Very rapid DNA-templated reaction for efficient signal amplification and its steady-state kinetic analysis of the turnover cycle

Article abstract of DOI:10.1021/ja404743m

Oligonucleotide-templated reactions are powerful tools for the detection of nucleic acid sequences. One of the major scientific challenges associated with this technique is the rational design of non-enzyme-mediated catalytic templated reactions capable of multiple turnovers that provide high levels of signal amplification. Herein, we report the development of a nucleophilic aromatic substitution reaction-triggered fluorescent probe. The probe underwent a rapid templated reaction without any of the undesired background reactions. The fluorogenic reaction conducted in the presence of a template provided a 223-fold increase in fluorescence after 30 s compared with the nontemplated reaction. The probe provided an efficient level of signal amplification that ultimately enabled particularly sensitive levels of detection. Assuming a simple model for the templated reactions, it was possible to estimate the rate constants of the chemical reaction in the presence and in the absence of the template. From these kinetic analyses, it was possible to confirm that an efficient turnover cycle had been achieved, on the basis of the dramatic enhancement in the rate of the chemical reaction considered to be the rate-determining step. With maximized turnover efficiency, it was demonstrated that the probe could offer a high turnover number of 1500 times to enable sensitive levels of detection with a detection limit of 0.5 pM in the catalytic templated reactions.

Full text of DOI:10.1021/ja404743m

Article
pubs.acs.org/JACS
Very Rapid DNA-Templated Reaction for Efficient Signal
Amplification and Its Steady-State Kinetic Analysis of the Turnover
Cycle
Aya Shibata,† Takanori Uzawa,†,⊥ Yuko Nakashima,† Mika Ito,†,‡ Yukiko Nakano,† Satoshi Shuto,‡
Yoshihiro Ito,*,†,⊥ and Hiroshi Abe*,†,§,‡,⊥
†Nano Medical Engineering Laboratory, RIKEN Advanced Science Institute, 2-1, Hirosawa, Wako-Shi, Saitama 351-0198, Japan
‡Faculty of Pharmaceutical Sciences, Hokkaido University, Kita-12, Nishi-6, Kita-ku, Sapporo 060-0812, Japan
§PRESTO, Japan Science and Technology Agency, 4-1-8, Honcho, Kawaguchi, Saitama 332-0012, Japan
⊥Emergent Bioengineering Materials Research Team, RIKEN Center for Emergent Matter Science, 2-1, Hirosawa, Wako-Shi, Saitama
351-0198, Japan
S
* Supporting Information
ABSTRACT: Oligonucleotide-templated reactions are power-
ful tools for the detection of nucleic acid sequences. One of the
major scientific challenges associated with this technique is the
rational design of non-enzyme-mediated catalytic templated
reactions capable of multiple turnovers that provide high levels
of signal amplification. Herein, we report the development of a
nucleophilic aromatic substitution reaction-triggered fluores-
cent probe. The probe underwent a rapid templated reaction
without any of the undesired background reactions. The
fluorogenic reaction conducted in the presence of a template
provided a 223-fold increase in fluorescence after 30 s compared with the nontemplated reaction. The probe provided an efficient
level of signal amplification that ultimately enabled particularly sensitive levels of detection. Assuming a simple model for the
templated reactions, it was possible to estimate the rate constants of the chemical reaction in the presence and in the absence of
the template. From these kinetic analyses, it was possible to confirm that an efficient turnover cycle had been achieved, on the
basis of the dramatic enhancement in the rate of the chemical reaction considered to be the rate-determining step. With
maximized turnover efficiency, it was demonstrated that the probe could offer a high turnover number of 1500 times to enable
sensitive levels of detection with a detection limit of 0.5 pM in the catalytic templated reactions.
application of continuous external stimuli to provide a
photosensitization step for the signal amplification process.
We believe that the rational design of such molecular
mechanisms represents an important scientific challenge.
