Coumarin-based thiol chemosensor: Synthesis, turn-on mechanism, and its biological application

Article abstract of DOI:10.1021/ol2001864

A new chemodosimetric probe (1) is reported that selectively detects thiols over other relevant biological species by the turning on of its fluorescence through a Michael type reaction. The fluorogenic process upon its reaction was revealed to be mediated by intramolecular charge transfer, as confirmed by time-dependent density functional theory calculations. The application of probe 1 to cells is also examined by confocal microscopy, and its cysteine preference was observed by an ex vivo LC-MS analysis of the cellular metabolite.

Full text of DOI:10.1021/ol2001864

ORGANIC
LETTERS
2011
Vol. 13, No. 6
1498–1501
Coumarin-Based Thiol Chemosensor:
Synthesis, Turn-On Mechanism, and Its
Biological Application
Hyo Sung Jung,† Kyoung Chul Ko,‡ Gun-Hee Kim,§ Ah-Rah Lee,§ Yun-Cheol Na,
Chulhun Kang,*,§ Jin Yong Lee,*,‡ and Jong Seung Kim*,†
Department of Chemistry, Korea University, Seoul, 136-704, Korea, Department of
Chemistry, Sungkyunkwan University, Suwon 440-746, Korea, The School of East-West
Medical Science, Kyung Hee University, Yongin 446-701, Korea, and Seoul Center,
Korea Basic Science Institute, Seoul 136-713, Korea
kangch@khu.ac.kr; jinylee@skku.edu; jongskim@korea.ac.kr
Received January 21, 2011
ABSTRACT
A new chemodosimetric probe (1) is reported that selectively detects thiols over other relevant biological species by the turning on of its
fluorescence through a Michael type reaction. The fluorogenic process upon its reaction was revealed to be mediated by intramolecular charge
transfer, as confirmed by time-dependent density functional theory calculations. The application of probe 1 to cells is also examined by confocal
microscopy, and its cysteine preference was observed by an ex vivo LC-MS analysis of the cellular metabolite.
Thiols play important roles in the cellular antioxidant
defense system.1 Thiol groups play crucial roles in main-
taining biological redox homeostasis through the equili-
brium established between reduced free thiols (RSH) and
oxidized disulfides (RSSR).2 Moreover, their abnormal
levels are known to be associated with human diseases such
as slow growth, liver damage, skin lesions, Alzheimer’s
disease, cardiovascular disease, and coronary heart disease.3
Fluorescent probes possessinnate advantages over other
types of probes including their high sensitivity, specificity,
simplicity of implementation, and ability to allow real-time
monitoring due to their fast response times.4 Accordingly,
considerable effort has been made to develop fluorescent
probes,5 particularly for the detection of thiol species.6-12
However, the practical utility of most of these probes has
† Korea University.
‡ Sungkyunkwan University.
§ Kyung Hee University.
Korea Basic Science Institute.
(1) (a) Chen, X.; Zhou, Y.; Peng, X.; Yoon, J. Chem. Soc. Rev. 2010,
39, 2120–2135. (b) Fersht, A., Ed. Enzyme Structure and Mechanism, 2nd
ed.; Freeman, Co.: New York, 1984; pp 2-4.
(5) (a) Bozdemir, O. A.; Guliyev, R.; Buyukcakir, O.; Selcuk, S.;
Kolemen, S.; Gulseren, G.; Nalbantoglu, T.; Boyaci, H.; Akkaya, E. U.
J. Am. Chem. Soc. 2010, 132, 8029–8036. (b) Liu, B.; Tian, H. Chem.
Commun. 2005, 3156–3158. (c)Kim, S. K.; Lee, S. H.; Lee, J. Y.;Lee, J. Y.;
Bartsch, R. A.; Kim, J. S. J. Am. Chem. Soc. 2004, 126, 16499–16506.
(6) (a) Chen, X.; Ko, S.-K.; Kim, M. J.; Shin, I.; Yoon, J. Chem.
Commun 2010, 46, 2751–2753. (b) Lee, K. S.; Kim, T. K.; Lee, J. H.; Kim,
H. J.; Hong, J. I. Chem. Commun. 2008, 6173–6175. (c) Yi, L.; Li, H.;
Sun, L.; Liu, L.; Zhang, C.; Xi, Z. Angew. Chem., Int. Ed. 2009, 48, 4034–
4037. (d) Hong, V.; Kislukhin, A. A.; Finn, M. G. J. Am. Chem. Soc.
