SCIENCE CHINA Chemistry •  ARTICLES  •

August 2016

  Vol.59   No.8: 1051–1058

doi: 10.1007/s11426-015-0504-0

Ultrasensitive quantitation of MicroRNAs via magnetic beads-based chemiluminesent assay Bingcong Zhou1,2†, Haowen Yang1,3†, Yan Deng1,4, Ming Liu1, Bin Liu5, Nongyue He1,4* & Zhiyang Li6* 1

State Key Laboratory of Bioelectronics; School of Biological Science and Medical Engineering, Southeast University, Nanjing 210096, China 2 Laboratory of Neuro-Oncology, Tianjin Neurological Institute, Tianjin Medical University General Hospital, Tianjin 300052, China 3 Institute for Chemical and Bioengineering, Department of Chemistry and Applied Biosciences, ETH Zürich, Zürich 8093, Switzerland 4 Economical Forest Cultivation and Utilization of 2011 Collaborative Innovation Center in Hunan Province, Hunan Key Laboratory of Green Packaging and Application of Biological Nanotechnology, Hunan University of Technology, Zhuzhou 412007, China 5 Department of Biomedical Engineering, School of Basic Medical Sciences, Nanjing Medical University, Nanjing 210029, China 6 Department of Laboratory Medicine, Nanjing Drum Tower Hospital Clinical College of Nanjing Medical University, Nanjing 210008, China Received November 30, 2015; accepted February 10, 2016; published online July 4, 2016

MicroRNAs (miRNAs) are small non-coding RNAs that regulate gene expression at the post-transcriptional level, and their aberrant expression occurs during the development of malignant diseases. Recently, miRNAs have been proposed as potential prognostic and predictive biomarkers for early diagnosis. However, a major obstacle in rapid miRNA analysis from real samples is the lack of ultrasensitive and quantitative techniques. In this regard, the use of chemiluminescence (CL) system offers a highly sensitive strategy for detecting miRNAs. In this article, an ultrasensitive approach has been established for the quantification of miRNAs, using magnetic beads (MBs) and alkaline phosphatase (AP)-based CL system. This technique depends on sandwich hybridization among MBs-labeled capture probes, target miRNAs and biotin-labeled reporter probes, conjugation of streptavidin-alkaline phosphatase (SA-AP) to biotin-labeled reporter probes, and CL detection of AP-linked targets. Detection of miR-21 with this technique demonstrated a high selectivity and an ultralow limit of detection (LOD) of 60 fM with an extraordinarily wide range of six orders of magnitudes. The quantitation could be achieved by direct detecting target miRNA in serum samples within a total time of 1.5 h and did not require reverse transcription and polymerase chain reaction (PCR) amplification. Therefore, this developed method shows great potential for early cancer diagnosis based on miRNAs as biomarkers. MicroRNA, quantitation, magnetic bead, chemiluminescence, liver cancer

Citation:

Zhou B, Yang H, Deng Y, Liu M, Liu B, He N, Li Z. Ultrasensitive quantitation of MicroRNAs via magnetic beads-based chemiluminesent assay. Sci China Chem, 2016, 59: 1051–1058, doi: 10.1007/s11426-015-0504-0

   Introduction 1        MicroRNAs (miRNAs), a class of 19–24 nucleotides noncoding RNAs, can regulate gene expression by binding to the *Corresponding authors (email: [email protected]; [email protected]) †These authors contributed equally to this work.

© Science China Press and Springer-Verlag Berlin Heidelberg 2016

3'-untranslated region (3'-UTR) of messenger RNAs [1,2]. MiRNAs are known to perform key functions in a variety of cellular processes, such as proliferation, differentiation and apoptosis [3–5]. Emerging evidence has demonstrated that miRNAs regulate multiple genes associated with various human tumorigenesis [6–8]. This discovery has fueled tremendous interests in miRNAs as promising noninvasive diagnostic biomarkers, accurate pathological status predictors and chem.scichina.com   link.springer.com

 

1052

   Sci China Chem   August (2016)  Vol.59  No.8

Zhou et al.

