Dual-color imaging of cytosolic and mitochondrial zinc ions in live tissues with two-photon fluorescent probes

Article abstract of DOI:10.1039/c4ob00101j

We report two-photon probes for Zn²⁺ ions that can simultaneously detect cytosolic and mitochondrial Zn²⁺ ions in live cells and living tissues at 115 mm depth by dual-color TPM imaging with minimum interference from other biological

Full text of DOI:10.1039/c4ob00101j

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Dual-color imaging of cytosolic and
mitochondrial zinc ions in live tissues
with two-photon fluorescent probes†
Cite this: Org. Biomol. Chem., 2014,
12, 3406
Kailash Rathore,‡ Chang Su Lim,‡ Young Lee and Bong Rae Cho*
Received 14th January 2014,
Accepted 24th March 2014
We report two-photon probes for Zn2+ ions that can simultaneously detect cytosolic and mitochondrial
Zn2+ ions in live cells and living tissues at 115 mm depth by dual-color TPM imaging with minimum inter-
ference from other biologically relevant species.
DOI: 10.1039/c4ob00101j
www.rsc.org/obc
Introduction
The zinc ion is an essential trace metal ion that plays impor-
tant roles in mammalian cells as catalytic or structural co-
factors.1 The total zinc ion content in mammalian cells is
approximately 0.2 mM, of which a small fraction exists as cyto-
solic free Zn2+ ([Zn2+]cyto).2 Since a low nanomolar concen-
tration of free Zn2+ can be cytotoxic, Zn2+ homeostasis must be
tightly controlled for proper cell function.1 If the [Zn2+]cyto
level is elevated, the excess Zn2+ is removed by transferring
Zn2+ to the extracellular space and shuffling Zn2+ to subcellular
stores such as endoplasmic reticulum (ER) and mitochondria.3
To visualize such processes in a live tissue, it is crucial to
develop efficient two-photon (TP) probes exhibiting significant
TP action cross section (Φδmax) values that can simultaneously
detect the Zn2+ in different organelles by two-photon
microscopy (TPM). TPM, which utilizes two near-infrared
photons as the excitation source, has become an indispensable
tool in biomedical research due to the capability of visualizing
the targets deep inside a live tissue with intrinsically localized
emission for a long period of time.4
Chart 1 Structures of BZn-Cyto and FZn-Mito.
widely-separated wavelength regions upon binding with Zn2+
(Chart 1). BZn-Cyto was derived from 6-(benzo[d]oxazol-2′-yl)-2-
(N,N-dimethylamino)naphthalene (BODAN) as the fluorophore
and 2-methoxy-N-{2-[N,′N′-bis(2-picolyl)]aminoethyl}-1,4-phenyl-
enediamine (BPEA) as a Zn2+ chelator, while FZn-Mito was
derived from 7-(benzo[d]thiazol-2-yl)-9,9-dimethyl-9H-fluorene
(BMF) as the reporter, 2-methoxy-N-{2-[N′-(2-picolyl)-N′-
(2-pyridin-2-yl)ethyl]aminoethyl}-1,4-phenylenediamine (PPEA)
as a Zn2+ chelator, and the triphenylphosphonium ion as the
mitochondrial targeting moiety.7 We have employed BODAN
and BMF as TP fluorophores because they exhibited significant
Φδmax values and emitted TPEF at 470 and 550 nm in aqueous
buffer,8,9 respectively, and we have adopted BPEA and PPEA as
the Zn2+ receptors from our earlier work because their Kd
values are 0.5 and 21 nM, respectively,5 which are within the
range of expected Zn2+ concentrations in the two organelles.6
Recently, TP probes for Zn2+ ions that can independently
detect Zn2+ in the cytoplasm and mitochondria were
reported.5,6 However, simultaneous detection of Zn2+ in the
two organelles was not possible, due to the lack of probes for
such a purpose. Moreover, it is a crucial task to develop
organelle-specific probes that emit fluorescence over a wide
range of wavelengths for multi-color imaging. We, therefore,
have developed TP probes for cytosolic (BZn-Cyto) and mito-
chondrial Zn2+ ([Zn2+]mito) (FZn-Mito), which emit TPEF at
Results and discussion
Synthesis
Department of Chemistry, Korea University, 1-Anamdong, Seoul, Korea.
E-mail: chobr@korea.ac.kr; Fax: (+82) 2-3290-3544; Tel: (+82) 2-3290-3129
†Electronic supplementary information (ESI) available: Experimental details for
Synthesis of BZn-Cyto is summarized in Scheme 1.
Compound 2 was prepared by the Pd-catalyzed coupling of
1 with 5-methoxybenzo[d]oxazole. Compound 3 was obtained
by the reaction of 2 with BBr3. BZn-Cyto was prepared by HATU
the synthesis, photophysical studies, and cell imaging. See DOI:
10.1039/c4ob00101j
‡K.R. and C.S.L. contributed equally.
3406 | Org. Biomol. Chem., 2014, 12, 3406–3412
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 1 (a) Normalized one-photon (dotted line) and two-photon
excited fluorescence spectra (solid line) of BZn-Cyto (blue) and FZn-
Mito (red) measured in MOPS buffer and HeLa cells, respectively. The
excitation wavelengths for the one-photon and two-photon processes
were 380 and 750 nm, respectively.
Scheme 1 (a) 5-Methoxybenzo[d]oxazole, Pd(OAc)2, t-Bu3P, CuBr,
Cs2CO3, toluene, 90 °C, 16 h, 56%; (b) BBr3, CH2Cl2, −78 °C–rt, 16 h,
63%; (c) HATU, DIPEA, BPEA, DMF, rt, 16 h, 37%; (d) CH3OCH2CH2-
OCH2CH2OTs, K2CO3, DMF, rt, 22 h, 61%.
FZn-Mito increased gradually by 49 and 73 nm, respectively, as
the solvent polarity was increased in the order, 1,4-dioxane <
dioxane–EtOH (1/1) < DMF < MOPS containing 1.5 mM
sodium dodecylsulfate (MOPS/SDS) < EtOH < MOPS (Fig. S2
and Table S1†).
Upon 750 nm TP excitation in a scanning lambda mode,
HeLa cells labeled with BZn-Cyto and FZn-Mito emitted broad
spectra with emission maxima at 450 and 530 nm, respectively,
which are slightly blue-shifted from those measured in MOPS
buffer (Fig. 1). This outcome indicates that the microenviron-
ments of the probes in the organelles are more hydrophobic
Scheme 2 (a) CH3I, K2CO3, DMF, 0 °C, 5 h, 41%; (b) BrCH2CO2-t-Bu,
proton sponge, CH3CN, reflux, 6 h, 87%; (c) (i) TFA–CH2Cl2 (1/1), 0 °C–
rt, 5 h; (ii) HATU, DIPEA, PPEA, DMF, rt, 16 h, 53% (over two steps);
(d) i-PrMgCl·LiCl, DMF, THF, 0 °C–rt, 4.5 h, 43%; (e) 12, p-toluenesulfonic
acid, CHCl3, reflux, 16 h, 25%; (f) DCC, HOBT, CH2Cl2, rt, 16 h, 63%;
(g) hydrazine monohydrate, EtOH, 80 °C, 3 h, 47%.
than MOPS buffer. Moreover, the TPEF spectrum of BZn-Cyto
fl
was similar to those measured in EtOH (λ
= 445 nm) and
max
fl
DMF (λ
= 451 nm), whereas that of FZn-Mito was similar to
max
fl
max
those measured in MOPS/SDS (λ
= 520 nm) and dioxane–
= 536 nm) (Fig. 1, S2 and Table S1†), respect-
fl
EtOH (1/1) (λ
max
ively. This result indicates that EtOH and DMF can reasonably
represent the microenvironment of BZn-Cyto, while MOPS/SDS
and dioxane–EtOH (1/1) are good models for that of FZn-Mito.
coupling of 3 with BPEA to obtain 4 followed by alkylation
with 2-(2-methoxyethoxy)ethyl tosylate.
FZn-Mito was prepared by the multi-step synthesis as
shown in Scheme 2. N-methylation of 5 followed by alkylation
of 6 with t-butyl bromoacetate produced 7. The t-Bu group was
removed and the resulting carboxylic acid derivative was
coupled with PPEA to obtain 8. Compound 8 was then formy-
lated to afford 9. FZn-Mito was synthesized by the reaction
between 9 and 12, which was prepared by DCC coupling of 10
with (2-aminoethyl)triphenylphosphonium bromide followed
by cleavage of the thiazol ring in 11 (see ESI† for details).