2009, 131, 9986–9994. (e) Shiu, H.-Y.; Chong, H.-C.; Leung, Y.-C.;
Wong, M.-K.; Che, C.-M. Chem.-Eur. J. 2010, 16, 3308–3313. (f)
Sreejith, S.; Divya, K. P.; Ajayaghosh, A. Angew. Chem., Int. Ed.
2008, 47, 7883–7887.
(2) (a) Meister, A. Methods Enzymol. 1995, 251, 3–7. (b) Finkelstein,
J. D.; Martin, J. J. Int. J. Biochem. Cell Biol. 2000, 32, 385–389. (c)
Deneke, S. M. Curr. Top. Cell Regul. 2000, 36, 151–180.
(3) (a) Townsend, D. M.; Tew, K. D.; Tapiero, H. Biomed. Pharmac-
other. 2003, 57, 145–155. (b) Shahrokhian, S. Anal. Chem. 2001, 73,
5972–5978. (c) Carmel, R., Jacobsen, D. W., Eds. Homocysteine in
Health and Disease; Cambridge University Press: Cambridge, U.K., 2001.
(4) (a) Kim, H. N.; Lee, M. H.; Kim, H. J.; Kim, J. S.; Yoon, J. Chem.
Soc. Rev. 2008, 37, 1465–1472. (b) Kim, J. S.; Quang, D. T. Chem. Rev.
2007, 107, 3780–3799. (c) Quang, D. T.; Kim, J. S. Chem. Rev. 2010, 110,
6280–6301.
r
10.1021/ol2001864
Published on Web 02/16/2011
2011 American Chemical Society

been thwarted by their cross-sensitivities toward other
biological analytes, narrow pH span,7 and slow response8
under physiological conditions. Therefore, it remains a
challenging task to develop highly selective and efficient
fluorescent probes for the detection of thiols. Recently, Lin
et al. reported a ketocoumarin-based thiol detection probe
featuring the 1,4-addition reaction of thiols to R,β-unsa-
turated ketones based on intramolecular charge transfer
(ICT).9 Liet al.10 and Wu etal.11 reported results regarding
cysteine (Cys) selective fluorescent chemosensors.
Herein, the design and synthesis of a new fluorogenic
probe (1) that emits fluorescence through its reaction with
thiols, showing a preference for Cys, are reported. Upon its
reaction with Cys, the representative thiol in this study, the
fluorogenic process was revealed to be mediated by ICT, as
confirmedbythe time-dependent density functionaltheory
(TDDFT) calculation performed in this study.
adducts, are depicted in Figures S1 and S2, respectively.
The signals of the H1-H3 protons of 1 appear at 7.65, 7.75,
and 6.99 ppm, respectively. Then, the 1H NMR spectrum
of 1 in a solution of D2O/CD3CN (3:7) was monitored
upon the addition of Cys at room temperature. As the
vinylic proton (H1) at 7.65 ppm disappeared, two new
peaks appeared at 4.35 and 4.27 ppm, obviously caused by
the formation of the 1-Cys adduct (Figure S2). The FAB-
MS spectrometry analysis of probe 1 treated with Cys in
aqueous solution (10 mM PBS buffer, pH 7.4, 10%
DMSO) also confirms the formation of the 1-Cys adduct.
The mass spectrum displayed a peak at m/z 533.41
[MþH]þ (Figure S21). This Michael type reaction was
further confirmed by the reaction of probe 1 with 2-mer-
captoethanol, whose detailed 1H NMR and FAB-MS data
are presented in Figures S3 and S22, respectively.
Figure 1. X-ray crystal structure of 1 and its fluorescent turn-on
function for thiols. All hydrogen atoms are omitted for clarity.
The thermal ellipsoids are shown at the 50% probability level.
Figure 2. (A) UV-vis and (B) fluorescence spectra of 1 (5.0 μM)
in aqueous solution (10 mM PBS buffer, pH 7.4, 10% DMSO)
upon addition of various concentrations of Cys (0-100 equiv).
