Multi-Amplified Sensing of MicroRNA by a Small ... - ACS Publications

Oct 10, 2016 - In this work, we describe a microRNA (miRNA) detection assay that combines target recycling and isothermal amplification in an elegantl...
0 downloads 0 Views 810KB Size
Download PDF
Subscriber access provided by CORNELL UNIVERSITY LIBRARY

Article

Multi-Amplified Sensing of MicroRNA by a Small DNA Fragment-Driven Enzymatic Cascade Reaction Eunjung Kim, Philip D. Howes, Spencer W. Crowder, and Molly M. Stevens ACS Sens., Just Accepted Manuscript • DOI: 10.1021/acssensors.6b00601 • Publication Date (Web): 10 Oct 2016 Downloaded from http://pubs.acs.org on October 15, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Sensors is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 9

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

Multi-Amplified Sensing of MicroRNA by a Small DNA Fragment-Driven Enzymatic Cascade Reaction Eunjung Kim, Philip D. Howes, Spencer W. Crowder and Molly M. Stevens* Department of Materials, Department of Bioengineering and Institute for Biomedical Engineering, Imperial College London, London, SW7 2AZ, UK

KEYWORDS: biosensing, miRNA, DNA fragments, target recycling, isothermal amplification

ABSTRACT: Combining technological developments such as nanomaterials, DNA nanotechnology and functional enzymes has great potential to facilitate next generation high performance molecular diagnostic systems. In this work we describe a microRNA (miRNA) detection assay that combines target recycling and isothermal amplification in an elegantly designed enzyme-mediated cascade reaction. Target recycling is driven by the action of duplex-specific nuclease (DSN), resulting in highly amplified translation of input miRNA to short output DNA fragments. These fragments act as highly specific initiators of rolling circle amplification (RCA), an isothermal reaction that outputs a large volume of polymeric DNAzymes per initiator, and finally a fluorogenic output signal. Based on careful electrophoretic analysis we observed that the DSN produces ca. 10 nt DNA fragments from DNA/miRNA duplexes, regardless of the length of DNA strands. Target recycling yielded ca. five orders of magnitude amplification through DSN-assisted recycling system on magnetic particles, and the RCA a further two orders of magnitude. The final assay exhibited a limit of detection of 1.8 fM of miRNA spiked into 20% human serum, and showed excellent selectivity for miR-21 versus single base-mismatched sequences and other cancer-related miRNAs. The developed assay was further employed to determine accurate amounts of miR-21 in total RNA samples extracted from human cancer cell lines and normal cells, confirming the applicability of the assay for direct and absolute quantification of mature specific miRNA in real biological samples.

Over the last decade researchers have been revealing intricate relationships between the expression of noncoding RNAs with particular disease states.1,2 Such studies have provided an excellent foundation for translating fundamental biochemistry into clinically relevant ‘panels’ of molecular markers to form the basis of next generation medical diagnostics. Of particular interest has been microRNAs (miRNAs), as they are highly conserved and often present in a stable form in bodily fluids.3,4 For miRNAs to become clinically applicable molecular biomarkers, there is a pressing need for the development of technologies that are capable of accurate (sensitive and specific) and cost-effective testing. Existing technologies, such as qRT-PCR, microarrays and small RNA sequencing,5-7 are not suitable for wide-spread clinical application, thus it is necessary for us to look for emerging technologies to form the foundations of new clinically applicable assays. Various enzyme-assisted amplification strategies have emerged as alternatives to existing methods. Combining different functional enzymes provides a broad range of assay design opportunities.8 Such methods use a variety of nucleases to recognize specific sequences or structures of DNA or RNA substrates. They can be employed either alone or in combination with target recycling, probe amplification, signal amplification, and/or catalytic cascades, allowing very low limits of detection. Of specific interest here is duplex specific endonuclease (DSN), which has exhibited outstanding performance in the quantitative detection of miRNAs,9-23

single-nucleotide polymorphisms,24 and telomeric DNA overhangs.25 DSN has a strong preference for cleaving dsDNA or DNA in DNA/RNA heteroduplexes in a sequenceindependent manner, leaving RNA molecules intact.24,26 In particular, DSN needs a substrate of minimum 10 bp for recognition, and shows an excellent ability to distinguish between perfectly matched and partially matched duplexes with up to 1 bp mismatch. Inspired by these unique features, various miRNA detection assays using DSN have been developed to achieve high sensitivity and selectivity. Many approaches involve the design of DNA probes such as dyelabeled probes,14, 16 Taqman probes,9 and molecular beacons that directly induce a readable signal or trigger to release Gquardruplexes or DNAzyme-forming fragments for signal amplification after DSN digestion.10,12,15,17,21 Although such methods are very simple and fast, their detection limits have been relatively low (ranging from 0.1 pM to 2 nM) and their performance in complex biological samples was not reported. This may due to non-specific degradation, possibly by other endonucleases from biological origins, which is a significant drawback for direct detection of miRNA in serum or cell lysates. In addition to using DNA probes as a signal readout, various nanomaterials such as gold nanoparticles,11,13 quantum dots,19 reduced graphene oxide14 or tungsten disulfide nanosheets,16 magnetic micro/nano particles18, 20, 22-23 have been adopted to mediate the generated fluorescent signals or increase the quantity of the DNA probes. These nanomaterialbased DSN assays have enabled the determination of miRNA