Fluorescence titration
We next conducted the fluorescence titration for the complexa-
tion reaction of BZn-Cyto and FZn-Mito with Zn2+. When small
increments of free Zn2+ were added to BZn-Cyto in MOPS
buffer, the TP excited fluorescence (TPEF) intensity increased
gradually with the metal ion concentration (Fig. 2a), presum-
ably because of the blocking of the photo-induced electron
transfer (PeT) process by metal ion complexation.11 Nearly
identical results were observed in the one-photon (OP) process
(Fig. S3a†). The fluorescence enhancement factors [FEF = (F −
Photophysical properties
The solubilities of BZn-Cyto and FZn-Mito in 3-(N-morpholino)- Fmin)/Fmin] of BZn-Cyto measured by OP and TP processes were
propanesulfonic acid (MOPS) buffer (30 mM MOPS, 100 mM 27 and 28 in the presence of 22 nM Zn2+, respectively, indicat-
KCl, pH 7.2) as determined by the fluorescence method10 were ing a high sensitivity of BZn-Cyto to changes in the Zn2+ con-
approximately 10 and 14 nmol, which correspond to 5 and centration (Table 1). Moreover, the titration curves fitted well
7 μM in MOPS buffer, respectively (Fig. S1†). In MOPS buffer, with a 1 : 1 binding model (Fig. 2b), the Hill plot was linear
BZn-Cyto and FZn-Mito showed absorption maxima (λmax) at with a slope of 1.0 (Fig. 2c), and the Job plot exhibited a
367 (ε = 2.35 × 104 M−1 cm−1) and 392 nm (ε = 1.32 × 104 M−1 maximum point at the mole fraction 0.50 (Fig. S3b†), indicat-
fl
cm−1), and emission maxima (λ ) at 470 (Φ = 0.025) and ing 1 : 1 complexation between the probe and Zn2+.12
max
559 nm (Φ = 0.023), respectively (Table S1†). The λmax
remained nearly the same, while the λ
The fluorescence titration of FZn-Mito with Zn2+ was con-
of BZn-Cyto and ducted in MOPS/SDS, which is a good model for the intracellu-
fl
max
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 3406–3412 | 3407

View Article Online
Paper
Organic & Biomolecular Chemistry
by the negligible fluorescence intensity in the presence of
5 mM concentration of Na+, Ca2+, and Mg2+ and 300 µM of
Mn2+, Fe2+, Co2+, Ni2+, Cu2+, and Cd2+, as well as a large
enhancement in the fluorescence intensity upon the addition
of 22 nM of Zn2+ (Fig. 1d). Although the selectivity of BZn-Cyto
for Zn2+ over Co2+, Cu2+, and Cd2+ is not as high as that over
other metal ions, it will not be a problem because the intra-
cellular concentration of these metal ions should be much
lower than 300 µM. A similar result was observed for FZn-Mito
(Fig. S4f†). In addition, BZn-Cyto and FZn-Mito were relatively
pH-insensitive at pH > 4.0 (Fig. S5†). These results established
that BZn-Cyto and FZn-Mito can detect Zn2+ with minimum
interference from other competing metal ions and pH.
Two-photon action cross section
Fig. 2 (a) Two-photon fluorescence spectra of 3 μM BZn-Cyto in the
presence of free Zn2+ (0–22 nM). (b) One-photon ( ) and two-photon
We next evaluated the abilities of BZn-Cyto and FZn-Mito to
●
detect Zn2+ in a TP mode. The TP action cross sections (Φδmax
)
○
(
) fluorescence titration curves for the complexation of BZn-Cyto with
free Zn2+ (0–22 nM). (c) Hill plots for the complexation of BZn-Cyto
with free Zn2+ (0–22 nM) measured by one-photon and two-photon
processes. (d) The relative fluorescence intensity of 3.0 μM of BZn-Cyto
of BZn-Cyto and FZn-Mito were measured by the fluorescence
method.15,16 The Φδmax values of BZn-Cyto were 4 GM in the
absence and 103 GM at 740 nm in the presence of excess Zn2+
in MOPS buffer (Fig. 3 and Table 1). The value increased to
135 GM in EtOH in the presence of excess Zn2+ (Fig. 3 and
Table 1), presumably because of the enhanced fluorescence
quantum of the BZn-Cyto yield (0.68 in MOPS vs. 0.90 in
EtOH) in a less polar solvent (Table S1†). Since it was difficult
to measure the Φδmax value of FZn-Mito in MOPS/SDS due
to the scattering of TPEF signals by the SDS aggregates,
the Φδmax value was determined in dioxane–EtOH (1/1), which
is a good model for the intracellular environment (see above).
The Φδmax value of FZn-Mito in dioxane–EtOH (1/1) was 80 GM
in the presence of excess Zn2+. These values allowed us to
obtain bright TPM images of the probe-labeled cells and
tissues by TPM using BZn-Cyto and FZn-Mito as the TP
probes.
in the presence of 5.0 mM for Na+, K+, Ca2+, Mg2+; 300 μM for Mn2+
,
Fe2+, Co2+, Ni2+, Cu2+, Cd2+ (empty bars) followed by addition of 22 nM
of free Zn2+ (filled bars). All data were collected in MOPS buffer. The
excitation wavelengths for one-photon and two-photon processes were
367 and 750 nm, respectively.
Table 1 Photophysical properties of BZn-Cyto and FZn-Mitoa
b
(2)
d
e
Compd.
λmax/λfl
Φc
λ
max
Φδmax
FEFTP f
KTdP g
BZn-Cyto
FZn-Mito
367/470
392/559
0.025
0.023
740
103h
80i
27
16
1.8
17
800i
a All measurements were made in MOPS buffer (BZn-Cyto) and MOPS/
b
SDS (FZn-Mito) unless otherwise noted. λmax of the one-photon
absorption and emission spectra in nm. c Fluorescence quantum yield.
The uncertainty is 15%. d λmax of the two-photon excitation spectra in
nm. e Two-photon action cross section measured in the presence of
excess Zn2+ in 10−50 cm4 s per photon (GM), 15%. f Fluorescence
enhancement factors measured by the two-photon process.
g Dissociation constants for Zn2+ measured by the two-photon process
in nM, 10%. h The Φδmax value was 135 GM in EtOH. i The solvent
was dioxane–EtOH (1/1).
