No-wash protein labeling with designed fluorogenic probes and application to real-time pulse-chase analysis

Article abstract of DOI:10.1021/ja208290f

Small molecule labeling techniques for cellular proteins under physiological conditions are very promising for revealing new biological functions. We developed a no-wash fluorogenic labeling system by exploiting fluorescence resonance energy transfer (FRET)-based fluorescein-cephalosporin- azopyridinium probes and a mutant β-lactamase tag. Fast quencher elimination, hydrophilicity, and high resistance against autodegradation were achieved by rational refinement of the structure. By applying the probe to real-time pulse-chase analysis, the trafficking of epidermal growth factor receptors between cell surface and intracellular region was imaged. In addition, membrane-permeable derivatization of the probe enabled no-wash fluorogenic labeling of intracellular proteins.

Full text of DOI:10.1021/ja208290f

Article
pubs.acs.org/JACS
No-Wash Protein Labeling with Designed Fluorogenic Probes and
Application to Real-Time Pulse-Chase Analysis
Shin Mizukami,†,‡ Shuji Watanabe,† Yuri Akimoto,† and Kazuya Kikuchi*,†,‡
†Division of Advanced Science and Biotechnology, Graduate School of Engineering, Osaka University, 2-1 Yamadaoka, Suita, Osaka
565-0871, Japan
‡Immunology Frontier Research Center (IFReC), Osaka University, 3-1 Yamadaoka, Suita, Osaka 565-0871, Japan
W
S
* Web-Enhanced Feature * Supporting Information
ABSTRACT: Small molecule labeling techniques for cellular
proteins under physiological conditions are very promising for
revealing new biological functions. We developed a no-wash
fluorogenic labeling system by exploiting fluorescence
resonance energy transfer (FRET)-based fluorescein-cepha-
losporin-azopyridinium probes and a mutant β-lactamase tag.
Fast quencher elimination, hydrophilicity, and high resistance
against autodegradation were achieved by rational refinement
of the structure. By applying the probe to real-time pulse-chase
analysis, the trafficking of epidermal growth factor receptors
between cell surface and intracellular region was imaged. In addition, membrane-permeable derivatization of the probe enabled
no-wash fluorogenic labeling of intracellular proteins.
property. In other words, the washout of the unlabeled probes
is required prior to microscopic imaging. Tetracysteine-tag
technology also requires the washing procedure for imaging
with a high signal/noise ratio to avoid nonspecific labeling.5
Because the washout procedure takes at least 15 min, precise
pulse-chase experiments with shorter time intervals are
practically impossible with conventional labeling technologies.
In addition, the complete washout of the unreacted probes
inside cells is difficult, because synthetic probes often
accumulate in various organelles.
We developed a fluorogenic protein labeling system that
exploits a mutant of 29-kDa TEM-1 β-lactamase (BL-tag) and
the synthetic β-lactam probes.6 This technology is quite
versatile, and various applications such as multicolor
fluorescence imaging,7 a pulse-chase experiment with lumines-
cent quantum dots via specific biotinylation,8 and time-resolved
protein imaging with a lanthanide-based probe9 have been
demonstrated. In addition, the use of FRET-based labeling
probes6 in this system conferred both specificity and a turn-on
fluorescence property. The labeling mechanism involves two
steps: an initial noncatalytic enzyme reaction and subsequent
quencher elimination via a self-immolative reaction (Scheme
1).
INTRODUCTION
■
By imaging protein dynamics, valuable information about target
proteins can be obtained, such as their intracellular localization,
expression and degradation timings, and interactions with other
proteins. Fluorescent proteins (FPs) have been utilized for this
purpose because advances in fluorescence microscopy have
enabled the real-time visualization of target proteins genetically
fused with FPs in living cells.1 Combining the use of FPs with
probe design strategies based on fluorescence resonance energy
transfer (FRET) has also led to the production of innovative
probes for biological studies.2 However, such genetic engineer-
ing approaches have critical limitations. For instance, we cannot
label target proteins expressing at specific timings by using FPs.
Although imaging the dynamics of cell cycle progression was
achieved by exploiting the cell-cycle dependent proteolysis of
two FP-labeled ubiquitination oscillators,3 the strategy cannot
be generally applied to other dynamic biological processes. In
addition, organic or inorganic chemistry-based synthetic
molecules provide a wider range of functions. Therefore,
techniques for labeling cellular proteins with synthetic
molecules under physiological conditions are obviously useful
and promising in revealing new biological functions.
Many research groups have reported tag-based protein
labeling techniques, in which the target proteins are genetically
fused with a peptide tag or a small-protein tag.4 These tags are
specifically modified with the corresponding small-molecule
probes by a chemical or enzyme reaction, coordination,
protein−ligand interaction, etc. By using such labeling
technologies, proteins expressed at specific timings can be
visualized. However, most labeling techniques, except the
tetracysteine-tag technology,4a do not exhibit a fluorogenic
Effective FRET quenching of the probes and fluorescence
recovery after in vitro protein labeling via a self-immolative
reaction have been observed. However, the no-wash labeling of
live-cell proteins has not been performed because of a slow
turn-on fluorescence response. In addition, the high hydro-
Received: September 2, 2011
Published: January 6, 2012
© 2012 American Chemical Society
1623
dx.doi.org/10.1021/ja208290f | J. Am. Chem.Soc. 2012, 134, 1623−1629

Journal of the American Chemical Society
Article
Scheme 1. Schematic Illustration of Fluorogenic Protein Labeling Mechanism Using Previous and New Probes Based on
a
Fluorescence Resonance Energy Transfer (FRET)
a
POI: protein of interest.
Chart 1. Structures of Synthesized Compounds
phobicity of the probes, mainly of the DABCYL group, caused
a tendency to accumulate in the hydrophobic regions of cells
such as plasma membranes, although nonspecific accumulation
disappeared after washing the cells. When analyzing the quick
dynamics of target proteins using pulse-chase experiments, the
labeling rate should be as high as possible. Thus, it is very
important to develop superior labeling probes that exhibit a
quick turn-on fluorescence response and thus enable real-time
pulse-chase experiments without the washing procedure.
However, no such probes have been reported thus far.
Herein, we report the rational development of a novel probe
that can exhibit a fast fluorogenic labeling response for BL-tag
technology. The probe did not require the washing procedure
in live cell imaging and thus provided an innovative protein
analytical method using real-time pulse-chase analysis. In
addition, the membrane-permeable derivatization of the probe
by the esterification enabled the no-wash fluorogenic labeling of
intracellular proteins.
fluorogenic labeling. As described above, the fluorogenic
labeling process is divided into two steps: namely, enzymatic
labeling and subsequent elimination reaction. Preliminary
absorption analysis of the synthesized labeling probes showed
that the nucleophilic enzyme reaction proceeds quickly and the
following elimination is the rate-determining step. Elimination
rates generally depend on the elimination ability of the leaving
groups, which is correlated with the pKa value of the conjugate
acid. The pKa of the thiophenol group of FCD (Chart 1) was
estimated to be about 6.5,10 and it was thus expected that a
decrease in pKa would accelerate the elimination rate. It has
been reported that azopyridinium-conjugated cephalosporin
induces fast elimination from the 3′-position after hydrolysis of
β-lactam, because the corresponding conjugate acid has a lower
pKa (4.3).11 Because of its fast elimination, hydrophilicity, and
absorption spectra, we chose 2-(4-dimethylaminophenylazo)-
pyridinium as the quencher. Although this azopyridinium
compound is not widely known as a FRET quencher, as
compared with DABCYL, the absorption spectral property
from 450 to 600 nm fulfills the requirement for quenching
green or orange fluorescence of fluorescein or tetramethylrhod-
amine, respectively. In addition, the hydrophilicity of
RESULTS
■
Design, Synthesis, And Photophysical and Biochem-
ical Properties of Novel Labeling Probes. We investigated
the optimization of the leaving quencher for no-wash
1624
dx.doi.org/10.1021/ja208290f | J. Am. Chem.Soc. 2012, 134, 1623−1629

Journal of the American Chemical Society
Article
Figure 1. Spectral and protein labeling properties of synthesized probes. All reactions were performed in 100 mM HEPES buffer (pH 7.4) at 25 °C.
(a) Time-dependent absorption spectra of 10 μM CAP1, obtained by incubation with 40 nM WT TEM. (inset) Time-dependent absorbance (λ =
570 nm) of CAP1 with (circle) or without (triangle) WT TEM. (b) Absorption spectra of 5 μM FCAP2 (dotted line) and 5 μM FCAPO2 (solid
line). (c, d) Time-dependent emission spectra of 500 nM FCAP2 (c, λex = 487 nm) or 500 nM FCAPO2 (d, λex = 490 nm), obtained by incubation
with 1 μM BL-tag. (inset) Time-dependent fluorescence intensity of 500 nM FCAP2 (c, λex = 487 nm, λem = 518 nm) or 500 nM FCAPO2 (d, λex
=
490 nm, λem = 518 nm) with (circle) or without (triangle) BL-tag. (e) Electrophoresis gel images of BL-tag incubated with FCAPO2 in the presence
or absence of HEK293T cell lysate. CBB: Coomassie Brilliant Blue staining. Fl: fluorescence. (f) Plot of initial labeling reaction rate v0 versus probe
concentration [FCAPO2]. [BL-tag] = 125 nM.
azopyridinium (calculated logP = −1.9) was expected to
enhance the practical utility in live cell imaging.
showed enough stability in neutral buffer solution (Figure
S1b, Supporting Information). Thus, the intramolecular
cyclization observed in the case of CAP1 could be completely
suppressed by linker elongation in CAP2.
We designed and synthesized CAP1 (Chart 1 and Scheme
S1, Supporting Information) as the key compound, which can
be conjugated with various fluorophores, to develop novel
fluorogenic probes. The elimination reaction rate of CAP1 was
investigated using wild-type TEM-1 β-lactamase (WT TEM).
The absorption maximum of CAP1 in HEPES buffer (pH 7.4)
was observed at 580 nm, and the peak quickly decreased by the
addition of WT TEM (Figure 1a). This spectral change
indicates that azopyridinium was eliminated, getting converted
into the neutral azopyridine form.11a However, the absorption
spectrum of CAP1 gradually changed in an enzyme-free buffer
solution (Figure 3a inset), probably because CAP1 degraded
automatically in neutral aqueous solution. This degradation was
also observed in dimethyl sulfoxide (DMSO)-d6. Furthermore,
fluorophore conjugation to the amino group was unsuccessful
because of the instability of CAP1. We suspected that the
instability might be caused by intramolecular aminolysis via the
six-membered ring formation (Scheme S2, Supporting
Information).
A new fluorescent probe, FCAP2 (Chart 1), was successfully
synthesized from CAP2 (Scheme S3, Supporting Information).
FCAP2 showed the absorption maxima at 487 nm (λ = 85 000
M−1 cm−1) and 589 nm (ε = 25 200 M−1 cm−1) for fluorescein
and azopyridinium, respectively (Figure 1b, dotted line). As
expected, the fluorescence of FCAP2 was largely quenched (Φ
= 0.03) by FRET and recovered by incubation with BL-tag
protein (Figure 1c). The fluorescence increase, as well as the
absorption spectral change, was very quick and completed
within a few minutes (Figure 1c inset, circle). This indicates the
great potential of FCAP2 in real-time fluorescence imaging.
However, the fluorescence of FCAP2 was unexpectedly
increased in the absence of BL-tag, probably as a result of
autohydrolysis (Figure 1c inset, triangle).
A previous report has suggested that stability of cepha-
losporin derivatives is improved by oxidizing the sulfides of
cephalosporins to sulfoxides.12 Therefore, FCAPO2 (Chart 1),
the oxidized compound of FCAP2, was designed and
synthesized (Scheme S4). FCAPO2 showed an absorption
spectrum very similar to that of FCAP2, with two peaks for
fluorescein (λ = 490 nm, ε = 89 000 M−1 cm−1) and
azopyridinium (λ = 581 nm, ε = 37 000 M−1 cm−1) (Figure
1b, solid line). The fluorescence of FCAPO2 was largely
quenched (Φ = 0.02) by FRET and quickly recovered by
To avoid the intramolecular cyclization, we designed an
improved synthetic scheme (Scheme S3, Supporting Informa-
tion) via CAP2 (Chart 1), which has an aminoethyl group
instead of the aminomethyl group of CAP1. As a result, CAP2
showed good stability in solution. CAP2 showed equally quick
absorption spectral change by the addition of WT TEM as
CAP1 (Figure S1a, Supporting Information). CAP2 also
1625
dx.doi.org/10.1021/ja208290f | J. Am. Chem.Soc. 2012, 134, 1623−1629

Journal of the American Chemical Society
Article
Figure 2. No-wash fluorogenic labeling of cell surface proteins. (a−d) HEK293T cells expressing BL-EGFR (a−c) or EGFR (d) were treated with
500 nM FA (a,b) or FCAPO2 (c,d) at 37 °C. Fluorescence images (λex = 473 nm) were captured using a confocal fluorescence microscope without
(a, c, d) or with (b) the washing procedure. (e, f) Time-lapse fluorogenic labeling of cell surface BL-EGFR with 10 nM FCAPO2 at 37 °C. Time-
course graph of the fluorescence images (e) and the averaged fluorescence intensity for the complete field of view (n = 4) (f). Scale bar: 20 μm. See
also Movie 1 in the HTML version of this paper.
incubation with BL-tag protein (Figure 1d). The fluorogenic
labeling and elimination reactions were very quick and
completed within a few minutes (Figure 1d inset, circle). The
stability of FCAPO2 in neutral aqueous solution was
considerably increased, as compared with that of FCAP2
(Figure 1d inset, triangle) and was considered enough for
practical use in physiological studies. With regard to specificity,
FCAPO2 labeled only BL-tag, even in the presence of
HEK293T cell lysate (Figure 1e). All these results indicate
that we successfully developed a highly stable labeling probe
that showed quick fluorogenic response after specific protein
labeling.
Then, the precise labeling kinetics of FCAPO2 toward BL-
tag was studied. To a solution of the excess tag protein (125
nM), 1.25−25 nM FCAPO2 was added, and the initial reaction
rate calculated from the fluorescence increase (Figure S4,
Supporting Information) was plotted against the probe
concentration (Figure 1f). As a result, the bimolecular labeling
rate constant was determined to be 7.8 × 104 M−1 s−1. This
value indicates that the combination of FCAPO2 and BL-tag is
the fastest fluorogenic labeling system among the highly
selective labeling technologies, as shown in Table S1 in the
Supporting Information.
Fluorogenic Labeling of Cell Surface Proteins with
FCAPO2. Fluorogenic labeling of cell surface proteins with
FCAPO2 was examined using living cells. BL-tag-fused
epidermal growth factor receptor (BL-EGFR) proteins were
expressed in HEK293T cells, following treatment with
FCAPO2 or nonfluorogenic FA (Figure S2a, Supporting
Information) at a concentration of 500 nM. After 10 min of
incubation, the cells were observed with a confocal fluorescence
microscope without washout of the probe. In the case of FA,
high background fluorescence signals were observed in the
medium (Figure 2a), although the fluorescence labeling of the
membrane proteins was visible after the washing procedure
(Figure 2b), as previously reported.7 On the other hand, when
the cells were incubated with FCAPO2, green fluorescence
increased along the plasma membranes within a few minutes
without the washing procedure (Figure 2c). This fluorescence
increase was not observed in the cells expressing EGFR without
BL-tag (Figure 2d).
Next, we performed time-lapse imaging of protein labeling
with FCAPO2. Under confocal fluorescence microscopy, 10
nM FCAPO2 was added to HEK293T cells expressing BL-
EGFR. Consecutive capture of cell images without the washing
procedure showed time-dependent fluorescence enhancement
along the plasma membranes (Figure 2e and Movie 1). The
fluorogenic labeling rate of the living cell surface proteins was
investigated in time-lapse fluorescence images. Even with highly
diluted probe ([FCAPO2] = 10 nM), the labeling rate was fast
and labeling was almost completed within 15 min (Figure 2f).
Based on the results of these in vitro and live cell experiments,
we conclude that this new labeling system is faster than
previously reported fluorogenic labeling systems, including our
previous reports and the SNAP-tag-based fluorogenic labeling
system reported.13
Real-Time Imaging of Cell Surface Expression and
Internalization of Target Proteins by Using Dual Probes.
Although imaging of protein trafficking is one of the promising
applications of protein labeling techniques, the washing
procedures in the current technologies restrict wide application
of the techniques. Therefore, we performed real-time pulse-
chase analysis of protein expression on cell surface by using
FCAPO2. As pulse-chase analysis visualizes target protein
expression at a specific time, quick protein labeling that does
not require the washing procedure would be very effective for
pulse-chase analysis of protein trafficking with high temporal
resolution.
We constructed a protein analytical method, in which
internalization of cell surface proteins and cell surface
1626
dx.doi.org/10.1021/ja208290f | J. Am. Chem.Soc. 2012, 134, 1623−1629

Journal of the American Chemical Society
Article
Figure 3. Real-time pulse-chase analysis and no-wash fluorogenic labeling of intracellular proteins. (a, b) Real-time fluorescence imaging of BL-
EGFR trafficking by using FCAPO2 and RA. HEK293T cells were labeled with 100 nM RA prior to incubation with 20 nM FCAPO2 at 37 °C. (a)
(top row) Fluorescence images (λex = 559 nm) for RA, (middle row) fluorescence images (λex = 473 nm) for FCAPO2, (bottom row) merged
images of fluorescence and phase contrast microscopy. (b) Magnified image of the square marked in the bottom row in part a. (c) Intracellular
protein labeling with FCAPO2-DA (λex = 473 nm). HEK293T cells expressing BL-NLS (top) and nontransfected cells (bottom) were incubated
with 5 μM FCAPO2-DA and 100 ng mL−1 Hoechst33342 at 37 °C for 2 h. Scale bar: 20 μm.
expression of genetically identical proteins can be discrim-
inatively visualized. In the protocol, BL-EGFR proteins
displayed on the cell surface were labeled in advance with a
membrane-impermeable rhodamine probe, RA (Figure S2b,
Supporting Information). After the culture medium was
replaced with a fresh medium, FCAPO2 was added, and cell
images were captured using a confocal fluorescence microscope
(Figure 3a and b). Initially, only the red fluorescence of RA was
detected around the plasma membranes. Thereafter, the red
fluorescence signals gradually internalized and formed bright
spots in the cells in a time-dependent manner. This trafficking
indicates the endocytosis of EGFR caused by the stimulation of
epidermal growth factor (EGF) included in the serum. On the
other hand, the green fluorescence of FCAPO2 was scarcely
observed in the beginning. However, the green fluorescence
signals gradually increased along the plasma membranes,
probably due to the fluorogenic labeling of newly expressed
proteins on the cell surface. These results suggest that this
imaging methodology can discriminately visualize EGFR
trafficking from the cell surface to the intracellular region, as
well as in the opposite direction.
Fluorogenic Labeling of Intracellular Proteins with
the Cell-Permeable Probe FCAPO2-DA. We developed cell-
permeable labeling probes for intracellular proteins by utilizing
bacampicillin, which is a clinical penicillin-type β-lactam
prodrug developed for improving intestinal absorption.14
However, these probes do not have fluorogenic property
because it is very difficult to incorporate leaving groups in the
bacampicillin structure. Therefore, we extended the probe
design of FCAPO2 to achieve fluorogenic labeling of
intracellular proteins. FCAPO2 has three anionic groups and
one cationic group in its structure. Of the three anionic groups,
two are in the fluorescein structure, and the remaining anionic
carboxylate and the cationic group are in cephalosporin and
azopyridinium, respectively. Because the latter two groups are
very close, they are likely to form an intramolecular ion pair.
Hence, we hypothesized that the protection of only two anionic
groups in fluorescein in FCAPO2 might enable probe
introduction into living cells, although previous reports suggest
that the carboxylate of cephalosporin-based probes needs to be
protected to permeate living cell membranes.15
We acetylated FCAPO2 to obtain FCAPO2-DA (Chart 1
and Scheme S5 in the Supporting Information). The absorption
spectra of FCAPO2-DA indicated that it was quickly hydro-
lyzed by WT TEM (Figure S3a, Supporting Information),
although it was considerably stable for several hours in aqueous
solution (Figure S3b, Supporting Information). FCAPO2-DA
showed no fluorescence (Figure S3c, Supporting Information),
similar to the other diacetylated fluorescein derivatives.16
To assess the intracellular labeling property of FCAPO2-DA,
we used a BL-tag protein that localizes in cell nuclei, BL-tag-
fused nuclear localization signal (BL-NLS).14 BL-NLS has three
consecutive simian virus 40 (SV40) large T antigen NLS at the
C terminus of BL-tag. HEK293T cells expressing BL-NLS were
incubated with FCAPO2-DA, and the cell permeability and the
fluorogenic labeling property of FCAPO2-DA were confirmed
by confocal fluorescence microscopy. As expected, FCAPO2-
DA exhibited cell permeability, and the fluorescence gradually
accumulated at the cell nuclei (Figure 3c top). No noteworthy
fluorescence accumulation was detected in the nuclei of the
control cells transfected with empty vector plasmids (Figure 3c
bottom), even though a slight fluorescence increase was
observed in whole cells, which is now under investigation. As
a result, we finally developed a cell-permeable fluorogenic
probe to label intracellular proteins. This probe, FCAPO2-DA,
enabled no-wash intracellular protein labeling, as in the case of
extracellular protein labeling with FCAPO2.
DISCUSSION
■
The key problem in developing a practical fluorogenic protein
labeling technology by extending our previous strategy6 was the
slow quencher elimination after labeling. In principle,
elimination reaction rate is correlated with the pKa of the
conjugate acid of the leaving group. The pKa of 2-(4-
dimethylaminophenylazo)pyridinium, which is the conjugate
acid of the leaving group in FCAPO2, is 4.3.10 This value is
much smaller than that of thiophenol, which is the leaving
group in our previous probes such as CCD and FCD. The fast
turn-on fluorescence property of FCAP2 indicates that the pKa
1627
dx.doi.org/10.1021/ja208290f | J. Am. Chem.Soc. 2012, 134, 1623−1629

Journal of the American Chemical Society
Article
of the leaving quencher was indeed an important factor for the
probe design strategy. However, the fast elimination step
induced undesirable autodegradation of FCAP2. This instability
was overcome by oxidizing the sulfur atom in the design of
FCAPO2. In addition, the linker length between the
fluorophore and the β-lactam was optimized to achieve high
synthetic yields.
tization of FCAPO2 to FCAPO2-DA enabled no-wash
fluorogenic labeling of intracellular proteins. No-wash fluoro-
genic labeling for intracellular proteins is very rare, although we
have developed an alternative no-wash labeling method by
using highly diluted prodrug-based β-lactam probes.14 Our no-
wash fluorogenic labeling system described herein would
evidently be an innovative technique that provides various
applications in cell biology, one of which is described in this
paper. Moreover, the fluorophore can be extended to near-
infrared (NIR) fluorescent dyes, which are more suitable for in
vivo studies. These results and discussion confirm the practical
utility and promise of the mutant β-lactamase-tag-based protein
labeling technology in biological studies, especially those on
protein trafficking, gene expression, protein degradation, etc.
Another essence of the probe design is the physical
properties of the leaving quencher. The quencher in
FCAPO2 is the cationic azopyridinium form, and it converts
into the neutral azopyridine form after elimination. The
absorption spectrum of the azopyridinium shows the maximum
at 581 nm and broadly ranges from 500 to 600 nm. Thus, this
compound works as an efficient quencher for fluorescein. This
wavelength region is also suitable for the quenching of orange
to light-red fluorescent dyes such as rhodamines. Although
DABCYL is often used as the quencher in many quenching
FRET-based probes,17 the shorter-wavelength absorption (λmax
= 462 nm) restricts its application to red fluorescent dyes. The
higher molar extinction coefficient of the azopyridinium (ε =
37 000 M−1 cm−1) than that of DABCYL (ε = 25 500 M−1
cm−1)7 also confirms its utility in quenching FRET applications.
Hydrophilicity of azopyridinium provides another practical
advantage in biological experiments. In live cell imaging,
hydrophobic probes often cause localization at membranes or
subcellular organelles. With regard to this point, FCAPO2
including azopyridinium showed less accumulative properties,
as compared with DABCYL-based probes. In addition, the
cationic aromatic heterocycle may compensate the anionic
charge of the adjacent carboxy group in cephalosporin by
forming an inner salt, because FCAPO2-DA permeates living
cells without the protection of the carboxylate in cephalosporin.
In widely used protocols for other protein labeling
techniques, at least 15-min incubation with the probes and
subsequent washing procedure are required.18 Such procedures
restrict temporal resolution and diverse experimental designs.
In this study, we achieved real-time pulse-chase analysis, and
discriminatively visualized the internalization of cell surface-
displayed proteins and the translocation of newly expressed
proteins to plasma membranes. This innovative real-time
imaging provided the information about how fast protein
expression occurs on the plasma membranes and how fast the
proteins enter the cells. Conventional methods using FPs
cannot afford to discriminate proteins that have the same
amino acid sequence but different functions or history. Urano
et al. and Correa, Jr. et al. separately reported similar real-time
fluorogenic labeling probes based on SNAP-tag technology.13,19
However, our BL-tag-based system with FCAPO2 exhibited
much faster labeling kinetics (Table S1, Supporting Informa-
tion) and requires a lower probe concentration.
ASSOCIATED CONTENT
■
S
* Supporting Information
Materials, instruments, synthesis and characterization of
compounds, experimental procedures, and supplementary
figures and schemes. This material is available free of charge
via the Internet at http://pubs.acs.org.
W
* Web-Enhanced Features
Movie 1, showing fluorogenic labeling of cell surface BL-EGFR
with 10 nM FCAPO2 at 37 °C over 20 min, is available in the
HTML version of this paper.
AUTHOR INFORMATION
■
Corresponding Author
kkikuchi@mls.eng.osaka-u.ac.jp
ACKNOWLEDGMENTS
■
This work was supported by the Japan Society for the
Promotion of Science (JSPS) through its Funding Program
for World-Leading Innovative R&D on Science and Technol-
ogy (FIRST Program), the Grant-in-Aid for Scientific Research
from the Ministry of Education, Culture, Sports, Science, and
Technology (MEXT) of Japan, by the CREST funding
program from the Japan Science and Technology Agency
(JST), and by the Grant-in-Aid from the Ministry of Health,
Labor, and Welfare (MHLW) of Japan. K.K. is thankful for
support from the Takeda Science Foundation and from the
Naito Foundation. S.M. is thankful for support from Inamori
Foundation and from Sumitomo Foundation. K.K. and S.M. are
thankful for support from Asahi Glass Foundation. S.W. is
thankful for support from a Global COE Fellowship of Osaka
University and a JSPS Research Fellowship.
REFERENCES
■
(1) Chudakov, D. M.; Lukyanov, S.; Lukyanov, K. A. Trends
Biotechnol. 2005, 23, 605−613.
CONCLUSION
■
(2) Truong, K.; Ikura, M. Curr. Opin. Struct. Biol. 2001, 11, 573−578.
(3) Sakaue-Sawano, A.; Kurokawa, H.; Morimura, T.; Hanyu, A.;
Hama, H.; Osawa, H.; Kashiwagi, S.; Fukami, K.; Miyata, T.; Miyoshi,
H.; Imamura, T.; Ogawa, M.; Masai, H.; Miyawaki, A. Cell 2008, 132,
487−498.
(4) (a) Griffin, B. A.; Adams, S. R.; Tsien, R. Y. Science 1998, 281,
269−272. (b) Keppler, A.; Gendreizig, S.; Gronemeyer, T.; Pick, H.;
Vogel, H.; Johnsson, K. Nat. Biotechnol. 2003, 21, 86−89. (c) Chen, I.;
Howarth, M.; Lin, W.; Ting, A. Y. Nat. Methods 2005, 2, 99−104.
(d) Yin, J.; Liu, F.; Li, X.; Walsh, C. T. J. Am. Chem. Soc. 2004, 126,
7754−7755. (e) O’Hare, H. M.; Johnsson, K.; Gautier, A. Curr. Opin.
Struct. Biol. 2007, 17, 488−494. (f) Hinner, M. K.; Johnsson, K. Curr.
Opin. Biotechnol. 2010, 21, 766−776. (g) Rabuka., D. Curr. Opin.
In conclusion, we developed second-generation fluorogenic
labeling probes by exploiting an azopyridinium compound as
the leaving quencher. By rational optimization of the probe
structures, we reached the design of FCAPO2, which
specifically labeled BL-tag proteins and very quickly recovered
its fluorescence. In a detailed kinetic study of fluorogenic
labeling, the bimolecular reaction rate was the highest among
the known selective and fluorogenic protein labeling methods.
The outstanding properties of FCAPO2 provided a versatile
cell surface protein labeling system that does not require any
washing procedure. In addition, membrane-permeable deriva-
1628
dx.doi.org/10.1021/ja208290f | J. Am. Chem.Soc. 2012, 134, 1623−1629

Journal of the American Chemical Society
Article
Chem. Biol. 2010, 14, 790−796. (h) Sunbul, M.; Yin, J. Org. Biomol.
Chem. 2009, 7, 3361−3371.
(5) (a) Stroffekova, K.; Proenza, C.; Beam, K. G. Pfluegers Arch. 2001,
442, 859−886. (b) Machleidt, T.; Robers, M.; Hanson, G. T. Methods
in Molecular Biology; Humana Press: Totowa, 2007; Vol. 356, pp 209−
220.
(6) Mizukami, S.; Watanabe, S.; Hori, Y.; Kikuchi, K. J. Am. Chem.
Soc. 2009, 131, 5016−5017.
(7) Watanabe, S.; Mizukami, S.; Hori, Y.; Kikuchi, K. Bioconjug.
Chem. 2010, 21, 2320−2326.
(8) Yoshimura, A.; Mizukami, S.; Hori, Y.; Watanabe, S.; Kikuchi, K.
ChemBioChem 2011, 12, 1031−1034.
(9) Mizukami, S.; Yamamoto, T.; Yoshimura, A.; Watanabe, S.;
Kikuchi, K. Angew. Chem., Int. Ed. 2011, 50, 8750−8752.
(10) Dean, J. A. Lange’s Handbook of Chemistry, 14th ed.; McGraw-
Hill: New York, 1972.
(11) (a) Jones, R. N.; Wilson, H. W.; Novick, W. J. J. Clin. Microbiol.
1982, 15, 677−683. (b) Faraci, W. S.; Pratt, R. F. J. Am. Chem. Soc.
1984, 75, 1489−1490.
(12) Gao, W.; Xing, B.; Tsien, R. Y.; Rao, J. J. Am. Chem. Soc. 2003,
125, 11146−11147.
(13) Komatsu, T.; Johnsson, K.; Okuno, H.; Bito, H.; Inoue, T.;
Nagano, T.; Urano, Y. J. Am. Chem. Soc. 2011, 133, 6745−6751.
(14) Watanabe, S.; Mizukami, S.; Akimoto, Y.; Hori, Y.; Kikuchi, K.
Chem.Eur. J. 2011, 17, 8342−8349.
(15) Zlokarnik, G.; Negulescu, P. A.; Knapp, T. E.; Mere, L.; Burres,
N.; Feng, L.; Whitney, M.; Roemer, K.; Tsien, R. Y. Science 1998, 279,
84−88.
(16) Takakusa, H.; Kikuchi, K.; Urano, Y.; Kojima, H.; Nagano, T.
Chem.Eur. J. 2003, 9, 1479−1485.
(17) Marras, S. A. E. Mol. Biotechnol. 2008, 38, 247−255.
(18) HaloTag Technology: Focus on Imaging, Technical Manual;
Promega: U.S.A., 2009; Available online: http://www.promega.com/.
(19) Sun, X.; Zhang, A.; Baker, B.; Sun, L.; Howard, A.; Buswell, J.;
Maurel, D.; Masharina, A.; Johnsson, K.; Noren, C. J.; Xu, M. Q.;
Correa, I. R. ChemBioChem. 2011, 12, 2217−2226.
1629
dx.doi.org/10.1021/ja208290f | J. Am. Chem.Soc. 2012, 134, 1623−1629