Simultaneous detection of ATP and GTP by covalently linked fluorescent ribonucleopeptide sensors

Article abstract of DOI:10.1021/ja3097652

A noncovalent RNA complex embedding an aptamer function and a fluorophore-labeled peptide affords a fluorescent ribonucleopeptide (RNP) framework for constructing fluorescent sensors. By taking an advantage of the noncovalent properties of the RNP complex, the ligand-binding and fluorescence characteristics of the fluorescent RNP can be independently tuned by taking advantage of the nature of the RNA and peptide subunits, respectively. Fluorescent sensors tailored for given measurement conditions, such as a detection wavelength and a detection concentration range for a ligand of interest can be easily identified by screening of fluorescent RNP libraries. The noncovalent configuration of a RNP becomes a disadvantage when the sensor is to be utilized at very low concentrations or when multiple sensors are applied to the same solution. Here, we report a strategy to convert a fluorescent RNP sensor in the noncovalent configuration into a covalently linked stable fluorescent RNP sensor. This covalently linked fluorescent RNP sensor enabled ligand detection at a low sensor concentration, even in cell extracts. Furthermore, application of both ATP and GTP sensors enabled simultaneous detection of ATP and GTP by monitoring each wavelength corresponding to the respective sensor. Importantly, when a fluorescein-modified ATP sensor and a pyrene-modified GTP sensor were co-incubated in the same solution, the ATP sensor responded at 535 nm only to changes in the concentration of ATP, whereas the GTP sensor detected GTP at 390 nm without any effect on the ATP sensor. Finally, simultaneous monitoring by these sensors enabled real-time measurement of adenosine deaminase enzyme reactions.

Full text of DOI:10.1021/ja3097652

Article
pubs.acs.org/JACS
Simultaneous Detection of ATP and GTP by Covalently Linked
Fluorescent Ribonucleopeptide Sensors
Shun Nakano,† Masatora Fukuda,† Tomoki Tamura,† Reiko Sakaguchi,† Eiji Nakata,†,‡
and Takashi Morii*,†,‡
‡
†Institute of Advanced Energy, Kyoto University, and CREST, JST, Uji, Kyoto 611-0011, Japan
S
* Supporting Information
ABSTRACT: A noncovalent RNA complex embedding an
aptamer function and a fluorophore-labeled peptide affords a
fluorescent ribonucleopeptide (RNP) framework for con-
structing fluorescent sensors. By taking an advantage of the
noncovalent properties of the RNP complex, the ligand-
binding and fluorescence characteristics of the fluorescent
RNP can be independently tuned by taking advantage of the
nature of the RNA and peptide subunits, respectively.
Fluorescent sensors tailored for given measurement con-
ditions, such as a detection wavelength and a detection concentration range for a ligand of interest can be easily identified by
screening of fluorescent RNP libraries. The noncovalent configuration of a RNP becomes a disadvantage when the sensor is to be
utilized at very low concentrations or when multiple sensors are applied to the same solution. Here, we report a strategy to
convert a fluorescent RNP sensor in the noncovalent configuration into a covalently linked stable fluorescent RNP sensor. This
covalently linked fluorescent RNP sensor enabled ligand detection at a low sensor concentration, even in cell extracts.
Furthermore, application of both ATP and GTP sensors enabled simultaneous detection of ATP and GTP by monitoring each
wavelength corresponding to the respective sensor. Importantly, when a fluorescein-modified ATP sensor and a pyrene-modified
GTP sensor were co-incubated in the same solution, the ATP sensor responded at 535 nm only to changes in the concentration
of ATP, whereas the GTP sensor detected GTP at 390 nm without any effect on the ATP sensor. Finally, simultaneous
monitoring by these sensors enabled real-time measurement of adenosine deaminase enzyme reactions.
often results in a decrease or loss of the affinity or selectivity to
the substrate.
INTRODUCTION
■
Fluorescent biosensors that are able to detect specific ligands
with high sensitivity have been developed from macromolecular
receptors.1 The first step in the construction of fluorescent
biosensors often starts with the search for natural receptors that
possess ideal affinity and selectivity for the target molecule.
Otherwise, a novel receptor with necessary affinity and
selectivity has to be constructed. For conversion of receptors
into fluorescent biosensors, mutants of the receptor are
designed by examination of the position of mutation in which
substrate binding is effectively converted into optical signals
based on the 3-dimensional structural information of the
receptors.2 The sensor that has the desired substrate-binding
and fluorescent properties is then constructed by chemical
modification of the fluorophore to the mutant, followed by
evaluation of its function as a sensor. Fluorescent biosensors
have been developed by various methods based on this general
concept. However, it is still difficult to predict whether a
receptor that is modified by a fluorophore at a specific position
could effectively respond to a ligand of interest at the desired
fluorescent wavelength, even based on detailed 3-dimensional
structural information of the receptor. Indeed, many designed
sensors do not necessarily show desired fluorescent responses.
Moreover, modification of the receptor with a fluorophore
We previously reported a candidate fluorescent sensor that
utilized a ribonucleopeptide (RNP) framework.3 In this prior
study, an RNA-derived RNP library was prepared by
introducing a randomized nucleotide region as a ligand-binding
domain adjacent to the Rev Responsive Element (RRE) RNA
segment of an RRE-Rev peptide complex4 in a structure-based
manner. An in vitro selection method5 was applied to the RNA-
based RNP library to afford a series of RNP receptors for given
targets.3a−e In the second step, substitution of the Rev peptide
by the fluorophore-modified Rev peptide of the selected RNP
receptor afforded a fluorescent RNP complex that showed
fluorescence intensity changes upon binding to the target
molecule without loss of affinity or selectivity of the parent
RNP receptor. The RNA subunits obtained by the in vitro
selection of the RNP library usually display many kinds of
sequences, even sharing the same consensus sequence. Because
each RNP shows a different affinity to the target molecule,
these RNA subunits are used as a library that binds to the target
with various affinities and structures. On the other hand, a
library of fluorophore-modified Rev peptides is easily
Received: October 3, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja3097652 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
constructed by modifying the Rev peptide with fluorophores
that have a variety of excitation and emission wavelengths. By
taking advantage of the characteristic of the RNP as a
noncovalent complex, combinations of the RNA subunit library
and the fluorescent Rev library afford a pool of fluorescent
RNPs with various characters. From the fluorescent RNP
library, fluorescent RNP sensors for targets, such as ATP,3c,d
phosphotyrosine residues in defined amino acid sequences,3e
and dopamine,3h were obtained that respond at detection or
excitation wavelengths and have ligand affinities and ligand
concentration ranges of interest. For example, ATP sensors
were obtained with the dissociation constants ranging from 2.2
to 156 μM, and with the detection wavelengths from 390 to
670 nm.
covalent RNP sensors are capable of detecting multiple ligands
simultaneously in the same solution at different wavelengths
with low sensor concentrations, even in cellular extracts.
