A novel design method of ratiometric fluorescent probes based on fluorescence resonance energy transfer switching by spectral overlap integral

Article abstract of DOI:10.1002/chem.200390167

A ratiometric measurement, namely, simultaneous recording of the fluorescence intensities at two wave-lengths and calculation of their ratio, allows greater precision than measurements at a single wavelength, and is suitable for cellular imaging studies.

Full text of DOI:10.1002/chem.200390167

FULL PAPER
A Novel Design Method of Ratiometric Fluorescent Probes
Based on Fluorescence Resonance Energy Transfer Switching
by Spectral Overlap Integral
Hideo Takakusa,[a] Kazuya Kikuchi,[a, b] Yasuteru Urano,[a] Hirotatsu Kojima,[a] and
Tetsuo Nagano*[a]
Abstract: A ratiometric measurement,
namely, simultaneous recording of the
fluorescence intensities at two wave-
lengths and calculation of their ratio,
allows greater precision than measure-
ments at a single wavelength, and is
suitable for cellular imaging studies.
Here we describe a novel method of
designing probes for ratiometric meas-
urement of hydrolytic enzyme activity
based on switching of fluorescence res-
onance energy transfer (FRET). This
method employs fluorescent probes with
accept the excitation energy of the
donor moiety and the donor fluores-
cence can be observed. After cleavage of
the protective groups by hydrolytic en-
zymes, the fluorescein moiety shows a
strong absorption in the coumarin emis-
sion region, and then acceptor fluores-
cence due to FRETis observed. Based
on this mechanism, we have developed
novel ratiometric fluorescent probes
(1 3) for protein tyrosine phosphatase
(PTP) activity. They exhibit a large shift
in their emission wavelength after reac-
tion with PTPs. The fluorescence
quenching problem that usually occurs
with FRETprobes is overcome by using
the
coumarin cyclohexane fluores-
cein FRETcassette moiety, in which
close contact of the two dyes is hindered.
After study of their chemical and kinetic
properties, we have concluded that com-
pounds 1 and 2 bearing a rigid cyclo-
hexane linker are practically useful for
the ratiometric measurement of PTPs
activity. The design concept described in
this paper, using FRETswitching by
spectral overlap integral and a rigid link
that prevents close contact of the two
dyes, should also be applicable to other
hydrolytic enzymes by introducing other
appropriate enzyme-cleavable groups
into the fluorescein acceptor.
a
3'-O,6'-O-protected fluorescein ac-
ceptor linked to a coumarin donor
through a linker moiety. As there is no
spectral overlap integral between the
coumarin emission and fluorescein ab-
sorption, the fluorescein moiety cannot
Keywords: fluorescent probes ¥ flu-
orescence resonance energy transfer
¥
protein tyrosine phosphatase
¥
ratiometric measurement
of such factors, a ratiometric measurement is to be utilized,
namely, simultaneous recording of the fluorescence intensities
at two wavelengths and calculation of their ratio. This
technique provides greater precision than measurement at a
single wavelength, and is suitable for cellular imaging studies.
To carry out a ratiometric measurement, the probe must
exhibit a large shift in its emission or excitation spectrum after
it reacts (or binds) with the target molecule. Fluorescence
resonance energy transfer[2] (FRET) is one mechanism used
as a basis to obtain a large shift in the spectral peak. FRETis
an interaction between the electronic excited states of two
fluorophores, in which excitation energy is transferred from a
donor to an acceptor without emission of a photon. The
efficiency of FRETdepends on the donor acceptor distance,
the relative orientation of the donor and acceptor transition
dipoles, the extent of spectral overlap of the emission
spectrum of the donor with the absorption spectrum of the
acceptor (spectral overlap integral), and other factors. Re-
cently, several ratiometric fluorescent probes using FRET
have been developed,[3] such as the b-lactamase probe CCF2[4]
Introduction
In recent years, many fluorescent probes have been developed
to study biological phenomena in living cells.[1] A fluorescent
probe is advantageous due to its high sensitivity; however, the
fluorescence measurement can be influenced by many factors,
such as the localization of the probe, changes of environment
around the probe (e.g., pH, polarity, temperature) and
changes in the excitation intensity. To reduce the influence
[a] Prof. T. Nagano, H. Takakusa, Dr. K. Kikuchi,
Dr. Y. Urano, Dr. H. Kojima
Graduate School of Pharmaceutical Sciences
University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, 113-0033 (Japan)
Fax : (‡81)3-5841-4855
E-mail: tlong@mol.f.u-tokyo.ac.jp
[b] Dr. K. Kikuchi
Presto, JSTCorporation
Kawaguchi, Saitama (Japan)
Supporting information for this article is available on the WWW under
http://www.chemeurj.org or from the author.
Chem. Eur. J. 2003, 9, No. 7
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0907-1479 $ 20.00+.50/0
1479

T. Nagano et al.
FULL PAPER
and the phosphodiesterase probe CPF4.[5] They are all based
on the same principle, that is, that FRETefficiency is
dependent upon the donor acceptor distance. To design such
probes, it is necessary that the donor and acceptor should be
located on opposite sides of an enzyme-cleavable structure. It
is difficult to apply this design method widely, because of the
fluorescence quenching problem due to dye-to-dye close
contact[6] and a decrease in affinity for the target enzymes.[5]
To expand the range of application of FRET-based fluores-
cent probes, we have utilized a novel approach to design
probes whose absorption characteristics are altered by
enzymatic reaction, but which do not suffer from the above
problems.
Excitation
HO
Emission
O
O
O
O
H
N
N
H
O
O
RO
O
OR
Lactone form
Hydrolytic enzyme
FRET
Excitation
HO
Emission
O
O
O
O
H
N
N
H
COOH
HO
O
O
R : Enzyme-cleavable group
Results and Discussion
Quinoid form
Scheme 1. The FRET-based detection mechanism.
