4 via a cation in-
relevant anions is very crucial for a new fluoride sensor. For this reason,
termediate (corresponding to the peak 342.0). It would experience a 1,6-
oxidative elimination reaction to release methylene benzoquinone
the anions response property of BIQS (10
μ
M) was subsequently inves-
À
À
À
À
tigated by adding F and other various anions, such as Br , Cl , IO
4
,
5

Y.-H. Zhao et al.
Dyes and Pigments 181 (2020) 108547
1
Fig. 5. The changes of H Chemical shift (ppm) under the presence of fluoride with increasing of reaction time.
Fig. 6. The intermediates of this determination process were detected by MS.
corresponding to the peak 107.0) and provided another fluorescent DMEM with BIQS (10 M) for half an hour, followed by washing three
(
μ
molecule isoquinoline (corresponding to the peak 236.1) because of the
driving of aromatization (Fig. 6). The reaction process was completely
consistent with our assumption.
times with PBS buffer. The cells were then incubated with 10 and 30
μ
M
À
F
for half an hour. Immediate visualization of fluorescence response of
the probe BIQS to fluoride was carried out after been washed twice with
PBS buffer. As showed in Fig. 7, the control cells displayed a strong green
fluorescence after been stained with the probe BIQS. However, the
cellular fluorescence intensity decreased obviously after the addition of
Enlightened by the positive results, the fluorescence response of
fluoride ion in HeLa cells using the probe BIQS was further investigated
through fluorescence confocal microscopy. HeLa cells were incubated in
6

Y.-H. Zhao et al.
Dyes and Pigments 181 (2020) 108547
Fig. 7. Fluorescence images (λex ¼ 405 nm) of HeLa cells incubated with BIQS (10
μ
M) for 30 min and then further incubated with fluoride (10, 30
μ
M) for another
ꢀ
3
0 min at 37 C. Image of green channel (left), bright field (middle) and their composition. Scale bar ¼ 30
μ
m. (For interpretation of the references to color in this
figure legend, the reader is referred to the Web version of this article.)
fluoride ion. In addition, higher concentrations of fluoride (30
μ
M)
Provincial Education Department Scientific Research Fund (No.
18B221) are gratefully acknowledged.
caused more pronounced fluorescence quenching of the probe compared
to lower concentrations (10 M). These results indicated that the dye
μ
BIQS had membrane-permeability and could be utilized to the recog-
Appendix A. Supplementary data
nition and image of fluoride in living cells.
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.dyepig.2020.108547.
4
. Conclusion
In summary, we have described the design and synthesis of a novel
References
fluorescent probe BIQS for rapid and convenient recognition of fluoride.
The structure-activity relationship of the substituents on the dyes was
studied in detail, and it was found that the dye had the maximum
fluorescence emission when the R2 position was methoxy. The probe
exhibited higher fluoride selectivity over other anions, which was
ascribed to the high affinity of fluoride for silicone. The detection
mechanism was verified by mass spectroscopy analysis, and the dynamic
detection process was further monitored by a spectrum analysis of nu-
clear magnetic resonance. Due to its high affinity to fluoride and aro-
mitization drive, the probe was very sensitive to fluoride and has a very
low detection limit. BIQS had also been successfully applied for moni-
toring and imaging fluoride in HeLa cells under physiological
conditions.
[
1] Wang F, Wang L, Chen X, Yoon J. Recent progress in the development of
fluorometric and colorimetric chemosensors for detection of cyanide ions. Chem
Soc Rev 2014;43:4312–24.
0
0
[2] Ding Y, Li X, Li T, Zhu W, Xie Y.
α-Monoacylated and
α,
α
- and
α
,β -diacylated
dipyrrins as highly sensitive fluorescence “turn-on” Zn2þ probes. J Org Chem
2
013;78:5328–38.
[
[
3] Lee MH, Kim JS, Sessler JL. Small molecule-based ratiometric fluorescence probes
for cations, anions, and biomolecules. Chem Soc Rev 2015;44:4185–91.
4] Liu H-W, Chen L, Xu C, Li Z, Zhang H, Zhang X-B, Tan W. Recent progresses in
small-molecule enzymatic fluorescent probes for cancer imaging. Chem Soc Rev
2
018;47:7140–80.
