Mild acidic-enhanced mitochondrial-targeting by a neutral thiophene based terpyridine molecule with large two-photon action cross-section

Article abstract of DOI:10.1016/j.dyepig.2016.12.052

Compared to cationic molecules, neutral mitochondrial targeting probes rarely exist. Herein, neutral thiophene based terpyridine (-β, -γ) organic probes have been synthesized and systematically characterized. The results from photophysical property invest

Full text of DOI:10.1016/j.dyepig.2016.12.052

Dyes and Pigments 139 (2017) 431e439
Contents lists available at ScienceDirect
Dyes and Pigments
journal homepage: www.elsevier.com/locate/dyepig
Mild acidic-enhanced mitochondrial-targeting by a neutral thiophene
based terpyridine molecule with large two-photon action cross-
section
,
b
,
1
,
,
,
Xiaohe Tian a
*, Yingzhong Zhu b 1, Mingzhu Zhang a b, Jingyun Tan b, Qiong Zhang b,
Xingyu Wang b, Jiaxiang Yang b, Hongping Zhou b, Jieying Wu b, Yupeng Tian b
, c, **
a School of Life Science, Anhui University, Hefei 230039, China
b Department of Chemistry, Key Laboratory of Functional Inorganic Material Chemistry of Anhui Province, Anhui University, Hefei 230039, China
c School of Life Science, Center for Stem Cell Research and Translational Medicine, Anhui University, Hefei 230039, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Compared to cationic molecules, neutral mitochondrial targeting probes rarely exist. Herein, neutral
Received 28 September 2016
Received in revised form
21 November 2016
Accepted 4 December 2016
Available online 23 December 2016
thiophene based terpyridine (-b, -g) organic probes have been synthesized and systematically charac-
terized. The results from photophysical property investigations indicate that all the compounds possess
considerable two-photon activities, and one of these probes possesses the largest two-photon action
cross section (Fdmax) value (OT1, 850 nm, ~370 GM) compared with previously reported mitochondrial
imaging agents. Importantly, such probe in mild acidic external environment can trigger significant
protonation, and subsequently enhance mitochondrial-labeling using around 10ꢀ8 M range concentra-
tions for days. This result may offer a novel strategy to design the probes for mitochondria.
© 2016 Elsevier Ltd. All rights reserved.
Keywords:
Mitochondria
Two-photon action cross-section
Neutral mitochondrial dyes
Thiophene based terpyridine
1. Introduction
commonly observed in one-photon (1P) technique, two-photon
(2P) mitochondrial fluorescent probes using longer exciting
Mitochondria are essential organelles found in most eukaryotic
cells [1,2]. As the ‘power house’ of cells, mitochondria are the key
place of synthesizing adenosine triphosphate (ATP) and supplying
energy to maintain normal cell function such as signalling and cell
proliferation [3e5]. It is also the major source of reaction oxygen
species (ROS) [6e8] which play crucial roles in apoptosis [9].
Dysfunction of mitochondria is highly related with neurodegener-
ative disorder including Alzheimer's disease [10] and Parkinson's
disease [11]. Subsequently, molecules that capable of imaging
mitochondria in living cells and tissues are useful. In particular, for
the purpose of avoiding photobleaching, endogenous auto-
wavelength and lower energy have drawn significantly attention
[12e16].
Traditionally, fluorescent probes including Rhodamine 123,
pyridinium, and triphenylphosphonium (TPP) salts [6,17] with
cationic moiety are essential for mitochondria targeting due to high
membrane mitochondrial potential (Djm) [18e20]. These brands of
probes have been successfully applied for imaging mitochondria in
live cell, as well as indicating various substrates in mitochondrial
matrix [21]. However, it is not avoidable that excess cationic sub-
stances highly influence the extracellular/intracellular charge bal-
ance and bring unwanted cytotoxicity, which is a drawback in this
research field. However, mitochondrial-specific probes without
cationic targeting moiety rarely exist. Therefore, the preparation of
neutral fluorescent probes may be another alternative attractive
approach. In addition, most of these dyes are not optimized for two-
photon imaging or possess relatively low two-photon (2P) action
cross-section (Fdmax), resulting in using relatively higher concen-
fluorescence and shallow penetration depth (<100
mm) that
* Corresponding author. School of Life Science, Anhui University, 230039 Hefei,
China.
** Corresponding author. Department of Chemistry, Key Laboratory of Functional
Inorganic Material Chemistry of Anhui Province, Anhui University, 230039 Hefei,
China.
tration working solution (100 nM - 10 mM) and stronger laser po-
wer, which often lead to fast cell apoptosis induced by photo-
toxicity, particularly in real-time or long-term monitoring [21].
E-mail addresses: xiaohe.t@ahu.edu.cn (X. Tian), yptian@ahu.edu.cn (Y. Tian).
These authors contributed equally to this work.
1
http://dx.doi.org/10.1016/j.dyepig.2016.12.052
0143-7208/© 2016 Elsevier Ltd. All rights reserved.

432
X. Tian et al. / Dyes and Pigments 139 (2017) 431e439
To tackle above issues, a series of neutral mitochondrial probes
2.3. Cell culture
based on thiophene and terpyridine are designed and synthesized
at present work. In these strategies, thiophene group is adopted to
serve as p-bridge to extend the length of conjugated chain as well
as enhance the electron density of molecule, due to the high
polarizability and rich-electronic properties of the sulfur atom
[22,23]. Diphenylamine, electron-donor group, connected with
terpyridine, electron-withdrawing group, by rich-electronic thio-
phene-bridge will provide good non-linear optical properties.