RESULTS AND DISCUSSION
■
Construction of Covalently Linked Fluorescent RNP
Sensors. Noncovalent fluorescent RNP sensors for nucleotide
triphosphates were constructed as previously reported.3c To
construct a stable fluorescent RNP sensor, the 3′-terminal of
RNA subunit and the C-terminal of Rev peptide were
covalently tethered by a hydrazone bond formation12 (Figure
1). This reaction has been utilized as an immobilization method
The noncovalent RNP sensor complex is also characterized
by a modular structure in which the ligand-binding RNA
module and the fluorophore-labeled Rev peptide and RRE
RNA complex, a sensing module, are connected through a
spacer RNA. Modular design of fluorescent sensors is therefore
possible by using an existing RNA aptamer as the sensing
module. A GTP-binding RNA aptamer isolated by in vitro
selection is utilized as the ligand-binding module of the RNP
sensor by fusing it to RRE RNA, and successive complexation
with the fluorophore-labeled Rev peptide affords a fluorescent
GTP sensor.3g,i Almost all RNP sensors exhibited fluorescence
quenching upon RNP complex formation and restored their
intrinsic fluorescence after substrate binding.3c−i
This stepwise strategy conveniently provides fluorescent
RNP sensors with various detection wavelengths and detection
ranges for a given target by combination of RNP aptamers and
fluorophore-modified peptides, but the noncovalent config-
uration becomes a disadvantage for practical measurements, for
instance, when the sensor concentration is reduced. Previously
reported bimolecular sensors, such as signaling aptamers6 and
aptamer sensors noncovalently embedded with a fluorophore,7
would face the same situation. In addition, simultaneous
application of multiple noncovalent sensors that consist of the
same Rev/RRE scaffold would suffer from exchanges between
the subunits. Therefore, this system needs an improvement for
its orthogonal usage. Furthermore, development of a strategy to
construct stable fluorescent biosensors will facilitate simulta-
neous detection of two or more cellular signaling molecules at
different detection wavelengths. Such a set of biosensors will be
quite useful for understanding cellular signaling pathways, even
when applied to cellular extracts in vitro.
An effective method to overcome the disadvantages of the
noncovalent character of receptors has been reported for
antibody engineering,8 such as in the modification of
immunoglobulin Fv fragments that form as a result of the
noncovalent complexing of heterodimers of the heavy-chain
variable domain (VH) and the light-chain variable domain
(VL).9 The stability of Fv fragments was increased by covalently
linking each chain with a disulfide bond10 or connecting VH and
VL proteins with a flexible linker peptide to form a single-chain
Fv molecule.11 This covalent linking strategy would also be
applicable to the modular RNP sensor. A covalent linkage
between the RNA and the peptide subunits is expected not only
to enhance the stability of the RNP complex, but also to
prevent the exchange of the subunits between multiple RNP
sensors.
Figure 1. Schematic illustration of the construction of covalently
linked fluorescent RNP sensor. 3′-Ribose of the RNA subunit was
converted to 3′-dialdehyde by treating with sodium periodate. A
flexible peptide linker (GGSGGSGGSG) with a carboxyhydrazine was
introduced at the C-terminal of the fluorophore-modified Rev peptide.
A covalently linked fluorescent RNP sensor was synthesized through a
coupling of these modified RNA and peptide subunits via hydrazone
bond formation.
of nucleic acids or proteins to solid supports12a,e,h or as a cross-
linking method of nucleic acids and/or proteins.12f The RNA
subunit of ATP-binding RNP was treated with sodium
periodate to form a 3′-dialdehyde group. A flexible linker
with ten amino acid residues (GGSGGSGGSG) was introduced
at the C-terminal of the Rev peptide, and a hydrazine group was
introduced to the C-terminal. The length of the linker
GGSGGSGGSG was determined by reference to the 3-
dimensional structure of the complex of the Rev peptide and
RRE RNA (Figure S1).4 A linker with seven amino acid
residues (GGSGGSG) seemed too short to tether the Rev
peptide to RRE RNA (Figure S1a). No obvious steric
hindrance was observed for the Rev/RRE complexes with
linker over seven amino acid residues (Figure S1b,c).
Therefore, the linker with ten amino acid residues was utilized
for the covalent linkage between the C-terminal of Rev peptide
and the 3′-end of RRE RNA (Figure S1c).
The noncovalent complexes of fluorescent RNP sensors for
ATP and GTP were converted to covalently linked RNPs to
confirm the above strategy. Among the ATP-binding RNP
receptors with various affinity to ATP selected from the RNP
library,3c A26 RNP was chosen as a representative RNP because
A26 RNP binds to ATP and ATP-derivatives including Ado
with moderate affinities,3c,f which is ideal for the measurement
of the changes in ATP concentration to reduce the interference
from the ATP endogenous to the cellular extracts. In addition,
A26 RNP shows measurable changes in fluorescence intensities
upon binding to the targets with many types of fluorophore
(Pyr, NBD, Cy5, etc.).3c In a similar manner, G23 RNP was
Here, we report the synthesis of stable fluorescent RNP
sensors by covalently linking the RNA and fluorophore-
modified Rev peptide subunits and demonstrate that the
B
dx.doi.org/10.1021/ja3097652 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
chosen from the selected GTP-binding RNP receptors3c
because of its moderate affinity to GTP and fluorescent
response. Purified RNA subunits of these RNPs, A26 RNA and
G23 RNA, were treated with freshly prepared sodium periodate
to modify the 3′-terminal ribose to 3′-dialdehyde by periodate
oxidation. The peptide subunit for the covalent linkage was
synthesized by introducing a peptide linker (GGSGGSGGSG)
at the C-terminus, coupling a fluorophore to the N-terminal of
the Rev peptide, and derivatized to hydrazide by hydrazine
treatment of the resin. 5-Carboxyfluorescein, 1-pyrenesulfonyl
chloride, and Cy5 were conjugated to the N-terminal amino
group of the Rev peptide to give 5FAM-Rev-HZ, Pyr-Rev-HZ,
and Cy5-Rev-HZ, respectively. These peptides were purified by
HPLC and characterized by MALDI-TOF mass spectrometry.
Subsequently, the modified RNA and peptide subunits were
covalently linked together. Coupling reactions of A26 RNA 3′-
dialdehyde and 5FAM-Rev-HZ or G23 RNA 3′-dialdehyde and
Pyr-Rev-HZ were performed in sodium acetate buffer (pH 5.2)
at 37 °C. The resulting covalently linked RNP complex was
purified by phenol-chloroform extraction, ethanol precipitation,
and denaturing PAGE to remove the unreacted RNA and
peptide. The isolated yields of covalently linked RNPs of A26
RNA with 5FAM-Rev-HZ (c-A26/5FAM-Rev) and G23 RNA
with Pyr-Rev-HZ (c-G23/Py-Rev) were 36% and 32%,
respectively. The purities of the covalently linked RNP sensors
were checked by denaturing PAGE (Figure S2). After PAGE
purification and isolation, c-A26/5FAM-Rev and c-G23/Pyr-
Rev were characterized by MALDI-TOF mass spectroscopy. c-
An16/Cy5-Rev and c-G23/5FAM-Rev were prepared using the
same procedure. The coupling yields for the native and
denaturing conditions for the RNP complex were similar to
each other (Figure S3), indicating that the formation of RNP is
not a prerequisite for the reaction.
Fluorescent Responses of the Covalent RNP Com-
plexes. The fluorescence responses of the covalently linked
ATP sensor (c-A26/5FAM-Rev), the covalently linked GTP
sensor (c-G23/Pyr-Rev), and each corresponding noncovalent
derivative (i.e., A26/5FAM-Rev and G23/Pyr-Rev, respec-
tively) were evaluated in the presence of increasing
concentrations of ATP or GTP (Figure 2). Relative
fluorescence intensities (I/I0) of ATP sensors were calculated
by dividing the fluorescence intensity in the presence (I) of
ATP by the fluorescence intensity in the absence (I0) of ATP
(Figure 2). Relative fluorescence intensities were plotted
against the ATP concentration to calculate the dissociation
constant (KD) for the complex between the RNP sensor and
ATP. The saturation value of the change in relative fluorescence
intensity was found to be 1.6 for c-A26/5FAM-Rev (100 nM),
which was similar to the value of 1.7 for A26/5FAM-Rev (1000
nM). KD values for the ATP complexes of c-A26/5FAM-Rev
(100 nM) and A26/5FAM-Rev (1000 nM) were 6.3 and 13.4
μM, respectively (Figure 2a,b). The binding affinity of
covalently linked c-A26/5FAM-Rev was higher than that of
noncovalent A26/5FAM-Rev.
Figure 2. Saturation curves for the relative fluorescence intensity
changes of c-A26/5FAM-Rev and A26/5FAM-Rev by titration with
ATP in a buffer containing 10 mM Tris-HCl (pH 7.6), 100 mM NaCl,
10 mM MgCl2, and 0.005% Tween 20. (a) Titration of the ATP-
binding fluorescent RNP complexes c-A26/5FAM-Rev (1 nM, purple
reverse triangles; 10 nM, green triangles; 50 nM, blue squares; 100
nM, red circles) with ATP at 4 °C and (b) A26/5FAM-Rev (50 nM,
purple reverse triangles; 100 nM, green triangles; 500 nM, blue
squares; 1000 nM, red circles) with ATP at 4 °C are shown. (c)
Titration profiles of the ATP-binding fluorescent RNP complexes c-
A26/5FAM-Rev and (d) A26/5FAM-Rev with ATP at different
temperatures (4 °C, red circles; 15 °C, blue squares; 25 °C, green
triangles; 37 °C, purple reverse triangles) are shown. The dissociation
constant (KD) and the maximum relative fluorescence intensity for the
ATP complex of each RNP are shown in the inset.
covalently linked sensor c-A26/5FAM-Rev showed higher
tolerance to increasing temperatures, even at a sensor
concentration of 100 nM (Figure 2c), than that of A26/
5FAM-Rev (1000 nM; Figure 2d).
The covalently linked GTP sensor c-G23/Pyr-Rev showed a
higher affinity to GTP than that of the noncovalent sensor
G23/Pyr-Rev (Figure 3a). The I/I0 value for c-G23/Pyr-Rev
and GTP at the saturation was 3.1, which was significantly
higher than that of the parent noncovalent GTP sensor G23/
Pyr-Rev (1.8). Again, the covalently linked sensor c-G23/Pyr-
Rev showed higher tolerance to increasing temperatures
(Figure 3b) than that of G23/Pyr-Rev (Figure 3c).
These results indicated that the noncovalent RNP sensors for
ATP or GTP were converted to a covalently linked RNP sensor
without diminishing sensing functions for ATP or GTP,
respectively. The covalently linked RNP sensor exhibited
much higher stability at low concentrations and maintained
the fluorescence response of individual RNP sensors.
Simultaneous Detection of ATP and GTP by Fluo-
rescent RNP Sensors. Because the fluorophore-modified Rev
peptide binds to a common RRE RNA sequence, a unique
noncovalent RNP sensor must be applied for each separate
sample solution to reduce the possible exchange between
different subunits of multiple RNP species. In contrast, the
The KD value of the Rev-RRE complex has been shown to be
a few nanomolar at similar buffer conditions to those used in
this study.13 The covalently linked sensor c-A26/5FAM-Rev
provided distinct fluorescent responses, even at 1 nM (Figure
2a). On the other hand, the noncovalent sensor A26/5FAM-
Rev exhibited almost no fluorescent response when the RNP
concentration was reduced to 100 nM (Figure 2b), indicating
that the noncovalent RNP complex was mostly dissociated into
each subunit. Furthermore, the fluorescence response of the
C
dx.doi.org/10.1021/ja3097652 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 4. (a,b) Simultaneous detection of ATP and GTP by covalently
linked ATP and GTP sensors at different wavelengths in the same
solution. A solution containing both the ATP sensor c-A26/5FAM-
Rev (100 nM) and the GTP sensor c-G23/Pyr-Rev (100 nM) in the
presence of 100 μM ATP and/or 100 μM GTP was measured at (a)
535 nm or at (b) 390 nm at 4 °C to selectively detect ATP or GTP.
(c,d) Fluorescence responses of noncovalent ATP and GTP sensors at
different wavelengths in the same solution in the presence of ATP,
GTP, or both ATP and GTP. A solution containing both the ATP
sensor A26/5FAM-Rev (1000 nM) and the GTP sensor G23/Pyr-Rev
(1000 nM) in the presence of 100 μM ATP and/or 100 μM GTP was
measured at (c) 535 nm or at (d) 390 nm at 4 °C under the mixing
condition.
Figure 3. (a) Saturation curves for the relative fluorescence intensity
changes of c-G23/Pyr-Rev (100 nM, red circles) and G23/Pyr-Rev
(100 nM, blue squares) by titration with GTP in a buffer containing 10
mM Tris-HCl (pH 7.6), 100 mM NaCl, 10 mM MgCl2, and 0.005%
Tween 20 at 4 °C. (b) Titration profiles of c-G23/Pyr-Rev and (c)
G23/Pyr-Rev with GTP at different temperatures (4 °C, red circles; 15
°C, blue squares; 25 °C, green triangles; 37 °C, purple reverse
triangles) are shown. The dissociation constant (KD) and the
maximum relative fluorescence intensity for the GTP complex of
each RNP are shown in the inset.
covalently linked RNP sensor forms an intramolecular complex
that prevents such an exchange between the subunits, even
when several types of RNP sensors coexist in the solution.
To test this, the covalently linked sensors c-A26/5FAM-Rev
and c-G23/Pyr-Rev were applied to monitor the concentrations
of ATP and GTP in the same solution at different wavelengths.
The ATP sensor c-A26/5FAM-Rev shares the same consensus
sequence with the ATP-binding RNP A26/5FAM-Rev, which
selectively binds ATP over GTP.3f The parent RNP of the
covalent GTP sensor, G23/Pyr-Rev, has been shown to
distinguish GTP from ATP.3c When monitored at 535 nm,
the ATP sensor c-A26/5FAM-Rev responded to concentration
changes in ATP, but not to changes in the concentration of
GTP (Figure 4a), and showed a similar response that was
observed in the absence of the GTP sensor c-G23/Pyr-Rev
(Figure 2a). Similarly, the GTP sensor c-G23/Pyr-Rev
specifically detected GTP at 390 nm without any interference
from ATP and the ATP sensor c-A26/5FAM-Rev in the same
solution (Figure 4b). In contrast, the solution containing both
the noncovalent sensors A26/5FAM-Rev and G23/Pyr-Rev
showed fluorescence responses to both ATP and GTP at the
detection wavelengths of 5FAM and Pyr (Figure 4c,d). This
result suggests that exchange of 5FAM-Rev and Pyr-Rev has
occurred for the noncovalent RNP sensors in the solution.
These results indicate that ATP and GTP in the same solution
are specifically detected by c-A26/5FAM-Rev and c-G23/Pyr-
Rev, respectively, in the presence of other substrate and/or the
other sensor.
afforded a set of covalently linked fluorescent sensors for the
simultaneous detection of ATP and GTP. By using the RNA
subunit An16 derived from an ATP-binding RNP3a and a Cy5-
modified Rev peptide, the covalently linked RNP sensor c-
An16/Cy5-Rev, which responded to ATP with a KD lower than
200 nM at 670 nm, was obtained (Figure S4a). Likewise,
covalent linking of G23 RNA with a fluorescein-labeled Rev
peptide (5FAM-Rev) afforded c-G23/5FAM-Rev, which
detected GTP at 535 nm with a KD of 2.0 μM (Figure S4b).
When the two sensors exist in the same solution, the ATP
sensor c-An16/Cy5-Rev responded at 670 nm to ATP with a
similar KD value determined under the single sensor condition
(KD < 200 nM) (Figures S4c and S4e). Furthermore, ATP
titration in the presence of equal amounts of GTP did not show
significant interference of GTP (KD for ATP < 200 nM)
(Figure S4 c and e). Similarly, the binding ability of GTP sensor
c-G23/5FAM-Rev to GTP was not interfered in the presence of
ATP sensor (KD = 2.0 μM) and ATP (KD = 2.3 μM). Thus, the
two sensors c-An16/Cy5-Rev and c-G23/5FAM-Rev were
amenable to simultaneous measurement of their respective
ligands at different detection wavelengths in the same solution.
These results demonstrate that simultaneous detection of
multiple ligands is possible by applying covalently linked RNP
sensors that sense their respective ligands at different detection
wavelengths. In addition, a variety of sensors that respond at
the target concentration range with the desired detection
wavelength would be prepared by the combination of an
appropriate RNA subunit and a fluorophore-labeled Rev
peptide.
The combination of an appropriate RNA subunit of an ATP-
or GTP-binding RNP and a fluorophore-modified Rev peptide
D
dx.doi.org/10.1021/ja3097652 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Simultaneous Detection of a Substrate and a Product
in an Enzymatic Reaction by Covalently Linked
Fluorescent RNP Sensors. Covalent RNP sensors were
next applied for real-time monitoring of an enzymatic
conversion of adenosine (Ado) to inosine (Ino) catalyzed by
adenosine deaminase (ADA; Figure 5a).14 In order to monitor
affinity and sufficient relative fluorescence intensity changes to
Ino, and importantly, G23/Pyr-Rev possessed distinct
selectivity to Ino over Ado in our GTP-binding RNPs.3c
Titration analyses for each sensor revealed that c-A26/5FAM-
Rev responded specifically to changes in Ado concentrations at
535 nm, with a KD of 11.4 μM (Figure 5b), while c-G23/Pyr-
Rev monitored Ino concentrations at 380 nm, with a KD of 108
μM (Figure 5c). The GTP sensor c-G23/Pyr-Rev showed
much lower affinity to Ino than that to GTP, but exhibited a
distinctive selectivity to Ino over Ado (Figure 5c).
Next, the covalent sensors c-A26/5FAM-Rev and c-G23/Pyr-
Rev were mixed in a solution containing Ado (100 μM). After
incubation for 30 min at 15 °C, ADA (0.1 mU) was added to
the solution to start the reaction. When monitored at 535 nm,
the fluorescence intensity of the ATP sensor c-A26/5FAM-Rev
decreased with the reaction time, indicating that c-A26/5FAM-
Rev responded to decreases in Ado concentrations (Figure 5d).
Measurement of the fluorescence intensity of c-G23/Pyr-Rev at
380 nm showed a gradual increase with incubation time, which
corresponded to the time-dependent production of Ino. The
reaction of ADA was also evaluated by HPLC to quantitatively
assess the consumption of Ado and the production of Ino
(Figures S6 and S7). Quantification of the peak area of each
nucleoside (i.e., Ado and Ino) in the enzymatic reaction
afforded actual changes in the concentrations of each
nucleoside. The fluorescence intensity changes monitored by
the RNP sensors were also converted to time-dependent
concentration changes for Ado and Ino using the quantitative
titration curves (Figure S8).
These time-course profiles were overlaid to compare changes
in the concentrations of Ado and Ino as obtained by HPLC and
RNP sensors (Figure 5e). Both profiles for changes in Ado
concentrations obtained by the RNP sensor and the HPLC
analysis were almost identical, other than the initial phase (up
to 30 min) of the enzymatic reaction. In addition, the time-
course profile for the formation of Ino monitored by c-G23/
Pyr-Rev was also almost identical to the profile obtained by
quantitative HPLC analysis. Therefore, time-dependent
changes in the concentrations of Ado and Ino in the ADA
reaction were simultaneously quantified using the covalent
sensors c-A26/5FAM-Rev and c-G23/Pyr-Rev, respectively,
across the range of times and substrate concentrations studied.
Thus, this covalent strategy for generating RNP sensors
afforded multiple fluorescent detecting system that were
suitable for monitoring the conversion process within a given
concentration range. Further practical employment of the
covalent RNP sensors including the real-time monitoring of
other enzymatic reaction process, will be promised by
preparation of RNA subunit with various characteristics such
as the specificity and/or selectivity toward ligands.
Enhancement of the Stability of Covalent RNP
Sensors in Cellular Extracts. The covalent linkage of RNP
subunits increased the thermodynamic stability of RNP. At the
same time, such a modification was expected to enhance the
tolerance of RNP to degradation in cellular culture media. To
evaluate the stability of the covalent RNP sensor in the cellular
media, the covalent ATP sensor was titrated by ATP in HeLa
cell extracts at ambient temperature. An ATP-binding RNP
receptor (A26/Rev) with a moderate KD value3a was utilized for
the titration experiments to reduce the interference from the
ATP endogenous to the cellular extracts (Figure 6a).
Figure 5. (a) Conversion of adenosine (Ado) to inosine (Ino)
catalyzed by adenosine deaminase (ADA). The covalently linked RNP
sensors for ATP and GTP responded to changes in the concentrations
of Ado and Ino, respectively. Saturation curves for the relative
fluorescence intensity changes of c-A26/5FAM-Rev and c-G23/Pyr-
Rev by Ado and Ino in a buffer containing 10 mM Tris-HCl (pH 7.6),
100 mM NaCl, 10 mM MgCl2, and 0.005% Tween 20 at 15 °C are
shown. Titration analyses were performed for (b) c-A26/5FAM-Rev
(100 nM) and (c) c-G23/Pyr-Rev (100 nM) with Ado (circles) or Ino
(squares). Observed equilibrium dissociation constants (KD) and the
maximum relative fluorescence intensity change (I/I0) are shown in
the inset. (d) Time-dependent changes in the fluorescence intensities
of the covalent sensors c-A26/5FAM-Rev (circles) and c-G23/Pyr-Rev
(squares), which monitored the conversion of Ado to Ino by ADA. (e)
Overlay plots of time-dependent changes in the concentrations of Ado
and Ino were quantified by the covalently linked RNP sensors c-A26/
5FAM-Rev and c-G23/Pyr-Rev and by HPLC. HPLC quantification
data were converted to the corresponding concentrations of Ado (pink
filled triangles) and Ino (purple filled diamonds). Fluorescence
intensity changes of c-RNP sensors were converted to changes in
the concentrations of Ado (red open circles) and Ino (blue open
squares) using their respective titration curves (Figure S8).
the concentration changes of Ado and Ino in the same solution
at different wavelengths, c-A26/5FAM-Rev and c-G23/Pyr-Rev
were utilized. ATP sensors can recognize Ado with similar
affinity and fluorescence response to ATP3f (Figure S5).
Covalently linked c-G23/Pyr-Rev was used for Ino detection in
this assay because the parent G23/Pyr-Rev showed moderate
The noncovalent A26/5FAM-Rev complex showed higher
stability than A26RNA alone in the cell extract at 4 °C,3e but an
E
dx.doi.org/10.1021/ja3097652 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Application of the covalently linked ATP and GTP sensors
enabled the simultaneous real-time quantification of Ado and
Ino in the reaction catalyzed by ADA. Each sensor responded
exclusively to its ligand and permitted the simultaneous
detection of changes in the concentration of ligands induced
by an enzymatic reaction in the restricted concentration range.
Because the combination of RNA and a fluorophore-modified
peptide library afforded fluorescence responses in various
concentration ranges, this strategy provided an appropriate set
of fluorescent sensors for the detection of multiple ligands in
the desired concentration range. Once a fluorescent RNP
sensor with a desired property is obtained, the covalent linking
strategy reported here allows development of fluorescent RNP
sensors for simultaneous detection of multiple ligands by
monitoring each wavelength corresponding to the respective
ligand.
Unlike the RNA-based aptamer sensors, the covalently
modified RNP sensors cannot be in situ generated inside the
cell. However, the cell-permeable Rev peptide might be utilized
to translocate the sensor into the cell. The covalently linked
RNP sensors studied here show moderate relative fluorescence
intensity changes of 1.2−3-fold upon binding the substrate.
Because the fluorescent sensors with such moderate relative
intensity changes were utilized as detection tools of signaling
molecules in the cell,16 further optimization in the response and
stability of covalently linked RNP sensors would open their
application in the cell.
Figure 6. (a) Titration analyses of noncovalent A26/5FAM-Rev (200
nM, open triangles) and covalent c-A26/5FAM-Rev (200 nM, open
circles) following addition of ATP in HeLa cell extracts upon
incubation for 30 min at ambient temperature. (b) Evaluation of the
stabilities of A26/5FAM-Rev (red bars) and c-A26/5FAM-Rev (blue
bars) in HeLa cell extracts. Each sensor was mixed with ATP (1 mM)
and incubated in HeLa cell extracts at ambient temperature for the
indicated times. The change in relative fluorescent intensity was
measured by a Wallac ARVOsx 1420 multilabel counter (Perkin-
Elmer).
incubation of A26/5FAM-Rev in HeLa cell extracts at ambient
temperature for 30 min led to a complete loss of the
fluorescence response to the ATP titration (Figure 6a). In
contrast, c-A26/5FAM-Rev in the cell extract at ambient
temperature responded to ATP with a KD of 107 μM and
revealed a relative fluorescent intensity change of more than
1.3-fold (Figure 6a); these values were comparable to those
obtained under buffered conditions (Figure 2c). The slight
increase in the KD value of c-A26/5FAM-Rev to ATP in the cell
extract may reflect that a small portion of RNP was degraded or
that the sensing properties of the complex were affected by the
factors in the cell media. A time course of changes in the
fluorescence intensities of covalent and noncovalent sensors in
the cell extracts was evaluated to assess the stability of the
complexes (Figure 6b). Even after incubation for 120 min in
the cell extract, c-A26/5FAM-Rev retained a sufficient response
to ATP. These results indicated that the covalent linkage
between the two subunits of the fluorescent RNP sensor
enhanced the ability of the sensor to detect its target in cellular
extracts at ambient temperature. The modification of the 3′-end
of RNA would reduce the reactivity of 3′-exo-ribonucleases,15
and the covalent linkage between the C-terminal of Rev and the
3′-end of RNA enhances the stability of Rev-RRE RNA, which
in turn prevents the endonuclease reaction to some extent. The
covalent modification thus converted the RNP sensor to a
usable sensor for fluorescent sensing under physiological
conditions.
MATERIALS AND METHODS
■
Pyrobest DNA polymerase for PCR reactions was obtained from
TaKaRa Bio Inc. (Shiga, Japan), and the AmpliScribe T7 Kit for RNA
preparation was from Epicentre (Madison, WI). N-α-Fmoc-protected
amino acids, 2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyluronium
hexafluorophosphate (HBTU), 1-hydroxybenzotriazole (HOBt),
N,N-diisopropylethylamine (DIEA), distilled N,N-dimethylformamide
(DMF), diisopropylcarbodiimide (DIC), and N,N-dimethyl-4-amino-
pyridine (DMAP) were obtained from Watanabe Chemical Industries
(Hiroshima, Japan). Dichloromethane (DCM) and HPLC-grade
acetonitrile were purchased from Nacalai Tesque (Kyoto, Japan). 5-
Carboxyfluorescein N-succinimidyl ester and 1-pyrenesulfonyl chloride
were from Life technologies, Molecular Probes (Grand Island, NY).
Cy5 mono NHS ester was from GE Healthcare Japan Inc. (Tokyo,
Japan). 4-Hydroxymethylbenzoic acid PEGA resin (HMBA-PEGA
resin) was from Merck, Novabiochem (Darmstadt, Germany). A
reversed-phase C18 column (20 × 250 mm, Ultron VX Peptide;
Shinwa Chemical Industries, Kyoto, Japan) and a RESOURCE-RPC
column (3 mL, GE Healthcare) were used for purification of peptides
for preparative purposes. Dulbecco’s modified Eagle’s medium
(DMEM) was obtained from Sigma-Aldrich (St. Louis, MO). Trypsin
was purchased from Life technologiesTM, Gibco (Grand Island, NY).
Streptomycin was purchased from Meiji Seika Pharma, Co., Ltd.
(Tokyo, Japan). Sodium periodate, hydrazine, gel electrophoresis
grade acrylamide, bisacrylamide, phenol, thioanisol, 1,2-ethanditiol,
and fetal bovine serum (FBS) were purchased from Wako Chemicals
(Tokyo, Japan).
CONCLUSION
■
Fluorescent sensors with optical characteristics tailored for
given measurement conditions, such as the detection (or
excitation) wavelength and ligand concentration range, are
easily obtained from fluorescent ribonucleopeptide (RNP)
libraries by taking advantage of the noncovalent properties of
RNP complexes.3a,d,e The noncovalent configuration of the
RNP needs an improvement for practical measurements, for
instance, when one would like to reduce the sensor
concentration or to apply multiple sensors in the same solution.
For this purpose, formation of a covalent linkage between the
ligand-binding RNA and the fluorophore-modified peptide
provides a stable RNP sensor. Covalently linked RNP sensors
showed enhanced thermostability and enabled the detection of
ligands at lower concentrations of sensors, even in cell extracts.
Construction of Noncovalent Fluorescent RNP Sensors. An
RNA library was constructed with a randomized region containing up
to 40 nucleotides adjacent to the RRE region and was added to the
Rev peptide to afford an RNP library (RRENn RNP library), as
described previously.3e ATP-binding RNP receptors were isolated
from the RREN30 RNP library or the RRENn RNP library by in vitro
selection as described.3c,f The nucleotide sequences used in this paper,
except the RRE region, were as follows: A26, 5′-UUCCGGUAGUG-
GUUGUGUGUGUGCGGUUUU-3′; G23, 5′-UGGAGCGGCUU-
GUUGCGGAGUGUCUACUUG-3′; An16, 5′-CGUAGUGGUGU-
GUGUGUGG-3′.3c For the conversion of RNP receptors to
F
dx.doi.org/10.1021/ja3097652 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
noncovalent fluorescent RNP sensors, the RNA subunit of the RNP
receptor was combined with the fluorophore-modified Rev peptide to
form a noncovalent complex.3c
MALDI-TOF MS Analysis of the Covalent RNP Complex. The
covalently linked RNP complex was analyzed by a MALDI-TOF mass
spectrometer (AXIMA-LNR, Shimadzu). A sample solution was
desalted by a Sep-Pak C18 column (Waters), and a matrix solution
was prepared with 0.07 M 3-hydroxypicolinic acid and 0.07 M
diammonium hydrogen citrate in acetonitrile/H2O (v/v: 1/1) for
negative mode and with 2′,4′,6′-trihydroxyacetophenone monohydrate
and diammonium hydrogen citrate in acetonitrile/H2O/EtOH (v/v:
5/50/45) for positive mode. The sample solution was mixed with the
matrix solution at a 1:1 (v:v) ratio on the MALDI sample plate and
allowed to air-dry. Calibration of the instrument was carried out using
oligodeoxynucleotides (molecular weights: 9145.8, 14 547.6, and 18
204.0) as external standard samples. Each mass spectrum was
representative of an average of about 50 laser shots, and the laser
power density was held constant at near threshold power density
throughout these investigations. c-A26/5FAM-Rev: M+ calcd 21557.3,
found 21410.7; c-G23/Pyr-Rev: M+ calcd 21507.4, found 21333.6
(Figure S9).
Fluorescence Measurements on the Microplate. Fluorescence
measurements in 96-well plates were performed on a Wallac ARVOsx
1420 multilabel counter or Tecan Infinite M200 plate reader. A
binding solution (100 μL) containing noncovalent or covalent
fluorescent RNPs in 10 mM Tris-HCl (pH 7.6), 100 mM NaCl, 10
mM MgCl2, and 0.005% Tween 20 with an indicated amount of
substrate was gently swirled for a few minutes and allowed to sit for 30
min at the indicated temperature. Emission spectra were measured
with an appropriate filter set for each fluorophore. Respective
excitation and emission wavelengths were Pyr-Rev (350, 390 nm),
5FAM-Rev (485, 535 nm), and Cy5-Rev (650, 670 nm).
Construction of a Stable Fluorescent RNP Sensor by the
Covalent Linking Method. RNA subunits of RNP receptors were
purified by denaturing 8% polyacrylamide gel electrophoresis. The
purified RNA subunit was treated with freshly prepared sodium
periodate to convert the cis-diol of the 3′-terminal ribose to 3′-
dialdehyde by periodate oxidation.12 Freshly prepared 50 equiv of 0.01
M sodium periodate (10 μL; 100 nmol) was added to 100 μM RNA
(20 μL; 2 nmol) in 32 μL of 0.01 M sodium phosphate (pH 7.5), and
the reaction mixture was incubated for 1 h at 37 °C in the dark. After
the reaction, an excess of sodium periodate was reduced by adding
glycerol (final concentration 1 M), and the resulting oxidized RNA
was purified by ethanol precipitation.
The peptide subunit for the formation of a covalent linkage was
synthesized by introducing a fluorophore to the N-terminus and the
peptide linker (GGSGGSGGSG) with a hydrazide group at the C-
terminus of the Rev peptide TRQARRNRRRRWRERQR
GGSGGSGGSG. The modified peptide was synthesized as follows.
A 4-hydroxymethylbenzoic acid PEGA resin (HMBA-PEGA resin;
Novabiochem) was placed in a dry flask, and a sufficient amount of
DMF was added to soak the resin; this mixture was allowed to swell
for 30 min. N-α-Fmoc-glycine (10 equiv relative to resin loading) was
dissolved in dry DCM with 1 or 2 drops of DMF and then mixed with
a solution of DIC (5 equiv relative to resin loading) in dry DMF on
ice. After incubating for 20 min, DCM was removed by evaporation
under the reduced pressure. The residue was dissolved by minimum
volume of DMF and added to the resin prepared above. A DMF
solution of DMAP (0.1 equiv relative to resin loading) was added to
the resin/amino acid mixture and incubated at room temperature for 1
h with occasional swirling. This first attachment procedure was
repeated twice. The resin with the first amino acid was loaded onto an
automated peptide synthesizer (PSSM-8; Shimadzu, Kyoto, Japan),
and the subsequent synthesis was performed according to Fmoc
chemistry protocols using protected Fmoc-amino acids and HBTU. A
fluorophore with an activated group (5-carboxyfluorescein N-
succinimidyl ester, 1-pyrenesulfonyl chloride, or Cy5 mono NHS
ester) was directly coupled to the Fmoc-deprotected synthetic peptide
on the resin. The fluorophore-labeled peptide was cleaved from the
resin with 0.1 M hydrazine hydrate in DMF for 30 min at room
temperature, washed with DMF, precipitated by ether, and then
deprotected by a solution containing phenol (0.75 g), distilled water
(0.5 mL), thioanisol (0.5 mL), 1,2-ethanditiol (0.25 mL), and TFA
(10 mL) for 3 h. The N- and C-terminal modified peptide was gel
filtrated with a G-10 resin, purified by reversed-phase HPLC, and
characterized by MALDI-TOF mass spectrometry (AXIMA-LNR,
Shimadzu) as follows: 5-carboxyfluorescein-modified Rev peptide
hydrazide (5FAM-Rev-HZ), m/z 3471.8 (calcd for [M+H]+ 3471.7);
pyrene-modified Rev peptide hydrazide (Pyr-Rev-HZ), m/z 3376.1
(calcd for [M+H]+ 3376.6); Cy5-modified Rev peptide hydrazide
(Cy5-Rev-HZ), m/z 3747.3 (calcd for [M+H]+ 3747.9).
Determination of Ligand-Binding Affinity. The ligand-binding
affinity of fluorescent RNP was calculated by fitting the ligand titration
data using the following equation:
F
= Fmin + A{([FRNP]T + [substrate]T + KD)
obs
− (([FRNP]T + [substrate]T + KD)2
− 4[FRNP]T[substrate]T)1/2}/2[FRNP]T
where Fobs is the observed fluorescence intensity with each
concentration of substrate, A is the increase in fluorescence at
saturating substrate concentrations (Fmax − Fmin), KD is the equilibrium
dissociation constant, and [FRNP]T and [substrate]T are the total
concentrations of fluorescent RNP and the substrate, respectively.
Simultaneous Detection of ATP and GTP in the Same
Solution. Samples were prepared by mixing c-A26/5FAM-Rev (100
nM) and c-G23/Pyr-Rev (100 nM), or A26/5FAM-Rev (1000 nM)
and G23/Pyr-Rev (1000 nM), and the indicated substrates in 10 mM
Tris-HCl (pH 7.6), 100 mM NaCl, 10 mM MgCl2, and 0.005% Tween
20. The samples were incubated for 30 min at 4 °C. When two
noncovalent sensors were applied in the same solution, each
noncovalent sensor was incubated in an independent tube for 30
min at 4 °C to secure the complex formation, then the solution
containing the noncovalent sensor was added to the assay solution and
incubated for 30 min at 4 °C. Fluorescence intensities were recorded
on a Tecan Infinite M200 plate reader. The excitation wavelengths
were 485 nm for 5FAM-Rev, 350 nm for Pyr-Rev, and 650 nm for
Cy5-Rev. The emission wavelengths were 535 nm for 5FAM-Rev, 390
nm for Pyr-Rev, and 670 nm for Cy5-Rev and were used for the
determination of relative fluorescence intensity.
Simultaneous Detection of a Substrate and a Product in an
Enzymatic Reaction by Covalently Linked Fluorescent RNP
Sensors. A sample was prepared by mixing c-A26/5FAM-Rev, c-G23/
Pyr-Rev, and 100 μM Ado in 10 mM Tris-HCl (pH 7.6) containing
100 mM NaCl, 10 mM MgCl2, and 0.005% Tween 20. The sample
was incubated for 30 min at 15 °C, and then 0.1 mU of ADA was
added to start the reaction. For the detection of changes in the
concentrations of Ado and Ino, the fluorescence intensities were
measured with Hitachi a F-4500 fluorescence spectrofluorophotometer
with an excitation and emission bandwidth of 5 nm (ex.: 485 nm, em.:
535 nm for 5FAM-Rev; ex.: 350 nm, em.: 380 nm for Pyr-Rev) with
A coupling reaction between the 3′-modified RNA (50 μM) and
fluorophore-modified Rev-HZ (75 μM) was performed in 0.02 M
sodium acetate (pH 5.2) containing 0.01 M NaCl (total 40 μL) at 37
°C in the dark. After 5 h, the reaction mixture was extracted by phenol-
chloroform and then purified by ethanol precipitation. The precipitate
was dissolved in 0.01 M sodium phosphate (pH 7.5) containing 0.1 M
NaCl (40 μL). The sample solution was treated with a cation-exchange
resin (SP Sepharose Fast Flow, GE Healthcare) for 30 min at ambient
temperature to remove the excess peptide, and the unbound fraction
was collected to obtain a covalently linked RNP complex. The sample
solution was purified by 15% denaturing PAGE (6 M urea) and was
subsequently quantified by measuring the absorption at 260 nm (A26
ε = 541 600 M−1 cm−1, G23 ε = 538 000 M−1 cm−1, An16 = 439 500
M
−1 cm−1). As a result, about 50 μL of 15 μM covalently linked RNP
complex (c-A26/5FAM-Rev, c-G23/Pyr-Rev) was recovered (yield:
30−40%). c-An16/Cy5-Rev and c-G23/5FAM-Rev were prepared
using the same procedure (yield: 75%).
G
dx.doi.org/10.1021/ja3097652 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
reaction time. Amounts of the reacted substrate (Ado) and product
(Ino) were converted using the standard curves calculated from the
results of fluorescence spectroscopy of each substrate with the known
concentration (Figure S8). The actual amounts of Ado and Ino with
the reaction of ADA were measured by HPLC analysis for comparing
to the results traced by the cRNP sensors (Figures 6b, S6, and S7).
Preparation of HeLa Cell Extracts. HeLa cells were seeded on
15-cm dishes and cultured in DMEM supplemented with 10% FBS,
penicillin (30 units/mL), and streptomycin (30 μg/mL) at 37 °C in a
humidified atmosphere consisting of 5% CO2 and 95% air. After a 3-
day culture, cells were harvested (total 2.5 × 108 cells). The pelleted
cells were suspended in a binding buffer (10 mM sodium phosphate,
pH 7.6, containing 100 mM NaCl, 10 mM MgCl2, and 0.005% Tween
20) and sonicated for 10 min to obtain the cell lysate.17 The efficiency
of the cell homogenization was checked microscopically for cell lyses.
The whole cell lysate was centrifuged at 12000g for 15 min. The
supernatant was then passed through a 0.45 μm filter, and the protein
concentration was measured by an RC DC protein assay kit (Bio-Rad,
Hercules, CA). For a typical sample solution, the total protein of the
supernatant was diluted to a final concentration of 1.0 mg/mL.
Titration Analysis of the RNP Sensors in HeLa Cell Extracts.
Noncovalent A26/5FAM-Rev and covalent A26/5FAM-Rev (0.2 μM)
were titrated with given concentrations of ATP in HeLa cell extracts.
After 30, 60, or 120 min incubations, fluorescence measurements were
performed in a Wallac ARVOsx 1420 multilabel counter (Perkin-
Elmer) at 25 °C. Emission intensities were measured with a filter set
for the fluorescein chromophore (ex.: 485 nm, em.: 535 nm).
Sakaguchi, R.; Hayashi, H.; Nakano, S.; Liew, F. F.; Morii, T. Sensors
2010, 10, 1355−1376.
(2) (a) Morii, T.; Sugimoto, K.; Makino, K.; Otsuka, M.; Imoto, K.;
Mori, Y. J. Am. Chem. Soc. 2002, 124, 1139−1140. (b) de Lorimier, R.
M.; Smith, J. J.; Dwyer, M. A.; Looger, L. L.; Sali, K. M.; Paavola, C.
D.; Rizk, S. S.; Sadigov, S.; Conrad, D. W.; Loew, L.; Helinga, H. W.
Protein Sci. 2002, 11, 2655−2675. (c) Toutchkine, A.; Kraynov, V.;
Hahn, K. J. Am. Chem. Soc. 2003, 125, 4132−4145. (d) Nalbant, P.;
Hodgson, L.; Kraynov, V.; Toutchkine, A.; Hahn, K. M. Science 2004,
305, 1615−1619. (e) Chan, P. H.; Liu, H. B.; Chen, Y. W.; Chan, K.
C.; Tsang, C. W.; Leung, Y. C.; Wong, K. Y. J. Am. Chem. Soc. 2004,
126, 4074−4075. (f) Chan, P. H.; So, P. K.; Ma, D. L.; Zhao, Y.; Lai,
T. S.; Chung, W. H.; Chan, K. C.; Yiu, K. F.; Chan, H. W.; Siu, F. M.;
Tsang, C. W.; Leung, Y. C.; Wong, K. Y. J. Am. Chem. Soc. 2008, 130,
6351−6361. (g) Sakaguchi, R.; Endoh, T.; Yamamoto, S.; Tainaka, K.;
Sugimoto, K.; Fujieda, N.; Kiyonaka, S.; Mori, Y.; Morii, T. Bioorg.
Med. Chem. 2009, 17, 7381−7386. (h) Sakaguchi, R.; Tainaka, K.;
Shimada, N.; Nakano, S.; Inoue, M.; Kiyonaka, S.; Mori, Y.; Morii, T.
Angew. Chem., Int. Ed. 2010, 49, 2150−2153.
(3) (a) Morii, T.; Hagihara, M.; Sato, S.; Makino, K. J. Am. Chem. Soc.
2002, 124, 4617−4622. (b) Sato, S.; Fukuda, M.; Hagihara, M.;
Tanabe, Y.; Ohkubo, K.; Morii, T. J. Am. Chem. Soc. 2005, 127, 30−31.
(c) Hagihara, M.; Fukuda, M.; Hasegawa, T.; Morii, T. J. Am. Chem.
Soc. 2006, 128, 12932−12940. (d) Hasegawa, T.; Hagihara, M.;
Fukuda, M.; Morii, T. Nucleosides Nucleotides Nucleic Acids 2007, 26,
1277−1281. (e) Hasegawa, T.; Hagihara, M.; Fukuda, M.; Nakano, S.;
Fujieda, N.; Morii, T. J. Am. Chem. Soc. 2008, 130, 8804−8812.
(f) Nakano, S.; Mashima, T.; Matsugami, A.; Inoue, M.; Katahira, M.;
Morii, T. J. Am. Chem. Soc. 2011, 133, 4567−4579. (g) Nakano, S.;
Nakata, E.; Morii, T. Bioorg. Med. Chem. Lett. 2011, 21, 4503−4506.
(h) Liew, F. F.; Hasegawa, T.; Fukuda, M.; Nakata, E.; Morii, T.
Bioorg. Med. Chem. 2011, 19, 4473−4481. (i) Liew, F. F.; Hayashi, H.;
Nakano, S.; Nakata, E.; Morii, T. Bioorg. Med. Chem. 2011, 19, 5771−
5775.
(4) Battiste, J. L.; Mao, H.; Rao, N. S.; Tan, R.; Muhandiram, D. R.;
Kay, L. E.; Frankel, A. D.; Williamson, J. R. Science 1996, 273, 1547−
1551.
(5) (a) Ellington, A. D.; Szostak, J. W. Nature 1990, 346, 818−822.
(b) Tuerk, C.; Gold, L. Science 1990, 249, 505−510.
(6) (a) Stojanovic, M. N.; de Prada, P.; Landry, D. W. J. Am. Chem.
Soc. 2000, 122, 11547−11548. (b) Nutiu, R.; Li, Y. J. Am. Chem. Soc.
2003, 125, 4771−4778. (c) Nutiu, R.; Li, Y. Angew. Chem., Int. Ed.
2005, 44, 1061−1065. (d) Li, N.; Ho, C. M. J. Am. Chem. Soc. 2008,
130, 2380−2381.
ASSOCIATED CONTENT
■
S
* Supporting Information
Structural models of covalent linking of the Rev-RRE complex
(Figure S1), purity confirmation for the covalently linked RNP
sensors (Figure S2), reaction of c-A26/5FAM-Rev under
denature conditions (Figure S3), fluorescence titration analyses
of c-An16/Cy5-Rev and c-G23/5FAM-Rev by ATP and GTP
(Figure S4), saturation curves for the relative fluorescence
intensity changes of c-A26/5FAM-Rev and A26/5FAM-Rev by
titration with Ado (Figure S5), standard curves for the HPLC
analysis of adenosine and inosine (Figure S6), time course
profiles of the ADA reaction analyzed by HPLC (Figure S7),
quantitative titration curves for fluorescent c-RNP sensors by
adenosine and inosine (Figure S8), MALDI-TOF MS spec-
troscopy of c-RNP sensors (Figure S9), and chromatograms of
reversed-phase HPLC analyses of c-RNP sensors (Figure S10).
This material is available free of charge via the Internet at
http://pubs.acs.org.
(7) (a) Stojanovic, M. N.; Landry, D. W. J. Am. Chem. Soc. 2002, 124,
9678−9679. (b) Stojanovic, M. N.; Kolpashchikov, D. M. J. Am. Chem.
Soc. 2004, 126, 9266−9270. (c) Jiang, Y.; Fang, X.; Bai, C. Anal. Chem.
2004, 76, 5230−5235. (d) Zhou, C.; Jiang, Y.; Hou, S.; Ma, B.; Fang,
X.; Li, M. Anal. Bioanal. Chem. 2006, 384, 1175−1180. (e) Li, B. L.;
Wei, H.; Dong, S. J. Chem. Commun. 2007, 73−75.
AUTHOR INFORMATION
■
(8) (a) Maynard, J.; Georgiou, G. Annu. Rev. Biomed. Eng. 2000, 2,
Corresponding Author
t-morii@iae.kyoto-u.ac.jp
339−376. (b) Worn, A.; Pluckthun, A. J. Mol. Biol. 2001, 305, 989−
̈
1010.
(9) Skerra, A.; Pluckthun, A. Science 1988, 240, 1038−1041.
̈
Notes
(10) (a) Glockshuber, R.; Malia, M.; Pfitzinger, I.; Pluckthun, A.
̈
The authors declare no competing financial interest.
Biochemistry 1990, 29, 1362−1367. (b) Brinkmann, U.; Gallo, M.;
Brinkmann, E.; Kunwar, S.; Pastan, I. Proc. Natl. Acad. Sci. U.S.A. 1993,
90, 547−551.
ACKNOWLEDGMENTS
■
(11) (a) Bird, R. E.; Hardman, K. D.; Jacobson, J. W.; Johnson, S.;
Kaufman, B. M.; Lee, S. M.; Lee, T.; Pope, S. H.; Riordan, G. S.;
Whitlow, M. Science 1988, 242, 423−426. (b) Huston, J. S.; Levinson,
This work was supported in part by a Grant-in-Aid for Scientific
Research from the Ministry of Education, Culture, Sports,
Science, and Technology, Japan to T.M. (No. 24121717).
D.; Mudgett-Hunter, M.; Tai, M. S.; Novotny, J.; Margolies, M. N.;
́
Ridge, R. J.; Bruccoleri, R. E.; Haber, E.; Crea, R.; Oppermann, H.
Proc. Natl. Acad. Sci. U.S.A. 1988, 85, 5879−5883.
REFERENCES
■
(12) (a) Robberson, D. L.; Davidson, N. Biochemistry 1972, 11, 533−
537. (b) Hansske, F.; Cramer, F. Methods Enzymol. 1979, 59, 172−
181. (c) Proudnikov, D.; Mirzabekov, A. Nucleic Acids Res. 1996, 24,
4535−4542. (d) Cabezas, E.; Satterthwait, A. C. J. Am. Chem. Soc.
(1) (a) Weiss, S. Science 1999, 283, 1676−1683. (b) Zhang, J.;
Campbell, R. E.; Ting, A. Y.; Tsien, R. Y. Nat. Rev. Mol. Cell Biol. 2002,
3, 906−918. (c) Giepmans, B. N.; Adams, S. R.; Ellisman, M. H.;
Tsien, R. Y. Science 2006, 312, 217−224. (d) Wang, H.; Nakata, E.;
Hamachi, I. Chembiochem 2009, 10, 2560−2577. (e) Tainaka, K.;
1999, 121, 3862−3875. (e) Kirchhoff, J. H.; Brase, S.; Enders, D. J.
̈
H
dx.doi.org/10.1021/ja3097652 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Comb. Chem. 2001, 3, 71−77. (f) Kozlov, I. A.; Melnyk, P. C.;
Stromsborg, K. E.; Chee, M. S.; Barker, D. L.; Zhao, C. Biopolymers
2004, 73, 621−630. (g) Pfander, S.; Fiammengo, R.; Kirin, S. I.;
Metzler-Nolte, N.; Jaschke, A. Nucleic Acids Res. 2007, 35, e25.
̈
(h) Sharon, J. L.; Puelo, D. A. Biomaterials 2008, 29, 3137−3142.
(13) (a) Heaphy, S.; Dingwall, C.; Ernberg, I.; Gait, M. J.; Green, S.
M.; Karn, J.; Lowe, A. D.; Singh, M.; Skinner, M. A. Cell 1990, 60,
685−693. (b) Jain, C.; Belasco, J. G. Cell 1996, 87, 115−125.
(c) Wang, Y.; Hamasaki, K.; Rando, R. R. Biochemistry 1997, 36, 768−
779. (d) Kirk, S. R.; Luedtke, N. W.; Tor, Y. J. Am. Chem. Soc. 2000,
122, 980−981. (e) Zhang, C.-Y.; Johnson, L. W. J. Am. Chem. Soc.
2006, 128, 5324−5325.
(14) (a) Wolfenden, R.; Sharpless, T. K.; Allan, R. J. Biol. Chem.
1967, 242, 977−983. (b) Kati, W. M.; Acheson, S. A.; Wolfenden, R.
Biochemistry 1992, 31, 7356−7366. (c) Ajloo, D.; Taghizadeh, E.;
Saboury, A. A.; Bazyari, E.; Mahnam, K. Int. J. Biol. Macromol. 2008,
43, 151−158.
(15) (a) Osako, M. K.; Tomita, N.; Nakagami, H.; Kunugiza, Y.;
Yoshino, M.; Yuyama, K.; Tomita, T.; Yoshikawa, H.; Ogihara, T.;
Morishita, R. J. Gene Med. 2007, 9, 812−819. (b) Zhou, C.;
Chattopadhyaya, J. J. Org. Chem. 2010, 75, 2341−2349. (c) Steen,
K. A.; Malhotra, A.; Weeks, K. M. J. Am. Chem. Soc. 2010, 132, 9940−
9943.
(16) (a) Namiki, S.; Sakamoto, H.; Iinuma, S.; Iino, M.; Hirose, K.
Eur. J. Neurosci. 2007, 25, 2249−2259. (b) Tokunaga, T.; Namiki, S.;
Yamada, K.; Imaishi, T.; Nonaka, H.; Hirose, K.; Sando, S. J. Am.
Chem. Soc. 2012, 134, 9561−9564. (c) Kang, W. J.; Cho, Y. L.; Chae, J.
R.; Lee, J. D.; Ali, B. A.; Al-Khedhairy, A. A.; Lee, C. H.; Kim, S.
Biomaterials 2012, 33, 6430−6437. (d) Lee, M.-S.; Park, W.-S.; Kim, Y.
H.; Ahn, W. G.; Kwon, S.-H.; Her, S. Sensors 2012, 12, 15628−15637.
(17) Endoh, T.; Funabashi, H.; Mie, M.; Kobatake, E. Anal. Chem.
2005, 77, 4308−4314.
I
dx.doi.org/10.1021/ja3097652 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX