Reversible fluorescent probe for highly selective and sensitive detection of mercapto biomolecules

Article abstract of DOI:10.1021/ic200181p

A coumarin-derived complex, Hg₂L₂, was reported as a highly sensitive and selective probe for the detection of mercapto biomolecules in aqueous solution. The addition of Cys to a 99% aqueous solution of Hg ₂L₂ resulted in rapid and remarkable fluorescence OFF-ON (emission at 525 nm) due to the ligand-exchange reaction of Cys with L coordinated to Hg²⁺. The increased fluorescence can be completely quenched by Hg²⁺ and recovered again by the subsequent addition of Cys. Such a fluorescence OFF-ON circle can be repeated at least 10 times by the alterative addition of Cys and Hg²⁺ to the solution of Hg ₂L₂, indicating that it can be used as a convertible and reversible probe for the detection of Cys. The interconversion of Hg ₂L₂ and L via the decomplexation/complexation by the modulation of Cys/Hg²⁺ was definitely verified from their crystal structures. Other competitive amino acids without a thiol group cannot induce any fluorescence changes, implying that Hg₂L₂ can selectively determine mercapto biomolecules. Using confocal fluorescence imaging, L/Hg₂L₂ as a pair of reversible probes can be further applied to track and monitor the self-detoxification process of Hg ²⁺ ions in SYS5 cells.

Full text of DOI:10.1021/ic200181p

ARTICLE
pubs.acs.org/IC
Reversible Fluorescent Probe for Highly Selective and Sensitive
Detection of Mercapto Biomolecules
Jiasheng Wu, Ruilong Sheng, Weimin Liu, Pengfei Wang,* Jingjin Ma, Hongyan Zhang, and Xiaoqing Zhuang
Key Laboratory of Photochemical Conversion and Optoelectronic Materials, Technical Institute of Physics and Chemistry,
Chinese Academy of Sciences, Beijing 100190, People's Republic of China
S Supporting Information
b
ABSTRACT: A coumarin-derived complex, Hg2L2, was re-
ported as a highly sensitive and selective probe for the detection
of mercapto biomolecules in aqueous solution. The addition of
Cys to a 99% aqueous solution of Hg2L2 resulted in rapid and
remarkable fluorescence OFFꢀON (emission at 525 nm) due
to the ligand-exchange reaction of Cys with L coordinated to
Hg2+. The increased fluorescence can be completely quenched
by Hg2+ and recovered again by the subsequent addition of Cys. Such a fluorescence OFFꢀON circle can be repeated at least 10
times by the alterative addition of Cys and Hg2+ to the solution of Hg2L2, indicating that it can be used as a convertible and reversible
probe for the detection of Cys. The interconversion of Hg2L2 and L via the decomplexation/complexation by the modulation of
Cys/Hg2+ was definitely verified from their crystal structures. Other competitive amino acids without a thiol group cannot induce
any fluorescence changes, implying that Hg2L2 can selectively determine mercapto biomolecules. Using confocal fluorescence
imaging, L/Hg2L2 as a pair of reversible probes can be further applied to track and monitor the self-detoxification process of Hg2+
ions in SYS5 cells.
Fluorescein/rhodamine-based fluorescent probes have been
receiving considerable attention because of their self-modulation
of OFFꢀON fluorescence, good water solubility, and longer
emission wavelength (over 500 nm).10 In view of their attractive
advantages, there is still a high demand to develop new fluoro-
phores with such properties. As we notice, coumarin derivatives
as important fluorescent dyes are commonly considered to give
rise to strong emission only in organic solvents.11 In aqueous
solution, the emission quantum yield is drastically decreased as a
result of solvation. In addition, the poor solubility and relatively
shorter emission wavelength (often below 500 nm) of conven-
tional coumarin dyes also restrict their practical applications,
especially in biological systems.12 In this work, we developed a
new coumarin-based imine, L, bearing an aminothiourea unit
with an extensively green emission (Figure 1). Compared with
conventional coumarin dyes, both 7-diethylamino and imine
units in L extended the conjugation structure of the fluorophore.
The aminothiourea unit was incorporated into L to increase its
water compatibility and emitting ability in the hydrophilic
environment. As expected, L gives rise to an extensive emission
at 525 nm in aqueous solution with a quantum yield of 0.50. The
emission can be completely quenched (over 98%) by 1 equiv of
Hg2+ and recovered again (over 95%) with the addition of Cys.
Both L and its complex with Hg2+ exhibit good stabilities under a
wide pH span from 6 to 10, covering physiological conditions.
These properties of L including high water solubility, longer
1. INTRODUCTION
Mercapto biomolecules, such as cysteine (Cys), homocysteine
(Hcy), and glutathione (GSH), play a crucial role in maintaining
biological systems.1ꢀ3 Cys deficiency is involved in many syn-
dromes including neurotoxicity, edema, liver damage, and hair
depigmentation.1 An elevated level of Hcy in plasma is a risk
factor for neural tube defects, osteoporosis, and Alzheimer’s and
cardiovascular diseases.2 GSH, as the most abundant intracellular
nonprotein thiol, plays a pivotal role in maintaining the reductive
environment in cells and acts as the redox regulator.3 Because
of its important role in biological systems, much attention has
been paid to the detection of mercapto biomolecules. Many
analytical techniques, including UVꢀvis detection assays, mass
spectrometry (MS),4 gas chromatography,5 high-performance
liquid chromatography,6 and electrochemical methods,7 have
been available techniques, fluorescent probes have been widely
used to sense mercapto biomolecules because of their simplicity,
high selectivity, and sensitivity. Currently, a number of thiol-
reactive fluorescent probes have been reported.8 However,
most of these probes are generally irreversible and thus remain
in living cells, which will cause some negative impacts on cells,
such as restricting further intracellular imaging of the desired
analytes, preventing an understanding of the detailed interac-
tion process in cells, and imparting possible damage to cells.9
This requires a reversible fluorescent probe that can control
the toxic/detox process to avoid toxic cellular uptake. As a result,
it is highly desirable to develop reversible fluorescent probes
for the sensitive and selective determination of mercapto
biomolecules.
Received: January 26, 2011
Published: June 21, 2011
r
2011 American Chemical Society
6543
dx.doi.org/10.1021/ic200181p Inorg. Chem. 2011, 50, 6543–6551
|

Inorganic Chemistry
ARTICLE
Figure 1. Synthetic pathway of ligand L and complex Hg2L2.
emission wavelength (525 nm), and fluorescence OFFꢀON to
the guests are comparable with those of fluorescein/rhodamine-
based fluorescent probes.10
solution was kept at 50 °C for another 10 h, the crude product was
collected and washed with a small amount of ethanol (504 mg, yield
70%). 1H NMR (400 MHz, DMSO-d6, δ): 12.55 (s, 1H, NH), 9.31 (s,
1H, NH), 9.12 (s, 1H, NH), 8.65 (s, 1H, CHdN), 8.22 (s, 1H, ArH),
7.41 (d, J = 8.9 Hz, 1H, ArH), 6.80 (d, J = 7.6 Hz, 1H, ArH), 6.58 (d, J =
8.0 Hz, 1H, ArH), 3.48 (t, J = 6.8 Hz, 4H, CH2), 1.15 (t, J = 6.8 Hz, 6H,
CH3). 13C NMR (100 MHz, DMSO-d6, δ): 166.5, 160.6, 157.1, 152.1,
144.7, 140.9, 131.0, 110.7, 110.3, 108.0, 96.6, 44.5, 12.5. ESI-MS
([Hg2L2 + H]+, m/z): 1182.7 (calcd 1180.9). Anal. Calcd for
C30H36Cl4Hg2N8O4S2: C, 30.54; H, 3.08; N, 9.50. Found: C, 30.45;
H, 3.10; N, 9.54.
Crystal Structure. Crystal data and details of the data collection are
provided in Table 1. Diffraction data for L and Hg2L2 were collected on a
Bruker SMART D8 goniometer with an APEX CCD detector, using Mo
Ka radiation λ = 0.710 73 Å (graphite monochromator). The structures
were solved by direct methods (SHELXTL) and refined on F2 by full-
matrix least-squares techniques.14 Hydrogen atoms were included by
using a riding model. CCDC 783478 (L) and CCDC 783479 (Hg2L2)
contain the supplementary crystallographic data for this paper. These
data can be obtained free of charge from the Cambridge Crystallographic
Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
With excellent properties of L in hand, herein, we further
developed the complex [Hg2L2]4+Cl4ꢀ (abbreviated as Hg2L2 in
the text, vide infra) as a convertible fluorescent probe, which
exhibited high sensitivity and selectivity to mercapto biomole-
cules via a reversible decomplexation (Figure 1). Interconversion
of Hg2L2 and L can repeatedly occur and is thus applied for the
reversible fluorescent determination of mercapto biomolecules,
such as Cys, Hcy, and GSH. Confocal fluorescence imaging
reveals that L/Hg2L2 can be applied to monitor the intracellular
self-detoxification of foreign Hg2+ ions in SYS5 cells.
2. EXPERIMENTAL SECTION
Instruments. 1H and 13C NMR spectra were recorded on an Advance
Bruker instrument (400 MHz). UVꢀvis absorption and fluorescence
spectra were recorded on a Hitachi U-3010 spectrometer and a Hitachi
F-4500 fluorescence spectrometer, respectively. X-ray analysis was measured
on a Rigaku R-Axis Rapid IP diffractometer. Mass spectra were recorded on
a Finigan 4021C MS-spectrometer. High-resolution MS (HRMS) was
recorded on a Bruker Daltonics Inc. APEX II FT-ICR mass spectrometer.
Elemental analysis was measured on a FLASH EA1112 instrument.
Synthesis. 7-(N,N-Diethylamino)coumarin and 7-(N,N-diethyla-
mino)coumarin-3-aldehyde were synthesized according to published
procedures.13
Spectral Measurement. L (319 mg, 1 mmol) and Hg2L2 (590 mg,
0.5 mmol) were dissolved in 1 mL of dimethyl sulfoxide (DMSO) or
DMF and then diluted in a buffer solution (10 mMTris-HCl, pH = 7.6) to
10 μM as the stock solution. The stock solutions of Cys, Hcy, GSH, and
other competitive amino acids (1 mM) were prepared in a Tris buffer
(10 mM Tris-HCl, pH = 7.6). These solutions were all kept at about 4 °C
for further determination.
Fluorescence quantum yields were determined by comparing the
emission integral area of the sample with that of a fluorescence standard
by the following equation:
(E)-2-[[7-(Diethylamino)-2-oxo-2H-chromen-3-yl]methylene]hydra-
zinecarbothioamide (Ligand L). To 7-(N,N-diethylamino)coumarin-3-
aldehyde (250 mg, 1 mmol) dissolved in 20 mL of ethanol was added
dropwise thiosemicarbazone (140 mg, 1.25 mmol) in 10 mL of ethanol.
After the reaction refluxed for 5 h, orange crystals precipitated out. The
needle crystals were collected and washed with ethanol (259 mg, yield
!
ꢀ
ꢁ
2
2
AU
AR
nU
nR
ΦU ¼ ΦR
1
75%). H NMR (400 MHz, DMSO-d6, δ): 11.53 (s, 1H, NH), 8.65
(s, 1H, CHdN), 8.24 (s, 1H, ArH), 8.06 (s, 1H, NH), 8.01 (s, 1H, NH),
7.42(d, J = 8.8Hz, 1H, ArH), 6.78(d, J =7.2 Hz, 1H, ArH), 6.57(d, J =6.8
Hz, 1H, ArH), 3.48 (t, J = 6.8 Hz, 4H, CH2), 1.15 (t, J = 6.8 Hz, 6H, CH3).
13C NMR (100 MHz, DMSO-d6, δ): 177.6, 160.7, 156.4, 151.2, 138.9,
136.5, 130.3, 112.7, 109.8, 108.2, 96.5, 44.3, 12.4. ESI-MS ([L + H]+,
m/z): 319.1 (calcd 319.1). ESI-HRMS ([L + H]+, m/z): 319.1219 (calcd
319.1223). Anal. Calcd for C15H18N4O2S: C, 56.58; H, 5.70; N, 17.60.
Found: C, 56.58; H, 5.71; N, 17.72.
