Fluorescent probe for detection of fluoride in water and bioimaging in A549 human lung carcinoma cells

Article abstract of DOI:10.1039/b908745a

We have successfully developed a fluoride ion probe for fluorescence cell bioimaging - desirable properties include retention of the fluorophore inside cells, non-cytotoxicity to mammalian cells, appreciable solubility in water, and stoichiometric reactio

Full text of DOI:10.1039/b908745a

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Fluorescent probe for detection of fluoride in water and bioimaging
in A549 human lung carcinoma cellsw
a
a
Soon Young Kim,z Jongmin Park,z Minseob Koh,a Seung Bum Park*ab
and Jong-In Hong*a
Received (in Cambridge, UK) 5th May 2009, Accepted 9th June 2009
First published as an Advance Article on the web 29th June 2009
DOI: 10.1039/b908745a
We have successfully developed a fluoride ion probe for fluores-
cence cell bioimaging—desirable properties include retention of
the fluorophore inside cells, non-cytotoxicity to mammalian
cells, appreciable solubility in water, and stoichiometric reaction
with analytes.
our knowledge, no fluoride chemosensors or probes satisfying
the above-mentioned requirements have been reported.
In an effort to address these issues for biological
applications, we have explored the possibility of developing
a novel fluoride chemodosimeter in water. As part of previous
contributions to the development of chemodosimeters,8b,e,f,9
we recently demonstrated a resorufin-based chemodosimeter
incorporating a tert-butyldiphenylsilyl (TBDPS) moiety for
the detection of fluoride anions, with high selectivity in aqueous
solution.9b However, our initial attempt to apply the resorufin-
based system for the detection of fluoride ion in 100% water
failed due to its poor solubility in water.
Fluoride ions are widely used as an essential ingredient in
toothpaste and pharmaceutical agents and are even added to
drinking water owing to their tendency to prevent dental
caries1 and enamel demineralization resulting from wearing
orthodontic appliances; they are also used for the treatment of
osteoporosis.2 However, a high intake of fluoride may cause
fluorosis,3 and also lead to nephrotoxic changes4 and uro-
lithiasis in humans.5 In molecular and cell biology, NaF is
known to influence various cell signaling processes6 and to
induce apoptosis at high concentrations for 24 h in mammalian
cells.7 For these reasons, considerable effort has been devoted
to the development of novel methods for the detection of
fluorides, particularly NaF.8,9 In fact, the development of
receptor-based sensors for the detection of fluoride ions is
challenging owing to the small size, high electronegativity, and
high hydration enthalpy of fluoride ion. Therefore, while most
chemosensors recognize fluoride ions in a tetrabutylammonium
fluoride (TBAF) salt in organic solvents,8 only a few can detect
fluoride ions, as in NaF, in aqueous solutions.9 Recently,
In our endeavour to develop a fluoride sensor in water, we
moved forward to design a new fluoride detection system
working under physiological conditions. In order to improve
the water solubility, we planned to downsize our sensor system
and introduce hydrophilic moieties. This analysis led us to
design
a 7-hydroxycoumarin-based system containing a
tert-butyldimethylsilyl (TBDMS) moiety, TBMCA (1,
tert-butyldimethylsilyl 7-hydroxycourmarin-4-acetic acid
methyl ester), based on a previous approach.9d In addition,
we introduced a methyl ester group to 4-acetic acid on the
fluorescent coumarin moiety to increase water solubility and
to enhance cell permeability as well as to detain the fluoro-
phore inside the cell after hydrolysis by using a negatively
charged carboxylate group.10 TBMCA was prepared by silyl
protection of 7-hydroxycoumarin-4-acetic acid methyl ester
with TBDMSCl and imidazole in anhydrous DMF (see ESIw).11
7-Hydroxycoumarin-4-acetic acid methyl ester was syn-
thesized in accordance with a known procedure.12 Unfortunately,
the fluorescence intensity of TBMCA immediately showed
strong enhancement upon the exposure to phosphate buffered
saline (PBS) (Fig. S4, ESIw). In addition, the fluorescence
intensity of TBMCA was increased only 1.2-fold after treatment
of 1 mM NaF for 3 h in PBS (Fig. S1 and S4, ESIw), which is
consistent with a previous report.9d We initially conjectured
that the strong enhancement of fluorescence intensity in PBS
was due to the instability of the TBDMS moiety in TBMCA.
