A 3,7-Dihydroxyphenoxazine-based Fluorescent Probe for Selective Detection of Intracellular Hydrogen Peroxide

Article abstract of DOI:10.1002/asia.201501304

A novel N-borylbenzyloxycarbonyl-3,7-dihydroxyphenoxazine fluorescent probe (NBCD) for detecting H₂O₂ in living cells is described. The probe could achieve high selectivity for detecting H₂O₂ over other biological reactive oxygen species (ROS). In addition, upon addition of H₂O₂, NBCD exhibited color change from colorless to pink, which makes it a "naked-eye" probe for H₂O₂ detection. NBCD could not only be used to detect enzymatically generated H₂O₂ but also to detect H₂O₂ in living systems by using fluorescence spectroscopy, with a detection limit of 2 μm. Importantly, NBCD enabled the visualization of epidermal growth factor (EGF)-stimulated H₂O₂ generation inside the cells.

Full text of DOI:10.1002/asia.201501304

DOI: 10.1002/asia.201501304
Communication
Fluorescent Probes
A 3,7-Dihydroxyphenoxazine-based Fluorescent Probe for
Selective Detection of Intracellular Hydrogen Peroxide
Zhiqiang Han,[a, b] Xiao Liang,[a, b] Xuejiao Ren,[a, b] Luqing Shang,*[a, b] and Zheng Yin*[a, b]
Several types of probes for the detection of H2O2 have been
Abstract: A novel N-borylbenzyloxycarbonyl-3,7-dihydrox-
yphenoxazine fluorescent probe (NBCD) for detecting
reported, such as fluorescein,[6] coumarin,[7] and luciferin[8]. As
one of the important fluorescent reporters, resorufin, a highly
H2O2 in living cells is described. The probe could achieve
fluorescent molecule at pH 8 (fF =0.75),[9] has excitation/emis-
high selectivity for detecting H2O2 over other biological re-
sion maxima of ~570/585 nm, and there is much less interfer-
ence from autofluorescence in most biological samples. N-Pro-
active oxygen species (ROS). In addition, upon addition of
H2O2, NBCD exhibited color change from colorless to pink,
tected 3,7-dihydroxyphenoxazine[10] and O-alkyl resorufin deriv-
which makes it a “naked-eye” probe for H2O2 detection.
atives[1c,11] have been described as chemical probes for en-
NBCD could not only be used to detect enzymatically gen-
zymes in the literature. As a precursor of a fluorescent probe,
erated H2O2 but also to detect H2O2 in living systems by
the former has an advantage of lower fluorescence back-
using fluorescence spectroscopy, with a detection limit of
ground, which is less than that of the latter because of an in-
2 mm. Importantly, NBCD enabled the visualization of epi-
dermal growth factor (EGF)-stimulated H2O2 generation
terrupted conjugated system. A combination of N-acetyl-3,7-di-
hydroxyphenoxazine (Amplex Red) and horseradish peroxidase
inside the cells.
(HRP) was used in tests of enzymes that produce H2O2.[12] How-
ever, Amplex Red could only detect H2O2 indirectly because it
could not be oxidized by H2O2 itself.[10] Hitomi and co-workers
Hydrogen peroxide (H2O2), an important reactive oxygen spe-
cies (ROS), has been correlated with various functions in the
regulation of many physiological processes.[1] For a long time,
H2O2 has been considered as a harmful metabolic product and
an important molecule involved in pathophysiological path-
ways associated with various diseases.[2] Recently, H2O2 has
been found to act as a signaling molecule in many biological
processes such as cell proliferation, migration, and differentia-
tion.[3] Moderate levels of H2O2 play a crucial role in physiology,
aging, and various diseases in living organisms.[4] Overproduc-
tion of H2O2 at the cellular level has been linked to cancer,
neurodegenerative diseases, DNA damage, mutation, and ge-
netic instability.[5] Thus, the significant physiological impor-
tance of H2O2 has led to numerous studies aimed at develop-
ing high-efficiency fluorescent probes for detecting H2O2 in
living systems.
presented a metal-based fluorescent probe for H2O2, named
MBFh1, containing an iron complex and a non-fluorescent 3,7-
dihydroxyphenoxazine derivative.[10] Unfortunately, MBFh1 is
decomposed readily under cell culture conditions.[1c] The appli-
cation of these two resorufin-based probes is limited to only
cell-free systems, even though the detection of H2O2 in a cellu-
lar system is of great importance. To our knowledge, an N-pro-
tected 3,7-dihydroxyphenoxazine-derived chemical probe has
not yet been described for the selective detection of H2O2 in
living cells. Therefore, we aimed to design a novel chemical
probe with good stability and high selectivity for the detection
of H2O2 in living cells, taking advantage of the attributes of
unique fluorescent properties of resorufin. Herein, we report
a novel N-protected 3,7-dihydroxyphenoxazine probe (NBCD)
comprising a 3,7-dihydroxyphenoxazine as the reporter and
a borate-based carbamate leaving group[8a,13] as the H2O2 re-
sponse site. This probe features a good stability and high se-
lectivity, and detection of H2O2 was achieved not only in a cell-
free system but also in living cells.
[a] Dr. Z. Han, X. Liang, X. Ren, Prof. L. Shang, Prof. Z. Yin
College of Pharmacy & State Key Laboratory of Elemento-Organic Chemis-
The synthesis of the designed probe NBCD was initiated
from resazurin sodium salt (Scheme 1a and Scheme S1, Sup-
porting Information). Reduction of resazurin sodium salt with
zinc powder and glacial acetic acid afforded 3,7-dihydroxyphe-
noxazine (1). It reacted with acetic anhydride and N,N-dimeth-
yl-4-aminopyridine (DMAP) to generate 3,7-diacetoxyphenoxa-
zine (2),[10] and the subsequent treatment with triphosgene
yielded N-carbonyl-3,7-diacetoxyphenoxazine (3). Further reac-
tion of 3 with borylbenzyl alcohol (4) followed by hydrolysis
led to the desired N-borylbenzyloxycarbonyl-3,7-dihydroxyphe-
noxazine (NBCD). A possible mechanism of resorufin genera-
try
Nankai University
Tianjin 300071 (China)
E-mail: shanglq@nankai.edu.cn
zheng_yin@nankai.edu.cn
[b] Dr. Z. Han, X. Liang, X. Ren, Prof. L. Shang, Prof. Z. Yin
Collaborative Innovation Center of Chemical Science and Engineering (Tian-
jin)
Nankai University
Tianjin 300071 (China)
E-mail: zheng_yin@nankai.edu.cn
Supporting information for this article can be found under http://
dx.doi.org/10.1002/asia.201501304.
Chem. Asian J. 2016, 11, 818 – 822
818
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
Scheme 1. a) Synthesis and activation of NBCD. b) Possible mechanism of resorufin generation by the reaction of NBCD with H2O2.
tion by the reaction of NBCD with H2O2 is proposed in
Scheme 1b.
tensity at around 590 nm, which is visible with a clear color
change from colorless to pink (Figure 1b, inset, and Figure S3).
Moreover, there was a good linearity between the fluorescence
intensity (590 nm) and incubation time in the range of 0 to
60 min (Figure 1b inset). These results demonstrated that
NBCD could fluorescently detect H2O2 with excellent sensitivity.
We also investigated the concentration-dependent fluores-
cence response of NBCD at 30 min after the addition of H2O2
in range of 0–200 mm (Figure S4). The absorption and fluores-
cence spectra of the reaction mixture were identical to those
observed with resorufin. The generation of resorufin was also
confirmed chemically, and other byproducts were also found
by NMR spectroscopy and mass spectrometry (Figure S5A and
S6A, B). Moreover, the conversion efficiency of NBCD was
quantified (Figure S5B).
NBCD was expected to be able to display red fluorescence
in the presence of H2O2 because of the release of resorufin, al-
lowing for its use as a probe for detecting H2O2. Experimental-
ly, upon treatment with H2O2, NBCD displayed a remarkable
red fluorescence. Subsequently, the UV/Vis spectroscopic prop-
erties of NBCD were assessed under in vitro conditions (20 mm
HEPES buffer containing 0.5% DMSO, pH 7.3). Upon addition
of H2O2 (20 equiv) to a solution of NBCD, a gradual increase in
the absorption peak in the UV region (450–610 nm) over
45 min was observed, which suggested the generation of
a new chemical entity (Figure 1a). Moreover, there was a good
linearity between the absorption intensity and time in the
range of 0 to 45 min (Figure 1a, inset). NBCD exhibited no ab-
sorption in the UV region (450–610 nm) in the absence of
H2O2, and a concentration-dependent absorption response of
NBCD was observed at 30 min after the addition of H2O2 (Fig-
ure S1, Supporting Information). The absorption intensity of
NBCD increased significantly as a function of H2O2 concentra-
tion. Under pseudo-first-order conditions (5 mm NBCD, 5 mm
H2O2), the observed rate constant for H2O2 is kobs =2.266”
10À3 sÀ1 (Figure S2, Supporting Information).
In addition, we investigated the chemoselectivity of NBCD
toward H2O2 under the above-mentioned analytical conditions
and compared its relative reactivity of probe towards various
ROS. Selectivity data were collected at several time points over
60 min. As expected, only H2O2 significantly increased the fluo-
rescence intensity (590 nm). Fluorescence emission changes
were hardly observed in the presence of other ROS, including
C
hydroxyl radical ( OH), tert-butylhydroperoxide (TBHP), tert-
2À
À
C
Next, we investigated the fluorescence emission property of
NBCD (5 mm) in 20 mm HEPES buffer (pH 7.3) that is close to
physiological conditions. When excited at 530 nm, NBCD ex-
hibited no fluorescence emission at around 590 nm in the ab-
sence of H2O2. The subsequent addition of H2O2 to the solution
of NBCD led to a gradual increase in bright red fluorescence in-
butoxy radical ( OtBu), superoxide (O ), hypochlorite (ClO ),
and nitric oxide (NO) (Figure 2). Thus, these experimental re-
sults demonstrated that NBCD possessed excellent chemose-
lectivity towards H2O2 over other analytes.
As H2O2 is an important co-product of many oxidases, the
activity of the responsible enzymes or the concentration of
Chem. Asian J. 2016, 11, 818 – 822
www.chemasianj.org
819
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
lution containing NBCD quickly induced a progressive increase
of the fluorescence peak centered at around 590 nm because
of the generation of H2O2. However, addition of d-glucose to
a mixture solution of glucose oxidase containing NBCD in the
presence of catalase caused almost no fluorescence emission
at around 590 nm because H2O2 was consumed by catalase.
These results proved the ability of NBCD to assess enzymatical-
ly generated H2O2 (Figure 3a). We also investigated the fluores-
Figure 1. (a) Time-dependent absorption spectra during the reaction of
NBCD (10 mm) with 20 equiv. of H2O2 over 45 min. Inset: Linear relationship
between absorbance and time. (b) Time-dependent fluorescence intensity
changes of NBCD (5 mm) incubated with H2O2 (5 mm) over 60 min. Inset:
Linear relationship between fluorescence intensity and time; photographs of
the cuvettes containing NBCD before (left) and after (right) addition of H2O2.
Experiments were conducted at 378C in 20 mm HEPES buffer containing
0.5% DMSO (pH 7.3).
Figure 3. (a) Fluorescence detection of enzymatically generated H2O2 by
NBCD. Red trace: 10 mm NBCD, 50 mm d-glucose, 8 mgmLÀ1 glucose oxidase,
lex =530 nm. Black trace: as in panel in the presence of catalase 80 mgmLÀ1
(b) Fluorescence responses of NBCD (10 mm) towards different concentra-
.
tions of d-glucose at 30 min. Inset: Linear relationship between fluorescence
intensity and concentration of d-Glucose; conditions: 8 mgmLÀ1 glucose ox-
idase, lex =530 nm at 378C in 20 mm HEPES buffer containing 0.5% DMSO
(pH 7.3).
cence responses of NBCD towards different amounts of d-Glu-
cose (0–50 mm) after the reaction for 30 min. The addition of d-
Glucose to the solution of NBCD and glucose oxidase resulted
in a gradual increase of the fluorescence peak at around
590 nm with a good linearity between the fluorescence intensi-
ty and the concentration of d-glucose (0–50 mm) (Figure 3b
and Figure S7).
Figure 2. Fluorescence responses of NBCD (5 mm) towards ROS in 20 mm
HEPES, containing 0.5% DMSO, pH7.3, at 378C. Bars represent the relative
responses at 0, 10, 20, 30, 40, 50, and 60 min after addition of each oxidant.
Data were acquired at 10 mm H2O2, excess NO, and at a concentration of
500 mm for all other oxidants, with lexc =530 nm and lem =590 nm.
The dependence of the fluorescence intensity on pH was in-
vestigated by the reaction of NBCD with H2O2 at different pH
conditions in a range of pH from 4 to 9 (Figure S8). No fluores-
cence emission was observed in the range of pH 4–6 since de-
protection is impeded by the protonation of phenol, while the
fluorescence intensity significantly increased at pH 6–9 and
these substrates could be evaluated by monitoring H2O2.
Therefore, we investigated the ability of NBCD for detection of
enzymatically generated H2O2 to extend its application.
We used NBCD to detect H2O2 generated by glucose oxidase
and d-glucose. Addition of d-glucose to a glucose oxidase so-
Chem. Asian J. 2016, 11, 818 – 822
www.chemasianj.org
820
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
reached the maximum near pH 8. In addition, NBCD had excel-
lent stability in different pH buffers and cell culture media (Fig-
ure S9A, B).
Finally, we evaluated the ability of NBCD for application in
fluorescence imaging of intracellular H2O2 generated after stim-
ulation by the epidermal growth factor (EGF). In the experi-
ments, the detection of EGF-stimulated H2O2 inside A431 cells
was investigated according to the literature.[1c] Red fluorescent
signals were observed inside the cells once the cells were
stimulated with EGF, which proved the good sensitivity of
NBCD for EGF-stimulated H2O2 inside living cells (Figure S13).
In conclusion, we have described the design, synthesis,
properties and biological applications of a novel N-borylbenzy-
loxycarbonyl-3,7-dihydroxyphenoxazine fluorescent probe for
the detection of H2O2 in both in vitro systems and in living sys-
tems. NBCD displayed high H2O2 selectivity towards various
ROS, owing to the adoption of a boronate moiety. Moreover,
NBCD exhibited an apparent color change from colorless to
pink upon addition of H2O2, and thus could be used as
a “naked-eye” probe for H2O2. NBCD could be used to detect
H2O2 in living cells by using fluorescence spectroscopy with
a detection limit of 2 mm. This probe is stable under cell culture
conditions and could also detect enzymatically generated
H2O2. Most importantly, NBCD also enabled the visualization of
H2O2 generated by EGF stimulation inside the cells. The excel-
lent chemoselectivity for H2O2, good stability in the biological
pH range, and low cytotoxicity assure that this novel N-pro-
tected 3,7-dihydroxyphenoxazine probe could well serve as an
effective tool to investigate the detailed biological function of
H2O2 in living systems.
Based on the attractive properties of NBCD including the ex-
cellent ON/OFF ratio, high selectivity, and satisfactory stability,
NBCD was further evaluated for its application in fluorescence
imaging of H2O2 inside living systems. The experiment was ini-
tiated with the cytotoxicity study of NBCD in cells. The cell via-
bility was investigated by using 3-(4,5-dimethyl-2-thiazolyl)-2,5-
diphenyl-2-H-tetrazolium bromide (MTT assay) after incubation
of cells with NBCD at various concentrations for 24 h. Little cy-
totoxicity was observed at the maximum tested concentration
of 100 mm (Figure S10A). Then, H2O2 imaging experiments
were investigated using a previously reported method.[1c] HeLa
cells were incubated with 5 mm NBCD for 15 min at 378C, fol-
lowed by washing for four times to remove NBCD from the so-
lution outside of the cells. The cells were then incubated with
H2O2 at various concentrations for 40 min. The fluorescence
imaging showed clearly red fluorescence inside the cells (Fig-
ure 4b), which proved the good sensitivity of NBCD for H2O2
Experimental Section
Additional preparation details and fluorescent confocal microscopy
images, UV/Vis spectra, fluorescence spectra, details of the synthe-
1
sis, H NMR, 13C NMR, and MS spectra of compounds are included
in Supporting Information.
Acknowledgements
This work was supported by the National Basic Research Pro-
gram of China (973 program, Grant No. 2013CB911104), the Na-
tional Natural Science Foundation of China (Grant No.
21202087, 31200586), the Tianjin Science and Technology Pro-
gram (Grant No. 13JCYBJC24300, 13JCQNJC13100), the “111”
Project of the Ministry of Education of China (Project No.
B06005).
Figure 4. Confocal fluorescence and bright field images of live HeLa cells
after incubation with 5 mm NBCD for 15 min, followed by washing and fur-
ther incubation without H2O2 (a), with 500 mm H2O2 (b), with 500 mm H2O2 to-
gether with catalase (c) for 40 min. Scale bars, 25 mm.
inside living cells. We also investigated the time-dependent
fluorescence response of NBCD over 45 min after the addition
of H2O2 (Figure S11). To demonstrate that the red fluorescence
was caused by the reaction of NBCD with H2O2, control experi-
ments were performed by co-incubation of catalase with H2O2
that showed an extremely low level of fluorescence (Figure 4c).
