A ratiometric lysosomal pH chemosensor based on fluorescence resonance energy transfer

Article abstract of DOI:10.1016/j.dyepig.2013.06.032

In this work, we presented a naphthalimide-rhodamine based fluorescence resonance energy transfer system (FRET) NR1 as a ratiometric and intracellular pH probe, in which 1,2,3-triazole was identified as an ideal bridge and biocompatibility. It could selectively monitor pH variations in methanol/HEPES solution at room temperature. When the pH changed from 6.20 to 2.00, both the fluorescence intensities at 580 nm and the intensity ratios, R (I _{580 nm}/I_{538 nm}) increased, which allowed the detection of pH changes by both normal fluorescence and ratiometric fluorescence methods. The observation is consistent with the increased FRET from the 1,8-naphthalimide (donor) to the ring-opened, colored form of rhodamine (acceptor). Moreover, as NR1 is lysosomal with low cytotoxicity, it will be helpful for investigating the pivotal role of H⁺ in a biological context, especially in lysosomes through direct intracellular imaging.

Full text of DOI:10.1016/j.dyepig.2013.06.032

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Dyes and Pigments 99 (2013) 620e626
Contents lists available at SciVerse ScienceDirect
Dyes and Pigments
journal homepage: www.elsevier.com/locate/dyepig
A ratiometric lysosomal pH chemosensor based on fluorescence
resonance energy transfer
Jiangli Fan a, Chunying Lin a, Honglin Li a, Peng Zhan a, Jingyun Wang b, Shuang Cui b,
,
Mingming Hu a, Guanghui Cheng a, Xiaojun Peng a
*
a State Key Laboratory of Fine Chemicals, Dalian University of Technology, No. 2 Linggong Road, High-tech District, Dalian 116024, PR China
b School of Life Science and Biotechnology, Dalian University of Technology, 2 Linggong Road, Dalian 116024, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
In this work, we presented a naphthalimideerhodamine based fluorescence resonance energy transfer
system (FRET) NR1 as a ratiometric and intracellular pH probe, in which 1,2,3-triazole was identified as
an ideal bridge and biocompatibility. It could selectively monitor pH variations in methanol/HEPES so-
lution at room temperature. When the pH changed from 6.20 to 2.00, both the fluorescence intensities at
580 nm and the intensity ratios, R (I580 nm/I538 nm) increased, which allowed the detection of pH changes
by both normal fluorescence and ratiometric fluorescence methods. The observation is consistent with
the increased FRET from the 1,8-naphthalimide (donor) to the ring-opened, colored form of rhodamine
(acceptor). Moreover, as NR1 is lysosomal with low cytotoxicity, it will be helpful for investigating the
pivotal role of Hþ in a biological context, especially in lysosomes through direct intracellular imaging.
Ó 2013 Elsevier Ltd. All rights reserved.
Received 8 June 2013
Received in revised form
27 June 2013
Accepted 29 June 2013
Available online 13 July 2013
Keywords:
Naphthalimideerhodamine
FRET
Ratiometric sensing
pH
Fluorescent sensor
Lysosome localization
1. Introduction
Most reported fluorescent pH sensors for live cells function by
the enhancement of fluorescence signals. As the change in fluo-
Suitable intracellular pH (pHi) [1] is critical for many tissue, cell
and enzyme activities. However, abnormal pHi values are often
associated with inappropriate cell function, growth, and division;
and are observed in some common pathological states, such as
cancers [2] and Alzheimer’s disease [3]. Intracellular acidification is
also an early feature of apoptosis. Thus, Hþ forms one of the most
important targets among intracellular species.