Moreover, the nitrogen positions in the terpyridine unit are located
HepG2, A549, and 3T3 cells were cultured in Dulbecco's modi-
fied Eagle's medium supplemented with penicillin/streptomycin
and 10% bovine calf-serum in a 5% CO2 incubator at 37 ꢁC. The four
compounds were dissolved in DMSO with concentration of 1 mM as
stock solution, and the commercial dyes were prepared as 1 mM
PBS solution and diluted to working concentration as protocol
required.
at
b
- and
g-positions, which can reduce steric hindrance and lead to
2.4. Synthetic routes
protonation even in a mild acidic pH value. Consequently, the
probes are inducing membrane-dependent cationic molecules and
leading to target mitochondria without introducing excess cationic
species. Herein, it is showed that a temporary mild acidic extra-
cellular environment can enhance diphenylamine thiophene
compounds OT1 label mitochondria in cellulo in a highly effective
manner without interfering cell viability. Since OT1 owns the
2.4.1. Preparation for 5-(diphenylamino)thiophene-2-carbaldehyde
100 mL DMSO and 0.45 g (2.3 mmol) 1,10-Phenanthroline
monohydrate were added to the three neck flask and stirred for
30 min under the N2 atmosphere. After that 0.46 g (2.4 mmol) CuI
and 5.00 g (36 mmol) K2CO3 were added and stirred for another
30 min. 2.8 g (11 mmol) diphenylamine, 1.90 g (10 mmol) 5-bromo-
2-thiophenecarbaldehyde and 18-C-6 were added to the flask and
reflexed for 48 h. After the reaction, the reaction solution was
dispersed into 1000 mL water. The brown solid was received by
filtering. The pure production was obtained by column chroma-
tography (silica gel, Vpetroleum ether/Vacetic ether ¼ 30/1). Bright yellow
solid, 1.25 g. Yield 45%.
largest 2P action cross-section (~370
FGM), it offers the possibility
for long-term (48 h) mitochondria monitoring in living cells with
low molar concentration (10 nM). Current study not only provides a
new idea to design small molecules with significant 2P brightness,
also offers neutral compounds for staining mitochondria in living
cells.
2. Experimental section
2.4.2. Preparation of 5-(bis(4-ethoxyphenyl)amino)thiophene-2-
carbaldehyde
2.1. Apparatus and general methods
The route for this compound was similar to that of 5-(diphe-
nylamino)thiophene-2-carbaldehyde. Yellow brown sold, 1.65 g.
Yield 45%.
Nuclear magnetic resonance spectra (1H and 13C) were obtained
using a Bruker Avance 400 spectrometer. MALDI-TOF mass spectra
were recorded using Bruker Autoflex III Smartbeam. Elemental
analyses were carried out on a Perkin-Elmer 240 analyzer. Melting
point was measured on FP62 instruments. The UVevisible ab-
sorption spectra of dilute solutions were recorded using a SHI-
MADZU UV-3600 spectrophotometer using a quartz cuvette with a
1 cm path length. One-photon fluorescence spectra were obtained
using a HITACHI F-7000 spectrofluorimeter equipped with a 450 W
Xe lamp. Time-resolved fluorescence measurements were recorded
on HORIBA HuoroMax-4P. Two-photon excited fluorescence was
measured using a SpectroPro300i and the pump laser beam came
from a modelocked Ti:sapphire laser system at the pulse duration
of 140 fs and a repetition rate of 80 MHz (Coherent Mira900-D). Cell
survival was assayed using the cell proliferation Kit I with the
absorbance at 492 nm detected using a Perkin-Elmer Victor plate
reader. Confocal fluorescence imaging was obtained with an LSM
710 (Zeiss) confocal laser-scanning microscope.
2.4.3. Preparation of T1
2.8 g (10 mmol) 5-(diphenylamino)-thiophene-2-carbaldehyde,
2.70 g (22 mmol) 3-Acetylpyridine, and 2.90 g (50 mmol) KOH were
added to 80 mL ethanol and stirred for 30 min after that aqueous
ammonia was added to the solution in three instalments. Then the
mixture was heated to 85 ꢁC for 4 h. The residual was obtained by
filtering after the reaction solution cooled to room-temperature.
The pure production was received by re-crystalized from ethanol.
Yellow powder, 1.95 g. Yield 40%. M.p. ¼ 173e175 ꢁC. 1H NMR
(DMSO-d6, 400 MHz, ppm)
d
¼ 9.46 (d, J ¼ 1.3 Hz, 2H), 8.67 (dd,
J ¼ 4.7, 1.2 Hz, 2H), 8.64 (d, J ¼ 8.1 Hz, 2H), 8.14 (s, 2H), 8.00 (d,
J ¼ 4.0 Hz, 1H), 7.55 (dd, J ¼ 7.9, 4.8 Hz, 2H), 7.39 (t, J ¼ 7.9 Hz, 4H),
7.17 (m, 6H), 6.75 (d, J ¼ 4.0 Hz, 1H). 13C NMR (DMSO-d6, 100 MHz,
ppm)
d
¼ 154.54, 153.96, 150.11, 148.13, 146.86, 143.50, 134.36,
133.73, 131.70, 129.70, 127.32, 124.15, 123.70, 123.21, 119.56, 114.57.
IR(KBr, cm-1): 3033(w), 3003(w), 1598(s), 1545(m), 1488(s),
1470(s), 1452(s), 1390(m), 1289(m), 1276(m), 1078(w), 988(w),
754(m), 698(m), 508(m). Anal. Calcd. for C31H22N4S(482.60): C,
75.15; H, 4.59; N, 11.61%. Found: C, 75.10; H, 4.58; N, 11.89%. MALDI-
TOF: m/z, cal: 482.16, found: 481.76 [Mþ].
