PH-activatable near-infrared fluorescent probes for detection of lysosomal pH inside living cells

Article abstract of DOI:10.1039/c4tb00475b

Four near-infrared fluorescent probes (A, B, C and D) have been synthesized, characterized, and evaluated for detection of lysosomal pH inside living cells. The fluorescent probes display highly sensitive and selective fluorescent response to acidic pH as the acidic pH results in drastic structural changes from spirocyclic (non-fluorescent) forms to ring-opening (fluorescent) forms of the fluorescent probes. The fluorescence intensities of the fluorescent probes (B, C and D) increase significantly by more than 200-fold from pH 7.4 to 4.2. The fluorescent probe D bearing the N-(2-hydroxyethyl) ethylene amide residue possesses the advantages of high sensitivity, excellent photostability, good cell membrane permeability, strong pH dependence, and low auto-fluorescence background. It has been successfully applied to selectively stain lysosomes and detect lysosomal pH changes inside normal endothelial and breast cancer cells. This journal is the Partner Organisations 2014.

Full text of DOI:10.1039/c4tb00475b

Journal of
Materials Chemistry B
View Article Online
View Journal | View Issue
PAPER
pH-activatable near-infrared fluorescent probes for
detection of lysosomal pH inside living cells†
Cite this: J. Mater. Chem. B, 2014, 2,
4500
a
a
a
Giri K. Vegesna, Jagadeesh Janjanam, Jianheng Bi, Fen-Tair Luo, b Jingtuo Zhang,a
*
a
a
a
*
*
Connor Olds, Ashutosh Tiwari and Haiying Liu
Four near-infrared fluorescent probes (A, B, C and D) have been synthesized, characterized, and evaluated
for detection of lysosomal pH inside living cells. The fluorescent probes display highly sensitive and selective
fluorescent response to acidic pH as the acidic pH results in drastic structural changes from spirocyclic
(non-fluorescent) forms to ring-opening (fluorescent) forms of the fluorescent probes. The fluorescence
intensities of the fluorescent probes (B, C and D) increase significantly by more than 200-fold from pH
7.4 to 4.2. The fluorescent probe D bearing the N-(2-hydroxyethyl) ethylene amide residue possesses
the advantages of high sensitivity, excellent photostability, good cell membrane permeability, strong pH
dependence, and low auto-fluorescence background. It has been successfully applied to selectively stain
lysosomes and detect lysosomal pH changes inside normal endothelial and breast cancer cells.
Received 25th March 2014
Accepted 15th May 2014
DOI: 10.1039/c4tb00475b
www.rsc.org/MaterialsB
most powerful tools for monitoring intracellular pH, and
possesses many technical and practical advantages over other
Introduction
methods because it can detect the intracellular pH of intact cells
and subcellular regions with operational simplicity, high
sensitivity, and excellent spatial and temporal resolution.11–21
Although several uorescent probes for pH have been devel-
oped,23 only a few of them have been applied to detect lysosomal
pH inside living cells. Most uorescent probes including the
commercial ones take advantages of lysosomotropism where
their weak bases of tertiary amine groups help the probes
selectively accumulate in acidic lysosomes through the
protonation of the amine groups in a cellular acidic environ-
ment.18 Their pH sensitivity primarily results from protonation
of ionizable tertiary amine groups on the uorophores in acidic
lysosomes enhancing the probe uorescence through the
suppression of photo-induced electron transfer from the
tertiary amine to the probe uorophores.15 The potential
drawback for these uorescent probes is their broad pH
response and relative high uorescent background at pH 7.4. In
order to address the uorescent background issue, a few uo-
rescent probes for pH based on uorescein and Rhodamine
dyes have been developed to take advantages of low uores-
cence background at pH 7.4 because spirolactam rings of the
uorophores exists in the “ring-closed” state at pH 7.4.12,13,20,24
However, some uorescent probes can result in cell damage
because of their short absorption and emission wavelengths
with less than 600 nm. In this paper, we report four near-
infrared uorescent probes (A, B, C and D) with spirocyclic
structures which are highly sensitive to the pH changes of the
solutions. We chose four different residues of uorescent
probes in order to investigate effect of hydrophobic and
hydrophilic residues on the cell imaging experiments. The
Intracellular pH, known as pHi, plays a pivotal role in cell
function and regulation such as cell volume regulation, vesicle
trafficking, cellular metabolism, cell membrane polarity,
cellular signaling, cell activation, growth, proliferation and
apoptosis.1–4 Abnormal pH values in organelles are associated
with many diseases such as cancer and Alzheimer's.5,6 A low
intra-compartmental pH in organelles functions to denature
proteins or activate enzymes that are normally inactive around
neutral pH.7 For example, the membrane around lysosomes at
pH 4.5–5.0 enables digestive enzymes to degrade proteins, DNA,
RNA, polysaccharides, lipids, viruses and bacteria.4,7,8 Signi-
cant disruptive changes in the lysosomal pH can lead to lyso-
some malfunction and consequently result in lysosomal storage
diseases.9 Therefore, it is very important to precisely monitor
lysosomal pH inside living cells in order to investigate cellular
functions and understand physiological and pathological
processes. A variety of techniques such as microelectrodes,
nuclear magnetic resonance,10 absorbance spectroscopy, and
uorescence imaging and spectroscopy11–22 have been devel-
oped to measure intracellular pH. Fluorescence spectroscopy
using pH-sensitive uorescent probes is becoming one of the
aDepartment of Chemistry, Michigan Technological University, Houghton, MI 49931,
USA. E-mail: hyliu@mtu.edu; tiwari@mtu.edu
bInstitute of Chemistry, Academia Sinica, Taipei, Taiwan 11529, Republic of China.
