Article pubs.acs.org/ac
Autonomous Replication of Nucleic Acids by Polymerization/Nicking Enzyme/DNAzyme Cascades for the Amplified Detection of DNA and the Aptamer−Cocaine Complex Fuan Wang, Lina Freage, Ron Orbach, and Itamar Willner* Institute of Chemistry, The Center for Nanoscience and Nanotechnology, The Hebrew University of Jerusalem, Jerusalem 91904, Israel S Supporting Information *
ABSTRACT: The progressive development of amplified DNA sensors and aptasensors using replication/nicking enzymes/DNAzyme machineries is described. The sensing platforms are based on the tailoring of a DNA template on which the recognition of the target DNA or the formation of the aptamer−substrate complex trigger on the autonomous isothermal replication/nicking processes and the displacement of a Mg2+-dependent DNAzyme that catalyzes the generation of a fluorophore-labeled nucleic acid acting as readout signal for the analyses. Three different DNA sensing configurations are described, where in the ultimate configuration the target sequence is incorporated into a nucleic acid blocker structure associated with the sensing template. The target-triggered isothermal autonomous replication/nicking process on the modified template results in the formation of the Mg2+-dependent DNAzyme tethered to a free strand consisting of the target sequence. This activates additional template units for the nucleic acid self-replication process, resulting in the ultrasensitive detection of the target DNA (detection limit 1 aM). Similarly, amplified aptamer-based sensing platforms for cocaine are developed along these concepts. The modification of the cocaine-detection template by the addition of a nucleic acid sequence that enables the autonomous secondary coupled activation of a polymerization/nicking machinery and DNAzyme generation path leads to an improved analysis of cocaine (detection limit 10 nM).
T
of the recognition probe−analyte duplex and the regeneration of the target analyte proved as an effective means for the amplified analysis of DNA.15 Also, the analyte-induced rolling circle amplification (RCA) process and the polymerasemediated synthesis of DNAzyme wires were reported as amplification routes for DNA or RNA analysis.16 Similar to the amplification paths for analyzing DNA, analogous systems for the amplified detection of aptamer− substrate complexes were developed. Besides a variety of electrochemical17 or optical18 aptasensors, different amplified aptamer-based sensing platforms were developed. These included the application of enzymes19 or DNAzymes20 as amplifying labels, the enzyme-free hybridization chain reaction,21 the enzyme-aided regeneration of the target analyte22 or the aptamer substrate-stimulated rolling circle amplification synthesis of DNAzyme wires. An important route for the amplified detection of DNA or of aptamer−substrate complexes has involved the application of coupled polymerization/nicking enzyme processes that release
he amplified detection of DNA has attracted substantial research efforts, and numerous electrical1 or optical2 sensing platforms were developed. These included the use of enzymes3 or metal nanoparticles4 as electrocatalytic labels or the use of metal nanoparticles as plasmonic amplifiers.5 Catalytic nucleic acids (DNAzymes or ribozymes)6 were also implemented as catalytic labels for the amplified colorimetric,7 chemiluminescence,8 or electrochemical9 detection of DNA, and for the development of logic gates and logic gate cascades.10 The isothermal autonomous synthesis of DNAzymes or DNAzyme chains as a result of the DNA recognition events is a subject of intensive research efforts recently. Such autonomous synthesis of the DNAzymes may proceed in the absence or presence of coadded DNAzymes. For example, the hybridization chain reaction process11 that involves the crossopening of two tailored DNA hairpins and the formation of DNAzyme wires (Mg2+-dependent DNAzyme12 or hemin/Gquadruplex DNAzyme13) was reported for the amplified detection of DNA. Also, the target-induced catabolic replication of DNAzyme units has been reported as a means to amplify DNA analysis.14 The enzyme-aided isothermal, amplified detection of DNA has been reported using several strategies. The use of exonucleases (Exo III) as biocatalyst for the cleavage © 2013 American Chemical Society
Received: April 24, 2013 Accepted: July 25, 2013 Published: July 25, 2013 8196
dx.doi.org/10.1021/ac4013094 | Anal. Chem. 2013, 85, 8196−8203
Analytical Chemistry
Article
a DNAzyme as catalytic label acting as reporter for the recognition events.23 According to this method, a template consisting of a recognition sequence for the analyte DNA or an aptamer sequence for the substrate is conjugated to an encoded sequence that, upon replication, yields a nicking site and the DNAzyme sequence. Thus, the formation of the DNA recognition complex or the aptamer−substrate complex initiates, in the presence of polymerase/dNTPs, the replication of the template. The nicking of the replicated strand releases the replicated DNAzyme that provides the catalytic reporting units for the sensing events. Although this method represents an autonomous isothermal amplification route for analyzing DNA or aptamer−substrate complexes, it suffers from an intrinsic limitation. The replicated DNAzyme sequence is complementary to the template units. Thus, the relatively high concentration of the template adversely affects the resulting DNAzyme activity due to partial deactivation of the DNAzyme sequence by its hybridization to free template units. These analytical barriers limit the sensitivity of the analytical platform and results in relatively long analysis times. Realizing these difficulties, we present here a systematic study of the one-pot systems to improve the amplified detection of DNA and aptamer−substrate complexes by the polymerization/nicking/ DNAzyme cascade by introducing two new elements: (i) We introduce a simple strategy to prohibit the deactivation of the released DNAzyme by its hybridization to the template. (ii) We further introduce a positive feedback method to replicate the target sequence, thereby enhancing the polymerization/ nicking/DNAzyme cascade.
■
Figure 1. (A) Amplified analysis of a target DNA through a polymerization/nicking machinery that synthesizes the Mg2+-dependent DNAzyme. The cleavage of substrate (4) yields the fluorophorelabeled nucleic acid that provides the readout signal. (B) Timedependent fluorescence changes at λem = 605 nm upon analyzing different concentrations of the target (2) according to the scheme outlined in part A. (C) Fluorescence spectra recorded upon analyzing different concentrations of the target (2) according to part A for a fixed time-interval of 12 h. The concentration of (2) for the systems presented in parts B and C correspond to (a) 0 M, (b) 1 × 10−12 M, (c) 1 × 10−11 M, (d) 1 × 10−10 M, (e) 1 × 10−9 M, (f) 1 × 10−8 M, (g) 1 × 10−7 M, and (h) 1 × 10−6 M. (Fluorescence spectra were reproduced with an error of ±4% in a set of N = 5 experiments).
EXPERIMENTAL SECTION
Materials. Klenow fragment (3→5 exo-), dNTPs, and Nt.BbvCI nicking enzyme were purchased from New England Biolabs Inc. (Beverly, MA). All DNA oligonucleotides were purchased from Integrated DNA Technologies Inc. (Coralville, IA). Table S1 in the Supporting Information shows the sequences of the oligonucleotides used in the study. The oligonucleotides were HPLC-purified and dissolved in phosphate buffer (10 mM, pH = 7.0) to yield stock solutions of 100 μM. Ultrapure water from a NANOpure Diamond (Barnstead) source was used in all of the experiments. Detection of DNA or Cocaine by Using the Polymerization/Nicking/DNAzyme Cascade System. All the assays were prepared in 1× NEBuffer 2 buffer. The working solution included 0.5 μM of DNA template, 1 μM of the DNAzyme substrate, and for the detection of cocaine, 1 μM of the aptamer blocker. The solution was annealed by warming the solution to 95 °C for 5 min and subsequently cooling to 30 °C for 2 h. Next, dNTPs (0.5 mM), MgCl2 (20 mM), Klenow fragment (0.2 U/μL), Nt.BbvCI (0.2 U/μL), and different concentrations of the target DNA or cocaine were added. The mixture was transferred to a cuvette, and the time-dependent fluorescence changes were monitored spectroscopically at 30 °C. The ROX dye was excited at 580 nm, and the fluorescence emission spectra were recorded in the region 595−645 nm.
to yield the replicated strand that includes the regions I′, II′, and III′. The domain II′ includes the nicking site of Nt.BbvCI, and the region III′ consists of the Mg2+-dependent DNAzyme sequence. The Nt.BbvCI nicking enzyme cleaves domain II′, resulting in an opening for the autonomous polymerase/ dNTPs rereplication of the template with the concomitant displacement of the replicated product strand (3). The displaced strand (3) corresponds to the Mg2+-dependent DNAzyme, and the assembly of the DNAzyme cleaves the fluorophore/quencher-functionalized ribonucleobase-containing substrate (4) to yield the fluorophore-functionalized nucleic acid fragment (5). The resulting fluorescence provides the readout signal for the analysis of the target DNA (2). Figure 1B depicts the time-dependent fluorescence changes at λem = 605 nm upon analyzing different concentrations of the target DNA (2). The time interval for the onset of the fluorescence is longer as the concentration of the target is lower. Also, in order to analyze low concentrations of the target, long analysis times are required. For example, to detect the target DNA (2) at a concentration of 1 × 10−12 M with a S/N = 3, a time-interval of 12 h is required. Figure 1C depicts the fluorescence spectra of the system upon analyzing different concentrations of the target DNA (2) for a fixed time-interval of 12 h. Evidently, the fluorescence is intensified as the concentration of (2) is
■
RESULTS AND DISCUSSION Figure 1A presents the basic sensing platform that includes the DNA template (1) consisting of domains I, II, and III. Hybridization of the target DNA (2), a TP53 gene-specific DNA sequence, to the recognition domain I triggers, in the presence of polymerase/dNTPs, the replication of the template 8197
dx.doi.org/10.1021/ac4013094 | Anal. Chem. 2013, 85, 8196−8203
Analytical Chemistry
Article
recognition domain I of the template triggers on, in the presence of polymerase/dNTPs and the nicking enzyme (Nt.BbvCI), the autonomous isothermal replication/nicking process and the displacement of (3). The released strand (3) generates the Mg2+-dependent DNAzyme structure. This structure cleaves the DNAzyme substrate (4) and yields the fluorophore-labeled DNA fragment (5) that provides the readout signal. Note, however, that the blocked hybrid (1)/ (3) duplex structure prohibits, upon the analyte-initiated isothermal autonomous replication/nicking reaction, the destructive hybridization of the displaced strand (3) with the template strand (1), thus fully exposing the synthesized DNAzyme for the desired DNAzyme cleavage of the substrate (4) and releasing an increased content of active DNAzyme units for generating the fluorophore-labeled reporter units. Figure 2B shows the time-dependent fluorescence changes at λem = 605 nm upon analyzing different concentrations of (2) by the amplified hybrid (1)/(3) duplex template system. Figure 2C shows the fluorescence spectra generated by the blocked hybrid (1)/(3) template system upon analyzing variable concentrations of the target (2) for a fixed time-interval of 12 h. Two major advantages of this system are obvious: (i) The onset fluorescence is observed at substantially lower time intervals. For example, for analyzing the analyte (2) at a concentration of 1 nM the onset fluorescence is observed after ∼4 h, whereas the onset fluorescence in the unblocked template system, as shown in Figure 1A, is observed only after ∼7 h. (ii) The fluorescence intensities generated by the system are substantially higher than that generated by the unblocked template system. For example, the blocked hybrid (1)/(3) duplex template system shows, after a time-interval of 12 h, a ∼10-fold higher fluorescence intensity for the analysis of 10 pM target (2) as compared to the unblocked template system as shown in Figure 1A. These advantages are reflected with an improved sensitivity of the blocked (1)/(3) hybrid template system that enables the detection of (2) at a concentration of 10 fM (S/N = 3) as compared to 1 pM for the unblocked template system. These results are consistent with the original strategy that blocking of the template (1) with strand (3) prohibits the destructive inhibition of the synthesized and displaced product DNAzyme sequence that is, now, fully available for the catalytic transduction process. The further improvement of the analytical platform is depicted in Figure 3A. The elements that advance this polymerization/nicking/DNAzyme cascade machinery system include: (i) The template is blocked to prohibit the hybridization of the synthesized and replicated DNAzyme units with the template. (ii) The template and the blocking units are further modified to allow the catalytic replication of the target sequence. The modified template (6) includes the domains I-II-III-I, and the modified blocker unit (7) is composed of the sequences III′-I′. The directionality of the two strands and the number of base pairs ensure the formation of the stable DNA duplex (6)/(7). The cooperative synergetic stabilization of the duplex domains (III)(III′) and (I/I′) prohibits not only the activity of the DNAzyme, but also prevents the undesired binding of the sequence I′ of the blocker unit (7) to the recognition sequence of the template and the initiation of a nontarget polymerization/nicking process (that could lead to a high background signal). The analyte (2) triggers the polymerase/dNTPs replication of the template (6) and the concomitant displacement of (7). Thus, the analyte activates the isothermal autonomous polymerization
elevated. The obvious limitation of the system is the fact that the displaced replicated product strand (3) is complementary to the free templates units, and thus, the formation of the active DNAzyme reporter units is prohibited, especially at a lower concentration of the analyte (2) (for details see Figure S1 in the Supporting Information). A further limitation of the system involves the fact that only the templates that hybridized with the analyte are involved in the polymerase/nicking process, and thus, at low concentrations of the analyte most of the template units are inactive. To overcome the first limitation, a simple approach was undertaken as shown in Figure 2A. The template (1) is now
Figure 2. (A) Amplified analysis of a target DNA by the polymerization/nicking machinery using a blocked reaction template. The DNA machinery autonomously synthesizes the Mg2+-dependent DNAzyme that catalyzes the generation of the fluorophore-labeled product (5) as readout signal. (B) Time-dependent fluorescence changes at λem = 605 nm upon analyzing different concentrations of the target (2) according to the scheme outlined in part A. (C) Fluorescence spectra recorded upon analyzing different concentrations of the target (2) according to part A for a fixed time-interval of 12 h. The concentration of (2) for the systems presented in parts B and C correspond to (a) 0 M, (b) 1 × 10−14 M, (c) 1 × 10−13 M, (d) 1 × 10−12 M, (e) 1 × 10−11 M, (f) 1 × 10−10 M, (g) 1 × 10−9 M, (h) 1 × 10−8 M, (i) 1 × 10−7 M, and (j) 1 × 10−6 M. (Fluorescence spectra were reproduced with an error of ±4% in a set of N = 5 experiments).
blocked stoichiometrically by the coadded sequence (3), and the (1)/(3) duplex provides the machinery for the autonomous replication/nicking process upon the sensing of the analyte (2). It should be noted that the added blocker sequence (3) consists of the same sequence as the displaced replicated product strand (3), and it is encoded with the Mg2+-dependent DNAzyme sequence. The activity of the DNAzyme sequence (3) is, however, prohibited as it is fully hybridized with the DNA template (1). Binding of the analyte (2) to the single-stranded 8198
dx.doi.org/10.1021/ac4013094 | Anal. Chem. 2013, 85, 8196−8203
Analytical Chemistry
Article
Figure 3. (A) Amplified analysis of a target DNA by a target-stimulated autonomous polymerization/nicking process using a tailored template being blocked by a nucleic acid strand that includes the target sequence. (B) Time-dependent fluorescence changes at λem = 605 nm upon analyzing different concentrations of the target (2) according to the scheme outlined in part A. (C) Fluorescence spectra recorded upon analyzing different concentrations of the target (2) according to (A) for a fixed time-interval of 7 h. The concentration of (2) for the systems presented in parts B and C correspond to (a) 0 M, (b) 1 × 10−18 M, (c) 1 × 10−17 M, (d) 1 × 10−16 M, (e) 1 × 10−15 M, (f) 1 × 10−14 M, (g) 1 × 10−13 M, (h) 1 × 10−12 M, (i) 1 × 10−11 M, (j) 1 × 10−10 M, (k) 1 × 10−9 M, (l) 1 × 10−8 M, (m) 1 × 10−7 M, and (n) 1 × 10−6 M. (Fluorescence spectra were reproduced with an error of ±8% in a set of N = 5 experiments).
observed at a significantly shorter time-interval. For example, analysis of the target DNA, 1 nM, reveals an onset fluorescence after ∼1 h, while the onset fluorescence for analyzing (2), 1 nM, by the blocked hybrid template (1)/(3) shown in Figure 2A was observed only after ∼4 h. (ii) The resulting fluorescence intensities are substantially higher. For example, upon analyzing the target (2), 1 nM, by the modified hybrid (6)/(7) duplex template for a time-interval of 6 h, a 6-fold higher fluorescence intensity is observed as compared to the fluorescence generated by the simple (1)/(3) blocked system. Thus, the use of the modified blocked hybrid (6)/(7) duplex template enables faster analysis of the target DNA (2) with enhanced sensitivity. In fact, the sensing platform outlined in Figure 3A enabled the detection of the target DNA (2) within a time-interval of 7 h with a sensitivity that corresponds to 1 × 10−18 M (S/N = 3). This unprecedented sensitivity originates from the isothermal autonomous replication/nicking machineries that yield analogues of the target analyte and the simultaneous synthesis of the Mg2+-dependent DNAzyme units. Figure S2 in the Supporting Information shows the timedependent fluorescence changes at λem = 605 nm upon analyzing the target DNA (2), 10 pM, according to the platforms outlined in Figures 1A, 2A, and 3A. The results illuminate clearly the effect of the structure of the template/ blocker machineries on the sensitivities and response times of the different sensing platforms. It should be noted that the experimental procedure for activating the DNA detection machinery, shown in Figure 3A, represents optimized conditions. Attempts to change the relative concentrations of
and nicking cycle accompanied by the continuous release of replicated product (7). The displaced strand (7) is composed, as before, of the Mg2+-dependent DNAzyme sequence, but it also includes an overhang region I′ that includes the sequence of the analyte (2). Thus, the displaced strand (7) is not only acting as catalyst for reporting the primary recognition events, but it includes also the encoded target sequence to activate independently a new round of replication/nicking machinery through its binding to “empty”, inactive hybrid (6)/(7) duplex template units. In fact, the present analyte-initiated polymerization/nicking reaction enabled the conversation of a small amount of analyte to a large number of analyte analogue units that trigger the exponential amplification machinery. The displaced strand (7)-initiated polymerization/nicking reaction on hybrid (6)/(7) duplex template is a self-replication process. Note that the template (6) includes single-stranded tethers at its 3′ end to prevent undesired replication processes on the DNA template track. The Mg2+-dependent DNAzyme unit generated autonomously by the different machinery paths catalyzes, then, the cleavage of the substrate (4), leading to the fluorescence as readout signal. Figure 3B depicts the timedependent fluorescence changes at λem = 605 nm upon analyzing different concentrations of the analyte (2) according to the scheme outlined in Figure 3A. Figure 3C shows the fluorescence spectra generated by the system upon analyzing different concentrations of the analyte (2) for a fixed timeinterval of 7 h. This analytical platform reveals important advances as compared to the simple template-blocking approach shown in Figure 2A: (i) The onset fluorescence is 8199
dx.doi.org/10.1021/ac4013094 | Anal. Chem. 2013, 85, 8196−8203
Analytical Chemistry
Article
the components (e.g., decreasing the concentration of the template (6)/(7) and increasing the concentration of the DNAzyme substrate (4)), led to an inefficient amplification machinery and low sensitivity. The reproducibility of analyzing the target DNA (2) at a concentration of 1.0 × 10−18 M was addressed. We found that in a set of N = 5 experiments, the florescence spectra could be reproduced with a ±8% error. A further aspect to consider involves the selectivity of the present amplified sensing platform toward mutants of the target DNA, (2). Toward this end we have examined two different one-base mutants (2a) and (2b) and one two-base mutant (2c) (for details see Figure S3 in the Supporting Information). We find that the sensing platform reveals selectivity, and the fluorescence signals upon analyzing (2a) and (2b) are 40% and 25% of the fluorescence signal generated by the target (2), respectively, whereas the fluorescence signal generated by the two-base mismatched mutant (2c) is
