Disassembly-driven turn-on fluorescent nanoprobes for selective protein detection

Article abstract of DOI:10.1021/ja101879g

"Switchable" fluorescent probes, which induce changes in the fluorescence properties (e.g., intensity and/or wavelength) only at the intended target protein, are particularly useful for selective protein detection or imaging. However, the strategy for designing such smart probes remains very limited. We report herein a novel mechanism for generating protein-specific "turn-on" fluorescent probes. Our approach uses an amphiphilic, self-assembling compound consisting of a fluorophore and a protein ligand. In the absence of target protein, the probe forms self-assembled aggregates in aqueous solution and displays almost no fluorescence because of efficient quenching. On the other hand, it emits bright fluorescence in response to the target protein through recognition-induced disassembly of the probe. On the basis of this strategy, we successfully developed three types of fluorescent probes that allow the detection of carbonic anhydrase, avidin, and trypsin via turn-on emission signals. It is anticipated that the present supramolecular approach may facilitate the development of new protein-specific switchable fluorescent probes that are useful for a wide range of applications, such as diagnosis and molecular imaging.

Full text of DOI:10.1021/ja101879g

Published on Web 05/12/2010
Disassembly-Driven Turn-On Fluorescent Nanoprobes for Selective Protein
Detection
Keigo Mizusawa,† Yoshiyuki Ishida,† Yousuke Takaoka,† Masayoshi Miyagawa,† Shinya Tsukiji,†,‡,§
and Itaru Hamachi*,†,‡
Department of Synthetic Chemistry and Biological Chemistry, Graduate School of Engineering, Kyoto UniVersity,
Katsura, Kyoto 615-8510, Japan, and Core Research for EVolutional Science and Technology (CREST), Japan
Science and Technology Agency, 5 Sanbancho, Chiyoda-ku, Tokyo 102-0075, Japan
Received March 5, 2010; E-mail: ihamachi@sbchem.kyoto-u.ac.jp
Scheme 1. Disassembly-Based Strategy for Specific Protein
Detection with Self-Assembling Turn-On Fluorescent Probes
Selective protein detection or imaging is of fundamental impor-
tance in basic biological research as well as medical diagnosis.
Small-molecule “switchable” fluorescent probes, which induce
changes in the fluorescence properties (intensity and/or wavelength)
only at the intended target protein, are particularly useful for this
purpose because they allow sensitive, simple, and specific detection
with high target-to-background ratios.1 Several strategies for
generating such smart probes have been described. The most
prevalent probes are fluorogenic substrates that allow the monitoring
of specific hydrolases (e.g., ꢀ-lactamase, glycosidases, and pro-
teases) by either turn-on fluorescence2 or fluorescence resonance
energy transfer (FRET).3 The fluorescence switching of this type
of probe is based on the catalytic cleavage of the molecule by the
target enzyme. Substrate-based probes for other enzymes, such as
kinases, have also been reported.4 On the other hand, the design of
switchable probes for nonenzymatic protein targets remains more
challenging. A typical approach involves the incorporation of a
solvatochromic fluorophore to a ligand specific to the target protein.5
Unfortunately, in most cases, the resultant probes are always
fluorescent, causing high background signals, and induce only small-
to-moderate fluorescence changes in protein sensing. Recently,
peptide-based molecular beacons were prepared as switchable
probes for proteins that bind specific short peptide sequences.6 A
limitation of the beacon system is its applicability only to
polypeptide-binding protein receptors. Clearly, the establishment
of new switching strategies for protein-specific fluorescent probes
is highly desired.
Here we introduce a novel mechanism for the design of “turn-
on” fluorescent probes that are selective for target proteins. The
strategy was inspired by our recent work on self-assembling 19F
NMR/MRI probes developed for protein imaging.7 We demon-
strated that an amphiphilic compound consisting of an 19F reporter
and a protein ligand forms self-assembled aggregates in aqueous
solution (signal off) and that the aggregates disassemble in response
to the target protein through the specific protein-ligand interaction
(signal on). This switching mechanism based on the self-assembly
and recognition-driven disassembly of amphiphilic probes is
considered to be unique and applicable to the generation of a new
type of switchable fluorescent probe. It is well-known that many
fluorophores undergo quenching when they form dimers or ag-
gregates.8 Therefore, we reasoned that a self-assembling, ligand-
tethered fluorophore would function as a protein-specific probe as
shown in Scheme 1. Namely, it was expected that the self-
assembling probe would display no or only weak fluorescence when
Figure 1. Self-assembling turn-on fluorescent probes for protein detection.
(a) Chemical structures of fluorescent probes 1 and 2 for hCA, 3 for avidin,
and 4 for trypsin. (b) Fluorescence spectra (λex ) 468 nm) of probe 1 (25
µM) in 50 mM HEPES buffer (pH 7.2) in the absence (blue) and presence
(red) of hCA (25 µM) and after addition of EZA (125 µM) to the solution
of 1 and hCA (black). (c) Photograph of probe 1 (25 µM) in the absence
(left) and presence (middle) of hCA (25 µM) and after addition of EZA
(125 µM) to the solution of 1 and hCA (right). The image was obtained
with UV excitation (λex ) 365 nm).
aggregated, whereas it would emit bright fluorescence in response
to the target protein through the recognition-induced disassembly
of the probe.
To test this idea, we chose human carbonic anhydrase I (hCA)
as a model protein and designed compounds 1 and 2 (Figure 1a
and Schemes S1 and S2 in the Supporting Information). On the
basis of the previous study,7 a fluorophore, BODIPY in 1 or
† Kyoto University.
‡ CREST.
§ Present address: Top Runner Incubation Center for Academia-Industry Fusion,
Nagaoka University of Technology, 1603-1 Kamitomioka, Nagaoka, Niigata 940-
2188, Japan.
9
10.1021/ja101879g  2010 American Chemical Society
J. AM. CHEM. SOC. 2010, 132, 7291–7293 7291

C O M M U N I C A T I O N S
nm (0.5 µM) to 572 nm (25 µM) with increasing the probe
concentration. Probe 1 dissolved in DMSO (i.e., in a monomeric
state) showed an absorption maximum at 565 nm (data not shown).
Therefore, the observed red shift of the absorption maximum
suggested that the BODIPY fluorophore changed from the mono-
meric state to the aggregated state, in particular, in a J-type
orientation,8 as the concentration increased. In agreement with this
behavior, a substantial decrease in the fluorescence was observed
with increasing probe concentration (above 2.5 µM) after a linear
increase in the fluorescence intensity up to a probe concentration
of 1 µM (Figure S3). These data strongly support the conclusion
that 1 undergoes self-assembly in aqueous solution, which causes
efficient fluorescence quenching. Subsequently, an hCA titration
experiment was carried out. Absorption spectra of a solution
containing 1 (25 µM) showed that when the hCA concentration
was increased, the broad red-shifted absorption at 572 nm gradually
decreased and a sharp peak at 565 nm linearly increased (Figure
2b). These data suggest the disruption of the aggregates of 1 by
hCA. Consistent with this, the emission of 1 at 580 nm, which
was originally quenched in the absence of hCA, was significantly
increased by the addition of hCA (Figure 2c). The fluorescence
response was rapid, becoming complete within 1 min following
the hCA addition. The change in the emission intensity was
saturated at a 1:1 1/hCA ratio. The fluorescence enhancement
showed a linear relationship with hCA concentration up to 25
µM (1 equiv). From this plot, we determined the detection limit
for hCA to be 70 nM (Figure S4), which is much superior (by
∼70-fold) to that achieved with the hCA-specific 19F NMR probe
(5 µM).7
Figure 2. Spectroscopic analyses of hCA-specific probe 1. (a) UV-vis
absorption spectra of probe 1 alone: 0.5 µM (red); 25 µM (blue). (b) UV-vis
absorption spectral changes of probe 1 (25 µM) upon addition of hCA (0-75
µM). (c) Fluorescence spectral changes of probe 1 (25 µM) upon the addition
of hCA (0-75 µM) (λex ) 468 nm). Fluorescence measurements were
performed 1 min after adding hCA to the probe solution. The inset shows
the fluorescence titration curve (λem ) 580 nm); error bars represent standard
deviations of three experiments. (d) DLS analysis of the particle size
distribution of self-assembled probe 1 (25 µM). All of the experiments were
performed in 50 mM HEPES buffer (pH 7.2).
The self-assembly/disassembly properties of 1 were further
confirmed with several other techniques. Atomic force microscopy
(AFM) revealed the formation of spherical or oval aggregates of 1
with diameters ranging from 100 to 200 nm (Figure S5). In dynamic
light scattering (DLS) measurements, a buffer solution containing
only 1 (25 µM) showed aggregates with a mean diameter of 100
nm (Figure 2d), whereas negligible DLS intensity was observed
after addition of hCA to the solution. On the basis of all the data,
we can illustrate the fluorescence off/on switching mechanism as
follows: (1) probe 1 self-assembles to form submicrometer-sized
aggregates in aqueous solution;10 (2) in the aggregated state, 1 emits
almost no fluorescence because of efficient quenching; (3) the
aggregates disassemble in the presence of hCA through binding of
the protein to the ligand moiety of 1; and (4) the disassembly
process leads to the recovery of fluorescence.
tetramethylrhodamine (TMR) in 2, was attached as a hydrophobic
moiety to benzenesulfonamide (SA), an hCA-targeting ligand
(hydrophilic head),9 through a relatively hydrophobic linker. We
first examined the response of each probe to hCA by monitoring
the UV and fluorescence spectral changes in aqueous solution. As
shown in Figure 1b, the emission was extremely weak [the
fluorescence quantum yield (Φ) was 0.001] when 1 was dissolved
in aqueous buffer but dramatically increased (by 38-fold) upon the
addition of hCA (Φ ) 0.04). The fluorescence recovery was so
efficient that the brighter emission in the presence of hCA was easily
detectable by the naked eye (Figure 1c). On the other hand, probe
2 showed a relatively large fluorescence even in the absence of
hCA, and thus, no significant fluorescence enhancement occurred
upon addition of the protein (only by 1.5-fold) (Figure S1 in the
Supporting Information). These results indicated that the rather
hydrophobic BODIPY, not the positively charged TMR, was
suitable as a fluorophore for clear off/on response. Interestingly,
the enhanced fluorescence of 1 was almost completely quenched
again by the addition of EZA, a strong competitive inhibitor for
hCA (Figure 1b).9 This demonstrates that the turn-on fluorescence
switching is reversible and controlled by the recognition of the
ligand moiety of the probe by hCA. The target specificity of probe
1 was next assessed under protein mixture conditions. Only weak
fluorescence was observed when 1 was mixed with a solution
containing bovine serum albumin, hemoglobin, concanavalin A,
and chymotrypsin. In sharp contrast, a strong emission appeared
upon the addition of 1 to a mixture containing the four proteins
and hCA, which reveals that 1 is hCA-specific (Figure S2).
Next we performed titration experiments in detail. First, the
concentration-dependent changes in the absorption and emission
properties of 1 alone in buffer were investigated. As shown in Figure
2a, we observed a red shift of the absorption maximum from 565
Because of the modular feature of the probe design, we
successfully synthesized the similar probes targeting different
proteins simply by replacing the ligand motif. In order to target
avidin, for example, compound 3 containing the biotin ligand was
prepared (Figure 1a and Scheme S3).11 Similarly, a benzamidine
derivative was used as a ligand for the detection of trypsin (i.e.,
compound 4) (Scheme S4).12 As expected, in both cases, the
fluorescence intensity was very weak when the probe was alone in
solution. On the other hand, the fluorescence was dramatically
enhanced by 22- and 23-fold upon the addition of the corresponding
target proteins avidin and trypsin (Figure 3a and Figure S6). In the
latter case, the enhanced fluorescence was quenched again by the
addition of an excess amount of free benzamidine (data not shown).
More significantly, none of the probes 1, 3, or 4 responded to
nontarget proteins, demonstrating the high orthogonality and target
specificity of these probes (Figure 3b and Figure S7). A series of
orthogonal, turn-on fluorescence detection of proteins was clearly
observable by the naked eye.
9
7292 J. AM. CHEM. SOC. VOL. 132, NO. 21, 2010

C O M M U N I C A T I O N S
was partly supported by the CK Integrated Medical Bioimaging
Project (MEXT) and CREST (Japan Science and Technology
Agency).
Supporting Information Available: Experimental details, Figures
S1-S7, and Schemes S1-S4. This material is available free of charge
via the Internet at http://pubs.acs.org.
References
(1) (a) Lavis, L. D.; Raines, R. T. ACS Chem. Biol. 2008, 3, 142. (b) Kobayashi,
H.; Ogawa, M.; Alford, R.; Choyke, P. L.; Urano, Y. Chem. ReV. 2010,
110, 2620.
(2) (a) Haugland, R. P. The Handbook: A Guide to Fluorescent Probes and
Labeling Technologies, 10th ed.; Invitrogen: Carlsbad, CA, 2005. (b) Gao,
W.; Xing, B.; Tsien, R. Y.; Rao, J. J. Am. Chem. Soc. 2003, 125, 11146.
(c) Kamiya, M.; Kobayashi, H.; Hama, Y.; Koyama, Y.; Bernardo, M.;
Nagano, T.; Choyke, P. L.; Urano, Y. J. Am. Chem. Soc. 2007, 129, 3918.
(3) (a) 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.
(b) Mizukami, S.; Kikuchi, K.; Higuchi, T.; Urano, Y.; Mashima, T.;
Tsuruo, T.; Nagano, T. FEBS Lett. 1999, 453, 356. (c) Takakusa, H.;
Kikuchi, K.; Urano, Y.; Sakamoto, S.; Yamaguchi, K.; Nagano, T. J. Am.
Chem. Soc. 2002, 124, 1653. (d) Pham, W.; Choi, Y.; Weissleder, R.; Tung,
C.-H. Bioconjugate Chem. 2004, 15, 1403.
Figure 3. Detection of several different proteins using turn-on fluorescent
probes. (a) Fluorescence spectral changes of probe 4 (10 µM) upon the
addition of trypsin (0-100 µM) in 50 mM HEPES buffer, 0.5 M NaCl
(pH 7.2) (λex ) 480 nm). The inset shows the fluorescence titration curve
(λem ) 575 nm). (b) Photographs showing the orthogonal, specific protein
detection of hCA, avidin, and trypsin using probes 1, 3, and 4, respectively.
The images were obtained with UV excitation (λex ) 365 nm). For detailed
conditions, see the Supporting Information.
(4) Lawrence, D. S.; Wang, Q. ChemBioChem 2007, 8, 373.
(5) Loving, G. S.; Sainlos, M.; Imperiali, B. Trends Biotechnol. 2010, 28, 73.
(6) (a) Oh, K. J.; Cash, K. J.; Plaxco, K. W. J. Am. Chem. Soc. 2006, 128,
14018. (b) Thurley, S.; Ro¨glin, L.; Seitz, O. J. Am. Chem. Soc. 2007, 129,
12693.
(7) Takaoka, Y.; Sakamoto, T.; Tsukiji, S.; Narazaki, M.; Matsuda, T.; Tochio,
H.; Shirakawa, M.; Hamachi, I. Nat. Chem. 2009, 1, 557. Also see the
accompanying paper: Azagarsamy, M. A.; Thayumanavan, S. Nat. Chem.
2009, 1, 523.
(8) (a) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 2nd ed.;
Kluwer Academic/Plenum: New York, 1999. (b) Ogawa, M.; Kosaka, N.;
Choyke, P. L.; Kobayashi, H. ACS Chem. Biol. 2009, 4, 535.
(9) (a) Taylor, P. W.; King, R. W.; Burgen, A. S. V. Biochemistry 1970, 9,
2638. (b) Winum, J.-Y.; Dogne´, J.-M.; Casini, A.; de Leval, X.; Montero,
J.-L.; Scozzafava, A.; Vullo, D.; Innocenti, A.; Supuran, C. T. J. Med.
Chem. 2005, 48, 2121.
In summary, we have developed new fluorescent molecules that
detect specific proteins with turn-on fluorescence signals. The
switching mechanism is based on the self-assembly (fluorescence
off) and recognition-driven disassembly (fluorescence on) of
amphiphilic, ligand-tethered fluorophores. The probe design is
modular and thus applicable for specific detection of various
proteins, including enzymes and nonenzymatic proteins. It is
expected that not only the ligand motif but also the fluorophore
unit can be replaced with other fluorescent dyes with different
emission wavelengths. We anticipate that the present supramolecular
approach13 may facilitate the development of new protein-specific
switchable fluorescent probes that are useful for a wide range of
applications, such as diagnosis and molecular imaging. In addition,
along with widely used inorganic particles and polymer-based
assemblies, self-assembled aggregates formed by small organic
compounds may hold great promise as a new type of nanometer-
sized materials in the field of nanobiotechnology.
(10) From the concentration dependence of the fluorescence and DLS measure-
ments, the critical aggregation concentration for self-assembly of 1 was
found to be less than 2.5 µM.
(11) Green, N. M. Biochem. J. 1963, 89, 585.
(12) Talhout, R.; Villa, A.; Mark, A. E.; Engberts, J. B. F. N. J. Am. Chem.
Soc. 2003, 125, 10570.
(13) Recent important examples of supramolecular-disassembly-based protein-
sensing systems: (a) Nam, J.-M.; Thaxton, C. S.; Mirkin, C. A. Science
2003, 301, 1884. (b) You, C.-C.; Miranda, O. R.; Gider, B.; Ghosh, P. S.;
Kim, I.-B.; Erdogan, B.; Krovi, S. A.; Bunz, U. H. F.; Rotello, V. M. Nat.
Nanotechnol. 2007, 2, 318. (c) Savariar, E. N.; Ghosh, S.; Gonza´lez, D. C.;
Thayumanavan, S. J. Am. Chem. Soc. 2008, 130, 5416. (d) Azagarsamy,
M. A.; Sokkalingam, P.; Thayumanavan, S. J. Am. Chem. Soc. 2009, 131,
14184. (e) Azagarsamy, M. A.; Yesilyult, V.; Thayumanavan, S. J. Am.
Chem. Soc. 2010, 132, 4550.
Acknowledgment. We thank Prof. Kenji Matsuda and Mr.
Takashi Hirose for help with DLS measurements and Dr. Masato
Ikeda for help with AFM measurements. K.M. and Y.T. acknowl-
edge JSPS Research Fellowships for Young Scientists. This work
JA101879G
9
J. AM. CHEM. SOC. VOL. 132, NO. 21, 2010 7293