Detection mechanism: FRETefficiency is dependent on the
spectral overlap integral between the donor emission and
acceptor absorption. We thought that if the acceptor absorp-
tion could be dramatically changed by hydrolytic enzymes,
FRETswitching by spectral overlap integral would be
possible. It was anticipated that fluorescein would be appro-
priate as such an acceptor, because it has two conformations
(lactone form and quinoid form) with distinctly different
absorption properties. Fluorescein in the quinoid form has
strong absorption at around 490 nm, whereas the lactone form
has absorption only in the UV region. If a coumarin derivative
is chosen as the donor, there is a significant difference in the
spectral overlap between the two forms of fluorescein (Fig-
ure 1). So, we propose a novel ratiometric detection method
energy of the donor moiety, and the donor fluorescence can be
observed. After cleavage of the protective groups by hydro-
lytic enzymes, the fluorescein structure changes to the quinoid
form, which has a strong absorption in the coumarin emission
region. Then, acceptor fluorescence due to FRET will occur,
and so the probes will exhibit a large red shift in their emission
spectrum after the reaction with the target enzymes. To
validate the proposed strategy, we designed and synthesized
novel fluorescent compounds bearing phosphate groups in the
fluorescein moiety.
Design and synthesis of the ratiometric fluorescent probes: As
shown in Figure 2, three compounds which have a coumarin
moiety as the donor and a phosphorylated fluorescein moiety
as the acceptor were designed and synthesized as candidate
probes for protein tyrosine phosphatase (PTP). PTPs are
involved in many biologically important processes,[7] including
cell differentiation, synaptic function, cytoskeletal function
HO
O
O
O
O
H
N
N
H
O
O
O
P
O
P
HO
O
O
O
OH
OH
OH
Figure 1. Spectral overlap of the coumarin emission with the fluorescein
1
*
absorption. : normalized emission spectrum of 7-hydroxycoumarin-3-
HO
O
O
O
O
*
carboxylic acid, lex ˆ 400 nm; : absorption spectrum of 10 mm 6-carboxy-
H
N
&
N
H
fluorescein diacetate (lactone form); : absorption spectrum of 10 mm
6-carboxyfluorescein (quinoid form).
O
O
O
P
MeO
O
O
OH
for hydrolytic enzyme activity based on resonance energy
transfer switching by spectral overlap integral (Scheme 1).
This method employs fluorescent probes with 3'-O,6'-O-
protected fluorescein linked to coumarin through an appro-
priate linker. As fluorescein is locked in the lactone form, it
has no absorption band in the wavelength region of the
coumarin emission, so there will be no overlap integral
between the coumarin emission and fluorescein absorption.
Thus, the fluorescein moiety cannot accept the excitation
OH
2
HO
O
O
O
O
H
N
N
H
O
O
O
P
O
P
HO
O
O
O
OH
OH
OH
3
Figure 2. The structures of compounds 1 3.
1480
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0907-1480 $ 20.00+.50/0
Chem. Eur. J. 2003, 9, No. 7

Spectral Overlap Integral
1479 1485
and immune responses, and also in human diseases[8] such as
cancers and diabetes. However, there is little effective real-
time method to monitor their activity in living cells. In
compound 1, two phosphate residues are introduced into the
fluorescein moiety as enzyme-cleavable groups, and this
moiety is linked to the coumarin donor through a cyclohexane
moiety as a rigid linker. In compound 2, the second phosphate
group was replaced by a methyl residue in order to eliminate
the complication of two dephosphorylation steps. To inves-
tigate the effect of the linker structure on the fluorescence
properties and the affinity for PTPs, an ethylene moiety was
introduced as a flexible linker in compound 3. The synthetic
schemes are shown in Scheme 2.
Spectroscopic properties of 1 3: The emission spectra of 1
3
in an aqueous buffer excited at the excitation wavelength of
the coumarin donor exhibited the emission of the coumarin
donor at around 450 nm before the enzymatic reaction.
Addition of PTP1B, an intracellular PTP,[9] to aqueous
solutions of 1 3 resulted in a decrease in the donor
O
O
O
O
HO
N
O
HO
O
O
b)
COOH
a)
O
O
O
O
P
O
P
O
O
P
P
BnO
O
O
O
OBn
BnO
O
O
O
O
OBn
HO
O
O
OBn
OBn
OBn
OBn
5
4
HO
O
O
HO
O
O
O
O
H
N
H
N
H
N
N
H
O
O
O
d)
c)
O
O
O
O
P
O
P
O
P
P
HO
O
O
O
OH
BnO
O
O
O
OBn
OH
OH
OBn
OBn
6
1
O
O
O
O
HO
HO
MeO
HO
a)
f)
O
O
COOH
COOMe
O
e)
COOH
O
O
O
P
MeO
O
OBn
MeO
O
O
HO
O
OBn
HO
O
9
7
8
HO
O
O
O
O
O
O
H
N
N
N
O
H
O
c)
b)
O
O
O
O
O
O
P
P
MeO
11
O
O
OBn
MeO
O
O
OBn
OBn
OBn
10
HO
O
O
O
O
H
N
N
H
d)
O
O
O
P
MeO
O
O
OH
OH
2
HO
O
O
O
HO
O
O
O
H
O
H
N
N
N
H
N
H
O
h)
O
g)
O
O
5
O
O
P
O
P
O
P
O
P
HO
O
O
O
OH
BnO
O
O
O
OBn
OH
OH
OBn
OBn
3
12
Scheme 2. Synthesis of compounds 1 3: a) (BnO)2PN(iPr)2, 1H-tetrazole, NEt3, mCPBA, CHCl3; b) N-hydroxysuccinimide, EDC ¥ HCl, CHCl3; c) 7-
hydroxycoumarin-3-carboxylic succinimidyl ester, trans-1,4-cyclohexanediamine, DMF; d) Pd/C, H2, MeOH; e) H2SO4, MeOH; f) CH3I, Cs2CO3, DMF, aq
NaOH; g) 7-hydroxycoumarin-3-carboxylic succinimidyl ester, ethylenediamine, DMF; h) Pd/C, H2, MeOH/CHCl3.
