Coupling of DNA circuit and templated reactions for quadratic amplification and release of functional molecules

Article abstract of DOI:10.1021/jacs.9b05688

DNA-based circuitry empowers logic gated operations and amplifications but is restricted to nucleic acid output. Templated reactions enable the translation of nucleic acid cues into diverse small-molecule outputs but are more limited in their amplification. Herein, we demonstrate the coupling of a DNA circuit to templated reactions in order to achieve high levels of amplification in the output of small molecules, in response to nucleic acid input. We demonstrate that the coupling of the DNA circuit to templated reactions allows for the detection of the fM concentration of analyte and can respond with the release of a cytotoxic drug.

Full text of DOI:10.1021/jacs.9b05688

Subscriber access provided by BUPMC - Bibliothèque Universitaire Pierre et Marie Curie
Article
Coupling of DNA Circuit and Templated Reactions for
Quadratic Amplification and Release of Functional Molecules
Ki Tae Kim, Simona Angerani, Dalu Chang, and Nicolas Winssinger
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.9b05688 • Publication Date (Web): 20 Sep 2019
Downloaded from pubs.acs.org on September 20, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 10
Journal of the American Chemical Society
1
2
3
4
5
6
Coupling of DNA Circuit and Templated Reactions for Quadratic
Amplification and Release of Functional Molecules
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Ki Tae Kim, Simona Angerani, Dalu Chang and Nicolas Winssinger*
Department of Organic Chemistry, NCCR Chemical Biology, Faculty of Science, University of Geneva, 30 quai
Ernest Ansermet, 1205 Geneva, Switzerland
ABSTRACT: DNA-based circuitry empowers logic gated operations and amplifications but are restricted to a nucleic acid
output. Templated reactions enable the translation of a nucleic acid cues into diverse small molecule outputs but are more
limited in their amplification. Herein, we demonstrate the coupling of a DNA circuit to templated reactions in order to
achieve high levels of amplification in the output of small molecules, in response to a nucleic acid input. We demonstrate
that the coupling of the DNA circuit to templated reactions allow for the detection of fM concentration of analyte and can
respond with the release of a cytotoxic drug.
INTRODUCTION
cascades.25 Herein, we report the use of a DNA circuit that
respond to an oligonucleotide input with the formation
and amplification of a competent catalyst for templated
reactions (Figure 1). The DNA circuit is designed such that
the input is catalytic and thus amplifies the formation of
the product, which in turns templates a reaction capable of
turn-over (second amplification). The combined cycles of
the DNA-circuit with templated reactions result in
quadratic amplification.
The programmability and fidelity of hybridization makes
DNA an attractive platform to design networks and
assemblies with emergent properties or functions.
Developments
in
DNA
nanotechnologies
have
transformed our thinking about the applications of nucleic
1
-2
3
acids.
Like supramolecular chemistry,
DNA
nanotechnology is leveraged on reversible interactions that
exchange, error correct and respond. Exchanges between
strands can be programmed through toehold (a single
4
-5
strand overhang) displacements with controlled kinetics.
Strand displacement mechanisms have been used to
engineer networks capable of solving mathematical
operations and circuits capable of amplification or
6-15
molecular motion.
In parallel, DNA-templated
reactions have also been shown to be a powerful strategy
to translate a nucleic acid cue into a chemical reaction.16-17
Whereas the output of strand-displacement networks are
1
8
inherently oligonucleotides, templated reactions can be
used to uncage or synthesize functional molecules. It has
been shown that strand displacement and templated
reactions can be coupled to achieve logic-gated chemical
Figure 1. Schematic overview of DNA-fueled amplifier circuit
coupled to templated reaction
1
3, 19
transformation however,
only a fluorescent output has
RESULTS AND DISCUSSION
been demonstrated thus far and templated reactions have
not been coupled to DNA-based circuits for amplification.
This would be particularly relevant for amplifications in
which only low levels of inputs are available and
insufficient to generate critical threshold of output. For
instance, circulating microRNA are emerging as powerful
We hypothesized that templated reactions operating on
the products of a DNA amplification circuit would deliver
exponential amplifications that surpass the performance of
regular templated reactions (Figure 2). A prominent motif
in the products of DNA amplification circuits is a sticky
end, an unpaired short oligonucleotide at the end of a
duplex. We envisioned that this short stretch of DNA
would afford sites for templated reactions. The
amplification circuit is based on two metastable hairpins
functionalized with a ruthenium complex for templated
photocatalysis. Since the sequence for templated reaction
is part of the double-stranded stem region of the hairpin,
biomarkers but are typically present in sub-nanomolar
20-21
concentrations
be identified.
and other noncoding RNAs continue to
While templated reactions are capable of
22-23
signal amplification through template turn-over, few
examples surpass 100-fold amplification, and this level of
amplification requires long reaction times. Recently,
several strategies have been reported in attempts to
maximize amplifications in templated reactions, including
designs that maximize template turn-over2 and reaction
26
it is not available for templated reaction.
oligonucleotide trigger
An
4
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 10
for the templated reactions was unclear at the onset of this
work. Prior analysis of the rate of templated reactions
using PNA probes showed an optimal tradeoff between
affinity (K ) and turnover at 5-mer; reactions on shorter
M
sequences were limited by the weak affinity (koff too fast)
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
while longer sequences were limited by product
26
dissociation (koff too slow).
We began our study by optimizing the hairpin sequence
for the DNA circuit, using an architecture inspired by a
8
previous example of DNA fuel system. We designed the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
sequence to respond to miR-21 and use the corresponding
DNA sequence as initiator (I, miR-21-5p DNA). The length
of stem and loop parts of two hairpins were adjusted to
have the best meta-stability in the absence of I without
compromising the rate of the DNA circuit. Several pairs of
hairpins were tested, and A + B hairpin pair having 7-mer
overhang, 14-mer stem, and 9-mer loop was found to
perform best (Figure 3A, see Figure S1 for different
combinations tested). At 37°C, hairpin A and B (50 nM)
yielded >95% conversion into AB duplex within 1 h in the
presence of 0.125 eq of I, without any leakage products
(
uninitiated duplex formation, Figure 3B). Based on this
successful design of DNA hairpins, ruthenium
photocatalyst (Ru(bpy) phen) was introduced into 5ʹ end
a
2
of each A and B hairpin to produce RuA and RuB for
templated reactions. Likewise, the formation of the RuAB
duplex in the presence of I was monitored by the gel
electrophoresis. While the behavior of the RuA and RuB
circuit paralleled the performance of unmodified hairpins
(
(
A+B), Ru-modified hairpins exhibited slower kinetics
Figure 3C, less than 50 % conversion with 0.125 eq of I),
presumably due to additional stabilization of hairpins by
the electrostatic interaction between DNA and the cationic
ruthenium conjugate. To counter this effect and restore
the rate of the DNA circuit, we investigate the addition of
magnesium chloride in the buffer and found that indeed,
there was a positive correlation between the circuit rate
Figure 2. Schematic representation of linear templated
amplification (top) and exponential amplification used
designed in the present work (bottom).
(
initiator: I) opens the first hairpin by toehold
displacement resulting in a duplex with a long single stand
overhang. This product hybridizes to the second hairpin
which in turns displaces the initiator.
2
and MgCl (Figure 3C, see Figure S2 for titration); 8-12 mM
magnesium was sufficient to restore the rate of the circuit.
This concentration is in line with biological conditions
considering cell culture buffers often contain up to 10 mM
The dsDNA product now has two hybridization sites
open for templated reactions and the initiator turns over
in further cycles of dsDNA formation. A central point to
this double amplification is that the templated reactions
make use of a catalytic transformation such that each
dsDNA formed can convert multiple substrates. The
ruthenium photocatalysis is well suited for this purpose as
it is capable of high turn-overs in photocatalyzed reductive
reactions and has been shown to operate in complex
31-32
33
of magnesium
and seawater contain ~50 mM.
Importantly, the magnesium had no impact on the leakage
of the DNA circuit (Figure 3C, lane 7 from left).
To quantify the kinetics of the DNA circuit, fluorescence
quenching experiments were carried out using dabsyl-
conjugated hairpin (Dab-A) and fluorescein-conjugated
initiator (FAM-I). The hybridization of FAM-I and Dab-A
results in the quenching of the fluorescein signal by dabcyl
while introduction of RuB leads to release of FAM-I from
Dab-A and recovery of fluorescence signal. The reactions
27-29
biological environments.
The ground-state stability of
the catalyst coupled to its excited state reactivity is key to
3
0
its bioorthogonal chemistry. Upon photoexcitation with
st
were approximated as pseudo-1 -order when one of the
*
4
50 nm light and intersystem crossing, the Ru(II) reacts
reactants is in large excess (e.g., [Dab-A] >> [FAM-I]). We
measured a half-life of each step, finding the rate constants
with a stoichiometric reducing agent (NaAsc or NADPH in
a biological environment) to afford a Ru(I) complex, a
powerful single electron transfer (SET) reductant. The
pyridinium linker is an excellent acceptor in a SET reaction
leading to elimination of the coumarin and the pyridinium
with a reduced benzylic position. While these reactions are
very fast, the minimal length of the sticky end necessary
6
5
-1 -1
6
A B
k and k of 2.3×10 and 5.1×10 M s at 37 °C and 1.6×10
5
-1 -1
and 3.0× 10 M s at 25 °C, respectively (Figure 3D, see
st
Figure S3 for expanded analysis of pseudo 1 order analysis).
The circuit works both at room temperature or
ACS Paragon Plus Environment

Page 3 of 10
Journal of the American Chemical Society
physiological temperature and, as expected, is slightly
faster in the
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 3. (A) Sequences of A, B, and I with segment names. RuA and RuB have exactly same sequence with A and B, respectively.
DNA-fueled amplification of (B) AB or (C) RuAB duplexes bearing two sticky-ends, the sites for templated reaction, monitored
by 2 % agarose gel electrophoresis. The RuAB circuitry is accelerated by addition of 12 mM of MgCl
experiments for determination of kinetic parameters (k and k ) of DNA circuit. (E) Plot of fluorescence quenching experiments
performed with stoichiometric additions of Dab-A to FAM-I followed by RuB (second order kinetic measurments). Every
2
. (D) Fluorescence quenching
A
B
experiment was carried out in general Tris buffer conditions: 20 mM pH 7.5 Tris, 140 mM NaCl, 5 mM KCl, 12 mM MgCl
tween-20, 37 °C (reaction performed at 10 nM or reagents).
2
, 0.02 %
latter case (~2-times). The kinetic was also measured using
a second order rate constant analysis by the addition of
stochiometric amount of Dab-A to FAM-I followed by
stochiometric amount of RuB (Figure 3E). Interestingly,
comparison of rates for B vs RuB in the second step of the
RuA:RuB duplex in order to study each catalytic site
individually (RuS1:PC and RuS2:PC). We measured the
rate of templated coumarin uncaging as a function of time.
Among different types of PNA modifications tested, we
3
5-36
found that N-acetylated serine-modified
4-mer PNA
performed best (See Figure S5 for comparison of different
PNA backbones). Gratifyingly, the templated reaction on a
sticky end afforded fast conversion (Figure 4A,
t1/2<3 min) whereas ssDNA template
absence of sticky end) showed significantly slower
conversion (P1+RuS1, t1/2~40 min). This comparison
demonstrates the importance of base stacking in the
templated reactions. However, in the case of RuS2+P2ʹ+PC
circuit (I:A + B → A:B + I) showed that the cationic
ruthenium had a positive impact on the rate of that step
(2-fold acceleration, see Figure S4 for rate analysis),
P1+RuS1+PC,
(
3
4
presumably due to increased ionic interactions. Notably,
the kinetics of the present system, featuring 7-mer toehold,
is 3 to 23-times faster than that observed in prior DNA-
5
-1
-
fueled amplifier with shorter toehold (6-mer: 1.0×10 M s
1
8
)
. The overall data clearly show that the first amplification
(
the second sticky end site of RuA:RuB duplex) the
step, a toehold-fueled circuit, is successfully performed in
the presence of miR-21 DNA, exhibiting fast kinetics and
negligible leakage products.
reaction was found to be slower (t1/2 ~16 min, Figure S6).
This was attributed to the weaker stacking interaction of
pyrimidine (C) at N-terminal of P2ʹ with 3ʹ-end of PC vs a
purine in P1. This loss of stacking interaction was
overcome with the replacement of cytosine with a
phenoxazine nucleobase (Pz, Figure 4) which has an
extend -surface area and improves stacking onto the
sticky end. The Pz-containing PNA (P2) performed
comparably to P1 in templated reaction (Figure 4B and S6
for detailed analysis). Importantly, the 4-mer reaction was
The product of the DNA circuit is a duplex having two 4-
mer sticky ends at 5ʹ-ends, which serve as sites for
templated reactions. While this is generally too short for
DNA-templated reaction, we anticipated that the base-
staking arising from hybridization at a sticky-end may
offset the length deficiency. To evaluate the impact of this
additional stability, we used a truncated version of the
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 10
highly sequence-selective (Figure S7) and the use of a full
complementary sequence (FC) without sticky-ends
afforded catalytically inactive duplexes and negligible
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. Templated reaction on DNA sticky-ends using truncated versions of the RuA:RuB duplex. Underlined residues represent
serine-modified PNA. (A) Templated reaction between P1 and RuS1 using 1:1 stoichiometry (50 nM) at 25 °C in the presence or
absence of PC strand. (B) kapp of P1 and P2 with RuS1:PC and RuS2:PC respectively (6 μM, 1:1 stoichiometry). (C) Catalytic
M
performance of RuS and measurments of kcat and K from Lineweaver–Burk plot.
templated reaction (Figure S8). Taken together, the data
established that the templated reaction arises from duplex
formation at 4-mer sticky ends and this reaction is shut
down by masking the templating sequence in dsDNA.
individual steps of the system, we calculated the
theoretical kinetics of amplification and found a good
agreement with the experimental results (Figure 5B).
Within 40 min, this amplification afforded > 80 nM of
product from 1 nM of input. The maximum rate
determined by the slope at 60 min is 6-fold faster than the
linear templated reaction. This value is slightly lower than
calculated by the theoretical model, however, this model
does not take into consideration the competition of spent
PNA reagent competing with unreacted PNA reagent for
templated hybridization (product inhibition). Considering
that the system has already reached >25% conversion at 60
min, some product inhibition is to be expected. Similar
quadratic progressions were observed across different
initiator concentrations (31 pM-5 nM of I, Figure 5C) and
also at 37 °C (Figure S12). Notably, in case of 31 pM of I, the
yield (30 nM) corresponded to 960-fold amplification.
Increasing the PNA concentration to 5 μM further
accelerated the reaction, producing a total of 3.5 μM
concentration of uncaged coumarin (corrected for the
background reaction) after 80 min of irradiation (Figure
S13, 5 μM of P1 and P2, 10 nM of RuA and RuB, 1 nM of I,
We next measured the reactions’ half-life at higher
template loading (1-6 M), measuring an apparent rate
-1
constants (kapp) of 0.087 s for both reactions (Figure 4B
and S9, P1 and P2) at 25 °C. Notably, this rate 38-fold faster
than that of PNA-PNA templated reaction with 4-mer
26
(
reaction lacking the stabilization of a sticky end). To
investigate the kinetics in the presence of catalytic amount
of ruthenium complexes, initial rate of each reaction was
monitored at increasing concentrations of P1 or P2
(substrate) at 25 °C. Assuming Michaelis−Menten kinetics,
we determined kcat and K
(Figure 4C and S10): kcat = 0.053 and 0.052 s ; K
0.4 μM for P1 and P2, respectively. The reaction performed
M
based on Lineweaver–Burk plot
-1
M
= 0.24 and
well at both 25 and 37 °C. As expected, the K
templated reaction were higher at 37 °C vs 25 °C.
M
of the
We next investigated the coupling of DNA-fueled circuit
to the templated reactions. This is expected to be quadratic
amplification, in which templated reactions accelerate
with the accumulation of RuA:RuB generated by the DNA
circuit. As shown in Figure 5A, a clear rate acceleration is
indeed observed using 20% of initiator relatively to RuA
and B (5 nM) and 30-fold excess substrates (150 nM, see
also Figure S11 for reaction performed at different hairpin
concentrations). Comparing this exponential amplification
reaction to a linear amplification reaction with the
preformed catalyst (RuA:RuB) at the same loading of the
initiator, we observed that the exponential amplification
surpasses the linear amplification within a few minutes
3
0 min preincubation at 37 °C). This represent a 3500-fold
amplification and exceed background by 350%. Based on
the DNA circuit kinetics, conversion to RuA:RuB
represent > 9 turnovers of I, while each catalytic sites
performed >195 turnovers in the templated reaction. The
amplification also showed high selectivity for target I over
even single mismatched (SMM) DNA or RNA sequences
(
Figure 5D). Taken together, the results demonstrate
unprecedented amplification by the concerted action of a
DNA-amplification circuit with a templated reaction.
(
Figure 5B). Using the kinetic data measured for the
ACS Paragon Plus Environment

Page 5 of 10
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 5. Quadratic behavior of templated amplification coupled with toehold-fueled amplification at 25 °C. (A) Templated
reaction performed with or without 0.2 eq. of I. (B) Initial stage of the quadratic (5 nM of RuA and RuB) and linear templated
reaction (1 nM of RuA and RuB) in the presence of 1 nM of I. Templated reaction yield in the presence of different (C)
concentration of I or (D) types of oligonucleotide sequences (1 nM, 90 min of irradiation time). General conditions: 150 nM of
2
PNAs, 5 nM of DNAs, 20 mM pH 7.5 Tris, 140 mM NaCl, 5 mM KCl , 12 mM MgCl , 0.02 % tween-20, 5 mM sodium ascorbate, 25 °C,
200 μL, data was taken right after adding I. miR-21, SMM-miR-21, and TMM-miR-21 are DNA sequences. miR-31-5p and 3p are RNA
sequences.
To gain insight into the detection threshold of this system,
we measured endpoint fluorescence of amplification at
diminishing concentration of input sequence I. After 3 h,
discrimination of I vs control (no I) was possible down to
that suggest that >200 fM of I are required to produce the
minimum concentration of RuA:RuB (63 pM) for
detection (Figure S16). This kinetic model also provides a
framework to predict the reaction time necessary for a
given detection threshold.
8
pM (1.6 fmol, Figure 6). Pushing the reaction to 8 h led
to further gain in sensitivity with detection down to 2 pM
(400 amol, Figure S14). However, the kinetic models clearly
indicated that at low pM concentration, the DNA circuit is
much slower than the templated reaction and the system
would benefit from preincubation before irradiation.
Preincubation of the reaction for 24h followed by 8h
irradiation led to a significant improvement of detection in
the sub-picomolar range of I (Figure 6, see figure S14-15 for
intermediate preincubation times). This detection
threshold is consistent with the kinetic model calculation
Based on properties of this quadratic amplification with
a DNA circuit, this system was expected to be suitable for
micromolar release of bioactive molecules using nano- or
sub-nanomolar concentration of nucleic acid input. To this
end, 5-fluorouracil (5-FU), an anti-cancer drug, was
selected as a target of release. This drug has IC50 range in
37-39
the high nM to low μM in the most of cell lines.
5-FU
was caged with the pyridinium linker through O2 position
and conjugated to the PNA sequences required for
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 10
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 6. Detection threshold of quadratic amplification reaction tested with different irradiation and pre-incubation times.
Experiments were performed in general conditions. Modeled kinetics of RuAB DNA circuitry indicated that >200 fM of I produced
approximately 60 pM of RuAB, corresponding to the detection threshold of linear templated reaction, after 24 h of pre-incubation
at 37 °C. Statistics were calculated using an unpaired two-sample t-test (*p < 0.05, **p < 0.01). Detection threshold was defined as
the average of the no initiator (none) plus three times the standard deviation.
Figure 7. Release of functional molecule, 5-FU, by templated reaction coupled with DNA circuit. (A) HPLC analysis of released 5-
FU from the templated reaction. Conditions: 2 h of irradiation, 20 μM of 5FUP2 , 1 μM of RuA and RuB, 100 nM of I, 5 mM sodium
ascorbate, total V. 100 μL in general buffer. (B) Cytotoxicity assay of 5-FU and caged drugs (5FUP1+5FUP2) using HT-29 cells. (C)
HPLC analysis of 5-FU released by the coupled system in the presence of 1 nM of miR-21-5 with irradiation (light) or without
irradiation (no light). The area of 0.25 μM 5-FU (0.25 nmol / 100 μL) was used as the reference. Reaction conditions before direct
injection: 0.5 h of irradiation, 4 h of pre-incubation, 5 μM of each 5FUP1 and 5FUP2, 60 nM of RuA and RuB, 1 nM of miR-21-5, 5
mM of sodium ascorbate, total V. 100 μL in general buffer.
ACS Paragon Plus Environment

Page 7 of 10
1
Journal of the American Chemical Society
framework for the performance of the system. We
demonstrated its predictive value to establish the reaction
times required for a given output yield from a given input
loading. We demonstrated that the amplification allowed
detection of nucleic acid analytes down to 250 fM
concentration and the system offered strong
discrimination for even a single base pair mismatch in the
input nucleic acid sequence. The system was also capable
of reaching micromolar release of molecules in the
presence of nanomolar or sub-nanomolar trigger signal
templated reaction (5FUP1 & 2, Figure 7). To confirm
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
release of 5-FU, the reaction was first performed at high
concentration (20 μM of 5FUP2) in order to be analyzable
by HPLC. A near quantitative conversion of the caged 5-FU
to 5-FU was observed based on the HPLC traces (Figure 7A).
We confirmed that the pyridinium-linked PNA did mask
the function of 5-FU in a cytotoxicity assay against HT-29
40-42
(
colon adenocarcinoma).
As shown in Figure 7B, while
-FU showed the expected cytotoxicity with an IC50 of 350
5
(
nucleic acid sequence). This high level of trigger
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
nM
amplification was used to release a cytotoxic drug (5-FU)
using 1 nM of input (sequence of microRNA21-5p). While
the present work made use of the microRNA-21 sequence,
this design can be adapted to any nucleic acid trigger of 20-
in HT-29 (colorectal adenocarcinoma), the caged version
did not reach the measurable toxicity at 10 μM (Figure 7B).
We next set out to test the reaction in conditions that
emulate the trigger by a circulating RNA. To this end, the
reaction was performed with microRNA-21-5p, the mature
bioactive form of miR-21, an abundant miR biomarker of
2
5 nt. The high sensitivity, sequence selectivity, and large
amplification stem from the quadratic behavior of a DNA-
amplifier circuit coupled to DNA-templated reactions. The
major limitation of templated reactions is the trade-off
between probe affinity and turnover frequency; longer
probes having high template affinity show slow turnover
kinetics. The present system reconciles this trade off by
decoupling the response to templated reaction. First, the
DNA circuit generates the catalyst for templated reactions
with fast kinetics and turnover for long trigger sequences.
Second, 4-mer sticky ends are utilized as sites for
23
cancer. The reaction was performed at 5 μM of PNA
FUP1 and 5FUP2 (i.e. 10 μM prodrug concentration). The
5
uncaging was performed with 30 min irradiation following
a 4 h incubation in the presence or absence of 1 nM of miR-
2
1-5p. In the presence of miR-21-5p, the reaction yielded a
1
.4 μM of 5-FU, a concentration significantly above the IC50
,
while in the absence of miR-21-5p, the reaction resulted in
a concentration inferior to a significant cytotoxicity
threshold (0.19 μM. Figure 7C, see S18 for detailed
quantification). In the absence of light, no measurable
reaction was observed. This experiment demonstrates a
>1000-fold amplification in the output of a bioactive small
molecule from 1 nM input. Finally, the products of
amplification reactions were added to cell culture to
confirm the cytotoxicity of the uncaged product and lack
of cytotoxicity of the reaction components (See Figure S19,
RuA, RuB, prodrug) in the absence of light. While this is
only a proof of principle and the system likely needs
further optimization to be operative in live organisms, it
does demonstrate that a nucleic acid input can be
converted into a small molecule cytotoxic with >1000-fold
amplification.
-1
templated reaction affording one of the fastest kcat (0.05 s )
amongst templated reactions.
ASSOCIATED CONTENT
The Supporting Information is available free of charge on the
ACS Publications website at DOI:.
Supplementary figures, experimental protocols, synthetic
procedures and compound characterization.Raw data
associated with experiments reported has been deposited and
is available (www.zenodo.org)
AUTHOR INFORMATION
Corresponding Author
*
nicolas.winssinger@unige.ch.
CONCLUSION
Author Contributions
We report a system to convert a nucleic acid input to a
functional small molecule output with greater than 1000-
fold amplification in less than 3h under isothermal
conditions. This level of amplification was enabled by the
coupling of a DNA circuit and templated chemical
reactions. A critical finding is that templated reactions are
enhanced by sticky-end hybridization, thus extending the
scope of templated reaction to very short overhang
sequences (4-mer). This effect is accentuated by a purine
at the position interacting with the sticky end of the
template. However, nucleobase surrogates of pyrimidine
with larger -surface area (e.g. C to Pz substitution) can be
used for pyrimidines. The kinetics of all the steps in the
network were characterized, thus providing a theoretical
The manuscript was written through contributions of all
authors. All authors have given approval to the final version
of the manuscript.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENT
The authors thank the SNSF (grant: 200020_169141) and the
NCCR Chemical Biology for financial support.
REFERENCES
(
1) Seeman, N. C., DNA in a material world. Nature 2003, 421, 427-
(2) Seeman, N. C.; Sleiman, H. F., DNA nanotechnology. Nat. Rev.
Mater. 2018, 3, 17068.
4
31.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 10
(
3) Lehn, J. M., Toward self-organization and complex matter.
Science 2002, 295, 2400-2403.
4) Zhang, D. Y.; Winfree, E., Control of DNA Strand Displacement
Kinetics Using Toehold Exchange. J. Am. Chem. Soc. 2009, 131, 17303-
7314.
5) Srinivas, N.; Ouldridge, T. E.; Sulc, P.; Schaeffer, J. M.; Yurke, B.;
(24) Chang, D. L.; Kim, K. T.; Lindberg, E.; Winssinger, N.,
Accelerating Turnover Frequency in Nucleic Acid Templated
Reactions. Bioconj. Chem. 2018, 29, 158-163.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
(25) Velema, W. A.; Kool, E. T., Fluorogenic Templated Reaction
Cascades for RNA Detection. J. Am. Chem. Soc. 2017, 139, 5405-5411.
(26) Chang, D.; Lindberg, E.; Winssinger, N., Critical Analysis of
Rate Constants and Turnover Frequency in Nucleic Acid-Templated
Reactions: Reaching Terminal Velocity. J. Am. Chem. Soc. 2017, 139,
1444-1447.
(27) Holtzer, L.; Oleinich, I.; Anzola, M.; Lindberg, E.; Sadhu, K. K.;
Gonzalez-Gaitan, M.; Winssinger, N., Nucleic Acid Templated
Chemical Reaction in a Live Vertebrate. ACS Cent. Sci. 2016, 2, 394-
400.
(28) Lindberg, E.; Angerani, S.; Anzola, M.; Winssinger, N.,
Luciferase-induced photoreductive uncaging of small-molecule
effectors. Nat. Commun. 2018, 9.
(29) Anzola, M.; Winssinger, N., Turn On of a Ruthenium(II)
Photocatalyst by DNA-Templated Ligation. Chem. Eur. J. 2019, 25, 334-
342.
1
(
Louis, A. A.; Doye, J. P.; Winfree, E., On the biophysics and kinetics of
toehold-mediated DNA strand displacement. Nucleic Acids Res. 2013,
4
1, 10641-58.
6) Zhang, D. Y.; Seelig, G., Dynamic DNA nanotechnology using
(
strand-displacement reactions. Nat. Chem. 2011, 3, 103-113.
(7) Zhang, D. Y.; Turberfield, A. J.; Yurke, B.; Winfree, E.,
Engineering entropy-driven reactions and networks catalyzed by
DNA. Science 2007, 318, 1121-5.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
8) Yin, P.; Choi, H. M. T.; Calvert, C. R.; Pierce, N. A., Programming
biomolecular self-assembly pathways. Nature 2008, 451, 318-324.
(9) Qian, L.; Winfree, E., Scaling up digital circuit computation with
DNA strand displacement cascades. Science 2011, 332, 1196-201.
(
10) Krishnan, Y.; Simmel, F. C., Nucleic acid based molecular
devices. Angew. Chem. Int. Ed. Engl. 2011, 50, 3124-56.
11) Wang, F.; Lu, C. H.; Willner, I., From cascaded catalytic nucleic
(30) Angerani, S.; Winssinger, N., Visible Light Photoredox Catalysis
Using Ruthenium Complexes in Chemical Biology. Chem. Eur. J. 2019,
25, 6661-6672.
(
acids to enzyme-DNA nanostructures: controlling reactivity, sensing,
logic operations, and assembly of complex structures. Chem. Rev.
(31) Mooren, F. C.; Geada, M. M.; Singh, J.; Stoll, R.; Beil, W.;
Domschke, W., Effects of extracellular Mg2+ concentration on
intracellular signalling and acid secretion in rat gastric parietal cells.
Biochim. Biophys. Acta 1997, 1358, 279-88.
(32) Farruggia, G.; Castiglioni, S.; Sargenti, A.; Marraccini, C.;
Cazzaniga, A.; Merolle, L.; Iotti, S.; Cappadone, C.; Maier, J. A. M.,
Effects of supplementation with different Mg salts in cells: is there a
clue? Magnesium Res. 2014, 27, 25-33.
(33) Mewes, A.; Langer, G.; de Nooijer, L. J.; Bijma, J.; Reichart, G. J.,
Effect of different seawater Mg(2 +) concentrations on calcification in
two benthic foraminifers. Mar. Micropaleontol. 2014, 113, 56-64.
(34) Kim, W. J.; Akaike, T.; Maruyama, A., DNA strand exchange
stimulated by spontaneous complex formation with cationic comb-
type copolymer. J. Am. Chem. Soc. 2002, 124, 12676-12677.
(35) Dragulescu-Andrasi, A.; Rapireddy, S.; Frezza, B. M.; Gayathri,
C.; Gil, R. R.; Ly, D. H., A simple gamma-backbone modification
preorganizes peptide nucleic acid into a helical structure. J. Am. Chem.
Soc. 2006, 128, 10258-10267.
(36) Chouikhi, D.; Ciobanu, M.; Zambaldo, C.; Duplan, V.;
Barluenga, S.; Winssinger, N., Expanding the Scope of PNA-Encoded
Synthesis (PES): Mtt-Protected PNA Fully Orthogonal to Fmoc
Chemistry and a Broad Array of Robust Diversity-Generating
Reactions. Chem. Eur. J. 2012, 18, 12698-12704.
(37) Wiebke, E. A.; Grieshop, N. A.; Loehrer, P. J.; Eckert, G. J.;
Sidner, R. A., Antitumor effects of 5-fluorouracil on human colon
cancer cell lines: antagonism by levamisole. J. Surg. Res. 2003, 111, 63-
69.
(38) Flis, S.; Splwinski, J., Inhibitory effects of 5-fluorouracil and
oxaliplatin on human colorectal cancer cell survival are synergistically
enhanced by sulindac sulfide. Anticancer Res. 2009, 29, 435-441.
(39) Tong, J.; Xie, G.; He, J.; Li, J.; Pan, F.; Liang, H., Synergistic
antitumor effect of dichloroacetate in combination with 5-fluorouracil
in colorectal cancer. J. Biomed. Biotechnol. 2011, 2011, 740564.
(40) Violette, S.; Poulain, L.; Dussaulx, E.; Pepin, D.; Faussat, A. M.;
Chambaz, J.; Lacorte, J. M.; Staedel, C.; Lesuffleur, T., Resistance of
colon cancer cells to long-term 5-fluorouracil exposure is correlated
to the relative level of Bcl-2 and Bcl-X(L) in addition to Bax and p53
status. Int. J. Cancer 2002, 98, 498-504.
(41) Touil, Y.; Igoudjil, W.; Corvaisier, M.; Dessein, A. F.;
Vandomme, J.; Monte, D.; Stechly, L.; Skrypek, N.; Langlois, C.; Grard,
G.; Millet, G.; Leteurtre, E.; Dumont, P.; Truant, S.; Pruvot, F. R.;
Hebbar, M.; Fan, F.; Ellis, L. M.; Formstecher, P.; Van Seuningen, I.;
Gespach, C.; Polakowska, R.; Huet, G., Colon cancer cells escape 5FU
chemotherapy-induced cell death by entering stemness and
quiescence associated with the c-Yes/YAP axis. Clin. Cancer Res. 2014,
20, 837-846.
2
014, 114, 2881-941.
12) Wu, C. C.; Cansiz, S.; Zhang, L. Q.; Teng, I. T.; Qiu, L. P.; Li, J.;
(
Liu, Y.; Zhou, C. S.; Hu, R.; Zhang, T.; Cui, C.; Cui, L.; Tan, W. H., A
Nonenzymatic Hairpin DNA Cascade Reaction Provides High Signal
Gain of mRNA Imaging inside Live Cells. J. Am. Chem. Soc. 2015, 137,
4
900-4903.
13) Morihiro, K.; Ankenbruck, N.; Lukasak, B.; Deiters, A., Small
(
Molecule Release and Activation through DNA Computing. J. Am.
Chem. Soc. 2017, 139, 13909-13915.
(14) Sun, X.; Wei, B.; Guo, Y.; Xiao, S.; Li, X.; Yao, D.; Yin, X.; Liu, S.;
Liang, H., A Scalable "Junction Substrate" to Engineer Robust DNA
Circuits. J. Am. Chem. Soc. 2018, 140, 9979-9985.
(
15) Simmel, F. C.; Yurke, B.; Singh, H. R., Principles and
Applications of Nucleic Acid Strand Displacement Reactions. Chem.
Rev. 2019, 119, 6326-6369 .
(16) Gorska, K.; Winssinger, N., Reactions Templated by Nucleic
Acids: More Ways to Translate Oligonucleotide-Based Instructions
into Emerging Function. Angew. Chem. Int. Ed. Engl. 2013, 52, 6820-
6843.
(
17) Di Pisa, M.; Seitz, O., Nucleic Acid Templated Reactions for
Chemical Biology. ChemMedChem 2017, 12, 872-882.
(18) Benenson, Y.; Gil, B.; Ben-Dor, U.; Adar, R.; Shapiro, E., An
autonomous molecular computer for logical control of gene
expression. Nature 2004, 429, 423-429.
(
19) Rothlingshofer, M.; Gorska, K.; Winssinger, N., Nucleic Acid-
Templated Energy Transfer Leading to a Photorelease Reaction and
its Application to a System Displaying a Nonlinear Response. J. Am.
Chem. Soc. 2011, 133, 18110-18113.
(
20) Mitchell, P. S.; Parkin, R. K.; Kroh, E. M.; Fritz, B. R.; Wyman,
S. K.; Pogosova-Agadjanyan, E. L.; Peterson, A.; Noteboom, J.;
O'Briant, K. C.; Allen, A.; Lin, D. W.; Urban, N.; Drescher, C. W.;
Knudsen, B. S.; Stirewalt, D. L.; Gentleman, R.; Vessella, R. L.; Nelson,
P. S.; Martin, D. B.; Tewari, M., Circulating microRNAs as stable
blood-based markers for cancer detection. Proc. Natl. Acad. Sci. USA
2
008, 105, 10513-10518.
21) Dong, H. F.; Lei, J. P.; Ding, L.; Wen, Y. Q.; Ju, H. X.; Zhang, X.
(
J., MicroRNA: Function, Detection, and Bioanalysis. Chem. Rev. 2013,
113, 6207-6233.
(
22) Fish, L.; Zhang, S.; Yu, J. X.; Culbertson, B.; Zhou, A. Y.; Goga,
A.; Goodarzi, H., Cancer cells exploit an orphan RNA to drive
metastatic progression. Nat. Med. 2018, 24, 1743-1751.
(
23) Umu, S. U.; Langseth, H.; Bucher-Johannessen, C.; Fromm, B.;
Keller, A.; Meese, E.; Lauritzen, M.; Leithaug, M.; Lyle, R.; Rounge, T.
B., A comprehensive profile of circulating RNAs in human serum. RNA
Biol. 2018, 15, 242-250.
8
ACS Paragon Plus Environment

Page 9 of 10
Journal of the American Chemical Society
(
42) Leong, K. H.; Looi, C. Y.; Loong, X. M.; Cheah, F. K.; Supratman,
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
U.; Litaudon, M.; Mustafa, M. R.; Awang, K., Cycloart-24-ene-26-ol-3-
one, a New Cycloartane Isolated from Leaves of Aglaia exima Triggers
Tumour Necrosis Factor-Receptor 1-Mediated Caspase-Dependent
Apoptosis in Colon Cancer Cell Line. PLoS One 2016, 11, e0152652.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 10
TOC
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
10
ACS Paragon Plus Environment