ACS Paragon Plus Environment

ACS Sensors

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

at the femtomolar level, the recognition of single base mismatched miRNA, and good performances comparable to qRT-PCR in real samples. Among them, magnetic particles provide a useful means of rapid and easy separation of undesired components in the reaction mixtures, allowing an increase in the signal to noise ratio. However, such sensing approaches rely largely or solely on the DNase activity and mismatch discrimination ability of DSN. Indeed, there are few investigations on the exact cleaving mode of DSN on DNA/miRNA substrates with varying their lengths or mismatches under different reaction media. To create the optimized assay with improved detection limits, there is a need to systemically study DSN-based miRNA reactions, thereby looking into different reaction conditions as well as the cleaving mode (random or specific-site cutting). To build rationally designed multi-amplified reactions - linking DSNmediated reaction to further amplification steps - it is useful to have a comprehensive understanding of DNA products produced by DSN, as well as its enzymatic characteristics including the required substrate length and optimal process conditions. Herein we propose a new assay for the detection of miRNA, choosing miR-21 due to its relevance in cancer research – overexpression of miR-21 has been consistently linked with a poor outcome for cancer sufferers.29,30 The proposed approach is based on a series of enzymatic cascade reactions, as illustrated in Figure 1. The assay co-employs DSN-mediated target recycling, and the rolling circle amplification reaction (RCA) in three steps: 1) DSN-mediated generation of DNA fragments (Tc) through target recycling and magnetic separation, 2) Tc-primed RCA, and 3) detection of RCA products using DNAzyme substrates. In the presence of target miRNA, miRNA/DNA heteroduplexes are formed on the surface of magnetic microparticles and are subjected to a cleavage reaction by DSN. The DSN cleaves the DNA probe from its 5’-terminus in a heteroduplex into two DNA fragments, preserving the 3’ overhang, plus a few nucleotides (nt). Critically, the miRNA is left intact, and through miRNA recycling a large number of ca. 10 nt fragments are generated per single input miRNA (up to ca. five orders of magnitude in our experiments). The fragments are then easily isolated by removal of the magnetic particles with a magnet. The generated Tc fragments are then used to initiate a second amplification scheme based on RCA. A circular template DNA, dNTPs, and phi29 DNA polymerase (DNAP) together form tandem repeat units of ssDNA strands that form a Mg2+dependent RNA cleaving DNAzyme. By introducing DNAzyme substrates with Förster resonance energy transfer (FRET) pairs, the RCA products cut the substrate to dissociate the dye and quencher, leading to significantly increased fluorescence intensity as a signal readout. Overall, our rationally designed signal amplification approach using DNA fragments enables highly enhanced miRNA detection sensitivity by converting a single target binding event into a large abundance of functional DNA amplicons. Below we describe first the optimization of the DSN-mediated Tc fragment generation, and secondly the optimization of the fragment-initiated RCA assay. Finally, we explain how these two were combined to form the complete assay. EXPERIMENTAL SECTION Materials. HPLC-purified oligonucleotides were purchased

Page 2 of 9

Figure 1. Schematic illustration of the amplified detection of miRNA based on cascade enzymatic reactions composed of Tc fragment production by DSN-mediated catalysis, consecutive signal amplification using RCA and DNAzyme-mediated signal readout.

from Integrated DNA Technologies (USA) and used without further purification. The sequences of the oligonucleotides used in this study are listed in Table S1. Duplex-specific nuclease (DSN) was obtained from Evrogen (Moscow, Russia). T4 DNA ligase, exonuclease I (Exo I), exonuclease III (Exo III), exonuclease VII (Exo VII), phi29 DNA polymerase (phi29 DNAP), and deoxynucleotides (dNTPs) solution mix were obtained from New England Biolabs (USA). Streptavidin-coated magnetic microparticles (Dynabeads MyOne Streptavidin C1), dithiothreitol (DTT), Tris-HCl buffer (pH 8.0), magnesium chloride (MgCl2) solution, ethylenediaminetetraacetic acid (EDTA) solution, Tris/boric acid/EDTA (TBE) buffer, DNA ladders (10 bp, 100 bp, and 1 kb plus), BlueJuice gel loading buffer, SYBR gold nucleic acid gel stain, SYBR green I, and Quant-iT OliGreen ssDNA assay kit were purchased from Thermo Fisher Scientific (USA). 30% Acrylamide/bis solution (29:1), 40% acrylamide/bis solution (29:1), and TBE-urea sample buffer were obtained from Bio-Rad (USA). Sodium chloride (NaCl), ammonium persulfate (APS) and tetramethylethylenediamine (TEMED) were purchased from Sigma-Aldrich (UK). Amicon Ultra-0.5 Centrifugal Filter Units with Ultracel-10 membrane (MWCO 10 kDa) were purchased from MERCK Millipore (Germany). Human serum was purchased from Sera Laboratories International Ltd (UK). RNase/DNase-free distilled water (nuclease free water, NF water, Thermo Fisher Scientific) was used for the preparation of all aqueous solution and the Eppendorf RNA/DNA LoBind microcentrifuge tubes with RNase/DNase-free were used to maximize the sample recovery. Preparation of Magnetic Microparticles Modified with Probe DNA (MPs-A10 DNA). MPs-A10 DNA were prepared by direct coupling of 1 µm streptavidin-functionalized MPs (Dynabeads MyOne Streptavidin C1) to 3’-biotin-labeled oligonucleotides (A10 DNA) according to the manufacturer’s