Chem. Eur. J. 2003, 9, No. 7
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0907-1481 $ 20.00+.50/0
1481

T. Nagano et al.
FULL PAPER
fluorescence at around 450 nm and an increase in the acceptor
fluorescence at around 515 nm, as shown in Figure 3. The
same spectral changes were also observed upon the addition
of CD45, a receptor-like PTP.[10] These observations demon-
Figure 4. The absorption spectra of compounds 1 3 (1.0 mm) after the
addition of PTP1B (10 mgmLÀ1) in 0.1m HEPES buffer (pH 7.4) containing
1.0 mm DTT and EDTA at 308C. a) Absorption spectra of 1 measured at 1,
5, 10, 20, 45, 60, and 120 min after the addition. b) Absorption spectra of 2
measured at 1, 5, 10, 20, 30, 60, and 120 min after the addition.
c) Absorption spectra of 3 measured at 1, 5, 10, 20, 120, 240, and 360 min
after the addition.
Figure 3. The emission spectra of compounds 1 3 (1.0 mm) after the
addition of PTP1B (10 mgmLÀ1) in 0.1m HEPES buffer (pH 7.4) containing
1.0 mm DTT and EDTA at 308C. a) Emission spectra of 1 measured at 1, 5,
10, 20, 30, 60, and 120 min after the addition, lex ˆ 402 nm. b) Emission
spectra of 2 measured at 1, 5, 10, 20, 30, 60, and 180 min after the addition,
lex ˆ 402 nm. c) Emission spectra of 3 measured at 1, 5, 10, 30, 60, 120, and
180 min after the addition, lex ˆ 407 nm.
500 nm was enhanced with a slight decrease of the absorption
at around 450 nm. These observations in the fluorescence and
absorption spectra indicate that the two phosphate groups of 1
and 3 are hydrolyzed in two steps through the monophosphate
intermediates, whereas 2 is hydrolyzed in one step. The
fluorescence and absorption properties of compounds 1
3
strate that the proposed FRETswitching by spectral overlap
integral worked as expected. As the time courses of the
changes in the emission spectra show, the enhancement of the
acceptor fluorescence was slower than the decrease of the
donor fluorescence in the cases of 1 and 3; on the other hand,
they occurred simultaneously in
before and after hydrolysis by the enzymes are summarized in
Table 1. Compound 3 with the flexible ethylene linker
exhibited weaker fluorescence and longer wavelength of the
absorption maxima compared to 1 and 2 with the rigid
the emission spectrum of 2.
Table 1. Chemical Properties of Compounds 1 3 before and after enzymatic hydrolysis.
The absorption spectra of 1
Compound
Absorption
max [nm][a]
Emission
Quantum
FRET
max [nm][a]
efficiency[a]
efficiency [%][b]
3 measured in an aqueous buf-
fer exhibited the donor absorp-
tion peak at around 400 nm
(Figure 4). After the addition
of PTP1B or CD45, the absorp-
tion peaks of the fluorescein
moiety appeared and increased
with time. In the cases of 1 and
3, the absorption at around
450 nm was initially enhanced,
then the absorption at around
1
before hydrolysis
after hydrolysis
402
402
494
402
402
458
404
407
498
445
515
515
445
515
515
445
515
520
0.53
0.32
0.76
0.36
0.11
0.20
0.11
0.033
0.088
0
96
2
3
before hydrolysis
after hydrolysis
0
96
before hydrolysis
after hydrolysis
0
[a] Data were measured in HEPES buffer (0.1m, pH 7.4). [b] Data were obtained from the fluorescence quantum
efficiency of the donor in the presence and in the absence of acceptor.
1482
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0907-1482 $ 20.00+.50/0
Chem. Eur. J. 2003, 9, No. 7

Spectral Overlap Integral
1479 1485
cyclohexane linker. These observations indicate that fluores-
cence quenching due to dye-to-dye close contact[6] occurred in
3. In other words, the cyclohexane linker is sufficiently rigid to
prevent dye-to-dye close contact. After hydrolysis by PTPs,
efficient FRETcould be observed in 1 and 2. Although
compound 2 after enzymatic cleavage shows weaker fluo-
rescence intensity in the acceptor emission than compound 1
due to the effect of the methyl group incorporated into
fluorescein acceptor, the fluorescence quantum yield of the
fluorescein moiety in compound 2 is equivalent to that of
some widely used fluorescent indicators, and so, the fluo-
rescence of 2 should be sufficient to be detected in living cells.
ate better into the cells, where it should be hydrolyzed to
compound 1 by esterase in the cytosol. In order to determine
whether 1-AM can detect intracellular PTPs activity, it was
used for ratiometric imaging in human umbilical vein
endothelial cells (HUVEC). The fluorescence ratio of the
acceptor and donor within the cells was monitored, in the
presence and absence of sodium orthovanadate, a membrane-
permeable PTP inhibitor, as shown in Figure 5. We observed a
significant difference in the increase of the fluorescence ratio
between inhibitor-treated and untreated cells. Thus, 1-AM
should be practically useful for the ratiometric measurement
of intracellular PTP activity.
Kinetic studies: Kinetic parameters such as Km and kcat were
determined by direct fitting of the initial velocity versus
substrate concentration data to the Michaelis Menten equa-
tion (Table 2). As a standard, Km and kcat of 4-nitrophenyl
phosphate were also measured under the same conditions.