2.2. Crystal analysis and theory calculation
The X-ray diffraction measurements were performed on a
Bruker SMART CCD area detector using graphite monochromated
Mo-Ka radiation (
collected in the variable
l
¼ 0.71069 Å) at 298(2) K. Intensity data were
2.4.4. Preparation of T2
u
-scan mode. The structures were solved
The route for this compound was similar to that of T1. Yellow
by direct methods and difference Fourier transformations. The non-
hydrogen atoms were refined anisotropically and hydrogen atoms
were introduced geometrically. Calculations were performed with
SHELXTL-97 program package.
The ground states for each molecule were calculated using the
density functional theory level with the B3LYP functional employ-
ing a 6-31G* basis set. No symmetry or internal coordinate con-
straints were applied during optimization. The absorption energies
were investigated by time-dependent density functional theory
(TD-DFT). All calculations were performed by the use of the Gauss
03 suite of programs.
powder. Yield 33%. M.p. ¼ 225e227 ꢁC. 1H NMR (DMSO-d6,
400 MHz, ppm)
d
¼ 8.74 (d, J ¼ 4.9 Hz, 4H), 8.26 (d, J ¼ 4.9 Hz, 4H),
8.22 (s, 2H), 8.02 (d, J ¼ 3.9 Hz, 1H), 7.40 (t, J ¼ 7.8 Hz, 4H), 7.18 (dd,
J ¼ 18.1, 7.6 Hz, 6H), 6.74 (d, J ¼ 3.9 Hz, 1H). 13C NMR (DMSO-d6,
100 MHz, ppm)
d
¼ 154.3, 154.2, 150.4, 150.3, 146.8, 145.0, 143.8,
131.0, 129.7, 127.7, 124.3, 123.3, 121.0, 119.2, 115.8. IR(KBr, cm-1):
3053(w), 3030(w), 1593(s), 1541(m), 1493(m), 1467(s), 1424(m),
1294(m), 1221(m), 1061(w), 993(w), 754(m), 694(m), 511(m). Anal.
Calcd. for C31H22N4S(482.60): C, 75.15; H, 4.59; N, 11.61%. Found:
C, 75.22; H, 4.67; N, 11.61%. MALDI-TOF: m/z, cal: 482.16, found:
482.05 [Mþ].

X. Tian et al. / Dyes and Pigments 139 (2017) 431e439
433
2.4.5. Preparation of OT1
electron delocalization in the
p
conjugated system which may
The route for this compound was similar to that of T1. Yellow
benefit for two-photon absorption properties.
powder. Yield 37%. M.p. ¼ 156e158 ꢁC. 1H NMR (DMSO-d6,
400 MHz, ppm)
d
¼ 9.43 (d, J ¼ 1.3 Hz, 2H), 8.66 (d, J ¼ 3.4 Hz, 2H),
3.3. Photophysical properties and DFT calculation
8.61 (s, 2H), 8.04 (s, 2H), 7.91 (d, J ¼ 4.1 Hz, 1H), 7.54 (dd, J ¼ 7.9,
4.8 Hz, 2H), 7.19 (d, J ¼ 8.9 Hz, 4H), 6.96 (d, J ¼ 8.9 Hz, 4H), 6.23 (d,
J ¼ 4.1 Hz, 1H), 4.03 (q, J ¼ 6.9 Hz, 4H), 1.33 (t, J ¼ 6.9 Hz, 6H). 13C
The absorption and emission spectra of the four compounds in
six different solvents have been shown in Supporting information
(Fig. S2eS4), and the corresponding data are listed in Table 1.
Overlay of UVevis absorption and fluorescence emission spectra of
the four compounds in DMF are profiled in Fig. 2.
NMR (DMSO-d6, 100 MHz, ppm)
d
¼ 155.9, 154.2,149.9, 148.0, 139.8,
134.4, 133.8, 126.0, 123.5, 115.4, 114.0, 63.2, 14.7. IR(KBr, cm-1):
3055(w), 3037(w), 2979(m), 2930(w), 2898(w), 2870(w),
1601(m), 1545(m), 1504(s), 1469(s), 1471(m), 1239(s), 1044(w),
530(w). Anal. Calcd. for C35H30N4O2S (482.60): C, 75.15; H, 4.59;
N, 11.61%. Found: C, 75.85; H, 4.10; N, 11.73%. MALDI-TOF: m/z, cal:
570.21, found: 570.56 [Mþ].
As depicted in Fig. S2, all compounds show two intense ab-
sorption bonds at about 300 nm and 410 nm. The bands around
300 nm should be assigned to
p -
p* transition of diphenylamine
thiophene group, while the bands at 400e420 nm are attributed to
intramolecular charge transfer (ICT) of the molecule. In addition,
solvent polarity has slight effects on their linear absorption. With
increasing solvent polarity, the maximum absorption wavelengths
are 396 4 nm, 404 5 nm, 414 4 nm, and 420 5 nm, as a result
of no significant permanent ground-state dipole moment in polar
solvents. By contrast, the substitutes (ethyloxyl) have obvious in-
fluence on their absorption properties (Fig. 2). Compared with T
class, OT presents more than 10 nm bathochromic-shift. This
should be explained as follows. Introducing ethyoxyl unit can in-
crease the electron donating ability of terminal group and extend
the conjugated system, resulting in lowering the energy of the
entire molecule.