appealing biomedical research targets. Recently, a number of miRNAs have been observed circulating in serum samples with dramatic stability [9–12], and their expression levels closely reflect the pathological status [13–16], indicating the possibility of using cell-free miRNAs as informative biomarkers in non-invasive detection methods. Therefore, the development of reliable, sensitive and simple detection and quantitation strategies for clinical diagnosis and therapeutic monitor is in urgent demand. An ideal miRNA detection method should fulfill several characteristics, such as high sensitivity in detecting miRNAs with low abundance of starting materials, high selectivity in distinguishing a specific miRNA sequence from family members, easy performance with readily available reagents and simple equipments, and capability in processing multiple samples in parallel [17]. However, current methodological approaches for miRNA analysis are not perfect and have inherent limitations (Table 1). For example, northern blotting analysis with fluorescent or radiolabeled probes, the gold standard of miRNA validation and quantitation, suffers from poor sensitivity due to the short length of mature miRNAs. It is also time-consuming and cumbersome involving laborious handling procedures [18]. Microarray, the high-throughput technique which is able to screen large numbers of miRNAs simultaneously, is particularly appealing recently. However, besides the necessity of extra expenses in fabricating microarray chips, microarray techniques lack availability in quantitative analysis and specificity in differentiating miRNAs sharing similar sequences [18]. All of these factors restrict their widespread applications. Quantitative reverse transcription-polymerase chain reaction (qRT-PCR) has been popular as a powerful approach in nucleic acid quantitation with high sensitivity and accuracy. However, qRT-PCR is preferable to detect miRNA precursors with long sequences

rather than mature miRNAs with short sequences [19,20]. Although many improved methods have been developed afterwards, including stem-loop RT-PCR using Taqman primers [21], SYBR Green RT-PCR using poly-A polymerase [22] or locked nucleic acid (LNA)-modified primers [23], and enzymatic ligation-based RT-PCR [24], these assays require special reagents or complicated modifications and are rather costly and not conventional for routine use in diagnostics. Moreover, reverse transcription, an essential step in qRTPCR, may produce potential bias to influence the accuracy of this methodology [25]. Therefore, there remains an urgent demand for a more accurate, simple, sensitive and inexpensive miRNA detection method for early-stage medical diagnostics. In addition to the conventional strategies, some feasible alternatives have been developed recently [25–27]. For example, Deo’s group [26] reported a competitive hybridization assay employing the bioluminescent enzyme Renilla luciferase (Rluc) as a label for detection and quantitation of miRNAs. This method successfully discriminated miR-21 levels in cancerous and noncancerous cells with a detection limit of 1 fmol. However, the competitive strategy measured a decrease rather than an increase bioluminescent intensity [18], and the solid-phase neutravidin-coated plate showed a limited specific surface area. Recently, magnetic nanoparticles have found more and more application [28–32]. To overcome these drawbacks, we proposed the use of magnetic beads (MBs) and chemiluminescence (CL) platform for rapid and sensitive miRNA detection. The large specific surface area and excellent separation capability of MBs, and the high signal-to-noise ratio and wide linear range of CL have contributed to powerful strategies for molecular detection [33–38]. However, this platform was rarely applied in miRNA detection previously. In this article,  we described a

Table 1   Comparison of current miRNA detection methods Assay type

Mechanism

Advantage

Disadvantage

Detection limit

References

Northern blotting

Solid-phase

Hybridization between fluorescentor radio-labeled probes and the target miRNA on the nitrocellulose membrane

Widely used

Lowly sensitive, time-consuming

Nanomolar range

[18]

Microarray

Solid-phase

Hybridization between probes on the microarray chips and the target miRNA

High-throughput

Non-quantitative, lowly specific

Femtomolar range

[18]

Real-time PCR

Solutionphase

Primer extension PCR with fluorescent detection by SYBR Green or stem-loop

Highly-sensitive, highly-accurate, quantitative

Costly

Femtomolar range

[21–24]

Bioluminescent miRNA Detection

Solid-phase

Hybridization between bioluminescent enzyme-labeled probes and the target miRNA

Highly-sensitive, quantitative

Decrease signal, low specific surface area

1 fM

[26]

Electrochemical detection

Solid-phase

Oxidation of miRNA-ligated Os(VI) bipy label

Highly-selective

Waste mercury production

Femtomolar range

[27]

Graphene oxide detection

Solid-phase

Hybridization between fluorescent-labeled probes and the target miRNA

Highly-specific

Non-quantitative

Nanomolar range

[25]

Detection method

 

   Sci China Chem   August (2016)  Vol.59  No.8

Zhou et al.

comprehensive strategy for miRNA detection and quantification based on MBs and CL. This approach depends on sandwich hybridization among MBs-labeled capture probes, target miRNAs and biotin-labeled reporter probes, conjugation of streptavidin-alkeline phosphatase (SA-AP) to biotin-labeled reporter probes, and CL detection of AP-linked targets (Scheme 1). The AP-labeled targets are detected upon mixing with 3-(2'-spiroadamantyl)-4-methoxy-4-(3''-phosphoryloxy)phenyl-1,2-dioxetane (AMPPD), as the mixture can generate a prolonged CL signal reflecting the concentration of target miRNA. Herein, miR-21 was chosen as a proof of concept for the reason that the aberrant expression levels of miR-21 had been proved to be related to various tumor cell proliferation, migration, and invasion, and showed great significance for cancer diagnosis and therapy [39–42].

   Experimental 2           Chemicals 2.1        and reagents A total of 35 serum samples including 15 liver cancer subjects (a )

(b)

(c)

1053

confirmed by histological evaluation, and 20 healthy controls, were friendly donated by the Second Affiliated Hospital of Nanjing Medical University with approval by the Review Board of Hospital Ethics Committee. Magnetic beads (~400 nm in diameter) were provided by Nanjing Longliang Biological Science and Technology Co., Ltd (China). (3-Aminopropyl)triethoxysilane (APTES), N-(3-dimethylaminopropyl)-N-ethylcarbodiimide hydrochloride (EDC∙HCl) and hybridization solution were purchased from Sigma-Aldrich (USA). 3-(2'-Spiroadamantane)-4-methoxy-4-(3''-phosphoryloxy) phenyl-1,2-dioxetane (AMPPD) was purchased from Nanjing Duly Biotech Co., Ltd (China). MiRNeasy Serum/Plasma Kit was purchased from QIAGEN (Germany). N,N-dimethylformamide (DMF), succinic anhydride, 2-(N-morpholino) ethanesulfonic acid (MES), Tris, bovine serum albumin (BSA), sodium dodecylsulfate (SDS), salt-sodium citrate (SSC), Tween 20, streptavidin-alkaline phosphatase (SA-AP), diethylpyrocarbonate (DEPC), miRNA First-Strand cDNA synthesis Kit and miRNA SYBR Green qPCR Kit were purchased from Sangon Biotech (Shanghai) Co., Ltd (China). The oligonucleotide probes (capture probe: 5'-CTGATAAGCTA-(T)15-NH2-3'; reporter probe: 5'-Biotin-(T)15-TCAACATCAG-3') and artificial miRNAs (miR-21: 5'-UAGCUUAUCAGACUGAUGUUGA-3'; single-mismatch sequence: 5'-UAGCUUAUCGGACUGAUGUUGA-3'; non-complementary miRNA: 5'-UUGUACUACACAAAAGUACUG-3') were respectively chemically synthesized and HPLC purified by Sangon Biotech (Shanghai) Co., Ltd (China) and Shanghai GenePharma Co., Ltd (China). Other unnamed chemicals were of analytical reagent grade. All of the buffers, tubes and tips were autoclaved and DEPC treated throughout the experiment to protect miRNA from RNase degradation.    Preparation 2.2        of MBs-capture probe

(d)

   Schematic Scheme 1        diagram of chemiluminescent (CL) detection of miRNAs based on magnetic beads (MBs). (a) Extracted target miRNA was first captured by the capture probe conjugated with MBs, (b) and then recognized by the biotin-labeled reporter probe. (c) Through the biotin-streptavidin interaction, alkaline phosphatase (AP) was conjugated to MBs, (d) and then interacted with its substrate AMPPD to generate CL signals. The CL intensity reflected the expression levels of miRNA in real samples.