Compounds 1 and 3 had significantly lower Km values than
compound 2 and 4-nitrophenyl phosphate, especially with
CD45. The higher affinity of 1 and 3 would be due to the
presence of the second phosphate group with its negative
charges, that is, electrostatic interaction between the second
phosphate group and PTPs is important for the enhanced
binding.[11] A comparison of the Km values of 1 and 3 shows
that there is no significant difference in their affinities for
PTP1B or CD45. This indicates that the rigid and bulky
structure of the cyclohexane linker does not affect substrate
binding.
Table 2. Kinetic parameters of compounds 1 3 with PTPs.[a]
Compound
Enzyme
Km
[mm]
kcat
kcat/Km
Figure 5. Ratiometric imaging of PTPs activity in HUVEC by using 1-AM.
The fluorescence ratio (acceptor/donor) is shown in pseudo color.
a) Ratiometric image of the inhibitor-untreated cells 20 min after the
probe labeling. b) Ratiometric image of cells in the presence of 1.0 mm
sodium orthovanadate 20 min after the probe labeling. c) Time course of
[sÀ1
]
[mmÀ1 sÀ1
]
1
2
3
PTP1B
CD45
PTP1B
CD45
PTP1B
CD45
PTP1B
CD45
67
12
87
1000
27
14
0.8
3.7
1.6
4.1
0.33
6.7
12
300
18
4.1
12
480
12
10
*
the changes in the fluorescence ratio. : the fluorescence ratio at randomly
*
:
selected regions in HUVEC in the absence of sodium orthovanadate;
the fluorescence ratio at randomly selected regions in HUVEC in the
presence of 1.0 mm sodium orthovanadate.
4-nitrophenyl
phosphate
1100
6500
13
63
[a] Data were measured at 308C in HEPES buffer (0.1m, pH 7.4),
containing DTT (1.0 mm) and EDTA (1.0 mm).
Conclusion
Development of membrane-permeable derivative: Since
compounds 1 3 have low membrane permeability due to
their negative charges, we derivatized compound 1 to the
membrane-permeable 1-AM, which bears acetoxymethyl
residues on the phosphate and hydroxyl groups (Scheme 3).
Compound 1-AM is more lipophilic, and should thus perme-
Our results indicate that FRETswitching by spectral overlap
integral is feasible for practical use. It should be possible to
develop novel ratiometric fluorescent probes for various
hydrolytic enzymes by introducing other appropriate enzyme-
cleavable groups into the fluorescein acceptor. The fluores-
cence quenching problem that usually arises in developing
AMO
O
O
O
HO
O
O
O
O
O
H
H
N
N
N
N
H
H
O
O
a)
O
O
O
P
O
P
O
P
O
P
AMO
O
HO
O
O
O
OAM
O
O
OH
OH
OAM
OH
OAM
1-AM
1
Scheme 3. Synthesis of compound 1-AM: a) acetoxymethyl bromide, diisopropylethylamine, DMF; AM ˆ CH3COOCH2.
Chem. Eur. J. 2003, 9, No. 7 ¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 0947-6539/03/0907-1483 $ 20.00+.50/0
1483

T. Nagano et al.
FULL PAPER
Preparation of 5: Compound 4 (600 mg, 0.67 mmol) was added to a
solution of 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride
(EDC ¥ HCl) (190 mg, 0.99 mmol) and N-hydroxysuccinimide (120 mg,
1.0 mmol) in chloroform. The reaction mixture was stirred for 2 h at room
temperature, washed with an aqueous solution of citric acid and brine, and
dried over sodium sulfate. After evaporation of the chloroform, the residue
was purified by chromatography on silica gel to afford 5 (650 mg, 98%).
1H NMR (300 MHz, CDCl3): d ˆ 2.89 (s, 4H), 5.14 (d, J ˆ 9.0 Hz, 8H), 6.66
(d, J ˆ 8.8 Hz, 2H), 6.82 (dd, J ˆ 2.4, 8.8 Hz, 2H), 7.08 (d, J ˆ 2.4 Hz, 2H),
7.32 (m, 20H), 7.86 (d, J ˆ 1.3 Hz, 1H), 8.18 (d, J ˆ 8.1 Hz, 1H), 8.40 (dd,
FRETprobes can be overcome by using the coumarin-
cyclohexane fluorescein FRETcassette moiety. In addition
to fluorescein, it should be possible to use rhodamine as the
acceptor, because rhodamine also exhibits a large shift in its
absorption spectrum. Consequently, this study has provided
important information about the design and synthesis of
FRET-based fluorescent probes.
Furthermore, by using the above design approach we have
succeeded in developing the PTP probes 1 3. Based on their
chemical and kinetic properties, we anticipate that com-
pounds 1 and 2 bearing the rigid cyclohexane linker are
practically useful for the ratiometric measurement of PTPs
activity. Although compound 1 has the two dephosphoryla-
tion sites, it exhibits PTP activity-dependent increase in the
ratio value of the acceptor and donor fluorescence with a
large dynamic range. So, it should be useful for monitoring the
PTP activity with high sensitivity, especially for comparing the
PTP activity between the different regions in cells or tissues.
Meanwhile, compound 2 does have the advantage of only one
dephosphorylation step, and this could be especially impor-
tant for quantitative detection. The range of biological
applications could be extended by selecting a suitable probe
according to the experimental aim. In summary, chemically
synthesized FRET-based probes for ratiometric measurement
of PTP activity have been developed for the first time by using
spectral overlap integral switching, with a rigid linker moiety
to prevent close contact of the two dyes. These probes should
allow precise, quantitative detection of PTPs activity, and so
should be useful for studies on the biological functions of
PTPs and for studies aimed at developing effective therapeu-
tics for PTP-related diseases.
‡
J ˆ 1.3, 8.1 Hz, 1H); MS (FAB): m/z: 994 [M ‡H].
Preparation of 6: trans-1,4-Cyclohexanediamine (82 mg, 0.73 mmol) was
added to a solution of 5 (600 mg, 0.60 mmol) and the N-hydroxysuccini-
midyl ester of 7-hydroxycoumarin-3-carboxylic acid (190 mg, 0.63 mmol) in
DMF, and the mixture was stirred for 1 h at 08C. Then it was extracted with
ethyl acetate, and the extract was washed with an aqueous solution of citric
acid and brine, and dried over sodium sulfate. After evaporation of the
ethyl acetate, the residue was purified by chromatography on silica gel to
afford 6 (240 mg, 34%). 1H NMR (300 MHz, CDCl3): d ˆ 1.45 (m, 4H),
2.12 (m, 4H), 3.90 (m, 2H), 5.15 (d, J ˆ 9.0 Hz, 8H), 6.22 (d, J ˆ 7.7 Hz,
1H), 6.68 (d, J ˆ 8.8 Hz, 2H), 6.80 (dd, J ˆ 2.2, 8.8 Hz, 2H), 6.85 (dd, J ˆ
2.4, 8.2 Hz, 1H), 6.87 (d, J ˆ 2.4 Hz, 1H), 7.02 (d, J ˆ 2.2 Hz, 2H), 7.33 (m,
20H), 7.44 (s, 1H), 7.47 (d, J ˆ 8.2 Hz, 1H), 8.08 (s, 2H), 8.65 (d, J ˆ 7.9 Hz,
‡
1H), 8.75 (s, 1H); MS (FAB): m/z: 1181 [M ‡H].
Preparation of 1: 10% Pd/C (10 mg) was added to a solution of 6 (35 mg,
30 mmol) in methanol, and the reaction mixture was stirred for 30 min
under a slight positive pressure of H2 gas using a balloon. The progress of
the reduction was followed by reverse-phase HPLC. When no more
starting material was evident and only the product peak was observed, the
reaction mixture was filtered and the filtrate was evaporated to give 1
(20 mg, 82%). 1H NMR (300 MHz, CD3OD): d ˆ 1.48 (m, 4H), 2.00 (m,
4H), 3.75 (m, 2H), 6.74 (d, J ˆ 2.0 Hz, 1H), 6.86 (d, J ˆ 8.8 Hz, 2H), 6.87
(dd, J ˆ 2.0, 8.6 Hz, 1H), 7.00 (dd, J ˆ 2.4, 8.8 Hz, 2H), 7.25 (d, J ˆ 2.4 Hz,
2H), 7.63 (d, J ˆ 8.6 Hz, 1H), 7.63 (d, J ˆ 1.3 Hz, 1H), 8.10 (d, J ˆ 8.1 Hz,
1H), 8.16 (dd, J ˆ 1.3, 8.1 Hz, 1H), 8.71 (s, 1H); 13C NMR (125 MHz,
CD3OD): d ˆ 32.0, 32.4, 49.8, 49.9, 83.5, 103.1, 109.7, 112.8, 114.4, 115.7,
115.9, 118.1, 123.9, 126.4, 129.5, 130.5, 130.9, 132.9, 142.7, 149.6, 153.0, 154.6,
154.7, 158.2, 163.2, 163.7, 165.7, 167.3, 170.3; HRMS (FAB ‡ ): m/z: calcd for
‡
[M ‡H]: 821.1149; found: 821.1166.
Preparation of 7: A solution of 6-carboxyfluorescein (600 mg, 1.6 mmol)
and H2SO4 (1.0 mL) in methanol (30 mL) was heated under reflux
overnight. The reaction mixture was concentrated and diluted with
dichloromethane, then washed with sodium phosphate buffer (pH 7.0)
and brine, and dried over sodium sulfate. After evaporation of the
dichloromethane, compound 7 (330 mg, 52%) was obtained. 1H NMR
(300 MHz, CDCl3): d ˆ 3.64 (s, 3H), 3.94 (s, 3H), 6.79 (dd, J ˆ 2.0, 9.2 Hz,
2H), 6.90 (d, J ˆ 2.0 Hz, 1H), 6.91 (d, J ˆ 9.2 Hz, 2H), 7.98 (s, 1H), 8.31 (s,
Experimental Section
Materials and general methods: PTP1B and CD45 were purchased from
BIOMOL Research Laboratories Inc. (Plymouth Meeting, PA). Fluores-
cence spectra were measured using an F4500 spectrometer (Hitachi, Tokyo,
Japan). Slit width was 2.5 nm for both excitation and emission. The
photomultiplier voltage was 950 V. Absorption spectra were measured
using a UV-1600 spectrometer (Hitachi, Japan). Compounds were dis-
solved in DMSO to make stock solutions, which were diluted to the
‡
2H); MS (FAB): m/z: 405 [M ‡H].
Preparation of 8: MeI (100 mg, 0.70 mmol) was added to a solution of 7
(130 mg, 0.32 mmol) and cesium carbonate (130 mg, 0.4 mmol) in DMF
(5.0 mL). The reaction mixture was stirred for 20 min at room temperature
and then diluted with ethyl acetate. The mixture was washed with an
aqueous solution of citric acid and brine, and dried over sodium sulfate. The
organic phase was concentrated and diluted in methanol. To the methanol
solution, 2n NaOH (5.0 mL) was added and the mixture was stirred for 1 h
at room temperature. The methanol was evaporated and the aqueous
residue was acidified with 2n HCl. The resulting precipitate was collected
by filtration and dried to afford 8 (110 mg, 90%). 1H NMR (300 MHz,
CD3OD): d ˆ 3.95 (s, 3H), 6.78 7.14 (m, 6H), 7.84 (s, 1H), 8.22 (d, J ˆ
1
required concentration for measurement. H NMR and 13C NMR spectra
were recorded on a JEOL JNM-LA300 instrument; d values are given
relative to tetramethylsilane. Fast atom bombardment mass spectroscopy
(FAB-MS) was conducted with a JEOL SX-102A mass spectrometer. High-
resolution mass spectroscopy (HRMS) was done with a JEOL JMS-700T
mass spectrometer. Silica gel column chromatography was performed using
BW-300 (Fuji Silysia Chemical Ltd.).