B3LYP exchange functional employing 6-31G* basis set using a
suit of Gaussian 03 programs were performed. The frontier mo-
lecular orbital plots are shown in Fig. 3 and corresponding data of
the four compounds are presented in Table S4. The four compounds
display similar feature in absorption properties. As shown in Fig. 3,
set T1, as an example, the calculated bands appear at 303 nm with
oscillator strengths, f ¼ 0.0374, which is closely resemble with the
experimental data. This band originates from the HOMO to
2.4.6. Preparation of OT2
The route for this compound was similar to that of T1. Yellow
powder, Yield 31%. M.p. ¼ 176 ꢁC. 1H NMR (DMSO-d6, 400 MHz,
ppm)
d
¼ 8.73 (d, J ¼ 3.8 Hz, 4H), 8.25 (d, J ¼ 5.9 Hz, 4H), 8.13 (s, 2H),
7.93 (d, J ¼ 4.1 Hz, 1H), 7.21 (d, J ¼ 8.9 Hz, 4H), 6.96 (d, J ¼ 8.9 Hz,
4H), 6.32 (d, J ¼ 4.1 Hz,1H), 4.03 (q, J ¼ 6.9 Hz, 4H),1.34 (t, J ¼ 7.0 Hz,
6H). 13C NMR (DMSO-d6, 100 MHz, ppm)
d
¼ 157.5, 156.0, 154.1,
150.3, 145.2, 144.2, 139.7, 128.2, 126.2, 121.0, 115.5, 115.2, 112.3, 63.2,
14.7. IR(KBr, cm-1): 3063(w), 3035(w), 2975(w), 2926(w), 2884(w),
1592(m), 1541(w), 1506(m), 1464(s), 1243(m), 1176(w), 1044(w),
825(m), 533(w). Anal. Calcd. for C35H30N4O2S (482.60): C, 75.15;
H, 4.59; N, 11.61%. Found: C, 75.32; H, 4.51; N, 11.97%. MALDI-TOF:
m/z, cal: 570.21, found: 570.88 [Mþ].
3. Results and discussion
3.1. Synthesis
The synthetic routes and molecular structures of the four
compounds are shown in Scheme 1. The precursor diphenylamine
thiophene formaldehyde was prepared by modified Ullman reac-
tion using CuI and phenanthroline as catalyst. Then an effective
method was chosen to produce crude target product. Pure com-
pounds as yellow powder were obtained by recrystallization from
ethanol. Details of their synthetic routes, corresponding crystal
data are described in the supporting information.
LUMO þ4 which contains electronic charge transfer not only from
* of diphenylamine moiety but also from the orbit of thiophene
* orbit of diphenylamine. The S atom of thiophene ring bearing
high polarizability and electron-rich properties is responsible for
the transfer from the orbit of thiophene to * orbit of diphenyl-
p
-
p
p
to
p
p
p
amine. Meanwhile, the calculated absorption band in low energy
region locate at about 400e460 nm. It is clearly shown in Fig. 3 and
Table S3, these low energy bonds are dominated by ICT transition:
3.2. Structural character description
the electronic charge transfer from the
p orbit of diphenylamine
group to the * orbit of terpyridine group. In addition, the theo-
p
Single crystals of these four compounds, suitable for single
crystal X-ray diffraction analysis, were obtained by slow evapora-
tion from dichloromethane/methanol solution at room tempera-
ture after five days. The crystal structures of the four compounds
are presented in Fig. 1, and corresponding data are listed in Table S1.
Single-crystal structure of OT1, as a representative, belongs to
retical calculated data indicate the oxethyl unit extends the con-
jugated system within the molecule which will cause the
bathochromic shift.
The emission band between 400 and 620 nm of the four com-
pounds results mainly from the absorptive transition from the
ground state S0 to the S1 state, which undergoes an efficient
vibrational relaxation upon 1P excitation. In contrast to the ab-
sorption spectra, a remarkable red shift in emission spectra with
increasing polarity is shown in Fig. S3. When increasing the polarity
of the solvents, the charge separation increases in the excited state,
resulting in a larger dipole moment than that of ground state,
which also means the stabilization effect of a highly polar solvent
triclinic system with Pꢀ space group, whereas the other three
ı
compounds crystalize in monoclinic system with P21/c or P21/n
space group. With regard to the diphenylamine thiophene moiety
of the four compounds, the central N-atom adopts sp2 hybridiza-
tion to form a trigonal planar which can introduce its lone pair
delocalize to the propeller conjugated system constituted by three
aryl-rings. Compared with reported triphenylamine terpyridine
[24], the compounds in this paper show significant advantages as
follows. The dihedral angles between the bridge aryl-ring and
central pyridine are 7.9ꢁ, 10.1ꢁ, 2.3ꢁ, and 10.6ꢁ (shown in Table S3),
respectively, much smaller than 32.7ꢁ in triphenylamine terpyr-
idine. In addition, all bond lengths in the four compounds,
including CeC, CeN, and CeS bands, are located in the range of the
normal single and double bond (Table S2). All the crystal data in-
dicates that thiophene-terpyridine bear better co-planarity and
on the excited state is stronger. Compared with g-terpyridine, b-
terpyridine shows blue shift, which is agreed with absorption
spectra and large quantum yield, such as T1 bearing the largest
value of quantum yield in DMF (31.7%). Meanwhile, ethylox group
can decrease intensity of the emission band and broad the
bandwidth.