E-mail: luo@gate.sinica.edu.tw
† Electronic supplementary information (ESI) available: 1H NMR and 13C NMR,
absorption and emission spectra of the uorescent probes, absorption and
uorescent spectra of the uorescent probes at different pH values, and
uorescent spectra of uorescent probes at pH 7.0 and 4.1 in the absence and
presence of different metal ions, respectively. See DOI: 10.1039/c4tb00475b
4500 | J. Mater. Chem. B, 2014, 2, 4500–4508
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry B
solution was used every single time under different pH envi-
ronment. The citric acid sodium phosphate buffer solution was
prepared freshly. The excitation wavelength at 670 nm was used
to excite all uorescent dyes.
Materials
Unless otherwise indicated, all reagents and solvents were
obtained from commercial suppliers and used without further
purication. Air- and moisture-sensitive reactions were con-
ducted in oven-dried glassware using a standard Schlenk line or
drybox techniques under an inert atmosphere of dry nitrogen.
Fluorescent probe A. When phosphorous oxychloride (0.68 g,
4.4 mmol) was added to the compound 5 (0.5 g, 0.89 mmol) in
1,2-dichloroethane (20 mL) under a nitrogen atmosphere, the
resulting reaction mixture was reuxed for 4 hours. Aer the
solvent was removed under reduced pressure, dry acetonitrile
(30 mL) was added to the reaction residue. When 1,2-dia-
minobenzene (6) (0.38 g, 3.5 mmol) and triethylamine (0.5 mL)
was added to the reaction solution, the reaction mixture was
stirred at room temperature for overnight. Aer the solvent was
removed under reduced pressure, the reaction residue was
dissolved in dichloromethane (50 mL), washed with water (2 ꢁ
20 mL) and brine solution (2 ꢁ 20 mL), respectively. Organic
layer was collected, dried over anhydrous sodium sulfate,
ltered and concentrated under reduced pressure. The crude
compound was puried by ash column chromatography using
EtOAc/hexane (50/50) to yield yellow crystal (0.34 g, 60%). 1H
NMR (400 MHz, CDCl3): d 7.97–7.95 (d, J ¼ 8.0 Hz, 1H), 7.54–
7.45 (m, 2H), 7.34–7.31 (d, J ¼ 12.0 Hz, 1H), 7.24–7.22 (d, J ¼ 8.0
Hz, 1H), 7.16–7.11 (m, 2H), 6.99–6.95 (t, J ¼ 8.0 Hz, 1H), 6.83–
6.80 (t, J ¼ 8.0 Hz, 1H), 6.66–6.52 (m, 5H), 6.31–6.24 (m, 2H),
5.34–5.31 (d, J ¼ 12.0 Hz, 1H), 3.36–3.31 (q, J ¼ 8.0 Hz, 4H), 3.11
(s, 3H), 2.63–2.57 (m, 1H), 2.38–2.30 (m, 1H), 2.15–2.04 (m, 1H),
1.66–1.61 (m, 7H), 1.37–1.25 (m, 2H), 1.18–1.14 (t, J ¼ 8.0 Hz,
6H). 13C NMR (100 MHz, CDCl3): d 167.10, 157.65, 153.41,
152.40, 148.97, 148.32, 145.56, 144.14, 139.06, 132.66, 131.63,
128.79, 128.42, 127.88, 127.38, 123.96, 123.54, 123.47, 121.71,
120.85, 119.42, 119.34, 118.93, 118.10, 108.57, 105.85, 104.38,
Scheme 1 Spirocyclic ring closing and opening forms of near-infrared
fluorescent probes upon pH changes.
uorescent probes A and C possess hydrophobic residues such
as aminophenyl amide, and ethynyl amide residues while the
uorescent probes B and D bear hydrophilic residues such as
amino amide and N-(2-hydroxyethyl) ethylene amide residues,
respectively. Under neutral or basic conditions, the uorescent
probes retain the spirocyclic form that is non-uorescent and
colorless. Acidic environment effectively triggers ring opening
of the spirocyclic form in uorescent probes, and results in
strong uorescence (Scheme 1). Three sensitive near-infrared
uorescent probes (B, C and D) have been used to detect lyso-
somal pH inside living cells with advantages of deep tissue light
penetration and low autouorescence. The uorescent probes
display extremely weak uorescence at the extracellular pH at
7.4 and become strongly uorescent at the intracellular lyso-
somal pH at 4.5. However, the uorescent probe A did not work
in cell imaging experiments because of its aminophenyl amide
hydrophobic residue.