In recent years, small fluorescent organic molecules, quantum
dots and other nanoparticles [4e14] have been widely used for
measuring pH changes in solutions or live cells. However, it is
notable that there are only a few reagents for pH detection in
weakly acidic environments. Moreover of these even fewer are cell
membrane permeable, and so can monitor Hþ activities by cell
imaging. In particular, since media such as those found in some
organelles, e.g., endosomes and plant vacuoles with intra-
compartmental pH of 4e6, lysosomes with pH of 4.5e5.5, devel-
opment of a wide variety of membrane-permeable “acidic” probes
of high sensitivity and selectivity is desirable.
rescence intensity is the only detection signal, factors such as
instrumental efficiency, environmental conditions, and probe
concentration can interfere with the signal output. Accordingly,
ratiometric sensors, which can eliminate most or all ambiguities by
self calibration of two emission bands, are advantageous. Conse-
quently, fluorescence (Förster) resonance energy transfer (FRET) is
an optimal strategy, which can supply two fluorescence indicators
(energy transfer donor and acceptor) [15,16]. In addition, the
pseudo-Stokes shift of a FRET-based energy cassette is larger than
the Stokes shifts of either the donor or acceptor dyes. This greatly
reduces self-quenching, and fluorescence detection error due to the
excitation backscattering effects.
Here, we present a naphthalimideerhodamine FRET system
NR1 as a ratiometric and intracellular pH sensor. In recent years
the strategy of connecting a shorter-wavelength fluorophore to a
rhodamine spirolactam which can act as the energy acceptor has
provided several ratiometric sensors based on ring-opening re-
actions induced by metal ions [17e19]. Furthermore, we chose
naphthalimide as the energy donor since the emission spectrum of
naphthalimide and the excitation spectrum of rhodamine B have
substantial overlap, giving effective fluorescence energy transfer
* Corresponding author. Tel./fax: þ86 411 84986306.
E-mail address: pengxj@dlut.edu.cn (X. Peng).
0143-7208/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.dyepig.2013.06.032

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
J. Fan et al. / Dyes and Pigments 99 (2013) 620e626
621
with a single excitation wavelength. Given this molecular design,
click chemistry provides a convenient synthetic method for unit-
ing the two fluorophores, and the resultant 1,2,3-triazole was
identified as an ideal bridge and biocompatibility [20,21]. Then, a
Hþ-induced process can change the emission maximum of the
system from 538 nm (the characteristic peak of naphthalimide) to
580 nm (the characteristic peak of rhodamine). This wavelength
shift allows the ratiometric detection of pH both in aqueous so-
lution and live cells.
2.2.4. Synthesis of 2
Compound 4 (100 mg, 0.27 mmol) and NaN3 (87.75 mg,
1.35 mmol) were added to 20 mL ethanol. After heating at 110 ꢀC for
24 h, the solvent was removed on a rotary evaporator. The resulting
crude product was purified on a silica gel column with CH2Cl2/ethyl
acetate as eluent to give compound 2 (54.08 mg, yield: 59.4%). 1H
NMR (400 MHz, CDCl3), dH (ppm): 8.56 (m, 1H), 8.44 (d, 1H,
J ¼ 8.0 Hz), 8.08 (t, 1H, J ¼ 4.0 Hz), 7.59 (m, 1H), 6.69 (d, 1H,
J ¼ 4.0 Hz), 5.36 (s, 1H), 4.42 (t, 2H, J ¼ 8.0 Hz), 3.64 (t, 2H,
J ¼ 8.0 Hz), 3.40 (m, 2H), 1.80 (m, 2H), 1.55 (m, 2H), 1.02 (t, 3H,
J ¼ 8.0 Hz), Q-TOFMS: [M ꢁ H]þ: 336.1465; found: 336.1526.
2. Experimental section
2.1. Apparatus and general methods
2.2.5. Synthesis of 1
Rhodamine hydrazide 3 was synthesized from rhodamine B by
the procedure published in literature [24]. Compound 3 (300.0 mg,
0.66 mmol), K2CO3 (96.6 mg, 0.7 mmol) was dissolved in 20 mL
ethyl acetate in a 50 mL flask, excess (1.0 mL) 3-bromopropyne was
then added dropwise with vigorous stirring. The mixture was
refluxed overnight. After removal of ethyl acetate under vacuum,
the residue was purified by flash chromatography with CH2Cl2/
ethyl acetate as eluent to give compound 1 as a white powder
(100 mg, yield: 30.6%). TLC analysis: Rf ¼ 0.6 in 5.0% ethyl acetate in
CH2Cl2. 1H NMR (400 MHz, CDCl3), dH (ppm): 7.93 (t, 1H, J ¼ 4.0 Hz),
7.47 (m, 2H), 7.11 (d, 1H, J ¼ 4.0 Hz), 6.48 (d, 2H, J ¼ 8.0 Hz), 6.41 (s,
2H), 6.28 (d, 2H, J ¼ 4.0 Hz), 4.58 (t, 1H, J ¼ 8.0 Hz), 3.33 (q, 8H), 3.32
(d, 2H, J ¼ 4.0 Hz), 2.10 (s, 1H), 1.16 (t, 12H, J ¼ 8.0 Hz). 13C NMR
(100 MHz, CDCl3), dC: 166.79, 153.81, 151.90, 148.95, 132.90, 129.98,
128.44, 127.38, 124.11, 122.99, 107.97, 105.72, 98.11, 80.21, 72.54,
65.49, 44.47, 40.55, 12.76. Q-TOFMS: [M þ H]þ: 495.2760; found:
495.2760.
All solvents used were of analytical grade without further pu-
rification. 1H NMR and 13C NMR spectra were recorded on a VAR-
IAN INOVA-400 spectrometer, using TMS as an internal standard.
Mass spectrometry data were obtained with a HP1100LC/MSD
mass spectrometer and a LC/Q-TOF MS spectrometer. UVevisible
spectra were collected on a Perkin Elmer Lambda 35 UVeVis
spectrophotometer. Fluorescence measurements were performed
on a VAEIAN CARY Eclipse Fluorescence Spectrophotometer with a
slit width of 5 nm for excitation and 2.5 nm for emission, respec-
tively. All pH measurements were made with a Model PHS-3C
meter (SHANGHAI PRECISION & SCIENTIFIC INSTRUMENT CO.,
LTD). The fluorescence images of cells for the staining experiments
were performed with an Olympus FV1000 confocal laser scanning
microscope.
2.2. Synthetic routes of NR1
2.2.1. Synthesis of 6
2.2.6. Synthesis of NR1
According to the literature [22], compound 6 was synthesized by
refluxing compound 7 and ethanolamine for 4 h in ethanol. After
filtration, the crude product was recrystallized from ethanol as a
white product with a yield of 70.3%. 1H NMR (400 MHz, CDCl3), dH
(ppm): 8.68 (d, 1H, J ¼ 8.0 Hz), 8.60 (d, 1H, J ¼ 8.0 Hz), 8.44 (d, 1H,
J ¼ 4.0 Hz), 8.06 (d, 1H, J ¼ 8.0 Hz), 7.87 (t, 1H, J ¼ 8.0), 4.46 (t, 2H,
J ¼ 4.0 Hz), 3.99 (t, 2H, J ¼ 4.0 Hz). Q-TOFMS: [M þ H]þ: 320.1432,
found: 320.1430.
Compound 2 (42 mg, 0.124 mmol) and compound 1 (61.29 mg,
0.124 mmol) were added to 10 mL DMF. After CuSO4$5H2O and
sodium ascorbate (Sodium AC) were added to the solution, the
mixture was stirred under N2 at room temperature for 12 h. Excess
water was added to the mixture, which was extracted with CH2Cl2
and dried over magnesium sulfate. After removal of solvent under
vacuum, the residue was purified on a silica gel column with
CH2Cl2/ethyl acetate as eluant to give NR1 (74 mg, yield: 72.0%). 1H
NMR (400 MHz, CDCl3), dH (ppm): 8.48 (d, 1H, J ¼ 4.0 Hz), 8.39 (d,
1H, J ¼ 8.0 Hz), 8.15 (d, 1H, J ¼ 16.0 Hz), 7.99 (d, 1H, J ¼ 14.0 Hz), 7.52
(t,1H, J ¼ 8.0 Hz), 7.44 (t, 2H, J ¼ 8.0 Hz), 7.08 (d, 1H, J ¼ 8.0 Hz), 6.68
(d, 2H, J ¼ 8.0 Hz), 6.59 (d, 2H, J ¼ 4.0 Hz), 6.43 (s, 2H), 5.60 (s, 1H),
5.34 (s, 1H), 4.61 (d, 2H, J ¼ 4.0 Hz), 4.54 (d, 2H, J ¼ 8.0 Hz), 3.35 (q,
8H), 2.22 (t, 2H, J ¼ 8.0 Hz), 1.77 (t, 3H, J ¼ 8.0 Hz), 1.54 (q, 2H), 1.16
(t, 12H, J ¼ 8.0 Hz). 13C NMR (100 MHz, CDCl3), dC: 175.60, 165.53,
164.45, 163.67, 152.57, 150.04, 149.05, 145.49, 134.93, 133.69, 131.46,
130.01, 129.90, 129.75, 128.84, 128.16, 127.37, 126.61, 124.56, 123.47,
122.25, 120.16, 109.10, 108.10, 104.95, 104.27, 98.41, 71.82, 65.53,
47.92, 44.34, 43.40, 39.30, 35.93, 31.92, 30.91, 29.70, 29.53, 29.34,
29.24, 29.13, 27.22, 25.54, 22.70, 20.35, 19.17, 14.14, 13.87, 12.63.
Q-TOFMS: [M þ H]þ: 830.4148; found: 830.4152.
2.2.2. Synthesis of 5
n-Butylamine (146.3 mg, 2 mmol) was added to compound 6
(160 mg, 0.5 mmol) in acetonitrile (30 mL), and the mixture
refluxed at 90 ꢀC for 20 h under nitrogen (TLC monitoring). After
removal of acetonitrile and residual n-butylamine under vacuum,
the residue was purified by flash chromatography with petroleum
ether/ethyl acetate as eluent to give a bright yellow powder, com-
pound 5 (122.13 mg, yield: 78.2%). TLC analysis: Rf ¼ 0.6 in 50% ethyl
acetate in petroleum ether. The structure of compound 5 was
established by 1H NMR, 13C NMR, and MS [23].
2.2.3. Synthesis of 4
Excess (1.0 mL) PBr3 was added dropwise to compound 5
(60 mg, 0.19 mmol) in ClCH2CH2Cl (10 mL). This was heated at 45 ꢀC
for 4 h. A little water was added to the mixture, which was then
extracted with CH2Cl2 and dried over magnesium sulfate. After
removal of solvent under vacuum, the residue was purified by flash
chromatography with CH2Cl2 as eluent to give compound 4 as a
bright yellow powder (17 mg, yield: 23.8%). 1H NMR (400 MHz,
CDCl3), dH (ppm): 8.56 (q, 1H), 8.45 (d, 1H, J ¼ 8.0), 8.09 (t, 1H,
J ¼ 4.0), 7.59 (q, 1H), 6.70 (d, 1H, J ¼ 8.0 Hz), 4.42 (t, 2H, J ¼ 4.0 Hz),
3.64 (t, 2H, J ¼ 4.0 Hz), 3.40 (t, 2H, J ¼ 8.0 Hz), 1.80 (t, 2H, J ¼ 8.0 Hz),
1.53 (q, 2H), 1.02 (t, 3H, J ¼ 8.0 Hz). Q-TOFMS: [M ꢁ H]þ: 373.0796;
found: 373.0627.
2.3. FRET efficiency (E)
FRET efficiency can be obtained by measuring either the fluo-
rescence intensities of the donor with and without an acceptor
probe [25]. Thus
ID
E ¼ 1 ꢁ
ID0
where ID and ID0 are the intensities in the presence and the absence
of acceptor.

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
622
J. Fan et al. / Dyes and Pigments 99 (2013) 620e626
2.4. Theoretical calculations
2.6. Measurement of cell viability
The structures of NR1 in the absence and presence of Hþ were
optimized using density functional theory (DFT) by the B3LYP
method with the 6-31G basis set. The DFT calculations were per-
formed using the Gaussian 09 program.
Cell viability was evaluated by the reduction of MTT (3-(4,5)-
dimethylthiahiazo(-z-y1)-3,5-diphenytetrazoliumromide) to for-
mazan crystals by mitochondrial dehydrogenases (Mosmann,
1983). Briefly, MCF-7 cells were seeded in 96-well microplates
(Nunc, Denmark) at a density of 1 ꢂ 105cells/mL in 100
mL medium
containing 10% FBS. After 24 h of cell attachment, plates were
2.5. pH bioimaging and lysosomes localization in MCF-7 cells
washed with 100
cells were cultured in medium with various concentrations (0, 1, 5
and 10 M) of NR1 and lysosensor green for 12 h. Cells in culture
mL/well phosphate buffered saline (PBS) and then
MCF-7 cells were cultured in Dulbecco’s modified Eagle’s me-
dium (DMEM, Gibco; Invitrogen), supplemented with 100 units/mL
penicillin, 100 g/mL streptomycin, and 10% heat-inactivated fetal
bovine serum at 37 ꢀC in a 5% CO2 atmosphere. The cells were
seeded on a Ø 35 mm confocal laser dish at a density of
2 ꢂ 104 cells mLꢁ1 in culture medium overnight. Then the MCF-7
m
medium without NR1 and lysosensor green were used as the
control. Six replicate wells were used for each control and test
concentration. 10 mL of MTT (5 mg/mL) prepared in PBS was added
to each well and the plates were incubated at 37 ꢀC for 4 h in a 5%
CO2 humidified incubator. The medium was then carefully
removed, and the purple products were lysed in 200 mL dimethyl
cells were treated with 10
mM of NR1 for 0.5 h and washed 3
times with prewarmed PBS (pH 7.20). To increase the concentration
of cellular Hþ, MCF-7 cells were treated with PBS (pH 4.45). Then
the MCF-7 cells in different pH media were excited at 405 nm to
obtain images with both white light and fluorescence, by using a
digital color camera system (Olympus). To confirm the sensor’s
subcellular localization, acidic organelle specific lysosensor green
was used to co-stain cells with NR1. MCF-7 cells were incubated for
sulfoxide (DMSO). The plate was shaken for 10 min and the
absorbance was measured at 405 nm and 488 nm using a micro-
plate reader (Thermo Fisher Scientific). Cell viability was expressed
as a percent of the control culture value.
3. Results and discussions
30 min in the presence of lysosensor green (1 mM) and NR1 (10 mM)
simultaneously. The medium was removed, washed 3 times with
prewarmed PBS (pH 7.20) and replaced with fresh medium. Cells
were then analyzed with Olympus FV1000 confocal microscope.
Images were obtained with both white light and fluorescence.
Excitation wavelengths of lysosensor green and NR1 were 488 nm
and 405 nm respectively.
Both 1 [26] with a propagyl group and 2 with an azide group
were efficiently synthesized and well characterized. Then, NR1 was
obtained easily by a click reaction between and 1 and 2 with 72.0%
yield (Scheme 1). It was notable that this formation also contained
an oxidative dehydrogenation of amine to imine moiety, which was
a well known reaction in biochemistry promoted by coordination of
Scheme 1. The synthetic route of NR1.

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
J. Fan et al. / Dyes and Pigments 99 (2013) 620e626
623
NR1 is both hydrophilic and lipophilic, thus an optimized
methanol/HEPES (9/1, v/v, pH 7.2) solution was selected as a testing
system to investigate the optical response of NR1 to pH at room
temperature. NR1 can instantly respond to a change of Hþ con-
centration and the resulting fluorescence is stable for a consider-
able time at pH 4.45 and 2.90 respectively (Fig. S3).
As shown in Fig. 1a, NR1 showed green fluorescence and a
yellow color when pH > 6.2, which was mainly ascribed to the
optical properties of 1,8-naphthalimide. When the pH changed
from 7.20 to 2.00, the intensity of the green fluorescent band
centered at 538 nm gradually decreased and a new pink fluorescent
band centered at 580 nm gradually increased. (Excitation wave-
length is 430 nm) Accordingly, UV-absorption of the rhodamine
moiety gradually increased, thus the solution of the probe under-
went a distinct color change from yellow to pink, indicating that the
probe NR1 can serve as a “naked-eye” indicator for Hþ concentra-
tion (Fig. 1b). The observation is consistent with the increased FRET
from the 1,8-naphthalimide (donor) to the ring-opened, colored
form of rhodamine (acceptor). Both the fluorescence intensities at
580 nm (Fig. 1a) and the intensity ratios, R (I580 nm/I538 nm) (Fig. S4)
increased, which allowed the detection of Hþ by both normal
fluorescence and ratiometric fluorescence methods. Actually, this
took place mainly over the pH range 2.00e6.20, and the ratiometric
calibration curve of R (I580 nm/I538 nm) indicates the pKa is 2.79
(Fig. S4). This is mainly because the spirolactam ting of NR1 is
“open” in the acidic conditions. The fluorescence ratio changes of
NR1 were fully reversible when it was treated with acid or base
between pH 2.90 and 7.20 as shown in Fig. S5, which is of consid-
erable potential for diagnosis and understanding of pathology.
To further discuss the FRET process between naphthalimide and
rhodamine fluorophore, the structures of NR1 before and after Hþ-
induced spirolactam ring open were optimized by Guassian 09 as
shown in Fig. 2. Although the two distances between the donor and
acceptor (1.31 nm and 1.01 nm) are similar, which are both in the
range of 1e10 nm for FRET to occur, FRET only took place when the
two fluorophores tended to parallel. So we speculated that the
space arrangement between the energy transfer donor and
acceptor also would influence the FRET or not. In this case, the FRET
efficiency was 72%, that is why the fluorescence intensity of
naphthalimide could not be quenched completely but ensured
the double-wavelength emissions with ratiometric fluorescence
detection of pH changes in live cells.
Fig. 1. Emission spectra of NR1 (5
addition of HCl (1 M) in solution (methanol/HEPES ¼ 9/1, v/v, pH 7.2). Inset: Change in
fluorescence of NR1 (5 M). Excitation wavelength is
M) upon addition of Hþ (100
430 nm. Slit: 5 nm/2.5 nm.
mM) at pH approximately range from 7.2 to 2.0 upon
m
m
a transition metal cation such as Cu2þ [27,28]. A peak of m/z
Considering that nitrogen and oxygen can bind to many metal
ions in solution, it is important to determine whether non-Hþ ions
were potential interferents. As shown in Fig. 3a, when keeping
experimental conditions unchanged at pH 7.2, there was no obvious
change of fluorescence ratio (I580 nm/I538 nm) upon the addition of
830.4148 in Q-TOFMS and the elemental composition of
C
49H52N9O4 suggested that this dehydrogenation reaction of amine
group (Fig. S1). It was also confirmed by IR spectrum of NR1, in
which a sharp band assigned to the C]N stretching vibrations at ca.
1636 cmꢁ1 (Fig. S2).
Fig. 2. The lowest energy structure of NR1 in the absence (a) and presence (b) of Hþ by DFT calculations at B3LYP/6-31G (d, p) level.