Detection of Zn2+ ions in a live cell
We then sought to apply BZn-Cyto and FZn-Mito as TP probes
for detecting [Zn2+]cyto and [Zn2+]mito in cellular environments
by dual-color imaging. The TPM images of the HeLa cells
lar environment (see above). The titration behavior was similar
to that for BZn-Cyto; that is, the OP and TP titration curves
fitted well with the 1 : 1 binding model with FEFTP = 16 and
FEFOP = 16 in the presence of 130 nM Zn2+ (Fig. S4a–c†). The
Hill plots (Fig. S4d†) and the Job plots (Fig. S4e†) also indi-
cated 1 : 1 complexation between the probe and Zn2+ 12
The
.
dissociation constants (KTdP) for the complexation reactions of
BZn-Cyto and FZn-Mito with Zn2+ in a TP mode were calcu-
lated by the literature method (Fig. 1b and S4c†).13,14 The KdTP
values of BZn-Cyto and FZn-Mito were 1.8 0.3 and 17 2 nM,
respectively (Table 1).
Selectivity
■
Fig. 3 Two-photon action spectra of BZn-Cyto in EtOH ( ) and MOPS
We then studied the selectivity of BZn-Cyto and FZn-Mito for
●
○
buffer ( ) and of FZn-Mito in 1,4-dioxane–EtOH (1/1) ( ) in the absence
Zn2+. BZn-Cyto showed a high selectivity for Zn2+, as revealed (dotted line) and presence (solid line) of excess Zn2+
.
3408 | Org. Biomol. Chem., 2014, 12, 3406–3412
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
labeled with BZn-Cyto and FZn-Mito were bright, presumably
To confirm whether the bright regions reflected [Zn2+]cyto
because of the easy loading and large Φδmax values. Since the and [Zn2+]mito, we treated the HeLa cells co-labeled with BZn-
TPEF spectra of BZn-Cyto and FZn-Mito measured in HeLa Cyto and FZn-Mito with 2,2′-dithiodipyridine (DTDP; 150 μM),
cells showed appreciable overlap, we chose half of the bands, a reagent that promotes the release of Zn2+ from Zn2+-binding
that is, 400–450 nm (Ch1, BZn-Cyto) and 550–650 nm (Ch2, proteins,19 and carbonyl cyanide m-chlorophenylhydrazone
FZn-Mito), as the detection windows (Fig. 1). Under these con- (CCCP; 10 μM), a reagent that collapses mitochondrial poten-
ditions, the TPEF of FZn-Mito contributed 6% of the total tial and then depolarizes the plasma membrane, thereby
TPEF in Ch1, while that of BZn-Cyto contributed 4% of the releasing the Zn2+ into the extracellular space (Fig. 5).20 In
total TPEF in Ch2. Hence, it was possible to simultaneously both channels, the TPEF intensities increased upon addition
detect [Zn2+]cyto and [Zn2+]mito by dual-color imaging with of DTDP and decreased upon treatment with CCCP. The only
minimum interference from each other.
difference was the higher sensitivity of BZn-Cyto to the stimuli
To test whether FZn-Mito can specifically stain the mito- that can be attributed to its smaller KTdP value. It was not poss-
chondria, we performed a co-localization experiment with the ible to visualize the transport of Zn2+ between the two regions
HeLa cells co-labeled with FZn-Mito and MitoTracker Red FM probably because the process was too fast to monitor with
(MTR), a commercially available one-photon probe for mito- our instrument. Nevertheless, BZn-Cyto and FZn-Mito were
chondria, by using the detection windows at 550–650 (FZn- clearly capable of detecting [Zn2+]cyto and [Zn2+]mito in live cells
Mito) and 600–700 nm (MTR), respectively (Fig. S6†). Since by dual-color TPM imaging (Fig. 5h). Moreover, the assays
FZn-Mito did not emit fluorescence and the TPEF of MTR was with a CCK-8 kit revealed that both probes exhibited low cyto-
not detected at 450–550 nm when excited at 543 and 800 nm, toxicity (Fig. S7†). Further, the TPEF intensity at a given spot
respectively, the co-localization experiment could be con- on the HeLa cells labeled with BZn-Cyto and FZn-Mito
ducted without interference from each other (Fig. 4d).17 The remained nearly the same after continuous irradiation of the
TPM image matched well with the OPM image of mitochon- fs-pulses for
1 h, indicating their high photostabilities
dria. The Pearson’s colocalization coefficient (A), which (Fig. S8†). These results established that BZn-Cyto and FZn-
describes the correlation of the intensity distribution between Mito can simultaneously detect [Zn2+]cyto and [Zn2+]mito by
the channels,18 of FZn-Mito with MitoTracker Red FM was dual-color TPM imaging for a long period of time with
0.86, indicating that FZn-Mito exists predominantly in mito-
chondria (Fig. 4).
Fig. 5 TPM images of HeLa cells co-labeled with BZn-Cyto (3 µM) and
Fig. 4 (a) TPM and (b) OPM images of HeLa cells co-labeled with FZn-
Mito (3 µM) and MTR (1 µM). (c) Colocalized image. (d) Two-photon and
one-photon excited fluorescence spectra of FZn-Mito and MTR col-
lected from (a) and (b). FZn-Mito did not emit fluorescence when
excited at 543 nm. The wavelengths for one-photon and two-photon
excitation were 543 and 800 nm, respectively, and the emission was
collected at 450–550 nm (a) and 600–700 nm (b), respectively. Scale
bar: 25 µm.
FZn-Mito (3 µM) collected at Ch1 (a–c) and Ch2 (d–f), respectively.
Images were obtained before (a, d) and after (b, e) addition of 150 µM
DTDP. (c, f) After addition of 10 µM CCCP to (b, e). (g) Merged image of
(a) and (d). (h) Enlarged images of the white box in (g). (i) Relative TPEF
intensity of HeLa cells co-labeled with BZn-Cyto and FZn-Mito as a
function of time. The excitation wavelength was 750 nm and TPEF inten-
sities were collected at Ch1 (BZn-Cyto) and Ch2 (FZn-Mito), respectively.
Scale bars: (f) 30 and (h) 5 µm.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 3406–3412 | 3409

View Article Online
Paper
Organic & Biomolecular Chemistry
minimum interference from other competing metal ions, pH,
and cytotoxicity.
Conclusions
We have developed TP probes for [Zn2+]cyto and [Zn2+]mito
,
Detection of Zn2+ ions in a live tissue
which emit TPEF at widely-separated wavelength regions upon
binding with Zn2+, and show significant TP cross sections, low
cytotoxicity, high photostability, and insensitivity to pH in the
biologically relevant pH range. The new probes can simul-
taneously detect [Zn2+]cyto and [Zn2+]mito in live cells, as well as
in living tissues at 115 mm depth by dual-color TPM imaging
with minimum interference from other biologically relevant
species.
We next assessed the ability of BZn-Cyto and FZn-Mito to
detect Zn2+ deep inside live tissue. For this experiment, the
hippocampus was isolated from a 2-week-old rat, and a slice
was incubated with 20 µM of BZn-Cyto and FZn-Mito for 1 h
at 37 °C. Since it took a longer time to stain the tissues than
cells during which time they may be deformed, excess
amounts of BZn-Cyto and FZn-Mito were used to facilitate
staining. The bright-field image of a part of this slice revealed
the CA2 and CA3 regions (Fig. 6g). The TPM images collected
at Ch1 and Ch2 revealed the distribution of [Zn2+
]
and
cyto
Experimental section
Synthesis
[Zn2+]mito in the same region (Fig. 6a,b and d,e). At a higher
magnification [Zn2+]cyto and [Zn2+]mito distribution could be
clearly visualized at 115 mm depth in a live tissue (Fig. 6f).
Moreover, the TPEF intensity increased upon addition of
300 µM DTDP, and it decreased upon treatment with 10 µM
CCCP (Fig. 6h), a result similar to that observed in the cells.
These findings confirmed that BZn-Cyto and FZn-Mito can
simultaneously detect the [Zn2+]cyto and [Zn2+]mito in live
tissues at 115 mm depth by dual-color TPM imaging.
BZn-Cyto. A solution of diethyleneglycol monomethyl ether
tosylate (14 mg, 0.050 mmol) in DMF (1 mL) was added drop-
wise to a mixture of compound 4 (29 mg, 0.042 mmol) and
K2CO3 (7.0 mg, 0.050 mmol) in DMF (2 mL) at rt under an
argon atmosphere. The resulting mixture was stirred for 22 h
while TLC examination revealed complete consumption of the
starting material. The reaction mixture was diluted with water
(10 mL) and extracted with EtOAc (4 × 5 mL). Combined
organic extracts were washed with brine (5 mL), dried over
anhydrous MgSO4 and filtered. Evaporation of the solvent
in vacuo afforded a crude residue which was purified by silica gel
column chromatography using MeOH–CHCl3–acetone (1/9/10)
as the eluent to furnish BZn-Cyto (20 mg, 0.025 mmol) in 61%
yield as a colorless viscous oil. 1H NMR (300 MHz, CDCl3):
8.61 (s, 1H), 8.52–8.48 (m, 2H), 8.20 (dd, J = 8.8, 2.2 Hz, 1H),
8.11 (s, 1H), 7.89 (d, J = 8.8 Hz, 1H), 7.79 (d, J = 8.8 Hz, 1H),
7.66–7.58 (m, 2H), 7.52–7.43 (m, 4H), 7.18 (dd, J = 8.8, 2.2 Hz,
1H), 7.16–7.07 (m, 3H), 6.97 (dd, J = 8.8, 2.2 Hz, 1H), 6.70 (dd,
J = 8.8, 2.2 Hz, 1H), 6.38 (d, J = 8.8 Hz, 1H), 4.21 (t, J = 6.0 Hz,
2H), 4.10 (s, 2H), 3.94–3.81 (m, 9H), 3.78–3.72 (m, 2H),
3.63–3.57 (m, 2H), 3.40 (s, 3H), 3.23 (s, 3H), 3.16 (t, J = 6.0 Hz,
2H), 2.86 (t, J = 6.0 Hz, 2H). 13C NMR (100 MHz, acetone-d6):
169.1, 166.1, 161.5, 158.7, 151.2, 150.7, 148.5, 147.2, 145.2,
138.5, 138.0, 137.2, 131.8, 130.3, 129.3, 128.8, 128.1, 126.0,
124.7, 123.8, 122.3, 118.6, 115.4, 114.3, 112.3, 110.9, 108.0,
105.5, 105.2, 73.7, 72.2, 71.3, 70.2, 61.8, 59.8, 58.9, 56.9, 54.5,
42.8, 41.2. HRMS (FAB): calcd for C46H50N7O6 [M + H]+:
796.3823, Found 796.3820.
FZn-Mito. A solution of amino thiol 12 (19 mg, 0.035 mmol)
in CHCl3 (1 mL) was added dropwise to a stirred solution of
aldehyde 9 (24 mg, 0.035 mmol) and p-toluenesulfonic acid
monohydrate (PTSA, 2 mg, 0.0003 mmol) in CHCl3 (4 mL)
under an argon atmosphere. The reaction mixture was refluxed
for 16 h while accumulation of a brown solid was observed on
Fig. 6 (a–f) TPM image of a rat hippocampal slice co-stained with 20
µM each of BZn-Cyto and FZn-Mito collected at Ch1 (a, b) and Ch2 (d,
e), respectively. (c) Merged image of (b) and (e). TPM images were
acquired before (a, d) and after (b, e) addition of DTDP by 10× magnifi-
cation. (f) Magnification at 100× in the pyramidal neuron layer of CA3
regions (white box in (c)) at a depth of 115 µm. (g) Bright-field image of the wall of the RB flask. Mother liquor containing the desired
CA2–CA3. (h) Relative TPEF intensity of a rat hippocampal slice co-
labeled with BZn-Cyto- and FZn-Mito upon addition of 300 µM DTDP
and 10 µM CCCP as a function of time. The excitation wavelength was
750 nm and TPEF intensities were collected at Ch1 (BZn-Cyto) and Ch2
product was decanted carefully and the solid residue was
rinsed twice with cold CHCl3 (2 × 3 mL). Evaporation of the
solvent under reduced pressure yielded a crude residue which
was purified by reverse-phase HPLC using the following con-
ditions: YMC-Pack ODS-A, (20 × 250 mm), 5 µm, 12 nm;
(FZn-Mito), respectively. Scale bars: 30 µm (f, 100×) and 300 µm (g,
10×).
3410 | Org. Biomol. Chem., 2014, 12, 3406–3412
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
mobile phase, MeOH–H2O (0.1% TFA) = 20/80 to 100/0 to 100/0 Measurement of two-photon cross section
(linear gradient, 60 then 70 min); flow rate, 8 mL min−1
;
The two-photon cross sections (δ) were determined by using
the femtosecond (fs) fluorescence measurement technique as
described.15 BZn-Cyto and FZn-Mito were dissolved in solvent
at concentrations of 3.0 × 10−6 M and then the two-photon
induced fluorescence intensity was measured at 720–900 nm
by using Rhodamine 6G in MeOH as the reference, whose two-
photon property has been well characterized in the literature.16
The TPA cross section was calculated using δ = δr(SsФrϕrcr)/
(SrФsϕscs) where the subscripts s and r stand for the sample
and reference molecules, respectively. The intensity of the
signal collected using a CCD detector was denoted as S. Φ is
the fluorescence quantum yield. ϕ is the overall fluorescence
collection efficiency of the experimental apparatus. The
number density of the molecules in solution was denoted as c.
δr is the TPA cross section of the reference molecule.
temperature, 40 °C; injection, 0.1 mL, detection, UV (210 nm).
The fraction of interest at retention time 49.5 min was col-
lected and lyophilized to furnish FZn-mito (9 mg, 0.007 mmol)
in 21% yield as a dark brown, gummy product. 1H NMR
(400 MHz, acetone-d6): δ 8.61–8.49 (m, 2H), 8.23 (s, 1H),
8.14–7.99 (m, 10H), 7.99–7.90 (m, 4H), 7.90–7.68 (m, 11H),
7.54 (d, J = 7.8 Hz, 1H), 7.45–7.34 (m, 2H), 7.11–6.99 (m, 2H),
6.82 (d, J = 7.8, 1H), 6.65 (d, J = 7.8, 1H), 4.65 (s, 2H), 4.18 (s,
2H), 4.09–3.85 (m, 6H), 3.84–3.65 (m, 7H), 3.65–3.51 (m, 2H),
3.23 (s, 3H), 1.56 (s, 6H). 31P NMR (162 MHz, DMSO-d6):
22.3 ppm. HRMS (FAB+): m/z calcd for [C68H66N8O3PS]+:
1105.4716, Found 1105.4711.
Spectroscopic measurements
Cell culture
Absorption spectra were recorded on a S-3100 UV-Vis spectro-
photometer and fluorescence spectra were obtained with a
FluoroMate FS-2 fluorescence spectrometer with a 1 cm stan-
dard quartz cell. The fluorescence quantum yield was deter-
mined by using coumarin 307 (Φ = 0.95 in MeOH) as the
reference by the literature method.21
HeLa human cervical carcinoma cells were obtained from
American Type Culture Collection (ATCC, Manassas, VA, USA).
The cells were cultured in DMEM (WelGene Inc., Seoul, Korea)
supplemented with heat-inactivated 10% FBS (WelGene), peni-
cillin (100 units mL−1), and streptomycin (100 μg mL−1). All
the cell lines were maintained in a humidified atmosphere of
5% CO2 and 95% air at 37 °C. Two days before imaging, the
cells were detached and were replaced on glass-bottomed
dishes (MatTek). For labeling, the growth medium was
removed and replaced with DMEM without FBS. The cells were
incubated with BZn-Cyto and FZn-Mito (3 μM) for 30 min at
37 °C. Following this incubation, the cells were washed three
times with DMEM without FBS and imaged.
Water solubility
A small amount of dye was dissolved in DMSO to prepare the
stock solutions (1.0 × 10−2 M). The solution was diluted to 6.0
× 10−3–6.0 × 10−5 M and added to a cuvette containing 2.0 mL
of buffer (30 mM MOPS, 100 mM KCl, pH 7.2) using a micro
syringe. In all cases, the concentration of DMSO in buffer was
maintained at 0.2%.10 The plots of fluorescence intensity
against the total amount of the dye injected into the cuvette
were linear at low dye content and showed downward curvature
as more dye was added (Fig. S1†). The maximum point in the
linear region was taken as the solubility. The solubility of BZn-
Cyto was 5.0 μM and that of FZn-Mito was 7.0 μM.
Two-photon fluorescence microscopy
Two-photon fluorescence microscopy images of probe-labeled
HeLa cells and tissues were obtained with spectral confocal
and multiphoton microscopes (Leica TCS SP2) with a ×10 and
×100 objective, numerical aperture (NA) = 0.30 and 1.30. The
two-photon fluorescence microscopy images were obtained
with a DM IRE2 Microscope (Leica) by exciting the probes with
Determination of apparent dissociation constants (Kd)
a
mode-locked titanium–sapphire laser source (Coherent
Chameleon, 90 MHz, 200 fs) set at wavelength 740 nm and output
power 1305 mW, which corresponded to 2.28 × 108 mW nm−2
(×10) and 3.80 × 109 mW nm−2 (×100) power at the focal plane.
To obtain images in the 400–450 nm (channel 1) and
550–650 nm (channel 2) ranges, internal PMTs were used to
collect the signals in an 8 bit unsigned 512 × 512 pixels at 400
Hz scan speed.
A series of solutions containing various concentrations of free
Zn2+ ([Zn2+]free) was prepared by dissolving various amounts of
ZnSO4 (0–9.5 mM) and 10 mM EGTA in MOPS (30 mM, pH 7.2,
0.1 M KCl) by the literature method.13,22 To determine the Kd
values for the probe-Zn2+ complexes, the fluorescence titration
curves (Fig. 1a and S3b†) were obtained and fitted to the fol-
lowing equation (Fig. 1b):23,24
2þ
½Zn
ꢁ
free
2þ
Co-localization experiments
F ¼ F0 þ ðFmax ꢀ F0Þ
Kd þ ½Zn
ꢁ
free
Co-localization experiments were conducted by co-staining
where F is the fluorescence intensity, Fmax is the maximum HeLa cells with FZn-Mito (3 µM) and MTR (1 µM) for 30 min.
fluorescence intensity, F0 is the fluorescence intensity in the The emission was collected at 550–650 (FZn-Mito) and
absence of Zn2+, and [Zn2+]free is the free Zn2+ concentration. 600–700 nm (MTR) regions upon excitation at 543 (MTR) and
The Kd value that best fits the titration curve with the above 740 (BZn-Cyto) nm, respectively. The background images were
equation was calculated using the Excel program.
corrected, and the distribution of pixels in the TPM and OPM
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 3406–3412 | 3411

View Article Online
Paper
Organic & Biomolecular Chemistry
images acquired in the shorter and longer wavelength chan-
nels, respectively, was compared by using scatter gram. The
Pearson’s co-localization coefficients were calculated by using
the Autoquant X2 program.
5 A. Antony, C. S. Lim, Y. S. Tian, J. H. Han, M. Y. Kang and
B. R. Cho, Chem. – Asian J., 2011, 6, 1234.
6 G. Masanta, C. S. Lim, H. J. Kim, J. H. Han, H. M. Kim and
B. R. Cho, J. Am. Chem. Soc., 2011, 133, 5698.
7 (a) B. C. Dickinson, D. Srikun and C. J. Chang, Curr. Opin.
Chem. Biol., 2010, 14, 50; (b) B. C. Dickinson and
C. J. Chang, J. Am. Chem. Soc., 2008, 130, 9638.
8 J. H. Han, S. K. Park, C. S. Lim, M. K. Park, H. M. Kim and
B. R. Cho, Chem. – Eur. J., 2012, 18, 15246.
Cell viability
The viability of HeLa cells was determined using a CCK-8 kit
(Cell Counting Kit-8, Dojindo, Japan) according to the manu-
facturer’s protocol.
9 S. K. Das, C. S. Lim, S. Y. Yang, J. H. Han and B. R. Cho,
Chem. Commun., 2012, 48, 8395.
Photostability
The photostabilities of BZn-Cyto and FZn-Mito were deter- 10 H. M. Kim, H. J. Choo, S. Y. Jung, Y. G. Ko, W. H. Park,
mined by monitoring the changes in TPEF intensity with HeLa
cells chosen without bias.
S. J. Jeon, C. H. Kim, T. Joo and B. R. Cho, ChemBioChem,
2007, 8, 553.
11 R. A. Bissell, E. Calle, A. P. de Silva, S. A. de Silva,
H. Q. N. Gunaratne, J.-L. Habib-Jiwan, S. L. A. Peiris,
R. A. D. D. Rupasinghe, T. K. S. D. Samarasinghe, K. R. A.
S. Sandanayake and J.-P. Soumillion, J. Chem. Soc., Perkin
Trans 2, 1992, 1559.
Preparation and staining of fresh rat hippocampal and
hypothalamic slices
Slices were prepared from the hippocampus of a 2-week-old rat
(SD). Coronal slices were cut into 400 μm-thick using a vibrat-
ing-blade microtome in artificial cerebrospinal fluid (ACSF; 12 K. A. Connors, Binding Constants, Wiley, New York, 1987.
138.6 mM NaCl, 3.5 mM KCl, 21 mM NaHCO3, 0.6 mM 13 C. J. Fahrni and T. V. O’Halloran, J. Am. Chem. Soc., 1999,
NaH2PO4, 9.9 mM D-glucose, 1 mM CaCl2, and 3 mM MgCl2).
Slices were incubated with 20 μM BZn-Cyto and FZn-Mito in 14 T. Hirano, K. Y. Urano and T. Nagano, J. Am. Chem. Soc.,
ACSF bubbled with 95% O2 and 5% CO2 for 30 min at 37 °C. 2002, 124, 6555.
Slices were then washed three times with ACSF and transferred 15 S. K. Lee, W. J. Yang, J. J. Choi, C. H. Kim, S.-J. Jeon and
121, 11448.
to glass-bottomed dishes (MatTek) and observed using a spec-
tral confocal two-photon microscope.
B. R. Cho, Org. Lett., 2005, 7, 323.
16 N. S. Makarov, M. Drobizhev and A. Rebane, Opt. Express,
2008, 6, 4029.
17 A Guide to Fluorescent Probes and Labeling Technologies,
ed. R. P. Haugland, Molecular Probes, Eugene, OR, 10th
edn, 2005.
18 K. Emmerson and K. Roehrig, Comp. Biochem. Physiol., B:
Biochem. Mol. Biol., 1992, 103, 663.
19 E. Aizenman, A. K. Stout, K. A. Hartnett, K. E. Dineley,
B. McLaughlin and I. J. Reynolds, J. Neurochem., 2000, 75,
1878.
20 T. Caporale, D. Ciavardelli, C. Di Ilio, P. Lanuti, D. Drago
and S. L. Sensi, Exp. Neurol., 2009, 218, 228.
21 J. N. Demas and G. A. Crosby, J. Phys. Chem., 1971, 75,
991.
Acknowledgements
This work was supported by Korea University, the National
Research Foundation (NRF) grant (NRF2012007850), the Basic
Science Research Program of the NRF funded by the Ministry
of Education (NRF20100020209), and the Korea Healthcare
Technology R&D Project, Ministry of Health & Welfare, Republic
of Korea (A111182).
Notes and references
1 (a) J. L. Vinkenborg, T. J. Nicolson, E. A. Bellomo, 22 M. Taki, J. L. Wolford and T. V. O’Halloran, J. Am. Chem.
M. S. Koay, G. A. Rutter and M. Merkx, Nat. Methods, 2009,
6, 737; (b) W. Maret, Biometals, 2009, 22, 149.
2 C. E. Outten and T. V. O’Halloran, Science, 2001, 292, 2488.
3 R. A. Colvin, W. R. Holmes, C. P. Fontaine and W. Maret,
Metallomics, 2010, 2, 306.
Soc., 2004, 126, 712.
23 A. E. Martell and R. M. Smith, NIST Critical Stability
Constants of Metal Complexes. NIST Standard Reference Data-
base 46, Version 7.0, 2003.
24 T. Hirano, K. Kikuchi, Y. Urano, T. Higuchi and T. Nagano,
J. Am. Chem. Soc., 2000, 122, 12399.
4 F. Helmchen and W. Denk, Nat. Methods, 2005, 2, 932.
3412 | Org. Biomol. Chem., 2014, 12, 3406–3412
This journal is © The Royal Society of Chemistry 2014