Experimental section
Instrumentation
1H NMR and 13C NMR spectra were obtained by using a 98.24, 92.24, 70.89, 45.60, 44.54, 29.50, 29.29, 28.66, 28.36,
400 MHz Varian Unity Inova NMR spectrophotometer instru- 25.52, 24.60, 23.98, 22.28, 12.76. IR (cmꢀ1): 3346.38, 2968.95,
ment. H NMR and 13C NMR spectra were recorded in CDCl3, 2928.01, 1681.57, 1619.38, 1592.07, 1493.12, 1354.13, 1316.93,
1
chemical shis (d) were given in ppm relative to solvent peaks 1263.72, 1216.24, 1192.80, 1126.33, 1078.00, 1019.85, 930.06,
(1H: d 7.26; 13C: d 77.3) as internal standard. HRMS were 735.45, 702.26. HRMS (ESI) calcd for C43H45N4O2 [M + H]+,
measured with fast atom bombardment (FAB) ionization mass 649.3542; found, 649.3559.
spectrometer, double focusing magnetic mass spectrometer or
Fluorescent probe B. Aer compound 5 (0.30 g, 0.53 mmol)
matrix assisted laser desorption/ionization time of ight mass was dissolved in dry dichloromethane (30 mL) under nitrogen
spectrometer. Infrared (IR) spectra were obtained in the 400– atmosphere at room temperature, dicyclohexylcarbodiimide
4000 cmꢀ1 range using Perkin Elmer Spectrum One FTIR (DCC) (0.10 g, 0.53 mmol), and 4-dimethylaminopyridine (0.007
spectrometer. Absorption spectra were taken on a Perkin Elmer g, 0.06 mmol) were added to the reaction mixture at room
Lambda 35 UV/VIS spectrometer. Fluorescence spectra were temperature. When hydrazine hydrate (0.2 mL, 6.25 mmol) was
recorded on a Jobin Yvon Fluoromax-4 spectrouorometer. added to the mixture, the reaction mixture was stirred for 60
Fluorescence spectra of uorescent probes were measured by minutes at room temperature. Workup procedure was the same
using standard 1 cm path-length uorescence quartz cuvette at as that for uorescent probe A. The crude compound was
room temperature. All samples were scanned with increments puried by ash column chromatography using EtOAc/Hexane
1
of 1 nm. Each spectrum for absorbance and uorescence was (50/50) to get a yellow solid (0.15 g, 50%). H NMR (400 MHz,
measured at room temperature. 5 mM uorescent probe CDCl3): d 7.89–7.87 (d, J ¼ 8.0 Hz, 1H), 7.49–7.40 (m, 3H),
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. B, 2014, 2, 4500–4508 | 4501

View Article Online
Journal of Materials Chemistry B
Paper
7.19–7.13 (m, 3H), 6.86–6.79 (m, 1H), 6.60–6.59 (m, 1H), 6.39– 8H), 3.10 (s, 3H), 2.67–2.38 (m, 6H), 1.72–1.60 (m, 7H), 1.39–1.23
6.26 (m, 3H), 5.37–5.34 (d, J ¼ 12.0 Hz, 1H), 3.64 (brs, 2H), 3.36– (m, 3H), 1.16–1.12 (t, J ¼ 8.0 Hz, 6H). 13C NMR (100 MHz,
3.31 (q, J ¼ 8.0 Hz, 4H), 3.13 (s, 3H), 2.60–2.43 (m, 2H), 1.93–1.89 CDCl3): d 169.13, 157.96, 152.96, 151.81, 148.90, 148.26, 145.50,
(m, 1H), 1.71–1.69 (m, 6H), 1.36–1.24 (m, 3H), 1.18–1.14 (t, J ¼ 138.99, 132.42, 132.04, 128.60, 128.36, 127.93, 123.62, 123.01,
8.0 Hz, 6H). 13C NMR (100 MHz, CDCl3): d 166.50, 157.96, 121.72, 120.27, 119.87, 119.55, 108.66, 107.87, 105.93, 105.46,
153.45, 149.82, 148.99, 148.75, 145.55, 139.05, 132.50, 131.01, 103.60, 97.95, 92.24, 67.24, 61.06, 51.17, 48.14, 45.65, 44.57,
128.42, 127.91, 123.66, 123.17, 121.73, 120.32, 119.86, 119.49, 40.03, 30.58, 30.16, 29.89, 29.32, 28.71, 28.52, 25.55, 23.25,
108.61, 105.88, 104.29, 102.93, 98.17, 92.22, 68.02, 49.20, 45.64, 22.44, 12.78. IR (cmꢀ1): 3364.27, 2968.80, 2927.21, 1682.60,
44.59, 34.19, 30.26, 30.14, 29.33, 28.58, 28.50, 25.87, 25.55, 1621.06, 1592.96, 1515.14, 1492.83, 1466.55, 1352.63, 1316.87,
25.19, 23.20, 22.50, 12.78. IR (cmꢀ1): 2929.64, 1700.81, 1622.93, 1264.54, 1214.76, 1192.54, 1125.75, 1077.46, 929.65, 817.77,
1595.03, 1517.70, 1493.73, 1318.14, 1264.97, 1128.16, 735.27, 733.88, 701.34. HRMS (ESI) calcd for C41H49N4O3 [M + H]+,
702.01. HRMS (ESI) calcd for C37H41N4O2 [M + H]+, 573.3229; 645.3804; found, 645.3823.
found, 573.3247.
Fluorescent probe C. When compound 5 (0.30 g, 0.53 mmol)
was dissolved in dry dichloromethane (20 mL) under nitrogen
atmosphere at room temperature, N-hydroxy-succinimide (0.08
g, 0.69 mmol), dicyclohexylcarbodiimide (DCC) (0.11 g, 0.53
mmol) were added to the reaction mixture sequentially. Aer 30
minutes stirring, propargylamine (0.04 g, 0.80 mmol) was added
through a syringe under nitrogen atmosphere and the reaction
mixture was stirred at room temperature for 2 hours. Aer the
reaction mixture was washed with water (2 ꢁ 20 mL), the
organic layer was collected, dried over anhydrous sodium
sulfate and ltered. The ltrate was concentrated under
reduced pressure. The crude compound was puried by ash
column chromatography using EtOAc/Hexane (25/75) to afford
pale yellow syrupy compound (0.19 g, 60%). 1H NMR (400 MHz,
CDCl3): d 7.89–7.87 (d, J ¼ 8.0 Hz, 1H), 7.50–7.39 (m, 3H), 7.20–
7.14 (m, 3H), 6.88–6.79 (m, 1H), 6.65–6.53 (m, 1H), 6.34–6.24
(m, 3H), 5.44–5.27 (m, 1H), 4.24–4.19 (dd, J ¼ 16.0, 4.0 Hz, 1H),
3.76–3.71 (dd, J ¼ 16.0, 4.0 Hz, 1H), 3.36–3.31 (q, J ¼ 8.0 Hz, 4H),
3.20–3.03 (brs, 3H), 2.67–2.30 (m, 2H), 2.01–1.95 (m, 1H), 1.75–
Cell culture and confocal uorescence imaging
Breast cancer (MDA-MB-231) and normal endothelial (HUVEC-
C) cell lines were obtained from ATCC. The cells were cultured
according to the published procedures.25 Briey, the cells were
plated on 12-well culture plates or 35 mm glass bottom culture
dishes (MatTek Corp.) at a density of 1 ꢁ 105 cells per mL for
live cell imaging. Aer 24 h incubation at 37 ꢂC in 5% CO2
incubator, the media was removed and cells were rinsed with
1ꢁ PBS. Fresh serum free media with 5 or 20 mM of uorescent
probes A, B, C, D, and 5 were added and incubated for 2 hours.
Live cell imaging was performed with inverted uorescence
microscope (Model AMF-4306; EVOS, AMG) for initial dye
concentration standardization. The nal cell images were
obtained with confocal laser scanning microscope (Olympus
FV1000) with excitation wavelengths at 405 nm for Hoechst
33342 (Sigma-Aldrich), at 488 nm for LysoSensor Green DND-
189 (Invitrogen), and at 635 nm for uorescent probes A, B, C, D,
and 5. The uorescence images were obtained at 60ꢁ magni-
cation and the exposure times for each laser were kept constant
for each image series.
1.70 (m, 6H), 1.61–1.24 (m, 4H), 1.18–1.15 (t, J ¼ 8.0 Hz, 6H). 13
C
NMR (100 MHz, CDCl3): d 167.79, 153.26, 148.97, 132.66,
131.56, 128.70, 128.36, 127.97, 123.67, 123.26, 121.76, 119.55,
108.67, 105.93, 104.59, 97.99, 79.28, 70.43, 45.66, 44.60, 28.66,
25.56, 24.62, 23.43, 22.19, 12.81. IR (cmꢀ1): 3303.82, 2981.38,
1699.76, 1501.13, 1393.66, 1368.38, 1250.46, 1146.75, 1049.46,
855.42, 736.62, 702.53. HRMS (FAB) calcd for C40H42N3O2 [M]+,
596.3277; found, 596.3276.
Fluorescent probe D. Aer 1,10-carbonyldiimidazole (9)
(0.06 g, 0.4 mmol) was added to the solution of compound 5
(0.15 g, 0.26 mmol) in anhydrous dichloromethane (20 mL)
under nitrogen atmosphere, the reaction mixture was stirred at
room temperature for 4 hours. Aer the complete consumption
of compound 5 (conrmed by TLC), N-(2-hydroxyethyl)ethyl-
enediamine (0.05 mL, 0.53 mmol) was added to the mixture, the
reaction mixture was further stirred overnight. Aer work-up,
the organic layer was washed with water (2 ꢁ 30 mL) and brine
solution (2 ꢁ 30 mL), respectively, dried over anhydrous sodium
sulfate, and ltered. The ltrate was concentrated under
MTS assay
MDA-MB-231 cell lines were procured from ATCC. The cells
were cultured as described previously.25 Briey, the cells were
plated on 96-well culture plates at a density of 5000 cells per
well. Aer incubating the cells for 4 h to attach to the surface
fresh media with different concentrations (0, 5, 25 and 50 mM)
of probes A, B, C, D and 5 with 0.5% ethanol were added to the
wells. Each sample concentrations werꢂe repeated with 6 repli-
cates. The plates were incubated at 37 C in an incubator with
5% CO2 for 72 h. MTS assay was performed as described
previously.26
Results and discussion
Design and synthesis of near-infrared uorescent probes for
pH
reduced pressure. The crude compound was puried by Fluorescent dye (5) ((E)-2-(2-(9-(2-carboxyphenyl)-6-(diethyl-
preparative TLC plate with dichloromethane/EtOH (20/1) to amino)-2,3-dihydro-1H-xanthen-4-yl)vinyl)-1,3,3-trimethyl-3H-
afford a brown solid (0.09 g, 52%). 1H NMR (400 MHz, CDCl3): d indol-1-ium perchlorate) was chosen as a near-infrared uoro-
7.81–7.79 (d, J ¼ 8.0 Hz, 1H), 7.49–7.34 (m, 3H), 7.16–7.09 (m, phore to prepare near-infrared uorescent probes for lysosomal
3H), 6.81–6.77 (t, J ¼ 8.0 Hz, 1H), 6.57–6.55 (d, J ¼ 8.0 Hz, 1H), pH in living cells because of its advantageous photophysical
6.35–6.23 (m, 3H), 5.37–5.34 (d, J ¼ 12.0 Hz, 1H), 3.53–3.26 (m, properties including a large absorption extinction coefficient
4502 | J. Mater. Chem. B, 2014, 2, 4500–4508
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry B
(1.4 ꢁ 105 Mꢀ1 cmꢀ1), high uorescence quantum yield (41% in
methanol) with near-infrared emission peak at 720 nm, good
photostability and chemical stability.27 It displays absorption
maximum peak at 710 nm due to S0 / S1 transition, a shoulder
peak at 650 nm, and near-infrared emission peak at 731 nm in
ethanol mixed solution (please see Fig. S24 and S25 in ESI†).
Fluorescent dye (5) was prepared by condensation of 2-(4-
(diethylamino)-2-hydroxybenzoyl)benzoic acid (1) with cyclo-
hexanone (2) in acidic conditions, yielding 9-(2-carboxyphenyl)-
6-(diethylamino)-1,2,3,4-tetrahydroxanthylium perchlorate (3),
and followed by condensation of compound 3 with Fisher's
aldehyde (4) in acetic anhydride at 50 ꢂC (Scheme 2).27 Near-
infrared uorescent probes were readily synthesized from
uorescent dye 5 through one- or two-step procedures. Fluo-
rescent probe A was synthesized by treating uorescent dye 5
with phosphoryl chloride (POCl3), and followed by further
reaction with 1,2-diaminobenzene. Fluorescent probe B was
prepared by coupling uorescent dye 5 with dicyclohex-
ylcarbodiimide (DCC) and 4-dimethylaminopyridine (DMAP) in
dichloromethane for 30 minutes, and followed by further
reaction with hydrazine hydrate for one hour. Fluorescent probe
C was synthesized by coupling uorescent dye 5 with N-
hydroxysuccinimide in the presence of dicyclohexyl-
carbodiimide (DCC) for 30 minutes, and followed by further
reaction with propargylamine (Scheme 2). Fluorescent probe D
was prepared by reacting uorescent dye 5 with 1,10-carbon-
yldiimidazole (9) in dry dichloromethane for four hours, and
followed by further reaction with N-(2-hydroxyethyl)ethylenedi-
amine for overnight. Introduction of N-(2-hydroxyethyl)ethyl-
enediamine to uorescent dye 5 is expected to enhance
hydrophilic property of uorescent probe and facilitate selective
accumulation of uorescent probe D in lysosome of living cells
via protonation of the secondary amine in an acidic environ-
ment at pH 4.5 inside lysosome.
Fig. 1 Absorption spectra of 5 mM fluorescent probe A at different pH
values in 40 mM citrate–phosphate buffer solution containing 40%
ethanol (A) and effect of pH on absorbance of the fluorescent probe A
at 718 nm with three-repeated measurements (B).
accompanying with a shoulder peak at 663 nm (Fig. 1A and B),
indicating that the spirolactam ring of the uorophore was
opened, leading to signicant extension of p-conjugation of the
uorophore. Consequently, the solution of the uorescent
probe A displays distinct color changes from colorless to green
along with the pH titration from 6.9 to 4.0. However, further
decrease of pH below 2.0 causes decrease in the absorbance of
uorescent probe A at 720 nm (Fig. 1A). This may be due to
charge imbalance of the uorophore as the tertiary amine group
of the uorophore becomes protonated at extremely low pH
(Scheme 3). Fluorescent probe A displays full uorescent
reversible responses between pH 3.0 and 7.2 when it is treated
Optical responses of uorescent probes to pH
We investigated effect of pH on absorption spectra of the uo-
rescent probes. Fluorescent probe A displays a strong absorp-
tion peak at 379 nm, a moderate absorption at 479 nm, and an
extremely weak absorption peak at 720 nm in 40 mM citrate–
phosphate buffer solution containing 40% ethanol at pH 7.48
(Fig. 1A). Gradual decrease of pH from pH 7.0 to pH 4.0 results
in signicant absorbance enhancement at 720 nm
Scheme 3 Chemical structures of fluorescent probe A at different pH
Scheme 2 Synthetic routes to near-infrared fluorescent probe for pH. values.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. B, 2014, 2, 4500–4508 | 4503

View Article Online
Journal of Materials Chemistry B
Paper
with acid or base. The similar pH effect on absorbance of
uorescent probes B, C and D were also observed (Fig. 2, and
S17–S19 in ESI†).
In order to evaluate uorescent probes for pH sensing
application, we investigated pH effect on uorescence intensity
of 5 mM uorescent probes in 40 mM citrate–phosphate buffer
solution containing 40% ethanol. Fig. 3 displays the uores-
cence spectra of uorescent probe A at different pH values. The
probe was non-uorescent when the buffer pH is greater than
7.4. However, gradual decrease of pH from pH 7.4 to pH 4.0
results in appearance of a new uorescence peak at 743 nm, and
signicantly enhances uorescence peak intensity. There is
more than 71-fold increase in the uorescence intensity of
uorescent probe A at 743 nm with pH decrease from 7.4 to 4.1,
indicating the probe is very sensitive to acidic pH because of the
H+-induced spirolactam ring opening of the uorophore. In
addition, there are slightly red shis of the uorescence peak
with pH decrease from 7.4 to 4.1, which may be due to enhanced
p-conjugation of the uorescent probe A during these pH
changes. The pKcycl value of the probe A is 5.8 related to spi-
rolactam ring opening, which was obtained according to the
Henderson–Hasselbach-type mass action equation. The
Fig. 3 Fluorescent spectra of 5 mM fluorescent probe A at different pH
values in 40 mM citrate–phosphate buffer solution containing 40%
ethanol (A) and pH effect on fluorescence intensity of fluorescent
probe A at 743 nm with three repeated measurements (B).
uorescence intensity of the probe A displayed linear responses
to pH values in the range from 4.9 to 6.6. In addition, it showed
excellent reversible responses to pH between 4.0 and 7.4.
Compared with uorescent probe A, uorescent probes B, C and
D exhibit much more sensitive uorescent responses to pH with
395-, 592- and 229-fold increases in the uorescence intensity at
743 nm with pH decrease from 7.4 to 4.1, respectively. The pKcycl
values of the probe B, C and D related to the spirolactam ring
opening are 4.6, 4.9 and 5.4, respectively, which indicates that
the probes B, C and D are more suitable for lysosome imaging
application. However, further decrease of pH to strong acidic
conditions triggers signicant uorescence decreases of uo-
rescent probe A because the protonation of the nitrogen atom of
the uorophore signicantly reduces the electron donating
ability of the nitrogen atom, and results in charge imbalance
through the resonance structure (Scheme 3). The analysis of
uorescence intensity changes of uorescent probe A as a
function of pH by using the Henderson–Hasselbach-type mass
action equation yielded pKa value of 1.8. The similar pH effect
on uorescent probes B, C and D were also observed with
further decrease of pH to strong acidic conditions. The analysis
Fig. 2 Absorption spectra of 5 mM fluorescent probe D at different pH
values in 40 mM citrate–phosphate buffer solution containing 40%
ethanol (A) and effect of pH on absorbance of the fluorescent probe D
at 718 nm (B).
4504 | J. Mater. Chem. B, 2014, 2, 4500–4508
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry B
of uorescence intensity changes of uorescent probes B, C and
D gave almost the same pKa value of 1.7 (Fig. 4).
Effect of ethanol amount on uorescent responses of the
uorescent probes to pH
We studied effect of ethanol amount on uorescent responses
of the uorescent probes to pH. As the uorescent probes are
not completely soluble in aqueous solution, organic solvent
such as ethanol is needed to add to the buffer solution in order
to dissolve the probes. An increase of ethanol concentration in
the buffer solution enhances uorescence intensity of the
uorescent probe D as the probe solubility become improved
with an increase of ethanol concentration in 40 mM citrate–
phosphate buffer solution (Fig. 5). The similar results were
Fig. 5 Fluorescent spectra of 5 mM fluorescent probe D in 40 mM
observed in the uorescent probes A, B and C (Fig. S17, S21 and citrate–phosphate buffer solution at pH
S25 in ESI†).
4 containing different
concentrations of ethanol.
The selectivity experiments of uorescent probes to pH over
metal ions
uorescent probes with heavy, transition, and main group
metal ions (Fig. 6). Fluorescent probes A, B, C and D display no
responses to 200 mM alkali and alkaline-earth metal ions such
as Na+, K+, Ca2+ and Mg2+, as well as some transitional metal
ions (200 mM) such as Cu2+, Zn2+, Fe3+, Fe2+, Co2+, Ag+, Ba2+,
Cd2+, Pb2+, Ni2+ and Mn2+ at pH 7.0 and 4.1 (Fig. 6, and S27–S29
in ESI†), which indicates that the uorescent probes display
high selectivity to pH over these alkali, alkaline-earth metal
ions, and transitional metal ions.
We investigated effect of metal ions on uorescent response of
uorescent probes to pH by evaluating potential coordination of
Photostability of the uorescent probes
Fluorescent probes were excited continuously at their optimal
excitation for 5 min intervals and uorescence intensity was
measured every 5 min. The result in Fig. 7 clearly shows that the
uorescent probe D displays good photostability with its uo-
rescence decrease by 3.8% under one-hour excitation. The
uorescence intensity of uorescent probes A and C decrease by
5.2% and 10.6% under one-hour excitation (Fig. S18 and S26 in
ESI†). However, the uorescent probe B shows poor photo-
stability as its uorescence intensity decreased by 44% under
one-hour excitation (Fig. S22 in ESI†).
Fig. 4 Fluorescent spectra of 5 mM fluorescent probe D at different pH
values in 40 mM citrate–phosphate buffer solution containing 40% Fig. 6 Fluorescent responses of 5 mM fluorescent probe A to pH at 4.1
ethanol (A) and pH effect on fluorescence intensity of fluorescent and 7.0 in the absence and presence of different metal ions (200 mM),
probe D at 743 nm (B).
respectively.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. B, 2014, 2, 4500–4508 | 4505

View Article Online
Journal of Materials Chemistry B
Paper
Fig. 9 Fluorescence images of HUVEC-C cells incubated with fluo-
rescent probes B, C, and D. Cells were incubated with 20 mM of all
three dyes for 2 h and imaged for co-localization with 5 mM Lyso-
Sensor Green and Hoechst stains. The images were acquired using
confocal fluorescence microscope at 60ꢁ magnification.
Fig. 7 Photostability of 5 mM fluorescent probe D at pH 4.1 in 40%
ethanol solution. Sample was exposed under respective optimal
excitation wavelength and fluorescence intensities were measured at
5 min intervals.
these new probes. Images were captured for the same image
eld with different excitation and emission wavelengths of near-
infrared uorescent probes as compared to those of probe DND-
189, which enables simultaneous visualization of both probes
(near-infrared probes and DND-189) from the same intracellular
compartment. When the images were overlaid they showed co-
localization in lysosomes (Fig. 8–10). The probe D was most
uorescent and showed a very high signal even at 5 mM
concentration and was clearly co-localized to lysosomes
(Fig. 10). The uorescence intensity of the probe D is close to
probe DND-189 as the areas of low and high uorescence of the
probe match those of the probe DND-189, indicating the probe
D can effectively distinguish between different pH values in the
cell in a similar manner to the commercial probe DND-189.
Probe B showed a slightly weaker signal when compared to
probe D (Fig. 8 and 9). The close-up of co-localization of cells
image clearly shows that the lysosomes were mostly localized as
perinuclear clusters both in cancer and normal cell lines
(Fig. 10). These results are consistent with a previous report that
showed lysosomes in serum-starved cells relocate towards the
perinuclear position and forms clusters.28 In this study, the
probes were incubated in serum-free media thus the lysosomes
localize to perinuclear clusters. This data also conrms that the
Live cell imaging of uorescent probes
All four uorescent probes were used as near-infrared uores-
cent probes in cultured cells and compared to commercial
LysoSensor Green DND-189 and probe 5 to investigate if these
probes could be used to target lysosomes/acidic organelles
inside the cells as LysoSensor Green DND-189 is known to be
retained specically in acidic organelles. A series of experiments
were performed: a breast cancer line (MDA-MB-231) and normal
endothelial cell line (HUVEC) were loaded with uorescent
probes A, B, C and D and with LysoSensor Green DND-189
(Fig. 8 and 9), respectively. All the probes except probe A showed
a uorescence signal inside the cell, specically localized in
lysosomes with maximum signal with probe D followed by
probe B (Fig. 8–10). The uorescence signals of these probes
were compared with the commercially available and well char-
acterized lysosome specic probe, LysoSensor Green DND-189
(Lyso-green). To conrm the co-localization of these probes to
lysosomes, we incubated both the cells with Lyso-green and
Fig. 8 Fluorescence images of MDA-MB-231 cells incubated with
fluorescent probes B, C, and D. Cells were incubated with 20 mM of Fig. 10 Enlarged images of MDA-MB-231 and HUVEC-C cells with
dyes B, C, and D for 2 h and imaged for co-localization in presence of fluorescent probe D showing co-localization of fluorescent probe D in
5 mM LysoSensor Green, a lysosomal stain and Hoechst, a nuclear stain. lysosomes. The probe D shows much stronger signal in HUVEC cells
The images were acquired using confocal fluorescence microscope at compared to MDA-MB-231 cell. The lysosomes are mostly perinuclear
60ꢁ magnification.
in both cells due to serum-starvation for 2 hours.
4506 | J. Mater. Chem. B, 2014, 2, 4500–4508
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry B
probes are localized only in lysosomes and are highly sensitive
to pH environment. These uorescent probes B, C, and D were
compared with their precursor probe 5 (Scheme 2) and tested
for co-localization (Fig. 11) and cell toxicity (Fig. 12). Probe 5
shows non-specic distribution in cell and stains lysosomes as
well as other parts of the cell (Fig. 11). In addition, probe 5 is
very toxic and shows >30% cell toxicity for MDA-MB-231 cells at
5 mM concentration compared to nearly 100% viability observed
for cells in presence of probes B, C, and D at the same
concentration (Fig. 12).
Summary
We have prepared four near-infrared uorescent probes for pH
(A–D). The response mechanism of the uorescent probes to pH
value relies on the structural changes between spirocyclic and
ring-opening forms of the near-infrared uorophore. These
probes are not uorescent with a spirocyclic form at neutral pH
and highly uorescent with a spirocycle-opening at low pH
value of ꢃ4.5. Fluorescent probes have absorption and emission
peaks at 718 nm and 743 nm, respectively. The uorescent
probes B, C and D are cell-permeable and are capable of selec-
tive and sensitive labeling of lysosomes, and may offer potential
noninvasive monitoring of lysosomal pH changes during
physiological and pathological processes.
Conflict of interest
The authors declare no competing nancial interest.
Acknowledgements
This work was supported partially by National Science Foun-
dation (to H. Y. Liu), ALS Therapy Alliance (to A. Tiwari), and
Pruett postdoctoral fellowship support for J. Janjanam (to A.
Tiwari).
References
1 S. C. Burleigh, T. van de Laar, C. J. M. Stroop, W. M. J. van
Grunsven, N. O'Donoghue, P. M. Rudd and G. P. Davey,
BMC Biotechnol., 2011, 11, 95, DOI: 10.1186/1472-6750-11-95.
2 S. Humez, M. Monet, F. van Coppenolle, P. Delcourt and
N. Prevarskaya, Am. J. Physiol.: Cell Physiol., 2004, 287,
C1733–C1746.
Fig. 11 Fluorescence images of MDA-MB-231 cells incubated for 1 h
with 2.5, 5.0, and 10.0 mM concentrations of probe 5. Cells were
imaged for co-localization with LysoSensor Green and Hoechst stains
(A). Enlarged image (B) shows localization of probe 5 is not specific to
lysosomes. The images were acquired using confocal fluorescence
microscope at 60ꢁ magnification.
3 A. L. Edinger and C. B. Thompson, Curr. Opin. Cell Biol.,
2004, 16, 663–669.
4 A. C. Johansson, H. Appelqvist, C. Nilsson, K. Kagedal,
K. Roberg and K. Ollinger, Apoptosis, 2010, 15, 527–540.
5 E. S. Trombetta, M. Ebersold, W. Garrett, M. Pypaert and
I. Mellman, Science, 2003, 299, 1400–1403.
6 Y. Urano, D. Asanuma, Y. Hama, Y. Koyama, T. Barrett,
M. Kamiya, T. Nagano, T. Watanabe, A. Hasegawa,
P. L. Choyke and H. Kobayashi, Nat. Med., 2009, 15, 104–109.
7 B. Turk and V. Turk, J. Biol. Chem., 2009, 284, 21783–21787.
8 J. Stinchcombe, G. Bossi and G. M. Griffiths, Science, 2004,
305, 55–59.
9 E. J. Blott and G. M. Griffiths, Nat. Rev. Mol. Cell Biol., 2002, 3,
122–131.
10 S. He, R. P. Mason, S. Hunjan, V. D. Mehta, V. Arora,
R. Katipally, P. V. Kulkarni and P. P. Antich, Biorg. Med.
Chem., 1998, 6, 1631–1639.
Fig. 12 Cytotoxicity and cell proliferation effect of probes A, B, C, D, 11 Z. J. Diwu, C. S. Chen, C. L. Zhang, D. H. Klaubert and
and 5 (Scheme 2) were tested by MTS assay. The MDA-MB-231 cells
R. P. Haugland, Chem. Biol., 1999, 6, 411–418.
were incubated with 5, 25, and 50 mM concentrations of probes for
12 H. S. Lv, J. Liu, J. Zhao, B. X. Zhao and J. Y. Miao, Sens.
72 h and their viability was tested by adding MTS reagent and then
Actuators, B, 2013, 177, 956–963.
13 H. Zhu, J. L. Fan, Q. L. Xu, H. L. Li, J. Y. Wang, P. Gao and
reading absorbance at 490 nm for the colored formazon product
formed. The absorbance measured at 490 nm is directly proportional
to the number of living cells in the culture.
X. J. Peng, Chem. Commun., 2012, 48, 11766–11768.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. B, 2014, 2, 4500–4508 | 4507

View Article Online
Journal of Materials Chemistry B
Paper
14 Z. Li, Y. L. Song, Y. H. Yang, L. Yang, X. H. Huang, J. H. Han 22 J. L. Fan, C. Y. Lin, H. L. Li, P. Zhan, J. Y. Wang, S. Cui,
and S. F. Han, Chem. Sci., 2012, 3, 2941–2948.
15 L. Q. Ying and B. P. Branchaud, Bioorg. Med. Chem. Lett.,
2011, 21, 3546–3549.
M. M. Hu, G. H. Cheng and X. J. Peng, Dyes Pigm., 2013,
99, 620–626.
23 J. Y. Han and K. Burgess, Chem. Rev., 2010, 110, 2709–2728.
16 D. G. Smith, B. K. McMahon, R. Pal and D. Parker, Chem. 24 Z. Q. Hu, M. Li, M. D. Liu, W. M. Zhuang and G. K. Li, Dyes
Commun., 2012, 48, 8520–8522. Pigm., 2013, 96, 71–75.
17 L. J. Ma, W. G. Cao, J. L. Liu, D. Y. Deng, Y. Q. Wu, Y. H. Yan 25 S. L. Zhu, J. T. Zhang, J. Janjanam, J. H. Bi, G. Vegesna,
and L. T. Yang, Sens. Actuators, B, 2012, 169, 243–247.
18 F. Galindo, M. I. Burguete, L. Vigara, S. V. Luis, N. Kabir,
A. Tiwari, F. T. Luo, J. J. Wei and H. Y. Liu, Anal. Chim.
Acta, 2013, 758, 138–144.
J. Gavrilovic and D. A. Russell, Angew. Chem., Int. Ed., 2005, 26 X. Ding, J. Janjanam, A. Tiwari, M. Thompson and
44, 6504–6508.
P. A. Heiden, Macromol. Biosci., 2014, DOI: 10.1002/
mabi.201300569.
27 L. Yuan, W. Y. Lin, Y. T. Yang and H. Chen, J. Am. Chem. Soc.,
2012, 134, 1200–1211.
19 H. M. DePedro and P. Urayama, Anal. Biochem., 2009, 384,
359–361.
20 T. Hasegawa, Y. Kondo, Y. Koizumi, T. Sugiyama, A. Takeda,
S. Ito and F. Hamada, Biorg. Med. Chem., 2009, 17, 6015– 28 V. I. Korolchuk, S. Saiki, M. Lichtenberg, F. H. Siddiqi,
6019.
E. A. Roberts, S. Imarisio, L. Jahreiss, S. Sarkar, M. Futter,
F. M. Menzies, C. J. O'Kane, V. Deretic and
D. C. Rubinsztein, Nat. Cell Biol., 2011, 13, 453–460.
21 H. J. Lin, P. Herman, J. S. Kang and J. R. Lakowicz, Anal.
Biochem., 2001, 294, 118–125.
4508 | J. Mater. Chem. B, 2014, 2, 4500–4508
This journal is © The Royal Society of Chemistry 2014