ACS Paragon Plus Environment

Page 3 of 9

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

instructions. Briefly, after 2-min of vortexing, 200 µL of MPs (10 mg/mL, 2 mg) were transferred to a new tube, washed three times with 1× binding and washing buffer (5 mM TrisHCl, 0.5 mM EDTA, 1 M NaCl, pH 7.5) and resuspended in 400 µL of 2× binding and washing buffer. The biotinylated A10 DNA (1 nmol) dispersed in 400 µL of NF water was then added to the washed MPs and incubated for 20 min at room temperature using gentle rotation. After incubation, the supernatant was separated from the solution by placing them on a permanent magnet and used for DNA quantification. The DNA-coated MPs were washed once with 1× binding and washing buffer and twice with NF water, then resuspended in 200 µL of NF water and kept at 4 oC until further use. The coupling of 3’-biotin and 5’-Cy5-modified A10 DNA (Cy5A10 DNA) with MPs was also carried out in the same way. DSN Reaction of MPs-A10 DNA in the Presence of Target miRNA. Before the detection of target miRNA, the DSN reaction was optimized by varying the reaction conditions, including the concentration of MgCl2, incubation temperature and time. For these optimization experiments, Cy5-A10 DNA-coupled MPs (MPs-Cy5-A10 DNA) were used to measure fluorescence intensity. In a typical DSN reaction, each reaction was conducted in a final volume of 50 µL consisting of 15 µL MPs-A10 DNA (2 µM), 5 µL of target miRNA (the specified concentration), and 0.5 µL of DSN (0.5 U/µL) in the optimized reaction buffer (50 mM Tris-HCl, 25 mM MgCl2, 1 mM DTT, pH 8.0). The reaction mixtures were incubated at 55 oC for 2 hr using a thermomixer. The reaction was then stopped by separating the cleaved DNA fragments (Tc) from each mixture using a permanent magnet. The collected Tc solution was subsequently heated at 95 oC for 30 min to inactivate DSN. Fluorescence measurements were performed using Fluorolog-3 spectrofluorometer (HORIBA Scientific, UK) with a sub-micro quartz cuvette (path length = 10 mm, Starna, UK). The Cy5 was excited at λex = 630 nm and the emission spectra were obtained from 645 nm to 800 nm. The fluorescent intensity of each solution was normalized using (F-F0)/(Fmax-F0), where F and F0 are the fluorescent intensity of solution in the presence and absence of a target and Fmax is the maximum fluorescent intensity of solution at λ = 663 nm. Tc Fragment-Assisted RCA. For the RCA reaction with Tc, 45 µL of Tc solution was added to 14.5 µL of the stock solution containing circular template DNA (0.2 µM), dNTPs (0.3 mM), and BSA (83.3 µg/mL) in 1× phi29 DNAP reaction buffer (50 mM Tris-HCl, 10 mM MgCl2, 10 mM (NH4)2SO4, 4 mM DTT, pH 7.5) and incubated at 15 oC for 30 min to allow DNA hybridization between Tc and template DNA. The RCA process was initiated by adding 0.5 µL of phi29 DNAP (10 U/µL) to the mixture and incubated at 15 oC for 3 hr before heating at 65 oC for 10 min. To detect the RCA products, the above reaction mixtures were further incubated with 0.5 µL of DNAzyme substrate (100 µM) at room temperature for 1 hr. To stop Mg2+responsive cleavage of DNAzyme substrates, 2.5 µL of EDTA (500 mM) was added. Fluorescence emission spectra were recorded using a Fluorolog-3 spectrofluorometer from 645 nm to 800 nm at λex = 630 nm. The fluorescent intensity of each solution was normalized using (F-F0)/(Fmax-F0), where F and F0 are the fluorescent intensity of solution in the presence and absence of a target and Fmax is the maximum fluorescent intensity of solution at λ = 665 nm.

Cell Culture and RNA Isolation. Human cancerous cell lines including MCF-7, MDA-MB-231, and HeLa and noncancerous cells, human embryonic kidney cells HEK 293T, were obtained from ATCC. All cells were maintained in Dulbecco’s Modified Essential Medium (DMEM, Gibco, ThermoFisher Scientific, UK) with 10% (v/v) fetal bovine serum (FBS, Gibco) and 1% (v/v) penicillin/streptomycin (Gibco). All reagents and kits for RNA isolation, cDNA synthesis, and qRT-PCR were purchased from QIAGEN (Manchester, UK) and used according to the manufacturer’s instructions. Cells were lysed with the Qiazol reagent, passed through a 21 gauge needle for homogenization, mixed with chloroform (1:5 chloroform/ Qiazol), and separated by centrifugation at 12,000 × g (15 min, 4 oC). Total RNA contained within the aqueous phase, including mRNA and miRNA, was then isolated with miRNeasy columns. The RNA concentration was measured with a NanoDrop 2000c spectrophotometer (ThermoFisher Scientific, UK). Quantification of miR-21 in RNA Extracts Using qRTPCR and Tc-Primed RCA. All reagents were obtained from QIAGEN and used according to the manufacturer’s instructions. For quantification of miR-21 by quantitative realtime polymerase chain reaction (qRT-PCR), cDNA was synthesized from the extracted RNA using the miScript II RT kit. qRT-PCR was performed with the miScript SYBR Green PCR kit and the Hs_miR-21_2 miScript Primer Assay with 3 ng input cDNA reaction, using a QuantStudio 6 Flex RealTime PCR System (Applied Biosystems, ThermoFisher Scientific, UK). The qRT-PCR reaction was as follow: 95 oC for 15 min, followed by 40 cycles of denaturation at 94 oC for 15 sec, annealing at 55 oC for 30 sec, and extension at 70 oC for 30 sec. Synthetic miR-21 of known concentration was also used to prepare cDNA for a qRT-PCR standard curve, ranging from 2.5×10-17 to 2.5×10-9 M (Figure S14). The cycles to threshold (Ct) values of unknown samples from cell-derived RNA/cDNA were compared against the standard curve, and the concentration of miR-21 was calculated. The data presented are for N = 4 biological replicates. For miR-21 detection by Tc-primed RCA assay, the RNA extract (2 µL, 1~4 µg RNA per assay) was directly added into 50 µL of mixture composed of MPs-A10 DNA (600 nM) and DSN (0.25 U) in the reaction buffer and reacted at 55 oC for 2 hr. The Tc fragments were separated using a magnet and heated to 95 oC for 30 min, followed by the RCA reaction and detection of the RCA products in the same procedure, as described in the method of Tc Fragment-Assisted RCA. The data presented are for N = 4 biological replicates and N = 4 assay replicates. RESULTS AND DISCUSSION DSN Catalytic Activity on DNA/miRNA Duplexes. First we assessed DSN catalytic activity on three DNA/miRNA substrates by varying the length of probe DNA with poly(A) sequences. The DNA/miRNA heteroduplexes (1.5 µM) were incubated with DSN (0.25 units) under various temperatures and incubation times, then analyzed by 20% native polyacrylamide gel electrophoresis (PAGE) (Figures S1 and S2). The DSN showed strong cleavage preference for DNA in DNA/miRNA duplexes, leaving the miRNA intact, as previously reported in the literature.24,26 The DNA strands were cut into two major fragments, the smaller of which was ca. 10 nt, regardless of the length of each DNA strand (lane 5-

ACS Paragon Plus Environment

ACS Sensors

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 9

Figure 2. Tc fragment generation by DSN-mediated cleavage. Three DNA/miRNA substrates (1)-(3) before and after DSN treatment were resolved by 20% denaturing PAGE. Lane 1: 10 bp DNA marker, lane 2: miRNA, lane 3: A0 DNA, lane 4: A0 DNA/miRNA duplexes, lane 5: A0 DNA/miRNA duplexes treated with DSN, lane 6: A10 DNA, lane 7: A10 DNA/miRNA duplexes, lane 8: A10 DNA/miRNA duplexes treated with DSN, lane 9: A20 DNA, lane 10: A20 DNA/miRNA duplexes, lane 11: A20 DNA/miRNA duplexes treated with DSN, lane 12: 9 nt Tc,syn, lane 13: 10 nt Tc,syn.