Preparation of 4: Dibenzyl diisopropylphosphoramidite (1.8 g, 5.2 mmol)
was added to a solution of 6-carboxyfluorescein (500 mg, 1.3 mmol),
triethylamine (TEA) (1.3 g, 13 mmol) and 1H-tetrazole (930 mg, 13 mmol)
in chloroform (20 mL). The reaction mixture was stirred for 30 min at room
temperature and then cooled to 08C. The phosphite intermediate was
oxidized by adding mCPBA (2.0 g, 12 mmol) in portions and the mixture
was stirred for 30 min at room temperature. The mixture was diluted with
chloroform, then washed with 1.5m Na2SO3, 2n HCl and brine, and dried
over sodium sulfate. After evaporation of the chloroform, the residue was
purified by chromatography on silica gel to afford 4 (960 mg, 81%).
1H NMR (300 MHz, CDCl3): d ˆ 5.14 (d, J ˆ 9.2 Hz, 8H), 6.66 (d, J ˆ
8.8 Hz, 2H), 6.81 (dd, J ˆ 2.5, 8.8 Hz, 2H), 7.02 (d, J ˆ 2.5 Hz, 2H), 7.31
(m, 20H), 7.79 (d, J ˆ 1.3 Hz, 1H), 8.09 (d, J ˆ 8.1 Hz, 1H), 8.34 (dd, J ˆ 1.3,
‡
8.4 Hz, 1H), 8.37 (d, J ˆ 8.4 Hz, 1H); MS (FAB): m/z: 391 [M ‡H].
Preparation of 9: Dibenzyl diisopropylphosphoramidite (510 mg,
1.5 mmol) was added to a solution of 8 (290 mg, 0.74 mmol), TEA
(730 mg, 7.2 mmol) and 1H-tetrazole (530 mg, 7.6 mmol) in chloroform
(10 mL). The reaction mixture was stirred for 30 min at room temperature
and then cooled to 08C. The phosphite intermediate was oxidized by
adding mCPBA (1.5 g, 8.7 mmol) in portions and the mixture was stirred
for 30 min at room temperature. Then it was diluted with chloroform,
washed with 1.5m Na2SO3, 2n HCl and brine, and dried over sodium
sulfate. After evaporation of the chloroform, the residue was purified by
chromatography on silica gel to afford 9 (400 mg, 83%). 1H NMR
‡
8.1 Hz, 1H); MS (FAB): m/z: 897 [M ‡H].
1484
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0907-1484 $ 20.00+.50/0
Chem. Eur. J. 2003, 9, No. 7

Spectral Overlap Integral
1479 1485
‡
(300 MHz, CDCl3): d ˆ 3.81 (s, 3H), 5.14 (d, J ˆ 9.0 Hz, 4H), 6.61 7.06 (m,
6H), 7.29 (m, 5H), 7.83 (d, J ˆ 1.1 Hz, 1H), 8.08 (d, J ˆ 7.9 Hz, 1H), 8.31
165.0, 165.7, 170.0; HRMS (FAB ‡ ): m/z: calcd for [M ‡H]: 765.0523;
found: 765.0507.
‡
(dd, J ˆ 1.1, 7.9 Hz, 1H); MS (FAB): m/z: 651 [M ‡H].
Preparation of 1-AM: Acetoxymethyl bromide (30 mL, 300 mmol) and
diisopropylethylamine (100 mL, 600 mmol) were added to a solution of 1
(5.0 mg, 6.1 mmol) in DMF, and the reaction mixture was stirred for 2 h at
room temperature under Ar. Subsequently the mixture was extracted with
ethyl acetate, and the extract was washed with HEPES buffer (pH 7.4), and
dried over sodium sulfate. After evaporation of ethyl acetate, the residue
was purified by chromatography on silica gel to afford 1-AM (3.2 mg,
45%). 1H NMR (300 MHz, CDCl3): d ˆ 1.44 (m, 4H), 2.07 (m, 4H), 2.11 (s,
12H), 2.14 (s, 3H), 3.95 (m, 2H), 5.68 5.78 (m, 8H), 5.83 (s, 2H), 6.08 (d,
J ˆ 8.3 Hz, 1H), 6.81 (d, J ˆ 8.8 Hz, 2H), 6.97 (dd, J ˆ 2.2, 8.8 Hz, 2H), 7.04
(dd, J ˆ 2.3, 8.1 Hz, 1H), 7.06 (d, J ˆ 2.3 Hz, 1H), 7.23 (d, J ˆ 2.2 Hz, 1H),
7.41 (s, 1H), 7.62 (d, J ˆ 8.1 Hz, 1H), 8.05 (dd, J ˆ 1.5, 7.5 Hz, 1H), 8.10 (d,
J ˆ 7.5 Hz, 1H), 8.65 (d, J ˆ 7.9 Hz, 1H), 8.81 (s, 1H); MS (FAB): m/z: 1181
Preparation of 10: Compound 9 (300 mg, 0.46 mmol) was added to a
solution of EDC ¥ HCl (130 mg, 0.68 mmol) and N-hydroxysuccinimide
(80 mg, 0.70 mmol) in chloroform and the mixture was stirred for 2 h at
room temperature. Subsequently the reaction mixture was washed with an
aqueous solution of citric acid and brine, and dried over sodium sulfate.