However, the fluorescence intensity of TBMCA was barely
changed in PBS over a 24 h period (Fig. S2, ESIw). Then, we
examined fluorescence emission changes of TBMCA as a
function of solvent polarity and observed increased fluorescence
intensity of TBMCA in accordance with increased polarity of
solvent (Fig. S3 and S4, ESIw). Based on this observation, we
postulated that the covalent bond character of the Si–O bond
of TBMCA is much weaker in water, due to its interaction
with water molecules, which renders mimicking the
Gabbaı successfully demonstrated a borane-based fluoride
¨
receptor working in DMSO–H2O (4 : 6, v/v).9a Swager, Yang
and their co-workers also developed fluoride-detecting systems
based on a coumarin moiety performing in organic solvents
such as THF, CH2Cl2 or in acetone–water (7 : 3, v/v) through
Si–O bond cleavage.8e,9d However, none of the fluoride chemo-
sensors meet the requirements for biological applications,
which are as follows; (1) to be able to selectively detect fluoride
ions in 100% water, (2) to be capable of permeating the cell
membrane, (3) to be non-toxic and display fluorescence upon
the detection of fluoride ions in cellular systems. To the best of
a Department of Chemistry, College of Natural Sciences,
Seoul National University, Seoul 151-747, Korea
b Department of Biophysics and Chemical Biology, College of Natural
Sciences, Seoul National University, Seoul 151-747, Korea.
E-mail: sbpark@snu.ac.kr, jihong@snu.ac.kr; Fax: +82 2 889 1568;
Tel: +82 2 880 9090 or +82 2 880 6682
w Electronic supplementary information (ESI) available: Synthesis
and characterization of TBMCA and TBPCA, fluorescence emission
and optical changes of TBPCA, CCK8 assay, quantification of
fluoride ions in cells, theoretical rationalization by computer-assisted
analysis. See DOI: 10.1039/b908745a
z These authors equally contributed to this work.
ꢀc
This journal is The Royal Society of Chemistry 2009
Chem. Commun., 2009, 4735–4737 | 4735

View Article Online
Scheme 1 Molecular structures of TBMCA and TBPCA, and the
sensing mechanism of probe TBPCA for the detection of NaF.
intramolecular charge transfer (ICT) events via stronger Si–O
bond polarization and leads to turn-on of fluorescence in
water without actual desilylation. Consequently, there was
only a marginal enhancement of fluorescence when TBMCA
was treated with fluoride anion in water (Fig. S1, ESIw).
Therefore, we decided to replace TBDMS (dimethyl) moiety
with a bulkier TBDPS (diphenyl) moiety to reduce the accessibility
of water molecules to the silicon atom.
Fig. 1 (a) Fluorescence emission change of TBPCA (2 mM) recorded
4 h after reaction with various concentrations of NaF (0–1.3 mM) in
10 mM HEPES buffer (pH 7.4) at 25 1C. (b) Comparison of fluores-
cence emission intensity of TBPCA (2 mM) after 4 h for various
sodium salts (1 mM) in 10 mM HEPES buffer (pH 7.4) at 25 1C
(blue line represents NaF).
TBPCA (2, tert-butyldiphenylsilyl 7-hydroxycourmarin-4-
acetic acid methyl ester) was prepared using TBDPSCl by the
same procedure as that for TBMCA. TBPCA was prepared for
the selective turn-on of quenched fluorescence by an ICT
mechanism when 7-hydroxycoumarin 3 was released upon
the attack of fluoride ion on the silyl ether moiety
(Scheme 1). The fluorescence intensity of TBPCA is also
affected by the solvent polarity, but significantly less than that
of TBMCA (Fig, S3 and S4, ESIw). In terms of sensitivity
toward fluoride ion, we observed more than 4-fold enhancement
of fluorescence intensity of TBPCA after treatment of 1 mM
NaF for 3 h in PBS (Fig. S1, ESIw). Therefore, we selected
TBPCA as a fluoride sensor with excellent properties for the
application in physiological conditions.
of fluoride in aqueous environments; therefore, an extended
incubation time is required for the consistent determination of
fluoride concentration through sufficient reaction of fluoride
ions with TBPCA.9b For instance, the fluorescence emission
intensity of TBPCA was maximized only 4 h after exposure to
1 mM NaF in the HEPES buffer at 25 1C (Fig. S6, ESIw). The
selectivity of TBPCA for Fꢁ was confirmed by treatment of
various anions (1 mM) as sodium salts such as Clꢁ, Brꢁ, Iꢁ,
AcOꢁ, NO3ꢁ, N3ꢁ, H2PO4ꢁ. As shown in Fig. 1b, only NaF
can efficiently exhibit fluorescence emission (lex: 375 nm) and
the fluorescence intensity in the case of NaF is more than seven
times that produced in the case of the other anions. Fig. 2
clearly shows the difference between the optical image of
TBPCA in the presence of NaF and UV light (lex: 365 nm)
and the optical image in the presence of other anions. Finally,
we examined whether TBPCA can influence the cell viability in
mammalian cells. After the treatment of the A549 human
epithelial lung carcinoma cell line with TBPCA (20 mM) for
24 h, there was no reduction in the cell viability: the cell
The enhanced fluorescence intensity of TBMCA in water,
but not in the case of TBPCA, is consistent with the computer-
assisted rationalization with ab initio calculation and molecular
modelling (see ESIw).13 In fact, ab initio calculation intimates
the importance of the electronic effect by the TBDPS and
TBDMS group in polar solvents; the Si–O polarity difference
(|DQ| = 0.021 C) and the Si–O bond length difference (0.007 A)
of TBPCA between hexane and water were significantly
smaller than those (|DQ| = 0.103 C, 0.009 A) of TBMCA.
In addition, the water-accessible area on silicon atom of
TBPCA was 0.23 A2, which is significantly smaller than that
(1.67 A2) of TBMCA, as calculated by molecular modelling
software, Naccess, which means that the Si–O bond of
TBMCA is quite accessible to water. These results also support
our initial rationale above.
We confirmed the linearity of the fluorescence emission
intensity (lem: 461 nm) of TBPCA (2 mM) relative to fluoride
concentrations in HEPES buffer at 25 1C (Fig. 1(a)). On the
basis of this calibration curve at 461 nm, the concentration of
NaF can be confidently predicted by measuring the fluores-
cence emission of TBPCA. Even though fluoride anions have
high selectivity and affinity towards the silicon atoms of
TBDPS, it is difficult to overcome the strong hydration effect
Fig. 2 Optical changes in fluorescence emission of TBPCA (2 mM)
when TBPCA is excited by UV irradiation (lex: 365 nm) after 4 h for
various sodium salts (1 mM) in 10 mM HEPES buffer (pH 7.4) at
25 1C. ꢁLeft to right: TBPCA, Fꢁ, Clꢁ, Brꢁ, Iꢁ, AcOꢁ, NO3ꢁ, N3
,
ꢁ
H2PO4 (sodium salts).
ꢀc
This journal is The Royal Society of Chemistry 2009
4736 | Chem. Commun., 2009, 4735–4737

View Article Online
viability was measured by considering the mitochondrial
function using Cell Counting Kit-8 (CCK8),14 and this result
indicates the possibility of using TBPCA for bioimaging live
cells in aqueous media.15
This work was supported by the KRF (Grant No. KRF-
2006-312-C00592), Seoul R&BD, the KOSEF, and the WCU
program through the KOSEF funded by the Korean Ministry
of Education, Science, and Technology (MEST). S. Y. K., J. P.
and M. K. are grateful for the fellowship award of the BK 21
Program and the Seoul Science Fellowship. We would like to
thank Dong-Seon Lee and Sangyun Lee for helpful
discussions regarding calculations.
TBPCA fulfills the requirements for displaying fluorescence
in in vitro cell imaging: it can be retained in a cell, it is non-
cytotoxic, and it can exhibit fluorescence upon sensing the
appropriate physiological conditions. On the basis of these
properties of TBPCA, which indicate that this probe is a
selective chemodosimeter for fluoride anions, we explored
the possibility of its use in biological systems by its application
to A549 human lung carcinoma cell lines. The addition of
50 mM NaF to A549 cells loaded with TBPCA (20 mM) leads
to a significant increase in the fluorescence intensity as
compared to control experiments (Fig. 3(b) and (c)). Due to
the slow rate of this reaction, the TBPCA-loaded A549 cells
with NaF have to be incubated for 3 h at 37 1C to obtain the
maximum fluorescence intensity; incubation for more than 3 h
causes the deterioration of the fluorescence signal.
Notes and references
1 (a) K. L. Kirk, Biochemistry of the Halogens and Inorganic Halides,
Plenum Press, New York, 1991, p. 58; (b) H. S. Horowitz, J. Public
Health Dent., 2003, 63, 3–8.
2 (a) B. L. Riggs, Bone and Mineral Research, Annual 2, Elsevier,
Amsterdam, 1984, pp. 366–393; (b) J. R. Farley, J. E. Wergedal
and D. J. Baylink, Science, 1983, 222, 330–332; (c) M. Kleerekoper,
Endocrinol. Metab. Clin. North Am., 1998, 27, 441–452.
3 (a) A. Wiseman, Handbook of Experimental Pharmacology XX/2,
Springer-Verlag, Berlin, 1970, Part 2, pp. 48–97; (b) J. A. Weatherall,
Pharmacology of Fluorides. in Handbook of Experimental
Pharmacology XX/1, Springer-Verlag, Berlin, 1969, Part 1,
pp. 141–172; (c) R. H. Dreisbuch, Handbook of Poisoning, Lange
Medical Publishers, Los Altos, CA, 1980.
4 M. L. Cittanova, B. Lelongt and M. C. Verpont, Anesthesiology,
1996, 84, 428–435.
5 P. P. Singh, M. K. Barjatiya, S. Dhing, R. Bhatnagar, S. Kothari
and V. Dhar, Urol. Res., 2001, 29, 238–244.
6 (a) T.-J. Chen, T.-M. Chen, C.-H. Chen and Y.-K. Lai, J. Cell.
Biochem., 1998, 69, 221–231; (b) M. H. Arhima, O. P. Gulati and
S. C. Sharma, Phytother. Res., 2004, 18, 244–246.
7 (a) C. D. Anuradha, S. Kanno and S. Hirano, Arch. Toxicol., 2000,
74, 226–230; (b) M. Refsnes, P. E. Schwarze, J. A. Holme and
M. Lag, Hum. Exp. Toxicol., 2003, 22, 111–123; (c) M. Matsui,
M. Morimoto, K. Horimoto and Y. Nishimura, Toxicol. In vitro,
2007, 21, 1113–1120.
To take advantage of the chemodosimeter, TBPCA was also
used for the quantification of fluoride ions in the cells. After
3 h incubation of the A549 cells with NaF (50 mM) under the
physiological conditions, the cells were harvested and
thoroughly washed. The harvested cells were sonicated and
centrifuged for the preparation of cell lysate in PBS buffer.
The resulting lysate was treated with TBPCA (2 mM) for
4 h at 25 1C to quantify the fluoride ions in the cell lysates;
the quantification was performed on the basis of the
fluorescence intensity and a standard curve prepared with
NaF-doped cell lysates. This new quantification method
reveals that 1.86 ꢂ 10ꢁ2 pmol fluoride anion is present
in the cytoplasm per cell under physiological conditions
(Fig. S12, ESIw).
8 (a) C. B. Black, B. Andrioletti, A. C. Try, C. Ruiperez and
J. L. Sessler, J. Am. Chem. Soc., 1999, 121, 10438–10439;
(b) S. Yamaguchi, S. Akiyama and K. Tamao, J. Am. Chem. Soc.,
2000, 122, 6793–6794; (c) E. J. Cho, J. W. Moon, S. W. Ko, J. Y. Lee,
S. K. Kim, J. Yoon and K. C. Nam, J. Am. Chem. Soc., 2003, 125,
In conclusion, we have successfully developed TBPCA as a
fluoride ion probe for fluorescence cell bioimaging with desired
properties, such as the detaining of the fluorophore inside a
cell, non-cytotoxicity to mammalian cells, fluorescence upon
sensing, appreciable solubility in water, and stoichiometric
reaction with analytes. We also demonstrated fluorescence cell
bioimaging using TBPCA for the detection of NaF in A549
human epithelial lung cancer cells under physiological
conditions. Moreover, TBPCA can be utilized for the
quantification of fluoride ions in living systems.
12376–12377; (d) M. Va
´
R. M. Pedrido, A. M. Gonza
zquez, L. Fabbrizzi, A. Taglietti,
lez-Noya and M. R. Bermejo, Angew.
´
Chem., Int. Ed., 2004, 43, 1962–1965; (e) T.-H. Kim and T. M. Swager,
Angew. Chem., Int. Ed., 2003, 42, 4803–4806; (f) X. Jiang,
M. C. Vieweger, J. C. Bollinger, B. Dragnea and D. Lee, Org. Lett.,
2007, 9, 3579–3582; (g) X. Y. Liu, D. R. Bai and S. Wang, Angew.
Chem., Int. Ed., 2006, 45, 5475–5478; (h) T. W. Hudnall and
F. P. Gabbaı, Chem. Commun., 2008, 4596–4597.
¨
11978–11986; (b) S. Y. Kim and J.-I. Hong, Org. Lett., 2007, 9,
¨
9 (a) T. W. Hudnall and F. P. Gabbaı, J. Am. Chem. Soc., 2007, 129,
3109–3112; (c) C.-W. Chiu and F. P. Gabbaı, J. Am. Chem. Soc.,
¨
2006, 128, 14248–14249; (d) X.-F. Yang, Spectrochim. Acta, Part A,
2007, 67, 321–326; (e) A. B. Descalzo, D. Jime
´
D. Beltran, P. Amoros, M. D. Marcos, R. Martı
nez, J. E. Haskouri,
nez-Manez and
´
´
´
´
J. Soto, Chem. Commun., 2002, 562–563; (f) M. Cametti and
K. Rissanen, Chem. Commun., 2009, 2809–2829.
10 C. C. Woodroofe and S. J. Lippard, J. Am. Chem. Soc., 2003, 125,
11458–11459.
11 Protective Groups in Organic Synthesis, ed. T. W. Greene and
P. G. M. Wuts, Wiley, New York, 3rd edn, 1999, pp. 113–148.
12 Q. Zhu, M. Uttamchandani, D. Li, M. L. Lesaicherre and
S. Q. Yao, Org. Lett., 2003, 5, 1257–1260.
13 The ab initio calculation of the Si–O bond in two different solvent
systems (water and n-hexane) was done using the Material Studio
4.2s program and the water-accessible area of the Si–O bond was
calculated using Naccess (V2.1.1) with stable conformers of
TBMCA and TBPCA. See ESIw for detailed procedures.
14 T. Mosmann, J. Immunol. Methods, 1983, 65, 55–63.
15 J. Park, H. Y. Lee, M.-H. Cho and S. B. Park, Angew. Chem., Int.
Ed., 2007, 46, 2018–2022.
Fig. 3 Brightfield image and fluorescence cell images of A549, human
epithelial lung carcinoma. (a) Bright-field image of A549 cells incu-
bated with TBPCA (20 mM) for 30 min and subsequently incubated for
3 h at 37 1C. (b) Fluorescence image of A549 incubated with TBPCA
(20 mM) for 30 min and subsequently incubated without NaF for 3 h at
37 1C. (c) Fluorescence image of A549 cells incubated with TBPCA
(20 mM) for 30 min and subsequently treated with 50 mM NaF for 3 h
at 37 1C. The scale bar represents 20 mm.
ꢀc
This journal is The Royal Society of Chemistry 2009
Chem. Commun., 2009, 4735–4737 | 4737