Mercury Complex Hg2L2. To a solution of L (320 mg, 1 mmol)
dissolved in 5 mL of N,N-dimethylformamide (DMF)/EtOH (1:1, v/v)
was added dropwise within 20 min 10 mL of HgCl2 (400 mg, 1.5 mmol)
in ethanol. An orange solid immediately precipitated out. After the
where AU and AR are the integrated areas under the corrected fluores-
cence spectrum for the sample and reference, respectively, nU and nR are
the refractive indices of the sample and reference, respectively. The
fluorescence quantum yield of the standard compound of fluorescein in
0.1 N NaOH aqueous solutions is 0.85.15
Cell Culture. HEK293 and SYS5 cell lines were prepared from a
continuous culture in MEM, mixed with 10% (v/v) fetal bovine serum
and 1% penicillin/streptomycin, at 37 °C in 5% CO2 humidified air.
Following cell adhesion (24 h), the cell culture medium was removed.
After the cell was digested with trypsin, 1.0 mL of the cell suspension was
transferred and incubated in cell walls of the microplate at 37 °C in 5%
CO2 humidified air for 24 h. The cell layer was washed twice with
6544
dx.doi.org/10.1021/ic200181p |Inorg. Chem. 2011, 50, 6543–6551

Inorganic Chemistry
ARTICLE
Table 1. Crystal Data and Structure Refinement for Ligand L and Complex Hg2L2
compound
L
Hg2L2
empirical formula
fw
C19H30N4O4S3
474.65
C30H36Cl4Hg2N8O4S2
1179.79
296(2)
0.710 73
triclinic
P1
temperature (K)
wavelength (Å)
cryst syst
296(2)
0.710 69
triclinic
P1
space group
unit cell dimens
a = 7.1206(14) Å, R = 107.73(3)°,
b = 12.751(3) Å, β = 99.35(3)°,
c = 14.523(3) Å,
a = 8.2194(16) Å, R = 97.93(3)°, b = 9.3156(19) Å,
β = 105.24(3)°, c = 13.358(3) Å, γ = 94.29(3)°,
970.8(3) Å3
γ = 99.80(3)°
1204.9(4)
volume (Å3)
970.8(3)
Z, calcd density (Mg/m3)
2, 1.308
1, 2.018
abs coeff (mmꢀ1
)
0.339
8.326
F(000)
504
564
θ range (deg)
limiting indices
2.65ꢀ25.00
2.22ꢀ27.48
ꢀ8 e h e 8, ꢀ15 ek e 15, ꢀ17 e l e 17
7847/4244 [R(int) = 0.0291]
25.00 (99.9%)
ꢀ10 e h e 10, ꢀ12 ek e 12, ꢀ17 e l e 17
8055/4440 [R(int) = 0.0593]
27.48 (99.6%)
reflns collected/unique
completeness to theta
abs corrn
semiempirical from equivalents
0.9605 and 0.8906
full-matrix least squares on F2
4244/70/301
semiempirical from equivalents
0.7883 and 0.2121
full-matrix least squares on F2
4440/0/226
max and min transmn
refinement method
data/restraints/param
GOF on F2
1.108
0.989
final R indices [I > 2σ(I)]
R indices (all data)
largest diff peak/hole
(e/Å3)
R1 = 0.0614, wR2 = 0.1238
R1 = 0.0912, wR2 = 0.1338
0.304/ꢀ0.186
R1 = 0.0624, wR2 = 0.1529
R1 = 0.0905, wR2 = 0.1656
1.236/ꢀ1.498
phosphate-buffered saline (PBS), and then 1.0 mL of PBS was added in
each well for fluorescence imaging.
diethylamino group has an excellent coplanar configuration.
Thus, its emission ability is extensively increased. However,
when bound with Hg2+, L and Hg2+ can form a 2 + 2 complex
with a chlorine-bridged structure. From Figure 2b, Hg2L2 is a
dimer structure in which ligand L serves as a monodentate to
form HgꢀS bonding. The coordination of Hg2+ is completed by
a sulfur atom, a terminated chlorine atom, and two bridging
chlorine atoms. Two mercury atoms and two chlorine atoms
form a parallelogram configuration, and two HgꢀS bonds are
almost vertical to this parallelogram (the dihedral angels are
99.3° and 105.2°, respectively). The distance of two mercury
atoms is 3.773(13) Å, and the bond length of HgꢀS is 2.411(3) Å.
In addition, some slight conformational changes of ligand L are
observed when bound with Hg2+ (Table 2), indicating that the
bound L still keeps a similar coplanar structure. As a result,
complex Hg2L2 is composed of three paralleling planes (two L
planes and a HgꢀClꢀHgꢀCl plane), and the distance of each
plane is 2.411(3) Å. Three planes are linked by two HgꢀS bonds,
which are vertical to these planes. The formed structure can be
explained by the soft and hard acids and bases.16 According to the
IrvingꢀWilliams rule, the sulfur and chlorine atoms belong to the
soft base, while the oxygen and nitrogen atoms in L are the hard
base. Thus, Hg2+, as a soft acid, can be superior to bind with sulfur
and chlorine atoms to form a more stable dimer structure in
Figure 2b.
Confocal Imaging. HEK293 and SYS5 cells were incubated with
ligand L or complex Hg2L2 (10 μM) for several minutes at room
temperature. After the cell layer was carefully washed twice with PBS, the
fluorescence emission was imaged using a Nikon confocal microscope.
The excitation wavelength of the laser was 408 nm, and the emission was
recorded at 515 nm. The same procedures were repeated when SYS5
cells were continuously treated with 100ꢀ300 μM HgCl2 or Cys. EZ-C1
3.20 free viewer was used as a platform for data analysis.
3. RESULTS AND DISCUSSION
Synthesis and Crystal Structure. Ligand L is a coumarin-
based Schiff base synthesized from condensation of 7-(N, N-
diethylamino)coumarin-3-aldehyde and aminothiourea (Figure 1).