Surface amination of Fe3O4@SiO2 MBs was initially performed with APTES, followed by modifying carboxylic groups on the surface of MBs using succinic anhydride solution (dissolved in DMF) [34]. The obtained carboxylated MBs were blocked with 2 mL of acetonitrile containing 10% (v/v) acetic anhydride for 1 h and resuspended in deionized water at the concentration of 10 mg/mL [35]. Afterward, the capture probes were covalently immobilized to the carboxylic MBs. Carboxylated MBs (1.5 mg) were washed three times with 150 μL of MES buffer (25 mM, pH 6.0), and incubated with 30 μL of amino-modified capture probes diluted in MES buffer and 20 μL of EDC·HCl (5 mg/mL) for 4 h at 4 °C with continuous shaking. The resultant MBs-capture probe conjugates were magnetically separated, washed three times with 150 μL of Tris buffer (50 mM, pH 7.4), and resuspended in 150 μL of PBS buffer at the concentration of 10 mg/mL to form a ready-to-use capture probe.

1054

 

   Sci China Chem   August (2016)  Vol.59  No.8

Zhou et al.

   Detection 2.3        of miR-21 based on sandwich hybridization and chemiluminescence The detection of miR-21 was performed in the 0.2 mL of Eppendorf tubes. MBs-capture probe conjugates (30 μL, 10 mg/mL) were mixed with 10 μL of hybridization solution, 9 μL of deionized water and 1 μL of chemically synthetic miR-21 solution. The hybridization reaction was conducted for 10 min at 37 °C, and then the hybridized mixtures were washed two times with 2×SSC-0.1% SDS buffer and wash buffer (50 mM Tris, 0.15 M NaCl, pH 7.5) respectively. The obtained MBs-capture probe-target mixtures were mixed with 10 μL of hybridization solution, 9 μL of deionized water and 1 μL of reporter probe solution for 10 min at 37 °C. After a series of washing steps, 60 μL of wash buffer containing 0.25% BSA were added to block the hybridized complex. Subsequently, the hybridized complex was incubated with SA-AP. The CL signals were detected by a Victor X3 Multilabel Plate Reader (PerkinElmer, USA) upon the addition of 100 μL of AMPPD solution (0.25 mM), which is the substrate reagent of AP. The assays of negative and blank controls were respectively carried out using artificial single-mismatch sequence, non-complementary sequence and deionized water instead of miR-21. Each of CL signal readouts was repeated in triplicate.    Hybridization 2.4        parameters investigation and sensitivity test Several parameters, including the concentration of succinic anhydride and capture probe, capture probe-target hybridization temperature (I) and time (I), and reporter probe-target hybridization temperature (II) and time (II), were systematically investigated to establish optimal conditions for CL detection of miR-21. Subsequently, the synthetic miR-21 was diluted into a series of concentrations and a dose-response curve was generated by plotting the CL intensity against the concentrations of miR-21. The CL intensities were the average of three replicates within a single assay, and all the CL intensities were normalized to the blank control value.    MiRNA 2.5        extraction All serum samples were stored at –80 °C until used. The frozen sera were thawed and transferred into Eppendorf tubes. The extraction was carried out by miRNeasy Serum/Plasma Kit (QIAGEN, Germany) according to the protocol provided by the manufacturer. Briefly, 1 mL of QIAzol and 200 μL of chloroform were added to the serum sample, and the mixture was vigorously shaken for 15 s and incubated at room temperature for 2–3 min. After centrifugation at 12000×g for 15 min at 4 °C, the aqueous phase was transferred to a new tube, and 1.5 volume of 100% ethanol was added (e.g. for 600 μL of aqueous phase, 900