After evaporation of the chloroform, the residue was purified by
chromatography on silica gel to afford 10 (330 mg, 96%). 1H NMR
(300 MHz, CDCl3): d ˆ 2.82 (s, 4H), 3.81 (s, 3H), 5.14 (d, J ˆ 9.2 Hz, 4H),
6.66 7.12 (m, 6H), 7.31 (m, 10H), 7.91 (d, J ˆ 1.3 Hz, 1H), 8.16 (d, J ˆ
‡
8.0 Hz, 1H), 8.36 (dd, J ˆ 1.3, 8.0 Hz, 1H); MS (FAB): m/z: 748 [M ‡H].
Preparation of 11: trans-1,4-Cyclohexanediamine (64 mg, 0.56 mmol) was
added to a solution of 10 (350 mg, 0.47 mmol) and the N-hydroxysuccini-
midyl ester of 7-hydroxycoumarin-3-carboxylic acid (140 mg, 0.46 mmol) in
DMF, and the mixture was stirred for 1 h at room temperature. The
reaction mixture was then extracted with ethyl acetate and the extract was
washed with an aqueous solution of citric acid and brine, and dried over
sodium sulfate. After evaporation of the ethyl acetate, the residue was
purified by chromatography on silica gel to afford 11 (100 mg, 24%).
1H NMR (300 MHz, CDCl3): d ˆ 1.42 (m, 4H), 2.04 (m, 4H), 3.81 (s, 3H),
3.94 (m, 2H), 5.14 (dd, J ˆ 5.1, 9.0 Hz, 4H), 6.57 7.03 (m, 6H), 6.68 (d, J ˆ
2.2 Hz, 1H), 6.73 (dd, J ˆ 2.2, 8.4 Hz, 1H), 7.33 (m, 10H), 7.47 (d, J ˆ
8.4 Hz, 1H), 7.60 (s, 1H), 8.03 (d, J ˆ 8.0 Hz, 1H), 8.12 (d, J ˆ 8.0 Hz, 1H),
‡
[M ‡H].
Ratiometric imaging: HUVEC were cultured in EGM-2 medium (Sanko-
Junyaku, Japan). The cells were incubated with PBS containing 5.0 mm 1-
AM for 10 min, in the presence or absence of 1.0 mm sodium orthovanadate
(a PTP inhibitor), and the fluorescence ratio was measured every 2 min.
The imaging system comprised an Olympus IX-70 inverted fluorescence
microscope, a Hamamatsu Photonics ICCD camera C2400, a Hamamatsu
Photonics Argus 50 image processor, and
a Hamamatsu Photonics
W-VIEW system A4313. The microscope was equipped with a xenon
lamp, an objective lens X 40, a 400DF15 excitation filter (OMEGA), a
420DCLP dichroic mirror (OMEGA), and a 435ALP emission filter
(OMEGA).
‡
8.71 (s, 1H); MS (FAB): m/z: 935 [M ‡H].
Preparation of 2: 10% Pd/C (10 mg) was added to a solution of 11 (20 mg,
21 mmol) in methanol, and the reaction mixture was stirred for 30 min
under a slight positive pressure of H2 gas using a balloon. The progress of
the reduction was followed by reverse-phase HPLC. When no more
starting material was evident and only the product peak was observed, the
reaction mixture was filtered and the filtrate was evaporated to give 2
(12 mg, 75%). 1H NMR (300 MHz, CD3OD): d ˆ 1.44 (m, 4H), 2.00 (m,
4H), 3.77 (s, 3H), 3.78 (m, 2H), 6.56 (dd, J ˆ 2.6, 8.6 Hz, 1H), 6.69 (d, J ˆ
2.6 Hz, 1H), 6.75 (d, J ˆ 2.4 Hz, 1H), 6.82 (d, J ˆ 8.4 Hz, 1H), 6.86 (dd, J ˆ
2.4, 8.6 Hz, 1H), 6.90 (d, J ˆ 8.6 Hz, 1H), 7.00 (s, 1H), 7.03 (d, J ˆ 8.6 Hz,
1H), 7.49 (d, J ˆ 1.5 Hz, 1H), 7.62 (dd, J ˆ 1.5, 8.2 Hz, 1H), 7.65 (d, J ˆ
8.4 Hz, 1H), 7.90 (d, J ˆ 8.2 Hz, 1H), 8.73 (s, 1H); 13C NMR (125 MHz,
CD3OD): d ˆ 31.9, 32.4, 49.3, 49.9, 56.2, 79.5, 102.0, 103.1, 103.5, 109.7,
111.6, 112.8, 113.3, 114.4, 115.7, 116.3, 117.8, 123.9, 126.2, 129.8, 130.2, 130.5,
130.8, 132.9, 142.7, 149.6, 153.3, 153.6, 154.5, 158.2, 163.2, 163.4, 163.7, 165.7,
Acknowledgement
This work was supported in part by the Ministry of Education, Science,
Sports and Culture of Japan (grants 11794026, 12470475 & 12557217 to
T.N., 13024217, 13558078, 1367232
& 14045210 to K.K.), by the
Mitsubishi Foundation, and by the Research Foundation for Opt-Science
and Technology. We also thank Dr. Kentaro Yamaguchi and Mr. Shigeru
Sakamoto for mass spectrometry.
[1] a) H. Kojima, N. Nakatsubo, K. Kikuchi, S. Kawahara, Y. Kirino, H.
Nagoshi, Y. Hirata, T. Nagano, Anal. Chem. 1998, 70, 2446 2453;
b) N. Umezawa, K. Tanaka, Y. Urano, K. Kikuchi, T. Higuchi, T.
Nagano, Angew. Chem. 1999, 111, 3076 3079; Angew. Chem. Int. Ed.
