An interrupted PET coupled TBET process for the design of a specific receptor for Hg2+ and its intracellular detection in MCF7 cells

Article abstract of DOI:10.1039/c2cc34490d

A new coumarin-rhodamine conjugate constitutes a unique example of the interrupted PET coupled TBET response for developing an imaging reagent for determining the intracellular distribution of Hg²⁺ in MCF7 cells exposed to [Hg²⁺] as low as 2 ppb.

Full text of DOI:10.1039/c2cc34490d

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 9293–9295
www.rsc.org/chemcomm
COMMUNICATION
An interrupted PET coupled TBET process for the design of a specific
receptor for Hg2+ and its intracellular detection in MCF7 cellsw
Sukdeb Saha,a Prasenjit Mahato,a Mithu Baidya,b Sudip K. Ghosh*b and Amitava Das*a
Received 22nd June 2012, Accepted 26th July 2012
DOI: 10.1039/c2cc34490d
A new coumarin–rhodamine conjugate constitutes a unique example
of the interrupted PET coupled TBET response for developing an
imaging reagent for determining the intracellular distribution of
Hg2+ in MCF7 cells exposed to [Hg2+] as low as 2 ppb.
Small organic dye molecules have found wide applications as
fluorescent probes, biomarkers, solar energy harvesting, light
emitting materials, etc. Among various options, dye molecules
derived from rhodamine, boron dipyrromethane and cyanine
moieties are most prominent owing to their high fluorescence
Scheme 1
quantum yield and emission bands at the red region of the
spectrum. However, small Stokes shifts limit their applications
in certain cases.1 Forster Resonance Energy Transfer (FRET)
coumarin moiety subsequently initiates the TBET process and
fluorescence ON response of the Hg2+-bound acceptor rhod-
amine moiety. To the best of our knowledge, no such example
is there in the literature.
¨
is generally the most adopted methodology for addressing this
issue. For FRET to be effective, a substantial spectral overlap
and favourable orientations of the transition dipoles for the
donor emission and acceptor absorption bands are required,
which sometimes restrict the choice and the design of the
FRET-based probe molecules. In contrast, for a through-bond
energy transfer (TBET) system there is no such prerequisite,
though it exhibits fast energy transfer rates and large pseudo-
Stokes shifts. These provide the desired flexibility in designing
TBET-based cassettes which showed large quasi-Stokes
shifts.2 Despite such flexibility, reports on TBET cassettes
for developing biomarkers and/or fluorescent probes for
intracellular imaging applications are actually rare.3
Detailed synthetic procedures and various characterization
data for the new reagent L1 and the model coumarin derivative
(L2) (Scheme 1) are provided in the ESI.w Reagent L was
synthesised by following a literature procedure.6 L1 was found
to have a limited solubility in pure aqueous medium and thus
MeOH–aq. HEPES solution (1 : 1, v/v; pH 7.2) was used for
our studies, unless mentioned otherwise. Solution of L1 in this
aqueous buffer medium showed an absorption spectral band
at 320 nm, while no detectable absorbance at l > 400 nm was
observed (Fig. 1 inset). The electronic spectral band for the
model compound L2 also appeared at 322 nm. These spectral
observations established two crucial facts: firstly, there was no
or very little electronic interaction between two chromophores
Hg2+ and its various derivatives are considered to be the
most potent neurotoxins for human health.4 Hg2+ is also
known to have an adverse influence on the photosynthesis
process and thus, the global carbon cycle.5
Considering the fact that Hg2+ is known to quench mole-
cular fluorescence due to a facile spin–orbit coupling process,
the present example that allows the dual detection, colori-
metric as well as TBET-based fluorescence ON response with
large quasi-Stokes shifts has certain edge over other chemo-
sensors reported for Hg2+. Unique design of the receptor L1
helps in initiating an interrupted PET process upon binding of
Hg2+ to L1 and this fluorescence ON response of the donor
Fig. 1 Emission spectra and electronic spectra (inset i) of L, L1 and
L2 (6.25 Â 10À6 M) in MeOH–aq. HEPES buffer (1 : 1, v/v) at pH 7.2
and their respective protonated form at pH 2.0 for lExt = 320 nm.
Inset ii: solution colour for (a) L1, (b) L1 + Hg2+, (c) L1 + Mn+ and
(d) L1 + H+ ([L1] = 6.25 Â 10À6 M, [Mn+] or [Hg2+] = 1.2 mM,
[H+] = 20 mM).
a CSIR-Central Salt and Marine Chemicals Research Institute,
Bhavnagar - 364002, Gujarat, India. E-mail: amitava@csmcri.org
b Department of Biotechnology, Indian Institute of Technology,
Kharagpur, West Bengal - 721302, India
w Electronic supplementary information (ESI) available. See DOI:
10.1039/c2cc34490d
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 9293–9295 9293

View Article Online
in L1 and secondly, L1 was present almost exclusively in the
spirolactam form under the experimental conditions. A char-
acteristic signal at B64.16 ppm for the tertiary C21-atom in the
13C NMR spectrum for L1 also confirmed its spirolactam
structure (ESIw, Fig. S8).7
Table 1 Spectral bands for L, L1 and L2 in free and Hg2+-bound
forms in MeOH–aq. HEPES buffer (1 : 1, v/v) of pH 7.2
L
L1
L2
LÁHg2+
L1ÁHg2+
L2ÁHg2+
lAMbasx/nm
lEMmasx/nm
—
—
320
402
322
405
561
575
323, 561
583
324
410
Emission spectral studies for L1 (6.25 Â 10À6 M) in
MeOH–aq. HEPES buffer medium at different pH revealed that
the spirocyclic form of L1 was stable in the pH range 6.5–12.0
and only a weak luminescence band at 405 nm was observed for
lExt of 320 nm. Upon lowering the solution pH, a new distinct
band appeared at 585 nm due to the formation of the acyclic
xanthene form of L1. This was associated with an increase in
emission intensity at 402 nm with subsequent change in solution
colour (Fig. 1). Two fluorescence spectral bands at 405 and
585 nm for L1 at pH 2 appeared almost at the same position as
they appeared for two model compounds L2 and L, respectively,
at similar pH. Protonation at the secondary amine fragment of
L1 and L2 initiated an interrupted PET based process, which was
otherwise operational for L1 and L2. This caused an increase in
luminescence intensity at B402 nm at pH 2. For L and L1, the
new emission band at B580 nm (at pH 2 and lExt of B500 nm)
was attributed to the generation of the acyclic xanthene form.
However, the quantum yield for the 405 nm emission band
(lExt = 320 nm) was much higher for the corresponding
protonated form of L2 than that for L1 at pH 2.0 (Fig. 2B).
Further, L did not show any detectable emission at B580 nm
upon excitation at 320 nm (Fig. 2B). These collectively suggested
the possibility of an efficient energy transfer from the coumarin
moiety (donor) to the acyclic xanthene form of the rhodamine
moiety (acceptor). The inset of Fig. 2B clearly reveals that
these two chromophores do not have spectral band overlap.
All these support a TBET-based process.
emission bands appeared with maxima at B560 and 580 nm,
respectively. These signified the selective binding of Hg2+ to
the acyclic xanthene form of L1 in MeOH–aq. HEPES buffer
medium. Absorption band maxima at B322 nm for L1
remained almost unaffected upon binding to Hg2+ ions
(Table 1), while increase in emission band intensity at 402 nm
for L1 was also observed (lExt of 320 nm) (Fig. 2B and Table 1).
Such a behavior is characteristic of a PET based process.8
Subsequently, an intense emission band with maxima at
585 nm appeared upon binding to Hg2+ (Fig. 2B and
Ems
Table 1). Relative quantum yield (F = 0.69 at l
of 585 nm
Max
for lExt = 500 nm) for Hg2+ÁL1 was evaluated using the acyclic
xanthene form of rhodamine B as reference. The association
constant (KHg2+ Abs = (6.5 Æ 0.55) Â 104 MÀ1) for the Hg2+ÁL1
ÁL1
formation was evaluated from data obtained from the spectro-
photometric titration and corresponding Benesi–Hildebrand
(B–H) plot (ESIw, Fig. S9 and S10). A good linear fit of the
B–H plot supported the 1 : 1 binding stoichiometry.
