Design and synthesis of a library of BODIPY-based environmental polarity sensors utilizing photoinduced electron-transfer-controlled fluorescence ON/OFF switching

Article abstract of DOI:10.1021/ja068551y

We systematically examined the mechanism of the solvent polarity dependence of the fluorescence ON/OFF threshold of the BODIPY (boron dipyrromethene) fluorophore and the role of photoinduced electron transfer (PeT). In a series of BODIPY derivatives with

Full text of DOI:10.1021/ja068551y

Published on Web 04/11/2007
Design and Synthesis of a Library of BODIPY-Based
Environmental Polarity Sensors Utilizing Photoinduced
Electron-Transfer-Controlled Fluorescence ON/OFF Switching
Hisato Sunahara,†,§ Yasuteru Urano,†,‡ Hirotatsu Kojima,†,§ and Tetsuo Nagano*,†,§
Contribution from the Graduate School of Pharmaceutical Sciences, the UniVersity of Tokyo,
7
-3-1, Hongo, Bunkyo-ku, Tokyo 113-0033, Japan, CREST, JST, and PRESTO, JST, Honcho,
Kawaguchi, Saitama 332-0013, Japan
Received November 28, 2006; E-mail: ff47015@mail.ecc.u-tokyo.ac.jp
Abstract: We systematically examined the mechanism of the solvent polarity dependence of the
fluorescence ON/OFF threshold of the BODIPY (boron dipyrromethene) fluorophore and the role of
photoinduced electron transfer (PeT). In a series of BODIPY derivatives with variously substituted benzene
moieties at the 8-position, the oxidation potential of the benzene moiety became more positive and the
reduction potential of the BODIPY fluorophore became more negative as the solvent polarity was decreased;
consequently, the free energy change of PeT from the benzene moiety becomes larger in a more nonpolar
environment. Utilizing this finding, we designed and synthesized a library of probes in which the threshold
of fluorescence ON/OFF switching corresponds to different levels of solvent polarity. These environment-
sensitive probes were used to examine bovine serum albumin (BSA) and living cells. The polarity at the
surface of albumin was concluded to be similar to that of acetone, while the polarity of the internal membranes
of HeLa cells was similar to that of dichloromethane.
8
Introduction
acceptor fluorophore. Recently, we have established rational
design strategies for fluorescence probes based on the concept
Environment-sensitive fluorescence probes are functional
molecules whose fluorescence properties are dependent upon
the hydrophobicity of the environments, and they are widely
used in biochemical research.1-6 These probes generally exhibit
a low fluorescence quantum yield in aqueous solution but
become highly fluorescent in nonpolar solution (e.g.,
9-11
of PeT.
These findings allow us to design new fluorescence
probes for various target analytes. The donor moiety of these
fluorescence probes is structurally altered by the reaction or by
binding with target molecules, resulting in a marked change of
fluorescence properties.
Here we report a novel environment-sensitive BODIPY
8
-anilino-1-naphthalenesulfonic acid (ANS) and thioflavin T)
(boron dipyrromethene) probe library in which the fluorescence
or exhibit emission at longer wavelength in aqueous
solution but at shorter wavelength in nonpolar solution (e.g.,
properties are well controlled by the PeT mechanism. Daub and
1
2-16
co-workers
have described the solvent-dependent change
7
Prodan, Nile red). However, the mechanism of the fluorescence
of the optical properties of some BODIPY derivatives. They
focused on the charge-transfer emission, which depends on
solvent polarity, but the precise relationship between PeT-based
quenching of fluorescence and the polarity of the solvent has
not been established. We examined the mechanism of the solvent
polarity-dependent changes in the fluorescence ON/OFF thresh-
changes is not well established, so it is difficult to develop
new probes having different color and fluorescence ON/OFF
threshold.
Photoinduced electron transfer (PeT) is a well-known mech-
anism through which the fluorescence of a fluorophore is
quenched by electron transfer from the donor moiety to the
(
8) deSilva, A. P.; Gunaratne, H. Q. N.; Gunnlaugsson, T.; Huxley, A. J. M.;
McCoy, C. P.; Rademacher, J. T.; Rice, T. E. Chem. ReV. 1997, 97, 1515-
†
University of Tokyo.
CREST, JST.
PRESTO, JST.
1566.
‡
(9) Tanaka, K.; Miura, T.; Umezawa, N.; Urano, Y.; Kikuchi, K.; Higuchi,
§
T.; Nagano, T. J. Am. Chem. Soc. 2001, 123, 2530-2536.
(
(
(
(
(
(
(
1) Yeh, R. H.; Yan, X. W.; Cammer, M.; Bresnick, A. R.; Lawrence, D. S. J.
(10) Miura, T.; Urano, Y.; Tanaka, K.; Nagano, T.; Ohkubo, K.; Fukuzumi, S.
J. Am. Chem. Soc. 2003, 125, 8666-8671.
Biol. Chem. 2002, 277, 11527-11532.
2) Parasassi, T.; Krasnowska, E. K.; Bagatolli, L.; Gratton, E. J. Fluoresc.
(11) Urano, Y.; Kamiya, M.; Kanda, K.; Ueno, T.; Hirose, K.; Nagano, T. J.
Am. Chem. Soc. 2005, 127, 4888-4894.
1
998, 8, 365-373.