Moreover, fluorescence could still be observed upon exposure
to only 2 mm H2O2, which further demonstrated the good sensi-
tivity of NBCD (Figure S12), although the quantification of fluo-
rescence intensity indicated that only small amounts of NBCD
could pass through the cell membrane (Figure S14).[1c]
Keywords: fluorescent probes · hydrogen peroxide · imaging
agents
· N-protected 3,7-dihydroxyphenoxazine · reactive
oxygen species · sensors
[1] a) S. G. Rhee, Science 2006, 312, 1882–1883; b) E. A. Veal, A. M. Day,
B. A. Morgan, Mol. Cell 2007, 26, 1–14; c) Y. Hitomi, T. Takeyasu, M.
Kodera, Chem. Commun. 2013, 49, 9929–9931; d) L. Yuan, W. Lin, Y. Xie,
B. Chen, S. Zhu, J. Am. Chem. Soc. 2012, 134, 1305–1315.
[2] J. A. Imlay, Annu. Rev. Biochem. 2008, 77, 755.
[3] a) T. R. Hurd, M. DeGennaro, R. Lehmann, Trends cell biol. 2012, 22, 107–
115; b) E. Veal, A. Day, Antioxid. Redox Signaling 2011, 15, 147–151.
Chem. Asian J. 2016, 11, 818 – 822
www.chemasianj.org
821
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
[4] a) B. C. Dickinson, Y. Tang, Z. Chang, C. J. Chang, Chem. Biol. 2011, 18,
943–948; b) W.-K. Oh, Y. S. Jeong, S. Kim, J. Jang, ACS Nano 2012, 6,
8516–8524.
Anal. Chem. 2013, 85, 3926–3932; c) W. Gao, B. Xing, R. Y. Tsien, J. Rao,
J. Am. Chem. Soc. 2003, 125, 11146–11147; d) A. E. Albers, K. A. Rawls,
C. J. Chang, Chem. Commun. 2007, 4647–4649.
[5] a) T. Finkel, M. Serrano, M. A. Blasco, Nature 2007, 448, 767–774; b) K. J.
Barnham, C. L. Masters, A. I. Bush, Nat. Rev. Drug Discovery 2004, 3,
205–214; c) M. T. Lin, M. F. Beal, Nature 2006, 443, 787–795.
[6] B. C. Dickinson, C. Huynh, C. J. Chang, J. Am. Chem. Soc. 2010, 132,
5906–5915.
[7] L. C. Lo, C. Y. Chu, Chem. Commun. 2003, 2728–2729.
[8] a) W. X. Wu, J. Li, L. Z. Chen, Z. Ma, W. Zhang, Z. Z. Liu, Y. N. Cheng, L. P.
Du, M. Y. Li, Anal. Chem. 2014, 86, 9800–9806; b) M. Sekiya, K. Umeza-
wa, A. Sato, D. Citterio, K. Suzuki, Chem. Commun. 2009, 3047–3049.
[9] C. Bueno, M. Villegas, S. Bertolotti, C. Previtali, M. Neumann, M. Encinas,
Photochem. Photobiol. 2002, 76, 385–390.
[12] M. Zhou, Z. Diwu, N. Panchuk-Voloshina, R. P. Haugland, Anal. Biochem.
1997, 253, 162–168.
[13] a) M. C. Y. Chang, A. Pralle, E. Y. Isacoff, C. J. Chang, J. Am. Chem. Soc.
2004, 126, 15392–15393; b) D. Srikun, E. W. Miller, D. W. Dornaille, C. J.
Chang, J. Am. Chem. Soc. 2008, 130, 4596; c) V. Carroll, B. W. Michel, J.
Blecha, H. VanBrocklin, K. Keshari, D. Wilson, C. J. Chang, J. Am. Chem.
Soc. 2014, 136, 14742–14745; d) Y. Wen, K. Liu, H. Yang, Y. Li, H. Lan, Y.
Liu, X. Zhang, T. Yi, Anal. Chem. 2014, 86, 9970–9976; e) Y. Wen, K. Liu,
H. Yang, Y. Liu, L. Chen, Z. Liu, C. Huang, T. Yi, Anal. Chem. 2015, 87,
10579–10584.
[10] Y. Hitomi, T. Takeyasu, T. Funabiki, M. Kodera, Anal. Chem. 2011, 83,
9213–9216.
[11] a) D. J. Simpson, C. J. Unkefer, T. W. Whaley, B. L. Marrone, J. Org. Chem.
1991, 56, 5391–5396; b) Z. Li, X. Li, X. Gao, Y. Zhang, W. Shi, H. Ma,
Manuscript received: November 23, 2015
Accepted Article published: January 25, 2016
Final Article published: February 10, 2016
Chem. Asian J. 2016, 11, 818 – 822
www.chemasianj.org
822
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim