A near-infrared and mitochondria-targeted fluorescence probe for ratiometric monitoring of sulfur dioxide derivatives in living cells

Article abstract of DOI:10.1039/c9tb01686d

A promising near-infrared emissive and mitochondria-targeted fluorescence probe (SNB) for the ratiometric detection of sulfur dioxide derivatives with a novel reaction mechanism was developed on the basis of FRET and the ICT platform. Probe SNB showed fav

Full text of DOI:10.1039/c9tb01686d

Journal of
Materials Chemistry B
View Article Online
View Journal | View Issue
PAPER
A near-infrared and mitochondria-targeted
fluorescence probe for ratiometric monitoring
of sulfur dioxide derivatives in living cells†
Cite this: J. Mater. Chem. B, 2019,
7, 6585
b
a
Ye-Hao Yan,a Xiao-Ying He,b Jun-Ying Miao
and Bao-Xiang Zhao
*
A promising near-infrared emissive and mitochondria-targeted fluorescence probe (SNB) for the ratiometric
detection of sulfur dioxide derivatives with a novel reaction mechanism was developed on the basis of FRET
and the ICT platform. Probe SNB showed favorable fluorescence properties, including an active NIR
emission signal (671 nm), ultra-large Stokes shift (251 nm) and an ultra-broad emission band gap (193 nm).
Furthermore, SNB was applied to detect sulfur dioxide derivatives with a low detection limit (17 nM),
excellent sensitivity and exceptional selectivity. Of greater significance, SNB also successfully implemented
the real-time and ratiometric surveillance for exogenous or endogenous sulfur dioxide derivatives in living
HeLa, L-O2 and HepG2 cells. Interestingly, SNB possessed a subcellular organelle targeting ability with a
preferred co-localization coefficient of 0.91 for mitochondria, which revealed its promising potential in the
detection of subcellular sulfur dioxide derivatives.
Received 9th August 2019,
Accepted 23rd September 2019
DOI: 10.1039/c9tb01686d
rsc.li/materials-b
state intramolecular proton transfer (ESIPT)7 and through bond
energy transfer (TBET)8 mechanisms were usually utilized to
Introduction
Sulfur dioxide (SO2) is a serious environmental pollutant
develop organic fluorescence functional probes. Nevertheless,
2À
that widely exists as HSO3À/SO3 in aqueous solutions, food
¨
the Forster resonance energy transfer (FRET) platform readily
sources, and even organisms.1 Nevertheless, some research
studies have reported that the disordered concentration of
realized the design of the ratiometric probe due to the optional
dual fluorophores in the probe molecule, which could be
constructed by a donor, linker and acceptor. Moreover, the
FRET-based probe exhibited the enhanced fluorescence of
the acceptor when it was excited by the excitation light of the
donor, which also offered a large Stokes shift and active
emission signal.9 Therefore, FRET platforms, for the develop-
ment of functional probes, have attracted increasing scientific
attention because of their high energy transfer efficiency, broad
emission window gap, and extended Stokes shift.10
Organic fluorescence probes with a near-infrared emission
window (NIR, lem = 650 nm) have received increasing techno-
logical attention in the areas of photodynamic therapy, gene
delivery and bioimaging due to their various advantages in
eliminating light scattering, photo-bleaching, optical absorption,
and the auto-fluorescence of an organism.11 In particular, human
tissue is nearly transparent to NIR emission. As important
subcellular organelles, mitochondria are power workshops for
metabolism.12 In general, HSO3À/SO32À is principally generated
in the process of protein metabolism.13 Therefore, it is very
2À
endogenous HSO3À/SO3 in an organism may trigger some
diseases, including angiocardiopathy, endocrine abnormalities
and cancer.2 Recently, many methods for the monitoring of
HSO3À/SO32À have been developed, including electrochemistry,
electroluminescence and chromatography. In comparison with
the above methods, organic fluorescent probes have acquired
increasing attention because of their various advantages,
including non-invasive visualization, facile accessibility, low
detection limit, excellent sensitivity and selectivity.3
In recent years, some fluorescent probes have been reported
for the detection of HSO3À/SO32À, but most depended on the
changes of a single emission signal. Ratiometric probes achieve
an accurate detection for the species by calculating the ratio of
the two emission bands, which is more suited to alleviate the
interference of conditions, concentrations and instruments,
and realize self-calibration.4 Generally, intramolecular charge
transfer (ICT),5 photo-induced electron transfer (PET),6 excited
a Institute of Organic Chemistry, School of Chemistry and Chemical Engineering,
Shandong University, Jinan 250100, P. R. China. E-mail: bxzhao@sdu.edu.cn
b Shandong Provincial Key Laboratory of Animal Cells and Developmental Biology,
School of Life Science, Shandong University, Qingdao 266237, P. R. China
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
c9tb01686d
urgent to develop a NIR emission and mitochondria-targeted
2À
probe for the ratiometric detection of HSO3À/SO3
.
Taking the abovementioned points into consideration, a
novel fluorescence probe named SNB was developed for
2À
the ratiometric detection of HSO3À/SO3 (Scheme 1). SNB was
This journal is ©The Royal Society of Chemistry 2019
J. Mater. Chem. B, 2019, 7, 6585--6591 | 6585

View Article Online
Paper
Journal of Materials Chemistry B
an excitation source, measurement type: wavelength scan, scan
mode: emission, excitation and emission slit widths: 5/5 nm,
scanning speed: 1200 nm minÀ1; voltage: 700 V). Quantum yields
were determined in a DCM solution by an Edinburgh-FLS920
spectrometer with an excitation wavelength of 420 nm (emission
region: 430–800 nm).
Synthesis route of probe SNB
À
Scheme 1 Structure of the probe SNB and its response toward HSO3
/
4-(4-(7-(Diethylamino)-2-oxo-2H-chromene-3-carbonyl)piperazin-1-yl)-
benzaldehyde (compound 1) and 1,4-dimethylquinolin-1-ium iodide
(compound 2) were obtained according to previous reports.15
Compound 1 (1.0 mmol, 433.5 mg) and compound 2 (1.0 mmol,
285.2 mg) were dissolved in EtOH (30.00 mL) and refluxed for
8 h under the catalysis of piperidine (0.5 mL). By column
chromatography (dichloromethane : methanol = 25 : 1), the targeted
compound (SNB) was obtained in 41% yield (Scheme 2). SNB was
2À
SO3
.
constructed by coumarin (donor), piperazine (linker) and (E)-4-(4-
(dimethylamino)styryl)-1-methylquinolin-1-ium (acceptor) based
on FRET and the ICT platform. SNB exhibited a clearly enhanced
NIR emission signal (671 nm) of the acceptor and nearly
quenched the emission signal of the donor (478 nm) due to the
high energy transfer efficiency of the FRET process from the
donor to the acceptor under the excitation of 420 nm light, which
provided a large Stokes shift (251 nm) and stable signal output.
The detection mechanism depended on the nucleophilic addition
1
confirmed by the following methods (Fig. S1–S4, ESI†). H NMR
(300 MHz, DMSO-d6, ppm): d = 1.14 (t, J = 7.05 Hz, 6H), 3.48–3.75
(m, 12H), 4.48 (s, 3H), 6.57 (d, J = 1.8 Hz, 1H), 6.75 (dd, J = 9.0 and
2.1 Hz, 1H), 7.07 (d, J = 9.0 Hz, 2H), 7.50 (d, J = 9.0 Hz, 1H), 7.89 (d,
J = 9.0 Hz, 2H), 7.98 (t, J = 7.8 Hz, 2H), 8.13 (d, J = 5.7 Hz, 2H), 8.23
(t, J = 8.7 Hz, 1H), 8.37 (t, J = 4.2 Hz, 2H), 9.04 (d, J = 8.4 Hz, 1H),
9.19 (d, J = 8.4 Hz, 1H); 13C NMR (75 MHz, DMSO-d6, ppm): d =
12.77, 44.65, 96.80, 107.63, 109.93, 114.96, 115.17, 115.53, 116.23,
119.60, 126.00, 126.45, 126.88, 129.25, 130.63, 131.43, 135.21,
139.26, 144.30, 144.51, 147.50, 151.82, 152.72, 153.56, 157.14,
158.94, 164.68; IR (KBr) n cmÀ1: 3449.35, 2962.89, 2923.40,
2854.32, 1705.49, 1619.25, 1586.25, 1513.06, 1222.80, 1126.65;
HR-MS: calcd 573.2866, found 573.2925.
2À
À
reaction of HSO3À/SO3 to SNB. With the increment of HSO3
/
SO32À, the dual emission band of SNB displayed an opposite
trend, wherein the emission intensity at 478 nm was gradually
enhanced and the emission intensity at 671 nm was slowly
reduced. SNB realized the ratiometric detection of HSO3À/SO3
2À
by plotting the changes in the fluorescence intensity ratio
2À
(I478/I671) as a function of the concentration of HSO3À/SO3
.
The large emission band gap (193 nm) also ensured the weak
interference between the dual emission bands. Therefore, SNB
2À
was used to monitor HSO3À/SO3 with a low detection limit
Testing solution preparation and cell culturing
(17 nM), wide pH range, high sensitivity and selectivity.
As a cationic organic fluorescent probe, SNB preferred to
attach onto mitochondria in living cells because of the electro-
static interaction between the 1,4-dimethylquinolin-1-ium
moiety and mitochondrial membrane.14 Incorporating its NIR
active emission, SNB was used to obtain the ratiometric biÀo-
Probe SNB (7.00 mg) was dissolved in DMSO and diluted to
10 mL for a stock solution. The testing solution was prepared by
transferring the stock solution (50.0 mL) into a volumetric flask
(10.0 mL) and diluting to 10.0 mL by DMSO and PBS (VDMSO
:
VPBS = 4 : 6, pH = 8.5). To confirm the selectivity and inter-
ference of SNB toward HSO3À/SO32À, various species including
Cys, HClO, t-BuOOH, t-BuOOꢀ, HOꢀ, H2O2, S2À, 1O2, Fe2+, ONOOÀ,
HSO3À, GSH, FÀ, NO2À, SO42À, Cu2+, S2O32À, SCNÀ, NO3À, Ca2+,
Mg2+ were prepared according to previous reports.16
imaging and detection of exogenous and endogenous HSO3
/
2À
SO3
with high photo-stability, good biocompatibility and
preferable mitochondria-targeted capacity.
Cells were cultivated with Dulbecco’s modified Eagle’s
medium (DMEM), which contained a supplement of 10% FBS
(Fetal Bovine Serum) in the carbon dioxide incubator with an
Experimental
Reagents and instruments
The experimental reagents and materials used in all experimental
procedures were purchased from commercial companies without
further purification. Distilled water was used in all experimental
steps. Silica gel (200–300 mesh) was used in the purification
process. Thin-layer chromatography (TLC) with silica gel F254
was purchased from Merck KGaA. An Agilent Q-TOF6510 mass
spectrometer was used to conduct MS analysis. A Hitachi U-4100
UV-vis spectrophotometer was applied to record the UV-Vis spec-
trum. The IR spectrum was recorded by a VERTEX 70 FT-IR
(Bruker Optics) spectrophotometer. Various pH solutions were
demarcated by a PHS-3C meter. A fluorescence spectrum was
recorded by a Hitachi F-7000 FL spectrophotometer (Xe lamp as
Scheme 2 Synthesis route of probe SNB.
6586 | J. Mater. Chem. B, 2019, 7, 6585--6591
This journal is ©The Royal Society of Chemistry 2019

View Article Online
Journal of Materials Chemistry B
Paper
atmosphere of 5% CO2 and 95% air at 37 1C Cells were placed 677.2359. Compared with probe SNB, the diagnostic hydrogens
in a 96-well plate in the concentration of 40 000 per mL for 24 h, were observed with clear variation; the H1 (d = 7.528, J = 9.0 Hz,
and then incubated with probe SNB.
1H), H2 (d = 9.060, J = 9.0 Hz, 1H), H3 (d = 4.491, 3H) and H4
(d = 9.206, J = 6.6 Hz, 1H) disappeared when 2.5 equiv. NaHSO3
was added (Fig. S8, ESI†). In addition, some new and representa-
tive hydrogens were observed, in which H5 (d = 5.599, J = 9.3 Hz,
1H), H6 (d = 5.480, J = 8.1 Hz, 1H), H7 (d = 4.482, J = 9.6 Hz, 1H) and
H8 (d = 3.061, 3H) were clearly generated.
Results and discussion
Design of probe SNB
In general, coumarin and its derivatives displayed promising proper-
ties involving a preferable fluorescence, good biocompatibility, facile
functionalization, and high photo and chemical stability. Moreover,
4-(4-(dimethylamino)styryl)-1-methylquinolin-1-ium derivatives show
NIR emission signals, water-solubility, and preferable biocompat-
ibility (i.e., low toxicity and membrane permeability).
Property analysis of probe SNB
2À
Probe SNB responded well toward HSO3À/SO3 with a clearly
ratiometric fluorescence emission that drastically enhanced the
blue fluorescence at 478 nm and clearly reduced near-infrared
emissive fluorescence at 671 nm (Fig. 1(a)). The results revealed
that probe SNB possessed good potency for the real-time and
The intramolecular positive charge endows the probe to
preferentially locate on mitochondria due to the electrostatic
interaction between the quinoline moiety and mitochondrial
membrane, which promote them for applications in bioimaging
and subcellular localization.14a Notably, the emission band of the
coumarin and the absorbance band of 1,4-dimethylquinolin-1-
ium iodide are overlapped well (75.2%) (Fig. S5, ESI†), which
ensures a high energy transfer efficiency between them (92%)
(Fig. S6, ESI†). Consequently, probe SNB was constructed from the
coumarin donor and the 4-(4-(dimethylamino)styryl)-1-methyl-
quinolin-1-ium acceptor by the piperazine linker based on the FRET
platform. Remarkably, SNB displayed an enhanced NIR emission
signal of the acceptor (671 nm) under the excitation of the donor
(420 nm) because of the FRET process from the donor to acceptor,
which also offered a large Stokes shift (253 nm). There is rarely
interference between the dual emission bands due to the large
fluorescence emission band gap (193 nm). More importantly, the
‘‘CQC’’ bond in the acceptor with a strong electron-deficient
ratiometric detection of HSO3À/SO32À. The fluorescence quan-
À
tum yields of this probe before and after the addition of HSO3
/
2À
SO3 were determined as 15% and 29%, respectively. To gain
more insight into the ratiometric detection platform of the
FRET mechanism, a chromaticity diagram was obtained from
the fluorescence emission spectra by the CIE-1931 RGB coordi-
nates (Fig. 1(b)). The donor, acceptor and probe SNB showed
bright blue (0.14, 0.23), red (0.62, 0.38) and purple red
(0.34, 0.26), respectively. The results indicated that the ratio-
2À
metric detection mechanism of the probe toward HSO3À/SO3
was based on the FRET platform. In comparison with the
acceptor, the reduced red chrominance of probe SNB was
derived from the energy surplus of the donor. However, the
reduced red chrominance did not impact the near-infrared emissive
fluorescence, and only represented the visual fluorescent colour.
As a significant evaluation index, a selectivity analysis of
2À
probe SNB toward HSO3À/SO3
was conducted. Compared
property is easily attacked by HSO3À/SO3 based on the nucleo-
2À
philic addition reaction. With the increment of HSO3À/SO32À, the
emission signal at 478 nm enhanced clearly and the emission signal
at 671 nm decreased slowly, which was attributed to the prevention
of FRET process by the addition reaction of HSO3À/SO32À. The
positive charge endows them to attach onto mitochondria and
realize ratiometric bioimaging for the subcellular organelles.
with the blank group, the fluorescence emission signal of these
groups had no obvious changes in the presence of CyÀs, HClO,
1
À
t-BuOOH, t-BuOOꢀ, HOꢀ, H2O2, H2S, O2, Fe2+, ONOO , NO2
,
GSH, Fe3+, Cu2+, Hg2+, Ca2+, FÀ, CO32À, S2O32À, Mn2+, BrÀ, Ag+
and CNÀ 10 mM (Fig. 2(a)). In contrast, the fluorescence ratio
2À
(I478/I671) realized 12-fold enhancements when HSO3À/SO3
(1.0 equiv.) was added. The emission signal displayed a clear
variation when Cys (10.0 equiv.) was added, but the ratio
(I478/I671) rarely changed. Compared with the addition of
HSO3À/SO32À, the fluctuation can be ignored completely, which
2À
Detection mechanism for HSO3À/SO3
Based on previous reports,17 a possible mechanism was
1
proposed (Scheme 3) and confirmed Àby MS and H NMR. MS
(Fig. S7, ESI†): calcd for [M+ + HSÀO3 + H+], 655.2520; found
655.2558. Calcd for [M+ + HSO3 + Na+], 677.2410, found
2À
Fig. 1 (a) Fluorescence spectrum of SNB toward 10 equiv. HSO3À/SO3
.
(b) The CIE chromaticity diagram of donor, acceptor and probe SNB.
(Excitation wavelength: 420 nm, speed: 1200 nm minÀ1, excitation and
emission slit widths: 5/5 nm).
2À
Scheme 3 Recognition mechanism of SNB toward HSO3À/SO3
.
This journal is ©The Royal Society of Chemistry 2019
J. Mater. Chem. B, 2019, 7, 6585--6591 | 6587

View Article Online
Paper
Journal of Materials Chemistry B
was added into the testing solution, the fluorescent intensity
ratio (I478/I671) gradually increased and nearly reacted comple-
tely at about 50 min (Fig. S10, ESI†). The response of probe SNB
2À
to HSO3À/SO3
depending on pH changes was explored
(Fig. S11, ESI†). Probe SNB existed stably in a wide pH range
from 5 to 10. I478/I671 clearly increased with increasing pH
and reached a maximum when pH = 8.5 in the presence of
HSO3À/SO32À (5.0 mM). Taking this into consideration, pH = 8.5
was employed as the testing pH, which was more suitable for
2À
the nucleophilic addition of HSO3À/SO3 to SNB. Probe SNB
displayed an enhanced fluorescence emission signal of the
acceptor at the NIR region and quenched the emission signal
of the donor, owing to the FRET process from the donor to
the acceptor. With increasing HSO3À/SO32À, the fluorescence
emission at 671 nm gradually decreased and the emission
signal at 478 nm was clearly enhanced, which was attributed
to the prevention of the FRET process between the donor and
2À
the acceptor by the addition reaction of HSO3À/SO3
(Fig. 3(a)). In visible light, the testing solution changed from
2À
a deep red to colorless when 3 equiv. of HSO3À/SO3 was
À
added. To determine the quantitative detection of HSO3
/
SO32À, the relationship between the fluorescence intensity ratio
(I478/I671) and the concentration of HSO3À/SO32À was plotted. A
good linearity was obtained from 0 to 7 mM with R2 = 0.9904,
K = 1.2703, and s = 0.0368 (Fig. 3(b)), which revealed that the
2À
quantitative detection of HSO3À/SO3 could be realized. The
detection limit was calculated as 17 nM by the method of
3s/slope,19 which was lower than that reported for many probes
(Table S1, ESI†). Moreover, the UV-vis absorbance spectra of the
2À
SNB response to HSO3À/SO3 displayed a gradual decrease in
Fig. 2 (a) Fluorescence responses of SNB (5 mM) toward various species
(blank, Cys, HClO, t-BuOOH, t-BuOOꢀ, HOꢀ, H2O2, H2S, 1O2, Fe2+
,
,
ONOOÀ, NO2À, GSH, Fe3+, Cu2+, Hg2+, Ca2+, FÀ, CO32À, S2O32À, Mn2+
BrÀ, Ag+, CNÀ, HSO3À. HSO3À: 5 mM; others: 10 mM). (b) Fluorescence
response of SNB to HSO3À/SO3
(5 mM) in the presence of various
2À
interference (blank, Cys, GSH, t-BuOOH, NO3À, Fe2+, Pb2+, H2S, Cu2+
,
K+, Co3+, Mg2+, Hg2+, Fe3+, Ca2+, Ag+, Mn2+, FÀ, BrÀ, CNÀ, S2O32À, OHÀ,
SCNÀ, ClÀ and SO42À; 10 mM). (Excitation wavelength: 420 nm, speed:
1200 nm minÀ1, excitation and emission slit widths: 5/5 nm, DMSO : PBS =
4 : 6, pH = 8.5).
further revealed the high selectivity of SNB to HSO3À/SO32À and
the advantages of the ratiometric probe.18 In the UV-vis spectra,
the absorbance changed slightly in the presence of the above
À
analytes, while it clearly decreased when 1 equiv. of HSO3
/
2À
SO3 was added (Fig. S9, ESI†). The results were consistent
with the changes observed in the fluorescence spectra. More-
over, the anti-interference ability of SNB was also explored in
the presence of various species, including Cys, GSH, t-BuOOH,
NO3À, Fe2+, Pb2+, H2S, Cu2+, K+, Co3+, Mg2+, Hg2+, Fe3+, Ca2+,
Ag+, Mn2+, FÀ, BrÀ, CNÀ, S2O32À, OHÀ, SCNÀ, ClÀ and SO42À. In
Fig. 2(b), SNB reacted well with HSO3À/SO3 and retained a
relatively stable fluorescence emission signal output regardless to HSO3À/SO3 from 0 to 1.6 equiv. (Excitation wavelength: 420 nm,
2À
Fig. 3 (a) The fluorescence emission spectrum of SNB (5 mM) in response
2À
speed: 1200 nm minÀ1, excitation and emission slit widths: 5/5 nm). (b) The
of the presence of an interference species. Therefore, SNB
2À
fluorescence ratio of I478/I671 depended on the variation of HSO3À/SO3
possessed a preferable anti-interference ability in the detection
from 0 to 1.4 equiv. (c) The UV-vis absorbance spectrum of SNB (10 mM) in
of HSO3À/SO3 and provided a precondition for bioimaging.
2À
response to HSO3À/SO3 from 0 to 1.6 equiv. (Speed: 600 nm minÀ1, slit:
2À
The changes of I478/I671 were mapped as a function of the
5 nm). (d) The linearity of the absorbance intensity depended on the variation
2À
response time to conduct data analysis. When HSO3À/SO3
of HSO3À/SO3 from 0 to 1.4 equiv. (DMSO:PBS = 4 :6, pH = 8.5).
2À
6588 | J. Mater. Chem. B, 2019, 7, 6585--6591
This journal is ©The Royal Society of Chemistry 2019

View Article Online
Journal of Materials Chemistry B
Paper
the absorbance intensity at 490 nm and a negligible change
in the absorbance intensity at 410 nm (Fig. 3(c)), which
were consistent with the fluorescence emission changes. The
decreased absorbance at 490 nm was attributed to the nucleo-
2À
philic reaction of HSO3À/SO3 to SNB. More importantly, the
À
function between the absorbance and concentration of HSO3
/
SO32À was plotted with an excellent relationship from 0–7.0 mM
with R2 = 0.9968, K = À0.0351, and s = 0.0030 (Fig. 3(d)).
Therefore, the fluorescence emission and UV-vis absorbance
changes of SNB were dependent on the concentration of
HSO3À/SO32À. This demonstrated that probe SNB should be a
promising candidate for the quantitative detection of the
2À
Fig. 4 Imaging of exogenous HSO3À/SO3 in living HeLa cells by probe
SNB (1 mM). (a) Fluorescence, bright and merge field images of HeLa cells
treated by various concentration of NaHSO3 (0, 20, 100, 500 and 750 mM).
(b) The changes of the blue/red fluorescence intensity ratio with increasing
NaHSO3 (0, 20, 100, 500 and 750 mM). (Blue fluorescence: 450–610 nm,
red fluorescence: 610–700 nm, excitation wavelength: 405 nm, *p o 0.05,
**p o 0.01, n = 3).
2À
HSO3À/SO3 with a low detection limit, high sensitivity and
excellent NIR emission signal.
Bioimaging in living cells
SRB analysis was conducted in living HeLa cells, and showed
that the cell viability had no obvious fluctuation with increasing
SNB even up to 10 mM (Fig. S12, ESI†). Therefore, SNB possessed the liver cells (including HepG2 cells). Subsequently, probe SNB
2À
negligible toxicity and good biocompatibility in bioimaging. was applied to detect endogenous HSO3À/SO3
in living
Although SNB was continuously excited in living HeLa cells for HepG2 cells by confocal bio-imaging. HepG2 cells exhibited
0, 30, 60, 90, 150, 240 and 300 s by 405 nm light, the emission deep red fluorescence emission and weak blue fluorescence
signal output remained stable (Fig. S13, ESI†). Probe SNB dis- emission when hatched by SNB (1 mM). HepG2 cells were
played excellent photo and chemical stability in living cells, pre-treated with SNB (1 mM) for 1 h, and further treated by
indicating the feasibility of conducting bioimaging via confocal GSH (500 mM) and Na2S2O3 (250 mM) for another 0.5 h before
microscopy. The photostability experiment of probe SNB was the confocal imaging. Clear ratiometric fluorescence changes
conducted in daylight conditions, and the result is shown in were observed in the cells. No obvious fluorescence change was
Fig. S14 (ESI†). After exposure for 90 min in daylight, the observed compared with the control group, if HepG2 cells were
fluorescence ratio of I478/I671 remained unchanged, which further just hatched with SNB (1 mM) and GSH (500 mM). In addition,
revealed the high photostability of SNB. In addition, the thermo- the HepG2 cells were pre-incubated with TNBS (10 mM) for
stability of SNB was determined by temperature-fluorescence 30 min before 1 h pre-treatment with SNB (1 mM), and followed
cycling experiments (Fig. S15, ESI†). The temperatures 37 1C by GSH (500 mM) and Na2S2O3 (250 mM) for 0.5 h. There were
and 60 1C were used to conduct the temperature-fluorescence still no clear fluorescence changes observed compared to the
cycle. The fluorescence spectra of probe SNB at 37 1C was initially
recorded, then the probe was heated to 60 1C and cooled to 37 1C,
upon which a fluorescence measurement was conducted again.
After five cycles, the fluorescence ratio of probe SNB was still
stable, which further revealed the high thermal cycling ability.
2À
The response of probe SNB to exogenous HSO3À/SO3 was
studied in living HeLa cells. The cells were pre-incubated with
SNB (1 mM) for 1 h, followed by treatment with various
concentrations of NaHSO3 (0, 20, 100, 500 and 750 mM) for
another 1 h, and then confocal imaging was conducted.
As shown in Fig. 4(a), the red fluorescence emission signal
(610–700 nm) gradually receded and blue fluorescence emission
signal (450–610 nm) was clearly enhanced with increasing exogen-
ous HSO3À/SO32À. The changes were attributed to the destruction of
the FRET process from the donor to the acceptor, which was caused
Fig. 5 (a) First column: HepG2 cells were treated with SNB (1 mM) for 1 h.
2À
by the additional reaction of exogenous HSO3À/SO3 to SNB in
Second column: HepG2 cells were pre-treated with SNB (1 mM) for 1 h
living HeLa cells. In Fig. 4(b), the blue/red fluorescence ratio clearly
before incubation with GSH (500 mM) and Na2S2O3 (250 mM) for another
0.5 h. Third column: HepG2 cells were pre-cultured with SNB (1 mM) for
1 h, and then cultured with GSH (500 mM) for 0.5 h. Fourth column: HepG2
cells were treated for 1 h by SNB (1 mM) after 30 min pre-treatment with
TNBS (10 mM), followed by GSH (500 mM) and Na2S2O3 (250 mM) for 0.5 h.
(b) The fluorescence intensity ratio of blue/red of column 1, 2, 3 and 4 in
(a). Excitation wavelength: 405 nm; blue channel: 450–610 nm, red
channel: 610–700 nm (***p o 0.001, n = 3).
2À
increased with increasing HSO3À/SO3 and realized 10-fold
enhancements when 750 mM NaHSO3 was added, which was
consistent with the imaging in Fig. 3(a). The results demonstrated
the excellent properties of probe SNB in living cells.
Endogenous HSO3À/SO32À can be generated in an organism
under the catalysis of the TST enzyme,20 which is widespread in
This journal is ©The Royal Society of Chemistry 2019
J. Mater. Chem. B, 2019, 7, 6585--6591 | 6589

View Article Online
Paper
Journal of Materials Chemistry B
SNB (3 mM) for 1 h and followed by treatment with Mito Tracker
Deep Red (1 mM) for 20 min, then confocal imaging was conducted.
The fluorescence emission signal distribution of SNB and Mito
Tracker Deep Red are clearly overlapped with a high localization
coefficient of 0.91 (Fig. 7). Hence, probe SNB successfully realized
ratiometric images for exogenous and endogenous HSO3À/SO32À in
living cells, and preferred to adhere to mitochondria with a high
localization coefficient of 0.91.
Conclusion
Fig. 6 (a) First column: L-O2 cells were treated with SNB (1 mM) for 1 h.
Second column: L-O2 cells were pre-treated with SNB (1 mM) for 1 h
before incubation with GSH (500 mM) and Na2S2O3 (500 mM) for another
0.5 h. Third column: L-O2 cells were pre-cultured with SNB (1 mM) for 1 h,
and then cultured with NaHSO3 (20 mM) for 0.5 h. Fourth column: L-O2
cells were pre-cultured with SNB (1 mM) for 1 h, and then cultured with
NaHSO3 (100 mM) for 0.5 h. (b) The relative ratio (blue/red) of fluorescence
intensity of column 1, 2, 3 and 4 in (a). Excitation wavelength: 405 nm; the
blue channel: 450–610 nm; the red channel: 610–700 nm. (**p o 0.01,
n = 3).
In summary, a unique NIR probe SNB for the ratiometric
2À
monitoring of HSO3À/SO3 was developed based on the FRET
platform. The probe showed an active NIR emission signal,
extended Stokes shift and a broad emission band gap. SNB
2À
realized the ratiometric detection of HSO3À/SO3 with a low
detection limit, high sensitivity and selectivity. Importantly,
probe SNB realized the ratiometric imaging for exogenous and
2À
endogenous HSO3À/SO3 in living cancer cells due to their
good biocompatibility, preferable fluorescence, and good chemical
and photo stability. Remarkably, SNB possessed a prominent
mitochondria-targeted capacity in living cells. Therefore, SNB could
be a promising candidate for the probable diagnosis of cancer cells.
control group (Fig. 5). The results demonstrated that probe SNB
could be used for the ratiometric imaging of endogenous
2À
HSO3À/SO3 in living HepG2 cells.
Probe SNB hatched L-O2 cells exhibited deep red and weak
blue fluorescence emissions. After the SNB-loaded L-O2 cells
were treated by exogenous NaHSO3 (20 and 100 mM) for 0.5 h,
dramatic fluorescence changes were observed of reduced red
and enhanced blue fluorescence emissions. In addition, the
ratio of the blue/red fluorescence intensity realized a 6-fold
enhancement when 100 mM NaHSO3 was added, which is
clearly more than that reported by many research studies.
While the L-O2 cells were pre-treated with SNB (1 mM) for 1 h
before incubation with GSH (500 mM) and Na2S2O3 (500 mM) for
another 0.5 h, no clear fluorescence changes were observed
(Fig. 6). Therefore, probe SNB could realize the real-time
Conflicts of interest
There are no conflicts to declare.
Acknowledgements
This study was supported by the National Key Research and
Development Program of China (2017YFA0104604), the National
Natural Science Foundation of China (no. 31871407, 31741083,
31870831), and the Natural Science Foundation of Shandong
Province (ZR2018MB042).
2À
detection of endogenous HSO3À/SO3 in living cancer cells
and differentiate normal L-O2 cells with cancer cells.
Notes and references
The localization study was conducted in living HeLa cells. As
a cationic mitochondrial cyanine dye, Mito Tracker Deep Red
was often used to conduct mitochondrial localization. In general,
cationic organic fluorophores prefer to locate on the mitochondria
because of the electrostatic interaction between the cationic moiety
and mitochondria membrane.21 HeLa cells were incubated with
´
1 (a) C. Ruiz-Capillas and F. Jimenez-Colmenero, Food Chem.,
2009, 112, 487–493; (b) S. Samanta, S. Halder, P. Dey,
U. Manna, A. Ramesh and G. Das, Analyst, 2017, 143, 250–257.
2 (a) N. Sang, Y. Yun, H. Li, L. Hou, M. Han and G. Li, Toxicol.
Sci., 2010, 114, 226–236; (b) Z. Meng, Z. Yang, J. Li and
Q. Zhang, Chemosphere, 2012, 89, 579–584; (c) J. Li, R. Li and
Z. Meng, Eur. J. Pharmacol., 2010, 645, 143–150.
3 (a) G. Chen, W. Zhou, C. Zhao, Y. Liu, T. Chen, Y. Li and
B. Tang, Anal. Chem., 2018, 90, 12442–12448; (b) Y. Zhao,
Y. Ma and W. Lin, Sens. Actuators, B, 2018, 268, 157–163;
(c) S. Wang, S. Ma, J. Zhang, M. She, P. Liu, S. Zhang and
J. Li, Sens. Actuators, B, 2018, 261, 418–424; (d) M. Li,
W. Feng, H. Zhang and G. Feng, Sens. Actuators, B, 2017,
243, 51–58.
4 (a) Y. Liu, K. Li, M. Y. Wu, Y. H. Liu, Y. M. Xie and X. Q. Yu,
Chem. Commun., 2015, 51, 10236–10239; (b) J. Xu, J. Pan,
X. Jiang, C. Qin, L. Zeng, H. Zhang and J. F. Zhang, Biosens.
Fig. 7 (a) Fluorescence image of HeLa cells with SNB (3 mM) from the blue
channel (450–610 nm). (b) Fluorescence image of HeLa cells with Mito
Tracker Deep Red (1 mM) from the red channel (640–710 nm). (c) The
merged image of (a) and (b). (d) Co-localization images (co-localization
coefficient 0.91). (Excitation wavelength: 405 nm).
6590 | J. Mater. Chem. B, 2019, 7, 6585--6591
This journal is ©The Royal Society of Chemistry 2019

View Article Online
Journal of Materials Chemistry B
Paper
Bioelectron., 2016, 77, 725–732; (c) X. Yang, Y. Zhou,
Y. Geng, W. Chen and X. Song, Sens. Actuators, B, 2018, 276,
X. Zhang, S. Yang, Y. Chen, J. Guo, X. Li, Z. Qing and
247–253.
R. Yang, Chem. Commun., 2016, 52, 10289–10292; (d) H. Li, 13 (a) H. Li, J. Fan, S. Long, J. Du, J. Wang and X. Peng, Sens.
X. Zhou, J. Fan, S. Long, J. Du, J. Wang and X. Peng, Sens.
Actuators, B, 2018, 254, 709–718.
5 (a) X. Liu, Y. Su, H. Tian, L. Yang, H. Zhang, X. Song and
Actuators, B, 2018, 273, 899–905; (b) J. Li, Y. Gao, H. Guo,
X. Li, H. Tang, J. Li and Y. Guo, Dyes Pigm., 2019, 163,
285–290.
J. W. Foley, Anal. Chem., 2017, 89, 7038–7045; (b) C. Yin, 14 (a) G. Yin, Y. Gan, T. Yu, T. Niu, P. Yin, H. Chen, Y. Zhang,
X. Li, Y. Yue, J. Chao, Y. Zhang and F. Huo, Sens. Actuators,
B, 2017, 246, 615–622.
6 (a) H. Zhu, J. Fan, J. Wang, H. Mu and X. Peng, J. Am. Chem.
H. Li and S. Yao, Talanta, 2019, 191, 428–434; (b) F. Tian,
Y. Jia, Y. Zhang, W. Song, G. Zhao, Z. Qu, C. Li, Y. Chen and
P. Li, Biosens. Bioelectron., 2016, 86, 68–74.
Soc., 2014, 136, 12820–12823; (b) L. He, X. Yang, K. Xu and 15 (a) X. Feng, T. Zhang, J. T. Liu, J. Y. Miao and B. X. Zhao,
W. Lin, Chem. Commun., 2017, 53, 4080–4083.
7 (a) A. C. Sedgwick, L. Wu, H. H. Han, S. D. Bull, X. P. He,
T. D. James, J. L. Sessler, B. Z. Tang, H. Tian and J. Yoon,
Chem. Commun., 2016, 52, 3131–3134; (b) Y. J. Lu, X. L. Guo,
M. H. Xu, W. W. Chen, W. L. Wong, K. Zhang and
C. F. Chow, Dyes Pigm., 2017, 143, 331–341.
Chem. Soc. Rev., 2018, 47, 8842–8880; (b) Y. Liu, J. Nie, J. Niu, 16 (a) H. Song, Y. Zhou, C. Xu, X. Wang, J. Zhang, Y. Wang,
W. Wang and W. Lin, J. Mater. Chem. B, 2018, 6, 1973–1983.
8 (a) J. Wang, S. Xia, J. Bi, Y. Zhang, M. Fang, R. L. Luck,
Y. Zeng, T. H. Chen, H. M. Lee and H. Liu, J. Mater. Chem. B,
2019, 7, 198–209; (b) N. R. Chereddy, M. V. N. Raju, B. M.
Reddy, V. R. Krishnaswamy, P. S. Korrapati, B. J. M. Reddy
and V. J. Rao, Sens. Actuators, B, 2016, 237, 605–612.
X. Liu, M. Guo and X. Peng, Dyes Pigm., 2019, 162, 160–167;
(b) G. J. Mao, Z. Z. Liang, J. Bi, H. Zhang, H. M. Meng, L. Su,
Y. J. Gong, S. Feng and G. Zhang, Anal. Chim. Acta, 2019,
1048, 143–153; (c) Z. Mao, M. Ye, W. Hu, X. Ye, Y. Wang,
H. Zhang, C. Li and Z. Liu, Chem. Sci., 2018, 9, 6035–6040.
17 (a) Z. Ye, C. Duan, R. Sheng, J. Xu, H. Wang and L. Zeng,
Talanta, 2018, 176, 389–396; (b) H. Yu, Y. Wu, Y. Hu, X. Gao,
Q. Liang, J. Xu and S. Shao, Talanta, 2017, 165, 625–631.
9 (a) G. Song, A. Liu, H. Jiang, R. Ji, J. Dong and Y. Ge, Anal.
Chim. Acta, 2019, 1053, 148–154; (b) Q. Zhao, F. Huo,
J. Kang, Y. Zhang and C. Yin, J. Mater. Chem. B, 2018, 6, 18 (a) L. Tang, P. He, X. Yan, J. Sun, K. Zhong, S. Hou and
4903–4908.
Y. Bian, Sens. Actuators, B, 2017, 247, 421–427; (b) J. S. Lan,
R. F. Zeng, Y. Ding, Y. Zhang, T. Zhang and T. Wu, Sens.
Actuators, B, 2018, 268, 328–337.
10 (a) Y. H. Yan, H. L. Ma, J. Y. Miao, B. X. Zhao and Z. M. Lin,
Anal. Chim. Acta, 2019, 1064, 87–93; (b) X. Zhu, L. Zhu,
H. W. Liu, X. Hu, R. Z. Peng, J. Zhang, X. B. Zhang and 19 (a) M. Zhao, D. Liu, L. Zhou, B. Wu, X. Tian, Q. Zhang,
W. Tan, Anal. Chim. Acta, 2016, 937, 136–142; (c) W. Liu,
D. Zhang, B. Ni, J. Li, H. Weng and Y. Ye, Sens. Actuators, B,
2019, 284, 330–336.
H. Zhou, J. Yang, J. Wu and Y. Tian, Sens. Actuators, B, 2018,
255, 1228–1237; (b) Z. Wang, Y. Zhang, J. Song, Y. Yang,
X. Xu, M. Li, H. Xu and S. Wang, Anal. Chim. Acta, 2019,
1051, 169–178.
11 (a) X. Tian, Z. Li, Y. Sun, P. Wang and H. Ma, Anal. Chem.,
2018, 90, 13759–13766; (b) B. Zhu, L. Wu, M. Zhang, 20 (a) Q. Liu, A. Li, X. Li, B. Li, Y. Zhang, J. Li and Y. Guo, Sens.
Y. Wang, C. Liu, Z. Wang, Q. Duan and P. Jia, Biosens.
Bioelectron., 2018, 107, 218–223.
Actuators, B, 2019, 283, 820–830; (b) K. Liu, Y. Chen, H. Sun,
S. Wang and F. Kong, J. Mater. Chem. B, 2018, 6, 7060–7065.
12 (a) F. Yu, P. Li, B. Wang and K. Han, J. Am. Chem. Soc., 2013, 21 D. P. Li, Z. Y. Wang, H. Su, J. Y. Miao and B. X. Zhao, Chem.
135, 7674–7680; (b) L. He, X. Liu, Y. Zhang, L. Yang, Q. Fang,
Commun., 2017, 53, 577–580.
This journal is ©The Royal Society of Chemistry 2019
J. Mater. Chem. B, 2019, 7, 6585--6591 | 6591