Development of a cholesterol-conjugated fluorescent sensor for site-specific detection of zinc ion at the plasma membrane

Article abstract of DOI:10.1021/ol201746p

A cholesterol-conjugated fluorescence Zn²⁺ sensor based on the fluorescein platform was designed and synthesized. The cholesterol moiety is essential for localizing the Zn²⁺ sensor to the cell membrane, allowing the sensor to probe c

Full text of DOI:10.1021/ol201746p

ORGANIC
LETTERS
2011
Vol. 13, No. 17
4558–4561
Development of a Cholesterol-Conjugated
Fluorescent Sensor for Site-Specific
Detection of Zinc Ion at the Plasma
Membrane
Shohei Iyoshi,† Masayasu Taki,*,†,‡ and Yukio Yamamoto†
Graduate School of Human and Environmental Studies and Graduate School of Global
Environmental Studies, Kyoto University, Yoshida, Sakyo-ku, Kyoto 606-8501, Japan
taki.masayasu.4c@kyoto-u.ac.jp
Received June 28, 2011
ABSTRACT
A cholesterol-conjugated fluorescence Zn2þ sensor based on the fluorescein platform was designed and synthesized. The cholesterol moiety is
essential for localizing the Zn2þ sensor to the cell membrane, allowing the sensor to probe changes in the Zn2þ concentration in a localized area of
the cell.
Zinc is the second most abundant heavy metal in living
organisms and plays a significant role as a component of
many metalloenzymes and transcription factors.1 In recent
years, the importance of zinc ions in brain function,2 the
immune system,3 and apoptosis regulation4 has also been
revealed, and these findings have attracted the attention of
many researchers. However, details of the mechanisms of
zinc homeostasis have yet to be clarified. In order to
understand this equilibrium between zinc uptake and
efflux, chemical tools that can detect a change in zinc
† Graduate School of Human and Environmental Studies.
‡ Graduate School of Global Environmental Studies.
(1) (a) Pavletich, N. P.; Pabo, C. O. Science 1991, 252, 809–817. (b)
Pabo, C. O.; Sauer, R. T. Annu. Rev. Biochem. 1992, 61, 1053–1095. (c)
Berg, J. M.; Shi, Y. G. Science 1996, 271, 1081–1085. (d) McCall, K. A.;
Huang, C. C.; Fierke, C. A. J. Nutr. 2000, 130, 1437S–1446S.
(2) (a) Bush, A. I.; Pettingell, W. H.; Multhaup, G.; Paradis, M. D.;
Vonsattel, J. P.; Gusella, J. F.; Beyreuther, K.; Masters, C. L.; Tanzi,
R. E. Science 1994, 265, 1464–1467. (b) Koh, J. Y.; Suh, S. W.; Gwag,
B. J.; He, Y. Y.; Hsu, C. Y.; Choi, D. W. Science 1996, 272, 1013–1016.
(c) Frederickson, C. J.; Suh, S. W.; Silva, D.; Frederickson, C. J.;
Thompson, R. B. J. Nutr. 2000, 130, 1471S–1483S. (d) Flinn, J. M.;
Hunter, D.; Linkous, D. H.; Lanzirotti, A.; Smith, L. N.; Brightwell, J.;
Jones, B. F. Physiol. Behav. 2005, 83, 793–803. (e) Frederickson, C. J.;
Koh, J. Y.; Bush, A. I. Nat. Rev. Neurosci. 2005, 6, 449–462. (f)
Mocchegiani, E.; Bertoni-Freddari, C.; Marcellini, F.; Malavolta, M.
Prog. Neurobiol. 2005, 75, 367–390.
(3) (a) Shankar, A. H.; Prasad, A. S. Am. J. Clin. Nutr. 1998, 68,
447S–463S. (b) Raqib, R.; Roy, S. K.; Rahman, M. J.; Azim, T.; Ameer,
S. S.; Chisti, J.; Andersson, J. Am. J. Clin. Nutr. 2004, 79, 444–450. (c)
Kitamura, H.; Morikawa, H.; Kamon, H.; Iguchi, M.; Hojyo, S.;
Fukada, T.; Yamashita, S.; Kaisho, T.; Akira, S.; Murakami, M.;
Hirano, T. Nat. Immunol. 2006, 7, 971–977. (d) Yamasaki, S.; Sakata-
Sogawa, K.; Hasegawa, A.; Suzuki, T.; Kabu, K.; Sato, E.; Kurosaki, T.;
Yamashita, S.; Tokunaga, M.; Nishida, K.; Hirano, T. J. Cell Biol. 2007,
177, 637–645.
(4) (a) Chai, F. G.; Truong-Tran, A. Q.; Ho, L. H.; Zalewski, P. D.
Immunol. Cell Biol. 1999, 77, 272–278. (b) Truong-Tran, A. Q.; Ho,
L. H.; Chai, F.; Zalewski, P. D. J. Nutr. 2000, 130, 1459S–1466S. (c)
Kimura, E.; Aoki, S.; Kikuta, E.; Koike, T. Proc. Natl. Acad. Sci. U.S.A.
2003, 100, 3731–3736.
r
10.1021/ol201746p
Published on Web 08/01/2011
2011 American Chemical Society

