DNA sensors and aptasensors based on the hemin/G-quadruplex-controlled aggregation of Au NPs in the presence of L-cysteine

Article abstract of DOI:10.1002/smll.201400002

L-cysteine induces the aggregation of Au nanoparticles (NPs), resulting in a color transition from red to blue due to interparticle plasmonic coupling in the aggregated structure. The hemin/G-quadruplex horseradish peroxidase-mimicking DNAzyme catalyzes the aerobic oxidation of L-cysteine to cystine, a process that inhibits the aggregation of the NPs. The degree of inhibition of the aggregation process is controlled by the concentration of the DNAzyme in the system. These functions are implemented to develop sensing platforms for the detection of a target DNA, for the analysis of aptamer-substrate complexes, and for the analysis of L-cysteine in human urine samples. A hairpin DNA structure that includes a recognition site for the DNA analyte and a caged G-quadruplex sequence, is opened in the presence of the target DNA. The resulting self-assembled hemin/G-quadruplex acts as catalyst that controls the aggregation of the Au NPs. Also, the thrombin-binding aptamer folds into a G-quadruplex nanostructure upon binding to thrombin. The association of hemin to the resulting G-quadruplex aptamer-thrombin complex leads to a catalytic label that controls the L-cysteine-mediated aggregation of the Au NPs. The hemin/G-qaudruplex-controlled aggregation of Au NPs process is further implemented for visual and spectroscopic detection of L-cysteine concentration in urine samples.

Full text of DOI:10.1002/smll.201400002

DNA Sensors
DNA Sensors and Aptasensors Based on the
Hemin/G-quadruplex-Controlled Aggregation
of Au NPs in the Presence of L-Cysteine
Angelica Niazov-Elkan, Eyal Golub, Etery Sharon, Dora Balogh, and Itamar Willner*
L -cysteine induces the aggregation of Au nanoparticles (NPs), resulting in a color
transition from red to blue due to interparticle plasmonic coupling in the aggregated
structure. The hemin/G-quadruplex horseradish peroxidase-mimicking DNAzyme
catalyzes the aerobic oxidation of L-cysteine to cystine, a process that inhibits
the aggregation of the NPs. The degree of inhibition of the aggregation process is
controlled by the concentration of the DNAzyme in the system. These functions are
implemented to develop sensing platforms for the detection of a target DNA, for the
analysis of aptamer-substrate complexes, and for the analysis of L-cysteine in human
urine samples. A hairpin DNA structure that includes a recognition site for the
DNA analyte and a caged G-quadruplex sequence, is opened in the presence of the
target DNA. The resulting self-assembled hemin/G-quadruplex acts as catalyst that
controls the aggregation of the Au NPs. Also, the thrombin-binding aptamer folds
into a G-quadruplex nanostructure upon binding to thrombin. The association of
hemin to the resulting G-quadruplex aptamer-thrombin complex leads to a catalytic
label that controls the L-cysteine-mediated aggregation of the Au NPs. The hemin/G-
qaudruplex-controlled aggregation of Au NPs process was further implemented for
visual and spectroscopic detection of L-cysteine concentration in urine samples.
different sensing platforms.[2] These included complementary
1
. Introduction
[
3]
H-bonds between molecularly modified NPs, ion-induced
[
4]
The aggregation of Au nanoparticles (Au NPs) has been a key aggregation of ligand-functionalized Au NPs, host-guest
[
5]
[6]
phenomenon in the development of numerous molecular and interactions,
donor-acceptor affinity interactions
and
[
1]
biomolecular sensors. The color changes associated with complementary bio-recognition interactions (e.g., biotin-
[
7]
the aggregation of Au NPs, originating from the inter-particle avidin or the formation of antigen-antibody complexes).
plasmonic coupling between the localized plasmons of the Also, controlling the surface charge associated with Au NPs
NPs, provided a general mechanism for the development of provided a general means to induce aggregation.[ The
most extensively studied Au NPs aggregation systems have
8]
A. Niazov-Elkan,[+] E. Golub,[+] E. Sharon,
D. Balogh, Prof. I. Willner
The Institute of Chemistry
The Minerva Center for Biohybrid Complex Systems
The Hebrew University of Jerusalem
Jerusalem 91904, Israel
involved nucleic acid-functionalized Au NPs. The bridging of
nucleic acid-modified Au NPs by a complementary nucleic
acid has been a versatile route to design various DNA sen-
[9]
sors. Also, the aggregation of Au NPs functionalized with
[10]
2+
aptamer subunits,
or the Hg -induced aggregation of
nucleic acid-modified Au NPs through the bridging of the
E-mail: willnea@vms.huji.ac.il
2
+
[11]
NPs by forming thymine-Hg -thymine complexes,
were
[
+]Angelica Niazov-Elkan and Eyal Golub contributed equally to this work.
used to develop aptasensors and Hg2+ sensors. Alternatively,
DNAzyme-bridged aggregated Au NPs were implemented to
DOI: 10.1002/smll.201400002
small 2014,
DOI: 10.1002/smll.201400002
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
wileyonlinelibrary.com
1

full papers
A. Niazov-Elkan et al.
analyze Pb2+ ions by the ion-induced DNAzyme cleavage of and H O ,
[21]
the polymerization of aniline to polyaniline in
2
2
[
32]
the bridging nucleic acids, and the resulting deaggregation of the presence of H O ,
the epoxidation of olefins and the
2
2
[
12]
[33]
the Au NPs.
In addition, metal NPs assemblies have been used to
oxidative coupling of carbon-carbon bonds.
In the present study we implement the hemin/G-quadru-
develop other optical or electronic sensors. The imprinting of plex-catalyzed aerobic oxidation of thiols to disulfides as a
selective molecular binding sites by the electropolymeriza- means to control the L-cysteine-mediated aggregation of Au
tion of Au NPs matrices, in the presence of analogues of the NPs. We use the color changes occurring upon the hemin/G-
analytes, onto Au-coated glass surfaces provided a general quadruplex-inhibited aggregation of the NPs to develop a
means for the selective, surface plasmon resonance (SPR)- DNA sensor, aptasensors, and to sense L-cysteine in human
mediated detection of the analytes in the imprinted sites. urine samples.
The coupling between the localized plasmon of the NPs and
the surface plasmon wave provided a means to amplify the
dielectric changes in the sensing matrices upon the binding 2. Results and Discussion
of the analytes to the respective imprinted sites. These
[
13]
imprinted matrices were used to detect explosives,
antibi- L-Cysteine plays a key role in the folding of proteins and
[
14]
[15]
[16]
otics, saccharides and ions. Also, receptor-crosslinked was identified as an endogenous excitotoxin that damages
Au NPs were used for the electrochemical detection of ana- the nervous system. Also, elevated amounts of L-cysteine
lytes through the formation of donor-acceptor complexes.
Recently, we have implemented the iodide (I )-induced cata- as Parkinson's or Alzheimer’s diseases. Indeed, research
lyzed oxidation of L-cysteine to cystine by H O as a means efforts were directed in recent years toward the develop-
to control the L-cysteine-mediated aggregation of Au NPs. ment of sensors for L-cysteine such as electrochemical
This aggregation process was used to develop a glucose and optical sensors. The general platform for the hemin/G-
sensor and a sensor that probes inhibitors for acetylcholine quadruplex-inhibited L-cysteine-induced aggregation of
[
34]
[
17]
were detected in different neurodegenerative diseases such
−
[35]
[36]
2
2
[
37]
[
38]
[
18]
esterase.
Au NPs is depicted in Figure 1(A). Upon the addition of
The hemin/G-quadruplex horseradish peroxidase-mim- L-cysteine, (1), to the system, the L-cysteine binds to Au NPs
icking DNAzyme[19] has attracted substantial interest in via the thiol functional group, and the resulting cooperative
the past decade, and its catalytic activity to oxidize various H-bonds between the amino acid residue associated with the
substrates in the presence of H O2 have been extensively Au NPs lead to the aggregation of the Au NPs, accompanied
2
used to develop numerous optical or electrical sensing plat- with a red absorbance shift of the NPs, Figure 1(A), path (a).
[
20]
forms,
tions.
and to drive different biocatalytic transforma- However, in the presence of the hemin/G-quadruplex, the
The hemin/G-quadruplex was found to mimic the L-cysteine, (1), is oxidized to cystine, (2), thus inhibiting the
[
21]
functions of horseradish peroxidase by catalyzing the oxi- aggregation of the Au NPs, Figure 1(A), path (b). Figure 1(B),
dation of 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulphonic curve (a) and curve (b), show the absorption spectra of the
2
−
.−
[19a]
acid), ABTS , to the colored product of ABTS by H O ,
Au NPs in the absence and in the presence of L-cysteine
2
2
and to generate chemiluminescence in the presence of H O
respectively. In the absence of L-cysteine the resulting spec-
2
2
[
22]
and luminol.
develop optical sensors for DNA,
These processes were extensively used to trum presents a characteristic band at λmax = 520 nm that is
[
23]
aptamer-substrate associated with individual Au NPs, Figure 1(B), curve (a),
[
24]
[25]
and for enzyme assays.[26] Also, the while in the presence of L-cysteine, an aggregation is evident
complexes,
ions
DNAzyme-catalyzed generation of chemiluminescence was by the broad red-shifted absorbance band at λmax = 650 nm,
used to develop, in the presence of different sized semicon- corresponding to the interparticle-coupled plasmon exciton,
ductor quantum dots, multiplexed analysis assays for DNA Figure 1(B), curve (b). Upon reacting the hemin/G-quadru-
[
27]
[28]
analytes,
or for aptamer-substrate complexes,
using plex DNAzyme with L-cysteine for 45 minutes, Figure 1(B),
the chemiluminescence resonance energy transfer (CRET) curve (c), only the band at λmax = 520 nm appeared, implying
process as a sensing mechanism. The hemin/G-quadruplex that, in the presence of the DNAzyme, the aggregation of
was, additionally, implemented as a quencher of the lumi- the Au NPs is prohibited. These results imply that only in
nescence of QDs via an electron-transfer mechanism for the the presence of the complex formed between hemin and
[
29]
sensing of DNA analytes and aptamers.
The hemin/G- the G-quadruplex sequence, (3), the aggregation of the NPs
quadruplex acted, also, as a catalyst for the development of is eliminated. The mechanism whereby the hemin/G-quad-
electronic sensors. The electrocatalyzed reduction of H O2 ruplex retards the aggregation of the Au NPs may be for-
2
was used to construct amplified electrochemical DNA sen- mulated based on previous reports: (i) While L-cysteine is
[
30]
sors and aptasensors. Alternatively, in the presence of CdS known to aggregate Au NPs, the oxidized disulfide, cystine,
quantum dots/hemin/G-quadruplex hybrid systems associ- is inactive in stimulating the aggregation of the Au NPs on
ated with electrodes and H O , the CRET process stimulated the time-scale of the L-cysteine-stimulated aggregation of
2
2
[
18]
the generation of photocurrents, and the platform was used the NPs.
(ii) The hemin/G-quadruplex acts as catalyst for
[
39]
to develop a photoelectrochemical sensor for a DNA target the aerobic oxidation of thiols to disulfides.
and to monitor the biocatalytic activity of glucose oxidase.
The mecha-
nism leading to the hemin/G-quadruplex-catalyzed aerobic
Additionally, numerous chemical reactions were found to be oxidation of L-cysteine to cystine yields H O , and this acts
[
31]
2
2
catalyzed by the hemin/G-quadruplex DNAzyme. Examples as the oxidant for the catalyzed-formation of the disulfide
include the DNAzyme-catalyzed oxidation of NADH by O2 as depicted in Equation (1) to Equation (10). Indeed, the
2
www.small-journal.com
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2014,
DOI: 10.1002/smll.201400002

DNA Sensors and Aptasensors Based on the Hemin/G-quadruplex-Controlled Aggregation of Au NPs in the Presence of L-Cysteine
addition of catalase, 10U, to the system consisting of the Au
NPs, L-cysteine and the hemin/G-quadruplex, leads to the
partial aggregation of the Au NPs, Figure 1(B), curve (d).
That is, the catalase-induced decomposition of the thiol-
mediated generation of H O competes with the oxidation
2
2
of the thiol to the disulfide, thus favoring the L-cysteine-
stimulated aggregation of the NPs. TEM images confirmed
the L-cysteine-mediated aggregation of the NPs and the
elimination of the aggregation process, in the presence of
the hemin/G-quadruplex. While the addition of L-cysteine
to the Au NPs leads to highly aggregated NPs structures
after a fixed time-interval of 45 minutes, Figure 1(C), panel
I, the Au NPs system that included the hemin/G-quadruplex
DNAzyme shows, after interacting with L-cysteine for a fixed
time-interval of 45 minutes, that only individual Au NPs are
present, Figure 1(C), panel II.
2
4
RSH + O2 → RSSR + H2O2
(1)
(2)
RSH + O2 → 2RSSR + H2O
Hem
(
Fe+
)
/ GQ + H2O2 → Hem
(
Fe+4 = O •
)
+
/ GQ + H2O
3
(
3)
Hem(Fe+ = O) /GQ + RSH → Hem(Fe = O) /GQ + RS + H
4
+
+4
+
•
+
(
4)
+
Fe+ = O
4
/ GQ + RSH → Hem
Fe+3
/ GQ + RS• + H+
Hem
(
)
(
)
(
5)
RS + RSH ↔ RSSR + H+
•
•−
(6)
(7)
RS + O2 → RSOO•
•
•
•
RS + RS → RSSR
(8)
RSSR• + O2 → RSSR + O2
−
•−
(9)
•
−
•−
+
O
+ O2 + 2H → H2O2 + O2
2
(10)
The hemin/G-quadruplex-controlled aggregation of Au
NPs, in the presence of L-cysteine, allows the development of
new sensing platforms. It was reported that hairpin structures
that include the G-quadruplex sequence in a “caged” inactive
configuration and that also encode recognition sequences
either for the detection of DNA analytes, or for the forma-
tion of aptamer-substrate complexes, may act as functional
structures for sensing. The opening of the hairpins by the
hybridization of the target DNA, or upon the formation of
the aptamer-substrate complexes, uncages the G-quadru-
plex, and in the presence of hemin, it self-assembles into the
hemin/G-quadruplex DNAzyme that acts as catalytic label
for the amplification of the sensing events. Indeed, many dif-
Figure 1. (A) Hemin/G-quadruplex-controlled aggregation of Au NPs in the
presenceofL-cysteineasacrosslinker.(B)Absorptionspectracorresponding
to: (a) Au NPs with no added L-cysteine (b) Au NPs with added L-cysteine, ferent G-quadruplex nanostructures, e.g., telomeres, aptamers
−
5
−5
1
× 10 M. (c) Au NPs with added L-cysteine, 1 × 10 M, and the hemin/G- against ATP or VEGF, are known. Thus, the formation of the
6
−
−5
quadruplex, 1 × 10 M. (d) Au NPs with added L-cysteine, 1 × 10 M,
hemin/G-quadruplex, 1 × 10 M and catalase, 10 units. The spectra
were recorded after a fixed time-interval of 45 minutes. (C) TEM images
corresponding to: (I) The L-cysteine, 1 × 10 M, stimulated aggregated Au
NPs. (II) The Au NPs in the presence of L-cysteine, 1 × 10 , and added
hemin/G-quadruplex, 1 × 10 M. TEM images were recorded after allowing
the Au NPs to react with L-cysteine for a fixed time-interval of 10 minutes.
hemin/G-quadruplex DNAzyme structures may, then, be used
as catalytic labels for the sensing events through the con-
trolled aggregation of Au NPs in the presence of L-cysteine.
This is exemplified here with the sensing of a DNA analyte
and the analysis of thrombin by an aptamer sensing platform.
Figure 2(A) shows the layout for the analysis of a DNA ana-
−6
−5
−5
−6
small 2014,
DOI: 10.1002/smll.201400002
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com 3

full papers
A. Niazov-Elkan et al.
Figure 2. (A) The analysis of an analyte DNA that opens a functional hairpin probe to yield the hemin/G-quadruplex structure through the controlled
−5
L-cysteine-mediated aggregation of Au NPs. (B) Absorption spectra corresponding to: (a) The Au NPs in the presence of L-cysteine, 1 × 10 M,
−
6
−7
−7
−5
the hairpin probe, (4), 1 × 10 M, the analyte, (5), 1 × 10 M, and hemin, 7 × 10 M (b) The Au NPs in the presence of L-cysteine, 1 × 10 M,
−6
−6
the hairpin probe (4), 1 × 10 M, hemin 1 × 10 M, but in the absence of the analyte. (C) Absorption spectra corresponding to the analysis of
−
9
−8
the variable concentrations of the target DNA, (5), according to the scheme outlined in Figure 2(A): (a) 0 M, (b) 1 × 10 M, (c) 1 × 10 M, (d)
× 10 M, (e) 1 × 10 M, (f) 5 × 10 M, (g) 1 × 10 M. Spectra were recorded after a fixed time-interval of 45 minutes. (D) Calibration curve
−
8
−7
−7
−6
5
corresponding to the absorbance ratio A520/A620 for different concentrations of the target. Error bars correspond to N = 3 experiments. (E) TEM
images corresponding to: (I) The aggregated Au NPs generated by the L-cysteine-stimulated crosslinking process in the absence of DNA target. (II)
−7
The non-aggregated Au NPs in the presence of the DNA analyte, 1 × 10 M. For all experiments the Au NPs samples were allowed to interact for
0 minutes prior to the recording of the TEM images.
1
lyte by controlling the L-cysteine-mediated aggregation of Au the absence of the DNA analyte, the L-cysteine-stimulated
NPs, in the presence of the hemin/G-quadruplex. The hairpin aggregation process of the Au NPs proceeds, Figure 2(A),
structure, (4), includes in domain I the recognition sequence path (a). However, in the presence of the analyte, (5), the
complementary to the DNA analyte, (5), and in domain II the hairpin opens, and in the presence of hemin, this results in the
G-quadruplex sequence in a caged, inactive, configuration. In self-assembly of the hemin/G-quadruplex DNAzyme. Subse-
4
www.small-journal.com
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2014,
DOI: 10.1002/smll.201400002

DNA Sensors and Aptasensors Based on the Hemin/G-quadruplex-Controlled Aggregation of Au NPs in the Presence of L-Cysteine
quently, the self-assembled DNAzyme catalyzes the oxida- results imply that the thrombin/hemin/TBA complex act as
tion of L-cysteine to cystine, thus prohibiting the aggregation a catalyst that prohibits the aggregation of the NPs, through
of the NPs, Figure 2(A), path (b). Indeed, the resulting spec- the oxidation of L-cysteine to cystine. As the content of the
trum of the system in the presence of the DNA analyte, (5), DNAzyme is controlled by the concentration of thrombin,
−
7
1
× 10 M, presents an absorbance band at λmax = 520 nm, the degree of aggregation of the NPs relates to the concentra-
implying that the aggregation of the NPs is prohibited, tion of thrombin. Figure 3(C) shows the absorption spectra of
Figure 2(B), curve (a). On the contrary, in the absence of the the L-cysteine/Au NPs/hemin/TBA system upon interacting
analyte, the aggregation of the Au NPs occurs as can be seen with different concentrations of thrombin after a fixed time-
by the appearance of an absorbance band at λmax = 650 nm, interval of 45 minutes. Although in the absence of thrombin
Figure 2(B), curve (b). These experiments confirm that the an effective aggregation of the NPs is observed, curve (a), as
formation of the hemin/G-quadruplex is essential to pro- the concentration of thrombin increases, the degree of aggre-
hibit the aggregation of the NPs, and that the opening of the gation decreases, curves (b) – (f), and at a thrombin concen-
−
8
hairpin, (4), by the target, (5), followed by the self-assembly tration of 5 × 10 M no aggregated NPs are observed and
of the DNAzyme, in the presence of hemin, prevents an effec- only the absorbance band at λmax = 520 nm characteristic to
tive aggregation. As the degree of opening of the hairpin, (4), the individual Au NPs is observed, curve (f). These results
is controlled by the concentration of the analyte, the extent of indicate that as the concentration of thrombin increases, the
inhibition of the aggregation process provides a quantitative content of the generated thrombin/hemin/TBA-G-quadru-
measure for the concentration of the analyte, (5). Figure 2(C) plex catalyst is higher, resulting in a more effective oxidation
depicts the absorption spectra of the Au NPs observed upon of L-cysteine to cystine, a process that eliminates the aggre-
challenging the Au NPs with L-cysteine, the hairpin structure, gation of the NPs. Thus, the degree of aggregation of the Au
(
(
4), hemin, and variable concentrations of the DNA analyte, NPs relates to the concentration of thrombin. Figure 3(D)
5). As the concentration of the analyte increases, the degree shows the resulting calibration curve corresponding to the
of aggregation decreases, consistent with the higher absorb- A520/A650 ratio after a fixed time-interval of 45 minutes. The
ance of the aggregated NPs at λmax = 650 nm and the lower system enabled the analysis of thrombin with a detection
−
10
absorbance-band of the individual Au NPs at λmax= 520 nm. limit that corresponds to 3 × 10 M. The inhibition of the
The resulting calibration curve is shown in Figure 2(D), cor- aggregation process in the presence of thrombin was con-
responding to the A520/A650 ratio after a fixed time-interval of firmed, also, by TEM images, Figure 4, panels I – III. In the
4
5 minutes. The system enabled the detection of the analyte absence of thrombin the Au NPs are assembled into large
−9
DNA with a detection limit that corresponded to 4.5 × 10 M. aggregated structures, panel I. However, as the concentration
Figure 2(E), Panels I and II, depicts the TEM images of the of thrombin increases, the size of the aggregates decreases,
−
10
Au NPs for the system consisting of L-cysteine/Au NPs/ and already in the presence of thrombin, 5 × 10 M, smaller
hemin/DNA hairpin in the absence of the target, or in the aggregates appear, panel II. Eventually, in the presence of 5
presence of target, 1 × 10− M, respectively. The exclusion of × 10 M of thrombin, panel III, almost only individual par-
the DNA target from the system leads to an effective aggre- ticles can be seen. These results indicate that the thrombin/
gation of the NPs. However, upon adding the analyte to the hemin/TBA DNAzyme catalyzes the oxidation of L-cysteine
system most of the Au NPs appear as individual NPs and, to cystine, thus prohibiting the L-cysteine-mediated aggrega-
eventually, as dimers. These results imply that the opening of tion process of the Au NPs.
7
−8
the hairpin by the analyte, in the presence of hemin, gener-
ates the biocatalytic DNAzyme that oxidizes L-cysteine to highlighted the use of the hemin/G-quadrupex-catalyzed oxi-
cystine, therefore inhibiting the aggregation process. dation of L-cysteine and the subsequent controlled aggrega-
The thrombin binding-aptamer, TBA, (6), was found to tion of the Au NPs, as a means to detect L-cysteine, thrombin
The different systems described in the present study have
[
40]
generate a G-quadruplex upon its binding to thrombin.
and DNA. The value of these sensing platforms should be
The incorporation of hemin into the TBA-thrombin com- supported by two further experiments: (i) The possibility
[
41]
plex generates an active DNAzyme.
Accordingly, the to implement the method in a real biological fluid environ-
thrombin/hemin/TBA complex was used as a catalyst for the ments. (ii) The application of the unique absorbance fea-
analysis of thrombin through the L-cysteine-mediated aggre- tures of aggregated/deaggregated Au NPs for visual sample
gation of Au NPs, Figure 3(A). In the absence of thrombin, analysis, particularly for rapid point-of-care visual analytical
Figure 3(A), path (a), the aptamer does not fold into a assays. To address these challenges we attempted to quanti-
G-quadruplex structure and does not form the DNAzyme, tatively analyze L-cysteine in urine samples. High levels of
thus resulting in the L-cysteine-induced aggregation of the L-cysteine appear in the urine of individuals suffering from
NPs. In the presence of thrombin, the aggregation of the Au cystinuria that causes the formation of kidney-stones.[42]
NPs is prohibited due to the thrombin/hemin/TBA-catalyzed The present methods to analyze L-cysteine in urine include
[
42b,43]
[44]
aerobic oxidation of L-cysteine to cystine, Figure 3(A), path HPLC analyses,
(
electrochemical sensing
and optical
[
45]
b). Figure 3(B) shows the absorption spectra of a system sensing. Thus, the visual (or quantitative optical) analysis
containing the Au NPs, hemin and TBA either in the pres- of L-cysteine may provide a point-of-care diagnostic method.
ence of thrombin, 5 × 10− M, curve (a), and in the absence Figure 5(A) shows the analysis of variable concentrations of
of thrombin, curve (b). Clearly, in the absence of thrombin L-cysteine in urine samples. In these experiments the urine
an effective aggregation of the Au NPs proceeds, while in samples were diluted 10-fold and the native L-cysteine was
the presence of thrombin the NPs do not aggregate. These consumed by the addition of hemin/G-quadruplex under O2.
8
small 2014,
DOI: 10.1002/smll.201400002
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com 5

full papers
A. Niazov-Elkan et al.
Figure 3. (A) Analysis of thrombin through the hemin/thrombin-binding aptamer (TBA) complex-controlled L-cysteine-mediated aggregation of
−
6
−8
the Au NPs. (B) Absorption spectra corresponding to: (a) The Au NPs in the presence of TBA, (6), 1 × 10 M, thrombin, 5 × 10 M, hemin, 2 ×
−
6
−5
−6
−5
10
M, and L-cysteine, 1 × 10 M. (b) The Au NPs in the presence of TBA, 1 × 10 M, L-cysteine, 1 × 10 M, yet in the absence of thrombin. (C)
Absorption spectra corresponding to the analysis of different concentration of thrombin according to the L-cysteine-mediated aggregation of the
−
10
−9
Au NPs as shown in Figure 3(A). The spectra were recorded after a fixed time-interval of 45 minutes: (a) 0 M, (b) 1 × 10 M, (c) 1 × 10 M, (d) 5
×
−
9
−8
−8
10 M, (e) 1 × 10 M, (f) 5 × 10 M. (D) Derived calibration curve for the analysis of thrombin presented by the A520/A650 values at different
concentrations of thrombin. Error bars correspond to N = 3 experiments.
Different concentrations of L-cysteine were, then, added to the absorption spectra of the resulting Au NPs aggregates.
the urine samples, allowed to interact for 45 minutes, and sub- While in the absence of L-cysteine no aggregation of the NPs
sequently, the Au NPs were added to the samples and these is observed, as the concentration of L-cysteine is elevated,
were allowed to aggregate for 10 minutes. Figure 5(A) shows the degree of aggregation of the Au NPs is increasing. Within
Figure 4. TEM images of Au NPs treated with L-cysteine, 1 × 10 M, in the presence of hemin, 1 × 10 M, TBA, 1 × 10−6 M, and of variable
−
5
−6
−10
−8
concentrations of thrombin: (I) no thrombin, (II) 5 × 10 M, (III) 5 × 10 M. For all experiments the Au NPs samples were allowed to interact for
0 minutes prior to the recording of the TEM images.
1
6
www.small-journal.com
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2014,
DOI: 10.1002/smll.201400002

DNA Sensors and Aptasensors Based on the Hemin/G-quadruplex-Controlled Aggregation of Au NPs in the Presence of L-Cysteine
easily visually mapped, Figure 5(C). While the normal level
of L-cysteine yield a red solution of the Au NPs, the samples
including a high level of L-cysteine (100 µM) generate a blue
aggregated Au NPs solution.
3
. Conclusion
The present study has introduced a new paradigm for the
development of biosensor systems by controlling the aggre-
gation of Au NPs by the hemin/G-quadruplex DNAzyme.
While L-cysteine stimulates the aggregation of Au NPs by
the binding of the sulfhydryl group of L-cysteine to the NPs
and the formation of interparticle electrostatic attraction
interactions and H-bonds, the hemin/G-quadruplex-cata-
lyzed oxidation of L-cysteine to cystine, therefore prohibiting
the aggregation process. The concentration of the hemin/G-
quadruplex was found to control the degree of aggregation
of the Au NPs, thus providing a spectroscopic readout signal
in the different sensing platforms. A DNA sensing method
was established by designing a functional nucleic acid
hairpin structure that includes the recognition sequence for
the target DNA, and also the G-quadruplex sequence in a
“
caged”, inactive, configuration. The opening of the hairpin
structure by the target DNA results in the self-assembly
of the hemin/G-quadruplex DNAzyme that controls the
degree of aggregation of the Au NPs. This method enabled
the analysis of DNA with a detection limit corresponding
−
9
to 4.5 × 10 M. By changing the recognition region of the
hairpin structure the detection of other DNA targets may be
envisaged. Also, the DNAzyme-controlled L-cysteine-medi-
ated aggregation of Au NPs was implemented to develop a
thrombin/TBA sensing system. The TBA folds upon binding
to thrombin into a G-quadruplex structure. The association
of hemin to the aptamer-thrombin complex yields a cata-
lytic hemin/G-quadruplex that controls the aggregation of
the Au NPs, a process that provides a spectroscopic readout
signal for analyzing thrombin, with a detection limit corre-
−10
sponding to 3 × 10 M. As many other aptamers generate
G-quadruplex structures upon binding to their substrates
(
e.g., VEGF or ATP), this method could be broadened to
develop other aptasensor systems. We also demonstrated
the applicability of the present system for the detection of
L-cysteine in a biological sample. The hemin/G-quadruplex-
Figure 5. (A) Absorption spectra corresponding to the analysis of the
−
5
variable concentrations of L-cysteine in urine : (a) 0 M, (b) 2 × 10 M,
−
5
−5
−5
−5
(
1
c) 3.5 × 10 M, (d) 5 × 10 M, (e) 6.5 × 10 M, (f) 8 × 10 M, (g) controlled L-cysteine-stimulated aggregation of Au NPs
× 10 M. Spectra were recorded after a fixed time-interval of allowed the detection of L-cysteine in urine samples with a
−
4
45 minutes. (B) Calibration curve corresponding to the absorbance
detection range corresponding to that of a healthy adult, thus
ratio A520/A620 for different concentrations of the target. Error bars
correspond to N = 3 experiments. (C) Visual detection of L-cysteine in
urine samples in the presence of: (a) 5 × 10−5 M L-cysteine, (b) 1 × 10−4
M L-cysteine.
providing a means for the visual, naked-eye, detection of
excess of L-cysteine in urine that may be indicative of several
diseases. The results demonstrate the versatility in the use of
hemin/G-quadruplexes in the presence of L-cysteine and Au
NPs for sensing applications, and the design of many other
the concentration range of 20 µM – 80 µM, the spectral fea- sensor configurations may be envisaged. Furthermore, dif-
tures of the system reveal the co-existence of individual and ferent G-quadruplex structures exist (parallel, anti-parallel,
aggregated NPs (A650/A520 = 0.6 – 1.0), Figure 5(B). This is assembly of subunits, variable numbers of G-quadruplex
the normal concentration range of L-cysteine in healthy layers). The systematic analysis of the catalytic activities of
[
46]
adults.
At L-cysteine concentrations >100 µM (suspected the resulting hemin/G-quadruplexes toward the oxidation of
for cystinuria) the spectral features of only aggregated parti- L-cysteine and the resulting effect on the aggregation of the
cles is observed, A650/A520 = 1.4. These special features can be Au NPs is desirable.
small 2014,
DOI: 10.1002/smll.201400002
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com 7

full papers
A. Niazov-Elkan et al.
immediately recorded. The detection limit was calculated using the
4
. Experimental Section
3
σ method.
Materials and Reagents: Ultrapure water from NANOpure Dia-
Thrombin Detection: The thrombin-binding aptamer, TBA, (6),
mond (Barnstead Int., Dubuque, IA) source was used throughout
the experiments. Hemin was purchased from Frontier Scientific
and was used without further purification. Hemin stock solution
was prepared in DMSO and stored in the dark at −20 °C. L-cysteine
dihydrochloride anhydrous was purchased from Fluka. Thrombin,
from human plasma, 250 units, was purchased from Sigma.
The DNA strands were purchased from Integrated DNA Technol-
ogies Inc. (IDT). All oligonucleotides were HPLC-purified and freeze-
dried by the supplier. The oligonucleotides were used as provided
and diluted in 10 mM phosphate buffer solution, pH = 7.4, to give
stock solutions of 100 μM. The sequences of the oligomers used in
this study are as follows:
× 10−6 M, was incubated in a HEPES buffer solution, 5 mM,
1
pH = 7.2, containing KNO , 20 mM, NaNO , 200 mM, and hemin,
3
3
−6
2
× 10 M, for 3 hours at room temperature in the presence of
various concentrations of thrombin, in order to stabilize the
thrombin/hemin/TBA complex. The system was incubated, then,
with L-cysteine, 1 × 10−5 M, for 45 min at room temperature. 20 µl
−9
from the resulting system were incubated with Au NPs, 6 × 10 M,
in HEPES buffer solution, 5 mM, pH = 7.2, for 10 min, and the
resulting spectrum was immediately recorded. The detection limit
was calculated using the 3σ method.
−6
Analysis of L-cysteine in Urine: The nucleic acid, (3), 1 × 10 M,
was incubated in a HEPES buffer solution, 5 mM, pH = 7.2, con-
(
(
3) 5′– GGGTTGGGCGGGATGGG – 3′
4) 5′– GGGTTGGGCGGGATGGGTTACCTCAGTGCTTATTCGAAACC-
taining KNO , 20 mM, NaNO , 200 mM, 10% (v/v) of urine sample,
3
3
taken from a male adult volunteer and hemin, 7 × 10−7 M, for
CATC – 3′
4
5 minutes at room temperature in order to stabilize the hemin/G-
(
(
5) 5′– GGTTTCGAATAAGCACTGAGGT – 3′
6) 5′ – GGTTGGTGTGGTTGG – 3′
quadruplex complex. The system was incubated, then, with
L-cysteine, at variable concentrations for 45 min at room tempera-
ture. 20 µl from the resulting system were incubated with Au NPs,
Preparation and Functionalization of Au-NPs: The 13 nm Au
NPs were prepared using a standard citrate method.[47] Briefly, a
solution of 38 mM sodium citrate was added to a rapidly stirred
× 10−9 M, in HEPES buffer solution, 5 mM, pH = 7.2, for 10 min-
6
utes, and the resulting spectrum was immediately recorded.
boiling aqueous solution of 1 mM HAuCl . After 30 min of boiling,
4
the red mixture was allowed to cool to room temperature, and the
Au NPs were collected by filtering through a 0.45 µm membrane.
Finally, 50 mL of Au NPs were mixed with 2 mL of 1% surfactant
Tween-20 to yield well-dispersed Au NPs and were stored at 4 °C.
The concentration of the 13 nm Au NPs was determined by fol-
lowing the absorbance spectra, λmax = 520 nm, and by using the
appropriate extinction coefficient. The concentration of the Au NPs
solution was found to be 12 nM.
Acknowledgements
This work was funded by the Israel Science Foundation.
Hemin/G-quadruplex DNAzyme-Inhibited Aggregation of Au
NPs: The nucleic acid, (3), 1 × 10−6 M, was incubated in a HEPES
[1] a) R. A. Sperling, P. R. Gil, F. Zhang, M. Zanella, W. J. Parak, Chem.
Soc. Rev. 2008, 37, 1896; b) K. Saha, S. S. Agasti, C. Kim, X. Li,
V. M . R o t e l l o , Chem. Rev. 2012, 112, 2739; c) H. Jans, Q. Huo,
Chem. Soc. Rev. 2012, 41, 2849.
buffer solution, 5 mM, pH = 7.2, containing KNO , 20 mM, NaNO ,
3
3
−7
2
00 mM, and hemin, 7 × 10 M, for 30 minutes at room tem-
perature in order to stabilize the hemin/G-quadruplex complex.
The system was incubated, then, with L-cysteine, (1), 1 × 10−
M, for 45 min at room temperature (This time-interval represents
the optimized time-interval of interaction between the hemin/G-
quadruplex and L-cysteine. At lower interaction time the difference
between the L-cysteine-containing system and the control experi-
ments was too low. At longer interaction time-intervals, the signal
of the control system is high and close to the signal generated by
the hemin/G-quadruplex, due to the hemin-catalyzed oxidation of
[
2] a) K. H. Su, Q. H. Wei, X. Zhang, J. J. Mock, D. R. Smith, S. Schultz,
Nano Lett. 2003, 3, 1087; b) S. K. Ghosh, T. Pal, Chem. Rev. 2007,
107, 4797.
3] A. K. Boal, F. Ilhan, J. E. DeRouchey, T. Thurn-Albrecht, T. P. Russell,
V. M . R o t e l l o , Nature 2000, 404, 746–748.
5
[
[
4] a) Y. W. Lin, C. C. Huang, H. T. Chang, Analyst 2011, 136, 863;
b) S. O. Obare, R. E. Hollowell, C. J. Murphy, Langmuir 2002, 18,
1
0407.
[
5] a) J. Liu, S. Mendoza, E. Roman, M. J. Lyn, R. Xu, A. E. Kaifer, J.
Am. Chem. Soc. 1999, 121, 4304; b) S. Y. Lin, S. W. Liu, C. M. Lin,
C. H. Chen, Anal. Chem. 2002, 74, 330.
L-cysteine). 20 µl from the resulting system were incubated with
Au NPs, 6 × 10−9 M, in HEPES buffer solution, 5 mM, pH = 7.2, for
[6] a) D. Balogh, Z. Zhang, A. Cecconello, J. Vavra, L. Severa, F. Teply,
I. Willner, Nano Lett. 2012, 12, 5835; b) Y. Jiang, H. Zhao, N. Zhu,
Y. Lin, P. Yu, L. Mao, Angew. Chem. 2008, 120, 8729; c) Y. Jiang,
H. Zhao, N. Zhu, Y. Lin, P. Yu, L. Mao, Angew. Chem., Int. Ed. 2008,
47, 8601.
[7] a) W. Shenton, S. A. Davis, S. Mann, Adv. Mater. 1999, 11, 449;
b) H. Otsuka, Y. Akiyama, Y. Nagasaki, K. Kataoka, J. Am. Chem.
Soc. 2001, 123, 8226.
10 minutes, and the resulting spectrum was immediately recorded.
DNA Target Sensing: HEPES buffer solution, 5 mM, pH = 7.2,
−
6
containing the nucleic acid hairpin, (4), 1 × 10 M, KNO , 20 mM
3
and NaNO , 200 mM, was first annealed at 90° C for 5 minutes and
3
then allowed to cool-down to room temperature for additional 30
minutes. The solution was, then, added with the DNA analyte, (5),
with various concentrations for additional 90 minutes. Hemin, 7 ×
[8] H. Li, L. Rothberg, Proc. Natl. Acad. Sci. U. S. A. 2004, 101, 14036.
−
7
[9] N. L. Rosi, C. A. Mirkin, Chem. Rev. 2005, 105, 1547.
10] a) W. Zhao, W. Chiuman, M. A. Brook, Y. Li, ChemBioChem. 2007,
10
M, was subsequently added for 30 minutes at room tempera-
[
−5
ture. The system was incubated, then, with L-cysteine, 1 × 10 M,
8
, 727; b) J. Wang, L. Wang, X. Liu, Z. Liang, S. Song, W. Li, G. Li,
for 45 min at room temperature. 20 µl from the resulting system
C. Fan, Adv. Mater. 2007, 19, 3943.
were incubated with Au NPs, 6 × 10− M, in HEPES buffer solu- [11] a) J. S. Lee, M. S. Han, C. A. Mirkin, Angew. Chem. 2007, 119,
9
tion, 5 mM, pH = 7.2, for 10 min, and the resulting spectrum was
4171; b) J. S. Lee, M. S. Han, C. A. Mirkin, Angew. Chem., Int. Ed.
8
www.small-journal.com
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2014,
DOI: 10.1002/smll.201400002

DNA Sensors and Aptasensors Based on the Hemin/G-quadruplex-Controlled Aggregation of Au NPs in the Presence of L-Cysteine
2
2
007, 46, 4093; c) D. Li, A. Wieckowska, I. Willner, Angew. Chem. [29] E. Sharon, R. Freeman, I. Willner, Anal. Chem. 2010, 82, 7073.
008, 120, 3991; d) D. Li, A. Wieckowska, I. Willner, Angew. [30] G. Pelossof, R. Tel-Vered, J. Elbaz, I. Willner, Anal. Chem. 2010,
Chem., Int. Ed. 2008, 47, 3927.
82, 4396.
[
[
12] J. Liu, Y. A. Lu, J. Am. Chem. Soc. 2003, 125, 6642.
13] a) M. Riskin, R. Tel-Vered, O. Lioubashevski, I. Willner, J. Am.
[31] E. Golub, A. Niazov, R. Freeman, M. Zatsepin, I. Willner, J. Phys.
Chem. C 2012, 116, 13827.
Chem. Soc. 2009, 131, 7368; b) M. Riskin, R. Tel-Vered, I. Willner, [32] Z.-G. Wang, P. Zhan, B. Ding, ACS Nano 2013, 7, 1591.
Adv. Mater. 2010, 22, 1387.
14] M. Frasconi, R. Tel-Vered, M. Riskin, I. Willner, Anal. Chem. 2010,
[33] L. C. H. Poon, S. P Methot, W. Morabi-Pzaooki, F. Pio, A. J. Bennet,
D. Sen, J. Am. Chem. Soc. 2011, 133, 1877.
[
8
2, 2512.
[34] R. Janaky, V. Varga, A. Hermann, P. Saransaari, S. S. Oja, Neuro-
chem. Res. 2000, 25, 1397.
[
[
15] Y. Ben-Amram, M. Riskin, I. Willner, Analyst 2010135, 2952.
16] Y. Ben-Amram, R. Tel-Vered, M. Riskin, Z.-G. Wang, I. Willner, [35] M. T. Heafield, S. Fearn, G. B. Steventon, R. H. Waring,
Chem. Sci. 20123, 162. A. C. Williams, S. G. Sturman, Neurosci. Lett. 1990, 110, 216.
17] A. N. Shipway, M. Lahav, R. Blonder, I. Willner, Chem. Mater. [36] T. L. Perry, M. G. Norman, V. W. Young, S. Whiting, J. U. Crichton,
[
1
999, 11, 13.
S. Hansen, S. Kish, J. Ann. Neurol. 1985, 18, 482.
[37] X. Chen, Y. Zhou, X. Peng, J. Yoon, Chem. Soc. Rev. 2010, 39,
2120.
[
[
18] F. Wang, X. Liu, C.-H. Lu, I. Willner, ACS Nano 2013, 7, 7278.
19] a) P. Travascio, Y. F. Li, D. Sen, Chem. Biol. 1998, 5, 505;
b) P. K. Witting, P. Travascio, D. Sen, A. G. Mauk, Inorg. Chem. [38] T. Inoue, J. R. Kirchhoff, Anal. Chem. 2000, 72, 5755.
001, 40, 5017; c) D. Sen, L. C. H. Poon, Crit. Rev. Biochem. Mol. [39] E. Golub, R. Freeman, I. Willner, Anal. Chem. 2013, 85, 12126.
2
Biol. 2011, 46, 478.
[40] L. C. Bock, L. C. Griffin, J. A. Latham, E. H. Vermaas, J. J. Toole,
[
[
20] a) I. Willner, B. Shlyahovsky, M. Zayats, B. Willner, Chem. Soc.
Nature 1992, 355, 564.
Rev. 2008, 37, 1153; b) J. Kosman, B. Juskowiak, Anal. Chim. Acta [41] T. Li, E. Wang, S. Dong, Chem. Commun. 2008, 31, 3654.
2
011, 707, 7.
21] a) E. Golub, R. Freeman, I. Willner, Angew. Chem., Int. Ed. 2011,
23, 11914; b) E. Golub, R. Freeman, I. Willner, Angew. Chem.,
[42] a) J. Chrastil, Analyst 1990, 115, 1383; b) A. Pastore, R. Massoud,
C. Motti, A. Lo Russo, G. Fucci, C. Cortese, G. Federici, Clin. Chem.
1998, 44, 825; c) S. Segal, S. O. Their, in The Metabolic and
Molecular Bases of Inherited Disease, 7th ed. (Eds: C. R. Scriver,
A. L. Beaudet, W. S. Sly, D. Valle) McGraw-Hill, New York, 1995,
pp. 3581–3601.
1
Int. Ed. 2011, 50, 11710.
[
[
22] a) T. Niazov, V. Pavlov, Y. Xiao, R. Gill, I. Willner, Nano Lett. 2004,
4
, 1683; b) T. Li, E. Wang, S. Dong, Anal. Chem. 2010, 82, 1515.
23] a) Y. Xiao, V. Pavlov, T. Niazov, A. Dishon, M. Kotler, I. Willner, [43] B. Seiwert, U. Karst, Anal. Chem. 2007, 79, 7131.
J. Am. Chem. Soc. 2004, 126, 7430; b) S. Shimron, F. Wang, [44] C. Zhao, J. Zhang, J. Song, Anal. Biochem. 2001, 297, 170.
R. Orbach, I. Willner, Anal. Chem. 2012, 84, 1042.
[45] KH .. eL ung , H .Z . He .P, Y. Ma , D. S. H. C ah n , C .H . eL ung , D. L. Ma ,
[
[
24] C. Teller, S. Shimron, I. Willner, Anal. Chem. 2009, 81, 9114.
Chem. Commun. 2013, 49, 771.
25] a) T. Li, L. L. Shi, E. K. Wang, S. J. Dong, Chem. Eur. J. 2009, 15, [46] M. P. Brigham, W. H. Stein, S. J. Moore, Clin. Invest. 1960, 39,
347; b) T. Li, E. Wang, S. Dong, Anal. Chem. 2010, 82, 1515. 1633.
26] a) B. Shlyahovsky, D. Li, E. Katz, I. Willner, Biosens. Bioelectron. [47] J. Turkevich, P. C. Stevenson, J. A. Hillier, Discuss. Faraday Soc.
3
[
2
007, 22, 2570; b) L. Stefan, F. Denat, D. Monchaud, J. Am. Chem.
1951, 11, 55.
Soc. 2011, 133, 20405.
[
27] R. Freeman, X. Liu, I. Willner, J. Am. Chem. Soc. 2011, 133,
Received: January 1, 2014
Revised: February 20, 2014
Published online:
1
1597.
[
28] X. Liu, R. Freeman, E. Golub, I. Willner, ACS Nano 2011, 5, 7648.
small 2014,
DOI: 10.1002/smll.201400002
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com 9