3) Saito, Y.; Miyauchi, Y.; Okamoto, A.; Saito, I. Tetrahedron Lett. 2004,
5, 7827-7831.
4) Toutchkine, A.; Kraynov, V.; Hahn, K. J. Am. Chem. Soc. 2003, 125, 4132-
145.
5) Ban, T.; Hamada, D.; Hasegawa, K.; Naiki, H.; Goto, Y. J. Biol. Chem.
(12) Kollmannsberger, M.; Rurack, K.; Resch-Genger, U.; Daub, J. J. Phys.
Chem. A 1998, 102, 10211-10220.
4
(13) Kollmannsberger, M.; Rurack, K.; Resch-Genger, U.; Rettig, W.; Daub, J.
Chem. Phys. Lett. 2000, 329, 363-369.
4
(14) Rurack, K.; Kollmannsberger, M.; Resch-Genger, U.; Daub, J. J. Am. Chem.
Soc. 2000, 122, 968-969.
2
003, 278, 16462-16465.
6) Ali, V.; Prakash, K.; Kulkarni, S.; Ahmad, A.; Madhusudan, K. P.; Bhakuni,
V. Biochemistry 1999, 38, 13635-13642.
(15) Rurack, K.; Kollmannsberger, M.; Daub, J. Angew. Chem., Int. Ed. 2001,
40, 385-387.
7) Grabowski, Z. R.; Rotkiewicz, K.; Rettig, W. Chem. ReV. 2003, 103, 3899-
(16) Rurack, K.; Kollmannsberger, M.; Daub, J. New J. Chem. 2001, 25, 289-
292.
4
031.
10.1021/ja068551y CCC: $37.00 © 2007 American Chemical Society
J. AM. CHEM. SOC. 2007, 129, 5597-5604
9
5597

A R T I C L E S
Sunahara et al.
old of the BODIPY fluorophore thermodynamically, and the
results were applied to establish a library of environment-
sensitive fluorescence probes, which we employed to estimate
the polarity at the surface of a protein and at the membranes in
living cells.
aldehyde, a solution of 5.1 mmol of DDQ was added, and stirring was
continued overnight. The reaction mixture was washed with water, dried
over MgSO , filtered, and evaporated. The crude compound was roughly
4
purified by column chromatography on aluminum oxide (CH
2 2
Cl ) to
afford a brown powder.
The brown powder (129 mg, 0.42 mmol) and DIEA (5.0 mL, 5.7
mmol) were dissolved in 20 mL of absolute CH Cl (or anhydrous
toluene) under an Ar atmosphere, and the solution was stirred at room
temperature for 5 min. BF -OEt (8.0 mL, 7.9 mmol) was added, and
stirring was continued for 3 h. The reaction mixture was washed with
water. The aqueous solution was extracted with CH Cl . The combined
organic extracts were dried over anhydrous MgSO , filtered, and
evaporated. The crude compound was purified by silica gel chroma-
Experimental Section
2
2
Materials. General chemicals were of the best grade available,
supplied by Tokyo Chemical Industries, Wako Pure Chemical, or
Aldrich Chemical Co., and were used without further purification.
Special chemicals were tetrabutylammonium perchlorate (TBAP,
electrochemical grade, Fluka). Bovine serum albumin (BSA, product
no. A-7511, Sigma-Aldrich) was used without further purification.
Organelle markers were from Molecular Probes. All the solvents
employed were of spectrometric grade.
3
2
2
2
4
2 2
tography (CH Cl /MeOH, 10:3) to afford an orange powder (90 mg,
yield 9.3%).
Synthesis of 4,4-Difluoro-2,6-diacetyl-8-(5-amino-2-methoxyphe-
nyl)-1,3,5,7-tetramethyl-4-bora-3a,4a-diaza-s-indacene. Compound
General Procedure for the Synthesis of TetramethylBODIPY
Derivatives. An appropriate aldehyde (2 mmol) and 2,4-dimethylpyrrole
1
3 (39 mg, 0.08 mmol) was dissolved in 10 mL of MeOH. After the
addition of 10% Pd-C (8 mg), the mixture was stirred vigorously under
a H atmosphere. When TLC monitoring (silica; CH Cl /MeOH, 20:1)
2 2
(4 mmol) were dissolved in 250 mL of absolute CH Cl under an Ar
atmosphere. One drop of trifluoroacetic acid (TFA) was added, and
the solution was stirred at room temperature overnight. When thin-
2
2
2
showed complete consumption of 13, the Pd-C was filtered off and
washed with MeOH. The residue after evaporation of the filtrate was
layer chromatography (TLC) monitoring (silica; CH
complete consumption of the aldehyde, a solution of 2 mmol of 2,3-
dichloro-5,6-dicyano-1,4-benzoquinone (DDQ) in CH Cl was added,
and stirring was continued for 15 (or 20) min. The reaction mixture
was washed with water, dried over MgSO , filtered, and evaporated.
The crude compound was roughly purified by column chromatography
on aluminum oxide (CH Cl ) to afford a brown-orange solid.