[5] Li W, Gong X, Fan X, Yin S, Su D, Zhang X, Yuan L. Recent advances in molecular
fluorescent probes for organic phosphate biomolecules recognition. Chin Chem Lett
2
019;30:1775–90.
[
[
[
[
6] Chen X, Zhou Y, Peng X, Yoon J. Fluorescent and colorimetric probes for detection
of thiols. Chem Soc Rev 2010;39:2120–35.
7] Zhou Y, Xu Z, Yoon J. Fluorescent and colorimetric chemosensors for detection of
nucleotides. Chem Soc Rev 2011;40:2222–35.
Declaration of competing interest
The authors declare no conflict of interest.
Acknowledgements
8] Kim HN, Guo Z, Zhu W, Yoon J, Tian H. Recent progress on polymer-based
fluorescent and colorimetric chemosensors. Chem Soc Rev 2011;40:79–93.
9] Kim HN, Ren WX, Kim JS, Yoon J. Fluorescent and colorimetric sensors for
detection of lead, cadmium, and mercury ions. Chem Soc Rev 2012;41:3210–44.
[
[
10] Sun W, Guo S, Hu C, Fan J, Peng X. Recent development of chemosensors based on
cyanine platforms. Chem Rev 2016;116:7768–817.
11] Matsui H, Morimoto M, Horimoto K, Nishimura Y. Some characteristics of fluoride-
induced cell death in rat thymocytes: cytotoxicity of sodium fluoride. Toxicol Vitro
The generous financial support from the National Natural Science
Foundation of China (Nos. 51773057 and 21877034), the Hunan
2
007;21:1113–20.
7

Y.-H. Zhao et al.
Dyes and Pigments 181 (2020) 108547
[
[
[
[
[
[
[
12] Horowitz HS. The 2001 CDC recommendations for using fluoride to prevent and
control dental caries in the United States. J Publ Health Dent 2003;63:3–8.
13] Kleerekoper M. The role of fluoride in the prevention of osteoporosis. Endocrinol
Metab Clin N Am 1998;27:441–52.
[36] Ke B, Chen W, Ni N, Cheng Y, Dai C, Dinh H, Wang B. A fluorescent probe for rapid
aqueous fluoride detection and cell imaging. Chem Commun 2013;49:2494–6.
[37] Zhu B, Kan H, Liu J, Liu H, Wei Q, Du B. A highly selective ratiometric visual and
red-emitting fluorescent dual-channel probe for imaging fluoride anions in living
cells. Biosens Bioelectron 2014;52:298–303.
14] Newbrun EJ. What we know and do not know about fluoride. Public Health Dent
2
010;70:227–33.
[38] Selvaraj M, Rajalakshmi K, Nam Y-S, Lee Y, Song J-W, Lee H-J, Lee K-B. On-off-on
relay fluorescence recognition of ferric and fluoride ions based on indicator
displacement in living cells. Anal Chim Acta 2019;1066:112–20.
15] Arhima MH, Gulati OP, Sharma SC. The effect of Pycnogenol on fluoride induced
rat kidney lysosomal damage in vitro. Phytother Res 2004;18:244–6.
16] Ayoob S, Gupta AK. Fluoride in drinking water: a review on the status and stress
effects. Crit Rev Environ Sci Technol 2006;36:433–87.
[39] Krishnaveni K, Murugesan S, Siva A. Dual-mode recognition of biogenic amine
tryptamine and fluoride ions by a naphthyl hydrazone platform: application in
fluorescence imaging of HeLa cells and zebrafish embryos. New J Chem 2019;43:
9021–31.
17] Anuradha CD, Kanno S, Hirano S. Fluoride induces apoptosis by caspase-3
activation in human leukemia HL-60 cells. Arch Toxicol 2000;74:226–30.
18] Refsnes M, Schwarze PE, Holme JA, Lag M. Fluoride-induced apoptosis in human
epithelial lung cells (A549 cells): role of different G protein-linked signal systems.
Hum Exp Toxicol 2003;22:111–23.
[40] Zhou G, Cheng Y, Wang L, Jing X, Wang F. Novel polyphenylenes containing
phenol-substituted oxadiazole moieties as fluorescent chemosensors for fluoride
ion. Macromolecules 2005;38:2148–53.
[
[
19] Barbier O, Arreola-Mendoza L, Del Razo LM. Molecular mechanisms of fluoride
toxicity. Chem Biol Interact 2010;188:319–33.
[41] Hu R, Feng J, Hu D, Wang S, Li S, Li Y, Yang G. A rapid aqueous fluoride ion sensor
with dual output modes. Angew Chem, Int Ed 2010;49:4915–8.
À
20] Capka V, Bowers CP, Narvesen JN, Rossi RE. Determination of total fluorine in
blood at trace concentration levels by the Wickbold decomposition method with
direct potentiometric detection. Talanta 2004;64:869–78.
[42] Xu J, Sun S, Li Q, Yue Y, Li Y, Shao S. A novel “Turn-On” fluorescent probe for F
detection in aqueous solution and its application in live-cell imaging. Anal Chim
Acta 2014;849:36–42.
[
[
21] Ruiz-Payan A, Ortiz M, Duarte-Gardea M. Determination of fluoride in drinking
water and in urine of adolescents living in three counties in Northern Chihuahua
Mexico using a fluoride ion selective electrode. Microchem J 2005;1:19–22.
22] Hutchinson JP, Evenhuis CJ, Johns C, Kazarian AA, Breadmore MC, Macka M,
Hilder EF, Guijt RM, Dicinoski GW, Haddad PR. Identification of inorganic
improvised explosive devices by analysis of postblast residues using portable
capillary electrophoresis instrumentation and indirect photometric detection with
a light-emitting diode. Anal Chem 2007;79:7005–13.
[43] Shen Y, Zhang X, Zhang Y, Li H, Chen Y. An ICT-Modulated strategy to construct
colorimetric and ratiometric fluorescent sensor for mitochondria-targeted fluoride
ion in cell living. Sensor Actuator B Chem 2018;258:544–9.
[44] Xu J, Zhang Y, Yu H, Gao X, Shao S. Mitochondria-targeted fluorescent probe for
imaging hydrogen peroxide in living cells. Anal Chem 2016;88:1455–61.
[45] Tang Y, Yu Y, Wei X, Yang J, Zhu Y, Zhao Y-H, Tang Z, Zhou Z, Li X, Yu X. An
3
efficient approach to isoquinoline via AgNO -promoted 6-endo-dig cyclization
followed by oxidative elimination of o-alkynylarylaldimines and its application in
fluoride recognition. Tetrahedron Lett 2019. https://doi.org/10.1016/j.
tetlet.2019.151187.
[
23] Nishimoto J, Yamada T, Tabata M. Solvent extraction and fluorometric
determination of fluoride ion at ppb level in the presence of large excess of
aluminum(III) and iron(III) by using an expanded porphyrin, sapphyrin. Anal Chim
Acta 2001;428:201–8.
[46] Richard JP, Amyes TL, Toteva MM. Formation and stability of carbocations and
carbanions in water and intrinsic barriers to their reactions. Acc Chem Res 2001;
34:981–8.
[
[
24] Garrido M, Lista AG, Palomeque M, Band BSF. Fluorimetric determination of
fluoride in a flow assembly integrated on-line to an open/closed FIA system to
remove interference by solid phase extraction. Talanta 2002;58:849–53.
25] Xu XR, Li HB, Gu JD, Peng KJ. Determination of fluoride in water by reversed-
[47] Richard JP, Toteva MM, Crugeiras J. StructureÀ reactivity relationships and
intrinsic reaction barriers for nucleophile additions to a quinone methide: a
strongly resonance-stabilized carbocation. J Am Chem Soc 2000;122:1664–74.
[48] Toteva MM, Richard JP. StructureÀ reactivity relationships for addition of sulfur
nucleophiles to electrophilic carbon: resonance, polarization, and steric/
electrostatic effects. J Am Chem Soc 2000;122:11073–83.
À
3þ
phase high-performance liquid chromatography using F -La -alizarin
complexone ternary complex. Chromatographia 2004;59:745–7.
[
[
26] Li HB, Xu XR. Separation and determination of fluoride in plant samples. Talanta
1
999;48:57–62.
[49] Bai WJ, David JG, Feng ZG, Weaver MG, Wu KL, Pettus TRR. The domestication of
ortho-quinone methides. Acc Chem Res 2014;47:3655–64.