Excitation at 440 nm (slit = 1.5/3).
Probe 1 was synthesized by the condensation of coumar-
inaldehyde and diethyl malonate, as shown in Scheme S1,
and its structural identification was confirmed by 1H
NMR, 13C NMR, and FAB-MS spectroscopy (Figures S20,
S23, and S24). As shown in Figure 1, Cys can react with the
β-carbon of R,β-unsaturated diesters via a Michael type
addition reaction to generate the 1-Cys adduct. The de-
tailed NMR spectra, including 2D HSQC (Hetero-
nuclear Single Quantum Coherence), for the 1 and 1-Cys
During the 1,4-addition reaction to 1, the conjugation of
the coumarin ring system to the R,β-unsaturated carbonyl
group was broken, resulting in changes in both the UV/vis
and fluorescence spectra. Indeed, as shown in Figure 2, the
addition of Cys to 1 induced a 62 nm hypsochromic shift of
the absorption maximum at λab = 426 nm, resulting in a
perceived color change from dark orange to green, while the
fluorescence maximum at λem = 502 was enhanced ca. 107-
fold (Φf = 0.3268; see Table S1). Homocysteine and glu-
tathione (Hcy and GSH) demonstrated similar characteris-
tics. In contrast, probe 1 exhibited nonfluorescence (Φf =
0.0056; see Table S1) due to the ICT from the fluorophore
(coumarin) to the nearby conjugated diester (vide infra).
To investigate the mechanism of the fluorescence enhance-
ment during the formation of the 1-thiol adduct, DFT
calculations were carried out for the 1-Cys adduct with
6-31G* basis sets using a suite of Gaussian 03 programs.12
The optimized structures of 1 and 1-Cys are shown in Figure
3. Probe 1 has a conjugated bridge (-CdC-) between the
coumarin and diester groups, while 1-Cys has a saturated
bridge (-C-C-). The calculated structure of 1 was in good
agreement with the experimental crystal structures shown in
Figure 1. It is of note that one of the two ester groups of 1 is
locatedinthesameplaneasthecoumarin ring system, unlike
the corresponding ester groups in 1-Cys. This structural
difference leads to the strong, anticipated ICT process in 1
(7) Tanaka, F.; Mase, N.; Barbas, C. F., II. Chem. Commun. 2004,
1762–1763.
(8) Pires, M. M.; Chmielewski, J. Org. Lett. 2008, 10, 837–840.
(9) Lin, W.; Yuan, L.; Cao, Z.; Feng, Y.; Long, L. Chem.-Eur. J.
2009, 15, 5096–5103.
(10) Li, H.; Fan, J.; Wang, J.; Tian, M.; Du, J.; Sun, S.; Sunb, P.;
Peng, X. Chem. Commun. 2009, 5904–5906.
(11) Wu, H.; Huang, C.; Cheng, T.; Tseng, W. Talanta 2008, 76, 347–
352.
(12) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.;
Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.;
Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson,
G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.;
Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai,
H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken,
V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev,
O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.;
Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakr-
zewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.;
Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.;
Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov,
B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.;
Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.;
Gonzalez, C.; Pople, J. A. Gaussian 03, revision C02; Gaussian Inc.:
Pittsburgh, PA, 2004.
Org. Lett., Vol. 13, No. 6, 2011
1499

Figure 5. (A) Time course of the fluorescence response of
probe 1 (5.0 μM) in aqueous solution (10 mM PBS buffer,
pH 7.4, 10% DMSO), regarding the presence of 100 equiv
of Cys, Hcy, or GSH, with an excitation at 440 nm. The
fluorescence intensity was recorded near 502 nm at room
temperature. (B) Fluorescence spectra of 1 (5.0 μM) in aqu-
eous solution (10 mM PBS buffer, pH 7.4, 10% DMSO) in
the presence of various amino acids, metals, ROS, and
glucose (100 equiv, respectively) at an excitation
Figure 3. Optimized structures of 1, 1-Cys, and the first excited
state of 1.
and weakened ICT in 1-Cys. In addition, as shown in
Figure 3, the optimized structure for the first excited state
of 1 shows that both ester groups have a greater planar
geometry with the coumarin plane than that of 1-Cys, which
inhibits the fluorescence emission from 1.
wavelength of 440 nm after 1 min.