Ems
Max
Evaluation of the relative emission quantum yield at l
of
402 nm (lExt of 320 nm) for L2 (F = 0.007) and Hg2+ÁL2
(F = 0.8) revealed an interruption of the PET process in
Hg2+ÁL2, which was otherwise operational for free L2 involving
a lone pair of electrons of the amine functionality as a donor and
the coumarin fragment as an acceptor. Unsubstituted coumarin
was used as a reference for this purpose. Thus, the binding of
Hg2+ induced an B11 fold enhancement in emission intensity at
402 nm (lExt = 320 nm) for L2. For L1 this could induce only a 2.3
fold enhancement in emission intensity at 402 nm (lExt = 320 nm).
Further, F (lExt = 320 nm, lEMmaxs of 585 nm) for Hg2+ÁL was
evaluated as 0.035, while that for Hg2+ÁL1 under identical
experimental conditions was 0.61. These data clearly indicate
that an efficient energy transfer process is operational from
the donor Hg2+-bound coumarin moiety to the acceptor
Hg2+-bound rhodamine fragment in Hg2+ÁL1. Minimal spectral
overlap of donor and acceptor fragments (Fig. 2B inset)
disapproves the possibility of any FRET process to be opera-
tive. Thus, this is a unique example of the metal ion mediated
TBET process. Efficiency of the TBET process was evaluated
as 88% from the ratio of the quantum yields of the acceptor
fragment in the energy transfer cassette after excitation at the
donor to acceptor centre (ESIw).9 Systematic emission titration
profiles recorded using lExt of 320 nm revealed that with
increase in [Hg2+] a gradual increase in emission intensity at
585 nm was observed, while the emission responses at 402 nm
were rather weak. The TBET process and the spin–orbit
coupling of the Hg2+ could be proposed to account for this
little enhancement in emission intensities at 402 nm. Thus, the
present example could demonstrate a rare example of the
Hg2+ induced interrupted PET coupled TBET process. This
Binding affinity of the reagent L1 towards a series of groups
1A (Li+, Na+, K+), 2A (Mg2+, Ba2+, Ca2+, Sr2+), common
transition metal ions (Fe2+, Co2+, Ni2+, Cu2+, Zn2+, Cd2+
,
Pb2+, Fe3+, Cr3+ and Hg2+) was checked by monitoring
changes in its electronic as well as its fluorescence spectral
(lExt of 320 or 500 nm) pattern in MeOH–aq. HEPES buffer
medium, which remained unchanged for all metal ions, except
for Hg2+ (Fig. 2A and B). For Hg2+ new absorption and
Fig. 2 (A) Electronic spectra of L1 (6.7 Â 10À6 M) in the absence and
presence of different metal ions (1.0 Â 10À4 M); Mn+ = Li+, Na+
,
,
K+, Mg2+, Ca2+, Ba2+, Sr2+, Cu2+, Ni2+, Zn2+, Cd2+, Co2+
Fe2+, Fe3+, Cr3+ and Pb2+; inset: changes in optical spectra of L1 in
the presence of varying [Hg2+]; (B) emission spectra for L/L1/L2/L +
Hg2+/L1 + Mn+/Hg2+ and L2 + Hg2+. [Lx] (x = 0/1/2) = 6.7 Â
10À6 M) and [M2+] or [Hg2+] = 1.5 Â 10À4 M (lExt = 320 nm). Inset:
emission spectra for L2 + Hg2+ and absorption spectra for L + Hg2+
showing poor spectral overlap. All spectral studies were performed in
MeOH–aq. HEPES buffer (1 : 1, v/v) at pH 7.2.
also helped to achieve a large quasi-Stokes shift of 265 nm.
Flu
The association constant (KHg2+
104 MÀ1) for the Hg2+ÁL1 formation was evaluated from the
= (7.0 Æ 0.5) Â
ÁL1
c
9294 Chem. Commun., 2012, 48, 9293–9295
This journal is The Royal Society of Chemistry 2012

View Article Online
2 ppb, which is the concentration limit for safe drinking water
recommended by Environmental Protection Agency (EPA).12
Cytotoxicity of the reagent L1 towards live MCF7 cells was
also established using MTT assay, which revealed that 80% of
these cells remained viable even upon exposure to a 10 mM
concentration of L1 for 12 h (ESIw, Fig. S25).
In brief we have demonstrated a completely new optical
response mechanism, where an interrupted PET coupled
TBET process has been used for recognition of Hg2+ under
physiological conditions and its intracellular distribution in
live MCF7 cells. Intake of [Hg2+] as low as 2 ppb could be
detected using this new imaging probe L1. MTT assay showed
no significant cytotoxicity of L1 towards MCF7 cells.
Fig. 3 Changes in the fluorescence spectral pattern for the receptor
(A) L1 (6.7 Â 10À6 M) in the presence of varying [Hg(ClO4)2] for
l
Ext = 320 nm); inset: (i) L1 (6.7 Â 10À6 M) in the presence of varying
A. D acknowledges DST and CSIR for funding. S. S. and
P. M. thank CSIR, while M. B. thanks UGC for their fellow-
ships. S. K. G. thanks DBT for partial support and DST for
Confocal facilities at IIT, Kharagpur, India. The authors also
acknowledge the Analytical Discipline and Centralised Instru-
mental Facility of CSMCRI for instrumental analysis.
[Hg(ClO4)2] for lExt = 500 nm in MeOH–HEPES buffer medium
(1 mM) (1 : 1, v/v; pH 7.2), (ii) photograph of the visually detectable
solution emission of L1 in (a) the absence and (b) in the presence of
Hg2+ (1.5 Â 10À4 M). (B) Proposed mode of binding in L1ÁHg2+ and
direction of the energy transfer pathway.
B–H plot using data obtained from a systematic luminescence
titration (lExt = 320 nm, lMon = 585 nm) in MeOH–aq.
HEPES buffer medium (Fig. 3A). Linear fit of this plot
Notes and references
1 (a) X. Zhang, Y. Xiao and X. Qian, Angew. Chem., 2008,
120, 8145; (b) W. Lin, L. Yuan, Z. Cao, Y. Feng and J. Song,
Angew. Chem., Int. Ed., 2010, 49, 375.
2 (a) E. Kimura and T. Koike, Chem. Soc. Rev., 1998, 27, 179;
(b) X. Qu, Q. Liu, X. Ji, H. Chen, Z. Zhou and Z. Shen, Chem.
Commun., 2012, 48, 4600; (c) Z. Tan, R. Kote, W. N. Samaniego,
S. J. Weininger and W. G. McGimpsey, J. Phys. Chem. A, 1999,
103, 7612; (d) G.-S. Jiao, L. H. Thoresen and K. Burgess, J. Am.
Chem. Soc., 2003, 125, 14668.
3 (a) V. Bhalla, Roopa, M. Kumar, P. R. Sharma and T. Kaur,
Inorg. Chem., 2012, 51, 2150 and references therein;
(b) R. Bandichhor, A. D. Petrescu, A. Vespa, A. B. Kier,
F. Schroeder and K. Burgess, J. Am. Chem. Soc., 2006, 128, 10688.
4 (a) Q. Wanga, D. Kima, D. D. Dionysiou, G. A. Soriala and
D. Timberlake, Environ. Pollut., 2004, 131, 323; (b) B. N. Ahamed,
M. Arunachalam and P. Ghosh, Inorg. Chem., 2010, 49, 4447;
(c) M. Suresh, A. Shrivastav, S. Mishra, E. Suresh and A. Das,
Org. Lett., 2008, 10, 3013; (d) R. Von Burg, J. Appl. Toxicol., 1995,
15, 483; (e) ATSDR, Toxicological Profile for Mercury, U.S.
Department of Health and Human Services, Atlanta, GA, 1999,
pp. 1–696; (f) X. Ma, J. Wang, Q. Shan, Z. Tan, G. Wei and
D. Wei, Org. Lett., 2012, 14, 820.
corroborated the 1 : 1 binding stoichiometry. Evaluation of
Flu
the KHg2+
value for lExt = 500 nm, lMon = 585 nm in
ÁL1
MeOH–aq. HEPES buffer medium also resulted in an analogous
value ((7.3 Æ 0.6) Â 104 MÀ1). A binding stoichiometry of 1 : 1
was further confirmed from Job’s plot as well as from the results
of the ESI-MS analysis. Changing the counter anion did not
influence the binding processes (ESIw, Fig. S11).
Binding of L1 to Hg2+ induced a higher downfield shift for
the H2 proton, compared to other protons of the coumarin
moiety. This suggests an unusual Hg(II)–Z2-arene p-interaction
(ESIw, Fig. 12). Such an interaction between a relatively electron
deficient arene ring (coumarin) and Hg2+ is unprecedented.10
Based on the 1H NMR spectral data and other optical responses,
binding mode that is proposed for Hg2+ÁL1 is shown in Fig. 3B.
No other metal ion was found to interfere with the binding
of Hg2+ to L1. Details of this interference study, along with
the results for establishing the reversibility in the binding
process, are provided in the ESIw (Fig. S13 and S14).
5 E. M. Nolan and S. J. Lippard, Chem. Rev., 2008, 108, 3443.
6 J.-H. Soh, K. M. K. Swamy, S. K. Kim, S. Kim, S.-H. Lee and
J. Yoon, Tetrahedron Lett., 2007, 48, 5966.
Hg2+ is known to have an estrogenic effect on breast cancer
cells.11 Thus, we checked the possibility of using the reagent L1 as
an imaging agent for the detection of the intracellular distribu-
tion of Hg2+ in live MCF7 cells using a laser scanning confocal
microscope (FV1000, Olympus). Control experiments with live
MCF7 cells pre-treated only with L1 did not show any bright
fluorescence images (ESIw, Fig. S22), while intracellular red
fluorescence was observed for these cells pre-treated with Hg2+
(Fig. 4). Intracellular distribution of Hg2+ was evident in the
confocal images even when cells were exposed to [Hg2+] of
7 (a) P. Mahato, S. Saha, E. Suresh, R. D. Liddo, P. P. Parnigotto,
M. T. Conconi, M. K. Kesharwani, B. Ganguly and A. Das, Inorg.
Chem., 2012, 51, 1769; (b) S. Saha, P. Mahato, U. G. Reddy,
E. Suresh, A. Chakrabarty, M. Baidya, S. K. Ghosh and A. Das,
Inorg. Chem., 2012, 51, 336; (c) M. Suresh, A. K. Mandal, S. Saha,
E. Suresh, A. Mandoli, R. D. Liddo, P. P. Parnigotto and A. Das,
Org. Lett., 2010, 12, 5406; (d) H. N. Kim, M. H. Lee, H. J. Kim,
J. S. Kim and J. Yoon, Chem. Soc. Rev., 2008, 1465.
8 (a) J. R. Lokowize, Principles of Fluorescence Spectroscopy,
Springer, Berlin, 3rd edn, 2006; (b) B. Valeur, Molecular Fluorescence,
Wiley-VCH, New York, 2nd edn, 2005.
9 C. Thivierge, J. Han, R. M. Jenkins and K. Burgess, J. Org. Chem.,
2011, 76, 5219.
10 (a) W. Lin, X. Cao, Y. Ding, L. Yuan and Q. Yu, Org. Biomol.
Chem., 2010, 8, 3618; (b) I.-T. Ho, G.-H. Lee and W.-S. Chung,
J. Org. Chem., 2007, 72, 2434; (c) Y. Liu, X. Wan and F. Xu,
Organometallics, 2009, 28, 5590.
¨
11 J. G. Ionescu, J. Novotny, V. Stejskal, A. Latsch, E. Blaurock-
Busch and M. Eisenmann-Klein, Neuroendocrinol. Lett., 2006,
27, 36.
12 Risk Assessment, Management and Communication of Drinking
Water Contamination, U.S. Environmental Protection Agency,
Washington, DC, US EPA625/4-89/024, EPA: 1989.
Fig. 4 Confocal images of Hg2+-treated MCF7 cells. The cells were
supplemented with Hg(ClO4)2 of (a–c) 2 ppb and (d–f) 10 ppb in the
growth media for 1.0 h at 37 1C followed by staining with 10.0 mM L1 for
1.0 h at 37 1C (lExt = 543 nm). Images (a–c) and (d–f) are in different scales.
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 9293–9295 9295