27] Themelis DG, Tzanavaras PD. Simultaneous spectrophotometric determination of
fluoride and monofluorophosphate ions in toothpastes using a reversed flow
injection manifold. Anal Chim Acta 2001;429:111–6.
[50] Zhao Y-H, Li Y, Guo T, Tang Z, Xie W, Zhao G. Selective synthesis of pyrazolo[5,1-
a]isoquinolines via 1,3-dipolar cycloaddition reaction. Tetrahedron Lett 2016;57:
2257–61.
[
28] Zhao Y-H, Li Y, Long Y, Zhou Z, Tang Z, Deng K, Zhang S. Highly selective
fluorescence turn-on determination of fluoride ions via chromogenic aggregation
of a silyloxy-functionalized salicylaldehyde azine. Tetrahedron Lett 2017;58:
[51] Zhao Y-H, Luo M, Li Y, Liu X, Tang Z, Deng K, Zhao G. Efficient synthesis of
isoquinolines by AgNO
aldehydes with ammonium bicarbonate, Chin. J Chem 2016;34:857–60.
[52] Zhao Y-H, Li Y, Luo M, Tang Z, Deng K. AgNO -mediated cyclization/N-N bond
3
-catalyzed sequential iminationÀ annulation of 2-alkynyl
1
351–5.
[
[
29] Mohapatra S, Das RK. Dopamine integrated B, N, S doped CQD nanoprobe for rapid
and selective detection of fluoride ion. Anal Chim Acta 2019;1058:146–54.
30] Qi F, Zhang F, Mo L, Ren X, Wang Y, Li X, Liu X, Zhang Y, Yang Z, Song X. A HBT-
based bifunctional fluorescent probe for the ratiometric detection of fluoride and
sulphite in real samples. Spectrochim Acta A 2019;219:547–51.
3
cleavage reaction for the synthesis of 3-aryl isoquinolines. Synlett 2016;27:
2597–600.
[53] Zhao Y-H, Luo Y, Zhu Y, Wang H, Zhou H, Tan H, Zhou Z, Ma Y-C, Xie W, Tang Z.
3
Synthesis of potential anticancer 1-(1H-indol-3-yl)-isoquinolines via AgNO -
[
[
31] Zhou Y, Zhang JF, Yoon J. Fluorescence and colorimetric chemosensors for
fluoride-ion detection. Chem Rev 2014;114:5511–71.
mediated tandem reactions of 2-alkynylbenzaldehyde azines and indoles. Synlett
2018;29:773–8.
32] Wu X, Wang H, Yang S, Tian H, Liu Y, Sun B. Highly sensitive ratiometric
fluorescent paper sensors for the detection of fluoride ions. ACS Omega 2019;4(3):
[54] Tian M, Zheng G, Fan X, Li X. Rhodium (III)-Catalyzed redox-neutral synthesis of
isoquinolinium salts via C–H activation of imines. J Org Chem 2018;83:6477–88.
[55] Ye Z-S, Guo R-N, Cai X-F, Chen M-W, Shi L, Zhou Y-G. Enantioselective iridium-
catalyzed hydrogenation of 1- and 3-substituted isoquinolinium salts. Angew Chem
Int Ed 2013;52:3685–9.
4
918–26.
[
[
33] Han J, Zhang J, Gao M, Hao H, Xu X. Recent advances in chromo-fluorogenic
probes for fluoride detection. Dyes Pigments 2019;162:412–39.
34] Zhang Y, Tu Q, Chen L, Li N, Yang L, Zhang X, Yuan M-S, Wang J. A fluorescein-
[56] Zhao Y-H, Luo Y, Wang H, Wei H, Guo T, Tan H, Yuan L, Zhang X-B. A novel
2
þ
À
2þ
based AND-logic FPSi probe for the simultaneous detection of Hg and F . Talanta
ratiometric and reversible fluorescence probe with a large Stokes shift for Cu
2
019;202:323–8.
based on a new clamp-on unit. Anal Chim Acta 2019;1065:134–41.
[
35] Zeng L, Yuan Y, Jiang C, Mu J, Li F, Wan Y, Xu H, Qu J, Huang P, Lin J. A near-
infrared turn-on probe for in vivo chemoselective photoacoustic detection of
fluoride ion. Dyes Pigments 2019;165:408–14.
8