1999, 38, 2899 2901; c) T. Hirano, K. Kikuchi, Y. Urano, T. Higuchi,
T. Nagano, Angew. Chem. 2000, 112, 1094 1096; Angew. Chem. Int.
Ed. 2000, 39, 1052 1054
[2] J. R. Lakowicz, Principles of Fluorescence Spectroscopy, 2nd ed.,
Plenum, New York, 1999.
[3] A. Miyawaki, J. Llopis, R. Heim, J. M. McCaffery, J. A. Adams, M.
Ikura, R. Y. Tsien, Nature 1997, 388, 882 887.
[4] G. Zlokarnik, P. A. Negulescu, T. E. Knapp, L. Mere, N. Burres, L. X.
Feng, M. Whitney, K. Roemer, R. Y. Tsien, Science 1998, 279, 84 88.
[5] H. Takakusa, K. Kikuchi, Y. Urano, S. Sakamoto, K. Yamaguchi, T.
Nagano, J. Am. Chem. Soc. 2002, 124, 1653 1657.
[6] a) Y. Kawanishi, K. Kikuchi, H. Takakusa, S. Mizukami, Y. Urano, T.
Higuchi, T. Nagano, Angew. Chem. 2000, 112, 3580 3582; Angew.
Chem. Int. Ed. 2000, 39, 3438 3440; b) H. Takakusa, K. Kikuchi, Y.
Urano, T. Higuchi, T. Nagano,Anal. Chem. 2001, 73, 939 942.
[7] a) J. W. Gurd, Neurochem. Int. 1997, 31, 635 649; b) M. D. Haskell,
J. K. Slack, J. T. Parsons, S. J. Parsons, Chem. Rev. 2001, 101, 2425
2440; c) T. Hunter, Cell 1995, 80, 225 236.
[8] Z. Y. Zhang, Curr. Opin. Chem. Biol. 2001, 5, 416 423.
[9] X. L. Zhan, M. J. Wishart, K. L. Guan, Chem. Rev. 2001, 101, 2477
2496.
[10] R. Majeti, A. Weiss, Chem. Rev. 2001, 101, 2441 2448.
[11] Q. P. Wang, J. Scheigetz, M. Gilbert, J. Snider, C. Ramachandran,
Biochim. Biophys. Acta Protein Struct. Mol. Enzymol. 1999, 1431, 14
23.
‡
167.4, 170.2; HRMS (FAB ‡ ): m/z: calcd for [M ‡H]: 755.1642; found:
755.1638.
Preparation of 12: Ethylenediamine (60 mg, 1.0 mmol) was added to a
solution of 5 (800 mg, 0.81 mmol) and the N-hydroxysuccinimidyl ester of
7-hydroxycoumarin-3-carboxylic acid (240 mg, 0.79 mmol) in DMF, and
the mixture was stirred for 30 min at 08C. The reaction mixture was
extracted with ethyl acetate and the extract was washed with an aqueous
solution of citric acid and brine, and dried over sodium sulfate. After
evaporation of the ethyl acetate, the residue was purified by chromatog-
raphy on silica gel to afford 12 (180 mg, 20%). 1H NMR (300 MHz,
CDCl3): d ˆ 3.61 (m, 4H), 5.13 (d, J ˆ 9.0 Hz, 8H), 6.58 (d, J ˆ 8.8 Hz, 2H),
6.65 (dd, J ˆ 2.2, 7.3 Hz, 1H), 6.67 (d, J ˆ 2.2 Hz, 1H), 6.72 (dd, J ˆ 1.3,
8.8 Hz, 2H), 6.94 (d, J ˆ 1.3 Hz, 2H), 7.27 7.36 (m, 21H), 7.73 (s, 1H), 8.15
(s, 2H), 8.41 (m, 1H), 8.63 (s, 1H), 9.12 (m, 1H); MS (FAB): m/z: 1127
‡
[M ‡H].
Preparation of 3: 10% Pd/C (15 mg) was added to a solution of 12 (35 mg,
31 mmol) in methanol/chloroform 1:1 (5 mL), and the reaction mixture was
stirred for 10 min under a slight positive pressure of H2 gas using a balloon.
The progress of the reduction was followed by reverse-phase HPLC. When
no more starting material was evident and only the product peak was
observed, the reaction mixture was filtered and the filtrate was evaporated
1
to give 3 (21 mg, 90%). H NMR (300 MHz, CD3OD): d ˆ 3.50 3.65 (m,
4H), 6.55 (dd, J ˆ 2.4, 8.8 Hz, 1H), 6.63 (d, J ˆ 8.8 Hz, 1H), 6.71 (d, J ˆ
2.4 Hz, 1H), 6.75 (d, J ˆ 2.4 Hz, 1H), 6.79 (d, J ˆ 8.8 Hz, 1H), 6.87 (dd, J ˆ
2.4, 8.8 Hz, 1H), 6.93 (dd, J ˆ 2.4, 8.7 Hz, 1H), 7.20 (d, J ˆ 2.4 Hz, 1H), 7.62
(d, J ˆ 1.3 Hz, 1H), 7.63 (d, J ˆ 8.7 Hz, 1H), 8.07 (d, J ˆ 8.6 Hz, 1H), 8.12
(dd, J ˆ 1.3, 8.6 Hz, 1H), 8.62 (s, 1H); 13C NMR (75 MHz, CD3OD): d ˆ
40.0, 40.9, 83.6, 103.1, 103.5; 109.6, 112.7, 114.4, 115.7, 116.0, 118.1, 124.0,
126.4, 129.6, 130.4, 133.0, 142.8, 149.5, 153.1, 154.5, 154.7, 158.2, 161.0, 162.9,
Received: August 7, 2002
Revised: November 11, 2002 [F4326]
Chem. Eur. J. 2003, 9, No. 7
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0907-1485 $ 20.00+.50/0
1485