concentration in particular subdomains within a cell are
desired.
Fluorescence imaging is the most suitable method for
monitoring the local dynamics of zinc in living cells
because of its high spatial resolution.5 A straightforward
approach to achieve this is to attach a fluorescent zinc
sensor, which includes small organic molecules and pro-
tein-based indicators, to a target protein known to be
expressed on specific compartments within the cell. For
instance, Lippard et al. have reported SNAP-tag-based
site-specific zinc sensors, which have a benzylguanine unit
as a substrate of O6-alkylguanine transferase (AGT).6
They successfully carried out chemical labeling of the
AGT-fused protein expressed on mitochondria or Golgi
apparatuses and visualization of a change in zinc concen-
trations at a local intracellular region. Palmer et al. have
developed genetically targeted ratiometric zinc sensors
based on the fluorescence resonance energy transfer
(FRET) mechanism and have monitored the releasable
pool of Zn2þ in mitochondria.7 Although these genetically
encoded techniques are advantageous in that the sensor
distributions can be controlled precisely, the localization
strategies require a complex gene modification procedure,
including plasmid preparation and gene transfection, to
arrange the observation objects. Another strategy to loca-
lize fluorescent molecules to cellular subdomains is to use
fluorescent zinc sensors with a functional group that can
interact with specific organelles.8 A labile Zn2þ pool within
mitochondria has been monitored by using the mitochon-
dria-specific fluorescent zinc sensor Rhodzin-3, which is
based on a positively charged rhodamine fluorophore.8a
Although this approach may lack the certainty of sensor
localization within the target organelle, this method
requires no gene modification and, more importantly,
enables multicolor imaging of a single cell by changing
the fluorescence sensor moiety and organelle-targeting
group.9 However, only a few zinc sensors based on this
approach have been reported, and none have yet been
reported for the plasma membrane.
Figure 1. Molecular design of LF-Chol and its reference com-
pounds, LF-Me, F-chol, and F-Me.
O-triacetic acid)-based zinc-chelating structure in which
one acetic acid ligand is substituted with a 2-pyridylmethyl
group (L in Figure 1), because this pyridine-containing
chelatorhassufficient selectivityandhighaffinity(dissocia-
tion constant Kd sub-nM) for zinc ions, as we reported
previously.11 In addition, owing to the highly hydrophilic
character of the two remaining carboxyl groups of the zinc
chelator, the zinc sensor moiety is expected to be located at
the extracellular region in the vicinity of the plasma
membrane, which may allow for detection of zinc efflux
from the cell.
In order to evaluate the photophysical properties of
LF-Chol as a fluorescent zinc sensor, a reference molecule
LF-Me (shown in Figure 1), in which the TEG-cholesterol
moiety was substituted with a simple methyl group,
was initially prepared (Scheme 1). Compound 2 was pre-
pared according to the improved procedure we reported
previously.11 It was necessary to transform the tert-butyl
esters in compound 2 into the corresponding ethyl esters 3
because the formation of the xanthene platform as a
fluorophore by FriedelꢀCrafts reaction would be carried
out under acidic conditions. Formylation of ethyl ester 3
under Vilsmeier conditions followed bythe FriedelꢀCrafts
reaction afforded fluorescein derivative 5, which is also
used for the synthesis of LF-Chol. Methylation of the
phenolic oxygen atom and subsequent hydrolysis afforded
LF-Me, which was purified by semipreparative reversed-
phase HPLC (H2O/CH3CN containing 0.1% TFA) for
spectroscopic measurements.12
In this context, we designed and synthesized a new
small-molecule-based fluorescent zinc sensor, LF-Chol
(Figure 1). Because cholesterol has been known to prefer-
entially interact with a subset of membrane lipids,10 we
expected that the conjugated fluorescent zinc sensor moi-
ety of LF-Chol would localize specifically to the cell
membrane. We chose an APTRA (o-aminophenol-N,N,
(5) (a) Jiang, P. J.; Guo, Z. J. Coord. Chem. Rev. 2004, 248, 205–229.
(b) Kikuchi, K.; Komatsu, K.; Nagano, T. Curr. Opin. Chem. Biol. 2004,
8, 182–191. (c) Que, E. L.; Domaille, D. W.; Chang, C. J. Chem. Rev.
2008, 108, 1517–1549. (d) Tomat, E.; Lippard, S. J. Curr. Opin. Chem.
Biol. 2010, 14, 225–230.
(6) Tomat, E.; Nolan, E. M.; Jaworski, J.; Lippard, S. J. J. Am. Chem.
Soc. 2008, 130, 15776–15777.
(7) Dittmer, P. J.; Miranda, J. G.; Gorski, J. A.; Palmer, A. E. J. Biol.
Chem. 2009, 284, 16289–16297.
(8) (a) Sensi, S. L.; Ton-That, D.; Weiss, J. H.; Rothe, A.; Gee, K. R.
Cell Calcium 2003, 34, 281–284. (b) Dodani, S. C.; Leary, S. C.; Cobine,
P. A.; Winge, D. R.; Chang, C. J. J. Am. Chem. Soc. 2011, 133, 8606–
8616.
(9) Rosania, G. R.; Lee, J. W.; Ding, L.; Yoon, H. S.; Chang, Y. T. J.
Am. Chem. Soc. 2003, 125, 1130–1131.
(11) Taki, M.; Watanabe, Y.; Yamamoto, Y. Tetrahedron Lett. 2009,
50, 1345–1347.
(12) Since available amounts of LF-Me and LF-Chol were too small
for 13C and 1H/13C NMR analyses, respectively, the overall purity of the
product was confirmed by HPLC and the product was identified by mass
analysis (see Supporting Information).
(10) Wustner, D. Chem. Phys. Lipids 2007, 146, 1–25.
Org. Lett., Vol. 13, No. 17, 2011
4559

Scheme 1. Synthesis of LF-Me and LF-Chol
Figure 2. Emission spectra of LF-Me (5 μM) excited at 470 nm in
Zn2þ/NTA buffered system (50 mM HEPES, pH 7.20, 0.1 M
KNO3; 10 mM NTA, 0ꢀ9.5 mM ZnSO4) and in 50 mM HEPES
buffer (pH 7.20) containing 4 μM ZnSO4. Inset: plots of
fluorescence intensities at 517 nm with best-fit curves for the
dissociation constant of 1.26 ꢁ 10ꢀ7 M.
To confirm the localization ability of LF-Chol to the
plasma membrane, we next conducted in vivo imaging
experiments with a fluorescent reference compound,
F-Chol (Figure 1), in which the Zn2þ-binding moiety L
was replaced with a 2-carboxyphenyl group. HeLa cells
stained with F-Chol exhibit strong emission from the
plasma membrane region as determined by confocal scan-
ning laser microscopy (Figure 3). This result indicates that
the cholesterol moiety strongly interacts with a subset of
membrane lipids, whereas the fluorescein moiety is ex-
pected to be located at the extracellular region because of
the high polarity of the carboxyl group. In contrast, no
fluorescence signal was detected in any area (Figure S3)
when the cells were incubated in the presence of F-Me
(Figure 1), which lacks the cholesterol moiety from
F-Chol. This finding also supports the crucial role of
cholesterol in dye localization. On the basis of the above
observations, we concluded that LF-Chol is sufficiently
capable of detecting extracellular Zn2þ in the vicinity of the
plasma membrane.
a For LF-Me Et ester: MeI, K2CO3, DMF, 60 °C, overnight, 51%.
For LF-Chol Et ester: 7, K2CO3, DMF, 60 °C, overnight, 41%.
Under physiological conditions (50 mM HEPES, pH
7.20, 0.1 M KNO3), the fluorescence of LF-Me is nearly
quenched (Φ < 0.005) because of the photoinduced
electron transfer (PET) process and is enhanced (ca. 45-
fold) upon coordination with Zn2þ (Φ = 0.18), as shown
inFigure2. A metal selectivity experimentrevealedthatthe
fluorescent response of LF-Me was highly Zn2þ-selective
and that the existence of other biologically relevant metal
ions did not disturb the Zn2þ sensing process (Figure S1).
A 1:1 metal-to-ligand complex was confirmedby Job’s plot
(Figure S2), and on the basis of this finding, the dissocia-
tion constant Kd for Zn2þ was determined to be 126 nM by
a curve-fitting analysisofthe fluorescenceintensityspotsat
517 nm against the free Zn2þ concentration, [Zn2þ
]free, in a
Zn2þ/NTA (NTA = nitrilotriacetic acid) buffer solution
system (Figure 2 inset).11 This Kd value reveals that the
sensor is suitable for detecting [Zn2þ
]
between 32 and
free
520 nM,13 which is appropriate to the amount of Zn2þ
secreted by exocytosis to the extracellular milieu.14 These
profiles of LF-Me support the view that the Zn2þ-sensing
moiety of LF-Chol possesses sufficient ability to detect
Zn2þ under physiological conditions.
Figure 3. (a) Confocal fluorescence image of HeLa cells loaded
2.5 μM F-Chol. (b) Bright-field transmission image of the cells
shown in (a). (c) Overlay of images (a) and (b).
(13) Henary, M. M.; Wu, Y. G.; Fahrni, C. J. Chem.;Eur. J. 2004,
10, 3015–3025.
(14) (a) Qian, W. J.; Aspinwall, C. A.; Battiste, M. A.; Kennedy, R. T.
Anal. Chem. 2000, 72, 711–717. (b) Gee, K. R.; Zhou, Z. L.; Qian, W. J.;
Kennedy, R. J. Am. Chem. Soc. 2002, 124, 776–778. (c) Kay, A. R. J.
Neurosci. 2003, 23, 6847–6855. (d)Qian, W. J.; Gee, K. R.;Kennedy, R. T.
Anal. Chem. 2003, 75, 3468–3475. (e) Komatsu, K.; Kikuchi, K.; Urano,
Y.; Kojima, H.; Nagano, T. J. Am. Chem. Soc. 2005, 127, 10197–10204.
LF-Chol was prepared by using the common compound
5 for LF-Me (Scheme 1). A coupling reaction with
a cholesterol derivative 7 followed by hydrolysis of the
4560
Org. Lett., Vol. 13, No. 17, 2011

fluorescence increase in the local region of the cells, indicat-
ing that the zinc-sensing moiety of LF-Chol complexed
with extracellular Zn2þ while retaining sensor localization
in the plasma membrane (Figure 4c). In fact, a decrease in
emission was observed upon addition of 100 μM EDTA,
which is a membrane-impermeable chelator for Zn2þ. This
strongly supports the view that the Zn2þ-binding moiety of
LF-Chol existed on the extracellular side and detected
extracellular Zn2þ in a reversible manner (Figure 4d).15
We expect that LF-Chol can contribute to the visualiza-
tion of zinc ion influx or efflux in the extracellular plasma
membrane region, including zinc release from neurons and
zinc secretion from pancreatic β-cells. Currently, we are
working on the preparation of LF-Chol derivatives for
detecting Zn2þ in other intracellular organelles such as the
nucleus and mitochondria, which may be achieved with
Hoechst derivatives and cationic functional groups, re-
spectively. In addition, because this sensor-localization
technique enables Zn2þ imaging without any genetic en-
coding techniques, it should be possible to perform multi-
color imaging of local Zn2þ at more than one cellular
compartment in a single cell by using zinc sensors with
differently colored fluorophores. Such an imaging techni-
que may help in understanding the Zn2þ-trafficking path-
way between cell compartments.
Figure 4. Live-cell fluorescence imaging of extracellular Zn2þ at
the plasma membrane by confocal microscopy. (a) Bright-field
transmission image of HeLa cells. (b) Cells stained with 2.5 μM
LF-Chol. (c) Images obtained immediately after the addition of
20 μM Zn2þ to the same media shown in (b). (d) After addition
of 100 μM EDTA to the same cells shown in (c).
Acknowledgment. We thank Professor I. Hamachi and
Dr. Y. Takaoka at Kyoto University, as well as Professor
A. Ojida at Kyushu University, for helpful advice. This
work was financially supported by a Grant-in-Aid for
Young Scientists (B) from MEXT (No. 70378850 for M.
T.), the Naito Foundation (M.T.), and a Grant-in-Aid for
JSPS Fellows (S.I.).
ethyl esters with LiOH gave the final fluorescence sensor
LF-Chol as a brown solid. The crude product was purified
by ODS column chromatography. For application to in
vivo imaging, a sample of the compound was further
purified by HPLC.12 Cells loaded with 2.5 μM of LF-Chol
emitted slight but detectable fluorescence in the plasma
membrane, as shown in Figure 4b. The staining pattern is
similar to that observed with F-Chol, demonstrating mem-
brane localization of this zinc sensor. The addition of Zn2þ
(final concentration: 20 μM) to the same media yielded a
Supporting Information Available. Detailed description
of synthetic procedure and NMR spectra of the new
compounds, the results of a metal selectivity experiment
and Job’s plot, and fluorescence images of cells with
F-Me. This material is available free of charge via the
Internet at http://pubs.acs.org.
(15) It should be noted that some bright fluorescent spots still
remained in Figure 4d. Because LF-Chol becomes neutral and increases
the hydrophilicity upon complexation with Zn2þ, a part of the metal-
lated sensor may permeate the cell membrane.
Org. Lett., Vol. 13, No. 17, 2011
4561