The Lippert-Magata equation is used to evaluate the transition
dipole, the changes in the dipole moment between ground and
excited states
(me-mg), and corresponding permanent dipole

434
X. Tian et al. / Dyes and Pigments 139 (2017) 431e439
Scheme 1. The synthetic routes of compounds T1, T2, OT1, and OT2.
Fig. 1. Crystal structures of the compounds (T1, T2, OT1 and OT2). Hydrogen atoms are omitted for clarity.
moment [25]. Due to the specific solvent-solute interactions (site-
specific hydrogen-bonding) between compounds and protic sol-
vents (ethanol), a poor linear relationship between the Stokes shift
and the reaction field parameter (ε, n) is found [26], indicating that
the four thiophene-based terpyridine compounds are easier pro-
tonation with Hþ at a mild pH and the site of interaction may occur
at the nitrogen atom in the pyridine rings.

X. Tian et al. / Dyes and Pigments 139 (2017) 431e439
435
Table 1
Linear photophysical properties of T1, T2, OT1, and OT2 in different solvents.
lmaxa/εb
lmax
Fd
Stokes shifte
t
/nsf
c
Compounds
Solvents
T1
Benzene
301(3.46) 397(3.35)
301(3.36) 400(3.34)
301(3.47) 397(3.43)
304(3.42) 398(3.27)
302(3.28) 401(3.50)
300(3.46) 393(3.44)
296(2.35) 403(2.21)
297(3.15) 407(2.89)
296(3.29) 403(3.13)
295(3.26) 405(2.82)
298(3.41) 408(3.18)
294(3.58) 399(3.16)
303(3.66) 412(4.34)
301(3.93) 415(4.58)
300(3.47) 414(4.29)
300(3.87) 413(4.59)
299(4.16) 418(5.09)
298(4.00) 410(4.69)
301(3.50) 419(3.79)
302(3.29) 422(3.28)
300(3.19) 421(3.66)
298(3.52) 416(3.76)
301(3.22) 425(3.66)
301(3.23) 420(3.36)
459
479
471
490
495
468
465
493
485
510
515
478
495
523
515
528
533
510
503
532
522
520
539
546
29.4
28.5
30.2
22.5
23.1
31.7
35.5
11.6
12.7
4.8
62
79
74
92
94
75
62
86
2.70
3.19
3.00
3.07
3.16
3.04
2.60
2.18
2.09
1.93
0.48
1.88
3.67
4.00
3.96
3.14
2.00
3.72
3.72
3.30
3.50
3.57
0.22
1.85
Dichloromethane
Tetrahydrofuran
Ethyl acetate
Ethanol
Dimethylformamide
Benzene
Dichloromethane
Tetrahydrofuran
Ethyl acetate
Ethanol
Dimethylformamide
Benzene
Dichloromethane
Tetrahydrofuran
Ethyl acetate
Ethanol
Dimethylformamide
Benzene
Dichloromethane
Tetrahydrofuran
Ethyl acetate
Ethanol
T2
82
105
107
79
0.9
18.1
15.4
8.0
12.2
4.6
OT1
OT2
83
108
101
115
115
100
84
110
101
104
114
126
2.4
11.4
14.9
4.9
7.0
9.1
0.2
1.6
Dimethylformamide
a
Peak position of the largest absorption band in nm (1.0 ꢂ 10ꢀ5 mol/L).
Maximum molar absorbance in 104 L/mol$cm.
Peak position of SPEF, exited at the absorption maximum.
Quantum yields determined by using quinine sulfate as standard (%).
Stokes shift in nm.
b
c
d
e
f
The fitted fluorescence lifetime.
Fig. 4. The maximum values of the four compounds are appeared at
about 850 nm with 197 GM, 80 GM, 371 GM, and 74 GM,
F
F
F
F
respectively (Table 2). It is noteworthy that the 2P action cross-
section value of OT1 is ~17-folds greater than those commercial
mitotracker and 2.12 folds greater than the FMT-green that previ-
ously own the largest Fdmax value for mitochondrial targeting
[27,28] (Table 2), that potentially useful for two-photon microscopy
live cell imaging [29].
As shown in absorption spectra (Fig. S8), the absorption peak in
low energy range will decrease with decreasing the pH value,
indicating that OT1 is easy protonated even in a mild acidic pH
value. In order to continue in-depth study of potential application
in bioimaging, the one- and two-photon excited fluorescence
spectra in different pH solutions (Vwater/Vdimethylformamide ¼ 1/1) are
carried out (Fig. S9). The 1P excited fluorescence intensity (Fig. S9a)
and 2P action cross-section (Fig. S9b) are both increase first and
then decrease with decreasing the pH value from 7 to 3. These ef-
fects are mainly caused by the protonation of terpyridine moiety. In
high acidic pH solution, the compound is partially protonated
which will enhance intramolecular charge transfer (ICT) accom-
panied by an enhancement in fluorescence spectra [29,30]. How-
ever, when pH values are decreasing, intramolecular electron
transfer will alter ICT which always lead to fluorescence quenching.
Hence, the largest 2P action cross-section value of OT1 appears at
pH ¼ 6 (~410 GM), and then decreases to ~50 GM at pH ¼ 3.
Fig. 2. Overlay of UVevis absorption spectra (solid line) and fluorescence emission
spectra (dotted line) of compounds T1, T2, OT1, and OT2 in DMF.
3.4. Two-photon absorption characters
The two-photon absorption (2 PA) spectra of the four com-
pounds were recorded in DMF from 700 nm to 950 nm. The two-
photon excited fluorescent (2PEF) spectra obtained by using
optimal excitation wavelength are shown in Fig. S5. The experi-
ments confirm that the absorption in the region of 700e950 nm is
2 PA process (Fig. S6). The 2P action cross-sections were obtained
by 2PEF spectra of four compounds with fluorescein as reference.
Fig. S7 clearly displays that the bathochromic shift occurred
from T to OT series with more than 40 nm shift and OT series has a
wide width of emission with half peak width 60 nm, which pre-
sents highly accordance with their 1P emission spectra. The 2P
action cross-sections (Fd) spectra of four compounds are shown in
3.5. Cell uptake at mild-acidic condition and cell entry mechanism
All these four compounds are effectively taken up within living
HepG2 cells (Fig. S10), whereas OT1 is selected for further biological
assessment since it possesses the largest two-photon action cross
section (Fd). The standard MTT assay (Fig. S11) was first evaluated
and showed low toxic towards HepG2 cell lines under microscopic
working concentration (10 nM-1 mM). Due to the easy protonation

436
X. Tian et al. / Dyes and Pigments 139 (2017) 431e439
Fig. 4. Two-photon action cross-section of T1, T2, OT1, and OT2 in DMF solutions from
700 nm to 950 nm.
Table 2
Two-photon related photophysical properties of T1, T2, OT1, OT2, FMT-green, and
MTR.
ex
em
Compounds
l
(nm)
l
(nm)
Fd2pa (GM)
2PEF
2PEF
T1
T2
OT1
OT2
FMT-green [27]
MTR [28]
840
850
850
860
750
940
509
527
556
582
540
642
197
80
371
74
175
21
of terpyridine moiety at mild acidic pH value (Fig. S8), it is expected
that the accumulation of neutral OT1 in the mitochondria depends
on the mitochondrial membrane potential (Fig. S12), Djm, which
can be enhanced after protonating (pH ¼ 6.5). Therefore, OT1
(1
mM) was incubated with HepG2 cells in pH ¼ 6.5 for 5 min,
neutral incubation (pH ¼ 7.0) and mild basic condition (pH ¼ 7.4).
Cells were washed with buffer solution and regained the neutral
culture medium, then imaged under two-photon confocal micro-
scopy (lEx ¼ 850, lEm ¼ 500e550 nm) without fixation. Fig. 5a
shows that at both neutral and pre-mild acidic medium treated
cells are clearly internalized with OT1 whereas under the basic
condition (pH ¼ 7.4) cells present weaker uptake; the Differential
Interference Contrast (DIC) micrographs displays good cell shape.
The fluorescence activated cell-sorting (FACS) experiments have
quantitatively suggested that the mild acidic (pH ¼ 6.5) pre-treated
cells accelerated the entry of OT1, since the fluorescent intensity is
apparently higher than neutral medium cultured cells (Fig. 5b). The
real-time imaging (Fig. 5c) performed with mini-incubator (5% CO2/
37 ꢁC) that is attached with the confocal suggests effective uptake
(<3 min, pH ¼ 6.5) of OT1 by living cells. The possible cell uptake
mechanism (Fig. 5d) of OT1 was also evaluated using different cell
entry inhibitor [31,32] including general endocytosis, receptor-
mediated endocytosis and active transport, as well as lower tem-
perature (inhibitor energy-dependent cell entry) assessment
(25 ꢁC, 4 ꢁC). Apparently, the cell entry shows almost no alternation
under all inhibitors and at lower temperature, suggesting that the
small organic molecule OT1 is highly membrane-permeable and is
not energy dependent, and the possible uptake mechanism is
membrane diffusion (e.g. free diffusion, passive diffusion). Given
the credit to its low toxicity, fast cell entry and large Fd, this offers

X. Tian et al. / Dyes and Pigments 139 (2017) 431e439
437
Fig. 5. (a) Confocal micrograph of OT1 internalized with HepG2 cells at mild acidic, neutral and mild basic pre-treatment and (b) correspondent cell uptake intensity tested by flow
cytometry (FACS) (c) Real-time uptake of OT1 in HepG2 cells at pH ¼ 6.5 over 180 s (d) cell entry studies of OT1 under inhibitors' treatment and low-temperature incubation. Scale
bar ¼ 10
mm.
the possibility to apply such probe for long-term cell monitoring
under confocal microscopy.
and negative PI signal (Fig. 6c). Furthermore, the stained cells are
continued to be scanned under 850 nm laser, suggesting that OT1
displays relatively better photostability than commercial dye MDR
after 150 s irradiation (30 scans) (Fig. 6d) which also confirmed by
photogregradation test in vitro (Fig. S14).
3.6. Colocolization study and photon-resistance
It is noteworthy that although both neutral and acidic medium
treated cells display cell uptake and locat within intracellular
cytosolic region, cells incubated within acidic medium clearly show
tubular-like structure while the neutral group shows less specificity
(Fig. 6a and c). To confirm the precise subcellular compartment that
OT1 stained under potentization, the colocolization study under
higher magnification (X200) was performed. HepG2 cells were
3.7. Long-term mitochondrial-staining under low molar
concentration
To further explore the potential practical applications using OT1
for mitochondrial targeting under two-photon microscopy, inter-
nalization by human embryo liver fibroblast (HELF) cells and
incubated with 1
mM OT1 (5 min, pH ¼ 6.5) then co-stained with
A549 cell lines was also performed (1
mM, 5 min, pH ¼ 6.5), as
Mitotracker Deep-Red (MDR) and LysoTracker® Red DND-99
(LysoTracker), commercial dyes that specifically mark mitochon-
dria and lysosome in living cells, respectively. The higher resolution
micrographs clearly show that OT1 displays irregular fibroid
structures throughout the cytosolic space (see also 3D imaging,
Fig. 6b), and highly co-localizes with the signal from MDR, sug-
gesting that OT1 has the affinity with intracellular mitochondria at
mild-acidic condition (also refer to Fig. S10 for T1, T2 and OT2).
However, when incubated with LysoTracker, the profile presents a
negative Pearson coefficient Rr ¼ ꢀ0.2122, suggesting that no acidic
lysosomal trafficking is involved (Fig. S13). Live/Dead cell imaging
assay using Syto-select (for living cell) and propidium iodide (PI, for
dead cell) demonstrates the short mild-acidic window (5-min) will
not alter the cell viability, as indicated by the positive Syto signal
representatives for normal cells and cancerous cells, respectively.
Fig. 7a shows that both normal cells and cancer cells successfully
take up OT1 and display fibroid mitochondrial staining (also refer to
Fig. S15eS16 for T1, T2 and OT2). It is also clearly observed that the
auto-fluorescence free two-photon channel possesses higher
signal-to-noise ratio compare to 1P excited channel (lEx ¼ 850 nm,
lEm ¼ 500e550 nm). Owning to its standout Fd, it offers significant
brightness under two-photon confocal microscopy. Therefore, we
were motivated to perform mitochondria staining in living cell
using low molar concentration OT1. To out delight, even at 10 nM
(5 min, pH ¼ 6.5 pre-treated), legible mitochondrial morphology is
still observed (Fig. 7c), suggesting that OT1 at nano-molar range is
sufficient to label living cell mitochondria effectively. To investigate
the influence of experimental parameters on concentration-

438
X. Tian et al. / Dyes and Pigments 139 (2017) 431e439
Fig. 6. (a) Confocal micrograph of OT1 co-localized with commercial dyes Mito Deep Red; (b) the 3D confocal micrograph of OT1; (c) Live/Dead cell imaging assay using Syto-select
(for living cell) and propidium iodide (PI for dead cell) (d) the photostability of OT1 compared with commercial dyes Mito Deep Red. Scale bar ¼ 10 m.
m
Fig. 7. The confocal micrograph of OT1 on Helf cells (a) and A549 cells (b); (c) The A549 cells incubated with different concentrations of OT1; (d) The A549 cells incubated with OT1
for 24 h and 48 h. Scale bar ¼ 10 m.
m
depended cell staining, confocal fluorescence imaging of OT1 at
M and 10 nM under the same experimental parameters have
been carried out (Fig. S17a). When incubated with 1 M OT1, bright
photon microscopy at such diluted condition is indeed a significant
improvement compared with previous reported probes, which
normally use 1e4 orders of magnitudes higher concentration. The
above natures of OT1 offer a noninvasiveness (low cytotoxicity and
photon toxicity) and organic-solvent free (<0.001% DMSO) manner
for long-term mitochondria trafficking under two-photon
microscopy.
1
m
m
fluorescence emission can be observed from the mitochondria of
the cell (normalized fluorescence intensity value 1116). Likewise,
after being cultured with 10 nM OT1 for 30 min, the mitochondria
still can be well stained (normalized fluorescence intensity value
316). In addition, OT1 presents a good photostability under this
experimental condition (Fig. S17b). Since OT1 possessed almost no
toxicity at low concentration (MTT assay, Fig. S11), the mitochon-
drial fluorescence still remains after 48 h and no obvious fading is
observed (Fig. 7d). To visualize living cell mitochondria under two-
4. Conclusions
In summary, four thiophene based terpyridine (-
b, -g) organic
probes with considerable two-photon activity have been designed,

X. Tian et al. / Dyes and Pigments 139 (2017) 431e439
439
complex I deficiency in Parkinson's disease. J Neurochem 1990;54:823e7.
synthesized and fully characterized. The nonlinear optical proper-
ties studies demonstrated that OT1 possess the largest 2P action
cross-section (~370 GM), compare to reported mitochondria two-
photon probes. We demonstrated a short mild-acidic window
would trigger protonation of OT1, consequently resulting effective
mitochondrial membrane potential targeting. Own to its large 2P
action cross-section (~370 GM) and low working concentration
(10 nM). The staining for intracellular mitochondria remained
stable without interfering the cell viability up for 48 h. Current
results have provided a new idea for designing neutral mitochon-
drial probes with significant two-photon brightness.
[12] Xu Q, Heo CH, Kim G, Lee HW, Kim HM, Yoon J. Development of imidazoline-
2-thiones based two-photon fluorescence probes for imaging hypochlorite
generation in a Co-Culture system. Angew Chem Int Ed 2015;127:4972e6.
[13] Sarkar AR, Kang DE, Kim HM, Cho BR. Two-photon fluorescent probes for
metal ions in live tissues. Inorg Chem 2013;53:1794e803.
[14] Zhou L, Zhang X, Wang Q, Lv Y, Mao G, Luo A, et al. Molecular engineering of a
TBET-based two-photon fluorescent probe for ratiometric imaging of living
cells and tissues. J Am Chem Soc 2014;136:9838e41.
[15] Kim D, Ryu HG, Ahn KH. Recent development of two-photon fluorescent
probes for bioimaging. Org Biomol Chem 2014;12:4550e66.
[16] Niu W, Guo L, Li Y, Shuang S, Dong C, Wong MS. Highly selective two-photon
fluorescent probe for ratiometric sensing and imaging cysteine in mitochon-
dria. Anal Chem 2016;88:1908e14.
[17] Yousif LF, Stewart KM, Kelley SO. Targeting mitochondria with organelle-
specific compounds: strategies and applications. ChemBioChem 2009;10:
1939e50.
Acknowledgments
[18] Zhao N, Chen S, Hong Y, Tang BZ. A red emitting mitochondria-targeted AIE
probe as an indicator for membrane potential and mouse sperm activity.
Chem Commun 2015;51:13599e602.
This work was supported by the National Natural Science
Foundation of China (21602003, 21271004, 21501001, 51372003,
51432001, 51472002), Focus on returned overseas scholar of Min-
istry of Education of China, the Higher Education Revitalization
Plan Talent Project (2013). Anhui University Doctor Startup Fund
(J01001962).
[19] Liu Z, Zhang C, Chen Y, He W, Guo Z. An excitation ratiometric Zn2þ sensor
with mitochondria-targetability for monitoring of mitochondrial Zn2þ release
upon different stimulations. Chem Commun 2012;48:8365e7.
[20] Wu M-Y, Li K, Liu Y-H, Yu K-K, Xie Y-M, Zhou X-D, et al. Mitochondria-tar-
geted ratiometric fluorescent probe for real time monitoring of pH in living
cells. Biomaterials 2015;53:669e78.
[21] Xu Z, Xu L. Fluorescent probes for the selective detection of chemical species
inside mitochondria. Chem Commun 2016;52:1094e119.
[22] Mishra A, Ma C-Q, Bauerle P. Functional oligothiophenes: molecular design for
multidimensional nanoarchitectures and their applications. Chem Rev
2009;109:1141e276.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.dyepig.2016.12.052
[23] Ellinger S, Graham KR, Shi P, Farley RT, Steckler TT, Brookins RN, et al.
Donoreacceptoredonor-based
p-conjugated oligomers for nonlinear optics
and near-IR emission. Chem Mater 2011;23:3805e17.
[24] Liu B, Zhang Q, Ding H, Hu G, Du Y, Wang C, et al. Synthesis, crystal structures
and two-photon absorption properties of a series of terpyridine-based chro-
mophores. Dyes Pigments 2012;95:149e60.
[25] Ji S, Yang J, Yang Q, Liu S, Chen M, Zhao J. Tuning the intramolecular charge
transfer of alkynylpyrenes: effect on photophysical properties and its appli-
cation in design of OFF-ON fluorescent thiol probes. J Org Chem 2009;74:
4855e65.
[26] Zhao GJ, Liu JY, Zhou LC, Han KL. Site-selective photoinduced electron transfer
from alcoholic solvents to the chromophore facilitated by hydrogen bonding:
a new fluorescence quenching mechanism. J Phys Chem B 2007;111:8940e5.
[27] Han JH, Park SK, Lim CS, Park MK, Kim HJ, Kim HM. Simultaneous imaging of
mitochondria and lysosomes by using two-photon fluorescent probes. Chem
Eur J 2012;18:15246e9.
[28] Haugland RP. A guide to fluorescent probes and labeling technologies. Invit
Carlsbad 2005:11e37.
[29] Kim HM, Cho BR. Small-molecule two-photon probes for bioimaging appli-
cations. Chem Rev 2015;115:5014e55.
[30] Sumalekshmy S, Henary MM, Siegel N, Lawson PDV, Wu Y, Schmidt K, et al.
Design of emission ratiometric metal-ion sensors with enhanced two-photon
cross section and brightness. J Am Chem Soc 2007;129:11888e9.
[31] Ziegler HK, Unanue ER. Decrease in macrophage antigen catabolism caused by
ammonia and chloroquine is associated with inhibition of antigen presenta-
tion to T cells. P Nat Acad Sci U. S. A 1982;79:175e8.
References
[1] Chan DC. Mitochondria: dynamic organelles in disease, aging, and develop-
ment. Cell 2006;125:1241e52.
[2] Alberts B, Bray D, Hopkin K, Johnson A, Lewis J, Raff M, et al. Essential cell
biology: Garland science. 2013.
[3] Leist M, Single B, Castoldi AF, Kühnle S, Nicotera P. Intracellular adenosine
triphosphate (ATP) concentration: a switch in the decision between apoptosis
and necrosis. J Exp Med 1997;185:1481e6.
[4] Vander Heiden MG, Cantley LC, Thompson CB. Understanding the Warburg
effect: the metabolic requirements of cell proliferation. Science 2009;324:
1029e33.
[5] Cadenas E. Mitochondrial free radical production and cell signaling. Mol Asp
Med 2004;25:17e26.
[6] Dickinson BC, Srikun D, Chang CJ. Mitochondrial-targeted fluorescent probes
for reactive oxygen species. Curr Opin Chem Biol 2010;14:50e6.
[7] Murphy MP, Smith RA. Drug delivery to mitochondria: the key to mitochon-
drial medicine. Adv Drug Deliv Rev 2000;41:235e50.
[8] Murphy MP. How mitochondria produce reactive oxygen species. Biochem J
2009;417:1e13.
[9] Green DR, Reed JC. Mitochondria and apoptosis. Science 1998;281:1309.
[10] Santos RX, Correia SC, Wang X, Perry G, Smith MA, Moreira PI, et al.
A
synergistic dysfunction of mitochondrial fission/fusion dynamics and
[32] Puckett CA, Barton JK. Mechanism of cellular uptake of a ruthenium poly-
pyridyl complex. Biochemistry 2008;47:11711e6.
mitophagy in Alzheimer's disease. J Alzheimers Dis 2010;20:401e12.
[11] Schapira A, Cooper J, Dexter D, Clark J, Jenner P, Marsden C. Mitochondrial