Complex Hg2L2 was obtained from the coordination reaction of
ligand L and HgCl2 in an ethanol/DMF solvent mixture. X-ray
block crystals were grown from the vapor diffusion of ether into
an ethanol/DMF solution (10% DMF). ORTEP diagrams of L
and Hg2L2 are depicted in Figure 2, and the crystallographic data
and selected geometric parameters are also given in Tables 1 and
2. L exists as an S-trans-type Schiff base chain with a thiocarbonyl
group extended out of the coumarin ring (Figure 2a). From the
dihedral angles of L, all atoms of aminothiourea in L were located
almost in the same plane. It is also found that the dihedral angle
between the coumarin-ring and aminothiourea planes is 179.9°,
indicating that the whole molecule in L except for the
Absorption and Fluorescence Behaviors. Generally, imine-
derived molecules are difficult to directly design as probes
because of their instability upon irradiation and/or in acidic/
6545
dx.doi.org/10.1021/ic200181p |Inorg. Chem. 2011, 50, 6543–6551

Inorganic Chemistry
ARTICLE
Figure 2. ORTEP diagrams of (a) ligand L and (b) complex Hg2L2 with displacement atomic ellipsoids drawn at the 30% probability level. Hydrogen
atoms are omitted for clarity.
Table 2. Selected Interatomic Distances [Å] and Bond Angles [deg]
L
Hg2L2
L
Hg2L2
Hg1ꢀCl1
2.385(3)
Cl2ꢀHg1ꢀCl2#1
Cl1ꢀHg1ꢀHg1#1
S1ꢀHg1ꢀHg1#1
Cl2ꢀHg1ꢀHg1#1
Cl2#1ꢀHg1ꢀHg1#1
Hg1ꢀCl2ꢀHg1#1
C13ꢀO1ꢀC7
91.62(8)
109.59(9)
106.57(8)
48.88(7)
42.75(5)
88.38(8)
122.7(8)
115.3(8)
118.6(8)
118.0(9)
119.7(9)
124.9(9)
119.4(9)
115.7(8)
117.4(9)
125.0(9)
117.5(8)
121.2(8)
120.9(9)
118.6(7)
120.6(7)
Hg1ꢀS1
2.411(3)
Hg1ꢀCl2
2.562(3)
Hg1ꢀCl2#1
Hg1ꢀHg1#1
Cl2ꢀHg1#1
S1ꢀC15
2.843(3)
3.773(13)
2.843(3)
1.691(3)
1.376(3)
1.383(3)
1.198(3)
1.271(4)
1.370(3)
1.342(4)
1.320(4)
1.366(4)
1.446(4)
1.455(4)
1.731(9)
122.8(2)
116.3(2)
119.6(2)
117.7(3)
121.5(3)
123.5(3)
119.8(3)
116.7(3)
116.2(3)
126.7(3)
117.1(3)
121.1(3)
117.8(3)
122.9(2)
119.4(2)
O1ꢀC13
1.364(11)
1.381(11)
1.233(12)
1.275(13)
1.392(11)
1.310(12)
1.295(11)
1.350(13)
1.443(13)
1.458(13)
112.80(9)
104.80(9)
95.50(10)
98.53(10)
C14ꢀN2ꢀN3
O1ꢀC7
C15ꢀN3ꢀN2
O2ꢀC13
C6ꢀC5ꢀC10
N2ꢀC14
C7ꢀC6ꢀC5
N2ꢀN3
C11ꢀC12ꢀC14
C11ꢀC12ꢀC13
C14ꢀC12ꢀC13
O2ꢀC13ꢀO1
O2ꢀC13ꢀC12
O1ꢀC13ꢀC12
N2ꢀC14ꢀC12
N4ꢀC15ꢀN3
N3ꢀC15
N4ꢀC15
C11ꢀC12
C12ꢀC14
C12ꢀC13
Cl1ꢀHg1ꢀCl2
S1ꢀHg1ꢀCl2
Cl1ꢀHg1ꢀCl2#1
S1ꢀHg1ꢀCl2#
N4ꢀC15ꢀS1
N3ꢀC15ꢀS1
basic solution. In a summary of previous literatures, most
unstable imine intermediates were reduced to the corresponding
amines for fluorescent sensing.17 In this work, we found that
coumarin-derived imine and its complex, L/Hg2L2, were stable
enough even within a wide pH range (6ꢀ10). Because of the
structural differences of L and Hg2L2, their photophysical
properties are thus remarkably varied (Table 3). Ligand L
displays a characteristic absorption band peaked at 452 nm (ε =
5.3 ꢁ 104 Mꢀ1 cmꢀ1) in a 99% aqueous solution. Upon the
addition of Hg2+, a distinct decrease in the absorbance at 452 nm
and three obvious isosbestic points at 333, 375, and 497 nm were
observed, as shown in Figure 3a. Corresponding to its absorption
spectra, L gave an intensive emission band at 525 nm and its
fluorescence was completely quenched (over 98%) immediately
upon the addition of 1 equiv of Hg2+ (Figure 3b). The quantum
yields of free ligand L and Hg2+-bound forms were determined to
be 0.50 and 0.031, respectively.15 From Job’s plot (inset of
Figure 3b) and electrospray ionization (ESI; Figure S2 in the
Supporting Information) analyses, the stoichiometric ratio of L
with Hg2+ appeared to be 2:2, which was consistent with its
crystal structure.
Table 3. Photophysical Properties of L and Hg2L2 in DMSO/
H2O (1:99, v/v)
absorption
emission
ε
Stokes shift quantum
compound λmax (nm) λmax (nm) (Mꢀ1 cmꢀ1
)
(nm)
yield (Φf)
L
452
450
525
521
5.3 ꢁ 104
3.0 ꢁ 104
73
71
0.50
Hg2L2
0.031
It is generally believed that, in most fluorescent molecules, the
introduction of an extended conjugation structure to the rigid
aryl ring will result in a red shift of the emission wavelength, as
well as a drastic decrease of the fluorescent quantum yield
because of its weaker coplanar effect.18 However, in our sensor
L, the extended conjugation structure (aminothiourea group)
well sites the same plane with the coumarin ring (Figure 2a).
Thus, both the emission wavelength and quantum yield of L are
obviously increased. The fluorescence quenching of L bound
with Hg2+ may arise from the heavy-metal effect. In addition, the
decreased coplanar effect of Hg2L2 will also lead to fluorescence
quenching.
6546
dx.doi.org/10.1021/ic200181p |Inorg. Chem. 2011, 50, 6543–6551

Inorganic Chemistry
ARTICLE
Figure 3. (a) UVꢀvis absorption titration spectra of L (10 μM) in DMSO/H2O (1:99, v/v) upon the addition of Hg2+ (1, 2, 3, 4, 5, 6, 8, 10, 12, 15, and
20 μM, respectively). (b) Fluorescence titration spectra of L (10 μM) in DMSO/H2O (1:99, v/v) upon the addition of Hg2+ (0, 1, 2, 3, 4, 5, 6, 7, 8, and
10 μM, respectively). Excitation wavelength: 450 nm. Inset: Job’s plot analysis of L with Hg2+.
Figure 4. (a) Absorption spectra of Hg2L2 (5 μM) in DMSO/H2O (1:99, v/v) upon the gradual addition of Cys (0, 5, 10, 15, 20, 25, 30, 35, 40, 45, and
50 μM, respectively). (b) Fluorescence spectra of Hg2L2 (5 μM) in DMSO/H2O (1:99, v/v) upon the gradual addition of Cys (0, 1, 2.5, 5, 7.5, 10, 15, 20,
25, and 30 μM, respectively). Excitation wavelength: 450 nm. Inset: Fluorescence intensity of Hg2L2 (5 μM) at 520 nm versus the concentration of
Cys added.
The addition of mercapto biomolecules such as Cys, Hcy, or
GSH to the aqueous solution of Hg2L2 resulted in obvious
spectral changes. UVꢀvis absorption and fluorescence spectra of
Hg2L2 in aqueous solution gradually returned to those charac-
teristic of free L when titrated with Cys (Figure 4). An obvious
fluorescence increase (about 30-fold) was observed, and the
fluorescence quantum yield could be revived to 0.49, indicating
that Hg2L2 could be applied as a fluorescent OFFꢀON probe
for Cys in aqueous solution. The absorption and fluorescence
spectra of L/Hg2L2 clearly illustrate that the addition of Cys
results in a complete release of L from Hg2L2, and the fluores-
cence is thus recovered, which is also confirmed from the crystal
structures.
Figure 5. Fluorescent intensity of Hg2L2 (5 μM) in DMSO/H2O
Interestingly, the alternate addition of a constant amount of
(1:99, v/v, 10 mM Tris-HCl, pH = 7.4) upon the alternate addition of
Cys and Hg2+ to the aqueous solution of Hg2L2 gives rise to a
Cys/HgCl2 with several concentrations (0:0, 10:0, 10:15, 20:15, 20:30,
switchable change in the fluorescence intensity at 525 nm.
Such a reversible interconversion of Hg2L2/L can be repeated
more than 10 times by the modulation of Cys/Hg2+ added,
indicating that Hg2L2 can be developed as a reversible fluores-
cence OFFꢀON probe for Cys. Their corresponding fluores-
cence and visual color changes were also shown in Figure 5.
Reversible interconversions between Hg2L2 and L upon mod-
ulation of Cys/Hg2+ are illustrated in Figure 6.
40:30, 40:60, 60:60, 60:90, and 100:90 μM, respectively). Excitation at
420 nm. Inset: Their corresponding fluorescence profiles.
1H NMR and ESI-MS Studies. The conversion of Hg2L2/L
was also verified from 1H NMR titration spectra (Figure S1 in the
Supporting Information). The proton chemical shifts of the
coumarin ring in L were distinctly shifted downfield upon the
6547
dx.doi.org/10.1021/ic200181p |Inorg. Chem. 2011, 50, 6543–6551

Inorganic Chemistry
ARTICLE
Figure 6. Reversible interconversions between Hg2L2 and L upon modulation of Cys/Hg2+ Color code: pale, H; gray, C; red, O; blue, N; yellow, S;
green, Cl; pink, Hg).
Figure 7. Fluorescent intensity of Hg2L2 (5 μM) at 525 nm in DMSO/H2O (1:99, v/v) (a) with the addition of Cys, Hcy, and GSH (10 μM each) and
(b) with the addition of (1) none, (2) 10 μM Cys, (3) other amino acids including glycine, alanine, valine, leucine, isoleucine, methionine, proline,
tyrosine, lysine, histidine, serine, and tryptophan (each 10 μM), and (4) 10 μM Cys in the presence of other amino acids.
addition of 2 equiv of HgCl2 and returned to the original values
with the subsequent addition of 2.5 equiv of Cys. This reversible
interaction was further evidenced by ESI-MS detection of a
mixed solution. The peak at m/z 1182.7 corresponding to
[Hg2L2 + H]+ (calcd m/z 1180.9) was found after 1 equiv of
HgCl2 was added to the aqueous solution of L (Figure S2 in the
Supporting Information), and the peak at m/z443.0 corresponding
to [2Cys + Hg2+-H]+ (calcd m/z 442.9) was observed upon the
addition of 2 equiv of Cys (Figure S3 in the Supporting
Information). Thus, 1H NMR and ESI-MS results firmly support
the conclusion that the interconversion of Hg2L2/L can be
modulated by the decomplexation/complexation interaction
upon the addition of Cys/Hg2+.
Figure 8. Confocal images of HEK 293 cell lines. Upper panel: (a)
Quantitative Determination of Cys. Both L and Hg2L2 exhibit
Bright-field image; (b) fluorescence imaging after incubation with L
excellent water solubility and biocompatibility, as well as good
[10 μM, H2O/DMF (7:3, v/v), 50 mM Tris-HCl, pH = 7.4] for 15 min;
(c) overlay of the bright-field and fluorescent images. Lower panel:
Fluorescence imaging of (d) HEK 293 cells after incubation with L
stability under a wide pH span from 6 to 10 covering physiological
conditions (Figure S4 in the Supporting Information). These
features of L and Hg2L2 facilitate their practical applications for
the determination of Cys. Hg2L2 displays a high sensitivity to Cys
(the reaction can be completed within several seconds), which can
be used to create a calibration curve for a quick quantitative
measurement of Cys. Therefore, Cys was added at different
concentrations from 0 to 30 μM, and the fluorescence intensity
of Hg2L2 (5.0 μM) at 525 nm was recorded to generate a
calibration curve. A good linearity (R = 0.995) was found between
the fluorescence intensity of the solution and the Cys concentra-
tion (inset of Figure 4b). This linear fitting analysis reveals that
Hg2L2 is suitable for determining Cys from 0.5 to 30 μM. When
constant amounts of Cys (10 μM) and HgCl2 (10 μM) were
alternatively added to an aqueous solution of Hg2L2, the fluores-
cence intensity at 525 nm varied with alternating increases and
decreases to over 95%. Such reversible interconversions between
(10 μM), (e) HEK 293 cells in part d after treatment with HgCl2
(100 μM) for 12 min, and (f) HEK 293 cells in part e after incubation
with Cys (200 μM) for 12 min.