Of the variety of different OTRC chemistries currently
available, ligation reactions in particular suffer from several
limitations, including the increased affinity of the product for
the target oligonucleotide. Under thermodynamic control
conditions, inhibition from the product prevents the realization
of a high catalytic turnover reaction on the target oligonucleo-
tide. Unligated chemistries have therefore been reported to
suppress such product inhibition. In this case, the affinity of the
product for the target oligonucleotide remains unchanged after
the reaction, and the unreacted probe can therefore be readily
replaced by the product on an oligonucleotide target under
thermodynamic control. Grossman et al.5a reported the
achievement of a particularly good result from this approach
INTRODUCTION
■
Oligonucleotide-templated reactions have received considerable
levels of attention as a potential approach for nucleic acid
sensing, with applications including the detection of RNA in
living cells.1 This strategy exploits the target strand as a
template for two functionalized probes. For example, native
chemical ligation,2 hydrolysis,3 the Staudinger reaction,1c,d,4
transfer of a reporter group,5 DNAzyme,6 SN2 quencher
displacement ligation reactions (QUAL probe),1a,b,7 metal-
activated reactions,8 photocaged reactions,9 photocatalyst
reactions,10 and metal catalysts11 have all been applied to the
development of fluorescent nucleic acid reactions. These
oligonucleotide-templated reactions can provide signal ampli-
fication through multiple turnover reactions,12 and it is
generally envisaged that this will enable particularly sensitive
levels of detection. Oligonucleotide-templated reactions using a
target oligonucleotide as a catalyst (OTRC) provide a simple
molecular mechanism for multiple turnover reactions. Fur-
thermore, this strategy only requires two probe pairs and does
not require extra reagents to catalyze the reaction or the
Received: May 16, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja404743m | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 1. Fluorescence emission of DNs or CNs protected coumarin by nucleophilic substitution reaction. (A) Turnover cycle, (B) Energy diagram
of the templated reaction.
for an OTRC where the turnover number reached 402 within
24 h when a 104-fold excess of the probes was used for the
quencher transfer reaction.
suppressing the undesired background signal, and consequently
enables sensitive levels of detection with maximized signal
amplification. We also demonstrated that the probe could offer
1500 turnovers. Therefore, to the best of our knowledge, the
current system represents the best result ever reported to date
for an OTRC.
The catalytic turnover cycle for an OTRC is composed of
three steps (Figure 1), including (1) the association of the
probes to the template; (2) the templated chemical reaction
accompanied by a fluorescence signal; and (3) the dissociation
of the fluorescent product from the template. To design an
efficient signal amplification system, it would be necessary to
accelerate the reaction cycle under kinetic control by enhancing
the rate of the rate-determining step of the OTRC. To achieve
this goal, it would also be necessary to determine which of the
three steps of the OTRC that the rate-determining step is. For
example, the association or dissociation constants of the short
DNA fragment are reported to be 2.0 × 107 M−1 s−1 and 0.2
s−1, respectively.13 Staudinger, hydrolysis, and transfer reactions
typically exhibit rate constants in the range of 8.32 × 10−8 (ref
5a) to 7.5 × 10−3 s−1 (ref 4a). Thus, the chemical reaction
providing the rate-determining step in the OTRC cycle is
represented by the highest activation barrier in the reaction
diagram (Figure 1B). Any acceleration in the rate of the rate-
determining chemical reaction should therefore provide an
increase in the overall efficiency of the turnover cycle over a
certain time period.
An ideal probe for an OTRC would effectively facilitate a
high speed templated reaction, a low level of background
intermolecular reactions without template assistance together
with highly efficient fluorogenic properties. We recently
reported the development of a fluorogenic system for a
nucleophilic aromatic substitution (SNAr) reaction-triggered
fluorescence probe bearing a 2,4-dinitrobenzenesulfonyl
(DNs)-protected aminocoumarin (DNs-AMCA) as the fluoro-
genic molecule and phosphorothioate (PS) as nucleophilic
molecule for an oligonucleotide-templated reaction.14 Although
the fluorogenic system offered a high signal/background (S/B)
ratio, the rate of the reaction was slow and did not sufficiently
accelerate the turnover cycle. Herein, we report the develop-
ment of a new ultrarapid fluorogenic-reaction for an OTRC
based on a SNAr reaction between 2-cyano-4-nitrobenzene-
sulfonyl (CNs)-protected aminocoumarin (CNs-AMCA) as a
prefluorophore molecule and 4-mercaptobenzoic acid (MBA)
as a nucleophilic molecule. The fluorogenic SNAr reaction of
this new probe occurs rapidly and is 668 times faster than our
previously reported DNs-AMCA/PS pair, which was com-
parable in speed to dissociation step. Furthermore, this new
reaction yields a significant fluorescent signal as well as
RESULTS
■
Design and Synthesis of the Probes. We have previously
reported the use of DNs-AMCA and PS as fluorogenic and
nucleophilic molecules, respectively, for an oligonucleotide-
templated reaction.14 The rate of the chemical reaction in that
particular case, however, was too slow for effective application.
To accelerate the rate of this reaction, the use of an MBA group
as the nucleophilic molecule was evaluated, because thiophe-
nols are generally expected to be more nucleophilic than PS as
a consequence of reduced steric hindrance around the
nucleophilic center. We designed a novel CNs-AMCA as the
prefluorophore molecule with the expectation that this would
exhibit a slightly lower level of reactivity than DNs-AMCA. It
was envisaged that this approach would offer a high S/B ratio
for the appropriate level of reactivity. The syntheses of the
CNs-AMCA and MBA-NHS are shown in Scheme S1
(Supporting Information). The treatment of the AMCA methyl
ester 1 with 2-cyano-4-nitrobenzenesulfonyl chloride15 in a
mixture of pyridine/CH2Cl2 gave the corresponding CNs-
protected material 2 in a 51% yield. Subsequent hydrolysis of 2
with lithium hydroxide afforded CNs-AMCA 3 in a 95% yield.
In a separate scheme, the treatment of MBA with N-
hydroxysuccinimide in DMF for 16 h afforded MBA-NHS 4
in a 19% yield. The fluorogenic DNA probe was synthesized via
modification at the 5′ terminal with the DNs- or CNs-AMCA,
where the N-hydroxysuccinimide ester of the DNs- or CNs-
AMCA was reacted with the amino group at the 5′ terminal of
the DNA probe (Scheme S2, Supporting Information). The PS
and the MBA probes were synthesized via the modification of
the 3′ terminal with a phosphorothioate group or the MBA
group. The template originating from the 23S rRNA gene
sequence in the Escherichia coli (E. coli) K12 strain was
designed in this way.16
Reaction of the Fluorogenic Probes with Nucleophilic
Molecules. To compare the reactivities of different nucleo-
philic molecules, thymidine 3′-O-(12-(phosphoryloxy) dodecyl
O,S-dihydrogenphosphorothioate) (PS dT) and MBA were
tested for their SNAr reactions using the DNs- and CNs-AMCA
probes as the substrates. The pKa values of these nucleophilic
B
dx.doi.org/10.1021/ja404743m | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
groups are 5.3 (PS dT)17 and 6.0 (MBA),18 respectively. These
nucleophilic groups would therefore exist as the corresponding
thioxide anions under neutral conditions at pH 7.2, and should
consequently be nucleophilic enough to accelerate the SNAr
reactions with caged AMCA derivatives. The uncaging reaction
was evaluated by observing the increase in the intensity of
fluorescence signal associated with this transformation. The
apparent first-order rate constant, kapp, was consequently
obtained from the gradient of the plot describing the treatment
of the caged AMCA probe with an excess of the nucleophilic
molecule (Figure 2, Figure S1, Supporting Information). The
Rapid DNA Templated SNAr reactions. A systematic
study of the SNAr reaction was conducted by changing the
probe pairing between the electrophilic probe (DNs-AMCA or
CNs-AMCA) and the nucleophilic probe (PS or MBA) to
determine the best partnership of probes that would offer the
most rapid templated reaction together with the lowest level of
background reactions.
The electrophilic and nucleophilic probes were incubated
both in the presence and in the absence of the template under
equimolar conditions (500 nM). The fluorescence signals at an
emission wavelength of 450 nm were monitored in a time-
dependent manner over 60 s, and the conversion yields were
calculated on the basis of a comparison with an authentic
sample (Figure 3). For the DNs-AMCA probe, the MBA probe
Figure 2. (A) The first-order rate constant of DNs- or CNs-AMCA
probe with nucleophilic molecule. (B), (C) Time course of the
fluorescence intensity (B: DNs-AMCA probe, C: CNs-AMCA probe).
All measurements were done in 20 mM Tris-HCl buffer (pH 7.2)
containing 500 nM R-AMCA probe, 500 μM nucleophilic molecule
and 100 mM MgCl2 at 22 °C. Reaction was monitored by excitation at
a
345 nm and emission at 450 nm. Analyzed by linear regression as
simple first-order kinetics by Microsoft Excel.
Figure 3. (A) Sequences of oligonucleotides used in this study. The
DNs- or CNs-AMCA probe was an 8 mer ODN modified with DNs-
or CNs-AMCA at the 5′ terminal. The nucleophilic probe was a 7 mer
ODN modified with MBA or phosphorothioate at the 3′ terminal. The
DNA targets were 52 mer ODNs (SR). The target sequence is
underlined. (B), (C) Time course of the yield in the reaction between
500 nM MBA or PS probe and 500 nM R-AMCA probe (B: DNs-
AMCA, C: CNs-AMCA) with or without 500 nM SR in 20 mM Tris-
HCl buffer (pH 7.2) containing 100 mM MgCl2 at 22 °C. Reaction
was monitored by excitation at 345 nm and emission at 450 nm. (D)
The yield of templated reaction after 30 s.
DNs- and CNs-AMCA probes reacted rapidly with MBA to
provide a fluorescence signal from the free AMCA, although the
rate of the reaction of the DNs-AMCA probe (kapp 11.6 × 10−3
s−1) was greater than that of the CNs-AMCA probe (kapp 2.6 ×
10−3 s−1).
In contrast, the PS dT reacted at a much slower rate with
both of the caged probes, with the corresponding kapp values
determined to be 4 × 10−4 or 8 × 10−6 s−1, respectively. MBA
drastically increased the rate of the reactions with the DNs-
AMCA and CNs-AMCA probes, with the rates increasing 29-
and 325-fold, respectively, in comparison with PS dT.
A density functional theory (DFT) calculation was
conducted to develop a greater understanding of the reasons
behind the observed increases in the rates of the reactions using
a series of model compounds (Figure S2, Supporting
Information). The activation energies (ΔG‡) of the transition
states associated with a series of SNAr reactions, including
CNFB (a model of CNs-AMCA), thiophenol (a model of
MBA), and phosphorothioate (model of PS), were calculated at
showed a surprisingly rapid rate of reaction, with the reaction
progressing to 60% conversion after 20 s and reaching the
saturation phase (Figure 3B). The inflection point of the DNs-
AMCA probe was determined to be 37.8 s. In contrast, the rate
of the reaction with the PS probe was much slower, with the
reaction only progressing to 1% conversion after 30 min. One
of the major limitations associated with the use of the DNs-
AMCA/MBA probe pair, however, was the occurrence of a
slight background reaction in the absence of the template
(Figure 3B, Figure S3, Supporting Information). The S/B ratio
was calculated to be 41 after 30 s (Figure 3D).
B3LYP/6-31G*. The results revealed that the ΔG‡ (CNFB
1
and thiophenol) and ΔG‡ (CNFB and phosphorothioate)
2
values were 9.2 and 11.5 kcal/mol, respectively. The rate
enhancement between the thiophenol and phosphorothioate
was calculated to be 48-fold using the Eyring equation.
Theoretical calculations also supported that thiophenol was
more nucleophilic than phosphorothioate.
We then proceeded to evaluate the reaction of the CNs-
AMCA probe with the MBA probe (Figure 3C). Once again,
the MBA probe provided a rapid rate of reaction that was
similar to the rate observed from its combination with DNs-
C
dx.doi.org/10.1021/ja404743m | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 4. Turnover experiment. (A) Time course of the yield in the reaction between 50 nM CNs-AMCA and MBA probe at shown SR
concentrations (50 nM to 0.5 pM) in Tris-HCl buffer (pH 7.2) containing 100 mM MgCl2 at 22 °C. (B) The reaction yields after 15 h and turnover
numbers (TO). The reaction yields were reported as average of values from four times.
AMCA. The inflection point of the CNs-AMCA probe was 50.0
s. In contrast to the background reactions suffered by the DNs-
AMCA/MBA probe pair, however, the CNs-AMCA/MBA
probe pair pleasingly provided almost no background reactions
in the absence of the template, and the S/B ratio was found to
be 223 after 30 s (Figure 3D).
ideal reaction mechanism for the turnover cycle would use the
DNA target as a catalyst. These conditions would typically use a
large excess of the probe relative to the number of template
molecules to drive the replacement of the fluorescent product
with the unreacted probe. In this way, the target concentrations
were varied over the range of 50 nM to 0.5 pM, and the probe
concentration was held constant at a considerable excess (50
nM). Under these conditions, the fluorescence signal was
continuously monitored as a function of time (Figure 4A). The
turnover number was subsequently obtained by dividing the
number of moles of the fluorescent product by the number of
moles of the template following a reaction period of 15 h
(Figure 4B). To provide a precise estimate of the actual
turnover number from the templated reaction, the slight
background signals were subtracted from those of identical
reactions that had been conducted in the absence of a template.
To determine the optimum reaction temperature, we
evaluated the effect of temperature on the turnover of the
reaction using different concentrations of the DNA template
(Figure S7, Supporting Information). The reaction temperature
was set at 15, 22, or 37 °C. A reaction temperature of 22 °C
was found to provide the best turnover number for the probe,
and an additional turnover experiment was carried out at 22 °C.
When the reaction was conducted with 0.1 equiv of the
template (5 nM), the fluorescence signal increased rapidly and
efficiently (Figure 4A), and provided a 94.2% yield of reaction
product after 15 h, as calculated from the fluorescence standard
(Figure 4B). In the absence of the template, a 4.5% yield of the
background reaction was observed after 15 h (Figure 4B).
When 0.1 equiv of the template was used, the turnover number
was estimated to be 9.0 following a background signal
correction. When even lower concentrations of the template
were used in the range of 10−2 to 10−5 equiv, significant
fluorescent signals originating from multiple turnover reactions
were observed over the background signals. The experiments
involving 10−4 (5 pM) and 10−5 equiv (0.5 pM) of the template
indicated that the turnover numbers were 624 and 1505,
respectively. To the best of our knowledge, these turnover
numbers represent the highest numbers reported to date for an
OTRC process. It is noteworthy that the detection limit for this
process was particularly low for an OTRC-based detection
system.
The equilibrium constants for the formation of the double
strands between the CNs-AMCA or MBA probes and the
template were calculated to be 47.1 × 106 and 9.6 × 106 M−1,
respectively, on the basis of a thermodynamic study (Figure S4,
Supporting Information). On the basis of these results, 52% of
the template formed a reaction complex in the solution with the
two probes binding with template at the same time at 22 °C.
Although the speed of the reaction was not maximized under
these experimental conditions because of the relatively low
levels of the reaction complex, the apparent rate constant of the
CNs-AMCA/MBA probe pair was calculated to be 2.45 × 10−2
s−1 (Figure S3, Supporting Information). Although an
estimation of the rate constant of this chemical reaction has
been provided in a later section, it is worth pointing out at this
stage that the rate of this chemical reaction was much greater
than that of any of the chemistries previously reported in this
area of research.1c,2b,c,3b,d,4a,5,6 For this reason, the decision was
taken to use the probe pair of CNs-AMCA and MBA in the
next round of turnover experiments.
For the CNs-AMCA probe to be successfully applied to the
detection of a biological sample, it needs to be stable to
biological compounds containing thiols, such as cysteine
residues. To determine the stability of the CNs-AMCA
probe, we treated the probe with different concentrations (1
μM to 1 mM) of glutathione (GSH) or dithiothreitol (DTT)
and monitored the resulting fluorescence signal for 30 min
(Figure S5, Supporting Information). The probe did not
provide significant signal even at a 1 mM concentration of GSH
or DTT. We also applied the probe to the detection of RNA in
living E. coli cells (Figure S6, Supporting Information). Briefly,
E. coli JM109 cells were treated with a CNs-AMCA and MBA
probe pair, which targets 23S rRNA, in buffer containing a low
concentration of SDS for 10 min at 25 °C, and the resulting
fluorescence images were recorded using a fluorescence
microscope. The CNs-AMCA and MBA matched probe pair
showed strong fluorescence signals, whereas the mismatched
probe pair did not provide a significant fluorescence signal.
These results indicated that the probe can detect specific RNA
sequences in living bacterial cells.
Steady-State Kinetic Analysis of Turnover Reaction.
The current probes were designed to kinetically favor the
turnover process in two different ways. First, the lengths of the
probes were designed to be short, with the MBA probe existing
as a 7-mer and the CNs-AMCA probe existing as an 8-mer. The
shorter DNA normally provides a greater dissociation constant
Turnover Measurements. We evaluated the potential for
the new probe pair to provide a level of catalytic turnover that
would enhance fluorescent signal amplification (Figure 1). An
D
dx.doi.org/10.1021/ja404743m | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
without changing the association constant.19 Second, the probe
pairing of CNs-AMCA and MBA was designed to provide a
rapid chemical reaction. It was envisaged that these properties
would contribute to an efficient turnover.
Unfortunately, however, the kinetic mechanisms associated
with turnover processes consisting of multiple steps remain
unclear. The steady state of the turnover reaction of the OTRC
has not been carefully analyzed from a kinetic perspective. With
this in mind, the decision was taken in the current study to
evaluate the kinetics of the turnover reaction shown in Figure 5
values for the probes were calculated by multiplying the
corresponding kon values by the concentration of the probe
(CNs-AMCA or MBA), as shown in Figure 4. The association
values calculated for the AMCA probe/SR and MBA probe/SR
were 51.6 × 10−2 and 89.6 × 10−2 s−1, respectively. These
values were found to be slightly faster than the corresponding
dissociation values, which were 2.2 × 10−2 and 18.6 × 10−2 s−1,
respectively, which allowed for the formation of the reaction
complexes between the AMCA and MBA probes and the SR. It
is noteworthy that a greater difference was observed between
the dissociation constants than the association constants. This
difference implied that the length of the probe DNA could be
used to tune the dissociation constant to a much greater extent
than it could be used to tune the association constant.
We then proceeded to analyze the template-concentration
dependence on the OTRC using a simple scheme (Figure S8,
Supporting Information) to estimate the reaction rates both in
the presence and in the absence of the template (kr+ and kr−,
respectively). We adopted three simple assumptions: (1) the
two probes independently associate with and dissociate from
the template, in that the order of the association or dissociation
of a probe does not affect the association or dissociation rates of
another probe to the same target; (2) the association and
dissociation rates do not change before or after the fluorogenic
reaction between the probes; and (3) the absorption coefficient
and quantum yield values of the fluorogenic probe remain the
same in the two different states of the probe, in that a template-
bound probe provides the same values as a free probe. On the
basis of these assumptions, we constructed a simple scheme
(Figure S8, Supporting Information) and analyzed all of the
data obtained from the template-concentration dependence. A
global fitting was constructed for all of these data over a long
time period (2.8 h) using a set of differential equations with the
association and dissociation rates provided for the two probes,
and estimated the rate constants of the chemical reaction both
in the presence and in the absence of the template to be 8.5 ×
10−2 and 64 M−1 s−1, respectively (Figure 5C and D). To fit the
experimental data, we also set a variable correlating with the
value for the absorption coefficient and quantum yield of the
fluorogenic probe (i.e., this constant converts the concentration
of the fluorogenic probe to the fluorescence intensity). The
constant was estimated to be 2.1 × 109 M−1 from the fitting
process. This value was similar to the equivalent experimentally
determined constant of 1.6 × 109 M−1, supporting the validity
of our estimations based on a simple model. Furthermore, the
fitting curves provided an effective description of the traces
obtained experimentally for the template concentrations across
5 orders of magnitude. On the basis of these results, we believe
that our model has succeeded in estimating the reaction rates in
the presence and in the absence of a template.
Figure 5. (A) Scheme of kinetic parameters. (B) Typical stopped-flow
kinetics runs in this study at 22 °C, showing primary data (blue) with
the best-fit single-exponential curve (red) superimposed. (C) Curve
fitting analysis. Time course of the yield, showing primary data with
the fitting curve (red) superimposed. The fitting was calculated by
Mathematica 7 (Wolfrum). (D) Kinetic parameters for the hybrid-
ization of each probe and DNA-templated reaction. SR 2 was 21 mer
a
ODN containing target sequence. van’t Hoff plot (see Figure S4,
b
c
Supporting Information). koff = kon/K. kr− ′ = kr− × 50 nM.
using a model scheme (see more detail in Figure S8,
Supporting Information) and estimate the rate constant of
the SNAr reaction on the basis of the steady state of the
turnover reaction.
We initially evaluated the association (kon) and dissociation
(koff) rate constants from the hybridization processes of the
probes and the template. Using a UV absorption spectrometer
equipped with a stopped-flow, the kon values for the
hybridization processes of each probe were directly measured
under pseudo-first-order conditions (Figure 5B). The equili-
brium constants (K) of CNs-AMCA probe/SR and MBA
probe/SR were also evaluated using a thermal denaturation
experiment (Figure S4, Supporting Information). The koff
values were calculated from the K and kon values by assuming
normal two-state binding for the complex. The resulting kinetic
parameters from the experiment, including the K, kon, and koff
values at 22 °C, have been summarized in Figure 5D. The kon′
DISCUSSION
■
New chemistry based on the reactions between the CNs and
the MBA groups offered a low detection limit of 0.5 pM, which
was made possible by the efficient turnover. The turnover
number is controlled by both thermodynamic and kinetic
factors. The thermodynamic factors effectively maximize the
theoretical turnover number without considering the total
reaction time. For example, nonligation type chemistry relieves
product inhibition by favoring certain thermodynamic factors,
whereas ligation type chemistry provides thermodynamic
drawback. The kinetic factors contribute to rate of a single
turnover (Figure 1). In principle, the rate of a single turnover
E
dx.doi.org/10.1021/ja404743m | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
could be increased by lowering the energy barrier of the rate
limiting step to result in an increase of the turnover number
over a given period of time.
there is still scope for improving the overall rapidity of the
reaction through the development of new chemistry, with
particular emphasis on important kinetic parameters such as koff
and kr+ toward more efficient turnovers in the nonligation type
chemistry.
The CNs-AMCA/MBA probe pair was successfully applied
to the detection of a specific RNA sequence in living bacterial
cells. The observation of a short emission wavelength from the
AMCA, however, led to an increase in the background signal
derived from the biological sample. As a possible solution to
this issue, it could be possible to use fluorescence dyes with
longer emission wavelengths such as rhodamine or cyanine
derivatives.
Consideration of the model scheme using steady state
reaction kinetics revealed the rates of turnover for the different
steps of the reaction, including the association (kon), chemical
reaction (kr), and dissociation (koff) steps, as shown in Figure 5.
The CNs/MBA reaction was rapid (kr+, 8.5 × 10−2 s−1) and
occurred with a very low level of background reactivity (kr−′,
3.2 × 10−6 s−1). In contrast to other chemistries previously
published in this area, which are slowest step on OTRC cycle,
the rate of the CNs/MBA chemical reaction step was similar to
that of dissociation step. The rate of the association (kon′) step
was one order faster step than other two steps under the
current reaction conditions (i.e., 50 nM of the probe). The rate
limiting step of the turnover cycle was therefore determined to
be the chemical reaction step and the dissociation step in this
particular chemistry.
In conclusion, we have developed a new fluorogenic probe
based on an oligonucleotide-templated SNAr reaction. The
probe completed its reaction within 30 s of being started and
amplified the fluorescence signal from the very low target
concentration by 1500-fold under isothermal conditions.
We then proceeded to simulate how each of the different
kinetic parameters (kon, kr+, koff,) and the template concen-
tration would impact on the turnover number using the model
scheme (Figure S8, Supporting Information). Figure 6D shows
ASSOCIATED CONTENT
* Supporting Information
Supporting figures, schemes, and experimental data. This
material is available free of charge via the Internet at http://
pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
h-abe@pharm.hokudai.ac.jp; y-ito@riken.jp
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
H.A. was financially supported by the Ministry of Education,
Culture, Sports, Science, and Technology (MEXT), Precursory
Research for Embryonic Science and Technology (PREST) and
the New Energy and Industrial Technology Development
Organization (NEDO). A.S. was financially supported by a
Grant-in-Aid for Young Scientists (B) (Grant No. 22790124)
and the Special Postdoctoral Researcher Program of RIKEN.
We are grateful for the support of BSI’s Research Resources
Center for mass spectra analysis.
Figure 6. Simulation of turnover number. (A−C) kr+ versus koff at 5 ×
10−13 M template. (A) kon = 1 × 105 M−1 s−1, (B) kon = 1 × 106 M−1
s−1, (C) kon = 1 × 107 M−1 s−1, (D) concentration of template versus
kr+ at kon = 1 × 106 M−1 s−1 and koff = 1 × 10−2 s−1.
REFERENCES
■
(1) (a) Sando, S.; Kool, E. T. J. Am. Chem. Soc. 2002, 124, 9686.
(b) Abe, H.; Kool, E. T. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 263.
(c) Furukawa, K.; Abe, H.; Hibino, K.; Sako, Y.; Tsuneda, S.; Ito, Y.
Bioconjugate Chem. 2009, 20, 1026. (d) Pianowski, Z.; Gorska, K.;
Oswald, L.; Merten, C. A.; Winssinger, N. J. Am. Chem. Soc. 2009, 131,
6492.
(2) (a) Dose, C.; Seitz, O. Org. Lett. 2005, 7, 4365. (b) Dose, C.;
Ficht, S.; Seitz, O. Angew. Chem., Int. Ed. 2006, 45, 5369. (c) Dose, C.;
Seitz, O. Bioorg. Med. Chem. 2008, 16, 65.
(3) (a) Ma, Z.; Taylor, J. S. Proc. Natl. Acad. Sci. U. S. A. 2000, 97,
11159. (b) Ma, Z.; Taylor, J. S. Bioorg. Med. Chem. 2001, 9, 2501.
(c) Ma, Z.; Taylor, J. S. Bioconjugate Chem 2003, 14, 679. (d) Cai, J.;
Li, X.; Taylor, J. S. Org. Lett. 2005, 7, 751.
(4) (a) Cai, J.; Li, X.; Yue, X.; Taylor, J. S. J. Am. Chem. Soc. 2004,
126, 16324. (b) Pianowski, Z. L.; Winssinger, N. Chem. Commun.
2007, 3820. (c) Franzini, R. M.; Kool, E. T. ChemBioChem 2008, 9,
2981. (d) Abe, H.; Wang, J.; Furukawa, K.; Oki, K.; Uda, M.; Tsuneda,
S.; Ito, Y. Bioconjugate Chem. 2008, 19, 1219. (e) Li, H.; Franzini, R.
M.; Bruner, C.; Kool, E. T. ChemBioChem 2010, 11, 2132.
(5) (a) Grossmann, T. N.; Seitz, O. J. Am. Chem. Soc. 2006, 128,
15596. (b) Grossmann, T. N.; Seitz, O. Chem.Eur. J. 2009, 15, 6723.
that although the turnover number increased with a reduction
in the template concentration, it reached a limit at a certain
concentration that was defined by kr+. Figure 6A,B,C indicates
that an increase in the kr+ value together with a moderate koff
value maximized the turnover number. In addition, increases in
the kon value led to a significant increase in the turnover
number. In practice, it is difficult to control the kon value
through the design of the nucleic acid probe. For example,
although the application of a locked nucleic acid monomer
enhanced the binding affinity for the complementary
oligonucleotide, it is provided not from increase of kon value
but from decrease of koff value.13 In contrast, the koff value could
be readily controlled by changing the probe length. For
example, the off-rate of the shorter MBA probe (7-mer) was
more rapid than that of the CNs-AMCA probe (8-mer) (Figure
5D). If the target sequence is GC-rich, the probe length should
be made shorter to allow for a quicker koff value and the
achievement of a higher turnover number. Finally, we believe
F
dx.doi.org/10.1021/ja404743m | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(6) Sando, S.; Sasaki, T.; Kanatani, K.; Aoyama, Y. J. Am. Chem. Soc.
2003, 125, 15720.
(7) (a) Silverman, A. P.; Kool, E. T. Chem. Rev. 2006, 106, 3775.
(b) Silverman, A. P.; Kool, E. T. Adv. Clin. Chem. 2007, 43, 79.
(8) Franzini, R. M.; Kool, E. T. Org. Lett. 2008, 10, 2935.
(9) (a) Okamoto, A.; Tanabe, K.; Inasaki, T.; Saito, I. Angew. Chem.,
Int. Ed. 2003, 42, 2502. (b) Tanabe, K.; Nakata, H.; Mukai, S.;
Nishimoto, S. I. Org. Biomol. Chem. 2005, 3, 3893. (c) Rothlingshofer,
̈
̈
M.; Gorska, K.; Winssinger, N. J. Am. Chem. Soc. 2011, 133, 18110.
(10) (a) Arian, D.; Clo,
E.; Gothelf, K. V.; Mokhir, A. Chem.Eur. J.
2010, 16, 288. (b) Arian, D.; Kovbasyuk, L.; Mokhir, A. Inorg. Chem.
2011, 50, 12010. (c) Rothlingshofer, M.; Gorska, K.; Winssinger, N.
́
̈
̈
Org. Lett. 2011, 14, 482. (d) Sadhu, K. K.; Winssinger, N. Chem.Eur.
J. 2013, 19, 8182.
(11) (a) Prusty, D. K.; Herrmann, A. J. Am. Chem. Soc. 2010, 132,
12197. (b) Prusty, D. K.; Kwak, M.; Wildeman, J.; Herrmann, A.
Angew. Chem., Int. Ed. 2012, 51, 11894.
(12) Grossmann, T. N.; Strohbach, A.; Seitz, O. ChemBioChem 2008,
9, 2185.
(13) Christensen, U.; Jacobsen, N.; Rajwanshi, V. K.; Wengel, J.;
Koch, T. Biochem. J. 2001, 354, 481.
(14) Shibata, A.; Abe, H.; Ito, M.; Kondo, Y.; Shimizu, S.; Aikawa, K.;
Ito, Y. Chem. Commun. 2009, 6586.
(15) Crich, D.; Sharma, I. Angew. Chem., Int. Ed. 2009, 48, 7591.
(16) Fuchs, B. M.; Syutsubo, K.; Ludwig, W.; Amann, R. Appl.
Environ. Microbiol. 2001, 67, 961.
(17) Jaffe, E. K.; Cohn, M. Biochemistry 1978, 17, 652.
(18) Gough, J. D.; Gargano, J. M.; Donofrio, A. E.; Lees, W. J.
Biochemistry 2003, 42, 11787.
(19) Koval, V. V.; Gnedenko, O. V.; Ivanov, Y. D.; Fedorova, O. S.;
Archakov, A. I.; Knorre, D. G. IUBMB Life 1999, 48, 317.
G
dx.doi.org/10.1021/ja404743m | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX