Multi-level logic gate operation based on amplified aptasensor performance

Article abstract of DOI:10.1002/anie.201502315

Conventional electronic circuits can perform multi-level logic operations; however, this capability is rarely realized by biological logic gates. In addition, the question of how to close the gap between biomolecular computation and silicon-based electrical circuitry is still a key issue in the bioelectronics field. Here we explore a novel split aptamer-based multi-level logic gate built from INHIBIT and AND gates that performs a net XOR analysis, with electrochemical signal as output. Based on the aptamer-target interaction and a novel concept of electrochemical rectification, a relayed charge transfer occurs upon target binding between aptamer-linked redox probes and solution-phase probes, which amplifies the sensor signal and facilitates a straightforward and reliable diagnosis. This work reveals a new route for the design of bioelectronic logic circuits that can realize multi-level logic operation, which has the potential to simplify an otherwise complex diagnosis to a "yes" or "no" decision. Yes or no? A novel split aptamer-based multi-level logic gate, which is built from INHIBIT and AND gates, performs a net XOR analysis, with an electrochemical signal as output. This work reveals a new route for the design of bioelectronic logic circuits that can realize multi-level logic operations and which simplify an otherwise complex diagnosis to a yes/no decision.

Full text of DOI:10.1002/anie.201502315

Angewandte
Chemie
International Edition: DOI: 10.1002/anie.201502315
German Edition: DOI: 10.1002/ange.201502315
Logic Gates
Multi-Level Logic Gate Operation Based on Amplified Aptasensor
Performance**
Lingyan Feng, Zhaozi Lyu, Andreas Offenhäusser, and Dirk Mayer*
Abstract: Conventional electronic circuits can perform multi-
level logic operations; however, this capability is rarely realized
by biological logic gates. In addition, the question of how to
close the gap between biomolecular computation and silicon-
based electrical circuitry is still a key issue in the bioelectronics
field. Here we explore a novel split aptamer-based multi-level
logic gate built from INHIBIT and AND gates that performs
a net XOR analysis, with electrochemical signal as output.
Based on the aptamer–target interaction and a novel concept of
electrochemical rectification, a relayed charge transfer occurs
upon target binding between aptamer-linked redox probes and
solution-phase probes, which amplifies the sensor signal and
facilitates a straightforward and reliable diagnosis. This work
reveals a new route for the design of bioelectronic logic circuits
that can realize multi-level logic operation, which has the
potential to simplify an otherwise complex diagnosis to a “yes”
or “no” decision.
close the gap between DNA computation and electrical
circuitries, the future of such logic elements is closely related
to the successful linkage of nucleic acid to a conductive or
semiconductive support with electrochemical (electric) sig-
nals as their outputs, as well as functional DNA recognition
motifs to implement special biological applications. Just a few
pioneering attempts have been performed to immobilize
DNA logic gates on electrochemical supports, however, with
only one- or two-level logic operations.[5]
Recently, a novel kind of molecular recognition element
has been developed that is based on specific single-strand
DNA/RNA molecules selected from libraries with random
sequences.[6] The so-called aptamers offer great flexibility in
terms of structure variants to bind to a variety of targets with
high affinity and specificity, and are promising candidates for
bioanalytical applications and new computing systems.[7]
Moreover, when a single aptamer splits into two fragments,
they usually can form associated sandwich-type complexes
with their target.[8] Herein we extend this concept of a split
aptamer-based electrochemical sensor by incorporating
a novel electronic element, namely electrochemical current
rectification (ECR) for signal amplification capabilities and
its application in multi-level logic circuits.
Interest in ECR has increased rapidly due to its ability to
facilitate unidirectional current flow across an electrochem-
ical interface.[9] A characteristic property of the ECR process
reported here is that the entire charge transfer occurs only in
one direction between solution-phase redox probes and
electrode via surface-confined redox groups. It has been
successfully implemented in bioelectronic and bioelectro-
chemical systems through the incorporation and tailoring of
biomolecules into supramolecular bioassemblies to perform
electronic operations.[10] Furthermore, our group recently
reported the application of ECR for mimicking molecular
logic gate operations, as well as the realization of transistor
functions with signal amplification.[11]
We set out to design a multi-level aptamer-based logic
gate, which is built from sequential INHIBIT (INH) and
AND gates, that performs a net XOR analysis, with electro-
chemical signal as output. We used ATP and its split aptamers
as a model; one INH logic gate is initially constructed as the
starting point of the logic operation (Figure 1A). The 27-mer
ATP aptamer is divided into two parts with different
sequences.[8,12] First, the electrode surface is modified by
a 14-mer split aptamer (Apt-1) through covalent thiol–gold
bonds (Figure 1B, curve a). When the second 13-mer split
aptamer strand (Apt-2) bearing a ferrocene (Fc) tag is
A
wide and diverse range of DNA molecular devices have
been realized in the past decade with the goal of harnessing
specific properties of molecules as electric components for
molecular-scale computing.[1] Various Boolean logic opera-
tions have been developed with different applications in
biology, ranging from biosensors to bioimaging and thera-
peutics, which primarily focus on the basic logic functions,
AND, OR, and XOR.[2] For integrated electronic circuits,
a multi-level logic operation can be easily performed by
wiring different logic gates together electrically. However, the
connection of biological logic gates to form a multi-level
circuit poses significant challenges. Although enzymatic net-
works with hierarchical circuit design,[3] as well as modular
DNA-based Boolean logic gates wired into a three-level
circuit have been reported that exhibited a Boolean XOR
function,[4] most logic gates (mainly DNA-based) employ
fluorescent or colorimetric signals as their outputs. These
optical signals prove cumbersome for further linking of
different logic gates and their integration with electronic
devices. In order to design multi-level logic gates and further
[*] Dr. L. Feng, Z. Lyu, Prof. A. Offenhäusser, Dr. D. Mayer
Peter Grünberg Institute, PGI-8, Research Center Jülich
JARA – Fundamentals of Future Information Technology
Jülich 52425 (Germany)
E-mail: dirk.mayer@fz-juelich.de
[**] We thank Dr. Youlia Mourzina for helpful comments on the
manuscript and the Federal Ministry of Education and Research
(Germany), funding code 031A095B. Dr. L. Feng also thanks the
Alexander von Humboldt Foundation for a Humboldt fellowship.
applied,
a stable sandwiched structure forms through
target–ATP binding. An anodic peak at around 0.35 V and
a cathodic peak at around 0.28 V (vs. Ag/AgCl) can be
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201502315.
Angew. Chem. Int. Ed. 2015, 54, 7693 –7697
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
7693

.
Angewandte
Communications
together, but also to supply diagnostic information on
a coupled biological system.
To produce an INH logic gate, high levels of ATP (In1)
and ADA (In2) are defined at ꢀ 10 mm and at ꢀ 0.03 UmLÀ1,
respectively, as the “1” states while lower concentrations of
each of these are defined as the “0” states. For the output of
the INH gate, the electrochemical signal in the presence of
10 mm ATP represents the threshold. If the current value is
higher than this, it is considered as “1”, while lower values are
considered as “0”. In evaluating the four different states in the
presence and/or absence of ATP and ADA, namely (0,0) (1,0)
(0,1) (1,1), only (1,0) induces a significant relative current
change, resulting in a “1” output. Thus, an INH logic
operation can be realized by controlling the concentrations
of the ATP and ADA activity unit (Figure 1C and Figure S4).
One important concern about such a design is that the
signal obtained by this method is strongly dependent on the
surface probe density, which is typically in the range of a few
pmolcmÀ2 or even less. The low probe density usually affects
the platformꢀs sensitivity and applicability.[14] Therefore we
introduce the concept of electrochemical rectification (ECR)
for signal amplification.
Figure 1. A) Schematic illustration of the ATP split aptamer-based INH
logic gate. In the presence of ATP, a sandwich complex is formed
between the surface-tethered split aptamer (Apt-1) and the ferrocene-
containing split aptamer (Apt-2). ADA is able to catalyze the conver-
sion of ATP to inosine triphosphate (ITP). The latter shows no
interaction with the ATP aptamer and the sandwich aptamer complex
dissociates. B) CV curves of different electrode surfaces: a) Apt-1-
modified gold electrode; b) Apt-1-modified electrode after incubation
in 1 mm Apt-2 solution in the presence of 500 mm ATP and c) in the
presence of both ATP and 0.1 UmLÀ1 ADA. C) The INH logic scheme
and truth table.
Starting from the outputs of the INH gate, a third input,
ferrocyanide (In3), is then introduced into the system to
perform the second-level logic operation. The ECR effect is
realized through a relayed charge transport between split
aptamer-linked Fc groups and solution-phase redox probes as
shown in Figure 2A. Due to their different redox potentials,
the charge transfer can be efficiently relayed from the
measured (Figure 1B, curve ). The system
responds quite differently in the absence of ATP
molecules due to the thermal instability of target-
free aptamer complex.[13] The peak position and
the equal peak heights clearly indicate a reversible
and surface-confined electron transfer between Fc
and the electrode (Figure S1). Under optimized
modification conditions (Figures S2 and S3), the
DNA surface density is calculated to be about
(3.85 Æ 0.33) ” 1011 moleculescmÀ2, which corre-
sponds to previously reported DNA densities.[14]
Beyond that, an ATP–aptamer complex that has
already been formed can be dissociated by apply-
ing adenosine deaminase (ADA) (Figure 1B,
curve c). ADA is a key hydrolytic enzyme in
purine metabolism that can efficiently catalyze
the deamination of adenosine (deoxyadenosine)
Figure 2. A) Electrochemical current rectification (ECR) scheme. The ECR effect is
realized through a relayed charge transport between the split aptamer-linked Fc tag
and the solution-phase redox probes. B) CV curves in K4[Fe(CN)6] solutions of
different concentrations. Inset: the linear relationship between peak current and
into inosine (deoxyinosine), for which the ATP
aptamer shows no detectable affinity. This makes it
possible for ADA to break the formed sandwich
structure.[15] Inherited ADA deficiency is one main K4[Fe(CN)6] concentration. C) The two-level INH–AND logic scheme and truth table.
cause for severe combined immune deficiency
disease (SCID).[16] In contrast, ADA plethora
may also cause diseases, such as tuberculosis and hemolytic
anemia.[17] In healthy human plasma the ATP concentration is
roughly on the mm level, and the ADA activity is on the level
of 0.03 UmLÀ1.[18] Too high or too low ATP/ADA concen-
trations are critically harmful for human beings. Our present
logic gate is expected not only to link multi-level logic gates
electrode via the Fc center (Fc+/Fc redox potential around
0.32 V vs. Ag/AgCl) to ferrocyanide (Fe3+/Fe2+ redox poten-
tial 0.21 V vs. Ag/AgCl), which leads to a significant increase
in the anodic current (Figure 2A). In the reverse direction,
however, the electron-transfer process is thermodynamically
forbidden.[11,19] The surrounding of the surface-tethered
7694 www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 7693 –7697

Angewandte
Chemie
aptamer strand was backfilled with insulating 1-undecane-
thiol molecules to eliminate the direct charge transfer
between the electrode and the solution-phase redox probes
via defects in the self-assembled monolayer (Figure S5).[19]
The height of the sensor signal can be easily amplified by
tuning the concentration of electron-donating redox probes in
the solution (Figure 2B). The solution-phase redox probes
diffuse to the electrode to continuously replenish electrons to
the charge-transfer mediator and thus amplify the current
response with a linear relationship (Figure 2B, inset). A
concentration of 1 mm is selected for the solution-phase redox
probes in the following sections (Figures S6 and S7). The
whole rectified current (j) can be tuned through two adjust-
able chemical parameters [Eq. (1)]: the concentration of the
redox probes in solution as discussed above, and the
population of redox sites on the electrode introduced by
target binding and the formation of the aptamer sandwich
structure (Figure 2A).[10]
j ¼ Fkcross G0 cD
ð1Þ
Figure 3. A) CV curves with amplified anodic currents caused by
increasing ATP concentrations from 0.01 to 50 mm, in the presence of
1 mmK4[Fe(CN)6]. Inset: Enlargement of the linear region at low ATP
concentrations. B) CVs recorded from an as-prepared electrode with
the surface-tethered ATP/aptamer complex during ADA activity detec-
tion with different activity units (0, 0.0025, 0.005, 0.01, 0.03, 0.05,
0.1) UmLÀ1. Inset: Linear region of peak current and ADA activity unit.
Equation (1) includes the Faraday constant F and the rate
constant kcross for the reaction between surface-confined Fc
groups and the solution-phase redox probe. G0 is the coverage
of redox-active sites on the surface, and cD is the concen-
tration of the solution-phase donor (or acceptor) probe. The
number of surface redox sites on the electrode depends
exclusively on the amount of Fc-labeled Apt-2 aptamer in the
presence of ATP. The relationship between the amplified
peak current and the ATP target concentrations was deter-
mined, and it exhibited a sensitive response for the target with
a detection limit of about 7.5 nm (Figure 3A; Figure S8). This
result represents an improvement of around 130 times
compared with the detection limit of 1 mm achieved in the
absence of ECR-based amplification (Figure S9). The apta-
sensor selectivity was proven through a comparison with
other target analogues and a direct detection in complicated
serum solutions (Figure S10). Based on the same mechanism,
this aptasensor is also sensitive to ADA activity response,
since Fc-linked Apt-2 is released from the surface after
enzyme treatment (Figure S11). An ADA activity detection
as low as 0.002 UmlÀ1 can be obtained with a higher
sensitivity and an excellent selectivity compared with pre-
viously reports (Figure 3B, Figure S12, and Table S1).
Neither the detection of ATP nor ADA activity is affected
by the presence of interfering compounds in complex media.
Furthermore, the magnification of signal induced by ATP
binding occurs without the permeation of the mixed self-
assembled monolayer by solution-phase probes. This is an
important advantage of ECR-based amplification compared
with previously reported electrocatalytic amplification prin-
ciples, since a permeation reaction would reduce the charge-
transfer efficacy between surface-confined probe and solu-
tion-phase probe.[20] Our simple ECR amplification scheme
does not rely on comparatively expensive enzymes, stringent
environmental conditions, or complicated synthesis process-
es.[21] Therefore it can be concluded that the concept of ECR-
based signal amplification provides a suitable and more
general tool for the enhancement of aptamer-based biosensor
performance.
However, the additional third input (In3) not only enables
signal amplification but also introduces a second-level logic
gate, which facilitates distinguishable output signals by raising
the output threshold values. Therefore, the output of the INH
logic gate described above and In3 can be considered as
inputs to enter into the next INH–AND logic operation, with
logic scheme and corresponding truth table shown in Fig-
ure 2C. Noteworthy, here we defined different threshold
values for the final output current produced by the split
aptamer-based electrode due to ECR-based signal amplifica-
tion. The region with an absolute current value higher than
1.5 mA (when ATP ꢀ 10 mm or ADA < 0.01 UmLÀ1) was
defined as “1”, and the second region with a current value
lower than 0.1 mA (when ATP < 0.1 mm or ADA ꢀ
0.03 UmLÀ1) was defined as “0”, with a difference of
a factor of 15 between them to avoid ambiguous states
(Figure 3). This gray region between “1” and “0”, which lacks
any definite response, corresponds to the usual amount of
ATP (0.1 to 10 mm) and ADA activity unit (0.01 to
0.03 UmLÀ1) in normal human serum.[18] This definition is
also analogous to the approach used in electronics where only
appropriate voltage regions are employed to define the logic
functions. With the help of ECR, the different sensor
responses to ATP concentration and ADA activity are
much easier to analyze and the logic gate directly indicates
different states of health. The current increases significantly
only when In1 = 1 (ATP ꢀ 10 mm), In2 = 0 (ADA <
0.01 UmLÀ1), and In3 = 1 (1 mm ferrocyanide), yielding
Angew. Chem. Int. Ed. 2015, 54, 7693 –7697
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7695

.
Angewandte
Communications
a “1” output in the presence of solution-phase probes for
ECR-aided signal amplification. This usually indicates a dis-
ease status, for instance, the presence of a severe combined
immunodeficiency.[22] Otherwise, the current intensity is lower
which corresponds to an output signal “0” (Figure 2C and
Figure S13). According to the clinical studies,[18b] patients with
tuberculosis usually have higher levels of ADA activity
(ADA ꢀ 0.03 UmLÀ1) in their serum, which corresponds to
inputs (0,1,1) or (1,1,1). Both of the output signals, “0” and
“1”, supply diagnosis information that indicate a physical
impairment. Only if the inputs are within the gray range is the
diagnosis without findings.
as discussed above in the INH–AND logic gate, three regions
of “1”, “0”, and “gray” are adopted for the intensity of output
signals. A three-level logic function can be realized by
adjusting different input values. When only one type of
redox probe is added to the solution, the aptamer-modified
electrode can be considered as an anodic or cathodic ECR.
This ECR functions similar to a diode, which facilitates an
electric current in one direction (forward-biased conditions)
and blocks the current in the opposite direction (reverse-
biased conditions), resulting in an INH–AND logic gate
(Figure 4B, curves b and c). When both In3 and In4 are
applied simultaneously, the current responses decrease
largely because of the interplay of different interfacial and
solution-based charge-transfer processes (Figure 4B curve d
and Figure 4C).[11] The multi-level logic gate performs a final
XOR logic-gate function which results from the combination
of INH and AND gates (Figure 4D and Figure S14). Two
inputs sets (1,0,1,0) and (1,0,0,1) give a final “1” output, in
which the current is recorded as an absolute value
independent of the current direction. The XOR logic
gate can be used to eliminate inaccurate diagnoses caused
by the system malfunctions. High noise components in the
sensor signal could spuriously lead to a false “1” readout.
However, this output should remain only if In3 and In4
are added together for a defective system; it disappears if
the system works properly.
Based on these amplified aptasensor performances with
the ECR effect, we continue to construct a multi-level logic
gate and combine the INH–AND gate with an XOR function
(Figure 4A). Instead of an anodic rectification based on
The full function of the multi-level logic gate corre-
sponds to an INH–AND–XOR operation. It combines
the aptamer-based biochemical logic gate responses of
the sensor receptor (INH) with the logic gates of the
electrochemical transduction scheme (AND–XOR). By
this means a biochemical binding process is transduced
into an electrical signal, the sensor signal is enhanced by
the ECR effect, and several (bio)chemical input signals
are converted into one output signal which reports on the
overall status of the system. Different target ranges
induce distinguishable and easy to analyze “yes” or “no”
outputs, which is of importance in particular for point-of-
care diagnostics.
Figure 4. A) The INH–AND–XOR logic scheme built from INHIBIT and AND
gates that performs a net XOR analysis. B) The as-prepared split aptamer
sensor in the presence of different inputs: a (1,0,0,0), b (1,0,1,0), c (1,0,0,1),
and d (1,0,1,1). The anodic and cathodic current amplification are realized
with 1 mm K4[Fe(CN)6] and 1 mmK2IrCl6, respectively, in the same range of
CV potentials. C) Schematic illustration of the electron-transfer processes
between redox species and the aptamer-modified electrode surface. The
chemical equation demonstrates the direct charge transfer between the two
redox probes in solution. D) Truth table of the multi-level logic gate
designed for this work.
In conclusion, we have successfully established
a novel multi-level logic gate based on a sensitive
aptamer-binding reaction and the concept of electro-
chemical rectification, with a final electrical signal output
and amplification strategy for logic performance and
detection. A rectified charge transport originates from
the consecutive charge transfer between the electrode,
aptamer-confined surface redox tag, and the redox probes
in solution. With ECR as a novel signal amplification strategy,
our work offers a sensitive, straightforward, and robust
detection technique for aptamer-based electrochemical bio-
sensors. In addition, by carefully considering the concentra-
tions of the ATP and ADA activity unit and integrating the
resulting anodic and cathodic current rectifier, we have
realized an INH–AND–XOR multi-level logic gate that
shows a high switching ratio between output signals “1” and
“0” with a final net XOR analysis, which makes it possible to
distinguish different health statuses and facilitates fast and
accurate diagnosis at the same time. Our sensor may serve as
a promising proof of principle that demonstrates increased
ferrocyanide, the cathodic current can also be amplified
simply by changing the solution redox species from ferrocya-
nide (0.21 V vs. Ag/AgCl) to hexachloriridate(IV) (0.68 V vs.
Ag/AgCl). The different redox potentials of the solution-
phase redox probes lead to different directions of the charge
transfer between the surface-bound Fc center (Fc+/Fc redox
potential 0.32 V vs. Ag/AgCl) and electron-donating (ferro-
cyanide, Fe3+/Fe2+)/electron-accepting (hexachloriridate, Ir4+/
Ir3+) species in solution (Figure 4B). In3 and In4 correspond
to the addition of 1 mm ferrocyanide and iridate(IV) ions,
respectively, while the absence of the respective species is
considered to constitute “0”. With the same output definition
7696 www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 7693 –7697

Angewandte
Chemie
[10] a) O. Azzaroni, M. lvarez, A. I. Abou-Kandil, B. Yameen, W.
computational complexity and enhanced sensor performance
by linking multiple logic gates together.
Knoll, Adv. Funct. Mater. 2008, 18, 3487 – 3496; b) O. Azzaroni,
M. Mir, M. lvarez, L. Tiefenauer, W. Knoll, Langmuir 2008, 24,
2878 – 2883; c) D. Rong, H.-G. Hong, Y. I. Kim, J. S. Krueger,
J. E. Mayer, T. E. Mallouk, Coord. Chem. Rev. 1990, 97, 237 –
248; d) Y. Liu, A. Offenhäusser, D. Mayer, Bioelectrochemistry
2010, 77, 89 – 93.
Keywords: aptamers · diagnosis · electrochemical rectification ·
logic gates · signal amplification
How to cite: Angew. Chem. Int. Ed. 2015, 54, 7693–7697
Angew. Chem. 2015, 127, 7803–7808
[11] a) Y. Liu, A. Offenhäusser, D. Mayer, Angew. Chem. Int. Ed.
2010, 49, 2595 – 2598; Angew. Chem. 2010, 122, 2649 – 2652; b) Y.
Liu, B. Wolfrum, M. Hüske, A. Offenhäusser, E. Wang, D.
Mayer, Angew. Chem. Int. Ed. 2013, 52, 4029 – 4032; Angew.
Chem. 2013, 125, 4121 – 4124.
[1] a) X. P. Su, L. M. Smith, Nucleic Acids Res. 2004, 32, 3115 – 3123;
b) Y. Krishnan, F. C. Simmel, Angew. Chem. Int. Ed. 2011, 50,
3124 – 3156; Angew. Chem. 2011, 123, 3180 – 3215; c) L. Qian, E.
Winfree, J. Bruck, Nature 2011, 475, 368 – 372.
[12] A. K. Sharma, J. M. Heemstra, J. Am. Chem. Soc. 2011, 133,
12426 – 12429.
[2] a) F. Wang, X. Liu, I. Willner, Angew. Chem. Int. Ed. 2015, 54,
1098 – 1129; Angew. Chem. 2015, 127, 1112 – 1114; b) K. Szaci-
lowski, Infochemistry, Wiley, Chichester, 2012; c) A. P. de Silva,
Molecular Logic-Based Computation, Royal Society of Chemis-
try, Cambridge, 2013; d) Molecular and Supramolecular Infor-
mation Processing-From Molecular Swithces to Uncoventional
Computing (Ed.: E. Katz), Willey-VCH, Weinheim, 2012;
e) Bimolecular Computing-From Logic Systems to Smart Sensors
and Actuators (Ed.: E. Katz), Willey-VCH, Weinheim, 2012;
f) M. N. Stojanovic, T. E. Mitchell, D. Stefanovic, J. Am. Chem.
Soc. 2002, 124, 3555 – 3561; g) J. Elbaz, F. Wang, F. Remacle, I.
Willner, Nano Lett. 2012, 12, 6049 – 6054; h) T. Li, E. Wang, S.
Dong, J. Am. Chem. Soc. 2009, 131, 15082 – 15083; i) F. Pu, J.
Ren, X. Qu, Adv. Mater. 2014, 26, 5742 – 5757.
[3] a) M. N. Stojanovic, D. Stefanovic, J. Am. Chem. Soc. 2003, 125,
6673 – 6676; b) E. Katz, Curr. Opin. Chem. Biol. 2015, 34, 202 –
208; c) E. Katz, V. Privman, Chem. Soc. Rev. 2010, 39, 1835 –
1857.
[4] B. M. Frezza, S. L. Cockroft, M. R. Ghadiri, J. Am. Chem. Soc.
2007, 129, 14875 – 14879.
[5] a) Y. Jia, R. Duan, F. Hong, B. Wang, N. Liu, F. Xia, Soft Matter
2013, 9, 6571 – 6577; b) D. Kang, R. J. White, F. Xia, X. Zuo, A.
Vallee-Belisle, K. W. Plaxco, NPG Asia Mater. 2012, 4, e1; c) F.
Xia, X. Zuo, R. Yang, R. J. White, Y. Xiao, D. Kang, X. Gong,
A. A. Lubin, A. VallØe-BØlisle, J. D. Yuen, B. Y. B. Hsu, K. W.
Plaxco, J. Am. Chem. Soc. 2010, 132, 8557 – 8559.
[6] A. D. Ellington, J. W. Szostak, Nature 1990, 346, 818 – 822.
[7] a) M. Famulok, G. Mayer, Acc. Chem. Res. 2011, 44, 1349 – 1358;
b) A. A. Lubin, K. W. Plaxco, Acc. Chem. Res. 2010, 43, 496 –
505.
[8] a) X. Zuo, Y. Xiao, K. W. Plaxco, J. Am. Chem. Soc. 2009, 131,
6944 – 6945; b) J. Elbaz, M. Moshe, I. Willner, Angew. Chem. Int.
Ed. 2009, 48, 3834 – 3837; Angew. Chem. 2009, 121, 3892 – 3895;
c) Z. Liu, W. Zhang, L. Hu, H. Li, S. Zhu, G. Xu, Chem. Eur. J.
2010, 16, 13356 – 13359.
[13] H. Urata, K. Nomura, S.-i. Wada, M. Akagi, Biochem. Biophys.
Res. Commun. 2007, 360, 459 – 463.
[14] a) R. J. White, N. Phares, A. A. Lubin, Y. Xiao, K. W. Plaxco,
Langmuir 2008, 24, 10513 – 10518; b) S. J. P. Canete, W. Yang,
R. Y. Lai, Chem. Commun. 2009, 4835 – 4837.
[15] a) R. Nutiu, J. M. Y. Yu, Y. Li, ChemBioChem 2004, 5, 1139 –
1144; b) L. Feng, Z. Zhang, J. Ren, X. Qu, Biosens. Bioelectron.
2014, 62, 52 – 58.
[16] J. J. Sanchez, G. Monaghan, C. Børsting, G. Norbury, N. Morling,
H. B. Gaspar, Ann. Hum. Genet. 2007, 71, 336 – 347.
[17] X.-J. Xing, X.-G. Liu, H. Yue, Q.-Y. Luo, H.-W. Tang, D.-W.
Pang, Biosens. Bioelectron. 2012, 37, 61 – 67.
[18] a) M. W. Gorman, E. O. Feigl, C. W. Buffington, Clin. Chem.
2007, 53, 318 – 325; b) M. F. Baganha, A. PÞgo, M. A. Lima, E. V.
Gaspar, A. R. Cordeiro, Chest 1990, 97, 605 – 610.
[19] a) H. Li, D. Li, J. Liu, Y. Qin, J. Ren, S. Xu, Y. Liu, D. Mayer, E.
Wang, Chem. Commun. 2012, 48, 2594 – 2596; b) L. Feng, A.
Sivanesan, Z. Lyu, A. Offenhäusser, D. Mayer, Biosens. Bio-
electron. 2015, 66, 62 – 68.
[20] J. Zhang, L. Wang, D. Pan, S. Song, C. Fan, Chem. Commun.
2007, 1154 – 1156.
[21] a) E. J. Cho, L. Yang, M. Levy, A. D. Ellington, J. Am. Chem.
Soc. 2005, 127, 2022 – 2023; b) K. Hsieh, Y. Xiao, H. T. Soh,
Langmuir 2010, 26, 10392 – 10396; c) X. Chen, L. Ge, B. Guo, M.
Yan, N. Hao, L. Xu, Biosens. Bioelectron. 2014, 58, 48 – 56; d) K.-
J. Huang, Y.-J. Liu, H.-B. Wang, Y.-Y. Wang, Y.-M. Liu, Biosens.
Bioelectron. 2014, 55, 195 – 202; e) I. Palchetti, M. Mascini, Anal.
Bioanal. Chem. 2012, 402, 3103 – 3114.
[22] a) M. S. Coleman, J. Donofrio, J. J. Hutton, L. Hahn, J. Biol.
Chem. 1978, 253, 1619 – 1626; b) H. A. Simmonds, R. J. Levin-
sky, D. Perrett, D. R. Webster, Biochem. Pharmacol. 1982, 31,
947 – 951.
[9] P. Denisevich, K. W. Willman, R. W. Murray, J. Am. Chem. Soc.
1981, 103, 4727 – 4737.
Received: March 12, 2015
Published online: May 8, 2015
Angew. Chem. Int. Ed. 2015, 54, 7693 –7697
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 7697