10 in Figures S1 and S2). Zhao et al. also reported a digestion limit of DSN, showing that 5-6 nt of dsDNA at the junction of ssDNA and dsDNA are resistant to DSN cleavage, demonstrating the use of DSN to determine the size of 3’ overhangs of telomeric dsDNA.25 Similarly, we observed that DSN preserved the 12 nt of DNA at the miRNA-DNA junction, releasing 10 nt DNA Tc fragments. DSN is extremely thermostable, having a wide range of reaction temperatures from 30 oC to 70 oC. Many DSN-based assays have been set up at its optimal temperature of 40-60 oC and reaction time of 30-120 min.11,13,15-23 Therefore, we sought the optimum condition of DSN to leverage its cleaving capability. The DSN digested more than 90% of DNA/miRNA at 55 oC during a 2 hr incubation, based on quantitative analysis of relative band intensities corresponding to each substrate. We note that 1 mM of dithiothreitol (DTT) in the DSN reaction buffer (50 mM Tris-HCl, 5 mM MgCl2, pH 8.0) was an essential element to obtain the DNA fragments. As shown in Figure S3, without DTT in the reaction buffer DNA strands were degraded, but no Tc fragments were seen on the gel images (lane 4). DTT as a reducing agent is commonly used to protect an enzyme activity loss and seems to ensure that the DSN can recognize 10 nt (at least) DNA and cleave them. Therefore, DTT is notably necessary for DNA fragment generation in our assay. In consideration of the RCA step in the final assay, it was important to accurately determine the size of the Tc fragments to inform the design of the circular RCA DNA template. This was done by resolving DSN-treated DNA/miRNA duplexes under the optimal condition with 20% denatured PAGE (Figure 2). We used 9 nt and 10 nt synthetic DNA fragments (Tc,syn) by selecting the 9 and 10 sequences from the 5’-end of probe DNA (Table S1) and compared their size with the obtained Tc fragments. As expected, the produced Tc fragments from all substrates were 10 nt (lane 5, 8, and 11), indicating fragment release independent of the DNA length. Optimization of DSN-Triggered Tc Fragment Generation. In the assay it is necessary to separate the DNA

Figure 3. DSN-assisted Tc fragment generation and isolation using magnetic separation. (A) 20% denatured PAGE analysis of DSN-treated MPs-A10 DNA/miRNA. Lane 1: 10 bp DNA marker, lane 2: miRNA, lane 3: A10 DNA, lane 4: A10 DNA/miRNA duplexes, lane 5: A10 DNA/miRNA duplexes (1) with DSN, lane 6: supernatants after magnetic separation of MPsA10 DNA/miRNA complexes (2) with DSN. (B) Time-dependent normalized emission spectra of Tc solution separated from MPsCy5-A10 DNA (400 nM) with miRNA (10 nM) and DSN at different time intervals. Inset: normalized fluorescence intensity at λ = 663 nm plotted as a function of incubation time. (C) Targetresponsive normalized emission spectra and (D) fluorescence intensities of Tc solution separated from MPs-CY5-A10 DNA (400 nM) incubated with DSN for 2 hr. Here N=3, and error bars refer to the standard deviation.

fragments from the intact DNA substrate after the DSN reaction, and magnetic separation allowed a quick and efficient way of achieving this. 3’-biotin-conjugated A10 DNA was conjugated to streptavidin-functionalized magnetic microparticles (termed MPs-A10 DNA hereafter). The MPsA10 DNA were tested by mixing them (400 nM) with target miRNA (10 nM), then incubating with DSN at 55 oC for 2 hr. The gel image in Figure 3a confirmed that Tc fragments could be successfully recovered from the reaction mixture using magnetic separation (lane 6), and the same size Tc fragments are produced compared with the free DNA/miRNA duplexes (lane 5). For further optimization of DSN-promoted Tc fragment generation, we repeated the DSN reaction whilst varying Mg2+ concentration and reaction temperature. To measure the amount of Tc fragments released from MPs-A10 DNA, Cy5labeled A10 DNA was used. Here, cleavage of the A10 DNA by DSN released fluorescent dye-labeled fragments. After removal of the magnetic particles, the fluorescence of the remaining solution was indicative of the concentration of Tc fragments obtained after cleavage (using (F-F0)/(Fmax-F0), where F and F0 are the fluorescent intensity of solution in the presence and absence of a target and Fmax is the maximum