(10 mM PBS buffer, pH 7.4, 10% DMSO), and the results
are depicted in Figure 5A. The time course of the fluores-
cence intensity in a mixture of Cys and 1 seems to follow a
single exponential function. Hcy and GSH also showed
increases in their fluorescence intensity, but at much lower
rates. Indeed, the kinetic analysis based on a single ex-
ponential decay model showed that 1 reacts with Cys 13-
and 21-fold faster than with Hcy and GSH, respectively
(Figure S5). This result leads to the conclusion that 1 has a
preference for Cys over other biologically important thiols.
We found that the fluorescence intensity is proportional to
the amount of Cys added at a submicromolar level, with a
coefficient of R = 0.99557 (Figure S9) and a detection limit
of 30 nM (Figure S10). In addition, as indicated in Figure
S11, the fluorescence change of 1 with Cys is independent of
the pH in the range of 6 to 11. These results again imply that
1 may prefer Cys over other thiols in biological systems.
To further investigate the preferences of probe 1 toward
different thiols in biological media, the fluorescence spec-
tral changes were examined during the reaction of 1 with
various biologically relevant analytes in aqueous solution
(10mM PBSbuffer, pH 7.4, 10% DMSO). Probe 1 showed
a marked enhancement in its fluorescence intensity upon
its selective reaction with thiol containing amino acids over
other amino acids (Ala, Arg, Asn, Asp, Gln, Gluc, Glu,
Gly, His, Ile, Leu, Lys, Met, Phe, Pro, Ser, Tau, Thr, Trp,
Figure 4. Calculated HOMOs and LUMOs of 1 (left column)
and 1-Cys (right column).
Detailed information regarding the marked fluorescence
enhancement upon the formation of 1-Cys can be obtained
from TDDFT calculations. The calculated excitation wa-
velength of 1-Cys was 355 nm, which was blue-shifted by
56 nm from that of 1 (411 nm). The magnitude of the
calculated blue shift is in agreement with the experimentally
observed one (62 nm). The HOMOs and LUMOs of 1 and
1-Cys are shown in Figure 4. The HOMOfLUMO transi-
tion contributed 100% and 97.4% to the excitation of 1 and
1-Cys, respectively. Thus, the electron density distribution
is confined to the LUMO after its excitation. From the
LUMOs, it is clear that the ICT takes place through a
conjugated bridge between the coumarin and diester groups
in 1, while this is obviously prohibited in 1-Cys due to the
nonconjugated bridge. This result supports the hypothesis
that the strong fluorescence emissions from the 1-Cys adduct
are due to its weakened ICT process relative to that in 1.
A comparison of the reactivity of 1 toward Cys, Hcy, or
GSH was performed by monitoring the fluorescence
changes of the reaction mixture in aqueous solution
Tyr, Val), biologically common metal ions (Kþ, Ca2þ
,
Mg2þ, Naþ, Zn2þ, Fe2þ, Fe3þ), oxidoreducing agents
(H2O2, NADH), and glucose (Figures 5B and S13). In
addition, competitive experiments for the reaction of 1
with Cys in the presence of other biologically relevant
analytes revealed no significant influence of them on the
Cys sensing (Figures S12 and S13).
Regarding the Cys preference of 1, it is interesting to
point out that the pKa of Cys (8.30) is lower than that of
Hcy (8.87) or GSH (9.20).13 Thus, under the experimental
(13) Iciek, M.; Chwatko, G.; Lorenc-Koci, E.; Bald, E.; Wlodek, L.
Acta Biochim. Polonica 2004, 51, 815–824.
1500
Org. Lett., Vol. 13, No. 6, 2011

conditions in this study (pH 7.4), the thiol of Cys is
presumably changed to the thiolate which becomes a better
nucleophile in Michael addition reactions. Therefore, it
can be assumed that, at higher pH, this selectivity would
disappear. In fact, the relative ratio of the fluorescence
intensities of probe 1 with GSH to that with Cys was
largely increased as the solution pH shifted to the basic side
(Figure S14), supporting the hypothesis that the Cys
selectivity of probe 1 may be partly due to the relatively
lower pKa value of Cys compared to that of the other
thiols. However, this explanation may be applicable to
many other thiol probes and not be unique to probe 1.
Another possible origin of this preference could be a steric
effect, according to a report that the thiol group of Cys is
sterically less hindered than those of Hcy and GSH.9
LC/MS spectroscopy was also employed to investigate
theCys preferenceoverHcy and GSH. Inthe copresence of
Cys, Hcy, and GSH (100 equiv each), 1 preferably reacted
with Cys to give 1-Cys as the major peak in the ion
monitoring chromatogram, whereas the peak intensities
of 1-Hcy and 1-GSH were relatively small (Figure S18).
These apparent differences might be caused by the differ-
ent efficiencies of ionization. Under similar LC-MS con-
ditions, a longer reaction time allows 1 to form significant
amounts of the Cys adduct, Hcy, and GSH, and each
corresponding adduct was reasonably detectable (Figure
S17). These results imply that the differences in the peak
intensities of the adducts in Figures S18 and S19 are not
caused by the differences in their ionization efficiencies in
the MScompartment in thisinstrument, butrather by their
reactivity differences shown in Figure 5A.
With this more practical evidence of the preference of 1
toward Cys taken into account, the reaction of cellular
metabolites with probe 1 was examined by LC-MS spec-
troscopy. To obtain the metabolites, the cytosolic cell
extracts were treated by removing the macromolecules
through acetone precipitation. After the reaction of 1 with
the prepared metabolite mixture, LC-MS analysis was
performed and parts of the resulting LC-MS profiles are
shown in Figure S19. The major fluorescent adduct was
apparently from 1-Cys. The profiles in Figure S19 are
similar to those in Figure S18. These results confirm that
the major species detected by LC-MS analysis is the Cys-1
adduct, although the most abundant thiol species within
the cells may be GSH, suggesting that 1 may prefer Cys
over other thiols in biological systems. A similar analysis of
the cellular adducts from the cells treated with probe 1
would provide pivotal evidence for its Cys selectivity.
However, this was not attempted, because the Cys pre-
ference of probe 1 is time-dependent and its reactions with
the intracellular thiol species could not be stopped during
preparation of the samples for analysis.
Figure 6. Confocal microscopic analysis of HepG2 cells treated
with compound 1. The images of the cells were obtained using
excitation at 488 nm and a long-path (>560 nm) emission filter.
For the NEM-treated samples, before the media were finally
replaced with PBS containing compound 1, the cells were
incubated with media containing NEM at various concentra-
tions for 1 h at 37 °C.
In order to further demonstrate that probe 1 can detect
intracellular thiols, confocal microscopy experiments were
conducted. Figure 6 shows the fluorescence images of
HepG2, a human hepatoma cell line, a selective thiol
modifying reagent. The fluorescence intensity decreased
upon the addition of NEM, inferring that the species
containing thiol groups, such as Cys, Hcy, GSH, or the
proteins in the cells are responsible for the fluorescence
enhancement. The incubation of probe 1 with dialyzed
cytosolic protein extract failed to show any intensity in the
fluorescence spectroscopic experiments (data not shown).
In conclusion, the new coumarin-based chemodosimeter
1 effectively and selectively recognizes thiols based on a
Michael type reaction, showing a preference for cysteine
over other biological materials including homocysteine
and glutathione. From the TDDFT calculations, it was
determined that the fluorescence amplification of 1 upon
its reaction with thiols is due to an ICT blocking event. The
cysteinepreference of1 over homocysteineand glutathione
was also proven by LC-MS spectroscopic methods using
the metabolite from the HepG2 cell line. In the confocal
microscopy study, 1 showed a marked fluorescence en-
hancement for cysteine over other thiol functionalized
amino acids.
Acknowledgment. This work was supported by the CRI
program (No. 2010-0000728) of the National Research
Foundation of Korea (J.S.K.). The work at SKKU was
supported by an NRF grant (No. 20100001630) funded by
MEST and a Samsung Research Fund, Sungkyunkwan
University, 2010.
Supporting Information Available. Text and figures
giving details of the UV-vis, fluorescence spectral data,
kinetic data, calculations, and characterization. This
material is available free of charge via the Internet at
http://pubs.acs.org.
Org. Lett., Vol. 13, No. 6, 2011
1501