Hg2L2 and L can be repeated in at least five cycles by modulation
of the Cys and Hg2+ added. Their corresponding fluorescence
changes are also given in Figure 5.
High Selectivity to Mercapto Biomolecules. With the excep-
tion of Cys, other mercapto biomolecules such as Hcy and GSH
also induced similar variations in the absorption and fluorescence
spectra of Hg2L2. Among these, Cys led to the largest fluorescence
increase, whereas Hcy and GSH gave relatively smaller increases in
fluorescence (Figure 7a). For practical applications, an important
consideration is its selective detection in the presence of other
amino acids without thiol groups, such as glycine, alanine, valine,
6548
dx.doi.org/10.1021/ic200181p |Inorg. Chem. 2011, 50, 6543–6551

Inorganic Chemistry
ARTICLE
leucine, isoleucine, methionine, proline, tyrosine, lysine, histidine,
serine, and tryptophan. As shown in Figure 7b, the addition of
other competitive amino acids to the solution of Hg2L2 did not
result in any notable fluorescent increases. However, when Cys
was added to the aqueous solution of Hg2L2 containing other
competitive amino acids, an obvious increase in fluorescence was
observed. This result indicates that Hg2L2 can detect Cys with a
high selectivity over other coexisting amino acids.
Cell Culture. The foreign uptake of mercapto biomolecules in
HEK293 cells (human embryotic kidney cell, without thiols) and
intracellular mercapto biomolecules in SYS5 cells (human neu-
roblastoma SH cell) was examined using confocal fluorescence
imaging. As shown in Figure 8, HEK293 cell lines after incuba-
tion with L (10 μM) for 15 min were conveniently imaged using a
confocal fluorescence microscope. The matched fluorescence
and bright-field images elicited the intracellular uptake of L by
HEK293 cells (Figure 8aꢀc). The HEK 293 cells incubated with
L (10 μM; Figure 8d) were treated with HgCl2 (100 μM) for
12 min. Their fluorescence images became dim (Figure 8e),
implying that the intracellular uptake of Hg2+ ions complexed
with L yielded nonfluorescent Hg2L2. Upon further incubation
of these cells with foreign uptake of Cys (200 μM) for 12 min,
green fluorescence imaging was recovered (Figure 8f). The
recurrent imaging indicated that the uptake of Cys resulted in
the decomplexation of intracellular Hg2L2 to fluorescent L.
Through reversible fluorescence imaging, intracellular intercon-
version of Hg2L2/L was explicitly illustrated. Therefore, the
ONꢀOFFꢀON fluorescence imaging of L was accomplished
in HEK293 cell lines by the intracellular complexation/decom-
plexation interaction modulated by Hg2+/Cys.
Intracellular thiol-enriched cells such as SYS5 also induced
nonfluorescent Hg2L2 to give distinct fluorescence imaging like
Cys did (Figure 9). After SYS5 cell lines were incubated with
Hg2L2 (10 μM) for 1 min, referring to its bright-field image
(Figure 9a), only a weak fluorescence emission was imaged in
certain loci of the cell (Figure 9b). After 6 min, the intensity of
imaged fluorescence emission increased and the area expanded
(Figure 9d). After 15 min, the bright emission from the whole cell
was imaged (Figure 9f). Likewise, the initially weak fluorescence
emission originated from the decomplexation of Hg2L2 in the
thiol-enriched portion of the SYS5 cells. As the incubation time
increased, the concentration of intracellular L increased and
subsequently the concentrated L diffused to the whole cell,
analogous to the uptake of free L in HEK293 cells. These results
indicate that intracellular fluorescent recurrence could also be
accomplished by thiol-enriched cells instead of foreign Cys,
which provides the possibility of exploring the self-detoxification
process of Hg2+ ions in living cells.
On the basis of the result of confocal fluorescence imaging
above, a pair of sensors, L/Hg2L2, was further used to illustrate
the intracellular self-detoxification process by the toxic uptake of
Hg2+ ions (Figure 10). The SYS5 cell lines incubated with Hg2L2
(10 μM) for 10 min were treated with HgCl2 (10 μM; entry 1).
Figure 9. Confocal images of SYS5 cell lines. Bright-field image (a) and
fluorescence images after incubation with Hg2L2 [10 μM, H2O/DMF
(7:3, v/v), 50 mM Tris-HCl, pH = 7.4] for 1 min (b), 3 min (c), 6 min
(d), 9 min (e), and 15 min (f).
Figure 10. Confocal fluorescence imaging in the SYS5 cell lines incubated with Hg2L2 [10 μM, H2O/DMF (7/3, v/v), 50 mM Tris-HCl, pH = 7.4]
upon the multistep addition of HgCl2 (10 μM each) with regular time intervals (entries 1ꢀ10). Upper: In situ determination of their fluorescence
imaging (fluorescence changes of region 4 were recorded for comparison). Lower: Their corresponding changes in the fluorescence intensity with time.
6549
dx.doi.org/10.1021/ic200181p |Inorg. Chem. 2011, 50, 6543–6551

Inorganic Chemistry
ARTICLE
With increasing incubation time, the intracellular fluorescence
intensity was enhanced because the strongly fluorescent L
(released from Hg2L2) diffused to the whole cell (entry 2). With
continued treatment of the SYS5 cell lines above with HgCl2 (10
μM), their fluorescence images became dim quickly (entry 3).
Surprisingly, after about 2 min, the intracellular fluorescence was
revived to its original intensity again (entry 4). We repeated such
experiments to find that the intracellular fluorescent decreases
(entries 1, 3, 5, 7, and 9) and increases (entries 2, 4, 6, 8, and 10)
could be recurrent at least 10 times. For illustration of an
alternate fluorescence variation, the in situ determination of their
corresponding fluorescence in vivo in the SYS5 cell lines was
also shown in Figure 10. Obviously, the reversible intracellular
fluorescence revival upon the multistep addition of HgCl2 was
relevant to the increase of the intracellular active mercapto
biomolecules. As we know from cell biology, upon intracellular
uptake of some toxic species such as Hg2+ ions, the cells will start
the transduction and expression of antigene, and those detox
species are biosynthesized and secreted into the cells. SYS5, as
a kind of intracellular thiol-enriched cells, can always secrete
mercapto biomolecules for Hg2+ complexation when the toxic
Hg2+ ions are added repeatedly. These mercapto biomolecules
can form a stable complex with Hg2+ ions and excrete out of the
cells via metabolism to complete a self-detoxification process.
Therefore, L/Hg2L2 can be applied to track the intracellular self-
detoxification process, which will give a simple and convenient
method to study the cell toxicity. This also provides a new way of
communicating with living systems and adjusting and controlling
the chemical species inside cells that are critical for intracellular
engineering, manipulation, and the probe.19
’ ACKNOWLEDGMENT
This work was supported by the National Natural Science
Foundation of China (Grants 20953002, 20903110, and
60978034), the Research Grants Council of the Hong Kong
SAR (Project CityU 123607), and the National Basic Research
Program of China (Grant 2007CB936001).
’ REFERENCES
(1) Shahrokhian, S. Anal. Chem. 2001, 73, 5972–5978.
(2) (a) Seshadri, S.; Beiser, A.; Selhub, J.; Jacques, P. F.; Rosenberg,
I. H.; D’Agostino, R. B.; Wilson, P. W.; Wolf., P. A. N. Engl. J. Med. 2002,
346, 476–483. (b) Ueland, P. M.; Vollset, S. E. Clin. Chem. 2004,
50, 1293–1295.
(3) (a) Hassan, S. S. M.; Rechnitz, G. A. Anal. Chem. 1982,
54, 1972–1976. (b) Hwang, C.; Sinskey, A. J.; Lodish, H. F. Science
1992, 257, 1496–1502. (c) Hong, R.; Han, G.; Fernandez, J. M.; Kim,
B.-J.; Forbes, N. S.; Rotello, V. M. J. Am. Chem. Soc. 2006, 128,
1078–1079.
(4) Guan, X.; Hoffman, B.; Dwivedi, C.; Matthees, D. P. J. Pharm.
Biomed. Anal. 2003, 31, 251–261.
(5) Capitan, P.; Malmezat, T.; Breuille, D.; Obled, C. J. Chromatogr.,
B: Biomed. Sci. Appl. 1999, 732, 127–135.
(6) Qian, X.-X.; Nagashima, K.; Hobo, T.; Guo, Y.-Y.; Yamaguchi, C.
J. Chromatogr., A 1990, 515, 257–264.
(7) (a) Melnyk, S.; Pogribna, M.; Pogribny, I.; Hine, R. J.; James, S. J.
J. Nutr. Biochem. 1999, 10, 490–497. (b) Hiraku, Y.; Murata, M.;
Kawanishi, S. Biochim. Biophys. Acta 2002, 1570, 47–52.
(8) (a) Wang, W.; Rusin, O.; Xu, X.; Kim, K. K.; Escobedo, J. O.;
Fakayode, S. O.; Fletcher, K. A.; Lowry, M.; Schowalter, C. M.;
Lawrence, C. M.; Fronczek, F. R.; Warner, I. M.; Strongin, R. M.
J. Am. Chem. Soc. 2005, 127, 15949–15958. (b) Li, H.; Fan, J.; Wang,
J.; Tian, M.; Du, J.; Sun, S.; Sun, P.; Peng, X. Chem. Commun.
2009, 5904–5906. (c) Maeda, H.; Matsuno, H.; Ushida, M.; Katayama,
K.; Saeki, K.; Itoh, N. Angew. Chem., Int. Ed. 2005, 44, 2922–2925.
(d) Kim, T.-K.; Lee, D.-N.; Kim, H.-J. Tetrahedron Lett. 2008,
49, 4879–4881. (e) Lin, W. Y.; Long, L. L.; Yuan, L.; Cao, Z. M.; Chen,
B. B.; Tan, W. Org. Lett. 2008, 10, 5577–5580. (f) Lee, K. S.; Kim, T. K.;
Lee, J. H.; Kim, H. J.; Hong, J. I. Chem. Commun. 2008, 6173–6175.
(g) Zhang, X. J.; Ren, X. S.; Xu, Q. H.; Loh, K. P.; Chen, Z. K. Org. Lett.
2009, 11, 1257–1260. (h) Wang, W.; Escobedo, J. O.; Lawrence, C. M.;
Strongin, R. M. J. Am. Chem. Soc. 2004, 126, 3400–3401. (i) Chen, X.;
Zhou, Y.; Peng, X.; Yoon, J. Chem. Soc. Rev. 2010, 39, 2120–2135. (j)
Chen, X.; Ko, S.-K.; Kim, M. J.; Shin, I.; Yoon, J. Chem. Commun. 2010,
46, 2751–2753.
(9) (a) Berezin, M. Y.; Achilefu, S. Chem. Rev. 2010, 110
2641–2684. (b) Lee, J. H.; Lim, C. S.; Tian, Y. S.; Han, J. H.; Cho,
B. R. J. Am. Chem. Soc. 2010, 132, 1216–1217. (c) Bouffard, J.; Kim, Y.;
Swager, T. M.; Weissleder, R.; Hilderbrand, S. A. Org. Lett. 2008,
10, 37–40. (d) Tang, B.; Xing, Y.; Li, P.; Zhang, N.; Yu, F.; Yang, G.
J. Am. Chem. Soc. 2007, 129, 11666–11667. (e) Zhang, M.; Yu, M.; Li, F.;
Zhu, M.; Li, M.; Gao, Y.; Li, L.; Liu, Z.; Zhang, J.; Zhang, D.; Yi, T.;
Huang, C. J. Am. Chem. Soc. 2007, 129, 10322–10323.
(10) (a) Zheng, H.; Qian, Z.-H.; Xu, L.; Yuan, F.-F.; Lan, L.-D.; Xu,
J.-G. Org. Lett. 2006, 8, 859–861. (b) Wu, J.-S.; Hwang, I.-C.; Kim, K. S.;
Kim, J. S. Org. Lett. 2007, 9, 907–910. (c) Xiang, Y.; Tong, A.; Jin, P.; Ju,
Y. Org. Lett. 2006, 8, 2863–2866. (d) Yang, Y.-K.; Yook, K.-J; Tae, J.
J. Am. Chem. Soc. 2005, 127, 16760–16761. (e) Xiang, Y.; Tong, A. Org.
Lett. 2006, 8, 1549–1552. (f) Kwon, J. Y.; Jang, Y. J.; Lee, Y. J.; Kim,
K. M.; Seo, M. S.; Nam, W.; Yoon, J. J. Am. Chem. Soc. 2005,
127, 10107–10111. (g) Dujols, V.; Ford, F.; Czarnik, A. W. J. Am. Chem.
Soc. 1997, 119, 7386–7387. (h) Kim, H. N.; Lee, M. H.; Kim, H. J.; Kim,
J. S.; Yoon, J. Chem. Soc. Rev. 2008, 37, 1465–1472. (i) Chen, X.; Nam,
S.-W.; Jou, M. J.; Kim, Y.; Kim, S.-J.; Park, S.; Yoon, J. Org. Lett. 2008,
10, 5235–5238.
4. CONCLUSIONS
In summary, a coumarin-derived imine (L) was synthesized
and its complex, Hg2L2, was developed as a reversible fluorescent
probe for selective sensing of mercapto biomolecules such as
Cys, Hcy, and GSH. Hg2L2 exhibited a series of advantages as
fluorescent probes including highly sensitive detection, fluores-
cence OFFꢀON, reversible interconversion, good water solubi-
lity, high quantum yield of L (0.50), longer emission wavelength
(525 nm), and a wide pH span (6ꢀ10). The interconversion
of Hg2L2 and L in aqueous solution via the decomplexation/
complexation was definitely verified from crystal structures
and ESI-MS, NMR, UVꢀvis, and fluorescence spectra. Confocal
fluorescence imaging in the SYS5 cells reveals that L/Hg2L2
can be applied to monitor the intracellular self-detoxification
process to avoid toxic intracellular uptake. This will provide a
new strategy for the design of reversible fluorescent probes
to study the self-detoxification mechanism of heavy metals in
living cells.
’ ASSOCIATED CONTENT
S
Supporting Information. X-ray crystallographic data of
b
ligand L and complex Hg2L2 in CIF format, additional spectra,
NMR copies, and visual fluorescence imaging. This material is
available free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: wangpf@mail.ipc.ac.cn.
(11) (a) Wallace, K. J.; Fagbemi, R. I.; Folmer-Andersen, F. J.; Morey,
J.; Lyntha, V. M.; Anslyn, E. V. Chem. Commun. 2006, 3886–3888.
6550
dx.doi.org/10.1021/ic200181p |Inorg. Chem. 2011, 50, 6543–6551

Inorganic Chemistry
ARTICLE
(b) Dahiya, P.; Kumbhakar, M.; Maity, D. K.; Mukherjee, T.; Mittal, J. P.;
Tripathi, A. B. R.; Chattopadhyay, N.; Pal, H. Photochem. Photobiol. Sci.
2005, 4, 100–105. (c) Murata, C.; Masuda, T.; Kamochi, Y.; Todoroki,
K.; Yoshida, H.; Nohta, H.; Yamaguchi, M.; Takadate, A. Chem. Pharm.
Bull. 2005, 53, 750–758. (d) Kim, H. M.; Yang, P. R.; Seo, M. S.; Yi, J.-S.;
Hong, J. H.; Jeon, S.-J.; Ko, Y.-G.; Lee, K. J.; Cho, B. R.
J. Org. Chem. 2007, 72, 2088–2096.
(12) (a) Hagen, V.; Dekowski, B.; Nache, V.; Schmidt, R.; Geibler,
B.; Lorenz, D.; Eichhorst, J.; Keller, S.; Kaneko, H.; Benndorf, K.;
Wiesner, B. Angew. Chem., Int. Ed. 2005, 44, 7887–7891. (b) Secor, K. E.;
Glass, T. E. Org. Lett. 2004, 6, 3727–3730. (c) Akita, S.; Umezawa, N.;
Higuchi, T. Org. Lett. 2005, 7, 5565–5568. (d) Hirano, T.; Hiromoto, K.;
Kagechika, H. Org. Lett. 2007, 9, 1315–1318. (e) Wang, J.; Xie, J.;
Schultz, P. G. J. Am. Chem. Soc. 2006, 128, 8738–8739. (f) Chattopadhyay,
N.; Mallick, A.; Sengupta, S. J. Photochem. Photobiol. A 2006, 177, 55–60.
(g) Gawley, R. E.; Shanmugasundarama, M.; Thorne, J. B.; Tarkk, R. M.
Toxicon 2005, 45, 783–787.
(13) Wu, J.; Liu, W.; Zhuang, X.; Wang, F.; Wang, P.; Tao, S.; Zhang,
X. H.; Wu, S.; Lee, S. Org. Lett. 2007, 9, 33–36.
(14) Sheldrick, G. M. Acta Crystallogr., Sect. A 2008, 64, 112–122.
(15) (a) Paeker, C. A.; Rees, W. T. Analyst 1960, 85, 587–600.
(b) Gabe, T.; Urano, Y.; Kikuchi, K.; Kojima, H.; Nagano, T. J. Am.
Chem. Soc. 2004, 126, 3357–3367.
(16) (a) Schwarzenbach, G. In Soft and Hard Acids and Bases;
Pearson., R. G., Ed.; John Wiley: New York, 1973; p 20. (b) Kolthoff,
I. M. Treatise Anal. Chem. 1979, 129.
(17) (a) Nolan, E. M.; Burdette, S. C.; Harvey, J. H.; Hilderbrand,
S. A.; Lippard, S. J. Inorg. Chem. 2004, 43, 2624–2635. (b) Nolan, E. M.;
Lippard, S. J. Inorg. Chem. 2004, 43, 8310–8317. (c) Nolan, E. M.;
Lippard, S. J. J. Am. Chem. Soc. 2003, 125, 14270–14271.
(18) (a) Coskun, A.; Deniz, E.; Akkaya, E. U. Org. Lett. 2005,
7, 5187–5189. (b) Coskun, A.; Akkaya, E. U. J. Am. Chem. Soc. 2006,
128, 14474–14475.
(19) Win, M. N.; Smolke, C. D. Science 2008, 322, 456–460.
6551
dx.doi.org/10.1021/ic200181p |Inorg. Chem. 2011, 50, 6543–6551