ACS Paragon Plus Environment

Page 5 of 9

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

fluorescent intensity of solution). As shown in Figures S4 and S5, the highest signal was observed in 25 mM of Mg2+ in buffer at 55 oC, which is 13.2 oC lower than the estimated melting temperature (Tm) of the A10 DNA/miRNA duplex (Figure S6a). The reaction temperature optimized in our DSN reaction is slightly different to the results obtained in previous studies.13,15-18, 20-23 This might be driven by the compromise between optimal enzymatic activity and DNA/miRNA hybridization. Given that typical hybridization is performed at a temperature of 10-15 oC below Tm, a 55 oC incubation temperature seems to make both A10 DNA/miRNA duplexes and DSN activity most preferable. Furthermore, time-dependent fluorescence changes of the Tc fragment solution are observed upon analyzing 10 nM of target miRNA with 400 nM of MPs-Cy5-A10 DNA and 0.25 units of DSN (Figure 3b). The fluorescence signal from the isolated fragments gradually increased and reached saturation after 2 hr. From these data, we concluded the optimal conditions for DSN-based Tc fragment generation to be a 2 hr incubation at 55 oC in 50 mM Tris-HCl, 25 mM MgCl2, 1 mM DTT at pH 8.0. Under these conditions, we obtained a detection limit of 0.6 pM target miRNA with a good linear dynamic range from 9 pM to 1 nM based on 3σ (Figures 3c and 3d), which compares favorably with other single enzymebased detection approaches using DSN.9,10,13,18,19 To analyze the selectivity of DSN target recycling reaction we selected four other miRNAs involved in cancer and metastasis (miR-34a, miR-155, miR-10b, and let-7a),27 and also used 1 nt mismatched miR-21 at different positions, miR21 1m (1) and miR-21 1m (2). The fluorescence signal response of the MPs-Cy5-A10 DNA (400 nM) to the seven different miRNAs (10 nM) is depicted in Figure 4. The signal in the presence of miR-21 was much higher than that from miR-34a (26-fold), miR-155 (28-fold), miR-10 (32-fold), and let-7a (46-fold), respectively. In contrast, both miR-21 with 1 nt mismatches showed only a slight decrease in the signal, meaning that they can partially hybridize with the probe DNA, allowing some cleavage by DSN. DSN has previously been reported to allow differentiation between variations of only a single nucleotide in short DNA duplexes.24 However, it is likely that DSN still has partial cleavage activity on DNA in 1 nt mismatched miRNA-containing duplexes as it has very strong affinity to DNA molecules. To verify this, we conducted a DSN digestion study on perfectly matched or 1 nt mismatched DNA/miRNA duplexes using the denatured PAGE assay (Figure S7). As expected, we found that DSN can cleave some DNA strands with different cleavage efficiency (96% for miR-21, 93% for miR-21 1m (1), and 85% for DNA/miR-21 1m (2)).This is because the increased thermal stability of perfectly matched duplexes (Tm = 68.1 oC) compared to the others (Tm = 63.4 oC for miR-21 1m (1) and 62.5 oC for miR-21 1m (2)) (Figure S6b) led to a higher possibility to form duplexes during the reaction, thereby showing higher cleavage efficiency. Here we should note that there was no target recycling, therefore, we expect that the ability to discriminate a single-base mismatch would be greatly enhanced by the combination of DSN-based target cycling and RCA-mediated signal enhancement. Optimization of Tc Fragment-Mediated RCA Assay. As the second step in the full assay development, we optimized the Tc fragment-triggered RCA reaction that serves as a second amplification step. Firstly, we prepared circular

Figure 4. Target miRNA-specific Tc recovery. Normalized fluorescence intensities at λ = 663 nm of Tc solution separated from MPs-Cy5-A10 DNA (400 nM) in the presence of various miRNAs with DSN for 2 hr. Inset: normalized emission spectra of each Tc solution. [miRNA] = 10 nM. N=3, standard deviation shown in the error bar.

template DNA containing sequences complementary to the Tc fragment, and the sequence of an Mg2+-dependent RNAcleaving DNAzyme, by ligation of 5’-phosphorylated linear template DNA with T4 DNA ligase. The ligated template was additionally treated with Exo I/III to remove any excess ssDNA or dsDNA, and we confirmed circularization using both native and denatured 20% PAGE assays (Figure S8). Residual probe DNA produced during the DSN reaction is readily removed using magnetic separation. However, active DSN and miRNA are still present in the Tc fragment solution. The DSN could potentially hinder the intrinsic strand displacement and processive polymerization capacity of phi29 DNAP in RCA reaction buffer (50 mM Tris-HCl, 10 mM MgCl2, 10 mM (NH4)2SO4 and 4 mM DTT, at pH 7.5). To test this, the RCA reaction was performed with Tc fragments (obtained from [miRNA] = 10 nM) and circular template DNA (600 nM) in the presence or absence of DSN (0.25 units) at 30 o C. As shown in Figure S9, no RCA products were observed in the RCA reaction with DSN (lane 4), compared with RCA without DSN (lane 3). This indicated that the DSN was still active and inhibited phi29 DNAP activity. DSN is highly stable over a broad range of temperatures (30 - 70 oC) and pH (pH 3 - 9), however addition of 0.2 M NaCl decreases its enzymatic activity 10-fold.26 Interestingly, it was shown that the addition of excess NaCl still did not allow formation of RCA products (lane 5, 6), as it could also impede RCA-based amplification even in the absence of DSN (lane 7).28 Finally, we achieved successful RCA by inactivating DSN by heating the Tc fragment solution up to 95 oC for 30 min (lane 8). The effect of the RCA reaction temperature on product production was evaluated based on the consideration that the observed Tm for the hybridization of Tc (400 nM) with circular template DNA (1 µM) (Figure S10) was around 30 oC. This is the optimal temperature for phi29 DNAP. As depicted in Figure S11, RCA at 15 oC for 3 hr gave us an excellent Tc dose-responsive formation of RCA products, compared to 30 o C for 1 hr, implying that the low reaction temperature allowed short Tc to be well hybridized with template DNA. Detection Performance of the Full Assay. As a final step in assay development, we combined the DSN and RCA reactions into a multi-amplified enzymatic cascade reaction. Under the optimized conditions discussed above, we carried

ACS Paragon Plus Environment

ACS Sensors

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 5. Ultrasensitive and selective detection of miRNA based on our amplified Tc-primed RCA assay. (A) Target doseresponsive normalized fluorescence intensities at λ = 665 nm. [MPs-A10 DNA] = 600 nM, [DSN] = 0.005 U/µL, DSN reaction was carried out at 55 oC for 2 hr. [circular template DNA] = 167 nM, [phi29 DNAP] = 0.083 U/µL. RCA was performed at 15 oC for 3 hr. [DNAzyme substrate] = 1 µM. (B) Target-specific normalized fluorescence intensities at λ = 665 nm. Inset: normalized emission spectra of each sample. [miRNA] = 10 nM. N=3, standard deviation shown in the error bar.

out the full assay for miR-21 detection in four steps: 1) DSN reaction with MPs-A10 DNA (600 nM) with a range of miRNA concentrations (10 fM - 100 nM); 2) Tc fragment isolation by magnetic separation and heat inactivation of DSN; 3) Tc-primed RCA; 4) detection of RCA products with DNAzyme substrates. For the final detection step, we used a Mg2+-dependent RNA cleaving DNAzyme that can specifically recognize and cleave a single ribonucleotide (rA)containing substrate for visualization of the RCA process.29-31 Mg2+ ion-abundant reaction conditions (ca. 30 mM) resulting from both the DSN and RCA reactions yields enhanced catalytic activity of the DNAzyme without additional MgCl2. Remarkably, the Tc-primed RCA offered a highly improved limit of detection of 6.8 fM (based on 3σ), shown in Figure 5a. This is a near two orders of magnitude improvement over the DSN-based assay alone (0.6 pM). This final assay also gave rise to greatly enhanced mismatch discrimination. As shown in Figure 5b, the change in fluorescence produced by miR-21 was 1.8- and 4.6-fold higher than that of miR-211m (1) and miR-21 1m (2), respectively. Thus the subsequent signal enhancement of the released Tc simultaneously improves both the specificity and sensitivity of miRNA detection. We also found that the amplified Tc-primed RCA presents an amplification of more than 188,000-fold even if the amount of target miRNA is as low as 200 fM compared with nonamplified Tc-based RCA (Figure S12). Thus, approximately 5 orders of magnitude amplification through DSN-mediated miRNA recycling is observed, which is consistent with the previous findings reported by Yin et al. and Degliangeli et al.9,13 These results indicate that a majority of the amplification in our assay is from the first step of DSN-based Tc amplification and efficient Tc recovery through magnetic separation. Detection Performance with Serum Samples. The above results demonstrate that the assay is highly sensitive and selective in clean conditions. Next, we sought to show that amplified Tc-primed RCA is capable of sensitive detection in a biologically relevant matrix, a critical and non-trivial validator of all new diagnostic assays. We performed the assay with varying concentrations of miR-21 spiked into 20% (v/v)

Page 6 of 9

human serum. The fluorescence intensity of the RCA products showed a good dependence on target concentration and a limit of detection of 1.8 fM (based on 3σ) (Figure S13), which is comparable to that observed for miR-21 in buffer (Figure 5a). These results show that our detection strategy is capable of directly detecting miRNAs as low as 2 fM by eliminating interferences from coexisting biological molecules in a real sample. Detection of miR-21 in Total RNA Extracted From Cells. Finally, we employed the full assay to determine the content of endogenous mature miR-21 in various cell lines. MCF-7, MDA-MB-231 (human breast cancer cells) and HeLa (human cervical cancer cells) were chosen as cancerous cell lines, and HEK 293T (human embryonic kidney cells) were selected as a non-cancerous cell line. For absolute quantification, 2 µL of total RNA (1~4 µg) extracted from each cell line was directly used in the full assay and the expression levels of miR-21 were quantified through the standard curve in Figure 5a. The obtained values were further assessed with the results of qRTPCR as a reference. We observed stronger fluorescent signals in cancer cells than normal cells (Figure S15), confirming that miR-21 is more abundant in cancer cells. The amount of miR21 determined by the assay was (0.24 ± 0.10)×104 copies/ng RNA in HEK 293T, (86.53 ± 9.07)×104 copies/ng RNA in MCF-7, (11.12 ± 1.08)×104 copies/ng RNA in MDA-MB-231, and (21.35 ± 3.30)×104 copies/ng RNA in HeLa cells (Table 1). This is in good agreement with qRT-PCR-based quantification with the established standard curve using synthetic miR-21 (Figure S14). In particular, the contents of miR-21 in MCF-7 and HeLa cells agree well with the values reported by Degliangeli et al. (75×104 copies/ng for MCF-7, 24×104 copies/ng for HeLa), Si Li et al. (1.13 amol/ng or 68.05×104 copies/ng for MCF-7, 0.39 amol/ng or 23.49×104 copies/ng for HeLa), and Zhao et al. (12.11 fmol/10 µg or 72.93×104 copies/ng for MCF-7, 3.18 fmol/10 µg or 19.16×104 copies/ng for HeLa).13,32,33 Therefore, these results indicate the universal applicability of the designed assay for the quantification of specific miRNA in cells without the need to synthesize miRNA into cDNA and perform several cycles of amplification. Table 1. Quantification of miR-21 (104 copies/ng RNA) in total RNA extracted from each cell line. Cell line

Tc-primed RCAa

qRT-PCRb

HEK 293T

0.24 ± 0.10

0.39 ± 0.04

MCF-7

86.53 ± 9.07

94.06 ± 13.56

MDA-MB-231

11.12 ± 1.08

12.16 ± 1.77

HeLa

21.35 ± 3.30

22.94 ± 4.20

a 1~4 µg of total RNA per assay. Data represent average ± s.d. for N=4 biological replicates and N=4 assay replicates. b

Data represent average ± s.d. for N=4 biological replicates.

CONCLUSIONS We have developed a multi-amplified enzymatic cascade assay that allows highly sensitive (1.8 fM LOD) and specific (single base mismatch discrimination) detection of miR-21. The assay combines functional nanoparticles, enzymes and DNAzymes, performing miRNA target recycling, isothermal

ACS Paragon Plus Environment

Page 7 of 9

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

amplification and fluorescence readout, into an elegant workflow. Further, we have demonstrated that the assay is capable of detecting miRNA-21 spiked into 20% human serum with no degradation in performance. Even though the error bars at the low concentrations of target miRNA are quite large, excellent results were obtained in the analysis of real biological samples (total RNA isolated from cells) and they showed a good correlation with the conventional qRT-PCR assay. This demonstrates that the developed methodology is applicable for direct and precise detection of mature miRNA in cells. We envisage that, with suitable automation and optimization, this assay could form the basis of a fast and efficient clinical diagnostics system.

ASSOCIATED CONTENT Supporting Information. Supporting Information Available: The following files are available free of charge. Details of experiments, a table of oligonucleotide sequences, additional figures, including temperature-dependent DSN activity results, time-dependent DSN activity results, DTT effect on DSN activity, the effect of MgCl2 concentrations on DSN activity, the effect of reaction temperatures on DSN activity, melt curves of DNA/miRNA duplexes, DSN activity on 1 nt mismatched DNA in DNA/miRNA duplexes, characterization of template DNA, PAGE analysis of the RCA products in the presence of DSN, melt curve analysis of Tc hybridization with template DNA, PAGE analysis of RCA products at different incubation temperatures, comparison of detection limit between Tc- and Tc,syn-mediated RCA, detection of miR-21 spiked in 20% human serum, and quantification of miR-21 with qRT-PCR and Tc-primed RCA.

AUTHOR INFORMATION Corresponding Author * E-mail: [email protected]

Notes The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. The authors declare no competing financial interest. Data requests can be made to [email protected].

ACKNOWLEDGMENTS E.K. acknowledges support from Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (2015R1A6A3A03018919). M.M.S. and P.H. acknowledge support from EPSRC through the Interdisciplinary Research Centre (IRC) “Early-Warning Sensing Systems for Infectious Diseases” (EP/K031953/1). M.M.S. also acknowledges EPSRC research grant “Bio-functionalised Nanomaterials for Ultrasensitive Biosensing” (EP/K020641/1).

REFERENCES (1) Huang, Y.; Wang, J. P.; Yu, X. L.; Wang, Z. B.; Xu, T. S.; Cheng, X. C. Non-Coding RNAs and Diseases. Mol. Biol. 2013, 47, 465-475. (2) Shi, X.; Sun, M.; Liu, H.; Yao, Y.; Song, Y. Long Non-Coding RNAs: A New Frontier in the Study of Human Diseases. Cancer Lett. 2013, 339, 159-166.

(3) Pritchard, C. C.; Cheng, H. H.; Tewari, M. MicroRNA Profiling: Approaches and Considerations. Nat. Rev. Genet. 2012, 13, 358-369. (4) Dong, H.; Lei, J.; Ding, L.; Wen, Y.; Ju, H.; Zhang, X. MicroRNA: Function, Detection, and Bioanalysis. Chem. Rev. 2013, 113, 6207-6233. (5) de Planell-Saguer, M.; Rodicio, M. C. Analytical Aspects of MicroRNA in Diagnostics: A Review. Anal. Chim. Acta 2011, 699, 134-152. (6) de Planell-Saguer, M.; Rodicio, M. C. Detection Methods for MicroRNAs in Clinic Practice. Clin. Biochem. 2013, 46, 869-878. (7) Mestdagh, P.; Hartmann, N.; Baeriswyl, L.; Andreasen, D.; Bernard, N.; Chen, C.; Cheo, D.; D'Andrade, P.; DeMayo, M.; Dennis, L.; Derveaux, S.; Feng, Y.; Fulmer-Smentek, S.; Gerstmayer, B.; Gouffon, J.; Grimley, C.; Lader, E.; Lee, K. Y.; Luo, S.; Mouritzen, P.; Narayanan, A.; Patel, S.; Peiffer, S.; Ruberg, S.; Schroth, G.; Schuster, D.; Shaffer, J. M.; Shelton, E. J.; Silveria, S.; Ulmanella, U.; Veeramachaneni, V.; Staedtler, F.; Peters, T.; Guettouche, T.; Wong, L.; Vandesompele, J. Evaluation of Quantitative miRNA Expression Platforms in the MicroRNA Quality Control (miRQC) Study. Nat. Methods 2014, 11, 809-815. (8) Gerasimova, Y. V.; Kolpashchikov, D. M. Enzyme-Assisted Target Recycling (EATR) for Nucleic Acid Detection. Chem. Soc. Rev. 2014, 43, 6405-6438. (9) Yin, B.-C.; Liu, Y.-Q.; Ye, B.-C. One-Step, Multiplexed Fluorescence Detection of microRNAs Based on Duplex-Specific Nuclease Signal Amplification. J. Am. Chem. Soc. 2012, 134, 50645067. (10) Lin, X.; Zhang, C.; Huang, Y.; Zhu, Z.; Chen, X.; Yang, C. J. Backbone-Modified Molecular Beacons for Highly Sensitive and Selective Detection of MicroRNAs Based on Duplex Specific Nuclease Signal Amplification. Chem. Commun. 2013, 49, 72437245. (11) Shen, W.; Deng, H.; Ren, Y.; Gao, Z. A Real-Time Colorimetric Assay for Label-Free Detection of MicroRNAs Down to Sub-Femtomolar Levels. Chem. Commun. 2013, 49, 4959-4961. (12) Tian, T.; Xiao, H.; Zhang, Z.; Long, Y.; Peng, S.; Wang, S.; Zhou, X.; Liu, S.; Zhou, X. Sensitive and Convenient Detection of MicroRNAs Based on Cascade Amplification by Catalytic DNAzymes. Chem. Eur. J. 2013, 19, 92-95. (13) Degliangeli, F.; Kshirsagar, P.; Brunetti, V.; Pompa, P. P.; Fiammengo, R. Absolute and Direct MicroRNA Quantification Using DNA–Gold Nanoparticle Probes. J. Am. Chem. Soc. 2014, 136, 22642267. (14) Guo, S.; Yang, F.; Zhang, Y.; Ning, Y.; Yao, Q.; Zhang, G.-J. Amplified Fluorescence Sensing of miRNA by Combination of Graphene Oxide with Duplex-Specific Nuclease. Anal. Methods 2014, 6, 3598-3603. (15) Wang, S.; Fu, B.; Wang, J.; Long, Y.; Zhang, X.; Peng, S.; Guo, P.; Tian, T.; Zhou, X. Novel Amplex Red Oxidases Based on Noncanonical DNA Structures: Property Studies and Applications in MicroRNA Detection. Anal. Chem. 2014, 86, 2925-2930. (16) Xi, Q.; Zhou, D.-M.; Kan, Y.-Y.; Ge, J.; Wu, Z.-K.; Yu, R.-Q.; Jiang, J.-H. Highly Sensitive and Selective Strategy for MicroRNA Detection Based on WS2 Nanosheet Mediated Fluorescence Quenching and Duplex-Specific Nuclease Signal Amplification. Anal. Chem. 2014, 86, 1361-1365. (17) Yan, L.; Yan, Y.; Pei, L.; Wei, W.; Zhao, J. A G-Quadruplex DNA-Based, Label-Free and Ultrasensitive Strategy for MicroRNA Detection. Sci. Rep. 2014, 4, 7400. (18) Hao, N.; Dai, P.-P.; Yu, T.; Xu, J.-J.; Chen, H.-Y. A Dual Target-Recycling Amplification Strategy for Sensitive Detection of MicroRNAs Based on Duplex-Specific Nuclease and Catalytic Hairpin Assembly. Chem. Commun. 2015, 51, 13504-13507. (19) Jou, A. F.-j.; Lu, C.-H.; Ou, Y.-C.; Wang, S.-S.; Hsu, S.-L.; Willner, I.; Ho, J.-a. A. Diagnosing the miR-141 Prostate Cancer Biomarker Using Nucleic Acid-Functionalized CdSe/ZnS QDs and Telomerase. Chem. Sci. 2015, 6, 659-665. (20) Shen, W.; Yeo, K. H.; Gao, Z. A Simple and Highly Sensitive Fluorescence Assay for MicroRNAs. Analyst 2015, 140, 1932-1938.

ACS Paragon Plus Environment

ACS Sensors

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(21) Li, Y.; Zhang, J.; Zhao, J.; Zhao, L.; Cheng, Y.; Li, Z. A Simple Molecular Beacon with Duplex-Specific Nuclease Amplification for Detection of MicroRNA. Analyst 2016, 141, 10711076. (22) Lu, W.; Chen, Y.; Liu, Z.; Tang, W.; Feng, Q.; Sun, J.; Jiang, X. Quantitative Detection of MicroRNA in One Step via Next Generation Magnetic Relaxation Switch Sensing. ACS Nano 2016, 10, 6685-6692. (23) Lv, W.; Zhao, J.; Situ, B.; Li, B.; Ma, W.; Liu, J.; Wu, Z.; Wang, W.; Yan, X.; Zheng, L. A Target-Triggered Dual Amplification Strategy for Sensitive Detection of MicroRNA. Biosens. Bioelectron. 2016, 83, 250-255. (24) Shagin, D. A.; Rebrikov, D. V.; Kozhemyako, V. B.; Altshuler, I. M.; Shcheglov, A. S.; Zhulidov, P. A.; Bogdanova, E. A.; Staroverov, D. B.; Rasskazov, V. A.; Lukyanov, S. A Novel Method for SNP Detection Using a New Duplex-Specific Nuclease From Crab Hepatopancreas. Genome Res. 2002, 12, 1935-1942. (25) Zhao, Y.; Hoshiyama, H.; Shay, J. W.; Wright, W. E. Quantitative Telomeric Overhang Determination Using a DoubleStrand Specific Nuclease. Nucleic Acids Res. 2008, 36, e14. (26) Anisimova, V. E.; Rebrikov, D. V.; Shagin, D. A.; Kozhemyako, V. B.; Menzorova, N. I.; Staroverov, D. B.; Ziganshin, R.; Vagner, L. L.; Rasskazov, V. A.; Lukyanov, S. A.; Shcheglov, A. S. Isolation, Characterization and Molecular Cloning of DuplexSpecific Nuclease from the Hepatopancreas of the Kamchatka Crab. BMC Boichem. 2008, 9, 1-12. (27) Conde, J.; Edelman, E. R.; Artzi, N. Target-Responsive DNA/RNA Nanomaterials for MicroRNA Sensing and Inhibition: The Jack-of-All-Trades in Cancer Nanotheranostics? Adv. Drug Deliv. Rev. 2015, 81, 169-183. (28) Blanco, L.; Bernad, A.; Lázaro, J. M.; Martín, G.; Garmendia, C.; Salas, M. Highly Efficient DNA Synthesis by the Phage Phi 29 DNA Polymerase. Symmetrical Mode of DNA Replication. J. Biol. Chem. 1989, 264, 8935-8940. (29) Breaker, R. R.; Joyce, G. F. A DNA Enzyme with Mg2+Dependent RNA Phosphoesterase Activity. Chem. Biol 1995, 2, 655660. (30) Wang, F.; Elbaz, J.; Teller, C.; Willner, I. Amplified Detection of DNA through an Autocatalytic and Catabolic DNAzyme-Mediated Process. Angew. Chem.-Int. Edit. 2011, 50, 295-299. (31) Orbach, R.; Willner, B.; Willner, I. Catalytic Nucleic Acids (DNAzymes) as Functional Units for Logic Gates and Computing Circuits: From Basic Principles to Practical Applications. Chem. Commun. 2015, 51, 4144-4160. (32) Li, S.; Xu, L.; Ma, W.; Wu, X.; Sun, M.; Kuang, H.; Wang, L.; Kotov, N. A.; Xu, C. Dual-Mode Ultrasensitive Quantification of MicroRNA in Living Cells by Chiroplasmonic Nanopyramids SelfAssembled from Gold and Upconversion Nanoparticles. J. Am. Chem. Soc. 2016, 138, 306-312. (33) Zhao, X.; Xu, L.; Sun, M.; Ma, W.; Wu, X.; Kuang, H.; Wang, L.; Xu, C. Gold-Quantum Dot Core–Satellite Assemblies for Lighting Up MicroRNA In Vitro and In Vivo. Small 2016, 12, 4662-4668.

ACS Paragon Plus Environment

Page 8 of 9

Page 9 of 9

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

for TOC only

ACS Paragon Plus Environment