μL of ethanol was added). The mixture was centrifuged at ≥8000×g for 15 s at room temperature and the flow-through was discarded. Subsequently, the RNA pellet was washed by washing buffer and 80% ethanol for several times, and air-dried at full speed for 5 min. Finally, 14 μL of RNase-free water was directly added to dissolve the extracted miRNA.    qRT-PCR 2.6        analysis of miR-21 The extracted miR-21 was quantitatively detected by an ABI Prism7500 Real-Time Thermocycler (Applied Biosystems, USA) following the manufacturer’s instructions. RT reaction mixture (20 μL) included 100 ng of miRNAs, 2×miRNA RT Solution mix, miRNA RT Enzyme mix and RNase-free water. The mixture was incubated at 37 °C for 60 min and 85 °C for 5 min. Real-time PCR was carried out in 20 μL of reaction volume involving resultant cDNA templates, 2×miRNA qPCR master mix, a forward primer (5'-ACACTCCAGCTGGGTAGCTTATCAGACTGA-3') [43] and a universal primer (provided by the kit), ROX Reference Dye and RNase-free water by sequentially incubating in a 96-well plate with an initial denaturation at 95 °C for 30 s, followed by 40 cycles of 95 °C for 5 s and 60 °C for 35 s. A dissolving step was added in the end to validate the specificity of the expected PCR product. MiRNA expression level was measured in triplicate using the threshold cycle (Ct), i.e., the cycle number at which the fluorescence of each sample passed the fixed threshold of 0.2.

   Results 3        and discussion    Detection 3.1        of miR-21 based on magnetic beads and chemiluminescence Scheme 1 presents a schematic procedure of CL detection and quantitation of miRNA from real samples. In the presence of the target miRNA, it is first captured to a MBs-capture probe from extracted miRNA solution, followed by hybridization with a biotin-labeled reporter probe. The SA-AP is then conjugated to the hybridized complexes through the biotin-streptavidin interaction. After wash steps, AMPPD, the substrate of AP, is added for CL signal generation. The obtained signal reflects the concentration of target miRNA. As a proof of concept, miR-21 was detected to evaluate the principle and specificity of this method. Figure 1, depicting the CL intensity against the reaction time, illustrated that the signal intensity rapidly increased at the initial stage and then tended stable after 30 min. This indicated that our approach was successfully established. The strong CL intensity of synthetic miR-21 was about 4-, 12- and 20-fold compared with that of single mismatch, negative control and blank respectively, suggesting that this strategy was highly specific to distinguish miR-21 from other nucleic acids.

 

   Sci China Chem   August (2016)  Vol.59  No.8

Zhou et al.

   Plots Figure 1        of CL intensities of synthetic miR-21, single mismatch, negative control (NC) and blank. The CL intensities for each group were an average of three measurements. Error bar: standard deviation of three replicates (color online).

   Hybridization 3.2        parameters investigation Several parameters that may influence the hybridization efficiency were systematically investigated by monitoring CL

1055

intensity changes to optimize the detection method. The density of carboxyl groups on the surface of MBs significantly influenced the number of immobilized probes. As shown in Figure 2(a), the CL intensity evidently increased with the increase in succinic anhydride concentration, reached a maximum at 1 mM, and then decreased, possibly due to the negative charge repulsive interaction by the congested carboxyl groups on the surface of MBs. Hence, 1 mM of the succinic anhydride concentration was used in the subsequent work. The concentration of capture probes on the surface of MBs was one of the key factors to influence the hybridization efficiency. As shown in Figure 2(b), CL intensity was observed to increase over the range of 0.001–0.1 μM and then decreased, possibly due to that the binding sites on the MBs were progressively saturated and extra probes produced steric hindrance to restrict the hybridization efficiency. Thus, 0.1 μM capture probe was selected as the optimal  concentration

(a)

(b)

(c)

(d)

(e)

(f)

   Optimization Figure 2        of assay parameters. Normalized CL intensity vs. (a) concentration of succinic anhydride, (b) concentration of capture probe, (c) hybridization temperature I, (d) hybridization temperature II, (e) hybridization time I, and (f) hybridization time II. Error bar: standard deviation of three replicates.

1056

 

   Sci China Chem   August (2016)  Vol.59  No.8

Zhou et al.

in the subsequent experiments. Additionally, it was reported that temperature had a great impact on the hybridization efficiency. Low temperature may lead to non-specific adsorption, while high temperature may unzip the duplex sequences. The effect of different temperatures on CL intensity was shown in Figure 2(c, d). The CL intensity initially increased and then decreased with the increase of reaction temperature. The figures demonstrated that the optimal hybridization temperature I and II were both 35 °C, which were used in the subsequent work. At the same time, the hybridization time I and II were investigated to achieve the best hybridization performance. It was observed that the CL intensity increased rapidly with the increase of reaction duration, and there was no further increase after 10 min (Figure 2(e)) and 8 min (Figure 2(f)), indicating that the hybridization reaction had reached saturation. Thus, the optimal hybridization time I and II were 10 and 8 min respectively.

   Discrimination 3.4        of HCC samples from healthy controls Encouraged by the high sensitivity and wide linear range for miR-21 detection, 15 patient sera with liver cancer and 20 healthy sera were evaluated by the assay. As shown in Figure 4(a), the distinction of disease and health signals was significant.  The  average  CL  intensity of the cancer group was about 5-fold increase compared with that of the healthy

(a)

   Assay 3.3        sensitivity and calibration curve establishment Under the optimal conditions, a dose-response curve was generated by monitoring the dependence of CL intensity on the miR-21 concentration. The curve in Figure 3 depicted that the CL intensity increased with a linear correlation in an extraordinarily wide range from 10–2 to 104 pM (six orders of magnitude) and leveled off at concentrations higher than 104 pM. The linear part of the dose-response curve was represented as the calibration curve with a linear equation: y = 5328.3x + 12356.8 (R 2 = 0.9827)

(b)

(1)

where y was the CL intensity and x was the logarithm of the miR-21 concentration, which was utilized in the following quantitative detection. The plateau appeared probably due to the saturation of target loading. On the basis of a signal-tonoise ratio of 3, the limit of detection (LOD) was estimated to be 60 fM, indicating the developed method was as sensitive as other reported miRNA detection methods.

   CL Figure 3        intensities plotted versus the log of diluted concentrations of miR-21. Error bar: standard deviation of three replicates.

(c)

   Measurement Figure 4        of miR-21 in clinical serum samples utilizing the developed quantitative CL assay and comparison with qRT-PCR. MiRNAs extracted from 15 liver cancer patient sera and 20 healthy sera (control) were detected by the (a) CL quantitation technique and (b) qRT-PCR. In this box-and-whisker plot, the lines within the boxes represent median values; the upper and lower lines of the boxes represent the 25th and 75th percentiles, respectively; and the upper and lower bars outside the boxes represent the 90th and 10th percentiles, respectively. (c) Comparison of quantitative results of miR-21 in liver cancer serum samples between the CL quantitation technique and qRT-PCR. Error bar: standard deviation of three replicates (color online).

 

   Sci China Chem   August (2016)  Vol.59  No.8

Zhou et al.

1057

group. CL intensity of miR-21 in patient sera was above 10000, while that of in healthy sera was below 10000, indicating that the developed method could positively detect miR-21 from all serum samples with 100% specificity and 100% sensitivity, and successfully discriminated disease samples from healthy subjects according to the levels of miR-21. Meanwhile, to evaluate the quantification performance of this method, the CL detection results were compared with the qRT-PCR analysis results (Figure 4(b)). Higher CL intensities reflected higher miR-21 levels while higher Ct values indicated lower miR-21 levels in sera, in which the tendencies were consistent. Besides, the obtained concentrations from these two methods were excellent correlated with each other (Figure 4(c)). These results demonstrated that the developed technique offers unambiguous accuracy and repeatability in the detection of miR-21. Furthermore, levels of miR-21 are greatly up-regulated in liver cancer, and thus can be proposed as a potential biomarker in liver cancer diagnostics.

    The authors declare that they have no conflict of Conflict of interest      interest.

   Conclusions 4       

12

In summary, we have developed a novel miRNA quantitative approach based on MBs and CL. To our best knowledge, unlike previous researches that required the target amplification procedure, this is the first attempt to apply MBs and CL platform to quantitatively detect miRNAs directly from clinical serum samples without any reverse transcription and nucleic acid amplification procedures, making it particularly suitable for miRNA rapid analysis. Detection of miR-21 using the developed method revealed an ultralow LOD of 60 fM with an extraordinarily wide range of six orders of magnitudes. Moreover, the quantitative accuracy was proved by comparing with the qRT-PCR detection results. Because of its high sensitivity and wide detection range, this novel method facilitated the quantitative analysis of miRNA from clinical samples and showed high selectivity and excellent repeatability. Although only miR-21 was detected as a proof of concept, this approach could be readily applicable to the quantification of other miRNAs.     This Acknowledgments          work was supported by the National Key Program for Developing Basic Research (2014CB744501), the National High Technology Research and Development Program of China (2012AA022703), the National Key Special Science Program (2013ZX10004103-002), the National Natural Science Foundation of China (61471168, 61527806, 61271056), the Special Projects in Jiangsu Province (BL2014094), the Economical Forest Cultivation and Utilization of 2011 Collaborative Innovation Center in Hunan Province [(2013) 448], the Talents Planning of Six Summit Fields of Jiangsu Province (2013-WSN-056), Tianjin Medical University General Hospital Funding (ZYYFY2015029), China Postdoctoral Science Foundation funded project (2015T80487), Natural Science Foundation of Jiangsu Province (BK20140900), and the Open Project of State Key Laboratory of Bioelectronics (2014HX12).

1 2 3 4 5 6 7

8 9 10

11

13

14 15 16

17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32

Nilsen TW. Trends Genets, 2007, 23: 243–249 Bartel DP. Cell, 2009, 136: 215–233 Croce CM, Calin GA. Cell, 2005, 122: 6–7 Kloosterman WP, Plasterk RHA. Dev Cell, 2006, 11: 441–450 Sun K, Lai EC. Nat Rev Genet, 2013, 14: 535–548 Meng F, Henson R, Wehbe–Janek H, Ghoshal K, Jacob ST, Patel T. Gastroenterology, 2007, 133: 647–658 Ji J, Zheng X, Forgues M, Yamashita T, Wauthier EL, Reid LM, Wen X, Song Y, Wei JS, Khan J, Thorgeirsson SS, Wang XW. Hepatology, 2015, 62: 829–840 Cheng G. Adv Drug Deliver Rev, 2015, 81: 75–93 Chim SSC, Shing TKF, Hung ECW, Leung T, Lau T, Chiu RWK, Dennis Lo YM. Clin Chem, 2008, 54: 482–490 Diehl P, Fricke A, Sander L, Stamm J, Bassler N, Htun N, Ziemann M, Helbing T, El-Osta A, Jowett JBM, Peter K. Cardiovasc Res, 2012, 93: 633–644 Arroyo JD, Chevillet JR, Kroh EM, Ruf IK, Pritchard CC, Gibson DF, Mitchell PS, Bennett CF, Pogosova-agadjanyan EL, Stirewalt DL, Tait JF, Tewari M. Proc Natl Acad Sci, 2011, 108: 5003–5008 Turchinovich A, Weiz L, Langheinz A, Burwinkel B. Nucleic Acids Res, 2011, 39: 7223–7233 Mitchell PS, Parkin RK, Kroh EM, Fritz BR, Wyman SK, PogosovaAgadjanyan EL, Peterson A, Noteboom J, O’briant KC, Allen A, Lin DW, Urban N, Drescher CW, Knudsen BS, Stirewalt DL, Gentleman R, Vessella RL, Nelson PS, Martin DB, Tewari M. Proc Natl Acad Sci, 2008, 105: 10513–10518 Alečković M, Kang Y. BBA-Rev Cancer, 2015, 1855: 24–42 Zhou B, Li Z, Yang H, He N. J Biomed Nanotechnol, 2014, 10: 2865–2890 Cuk K, Zucknick M, Heil J, Madhavan D, Schott S, Turchinovich A, Arlt D, Rath M, Sohn C, Benner A, Junkermann H, Schneeweiss A, Burwinkel B. Int J Cancer, 2013, 132: 1602–1612 Takada S, Mano H. Nat Protoc, 2007, 2: 3136–3145 De Planell-saguer M, Rodicio MC. Anal Chim Acta, 2011, 699: 134–152 Schmittgen TD. Nucleic Acids Res, 2004, 32: e43 Raymond CK. RNA, 2005, 11: 1737–1744 Chen C. Nucleic Acids Res, 2005, 33: e179 Shi R, Chiang V. Biotechniques, 2005, 39: 519–525 Castoldi M. RNA, 2006, 12: 913–920 Li J, Yao B, Huang H, Wang Z, Sun C, Fan Y, Chang Q, Li S, Wang X, Xi J. Anal Chem, 2009, 81: 5446–5451 Lu Z, Zhang L, Deng Y, Li S, He N. Nanoscale, 2012, 4: 5840–5842 Cissell KA, Rahimi Y, Shrestha S, Hunt EA, Deo SK. Anal Chem, 2008, 80: 2319–2325 Bartosik M, Hrstka R, Palecek E, Vojtesek B. Anal Chim Acta, 2014, 813: 35–40 Yang H, Ji SF, Liu XF, Zhang DN, Shi D. Sci China Chem, 2014, 57: 866–872 Wang J, Ali Z, Wang N, Liang W, Liu H, Li F, Yang H, He L, Nie L, He N, Li Z. Sci China Chem, 2015, 58: 1774–1778 Liu JW, Xu JJ, Liu ZW, Liu XL, Che RC. Sci China Chem, 2014, 57: 3–12 Liu P, Qi W, Du YF, Li Z, Wang J, Bi JJ, Wu WS. Sci China Chem, 2014, 57: 1483–1490 Fan J, Yang H, Liu M, Wu D, Jiang H, Zeng X, Elingarami S, Li Z, Li

1058

33 34 35 36 37

 

   Sci China Chem   August (2016)  Vol.59  No.8

Zhou et al.

S, Liu H, He N. J Nanosci Nanotechnol, 2015, 15: 1123–1129 He N, Wang F, Ma C, Li C, Zeng X, Deng Y, Zhang L, Li Z. J Biomed Nanotechnol, 2013, 9: 267–273 Tang Y, Zou J, Ma C, Ali Z, Li Z, Li X, Ma N, Mou X, Deng Y, Zhang L, Li K, Lu G, Yang H,  He N. Theranostics, 2013, 3: 85–92 Liu M, Hu P, Zhang G, Zeng Y, Yang H, Fan J, Jin L, Liu H, Deng Y, Li S, Zeng X, Elingarami S, He N. Theranostics, 2015, 5: 71–85 Li Z, Yang H, He N, Liang W, Ma C, Shah MAA, Tang Y, Li S, Liu H, Jiang H, Guo Y. J Biomed Nanotechnol, 2013, 9: 1945–1949 Yang H, Liang W, He N, Deng Y, Li Z. ACS Appl Mater Interf, 2015, 7: 774–781

38 39 40 41 42 43

Yang H, Liang W, Si J, Li Z, He N. J Biomed Nanotechnol, 2014, 10: 3610–3619 Ren WH, Qiang C, Gao L, Li SM, Zhang LM, Wang XL, Dong JW, Chen C, Liu CY, Zhi KQ. Biomarkers, 2014, 19: 590–596 Frankel LB, Christoffersen NR, Jacobsen A, Lindow M, Krogh A, Lund AH. J Biol Chem, 2008, 283: 1026–1033 Guven B, Dudak FC, Boyaci IH, Tamer U, Ozsoz M. Analyst, 2014, 139: 1141–1147 Wang DC, Hu LH, Zhou YH, Huang YT, Li X, Zhu JJ. J Biomed Nanotechnol, 2014, 10: 615–621 Tang F. Nucleic Acids Res, 2006, 34: e9