The solid and 4 mL of N,N-diisopropylethylamine (DIEA) were
dissolved in 150 mL of absolute CH Cl (or anhydrous toluene) under
an Ar atmosphere, and the solution was stirred at room temperature
for 5 min. Then 4 mL of boron trifluoride etherate (BF -OEt ) was
added, and stirring was continued for 30 min. The reaction mixture
was washed with water and dried over anhydrous MgSO , filtered, and
evaporated. The crude compound was purified by silica gel chroma-
2 2
Cl ) showed
purified by column chromatography over silica gel (CH
0:1) to afford a violet powder (recrystallized from EtOH to afford
orange needles) (14 mg, 0.03 mmol, yield 38%).
2 2
Cl /MeOH,
2
2
2
4
General Instruments. NMR spectra were recorded on a JNM-
1
LA300 (JEOL) instrument at 300 MHz for H NMR and at 75 MHz
2
2
13
for C NMR. Mass spectra (MS) were measured with a JMS-DX300
(JEOL) for EI and a JMS-T100LC AccuTOF (JEOL) for ESI.
2
2
Cyclic Voltammetry. Cyclic voltammetry was performed on a 600A
electrochemical analyzer (ALS). A three-electrode arrangement in a
single cell was used for the measurement: a Pt wire as the auxiliary
3
2
+
electrode, a Pt electrode as the working electrode, and a Ag/Ag
4
electrode as the reference electrode. The sample solution contained 0.1
M TBAP as a supporting electrolyte in the solvent, and argon was
bubbled for 2 min before each measurement. Obtained potentials (Ag/
2 2 2 2
tography (CH Cl /MeOH or CH Cl /hexane) to afford an analytically
pure sample.
+
Ag ) were converted to those versus SCE by adding 0.25 V.
Synthesis of 2-Methoxy-3-methylbenzaldehyde. A solution of
Fluorescence Properties and Quantum Yield of Fluorescence.
Steady-state fluorescence spectroscopic studies were performed on an
F 4500 (Hitachi). The slit width was 2.5 nm for both excitation and
emission. The photon multiplier voltage was 700 V. UV-vis spectra
were obtained on a UV-1600 and a UV-1650PC (Shimazu). Each
2
-hydroxy-3-methylbenzaldehyde (1.10 g, 8.05 mmol) and cesium
carbonate (8.00 g, 24.6 mmol) in dimethylformamide (DMF) (15 mL)
was treated with iodomethane (3.47 g, 24.4 mmol). The mixture was
stirred at room temperature for 10 h, water was added, and the reaction
mixture was extracted with AcOEt (3 × 30 mL). The organic layer
solution contained up to 1.0% (v/v) DMSO or CHCl
3
as a cosolvent
2 4
was dried over anhydrous Na SO and concentrated in vacuo. The crude
(see the Supporting Information). The relative quantum efficiencies of
product was purified by silica gel chromatography to afford a colorless
oil (667 mg, yield 55%).
fluorescence of BODIPY derivatives were obtained by comparing the
area under the corrected emission spectrum of the test sample with
that of a solution of fluorescein excited at 490 nm in 0.1 N NaOH,
Synthesis of 10-Phenyl-9-anthracenecarboxyaldehyde. This was
1
7
synthesized as previously reported.
18
which has a quantum efficiency of 0.85 according to the literature.
General Procedure for the Synthesis of AminoBODIPY Deriva-
tives from NitroBODIPY Derivatives. Nitro derivatives were dis-
solved in MeOH. After the addition of 10% palladium-carbon (Pd-
The quantum efficiencies of fluorescence (Φfl) were obtained with the
following equation (F denotes the area under the fluorescence band (F
)
ΣIfl(λ), where Ifl(λ) is the fluorescence intensity at each emission
2
C), the mixture was stirred vigorously under a H atmosphere. When
wavelength), Abs denotes the absorbance at the excitation wavelength,
TLC monitoring (silica) showed complete consumption of the starting
material, the Pd-C was filtered off and washed with MeOH. The
residue after evaporation of the filtrate was purified by column
and n denotes the refractive index of the solvent)
sample
standard
sample standard
sample standard 2
Φfl
) Φfl
× (F
/F
) × (n
/n
standard
) ×sample
2 2
chromatography over silica gel (CH Cl /MeOH) to afford a violet
(Abs
/Abs
)
powder.
Synthesis of 4,4-Difluoro-2,6-diacetyl-8-(2-methoxy-5-nitrophe-
nyl)-1,3,5,7-tetramethyl-4-bora-3a,4a-diaza-s-indacene. 2-Methoxy-
Computational Methods. All structures were computed using hybrid
density functional theory (B3LYP) with the 6-31G* basis set as
implemented in Gaussian 98W. Several starting geometries were used
for the geometry optimization to ensure that the optimized structure
corresponds to a global minimum.
5
-nitrobenzaldehyde (312 mg, 1.72 mmol) and 3-acetyl-2,4-dimeth-
ylpyrrole (548 mg, 4.00 mmol) were dissolved in 250 mL of absolute
CH Cl under an Ar atmosphere. Two drops of TFA were added, and
the solution was stirred at room temperature overnight. When TLC
monitoring (silica; CH Cl ) showed complete consumption of the
2
2
Preparation of Cells. HeLa cells were cultured in Dulbecco’s
modified Eagle’s medium (DMEM) (Invitrogen Corp., Carlsbad, CA),
2
2
(
17) Soutar, I.; Swanson, L.; Guillet, J. E.; Takahashi, Y. Macromolecules 1991,
2
4, 2815-2821.
(18) Parker, C. A.; Rees, W. T. Analyst 1960, 85, 587-600.
5598 J. AM. CHEM. SOC.
9
VOL. 129, NO. 17, 2007

Design and Synthesis of a BODIPY-Based Library
A R T I C L E S
Figure 1. Structures of compounds used in this study: (A) BODIPY structure divided into two parts, the benzene moiety and the fluorophore; (B) structures
of the benzene moiety at the 8-position of the BODIPY fluorophore. These are arranged according to HOMO energy level from 1 to 15.
supplemented with 10% fetal bovine serum (Invitrogen Corp.), 1%
penicillin, and 1% streptomycin (Invitrogen Corp.) at 37 °C in a 5%
CO /95% air incubator. The cells were grown on an uncoated 35 mm
2
molecular rigidity arising from the introduction of methyl
groups at the ortho positions of aryl-substituted BODIPYs
minimizes the decrease of fluorescence intensity due to the
nonradiative pathway. Thus, the BODIPY derivatives employed
in this study commonly contain ortho substituents to block the
nonradiative pathway and maintain a high fluorescence quantum
yield.
diameter glass-bottomed dish (MatTek, Ashland, MA), and washed
twice with Hanks’ balanced salt solution (HBSS) buffer (Invitrogen
Corp.). The medium was replaced with HBSS buffer containing 50
nM BODIPY derivative dissolved in DMSO before imaging.
Confocal Laser Scanning Microscopy. Confocal imaging was
performed on an Olympus Fluoview 500 confocal microscope. BODIPY
derivatives were excited with the 488 nm line of an argon laser, and
emission was detected over a range of 515-530 nm. For BODIPY
derivative and organelle marker double detection, the samples were
sequentially excited at 488 and 543 nm with argon lasers. The emission
signals from BODIPY derivatives and organelle markers were detected
over a range of 515-530 nm and over a range of 560-600 nm,
respectively, to minimize overlap between the two signals.
2
. Spectroscopic Properties of the BODIPY Derivatives.
The absorbance and fluorescence emission spectra of the
BODIPY derivatives were measured in various solvents. The
HOMO energy levels of the benzene moieties, the absorbance
and fluorescence properties, and the relative quantum yields of
fluorescence (Φfl) in various solvents are summarized in Table
1
. Figure 2, parts A and B, shows that the absorption and
fluorescence spectra remained almost identical when the benzene
moiety at the 8-position was changed even from benzene to
naphthalene or anthracene, which suggests that there is no
significant electronic coupling between the donor and fluoro-
phore in the ground-state configuration. Figure 2, parts C and
D, shows absorption and fluorescence spectra of 8-phenyl-
BODIPY measured in various solvents. Other derivatives also
showed almost identical absorption and fluorescence spectra in
the various solvents tested here.
Results and Discussion
1
. Design and Synthesis of Pilot Compounds. The boron
dipyrromethene (BODIPY) fluorophore (4,4-difluoro-4-bora-
a,4a-diaza-s-indacene) (Figure 1A) has a high extinction
3
-
1
-1
coefficient (>80 000 cm M ), a high fluorescence quantum
yield (often approaching 1.0, even in water), and is relatively
insensitive to solvent polarity and pH.19 Therefore, BODIPY
derivatives are used as fluorescence tags and probes which
function in water or organic solvent. Conceptually, the
structure of BODIPY derivatives can be divided into two parts,
i.e., the benzene moiety and BODIPY moiety, since they are
twisted and conjugatively uncoupled.2
Figure 3 shows the relationship between the fluorescence
quantum yield and the HOMO energy level of the benzene
moiety calculated at the B3LYP/6-31G level. In DMSO,
derivatives in which the HOMO energy level of the benzene
moiety is more negative than -0.21 hartrees showed very high
Φfl values. At HOMO energy levels more positive than -0.20
hartrees, the Φfl values dropped sharply, finally reaching Φfl ∼
2
0
21
2,23
To systematically investigate the PeT-based fluorescence
quenching of BODIPY derivatives in different environments,
we synthesized various BODIPY derivatives in which the
highest occupied molecular orbital (HOMO) energy levels of
the benzene moiety at the 8-position were tuned over a wide
range (Figure 1B). A previous report24,25 suggested that the
0
. Thus, the threshold level between fluorescence OFF and ON
(green column in the figure) was determined to be around -0.21
to -0.20 hartrees in the case of DMSO.
Similar behavior was observed in other solvents, but the
position of the fluorescence ON/OFF threshold moved to
higher HOMO energy regions with decreasing dielectric
constant of the solvents. Finally, in the case of hexane, all
the derivatives were highly fluorescent and the threshold could
not be determined. These results suggested that PeT hardly
occurred in nonpolar solvents, and an even higher HOMO
energy would be necessary to quench fluorescence by PeT in
hexane.
(
(
(
19) Haugland, R. P. In Handbook of Fluorescent Probes and Research
Chemicals, 9th ed.; Molecular Probes, Inc.: Eugene, OR, 2002.
20) Gabe, Y.; Urano, Y.; Kikuchi, K.; Kojima, H.; Nagano, T. J. Am. Chem.
Soc. 2004, 126, 3357-3367.
21) Pap, E. H. W.; Drummen, G. P. C.; Winter, V. J.; Kooij, T. W. A.; Rijken,
P.; Wirtz, K. W. A.; Op den Kamp, J. A. F.; Hage, W. J.; Post, J. A. FEBS
Lett. 1999, 453, 278-282.
(
22) Kollmannsberger, M.; Gareis, T.; Heinl, S.; Breu, J.; Daub, J. Angew. Chem.,
Int. Ed. Engl. 1997, 36, 1333-133.
(23) Montalban, A. G.; Herrera, A. J.; Johannsen, J.; Beck, J.; Godet, T.; Vrettou,
M.; White, A. J. P.; Williams, D. J. Tetrahedron Lett. 2002, 43, 1751-
1
75.
3
. Quantitative Study of the PeT Process. It was plausible
(
24) Yamada, K.; Toyota, T.; Takakura, K.; Ishimaru, M.; Sugawara, T. New J.
Chem. 2001, 25, 667-66.
that the fluorescence of BODIPY was quenched by PeT in all
(
25) Prieto, J. B.; Arbeloa, F. L.; Martinez, V. M.; Lopez, T. A.; Amat-Guerri,
F.; Liras, M.; Arbeloa, I. L. Chem. Phys. Lett. 2004, 385, 29-35.
26) Marcus, R. A.; Sutin, N. Biochim. Biophys. Acta 1985, 811, 265-322.
27) Marcus, R. A. Angew. Chem., Int. Ed. Engl. 1993, 32, 1111-1121.
(
(
(28) Dean, J. A. Langre’s Handbook of Chemistry; McGraw-Hill, Inc.: New
York, 1992.
J. AM. CHEM. SOC.
9
VOL. 129, NO. 17, 2007 5599

A R T I C L E S
Sunahara et al.
Table 1. HOMO Energy Levels and Spectroscopic Data of BODIPY Derivatives in Various Solvents
HOMO energy
(hartrees)
λ
(nm)
abs
λ
(nm)
em
HOMO energy
(hartrees)
λ
(nm)
abs
λ
em
(nm)
compd
solvent
Φ
fl
compd
solvent
Φ
fl
1
-0.2365
-0.2303
-0.2174
-0.2141
-0.2127
-0.2100
-0.2063
-0.1997
DMSO
CH3CN
MeOH
acetone
CH2Cl2
CHCl3
503
498
499
502
502
503
505
502
514
509
509
512
512
514
515
511
0.875
0.923
0.915
0.849
0.874
0.862
0.950
0.925
9
-0.1975
-0.1931
-0.1901
-0.1901
-0.1894
-0.1887
-0.1744
DMSO
CH3CN
MeOH
acetone
CH2Cl2
CHCl3
504
499
500
501
503
505
506
504
517
508
511
511
514
516
517
513
0.011
0.011
0.053
0.127
0.818
0.979
0.815
0.975
benzene
hexane
benzene
hexane
2
3
4
5
6
7
8
DMSO
CH3CN
MeOH
acetone
CH2Cl2
CHCl3
504
499
500
500
503
504
505
502
514
509
510
510
513
514
515
511
0.903
0.910
0.959
0.891
0.858
0.903
0.921
0.986
10
11
12
13
14
15
DMSO
CH3CN
MeOH
acetone
CH2Cl2
CHCl3
507
502
502
503
505
507
508
505
518
513
513
513
515
518
516
511
0.003
0.003
0.007
0.022
0.094
0.595
0.924
0.974
benzene
hexane
benzene
hexane
DMSO
CH3CN
MeOH
acetone
CH2Cl2
CHCl3
504
499
500
500
503
505
506
504
515
509
510
510
513
515
516
512
0.889
0.860
0.950
0.880
0.961
0.889
0.970
0.855
DMSO
CH3CN
MeOH
acetone
CH2Cl2
CHCl3
505
500
502
504
504
507
507
505
n.d.
n.d.
n.d.
n.d.
515
517
514
511
<0.001
<0.001
<0.001
<0.001
0.002
0.101
0.500
benzene
hexane
benzene
hexane
0.880
DMSO
CH3CN
MeOH
acetone
CH2Cl2
CHCl3
504
500
500
500
503
505
505
503
515
510
511
510
513
515
516
512
0.728
0.892
0.964
0.966
0.965
0.961
0.922
0.943
DMSO
CH3CN
MeOH
acetone
CH2Cl2
CHCl3
507
502
503
503
506
508
509
505
n.d..
n.d.
515
513
519
519
516
511
<0.001
<0.001
0.004
0.011
0.019
0.108
0.685
0.834
benzene
hexane
benzene
hexane
DMSO
CH3CN
MeOH
acetone
CH2Cl2
CHCl3
505
500
500
501
504
505
507
504
517
511
512
512
516
518
518
514
0.857
0.873
0.893
0.814
0.885
0.904
0.844
0.916
DMSO
CH3CN
MeOH
acetone
CH2Cl2
CHCl3
505
501
501
502
504
505
506
502
n.d.
n.d.
n.d.
n.d.
517
517
515
512
<0.001
<0.001
<0.001
<0.001
0.004
0.024
0.149
benzene
hexane
benzene
hexane
0.875
DMSO
CH3CN
MeOH
acetone
CH2Cl2
CHCl3
503
498
499
499
503
505
505
502
514
510
511
511
514
515
518
513
0.717
0.891
0.949
0.854
0.899
0.971
0.925
0.852
DMSO
CH3CN
MeOH
acetone
CH2Cl2
CHCl3
506
502
502
503
505
507
508
505
n.d.
n.d.
n.d..
n.d.
516
517
516
511
<0.001
<0.001
<0.001
<0.001
0.012
0.049
0.539
benzene
hexane
benzene
hexane
0.926
DMSO
CH3CN
MeOH
acetone
CH2Cl2
CHCl3
504
499
500
501
503
505
506
504
514
509
510
510
514
514
516
513
0.785
0.894
0.942
0.965
0.923
0.929
0.859
0.948
DMSO
CH3CN
MeOH
acetone
CH2Cl2
CHCl3
502
498
499
499
503
505
505
503
n.d.
n.d.
n.d.
n.d.
n.d.
515
514
513
<0.001
<0.001
<0.001
<0.001
<0.001
0.003
benzene
hexane
benzene
hexane
0.022
0.861
DMSO
CH3CN
MeOH
acetone
CH2Cl2
CHCl3
506
501
502
502
505
506
507
504
517
507
507
513
515
516
514
509
0.142
0.240
0.573
0.715
0.886
0.857
0.763
0.882
benzene
hexane
a
All data were measured in the presence of up to 1% CHCl3 as a cosolvent. The fluorescence quantum yield was determined using fluorescein as a
standard (Φfl ) 0.85 in 0.1 M NaOH) (ref 18). HOMO energy levels of the corresponding benzene moieties were calculated with B3LYP/6-31G(d)//B3LYP/
6
-31G(d) by Gaussian 98W. n.d. ) not detectable.
the solvents except hexane because only Φfl changed and no
change in absorbance or emission wavelength was observed.
In general, the feasibility of electron transfer between a
fluorophore and a quencher can be judged from the change in
5600 J. AM. CHEM. SOC.
9
VOL. 129, NO. 17, 2007

Design and Synthesis of a BODIPY-Based Library
A R T I C L E S
Figure 2. Absorption and fluorescence spectra of BODIPY derivatives:
absorption (A) and fluorescence (B) spectra of 8-position-substituted
BODIPY (red, benzene; green, naphthalene; light blue, anthracene) measured
in CHCl3; absorption (C) and fluorescence (D) spectra of 8-phenyl BODIPY
in various solvents (red, hexane; orange, benzene; yellow, CHCl3; yellow-
green, CH2Cl2; green, acetone; light blue, MeOH; blue, CH3CN; purple,
DMSO).
Table 2. Oxidation Potential of the Benzene Moiety and
a
Reduction Potential of BODIPY in Various Solvents
DMSO
CH
3
CN
MeOH
2 2
CH Cl
Eox 1/2 (V vs SCE)
ferrocene
0.31
1.00
n.d.
0.37
1.04
1.25
0.40
1.06
1.29
0.51
1.15
1.40b
b
b
9
,10-dimethylanthracene
,4-dimethoxybenzene
c
b
1
Ered 1/2 (V vs SCE)
-1.21 -1.19
compd 2
n.d.c
-1.25
a
The redox potentials were measured by cyclic voltammetry in DMSO,
CH3CN, MeOH, and CH2Cl2, containing 0.1 M TBAP with a scan rate of
-
1
b
-1 c
0
.1 V/s . Scan rate of 500 V/s . n.d. ) not detectable.
free energy (∆GeT). The ∆GeT value can be calculated from
the Rehm-Weller equation,29
∆
GeT ) E - E - E - C
ox red 00
where Eox and Ered are the oxidation potential of the electron
donor and the reduction potential of the acceptor, respectively,
E00 is the singlet excited energy of the fluorophore, and C is
the work term for the charge separation state.
We next tried to elucidate the reason why the fluorescence
ON/OFF threshold was shifted in different solvents. Judging
from the solvent-dependent change of Φfl, the electron-transfer
process becomes thermodynamically disadvantageous in lower
polarity solvents. Therefore, we measured Eox and Ered in every
solvent.
Figure 3. Relationships between the HOMO energy level of the benzene
moiety and the Φfl of BODIPY derivatives. The curves represent the best
fit to the Marcus equation (refs 26 and 27). Depending upon the dielectric
constant (ꢀ) of the solvents (ref 28), the fluorescence ON/OFF threshold
(green column) gradually changes.
(from DMSO to CH Cl ), the oxidation potential of the donor
2
2
3
.1. Electrochemistry. Cyclic voltammetry was carried out
moiety increased (for example for 9,10-dimethylanthracene,
from 1.00 V in DMSO to 1.15 V in CH2Cl2) and the reduction
potential of the BODIPY fluorophore decreased (from -1.21
V in DMSO to -1.26 V in CH2Cl2). These results indicate that
it is harder for the donor moiety to be oxidized and for the
fluorophore to be reduced in less polar solvents.
to determine the redox potentials of BODIPY derivatives in
various solvents (Table 2). BODIPY derivatives could not be
used for the comparison of Eox of the benzene moiety in each
solvent, because the BODIPY fluorophore can be oxidized
earlier than the benzene moiety. Therefore, we compared the
half-wave potential of the corresponding benzene moiety of the
BODIPY derivatives; for example, p-dimethoxybenzene was
measured for compound 9. As the solvent polarity decreased
3
.2. Other Parameters in the Rehm-Weller Equation. In
the case of BODIPY derivatives, all compounds showed almost
the same absorption maxima and emission maxima, which
indicated that E00 was essentially constant at 2.44 ( 0.02 eV.
(
29) Rehm, D.; Weller, A. Isr. J. Chem. 1970, 8, 259-271.
J. AM. CHEM. SOC.
9
VOL. 129, NO. 17, 2007 5601

A R T I C L E S
Sunahara et al.
Figure 4. Electrochemical and fluorescence properties of compound 16: (A) synthesis of compound 16; (B) electrochemical properties of the fluorophore
of BODIPY derivatives in acetonitrile (V vs SCE); (C) relationship between the fluorescence quantum efficiency (Φfl) in hexane and the free energy change
(∆G). ∆G was calculated from eq 1 (E00 ) 2.44, C ) 0.72; see text) in acetonitrile.
Figure 5. Fluorescence change of BODIPY derivatives (0.25 µM, containing 0.2% DMSO as a cosolvent) with or without BSA (20 µM) in sodium phosphate
buffer (pH 7.4): fluorescence of compound 3 (a), compound 7 (b), compound 9 (c), and compound 11 (d) excited at 365 nm; absorbance spectra of each
solution (e-h) without BSA (broken lines) or with BSA (solid lines); fluorescence spectra of each solution excited at 490 nm (i-l).
Previously, we showed that the electron-transfer process of
fluorescein could be reasonably described by assuming C to
have a constant value.10 In a directly linked donor-acceptor
system, the dielectric constant of the solvent becomes unrelated
to C. So, we considered C to be a constant for the BODIPYs
tested here.
These findings suggest that the redox potential influences
∆GeT and determines the rate of electron transfer. Therefore,
the solvent-dependent change of fluorescence ON/OFF threshold
is concluded to be due primarily to the change of ∆GeT caused
by the influence of solvent polarity on the redox potentials.
Electron transfer occurs more readily in high-polarity solvents.
5602 J. AM. CHEM. SOC.
9
VOL. 129, NO. 17, 2007

Design and Synthesis of a BODIPY-Based Library
A R T I C L E S
Figure 6. Fluorescence micrographs of HeLa cells colabeled for 10 min
with 50 nM compound 3 (containing 0.1% DMSO as a cosolvent) in HBSS
and various organelle-specific probes: (A) mitochondrial staining with 100
nM MitoTracker red CM-H2XRos; (B) lysosomal staining with 75 nM
LysoTracker red DND-99; (C) staining of the Golgi apparatus with 5 µM
BODIPY TR ceramide.
Table 3. Comparison of Spectroscopic Properties of BODIPYs
with or without Serum Albumin
compd 3
compd 7
compd 9
compd 11
BSA
(−)
(
+)
(
−)
(
+)
(
−)
(
+)
(
−)
(+)
λabs (nm) 498
λem (nm) 510
505
513
498
512
505
515
499
n.d.
505
515
499
505
a
n.d.a 516
Φfl
0.871 0.330 0.020 0.218 0.002 0.187 0.001 0.023
a
n.d. ) not detectable.
4
. Quenching of Fluorescence in Hexane. To prepare an
environment-sensitive fluorescence probe library of BODIPY
derivatives having different fluorescence ON/OFF thresholds
of solvent polarity, we required a compound whose fluorescence
would be quenched in every solvent. Since the fluorescence of
the 15 BODIPY derivatives described above was not quenched
in hexane, we tried to design a compound whose fluorescence
is quenched in hexane.
Figure 7. Fluorescence intensity change of BODIPY derivatives in HeLa
cells. (A) HeLa cells were incubated with 50 nM BODIPY derivatives
(containing 0.1% DMSO as a cosolvent) for 10 min at room temperature,
and the fluorescence excited at 488 nm was measured. Fluorescence imaging
and transmission imaging are shown. (B) The fluorescence ON/OFF
threshold for HeLa cells loaded with BODIPY derivatives is compared with
results from Figure 3.
Because the ∆GeT value in a nonpolar solvent becomes more
positive, quenching of fluorescence in a nonpolar solvent
requires a more negative value of ∆GeT. According to the
Rehm-Weller equation, the free energy change of the PeT
process is determined not only by the electron-donating ability
of the benzene moiety but also by the reduction potential and
the excitation energy of the fluorophore.
5
. Biological Applications of the Library of BODIPY
Derivatives. 5.1. Serum Albumin. We examined the application
of our library of BODIPY derivatives to assess the polarity of
the surface of a protein (BSA) in vitro.
BODIPY derivatives were dissolved in sodium phosphate
buffer (pH 7.4) with or without BSA. As shown in Figure 5,
parts a-d, compounds 7 and 9 showed a marked fluorescence
increase in the presence of BSA, i.e., almost no fluorescence
without BSA and strong fluorescence with BSA. In contrast,
compounds 3 and 11 showed almost no fluorescence change in
the presence of BSA. The absorption maximum of each
BODIPY (Figure 5, parts e-h) showed a red-shift in the
presence of BSA, indicating that each BODIPY interacts with
So we synthesized a BODIPY derivative whose benzene
moiety has a higher HOMO energy level as a PeT donor and
whose fluorophore has acetyl moieties in the 2,6-positions of
BODIPY as a readily reducible fluorophore (Figure 4A). The
reduction potentials of the fluorophore were measured as -1.19
and -0.85 V versus SCE for 2,6H-BODIPY and 2,6Ac-
BODIPY in acetonitrile, respectively (Figure 4B). Because the
reduction potential of the BODIPY fluorophore was made more
positive by introduction of the acetyl groups, Φfl of compound
5
6
-1
BSA with a binding constant of about 2 × 10 to 1 × 10 M
(
see the Supporting Information), which is similar to those of
30,31
conventional probes.
1
6 became as small as 0.06 even in hexane (Figure 4C).
The fluorescence properties of the BODIPY derivatives were
shown in Figure 5, parts i-l, and Table 3. Fluorescence emission
of compound 3 showed no increase, because there is no
favorable PeT pathway in the absence or presence of BSA. The
Thus, we successfully utilized our finding that the solvent
dependence of fluorescence ON/OFF switching of BODIPY
derivatives is due to the influence of environmental polarity
on the PeT process to prepare a library of environment-
sensitive fluorescence probes usable over a wide range of solvent
polarity.
(
30) Haskard, C. A.; Li-Chan, E. C. Y. J. Agric. Food Chem. 1998, 46, 2671-
677.
(31) Cardamone, M.; Puri, N. K. Biochem. J. 1992, 282, 589-593.
2
J. AM. CHEM. SOC.
9
VOL. 129, NO. 17, 2007 5603

A R T I C L E S
Sunahara et al.
reason for the fluorescence decrease is presumably an increase
of the nonradiative pathway owing to attachment to the protein,
that the polarity of the internal membranes is close to that of
dichloromethane.
It is thought that the polarity of the lipid in membranes is
similar to that of a long-chain alkane, so BODIPY derivatives
may interact with membrane surface proteins.
3
2
as is generally observed with known fluorescence tags. In the
case of compounds 7 and 9, fluorescence emission is quenched
by PeT in the buffer, but at the surface of BSA, PeT is less
favorable due to the hydrophobic environment, so fluorescence
is increased. Compound 11, in which PeT readily occurs in
buffer, showed no fluorescence increase upon binding to BSA,
because it showed almost no fluorescence even in a hydrophobic
solvent such as CHCl3.
Thus, fluorescence ON/OFF switching is caused by the
detection of the protein surface polarity. The surface polarity
of BSA is concluded to be similar to the polarity of acetone.
Therefore, our environment-sensitive BODIPY library can be
used for sensing the hydrophobicity at a protein surface.
These results demonstrate that our library of BODIPY
derivatives has potential for biological applications.
Conclusion
We have shown that the fluorescence ON/OFF threshold of
BODIPY derivatives is dependent upon solvent polarity, i.e.,
PeT-dependent fluorescence quenching occurred more easily in
highly polar media than in less polar media. The oxidation
potential of the donor moiety became more positive and the
reduction potential of the fluorophore became more negative
in less polar media so that the ∆GeT value of PeT became larger.
Hence, PeT is decreased in a nonpolar environment. Based on
these findings, we developed a library of environment-sensitive
fluorescence probes consisting of a series of BODIPYs bearing
different electron donors. We showed that it could be used to
estimate the polarity of the surface of BSA and internal
membranes of HeLa cells.
5
.2. Living Cells. Next, we examined the application of
BODIPY derivatives to cultured living cells to estimate the
polarity of the cell membrane.
To investigate the intracellular localization of BODIPY
derivatives, we examined the colocalization of our BODIPYs
with known organelle probes (Figure 6). HeLa cells were
incubated first with organelle probes (MitoTracker red CM-
H2XRos for 30 min at 37 °C, LysoTracker red DND-99 for 30
min at 37 °C, and BODIPY TR ceramide for 30 min at 4 °C
then for 30 min at 37 °C), followed by BODIPY derivatives
for 10 min at room temperature. BODIPY derivatives mainly
colocalized with Golgi apparatus marker, and partially with
mitochondrial and lysosomal markers. These results suggested
that BODIPY derivatives accumulate generally on internal
membranes in living cells.
We next examined the membrane polarity by measuring the
fluorescence intensity of BODIPY derivatives loaded into HeLa
cells (Figure 7). Cells loaded with compound 9 gave bright
images, while those loaded with compound 11 gave dark images.
These two compounds have almost the same methoxybenzene
moiety, so it is plausible that they share almost the same
localization. Comparison with the data in Figure 3 suggested
Our environment-sensitive fluorescence probe library should
be a useful tool to detect local polarity changes in a wide variety
of biological samples, including proteins, membranes, and
receptors.
Acknowledgment. This study was supported in part by a
Grant from Hoansha Foundation to T.N., and also by research
Grants (Grant Nos. 14103018, 16651106, 16689002, and
18038008 to Y.U.) from the Ministry of Education, Culture,
Sports, Science and Technology of the Japanese Government,
and by a Grant from the Kato Memorial Bioscience Foundation
to Y.U.
Supporting Information Available: Characterization data for
synthetic compounds, effect of cosolvent concentration, and
determination of the binding constant. This material is available
free of charge via the Internet at http://pubs.acs.org.
(32) Lancet, D.; Pecht, I. Biochemistry 1977, 16, 5150-5157.
JA068551Y
5604 J. AM. CHEM. SOC.
9
VOL. 129, NO. 17, 2007